From b64f34cdfe5bef9dfed1304c513220b0f2862eca Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Tue, 29 Jan 2008 14:27:30 +0900
Subject: [IA64] VIRT_CPU_ACCOUNTING (accurate cpu time accounting)

This patch implements VIRT_CPU_ACCOUNTING for ia64,
which enable us to use more accurate cpu time accounting.

The VIRT_CPU_ACCOUNTING is an item of kernel config, which s390
and powerpc arch have.  By turning this config on, these archs
change the mechanism of cpu time accounting from tick-sampling
based one to state-transition based one.

The state-transition based accounting is done by checking time
(cycle counter in processor) at every state-transition point,
such as entrance/exit of kernel, interrupt, softirq etc.
The difference between point to point is the actual time consumed
during in the state. There is no doubt about that this value is
more accurate than that of tick-sampling based accounting.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/cputime.h     | 104 +++++++++++++++++++++++++++++++++++++++++
 include/asm-ia64/system.h      |  12 +++++
 include/asm-ia64/thread_info.h |  14 ++++++
 3 files changed, 130 insertions(+)

(limited to 'include')

diff --git a/include/asm-ia64/cputime.h b/include/asm-ia64/cputime.h
index 72400a78002a..f9abdec6577a 100644
--- a/include/asm-ia64/cputime.h
+++ b/include/asm-ia64/cputime.h
@@ -1,6 +1,110 @@
+/*
+ * include/asm-ia64/cputime.h:
+ *		Definitions for measuring cputime on ia64 machines.
+ *
+ * Based on <asm-powerpc/cputime.h>.
+ *
+ * Copyright (C) 2007 FUJITSU LIMITED
+ * Copyright (C) 2007 Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * If we have CONFIG_VIRT_CPU_ACCOUNTING, we measure cpu time in nsec.
+ * Otherwise we measure cpu time in jiffies using the generic definitions.
+ */
+
 #ifndef __IA64_CPUTIME_H
 #define __IA64_CPUTIME_H
 
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
 #include <asm-generic/cputime.h>
+#else
+
+#include <linux/time.h>
+#include <linux/jiffies.h>
+#include <asm/processor.h>
+
+typedef u64 cputime_t;
+typedef u64 cputime64_t;
+
+#define cputime_zero			((cputime_t)0)
+#define cputime_max			((~((cputime_t)0) >> 1) - 1)
+#define cputime_add(__a, __b)		((__a) +  (__b))
+#define cputime_sub(__a, __b)		((__a) -  (__b))
+#define cputime_div(__a, __n)		((__a) /  (__n))
+#define cputime_halve(__a)		((__a) >> 1)
+#define cputime_eq(__a, __b)		((__a) == (__b))
+#define cputime_gt(__a, __b)		((__a) >  (__b))
+#define cputime_ge(__a, __b)		((__a) >= (__b))
+#define cputime_lt(__a, __b)		((__a) <  (__b))
+#define cputime_le(__a, __b)		((__a) <= (__b))
+
+#define cputime64_zero			((cputime64_t)0)
+#define cputime64_add(__a, __b)		((__a) + (__b))
+#define cputime64_sub(__a, __b)		((__a) - (__b))
+#define cputime_to_cputime64(__ct)	(__ct)
+
+/*
+ * Convert cputime <-> jiffies (HZ)
+ */
+#define cputime_to_jiffies(__ct)	((__ct) / (NSEC_PER_SEC / HZ))
+#define jiffies_to_cputime(__jif)	((__jif) * (NSEC_PER_SEC / HZ))
+#define cputime64_to_jiffies64(__ct)	((__ct) / (NSEC_PER_SEC / HZ))
+#define jiffies64_to_cputime64(__jif)	((__jif) * (NSEC_PER_SEC / HZ))
+
+/*
+ * Convert cputime <-> milliseconds
+ */
+#define cputime_to_msecs(__ct)		((__ct) / NSEC_PER_MSEC)
+#define msecs_to_cputime(__msecs)	((__msecs) * NSEC_PER_MSEC)
+
+/*
+ * Convert cputime <-> seconds
+ */
+#define cputime_to_secs(__ct)		((__ct) / NSEC_PER_SEC)
+#define secs_to_cputime(__secs)		((__secs) * NSEC_PER_SEC)
+
+/*
+ * Convert cputime <-> timespec (nsec)
+ */
+static inline cputime_t timespec_to_cputime(const struct timespec *val)
+{
+	cputime_t ret = val->tv_sec * NSEC_PER_SEC;
+	return (ret + val->tv_nsec);
+}
+static inline void cputime_to_timespec(const cputime_t ct, struct timespec *val)
+{
+	val->tv_sec  = ct / NSEC_PER_SEC;
+	val->tv_nsec = ct % NSEC_PER_SEC;
+}
+
+/*
+ * Convert cputime <-> timeval (msec)
+ */
+static inline cputime_t timeval_to_cputime(struct timeval *val)
+{
+	cputime_t ret = val->tv_sec * NSEC_PER_SEC;
+	return (ret + val->tv_usec * NSEC_PER_USEC);
+}
+static inline void cputime_to_timeval(const cputime_t ct, struct timeval *val)
+{
+	val->tv_sec = ct / NSEC_PER_SEC;
+	val->tv_usec = (ct % NSEC_PER_SEC) / NSEC_PER_USEC;
+}
+
+/*
+ * Convert cputime <-> clock (USER_HZ)
+ */
+#define cputime_to_clock_t(__ct)	((__ct) / (NSEC_PER_SEC / USER_HZ))
+#define clock_t_to_cputime(__x)		((__x) * (NSEC_PER_SEC / USER_HZ))
+
+/*
+ * Convert cputime64 to clock.
+ */
+#define cputime64_to_clock_t(__ct)      cputime_to_clock_t((cputime_t)__ct)
 
+#endif /* CONFIG_VIRT_CPU_ACCOUNTING */
 #endif /* __IA64_CPUTIME_H */
diff --git a/include/asm-ia64/system.h b/include/asm-ia64/system.h
index 595112bca3cc..dff8128fa58e 100644
--- a/include/asm-ia64/system.h
+++ b/include/asm-ia64/system.h
@@ -210,6 +210,13 @@ struct task_struct;
 extern void ia64_save_extra (struct task_struct *task);
 extern void ia64_load_extra (struct task_struct *task);
 
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+extern void ia64_account_on_switch (struct task_struct *prev, struct task_struct *next);
+# define IA64_ACCOUNT_ON_SWITCH(p,n) ia64_account_on_switch(p,n)
+#else
+# define IA64_ACCOUNT_ON_SWITCH(p,n)
+#endif
+
 #ifdef CONFIG_PERFMON
   DECLARE_PER_CPU(unsigned long, pfm_syst_info);
 # define PERFMON_IS_SYSWIDE() (__get_cpu_var(pfm_syst_info) & 0x1)
@@ -222,6 +229,7 @@ extern void ia64_load_extra (struct task_struct *task);
 	 || IS_IA32_PROCESS(task_pt_regs(t)) || PERFMON_IS_SYSWIDE())
 
 #define __switch_to(prev,next,last) do {							 \
+	IA64_ACCOUNT_ON_SWITCH(prev, next);							 \
 	if (IA64_HAS_EXTRA_STATE(prev))								 \
 		ia64_save_extra(prev);								 \
 	if (IA64_HAS_EXTRA_STATE(next))								 \
@@ -266,6 +274,10 @@ void cpu_idle_wait(void);
 
 void default_idle(void);
 
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+extern void account_system_vtime(struct task_struct *);
+#endif
+
 #endif /* __KERNEL__ */
 
 #endif /* __ASSEMBLY__ */
diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h
index 93d83cbe0c8c..6da8069a0f77 100644
--- a/include/asm-ia64/thread_info.h
+++ b/include/asm-ia64/thread_info.h
@@ -31,6 +31,12 @@ struct thread_info {
 	mm_segment_t addr_limit;	/* user-level address space limit */
 	int preempt_count;		/* 0=premptable, <0=BUG; will also serve as bh-counter */
 	struct restart_block restart_block;
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+	__u64 ac_stamp;
+	__u64 ac_leave;
+	__u64 ac_stime;
+	__u64 ac_utime;
+#endif
 };
 
 #define THREAD_SIZE			KERNEL_STACK_SIZE
@@ -62,9 +68,17 @@ struct thread_info {
 #define task_stack_page(tsk)	((void *)(tsk))
 
 #define __HAVE_THREAD_FUNCTIONS
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+#define setup_thread_stack(p, org)			\
+	*task_thread_info(p) = *task_thread_info(org);	\
+	task_thread_info(p)->ac_stime = 0;		\
+	task_thread_info(p)->ac_utime = 0;		\
+	task_thread_info(p)->task = (p);
+#else
 #define setup_thread_stack(p, org) \
 	*task_thread_info(p) = *task_thread_info(org); \
 	task_thread_info(p)->task = (p);
+#endif
 #define end_of_stack(p) (unsigned long *)((void *)(p) + IA64_RBS_OFFSET)
 
 #define __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
-- 
cgit v1.2.3


From e5b13cb10de209f924fdf9478214bcf7e4008d6d Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 28 Feb 2008 20:51:43 -0800
Subject: [NETNS]: Process devinet ioctl in the correct namespace.

Add namespace parameter to devinet_ioctl and locate device inside it for
state changes.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h | 2 +-
 net/ipv4/af_inet.c         | 7 ++++---
 net/ipv4/devinet.c         | 6 +++---
 net/ipv4/ipconfig.c        | 2 +-
 4 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index fc4e3db649e8..da05ab47ff2f 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -129,7 +129,7 @@ extern int unregister_inetaddr_notifier(struct notifier_block *nb);
 
 extern struct net_device *ip_dev_find(struct net *net, __be32 addr);
 extern int		inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b);
-extern int		devinet_ioctl(unsigned int cmd, void __user *);
+extern int		devinet_ioctl(struct net *net, unsigned int cmd, void __user *);
 extern void		devinet_init(void);
 extern struct in_device	*inetdev_by_index(struct net *, int);
 extern __be32		inet_select_addr(const struct net_device *dev, __be32 dst, int scope);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 09ca5293d08f..c270080f370e 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -784,6 +784,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 {
 	struct sock *sk = sock->sk;
 	int err = 0;
+	struct net *net = sk->sk_net;
 
 	switch (cmd) {
 		case SIOCGSTAMP:
@@ -795,12 +796,12 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		case SIOCADDRT:
 		case SIOCDELRT:
 		case SIOCRTMSG:
-			err = ip_rt_ioctl(sk->sk_net, cmd, (void __user *)arg);
+			err = ip_rt_ioctl(net, cmd, (void __user *)arg);
 			break;
 		case SIOCDARP:
 		case SIOCGARP:
 		case SIOCSARP:
-			err = arp_ioctl(sk->sk_net, cmd, (void __user *)arg);
+			err = arp_ioctl(net, cmd, (void __user *)arg);
 			break;
 		case SIOCGIFADDR:
 		case SIOCSIFADDR:
@@ -813,7 +814,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		case SIOCSIFPFLAGS:
 		case SIOCGIFPFLAGS:
 		case SIOCSIFFLAGS:
-			err = devinet_ioctl(cmd, (void __user *)arg);
+			err = devinet_ioctl(net, cmd, (void __user *)arg);
 			break;
 		default:
 			if (sk->sk_prot->ioctl)
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 90210a74d638..af752fc5d0ab 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -595,7 +595,7 @@ static __inline__ int inet_abc_len(__be32 addr)
 }
 
 
-int devinet_ioctl(unsigned int cmd, void __user *arg)
+int devinet_ioctl(struct net *net, unsigned int cmd, void __user *arg)
 {
 	struct ifreq ifr;
 	struct sockaddr_in sin_orig;
@@ -624,7 +624,7 @@ int devinet_ioctl(unsigned int cmd, void __user *arg)
 		*colon = 0;
 
 #ifdef CONFIG_KMOD
-	dev_load(&init_net, ifr.ifr_name);
+	dev_load(net, ifr.ifr_name);
 #endif
 
 	switch (cmd) {
@@ -665,7 +665,7 @@ int devinet_ioctl(unsigned int cmd, void __user *arg)
 	rtnl_lock();
 
 	ret = -ENODEV;
-	if ((dev = __dev_get_by_name(&init_net, ifr.ifr_name)) == NULL)
+	if ((dev = __dev_get_by_name(net, ifr.ifr_name)) == NULL)
 		goto done;
 
 	if (colon)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 10013ccee8dd..c90e75a66e81 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -291,7 +291,7 @@ static int __init ic_dev_ioctl(unsigned int cmd, struct ifreq *arg)
 
 	mm_segment_t oldfs = get_fs();
 	set_fs(get_ds());
-	res = devinet_ioctl(cmd, (struct ifreq __user *) arg);
+	res = devinet_ioctl(&init_net, cmd, (struct ifreq __user *) arg);
 	set_fs(oldfs);
 	return res;
 }
-- 
cgit v1.2.3


From 99cd07a537bf4c7c954f31611e30064407104410 Mon Sep 17 00:00:00 2001
From: Juha-Matti Tapio <jmtapio@verkkotelakka.net>
Date: Thu, 28 Feb 2008 20:55:46 -0800
Subject: [IPV6]: Fix source address selection for ORCHID addresses

Skip the prefix length matching in source address selection for
orchid -> non-orchid addresses.

Overlay Routable Cryptographic Hash IDentifiers (RFC 4843,
2001:10::/28) are currenty not globally reachable. Without this
check a host with an ORCHID address can end up preferring those over
regular addresses when talking to other regular hosts in the 2001::/16
range thus breaking non-orchid connections.

Signed-off-by: Juha-Matti Tapio <jmtapio@verkkotelakka.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h  | 10 ++++++++++
 net/ipv6/addrconf.c |  5 +++++
 2 files changed, 15 insertions(+)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index c0c019f72ba9..8b05c65415cb 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -383,6 +383,16 @@ static inline int ipv6_addr_v4mapped(const struct in6_addr *a)
 		 a->s6_addr32[2] == htonl(0x0000ffff));
 }
 
+/*
+ * Check for a RFC 4843 ORCHID address
+ * (Overlay Routable Cryptographic Hash Identifiers)
+ */
+static inline int ipv6_addr_orchid(const struct in6_addr *a)
+{
+	return ((a->s6_addr32[0] & htonl(0xfffffff0))
+		== htonl(0x20010010));
+}
+
 /*
  * find the first different bit between two addresses
  * length of address must be a multiple of 32bits
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 101e0e70ba27..18e3a9825d81 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1125,6 +1125,11 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev,
 			if (hiscore.rule < 7)
 				hiscore.rule++;
 #endif
+
+			/* Skip rule 8 for orchid -> non-orchid address pairs. */
+			if (ipv6_addr_orchid(&ifa->addr) && !ipv6_addr_orchid(daddr))
+				continue;
+
 			/* Rule 8: Use longest matching prefix */
 			if (hiscore.rule < 8) {
 				hiscore.matchlen = ipv6_addr_diff(&ifa_result->addr, daddr);
-- 
cgit v1.2.3


From 4c563f7669c10a12354b72b518c2287ffc6ebfb3 Mon Sep 17 00:00:00 2001
From: Timo Teras <timo.teras@iki.fi>
Date: Thu, 28 Feb 2008 21:31:08 -0800
Subject: [XFRM]: Speed up xfrm_policy and xfrm_state walking

Change xfrm_policy and xfrm_state walking algorithm from O(n^2) to O(n).
This is achieved adding the entries to one more list which is used
solely for walking the entries.

This also fixes some races where the dump can have duplicate or missing
entries when the SPD/SADB is modified during an ongoing dump.

Dumping SADB with 20000 entries using "time ip xfrm state" the sys
time dropped from 1.012s to 0.080s.

Signed-off-by: Timo Teras <timo.teras@iki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h   |  3 +-
 include/net/xfrm.h     | 52 +++++++++++++++++++++++++++++++--
 net/key/af_key.c       | 24 ++++++++++++---
 net/xfrm/xfrm_policy.c | 79 +++++++++++++++++++++++++++++---------------------
 net/xfrm/xfrm_state.c  | 53 ++++++++++++++++++++++-----------
 net/xfrm/xfrm_user.c   | 71 +++++++++++++++++++++++++++------------------
 6 files changed, 197 insertions(+), 85 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index e31b8c84f2c9..0c82c80b277f 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -113,7 +113,8 @@ enum
 {
 	XFRM_POLICY_TYPE_MAIN	= 0,
 	XFRM_POLICY_TYPE_SUB	= 1,
-	XFRM_POLICY_TYPE_MAX	= 2
+	XFRM_POLICY_TYPE_MAX	= 2,
+	XFRM_POLICY_TYPE_ANY	= 255
 };
 
 enum
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index eea7785cc757..9b6205665190 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -121,6 +121,7 @@ extern struct mutex xfrm_cfg_mutex;
 struct xfrm_state
 {
 	/* Note: bydst is re-used during gc */
+	struct list_head	all;
 	struct hlist_node	bydst;
 	struct hlist_node	bysrc;
 	struct hlist_node	byspi;
@@ -424,6 +425,7 @@ struct xfrm_tmpl
 struct xfrm_policy
 {
 	struct xfrm_policy	*next;
+	struct list_head	bytype;
 	struct hlist_node	bydst;
 	struct hlist_node	byidx;
 
@@ -1160,6 +1162,18 @@ struct xfrm6_tunnel {
 	int priority;
 };
 
+struct xfrm_state_walk {
+	struct xfrm_state *state;
+	int count;
+	u8 proto;
+};
+
+struct xfrm_policy_walk {
+	struct xfrm_policy *policy;
+	int count;
+	u8 type, cur_type;
+};
+
 extern void xfrm_init(void);
 extern void xfrm4_init(void);
 extern void xfrm_state_init(void);
@@ -1184,7 +1198,23 @@ static inline void xfrm6_fini(void)
 extern int xfrm_proc_init(void);
 #endif
 
-extern int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*), void *);
+static inline void xfrm_state_walk_init(struct xfrm_state_walk *walk, u8 proto)
+{
+	walk->proto = proto;
+	walk->state = NULL;
+	walk->count = 0;
+}
+
+static inline void xfrm_state_walk_done(struct xfrm_state_walk *walk)
+{
+	if (walk->state != NULL) {
+		xfrm_state_put(walk->state);
+		walk->state = NULL;
+	}
+}
+
+extern int xfrm_state_walk(struct xfrm_state_walk *walk,
+			   int (*func)(struct xfrm_state *, int, void*), void *);
 extern struct xfrm_state *xfrm_state_alloc(void);
 extern struct xfrm_state *xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, 
 					  struct flowi *fl, struct xfrm_tmpl *tmpl,
@@ -1306,7 +1336,25 @@ static inline int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb)
 #endif
 
 struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp);
-extern int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*), void *);
+
+static inline void xfrm_policy_walk_init(struct xfrm_policy_walk *walk, u8 type)
+{
+	walk->cur_type = XFRM_POLICY_TYPE_MAIN;
+	walk->type = type;
+	walk->policy = NULL;
+	walk->count = 0;
+}
+
+static inline void xfrm_policy_walk_done(struct xfrm_policy_walk *walk)
+{
+	if (walk->policy != NULL) {
+		xfrm_pol_put(walk->policy);
+		walk->policy = NULL;
+	}
+}
+
+extern int xfrm_policy_walk(struct xfrm_policy_walk *walk,
+	int (*func)(struct xfrm_policy *, int, int, void*), void *);
 int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl);
 struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir,
 					  struct xfrm_selector *sel,
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 8b5f486ac80f..7cb6f1213360 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -1742,12 +1742,18 @@ static int pfkey_dump(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr
 {
 	u8 proto;
 	struct pfkey_dump_data data = { .skb = skb, .hdr = hdr, .sk = sk };
+	struct xfrm_state_walk walk;
+	int rc;
 
 	proto = pfkey_satype2proto(hdr->sadb_msg_satype);
 	if (proto == 0)
 		return -EINVAL;
 
-	return xfrm_state_walk(proto, dump_sa, &data);
+	xfrm_state_walk_init(&walk, proto);
+	rc = xfrm_state_walk(&walk, dump_sa, &data);
+	xfrm_state_walk_done(&walk);
+
+	return rc;
 }
 
 static int pfkey_promisc(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
@@ -1780,7 +1786,9 @@ static int check_reqid(struct xfrm_policy *xp, int dir, int count, void *ptr)
 
 static u32 gen_reqid(void)
 {
+	struct xfrm_policy_walk walk;
 	u32 start;
+	int rc;
 	static u32 reqid = IPSEC_MANUAL_REQID_MAX;
 
 	start = reqid;
@@ -1788,8 +1796,10 @@ static u32 gen_reqid(void)
 		++reqid;
 		if (reqid == 0)
 			reqid = IPSEC_MANUAL_REQID_MAX+1;
-		if (xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, check_reqid,
-				     (void*)&reqid) != -EEXIST)
+		xfrm_policy_walk_init(&walk, XFRM_POLICY_TYPE_MAIN);
+		rc = xfrm_policy_walk(&walk, check_reqid, (void*)&reqid);
+		xfrm_policy_walk_done(&walk);
+		if (rc != -EEXIST)
 			return reqid;
 	} while (reqid != start);
 	return 0;
@@ -2665,8 +2675,14 @@ static int dump_sp(struct xfrm_policy *xp, int dir, int count, void *ptr)
 static int pfkey_spddump(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
 {
 	struct pfkey_dump_data data = { .skb = skb, .hdr = hdr, .sk = sk };
+	struct xfrm_policy_walk walk;
+	int rc;
+
+	xfrm_policy_walk_init(&walk, XFRM_POLICY_TYPE_MAIN);
+	rc = xfrm_policy_walk(&walk, dump_sp, &data);
+	xfrm_policy_walk_done(&walk);
 
-	return xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, dump_sp, &data);
+	return rc;
 }
 
 static int key_notify_policy_flush(struct km_event *c)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 9fc4c315f6cd..bae94a8031a2 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -46,6 +46,7 @@ EXPORT_SYMBOL(xfrm_cfg_mutex);
 
 static DEFINE_RWLOCK(xfrm_policy_lock);
 
+static struct list_head xfrm_policy_bytype[XFRM_POLICY_TYPE_MAX];
 unsigned int xfrm_policy_count[XFRM_POLICY_MAX*2];
 EXPORT_SYMBOL(xfrm_policy_count);
 
@@ -208,6 +209,7 @@ struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp)
 	policy = kzalloc(sizeof(struct xfrm_policy), gfp);
 
 	if (policy) {
+		INIT_LIST_HEAD(&policy->bytype);
 		INIT_HLIST_NODE(&policy->bydst);
 		INIT_HLIST_NODE(&policy->byidx);
 		rwlock_init(&policy->lock);
@@ -230,6 +232,10 @@ void xfrm_policy_destroy(struct xfrm_policy *policy)
 	if (del_timer(&policy->timer))
 		BUG();
 
+	write_lock_bh(&xfrm_policy_lock);
+	list_del(&policy->bytype);
+	write_unlock_bh(&xfrm_policy_lock);
+
 	security_xfrm_policy_free(policy);
 	kfree(policy);
 }
@@ -584,6 +590,7 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
 	policy->curlft.use_time = 0;
 	if (!mod_timer(&policy->timer, jiffies + HZ))
 		xfrm_pol_hold(policy);
+	list_add_tail(&policy->bytype, &xfrm_policy_bytype[policy->type]);
 	write_unlock_bh(&xfrm_policy_lock);
 
 	if (delpol)
@@ -822,57 +829,60 @@ out:
 }
 EXPORT_SYMBOL(xfrm_policy_flush);
 
-int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*),
+int xfrm_policy_walk(struct xfrm_policy_walk *walk,
+		     int (*func)(struct xfrm_policy *, int, int, void*),
 		     void *data)
 {
-	struct xfrm_policy *pol, *last = NULL;
-	struct hlist_node *entry;
-	int dir, last_dir = 0, count, error;
+	struct xfrm_policy *old, *pol, *last = NULL;
+	int error = 0;
+
+	if (walk->type >= XFRM_POLICY_TYPE_MAX &&
+	    walk->type != XFRM_POLICY_TYPE_ANY)
+		return -EINVAL;
 
+	if (walk->policy == NULL && walk->count != 0)
+		return 0;
+
+	old = pol = walk->policy;
+	walk->policy = NULL;
 	read_lock_bh(&xfrm_policy_lock);
-	count = 0;
 
-	for (dir = 0; dir < 2*XFRM_POLICY_MAX; dir++) {
-		struct hlist_head *table = xfrm_policy_bydst[dir].table;
-		int i;
+	for (; walk->cur_type < XFRM_POLICY_TYPE_MAX; walk->cur_type++) {
+		if (walk->type != walk->cur_type &&
+		    walk->type != XFRM_POLICY_TYPE_ANY)
+			continue;
 
-		hlist_for_each_entry(pol, entry,
-				     &xfrm_policy_inexact[dir], bydst) {
-			if (pol->type != type)
+		if (pol == NULL) {
+			pol = list_first_entry(&xfrm_policy_bytype[walk->cur_type],
+					       struct xfrm_policy, bytype);
+		}
+		list_for_each_entry_from(pol, &xfrm_policy_bytype[walk->cur_type], bytype) {
+			if (pol->dead)
 				continue;
 			if (last) {
-				error = func(last, last_dir % XFRM_POLICY_MAX,
-					     count, data);
-				if (error)
+				error = func(last, xfrm_policy_id2dir(last->index),
+					     walk->count, data);
+				if (error) {
+					xfrm_pol_hold(last);
+					walk->policy = last;
 					goto out;
-			}
-			last = pol;
-			last_dir = dir;
-			count++;
-		}
-		for (i = xfrm_policy_bydst[dir].hmask; i >= 0; i--) {
-			hlist_for_each_entry(pol, entry, table + i, bydst) {
-				if (pol->type != type)
-					continue;
-				if (last) {
-					error = func(last, last_dir % XFRM_POLICY_MAX,
-						     count, data);
-					if (error)
-						goto out;
 				}
-				last = pol;
-				last_dir = dir;
-				count++;
 			}
+			last = pol;
+			walk->count++;
 		}
+		pol = NULL;
 	}
-	if (count == 0) {
+	if (walk->count == 0) {
 		error = -ENOENT;
 		goto out;
 	}
-	error = func(last, last_dir % XFRM_POLICY_MAX, 0, data);
+	if (last)
+		error = func(last, xfrm_policy_id2dir(last->index), 0, data);
 out:
 	read_unlock_bh(&xfrm_policy_lock);
+	if (old != NULL)
+		xfrm_pol_put(old);
 	return error;
 }
 EXPORT_SYMBOL(xfrm_policy_walk);
@@ -2365,6 +2375,9 @@ static void __init xfrm_policy_init(void)
 			panic("XFRM: failed to allocate bydst hash\n");
 	}
 
+	for (dir = 0; dir < XFRM_POLICY_TYPE_MAX; dir++)
+		INIT_LIST_HEAD(&xfrm_policy_bytype[dir]);
+
 	INIT_WORK(&xfrm_policy_gc_work, xfrm_policy_gc_task);
 	register_netdevice_notifier(&xfrm_dev_notifier);
 }
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 7ba65e82941c..9880b792e6a5 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -50,6 +50,7 @@ static DEFINE_SPINLOCK(xfrm_state_lock);
  * Main use is finding SA after policy selected tunnel or transport mode.
  * Also, it can be used by ah/esp icmp error handler to find offending SA.
  */
+static LIST_HEAD(xfrm_state_all);
 static struct hlist_head *xfrm_state_bydst __read_mostly;
 static struct hlist_head *xfrm_state_bysrc __read_mostly;
 static struct hlist_head *xfrm_state_byspi __read_mostly;
@@ -510,6 +511,7 @@ struct xfrm_state *xfrm_state_alloc(void)
 	if (x) {
 		atomic_set(&x->refcnt, 1);
 		atomic_set(&x->tunnel_users, 0);
+		INIT_LIST_HEAD(&x->all);
 		INIT_HLIST_NODE(&x->bydst);
 		INIT_HLIST_NODE(&x->bysrc);
 		INIT_HLIST_NODE(&x->byspi);
@@ -533,6 +535,10 @@ void __xfrm_state_destroy(struct xfrm_state *x)
 {
 	BUG_TRAP(x->km.state == XFRM_STATE_DEAD);
 
+	spin_lock_bh(&xfrm_state_lock);
+	list_del(&x->all);
+	spin_unlock_bh(&xfrm_state_lock);
+
 	spin_lock_bh(&xfrm_state_gc_lock);
 	hlist_add_head(&x->bydst, &xfrm_state_gc_list);
 	spin_unlock_bh(&xfrm_state_gc_lock);
@@ -909,6 +915,8 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 
 	x->genid = ++xfrm_state_genid;
 
+	list_add_tail(&x->all, &xfrm_state_all);
+
 	h = xfrm_dst_hash(&x->id.daddr, &x->props.saddr,
 			  x->props.reqid, x->props.family);
 	hlist_add_head(&x->bydst, xfrm_state_bydst+h);
@@ -1518,36 +1526,47 @@ unlock:
 }
 EXPORT_SYMBOL(xfrm_alloc_spi);
 
-int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
+int xfrm_state_walk(struct xfrm_state_walk *walk,
+		    int (*func)(struct xfrm_state *, int, void*),
 		    void *data)
 {
-	int i;
-	struct xfrm_state *x, *last = NULL;
-	struct hlist_node *entry;
-	int count = 0;
+	struct xfrm_state *old, *x, *last = NULL;
 	int err = 0;
 
+	if (walk->state == NULL && walk->count != 0)
+		return 0;
+
+	old = x = walk->state;
+	walk->state = NULL;
 	spin_lock_bh(&xfrm_state_lock);
-	for (i = 0; i <= xfrm_state_hmask; i++) {
-		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
-			if (!xfrm_id_proto_match(x->id.proto, proto))
-				continue;
-			if (last) {
-				err = func(last, count, data);
-				if (err)
-					goto out;
+	if (x == NULL)
+		x = list_first_entry(&xfrm_state_all, struct xfrm_state, all);
+	list_for_each_entry_from(x, &xfrm_state_all, all) {
+		if (x->km.state == XFRM_STATE_DEAD)
+			continue;
+		if (!xfrm_id_proto_match(x->id.proto, walk->proto))
+			continue;
+		if (last) {
+			err = func(last, walk->count, data);
+			if (err) {
+				xfrm_state_hold(last);
+				walk->state = last;
+				goto out;
 			}
-			last = x;
-			count++;
 		}
+		last = x;
+		walk->count++;
 	}
-	if (count == 0) {
+	if (walk->count == 0) {
 		err = -ENOENT;
 		goto out;
 	}
-	err = func(last, 0, data);
+	if (last)
+		err = func(last, 0, data);
 out:
 	spin_unlock_bh(&xfrm_state_lock);
+	if (old != NULL)
+		xfrm_state_put(old);
 	return err;
 }
 EXPORT_SYMBOL(xfrm_state_walk);
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index f971ca5645f8..f5fd5b3147cc 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -532,8 +532,6 @@ struct xfrm_dump_info {
 	struct sk_buff *out_skb;
 	u32 nlmsg_seq;
 	u16 nlmsg_flags;
-	int start_idx;
-	int this_idx;
 };
 
 static int copy_sec_ctx(struct xfrm_sec_ctx *s, struct sk_buff *skb)
@@ -600,9 +598,6 @@ static int dump_one_state(struct xfrm_state *x, int count, void *ptr)
 	struct nlmsghdr *nlh;
 	int err;
 
-	if (sp->this_idx < sp->start_idx)
-		goto out;
-
 	nlh = nlmsg_put(skb, NETLINK_CB(in_skb).pid, sp->nlmsg_seq,
 			XFRM_MSG_NEWSA, sizeof(*p), sp->nlmsg_flags);
 	if (nlh == NULL)
@@ -615,8 +610,6 @@ static int dump_one_state(struct xfrm_state *x, int count, void *ptr)
 		goto nla_put_failure;
 
 	nlmsg_end(skb, nlh);
-out:
-	sp->this_idx++;
 	return 0;
 
 nla_put_failure:
@@ -624,18 +617,32 @@ nla_put_failure:
 	return err;
 }
 
+static int xfrm_dump_sa_done(struct netlink_callback *cb)
+{
+	struct xfrm_state_walk *walk = (struct xfrm_state_walk *) &cb->args[1];
+	xfrm_state_walk_done(walk);
+	return 0;
+}
+
 static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	struct xfrm_state_walk *walk = (struct xfrm_state_walk *) &cb->args[1];
 	struct xfrm_dump_info info;
 
+	BUILD_BUG_ON(sizeof(struct xfrm_state_walk) >
+		     sizeof(cb->args) - sizeof(cb->args[0]));
+
 	info.in_skb = cb->skb;
 	info.out_skb = skb;
 	info.nlmsg_seq = cb->nlh->nlmsg_seq;
 	info.nlmsg_flags = NLM_F_MULTI;
-	info.this_idx = 0;
-	info.start_idx = cb->args[0];
-	(void) xfrm_state_walk(0, dump_one_state, &info);
-	cb->args[0] = info.this_idx;
+
+	if (!cb->args[0]) {
+		cb->args[0] = 1;
+		xfrm_state_walk_init(walk, 0);
+	}
+
+	(void) xfrm_state_walk(walk, dump_one_state, &info);
 
 	return skb->len;
 }
@@ -654,7 +661,6 @@ static struct sk_buff *xfrm_state_netlink(struct sk_buff *in_skb,
 	info.out_skb = skb;
 	info.nlmsg_seq = seq;
 	info.nlmsg_flags = 0;
-	info.this_idx = info.start_idx = 0;
 
 	if (dump_one_state(x, 0, &info)) {
 		kfree_skb(skb);
@@ -1232,9 +1238,6 @@ static int dump_one_policy(struct xfrm_policy *xp, int dir, int count, void *ptr
 	struct sk_buff *skb = sp->out_skb;
 	struct nlmsghdr *nlh;
 
-	if (sp->this_idx < sp->start_idx)
-		goto out;
-
 	nlh = nlmsg_put(skb, NETLINK_CB(in_skb).pid, sp->nlmsg_seq,
 			XFRM_MSG_NEWPOLICY, sizeof(*p), sp->nlmsg_flags);
 	if (nlh == NULL)
@@ -1250,8 +1253,6 @@ static int dump_one_policy(struct xfrm_policy *xp, int dir, int count, void *ptr
 		goto nlmsg_failure;
 
 	nlmsg_end(skb, nlh);
-out:
-	sp->this_idx++;
 	return 0;
 
 nlmsg_failure:
@@ -1259,21 +1260,33 @@ nlmsg_failure:
 	return -EMSGSIZE;
 }
 
+static int xfrm_dump_policy_done(struct netlink_callback *cb)
+{
+	struct xfrm_policy_walk *walk = (struct xfrm_policy_walk *) &cb->args[1];
+
+	xfrm_policy_walk_done(walk);
+	return 0;
+}
+
 static int xfrm_dump_policy(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	struct xfrm_policy_walk *walk = (struct xfrm_policy_walk *) &cb->args[1];
 	struct xfrm_dump_info info;
 
+	BUILD_BUG_ON(sizeof(struct xfrm_policy_walk) >
+		     sizeof(cb->args) - sizeof(cb->args[0]));
+
 	info.in_skb = cb->skb;
 	info.out_skb = skb;
 	info.nlmsg_seq = cb->nlh->nlmsg_seq;
 	info.nlmsg_flags = NLM_F_MULTI;
-	info.this_idx = 0;
-	info.start_idx = cb->args[0];
-	(void) xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, dump_one_policy, &info);
-#ifdef CONFIG_XFRM_SUB_POLICY
-	(void) xfrm_policy_walk(XFRM_POLICY_TYPE_SUB, dump_one_policy, &info);
-#endif
-	cb->args[0] = info.this_idx;
+
+	if (!cb->args[0]) {
+		cb->args[0] = 1;
+		xfrm_policy_walk_init(walk, XFRM_POLICY_TYPE_ANY);
+	}
+
+	(void) xfrm_policy_walk(walk, dump_one_policy, &info);
 
 	return skb->len;
 }
@@ -1293,7 +1306,6 @@ static struct sk_buff *xfrm_policy_netlink(struct sk_buff *in_skb,
 	info.out_skb = skb;
 	info.nlmsg_seq = seq;
 	info.nlmsg_flags = 0;
-	info.this_idx = info.start_idx = 0;
 
 	if (dump_one_policy(xp, dir, 0, &info) < 0) {
 		kfree_skb(skb);
@@ -1891,15 +1903,18 @@ static const struct nla_policy xfrma_policy[XFRMA_MAX+1] = {
 static struct xfrm_link {
 	int (*doit)(struct sk_buff *, struct nlmsghdr *, struct nlattr **);
 	int (*dump)(struct sk_buff *, struct netlink_callback *);
+	int (*done)(struct netlink_callback *);
 } xfrm_dispatch[XFRM_NR_MSGTYPES] = {
 	[XFRM_MSG_NEWSA       - XFRM_MSG_BASE] = { .doit = xfrm_add_sa        },
 	[XFRM_MSG_DELSA       - XFRM_MSG_BASE] = { .doit = xfrm_del_sa        },
 	[XFRM_MSG_GETSA       - XFRM_MSG_BASE] = { .doit = xfrm_get_sa,
-						   .dump = xfrm_dump_sa       },
+						   .dump = xfrm_dump_sa,
+						   .done = xfrm_dump_sa_done  },
 	[XFRM_MSG_NEWPOLICY   - XFRM_MSG_BASE] = { .doit = xfrm_add_policy    },
 	[XFRM_MSG_DELPOLICY   - XFRM_MSG_BASE] = { .doit = xfrm_get_policy    },
 	[XFRM_MSG_GETPOLICY   - XFRM_MSG_BASE] = { .doit = xfrm_get_policy,
-						   .dump = xfrm_dump_policy   },
+						   .dump = xfrm_dump_policy,
+						   .done = xfrm_dump_policy_done },
 	[XFRM_MSG_ALLOCSPI    - XFRM_MSG_BASE] = { .doit = xfrm_alloc_userspi },
 	[XFRM_MSG_ACQUIRE     - XFRM_MSG_BASE] = { .doit = xfrm_add_acquire   },
 	[XFRM_MSG_EXPIRE      - XFRM_MSG_BASE] = { .doit = xfrm_add_sa_expire },
@@ -1938,7 +1953,7 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		if (link->dump == NULL)
 			return -EINVAL;
 
-		return netlink_dump_start(xfrm_nl, skb, nlh, link->dump, NULL);
+		return netlink_dump_start(xfrm_nl, skb, nlh, link->dump, link->done);
 	}
 
 	err = nlmsg_parse(nlh, xfrm_msg_min[type], attrs, XFRMA_MAX,
-- 
cgit v1.2.3


From 9b0f976f27f00a81cf47643d90854659626795b4 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 29 Feb 2008 11:13:15 -0800
Subject: [INET]: Remove struct net_proto_family* from _init calls.

struct net_proto_family* is not used in icmp[v6]_init, ndisc_init,
igmp_init and tcp_v4_init. Remove it.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/icmpv6.h | 2 +-
 include/net/icmp.h     | 2 +-
 include/net/ndisc.h    | 4 ++--
 include/net/tcp.h      | 2 +-
 net/ipv4/af_inet.c     | 4 ++--
 net/ipv4/icmp.c        | 2 +-
 net/ipv4/tcp_ipv4.c    | 2 +-
 net/ipv6/af_inet6.c    | 6 +++---
 net/ipv6/icmp.c        | 2 +-
 net/ipv6/mcast.c       | 2 +-
 net/ipv6/ndisc.c       | 2 +-
 11 files changed, 15 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h
index 7c5e9817e998..8f86d6b621c8 100644
--- a/include/linux/icmpv6.h
+++ b/include/linux/icmpv6.h
@@ -176,7 +176,7 @@ extern void				icmpv6_send(struct sk_buff *skb,
 						    __u32 info, 
 						    struct net_device *dev);
 
-extern int				icmpv6_init(struct net_proto_family *ops);
+extern int				icmpv6_init(void);
 extern int				icmpv6_err_convert(int type, int code,
 							   int *err);
 extern void				icmpv6_cleanup(void);
diff --git a/include/net/icmp.h b/include/net/icmp.h
index 9f7ef3c8baef..7bf714d9d7c7 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -48,7 +48,7 @@ struct sk_buff;
 extern void	icmp_send(struct sk_buff *skb_in,  int type, int code, __be32 info);
 extern int	icmp_rcv(struct sk_buff *skb);
 extern int	icmp_ioctl(struct sock *sk, int cmd, unsigned long arg);
-extern void	icmp_init(struct net_proto_family *ops);
+extern void	icmp_init(void);
 extern void	icmp_out_count(unsigned char type);
 
 /* Move into dst.h ? */
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index 59b70624b056..5aedf324de66 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -77,7 +77,7 @@ struct nd_opt_hdr {
 } __attribute__((__packed__));
 
 
-extern int			ndisc_init(struct net_proto_family *ops);
+extern int			ndisc_init(void);
 
 extern void			ndisc_cleanup(void);
 
@@ -107,7 +107,7 @@ extern int			ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *d
 /*
  *	IGMP
  */
-extern int			igmp6_init(struct net_proto_family *ops);
+extern int			igmp6_init(void);
 
 extern void			igmp6_cleanup(void);
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 7de4ea3a04d9..ae9774b478f5 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1373,7 +1373,7 @@ struct tcp_request_sock_ops {
 #endif
 };
 
-extern void tcp_v4_init(struct net_proto_family *ops);
+extern void tcp_v4_init(void);
 extern void tcp_init(void);
 
 #endif	/* _TCP_H */
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index c270080f370e..a7a99ac856dc 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1415,7 +1415,7 @@ static int __init inet_init(void)
 
 	ip_init();
 
-	tcp_v4_init(&inet_family_ops);
+	tcp_v4_init();
 
 	/* Setup TCP slab cache for open requests. */
 	tcp_init();
@@ -1430,7 +1430,7 @@ static int __init inet_init(void)
 	 *	Set the ICMP layer up
 	 */
 
-	icmp_init(&inet_family_ops);
+	icmp_init();
 
 	/*
 	 *	Initialise the multicast router
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index a13c074dac09..98372db66c66 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -1139,7 +1139,7 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = {
 	},
 };
 
-void __init icmp_init(struct net_proto_family *ops)
+void __init icmp_init(void)
 {
 	struct inet_sock *inet;
 	int i;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 00156bf421ca..256032a41069 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2443,7 +2443,7 @@ struct proto tcp_prot = {
 	REF_PROTO_INUSE(tcp)
 };
 
-void __init tcp_v4_init(struct net_proto_family *ops)
+void __init tcp_v4_init(void)
 {
 	if (inet_csk_ctl_sock_create(&tcp_socket, PF_INET, SOCK_RAW,
 				     IPPROTO_TCP) < 0)
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index f0aa97738746..9869f87243cf 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -808,13 +808,13 @@ static int __init inet6_init(void)
 	if (err)
 		goto sysctl_fail;
 #endif
-	err = icmpv6_init(&inet6_family_ops);
+	err = icmpv6_init();
 	if (err)
 		goto icmp_fail;
-	err = ndisc_init(&inet6_family_ops);
+	err = ndisc_init();
 	if (err)
 		goto ndisc_fail;
-	err = igmp6_init(&inet6_family_ops);
+	err = igmp6_init();
 	if (err)
 		goto igmp_fail;
 	err = ipv6_netfilter_init();
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 121d517bf91c..b9b13a77ba30 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -780,7 +780,7 @@ drop_no_count:
  */
 static struct lock_class_key icmpv6_socket_sk_dst_lock_key;
 
-int __init icmpv6_init(struct net_proto_family *ops)
+int __init icmpv6_init(void)
 {
 	struct sock *sk;
 	int err, i, j;
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index ab228d1ea114..8ce894d90063 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -2597,7 +2597,7 @@ static const struct file_operations igmp6_mcf_seq_fops = {
 };
 #endif
 
-int __init igmp6_init(struct net_proto_family *ops)
+int __init igmp6_init(void)
 {
 	struct ipv6_pinfo *np;
 	struct sock *sk;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 0d33a7d32125..1fc33c8c7232 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1733,7 +1733,7 @@ static int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name,
 
 #endif
 
-int __init ndisc_init(struct net_proto_family *ops)
+int __init ndisc_init(void)
 {
 	struct ipv6_pinfo *np;
 	struct sock *sk;
-- 
cgit v1.2.3


From a5710d6582868a96cf0fe02eac11d34cfbc783c9 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 29 Feb 2008 11:14:50 -0800
Subject: [ICMP]: Add return code to icmp_init.

icmp_init could fail and this is normal for namespace other than initial.
So, the panic should be triggered only on init_net initialization path.

Additionally create rollback path for icmp_init as a separate function.
It will also be used later during namespace destruction.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h |  2 +-
 net/ipv4/af_inet.c |  3 ++-
 net/ipv4/icmp.c    | 26 ++++++++++++++++++++++----
 3 files changed, 25 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index 7bf714d9d7c7..faba64db8ff0 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -48,7 +48,7 @@ struct sk_buff;
 extern void	icmp_send(struct sk_buff *skb_in,  int type, int code, __be32 info);
 extern int	icmp_rcv(struct sk_buff *skb);
 extern int	icmp_ioctl(struct sock *sk, int cmd, unsigned long arg);
-extern void	icmp_init(void);
+extern int	icmp_init(void);
 extern void	icmp_out_count(unsigned char type);
 
 /* Move into dst.h ? */
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index a7a99ac856dc..4f539bd48718 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1430,7 +1430,8 @@ static int __init inet_init(void)
 	 *	Set the ICMP layer up
 	 */
 
-	icmp_init();
+	if (icmp_init() < 0)
+		panic("Failed to create the ICMP control socket.\n");
 
 	/*
 	 *	Initialise the multicast router
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 98372db66c66..b345b3d9bcaa 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -1139,19 +1139,32 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = {
 	},
 };
 
-void __init icmp_init(void)
+static void __exit icmp_exit(void)
 {
-	struct inet_sock *inet;
 	int i;
 
 	for_each_possible_cpu(i) {
-		int err;
+		struct socket *sock;
+
+		sock = per_cpu(__icmp_socket, i);
+		if (sock == NULL)
+			continue;
+		per_cpu(__icmp_socket, i) = NULL;
+		sock_release(sock);
+	}
+}
 
+int __init icmp_init(void)
+{
+	struct inet_sock *inet;
+	int i, err;
+
+	for_each_possible_cpu(i) {
 		err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_ICMP,
 				       &per_cpu(__icmp_socket, i));
 
 		if (err < 0)
-			panic("Failed to create the ICMP control socket.\n");
+			goto fail;
 
 		per_cpu(__icmp_socket, i)->sk->sk_allocation = GFP_ATOMIC;
 
@@ -1171,6 +1184,11 @@ void __init icmp_init(void)
 		 */
 		per_cpu(__icmp_socket, i)->sk->sk_prot->unhash(per_cpu(__icmp_socket, i)->sk);
 	}
+	return 0;
+
+fail:
+	icmp_exit();
+	return err;
 }
 
 EXPORT_SYMBOL(icmp_err_convert);
-- 
cgit v1.2.3


From edf0208702007ec1f6a36756fdd005f771a4cf17 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 29 Feb 2008 11:18:32 -0800
Subject: [NET]: Make netlink_kernel_release publically available as
 sk_release_kernel.

This staff will be needed for non-netlink kernel sockets, which should
also not pin a namespace like tcp_socket and icmp_socket.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h       | 13 +++++++++++++
 net/core/sock.c          | 18 ++++++++++++++++++
 net/netlink/af_netlink.c | 18 ++----------------
 3 files changed, 33 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index fd9876087651..39112e75411c 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -850,6 +850,7 @@ extern struct sock		*sk_alloc(struct net *net, int family,
 					  gfp_t priority,
 					  struct proto *prot);
 extern void			sk_free(struct sock *sk);
+extern void			sk_release_kernel(struct sock *sk);
 extern struct sock		*sk_clone(const struct sock *sk,
 					  const gfp_t priority);
 
@@ -1333,6 +1334,18 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_e
 }
 #endif
 
+/*
+ * Kernel sockets, f.e. rtnl or icmp_socket, are a part of a namespace.
+ * They should not hold a referrence to a namespace in order to allow
+ * to stop it.
+ * Sockets after sk_change_net should be released using sk_release_kernel
+ */
+static inline void sk_change_net(struct sock *sk, struct net *net)
+{
+	put_net(sk->sk_net);
+	sk->sk_net = net;
+}
+
 extern void sock_enable_timestamp(struct sock *sk);
 extern int sock_get_timestamp(struct sock *, struct timeval __user *);
 extern int sock_get_timestampns(struct sock *, struct timespec __user *);
diff --git a/net/core/sock.c b/net/core/sock.c
index 09cb3a74de7f..c71b645a78f0 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -987,6 +987,24 @@ void sk_free(struct sock *sk)
 	sk_prot_free(sk->sk_prot_creator, sk);
 }
 
+/*
+ * Last sock_put should drop referrence to sk->sk_net. It has already
+ * been dropped in sk_change_net. Taking referrence to stopping namespace
+ * is not an option.
+ * Take referrence to a socket to remove it from hash _alive_ and after that
+ * destroy it in the context of init_net.
+ */
+void sk_release_kernel(struct sock *sk)
+{
+	if (sk == NULL || sk->sk_socket == NULL)
+		return;
+
+	sock_hold(sk);
+	sock_release(sk->sk_socket);
+	sk->sk_net = get_net(&init_net);
+	sock_put(sk);
+}
+
 struct sock *sk_clone(const struct sock *sk, const gfp_t priority)
 {
 	struct sock *newsk;
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index e6b636d53633..524e826bb976 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1372,8 +1372,7 @@ netlink_kernel_create(struct net *net, int unit, unsigned int groups,
 		goto out_sock_release_nosk;
 
 	sk = sock->sk;
-	put_net(sk->sk_net);
-	sk->sk_net = net;
+	sk_change_net(sk, net);
 
 	if (groups < 32)
 		groups = 32;
@@ -1421,20 +1420,7 @@ EXPORT_SYMBOL(netlink_kernel_create);
 void
 netlink_kernel_release(struct sock *sk)
 {
-	/*
-	 * Last sock_put should drop referrence to sk->sk_net. It has already
-	 * been dropped in netlink_kernel_create. Taking referrence to stopping
-	 * namespace is not an option.
-	 * Take referrence to a socket to remove it from netlink lookup table
-	 * _alive_ and after that destroy it in the context of init_net.
-	 */
-	if (sk == NULL || sk->sk_socket == NULL)
-		return;
-
-	sock_hold(sk);
-	sock_release(sk->sk_socket);
-	sk->sk_net = get_net(&init_net);
-	sock_put(sk);
+	sk_release_kernel(sk);
 }
 EXPORT_SYMBOL(netlink_kernel_release);
 
-- 
cgit v1.2.3


From 4a6ad7a141cbee2cf074e6cf8dc527b231b69ece Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 29 Feb 2008 11:19:58 -0800
Subject: [NETNS]: Make icmp_sk per namespace.

All preparations are done. Now just add a hook to perform an
initialization on namespace startup and replace icmp_sk macro with
proper inline call.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  2 ++
 net/ipv4/icmp.c          | 49 +++++++++++++++++++++++++++++++-----------------
 2 files changed, 34 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index a9b4f6086294..504fde174525 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -26,6 +26,8 @@ struct netns_ipv4 {
 	struct hlist_head	*fib_table_hash;
 	struct sock		*fibnl;
 
+	struct sock		**icmp_sk;
+
 	struct netns_frags	frags;
 #ifdef CONFIG_NETFILTER
 	struct xt_table		*iptable_filter;
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 97d97ada4ce6..b51f4b0a3264 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -229,8 +229,10 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES+1];
  *
  *	On SMP we have one ICMP socket per-cpu.
  */
-static struct sock **__icmp_sk = NULL;
-#define icmp_sk		(__icmp_sk[smp_processor_id()])
+static struct sock *icmp_sk(struct net *net)
+{
+	return net->ipv4.icmp_sk[smp_processor_id()];
+}
 
 static inline int icmp_xmit_lock(struct sock *sk)
 {
@@ -349,7 +351,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
 	struct sock *sk;
 	struct sk_buff *skb;
 
-	sk = icmp_sk;
+	sk = icmp_sk(rt->u.dst.dev->nd_net);
 	if (ip_append_data(sk, icmp_glue_bits, icmp_param,
 			   icmp_param->data_len+icmp_param->head_len,
 			   icmp_param->head_len,
@@ -378,10 +380,11 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
 
 static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 {
-	struct sock *sk = icmp_sk;
-	struct inet_sock *inet = inet_sk(sk);
 	struct ipcm_cookie ipc;
 	struct rtable *rt = (struct rtable *)skb->dst;
+	struct net *net = rt->u.dst.dev->nd_net;
+	struct sock *sk = icmp_sk(net);
+	struct inet_sock *inet = inet_sk(sk);
 	__be32 daddr;
 
 	if (ip_options_echo(&icmp_param->replyopts, skb))
@@ -407,7 +410,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 						.tos = RT_TOS(ip_hdr(skb)->tos) } },
 				    .proto = IPPROTO_ICMP };
 		security_skb_classify_flow(skb, &fl);
-		if (ip_route_output_key(rt->u.dst.dev->nd_net, &rt, &fl))
+		if (ip_route_output_key(net, &rt, &fl))
 			goto out_unlock;
 	}
 	if (icmpv4_xrlim_allow(rt, icmp_param->data.icmph.type,
@@ -440,11 +443,12 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 	__be32 saddr;
 	u8  tos;
 	struct net *net;
-	struct sock *sk = icmp_sk;
+	struct sock *sk;
 
 	if (!rt)
 		goto out;
 	net = rt->u.dst.dev->nd_net;
+	sk = icmp_sk(net);
 
 	/*
 	 *	Find the original header. It is expected to be valid, of course.
@@ -1142,22 +1146,23 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = {
 	},
 };
 
-static void __exit icmp_exit(void)
+static void __net_exit icmp_sk_exit(struct net *net)
 {
 	int i;
 
 	for_each_possible_cpu(i)
-		sk_release_kernel(__icmp_sk[i]);
-	kfree(__icmp_sk);
-	__icmp_sk = NULL;
+		sk_release_kernel(net->ipv4.icmp_sk[i]);
+	kfree(net->ipv4.icmp_sk);
+	net->ipv4.icmp_sk = NULL;
 }
 
-int __init icmp_init(void)
+int __net_init icmp_sk_init(struct net *net)
 {
 	int i, err;
 
-	__icmp_sk = kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL);
-	if (__icmp_sk == NULL)
+	net->ipv4.icmp_sk =
+		kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL);
+	if (net->ipv4.icmp_sk == NULL)
 		return -ENOMEM;
 
 	for_each_possible_cpu(i) {
@@ -1169,8 +1174,8 @@ int __init icmp_init(void)
 		if (err < 0)
 			goto fail;
 
-		__icmp_sk[i] = sk = sock->sk;
-		sk_change_net(sk, &init_net);
+		net->ipv4.icmp_sk[i] = sk = sock->sk;
+		sk_change_net(sk, net);
 
 		sk->sk_allocation = GFP_ATOMIC;
 
@@ -1193,10 +1198,20 @@ int __init icmp_init(void)
 	return 0;
 
 fail:
-	icmp_exit();
+	icmp_sk_exit(net);
 	return err;
 }
 
+static struct pernet_operations __net_initdata icmp_sk_ops = {
+       .init = icmp_sk_init,
+       .exit = icmp_sk_exit,
+};
+
+int __init icmp_init(void)
+{
+	return register_pernet_device(&icmp_sk_ops);
+}
+
 EXPORT_SYMBOL(icmp_err_convert);
 EXPORT_SYMBOL(icmp_send);
 EXPORT_SYMBOL(icmp_statistics);
-- 
cgit v1.2.3


From 98c6d1b261e74750e6c56ede26dc295201c4a8b2 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 29 Feb 2008 11:21:22 -0800
Subject: [NETNS]: Make icmpv6_sk per namespace.

All preparations are done. Now just add a hook to perform an
initialization on namespace startup and replace icmpv6_sk macro with
proper inline call.  Actual namespace the packet belongs too will be
passed later along with the one for the routing.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |  1 +
 net/ipv6/icmp.c          | 68 +++++++++++++++++++++++++++++++++---------------
 2 files changed, 48 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 1dd7de4e4195..82623d3a8e35 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -36,5 +36,6 @@ struct netns_ipv6 {
 	struct xt_table		*ip6table_mangle;
 	struct xt_table		*ip6table_raw;
 #endif
+	struct sock		**icmp_sk;
 };
 #endif
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 3eb594d54172..9f55a965c884 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -80,8 +80,10 @@ EXPORT_SYMBOL(icmpv6msg_statistics);
  *
  *	On SMP we have one ICMP socket per-cpu.
  */
-static struct sock **__icmpv6_sk = NULL;
-#define icmpv6_sk	(__icmpv6_sk[smp_processor_id()])
+static inline struct sock *icmpv6_sk(struct net *net)
+{
+	return net->ipv6.icmp_sk[smp_processor_id()];
+}
 
 static int icmpv6_rcv(struct sk_buff *skb);
 
@@ -389,7 +391,7 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 	fl.fl_icmp_code = code;
 	security_skb_classify_flow(skb, &fl);
 
-	sk = icmpv6_sk;
+	sk = icmpv6_sk(&init_net);
 	np = inet6_sk(sk);
 
 	if (icmpv6_xmit_lock(sk))
@@ -535,7 +537,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
 	fl.fl_icmp_type = ICMPV6_ECHO_REPLY;
 	security_skb_classify_flow(skb, &fl);
 
-	sk = icmpv6_sk;
+	sk = icmpv6_sk(&init_net);
 	np = inet6_sk(sk);
 
 	if (icmpv6_xmit_lock(sk))
@@ -780,13 +782,14 @@ drop_no_count:
  */
 static struct lock_class_key icmpv6_socket_sk_dst_lock_key;
 
-int __init icmpv6_init(void)
+static int __net_init icmpv6_sk_init(struct net *net)
 {
 	struct sock *sk;
 	int err, i, j;
 
-	__icmpv6_sk = kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL);
-	if (__icmpv6_sk == NULL)
+	net->ipv6.icmp_sk =
+		kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL);
+	if (net->ipv6.icmp_sk == NULL)
 		return -ENOMEM;
 
 	for_each_possible_cpu(i) {
@@ -801,8 +804,8 @@ int __init icmpv6_init(void)
 			goto fail;
 		}
 
-		__icmpv6_sk[i] = sk = sock->sk;
-		sk_change_net(sk, &init_net);
+		net->ipv6.icmp_sk[i] = sk = sock->sk;
+		sk_change_net(sk, net);
 
 		sk->sk_allocation = GFP_ATOMIC;
 		/*
@@ -822,33 +825,56 @@ int __init icmpv6_init(void)
 
 		sk->sk_prot->unhash(sk);
 	}
-
-
-	if (inet6_add_protocol(&icmpv6_protocol, IPPROTO_ICMPV6) < 0) {
-		printk(KERN_ERR "Failed to register ICMP6 protocol\n");
-		err = -EAGAIN;
-		goto fail;
-	}
-
 	return 0;
 
  fail:
 	for (j = 0; j < i; j++)
-		sk_release_kernel(__icmpv6_sk[j]);
-
+		sk_release_kernel(net->ipv6.icmp_sk[j]);
+	kfree(net->ipv6.icmp_sk);
 	return err;
 }
 
-void icmpv6_cleanup(void)
+static void __net_exit icmpv6_sk_exit(struct net *net)
 {
 	int i;
 
 	for_each_possible_cpu(i) {
-		sk_release_kernel(__icmpv6_sk[i]);
+		sk_release_kernel(net->ipv6.icmp_sk[i]);
 	}
+	kfree(net->ipv6.icmp_sk);
+}
+
+static struct pernet_operations __net_initdata icmpv6_sk_ops = {
+       .init = icmpv6_sk_init,
+       .exit = icmpv6_sk_exit,
+};
+
+int __init icmpv6_init(void)
+{
+	int err;
+
+	err = register_pernet_subsys(&icmpv6_sk_ops);
+	if (err < 0)
+		return err;
+
+	err = -EAGAIN;
+	if (inet6_add_protocol(&icmpv6_protocol, IPPROTO_ICMPV6) < 0)
+		goto fail;
+	return 0;
+
+fail:
+	printk(KERN_ERR "Failed to register ICMP6 protocol\n");
+	unregister_pernet_subsys(&icmpv6_sk_ops);
+	return err;
+}
+
+void __exit icmpv6_cleanup(void)
+{
+	unregister_pernet_subsys(&icmpv6_sk_ops);
 	inet6_del_protocol(&icmpv6_protocol, IPPROTO_ICMPV6);
 }
 
+
 static const struct icmp6_err {
 	int err;
 	int fatal;
-- 
cgit v1.2.3


From 58fbbed4fbc0094fc808a568fe99a915f85402ee Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Fri, 29 Feb 2008 11:40:56 -0800
Subject: [SCTP]: extend exported data in /proc/net/sctp/assoc

RFC 3873 specifies several MIB objects that can't be obtained by the
current data set exported by /proc/sys/net/sctp/assoc.  This patch
adds the missing pieces of data that allow us to compute all the
objects in the sctpAssocTable object.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  8 ++++++++
 net/sctp/outqueue.c        |  3 +++
 net/sctp/proc.c            | 20 ++++++++++++++------
 net/sctp/sm_statefuns.c    |  2 ++
 4 files changed, 27 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 9c827a749b6f..8966599ddb9f 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1661,6 +1661,9 @@ struct sctp_association {
 	/* Transport to which SHUTDOWN chunk was last sent.  */
 	struct sctp_transport *shutdown_last_sent_to;
 
+	/* How many times have we resent a SHUTDOWN */
+	int shutdown_retries;
+
 	/* Transport to which INIT chunk was last sent.  */
 	struct sctp_transport *init_last_sent_to;
 
@@ -1695,6 +1698,11 @@ struct sctp_association {
 	 */
 	__u16 unack_data;
 
+	/* The total number of data chunks that we've had to retransmit
+	 * as the result of a T3 timer expiration
+	 */
+	__u32 rtx_data_chunks;
+
 	/* This is the association's receive buffer space.  This value is used
 	 * to set a_rwnd field in an INIT or a SACK chunk.
 	 */
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 1bb3c5c35d2a..fd4deefab3cf 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -494,6 +494,8 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport,
 		 */
 		if (transport == transport->asoc->peer.retran_path)
 			sctp_assoc_update_retran_path(transport->asoc);
+		transport->asoc->rtx_data_chunks +=
+			transport->asoc->unack_data;
 		break;
 	case SCTP_RTXR_FAST_RTX:
 		SCTP_INC_STATS(SCTP_MIB_FAST_RETRANSMITS);
@@ -504,6 +506,7 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport,
 		break;
 	case SCTP_RTXR_T1_RTX:
 		SCTP_INC_STATS(SCTP_MIB_T1_RETRANSMITS);
+		transport->asoc->init_retries++;
 		break;
 	default:
 		BUG();
diff --git a/net/sctp/proc.c b/net/sctp/proc.c
index 9e214da82d9e..82bea300308d 100644
--- a/net/sctp/proc.c
+++ b/net/sctp/proc.c
@@ -281,8 +281,10 @@ static void * sctp_assocs_seq_start(struct seq_file *seq, loff_t *pos)
 		*pos = 0;
 
 	if (*pos == 0)
-		seq_printf(seq, " ASSOC     SOCK   STY SST ST HBKT ASSOC-ID TX_QUEUE RX_QUEUE UID INODE LPORT "
-				"RPORT LADDRS <-> RADDRS\n");
+		seq_printf(seq, " ASSOC     SOCK   STY SST ST HBKT "
+				"ASSOC-ID TX_QUEUE RX_QUEUE UID INODE LPORT "
+				"RPORT LADDRS <-> RADDRS "
+				"HBINT INS OUTS MAXRT T1X T2X RTXC\n");
 
 	return (void *)pos;
 }
@@ -321,15 +323,21 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v)
 		assoc = sctp_assoc(epb);
 		sk = epb->sk;
 		seq_printf(seq,
-			   "%8p %8p %-3d %-3d %-2d %-4d %4d %8d %8d %7d %5lu %-5d %5d ",
+			   "%8p %8p %-3d %-3d %-2d %-4d "
+			   "%4d %8d %8d %7d %5lu %-5d %5d "
+			   "%8d %5d %5d %4d %4d %4d %8d ",
 			   assoc, sk, sctp_sk(sk)->type, sk->sk_state,
-			   assoc->state, hash, assoc->assoc_id,
+			   assoc->state, hash,
+			   assoc->assoc_id,
 			   assoc->sndbuf_used,
 			   atomic_read(&assoc->rmem_alloc),
 			   sock_i_uid(sk), sock_i_ino(sk),
 			   epb->bind_addr.port,
-			   assoc->peer.port);
-
+			   assoc->peer.port,
+			   assoc->hbinterval, assoc->c.sinit_max_instreams,
+			   assoc->c.sinit_num_ostreams, assoc->max_retrans,
+			   assoc->init_retries, assoc->shutdown_retries,
+			   assoc->rtx_data_chunks);
 		seq_printf(seq, " ");
 		sctp_seq_dump_local_addrs(seq, epb);
 		seq_printf(seq, "<-> ");
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index f2ed6473feef..ade0cbd3a52b 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -5312,6 +5312,8 @@ sctp_disposition_t sctp_sf_t2_timer_expire(const struct sctp_endpoint *ep,
 	SCTP_DEBUG_PRINTK("Timer T2 expired.\n");
 	SCTP_INC_STATS(SCTP_MIB_T2_SHUTDOWN_EXPIREDS);
 
+	((struct sctp_association *)asoc)->shutdown_retries++;
+
 	if (asoc->overall_error_count >= asoc->max_retrans) {
 		sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR,
 				SCTP_ERROR(ETIMEDOUT));
-- 
cgit v1.2.3


From fd80eb942ad9761f241c9b287b3b9a342b20690d Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 29 Feb 2008 11:43:03 -0800
Subject: [INET]: Remove struct dst_entry *dst from
 request_sock_ops.rtx_syn_ack.

It looks like dst parameter is used in this API due to historical
reasons.  Actually, it is really used in the direct call to
tcp_v4_send_synack only.  So, create a wrapper for tcp_v4_send_synack
and remove dst from rtx_syn_ack.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/request_sock.h      |  3 +--
 net/dccp/ipv4.c                 | 11 +++++------
 net/dccp/ipv6.c                 | 38 ++++++++++++++++++--------------------
 net/dccp/minisocks.c            |  2 +-
 net/ipv4/inet_connection_sock.c |  2 +-
 net/ipv4/tcp_ipv4.c             | 14 +++++++++-----
 net/ipv4/tcp_minisocks.c        |  2 +-
 net/ipv6/tcp_ipv6.c             | 36 +++++++++++++++++-------------------
 8 files changed, 53 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index cff4608179c1..040780add355 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -31,8 +31,7 @@ struct request_sock_ops {
 	int		obj_size;
 	struct kmem_cache	*slab;
 	int		(*rtx_syn_ack)(struct sock *sk,
-				       struct request_sock *req,
-				       struct dst_entry *dst);
+				       struct request_sock *req);
 	void		(*send_ack)(struct sk_buff *skb,
 				    struct request_sock *req);
 	void		(*send_reset)(struct sock *sk,
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 474075adbde4..514a40b7fc7f 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -471,15 +471,14 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk,
 	return &rt->u.dst;
 }
 
-static int dccp_v4_send_response(struct sock *sk, struct request_sock *req,
-				 struct dst_entry *dst)
+static int dccp_v4_send_response(struct sock *sk, struct request_sock *req)
 {
 	int err = -1;
 	struct sk_buff *skb;
+	struct dst_entry *dst;
 
-	/* First, grab a route. */
-
-	if (dst == NULL && (dst = inet_csk_route_req(sk, req)) == NULL)
+	dst = inet_csk_route_req(sk, req);
+	if (dst == NULL)
 		goto out;
 
 	skb = dccp_make_response(sk, dst, req);
@@ -620,7 +619,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	dreq->dreq_iss	   = dccp_v4_init_sequence(skb);
 	dreq->dreq_service = service;
 
-	if (dccp_v4_send_response(sk, req, NULL))
+	if (dccp_v4_send_response(sk, req))
 		goto drop_and_free;
 
 	inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 490333d47c7b..1a5e50b90677 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -224,8 +224,7 @@ out:
 }
 
 
-static int dccp_v6_send_response(struct sock *sk, struct request_sock *req,
-				 struct dst_entry *dst)
+static int dccp_v6_send_response(struct sock *sk, struct request_sock *req)
 {
 	struct inet6_request_sock *ireq6 = inet6_rsk(req);
 	struct ipv6_pinfo *np = inet6_sk(sk);
@@ -234,6 +233,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req,
 	struct in6_addr *final_p = NULL, final;
 	struct flowi fl;
 	int err = -1;
+	struct dst_entry *dst;
 
 	memset(&fl, 0, sizeof(fl));
 	fl.proto = IPPROTO_DCCP;
@@ -245,28 +245,26 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req,
 	fl.fl_ip_sport = inet_sk(sk)->sport;
 	security_req_classify_flow(req, &fl);
 
-	if (dst == NULL) {
-		opt = np->opt;
+	opt = np->opt;
 
-		if (opt != NULL && opt->srcrt != NULL) {
-			const struct rt0_hdr *rt0 = (struct rt0_hdr *)opt->srcrt;
+	if (opt != NULL && opt->srcrt != NULL) {
+		const struct rt0_hdr *rt0 = (struct rt0_hdr *)opt->srcrt;
 
-			ipv6_addr_copy(&final, &fl.fl6_dst);
-			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-			final_p = &final;
-		}
+		ipv6_addr_copy(&final, &fl.fl6_dst);
+		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
+		final_p = &final;
+	}
 
-		err = ip6_dst_lookup(sk, &dst, &fl);
-		if (err)
-			goto done;
+	err = ip6_dst_lookup(sk, &dst, &fl);
+	if (err)
+		goto done;
 
-		if (final_p)
-			ipv6_addr_copy(&fl.fl6_dst, final_p);
+	if (final_p)
+		ipv6_addr_copy(&fl.fl6_dst, final_p);
 
-		err = xfrm_lookup(&dst, &fl, sk, 0);
-		if (err < 0)
-			goto done;
-	}
+	err = xfrm_lookup(&dst, &fl, sk, 0);
+	if (err < 0)
+		goto done;
 
 	skb = dccp_make_response(sk, dst, req);
 	if (skb != NULL) {
@@ -448,7 +446,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 	dreq->dreq_iss	   = dccp_v6_init_sequence(skb);
 	dreq->dreq_service = service;
 
-	if (dccp_v6_send_response(sk, req, NULL))
+	if (dccp_v6_send_response(sk, req))
 		goto drop_and_free;
 
 	inet6_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index 027d1814e1ab..33ad48321b08 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -216,7 +216,7 @@ struct sock *dccp_check_req(struct sock *sk, struct sk_buff *skb,
 			 * counter (backoff, monitored by dccp_response_timer).
 			 */
 			req->retrans++;
-			req->rsk_ops->rtx_syn_ack(sk, req, NULL);
+			req->rsk_ops->rtx_syn_ack(sk, req);
 		}
 		/* Network Duplicate, discard packet */
 		return NULL;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index b189278c7bc1..c0e0fa03fce1 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -463,7 +463,7 @@ void inet_csk_reqsk_queue_prune(struct sock *parent,
 			if (time_after_eq(now, req->expires)) {
 				if ((req->retrans < thresh ||
 				     (inet_rsk(req)->acked && req->retrans < max_retries))
-				    && !req->rsk_ops->rtx_syn_ack(parent, req, NULL)) {
+				    && !req->rsk_ops->rtx_syn_ack(parent, req)) {
 					unsigned long timeo;
 
 					if (req->retrans++ == 0)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 256032a41069..3b26f9586dcb 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -723,8 +723,8 @@ static void tcp_v4_reqsk_send_ack(struct sk_buff *skb,
  *	This still operates on a request_sock only, not on a big
  *	socket.
  */
-static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
-			      struct dst_entry *dst)
+static int __tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
+				struct dst_entry *dst)
 {
 	const struct inet_request_sock *ireq = inet_rsk(req);
 	int err = -1;
@@ -732,7 +732,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
 
 	/* First, grab a route. */
 	if (!dst && (dst = inet_csk_route_req(sk, req)) == NULL)
-		goto out;
+		return -1;
 
 	skb = tcp_make_synack(sk, dst, req);
 
@@ -751,11 +751,15 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
 		err = net_xmit_eval(err);
 	}
 
-out:
 	dst_release(dst);
 	return err;
 }
 
+static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req)
+{
+	return __tcp_v4_send_synack(sk, req, NULL);
+}
+
 /*
  *	IPv4 request_sock destructor.
  */
@@ -1380,7 +1384,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	}
 	tcp_rsk(req)->snt_isn = isn;
 
-	if (tcp_v4_send_synack(sk, req, dst))
+	if (__tcp_v4_send_synack(sk, req, dst))
 		goto drop_and_free;
 
 	if (want_cookie) {
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index b61b76847ad9..0fdd1db641ac 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -536,7 +536,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
 		 * Enforce "SYN-ACK" according to figure 8, figure 6
 		 * of RFC793, fixed by RFC1122.
 		 */
-		req->rsk_ops->rtx_syn_ack(sk, req, NULL);
+		req->rsk_ops->rtx_syn_ack(sk, req);
 		return NULL;
 	}
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 12750f2b05ab..1cbbb87dbad2 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -455,8 +455,7 @@ out:
 }
 
 
-static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
-			      struct dst_entry *dst)
+static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req)
 {
 	struct inet6_request_sock *treq = inet6_rsk(req);
 	struct ipv6_pinfo *np = inet6_sk(sk);
@@ -464,6 +463,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
 	struct ipv6_txoptions *opt = NULL;
 	struct in6_addr * final_p = NULL, final;
 	struct flowi fl;
+	struct dst_entry *dst;
 	int err = -1;
 
 	memset(&fl, 0, sizeof(fl));
@@ -476,24 +476,22 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
 	fl.fl_ip_sport = inet_sk(sk)->sport;
 	security_req_classify_flow(req, &fl);
 
-	if (dst == NULL) {
-		opt = np->opt;
-		if (opt && opt->srcrt) {
-			struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
-			ipv6_addr_copy(&final, &fl.fl6_dst);
-			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-			final_p = &final;
-		}
-
-		err = ip6_dst_lookup(sk, &dst, &fl);
-		if (err)
-			goto done;
-		if (final_p)
-			ipv6_addr_copy(&fl.fl6_dst, final_p);
-		if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0)
-			goto done;
+	opt = np->opt;
+	if (opt && opt->srcrt) {
+		struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
+		ipv6_addr_copy(&final, &fl.fl6_dst);
+		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
+		final_p = &final;
 	}
 
+	err = ip6_dst_lookup(sk, &dst, &fl);
+	if (err)
+		goto done;
+	if (final_p)
+		ipv6_addr_copy(&fl.fl6_dst, final_p);
+	if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0)
+		goto done;
+
 	skb = tcp_make_synack(sk, dst, req);
 	if (skb) {
 		struct tcphdr *th = tcp_hdr(skb);
@@ -1294,7 +1292,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	security_inet_conn_request(sk, skb, req);
 
-	if (tcp_v6_send_synack(sk, req, NULL))
+	if (tcp_v6_send_synack(sk, req))
 		goto drop;
 
 	inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT);
-- 
cgit v1.2.3


From a90bcbd651b453d8259243115696c11b864b30fe Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Fri, 29 Feb 2008 11:45:34 -0800
Subject: [SCTP]: Kill unused static inline sctp_sysctl_jiffies_ms
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

After the patch:
$ git-grep sctp_sysctl_jiffies_ms | wc -l
0

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Acked-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/sctp.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 57df27f19588..a653eb3e1e70 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -368,11 +368,6 @@ void sctp_sysctl_unregister(void);
 #else
 static inline void sctp_sysctl_register(void) { return; }
 static inline void sctp_sysctl_unregister(void) { return; }
-static inline int sctp_sysctl_jiffies_ms(ctl_table *table, int __user *name, int nlen,
-		void __user *oldval, size_t __user *oldlenp,
-		void __user *newval, size_t newlen) {
-	return -ENOSYS;
-}
 #endif
 
 /* Size of Supported Address Parameter for 'x' address types. */
-- 
cgit v1.2.3


From 03a64c93b68e1eff299b9bbbb0d13105171cddc4 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Fri, 29 Feb 2008 11:46:17 -0800
Subject: [LLC]: Kill static inline llc_addrany
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

After the patch:
$ git-grep llc_addrany | wc -l
0

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/llc_if.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/net/llc_if.h b/include/net/llc_if.h
index c608812a8e89..b595a004d31b 100644
--- a/include/net/llc_if.h
+++ b/include/net/llc_if.h
@@ -74,11 +74,6 @@ static inline int llc_mac_null(const u8 *mac)
 	return is_zero_ether_addr(mac);
 }
 
-static inline int llc_addrany(const struct llc_addr *addr)
-{
-	return llc_mac_null(addr->mac) && !addr->lsap;
-}
-
 static inline int llc_mac_multicast(const u8 *mac)
 {
 	return is_multicast_ether_addr(mac);
-- 
cgit v1.2.3


From 0df3ef45a3d7b59cc53ce4e3611033c6e3b51a1b Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Mon, 28 Jan 2008 14:07:15 +0200
Subject: mac80211: A-MPDU Tx add session's and low level driver's API

This patch adds the API for 3 stages in A-MPDU Tx session flow:
- request mac80211 to start/stop A-MPDU Tx session for specific TID. such a
  request should be issued by a load aware element, either mac80211 itself
  or external element.
- requests by mac80211 to low-level driver to start/stop Tx aggregation.
  notice that low level driver responds now with Starting Sequence Number.
- async feedback by low-level to mac80211 to inform that HW is ready for
  next A-MPDU Tx state.
Changes in API to Rx A-MPDU were also made, reflected in iwlwifi changes as
well.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-4965.c |  4 +-
 drivers/net/wireless/iwlwifi/iwl-4965.h |  2 +-
 include/net/mac80211.h                  | 86 ++++++++++++++++++++++++++++++++-
 net/mac80211/ieee80211_sta.c            |  4 +-
 4 files changed, 89 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index d727de8b96fe..cd4f5e383e60 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -4732,7 +4732,7 @@ static void iwl4965_sta_modify_del_ba_tid(struct iwl4965_priv *priv,
 
 int iwl4965_mac_ampdu_action(struct ieee80211_hw *hw,
 			     enum ieee80211_ampdu_mlme_action action,
-			     const u8 *addr, u16 tid, u16 ssn)
+			     const u8 *addr, u16 tid, u16 *ssn)
 {
 	struct iwl4965_priv *priv = hw->priv;
 	int sta_id;
@@ -4744,7 +4744,7 @@ int iwl4965_mac_ampdu_action(struct ieee80211_hw *hw,
 	switch (action) {
 	case IEEE80211_AMPDU_RX_START:
 		IWL_DEBUG_HT("start Rx\n");
-		iwl4965_sta_modify_add_ba_tid(priv, sta_id, tid, ssn);
+		iwl4965_sta_modify_add_ba_tid(priv, sta_id, tid, *ssn);
 		break;
 	case IEEE80211_AMPDU_RX_STOP:
 		IWL_DEBUG_HT("stop Rx\n");
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.h b/drivers/net/wireless/iwlwifi/iwl-4965.h
index 9cb82be0ff80..6edf869d4d54 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.h
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.h
@@ -777,7 +777,7 @@ extern void iwl4965_set_ht_add_station(struct iwl4965_priv *priv, u8 index,
 				struct ieee80211_ht_info *sta_ht_inf);
 extern int iwl4965_mac_ampdu_action(struct ieee80211_hw *hw,
 				    enum ieee80211_ampdu_mlme_action action,
-				    const u8 *addr, u16 tid, u16 ssn);
+				    const u8 *addr, u16 tid, u16 *ssn);
 #ifdef CONFIG_IWL4965_HT_AGG
 extern int iwl4965_mac_ht_tx_agg_start(struct ieee80211_hw *hw, u8 *da,
 				   u16 tid, u16 *start_seq_num);
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 9083bafb63ca..3bbc33cc2b96 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -967,10 +967,14 @@ enum ieee80211_filter_flags {
  * &struct ieee80211_ops to indicate which action is needed.
  * @IEEE80211_AMPDU_RX_START: start Rx aggregation
  * @IEEE80211_AMPDU_RX_STOP: stop Rx aggregation
+ * @IEEE80211_AMPDU_TX_START: start Tx aggregation
+ * @IEEE80211_AMPDU_TX_STOP: stop Tx aggregation
  */
 enum ieee80211_ampdu_mlme_action {
 	IEEE80211_AMPDU_RX_START,
 	IEEE80211_AMPDU_RX_STOP,
+	IEEE80211_AMPDU_TX_START,
+	IEEE80211_AMPDU_TX_STOP,
 };
 
 /**
@@ -1111,7 +1115,8 @@ enum ieee80211_ampdu_mlme_action {
  * 	The RA/TID combination determines the destination and TID we want
  * 	the ampdu action to be performed for. The action is defined through
  * 	ieee80211_ampdu_mlme_action. Starting sequence number (@ssn)
- * 	is the first frame we expect to perform the action on.
+ * 	is the first frame we expect to perform the action on. notice
+ * 	that TX/RX_STOP can pass NULL for this parameter.
  */
 struct ieee80211_ops {
 	int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb,
@@ -1162,7 +1167,7 @@ struct ieee80211_ops {
 	int (*conf_ht)(struct ieee80211_hw *hw, struct ieee80211_conf *conf);
 	int (*ampdu_action)(struct ieee80211_hw *hw,
 			    enum ieee80211_ampdu_mlme_action action,
-			    const u8 *ra, u16 tid, u16 ssn);
+			    const u8 *addr, u16 tid, u16 *ssn);
 };
 
 /**
@@ -1574,4 +1579,81 @@ void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw,
 						struct ieee80211_vif *vif),
 					 void *data);
 
+/**
+ * ieee80211_start_tx_ba_session - Start a tx Block Ack session.
+ * @hw: pointer as obtained from ieee80211_alloc_hw().
+ * @ra: receiver address of the BA session recipient
+ * @tid: the TID to BA on.
+ * @return: success if addBA request was sent, failure otherwise
+ *
+ * Although mac80211/low level driver/user space application can estimate
+ * the need to start aggregation on a certain RA/TID, the session level
+ * will be managed by the mac80211.
+ */
+int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid);
+
+/**
+ * ieee80211_start_tx_ba_cb - low level driver ready to aggregate.
+ * @hw: pointer as obtained from ieee80211_alloc_hw().
+ * @ra: receiver address of the BA session recipient.
+ * @tid: the TID to BA on.
+ *
+ * This function must be called by low level driver once it has
+ * finished with preparations for the BA session.
+ */
+void ieee80211_start_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u16 tid);
+
+/**
+ * ieee80211_start_tx_ba_cb_irqsafe - low level driver ready to aggregate.
+ * @hw: pointer as obtained from ieee80211_alloc_hw().
+ * @ra: receiver address of the BA session recipient.
+ * @tid: the TID to BA on.
+ *
+ * This function must be called by low level driver once it has
+ * finished with preparations for the BA session.
+ * This version of the function is irq safe.
+ */
+void ieee80211_start_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra,
+				      u16 tid);
+
+/**
+ * ieee80211_stop_tx_ba_session - Stop a Block Ack session.
+ * @hw: pointer as obtained from ieee80211_alloc_hw().
+ * @ra: receiver address of the BA session recipient
+ * @tid: the TID to stop BA.
+ * @initiator: if indicates initiator DELBA frame will be sent.
+ * @return: error if no sta with matching da found, success otherwise
+ *
+ * Although mac80211/low level driver/user space application can estimate
+ * the need to stop aggregation on a certain RA/TID, the session level
+ * will be managed by the mac80211.
+ */
+int ieee80211_stop_tx_ba_session(struct ieee80211_hw *hw,
+				 u8 *ra, u16 tid,
+				 enum ieee80211_back_parties initiator);
+
+/**
+ * ieee80211_stop_tx_ba_cb - low level driver ready to stop aggregate.
+ * @hw: pointer as obtained from ieee80211_alloc_hw().
+ * @ra: receiver address of the BA session recipient.
+ * @tid: the desired TID to BA on.
+ *
+ * This function must be called by low level driver once it has
+ * finished with preparations for the BA session tear down.
+ */
+void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid);
+
+/**
+ * ieee80211_stop_tx_ba_cb_irqsafe - low level driver ready to stop aggregate.
+ * @hw: pointer as obtained from ieee80211_alloc_hw().
+ * @ra: receiver address of the BA session recipient.
+ * @tid: the desired TID to BA on.
+ *
+ * This function must be called by low level driver once it has
+ * finished with preparations for the BA session tear down.
+ * This version of the function is irq safe.
+ */
+void ieee80211_stop_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra,
+				     u16 tid);
+
 #endif /* MAC80211_H */
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index 79dadd2976cd..d5d5610db1bd 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -1128,7 +1128,7 @@ static void ieee80211_sta_process_addba_request(struct net_device *dev,
 
 	if (local->ops->ampdu_action)
 		ret = local->ops->ampdu_action(hw, IEEE80211_AMPDU_RX_START,
-					       sta->addr, tid, start_seq_num);
+					       sta->addr, tid, &start_seq_num);
 #ifdef CONFIG_MAC80211_HT_DEBUG
 	printk(KERN_DEBUG "Rx A-MPDU on tid %d result %d", tid, ret);
 #endif /* CONFIG_MAC80211_HT_DEBUG */
@@ -1230,7 +1230,7 @@ void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *ra, u16 tid,
 	BUG_ON(!local->ops->ampdu_action);
 
 	ret = local->ops->ampdu_action(hw, IEEE80211_AMPDU_RX_STOP,
-					ra, tid, EINVAL);
+					ra, tid, NULL);
 	if (ret)
 		printk(KERN_DEBUG "HW problem - can not stop rx "
 				"aggergation for tid %d\n", tid);
-- 
cgit v1.2.3


From 9e7234923789897858e1a475c579b5e2e6ad5b74 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Mon, 28 Jan 2008 14:07:18 +0200
Subject: mac80211: A-MPDU Tx adding qdisc support

This patch allows qdisc support in A-MPDU Tx. a method to
handle QoS <-> TID switches is present in this patch.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h     |   8 ++-
 net/mac80211/ieee80211.c   |   8 +--
 net/mac80211/ieee80211_i.h |   5 +-
 net/mac80211/tx.c          |   2 +
 net/mac80211/wme.c         | 135 +++++++++++++++++++++++++++++++++++++++++----
 net/mac80211/wme.h         |  23 +++++++-
 6 files changed, 161 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 3bbc33cc2b96..0ce2e94dc844 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -246,6 +246,7 @@ struct ieee80211_tx_queue_stats_data {
  * @IEEE80211_TX_QUEUE_AFTER_BEACON: transmit queue for frames to be
  *	sent after a beacon
  * @IEEE80211_TX_QUEUE_BEACON: transmit queue for beacon frames
+ * @NUM_TX_DATA_QUEUES_AMPDU: adding more queues for A-MPDU
  */
 enum ieee80211_tx_queue {
 	IEEE80211_TX_QUEUE_DATA0,
@@ -261,11 +262,12 @@ enum ieee80211_tx_queue {
  * this struct need to have fixed values. As soon as it is removed, we can
  * fix these entries. */
 	IEEE80211_TX_QUEUE_AFTER_BEACON = 6,
-	IEEE80211_TX_QUEUE_BEACON = 7
+	IEEE80211_TX_QUEUE_BEACON = 7,
+	NUM_TX_DATA_QUEUES_AMPDU = 16
 };
 
 struct ieee80211_tx_queue_stats {
-	struct ieee80211_tx_queue_stats_data data[NUM_TX_DATA_QUEUES];
+	struct ieee80211_tx_queue_stats_data data[NUM_TX_DATA_QUEUES_AMPDU];
 };
 
 struct ieee80211_low_level_stats {
@@ -348,6 +350,8 @@ struct ieee80211_tx_control {
 #define IEEE80211_TXCTL_EAPOL_FRAME	(1<<11) /* internal to mac80211 */
 #define IEEE80211_TXCTL_SEND_AFTER_DTIM	(1<<12) /* send this frame after DTIM
 						 * beacon */
+#define IEEE80211_TXCTL_AMPDU		(1<<13) /* this frame should be sent
+						 * as part of an A-MPDU */
 	u32 flags;			       /* tx control flags defined
 						* above */
 	u8 key_idx;		/* keyidx from hw->set_key(), undefined if
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index c323c9af9b8f..f8e734f0da1a 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -462,7 +462,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 	spin_lock_bh(&local->mdev->queue_lock);
 
 	/* create a new queue for this aggregation */
-	/* ret = ieee80211_ht_agg_queue_add(local, sta, tid); */
+	ret = ieee80211_ht_agg_queue_add(local, sta, tid);
 
 	/* case no queue is available to aggregation
 	 * don't switch to aggregation */
@@ -488,7 +488,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 		/* No need to requeue the packets in the agg queue, since we
 		 * held the tx lock: no packet could be enqueued to the newly
 		 * allocated queue */
-		/* ieee80211_ht_agg_queue_remove(local, sta, tid, 0); */
+		 ieee80211_ht_agg_queue_remove(local, sta, tid, 0);
 #ifdef CONFIG_MAC80211_HT_DEBUG
 		printk(KERN_DEBUG "BA request denied - HW or queue unavailable"
 				" for tid %d\n", tid);
@@ -499,7 +499,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 	}
 
 	/* Will put all the packets in the new SW queue */
-	/* ieee80211_requeue(local, ieee802_1d_to_ac[tid]); */
+	ieee80211_requeue(local, ieee802_1d_to_ac[tid]);
 	spin_unlock_bh(&local->mdev->queue_lock);
 
 	/* We have most probably almost emptied the legacy queue */
@@ -675,7 +675,7 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 	 * the content of the qdiscs */
 	spin_lock_bh(&local->mdev->queue_lock);
 	/* remove the queue for this aggregation */
-	/* ieee80211_ht_agg_queue_remove(local, sta, tid, 1); */
+	ieee80211_ht_agg_queue_remove(local, sta, tid, 1);
 	spin_unlock_bh(&local->mdev->queue_lock);
 
 	/* we just requeued the all the frames that were in the removed
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 8a24c2c6ebc1..cfd0717c0033 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -165,6 +165,7 @@ struct ieee80211_txrx_data {
 #define IEEE80211_TXPD_DO_NOT_ENCRYPT	BIT(1)
 #define IEEE80211_TXPD_REQUEUE		BIT(2)
 #define IEEE80211_TXPD_EAPOL_FRAME	BIT(3)
+#define IEEE80211_TXPD_AMPDU		BIT(4)
 /* Stored in sk_buff->cb */
 struct ieee80211_tx_packet_data {
 	int ifindex;
@@ -452,8 +453,8 @@ struct ieee80211_local {
 	struct sta_info *sta_hash[STA_HASH_SIZE];
 	struct timer_list sta_cleanup;
 
-	unsigned long state[NUM_TX_DATA_QUEUES];
-	struct ieee80211_tx_stored_packet pending_packet[NUM_TX_DATA_QUEUES];
+	unsigned long state[NUM_TX_DATA_QUEUES_AMPDU];
+	struct ieee80211_tx_stored_packet pending_packet[NUM_TX_DATA_QUEUES_AMPDU];
 	struct tasklet_struct tx_pending_tasklet;
 
 	/* number of interfaces with corresponding IFF_ flags */
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 85d01646abf5..38e1b2bd8245 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -1260,6 +1260,8 @@ int ieee80211_master_start_xmit(struct sk_buff *skb,
 		control.flags |= IEEE80211_TXCTL_REQUEUE;
 	if (pkt_data->flags & IEEE80211_TXPD_EAPOL_FRAME)
 		control.flags |= IEEE80211_TXCTL_EAPOL_FRAME;
+	if (pkt_data->flags & IEEE80211_TXPD_AMPDU)
+		control.flags |= IEEE80211_TXCTL_AMPDU;
 	control.queue = pkt_data->queue;
 
 	ret = ieee80211_tx(odev, skb, &control);
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 4e236599dd31..425aa8588ea0 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -19,10 +19,13 @@
 #include "wme.h"
 
 /* maximum number of hardware queues we support. */
-#define TC_80211_MAX_QUEUES 8
+#define TC_80211_MAX_QUEUES 16
+
+const int ieee802_1d_to_ac[8] = { 2, 3, 3, 2, 1, 1, 0, 0 };
 
 struct ieee80211_sched_data
 {
+	unsigned long qdisc_pool;
 	struct tcf_proto *filter_list;
 	struct Qdisc *queues[TC_80211_MAX_QUEUES];
 	struct sk_buff_head requeued[TC_80211_MAX_QUEUES];
@@ -98,7 +101,6 @@ static inline int classify80211(struct sk_buff *skb, struct Qdisc *qd)
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	unsigned short fc = le16_to_cpu(hdr->frame_control);
 	int qos;
-	const int ieee802_1d_to_ac[8] = { 2, 3, 3, 2, 1, 1, 0, 0 };
 
 	/* see if frame is data or non data frame */
 	if (unlikely((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA)) {
@@ -146,9 +148,25 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 	unsigned short fc = le16_to_cpu(hdr->frame_control);
 	struct Qdisc *qdisc;
 	int err, queue;
+	struct sta_info *sta;
+	u8 tid;
 
 	if (pkt_data->flags & IEEE80211_TXPD_REQUEUE) {
-		skb_queue_tail(&q->requeued[pkt_data->queue], skb);
+		queue = pkt_data->queue;
+		sta = sta_info_get(local, hdr->addr1);
+		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
+		if (sta) {
+			int ampdu_queue = sta->tid_to_tx_q[tid];
+			if ((ampdu_queue < local->hw.queues) &&
+			    test_bit(ampdu_queue, &q->qdisc_pool)) {
+				queue = ampdu_queue;
+				pkt_data->flags |= IEEE80211_TXPD_AMPDU;
+			} else {
+				pkt_data->flags &= ~IEEE80211_TXPD_AMPDU;
+			}
+			sta_info_put(sta);
+		}
+		skb_queue_tail(&q->requeued[queue], skb);
 		qd->q.qlen++;
 		return 0;
 	}
@@ -159,14 +177,28 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 	 */
 	if (WLAN_FC_IS_QOS_DATA(fc)) {
 		u8 *p = skb->data + ieee80211_get_hdrlen(fc) - 2;
-		u8 qos_hdr = skb->priority & QOS_CONTROL_TAG1D_MASK;
+		u8 ack_policy = 0;
+		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
 		if (local->wifi_wme_noack_test)
-			qos_hdr |= QOS_CONTROL_ACK_POLICY_NOACK <<
+			ack_policy |= QOS_CONTROL_ACK_POLICY_NOACK <<
 					QOS_CONTROL_ACK_POLICY_SHIFT;
 		/* qos header is 2 bytes, second reserved */
-		*p = qos_hdr;
+		*p = ack_policy | tid;
 		p++;
 		*p = 0;
+
+		sta = sta_info_get(local, hdr->addr1);
+		if (sta) {
+			int ampdu_queue = sta->tid_to_tx_q[tid];
+			if ((ampdu_queue < local->hw.queues) &&
+				test_bit(ampdu_queue, &q->qdisc_pool)) {
+				queue = ampdu_queue;
+				pkt_data->flags |= IEEE80211_TXPD_AMPDU;
+			} else {
+				pkt_data->flags &= ~IEEE80211_TXPD_AMPDU;
+			}
+			sta_info_put(sta);
+		}
 	}
 
 	if (unlikely(queue >= local->hw.queues)) {
@@ -184,6 +216,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 			kfree_skb(skb);
 			err = NET_XMIT_DROP;
 	} else {
+		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
 		pkt_data->queue = (unsigned int) queue;
 		qdisc = q->queues[queue];
 		err = qdisc->enqueue(skb, qdisc);
@@ -235,10 +268,11 @@ static struct sk_buff *wme_qdiscop_dequeue(struct Qdisc* qd)
 	/* check all the h/w queues in numeric/priority order */
 	for (queue = 0; queue < hw->queues; queue++) {
 		/* see if there is room in this hardware queue */
-		if (test_bit(IEEE80211_LINK_STATE_XOFF,
-			     &local->state[queue]) ||
-		    test_bit(IEEE80211_LINK_STATE_PENDING,
-			     &local->state[queue]))
+		if ((test_bit(IEEE80211_LINK_STATE_XOFF,
+				&local->state[queue])) ||
+		    (test_bit(IEEE80211_LINK_STATE_PENDING,
+				&local->state[queue])) ||
+			 (!test_bit(queue, &q->qdisc_pool)))
 			continue;
 
 		/* there is space - try and get a frame */
@@ -360,6 +394,10 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 		}
 	}
 
+	/* reserve all legacy QoS queues */
+	for (i = 0; i < min(IEEE80211_TX_QUEUE_DATA4, queues); i++)
+		set_bit(i, &q->qdisc_pool);
+
 	return err;
 }
 
@@ -605,3 +643,80 @@ void ieee80211_wme_unregister(void)
 {
 	unregister_qdisc(&wme_qdisc_ops);
 }
+
+int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
+			struct sta_info *sta, u16 tid)
+{
+	int i;
+	struct ieee80211_sched_data *q =
+			qdisc_priv(local->mdev->qdisc_sleeping);
+	DECLARE_MAC_BUF(mac);
+
+	/* prepare the filter and save it for the SW queue
+	 * matching the recieved HW queue */
+
+	/* try to get a Qdisc from the pool */
+	for (i = IEEE80211_TX_QUEUE_BEACON; i < local->hw.queues; i++)
+		if (!test_and_set_bit(i, &q->qdisc_pool)) {
+			ieee80211_stop_queue(local_to_hw(local), i);
+			sta->tid_to_tx_q[tid] = i;
+
+			/* IF there are already pending packets
+			 * on this tid first we need to drain them
+			 * on the previous queue
+			 * since HT is strict in order */
+#ifdef CONFIG_MAC80211_HT_DEBUG
+			if (net_ratelimit())
+				printk(KERN_DEBUG "allocated aggregation queue"
+					" %d tid %d addr %s pool=0x%lX\n",
+					i, tid, print_mac(mac, sta->addr),
+					q->qdisc_pool);
+#endif /* CONFIG_MAC80211_HT_DEBUG */
+			return 0;
+		}
+
+	return -EAGAIN;
+}
+
+/**
+ * the caller needs to hold local->mdev->queue_lock
+ */
+void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
+				   struct sta_info *sta, u16 tid,
+				   u8 requeue)
+{
+	struct ieee80211_sched_data *q =
+		qdisc_priv(local->mdev->qdisc_sleeping);
+	int agg_queue = sta->tid_to_tx_q[tid];
+
+	/* return the qdisc to the pool */
+	clear_bit(agg_queue, &q->qdisc_pool);
+	sta->tid_to_tx_q[tid] = local->hw.queues;
+
+	if (requeue)
+		ieee80211_requeue(local, agg_queue);
+	else
+		q->queues[agg_queue]->ops->reset(q->queues[agg_queue]);
+}
+
+void ieee80211_requeue(struct ieee80211_local *local, int queue)
+{
+	struct Qdisc *root_qd = local->mdev->qdisc_sleeping;
+	struct ieee80211_sched_data *q = qdisc_priv(root_qd);
+	struct Qdisc *qdisc = q->queues[queue];
+	struct sk_buff *skb = NULL;
+	u32 len = qdisc->q.qlen;
+
+	if (!qdisc || !qdisc->dequeue)
+		return;
+
+	printk(KERN_DEBUG "requeue: qlen = %d\n", qdisc->q.qlen);
+	for (len = qdisc->q.qlen; len > 0; len--) {
+		skb = qdisc->dequeue(qdisc);
+		root_qd->q.qlen--;
+		/* packet will be classified again and */
+		/* skb->packet_data->queue will be overridden if needed */
+		if (skb)
+			wme_qdiscop_enqueue(skb, root_qd);
+	}
+}
diff --git a/net/mac80211/wme.h b/net/mac80211/wme.h
index 76c713a6450c..fcc6b05508cc 100644
--- a/net/mac80211/wme.h
+++ b/net/mac80211/wme.h
@@ -24,6 +24,8 @@
 
 #define QOS_CONTROL_TAG1D_MASK 0x07
 
+extern const int ieee802_1d_to_ac[8];
+
 static inline int WLAN_FC_IS_QOS_DATA(u16 fc)
 {
 	return (fc & 0x8C) == 0x88;
@@ -32,7 +34,12 @@ static inline int WLAN_FC_IS_QOS_DATA(u16 fc)
 #ifdef CONFIG_NET_SCHED
 void ieee80211_install_qdisc(struct net_device *dev);
 int ieee80211_qdisc_installed(struct net_device *dev);
-
+int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
+			       struct sta_info *sta, u16 tid);
+void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
+				   struct sta_info *sta, u16 tid,
+				   u8 requeue);
+void ieee80211_requeue(struct ieee80211_local *local, int queue);
 int ieee80211_wme_register(void);
 void ieee80211_wme_unregister(void);
 #else
@@ -43,7 +50,19 @@ static inline int ieee80211_qdisc_installed(struct net_device *dev)
 {
 	return 0;
 }
-
+static inline int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
+					     struct sta_info *sta, u16 tid)
+{
+	return -EAGAIN;
+}
+static inline void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
+						 struct sta_info *sta, u16 tid,
+						 u8 requeue)
+{
+}
+static inline void ieee80211_requeue(struct ieee80211_local *local, int queue)
+{
+}
 static inline int ieee80211_wme_register(void)
 {
 	return 0;
-- 
cgit v1.2.3


From 483fdcecc564ae6b011148a758517cf561f65678 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Mon, 28 Jan 2008 14:07:21 +0200
Subject: mac80211: A-MPDU Tx change tx_status to support Block Ack data

This patch adds fields to ieee80211_tx_status in order to allow block ack
information exchange between low-level driver,mac80211 and rate scaling
module.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 0ce2e94dc844..277488176a44 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -445,12 +445,14 @@ struct ieee80211_rx_status {
  *
  * @IEEE80211_TX_STATUS_TX_FILTERED: The frame was not transmitted
  *	because the destination STA was in powersave mode.
- *
  * @IEEE80211_TX_STATUS_ACK: Frame was acknowledged
+ * @IEEE80211_TX_STATUS_AMPDU: The frame was aggregated, so status
+ * 	is for the whole aggregation.
  */
 enum ieee80211_tx_status_flags {
 	IEEE80211_TX_STATUS_TX_FILTERED	= 1<<0,
 	IEEE80211_TX_STATUS_ACK		= 1<<1,
+	IEEE80211_TX_STATUS_AMPDU	= 1<<2,
 };
 
 /**
@@ -461,24 +463,25 @@ enum ieee80211_tx_status_flags {
  *
  * @control: a copy of the &struct ieee80211_tx_control passed to the driver
  *	in the tx() callback.
- *
  * @flags: transmit status flags, defined above
- *
- * @ack_signal: signal strength of the ACK frame
- *
+ * @retry_count: number of retries
  * @excessive_retries: set to 1 if the frame was retried many times
  *	but not acknowledged
- *
- * @retry_count: number of retries
- *
+ * @ampdu_ack_len: number of aggregated frames.
+ * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
+ * @ampdu_ack_map: block ack bit map for the aggregation.
+ * 	relevant only if IEEE80211_TX_STATUS_AMPDU was set.
+ * @ack_signal: signal strength of the ACK frame
  * @queue_length: ?? REMOVE
  * @queue_number: ?? REMOVE
  */
 struct ieee80211_tx_status {
 	struct ieee80211_tx_control control;
 	u8 flags;
-	bool excessive_retries;
 	u8 retry_count;
+	bool excessive_retries;
+	u8 ampdu_ack_len;
+	u64 ampdu_ack_map;
 	int ack_signal;
 	int queue_length;
 	int queue_number;
-- 
cgit v1.2.3


From 8318d78a44d49ac1edf2bdec7299de3617c4232e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 24 Jan 2008 19:38:38 +0100
Subject: cfg80211 API for channels/bitrates, mac80211 and driver conversion

This patch creates new cfg80211 wiphy API for channel and bitrate
registration and converts mac80211 and drivers to the new API. The
old mac80211 API is completely ripped out. All drivers (except ath5k)
are updated to the new API, in many cases I expect that optimisations
can be done.

Along with the regulatory code I've also ripped out the
IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED flag, I believe it to be
unnecessary if the hardware simply gives us whatever channels it wants
to support and we then enable/disable them as required, which is pretty
much required for travelling.

Additionally, the patch adds proper "basic" rate handling for STA
mode interface, AP mode interface will have to have new API added
to allow userspace to set the basic rate set, currently it'll be
empty... However, the basic rate handling will need to be moved to
the BSS conf stuff.

I do expect there to be bugs in this, especially wrt. transmit
power handling where I'm basically clueless about how it should work.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/Kconfig                 |   1 +
 drivers/net/wireless/adm8211.c               |  80 ++---
 drivers/net/wireless/adm8211.h               |  65 +---
 drivers/net/wireless/b43/b43.h               |   5 -
 drivers/net/wireless/b43/main.c              | 119 +++----
 drivers/net/wireless/b43/sysfs.c             |  89 +-----
 drivers/net/wireless/b43/xmit.c              |  81 ++---
 drivers/net/wireless/b43legacy/b43legacy.h   |   4 -
 drivers/net/wireless/b43legacy/main.c        | 159 +++++-----
 drivers/net/wireless/b43legacy/xmit.c        |  64 ++--
 drivers/net/wireless/iwlwifi/iwl-3945-rs.c   |  88 +++---
 drivers/net/wireless/iwlwifi/iwl-3945.c      |  35 +--
 drivers/net/wireless/iwlwifi/iwl-3945.h      |  13 +-
 drivers/net/wireless/iwlwifi/iwl-4965-rs.c   |  82 ++---
 drivers/net/wireless/iwlwifi/iwl-4965.c      |  43 ++-
 drivers/net/wireless/iwlwifi/iwl-4965.h      |  16 +-
 drivers/net/wireless/iwlwifi/iwl3945-base.c  | 445 ++++++++-------------------
 drivers/net/wireless/iwlwifi/iwl4965-base.c  | 438 +++++++-------------------
 drivers/net/wireless/p54.h                   |   4 -
 drivers/net/wireless/p54common.c             |  74 +++--
 drivers/net/wireless/p54common.h             |  75 -----
 drivers/net/wireless/rt2x00/rt2x00.h         |  11 +-
 drivers/net/wireless/rt2x00/rt2x00config.c   |  33 +-
 drivers/net/wireless/rt2x00/rt2x00dev.c      | 149 ++++-----
 drivers/net/wireless/rt2x00/rt61pci.c        |  23 +-
 drivers/net/wireless/rt2x00/rt73usb.c        |  23 +-
 drivers/net/wireless/rtl8180.h               |   2 +-
 drivers/net/wireless/rtl8180_dev.c           |  93 ++++--
 drivers/net/wireless/rtl8180_grf5101.c       |   5 +-
 drivers/net/wireless/rtl8180_max2820.c       |   5 +-
 drivers/net/wireless/rtl8180_rtl8225.c       |  15 +-
 drivers/net/wireless/rtl8180_sa2400.c        |   5 +-
 drivers/net/wireless/rtl8187.h               |   2 +-
 drivers/net/wireless/rtl8187_dev.c           |  82 +++--
 drivers/net/wireless/rtl8187_rtl8225.c       |  15 +-
 drivers/net/wireless/rtl818x.h               |  70 -----
 drivers/net/wireless/zd1211rw/zd_chip.c      |  15 +-
 drivers/net/wireless/zd1211rw/zd_ieee80211.c |  11 +-
 drivers/net/wireless/zd1211rw/zd_mac.c       | 170 +++++-----
 drivers/net/wireless/zd1211rw/zd_mac.h       |   2 +-
 include/net/mac80211.h                       | 197 ++----------
 include/net/wireless.h                       | 168 ++++++++++
 net/mac80211/Makefile                        |   1 -
 net/mac80211/cfg.c                           |  11 +-
 net/mac80211/debugfs.c                       |  47 +--
 net/mac80211/debugfs_sta.c                   |  18 --
 net/mac80211/ieee80211.c                     | 112 +++----
 net/mac80211/ieee80211_i.h                   |  63 +---
 net/mac80211/ieee80211_iface.c               |   2 +
 net/mac80211/ieee80211_ioctl.c               | 127 ++++----
 net/mac80211/ieee80211_rate.c                |  15 +-
 net/mac80211/ieee80211_rate.h                |  28 +-
 net/mac80211/ieee80211_sta.c                 | 384 ++++++++++++-----------
 net/mac80211/rc80211_pid_algo.c              |  76 ++---
 net/mac80211/rc80211_simple.c                |  66 ++--
 net/mac80211/regdomain.c                     | 152 ---------
 net/mac80211/rx.c                            |  82 ++---
 net/mac80211/sta_info.c                      |  24 --
 net/mac80211/sta_info.h                      |  10 +-
 net/mac80211/tx.c                            | 164 ++++++----
 net/mac80211/util.c                          | 142 ++-------
 net/wireless/Makefile                        |   2 +-
 net/wireless/core.c                          |  41 +++
 net/wireless/core.h                          |   3 +
 net/wireless/reg.c                           | 153 +++++++++
 net/wireless/util.c                          |  98 ++++++
 66 files changed, 2116 insertions(+), 2781 deletions(-)
 delete mode 100644 net/mac80211/regdomain.c
 create mode 100644 net/wireless/reg.c
 create mode 100644 net/wireless/util.c

(limited to 'include')

diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig
index 714a6ca30ad2..cd5fcc67e954 100644
--- a/drivers/net/wireless/Kconfig
+++ b/drivers/net/wireless/Kconfig
@@ -735,6 +735,7 @@ config P54_PCI
 config ATH5K
 	tristate "Atheros 5xxx wireless cards support"
 	depends on PCI && MAC80211 && WLAN_80211 && EXPERIMENTAL
+	depends on BROKEN
 	---help---
 	  This module adds support for wireless adapters based on
 	  Atheros 5xxx chipset.
diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c
index 79796186713e..7d4218206c47 100644
--- a/drivers/net/wireless/adm8211.c
+++ b/drivers/net/wireless/adm8211.c
@@ -48,6 +48,32 @@ static struct pci_device_id adm8211_pci_id_table[] __devinitdata = {
 	{ 0 }
 };
 
+static struct ieee80211_rate adm8211_rates[] = {
+	{ .bitrate = 10, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 20, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 55, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 110, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 220, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, /* XX ?? */
+};
+
+static const struct ieee80211_channel adm8211_channels[] = {
+	{ .center_freq = 2412},
+	{ .center_freq = 2417},
+	{ .center_freq = 2422},
+	{ .center_freq = 2427},
+	{ .center_freq = 2432},
+	{ .center_freq = 2437},
+	{ .center_freq = 2442},
+	{ .center_freq = 2447},
+	{ .center_freq = 2452},
+	{ .center_freq = 2457},
+	{ .center_freq = 2462},
+	{ .center_freq = 2467},
+	{ .center_freq = 2472},
+	{ .center_freq = 2484},
+};
+
+
 static void adm8211_eeprom_register_read(struct eeprom_93cx6 *eeprom)
 {
 	struct adm8211_priv *priv = eeprom->data;
@@ -155,17 +181,17 @@ static int adm8211_read_eeprom(struct ieee80211_hw *dev)
 	printk(KERN_DEBUG "%s (adm8211): Channel range: %d - %d\n",
 	       pci_name(priv->pdev), (int)chan_range.min, (int)chan_range.max);
 
-	priv->modes[0].num_channels = chan_range.max - chan_range.min + 1;
-	priv->modes[0].channels = priv->channels;
+	BUILD_BUG_ON(sizeof(priv->channels) != sizeof(adm8211_channels));
 
-	memcpy(priv->channels, adm8211_channels, sizeof(adm8211_channels));
+	memcpy(priv->channels, adm8211_channels, sizeof(priv->channels));
+	priv->band.channels = priv->channels;
+	priv->band.n_channels = ARRAY_SIZE(adm8211_channels);
+	priv->band.bitrates = adm8211_rates;
+	priv->band.n_bitrates = ARRAY_SIZE(adm8211_rates);
 
 	for (i = 1; i <= ARRAY_SIZE(adm8211_channels); i++)
-		if (i >= chan_range.min && i <= chan_range.max)
-			priv->channels[i - 1].flag =
-				IEEE80211_CHAN_W_SCAN |
-				IEEE80211_CHAN_W_ACTIVE_SCAN |
-				IEEE80211_CHAN_W_IBSS;
+		if (i < chan_range.min || i > chan_range.max)
+			priv->channels[i - 1].flags |= IEEE80211_CHAN_DISABLED;
 
 	switch (priv->eeprom->specific_bbptype) {
 	case ADM8211_BBP_RFMD3000:
@@ -347,7 +373,6 @@ static void adm8211_interrupt_rci(struct ieee80211_hw *dev)
 	unsigned int pktlen;
 	struct sk_buff *skb, *newskb;
 	unsigned int limit = priv->rx_ring_size;
-	static const u8 rate_tbl[] = {10, 20, 55, 110, 220};
 	u8 rssi, rate;
 
 	while (!(priv->rx_ring[entry].status & cpu_to_le32(RDES0_STATUS_OWN))) {
@@ -425,12 +450,10 @@ static void adm8211_interrupt_rci(struct ieee80211_hw *dev)
 			else
 				rx_status.ssi = 100 - rssi;
 
-			if (rate <= 4)
-				rx_status.rate = rate_tbl[rate];
+			rx_status.rate_idx = rate;
 
-			rx_status.channel = priv->channel;
-			rx_status.freq = adm8211_channels[priv->channel - 1].freq;
-			rx_status.phymode = MODE_IEEE80211B;
+			rx_status.freq = adm8211_channels[priv->channel - 1].center_freq;
+			rx_status.band = IEEE80211_BAND_2GHZ;
 
 			ieee80211_rx_irqsafe(dev, skb, &rx_status);
 		}
@@ -1054,7 +1077,7 @@ static int adm8211_set_rate(struct ieee80211_hw *dev)
 	if (priv->pdev->revision != ADM8211_REV_BA) {
 		rate_buf[0] = ARRAY_SIZE(adm8211_rates);
 		for (i = 0; i < ARRAY_SIZE(adm8211_rates); i++)
-			rate_buf[i + 1] = (adm8211_rates[i].rate / 5) | 0x80;
+			rate_buf[i + 1] = (adm8211_rates[i].bitrate / 5) | 0x80;
 	} else {
 		/* workaround for rev BA specific bug */
 		rate_buf[0] = 0x04;
@@ -1303,9 +1326,10 @@ static int adm8211_set_ssid(struct ieee80211_hw *dev, u8 *ssid, size_t ssid_len)
 static int adm8211_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
 {
 	struct adm8211_priv *priv = dev->priv;
+	int channel = ieee80211_frequency_to_channel(conf->channel->center_freq);
 
-	if (conf->channel != priv->channel) {
-		priv->channel = conf->channel;
+	if (channel != priv->channel) {
+		priv->channel = channel;
 		adm8211_rf_set_channel(dev, priv->channel);
 	}
 
@@ -1680,10 +1704,10 @@ static int adm8211_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 
 	if (control->tx_rate < 0) {
 		short_preamble = 1;
-		plcp_signal = -control->tx_rate;
+		plcp_signal = -control->tx_rate->bitrate;
 	} else {
 		short_preamble = 0;
-		plcp_signal = control->tx_rate;
+		plcp_signal = control->tx_rate->bitrate;
 	}
 
 	hdr = (struct ieee80211_hdr *)skb->data;
@@ -1880,18 +1904,11 @@ static int __devinit adm8211_probe(struct pci_dev *pdev,
 	SET_IEEE80211_PERM_ADDR(dev, perm_addr);
 
 	dev->extra_tx_headroom = sizeof(struct adm8211_tx_hdr);
-	dev->flags = IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED;
-	/* IEEE80211_HW_RX_INCLUDES_FCS in promisc mode */
+	/* dev->flags = IEEE80211_HW_RX_INCLUDES_FCS in promisc mode */
 
 	dev->channel_change_time = 1000;
 	dev->max_rssi = 100;	/* FIXME: find better value */
 
-	priv->modes[0].mode = MODE_IEEE80211B;
-	/* channel info filled in by adm8211_read_eeprom */
-	memcpy(priv->rates, adm8211_rates, sizeof(adm8211_rates));
-	priv->modes[0].num_rates = ARRAY_SIZE(adm8211_rates);
-	priv->modes[0].rates = priv->rates;
-
 	dev->queues = 1; /* ADM8211C supports more, maybe ADM8211B too */
 
 	priv->retry_limit = 3;
@@ -1917,14 +1934,7 @@ static int __devinit adm8211_probe(struct pci_dev *pdev,
 		goto err_free_desc;
 	}
 
-	priv->channel = priv->modes[0].channels[0].chan;
-
-	err = ieee80211_register_hwmode(dev, &priv->modes[0]);
-	if (err) {
-		printk(KERN_ERR "%s (adm8211): Can't register hwmode\n",
-		       pci_name(pdev));
-		goto err_free_desc;
-	}
+	priv->channel = 1;
 
 	err = ieee80211_register_hw(dev);
 	if (err) {
diff --git a/drivers/net/wireless/adm8211.h b/drivers/net/wireless/adm8211.h
index ef326fed42e4..8d7c564b3b04 100644
--- a/drivers/net/wireless/adm8211.h
+++ b/drivers/net/wireless/adm8211.h
@@ -534,61 +534,6 @@ struct adm8211_eeprom {
 	u8	cis_data[0];		/* 0x80, 384 bytes */
 } __attribute__ ((packed));
 
-static const struct ieee80211_rate adm8211_rates[] = {
-	{ .rate = 10,
-	  .val = 10,
-	  .val2 = -10,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 20,
-	  .val = 20,
-	  .val2 = -20,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 55,
-	  .val = 55,
-	  .val2 = -55,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 110,
-	  .val = 110,
-	  .val2 = -110,
-	  .flags = IEEE80211_RATE_CCK_2 }
-};
-
-struct ieee80211_chan_range {
-	u8 min;
-	u8 max;
-};
-
-static const struct ieee80211_channel adm8211_channels[] = {
-	{ .chan = 1,
-	  .freq = 2412},
-	{ .chan = 2,
-	  .freq = 2417},
-	{ .chan = 3,
-	  .freq = 2422},
-	{ .chan = 4,
-	  .freq = 2427},
-	{ .chan = 5,
-	  .freq = 2432},
-	{ .chan = 6,
-	  .freq = 2437},
-	{ .chan = 7,
-	  .freq = 2442},
-	{ .chan = 8,
-	  .freq = 2447},
-	{ .chan = 9,
-	  .freq = 2452},
-	{ .chan = 10,
-	  .freq = 2457},
-	{ .chan = 11,
-	  .freq = 2462},
-	{ .chan = 12,
-	  .freq = 2467},
-	{ .chan = 13,
-	  .freq = 2472},
-	{ .chan = 14,
-	  .freq = 2484},
-};
-
 struct adm8211_priv {
 	struct pci_dev *pdev;
 	spinlock_t lock;
@@ -603,9 +548,8 @@ struct adm8211_priv {
 	unsigned int cur_tx, dirty_tx, cur_rx;
 
 	struct ieee80211_low_level_stats stats;
-	struct ieee80211_hw_mode modes[1];
-	struct ieee80211_channel channels[ARRAY_SIZE(adm8211_channels)];
-	struct ieee80211_rate rates[ARRAY_SIZE(adm8211_rates)];
+	struct ieee80211_supported_band band;
+	struct ieee80211_channel channels[14];
 	int mode;
 
 	int channel;
@@ -643,6 +587,11 @@ struct adm8211_priv {
 	} transceiver_type;
 };
 
+struct ieee80211_chan_range {
+	u8 min;
+	u8 max;
+};
+
 static const struct ieee80211_chan_range cranges[] = {
 	{1,  11},	/* FCC */
 	{1,  11},	/* IC */
diff --git a/drivers/net/wireless/b43/b43.h b/drivers/net/wireless/b43/b43.h
index f13346ba9dd2..3e40323cd43f 100644
--- a/drivers/net/wireless/b43/b43.h
+++ b/drivers/net/wireless/b43/b43.h
@@ -468,10 +468,6 @@ struct b43_phy {
 	u8 possible_phymodes;
 	/* GMODE bit enabled? */
 	bool gmode;
-	/* Possible ieee80211 subsystem hwmodes for this PHY.
-	 * Which mode is selected, depends on thr GMODE enabled bit */
-#define B43_MAX_PHYHWMODES	2
-	struct ieee80211_hw_mode hwmodes[B43_MAX_PHYHWMODES];
 
 	/* Analog Type */
 	u8 analog;
@@ -727,7 +723,6 @@ struct b43_wldev {
 
 	bool bad_frames_preempt;	/* Use "Bad Frames Preemption" (default off) */
 	bool dfq_valid;		/* Directed frame queue valid (IBSS PS mode, ATIM) */
-	bool short_preamble;	/* TRUE, if short preamble is enabled. */
 	bool short_slot;	/* TRUE, if short slot timing is enabled. */
 	bool radio_hw_enable;	/* saved state of radio hardware enabled state */
 	bool suspend_in_progress;	/* TRUE, if we are in a suspend/resume cycle */
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index 51dfce16178a..017a041d07d0 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -96,25 +96,29 @@ MODULE_DEVICE_TABLE(ssb, b43_ssb_tbl);
  * data in there. This data is the same for all devices, so we don't
  * get concurrency issues */
 #define RATETAB_ENT(_rateid, _flags) \
-	{							\
-		.rate	= B43_RATE_TO_BASE100KBPS(_rateid),	\
-		.val	= (_rateid),				\
-		.val2	= (_rateid),				\
-		.flags	= (_flags),				\
+	{								\
+		.bitrate	= B43_RATE_TO_BASE100KBPS(_rateid),	\
+		.hw_value	= (_rateid),				\
+		.flags		= (_flags),				\
 	}
+
+/*
+ * NOTE: When changing this, sync with xmit.c's
+ *	 b43_plcp_get_bitrate_idx_* functions!
+ */
 static struct ieee80211_rate __b43_ratetable[] = {
-	RATETAB_ENT(B43_CCK_RATE_1MB, IEEE80211_RATE_CCK),
-	RATETAB_ENT(B43_CCK_RATE_2MB, IEEE80211_RATE_CCK_2),
-	RATETAB_ENT(B43_CCK_RATE_5MB, IEEE80211_RATE_CCK_2),
-	RATETAB_ENT(B43_CCK_RATE_11MB, IEEE80211_RATE_CCK_2),
-	RATETAB_ENT(B43_OFDM_RATE_6MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43_OFDM_RATE_9MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43_OFDM_RATE_12MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43_OFDM_RATE_18MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43_OFDM_RATE_24MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43_OFDM_RATE_36MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43_OFDM_RATE_48MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43_OFDM_RATE_54MB, IEEE80211_RATE_OFDM),
+	RATETAB_ENT(B43_CCK_RATE_1MB, 0),
+	RATETAB_ENT(B43_CCK_RATE_2MB, IEEE80211_RATE_SHORT_PREAMBLE),
+	RATETAB_ENT(B43_CCK_RATE_5MB, IEEE80211_RATE_SHORT_PREAMBLE),
+	RATETAB_ENT(B43_CCK_RATE_11MB, IEEE80211_RATE_SHORT_PREAMBLE),
+	RATETAB_ENT(B43_OFDM_RATE_6MB, 0),
+	RATETAB_ENT(B43_OFDM_RATE_9MB, 0),
+	RATETAB_ENT(B43_OFDM_RATE_12MB, 0),
+	RATETAB_ENT(B43_OFDM_RATE_18MB, 0),
+	RATETAB_ENT(B43_OFDM_RATE_24MB, 0),
+	RATETAB_ENT(B43_OFDM_RATE_36MB, 0),
+	RATETAB_ENT(B43_OFDM_RATE_48MB, 0),
+	RATETAB_ENT(B43_OFDM_RATE_54MB, 0),
 };
 
 #define b43_a_ratetable		(__b43_ratetable + 4)
@@ -126,14 +130,8 @@ static struct ieee80211_rate __b43_ratetable[] = {
 
 #define CHANTAB_ENT(_chanid, _freq) \
 	{							\
-		.chan	= (_chanid),				\
-		.freq	= (_freq),				\
-		.val	= (_chanid),				\
-		.flag	= IEEE80211_CHAN_W_SCAN |		\
-			  IEEE80211_CHAN_W_ACTIVE_SCAN |	\
-			  IEEE80211_CHAN_W_IBSS,		\
-		.power_level	= 0xFF,				\
-		.antenna_max	= 0xFF,				\
+		.center_freq	= (_freq),			\
+		.hw_value	= (_chanid),			\
 	}
 static struct ieee80211_channel b43_2ghz_chantable[] = {
 	CHANTAB_ENT(1, 2412),
@@ -151,9 +149,8 @@ static struct ieee80211_channel b43_2ghz_chantable[] = {
 	CHANTAB_ENT(13, 2472),
 	CHANTAB_ENT(14, 2484),
 };
-#define b43_2ghz_chantable_size	ARRAY_SIZE(b43_2ghz_chantable)
 
-#if 0
+#ifdef NOTYET
 static struct ieee80211_channel b43_5ghz_chantable[] = {
 	CHANTAB_ENT(36, 5180),
 	CHANTAB_ENT(40, 5200),
@@ -169,9 +166,22 @@ static struct ieee80211_channel b43_5ghz_chantable[] = {
 	CHANTAB_ENT(161, 5805),
 	CHANTAB_ENT(165, 5825),
 };
-#define b43_5ghz_chantable_size	ARRAY_SIZE(b43_5ghz_chantable)
+
+static struct ieee80211_supported_band b43_band_5GHz = {
+	.channels = b43_5ghz_chantable,
+	.n_channels = ARRAY_SIZE(b43_5ghz_chantable),
+	.bitrates = b43_a_ratetable,
+	.n_bitrates = b43_a_ratetable_size,
+};
 #endif
 
+static struct ieee80211_supported_band b43_band_2GHz = {
+	.channels = b43_2ghz_chantable,
+	.n_channels = ARRAY_SIZE(b43_2ghz_chantable),
+	.bitrates = b43_g_ratetable,
+	.n_bitrates = b43_g_ratetable_size,
+};
+
 static void b43_wireless_core_exit(struct b43_wldev *dev);
 static int b43_wireless_core_init(struct b43_wldev *dev);
 static void b43_wireless_core_stop(struct b43_wldev *dev);
@@ -1222,17 +1232,18 @@ static void b43_write_beacon_template(struct b43_wldev *dev,
 }
 
 static void b43_write_probe_resp_plcp(struct b43_wldev *dev,
-				      u16 shm_offset, u16 size, u8 rate)
+				      u16 shm_offset, u16 size,
+				      struct ieee80211_rate *rate)
 {
 	struct b43_plcp_hdr4 plcp;
 	u32 tmp;
 	__le16 dur;
 
 	plcp.data = 0;
-	b43_generate_plcp_hdr(&plcp, size + FCS_LEN, rate);
+	b43_generate_plcp_hdr(&plcp, size + FCS_LEN, rate->hw_value);
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
 					       dev->wl->vif, size,
-					       B43_RATE_TO_BASE100KBPS(rate));
+					       rate);
 	/* Write PLCP in two parts and timing for packet transfer */
 	tmp = le32_to_cpu(plcp.data);
 	b43_shm_write16(dev, B43_SHM_SHARED, shm_offset, tmp & 0xFFFF);
@@ -1247,7 +1258,8 @@ static void b43_write_probe_resp_plcp(struct b43_wldev *dev,
  * 3) Stripping TIM
  */
 static const u8 * b43_generate_probe_resp(struct b43_wldev *dev,
-					  u16 *dest_size, u8 rate)
+					  u16 *dest_size,
+					  struct ieee80211_rate *rate)
 {
 	const u8 *src_data;
 	u8 *dest_data;
@@ -1292,7 +1304,7 @@ static const u8 * b43_generate_probe_resp(struct b43_wldev *dev,
 					 IEEE80211_STYPE_PROBE_RESP);
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
 					       dev->wl->vif, *dest_size,
-					       B43_RATE_TO_BASE100KBPS(rate));
+					       rate);
 	hdr->duration_id = dur;
 
 	return dest_data;
@@ -1300,7 +1312,8 @@ static const u8 * b43_generate_probe_resp(struct b43_wldev *dev,
 
 static void b43_write_probe_resp_template(struct b43_wldev *dev,
 					  u16 ram_offset,
-					  u16 shm_size_offset, u8 rate)
+					  u16 shm_size_offset,
+					  struct ieee80211_rate *rate)
 {
 	const u8 *probe_resp_data;
 	u16 size;
@@ -1313,14 +1326,15 @@ static void b43_write_probe_resp_template(struct b43_wldev *dev,
 	/* Looks like PLCP headers plus packet timings are stored for
 	 * all possible basic rates
 	 */
-	b43_write_probe_resp_plcp(dev, 0x31A, size, B43_CCK_RATE_1MB);
-	b43_write_probe_resp_plcp(dev, 0x32C, size, B43_CCK_RATE_2MB);
-	b43_write_probe_resp_plcp(dev, 0x33E, size, B43_CCK_RATE_5MB);
-	b43_write_probe_resp_plcp(dev, 0x350, size, B43_CCK_RATE_11MB);
+	b43_write_probe_resp_plcp(dev, 0x31A, size, &b43_b_ratetable[0]);
+	b43_write_probe_resp_plcp(dev, 0x32C, size, &b43_b_ratetable[1]);
+	b43_write_probe_resp_plcp(dev, 0x33E, size, &b43_b_ratetable[2]);
+	b43_write_probe_resp_plcp(dev, 0x350, size, &b43_b_ratetable[3]);
 
 	size = min((size_t) size, 0x200 - sizeof(struct b43_plcp_hdr6));
 	b43_write_template_common(dev, probe_resp_data,
-				  size, ram_offset, shm_size_offset, rate);
+				  size, ram_offset, shm_size_offset,
+				  rate->hw_value);
 	kfree(probe_resp_data);
 }
 
@@ -1388,7 +1402,7 @@ static void handle_irq_beacon(struct b43_wldev *dev)
 			b43_write_beacon_template(dev, 0x68, 0x18,
 						  B43_CCK_RATE_1MB);
 			b43_write_probe_resp_template(dev, 0x268, 0x4A,
-						      B43_CCK_RATE_11MB);
+						      &__b43_ratetable[3]);
 			wl->beacon0_uploaded = 1;
 		}
 		cmd |= B43_MACCMD_BEACON0_VALID;
@@ -2830,14 +2844,11 @@ static int b43_op_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf)
 	mutex_lock(&wl->mutex);
 
 	/* Switch the PHY mode (if necessary). */
-	switch (conf->phymode) {
-	case MODE_IEEE80211A:
+	switch (conf->channel->band) {
+	case IEEE80211_BAND_5GHZ:
 		new_phymode = B43_PHYMODE_A;
 		break;
-	case MODE_IEEE80211B:
-		new_phymode = B43_PHYMODE_B;
-		break;
-	case MODE_IEEE80211G:
+	case IEEE80211_BAND_2GHZ:
 		new_phymode = B43_PHYMODE_G;
 		break;
 	default:
@@ -2863,8 +2874,8 @@ static int b43_op_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf)
 
 	/* Switch to the requested channel.
 	 * The firmware takes care of races with the TX handler. */
-	if (conf->channel_val != phy->channel)
-		b43_radio_selectchannel(dev, conf->channel_val, 0);
+	if (conf->channel->hw_value != phy->channel)
+		b43_radio_selectchannel(dev, conf->channel->hw_value, 0);
 
 	/* Enable/Disable ShortSlot timing. */
 	if ((!!(conf->flags & IEEE80211_CONF_SHORT_SLOT_TIME)) !=
@@ -3810,9 +3821,7 @@ static int b43_setup_modes(struct b43_wldev *dev,
 			   bool have_2ghz_phy, bool have_5ghz_phy)
 {
 	struct ieee80211_hw *hw = dev->wl->hw;
-	struct ieee80211_hw_mode *mode;
 	struct b43_phy *phy = &dev->phy;
-	int err;
 
 	/* XXX: This function will go away soon, when mac80211
 	 *      band stuff is rewritten. So this is just a hack.
@@ -3821,15 +3830,7 @@ static int b43_setup_modes(struct b43_wldev *dev,
 	 *      This assumption is OK, as any B, N or A PHY will already
 	 *      have died a horrible sanity check death earlier. */
 
-	mode = &phy->hwmodes[0];
-	mode->mode = MODE_IEEE80211G;
-	mode->num_channels = b43_2ghz_chantable_size;
-	mode->channels = b43_2ghz_chantable;
-	mode->num_rates = b43_g_ratetable_size;
-	mode->rates = b43_g_ratetable;
-	err = ieee80211_register_hwmode(hw, mode);
-	if (err)
-		return err;
+	hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &b43_band_2GHz;
 	phy->possible_phymodes |= B43_PHYMODE_G;
 
 	return 0;
diff --git a/drivers/net/wireless/b43/sysfs.c b/drivers/net/wireless/b43/sysfs.c
index f4faff6a7d6c..275095b8cbe7 100644
--- a/drivers/net/wireless/b43/sysfs.c
+++ b/drivers/net/wireless/b43/sysfs.c
@@ -47,29 +47,6 @@ static int get_integer(const char *buf, size_t count)
 	return ret;
 }
 
-static int get_boolean(const char *buf, size_t count)
-{
-	if (count != 0) {
-		if (buf[0] == '1')
-			return 1;
-		if (buf[0] == '0')
-			return 0;
-		if (count >= 4 && memcmp(buf, "true", 4) == 0)
-			return 1;
-		if (count >= 5 && memcmp(buf, "false", 5) == 0)
-			return 0;
-		if (count >= 3 && memcmp(buf, "yes", 3) == 0)
-			return 1;
-		if (count >= 2 && memcmp(buf, "no", 2) == 0)
-			return 0;
-		if (count >= 2 && memcmp(buf, "on", 2) == 0)
-			return 1;
-		if (count >= 3 && memcmp(buf, "off", 3) == 0)
-			return 0;
-	}
-	return -EINVAL;
-}
-
 static ssize_t b43_attr_interfmode_show(struct device *dev,
 					struct device_attribute *attr,
 					char *buf)
@@ -155,82 +132,18 @@ static ssize_t b43_attr_interfmode_store(struct device *dev,
 static DEVICE_ATTR(interference, 0644,
 		   b43_attr_interfmode_show, b43_attr_interfmode_store);
 
-static ssize_t b43_attr_preamble_show(struct device *dev,
-				      struct device_attribute *attr, char *buf)
-{
-	struct b43_wldev *wldev = dev_to_b43_wldev(dev);
-	ssize_t count;
-
-	if (!capable(CAP_NET_ADMIN))
-		return -EPERM;
-
-	mutex_lock(&wldev->wl->mutex);
-
-	if (wldev->short_preamble)
-		count =
-		    snprintf(buf, PAGE_SIZE, "1 (Short Preamble enabled)\n");
-	else
-		count =
-		    snprintf(buf, PAGE_SIZE, "0 (Short Preamble disabled)\n");
-
-	mutex_unlock(&wldev->wl->mutex);
-
-	return count;
-}
-
-static ssize_t b43_attr_preamble_store(struct device *dev,
-				       struct device_attribute *attr,
-				       const char *buf, size_t count)
-{
-	struct b43_wldev *wldev = dev_to_b43_wldev(dev);
-	unsigned long flags;
-	int value;
-
-	if (!capable(CAP_NET_ADMIN))
-		return -EPERM;
-
-	value = get_boolean(buf, count);
-	if (value < 0)
-		return value;
-	mutex_lock(&wldev->wl->mutex);
-	spin_lock_irqsave(&wldev->wl->irq_lock, flags);
-
-	wldev->short_preamble = !!value;
-
-	spin_unlock_irqrestore(&wldev->wl->irq_lock, flags);
-	mutex_unlock(&wldev->wl->mutex);
-
-	return count;
-}
-
-static DEVICE_ATTR(shortpreamble, 0644,
-		   b43_attr_preamble_show, b43_attr_preamble_store);
-
 int b43_sysfs_register(struct b43_wldev *wldev)
 {
 	struct device *dev = wldev->dev->dev;
-	int err;
 
 	B43_WARN_ON(b43_status(wldev) != B43_STAT_INITIALIZED);
 
-	err = device_create_file(dev, &dev_attr_interference);
-	if (err)
-		goto out;
-	err = device_create_file(dev, &dev_attr_shortpreamble);
-	if (err)
-		goto err_remove_interfmode;
-
-      out:
-	return err;
-      err_remove_interfmode:
-	device_remove_file(dev, &dev_attr_interference);
-	goto out;
+	return device_create_file(dev, &dev_attr_interference);
 }
 
 void b43_sysfs_unregister(struct b43_wldev *wldev)
 {
 	struct device *dev = wldev->dev->dev;
 
-	device_remove_file(dev, &dev_attr_shortpreamble);
 	device_remove_file(dev, &dev_attr_interference);
 }
diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index 7caa26eb4105..4014b6c8272b 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -32,46 +32,48 @@
 #include "dma.h"
 
 
-/* Extract the bitrate out of a CCK PLCP header. */
-static u8 b43_plcp_get_bitrate_cck(struct b43_plcp_hdr6 *plcp)
+/* Extract the bitrate index out of a CCK PLCP header. */
+static int b43_plcp_get_bitrate_idx_cck(struct b43_plcp_hdr6 *plcp)
 {
 	switch (plcp->raw[0]) {
 	case 0x0A:
-		return B43_CCK_RATE_1MB;
+		return 0;
 	case 0x14:
-		return B43_CCK_RATE_2MB;
+		return 1;
 	case 0x37:
-		return B43_CCK_RATE_5MB;
+		return 2;
 	case 0x6E:
-		return B43_CCK_RATE_11MB;
+		return 3;
 	}
 	B43_WARN_ON(1);
-	return 0;
+	return -1;
 }
 
-/* Extract the bitrate out of an OFDM PLCP header. */
-static u8 b43_plcp_get_bitrate_ofdm(struct b43_plcp_hdr6 *plcp)
+/* Extract the bitrate index out of an OFDM PLCP header. */
+static u8 b43_plcp_get_bitrate_idx_ofdm(struct b43_plcp_hdr6 *plcp, bool aphy)
 {
+	int base = aphy ? 0 : 4;
+
 	switch (plcp->raw[0] & 0xF) {
 	case 0xB:
-		return B43_OFDM_RATE_6MB;
+		return base + 0;
 	case 0xF:
-		return B43_OFDM_RATE_9MB;
+		return base + 1;
 	case 0xA:
-		return B43_OFDM_RATE_12MB;
+		return base + 2;
 	case 0xE:
-		return B43_OFDM_RATE_18MB;
+		return base + 3;
 	case 0x9:
-		return B43_OFDM_RATE_24MB;
+		return base + 4;
 	case 0xD:
-		return B43_OFDM_RATE_36MB;
+		return base + 5;
 	case 0x8:
-		return B43_OFDM_RATE_48MB;
+		return base + 6;
 	case 0xC:
-		return B43_OFDM_RATE_54MB;
+		return base + 7;
 	}
 	B43_WARN_ON(1);
-	return 0;
+	return -1;
 }
 
 u8 b43_plcp_get_ratecode_cck(const u8 bitrate)
@@ -191,6 +193,7 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 	    (const struct ieee80211_hdr *)fragment_data;
 	int use_encryption = (!(txctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT));
 	u16 fctl = le16_to_cpu(wlhdr->frame_control);
+	struct ieee80211_rate *fbrate;
 	u8 rate, rate_fb;
 	int rate_ofdm, rate_fb_ofdm;
 	unsigned int plcp_fragment_len;
@@ -200,9 +203,11 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 
 	memset(txhdr, 0, sizeof(*txhdr));
 
-	rate = txctl->tx_rate;
+	WARN_ON(!txctl->tx_rate);
+	rate = txctl->tx_rate ? txctl->tx_rate->hw_value : B43_CCK_RATE_1MB;
 	rate_ofdm = b43_is_ofdm_rate(rate);
-	rate_fb = (txctl->alt_retry_rate == -1) ? rate : txctl->alt_retry_rate;
+	fbrate = txctl->alt_retry_rate ? : txctl->tx_rate;
+	rate_fb = fbrate->hw_value;
 	rate_fb_ofdm = b43_is_ofdm_rate(rate_fb);
 
 	if (rate_ofdm)
@@ -221,11 +226,10 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		 * use the original dur_id field. */
 		txhdr->dur_fb = wlhdr->duration_id;
 	} else {
-		int fbrate_base100kbps = B43_RATE_TO_BASE100KBPS(rate_fb);
 		txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw,
 								 txctl->vif,
 								 fragment_len,
-								 fbrate_base100kbps);
+								 fbrate);
 	}
 
 	plcp_fragment_len = fragment_len + FCS_LEN;
@@ -287,7 +291,7 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		phy_ctl |= B43_TXH_PHY_ENC_OFDM;
 	else
 		phy_ctl |= B43_TXH_PHY_ENC_CCK;
-	if (dev->short_preamble)
+	if (txctl->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
 		phy_ctl |= B43_TXH_PHY_SHORTPRMBL;
 
 	switch (b43_ieee80211_antenna_sanitize(dev, txctl->antenna_sel_tx)) {
@@ -332,7 +336,8 @@ int b43_generate_txhdr(struct b43_wldev *dev,
 		int rts_rate_ofdm, rts_rate_fb_ofdm;
 		struct b43_plcp_hdr6 *plcp;
 
-		rts_rate = txctl->rts_cts_rate;
+		WARN_ON(!txctl->rts_cts_rate);
+		rts_rate = txctl->rts_cts_rate ? txctl->rts_cts_rate->hw_value : B43_CCK_RATE_1MB;
 		rts_rate_ofdm = b43_is_ofdm_rate(rts_rate);
 		rts_rate_fb = b43_calc_fallback_rate(rts_rate);
 		rts_rate_fb_ofdm = b43_is_ofdm_rate(rts_rate_fb);
@@ -506,6 +511,7 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr)
 	u16 phystat0, phystat3, chanstat, mactime;
 	u32 macstat;
 	u16 chanid;
+	u16 phytype;
 	u8 jssi;
 	int padding;
 
@@ -518,6 +524,7 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr)
 	macstat = le32_to_cpu(rxhdr->mac_status);
 	mactime = le16_to_cpu(rxhdr->mac_time);
 	chanstat = le16_to_cpu(rxhdr->channel);
+	phytype = chanstat & B43_RX_CHAN_PHYTYPE;
 
 	if (macstat & B43_RX_MAC_FCSERR)
 		dev->wl->ieee_stats.dot11FCSErrorCount++;
@@ -575,9 +582,10 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr)
 	/* the next line looks wrong, but is what mac80211 wants */
 	status.signal = (jssi * 100) / B43_RX_MAX_SSI;
 	if (phystat0 & B43_RX_PHYST0_OFDM)
-		status.rate = b43_plcp_get_bitrate_ofdm(plcp);
+		status.rate_idx = b43_plcp_get_bitrate_idx_ofdm(plcp,
+						phytype == B43_PHYTYPE_A);
 	else
-		status.rate = b43_plcp_get_bitrate_cck(plcp);
+		status.rate_idx = b43_plcp_get_bitrate_idx_cck(plcp);
 	status.antenna = !!(phystat0 & B43_RX_PHYST0_ANT);
 
 	/*
@@ -601,29 +609,28 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr)
 	chanid = (chanstat & B43_RX_CHAN_ID) >> B43_RX_CHAN_ID_SHIFT;
 	switch (chanstat & B43_RX_CHAN_PHYTYPE) {
 	case B43_PHYTYPE_A:
-		status.phymode = MODE_IEEE80211A;
+		status.band = IEEE80211_BAND_5GHZ;
 		B43_WARN_ON(1);
 		/* FIXME: We don't really know which value the "chanid" contains.
 		 *        So the following assignment might be wrong. */
-		status.channel = chanid;
-		status.freq = b43_channel_to_freq_5ghz(status.channel);
+		status.freq = b43_channel_to_freq_5ghz(chanid);
 		break;
 	case B43_PHYTYPE_G:
-		status.phymode = MODE_IEEE80211G;
+		status.band = IEEE80211_BAND_2GHZ;
 		/* chanid is the radio channel cookie value as used
 		 * to tune the radio. */
 		status.freq = chanid + 2400;
-		status.channel = b43_freq_to_channel_2ghz(status.freq);
 		break;
 	case B43_PHYTYPE_N:
-		status.phymode = 0xDEAD /*FIXME MODE_IEEE80211N*/;
 		/* chanid is the SHM channel cookie. Which is the plain
 		 * channel number in b43. */
-		status.channel = chanid;
-		if (chanstat & B43_RX_CHAN_5GHZ)
-			status.freq = b43_freq_to_channel_5ghz(status.freq);
-		else
-			status.freq = b43_freq_to_channel_2ghz(status.freq);
+		if (chanstat & B43_RX_CHAN_5GHZ) {
+			status.band = IEEE80211_BAND_5GHZ;
+			status.freq = b43_freq_to_channel_5ghz(chanid);
+		} else {
+			status.band = IEEE80211_BAND_2GHZ;
+			status.freq = b43_freq_to_channel_2ghz(chanid);
+		}
 		break;
 	default:
 		B43_WARN_ON(1);
diff --git a/drivers/net/wireless/b43legacy/b43legacy.h b/drivers/net/wireless/b43legacy/b43legacy.h
index 93d45b71799a..5f217d6d2e6a 100644
--- a/drivers/net/wireless/b43legacy/b43legacy.h
+++ b/drivers/net/wireless/b43legacy/b43legacy.h
@@ -392,10 +392,6 @@ struct b43legacy_phy {
 	u8 possible_phymodes;
 	/* GMODE bit enabled in MACCTL? */
 	bool gmode;
-	/* Possible ieee80211 subsystem hwmodes for this PHY.
-	 * Which mode is selected, depends on thr GMODE enabled bit */
-#define B43legacy_MAX_PHYHWMODES	2
-	struct ieee80211_hw_mode hwmodes[B43legacy_MAX_PHYHWMODES];
 
 	/* Analog Type */
 	u8 analog;
diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c
index c39de422e220..d2a72a2cd178 100644
--- a/drivers/net/wireless/b43legacy/main.c
+++ b/drivers/net/wireless/b43legacy/main.c
@@ -95,28 +95,29 @@ MODULE_DEVICE_TABLE(ssb, b43legacy_ssb_tbl);
  * data in there. This data is the same for all devices, so we don't
  * get concurrency issues */
 #define RATETAB_ENT(_rateid, _flags) \
-	{							\
-		.rate	= B43legacy_RATE_TO_100KBPS(_rateid),	\
-		.val	= (_rateid),				\
-		.val2	= (_rateid),				\
-		.flags	= (_flags),				\
+	{								\
+		.bitrate	= B43legacy_RATE_TO_100KBPS(_rateid),	\
+		.hw_value	= (_rateid),				\
+		.flags		= (_flags),				\
 	}
+/*
+ * NOTE: When changing this, sync with xmit.c's
+ *	 b43legacy_plcp_get_bitrate_idx_* functions!
+ */
 static struct ieee80211_rate __b43legacy_ratetable[] = {
-	RATETAB_ENT(B43legacy_CCK_RATE_1MB, IEEE80211_RATE_CCK),
-	RATETAB_ENT(B43legacy_CCK_RATE_2MB, IEEE80211_RATE_CCK_2),
-	RATETAB_ENT(B43legacy_CCK_RATE_5MB, IEEE80211_RATE_CCK_2),
-	RATETAB_ENT(B43legacy_CCK_RATE_11MB, IEEE80211_RATE_CCK_2),
-	RATETAB_ENT(B43legacy_OFDM_RATE_6MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43legacy_OFDM_RATE_9MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43legacy_OFDM_RATE_12MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43legacy_OFDM_RATE_18MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43legacy_OFDM_RATE_24MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43legacy_OFDM_RATE_36MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43legacy_OFDM_RATE_48MB, IEEE80211_RATE_OFDM),
-	RATETAB_ENT(B43legacy_OFDM_RATE_54MB, IEEE80211_RATE_OFDM),
+	RATETAB_ENT(B43legacy_CCK_RATE_1MB, 0),
+	RATETAB_ENT(B43legacy_CCK_RATE_2MB, IEEE80211_RATE_SHORT_PREAMBLE),
+	RATETAB_ENT(B43legacy_CCK_RATE_5MB, IEEE80211_RATE_SHORT_PREAMBLE),
+	RATETAB_ENT(B43legacy_CCK_RATE_11MB, IEEE80211_RATE_SHORT_PREAMBLE),
+	RATETAB_ENT(B43legacy_OFDM_RATE_6MB, 0),
+	RATETAB_ENT(B43legacy_OFDM_RATE_9MB, 0),
+	RATETAB_ENT(B43legacy_OFDM_RATE_12MB, 0),
+	RATETAB_ENT(B43legacy_OFDM_RATE_18MB, 0),
+	RATETAB_ENT(B43legacy_OFDM_RATE_24MB, 0),
+	RATETAB_ENT(B43legacy_OFDM_RATE_36MB, 0),
+	RATETAB_ENT(B43legacy_OFDM_RATE_48MB, 0),
+	RATETAB_ENT(B43legacy_OFDM_RATE_54MB, 0),
 };
-#define b43legacy_a_ratetable		(__b43legacy_ratetable + 4)
-#define b43legacy_a_ratetable_size	8
 #define b43legacy_b_ratetable		(__b43legacy_ratetable + 0)
 #define b43legacy_b_ratetable_size	4
 #define b43legacy_g_ratetable		(__b43legacy_ratetable + 0)
@@ -124,14 +125,8 @@ static struct ieee80211_rate __b43legacy_ratetable[] = {
 
 #define CHANTAB_ENT(_chanid, _freq) \
 	{							\
-		.chan	= (_chanid),				\
-		.freq	= (_freq),				\
-		.val	= (_chanid),				\
-		.flag	= IEEE80211_CHAN_W_SCAN |		\
-			  IEEE80211_CHAN_W_ACTIVE_SCAN |	\
-			  IEEE80211_CHAN_W_IBSS,		\
-		.power_level	= 0x0A,				\
-		.antenna_max	= 0xFF,				\
+		.center_freq	= (_freq),			\
+		.hw_value	= (_chanid),			\
 	}
 static struct ieee80211_channel b43legacy_bg_chantable[] = {
 	CHANTAB_ENT(1, 2412),
@@ -149,7 +144,20 @@ static struct ieee80211_channel b43legacy_bg_chantable[] = {
 	CHANTAB_ENT(13, 2472),
 	CHANTAB_ENT(14, 2484),
 };
-#define b43legacy_bg_chantable_size	ARRAY_SIZE(b43legacy_bg_chantable)
+
+static struct ieee80211_supported_band b43legacy_band_2GHz_BPHY = {
+	.channels = b43legacy_bg_chantable,
+	.n_channels = ARRAY_SIZE(b43legacy_bg_chantable),
+	.bitrates = b43legacy_b_ratetable,
+	.n_bitrates = b43legacy_b_ratetable_size,
+};
+
+static struct ieee80211_supported_band b43legacy_band_2GHz_GPHY = {
+	.channels = b43legacy_bg_chantable,
+	.n_channels = ARRAY_SIZE(b43legacy_bg_chantable),
+	.bitrates = b43legacy_g_ratetable,
+	.n_bitrates = b43legacy_g_ratetable_size,
+};
 
 static void b43legacy_wireless_core_exit(struct b43legacy_wldev *dev);
 static int b43legacy_wireless_core_init(struct b43legacy_wldev *dev);
@@ -969,18 +977,18 @@ static void b43legacy_write_beacon_template(struct b43legacy_wldev *dev,
 
 static void b43legacy_write_probe_resp_plcp(struct b43legacy_wldev *dev,
 					    u16 shm_offset, u16 size,
-					    u8 rate)
+					    struct ieee80211_rate *rate)
 {
 	struct b43legacy_plcp_hdr4 plcp;
 	u32 tmp;
 	__le16 dur;
 
 	plcp.data = 0;
-	b43legacy_generate_plcp_hdr(&plcp, size + FCS_LEN, rate);
+	b43legacy_generate_plcp_hdr(&plcp, size + FCS_LEN, rate->bitrate);
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
 					       dev->wl->vif,
 					       size,
-					       B43legacy_RATE_TO_100KBPS(rate));
+					       rate);
 	/* Write PLCP in two parts and timing for packet transfer */
 	tmp = le32_to_cpu(plcp.data);
 	b43legacy_shm_write16(dev, B43legacy_SHM_SHARED, shm_offset,
@@ -998,7 +1006,8 @@ static void b43legacy_write_probe_resp_plcp(struct b43legacy_wldev *dev,
  * 3) Stripping TIM
  */
 static u8 *b43legacy_generate_probe_resp(struct b43legacy_wldev *dev,
-					 u16 *dest_size, u8 rate)
+					 u16 *dest_size,
+					 struct ieee80211_rate *rate)
 {
 	const u8 *src_data;
 	u8 *dest_data;
@@ -1046,7 +1055,7 @@ static u8 *b43legacy_generate_probe_resp(struct b43legacy_wldev *dev,
 	dur = ieee80211_generic_frame_duration(dev->wl->hw,
 					       dev->wl->vif,
 					       *dest_size,
-					       B43legacy_RATE_TO_100KBPS(rate));
+					       rate);
 	hdr->duration_id = dur;
 
 	return dest_data;
@@ -1054,7 +1063,8 @@ static u8 *b43legacy_generate_probe_resp(struct b43legacy_wldev *dev,
 
 static void b43legacy_write_probe_resp_template(struct b43legacy_wldev *dev,
 						u16 ram_offset,
-						u16 shm_size_offset, u8 rate)
+						u16 shm_size_offset,
+						struct ieee80211_rate *rate)
 {
 	u8 *probe_resp_data;
 	u16 size;
@@ -1069,19 +1079,19 @@ static void b43legacy_write_probe_resp_template(struct b43legacy_wldev *dev,
 	 * all possible basic rates
 	 */
 	b43legacy_write_probe_resp_plcp(dev, 0x31A, size,
-					B43legacy_CCK_RATE_1MB);
+					&b43legacy_b_ratetable[0]);
 	b43legacy_write_probe_resp_plcp(dev, 0x32C, size,
-					B43legacy_CCK_RATE_2MB);
+					&b43legacy_b_ratetable[1]);
 	b43legacy_write_probe_resp_plcp(dev, 0x33E, size,
-					B43legacy_CCK_RATE_5MB);
+					&b43legacy_b_ratetable[2]);
 	b43legacy_write_probe_resp_plcp(dev, 0x350, size,
-					B43legacy_CCK_RATE_11MB);
+					&b43legacy_b_ratetable[3]);
 
 	size = min((size_t)size,
 		   0x200 - sizeof(struct b43legacy_plcp_hdr6));
 	b43legacy_write_template_common(dev, probe_resp_data,
 					size, ram_offset,
-					shm_size_offset, rate);
+					shm_size_offset, rate->bitrate);
 	kfree(probe_resp_data);
 }
 
@@ -1106,7 +1116,7 @@ static void b43legacy_update_templates(struct b43legacy_wldev *dev)
 	b43legacy_write_beacon_template(dev, 0x468, 0x1A,
 					B43legacy_CCK_RATE_1MB);
 	b43legacy_write_probe_resp_template(dev, 0x268, 0x4A,
-					    B43legacy_CCK_RATE_11MB);
+					    &b43legacy_b_ratetable[0]);
 
 	status = b43legacy_read32(dev, B43legacy_MMIO_MACCMD);
 	status |= 0x03;
@@ -2550,14 +2560,16 @@ static int b43legacy_op_dev_config(struct ieee80211_hw *hw,
 	antenna_rx = b43legacy_antenna_from_ieee80211(conf->antenna_sel_rx);
 
 	mutex_lock(&wl->mutex);
+	dev = wl->current_dev;
+	phy = &dev->phy;
 
 	/* Switch the PHY mode (if necessary). */
-	switch (conf->phymode) {
-	case MODE_IEEE80211B:
-		new_phymode = B43legacy_PHYMODE_B;
-		break;
-	case MODE_IEEE80211G:
-		new_phymode = B43legacy_PHYMODE_G;
+	switch (conf->channel->band) {
+	case IEEE80211_BAND_2GHZ:
+		if (phy->type == B43legacy_PHYTYPE_B)
+			new_phymode = B43legacy_PHYMODE_B;
+		else
+			new_phymode = B43legacy_PHYMODE_G;
 		break;
 	default:
 		B43legacy_WARN_ON(1);
@@ -2565,8 +2577,6 @@ static int b43legacy_op_dev_config(struct ieee80211_hw *hw,
 	err = b43legacy_switch_phymode(wl, new_phymode);
 	if (err)
 		goto out_unlock_mutex;
-	dev = wl->current_dev;
-	phy = &dev->phy;
 
 	/* Disable IRQs while reconfiguring the device.
 	 * This makes it possible to drop the spinlock throughout
@@ -2582,8 +2592,8 @@ static int b43legacy_op_dev_config(struct ieee80211_hw *hw,
 
 	/* Switch to the requested channel.
 	 * The firmware takes care of races with the TX handler. */
-	if (conf->channel_val != phy->channel)
-		b43legacy_radio_selectchannel(dev, conf->channel_val, 0);
+	if (conf->channel->hw_value != phy->channel)
+		b43legacy_radio_selectchannel(dev, conf->channel->hw_value, 0);
 
 	/* Enable/Disable ShortSlot timing. */
 	if ((!!(conf->flags & IEEE80211_CONF_SHORT_SLOT_TIME))
@@ -3398,48 +3408,19 @@ static int b43legacy_setup_modes(struct b43legacy_wldev *dev,
 				 int have_gphy)
 {
 	struct ieee80211_hw *hw = dev->wl->hw;
-	struct ieee80211_hw_mode *mode;
 	struct b43legacy_phy *phy = &dev->phy;
-	int cnt = 0;
-	int err;
 
 	phy->possible_phymodes = 0;
-	for (; 1; cnt++) {
-		if (have_bphy) {
-			B43legacy_WARN_ON(cnt >= B43legacy_MAX_PHYHWMODES);
-			mode = &phy->hwmodes[cnt];
-
-			mode->mode = MODE_IEEE80211B;
-			mode->num_channels = b43legacy_bg_chantable_size;
-			mode->channels = b43legacy_bg_chantable;
-			mode->num_rates = b43legacy_b_ratetable_size;
-			mode->rates = b43legacy_b_ratetable;
-			err = ieee80211_register_hwmode(hw, mode);
-			if (err)
-				return err;
-
-			phy->possible_phymodes |= B43legacy_PHYMODE_B;
-			have_bphy = 0;
-			continue;
-		}
-		if (have_gphy) {
-			B43legacy_WARN_ON(cnt >= B43legacy_MAX_PHYHWMODES);
-			mode = &phy->hwmodes[cnt];
-
-			mode->mode = MODE_IEEE80211G;
-			mode->num_channels = b43legacy_bg_chantable_size;
-			mode->channels = b43legacy_bg_chantable;
-			mode->num_rates = b43legacy_g_ratetable_size;
-			mode->rates = b43legacy_g_ratetable;
-			err = ieee80211_register_hwmode(hw, mode);
-			if (err)
-				return err;
-
-			phy->possible_phymodes |= B43legacy_PHYMODE_G;
-			have_gphy = 0;
-			continue;
-		}
-		break;
+	if (have_bphy) {
+		hw->wiphy->bands[IEEE80211_BAND_2GHZ] =
+			&b43legacy_band_2GHz_BPHY;
+		phy->possible_phymodes |= B43legacy_PHYMODE_B;
+	}
+
+	if (have_gphy) {
+		hw->wiphy->bands[IEEE80211_BAND_2GHZ] =
+			&b43legacy_band_2GHz_GPHY;
+		phy->possible_phymodes |= B43legacy_PHYMODE_G;
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/b43legacy/xmit.c b/drivers/net/wireless/b43legacy/xmit.c
index d84408a82db9..47e130e9fdf2 100644
--- a/drivers/net/wireless/b43legacy/xmit.c
+++ b/drivers/net/wireless/b43legacy/xmit.c
@@ -37,45 +37,48 @@
 
 
 /* Extract the bitrate out of a CCK PLCP header. */
-static u8 b43legacy_plcp_get_bitrate_cck(struct b43legacy_plcp_hdr6 *plcp)
+static u8 b43legacy_plcp_get_bitrate_idx_cck(struct b43legacy_plcp_hdr6 *plcp)
 {
 	switch (plcp->raw[0]) {
 	case 0x0A:
-		return B43legacy_CCK_RATE_1MB;
+		return 0;
 	case 0x14:
-		return B43legacy_CCK_RATE_2MB;
+		return 1;
 	case 0x37:
-		return B43legacy_CCK_RATE_5MB;
+		return 2;
 	case 0x6E:
-		return B43legacy_CCK_RATE_11MB;
+		return 3;
 	}
 	B43legacy_BUG_ON(1);
-	return 0;
+	return -1;
 }
 
 /* Extract the bitrate out of an OFDM PLCP header. */
-static u8 b43legacy_plcp_get_bitrate_ofdm(struct b43legacy_plcp_hdr6 *plcp)
+static u8 b43legacy_plcp_get_bitrate_idx_ofdm(struct b43legacy_plcp_hdr6 *plcp,
+					      bool aphy)
 {
+	int base = aphy ? 0 : 4;
+
 	switch (plcp->raw[0] & 0xF) {
 	case 0xB:
-		return B43legacy_OFDM_RATE_6MB;
+		return base + 0;
 	case 0xF:
-		return B43legacy_OFDM_RATE_9MB;
+		return base + 1;
 	case 0xA:
-		return B43legacy_OFDM_RATE_12MB;
+		return base + 2;
 	case 0xE:
-		return B43legacy_OFDM_RATE_18MB;
+		return base + 3;
 	case 0x9:
-		return B43legacy_OFDM_RATE_24MB;
+		return base + 4;
 	case 0xD:
-		return B43legacy_OFDM_RATE_36MB;
+		return base + 5;
 	case 0x8:
-		return B43legacy_OFDM_RATE_48MB;
+		return base + 6;
 	case 0xC:
-		return B43legacy_OFDM_RATE_54MB;
+		return base + 7;
 	}
 	B43legacy_BUG_ON(1);
-	return 0;
+	return -1;
 }
 
 u8 b43legacy_plcp_get_ratecode_cck(const u8 bitrate)
@@ -192,7 +195,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 	int use_encryption = (!(txctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT));
 	u16 fctl;
 	u8 rate;
-	u8 rate_fb;
+	struct ieee80211_rate *rate_fb;
 	int rate_ofdm;
 	int rate_fb_ofdm;
 	unsigned int plcp_fragment_len;
@@ -204,16 +207,16 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 
 	memset(txhdr, 0, sizeof(*txhdr));
 
-	rate = txctl->tx_rate;
+	rate = txctl->tx_rate->hw_value;
 	rate_ofdm = b43legacy_is_ofdm_rate(rate);
-	rate_fb = (txctl->alt_retry_rate == -1) ? rate : txctl->alt_retry_rate;
-	rate_fb_ofdm = b43legacy_is_ofdm_rate(rate_fb);
+	rate_fb = txctl->alt_retry_rate ? : txctl->tx_rate;
+	rate_fb_ofdm = b43legacy_is_ofdm_rate(rate_fb->hw_value);
 
 	txhdr->mac_frame_ctl = wlhdr->frame_control;
 	memcpy(txhdr->tx_receiver, wlhdr->addr1, 6);
 
 	/* Calculate duration for fallback rate */
-	if ((rate_fb == rate) ||
+	if ((rate_fb->hw_value == rate) ||
 	    (wlhdr->duration_id & cpu_to_le16(0x8000)) ||
 	    (wlhdr->duration_id == cpu_to_le16(0))) {
 		/* If the fallback rate equals the normal rate or the
@@ -221,11 +224,10 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 		 * use the original dur_id field. */
 		txhdr->dur_fb = wlhdr->duration_id;
 	} else {
-		int fbrate_base100kbps = B43legacy_RATE_TO_100KBPS(rate_fb);
 		txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw,
 							 txctl->vif,
 							 fragment_len,
-							 fbrate_base100kbps);
+							 rate_fb);
 	}
 
 	plcp_fragment_len = fragment_len + FCS_LEN;
@@ -266,7 +268,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 				    rate);
 	b43legacy_generate_plcp_hdr((struct b43legacy_plcp_hdr4 *)
 				    (&txhdr->plcp_fb), plcp_fragment_len,
-				    rate_fb);
+				    rate_fb->hw_value);
 
 	/* PHY TX Control word */
 	if (rate_ofdm)
@@ -310,7 +312,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev,
 		int rts_rate_ofdm;
 		int rts_rate_fb_ofdm;
 
-		rts_rate = txctl->rts_cts_rate;
+		rts_rate = txctl->rts_cts_rate->hw_value;
 		rts_rate_ofdm = b43legacy_is_ofdm_rate(rts_rate);
 		rts_rate_fb = b43legacy_calc_fallback_rate(rts_rate);
 		rts_rate_fb_ofdm = b43legacy_is_ofdm_rate(rts_rate_fb);
@@ -536,10 +538,11 @@ void b43legacy_rx(struct b43legacy_wldev *dev,
 				      (phystat3 & B43legacy_RX_PHYST3_TRSTATE));
 	status.noise = dev->stats.link_noise;
 	status.signal = (jssi * 100) / B43legacy_RX_MAX_SSI;
+	/* change to support A PHY */
 	if (phystat0 & B43legacy_RX_PHYST0_OFDM)
-		status.rate = b43legacy_plcp_get_bitrate_ofdm(plcp);
+		status.rate_idx = b43legacy_plcp_get_bitrate_idx_ofdm(plcp, false);
 	else
-		status.rate = b43legacy_plcp_get_bitrate_cck(plcp);
+		status.rate_idx = b43legacy_plcp_get_bitrate_idx_cck(plcp);
 	status.antenna = !!(phystat0 & B43legacy_RX_PHYST0_ANT);
 
 	/*
@@ -564,14 +567,9 @@ void b43legacy_rx(struct b43legacy_wldev *dev,
 		  B43legacy_RX_CHAN_ID_SHIFT;
 	switch (chanstat & B43legacy_RX_CHAN_PHYTYPE) {
 	case B43legacy_PHYTYPE_B:
-		status.phymode = MODE_IEEE80211B;
-		status.freq = chanid + 2400;
-		status.channel = b43legacy_freq_to_channel_bg(chanid + 2400);
-		break;
 	case B43legacy_PHYTYPE_G:
-		status.phymode = MODE_IEEE80211G;
+		status.band = IEEE80211_BAND_2GHZ;
 		status.freq = chanid + 2400;
-		status.channel = b43legacy_freq_to_channel_bg(chanid + 2400);
 		break;
 	default:
 		b43legacywarn(dev->wl, "Unexpected value for chanstat (0x%X)\n",
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
index 80d31ae51e77..f018ce464d97 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c
@@ -100,14 +100,6 @@ static struct iwl3945_tpt_entry iwl3945_tpt_table_a[] = {
 	{-89, IWL_RATE_6M_INDEX}
 };
 
-static struct iwl3945_tpt_entry iwl3945_tpt_table_b[] = {
-	{-86, IWL_RATE_11M_INDEX},
-	{-88, IWL_RATE_5M_INDEX},
-	{-90, IWL_RATE_2M_INDEX},
-	{-92, IWL_RATE_1M_INDEX}
-
-};
-
 static struct iwl3945_tpt_entry iwl3945_tpt_table_g[] = {
 	{-60, IWL_RATE_54M_INDEX},
 	{-64, IWL_RATE_48M_INDEX},
@@ -129,7 +121,7 @@ static struct iwl3945_tpt_entry iwl3945_tpt_table_g[] = {
 #define IWL_RATE_MIN_SUCCESS_TH       8
 #define IWL_RATE_DECREASE_TH       1920
 
-static u8 iwl3945_get_rate_index_by_rssi(s32 rssi, u8 mode)
+static u8 iwl3945_get_rate_index_by_rssi(s32 rssi, enum ieee80211_band band)
 {
 	u32 index = 0;
 	u32 table_size = 0;
@@ -138,21 +130,19 @@ static u8 iwl3945_get_rate_index_by_rssi(s32 rssi, u8 mode)
 	if ((rssi < IWL_MIN_RSSI_VAL) || (rssi > IWL_MAX_RSSI_VAL))
 		rssi = IWL_MIN_RSSI_VAL;
 
-	switch (mode) {
-	case MODE_IEEE80211G:
+	switch (band) {
+	case IEEE80211_BAND_2GHZ:
 		tpt_table = iwl3945_tpt_table_g;
 		table_size = ARRAY_SIZE(iwl3945_tpt_table_g);
 		break;
 
-	case MODE_IEEE80211A:
+	case IEEE80211_BAND_5GHZ:
 		tpt_table = iwl3945_tpt_table_a;
 		table_size = ARRAY_SIZE(iwl3945_tpt_table_a);
 		break;
 
 	default:
-	case MODE_IEEE80211B:
-		tpt_table = iwl3945_tpt_table_b;
-		table_size = ARRAY_SIZE(iwl3945_tpt_table_b);
+		BUG();
 		break;
 	}
 
@@ -340,17 +330,17 @@ static void rs_rate_init(void *priv_rate, void *priv_sta,
 	 * after assoc.. */
 
 	for (i = IWL_RATE_COUNT - 1; i >= 0; i--) {
-		if (sta->supp_rates & (1 << i)) {
-			sta->txrate = i;
+		if (sta->supp_rates[local->hw.conf.channel->band] & (1 << i)) {
+			sta->txrate_idx = i;
 			break;
 		}
 	}
 
-	sta->last_txrate = sta->txrate;
+	sta->last_txrate_idx = sta->txrate_idx;
 
-	/* For MODE_IEEE80211A mode it start at IWL_FIRST_OFDM_RATE */
-        if (local->hw.conf.phymode == MODE_IEEE80211A)
-                sta->last_txrate += IWL_FIRST_OFDM_RATE;
+	/* For 5 GHz band it start at IWL_FIRST_OFDM_RATE */
+	if (local->hw.conf.channel->band == IEEE80211_BAND_5GHZ)
+		sta->last_txrate_idx += IWL_FIRST_OFDM_RATE;
 
 	IWL_DEBUG_RATE("leave\n");
 }
@@ -429,17 +419,19 @@ static int rs_adjust_next_rate(struct iwl3945_priv *priv, int rate)
 {
 	int next_rate = iwl3945_get_prev_ieee_rate(rate);
 
-	switch (priv->phymode) {
-	case MODE_IEEE80211A:
+	switch (priv->band) {
+	case IEEE80211_BAND_5GHZ:
 		if (rate == IWL_RATE_12M_INDEX)
 			next_rate = IWL_RATE_9M_INDEX;
 		else if (rate == IWL_RATE_6M_INDEX)
 			next_rate = IWL_RATE_6M_INDEX;
 		break;
+/* XXX cannot be invoked in current mac80211 so not a regression
 	case MODE_IEEE80211B:
 		if (rate == IWL_RATE_11M_INDEX_TABLE)
 			next_rate = IWL_RATE_5M_INDEX_TABLE;
 		break;
+ */
 	default:
 		break;
 	}
@@ -465,15 +457,17 @@ static void rs_tx_status(void *priv_rate,
 	struct iwl3945_priv *priv = (struct iwl3945_priv *)priv_rate;
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct iwl3945_rs_sta *rs_sta;
+	struct ieee80211_supported_band *sband;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
 	IWL_DEBUG_RATE("enter\n");
 
 	retries = tx_resp->retry_count;
 
-	first_index = tx_resp->control.tx_rate;
+	first_index = &sband->bitrates[0] - tx_resp->control.tx_rate;
 	if ((first_index < 0) || (first_index >= IWL_RATE_COUNT)) {
-		IWL_DEBUG_RATE("leave: Rate out of bounds: %0x for %d\n",
-			       tx_resp->control.tx_rate, first_index);
+		IWL_DEBUG_RATE("leave: Rate out of bounds: %d\n", first_index);
 		return;
 	}
 
@@ -561,14 +555,14 @@ static void rs_tx_status(void *priv_rate,
 }
 
 static u16 iwl3945_get_adjacent_rate(struct iwl3945_rs_sta *rs_sta,
-				 u8 index, u16 rate_mask, int phymode)
+				 u8 index, u16 rate_mask, enum ieee80211_band band)
 {
 	u8 high = IWL_RATE_INVALID;
 	u8 low = IWL_RATE_INVALID;
 
 	/* 802.11A walks to the next literal adjacent rate in
 	 * the rate table */
-	if (unlikely(phymode == MODE_IEEE80211A)) {
+	if (unlikely(band == IEEE80211_BAND_5GHZ)) {
 		int i;
 		u32 mask;
 
@@ -639,7 +633,8 @@ static u16 iwl3945_get_adjacent_rate(struct iwl3945_rs_sta *rs_sta,
  *
  */
 static void rs_get_rate(void *priv_rate, struct net_device *dev,
-			struct ieee80211_hw_mode *mode, struct sk_buff *skb,
+			struct ieee80211_supported_band *band,
+			struct sk_buff *skb,
 			struct rate_selection *sel)
 {
 	u8 low = IWL_RATE_INVALID;
@@ -672,16 +667,16 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 	    is_multicast_ether_addr(hdr->addr1) ||
 	    !sta || !sta->rate_ctrl_priv) {
 		IWL_DEBUG_RATE("leave: No STA priv data to update!\n");
-		sel->rate = rate_lowest(local, local->oper_hw_mode, sta);
+		sel->rate = rate_lowest(local, band, sta);
 		if (sta)
 			sta_info_put(sta);
 		return;
 	}
 
-	rate_mask = sta->supp_rates;
-	index = min(sta->last_txrate & 0xffff, IWL_RATE_COUNT - 1);
+	rate_mask = sta->supp_rates[band->band];
+	index = min(sta->last_txrate_idx & 0xffff, IWL_RATE_COUNT - 1);
 
-	if (priv->phymode == (u8) MODE_IEEE80211A)
+	if (priv->band == IEEE80211_BAND_5GHZ)
 		rate_mask = rate_mask << IWL_FIRST_OFDM_RATE;
 
 	rs_sta = (void *)sta->rate_ctrl_priv;
@@ -732,7 +727,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 	current_tpt = window->average_tpt;
 
 	high_low = iwl3945_get_adjacent_rate(rs_sta, index, rate_mask,
-					 local->hw.conf.phymode);
+					     band->band);
 	low = high_low & 0xff;
 	high = (high_low >> 8) & 0xff;
 
@@ -810,11 +805,11 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 
  out:
 
-	sta->last_txrate = index;
-	if (priv->phymode == (u8) MODE_IEEE80211A)
-		sta->txrate = sta->last_txrate - IWL_FIRST_OFDM_RATE;
+	sta->last_txrate_idx = index;
+	if (priv->band == IEEE80211_BAND_5GHZ)
+		sta->txrate_idx = sta->last_txrate_idx - IWL_FIRST_OFDM_RATE;
 	else
-		sta->txrate = sta->last_txrate;
+		sta->txrate_idx = sta->last_txrate_idx;
 
 	sta_info_put(sta);
 
@@ -945,8 +940,9 @@ void iwl3945_rate_scale_init(struct ieee80211_hw *hw, s32 sta_id)
 	spin_lock_irqsave(&rs_sta->lock, flags);
 
 	rs_sta->tgg = 0;
-	switch (priv->phymode) {
-	case MODE_IEEE80211G:
+	switch (priv->band) {
+	case IEEE80211_BAND_2GHZ:
+		/* TODO: this always does G, not a regression */
 		if (priv->active_rxon.flags & RXON_FLG_TGG_PROTECT_MSK) {
 			rs_sta->tgg = 1;
 			rs_sta->expected_tpt = iwl3945_expected_tpt_g_prot;
@@ -954,14 +950,11 @@ void iwl3945_rate_scale_init(struct ieee80211_hw *hw, s32 sta_id)
 			rs_sta->expected_tpt = iwl3945_expected_tpt_g;
 		break;
 
-	case MODE_IEEE80211A:
+	case IEEE80211_BAND_5GHZ:
 		rs_sta->expected_tpt = iwl3945_expected_tpt_a;
 		break;
-
-	default:
-		IWL_WARNING("Invalid phymode.  Defaulting to 802.11b\n");
-	case MODE_IEEE80211B:
-		rs_sta->expected_tpt = iwl3945_expected_tpt_b;
+	case IEEE80211_NUM_BANDS:
+		BUG();
 		break;
 	}
 
@@ -974,8 +967,7 @@ void iwl3945_rate_scale_init(struct ieee80211_hw *hw, s32 sta_id)
 
 	IWL_DEBUG(IWL_DL_INFO | IWL_DL_RATE, "Network RSSI: %d\n", rssi);
 
-	rs_sta->start_rate =
-			iwl3945_get_rate_index_by_rssi(rssi, priv->phymode);
+	rs_sta->start_rate = iwl3945_get_rate_index_by_rssi(rssi, priv->band);
 
 	IWL_DEBUG_RATE("leave: rssi %d assign rate index: "
 		       "%d (plcp 0x%x)\n", rssi, rs_sta->start_rate,
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index 8d4d91d35fd2..50d927bb2170 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -247,7 +247,7 @@ static void iwl3945_add_radiotap(struct iwl3945_priv *priv,
 	 * the information provided in the skb from the hardware */
 	s8 signal = stats->ssi;
 	s8 noise = 0;
-	int rate = stats->rate;
+	int rate = stats->rate_idx;
 	u64 tsf = stats->mactime;
 	__le16 phy_flags_hw = rx_hdr->phy_flags;
 
@@ -315,7 +315,6 @@ static void iwl3945_add_radiotap(struct iwl3945_priv *priv,
 					  IEEE80211_CHAN_2GHZ),
 			      &iwl3945_rt->rt_chbitmask);
 
-	rate = iwl3945_rate_index_from_plcp(rate);
 	if (rate == -1)
 		iwl3945_rt->rt_rate = 0;
 	else
@@ -387,11 +386,10 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 	struct ieee80211_rx_status stats = {
 		.mactime = le64_to_cpu(rx_end->timestamp),
 		.freq = ieee80211chan2mhz(le16_to_cpu(rx_hdr->channel)),
-		.channel = le16_to_cpu(rx_hdr->channel),
-		.phymode = (rx_hdr->phy_flags & RX_RES_PHY_FLAGS_BAND_24_MSK) ?
-		MODE_IEEE80211G : MODE_IEEE80211A,
+		.band = (rx_hdr->phy_flags & RX_RES_PHY_FLAGS_BAND_24_MSK) ?
+		IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ,
 		.antenna = 0,
-		.rate = rx_hdr->rate,
+		.rate_idx = iwl3945_rate_index_from_plcp(rx_hdr->rate),
 		.flag = 0,
 	};
 	u8 network_packet;
@@ -450,8 +448,6 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 			stats.ssi, stats.noise, stats.signal,
 			rx_stats_sig_avg, rx_stats_noise_diff);
 
-	stats.freq = ieee80211chan2mhz(stats.channel);
-
 	/* can be covered by iwl3945_report_frame() in most cases */
 /*      IWL_DEBUG_RX("RX status: 0x%08X\n", rx_end->status); */
 
@@ -464,8 +460,9 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv,
 		IWL_DEBUG_STATS
 		    ("[%c] %d RSSI: %d Signal: %u, Noise: %u, Rate: %u\n",
 		     network_packet ? '*' : ' ',
-		     stats.channel, stats.ssi, stats.ssi,
-		     stats.ssi, stats.rate);
+		     le16_to_cpu(rx_hdr->channel),
+		     stats.ssi, stats.ssi,
+		     stats.ssi, stats.rate_idx);
 
 	if (iwl3945_debug_level & (IWL_DL_RX))
 		/* Set "1" to report good data frames in groups of 100 */
@@ -689,7 +686,7 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 			      struct ieee80211_hdr *hdr, int sta_id, int tx_id)
 {
 	unsigned long flags;
-	u16 rate_index = min(ctrl->tx_rate & 0xffff, IWL_RATE_COUNT - 1);
+	u16 rate_index = min(ctrl->tx_rate->hw_value & 0xffff, IWL_RATE_COUNT - 1);
 	u16 rate_mask;
 	int rate;
 	u8 rts_retry_limit;
@@ -1552,14 +1549,14 @@ int iwl3945_hw_reg_send_txpower(struct iwl3945_priv *priv)
 		.channel = priv->active_rxon.channel,
 	};
 
-	txpower.band = (priv->phymode == MODE_IEEE80211A) ? 0 : 1;
+	txpower.band = (priv->band == IEEE80211_BAND_5GHZ) ? 0 : 1;
 	ch_info = iwl3945_get_channel_info(priv,
-				       priv->phymode,
+				       priv->band,
 				       le16_to_cpu(priv->active_rxon.channel));
 	if (!ch_info) {
 		IWL_ERROR
 		    ("Failed to get channel info for channel %d [%d]\n",
-		     le16_to_cpu(priv->active_rxon.channel), priv->phymode);
+		     le16_to_cpu(priv->active_rxon.channel), priv->band);
 		return -EINVAL;
 	}
 
@@ -2241,8 +2238,8 @@ int iwl3945_init_hw_rate_table(struct iwl3945_priv *priv)
 		table[index].next_rate_index = iwl3945_rates[prev_index].table_rs_index;
 	}
 
-	switch (priv->phymode) {
-	case MODE_IEEE80211A:
+	switch (priv->band) {
+	case IEEE80211_BAND_5GHZ:
 		IWL_DEBUG_RATE("Select A mode rate scale\n");
 		/* If one of the following CCK rates is used,
 		 * have it fall back to the 6M OFDM rate */
@@ -2257,8 +2254,8 @@ int iwl3945_init_hw_rate_table(struct iwl3945_priv *priv)
 		    iwl3945_rates[IWL_FIRST_OFDM_RATE].table_rs_index;
 		break;
 
-	case MODE_IEEE80211B:
-		IWL_DEBUG_RATE("Select B mode rate scale\n");
+	case IEEE80211_BAND_2GHZ:
+		IWL_DEBUG_RATE("Select B/G mode rate scale\n");
 		/* If an OFDM rate is used, have it fall back to the
 		 * 1M CCK rates */
 		for (i = IWL_RATE_6M_INDEX_TABLE; i <= IWL_RATE_54M_INDEX_TABLE; i++)
@@ -2269,7 +2266,7 @@ int iwl3945_init_hw_rate_table(struct iwl3945_priv *priv)
 		break;
 
 	default:
-		IWL_DEBUG_RATE("Select G mode rate scale\n");
+		WARN_ON(1);
 		break;
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.h b/drivers/net/wireless/iwlwifi/iwl-3945.h
index 1da14f9bbe0f..1beb5b676797 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.h
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.h
@@ -195,7 +195,7 @@ struct iwl3945_channel_info {
 
 	u8 group_index;	  /* 0-4, maps channel to group1/2/3/4/5 */
 	u8 band_index;	  /* 0-4, maps channel to band1/2/3/4/5 */
-	u8 phymode;	  /* MODE_IEEE80211{A,B,G} */
+	enum ieee80211_band band;
 
 	/* Radio/DSP gain settings for each "normal" data Tx rate.
 	 * These include, in addition to RF and DSP gain, a few fields for
@@ -699,14 +699,14 @@ struct iwl3945_priv {
 	struct list_head free_frames;
 	int frames_count;
 
-	u8 phymode;
+	enum ieee80211_band band;
 	int alloc_rxb_skb;
 	bool add_radiotap;
 
 	void (*rx_handlers[REPLY_MAX])(struct iwl3945_priv *priv,
 				       struct iwl3945_rx_mem_buffer *rxb);
 
-	const struct ieee80211_hw_mode *modes;
+	struct ieee80211_supported_band bands[IEEE80211_NUM_BANDS];
 
 #ifdef CONFIG_IWL3945_SPECTRUM_MEASUREMENT
 	/* spectrum measurement report caching */
@@ -937,13 +937,12 @@ static inline int is_channel_radar(const struct iwl3945_channel_info *ch_info)
 
 static inline u8 is_channel_a_band(const struct iwl3945_channel_info *ch_info)
 {
-	return ch_info->phymode == MODE_IEEE80211A;
+	return ch_info->band == IEEE80211_BAND_5GHZ;
 }
 
 static inline u8 is_channel_bg_band(const struct iwl3945_channel_info *ch_info)
 {
-	return ((ch_info->phymode == MODE_IEEE80211B) ||
-		(ch_info->phymode == MODE_IEEE80211G));
+	return ch_info->band == IEEE80211_BAND_2GHZ;
 }
 
 static inline int is_channel_passive(const struct iwl3945_channel_info *ch)
@@ -967,7 +966,7 @@ static inline int iwl3945_rate_index_from_plcp(int plcp)
 }
 
 extern const struct iwl3945_channel_info *iwl3945_get_channel_info(
-	const struct iwl3945_priv *priv, int phymode, u16 channel);
+	const struct iwl3945_priv *priv, enum ieee80211_band band, u16 channel);
 
 /* Requires full declaration of iwl3945_priv before including */
 #include "iwl-3945-io.h"
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
index 660671f17a3b..48a6a85355ec 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
@@ -139,7 +139,7 @@ struct iwl4965_lq_sta {
 	u8 valid_antenna;
 	u8 is_green;
 	u8 is_dup;
-	u8 phymode;
+	enum ieee80211_band band;
 	u8 ibss_sta_added;
 
 	/* The following are bitmaps of rates; IWL_RATE_6M_MASK, etc. */
@@ -563,7 +563,8 @@ static void rs_mcs_from_tbl(struct iwl4965_rate *mcs_rate,
  * fill "search" or "active" tx mode table.
  */
 static int rs_get_tbl_info_from_mcs(const struct iwl4965_rate *mcs_rate,
-				    int phymode, struct iwl4965_scale_tbl_info *tbl,
+				    enum ieee80211_band band,
+				    struct iwl4965_scale_tbl_info *tbl,
 				    int *rate_idx)
 {
 	int index;
@@ -588,7 +589,7 @@ static int rs_get_tbl_info_from_mcs(const struct iwl4965_rate *mcs_rate,
 			tbl->lq_type = LQ_NONE;
 		else {
 
-			if (phymode == MODE_IEEE80211A)
+			if (band == IEEE80211_BAND_5GHZ)
 				tbl->lq_type = LQ_A;
 			else
 				tbl->lq_type = LQ_G;
@@ -766,7 +767,7 @@ static void rs_get_lower_rate(struct iwl4965_lq_sta *lq_sta,
 	if (!is_legacy(tbl->lq_type) && (!ht_possible || !scale_index)) {
 		switch_to_legacy = 1;
 		scale_index = rs_ht_to_legacy[scale_index];
-		if (lq_sta->phymode == MODE_IEEE80211A)
+		if (lq_sta->band == IEEE80211_BAND_5GHZ)
 			tbl->lq_type = LQ_A;
 		else
 			tbl->lq_type = LQ_G;
@@ -784,7 +785,7 @@ static void rs_get_lower_rate(struct iwl4965_lq_sta *lq_sta,
 	/* Mask with station rate restriction */
 	if (is_legacy(tbl->lq_type)) {
 		/* supp_rates has no CCK bits in A mode */
-		if (lq_sta->phymode == (u8) MODE_IEEE80211A)
+		if (lq_sta->band == IEEE80211_BAND_5GHZ)
 			rate_mask  = (u16)(rate_mask &
 			   (lq_sta->supp_rates << IWL_FIRST_OFDM_RATE));
 		else
@@ -883,9 +884,9 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	search_win = (struct iwl4965_rate_scale_data *)
 	    &(search_tbl->win[0]);
 
-	tx_mcs.rate_n_flags = tx_resp->control.tx_rate;
+	tx_mcs.rate_n_flags = tx_resp->control.tx_rate->hw_value;
 
-	rs_get_tbl_info_from_mcs(&tx_mcs, priv->phymode,
+	rs_get_tbl_info_from_mcs(&tx_mcs, priv->band,
 				  &tbl_type, &rs_index);
 	if ((rs_index < 0) || (rs_index >= IWL_RATE_COUNT)) {
 		IWL_DEBUG_RATE("bad rate index at: %d rate 0x%X\n",
@@ -918,7 +919,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 		 * Each tx attempt steps one entry deeper in the rate table. */
 		tx_mcs.rate_n_flags =
 		    le32_to_cpu(table->rs_table[index].rate_n_flags);
-		rs_get_tbl_info_from_mcs(&tx_mcs, priv->phymode,
+		rs_get_tbl_info_from_mcs(&tx_mcs, priv->band,
 					  &tbl_type, &rs_index);
 
 		/* If type matches "search" table,
@@ -959,12 +960,12 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	 * else look up the rate that was, finally, successful.
 	 */
 	if (!tx_resp->retry_count)
-		tx_mcs.rate_n_flags = tx_resp->control.tx_rate;
+		tx_mcs.rate_n_flags = tx_resp->control.tx_rate->hw_value;
 	else
 		tx_mcs.rate_n_flags =
 			le32_to_cpu(table->rs_table[index].rate_n_flags);
 
-	rs_get_tbl_info_from_mcs(&tx_mcs, priv->phymode,
+	rs_get_tbl_info_from_mcs(&tx_mcs, priv->band,
 				  &tbl_type, &rs_index);
 
 	/* Update frame history window with "success" if Tx got ACKed ... */
@@ -1801,7 +1802,7 @@ static void rs_rate_scale_perform(struct iwl4965_priv *priv,
 	is_green = lq_sta->is_green;
 
 	/* current tx rate */
-	index = sta->last_txrate;
+	index = sta->last_txrate_idx;
 
 	IWL_DEBUG_RATE("Rate scale index %d for type %d\n", index,
 		       tbl->lq_type);
@@ -1814,7 +1815,7 @@ static void rs_rate_scale_perform(struct iwl4965_priv *priv,
 
 	/* mask with station rate restriction */
 	if (is_legacy(tbl->lq_type)) {
-		if (lq_sta->phymode == (u8) MODE_IEEE80211A)
+		if (lq_sta->band == IEEE80211_BAND_5GHZ)
 			/* supp_rates has no CCK bits in A mode */
 			rate_scale_index_msk = (u16) (rate_mask &
 				(lq_sta->supp_rates << IWL_FIRST_OFDM_RATE));
@@ -2134,15 +2135,15 @@ static void rs_rate_scale_perform(struct iwl4965_priv *priv,
 out:
 	rs_mcs_from_tbl(&tbl->current_rate, tbl, index, is_green);
 	i = index;
-	sta->last_txrate = i;
+	sta->last_txrate_idx = i;
 
-	/* sta->txrate is an index to A mode rates which start
+	/* sta->txrate_idx is an index to A mode rates which start
 	 * at IWL_FIRST_OFDM_RATE
 	 */
-	if (lq_sta->phymode == (u8) MODE_IEEE80211A)
-		sta->txrate = i - IWL_FIRST_OFDM_RATE;
+	if (lq_sta->band == IEEE80211_BAND_5GHZ)
+		sta->txrate_idx = i - IWL_FIRST_OFDM_RATE;
 	else
-		sta->txrate = i;
+		sta->txrate_idx = i;
 
 	return;
 }
@@ -2164,7 +2165,7 @@ static void rs_initialize_lq(struct iwl4965_priv *priv,
 		goto out;
 
 	lq_sta = (struct iwl4965_lq_sta *)sta->rate_ctrl_priv;
-	i = sta->last_txrate;
+	i = sta->last_txrate_idx;
 
 	if ((lq_sta->lq.sta_id == 0xff) &&
 	    (priv->iw_mode == IEEE80211_IF_TYPE_IBSS))
@@ -2188,7 +2189,7 @@ static void rs_initialize_lq(struct iwl4965_priv *priv,
 		mcs_rate.rate_n_flags |= RATE_MCS_CCK_MSK;
 
 	tbl->antenna_type = ANT_AUX;
-	rs_get_tbl_info_from_mcs(&mcs_rate, priv->phymode, tbl, &rate_idx);
+	rs_get_tbl_info_from_mcs(&mcs_rate, priv->band, tbl, &rate_idx);
 	if (!rs_is_ant_connected(priv->valid_antenna, tbl->antenna_type))
 	    rs_toggle_antenna(&mcs_rate, tbl);
 
@@ -2202,7 +2203,8 @@ static void rs_initialize_lq(struct iwl4965_priv *priv,
 }
 
 static void rs_get_rate(void *priv_rate, struct net_device *dev,
-			struct ieee80211_hw_mode *mode, struct sk_buff *skb,
+			struct ieee80211_supported_band *sband,
+			struct sk_buff *skb,
 			struct rate_selection *sel)
 {
 
@@ -2224,14 +2226,14 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 	fc = le16_to_cpu(hdr->frame_control);
 	if (!ieee80211_is_data(fc) || is_multicast_ether_addr(hdr->addr1) ||
 	    !sta || !sta->rate_ctrl_priv) {
-		sel->rate = rate_lowest(local, local->oper_hw_mode, sta);
+		sel->rate = rate_lowest(local, sband, sta);
 		if (sta)
 			sta_info_put(sta);
 		return;
 	}
 
 	lq_sta = (struct iwl4965_lq_sta *)sta->rate_ctrl_priv;
-	i = sta->last_txrate;
+	i = sta->last_txrate_idx;
 
 	if ((priv->iw_mode == IEEE80211_IF_TYPE_IBSS) &&
 	    !lq_sta->ibss_sta_added) {
@@ -2256,7 +2258,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 
  done:
 	if ((i < 0) || (i > IWL_RATE_COUNT)) {
-		sel->rate = rate_lowest(local, local->oper_hw_mode, sta);
+		sel->rate = rate_lowest(local, sband, sta);
 		return;
 	}
 	sta_info_put(sta);
@@ -2291,13 +2293,15 @@ static void rs_rate_init(void *priv_rate, void *priv_sta,
 {
 	int i, j;
 	struct ieee80211_conf *conf = &local->hw.conf;
-	struct ieee80211_hw_mode *mode = local->oper_hw_mode;
+	struct ieee80211_supported_band *sband;
 	struct iwl4965_priv *priv = (struct iwl4965_priv *)priv_rate;
 	struct iwl4965_lq_sta *lq_sta = priv_sta;
 
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
 	lq_sta->flush_timer = 0;
-	lq_sta->supp_rates = sta->supp_rates;
-	sta->txrate = 3;
+	lq_sta->supp_rates = sta->supp_rates[sband->band];
+	sta->txrate_idx = 3;
 	for (j = 0; j < LQ_SIZE; j++)
 		for (i = 0; i < IWL_RATE_COUNT; i++)
 			rs_rate_scale_clear_window(&(lq_sta->lq_info[j].win[i]));
@@ -2332,15 +2336,15 @@ static void rs_rate_init(void *priv_rate, void *priv_sta,
 	}
 
 	/* Find highest tx rate supported by hardware and destination station */
-	for (i = 0; i < mode->num_rates; i++) {
-		if ((sta->supp_rates & BIT(i)) &&
-		    (mode->rates[i].flags & IEEE80211_RATE_SUPPORTED))
-			sta->txrate = i;
-	}
-	sta->last_txrate = sta->txrate;
+	for (i = 0; i < sband->n_bitrates; i++)
+		if (sta->supp_rates[sband->band] & BIT(i))
+			sta->txrate_idx = i;
+
+	sta->last_txrate_idx = sta->txrate_idx;
+	/* WTF is with this bogus comment? A doesn't have cck rates */
 	/* For MODE_IEEE80211A, cck rates are at end of rate table */
-	if (local->hw.conf.phymode == MODE_IEEE80211A)
-		sta->last_txrate += IWL_FIRST_OFDM_RATE;
+	if (local->hw.conf.channel->band == IEEE80211_BAND_5GHZ)
+		sta->last_txrate_idx += IWL_FIRST_OFDM_RATE;
 
 	lq_sta->is_dup = 0;
 	lq_sta->valid_antenna = priv->valid_antenna;
@@ -2349,7 +2353,7 @@ static void rs_rate_init(void *priv_rate, void *priv_sta,
 	lq_sta->active_rate = priv->active_rate;
 	lq_sta->active_rate &= ~(0x1000);
 	lq_sta->active_rate_basic = priv->active_rate_basic;
-	lq_sta->phymode = priv->phymode;
+	lq_sta->band = priv->band;
 #ifdef CONFIG_IWL4965_HT
 	/*
 	 * active_siso_rate mask includes 9 MBits (bit 5), and CCK (bits 0-3),
@@ -2401,7 +2405,7 @@ static void rs_fill_link_cmd(struct iwl4965_lq_sta *lq_sta,
 	rs_dbgfs_set_mcs(lq_sta, tx_mcs, index);
 
 	/* Interpret rate_n_flags */
-	rs_get_tbl_info_from_mcs(tx_mcs, lq_sta->phymode,
+	rs_get_tbl_info_from_mcs(tx_mcs, lq_sta->band,
 				  &tbl_type, &rate_idx);
 
 	/* How many times should we repeat the initial rate? */
@@ -2455,7 +2459,7 @@ static void rs_fill_link_cmd(struct iwl4965_lq_sta *lq_sta,
 			index++;
 		}
 
-		rs_get_tbl_info_from_mcs(&new_rate, lq_sta->phymode, &tbl_type,
+		rs_get_tbl_info_from_mcs(&new_rate, lq_sta->band, &tbl_type,
 						&rate_idx);
 
 		/* Indicate to uCode which entries might be MIMO.
@@ -2542,7 +2546,7 @@ static void rs_dbgfs_set_mcs(struct iwl4965_lq_sta *lq_sta,
 {
 	u32 base_rate;
 
-	if (lq_sta->phymode == (u8) MODE_IEEE80211A)
+	if (lq_sta->band == IEEE80211_BAND_5GHZ)
 		base_rate = 0x800D;
 	else
 		base_rate = 0x820A;
@@ -2802,7 +2806,7 @@ int iwl4965_fill_rs_info(struct ieee80211_hw *hw, char *buf, u8 sta_id)
 
 	cnt += sprintf(&buf[cnt], "\nrate scale type %d antenna %d "
 			 "active_search %d rate index %d\n", lq_type, antenna,
-			 lq_sta->search_better_tbl, sta->last_txrate);
+			 lq_sta->search_better_tbl, sta->last_txrate_idx);
 
 	sta_info_put(sta);
 	return cnt;
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index 0bded8570275..a89439320498 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -339,14 +339,15 @@ static int iwl4965_kw_alloc(struct iwl4965_priv *priv)
  *
  * Does not set up a command, or touch hardware.
  */
-int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv, int phymode, u16 channel,
+int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv,
+			      enum ieee80211_band band, u16 channel,
 			      const struct iwl4965_eeprom_channel *eeprom_ch,
 			      u8 fat_extension_channel)
 {
 	struct iwl4965_channel_info *ch_info;
 
 	ch_info = (struct iwl4965_channel_info *)
-			iwl4965_get_channel_info(priv, phymode, channel);
+			iwl4965_get_channel_info(priv, band, channel);
 
 	if (!is_channel_valid(ch_info))
 		return -1;
@@ -1939,11 +1940,12 @@ static s32 iwl4965_get_voltage_compensation(s32 eeprom_voltage,
 }
 
 static const struct iwl4965_channel_info *
-iwl4965_get_channel_txpower_info(struct iwl4965_priv *priv, u8 phymode, u16 channel)
+iwl4965_get_channel_txpower_info(struct iwl4965_priv *priv,
+				 enum ieee80211_band band, u16 channel)
 {
 	const struct iwl4965_channel_info *ch_info;
 
-	ch_info = iwl4965_get_channel_info(priv, phymode, channel);
+	ch_info = iwl4965_get_channel_info(priv, band, channel);
 
 	if (!is_channel_valid(ch_info))
 		return NULL;
@@ -2392,7 +2394,7 @@ static int iwl4965_fill_txpower_tbl(struct iwl4965_priv *priv, u8 band, u16 chan
 
 	/* Get current (RXON) channel, band, width */
 	ch_info =
-		iwl4965_get_channel_txpower_info(priv, priv->phymode, channel);
+		iwl4965_get_channel_txpower_info(priv, priv->band, channel);
 
 	IWL_DEBUG_TXPOWER("chan %d band %d is_fat %d\n", channel, band,
 			  is_fat);
@@ -2619,8 +2621,7 @@ int iwl4965_hw_reg_send_txpower(struct iwl4965_priv *priv)
 		return -EAGAIN;
 	}
 
-	band = ((priv->phymode == MODE_IEEE80211B) ||
-		(priv->phymode == MODE_IEEE80211G));
+	band = priv->band == IEEE80211_BAND_2GHZ;
 
 	is_fat =  is_fat_channel(priv->active_rxon.flags);
 
@@ -2650,10 +2651,9 @@ int iwl4965_hw_channel_switch(struct iwl4965_priv *priv, u16 channel)
 	struct iwl4965_channel_switch_cmd cmd = { 0 };
 	const struct iwl4965_channel_info *ch_info;
 
-	band = ((priv->phymode == MODE_IEEE80211B) ||
-		(priv->phymode == MODE_IEEE80211G));
+	band = priv->band == IEEE80211_BAND_2GHZ;
 
-	ch_info = iwl4965_get_channel_info(priv, priv->phymode, channel);
+	ch_info = iwl4965_get_channel_info(priv, priv->band, channel);
 
 	is_fat = is_fat_channel(priv->staging_rxon.flags);
 
@@ -2698,7 +2698,7 @@ void iwl4965_hw_build_tx_cmd_rate(struct iwl4965_priv *priv,
 	u16 fc = le16_to_cpu(hdr->frame_control);
 	u8 rate_plcp;
 	u16 rate_flags = 0;
-	int rate_idx = min(ctrl->tx_rate & 0xffff, IWL_RATE_COUNT - 1);
+	int rate_idx = min(ctrl->tx_rate->hw_value & 0xffff, IWL_RATE_COUNT - 1);
 
 	rate_plcp = iwl4965_rates[rate_idx].plcp;
 
@@ -3178,7 +3178,7 @@ static void iwl4965_add_radiotap(struct iwl4965_priv *priv,
 {
 	s8 signal = stats->ssi;
 	s8 noise = 0;
-	int rate = stats->rate;
+	int rate = stats->rate_idx;
 	u64 tsf = stats->mactime;
 	__le16 phy_flags_hw = rx_start->phy_flags;
 	struct iwl4965_rt_rx_hdr {
@@ -3246,7 +3246,6 @@ static void iwl4965_add_radiotap(struct iwl4965_priv *priv,
 					  IEEE80211_CHAN_2GHZ),
 			      &iwl4965_rt->rt_chbitmask);
 
-	rate = iwl4965_rate_index_from_plcp(rate);
 	if (rate == -1)
 		iwl4965_rt->rt_rate = 0;
 	else
@@ -3542,12 +3541,13 @@ static void iwl4965_rx_reply_rx(struct iwl4965_priv *priv,
 	u16 fc;
 	struct ieee80211_rx_status stats = {
 		.mactime = le64_to_cpu(rx_start->timestamp),
-		.channel = le16_to_cpu(rx_start->channel),
-		.phymode =
+		.freq = ieee80211chan2mhz(le16_to_cpu(rx_start->channel)),
+		.band =
 			(rx_start->phy_flags & RX_RES_PHY_FLAGS_BAND_24_MSK) ?
-			MODE_IEEE80211G : MODE_IEEE80211A,
+			IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ,
 		.antenna = 0,
-		.rate = iwl4965_hw_get_rate(rx_start->rate_n_flags),
+		.rate_idx = iwl4965_hw_get_rate(
+				le32_to_cpu(rx_start->rate_n_flags)),
 		.flag = 0,
 	};
 	u8 network_packet;
@@ -3598,8 +3598,6 @@ static void iwl4965_rx_reply_rx(struct iwl4965_priv *priv,
 
 	priv->ucode_beacon_time = le32_to_cpu(rx_start->beacon_time_stamp);
 
-	stats.freq = ieee80211chan2mhz(stats.channel);
-
 	/* Find max signal strength (dBm) among 3 antenna/receiver chains */
 	stats.ssi = iwl4965_calc_rssi(rx_start);
 
@@ -4185,7 +4183,7 @@ void iwl4965_add_station(struct iwl4965_priv *priv, const u8 *addr, int is_ap)
 	 * all the way down to 1M in IEEE order, and then spin on 1M */
 	if (is_ap)
 		r = IWL_RATE_54M_INDEX;
-	else if (priv->phymode == MODE_IEEE80211A)
+	else if (priv->band == IEEE80211_BAND_5GHZ)
 		r = IWL_RATE_6M_INDEX;
 	else
 		r = IWL_RATE_1M_INDEX;
@@ -4218,12 +4216,13 @@ void iwl4965_add_station(struct iwl4965_priv *priv, const u8 *addr, int is_ap)
 
 #ifdef CONFIG_IWL4965_HT
 
-static u8 iwl4965_is_channel_extension(struct iwl4965_priv *priv, int phymode,
+static u8 iwl4965_is_channel_extension(struct iwl4965_priv *priv,
+				       enum ieee80211_band band,
 				   u16 channel, u8 extension_chan_offset)
 {
 	const struct iwl4965_channel_info *ch_info;
 
-	ch_info = iwl4965_get_channel_info(priv, phymode, channel);
+	ch_info = iwl4965_get_channel_info(priv, band, channel);
 	if (!is_channel_valid(ch_info))
 		return 0;
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.h b/drivers/net/wireless/iwlwifi/iwl-4965.h
index de5c1bf8fc42..cb8f7f2a8d48 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.h
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.h
@@ -206,7 +206,7 @@ struct iwl4965_channel_info {
 
 	u8 group_index;	  /* 0-4, maps channel to group1/2/3/4/5 */
 	u8 band_index;	  /* 0-4, maps channel to band1/2/3/4/5 */
-	u8 phymode;	  /* MODE_IEEE80211{A,B,G} */
+	enum ieee80211_band band;
 
 	/* Radio/DSP gain settings for each "normal" data Tx rate.
 	 * These include, in addition to RF and DSP gain, a few fields for
@@ -764,7 +764,8 @@ extern void iwl4965_update_rate_scaling(struct iwl4965_priv *priv, u8 mode);
 extern void iwl4965_chain_noise_reset(struct iwl4965_priv *priv);
 extern void iwl4965_init_sensitivity(struct iwl4965_priv *priv, u8 flags,
 				     u8 force);
-extern int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv, int phymode,
+extern int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv,
+				enum ieee80211_band band,
 				u16 channel,
 				const struct iwl4965_eeprom_channel *eeprom_ch,
 				u8 fat_extension_channel);
@@ -977,14 +978,14 @@ struct iwl4965_priv {
 	struct list_head free_frames;
 	int frames_count;
 
-	u8 phymode;
+	enum ieee80211_band band;
 	int alloc_rxb_skb;
 	bool add_radiotap;
 
 	void (*rx_handlers[REPLY_MAX])(struct iwl4965_priv *priv,
 				       struct iwl4965_rx_mem_buffer *rxb);
 
-	const struct ieee80211_hw_mode *modes;
+	struct ieee80211_supported_band bands[IEEE80211_NUM_BANDS];
 
 #ifdef CONFIG_IWL4965_SPECTRUM_MEASUREMENT
 	/* spectrum measurement report caching */
@@ -1243,13 +1244,12 @@ static inline int is_channel_radar(const struct iwl4965_channel_info *ch_info)
 
 static inline u8 is_channel_a_band(const struct iwl4965_channel_info *ch_info)
 {
-	return ch_info->phymode == MODE_IEEE80211A;
+	return ch_info->band == IEEE80211_BAND_5GHZ;
 }
 
 static inline u8 is_channel_bg_band(const struct iwl4965_channel_info *ch_info)
 {
-	return ((ch_info->phymode == MODE_IEEE80211B) ||
-		(ch_info->phymode == MODE_IEEE80211G));
+	return ch_info->band == IEEE80211_BAND_2GHZ;
 }
 
 static inline int is_channel_passive(const struct iwl4965_channel_info *ch)
@@ -1263,7 +1263,7 @@ static inline int is_channel_ibss(const struct iwl4965_channel_info *ch)
 }
 
 extern const struct iwl4965_channel_info *iwl4965_get_channel_info(
-	const struct iwl4965_priv *priv, int phymode, u16 channel);
+	const struct iwl4965_priv *priv, enum ieee80211_band band, u16 channel);
 
 /* Requires full declaration of iwl4965_priv before including */
 #include "iwl-4965-io.h"
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 40b71bc2c4a4..57a1d70f2abf 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -116,16 +116,10 @@ static __le16 *ieee80211_get_qos_ctrl(struct ieee80211_hdr *hdr)
 	return NULL;
 }
 
-static const struct ieee80211_hw_mode *iwl3945_get_hw_mode(
-		struct iwl3945_priv *priv, int mode)
+static const struct ieee80211_supported_band *iwl3945_get_band(
+		struct iwl3945_priv *priv, enum ieee80211_band band)
 {
-	int i;
-
-	for (i = 0; i < 3; i++)
-		if (priv->modes[i].mode == mode)
-			return &priv->modes[i];
-
-	return NULL;
+	return priv->hw->wiphy->bands[band];
 }
 
 static int iwl3945_is_empty_essid(const char *essid, int essid_len)
@@ -547,7 +541,7 @@ u8 iwl3945_add_station(struct iwl3945_priv *priv, const u8 *addr, int is_ap, u8
 	station->sta.sta.sta_id = index;
 	station->sta.station_flags = 0;
 
-	if (priv->phymode == MODE_IEEE80211A)
+	if (priv->band == IEEE80211_BAND_5GHZ)
 		rate = IWL_RATE_6M_PLCP;
 	else
 		rate =	IWL_RATE_1M_PLCP;
@@ -894,35 +888,37 @@ int iwl3945_send_statistics_request(struct iwl3945_priv *priv)
 
 /**
  * iwl3945_set_rxon_channel - Set the phymode and channel values in staging RXON
- * @phymode: MODE_IEEE80211A sets to 5.2GHz; all else set to 2.4GHz
- * @channel: Any channel valid for the requested phymode
+ * @band: 2.4 or 5 GHz band
+ * @channel: Any channel valid for the requested band
 
- * In addition to setting the staging RXON, priv->phymode is also set.
+ * In addition to setting the staging RXON, priv->band is also set.
  *
  * NOTE:  Does not commit to the hardware; it sets appropriate bit fields
- * in the staging RXON flag structure based on the phymode
+ * in the staging RXON flag structure based on the band
  */
-static int iwl3945_set_rxon_channel(struct iwl3945_priv *priv, u8 phymode, u16 channel)
+static int iwl3945_set_rxon_channel(struct iwl3945_priv *priv,
+				    enum ieee80211_band band,
+				    u16 channel)
 {
-	if (!iwl3945_get_channel_info(priv, phymode, channel)) {
+	if (!iwl3945_get_channel_info(priv, band, channel)) {
 		IWL_DEBUG_INFO("Could not set channel to %d [%d]\n",
-			       channel, phymode);
+			       channel, band);
 		return -EINVAL;
 	}
 
 	if ((le16_to_cpu(priv->staging_rxon.channel) == channel) &&
-	    (priv->phymode == phymode))
+	    (priv->band == band))
 		return 0;
 
 	priv->staging_rxon.channel = cpu_to_le16(channel);
-	if (phymode == MODE_IEEE80211A)
+	if (band == IEEE80211_BAND_5GHZ)
 		priv->staging_rxon.flags &= ~RXON_FLG_BAND_24G_MSK;
 	else
 		priv->staging_rxon.flags |= RXON_FLG_BAND_24G_MSK;
 
-	priv->phymode = phymode;
+	priv->band = band;
 
-	IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, phymode);
+	IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, band);
 
 	return 0;
 }
@@ -1210,8 +1206,7 @@ static int iwl3945_commit_rxon(struct iwl3945_priv *priv)
 			return -EIO;
 		}
 
-	/* Init the hardware's rate fallback order based on the
-	 * phymode */
+	/* Init the hardware's rate fallback order based on the band */
 	rc = iwl3945_init_hw_rate_table(priv);
 	if (rc) {
 		IWL_ERROR("Error setting HW rate table: %02X\n", rc);
@@ -2461,9 +2456,10 @@ static int iwl3945_set_rxon_hwcrypto(struct iwl3945_priv *priv, int hw_decrypt)
 	return 0;
 }
 
-static void iwl3945_set_flags_for_phymode(struct iwl3945_priv *priv, u8 phymode)
+static void iwl3945_set_flags_for_phymode(struct iwl3945_priv *priv,
+					  enum ieee80211_band band)
 {
-	if (phymode == MODE_IEEE80211A) {
+	if (band == IEEE80211_BAND_5GHZ) {
 		priv->staging_rxon.flags &=
 		    ~(RXON_FLG_BAND_24G_MSK | RXON_FLG_AUTO_DETECT_MSK
 		      | RXON_FLG_CCK_MSK);
@@ -2526,7 +2522,7 @@ static void iwl3945_connection_init_rx_config(struct iwl3945_priv *priv)
 		priv->staging_rxon.flags |= RXON_FLG_SHORT_PREAMBLE_MSK;
 #endif
 
-	ch_info = iwl3945_get_channel_info(priv, priv->phymode,
+	ch_info = iwl3945_get_channel_info(priv, priv->band,
 				       le16_to_cpu(priv->staging_rxon.channel));
 
 	if (!ch_info)
@@ -2542,11 +2538,11 @@ static void iwl3945_connection_init_rx_config(struct iwl3945_priv *priv)
 
 	priv->staging_rxon.channel = cpu_to_le16(ch_info->channel);
 	if (is_channel_a_band(ch_info))
-		priv->phymode = MODE_IEEE80211A;
+		priv->band = IEEE80211_BAND_5GHZ;
 	else
-		priv->phymode = MODE_IEEE80211G;
+		priv->band = IEEE80211_BAND_2GHZ;
 
-	iwl3945_set_flags_for_phymode(priv, priv->phymode);
+	iwl3945_set_flags_for_phymode(priv, priv->band);
 
 	priv->staging_rxon.ofdm_basic_rates =
 	    (IWL_OFDM_RATES_MASK >> IWL_FIRST_OFDM_RATE) & 0xFF;
@@ -2560,7 +2556,7 @@ static int iwl3945_set_mode(struct iwl3945_priv *priv, int mode)
 		const struct iwl3945_channel_info *ch_info;
 
 		ch_info = iwl3945_get_channel_info(priv,
-			priv->phymode,
+			priv->band,
 			le16_to_cpu(priv->staging_rxon.channel));
 
 		if (!ch_info || !is_channel_ibss(ch_info)) {
@@ -2792,7 +2788,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv,
 		goto drop_unlock;
 	}
 
-	if ((ctl->tx_rate & 0xFF) == IWL_INVALID_RATE) {
+	if ((ctl->tx_rate->hw_value & 0xFF) == IWL_INVALID_RATE) {
 		IWL_ERROR("ERROR: No TX rate available.\n");
 		goto drop_unlock;
 	}
@@ -2992,12 +2988,12 @@ drop:
 
 static void iwl3945_set_rate(struct iwl3945_priv *priv)
 {
-	const struct ieee80211_hw_mode *hw = NULL;
+	const struct ieee80211_supported_band *sband = NULL;
 	struct ieee80211_rate *rate;
 	int i;
 
-	hw = iwl3945_get_hw_mode(priv, priv->phymode);
-	if (!hw) {
+	sband = iwl3945_get_band(priv, priv->band);
+	if (!sband) {
 		IWL_ERROR("Failed to set rate: unable to get hw mode\n");
 		return;
 	}
@@ -3005,24 +3001,17 @@ static void iwl3945_set_rate(struct iwl3945_priv *priv)
 	priv->active_rate = 0;
 	priv->active_rate_basic = 0;
 
-	IWL_DEBUG_RATE("Setting rates for 802.11%c\n",
-		       hw->mode == MODE_IEEE80211A ?
-		       'a' : ((hw->mode == MODE_IEEE80211B) ? 'b' : 'g'));
-
-	for (i = 0; i < hw->num_rates; i++) {
-		rate = &(hw->rates[i]);
-		if ((rate->val < IWL_RATE_COUNT) &&
-		    (rate->flags & IEEE80211_RATE_SUPPORTED)) {
-			IWL_DEBUG_RATE("Adding rate index %d (plcp %d)%s\n",
-				       rate->val, iwl3945_rates[rate->val].plcp,
-				       (rate->flags & IEEE80211_RATE_BASIC) ?
-				       "*" : "");
-			priv->active_rate |= (1 << rate->val);
-			if (rate->flags & IEEE80211_RATE_BASIC)
-				priv->active_rate_basic |= (1 << rate->val);
-		} else
-			IWL_DEBUG_RATE("Not adding rate %d (plcp %d)\n",
-				       rate->val, iwl3945_rates[rate->val].plcp);
+	IWL_DEBUG_RATE("Setting rates for %s GHz\n",
+		       sband->band == IEEE80211_BAND_2GHZ ? "2.4" : "5");
+
+	for (i = 0; i < sband->n_bitrates; i++) {
+		rate = &sband->bitrates[i];
+		if ((rate->hw_value < IWL_RATE_COUNT) &&
+		    !(rate->flags & IEEE80211_CHAN_DISABLED)) {
+			IWL_DEBUG_RATE("Adding rate index %d (plcp %d)\n",
+				       rate->hw_value, iwl3945_rates[rate->hw_value].plcp);
+			priv->active_rate |= (1 << rate->hw_value);
+		}
 	}
 
 	IWL_DEBUG_RATE("Set active_rate = %0x, active_rate_basic = %0x\n",
@@ -3436,8 +3425,6 @@ static void iwl3945_rx_reply_tx(struct iwl3945_priv *priv,
 	tx_status->flags =
 	    iwl3945_is_tx_success(status) ? IEEE80211_TX_STATUS_ACK : 0;
 
-	tx_status->control.tx_rate = iwl3945_rate_index_from_plcp(tx_resp->rate);
-
 	IWL_DEBUG_TX("Tx queue %d Status %s (0x%08x) plcp rate %d retries %d\n",
 			txq_id, iwl3945_get_tx_fail_reason(status), status,
 			tx_resp->rate, tx_resp->failure_frame);
@@ -5026,24 +5013,24 @@ static void iwl3945_init_band_reference(const struct iwl3945_priv *priv, int ban
  * Based on band and channel number.
  */
 const struct iwl3945_channel_info *iwl3945_get_channel_info(const struct iwl3945_priv *priv,
-						    int phymode, u16 channel)
+						    enum ieee80211_band band, u16 channel)
 {
 	int i;
 
-	switch (phymode) {
-	case MODE_IEEE80211A:
+	switch (band) {
+	case IEEE80211_BAND_5GHZ:
 		for (i = 14; i < priv->channel_count; i++) {
 			if (priv->channel_info[i].channel == channel)
 				return &priv->channel_info[i];
 		}
 		break;
 
-	case MODE_IEEE80211B:
-	case MODE_IEEE80211G:
+	case IEEE80211_BAND_2GHZ:
 		if (channel >= 1 && channel <= 14)
 			return &priv->channel_info[channel - 1];
 		break;
-
+	case IEEE80211_NUM_BANDS:
+		WARN_ON(1);
 	}
 
 	return NULL;
@@ -5106,8 +5093,8 @@ static int iwl3945_init_channel_map(struct iwl3945_priv *priv)
 		/* Loop through each band adding each of the channels */
 		for (ch = 0; ch < eeprom_ch_count; ch++) {
 			ch_info->channel = eeprom_ch_index[ch];
-			ch_info->phymode = (band == 1) ? MODE_IEEE80211B :
-			    MODE_IEEE80211A;
+			ch_info->band = (band == 1) ? IEEE80211_BAND_2GHZ :
+			    IEEE80211_BAND_5GHZ;
 
 			/* permanently store EEPROM's channel regulatory flags
 			 *   and max power in channel info database. */
@@ -5203,18 +5190,20 @@ static void iwl3945_free_channel_map(struct iwl3945_priv *priv)
 #define IWL_PASSIVE_DWELL_BASE      (100)
 #define IWL_CHANNEL_TUNE_TIME       5
 
-static inline u16 iwl3945_get_active_dwell_time(struct iwl3945_priv *priv, int phymode)
+static inline u16 iwl3945_get_active_dwell_time(struct iwl3945_priv *priv,
+						enum ieee80211_band band)
 {
-	if (phymode == MODE_IEEE80211A)
+	if (band == IEEE80211_BAND_5GHZ)
 		return IWL_ACTIVE_DWELL_TIME_52;
 	else
 		return IWL_ACTIVE_DWELL_TIME_24;
 }
 
-static u16 iwl3945_get_passive_dwell_time(struct iwl3945_priv *priv, int phymode)
+static u16 iwl3945_get_passive_dwell_time(struct iwl3945_priv *priv,
+					  enum ieee80211_band band)
 {
-	u16 active = iwl3945_get_active_dwell_time(priv, phymode);
-	u16 passive = (phymode != MODE_IEEE80211A) ?
+	u16 active = iwl3945_get_active_dwell_time(priv, band);
+	u16 passive = (band == IEEE80211_BAND_2GHZ) ?
 	    IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_24 :
 	    IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_52;
 
@@ -5234,28 +5223,29 @@ static u16 iwl3945_get_passive_dwell_time(struct iwl3945_priv *priv, int phymode
 	return passive;
 }
 
-static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode,
+static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv,
+					 enum ieee80211_band band,
 				     u8 is_active, u8 direct_mask,
 				     struct iwl3945_scan_channel *scan_ch)
 {
 	const struct ieee80211_channel *channels = NULL;
-	const struct ieee80211_hw_mode *hw_mode;
+	const struct ieee80211_supported_band *sband;
 	const struct iwl3945_channel_info *ch_info;
 	u16 passive_dwell = 0;
 	u16 active_dwell = 0;
 	int added, i;
 
-	hw_mode = iwl3945_get_hw_mode(priv, phymode);
-	if (!hw_mode)
+	sband = iwl3945_get_band(priv, band);
+	if (!sband)
 		return 0;
 
-	channels = hw_mode->channels;
+	channels = sband->channels;
 
-	active_dwell = iwl3945_get_active_dwell_time(priv, phymode);
-	passive_dwell = iwl3945_get_passive_dwell_time(priv, phymode);
+	active_dwell = iwl3945_get_active_dwell_time(priv, band);
+	passive_dwell = iwl3945_get_passive_dwell_time(priv, band);
 
-	for (i = 0, added = 0; i < hw_mode->num_channels; i++) {
-		if (channels[i].chan ==
+	for (i = 0, added = 0; i < sband->n_channels; i++) {
+		if (channels[i].hw_value ==
 		    le16_to_cpu(priv->active_rxon.channel)) {
 			if (iwl3945_is_associated(priv)) {
 				IWL_DEBUG_SCAN
@@ -5266,9 +5256,9 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode,
 		} else if (priv->only_active_channel)
 			continue;
 
-		scan_ch->channel = channels[i].chan;
+		scan_ch->channel = channels[i].hw_value;
 
-		ch_info = iwl3945_get_channel_info(priv, phymode, scan_ch->channel);
+		ch_info = iwl3945_get_channel_info(priv, band, scan_ch->channel);
 		if (!is_channel_valid(ch_info)) {
 			IWL_DEBUG_SCAN("Channel %d is INVALID for this SKU.\n",
 				       scan_ch->channel);
@@ -5276,7 +5266,7 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode,
 		}
 
 		if (!is_active || is_channel_passive(ch_info) ||
-		    !(channels[i].flag & IEEE80211_CHAN_W_ACTIVE_SCAN))
+		    (channels[i].flags & IEEE80211_CHAN_PASSIVE_SCAN))
 			scan_ch->type = 0;	/* passive */
 		else
 			scan_ch->type = 1;	/* active */
@@ -5295,7 +5285,7 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode,
 		/* scan_pwr_info->tpc.dsp_atten; */
 
 		/*scan_pwr_info->tpc.tx_gain; */
-		if (phymode == MODE_IEEE80211A)
+		if (band == IEEE80211_BAND_5GHZ)
 			scan_ch->tpc.tx_gain = ((1 << 5) | (3 << 3)) | 3;
 		else {
 			scan_ch->tpc.tx_gain = ((1 << 5) | (5 << 3));
@@ -5319,41 +5309,23 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode,
 	return added;
 }
 
-static void iwl3945_reset_channel_flag(struct iwl3945_priv *priv)
-{
-	int i, j;
-	for (i = 0; i < 3; i++) {
-		struct ieee80211_hw_mode *hw_mode = (void *)&priv->modes[i];
-		for (j = 0; j < hw_mode->num_channels; j++)
-			hw_mode->channels[j].flag = hw_mode->channels[j].val;
-	}
-}
-
 static void iwl3945_init_hw_rates(struct iwl3945_priv *priv,
 			      struct ieee80211_rate *rates)
 {
 	int i;
 
 	for (i = 0; i < IWL_RATE_COUNT; i++) {
-		rates[i].rate = iwl3945_rates[i].ieee * 5;
-		rates[i].val = i; /* Rate scaling will work on indexes */
-		rates[i].val2 = i;
-		rates[i].flags = IEEE80211_RATE_SUPPORTED;
-		/* Only OFDM have the bits-per-symbol set */
-		if ((i <= IWL_LAST_OFDM_RATE) && (i >= IWL_FIRST_OFDM_RATE))
-			rates[i].flags |= IEEE80211_RATE_OFDM;
-		else {
+		rates[i].bitrate = iwl3945_rates[i].ieee * 5;
+		rates[i].hw_value = i; /* Rate scaling will work on indexes */
+		rates[i].hw_value_short = i;
+		rates[i].flags = 0;
+		if ((i > IWL_LAST_OFDM_RATE) || (i < IWL_FIRST_OFDM_RATE)) {
 			/*
-			 * If CCK 1M then set rate flag to CCK else CCK_2
-			 * which is CCK | PREAMBLE2
+			 * If CCK != 1M then set short preamble rate flag.
 			 */
 			rates[i].flags |= (iwl3945_rates[i].plcp == 10) ?
-				IEEE80211_RATE_CCK : IEEE80211_RATE_CCK_2;
+				0 : IEEE80211_RATE_SHORT_PREAMBLE;
 		}
-
-		/* Set up which ones are basic rates... */
-		if (IWL_BASIC_RATES_MASK & (1 << i))
-			rates[i].flags |= IEEE80211_RATE_BASIC;
 	}
 }
 
@@ -5363,67 +5335,41 @@ static void iwl3945_init_hw_rates(struct iwl3945_priv *priv,
 static int iwl3945_init_geos(struct iwl3945_priv *priv)
 {
 	struct iwl3945_channel_info *ch;
-	struct ieee80211_hw_mode *modes;
+	struct ieee80211_supported_band *band;
 	struct ieee80211_channel *channels;
 	struct ieee80211_channel *geo_ch;
 	struct ieee80211_rate *rates;
 	int i = 0;
-	enum {
-		A = 0,
-		B = 1,
-		G = 2,
-	};
-	int mode_count = 3;
 
-	if (priv->modes) {
+	if (priv->bands[IEEE80211_BAND_2GHZ].n_bitrates ||
+	    priv->bands[IEEE80211_BAND_5GHZ].n_bitrates) {
 		IWL_DEBUG_INFO("Geography modes already initialized.\n");
 		set_bit(STATUS_GEO_CONFIGURED, &priv->status);
 		return 0;
 	}
 
-	modes = kzalloc(sizeof(struct ieee80211_hw_mode) * mode_count,
-			GFP_KERNEL);
-	if (!modes)
-		return -ENOMEM;
-
 	channels = kzalloc(sizeof(struct ieee80211_channel) *
 			   priv->channel_count, GFP_KERNEL);
-	if (!channels) {
-		kfree(modes);
+	if (!channels)
 		return -ENOMEM;
-	}
 
 	rates = kzalloc((sizeof(struct ieee80211_rate) * (IWL_MAX_RATES + 1)),
 			GFP_KERNEL);
 	if (!rates) {
-		kfree(modes);
 		kfree(channels);
 		return -ENOMEM;
 	}
 
-	/* 0 = 802.11a
-	 * 1 = 802.11b
-	 * 2 = 802.11g
-	 */
-
 	/* 5.2GHz channels start after the 2.4GHz channels */
-	modes[A].mode = MODE_IEEE80211A;
-	modes[A].channels = &channels[ARRAY_SIZE(iwl3945_eeprom_band_1)];
-	modes[A].rates = &rates[4];
-	modes[A].num_rates = 8;	/* just OFDM */
-	modes[A].num_channels = 0;
-
-	modes[B].mode = MODE_IEEE80211B;
-	modes[B].channels = channels;
-	modes[B].rates = rates;
-	modes[B].num_rates = 4;	/* just CCK */
-	modes[B].num_channels = 0;
-
-	modes[G].mode = MODE_IEEE80211G;
-	modes[G].channels = channels;
-	modes[G].rates = rates;
-	modes[G].num_rates = 12;	/* OFDM & CCK */
-	modes[G].num_channels = 0;
+	band = &priv->bands[IEEE80211_BAND_5GHZ];
+	band->channels = &channels[ARRAY_SIZE(iwl3945_eeprom_band_1)];
+	band->bitrates = &rates[4];
+	band->n_bitrates = 8;	/* just OFDM */
+
+	band = &priv->bands[IEEE80211_BAND_2GHZ];
+	band->channels = channels;
+	band->bitrates = rates;
+	band->n_bitrates = 12;	/* OFDM & CCK */
 
 	priv->ieee_channels = channels;
 	priv->ieee_rates = rates;
@@ -5442,37 +5388,32 @@ static int iwl3945_init_geos(struct iwl3945_priv *priv)
 		}
 
 		if (is_channel_a_band(ch))
-			geo_ch = &modes[A].channels[modes[A].num_channels++];
-		else {
-			geo_ch = &modes[B].channels[modes[B].num_channels++];
-			modes[G].num_channels++;
-		}
+			geo_ch = &priv->bands[IEEE80211_BAND_5GHZ].channels[priv->bands[IEEE80211_BAND_5GHZ].n_channels++];
+		else
+			geo_ch = &priv->bands[IEEE80211_BAND_2GHZ].channels[priv->bands[IEEE80211_BAND_2GHZ].n_channels++];
 
-		geo_ch->freq = ieee80211chan2mhz(ch->channel);
-		geo_ch->chan = ch->channel;
-		geo_ch->power_level = ch->max_power_avg;
-		geo_ch->antenna_max = 0xff;
+		geo_ch->center_freq = ieee80211chan2mhz(ch->channel);
+		geo_ch->max_power = ch->max_power_avg;
+		geo_ch->max_antenna_gain = 0xff;
 
 		if (is_channel_valid(ch)) {
-			geo_ch->flag = IEEE80211_CHAN_W_SCAN;
-			if (ch->flags & EEPROM_CHANNEL_IBSS)
-				geo_ch->flag |= IEEE80211_CHAN_W_IBSS;
+			if (!(ch->flags & EEPROM_CHANNEL_IBSS))
+				geo_ch->flags |= IEEE80211_CHAN_NO_IBSS;
 
-			if (ch->flags & EEPROM_CHANNEL_ACTIVE)
-				geo_ch->flag |= IEEE80211_CHAN_W_ACTIVE_SCAN;
+			if (!(ch->flags & EEPROM_CHANNEL_ACTIVE))
+				geo_ch->flags |= IEEE80211_CHAN_PASSIVE_SCAN;
 
 			if (ch->flags & EEPROM_CHANNEL_RADAR)
-				geo_ch->flag |= IEEE80211_CHAN_W_RADAR_DETECT;
+				geo_ch->flags |= IEEE80211_CHAN_RADAR;
 
 			if (ch->max_power_avg > priv->max_channel_txpower_limit)
 				priv->max_channel_txpower_limit =
 				    ch->max_power_avg;
-		}
-
-		geo_ch->val = geo_ch->flag;
+		} else
+			geo_ch->flags |= IEEE80211_CHAN_DISABLED;
 	}
 
-	if ((modes[A].num_channels == 0) && priv->is_abg) {
+	if ((priv->bands[IEEE80211_BAND_5GHZ].n_channels == 0) && priv->is_abg) {
 		printk(KERN_INFO DRV_NAME
 		       ": Incorrectly detected BG card as ABG.  Please send "
 		       "your PCI ID 0x%04X:0x%04X to maintainer.\n",
@@ -5482,24 +5423,12 @@ static int iwl3945_init_geos(struct iwl3945_priv *priv)
 
 	printk(KERN_INFO DRV_NAME
 	       ": Tunable channels: %d 802.11bg, %d 802.11a channels\n",
-	       modes[G].num_channels, modes[A].num_channels);
+	       priv->bands[IEEE80211_BAND_2GHZ].n_channels,
+	       priv->bands[IEEE80211_BAND_5GHZ].n_channels);
 
-	/*
-	 * NOTE:  We register these in preference of order -- the
-	 * stack doesn't currently (as of 7.0.6 / Apr 24 '07) pick
-	 * a phymode based on rates or AP capabilities but seems to
-	 * configure it purely on if the channel being configured
-	 * is supported by a mode -- and the first match is taken
-	 */
-
-	if (modes[G].num_channels)
-		ieee80211_register_hwmode(priv->hw, &modes[G]);
-	if (modes[B].num_channels)
-		ieee80211_register_hwmode(priv->hw, &modes[B]);
-	if (modes[A].num_channels)
-		ieee80211_register_hwmode(priv->hw, &modes[A]);
+	priv->hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->bands[IEEE80211_BAND_2GHZ];
+	priv->hw->wiphy->bands[IEEE80211_BAND_5GHZ] = &priv->bands[IEEE80211_BAND_5GHZ];
 
-	priv->modes = modes;
 	set_bit(STATUS_GEO_CONFIGURED, &priv->status);
 
 	return 0;
@@ -5510,7 +5439,6 @@ static int iwl3945_init_geos(struct iwl3945_priv *priv)
  */
 static void iwl3945_free_geos(struct iwl3945_priv *priv)
 {
-	kfree(priv->modes);
 	kfree(priv->ieee_channels);
 	kfree(priv->ieee_rates);
 	clear_bit(STATUS_GEO_CONFIGURED, &priv->status);
@@ -6519,7 +6447,7 @@ static void iwl3945_bg_request_scan(struct work_struct *data)
 	struct iwl3945_scan_cmd *scan;
 	struct ieee80211_conf *conf = NULL;
 	u8 direct_mask;
-	int phymode;
+	enum ieee80211_band band;
 
 	conf = ieee80211_get_hw_conf(priv->hw);
 
@@ -6651,13 +6579,13 @@ static void iwl3945_bg_request_scan(struct work_struct *data)
 		scan->flags = RXON_FLG_BAND_24G_MSK | RXON_FLG_AUTO_DETECT_MSK;
 		scan->tx_cmd.rate = IWL_RATE_1M_PLCP;
 		scan->good_CRC_th = 0;
-		phymode = MODE_IEEE80211G;
+		band = IEEE80211_BAND_2GHZ;
 		break;
 
 	case 1:
 		scan->tx_cmd.rate = IWL_RATE_6M_PLCP;
 		scan->good_CRC_th = IWL_GOOD_CRC_TH;
-		phymode = MODE_IEEE80211A;
+		band = IEEE80211_BAND_5GHZ;
 		break;
 
 	default:
@@ -6680,7 +6608,7 @@ static void iwl3945_bg_request_scan(struct work_struct *data)
 
 	scan->channel_count =
 		iwl3945_get_channels_for_scan(
-			priv, phymode, 1, /* active */
+			priv, band, 1, /* active */
 			direct_mask,
 			(void *)&scan->data[le16_to_cpu(scan->tx_cmd.len)]);
 
@@ -6825,7 +6753,7 @@ static void iwl3945_bg_post_associate(struct work_struct *data)
 		iwl3945_add_station(priv, iwl3945_broadcast_addr, 0, 0);
 		iwl3945_add_station(priv, priv->bssid, 0, 0);
 		iwl3945_sync_sta(priv, IWL_STA_ID,
-				 (priv->phymode == MODE_IEEE80211A)?
+				 (priv->band == IEEE80211_BAND_5GHZ) ?
 				 IWL_RATE_6M_PLCP : IWL_RATE_1M_PLCP,
 				 CMD_ASYNC);
 		iwl3945_rate_scale_init(priv->hw, IWL_STA_ID);
@@ -7020,7 +6948,7 @@ static int iwl3945_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	}
 
 	IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len,
-		     ctl->tx_rate);
+		     ctl->tx_rate->bitrate);
 
 	if (iwl3945_tx_skb(priv, skb, ctl))
 		dev_kfree_skb_any(skb);
@@ -7079,7 +7007,7 @@ static int iwl3945_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co
 	int ret = 0;
 
 	mutex_lock(&priv->mutex);
-	IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel);
+	IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel->hw_value);
 
 	priv->add_radiotap = !!(conf->flags & IEEE80211_CONF_RADIOTAP);
 
@@ -7099,19 +7027,20 @@ static int iwl3945_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co
 
 	spin_lock_irqsave(&priv->lock, flags);
 
-	ch_info = iwl3945_get_channel_info(priv, conf->phymode, conf->channel);
+	ch_info = iwl3945_get_channel_info(priv, conf->channel->band,
+					   conf->channel->hw_value);
 	if (!is_channel_valid(ch_info)) {
 		IWL_DEBUG_SCAN("Channel %d [%d] is INVALID for this SKU.\n",
-			       conf->channel, conf->phymode);
+			       conf->channel->hw_value, conf->channel->band);
 		IWL_DEBUG_MAC80211("leave - invalid channel\n");
 		spin_unlock_irqrestore(&priv->lock, flags);
 		ret = -EINVAL;
 		goto out;
 	}
 
-	iwl3945_set_rxon_channel(priv, conf->phymode, conf->channel);
+	iwl3945_set_rxon_channel(priv, conf->channel->band, conf->channel->hw_value);
 
-	iwl3945_set_flags_for_phymode(priv, conf->phymode);
+	iwl3945_set_flags_for_phymode(priv, conf->channel->band);
 
 	/* The list of supported rates and rate mask can be different
 	 * for each phymode; since the phymode may have changed, reset
@@ -7892,65 +7821,6 @@ static ssize_t store_filter_flags(struct device *d,
 static DEVICE_ATTR(filter_flags, S_IWUSR | S_IRUGO, show_filter_flags,
 		   store_filter_flags);
 
-static ssize_t show_tune(struct device *d,
-			 struct device_attribute *attr, char *buf)
-{
-	struct iwl3945_priv *priv = (struct iwl3945_priv *)d->driver_data;
-
-	return sprintf(buf, "0x%04X\n",
-		       (priv->phymode << 8) |
-			le16_to_cpu(priv->active_rxon.channel));
-}
-
-static void iwl3945_set_flags_for_phymode(struct iwl3945_priv *priv, u8 phymode);
-
-static ssize_t store_tune(struct device *d,
-			  struct device_attribute *attr,
-			  const char *buf, size_t count)
-{
-	struct iwl3945_priv *priv = (struct iwl3945_priv *)d->driver_data;
-	char *p = (char *)buf;
-	u16 tune = simple_strtoul(p, &p, 0);
-	u8 phymode = (tune >> 8) & 0xff;
-	u16 channel = tune & 0xff;
-
-	IWL_DEBUG_INFO("Tune request to:%d channel:%d\n", phymode, channel);
-
-	mutex_lock(&priv->mutex);
-	if ((le16_to_cpu(priv->staging_rxon.channel) != channel) ||
-	    (priv->phymode != phymode)) {
-		const struct iwl3945_channel_info *ch_info;
-
-		ch_info = iwl3945_get_channel_info(priv, phymode, channel);
-		if (!ch_info) {
-			IWL_WARNING("Requested invalid phymode/channel "
-				    "combination: %d %d\n", phymode, channel);
-			mutex_unlock(&priv->mutex);
-			return -EINVAL;
-		}
-
-		/* Cancel any currently running scans... */
-		if (iwl3945_scan_cancel_timeout(priv, 100))
-			IWL_WARNING("Could not cancel scan.\n");
-		else {
-			IWL_DEBUG_INFO("Committing phymode and "
-				       "rxon.channel = %d %d\n",
-				       phymode, channel);
-
-			iwl3945_set_rxon_channel(priv, phymode, channel);
-			iwl3945_set_flags_for_phymode(priv, phymode);
-
-			iwl3945_set_rate(priv);
-			iwl3945_commit_rxon(priv);
-		}
-	}
-	mutex_unlock(&priv->mutex);
-
-	return count;
-}
-
-static DEVICE_ATTR(tune, S_IWUSR | S_IRUGO, show_tune, store_tune);
-
 #ifdef CONFIG_IWL3945_SPECTRUM_MEASUREMENT
 
 static ssize_t show_measurement(struct device *d,
@@ -8165,73 +8035,8 @@ static DEVICE_ATTR(power_level, S_IWUSR | S_IRUSR, show_power_level,
 static ssize_t show_channels(struct device *d,
 			     struct device_attribute *attr, char *buf)
 {
-	struct iwl3945_priv *priv = dev_get_drvdata(d);
-	int len = 0, i;
-	struct ieee80211_channel *channels = NULL;
-	const struct ieee80211_hw_mode *hw_mode = NULL;
-	int count = 0;
-
-	if (!iwl3945_is_ready(priv))
-		return -EAGAIN;
-
-	hw_mode = iwl3945_get_hw_mode(priv, MODE_IEEE80211G);
-	if (!hw_mode)
-		hw_mode = iwl3945_get_hw_mode(priv, MODE_IEEE80211B);
-	if (hw_mode) {
-		channels = hw_mode->channels;
-		count = hw_mode->num_channels;
-	}
-
-	len +=
-	    sprintf(&buf[len],
-		    "Displaying %d channels in 2.4GHz band "
-		    "(802.11bg):\n", count);
-
-	for (i = 0; i < count; i++)
-		len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n",
-			       channels[i].chan,
-			       channels[i].power_level,
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_RADAR_DETECT ?
-			       " (IEEE 802.11h required)" : "",
-			       (!(channels[i].flag & IEEE80211_CHAN_W_IBSS)
-				|| (channels[i].
-				    flag &
-				    IEEE80211_CHAN_W_RADAR_DETECT)) ? "" :
-			       ", IBSS",
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_ACTIVE_SCAN ?
-			       "active/passive" : "passive only");
-
-	hw_mode = iwl3945_get_hw_mode(priv, MODE_IEEE80211A);
-	if (hw_mode) {
-		channels = hw_mode->channels;
-		count = hw_mode->num_channels;
-	} else {
-		channels = NULL;
-		count = 0;
-	}
-
-	len += sprintf(&buf[len], "Displaying %d channels in 5.2GHz band "
-		       "(802.11a):\n", count);
-
-	for (i = 0; i < count; i++)
-		len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n",
-			       channels[i].chan,
-			       channels[i].power_level,
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_RADAR_DETECT ?
-			       " (IEEE 802.11h required)" : "",
-			       (!(channels[i].flag & IEEE80211_CHAN_W_IBSS)
-				|| (channels[i].
-				    flag &
-				    IEEE80211_CHAN_W_RADAR_DETECT)) ? "" :
-			       ", IBSS",
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_ACTIVE_SCAN ?
-			       "active/passive" : "passive only");
-
-	return len;
+	/* all this shit doesn't belong into sysfs anyway */
+	return 0;
 }
 
 static DEVICE_ATTR(channels, S_IRUSR, show_channels, NULL);
@@ -8411,7 +8216,6 @@ static struct attribute *iwl3945_sysfs_entries[] = {
 	&dev_attr_statistics.attr,
 	&dev_attr_status.attr,
 	&dev_attr_temperature.attr,
-	&dev_attr_tune.attr,
 	&dev_attr_tx_power.attr,
 
 	NULL
@@ -8532,7 +8336,7 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
 	priv->data_retry_limit = -1;
 	priv->ieee_channels = NULL;
 	priv->ieee_rates = NULL;
-	priv->phymode = -1;
+	priv->band = IEEE80211_BAND_2GHZ;
 
 	err = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
 	if (!err)
@@ -8614,7 +8418,7 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
 	priv->qos_data.qos_cap.val = 0;
 #endif /* CONFIG_IWL3945_QOS */
 
-	iwl3945_set_rxon_channel(priv, MODE_IEEE80211G, 6);
+	iwl3945_set_rxon_channel(priv, IEEE80211_BAND_2GHZ, 6);
 	iwl3945_setup_deferred_work(priv);
 	iwl3945_setup_rx_handlers(priv);
 
@@ -8665,7 +8469,6 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
 		IWL_ERROR("initializing geos failed: %d\n", err);
 		goto out_free_channel_map;
 	}
-	iwl3945_reset_channel_flag(priv);
 
 	iwl3945_rate_control_register(priv->hw);
 	err = ieee80211_register_hw(priv->hw);
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 5f38fc585eda..6de969de4c84 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -115,16 +115,10 @@ __le16 *ieee80211_get_qos_ctrl(struct ieee80211_hdr *hdr)
 	return NULL;
 }
 
-static const struct ieee80211_hw_mode *iwl4965_get_hw_mode(
-		struct iwl4965_priv *priv, int mode)
+static const struct ieee80211_supported_band *iwl4965_get_hw_mode(
+		struct iwl4965_priv *priv, enum ieee80211_band band)
 {
-	int i;
-
-	for (i = 0; i < 3; i++)
-		if (priv->modes[i].mode == mode)
-			return &priv->modes[i];
-
-	return NULL;
+	return priv->hw->wiphy->bands[band];
 }
 
 static int iwl4965_is_empty_essid(const char *essid, int essid_len)
@@ -937,28 +931,29 @@ static int iwl4965_rxon_add_station(struct iwl4965_priv *priv,
  * NOTE:  Does not commit to the hardware; it sets appropriate bit fields
  * in the staging RXON flag structure based on the phymode
  */
-static int iwl4965_set_rxon_channel(struct iwl4965_priv *priv, u8 phymode,
+static int iwl4965_set_rxon_channel(struct iwl4965_priv *priv,
+				    enum ieee80211_band band,
 				 u16 channel)
 {
-	if (!iwl4965_get_channel_info(priv, phymode, channel)) {
+	if (!iwl4965_get_channel_info(priv, band, channel)) {
 		IWL_DEBUG_INFO("Could not set channel to %d [%d]\n",
-			       channel, phymode);
+			       channel, band);
 		return -EINVAL;
 	}
 
 	if ((le16_to_cpu(priv->staging_rxon.channel) == channel) &&
-	    (priv->phymode == phymode))
+	    (priv->band == band))
 		return 0;
 
 	priv->staging_rxon.channel = cpu_to_le16(channel);
-	if (phymode == MODE_IEEE80211A)
+	if (band == IEEE80211_BAND_5GHZ)
 		priv->staging_rxon.flags &= ~RXON_FLG_BAND_24G_MSK;
 	else
 		priv->staging_rxon.flags |= RXON_FLG_BAND_24G_MSK;
 
-	priv->phymode = phymode;
+	priv->band = band;
 
-	IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, phymode);
+	IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, band);
 
 	return 0;
 }
@@ -2571,9 +2566,10 @@ static int iwl4965_set_rxon_hwcrypto(struct iwl4965_priv *priv, int hw_decrypt)
 	return 0;
 }
 
-static void iwl4965_set_flags_for_phymode(struct iwl4965_priv *priv, u8 phymode)
+static void iwl4965_set_flags_for_phymode(struct iwl4965_priv *priv,
+					  enum ieee80211_band band)
 {
-	if (phymode == MODE_IEEE80211A) {
+	if (band == IEEE80211_BAND_5GHZ) {
 		priv->staging_rxon.flags &=
 		    ~(RXON_FLG_BAND_24G_MSK | RXON_FLG_AUTO_DETECT_MSK
 		      | RXON_FLG_CCK_MSK);
@@ -2636,7 +2632,7 @@ static void iwl4965_connection_init_rx_config(struct iwl4965_priv *priv)
 		priv->staging_rxon.flags |= RXON_FLG_SHORT_PREAMBLE_MSK;
 #endif
 
-	ch_info = iwl4965_get_channel_info(priv, priv->phymode,
+	ch_info = iwl4965_get_channel_info(priv, priv->band,
 				       le16_to_cpu(priv->staging_rxon.channel));
 
 	if (!ch_info)
@@ -2651,12 +2647,9 @@ static void iwl4965_connection_init_rx_config(struct iwl4965_priv *priv)
 		ch_info = &priv->channel_info[0];
 
 	priv->staging_rxon.channel = cpu_to_le16(ch_info->channel);
-	if (is_channel_a_band(ch_info))
-		priv->phymode = MODE_IEEE80211A;
-	else
-		priv->phymode = MODE_IEEE80211G;
+	priv->band = ch_info->band;
 
-	iwl4965_set_flags_for_phymode(priv, priv->phymode);
+	iwl4965_set_flags_for_phymode(priv, priv->band);
 
 	priv->staging_rxon.ofdm_basic_rates =
 	    (IWL_OFDM_RATES_MASK >> IWL_FIRST_OFDM_RATE) & 0xFF;
@@ -2678,7 +2671,7 @@ static int iwl4965_set_mode(struct iwl4965_priv *priv, int mode)
 		const struct iwl4965_channel_info *ch_info;
 
 		ch_info = iwl4965_get_channel_info(priv,
-			priv->phymode,
+			priv->band,
 			le16_to_cpu(priv->staging_rxon.channel));
 
 		if (!ch_info || !is_channel_ibss(ch_info)) {
@@ -2918,7 +2911,7 @@ static int iwl4965_tx_skb(struct iwl4965_priv *priv,
 		goto drop_unlock;
 	}
 
-	if ((ctl->tx_rate & 0xFF) == IWL_INVALID_RATE) {
+	if ((ctl->tx_rate->hw_value & 0xFF) == IWL_INVALID_RATE) {
 		IWL_ERROR("ERROR: No TX rate available.\n");
 		goto drop_unlock;
 	}
@@ -3125,11 +3118,11 @@ drop:
 
 static void iwl4965_set_rate(struct iwl4965_priv *priv)
 {
-	const struct ieee80211_hw_mode *hw = NULL;
+	const struct ieee80211_supported_band *hw = NULL;
 	struct ieee80211_rate *rate;
 	int i;
 
-	hw = iwl4965_get_hw_mode(priv, priv->phymode);
+	hw = iwl4965_get_hw_mode(priv, priv->band);
 	if (!hw) {
 		IWL_ERROR("Failed to set rate: unable to get hw mode\n");
 		return;
@@ -3138,24 +3131,10 @@ static void iwl4965_set_rate(struct iwl4965_priv *priv)
 	priv->active_rate = 0;
 	priv->active_rate_basic = 0;
 
-	IWL_DEBUG_RATE("Setting rates for 802.11%c\n",
-		       hw->mode == MODE_IEEE80211A ?
-		       'a' : ((hw->mode == MODE_IEEE80211B) ? 'b' : 'g'));
-
-	for (i = 0; i < hw->num_rates; i++) {
-		rate = &(hw->rates[i]);
-		if ((rate->val < IWL_RATE_COUNT) &&
-		    (rate->flags & IEEE80211_RATE_SUPPORTED)) {
-			IWL_DEBUG_RATE("Adding rate index %d (plcp %d)%s\n",
-				       rate->val, iwl4965_rates[rate->val].plcp,
-				       (rate->flags & IEEE80211_RATE_BASIC) ?
-				       "*" : "");
-			priv->active_rate |= (1 << rate->val);
-			if (rate->flags & IEEE80211_RATE_BASIC)
-				priv->active_rate_basic |= (1 << rate->val);
-		} else
-			IWL_DEBUG_RATE("Not adding rate %d (plcp %d)\n",
-				       rate->val, iwl4965_rates[rate->val].plcp);
+	for (i = 0; i < hw->n_bitrates; i++) {
+		rate = &(hw->bitrates[i]);
+		if (rate->hw_value < IWL_RATE_COUNT)
+			priv->active_rate |= (1 << rate->hw_value);
 	}
 
 	IWL_DEBUG_RATE("Set active_rate = %0x, active_rate_basic = %0x\n",
@@ -3775,9 +3754,6 @@ static void iwl4965_rx_reply_tx(struct iwl4965_priv *priv,
 	tx_status->flags =
 	    iwl4965_is_tx_success(status) ? IEEE80211_TX_STATUS_ACK : 0;
 
-	tx_status->control.tx_rate =
-		iwl4965_hw_get_rate_n_flags(tx_resp->rate_n_flags);
-
 	IWL_DEBUG_TX("Tx queue %d Status %s (0x%08x) rate_n_flags 0x%x "
 		     "retries %d\n", txq_id, iwl4965_get_tx_fail_reason(status),
 		     status, le32_to_cpu(tx_resp->rate_n_flags),
@@ -5419,24 +5395,23 @@ static void iwl4965_init_band_reference(const struct iwl4965_priv *priv,
  * Based on band and channel number.
  */
 const struct iwl4965_channel_info *iwl4965_get_channel_info(const struct iwl4965_priv *priv,
-						    int phymode, u16 channel)
+						    enum ieee80211_band band, u16 channel)
 {
 	int i;
 
-	switch (phymode) {
-	case MODE_IEEE80211A:
+	switch (band) {
+	case IEEE80211_BAND_5GHZ:
 		for (i = 14; i < priv->channel_count; i++) {
 			if (priv->channel_info[i].channel == channel)
 				return &priv->channel_info[i];
 		}
 		break;
-
-	case MODE_IEEE80211B:
-	case MODE_IEEE80211G:
+	case IEEE80211_BAND_2GHZ:
 		if (channel >= 1 && channel <= 14)
 			return &priv->channel_info[channel - 1];
 		break;
-
+	default:
+		BUG();
 	}
 
 	return NULL;
@@ -5499,8 +5474,8 @@ static int iwl4965_init_channel_map(struct iwl4965_priv *priv)
 		/* Loop through each band adding each of the channels */
 		for (ch = 0; ch < eeprom_ch_count; ch++) {
 			ch_info->channel = eeprom_ch_index[ch];
-			ch_info->phymode = (band == 1) ? MODE_IEEE80211B :
-			    MODE_IEEE80211A;
+			ch_info->band = (band == 1) ? IEEE80211_BAND_2GHZ :
+			    IEEE80211_BAND_5GHZ;
 
 			/* permanently store EEPROM's channel regulatory flags
 			 *   and max power in channel info database. */
@@ -5559,14 +5534,14 @@ static int iwl4965_init_channel_map(struct iwl4965_priv *priv)
 
 	/* Two additional EEPROM bands for 2.4 and 5 GHz FAT channels */
 	for (band = 6; band <= 7; band++) {
-		int phymode;
+		enum ieee80211_band ieeeband;
 		u8 fat_extension_chan;
 
 		iwl4965_init_band_reference(priv, band, &eeprom_ch_count,
 					&eeprom_ch_info, &eeprom_ch_index);
 
 		/* EEPROM band 6 is 2.4, band 7 is 5 GHz */
-		phymode = (band == 6) ? MODE_IEEE80211B : MODE_IEEE80211A;
+		ieeeband = (band == 6) ? IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ;
 
 		/* Loop through each band adding each of the channels */
 		for (ch = 0; ch < eeprom_ch_count; ch++) {
@@ -5580,13 +5555,13 @@ static int iwl4965_init_channel_map(struct iwl4965_priv *priv)
 				fat_extension_chan = HT_IE_EXT_CHANNEL_ABOVE;
 
 			/* Set up driver's info for lower half */
-			iwl4965_set_fat_chan_info(priv, phymode,
+			iwl4965_set_fat_chan_info(priv, ieeeband,
 						  eeprom_ch_index[ch],
 						  &(eeprom_ch_info[ch]),
 						  fat_extension_chan);
 
 			/* Set up driver's info for upper half */
-			iwl4965_set_fat_chan_info(priv, phymode,
+			iwl4965_set_fat_chan_info(priv, ieeeband,
 						  (eeprom_ch_index[ch] + 4),
 						  &(eeprom_ch_info[ch]),
 						  HT_IE_EXT_CHANNEL_BELOW);
@@ -5628,18 +5603,20 @@ static void iwl4965_free_channel_map(struct iwl4965_priv *priv)
 #define IWL_PASSIVE_DWELL_BASE      (100)
 #define IWL_CHANNEL_TUNE_TIME       5
 
-static inline u16 iwl4965_get_active_dwell_time(struct iwl4965_priv *priv, int phymode)
+static inline u16 iwl4965_get_active_dwell_time(struct iwl4965_priv *priv,
+						enum ieee80211_band band)
 {
-	if (phymode == MODE_IEEE80211A)
+	if (band == IEEE80211_BAND_5GHZ)
 		return IWL_ACTIVE_DWELL_TIME_52;
 	else
 		return IWL_ACTIVE_DWELL_TIME_24;
 }
 
-static u16 iwl4965_get_passive_dwell_time(struct iwl4965_priv *priv, int phymode)
+static u16 iwl4965_get_passive_dwell_time(struct iwl4965_priv *priv,
+					  enum ieee80211_band band)
 {
-	u16 active = iwl4965_get_active_dwell_time(priv, phymode);
-	u16 passive = (phymode != MODE_IEEE80211A) ?
+	u16 active = iwl4965_get_active_dwell_time(priv, band);
+	u16 passive = (band != IEEE80211_BAND_5GHZ) ?
 	    IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_24 :
 	    IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_52;
 
@@ -5659,28 +5636,29 @@ static u16 iwl4965_get_passive_dwell_time(struct iwl4965_priv *priv, int phymode
 	return passive;
 }
 
-static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode,
+static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv,
+					 enum ieee80211_band band,
 				     u8 is_active, u8 direct_mask,
 				     struct iwl4965_scan_channel *scan_ch)
 {
 	const struct ieee80211_channel *channels = NULL;
-	const struct ieee80211_hw_mode *hw_mode;
+	const struct ieee80211_supported_band *sband;
 	const struct iwl4965_channel_info *ch_info;
 	u16 passive_dwell = 0;
 	u16 active_dwell = 0;
 	int added, i;
 
-	hw_mode = iwl4965_get_hw_mode(priv, phymode);
-	if (!hw_mode)
+	sband = iwl4965_get_hw_mode(priv, band);
+	if (!sband)
 		return 0;
 
-	channels = hw_mode->channels;
+	channels = sband->channels;
 
-	active_dwell = iwl4965_get_active_dwell_time(priv, phymode);
-	passive_dwell = iwl4965_get_passive_dwell_time(priv, phymode);
+	active_dwell = iwl4965_get_active_dwell_time(priv, band);
+	passive_dwell = iwl4965_get_passive_dwell_time(priv, band);
 
-	for (i = 0, added = 0; i < hw_mode->num_channels; i++) {
-		if (channels[i].chan ==
+	for (i = 0, added = 0; i < sband->n_channels; i++) {
+		if (ieee80211_frequency_to_channel(channels[i].center_freq) ==
 		    le16_to_cpu(priv->active_rxon.channel)) {
 			if (iwl4965_is_associated(priv)) {
 				IWL_DEBUG_SCAN
@@ -5691,9 +5669,9 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode,
 		} else if (priv->only_active_channel)
 			continue;
 
-		scan_ch->channel = channels[i].chan;
+		scan_ch->channel = ieee80211_frequency_to_channel(channels[i].center_freq);
 
-		ch_info = iwl4965_get_channel_info(priv, phymode,
+		ch_info = iwl4965_get_channel_info(priv, band,
 					 scan_ch->channel);
 		if (!is_channel_valid(ch_info)) {
 			IWL_DEBUG_SCAN("Channel %d is INVALID for this SKU.\n",
@@ -5702,7 +5680,7 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode,
 		}
 
 		if (!is_active || is_channel_passive(ch_info) ||
-		    !(channels[i].flag & IEEE80211_CHAN_W_ACTIVE_SCAN))
+		    (channels[i].flags & IEEE80211_CHAN_PASSIVE_SCAN))
 			scan_ch->type = 0;	/* passive */
 		else
 			scan_ch->type = 1;	/* active */
@@ -5721,7 +5699,7 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode,
 		/* scan_pwr_info->tpc.dsp_atten; */
 
 		/*scan_pwr_info->tpc.tx_gain; */
-		if (phymode == MODE_IEEE80211A)
+		if (band == IEEE80211_BAND_5GHZ)
 			scan_ch->tpc.tx_gain = ((1 << 5) | (3 << 3)) | 3;
 		else {
 			scan_ch->tpc.tx_gain = ((1 << 5) | (5 << 3));
@@ -5745,41 +5723,23 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode,
 	return added;
 }
 
-static void iwl4965_reset_channel_flag(struct iwl4965_priv *priv)
-{
-	int i, j;
-	for (i = 0; i < 3; i++) {
-		struct ieee80211_hw_mode *hw_mode = (void *)&priv->modes[i];
-		for (j = 0; j < hw_mode->num_channels; j++)
-			hw_mode->channels[j].flag = hw_mode->channels[j].val;
-	}
-}
-
 static void iwl4965_init_hw_rates(struct iwl4965_priv *priv,
 			      struct ieee80211_rate *rates)
 {
 	int i;
 
 	for (i = 0; i < IWL_RATE_COUNT; i++) {
-		rates[i].rate = iwl4965_rates[i].ieee * 5;
-		rates[i].val = i; /* Rate scaling will work on indexes */
-		rates[i].val2 = i;
-		rates[i].flags = IEEE80211_RATE_SUPPORTED;
-		/* Only OFDM have the bits-per-symbol set */
-		if ((i <= IWL_LAST_OFDM_RATE) && (i >= IWL_FIRST_OFDM_RATE))
-			rates[i].flags |= IEEE80211_RATE_OFDM;
-		else {
+		rates[i].bitrate = iwl4965_rates[i].ieee * 5;
+		rates[i].hw_value = i; /* Rate scaling will work on indexes */
+		rates[i].hw_value_short = i;
+		rates[i].flags = 0;
+		if ((i > IWL_LAST_OFDM_RATE) || (i < IWL_FIRST_OFDM_RATE)) {
 			/*
-			 * If CCK 1M then set rate flag to CCK else CCK_2
-			 * which is CCK | PREAMBLE2
+			 * If CCK != 1M then set short preamble rate flag.
 			 */
 			rates[i].flags |= (iwl4965_rates[i].plcp == 10) ?
-				IEEE80211_RATE_CCK : IEEE80211_RATE_CCK_2;
+				0 : IEEE80211_RATE_SHORT_PREAMBLE;
 		}
-
-		/* Set up which ones are basic rates... */
-		if (IWL_BASIC_RATES_MASK & (1 << i))
-			rates[i].flags |= IEEE80211_RATE_BASIC;
 	}
 }
 
@@ -5789,74 +5749,47 @@ static void iwl4965_init_hw_rates(struct iwl4965_priv *priv,
 static int iwl4965_init_geos(struct iwl4965_priv *priv)
 {
 	struct iwl4965_channel_info *ch;
-	struct ieee80211_hw_mode *modes;
+	struct ieee80211_supported_band *band;
 	struct ieee80211_channel *channels;
 	struct ieee80211_channel *geo_ch;
 	struct ieee80211_rate *rates;
 	int i = 0;
-	enum {
-		A = 0,
-		B = 1,
-		G = 2,
-	};
-	int mode_count = 3;
 
-	if (priv->modes) {
+	if (priv->bands[IEEE80211_BAND_2GHZ].n_bitrates ||
+	    priv->bands[IEEE80211_BAND_5GHZ].n_bitrates) {
 		IWL_DEBUG_INFO("Geography modes already initialized.\n");
 		set_bit(STATUS_GEO_CONFIGURED, &priv->status);
 		return 0;
 	}
 
-	modes = kzalloc(sizeof(struct ieee80211_hw_mode) * mode_count,
-			GFP_KERNEL);
-	if (!modes)
-		return -ENOMEM;
-
 	channels = kzalloc(sizeof(struct ieee80211_channel) *
 			   priv->channel_count, GFP_KERNEL);
-	if (!channels) {
-		kfree(modes);
+	if (!channels)
 		return -ENOMEM;
-	}
 
 	rates = kzalloc((sizeof(struct ieee80211_rate) * (IWL_MAX_RATES + 1)),
 			GFP_KERNEL);
 	if (!rates) {
-		kfree(modes);
 		kfree(channels);
 		return -ENOMEM;
 	}
 
-	/* 0 = 802.11a
-	 * 1 = 802.11b
-	 * 2 = 802.11g
-	 */
-
 	/* 5.2GHz channels start after the 2.4GHz channels */
-	modes[A].mode = MODE_IEEE80211A;
-	modes[A].channels = &channels[ARRAY_SIZE(iwl4965_eeprom_band_1)];
-	modes[A].rates = rates;
-	modes[A].num_rates = 8;	/* just OFDM */
-	modes[A].rates = &rates[4];
-	modes[A].num_channels = 0;
 #ifdef CONFIG_IWL4965_HT
 	iwl4965_init_ht_hw_capab(&modes[A].ht_info, MODE_IEEE80211A);
 #endif
-
-	modes[B].mode = MODE_IEEE80211B;
-	modes[B].channels = channels;
-	modes[B].rates = rates;
-	modes[B].num_rates = 4;	/* just CCK */
-	modes[B].num_channels = 0;
-
-	modes[G].mode = MODE_IEEE80211G;
-	modes[G].channels = channels;
-	modes[G].rates = rates;
-	modes[G].num_rates = 12;	/* OFDM & CCK */
-	modes[G].num_channels = 0;
 #ifdef CONFIG_IWL4965_HT
 	iwl4965_init_ht_hw_capab(&modes[G].ht_info, MODE_IEEE80211G);
 #endif
+	band = &priv->bands[IEEE80211_BAND_5GHZ];
+	band->channels = &channels[ARRAY_SIZE(iwl4965_eeprom_band_1)];
+	band->bitrates = &rates[4];
+	band->n_bitrates = 8;	/* just OFDM */
+
+	band = &priv->bands[IEEE80211_BAND_2GHZ];
+	band->channels = channels;
+	band->bitrates = rates;
+	band->n_bitrates = 12;	/* OFDM & CCK */
 
 	priv->ieee_channels = channels;
 	priv->ieee_rates = rates;
@@ -5875,37 +5808,32 @@ static int iwl4965_init_geos(struct iwl4965_priv *priv)
 		}
 
 		if (is_channel_a_band(ch)) {
-			geo_ch = &modes[A].channels[modes[A].num_channels++];
-		} else {
-			geo_ch = &modes[B].channels[modes[B].num_channels++];
-			modes[G].num_channels++;
-		}
+			geo_ch = &priv->bands[IEEE80211_BAND_5GHZ].channels[priv->bands[IEEE80211_BAND_5GHZ].n_channels++];
+		} else
+			geo_ch = &priv->bands[IEEE80211_BAND_2GHZ].channels[priv->bands[IEEE80211_BAND_2GHZ].n_channels++];
 
-		geo_ch->freq = ieee80211chan2mhz(ch->channel);
-		geo_ch->chan = ch->channel;
-		geo_ch->power_level = ch->max_power_avg;
-		geo_ch->antenna_max = 0xff;
+		geo_ch->center_freq = ieee80211chan2mhz(ch->channel);
+		geo_ch->max_power = ch->max_power_avg;
+		geo_ch->max_antenna_gain = 0xff;
 
 		if (is_channel_valid(ch)) {
-			geo_ch->flag = IEEE80211_CHAN_W_SCAN;
-			if (ch->flags & EEPROM_CHANNEL_IBSS)
-				geo_ch->flag |= IEEE80211_CHAN_W_IBSS;
+			if (!(ch->flags & EEPROM_CHANNEL_IBSS))
+				geo_ch->flags |= IEEE80211_CHAN_NO_IBSS;
 
-			if (ch->flags & EEPROM_CHANNEL_ACTIVE)
-				geo_ch->flag |= IEEE80211_CHAN_W_ACTIVE_SCAN;
+			if (!(ch->flags & EEPROM_CHANNEL_ACTIVE))
+				geo_ch->flags |= IEEE80211_CHAN_PASSIVE_SCAN;
 
 			if (ch->flags & EEPROM_CHANNEL_RADAR)
-				geo_ch->flag |= IEEE80211_CHAN_W_RADAR_DETECT;
+				geo_ch->flags |= IEEE80211_CHAN_RADAR;
 
 			if (ch->max_power_avg > priv->max_channel_txpower_limit)
 				priv->max_channel_txpower_limit =
 				    ch->max_power_avg;
-		}
-
-		geo_ch->val = geo_ch->flag;
+		} else
+			geo_ch->flags |= IEEE80211_CHAN_DISABLED;
 	}
 
-	if ((modes[A].num_channels == 0) && priv->is_abg) {
+	if ((priv->bands[IEEE80211_BAND_5GHZ].n_channels == 0) && priv->is_abg) {
 		printk(KERN_INFO DRV_NAME
 		       ": Incorrectly detected BG card as ABG.  Please send "
 		       "your PCI ID 0x%04X:0x%04X to maintainer.\n",
@@ -5915,24 +5843,12 @@ static int iwl4965_init_geos(struct iwl4965_priv *priv)
 
 	printk(KERN_INFO DRV_NAME
 	       ": Tunable channels: %d 802.11bg, %d 802.11a channels\n",
-	       modes[G].num_channels, modes[A].num_channels);
-
-	/*
-	 * NOTE:  We register these in preference of order -- the
-	 * stack doesn't currently (as of 7.0.6 / Apr 24 '07) pick
-	 * a phymode based on rates or AP capabilities but seems to
-	 * configure it purely on if the channel being configured
-	 * is supported by a mode -- and the first match is taken
-	 */
+	       priv->bands[IEEE80211_BAND_2GHZ].n_channels,
+	       priv->bands[IEEE80211_BAND_5GHZ].n_channels);
 
-	if (modes[G].num_channels)
-		ieee80211_register_hwmode(priv->hw, &modes[G]);
-	if (modes[B].num_channels)
-		ieee80211_register_hwmode(priv->hw, &modes[B]);
-	if (modes[A].num_channels)
-		ieee80211_register_hwmode(priv->hw, &modes[A]);
+	priv->hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->bands[IEEE80211_BAND_2GHZ];
+	priv->hw->wiphy->bands[IEEE80211_BAND_5GHZ] = &priv->bands[IEEE80211_BAND_5GHZ];
 
-	priv->modes = modes;
 	set_bit(STATUS_GEO_CONFIGURED, &priv->status);
 
 	return 0;
@@ -5943,7 +5859,6 @@ static int iwl4965_init_geos(struct iwl4965_priv *priv)
  */
 static void iwl4965_free_geos(struct iwl4965_priv *priv)
 {
-	kfree(priv->modes);
 	kfree(priv->ieee_channels);
 	kfree(priv->ieee_rates);
 	clear_bit(STATUS_GEO_CONFIGURED, &priv->status);
@@ -6945,7 +6860,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data)
 	struct iwl4965_scan_cmd *scan;
 	struct ieee80211_conf *conf = NULL;
 	u8 direct_mask;
-	int phymode;
+	enum ieee80211_band band;
 
 	conf = ieee80211_get_hw_conf(priv->hw);
 
@@ -7075,7 +6990,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data)
 				RATE_MCS_ANT_B_MSK|RATE_MCS_CCK_MSK);
 
 		scan->good_CRC_th = 0;
-		phymode = MODE_IEEE80211G;
+		band = IEEE80211_BAND_2GHZ;
 		break;
 
 	case 1:
@@ -7083,7 +6998,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data)
 				iwl4965_hw_set_rate_n_flags(IWL_RATE_6M_PLCP,
 				RATE_MCS_ANT_B_MSK);
 		scan->good_CRC_th = IWL_GOOD_CRC_TH;
-		phymode = MODE_IEEE80211A;
+		band = IEEE80211_BAND_5GHZ;
 		break;
 
 	default:
@@ -7113,7 +7028,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data)
 
 	scan->channel_count =
 		iwl4965_get_channels_for_scan(
-			priv, phymode, 1, /* active */
+			priv, band, 1, /* active */
 			direct_mask,
 			(void *)&scan->data[le16_to_cpu(scan->tx_cmd.len)]);
 
@@ -7463,7 +7378,7 @@ static int iwl4965_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	}
 
 	IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len,
-		     ctl->tx_rate);
+		     ctl->tx_rate->bitrate);
 
 	if (iwl4965_tx_skb(priv, skb, ctl))
 		dev_kfree_skb_any(skb);
@@ -7522,7 +7437,7 @@ static int iwl4965_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co
 	int ret = 0;
 
 	mutex_lock(&priv->mutex);
-	IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel);
+	IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel->hw_value);
 
 	priv->add_radiotap = !!(conf->flags & IEEE80211_CONF_RADIOTAP);
 
@@ -7542,10 +7457,9 @@ static int iwl4965_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co
 
 	spin_lock_irqsave(&priv->lock, flags);
 
-	ch_info = iwl4965_get_channel_info(priv, conf->phymode, conf->channel);
+	ch_info = iwl4965_get_channel_info(priv, conf->channel->band,
+			ieee80211_frequency_to_channel(conf->channel->center_freq));
 	if (!is_channel_valid(ch_info)) {
-		IWL_DEBUG_SCAN("Channel %d [%d] is INVALID for this SKU.\n",
-			       conf->channel, conf->phymode);
 		IWL_DEBUG_MAC80211("leave - invalid channel\n");
 		spin_unlock_irqrestore(&priv->lock, flags);
 		ret = -EINVAL;
@@ -7564,12 +7478,13 @@ static int iwl4965_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co
 		priv->staging_rxon.flags = 0;
 #endif /* CONFIG_IWL4965_HT */
 
-	iwl4965_set_rxon_channel(priv, conf->phymode, conf->channel);
+	iwl4965_set_rxon_channel(priv, conf->channel->band,
+		ieee80211_frequency_to_channel(conf->channel->center_freq));
 
-	iwl4965_set_flags_for_phymode(priv, conf->phymode);
+	iwl4965_set_flags_for_phymode(priv, conf->channel->band);
 
 	/* The list of supported rates and rate mask can be different
-	 * for each phymode; since the phymode may have changed, reset
+	 * for each band; since the band may have changed, reset
 	 * the rate mask to what mac80211 lists */
 	iwl4965_set_rate(priv);
 
@@ -7839,7 +7754,7 @@ static void iwl4965_bss_info_changed(struct ieee80211_hw *hw,
 	}
 
 	if (changes & BSS_CHANGED_ERP_CTS_PROT) {
-		if (bss_conf->use_cts_prot && (priv->phymode != MODE_IEEE80211A))
+		if (bss_conf->use_cts_prot && (priv->band != IEEE80211_BAND_5GHZ))
 			priv->staging_rxon.flags |= RXON_FLG_TGG_PROTECT_MSK;
 		else
 			priv->staging_rxon.flags &= ~RXON_FLG_TGG_PROTECT_MSK;
@@ -8277,7 +8192,6 @@ static void iwl4965_set_ht_capab(struct ieee80211_hw *hw,
 			u8 use_current_config)
 {
 	struct ieee80211_conf *conf = &hw->conf;
-	struct ieee80211_hw_mode *mode = conf->mode;
 
 	if (use_current_config) {
 		ht_cap->cap_info = cpu_to_le16(conf->ht_conf.cap);
@@ -8488,65 +8402,6 @@ static ssize_t store_filter_flags(struct device *d,
 static DEVICE_ATTR(filter_flags, S_IWUSR | S_IRUGO, show_filter_flags,
 		   store_filter_flags);
 
-static ssize_t show_tune(struct device *d,
-			 struct device_attribute *attr, char *buf)
-{
-	struct iwl4965_priv *priv = (struct iwl4965_priv *)d->driver_data;
-
-	return sprintf(buf, "0x%04X\n",
-		       (priv->phymode << 8) |
-			le16_to_cpu(priv->active_rxon.channel));
-}
-
-static void iwl4965_set_flags_for_phymode(struct iwl4965_priv *priv, u8 phymode);
-
-static ssize_t store_tune(struct device *d,
-			  struct device_attribute *attr,
-			  const char *buf, size_t count)
-{
-	struct iwl4965_priv *priv = (struct iwl4965_priv *)d->driver_data;
-	char *p = (char *)buf;
-	u16 tune = simple_strtoul(p, &p, 0);
-	u8 phymode = (tune >> 8) & 0xff;
-	u16 channel = tune & 0xff;
-
-	IWL_DEBUG_INFO("Tune request to:%d channel:%d\n", phymode, channel);
-
-	mutex_lock(&priv->mutex);
-	if ((le16_to_cpu(priv->staging_rxon.channel) != channel) ||
-	    (priv->phymode != phymode)) {
-		const struct iwl4965_channel_info *ch_info;
-
-		ch_info = iwl4965_get_channel_info(priv, phymode, channel);
-		if (!ch_info) {
-			IWL_WARNING("Requested invalid phymode/channel "
-				    "combination: %d %d\n", phymode, channel);
-			mutex_unlock(&priv->mutex);
-			return -EINVAL;
-		}
-
-		/* Cancel any currently running scans... */
-		if (iwl4965_scan_cancel_timeout(priv, 100))
-			IWL_WARNING("Could not cancel scan.\n");
-		else {
-			IWL_DEBUG_INFO("Committing phymode and "
-				       "rxon.channel = %d %d\n",
-				       phymode, channel);
-
-			iwl4965_set_rxon_channel(priv, phymode, channel);
-			iwl4965_set_flags_for_phymode(priv, phymode);
-
-			iwl4965_set_rate(priv);
-			iwl4965_commit_rxon(priv);
-		}
-	}
-	mutex_unlock(&priv->mutex);
-
-	return count;
-}
-
-static DEVICE_ATTR(tune, S_IWUSR | S_IRUGO, show_tune, store_tune);
-
 #ifdef CONFIG_IWL4965_SPECTRUM_MEASUREMENT
 
 static ssize_t show_measurement(struct device *d,
@@ -8736,73 +8591,8 @@ static DEVICE_ATTR(power_level, S_IWUSR | S_IRUSR, show_power_level,
 static ssize_t show_channels(struct device *d,
 			     struct device_attribute *attr, char *buf)
 {
-	struct iwl4965_priv *priv = dev_get_drvdata(d);
-	int len = 0, i;
-	struct ieee80211_channel *channels = NULL;
-	const struct ieee80211_hw_mode *hw_mode = NULL;
-	int count = 0;
-
-	if (!iwl4965_is_ready(priv))
-		return -EAGAIN;
-
-	hw_mode = iwl4965_get_hw_mode(priv, MODE_IEEE80211G);
-	if (!hw_mode)
-		hw_mode = iwl4965_get_hw_mode(priv, MODE_IEEE80211B);
-	if (hw_mode) {
-		channels = hw_mode->channels;
-		count = hw_mode->num_channels;
-	}
-
-	len +=
-	    sprintf(&buf[len],
-		    "Displaying %d channels in 2.4GHz band "
-		    "(802.11bg):\n", count);
-
-	for (i = 0; i < count; i++)
-		len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n",
-			       channels[i].chan,
-			       channels[i].power_level,
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_RADAR_DETECT ?
-			       " (IEEE 802.11h required)" : "",
-			       (!(channels[i].flag & IEEE80211_CHAN_W_IBSS)
-				|| (channels[i].
-				    flag &
-				    IEEE80211_CHAN_W_RADAR_DETECT)) ? "" :
-			       ", IBSS",
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_ACTIVE_SCAN ?
-			       "active/passive" : "passive only");
-
-	hw_mode = iwl4965_get_hw_mode(priv, MODE_IEEE80211A);
-	if (hw_mode) {
-		channels = hw_mode->channels;
-		count = hw_mode->num_channels;
-	} else {
-		channels = NULL;
-		count = 0;
-	}
-
-	len += sprintf(&buf[len], "Displaying %d channels in 5.2GHz band "
-		       "(802.11a):\n", count);
-
-	for (i = 0; i < count; i++)
-		len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n",
-			       channels[i].chan,
-			       channels[i].power_level,
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_RADAR_DETECT ?
-			       " (IEEE 802.11h required)" : "",
-			       (!(channels[i].flag & IEEE80211_CHAN_W_IBSS)
-				|| (channels[i].
-				    flag &
-				    IEEE80211_CHAN_W_RADAR_DETECT)) ? "" :
-			       ", IBSS",
-			       channels[i].
-			       flag & IEEE80211_CHAN_W_ACTIVE_SCAN ?
-			       "active/passive" : "passive only");
-
-	return len;
+	/* all this shit doesn't belong into sysfs anyway */
+	return 0;
 }
 
 static DEVICE_ATTR(channels, S_IRUSR, show_channels, NULL);
@@ -8981,7 +8771,6 @@ static struct attribute *iwl4965_sysfs_entries[] = {
 	&dev_attr_statistics.attr,
 	&dev_attr_status.attr,
 	&dev_attr_temperature.attr,
-	&dev_attr_tune.attr,
 	&dev_attr_tx_power.attr,
 
 	NULL
@@ -9109,7 +8898,7 @@ static int iwl4965_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
 	priv->data_retry_limit = -1;
 	priv->ieee_channels = NULL;
 	priv->ieee_rates = NULL;
-	priv->phymode = -1;
+	priv->band = IEEE80211_BAND_2GHZ;
 
 	err = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
 	if (!err)
@@ -9175,7 +8964,7 @@ static int iwl4965_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
 	priv->qos_data.qos_cap.val = 0;
 #endif /* CONFIG_IWL4965_QOS */
 
-	iwl4965_set_rxon_channel(priv, MODE_IEEE80211G, 6);
+	iwl4965_set_rxon_channel(priv, IEEE80211_BAND_2GHZ, 6);
 	iwl4965_setup_deferred_work(priv);
 	iwl4965_setup_rx_handlers(priv);
 
@@ -9226,7 +9015,6 @@ static int iwl4965_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
 		IWL_ERROR("initializing geos failed: %d\n", err);
 		goto out_free_channel_map;
 	}
-	iwl4965_reset_channel_flag(priv);
 
 	iwl4965_rate_control_register(priv->hw);
 	err = ieee80211_register_hw(priv->hw);
diff --git a/drivers/net/wireless/p54.h b/drivers/net/wireless/p54.h
index 744c866066c5..06d2c67f4c81 100644
--- a/drivers/net/wireless/p54.h
+++ b/drivers/net/wireless/p54.h
@@ -64,10 +64,6 @@ struct p54_common {
 	unsigned int tx_hdr_len;
 	void *cached_vdcf;
 	unsigned int fw_var;
-	/* FIXME: this channels/modes/rates stuff sucks */
-	struct ieee80211_channel channels[14];
-	struct ieee80211_rate rates[12];
-	struct ieee80211_hw_mode modes[2];
 	struct ieee80211_tx_queue_stats tx_stats;
 };
 
diff --git a/drivers/net/wireless/p54common.c b/drivers/net/wireless/p54common.c
index 5cda49aff3a8..218ff7770ef6 100644
--- a/drivers/net/wireless/p54common.c
+++ b/drivers/net/wireless/p54common.c
@@ -27,6 +27,46 @@ MODULE_DESCRIPTION("Softmac Prism54 common code");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("prism54common");
 
+static struct ieee80211_rate p54_rates[] = {
+	{ .bitrate = 10, .hw_value = 0, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 20, .hw_value = 1, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 55, .hw_value = 2, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 110, .hw_value = 3, .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 60, .hw_value = 4, },
+	{ .bitrate = 90, .hw_value = 5, },
+	{ .bitrate = 120, .hw_value = 6, },
+	{ .bitrate = 180, .hw_value = 7, },
+	{ .bitrate = 240, .hw_value = 8, },
+	{ .bitrate = 360, .hw_value = 9, },
+	{ .bitrate = 480, .hw_value = 10, },
+	{ .bitrate = 540, .hw_value = 11, },
+};
+
+static struct ieee80211_channel p54_channels[] = {
+	{ .center_freq = 2412, .hw_value = 1, },
+	{ .center_freq = 2417, .hw_value = 2, },
+	{ .center_freq = 2422, .hw_value = 3, },
+	{ .center_freq = 2427, .hw_value = 4, },
+	{ .center_freq = 2432, .hw_value = 5, },
+	{ .center_freq = 2437, .hw_value = 6, },
+	{ .center_freq = 2442, .hw_value = 7, },
+	{ .center_freq = 2447, .hw_value = 8, },
+	{ .center_freq = 2452, .hw_value = 9, },
+	{ .center_freq = 2457, .hw_value = 10, },
+	{ .center_freq = 2462, .hw_value = 11, },
+	{ .center_freq = 2467, .hw_value = 12, },
+	{ .center_freq = 2472, .hw_value = 13, },
+	{ .center_freq = 2484, .hw_value = 14, },
+};
+
+struct ieee80211_supported_band band_2GHz = {
+	.channels = p54_channels,
+	.n_channels = ARRAY_SIZE(p54_channels),
+	.bitrates = p54_rates,
+	.n_bitrates = ARRAY_SIZE(p54_rates),
+};
+
+
 void p54_parse_firmware(struct ieee80211_hw *dev, const struct firmware *fw)
 {
 	struct p54_common *priv = dev->priv;
@@ -308,10 +348,10 @@ static void p54_rx_data(struct ieee80211_hw *dev, struct sk_buff *skb)
 	u16 freq = le16_to_cpu(hdr->freq);
 
 	rx_status.ssi = hdr->rssi;
-	rx_status.rate = hdr->rate & 0x1f; /* report short preambles & CCK too */
-	rx_status.channel = freq == 2484 ? 14 : (freq - 2407)/5;
+	/* XX correct? */
+	rx_status.rate_idx = hdr->rate & 0xf;
 	rx_status.freq = freq;
-	rx_status.phymode = MODE_IEEE80211G;
+	rx_status.band = IEEE80211_BAND_2GHZ;
 	rx_status.antenna = hdr->antenna;
 	rx_status.mactime = le64_to_cpu(hdr->timestamp);
 	rx_status.flag |= RX_FLAG_TSFT;
@@ -547,7 +587,9 @@ static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	txhdr->padding2 = 0;
 
 	/* TODO: add support for alternate retry TX rates */
-	rate = control->tx_rate;
+	rate = control->tx_rate->hw_value;
+	if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
+		rate |= 0x10;
 	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS)
 		rate |= 0x40;
 	else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
@@ -849,7 +891,7 @@ static int p54_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf)
 {
 	int ret;
 
-	ret = p54_set_freq(dev, cpu_to_le16(conf->freq));
+	ret = p54_set_freq(dev, cpu_to_le16(conf->channel->center_freq));
 	p54_set_vdcf(dev);
 	return ret;
 }
@@ -944,7 +986,6 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len)
 {
 	struct ieee80211_hw *dev;
 	struct p54_common *priv;
-	int i;
 
 	dev = ieee80211_alloc_hw(priv_data_len, &p54_ops);
 	if (!dev)
@@ -953,18 +994,7 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len)
 	priv = dev->priv;
 	priv->mode = IEEE80211_IF_TYPE_INVALID;
 	skb_queue_head_init(&priv->tx_queue);
-	memcpy(priv->channels, p54_channels, sizeof(p54_channels));
-	memcpy(priv->rates, p54_rates, sizeof(p54_rates));
-	priv->modes[1].mode = MODE_IEEE80211B;
-	priv->modes[1].num_rates = 4;
-	priv->modes[1].rates = priv->rates;
-	priv->modes[1].num_channels = ARRAY_SIZE(p54_channels);
-	priv->modes[1].channels = priv->channels;
-	priv->modes[0].mode = MODE_IEEE80211G;
-	priv->modes[0].num_rates = ARRAY_SIZE(p54_rates);
-	priv->modes[0].rates = priv->rates;
-	priv->modes[0].num_channels = ARRAY_SIZE(p54_channels);
-	priv->modes[0].channels = priv->channels;
+	dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &band_2GHz;
 	dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING | /* not sure */
 		    IEEE80211_HW_RX_INCLUDES_FCS;
 	dev->channel_change_time = 1000;	/* TODO: find actual value */
@@ -986,14 +1016,6 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len)
 
 	p54_init_vdcf(dev);
 
-	for (i = 0; i < 2; i++) {
-		if (ieee80211_register_hwmode(dev, &priv->modes[i])) {
-			kfree(priv->cached_vdcf);
-			ieee80211_free_hw(dev);
-			return NULL;
-		}
-	}
-
 	return dev;
 }
 EXPORT_SYMBOL_GPL(p54_init_common);
diff --git a/drivers/net/wireless/p54common.h b/drivers/net/wireless/p54common.h
index a721334e20d9..dc9f4cef585e 100644
--- a/drivers/net/wireless/p54common.h
+++ b/drivers/net/wireless/p54common.h
@@ -251,79 +251,4 @@ struct p54_tx_control_vdcf {
 	__le16 frameburst;
 } __attribute__ ((packed));
 
-static const struct ieee80211_rate p54_rates[] = {
-	{ .rate = 10,
-	  .val = 0,
-	  .val2 = 0x10,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 20,
-	  .val = 1,
-	  .val2 = 0x11,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 55,
-	  .val = 2,
-	  .val2 = 0x12,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 110,
-	  .val = 3,
-	  .val2 = 0x13,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 60,
-	  .val = 4,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 90,
-	  .val = 5,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 120,
-	  .val = 6,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 180,
-	  .val = 7,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 240,
-	  .val = 8,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 360,
-	  .val = 9,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 480,
-	  .val = 10,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 540,
-	  .val = 11,
-	  .flags = IEEE80211_RATE_OFDM },
-};
-
-// TODO: just generate this..
-static const struct ieee80211_channel p54_channels[] = {
-	{ .chan = 1,
-	  .freq = 2412},
-	{ .chan = 2,
-	  .freq = 2417},
-	{ .chan = 3,
-	  .freq = 2422},
-	{ .chan = 4,
-	  .freq = 2427},
-	{ .chan = 5,
-	  .freq = 2432},
-	{ .chan = 6,
-	  .freq = 2437},
-	{ .chan = 7,
-	  .freq = 2442},
-	{ .chan = 8,
-	  .freq = 2447},
-	{ .chan = 9,
-	  .freq = 2452},
-	{ .chan = 10,
-	  .freq = 2457},
-	{ .chan = 11,
-	  .freq = 2462},
-	{ .chan = 12,
-	  .freq = 2467},
-	{ .chan = 13,
-	  .freq = 2472},
-	{ .chan = 14,
-	  .freq = 2484}
-};
-
 #endif /* PRISM54COMMON_H */
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index b0e4ea7c9dca..4fa762bdb734 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -390,6 +390,10 @@ static inline struct rt2x00_intf* vif_to_intf(struct ieee80211_vif *vif)
 	return (struct rt2x00_intf *)vif->drv_priv;
 }
 
+#define HWMODE_B	0
+#define HWMODE_G	1
+#define HWMODE_A	2
+
 /*
  * Details about the supported modes, rates and channels
  * of a particular chipset. This is used by rt2x00lib
@@ -644,11 +648,8 @@ struct rt2x00_dev {
 	 * IEEE80211 control structure.
 	 */
 	struct ieee80211_hw *hw;
-	struct ieee80211_hw_mode *hwmodes;
-	unsigned int curr_hwmode;
-#define HWMODE_B	0
-#define HWMODE_G	1
-#define HWMODE_A	2
+	struct ieee80211_supported_band bands[IEEE80211_NUM_BANDS];
+	enum ieee80211_band curr_band;
 
 	/*
 	 * rfkill structure for RF state switching support.
diff --git a/drivers/net/wireless/rt2x00/rt2x00config.c b/drivers/net/wireless/rt2x00/rt2x00config.c
index 20231e0c53fa..9fba485a40ac 100644
--- a/drivers/net/wireless/rt2x00/rt2x00config.c
+++ b/drivers/net/wireless/rt2x00/rt2x00config.c
@@ -152,7 +152,7 @@ void rt2x00lib_config(struct rt2x00_dev *rt2x00dev,
 		      struct ieee80211_conf *conf, const int force_config)
 {
 	struct rt2x00lib_conf libconf;
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *band;
 	struct ieee80211_rate *rate;
 	struct antenna_setup *default_ant = &rt2x00dev->default_ant;
 	struct antenna_setup *active_ant = &rt2x00dev->link.ant.active;
@@ -172,9 +172,9 @@ void rt2x00lib_config(struct rt2x00_dev *rt2x00dev,
 	 * Check which configuration options have been
 	 * updated and should be send to the device.
 	 */
-	if (rt2x00dev->rx_status.phymode != conf->phymode)
+	if (rt2x00dev->rx_status.band != conf->channel->band)
 		flags |= CONFIG_UPDATE_PHYMODE;
-	if (rt2x00dev->rx_status.channel != conf->channel)
+	if (rt2x00dev->rx_status.freq != conf->channel->center_freq)
 		flags |= CONFIG_UPDATE_CHANNEL;
 	if (rt2x00dev->tx_power != conf->power_level)
 		flags |= CONFIG_UPDATE_TXPOWER;
@@ -229,33 +229,31 @@ config:
 	memset(&libconf, 0, sizeof(libconf));
 
 	if (flags & CONFIG_UPDATE_PHYMODE) {
-		switch (conf->phymode) {
-		case MODE_IEEE80211A:
+		switch (conf->channel->band) {
+		case IEEE80211_BAND_5GHZ:
 			libconf.phymode = HWMODE_A;
 			break;
-		case MODE_IEEE80211B:
-			libconf.phymode = HWMODE_B;
-			break;
-		case MODE_IEEE80211G:
+		case IEEE80211_BAND_2GHZ:
+			/* Uh oh. what about B? */
 			libconf.phymode = HWMODE_G;
 			break;
 		default:
 			ERROR(rt2x00dev,
 			      "Attempt to configure unsupported mode (%d)"
-			      "Defaulting to 802.11b", conf->phymode);
+			      "Defaulting to 802.11b", conf->channel->band);
 			libconf.phymode = HWMODE_B;
 		}
 
-		mode = &rt2x00dev->hwmodes[libconf.phymode];
-		rate = &mode->rates[mode->num_rates - 1];
+		band = &rt2x00dev->bands[conf->channel->band];
+		rate = &band->bitrates[band->n_bitrates - 1];
 
 		libconf.basic_rates =
-		    DEVICE_GET_RATE_FIELD(rate->val, RATEMASK) & DEV_BASIC_RATEMASK;
+		    DEVICE_GET_RATE_FIELD(rate->hw_value, RATEMASK) & DEV_BASIC_RATEMASK;
 	}
 
 	if (flags & CONFIG_UPDATE_CHANNEL) {
 		memcpy(&libconf.rf,
-		       &rt2x00dev->spec.channels[conf->channel_val],
+		       &rt2x00dev->spec.channels[conf->channel->hw_value],
 		       sizeof(libconf.rf));
 	}
 
@@ -301,12 +299,11 @@ config:
 		rt2x00lib_reset_link_tuner(rt2x00dev);
 
 	if (flags & CONFIG_UPDATE_PHYMODE) {
-		rt2x00dev->curr_hwmode = libconf.phymode;
-		rt2x00dev->rx_status.phymode = conf->phymode;
+		rt2x00dev->curr_band = conf->channel->band;
+		rt2x00dev->rx_status.band = conf->channel->band;
 	}
 
-	rt2x00dev->rx_status.freq = conf->freq;
-	rt2x00dev->rx_status.channel = conf->channel;
+	rt2x00dev->rx_status.freq = conf->channel->center_freq;
 	rt2x00dev->tx_power = conf->power_level;
 
 	if (flags & CONFIG_UPDATE_ANTENNA) {
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index 0df8062b1a8e..83a72ae36638 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -550,19 +550,19 @@ void rt2x00lib_rxdone(struct queue_entry *entry,
 {
 	struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev;
 	struct ieee80211_rx_status *rx_status = &rt2x00dev->rx_status;
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	struct ieee80211_rate *rate;
 	struct ieee80211_hdr *hdr;
 	unsigned int i;
-	int val = 0;
+	int val = 0, idx = -1;
 	u16 fc;
 
 	/*
 	 * Update RX statistics.
 	 */
-	mode = &rt2x00dev->hwmodes[rt2x00dev->curr_hwmode];
-	for (i = 0; i < mode->num_rates; i++) {
-		rate = &mode->rates[i];
+	sband = &rt2x00dev->bands[rt2x00dev->curr_band];
+	for (i = 0; i < sband->n_bitrates; i++) {
+		rate = &sband->bitrates[i];
 
 		/*
 		 * When frame was received with an OFDM bitrate,
@@ -570,12 +570,12 @@ void rt2x00lib_rxdone(struct queue_entry *entry,
 		 * a CCK bitrate the signal is the rate in 0.5kbit/s.
 		 */
 		if (!rxdesc->ofdm)
-			val = DEVICE_GET_RATE_FIELD(rate->val, RATE);
+			val = DEVICE_GET_RATE_FIELD(rate->hw_value, RATE);
 		else
-			val = DEVICE_GET_RATE_FIELD(rate->val, PLCP);
+			val = DEVICE_GET_RATE_FIELD(rate->hw_value, PLCP);
 
 		if (val == rxdesc->signal) {
-			val = rate->val;
+			idx = i;
 			break;
 		}
 	}
@@ -590,7 +590,7 @@ void rt2x00lib_rxdone(struct queue_entry *entry,
 
 	rt2x00dev->link.qual.rx_success++;
 
-	rx_status->rate = val;
+	rx_status->rate_idx = idx;
 	rx_status->signal =
 	    rt2x00lib_calculate_link_signal(rt2x00dev, rxdesc->rssi);
 	rx_status->ssi = rxdesc->rssi;
@@ -639,7 +639,7 @@ void rt2x00lib_write_tx_desc(struct rt2x00_dev *rt2x00dev,
 	frame_control = le16_to_cpu(ieee80211hdr->frame_control);
 	seq_ctrl = le16_to_cpu(ieee80211hdr->seq_ctrl);
 
-	tx_rate = control->tx_rate;
+	tx_rate = control->tx_rate->hw_value;
 
 	/*
 	 * Check whether this frame is to be acked
@@ -658,7 +658,7 @@ void rt2x00lib_write_tx_desc(struct rt2x00_dev *rt2x00dev,
 		} else
 			__clear_bit(ENTRY_TXD_ACK, &txdesc.flags);
 		if (control->rts_cts_rate)
-			tx_rate = control->rts_cts_rate;
+			tx_rate = control->rts_cts_rate->hw_value;
 	}
 
 	/*
@@ -760,54 +760,45 @@ static void rt2x00lib_channel(struct ieee80211_channel *entry,
 			      const int channel, const int tx_power,
 			      const int value)
 {
-	entry->chan = channel;
 	if (channel <= 14)
-		entry->freq = 2407 + (5 * channel);
+		entry->center_freq = 2407 + (5 * channel);
 	else
-		entry->freq = 5000 + (5 * channel);
-	entry->val = value;
-	entry->flag =
-	    IEEE80211_CHAN_W_IBSS |
-	    IEEE80211_CHAN_W_ACTIVE_SCAN |
-	    IEEE80211_CHAN_W_SCAN;
-	entry->power_level = tx_power;
-	entry->antenna_max = 0xff;
+		entry->center_freq = 5000 + (5 * channel);
+	entry->hw_value = value;
+	entry->max_power = tx_power;
+	entry->max_antenna_gain = 0xff;
 }
 
 static void rt2x00lib_rate(struct ieee80211_rate *entry,
 			   const int rate, const int mask,
 			   const int plcp, const int flags)
 {
-	entry->rate = rate;
-	entry->val =
+	entry->bitrate = rate;
+	entry->hw_value =
 	    DEVICE_SET_RATE_FIELD(rate, RATE) |
 	    DEVICE_SET_RATE_FIELD(mask, RATEMASK) |
 	    DEVICE_SET_RATE_FIELD(plcp, PLCP);
 	entry->flags = flags;
-	entry->val2 = entry->val;
-	if (entry->flags & IEEE80211_RATE_PREAMBLE2)
-		entry->val2 |= DEVICE_SET_RATE_FIELD(1, PREAMBLE);
-	entry->min_rssi_ack = 0;
-	entry->min_rssi_ack_delta = 0;
+	entry->hw_value_short = entry->hw_value;
+	if (entry->flags & IEEE80211_RATE_SHORT_PREAMBLE)
+		entry->hw_value_short |= DEVICE_SET_RATE_FIELD(1, PREAMBLE);
 }
 
 static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev,
 				    struct hw_mode_spec *spec)
 {
 	struct ieee80211_hw *hw = rt2x00dev->hw;
-	struct ieee80211_hw_mode *hwmodes;
+	struct ieee80211_supported_band *sbands;
 	struct ieee80211_channel *channels;
 	struct ieee80211_rate *rates;
 	unsigned int i;
 	unsigned char tx_power;
 
-	hwmodes = kzalloc(sizeof(*hwmodes) * spec->num_modes, GFP_KERNEL);
-	if (!hwmodes)
-		goto exit;
+	sbands = &rt2x00dev->bands[0];
 
 	channels = kzalloc(sizeof(*channels) * spec->num_channels, GFP_KERNEL);
 	if (!channels)
-		goto exit_free_modes;
+		return -ENOMEM;
 
 	rates = kzalloc(sizeof(*rates) * spec->num_rates, GFP_KERNEL);
 	if (!rates)
@@ -817,31 +808,31 @@ static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev,
 	 * Initialize Rate list.
 	 */
 	rt2x00lib_rate(&rates[0], 10, DEV_RATEMASK_1MB,
-		       0x00, IEEE80211_RATE_CCK);
+		       0x00, 0);
 	rt2x00lib_rate(&rates[1], 20, DEV_RATEMASK_2MB,
-		       0x01, IEEE80211_RATE_CCK_2);
+		       0x01, IEEE80211_RATE_SHORT_PREAMBLE);
 	rt2x00lib_rate(&rates[2], 55, DEV_RATEMASK_5_5MB,
-		       0x02, IEEE80211_RATE_CCK_2);
+		       0x02, IEEE80211_RATE_SHORT_PREAMBLE);
 	rt2x00lib_rate(&rates[3], 110, DEV_RATEMASK_11MB,
-		       0x03, IEEE80211_RATE_CCK_2);
+		       0x03, IEEE80211_RATE_SHORT_PREAMBLE);
 
 	if (spec->num_rates > 4) {
 		rt2x00lib_rate(&rates[4], 60, DEV_RATEMASK_6MB,
-			       0x0b, IEEE80211_RATE_OFDM);
+			       0x0b, 0);
 		rt2x00lib_rate(&rates[5], 90, DEV_RATEMASK_9MB,
-			       0x0f, IEEE80211_RATE_OFDM);
+			       0x0f, 0);
 		rt2x00lib_rate(&rates[6], 120, DEV_RATEMASK_12MB,
-			       0x0a, IEEE80211_RATE_OFDM);
+			       0x0a, 0);
 		rt2x00lib_rate(&rates[7], 180, DEV_RATEMASK_18MB,
-			       0x0e, IEEE80211_RATE_OFDM);
+			       0x0e, 0);
 		rt2x00lib_rate(&rates[8], 240, DEV_RATEMASK_24MB,
-			       0x09, IEEE80211_RATE_OFDM);
+			       0x09, 0);
 		rt2x00lib_rate(&rates[9], 360, DEV_RATEMASK_36MB,
-			       0x0d, IEEE80211_RATE_OFDM);
+			       0x0d, 0);
 		rt2x00lib_rate(&rates[10], 480, DEV_RATEMASK_48MB,
-			       0x08, IEEE80211_RATE_OFDM);
+			       0x08, 0);
 		rt2x00lib_rate(&rates[11], 540, DEV_RATEMASK_54MB,
-			       0x0c, IEEE80211_RATE_OFDM);
+			       0x0c, 0);
 	}
 
 	/*
@@ -862,27 +853,27 @@ static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev,
 	/*
 	 * Intitialize 802.11b
 	 * Rates: CCK.
-	 * Channels: OFDM.
+	 * Channels: 2.4 GHz
 	 */
 	if (spec->num_modes > HWMODE_B) {
-		hwmodes[HWMODE_B].mode = MODE_IEEE80211B;
-		hwmodes[HWMODE_B].num_channels = 14;
-		hwmodes[HWMODE_B].num_rates = 4;
-		hwmodes[HWMODE_B].channels = channels;
-		hwmodes[HWMODE_B].rates = rates;
+		sbands[IEEE80211_BAND_2GHZ].n_channels = 14;
+		sbands[IEEE80211_BAND_2GHZ].n_bitrates = 4;
+		sbands[IEEE80211_BAND_2GHZ].channels = channels;
+		sbands[IEEE80211_BAND_2GHZ].bitrates = rates;
+		hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &rt2x00dev->bands[IEEE80211_BAND_2GHZ];
 	}
 
 	/*
 	 * Intitialize 802.11g
 	 * Rates: CCK, OFDM.
-	 * Channels: OFDM.
+	 * Channels: 2.4 GHz
 	 */
 	if (spec->num_modes > HWMODE_G) {
-		hwmodes[HWMODE_G].mode = MODE_IEEE80211G;
-		hwmodes[HWMODE_G].num_channels = 14;
-		hwmodes[HWMODE_G].num_rates = spec->num_rates;
-		hwmodes[HWMODE_G].channels = channels;
-		hwmodes[HWMODE_G].rates = rates;
+		sbands[IEEE80211_BAND_2GHZ].n_channels = 14;
+		sbands[IEEE80211_BAND_2GHZ].n_bitrates = spec->num_rates;
+		sbands[IEEE80211_BAND_2GHZ].channels = channels;
+		sbands[IEEE80211_BAND_2GHZ].bitrates = rates;
+		hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &rt2x00dev->bands[IEEE80211_BAND_2GHZ];
 	}
 
 	/*
@@ -891,39 +882,17 @@ static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev,
 	 * Channels: OFDM, UNII, HiperLAN2.
 	 */
 	if (spec->num_modes > HWMODE_A) {
-		hwmodes[HWMODE_A].mode = MODE_IEEE80211A;
-		hwmodes[HWMODE_A].num_channels = spec->num_channels - 14;
-		hwmodes[HWMODE_A].num_rates = spec->num_rates - 4;
-		hwmodes[HWMODE_A].channels = &channels[14];
-		hwmodes[HWMODE_A].rates = &rates[4];
+		sbands[IEEE80211_BAND_5GHZ].n_channels = spec->num_channels - 14;
+		sbands[IEEE80211_BAND_5GHZ].n_bitrates = spec->num_rates - 4;
+		sbands[IEEE80211_BAND_5GHZ].channels = &channels[14];
+		sbands[IEEE80211_BAND_5GHZ].bitrates = &rates[4];
+		hw->wiphy->bands[IEEE80211_BAND_5GHZ] = &rt2x00dev->bands[IEEE80211_BAND_5GHZ];
 	}
 
-	if (spec->num_modes > HWMODE_G &&
-	    ieee80211_register_hwmode(hw, &hwmodes[HWMODE_G]))
-		goto exit_free_rates;
-
-	if (spec->num_modes > HWMODE_B &&
-	    ieee80211_register_hwmode(hw, &hwmodes[HWMODE_B]))
-		goto exit_free_rates;
-
-	if (spec->num_modes > HWMODE_A &&
-	    ieee80211_register_hwmode(hw, &hwmodes[HWMODE_A]))
-		goto exit_free_rates;
-
-	rt2x00dev->hwmodes = hwmodes;
-
 	return 0;
 
-exit_free_rates:
-	kfree(rates);
-
-exit_free_channels:
+ exit_free_channels:
 	kfree(channels);
-
-exit_free_modes:
-	kfree(hwmodes);
-
-exit:
 	ERROR(rt2x00dev, "Allocation ieee80211 modes failed.\n");
 	return -ENOMEM;
 }
@@ -933,11 +902,11 @@ static void rt2x00lib_remove_hw(struct rt2x00_dev *rt2x00dev)
 	if (test_bit(DEVICE_REGISTERED_HW, &rt2x00dev->flags))
 		ieee80211_unregister_hw(rt2x00dev->hw);
 
-	if (likely(rt2x00dev->hwmodes)) {
-		kfree(rt2x00dev->hwmodes->channels);
-		kfree(rt2x00dev->hwmodes->rates);
-		kfree(rt2x00dev->hwmodes);
-		rt2x00dev->hwmodes = NULL;
+	if (likely(rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ])) {
+		kfree(rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ]->channels);
+		kfree(rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ]->bitrates);
+		rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ] = NULL;
+		rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_5GHZ] = NULL;
 	}
 }
 
diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c
index 59e87a1d96a4..1dd30510ed1e 100644
--- a/drivers/net/wireless/rt2x00/rt61pci.c
+++ b/drivers/net/wireless/rt2x00/rt61pci.c
@@ -426,12 +426,12 @@ static void rt61pci_config_antenna_5x(struct rt2x00_dev *rt2x00dev,
 	case ANTENNA_HW_DIVERSITY:
 		rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 2);
 		rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END,
-				  (rt2x00dev->curr_hwmode != HWMODE_A));
+				  (rt2x00dev->curr_band != IEEE80211_BAND_5GHZ));
 		break;
 	case ANTENNA_A:
 		rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1);
 		rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0);
-		if (rt2x00dev->curr_hwmode == HWMODE_A)
+		if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ)
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0);
 		else
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3);
@@ -446,7 +446,7 @@ static void rt61pci_config_antenna_5x(struct rt2x00_dev *rt2x00dev,
 	case ANTENNA_B:
 		rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1);
 		rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0);
-		if (rt2x00dev->curr_hwmode == HWMODE_A)
+		if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ)
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3);
 		else
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0);
@@ -602,7 +602,7 @@ static void rt61pci_config_antenna(struct rt2x00_dev *rt2x00dev,
 	unsigned int i;
 	u32 reg;
 
-	if (rt2x00dev->curr_hwmode == HWMODE_A) {
+	if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) {
 		sel = antenna_sel_a;
 		lna = test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags);
 	} else {
@@ -616,10 +616,9 @@ static void rt61pci_config_antenna(struct rt2x00_dev *rt2x00dev,
 	rt2x00pci_register_read(rt2x00dev, PHY_CSR0, &reg);
 
 	rt2x00_set_field32(&reg, PHY_CSR0_PA_PE_BG,
-			   (rt2x00dev->curr_hwmode == HWMODE_B ||
-			    rt2x00dev->curr_hwmode == HWMODE_G));
+			   rt2x00dev->curr_band == IEEE80211_BAND_2GHZ);
 	rt2x00_set_field32(&reg, PHY_CSR0_PA_PE_A,
-			   (rt2x00dev->curr_hwmode == HWMODE_A));
+			   rt2x00dev->curr_band == IEEE80211_BAND_5GHZ);
 
 	rt2x00pci_register_write(rt2x00dev, PHY_CSR0, reg);
 
@@ -698,9 +697,9 @@ static void rt61pci_enable_led(struct rt2x00_dev *rt2x00dev)
 
 	rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_RADIO_STATUS, 1);
 	rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_A_STATUS,
-			   (rt2x00dev->rx_status.phymode == MODE_IEEE80211A));
+			   rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ);
 	rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_BG_STATUS,
-			   (rt2x00dev->rx_status.phymode != MODE_IEEE80211A));
+			   rt2x00dev->rx_status.band != IEEE80211_BAND_5GHZ);
 
 	arg0 = rt2x00dev->led_reg & 0xff;
 	arg1 = (rt2x00dev->led_reg >> 8) & 0xff;
@@ -798,7 +797,7 @@ static void rt61pci_link_tuner(struct rt2x00_dev *rt2x00dev)
 	/*
 	 * Determine r17 bounds.
 	 */
-	if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) {
+	if (rt2x00dev->rx_status.band == IEEE80211_BAND_2GHZ) {
 		low_bound = 0x28;
 		up_bound = 0x48;
 		if (test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags)) {
@@ -1544,8 +1543,10 @@ static void rt61pci_write_tx_desc(struct rt2x00_dev *rt2x00dev,
 	rt2x00_desc_write(txd, 2, word);
 
 	rt2x00_desc_read(txd, 5, &word);
+/* XXX: removed for now
 	rt2x00_set_field32(&word, TXD_W5_TX_POWER,
 			   TXPOWER_TO_DEV(control->power_level));
+ */
 	rt2x00_set_field32(&word, TXD_W5_WAITING_DMA_DONE_INT, 1);
 	rt2x00_desc_write(txd, 5, word);
 
@@ -1637,7 +1638,7 @@ static int rt61pci_agc_to_rssi(struct rt2x00_dev *rt2x00dev, int rxd_w1)
 		return 0;
 	}
 
-	if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) {
+	if (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ) {
 		if (test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags))
 			offset += 14;
 
diff --git a/drivers/net/wireless/rt2x00/rt73usb.c b/drivers/net/wireless/rt2x00/rt73usb.c
index 4b5bde8b53de..9cbc879da037 100644
--- a/drivers/net/wireless/rt2x00/rt73usb.c
+++ b/drivers/net/wireless/rt2x00/rt73usb.c
@@ -439,13 +439,13 @@ static void rt73usb_config_antenna_5x(struct rt2x00_dev *rt2x00dev,
 	case ANTENNA_HW_DIVERSITY:
 		rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 2);
 		temp = !test_bit(CONFIG_FRAME_TYPE, &rt2x00dev->flags)
-		       && (rt2x00dev->curr_hwmode != HWMODE_A);
+		       && (rt2x00dev->curr_band != IEEE80211_BAND_5GHZ);
 		rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, temp);
 		break;
 	case ANTENNA_A:
 		rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1);
 		rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0);
-		if (rt2x00dev->curr_hwmode == HWMODE_A)
+		if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ)
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0);
 		else
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3);
@@ -460,7 +460,7 @@ static void rt73usb_config_antenna_5x(struct rt2x00_dev *rt2x00dev,
 	case ANTENNA_B:
 		rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1);
 		rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0);
-		if (rt2x00dev->curr_hwmode == HWMODE_A)
+		if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ)
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3);
 		else
 			rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0);
@@ -555,7 +555,7 @@ static void rt73usb_config_antenna(struct rt2x00_dev *rt2x00dev,
 	unsigned int i;
 	u32 reg;
 
-	if (rt2x00dev->curr_hwmode == HWMODE_A) {
+	if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) {
 		sel = antenna_sel_a;
 		lna = test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags);
 	} else {
@@ -569,10 +569,9 @@ static void rt73usb_config_antenna(struct rt2x00_dev *rt2x00dev,
 	rt73usb_register_read(rt2x00dev, PHY_CSR0, &reg);
 
 	rt2x00_set_field32(&reg, PHY_CSR0_PA_PE_BG,
-			   (rt2x00dev->curr_hwmode == HWMODE_B ||
-			    rt2x00dev->curr_hwmode == HWMODE_G));
+			   (rt2x00dev->curr_band == IEEE80211_BAND_2GHZ));
 	rt2x00_set_field32(&reg, PHY_CSR0_PA_PE_A,
-			   (rt2x00dev->curr_hwmode == HWMODE_A));
+			   (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ));
 
 	rt73usb_register_write(rt2x00dev, PHY_CSR0, reg);
 
@@ -644,9 +643,9 @@ static void rt73usb_enable_led(struct rt2x00_dev *rt2x00dev)
 
 	rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_RADIO_STATUS, 1);
 	rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_A_STATUS,
-			   (rt2x00dev->rx_status.phymode == MODE_IEEE80211A));
+			   (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ));
 	rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_BG_STATUS,
-			   (rt2x00dev->rx_status.phymode != MODE_IEEE80211A));
+			   (rt2x00dev->rx_status.band != IEEE80211_BAND_5GHZ));
 
 	rt2x00usb_vendor_request_sw(rt2x00dev, USB_LED_CONTROL, 0x0000,
 				    rt2x00dev->led_reg, REGISTER_TIMEOUT);
@@ -736,7 +735,7 @@ static void rt73usb_link_tuner(struct rt2x00_dev *rt2x00dev)
 	/*
 	 * Determine r17 bounds.
 	 */
-	if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) {
+	if (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ) {
 		low_bound = 0x28;
 		up_bound = 0x48;
 
@@ -1278,8 +1277,10 @@ static void rt73usb_write_tx_desc(struct rt2x00_dev *rt2x00dev,
 	rt2x00_desc_write(txd, 2, word);
 
 	rt2x00_desc_read(txd, 5, &word);
+/* XXX: removed for now
 	rt2x00_set_field32(&word, TXD_W5_TX_POWER,
 			   TXPOWER_TO_DEV(control->power_level));
+ */
 	rt2x00_set_field32(&word, TXD_W5_WAITING_DMA_DONE_INT, 1);
 	rt2x00_desc_write(txd, 5, word);
 
@@ -1370,7 +1371,7 @@ static int rt73usb_agc_to_rssi(struct rt2x00_dev *rt2x00dev, int rxd_w1)
 		return 0;
 	}
 
-	if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) {
+	if (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ) {
 		if (test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags)) {
 			if (lna == 3 || lna == 2)
 				offset += 10;
diff --git a/drivers/net/wireless/rtl8180.h b/drivers/net/wireless/rtl8180.h
index 2cbfe3c8081f..082a11f93beb 100644
--- a/drivers/net/wireless/rtl8180.h
+++ b/drivers/net/wireless/rtl8180.h
@@ -102,7 +102,7 @@ struct rtl8180_priv {
 	struct rtl8180_tx_ring tx_ring[4];
 	struct ieee80211_channel channels[14];
 	struct ieee80211_rate rates[12];
-	struct ieee80211_hw_mode modes[2];
+	struct ieee80211_supported_band band;
 	struct pci_dev *pdev;
 	u32 rx_conf;
 
diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c
index 5e9a8ace0d81..d0928c91c21c 100644
--- a/drivers/net/wireless/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl8180_dev.c
@@ -49,6 +49,41 @@ static struct pci_device_id rtl8180_table[] __devinitdata = {
 
 MODULE_DEVICE_TABLE(pci, rtl8180_table);
 
+static const struct ieee80211_rate rtl818x_rates[] = {
+	{ .bitrate = 10, .hw_value = 0, },
+	{ .bitrate = 20, .hw_value = 1, },
+	{ .bitrate = 55, .hw_value = 2, },
+	{ .bitrate = 110, .hw_value = 3, },
+	{ .bitrate = 60, .hw_value = 4, },
+	{ .bitrate = 90, .hw_value = 5, },
+	{ .bitrate = 120, .hw_value = 6, },
+	{ .bitrate = 180, .hw_value = 7, },
+	{ .bitrate = 240, .hw_value = 8, },
+	{ .bitrate = 360, .hw_value = 9, },
+	{ .bitrate = 480, .hw_value = 10, },
+	{ .bitrate = 540, .hw_value = 11, },
+};
+
+static const struct ieee80211_channel rtl818x_channels[] = {
+	{ .center_freq = 2412 },
+	{ .center_freq = 2417 },
+	{ .center_freq = 2422 },
+	{ .center_freq = 2427 },
+	{ .center_freq = 2432 },
+	{ .center_freq = 2437 },
+	{ .center_freq = 2442 },
+	{ .center_freq = 2447 },
+	{ .center_freq = 2452 },
+	{ .center_freq = 2457 },
+	{ .center_freq = 2462 },
+	{ .center_freq = 2467 },
+	{ .center_freq = 2472 },
+	{ .center_freq = 2484 },
+};
+
+
+
+
 void rtl8180_write_phy(struct ieee80211_hw *dev, u8 addr, u32 data)
 {
 	struct rtl8180_priv *priv = dev->priv;
@@ -99,10 +134,10 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev)
 			/* TODO: improve signal/rssi reporting */
 			rx_status.signal = flags2 & 0xFF;
 			rx_status.ssi = (flags2 >> 8) & 0x7F;
-			rx_status.rate = (flags >> 20) & 0xF;
-			rx_status.freq = dev->conf.freq;
-			rx_status.channel = dev->conf.channel;
-			rx_status.phymode = dev->conf.phymode;
+			/* XXX: is this correct? */
+			rx_status.rate_idx = (flags >> 20) & 0xF;
+			rx_status.freq = dev->conf.channel->center_freq;
+			rx_status.band = dev->conf.channel->band;
 			rx_status.mactime = le64_to_cpu(entry->tsft);
 			rx_status.flag |= RX_FLAG_TSFT;
 			if (flags & RTL8180_RX_DESC_FLAG_CRC32_ERR)
@@ -223,8 +258,9 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 				 skb->len, PCI_DMA_TODEVICE);
 
 	tx_flags = RTL8180_TX_DESC_FLAG_OWN | RTL8180_TX_DESC_FLAG_FS |
-		   RTL8180_TX_DESC_FLAG_LS | (control->tx_rate << 24) |
-		   (control->rts_cts_rate << 19) | skb->len;
+		   RTL8180_TX_DESC_FLAG_LS |
+		   (control->tx_rate->hw_value << 24) |
+		   (control->rts_cts_rate->hw_value << 19) | skb->len;
 
 	if (priv->r8185)
 		tx_flags |= RTL8180_TX_DESC_FLAG_DMA |
@@ -246,9 +282,9 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 		unsigned int remainder;
 
 		plcp_len = DIV_ROUND_UP(16 * (skb->len + 4),
-					(control->rate->rate * 2) / 10);
+					(control->tx_rate->bitrate * 2) / 10);
 		remainder = (16 * (skb->len + 4)) %
-			    ((control->rate->rate * 2) / 10);
+			    ((control->tx_rate->bitrate * 2) / 10);
 		if (remainder > 0 && remainder <= 6)
 			plcp_len |= 1 << 15;
 	}
@@ -261,8 +297,8 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 	entry->plcp_len = cpu_to_le16(plcp_len);
 	entry->tx_buf = cpu_to_le32(mapping);
 	entry->frame_len = cpu_to_le32(skb->len);
-	entry->flags2 = control->alt_retry_rate != -1 ?
-			control->alt_retry_rate << 4 : 0;
+	entry->flags2 = control->alt_retry_rate != NULL ?
+			control->alt_retry_rate->bitrate << 4 : 0;
 	entry->retry_limit = control->retry_limit;
 	entry->flags = cpu_to_le32(tx_flags);
 	__skb_queue_tail(&ring->queue, skb);
@@ -838,19 +874,19 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev,
 		goto err_free_dev;
 	}
 
+	BUILD_BUG_ON(sizeof(priv->channels) != sizeof(rtl818x_channels));
+	BUILD_BUG_ON(sizeof(priv->rates) != sizeof(rtl818x_rates));
+
 	memcpy(priv->channels, rtl818x_channels, sizeof(rtl818x_channels));
 	memcpy(priv->rates, rtl818x_rates, sizeof(rtl818x_rates));
-	priv->modes[0].mode = MODE_IEEE80211G;
-	priv->modes[0].num_rates = ARRAY_SIZE(rtl818x_rates);
-	priv->modes[0].rates = priv->rates;
-	priv->modes[0].num_channels = ARRAY_SIZE(rtl818x_channels);
-	priv->modes[0].channels = priv->channels;
-	priv->modes[1].mode = MODE_IEEE80211B;
-	priv->modes[1].num_rates = 4;
-	priv->modes[1].rates = priv->rates;
-	priv->modes[1].num_channels = ARRAY_SIZE(rtl818x_channels);
-	priv->modes[1].channels = priv->channels;
-	priv->mode = IEEE80211_IF_TYPE_INVALID;
+
+	priv->band.band = IEEE80211_BAND_2GHZ;
+	priv->band.channels = priv->channels;
+	priv->band.n_channels = ARRAY_SIZE(rtl818x_channels);
+	priv->band.bitrates = priv->rates;
+	priv->band.n_bitrates = 4;
+	dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->band;
+
 	dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
 		     IEEE80211_HW_RX_INCLUDES_FCS;
 	dev->queues = 1;
@@ -879,15 +915,10 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev,
 
 	priv->r8185 = reg & RTL818X_TX_CONF_R8185_ABC;
 	if (priv->r8185) {
-		if ((err = ieee80211_register_hwmode(dev, &priv->modes[0])))
-			goto err_iounmap;
-
+		priv->band.n_bitrates = ARRAY_SIZE(rtl818x_rates);
 		pci_try_set_mwi(pdev);
 	}
 
-	if ((err = ieee80211_register_hwmode(dev, &priv->modes[1])))
-		goto err_iounmap;
-
 	eeprom.data = dev;
 	eeprom.register_read = rtl8180_eeprom_register_read;
 	eeprom.register_write = rtl8180_eeprom_register_write;
@@ -950,8 +981,8 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev,
 	for (i = 0; i < 14; i += 2) {
 		u16 txpwr;
 		eeprom_93cx6_read(&eeprom, 0x10 + (i >> 1), &txpwr);
-		priv->channels[i].val = txpwr & 0xFF;
-		priv->channels[i + 1].val = txpwr >> 8;
+		priv->channels[i].hw_value = txpwr & 0xFF;
+		priv->channels[i + 1].hw_value = txpwr >> 8;
 	}
 
 	/* OFDM TX power */
@@ -959,8 +990,8 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev,
 		for (i = 0; i < 14; i += 2) {
 			u16 txpwr;
 			eeprom_93cx6_read(&eeprom, 0x20 + (i >> 1), &txpwr);
-			priv->channels[i].val |= (txpwr & 0xFF) << 8;
-			priv->channels[i + 1].val |= txpwr & 0xFF00;
+			priv->channels[i].hw_value |= (txpwr & 0xFF) << 8;
+			priv->channels[i + 1].hw_value |= txpwr & 0xFF00;
 		}
 	}
 
diff --git a/drivers/net/wireless/rtl8180_grf5101.c b/drivers/net/wireless/rtl8180_grf5101.c
index 8293e19c4c59..5d47935dbac3 100644
--- a/drivers/net/wireless/rtl8180_grf5101.c
+++ b/drivers/net/wireless/rtl8180_grf5101.c
@@ -73,8 +73,9 @@ static void grf5101_rf_set_channel(struct ieee80211_hw *dev,
 				   struct ieee80211_conf *conf)
 {
 	struct rtl8180_priv *priv = dev->priv;
-	u32 txpw = priv->channels[conf->channel - 1].val & 0xFF;
-	u32 chan = conf->channel - 1;
+	int channel = ieee80211_frequency_to_channel(conf->channel->center_freq);
+	u32 txpw = priv->channels[channel - 1].hw_value & 0xFF;
+	u32 chan = channel - 1;
 
 	/* set TX power */
 	write_grf5101(dev, 0x15, 0x0);
diff --git a/drivers/net/wireless/rtl8180_max2820.c b/drivers/net/wireless/rtl8180_max2820.c
index 98fe9fd64968..a34dfd382b6d 100644
--- a/drivers/net/wireless/rtl8180_max2820.c
+++ b/drivers/net/wireless/rtl8180_max2820.c
@@ -78,8 +78,9 @@ static void max2820_rf_set_channel(struct ieee80211_hw *dev,
 				   struct ieee80211_conf *conf)
 {
 	struct rtl8180_priv *priv = dev->priv;
-	unsigned int chan_idx = conf ? conf->channel - 1 : 0;
-	u32 txpw = priv->channels[chan_idx].val & 0xFF;
+	int channel = ieee80211_frequency_to_channel(conf->channel->center_freq);
+	unsigned int chan_idx = channel - 1;
+	u32 txpw = priv->channels[chan_idx].hw_value & 0xFF;
 	u32 chan = max2820_chan[chan_idx];
 
 	/* While philips SA2400 drive the PA bias from
diff --git a/drivers/net/wireless/rtl8180_rtl8225.c b/drivers/net/wireless/rtl8180_rtl8225.c
index ef3832bee85c..cd22781728a9 100644
--- a/drivers/net/wireless/rtl8180_rtl8225.c
+++ b/drivers/net/wireless/rtl8180_rtl8225.c
@@ -261,8 +261,8 @@ static void rtl8225_rf_set_tx_power(struct ieee80211_hw *dev, int channel)
 	u32 reg;
 	int i;
 
-	cck_power = priv->channels[channel - 1].val & 0xFF;
-	ofdm_power = priv->channels[channel - 1].val >> 8;
+	cck_power = priv->channels[channel - 1].hw_value & 0xFF;
+	ofdm_power = priv->channels[channel - 1].hw_value >> 8;
 
 	cck_power = min(cck_power, (u8)35);
 	ofdm_power = min(ofdm_power, (u8)35);
@@ -476,8 +476,8 @@ static void rtl8225z2_rf_set_tx_power(struct ieee80211_hw *dev, int channel)
 	const u8 *tmp;
 	int i;
 
-	cck_power = priv->channels[channel - 1].val & 0xFF;
-	ofdm_power = priv->channels[channel - 1].val >> 8;
+	cck_power = priv->channels[channel - 1].hw_value & 0xFF;
+	ofdm_power = priv->channels[channel - 1].hw_value >> 8;
 
 	if (channel == 14)
 		tmp = rtl8225z2_tx_power_cck_ch14;
@@ -716,13 +716,14 @@ static void rtl8225_rf_set_channel(struct ieee80211_hw *dev,
 				   struct ieee80211_conf *conf)
 {
 	struct rtl8180_priv *priv = dev->priv;
+	int chan = ieee80211_frequency_to_channel(conf->channel->center_freq);
 
 	if (priv->rf->init == rtl8225_rf_init)
-		rtl8225_rf_set_tx_power(dev, conf->channel);
+		rtl8225_rf_set_tx_power(dev, chan);
 	else
-		rtl8225z2_rf_set_tx_power(dev, conf->channel);
+		rtl8225z2_rf_set_tx_power(dev, chan);
 
-	rtl8225_write(dev, 0x7, rtl8225_chan[conf->channel - 1]);
+	rtl8225_write(dev, 0x7, rtl8225_chan[chan - 1]);
 	msleep(10);
 
 	if (conf->flags & IEEE80211_CONF_SHORT_SLOT_TIME) {
diff --git a/drivers/net/wireless/rtl8180_sa2400.c b/drivers/net/wireless/rtl8180_sa2400.c
index e08ace7b1cb7..0311b4ea124c 100644
--- a/drivers/net/wireless/rtl8180_sa2400.c
+++ b/drivers/net/wireless/rtl8180_sa2400.c
@@ -80,8 +80,9 @@ static void sa2400_rf_set_channel(struct ieee80211_hw *dev,
 				  struct ieee80211_conf *conf)
 {
 	struct rtl8180_priv *priv = dev->priv;
-	u32 txpw = priv->channels[conf->channel - 1].val & 0xFF;
-	u32 chan = sa2400_chan[conf->channel - 1];
+	int channel = ieee80211_frequency_to_channel(conf->channel->center_freq);
+	u32 txpw = priv->channels[channel - 1].hw_value & 0xFF;
+	u32 chan = sa2400_chan[channel - 1];
 
 	write_sa2400(dev, 7, txpw);
 
diff --git a/drivers/net/wireless/rtl8187.h b/drivers/net/wireless/rtl8187.h
index 8680a0b6433c..076d88b6db0e 100644
--- a/drivers/net/wireless/rtl8187.h
+++ b/drivers/net/wireless/rtl8187.h
@@ -71,7 +71,7 @@ struct rtl8187_priv {
 	/* rtl8187 specific */
 	struct ieee80211_channel channels[14];
 	struct ieee80211_rate rates[12];
-	struct ieee80211_hw_mode modes[2];
+	struct ieee80211_supported_band band;
 	struct usb_device *udev;
 	u32 rx_conf;
 	u16 txpwr_base;
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index f44505994a0e..6ef67990692d 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -45,6 +45,38 @@ static struct usb_device_id rtl8187_table[] __devinitdata = {
 
 MODULE_DEVICE_TABLE(usb, rtl8187_table);
 
+static const struct ieee80211_rate rtl818x_rates[] = {
+	{ .bitrate = 10, .hw_value = 0, },
+	{ .bitrate = 20, .hw_value = 1, },
+	{ .bitrate = 55, .hw_value = 2, },
+	{ .bitrate = 110, .hw_value = 3, },
+	{ .bitrate = 60, .hw_value = 4, },
+	{ .bitrate = 90, .hw_value = 5, },
+	{ .bitrate = 120, .hw_value = 6, },
+	{ .bitrate = 180, .hw_value = 7, },
+	{ .bitrate = 240, .hw_value = 8, },
+	{ .bitrate = 360, .hw_value = 9, },
+	{ .bitrate = 480, .hw_value = 10, },
+	{ .bitrate = 540, .hw_value = 11, },
+};
+
+static const struct ieee80211_channel rtl818x_channels[] = {
+	{ .center_freq = 2412 },
+	{ .center_freq = 2417 },
+	{ .center_freq = 2422 },
+	{ .center_freq = 2427 },
+	{ .center_freq = 2432 },
+	{ .center_freq = 2437 },
+	{ .center_freq = 2442 },
+	{ .center_freq = 2447 },
+	{ .center_freq = 2452 },
+	{ .center_freq = 2457 },
+	{ .center_freq = 2462 },
+	{ .center_freq = 2467 },
+	{ .center_freq = 2472 },
+	{ .center_freq = 2484 },
+};
+
 static void rtl8187_iowrite_async_cb(struct urb *urb)
 {
 	kfree(urb->context);
@@ -146,8 +178,8 @@ static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb,
 
 	flags = skb->len;
 	flags |= RTL8187_TX_FLAG_NO_ENCRYPT;
-	flags |= control->rts_cts_rate << 19;
-	flags |= control->tx_rate << 24;
+	flags |= control->rts_cts_rate->hw_value << 19;
+	flags |= control->tx_rate->hw_value << 24;
 	if (ieee80211_get_morefrag((struct ieee80211_hdr *)skb->data))
 		flags |= RTL8187_TX_FLAG_MORE_FRAG;
 	if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) {
@@ -225,10 +257,9 @@ static void rtl8187_rx_cb(struct urb *urb)
 	rx_status.antenna = (hdr->signal >> 7) & 1;
 	rx_status.signal = 64 - min(hdr->noise, (u8)64);
 	rx_status.ssi = signal;
-	rx_status.rate = rate;
-	rx_status.freq = dev->conf.freq;
-	rx_status.channel = dev->conf.channel;
-	rx_status.phymode = dev->conf.phymode;
+	rx_status.rate_idx = rate;
+	rx_status.freq = dev->conf.channel->center_freq;
+	rx_status.band = dev->conf.channel->band;
 	rx_status.mactime = le64_to_cpu(hdr->mac_time);
 	rx_status.flag |= RX_FLAG_TSFT;
 	if (flags & (1 << 13))
@@ -682,19 +713,22 @@ static int __devinit rtl8187_probe(struct usb_interface *intf,
 	usb_get_dev(udev);
 
 	skb_queue_head_init(&priv->rx_queue);
+
+	BUILD_BUG_ON(sizeof(priv->channels) != sizeof(rtl818x_channels));
+	BUILD_BUG_ON(sizeof(priv->rates) != sizeof(rtl818x_rates));
+
 	memcpy(priv->channels, rtl818x_channels, sizeof(rtl818x_channels));
 	memcpy(priv->rates, rtl818x_rates, sizeof(rtl818x_rates));
 	priv->map = (struct rtl818x_csr *)0xFF00;
-	priv->modes[0].mode = MODE_IEEE80211G;
-	priv->modes[0].num_rates = ARRAY_SIZE(rtl818x_rates);
-	priv->modes[0].rates = priv->rates;
-	priv->modes[0].num_channels = ARRAY_SIZE(rtl818x_channels);
-	priv->modes[0].channels = priv->channels;
-	priv->modes[1].mode = MODE_IEEE80211B;
-	priv->modes[1].num_rates = 4;
-	priv->modes[1].rates = priv->rates;
-	priv->modes[1].num_channels = ARRAY_SIZE(rtl818x_channels);
-	priv->modes[1].channels = priv->channels;
+
+	priv->band.band = IEEE80211_BAND_2GHZ;
+	priv->band.channels = priv->channels;
+	priv->band.n_channels = ARRAY_SIZE(rtl818x_channels);
+	priv->band.bitrates = priv->rates;
+	priv->band.n_bitrates = ARRAY_SIZE(rtl818x_rates);
+	dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->band;
+
+
 	priv->mode = IEEE80211_IF_TYPE_MNTR;
 	dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING |
 		     IEEE80211_HW_RX_INCLUDES_FCS;
@@ -703,10 +737,6 @@ static int __devinit rtl8187_probe(struct usb_interface *intf,
 	dev->max_rssi = 65;
 	dev->max_signal = 64;
 
-	for (i = 0; i < 2; i++)
-		if ((err = ieee80211_register_hwmode(dev, &priv->modes[i])))
-			goto err_free_dev;
-
 	eeprom.data = dev;
 	eeprom.register_read = rtl8187_eeprom_register_read;
 	eeprom.register_write = rtl8187_eeprom_register_write;
@@ -730,20 +760,20 @@ static int __devinit rtl8187_probe(struct usb_interface *intf,
 	for (i = 0; i < 3; i++) {
 		eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_CHAN_1 + i,
 				  &txpwr);
-		(*channel++).val = txpwr & 0xFF;
-		(*channel++).val = txpwr >> 8;
+		(*channel++).hw_value = txpwr & 0xFF;
+		(*channel++).hw_value = txpwr >> 8;
 	}
 	for (i = 0; i < 2; i++) {
 		eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_CHAN_4 + i,
 				  &txpwr);
-		(*channel++).val = txpwr & 0xFF;
-		(*channel++).val = txpwr >> 8;
+		(*channel++).hw_value = txpwr & 0xFF;
+		(*channel++).hw_value = txpwr >> 8;
 	}
 	for (i = 0; i < 2; i++) {
 		eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_CHAN_6 + i,
 				  &txpwr);
-		(*channel++).val = txpwr & 0xFF;
-		(*channel++).val = txpwr >> 8;
+		(*channel++).hw_value = txpwr & 0xFF;
+		(*channel++).hw_value = txpwr >> 8;
 	}
 
 	eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_BASE,
diff --git a/drivers/net/wireless/rtl8187_rtl8225.c b/drivers/net/wireless/rtl8187_rtl8225.c
index b713de17ba0a..9146387b4c5e 100644
--- a/drivers/net/wireless/rtl8187_rtl8225.c
+++ b/drivers/net/wireless/rtl8187_rtl8225.c
@@ -283,8 +283,8 @@ static void rtl8225_rf_set_tx_power(struct ieee80211_hw *dev, int channel)
 	u32 reg;
 	int i;
 
-	cck_power = priv->channels[channel - 1].val & 0xF;
-	ofdm_power = priv->channels[channel - 1].val >> 4;
+	cck_power = priv->channels[channel - 1].hw_value & 0xF;
+	ofdm_power = priv->channels[channel - 1].hw_value >> 4;
 
 	cck_power = min(cck_power, (u8)11);
 	ofdm_power = min(ofdm_power, (u8)35);
@@ -500,8 +500,8 @@ static void rtl8225z2_rf_set_tx_power(struct ieee80211_hw *dev, int channel)
 	u32 reg;
 	int i;
 
-	cck_power = priv->channels[channel - 1].val & 0xF;
-	ofdm_power = priv->channels[channel - 1].val >> 4;
+	cck_power = priv->channels[channel - 1].hw_value & 0xF;
+	ofdm_power = priv->channels[channel - 1].hw_value >> 4;
 
 	cck_power = min(cck_power, (u8)15);
 	cck_power += priv->txpwr_base & 0xF;
@@ -735,13 +735,14 @@ static void rtl8225_rf_set_channel(struct ieee80211_hw *dev,
 				   struct ieee80211_conf *conf)
 {
 	struct rtl8187_priv *priv = dev->priv;
+	int chan = ieee80211_frequency_to_channel(conf->channel->center_freq);
 
 	if (priv->rf->init == rtl8225_rf_init)
-		rtl8225_rf_set_tx_power(dev, conf->channel);
+		rtl8225_rf_set_tx_power(dev, chan);
 	else
-		rtl8225z2_rf_set_tx_power(dev, conf->channel);
+		rtl8225z2_rf_set_tx_power(dev, chan);
 
-	rtl8225_write(dev, 0x7, rtl8225_chan[conf->channel - 1]);
+	rtl8225_write(dev, 0x7, rtl8225_chan[chan - 1]);
 	msleep(10);
 }
 
diff --git a/drivers/net/wireless/rtl818x.h b/drivers/net/wireless/rtl818x.h
index 1e7d6f8278d7..4f7d38f506eb 100644
--- a/drivers/net/wireless/rtl818x.h
+++ b/drivers/net/wireless/rtl818x.h
@@ -175,74 +175,4 @@ struct rtl818x_rf_ops {
 	void (*set_chan)(struct ieee80211_hw *, struct ieee80211_conf *);
 };
 
-static const struct ieee80211_rate rtl818x_rates[] = {
-	{ .rate = 10,
-	  .val = 0,
-	  .flags = IEEE80211_RATE_CCK },
-	{ .rate = 20,
-	  .val = 1,
-	  .flags = IEEE80211_RATE_CCK },
-	{ .rate = 55,
-	  .val = 2,
-	  .flags = IEEE80211_RATE_CCK },
-	{ .rate = 110,
-	  .val = 3,
-	  .flags = IEEE80211_RATE_CCK },
-	{ .rate = 60,
-	  .val = 4,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 90,
-	  .val = 5,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 120,
-	  .val = 6,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 180,
-	  .val = 7,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 240,
-	  .val = 8,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 360,
-	  .val = 9,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 480,
-	  .val = 10,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 540,
-	  .val = 11,
-	  .flags = IEEE80211_RATE_OFDM },
-};
-
-static const struct ieee80211_channel rtl818x_channels[] = {
-	{ .chan = 1,
-	  .freq = 2412},
-	{ .chan = 2,
-	  .freq = 2417},
-	{ .chan = 3,
-	  .freq = 2422},
-	{ .chan = 4,
-	  .freq = 2427},
-	{ .chan = 5,
-	  .freq = 2432},
-	{ .chan = 6,
-	  .freq = 2437},
-	{ .chan = 7,
-	  .freq = 2442},
-	{ .chan = 8,
-	  .freq = 2447},
-	{ .chan = 9,
-	  .freq = 2452},
-	{ .chan = 10,
-	  .freq = 2457},
-	{ .chan = 11,
-	  .freq = 2462},
-	{ .chan = 12,
-	  .freq = 2467},
-	{ .chan = 13,
-	  .freq = 2472},
-	{ .chan = 14,
-	  .freq = 2484}
-};
-
 #endif /* RTL818X_H */
diff --git a/drivers/net/wireless/zd1211rw/zd_chip.c b/drivers/net/wireless/zd1211rw/zd_chip.c
index 99e5b03b3f51..e3fba6f09455 100644
--- a/drivers/net/wireless/zd1211rw/zd_chip.c
+++ b/drivers/net/wireless/zd1211rw/zd_chip.c
@@ -986,7 +986,7 @@ static int print_fw_version(struct zd_chip *chip)
 	return 0;
 }
 
-static int set_mandatory_rates(struct zd_chip *chip, int mode)
+static int set_mandatory_rates(struct zd_chip *chip, int gmode)
 {
 	u32 rates;
 	ZD_ASSERT(mutex_is_locked(&chip->mutex));
@@ -994,17 +994,12 @@ static int set_mandatory_rates(struct zd_chip *chip, int mode)
 	 * that the device is supporting. Until further notice we should try
 	 * to support 802.11g also for full speed USB.
 	 */
-	switch (mode) {
-	case MODE_IEEE80211B:
+	if (!gmode)
 		rates = CR_RATE_1M|CR_RATE_2M|CR_RATE_5_5M|CR_RATE_11M;
-		break;
-	case MODE_IEEE80211G:
+	else
 		rates = CR_RATE_1M|CR_RATE_2M|CR_RATE_5_5M|CR_RATE_11M|
 			CR_RATE_6M|CR_RATE_12M|CR_RATE_24M;
-		break;
-	default:
-		return -EINVAL;
-	}
+
 	return zd_iowrite32_locked(chip, rates, CR_MANDATORY_RATE_TBL);
 }
 
@@ -1108,7 +1103,7 @@ int zd_chip_init_hw(struct zd_chip *chip)
 	 * It might be discussed, whether we should suppport pure b mode for
 	 * full speed USB.
 	 */
-	r = set_mandatory_rates(chip, MODE_IEEE80211G);
+	r = set_mandatory_rates(chip, 1);
 	if (r)
 		goto out;
 	/* Disabling interrupts is certainly a smart thing here.
diff --git a/drivers/net/wireless/zd1211rw/zd_ieee80211.c b/drivers/net/wireless/zd1211rw/zd_ieee80211.c
index 7c277ec43f79..d8dc41ec0e5d 100644
--- a/drivers/net/wireless/zd1211rw/zd_ieee80211.c
+++ b/drivers/net/wireless/zd1211rw/zd_ieee80211.c
@@ -65,16 +65,14 @@ static const struct channel_range *zd_channel_range(u8 regdomain)
 
 static void unmask_bg_channels(struct ieee80211_hw *hw,
 	const struct channel_range *range,
-	struct ieee80211_hw_mode *mode)
+	struct ieee80211_supported_band *sband)
 {
 	u8 channel;
 
 	for (channel = range->start; channel < range->end; channel++) {
 		struct ieee80211_channel *chan =
-			&mode->channels[CHAN_TO_IDX(channel)];
-		chan->flag |= IEEE80211_CHAN_W_SCAN |
-			IEEE80211_CHAN_W_ACTIVE_SCAN |
-			IEEE80211_CHAN_W_IBSS;
+			&sband->channels[CHAN_TO_IDX(channel)];
+		chan->flags = 0;
 	}
 }
 
@@ -97,7 +95,6 @@ void zd_geo_init(struct ieee80211_hw *hw, u8 regdomain)
 		range = zd_channel_range(ZD_REGDOMAIN_FCC);
 	}
 
-	unmask_bg_channels(hw, range, &mac->modes[0]);
-	unmask_bg_channels(hw, range, &mac->modes[1]);
+	unmask_bg_channels(hw, range, &mac->band);
 }
 
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 76ef2d83919d..21b653458831 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -34,76 +34,61 @@
 
 /* This table contains the hardware specific values for the modulation rates. */
 static const struct ieee80211_rate zd_rates[] = {
-	{ .rate = 10,
-	  .val = ZD_CCK_RATE_1M,
-	  .flags = IEEE80211_RATE_CCK },
-	{ .rate = 20,
-	  .val = ZD_CCK_RATE_2M,
-	  .val2 = ZD_CCK_RATE_2M | ZD_CCK_PREA_SHORT,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 55,
-	  .val = ZD_CCK_RATE_5_5M,
-	  .val2 = ZD_CCK_RATE_5_5M | ZD_CCK_PREA_SHORT,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 110,
-	  .val = ZD_CCK_RATE_11M,
-	  .val2 = ZD_CCK_RATE_11M | ZD_CCK_PREA_SHORT,
-	  .flags = IEEE80211_RATE_CCK_2 },
-	{ .rate = 60,
-	  .val = ZD_OFDM_RATE_6M,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 90,
-	  .val = ZD_OFDM_RATE_9M,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 120,
-	  .val = ZD_OFDM_RATE_12M,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 180,
-	  .val = ZD_OFDM_RATE_18M,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 240,
-	  .val = ZD_OFDM_RATE_24M,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 360,
-	  .val = ZD_OFDM_RATE_36M,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 480,
-	  .val = ZD_OFDM_RATE_48M,
-	  .flags = IEEE80211_RATE_OFDM },
-	{ .rate = 540,
-	  .val = ZD_OFDM_RATE_54M,
-	  .flags = IEEE80211_RATE_OFDM },
+	{ .bitrate = 10,
+	  .hw_value = ZD_CCK_RATE_1M, },
+	{ .bitrate = 20,
+	  .hw_value = ZD_CCK_RATE_2M,
+	  .hw_value_short = ZD_CCK_RATE_2M | ZD_CCK_PREA_SHORT,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 55,
+	  .hw_value = ZD_CCK_RATE_5_5M,
+	  .hw_value_short = ZD_CCK_RATE_5_5M | ZD_CCK_PREA_SHORT,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 110,
+	  .hw_value = ZD_CCK_RATE_11M,
+	  .hw_value_short = ZD_CCK_RATE_11M | ZD_CCK_PREA_SHORT,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE },
+	{ .bitrate = 60,
+	  .hw_value = ZD_OFDM_RATE_6M,
+	  .flags = 0 },
+	{ .bitrate = 90,
+	  .hw_value = ZD_OFDM_RATE_9M,
+	  .flags = 0 },
+	{ .bitrate = 120,
+	  .hw_value = ZD_OFDM_RATE_12M,
+	  .flags = 0 },
+	{ .bitrate = 180,
+	  .hw_value = ZD_OFDM_RATE_18M,
+	  .flags = 0 },
+	{ .bitrate = 240,
+	  .hw_value = ZD_OFDM_RATE_24M,
+	  .flags = 0 },
+	{ .bitrate = 360,
+	  .hw_value = ZD_OFDM_RATE_36M,
+	  .flags = 0 },
+	{ .bitrate = 480,
+	  .hw_value = ZD_OFDM_RATE_48M,
+	  .flags = 0 },
+	{ .bitrate = 540,
+	  .hw_value = ZD_OFDM_RATE_54M,
+	  .flags = 0 },
 };
 
 static const struct ieee80211_channel zd_channels[] = {
-	{ .chan = 1,
-	  .freq = 2412},
-	{ .chan = 2,
-	  .freq = 2417},
-	{ .chan = 3,
-	  .freq = 2422},
-	{ .chan = 4,
-	  .freq = 2427},
-	{ .chan = 5,
-	  .freq = 2432},
-	{ .chan = 6,
-	  .freq = 2437},
-	{ .chan = 7,
-	  .freq = 2442},
-	{ .chan = 8,
-	  .freq = 2447},
-	{ .chan = 9,
-	  .freq = 2452},
-	{ .chan = 10,
-	  .freq = 2457},
-	{ .chan = 11,
-	  .freq = 2462},
-	{ .chan = 12,
-	  .freq = 2467},
-	{ .chan = 13,
-	  .freq = 2472},
-	{ .chan = 14,
-	  .freq = 2484}
+	{ .center_freq = 2412, .hw_value = 1 },
+	{ .center_freq = 2417, .hw_value = 2 },
+	{ .center_freq = 2422, .hw_value = 3 },
+	{ .center_freq = 2427, .hw_value = 4 },
+	{ .center_freq = 2432, .hw_value = 5 },
+	{ .center_freq = 2437, .hw_value = 6 },
+	{ .center_freq = 2442, .hw_value = 7 },
+	{ .center_freq = 2447, .hw_value = 8 },
+	{ .center_freq = 2452, .hw_value = 9 },
+	{ .center_freq = 2457, .hw_value = 10 },
+	{ .center_freq = 2462, .hw_value = 11 },
+	{ .center_freq = 2467, .hw_value = 12 },
+	{ .center_freq = 2472, .hw_value = 13 },
+	{ .center_freq = 2484, .hw_value = 14 },
 };
 
 static void housekeeping_init(struct zd_mac *mac);
@@ -503,7 +488,9 @@ static int fill_ctrlset(struct zd_mac *mac,
 
 	ZD_ASSERT(frag_len <= 0xffff);
 
-	cs->modulation = control->tx_rate;
+	cs->modulation = control->tx_rate->hw_value;
+	if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE)
+		cs->modulation = control->tx_rate->hw_value_short;
 
 	cs->tx_length = cpu_to_le16(frag_len);
 
@@ -631,6 +618,8 @@ int zd_mac_rx(struct ieee80211_hw *hw, const u8 *buffer, unsigned int length)
 	int bad_frame = 0;
 	u16 fc;
 	bool is_qos, is_4addr, need_padding;
+	int i;
+	u8 rate;
 
 	if (length < ZD_PLCP_HEADER_SIZE + 10 /* IEEE80211_1ADDR_LEN */ +
 	             FCS_LEN + sizeof(struct rx_status))
@@ -660,14 +649,19 @@ int zd_mac_rx(struct ieee80211_hw *hw, const u8 *buffer, unsigned int length)
 		}
 	}
 
-	stats.channel = _zd_chip_get_channel(&mac->chip);
-	stats.freq = zd_channels[stats.channel - 1].freq;
-	stats.phymode = MODE_IEEE80211G;
+	stats.freq = zd_channels[_zd_chip_get_channel(&mac->chip) - 1].center_freq;
+	stats.band = IEEE80211_BAND_2GHZ;
 	stats.ssi = status->signal_strength;
 	stats.signal = zd_rx_qual_percent(buffer,
 		                          length - sizeof(struct rx_status),
 		                          status);
-	stats.rate = zd_rx_rate(buffer, status);
+
+	rate = zd_rx_rate(buffer, status);
+
+	/* todo: return index in the big switches in zd_rx_rate instead */
+	for (i = 0; i < mac->band.n_bitrates; i++)
+		if (rate == mac->band.bitrates[i].hw_value)
+			stats.rate_idx = i;
 
 	length -= ZD_PLCP_HEADER_SIZE + sizeof(struct rx_status);
 	buffer += ZD_PLCP_HEADER_SIZE;
@@ -736,7 +730,7 @@ static void zd_op_remove_interface(struct ieee80211_hw *hw,
 static int zd_op_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf)
 {
 	struct zd_mac *mac = zd_hw_mac(hw);
-	return zd_chip_set_channel(&mac->chip, conf->channel);
+	return zd_chip_set_channel(&mac->chip, conf->channel->hw_value);
 }
 
 static int zd_op_config_interface(struct ieee80211_hw *hw,
@@ -894,7 +888,6 @@ struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf)
 {
 	struct zd_mac *mac;
 	struct ieee80211_hw *hw;
-	int i;
 
 	hw = ieee80211_alloc_hw(sizeof(struct zd_mac), &zd_ops);
 	if (!hw) {
@@ -912,19 +905,14 @@ struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf)
 
 	memcpy(mac->channels, zd_channels, sizeof(zd_channels));
 	memcpy(mac->rates, zd_rates, sizeof(zd_rates));
-	mac->modes[0].mode = MODE_IEEE80211G;
-	mac->modes[0].num_rates = ARRAY_SIZE(zd_rates);
-	mac->modes[0].rates = mac->rates;
-	mac->modes[0].num_channels = ARRAY_SIZE(zd_channels);
-	mac->modes[0].channels = mac->channels;
-	mac->modes[1].mode = MODE_IEEE80211B;
-	mac->modes[1].num_rates = 4;
-	mac->modes[1].rates = mac->rates;
-	mac->modes[1].num_channels = ARRAY_SIZE(zd_channels);
-	mac->modes[1].channels = mac->channels;
-
-	hw->flags = IEEE80211_HW_RX_INCLUDES_FCS |
-		     IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED;
+	mac->band.n_bitrates = ARRAY_SIZE(zd_rates);
+	mac->band.bitrates = mac->rates;
+	mac->band.n_channels = ARRAY_SIZE(zd_channels);
+	mac->band.channels = mac->channels;
+
+	hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &mac->band;
+
+	hw->flags = IEEE80211_HW_RX_INCLUDES_FCS;
 	hw->max_rssi = 100;
 	hw->max_signal = 100;
 
@@ -933,14 +921,6 @@ struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf)
 
 	skb_queue_head_init(&mac->ack_wait_queue);
 
-	for (i = 0; i < 2; i++) {
-		if (ieee80211_register_hwmode(hw, &mac->modes[i])) {
-			dev_dbg_f(&intf->dev, "cannot register hwmode\n");
-			ieee80211_free_hw(hw);
-			return NULL;
-		}
-	}
-
 	zd_chip_init(&mac->chip, hw, intf);
 	housekeeping_init(mac);
 	INIT_WORK(&mac->set_multicast_hash_work, set_multicast_hash_handler);
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h
index 2dde108df767..67dea9739c8f 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.h
+++ b/drivers/net/wireless/zd1211rw/zd_mac.h
@@ -185,7 +185,7 @@ struct zd_mac {
 	struct sk_buff_head ack_wait_queue;
 	struct ieee80211_channel channels[14];
 	struct ieee80211_rate rates[12];
-	struct ieee80211_hw_mode modes[2];
+	struct ieee80211_supported_band band;
 
 	/* Short preamble (used for RTS/CTS) */
 	unsigned int short_preamble:1;
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 277488176a44..460da54a0019 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -69,95 +69,6 @@
  * not do so then mac80211 may add this under certain circumstances.
  */
 
-#define IEEE80211_CHAN_W_SCAN 0x00000001
-#define IEEE80211_CHAN_W_ACTIVE_SCAN 0x00000002
-#define IEEE80211_CHAN_W_IBSS 0x00000004
-
-/* Channel information structure. Low-level driver is expected to fill in chan,
- * freq, and val fields. Other fields will be filled in by 80211.o based on
- * hostapd information and low-level driver does not need to use them. The
- * limits for each channel will be provided in 'struct ieee80211_conf' when
- * configuring the low-level driver with hw->config callback. If a device has
- * a default regulatory domain, IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED
- * can be set to let the driver configure all fields */
-struct ieee80211_channel {
-	short chan; /* channel number (IEEE 802.11) */
-	short freq; /* frequency in MHz */
-	int val; /* hw specific value for the channel */
-	int flag; /* flag for hostapd use (IEEE80211_CHAN_*) */
-	unsigned char power_level;
-	unsigned char antenna_max;
-};
-
-#define IEEE80211_RATE_ERP 0x00000001
-#define IEEE80211_RATE_BASIC 0x00000002
-#define IEEE80211_RATE_PREAMBLE2 0x00000004
-#define IEEE80211_RATE_SUPPORTED 0x00000010
-#define IEEE80211_RATE_OFDM 0x00000020
-#define IEEE80211_RATE_CCK 0x00000040
-#define IEEE80211_RATE_MANDATORY 0x00000100
-
-#define IEEE80211_RATE_CCK_2 (IEEE80211_RATE_CCK | IEEE80211_RATE_PREAMBLE2)
-#define IEEE80211_RATE_MODULATION(f) \
-	(f & (IEEE80211_RATE_CCK | IEEE80211_RATE_OFDM))
-
-/* Low-level driver should set PREAMBLE2, OFDM and CCK flags.
- * BASIC, SUPPORTED, ERP, and MANDATORY flags are set in 80211.o based on the
- * configuration. */
-struct ieee80211_rate {
-	int rate; /* rate in 100 kbps */
-	int val; /* hw specific value for the rate */
-	int flags; /* IEEE80211_RATE_ flags */
-	int val2; /* hw specific value for the rate when using short preamble
-		   * (only when IEEE80211_RATE_PREAMBLE2 flag is set, i.e., for
-		   * 2, 5.5, and 11 Mbps) */
-	signed char min_rssi_ack;
-	unsigned char min_rssi_ack_delta;
-
-	/* following fields are set by 80211.o and need not be filled by the
-	 * low-level driver */
-	int rate_inv; /* inverse of the rate (LCM(all rates) / rate) for
-		       * optimizing channel utilization estimates */
-};
-
-/**
- * enum ieee80211_phymode - PHY modes
- *
- * @MODE_IEEE80211A: 5GHz as defined by 802.11a/802.11h
- * @MODE_IEEE80211B: 2.4 GHz as defined by 802.11b
- * @MODE_IEEE80211G: 2.4 GHz as defined by 802.11g (with OFDM),
- *	backwards compatible with 11b mode
- * @NUM_IEEE80211_MODES: internal
- */
-enum ieee80211_phymode {
-	MODE_IEEE80211A,
-	MODE_IEEE80211B,
-	MODE_IEEE80211G,
-
-	/* keep last */
-	NUM_IEEE80211_MODES
-};
-
-/**
- * struct ieee80211_ht_info - describing STA's HT capabilities
- *
- * This structure describes most essential parameters needed
- * to describe 802.11n HT capabilities for an STA.
- *
- * @ht_supported: is HT supported by STA, 0: no, 1: yes
- * @cap: HT capabilities map as described in 802.11n spec
- * @ampdu_factor: Maximum A-MPDU length factor
- * @ampdu_density: Minimum A-MPDU spacing
- * @supp_mcs_set: Supported MCS set as described in 802.11n spec
- */
-struct ieee80211_ht_info {
-	u8 ht_supported;
-	u16 cap; /* use IEEE80211_HT_CAP_ */
-	u8 ampdu_factor;
-	u8 ampdu_density;
-	u8 supp_mcs_set[16];
-};
-
 /**
  * struct ieee80211_ht_bss_info - describing BSS's HT characteristics
  *
@@ -174,30 +85,6 @@ struct ieee80211_ht_bss_info {
 	u8 bss_op_mode; /* use IEEE80211_HT_IE_ */
 };
 
-/**
- * struct ieee80211_hw_mode - PHY mode definition
- *
- * This structure describes the capabilities supported by the device
- * in a single PHY mode.
- *
- * @list: internal
- * @channels: pointer to array of supported channels
- * @rates: pointer to array of supported bitrates
- * @mode: the PHY mode for this definition
- * @num_channels: number of supported channels
- * @num_rates: number of supported bitrates
- * @ht_info: PHY's 802.11n HT abilities for this mode
- */
-struct ieee80211_hw_mode {
-	struct list_head list;
-	struct ieee80211_channel *channels;
-	struct ieee80211_rate *rates;
-	enum ieee80211_phymode mode;
-	int num_channels;
-	int num_rates;
-	struct ieee80211_ht_info ht_info;
-};
-
 /**
  * struct ieee80211_tx_queue_params - transmit queue configuration
  *
@@ -320,11 +207,13 @@ struct ieee80211_bss_conf {
 
 struct ieee80211_tx_control {
 	struct ieee80211_vif *vif;
-	int tx_rate; /* Transmit rate, given as the hw specific value for the
-		      * rate (from struct ieee80211_rate) */
-	int rts_cts_rate; /* Transmit rate for RTS/CTS frame, given as the hw
-			   * specific value for the rate (from
-			   * struct ieee80211_rate) */
+	struct ieee80211_rate *tx_rate;
+
+	/* Transmit rate for RTS/CTS frame */
+	struct ieee80211_rate *rts_cts_rate;
+
+	/* retry rate for the last retries */
+	struct ieee80211_rate *alt_retry_rate;
 
 #define IEEE80211_TXCTL_REQ_TX_STATUS	(1<<0)/* request TX status callback for
 						* this frame */
@@ -343,6 +232,7 @@ struct ieee80211_tx_control {
 #define IEEE80211_TXCTL_REQUEUE		(1<<7)
 #define IEEE80211_TXCTL_FIRST_FRAGMENT	(1<<8) /* this is a first fragment of
 						* the frame */
+#define IEEE80211_TXCTL_SHORT_PREAMBLE	(1<<9)
 #define IEEE80211_TXCTL_LONG_RETRY_LIMIT (1<<10) /* this frame should be send
 						  * using the through
 						  * set_retry_limit configured
@@ -359,20 +249,11 @@ struct ieee80211_tx_control {
 	u8 retry_limit;		/* 1 = only first attempt, 2 = one retry, ..
 				 * This could be used when set_retry_limit
 				 * is not implemented by the driver */
-	u8 power_level;		/* per-packet transmit power level, in dBm */
 	u8 antenna_sel_tx; 	/* 0 = default/diversity, 1 = Ant0, 2 = Ant1 */
 	u8 icv_len;		/* length of the ICV/MIC field in octets */
 	u8 iv_len;		/* length of the IV field in octets */
 	u8 queue;		/* hardware queue to use for this frame;
 				 * 0 = highest, hw->queues-1 = lowest */
-	struct ieee80211_rate *rate;		/* internal 80211.o rate */
-	struct ieee80211_rate *rts_rate;	/* internal 80211.o rate
-						 * for RTS/CTS */
-	int alt_retry_rate; /* retry rate for the last retries, given as the
-			     * hw specific value for the rate (from
-			     * struct ieee80211_rate). To be used to limit
-			     * packet dropping when probing higher rates, if hw
-			     * supports multiple retry rates. -1 = not used */
 	int type;	/* internal */
 };
 
@@ -415,26 +296,24 @@ enum mac80211_rx_flags {
  * supported by hardware) to the 802.11 code with each received
  * frame.
  * @mactime: MAC timestamp as defined by 802.11
+ * @band: the active band when this frame was received
  * @freq: frequency the radio was tuned to when receiving this frame, in MHz
- * @channel: channel the radio was tuned to
- * @phymode: active PHY mode
  * @ssi: signal strength when receiving this frame
  * @signal: used as 'qual' in statistics reporting
  * @noise: PHY noise when receiving this frame
  * @antenna: antenna used
- * @rate: data rate
+ * @rate_idx: index of data rate into band's supported rates
  * @flag: %RX_FLAG_*
  */
 struct ieee80211_rx_status {
 	u64 mactime;
+	enum ieee80211_band band;
 	int freq;
-	int channel;
-	enum ieee80211_phymode phymode;
 	int ssi;
 	int signal;
 	int noise;
 	int antenna;
-	int rate;
+	int rate_idx;
 	int flag;
 };
 
@@ -509,41 +388,30 @@ enum ieee80211_conf_flags {
  *
  * @radio_enabled: when zero, driver is required to switch off the radio.
  *	TODO make a flag
- * @channel: IEEE 802.11 channel number
- * @freq: frequency in MHz
- * @channel_val: hardware specific channel value for the channel
- * @phymode: PHY mode to activate (REMOVE)
- * @chan: channel to switch to, pointer to the channel information
- * @mode: pointer to mode definition
- * @regulatory_domain: ??
  * @beacon_int: beacon interval (TODO make interface config)
  * @flags: configuration flags defined above
- * @power_level: transmit power limit for current regulatory domain in dBm
- * @antenna_max: maximum antenna gain
+ * @power_level: requested transmit power (in dBm)
+ * @max_antenna_gain: maximum antenna gain (in dBi)
  * @antenna_sel_tx: transmit antenna selection, 0: default/diversity,
  *	1/2: antenna 0/1
  * @antenna_sel_rx: receive antenna selection, like @antenna_sel_tx
  * @ht_conf: describes current self configuration of 802.11n HT capabilies
  * @ht_bss_conf: describes current BSS configuration of 802.11n HT parameters
+ * @channel: the channel to tune to
  */
 struct ieee80211_conf {
-	int channel;			/* IEEE 802.11 channel number */
-	int freq;			/* MHz */
-	int channel_val;		/* hw specific value for the channel */
-
-	enum ieee80211_phymode phymode;
-	struct ieee80211_channel *chan;
-	struct ieee80211_hw_mode *mode;
 	unsigned int regulatory_domain;
 	int radio_enabled;
 
 	int beacon_int;
 	u32 flags;
-	u8 power_level;
-	u8 antenna_max;
+	int power_level;
+	int max_antenna_gain;
 	u8 antenna_sel_tx;
 	u8 antenna_sel_rx;
 
+	struct ieee80211_channel *channel;
+
 	struct ieee80211_ht_info ht_conf;
 	struct ieee80211_ht_bss_info ht_bss_conf;
 };
@@ -764,15 +632,19 @@ enum sta_notify_cmd {
  *	%IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE is also not set because
  *	otherwise the stack will not know when the DTIM beacon was sent.
  *
- * @IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED:
- *	Channels are already configured to the default regulatory domain
- *	specified in the device's EEPROM
+ * @IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE:
+ *	Hardware is not capable of short slot operation on the 2.4 GHz band.
+ *
+ * @IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE:
+ *	Hardware is not capable of receiving frames with short preamble on
+ *	the 2.4 GHz band.
  */
 enum ieee80211_hw_flags {
 	IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE		= 1<<0,
 	IEEE80211_HW_RX_INCLUDES_FCS			= 1<<1,
 	IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING	= 1<<2,
-	IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED	= 1<<3,
+	IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE		= 1<<3,
+	IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE	= 1<<4,
 };
 
 /**
@@ -784,7 +656,8 @@ enum ieee80211_hw_flags {
  * @wiphy: This points to the &struct wiphy allocated for this
  *	802.11 PHY. You must fill in the @perm_addr and @dev
  *	members of this structure using SET_IEEE80211_DEV()
- *	and SET_IEEE80211_PERM_ADDR().
+ *	and SET_IEEE80211_PERM_ADDR(). Additionally, all supported
+ *	bands (with channels, bitrates) are registered here.
  *
  * @conf: &struct ieee80211_conf, device configuration, don't use.
  *
@@ -1062,7 +935,9 @@ enum ieee80211_ampdu_mlme_action {
  *	given local_address is enabled.
  *
  * @hw_scan: Ask the hardware to service the scan request, no need to start
- *	the scan state machine in stack.
+ *	the scan state machine in stack. The scan must honour the channel
+ *	configuration done by the regulatory agent in the wiphy's registered
+ *	bands.
  *
  * @get_stats: return low-level statistics
  *
@@ -1284,10 +1159,6 @@ static inline char *ieee80211_get_radio_led_name(struct ieee80211_hw *hw)
 #endif
 }
 
-/* Register a new hardware PHYMODE capability to the stack. */
-int ieee80211_register_hwmode(struct ieee80211_hw *hw,
-			      struct ieee80211_hw_mode *mode);
-
 /**
  * ieee80211_unregister_hw - Unregister a hardware device
  *
@@ -1461,7 +1332,7 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
  * @hw: pointer obtained from ieee80211_alloc_hw().
  * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf.
  * @frame_len: the length of the frame.
- * @rate: the rate (in 100kbps) at which the frame is going to be transmitted.
+ * @rate: the rate at which the frame is going to be transmitted.
  *
  * Calculate the duration field of some generic frame, given its
  * length and transmission rate (in 100kbps).
@@ -1469,7 +1340,7 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw,
 					struct ieee80211_vif *vif,
 					size_t frame_len,
-					int rate);
+					struct ieee80211_rate *rate);
 
 /**
  * ieee80211_get_buffered_bc - accessing buffered broadcast and multicast frames
diff --git a/include/net/wireless.h b/include/net/wireless.h
index d30c4ba8fd99..c7f805ee5545 100644
--- a/include/net/wireless.h
+++ b/include/net/wireless.h
@@ -12,6 +12,162 @@
 #include <linux/list.h>
 #include <net/cfg80211.h>
 
+/**
+ * enum ieee80211_band - supported frequency bands
+ *
+ * The bands are assigned this way because the supported
+ * bitrates differ in these bands.
+ *
+ * @IEEE80211_BAND_2GHZ: 2.4GHz ISM band
+ * @IEEE80211_BAND_5GHZ: around 5GHz band (4.9-5.7)
+ */
+enum ieee80211_band {
+	IEEE80211_BAND_2GHZ,
+	IEEE80211_BAND_5GHZ,
+
+	/* keep last */
+	IEEE80211_NUM_BANDS
+};
+
+/**
+ * enum ieee80211_channel_flags - channel flags
+ *
+ * Channel flags set by the regulatory control code.
+ *
+ * @IEEE80211_CHAN_DISABLED: This channel is disabled.
+ * @IEEE80211_CHAN_PASSIVE_SCAN: Only passive scanning is permitted
+ *	on this channel.
+ * @IEEE80211_CHAN_NO_IBSS: IBSS is not allowed on this channel.
+ * @IEEE80211_CHAN_RADAR: Radar detection is required on this channel.
+ */
+enum ieee80211_channel_flags {
+	IEEE80211_CHAN_DISABLED		= 1<<0,
+	IEEE80211_CHAN_PASSIVE_SCAN	= 1<<1,
+	IEEE80211_CHAN_NO_IBSS		= 1<<2,
+	IEEE80211_CHAN_RADAR		= 1<<3,
+};
+
+/**
+ * struct ieee80211_channel - channel definition
+ *
+ * This structure describes a single channel for use
+ * with cfg80211.
+ *
+ * @center_freq: center frequency in MHz
+ * @hw_value: hardware-specific value for the channel
+ * @flags: channel flags from &enum ieee80211_channel_flags.
+ * @orig_flags: channel flags at registration time, used by regulatory
+ *	code to support devices with additional restrictions
+ * @band: band this channel belongs to.
+ * @max_antenna_gain: maximum antenna gain in dBi
+ * @max_power: maximum transmission power (in dBm)
+ * @orig_mag: internal use
+ * @orig_mpwr: internal use
+ */
+struct ieee80211_channel {
+	enum ieee80211_band band;
+	u16 center_freq;
+	u16 hw_value;
+	u32 flags;
+	int max_antenna_gain;
+	int max_power;
+	u32 orig_flags;
+	int orig_mag, orig_mpwr;
+};
+
+/**
+ * enum ieee80211_rate_flags - rate flags
+ *
+ * Hardware/specification flags for rates. These are structured
+ * in a way that allows using the same bitrate structure for
+ * different bands/PHY modes.
+ *
+ * @IEEE80211_RATE_SHORT_PREAMBLE: Hardware can send with short
+ *	preamble on this bitrate; only relevant in 2.4GHz band and
+ *	with CCK rates.
+ * @IEEE80211_RATE_MANDATORY_A: This bitrate is a mandatory rate
+ *	when used with 802.11a (on the 5 GHz band); filled by the
+ *	core code when registering the wiphy.
+ * @IEEE80211_RATE_MANDATORY_B: This bitrate is a mandatory rate
+ *	when used with 802.11b (on the 2.4 GHz band); filled by the
+ *	core code when registering the wiphy.
+ * @IEEE80211_RATE_MANDATORY_G: This bitrate is a mandatory rate
+ *	when used with 802.11g (on the 2.4 GHz band); filled by the
+ *	core code when registering the wiphy.
+ * @IEEE80211_RATE_ERP_G: This is an ERP rate in 802.11g mode.
+ */
+enum ieee80211_rate_flags {
+	IEEE80211_RATE_SHORT_PREAMBLE	= 1<<0,
+	IEEE80211_RATE_MANDATORY_A	= 1<<1,
+	IEEE80211_RATE_MANDATORY_B	= 1<<2,
+	IEEE80211_RATE_MANDATORY_G	= 1<<3,
+	IEEE80211_RATE_ERP_G		= 1<<4,
+};
+
+/**
+ * struct ieee80211_rate - bitrate definition
+ *
+ * This structure describes a bitrate that an 802.11 PHY can
+ * operate with. The two values @hw_value and @hw_value_short
+ * are only for driver use when pointers to this structure are
+ * passed around.
+ *
+ * @flags: rate-specific flags
+ * @bitrate: bitrate in units of 100 Kbps
+ * @hw_value: driver/hardware value for this rate
+ * @hw_value_short: driver/hardware value for this rate when
+ *	short preamble is used
+ */
+struct ieee80211_rate {
+	u32 flags;
+	u16 bitrate;
+	u16 hw_value, hw_value_short;
+};
+
+/**
+ * struct ieee80211_ht_info - describing STA's HT capabilities
+ *
+ * This structure describes most essential parameters needed
+ * to describe 802.11n HT capabilities for an STA.
+ *
+ * @ht_supported: is HT supported by STA, 0: no, 1: yes
+ * @cap: HT capabilities map as described in 802.11n spec
+ * @ampdu_factor: Maximum A-MPDU length factor
+ * @ampdu_density: Minimum A-MPDU spacing
+ * @supp_mcs_set: Supported MCS set as described in 802.11n spec
+ */
+struct ieee80211_ht_info {
+	u16 cap; /* use IEEE80211_HT_CAP_ */
+	u8 ht_supported;
+	u8 ampdu_factor;
+	u8 ampdu_density;
+	u8 supp_mcs_set[16];
+};
+
+/**
+ * struct ieee80211_supported_band - frequency band definition
+ *
+ * This structure describes a frequency band a wiphy
+ * is able to operate in.
+ *
+ * @channels: Array of channels the hardware can operate in
+ *	in this band.
+ * @band: the band this structure represents
+ * @n_channels: Number of channels in @channels
+ * @bitrates: Array of bitrates the hardware can operate with
+ *	in this band. Must be sorted to give a valid "supported
+ *	rates" IE, i.e. CCK rates first, then OFDM.
+ * @n_bitrates: Number of bitrates in @bitrates
+ */
+struct ieee80211_supported_band {
+	struct ieee80211_channel *channels;
+	struct ieee80211_rate *bitrates;
+	enum ieee80211_band band;
+	int n_channels;
+	int n_bitrates;
+	struct ieee80211_ht_info ht_info;
+};
+
 /**
  * struct wiphy - wireless hardware description
  * @idx: the wiphy index assigned to this item
@@ -30,6 +186,8 @@ struct wiphy {
 	 * help determine whether you own this wiphy or not. */
 	void *privid;
 
+	struct ieee80211_supported_band *bands[IEEE80211_NUM_BANDS];
+
 	/* fields below are read-only, assigned by cfg80211 */
 
 	/* the item in /sys/class/ieee80211/ points to this,
@@ -136,4 +294,14 @@ extern void wiphy_unregister(struct wiphy *wiphy);
  */
 extern void wiphy_free(struct wiphy *wiphy);
 
+/**
+ * ieee80211_channel_to_frequency - convert channel number to frequency
+ */
+extern int ieee80211_channel_to_frequency(int chan);
+
+/**
+ * ieee80211_frequency_to_channel - convert frequency to channel number
+ */
+extern int ieee80211_frequency_to_channel(int freq);
+
 #endif /* __NET_WIRELESS_H */
diff --git a/net/mac80211/Makefile b/net/mac80211/Makefile
index 54f46bc80cfe..9d7a19581a29 100644
--- a/net/mac80211/Makefile
+++ b/net/mac80211/Makefile
@@ -19,7 +19,6 @@ mac80211-y := \
 	ieee80211_iface.o \
 	ieee80211_rate.o \
 	michael.o \
-	regdomain.o \
 	tkip.o \
 	aes_ccm.o \
 	cfg.o \
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 22c9619ba776..15b8cf94f510 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -498,7 +498,7 @@ static void sta_apply_parameters(struct ieee80211_local *local,
 {
 	u32 rates;
 	int i, j;
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 
 	if (params->station_flags & STATION_FLAG_CHANGED) {
 		sta->flags &= ~WLAN_STA_AUTHORIZED;
@@ -525,15 +525,16 @@ static void sta_apply_parameters(struct ieee80211_local *local,
 
 	if (params->supported_rates) {
 		rates = 0;
-		mode = local->oper_hw_mode;
+		sband = local->hw.wiphy->bands[local->oper_channel->band];
+
 		for (i = 0; i < params->supported_rates_len; i++) {
 			int rate = (params->supported_rates[i] & 0x7f) * 5;
-			for (j = 0; j < mode->num_rates; j++) {
-				if (mode->rates[j].rate == rate)
+			for (j = 0; j < sband->n_bitrates; j++) {
+				if (sband->bitrates[j].bitrate == rate)
 					rates |= BIT(j);
 			}
 		}
-		sta->supp_rates = rates;
+		sta->supp_rates[local->oper_channel->band] = rates;
 	}
 }
 
diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c
index 60514b2c97b9..4736c64937b4 100644
--- a/net/mac80211/debugfs.c
+++ b/net/mac80211/debugfs.c
@@ -19,41 +19,6 @@ int mac80211_open_file_generic(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static const char *ieee80211_mode_str(int mode)
-{
-	switch (mode) {
-	case MODE_IEEE80211A:
-		return "IEEE 802.11a";
-	case MODE_IEEE80211B:
-		return "IEEE 802.11b";
-	case MODE_IEEE80211G:
-		return "IEEE 802.11g";
-	default:
-		return "UNKNOWN";
-	}
-}
-
-static ssize_t modes_read(struct file *file, char __user *userbuf,
-			  size_t count, loff_t *ppos)
-{
-	struct ieee80211_local *local = file->private_data;
-	struct ieee80211_hw_mode *mode;
-	char buf[150], *p = buf;
-
-	/* FIXME: locking! */
-	list_for_each_entry(mode, &local->modes_list, list) {
-		p += scnprintf(p, sizeof(buf)+buf-p,
-			       "%s\n", ieee80211_mode_str(mode->mode));
-	}
-
-	return simple_read_from_buffer(userbuf, count, ppos, buf, p-buf);
-}
-
-static const struct file_operations modes_ops = {
-	.read = modes_read,
-	.open = mac80211_open_file_generic,
-};
-
 #define DEBUGFS_READONLY_FILE(name, buflen, fmt, value...)		\
 static ssize_t name## _read(struct file *file, char __user *userbuf,	\
 			    size_t count, loff_t *ppos)			\
@@ -80,10 +45,8 @@ static const struct file_operations name## _ops = {			\
 	local->debugfs.name = NULL;
 
 
-DEBUGFS_READONLY_FILE(channel, 20, "%d",
-		      local->hw.conf.channel);
 DEBUGFS_READONLY_FILE(frequency, 20, "%d",
-		      local->hw.conf.freq);
+		      local->hw.conf.channel->center_freq);
 DEBUGFS_READONLY_FILE(antenna_sel_tx, 20, "%d",
 		      local->hw.conf.antenna_sel_tx);
 DEBUGFS_READONLY_FILE(antenna_sel_rx, 20, "%d",
@@ -100,8 +63,6 @@ DEBUGFS_READONLY_FILE(long_retry_limit, 20, "%d",
 		      local->long_retry_limit);
 DEBUGFS_READONLY_FILE(total_ps_buffered, 20, "%d",
 		      local->total_ps_buffered);
-DEBUGFS_READONLY_FILE(mode, 20, "%s",
-		      ieee80211_mode_str(local->hw.conf.phymode));
 DEBUGFS_READONLY_FILE(wep_iv, 20, "%#06x",
 		      local->wep_iv & 0xffffff);
 DEBUGFS_READONLY_FILE(rate_ctrl_alg, 100, "%s",
@@ -294,7 +255,6 @@ void debugfs_hw_add(struct ieee80211_local *local)
 	local->debugfs.stations = debugfs_create_dir("stations", phyd);
 	local->debugfs.keys = debugfs_create_dir("keys", phyd);
 
-	DEBUGFS_ADD(channel);
 	DEBUGFS_ADD(frequency);
 	DEBUGFS_ADD(antenna_sel_tx);
 	DEBUGFS_ADD(antenna_sel_rx);
@@ -304,9 +264,7 @@ void debugfs_hw_add(struct ieee80211_local *local)
 	DEBUGFS_ADD(short_retry_limit);
 	DEBUGFS_ADD(long_retry_limit);
 	DEBUGFS_ADD(total_ps_buffered);
-	DEBUGFS_ADD(mode);
 	DEBUGFS_ADD(wep_iv);
-	DEBUGFS_ADD(modes);
 
 	statsd = debugfs_create_dir("statistics", phyd);
 	local->debugfs.statistics = statsd;
@@ -356,7 +314,6 @@ void debugfs_hw_add(struct ieee80211_local *local)
 
 void debugfs_hw_del(struct ieee80211_local *local)
 {
-	DEBUGFS_DEL(channel);
 	DEBUGFS_DEL(frequency);
 	DEBUGFS_DEL(antenna_sel_tx);
 	DEBUGFS_DEL(antenna_sel_rx);
@@ -366,9 +323,7 @@ void debugfs_hw_del(struct ieee80211_local *local)
 	DEBUGFS_DEL(short_retry_limit);
 	DEBUGFS_DEL(long_retry_limit);
 	DEBUGFS_DEL(total_ps_buffered);
-	DEBUGFS_DEL(mode);
 	DEBUGFS_DEL(wep_iv);
-	DEBUGFS_DEL(modes);
 
 	DEBUGFS_STATS_DEL(transmitted_fragment_count);
 	DEBUGFS_STATS_DEL(multicast_transmitted_frame_count);
diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c
index df25abf63137..49660f4e845d 100644
--- a/net/mac80211/debugfs_sta.c
+++ b/net/mac80211/debugfs_sta.c
@@ -33,22 +33,6 @@ static ssize_t sta_ ##name## _read(struct file *file,			\
 #define STA_READ_LU(name, field) STA_READ(name, 20, field, "%lu\n")
 #define STA_READ_S(name, field) STA_READ(name, 20, field, "%s\n")
 
-#define STA_READ_RATE(name, field)					\
-static ssize_t sta_##name##_read(struct file *file,			\
-				 char __user *userbuf,			\
-				 size_t count, loff_t *ppos)		\
-{									\
-	struct sta_info *sta = file->private_data;			\
-	struct ieee80211_local *local = wdev_priv(sta->dev->ieee80211_ptr);\
-	struct ieee80211_hw_mode *mode = local->oper_hw_mode;		\
-	char buf[20];							\
-	int res = scnprintf(buf, sizeof(buf), "%d\n",			\
-			    (sta->field >= 0 &&				\
-			    sta->field < mode->num_rates) ?		\
-			    mode->rates[sta->field].rate : -1);		\
-	return simple_read_from_buffer(userbuf, count, ppos, buf, res);	\
-}
-
 #define STA_OPS(name)							\
 static const struct file_operations sta_ ##name## _ops = {		\
 	.read = sta_##name##_read,					\
@@ -77,8 +61,6 @@ STA_FILE(rx_fragments, rx_fragments, LU);
 STA_FILE(rx_dropped, rx_dropped, LU);
 STA_FILE(tx_fragments, tx_fragments, LU);
 STA_FILE(tx_filtered, tx_filtered_count, LU);
-STA_FILE(txrate, txrate, RATE);
-STA_FILE(last_txrate, last_txrate, RATE);
 STA_FILE(tx_retry_failed, tx_retry_failed, LU);
 STA_FILE(tx_retry_count, tx_retry_count, LU);
 STA_FILE(last_rssi, last_rssi, D);
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 3961d4c4320c..de894b61a23c 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -876,37 +876,28 @@ int ieee80211_if_config_beacon(struct net_device *dev)
 
 int ieee80211_hw_config(struct ieee80211_local *local)
 {
-	struct ieee80211_hw_mode *mode;
 	struct ieee80211_channel *chan;
 	int ret = 0;
 
-	if (local->sta_sw_scanning) {
+	if (local->sta_sw_scanning)
 		chan = local->scan_channel;
-		mode = local->scan_hw_mode;
-	} else {
+	else
 		chan = local->oper_channel;
-		mode = local->oper_hw_mode;
-	}
 
-	local->hw.conf.channel = chan->chan;
-	local->hw.conf.channel_val = chan->val;
-	if (!local->hw.conf.power_level) {
-		local->hw.conf.power_level = chan->power_level;
-	} else {
-		local->hw.conf.power_level = min(chan->power_level,
-						 local->hw.conf.power_level);
-	}
-	local->hw.conf.freq = chan->freq;
-	local->hw.conf.phymode = mode->mode;
-	local->hw.conf.antenna_max = chan->antenna_max;
-	local->hw.conf.chan = chan;
-	local->hw.conf.mode = mode;
+	local->hw.conf.channel = chan;
+
+	if (!local->hw.conf.power_level)
+		local->hw.conf.power_level = chan->max_power;
+	else
+		local->hw.conf.power_level = min(chan->max_power,
+					       local->hw.conf.power_level);
+
+	local->hw.conf.max_antenna_gain = chan->max_antenna_gain;
 
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
-	printk(KERN_DEBUG "HW CONFIG: channel=%d freq=%d "
-	       "phymode=%d\n", local->hw.conf.channel, local->hw.conf.freq,
-	       local->hw.conf.phymode);
-#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
+	printk(KERN_DEBUG "%s: HW CONFIG: freq=%d\n",
+	       wiphy_name(local->hw.wiphy), chan->center_freq);
+#endif
 
 	if (local->open_count)
 		ret = local->ops->config(local_to_hw(local), &local->hw.conf);
@@ -924,11 +915,13 @@ int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
 			   struct ieee80211_ht_bss_info *req_bss_cap)
 {
 	struct ieee80211_conf *conf = &local->hw.conf;
-	struct ieee80211_hw_mode *mode = conf->mode;
+	struct ieee80211_supported_band *sband;
 	int i;
 
+	sband = local->hw.wiphy->bands[conf->channel->band];
+
 	/* HT is not supported */
-	if (!mode->ht_info.ht_supported) {
+	if (!sband->ht_info.ht_supported) {
 		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
 		return -EOPNOTSUPP;
 	}
@@ -938,17 +931,17 @@ int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
 		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
 	} else {
 		conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE;
-		conf->ht_conf.cap = req_ht_cap->cap & mode->ht_info.cap;
+		conf->ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap;
 		conf->ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
 		conf->ht_conf.cap |=
-			mode->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
+			sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
 		conf->ht_bss_conf.primary_channel =
 			req_bss_cap->primary_channel;
 		conf->ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
 		conf->ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
 		for (i = 0; i < SUPP_MCS_SET_LEN; i++)
 			conf->ht_conf.supp_mcs_set[i] =
-				mode->ht_info.supp_mcs_set[i] &
+				sband->ht_info.supp_mcs_set[i] &
 				  req_ht_cap->supp_mcs_set[i];
 
 		/* In STA mode, this gives us indication
@@ -1418,10 +1411,6 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len,
 	local->long_retry_limit = 4;
 	local->hw.conf.radio_enabled = 1;
 
-	local->enabled_modes = ~0;
-
-	INIT_LIST_HEAD(&local->modes_list);
-
 	INIT_LIST_HEAD(&local->interfaces);
 
 	INIT_DELAYED_WORK(&local->scan_work, ieee80211_sta_scan_work);
@@ -1466,6 +1455,25 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	struct ieee80211_local *local = hw_to_local(hw);
 	const char *name;
 	int result;
+	enum ieee80211_band band;
+
+	/*
+	 * generic code guarantees at least one band,
+	 * set this very early because much code assumes
+	 * that hw.conf.channel is assigned
+	 */
+	for (band = 0; band < IEEE80211_NUM_BANDS; band++) {
+		struct ieee80211_supported_band *sband;
+
+		sband = local->hw.wiphy->bands[band];
+		if (sband) {
+			/* init channel we're on */
+			local->hw.conf.channel =
+			local->oper_channel =
+			local->scan_channel = &sband->channels[0];
+			break;
+		}
+	}
 
 	result = wiphy_register(local->hw.wiphy);
 	if (result < 0)
@@ -1567,44 +1575,10 @@ fail_workqueue:
 }
 EXPORT_SYMBOL(ieee80211_register_hw);
 
-int ieee80211_register_hwmode(struct ieee80211_hw *hw,
-			      struct ieee80211_hw_mode *mode)
-{
-	struct ieee80211_local *local = hw_to_local(hw);
-	struct ieee80211_rate *rate;
-	int i;
-
-	INIT_LIST_HEAD(&mode->list);
-	list_add_tail(&mode->list, &local->modes_list);
-
-	local->hw_modes |= (1 << mode->mode);
-	for (i = 0; i < mode->num_rates; i++) {
-		rate = &(mode->rates[i]);
-		rate->rate_inv = CHAN_UTIL_RATE_LCM / rate->rate;
-	}
-	ieee80211_prepare_rates(local, mode);
-
-	if (!local->oper_hw_mode) {
-		/* Default to this mode */
-		local->hw.conf.phymode = mode->mode;
-		local->oper_hw_mode = local->scan_hw_mode = mode;
-		local->oper_channel = local->scan_channel = &mode->channels[0];
-		local->hw.conf.mode = local->oper_hw_mode;
-		local->hw.conf.chan = local->oper_channel;
-	}
-
-	if (!(hw->flags & IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED))
-		ieee80211_set_default_regdomain(mode);
-
-	return 0;
-}
-EXPORT_SYMBOL(ieee80211_register_hwmode);
-
 void ieee80211_unregister_hw(struct ieee80211_hw *hw)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_sub_if_data *sdata, *tmp;
-	int i;
 
 	tasklet_kill(&local->tx_pending_tasklet);
 	tasklet_kill(&local->tasklet);
@@ -1645,11 +1619,6 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw)
 	rate_control_deinitialize(local);
 	debugfs_hw_del(local);
 
-	for (i = 0; i < NUM_IEEE80211_MODES; i++) {
-		kfree(local->supp_rates[i]);
-		kfree(local->basic_rates[i]);
-	}
-
 	if (skb_queue_len(&local->skb_queue)
 			|| skb_queue_len(&local->skb_queue_unreliable))
 		printk(KERN_WARNING "%s: skb_queue not empty\n",
@@ -1696,7 +1665,6 @@ static int __init ieee80211_init(void)
 	}
 
 	ieee80211_debugfs_netdev_init();
-	ieee80211_regdomain_init();
 
 	return 0;
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 9d09ba8cc02b..54eea5f24474 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -79,8 +79,7 @@ struct ieee80211_sta_bss {
 	u8 ssid[IEEE80211_MAX_SSID_LEN];
 	size_t ssid_len;
 	u16 capability; /* host byte order */
-	int hw_mode;
-	int channel;
+	enum ieee80211_band band;
 	int freq;
 	int rssi, signal, noise;
 	u8 *wpa_ie;
@@ -136,13 +135,12 @@ struct ieee80211_txrx_data {
 	union {
 		struct {
 			struct ieee80211_tx_control *control;
-			struct ieee80211_hw_mode *mode;
+			struct ieee80211_channel *channel;
 			struct ieee80211_rate *rate;
 			/* use this rate (if set) for last fragment; rate can
 			 * be set to lower rate for the first fragments, e.g.,
 			 * when using CTS protection with IEEE 802.11g. */
 			struct ieee80211_rate *last_frag_rate;
-			int last_frag_hwrate;
 
 			/* Extra fragments (in addition to the first fragment
 			 * in skb) */
@@ -151,6 +149,7 @@ struct ieee80211_txrx_data {
 		} tx;
 		struct {
 			struct ieee80211_rx_status *status;
+			struct ieee80211_rate *rate;
 			int sent_ps_buffered;
 			int queue;
 			int load;
@@ -179,8 +178,6 @@ struct ieee80211_tx_stored_packet {
 	struct sk_buff *skb;
 	int num_extra_frag;
 	struct sk_buff **extra_frag;
-	int last_frag_rateidx;
-	int last_frag_hwrate;
 	struct ieee80211_rate *last_frag_rate;
 	unsigned int last_frag_rate_ctrl_probe;
 };
@@ -283,7 +280,7 @@ struct ieee80211_if_sta {
 
 	unsigned long ibss_join_req;
 	struct sk_buff *probe_resp; /* ProbeResp template for IBSS */
-	u32 supp_rates_bits;
+	u32 supp_rates_bits[IEEE80211_NUM_BANDS];
 
 	int wmm_last_param_set;
 };
@@ -293,6 +290,7 @@ struct ieee80211_if_sta {
 #define IEEE80211_SDATA_ALLMULTI	BIT(0)
 #define IEEE80211_SDATA_PROMISC		BIT(1)
 #define IEEE80211_SDATA_USERSPACE_MLME	BIT(2)
+#define IEEE80211_SDATA_OPERATING_GMODE	BIT(3)
 struct ieee80211_sub_if_data {
 	struct list_head list;
 
@@ -313,6 +311,11 @@ struct ieee80211_sub_if_data {
 	 */
 	int ieee802_1x_pac;
 
+	/*
+	 * basic rates of this AP or the AP we're associated to
+	 */
+	u64 basic_rates;
+
 	u16 sequence;
 
 	/* Fragment table for host-based reassembly */
@@ -420,9 +423,6 @@ struct ieee80211_local {
 
 	const struct ieee80211_ops *ops;
 
-	/* List of registered struct ieee80211_hw_mode */
-	struct list_head modes_list;
-
 	struct net_device *mdev; /* wmaster# - "master" 802.11 device */
 	int open_count;
 	int monitors;
@@ -462,11 +462,6 @@ struct ieee80211_local {
 
 	struct rate_control_ref *rate_ctrl;
 
-	/* Supported and basic rate filters for different modes. These are
-	 * pointers to -1 terminated lists and rates in 100 kbps units. */
-	int *supp_rates[NUM_IEEE80211_MODES];
-	int *basic_rates[NUM_IEEE80211_MODES];
-
 	int rts_threshold;
 	int fragmentation_threshold;
 	int short_retry_limit; /* dot11ShortRetryLimit */
@@ -488,12 +483,13 @@ struct ieee80211_local {
 	bool sta_sw_scanning;
 	bool sta_hw_scanning;
 	int scan_channel_idx;
+	enum ieee80211_band scan_band;
+
 	enum { SCAN_SET_CHANNEL, SCAN_SEND_PROBE } scan_state;
 	unsigned long last_scan_completed;
 	struct delayed_work scan_work;
 	struct net_device *scan_dev;
 	struct ieee80211_channel *oper_channel, *scan_channel;
-	struct ieee80211_hw_mode *oper_hw_mode, *scan_hw_mode;
 	u8 scan_ssid[IEEE80211_MAX_SSID_LEN];
 	size_t scan_ssid_len;
 	struct list_head sta_bss_list;
@@ -562,14 +558,8 @@ struct ieee80211_local {
 	int wifi_wme_noack_test;
 	unsigned int wmm_acm; /* bit field of ACM bits (BIT(802.1D tag)) */
 
-	unsigned int enabled_modes; /* bitfield of allowed modes;
-				      * (1 << MODE_*) */
-	unsigned int hw_modes; /* bitfield of supported hardware modes;
-				* (1 << MODE_*) */
-
 #ifdef CONFIG_MAC80211_DEBUGFS
 	struct local_debugfsdentries {
-		struct dentry *channel;
 		struct dentry *frequency;
 		struct dentry *antenna_sel_tx;
 		struct dentry *antenna_sel_rx;
@@ -579,9 +569,7 @@ struct ieee80211_local {
 		struct dentry *short_retry_limit;
 		struct dentry *long_retry_limit;
 		struct dentry *total_ps_buffered;
-		struct dentry *mode;
 		struct dentry *wep_iv;
-		struct dentry *modes;
 		struct dentry *statistics;
 		struct local_debugfsdentries_statsdentries {
 			struct dentry *transmitted_fragment_count;
@@ -692,23 +680,6 @@ static inline void bss_tim_clear(struct ieee80211_local *local,
 	read_unlock_bh(&local->sta_lock);
 }
 
-/**
- * ieee80211_is_erp_rate - Check if a rate is an ERP rate
- * @phymode: The PHY-mode for this rate (MODE_IEEE80211...)
- * @rate: Transmission rate to check, in 100 kbps
- *
- * Check if a given rate is an Extended Rate PHY (ERP) rate.
- */
-static inline int ieee80211_is_erp_rate(int phymode, int rate)
-{
-	if (phymode == MODE_IEEE80211G) {
-		if (rate != 10 && rate != 20 &&
-		    rate != 55 && rate != 110)
-			return 1;
-	}
-	return 0;
-}
-
 static inline int ieee80211_bssid_match(const u8 *raddr, const u8 *addr)
 {
 	return compare_ether_addr(raddr, addr) == 0 ||
@@ -720,13 +691,9 @@ static inline int ieee80211_bssid_match(const u8 *raddr, const u8 *addr)
 int ieee80211_hw_config(struct ieee80211_local *local);
 int ieee80211_if_config(struct net_device *dev);
 int ieee80211_if_config_beacon(struct net_device *dev);
-void ieee80211_prepare_rates(struct ieee80211_local *local,
-			     struct ieee80211_hw_mode *mode);
 void ieee80211_tx_set_iswep(struct ieee80211_txrx_data *tx);
 int ieee80211_if_update_wds(struct net_device *dev, u8 *remote_addr);
 void ieee80211_if_setup(struct net_device *dev);
-struct ieee80211_rate *ieee80211_get_rate(struct ieee80211_local *local,
-					  int phymode, int hwrate);
 int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
 			   struct ieee80211_ht_info *req_ht_cap,
 			   struct ieee80211_ht_bss_info *req_bss_cap);
@@ -757,7 +724,7 @@ extern const struct iw_handler_def ieee80211_iw_handler_def;
 /* ieee80211_ioctl.c */
 int ieee80211_set_compression(struct ieee80211_local *local,
 			      struct net_device *dev, struct sta_info *sta);
-int ieee80211_set_channel(struct ieee80211_local *local, int channel, int freq);
+int ieee80211_set_freq(struct ieee80211_local *local, int freq);
 /* ieee80211_sta.c */
 void ieee80211_sta_timer(unsigned long data);
 void ieee80211_sta_work(struct work_struct *work);
@@ -810,10 +777,6 @@ int ieee80211_if_remove(struct net_device *dev, const char *name, int id);
 void ieee80211_if_free(struct net_device *dev);
 void ieee80211_if_sdata_init(struct ieee80211_sub_if_data *sdata);
 
-/* regdomain.c */
-void ieee80211_regdomain_init(void);
-void ieee80211_set_default_regdomain(struct ieee80211_hw_mode *mode);
-
 /* rx handling */
 extern ieee80211_rx_handler ieee80211_rx_handlers[];
 
diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c
index 92f1eb2da311..27cee580f9f1 100644
--- a/net/mac80211/ieee80211_iface.c
+++ b/net/mac80211/ieee80211_iface.c
@@ -118,6 +118,8 @@ void ieee80211_if_set_type(struct net_device *dev, int type)
 	sdata->bss = NULL;
 	sdata->vif.type = type;
 
+	sdata->basic_rates = 0;
+
 	switch (type) {
 	case IEEE80211_IF_TYPE_WDS:
 		/* nothing special */
diff --git a/net/mac80211/ieee80211_ioctl.c b/net/mac80211/ieee80211_ioctl.c
index 5024d3733834..54ad07aafe2d 100644
--- a/net/mac80211/ieee80211_ioctl.c
+++ b/net/mac80211/ieee80211_ioctl.c
@@ -129,22 +129,7 @@ static int ieee80211_ioctl_giwname(struct net_device *dev,
 				   struct iw_request_info *info,
 				   char *name, char *extra)
 {
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-
-	switch (local->hw.conf.phymode) {
-	case MODE_IEEE80211A:
-		strcpy(name, "IEEE 802.11a");
-		break;
-	case MODE_IEEE80211B:
-		strcpy(name, "IEEE 802.11b");
-		break;
-	case MODE_IEEE80211G:
-		strcpy(name, "IEEE 802.11g");
-		break;
-	default:
-		strcpy(name, "IEEE 802.11");
-		break;
-	}
+	strcpy(name, "IEEE 802.11");
 
 	return 0;
 }
@@ -156,7 +141,7 @@ static int ieee80211_ioctl_giwrange(struct net_device *dev,
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct iw_range *range = (struct iw_range *) extra;
-	struct ieee80211_hw_mode *mode = NULL;
+	enum ieee80211_band band;
 	int c = 0;
 
 	data->length = sizeof(struct iw_range);
@@ -191,24 +176,27 @@ static int ieee80211_ioctl_giwrange(struct net_device *dev,
 	range->enc_capa = IW_ENC_CAPA_WPA | IW_ENC_CAPA_WPA2 |
 			  IW_ENC_CAPA_CIPHER_TKIP | IW_ENC_CAPA_CIPHER_CCMP;
 
-	list_for_each_entry(mode, &local->modes_list, list) {
-		int i = 0;
 
-		if (!(local->enabled_modes & (1 << mode->mode)) ||
-		    (local->hw_modes & local->enabled_modes &
-		     (1 << MODE_IEEE80211G) && mode->mode == MODE_IEEE80211B))
+	for (band = 0; band < IEEE80211_NUM_BANDS; band ++) {
+		int i;
+		struct ieee80211_supported_band *sband;
+
+		sband = local->hw.wiphy->bands[band];
+
+		if (!sband)
 			continue;
 
-		while (i < mode->num_channels && c < IW_MAX_FREQUENCIES) {
-			struct ieee80211_channel *chan = &mode->channels[i];
+		for (i = 0; i < sband->n_channels && c < IW_MAX_FREQUENCIES; i++) {
+			struct ieee80211_channel *chan = &sband->channels[i];
 
-			if (chan->flag & IEEE80211_CHAN_W_SCAN) {
-				range->freq[c].i = chan->chan;
-				range->freq[c].m = chan->freq * 100000;
-				range->freq[c].e = 1;
+			if (!(chan->flags & IEEE80211_CHAN_DISABLED)) {
+				range->freq[c].i =
+					ieee80211_frequency_to_channel(
+						chan->center_freq);
+				range->freq[c].m = chan->center_freq;
+				range->freq[c].e = 6;
 				c++;
 			}
-			i++;
 		}
 	}
 	range->num_channels = c;
@@ -294,22 +282,29 @@ static int ieee80211_ioctl_giwmode(struct net_device *dev,
 	return 0;
 }
 
-int ieee80211_set_channel(struct ieee80211_local *local, int channel, int freq)
+int ieee80211_set_freq(struct ieee80211_local *local, int freqMHz)
 {
-	struct ieee80211_hw_mode *mode;
-	int c, set = 0;
+	int set = 0;
 	int ret = -EINVAL;
+	enum ieee80211_band band;
+	struct ieee80211_supported_band *sband;
+	int i;
+
+	for (band = 0; band < IEEE80211_NUM_BANDS; band ++) {
+		sband = local->hw.wiphy->bands[band];
 
-	list_for_each_entry(mode, &local->modes_list, list) {
-		if (!(local->enabled_modes & (1 << mode->mode)))
+		if (!sband)
 			continue;
-		for (c = 0; c < mode->num_channels; c++) {
-			struct ieee80211_channel *chan = &mode->channels[c];
-			if (chan->flag & IEEE80211_CHAN_W_SCAN &&
-			    ((chan->chan == channel) || (chan->freq == freq))) {
-				local->oper_channel = chan;
-				local->oper_hw_mode = mode;
+
+		for (i = 0; i < sband->n_channels; i++) {
+			struct ieee80211_channel *chan = &sband->channels[i];
+
+			if (chan->flags & IEEE80211_CHAN_DISABLED)
+				continue;
+
+			if (chan->center_freq == freqMHz) {
 				set = 1;
+				local->oper_channel = chan;
 				break;
 			}
 		}
@@ -347,13 +342,14 @@ static int ieee80211_ioctl_siwfreq(struct net_device *dev,
 					IEEE80211_STA_AUTO_CHANNEL_SEL;
 			return 0;
 		} else
-			return ieee80211_set_channel(local, freq->m, -1);
+			return ieee80211_set_freq(local,
+				ieee80211_channel_to_frequency(freq->m));
 	} else {
 		int i, div = 1000000;
 		for (i = 0; i < freq->e; i++)
 			div /= 10;
 		if (div > 0)
-			return ieee80211_set_channel(local, -1, freq->m / div);
+			return ieee80211_set_freq(local, freq->m / div);
 		else
 			return -EINVAL;
 	}
@@ -366,10 +362,7 @@ static int ieee80211_ioctl_giwfreq(struct net_device *dev,
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 
-	/* TODO: in station mode (Managed/Ad-hoc) might need to poll low-level
-	 * driver for the current channel with firmware-based management */
-
-	freq->m = local->hw.conf.freq;
+	freq->m = local->hw.conf.channel->center_freq;
 	freq->e = 6;
 
 	return 0;
@@ -566,15 +559,17 @@ static int ieee80211_ioctl_siwrate(struct net_device *dev,
 				  struct iw_param *rate, char *extra)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_hw_mode *mode;
-	int i;
+	int i, err = -EINVAL;
 	u32 target_rate = rate->value / 100000;
 	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_supported_band *sband;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	if (!sdata->bss)
 		return -ENODEV;
-	mode = local->oper_hw_mode;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
 	/* target_rate = -1, rate->fixed = 0 means auto only, so use all rates
 	 * target_rate = X, rate->fixed = 1 means only rate X
 	 * target_rate = X, rate->fixed = 0 means all rates <= X */
@@ -582,18 +577,20 @@ static int ieee80211_ioctl_siwrate(struct net_device *dev,
 	sdata->bss->force_unicast_rateidx = -1;
 	if (rate->value < 0)
 		return 0;
-	for (i=0; i < mode->num_rates; i++) {
-		struct ieee80211_rate *rates = &mode->rates[i];
-		int this_rate = rates->rate;
+
+	for (i=0; i< sband->n_bitrates; i++) {
+		struct ieee80211_rate *brate = &sband->bitrates[i];
+		int this_rate = brate->bitrate;
 
 		if (target_rate == this_rate) {
 			sdata->bss->max_ratectrl_rateidx = i;
 			if (rate->fixed)
 				sdata->bss->force_unicast_rateidx = i;
-			return 0;
+			err = 0;
+			break;
 		}
 	}
-	return -EINVAL;
+	return err;
 }
 
 static int ieee80211_ioctl_giwrate(struct net_device *dev,
@@ -603,18 +600,24 @@ static int ieee80211_ioctl_giwrate(struct net_device *dev,
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct sta_info *sta;
 	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_supported_band *sband;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+
 	if (sdata->vif.type == IEEE80211_IF_TYPE_STA)
 		sta = sta_info_get(local, sdata->u.sta.bssid);
 	else
 		return -EOPNOTSUPP;
 	if (!sta)
 		return -ENODEV;
-	if (sta->txrate < local->oper_hw_mode->num_rates)
-		rate->value = local->oper_hw_mode->rates[sta->txrate].rate * 100000;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
+	if (sta->txrate_idx < sband->n_bitrates)
+		rate->value = sband->bitrates[sta->txrate_idx].bitrate;
 	else
 		rate->value = 0;
+	rate->value *= 100000;
 	sta_info_put(sta);
 	return 0;
 }
@@ -625,7 +628,7 @@ static int ieee80211_ioctl_siwtxpower(struct net_device *dev,
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	bool need_reconfig = 0;
-	u8 new_power_level;
+	int new_power_level;
 
 	if ((data->txpower.flags & IW_TXPOW_TYPE) != IW_TXPOW_DBM)
 		return -EINVAL;
@@ -635,13 +638,15 @@ static int ieee80211_ioctl_siwtxpower(struct net_device *dev,
 	if (data->txpower.fixed) {
 		new_power_level = data->txpower.value;
 	} else {
-		/* Automatic power level. Get the px power from the current
-		 * channel. */
-		struct ieee80211_channel* chan = local->oper_channel;
+		/*
+		 * Automatic power level. Use maximum power for the current
+		 * channel. Should be part of rate control.
+		 */
+		struct ieee80211_channel* chan = local->hw.conf.channel;
 		if (!chan)
 			return -EINVAL;
 
-		new_power_level = chan->power_level;
+		new_power_level = chan->max_power;
 	}
 
 	if (local->hw.conf.power_level != new_power_level) {
diff --git a/net/mac80211/ieee80211_rate.c b/net/mac80211/ieee80211_rate.c
index b957e67c5fba..ebe29b716b27 100644
--- a/net/mac80211/ieee80211_rate.c
+++ b/net/mac80211/ieee80211_rate.c
@@ -163,7 +163,8 @@ static void rate_control_release(struct kref *kref)
 }
 
 void rate_control_get_rate(struct net_device *dev,
-			   struct ieee80211_hw_mode *mode, struct sk_buff *skb,
+			   struct ieee80211_supported_band *sband,
+			   struct sk_buff *skb,
 			   struct rate_selection *sel)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
@@ -174,17 +175,17 @@ void rate_control_get_rate(struct net_device *dev,
 
 	memset(sel, 0, sizeof(struct rate_selection));
 
-	ref->ops->get_rate(ref->priv, dev, mode, skb, sel);
+	ref->ops->get_rate(ref->priv, dev, sband, skb, sel);
 
 	/* Select a non-ERP backup rate. */
 	if (!sel->nonerp) {
-		for (i = 0; i < mode->num_rates - 1; i++) {
-			struct ieee80211_rate *rate = &mode->rates[i];
-			if (sel->rate->rate < rate->rate)
+		for (i = 0; i < sband->n_bitrates; i++) {
+			struct ieee80211_rate *rate = &sband->bitrates[i];
+			if (sel->rate->bitrate < rate->bitrate)
 				break;
 
-			if (rate_supported(sta, mode, i) &&
-			    !(rate->flags & IEEE80211_RATE_ERP))
+			if (rate_supported(sta, sband->band, i) &&
+			    !(rate->flags & IEEE80211_RATE_ERP_G))
 				sel->nonerp = rate;
 		}
 	}
diff --git a/net/mac80211/ieee80211_rate.h b/net/mac80211/ieee80211_rate.h
index 73f19e8aa51c..5f9a2ca49a57 100644
--- a/net/mac80211/ieee80211_rate.h
+++ b/net/mac80211/ieee80211_rate.h
@@ -18,6 +18,7 @@
 #include "ieee80211_i.h"
 #include "sta_info.h"
 
+/* TODO: kdoc */
 struct rate_selection {
 	/* Selected transmission rate */
 	struct ieee80211_rate *rate;
@@ -34,7 +35,8 @@ struct rate_control_ops {
 			  struct sk_buff *skb,
 			  struct ieee80211_tx_status *status);
 	void (*get_rate)(void *priv, struct net_device *dev,
-			 struct ieee80211_hw_mode *mode, struct sk_buff *skb,
+			 struct ieee80211_supported_band *band,
+			 struct sk_buff *skb,
 			 struct rate_selection *sel);
 	void (*rate_init)(void *priv, void *priv_sta,
 			  struct ieee80211_local *local, struct sta_info *sta);
@@ -66,7 +68,8 @@ void ieee80211_rate_control_unregister(struct rate_control_ops *ops);
 struct rate_control_ref *rate_control_alloc(const char *name,
 					    struct ieee80211_local *local);
 void rate_control_get_rate(struct net_device *dev,
-			   struct ieee80211_hw_mode *mode, struct sk_buff *skb,
+			   struct ieee80211_supported_band *sband,
+			   struct sk_buff *skb,
 			   struct rate_selection *sel);
 struct rate_control_ref *rate_control_get(struct rate_control_ref *ref);
 void rate_control_put(struct rate_control_ref *ref);
@@ -127,23 +130,23 @@ static inline void rate_control_remove_sta_debugfs(struct sta_info *sta)
 #endif
 }
 
-static inline int
-rate_supported(struct sta_info *sta, struct ieee80211_hw_mode *mode, int index)
+static inline int rate_supported(struct sta_info *sta,
+				 enum ieee80211_band band,
+				 int index)
 {
-	return (sta == NULL || sta->supp_rates & BIT(index)) &&
-	       (mode->rates[index].flags & IEEE80211_RATE_SUPPORTED);
+	return (sta == NULL || sta->supp_rates[band] & BIT(index));
 }
 
 static inline int
-rate_lowest_index(struct ieee80211_local *local, struct ieee80211_hw_mode *mode,
+rate_lowest_index(struct ieee80211_local *local,
+		  struct ieee80211_supported_band *sband,
 		  struct sta_info *sta)
 {
 	int i;
 
-	for (i = 0; i < mode->num_rates; i++) {
-		if (rate_supported(sta, mode, i))
+	for (i = 0; i < sband->n_bitrates; i++)
+		if (rate_supported(sta, sband->band, i))
 			return i;
-	}
 
 	/* warn when we cannot find a rate. */
 	WARN_ON(1);
@@ -152,10 +155,11 @@ rate_lowest_index(struct ieee80211_local *local, struct ieee80211_hw_mode *mode,
 }
 
 static inline struct ieee80211_rate *
-rate_lowest(struct ieee80211_local *local, struct ieee80211_hw_mode *mode,
+rate_lowest(struct ieee80211_local *local,
+	    struct ieee80211_supported_band *sband,
 	    struct sta_info *sta)
 {
-	return &mode->rates[rate_lowest_index(local, mode, sta)];
+	return &sband->bitrates[rate_lowest_index(local, sband, sta)];
 }
 
 
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index d0273ccbdbae..2628222a5085 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -74,7 +74,7 @@
 static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst,
 				     u8 *ssid, size_t ssid_len);
 static struct ieee80211_sta_bss *
-ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int channel,
+ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int freq,
 		     u8 *ssid, u8 ssid_len);
 static void ieee80211_rx_bss_put(struct net_device *dev,
 				 struct ieee80211_sta_bss *bss);
@@ -466,7 +466,7 @@ static void ieee80211_set_associated(struct net_device *dev,
 			return;
 
 		bss = ieee80211_rx_bss_get(dev, ifsta->bssid,
-					   local->hw.conf.channel,
+					   local->hw.conf.channel->center_freq,
 					   ifsta->ssid, ifsta->ssid_len);
 		if (bss) {
 			if (bss->has_erp_value)
@@ -593,7 +593,6 @@ static void ieee80211_send_assoc(struct net_device *dev,
 				 struct ieee80211_if_sta *ifsta)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_hw_mode *mode;
 	struct sk_buff *skb;
 	struct ieee80211_mgmt *mgmt;
 	u8 *pos, *ies;
@@ -601,6 +600,7 @@ static void ieee80211_send_assoc(struct net_device *dev,
 	u16 capab;
 	struct ieee80211_sta_bss *bss;
 	int wmm = 0;
+	struct ieee80211_supported_band *sband;
 
 	skb = dev_alloc_skb(local->hw.extra_tx_headroom +
 			    sizeof(*mgmt) + 200 + ifsta->extra_ie_len +
@@ -612,13 +612,19 @@ static void ieee80211_send_assoc(struct net_device *dev,
 	}
 	skb_reserve(skb, local->hw.extra_tx_headroom);
 
-	mode = local->oper_hw_mode;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
 	capab = ifsta->capab;
-	if (mode->mode == MODE_IEEE80211G) {
-		capab |= WLAN_CAPABILITY_SHORT_SLOT_TIME |
-			WLAN_CAPABILITY_SHORT_PREAMBLE;
+
+	if (local->hw.conf.channel->band == IEEE80211_BAND_2GHZ) {
+		if (!(local->hw.flags & IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE))
+			capab |= WLAN_CAPABILITY_SHORT_SLOT_TIME;
+		if (!(local->hw.flags & IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE))
+			capab |= WLAN_CAPABILITY_SHORT_PREAMBLE;
 	}
-	bss = ieee80211_rx_bss_get(dev, ifsta->bssid, local->hw.conf.channel,
+
+	bss = ieee80211_rx_bss_get(dev, ifsta->bssid,
+				   local->hw.conf.channel->center_freq,
 				   ifsta->ssid, ifsta->ssid_len);
 	if (bss) {
 		if (bss->capability & WLAN_CAPABILITY_PRIVACY)
@@ -657,23 +663,23 @@ static void ieee80211_send_assoc(struct net_device *dev,
 	*pos++ = ifsta->ssid_len;
 	memcpy(pos, ifsta->ssid, ifsta->ssid_len);
 
-	len = mode->num_rates;
+	len = sband->n_bitrates;
 	if (len > 8)
 		len = 8;
 	pos = skb_put(skb, len + 2);
 	*pos++ = WLAN_EID_SUPP_RATES;
 	*pos++ = len;
 	for (i = 0; i < len; i++) {
-		int rate = mode->rates[i].rate;
+		int rate = sband->bitrates[i].bitrate;
 		*pos++ = (u8) (rate / 5);
 	}
 
-	if (mode->num_rates > len) {
-		pos = skb_put(skb, mode->num_rates - len + 2);
+	if (sband->n_bitrates > len) {
+		pos = skb_put(skb, sband->n_bitrates - len + 2);
 		*pos++ = WLAN_EID_EXT_SUPP_RATES;
-		*pos++ = mode->num_rates - len;
-		for (i = len; i < mode->num_rates; i++) {
-			int rate = mode->rates[i].rate;
+		*pos++ = sband->n_bitrates - len;
+		for (i = len; i < sband->n_bitrates; i++) {
+			int rate = sband->bitrates[i].bitrate;
 			*pos++ = (u8) (rate / 5);
 		}
 	}
@@ -696,17 +702,18 @@ static void ieee80211_send_assoc(struct net_device *dev,
 		*pos++ = 0;
 	}
 	/* wmm support is a must to HT */
-	if (wmm && mode->ht_info.ht_supported) {
-		__le16 tmp = cpu_to_le16(mode->ht_info.cap);
+	if (wmm && sband->ht_info.ht_supported) {
+		__le16 tmp = cpu_to_le16(sband->ht_info.cap);
 		pos = skb_put(skb, sizeof(struct ieee80211_ht_cap)+2);
 		*pos++ = WLAN_EID_HT_CAPABILITY;
 		*pos++ = sizeof(struct ieee80211_ht_cap);
 		memset(pos, 0, sizeof(struct ieee80211_ht_cap));
 		memcpy(pos, &tmp, sizeof(u16));
 		pos += sizeof(u16);
-		*pos++ = (mode->ht_info.ampdu_factor |
-				(mode->ht_info.ampdu_density << 2));
-		memcpy(pos, mode->ht_info.supp_mcs_set, 16);
+		/* TODO: needs a define here for << 2 */
+		*pos++ = sband->ht_info.ampdu_factor |
+			 (sband->ht_info.ampdu_density << 2);
+		memcpy(pos, sband->ht_info.supp_mcs_set, 16);
 	}
 
 	kfree(ifsta->assocreq_ies);
@@ -789,7 +796,8 @@ static int ieee80211_privacy_mismatch(struct net_device *dev,
 	if (!ifsta || (ifsta->flags & IEEE80211_STA_MIXED_CELL))
 		return 0;
 
-	bss = ieee80211_rx_bss_get(dev, ifsta->bssid, local->hw.conf.channel,
+	bss = ieee80211_rx_bss_get(dev, ifsta->bssid,
+				   local->hw.conf.channel->center_freq,
 				   ifsta->ssid, ifsta->ssid_len);
 	if (!bss)
 		return 0;
@@ -899,7 +907,7 @@ static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst,
 				     u8 *ssid, size_t ssid_len)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	struct sk_buff *skb;
 	struct ieee80211_mgmt *mgmt;
 	u8 *pos, *supp_rates, *esupp_rates = NULL;
@@ -933,11 +941,10 @@ static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst,
 	supp_rates = skb_put(skb, 2);
 	supp_rates[0] = WLAN_EID_SUPP_RATES;
 	supp_rates[1] = 0;
-	mode = local->oper_hw_mode;
-	for (i = 0; i < mode->num_rates; i++) {
-		struct ieee80211_rate *rate = &mode->rates[i];
-		if (!(rate->flags & IEEE80211_RATE_SUPPORTED))
-			continue;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
+	for (i = 0; i < sband->n_bitrates; i++) {
+		struct ieee80211_rate *rate = &sband->bitrates[i];
 		if (esupp_rates) {
 			pos = skb_put(skb, 1);
 			esupp_rates[1]++;
@@ -950,7 +957,7 @@ static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst,
 			pos = skb_put(skb, 1);
 			supp_rates[1]++;
 		}
-		*pos = rate->rate / 5;
+		*pos = rate->bitrate / 5;
 	}
 
 	ieee80211_sta_tx(dev, skb, 0);
@@ -1146,9 +1153,11 @@ static void ieee80211_sta_process_addba_request(struct net_device *dev,
 	}
 	/* determine default buffer size */
 	if (buf_size == 0) {
-		struct ieee80211_hw_mode *mode = conf->mode;
+		struct ieee80211_supported_band *sband;
+
+		sband = local->hw.wiphy->bands[conf->channel->band];
 		buf_size = IEEE80211_MIN_AMPDU_BUF;
-		buf_size = buf_size << mode->ht_info.ampdu_factor;
+		buf_size = buf_size << sband->ht_info.ampdu_factor;
 	}
 
 	tid_agg_rx = &sta->ampdu_mlme.tid_rx[tid];
@@ -1718,15 +1727,16 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 {
 	struct ieee80211_local *local = sdata->local;
 	struct net_device *dev = sdata->dev;
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	struct sta_info *sta;
-	u32 rates;
+	u64 rates, basic_rates;
 	u16 capab_info, status_code, aid;
 	struct ieee802_11_elems elems;
 	struct ieee80211_bss_conf *bss_conf = &sdata->bss_conf;
 	u8 *pos;
 	int i, j;
 	DECLARE_MAC_BUF(mac);
+	bool have_higher_than_11mbit = false;
 
 	/* AssocResp and ReassocResp have identical structure, so process both
 	 * of them in this function. */
@@ -1796,10 +1806,6 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 	if (ifsta->assocresp_ies)
 		memcpy(ifsta->assocresp_ies, pos, ifsta->assocresp_ies_len);
 
-	/* set AID, ieee80211_set_associated() will tell the driver */
-	bss_conf->aid = aid;
-	ieee80211_set_associated(dev, ifsta, 1);
-
 	/* Add STA entry for the AP */
 	sta = sta_info_get(local, ifsta->bssid);
 	if (!sta) {
@@ -1811,7 +1817,7 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 			return;
 		}
 		bss = ieee80211_rx_bss_get(dev, ifsta->bssid,
-					   local->hw.conf.channel,
+					   local->hw.conf.channel->center_freq,
 					   ifsta->ssid, ifsta->ssid_len);
 		if (bss) {
 			sta->last_rssi = bss->rssi;
@@ -1825,20 +1831,46 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 	sta->flags |= WLAN_STA_AUTH | WLAN_STA_ASSOC | WLAN_STA_ASSOC_AP;
 
 	rates = 0;
-	mode = local->oper_hw_mode;
+	basic_rates = 0;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
 	for (i = 0; i < elems.supp_rates_len; i++) {
 		int rate = (elems.supp_rates[i] & 0x7f) * 5;
-		for (j = 0; j < mode->num_rates; j++)
-			if (mode->rates[j].rate == rate)
+
+		if (rate > 110)
+			have_higher_than_11mbit = true;
+
+		for (j = 0; j < sband->n_bitrates; j++) {
+			if (sband->bitrates[j].bitrate == rate)
 				rates |= BIT(j);
+			if (elems.supp_rates[i] & 0x80)
+				basic_rates |= BIT(j);
+		}
 	}
+
 	for (i = 0; i < elems.ext_supp_rates_len; i++) {
 		int rate = (elems.ext_supp_rates[i] & 0x7f) * 5;
-		for (j = 0; j < mode->num_rates; j++)
-			if (mode->rates[j].rate == rate)
+
+		if (rate > 110)
+			have_higher_than_11mbit = true;
+
+		for (j = 0; j < sband->n_bitrates; j++) {
+			if (sband->bitrates[j].bitrate == rate)
 				rates |= BIT(j);
+			if (elems.ext_supp_rates[i] & 0x80)
+				basic_rates |= BIT(j);
+		}
 	}
-	sta->supp_rates = rates;
+
+	sta->supp_rates[local->hw.conf.channel->band] = rates;
+	sdata->basic_rates = basic_rates;
+
+	/* cf. IEEE 802.11 9.2.12 */
+	if (local->hw.conf.channel->band == IEEE80211_BAND_2GHZ &&
+	    have_higher_than_11mbit)
+		sdata->flags |= IEEE80211_SDATA_OPERATING_GMODE;
+	else
+		sdata->flags &= ~IEEE80211_SDATA_OPERATING_GMODE;
 
 	if (elems.ht_cap_elem && elems.ht_info_elem && elems.wmm_param &&
 	    local->ops->conf_ht) {
@@ -1861,6 +1893,9 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 					 elems.wmm_param_len);
 	}
 
+	/* set AID, ieee80211_set_associated() will tell the driver */
+	bss_conf->aid = aid;
+	ieee80211_set_associated(dev, ifsta, 1);
 
 	sta_info_put(sta);
 
@@ -1901,7 +1936,7 @@ static void __ieee80211_rx_bss_hash_del(struct net_device *dev,
 
 
 static struct ieee80211_sta_bss *
-ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int channel,
+ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int freq,
 		     u8 *ssid, u8 ssid_len)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
@@ -1913,7 +1948,7 @@ ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int channel,
 	atomic_inc(&bss->users);
 	atomic_inc(&bss->users);
 	memcpy(bss->bssid, bssid, ETH_ALEN);
-	bss->channel = channel;
+	bss->freq = freq;
 	if (ssid && ssid_len <= IEEE80211_MAX_SSID_LEN) {
 		memcpy(bss->ssid, ssid, ssid_len);
 		bss->ssid_len = ssid_len;
@@ -1929,7 +1964,7 @@ ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int channel,
 
 
 static struct ieee80211_sta_bss *
-ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int channel,
+ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int freq,
 		     u8 *ssid, u8 ssid_len)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
@@ -1939,7 +1974,7 @@ ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int channel,
 	bss = local->sta_bss_hash[STA_HASH(bssid)];
 	while (bss) {
 		if (!memcmp(bss->bssid, bssid, ETH_ALEN) &&
-		    bss->channel == channel &&
+		    bss->freq == freq &&
 		    bss->ssid_len == ssid_len &&
 		    (ssid_len == 0 || !memcmp(bss->ssid, ssid, ssid_len))) {
 			atomic_inc(&bss->users);
@@ -2004,7 +2039,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee802_11_elems elems;
 	size_t baselen;
-	int channel, clen;
+	int freq, clen;
 	struct ieee80211_sta_bss *bss;
 	struct sta_info *sta;
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
@@ -2055,26 +2090,22 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates &&
 	    memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0 &&
 	    (sta = sta_info_get(local, mgmt->sa))) {
-		struct ieee80211_hw_mode *mode;
-		struct ieee80211_rate *rates;
+		struct ieee80211_supported_band *sband;
+		struct ieee80211_rate *bitrates;
 		size_t num_rates;
-		u32 supp_rates, prev_rates;
+		u64 supp_rates, prev_rates;
 		int i, j;
 
-		mode = local->sta_sw_scanning ?
-		       local->scan_hw_mode : local->oper_hw_mode;
-
-		if (local->sta_hw_scanning) {
-			/* search for the correct mode matches the beacon */
-			list_for_each_entry(mode, &local->modes_list, list)
-				if (mode->mode == rx_status->phymode)
-					break;
+		sband = local->hw.wiphy->bands[rx_status->band];
 
-			if (mode == NULL)
-				mode = local->oper_hw_mode;
+		if (!sband) {
+			WARN_ON(1);
+			sband = local->hw.wiphy->bands[
+					local->hw.conf.channel->band];
 		}
-		rates = mode->rates;
-		num_rates = mode->num_rates;
+
+		bitrates = sband->bitrates;
+		num_rates = sband->n_bitrates;
 
 		supp_rates = 0;
 		for (i = 0; i < elems.supp_rates_len +
@@ -2088,24 +2119,27 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 					[i - elems.supp_rates_len];
 			own_rate = 5 * (rate & 0x7f);
 			for (j = 0; j < num_rates; j++)
-				if (rates[j].rate == own_rate)
+				if (bitrates[j].bitrate == own_rate)
 					supp_rates |= BIT(j);
 		}
 
-		prev_rates = sta->supp_rates;
-		sta->supp_rates &= supp_rates;
-		if (sta->supp_rates == 0) {
+		prev_rates = sta->supp_rates[rx_status->band];
+		sta->supp_rates[rx_status->band] &= supp_rates;
+		if (sta->supp_rates[rx_status->band] == 0) {
 			/* No matching rates - this should not really happen.
 			 * Make sure that at least one rate is marked
 			 * supported to avoid issues with TX rate ctrl. */
-			sta->supp_rates = sdata->u.sta.supp_rates_bits;
+			sta->supp_rates[rx_status->band] =
+				sdata->u.sta.supp_rates_bits[rx_status->band];
 		}
-		if (sta->supp_rates != prev_rates) {
+		if (sta->supp_rates[rx_status->band] != prev_rates) {
 			printk(KERN_DEBUG "%s: updated supp_rates set for "
-			       "%s based on beacon info (0x%x & 0x%x -> "
-			       "0x%x)\n",
-			       dev->name, print_mac(mac, sta->addr), prev_rates,
-			       supp_rates, sta->supp_rates);
+			       "%s based on beacon info (0x%llx & 0x%llx -> "
+			       "0x%llx)\n",
+			       dev->name, print_mac(mac, sta->addr),
+			       (unsigned long long) prev_rates,
+			       (unsigned long long) supp_rates,
+			       (unsigned long long) sta->supp_rates[rx_status->band]);
 		}
 		sta_info_put(sta);
 	}
@@ -2114,14 +2148,14 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 		return;
 
 	if (elems.ds_params && elems.ds_params_len == 1)
-		channel = elems.ds_params[0];
+		freq = ieee80211_channel_to_frequency(elems.ds_params[0]);
 	else
-		channel = rx_status->channel;
+		freq = rx_status->freq;
 
-	bss = ieee80211_rx_bss_get(dev, mgmt->bssid, channel,
+	bss = ieee80211_rx_bss_get(dev, mgmt->bssid, freq,
 				   elems.ssid, elems.ssid_len);
 	if (!bss) {
-		bss = ieee80211_rx_bss_add(dev, mgmt->bssid, channel,
+		bss = ieee80211_rx_bss_add(dev, mgmt->bssid, freq,
 					   elems.ssid, elems.ssid_len);
 		if (!bss)
 			return;
@@ -2134,6 +2168,8 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 #endif
 	}
 
+	bss->band = rx_status->band;
+
 	if (bss->probe_resp && beacon) {
 		/* Do not allow beacon to override data from Probe Response. */
 		ieee80211_rx_bss_put(dev, bss);
@@ -2232,20 +2268,6 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 		bss->ht_ie_len = 0;
 	}
 
-	bss->hw_mode = rx_status->phymode;
-	bss->freq = rx_status->freq;
-	if (channel != rx_status->channel &&
-	    (bss->hw_mode == MODE_IEEE80211G ||
-	     bss->hw_mode == MODE_IEEE80211B) &&
-	    channel >= 1 && channel <= 14) {
-		static const int freq_list[] = {
-			2412, 2417, 2422, 2427, 2432, 2437, 2442,
-			2447, 2452, 2457, 2462, 2467, 2472, 2484
-		};
-		/* IEEE 802.11g/b mode can receive packets from neighboring
-		 * channels, so map the channel into frequency. */
-		bss->freq = freq_list[channel - 1];
-	}
 	bss->timestamp = timestamp;
 	bss->last_update = jiffies;
 	bss->rssi = rx_status->ssi;
@@ -2817,7 +2839,7 @@ static int ieee80211_sta_config_auth(struct net_device *dev,
 	}
 
 	spin_lock_bh(&local->sta_bss_lock);
-	freq = local->oper_channel->freq;
+	freq = local->oper_channel->center_freq;
 	list_for_each_entry(bss, &local->sta_bss_list, list) {
 		if (!(bss->capability & WLAN_CAPABILITY_ESS))
 			continue;
@@ -2848,7 +2870,7 @@ static int ieee80211_sta_config_auth(struct net_device *dev,
 	spin_unlock_bh(&local->sta_bss_lock);
 
 	if (selected) {
-		ieee80211_set_channel(local, -1, selected->freq);
+		ieee80211_set_freq(local, selected->freq);
 		if (!(ifsta->flags & IEEE80211_STA_SSID_SET))
 			ieee80211_sta_set_ssid(dev, selected->ssid,
 					       selected->ssid_len);
@@ -2881,10 +2903,12 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 	struct sk_buff *skb;
 	struct ieee80211_mgmt *mgmt;
 	struct ieee80211_tx_control control;
-	struct ieee80211_hw_mode *mode;
 	struct rate_selection ratesel;
 	u8 *pos;
 	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_supported_band *sband;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
 	/* Remove possible STA entries from other IBSS networks. */
 	sta_info_flush(local, NULL);
@@ -2904,12 +2928,11 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 	sdata->drop_unencrypted = bss->capability &
 		WLAN_CAPABILITY_PRIVACY ? 1 : 0;
 
-	res = ieee80211_set_channel(local, -1, bss->freq);
+	res = ieee80211_set_freq(local, bss->freq);
 
-	if (!(local->oper_channel->flag & IEEE80211_CHAN_W_IBSS)) {
-		printk(KERN_DEBUG "%s: IBSS not allowed on channel %d "
-		       "(%d MHz)\n", dev->name, local->hw.conf.channel,
-		       local->hw.conf.freq);
+	if (local->oper_channel->flags & IEEE80211_CHAN_NO_IBSS) {
+		printk(KERN_DEBUG "%s: IBSS not allowed on frequency "
+		       "%d MHz\n", dev->name, local->oper_channel->center_freq);
 		return -1;
 	}
 
@@ -2946,10 +2969,12 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 		*pos++ = rates;
 		memcpy(pos, bss->supp_rates, rates);
 
-		pos = skb_put(skb, 2 + 1);
-		*pos++ = WLAN_EID_DS_PARAMS;
-		*pos++ = 1;
-		*pos++ = bss->channel;
+		if (bss->band == IEEE80211_BAND_2GHZ) {
+			pos = skb_put(skb, 2 + 1);
+			*pos++ = WLAN_EID_DS_PARAMS;
+			*pos++ = 1;
+			*pos++ = ieee80211_frequency_to_channel(bss->freq);
+		}
 
 		pos = skb_put(skb, 2 + 2);
 		*pos++ = WLAN_EID_IBSS_PARAMS;
@@ -2967,19 +2992,18 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 		}
 
 		memset(&control, 0, sizeof(control));
-		rate_control_get_rate(dev, local->oper_hw_mode, skb, &ratesel);
+		rate_control_get_rate(dev, sband, skb, &ratesel);
 		if (!ratesel.rate) {
 			printk(KERN_DEBUG "%s: Failed to determine TX rate "
 			       "for IBSS beacon\n", dev->name);
 			break;
 		}
 		control.vif = &sdata->vif;
-		control.tx_rate =
-			(sdata->bss_conf.use_short_preamble &&
-			(ratesel.rate->flags & IEEE80211_RATE_PREAMBLE2)) ?
-			ratesel.rate->val2 : ratesel.rate->val;
+		control.tx_rate = ratesel.rate;
+		if (sdata->bss_conf.use_short_preamble &&
+		    ratesel.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE)
+			control.flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
 		control.antenna_sel_tx = local->hw.conf.antenna_sel_tx;
-		control.power_level = local->hw.conf.power_level;
 		control.flags |= IEEE80211_TXCTL_NO_ACK;
 		control.retry_limit = 1;
 
@@ -3004,14 +3028,14 @@ static int ieee80211_sta_join_ibss(struct net_device *dev,
 		}
 
 		rates = 0;
-		mode = local->oper_hw_mode;
+		sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 		for (i = 0; i < bss->supp_rates_len; i++) {
 			int bitrate = (bss->supp_rates[i] & 0x7f) * 5;
-			for (j = 0; j < mode->num_rates; j++)
-				if (mode->rates[j].rate == bitrate)
+			for (j = 0; j < sband->n_bitrates; j++)
+				if (sband->bitrates[j].bitrate == bitrate)
 					rates |= BIT(j);
 		}
-		ifsta->supp_rates_bits = rates;
+		ifsta->supp_rates_bits[local->hw.conf.channel->band] = rates;
 	} while (0);
 
 	if (skb) {
@@ -3035,7 +3059,7 @@ static int ieee80211_sta_create_ibss(struct net_device *dev,
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct ieee80211_sta_bss *bss;
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	u8 bssid[ETH_ALEN], *pos;
 	int i;
 	DECLARE_MAC_BUF(mac);
@@ -3057,28 +3081,28 @@ static int ieee80211_sta_create_ibss(struct net_device *dev,
 	printk(KERN_DEBUG "%s: Creating new IBSS network, BSSID %s\n",
 	       dev->name, print_mac(mac, bssid));
 
-	bss = ieee80211_rx_bss_add(dev, bssid, local->hw.conf.channel,
+	bss = ieee80211_rx_bss_add(dev, bssid,
+				   local->hw.conf.channel->center_freq,
 				   sdata->u.sta.ssid, sdata->u.sta.ssid_len);
 	if (!bss)
 		return -ENOMEM;
 
-	mode = local->oper_hw_mode;
+	bss->band = local->hw.conf.channel->band;
+	sband = local->hw.wiphy->bands[bss->band];
 
 	if (local->hw.conf.beacon_int == 0)
 		local->hw.conf.beacon_int = 100;
 	bss->beacon_int = local->hw.conf.beacon_int;
-	bss->hw_mode = local->hw.conf.phymode;
-	bss->freq = local->hw.conf.freq;
 	bss->last_update = jiffies;
 	bss->capability = WLAN_CAPABILITY_IBSS;
 	if (sdata->default_key) {
 		bss->capability |= WLAN_CAPABILITY_PRIVACY;
 	} else
 		sdata->drop_unencrypted = 0;
-	bss->supp_rates_len = mode->num_rates;
+	bss->supp_rates_len = sband->n_bitrates;
 	pos = bss->supp_rates;
-	for (i = 0; i < mode->num_rates; i++) {
-		int rate = mode->rates[i].rate;
+	for (i = 0; i < sband->n_bitrates; i++) {
+		int rate = sband->bitrates[i].bitrate;
 		*pos++ = (u8) (rate / 5);
 	}
 
@@ -3127,7 +3151,8 @@ static int ieee80211_sta_find_ibss(struct net_device *dev,
 	       "%s\n", print_mac(mac, bssid), print_mac(mac2, ifsta->bssid));
 #endif /* CONFIG_MAC80211_IBSS_DEBUG */
 	if (found && memcmp(ifsta->bssid, bssid, ETH_ALEN) != 0 &&
-	    (bss = ieee80211_rx_bss_get(dev, bssid, local->hw.conf.channel,
+	    (bss = ieee80211_rx_bss_get(dev, bssid,
+					local->hw.conf.channel->center_freq,
 					ifsta->ssid, ifsta->ssid_len))) {
 		printk(KERN_DEBUG "%s: Selected IBSS BSSID %s"
 		       " based on configured SSID\n",
@@ -3155,13 +3180,13 @@ static int ieee80211_sta_find_ibss(struct net_device *dev,
 		if (time_after(jiffies, ifsta->ibss_join_req +
 			       IEEE80211_IBSS_JOIN_TIMEOUT)) {
 			if ((ifsta->flags & IEEE80211_STA_CREATE_IBSS) &&
-			    local->oper_channel->flag & IEEE80211_CHAN_W_IBSS)
+			    (!(local->oper_channel->flags &
+					IEEE80211_CHAN_NO_IBSS)))
 				return ieee80211_sta_create_ibss(dev, ifsta);
 			if (ifsta->flags & IEEE80211_STA_CREATE_IBSS) {
-				printk(KERN_DEBUG "%s: IBSS not allowed on the"
-				       " configured channel %d (%d MHz)\n",
-				       dev->name, local->hw.conf.channel,
-				       local->hw.conf.freq);
+				printk(KERN_DEBUG "%s: IBSS not allowed on"
+				       " %d MHz\n", dev->name,
+				       local->hw.conf.channel->center_freq);
 			}
 
 			/* No IBSS found - decrease scan interval and continue
@@ -3180,7 +3205,7 @@ static int ieee80211_sta_find_ibss(struct net_device *dev,
 
 int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len)
 {
-	struct ieee80211_sub_if_data *sdata;
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct ieee80211_if_sta *ifsta;
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 
@@ -3194,18 +3219,23 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len)
 		int i;
 
 		memset(&qparam, 0, sizeof(qparam));
-		/* TODO: are these ok defaults for all hw_modes? */
+
 		qparam.aifs = 2;
-		qparam.cw_min =
-			local->hw.conf.phymode == MODE_IEEE80211B ? 31 : 15;
+
+		if (local->hw.conf.channel->band == IEEE80211_BAND_2GHZ &&
+		    !(sdata->flags & IEEE80211_SDATA_OPERATING_GMODE))
+			qparam.cw_min = 31;
+		else
+			qparam.cw_min = 15;
+
 		qparam.cw_max = 1023;
 		qparam.burst_time = 0;
+
 		for (i = IEEE80211_TX_QUEUE_DATA0; i < NUM_TX_DATA_QUEUES; i++)
-		{
 			local->ops->conf_tx(local_to_hw(local),
 					   i + IEEE80211_TX_QUEUE_DATA0,
 					   &qparam);
-		}
+
 		/* IBSS uses different parameters for Beacon sending */
 		qparam.cw_min++;
 		qparam.cw_min *= 2;
@@ -3214,7 +3244,6 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len)
 				   IEEE80211_TX_QUEUE_BEACON, &qparam);
 	}
 
-	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	ifsta = &sdata->u.sta;
 
 	if (ifsta->ssid_len != len || memcmp(ifsta->ssid, ssid, len) != 0)
@@ -3373,7 +3402,7 @@ void ieee80211_sta_scan_work(struct work_struct *work)
 		container_of(work, struct ieee80211_local, scan_work.work);
 	struct net_device *dev = local->scan_dev;
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	struct ieee80211_channel *chan;
 	int skip;
 	unsigned long next_delay = 0;
@@ -3383,44 +3412,47 @@ void ieee80211_sta_scan_work(struct work_struct *work)
 
 	switch (local->scan_state) {
 	case SCAN_SET_CHANNEL:
-		mode = local->scan_hw_mode;
-		if (local->scan_hw_mode->list.next == &local->modes_list &&
-		    local->scan_channel_idx >= mode->num_channels) {
+		/* get current scan band */
+		if (local->scan_band < IEEE80211_NUM_BANDS)
+			sband = local->hw.wiphy->bands[local->scan_band];
+		else
+			sband = NULL;
+
+		/* if we started at an unsupported one, advance */
+		while (!sband && local->scan_band < IEEE80211_NUM_BANDS) {
+			local->scan_band++;
+			sband = local->hw.wiphy->bands[local->scan_band];
+			local->scan_channel_idx = 0;
+		}
+
+		if (!sband ||
+		    (local->scan_channel_idx >= sband->n_channels &&
+		     local->scan_band >= IEEE80211_NUM_BANDS)) {
 			ieee80211_scan_completed(local_to_hw(local));
 			return;
 		}
-		skip = !(local->enabled_modes & (1 << mode->mode));
-		chan = &mode->channels[local->scan_channel_idx];
-		if (!(chan->flag & IEEE80211_CHAN_W_SCAN) ||
+		skip = 0;
+		chan = &sband->channels[local->scan_channel_idx];
+
+		if (chan->flags & IEEE80211_CHAN_DISABLED ||
 		    (sdata->vif.type == IEEE80211_IF_TYPE_IBSS &&
-		     !(chan->flag & IEEE80211_CHAN_W_IBSS)) ||
-		    (local->hw_modes & local->enabled_modes &
-		     (1 << MODE_IEEE80211G) && mode->mode == MODE_IEEE80211B))
+		     chan->flags & IEEE80211_CHAN_NO_IBSS))
 			skip = 1;
 
 		if (!skip) {
-#if 0
-			printk(KERN_DEBUG "%s: scan channel %d (%d MHz)\n",
-			       dev->name, chan->chan, chan->freq);
-#endif
-
 			local->scan_channel = chan;
 			if (ieee80211_hw_config(local)) {
-				printk(KERN_DEBUG "%s: failed to set channel "
-				       "%d (%d MHz) for scan\n", dev->name,
-				       chan->chan, chan->freq);
+				printk(KERN_DEBUG "%s: failed to set freq to "
+				       "%d MHz for scan\n", dev->name,
+				       chan->center_freq);
 				skip = 1;
 			}
 		}
 
 		local->scan_channel_idx++;
-		if (local->scan_channel_idx >= local->scan_hw_mode->num_channels) {
-			if (local->scan_hw_mode->list.next != &local->modes_list) {
-				local->scan_hw_mode = list_entry(local->scan_hw_mode->list.next,
-								 struct ieee80211_hw_mode,
-								 list);
-				local->scan_channel_idx = 0;
-			}
+		if (local->scan_channel_idx >= sband->n_channels) {
+			local->scan_band++;
+			local->scan_channel_idx = 0;
 		}
 
 		if (skip)
@@ -3431,13 +3463,14 @@ void ieee80211_sta_scan_work(struct work_struct *work)
 		local->scan_state = SCAN_SEND_PROBE;
 		break;
 	case SCAN_SEND_PROBE:
-		if (local->scan_channel->flag & IEEE80211_CHAN_W_ACTIVE_SCAN) {
-			ieee80211_send_probe_req(dev, NULL, local->scan_ssid,
-						 local->scan_ssid_len);
-			next_delay = IEEE80211_CHANNEL_TIME;
-		} else
-			next_delay = IEEE80211_PASSIVE_CHANNEL_TIME;
+		next_delay = IEEE80211_PASSIVE_CHANNEL_TIME;
 		local->scan_state = SCAN_SET_CHANNEL;
+
+		if (local->scan_channel->flags & IEEE80211_CHAN_PASSIVE_SCAN)
+			break;
+		ieee80211_send_probe_req(dev, NULL, local->scan_ssid,
+					 local->scan_ssid_len);
+		next_delay = IEEE80211_CHANNEL_TIME;
 		break;
 	}
 
@@ -3512,10 +3545,8 @@ static int ieee80211_sta_start_scan(struct net_device *dev,
 	} else
 		local->scan_ssid_len = 0;
 	local->scan_state = SCAN_SET_CHANNEL;
-	local->scan_hw_mode = list_entry(local->modes_list.next,
-					 struct ieee80211_hw_mode,
-					 list);
 	local->scan_channel_idx = 0;
+	local->scan_band = IEEE80211_BAND_2GHZ;
 	local->scan_dev = dev;
 
 	netif_tx_lock_bh(local->mdev);
@@ -3570,9 +3601,6 @@ ieee80211_sta_scan_result(struct net_device *dev,
 		       bss->last_update + IEEE80211_SCAN_RESULT_EXPIRE))
 		return current_ev;
 
-	if (!(local->enabled_modes & (1 << bss->hw_mode)))
-		return current_ev;
-
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
@@ -3600,12 +3628,15 @@ ieee80211_sta_scan_result(struct net_device *dev,
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWFREQ;
-	iwe.u.freq.m = bss->channel;
-	iwe.u.freq.e = 0;
+	iwe.u.freq.m = bss->freq;
+	iwe.u.freq.e = 6;
 	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
 					  IW_EV_FREQ_LEN);
-	iwe.u.freq.m = bss->freq * 100000;
-	iwe.u.freq.e = 1;
+
+	memset(&iwe, 0, sizeof(iwe));
+	iwe.cmd = SIOCGIWFREQ;
+	iwe.u.freq.m = ieee80211_frequency_to_channel(bss->freq);
+	iwe.u.freq.e = 0;
 	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
 					  IW_EV_FREQ_LEN);
 
@@ -3748,7 +3779,8 @@ struct sta_info * ieee80211_ibss_add_sta(struct net_device *dev,
 	if (!sta)
 		return NULL;
 
-	sta->supp_rates = sdata->u.sta.supp_rates_bits;
+	sta->supp_rates[local->hw.conf.channel->band] =
+		sdata->u.sta.supp_rates_bits[local->hw.conf.channel->band];
 
 	rate_control_rate_init(sta, local);
 
diff --git a/net/mac80211/rc80211_pid_algo.c b/net/mac80211/rc80211_pid_algo.c
index c339571632b2..c5a607ca8440 100644
--- a/net/mac80211/rc80211_pid_algo.c
+++ b/net/mac80211/rc80211_pid_algo.c
@@ -102,23 +102,23 @@ static void rate_control_pid_adjust_rate(struct ieee80211_local *local,
 					 struct rc_pid_rateinfo *rinfo)
 {
 	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	int newidx;
 	int maxrate;
 	int back = (adj > 0) ? 1 : -1;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
 
-	mode = local->oper_hw_mode;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 	maxrate = sdata->bss ? sdata->bss->max_ratectrl_rateidx : -1;
 
-	newidx = rate_control_pid_shift_adjust(rinfo, adj, sta->txrate,
-					       mode->num_rates);
+	newidx = rate_control_pid_shift_adjust(rinfo, adj, sta->txrate_idx,
+					       sband->n_bitrates);
 
-	while (newidx != sta->txrate) {
-		if (rate_supported(sta, mode, newidx) &&
+	while (newidx != sta->txrate_idx) {
+		if (rate_supported(sta, sband->band, newidx) &&
 		    (maxrate < 0 || newidx <= maxrate)) {
-			sta->txrate = newidx;
+			sta->txrate_idx = newidx;
 			break;
 		}
 
@@ -128,7 +128,7 @@ static void rate_control_pid_adjust_rate(struct ieee80211_local *local,
 #ifdef CONFIG_MAC80211_DEBUGFS
 	rate_control_pid_event_rate_change(
 		&((struct rc_pid_sta_info *)sta->rate_ctrl_priv)->events,
-		newidx, mode->rates[newidx].rate);
+		newidx, sband->bitrates[newidx].bitrate);
 #endif
 }
 
@@ -155,7 +155,7 @@ static void rate_control_pid_sample(struct rc_pid_info *pinfo,
 {
 	struct rc_pid_sta_info *spinfo = sta->rate_ctrl_priv;
 	struct rc_pid_rateinfo *rinfo = pinfo->rinfo;
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	u32 pf;
 	s32 err_avg;
 	u32 err_prop;
@@ -164,7 +164,7 @@ static void rate_control_pid_sample(struct rc_pid_info *pinfo,
 	int adj, i, j, tmp;
 	unsigned long period;
 
-	mode = local->oper_hw_mode;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 	spinfo = sta->rate_ctrl_priv;
 
 	/* In case nothing happened during the previous control interval, turn
@@ -190,18 +190,18 @@ static void rate_control_pid_sample(struct rc_pid_info *pinfo,
 	spinfo->tx_num_failed = 0;
 
 	/* If we just switched rate, update the rate behaviour info. */
-	if (pinfo->oldrate != sta->txrate) {
+	if (pinfo->oldrate != sta->txrate_idx) {
 
 		i = rinfo[pinfo->oldrate].rev_index;
-		j = rinfo[sta->txrate].rev_index;
+		j = rinfo[sta->txrate_idx].rev_index;
 
 		tmp = (pf - spinfo->last_pf);
 		tmp = RC_PID_DO_ARITH_RIGHT_SHIFT(tmp, RC_PID_ARITH_SHIFT);
 
 		rinfo[j].diff = rinfo[i].diff + tmp;
-		pinfo->oldrate = sta->txrate;
+		pinfo->oldrate = sta->txrate_idx;
 	}
-	rate_control_pid_normalize(pinfo, mode->num_rates);
+	rate_control_pid_normalize(pinfo, sband->n_bitrates);
 
 	/* Compute the proportional, integral and derivative errors. */
 	err_prop = (pinfo->target << RC_PID_ARITH_SHIFT) - pf;
@@ -242,8 +242,10 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev,
 	struct sta_info *sta;
 	struct rc_pid_sta_info *spinfo;
 	unsigned long period;
+	struct ieee80211_supported_band *sband;
 
 	sta = sta_info_get(local, hdr->addr1);
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
 	if (!sta)
 		return;
@@ -251,13 +253,13 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev,
 	/* Don't update the state if we're not controlling the rate. */
 	sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
 	if (sdata->bss && sdata->bss->force_unicast_rateidx > -1) {
-		sta->txrate = sdata->bss->max_ratectrl_rateidx;
+		sta->txrate_idx = sdata->bss->max_ratectrl_rateidx;
 		return;
 	}
 
 	/* Ignore all frames that were sent with a different rate than the rate
 	 * we currently advise mac80211 to use. */
-	if (status->control.rate != &local->oper_hw_mode->rates[sta->txrate])
+	if (status->control.tx_rate != &sband->bitrates[sta->txrate_idx])
 		goto ignore;
 
 	spinfo = sta->rate_ctrl_priv;
@@ -304,7 +306,7 @@ ignore:
 }
 
 static void rate_control_pid_get_rate(void *priv, struct net_device *dev,
-				      struct ieee80211_hw_mode *mode,
+				      struct ieee80211_supported_band *sband,
 				      struct sk_buff *skb,
 				      struct rate_selection *sel)
 {
@@ -322,7 +324,7 @@ static void rate_control_pid_get_rate(void *priv, struct net_device *dev,
 	fc = le16_to_cpu(hdr->frame_control);
 	if ((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA ||
 	    is_multicast_ether_addr(hdr->addr1) || !sta) {
-		sel->rate = rate_lowest(local, mode, sta);
+		sel->rate = rate_lowest(local, sband, sta);
 		if (sta)
 			sta_info_put(sta);
 		return;
@@ -331,23 +333,23 @@ static void rate_control_pid_get_rate(void *priv, struct net_device *dev,
 	/* If a forced rate is in effect, select it. */
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	if (sdata->bss && sdata->bss->force_unicast_rateidx > -1)
-		sta->txrate = sdata->bss->force_unicast_rateidx;
+		sta->txrate_idx = sdata->bss->force_unicast_rateidx;
 
-	rateidx = sta->txrate;
+	rateidx = sta->txrate_idx;
 
-	if (rateidx >= mode->num_rates)
-		rateidx = mode->num_rates - 1;
+	if (rateidx >= sband->n_bitrates)
+		rateidx = sband->n_bitrates - 1;
 
-	sta->last_txrate = rateidx;
+	sta->last_txrate_idx = rateidx;
 
 	sta_info_put(sta);
 
-	sel->rate = &mode->rates[rateidx];
+	sel->rate = &sband->bitrates[rateidx];
 
 #ifdef CONFIG_MAC80211_DEBUGFS
 	rate_control_pid_event_tx_rate(
 		&((struct rc_pid_sta_info *) sta->rate_ctrl_priv)->events,
-		rateidx, mode->rates[rateidx].rate);
+		rateidx, sband->bitrates[rateidx].bitrate);
 #endif
 }
 
@@ -359,28 +361,32 @@ static void rate_control_pid_rate_init(void *priv, void *priv_sta,
 	 * as we need to have IEEE 802.1X auth succeed immediately after assoc..
 	 * Until that method is implemented, we will use the lowest supported
 	 * rate as a workaround. */
-	sta->txrate = rate_lowest_index(local, local->oper_hw_mode, sta);
+	struct ieee80211_supported_band *sband;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+	sta->txrate_idx = rate_lowest_index(local, sband, sta);
 }
 
 static void *rate_control_pid_alloc(struct ieee80211_local *local)
 {
 	struct rc_pid_info *pinfo;
 	struct rc_pid_rateinfo *rinfo;
-	struct ieee80211_hw_mode *mode;
+	struct ieee80211_supported_band *sband;
 	int i, j, tmp;
 	bool s;
 #ifdef CONFIG_MAC80211_DEBUGFS
 	struct rc_pid_debugfs_entries *de;
 #endif
 
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
 	pinfo = kmalloc(sizeof(*pinfo), GFP_ATOMIC);
 	if (!pinfo)
 		return NULL;
 
-	/* We can safely assume that oper_hw_mode won't change unless we get
+	/* We can safely assume that sband won't change unless we get
 	 * reinitialized. */
-	mode = local->oper_hw_mode;
-	rinfo = kmalloc(sizeof(*rinfo) * mode->num_rates, GFP_ATOMIC);
+	rinfo = kmalloc(sizeof(*rinfo) * sband->n_bitrates, GFP_ATOMIC);
 	if (!rinfo) {
 		kfree(pinfo);
 		return NULL;
@@ -389,7 +395,7 @@ static void *rate_control_pid_alloc(struct ieee80211_local *local)
 	/* Sort the rates. This is optimized for the most common case (i.e.
 	 * almost-sorted CCK+OFDM rates). Kind of bubble-sort with reversed
 	 * mapping too. */
-	for (i = 0; i < mode->num_rates; i++) {
+	for (i = 0; i < sband->n_bitrates; i++) {
 		rinfo[i].index = i;
 		rinfo[i].rev_index = i;
 		if (pinfo->fast_start)
@@ -397,11 +403,11 @@ static void *rate_control_pid_alloc(struct ieee80211_local *local)
 		else
 			rinfo[i].diff = i * pinfo->norm_offset;
 	}
-	for (i = 1; i < mode->num_rates; i++) {
+	for (i = 1; i < sband->n_bitrates; i++) {
 		s = 0;
-		for (j = 0; j < mode->num_rates - i; j++)
-			if (unlikely(mode->rates[rinfo[j].index].rate >
-				     mode->rates[rinfo[j + 1].index].rate)) {
+		for (j = 0; j < sband->n_bitrates - i; j++)
+			if (unlikely(sband->bitrates[rinfo[j].index].bitrate >
+				     sband->bitrates[rinfo[j + 1].index].bitrate)) {
 				tmp = rinfo[j].index;
 				rinfo[j].index = rinfo[j + 1].index;
 				rinfo[j + 1].index = tmp;
diff --git a/net/mac80211/rc80211_simple.c b/net/mac80211/rc80211_simple.c
index 9a78b116acff..c4678905a142 100644
--- a/net/mac80211/rc80211_simple.c
+++ b/net/mac80211/rc80211_simple.c
@@ -35,8 +35,8 @@ static void rate_control_rate_inc(struct ieee80211_local *local,
 				  struct sta_info *sta)
 {
 	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_hw_mode *mode;
-	int i = sta->txrate;
+	struct ieee80211_supported_band *sband;
+	int i = sta->txrate_idx;
 	int maxrate;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
@@ -45,18 +45,17 @@ static void rate_control_rate_inc(struct ieee80211_local *local,
 		return;
 	}
 
-	mode = local->oper_hw_mode;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 	maxrate = sdata->bss ? sdata->bss->max_ratectrl_rateidx : -1;
 
-	if (i > mode->num_rates)
-		i = mode->num_rates - 2;
+	if (i > sband->n_bitrates)
+		i = sband->n_bitrates - 2;
 
-	while (i + 1 < mode->num_rates) {
+	while (i + 1 < sband->n_bitrates) {
 		i++;
-		if (sta->supp_rates & BIT(i) &&
-		    mode->rates[i].flags & IEEE80211_RATE_SUPPORTED &&
+		if (rate_supported(sta, sband->band, i) &&
 		    (maxrate < 0 || i <= maxrate)) {
-			sta->txrate = i;
+			sta->txrate_idx = i;
 			break;
 		}
 	}
@@ -67,8 +66,8 @@ static void rate_control_rate_dec(struct ieee80211_local *local,
 				  struct sta_info *sta)
 {
 	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_hw_mode *mode;
-	int i = sta->txrate;
+	struct ieee80211_supported_band *sband;
+	int i = sta->txrate_idx;
 
 	sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
 	if (sdata->bss && sdata->bss->force_unicast_rateidx > -1) {
@@ -76,15 +75,14 @@ static void rate_control_rate_dec(struct ieee80211_local *local,
 		return;
 	}
 
-	mode = local->oper_hw_mode;
-	if (i > mode->num_rates)
-		i = mode->num_rates;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+	if (i > sband->n_bitrates)
+		i = sband->n_bitrates;
 
 	while (i > 0) {
 		i--;
-		if (sta->supp_rates & BIT(i) &&
-		    mode->rates[i].flags & IEEE80211_RATE_SUPPORTED) {
-			sta->txrate = i;
+		if (rate_supported(sta, sband->band, i)) {
+			sta->txrate_idx = i;
 			break;
 		}
 	}
@@ -168,7 +166,7 @@ static void rate_control_simple_tx_status(void *priv, struct net_device *dev,
 		} else if (per_failed < RATE_CONTROL_NUM_UP) {
 			rate_control_rate_inc(local, sta);
 		}
-		srctrl->tx_avg_rate_sum += status->control.rate->rate;
+		srctrl->tx_avg_rate_sum += status->control.tx_rate->bitrate;
 		srctrl->tx_avg_rate_num++;
 		srctrl->tx_num_failures = 0;
 		srctrl->tx_num_xmit = 0;
@@ -201,7 +199,7 @@ static void rate_control_simple_tx_status(void *priv, struct net_device *dev,
 
 static void
 rate_control_simple_get_rate(void *priv, struct net_device *dev,
-			     struct ieee80211_hw_mode *mode,
+			     struct ieee80211_supported_band *sband,
 			     struct sk_buff *skb,
 			     struct rate_selection *sel)
 {
@@ -219,7 +217,7 @@ rate_control_simple_get_rate(void *priv, struct net_device *dev,
 	fc = le16_to_cpu(hdr->frame_control);
 	if ((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA ||
 	    is_multicast_ether_addr(hdr->addr1) || !sta) {
-		sel->rate = rate_lowest(local, mode, sta);
+		sel->rate = rate_lowest(local, sband, sta);
 		if (sta)
 			sta_info_put(sta);
 		return;
@@ -228,18 +226,18 @@ rate_control_simple_get_rate(void *priv, struct net_device *dev,
 	/* If a forced rate is in effect, select it. */
 	sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	if (sdata->bss && sdata->bss->force_unicast_rateidx > -1)
-		sta->txrate = sdata->bss->force_unicast_rateidx;
+		sta->txrate_idx = sdata->bss->force_unicast_rateidx;
 
-	rateidx = sta->txrate;
+	rateidx = sta->txrate_idx;
 
-	if (rateidx >= mode->num_rates)
-		rateidx = mode->num_rates - 1;
+	if (rateidx >= sband->n_bitrates)
+		rateidx = sband->n_bitrates - 1;
 
-	sta->last_txrate = rateidx;
+	sta->last_txrate_idx = rateidx;
 
 	sta_info_put(sta);
 
-	sel->rate = &mode->rates[rateidx];
+	sel->rate = &sband->bitrates[rateidx];
 }
 
 
@@ -247,21 +245,15 @@ static void rate_control_simple_rate_init(void *priv, void *priv_sta,
 					  struct ieee80211_local *local,
 					  struct sta_info *sta)
 {
-	struct ieee80211_hw_mode *mode;
-	int i;
-	sta->txrate = 0;
-	mode = local->oper_hw_mode;
+	struct ieee80211_supported_band *sband;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
 	/* TODO: This routine should consider using RSSI from previous packets
 	 * as we need to have IEEE 802.1X auth succeed immediately after assoc..
 	 * Until that method is implemented, we will use the lowest supported rate
 	 * as a workaround, */
-	for (i = 0; i < mode->num_rates; i++) {
-		if ((sta->supp_rates & BIT(i)) &&
-		    (mode->rates[i].flags & IEEE80211_RATE_SUPPORTED)) {
-			sta->txrate = i;
-			break;
-		}
-	}
+	sta->txrate_idx = rate_lowest_index(local, sband, sta);
 }
 
 
diff --git a/net/mac80211/regdomain.c b/net/mac80211/regdomain.c
deleted file mode 100644
index f42678fa62d1..000000000000
--- a/net/mac80211/regdomain.c
+++ /dev/null
@@ -1,152 +0,0 @@
-/*
- * Copyright 2002-2005, Instant802 Networks, Inc.
- * Copyright 2005-2006, Devicescape Software, Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-/*
- * This regulatory domain control implementation is known to be incomplete
- * and confusing. mac80211 regulatory domain control will be significantly
- * reworked in the not-too-distant future.
- *
- * For now, drivers wishing to control which channels are and aren't available
- * are advised as follows:
- *  - set the IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED flag
- *  - continue to include *ALL* possible channels in the modes registered
- *    through ieee80211_register_hwmode()
- *  - for each allowable ieee80211_channel structure registered in the above
- *    call, set the flag member to some meaningful value such as
- *    IEEE80211_CHAN_W_SCAN | IEEE80211_CHAN_W_ACTIVE_SCAN |
- *    IEEE80211_CHAN_W_IBSS.
- *  - leave flag as 0 for non-allowable channels
- *
- * The usual implementation is for a driver to read a device EEPROM to
- * determine which regulatory domain it should be operating under, then
- * looking up the allowable channels in a driver-local table, then performing
- * the above.
- */
-
-#include <linux/module.h>
-#include <linux/netdevice.h>
-#include <net/mac80211.h>
-#include "ieee80211_i.h"
-
-static int ieee80211_regdom = 0x10; /* FCC */
-module_param(ieee80211_regdom, int, 0444);
-MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain; 64=MKK");
-
-/*
- * If firmware is upgraded by the vendor, additional channels can be used based
- * on the new Japanese regulatory rules. This is indicated by setting
- * ieee80211_japan_5ghz module parameter to one when loading the 80211 kernel
- * module.
- */
-static int ieee80211_japan_5ghz /* = 0 */;
-module_param(ieee80211_japan_5ghz, int, 0444);
-MODULE_PARM_DESC(ieee80211_japan_5ghz, "Vendor-updated firmware for 5 GHz");
-
-
-struct ieee80211_channel_range {
-	short start_freq;
-	short end_freq;
-	unsigned char power_level;
-	unsigned char antenna_max;
-};
-
-static const struct ieee80211_channel_range ieee80211_fcc_channels[] = {
-	{ 2412, 2462, 27, 6 } /* IEEE 802.11b/g, channels 1..11 */,
-	{ 5180, 5240, 17, 6 } /* IEEE 802.11a, channels 36..48 */,
-	{ 5260, 5320, 23, 6 } /* IEEE 802.11a, channels 52..64 */,
-	{ 5745, 5825, 30, 6 } /* IEEE 802.11a, channels 149..165, outdoor */,
-	{ 0 }
-};
-
-static const struct ieee80211_channel_range ieee80211_mkk_channels[] = {
-	{ 2412, 2472, 20, 6 } /* IEEE 802.11b/g, channels 1..13 */,
-	{ 5170, 5240, 20, 6 } /* IEEE 802.11a, channels 34..48 */,
-	{ 5260, 5320, 20, 6 } /* IEEE 802.11a, channels 52..64 */,
-	{ 0 }
-};
-
-
-static const struct ieee80211_channel_range *channel_range =
-	ieee80211_fcc_channels;
-
-
-static void ieee80211_unmask_channel(int mode, struct ieee80211_channel *chan)
-{
-	int i;
-
-	chan->flag = 0;
-
-	for (i = 0; channel_range[i].start_freq; i++) {
-		const struct ieee80211_channel_range *r = &channel_range[i];
-		if (r->start_freq <= chan->freq && r->end_freq >= chan->freq) {
-			if (ieee80211_regdom == 64 && !ieee80211_japan_5ghz &&
-			    chan->freq >= 5260 && chan->freq <= 5320) {
-				/*
-				 * Skip new channels in Japan since the
-				 * firmware was not marked having been upgraded
-				 * by the vendor.
-				 */
-				continue;
-			}
-
-			if (ieee80211_regdom == 0x10 &&
-			    (chan->freq == 5190 || chan->freq == 5210 ||
-			     chan->freq == 5230)) {
-				    /* Skip MKK channels when in FCC domain. */
-				    continue;
-			}
-
-			chan->flag |= IEEE80211_CHAN_W_SCAN |
-				IEEE80211_CHAN_W_ACTIVE_SCAN |
-				IEEE80211_CHAN_W_IBSS;
-			chan->power_level = r->power_level;
-			chan->antenna_max = r->antenna_max;
-
-			if (ieee80211_regdom == 64 &&
-			    (chan->freq == 5170 || chan->freq == 5190 ||
-			     chan->freq == 5210 || chan->freq == 5230)) {
-				/*
-				 * New regulatory rules in Japan have backwards
-				 * compatibility with old channels in 5.15-5.25
-				 * GHz band, but the station is not allowed to
-				 * use active scan on these old channels.
-				 */
-				chan->flag &= ~IEEE80211_CHAN_W_ACTIVE_SCAN;
-			}
-
-			if (ieee80211_regdom == 64 &&
-			    (chan->freq == 5260 || chan->freq == 5280 ||
-			     chan->freq == 5300 || chan->freq == 5320)) {
-				/*
-				 * IBSS is not allowed on 5.25-5.35 GHz band
-				 * due to radar detection requirements.
-				 */
-				chan->flag &= ~IEEE80211_CHAN_W_IBSS;
-			}
-
-			break;
-		}
-	}
-}
-
-
-void ieee80211_set_default_regdomain(struct ieee80211_hw_mode *mode)
-{
-	int c;
-	for (c = 0; c < mode->num_channels; c++)
-		ieee80211_unmask_channel(mode->mode, &mode->channels[c]);
-}
-
-
-void ieee80211_regdomain_init(void)
-{
-	if (ieee80211_regdom == 0x40)
-		channel_range = ieee80211_mkk_channels;
-}
-
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 3aae8e9e4e0a..c9ff98a93211 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -82,10 +82,10 @@ static inline int should_drop_frame(struct ieee80211_rx_status *status,
  */
 static struct sk_buff *
 ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
-		     struct ieee80211_rx_status *status)
+		     struct ieee80211_rx_status *status,
+		     struct ieee80211_rate *rate)
 {
 	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_rate *rate;
 	int needed_headroom = 0;
 	struct ieee80211_radiotap_header *rthdr;
 	__le64 *rttsft = NULL;
@@ -194,14 +194,11 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 			rtfixed->rx_flags |=
 				cpu_to_le16(IEEE80211_RADIOTAP_F_RX_BADFCS);
 
-		rate = ieee80211_get_rate(local, status->phymode,
-					  status->rate);
-		if (rate)
-			rtfixed->rate = rate->rate / 5;
+		rtfixed->rate = rate->bitrate / 5;
 
 		rtfixed->chan_freq = cpu_to_le16(status->freq);
 
-		if (status->phymode == MODE_IEEE80211A)
+		if (status->band == IEEE80211_BAND_5GHZ)
 			rtfixed->chan_flags =
 				cpu_to_le16(IEEE80211_CHAN_OFDM |
 					    IEEE80211_CHAN_5GHZ);
@@ -320,34 +317,21 @@ static void ieee80211_verify_ip_alignment(struct ieee80211_txrx_data *rx)
 
 
 static u32 ieee80211_rx_load_stats(struct ieee80211_local *local,
-			      struct sk_buff *skb,
-			      struct ieee80211_rx_status *status)
+				   struct sk_buff *skb,
+				   struct ieee80211_rx_status *status,
+				   struct ieee80211_rate *rate)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	u32 load = 0, hdrtime;
-	struct ieee80211_rate *rate;
-	struct ieee80211_hw_mode *mode = local->hw.conf.mode;
-	int i;
 
 	/* Estimate total channel use caused by this frame */
 
-	if (unlikely(mode->num_rates < 0))
-		return TXRX_CONTINUE;
-
-	rate = &mode->rates[0];
-	for (i = 0; i < mode->num_rates; i++) {
-		if (mode->rates[i].val == status->rate) {
-			rate = &mode->rates[i];
-			break;
-		}
-	}
-
 	/* 1 bit at 1 Mbit/s takes 1 usec; in channel_use values,
 	 * 1 usec = 1/8 * (1080 / 10) = 13.5 */
 
-	if (mode->mode == MODE_IEEE80211A ||
-	    (mode->mode == MODE_IEEE80211G &&
-	     rate->flags & IEEE80211_RATE_ERP))
+	if (status->band == IEEE80211_BAND_5GHZ ||
+	    (status->band == IEEE80211_BAND_5GHZ &&
+	     rate->flags & IEEE80211_RATE_ERP_G))
 		hdrtime = CHAN_UTIL_HDR_SHORT;
 	else
 		hdrtime = CHAN_UTIL_HDR_LONG;
@@ -356,7 +340,8 @@ static u32 ieee80211_rx_load_stats(struct ieee80211_local *local,
 	if (!is_multicast_ether_addr(hdr->addr1))
 		load += hdrtime;
 
-	load += skb->len * rate->rate_inv;
+	/* TODO: optimise again */
+	load += skb->len * CHAN_UTIL_RATE_LCM / rate->bitrate;
 
 	/* Divide channel_use by 8 to avoid wrapping around the counter */
 	load >>= CHAN_UTIL_SHIFT;
@@ -1685,7 +1670,8 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata,
 static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw,
 					 struct sk_buff *skb,
 					 struct ieee80211_rx_status *status,
-					 u32 load)
+					 u32 load,
+					 struct ieee80211_rate *rate)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_sub_if_data *sdata;
@@ -1705,6 +1691,7 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw,
 
 	rx.u.rx.status = status;
 	rx.u.rx.load = load;
+	rx.u.rx.rate = rate;
 	rx.fc = le16_to_cpu(hdr->frame_control);
 	type = rx.fc & IEEE80211_FCTL_FTYPE;
 
@@ -1837,6 +1824,8 @@ u8 ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw,
 	u16 head_seq_num, buf_size;
 	int index;
 	u32 pkt_load;
+	struct ieee80211_supported_band *sband;
+	struct ieee80211_rate *rate;
 
 	buf_size = tid_agg_rx->buf_size;
 	head_seq_num = tid_agg_rx->head_seq_num;
@@ -1867,12 +1856,14 @@ u8 ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw,
 				memcpy(&status,
 					tid_agg_rx->reorder_buf[index]->cb,
 					sizeof(status));
+				sband = local->hw.wiphy->bands[status.band];
+				rate = &sband->bitrates[status.rate_idx];
 				pkt_load = ieee80211_rx_load_stats(local,
 						tid_agg_rx->reorder_buf[index],
-						&status);
+						&status, rate);
 				__ieee80211_rx_handle_packet(hw,
 					tid_agg_rx->reorder_buf[index],
-					&status, pkt_load);
+					&status, pkt_load, rate);
 				tid_agg_rx->stored_mpdu_num--;
 				tid_agg_rx->reorder_buf[index] = NULL;
 			}
@@ -1912,11 +1903,13 @@ u8 ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw,
 		/* release the reordered frame back to stack */
 		memcpy(&status, tid_agg_rx->reorder_buf[index]->cb,
 			sizeof(status));
+		sband = local->hw.wiphy->bands[status.band];
+		rate = &sband->bitrates[status.rate_idx];
 		pkt_load = ieee80211_rx_load_stats(local,
 					tid_agg_rx->reorder_buf[index],
-					&status);
+					&status, rate);
 		__ieee80211_rx_handle_packet(hw, tid_agg_rx->reorder_buf[index],
-						&status, pkt_load);
+					     &status, pkt_load, rate);
 		tid_agg_rx->stored_mpdu_num--;
 		tid_agg_rx->reorder_buf[index] = NULL;
 		tid_agg_rx->head_seq_num = seq_inc(tid_agg_rx->head_seq_num);
@@ -1997,6 +1990,25 @@ void __ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb,
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	u32 pkt_load;
+	struct ieee80211_rate *rate = NULL;
+	struct ieee80211_supported_band *sband;
+
+	if (status->band < 0 ||
+	    status->band > IEEE80211_NUM_BANDS) {
+		WARN_ON(1);
+		return;
+	}
+
+	sband = local->hw.wiphy->bands[status->band];
+
+	if (!sband ||
+	    status->rate_idx < 0 ||
+	    status->rate_idx >= sband->n_bitrates) {
+		WARN_ON(1);
+		return;
+	}
+
+	rate = &sband->bitrates[status->rate_idx];
 
 	/*
 	 * key references and virtual interfaces are protected using RCU
@@ -2011,17 +2023,17 @@ void __ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb,
 	 * if it was previously present.
 	 * Also, frames with less than 16 bytes are dropped.
 	 */
-	skb = ieee80211_rx_monitor(local, skb, status);
+	skb = ieee80211_rx_monitor(local, skb, status, rate);
 	if (!skb) {
 		rcu_read_unlock();
 		return;
 	}
 
-	pkt_load = ieee80211_rx_load_stats(local, skb, status);
+	pkt_load = ieee80211_rx_load_stats(local, skb, status, rate);
 	local->channel_use_raw += pkt_load;
 
 	if (!ieee80211_rx_reorder_ampdu(local, skb))
-		__ieee80211_rx_handle_packet(hw, skb, status, pkt_load);
+		__ieee80211_rx_handle_packet(hw, skb, status, pkt_load, rate);
 
 	rcu_read_unlock();
 }
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index ddc1f47194dd..746bbdea6b4c 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -74,30 +74,6 @@ struct sta_info *sta_info_get(struct ieee80211_local *local, u8 *addr)
 }
 EXPORT_SYMBOL(sta_info_get);
 
-int sta_info_min_txrate_get(struct ieee80211_local *local)
-{
-	struct sta_info *sta;
-	struct ieee80211_hw_mode *mode;
-	int min_txrate = 9999999;
-	int i;
-
-	read_lock_bh(&local->sta_lock);
-	mode = local->oper_hw_mode;
-	for (i = 0; i < STA_HASH_SIZE; i++) {
-		sta = local->sta_hash[i];
-		while (sta) {
-			if (sta->txrate < min_txrate)
-				min_txrate = sta->txrate;
-			sta = sta->hnext;
-		}
-	}
-	read_unlock_bh(&local->sta_lock);
-	if (min_txrate == 9999999)
-		min_txrate = 0;
-
-	return mode->rates[min_txrate].rate;
-}
-
 
 static void sta_info_release(struct kref *kref)
 {
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index 75573dc79d7a..3573743dfa59 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -133,10 +133,11 @@ struct sta_info {
 	unsigned int wep_weak_iv_count; /* number of RX frames with weak IV */
 
 	unsigned long last_rx;
-	u32 supp_rates; /* bitmap of supported rates in local->curr_rates */
-	int txrate; /* index in local->curr_rates */
-	int last_txrate; /* last rate used to send a frame to this STA */
-	int last_nonerp_idx;
+	/* bitmap of supported rates per band */
+	u64 supp_rates[IEEE80211_NUM_BANDS];
+	int txrate_idx;
+	/* last rates used to send a frame to this STA */
+	int last_txrate_idx, last_nonerp_txrate_idx;
 
 	struct net_device *dev; /* which net device is this station associated
 				 * to */
@@ -222,7 +223,6 @@ static inline void __sta_info_get(struct sta_info *sta)
 }
 
 struct sta_info * sta_info_get(struct ieee80211_local *local, u8 *addr);
-int sta_info_min_txrate_get(struct ieee80211_local *local);
 void sta_info_put(struct sta_info *sta);
 struct sta_info * sta_info_add(struct ieee80211_local *local,
 			       struct net_device *dev, u8 *addr, gfp_t gfp);
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 38e1b2bd8245..9e5359991985 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -92,9 +92,13 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr,
 	int rate, mrate, erp, dur, i;
 	struct ieee80211_rate *txrate = tx->u.tx.rate;
 	struct ieee80211_local *local = tx->local;
-	struct ieee80211_hw_mode *mode = tx->u.tx.mode;
+	struct ieee80211_supported_band *sband;
 
-	erp = txrate->flags & IEEE80211_RATE_ERP;
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
+
+	erp = 0;
+	if (tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
+		erp = txrate->flags & IEEE80211_RATE_ERP_G;
 
 	/*
 	 * data and mgmt (except PS Poll):
@@ -150,20 +154,36 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr,
 	 * Mandatory rates for IEEE 802.11g PHY: 1, 2, 5.5, 11, 6, 12, 24 Mbps
 	 */
 	rate = -1;
-	mrate = 10; /* use 1 Mbps if everything fails */
-	for (i = 0; i < mode->num_rates; i++) {
-		struct ieee80211_rate *r = &mode->rates[i];
-		if (r->rate > txrate->rate)
-			break;
+	/* use lowest available if everything fails */
+	mrate = sband->bitrates[0].bitrate;
+	for (i = 0; i < sband->n_bitrates; i++) {
+		struct ieee80211_rate *r = &sband->bitrates[i];
 
-		if (IEEE80211_RATE_MODULATION(txrate->flags) !=
-		    IEEE80211_RATE_MODULATION(r->flags))
-			continue;
+		if (r->bitrate > txrate->bitrate)
+			break;
 
-		if (r->flags & IEEE80211_RATE_BASIC)
-			rate = r->rate;
-		else if (r->flags & IEEE80211_RATE_MANDATORY)
-			mrate = r->rate;
+		if (tx->sdata->basic_rates & BIT(i))
+			rate = r->bitrate;
+
+		switch (sband->band) {
+		case IEEE80211_BAND_2GHZ: {
+			u32 flag;
+			if (tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
+				flag = IEEE80211_RATE_MANDATORY_G;
+			else
+				flag = IEEE80211_RATE_MANDATORY_B;
+			if (r->flags & flag)
+				mrate = r->bitrate;
+			break;
+		}
+		case IEEE80211_BAND_5GHZ:
+			if (r->flags & IEEE80211_RATE_MANDATORY_A)
+				mrate = r->bitrate;
+			break;
+		case IEEE80211_NUM_BANDS:
+			WARN_ON(1);
+			break;
+		}
 	}
 	if (rate == -1) {
 		/* No matching basic rate found; use highest suitable mandatory
@@ -184,7 +204,7 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr,
 		dur *= 2; /* ACK + SIFS */
 		/* next fragment */
 		dur += ieee80211_frame_duration(local, next_frag_len,
-				txrate->rate, erp,
+				txrate->bitrate, erp,
 				tx->sdata->bss_conf.use_short_preamble);
 	}
 
@@ -585,26 +605,28 @@ static ieee80211_txrx_result
 ieee80211_tx_h_rate_ctrl(struct ieee80211_txrx_data *tx)
 {
 	struct rate_selection rsel;
+	struct ieee80211_supported_band *sband;
+
+	sband = tx->local->hw.wiphy->bands[tx->local->hw.conf.channel->band];
 
 	if (likely(!tx->u.tx.rate)) {
-		rate_control_get_rate(tx->dev, tx->u.tx.mode, tx->skb, &rsel);
+		rate_control_get_rate(tx->dev, sband, tx->skb, &rsel);
 		tx->u.tx.rate = rsel.rate;
-		if (unlikely(rsel.probe != NULL)) {
+		if (unlikely(rsel.probe)) {
 			tx->u.tx.control->flags |=
 				IEEE80211_TXCTL_RATE_CTRL_PROBE;
 			tx->flags |= IEEE80211_TXRXD_TXPROBE_LAST_FRAG;
-			tx->u.tx.control->alt_retry_rate = tx->u.tx.rate->val;
+			tx->u.tx.control->alt_retry_rate = tx->u.tx.rate;
 			tx->u.tx.rate = rsel.probe;
 		} else
-			tx->u.tx.control->alt_retry_rate = -1;
+			tx->u.tx.control->alt_retry_rate = NULL;
 
 		if (!tx->u.tx.rate)
 			return TXRX_DROP;
 	} else
-		tx->u.tx.control->alt_retry_rate = -1;
+		tx->u.tx.control->alt_retry_rate = NULL;
 
-	if (tx->u.tx.mode->mode == MODE_IEEE80211G &&
-	    tx->sdata->bss_conf.use_cts_prot &&
+	if (tx->sdata->bss_conf.use_cts_prot &&
 	    (tx->flags & IEEE80211_TXRXD_FRAGMENTED) && rsel.nonerp) {
 		tx->u.tx.last_frag_rate = tx->u.tx.rate;
 		if (rsel.probe)
@@ -612,13 +634,13 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_txrx_data *tx)
 		else
 			tx->flags |= IEEE80211_TXRXD_TXPROBE_LAST_FRAG;
 		tx->u.tx.rate = rsel.nonerp;
-		tx->u.tx.control->rate = rsel.nonerp;
+		tx->u.tx.control->tx_rate = rsel.nonerp;
 		tx->u.tx.control->flags &= ~IEEE80211_TXCTL_RATE_CTRL_PROBE;
 	} else {
 		tx->u.tx.last_frag_rate = tx->u.tx.rate;
-		tx->u.tx.control->rate = tx->u.tx.rate;
+		tx->u.tx.control->tx_rate = tx->u.tx.rate;
 	}
-	tx->u.tx.control->tx_rate = tx->u.tx.rate->val;
+	tx->u.tx.control->tx_rate = tx->u.tx.rate;
 
 	return TXRX_CONTINUE;
 }
@@ -630,7 +652,6 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx)
 	u16 fc = le16_to_cpu(hdr->frame_control);
 	u16 dur;
 	struct ieee80211_tx_control *control = tx->u.tx.control;
-	struct ieee80211_hw_mode *mode = tx->u.tx.mode;
 
 	if (!control->retry_limit) {
 		if (!is_multicast_ether_addr(hdr->addr1)) {
@@ -657,14 +678,14 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx)
 		 * frames.
 		 * TODO: The last fragment could still use multiple retry
 		 * rates. */
-		control->alt_retry_rate = -1;
+		control->alt_retry_rate = NULL;
 	}
 
 	/* Use CTS protection for unicast frames sent using extended rates if
 	 * there are associated non-ERP stations and RTS/CTS is not configured
 	 * for the frame. */
-	if (mode->mode == MODE_IEEE80211G &&
-	    (tx->u.tx.rate->flags & IEEE80211_RATE_ERP) &&
+	if ((tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) &&
+	    (tx->u.tx.rate->flags & IEEE80211_RATE_ERP_G) &&
 	    (tx->flags & IEEE80211_TXRXD_TXUNICAST) &&
 	    tx->sdata->bss_conf.use_cts_prot &&
 	    !(control->flags & IEEE80211_TXCTL_USE_RTS_CTS))
@@ -674,10 +695,10 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx)
 	 * short preambles at the selected rate and short preambles are
 	 * available on the network at the current point in time. */
 	if (((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) &&
-	    (tx->u.tx.rate->flags & IEEE80211_RATE_PREAMBLE2) &&
+	    (tx->u.tx.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE) &&
 	    tx->sdata->bss_conf.use_short_preamble &&
 	    (!tx->sta || (tx->sta->flags & WLAN_STA_SHORT_PREAMBLE))) {
-		tx->u.tx.control->tx_rate = tx->u.tx.rate->val2;
+		tx->u.tx.control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
 	}
 
 	/* Setup duration field for the first fragment of the frame. Duration
@@ -690,19 +711,33 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx)
 
 	if ((control->flags & IEEE80211_TXCTL_USE_RTS_CTS) ||
 	    (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)) {
-		struct ieee80211_rate *rate;
+		struct ieee80211_supported_band *sband;
+		struct ieee80211_rate *rate, *baserate;
+		int idx;
+
+		sband = tx->local->hw.wiphy->bands[
+				tx->local->hw.conf.channel->band];
 
 		/* Do not use multiple retry rates when using RTS/CTS */
-		control->alt_retry_rate = -1;
+		control->alt_retry_rate = NULL;
 
 		/* Use min(data rate, max base rate) as CTS/RTS rate */
 		rate = tx->u.tx.rate;
-		while (rate > mode->rates &&
-		       !(rate->flags & IEEE80211_RATE_BASIC))
-			rate--;
+		baserate = NULL;
+
+		for (idx = 0; idx < sband->n_bitrates; idx++) {
+			if (sband->bitrates[idx].bitrate > rate->bitrate)
+				continue;
+			if (tx->sdata->basic_rates & BIT(idx) &&
+			    (!baserate ||
+			     (baserate->bitrate < sband->bitrates[idx].bitrate)))
+				baserate = &sband->bitrates[idx];
+		}
 
-		control->rts_cts_rate = rate->val;
-		control->rts_rate = rate;
+		if (baserate)
+			control->rts_cts_rate = baserate;
+		else
+			control->rts_cts_rate = &sband->bitrates[0];
 	}
 
 	if (tx->sta) {
@@ -726,10 +761,10 @@ static ieee80211_txrx_result
 ieee80211_tx_h_load_stats(struct ieee80211_txrx_data *tx)
 {
 	struct ieee80211_local *local = tx->local;
-	struct ieee80211_hw_mode *mode = tx->u.tx.mode;
 	struct sk_buff *skb = tx->skb;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	u32 load = 0, hdrtime;
+	struct ieee80211_rate *rate = tx->u.tx.rate;
 
 	/* TODO: this could be part of tx_status handling, so that the number
 	 * of retries would be known; TX rate should in that case be stored
@@ -740,9 +775,9 @@ ieee80211_tx_h_load_stats(struct ieee80211_txrx_data *tx)
 	/* 1 bit at 1 Mbit/s takes 1 usec; in channel_use values,
 	 * 1 usec = 1/8 * (1080 / 10) = 13.5 */
 
-	if (mode->mode == MODE_IEEE80211A ||
-	    (mode->mode == MODE_IEEE80211G &&
-	     tx->u.tx.rate->flags & IEEE80211_RATE_ERP))
+	if (tx->u.tx.channel->band == IEEE80211_BAND_5GHZ ||
+	    (tx->u.tx.channel->band == IEEE80211_BAND_2GHZ &&
+	     rate->flags & IEEE80211_RATE_ERP_G))
 		hdrtime = CHAN_UTIL_HDR_SHORT;
 	else
 		hdrtime = CHAN_UTIL_HDR_LONG;
@@ -756,14 +791,15 @@ ieee80211_tx_h_load_stats(struct ieee80211_txrx_data *tx)
 	else if (tx->u.tx.control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)
 		load += hdrtime;
 
-	load += skb->len * tx->u.tx.rate->rate_inv;
+	/* TODO: optimise again */
+	load += skb->len * CHAN_UTIL_RATE_LCM / rate->bitrate;
 
 	if (tx->u.tx.extra_frag) {
 		int i;
 		for (i = 0; i < tx->u.tx.num_extra_frag; i++) {
 			load += 2 * hdrtime;
 			load += tx->u.tx.extra_frag[i]->len *
-				tx->u.tx.rate->rate;
+				tx->u.tx.rate->bitrate;
 		}
 	}
 
@@ -816,10 +852,12 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_txrx_data *tx,
 	struct ieee80211_radiotap_iterator iterator;
 	struct ieee80211_radiotap_header *rthdr =
 		(struct ieee80211_radiotap_header *) skb->data;
-	struct ieee80211_hw_mode *mode = tx->local->hw.conf.mode;
+	struct ieee80211_supported_band *sband;
 	int ret = ieee80211_radiotap_iterator_init(&iterator, rthdr, skb->len);
 	struct ieee80211_tx_control *control = tx->u.tx.control;
 
+	sband = tx->local->hw.wiphy->bands[tx->local->hw.conf.channel->band];
+
 	control->flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT;
 	tx->flags |= IEEE80211_TXRXD_TX_INJECTED;
 	tx->flags &= ~IEEE80211_TXRXD_FRAGMENTED;
@@ -852,10 +890,12 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_txrx_data *tx,
 			 * ieee80211 rate int is in 100kbps units eg, 0x0a=1Mbps
 			 */
 			target_rate = (*iterator.this_arg) * 5;
-			for (i = 0; i < mode->num_rates; i++) {
-				struct ieee80211_rate *r = &mode->rates[i];
+			for (i = 0; i < sband->n_bitrates; i++) {
+				struct ieee80211_rate *r;
 
-				if (r->rate == target_rate) {
+				r = &sband->bitrates[i];
+
+				if (r->bitrate == target_rate) {
 					tx->u.tx.rate = r;
 					break;
 				}
@@ -870,9 +910,11 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_txrx_data *tx,
 			control->antenna_sel_tx = (*iterator.this_arg) + 1;
 			break;
 
+#if 0
 		case IEEE80211_RADIOTAP_DBM_TX_POWER:
 			control->power_level = *iterator.this_arg;
 			break;
+#endif
 
 		case IEEE80211_RADIOTAP_FLAGS:
 			if (*iterator.this_arg & IEEE80211_RADIOTAP_F_FCS) {
@@ -1054,8 +1096,8 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 			if (__ieee80211_queue_stopped(local, control->queue))
 				return IEEE80211_TX_FRAG_AGAIN;
 			if (i == tx->u.tx.num_extra_frag) {
-				control->tx_rate = tx->u.tx.last_frag_hwrate;
-				control->rate = tx->u.tx.last_frag_rate;
+				control->tx_rate = tx->u.tx.last_frag_rate;
+
 				if (tx->flags & IEEE80211_TXRXD_TXPROBE_LAST_FRAG)
 					control->flags |=
 						IEEE80211_TXCTL_RATE_CTRL_PROBE;
@@ -1114,7 +1156,7 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
 	rcu_read_lock();
 
 	sta = tx.sta;
-	tx.u.tx.mode = local->hw.conf.mode;
+	tx.u.tx.channel = local->hw.conf.channel;
 
 	for (handler = local->tx_handlers; *handler != NULL;
 	     handler++) {
@@ -1151,7 +1193,6 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb,
 			} else {
 				next_len = 0;
 				tx.u.tx.rate = tx.u.tx.last_frag_rate;
-				tx.u.tx.last_frag_hwrate = tx.u.tx.rate->val;
 			}
 			dur = ieee80211_duration(&tx, 0, next_len);
 			hdr->duration_id = cpu_to_le16(dur);
@@ -1188,7 +1229,6 @@ retry:
 		store->skb = skb;
 		store->extra_frag = tx.u.tx.extra_frag;
 		store->num_extra_frag = tx.u.tx.num_extra_frag;
-		store->last_frag_hwrate = tx.u.tx.last_frag_hwrate;
 		store->last_frag_rate = tx.u.tx.last_frag_rate;
 		store->last_frag_rate_ctrl_probe =
 			!!(tx.flags & IEEE80211_TXRXD_TXPROBE_LAST_FRAG);
@@ -1609,7 +1649,6 @@ void ieee80211_tx_pending(unsigned long data)
 		tx.u.tx.control = &store->control;
 		tx.u.tx.extra_frag = store->extra_frag;
 		tx.u.tx.num_extra_frag = store->num_extra_frag;
-		tx.u.tx.last_frag_hwrate = store->last_frag_hwrate;
 		tx.u.tx.last_frag_rate = store->last_frag_rate;
 		tx.flags = 0;
 		if (store->last_frag_rate_ctrl_probe)
@@ -1712,6 +1751,9 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 	struct ieee80211_if_ap *ap = NULL;
 	struct rate_selection rsel;
 	struct beacon_data *beacon;
+	struct ieee80211_supported_band *sband;
+
+	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
 	rcu_read_lock();
 
@@ -1750,8 +1792,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 		       beacon->tail_len);
 
 	if (control) {
-		rate_control_get_rate(local->mdev, local->oper_hw_mode, skb,
-				      &rsel);
+		rate_control_get_rate(local->mdev, sband, skb, &rsel);
 		if (!rsel.rate) {
 			if (net_ratelimit()) {
 				printk(KERN_DEBUG "%s: ieee80211_beacon_get: "
@@ -1764,12 +1805,11 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 		}
 
 		control->vif = vif;
-		control->tx_rate =
-			(sdata->bss_conf.use_short_preamble &&
-			(rsel.rate->flags & IEEE80211_RATE_PREAMBLE2)) ?
-			rsel.rate->val2 : rsel.rate->val;
+		control->tx_rate = rsel.rate;
+		if (sdata->bss_conf.use_short_preamble &&
+		    rsel.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE)
+			control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE;
 		control->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
-		control->power_level = local->hw.conf.power_level;
 		control->flags |= IEEE80211_TXCTL_NO_ACK;
 		control->retry_limit = 1;
 		control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK;
@@ -1874,7 +1914,7 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw,
 	}
 	sta = tx.sta;
 	tx.flags |= IEEE80211_TXRXD_TXPS_BUFFERED;
-	tx.u.tx.mode = local->hw.conf.mode;
+	tx.u.tx.channel = local->hw.conf.channel;
 
 	for (handler = local->tx_handlers; *handler != NULL; handler++) {
 		res = (*handler)(&tx);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 5e631ce98d7e..f64804fed0a9 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -41,92 +41,6 @@ const unsigned char bridge_tunnel_header[] =
 	{ 0xaa, 0xaa, 0x03, 0x00, 0x00, 0xf8 };
 
 
-static int rate_list_match(const int *rate_list, int rate)
-{
-	int i;
-
-	if (!rate_list)
-		return 0;
-
-	for (i = 0; rate_list[i] >= 0; i++)
-		if (rate_list[i] == rate)
-			return 1;
-
-	return 0;
-}
-
-void ieee80211_prepare_rates(struct ieee80211_local *local,
-			     struct ieee80211_hw_mode *mode)
-{
-	int i;
-
-	for (i = 0; i < mode->num_rates; i++) {
-		struct ieee80211_rate *rate = &mode->rates[i];
-
-		rate->flags &= ~(IEEE80211_RATE_SUPPORTED |
-				 IEEE80211_RATE_BASIC);
-
-		if (local->supp_rates[mode->mode]) {
-			if (!rate_list_match(local->supp_rates[mode->mode],
-					     rate->rate))
-				continue;
-		}
-
-		rate->flags |= IEEE80211_RATE_SUPPORTED;
-
-		/* Use configured basic rate set if it is available. If not,
-		 * use defaults that are sane for most cases. */
-		if (local->basic_rates[mode->mode]) {
-			if (rate_list_match(local->basic_rates[mode->mode],
-					    rate->rate))
-				rate->flags |= IEEE80211_RATE_BASIC;
-		} else switch (mode->mode) {
-		case MODE_IEEE80211A:
-			if (rate->rate == 60 || rate->rate == 120 ||
-			    rate->rate == 240)
-				rate->flags |= IEEE80211_RATE_BASIC;
-			break;
-		case MODE_IEEE80211B:
-			if (rate->rate == 10 || rate->rate == 20)
-				rate->flags |= IEEE80211_RATE_BASIC;
-			break;
-		case MODE_IEEE80211G:
-			if (rate->rate == 10 || rate->rate == 20 ||
-			    rate->rate == 55 || rate->rate == 110)
-				rate->flags |= IEEE80211_RATE_BASIC;
-			break;
-		case NUM_IEEE80211_MODES:
-			/* not useful */
-			break;
-		}
-
-		/* Set ERP and MANDATORY flags based on phymode */
-		switch (mode->mode) {
-		case MODE_IEEE80211A:
-			if (rate->rate == 60 || rate->rate == 120 ||
-			    rate->rate == 240)
-				rate->flags |= IEEE80211_RATE_MANDATORY;
-			break;
-		case MODE_IEEE80211B:
-			if (rate->rate == 10)
-				rate->flags |= IEEE80211_RATE_MANDATORY;
-			break;
-		case MODE_IEEE80211G:
-			if (rate->rate == 10 || rate->rate == 20 ||
-			    rate->rate == 55 || rate->rate == 110 ||
-			    rate->rate == 60 || rate->rate == 120 ||
-			    rate->rate == 240)
-				rate->flags |= IEEE80211_RATE_MANDATORY;
-			break;
-		case NUM_IEEE80211_MODES:
-			/* not useful */
-			break;
-		}
-		if (ieee80211_is_erp_rate(mode->mode, rate->rate))
-			rate->flags |= IEEE80211_RATE_ERP;
-	}
-}
-
 u8 *ieee80211_get_bssid(struct ieee80211_hdr *hdr, size_t len,
 			enum ieee80211_if_types type)
 {
@@ -262,7 +176,7 @@ int ieee80211_frame_duration(struct ieee80211_local *local, size_t len,
 	 * DIV_ROUND_UP() operations.
 	 */
 
-	if (local->hw.conf.phymode == MODE_IEEE80211A || erp) {
+	if (local->hw.conf.channel->band == IEEE80211_BAND_5GHZ || erp) {
 		/*
 		 * OFDM:
 		 *
@@ -304,15 +218,19 @@ int ieee80211_frame_duration(struct ieee80211_local *local, size_t len,
 /* Exported duration function for driver use */
 __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw,
 					struct ieee80211_vif *vif,
-					size_t frame_len, int rate)
+					size_t frame_len,
+					struct ieee80211_rate *rate)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
 	u16 dur;
 	int erp;
 
-	erp = ieee80211_is_erp_rate(hw->conf.phymode, rate);
-	dur = ieee80211_frame_duration(local, frame_len, rate, erp,
+	erp = 0;
+	if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
+		erp = rate->flags & IEEE80211_RATE_ERP_G;
+
+	dur = ieee80211_frame_duration(local, frame_len, rate->bitrate, erp,
 				       sdata->bss_conf.use_short_preamble);
 
 	return cpu_to_le16(dur);
@@ -332,17 +250,20 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw,
 
 	short_preamble = sdata->bss_conf.use_short_preamble;
 
-	rate = frame_txctl->rts_rate;
-	erp = !!(rate->flags & IEEE80211_RATE_ERP);
+	rate = frame_txctl->rts_cts_rate;
+
+	erp = 0;
+	if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
+		erp = rate->flags & IEEE80211_RATE_ERP_G;
 
 	/* CTS duration */
-	dur = ieee80211_frame_duration(local, 10, rate->rate,
+	dur = ieee80211_frame_duration(local, 10, rate->bitrate,
 				       erp, short_preamble);
 	/* Data frame duration */
-	dur += ieee80211_frame_duration(local, frame_len, rate->rate,
+	dur += ieee80211_frame_duration(local, frame_len, rate->bitrate,
 					erp, short_preamble);
 	/* ACK duration */
-	dur += ieee80211_frame_duration(local, 10, rate->rate,
+	dur += ieee80211_frame_duration(local, 10, rate->bitrate,
 					erp, short_preamble);
 
 	return cpu_to_le16(dur);
@@ -363,15 +284,17 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 
 	short_preamble = sdata->bss_conf.use_short_preamble;
 
-	rate = frame_txctl->rts_rate;
-	erp = !!(rate->flags & IEEE80211_RATE_ERP);
+	rate = frame_txctl->rts_cts_rate;
+	erp = 0;
+	if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)
+		erp = rate->flags & IEEE80211_RATE_ERP_G;
 
 	/* Data frame duration */
-	dur = ieee80211_frame_duration(local, frame_len, rate->rate,
+	dur = ieee80211_frame_duration(local, frame_len, rate->bitrate,
 				       erp, short_preamble);
 	if (!(frame_txctl->flags & IEEE80211_TXCTL_NO_ACK)) {
 		/* ACK duration */
-		dur += ieee80211_frame_duration(local, 10, rate->rate,
+		dur += ieee80211_frame_duration(local, 10, rate->bitrate,
 						erp, short_preamble);
 	}
 
@@ -379,27 +302,6 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw,
 }
 EXPORT_SYMBOL(ieee80211_ctstoself_duration);
 
-struct ieee80211_rate *
-ieee80211_get_rate(struct ieee80211_local *local, int phymode, int hw_rate)
-{
-	struct ieee80211_hw_mode *mode;
-	int r;
-
-	list_for_each_entry(mode, &local->modes_list, list) {
-		if (mode->mode != phymode)
-			continue;
-		for (r = 0; r < mode->num_rates; r++) {
-			struct ieee80211_rate *rate = &mode->rates[r];
-			if (rate->val == hw_rate ||
-			    (rate->flags & IEEE80211_RATE_PREAMBLE2 &&
-			     rate->val2 == hw_rate))
-				return rate;
-		}
-	}
-
-	return NULL;
-}
-
 void ieee80211_wake_queue(struct ieee80211_hw *hw, int queue)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
diff --git a/net/wireless/Makefile b/net/wireless/Makefile
index 65710a42e5a7..b9f943c45f3b 100644
--- a/net/wireless/Makefile
+++ b/net/wireless/Makefile
@@ -1,5 +1,5 @@
 obj-$(CONFIG_WIRELESS_EXT) += wext.o
 obj-$(CONFIG_CFG80211) += cfg80211.o
 
-cfg80211-y += core.o sysfs.o radiotap.o
+cfg80211-y += core.o sysfs.o radiotap.o util.o reg.o
 cfg80211-$(CONFIG_NL80211) += nl80211.o
diff --git a/net/wireless/core.c b/net/wireless/core.c
index cfc5fc5f9e75..80afacdae46c 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -232,6 +232,47 @@ int wiphy_register(struct wiphy *wiphy)
 {
 	struct cfg80211_registered_device *drv = wiphy_to_dev(wiphy);
 	int res;
+	enum ieee80211_band band;
+	struct ieee80211_supported_band *sband;
+	bool have_band = false;
+	int i;
+
+	/* sanity check supported bands/channels */
+	for (band = 0; band < IEEE80211_NUM_BANDS; band++) {
+		sband = wiphy->bands[band];
+		if (!sband)
+			continue;
+
+		sband->band = band;
+
+		if (!sband->n_channels || !sband->n_bitrates) {
+			WARN_ON(1);
+			return -EINVAL;
+		}
+
+		for (i = 0; i < sband->n_channels; i++) {
+			sband->channels[i].orig_flags =
+				sband->channels[i].flags;
+			sband->channels[i].orig_mag =
+				sband->channels[i].max_antenna_gain;
+			sband->channels[i].orig_mpwr =
+				sband->channels[i].max_power;
+			sband->channels[i].band = band;
+		}
+
+		have_band = true;
+	}
+
+	if (!have_band) {
+		WARN_ON(1);
+		return -EINVAL;
+	}
+
+	/* check and set up bitrates */
+	ieee80211_set_bitrate_flags(wiphy);
+
+	/* set up regulatory info */
+	wiphy_update_regulatory(wiphy);
 
 	mutex_lock(&cfg80211_drv_mutex);
 
diff --git a/net/wireless/core.h b/net/wireless/core.h
index eb0f846b40df..7a02c356d63d 100644
--- a/net/wireless/core.h
+++ b/net/wireless/core.h
@@ -78,4 +78,7 @@ extern void cfg80211_dev_free(struct cfg80211_registered_device *drv);
 extern int cfg80211_dev_rename(struct cfg80211_registered_device *drv,
 			       char *newname);
 
+void ieee80211_set_bitrate_flags(struct wiphy *wiphy);
+void wiphy_update_regulatory(struct wiphy *wiphy);
+
 #endif /* __NET_WIRELESS_CORE_H */
diff --git a/net/wireless/reg.c b/net/wireless/reg.c
new file mode 100644
index 000000000000..2b63c96dcf19
--- /dev/null
+++ b/net/wireless/reg.c
@@ -0,0 +1,153 @@
+/*
+ * Copyright 2002-2005, Instant802 Networks, Inc.
+ * Copyright 2005-2006, Devicescape Software, Inc.
+ * Copyright 2007	Johannes Berg <johannes@sipsolutions.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * This regulatory domain control implementation is highly incomplete, it
+ * only exists for the purpose of not regressing mac80211.
+ *
+ * For now, drivers can restrict the set of allowed channels by either
+ * not registering those channels or setting the IEEE80211_CHAN_DISABLED
+ * flag; that flag will only be *set* by this code, never *cleared.
+ *
+ * The usual implementation is for a driver to read a device EEPROM to
+ * determine which regulatory domain it should be operating under, then
+ * looking up the allowable channels in a driver-local table and finally
+ * registering those channels in the wiphy structure.
+ *
+ * Alternatively, drivers that trust the regulatory domain control here
+ * will register a complete set of capabilities and the control code
+ * will restrict the set by setting the IEEE80211_CHAN_* flags.
+ */
+#include <linux/kernel.h>
+#include <net/wireless.h>
+#include "core.h"
+
+static char *ieee80211_regdom = "US";
+module_param(ieee80211_regdom, charp, 0444);
+MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain code");
+
+struct ieee80211_channel_range {
+	short start_freq;
+	short end_freq;
+	int max_power;
+	int max_antenna_gain;
+	u32 flags;
+};
+
+struct ieee80211_regdomain {
+	const char *code;
+	const struct ieee80211_channel_range *ranges;
+	int n_ranges;
+};
+
+#define RANGE_PWR(_start, _end, _pwr, _ag, _flags)	\
+	{ _start, _end, _pwr, _ag, _flags }
+
+
+/*
+ * Ideally, in the future, these definitions will be loaded from a
+ * userspace table via some daemon.
+ */
+static const struct ieee80211_channel_range ieee80211_US_channels[] = {
+	/* IEEE 802.11b/g, channels 1..11 */
+	RANGE_PWR(2412, 2462, 27, 6, 0),
+	/* IEEE 802.11a, channels 52..64 */
+	RANGE_PWR(5260, 5320, 23, 6, 0),
+	/* IEEE 802.11a, channels 149..165, outdoor */
+	RANGE_PWR(5745, 5825, 30, 6, 0),
+};
+
+static const struct ieee80211_channel_range ieee80211_JP_channels[] = {
+	/* IEEE 802.11b/g, channels 1..14 */
+	RANGE_PWR(2412, 2484, 20, 6, 0),
+	/* IEEE 802.11a, channels 34..48 */
+	RANGE_PWR(5170, 5240, 20, 6, IEEE80211_CHAN_PASSIVE_SCAN),
+	/* IEEE 802.11a, channels 52..64 */
+	RANGE_PWR(5260, 5320, 20, 6, IEEE80211_CHAN_NO_IBSS |
+				     IEEE80211_CHAN_RADAR),
+};
+
+#define REGDOM(_code)							\
+	{								\
+		.code = __stringify(_code),				\
+		.ranges = ieee80211_ ##_code## _channels,		\
+		.n_ranges = ARRAY_SIZE(ieee80211_ ##_code## _channels),	\
+	}
+
+static const struct ieee80211_regdomain ieee80211_regdoms[] = {
+	REGDOM(US),
+	REGDOM(JP),
+};
+
+
+static const struct ieee80211_regdomain *get_regdom(void)
+{
+	static const struct ieee80211_channel_range
+	ieee80211_world_channels[] = {
+		/* IEEE 802.11b/g, channels 1..11 */
+		RANGE_PWR(2412, 2462, 27, 6, 0),
+	};
+	static const struct ieee80211_regdomain regdom_world = REGDOM(world);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(ieee80211_regdoms); i++)
+		if (strcmp(ieee80211_regdom, ieee80211_regdoms[i].code) == 0)
+			return &ieee80211_regdoms[i];
+
+	return &regdom_world;
+}
+
+
+static void handle_channel(struct ieee80211_channel *chan,
+			   const struct ieee80211_regdomain *rd)
+{
+	int i;
+	u32 flags = chan->orig_flags;
+	const struct ieee80211_channel_range *rg = NULL;
+
+	for (i = 0; i < rd->n_ranges; i++) {
+		if (rd->ranges[i].start_freq <= chan->center_freq &&
+		    chan->center_freq <= rd->ranges[i].end_freq) {
+			rg = &rd->ranges[i];
+			break;
+		}
+	}
+
+	if (!rg) {
+		/* not found */
+		flags |= IEEE80211_CHAN_DISABLED;
+		chan->flags = flags;
+		return;
+	}
+
+	chan->flags = flags;
+	chan->max_antenna_gain = min(chan->orig_mag,
+					 rg->max_antenna_gain);
+	chan->max_power = min(chan->orig_mpwr, rg->max_power);
+}
+
+static void handle_band(struct ieee80211_supported_band *sband,
+			const struct ieee80211_regdomain *rd)
+{
+	int i;
+
+	for (i = 0; i < sband->n_channels; i++)
+		handle_channel(&sband->channels[i], rd);
+}
+
+void wiphy_update_regulatory(struct wiphy *wiphy)
+{
+	enum ieee80211_band band;
+	const struct ieee80211_regdomain *rd = get_regdom();
+
+	for (band = 0; band < IEEE80211_NUM_BANDS; band++)
+		if (wiphy->bands[band])
+			handle_band(wiphy->bands[band], rd);
+}
diff --git a/net/wireless/util.c b/net/wireless/util.c
new file mode 100644
index 000000000000..0dcccbf3eb57
--- /dev/null
+++ b/net/wireless/util.c
@@ -0,0 +1,98 @@
+/*
+ * Wireless utility functions
+ *
+ * Copyright 2007	Johannes Berg <johannes@sipsolutions.net>
+ */
+#include <net/wireless.h>
+#include <asm/bitops.h>
+#include "core.h"
+
+int ieee80211_channel_to_frequency(int chan)
+{
+	if (chan < 14)
+		return 2407 + chan * 5;
+
+	if (chan == 14)
+		return 2484;
+
+	/* FIXME: 802.11j 17.3.8.3.2 */
+	return (chan + 1000) * 5;
+}
+EXPORT_SYMBOL(ieee80211_channel_to_frequency);
+
+int ieee80211_frequency_to_channel(int freq)
+{
+	if (freq == 2484)
+		return 14;
+
+	if (freq < 2484)
+		return (freq - 2407) / 5;
+
+	/* FIXME: 802.11j 17.3.8.3.2 */
+	return freq/5 - 1000;
+}
+EXPORT_SYMBOL(ieee80211_frequency_to_channel);
+
+static void set_mandatory_flags_band(struct ieee80211_supported_band *sband,
+				     enum ieee80211_band band)
+{
+	int i, want;
+
+	switch (band) {
+	case IEEE80211_BAND_5GHZ:
+		want = 3;
+		for (i = 0; i < sband->n_bitrates; i++) {
+			if (sband->bitrates[i].bitrate == 60 ||
+			    sband->bitrates[i].bitrate == 120 ||
+			    sband->bitrates[i].bitrate == 240) {
+				sband->bitrates[i].flags |=
+					IEEE80211_RATE_MANDATORY_A;
+				want--;
+			}
+		}
+		WARN_ON(want);
+		break;
+	case IEEE80211_BAND_2GHZ:
+		want = 7;
+		for (i = 0; i < sband->n_bitrates; i++) {
+			if (sband->bitrates[i].bitrate == 10) {
+				sband->bitrates[i].flags |=
+					IEEE80211_RATE_MANDATORY_B |
+					IEEE80211_RATE_MANDATORY_G;
+				want--;
+			}
+
+			if (sband->bitrates[i].bitrate == 20 ||
+			    sband->bitrates[i].bitrate == 55 ||
+			    sband->bitrates[i].bitrate == 110 ||
+			    sband->bitrates[i].bitrate == 60 ||
+			    sband->bitrates[i].bitrate == 120 ||
+			    sband->bitrates[i].bitrate == 240) {
+				sband->bitrates[i].flags |=
+					IEEE80211_RATE_MANDATORY_G;
+				want--;
+			}
+
+			if (sband->bitrates[i].bitrate == 10 ||
+			    sband->bitrates[i].bitrate == 20 ||
+			    sband->bitrates[i].bitrate == 55 ||
+			    sband->bitrates[i].bitrate == 110)
+				sband->bitrates[i].flags |=
+					IEEE80211_RATE_ERP_G;
+		}
+		WARN_ON(want != 0 && want != 6);
+		break;
+	case IEEE80211_NUM_BANDS:
+		WARN_ON(1);
+		break;
+	}
+}
+
+void ieee80211_set_bitrate_flags(struct wiphy *wiphy)
+{
+	enum ieee80211_band band;
+
+	for (band = 0; band < IEEE80211_NUM_BANDS; band++)
+		if (wiphy->bands[band])
+			set_mandatory_flags_band(wiphy->bands[band], band);
+}
-- 
cgit v1.2.3


From ee688b000d35f413f33561ec9c7d3355be561e2f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 24 Jan 2008 19:38:39 +0100
Subject: nl80211: export hardware bitrate/channel capabilities

This makes nl80211 export the hardware bitrate/channel capabilities
as registered in a wiphy.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 64 ++++++++++++++++++++++++++++++++++++++++++
 net/wireless/nl80211.c  | 74 +++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 138 insertions(+)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 9fecf902419c..63695060db9f 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -161,6 +161,9 @@ enum nl80211_commands {
  *	given for %NL80211_CMD_GET_STATION, nested attribute containing
  *	info as possible, see &enum nl80211_sta_stats.
  *
+ * @NL80211_ATTR_WIPHY_BANDS: Information about an operating bands,
+ *	consisting of a nested array.
+ *
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
  */
@@ -195,6 +198,8 @@ enum nl80211_attrs {
 	NL80211_ATTR_STA_VLAN,
 	NL80211_ATTR_STA_STATS,
 
+	NL80211_ATTR_WIPHY_BANDS,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
@@ -280,4 +285,63 @@ enum nl80211_sta_stats {
 	NL80211_STA_STAT_MAX = __NL80211_STA_STAT_AFTER_LAST - 1
 };
 
+/**
+ * enum nl80211_band_attr - band attributes
+ * @__NL80211_BAND_ATTR_INVALID: attribute number 0 is reserved
+ * @NL80211_BAND_ATTR_FREQS: supported frequencies in this band,
+ *	an array of nested frequency attributes
+ * @NL80211_BAND_ATTR_RATES: supported bitrates in this band,
+ *	an array of nested bitrate attributes
+ */
+enum nl80211_band_attr {
+	__NL80211_BAND_ATTR_INVALID,
+	NL80211_BAND_ATTR_FREQS,
+	NL80211_BAND_ATTR_RATES,
+
+	/* keep last */
+	__NL80211_BAND_ATTR_AFTER_LAST,
+	NL80211_BAND_ATTR_MAX = __NL80211_BAND_ATTR_AFTER_LAST - 1
+};
+
+/**
+ * enum nl80211_frequency_attr - frequency attributes
+ * @NL80211_FREQUENCY_ATTR_FREQ: Frequency in MHz
+ * @NL80211_FREQUENCY_ATTR_DISABLED: Channel is disabled in current
+ *	regulatory domain.
+ * @NL80211_FREQUENCY_ATTR_PASSIVE_SCAN: Only passive scanning is
+ *	permitted on this channel in current regulatory domain.
+ * @NL80211_FREQUENCY_ATTR_NO_IBSS: IBSS networks are not permitted
+ *	on this channel in current regulatory domain.
+ * @NL80211_FREQUENCY_ATTR_RADAR: Radar detection is mandatory
+ *	on this channel in current regulatory domain.
+ */
+enum nl80211_frequency_attr {
+	__NL80211_FREQUENCY_ATTR_INVALID,
+	NL80211_FREQUENCY_ATTR_FREQ,
+	NL80211_FREQUENCY_ATTR_DISABLED,
+	NL80211_FREQUENCY_ATTR_PASSIVE_SCAN,
+	NL80211_FREQUENCY_ATTR_NO_IBSS,
+	NL80211_FREQUENCY_ATTR_RADAR,
+
+	/* keep last */
+	__NL80211_FREQUENCY_ATTR_AFTER_LAST,
+	NL80211_FREQUENCY_ATTR_MAX = __NL80211_FREQUENCY_ATTR_AFTER_LAST - 1
+};
+
+/**
+ * enum nl80211_bitrate_attr - bitrate attributes
+ * @NL80211_BITRATE_ATTR_RATE: Bitrate in units of 100 kbps
+ * @NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE: Short preamble supported
+ *	in 2.4 GHz band.
+ */
+enum nl80211_bitrate_attr {
+	__NL80211_BITRATE_ATTR_INVALID,
+	NL80211_BITRATE_ATTR_RATE,
+	NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE,
+
+	/* keep last */
+	__NL80211_BITRATE_ATTR_AFTER_LAST,
+	NL80211_BITRATE_ATTR_MAX = __NL80211_BITRATE_ATTR_AFTER_LAST - 1
+};
+
 #endif /* __LINUX_NL80211_H */
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index e3a214f63f91..b123f58d3909 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -98,6 +98,13 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 			      struct cfg80211_registered_device *dev)
 {
 	void *hdr;
+	struct nlattr *nl_bands, *nl_band;
+	struct nlattr *nl_freqs, *nl_freq;
+	struct nlattr *nl_rates, *nl_rate;
+	enum ieee80211_band band;
+	struct ieee80211_channel *chan;
+	struct ieee80211_rate *rate;
+	int i;
 
 	hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_WIPHY);
 	if (!hdr)
@@ -105,6 +112,73 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 
 	NLA_PUT_U32(msg, NL80211_ATTR_WIPHY, dev->idx);
 	NLA_PUT_STRING(msg, NL80211_ATTR_WIPHY_NAME, wiphy_name(&dev->wiphy));
+
+	nl_bands = nla_nest_start(msg, NL80211_ATTR_WIPHY_BANDS);
+	if (!nl_bands)
+		goto nla_put_failure;
+
+	for (band = 0; band < IEEE80211_NUM_BANDS; band++) {
+		if (!dev->wiphy.bands[band])
+			continue;
+
+		nl_band = nla_nest_start(msg, band);
+		if (!nl_band)
+			goto nla_put_failure;
+
+		/* add frequencies */
+		nl_freqs = nla_nest_start(msg, NL80211_BAND_ATTR_FREQS);
+		if (!nl_freqs)
+			goto nla_put_failure;
+
+		for (i = 0; i < dev->wiphy.bands[band]->n_channels; i++) {
+			nl_freq = nla_nest_start(msg, i);
+			if (!nl_freq)
+				goto nla_put_failure;
+
+			chan = &dev->wiphy.bands[band]->channels[i];
+			NLA_PUT_U32(msg, NL80211_FREQUENCY_ATTR_FREQ,
+				    chan->center_freq);
+
+			if (chan->flags & IEEE80211_CHAN_DISABLED)
+				NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_DISABLED);
+			if (chan->flags & IEEE80211_CHAN_PASSIVE_SCAN)
+				NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_PASSIVE_SCAN);
+			if (chan->flags & IEEE80211_CHAN_NO_IBSS)
+				NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_NO_IBSS);
+			if (chan->flags & IEEE80211_CHAN_RADAR)
+				NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_RADAR);
+
+			nla_nest_end(msg, nl_freq);
+		}
+
+		nla_nest_end(msg, nl_freqs);
+
+		/* add bitrates */
+		nl_rates = nla_nest_start(msg, NL80211_BAND_ATTR_RATES);
+		if (!nl_rates)
+			goto nla_put_failure;
+
+		for (i = 0; i < dev->wiphy.bands[band]->n_bitrates; i++) {
+			nl_rate = nla_nest_start(msg, i);
+			if (!nl_rate)
+				goto nla_put_failure;
+
+			rate = &dev->wiphy.bands[band]->bitrates[i];
+			NLA_PUT_U32(msg, NL80211_BITRATE_ATTR_RATE,
+				    rate->bitrate);
+			if (rate->flags & IEEE80211_RATE_SHORT_PREAMBLE)
+				NLA_PUT_FLAG(msg,
+					NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE);
+
+			nla_nest_end(msg, nl_rate);
+		}
+
+		nla_nest_end(msg, nl_rates);
+
+		nla_nest_end(msg, nl_band);
+	}
+	nla_nest_end(msg, nl_bands);
+
 	return genlmsg_end(msg, hdr);
 
  nla_put_failure:
-- 
cgit v1.2.3


From 66f7ac50ed7cc5c19a62bc97e8f6e7891004a03a Mon Sep 17 00:00:00 2001
From: Michael Wu <flamingice@sourmilk.net>
Date: Thu, 31 Jan 2008 19:48:22 +0100
Subject: nl80211: Add monitor interface configuration flags

This allows precise control over what a monitor interface shows.

Signed-off-by: Michael Wu <flamingice@sourmilk.net>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 35 +++++++++++++++++++++++++++++++++++
 include/net/cfg80211.h  | 24 ++++++++++++++++++++++--
 net/mac80211/cfg.c      |  4 ++--
 net/wireless/nl80211.c  | 44 ++++++++++++++++++++++++++++++++++++++++++--
 4 files changed, 101 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 63695060db9f..a9f0b93324a2 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -164,6 +164,9 @@ enum nl80211_commands {
  * @NL80211_ATTR_WIPHY_BANDS: Information about an operating bands,
  *	consisting of a nested array.
  *
+ * @NL80211_ATTR_MNTR_FLAGS: flags, nested element with NLA_FLAG attributes of
+ *      &enum nl80211_mntr_flags.
+ *
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
  */
@@ -200,6 +203,8 @@ enum nl80211_attrs {
 
 	NL80211_ATTR_WIPHY_BANDS,
 
+	NL80211_ATTR_MNTR_FLAGS,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
@@ -344,4 +349,34 @@ enum nl80211_bitrate_attr {
 	NL80211_BITRATE_ATTR_MAX = __NL80211_BITRATE_ATTR_AFTER_LAST - 1
 };
 
+/**
+ * enum nl80211_mntr_flags - monitor configuration flags
+ *
+ * Monitor configuration flags.
+ *
+ * @__NL80211_MNTR_FLAG_INVALID: reserved
+ *
+ * @NL80211_MNTR_FLAG_FCSFAIL: pass frames with bad FCS
+ * @NL80211_MNTR_FLAG_PLCPFAIL: pass frames with bad PLCP
+ * @NL80211_MNTR_FLAG_CONTROL: pass control frames
+ * @NL80211_MNTR_FLAG_OTHER_BSS: disable BSSID filtering
+ * @NL80211_MNTR_FLAG_COOK_FRAMES: report frames after processing.
+ *	overrides all other flags.
+ *
+ * @__NL80211_MNTR_FLAG_AFTER_LAST: internal use
+ * @NL80211_MNTR_FLAG_MAX: highest possible monitor flag
+ */
+enum nl80211_mntr_flags {
+	__NL80211_MNTR_FLAG_INVALID,
+	NL80211_MNTR_FLAG_FCSFAIL,
+	NL80211_MNTR_FLAG_PLCPFAIL,
+	NL80211_MNTR_FLAG_CONTROL,
+	NL80211_MNTR_FLAG_OTHER_BSS,
+	NL80211_MNTR_FLAG_COOK_FRAMES,
+
+	/* keep last */
+	__NL80211_MNTR_FLAG_AFTER_LAST,
+	NL80211_MNTR_FLAG_MAX = __NL80211_MNTR_FLAG_AFTER_LAST - 1
+};
+
 #endif /* __LINUX_NL80211_H */
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index bcc480b8892a..ab4caf63954f 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -163,6 +163,26 @@ struct station_stats {
 	u32 tx_bytes;
 };
 
+/**
+ * enum monitor_flags - monitor flags
+ *
+ * Monitor interface configuration flags. Note that these must be the bits
+ * according to the nl80211 flags.
+ *
+ * @MONITOR_FLAG_FCSFAIL: pass frames with bad FCS
+ * @MONITOR_FLAG_PLCPFAIL: pass frames with bad PLCP
+ * @MONITOR_FLAG_CONTROL: pass control frames
+ * @MONITOR_FLAG_OTHER_BSS: disable BSSID filtering
+ * @MONITOR_FLAG_COOK_FRAMES: report frames after processing
+ */
+enum monitor_flags {
+	MONITOR_FLAG_FCSFAIL		= 1<<NL80211_MNTR_FLAG_FCSFAIL,
+	MONITOR_FLAG_PLCPFAIL		= 1<<NL80211_MNTR_FLAG_PLCPFAIL,
+	MONITOR_FLAG_CONTROL		= 1<<NL80211_MNTR_FLAG_CONTROL,
+	MONITOR_FLAG_OTHER_BSS		= 1<<NL80211_MNTR_FLAG_OTHER_BSS,
+	MONITOR_FLAG_COOK_FRAMES	= 1<<NL80211_MNTR_FLAG_COOK_FRAMES,
+};
+
 /* from net/wireless.h */
 struct wiphy;
 
@@ -213,10 +233,10 @@ struct wiphy;
  */
 struct cfg80211_ops {
 	int	(*add_virtual_intf)(struct wiphy *wiphy, char *name,
-				    enum nl80211_iftype type);
+				    enum nl80211_iftype type, u32 *flags);
 	int	(*del_virtual_intf)(struct wiphy *wiphy, int ifindex);
 	int	(*change_virtual_intf)(struct wiphy *wiphy, int ifindex,
-				       enum nl80211_iftype type);
+				       enum nl80211_iftype type, u32 *flags);
 
 	int	(*add_key)(struct wiphy *wiphy, struct net_device *netdev,
 			   u8 key_index, u8 *mac_addr,
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 15b8cf94f510..706418d477c2 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -34,7 +34,7 @@ nl80211_type_to_mac80211_type(enum nl80211_iftype type)
 }
 
 static int ieee80211_add_iface(struct wiphy *wiphy, char *name,
-			       enum nl80211_iftype type)
+			       enum nl80211_iftype type, u32 *flags)
 {
 	struct ieee80211_local *local = wiphy_priv(wiphy);
 	enum ieee80211_if_types itype;
@@ -69,7 +69,7 @@ static int ieee80211_del_iface(struct wiphy *wiphy, int ifindex)
 }
 
 static int ieee80211_change_iface(struct wiphy *wiphy, int ifindex,
-				  enum nl80211_iftype type)
+				  enum nl80211_iftype type, u32 *flags)
 {
 	struct ieee80211_local *local = wiphy_priv(wiphy);
 	struct net_device *dev;
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index b123f58d3909..5b3474798b8d 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -82,6 +82,7 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = {
 	[NL80211_ATTR_STA_SUPPORTED_RATES] = { .type = NLA_BINARY,
 					       .len = NL80211_MAX_SUPP_RATES },
 	[NL80211_ATTR_STA_VLAN] = { .type = NLA_U32 },
+	[NL80211_ATTR_MNTR_FLAGS] = { .type = NLA_NESTED },
 };
 
 /* message building helper */
@@ -336,12 +337,42 @@ static int nl80211_get_interface(struct sk_buff *skb, struct genl_info *info)
 	return -ENOBUFS;
 }
 
+static const struct nla_policy mntr_flags_policy[NL80211_MNTR_FLAG_MAX + 1] = {
+	[NL80211_MNTR_FLAG_FCSFAIL] = { .type = NLA_FLAG },
+	[NL80211_MNTR_FLAG_PLCPFAIL] = { .type = NLA_FLAG },
+	[NL80211_MNTR_FLAG_CONTROL] = { .type = NLA_FLAG },
+	[NL80211_MNTR_FLAG_OTHER_BSS] = { .type = NLA_FLAG },
+	[NL80211_MNTR_FLAG_COOK_FRAMES] = { .type = NLA_FLAG },
+};
+
+static int parse_monitor_flags(struct nlattr *nla, u32 *mntrflags)
+{
+	struct nlattr *flags[NL80211_MNTR_FLAG_MAX + 1];
+	int flag;
+
+	*mntrflags = 0;
+
+	if (!nla)
+		return -EINVAL;
+
+	if (nla_parse_nested(flags, NL80211_MNTR_FLAG_MAX,
+			     nla, mntr_flags_policy))
+		return -EINVAL;
+
+	for (flag = 1; flag <= NL80211_MNTR_FLAG_MAX; flag++)
+		if (flags[flag])
+			*mntrflags |= (1<<flag);
+
+	return 0;
+}
+
 static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *drv;
 	int err, ifindex;
 	enum nl80211_iftype type;
 	struct net_device *dev;
+	u32 flags;
 
 	if (info->attrs[NL80211_ATTR_IFTYPE]) {
 		type = nla_get_u32(info->attrs[NL80211_ATTR_IFTYPE]);
@@ -362,7 +393,11 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	rtnl_lock();
-	err = drv->ops->change_virtual_intf(&drv->wiphy, ifindex, type);
+	err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ?
+				  info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL,
+				  &flags);
+	err = drv->ops->change_virtual_intf(&drv->wiphy, ifindex,
+					    type, err ? NULL : &flags);
 	rtnl_unlock();
 
  unlock:
@@ -375,6 +410,7 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info)
 	struct cfg80211_registered_device *drv;
 	int err;
 	enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED;
+	u32 flags;
 
 	if (!info->attrs[NL80211_ATTR_IFNAME])
 		return -EINVAL;
@@ -395,8 +431,12 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	rtnl_lock();
+	err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ?
+				  info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL,
+				  &flags);
 	err = drv->ops->add_virtual_intf(&drv->wiphy,
-		nla_data(info->attrs[NL80211_ATTR_IFNAME]), type);
+		nla_data(info->attrs[NL80211_ATTR_IFNAME]),
+		type, err ? NULL : &flags);
 	rtnl_unlock();
 
  unlock:
-- 
cgit v1.2.3


From 3330d7be7008fa8e213648750fc13613eecc54bb Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sun, 10 Feb 2008 16:49:38 +0100
Subject: mac80211: give burst time in txop rather than 0.1msec units

This changes mac80211 to pass the burst time to conf_tx in txop
units rather than 0.1msec units. 0.1msec units are only required
by atheros hardware (according to current driver support), all
other drivers do other calculations or require the txop value.
Therefore, it results in fewer calculations and more precision
if we just pass the txop value through to the driver.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  2 +-
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  2 +-
 drivers/net/wireless/p54common.c            | 21 +++++++++++++--------
 include/net/mac80211.h                      | 12 ++++++------
 net/mac80211/ieee80211_sta.c                | 11 ++++++-----
 5 files changed, 27 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 4ad26f7613c5..bb3a5f633227 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -7442,7 +7442,7 @@ static int iwl3945_mac_conf_tx(struct ieee80211_hw *hw, int queue,
 	priv->qos_data.def_qos_parm.ac[q].cw_max = cpu_to_le16(params->cw_max);
 	priv->qos_data.def_qos_parm.ac[q].aifsn = params->aifs;
 	priv->qos_data.def_qos_parm.ac[q].edca_txop =
-			cpu_to_le16((params->burst_time * 100));
+			cpu_to_le16((params->txop * 32));
 
 	priv->qos_data.def_qos_parm.ac[q].reserved1 = 0;
 	priv->qos_data.qos_active = 1;
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index f8931993ab70..fb4c84e31932 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -7916,7 +7916,7 @@ static int iwl4965_mac_conf_tx(struct ieee80211_hw *hw, int queue,
 	priv->qos_data.def_qos_parm.ac[q].cw_max = cpu_to_le16(params->cw_max);
 	priv->qos_data.def_qos_parm.ac[q].aifsn = params->aifs;
 	priv->qos_data.def_qos_parm.ac[q].edca_txop =
-			cpu_to_le16((params->burst_time * 100));
+			cpu_to_le16((params->txop * 32));
 
 	priv->qos_data.def_qos_parm.ac[q].reserved1 = 0;
 	priv->qos_data.qos_active = 1;
diff --git a/drivers/net/wireless/p54common.c b/drivers/net/wireless/p54common.c
index 218ff7770ef6..0e9f887178ce 100644
--- a/drivers/net/wireless/p54common.c
+++ b/drivers/net/wireless/p54common.c
@@ -759,13 +759,12 @@ static int p54_set_leds(struct ieee80211_hw *dev, int mode, int link, int act)
 	return 0;
 }
 
-#define P54_SET_QUEUE(queue, ai_fs, cw_min, cw_max, burst)	\
+#define P54_SET_QUEUE(queue, ai_fs, cw_min, cw_max, _txop)	\
 do {	 							\
 	queue.aifs = cpu_to_le16(ai_fs);			\
 	queue.cwmin = cpu_to_le16(cw_min);			\
 	queue.cwmax = cpu_to_le16(cw_max);			\
-	queue.txop = (burst == 0) ? 				\
-		0 : cpu_to_le16((burst * 100) / 32 + 1);	\
+	queue.txop = cpu_to_le16(_txop);			\
 } while(0)
 
 static void p54_init_vdcf(struct ieee80211_hw *dev)
@@ -783,10 +782,16 @@ static void p54_init_vdcf(struct ieee80211_hw *dev)
 
 	vdcf = (struct p54_tx_control_vdcf *) hdr->data;
 
-	P54_SET_QUEUE(vdcf->queue[0], 0x0002, 0x0003, 0x0007, 0x000f);
-	P54_SET_QUEUE(vdcf->queue[1], 0x0002, 0x0007, 0x000f, 0x001e);
-	P54_SET_QUEUE(vdcf->queue[2], 0x0002, 0x000f, 0x03ff, 0x0014);
-	P54_SET_QUEUE(vdcf->queue[3], 0x0007, 0x000f, 0x03ff, 0x0000);
+	/*
+	 * FIXME: The default values in the spec (IEEE 802.11
+	 *	  7.3.2.19 Table 37) are 47, 94, 0, 0, why use
+	 *	  47, 94, 63, 0 here? Also, the default AIFS
+	 *	  values (second parameter) are 2, 2, 3, 7...
+	 */
+	P54_SET_QUEUE(vdcf->queue[0], 0x0002, 0x0003, 0x0007, 47);
+	P54_SET_QUEUE(vdcf->queue[1], 0x0002, 0x0007, 0x000f, 94);
+	P54_SET_QUEUE(vdcf->queue[2], 0x0002, 0x000f, 0x03ff, 63);
+	P54_SET_QUEUE(vdcf->queue[3], 0x0007, 0x000f, 0x03ff, 0);
 }
 
 static void p54_set_vdcf(struct ieee80211_hw *dev)
@@ -939,7 +944,7 @@ static int p54_conf_tx(struct ieee80211_hw *dev, int queue,
 
 	if ((params) && !((queue < 0) || (queue > 4))) {
 		P54_SET_QUEUE(vdcf->queue[queue], params->aifs,
-			params->cw_min, params->cw_max, params->burst_time);
+			params->cw_min, params->cw_max, params->txop);
 	} else
 		return -EINVAL;
 
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 460da54a0019..5ecf3cc8d977 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -89,19 +89,19 @@ struct ieee80211_ht_bss_info {
  * struct ieee80211_tx_queue_params - transmit queue configuration
  *
  * The information provided in this structure is required for QoS
- * transmit queue configuration.
+ * transmit queue configuration. Cf. IEEE 802.11 7.3.2.29.
  *
  * @aifs: arbitration interface space [0..255, -1: use default]
  * @cw_min: minimum contention window [will be a value of the form
  *	2^n-1 in the range 1..1023; 0: use default]
  * @cw_max: maximum contention window [like @cw_min]
- * @burst_time: maximum burst time in units of 0.1ms, 0 meaning disabled
+ * @txop: maximum burst time in units of 32 usecs, 0 meaning disabled
  */
 struct ieee80211_tx_queue_params {
-	int aifs;
-	int cw_min;
-	int cw_max;
-	int burst_time;
+	s16 aifs;
+	u16 cw_min;
+	u16 cw_max;
+	u16 txop;
 };
 
 /**
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index a8251a23917b..11f2ee6b2269 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -297,12 +297,13 @@ static void ieee80211_sta_wmm_params(struct net_device *dev,
 		params.aifs = pos[0] & 0x0f;
 		params.cw_max = ecw2cw((pos[1] & 0xf0) >> 4);
 		params.cw_min = ecw2cw(pos[1] & 0x0f);
-		/* TXOP is in units of 32 usec; burst_time in 0.1 ms */
-		params.burst_time = (pos[2] | (pos[3] << 8)) * 32 / 100;
+		params.txop = pos[2] | (pos[3] << 8);
+#ifdef CONFIG_MAC80211_DEBUG
 		printk(KERN_DEBUG "%s: WMM queue=%d aci=%d acm=%d aifs=%d "
-		       "cWmin=%d cWmax=%d burst=%d\n",
+		       "cWmin=%d cWmax=%d txop=%d\n",
 		       dev->name, queue, aci, acm, params.aifs, params.cw_min,
-		       params.cw_max, params.burst_time);
+		       params.cw_max, params.txop);
+#endif
 		/* TODO: handle ACM (block TX, fallback to next lowest allowed
 		 * AC for now) */
 		if (local->ops->conf_tx(local_to_hw(local), queue, &params)) {
@@ -3230,7 +3231,7 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len)
 			qparam.cw_min = 15;
 
 		qparam.cw_max = 1023;
-		qparam.burst_time = 0;
+		qparam.txop = 0;
 
 		for (i = IEEE80211_TX_QUEUE_DATA0; i < NUM_TX_DATA_QUEUES; i++)
 			local->ops->conf_tx(local_to_hw(local),
-- 
cgit v1.2.3


From d0f5afbe6de6b8c06f94a8a0b370252e3863afe7 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Tue, 12 Feb 2008 20:12:45 +0100
Subject: mac80211: Extend filter flag documentation about unsupported flags

This extends the filter flags documentation to make it clear
what clearing a flag really means.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5ecf3cc8d977..027d51f876ee 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -793,8 +793,18 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr)
  * parameter to see whether multicast frames should be accepted
  * or dropped.
  *
- * All unsupported flags in @total_flags must be cleared, i.e. you
- * should clear all bits except those you honoured.
+ * All unsupported flags in @total_flags must be cleared.
+ * Hardware does not support a flag if it is incapable of _passing_
+ * the frame to the stack. Otherwise the driver must ignore
+ * the flag, but not clear it.
+ * You must _only_ clear the flag (announce no support for the
+ * flag to mac80211) if you are not able to pass the packet type
+ * to the stack (so the hardware always filters it).
+ * So for example, you should clear @FIF_CONTROL, if your hardware
+ * always filters control frames. If your hardware always passes
+ * control frames to the kernel and is incapable of filtering them,
+ * you do _not_ clear the @FIF_CONTROL flag.
+ * This rule applies to all other FIF flags as well.
  */
 
 /**
-- 
cgit v1.2.3


From c132bec33c2eb5e46d8e4b80cfa5a9656d8e57e7 Mon Sep 17 00:00:00 2001
From: Bruno Randolf <bruno@thinktube.com>
Date: Mon, 18 Feb 2008 11:20:51 +0900
Subject: mac80211: better definition of mactime

define mactime as the time when the first data symbol arrived at the HW. the
old definition was questionable because 802.11 defines timestamp only for
beacon and probe response frames, and there it means the timestamp field.

a stricter definition of mactime is necessary for correct merging of IBSS.

note that it is up to the driver to convert whatever its hardware returns to
this definition. unfortunately we don't know for example when atheros hardware
takes its rx timestamp exactly :(

Signed-off-by: Bruno Randolf <bruno@thinktube.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 027d51f876ee..1a1d0d8203aa 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -295,7 +295,8 @@ enum mac80211_rx_flags {
  * The low-level driver should provide this information (the subset
  * supported by hardware) to the 802.11 code with each received
  * frame.
- * @mactime: MAC timestamp as defined by 802.11
+ * @mactime: value in microseconds of the 64-bit Time Synchronization Function
+ * 	(TSF) timer when the first data symbol (MPDU) arrived at the hardware.
  * @band: the active band when this frame was received
  * @freq: frequency the radio was tuned to when receiving this frame, in MHz
  * @ssi: signal strength when receiving this frame
-- 
cgit v1.2.3


From 9d9bf77d16ba527f6f63846ca18cf20ae6e8d697 Mon Sep 17 00:00:00 2001
From: Bruno Randolf <bruno@thinktube.com>
Date: Mon, 18 Feb 2008 11:21:36 +0900
Subject: mac80211: enable IBSS merging

enable IBSS cell merging. if an IBSS beacon with the same channel, same ESSID
and a TSF higher than the local TSF (mactime) is received, we have to join its
BSSID. while this might not be immediately apparent from reading the 802.11
standard it is compliant and necessary to make IBSS mode functional in many
cases. most drivers have a similar behaviour.

* move the relevant code section (previously only containing debug code) down
to the end of the function, so we can reuse the bss structure.

* we have to compare the mactime (TSF at the time of packet receive) rather
than the current TSF. since mactime is defined as the time the first data
symbol arrived we add the time until byte 24 where the timestamp resides, since
this is how the beacon timestamp is defined. as some some drivers are not able
to give a reliable mactime we fall back to use the current TSF, which will be
enough to catch most (but not all) cases where an IBSS merge is necessary.

* in IBSS mode we want to allow beacons to override probe response info so we
can correctly do merges.

* we don't only configure beacons based on scan results, so change that
message.

* to enable this we have to let all beacons thru in IBSS mode, even if they
have a different BSSID.

Signed-off-by: Bruno Randolf <bruno@thinktube.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h       |  3 +-
 net/mac80211/ieee80211_sta.c | 90 ++++++++++++++++++++++++++++++--------------
 net/mac80211/rx.c            |  5 ++-
 3 files changed, 67 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 1a1d0d8203aa..784ab769b00b 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -276,7 +276,8 @@ struct ieee80211_tx_control {
  * @RX_FLAG_FAILED_PLCP_CRC: Set this flag if the PCLP check failed on
  *	the frame.
  * @RX_FLAG_TSFT: The timestamp passed in the RX status (@mactime field)
- *	is valid.
+ *	is valid. This is useful in monitor mode and necessary for beacon frames
+ *	to enable IBSS merging.
  */
 enum mac80211_rx_flags {
 	RX_FLAG_MMIC_ERROR	= 1<<0,
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index a84ead49b1e6..6e5b57dccc50 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -2199,7 +2199,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 	struct ieee80211_sta_bss *bss;
 	struct sta_info *sta;
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	u64 timestamp;
+	u64 beacon_timestamp, rx_timestamp;
 	DECLARE_MAC_BUF(mac);
 	DECLARE_MAC_BUF(mac2);
 
@@ -2216,31 +2216,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 	if (baselen > len)
 		return;
 
-	timestamp = le64_to_cpu(mgmt->u.beacon.timestamp);
-
-	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && beacon &&
-	    memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0) {
-#ifdef CONFIG_MAC80211_IBSS_DEBUG
-		static unsigned long last_tsf_debug = 0;
-		u64 tsf;
-		if (local->ops->get_tsf)
-			tsf = local->ops->get_tsf(local_to_hw(local));
-		else
-			tsf = -1LLU;
-		if (time_after(jiffies, last_tsf_debug + 5 * HZ)) {
-			printk(KERN_DEBUG "RX beacon SA=%s BSSID="
-			       "%s TSF=0x%llx BCN=0x%llx diff=%lld "
-			       "@%lu\n",
-			       print_mac(mac, mgmt->sa), print_mac(mac2, mgmt->bssid),
-			       (unsigned long long)tsf,
-			       (unsigned long long)timestamp,
-			       (unsigned long long)(tsf - timestamp),
-			       jiffies);
-			last_tsf_debug = jiffies;
-		}
-#endif /* CONFIG_MAC80211_IBSS_DEBUG */
-	}
-
+	beacon_timestamp = le64_to_cpu(mgmt->u.beacon.timestamp);
 	ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems);
 
 	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates &&
@@ -2326,8 +2302,10 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 
 	bss->band = rx_status->band;
 
-	if (bss->probe_resp && beacon) {
-		/* Do not allow beacon to override data from Probe Response. */
+	if (sdata->vif.type != IEEE80211_IF_TYPE_IBSS &&
+	    bss->probe_resp && beacon) {
+		/* STA mode:
+		 * Do not allow beacon to override data from Probe Response. */
 		ieee80211_rx_bss_put(dev, bss);
 		return;
 	}
@@ -2424,13 +2402,67 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 		bss->ht_ie_len = 0;
 	}
 
-	bss->timestamp = timestamp;
+	bss->timestamp = beacon_timestamp;
 	bss->last_update = jiffies;
 	bss->rssi = rx_status->ssi;
 	bss->signal = rx_status->signal;
 	bss->noise = rx_status->noise;
 	if (!beacon)
 		bss->probe_resp++;
+
+	/* check if we need to merge IBSS */
+	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && beacon &&
+	    !local->sta_sw_scanning && !local->sta_hw_scanning &&
+	    mgmt->u.beacon.capab_info & WLAN_CAPABILITY_IBSS &&
+	    bss->freq == local->oper_channel->center_freq &&
+	    elems.ssid_len == sdata->u.sta.ssid_len &&
+	    memcmp(elems.ssid, sdata->u.sta.ssid, sdata->u.sta.ssid_len) == 0) {
+		if (rx_status->flag & RX_FLAG_TSFT) {
+			/* in order for correct IBSS merging we need mactime
+			 *
+			 * since mactime is defined as the time the first data
+			 * symbol of the frame hits the PHY, and the timestamp
+			 * of the beacon is defined as "the time that the data
+			 * symbol containing the first bit of the timestamp is
+			 * transmitted to the PHY plus the transmitting STA’s
+			 * delays through its local PHY from the MAC-PHY
+			 * interface to its interface with the WM"
+			 * (802.11 11.1.2) - equals the time this bit arrives at
+			 * the receiver - we have to take into account the
+			 * offset between the two.
+			 * e.g: at 1 MBit that means mactime is 192 usec earlier
+			 * (=24 bytes * 8 usecs/byte) than the beacon timestamp.
+			 */
+			int rate = local->hw.wiphy->bands[rx_status->band]->
+					bitrates[rx_status->rate_idx].bitrate;
+			rx_timestamp = rx_status->mactime + (24 * 8 * 10 / rate);
+		} else if (local && local->ops && local->ops->get_tsf)
+			/* second best option: get current TSF */
+			rx_timestamp = local->ops->get_tsf(local_to_hw(local));
+		else
+			/* can't merge without knowing the TSF */
+			rx_timestamp = -1LLU;
+#ifdef CONFIG_MAC80211_IBSS_DEBUG
+		printk(KERN_DEBUG "RX beacon SA=%s BSSID="
+		       "%s TSF=0x%llx BCN=0x%llx diff=%lld @%lu\n",
+		       print_mac(mac, mgmt->sa),
+		       print_mac(mac2, mgmt->bssid),
+		       (unsigned long long)rx_timestamp,
+		       (unsigned long long)beacon_timestamp,
+		       (unsigned long long)(rx_timestamp - beacon_timestamp),
+		       jiffies);
+#endif /* CONFIG_MAC80211_IBSS_DEBUG */
+		if (beacon_timestamp > rx_timestamp) {
+			if (CONFIG_MAC80211_IBSS_DEBUG || net_ratelimit())
+				printk(KERN_DEBUG "%s: beacon TSF higher than "
+				       "local TSF - IBSS merge with BSSID %s\n",
+				       dev->name, print_mac(mac, mgmt->bssid));
+			ieee80211_sta_join_ibss(dev, &sdata->u.sta, bss);
+			ieee80211_ibss_add_sta(dev, NULL,
+					       mgmt->bssid, mgmt->sa);
+		}
+	}
+
 	ieee80211_rx_bss_put(dev, bss);
 }
 
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 472b19fc9143..279ee493520d 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1695,7 +1695,10 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata,
 	case IEEE80211_IF_TYPE_IBSS:
 		if (!bssid)
 			return 0;
-		if (!ieee80211_bssid_match(bssid, sdata->u.sta.bssid)) {
+		if ((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT &&
+		    (rx->fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BEACON)
+			return 1;
+		else if (!ieee80211_bssid_match(bssid, sdata->u.sta.bssid)) {
 			if (!(rx->flags & IEEE80211_TXRXD_RXIN_SCAN))
 				return 0;
 			rx->flags &= ~IEEE80211_TXRXD_RXRA_MATCH;
-- 
cgit v1.2.3


From ffc7689ddae5cbe12bde437ae0f2b386d568b5cd Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Wed, 20 Feb 2008 19:08:10 +0100
Subject: ssb: Add support for 8bit register access

This adds support for 8bit wide register reads/writes.
This is needed in order to support the gigabit ethernet core.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/main.c      | 18 ++++++++++++++++++
 drivers/ssb/pci.c       | 28 ++++++++++++++++++++++++++++
 drivers/ssb/pcmcia.c    | 32 ++++++++++++++++++++++++++++++++
 include/linux/ssb/ssb.h | 10 ++++++++++
 4 files changed, 88 insertions(+)

(limited to 'include')

diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index bedb2b4ee9d2..8db40c4b86e9 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -505,6 +505,14 @@ error:
 	return err;
 }
 
+static u8 ssb_ssb_read8(struct ssb_device *dev, u16 offset)
+{
+	struct ssb_bus *bus = dev->bus;
+
+	offset += dev->core_index * SSB_CORE_SIZE;
+	return readb(bus->mmio + offset);
+}
+
 static u16 ssb_ssb_read16(struct ssb_device *dev, u16 offset)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -521,6 +529,14 @@ static u32 ssb_ssb_read32(struct ssb_device *dev, u16 offset)
 	return readl(bus->mmio + offset);
 }
 
+static void ssb_ssb_write8(struct ssb_device *dev, u16 offset, u8 value)
+{
+	struct ssb_bus *bus = dev->bus;
+
+	offset += dev->core_index * SSB_CORE_SIZE;
+	writeb(value, bus->mmio + offset);
+}
+
 static void ssb_ssb_write16(struct ssb_device *dev, u16 offset, u16 value)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -539,8 +555,10 @@ static void ssb_ssb_write32(struct ssb_device *dev, u16 offset, u32 value)
 
 /* Ops for the plain SSB bus without a host-device (no PCI or PCMCIA). */
 static const struct ssb_bus_ops ssb_ssb_ops = {
+	.read8		= ssb_ssb_read8,
 	.read16		= ssb_ssb_read16,
 	.read32		= ssb_ssb_read32,
+	.write8		= ssb_ssb_write8,
 	.write16	= ssb_ssb_write16,
 	.write32	= ssb_ssb_write32,
 };
diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index b434df75047f..1facc7620fc8 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -572,6 +572,19 @@ static inline int ssb_pci_assert_buspower(struct ssb_bus *bus)
 }
 #endif /* DEBUG */
 
+static u8 ssb_pci_read8(struct ssb_device *dev, u16 offset)
+{
+	struct ssb_bus *bus = dev->bus;
+
+	if (unlikely(ssb_pci_assert_buspower(bus)))
+		return 0xFF;
+	if (unlikely(bus->mapped_device != dev)) {
+		if (unlikely(ssb_pci_switch_core(bus, dev)))
+			return 0xFF;
+	}
+	return ioread8(bus->mmio + offset);
+}
+
 static u16 ssb_pci_read16(struct ssb_device *dev, u16 offset)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -598,6 +611,19 @@ static u32 ssb_pci_read32(struct ssb_device *dev, u16 offset)
 	return ioread32(bus->mmio + offset);
 }
 
+static void ssb_pci_write8(struct ssb_device *dev, u16 offset, u8 value)
+{
+	struct ssb_bus *bus = dev->bus;
+
+	if (unlikely(ssb_pci_assert_buspower(bus)))
+		return;
+	if (unlikely(bus->mapped_device != dev)) {
+		if (unlikely(ssb_pci_switch_core(bus, dev)))
+			return;
+	}
+	iowrite8(value, bus->mmio + offset);
+}
+
 static void ssb_pci_write16(struct ssb_device *dev, u16 offset, u16 value)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -626,8 +652,10 @@ static void ssb_pci_write32(struct ssb_device *dev, u16 offset, u32 value)
 
 /* Not "static", as it's used in main.c */
 const struct ssb_bus_ops ssb_pci_ops = {
+	.read8		= ssb_pci_read8,
 	.read16		= ssb_pci_read16,
 	.read32		= ssb_pci_read32,
+	.write8		= ssb_pci_write8,
 	.write16	= ssb_pci_write16,
 	.write32	= ssb_pci_write32,
 };
diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c
index 46816cda8b98..84b3a845a8a8 100644
--- a/drivers/ssb/pcmcia.c
+++ b/drivers/ssb/pcmcia.c
@@ -172,6 +172,22 @@ static int select_core_and_segment(struct ssb_device *dev,
 	return 0;
 }
 
+static u8 ssb_pcmcia_read8(struct ssb_device *dev, u16 offset)
+{
+	struct ssb_bus *bus = dev->bus;
+	unsigned long flags;
+	int err;
+	u8 value = 0xFF;
+
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (likely(!err))
+		value = readb(bus->mmio + offset);
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
+
+	return value;
+}
+
 static u16 ssb_pcmcia_read16(struct ssb_device *dev, u16 offset)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -206,6 +222,20 @@ static u32 ssb_pcmcia_read32(struct ssb_device *dev, u16 offset)
 	return (lo | (hi << 16));
 }
 
+static void ssb_pcmcia_write8(struct ssb_device *dev, u16 offset, u8 value)
+{
+	struct ssb_bus *bus = dev->bus;
+	unsigned long flags;
+	int err;
+
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (likely(!err))
+		writeb(value, bus->mmio + offset);
+	mmiowb();
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
+}
+
 static void ssb_pcmcia_write16(struct ssb_device *dev, u16 offset, u16 value)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -238,8 +268,10 @@ static void ssb_pcmcia_write32(struct ssb_device *dev, u16 offset, u32 value)
 
 /* Not "static", as it's used in main.c */
 const struct ssb_bus_ops ssb_pcmcia_ops = {
+	.read8		= ssb_pcmcia_read8,
 	.read16		= ssb_pcmcia_read16,
 	.read32		= ssb_pcmcia_read32,
+	.write8		= ssb_pcmcia_write8,
 	.write16	= ssb_pcmcia_write16,
 	.write32	= ssb_pcmcia_write32,
 };
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 20add65215af..860d28c6d149 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -72,8 +72,10 @@ struct ssb_device;
 /* Lowlevel read/write operations on the device MMIO.
  * Internal, don't use that outside of ssb. */
 struct ssb_bus_ops {
+	u8 (*read8)(struct ssb_device *dev, u16 offset);
 	u16 (*read16)(struct ssb_device *dev, u16 offset);
 	u32 (*read32)(struct ssb_device *dev, u16 offset);
+	void (*write8)(struct ssb_device *dev, u16 offset, u8 value);
 	void (*write16)(struct ssb_device *dev, u16 offset, u16 value);
 	void (*write32)(struct ssb_device *dev, u16 offset, u32 value);
 };
@@ -344,6 +346,10 @@ void ssb_device_disable(struct ssb_device *dev, u32 core_specific_flags);
 
 
 /* Device MMIO register read/write functions. */
+static inline u8 ssb_read8(struct ssb_device *dev, u16 offset)
+{
+	return dev->ops->read8(dev, offset);
+}
 static inline u16 ssb_read16(struct ssb_device *dev, u16 offset)
 {
 	return dev->ops->read16(dev, offset);
@@ -352,6 +358,10 @@ static inline u32 ssb_read32(struct ssb_device *dev, u16 offset)
 {
 	return dev->ops->read32(dev, offset);
 }
+static inline void ssb_write8(struct ssb_device *dev, u16 offset, u8 value)
+{
+	dev->ops->write8(dev, offset, value);
+}
 static inline void ssb_write16(struct ssb_device *dev, u16 offset, u16 value)
 {
 	dev->ops->write16(dev, offset, value);
-- 
cgit v1.2.3


From d46e144b65bf053b25d134ec9f52a38e63e04bb4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 20 Feb 2008 23:59:33 +0100
Subject: mac80211: rework TX filtered frame code

This reworks the code for TX filtered frames, splitting it out to
a new function to handle those cases, making the clear instruction
a flag and renaming a few things to be easier to understand and
less Atheros hardware specific. Finally, it also makes the comments
explain more.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h   |   3 +-
 net/mac80211/ieee80211.c | 121 +++++++++++++++++++++++++++++++----------------
 net/mac80211/sta_info.h  |   6 ++-
 net/mac80211/tx.c        |  12 ++---
 4 files changed, 91 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 784ab769b00b..cd4b1c1a0d48 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -228,7 +228,8 @@ struct ieee80211_tx_control {
 #define IEEE80211_TXCTL_NO_ACK		(1<<4) /* tell the low level not to
 						* wait for an ack */
 #define IEEE80211_TXCTL_RATE_CTRL_PROBE	(1<<5)
-#define IEEE80211_TXCTL_CLEAR_DST_MASK	(1<<6)
+#define IEEE80211_TXCTL_CLEAR_PS_FILT	(1<<6) /* clear powersave filter
+						* for destination station */
 #define IEEE80211_TXCTL_REQUEUE		(1<<7)
 #define IEEE80211_TXCTL_FIRST_FRAGMENT	(1<<8) /* this is a first fragment of
 						* the frame */
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 7df14799e38b..a00858dbab18 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -1178,6 +1178,77 @@ no_key:
 	}
 }
 
+static void ieee80211_handle_filtered_frame(struct ieee80211_local *local,
+					    struct sta_info *sta,
+					    struct sk_buff *skb,
+					    struct ieee80211_tx_status *status)
+{
+	sta->tx_filtered_count++;
+
+	/*
+	 * Clear the TX filter mask for this STA when sending the next
+	 * packet. If the STA went to power save mode, this will happen
+	 * happen when it wakes up for the next time.
+	 */
+	sta->flags |= WLAN_STA_CLEAR_PS_FILT;
+
+	/*
+	 * This code races in the following way:
+	 *
+	 *  (1) STA sends frame indicating it will go to sleep and does so
+	 *  (2) hardware/firmware adds STA to filter list, passes frame up
+	 *  (3) hardware/firmware processes TX fifo and suppresses a frame
+	 *  (4) we get TX status before having processed the frame and
+	 *	knowing that the STA has gone to sleep.
+	 *
+	 * This is actually quite unlikely even when both those events are
+	 * processed from interrupts coming in quickly after one another or
+	 * even at the same time because we queue both TX status events and
+	 * RX frames to be processed by a tasklet and process them in the
+	 * same order that they were received or TX status last. Hence, there
+	 * is no race as long as the frame RX is processed before the next TX
+	 * status, which drivers can ensure, see below.
+	 *
+	 * Note that this can only happen if the hardware or firmware can
+	 * actually add STAs to the filter list, if this is done by the
+	 * driver in response to set_tim() (which will only reduce the race
+	 * this whole filtering tries to solve, not completely solve it)
+	 * this situation cannot happen.
+	 *
+	 * To completely solve this race drivers need to make sure that they
+	 *  (a) don't mix the irq-safe/not irq-safe TX status/RX processing
+	 *	functions and
+	 *  (b) always process RX events before TX status events if ordering
+	 *      can be unknown, for example with different interrupt status
+	 *	bits.
+	 */
+	if (sta->flags & WLAN_STA_PS &&
+	    skb_queue_len(&sta->tx_filtered) < STA_MAX_TX_BUFFER) {
+		ieee80211_remove_tx_extra(local, sta->key, skb,
+					  &status->control);
+		skb_queue_tail(&sta->tx_filtered, skb);
+		return;
+	}
+
+	if (!(sta->flags & WLAN_STA_PS) &&
+	    !(status->control.flags & IEEE80211_TXCTL_REQUEUE)) {
+		/* Software retry the packet once */
+		status->control.flags |= IEEE80211_TXCTL_REQUEUE;
+		ieee80211_remove_tx_extra(local, sta->key, skb,
+					  &status->control);
+		dev_queue_xmit(skb);
+		return;
+	}
+
+	if (net_ratelimit())
+		printk(KERN_DEBUG "%s: dropped TX filtered frame, "
+		       "queue_len=%d PS=%d @%lu\n",
+		       wiphy_name(local->hw.wiphy),
+		       skb_queue_len(&sta->tx_filtered),
+		       !!(sta->flags & WLAN_STA_PS), jiffies);
+	dev_kfree_skb(skb);
+}
+
 void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
 			 struct ieee80211_tx_status *status)
 {
@@ -1202,11 +1273,16 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
 			if (sta->flags & WLAN_STA_PS) {
-				/* The STA is in power save mode, so assume
+				/*
+				 * The STA is in power save mode, so assume
 				 * that this TX packet failed because of that.
 				 */
 				status->excessive_retries = 0;
 				status->flags |= IEEE80211_TX_STATUS_TX_FILTERED;
+				ieee80211_handle_filtered_frame(local, sta,
+								skb, status);
+				sta_info_put(sta);
+				return;
 			}
 			sta_info_put(sta);
 		}
@@ -1216,47 +1292,8 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb,
 		struct sta_info *sta;
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
-			sta->tx_filtered_count++;
-
-			/* Clear the TX filter mask for this STA when sending
-			 * the next packet. If the STA went to power save mode,
-			 * this will happen when it is waking up for the next
-			 * time. */
-			sta->clear_dst_mask = 1;
-
-			/* TODO: Is the WLAN_STA_PS flag always set here or is
-			 * the race between RX and TX status causing some
-			 * packets to be filtered out before 80211.o gets an
-			 * update for PS status? This seems to be the case, so
-			 * no changes are likely to be needed. */
-			if (sta->flags & WLAN_STA_PS &&
-			    skb_queue_len(&sta->tx_filtered) <
-			    STA_MAX_TX_BUFFER) {
-				ieee80211_remove_tx_extra(local, sta->key,
-							  skb,
-							  &status->control);
-				skb_queue_tail(&sta->tx_filtered, skb);
-			} else if (!(sta->flags & WLAN_STA_PS) &&
-				   !(status->control.flags & IEEE80211_TXCTL_REQUEUE)) {
-				/* Software retry the packet once */
-				status->control.flags |= IEEE80211_TXCTL_REQUEUE;
-				ieee80211_remove_tx_extra(local, sta->key,
-							  skb,
-							  &status->control);
-				dev_queue_xmit(skb);
-			} else {
-				if (net_ratelimit()) {
-					printk(KERN_DEBUG "%s: dropped TX "
-					       "filtered frame queue_len=%d "
-					       "PS=%d @%lu\n",
-					       wiphy_name(local->hw.wiphy),
-					       skb_queue_len(
-						       &sta->tx_filtered),
-					       !!(sta->flags & WLAN_STA_PS),
-					       jiffies);
-				}
-				dev_kfree_skb(skb);
-			}
+			ieee80211_handle_filtered_frame(local, sta, skb,
+							status);
 			sta_info_put(sta);
 			return;
 		}
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index 4099ece143ef..f7e65fa3f9ed 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -32,6 +32,9 @@
  * @WLAN_STA_WME: Station is a QoS-STA.
  * @WLAN_STA_WDS: Station is one of our WDS peers.
  * @WLAN_STA_PSPOLL: Station has just PS-polled us.
+ * @WLAN_STA_CLEAR_PS_FILT: Clear PS filter in hardware (using the
+ *	IEEE80211_TXCTL_CLEAR_PS_FILT control flag) when the next
+ *	frame to this station is transmitted.
  */
 enum ieee80211_sta_info_flags {
 	WLAN_STA_AUTH		= 1<<0,
@@ -43,6 +46,7 @@ enum ieee80211_sta_info_flags {
 	WLAN_STA_WME		= 1<<6,
 	WLAN_STA_WDS		= 1<<7,
 	WLAN_STA_PSPOLL		= 1<<8,
+	WLAN_STA_CLEAR_PS_FILT	= 1<<9,
 };
 
 #define STA_TID_NUM 16
@@ -136,8 +140,6 @@ struct sta_info {
 	struct sk_buff_head tx_filtered; /* buffer of TX frames that were
 					  * already given to low-level driver,
 					  * but were filtered */
-	int clear_dst_mask;
-
 	unsigned long rx_packets, tx_packets; /* number of RX/TX MSDUs */
 	unsigned long rx_bytes, tx_bytes;
 	unsigned long tx_retry_failed, tx_retry_count;
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 69fdb763aa8b..1cd58e01f1ee 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -1020,10 +1020,10 @@ __ieee80211_tx_prepare(struct ieee80211_txrx_data *tx,
 	}
 
 	if (!tx->sta)
-		control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK;
-	else if (tx->sta->clear_dst_mask) {
-		control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK;
-		tx->sta->clear_dst_mask = 0;
+		control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT;
+	else if (tx->sta->flags & WLAN_STA_CLEAR_PS_FILT) {
+		control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT;
+		tx->sta->flags &= ~WLAN_STA_CLEAR_PS_FILT;
 	}
 
 	hdrlen = ieee80211_get_hdrlen(tx->fc);
@@ -1084,7 +1084,7 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb,
 	if (tx->u.tx.extra_frag) {
 		control->flags &= ~(IEEE80211_TXCTL_USE_RTS_CTS |
 				    IEEE80211_TXCTL_USE_CTS_PROTECT |
-				    IEEE80211_TXCTL_CLEAR_DST_MASK |
+				    IEEE80211_TXCTL_CLEAR_PS_FILT |
 				    IEEE80211_TXCTL_FIRST_FRAGMENT);
 		for (i = 0; i < tx->u.tx.num_extra_frag; i++) {
 			if (!tx->u.tx.extra_frag[i])
@@ -1806,7 +1806,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 		control->antenna_sel_tx = local->hw.conf.antenna_sel_tx;
 		control->flags |= IEEE80211_TXCTL_NO_ACK;
 		control->retry_limit = 1;
-		control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK;
+		control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT;
 	}
 
 	ap->num_beacons++;
-- 
cgit v1.2.3


From 28de57d1a9eb7e67badb731297197fcbef0cc19e Mon Sep 17 00:00:00 2001
From: Aurelien Jarno <aurelien@aurel32.net>
Date: Fri, 22 Feb 2008 16:14:58 +0100
Subject: ssb: Add CHIPCO IRQ access functions

This patch adds functions to setup and read the CHIPCO IRQ.

Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/driver_chipcommon.c           | 10 ++++++++++
 include/linux/ssb/ssb_driver_chipcommon.h |  4 ++++
 2 files changed, 14 insertions(+)

(limited to 'include')

diff --git a/drivers/ssb/driver_chipcommon.c b/drivers/ssb/driver_chipcommon.c
index e586321a473a..45b672a69003 100644
--- a/drivers/ssb/driver_chipcommon.c
+++ b/drivers/ssb/driver_chipcommon.c
@@ -353,6 +353,16 @@ void ssb_chipco_watchdog_timer_set(struct ssb_chipcommon *cc, u32 ticks)
 	chipco_write32(cc, SSB_CHIPCO_WATCHDOG, ticks);
 }
 
+void ssb_chipco_irq_mask(struct ssb_chipcommon *cc, u32 mask, u32 value)
+{
+	chipco_write32_masked(cc, SSB_CHIPCO_IRQMASK, mask, value);
+}
+
+u32 ssb_chipco_irq_status(struct ssb_chipcommon *cc, u32 mask)
+{
+	return chipco_read32(cc, SSB_CHIPCO_IRQSTAT) & mask;
+}
+
 u32 ssb_chipco_gpio_in(struct ssb_chipcommon *cc, u32 mask)
 {
 	return chipco_read32(cc, SSB_CHIPCO_GPIOIN) & mask;
diff --git a/include/linux/ssb/ssb_driver_chipcommon.h b/include/linux/ssb/ssb_driver_chipcommon.h
index 536851b946f6..b548a54ff1f5 100644
--- a/include/linux/ssb/ssb_driver_chipcommon.h
+++ b/include/linux/ssb/ssb_driver_chipcommon.h
@@ -390,6 +390,10 @@ extern void ssb_chipco_set_clockmode(struct ssb_chipcommon *cc,
 extern void ssb_chipco_watchdog_timer_set(struct ssb_chipcommon *cc,
 					  u32 ticks);
 
+void ssb_chipco_irq_mask(struct ssb_chipcommon *cc, u32 mask, u32 value);
+
+u32 ssb_chipco_irq_status(struct ssb_chipcommon *cc, u32 mask);
+
 /* Chipcommon GPIO pin access. */
 u32 ssb_chipco_gpio_in(struct ssb_chipcommon *cc, u32 mask);
 u32 ssb_chipco_gpio_out(struct ssb_chipcommon *cc, u32 mask, u32 value);
-- 
cgit v1.2.3


From 2485f7105f20f85c2dbebc67be6b2cb97175fa7e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Mon, 25 Feb 2008 16:27:41 +0100
Subject: mac80211: clarify use of TX status/RX callbacks

This patch clarifies the use of the irqsafe vs. non-irq-safe
functions and their respective locking requirements.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 37 ++++++++++++++++++++++++++++++++++---
 1 file changed, 34 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index cd4b1c1a0d48..7a80c3981237 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -38,7 +38,11 @@
  * called in hardware interrupt context. The low-level driver must not call any
  * other functions in hardware interrupt context. If there is a need for such
  * call, the low-level driver should first ACK the interrupt and perform the
- * IEEE 802.11 code call after this, e.g. from a scheduled workqueue function.
+ * IEEE 802.11 code call after this, e.g. from a scheduled workqueue or even
+ * tasklet function.
+ *
+ * NOTE: If the driver opts to use the _irqsafe() functions, it may not also
+ *	 use the non-irqsafe functions!
  */
 
 /**
@@ -1204,7 +1208,10 @@ void __ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb,
  * buffer in @skb must start with an IEEE 802.11 header or a radiotap
  * header if %RX_FLAG_RADIOTAP is set in the @status flags.
  *
- * This function may not be called in IRQ context.
+ * This function may not be called in IRQ context. Calls to this function
+ * for a single hardware must be synchronized against each other. Calls
+ * to this function and ieee80211_rx_irqsafe() may not be mixed for a
+ * single hardware.
  *
  * @hw: the hardware this frame came in on
  * @skb: the buffer to receive, owned by mac80211 after this call
@@ -1221,7 +1228,10 @@ static inline void ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb,
  * ieee80211_rx_irqsafe - receive frame
  *
  * Like ieee80211_rx() but can be called in IRQ context
- * (internally defers to a workqueue.)
+ * (internally defers to a tasklet.)
+ *
+ * Calls to this function and ieee80211_rx() may not be mixed for a
+ * single hardware.
  *
  * @hw: the hardware this frame came in on
  * @skb: the buffer to receive, owned by mac80211 after this call
@@ -1240,6 +1250,11 @@ void ieee80211_rx_irqsafe(struct ieee80211_hw *hw,
  * transmitted. It is permissible to not call this function for
  * multicast frames but this can affect statistics.
  *
+ * This function may not be called in IRQ context. Calls to this function
+ * for a single hardware must be synchronized against each other. Calls
+ * to this function and ieee80211_tx_status_irqsafe() may not be mixed
+ * for a single hardware.
+ *
  * @hw: the hardware the frame was transmitted by
  * @skb: the frame that was transmitted, owned by mac80211 after this call
  * @status: status information for this frame; the status pointer need not
@@ -1249,6 +1264,22 @@ void ieee80211_rx_irqsafe(struct ieee80211_hw *hw,
 void ieee80211_tx_status(struct ieee80211_hw *hw,
 			 struct sk_buff *skb,
 			 struct ieee80211_tx_status *status);
+
+/**
+ * ieee80211_tx_status_irqsafe - irq-safe transmit status callback
+ *
+ * Like ieee80211_tx_status() but can be called in IRQ context
+ * (internally defers to a tasklet.)
+ *
+ * Calls to this function and ieee80211_tx_status() may not be mixed for a
+ * single hardware.
+ *
+ * @hw: the hardware the frame was transmitted by
+ * @skb: the frame that was transmitted, owned by mac80211 after this call
+ * @status: status information for this frame; the status pointer need not
+ *	be valid after this function returns and is not freed by mac80211,
+ *	it is recommended that it points to a stack area
+ */
 void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw,
 				 struct sk_buff *skb,
 				 struct ieee80211_tx_status *status);
-- 
cgit v1.2.3


From 988b705077d8f922408913f4f521ae073256d4a1 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 3 Mar 2008 12:20:57 -0800
Subject: [ARP]: Introduce the arp_hdr_len helper.

There are some place, that calculate the ARP header length. These
calculations are correct, but
 a) some operate with "magic" constants,
 b) enlarge the code length (sometimes at the cost of coding style),
 c) are not informative from the first glance.

The proposal is to introduce a helper, that includes all the good
sides of these calculations.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bonding/bond_main.c | 5 +----
 include/linux/if_arp.h          | 6 ++++++
 net/core/netpoll.c              | 6 ++----
 net/ipv4/arp.c                  | 9 +++------
 net/ipv4/ipconfig.c             | 5 +----
 net/ipv4/netfilter/arp_tables.c | 5 +----
 6 files changed, 14 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 966643473da7..5fc9d8d58ece 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2646,10 +2646,7 @@ static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct pack
 	if (!slave || !slave_do_arp_validate(bond, slave))
 		goto out_unlock;
 
-	/* ARP header, plus 2 device addresses, plus 2 IP addresses.  */
-	if (!pskb_may_pull(skb, (sizeof(struct arphdr) +
-				 (2 * dev->addr_len) +
-				 (2 * sizeof(u32)))))
+	if (!pskb_may_pull(skb, arp_hdr_len(dev)))
 		goto out_unlock;
 
 	arp = arp_hdr(skb);
diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h
index 296e8e86e91d..4d3401812e6c 100644
--- a/include/linux/if_arp.h
+++ b/include/linux/if_arp.h
@@ -156,6 +156,12 @@ static inline struct arphdr *arp_hdr(const struct sk_buff *skb)
 {
 	return (struct arphdr *)skb_network_header(skb);
 }
+
+static inline int arp_hdr_len(struct net_device *dev)
+{
+	/* ARP header, plus 2 device addresses, plus 2 IP addresses. */
+	return sizeof(struct arphdr) + (dev->addr_len + sizeof(u32)) * 2;
+}
 #endif
 
 #endif	/* _LINUX_IF_ARP_H */
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 6faa128a4c8e..7ae98659d79d 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -384,9 +384,7 @@ static void arp_reply(struct sk_buff *skb)
 	if (skb->dev->flags & IFF_NOARP)
 		return;
 
-	if (!pskb_may_pull(skb, (sizeof(struct arphdr) +
-				 (2 * skb->dev->addr_len) +
-				 (2 * sizeof(u32)))))
+	if (!pskb_may_pull(skb, arp_hdr_len(skb->dev)))
 		return;
 
 	skb_reset_network_header(skb);
@@ -414,7 +412,7 @@ static void arp_reply(struct sk_buff *skb)
 	    ipv4_is_loopback(tip) || ipv4_is_multicast(tip))
 		return;
 
-	size = sizeof(struct arphdr) + 2 * (skb->dev->addr_len + 4);
+	size = arp_hdr_len(skb->dev);
 	send_skb = find_skb(np, size + LL_RESERVED_SPACE(np->dev),
 			    LL_RESERVED_SPACE(np->dev));
 
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 8e17f65f4002..69e80bd9774a 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -570,14 +570,13 @@ struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip,
 	 *	Allocate a buffer
 	 */
 
-	skb = alloc_skb(sizeof(struct arphdr)+ 2*(dev->addr_len+4)
-				+ LL_RESERVED_SPACE(dev), GFP_ATOMIC);
+	skb = alloc_skb(arp_hdr_len(dev) + LL_RESERVED_SPACE(dev), GFP_ATOMIC);
 	if (skb == NULL)
 		return NULL;
 
 	skb_reserve(skb, LL_RESERVED_SPACE(dev));
 	skb_reset_network_header(skb);
-	arp = (struct arphdr *) skb_put(skb,sizeof(struct arphdr) + 2*(dev->addr_len+4));
+	arp = (struct arphdr *) skb_put(skb, arp_hdr_len(dev));
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_ARP);
 	if (src_hw == NULL)
@@ -916,9 +915,7 @@ static int arp_rcv(struct sk_buff *skb, struct net_device *dev,
 		goto freeskb;
 
 	/* ARP header, plus 2 device addresses, plus 2 IP addresses.  */
-	if (!pskb_may_pull(skb, (sizeof(struct arphdr) +
-				 (2 * dev->addr_len) +
-				 (2 * sizeof(u32)))))
+	if (!pskb_may_pull(skb, arp_hdr_len(dev)))
 		goto freeskb;
 
 	arp = arp_hdr(skb);
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index c90e75a66e81..f84041d1f623 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -459,10 +459,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 	if (rarp->ar_pro != htons(ETH_P_IP))
 		goto drop;
 
-	if (!pskb_may_pull(skb,
-			   sizeof(struct arphdr) +
-			   (2 * dev->addr_len) +
-			   (2 * 4)))
+	if (!pskb_may_pull(skb, arp_hdr_len(dev)))
 		goto drop;
 
 	/* OK, it is all there and looks valid, process... */
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index a7591ce344d2..9b5904486184 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -233,10 +233,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 	void *table_base;
 	struct xt_table_info *private;
 
-	/* ARP header, plus 2 device addresses, plus 2 IP addresses.  */
-	if (!pskb_may_pull(skb, (sizeof(struct arphdr) +
-				 (2 * skb->dev->addr_len) +
-				 (2 * sizeof(u32)))))
+	if (!pskb_may_pull(skb, arp_hdr_len(skb->dev)))
 		return NF_DROP;
 
 	indev = in ? in->name : nulldevname;
-- 
cgit v1.2.3


From c6aefafb7ec620911d46174eed514f9df639e5a4 Mon Sep 17 00:00:00 2001
From: Glenn Griffin <ggriffin.kernel@gmail.com>
Date: Thu, 7 Feb 2008 21:49:26 -0800
Subject: [TCP]: Add IPv6 support to TCP SYN cookies

Updated to incorporate Eric's suggestion of using a per cpu buffer
rather than allocating on the stack.  Just a two line change, but will
resend in it's entirety.

Signed-off-by: Glenn Griffin <ggriffin.kernel@gmail.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/tcp.h        |   8 ++
 net/ipv4/syncookies.c    |   7 +-
 net/ipv4/tcp_input.c     |   1 +
 net/ipv4/tcp_minisocks.c |   2 +
 net/ipv4/tcp_output.c    |   1 +
 net/ipv6/Makefile        |   1 +
 net/ipv6/syncookies.c    | 267 +++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/tcp_ipv6.c      |  77 ++++++++++----
 8 files changed, 338 insertions(+), 26 deletions(-)
 create mode 100644 net/ipv6/syncookies.c

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index ae9774b478f5..11119e33acfe 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -29,6 +29,7 @@
 #include <linux/skbuff.h>
 #include <linux/dmaengine.h>
 #include <linux/crypto.h>
+#include <linux/cryptohash.h>
 
 #include <net/inet_connection_sock.h>
 #include <net/inet_timewait_sock.h>
@@ -434,11 +435,17 @@ extern int			tcp_disconnect(struct sock *sk, int flags);
 extern void			tcp_unhash(struct sock *sk);
 
 /* From syncookies.c */
+extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS];
 extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, 
 				    struct ip_options *opt);
 extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, 
 				     __u16 *mss);
 
+/* From net/ipv6/syncookies.c */
+extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb);
+extern __u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb,
+				     __u16 *mss);
+
 /* tcp_output.c */
 
 extern void __tcp_push_pending_frames(struct sock *sk, unsigned int cur_mss,
@@ -1332,6 +1339,7 @@ extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo);
 extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo);
 
 extern struct request_sock_ops tcp_request_sock_ops;
+extern struct request_sock_ops tcp6_request_sock_ops;
 
 extern int tcp_v4_destroy_sock(struct sock *sk);
 
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 177da14f6b14..4704f27f6c0b 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -10,8 +10,6 @@
  *      2 of the License, or (at your option) any later version.
  *
  *  $Id: syncookies.c,v 1.18 2002/02/01 22:01:04 davem Exp $
- *
- *  Missing: IPv6 support.
  */
 
 #include <linux/tcp.h>
@@ -23,14 +21,15 @@
 
 extern int sysctl_tcp_syncookies;
 
-static __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS];
+__u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS];
+EXPORT_SYMBOL(syncookie_secret);
 
 static __init int init_syncookies(void)
 {
 	get_random_bytes(syncookie_secret, sizeof(syncookie_secret));
 	return 0;
 }
-module_init(init_syncookies);
+__initcall(init_syncookies);
 
 #define COOKIEBITS 24	/* Upper bits store count */
 #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 19c449f62672..93e128c239c4 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5326,6 +5326,7 @@ discard:
 
 EXPORT_SYMBOL(sysctl_tcp_ecn);
 EXPORT_SYMBOL(sysctl_tcp_reordering);
+EXPORT_SYMBOL(sysctl_tcp_adv_win_scale);
 EXPORT_SYMBOL(tcp_parse_options);
 EXPORT_SYMBOL(tcp_rcv_established);
 EXPORT_SYMBOL(tcp_rcv_state_process);
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 0fdd1db641ac..8245247a6ceb 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -35,6 +35,8 @@
 #endif
 
 int sysctl_tcp_syncookies __read_mostly = SYNC_INIT;
+EXPORT_SYMBOL(sysctl_tcp_syncookies);
+
 int sysctl_tcp_abort_on_overflow __read_mostly;
 
 struct inet_timewait_death_row tcp_death_row = {
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index ed750f9ceb07..cbfef8b1f5e8 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2560,6 +2560,7 @@ void tcp_send_probe0(struct sock *sk)
 	}
 }
 
+EXPORT_SYMBOL(tcp_select_initial_window);
 EXPORT_SYMBOL(tcp_connect);
 EXPORT_SYMBOL(tcp_make_synack);
 EXPORT_SYMBOL(tcp_simple_retransmit);
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index 24f3aa0f2a35..ae14617e607f 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -16,6 +16,7 @@ ipv6-$(CONFIG_XFRM) += xfrm6_policy.o xfrm6_state.o xfrm6_input.o \
 ipv6-$(CONFIG_NETFILTER) += netfilter.o
 ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o
 ipv6-$(CONFIG_PROC_FS) += proc.o
+ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o
 
 ipv6-objs += $(ipv6-y)
 
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
new file mode 100644
index 000000000000..827c5aa7524c
--- /dev/null
+++ b/net/ipv6/syncookies.c
@@ -0,0 +1,267 @@
+/*
+ *  IPv6 Syncookies implementation for the Linux kernel
+ *
+ *  Authors:
+ *  Glenn Griffin	<ggriffin.kernel@gmail.com>
+ *
+ *  Based on IPv4 implementation by Andi Kleen
+ *  linux/net/ipv4/syncookies.c
+ *
+ *	This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ *
+ */
+
+#include <linux/tcp.h>
+#include <linux/random.h>
+#include <linux/cryptohash.h>
+#include <linux/kernel.h>
+#include <net/ipv6.h>
+#include <net/tcp.h>
+
+extern int sysctl_tcp_syncookies;
+extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS];
+
+#define COOKIEBITS 24	/* Upper bits store count */
+#define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1)
+
+/*
+ * This table has to be sorted and terminated with (__u16)-1.
+ * XXX generate a better table.
+ * Unresolved Issues: HIPPI with a 64k MSS is not well supported.
+ *
+ * Taken directly from ipv4 implementation.
+ * Should this list be modified for ipv6 use or is it close enough?
+ * rfc 2460 8.3 suggests mss values 20 bytes less than ipv4 counterpart
+ */
+static __u16 const msstab[] = {
+	64 - 1,
+	256 - 1,
+	512 - 1,
+	536 - 1,
+	1024 - 1,
+	1440 - 1,
+	1460 - 1,
+	4312 - 1,
+	(__u16)-1
+};
+/* The number doesn't include the -1 terminator */
+#define NUM_MSS (ARRAY_SIZE(msstab) - 1)
+
+/*
+ * This (misnamed) value is the age of syncookie which is permitted.
+ * Its ideal value should be dependent on TCP_TIMEOUT_INIT and
+ * sysctl_tcp_retries1. It's a rather complicated formula (exponential
+ * backoff) to compute at runtime so it's currently hardcoded here.
+ */
+#define COUNTER_TRIES 4
+
+static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb,
+					   struct request_sock *req,
+					   struct dst_entry *dst)
+{
+	struct inet_connection_sock *icsk = inet_csk(sk);
+	struct sock *child;
+
+	child = icsk->icsk_af_ops->syn_recv_sock(sk, skb, req, dst);
+	if (child)
+		inet_csk_reqsk_queue_add(sk, req, child);
+	else
+		reqsk_free(req);
+
+	return child;
+}
+
+static DEFINE_PER_CPU(__u32, cookie_scratch)[16 + 5 + SHA_WORKSPACE_WORDS];
+
+static u32 cookie_hash(struct in6_addr *saddr, struct in6_addr *daddr,
+		       __be16 sport, __be16 dport, u32 count, int c)
+{
+	__u32 *tmp = __get_cpu_var(cookie_scratch);
+
+	/*
+	 * we have 320 bits of information to hash, copy in the remaining
+	 * 192 bits required for sha_transform, from the syncookie_secret
+	 * and overwrite the digest with the secret
+	 */
+	memcpy(tmp + 10, syncookie_secret[c], 44);
+	memcpy(tmp, saddr, 16);
+	memcpy(tmp + 4, daddr, 16);
+	tmp[8] = ((__force u32)sport << 16) + (__force u32)dport;
+	tmp[9] = count;
+	sha_transform(tmp + 16, (__u8 *)tmp, tmp + 16 + 5);
+
+	return tmp[17];
+}
+
+static __u32 secure_tcp_syn_cookie(struct in6_addr *saddr, struct in6_addr *daddr,
+				   __be16 sport, __be16 dport, __u32 sseq,
+				   __u32 count, __u32 data)
+{
+	return (cookie_hash(saddr, daddr, sport, dport, 0, 0) +
+		sseq + (count << COOKIEBITS) +
+		((cookie_hash(saddr, daddr, sport, dport, count, 1) + data)
+		& COOKIEMASK));
+}
+
+static __u32 check_tcp_syn_cookie(__u32 cookie, struct in6_addr *saddr,
+				  struct in6_addr *daddr, __be16 sport,
+				  __be16 dport, __u32 sseq, __u32 count,
+				  __u32 maxdiff)
+{
+	__u32 diff;
+
+	cookie -= cookie_hash(saddr, daddr, sport, dport, 0, 0) + sseq;
+
+	diff = (count - (cookie >> COOKIEBITS)) & ((__u32) -1 >> COOKIEBITS);
+	if (diff >= maxdiff)
+		return (__u32)-1;
+
+	return (cookie -
+		cookie_hash(saddr, daddr, sport, dport, count - diff, 1))
+		& COOKIEMASK;
+}
+
+__u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp)
+{
+	struct ipv6hdr *iph = ipv6_hdr(skb);
+	const struct tcphdr *th = tcp_hdr(skb);
+	int mssind;
+	const __u16 mss = *mssp;
+
+	tcp_sk(sk)->last_synq_overflow = jiffies;
+
+	for (mssind = 0; mss > msstab[mssind + 1]; mssind++)
+		;
+	*mssp = msstab[mssind] + 1;
+
+	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESSENT);
+
+	return secure_tcp_syn_cookie(&iph->saddr, &iph->daddr, th->source,
+				     th->dest, ntohl(th->seq),
+				     jiffies / (HZ * 60), mssind);
+}
+
+static inline int cookie_check(struct sk_buff *skb, __u32 cookie)
+{
+	struct ipv6hdr *iph = ipv6_hdr(skb);
+	const struct tcphdr *th = tcp_hdr(skb);
+	__u32 seq = ntohl(th->seq) - 1;
+	__u32 mssind = check_tcp_syn_cookie(cookie, &iph->saddr, &iph->daddr,
+					    th->source, th->dest, seq,
+					    jiffies / (HZ * 60), COUNTER_TRIES);
+
+	return mssind < NUM_MSS ? msstab[mssind] + 1 : 0;
+}
+
+struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
+{
+	struct inet_request_sock *ireq;
+	struct inet6_request_sock *ireq6;
+	struct tcp_request_sock *treq;
+	struct ipv6_pinfo *np = inet6_sk(sk);
+	struct tcp_sock *tp = tcp_sk(sk);
+	const struct tcphdr *th = tcp_hdr(skb);
+	__u32 cookie = ntohl(th->ack_seq) - 1;
+	struct sock *ret = sk;
+	struct request_sock *req;
+	int mss;
+	struct dst_entry *dst;
+	__u8 rcv_wscale;
+
+	if (!sysctl_tcp_syncookies || !th->ack)
+		goto out;
+
+	if (time_after(jiffies, tp->last_synq_overflow + TCP_TIMEOUT_INIT) ||
+		(mss = cookie_check(skb, cookie)) == 0) {
+		NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESFAILED);
+		goto out;
+	}
+
+	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV);
+
+	ret = NULL;
+	req = inet6_reqsk_alloc(&tcp6_request_sock_ops);
+	if (!req)
+		goto out;
+
+	ireq = inet_rsk(req);
+	ireq6 = inet6_rsk(req);
+	treq = tcp_rsk(req);
+	ireq6->pktopts = NULL;
+
+	if (security_inet_conn_request(sk, skb, req)) {
+		reqsk_free(req);
+		goto out;
+	}
+
+	req->mss = mss;
+	ireq->rmt_port = th->source;
+	ipv6_addr_copy(&ireq6->rmt_addr, &ipv6_hdr(skb)->saddr);
+	ipv6_addr_copy(&ireq6->loc_addr, &ipv6_hdr(skb)->daddr);
+	if (ipv6_opt_accepted(sk, skb) ||
+	    np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo ||
+	    np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) {
+		atomic_inc(&skb->users);
+		ireq6->pktopts = skb;
+	}
+
+	ireq6->iif = sk->sk_bound_dev_if;
+	/* So that link locals have meaning */
+	if (!sk->sk_bound_dev_if &&
+	    ipv6_addr_type(&ireq6->rmt_addr) & IPV6_ADDR_LINKLOCAL)
+		ireq6->iif = inet6_iif(skb);
+
+	req->expires = 0UL;
+	req->retrans = 0;
+	ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0;
+	ireq->wscale_ok = ireq->sack_ok = 0;
+	treq->rcv_isn = ntohl(th->seq) - 1;
+	treq->snt_isn = cookie;
+
+	/*
+	 * We need to lookup the dst_entry to get the correct window size.
+	 * This is taken from tcp_v6_syn_recv_sock.  Somebody please enlighten
+	 * me if there is a preferred way.
+	 */
+	{
+		struct in6_addr *final_p = NULL, final;
+		struct flowi fl;
+		memset(&fl, 0, sizeof(fl));
+		fl.proto = IPPROTO_TCP;
+		ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr);
+		if (np->opt && np->opt->srcrt) {
+			struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt;
+			ipv6_addr_copy(&final, &fl.fl6_dst);
+			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
+			final_p = &final;
+		}
+		ipv6_addr_copy(&fl.fl6_src, &ireq6->loc_addr);
+		fl.oif = sk->sk_bound_dev_if;
+		fl.fl_ip_dport = inet_rsk(req)->rmt_port;
+		fl.fl_ip_sport = inet_sk(sk)->sport;
+		security_req_classify_flow(req, &fl);
+		if (ip6_dst_lookup(sk, &dst, &fl)) {
+			reqsk_free(req);
+			goto out;
+		}
+		if (final_p)
+			ipv6_addr_copy(&fl.fl6_dst, final_p);
+		if ((xfrm_lookup(&dst, &fl, sk, 0)) < 0)
+			goto out;
+	}
+
+	req->window_clamp = dst_metric(dst, RTAX_WINDOW);
+	tcp_select_initial_window(tcp_full_space(sk), req->mss,
+				  &req->rcv_wnd, &req->window_clamp,
+				  0, &rcv_wscale);
+
+	ireq->rcv_wscale = rcv_wscale;
+
+	ret = get_cookie_sock(sk, skb, req, dst);
+
+out:	return ret;
+}
+
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 1cbbb87dbad2..fd773ac7531a 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -512,6 +512,20 @@ done:
 	return err;
 }
 
+static inline void syn_flood_warning(struct sk_buff *skb)
+{
+#ifdef CONFIG_SYN_COOKIES
+	if (sysctl_tcp_syncookies)
+		printk(KERN_INFO
+		       "TCPv6: Possible SYN flooding on port %d. "
+		       "Sending cookies.\n", ntohs(tcp_hdr(skb)->dest));
+	else
+#endif
+		printk(KERN_INFO
+		       "TCPv6: Possible SYN flooding on port %d. "
+		       "Dropping request.\n", ntohs(tcp_hdr(skb)->dest));
+}
+
 static void tcp_v6_reqsk_destructor(struct request_sock *req)
 {
 	if (inet6_rsk(req)->pktopts)
@@ -915,7 +929,7 @@ done_opts:
 }
 #endif
 
-static struct request_sock_ops tcp6_request_sock_ops __read_mostly = {
+struct request_sock_ops tcp6_request_sock_ops __read_mostly = {
 	.family		=	AF_INET6,
 	.obj_size	=	sizeof(struct tcp6_request_sock),
 	.rtx_syn_ack	=	tcp_v6_send_synack,
@@ -1213,9 +1227,9 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb)
 		return NULL;
 	}
 
-#if 0 /*def CONFIG_SYN_COOKIES*/
+#ifdef CONFIG_SYN_COOKIES
 	if (!th->rst && !th->syn && th->ack)
-		sk = cookie_v6_check(sk, skb, &(IPCB(skb)->opt));
+		sk = cookie_v6_check(sk, skb);
 #endif
 	return sk;
 }
@@ -1231,6 +1245,11 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct request_sock *req = NULL;
 	__u32 isn = TCP_SKB_CB(skb)->when;
+#ifdef CONFIG_SYN_COOKIES
+	int want_cookie = 0;
+#else
+#define want_cookie 0
+#endif
 
 	if (skb->protocol == htons(ETH_P_IP))
 		return tcp_v4_conn_request(sk, skb);
@@ -1238,12 +1257,14 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 	if (!ipv6_unicast_destination(skb))
 		goto drop;
 
-	/*
-	 *	There are no SYN attacks on IPv6, yet...
-	 */
 	if (inet_csk_reqsk_queue_is_full(sk) && !isn) {
 		if (net_ratelimit())
-			printk(KERN_INFO "TCPv6: dropping request, synflood is possible\n");
+			syn_flood_warning(skb);
+#ifdef CONFIG_SYN_COOKIES
+		if (sysctl_tcp_syncookies)
+			want_cookie = 1;
+		else
+#endif
 		goto drop;
 	}
 
@@ -1264,29 +1285,39 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	tcp_parse_options(skb, &tmp_opt, 0);
 
+	if (want_cookie) {
+		tcp_clear_options(&tmp_opt);
+		tmp_opt.saw_tstamp = 0;
+	}
+
 	tmp_opt.tstamp_ok = tmp_opt.saw_tstamp;
 	tcp_openreq_init(req, &tmp_opt, skb);
 
 	treq = inet6_rsk(req);
 	ipv6_addr_copy(&treq->rmt_addr, &ipv6_hdr(skb)->saddr);
 	ipv6_addr_copy(&treq->loc_addr, &ipv6_hdr(skb)->daddr);
-	TCP_ECN_create_request(req, tcp_hdr(skb));
 	treq->pktopts = NULL;
-	if (ipv6_opt_accepted(sk, skb) ||
-	    np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo ||
-	    np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) {
-		atomic_inc(&skb->users);
-		treq->pktopts = skb;
-	}
-	treq->iif = sk->sk_bound_dev_if;
+	if (!want_cookie)
+		TCP_ECN_create_request(req, tcp_hdr(skb));
+
+	if (want_cookie) {
+		isn = cookie_v6_init_sequence(sk, skb, &req->mss);
+	} else if (!isn) {
+		if (ipv6_opt_accepted(sk, skb) ||
+		    np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo ||
+		    np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) {
+			atomic_inc(&skb->users);
+			treq->pktopts = skb;
+		}
+		treq->iif = sk->sk_bound_dev_if;
 
-	/* So that link locals have meaning */
-	if (!sk->sk_bound_dev_if &&
-	    ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL)
-		treq->iif = inet6_iif(skb);
+		/* So that link locals have meaning */
+		if (!sk->sk_bound_dev_if &&
+		    ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL)
+			treq->iif = inet6_iif(skb);
 
-	if (isn == 0)
 		isn = tcp_v6_init_sequence(skb);
+	}
 
 	tcp_rsk(req)->snt_isn = isn;
 
@@ -1295,8 +1326,10 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 	if (tcp_v6_send_synack(sk, req))
 		goto drop;
 
-	inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT);
-	return 0;
+	if (!want_cookie) {
+		inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT);
+		return 0;
+	}
 
 drop:
 	if (req)
-- 
cgit v1.2.3


From e898d4db2749c6052072e9bc4448e396cbdeb06a Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 1 Mar 2008 01:06:47 +0900
Subject: [UDP]: Allow users to configure UDP-Lite.

Let's give users an option for disabling UDP-Lite (~4K).

old:
|    text	   data	    bss	    dec	    hex	filename
|  286498	  12432	   6072	 305002	  4a76a	net/ipv4/built-in.o
|  193830	   8192	   3204	 205226	  321aa	net/ipv6/ipv6.o

new (without UDP-Lite):
|    text	   data	    bss	    dec	    hex	filename
|  284086	  12136	   5432	 301654	  49a56	net/ipv4/built-in.o
|  191835	   7832	   3076	 202743	  317f7	net/ipv6/ipv6.o

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/udp.h      | 11 +++++++++++
 include/net/ipv6.h       |  5 +++++
 include/net/transp_v6.h  |  5 +++++
 include/net/udplite.h    |  9 +++++++--
 net/ipv4/Kconfig         | 10 ++++++++++
 net/ipv4/Makefile        |  3 ++-
 net/ipv4/af_inet.c       |  7 ++++++-
 net/ipv4/proc.c          |  5 ++++-
 net/ipv4/udp.c           | 24 ++++++++++++++----------
 net/ipv6/Makefile        |  3 ++-
 net/ipv6/af_inet6.c      | 14 ++++++++++++++
 net/ipv6/ipv6_sockglue.c |  6 +++++-
 net/ipv6/proc.c          |  6 ++++++
 net/ipv6/udp.c           | 16 ++++++++--------
 14 files changed, 99 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 8ec703f462da..4144664d69d9 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -70,8 +70,10 @@ struct udp_sock {
 #define UDPLITE_BIT      0x1  		/* set by udplite proto init function */
 #define UDPLITE_SEND_CC  0x2  		/* set via udplite setsockopt         */
 #define UDPLITE_RECV_CC  0x4		/* set via udplite setsocktopt        */
+#ifdef CONFIG_IP_UDPLITE
 	__u8		 pcflag;        /* marks socket as UDP-Lite if > 0    */
 	__u8		 unused[3];
+#endif
 	/*
 	 * For encapsulation sockets.
 	 */
@@ -82,7 +84,16 @@ static inline struct udp_sock *udp_sk(const struct sock *sk)
 {
 	return (struct udp_sock *)sk;
 }
+
+#ifdef CONFIG_IP_UDPLITE
 #define IS_UDPLITE(__sk) (udp_sk(__sk)->pcflag)
+#define IS_PROTO_UDPLITE(__proto) ((__proto) == IPPROTO_UDPLITE)
+#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP || (level) == SOL_UDPLITE)
+#else
+#define IS_UDPLITE(__sk) 0
+#define IS_PROTO_UDPLITE(__proto) 0
+#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP)
+#endif
 
 #endif
 
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 8b05c65415cb..96b1763bfcaa 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -603,8 +603,13 @@ extern int  tcp6_proc_init(void);
 extern void tcp6_proc_exit(void);
 extern int  udp6_proc_init(void);
 extern void udp6_proc_exit(void);
+#ifdef CONFIG_IP_UDPLITE
 extern int  udplite6_proc_init(void);
 extern void udplite6_proc_exit(void);
+#else
+static inline int udplite6_proc_init(void) { return 0; }
+static inline void udplite6_proc_exit(void) { }
+#endif
 extern int  ipv6_misc_proc_init(void);
 extern void ipv6_misc_proc_exit(void);
 extern int snmp6_register_dev(struct inet6_dev *idev);
diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h
index 27394e0447d8..902e6c6bc793 100644
--- a/include/net/transp_v6.h
+++ b/include/net/transp_v6.h
@@ -27,8 +27,13 @@ extern int				rawv6_init(void);
 extern void				rawv6_exit(void);
 extern int				udpv6_init(void);
 extern void				udpv6_exit(void);
+#ifdef CONFIG_IP_UDPLITE
 extern int 				udplitev6_init(void);
 extern void 				udplitev6_exit(void);
+#else
+static inline int udplitev6_init(void) { return 0; }
+static inline void udplitev6_exit(void) { }
+#endif
 extern int				tcpv6_init(void);
 extern void				tcpv6_exit(void);
 
diff --git a/include/net/udplite.h b/include/net/udplite.h
index b76b2e377af4..01ddb2c20264 100644
--- a/include/net/udplite.h
+++ b/include/net/udplite.h
@@ -25,7 +25,9 @@ static __inline__ int udplite_getfrag(void *from, char *to, int  offset,
 /* Designate sk as UDP-Lite socket */
 static inline int udplite_sk_init(struct sock *sk)
 {
+#ifdef CONFIG_IP_UDPLITE
 	udp_sk(sk)->pcflag = UDPLITE_BIT;
+#endif
 	return 0;
 }
 
@@ -69,7 +71,7 @@ static inline int udplite_checksum_init(struct sk_buff *skb, struct udphdr *uh)
 static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh)
 {
 	int cscov = up->len;
-
+#ifdef CONFIG_IP_UDPLITE
 	/*
 	 * Sender has set `partial coverage' option on UDP-Lite socket
 	 */
@@ -93,13 +95,15 @@ static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh)
 	 *       illegal, we fall back to the defaults here.
 	 */
 	}
+#endif
 	return cscov;
 }
 
 static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb)
 {
-	int cscov = udplite_sender_cscov(udp_sk(sk), udp_hdr(skb));
 	__wsum csum = 0;
+#ifdef CONFIG_IP_UDPLITE
+	int cscov = udplite_sender_cscov(udp_sk(sk), udp_hdr(skb));
 
 	skb->ip_summed = CHECKSUM_NONE;     /* no HW support for checksumming */
 
@@ -112,6 +116,7 @@ static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb)
 		if ((cscov -= len) <= 0)
 			break;
 	}
+#endif
 	return csum;
 }
 
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 19880b086e71..efe3832c4ad8 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -632,5 +632,15 @@ config TCP_MD5SIG
 
 	  If unsure, say N.
 
+config IP_UDPLITE
+	bool "IP: UDP-Lite Protocol (RFC 3828)"
+	default n
+	---help---
+	  UDP-Lite (RFC 3828) is a UDP-like protocol with variable-length
+	  checksum.  Read <file:Documentation/networking/udplite.txt> for
+	  details.
+
+	  If unsure, say N.
+
 source "net/ipv4/ipvs/Kconfig"
 
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index ad40ef3f9ebc..e88cebdf3e30 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -8,7 +8,7 @@ obj-y     := route.o inetpeer.o protocol.o \
 	     inet_timewait_sock.o inet_connection_sock.o \
 	     tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \
 	     tcp_minisocks.o tcp_cong.o \
-	     datagram.o raw.o udp.o udplite.o \
+	     datagram.o raw.o udp.o \
 	     arp.o icmp.o devinet.o af_inet.o  igmp.o \
 	     fib_frontend.o fib_semantics.o \
 	     inet_fragment.o
@@ -49,6 +49,7 @@ obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
 obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
 obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
 obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
+obj-$(CONFIG_IP_UDPLITE) += udplite.o
 obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
 
 obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 4f539bd48718..67260c0eaaa8 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1317,15 +1317,18 @@ static int __init init_ipv4_mibs(void)
 	if (snmp_mib_init((void **)udp_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udp_mib;
+#ifdef CONFIG_IP_UDPLITE
 	if (snmp_mib_init((void **)udplite_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udplite_mib;
-
+#endif
 	tcp_mib_init();
 
 	return 0;
 
+#ifdef CONFIG_IP_UDPLITE
 err_udplite_mib:
+#endif
 	snmp_mib_free((void **)udp_statistics);
 err_udp_mib:
 	snmp_mib_free((void **)tcp_statistics);
@@ -1423,8 +1426,10 @@ static int __init inet_init(void)
 	/* Setup UDP memory threshold */
 	udp_init();
 
+#ifdef CONFIG_IP_UDPLITE
 	/* Add UDP-Lite (RFC 3828) */
 	udplite4_register();
+#endif
 
 	/*
 	 *	Set the ICMP layer up
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index d63474c6b400..d75ddb7fa4b8 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -59,7 +59,9 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 		   atomic_read(&tcp_memory_allocated));
 	seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse_get(&udp_prot),
 		   atomic_read(&udp_memory_allocated));
+#ifdef CONFIG_IP_UDPLITE
 	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot));
+#endif
 	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot));
 	seq_printf(seq,  "FRAG: inuse %d memory %d\n",
 			ip_frag_nqueues(&init_net), ip_frag_mem(&init_net));
@@ -349,6 +351,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 			   snmp_fold_field((void **)udp_statistics,
 					   snmp4_udp_list[i].entry));
 
+#ifdef CONFIG_IP_UDPLITE
 	/* the UDP and UDP-Lite MIBs are the same */
 	seq_puts(seq, "\nUdpLite:");
 	for (i = 0; snmp4_udp_list[i].name != NULL; i++)
@@ -359,7 +362,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 		seq_printf(seq, " %lu",
 			   snmp_fold_field((void **)udplite_statistics,
 					   snmp4_udp_list[i].entry));
-
+#endif
 	seq_putc(seq, '\n');
 	return 0;
 }
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 7ea1b67b6de1..acc353aa89eb 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1127,7 +1127,7 @@ static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh,
 	UDP_SKB_CB(skb)->partial_cov = 0;
 	UDP_SKB_CB(skb)->cscov = skb->len;
 
-	if (proto == IPPROTO_UDPLITE) {
+	if (IS_PROTO_UDPLITE(proto)) {
 		err = udplite_checksum_init(skb, uh);
 		if (err)
 			return err;
@@ -1175,7 +1175,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	if (ulen > skb->len)
 		goto short_packet;
 
-	if (proto == IPPROTO_UDP) {
+	if (IS_PROTO_UDPLITE(proto)) {
 		/* UDP validates ulen. */
 		if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
 			goto short_packet;
@@ -1217,7 +1217,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	if (udp_lib_checksum_complete(skb))
 		goto csum_error;
 
-	UDP_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+	UDP_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto));
 	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
 
 	/*
@@ -1229,7 +1229,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 
 short_packet:
 	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n",
-		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
+		       IS_PROTO_UDPLITE(proto) ? "-Lite" : "",
 		       NIPQUAD(saddr),
 		       ntohs(uh->source),
 		       ulen,
@@ -1244,14 +1244,14 @@ csum_error:
 	 * the network is concerned, anyway) as per 4.1.3.4 (MUST).
 	 */
 	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n",
-		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
+		       IS_PROTO_UDPLITE(proto) ? "-Lite" : "",
 		       NIPQUAD(saddr),
 		       ntohs(uh->source),
 		       NIPQUAD(daddr),
 		       ntohs(uh->dest),
 		       ulen);
 drop:
-	UDP_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+	UDP_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto));
 	kfree_skb(skb);
 	return 0;
 }
@@ -1279,7 +1279,9 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 	struct udp_sock *up = udp_sk(sk);
 	int val;
 	int err = 0;
+#ifdef CONFIG_IP_UDPLITE
 	int is_udplite = IS_UDPLITE(sk);
+#endif
 
 	if (optlen<sizeof(int))
 		return -EINVAL;
@@ -1315,6 +1317,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 		}
 		break;
 
+#ifdef CONFIG_IP_UDPLITE
 	/*
 	 * 	UDP-Lite's partial checksum coverage (RFC 3828).
 	 */
@@ -1340,6 +1343,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 		up->pcrlen = val;
 		up->pcflag |= UDPLITE_RECV_CC;
 		break;
+#endif
 
 	default:
 		err = -ENOPROTOOPT;
@@ -1352,7 +1356,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 int udp_setsockopt(struct sock *sk, int level, int optname,
 		   char __user *optval, int optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
 					  udp_push_pending_frames);
 	return ip_setsockopt(sk, level, optname, optval, optlen);
@@ -1362,7 +1366,7 @@ int udp_setsockopt(struct sock *sk, int level, int optname,
 int compat_udp_setsockopt(struct sock *sk, int level, int optname,
 			  char __user *optval, int optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
 					  udp_push_pending_frames);
 	return compat_ip_setsockopt(sk, level, optname, optval, optlen);
@@ -1416,7 +1420,7 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname,
 int udp_getsockopt(struct sock *sk, int level, int optname,
 		   char __user *optval, int __user *optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
 	return ip_getsockopt(sk, level, optname, optval, optlen);
 }
@@ -1425,7 +1429,7 @@ int udp_getsockopt(struct sock *sk, int level, int optname,
 int compat_udp_getsockopt(struct sock *sk, int level, int optname,
 				 char __user *optval, int __user *optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
 	return compat_ip_getsockopt(sk, level, optname, optval, optlen);
 }
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index ae14617e607f..81969479955f 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -6,7 +6,7 @@ obj-$(CONFIG_IPV6) += ipv6.o
 
 ipv6-objs :=	af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \
 		addrlabel.o \
-		route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \
+		route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o \
 		raw.o protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \
 		exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o
 
@@ -17,6 +17,7 @@ ipv6-$(CONFIG_NETFILTER) += netfilter.o
 ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o
 ipv6-$(CONFIG_PROC_FS) += proc.o
 ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o
+ipv6-$(CONFIG_IP_UDPLITE) += udplite.o
 
 ipv6-objs += $(ipv6-y)
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 9869f87243cf..243c42a6b80d 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -691,12 +691,16 @@ static int __init init_ipv6_mibs(void)
 		goto err_icmpmsg_mib;
 	if (snmp_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib)) < 0)
 		goto err_udp_mib;
+#ifdef CONFIG_IP_UDPLITE
 	if (snmp_mib_init((void **)udplite_stats_in6,
 			  sizeof (struct udp_mib)) < 0)
 		goto err_udplite_mib;
+#endif
 	return 0;
 
+#ifdef CONFIG_IP_UDPLITE
 err_udplite_mib:
+#endif
 	snmp_mib_free((void **)udp_stats_in6);
 err_udp_mib:
 	snmp_mib_free((void **)icmpv6msg_statistics);
@@ -715,7 +719,9 @@ static void cleanup_ipv6_mibs(void)
 	snmp_mib_free((void **)icmpv6_statistics);
 	snmp_mib_free((void **)icmpv6msg_statistics);
 	snmp_mib_free((void **)udp_stats_in6);
+#ifdef CONFIG_IP_UDPLITE
 	snmp_mib_free((void **)udplite_stats_in6);
+#endif
 }
 
 static int inet6_net_init(struct net *net)
@@ -760,9 +766,11 @@ static int __init inet6_init(void)
 	if (err)
 		goto out_unregister_tcp_proto;
 
+#ifdef CONFIG_IP_UDPLITE
 	err = proto_register(&udplitev6_prot, 1);
 	if (err)
 		goto out_unregister_udp_proto;
+#endif
 
 	err = proto_register(&rawv6_prot, 1);
 	if (err)
@@ -933,8 +941,10 @@ out_sock_register_fail:
 out_unregister_raw_proto:
 	proto_unregister(&rawv6_prot);
 out_unregister_udplite_proto:
+#ifdef CONFIG_IP_UDPLITE
 	proto_unregister(&udplitev6_prot);
 out_unregister_udp_proto:
+#endif
 	proto_unregister(&udpv6_prot);
 out_unregister_tcp_proto:
 	proto_unregister(&tcpv6_prot);
@@ -950,7 +960,9 @@ static void __exit inet6_exit(void)
 	rtnl_unregister_all(PF_INET6);
 
 	udpv6_exit();
+#ifdef CONFIG_IP_UDPLITE
 	udplitev6_exit();
+#endif
 	tcpv6_exit();
 
 	/* Cleanup code parts. */
@@ -982,7 +994,9 @@ static void __exit inet6_exit(void)
 	unregister_pernet_subsys(&inet6_net_ops);
 	cleanup_ipv6_mibs();
 	proto_unregister(&rawv6_prot);
+#ifdef CONFIG_IP_UDPLITE
 	proto_unregister(&udplitev6_prot);
+#endif
 	proto_unregister(&udpv6_prot);
 	proto_unregister(&tcpv6_prot);
 }
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index bf2a686aa13d..0a18fecb93d1 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -239,7 +239,9 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 			struct sk_buff *pktopt;
 
 			if (sk->sk_protocol != IPPROTO_UDP &&
+#ifdef CONFIG_IP_UDPLITE
 			    sk->sk_protocol != IPPROTO_UDPLITE &&
+#endif
 			    sk->sk_protocol != IPPROTO_TCP)
 				break;
 
@@ -279,7 +281,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 			} else {
 				struct proto *prot = &udp_prot;
 
-				if (sk->sk_protocol == IPPROTO_UDPLITE)
+				if (IS_PROTO_UDPLITE(sk->sk_protocol))
 					prot = &udplite_prot;
 				local_bh_disable();
 				sock_prot_inuse_add(sk->sk_prot, -1);
@@ -844,7 +846,9 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
 	switch (optname) {
 	case IPV6_ADDRFORM:
 		if (sk->sk_protocol != IPPROTO_UDP &&
+#ifdef CONFIG_IP_UDPLITE
 		    sk->sk_protocol != IPPROTO_UDPLITE &&
+#endif
 		    sk->sk_protocol != IPPROTO_TCP)
 			return -EINVAL;
 		if (sk->sk_state != TCP_ESTABLISHED)
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 199ef379e501..5ba7ae849d04 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -39,8 +39,10 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v)
 		       sock_prot_inuse_get(&tcpv6_prot));
 	seq_printf(seq, "UDP6: inuse %d\n",
 		       sock_prot_inuse_get(&udpv6_prot));
+#ifdef CONFIG_IP_UDPLITE
 	seq_printf(seq, "UDPLITE6: inuse %d\n",
 			sock_prot_inuse_get(&udplitev6_prot));
+#endif
 	seq_printf(seq, "RAW6: inuse %d\n",
 		       sock_prot_inuse_get(&rawv6_prot));
 	seq_printf(seq, "FRAG6: inuse %d memory %d\n",
@@ -111,6 +113,7 @@ static struct snmp_mib snmp6_udp6_list[] = {
 	SNMP_MIB_SENTINEL
 };
 
+#ifdef CONFIG_IP_UDPLITE
 static struct snmp_mib snmp6_udplite6_list[] = {
 	SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS),
 	SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS),
@@ -118,6 +121,7 @@ static struct snmp_mib snmp6_udplite6_list[] = {
 	SNMP_MIB_ITEM("UdpLite6OutDatagrams", UDP_MIB_OUTDATAGRAMS),
 	SNMP_MIB_SENTINEL
 };
+#endif
 
 static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib)
 {
@@ -176,7 +180,9 @@ static int snmp6_seq_show(struct seq_file *seq, void *v)
 		snmp6_seq_show_item(seq, (void **)icmpv6_statistics, snmp6_icmp6_list);
 		snmp6_seq_show_icmpv6msg(seq, (void **)icmpv6msg_statistics);
 		snmp6_seq_show_item(seq, (void **)udp_stats_in6, snmp6_udp6_list);
+#ifdef CONFIG_IP_UDPLITE
 		snmp6_seq_show_item(seq, (void **)udplite_stats_in6, snmp6_udplite6_list);
+#endif
 	}
 	return 0;
 }
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 53739de829db..55feac7ba717 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -400,7 +400,7 @@ static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh,
 	UDP_SKB_CB(skb)->partial_cov = 0;
 	UDP_SKB_CB(skb)->cscov = skb->len;
 
-	if (proto == IPPROTO_UDPLITE) {
+	if (IS_PROTO_UDPLITE(proto)) {
 		err = udplite_checksum_init(skb, uh);
 		if (err)
 			return err;
@@ -489,7 +489,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 
 		if (udp_lib_checksum_complete(skb))
 			goto discard;
-		UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+		UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto));
 
 		icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev);
 
@@ -510,11 +510,11 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 
 short_packet:
 	LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n",
-		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
+		       IS_PROTO_UDPLITE(proto) ? "-Lite" : "",
 		       ulen, skb->len);
 
 discard:
-	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto));
 	kfree_skb(skb);
 	return 0;
 }
@@ -890,7 +890,7 @@ int udpv6_destroy_sock(struct sock *sk)
 int udpv6_setsockopt(struct sock *sk, int level, int optname,
 		     char __user *optval, int optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
 					  udp_v6_push_pending_frames);
 	return ipv6_setsockopt(sk, level, optname, optval, optlen);
@@ -900,7 +900,7 @@ int udpv6_setsockopt(struct sock *sk, int level, int optname,
 int compat_udpv6_setsockopt(struct sock *sk, int level, int optname,
 			    char __user *optval, int optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
 					  udp_v6_push_pending_frames);
 	return compat_ipv6_setsockopt(sk, level, optname, optval, optlen);
@@ -910,7 +910,7 @@ int compat_udpv6_setsockopt(struct sock *sk, int level, int optname,
 int udpv6_getsockopt(struct sock *sk, int level, int optname,
 		     char __user *optval, int __user *optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
 	return ipv6_getsockopt(sk, level, optname, optval, optlen);
 }
@@ -919,7 +919,7 @@ int udpv6_getsockopt(struct sock *sk, int level, int optname,
 int compat_udpv6_getsockopt(struct sock *sk, int level, int optname,
 			    char __user *optval, int __user *optlen)
 {
-	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+	if (IS_SOL_UDPFAMILY(level))
 		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
 	return compat_ipv6_getsockopt(sk, level, optname, optval, optlen);
 }
-- 
cgit v1.2.3


From 662397fd7aaa10afdbdc55a0bfdb7e9701454c27 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 27 Feb 2008 23:14:03 +0900
Subject: [IPV6]: Move packet_type{} related bits to af_inet6.c.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/ipv6.h       |   4 --
 net/ipv6/af_inet6.c      | 123 +++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/ipv6_sockglue.c | 122 ----------------------------------------------
 3 files changed, 123 insertions(+), 126 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 96b1763bfcaa..5dc8164e5d3b 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -555,10 +555,6 @@ extern int			compat_ipv6_getsockopt(struct sock *sk,
 						char __user *optval,
 						int __user *optlen);
 
-extern int			ipv6_packet_init(void);
-
-extern void			ipv6_packet_cleanup(void);
-
 extern int			ip6_datagram_connect(struct sock *sk, 
 						     struct sockaddr *addr, int addr_len);
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 243c42a6b80d..73021d5baece 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -678,6 +678,129 @@ int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb)
 
 EXPORT_SYMBOL_GPL(ipv6_opt_accepted);
 
+static struct inet6_protocol *ipv6_gso_pull_exthdrs(struct sk_buff *skb,
+						    int proto)
+{
+	struct inet6_protocol *ops = NULL;
+
+	for (;;) {
+		struct ipv6_opt_hdr *opth;
+		int len;
+
+		if (proto != NEXTHDR_HOP) {
+			ops = rcu_dereference(inet6_protos[proto]);
+
+			if (unlikely(!ops))
+				break;
+
+			if (!(ops->flags & INET6_PROTO_GSO_EXTHDR))
+				break;
+		}
+
+		if (unlikely(!pskb_may_pull(skb, 8)))
+			break;
+
+		opth = (void *)skb->data;
+		len = ipv6_optlen(opth);
+
+		if (unlikely(!pskb_may_pull(skb, len)))
+			break;
+
+		proto = opth->nexthdr;
+		__skb_pull(skb, len);
+	}
+
+	return ops;
+}
+
+static int ipv6_gso_send_check(struct sk_buff *skb)
+{
+	struct ipv6hdr *ipv6h;
+	struct inet6_protocol *ops;
+	int err = -EINVAL;
+
+	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
+		goto out;
+
+	ipv6h = ipv6_hdr(skb);
+	__skb_pull(skb, sizeof(*ipv6h));
+	err = -EPROTONOSUPPORT;
+
+	rcu_read_lock();
+	ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);
+	if (likely(ops && ops->gso_send_check)) {
+		skb_reset_transport_header(skb);
+		err = ops->gso_send_check(skb);
+	}
+	rcu_read_unlock();
+
+out:
+	return err;
+}
+
+static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, int features)
+{
+	struct sk_buff *segs = ERR_PTR(-EINVAL);
+	struct ipv6hdr *ipv6h;
+	struct inet6_protocol *ops;
+
+	if (!(features & NETIF_F_V6_CSUM))
+		features &= ~NETIF_F_SG;
+
+	if (unlikely(skb_shinfo(skb)->gso_type &
+		     ~(SKB_GSO_UDP |
+		       SKB_GSO_DODGY |
+		       SKB_GSO_TCP_ECN |
+		       SKB_GSO_TCPV6 |
+		       0)))
+		goto out;
+
+	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
+		goto out;
+
+	ipv6h = ipv6_hdr(skb);
+	__skb_pull(skb, sizeof(*ipv6h));
+	segs = ERR_PTR(-EPROTONOSUPPORT);
+
+	rcu_read_lock();
+	ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);
+	if (likely(ops && ops->gso_segment)) {
+		skb_reset_transport_header(skb);
+		segs = ops->gso_segment(skb, features);
+	}
+	rcu_read_unlock();
+
+	if (unlikely(IS_ERR(segs)))
+		goto out;
+
+	for (skb = segs; skb; skb = skb->next) {
+		ipv6h = ipv6_hdr(skb);
+		ipv6h->payload_len = htons(skb->len - skb->mac_len -
+					   sizeof(*ipv6h));
+	}
+
+out:
+	return segs;
+}
+
+static struct packet_type ipv6_packet_type = {
+	.type = __constant_htons(ETH_P_IPV6),
+	.func = ipv6_rcv,
+	.gso_send_check = ipv6_gso_send_check,
+	.gso_segment = ipv6_gso_segment,
+};
+
+static int __init ipv6_packet_init(void)
+{
+	dev_add_pack(&ipv6_packet_type);
+	return 0;
+}
+
+static void ipv6_packet_cleanup(void)
+{
+	dev_remove_pack(&ipv6_packet_type);
+}
+
 static int __init init_ipv6_mibs(void)
 {
 	if (snmp_mib_init((void **)ipv6_statistics,
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 0a18fecb93d1..3bbfdff698d2 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -57,118 +57,6 @@
 
 DEFINE_SNMP_STAT(struct ipstats_mib, ipv6_statistics) __read_mostly;
 
-static struct inet6_protocol *ipv6_gso_pull_exthdrs(struct sk_buff *skb,
-						    int proto)
-{
-	struct inet6_protocol *ops = NULL;
-
-	for (;;) {
-		struct ipv6_opt_hdr *opth;
-		int len;
-
-		if (proto != NEXTHDR_HOP) {
-			ops = rcu_dereference(inet6_protos[proto]);
-
-			if (unlikely(!ops))
-				break;
-
-			if (!(ops->flags & INET6_PROTO_GSO_EXTHDR))
-				break;
-		}
-
-		if (unlikely(!pskb_may_pull(skb, 8)))
-			break;
-
-		opth = (void *)skb->data;
-		len = opth->hdrlen * 8 + 8;
-
-		if (unlikely(!pskb_may_pull(skb, len)))
-			break;
-
-		proto = opth->nexthdr;
-		__skb_pull(skb, len);
-	}
-
-	return ops;
-}
-
-static int ipv6_gso_send_check(struct sk_buff *skb)
-{
-	struct ipv6hdr *ipv6h;
-	struct inet6_protocol *ops;
-	int err = -EINVAL;
-
-	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
-		goto out;
-
-	ipv6h = ipv6_hdr(skb);
-	__skb_pull(skb, sizeof(*ipv6h));
-	err = -EPROTONOSUPPORT;
-
-	rcu_read_lock();
-	ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);
-	if (likely(ops && ops->gso_send_check)) {
-		skb_reset_transport_header(skb);
-		err = ops->gso_send_check(skb);
-	}
-	rcu_read_unlock();
-
-out:
-	return err;
-}
-
-static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, int features)
-{
-	struct sk_buff *segs = ERR_PTR(-EINVAL);
-	struct ipv6hdr *ipv6h;
-	struct inet6_protocol *ops;
-
-	if (!(features & NETIF_F_V6_CSUM))
-		features &= ~NETIF_F_SG;
-
-	if (unlikely(skb_shinfo(skb)->gso_type &
-		     ~(SKB_GSO_UDP |
-		       SKB_GSO_DODGY |
-		       SKB_GSO_TCP_ECN |
-		       SKB_GSO_TCPV6 |
-		       0)))
-		goto out;
-
-	if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h))))
-		goto out;
-
-	ipv6h = ipv6_hdr(skb);
-	__skb_pull(skb, sizeof(*ipv6h));
-	segs = ERR_PTR(-EPROTONOSUPPORT);
-
-	rcu_read_lock();
-	ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr);
-	if (likely(ops && ops->gso_segment)) {
-		skb_reset_transport_header(skb);
-		segs = ops->gso_segment(skb, features);
-	}
-	rcu_read_unlock();
-
-	if (unlikely(IS_ERR(segs)))
-		goto out;
-
-	for (skb = segs; skb; skb = skb->next) {
-		ipv6h = ipv6_hdr(skb);
-		ipv6h->payload_len = htons(skb->len - skb->mac_len -
-					   sizeof(*ipv6h));
-	}
-
-out:
-	return segs;
-}
-
-static struct packet_type ipv6_packet_type = {
-	.type = __constant_htons(ETH_P_IPV6),
-	.func = ipv6_rcv,
-	.gso_send_check = ipv6_gso_send_check,
-	.gso_segment = ipv6_gso_segment,
-};
-
 struct ip6_ra_chain *ip6_ra_chain;
 DEFINE_RWLOCK(ip6_ra_lock);
 
@@ -1132,13 +1020,3 @@ int compat_ipv6_getsockopt(struct sock *sk, int level, int optname,
 EXPORT_SYMBOL(compat_ipv6_getsockopt);
 #endif
 
-int __init ipv6_packet_init(void)
-{
-	dev_add_pack(&ipv6_packet_type);
-	return 0;
-}
-
-void ipv6_packet_cleanup(void)
-{
-	dev_remove_pack(&ipv6_packet_type);
-}
-- 
cgit v1.2.3


From 0e7b8dcd16eb91b9cd8ecc07c4094512f20d7e3c Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 28 Feb 2008 17:03:12 +0900
Subject: [IPV6]: Use htonl() instead of __constant_htonl() where appricable.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 496503c03846..a9ff97c120c7 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -214,29 +214,25 @@ static inline void addrconf_addr_solict_mult(const struct in6_addr *addr,
 					     struct in6_addr *solicited)
 {
 	ipv6_addr_set(solicited,
-		      __constant_htonl(0xFF020000), 0,
-		      __constant_htonl(0x1),
-		      __constant_htonl(0xFF000000) | addr->s6_addr32[3]);
+		      htonl(0xFF020000), 0,
+		      htonl(0x1),
+		      htonl(0xFF000000) | addr->s6_addr32[3]);
 }
 
 
 static inline void ipv6_addr_all_nodes(struct in6_addr *addr)
 {
-	ipv6_addr_set(addr,
-		      __constant_htonl(0xFF020000), 0, 0,
-		      __constant_htonl(0x1));
+	ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x1));
 }
 
 static inline void ipv6_addr_all_routers(struct in6_addr *addr)
 {
-	ipv6_addr_set(addr,
-		      __constant_htonl(0xFF020000), 0, 0,
-		      __constant_htonl(0x2));
+	ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x2));
 }
 
 static inline int ipv6_addr_is_multicast(const struct in6_addr *addr)
 {
-	return (addr->s6_addr32[0] & __constant_htonl(0xFF000000)) == __constant_htonl(0xFF000000);
+	return (addr->s6_addr32[0] & htonl(0xFF000000)) == htonl(0xFF000000);
 }
 
 static inline int ipv6_addr_is_ll_all_nodes(const struct in6_addr *addr)
-- 
cgit v1.2.3


From 8082c37cdc31fb0ed178d9d706bf7568ada0edd9 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Tue, 4 Mar 2008 14:55:03 +0900
Subject: [NET] NEIGHBOUR: Remove unpopular neigh_is_connected().

neigh_is_connected() is not popular at all, and the only user
drivers/net/cxgb3/l2t.c:t3_l2t_update() also have raw (expanded) expression.
Let's expand it and remove the inline function.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 drivers/net/cxgb3/l2t.c | 2 +-
 include/net/neighbour.h | 6 ------
 2 files changed, 1 insertion(+), 7 deletions(-)

(limited to 'include')

diff --git a/drivers/net/cxgb3/l2t.c b/drivers/net/cxgb3/l2t.c
index 865faee53e17..f510140885ae 100644
--- a/drivers/net/cxgb3/l2t.c
+++ b/drivers/net/cxgb3/l2t.c
@@ -407,7 +407,7 @@ found:
 			} else if (neigh->nud_state & (NUD_CONNECTED|NUD_STALE))
 				setup_l2e_send_pending(dev, NULL, e);
 		} else {
-			e->state = neigh_is_connected(neigh) ?
+			e->state = neigh->nud_state & NUD_CONNECTED ?
 			    L2T_STATE_VALID : L2T_STATE_STALE;
 			if (memcmp(e->dmac, neigh->ha, 6))
 				setup_l2e_send_pending(dev, NULL, e);
diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index ebbfb509822e..062281872064 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -284,12 +284,6 @@ static inline void neigh_confirm(struct neighbour *neigh)
 		neigh->confirmed = jiffies;
 }
 
-static inline int neigh_is_connected(struct neighbour *neigh)
-{
-	return neigh->nud_state&NUD_CONNECTED;
-}
-
-
 static inline int neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
 {
 	neigh->used = jiffies;
-- 
cgit v1.2.3


From 5e5f3f0f801321078c897a5de0b4b4304f234da0 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Mon, 3 Mar 2008 21:44:34 +0900
Subject: [IPV6] ADDRCONF: Convert ipv6_get_saddr() to ipv6_dev_get_saddr().

Since most users of ipv6_get_saddr() pass non-NULL as
dst argument, use ipv6_dev_get_saddr() directly.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h  | 3 ---
 net/ipv6/addrconf.c     | 9 +--------
 net/ipv6/fib6_rules.c   | 4 ++--
 net/ipv6/ip6_output.c   | 3 ++-
 net/ipv6/route.c        | 3 ++-
 net/ipv6/xfrm6_policy.c | 5 +++--
 net/sctp/ipv6.c         | 3 ++-
 7 files changed, 12 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index a9ff97c120c7..89e3c53c8886 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -73,9 +73,6 @@ extern struct inet6_ifaddr      *ipv6_get_ifaddr(struct net *net,
 						 struct net_device *dev,
 						 int strict);
 
-extern int			ipv6_get_saddr(struct dst_entry *dst, 
-					       struct in6_addr *daddr,
-					       struct in6_addr *saddr);
 extern int			ipv6_dev_get_saddr(struct net_device *dev, 
 					       struct in6_addr *daddr,
 					       struct in6_addr *saddr);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 18e3a9825d81..9b3a2d0e4269 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1167,14 +1167,7 @@ record_it:
 	return 0;
 }
 
-
-int ipv6_get_saddr(struct dst_entry *dst,
-		   struct in6_addr *daddr, struct in6_addr *saddr)
-{
-	return ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, daddr, saddr);
-}
-
-EXPORT_SYMBOL(ipv6_get_saddr);
+EXPORT_SYMBOL(ipv6_dev_get_saddr);
 
 int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr,
 		    unsigned char banned_flags)
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 695c0ca8a417..157db3a1ce00 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -85,8 +85,8 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 		if ((rule->flags & FIB_RULE_FIND_SADDR) &&
 		    r->src.plen && !(flags & RT6_LOOKUP_F_HAS_SADDR)) {
 			struct in6_addr saddr;
-			if (ipv6_get_saddr(&rt->u.dst, &flp->fl6_dst,
-					   &saddr))
+			if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev,
+					       &flp->fl6_dst, &saddr))
 				goto again;
 			if (!ipv6_prefix_equal(&saddr, &r->src.addr,
 					       r->src.plen))
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index a59d259cf97e..ff3971173e1e 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -919,7 +919,8 @@ static int ip6_dst_lookup_tail(struct sock *sk,
 		goto out_err_release;
 
 	if (ipv6_addr_any(&fl->fl6_src)) {
-		err = ipv6_get_saddr(*dst, &fl->fl6_dst, &fl->fl6_src);
+		err = ipv6_dev_get_saddr(ip6_dst_idev(*dst)->dev,
+					 &fl->fl6_dst, &fl->fl6_src);
 		if (err)
 			goto out_err_release;
 	}
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 1b15e1708409..6abe7da45ef7 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2122,7 +2122,8 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 		NLA_PUT_U32(skb, RTA_IIF, iif);
 	else if (dst) {
 		struct in6_addr saddr_buf;
-		if (ipv6_get_saddr(&rt->u.dst, dst, &saddr_buf) == 0)
+		if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev,
+				       dst, &saddr_buf) == 0)
 			NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf);
 	}
 
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 7d20199ee1f3..6ef56303e69e 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -57,8 +57,9 @@ static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
 	if (IS_ERR(dst))
 		return -EHOSTUNREACH;
 
-	ipv6_get_saddr(dst, (struct in6_addr *)&daddr->a6,
-		       (struct in6_addr *)&saddr->a6);
+	ipv6_dev_get_saddr(ip6_dst_idev(dst)->dev,
+			   (struct in6_addr *)&daddr->a6,
+			   (struct in6_addr *)&saddr->a6);
 	dst_release(dst);
 	return 0;
 }
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index 87f940587d5f..3e4878800b36 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -313,7 +313,8 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc,
 			  __FUNCTION__, asoc, dst, NIP6(daddr->v6.sin6_addr));
 
 	if (!asoc) {
-		ipv6_get_saddr(dst, &daddr->v6.sin6_addr,&saddr->v6.sin6_addr);
+		ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL,
+				   &daddr->v6.sin6_addr, &saddr->v6.sin6_addr);
 		SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " NIP6_FMT "\n",
 				  NIP6(saddr->v6.sin6_addr));
 		return;
-- 
cgit v1.2.3


From 95e41e93e18d8e1e272ce23d96bae4f17ce11d42 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 6 Dec 2007 15:43:30 -0800
Subject: [IPV6]: Make ndisc_flow_init() common for later use.

For later use, this patch is renaming ndisc_flow_init() to
icmpv6_flow_init() and putting it in common place.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/icmpv6.h |  8 ++++++++
 net/ipv6/icmp.c        | 16 ++++++++++++++++
 net/ipv6/ndisc.c       | 23 ++++-------------------
 3 files changed, 28 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h
index 8f86d6b621c8..e4d4300d768f 100644
--- a/include/linux/icmpv6.h
+++ b/include/linux/icmpv6.h
@@ -182,6 +182,14 @@ extern int				icmpv6_err_convert(int type, int code,
 extern void				icmpv6_cleanup(void);
 extern void				icmpv6_param_prob(struct sk_buff *skb,
 							  int code, int pos);
+
+struct flowi;
+extern void				icmpv6_flow_init(struct sock *sk,
+							 struct flowi *fl,
+							 u8 type,
+							 const struct in6_addr *saddr,
+							 const struct in6_addr *daddr,
+							 int oif);
 #endif
 
 #endif
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 12c0b85d6c46..cff74127ea32 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -777,6 +777,22 @@ drop_no_count:
 	return 0;
 }
 
+void icmpv6_flow_init(struct sock *sk, struct flowi *fl,
+		      u8 type,
+		      const struct in6_addr *saddr,
+		      const struct in6_addr *daddr,
+		      int oif)
+{
+	memset(fl, 0, sizeof(*fl));
+	ipv6_addr_copy(&fl->fl6_src, saddr);
+	ipv6_addr_copy(&fl->fl6_dst, daddr);
+	fl->proto	 	= IPPROTO_ICMPV6;
+	fl->fl_icmp_type	= type;
+	fl->fl_icmp_code	= 0;
+	fl->oif			= oif;
+	security_sk_classify_flow(sk, fl);
+}
+
 /*
  * Special lock-class for __icmpv6_sk:
  */
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 1fc33c8c7232..8db5f4a419aa 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -441,21 +441,6 @@ static void pndisc_destructor(struct pneigh_entry *n)
 /*
  *	Send a Neighbour Advertisement
  */
-
-static inline void ndisc_flow_init(struct flowi *fl, u8 type,
-			    struct in6_addr *saddr, struct in6_addr *daddr,
-			    int oif)
-{
-	memset(fl, 0, sizeof(*fl));
-	ipv6_addr_copy(&fl->fl6_src, saddr);
-	ipv6_addr_copy(&fl->fl6_dst, daddr);
-	fl->proto	 	= IPPROTO_ICMPV6;
-	fl->fl_icmp_type	= type;
-	fl->fl_icmp_code	= 0;
-	fl->oif			= oif;
-	security_sk_classify_flow(ndisc_socket->sk, fl);
-}
-
 static void __ndisc_send(struct net_device *dev,
 			 struct neighbour *neigh,
 			 struct in6_addr *daddr, struct in6_addr *saddr,
@@ -474,8 +459,8 @@ static void __ndisc_send(struct net_device *dev,
 
 	type = icmp6h->icmp6_type;
 
-	ndisc_flow_init(&fl, type, saddr, daddr,
-			dev->ifindex);
+	icmpv6_flow_init(ndisc_socket->sk, &fl, type,
+			 saddr, daddr, dev->ifindex);
 
 	dst = ndisc_dst_alloc(dev, neigh, daddr, ip6_output);
 	if (!dst)
@@ -1439,8 +1424,8 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
 		return;
 	}
 
-	ndisc_flow_init(&fl, NDISC_REDIRECT, &saddr_buf, &ipv6_hdr(skb)->saddr,
-			dev->ifindex);
+	icmpv6_flow_init(ndisc_socket->sk, &fl, NDISC_REDIRECT,
+			 &saddr_buf, &ipv6_hdr(skb)->saddr, dev->ifindex);
 
 	dst = ip6_route_output(NULL, &fl);
 	if (dst == NULL)
-- 
cgit v1.2.3


From 3b00944c5c73c49ef52bf17b66557c43c1d945fe Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 6 Dec 2007 16:11:48 -0800
Subject: [IPV6]: Make ndisc_dst_alloc() common for later use.

For later use, this patch is renaming ndisc_dst_alloc()
(and related function/structures) to icmp6_dst_alloc()
(and so on).  This patch also removing unused function-
pointer argument for it.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/ip6_route.h |  8 ++++----
 net/ipv6/ip6_fib.c      |  2 +-
 net/ipv6/ndisc.c        |  2 +-
 net/ipv6/route.c        | 27 +++++++++++++--------------
 4 files changed, 19 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index f99e4f0f568f..8d155a645aba 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -73,11 +73,11 @@ extern struct rt6_info		*rt6_lookup(struct in6_addr *daddr,
 					    struct in6_addr *saddr,
 					    int oif, int flags);
 
-extern struct dst_entry *ndisc_dst_alloc(struct net_device *dev,
+extern struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 					 struct neighbour *neigh,
-					 struct in6_addr *addr,
-					 int (*output)(struct sk_buff *));
-extern int ndisc_dst_gc(int *more);
+					 struct in6_addr *addr);
+extern int icmp6_dst_gc(int *more);
+
 extern void fib6_force_start_gc(void);
 
 extern struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index bab72b6f1444..c70fd38b54b7 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1464,7 +1464,7 @@ void fib6_run_gc(unsigned long dummy)
 	}
 	gc_args.more = 0;
 
-	ndisc_dst_gc(&gc_args.more);
+	icmp6_dst_gc(&gc_args.more);
 	fib6_clean_all(fib6_age, 0, NULL);
 
 	if (gc_args.more)
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 8db5f4a419aa..eb322959a3e1 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -462,7 +462,7 @@ static void __ndisc_send(struct net_device *dev,
 	icmpv6_flow_init(ndisc_socket->sk, &fl, type,
 			 saddr, daddr, dev->ifindex);
 
-	dst = ndisc_dst_alloc(dev, neigh, daddr, ip6_output);
+	dst = icmp6_dst_alloc(dev, neigh, daddr);
 	if (!dst)
 		return;
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 6abe7da45ef7..cd717450fb10 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -904,13 +904,12 @@ static inline unsigned int ipv6_advmss(unsigned int mtu)
 	return mtu;
 }
 
-static struct dst_entry *ndisc_dst_gc_list;
-static DEFINE_SPINLOCK(ndisc_lock);
+static struct dst_entry *icmp6_dst_gc_list;
+static DEFINE_SPINLOCK(icmp6_dst_lock);
 
-struct dst_entry *ndisc_dst_alloc(struct net_device *dev,
+struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 				  struct neighbour *neigh,
-				  struct in6_addr *addr,
-				  int (*output)(struct sk_buff *))
+				  struct in6_addr *addr)
 {
 	struct rt6_info *rt;
 	struct inet6_dev *idev = in6_dev_get(dev);
@@ -937,7 +936,7 @@ struct dst_entry *ndisc_dst_alloc(struct net_device *dev,
 	rt->u.dst.metrics[RTAX_HOPLIMIT-1] = 255;
 	rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev);
 	rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
-	rt->u.dst.output  = output;
+	rt->u.dst.output  = ip6_output;
 
 #if 0	/* there's no chance to use these for ndisc */
 	rt->u.dst.flags   = ipv6_addr_type(addr) & IPV6_ADDR_UNICAST
@@ -947,10 +946,10 @@ struct dst_entry *ndisc_dst_alloc(struct net_device *dev,
 	rt->rt6i_dst.plen = 128;
 #endif
 
-	spin_lock_bh(&ndisc_lock);
-	rt->u.dst.next = ndisc_dst_gc_list;
-	ndisc_dst_gc_list = &rt->u.dst;
-	spin_unlock_bh(&ndisc_lock);
+	spin_lock_bh(&icmp6_dst_lock);
+	rt->u.dst.next = icmp6_dst_gc_list;
+	icmp6_dst_gc_list = &rt->u.dst;
+	spin_unlock_bh(&icmp6_dst_lock);
 
 	fib6_force_start_gc();
 
@@ -958,7 +957,7 @@ out:
 	return &rt->u.dst;
 }
 
-int ndisc_dst_gc(int *more)
+int icmp6_dst_gc(int *more)
 {
 	struct dst_entry *dst, *next, **pprev;
 	int freed;
@@ -966,8 +965,8 @@ int ndisc_dst_gc(int *more)
 	next = NULL;
 	freed = 0;
 
-	spin_lock_bh(&ndisc_lock);
-	pprev = &ndisc_dst_gc_list;
+	spin_lock_bh(&icmp6_dst_lock);
+	pprev = &icmp6_dst_gc_list;
 
 	while ((dst = *pprev) != NULL) {
 		if (!atomic_read(&dst->__refcnt)) {
@@ -980,7 +979,7 @@ int ndisc_dst_gc(int *more)
 		}
 	}
 
-	spin_unlock_bh(&ndisc_lock);
+	spin_unlock_bh(&icmp6_dst_lock);
 
 	return freed;
 }
-- 
cgit v1.2.3


From 58f09b78b730cf0d936597272bf35b3d615e967c Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Mon, 3 Mar 2008 23:25:27 -0800
Subject: [NETNS][IPV6] ip6_fib - make it per network namespace

The fib table for ipv6 are moved to the network namespace structure.
All references to them are made relatively to the network namespace.

All external calls to the ip6_fib functions taking the network
namespace parameter are made using the init_net variable, so the
ip6_fib engine is ready for the namespaces but the callers not yet.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h    |   9 +--
 include/net/netns/ipv6.h |   5 ++
 net/ipv6/fib6_rules.c    |   8 +--
 net/ipv6/ip6_fib.c       | 161 +++++++++++++++++++++++++++--------------------
 net/ipv6/route.c         |  22 ++++---
 5 files changed, 119 insertions(+), 86 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 953d6040ff50..4d4c8aca8fb9 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -181,10 +181,11 @@ typedef struct rt6_info *(*pol_lookup_t)(struct fib6_table *,
  *	exported functions
  */
 
-extern struct fib6_table *	fib6_get_table(u32 id);
-extern struct fib6_table *	fib6_new_table(u32 id);
-extern struct dst_entry *	fib6_rule_lookup(struct flowi *fl, int flags,
-						 pol_lookup_t lookup);
+extern struct fib6_table        *fib6_get_table(struct net *net, u32 id);
+extern struct fib6_table        *fib6_new_table(struct net *net, u32 id);
+extern struct dst_entry         *fib6_rule_lookup(struct net *net,
+						  struct flowi *fl, int flags,
+						  pol_lookup_t lookup);
 
 extern struct fib6_node		*fib6_lookup(struct fib6_node *root,
 					     struct in6_addr *daddr,
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 82623d3a8e35..b0653261c5a3 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -35,6 +35,11 @@ struct netns_ipv6 {
 	struct xt_table		*ip6table_filter;
 	struct xt_table		*ip6table_mangle;
 	struct xt_table		*ip6table_raw;
+#endif
+	struct hlist_head       *fib_table_hash;
+	struct fib6_table       *fib6_main_tbl;
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	struct fib6_table       *fib6_local_tbl;
 #endif
 	struct sock		**icmp_sk;
 };
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 157db3a1ce00..03ad23a5fd3c 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -31,8 +31,8 @@ struct fib6_rule
 
 static struct fib_rules_ops fib6_rules_ops;
 
-struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
-				   pol_lookup_t lookup)
+struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
+				   int flags, pol_lookup_t lookup)
 {
 	struct fib_lookup_arg arg = {
 		.lookup_ptr = lookup,
@@ -71,7 +71,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 		goto discard_pkt;
 	}
 
-	table = fib6_get_table(rule->table);
+	table = fib6_get_table(&init_net, rule->table);
 	if (table)
 		rt = lookup(table, flp, flags);
 
@@ -151,7 +151,7 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 		if (rule->table == RT6_TABLE_UNSPEC)
 			goto errout;
 
-		if (fib6_new_table(rule->table) == NULL) {
+		if (fib6_new_table(&init_net, rule->table) == NULL) {
 			err = -ENOBUFS;
 			goto errout;
 		}
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 04d774963f3c..7b549f0bc428 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -166,16 +166,13 @@ static __inline__ void rt6_release(struct rt6_info *rt)
 		dst_free(&rt->u.dst);
 }
 
-static struct fib6_table *fib6_main_tbl;
-
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 #define FIB_TABLE_HASHSZ 256
 #else
 #define FIB_TABLE_HASHSZ 1
 #endif
-static struct hlist_head *fib_table_hash;
 
-static void fib6_link_table(struct fib6_table *tb)
+static void fib6_link_table(struct net *net, struct fib6_table *tb)
 {
 	unsigned int h;
 
@@ -191,13 +188,11 @@ static void fib6_link_table(struct fib6_table *tb)
 	 * No protection necessary, this is the only list mutatation
 	 * operation, tables never disappear once they exist.
 	 */
-	hlist_add_head_rcu(&tb->tb6_hlist, &fib_table_hash[h]);
+	hlist_add_head_rcu(&tb->tb6_hlist, &net->ipv6.fib_table_hash[h]);
 }
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 
-static struct fib6_table *fib6_local_tbl;
-
 static struct fib6_table *fib6_alloc_table(u32 id)
 {
 	struct fib6_table *table;
@@ -212,26 +207,27 @@ static struct fib6_table *fib6_alloc_table(u32 id)
 	return table;
 }
 
-struct fib6_table *fib6_new_table(u32 id)
+struct fib6_table *fib6_new_table(struct net *net, u32 id)
 {
 	struct fib6_table *tb;
 
 	if (id == 0)
 		id = RT6_TABLE_MAIN;
-	tb = fib6_get_table(id);
+	tb = fib6_get_table(net, id);
 	if (tb)
 		return tb;
 
 	tb = fib6_alloc_table(id);
 	if (tb != NULL)
-		fib6_link_table(tb);
+		fib6_link_table(net, tb);
 
 	return tb;
 }
 
-struct fib6_table *fib6_get_table(u32 id)
+struct fib6_table *fib6_get_table(struct net *net, u32 id)
 {
 	struct fib6_table *tb;
+	struct hlist_head *head;
 	struct hlist_node *node;
 	unsigned int h;
 
@@ -239,7 +235,8 @@ struct fib6_table *fib6_get_table(u32 id)
 		id = RT6_TABLE_MAIN;
 	h = id & (FIB_TABLE_HASHSZ - 1);
 	rcu_read_lock();
-	hlist_for_each_entry_rcu(tb, node, &fib_table_hash[h], tb6_hlist) {
+	head = &net->ipv6.fib_table_hash[h];
+	hlist_for_each_entry_rcu(tb, node, head, tb6_hlist) {
 		if (tb->tb6_id == id) {
 			rcu_read_unlock();
 			return tb;
@@ -250,33 +247,32 @@ struct fib6_table *fib6_get_table(u32 id)
 	return NULL;
 }
 
-static void __init fib6_tables_init(void)
+static void fib6_tables_init(struct net *net)
 {
-	fib6_link_table(fib6_main_tbl);
-	fib6_link_table(fib6_local_tbl);
+	fib6_link_table(net, net->ipv6.fib6_main_tbl);
+	fib6_link_table(net, net->ipv6.fib6_local_tbl);
 }
-
 #else
 
-struct fib6_table *fib6_new_table(u32 id)
+struct fib6_table *fib6_new_table(struct net *net, u32 id)
 {
-	return fib6_get_table(id);
+	return fib6_get_table(net, id);
 }
 
-struct fib6_table *fib6_get_table(u32 id)
+struct fib6_table *fib6_get_table(struct net *net, u32 id)
 {
-	return fib6_main_tbl;
+	  return net->ipv6.fib6_main_tbl;
 }
 
-struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
-				   pol_lookup_t lookup)
+struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
+				   int flags, pol_lookup_t lookup)
 {
-	return (struct dst_entry *) lookup(fib6_main_tbl, fl, flags);
+	return (struct dst_entry *) lookup(net->ipv6.fib6_main_tbl, fl, flags);
 }
 
-static void __init fib6_tables_init(void)
+static void fib6_tables_init(struct net *net)
 {
-	fib6_link_table(fib6_main_tbl);
+	fib6_link_table(net, net->ipv6.fib6_main_tbl);
 }
 
 #endif
@@ -357,11 +353,9 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 	struct fib6_walker_t *w;
 	struct fib6_table *tb;
 	struct hlist_node *node;
+	struct hlist_head *head;
 	int res = 0;
 
-	if (net != &init_net)
-		return 0;
-
 	s_h = cb->args[0];
 	s_e = cb->args[1];
 
@@ -390,7 +384,8 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 
 	for (h = s_h; h < FIB_TABLE_HASHSZ; h++, s_e = 0) {
 		e = 0;
-		hlist_for_each_entry(tb, node, &fib_table_hash[h], tb6_hlist) {
+		head = &net->ipv6.fib_table_hash[h];
+		hlist_for_each_entry(tb, node, head, tb6_hlist) {
 			if (e < s_e)
 				goto next;
 			res = fib6_dump_table(tb, skb, cb);
@@ -1360,12 +1355,13 @@ void fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
 {
 	struct fib6_table *table;
 	struct hlist_node *node;
+	struct hlist_head *head;
 	unsigned int h;
 
 	rcu_read_lock();
 	for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
-		hlist_for_each_entry_rcu(table, node, &fib_table_hash[h],
-					 tb6_hlist) {
+		head = &init_net.ipv6.fib_table_hash[h];
+		hlist_for_each_entry_rcu(table, node, head, tb6_hlist) {
 			write_lock_bh(&table->tb6_lock);
 			fib6_clean_tree(&table->tb6_root, func, prune, arg);
 			write_unlock_bh(&table->tb6_lock);
@@ -1466,55 +1462,88 @@ void fib6_run_gc(unsigned long dummy)
 	spin_unlock_bh(&fib6_gc_lock);
 }
 
-int __init fib6_init(void)
+static int fib6_net_init(struct net *net)
 {
-	int ret = -ENOMEM;
-	fib6_node_kmem = kmem_cache_create("fib6_nodes",
-					   sizeof(struct fib6_node),
-					   0, SLAB_HWCACHE_ALIGN,
-					   NULL);
-	if (!fib6_node_kmem)
-		goto out;
+	int ret;
 
-	fib_table_hash = kzalloc(sizeof(*fib_table_hash)*FIB_TABLE_HASHSZ,
-				 GFP_KERNEL);
-	if (!fib_table_hash)
-		goto out_kmem_cache_create;
+	ret = -ENOMEM;
+	net->ipv6.fib_table_hash =
+		kzalloc(sizeof(*net->ipv6.fib_table_hash)*FIB_TABLE_HASHSZ,
+			GFP_KERNEL);
+	if (!net->ipv6.fib_table_hash)
+		goto out;
 
-	fib6_main_tbl = kzalloc(sizeof(*fib6_main_tbl), GFP_KERNEL);
-	if (!fib6_main_tbl)
+	net->ipv6.fib6_main_tbl = kzalloc(sizeof(*net->ipv6.fib6_main_tbl),
+					  GFP_KERNEL);
+	if (!net->ipv6.fib6_main_tbl)
 		goto out_fib_table_hash;
 
-	fib6_main_tbl->tb6_id = RT6_TABLE_MAIN;
-	fib6_main_tbl->tb6_root.leaf = &ip6_null_entry;
-	fib6_main_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
+	net->ipv6.fib6_main_tbl->tb6_id = RT6_TABLE_MAIN;
+	net->ipv6.fib6_main_tbl->tb6_root.leaf = &ip6_null_entry;
+	net->ipv6.fib6_main_tbl->tb6_root.fn_flags =
+		RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	fib6_local_tbl = kzalloc(sizeof(*fib6_local_tbl), GFP_KERNEL);
-	if (!fib6_local_tbl)
+	net->ipv6.fib6_local_tbl = kzalloc(sizeof(*net->ipv6.fib6_local_tbl),
+					   GFP_KERNEL);
+	if (!net->ipv6.fib6_local_tbl)
 		goto out_fib6_main_tbl;
-
-	fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL;
-	fib6_local_tbl->tb6_root.leaf = &ip6_null_entry;
-	fib6_local_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
+	net->ipv6.fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL;
+	net->ipv6.fib6_local_tbl->tb6_root.leaf = &ip6_null_entry;
+	net->ipv6.fib6_local_tbl->tb6_root.fn_flags =
+		RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 #endif
+	fib6_tables_init(net);
 
-	fib6_tables_init();
-
-	ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
-	if (ret)
-		goto out_fib6_local_tbl;
+	ret = 0;
 out:
 	return ret;
 
-out_fib6_local_tbl:
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	kfree(fib6_local_tbl);
 out_fib6_main_tbl:
+	kfree(net->ipv6.fib6_main_tbl);
 #endif
-	kfree(fib6_main_tbl);
 out_fib_table_hash:
-	kfree(fib_table_hash);
+	kfree(net->ipv6.fib_table_hash);
+	goto out;
+ }
+
+static void fib6_net_exit(struct net *net)
+{
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	kfree(net->ipv6.fib6_local_tbl);
+#endif
+	kfree(net->ipv6.fib6_main_tbl);
+	kfree(net->ipv6.fib_table_hash);
+}
+
+static struct pernet_operations fib6_net_ops = {
+	.init = fib6_net_init,
+	.exit = fib6_net_exit,
+};
+
+int __init fib6_init(void)
+{
+	int ret = -ENOMEM;
+	fib6_node_kmem = kmem_cache_create("fib6_nodes",
+					   sizeof(struct fib6_node),
+					   0, SLAB_HWCACHE_ALIGN,
+					   NULL);
+	if (!fib6_node_kmem)
+		goto out;
+
+	ret = register_pernet_subsys(&fib6_net_ops);
+	if (ret)
+		goto out_kmem_cache_create;
+
+	ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
+	if (ret)
+		goto out_unregister_subsys;
+out:
+	return ret;
+
+out_unregister_subsys:
+	unregister_pernet_subsys(&fib6_net_ops);
 out_kmem_cache_create:
 	kmem_cache_destroy(fib6_node_kmem);
 	goto out;
@@ -1523,10 +1552,6 @@ out_kmem_cache_create:
 void fib6_gc_cleanup(void)
 {
 	del_timer(&ip6_fib_timer);
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	kfree(fib6_local_tbl);
-#endif
-	kfree(fib6_main_tbl);
-	kfree(fib_table_hash);
+	unregister_pernet_subsys(&fib6_net_ops);
 	kmem_cache_destroy(fib6_node_kmem);
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index cd717450fb10..09206f7ba525 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -571,7 +571,7 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
 		flags |= RT6_LOOKUP_F_HAS_SADDR;
 	}
 
-	dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_lookup);
+	dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_lookup);
 	if (dst->error == 0)
 		return (struct rt6_info *) dst;
 
@@ -758,7 +758,7 @@ void ip6_route_input(struct sk_buff *skb)
 	if (rt6_need_strict(&iph->daddr))
 		flags |= RT6_LOOKUP_F_IFACE;
 
-	skb->dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_input);
+	skb->dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_input);
 }
 
 static struct rt6_info *ip6_pol_route_output(struct fib6_table *table,
@@ -777,7 +777,7 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
 	if (!ipv6_addr_any(&fl->fl6_src))
 		flags |= RT6_LOOKUP_F_HAS_SADDR;
 
-	return fib6_rule_lookup(fl, flags, ip6_pol_route_output);
+	return fib6_rule_lookup(&init_net, fl, flags, ip6_pol_route_output);
 }
 
 EXPORT_SYMBOL(ip6_route_output);
@@ -1069,7 +1069,7 @@ int ip6_route_add(struct fib6_config *cfg)
 	if (cfg->fc_metric == 0)
 		cfg->fc_metric = IP6_RT_PRIO_USER;
 
-	table = fib6_new_table(cfg->fc_table);
+	table = fib6_new_table(&init_net, cfg->fc_table);
 	if (table == NULL) {
 		err = -ENOBUFS;
 		goto out;
@@ -1275,7 +1275,7 @@ static int ip6_route_del(struct fib6_config *cfg)
 	struct rt6_info *rt;
 	int err = -ESRCH;
 
-	table = fib6_get_table(cfg->fc_table);
+	table = fib6_get_table(&init_net, cfg->fc_table);
 	if (table == NULL)
 		return err;
 
@@ -1390,7 +1390,9 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest,
 	if (rt6_need_strict(dest))
 		flags |= RT6_LOOKUP_F_IFACE;
 
-	return (struct rt6_info *)fib6_rule_lookup((struct flowi *)&rdfl, flags, __ip6_route_redirect);
+	return (struct rt6_info *)fib6_rule_lookup(&init_net,
+						   (struct flowi *)&rdfl,
+						   flags, __ip6_route_redirect);
 }
 
 void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
@@ -1589,7 +1591,7 @@ static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixle
 	struct rt6_info *rt = NULL;
 	struct fib6_table *table;
 
-	table = fib6_get_table(RT6_TABLE_INFO);
+	table = fib6_get_table(&init_net, RT6_TABLE_INFO);
 	if (table == NULL)
 		return NULL;
 
@@ -1644,7 +1646,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
 	struct rt6_info *rt;
 	struct fib6_table *table;
 
-	table = fib6_get_table(RT6_TABLE_DFLT);
+	table = fib6_get_table(&init_net, RT6_TABLE_DFLT);
 	if (table == NULL)
 		return NULL;
 
@@ -1688,7 +1690,7 @@ void rt6_purge_dflt_routers(void)
 	struct fib6_table *table;
 
 	/* NOTE: Keep consistent with rt6_get_dflt_router */
-	table = fib6_get_table(RT6_TABLE_DFLT);
+	table = fib6_get_table(&init_net, RT6_TABLE_DFLT);
 	if (table == NULL)
 		return;
 
@@ -1851,7 +1853,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 
 	ipv6_addr_copy(&rt->rt6i_dst.addr, addr);
 	rt->rt6i_dst.plen = 128;
-	rt->rt6i_table = fib6_get_table(RT6_TABLE_LOCAL);
+	rt->rt6i_table = fib6_get_table(&init_net, RT6_TABLE_LOCAL);
 
 	atomic_set(&rt->u.dst.__refcnt, 1);
 
-- 
cgit v1.2.3


From f3db48517f59133610f558f29de8834d7b007691 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Mon, 3 Mar 2008 23:27:06 -0800
Subject: [NETNS][IPV6] ip6_fib - fib6_clean_all handle several network
 namespaces

The function fib6_clean_all takes the network namespace as
parameter. That allows to flush the routes related to a specific
network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h   |  3 ++-
 include/net/ip6_route.h |  2 +-
 net/ipv6/addrconf.c     |  3 ++-
 net/ipv6/ip6_fib.c      |  7 ++++---
 net/ipv6/route.c        | 28 ++++++++++++++++++++--------
 5 files changed, 29 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 4d4c8aca8fb9..e54075d902db 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -195,7 +195,8 @@ struct fib6_node		*fib6_locate(struct fib6_node *root,
 					     struct in6_addr *daddr, int dst_len,
 					     struct in6_addr *saddr, int src_len);
 
-extern void			fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
+extern void			fib6_clean_all(struct net *net,
+					       int (*func)(struct rt6_info *, void *arg),
 					       int prune, void *arg);
 
 extern int			fib6_add(struct fib6_node *root,
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 8d155a645aba..a158abe71933 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -121,7 +121,7 @@ struct rt6_rtnl_dump_arg
 };
 
 extern int rt6_dump_route(struct rt6_info *rt, void *p_arg);
-extern void rt6_ifdown(struct net_device *dev);
+extern void rt6_ifdown(struct net *net, struct net_device *dev);
 extern void rt6_mtu_change(struct net_device *dev, unsigned mtu);
 
 extern rwlock_t rt6_lock;
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 9b3a2d0e4269..a1d872dacad6 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -2431,6 +2431,7 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 {
 	struct inet6_dev *idev;
 	struct inet6_ifaddr *ifa, **bifa;
+	struct net *net = dev->nd_net;
 	int i;
 
 	ASSERT_RTNL();
@@ -2438,7 +2439,7 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 	if (dev == init_net.loopback_dev && how == 1)
 		how = 0;
 
-	rt6_ifdown(dev);
+	rt6_ifdown(net, dev);
 	neigh_ifdown(&nd_tbl, dev);
 
 	idev = __in6_dev_get(dev);
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 7b549f0bc428..0f9dc81f0e61 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1350,7 +1350,7 @@ static void fib6_clean_tree(struct fib6_node *root,
 	fib6_walk(&c.w);
 }
 
-void fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
+void fib6_clean_all(struct net *net, int (*func)(struct rt6_info *, void *arg),
 		    int prune, void *arg)
 {
 	struct fib6_table *table;
@@ -1360,7 +1360,7 @@ void fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
 
 	rcu_read_lock();
 	for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
-		head = &init_net.ipv6.fib_table_hash[h];
+		head = &net->ipv6.fib_table_hash[h];
 		hlist_for_each_entry_rcu(table, node, head, tb6_hlist) {
 			write_lock_bh(&table->tb6_lock);
 			fib6_clean_tree(&table->tb6_root, func, prune, arg);
@@ -1450,7 +1450,8 @@ void fib6_run_gc(unsigned long dummy)
 	gc_args.more = 0;
 
 	icmp6_dst_gc(&gc_args.more);
-	fib6_clean_all(fib6_age, 0, NULL);
+
+	fib6_clean_all(&init_net, fib6_age, 0, NULL);
 
 	if (gc_args.more)
 		mod_timer(&ip6_fib_timer, jiffies +
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 09206f7ba525..2e6da2afd948 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1870,9 +1870,9 @@ static int fib6_ifdown(struct rt6_info *rt, void *arg)
 	return 0;
 }
 
-void rt6_ifdown(struct net_device *dev)
+void rt6_ifdown(struct net *net, struct net_device *dev)
 {
-	fib6_clean_all(fib6_ifdown, 0, dev);
+	fib6_clean_all(net, fib6_ifdown, 0, dev);
 }
 
 struct rt6_mtu_change_arg
@@ -1928,7 +1928,7 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu)
 		.mtu = mtu,
 	};
 
-	fib6_clean_all(rt6_mtu_change_route, 0, &arg);
+	fib6_clean_all(dev->nd_net, rt6_mtu_change_route, 0, &arg);
 }
 
 static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = {
@@ -2318,13 +2318,25 @@ static int rt6_info_route(struct rt6_info *rt, void *p_arg)
 
 static int ipv6_route_show(struct seq_file *m, void *v)
 {
-	fib6_clean_all(rt6_info_route, 0, m);
+	struct net *net = (struct net *)m->private;
+	fib6_clean_all(net, rt6_info_route, 0, m);
 	return 0;
 }
 
 static int ipv6_route_open(struct inode *inode, struct file *file)
 {
-	return single_open(file, ipv6_route_show, NULL);
+	struct net *net = get_proc_net(inode);
+	if (!net)
+		return -ENXIO;
+	return single_open(file, ipv6_route_show, net);
+}
+
+static int ipv6_route_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq = file->private_data;
+	struct net *net = seq->private;
+	put_net(net);
+	return single_release(inode, file);
 }
 
 static const struct file_operations ipv6_route_proc_fops = {
@@ -2332,7 +2344,7 @@ static const struct file_operations ipv6_route_proc_fops = {
 	.open		= ipv6_route_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= single_release,
+	.release	= ipv6_route_release,
 };
 
 static int rt6_stats_seq_show(struct seq_file *seq, void *v)
@@ -2570,7 +2582,7 @@ xfrm6_init:
 out_proc_init:
 	ipv6_route_proc_fini(&init_net);
 out_fib6_init:
-	rt6_ifdown(NULL);
+	rt6_ifdown(&init_net, NULL);
 	fib6_gc_cleanup();
 out_kmem_cache:
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
@@ -2582,7 +2594,7 @@ void ip6_route_cleanup(void)
 	fib6_rules_cleanup();
 	ipv6_route_proc_fini(&init_net);
 	xfrm6_fini();
-	rt6_ifdown(NULL);
+	rt6_ifdown(&init_net, NULL);
 	fib6_gc_cleanup();
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
 }
-- 
cgit v1.2.3


From 5b7c931dff03621ae7ac524c4fa280d4e5f187a4 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Mon, 3 Mar 2008 23:28:58 -0800
Subject: [NETNS][IPV6] ip6_fib - add net to gc timer parameter

The fib tables are now relative to the network namespace. When the
garbage collector timer expires, we must have a network namespace
parameter in order to retrieve the tables. For now this is the
init_net, but we should be able to have a timer per namespace and use
the timer callback parameter to pass the network namespace from the
expired timer.

The timer callback, fib6_run_gc, is actually used to be called
synchronously by some functions and asynchronously when the timer
expires.

When the timer expires, the delay specified for fib6_run_gc parameter
is always zero. So, I changed fib6_run_gc to not be a timer callback
but a function called by the timer callback and I added a timer
callback where its work is just to retrieve from the data arg of the
timer the network namespace and call fib6_run_gc with zero expiring
time and the network namespace parameters. That makes the code cleaner
for the fib6_run_gc callers.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h |  3 ++-
 net/ipv6/ip6_fib.c    | 24 ++++++++++++++++--------
 net/ipv6/ndisc.c      |  5 +++--
 net/ipv6/route.c      |  8 +++++---
 4 files changed, 26 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index e54075d902db..fae267f65341 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -209,7 +209,8 @@ extern int			fib6_del(struct rt6_info *rt,
 extern void			inet6_rt_notify(int event, struct rt6_info *rt,
 						struct nl_info *info);
 
-extern void			fib6_run_gc(unsigned long dummy);
+extern void			fib6_run_gc(unsigned long expires,
+					    struct net *net);
 
 extern void			fib6_gc_cleanup(void);
 
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 0f9dc81f0e61..9b1c232a29ee 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -93,7 +93,10 @@ static int fib6_walk_continue(struct fib6_walker_t *w);
 
 static __u32 rt_sernum;
 
-static DEFINE_TIMER(ip6_fib_timer, fib6_run_gc, 0, 0);
+static void fib6_gc_timer_cb(unsigned long arg);
+
+static DEFINE_TIMER(ip6_fib_timer, fib6_gc_timer_cb, 0,
+		    (unsigned long)&init_net);
 
 static struct fib6_walker_t fib6_walker_list = {
 	.prev	= &fib6_walker_list,
@@ -1432,12 +1435,12 @@ static int fib6_age(struct rt6_info *rt, void *arg)
 
 static DEFINE_SPINLOCK(fib6_gc_lock);
 
-void fib6_run_gc(unsigned long dummy)
+void fib6_run_gc(unsigned long expires, struct net *net)
 {
-	if (dummy != ~0UL) {
+	if (expires != ~0UL) {
 		spin_lock_bh(&fib6_gc_lock);
-		gc_args.timeout = dummy ? (int)dummy :
-			init_net.ipv6.sysctl.ip6_rt_gc_interval;
+		gc_args.timeout = expires ? (int)expires :
+			net->ipv6.sysctl.ip6_rt_gc_interval;
 	} else {
 		local_bh_disable();
 		if (!spin_trylock(&fib6_gc_lock)) {
@@ -1445,17 +1448,17 @@ void fib6_run_gc(unsigned long dummy)
 			local_bh_enable();
 			return;
 		}
-		gc_args.timeout = init_net.ipv6.sysctl.ip6_rt_gc_interval;
+		gc_args.timeout = net->ipv6.sysctl.ip6_rt_gc_interval;
 	}
 	gc_args.more = 0;
 
 	icmp6_dst_gc(&gc_args.more);
 
-	fib6_clean_all(&init_net, fib6_age, 0, NULL);
+	fib6_clean_all(net, fib6_age, 0, NULL);
 
 	if (gc_args.more)
 		mod_timer(&ip6_fib_timer, jiffies +
-			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
+			  net->ipv6.sysctl.ip6_rt_gc_interval);
 	else {
 		del_timer(&ip6_fib_timer);
 		ip6_fib_timer.expires = 0;
@@ -1463,6 +1466,11 @@ void fib6_run_gc(unsigned long dummy)
 	spin_unlock_bh(&fib6_gc_lock);
 }
 
+static void fib6_gc_timer_cb(unsigned long arg)
+{
+	fib6_run_gc(0, (struct net *)arg);
+}
+
 static int fib6_net_init(struct net *net)
 {
 	int ret;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index eb322959a3e1..f1c95125100d 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1598,6 +1598,7 @@ int ndisc_rcv(struct sk_buff *skb)
 static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, void *ptr)
 {
 	struct net_device *dev = ptr;
+	struct net *net = dev->nd_net;
 
 	if (dev->nd_net != &init_net)
 		return NOTIFY_DONE;
@@ -1605,11 +1606,11 @@ static int ndisc_netdev_event(struct notifier_block *this, unsigned long event,
 	switch (event) {
 	case NETDEV_CHANGEADDR:
 		neigh_changeaddr(&nd_tbl, dev);
-		fib6_run_gc(~0UL);
+		fib6_run_gc(~0UL, net);
 		break;
 	case NETDEV_DOWN:
 		neigh_ifdown(&nd_tbl, dev);
-		fib6_run_gc(~0UL);
+		fib6_run_gc(~0UL, net);
 		break;
 	default:
 		break;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 2e6da2afd948..fd44721abebb 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -40,6 +40,7 @@
 #include <linux/if_arp.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
+#include <linux/nsproxy.h>
 #include <net/net_namespace.h>
 #include <net/snmp.h>
 #include <net/ipv6.h>
@@ -995,7 +996,7 @@ static int ip6_dst_gc(struct dst_ops *ops)
 		goto out;
 
 	expire++;
-	fib6_run_gc(expire);
+	fib6_run_gc(expire, &init_net);
 	last_gc = now;
 	if (atomic_read(&ip6_dst_ops.entries) < ip6_dst_ops.gc_thresh)
 		expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1;
@@ -2413,10 +2414,11 @@ static
 int ipv6_sysctl_rtcache_flush(ctl_table *ctl, int write, struct file * filp,
 			      void __user *buffer, size_t *lenp, loff_t *ppos)
 {
-	int delay = init_net.ipv6.sysctl.flush_delay;
+	struct net *net = current->nsproxy->net_ns;
+	int delay = net->ipv6.sysctl.flush_delay;
 	if (write) {
 		proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
-		fib6_run_gc(delay <= 0 ? ~0UL : (unsigned long)delay);
+		fib6_run_gc(delay <= 0 ? ~0UL : (unsigned long)delay, net);
 		return 0;
 	} else
 		return -EINVAL;
-- 
cgit v1.2.3


From 63152fc0de4dfe83da543bf133cef73d885a50fc Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Mon, 3 Mar 2008 23:31:11 -0800
Subject: [NETNS][IPV6] ip6_fib - gc timer per namespace

Move the timer initialization at the network namespace creation and
store the network namespace in the timer argument.

That enables multiple timers (one per network namespace) to do garbage
collecting.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h  |  2 +-
 include/net/netns/ipv6.h |  1 +
 net/ipv6/ip6_fib.c       | 56 ++++++++++++++++++++++++------------------------
 net/ipv6/route.c         |  5 ++++-
 4 files changed, 34 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index a158abe71933..79dce496f4d4 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -78,7 +78,7 @@ extern struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 					 struct in6_addr *addr);
 extern int icmp6_dst_gc(int *more);
 
-extern void fib6_force_start_gc(void);
+extern void fib6_force_start_gc(struct net *net);
 
 extern struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 					   const struct in6_addr *addr,
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index b0653261c5a3..5279cd6a00ba 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -36,6 +36,7 @@ struct netns_ipv6 {
 	struct xt_table		*ip6table_mangle;
 	struct xt_table		*ip6table_raw;
 #endif
+	struct timer_list       *ip6_fib_timer;
 	struct hlist_head       *fib_table_hash;
 	struct fib6_table       *fib6_main_tbl;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 77ad1002c904..4aa67988f518 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -95,8 +95,6 @@ static __u32 rt_sernum;
 
 static void fib6_gc_timer_cb(unsigned long arg);
 
-static struct timer_list *ip6_fib_timer;
-
 static struct fib6_walker_t fib6_walker_list = {
 	.prev	= &fib6_walker_list,
 	.next	= &fib6_walker_list,
@@ -663,19 +661,19 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
 	return 0;
 }
 
-static __inline__ void fib6_start_gc(struct rt6_info *rt)
+static __inline__ void fib6_start_gc(struct net *net, struct rt6_info *rt)
 {
-	if (ip6_fib_timer->expires == 0 &&
+	if (net->ipv6.ip6_fib_timer->expires == 0 &&
 	    (rt->rt6i_flags & (RTF_EXPIRES|RTF_CACHE)))
-		mod_timer(ip6_fib_timer, jiffies +
-			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
+		mod_timer(net->ipv6.ip6_fib_timer, jiffies +
+			  net->ipv6.sysctl.ip6_rt_gc_interval);
 }
 
-void fib6_force_start_gc(void)
+void fib6_force_start_gc(struct net *net)
 {
-	if (ip6_fib_timer->expires == 0)
-		mod_timer(ip6_fib_timer, jiffies +
-			  init_net.ipv6.sysctl.ip6_rt_gc_interval);
+	if (net->ipv6.ip6_fib_timer->expires == 0)
+		mod_timer(net->ipv6.ip6_fib_timer, jiffies +
+			  net->ipv6.sysctl.ip6_rt_gc_interval);
 }
 
 /*
@@ -762,7 +760,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info)
 	err = fib6_add_rt2node(fn, rt, info);
 
 	if (err == 0) {
-		fib6_start_gc(rt);
+		fib6_start_gc(info->nl_net, rt);
 		if (!(rt->rt6i_flags&RTF_CACHE))
 			fib6_prune_clones(pn, rt);
 	}
@@ -1443,7 +1441,7 @@ void fib6_run_gc(unsigned long expires, struct net *net)
 	} else {
 		local_bh_disable();
 		if (!spin_trylock(&fib6_gc_lock)) {
-			mod_timer(ip6_fib_timer, jiffies + HZ);
+			mod_timer(net->ipv6.ip6_fib_timer, jiffies + HZ);
 			local_bh_enable();
 			return;
 		}
@@ -1456,11 +1454,11 @@ void fib6_run_gc(unsigned long expires, struct net *net)
 	fib6_clean_all(net, fib6_age, 0, NULL);
 
 	if (gc_args.more)
-		mod_timer(ip6_fib_timer, jiffies +
+		mod_timer(net->ipv6.ip6_fib_timer, jiffies +
 			  net->ipv6.sysctl.ip6_rt_gc_interval);
 	else {
-		del_timer(ip6_fib_timer);
-		ip6_fib_timer->expires = 0;
+		del_timer(net->ipv6.ip6_fib_timer);
+		net->ipv6.ip6_fib_timer->expires = 0;
 	}
 	spin_unlock_bh(&fib6_gc_lock);
 }
@@ -1473,13 +1471,21 @@ static void fib6_gc_timer_cb(unsigned long arg)
 static int fib6_net_init(struct net *net)
 {
 	int ret;
+	struct timer_list *timer;
 
 	ret = -ENOMEM;
+	timer = kzalloc(sizeof(*timer), GFP_KERNEL);
+	if (!timer)
+		goto out;
+
+	setup_timer(timer, fib6_gc_timer_cb, (unsigned long)net);
+	net->ipv6.ip6_fib_timer = timer;
+
 	net->ipv6.fib_table_hash =
 		kzalloc(sizeof(*net->ipv6.fib_table_hash)*FIB_TABLE_HASHSZ,
 			GFP_KERNEL);
 	if (!net->ipv6.fib_table_hash)
-		goto out;
+		goto out_timer;
 
 	net->ipv6.fib6_main_tbl = kzalloc(sizeof(*net->ipv6.fib6_main_tbl),
 					  GFP_KERNEL);
@@ -1513,11 +1519,15 @@ out_fib6_main_tbl:
 #endif
 out_fib_table_hash:
 	kfree(net->ipv6.fib_table_hash);
+out_timer:
+	kfree(timer);
 	goto out;
  }
 
 static void fib6_net_exit(struct net *net)
 {
+	del_timer(net->ipv6.ip6_fib_timer);
+	kfree(net->ipv6.ip6_fib_timer);
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	kfree(net->ipv6.fib6_local_tbl);
 #endif
@@ -1533,6 +1543,7 @@ static struct pernet_operations fib6_net_ops = {
 int __init fib6_init(void)
 {
 	int ret = -ENOMEM;
+
 	fib6_node_kmem = kmem_cache_create("fib6_nodes",
 					   sizeof(struct fib6_node),
 					   0, SLAB_HWCACHE_ALIGN,
@@ -1540,16 +1551,9 @@ int __init fib6_init(void)
 	if (!fib6_node_kmem)
 		goto out;
 
-	ret = -ENOMEM;
-	ip6_fib_timer = kzalloc(sizeof(*ip6_fib_timer), GFP_KERNEL);
-	if (!ip6_fib_timer)
-		goto out_kmem_cache_create;
-
-	setup_timer(ip6_fib_timer, fib6_gc_timer_cb, (unsigned long)&init_net);
-
 	ret = register_pernet_subsys(&fib6_net_ops);
 	if (ret)
-		goto out_timer;
+		goto out_kmem_cache_create;
 
 	ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
 	if (ret)
@@ -1559,8 +1563,6 @@ out:
 
 out_unregister_subsys:
 	unregister_pernet_subsys(&fib6_net_ops);
-out_timer:
-	kfree(ip6_fib_timer);
 out_kmem_cache_create:
 	kmem_cache_destroy(fib6_node_kmem);
 	goto out;
@@ -1568,8 +1570,6 @@ out_kmem_cache_create:
 
 void fib6_gc_cleanup(void)
 {
-	del_timer(ip6_fib_timer);
-	kfree(ip6_fib_timer);
 	unregister_pernet_subsys(&fib6_net_ops);
 	kmem_cache_destroy(fib6_node_kmem);
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index fd44721abebb..b13eb0111270 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -952,7 +952,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 	icmp6_dst_gc_list = &rt->u.dst;
 	spin_unlock_bh(&icmp6_dst_lock);
 
-	fib6_force_start_gc();
+	fib6_force_start_gc(dev->nd_net);
 
 out:
 	return &rt->u.dst;
@@ -1230,6 +1230,9 @@ install_route:
 	rt->u.dst.dev = dev;
 	rt->rt6i_idev = idev;
 	rt->rt6i_table = table;
+
+	cfg->fc_nlinfo.nl_net = dev->nd_net;
+
 	return __ip6_ins_rt(rt, &cfg->fc_nlinfo);
 
 out:
-- 
cgit v1.2.3


From dcabb819a6eced95ef531b001e663d0d592c8d9f Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Mon, 3 Mar 2008 23:33:08 -0800
Subject: [NETNS][IPV6] fib6_rules - handle several network namespaces

The fib6_rules_ops is moved to the network namespace structure.  All
references are changed to have it relatively to it.

Each time a network namespace is created a new fib6_rules_ops is
allocated, initialized and stored into the network namespace
structure.

The common part of the fib rules is namespace aware, so it is quite
easy to retrieve the network namespace from the rules and use it in
the different callbacks.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |  1 +
 net/ipv6/fib6_rules.c    | 82 +++++++++++++++++++++++++++---------------------
 2 files changed, 47 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 5279cd6a00ba..66bf9c0f745b 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -41,6 +41,7 @@ struct netns_ipv6 {
 	struct fib6_table       *fib6_main_tbl;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	struct fib6_table       *fib6_local_tbl;
+	struct fib_rules_ops    *fib6_rules_ops;
 #endif
 	struct sock		**icmp_sk;
 };
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 60af08f12547..89cb092c9732 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -29,8 +29,6 @@ struct fib6_rule
 	u8			tclass;
 };
 
-static struct fib_rules_ops *fib6_rules_ops;
-
 struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
 				   int flags, pol_lookup_t lookup)
 {
@@ -38,7 +36,7 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
 		.lookup_ptr = lookup,
 	};
 
-	fib_rules_lookup(fib6_rules_ops, fl, flags, &arg);
+	fib_rules_lookup(net->ipv6.fib6_rules_ops, fl, flags, &arg);
 	if (arg.rule)
 		fib_rule_put(arg.rule);
 
@@ -71,7 +69,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 		goto discard_pkt;
 	}
 
-	table = fib6_get_table(&init_net, rule->table);
+	table = fib6_get_table(rule->fr_net, rule->table);
 	if (table)
 		rt = lookup(table, flp, flags);
 
@@ -145,13 +143,14 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 			       struct nlattr **tb)
 {
 	int err = -EINVAL;
+	struct net *net = skb->sk->sk_net;
 	struct fib6_rule *rule6 = (struct fib6_rule *) rule;
 
 	if (rule->action == FR_ACT_TO_TBL) {
 		if (rule->table == RT6_TABLE_UNSPEC)
 			goto errout;
 
-		if (fib6_new_table(&init_net, rule->table) == NULL) {
+		if (fib6_new_table(net, rule->table) == NULL) {
 			err = -ENOBUFS;
 			goto errout;
 		}
@@ -251,49 +250,60 @@ static struct fib_rules_ops fib6_rules_ops_template = {
 	.fro_net		= &init_net,
 };
 
-static int __init fib6_default_rules_init(void)
+static int fib6_rules_net_init(struct net *net)
 {
-	int err;
+	int err = -ENOMEM;
 
-	fib6_rules_ops = kmemdup(&fib6_rules_ops_template,
-				 sizeof(*fib6_rules_ops), GFP_KERNEL);
-	if (!fib6_rules_ops)
-		return -ENOMEM;
+	net->ipv6.fib6_rules_ops = kmemdup(&fib6_rules_ops_template,
+					   sizeof(*net->ipv6.fib6_rules_ops),
+					   GFP_KERNEL);
+	if (!net->ipv6.fib6_rules_ops)
+		goto out;
 
-	INIT_LIST_HEAD(&fib6_rules_ops->rules_list);
+	net->ipv6.fib6_rules_ops->fro_net = net;
+	INIT_LIST_HEAD(&net->ipv6.fib6_rules_ops->rules_list);
 
-	err = fib_default_rule_add(fib6_rules_ops, 0,
+	err = fib_default_rule_add(net->ipv6.fib6_rules_ops, 0,
 				   RT6_TABLE_LOCAL, FIB_RULE_PERMANENT);
-	if (err < 0)
-		return err;
-	err = fib_default_rule_add(fib6_rules_ops, 0x7FFE, RT6_TABLE_MAIN, 0);
-	if (err < 0)
-		return err;
-	return 0;
-}
-
-int __init fib6_rules_init(void)
-{
-	int ret;
+	if (err)
+		goto out_fib6_rules_ops;
 
-	ret = fib6_default_rules_init();
-	if (ret)
-		goto out;
+	err = fib_default_rule_add(net->ipv6.fib6_rules_ops,
+				   0x7FFE, RT6_TABLE_MAIN, 0);
+	if (err)
+		goto out_fib6_default_rule_add;
 
-	ret = fib_rules_register(fib6_rules_ops);
-	if (ret)
-		goto out_default_rules_init;
+	err = fib_rules_register(net->ipv6.fib6_rules_ops);
+	if (err)
+		goto out_fib6_default_rule_add;
 out:
-	return ret;
+	return err;
 
-out_default_rules_init:
-	fib_rules_cleanup_ops(fib6_rules_ops);
-	kfree(fib6_rules_ops);
+out_fib6_default_rule_add:
+	fib_rules_cleanup_ops(net->ipv6.fib6_rules_ops);
+out_fib6_rules_ops:
+	kfree(net->ipv6.fib6_rules_ops);
 	goto out;
 }
 
+static void fib6_rules_net_exit(struct net *net)
+{
+	fib_rules_unregister(net->ipv6.fib6_rules_ops);
+	kfree(net->ipv6.fib6_rules_ops);
+}
+
+static struct pernet_operations fib6_rules_net_ops = {
+	.init = fib6_rules_net_init,
+	.exit = fib6_rules_net_exit,
+};
+
+int __init fib6_rules_init(void)
+{
+	return register_pernet_subsys(&fib6_rules_net_ops);
+}
+
+
 void fib6_rules_cleanup(void)
 {
-	fib_rules_unregister(fib6_rules_ops);
-	kfree(fib6_rules_ops);
+	return unregister_pernet_subsys(&fib6_rules_net_ops);
 }
-- 
cgit v1.2.3


From 6cc118bd506ae8c6436f507b838a0e1f6185fec2 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Mon, 3 Mar 2008 23:33:43 -0800
Subject: [NETNS][IPV6] rt6_stats - dynamically allocate the routes statistics

This patch allocates the rt6_stats struct dynamically when the fib6 is
initialized. That provides the ability to create several instances of
this structure for the network namespaces.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h |  2 +-
 net/ipv6/ip6_fib.c | 22 +++++++++++++++-------
 net/ipv6/route.c   |  8 ++++----
 3 files changed, 20 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 5dc8164e5d3b..9a00f35ae773 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -611,7 +611,7 @@ extern void ipv6_misc_proc_exit(void);
 extern int snmp6_register_dev(struct inet6_dev *idev);
 extern int snmp6_unregister_dev(struct inet6_dev *idev);
 
-extern struct rt6_statistics rt6_stats;
+extern struct rt6_statistics *rt6_stats;
 #else
 static inline int snmp6_register_dev(struct inet6_dev *idev)
 {
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 350a970ca1e3..7d82ad540140 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -48,7 +48,7 @@
 #define RT6_TRACE(x...) do { ; } while (0)
 #endif
 
-struct rt6_statistics	rt6_stats;
+struct rt6_statistics *rt6_stats;
 
 static struct kmem_cache * fib6_node_kmem __read_mostly;
 
@@ -653,10 +653,10 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
 	rt->rt6i_node = fn;
 	atomic_inc(&rt->rt6i_ref);
 	inet6_rt_notify(RTM_NEWROUTE, rt, info);
-	rt6_stats.fib_rt_entries++;
+	rt6_stats->fib_rt_entries++;
 
 	if ((fn->fn_flags & RTN_RTINFO) == 0) {
-		rt6_stats.fib_route_nodes++;
+		rt6_stats->fib_route_nodes++;
 		fn->fn_flags |= RTN_RTINFO;
 	}
 
@@ -1093,8 +1093,8 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 	/* Unlink it */
 	*rtp = rt->u.dst.rt6_next;
 	rt->rt6i_node = NULL;
-	rt6_stats.fib_rt_entries--;
-	rt6_stats.fib_discarded_routes++;
+	rt6_stats->fib_rt_entries--;
+	rt6_stats->fib_discarded_routes++;
 
 	/* Reset round-robin state, if necessary */
 	if (fn->rr_ptr == rt)
@@ -1117,7 +1117,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 	/* If it was last route, expunge its radix tree node */
 	if (fn->leaf == NULL) {
 		fn->fn_flags &= ~RTN_RTINFO;
-		rt6_stats.fib_route_nodes--;
+		rt6_stats->fib_route_nodes--;
 		fn = fib6_repair_tree(fn);
 	}
 
@@ -1556,9 +1556,14 @@ int __init fib6_init(void)
 	if (!fib6_node_kmem)
 		goto out;
 
+	ret = -ENOMEM;
+	rt6_stats = kzalloc(sizeof(*rt6_stats), GFP_KERNEL);
+	if (!rt6_stats)
+		goto out_kmem_cache_create;
+
 	ret = register_pernet_subsys(&fib6_net_ops);
 	if (ret)
-		goto out_kmem_cache_create;
+		goto out_rt6_stats;
 
 	ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
 	if (ret)
@@ -1568,6 +1573,8 @@ out:
 
 out_unregister_subsys:
 	unregister_pernet_subsys(&fib6_net_ops);
+out_rt6_stats:
+	kfree(rt6_stats);
 out_kmem_cache_create:
 	kmem_cache_destroy(fib6_node_kmem);
 	goto out;
@@ -1576,5 +1583,6 @@ out_kmem_cache_create:
 void fib6_gc_cleanup(void)
 {
 	unregister_pernet_subsys(&fib6_net_ops);
+	kfree(rt6_stats);
 	kmem_cache_destroy(fib6_node_kmem);
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index b13eb0111270..370f2aef6909 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2354,11 +2354,11 @@ static const struct file_operations ipv6_route_proc_fops = {
 static int rt6_stats_seq_show(struct seq_file *seq, void *v)
 {
 	seq_printf(seq, "%04x %04x %04x %04x %04x %04x %04x\n",
-		      rt6_stats.fib_nodes, rt6_stats.fib_route_nodes,
-		      rt6_stats.fib_rt_alloc, rt6_stats.fib_rt_entries,
-		      rt6_stats.fib_rt_cache,
+		      rt6_stats->fib_nodes, rt6_stats->fib_route_nodes,
+		      rt6_stats->fib_rt_alloc, rt6_stats->fib_rt_entries,
+		      rt6_stats->fib_rt_cache,
 		      atomic_read(&ip6_dst_ops.entries),
-		      rt6_stats.fib_discarded_routes);
+		      rt6_stats->fib_discarded_routes);
 
 	return 0;
 }
-- 
cgit v1.2.3


From c572872f89e46e38cdb35a43b81122bfb7ff43fc Mon Sep 17 00:00:00 2001
From: Benjamin Thery <benjamin.thery@bull.net>
Date: Mon, 3 Mar 2008 23:34:17 -0800
Subject: [NETNS][IPV6] rt6_stats - make the stats per network namespace

The rt6_stats is now per namespace with this patch. It is allocated
when a network namespace is created and freed when the network
namespace exits and references are relative to the network namespace.

Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h       |  1 -
 include/net/netns/ipv6.h |  1 +
 net/ipv6/ip6_fib.c       | 32 +++++++++++++++-----------------
 net/ipv6/route.c         | 12 +++++++-----
 4 files changed, 23 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 9a00f35ae773..5f6df50a33a9 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -611,7 +611,6 @@ extern void ipv6_misc_proc_exit(void);
 extern int snmp6_register_dev(struct inet6_dev *idev);
 extern int snmp6_unregister_dev(struct inet6_dev *idev);
 
-extern struct rt6_statistics *rt6_stats;
 #else
 static inline int snmp6_register_dev(struct inet6_dev *idev)
 {
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 66bf9c0f745b..c6c9afff13ef 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -36,6 +36,7 @@ struct netns_ipv6 {
 	struct xt_table		*ip6table_mangle;
 	struct xt_table		*ip6table_raw;
 #endif
+	struct rt6_statistics   *rt6_stats;
 	struct timer_list       *ip6_fib_timer;
 	struct hlist_head       *fib_table_hash;
 	struct fib6_table       *fib6_main_tbl;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 7d82ad540140..1c2566e3d392 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -48,8 +48,6 @@
 #define RT6_TRACE(x...) do { ; } while (0)
 #endif
 
-struct rt6_statistics *rt6_stats;
-
 static struct kmem_cache * fib6_node_kmem __read_mostly;
 
 enum fib_walk_state_t
@@ -653,10 +651,10 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
 	rt->rt6i_node = fn;
 	atomic_inc(&rt->rt6i_ref);
 	inet6_rt_notify(RTM_NEWROUTE, rt, info);
-	rt6_stats->fib_rt_entries++;
+	info->nl_net->ipv6.rt6_stats->fib_rt_entries++;
 
 	if ((fn->fn_flags & RTN_RTINFO) == 0) {
-		rt6_stats->fib_route_nodes++;
+		info->nl_net->ipv6.rt6_stats->fib_route_nodes++;
 		fn->fn_flags |= RTN_RTINFO;
 	}
 
@@ -1087,14 +1085,15 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 {
 	struct fib6_walker_t *w;
 	struct rt6_info *rt = *rtp;
+	struct net *net = info->nl_net;
 
 	RT6_TRACE("fib6_del_route\n");
 
 	/* Unlink it */
 	*rtp = rt->u.dst.rt6_next;
 	rt->rt6i_node = NULL;
-	rt6_stats->fib_rt_entries--;
-	rt6_stats->fib_discarded_routes++;
+	net->ipv6.rt6_stats->fib_rt_entries--;
+	net->ipv6.rt6_stats->fib_discarded_routes++;
 
 	/* Reset round-robin state, if necessary */
 	if (fn->rr_ptr == rt)
@@ -1117,7 +1116,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 	/* If it was last route, expunge its radix tree node */
 	if (fn->leaf == NULL) {
 		fn->fn_flags &= ~RTN_RTINFO;
-		rt6_stats->fib_route_nodes--;
+		net->ipv6.rt6_stats->fib_route_nodes--;
 		fn = fib6_repair_tree(fn);
 	}
 
@@ -1486,11 +1485,15 @@ static int fib6_net_init(struct net *net)
 	setup_timer(timer, fib6_gc_timer_cb, (unsigned long)net);
 	net->ipv6.ip6_fib_timer = timer;
 
+	net->ipv6.rt6_stats = kzalloc(sizeof(*net->ipv6.rt6_stats), GFP_KERNEL);
+	if (!net->ipv6.rt6_stats)
+		goto out_timer;
+
 	net->ipv6.fib_table_hash =
 		kzalloc(sizeof(*net->ipv6.fib_table_hash)*FIB_TABLE_HASHSZ,
 			GFP_KERNEL);
 	if (!net->ipv6.fib_table_hash)
-		goto out_timer;
+		goto out_rt6_stats;
 
 	net->ipv6.fib6_main_tbl = kzalloc(sizeof(*net->ipv6.fib6_main_tbl),
 					  GFP_KERNEL);
@@ -1524,6 +1527,8 @@ out_fib6_main_tbl:
 #endif
 out_fib_table_hash:
 	kfree(net->ipv6.fib_table_hash);
+out_rt6_stats:
+	kfree(net->ipv6.rt6_stats);
 out_timer:
 	kfree(timer);
 	goto out;
@@ -1538,6 +1543,7 @@ static void fib6_net_exit(struct net *net)
 #endif
 	kfree(net->ipv6.fib6_main_tbl);
 	kfree(net->ipv6.fib_table_hash);
+	kfree(net->ipv6.rt6_stats);
 }
 
 static struct pernet_operations fib6_net_ops = {
@@ -1556,14 +1562,9 @@ int __init fib6_init(void)
 	if (!fib6_node_kmem)
 		goto out;
 
-	ret = -ENOMEM;
-	rt6_stats = kzalloc(sizeof(*rt6_stats), GFP_KERNEL);
-	if (!rt6_stats)
-		goto out_kmem_cache_create;
-
 	ret = register_pernet_subsys(&fib6_net_ops);
 	if (ret)
-		goto out_rt6_stats;
+		goto out_kmem_cache_create;
 
 	ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib);
 	if (ret)
@@ -1573,8 +1574,6 @@ out:
 
 out_unregister_subsys:
 	unregister_pernet_subsys(&fib6_net_ops);
-out_rt6_stats:
-	kfree(rt6_stats);
 out_kmem_cache_create:
 	kmem_cache_destroy(fib6_node_kmem);
 	goto out;
@@ -1583,6 +1582,5 @@ out_kmem_cache_create:
 void fib6_gc_cleanup(void)
 {
 	unregister_pernet_subsys(&fib6_net_ops);
-	kfree(rt6_stats);
 	kmem_cache_destroy(fib6_node_kmem);
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 370f2aef6909..3afc3f41f2d5 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2354,11 +2354,13 @@ static const struct file_operations ipv6_route_proc_fops = {
 static int rt6_stats_seq_show(struct seq_file *seq, void *v)
 {
 	seq_printf(seq, "%04x %04x %04x %04x %04x %04x %04x\n",
-		      rt6_stats->fib_nodes, rt6_stats->fib_route_nodes,
-		      rt6_stats->fib_rt_alloc, rt6_stats->fib_rt_entries,
-		      rt6_stats->fib_rt_cache,
-		      atomic_read(&ip6_dst_ops.entries),
-		      rt6_stats->fib_discarded_routes);
+		   init_net.ipv6.rt6_stats->fib_nodes,
+		   init_net.ipv6.rt6_stats->fib_route_nodes,
+		   init_net.ipv6.rt6_stats->fib_rt_alloc,
+		   init_net.ipv6.rt6_stats->fib_rt_entries,
+		   init_net.ipv6.rt6_stats->fib_rt_cache,
+		   atomic_read(&ip6_dst_ops.entries),
+		   init_net.ipv6.rt6_stats->fib_discarded_routes);
 
 	return 0;
 }
-- 
cgit v1.2.3


From 606a2b4862d4be31fa55cad89871fe52a422d511 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 4 Mar 2008 13:45:59 -0800
Subject: [NETNS][IPV6] route6 - Pass the network namespace parameter to
 rt6_lookup

Add a network namespace parameter to rt6_lookup().

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h |  3 ++-
 net/ipv6/addrconf.c     |  4 ++--
 net/ipv6/anycast.c      |  2 +-
 net/ipv6/ip6_tunnel.c   |  4 ++--
 net/ipv6/mcast.c        |  4 ++--
 net/ipv6/route.c        | 10 +++++-----
 net/ipv6/sit.c          |  4 ++--
 7 files changed, 16 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 79dce496f4d4..92004c5160c7 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -69,7 +69,8 @@ extern void			rt6_sndmsg(int type, struct in6_addr *dst,
 					   int dstlen, int srclen,
 					   int metric, __u32 flags);
 
-extern struct rt6_info		*rt6_lookup(struct in6_addr *daddr,
+extern struct rt6_info		*rt6_lookup(struct net *net,
+					    struct in6_addr *daddr,
 					    struct in6_addr *saddr,
 					    int oif, int flags);
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index a1d872dacad6..9d894e8c7b72 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -753,7 +753,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
 		struct rt6_info *rt;
 
 		ipv6_addr_prefix(&prefix, &ifp->addr, ifp->prefix_len);
-		rt = rt6_lookup(&prefix, NULL, ifp->idev->dev->ifindex, 1);
+		rt = rt6_lookup(&init_net, &prefix, NULL, ifp->idev->dev->ifindex, 1);
 
 		if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
 			if (onlink == 0) {
@@ -1700,7 +1700,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
 
 	if (pinfo->onlink) {
 		struct rt6_info *rt;
-		rt = rt6_lookup(&pinfo->prefix, NULL, dev->ifindex, 1);
+		rt = rt6_lookup(&init_net, &pinfo->prefix, NULL, dev->ifindex, 1);
 
 		if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
 			if (rt->rt6i_flags&RTF_EXPIRES) {
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index 9c7f83fbc3a1..96868b994b37 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -101,7 +101,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 	if (ifindex == 0) {
 		struct rt6_info *rt;
 
-		rt = rt6_lookup(addr, NULL, 0, 0);
+		rt = rt6_lookup(&init_net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 78f438880923..4e1981660b3c 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -602,7 +602,7 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		skb_reset_network_header(skb2);
 
 		/* Try to guess incoming interface */
-		rt = rt6_lookup(&ipv6_hdr(skb2)->saddr, NULL, 0, 0);
+		rt = rt6_lookup(&init_net, &ipv6_hdr(skb2)->saddr, NULL, 0, 0);
 
 		if (rt && rt->rt6i_dev)
 			skb2->dev = rt->rt6i_dev;
@@ -1112,7 +1112,7 @@ static void ip6_tnl_link_config(struct ip6_tnl *t)
 		int strict = (ipv6_addr_type(&p->raddr) &
 			      (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL));
 
-		struct rt6_info *rt = rt6_lookup(&p->raddr, &p->laddr,
+		struct rt6_info *rt = rt6_lookup(&init_net, &p->raddr, &p->laddr,
 						 p->link, strict);
 
 		if (rt == NULL)
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index a373b8e7f241..197ca390a15d 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -208,7 +208,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 
 	if (ifindex == 0) {
 		struct rt6_info *rt;
-		rt = rt6_lookup(addr, NULL, 0, 0);
+		rt = rt6_lookup(&init_net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
@@ -294,7 +294,7 @@ static struct inet6_dev *ip6_mc_find_dev(struct in6_addr *group, int ifindex)
 	if (ifindex == 0) {
 		struct rt6_info *rt;
 
-		rt = rt6_lookup(group, NULL, 0, 0);
+		rt = rt6_lookup(&init_net, group, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 5d9d293156cd..d9d840ced1c6 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -553,8 +553,8 @@ out:
 
 }
 
-struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
-			    int oif, int strict)
+struct rt6_info *rt6_lookup(struct net *net, struct in6_addr *daddr,
+			    struct in6_addr *saddr, int oif, int strict)
 {
 	struct flowi fl = {
 		.oif = oif,
@@ -572,7 +572,7 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
 		flags |= RT6_LOOKUP_F_HAS_SADDR;
 	}
 
-	dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_lookup);
+	dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_lookup);
 	if (dst->error == 0)
 		return (struct rt6_info *) dst;
 
@@ -1159,7 +1159,7 @@ int ip6_route_add(struct fib6_config *cfg)
 			if (!(gwa_type&IPV6_ADDR_UNICAST))
 				goto out;
 
-			grt = rt6_lookup(gw_addr, NULL, cfg->fc_ifindex, 1);
+			grt = rt6_lookup(&init_net, gw_addr, NULL, cfg->fc_ifindex, 1);
 
 			err = -EHOSTUNREACH;
 			if (grt == NULL)
@@ -1483,7 +1483,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 	struct rt6_info *rt, *nrt;
 	int allfrag = 0;
 
-	rt = rt6_lookup(daddr, saddr, dev->ifindex, 0);
+	rt = rt6_lookup(dev->nd_net, daddr, saddr, dev->ifindex, 0);
 	if (rt == NULL)
 		return;
 
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 1656c003b989..68720aa63f96 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -339,11 +339,11 @@ out:
 	skb_reset_network_header(skb2);
 
 	/* Try to guess incoming interface */
-	rt6i = rt6_lookup(&iph6->saddr, NULL, NULL, 0);
+	rt6i = rt6_lookup(&init_net, &iph6->saddr, NULL, NULL, 0);
 	if (rt6i && rt6i->rt6i_dev) {
 		skb2->dev = rt6i->rt6i_dev;
 
-		rt6i = rt6_lookup(&iph6->daddr, &iph6->saddr, NULL, 0);
+		rt6i = rt6_lookup(&init_net, &iph6->daddr, &iph6->saddr, NULL, 0);
 
 		if (rt6i && rt6i->rt6i_dev && rt6i->rt6i_dev->type == ARPHRD_SIT) {
 			struct ip_tunnel *t = netdev_priv(rt6i->rt6i_dev);
-- 
cgit v1.2.3


From 7b4da53229bb61469bdab321384b9a13406e3485 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 4 Mar 2008 13:47:14 -0800
Subject: [NETNS][IPV6] route6 - Pass the network namespace parameter to
 rt6_purge_dflt_routers

Add a network namespace parameter to rt6_purge_dflt_routers.  This is
needed to call fib6_get_table with the appropriate network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h | 2 +-
 net/ipv6/addrconf.c     | 2 +-
 net/ipv6/route.c        | 4 ++--
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 92004c5160c7..1444d358836d 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -95,7 +95,7 @@ extern struct rt6_info *	rt6_add_dflt_router(struct in6_addr *gwaddr,
 						    struct net_device *dev,
 						    unsigned int pref);
 
-extern void			rt6_purge_dflt_routers(void);
+extern void			rt6_purge_dflt_routers(struct net *net);
 
 extern int			rt6_route_rcv(struct net_device *dev,
 					      u8 *opt, int len,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 9d894e8c7b72..b37ae421b61b 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -493,7 +493,7 @@ static void addrconf_fixup_forwarding(struct ctl_table *table, int *p, int old)
 		dev_forward_change((struct inet6_dev *)table->extra1);
 
 	if (*p)
-		rt6_purge_dflt_routers();
+		rt6_purge_dflt_routers(net);
 }
 #endif
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 4278cec522c5..ad3d684e544a 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1697,13 +1697,13 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 	return rt6_get_dflt_router(gwaddr, dev);
 }
 
-void rt6_purge_dflt_routers(void)
+void rt6_purge_dflt_routers(struct net *net)
 {
 	struct rt6_info *rt;
 	struct fib6_table *table;
 
 	/* NOTE: Keep consistent with rt6_get_dflt_router */
-	table = fib6_get_table(&init_net, RT6_TABLE_DFLT);
+	table = fib6_get_table(net, RT6_TABLE_DFLT);
 	if (table == NULL)
 		return;
 
-- 
cgit v1.2.3


From 5578689a4e3c04f2d43ea39736fd3fa396d80c6e Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 4 Mar 2008 13:47:47 -0800
Subject: [NETNS][IPV6] route6 - make route6 per namespace

This patch makes the routing engine use the network namespaces to
access routing informations: Add a network namespace parameter to
ipv6_route_ioctl and propagate the network namespace value to all the
routing code that have not yet been changed.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h |  4 +-
 net/ipv6/af_inet6.c     |  3 +-
 net/ipv6/route.c        | 98 ++++++++++++++++++++++++-------------------------
 3 files changed, 54 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 1444d358836d..2bcbfb826530 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -49,7 +49,9 @@ extern struct dst_entry *	ip6_route_output(struct sock *sk,
 extern int			ip6_route_init(void);
 extern void			ip6_route_cleanup(void);
 
-extern int			ipv6_route_ioctl(unsigned int cmd, void __user *arg);
+extern int			ipv6_route_ioctl(struct net *net,
+						 unsigned int cmd,
+						 void __user *arg);
 
 extern int			ip6_route_add(struct fib6_config *cfg);
 extern int			ip6_ins_rt(struct rt6_info *);
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 73021d5baece..60b8a2254046 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -440,6 +440,7 @@ EXPORT_SYMBOL(inet6_getname);
 int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 {
 	struct sock *sk = sock->sk;
+	struct net *net = sk->sk_net;
 
 	switch(cmd)
 	{
@@ -452,7 +453,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 	case SIOCADDRT:
 	case SIOCDELRT:
 
-		return(ipv6_route_ioctl(cmd,(void __user *)arg));
+		return(ipv6_route_ioctl(net, cmd, (void __user *)arg));
 
 	case SIOCSIFADDR:
 		return addrconf_add_ifaddr((void __user *) arg);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index ad3d684e544a..b3ac4901af86 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -609,7 +609,7 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info)
 int ip6_ins_rt(struct rt6_info *rt)
 {
 	struct nl_info info = {
-		.nl_net = &init_net,
+		.nl_net = rt->rt6i_dev->nd_net,
 	};
 	return __ip6_ins_rt(rt, &info);
 }
@@ -746,6 +746,7 @@ static struct rt6_info *ip6_pol_route_input(struct fib6_table *table,
 void ip6_route_input(struct sk_buff *skb)
 {
 	struct ipv6hdr *iph = ipv6_hdr(skb);
+	struct net *net = skb->dev->nd_net;
 	int flags = RT6_LOOKUP_F_HAS_SADDR;
 	struct flowi fl = {
 		.iif = skb->dev->ifindex,
@@ -763,7 +764,7 @@ void ip6_route_input(struct sk_buff *skb)
 	if (rt6_need_strict(&iph->daddr))
 		flags |= RT6_LOOKUP_F_IFACE;
 
-	skb->dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_input);
+	skb->dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input);
 }
 
 static struct rt6_info *ip6_pol_route_output(struct fib6_table *table,
@@ -891,12 +892,12 @@ static void ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
 
 static int ipv6_get_mtu(struct net_device *dev);
 
-static inline unsigned int ipv6_advmss(unsigned int mtu)
+static inline unsigned int ipv6_advmss(struct net *net, unsigned int mtu)
 {
 	mtu -= sizeof(struct ipv6hdr) + sizeof(struct tcphdr);
 
-	if (mtu < init_net.ipv6.sysctl.ip6_rt_min_advmss)
-		mtu = init_net.ipv6.sysctl.ip6_rt_min_advmss;
+	if (mtu < net->ipv6.sysctl.ip6_rt_min_advmss)
+		mtu = net->ipv6.sysctl.ip6_rt_min_advmss;
 
 	/*
 	 * Maximal non-jumbo IPv6 payload is IPV6_MAXPLEN and
@@ -918,6 +919,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 {
 	struct rt6_info *rt;
 	struct inet6_dev *idev = in6_dev_get(dev);
+	struct net *net = dev->nd_net;
 
 	if (unlikely(idev == NULL))
 		return NULL;
@@ -940,7 +942,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 	atomic_set(&rt->u.dst.__refcnt, 1);
 	rt->u.dst.metrics[RTAX_HOPLIMIT-1] = 255;
 	rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev);
-	rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
+	rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst));
 	rt->u.dst.output  = ip6_output;
 
 #if 0	/* there's no chance to use these for ndisc */
@@ -956,7 +958,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 	icmp6_dst_gc_list = &rt->u.dst;
 	spin_unlock_bh(&icmp6_dst_lock);
 
-	fib6_force_start_gc(dev->nd_net);
+	fib6_force_start_gc(net);
 
 out:
 	return &rt->u.dst;
@@ -1049,6 +1051,7 @@ int ipv6_get_hoplimit(struct net_device *dev)
 int ip6_route_add(struct fib6_config *cfg)
 {
 	int err;
+	struct net *net = cfg->fc_nlinfo.nl_net;
 	struct rt6_info *rt = NULL;
 	struct net_device *dev = NULL;
 	struct inet6_dev *idev = NULL;
@@ -1063,7 +1066,7 @@ int ip6_route_add(struct fib6_config *cfg)
 #endif
 	if (cfg->fc_ifindex) {
 		err = -ENODEV;
-		dev = dev_get_by_index(&init_net, cfg->fc_ifindex);
+		dev = dev_get_by_index(net, cfg->fc_ifindex);
 		if (!dev)
 			goto out;
 		idev = in6_dev_get(dev);
@@ -1074,7 +1077,7 @@ int ip6_route_add(struct fib6_config *cfg)
 	if (cfg->fc_metric == 0)
 		cfg->fc_metric = IP6_RT_PRIO_USER;
 
-	table = fib6_new_table(&init_net, cfg->fc_table);
+	table = fib6_new_table(net, cfg->fc_table);
 	if (table == NULL) {
 		err = -ENOBUFS;
 		goto out;
@@ -1121,12 +1124,12 @@ int ip6_route_add(struct fib6_config *cfg)
 	if ((cfg->fc_flags & RTF_REJECT) ||
 	    (dev && (dev->flags&IFF_LOOPBACK) && !(addr_type&IPV6_ADDR_LOOPBACK))) {
 		/* hold loopback dev/idev if we haven't done so. */
-		if (dev != init_net.loopback_dev) {
+		if (dev != net->loopback_dev) {
 			if (dev) {
 				dev_put(dev);
 				in6_dev_put(idev);
 			}
-			dev = init_net.loopback_dev;
+			dev = net->loopback_dev;
 			dev_hold(dev);
 			idev = in6_dev_get(dev);
 			if (!idev) {
@@ -1163,7 +1166,7 @@ int ip6_route_add(struct fib6_config *cfg)
 			if (!(gwa_type&IPV6_ADDR_UNICAST))
 				goto out;
 
-			grt = rt6_lookup(&init_net, gw_addr, NULL, cfg->fc_ifindex, 1);
+			grt = rt6_lookup(net, gw_addr, NULL, cfg->fc_ifindex, 1);
 
 			err = -EHOSTUNREACH;
 			if (grt == NULL)
@@ -1230,7 +1233,7 @@ install_route:
 	if (!rt->u.dst.metrics[RTAX_MTU-1])
 		rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(dev);
 	if (!rt->u.dst.metrics[RTAX_ADVMSS-1])
-		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
+		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst));
 	rt->u.dst.dev = dev;
 	rt->rt6i_idev = idev;
 	rt->rt6i_table = table;
@@ -1271,7 +1274,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 int ip6_del_rt(struct rt6_info *rt)
 {
 	struct nl_info info = {
-		.nl_net = &init_net,
+		.nl_net = rt->rt6i_dev->nd_net,
 	};
 	return __ip6_del_rt(rt, &info);
 }
@@ -1283,7 +1286,7 @@ static int ip6_route_del(struct fib6_config *cfg)
 	struct rt6_info *rt;
 	int err = -ESRCH;
 
-	table = fib6_get_table(&init_net, cfg->fc_table);
+	table = fib6_get_table(cfg->fc_nlinfo.nl_net, cfg->fc_table);
 	if (table == NULL)
 		return err;
 
@@ -1382,6 +1385,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest,
 					   struct net_device *dev)
 {
 	int flags = RT6_LOOKUP_F_HAS_SADDR;
+	struct net *net = dev->nd_net;
 	struct ip6rd_flowi rdfl = {
 		.fl = {
 			.oif = dev->ifindex,
@@ -1398,8 +1402,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest,
 	if (rt6_need_strict(dest))
 		flags |= RT6_LOOKUP_F_IFACE;
 
-	return (struct rt6_info *)fib6_rule_lookup(&init_net,
-						   (struct flowi *)&rdfl,
+	return (struct rt6_info *)fib6_rule_lookup(net, (struct flowi *)&rdfl,
 						   flags, __ip6_route_redirect);
 }
 
@@ -1457,7 +1460,8 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 	nrt->rt6i_nexthop = neigh_clone(neigh);
 	/* Reset pmtu, it may be better */
 	nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
-	nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&nrt->u.dst));
+	nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(neigh->dev->nd_net,
+							dst_mtu(&nrt->u.dst));
 
 	if (ip6_ins_rt(nrt))
 		goto out;
@@ -1485,9 +1489,10 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 			struct net_device *dev, u32 pmtu)
 {
 	struct rt6_info *rt, *nrt;
+	struct net *net = dev->nd_net;
 	int allfrag = 0;
 
-	rt = rt6_lookup(dev->nd_net, daddr, saddr, dev->ifindex, 0);
+	rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0);
 	if (rt == NULL)
 		return;
 
@@ -1520,7 +1525,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		rt->u.dst.metrics[RTAX_MTU-1] = pmtu;
 		if (allfrag)
 			rt->u.dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
-		dst_set_expires(&rt->u.dst, init_net.ipv6.sysctl.ip6_rt_mtu_expires);
+		dst_set_expires(&rt->u.dst, net->ipv6.sysctl.ip6_rt_mtu_expires);
 		rt->rt6i_flags |= RTF_MODIFIED|RTF_EXPIRES;
 		goto out;
 	}
@@ -1546,7 +1551,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		 * which is 10 mins. After 10 mins the decreased pmtu is expired
 		 * and detecting PMTU increase will be automatically happened.
 		 */
-		dst_set_expires(&nrt->u.dst, init_net.ipv6.sysctl.ip6_rt_mtu_expires);
+		dst_set_expires(&nrt->u.dst, net->ipv6.sysctl.ip6_rt_mtu_expires);
 		nrt->rt6i_flags |= RTF_DYNAMIC|RTF_EXPIRES;
 
 		ip6_ins_rt(nrt);
@@ -1659,7 +1664,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
 	struct rt6_info *rt;
 	struct fib6_table *table;
 
-	table = fib6_get_table(&init_net, RT6_TABLE_DFLT);
+	table = fib6_get_table(dev->nd_net, RT6_TABLE_DFLT);
 	if (table == NULL)
 		return NULL;
 
@@ -1688,6 +1693,9 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 		.fc_ifindex	= dev->ifindex,
 		.fc_flags	= RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT |
 				  RTF_UP | RTF_EXPIRES | RTF_PREF(pref),
+		.fc_nlinfo.pid = 0,
+		.fc_nlinfo.nlh = NULL,
+		.fc_nlinfo.nl_net = dev->nd_net,
 	};
 
 	ipv6_addr_copy(&cfg.fc_gateway, gwaddr);
@@ -1720,7 +1728,8 @@ restart:
 	read_unlock_bh(&table->tb6_lock);
 }
 
-static void rtmsg_to_fib6_config(struct in6_rtmsg *rtmsg,
+static void rtmsg_to_fib6_config(struct net *net,
+				 struct in6_rtmsg *rtmsg,
 				 struct fib6_config *cfg)
 {
 	memset(cfg, 0, sizeof(*cfg));
@@ -1733,14 +1742,14 @@ static void rtmsg_to_fib6_config(struct in6_rtmsg *rtmsg,
 	cfg->fc_src_len = rtmsg->rtmsg_src_len;
 	cfg->fc_flags = rtmsg->rtmsg_flags;
 
-	cfg->fc_nlinfo.nl_net = &init_net;
+	cfg->fc_nlinfo.nl_net = net;
 
 	ipv6_addr_copy(&cfg->fc_dst, &rtmsg->rtmsg_dst);
 	ipv6_addr_copy(&cfg->fc_src, &rtmsg->rtmsg_src);
 	ipv6_addr_copy(&cfg->fc_gateway, &rtmsg->rtmsg_gateway);
 }
 
-int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
+int ipv6_route_ioctl(struct net *net, unsigned int cmd, void __user *arg)
 {
 	struct fib6_config cfg;
 	struct in6_rtmsg rtmsg;
@@ -1756,7 +1765,7 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
 		if (err)
 			return -EFAULT;
 
-		rtmsg_to_fib6_config(&rtmsg, &cfg);
+		rtmsg_to_fib6_config(net, &rtmsg, &cfg);
 
 		rtnl_lock();
 		switch (cmd) {
@@ -1835,21 +1844,22 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 				    const struct in6_addr *addr,
 				    int anycast)
 {
+	struct net *net = idev->dev->nd_net;
 	struct rt6_info *rt = ip6_dst_alloc();
 
 	if (rt == NULL)
 		return ERR_PTR(-ENOMEM);
 
-	dev_hold(init_net.loopback_dev);
+	dev_hold(net->loopback_dev);
 	in6_dev_hold(idev);
 
 	rt->u.dst.flags = DST_HOST;
 	rt->u.dst.input = ip6_input;
 	rt->u.dst.output = ip6_output;
-	rt->rt6i_dev = init_net.loopback_dev;
+	rt->rt6i_dev = net->loopback_dev;
 	rt->rt6i_idev = idev;
 	rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev);
-	rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
+	rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst));
 	rt->u.dst.metrics[RTAX_HOPLIMIT-1] = -1;
 	rt->u.dst.obsolete = -1;
 
@@ -1866,7 +1876,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 
 	ipv6_addr_copy(&rt->rt6i_dst.addr, addr);
 	rt->rt6i_dst.plen = 128;
-	rt->rt6i_table = fib6_get_table(&init_net, RT6_TABLE_LOCAL);
+	rt->rt6i_table = fib6_get_table(net, RT6_TABLE_LOCAL);
 
 	atomic_set(&rt->u.dst.__refcnt, 1);
 
@@ -1898,6 +1908,7 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg)
 {
 	struct rt6_mtu_change_arg *arg = (struct rt6_mtu_change_arg *) p_arg;
 	struct inet6_dev *idev;
+	struct net *net = arg->dev->nd_net;
 
 	/* In IPv6 pmtu discovery is not optional,
 	   so that RTAX_MTU lock cannot disable it.
@@ -1929,7 +1940,7 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg)
 	     (dst_mtu(&rt->u.dst) < arg->mtu &&
 	      dst_mtu(&rt->u.dst) == idev->cnf.mtu6))) {
 		rt->u.dst.metrics[RTAX_MTU-1] = arg->mtu;
-		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(arg->mtu);
+		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, arg->mtu);
 	}
 	return 0;
 }
@@ -2024,13 +2035,9 @@ errout:
 
 static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
 	struct fib6_config cfg;
 	int err;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	err = rtm_to_fib6_config(skb, nlh, &cfg);
 	if (err < 0)
 		return err;
@@ -2040,13 +2047,9 @@ static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *a
 
 static int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
 	struct fib6_config cfg;
 	int err;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	err = rtm_to_fib6_config(skb, nlh, &cfg);
 	if (err < 0)
 		return err;
@@ -2190,9 +2193,6 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 	struct flowi fl;
 	int err, iif = 0;
 
-	if (net != &init_net)
-		return -EINVAL;
-
 	err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy);
 	if (err < 0)
 		goto errout;
@@ -2222,7 +2222,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 
 	if (iif) {
 		struct net_device *dev;
-		dev = __dev_get_by_index(&init_net, iif);
+		dev = __dev_get_by_index(net, iif);
 		if (!dev) {
 			err = -ENODEV;
 			goto errout;
@@ -2252,7 +2252,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 		goto errout;
 	}
 
-	err = rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid);
+	err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).pid);
 errout:
 	return err;
 }
@@ -2260,6 +2260,7 @@ errout:
 void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
 {
 	struct sk_buff *skb;
+	struct net *net = info->nl_net;
 	u32 seq;
 	int err;
 
@@ -2278,11 +2279,11 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
 		kfree_skb(skb);
 		goto errout;
 	}
-	err = rtnl_notify(skb, &init_net, info->pid,
-				RTNLGRP_IPV6_ROUTE, info->nlh, gfp_any());
+	err = rtnl_notify(skb, net, info->pid, RTNLGRP_IPV6_ROUTE,
+			  info->nlh, gfp_any());
 errout:
 	if (err < 0)
-		rtnl_set_sk_err(&init_net, RTNLGRP_IPV6_ROUTE, err);
+		rtnl_set_sk_err(net, RTNLGRP_IPV6_ROUTE, err);
 }
 
 /*
@@ -2544,6 +2545,7 @@ static void ip6_route_net_exit(struct net *net)
 	proc_net_remove(net, "ipv6_route");
 	proc_net_remove(net, "rt6_stats");
 #endif
+	rt6_ifdown(net, NULL);
 }
 
 static struct pernet_operations ip6_route_net_ops = {
@@ -2592,7 +2594,6 @@ fib6_rules_init:
 xfrm6_init:
 	xfrm6_fini();
 out_fib6_init:
-	rt6_ifdown(&init_net, NULL);
 	fib6_gc_cleanup();
 out_kmem_cache:
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
@@ -2604,7 +2605,6 @@ void ip6_route_cleanup(void)
 	unregister_pernet_subsys(&ip6_route_net_ops);
 	fib6_rules_cleanup();
 	xfrm6_fini();
-	rt6_ifdown(&init_net, NULL);
 	fib6_gc_cleanup();
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
 }
-- 
cgit v1.2.3


From bdb3289f739e94bcae8b51972ae844ec66c2f4df Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 4 Mar 2008 13:48:10 -0800
Subject: [NETNS][IPV6] rt6_info - make rt6_info accessed as a pointer

This patch make mindless changes and prepares the code to use dynamic
allocation for rt6_info structure. The code accesses the rt6_info
structure as a pointer instead of a global static variable.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h |  6 ++---
 net/ipv6/addrconf.c     | 12 ++++-----
 net/ipv6/fib6_rules.c   | 12 ++++-----
 net/ipv6/ip6_fib.c      | 18 ++++++-------
 net/ipv6/route.c        | 70 ++++++++++++++++++++++++++++++++++++++-----------
 5 files changed, 78 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 2bcbfb826530..e0caed25bfba 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -34,11 +34,11 @@ struct route_info {
 #define RT6_LOOKUP_F_REACHABLE	0x2
 #define RT6_LOOKUP_F_HAS_SADDR	0x4
 
-extern struct rt6_info	ip6_null_entry;
+extern struct rt6_info	*ip6_null_entry;
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
-extern struct rt6_info	ip6_prohibit_entry;
-extern struct rt6_info	ip6_blk_hole_entry;
+extern struct rt6_info	*ip6_prohibit_entry;
+extern struct rt6_info	*ip6_blk_hole_entry;
 #endif
 
 extern void			ip6_route_input(struct sk_buff *skb);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index b37ae421b61b..3192a848a53a 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -4301,13 +4301,13 @@ int __init addrconf_init(void)
 	if (err)
 		goto errlo;
 
-	ip6_null_entry.u.dst.dev = init_net.loopback_dev;
-	ip6_null_entry.rt6i_idev = in6_dev_get(init_net.loopback_dev);
+	ip6_null_entry->u.dst.dev = init_net.loopback_dev;
+	ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	ip6_prohibit_entry.u.dst.dev = init_net.loopback_dev;
-	ip6_prohibit_entry.rt6i_idev = in6_dev_get(init_net.loopback_dev);
-	ip6_blk_hole_entry.u.dst.dev = init_net.loopback_dev;
-	ip6_blk_hole_entry.rt6i_idev = in6_dev_get(init_net.loopback_dev);
+	ip6_prohibit_entry->u.dst.dev = init_net.loopback_dev;
+	ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
+	ip6_blk_hole_entry->u.dst.dev = init_net.loopback_dev;
+	ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
 #endif
 
 	register_netdevice_notifier(&ipv6_dev_notf);
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 89cb092c9732..c00055f232c4 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -43,8 +43,8 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
 	if (arg.result)
 		return arg.result;
 
-	dst_hold(&ip6_null_entry.u.dst);
-	return &ip6_null_entry.u.dst;
+	dst_hold(&ip6_null_entry->u.dst);
+	return &ip6_null_entry->u.dst;
 }
 
 static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
@@ -58,14 +58,14 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 	case FR_ACT_TO_TBL:
 		break;
 	case FR_ACT_UNREACHABLE:
-		rt = &ip6_null_entry;
+		rt = ip6_null_entry;
 		goto discard_pkt;
 	default:
 	case FR_ACT_BLACKHOLE:
-		rt = &ip6_blk_hole_entry;
+		rt = ip6_blk_hole_entry;
 		goto discard_pkt;
 	case FR_ACT_PROHIBIT:
-		rt = &ip6_prohibit_entry;
+		rt = ip6_prohibit_entry;
 		goto discard_pkt;
 	}
 
@@ -73,7 +73,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 	if (table)
 		rt = lookup(table, flp, flags);
 
-	if (rt != &ip6_null_entry) {
+	if (rt != ip6_null_entry) {
 		struct fib6_rule *r = (struct fib6_rule *)rule;
 
 		/*
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 1c2566e3d392..f028f7a1a446 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -200,7 +200,7 @@ static struct fib6_table *fib6_alloc_table(u32 id)
 	table = kzalloc(sizeof(*table), GFP_ATOMIC);
 	if (table != NULL) {
 		table->tb6_id = id;
-		table->tb6_root.leaf = &ip6_null_entry;
+		table->tb6_root.leaf = ip6_null_entry;
 		table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 	}
 
@@ -717,8 +717,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info)
 			if (sfn == NULL)
 				goto st_failure;
 
-			sfn->leaf = &ip6_null_entry;
-			atomic_inc(&ip6_null_entry.rt6i_ref);
+			sfn->leaf = ip6_null_entry;
+			atomic_inc(&ip6_null_entry->rt6i_ref);
 			sfn->fn_flags = RTN_ROOT;
 			sfn->fn_sernum = fib6_new_sernum();
 
@@ -777,7 +777,7 @@ out:
 #if RT6_DEBUG >= 2
 			if (!pn->leaf) {
 				BUG_TRAP(pn->leaf != NULL);
-				pn->leaf = &ip6_null_entry;
+				pn->leaf = ip6_null_entry;
 			}
 #endif
 			atomic_inc(&pn->leaf->rt6i_ref);
@@ -962,7 +962,7 @@ struct fib6_node * fib6_locate(struct fib6_node *root,
 static struct rt6_info * fib6_find_prefix(struct fib6_node *fn)
 {
 	if (fn->fn_flags&RTN_ROOT)
-		return &ip6_null_entry;
+		return ip6_null_entry;
 
 	while(fn) {
 		if(fn->left)
@@ -1012,7 +1012,7 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
 #if RT6_DEBUG >= 2
 			if (fn->leaf==NULL) {
 				BUG_TRAP(fn->leaf);
-				fn->leaf = &ip6_null_entry;
+				fn->leaf = ip6_null_entry;
 			}
 #endif
 			atomic_inc(&fn->leaf->rt6i_ref);
@@ -1154,7 +1154,7 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info)
 		return -ENOENT;
 	}
 #endif
-	if (fn == NULL || rt == &ip6_null_entry)
+	if (fn == NULL || rt == ip6_null_entry)
 		return -ENOENT;
 
 	BUG_TRAP(fn->fn_flags&RTN_RTINFO);
@@ -1501,7 +1501,7 @@ static int fib6_net_init(struct net *net)
 		goto out_fib_table_hash;
 
 	net->ipv6.fib6_main_tbl->tb6_id = RT6_TABLE_MAIN;
-	net->ipv6.fib6_main_tbl->tb6_root.leaf = &ip6_null_entry;
+	net->ipv6.fib6_main_tbl->tb6_root.leaf = ip6_null_entry;
 	net->ipv6.fib6_main_tbl->tb6_root.fn_flags =
 		RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 
@@ -1511,7 +1511,7 @@ static int fib6_net_init(struct net *net)
 	if (!net->ipv6.fib6_local_tbl)
 		goto out_fib6_main_tbl;
 	net->ipv6.fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL;
-	net->ipv6.fib6_local_tbl->tb6_root.leaf = &ip6_null_entry;
+	net->ipv6.fib6_local_tbl->tb6_root.leaf = ip6_null_entry;
 	net->ipv6.fib6_local_tbl->tb6_root.fn_flags =
 		RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 #endif
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index b3ac4901af86..8f954c1e961f 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -127,7 +127,7 @@ static struct dst_ops ip6_dst_blackhole_ops = {
 	.entries		=	ATOMIC_INIT(0),
 };
 
-struct rt6_info ip6_null_entry = {
+static struct rt6_info ip6_null_entry_template = {
 	.u = {
 		.dst = {
 			.__refcnt	= ATOMIC_INIT(1),
@@ -138,7 +138,6 @@ struct rt6_info ip6_null_entry = {
 			.input		= ip6_pkt_discard,
 			.output		= ip6_pkt_discard_out,
 			.ops		= &ip6_dst_ops,
-			.path		= (struct dst_entry*)&ip6_null_entry,
 		}
 	},
 	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
@@ -146,12 +145,14 @@ struct rt6_info ip6_null_entry = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
+struct rt6_info *ip6_null_entry;
+
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 
 static int ip6_pkt_prohibit(struct sk_buff *skb);
 static int ip6_pkt_prohibit_out(struct sk_buff *skb);
 
-struct rt6_info ip6_prohibit_entry = {
+struct rt6_info ip6_prohibit_entry_template = {
 	.u = {
 		.dst = {
 			.__refcnt	= ATOMIC_INIT(1),
@@ -162,7 +163,6 @@ struct rt6_info ip6_prohibit_entry = {
 			.input		= ip6_pkt_prohibit,
 			.output		= ip6_pkt_prohibit_out,
 			.ops		= &ip6_dst_ops,
-			.path		= (struct dst_entry*)&ip6_prohibit_entry,
 		}
 	},
 	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
@@ -170,7 +170,9 @@ struct rt6_info ip6_prohibit_entry = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
-struct rt6_info ip6_blk_hole_entry = {
+struct rt6_info *ip6_prohibit_entry;
+
+static struct rt6_info ip6_blk_hole_entry_template = {
 	.u = {
 		.dst = {
 			.__refcnt	= ATOMIC_INIT(1),
@@ -181,7 +183,6 @@ struct rt6_info ip6_blk_hole_entry = {
 			.input		= dst_discard,
 			.output		= dst_discard,
 			.ops		= &ip6_dst_ops,
-			.path		= (struct dst_entry*)&ip6_blk_hole_entry,
 		}
 	},
 	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
@@ -189,6 +190,8 @@ struct rt6_info ip6_blk_hole_entry = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
+struct rt6_info *ip6_blk_hole_entry;
+
 #endif
 
 /* allocate dst with ip6_dst_ops */
@@ -271,7 +274,7 @@ static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt,
 			return local;
 
 		if (strict)
-			return &ip6_null_entry;
+			return ip6_null_entry;
 	}
 	return rt;
 }
@@ -437,7 +440,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 	RT6_TRACE("%s() => %p\n",
 		  __FUNCTION__, match);
 
-	return (match ? match : &ip6_null_entry);
+	return (match ? match : ip6_null_entry);
 }
 
 #ifdef CONFIG_IPV6_ROUTE_INFO
@@ -522,7 +525,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 
 #define BACKTRACK(saddr) \
 do { \
-	if (rt == &ip6_null_entry) { \
+	if (rt == ip6_null_entry) { \
 		struct fib6_node *pn; \
 		while (1) { \
 			if (fn->fn_flags & RTN_TL_ROOT) \
@@ -686,7 +689,7 @@ restart_2:
 restart:
 	rt = rt6_select(fn, oif, strict | reachable);
 	BACKTRACK(&fl->fl6_src);
-	if (rt == &ip6_null_entry ||
+	if (rt == ip6_null_entry ||
 	    rt->rt6i_flags & RTF_CACHE)
 		goto out;
 
@@ -704,7 +707,7 @@ restart:
 	}
 
 	dst_release(&rt->u.dst);
-	rt = nrt ? : &ip6_null_entry;
+	rt = nrt ? : ip6_null_entry;
 
 	dst_hold(&rt->u.dst);
 	if (nrt) {
@@ -1257,7 +1260,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 	int err;
 	struct fib6_table *table;
 
-	if (rt == &ip6_null_entry)
+	if (rt == ip6_null_entry)
 		return -ENOENT;
 
 	table = rt->rt6i_table;
@@ -1369,7 +1372,7 @@ restart:
 	}
 
 	if (!rt)
-		rt = &ip6_null_entry;
+		rt = ip6_null_entry;
 	BACKTRACK(&fl->fl6_src);
 out:
 	dst_hold(&rt->u.dst);
@@ -1415,7 +1418,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 
 	rt = ip6_route_redirect(dest, src, saddr, neigh->dev);
 
-	if (rt == &ip6_null_entry) {
+	if (rt == ip6_null_entry) {
 		if (net_ratelimit())
 			printk(KERN_DEBUG "rt6_redirect: source isn't a valid nexthop "
 			       "for redirect target\n");
@@ -1886,7 +1889,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 static int fib6_ifdown(struct rt6_info *rt, void *arg)
 {
 	if (((void*)rt->rt6i_dev == arg || arg == NULL) &&
-	    rt != &ip6_null_entry) {
+	    rt != ip6_null_entry) {
 		RT6_TRACE("deleted by ifdown %p\n", rt);
 		return -1;
 	}
@@ -2565,9 +2568,30 @@ int __init ip6_route_init(void)
 
 	ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops.kmem_cachep;
 
+	ret = -ENOMEM;
+	ip6_null_entry = kmemdup(&ip6_null_entry_template,
+				 sizeof(*ip6_null_entry), GFP_KERNEL);
+	if (!ip6_null_entry)
+		goto out_kmem_cache;
+	ip6_null_entry->u.dst.path = (struct dst_entry *)ip6_null_entry;
+
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
+				     sizeof(*ip6_prohibit_entry), GFP_KERNEL);
+	if (!ip6_prohibit_entry)
+		goto out_ip6_null_entry;
+	ip6_prohibit_entry->u.dst.path = (struct dst_entry *)ip6_prohibit_entry;
+
+	ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
+				     sizeof(*ip6_blk_hole_entry), GFP_KERNEL);
+	if (!ip6_blk_hole_entry)
+		goto out_ip6_prohibit_entry;
+	ip6_blk_hole_entry->u.dst.path = (struct dst_entry *)ip6_blk_hole_entry;
+#endif
+
 	ret = fib6_init();
 	if (ret)
-		goto out_kmem_cache;
+		goto out_ip6_blk_hole_entry;
 
 	ret = xfrm6_init();
 	if (ret)
@@ -2595,6 +2619,14 @@ xfrm6_init:
 	xfrm6_fini();
 out_fib6_init:
 	fib6_gc_cleanup();
+out_ip6_blk_hole_entry:
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	kfree(ip6_blk_hole_entry);
+out_ip6_prohibit_entry:
+	kfree(ip6_prohibit_entry);
+out_ip6_null_entry:
+#endif
+	kfree(ip6_null_entry);
 out_kmem_cache:
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
 	goto out;
@@ -2607,4 +2639,10 @@ void ip6_route_cleanup(void)
 	xfrm6_fini();
 	fib6_gc_cleanup();
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
+
+	kfree(ip6_null_entry);
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	kfree(ip6_prohibit_entry);
+	kfree(ip6_blk_hole_entry);
+#endif
 }
-- 
cgit v1.2.3


From 8ed677896752fff056f6cf3d7ce462adc6c464f0 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Tue, 4 Mar 2008 13:48:30 -0800
Subject: [NETNS][IPV6] rt6_info - move rt6_info structure inside the namespace

The rt6_info structures are moved inside the network namespace
structure. All references to these structures are now relative to the
initial network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h    |   3 +-
 include/net/netns/ipv6.h |   3 +
 net/ipv6/addrconf.c      |   9 ---
 net/ipv6/fib6_rules.c    |  17 ++--
 net/ipv6/ip6_fib.c       |  45 ++++++-----
 net/ipv6/route.c         | 202 +++++++++++++++++++++++++++++++----------------
 6 files changed, 171 insertions(+), 108 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index fae267f65341..7c5c0f79168a 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -174,7 +174,8 @@ struct fib6_table {
 #define RT6_TABLE_LOCAL		RT6_TABLE_MAIN
 #endif
 
-typedef struct rt6_info *(*pol_lookup_t)(struct fib6_table *,
+typedef struct rt6_info *(*pol_lookup_t)(struct net *,
+					 struct fib6_table *,
 					 struct flowi *, int);
 
 /*
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index c6c9afff13ef..311a942f36e9 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -36,11 +36,14 @@ struct netns_ipv6 {
 	struct xt_table		*ip6table_mangle;
 	struct xt_table		*ip6table_raw;
 #endif
+	struct rt6_info         *ip6_null_entry;
 	struct rt6_statistics   *rt6_stats;
 	struct timer_list       *ip6_fib_timer;
 	struct hlist_head       *fib_table_hash;
 	struct fib6_table       *fib6_main_tbl;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	struct rt6_info         *ip6_prohibit_entry;
+	struct rt6_info         *ip6_blk_hole_entry;
 	struct fib6_table       *fib6_local_tbl;
 	struct fib_rules_ops    *fib6_rules_ops;
 #endif
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 3192a848a53a..17b06d220e95 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -4301,15 +4301,6 @@ int __init addrconf_init(void)
 	if (err)
 		goto errlo;
 
-	ip6_null_entry->u.dst.dev = init_net.loopback_dev;
-	ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	ip6_prohibit_entry->u.dst.dev = init_net.loopback_dev;
-	ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
-	ip6_blk_hole_entry->u.dst.dev = init_net.loopback_dev;
-	ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
-#endif
-
 	register_netdevice_notifier(&ipv6_dev_notf);
 
 	addrconf_verify(0);
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index c00055f232c4..55137408f054 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -43,8 +43,8 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
 	if (arg.result)
 		return arg.result;
 
-	dst_hold(&ip6_null_entry->u.dst);
-	return &ip6_null_entry->u.dst;
+	dst_hold(&net->ipv6.ip6_null_entry->u.dst);
+	return &net->ipv6.ip6_null_entry->u.dst;
 }
 
 static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
@@ -52,28 +52,29 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 {
 	struct rt6_info *rt = NULL;
 	struct fib6_table *table;
+	struct net *net = rule->fr_net;
 	pol_lookup_t lookup = arg->lookup_ptr;
 
 	switch (rule->action) {
 	case FR_ACT_TO_TBL:
 		break;
 	case FR_ACT_UNREACHABLE:
-		rt = ip6_null_entry;
+		rt = net->ipv6.ip6_null_entry;
 		goto discard_pkt;
 	default:
 	case FR_ACT_BLACKHOLE:
-		rt = ip6_blk_hole_entry;
+		rt = net->ipv6.ip6_blk_hole_entry;
 		goto discard_pkt;
 	case FR_ACT_PROHIBIT:
-		rt = ip6_prohibit_entry;
+		rt = net->ipv6.ip6_prohibit_entry;
 		goto discard_pkt;
 	}
 
-	table = fib6_get_table(rule->fr_net, rule->table);
+	table = fib6_get_table(net, rule->table);
 	if (table)
-		rt = lookup(table, flp, flags);
+		rt = lookup(net, table, flp, flags);
 
-	if (rt != ip6_null_entry) {
+	if (rt != net->ipv6.ip6_null_entry) {
 		struct fib6_rule *r = (struct fib6_rule *)rule;
 
 		/*
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index f028f7a1a446..b0814b0082e7 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -79,8 +79,8 @@ static DEFINE_RWLOCK(fib6_walker_lock);
 
 static void fib6_prune_clones(struct net *net, struct fib6_node *fn,
 			      struct rt6_info *rt);
-static struct rt6_info * fib6_find_prefix(struct fib6_node *fn);
-static struct fib6_node * fib6_repair_tree(struct fib6_node *fn);
+static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn);
+static struct fib6_node *fib6_repair_tree(struct net *net, struct fib6_node *fn);
 static int fib6_walk(struct fib6_walker_t *w);
 static int fib6_walk_continue(struct fib6_walker_t *w);
 
@@ -193,14 +193,14 @@ static void fib6_link_table(struct net *net, struct fib6_table *tb)
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 
-static struct fib6_table *fib6_alloc_table(u32 id)
+static struct fib6_table *fib6_alloc_table(struct net *net, u32 id)
 {
 	struct fib6_table *table;
 
 	table = kzalloc(sizeof(*table), GFP_ATOMIC);
 	if (table != NULL) {
 		table->tb6_id = id;
-		table->tb6_root.leaf = ip6_null_entry;
+		table->tb6_root.leaf = net->ipv6.ip6_null_entry;
 		table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 	}
 
@@ -217,7 +217,7 @@ struct fib6_table *fib6_new_table(struct net *net, u32 id)
 	if (tb)
 		return tb;
 
-	tb = fib6_alloc_table(id);
+	tb = fib6_alloc_table(net, id);
 	if (tb != NULL)
 		fib6_link_table(net, tb);
 
@@ -267,7 +267,7 @@ struct fib6_table *fib6_get_table(struct net *net, u32 id)
 struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
 				   int flags, pol_lookup_t lookup)
 {
-	return (struct dst_entry *) lookup(net->ipv6.fib6_main_tbl, fl, flags);
+	return (struct dst_entry *) lookup(net, net->ipv6.fib6_main_tbl, fl, flags);
 }
 
 static void fib6_tables_init(struct net *net)
@@ -717,8 +717,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info)
 			if (sfn == NULL)
 				goto st_failure;
 
-			sfn->leaf = ip6_null_entry;
-			atomic_inc(&ip6_null_entry->rt6i_ref);
+			sfn->leaf = info->nl_net->ipv6.ip6_null_entry;
+			atomic_inc(&info->nl_net->ipv6.ip6_null_entry->rt6i_ref);
 			sfn->fn_flags = RTN_ROOT;
 			sfn->fn_sernum = fib6_new_sernum();
 
@@ -773,11 +773,11 @@ out:
 		 * super-tree leaf node we have to find a new one for it.
 		 */
 		if (pn != fn && !pn->leaf && !(pn->fn_flags & RTN_RTINFO)) {
-			pn->leaf = fib6_find_prefix(pn);
+			pn->leaf = fib6_find_prefix(info->nl_net, pn);
 #if RT6_DEBUG >= 2
 			if (!pn->leaf) {
 				BUG_TRAP(pn->leaf != NULL);
-				pn->leaf = ip6_null_entry;
+				pn->leaf = info->nl_net->ipv6.ip6_null_entry;
 			}
 #endif
 			atomic_inc(&pn->leaf->rt6i_ref);
@@ -793,7 +793,7 @@ out:
 	 */
 st_failure:
 	if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT)))
-		fib6_repair_tree(fn);
+		fib6_repair_tree(info->nl_net, fn);
 	dst_free(&rt->u.dst);
 	return err;
 #endif
@@ -959,10 +959,10 @@ struct fib6_node * fib6_locate(struct fib6_node *root,
  *
  */
 
-static struct rt6_info * fib6_find_prefix(struct fib6_node *fn)
+static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn)
 {
 	if (fn->fn_flags&RTN_ROOT)
-		return ip6_null_entry;
+		return net->ipv6.ip6_null_entry;
 
 	while(fn) {
 		if(fn->left)
@@ -981,7 +981,8 @@ static struct rt6_info * fib6_find_prefix(struct fib6_node *fn)
  *	is the node we want to try and remove.
  */
 
-static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
+static struct fib6_node *fib6_repair_tree(struct net *net,
+					   struct fib6_node *fn)
 {
 	int children;
 	int nstate;
@@ -1008,11 +1009,11 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
 		    || (children && fn->fn_flags&RTN_ROOT)
 #endif
 		    ) {
-			fn->leaf = fib6_find_prefix(fn);
+			fn->leaf = fib6_find_prefix(net, fn);
 #if RT6_DEBUG >= 2
 			if (fn->leaf==NULL) {
 				BUG_TRAP(fn->leaf);
-				fn->leaf = ip6_null_entry;
+				fn->leaf = net->ipv6.ip6_null_entry;
 			}
 #endif
 			atomic_inc(&fn->leaf->rt6i_ref);
@@ -1117,7 +1118,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 	if (fn->leaf == NULL) {
 		fn->fn_flags &= ~RTN_RTINFO;
 		net->ipv6.rt6_stats->fib_route_nodes--;
-		fn = fib6_repair_tree(fn);
+		fn = fib6_repair_tree(net, fn);
 	}
 
 	if (atomic_read(&rt->rt6i_ref) != 1) {
@@ -1129,7 +1130,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 		 */
 		while (fn) {
 			if (!(fn->fn_flags&RTN_RTINFO) && fn->leaf == rt) {
-				fn->leaf = fib6_find_prefix(fn);
+				fn->leaf = fib6_find_prefix(net, fn);
 				atomic_inc(&fn->leaf->rt6i_ref);
 				rt6_release(rt);
 			}
@@ -1145,6 +1146,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 
 int fib6_del(struct rt6_info *rt, struct nl_info *info)
 {
+	struct net *net = info->nl_net;
 	struct fib6_node *fn = rt->rt6i_node;
 	struct rt6_info **rtp;
 
@@ -1154,7 +1156,7 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info)
 		return -ENOENT;
 	}
 #endif
-	if (fn == NULL || rt == ip6_null_entry)
+	if (fn == NULL || rt == net->ipv6.ip6_null_entry)
 		return -ENOENT;
 
 	BUG_TRAP(fn->fn_flags&RTN_RTINFO);
@@ -1501,7 +1503,7 @@ static int fib6_net_init(struct net *net)
 		goto out_fib_table_hash;
 
 	net->ipv6.fib6_main_tbl->tb6_id = RT6_TABLE_MAIN;
-	net->ipv6.fib6_main_tbl->tb6_root.leaf = ip6_null_entry;
+	net->ipv6.fib6_main_tbl->tb6_root.leaf = net->ipv6.ip6_null_entry;
 	net->ipv6.fib6_main_tbl->tb6_root.fn_flags =
 		RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 
@@ -1511,7 +1513,7 @@ static int fib6_net_init(struct net *net)
 	if (!net->ipv6.fib6_local_tbl)
 		goto out_fib6_main_tbl;
 	net->ipv6.fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL;
-	net->ipv6.fib6_local_tbl->tb6_root.leaf = ip6_null_entry;
+	net->ipv6.fib6_local_tbl->tb6_root.leaf = net->ipv6.ip6_null_entry;
 	net->ipv6.fib6_local_tbl->tb6_root.fn_flags =
 		RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
 #endif
@@ -1536,6 +1538,7 @@ out_timer:
 
 static void fib6_net_exit(struct net *net)
 {
+	rt6_ifdown(net, NULL);
 	del_timer(net->ipv6.ip6_fib_timer);
 	kfree(net->ipv6.ip6_fib_timer);
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 8f954c1e961f..7ff66cebe77c 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -145,8 +145,6 @@ static struct rt6_info ip6_null_entry_template = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
-struct rt6_info *ip6_null_entry;
-
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 
 static int ip6_pkt_prohibit(struct sk_buff *skb);
@@ -170,8 +168,6 @@ struct rt6_info ip6_prohibit_entry_template = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
-struct rt6_info *ip6_prohibit_entry;
-
 static struct rt6_info ip6_blk_hole_entry_template = {
 	.u = {
 		.dst = {
@@ -190,8 +186,6 @@ static struct rt6_info ip6_blk_hole_entry_template = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
-struct rt6_info *ip6_blk_hole_entry;
-
 #endif
 
 /* allocate dst with ip6_dst_ops */
@@ -245,7 +239,8 @@ static inline int rt6_need_strict(struct in6_addr *daddr)
  *	Route lookup. Any table->tb6_lock is implied.
  */
 
-static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt,
+static inline struct rt6_info *rt6_device_match(struct net *net,
+						    struct rt6_info *rt,
 						    int oif,
 						    int strict)
 {
@@ -274,7 +269,7 @@ static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt,
 			return local;
 
 		if (strict)
-			return ip6_null_entry;
+			return net->ipv6.ip6_null_entry;
 	}
 	return rt;
 }
@@ -415,6 +410,7 @@ static struct rt6_info *find_rr_leaf(struct fib6_node *fn,
 static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 {
 	struct rt6_info *match, *rt0;
+	struct net *net;
 
 	RT6_TRACE("%s(fn->leaf=%p, oif=%d)\n",
 		  __FUNCTION__, fn->leaf, oif);
@@ -440,7 +436,8 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 	RT6_TRACE("%s() => %p\n",
 		  __FUNCTION__, match);
 
-	return (match ? match : ip6_null_entry);
+	net = rt0->rt6i_dev->nd_net;
+	return (match ? match : net->ipv6.ip6_null_entry);
 }
 
 #ifdef CONFIG_IPV6_ROUTE_INFO
@@ -523,9 +520,9 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 }
 #endif
 
-#define BACKTRACK(saddr) \
+#define BACKTRACK(__net, saddr)			\
 do { \
-	if (rt == ip6_null_entry) { \
+	if (rt == __net->ipv6.ip6_null_entry) {	\
 		struct fib6_node *pn; \
 		while (1) { \
 			if (fn->fn_flags & RTN_TL_ROOT) \
@@ -541,7 +538,8 @@ do { \
 	} \
 } while(0)
 
-static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table,
+static struct rt6_info *ip6_pol_route_lookup(struct net *net,
+					     struct fib6_table *table,
 					     struct flowi *fl, int flags)
 {
 	struct fib6_node *fn;
@@ -551,8 +549,8 @@ static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table,
 	fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src);
 restart:
 	rt = fn->leaf;
-	rt = rt6_device_match(rt, fl->oif, flags);
-	BACKTRACK(&fl->fl6_src);
+	rt = rt6_device_match(net, rt, fl->oif, flags);
+	BACKTRACK(net, &fl->fl6_src);
 out:
 	dst_use(&rt->u.dst, jiffies);
 	read_unlock_bh(&table->tb6_lock);
@@ -668,8 +666,8 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d
 	return rt;
 }
 
-static struct rt6_info *ip6_pol_route(struct fib6_table *table, int oif,
-					    struct flowi *fl, int flags)
+static struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif,
+				      struct flowi *fl, int flags)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt, *nrt;
@@ -688,8 +686,9 @@ restart_2:
 
 restart:
 	rt = rt6_select(fn, oif, strict | reachable);
-	BACKTRACK(&fl->fl6_src);
-	if (rt == ip6_null_entry ||
+
+	BACKTRACK(net, &fl->fl6_src);
+	if (rt == net->ipv6.ip6_null_entry ||
 	    rt->rt6i_flags & RTF_CACHE)
 		goto out;
 
@@ -707,7 +706,7 @@ restart:
 	}
 
 	dst_release(&rt->u.dst);
-	rt = nrt ? : ip6_null_entry;
+	rt = nrt ? : net->ipv6.ip6_null_entry;
 
 	dst_hold(&rt->u.dst);
 	if (nrt) {
@@ -740,10 +739,10 @@ out2:
 	return rt;
 }
 
-static struct rt6_info *ip6_pol_route_input(struct fib6_table *table,
+static struct rt6_info *ip6_pol_route_input(struct net *net, struct fib6_table *table,
 					    struct flowi *fl, int flags)
 {
-	return ip6_pol_route(table, fl->iif, fl, flags);
+	return ip6_pol_route(net, table, fl->iif, fl, flags);
 }
 
 void ip6_route_input(struct sk_buff *skb)
@@ -770,10 +769,10 @@ void ip6_route_input(struct sk_buff *skb)
 	skb->dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input);
 }
 
-static struct rt6_info *ip6_pol_route_output(struct fib6_table *table,
+static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table *table,
 					     struct flowi *fl, int flags)
 {
-	return ip6_pol_route(table, fl->oif, fl, flags);
+	return ip6_pol_route(net, table, fl->oif, fl, flags);
 }
 
 struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
@@ -1259,8 +1258,9 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 {
 	int err;
 	struct fib6_table *table;
+	struct net *net = rt->rt6i_dev->nd_net;
 
-	if (rt == ip6_null_entry)
+	if (rt == net->ipv6.ip6_null_entry)
 		return -ENOENT;
 
 	table = rt->rt6i_table;
@@ -1329,7 +1329,8 @@ struct ip6rd_flowi {
 	struct in6_addr gateway;
 };
 
-static struct rt6_info *__ip6_route_redirect(struct fib6_table *table,
+static struct rt6_info *__ip6_route_redirect(struct net *net,
+					     struct fib6_table *table,
 					     struct flowi *fl,
 					     int flags)
 {
@@ -1372,8 +1373,8 @@ restart:
 	}
 
 	if (!rt)
-		rt = ip6_null_entry;
-	BACKTRACK(&fl->fl6_src);
+		rt = net->ipv6.ip6_null_entry;
+	BACKTRACK(net, &fl->fl6_src);
 out:
 	dst_hold(&rt->u.dst);
 
@@ -1415,10 +1416,11 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 {
 	struct rt6_info *rt, *nrt = NULL;
 	struct netevent_redirect netevent;
+	struct net *net = neigh->dev->nd_net;
 
 	rt = ip6_route_redirect(dest, src, saddr, neigh->dev);
 
-	if (rt == ip6_null_entry) {
+	if (rt == net->ipv6.ip6_null_entry) {
 		if (net_ratelimit())
 			printk(KERN_DEBUG "rt6_redirect: source isn't a valid nexthop "
 			       "for redirect target\n");
@@ -1886,10 +1888,18 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 	return rt;
 }
 
+struct arg_dev_net {
+	struct net_device *dev;
+	struct net *net;
+};
+
 static int fib6_ifdown(struct rt6_info *rt, void *arg)
 {
-	if (((void*)rt->rt6i_dev == arg || arg == NULL) &&
-	    rt != ip6_null_entry) {
+	struct net_device *dev = ((struct arg_dev_net *)arg)->dev;
+	struct net *net = ((struct arg_dev_net *)arg)->net;
+
+	if (((void *)rt->rt6i_dev == dev || dev == NULL) &&
+	    rt != net->ipv6.ip6_null_entry) {
 		RT6_TRACE("deleted by ifdown %p\n", rt);
 		return -1;
 	}
@@ -1898,7 +1908,12 @@ static int fib6_ifdown(struct rt6_info *rt, void *arg)
 
 void rt6_ifdown(struct net *net, struct net_device *dev)
 {
-	fib6_clean_all(net, fib6_ifdown, 0, dev);
+	struct arg_dev_net adn = {
+		.dev = dev,
+		.net = net,
+	};
+
+	fib6_clean_all(net, fib6_ifdown, 0, &adn);
 }
 
 struct rt6_mtu_change_arg
@@ -2289,6 +2304,26 @@ errout:
 		rtnl_set_sk_err(net, RTNLGRP_IPV6_ROUTE, err);
 }
 
+static int ip6_route_dev_notify(struct notifier_block *this,
+				unsigned long event, void *data)
+{
+	struct net_device *dev = (struct net_device *)data;
+	struct net *net = dev->nd_net;
+
+	if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) {
+		net->ipv6.ip6_null_entry->u.dst.dev = dev;
+		net->ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(dev);
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+		net->ipv6.ip6_prohibit_entry->u.dst.dev = dev;
+		net->ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(dev);
+		net->ipv6.ip6_blk_hole_entry->u.dst.dev = dev;
+		net->ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(dev);
+#endif
+	}
+
+	return NOTIFY_OK;
+}
+
 /*
  *	/proc
  */
@@ -2535,11 +2570,47 @@ struct ctl_table *ipv6_route_sysctl_init(struct net *net)
 
 static int ip6_route_net_init(struct net *net)
 {
+	int ret = 0;
+
+	ret = -ENOMEM;
+	net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template,
+					   sizeof(*net->ipv6.ip6_null_entry),
+					   GFP_KERNEL);
+	if (!net->ipv6.ip6_null_entry)
+		goto out;
+	net->ipv6.ip6_null_entry->u.dst.path =
+		(struct dst_entry *)net->ipv6.ip6_null_entry;
+
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
+					       sizeof(*net->ipv6.ip6_prohibit_entry),
+					       GFP_KERNEL);
+	if (!net->ipv6.ip6_prohibit_entry) {
+		kfree(net->ipv6.ip6_null_entry);
+		goto out;
+	}
+	net->ipv6.ip6_prohibit_entry->u.dst.path =
+		(struct dst_entry *)net->ipv6.ip6_prohibit_entry;
+
+	net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
+					       sizeof(*net->ipv6.ip6_blk_hole_entry),
+					       GFP_KERNEL);
+	if (!net->ipv6.ip6_blk_hole_entry) {
+		kfree(net->ipv6.ip6_null_entry);
+		kfree(net->ipv6.ip6_prohibit_entry);
+		goto out;
+	}
+	net->ipv6.ip6_blk_hole_entry->u.dst.path =
+		(struct dst_entry *)net->ipv6.ip6_blk_hole_entry;
+#endif
+
 #ifdef CONFIG_PROC_FS
 	proc_net_fops_create(net, "ipv6_route", 0, &ipv6_route_proc_fops);
 	proc_net_fops_create(net, "rt6_stats", S_IRUGO, &rt6_stats_seq_fops);
 #endif
-	return 0;
+	ret = 0;
+out:
+	return ret;
 }
 
 static void ip6_route_net_exit(struct net *net)
@@ -2548,7 +2619,11 @@ static void ip6_route_net_exit(struct net *net)
 	proc_net_remove(net, "ipv6_route");
 	proc_net_remove(net, "rt6_stats");
 #endif
-	rt6_ifdown(net, NULL);
+	kfree(net->ipv6.ip6_null_entry);
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	kfree(net->ipv6.ip6_prohibit_entry);
+	kfree(net->ipv6.ip6_blk_hole_entry);
+#endif
 }
 
 static struct pernet_operations ip6_route_net_ops = {
@@ -2556,6 +2631,11 @@ static struct pernet_operations ip6_route_net_ops = {
 	.exit = ip6_route_net_exit,
 };
 
+static struct notifier_block ip6_route_dev_notifier = {
+	.notifier_call = ip6_route_dev_notify,
+	.priority = 0,
+};
+
 int __init ip6_route_init(void)
 {
 	int ret;
@@ -2568,30 +2648,24 @@ int __init ip6_route_init(void)
 
 	ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops.kmem_cachep;
 
-	ret = -ENOMEM;
-	ip6_null_entry = kmemdup(&ip6_null_entry_template,
-				 sizeof(*ip6_null_entry), GFP_KERNEL);
-	if (!ip6_null_entry)
+	ret = register_pernet_subsys(&ip6_route_net_ops);
+	if (ret)
 		goto out_kmem_cache;
-	ip6_null_entry->u.dst.path = (struct dst_entry *)ip6_null_entry;
-
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
-				     sizeof(*ip6_prohibit_entry), GFP_KERNEL);
-	if (!ip6_prohibit_entry)
-		goto out_ip6_null_entry;
-	ip6_prohibit_entry->u.dst.path = (struct dst_entry *)ip6_prohibit_entry;
-
-	ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
-				     sizeof(*ip6_blk_hole_entry), GFP_KERNEL);
-	if (!ip6_blk_hole_entry)
-		goto out_ip6_prohibit_entry;
-	ip6_blk_hole_entry->u.dst.path = (struct dst_entry *)ip6_blk_hole_entry;
-#endif
 
+	/* Registering of the loopback is done before this portion of code,
+	 * the loopback reference in rt6_info will not be taken, do it
+	 * manually for init_net */
+	init_net.ipv6.ip6_null_entry->u.dst.dev = init_net.loopback_dev;
+	init_net.ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
+  #ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	init_net.ipv6.ip6_prohibit_entry->u.dst.dev = init_net.loopback_dev;
+	init_net.ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
+	init_net.ipv6.ip6_blk_hole_entry->u.dst.dev = init_net.loopback_dev;
+	init_net.ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
+  #endif
 	ret = fib6_init();
 	if (ret)
-		goto out_ip6_blk_hole_entry;
+		goto out_register_subsys;
 
 	ret = xfrm6_init();
 	if (ret)
@@ -2607,9 +2681,10 @@ int __init ip6_route_init(void)
 	    __rtnl_register(PF_INET6, RTM_GETROUTE, inet6_rtm_getroute, NULL))
 		goto fib6_rules_init;
 
-	ret = register_pernet_subsys(&ip6_route_net_ops);
+	ret = register_netdevice_notifier(&ip6_route_dev_notifier);
 	if (ret)
 		goto fib6_rules_init;
+
 out:
 	return ret;
 
@@ -2619,14 +2694,8 @@ xfrm6_init:
 	xfrm6_fini();
 out_fib6_init:
 	fib6_gc_cleanup();
-out_ip6_blk_hole_entry:
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	kfree(ip6_blk_hole_entry);
-out_ip6_prohibit_entry:
-	kfree(ip6_prohibit_entry);
-out_ip6_null_entry:
-#endif
-	kfree(ip6_null_entry);
+out_register_subsys:
+	unregister_pernet_subsys(&ip6_route_net_ops);
 out_kmem_cache:
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
 	goto out;
@@ -2634,15 +2703,10 @@ out_kmem_cache:
 
 void ip6_route_cleanup(void)
 {
-	unregister_pernet_subsys(&ip6_route_net_ops);
+	unregister_netdevice_notifier(&ip6_route_dev_notifier);
 	fib6_rules_cleanup();
 	xfrm6_fini();
 	fib6_gc_cleanup();
+	unregister_pernet_subsys(&ip6_route_net_ops);
 	kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
-
-	kfree(ip6_null_entry);
-#ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	kfree(ip6_prohibit_entry);
-	kfree(ip6_blk_hole_entry);
-#endif
 }
-- 
cgit v1.2.3


From f2fc6a54585a1be6669613a31fbaba2ecbadcd36 Mon Sep 17 00:00:00 2001
From: Benjamin Thery <benjamin.thery@bull.net>
Date: Tue, 4 Mar 2008 13:49:23 -0800
Subject: [NETNS][IPV6] route6 - move ip6_dst_ops inside the network namespace

The ip6_dst_ops is moved inside the network namespace structure.  All
references to this structure are now relative to the initial network
namespace.

Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |  1 +
 net/ipv6/route.c         | 62 ++++++++++++++++++++++++------------------------
 2 files changed, 32 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 311a942f36e9..53b5a283a4b3 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -41,6 +41,7 @@ struct netns_ipv6 {
 	struct timer_list       *ip6_fib_timer;
 	struct hlist_head       *fib_table_hash;
 	struct fib6_table       *fib6_main_tbl;
+	struct dst_ops		*ip6_dst_ops;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	struct rt6_info         *ip6_prohibit_entry;
 	struct rt6_info         *ip6_blk_hole_entry;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index fa014d701c1a..d88b6ec3c5d1 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -113,8 +113,6 @@ static struct dst_ops ip6_dst_ops_template = {
 	.entries		=	ATOMIC_INIT(0),
 };
 
-static struct dst_ops *ip6_dst_ops;
-
 static void ip6_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu)
 {
 }
@@ -188,9 +186,9 @@ static struct rt6_info ip6_blk_hole_entry_template = {
 #endif
 
 /* allocate dst with ip6_dst_ops */
-static __inline__ struct rt6_info *ip6_dst_alloc(void)
+static inline struct rt6_info *ip6_dst_alloc(struct dst_ops *ops)
 {
-	return (struct rt6_info *)dst_alloc(ip6_dst_ops);
+	return (struct rt6_info *)dst_alloc(ops);
 }
 
 static void ip6_dst_destroy(struct dst_entry *dst)
@@ -925,7 +923,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 	if (unlikely(idev == NULL))
 		return NULL;
 
-	rt = ip6_dst_alloc();
+	rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
 	if (unlikely(rt == NULL)) {
 		in6_dev_put(idev);
 		goto out;
@@ -999,18 +997,18 @@ static int ip6_dst_gc(struct dst_ops *ops)
 	unsigned long now = jiffies;
 
 	if (time_after(last_gc + init_net.ipv6.sysctl.ip6_rt_gc_min_interval, now) &&
-	    atomic_read(&ip6_dst_ops->entries) <= init_net.ipv6.sysctl.ip6_rt_max_size)
+	    atomic_read(&init_net.ipv6.ip6_dst_ops->entries) <= init_net.ipv6.sysctl.ip6_rt_max_size)
 		goto out;
 
 	expire++;
 	fib6_run_gc(expire, &init_net);
 	last_gc = now;
-	if (atomic_read(&ip6_dst_ops->entries) < ip6_dst_ops->gc_thresh)
+	if (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) < init_net.ipv6.ip6_dst_ops->gc_thresh)
 		expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1;
 
 out:
 	expire -= expire>>init_net.ipv6.sysctl.ip6_rt_gc_elasticity;
-	return (atomic_read(&ip6_dst_ops->entries) > init_net.ipv6.sysctl.ip6_rt_max_size);
+	return (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) > init_net.ipv6.sysctl.ip6_rt_max_size);
 }
 
 /* Clean host part of a prefix. Not necessary in radix tree,
@@ -1084,7 +1082,7 @@ int ip6_route_add(struct fib6_config *cfg)
 		goto out;
 	}
 
-	rt = ip6_dst_alloc();
+	rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
 
 	if (rt == NULL) {
 		err = -ENOMEM;
@@ -1570,7 +1568,8 @@ out:
 
 static struct rt6_info * ip6_rt_copy(struct rt6_info *ort)
 {
-	struct rt6_info *rt = ip6_dst_alloc();
+	struct net *net = ort->rt6i_dev->nd_net;
+	struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
 
 	if (rt) {
 		rt->u.dst.input = ort->u.dst.input;
@@ -1849,7 +1848,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 				    int anycast)
 {
 	struct net *net = idev->dev->nd_net;
-	struct rt6_info *rt = ip6_dst_alloc();
+	struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
 
 	if (rt == NULL)
 		return ERR_PTR(-ENOMEM);
@@ -2407,7 +2406,7 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v)
 		   net->ipv6.rt6_stats->fib_rt_alloc,
 		   net->ipv6.rt6_stats->fib_rt_entries,
 		   net->ipv6.rt6_stats->fib_rt_cache,
-		   atomic_read(&ip6_dst_ops->entries),
+		   atomic_read(&net->ipv6.ip6_dst_ops->entries),
 		   net->ipv6.rt6_stats->fib_discarded_routes);
 
 	return 0;
@@ -2552,7 +2551,7 @@ struct ctl_table *ipv6_route_sysctl_init(struct net *net)
 
 	if (table) {
 		table[0].data = &net->ipv6.sysctl.flush_delay;
-		table[1].data = &ip6_dst_ops_template.gc_thresh;
+		table[1].data = &net->ipv6.ip6_dst_ops->gc_thresh;
 		table[2].data = &net->ipv6.sysctl.ip6_rt_max_size;
 		table[3].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval;
 		table[4].data = &net->ipv6.sysctl.ip6_rt_gc_timeout;
@@ -2571,14 +2570,21 @@ static int ip6_route_net_init(struct net *net)
 	int ret = 0;
 
 	ret = -ENOMEM;
+	net->ipv6.ip6_dst_ops = kmemdup(&ip6_dst_ops_template,
+					sizeof(*net->ipv6.ip6_dst_ops),
+					GFP_KERNEL);
+	if (!net->ipv6.ip6_dst_ops)
+		goto out;
+	net->ipv6.ip6_dst_ops->dst_net = net;
+
 	net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template,
 					   sizeof(*net->ipv6.ip6_null_entry),
 					   GFP_KERNEL);
 	if (!net->ipv6.ip6_null_entry)
-		goto out;
+		goto out_ip6_dst_ops;
 	net->ipv6.ip6_null_entry->u.dst.path =
 		(struct dst_entry *)net->ipv6.ip6_null_entry;
-	net->ipv6.ip6_null_entry->u.dst.ops = ip6_dst_ops;
+	net->ipv6.ip6_null_entry->u.dst.ops = net->ipv6.ip6_dst_ops;
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
@@ -2590,7 +2596,7 @@ static int ip6_route_net_init(struct net *net)
 	}
 	net->ipv6.ip6_prohibit_entry->u.dst.path =
 		(struct dst_entry *)net->ipv6.ip6_prohibit_entry;
-	net->ipv6.ip6_prohibit_entry->u.dst.ops = ip6_dst_ops;
+	net->ipv6.ip6_prohibit_entry->u.dst.ops = net->ipv6.ip6_dst_ops;
 
 	net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
 					       sizeof(*net->ipv6.ip6_blk_hole_entry),
@@ -2602,7 +2608,7 @@ static int ip6_route_net_init(struct net *net)
 	}
 	net->ipv6.ip6_blk_hole_entry->u.dst.path =
 		(struct dst_entry *)net->ipv6.ip6_blk_hole_entry;
-	net->ipv6.ip6_blk_hole_entry->u.dst.ops = ip6_dst_ops;
+	net->ipv6.ip6_blk_hole_entry->u.dst.ops = net->ipv6.ip6_dst_ops;
 #endif
 
 #ifdef CONFIG_PROC_FS
@@ -2612,6 +2618,10 @@ static int ip6_route_net_init(struct net *net)
 	ret = 0;
 out:
 	return ret;
+
+out_ip6_dst_ops:
+	kfree(net->ipv6.ip6_dst_ops);
+	goto out;
 }
 
 static void ip6_route_net_exit(struct net *net)
@@ -2625,6 +2635,7 @@ static void ip6_route_net_exit(struct net *net)
 	kfree(net->ipv6.ip6_prohibit_entry);
 	kfree(net->ipv6.ip6_blk_hole_entry);
 #endif
+	kfree(net->ipv6.ip6_dst_ops);
 }
 
 static struct pernet_operations ip6_route_net_ops = {
@@ -2641,20 +2652,12 @@ int __init ip6_route_init(void)
 {
 	int ret;
 
-	ip6_dst_ops = kmemdup(&ip6_dst_ops_template,
-			      sizeof(*ip6_dst_ops), GFP_KERNEL);
-	if (!ip6_dst_ops)
-		return -ENOMEM;
-
 	ret = -ENOMEM;
 	ip6_dst_ops_template.kmem_cachep =
 		kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0,
 				  SLAB_HWCACHE_ALIGN, NULL);
 	if (!ip6_dst_ops_template.kmem_cachep)
-		goto out_ip6_dst_ops;
-
-	ip6_dst_ops->kmem_cachep = ip6_dst_ops_template.kmem_cachep;
-	ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops_template.kmem_cachep;
+		goto out;;
 
 	ret = register_pernet_subsys(&ip6_route_net_ops);
 	if (ret)
@@ -2705,9 +2708,7 @@ out_fib6_init:
 out_register_subsys:
 	unregister_pernet_subsys(&ip6_route_net_ops);
 out_kmem_cache:
-	kmem_cache_destroy(ip6_dst_ops->kmem_cachep);
-out_ip6_dst_ops:
-	kfree(ip6_dst_ops);
+	kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
 	goto out;
 }
 
@@ -2718,6 +2719,5 @@ void ip6_route_cleanup(void)
 	xfrm6_fini();
 	fib6_gc_cleanup();
 	unregister_pernet_subsys(&ip6_route_net_ops);
-	kmem_cache_destroy(ip6_dst_ops->kmem_cachep);
-	kfree(ip6_dst_ops);
+	kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
 }
-- 
cgit v1.2.3


From 6891a346c387bd0a64afa50f4522f5fe8ba879d8 Mon Sep 17 00:00:00 2001
From: Benjamin Thery <benjamin.thery@bull.net>
Date: Tue, 4 Mar 2008 13:49:47 -0800
Subject: [NETNS][IPV6] route6 - make garbage collection work with multiple
 network namespaces

This patch makes the necessary changes to make IPv6 dst_entry garbage
collection work with multiple network namespaces.

In ip6_dst_gc(), static local variables are now declared
per-namespace.

Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |  2 ++
 net/ipv6/route.c         | 23 ++++++++++++-----------
 2 files changed, 14 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 53b5a283a4b3..90e6e24df858 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -42,6 +42,8 @@ struct netns_ipv6 {
 	struct hlist_head       *fib_table_hash;
 	struct fib6_table       *fib6_main_tbl;
 	struct dst_ops		*ip6_dst_ops;
+	unsigned int		 ip6_rt_gc_expire;
+	unsigned long		 ip6_rt_last_gc;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	struct rt6_info         *ip6_prohibit_entry;
 	struct rt6_info         *ip6_blk_hole_entry;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index d88b6ec3c5d1..10a6cc0aca6c 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -992,23 +992,22 @@ int icmp6_dst_gc(int *more)
 
 static int ip6_dst_gc(struct dst_ops *ops)
 {
-	static unsigned expire = 30*HZ;
-	static unsigned long last_gc;
+	struct net *net = ops->dst_net;
 	unsigned long now = jiffies;
 
-	if (time_after(last_gc + init_net.ipv6.sysctl.ip6_rt_gc_min_interval, now) &&
-	    atomic_read(&init_net.ipv6.ip6_dst_ops->entries) <= init_net.ipv6.sysctl.ip6_rt_max_size)
+	if (time_after(net->ipv6.ip6_rt_last_gc + net->ipv6.sysctl.ip6_rt_gc_min_interval, now) &&
+	    atomic_read(&net->ipv6.ip6_dst_ops->entries) <= net->ipv6.sysctl.ip6_rt_max_size)
 		goto out;
 
-	expire++;
-	fib6_run_gc(expire, &init_net);
-	last_gc = now;
-	if (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) < init_net.ipv6.ip6_dst_ops->gc_thresh)
-		expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1;
+	net->ipv6.ip6_rt_gc_expire++;
+	fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net);
+	net->ipv6.ip6_rt_last_gc = now;
+	if (atomic_read(&net->ipv6.ip6_dst_ops->entries) < net->ipv6.ip6_dst_ops->gc_thresh)
+		net->ipv6.ip6_rt_gc_expire = net->ipv6.sysctl.ip6_rt_gc_timeout>>1;
 
 out:
-	expire -= expire>>init_net.ipv6.sysctl.ip6_rt_gc_elasticity;
-	return (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) > init_net.ipv6.sysctl.ip6_rt_max_size);
+	net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>net->ipv6.sysctl.ip6_rt_gc_elasticity;
+	return (atomic_read(&net->ipv6.ip6_dst_ops->entries) > net->ipv6.sysctl.ip6_rt_max_size);
 }
 
 /* Clean host part of a prefix. Not necessary in radix tree,
@@ -2615,6 +2614,8 @@ static int ip6_route_net_init(struct net *net)
 	proc_net_fops_create(net, "ipv6_route", 0, &ipv6_route_proc_fops);
 	proc_net_fops_create(net, "rt6_stats", S_IRUGO, &rt6_stats_seq_fops);
 #endif
+	net->ipv6.ip6_rt_gc_expire = 30*HZ;
+
 	ret = 0;
 out:
 	return ret;
-- 
cgit v1.2.3


From af2849377e7b70afa1274e475be50286cd0ef6eb Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Wed, 5 Mar 2008 10:46:57 -0800
Subject: [NETNS][IPV6] addrconf - Pass the proper network namespace parameters
 to addrconf

This patch propagates the network namespace pointer to the address
configuration routines which need it, which means adding a new
parameter to these functions, and make them use it instead of using
the initial network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h |  9 ++++---
 net/ipv6/addrconf.c    | 66 ++++++++++++++++++++++++++++++++++++--------------
 net/ipv6/af_inet6.c    |  6 ++---
 3 files changed, 57 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 89e3c53c8886..232da20e7171 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -55,9 +55,12 @@ struct prefix_info {
 extern int			addrconf_init(void);
 extern void			addrconf_cleanup(void);
 
-extern int			addrconf_add_ifaddr(void __user *arg);
-extern int			addrconf_del_ifaddr(void __user *arg);
-extern int			addrconf_set_dstaddr(void __user *arg);
+extern int			addrconf_add_ifaddr(struct net *net,
+						    void __user *arg);
+extern int			addrconf_del_ifaddr(struct net *net,
+						    void __user *arg);
+extern int			addrconf_set_dstaddr(struct net *net,
+						     void __user *arg);
 
 extern int			ipv6_chk_addr(struct net *net,
 					      struct in6_addr *addr,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 17b06d220e95..127021bdd180 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1866,7 +1866,7 @@ ok:
  *	Special case for SIT interfaces where we create a new "virtual"
  *	device.
  */
-int addrconf_set_dstaddr(void __user *arg)
+int addrconf_set_dstaddr(struct net *net, void __user *arg)
 {
 	struct in6_ifreq ireq;
 	struct net_device *dev;
@@ -1878,7 +1878,7 @@ int addrconf_set_dstaddr(void __user *arg)
 	if (copy_from_user(&ireq, arg, sizeof(struct in6_ifreq)))
 		goto err_exit;
 
-	dev = __dev_get_by_index(&init_net, ireq.ifr6_ifindex);
+	dev = __dev_get_by_index(net, ireq.ifr6_ifindex);
 
 	err = -ENODEV;
 	if (dev == NULL)
@@ -1909,7 +1909,8 @@ int addrconf_set_dstaddr(void __user *arg)
 
 		if (err == 0) {
 			err = -ENOBUFS;
-			if ((dev = __dev_get_by_name(&init_net, p.name)) == NULL)
+			dev = __dev_get_by_name(net, p.name);
+			if (!dev)
 				goto err_exit;
 			err = dev_open(dev);
 		}
@@ -1924,8 +1925,9 @@ err_exit:
 /*
  *	Manual configuration of address on an interface
  */
-static int inet6_addr_add(int ifindex, struct in6_addr *pfx, int plen,
-			  __u8 ifa_flags, __u32 prefered_lft, __u32 valid_lft)
+static int inet6_addr_add(struct net *net, int ifindex, struct in6_addr *pfx,
+			  int plen, __u8 ifa_flags, __u32 prefered_lft,
+			  __u32 valid_lft)
 {
 	struct inet6_ifaddr *ifp;
 	struct inet6_dev *idev;
@@ -1939,7 +1941,8 @@ static int inet6_addr_add(int ifindex, struct in6_addr *pfx, int plen,
 	if (!valid_lft || prefered_lft > valid_lft)
 		return -EINVAL;
 
-	if ((dev = __dev_get_by_index(&init_net, ifindex)) == NULL)
+	dev = __dev_get_by_index(net, ifindex);
+	if (!dev)
 		return -ENODEV;
 
 	if ((idev = addrconf_add_dev(dev)) == NULL)
@@ -1984,13 +1987,15 @@ static int inet6_addr_add(int ifindex, struct in6_addr *pfx, int plen,
 	return PTR_ERR(ifp);
 }
 
-static int inet6_addr_del(int ifindex, struct in6_addr *pfx, int plen)
+static int inet6_addr_del(struct net *net, int ifindex, struct in6_addr *pfx,
+			  int plen)
 {
 	struct inet6_ifaddr *ifp;
 	struct inet6_dev *idev;
 	struct net_device *dev;
 
-	if ((dev = __dev_get_by_index(&init_net, ifindex)) == NULL)
+	dev = __dev_get_by_index(net, ifindex);
+	if (!dev)
 		return -ENODEV;
 
 	if ((idev = __in6_dev_get(dev)) == NULL)
@@ -2018,7 +2023,7 @@ static int inet6_addr_del(int ifindex, struct in6_addr *pfx, int plen)
 }
 
 
-int addrconf_add_ifaddr(void __user *arg)
+int addrconf_add_ifaddr(struct net *net, void __user *arg)
 {
 	struct in6_ifreq ireq;
 	int err;
@@ -2030,13 +2035,14 @@ int addrconf_add_ifaddr(void __user *arg)
 		return -EFAULT;
 
 	rtnl_lock();
-	err = inet6_addr_add(ireq.ifr6_ifindex, &ireq.ifr6_addr, ireq.ifr6_prefixlen,
-			     IFA_F_PERMANENT, INFINITY_LIFE_TIME, INFINITY_LIFE_TIME);
+	err = inet6_addr_add(net, ireq.ifr6_ifindex, &ireq.ifr6_addr,
+			     ireq.ifr6_prefixlen, IFA_F_PERMANENT,
+			     INFINITY_LIFE_TIME, INFINITY_LIFE_TIME);
 	rtnl_unlock();
 	return err;
 }
 
-int addrconf_del_ifaddr(void __user *arg)
+int addrconf_del_ifaddr(struct net *net, void __user *arg)
 {
 	struct in6_ifreq ireq;
 	int err;
@@ -2048,7 +2054,8 @@ int addrconf_del_ifaddr(void __user *arg)
 		return -EFAULT;
 
 	rtnl_lock();
-	err = inet6_addr_del(ireq.ifr6_ifindex, &ireq.ifr6_addr, ireq.ifr6_prefixlen);
+	err = inet6_addr_del(net, ireq.ifr6_ifindex, &ireq.ifr6_addr,
+			     ireq.ifr6_prefixlen);
 	rtnl_unlock();
 	return err;
 }
@@ -3061,7 +3068,7 @@ inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	if (pfx == NULL)
 		return -EINVAL;
 
-	return inet6_addr_del(ifm->ifa_index, pfx, ifm->ifa_prefixlen);
+	return inet6_addr_del(net, ifm->ifa_index, pfx, ifm->ifa_prefixlen);
 }
 
 static int inet6_addr_modify(struct inet6_ifaddr *ifp, u8 ifa_flags,
@@ -3137,7 +3144,7 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 		valid_lft = INFINITY_LIFE_TIME;
 	}
 
-	dev =  __dev_get_by_index(&init_net, ifm->ifa_index);
+	dev =  __dev_get_by_index(net, ifm->ifa_index);
 	if (dev == NULL)
 		return -ENODEV;
 
@@ -3150,8 +3157,9 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 		 * It would be best to check for !NLM_F_CREATE here but
 		 * userspace alreay relies on not having to provide this.
 		 */
-		return inet6_addr_add(ifm->ifa_index, pfx, ifm->ifa_prefixlen,
-				      ifa_flags, preferred_lft, valid_lft);
+		return inet6_addr_add(net, ifm->ifa_index, pfx,
+				      ifm->ifa_prefixlen, ifa_flags,
+				      preferred_lft, valid_lft);
 	}
 
 	if (nlh->nlmsg_flags & NLM_F_EXCL ||
@@ -4260,6 +4268,22 @@ int unregister_inet6addr_notifier(struct notifier_block *nb)
 
 EXPORT_SYMBOL(unregister_inet6addr_notifier);
 
+
+static int addrconf_net_init(struct net *net)
+{
+	return 0;
+}
+
+static void addrconf_net_exit(struct net *net)
+{
+	;
+}
+
+static struct pernet_operations addrconf_net_ops = {
+	.init = addrconf_net_init,
+	.exit = addrconf_net_exit,
+};
+
 /*
  *	Init / cleanup code
  */
@@ -4301,6 +4325,10 @@ int __init addrconf_init(void)
 	if (err)
 		goto errlo;
 
+	err = register_pernet_device(&addrconf_net_ops);
+	if (err)
+		return err;
+
 	register_netdevice_notifier(&ipv6_dev_notf);
 
 	addrconf_verify(0);
@@ -4334,6 +4362,7 @@ void addrconf_cleanup(void)
 	int i;
 
 	unregister_netdevice_notifier(&ipv6_dev_notf);
+	unregister_pernet_device(&addrconf_net_ops);
 
 	unregister_pernet_subsys(&addrconf_ops);
 
@@ -4370,6 +4399,7 @@ void addrconf_cleanup(void)
 	write_unlock_bh(&addrconf_hash_lock);
 
 	del_timer(&addr_chk_timer);
-
 	rtnl_unlock();
+
+	unregister_pernet_subsys(&addrconf_net_ops);
 }
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 430bbd2139c1..afe9276d0420 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -454,11 +454,11 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		return(ipv6_route_ioctl(net, cmd, (void __user *)arg));
 
 	case SIOCSIFADDR:
-		return addrconf_add_ifaddr((void __user *) arg);
+		return addrconf_add_ifaddr(net, (void __user *) arg);
 	case SIOCDIFADDR:
-		return addrconf_del_ifaddr((void __user *) arg);
+		return addrconf_del_ifaddr(net, (void __user *) arg);
 	case SIOCSIFDSTADDR:
-		return addrconf_set_dstaddr((void __user *) arg);
+		return addrconf_set_dstaddr(net, (void __user *) arg);
 	default:
 		if (!sk->sk_prot->ioctl)
 			return -ENOIOCTLCMD;
-- 
cgit v1.2.3


From 4591db4f37618f37a9f1f25d291c3c7a43a15a21 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Wed, 5 Mar 2008 10:48:10 -0800
Subject: [NETNS][IPV6] route6 - add netns parameter to ip6_route_output

Add an netns parameter to ip6_route_output. That will allow to access
to the right routing table for outgoing traffic.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h          | 3 ++-
 net/ipv6/icmp.c                  | 2 +-
 net/ipv6/ip6_output.c            | 4 ++--
 net/ipv6/ip6_tunnel.c            | 2 +-
 net/ipv6/ndisc.c                 | 2 +-
 net/ipv6/netfilter.c             | 4 ++--
 net/ipv6/netfilter/ip6t_REJECT.c | 2 +-
 net/ipv6/route.c                 | 7 ++++---
 net/ipv6/sit.c                   | 2 +-
 net/ipv6/xfrm6_policy.c          | 2 +-
 net/sctp/ipv6.c                  | 2 +-
 11 files changed, 17 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index e0caed25bfba..0e2895c8b270 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -43,7 +43,8 @@ extern struct rt6_info	*ip6_blk_hole_entry;
 
 extern void			ip6_route_input(struct sk_buff *skb);
 
-extern struct dst_entry *	ip6_route_output(struct sock *sk,
+extern struct dst_entry *	ip6_route_output(struct net *net,
+						 struct sock *sk,
 						 struct flowi *fl);
 
 extern int			ip6_route_init(void);
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index cff74127ea32..d4c4b50420d8 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -178,7 +178,7 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type,
 	 * XXX: perhaps the expire for routing entries cloned by
 	 * this lookup should be more aggressive (not longer than timeout).
 	 */
-	dst = ip6_route_output(sk, fl);
+	dst = ip6_route_output(&init_net, sk, fl);
 	if (dst->error) {
 		IP6_INC_STATS(ip6_dst_idev(dst),
 			      IPSTATS_MIB_OUTNOROUTES);
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index ff3971173e1e..161afd1142d0 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -913,7 +913,7 @@ static int ip6_dst_lookup_tail(struct sock *sk,
 	int err;
 
 	if (*dst == NULL)
-		*dst = ip6_route_output(sk, fl);
+		*dst = ip6_route_output(&init_net, sk, fl);
 
 	if ((err = (*dst)->error))
 		goto out_err_release;
@@ -954,7 +954,7 @@ static int ip6_dst_lookup_tail(struct sock *sk,
 				dst_release(*dst);
 				memcpy(&fl_gw, fl, sizeof(struct flowi));
 				memset(&fl_gw.fl6_dst, 0, sizeof(struct in6_addr));
-				*dst = ip6_route_output(sk, &fl_gw);
+				*dst = ip6_route_output(&init_net, sk, &fl_gw);
 				if ((err = (*dst)->error))
 					goto out_err_release;
 			}
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 4e1981660b3c..1e1ad1ed87e6 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -847,7 +847,7 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 	if ((dst = ip6_tnl_dst_check(t)) != NULL)
 		dst_hold(dst);
 	else {
-		dst = ip6_route_output(NULL, fl);
+		dst = ip6_route_output(&init_net, NULL, fl);
 
 		if (dst->error || xfrm_lookup(&dst, fl, NULL, 0) < 0)
 			goto tx_err_link_failure;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index f1c95125100d..b5b4fd173e98 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1427,7 +1427,7 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
 	icmpv6_flow_init(ndisc_socket->sk, &fl, NDISC_REDIRECT,
 			 &saddr_buf, &ipv6_hdr(skb)->saddr, dev->ifindex);
 
-	dst = ip6_route_output(NULL, &fl);
+	dst = ip6_route_output(&init_net, NULL, &fl);
 	if (dst == NULL)
 		return;
 
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 2e06724dc348..aed51bcc66b4 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -23,7 +23,7 @@ int ip6_route_me_harder(struct sk_buff *skb)
 		    .saddr = iph->saddr, } },
 	};
 
-	dst = ip6_route_output(skb->sk, &fl);
+	dst = ip6_route_output(&init_net, skb->sk, &fl);
 
 #ifdef CONFIG_XFRM
 	if (!(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) &&
@@ -86,7 +86,7 @@ static int nf_ip6_reroute(struct sk_buff *skb,
 
 static int nf_ip6_route(struct dst_entry **dst, struct flowi *fl)
 {
-	*dst = ip6_route_output(NULL, fl);
+	*dst = ip6_route_output(&init_net, NULL, fl);
 	return (*dst)->error;
 }
 
diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c
index b23baa635fe0..831708aeab37 100644
--- a/net/ipv6/netfilter/ip6t_REJECT.c
+++ b/net/ipv6/netfilter/ip6t_REJECT.c
@@ -93,7 +93,7 @@ static void send_reset(struct sk_buff *oldskb)
 	fl.fl_ip_sport = otcph.dest;
 	fl.fl_ip_dport = otcph.source;
 	security_skb_classify_flow(oldskb, &fl);
-	dst = ip6_route_output(NULL, &fl);
+	dst = ip6_route_output(&init_net, NULL, &fl);
 	if (dst == NULL)
 		return;
 	if (dst->error || xfrm_lookup(&dst, &fl, NULL, 0))
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index ec1fedb339cc..f31d7dc11e72 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -772,7 +772,8 @@ static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table
 	return ip6_pol_route(net, table, fl->oif, fl, flags);
 }
 
-struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
+struct dst_entry * ip6_route_output(struct net *net, struct sock *sk,
+				    struct flowi *fl)
 {
 	int flags = 0;
 
@@ -782,7 +783,7 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
 	if (!ipv6_addr_any(&fl->fl6_src))
 		flags |= RT6_LOOKUP_F_HAS_SADDR;
 
-	return fib6_rule_lookup(&init_net, fl, flags, ip6_pol_route_output);
+	return fib6_rule_lookup(net, fl, flags, ip6_pol_route_output);
 }
 
 EXPORT_SYMBOL(ip6_route_output);
@@ -2260,7 +2261,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 	skb_reset_mac_header(skb);
 	skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr));
 
-	rt = (struct rt6_info*) ip6_route_output(NULL, &fl);
+	rt = (struct rt6_info*) ip6_route_output(&init_net, NULL, &fl);
 	skb->dst = &rt->u.dst;
 
 	err = rt6_fill_node(skb, rt, &fl.fl6_dst, &fl.fl6_src, iif,
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 68720aa63f96..1b8196c8d145 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -393,7 +393,7 @@ isatap_srcok(struct sk_buff *skb, struct iphdr *iph, struct net_device *dev)
 	fl.oif = dev->ifindex;
 	security_skb_classify_flow(skb, &fl);
 
-	dst = ip6_route_output(NULL, &fl);
+	dst = ip6_route_output(&init_net, NULL, &fl);
 	if (!dst->error && (dst->dev == dev) && (neigh = dst->neighbour)) {
 
 		addr6 = (struct in6_addr*)&neigh->primary_key;
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 6ef56303e69e..e96dafdc7032 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -38,7 +38,7 @@ static struct dst_entry *xfrm6_dst_lookup(int tos, xfrm_address_t *saddr,
 	if (saddr)
 		memcpy(&fl.fl6_src, saddr, sizeof(fl.fl6_src));
 
-	dst = ip6_route_output(NULL, &fl);
+	dst = ip6_route_output(&init_net, NULL, &fl);
 
 	err = dst->error;
 	if (dst->error) {
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index 3e4878800b36..4862835b0c39 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -257,7 +257,7 @@ static struct dst_entry *sctp_v6_get_dst(struct sctp_association *asoc,
 			NIP6(fl.fl6_src));
 	}
 
-	dst = ip6_route_output(NULL, &fl);
+	dst = ip6_route_output(&init_net, NULL, &fl);
 	if (!dst->error) {
 		struct rt6_info *rt;
 		rt = (struct rt6_info *)dst;
-- 
cgit v1.2.3


From ee6b967301b4aa5d4a4b61e2f682f086266db9fb Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Wed, 5 Mar 2008 18:30:47 -0800
Subject: [IPV4]: Add 'rtable' field in struct sk_buff to alias 'dst' and avoid
 casts

(Anonymous) unions can help us to avoid ugly casts.

A common cast it the (struct rtable *)skb->dst one.

Defining an union like  :
union {
     struct dst_entry *dst;
     struct rtable *rtable;
};
permits to use skb->rtable in place.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h                  |  5 ++++-
 include/net/inet_sock.h                 |  2 +-
 net/bridge/br_netfilter.c               | 14 +++++++-------
 net/dccp/ipv4.c                         |  7 +++----
 net/ipv4/arp.c                          |  4 ++--
 net/ipv4/icmp.c                         | 10 +++++-----
 net/ipv4/igmp.c                         |  2 +-
 net/ipv4/ip_forward.c                   |  2 +-
 net/ipv4/ip_gre.c                       |  4 ++--
 net/ipv4/ip_input.c                     |  2 +-
 net/ipv4/ip_options.c                   | 16 ++++++++--------
 net/ipv4/ip_output.c                    | 10 +++++-----
 net/ipv4/ip_sockglue.c                  |  2 +-
 net/ipv4/ipip.c                         |  2 +-
 net/ipv4/ipmr.c                         |  6 +++---
 net/ipv4/netfilter/ipt_MASQUERADE.c     |  2 +-
 net/ipv4/netfilter/nf_nat_helper.c      |  4 ++--
 net/ipv4/route.c                        | 22 +++++++++++-----------
 net/ipv4/tcp_ipv4.c                     |  5 ++---
 net/ipv4/udp_ipv4.c                     |  2 +-
 net/netfilter/nf_conntrack_netbios_ns.c |  2 +-
 net/sched/em_meta.c                     |  4 ++--
 net/sctp/protocol.c                     |  8 ++++----
 23 files changed, 69 insertions(+), 68 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index bbd8d0027e2f..7beb239d2ee0 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -256,7 +256,10 @@ struct sk_buff {
 	ktime_t			tstamp;
 	struct net_device	*dev;
 
-	struct  dst_entry	*dst;
+	union {
+		struct  dst_entry	*dst;
+		struct  rtable		*rtable;
+	};
 	struct	sec_path	*sp;
 
 	/*
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 89cd011edb99..8660cb0fa0dd 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -195,7 +195,7 @@ static inline int inet_sk_ehashfn(const struct sock *sk)
 
 static inline int inet_iif(const struct sk_buff *skb)
 {
-	return ((struct rtable *)skb->dst)->rt_iif;
+	return skb->rtable->rt_iif;
 }
 
 #endif	/* _INET_SOCK_H */
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 1c0efd8ad9f3..0278a069c6f1 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -223,8 +223,8 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb)
 	}
 	nf_bridge->mask ^= BRNF_NF_BRIDGE_PREROUTING;
 
-	skb->dst = (struct dst_entry *)&__fake_rtable;
-	dst_hold(skb->dst);
+	skb->rtable = &__fake_rtable;
+	dst_hold(&__fake_rtable.u.dst);
 
 	skb->dev = nf_bridge->physindev;
 	nf_bridge_push_encap_header(skb);
@@ -388,8 +388,8 @@ bridged_dnat:
 			skb->pkt_type = PACKET_HOST;
 		}
 	} else {
-		skb->dst = (struct dst_entry *)&__fake_rtable;
-		dst_hold(skb->dst);
+		skb->rtable = &__fake_rtable;
+		dst_hold(&__fake_rtable.u.dst);
 	}
 
 	skb->dev = nf_bridge->physindev;
@@ -608,9 +608,9 @@ static unsigned int br_nf_local_in(unsigned int hook, struct sk_buff *skb,
 				   const struct net_device *out,
 				   int (*okfn)(struct sk_buff *))
 {
-	if (skb->dst == (struct dst_entry *)&__fake_rtable) {
-		dst_release(skb->dst);
-		skb->dst = NULL;
+	if (skb->rtable == &__fake_rtable) {
+		dst_release(&__fake_rtable.u.dst);
+		skb->rtable = NULL;
 	}
 
 	return NF_ACCEPT;
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 514a40b7fc7f..17ad69e90e48 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -450,7 +450,7 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk,
 					   struct sk_buff *skb)
 {
 	struct rtable *rt;
-	struct flowi fl = { .oif = ((struct rtable *)skb->dst)->rt_iif,
+	struct flowi fl = { .oif = skb->rtable->rt_iif,
 			    .nl_u = { .ip4_u =
 				      { .daddr = ip_hdr(skb)->saddr,
 					.saddr = ip_hdr(skb)->daddr,
@@ -511,7 +511,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 	if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
 		return;
 
-	if (((struct rtable *)rxskb->dst)->rt_type != RTN_LOCAL)
+	if (rxskb->rtable->rt_type != RTN_LOCAL)
 		return;
 
 	dst = dccp_v4_route_skb(dccp_v4_ctl_socket->sk, rxskb);
@@ -563,8 +563,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb);
 
 	/* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */
-	if (((struct rtable *)skb->dst)->rt_flags &
-	    (RTCF_BROADCAST | RTCF_MULTICAST))
+	if (skb->rtable->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
 		return 0;	/* discard, don't send a reset here */
 
 	if (dccp_bad_service_code(sk, service)) {
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 69e80bd9774a..efe01df8fc0e 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -475,7 +475,7 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb)
 		return 1;
 	}
 
-	paddr = ((struct rtable*)skb->dst)->rt_gateway;
+	paddr = skb->rtable->rt_gateway;
 
 	if (arp_set_predefined(inet_addr_type(&init_net, paddr), haddr, paddr, dev))
 		return 0;
@@ -814,7 +814,7 @@ static int arp_process(struct sk_buff *skb)
 	if (arp->ar_op == htons(ARPOP_REQUEST) &&
 	    ip_route_input(skb, tip, sip, 0, dev) == 0) {
 
-		rt = (struct rtable*)skb->dst;
+		rt = skb->rtable;
 		addr_type = rt->rt_type;
 
 		if (addr_type == RTN_LOCAL) {
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index cee77d606fbe..ff9a8e643fcc 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -381,7 +381,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
 static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 {
 	struct ipcm_cookie ipc;
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct net *net = rt->u.dst.dev->nd_net;
 	struct sock *sk = icmp_sk(net);
 	struct inet_sock *inet = inet_sk(sk);
@@ -438,7 +438,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 	struct iphdr *iph;
 	int room;
 	struct icmp_bxm icmp_param;
-	struct rtable *rt = (struct rtable *)skb_in->dst;
+	struct rtable *rt = skb_in->rtable;
 	struct ipcm_cookie ipc;
 	__be32 saddr;
 	u8  tos;
@@ -616,7 +616,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 					     RT_TOS(tos), rt2->u.dst.dev);
 
 			dst_release(&rt2->u.dst);
-			rt2 = (struct rtable *)skb_in->dst;
+			rt2 = skb_in->rtable;
 			skb_in->dst = odst;
 		}
 
@@ -943,7 +943,7 @@ static void icmp_address(struct sk_buff *skb)
 
 static void icmp_address_reply(struct sk_buff *skb)
 {
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct net_device *dev = skb->dev;
 	struct in_device *in_dev;
 	struct in_ifaddr *ifa;
@@ -988,7 +988,7 @@ static void icmp_discard(struct sk_buff *skb)
 int icmp_rcv(struct sk_buff *skb)
 {
 	struct icmphdr *icmph;
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 
 	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) {
 		int nh;
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index d3f34a772f3b..6a4ee8da6994 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -948,7 +948,7 @@ int igmp_rcv(struct sk_buff *skb)
 	case IGMPV2_HOST_MEMBERSHIP_REPORT:
 	case IGMPV3_HOST_MEMBERSHIP_REPORT:
 		/* Is it our report looped back? */
-		if (((struct rtable*)skb->dst)->fl.iif == 0)
+		if (skb->rtable->fl.iif == 0)
 			break;
 		/* don't rely on MC router hearing unicast reports */
 		if (skb->pkt_type == PACKET_MULTICAST ||
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 0b3b328d82db..9d6d3befd854 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -80,7 +80,7 @@ int ip_forward(struct sk_buff *skb)
 	if (!xfrm4_route_forward(skb))
 		goto drop;
 
-	rt = (struct rtable*)skb->dst;
+	rt = skb->rtable;
 
 	if (opt->is_strictroute && rt->rt_dst != rt->rt_gateway)
 		goto sr_failed;
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index e7821ba7a9a0..f9ee84420cb3 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -619,7 +619,7 @@ static int ipgre_rcv(struct sk_buff *skb)
 #ifdef CONFIG_NET_IPGRE_BROADCAST
 		if (ipv4_is_multicast(iph->daddr)) {
 			/* Looped back packet, drop it! */
-			if (((struct rtable*)skb->dst)->fl.iif == 0)
+			if (skb->rtable->fl.iif == 0)
 				goto drop;
 			tunnel->stat.multicast++;
 			skb->pkt_type = PACKET_BROADCAST;
@@ -699,7 +699,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 
 		if (skb->protocol == htons(ETH_P_IP)) {
-			rt = (struct rtable*)skb->dst;
+			rt = skb->rtable;
 			if ((dst = rt->rt_gateway) == 0)
 				goto tx_error_icmp;
 		}
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 65631391d479..d36e310b314d 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -351,7 +351,7 @@ static int ip_rcv_finish(struct sk_buff *skb)
 	if (iph->ihl > 5 && ip_rcv_options(skb))
 		goto drop;
 
-	rt = (struct rtable*)skb->dst;
+	rt = skb->rtable;
 	if (rt->rt_type == RTN_MULTICAST)
 		IP_INC_STATS_BH(IPSTATS_MIB_INMCASTPKTS);
 	else if (rt->rt_type == RTN_BROADCAST)
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index baaedd9689a0..df93a9c2efda 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -107,7 +107,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb)
 	sptr = skb_network_header(skb);
 	dptr = dopt->__data;
 
-	daddr = ((struct rtable*)skb->dst)->rt_spec_dst;
+	daddr = skb->rtable->rt_spec_dst;
 
 	if (sopt->rr) {
 		optlen  = sptr[sopt->rr+1];
@@ -258,7 +258,7 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb)
 	unsigned char * optptr;
 	int optlen;
 	unsigned char * pp_ptr = NULL;
-	struct rtable *rt = skb ? (struct rtable*)skb->dst : NULL;
+	struct rtable *rt = skb ? skb->rtable : NULL;
 
 	if (!opt) {
 		opt = &(IPCB(skb)->opt);
@@ -558,7 +558,7 @@ void ip_forward_options(struct sk_buff *skb)
 {
 	struct   ip_options * opt	= &(IPCB(skb)->opt);
 	unsigned char * optptr;
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	unsigned char *raw = skb_network_header(skb);
 
 	if (opt->rr_needaddr) {
@@ -606,7 +606,7 @@ int ip_options_rcv_srr(struct sk_buff *skb)
 	__be32 nexthop;
 	struct iphdr *iph = ip_hdr(skb);
 	unsigned char *optptr = skb_network_header(skb) + opt->srr;
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct rtable *rt2;
 	int err;
 
@@ -631,13 +631,13 @@ int ip_options_rcv_srr(struct sk_buff *skb)
 		}
 		memcpy(&nexthop, &optptr[srrptr-1], 4);
 
-		rt = (struct rtable*)skb->dst;
-		skb->dst = NULL;
+		rt = skb->rtable;
+		skb->rtable = NULL;
 		err = ip_route_input(skb, nexthop, iph->saddr, iph->tos, skb->dev);
-		rt2 = (struct rtable*)skb->dst;
+		rt2 = skb->rtable;
 		if (err || (rt2->rt_type != RTN_UNICAST && rt2->rt_type != RTN_LOCAL)) {
 			ip_rt_put(rt2);
-			skb->dst = &rt->u.dst;
+			skb->rtable = rt;
 			return -EINVAL;
 		}
 		ip_rt_put(rt);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 341779e685d9..dc494ea594a7 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -142,7 +142,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk,
 			  __be32 saddr, __be32 daddr, struct ip_options *opt)
 {
 	struct inet_sock *inet = inet_sk(sk);
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct iphdr *iph;
 
 	/* Build the IP header. */
@@ -240,7 +240,7 @@ static int ip_finish_output(struct sk_buff *skb)
 int ip_mc_output(struct sk_buff *skb)
 {
 	struct sock *sk = skb->sk;
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct net_device *dev = rt->u.dst.dev;
 
 	/*
@@ -321,7 +321,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok)
 	/* Skip all of this if the packet is already routed,
 	 * f.e. by something like SCTP.
 	 */
-	rt = (struct rtable *) skb->dst;
+	rt = skb->rtable;
 	if (rt != NULL)
 		goto packet_routed;
 
@@ -441,7 +441,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*))
 	unsigned int mtu, hlen, left, len, ll_rs, pad;
 	int offset;
 	__be16 not_last_frag;
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	int err = 0;
 
 	dev = rt->u.dst.dev;
@@ -1357,7 +1357,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 	} replyopts;
 	struct ipcm_cookie ipc;
 	__be32 daddr;
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 
 	if (ip_options_echo(&replyopts.opt, skb))
 		return;
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index de0572c88859..e7c9e4e72327 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -57,7 +57,7 @@
 static void ip_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb)
 {
 	struct in_pktinfo info;
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 
 	info.ipi_addr.s_addr = ip_hdr(skb)->daddr;
 	if (rt) {
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index dbaed69de06a..894bce96284a 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -528,7 +528,7 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (!dst) {
 		/* NBMA tunnel */
-		if ((rt = (struct rtable*)skb->dst) == NULL) {
+		if ((rt = skb->rtable) == NULL) {
 			tunnel->stat.tx_fifo_errors++;
 			goto tx_error;
 		}
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index a94f52c207a7..7d63d74ef62a 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1283,7 +1283,7 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local
 	if (vif_table[vif].dev != skb->dev) {
 		int true_vifi;
 
-		if (((struct rtable*)skb->dst)->fl.iif == 0) {
+		if (skb->rtable->fl.iif == 0) {
 			/* It is our own packet, looped back.
 			   Very complicated situation...
 
@@ -1357,7 +1357,7 @@ dont_forward:
 int ip_mr_input(struct sk_buff *skb)
 {
 	struct mfc_cache *cache;
-	int local = ((struct rtable*)skb->dst)->rt_flags&RTCF_LOCAL;
+	int local = skb->rtable->rt_flags&RTCF_LOCAL;
 
 	/* Packet is looped back after forward, it should not be
 	   forwarded second time, but still can be delivered locally.
@@ -1594,7 +1594,7 @@ int ipmr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait)
 {
 	int err;
 	struct mfc_cache *cache;
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 
 	read_lock(&mrt_lock);
 	cache = ipmr_cache_find(rt->rt_src, rt->rt_dst);
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index 313b3fcf387e..c6817b18366a 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -77,7 +77,7 @@ masquerade_tg(struct sk_buff *skb, const struct net_device *in,
 		return NF_ACCEPT;
 
 	mr = targinfo;
-	rt = (struct rtable *)skb->dst;
+	rt = skb->rtable;
 	newsrc = inet_select_addr(out, rt->rt_gateway, RT_SCOPE_UNIVERSE);
 	if (!newsrc) {
 		printk("MASQUERADE: %s ate my IP address\n", out->name);
diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c
index ca57f47bbd25..2fca727aa8ba 100644
--- a/net/ipv4/netfilter/nf_nat_helper.c
+++ b/net/ipv4/netfilter/nf_nat_helper.c
@@ -139,7 +139,7 @@ nf_nat_mangle_tcp_packet(struct sk_buff *skb,
 			 const char *rep_buffer,
 			 unsigned int rep_len)
 {
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct iphdr *iph;
 	struct tcphdr *tcph;
 	int oldlen, datalen;
@@ -217,7 +217,7 @@ nf_nat_mangle_udp_packet(struct sk_buff *skb,
 			 const char *rep_buffer,
 			 unsigned int rep_len)
 {
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct iphdr *iph;
 	struct udphdr *udph;
 	int datalen, oldlen;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 8c3e165f0034..1051326c36b2 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1289,7 +1289,7 @@ reject_redirect:
 
 static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst)
 {
-	struct rtable *rt = (struct rtable*)dst;
+	struct rtable *rt = (struct rtable *)dst;
 	struct dst_entry *ret = dst;
 
 	if (rt) {
@@ -1330,7 +1330,7 @@ static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst)
 
 void ip_rt_send_redirect(struct sk_buff *skb)
 {
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct in_device *in_dev = in_dev_get(rt->u.dst.dev);
 
 	if (!in_dev)
@@ -1379,7 +1379,7 @@ out:
 
 static int ip_error(struct sk_buff *skb)
 {
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	unsigned long now;
 	int code;
 
@@ -1548,7 +1548,7 @@ static void ipv4_link_failure(struct sk_buff *skb)
 
 	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_HOST_UNREACH, 0);
 
-	rt = (struct rtable *) skb->dst;
+	rt = skb->rtable;
 	if (rt)
 		dst_set_expires(&rt->u.dst, 0);
 }
@@ -1708,7 +1708,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 
 	in_dev_put(in_dev);
 	hash = rt_hash(daddr, saddr, dev->ifindex);
-	return rt_intern_hash(hash, rth, (struct rtable**) &skb->dst);
+	return rt_intern_hash(hash, rth, &skb->rtable);
 
 e_nobufs:
 	in_dev_put(in_dev);
@@ -1869,7 +1869,7 @@ static inline int ip_mkroute_input(struct sk_buff *skb,
 
 	/* put it into the cache */
 	hash = rt_hash(daddr, saddr, fl->iif);
-	return rt_intern_hash(hash, rth, (struct rtable**)&skb->dst);
+	return rt_intern_hash(hash, rth, &skb->rtable);
 }
 
 /*
@@ -2025,7 +2025,7 @@ local_input:
 	}
 	rth->rt_type	= res.type;
 	hash = rt_hash(daddr, saddr, fl.iif);
-	err = rt_intern_hash(hash, rth, (struct rtable**)&skb->dst);
+	err = rt_intern_hash(hash, rth, &skb->rtable);
 	goto done;
 
 no_route:
@@ -2091,7 +2091,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(in_hit);
 			rcu_read_unlock();
-			skb->dst = (struct dst_entry*)rth;
+			skb->rtable = rth;
 			return 0;
 		}
 		RT_CACHE_STAT_INC(in_hlist_search);
@@ -2598,7 +2598,7 @@ int ip_route_output_key(struct net *net, struct rtable **rp, struct flowi *flp)
 static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
 			int nowait, unsigned int flags)
 {
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct rtmsg *r;
 	struct nlmsghdr *nlh;
 	long expires;
@@ -2742,7 +2742,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 		err = ip_route_input(skb, dst, src, rtm->rtm_tos, dev);
 		local_bh_enable();
 
-		rt = (struct rtable*) skb->dst;
+		rt = skb->rtable;
 		if (err == 0 && rt->u.dst.error)
 			err = -rt->u.dst.error;
 	} else {
@@ -2762,7 +2762,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 	if (err)
 		goto errout_free;
 
-	skb->dst = &rt->u.dst;
+	skb->rtable = rt;
 	if (rtm->rtm_flags & RTM_F_NOTIFY)
 		rt->rt_flags |= RTCF_NOTIFY;
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3873c4dbeaeb..a79e324638eb 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -552,7 +552,7 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb)
 	if (th->rst)
 		return;
 
-	if (((struct rtable *)skb->dst)->rt_type != RTN_LOCAL)
+	if (skb->rtable->rt_type != RTN_LOCAL)
 		return;
 
 	/* Swap the send and the receive. */
@@ -1262,8 +1262,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 #endif
 
 	/* Never answer to SYNs send to broadcast or multicast */
-	if (((struct rtable *)skb->dst)->rt_flags &
-	    (RTCF_BROADCAST | RTCF_MULTICAST))
+	if (skb->rtable->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
 		goto drop;
 
 	/* TW buckets are converted to open requests without
diff --git a/net/ipv4/udp_ipv4.c b/net/ipv4/udp_ipv4.c
index 40978de7fb51..fd14c2c50ed4 100644
--- a/net/ipv4/udp_ipv4.c
+++ b/net/ipv4/udp_ipv4.c
@@ -893,7 +893,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	struct sock *sk;
 	struct udphdr *uh = udp_hdr(skb);
 	unsigned short ulen;
-	struct rtable *rt = (struct rtable*)skb->dst;
+	struct rtable *rt = skb->rtable;
 	__be32 saddr = ip_hdr(skb)->saddr;
 	__be32 daddr = ip_hdr(skb)->daddr;
 
diff --git a/net/netfilter/nf_conntrack_netbios_ns.c b/net/netfilter/nf_conntrack_netbios_ns.c
index 9810d81e2a06..60dedaded84e 100644
--- a/net/netfilter/nf_conntrack_netbios_ns.c
+++ b/net/netfilter/nf_conntrack_netbios_ns.c
@@ -47,7 +47,7 @@ static int help(struct sk_buff *skb, unsigned int protoff,
 {
 	struct nf_conntrack_expect *exp;
 	struct iphdr *iph = ip_hdr(skb);
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct in_device *in_dev;
 	__be32 mask = 0;
 
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index 3da4129b89d1..72cf86e3c090 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -256,10 +256,10 @@ META_COLLECTOR(int_rtclassid)
 
 META_COLLECTOR(int_rtiif)
 {
-	if (unlikely(skb->dst == NULL))
+	if (unlikely(skb->rtable == NULL))
 		*err = -1;
 	else
-		dst->value = ((struct rtable*) skb->dst)->fl.iif;
+		dst->value = skb->rtable->fl.iif;
 }
 
 /**************************************************************************
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index 8d9d929f6cea..1afef08f6c1d 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -363,7 +363,7 @@ static int sctp_v4_addr_valid(union sctp_addr *addr,
 		return 0;
 
 	/* Is this a broadcast address? */
-	if (skb && ((struct rtable *)skb->dst)->rt_flags & RTCF_BROADCAST)
+	if (skb && skb->rtable->rt_flags & RTCF_BROADCAST)
 		return 0;
 
 	return 1;
@@ -539,7 +539,7 @@ static void sctp_v4_get_saddr(struct sctp_association *asoc,
 /* What interface did this skb arrive on? */
 static int sctp_v4_skb_iif(const struct sk_buff *skb)
 {
-	return ((struct rtable *)skb->dst)->rt_iif;
+	return skb->rtable->rt_iif;
 }
 
 /* Was this packet marked by Explicit Congestion Notification? */
@@ -828,8 +828,8 @@ static inline int sctp_v4_xmit(struct sk_buff *skb,
 	SCTP_DEBUG_PRINTK("%s: skb:%p, len:%d, "
 			  "src:%u.%u.%u.%u, dst:%u.%u.%u.%u\n",
 			  __FUNCTION__, skb, skb->len,
-			  NIPQUAD(((struct rtable *)skb->dst)->rt_src),
-			  NIPQUAD(((struct rtable *)skb->dst)->rt_dst));
+			  NIPQUAD(skb->rtable->rt_src),
+			  NIPQUAD(skb->rtable->rt_dst));
 
 	SCTP_INC_STATS(SCTP_MIB_OUTSCTPPACKS);
 	return ip_queue_xmit(skb, ipfragok);
-- 
cgit v1.2.3


From f59d43899e279c77924a7ada4bec8c70e5aeca06 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 5 Mar 2008 20:58:10 -0800
Subject: [IPV6]: Fix powerpc allmodconfig build warnings.

Introduced by changeset 95e41e93e18d8e1e272ce23d96bae4f17ce11d42
("[IPV6]: Make ndisc_flow_init() common for later use.")

Reported by Stephen Rothwell.

In file included from net/ipv6/netfilter/ip6_tables.c:21:
include/linux/icmpv6.h:192: warning: 'struct in6_addr' declared inside parameter list
include/linux/icmpv6.h:192: warning: its scope is only this definition or declaration, which is probably not what you want

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/icmpv6.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h
index e4d4300d768f..03067443198a 100644
--- a/include/linux/icmpv6.h
+++ b/include/linux/icmpv6.h
@@ -184,6 +184,7 @@ extern void				icmpv6_param_prob(struct sk_buff *skb,
 							  int code, int pos);
 
 struct flowi;
+struct in6_addr;
 extern void				icmpv6_flow_init(struct sock *sk,
 							 struct flowi *fl,
 							 u8 type,
-- 
cgit v1.2.3


From 37c5798968d0ce4d479f114f1d5785551b57bfa5 Mon Sep 17 00:00:00 2001
From: Luis Carlos Cobo <luisca@cozybit.com>
Date: Sat, 23 Feb 2008 15:17:04 +0100
Subject: wireless: various definitions for mesh networking

Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

(limited to 'include')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index f577c8f1c66d..f27d11ab418b 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -97,6 +97,7 @@
 #define IEEE80211_MAX_FRAME_LEN		2352
 
 #define IEEE80211_MAX_SSID_LEN		32
+#define IEEE80211_MAX_MESH_ID_LEN	32
 
 struct ieee80211_hdr {
 	__le16 frame_control;
@@ -109,6 +110,16 @@ struct ieee80211_hdr {
 } __attribute__ ((packed));
 
 
+struct ieee80211s_hdr {
+	u8 flags;
+	u8 ttl;
+	u8 seqnum[3];
+	u8 eaddr1[6];
+	u8 eaddr2[6];
+	u8 eaddr3[6];
+} __attribute__ ((packed));
+
+
 struct ieee80211_mgmt {
 	__le16 frame_control;
 	__le16 duration;
@@ -206,6 +217,23 @@ struct ieee80211_mgmt {
 					__le16 params;
 					__le16 reason_code;
 				} __attribute__((packed)) delba;
+				struct{
+					u8 action_code;
+					/* capab_info for open and confirm,
+					 * reason for close
+					 */
+					__le16 aux;
+					/* Followed in plink_confirm by status
+					 * code, AID and supported rates,
+					 * and directly by supported rates in
+					 * plink_open and plink_close
+					 */
+					u8 variable[0];
+				} __attribute__((packed)) plink_action;
+				struct{
+					u8 action_code;
+					u8 variable[0];
+				} __attribute__((packed)) mesh_action;
 			} u;
 		} __attribute__ ((packed)) action;
 	} u;
@@ -437,6 +465,13 @@ enum ieee80211_eid {
 	WLAN_EID_TS_DELAY = 43,
 	WLAN_EID_TCLAS_PROCESSING = 44,
 	WLAN_EID_QOS_CAPA = 46,
+	/* 802.11s */
+	WLAN_EID_MESH_CONFIG = 36,      /* Pending IEEE 802.11 ANA approval */
+	WLAN_EID_MESH_ID = 37,          /* Pending IEEE 802.11 ANA approval */
+	WLAN_EID_PEER_LINK = 40,	/* Pending IEEE 802.11 ANA approval */
+	WLAN_EID_PREQ = 53,		/* Pending IEEE 802.11 ANA approval */
+	WLAN_EID_PREP = 54,		/* Pending IEEE 802.11 ANA approval */
+	WLAN_EID_PERR = 55,		/* Pending IEEE 802.11 ANA approval */
 	/* 802.11h */
 	WLAN_EID_PWR_CONSTRAINT = 32,
 	WLAN_EID_PWR_CAPABILITY = 33,
-- 
cgit v1.2.3


From cc0672a1066829be7e1b0128a13e36a2d0a15479 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sat, 23 Feb 2008 15:17:05 +0100
Subject: WEXT: add mesh interface type

This introduces a new WEXT type IW_MODE_MESH for mesh networks,
used for scan results.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/wireless.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 3160dfed73ca..2864b1699ecc 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -455,6 +455,7 @@
 #define IW_MODE_REPEAT	4	/* Wireless Repeater (forwarder) */
 #define IW_MODE_SECOND	5	/* Secondary master/repeater (backup) */
 #define IW_MODE_MONITOR	6	/* Passive monitor (listen only) */
+#define IW_MODE_MESH	7	/* Mesh (IEEE 802.11s) network */
 
 /* Statistics flags (bitmask in updated) */
 #define IW_QUAL_QUAL_UPDATED	0x01	/* Value was updated since last read */
-- 
cgit v1.2.3


From 2ec600d672e74488f8d1acf67a0a2baed222564c Mon Sep 17 00:00:00 2001
From: Luis Carlos Cobo <luisca@cozybit.com>
Date: Sat, 23 Feb 2008 15:17:06 +0100
Subject: nl80211/cfg80211: support for mesh, sta dumping

Added support for mesh id and mesh path operation as well as
station structure dumping.

Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 119 +++++++++++--
 include/net/cfg80211.h  | 139 +++++++++++++--
 net/mac80211/cfg.c      |  20 ++-
 net/wireless/nl80211.c  | 438 +++++++++++++++++++++++++++++++++++++++++++++---
 4 files changed, 650 insertions(+), 66 deletions(-)

(limited to 'include')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index a9f0b93324a2..ea6517e58b04 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -78,6 +78,18 @@
  *	or, if no MAC address given, all stations, on the interface identified
  *	by %NL80211_ATTR_IFINDEX.
  *
+ * @NL80211_CMD_GET_MPATH: Get mesh path attributes for mesh path to
+ * 	destination %NL80211_ATTR_MAC on the interface identified by
+ * 	%NL80211_ATTR_IFINDEX.
+ * @NL80211_CMD_SET_MPATH:  Set mesh path attributes for mesh path to
+ * 	destination %NL80211_ATTR_MAC on the interface identified by
+ * 	%NL80211_ATTR_IFINDEX.
+ * @NL80211_CMD_NEW_PATH: Add a mesh path with given attributes to the
+ *	the interface identified by %NL80211_ATTR_IFINDEX.
+ * @NL80211_CMD_DEL_PATH: Remove a mesh path identified by %NL80211_ATTR_MAC
+ *	or, if no MAC address given, all mesh paths, on the interface identified
+ *	by %NL80211_ATTR_IFINDEX.
+ *
  * @NL80211_CMD_MAX: highest used command number
  * @__NL80211_CMD_AFTER_LAST: internal use
  */
@@ -112,6 +124,11 @@ enum nl80211_commands {
 
 	/* add commands here */
 
+	NL80211_CMD_GET_MPATH,
+	NL80211_CMD_SET_MPATH,
+	NL80211_CMD_NEW_MPATH,
+	NL80211_CMD_DEL_MPATH,
+
 	/* used to define NL80211_CMD_MAX below */
 	__NL80211_CMD_AFTER_LAST,
 	NL80211_CMD_MAX = __NL80211_CMD_AFTER_LAST - 1
@@ -157,13 +174,21 @@ enum nl80211_commands {
  *	restriction (at most %NL80211_MAX_SUPP_RATES).
  * @NL80211_ATTR_STA_VLAN: interface index of VLAN interface to move station
  *	to, or the AP interface the station was originally added to to.
- * @NL80211_ATTR_STA_STATS: statistics for a station, part of station info
+ * @NL80211_ATTR_STA_INFO: information about a station, part of station info
  *	given for %NL80211_CMD_GET_STATION, nested attribute containing
- *	info as possible, see &enum nl80211_sta_stats.
+ *	info as possible, see &enum nl80211_sta_info.
  *
  * @NL80211_ATTR_WIPHY_BANDS: Information about an operating bands,
  *	consisting of a nested array.
  *
+ * @NL80211_ATTR_MESH_ID: mesh id (1-32 bytes).
+ * @NL80211_ATTR_PLINK_ACTION: action to perform on the mesh peer link.
+ * @NL80211_ATTR_MPATH_NEXT_HOP: MAC address of the next hop for a mesh path.
+ * @NL80211_ATTR_MPATH_INFO: information about a mesh_path, part of mesh path
+ * 	info given for %NL80211_CMD_GET_MPATH, nested attribute described at
+ *	&enum nl80211_mpath_info.
+ *
+ *
  * @NL80211_ATTR_MNTR_FLAGS: flags, nested element with NLA_FLAG attributes of
  *      &enum nl80211_mntr_flags.
  *
@@ -199,7 +224,7 @@ enum nl80211_attrs {
 	NL80211_ATTR_STA_LISTEN_INTERVAL,
 	NL80211_ATTR_STA_SUPPORTED_RATES,
 	NL80211_ATTR_STA_VLAN,
-	NL80211_ATTR_STA_STATS,
+	NL80211_ATTR_STA_INFO,
 
 	NL80211_ATTR_WIPHY_BANDS,
 
@@ -207,6 +232,11 @@ enum nl80211_attrs {
 
 	/* add attributes here, update the policy in nl80211.c */
 
+	NL80211_ATTR_MESH_ID,
+	NL80211_ATTR_STA_PLINK_ACTION,
+	NL80211_ATTR_MPATH_NEXT_HOP,
+	NL80211_ATTR_MPATH_INFO,
+
 	__NL80211_ATTR_AFTER_LAST,
 	NL80211_ATTR_MAX = __NL80211_ATTR_AFTER_LAST - 1
 };
@@ -223,6 +253,7 @@ enum nl80211_attrs {
  * @NL80211_IFTYPE_AP_VLAN: VLAN interface for access points
  * @NL80211_IFTYPE_WDS: wireless distribution interface
  * @NL80211_IFTYPE_MONITOR: monitor interface receiving all frames
+ * @NL80211_IFTYPE_MESH_POINT: mesh point
  * @NL80211_IFTYPE_MAX: highest interface type number currently defined
  * @__NL80211_IFTYPE_AFTER_LAST: internal use
  *
@@ -238,6 +269,7 @@ enum nl80211_iftype {
 	NL80211_IFTYPE_AP_VLAN,
 	NL80211_IFTYPE_WDS,
 	NL80211_IFTYPE_MONITOR,
+	NL80211_IFTYPE_MESH_POINT,
 
 	/* keep last */
 	__NL80211_IFTYPE_AFTER_LAST,
@@ -267,27 +299,78 @@ enum nl80211_sta_flags {
 };
 
 /**
- * enum nl80211_sta_stats - station statistics
+ * enum nl80211_sta_info - station information
  *
- * These attribute types are used with %NL80211_ATTR_STA_STATS
+ * These attribute types are used with %NL80211_ATTR_STA_INFO
  * when getting information about a station.
  *
- * @__NL80211_STA_STAT_INVALID: attribute number 0 is reserved
- * @NL80211_STA_STAT_INACTIVE_TIME: time since last activity (u32, msecs)
- * @NL80211_STA_STAT_RX_BYTES: total received bytes (u32, from this station)
- * @NL80211_STA_STAT_TX_BYTES: total transmitted bytes (u32, to this station)
- * @__NL80211_STA_STAT_AFTER_LAST: internal
- * @NL80211_STA_STAT_MAX: highest possible station stats attribute
+ * @__NL80211_STA_INFO_INVALID: attribute number 0 is reserved
+ * @NL80211_STA_INFO_INACTIVE_TIME: time since last activity (u32, msecs)
+ * @NL80211_STA_INFO_RX_BYTES: total received bytes (u32, from this station)
+ * @NL80211_STA_INFO_TX_BYTES: total transmitted bytes (u32, to this station)
+ * @__NL80211_STA_INFO_AFTER_LAST: internal
+ * @NL80211_STA_INFO_MAX: highest possible station info attribute
+ */
+enum nl80211_sta_info {
+	__NL80211_STA_INFO_INVALID,
+	NL80211_STA_INFO_INACTIVE_TIME,
+	NL80211_STA_INFO_RX_BYTES,
+	NL80211_STA_INFO_TX_BYTES,
+	NL80211_STA_INFO_LLID,
+	NL80211_STA_INFO_PLID,
+	NL80211_STA_INFO_PLINK_STATE,
+
+	/* keep last */
+	__NL80211_STA_INFO_AFTER_LAST,
+	NL80211_STA_INFO_MAX = __NL80211_STA_INFO_AFTER_LAST - 1
+};
+
+/**
+ * enum nl80211_mpath_flags - nl80211 mesh path flags
+ *
+ * @NL80211_MPATH_FLAG_ACTIVE: the mesh path is active
+ * @NL80211_MPATH_FLAG_RESOLVING: the mesh path discovery process is running
+ * @NL80211_MPATH_FLAG_DSN_VALID: the mesh path contains a valid DSN
+ * @NL80211_MPATH_FLAG_FIXED: the mesh path has been manually set
+ * @NL80211_MPATH_FLAG_RESOLVED: the mesh path discovery process succeeded
+ */
+enum nl80211_mpath_flags {
+	NL80211_MPATH_FLAG_ACTIVE =	1<<0,
+	NL80211_MPATH_FLAG_RESOLVING =	1<<1,
+	NL80211_MPATH_FLAG_DSN_VALID =	1<<2,
+	NL80211_MPATH_FLAG_FIXED =	1<<3,
+	NL80211_MPATH_FLAG_RESOLVED =	1<<4,
+};
+
+/**
+ * enum nl80211_mpath_info - mesh path information
+ *
+ * These attribute types are used with %NL80211_ATTR_MPATH_INFO when getting
+ * information about a mesh path.
+ *
+ * @__NL80211_MPATH_INFO_INVALID: attribute number 0 is reserved
+ * @NL80211_ATTR_MPATH_FRAME_QLEN: number of queued frames for this destination
+ * @NL80211_ATTR_MPATH_DSN: destination sequence number
+ * @NL80211_ATTR_MPATH_METRIC: metric (cost) of this mesh path
+ * @NL80211_ATTR_MPATH_EXPTIME: expiration time for the path, in msec from now
+ * @NL80211_ATTR_MPATH_FLAGS: mesh path flags, enumerated in
+ * 	&enum nl80211_mpath_flags;
+ * @NL80211_ATTR_MPATH_DISCOVERY_TIMEOUT: total path discovery timeout, in msec
+ * @NL80211_ATTR_MPATH_DISCOVERY_RETRIES: mesh path discovery retries
  */
-enum nl80211_sta_stats {
-	__NL80211_STA_STAT_INVALID,
-	NL80211_STA_STAT_INACTIVE_TIME,
-	NL80211_STA_STAT_RX_BYTES,
-	NL80211_STA_STAT_TX_BYTES,
+enum nl80211_mpath_info {
+	__NL80211_MPATH_INFO_INVALID,
+	NL80211_MPATH_INFO_FRAME_QLEN,
+	NL80211_MPATH_INFO_DSN,
+	NL80211_MPATH_INFO_METRIC,
+	NL80211_MPATH_INFO_EXPTIME,
+	NL80211_MPATH_INFO_FLAGS,
+	NL80211_MPATH_INFO_DISCOVERY_TIMEOUT,
+	NL80211_MPATH_INFO_DISCOVERY_RETRIES,
 
 	/* keep last */
-	__NL80211_STA_STAT_AFTER_LAST,
-	NL80211_STA_STAT_MAX = __NL80211_STA_STAT_AFTER_LAST - 1
+	__NL80211_MPATH_INFO_AFTER_LAST,
+	NL80211_MPATH_INFO_MAX = __NL80211_MPATH_INFO_AFTER_LAST - 1
 };
 
 /**
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index ab4caf63954f..e00750836ba5 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -12,6 +12,16 @@
  * Copyright 2006, 2007	Johannes Berg <johannes@sipsolutions.net>
  */
 
+/**
+ * struct vif_params - describes virtual interface parameters
+ * @mesh_id: mesh ID to use
+ * @mesh_id_len: length of the mesh ID
+ */
+struct vif_params {
+       u8 *mesh_id;
+       int mesh_id_len;
+};
+
 /* Radiotap header iteration
  *   implemented in net/wireless/radiotap.c
  *   docs in Documentation/networking/radiotap-headers.txt
@@ -108,6 +118,19 @@ enum station_flags {
 	STATION_FLAG_WME		= 1<<NL80211_STA_FLAG_WME,
 };
 
+/**
+ * enum plink_action - actions to perform in mesh peers
+ *
+ * @PLINK_ACTION_INVALID: action 0 is reserved
+ * @PLINK_ACTION_OPEN: start mesh peer link establishment
+ * @PLINK_ACTION_BLOCL: block traffic from this mesh peer
+ */
+enum plink_actions {
+	PLINK_ACTION_INVALID,
+	PLINK_ACTION_OPEN,
+	PLINK_ACTION_BLOCK,
+};
+
 /**
  * struct station_parameters - station parameters
  *
@@ -128,39 +151,52 @@ struct station_parameters {
 	int listen_interval;
 	u16 aid;
 	u8 supported_rates_len;
+	u8 plink_action;
 };
 
 /**
- * enum station_stats_flags - station statistics flags
+ * enum station_info_flags - station information flags
  *
- * Used by the driver to indicate which info in &struct station_stats
- * it has filled in during get_station().
+ * Used by the driver to indicate which info in &struct station_info
+ * it has filled in during get_station() or dump_station().
  *
- * @STATION_STAT_INACTIVE_TIME: @inactive_time filled
- * @STATION_STAT_RX_BYTES: @rx_bytes filled
- * @STATION_STAT_TX_BYTES: @tx_bytes filled
+ * @STATION_INFO_INACTIVE_TIME: @inactive_time filled
+ * @STATION_INFO_RX_BYTES: @rx_bytes filled
+ * @STATION_INFO_TX_BYTES: @tx_bytes filled
+ * @STATION_INFO_LLID: @llid filled
+ * @STATION_INFO_PLID: @plid filled
+ * @STATION_INFO_PLINK_STATE: @plink_state filled
  */
-enum station_stats_flags {
-	STATION_STAT_INACTIVE_TIME	= 1<<0,
-	STATION_STAT_RX_BYTES		= 1<<1,
-	STATION_STAT_TX_BYTES		= 1<<2,
+enum station_info_flags {
+	STATION_INFO_INACTIVE_TIME	= 1<<0,
+	STATION_INFO_RX_BYTES		= 1<<1,
+	STATION_INFO_TX_BYTES		= 1<<2,
+	STATION_INFO_LLID		= 1<<3,
+	STATION_INFO_PLID		= 1<<4,
+	STATION_INFO_PLINK_STATE	= 1<<5,
 };
 
 /**
- * struct station_stats - station statistics
+ * struct station_info - station information
  *
- * Station information filled by driver for get_station().
+ * Station information filled by driver for get_station() and dump_station.
  *
- * @filled: bitflag of flags from &enum station_stats_flags
+ * @filled: bitflag of flags from &enum station_info_flags
  * @inactive_time: time since last station activity (tx/rx) in milliseconds
  * @rx_bytes: bytes received from this station
  * @tx_bytes: bytes transmitted to this station
+ * @llid: mesh local link id
+ * @plid: mesh peer link id
+ * @plink_state: mesh peer link state
  */
-struct station_stats {
+struct station_info {
 	u32 filled;
 	u32 inactive_time;
 	u32 rx_bytes;
 	u32 tx_bytes;
+	u16 llid;
+	u16 plid;
+	u8 plink_state;
 };
 
 /**
@@ -183,6 +219,56 @@ enum monitor_flags {
 	MONITOR_FLAG_COOK_FRAMES	= 1<<NL80211_MNTR_FLAG_COOK_FRAMES,
 };
 
+/**
+ * enum mpath_info_flags -  mesh path information flags
+ *
+ * Used by the driver to indicate which info in &struct mpath_info it has filled
+ * in during get_station() or dump_station().
+ *
+ * MPATH_INFO_FRAME_QLEN: @frame_qlen filled
+ * MPATH_INFO_DSN: @dsn filled
+ * MPATH_INFO_METRIC: @metric filled
+ * MPATH_INFO_EXPTIME: @exptime filled
+ * MPATH_INFO_DISCOVERY_TIMEOUT: @discovery_timeout filled
+ * MPATH_INFO_DISCOVERY_RETRIES: @discovery_retries filled
+ * MPATH_INFO_FLAGS: @flags filled
+ */
+enum mpath_info_flags {
+	MPATH_INFO_FRAME_QLEN		= BIT(0),
+	MPATH_INFO_DSN			= BIT(1),
+	MPATH_INFO_METRIC		= BIT(2),
+	MPATH_INFO_EXPTIME		= BIT(3),
+	MPATH_INFO_DISCOVERY_TIMEOUT	= BIT(4),
+	MPATH_INFO_DISCOVERY_RETRIES	= BIT(5),
+	MPATH_INFO_FLAGS		= BIT(6),
+};
+
+/**
+ * struct mpath_info - mesh path information
+ *
+ * Mesh path information filled by driver for get_mpath() and dump_mpath().
+ *
+ * @filled: bitfield of flags from &enum mpath_info_flags
+ * @frame_qlen: number of queued frames for this destination
+ * @dsn: destination sequence number
+ * @metric: metric (cost) of this mesh path
+ * @exptime: expiration time for the mesh path from now, in msecs
+ * @flags: mesh path flags
+ * @discovery_timeout: total mesh path discovery timeout, in msecs
+ * @discovery_retries: mesh path discovery retries
+ */
+struct mpath_info {
+	u32 filled;
+	u32 frame_qlen;
+	u32 dsn;
+	u32 metric;
+	u32 exptime;
+	u32 discovery_timeout;
+	u8 discovery_retries;
+	u8 flags;
+};
+
+
 /* from net/wireless.h */
 struct wiphy;
 
@@ -230,13 +316,17 @@ struct wiphy;
  * @del_station: Remove a station; @mac may be NULL to remove all stations.
  *
  * @change_station: Modify a given station.
+ *
+ * @set_mesh_cfg: set mesh parameters (by now, just mesh id)
  */
 struct cfg80211_ops {
 	int	(*add_virtual_intf)(struct wiphy *wiphy, char *name,
-				    enum nl80211_iftype type, u32 *flags);
+				    enum nl80211_iftype type, u32 *flags,
+				    struct vif_params *params);
 	int	(*del_virtual_intf)(struct wiphy *wiphy, int ifindex);
 	int	(*change_virtual_intf)(struct wiphy *wiphy, int ifindex,
-				       enum nl80211_iftype type, u32 *flags);
+				       enum nl80211_iftype type, u32 *flags,
+				       struct vif_params *params);
 
 	int	(*add_key)(struct wiphy *wiphy, struct net_device *netdev,
 			   u8 key_index, u8 *mac_addr,
@@ -264,7 +354,22 @@ struct cfg80211_ops {
 	int	(*change_station)(struct wiphy *wiphy, struct net_device *dev,
 				  u8 *mac, struct station_parameters *params);
 	int	(*get_station)(struct wiphy *wiphy, struct net_device *dev,
-			       u8 *mac, struct station_stats *stats);
+			       u8 *mac, struct station_info *sinfo);
+	int	(*dump_station)(struct wiphy *wiphy, struct net_device *dev,
+			       int idx, u8 *mac, struct station_info *sinfo);
+
+	int	(*add_mpath)(struct wiphy *wiphy, struct net_device *dev,
+			       u8 *dst, u8 *next_hop);
+	int	(*del_mpath)(struct wiphy *wiphy, struct net_device *dev,
+			       u8 *dst);
+	int	(*change_mpath)(struct wiphy *wiphy, struct net_device *dev,
+				  u8 *dst, u8 *next_hop);
+	int	(*get_mpath)(struct wiphy *wiphy, struct net_device *dev,
+			       u8 *dst, u8 *next_hop,
+			       struct mpath_info *pinfo);
+	int	(*dump_mpath)(struct wiphy *wiphy, struct net_device *dev,
+			       int idx, u8 *dst, u8 *next_hop,
+			       struct mpath_info *pinfo);
 };
 
 #endif /* __NET_CFG80211_H */
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index e7535ffc8e1c..006da6a2e71b 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -34,7 +34,8 @@ nl80211_type_to_mac80211_type(enum nl80211_iftype type)
 }
 
 static int ieee80211_add_iface(struct wiphy *wiphy, char *name,
-			       enum nl80211_iftype type, u32 *flags)
+			       enum nl80211_iftype type, u32 *flags,
+			       struct vif_params *params)
 {
 	struct ieee80211_local *local = wiphy_priv(wiphy);
 	enum ieee80211_if_types itype;
@@ -78,7 +79,8 @@ static int ieee80211_del_iface(struct wiphy *wiphy, int ifindex)
 }
 
 static int ieee80211_change_iface(struct wiphy *wiphy, int ifindex,
-				  enum nl80211_iftype type, u32 *flags)
+				  enum nl80211_iftype type, u32 *flags,
+				  struct vif_params *params)
 {
 	struct ieee80211_local *local = wiphy_priv(wiphy);
 	struct net_device *dev;
@@ -296,7 +298,7 @@ static int ieee80211_config_default_key(struct wiphy *wiphy,
 }
 
 static int ieee80211_get_station(struct wiphy *wiphy, struct net_device *dev,
-				 u8 *mac, struct station_stats *stats)
+				 u8 *mac, struct station_info *sinfo)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	struct sta_info *sta;
@@ -307,13 +309,13 @@ static int ieee80211_get_station(struct wiphy *wiphy, struct net_device *dev,
 
 	/* XXX: verify sta->dev == dev */
 
-	stats->filled = STATION_STAT_INACTIVE_TIME |
-			STATION_STAT_RX_BYTES |
-			STATION_STAT_TX_BYTES;
+	sinfo->filled = STATION_INFO_INACTIVE_TIME |
+			STATION_INFO_RX_BYTES |
+			STATION_INFO_TX_BYTES;
 
-	stats->inactive_time = jiffies_to_msecs(jiffies - sta->last_rx);
-	stats->rx_bytes = sta->rx_bytes;
-	stats->tx_bytes = sta->tx_bytes;
+	sinfo->inactive_time = jiffies_to_msecs(jiffies - sta->last_rx);
+	sinfo->rx_bytes = sta->rx_bytes;
+	sinfo->tx_bytes = sta->tx_bytes;
 
 	sta_info_put(sta);
 
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 5b3474798b8d..64a7460af734 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -81,8 +81,12 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = {
 	[NL80211_ATTR_STA_LISTEN_INTERVAL] = { .type = NLA_U16 },
 	[NL80211_ATTR_STA_SUPPORTED_RATES] = { .type = NLA_BINARY,
 					       .len = NL80211_MAX_SUPP_RATES },
+	[NL80211_ATTR_STA_PLINK_ACTION] = { .type = NLA_U8 },
 	[NL80211_ATTR_STA_VLAN] = { .type = NLA_U32 },
 	[NL80211_ATTR_MNTR_FLAGS] = { .type = NLA_NESTED },
+	[NL80211_ATTR_MESH_ID] = { .type = NLA_BINARY,
+				.len = IEEE80211_MAX_MESH_ID_LEN },
+	[NL80211_ATTR_MPATH_NEXT_HOP] = { .type = NLA_U32 },
 };
 
 /* message building helper */
@@ -369,11 +373,14 @@ static int parse_monitor_flags(struct nlattr *nla, u32 *mntrflags)
 static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *drv;
+	struct vif_params params;
 	int err, ifindex;
 	enum nl80211_iftype type;
 	struct net_device *dev;
 	u32 flags;
 
+	memset(&params, 0, sizeof(params));
+
 	if (info->attrs[NL80211_ATTR_IFTYPE]) {
 		type = nla_get_u32(info->attrs[NL80211_ATTR_IFTYPE]);
 		if (type > NL80211_IFTYPE_MAX)
@@ -392,12 +399,18 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info)
 		goto unlock;
 	}
 
+	if (type == NL80211_IFTYPE_MESH_POINT &&
+	    info->attrs[NL80211_ATTR_MESH_ID]) {
+		params.mesh_id = nla_data(info->attrs[NL80211_ATTR_MESH_ID]);
+		params.mesh_id_len = nla_len(info->attrs[NL80211_ATTR_MESH_ID]);
+	}
+
 	rtnl_lock();
 	err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ?
 				  info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL,
 				  &flags);
 	err = drv->ops->change_virtual_intf(&drv->wiphy, ifindex,
-					    type, err ? NULL : &flags);
+					    type, err ? NULL : &flags, &params);
 	rtnl_unlock();
 
  unlock:
@@ -408,10 +421,13 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info)
 static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *drv;
+	struct vif_params params;
 	int err;
 	enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED;
 	u32 flags;
 
+	memset(&params, 0, sizeof(params));
+
 	if (!info->attrs[NL80211_ATTR_IFNAME])
 		return -EINVAL;
 
@@ -430,15 +446,22 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info)
 		goto unlock;
 	}
 
+	if (type == NL80211_IFTYPE_MESH_POINT &&
+	    info->attrs[NL80211_ATTR_MESH_ID]) {
+		params.mesh_id = nla_data(info->attrs[NL80211_ATTR_MESH_ID]);
+		params.mesh_id_len = nla_len(info->attrs[NL80211_ATTR_MESH_ID]);
+	}
+
 	rtnl_lock();
 	err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ?
 				  info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL,
 				  &flags);
 	err = drv->ops->add_virtual_intf(&drv->wiphy,
 		nla_data(info->attrs[NL80211_ATTR_IFNAME]),
-		type, err ? NULL : &flags);
+		type, err ? NULL : &flags, &params);
 	rtnl_unlock();
 
+
  unlock:
 	cfg80211_put_dev(drv);
 	return err;
@@ -866,10 +889,10 @@ static int parse_station_flags(struct nlattr *nla, u32 *staflags)
 
 static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq,
 				int flags, struct net_device *dev,
-				u8 *mac_addr, struct station_stats *stats)
+				u8 *mac_addr, struct station_info *sinfo)
 {
 	void *hdr;
-	struct nlattr *statsattr;
+	struct nlattr *sinfoattr;
 
 	hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_STATION);
 	if (!hdr)
@@ -878,20 +901,29 @@ static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq,
 	NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, dev->ifindex);
 	NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr);
 
-	statsattr = nla_nest_start(msg, NL80211_ATTR_STA_STATS);
-	if (!statsattr)
+	sinfoattr = nla_nest_start(msg, NL80211_ATTR_STA_INFO);
+	if (!sinfoattr)
 		goto nla_put_failure;
-	if (stats->filled & STATION_STAT_INACTIVE_TIME)
-		NLA_PUT_U32(msg, NL80211_STA_STAT_INACTIVE_TIME,
-			    stats->inactive_time);
-	if (stats->filled & STATION_STAT_RX_BYTES)
-		NLA_PUT_U32(msg, NL80211_STA_STAT_RX_BYTES,
-			    stats->rx_bytes);
-	if (stats->filled & STATION_STAT_TX_BYTES)
-		NLA_PUT_U32(msg, NL80211_STA_STAT_TX_BYTES,
-			    stats->tx_bytes);
-
-	nla_nest_end(msg, statsattr);
+	if (sinfo->filled & STATION_INFO_INACTIVE_TIME)
+		NLA_PUT_U32(msg, NL80211_STA_INFO_INACTIVE_TIME,
+			    sinfo->inactive_time);
+	if (sinfo->filled & STATION_INFO_RX_BYTES)
+		NLA_PUT_U32(msg, NL80211_STA_INFO_RX_BYTES,
+			    sinfo->rx_bytes);
+	if (sinfo->filled & STATION_INFO_TX_BYTES)
+		NLA_PUT_U32(msg, NL80211_STA_INFO_TX_BYTES,
+			    sinfo->tx_bytes);
+	if (sinfo->filled & STATION_INFO_LLID)
+		NLA_PUT_U16(msg, NL80211_STA_INFO_LLID,
+			    sinfo->llid);
+	if (sinfo->filled & STATION_INFO_PLID)
+		NLA_PUT_U16(msg, NL80211_STA_INFO_PLID,
+			    sinfo->plid);
+	if (sinfo->filled & STATION_INFO_PLINK_STATE)
+		NLA_PUT_U8(msg, NL80211_STA_INFO_PLINK_STATE,
+			    sinfo->plink_state);
+
+	nla_nest_end(msg, sinfoattr);
 
 	return genlmsg_end(msg, hdr);
 
@@ -899,17 +931,80 @@ static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq,
 	return genlmsg_cancel(msg, hdr);
 }
 
+static int nl80211_dump_station(struct sk_buff *skb,
+		struct netlink_callback *cb)
+{
+	int wp_idx = 0;
+	int if_idx = 0;
+	int sta_idx = cb->args[2];
+	int wp_start = cb->args[0];
+	int if_start = cb->args[1];
+	struct station_info sinfo;
+	struct cfg80211_registered_device *dev;
+	struct wireless_dev *wdev;
+	u8 mac_addr[ETH_ALEN];
+	int err;
+	int exit = 0;
+
+	/* TODO: filter by device */
+	mutex_lock(&cfg80211_drv_mutex);
+	list_for_each_entry(dev, &cfg80211_drv_list, list) {
+		if (exit)
+			break;
+		if (++wp_idx < wp_start)
+			continue;
+		if_idx = 0;
+
+		mutex_lock(&dev->devlist_mtx);
+		list_for_each_entry(wdev, &dev->netdev_list, list) {
+			if (exit)
+				break;
+			if (++if_idx < if_start)
+				continue;
+			if (!dev->ops->dump_station)
+				continue;
+
+			for (;; ++sta_idx) {
+				rtnl_lock();
+				err = dev->ops->dump_station(&dev->wiphy,
+						wdev->netdev, sta_idx, mac_addr,
+						&sinfo);
+				rtnl_unlock();
+				if (err) {
+					sta_idx = 0;
+					break;
+				}
+				if (nl80211_send_station(skb,
+						NETLINK_CB(cb->skb).pid,
+						cb->nlh->nlmsg_seq, NLM_F_MULTI,
+						wdev->netdev, mac_addr,
+						&sinfo) < 0) {
+					exit = 1;
+					break;
+				}
+			}
+		}
+		mutex_unlock(&dev->devlist_mtx);
+	}
+	mutex_unlock(&cfg80211_drv_mutex);
+
+	cb->args[0] = wp_idx;
+	cb->args[1] = if_idx;
+	cb->args[2] = sta_idx;
+
+	return skb->len;
+}
 
 static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *drv;
 	int err;
 	struct net_device *dev;
-	struct station_stats stats;
+	struct station_info sinfo;
 	struct sk_buff *msg;
 	u8 *mac_addr = NULL;
 
-	memset(&stats, 0, sizeof(stats));
+	memset(&sinfo, 0, sizeof(sinfo));
 
 	if (!info->attrs[NL80211_ATTR_MAC])
 		return -EINVAL;
@@ -926,15 +1021,18 @@ static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	rtnl_lock();
-	err = drv->ops->get_station(&drv->wiphy, dev, mac_addr, &stats);
+	err = drv->ops->get_station(&drv->wiphy, dev, mac_addr, &sinfo);
 	rtnl_unlock();
 
+	if (err)
+		goto out;
+
 	msg = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (!msg)
 		goto out;
 
 	if (nl80211_send_station(msg, info->snd_pid, info->snd_seq, 0,
-				 dev, mac_addr, &stats) < 0)
+				 dev, mac_addr, &sinfo) < 0)
 		goto out_free;
 
 	err = genlmsg_unicast(msg, info->snd_pid);
@@ -1005,6 +1103,10 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info)
 				&params.station_flags))
 		return -EINVAL;
 
+	if (info->attrs[NL80211_ATTR_STA_PLINK_ACTION])
+		params.plink_action =
+		    nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_ACTION]);
+
 	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
 	if (err)
 		return err;
@@ -1119,6 +1221,273 @@ static int nl80211_del_station(struct sk_buff *skb, struct genl_info *info)
 	return err;
 }
 
+static int nl80211_send_mpath(struct sk_buff *msg, u32 pid, u32 seq,
+				int flags, struct net_device *dev,
+				u8 *dst, u8 *next_hop,
+				struct mpath_info *pinfo)
+{
+	void *hdr;
+	struct nlattr *pinfoattr;
+
+	hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_STATION);
+	if (!hdr)
+		return -1;
+
+	NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, dev->ifindex);
+	NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, dst);
+	NLA_PUT(msg, NL80211_ATTR_MPATH_NEXT_HOP, ETH_ALEN, next_hop);
+
+	pinfoattr = nla_nest_start(msg, NL80211_ATTR_MPATH_INFO);
+	if (!pinfoattr)
+		goto nla_put_failure;
+	if (pinfo->filled & MPATH_INFO_FRAME_QLEN)
+		NLA_PUT_U32(msg, NL80211_MPATH_INFO_FRAME_QLEN,
+			    pinfo->frame_qlen);
+	if (pinfo->filled & MPATH_INFO_DSN)
+		NLA_PUT_U32(msg, NL80211_MPATH_INFO_DSN,
+			    pinfo->dsn);
+	if (pinfo->filled & MPATH_INFO_METRIC)
+		NLA_PUT_U32(msg, NL80211_MPATH_INFO_METRIC,
+			    pinfo->metric);
+	if (pinfo->filled & MPATH_INFO_EXPTIME)
+		NLA_PUT_U32(msg, NL80211_MPATH_INFO_EXPTIME,
+			    pinfo->exptime);
+	if (pinfo->filled & MPATH_INFO_FLAGS)
+		NLA_PUT_U8(msg, NL80211_MPATH_INFO_FLAGS,
+			    pinfo->flags);
+	if (pinfo->filled & MPATH_INFO_DISCOVERY_TIMEOUT)
+		NLA_PUT_U32(msg, NL80211_MPATH_INFO_DISCOVERY_TIMEOUT,
+			    pinfo->discovery_timeout);
+	if (pinfo->filled & MPATH_INFO_DISCOVERY_RETRIES)
+		NLA_PUT_U8(msg, NL80211_MPATH_INFO_DISCOVERY_RETRIES,
+			    pinfo->discovery_retries);
+
+	nla_nest_end(msg, pinfoattr);
+
+	return genlmsg_end(msg, hdr);
+
+ nla_put_failure:
+	return genlmsg_cancel(msg, hdr);
+}
+
+static int nl80211_dump_mpath(struct sk_buff *skb,
+		struct netlink_callback *cb)
+{
+	int wp_idx = 0;
+	int if_idx = 0;
+	int sta_idx = cb->args[2];
+	int wp_start = cb->args[0];
+	int if_start = cb->args[1];
+	struct mpath_info pinfo;
+	struct cfg80211_registered_device *dev;
+	struct wireless_dev *wdev;
+	u8 dst[ETH_ALEN];
+	u8 next_hop[ETH_ALEN];
+	int err;
+	int exit = 0;
+
+	/* TODO: filter by device */
+	mutex_lock(&cfg80211_drv_mutex);
+	list_for_each_entry(dev, &cfg80211_drv_list, list) {
+		if (exit)
+			break;
+		if (++wp_idx < wp_start)
+			continue;
+		if_idx = 0;
+
+		mutex_lock(&dev->devlist_mtx);
+		list_for_each_entry(wdev, &dev->netdev_list, list) {
+			if (exit)
+				break;
+			if (++if_idx < if_start)
+				continue;
+			if (!dev->ops->dump_mpath)
+				continue;
+
+			for (;; ++sta_idx) {
+				rtnl_lock();
+				err = dev->ops->dump_mpath(&dev->wiphy,
+						wdev->netdev, sta_idx, dst,
+						next_hop, &pinfo);
+				rtnl_unlock();
+				if (err) {
+					sta_idx = 0;
+					break;
+				}
+				if (nl80211_send_mpath(skb,
+						NETLINK_CB(cb->skb).pid,
+						cb->nlh->nlmsg_seq, NLM_F_MULTI,
+						wdev->netdev, dst, next_hop,
+						&pinfo) < 0) {
+					exit = 1;
+					break;
+				}
+			}
+		}
+		mutex_unlock(&dev->devlist_mtx);
+	}
+	mutex_unlock(&cfg80211_drv_mutex);
+
+	cb->args[0] = wp_idx;
+	cb->args[1] = if_idx;
+	cb->args[2] = sta_idx;
+
+	return skb->len;
+}
+
+static int nl80211_get_mpath(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	struct mpath_info pinfo;
+	struct sk_buff *msg;
+	u8 *dst = NULL;
+	u8 next_hop[ETH_ALEN];
+
+	memset(&pinfo, 0, sizeof(pinfo));
+
+	if (!info->attrs[NL80211_ATTR_MAC])
+		return -EINVAL;
+
+	dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->get_mpath) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->get_mpath(&drv->wiphy, dev, dst, next_hop, &pinfo);
+	rtnl_unlock();
+
+	if (err)
+		goto out;
+
+	msg = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+	if (!msg)
+		goto out;
+
+	if (nl80211_send_mpath(msg, info->snd_pid, info->snd_seq, 0,
+				 dev, dst, next_hop, &pinfo) < 0)
+		goto out_free;
+
+	err = genlmsg_unicast(msg, info->snd_pid);
+	goto out;
+
+ out_free:
+	nlmsg_free(msg);
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_set_mpath(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	u8 *dst = NULL;
+	u8 *next_hop = NULL;
+
+	if (!info->attrs[NL80211_ATTR_MAC])
+		return -EINVAL;
+
+	if (!info->attrs[NL80211_ATTR_MPATH_NEXT_HOP])
+		return -EINVAL;
+
+	dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
+	next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->change_mpath) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->change_mpath(&drv->wiphy, dev, dst, next_hop);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+static int nl80211_new_mpath(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	u8 *dst = NULL;
+	u8 *next_hop = NULL;
+
+	if (!info->attrs[NL80211_ATTR_MAC])
+		return -EINVAL;
+
+	if (!info->attrs[NL80211_ATTR_MPATH_NEXT_HOP])
+		return -EINVAL;
+
+	dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
+	next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->add_mpath) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->add_mpath(&drv->wiphy, dev, dst, next_hop);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
+static int nl80211_del_mpath(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *drv;
+	int err;
+	struct net_device *dev;
+	u8 *dst = NULL;
+
+	if (info->attrs[NL80211_ATTR_MAC])
+		dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
+
+	err = get_drv_dev_by_info_ifindex(info, &drv, &dev);
+	if (err)
+		return err;
+
+	if (!drv->ops->del_mpath) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rtnl_lock();
+	err = drv->ops->del_mpath(&drv->wiphy, dev, dst);
+	rtnl_unlock();
+
+ out:
+	cfg80211_put_dev(drv);
+	dev_put(dev);
+	return err;
+}
+
 static struct genl_ops nl80211_ops[] = {
 	{
 		.cmd = NL80211_CMD_GET_WIPHY,
@@ -1203,7 +1572,7 @@ static struct genl_ops nl80211_ops[] = {
 	{
 		.cmd = NL80211_CMD_GET_STATION,
 		.doit = nl80211_get_station,
-		/* TODO: implement dumpit */
+		.dumpit = nl80211_dump_station,
 		.policy = nl80211_policy,
 		.flags = GENL_ADMIN_PERM,
 	},
@@ -1225,6 +1594,31 @@ static struct genl_ops nl80211_ops[] = {
 		.policy = nl80211_policy,
 		.flags = GENL_ADMIN_PERM,
 	},
+	{
+		.cmd = NL80211_CMD_GET_MPATH,
+		.doit = nl80211_get_mpath,
+		.dumpit = nl80211_dump_mpath,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_SET_MPATH,
+		.doit = nl80211_set_mpath,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_NEW_MPATH,
+		.doit = nl80211_new_mpath,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = NL80211_CMD_DEL_MPATH,
+		.doit = nl80211_del_mpath,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
 };
 
 /* multicast groups */
-- 
cgit v1.2.3


From 6032f934c818e5c3435c9f17274fe1983f53c6b4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sat, 23 Feb 2008 15:17:07 +0100
Subject: mac80211: add mesh interface type

This adds the mesh interface type.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h         |  1 +
 net/mac80211/ieee80211.c       |  1 +
 net/mac80211/ieee80211_iface.c |  1 +
 net/mac80211/rx.c              | 10 ++++++++++
 net/mac80211/util.c            |  1 +
 5 files changed, 14 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 7a80c3981237..934cc25f757a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -443,6 +443,7 @@ enum ieee80211_if_types {
 	IEEE80211_IF_TYPE_AP,
 	IEEE80211_IF_TYPE_STA,
 	IEEE80211_IF_TYPE_IBSS,
+	IEEE80211_IF_TYPE_MESH_POINT,
 	IEEE80211_IF_TYPE_MNTR,
 	IEEE80211_IF_TYPE_WDS,
 	IEEE80211_IF_TYPE_VLAN,
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 2133c9fd27a4..1ddb8e1b6ab6 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -258,6 +258,7 @@ static int ieee80211_open(struct net_device *dev)
 	case IEEE80211_IF_TYPE_STA:
 	case IEEE80211_IF_TYPE_MNTR:
 	case IEEE80211_IF_TYPE_IBSS:
+	case IEEE80211_IF_TYPE_MESH_POINT:
 		/* no special treatment */
 		break;
 	case IEEE80211_IF_TYPE_INVALID:
diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c
index 677705046c6d..9523aeb71032 100644
--- a/net/mac80211/ieee80211_iface.c
+++ b/net/mac80211/ieee80211_iface.c
@@ -235,6 +235,7 @@ void ieee80211_if_reinit(struct net_device *dev)
 #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
 		}
 		break;
+	case IEEE80211_IF_TYPE_MESH_POINT:
 	case IEEE80211_IF_TYPE_STA:
 	case IEEE80211_IF_TYPE_IBSS:
 		kfree(sdata->u.sta.extra_ie);
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 48574f6c0e74..b7eeae0d3956 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1713,6 +1713,16 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata,
 			rx->sta = ieee80211_ibss_add_sta(sdata->dev, rx->skb,
 							 bssid, hdr->addr2);
 		break;
+	case IEEE80211_IF_TYPE_MESH_POINT:
+		if (!multicast &&
+		    compare_ether_addr(sdata->dev->dev_addr,
+				       hdr->addr1) != 0) {
+			if (!(sdata->dev->flags & IFF_PROMISC))
+				return 0;
+
+			rx->flags &= ~IEEE80211_TXRXD_RXRA_MATCH;
+		}
+		break;
 	case IEEE80211_IF_TYPE_VLAN:
 	case IEEE80211_IF_TYPE_AP:
 		if (!bssid) {
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index f64804fed0a9..790c32f894c0 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -382,6 +382,7 @@ void ieee80211_iterate_active_interfaces(
 		case IEEE80211_IF_TYPE_STA:
 		case IEEE80211_IF_TYPE_IBSS:
 		case IEEE80211_IF_TYPE_WDS:
+		case IEEE80211_IF_TYPE_MESH_POINT:
 			break;
 		}
 		if (sdata->dev == local->mdev)
-- 
cgit v1.2.3


From 902acc7896d7649fb30e4b22bd4e643c7f34b02c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sat, 23 Feb 2008 15:17:19 +0100
Subject: mac80211: clean up mesh code

Various cleanups, reducing the #ifdef mess and other things.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h          |   8 +++
 net/mac80211/cfg.c              |  32 +++--------
 net/mac80211/ieee80211.c        |   9 +--
 net/mac80211/ieee80211_i.h      |  56 ++++++++++++++++---
 net/mac80211/ieee80211_iface.c  |  66 ++++------------------
 net/mac80211/ieee80211_sta.c    |  72 +++++++++++-------------
 net/mac80211/mesh.c             |  67 ++++++++++++++++++++++
 net/mac80211/mesh.h             |  13 ++++-
 net/mac80211/mesh_plink.c       |  28 +++++++---
 net/mac80211/rc80211_pid_algo.c |   3 -
 net/mac80211/rx.c               |  28 +++++-----
 net/mac80211/sta_info.c         |  13 +----
 net/mac80211/sta_info.h         |   2 -
 net/mac80211/tx.c               | 120 +++++++++++++++++-----------------------
 net/mac80211/util.c             |  32 -----------
 15 files changed, 277 insertions(+), 272 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 934cc25f757a..6aca472d7a02 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -465,6 +465,14 @@ struct ieee80211_vif {
 	u8 drv_priv[0] __attribute__((__aligned__(sizeof(void *))));
 };
 
+static inline bool ieee80211_vif_is_mesh(struct ieee80211_vif *vif)
+{
+#ifdef CONFIG_MAC80211_MESH
+	return vif->type == IEEE80211_IF_TYPE_MESH_POINT;
+#endif
+	return false;
+}
+
 /**
  * struct ieee80211_if_init_conf - initial configuration of an interface
  *
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index b1befac1736a..6ac49231efa9 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -15,9 +15,7 @@
 #include "ieee80211_i.h"
 #include "cfg.h"
 #include "ieee80211_rate.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 
 #define DEFAULT_RATES 0
 
@@ -119,14 +117,10 @@ static int ieee80211_change_iface(struct wiphy *wiphy, int ifindex,
 	ieee80211_if_reinit(dev);
 	ieee80211_if_set_type(dev, itype);
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT &&
-	    params->mesh_id_len) {
-		sdata->u.sta.mesh_id_len = params->mesh_id_len;
-		memcpy(sdata->u.sta.mesh_id, params->mesh_id,
-		       params->mesh_id_len);
-	}
-#endif
+	if (ieee80211_vif_is_mesh(&sdata->vif) && params->mesh_id_len)
+		ieee80211_if_sta_set_mesh_id(&sdata->u.sta,
+					     params->mesh_id_len,
+					     params->mesh_id);
 
 	if (sdata->vif.type != IEEE80211_IF_TYPE_MNTR || !flags)
 		return 0;
@@ -317,9 +311,7 @@ static int ieee80211_config_default_key(struct wiphy *wiphy,
 
 static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo)
 {
-#ifdef CONFIG_MAC80211_MESH
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
-#endif
 
 	sinfo->filled = STATION_INFO_INACTIVE_TIME |
 			STATION_INFO_RX_BYTES |
@@ -329,8 +321,8 @@ static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo)
 	sinfo->rx_bytes = sta->rx_bytes;
 	sinfo->tx_bytes = sta->tx_bytes;
 
+	if (ieee80211_vif_is_mesh(&sdata->vif)) {
 #ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) {
 		sinfo->filled |= STATION_INFO_LLID |
 				 STATION_INFO_PLID |
 				 STATION_INFO_PLINK_STATE;
@@ -338,8 +330,8 @@ static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo)
 		sinfo->llid = le16_to_cpu(sta->llid);
 		sinfo->plid = le16_to_cpu(sta->plid);
 		sinfo->plink_state = sta->plink_state;
-	}
 #endif
+	}
 }
 
 
@@ -580,9 +572,7 @@ static void sta_apply_parameters(struct ieee80211_local *local,
 	u32 rates;
 	int i, j;
 	struct ieee80211_supported_band *sband;
-#ifdef CONFIG_MAC80211_MESH
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
-#endif
 
 	if (params->station_flags & STATION_FLAG_CHANGED) {
 		sta->flags &= ~WLAN_STA_AUTHORIZED;
@@ -621,9 +611,7 @@ static void sta_apply_parameters(struct ieee80211_local *local,
 		sta->supp_rates[local->oper_channel->band] = rates;
 	}
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT &&
-	    params->plink_action)
+	if (ieee80211_vif_is_mesh(&sdata->vif) && params->plink_action) {
 		switch (params->plink_action) {
 		case PLINK_ACTION_OPEN:
 			mesh_plink_open(sta);
@@ -632,7 +620,7 @@ static void sta_apply_parameters(struct ieee80211_local *local,
 			mesh_plink_block(sta);
 			break;
 		}
-#endif
+	}
 }
 
 static int ieee80211_add_station(struct wiphy *wiphy, struct net_device *dev,
@@ -655,11 +643,9 @@ static int ieee80211_add_station(struct wiphy *wiphy, struct net_device *dev,
 	} else
 		sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT)
+	if (ieee80211_vif_is_mesh(&sdata->vif))
 		sta = mesh_plink_add(mac, DEFAULT_RATES, dev);
 	else
-#endif
 		sta = sta_info_add(local, dev, mac, GFP_KERNEL);
 
 	if (IS_ERR(sta))
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 7106d651f4f9..727af295c969 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -26,9 +26,7 @@
 
 #include "ieee80211_i.h"
 #include "ieee80211_rate.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 #include "wep.h"
 #include "wme.h"
 #include "aes_ccm.h"
@@ -938,11 +936,9 @@ static int __ieee80211_if_config(struct net_device *dev,
 		conf.bssid = sdata->u.sta.bssid;
 		conf.ssid = sdata->u.sta.ssid;
 		conf.ssid_len = sdata->u.sta.ssid_len;
-#ifdef CONFIG_MAC80211_MESH
-	} else if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) {
+	} else if (ieee80211_vif_is_mesh(&sdata->vif)) {
 		conf.beacon = beacon;
 		ieee80211_start_mesh(dev);
-#endif
 	} else if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
 		conf.ssid = sdata->u.ap.ssid;
 		conf.ssid_len = sdata->u.ap.ssid_len;
@@ -1824,10 +1820,9 @@ static void __exit ieee80211_exit(void)
 	rc80211_simple_exit();
 	rc80211_pid_exit();
 
-#ifdef CONFIG_MAC80211_MESH
 	if (mesh_allocated)
 		ieee80211s_stop();
-#endif
+
 	ieee80211_wme_unregister();
 	ieee80211_debugfs_netdev_exit();
 }
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 49466b6996d1..7394c9b783b8 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -93,9 +93,8 @@ struct ieee80211_sta_bss {
 #ifdef CONFIG_MAC80211_MESH
 	u8 *mesh_id;
 	size_t mesh_id_len;
-#endif
-	/* mesh_cfg left out the ifdef to reduce clutter on bss handling */
 	u8 *mesh_cfg;
+#endif
 #define IEEE80211_MAX_SUPP_RATES 32
 	u8 supp_rates[IEEE80211_MAX_SUPP_RATES];
 	size_t supp_rates_len;
@@ -113,6 +112,30 @@ struct ieee80211_sta_bss {
 	u8 erp_value;
 };
 
+static inline u8 *bss_mesh_cfg(struct ieee80211_sta_bss *bss)
+{
+#ifdef CONFIG_MAC80211_MESH
+	return bss->mesh_cfg;
+#endif
+	return NULL;
+}
+
+static inline u8 *bss_mesh_id(struct ieee80211_sta_bss *bss)
+{
+#ifdef CONFIG_MAC80211_MESH
+	return bss->mesh_id;
+#endif
+	return NULL;
+}
+
+static inline u8 bss_mesh_id_len(struct ieee80211_sta_bss *bss)
+{
+#ifdef CONFIG_MAC80211_MESH
+	return bss->mesh_id_len;
+#endif
+	return 0;
+}
+
 
 typedef unsigned __bitwise__ ieee80211_tx_result;
 #define TX_CONTINUE	((__force ieee80211_tx_result) 0u)
@@ -233,7 +256,6 @@ struct ieee80211_if_vlan {
 	struct list_head list;
 };
 
-#ifdef CONFIG_MAC80211_MESH
 struct mesh_stats {
 	__u32 fwded_frames;		/* Mesh forwarded frames */
 	__u32 dropped_frames_ttl;	/* Not transmitted since mesh_ttl == 0*/
@@ -249,7 +271,6 @@ struct mesh_preq_queue {
 	u8 flags;
 };
 
-
 struct mesh_config {
 	/* Timeouts in ms */
 	/* Mesh plink management parameters */
@@ -268,7 +289,7 @@ struct mesh_config {
 	u32 path_refresh_time;
 	u16 min_discovery_timeout;
 };
-#endif
+
 
 /* flags used in struct ieee80211_if_sta.flags */
 #define IEEE80211_STA_SSID_SET		BIT(0)
@@ -361,6 +382,22 @@ struct ieee80211_if_sta {
 	int num_beacons; /* number of TXed beacon frames by this STA */
 };
 
+static inline void ieee80211_if_sta_set_mesh_id(struct ieee80211_if_sta *ifsta,
+						u8 mesh_id_len, u8 *mesh_id)
+{
+#ifdef CONFIG_MAC80211_MESH
+	ifsta->mesh_id_len = mesh_id_len;
+	memcpy(ifsta->mesh_id, mesh_id, mesh_id_len);
+#endif
+}
+
+#ifdef CONFIG_MAC80211_MESH
+#define IEEE80211_IFSTA_MESH_CTR_INC(sta, name)	\
+	do { (sta)->mshstats.name++; } while (0)
+#else
+#define IEEE80211_IFSTA_MESH_CTR_INC(sta, name) \
+	do { } while (0)
+#endif
 
 /* flags used in struct ieee80211_sub_if_data.flags */
 #define IEEE80211_SDATA_ALLMULTI	BIT(0)
@@ -472,7 +509,7 @@ struct ieee80211_sub_if_data {
 		struct dentry *dropped_frames_ttl;
 		struct dentry *dropped_frames_no_route;
 		struct dentry *estab_plinks;
-	struct timer_list mesh_path_timer;
+		struct timer_list mesh_path_timer;
 	} mesh_stats;
 
 	struct dentry *mesh_config_dir;
@@ -884,12 +921,17 @@ void sta_addba_resp_timer_expired(unsigned long data);
 u64 ieee80211_sta_get_rates(struct ieee80211_local *local,
 			    struct ieee802_11_elems *elems,
 			    enum ieee80211_band band);
-void ieee80211_start_mesh(struct net_device *dev);
 void ieee80211_sta_tx(struct net_device *dev, struct sk_buff *skb,
 		int encrypt);
 void ieee802_11_parse_elems(u8 *start, size_t len,
 				   struct ieee802_11_elems *elems);
 
+#ifdef CONFIG_MAC80211_MESH
+void ieee80211_start_mesh(struct net_device *dev);
+#else
+static inline void ieee80211_start_mesh(struct net_device *dev)
+{}
+#endif
 
 /* ieee80211_iface.c */
 int ieee80211_if_add(struct net_device *dev, const char *name,
diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c
index c2f92b78bfc9..b0f17a2b1a42 100644
--- a/net/mac80211/ieee80211_iface.c
+++ b/net/mac80211/ieee80211_iface.c
@@ -15,9 +15,7 @@
 #include "ieee80211_i.h"
 #include "sta_info.h"
 #include "debugfs_netdev.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 
 void ieee80211_if_sdata_init(struct ieee80211_sub_if_data *sdata)
 {
@@ -82,14 +80,11 @@ int ieee80211_if_add(struct net_device *dev, const char *name,
 	ieee80211_debugfs_add_netdev(sdata);
 	ieee80211_if_set_type(ndev, type);
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT &&
-	    params && params->mesh_id_len) {
-		sdata->u.sta.mesh_id_len = params->mesh_id_len;
-		memcpy(sdata->u.sta.mesh_id, params->mesh_id,
-		       params->mesh_id_len);
-	}
-#endif
+	if (ieee80211_vif_is_mesh(&sdata->vif) &&
+	    params && params->mesh_id_len)
+		ieee80211_if_sta_set_mesh_id(&sdata->u.sta,
+					     params->mesh_id_len,
+					     params->mesh_id);
 
 	/* we're under RTNL so all this is fine */
 	if (unlikely(local->reg_state == IEEE80211_DEV_UNREGISTERED)) {
@@ -170,47 +165,8 @@ void ieee80211_if_set_type(struct net_device *dev, int type)
 		msdata = IEEE80211_DEV_TO_SUB_IF(sdata->local->mdev);
 		sdata->bss = &msdata->u.ap;
 
-#ifdef CONFIG_MAC80211_MESH
-		if (type == IEEE80211_IF_TYPE_MESH_POINT) {
-			ifsta->mshcfg.dot11MeshRetryTimeout = MESH_RET_T;
-			ifsta->mshcfg.dot11MeshConfirmTimeout = MESH_CONF_T;
-			ifsta->mshcfg.dot11MeshHoldingTimeout = MESH_HOLD_T;
-			ifsta->mshcfg.dot11MeshMaxRetries = MESH_MAX_RETR;
-			ifsta->mshcfg.dot11MeshTTL = MESH_TTL;
-			ifsta->mshcfg.auto_open_plinks = true;
-			ifsta->mshcfg.dot11MeshMaxPeerLinks =
-				MESH_MAX_ESTAB_PLINKS;
-			ifsta->mshcfg.dot11MeshHWMPactivePathTimeout =
-				MESH_PATH_TIMEOUT;
-			ifsta->mshcfg.dot11MeshHWMPpreqMinInterval =
-				MESH_PREQ_MIN_INT;
-			ifsta->mshcfg.dot11MeshHWMPnetDiameterTraversalTime =
-				MESH_DIAM_TRAVERSAL_TIME;
-			ifsta->mshcfg.dot11MeshHWMPmaxPREQretries =
-				MESH_MAX_PREQ_RETRIES;
-			ifsta->mshcfg.path_refresh_time =
-				MESH_PATH_REFRESH_TIME;
-			ifsta->mshcfg.min_discovery_timeout =
-				MESH_MIN_DISCOVERY_TIMEOUT;
-			ifsta->accepting_plinks = true;
-			ifsta->preq_id = 0;
-			ifsta->dsn = 0;
-			atomic_set(&ifsta->mpaths, 0);
-			mesh_rmc_init(dev);
-			ifsta->last_preq = jiffies;
-			/* Allocate all mesh structures when creating the first
-			 * mesh interface.
-			 */
-			if (!mesh_allocated)
-				ieee80211s_init();
-			mesh_ids_set_default(ifsta);
-			setup_timer(&ifsta->mesh_path_timer,
-				    ieee80211_mesh_path_timer,
-				    (unsigned long) sdata);
-			INIT_LIST_HEAD(&ifsta->preq_queue.list);
-			spin_lock_init(&ifsta->mesh_preq_queue_lock);
-		}
-#endif
+		if (ieee80211_vif_is_mesh(&sdata->vif))
+			ieee80211_mesh_init_sdata(sdata);
 		break;
 	}
 	case IEEE80211_IF_TYPE_MNTR:
@@ -240,6 +196,10 @@ void ieee80211_if_reinit(struct net_device *dev)
 
 	ieee80211_if_sdata_deinit(sdata);
 
+	/* Need to handle mesh specially to allow eliding the function call */
+	if (ieee80211_vif_is_mesh(&sdata->vif))
+		mesh_rmc_free(dev);
+
 	switch (sdata->vif.type) {
 	case IEEE80211_IF_TYPE_INVALID:
 		/* cannot happen */
@@ -292,10 +252,6 @@ void ieee80211_if_reinit(struct net_device *dev)
 		}
 		break;
 	case IEEE80211_IF_TYPE_MESH_POINT:
-#ifdef CONFIG_MAC80211_MESH
-		mesh_rmc_free(dev);
-#endif
-		/* fall through */
 	case IEEE80211_IF_TYPE_STA:
 	case IEEE80211_IF_TYPE_IBSS:
 		kfree(sdata->u.sta.extra_ie);
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index d2dedcb5a954..9f933aeca719 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -31,9 +31,7 @@
 #include "ieee80211_i.h"
 #include "ieee80211_rate.h"
 #include "ieee80211_led.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 
 #define IEEE80211_AUTH_TIMEOUT (HZ / 5)
 #define IEEE80211_AUTH_MAX_TRIES 3
@@ -1897,12 +1895,13 @@ static void __ieee80211_rx_bss_hash_add(struct net_device *dev,
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	u8 hash_idx;
-#ifdef CONFIG_MAC80211_MESH
-	if (bss->mesh_cfg)
-		hash_idx = mesh_id_hash(bss->mesh_id, bss->mesh_id_len);
+
+	if (bss_mesh_cfg(bss))
+		hash_idx = mesh_id_hash(bss_mesh_id(bss),
+					bss_mesh_id_len(bss));
 	else
-#endif
 		hash_idx = STA_HASH(bss->bssid);
+
 	bss->hnext = local->sta_bss_hash[hash_idx];
 	local->sta_bss_hash[hash_idx] = bss;
 }
@@ -1967,7 +1966,8 @@ ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int freq,
 	spin_lock_bh(&local->sta_bss_lock);
 	bss = local->sta_bss_hash[STA_HASH(bssid)];
 	while (bss) {
-		if (!bss->mesh_cfg && !memcmp(bss->bssid, bssid, ETH_ALEN) &&
+		if (!bss_mesh_cfg(bss) &&
+		    !memcmp(bss->bssid, bssid, ETH_ALEN) &&
 		    bss->freq == freq &&
 		    bss->ssid_len == ssid_len &&
 		    (ssid_len == 0 || !memcmp(bss->ssid, ssid, ssid_len))) {
@@ -1991,8 +1991,8 @@ ieee80211_rx_mesh_bss_get(struct net_device *dev, u8 *mesh_id, int mesh_id_len,
 	spin_lock_bh(&local->sta_bss_lock);
 	bss = local->sta_bss_hash[mesh_id_hash(mesh_id, mesh_id_len)];
 	while (bss) {
-		if (bss->mesh_cfg &&
-		    !memcmp(bss->mesh_cfg, mesh_cfg, MESH_CFG_CMP_LEN) &&
+		if (bss_mesh_cfg(bss) &&
+		    !memcmp(bss_mesh_cfg(bss), mesh_cfg, MESH_CFG_CMP_LEN) &&
 		    bss->freq == freq &&
 		    mesh_id_len == bss->mesh_id_len &&
 		    (mesh_id_len == 0 || !memcmp(bss->mesh_id, mesh_id,
@@ -2053,10 +2053,8 @@ static void ieee80211_rx_bss_free(struct ieee80211_sta_bss *bss)
 	kfree(bss->rsn_ie);
 	kfree(bss->wmm_ie);
 	kfree(bss->ht_ie);
-#ifdef CONFIG_MAC80211_MESH
-	kfree(bss->mesh_id);
-	kfree(bss->mesh_cfg);
-#endif
+	kfree(bss_mesh_id(bss));
+	kfree(bss_mesh_cfg(bss));
 	kfree(bss);
 }
 
@@ -2322,16 +2320,14 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 	beacon_timestamp = le64_to_cpu(mgmt->u.beacon.timestamp);
 	ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems);
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT && elems.mesh_id
-			&& elems.mesh_config)
-		if (mesh_matches_local(&elems, dev)) {
-			u64 rates = ieee80211_sta_get_rates(local, &elems,
-							rx_status->band);
-			mesh_neighbour_update(mgmt->sa, rates, dev,
-				mesh_peer_accepts_plinks(&elems, dev));
-		}
-#endif
+	if (ieee80211_vif_is_mesh(&sdata->vif) && elems.mesh_id &&
+	    elems.mesh_config && mesh_matches_local(&elems, dev)) {
+		u64 rates = ieee80211_sta_get_rates(local, &elems,
+						rx_status->band);
+
+		mesh_neighbour_update(mgmt->sa, rates, dev,
+				      mesh_peer_accepts_plinks(&elems, dev));
+	}
 
 	if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates &&
 	    memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0 &&
@@ -2712,9 +2708,7 @@ static void ieee80211_rx_mgmt_action(struct net_device *dev,
 				     size_t len,
 				     struct ieee80211_rx_status *rx_status)
 {
-#ifdef CONFIG_MAC80211_MESH
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-#endif
 
 	if (len < IEEE80211_MIN_ACTION_SIZE)
 		return;
@@ -2747,17 +2741,14 @@ static void ieee80211_rx_mgmt_action(struct net_device *dev,
 			break;
 		}
 		break;
-#ifdef CONFIG_MAC80211_MESH
 	case PLINK_CATEGORY:
-		if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT)
+		if (ieee80211_vif_is_mesh(&sdata->vif))
 			mesh_rx_plink_frame(dev, mgmt, len, rx_status);
 		break;
-
 	case MESH_PATH_SEL_CATEGORY:
-		if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT)
+		if (ieee80211_vif_is_mesh(&sdata->vif))
 			mesh_rx_path_sel_frame(dev, mgmt, len);
 		break;
-#endif
 	default:
 		if (net_ratelimit())
 			printk(KERN_DEBUG "%s: Rx unknown action frame - "
@@ -3027,8 +3018,9 @@ void ieee80211_sta_work(struct work_struct *work)
 		ieee80211_sta_rx_queued_mgmt(dev, skb);
 
 #ifdef CONFIG_MAC80211_MESH
-	if (ifsta->preq_queue_len && time_after(jiffies, ifsta->last_preq +
-		msecs_to_jiffies(ifsta->mshcfg.dot11MeshHWMPpreqMinInterval)))
+	if (ifsta->preq_queue_len &&
+	    time_after(jiffies,
+		       ifsta->last_preq + msecs_to_jiffies(ifsta->mshcfg.dot11MeshHWMPpreqMinInterval)))
 		mesh_path_start_discovery(dev);
 #endif
 
@@ -3810,13 +3802,11 @@ ieee80211_sta_scan_result(struct net_device *dev,
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWESSID;
-	if (bss->mesh_cfg) {
-#ifdef CONFIG_MAC80211_MESH
-		iwe.u.data.length = bss->mesh_id_len;
+	if (bss_mesh_cfg(bss)) {
+		iwe.u.data.length = bss_mesh_id_len(bss);
 		iwe.u.data.flags = 1;
 		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->mesh_id);
-#endif
+						  bss_mesh_id(bss));
 	} else {
 		iwe.u.data.length = bss->ssid_len;
 		iwe.u.data.flags = 1;
@@ -3825,10 +3815,10 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	}
 
 	if (bss->capability & (WLAN_CAPABILITY_ESS | WLAN_CAPABILITY_IBSS
-	    || bss->mesh_cfg)) {
+	    || bss_mesh_cfg(bss))) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = SIOCGIWMODE;
-		if (bss->mesh_cfg)
+		if (bss_mesh_cfg(bss))
 			iwe.u.mode = IW_MODE_MESH;
 		else if (bss->capability & WLAN_CAPABILITY_ESS)
 			iwe.u.mode = IW_MODE_MASTER;
@@ -3919,9 +3909,9 @@ ieee80211_sta_scan_result(struct net_device *dev,
 		}
 	}
 
-	if (bss->mesh_cfg) {
+	if (bss_mesh_cfg(bss)) {
 		char *buf;
-		u8 *cfg = bss->mesh_cfg;
+		u8 *cfg = bss_mesh_cfg(bss);
 		buf = kmalloc(200, GFP_ATOMIC);
 		if (buf) {
 			memset(&iwe, 0, sizeof(iwe));
diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index 8ff533005d92..ebe1a7a80bad 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -381,3 +381,70 @@ endgrow:
 	else
 		return newtbl;
 }
+
+/**
+ * ieee80211_new_mesh_header - create a new mesh header
+ * @meshhdr:    uninitialized mesh header
+ * @sdata:	mesh interface to be used
+ *
+ * Return the header length.
+ */
+int ieee80211_new_mesh_header(struct ieee80211s_hdr *meshhdr,
+		struct ieee80211_sub_if_data *sdata)
+{
+	meshhdr->flags = 0;
+	meshhdr->ttl = sdata->u.sta.mshcfg.dot11MeshTTL;
+
+	meshhdr->seqnum[0] = sdata->u.sta.mesh_seqnum[0]++;
+	meshhdr->seqnum[1] = sdata->u.sta.mesh_seqnum[1];
+	meshhdr->seqnum[2] = sdata->u.sta.mesh_seqnum[2];
+
+	if (sdata->u.sta.mesh_seqnum[0] == 0) {
+		sdata->u.sta.mesh_seqnum[1]++;
+		if (sdata->u.sta.mesh_seqnum[1] == 0)
+			sdata->u.sta.mesh_seqnum[2]++;
+	}
+
+	return 5;
+}
+
+void ieee80211_mesh_init_sdata(struct ieee80211_sub_if_data *sdata)
+{
+	struct ieee80211_if_sta *ifsta = &sdata->u.sta;
+
+	ifsta->mshcfg.dot11MeshRetryTimeout = MESH_RET_T;
+	ifsta->mshcfg.dot11MeshConfirmTimeout = MESH_CONF_T;
+	ifsta->mshcfg.dot11MeshHoldingTimeout = MESH_HOLD_T;
+	ifsta->mshcfg.dot11MeshMaxRetries = MESH_MAX_RETR;
+	ifsta->mshcfg.dot11MeshTTL = MESH_TTL;
+	ifsta->mshcfg.auto_open_plinks = true;
+	ifsta->mshcfg.dot11MeshMaxPeerLinks =
+		MESH_MAX_ESTAB_PLINKS;
+	ifsta->mshcfg.dot11MeshHWMPactivePathTimeout =
+		MESH_PATH_TIMEOUT;
+	ifsta->mshcfg.dot11MeshHWMPpreqMinInterval =
+		MESH_PREQ_MIN_INT;
+	ifsta->mshcfg.dot11MeshHWMPnetDiameterTraversalTime =
+		MESH_DIAM_TRAVERSAL_TIME;
+	ifsta->mshcfg.dot11MeshHWMPmaxPREQretries =
+		MESH_MAX_PREQ_RETRIES;
+	ifsta->mshcfg.path_refresh_time =
+		MESH_PATH_REFRESH_TIME;
+	ifsta->mshcfg.min_discovery_timeout =
+		MESH_MIN_DISCOVERY_TIMEOUT;
+	ifsta->accepting_plinks = true;
+	ifsta->preq_id = 0;
+	ifsta->dsn = 0;
+	atomic_set(&ifsta->mpaths, 0);
+	mesh_rmc_init(sdata->dev);
+	ifsta->last_preq = jiffies;
+	/* Allocate all mesh structures when creating the first mesh interface. */
+	if (!mesh_allocated)
+		ieee80211s_init();
+	mesh_ids_set_default(ifsta);
+	setup_timer(&ifsta->mesh_path_timer,
+		    ieee80211_mesh_path_timer,
+		    (unsigned long) sdata);
+	INIT_LIST_HEAD(&ifsta->preq_queue.list);
+	spin_lock_init(&ifsta->mesh_preq_queue_lock);
+}
diff --git a/net/mac80211/mesh.h b/net/mac80211/mesh.h
index ac8923793908..d565b3fb9e6a 100644
--- a/net/mac80211/mesh.h
+++ b/net/mac80211/mesh.h
@@ -11,10 +11,10 @@
 #ifndef IEEE80211S_H
 #define IEEE80211S_H
 
-#include "ieee80211_i.h"
+#include <linux/types.h>
 #include <linux/jhash.h>
+#include "ieee80211_i.h"
 
-extern int mesh_allocated;
 
 /* Data structures */
 
@@ -211,6 +211,8 @@ void mesh_rmc_free(struct net_device *dev);
 int mesh_rmc_init(struct net_device *dev);
 void ieee80211s_init(void);
 void ieee80211s_stop(void);
+void ieee80211_mesh_init_sdata(struct ieee80211_sub_if_data *sdata);
+
 /* Mesh paths */
 int mesh_nexthop_lookup(u8 *next_hop, struct sk_buff *skb,
 		struct net_device *dev);
@@ -257,6 +259,9 @@ void mesh_path_timer(unsigned long data);
 void mesh_path_flush_by_nexthop(struct sta_info *sta);
 void mesh_path_discard_frame(struct sk_buff *skb, struct net_device *dev);
 
+#ifdef CONFIG_MAC80211_MESH
+extern int mesh_allocated;
+
 static inline int mesh_plink_free_count(struct ieee80211_sub_if_data *sdata)
 {
 	return sdata->u.sta.mshcfg.dot11MeshMaxPeerLinks -
@@ -278,6 +283,10 @@ static inline void mesh_path_activate(struct mesh_path *mpath)
 	for (i = 0; i <= x->hash_mask; i++) \
 		hlist_for_each_entry_rcu(node, p, &x->hash_buckets[i], list)
 
+#else
+#define mesh_allocated	0
+#endif
+
 #define MESH_PREQ(skb)	(skb->cb + 30)
 
 #endif /* IEEE80211S_H */
diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c
index 5cd97e99be62..0b0e8d7eb9c7 100644
--- a/net/mac80211/mesh_plink.c
+++ b/net/mac80211/mesh_plink.c
@@ -6,11 +6,11 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
-
+#include <linux/kernel.h>
+#include <linux/random.h>
 #include "ieee80211_i.h"
 #include "ieee80211_rate.h"
 #include "mesh.h"
-#include <linux/random.h>
 
 #ifdef CONFIG_MAC80211_VERBOSE_MPL_DEBUG
 #define mpl_dbg(fmt, args...)	printk(KERN_DEBUG fmt, ##args)
@@ -131,7 +131,7 @@ struct sta_info *mesh_plink_add(u8 *hw_addr, u64 rates, struct net_device *dev)
 }
 
 /**
- * mesh_plink_deactivate - deactivate mesh peer link
+ * __mesh_plink_deactivate - deactivate mesh peer link
  *
  * @sta: mesh peer link to deactivate
  *
@@ -139,7 +139,7 @@ struct sta_info *mesh_plink_add(u8 *hw_addr, u64 rates, struct net_device *dev)
  *
  * Locking: the caller must hold sta->plink_lock
  */
-void mesh_plink_deactivate(struct sta_info *sta)
+static void __mesh_plink_deactivate(struct sta_info *sta)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev);
 	if (sta->plink_state == ESTAB)
@@ -148,6 +148,20 @@ void mesh_plink_deactivate(struct sta_info *sta)
 	mesh_path_flush_by_nexthop(sta);
 }
 
+/**
+ * __mesh_plink_deactivate - deactivate mesh peer link
+ *
+ * @sta: mesh peer link to deactivate
+ *
+ * All mesh paths with this peer as next hop will be flushed
+ */
+void mesh_plink_deactivate(struct sta_info *sta)
+{
+	spin_lock_bh(&sta->plink_lock);
+	__mesh_plink_deactivate(sta);
+	spin_unlock_bh(&sta->plink_lock);
+}
+
 static int mesh_plink_frame_tx(struct net_device *dev,
 		enum plink_frame_type action, u8 *da, __le16 llid, __le16 plid,
 		__le16 reason) {
@@ -365,7 +379,7 @@ void mesh_plink_block(struct sta_info *sta)
 #endif
 
 	spin_lock_bh(&sta->plink_lock);
-	mesh_plink_deactivate(sta);
+	__mesh_plink_deactivate(sta);
 	sta->plink_state = BLOCKED;
 	spin_unlock_bh(&sta->plink_lock);
 }
@@ -390,7 +404,7 @@ int mesh_plink_close(struct sta_info *sta)
 		sta_info_put(sta);
 		return 0;
 	} else if (sta->plink_state == ESTAB) {
-		mesh_plink_deactivate(sta);
+		__mesh_plink_deactivate(sta);
 		/* The timer should not be running */
 		if (!mod_plink_timer(sta, dot11MeshHoldingTimeout(sdata)))
 			__sta_info_get(sta);
@@ -699,7 +713,7 @@ void mesh_rx_plink_frame(struct net_device *dev, struct ieee80211_mgmt *mgmt,
 		case CLS_ACPT:
 			reason = cpu_to_le16(MESH_CLOSE_RCVD);
 			sta->reason = reason;
-			mesh_plink_deactivate(sta);
+			__mesh_plink_deactivate(sta);
 			sta->plink_state = HOLDING;
 			llid = sta->llid;
 			if (!mod_plink_timer(sta,
diff --git a/net/mac80211/rc80211_pid_algo.c b/net/mac80211/rc80211_pid_algo.c
index 4a51647a41af..217c0f487bba 100644
--- a/net/mac80211/rc80211_pid_algo.c
+++ b/net/mac80211/rc80211_pid_algo.c
@@ -15,10 +15,7 @@
 #include <linux/debugfs.h>
 #include <net/mac80211.h>
 #include "ieee80211_rate.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
-
 #include "rc80211_pid.h"
 
 
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index cc4a896c617f..d0018fc40b09 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -20,9 +20,7 @@
 
 #include "ieee80211_i.h"
 #include "ieee80211_led.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 #include "wep.h"
 #include "wpa.h"
 #include "tkip.h"
@@ -439,6 +437,13 @@ ieee80211_rx_mesh_check(struct ieee80211_txrx_data *rx)
 	else
 		return RX_CONTINUE;
 }
+#undef msh_h_get
+#else
+static inline ieee80211_rx_result
+ieee80211_rx_mesh_check(struct ieee80211_txrx_data *rx)
+{
+	return RX_CONTINUE;
+}
 #endif
 
 
@@ -477,10 +482,8 @@ ieee80211_rx_h_check(struct ieee80211_txrx_data *rx)
 	 * responsible for filtering on both auth and assoc states.
 	 */
 
-#ifdef CONFIG_MAC80211_MESH
-	if (rx->sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT)
+	if (ieee80211_vif_is_mesh(&rx->sdata->vif))
 		return ieee80211_rx_mesh_check(rx);
-#endif
 
 	if (unlikely(((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA ||
 		      ((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL &&
@@ -1111,8 +1114,7 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx)
 
 	hdrlen = ieee80211_get_hdrlen(fc);
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) {
+	if (ieee80211_vif_is_mesh(&sdata->vif)) {
 		int meshhdrlen = ieee80211_get_mesh_hdrlen(
 				(struct ieee80211s_hdr *) (skb->data + hdrlen));
 		/* Copy on cb:
@@ -1126,7 +1128,6 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx)
 		memcpy(MESH_PREQ(skb), hdr->addr2, ETH_ALEN);
 		hdrlen += meshhdrlen;
 	}
-#endif
 
 	/* convert IEEE 802.11 header + possible LLC headers into Ethernet
 	 * header
@@ -1306,9 +1307,8 @@ ieee80211_deliver_skb(struct ieee80211_txrx_data *rx)
 		}
 	}
 
-#ifdef CONFIG_MAC80211_MESH
 	/* Mesh forwarding */
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) {
+	if (ieee80211_vif_is_mesh(&sdata->vif)) {
 		u8 *mesh_ttl = &((struct ieee80211s_hdr *)skb->cb)->ttl;
 		(*mesh_ttl)--;
 
@@ -1321,12 +1321,13 @@ ieee80211_deliver_skb(struct ieee80211_txrx_data *rx)
 				else
 					xmit_skb->pkt_type = PACKET_OTHERHOST;
 			} else
-				sdata->u.sta.mshstats.dropped_frames_ttl++;
-
+				IEEE80211_IFSTA_MESH_CTR_INC(&sdata->u.sta,
+							     dropped_frames_ttl);
 		} else if (skb->pkt_type != PACKET_OTHERHOST &&
 			compare_ether_addr(dev->dev_addr, skb->data) != 0) {
 			if (*mesh_ttl == 0) {
-				sdata->u.sta.mshstats.dropped_frames_ttl++;
+				IEEE80211_IFSTA_MESH_CTR_INC(&sdata->u.sta,
+							     dropped_frames_ttl);
 				dev_kfree_skb(skb);
 				skb = NULL;
 			} else {
@@ -1337,7 +1338,6 @@ ieee80211_deliver_skb(struct ieee80211_txrx_data *rx)
 			}
 		}
 	}
-#endif
 
 	if (skb) {
 		/* deliver to local stack */
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 1f3c9eb98500..81c4e3392f40 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -21,9 +21,7 @@
 #include "ieee80211_rate.h"
 #include "sta_info.h"
 #include "debugfs_sta.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 
 /* Caller must hold local->sta_lock */
 static void sta_info_hash_add(struct ieee80211_local *local,
@@ -309,10 +307,8 @@ void sta_info_remove(struct sta_info *sta)
 	}
 	local->num_sta--;
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT)
+	if (ieee80211_vif_is_mesh(&sdata->vif))
 		mesh_accept_plinks_update(sdata->dev);
-#endif
 }
 
 void sta_info_free(struct sta_info *sta)
@@ -329,13 +325,8 @@ void sta_info_free(struct sta_info *sta)
 	sta_info_remove(sta);
 	write_unlock_bh(&local->sta_lock);
 
-#ifdef CONFIG_MAC80211_MESH
-	if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) {
-		spin_lock_bh(&sta->plink_lock);
+	if (ieee80211_vif_is_mesh(&sdata->vif))
 		mesh_plink_deactivate(sta);
-		spin_unlock_bh(&sta->plink_lock);
-	}
-#endif
 
 	while ((skb = skb_dequeue(&sta->ps_tx_buf)) != NULL) {
 		local->total_ps_buffered--;
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index 9d1d7a0e3114..4ad500373d5a 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -107,7 +107,6 @@ struct tid_ampdu_rx {
 	struct timer_list session_timer;
 };
 
-#ifdef CONFIG_MAC80211_MESH
 enum plink_state {
 	LISTEN,
 	OPN_SNT,
@@ -117,7 +116,6 @@ enum plink_state {
 	HOLDING,
 	BLOCKED
 };
-#endif
 
 /**
  * struct sta_ampdu_mlme - STA aggregation information.
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index fc1ffb55ed5c..3b06e0d8f35c 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -26,9 +26,7 @@
 
 #include "ieee80211_i.h"
 #include "ieee80211_led.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 #include "wep.h"
 #include "wpa.h"
 #include "wme.h"
@@ -1460,7 +1458,7 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb,
 				goto fail;
 			}
 			meshhdrlen = ieee80211_new_mesh_header(&mesh_hdr,
-					sdata);
+							       sdata);
 		}
 		hdrlen = 30;
 		break;
@@ -1778,40 +1776,6 @@ static void ieee80211_beacon_add_tim(struct ieee80211_local *local,
 	read_unlock_bh(&local->sta_lock);
 }
 
-#ifdef CONFIG_MAC80211_MESH
-static struct sk_buff *ieee80211_mesh_beacon_get(struct net_device *dev)
-{
-	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
-	struct sk_buff *skb = dev_alloc_skb(local->hw.extra_tx_headroom + 400);
-	struct ieee80211_mgmt *mgmt;
-	u8 *pos;
-
-	if (!skb)
-		return NULL;
-	skb_reserve(skb, local->hw.extra_tx_headroom);
-	mgmt = (struct ieee80211_mgmt *)
-		skb_put(skb, 24 + sizeof(mgmt->u.beacon));
-	memset(mgmt, 0, 24 + sizeof(mgmt->u.beacon));
-	mgmt->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT,
-					   IEEE80211_STYPE_BEACON);
-	memset(mgmt->da, 0xff, ETH_ALEN);
-	memcpy(mgmt->sa, dev->dev_addr, ETH_ALEN);
-	/* BSSID is left zeroed, wildcard value */
-	mgmt->u.beacon.beacon_int =
-		cpu_to_le16(local->hw.conf.beacon_int);
-	mgmt->u.beacon.capab_info = 0x0; /* 0x0 for MPs */
-
-	pos = skb_put(skb, 2);
-	*pos++ = WLAN_EID_SSID;
-	*pos++ = 0x0;
-
-	mesh_mgmt_ies_add(skb, dev);
-
-	return skb;
-}
-#endif
-
-
 struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 				     struct ieee80211_vif *vif,
 				     struct ieee80211_tx_control *control)
@@ -1824,8 +1788,10 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 	struct rate_selection rsel;
 	struct beacon_data *beacon;
 	struct ieee80211_supported_band *sband;
+	struct ieee80211_mgmt *mgmt;
 	int *num_beacons;
-	int err = 0;
+	bool err = true;
+	u8 *pos;
 
 	sband = local->hw.wiphy->bands[local->hw.conf.channel->band];
 
@@ -1834,47 +1800,65 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw,
 	sdata = vif_to_sdata(vif);
 	bdev = sdata->dev;
 
-	switch (sdata->vif.type) {
-	case IEEE80211_IF_TYPE_AP:
+	if (sdata->vif.type == IEEE80211_IF_TYPE_AP) {
 		ap = &sdata->u.ap;
 		beacon = rcu_dereference(ap->beacon);
-		if (!ap || !beacon) {
-			err = -1;
-			break;
-		}
+		if (ap && beacon) {
+			/*
+			 * headroom, head length,
+			 * tail length and maximum TIM length
+			 */
+			skb = dev_alloc_skb(local->tx_headroom +
+					    beacon->head_len +
+					    beacon->tail_len + 256);
+			if (!skb)
+				goto out;
 
-		/* headroom, head length, tail length and maximum TIM length */
-		skb = dev_alloc_skb(local->tx_headroom + beacon->head_len +
-				    beacon->tail_len + 256);
-		if (!skb)
-			goto out;
+			skb_reserve(skb, local->tx_headroom);
+			memcpy(skb_put(skb, beacon->head_len), beacon->head,
+			       beacon->head_len);
 
-		skb_reserve(skb, local->tx_headroom);
-		memcpy(skb_put(skb, beacon->head_len), beacon->head,
-		       beacon->head_len);
+			ieee80211_include_sequence(sdata,
+					(struct ieee80211_hdr *)skb->data);
 
-		ieee80211_include_sequence(sdata,
-					   (struct ieee80211_hdr *)skb->data);
+			ieee80211_beacon_add_tim(local, ap, skb, beacon);
 
-		ieee80211_beacon_add_tim(local, ap, skb, beacon);
+			if (beacon->tail)
+				memcpy(skb_put(skb, beacon->tail_len),
+				       beacon->tail, beacon->tail_len);
 
-		if (beacon->tail)
-			memcpy(skb_put(skb, beacon->tail_len), beacon->tail,
-			       beacon->tail_len);
+			num_beacons = &ap->num_beacons;
 
-		num_beacons = &ap->num_beacons;
-		break;
+			err = false;
+		}
+	} else if (ieee80211_vif_is_mesh(&sdata->vif)) {
+		/* headroom, head length, tail length and maximum TIM length */
+		skb = dev_alloc_skb(local->tx_headroom + 400);
+		if (!skb)
+			goto out;
+
+		skb_reserve(skb, local->hw.extra_tx_headroom);
+		mgmt = (struct ieee80211_mgmt *)
+			skb_put(skb, 24 + sizeof(mgmt->u.beacon));
+		memset(mgmt, 0, 24 + sizeof(mgmt->u.beacon));
+		mgmt->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT,
+						   IEEE80211_STYPE_BEACON);
+		memset(mgmt->da, 0xff, ETH_ALEN);
+		memcpy(mgmt->sa, sdata->dev->dev_addr, ETH_ALEN);
+		/* BSSID is left zeroed, wildcard value */
+		mgmt->u.beacon.beacon_int =
+			cpu_to_le16(local->hw.conf.beacon_int);
+		mgmt->u.beacon.capab_info = 0x0; /* 0x0 for MPs */
+
+		pos = skb_put(skb, 2);
+		*pos++ = WLAN_EID_SSID;
+		*pos++ = 0x0;
+
+		mesh_mgmt_ies_add(skb, sdata->dev);
 
-#ifdef CONFIG_MAC80211_MESH
-	case IEEE80211_IF_TYPE_MESH_POINT:
-		skb = ieee80211_mesh_beacon_get(bdev);
 		num_beacons = &sdata->u.sta.num_beacons;
-		break;
-#endif
 
-	default:
-		err = -1;
-		break;
+		err = false;
 	}
 
 	if (err) {
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 6b50b6c12da3..b46496fa2e10 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -26,9 +26,7 @@
 
 #include "ieee80211_i.h"
 #include "ieee80211_rate.h"
-#ifdef CONFIG_MAC80211_MESH
 #include "mesh.h"
-#endif
 #include "wme.h"
 
 /* privid for wiphys to determine whether they belong to us or not */
@@ -149,7 +147,6 @@ int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb)
 }
 EXPORT_SYMBOL(ieee80211_get_hdrlen_from_skb);
 
-#ifdef CONFIG_MAC80211_MESH
 int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr)
 {
 	int ae = meshhdr->flags & IEEE80211S_FLAGS_AE;
@@ -167,7 +164,6 @@ int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr)
 		return 5;
 	}
 }
-#endif
 
 void ieee80211_tx_set_iswep(struct ieee80211_txrx_data *tx)
 {
@@ -418,31 +414,3 @@ void ieee80211_iterate_active_interfaces(
 	rcu_read_unlock();
 }
 EXPORT_SYMBOL_GPL(ieee80211_iterate_active_interfaces);
-
-#ifdef CONFIG_MAC80211_MESH
-/**
- * ieee80211_new_mesh_header - create a new mesh header
- * @meshhdr:    uninitialized mesh header
- * @sdata:	mesh interface to be used
- *
- * Return the header length.
- */
-int ieee80211_new_mesh_header(struct ieee80211s_hdr *meshhdr,
-		struct ieee80211_sub_if_data *sdata)
-{
-	meshhdr->flags = 0;
-	meshhdr->ttl = sdata->u.sta.mshcfg.dot11MeshTTL;
-
-	meshhdr->seqnum[0] = sdata->u.sta.mesh_seqnum[0]++;
-	meshhdr->seqnum[1] = sdata->u.sta.mesh_seqnum[1];
-	meshhdr->seqnum[2] = sdata->u.sta.mesh_seqnum[2];
-
-	if (sdata->u.sta.mesh_seqnum[0] == 0) {
-		sdata->u.sta.mesh_seqnum[1]++;
-		if (sdata->u.sta.mesh_seqnum[1] == 0)
-			sdata->u.sta.mesh_seqnum[2]++;
-	}
-
-	return 5;
-}
-#endif
-- 
cgit v1.2.3


From dbbea6713d6096cd1c411cb453a6b71292c78b33 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Tue, 26 Feb 2008 14:34:06 +0100
Subject: mac80211: add documentation book

Quite a while ago I started this book. The required kernel-doc
patches have since gone into the tree so it is now possible to
build the book in mainline.

The actual documentation is still rather incomplete and not all
things are linked into the book, but this enables us to edit
the documentation collaboratively, hopefully driver authors can
add documentation based on their experience with mac80211.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 Documentation/DocBook/Makefile      |   3 +-
 Documentation/DocBook/mac80211.tmpl | 335 ++++++++++++++++++++++++++++++++++++
 include/net/mac80211.h              |   7 +-
 net/mac80211/key.c                  |   4 +-
 net/mac80211/sta_info.c             |  16 +-
 5 files changed, 351 insertions(+), 14 deletions(-)
 create mode 100644 Documentation/DocBook/mac80211.tmpl

(limited to 'include')

diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile
index 300e1707893f..9ebd1f00c6e7 100644
--- a/Documentation/DocBook/Makefile
+++ b/Documentation/DocBook/Makefile
@@ -11,7 +11,8 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \
 	    procfs-guide.xml writing_usb_driver.xml networking.xml \
 	    kernel-api.xml filesystems.xml lsm.xml usb.xml \
 	    gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \
-	    genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml
+	    genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml \
+	    mac80211.xml
 
 ###
 # The build process is as follows (targets):
diff --git a/Documentation/DocBook/mac80211.tmpl b/Documentation/DocBook/mac80211.tmpl
new file mode 100644
index 000000000000..b651e0a4b1c0
--- /dev/null
+++ b/Documentation/DocBook/mac80211.tmpl
@@ -0,0 +1,335 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
+	"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
+
+<book id="mac80211-developers-guide">
+  <bookinfo>
+    <title>The mac80211 subsystem for kernel developers</title>
+
+    <authorgroup>
+      <author>
+        <firstname>Johannes</firstname>
+        <surname>Berg</surname>
+        <affiliation>
+          <address><email>johannes@sipsolutions.net</email></address>
+        </affiliation>
+      </author>
+    </authorgroup>
+
+    <copyright>
+      <year>2007</year>
+      <year>2008</year>
+      <holder>Johannes Berg</holder>
+    </copyright>
+
+    <legalnotice>
+      <para>
+        This documentation is free software; you can redistribute
+        it and/or modify it under the terms of the GNU General Public
+        License version 2 as published by the Free Software Foundation.
+      </para>
+
+      <para>
+        This documentation is distributed in the hope that it will be
+        useful, but WITHOUT ANY WARRANTY; without even the implied
+        warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+        See the GNU General Public License for more details.
+      </para>
+
+      <para>
+        You should have received a copy of the GNU General Public
+        License along with this documentation; if not, write to the Free
+        Software Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+        MA 02111-1307 USA
+      </para>
+
+      <para>
+        For more details see the file COPYING in the source
+        distribution of Linux.
+      </para>
+    </legalnotice>
+
+    <abstract>
+!Pinclude/net/mac80211.h Introduction
+!Pinclude/net/mac80211.h Warning
+    </abstract>
+  </bookinfo>
+
+  <toc></toc>
+
+<!--
+Generally, this document shall be ordered by increasing complexity.
+It is important to note that readers should be able to read only
+the first few sections to get a working driver and only advanced
+usage should require reading the full document.
+-->
+
+  <part>
+    <title>The basic mac80211 driver interface</title>
+    <partintro>
+      <para>
+        You should read and understand the information contained
+        within this part of the book while implementing a driver.
+        In some chapters, advanced usage is noted, that may be
+        skipped at first.
+      </para>
+      <para>
+        This part of the book only covers station and monitor mode
+        functionality, additional information required to implement
+        the other modes is covered in the second part of the book.
+      </para>
+    </partintro>
+
+    <chapter id="basics">
+      <title>Basic hardware handling</title>
+      <para>TBD</para>
+      <para>
+        This chapter shall contain information on getting a hw
+        struct allocated and registered with mac80211.
+      </para>
+      <para>
+        Since it is required to allocate rates/modes before registering
+        a hw struct, this chapter shall also contain information on setting
+        up the rate/mode structs.
+      </para>
+      <para>
+        Additionally, some discussion about the callbacks and
+        the general programming model should be in here, including
+        the definition of ieee80211_ops which will be referred to
+        a lot.
+      </para>
+      <para>
+        Finally, a discussion of hardware capabilities should be done
+        with references to other parts of the book.
+      </para>
+<!-- intentionally multiple !F lines to get proper order -->
+!Finclude/net/mac80211.h ieee80211_hw
+!Finclude/net/mac80211.h ieee80211_hw_flags
+!Finclude/net/mac80211.h SET_IEEE80211_DEV
+!Finclude/net/mac80211.h SET_IEEE80211_PERM_ADDR
+!Finclude/net/mac80211.h ieee80211_ops
+!Finclude/net/mac80211.h ieee80211_alloc_hw
+!Finclude/net/mac80211.h ieee80211_register_hw
+!Finclude/net/mac80211.h ieee80211_get_tx_led_name
+!Finclude/net/mac80211.h ieee80211_get_rx_led_name
+!Finclude/net/mac80211.h ieee80211_get_assoc_led_name
+!Finclude/net/mac80211.h ieee80211_get_radio_led_name
+!Finclude/net/mac80211.h ieee80211_unregister_hw
+!Finclude/net/mac80211.h ieee80211_free_hw
+    </chapter>
+
+    <chapter id="phy-handling">
+      <title>PHY configuration</title>
+      <para>TBD</para>
+      <para>
+        This chapter should describe PHY handling including
+        start/stop callbacks and the various structures used.
+      </para>
+!Finclude/net/mac80211.h ieee80211_conf
+!Finclude/net/mac80211.h ieee80211_conf_flags
+    </chapter>
+
+    <chapter id="iface-handling">
+      <title>Virtual interfaces</title>
+      <para>TBD</para>
+      <para>
+        This chapter should describe virtual interface basics
+        that are relevant to the driver (VLANs, MGMT etc are not.)
+        It should explain the use of the add_iface/remove_iface
+        callbacks as well as the interface configuration callbacks.
+      </para>
+      <para>Things related to AP mode should be discussed there.</para>
+      <para>
+        Things related to supporting multiple interfaces should be
+        in the appropriate chapter, a BIG FAT note should be here about
+        this though and the recommendation to allow only a single
+        interface in STA mode at first!
+      </para>
+!Finclude/net/mac80211.h ieee80211_if_types
+!Finclude/net/mac80211.h ieee80211_if_init_conf
+!Finclude/net/mac80211.h ieee80211_if_conf
+    </chapter>
+
+    <chapter id="rx-tx">
+      <title>Receive and transmit processing</title>
+      <sect1>
+        <title>what should be here</title>
+        <para>TBD</para>
+        <para>
+          This should describe the receive and transmit
+          paths in mac80211/the drivers as well as
+          transmit status handling.
+        </para>
+      </sect1>
+      <sect1>
+        <title>Frame format</title>
+!Pinclude/net/mac80211.h Frame format
+      </sect1>
+      <sect1>
+        <title>Alignment issues</title>
+        <para>TBD</para>
+      </sect1>
+      <sect1>
+        <title>Calling into mac80211 from interrupts</title>
+!Pinclude/net/mac80211.h Calling mac80211 from interrupts
+      </sect1>
+      <sect1>
+        <title>functions/definitions</title>
+!Finclude/net/mac80211.h ieee80211_rx_status
+!Finclude/net/mac80211.h mac80211_rx_flags
+!Finclude/net/mac80211.h ieee80211_tx_control
+!Finclude/net/mac80211.h ieee80211_tx_status_flags
+!Finclude/net/mac80211.h ieee80211_rx
+!Finclude/net/mac80211.h ieee80211_rx_irqsafe
+!Finclude/net/mac80211.h ieee80211_tx_status
+!Finclude/net/mac80211.h ieee80211_tx_status_irqsafe
+!Finclude/net/mac80211.h ieee80211_rts_get
+!Finclude/net/mac80211.h ieee80211_rts_duration
+!Finclude/net/mac80211.h ieee80211_ctstoself_get
+!Finclude/net/mac80211.h ieee80211_ctstoself_duration
+!Finclude/net/mac80211.h ieee80211_generic_frame_duration
+!Finclude/net/mac80211.h ieee80211_get_hdrlen_from_skb
+!Finclude/net/mac80211.h ieee80211_get_hdrlen
+!Finclude/net/mac80211.h ieee80211_wake_queue
+!Finclude/net/mac80211.h ieee80211_stop_queue
+!Finclude/net/mac80211.h ieee80211_start_queues
+!Finclude/net/mac80211.h ieee80211_stop_queues
+!Finclude/net/mac80211.h ieee80211_wake_queues
+      </sect1>
+    </chapter>
+
+    <chapter id="filters">
+      <title>Frame filtering</title>
+!Pinclude/net/mac80211.h Frame filtering
+!Finclude/net/mac80211.h ieee80211_filter_flags
+    </chapter>
+  </part>
+
+  <part id="advanced">
+    <title>Advanced driver interface</title>
+    <partintro>
+      <para>
+       Information contained within this part of the book is
+       of interest only for advanced interaction of mac80211
+       with drivers to exploit more hardware capabilities and
+       improve performance.
+      </para>
+    </partintro>
+
+    <chapter id="hardware-crypto-offload">
+      <title>Hardware crypto acceleration</title>
+!Pinclude/net/mac80211.h Hardware crypto acceleration
+<!-- intentionally multiple !F lines to get proper order -->
+!Finclude/net/mac80211.h set_key_cmd
+!Finclude/net/mac80211.h ieee80211_key_conf
+!Finclude/net/mac80211.h ieee80211_key_alg
+!Finclude/net/mac80211.h ieee80211_key_flags
+    </chapter>
+
+    <chapter id="qos">
+      <title>Multiple queues and QoS support</title>
+      <para>TBD</para>
+!Finclude/net/mac80211.h ieee80211_tx_queue_params
+!Finclude/net/mac80211.h ieee80211_tx_queue_stats_data
+!Finclude/net/mac80211.h ieee80211_tx_queue
+    </chapter>
+
+    <chapter id="AP">
+      <title>Access point mode support</title>
+      <para>TBD</para>
+      <para>Some parts of the if_conf should be discussed here instead</para>
+      <para>
+        Insert notes about VLAN interfaces with hw crypto here or
+        in the hw crypto chapter.
+      </para>
+!Finclude/net/mac80211.h ieee80211_get_buffered_bc
+!Finclude/net/mac80211.h ieee80211_beacon_get
+    </chapter>
+
+    <chapter id="multi-iface">
+      <title>Supporting multiple virtual interfaces</title>
+      <para>TBD</para>
+      <para>
+        Note: WDS with identical MAC address should almost always be OK
+      </para>
+      <para>
+        Insert notes about having multiple virtual interfaces with
+        different MAC addresses here, note which configurations are
+        supported by mac80211, add notes about supporting hw crypto
+        with it.
+      </para>
+    </chapter>
+
+    <chapter id="hardware-scan-offload">
+      <title>Hardware scan offload</title>
+      <para>TBD</para>
+!Finclude/net/mac80211.h ieee80211_scan_completed
+    </chapter>
+  </part>
+
+  <part id="rate-control">
+    <title>Rate control interface</title>
+    <partintro>
+      <para>TBD</para>
+      <para>
+       This part of the book describes the rate control algorithm
+       interface and how it relates to mac80211 and drivers.
+      </para>
+    </partintro>
+    <chapter id="dummy">
+      <title>dummy chapter</title>
+      <para>TBD</para>
+    </chapter>
+  </part>
+
+  <part id="internal">
+    <title>Internals</title>
+    <partintro>
+      <para>TBD</para>
+      <para>
+       This part of the book describes mac80211 internals.
+      </para>
+    </partintro>
+
+    <chapter id="key-handling">
+      <title>Key handling</title>
+      <sect1>
+        <title>Key handling basics</title>
+!Pnet/mac80211/key.c Key handling basics
+      </sect1>
+      <sect1>
+        <title>MORE TBD</title>
+        <para>TBD</para>
+      </sect1>
+    </chapter>
+
+    <chapter id="rx-processing">
+      <title>Receive processing</title>
+      <para>TBD</para>
+    </chapter>
+
+    <chapter id="tx-processing">
+      <title>Transmit processing</title>
+      <para>TBD</para>
+    </chapter>
+
+    <chapter id="sta-info">
+      <title>Station info handling</title>
+      <sect1>
+        <title>Programming information</title>
+!Fnet/mac80211/sta_info.h sta_info
+!Fnet/mac80211/sta_info.h ieee80211_sta_info_flags
+      </sect1>
+      <sect1>
+        <title>STA information lifetime rules</title>
+!Pnet/mac80211/sta_info.c STA information lifetime rules
+      </sect1>
+    </chapter>
+
+    <chapter id="synchronisation">
+      <title>Synchronisation</title>
+      <para>TBD</para>
+      <para>Locking, lots of RCU</para>
+    </chapter>
+  </part>
+</book>
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 6aca472d7a02..d002b1c6e78e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -407,7 +407,6 @@ enum ieee80211_conf_flags {
  * @channel: the channel to tune to
  */
 struct ieee80211_conf {
-	unsigned int regulatory_domain;
 	int radio_enabled;
 
 	int beacon_int;
@@ -437,6 +436,7 @@ struct ieee80211_conf {
  * @IEEE80211_IF_TYPE_WDS: interface in WDS mode.
  * @IEEE80211_IF_TYPE_VLAN: VLAN interface bound to an AP, drivers
  *	will never see this type.
+ * @IEEE80211_IF_TYPE_MESH_POINT: 802.11s mesh point
  */
 enum ieee80211_if_types {
 	IEEE80211_IF_TYPE_INVALID,
@@ -1096,8 +1096,9 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len,
 /**
  * ieee80211_register_hw - Register hardware device
  *
- * You must call this function before any other functions
- * except ieee80211_register_hwmode.
+ * You must call this function before any other functions in
+ * mac80211. Note that before a hardware can be registered, you
+ * need to fill the contained wiphy's information.
  *
  * @hw: the device to register as returned by ieee80211_alloc_hw()
  */
diff --git a/net/mac80211/key.c b/net/mac80211/key.c
index df0c04cedbe4..166d0f00d135 100644
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -20,8 +20,8 @@
 #include "aes_ccm.h"
 
 
-/*
- * Key handling basics
+/**
+ * DOC: Key handling basics
  *
  * Key handling in mac80211 is done based on per-interface (sub_if_data)
  * keys and per-station keys. Since each station belongs to an interface,
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index a767042ec4fd..42414b441592 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -31,10 +31,10 @@
  * for faster lookup and a list for iteration. They are managed using
  * RCU, i.e. access to the list and hash table is protected by RCU.
  *
- * Upon allocating a STA info structure with @sta_info_alloc() or
+ * Upon allocating a STA info structure with sta_info_alloc() or
  * mesh_plink_alloc(), the caller owns that structure. It must then either
- * destroy it using @sta_info_destroy() (which is pretty useless) or insert
- * it into the hash table using @sta_info_insert() which demotes the reference
+ * destroy it using sta_info_destroy() (which is pretty useless) or insert
+ * it into the hash table using sta_info_insert() which demotes the reference
  * from ownership to a regular RCU-protected reference; if the function
  * is called without protection by an RCU critical section the reference
  * is instantly invalidated.
@@ -42,19 +42,19 @@
  * Because there are debugfs entries for each station, and adding those
  * must be able to sleep, it is also possible to "pin" a station entry,
  * that means it can be removed from the hash table but not be freed.
- * See the comment in @__sta_info_unlink() for more information.
+ * See the comment in __sta_info_unlink() for more information.
  *
  * In order to remove a STA info structure, the caller needs to first
- * unlink it (@sta_info_unlink()) from the list and hash tables and
+ * unlink it (sta_info_unlink()) from the list and hash tables and
  * then wait for an RCU synchronisation before it can be freed. Due to
  * the pinning and the possibility of multiple callers trying to remove
- * the same STA info at the same time, @sta_info_unlink() can clear the
+ * the same STA info at the same time, sta_info_unlink() can clear the
  * STA info pointer it is passed to indicate that the STA info is owned
  * by somebody else now.
  *
- * If @sta_info_unlink() did not clear the pointer then the caller owns
+ * If sta_info_unlink() did not clear the pointer then the caller owns
  * the STA info structure now and is responsible of destroying it with
- * a call to @sta_info_destroy(), not before RCU synchronisation, of
+ * a call to sta_info_destroy(), not before RCU synchronisation, of
  * course. Note that sta_info_destroy() must be protected by the RTNL.
  *
  * In all other cases, there is no concept of ownership on a STA entry,
-- 
cgit v1.2.3


From aab547ce0d1493d400b6468c521a0137cd8c1edf Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Fri, 29 Feb 2008 11:36:12 +0100
Subject: ssb: Add Gigabit Ethernet driver

This adds the Gigabit Ethernet driver for the SSB
Gigabit Ethernet core. This driver actually is a frontend to
the Tigon3 driver. So the real work is done by tg3.
This device is used in the Linksys WRT350N.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/Kconfig                 |   9 ++
 drivers/ssb/Makefile                |   1 +
 drivers/ssb/driver_gige.c           | 294 ++++++++++++++++++++++++++++++++++++
 drivers/ssb/driver_mipscore.c       |   1 +
 drivers/ssb/driver_pcicore.c        | 160 +++++++++++---------
 drivers/ssb/embedded.c              |  90 +++++++++++
 drivers/ssb/main.c                  |  30 +++-
 drivers/ssb/ssb_private.h           |   2 +
 include/linux/ssb/ssb.h             |   7 +
 include/linux/ssb/ssb_driver_gige.h | 174 +++++++++++++++++++++
 include/linux/ssb/ssb_driver_pci.h  |  19 +++
 11 files changed, 715 insertions(+), 72 deletions(-)
 create mode 100644 drivers/ssb/driver_gige.c
 create mode 100644 include/linux/ssb/ssb_driver_gige.h

(limited to 'include')

diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig
index adea792fb675..f69ef0ba2613 100644
--- a/drivers/ssb/Kconfig
+++ b/drivers/ssb/Kconfig
@@ -125,4 +125,13 @@ config SSB_DRIVER_EXTIF
 
 	  If unsure, say N
 
+config SSB_DRIVER_GIGE
+	bool "SSB Broadcom Gigabit Ethernet driver"
+	depends on SSB_PCIHOST_POSSIBLE && SSB_EMBEDDED && MIPS
+	help
+	  Driver for the Sonics Silicon Backplane attached
+	  Broadcom Gigabit Ethernet.
+
+	  If unsure, say N
+
 endmenu
diff --git a/drivers/ssb/Makefile b/drivers/ssb/Makefile
index de94c2eb7a37..910f35e32fc9 100644
--- a/drivers/ssb/Makefile
+++ b/drivers/ssb/Makefile
@@ -11,6 +11,7 @@ ssb-y					+= driver_chipcommon.o
 ssb-$(CONFIG_SSB_DRIVER_MIPS)		+= driver_mipscore.o
 ssb-$(CONFIG_SSB_DRIVER_EXTIF)		+= driver_extif.o
 ssb-$(CONFIG_SSB_DRIVER_PCICORE)	+= driver_pcicore.o
+ssb-$(CONFIG_SSB_DRIVER_GIGE)		+= driver_gige.o
 
 # b43 pci-ssb-bridge driver
 # Not strictly a part of SSB, but kept here for convenience
diff --git a/drivers/ssb/driver_gige.c b/drivers/ssb/driver_gige.c
new file mode 100644
index 000000000000..172f90407b93
--- /dev/null
+++ b/drivers/ssb/driver_gige.c
@@ -0,0 +1,294 @@
+/*
+ * Sonics Silicon Backplane
+ * Broadcom Gigabit Ethernet core driver
+ *
+ * Copyright 2008, Broadcom Corporation
+ * Copyright 2008, Michael Buesch <mb@bu3sch.de>
+ *
+ * Licensed under the GNU/GPL. See COPYING for details.
+ */
+
+#include <linux/ssb/ssb.h>
+#include <linux/ssb/ssb_driver_gige.h>
+#include <linux/pci.h>
+#include <linux/pci_regs.h>
+
+
+/*
+MODULE_DESCRIPTION("SSB Broadcom Gigabit Ethernet driver");
+MODULE_AUTHOR("Michael Buesch");
+MODULE_LICENSE("GPL");
+*/
+
+static const struct ssb_device_id ssb_gige_tbl[] = {
+	SSB_DEVICE(SSB_VENDOR_BROADCOM, SSB_DEV_ETHERNET_GBIT, SSB_ANY_REV),
+	SSB_DEVTABLE_END
+};
+/* MODULE_DEVICE_TABLE(ssb, ssb_gige_tbl); */
+
+
+static inline u8 gige_read8(struct ssb_gige *dev, u16 offset)
+{
+	return ssb_read8(dev->dev, offset);
+}
+
+static inline u16 gige_read16(struct ssb_gige *dev, u16 offset)
+{
+	return ssb_read16(dev->dev, offset);
+}
+
+static inline u32 gige_read32(struct ssb_gige *dev, u16 offset)
+{
+	return ssb_read32(dev->dev, offset);
+}
+
+static inline void gige_write8(struct ssb_gige *dev,
+			       u16 offset, u8 value)
+{
+	ssb_write8(dev->dev, offset, value);
+}
+
+static inline void gige_write16(struct ssb_gige *dev,
+				u16 offset, u16 value)
+{
+	ssb_write16(dev->dev, offset, value);
+}
+
+static inline void gige_write32(struct ssb_gige *dev,
+				u16 offset, u32 value)
+{
+	ssb_write32(dev->dev, offset, value);
+}
+
+static inline
+u8 gige_pcicfg_read8(struct ssb_gige *dev, unsigned int offset)
+{
+	BUG_ON(offset >= 256);
+	return gige_read8(dev, SSB_GIGE_PCICFG + offset);
+}
+
+static inline
+u16 gige_pcicfg_read16(struct ssb_gige *dev, unsigned int offset)
+{
+	BUG_ON(offset >= 256);
+	return gige_read16(dev, SSB_GIGE_PCICFG + offset);
+}
+
+static inline
+u32 gige_pcicfg_read32(struct ssb_gige *dev, unsigned int offset)
+{
+	BUG_ON(offset >= 256);
+	return gige_read32(dev, SSB_GIGE_PCICFG + offset);
+}
+
+static inline
+void gige_pcicfg_write8(struct ssb_gige *dev,
+			unsigned int offset, u8 value)
+{
+	BUG_ON(offset >= 256);
+	gige_write8(dev, SSB_GIGE_PCICFG + offset, value);
+}
+
+static inline
+void gige_pcicfg_write16(struct ssb_gige *dev,
+			 unsigned int offset, u16 value)
+{
+	BUG_ON(offset >= 256);
+	gige_write16(dev, SSB_GIGE_PCICFG + offset, value);
+}
+
+static inline
+void gige_pcicfg_write32(struct ssb_gige *dev,
+			 unsigned int offset, u32 value)
+{
+	BUG_ON(offset >= 256);
+	gige_write32(dev, SSB_GIGE_PCICFG + offset, value);
+}
+
+static int ssb_gige_pci_read_config(struct pci_bus *bus, unsigned int devfn,
+				    int reg, int size, u32 *val)
+{
+	struct ssb_gige *dev = container_of(bus->ops, struct ssb_gige, pci_ops);
+	unsigned long flags;
+
+	if ((PCI_SLOT(devfn) > 0) || (PCI_FUNC(devfn) > 0))
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	if (reg >= 256)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	switch (size) {
+	case 1:
+		*val = gige_pcicfg_read8(dev, reg);
+		break;
+	case 2:
+		*val = gige_pcicfg_read16(dev, reg);
+		break;
+	case 4:
+		*val = gige_pcicfg_read32(dev, reg);
+		break;
+	default:
+		WARN_ON(1);
+	}
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int ssb_gige_pci_write_config(struct pci_bus *bus, unsigned int devfn,
+				     int reg, int size, u32 val)
+{
+	struct ssb_gige *dev = container_of(bus->ops, struct ssb_gige, pci_ops);
+	unsigned long flags;
+
+	if ((PCI_SLOT(devfn) > 0) || (PCI_FUNC(devfn) > 0))
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	if (reg >= 256)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	switch (size) {
+	case 1:
+		gige_pcicfg_write8(dev, reg, val);
+		break;
+	case 2:
+		gige_pcicfg_write16(dev, reg, val);
+		break;
+	case 4:
+		gige_pcicfg_write32(dev, reg, val);
+		break;
+	default:
+		WARN_ON(1);
+	}
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int ssb_gige_probe(struct ssb_device *sdev, const struct ssb_device_id *id)
+{
+	struct ssb_gige *dev;
+	u32 base, tmslow, tmshigh;
+
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+	dev->dev = sdev;
+
+	spin_lock_init(&dev->lock);
+	dev->pci_controller.pci_ops = &dev->pci_ops;
+	dev->pci_controller.io_resource = &dev->io_resource;
+	dev->pci_controller.mem_resource = &dev->mem_resource;
+	dev->pci_controller.io_map_base = 0x800;
+	dev->pci_ops.read = ssb_gige_pci_read_config;
+	dev->pci_ops.write = ssb_gige_pci_write_config;
+
+	dev->io_resource.name = SSB_GIGE_IO_RES_NAME;
+	dev->io_resource.start = 0x800;
+	dev->io_resource.end = 0x8FF;
+	dev->io_resource.flags = IORESOURCE_IO | IORESOURCE_PCI_FIXED;
+
+	if (!ssb_device_is_enabled(sdev))
+		ssb_device_enable(sdev, 0);
+
+	/* Setup BAR0. This is a 64k MMIO region. */
+	base = ssb_admatch_base(ssb_read32(sdev, SSB_ADMATCH1));
+	gige_pcicfg_write32(dev, PCI_BASE_ADDRESS_0, base);
+	gige_pcicfg_write32(dev, PCI_BASE_ADDRESS_1, 0);
+
+	dev->mem_resource.name = SSB_GIGE_MEM_RES_NAME;
+	dev->mem_resource.start = base;
+	dev->mem_resource.end = base + 0x10000 - 1;
+	dev->mem_resource.flags = IORESOURCE_MEM | IORESOURCE_PCI_FIXED;
+
+	/* Enable the memory region. */
+	gige_pcicfg_write16(dev, PCI_COMMAND,
+			    gige_pcicfg_read16(dev, PCI_COMMAND)
+			    | PCI_COMMAND_MEMORY);
+
+	/* Write flushing is controlled by the Flush Status Control register.
+	 * We want to flush every register write with a timeout and we want
+	 * to disable the IRQ mask while flushing to avoid concurrency.
+	 * Note that automatic write flushing does _not_ work from
+	 * an IRQ handler. The driver must flush manually by reading a register.
+	 */
+	gige_write32(dev, SSB_GIGE_SHIM_FLUSHSTAT, 0x00000068);
+
+	/* Check if we have an RGMII or GMII PHY-bus.
+	 * On RGMII do not bypass the DLLs */
+	tmslow = ssb_read32(sdev, SSB_TMSLOW);
+	tmshigh = ssb_read32(sdev, SSB_TMSHIGH);
+	if (tmshigh & SSB_GIGE_TMSHIGH_RGMII) {
+		tmslow &= ~SSB_GIGE_TMSLOW_TXBYPASS;
+		tmslow &= ~SSB_GIGE_TMSLOW_RXBYPASS;
+		dev->has_rgmii = 1;
+	} else {
+		tmslow |= SSB_GIGE_TMSLOW_TXBYPASS;
+		tmslow |= SSB_GIGE_TMSLOW_RXBYPASS;
+		dev->has_rgmii = 0;
+	}
+	tmslow |= SSB_GIGE_TMSLOW_DLLEN;
+	ssb_write32(sdev, SSB_TMSLOW, tmslow);
+
+	ssb_set_drvdata(sdev, dev);
+	register_pci_controller(&dev->pci_controller);
+
+	return 0;
+}
+
+bool pdev_is_ssb_gige_core(struct pci_dev *pdev)
+{
+	if (!pdev->resource[0].name)
+		return 0;
+	return (strcmp(pdev->resource[0].name, SSB_GIGE_MEM_RES_NAME) == 0);
+}
+EXPORT_SYMBOL(pdev_is_ssb_gige_core);
+
+int ssb_gige_pcibios_plat_dev_init(struct ssb_device *sdev,
+				   struct pci_dev *pdev)
+{
+	struct ssb_gige *dev = ssb_get_drvdata(sdev);
+	struct resource *res;
+
+	if (pdev->bus->ops != &dev->pci_ops) {
+		/* The PCI device is not on this SSB GigE bridge device. */
+		return -ENODEV;
+	}
+
+	/* Fixup the PCI resources. */
+	res = &(pdev->resource[0]);
+	res->flags = IORESOURCE_MEM | IORESOURCE_PCI_FIXED;
+	res->name = dev->mem_resource.name;
+	res->start = dev->mem_resource.start;
+	res->end = dev->mem_resource.end;
+
+	/* Fixup interrupt lines. */
+	pdev->irq = ssb_mips_irq(sdev) + 2;
+	pci_write_config_byte(pdev, PCI_INTERRUPT_LINE, pdev->irq);
+
+	return 0;
+}
+
+int ssb_gige_map_irq(struct ssb_device *sdev,
+		     const struct pci_dev *pdev)
+{
+	struct ssb_gige *dev = ssb_get_drvdata(sdev);
+
+	if (pdev->bus->ops != &dev->pci_ops) {
+		/* The PCI device is not on this SSB GigE bridge device. */
+		return -ENODEV;
+	}
+
+	return ssb_mips_irq(sdev) + 2;
+}
+
+static struct ssb_driver ssb_gige_driver = {
+	.name		= "BCM-GigE",
+	.id_table	= ssb_gige_tbl,
+	.probe		= ssb_gige_probe,
+};
+
+int ssb_gige_init(void)
+{
+	return ssb_driver_register(&ssb_gige_driver);
+}
diff --git a/drivers/ssb/driver_mipscore.c b/drivers/ssb/driver_mipscore.c
index 3d3dd32bf3ab..e3fad3123ecb 100644
--- a/drivers/ssb/driver_mipscore.c
+++ b/drivers/ssb/driver_mipscore.c
@@ -209,6 +209,7 @@ void ssb_mipscore_init(struct ssb_mipscore *mcore)
 			/* fallthrough */
 		case SSB_DEV_PCI:
 		case SSB_DEV_ETHERNET:
+		case SSB_DEV_ETHERNET_GBIT:
 		case SSB_DEV_80211:
 		case SSB_DEV_USB20_HOST:
 			/* These devices get their own IRQ line if available, the rest goes on IRQ0 */
diff --git a/drivers/ssb/driver_pcicore.c b/drivers/ssb/driver_pcicore.c
index 74b9a8aea52b..33a7d5620474 100644
--- a/drivers/ssb/driver_pcicore.c
+++ b/drivers/ssb/driver_pcicore.c
@@ -60,77 +60,6 @@ static DEFINE_SPINLOCK(cfgspace_lock);
 /* Core to access the external PCI config space. Can only have one. */
 static struct ssb_pcicore *extpci_core;
 
-static u32 ssb_pcicore_pcibus_iobase = 0x100;
-static u32 ssb_pcicore_pcibus_membase = SSB_PCI_DMA;
-
-int pcibios_plat_dev_init(struct pci_dev *d)
-{
-	struct resource *res;
-	int pos, size;
-	u32 *base;
-
-	ssb_printk(KERN_INFO "PCI: Fixing up device %s\n",
-		   pci_name(d));
-
-	/* Fix up resource bases */
-	for (pos = 0; pos < 6; pos++) {
-		res = &d->resource[pos];
-		if (res->flags & IORESOURCE_IO)
-			base = &ssb_pcicore_pcibus_iobase;
-		else
-			base = &ssb_pcicore_pcibus_membase;
-		res->flags |= IORESOURCE_PCI_FIXED;
-		if (res->end) {
-			size = res->end - res->start + 1;
-			if (*base & (size - 1))
-				*base = (*base + size) & ~(size - 1);
-			res->start = *base;
-			res->end = res->start + size - 1;
-			*base += size;
-			pci_write_config_dword(d, PCI_BASE_ADDRESS_0 + (pos << 2), res->start);
-		}
-		/* Fix up PCI bridge BAR0 only */
-		if (d->bus->number == 0 && PCI_SLOT(d->devfn) == 0)
-			break;
-	}
-	/* Fix up interrupt lines */
-	d->irq = ssb_mips_irq(extpci_core->dev) + 2;
-	pci_write_config_byte(d, PCI_INTERRUPT_LINE, d->irq);
-
-	return 0;
-}
-
-static void __init ssb_fixup_pcibridge(struct pci_dev *dev)
-{
-	u8 lat;
-
-	if (dev->bus->number != 0 || PCI_SLOT(dev->devfn) != 0)
-		return;
-
-	ssb_printk(KERN_INFO "PCI: Fixing up bridge %s\n", pci_name(dev));
-
-	/* Enable PCI bridge bus mastering and memory space */
-	pci_set_master(dev);
-	if (pcibios_enable_device(dev, ~0) < 0) {
-		ssb_printk(KERN_ERR "PCI: SSB bridge enable failed\n");
-		return;
-	}
-
-	/* Enable PCI bridge BAR1 prefetch and burst */
-	pci_write_config_dword(dev, SSB_BAR1_CONTROL, 3);
-
-	/* Make sure our latency is high enough to handle the devices behind us */
-	lat = 168;
-	ssb_printk(KERN_INFO "PCI: Fixing latency timer of device %s to %u\n",
-		   pci_name(dev), lat);
-	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
-}
-DECLARE_PCI_FIXUP_EARLY(PCI_ANY_ID, PCI_ANY_ID, ssb_fixup_pcibridge);
-
-int __init pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
-{
-	return ssb_mips_irq(extpci_core->dev) + 2;
-}
 
 static u32 get_cfgspace_addr(struct ssb_pcicore *pc,
 			     unsigned int bus, unsigned int dev,
@@ -320,6 +249,95 @@ static struct pci_controller ssb_pcicore_controller = {
 	.mem_offset	= 0x24000000,
 };
 
+static u32 ssb_pcicore_pcibus_iobase = 0x100;
+static u32 ssb_pcicore_pcibus_membase = SSB_PCI_DMA;
+
+/* This function is called when doing a pci_enable_device().
+ * We must first check if the device is a device on the PCI-core bridge. */
+int ssb_pcicore_plat_dev_init(struct pci_dev *d)
+{
+	struct resource *res;
+	int pos, size;
+	u32 *base;
+
+	if (d->bus->ops != &ssb_pcicore_pciops) {
+		/* This is not a device on the PCI-core bridge. */
+		return -ENODEV;
+	}
+
+	ssb_printk(KERN_INFO "PCI: Fixing up device %s\n",
+		   pci_name(d));
+
+	/* Fix up resource bases */
+	for (pos = 0; pos < 6; pos++) {
+		res = &d->resource[pos];
+		if (res->flags & IORESOURCE_IO)
+			base = &ssb_pcicore_pcibus_iobase;
+		else
+			base = &ssb_pcicore_pcibus_membase;
+		res->flags |= IORESOURCE_PCI_FIXED;
+		if (res->end) {
+			size = res->end - res->start + 1;
+			if (*base & (size - 1))
+				*base = (*base + size) & ~(size - 1);
+			res->start = *base;
+			res->end = res->start + size - 1;
+			*base += size;
+			pci_write_config_dword(d, PCI_BASE_ADDRESS_0 + (pos << 2), res->start);
+		}
+		/* Fix up PCI bridge BAR0 only */
+		if (d->bus->number == 0 && PCI_SLOT(d->devfn) == 0)
+			break;
+	}
+	/* Fix up interrupt lines */
+	d->irq = ssb_mips_irq(extpci_core->dev) + 2;
+	pci_write_config_byte(d, PCI_INTERRUPT_LINE, d->irq);
+
+	return 0;
+}
+
+/* Early PCI fixup for a device on the PCI-core bridge. */
+static void ssb_pcicore_fixup_pcibridge(struct pci_dev *dev)
+{
+	u8 lat;
+
+	if (dev->bus->ops != &ssb_pcicore_pciops) {
+		/* This is not a device on the PCI-core bridge. */
+		return;
+	}
+	if (dev->bus->number != 0 || PCI_SLOT(dev->devfn) != 0)
+		return;
+
+	ssb_printk(KERN_INFO "PCI: Fixing up bridge %s\n", pci_name(dev));
+
+	/* Enable PCI bridge bus mastering and memory space */
+	pci_set_master(dev);
+	if (pcibios_enable_device(dev, ~0) < 0) {
+		ssb_printk(KERN_ERR "PCI: SSB bridge enable failed\n");
+		return;
+	}
+
+	/* Enable PCI bridge BAR1 prefetch and burst */
+	pci_write_config_dword(dev, SSB_BAR1_CONTROL, 3);
+
+	/* Make sure our latency is high enough to handle the devices behind us */
+	lat = 168;
+	ssb_printk(KERN_INFO "PCI: Fixing latency timer of device %s to %u\n",
+		   pci_name(dev), lat);
+	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
+}
+DECLARE_PCI_FIXUP_EARLY(PCI_ANY_ID, PCI_ANY_ID, ssb_pcicore_fixup_pcibridge);
+
+/* PCI device IRQ mapping. */
+int ssb_pcicore_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
+{
+	if (dev->bus->ops != &ssb_pcicore_pciops) {
+		/* This is not a device on the PCI-core bridge. */
+		return -ENODEV;
+	}
+	return ssb_mips_irq(extpci_core->dev) + 2;
+}
+
 static void ssb_pcicore_init_hostmode(struct ssb_pcicore *pc)
 {
 	u32 val;
diff --git a/drivers/ssb/embedded.c b/drivers/ssb/embedded.c
index d3ade821555c..7dc3a6b41397 100644
--- a/drivers/ssb/embedded.c
+++ b/drivers/ssb/embedded.c
@@ -10,6 +10,9 @@
 
 #include <linux/ssb/ssb.h>
 #include <linux/ssb/ssb_embedded.h>
+#include <linux/ssb/ssb_driver_pci.h>
+#include <linux/ssb/ssb_driver_gige.h>
+#include <linux/pci.h>
 
 #include "ssb_private.h"
 
@@ -130,3 +133,90 @@ u32 ssb_gpio_polarity(struct ssb_bus *bus, u32 mask, u32 value)
 	return res;
 }
 EXPORT_SYMBOL(ssb_gpio_polarity);
+
+#ifdef CONFIG_SSB_DRIVER_GIGE
+static int gige_pci_init_callback(struct ssb_bus *bus, unsigned long data)
+{
+	struct pci_dev *pdev = (struct pci_dev *)data;
+	struct ssb_device *dev;
+	unsigned int i;
+	int res;
+
+	for (i = 0; i < bus->nr_devices; i++) {
+		dev = &(bus->devices[i]);
+		if (dev->id.coreid != SSB_DEV_ETHERNET_GBIT)
+			continue;
+		if (!dev->dev ||
+		    !dev->dev->driver ||
+		    !device_is_registered(dev->dev))
+			continue;
+		res = ssb_gige_pcibios_plat_dev_init(dev, pdev);
+		if (res >= 0)
+			return res;
+	}
+
+	return -ENODEV;
+}
+#endif /* CONFIG_SSB_DRIVER_GIGE */
+
+int ssb_pcibios_plat_dev_init(struct pci_dev *dev)
+{
+	int err;
+
+	err = ssb_pcicore_plat_dev_init(dev);
+	if (!err)
+		return 0;
+#ifdef CONFIG_SSB_DRIVER_GIGE
+	err = ssb_for_each_bus_call((unsigned long)dev, gige_pci_init_callback);
+	if (err >= 0)
+		return err;
+#endif
+	/* This is not a PCI device on any SSB device. */
+
+	return -ENODEV;
+}
+
+#ifdef CONFIG_SSB_DRIVER_GIGE
+static int gige_map_irq_callback(struct ssb_bus *bus, unsigned long data)
+{
+	const struct pci_dev *pdev = (const struct pci_dev *)data;
+	struct ssb_device *dev;
+	unsigned int i;
+	int res;
+
+	for (i = 0; i < bus->nr_devices; i++) {
+		dev = &(bus->devices[i]);
+		if (dev->id.coreid != SSB_DEV_ETHERNET_GBIT)
+			continue;
+		if (!dev->dev ||
+		    !dev->dev->driver ||
+		    !device_is_registered(dev->dev))
+			continue;
+		res = ssb_gige_map_irq(dev, pdev);
+		if (res >= 0)
+			return res;
+	}
+
+	return -ENODEV;
+}
+#endif /* CONFIG_SSB_DRIVER_GIGE */
+
+int ssb_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int res;
+
+	/* Check if this PCI device is a device on a SSB bus or device
+	 * and return the IRQ number for it. */
+
+	res = ssb_pcicore_pcibios_map_irq(dev, slot, pin);
+	if (res >= 0)
+		return res;
+#ifdef CONFIG_SSB_DRIVER_GIGE
+	res = ssb_for_each_bus_call((unsigned long)dev, gige_map_irq_callback);
+	if (res >= 0)
+		return res;
+#endif
+	/* This is not a PCI device on any SSB device. */
+
+	return -ENODEV;
+}
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index 8db40c4b86e9..49d7bbb9bea7 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -14,6 +14,7 @@
 #include <linux/io.h>
 #include <linux/ssb/ssb.h>
 #include <linux/ssb/ssb_regs.h>
+#include <linux/ssb/ssb_driver_gige.h>
 #include <linux/dma-mapping.h>
 #include <linux/pci.h>
 
@@ -68,6 +69,25 @@ found:
 }
 #endif /* CONFIG_SSB_PCIHOST */
 
+int ssb_for_each_bus_call(unsigned long data,
+			  int (*func)(struct ssb_bus *bus, unsigned long data))
+{
+	struct ssb_bus *bus;
+	int res;
+
+	ssb_buses_lock();
+	list_for_each_entry(bus, &buses, list) {
+		res = func(bus, data);
+		if (res >= 0) {
+			ssb_buses_unlock();
+			return res;
+		}
+	}
+	ssb_buses_unlock();
+
+	return -ENODEV;
+}
+
 static struct ssb_device *ssb_device_get(struct ssb_device *dev)
 {
 	if (dev)
@@ -1171,7 +1191,14 @@ static int __init ssb_modinit(void)
 	err = b43_pci_ssb_bridge_init();
 	if (err) {
 		ssb_printk(KERN_ERR "Broadcom 43xx PCI-SSB-bridge "
-			   "initialization failed");
+			   "initialization failed\n");
+		/* don't fail SSB init because of this */
+		err = 0;
+	}
+	err = ssb_gige_init();
+	if (err) {
+		ssb_printk(KERN_ERR "SSB Broadcom Gigabit Ethernet "
+			   "driver initialization failed\n");
 		/* don't fail SSB init because of this */
 		err = 0;
 	}
@@ -1185,6 +1212,7 @@ fs_initcall(ssb_modinit);
 
 static void __exit ssb_modexit(void)
 {
+	ssb_gige_exit();
 	b43_pci_ssb_bridge_exit();
 	bus_unregister(&ssb_bustype);
 }
diff --git a/drivers/ssb/ssb_private.h b/drivers/ssb/ssb_private.h
index 21eca2b5118b..d03b20983b1e 100644
--- a/drivers/ssb/ssb_private.h
+++ b/drivers/ssb/ssb_private.h
@@ -118,6 +118,8 @@ extern u32 ssb_calc_clock_rate(u32 plltype, u32 n, u32 m);
 extern int ssb_devices_freeze(struct ssb_bus *bus);
 extern int ssb_devices_thaw(struct ssb_bus *bus);
 extern struct ssb_bus *ssb_pci_dev_to_bus(struct pci_dev *pdev);
+int ssb_for_each_bus_call(unsigned long data,
+			  int (*func)(struct ssb_bus *bus, unsigned long data));
 
 /* b43_pci_bridge.c */
 #ifdef CONFIG_SSB_B43_PCI_BRIDGE
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 860d28c6d149..b7c388972fcf 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -422,5 +422,12 @@ extern int ssb_bus_powerup(struct ssb_bus *bus, bool dynamic_pctl);
 extern u32 ssb_admatch_base(u32 adm);
 extern u32 ssb_admatch_size(u32 adm);
 
+/* PCI device mapping and fixup routines.
+ * Called from the architecture pcibios init code.
+ * These are only available on SSB_EMBEDDED configurations. */
+#ifdef CONFIG_SSB_EMBEDDED
+int ssb_pcibios_plat_dev_init(struct pci_dev *dev);
+int ssb_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin);
+#endif /* CONFIG_SSB_EMBEDDED */
 
 #endif /* LINUX_SSB_H_ */
diff --git a/include/linux/ssb/ssb_driver_gige.h b/include/linux/ssb/ssb_driver_gige.h
new file mode 100644
index 000000000000..01fbdf5fef22
--- /dev/null
+++ b/include/linux/ssb/ssb_driver_gige.h
@@ -0,0 +1,174 @@
+#ifndef LINUX_SSB_DRIVER_GIGE_H_
+#define LINUX_SSB_DRIVER_GIGE_H_
+
+#include <linux/ssb/ssb.h>
+#include <linux/pci.h>
+#include <linux/spinlock.h>
+
+
+#ifdef CONFIG_SSB_DRIVER_GIGE
+
+
+#define SSB_GIGE_PCIIO			0x0000 /* PCI I/O Registers (1024 bytes) */
+#define SSB_GIGE_RESERVED		0x0400 /* Reserved (1024 bytes) */
+#define SSB_GIGE_PCICFG			0x0800 /* PCI config space (256 bytes) */
+#define SSB_GIGE_SHIM_FLUSHSTAT		0x0C00 /* PCI to OCP: Flush status control (32bit) */
+#define SSB_GIGE_SHIM_FLUSHRDA		0x0C04 /* PCI to OCP: Flush read address (32bit) */
+#define SSB_GIGE_SHIM_FLUSHTO		0x0C08 /* PCI to OCP: Flush timeout counter (32bit) */
+#define SSB_GIGE_SHIM_BARRIER		0x0C0C /* PCI to OCP: Barrier register (32bit) */
+#define SSB_GIGE_SHIM_MAOCPSI		0x0C10 /* PCI to OCP: MaocpSI Control (32bit) */
+#define SSB_GIGE_SHIM_SIOCPMA		0x0C14 /* PCI to OCP: SiocpMa Control (32bit) */
+
+/* TM Status High flags */
+#define SSB_GIGE_TMSHIGH_RGMII		0x00010000 /* Have an RGMII PHY-bus */
+/* TM Status Low flags */
+#define SSB_GIGE_TMSLOW_TXBYPASS	0x00080000 /* TX bypass (no delay) */
+#define SSB_GIGE_TMSLOW_RXBYPASS	0x00100000 /* RX bypass (no delay) */
+#define SSB_GIGE_TMSLOW_DLLEN		0x01000000 /* Enable DLL controls */
+
+/* Boardflags (low) */
+#define SSB_GIGE_BFL_ROBOSWITCH		0x0010
+
+
+#define SSB_GIGE_MEM_RES_NAME		"SSB Broadcom 47xx GigE memory"
+#define SSB_GIGE_IO_RES_NAME		"SSB Broadcom 47xx GigE I/O"
+
+struct ssb_gige {
+	struct ssb_device *dev;
+
+	spinlock_t lock;
+
+	/* True, if the device has an RGMII bus.
+	 * False, if the device has a GMII bus. */
+	bool has_rgmii;
+
+	/* The PCI controller device. */
+	struct pci_controller pci_controller;
+	struct pci_ops pci_ops;
+	struct resource mem_resource;
+	struct resource io_resource;
+};
+
+/* Check whether a PCI device is a SSB Gigabit Ethernet core. */
+extern bool pdev_is_ssb_gige_core(struct pci_dev *pdev);
+
+/* Convert a pci_dev pointer to a ssb_gige pointer. */
+static inline struct ssb_gige * pdev_to_ssb_gige(struct pci_dev *pdev)
+{
+	if (!pdev_is_ssb_gige_core(pdev))
+		return NULL;
+	return container_of(pdev->bus->ops, struct ssb_gige, pci_ops);
+}
+
+/* Returns whether the PHY is connected by an RGMII bus. */
+static inline bool ssb_gige_is_rgmii(struct pci_dev *pdev)
+{
+	struct ssb_gige *dev = pdev_to_ssb_gige(pdev);
+	return (dev ? dev->has_rgmii : 0);
+}
+
+/* Returns whether we have a Roboswitch. */
+static inline bool ssb_gige_have_roboswitch(struct pci_dev *pdev)
+{
+	struct ssb_gige *dev = pdev_to_ssb_gige(pdev);
+	if (dev)
+		return !!(dev->dev->bus->sprom.boardflags_lo &
+			  SSB_GIGE_BFL_ROBOSWITCH);
+	return 0;
+}
+
+/* Returns whether we can only do one DMA at once. */
+static inline bool ssb_gige_one_dma_at_once(struct pci_dev *pdev)
+{
+	struct ssb_gige *dev = pdev_to_ssb_gige(pdev);
+	if (dev)
+		return ((dev->dev->bus->chip_id == 0x4785) &&
+			(dev->dev->bus->chip_rev < 2));
+	return 0;
+}
+
+/* Returns whether we must flush posted writes. */
+static inline bool ssb_gige_must_flush_posted_writes(struct pci_dev *pdev)
+{
+	struct ssb_gige *dev = pdev_to_ssb_gige(pdev);
+	if (dev)
+		return (dev->dev->bus->chip_id == 0x4785);
+	return 0;
+}
+
+extern char * nvram_get(const char *name);
+/* Get the device MAC address */
+static inline void ssb_gige_get_macaddr(struct pci_dev *pdev, u8 *macaddr)
+{
+#ifdef CONFIG_BCM947XX
+	char *res = nvram_get("et0macaddr");
+	if (res)
+		memcpy(macaddr, res, 6);
+#endif
+}
+
+extern int ssb_gige_pcibios_plat_dev_init(struct ssb_device *sdev,
+					  struct pci_dev *pdev);
+extern int ssb_gige_map_irq(struct ssb_device *sdev,
+			    const struct pci_dev *pdev);
+
+/* The GigE driver is not a standalone module, because we don't have support
+ * for unregistering the driver. So we could not unload the module anyway. */
+extern int ssb_gige_init(void);
+static inline void ssb_gige_exit(void)
+{
+	/* Currently we can not unregister the GigE driver,
+	 * because we can not unregister the PCI bridge. */
+	BUG();
+}
+
+
+#else /* CONFIG_SSB_DRIVER_GIGE */
+/* Gigabit Ethernet driver disabled */
+
+
+static inline int ssb_gige_pcibios_plat_dev_init(struct ssb_device *sdev,
+						 struct pci_dev *pdev)
+{
+	return -ENOSYS;
+}
+static inline int ssb_gige_map_irq(struct ssb_device *sdev,
+				   const struct pci_dev *pdev)
+{
+	return -ENOSYS;
+}
+static inline int ssb_gige_init(void)
+{
+	return 0;
+}
+static inline void ssb_gige_exit(void)
+{
+}
+
+static inline bool pdev_is_ssb_gige_core(struct pci_dev *pdev)
+{
+	return 0;
+}
+static inline struct ssb_gige * pdev_to_ssb_gige(struct pci_dev *pdev)
+{
+	return NULL;
+}
+static inline bool ssb_gige_is_rgmii(struct pci_dev *pdev)
+{
+	return 0;
+}
+static inline bool ssb_gige_have_roboswitch(struct pci_dev *pdev)
+{
+	return 0;
+}
+static inline bool ssb_gige_one_dma_at_once(struct pci_dev *pdev)
+{
+	return 0;
+}
+static inline bool ssb_gige_must_flush_posted_writes(struct pci_dev *pdev)
+{
+	return 0;
+}
+
+#endif /* CONFIG_SSB_DRIVER_GIGE */
+#endif /* LINUX_SSB_DRIVER_GIGE_H_ */
diff --git a/include/linux/ssb/ssb_driver_pci.h b/include/linux/ssb/ssb_driver_pci.h
index 5e25bac4ed31..41e330e51c2a 100644
--- a/include/linux/ssb/ssb_driver_pci.h
+++ b/include/linux/ssb/ssb_driver_pci.h
@@ -1,6 +1,11 @@
 #ifndef LINUX_SSB_PCICORE_H_
 #define LINUX_SSB_PCICORE_H_
 
+#include <linux/types.h>
+
+struct pci_dev;
+
+
 #ifdef CONFIG_SSB_DRIVER_PCICORE
 
 /* PCI core registers. */
@@ -88,6 +93,9 @@ extern void ssb_pcicore_init(struct ssb_pcicore *pc);
 extern int ssb_pcicore_dev_irqvecs_enable(struct ssb_pcicore *pc,
 					  struct ssb_device *dev);
 
+int ssb_pcicore_plat_dev_init(struct pci_dev *d);
+int ssb_pcicore_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin);
+
 
 #else /* CONFIG_SSB_DRIVER_PCICORE */
 
@@ -107,5 +115,16 @@ int ssb_pcicore_dev_irqvecs_enable(struct ssb_pcicore *pc,
 	return 0;
 }
 
+static inline
+int ssb_pcicore_plat_dev_init(struct pci_dev *d)
+{
+	return -ENODEV;
+}
+static inline
+int ssb_pcicore_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
+{
+	return -ENODEV;
+}
+
 #endif /* CONFIG_SSB_DRIVER_PCICORE */
 #endif /* LINUX_SSB_PCICORE_H_ */
-- 
cgit v1.2.3


From 8c8696553aa3895c2ad4289537e4af45a8877b62 Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@windriver.com>
Date: Thu, 6 Mar 2008 15:05:07 -0800
Subject: [TIPC]: Removal of message header option code

This patch removes code associated with optional, user-specified
fields of the TIPC message header.  Such fields were never
utilized by TIPC, and have now been removed from the protocol
specification.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tipc/tipc_port.h |  7 -------
 net/tipc/core.c              |  1 -
 net/tipc/msg.h               | 23 -----------------------
 net/tipc/port.c              |  7 -------
 4 files changed, 38 deletions(-)

(limited to 'include')

diff --git a/include/net/tipc/tipc_port.h b/include/net/tipc/tipc_port.h
index cfc4ba46de8f..c9b36b77a0b9 100644
--- a/include/net/tipc/tipc_port.h
+++ b/include/net/tipc/tipc_port.h
@@ -86,13 +86,6 @@ u32 tipc_createport_raw(void *usr_handle,
 			void (*wakeup)(struct tipc_port *),
 			const u32 importance);
 
-/*
- * tipc_set_msg_option(): port must be locked.
- */
-int tipc_set_msg_option(struct tipc_port *tp_ptr,
-			const char *opt,
-			const u32 len);
-
 int tipc_reject_msg(struct sk_buff *buf, u32 err);
 
 int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode);
diff --git a/net/tipc/core.c b/net/tipc/core.c
index d2d7d32c02c7..9a5eb97684a2 100644
--- a/net/tipc/core.c
+++ b/net/tipc/core.c
@@ -277,7 +277,6 @@ EXPORT_SYMBOL(tipc_register_media);
 /* TIPC API for external APIs (see tipc_port.h) */
 
 EXPORT_SYMBOL(tipc_createport_raw);
-EXPORT_SYMBOL(tipc_set_msg_option);
 EXPORT_SYMBOL(tipc_reject_msg);
 EXPORT_SYMBOL(tipc_send_buf_fast);
 EXPORT_SYMBOL(tipc_acknowledge);
diff --git a/net/tipc/msg.h b/net/tipc/msg.h
index e9ef6df26562..12f8945e8a54 100644
--- a/net/tipc/msg.h
+++ b/net/tipc/msg.h
@@ -190,18 +190,6 @@ static inline void msg_set_lookup_scope(struct tipc_msg *m, u32 n)
 	msg_set_bits(m, 1, 19, 0x3, n);
 }
 
-static inline void msg_set_options(struct tipc_msg *m, const char *opt, u32 sz)
-{
-	u32 hsz = msg_hdr_sz(m);
-	char *to = (char *)&m->hdr[hsz/4];
-
-	if ((hsz < DIR_MSG_H_SIZE) || ((hsz + sz) > MAX_H_SIZE))
-		return;
-	msg_set_bits(m, 1, 16, 0x7, (hsz - 28)/4);
-	msg_set_hdr_sz(m, hsz + sz);
-	memcpy(to, opt, sz);
-}
-
 static inline u32 msg_bcast_ack(struct tipc_msg *m)
 {
 	return msg_bits(m, 1, 0, 0xffff);
@@ -330,17 +318,6 @@ static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m)
 	return (struct tipc_msg *)msg_data(m);
 }
 
-static inline void msg_expand(struct tipc_msg *m, u32 destnode)
-{
-	if (!msg_short(m))
-		return;
-	msg_set_hdr_sz(m, LONG_H_SIZE);
-	msg_set_orignode(m, msg_prevnode(m));
-	msg_set_destnode(m, destnode);
-	memset(&m->hdr[8], 0, 12);
-}
-
-
 
 /*
 		TIPC internal message header format, version 2
diff --git a/net/tipc/port.c b/net/tipc/port.c
index f508614ca59b..2a16ca43e953 100644
--- a/net/tipc/port.c
+++ b/net/tipc/port.c
@@ -413,13 +413,6 @@ static struct sk_buff *port_build_proto_msg(u32 destport, u32 destnode,
 	return buf;
 }
 
-int tipc_set_msg_option(struct tipc_port *tp_ptr, const char *opt, const u32 sz)
-{
-	msg_expand(&tp_ptr->phdr, msg_destnode(&tp_ptr->phdr));
-	msg_set_options(&tp_ptr->phdr, opt, sz);
-	return TIPC_OK;
-}
-
 int tipc_reject_msg(struct sk_buff *buf, u32 err)
 {
 	struct tipc_msg *msg = buf_msg(buf);
-- 
cgit v1.2.3


From 0e0609bbd2ab39a5964a70b409a5567ebbaf3700 Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@windriver.com>
Date: Thu, 6 Mar 2008 15:06:06 -0800
Subject: [TIPC]: Eliminate "sparse" symbol warnings

This patch eliminates warnings about undeclared symbols.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tipc/tipc_bearer.h | 9 +++++++++
 net/tipc/core.c                | 8 --------
 net/tipc/core.h                | 6 ++++++
 3 files changed, 15 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/tipc/tipc_bearer.h b/include/net/tipc/tipc_bearer.h
index 2151a80cdf30..ee2f304e4919 100644
--- a/include/net/tipc/tipc_bearer.h
+++ b/include/net/tipc/tipc_bearer.h
@@ -99,6 +99,9 @@ struct tipc_bearer {
 	char name[TIPC_MAX_BEARER_NAME];
 };
 
+/*
+ * TIPC routines available to supported media types
+ */
 
 int  tipc_register_media(u32 media_type,
 			 char *media_name, 
@@ -123,6 +126,12 @@ void tipc_continue(struct tipc_bearer *tb_ptr);
 int tipc_enable_bearer(const char *bearer_name, u32 bcast_scope, u32 priority);
 int tipc_disable_bearer(const char *name);
 
+/*
+ * Routines made available to TIPC by supported media types
+ */
+
+int  tipc_eth_media_start(void);
+void tipc_eth_media_stop(void);
 
 #endif
 
diff --git a/net/tipc/core.c b/net/tipc/core.c
index 9a5eb97684a2..b41b0ac67c0b 100644
--- a/net/tipc/core.c
+++ b/net/tipc/core.c
@@ -48,14 +48,6 @@
 #include "subscr.h"
 #include "config.h"
 
-int  tipc_eth_media_start(void);
-void tipc_eth_media_stop(void);
-int  tipc_handler_start(void);
-void tipc_handler_stop(void);
-int  tipc_socket_init(void);
-void tipc_socket_stop(void);
-int  tipc_netlink_start(void);
-void tipc_netlink_stop(void);
 
 #define TIPC_MOD_VER "1.6.2"
 
diff --git a/net/tipc/core.h b/net/tipc/core.h
index feabca580820..3fe9b70331d9 100644
--- a/net/tipc/core.h
+++ b/net/tipc/core.h
@@ -180,6 +180,12 @@ extern int  tipc_core_start(void);
 extern void tipc_core_stop(void);
 extern int  tipc_core_start_net(void);
 extern void tipc_core_stop_net(void);
+extern int  tipc_handler_start(void);
+extern void tipc_handler_stop(void);
+extern int  tipc_netlink_start(void);
+extern void tipc_netlink_stop(void);
+extern int  tipc_socket_init(void);
+extern void tipc_socket_stop(void);
 
 static inline int delimit(int val, int min, int max)
 {
-- 
cgit v1.2.3


From db8dac20d5199307dcfcf4e01dac4bda5edf9e89 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 6 Mar 2008 16:22:02 -0800
Subject: [UDP]: Revert udplite and code split.

This reverts commit db1ed684f6c430c4cdad67d058688b8a1b5e607c ("[IPV6]
UDP: Rename IPv6 UDP files."), commit
8be8af8fa4405652e6c0797db5465a4be8afb998 ("[IPV4] UDP: Move
IPv4-specific bits to other file.") and commit
e898d4db2749c6052072e9bc4448e396cbdeb06a ("[UDP]: Allow users to
configure UDP-Lite.").

First, udplite is of such small cost, and it is a core protocol just
like TCP and normal UDP are.

We spent enormous amounts of effort to make udplite share as much code
with core UDP as possible.  All of that work is less valuable if we're
just going to slap a config option on udplite support.

It is also causing build failures, as reported on linux-next, showing
that the changeset was not tested very well.  In fact, this is the
second build failure resulting from the udplite change.

Finally, the config options provided was a bool, instead of a modular
option.  Meaning the udplite code does not even get build tested
by allmodconfig builds, and furthermore the user is not presented
with a reasonable modular build option which is particularly needed
by distribution vendors.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h      |   10 -
 include/net/ipv6.h       |    5 -
 include/net/transp_v6.h  |    5 -
 include/net/udplite.h    |    9 +-
 net/ipv4/Kconfig         |   10 -
 net/ipv4/Makefile        |    3 +-
 net/ipv4/af_inet.c       |    7 +-
 net/ipv4/proc.c          |    5 +-
 net/ipv4/udp.c           | 1090 +++++++++++++++++++++++++++++++++++++++++++-
 net/ipv4/udp_ipv4.c      | 1134 ----------------------------------------------
 net/ipv4/udplite.c       |  121 +++++
 net/ipv4/udplite_ipv4.c  |  121 -----
 net/ipv6/Makefile        |    3 +-
 net/ipv6/af_inet6.c      |   14 -
 net/ipv6/ipv6_sockglue.c |    6 +-
 net/ipv6/proc.c          |    6 -
 net/ipv6/udp.c           | 1065 +++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/udp_ipv6.c      | 1065 -------------------------------------------
 net/ipv6/udplite.c       |  125 +++++
 net/ipv6/udplite_ipv6.c  |  125 -----
 20 files changed, 2404 insertions(+), 2525 deletions(-)
 delete mode 100644 net/ipv4/udp_ipv4.c
 create mode 100644 net/ipv4/udplite.c
 delete mode 100644 net/ipv4/udplite_ipv4.c
 create mode 100644 net/ipv6/udp.c
 delete mode 100644 net/ipv6/udp_ipv6.c
 create mode 100644 net/ipv6/udplite.c
 delete mode 100644 net/ipv6/udplite_ipv6.c

(limited to 'include')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 4144664d69d9..1e7b7cb5703b 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -70,10 +70,8 @@ struct udp_sock {
 #define UDPLITE_BIT      0x1  		/* set by udplite proto init function */
 #define UDPLITE_SEND_CC  0x2  		/* set via udplite setsockopt         */
 #define UDPLITE_RECV_CC  0x4		/* set via udplite setsocktopt        */
-#ifdef CONFIG_IP_UDPLITE
 	__u8		 pcflag;        /* marks socket as UDP-Lite if > 0    */
 	__u8		 unused[3];
-#endif
 	/*
 	 * For encapsulation sockets.
 	 */
@@ -85,15 +83,7 @@ static inline struct udp_sock *udp_sk(const struct sock *sk)
 	return (struct udp_sock *)sk;
 }
 
-#ifdef CONFIG_IP_UDPLITE
 #define IS_UDPLITE(__sk) (udp_sk(__sk)->pcflag)
-#define IS_PROTO_UDPLITE(__proto) ((__proto) == IPPROTO_UDPLITE)
-#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP || (level) == SOL_UDPLITE)
-#else
-#define IS_UDPLITE(__sk) 0
-#define IS_PROTO_UDPLITE(__proto) 0
-#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP)
-#endif
 
 #endif
 
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 5f6df50a33a9..8db06af1efbb 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -599,13 +599,8 @@ extern int  tcp6_proc_init(void);
 extern void tcp6_proc_exit(void);
 extern int  udp6_proc_init(void);
 extern void udp6_proc_exit(void);
-#ifdef CONFIG_IP_UDPLITE
 extern int  udplite6_proc_init(void);
 extern void udplite6_proc_exit(void);
-#else
-static inline int udplite6_proc_init(void) { return 0; }
-static inline void udplite6_proc_exit(void) { }
-#endif
 extern int  ipv6_misc_proc_init(void);
 extern void ipv6_misc_proc_exit(void);
 extern int snmp6_register_dev(struct inet6_dev *idev);
diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h
index 902e6c6bc793..27394e0447d8 100644
--- a/include/net/transp_v6.h
+++ b/include/net/transp_v6.h
@@ -27,13 +27,8 @@ extern int				rawv6_init(void);
 extern void				rawv6_exit(void);
 extern int				udpv6_init(void);
 extern void				udpv6_exit(void);
-#ifdef CONFIG_IP_UDPLITE
 extern int 				udplitev6_init(void);
 extern void 				udplitev6_exit(void);
-#else
-static inline int udplitev6_init(void) { return 0; }
-static inline void udplitev6_exit(void) { }
-#endif
 extern int				tcpv6_init(void);
 extern void				tcpv6_exit(void);
 
diff --git a/include/net/udplite.h b/include/net/udplite.h
index 01ddb2c20264..b76b2e377af4 100644
--- a/include/net/udplite.h
+++ b/include/net/udplite.h
@@ -25,9 +25,7 @@ static __inline__ int udplite_getfrag(void *from, char *to, int  offset,
 /* Designate sk as UDP-Lite socket */
 static inline int udplite_sk_init(struct sock *sk)
 {
-#ifdef CONFIG_IP_UDPLITE
 	udp_sk(sk)->pcflag = UDPLITE_BIT;
-#endif
 	return 0;
 }
 
@@ -71,7 +69,7 @@ static inline int udplite_checksum_init(struct sk_buff *skb, struct udphdr *uh)
 static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh)
 {
 	int cscov = up->len;
-#ifdef CONFIG_IP_UDPLITE
+
 	/*
 	 * Sender has set `partial coverage' option on UDP-Lite socket
 	 */
@@ -95,15 +93,13 @@ static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh)
 	 *       illegal, we fall back to the defaults here.
 	 */
 	}
-#endif
 	return cscov;
 }
 
 static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb)
 {
-	__wsum csum = 0;
-#ifdef CONFIG_IP_UDPLITE
 	int cscov = udplite_sender_cscov(udp_sk(sk), udp_hdr(skb));
+	__wsum csum = 0;
 
 	skb->ip_summed = CHECKSUM_NONE;     /* no HW support for checksumming */
 
@@ -116,7 +112,6 @@ static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb)
 		if ((cscov -= len) <= 0)
 			break;
 	}
-#endif
 	return csum;
 }
 
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 5098fd2ff4d0..9c7e5ffb223d 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -632,15 +632,5 @@ config TCP_MD5SIG
 
 	  If unsure, say N.
 
-config IP_UDPLITE
-	bool "IP: UDP-Lite Protocol (RFC 3828)"
-	default n
-	---help---
-	  UDP-Lite (RFC 3828) is a UDP-like protocol with variable-length
-	  checksum.  Read <file:Documentation/networking/udplite.txt> for
-	  details.
-
-	  If unsure, say N.
-
 source "net/ipv4/ipvs/Kconfig"
 
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index d5226241d5ed..ad40ef3f9ebc 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -8,7 +8,7 @@ obj-y     := route.o inetpeer.o protocol.o \
 	     inet_timewait_sock.o inet_connection_sock.o \
 	     tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \
 	     tcp_minisocks.o tcp_cong.o \
-	     datagram.o raw.o udp.o udp_ipv4.o \
+	     datagram.o raw.o udp.o udplite.o \
 	     arp.o icmp.o devinet.o af_inet.o  igmp.o \
 	     fib_frontend.o fib_semantics.o \
 	     inet_fragment.o
@@ -49,7 +49,6 @@ obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
 obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
 obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
 obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
-obj-$(CONFIG_IP_UDPLITE) += udplite_ipv4.o
 obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
 
 obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 25871c6c7444..4cb8a1385539 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1317,18 +1317,15 @@ static int __init init_ipv4_mibs(void)
 	if (snmp_mib_init((void **)udp_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udp_mib;
-#ifdef CONFIG_IP_UDPLITE
 	if (snmp_mib_init((void **)udplite_statistics,
 			  sizeof(struct udp_mib)) < 0)
 		goto err_udplite_mib;
-#endif
+
 	tcp_mib_init();
 
 	return 0;
 
-#ifdef CONFIG_IP_UDPLITE
 err_udplite_mib:
-#endif
 	snmp_mib_free((void **)udp_statistics);
 err_udp_mib:
 	snmp_mib_free((void **)tcp_statistics);
@@ -1426,10 +1423,8 @@ static int __init inet_init(void)
 	/* Setup UDP memory threshold */
 	udp_init();
 
-#ifdef CONFIG_IP_UDPLITE
 	/* Add UDP-Lite (RFC 3828) */
 	udplite4_register();
-#endif
 
 	/*
 	 *	Set the ICMP layer up
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index d75ddb7fa4b8..d63474c6b400 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -59,9 +59,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 		   atomic_read(&tcp_memory_allocated));
 	seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse_get(&udp_prot),
 		   atomic_read(&udp_memory_allocated));
-#ifdef CONFIG_IP_UDPLITE
 	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot));
-#endif
 	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot));
 	seq_printf(seq,  "FRAG: inuse %d memory %d\n",
 			ip_frag_nqueues(&init_net), ip_frag_mem(&init_net));
@@ -351,7 +349,6 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 			   snmp_fold_field((void **)udp_statistics,
 					   snmp4_udp_list[i].entry));
 
-#ifdef CONFIG_IP_UDPLITE
 	/* the UDP and UDP-Lite MIBs are the same */
 	seq_puts(seq, "\nUdpLite:");
 	for (i = 0; snmp4_udp_list[i].name != NULL; i++)
@@ -362,7 +359,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 		seq_printf(seq, " %lu",
 			   snmp_fold_field((void **)udplite_statistics,
 					   snmp4_udp_list[i].entry));
-#endif
+
 	seq_putc(seq, '\n');
 	return 0;
 }
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index c53d7673b57d..7ea1b67b6de1 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -246,6 +246,553 @@ int udp_get_port(struct sock *sk, unsigned short snum,
 	return  __udp_lib_get_port(sk, snum, udp_hash, scmp);
 }
 
+int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
+{
+	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);
+
+	return 	( !ipv6_only_sock(sk2)  &&
+		  (!inet1->rcv_saddr || !inet2->rcv_saddr ||
+		   inet1->rcv_saddr == inet2->rcv_saddr      ));
+}
+
+static inline int udp_v4_get_port(struct sock *sk, unsigned short snum)
+{
+	return udp_get_port(sk, snum, ipv4_rcv_saddr_equal);
+}
+
+/* UDP is nearly always wildcards out the wazoo, it makes no sense to try
+ * harder than this. -DaveM
+ */
+static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
+		__be16 sport, __be32 daddr, __be16 dport,
+		int dif, struct hlist_head udptable[])
+{
+	struct sock *sk, *result = NULL;
+	struct hlist_node *node;
+	unsigned short hnum = ntohs(dport);
+	int badness = -1;
+
+	read_lock(&udp_hash_lock);
+	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
+		struct inet_sock *inet = inet_sk(sk);
+
+		if (sk->sk_net == net && sk->sk_hash == hnum &&
+				!ipv6_only_sock(sk)) {
+			int score = (sk->sk_family == PF_INET ? 1 : 0);
+			if (inet->rcv_saddr) {
+				if (inet->rcv_saddr != daddr)
+					continue;
+				score+=2;
+			}
+			if (inet->daddr) {
+				if (inet->daddr != saddr)
+					continue;
+				score+=2;
+			}
+			if (inet->dport) {
+				if (inet->dport != sport)
+					continue;
+				score+=2;
+			}
+			if (sk->sk_bound_dev_if) {
+				if (sk->sk_bound_dev_if != dif)
+					continue;
+				score+=2;
+			}
+			if (score == 9) {
+				result = sk;
+				break;
+			} else if (score > badness) {
+				result = sk;
+				badness = score;
+			}
+		}
+	}
+	if (result)
+		sock_hold(result);
+	read_unlock(&udp_hash_lock);
+	return result;
+}
+
+static inline struct sock *udp_v4_mcast_next(struct sock *sk,
+					     __be16 loc_port, __be32 loc_addr,
+					     __be16 rmt_port, __be32 rmt_addr,
+					     int dif)
+{
+	struct hlist_node *node;
+	struct sock *s = sk;
+	unsigned short hnum = ntohs(loc_port);
+
+	sk_for_each_from(s, node) {
+		struct inet_sock *inet = inet_sk(s);
+
+		if (s->sk_hash != hnum					||
+		    (inet->daddr && inet->daddr != rmt_addr)		||
+		    (inet->dport != rmt_port && inet->dport)		||
+		    (inet->rcv_saddr && inet->rcv_saddr != loc_addr)	||
+		    ipv6_only_sock(s)					||
+		    (s->sk_bound_dev_if && s->sk_bound_dev_if != dif))
+			continue;
+		if (!ip_mc_sf_allow(s, loc_addr, rmt_addr, dif))
+			continue;
+		goto found;
+	}
+	s = NULL;
+found:
+	return s;
+}
+
+/*
+ * This routine is called by the ICMP module when it gets some
+ * sort of error condition.  If err < 0 then the socket should
+ * be closed and the error returned to the user.  If err > 0
+ * it's just the icmp type << 8 | icmp code.
+ * Header points to the ip header of the error packet. We move
+ * on past this. Then (as it used to claim before adjustment)
+ * header points to the first 8 bytes of the udp header.  We need
+ * to find the appropriate port.
+ */
+
+void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[])
+{
+	struct inet_sock *inet;
+	struct iphdr *iph = (struct iphdr*)skb->data;
+	struct udphdr *uh = (struct udphdr*)(skb->data+(iph->ihl<<2));
+	const int type = icmp_hdr(skb)->type;
+	const int code = icmp_hdr(skb)->code;
+	struct sock *sk;
+	int harderr;
+	int err;
+
+	sk = __udp4_lib_lookup(skb->dev->nd_net, iph->daddr, uh->dest,
+			iph->saddr, uh->source, skb->dev->ifindex, udptable);
+	if (sk == NULL) {
+		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
+		return;	/* No socket for error */
+	}
+
+	err = 0;
+	harderr = 0;
+	inet = inet_sk(sk);
+
+	switch (type) {
+	default:
+	case ICMP_TIME_EXCEEDED:
+		err = EHOSTUNREACH;
+		break;
+	case ICMP_SOURCE_QUENCH:
+		goto out;
+	case ICMP_PARAMETERPROB:
+		err = EPROTO;
+		harderr = 1;
+		break;
+	case ICMP_DEST_UNREACH:
+		if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */
+			if (inet->pmtudisc != IP_PMTUDISC_DONT) {
+				err = EMSGSIZE;
+				harderr = 1;
+				break;
+			}
+			goto out;
+		}
+		err = EHOSTUNREACH;
+		if (code <= NR_ICMP_UNREACH) {
+			harderr = icmp_err_convert[code].fatal;
+			err = icmp_err_convert[code].errno;
+		}
+		break;
+	}
+
+	/*
+	 *      RFC1122: OK.  Passes ICMP errors back to application, as per
+	 *	4.1.3.3.
+	 */
+	if (!inet->recverr) {
+		if (!harderr || sk->sk_state != TCP_ESTABLISHED)
+			goto out;
+	} else {
+		ip_icmp_error(sk, skb, err, uh->dest, info, (u8*)(uh+1));
+	}
+	sk->sk_err = err;
+	sk->sk_error_report(sk);
+out:
+	sock_put(sk);
+}
+
+void udp_err(struct sk_buff *skb, u32 info)
+{
+	__udp4_lib_err(skb, info, udp_hash);
+}
+
+/*
+ * Throw away all pending data and cancel the corking. Socket is locked.
+ */
+static void udp_flush_pending_frames(struct sock *sk)
+{
+	struct udp_sock *up = udp_sk(sk);
+
+	if (up->pending) {
+		up->len = 0;
+		up->pending = 0;
+		ip_flush_pending_frames(sk);
+	}
+}
+
+/**
+ * 	udp4_hwcsum_outgoing  -  handle outgoing HW checksumming
+ * 	@sk: 	socket we are sending on
+ * 	@skb: 	sk_buff containing the filled-in UDP header
+ * 	        (checksum field must be zeroed out)
+ */
+static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb,
+				 __be32 src, __be32 dst, int len      )
+{
+	unsigned int offset;
+	struct udphdr *uh = udp_hdr(skb);
+	__wsum csum = 0;
+
+	if (skb_queue_len(&sk->sk_write_queue) == 1) {
+		/*
+		 * Only one fragment on the socket.
+		 */
+		skb->csum_start = skb_transport_header(skb) - skb->head;
+		skb->csum_offset = offsetof(struct udphdr, check);
+		uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0);
+	} else {
+		/*
+		 * HW-checksum won't work as there are two or more
+		 * fragments on the socket so that all csums of sk_buffs
+		 * should be together
+		 */
+		offset = skb_transport_offset(skb);
+		skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);
+
+		skb->ip_summed = CHECKSUM_NONE;
+
+		skb_queue_walk(&sk->sk_write_queue, skb) {
+			csum = csum_add(csum, skb->csum);
+		}
+
+		uh->check = csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum);
+		if (uh->check == 0)
+			uh->check = CSUM_MANGLED_0;
+	}
+}
+
+/*
+ * Push out all pending data as one UDP datagram. Socket is locked.
+ */
+static int udp_push_pending_frames(struct sock *sk)
+{
+	struct udp_sock  *up = udp_sk(sk);
+	struct inet_sock *inet = inet_sk(sk);
+	struct flowi *fl = &inet->cork.fl;
+	struct sk_buff *skb;
+	struct udphdr *uh;
+	int err = 0;
+	int is_udplite = IS_UDPLITE(sk);
+	__wsum csum = 0;
+
+	/* Grab the skbuff where UDP header space exists. */
+	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
+		goto out;
+
+	/*
+	 * Create a UDP header
+	 */
+	uh = udp_hdr(skb);
+	uh->source = fl->fl_ip_sport;
+	uh->dest = fl->fl_ip_dport;
+	uh->len = htons(up->len);
+	uh->check = 0;
+
+	if (is_udplite)  				 /*     UDP-Lite      */
+		csum  = udplite_csum_outgoing(sk, skb);
+
+	else if (sk->sk_no_check == UDP_CSUM_NOXMIT) {   /* UDP csum disabled */
+
+		skb->ip_summed = CHECKSUM_NONE;
+		goto send;
+
+	} else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */
+
+		udp4_hwcsum_outgoing(sk, skb, fl->fl4_src,fl->fl4_dst, up->len);
+		goto send;
+
+	} else						 /*   `normal' UDP    */
+		csum = udp_csum_outgoing(sk, skb);
+
+	/* add protocol-dependent pseudo-header */
+	uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len,
+				      sk->sk_protocol, csum             );
+	if (uh->check == 0)
+		uh->check = CSUM_MANGLED_0;
+
+send:
+	err = ip_push_pending_frames(sk);
+out:
+	up->len = 0;
+	up->pending = 0;
+	if (!err)
+		UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite);
+	return err;
+}
+
+int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+		size_t len)
+{
+	struct inet_sock *inet = inet_sk(sk);
+	struct udp_sock *up = udp_sk(sk);
+	int ulen = len;
+	struct ipcm_cookie ipc;
+	struct rtable *rt = NULL;
+	int free = 0;
+	int connected = 0;
+	__be32 daddr, faddr, saddr;
+	__be16 dport;
+	u8  tos;
+	int err, is_udplite = IS_UDPLITE(sk);
+	int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
+	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
+
+	if (len > 0xFFFF)
+		return -EMSGSIZE;
+
+	/*
+	 *	Check the flags.
+	 */
+
+	if (msg->msg_flags&MSG_OOB)	/* Mirror BSD error message compatibility */
+		return -EOPNOTSUPP;
+
+	ipc.opt = NULL;
+
+	if (up->pending) {
+		/*
+		 * There are pending frames.
+		 * The socket lock must be held while it's corked.
+		 */
+		lock_sock(sk);
+		if (likely(up->pending)) {
+			if (unlikely(up->pending != AF_INET)) {
+				release_sock(sk);
+				return -EINVAL;
+			}
+			goto do_append_data;
+		}
+		release_sock(sk);
+	}
+	ulen += sizeof(struct udphdr);
+
+	/*
+	 *	Get and verify the address.
+	 */
+	if (msg->msg_name) {
+		struct sockaddr_in * usin = (struct sockaddr_in*)msg->msg_name;
+		if (msg->msg_namelen < sizeof(*usin))
+			return -EINVAL;
+		if (usin->sin_family != AF_INET) {
+			if (usin->sin_family != AF_UNSPEC)
+				return -EAFNOSUPPORT;
+		}
+
+		daddr = usin->sin_addr.s_addr;
+		dport = usin->sin_port;
+		if (dport == 0)
+			return -EINVAL;
+	} else {
+		if (sk->sk_state != TCP_ESTABLISHED)
+			return -EDESTADDRREQ;
+		daddr = inet->daddr;
+		dport = inet->dport;
+		/* Open fast path for connected socket.
+		   Route will not be used, if at least one option is set.
+		 */
+		connected = 1;
+	}
+	ipc.addr = inet->saddr;
+
+	ipc.oif = sk->sk_bound_dev_if;
+	if (msg->msg_controllen) {
+		err = ip_cmsg_send(msg, &ipc);
+		if (err)
+			return err;
+		if (ipc.opt)
+			free = 1;
+		connected = 0;
+	}
+	if (!ipc.opt)
+		ipc.opt = inet->opt;
+
+	saddr = ipc.addr;
+	ipc.addr = faddr = daddr;
+
+	if (ipc.opt && ipc.opt->srr) {
+		if (!daddr)
+			return -EINVAL;
+		faddr = ipc.opt->faddr;
+		connected = 0;
+	}
+	tos = RT_TOS(inet->tos);
+	if (sock_flag(sk, SOCK_LOCALROUTE) ||
+	    (msg->msg_flags & MSG_DONTROUTE) ||
+	    (ipc.opt && ipc.opt->is_strictroute)) {
+		tos |= RTO_ONLINK;
+		connected = 0;
+	}
+
+	if (ipv4_is_multicast(daddr)) {
+		if (!ipc.oif)
+			ipc.oif = inet->mc_index;
+		if (!saddr)
+			saddr = inet->mc_addr;
+		connected = 0;
+	}
+
+	if (connected)
+		rt = (struct rtable*)sk_dst_check(sk, 0);
+
+	if (rt == NULL) {
+		struct flowi fl = { .oif = ipc.oif,
+				    .nl_u = { .ip4_u =
+					      { .daddr = faddr,
+						.saddr = saddr,
+						.tos = tos } },
+				    .proto = sk->sk_protocol,
+				    .uli_u = { .ports =
+					       { .sport = inet->sport,
+						 .dport = dport } } };
+		security_sk_classify_flow(sk, &fl);
+		err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1);
+		if (err) {
+			if (err == -ENETUNREACH)
+				IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
+			goto out;
+		}
+
+		err = -EACCES;
+		if ((rt->rt_flags & RTCF_BROADCAST) &&
+		    !sock_flag(sk, SOCK_BROADCAST))
+			goto out;
+		if (connected)
+			sk_dst_set(sk, dst_clone(&rt->u.dst));
+	}
+
+	if (msg->msg_flags&MSG_CONFIRM)
+		goto do_confirm;
+back_from_confirm:
+
+	saddr = rt->rt_src;
+	if (!ipc.addr)
+		daddr = ipc.addr = rt->rt_dst;
+
+	lock_sock(sk);
+	if (unlikely(up->pending)) {
+		/* The socket is already corked while preparing it. */
+		/* ... which is an evident application bug. --ANK */
+		release_sock(sk);
+
+		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n");
+		err = -EINVAL;
+		goto out;
+	}
+	/*
+	 *	Now cork the socket to pend data.
+	 */
+	inet->cork.fl.fl4_dst = daddr;
+	inet->cork.fl.fl_ip_dport = dport;
+	inet->cork.fl.fl4_src = saddr;
+	inet->cork.fl.fl_ip_sport = inet->sport;
+	up->pending = AF_INET;
+
+do_append_data:
+	up->len += ulen;
+	getfrag  =  is_udplite ?  udplite_getfrag : ip_generic_getfrag;
+	err = ip_append_data(sk, getfrag, msg->msg_iov, ulen,
+			sizeof(struct udphdr), &ipc, rt,
+			corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
+	if (err)
+		udp_flush_pending_frames(sk);
+	else if (!corkreq)
+		err = udp_push_pending_frames(sk);
+	else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
+		up->pending = 0;
+	release_sock(sk);
+
+out:
+	ip_rt_put(rt);
+	if (free)
+		kfree(ipc.opt);
+	if (!err)
+		return len;
+	/*
+	 * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space.  Reporting
+	 * ENOBUFS might not be good (it's not tunable per se), but otherwise
+	 * we don't have a good statistic (IpOutDiscards but it can be too many
+	 * things).  We could add another new stat but at least for now that
+	 * seems like overkill.
+	 */
+	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
+		UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite);
+	}
+	return err;
+
+do_confirm:
+	dst_confirm(&rt->u.dst);
+	if (!(msg->msg_flags&MSG_PROBE) || len)
+		goto back_from_confirm;
+	err = 0;
+	goto out;
+}
+
+int udp_sendpage(struct sock *sk, struct page *page, int offset,
+		 size_t size, int flags)
+{
+	struct udp_sock *up = udp_sk(sk);
+	int ret;
+
+	if (!up->pending) {
+		struct msghdr msg = {	.msg_flags = flags|MSG_MORE };
+
+		/* Call udp_sendmsg to specify destination address which
+		 * sendpage interface can't pass.
+		 * This will succeed only when the socket is connected.
+		 */
+		ret = udp_sendmsg(NULL, sk, &msg, 0);
+		if (ret < 0)
+			return ret;
+	}
+
+	lock_sock(sk);
+
+	if (unlikely(!up->pending)) {
+		release_sock(sk);
+
+		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 3\n");
+		return -EINVAL;
+	}
+
+	ret = ip_append_page(sk, page, offset, size, flags);
+	if (ret == -EOPNOTSUPP) {
+		release_sock(sk);
+		return sock_no_sendpage(sk->sk_socket, page, offset,
+					size, flags);
+	}
+	if (ret < 0) {
+		udp_flush_pending_frames(sk);
+		goto out;
+	}
+
+	up->len += size;
+	if (!(up->corkflag || (flags&MSG_MORE)))
+		ret = udp_push_pending_frames(sk);
+	if (!ret)
+		ret = size;
+out:
+	release_sock(sk);
+	return ret;
+}
+
 /*
  *	IOCTL requests applicable to the UDP protocol
  */
@@ -286,6 +833,107 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg)
 	return 0;
 }
 
+/*
+ * 	This should be easy, if there is something there we
+ * 	return it, otherwise we block.
+ */
+
+int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+		size_t len, int noblock, int flags, int *addr_len)
+{
+	struct inet_sock *inet = inet_sk(sk);
+	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
+	struct sk_buff *skb;
+	unsigned int ulen, copied;
+	int peeked;
+	int err;
+	int is_udplite = IS_UDPLITE(sk);
+
+	/*
+	 *	Check any passed addresses
+	 */
+	if (addr_len)
+		*addr_len=sizeof(*sin);
+
+	if (flags & MSG_ERRQUEUE)
+		return ip_recv_error(sk, msg, len);
+
+try_again:
+	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
+				  &peeked, &err);
+	if (!skb)
+		goto out;
+
+	ulen = skb->len - sizeof(struct udphdr);
+	copied = len;
+	if (copied > ulen)
+		copied = ulen;
+	else if (copied < ulen)
+		msg->msg_flags |= MSG_TRUNC;
+
+	/*
+	 * If checksum is needed at all, try to do it while copying the
+	 * data.  If the data is truncated, or if we only want a partial
+	 * coverage checksum (UDP-Lite), do it before the copy.
+	 */
+
+	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
+		if (udp_lib_checksum_complete(skb))
+			goto csum_copy_err;
+	}
+
+	if (skb_csum_unnecessary(skb))
+		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
+					      msg->msg_iov, copied       );
+	else {
+		err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov);
+
+		if (err == -EINVAL)
+			goto csum_copy_err;
+	}
+
+	if (err)
+		goto out_free;
+
+	if (!peeked)
+		UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
+
+	sock_recv_timestamp(msg, sk, skb);
+
+	/* Copy the address. */
+	if (sin)
+	{
+		sin->sin_family = AF_INET;
+		sin->sin_port = udp_hdr(skb)->source;
+		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
+		memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
+	}
+	if (inet->cmsg_flags)
+		ip_cmsg_recv(msg, skb);
+
+	err = copied;
+	if (flags & MSG_TRUNC)
+		err = ulen;
+
+out_free:
+	lock_sock(sk);
+	skb_free_datagram(sk, skb);
+	release_sock(sk);
+out:
+	return err;
+
+csum_copy_err:
+	lock_sock(sk);
+	if (!skb_kill_datagram(sk, skb, flags))
+		UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
+	release_sock(sk);
+
+	if (noblock)
+		return -EAGAIN;
+	goto try_again;
+}
+
+
 int udp_disconnect(struct sock *sk, int flags)
 {
 	struct inet_sock *inet = inet_sk(sk);
@@ -308,6 +956,319 @@ int udp_disconnect(struct sock *sk, int flags)
 	return 0;
 }
 
+/* returns:
+ *  -1: error
+ *   0: success
+ *  >0: "udp encap" protocol resubmission
+ *
+ * Note that in the success and error cases, the skb is assumed to
+ * have either been requeued or freed.
+ */
+int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
+{
+	struct udp_sock *up = udp_sk(sk);
+	int rc;
+	int is_udplite = IS_UDPLITE(sk);
+
+	/*
+	 *	Charge it to the socket, dropping if the queue is full.
+	 */
+	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
+		goto drop;
+	nf_reset(skb);
+
+	if (up->encap_type) {
+		/*
+		 * This is an encapsulation socket so pass the skb to
+		 * the socket's udp_encap_rcv() hook. Otherwise, just
+		 * fall through and pass this up the UDP socket.
+		 * up->encap_rcv() returns the following value:
+		 * =0 if skb was successfully passed to the encap
+		 *    handler or was discarded by it.
+		 * >0 if skb should be passed on to UDP.
+		 * <0 if skb should be resubmitted as proto -N
+		 */
+
+		/* if we're overly short, let UDP handle it */
+		if (skb->len > sizeof(struct udphdr) &&
+		    up->encap_rcv != NULL) {
+			int ret;
+
+			ret = (*up->encap_rcv)(sk, skb);
+			if (ret <= 0) {
+				UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS,
+						 is_udplite);
+				return -ret;
+			}
+		}
+
+		/* FALLTHROUGH -- it's a UDP Packet */
+	}
+
+	/*
+	 * 	UDP-Lite specific tests, ignored on UDP sockets
+	 */
+	if ((is_udplite & UDPLITE_RECV_CC)  &&  UDP_SKB_CB(skb)->partial_cov) {
+
+		/*
+		 * MIB statistics other than incrementing the error count are
+		 * disabled for the following two types of errors: these depend
+		 * on the application settings, not on the functioning of the
+		 * protocol stack as such.
+		 *
+		 * RFC 3828 here recommends (sec 3.3): "There should also be a
+		 * way ... to ... at least let the receiving application block
+		 * delivery of packets with coverage values less than a value
+		 * provided by the application."
+		 */
+		if (up->pcrlen == 0) {          /* full coverage was set  */
+			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE: partial coverage "
+				"%d while full coverage %d requested\n",
+				UDP_SKB_CB(skb)->cscov, skb->len);
+			goto drop;
+		}
+		/* The next case involves violating the min. coverage requested
+		 * by the receiver. This is subtle: if receiver wants x and x is
+		 * greater than the buffersize/MTU then receiver will complain
+		 * that it wants x while sender emits packets of smaller size y.
+		 * Therefore the above ...()->partial_cov statement is essential.
+		 */
+		if (UDP_SKB_CB(skb)->cscov  <  up->pcrlen) {
+			LIMIT_NETDEBUG(KERN_WARNING
+				"UDPLITE: coverage %d too small, need min %d\n",
+				UDP_SKB_CB(skb)->cscov, up->pcrlen);
+			goto drop;
+		}
+	}
+
+	if (sk->sk_filter) {
+		if (udp_lib_checksum_complete(skb))
+			goto drop;
+	}
+
+	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
+		/* Note that an ENOMEM error is charged twice */
+		if (rc == -ENOMEM)
+			UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
+		goto drop;
+	}
+
+	return 0;
+
+drop:
+	UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite);
+	kfree_skb(skb);
+	return -1;
+}
+
+/*
+ *	Multicasts and broadcasts go to each listener.
+ *
+ *	Note: called only from the BH handler context,
+ *	so we don't need to lock the hashes.
+ */
+static int __udp4_lib_mcast_deliver(struct sk_buff *skb,
+				    struct udphdr  *uh,
+				    __be32 saddr, __be32 daddr,
+				    struct hlist_head udptable[])
+{
+	struct sock *sk;
+	int dif;
+
+	read_lock(&udp_hash_lock);
+	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
+	dif = skb->dev->ifindex;
+	sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
+	if (sk) {
+		struct sock *sknext = NULL;
+
+		do {
+			struct sk_buff *skb1 = skb;
+
+			sknext = udp_v4_mcast_next(sk_next(sk), uh->dest, daddr,
+						   uh->source, saddr, dif);
+			if (sknext)
+				skb1 = skb_clone(skb, GFP_ATOMIC);
+
+			if (skb1) {
+				int ret = 0;
+
+				bh_lock_sock_nested(sk);
+				if (!sock_owned_by_user(sk))
+					ret = udp_queue_rcv_skb(sk, skb1);
+				else
+					sk_add_backlog(sk, skb1);
+				bh_unlock_sock(sk);
+
+				if (ret > 0)
+					/* we should probably re-process instead
+					 * of dropping packets here. */
+					kfree_skb(skb1);
+			}
+			sk = sknext;
+		} while (sknext);
+	} else
+		kfree_skb(skb);
+	read_unlock(&udp_hash_lock);
+	return 0;
+}
+
+/* Initialize UDP checksum. If exited with zero value (success),
+ * CHECKSUM_UNNECESSARY means, that no more checks are required.
+ * Otherwise, csum completion requires chacksumming packet body,
+ * including udp header and folding it to skb->csum.
+ */
+static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh,
+				 int proto)
+{
+	const struct iphdr *iph;
+	int err;
+
+	UDP_SKB_CB(skb)->partial_cov = 0;
+	UDP_SKB_CB(skb)->cscov = skb->len;
+
+	if (proto == IPPROTO_UDPLITE) {
+		err = udplite_checksum_init(skb, uh);
+		if (err)
+			return err;
+	}
+
+	iph = ip_hdr(skb);
+	if (uh->check == 0) {
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	} else if (skb->ip_summed == CHECKSUM_COMPLETE) {
+	       if (!csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len,
+				      proto, skb->csum))
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+	}
+	if (!skb_csum_unnecessary(skb))
+		skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
+					       skb->len, proto, 0);
+	/* Probably, we should checksum udp header (it should be in cache
+	 * in any case) and data in tiny packets (< rx copybreak).
+	 */
+
+	return 0;
+}
+
+/*
+ *	All we need to do is get the socket, and then do a checksum.
+ */
+
+int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
+		   int proto)
+{
+	struct sock *sk;
+	struct udphdr *uh = udp_hdr(skb);
+	unsigned short ulen;
+	struct rtable *rt = (struct rtable*)skb->dst;
+	__be32 saddr = ip_hdr(skb)->saddr;
+	__be32 daddr = ip_hdr(skb)->daddr;
+
+	/*
+	 *  Validate the packet.
+	 */
+	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
+		goto drop;		/* No space for header. */
+
+	ulen = ntohs(uh->len);
+	if (ulen > skb->len)
+		goto short_packet;
+
+	if (proto == IPPROTO_UDP) {
+		/* UDP validates ulen. */
+		if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
+			goto short_packet;
+		uh = udp_hdr(skb);
+	}
+
+	if (udp4_csum_init(skb, uh, proto))
+		goto csum_error;
+
+	if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
+		return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable);
+
+	sk = __udp4_lib_lookup(skb->dev->nd_net, saddr, uh->source, daddr,
+			uh->dest, inet_iif(skb), udptable);
+
+	if (sk != NULL) {
+		int ret = 0;
+		bh_lock_sock_nested(sk);
+		if (!sock_owned_by_user(sk))
+			ret = udp_queue_rcv_skb(sk, skb);
+		else
+			sk_add_backlog(sk, skb);
+		bh_unlock_sock(sk);
+		sock_put(sk);
+
+		/* a return value > 0 means to resubmit the input, but
+		 * it wants the return to be -protocol, or 0
+		 */
+		if (ret > 0)
+			return -ret;
+		return 0;
+	}
+
+	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
+		goto drop;
+	nf_reset(skb);
+
+	/* No socket. Drop packet silently, if checksum is wrong */
+	if (udp_lib_checksum_complete(skb))
+		goto csum_error;
+
+	UDP_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
+
+	/*
+	 * Hmm.  We got an UDP packet to a port to which we
+	 * don't wanna listen.  Ignore it.
+	 */
+	kfree_skb(skb);
+	return 0;
+
+short_packet:
+	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n",
+		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
+		       NIPQUAD(saddr),
+		       ntohs(uh->source),
+		       ulen,
+		       skb->len,
+		       NIPQUAD(daddr),
+		       ntohs(uh->dest));
+	goto drop;
+
+csum_error:
+	/*
+	 * RFC1122: OK.  Discards the bad packet silently (as far as
+	 * the network is concerned, anyway) as per 4.1.3.4 (MUST).
+	 */
+	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n",
+		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
+		       NIPQUAD(saddr),
+		       ntohs(uh->source),
+		       NIPQUAD(daddr),
+		       ntohs(uh->dest),
+		       ulen);
+drop:
+	UDP_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+	kfree_skb(skb);
+	return 0;
+}
+
+int udp_rcv(struct sk_buff *skb)
+{
+	return __udp4_lib_rcv(skb, udp_hash, IPPROTO_UDP);
+}
+
+int udp_destroy_sock(struct sock *sk)
+{
+	lock_sock(sk);
+	udp_flush_pending_frames(sk);
+	release_sock(sk);
+	return 0;
+}
+
 /*
  *	Socket option code for UDP
  */
@@ -318,9 +1279,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 	struct udp_sock *up = udp_sk(sk);
 	int val;
 	int err = 0;
-#ifdef CONFIG_IP_UDPLITE
 	int is_udplite = IS_UDPLITE(sk);
-#endif
 
 	if (optlen<sizeof(int))
 		return -EINVAL;
@@ -356,7 +1315,6 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 		}
 		break;
 
-#ifdef CONFIG_IP_UDPLITE
 	/*
 	 * 	UDP-Lite's partial checksum coverage (RFC 3828).
 	 */
@@ -382,7 +1340,6 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 		up->pcrlen = val;
 		up->pcflag |= UDPLITE_RECV_CC;
 		break;
-#endif
 
 	default:
 		err = -ENOPROTOOPT;
@@ -392,6 +1349,26 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
 	return err;
 }
 
+int udp_setsockopt(struct sock *sk, int level, int optname,
+		   char __user *optval, int optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
+					  udp_push_pending_frames);
+	return ip_setsockopt(sk, level, optname, optval, optlen);
+}
+
+#ifdef CONFIG_COMPAT
+int compat_udp_setsockopt(struct sock *sk, int level, int optname,
+			  char __user *optval, int optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
+					  udp_push_pending_frames);
+	return compat_ip_setsockopt(sk, level, optname, optval, optlen);
+}
+#endif
+
 int udp_lib_getsockopt(struct sock *sk, int level, int optname,
 		       char __user *optval, int __user *optlen)
 {
@@ -436,6 +1413,23 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname,
 	return 0;
 }
 
+int udp_getsockopt(struct sock *sk, int level, int optname,
+		   char __user *optval, int __user *optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
+	return ip_getsockopt(sk, level, optname, optval, optlen);
+}
+
+#ifdef CONFIG_COMPAT
+int compat_udp_getsockopt(struct sock *sk, int level, int optname,
+				 char __user *optval, int __user *optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
+	return compat_ip_getsockopt(sk, level, optname, optval, optlen);
+}
+#endif
 /**
  * 	udp_poll - wait for a UDP event.
  *	@file - file struct
@@ -480,6 +1474,36 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait)
 
 }
 
+DEFINE_PROTO_INUSE(udp)
+
+struct proto udp_prot = {
+	.name		   = "UDP",
+	.owner		   = THIS_MODULE,
+	.close		   = udp_lib_close,
+	.connect	   = ip4_datagram_connect,
+	.disconnect	   = udp_disconnect,
+	.ioctl		   = udp_ioctl,
+	.destroy	   = udp_destroy_sock,
+	.setsockopt	   = udp_setsockopt,
+	.getsockopt	   = udp_getsockopt,
+	.sendmsg	   = udp_sendmsg,
+	.recvmsg	   = udp_recvmsg,
+	.sendpage	   = udp_sendpage,
+	.backlog_rcv	   = udp_queue_rcv_skb,
+	.hash		   = udp_lib_hash,
+	.unhash		   = udp_lib_unhash,
+	.get_port	   = udp_v4_get_port,
+	.memory_allocated  = &udp_memory_allocated,
+	.sysctl_mem	   = sysctl_udp_mem,
+	.sysctl_wmem	   = &sysctl_udp_wmem_min,
+	.sysctl_rmem	   = &sysctl_udp_rmem_min,
+	.obj_size	   = sizeof(struct udp_sock),
+#ifdef CONFIG_COMPAT
+	.compat_setsockopt = compat_udp_setsockopt,
+	.compat_getsockopt = compat_udp_getsockopt,
+#endif
+	REF_PROTO_INUSE(udp)
+};
 
 /* ------------------------------------------------------------------------ */
 #ifdef CONFIG_PROC_FS
@@ -612,6 +1636,62 @@ void udp_proc_unregister(struct udp_seq_afinfo *afinfo)
 	proc_net_remove(&init_net, afinfo->name);
 	memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops));
 }
+
+/* ------------------------------------------------------------------------ */
+static void udp4_format_sock(struct sock *sp, char *tmpbuf, int bucket)
+{
+	struct inet_sock *inet = inet_sk(sp);
+	__be32 dest = inet->daddr;
+	__be32 src  = inet->rcv_saddr;
+	__u16 destp	  = ntohs(inet->dport);
+	__u16 srcp	  = ntohs(inet->sport);
+
+	sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
+		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p",
+		bucket, src, srcp, dest, destp, sp->sk_state,
+		atomic_read(&sp->sk_wmem_alloc),
+		atomic_read(&sp->sk_rmem_alloc),
+		0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp),
+		atomic_read(&sp->sk_refcnt), sp);
+}
+
+int udp4_seq_show(struct seq_file *seq, void *v)
+{
+	if (v == SEQ_START_TOKEN)
+		seq_printf(seq, "%-127s\n",
+			   "  sl  local_address rem_address   st tx_queue "
+			   "rx_queue tr tm->when retrnsmt   uid  timeout "
+			   "inode");
+	else {
+		char tmpbuf[129];
+		struct udp_iter_state *state = seq->private;
+
+		udp4_format_sock(v, tmpbuf, state->bucket);
+		seq_printf(seq, "%-127s\n", tmpbuf);
+	}
+	return 0;
+}
+
+/* ------------------------------------------------------------------------ */
+static struct file_operations udp4_seq_fops;
+static struct udp_seq_afinfo udp4_seq_afinfo = {
+	.owner		= THIS_MODULE,
+	.name		= "udp",
+	.family		= AF_INET,
+	.hashtable	= udp_hash,
+	.seq_show	= udp4_seq_show,
+	.seq_fops	= &udp4_seq_fops,
+};
+
+int __init udp4_proc_init(void)
+{
+	return udp_proc_register(&udp4_seq_afinfo);
+}
+
+void udp4_proc_exit(void)
+{
+	udp_proc_unregister(&udp4_seq_afinfo);
+}
 #endif /* CONFIG_PROC_FS */
 
 void __init udp_init(void)
@@ -638,6 +1718,8 @@ EXPORT_SYMBOL(udp_hash);
 EXPORT_SYMBOL(udp_hash_lock);
 EXPORT_SYMBOL(udp_ioctl);
 EXPORT_SYMBOL(udp_get_port);
+EXPORT_SYMBOL(udp_prot);
+EXPORT_SYMBOL(udp_sendmsg);
 EXPORT_SYMBOL(udp_lib_getsockopt);
 EXPORT_SYMBOL(udp_lib_setsockopt);
 EXPORT_SYMBOL(udp_poll);
diff --git a/net/ipv4/udp_ipv4.c b/net/ipv4/udp_ipv4.c
deleted file mode 100644
index fd14c2c50ed4..000000000000
--- a/net/ipv4/udp_ipv4.c
+++ /dev/null
@@ -1,1134 +0,0 @@
-/*
- * INET		An implementation of the TCP/IP protocol suite for the LINUX
- *		operating system.  INET is implemented using the  BSD Socket
- *		interface as the means of communication with the user level.
- *
- *		UDP for IPv4.
- *
- *		For full credits, see net/ipv4/udp.c.
- *
- *		This program is free software; you can redistribute it and/or
- *		modify it under the terms of the GNU General Public License
- *		as published by the Free Software Foundation; either version
- *		2 of the License, or (at your option) any later version.
- */
-
-#include <asm/system.h>
-#include <asm/uaccess.h>
-#include <asm/ioctls.h>
-#include <linux/bootmem.h>
-#include <linux/types.h>
-#include <linux/fcntl.h>
-#include <linux/module.h>
-#include <linux/socket.h>
-#include <linux/sockios.h>
-#include <linux/igmp.h>
-#include <linux/in.h>
-#include <linux/errno.h>
-#include <linux/timer.h>
-#include <linux/mm.h>
-#include <linux/inet.h>
-#include <linux/netdevice.h>
-#include <net/tcp_states.h>
-#include <linux/skbuff.h>
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
-#include <net/net_namespace.h>
-#include <net/icmp.h>
-#include <net/route.h>
-#include <net/checksum.h>
-#include <net/xfrm.h>
-#include "udp_impl.h"
-
-int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
-{
-	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);
-
-	return 	( !ipv6_only_sock(sk2)  &&
-		  (!inet1->rcv_saddr || !inet2->rcv_saddr ||
-		   inet1->rcv_saddr == inet2->rcv_saddr      ));
-}
-
-static inline int udp_v4_get_port(struct sock *sk, unsigned short snum)
-{
-	return udp_get_port(sk, snum, ipv4_rcv_saddr_equal);
-}
-
-/* UDP is nearly always wildcards out the wazoo, it makes no sense to try
- * harder than this. -DaveM
- */
-static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
-		__be16 sport, __be32 daddr, __be16 dport,
-		int dif, struct hlist_head udptable[])
-{
-	struct sock *sk, *result = NULL;
-	struct hlist_node *node;
-	unsigned short hnum = ntohs(dport);
-	int badness = -1;
-
-	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
-		struct inet_sock *inet = inet_sk(sk);
-
-		if (sk->sk_net == net && sk->sk_hash == hnum &&
-				!ipv6_only_sock(sk)) {
-			int score = (sk->sk_family == PF_INET ? 1 : 0);
-			if (inet->rcv_saddr) {
-				if (inet->rcv_saddr != daddr)
-					continue;
-				score+=2;
-			}
-			if (inet->daddr) {
-				if (inet->daddr != saddr)
-					continue;
-				score+=2;
-			}
-			if (inet->dport) {
-				if (inet->dport != sport)
-					continue;
-				score+=2;
-			}
-			if (sk->sk_bound_dev_if) {
-				if (sk->sk_bound_dev_if != dif)
-					continue;
-				score+=2;
-			}
-			if (score == 9) {
-				result = sk;
-				break;
-			} else if (score > badness) {
-				result = sk;
-				badness = score;
-			}
-		}
-	}
-	if (result)
-		sock_hold(result);
-	read_unlock(&udp_hash_lock);
-	return result;
-}
-
-static inline struct sock *udp_v4_mcast_next(struct sock *sk,
-					     __be16 loc_port, __be32 loc_addr,
-					     __be16 rmt_port, __be32 rmt_addr,
-					     int dif)
-{
-	struct hlist_node *node;
-	struct sock *s = sk;
-	unsigned short hnum = ntohs(loc_port);
-
-	sk_for_each_from(s, node) {
-		struct inet_sock *inet = inet_sk(s);
-
-		if (s->sk_hash != hnum					||
-		    (inet->daddr && inet->daddr != rmt_addr)		||
-		    (inet->dport != rmt_port && inet->dport)		||
-		    (inet->rcv_saddr && inet->rcv_saddr != loc_addr)	||
-		    ipv6_only_sock(s)					||
-		    (s->sk_bound_dev_if && s->sk_bound_dev_if != dif))
-			continue;
-		if (!ip_mc_sf_allow(s, loc_addr, rmt_addr, dif))
-			continue;
-		goto found;
-	}
-	s = NULL;
-found:
-	return s;
-}
-
-/*
- * This routine is called by the ICMP module when it gets some
- * sort of error condition.  If err < 0 then the socket should
- * be closed and the error returned to the user.  If err > 0
- * it's just the icmp type << 8 | icmp code.
- * Header points to the ip header of the error packet. We move
- * on past this. Then (as it used to claim before adjustment)
- * header points to the first 8 bytes of the udp header.  We need
- * to find the appropriate port.
- */
-
-void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[])
-{
-	struct inet_sock *inet;
-	struct iphdr *iph = (struct iphdr*)skb->data;
-	struct udphdr *uh = (struct udphdr*)(skb->data+(iph->ihl<<2));
-	const int type = icmp_hdr(skb)->type;
-	const int code = icmp_hdr(skb)->code;
-	struct sock *sk;
-	int harderr;
-	int err;
-
-	sk = __udp4_lib_lookup(skb->dev->nd_net, iph->daddr, uh->dest,
-			iph->saddr, uh->source, skb->dev->ifindex, udptable);
-	if (sk == NULL) {
-		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
-		return;	/* No socket for error */
-	}
-
-	err = 0;
-	harderr = 0;
-	inet = inet_sk(sk);
-
-	switch (type) {
-	default:
-	case ICMP_TIME_EXCEEDED:
-		err = EHOSTUNREACH;
-		break;
-	case ICMP_SOURCE_QUENCH:
-		goto out;
-	case ICMP_PARAMETERPROB:
-		err = EPROTO;
-		harderr = 1;
-		break;
-	case ICMP_DEST_UNREACH:
-		if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */
-			if (inet->pmtudisc != IP_PMTUDISC_DONT) {
-				err = EMSGSIZE;
-				harderr = 1;
-				break;
-			}
-			goto out;
-		}
-		err = EHOSTUNREACH;
-		if (code <= NR_ICMP_UNREACH) {
-			harderr = icmp_err_convert[code].fatal;
-			err = icmp_err_convert[code].errno;
-		}
-		break;
-	}
-
-	/*
-	 *      RFC1122: OK.  Passes ICMP errors back to application, as per
-	 *	4.1.3.3.
-	 */
-	if (!inet->recverr) {
-		if (!harderr || sk->sk_state != TCP_ESTABLISHED)
-			goto out;
-	} else {
-		ip_icmp_error(sk, skb, err, uh->dest, info, (u8*)(uh+1));
-	}
-	sk->sk_err = err;
-	sk->sk_error_report(sk);
-out:
-	sock_put(sk);
-}
-
-void udp_err(struct sk_buff *skb, u32 info)
-{
-	__udp4_lib_err(skb, info, udp_hash);
-}
-
-/*
- * Throw away all pending data and cancel the corking. Socket is locked.
- */
-static void udp_flush_pending_frames(struct sock *sk)
-{
-	struct udp_sock *up = udp_sk(sk);
-
-	if (up->pending) {
-		up->len = 0;
-		up->pending = 0;
-		ip_flush_pending_frames(sk);
-	}
-}
-
-/**
- * 	udp4_hwcsum_outgoing  -  handle outgoing HW checksumming
- * 	@sk: 	socket we are sending on
- * 	@skb: 	sk_buff containing the filled-in UDP header
- * 	        (checksum field must be zeroed out)
- */
-static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb,
-				 __be32 src, __be32 dst, int len      )
-{
-	unsigned int offset;
-	struct udphdr *uh = udp_hdr(skb);
-	__wsum csum = 0;
-
-	if (skb_queue_len(&sk->sk_write_queue) == 1) {
-		/*
-		 * Only one fragment on the socket.
-		 */
-		skb->csum_start = skb_transport_header(skb) - skb->head;
-		skb->csum_offset = offsetof(struct udphdr, check);
-		uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0);
-	} else {
-		/*
-		 * HW-checksum won't work as there are two or more
-		 * fragments on the socket so that all csums of sk_buffs
-		 * should be together
-		 */
-		offset = skb_transport_offset(skb);
-		skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);
-
-		skb->ip_summed = CHECKSUM_NONE;
-
-		skb_queue_walk(&sk->sk_write_queue, skb) {
-			csum = csum_add(csum, skb->csum);
-		}
-
-		uh->check = csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum);
-		if (uh->check == 0)
-			uh->check = CSUM_MANGLED_0;
-	}
-}
-
-/*
- * Push out all pending data as one UDP datagram. Socket is locked.
- */
-static int udp_push_pending_frames(struct sock *sk)
-{
-	struct udp_sock  *up = udp_sk(sk);
-	struct inet_sock *inet = inet_sk(sk);
-	struct flowi *fl = &inet->cork.fl;
-	struct sk_buff *skb;
-	struct udphdr *uh;
-	int err = 0;
-	int is_udplite = IS_UDPLITE(sk);
-	__wsum csum = 0;
-
-	/* Grab the skbuff where UDP header space exists. */
-	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
-		goto out;
-
-	/*
-	 * Create a UDP header
-	 */
-	uh = udp_hdr(skb);
-	uh->source = fl->fl_ip_sport;
-	uh->dest = fl->fl_ip_dport;
-	uh->len = htons(up->len);
-	uh->check = 0;
-
-	if (is_udplite)  				 /*     UDP-Lite      */
-		csum  = udplite_csum_outgoing(sk, skb);
-
-	else if (sk->sk_no_check == UDP_CSUM_NOXMIT) {   /* UDP csum disabled */
-
-		skb->ip_summed = CHECKSUM_NONE;
-		goto send;
-
-	} else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */
-
-		udp4_hwcsum_outgoing(sk, skb, fl->fl4_src,fl->fl4_dst, up->len);
-		goto send;
-
-	} else						 /*   `normal' UDP    */
-		csum = udp_csum_outgoing(sk, skb);
-
-	/* add protocol-dependent pseudo-header */
-	uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len,
-				      sk->sk_protocol, csum             );
-	if (uh->check == 0)
-		uh->check = CSUM_MANGLED_0;
-
-send:
-	err = ip_push_pending_frames(sk);
-out:
-	up->len = 0;
-	up->pending = 0;
-	if (!err)
-		UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite);
-	return err;
-}
-
-int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
-		size_t len)
-{
-	struct inet_sock *inet = inet_sk(sk);
-	struct udp_sock *up = udp_sk(sk);
-	int ulen = len;
-	struct ipcm_cookie ipc;
-	struct rtable *rt = NULL;
-	int free = 0;
-	int connected = 0;
-	__be32 daddr, faddr, saddr;
-	__be16 dport;
-	u8  tos;
-	int err, is_udplite = IS_UDPLITE(sk);
-	int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
-	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
-
-	if (len > 0xFFFF)
-		return -EMSGSIZE;
-
-	/*
-	 *	Check the flags.
-	 */
-
-	if (msg->msg_flags&MSG_OOB)	/* Mirror BSD error message compatibility */
-		return -EOPNOTSUPP;
-
-	ipc.opt = NULL;
-
-	if (up->pending) {
-		/*
-		 * There are pending frames.
-		 * The socket lock must be held while it's corked.
-		 */
-		lock_sock(sk);
-		if (likely(up->pending)) {
-			if (unlikely(up->pending != AF_INET)) {
-				release_sock(sk);
-				return -EINVAL;
-			}
-			goto do_append_data;
-		}
-		release_sock(sk);
-	}
-	ulen += sizeof(struct udphdr);
-
-	/*
-	 *	Get and verify the address.
-	 */
-	if (msg->msg_name) {
-		struct sockaddr_in * usin = (struct sockaddr_in*)msg->msg_name;
-		if (msg->msg_namelen < sizeof(*usin))
-			return -EINVAL;
-		if (usin->sin_family != AF_INET) {
-			if (usin->sin_family != AF_UNSPEC)
-				return -EAFNOSUPPORT;
-		}
-
-		daddr = usin->sin_addr.s_addr;
-		dport = usin->sin_port;
-		if (dport == 0)
-			return -EINVAL;
-	} else {
-		if (sk->sk_state != TCP_ESTABLISHED)
-			return -EDESTADDRREQ;
-		daddr = inet->daddr;
-		dport = inet->dport;
-		/* Open fast path for connected socket.
-		   Route will not be used, if at least one option is set.
-		 */
-		connected = 1;
-	}
-	ipc.addr = inet->saddr;
-
-	ipc.oif = sk->sk_bound_dev_if;
-	if (msg->msg_controllen) {
-		err = ip_cmsg_send(msg, &ipc);
-		if (err)
-			return err;
-		if (ipc.opt)
-			free = 1;
-		connected = 0;
-	}
-	if (!ipc.opt)
-		ipc.opt = inet->opt;
-
-	saddr = ipc.addr;
-	ipc.addr = faddr = daddr;
-
-	if (ipc.opt && ipc.opt->srr) {
-		if (!daddr)
-			return -EINVAL;
-		faddr = ipc.opt->faddr;
-		connected = 0;
-	}
-	tos = RT_TOS(inet->tos);
-	if (sock_flag(sk, SOCK_LOCALROUTE) ||
-	    (msg->msg_flags & MSG_DONTROUTE) ||
-	    (ipc.opt && ipc.opt->is_strictroute)) {
-		tos |= RTO_ONLINK;
-		connected = 0;
-	}
-
-	if (ipv4_is_multicast(daddr)) {
-		if (!ipc.oif)
-			ipc.oif = inet->mc_index;
-		if (!saddr)
-			saddr = inet->mc_addr;
-		connected = 0;
-	}
-
-	if (connected)
-		rt = (struct rtable*)sk_dst_check(sk, 0);
-
-	if (rt == NULL) {
-		struct flowi fl = { .oif = ipc.oif,
-				    .nl_u = { .ip4_u =
-					      { .daddr = faddr,
-						.saddr = saddr,
-						.tos = tos } },
-				    .proto = sk->sk_protocol,
-				    .uli_u = { .ports =
-					       { .sport = inet->sport,
-						 .dport = dport } } };
-		security_sk_classify_flow(sk, &fl);
-		err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1);
-		if (err) {
-			if (err == -ENETUNREACH)
-				IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
-			goto out;
-		}
-
-		err = -EACCES;
-		if ((rt->rt_flags & RTCF_BROADCAST) &&
-		    !sock_flag(sk, SOCK_BROADCAST))
-			goto out;
-		if (connected)
-			sk_dst_set(sk, dst_clone(&rt->u.dst));
-	}
-
-	if (msg->msg_flags&MSG_CONFIRM)
-		goto do_confirm;
-back_from_confirm:
-
-	saddr = rt->rt_src;
-	if (!ipc.addr)
-		daddr = ipc.addr = rt->rt_dst;
-
-	lock_sock(sk);
-	if (unlikely(up->pending)) {
-		/* The socket is already corked while preparing it. */
-		/* ... which is an evident application bug. --ANK */
-		release_sock(sk);
-
-		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n");
-		err = -EINVAL;
-		goto out;
-	}
-	/*
-	 *	Now cork the socket to pend data.
-	 */
-	inet->cork.fl.fl4_dst = daddr;
-	inet->cork.fl.fl_ip_dport = dport;
-	inet->cork.fl.fl4_src = saddr;
-	inet->cork.fl.fl_ip_sport = inet->sport;
-	up->pending = AF_INET;
-
-do_append_data:
-	up->len += ulen;
-	getfrag  =  is_udplite ?  udplite_getfrag : ip_generic_getfrag;
-	err = ip_append_data(sk, getfrag, msg->msg_iov, ulen,
-			sizeof(struct udphdr), &ipc, rt,
-			corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
-	if (err)
-		udp_flush_pending_frames(sk);
-	else if (!corkreq)
-		err = udp_push_pending_frames(sk);
-	else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
-		up->pending = 0;
-	release_sock(sk);
-
-out:
-	ip_rt_put(rt);
-	if (free)
-		kfree(ipc.opt);
-	if (!err)
-		return len;
-	/*
-	 * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space.  Reporting
-	 * ENOBUFS might not be good (it's not tunable per se), but otherwise
-	 * we don't have a good statistic (IpOutDiscards but it can be too many
-	 * things).  We could add another new stat but at least for now that
-	 * seems like overkill.
-	 */
-	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
-		UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite);
-	}
-	return err;
-
-do_confirm:
-	dst_confirm(&rt->u.dst);
-	if (!(msg->msg_flags&MSG_PROBE) || len)
-		goto back_from_confirm;
-	err = 0;
-	goto out;
-}
-
-int udp_sendpage(struct sock *sk, struct page *page, int offset,
-		 size_t size, int flags)
-{
-	struct udp_sock *up = udp_sk(sk);
-	int ret;
-
-	if (!up->pending) {
-		struct msghdr msg = {	.msg_flags = flags|MSG_MORE };
-
-		/* Call udp_sendmsg to specify destination address which
-		 * sendpage interface can't pass.
-		 * This will succeed only when the socket is connected.
-		 */
-		ret = udp_sendmsg(NULL, sk, &msg, 0);
-		if (ret < 0)
-			return ret;
-	}
-
-	lock_sock(sk);
-
-	if (unlikely(!up->pending)) {
-		release_sock(sk);
-
-		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 3\n");
-		return -EINVAL;
-	}
-
-	ret = ip_append_page(sk, page, offset, size, flags);
-	if (ret == -EOPNOTSUPP) {
-		release_sock(sk);
-		return sock_no_sendpage(sk->sk_socket, page, offset,
-					size, flags);
-	}
-	if (ret < 0) {
-		udp_flush_pending_frames(sk);
-		goto out;
-	}
-
-	up->len += size;
-	if (!(up->corkflag || (flags&MSG_MORE)))
-		ret = udp_push_pending_frames(sk);
-	if (!ret)
-		ret = size;
-out:
-	release_sock(sk);
-	return ret;
-}
-
-/*
- * 	This should be easy, if there is something there we
- * 	return it, otherwise we block.
- */
-
-int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
-		size_t len, int noblock, int flags, int *addr_len)
-{
-	struct inet_sock *inet = inet_sk(sk);
-	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
-	struct sk_buff *skb;
-	unsigned int ulen, copied;
-	int peeked;
-	int err;
-	int is_udplite = IS_UDPLITE(sk);
-
-	/*
-	 *	Check any passed addresses
-	 */
-	if (addr_len)
-		*addr_len=sizeof(*sin);
-
-	if (flags & MSG_ERRQUEUE)
-		return ip_recv_error(sk, msg, len);
-
-try_again:
-	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
-				  &peeked, &err);
-	if (!skb)
-		goto out;
-
-	ulen = skb->len - sizeof(struct udphdr);
-	copied = len;
-	if (copied > ulen)
-		copied = ulen;
-	else if (copied < ulen)
-		msg->msg_flags |= MSG_TRUNC;
-
-	/*
-	 * If checksum is needed at all, try to do it while copying the
-	 * data.  If the data is truncated, or if we only want a partial
-	 * coverage checksum (UDP-Lite), do it before the copy.
-	 */
-
-	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
-		if (udp_lib_checksum_complete(skb))
-			goto csum_copy_err;
-	}
-
-	if (skb_csum_unnecessary(skb))
-		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
-					      msg->msg_iov, copied       );
-	else {
-		err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov);
-
-		if (err == -EINVAL)
-			goto csum_copy_err;
-	}
-
-	if (err)
-		goto out_free;
-
-	if (!peeked)
-		UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
-
-	sock_recv_timestamp(msg, sk, skb);
-
-	/* Copy the address. */
-	if (sin)
-	{
-		sin->sin_family = AF_INET;
-		sin->sin_port = udp_hdr(skb)->source;
-		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
-		memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
-	}
-	if (inet->cmsg_flags)
-		ip_cmsg_recv(msg, skb);
-
-	err = copied;
-	if (flags & MSG_TRUNC)
-		err = ulen;
-
-out_free:
-	lock_sock(sk);
-	skb_free_datagram(sk, skb);
-	release_sock(sk);
-out:
-	return err;
-
-csum_copy_err:
-	lock_sock(sk);
-	if (!skb_kill_datagram(sk, skb, flags))
-		UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
-	release_sock(sk);
-
-	if (noblock)
-		return -EAGAIN;
-	goto try_again;
-}
-
-
-/* returns:
- *  -1: error
- *   0: success
- *  >0: "udp encap" protocol resubmission
- *
- * Note that in the success and error cases, the skb is assumed to
- * have either been requeued or freed.
- */
-int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
-{
-	struct udp_sock *up = udp_sk(sk);
-	int rc;
-	int is_udplite = IS_UDPLITE(sk);
-
-	/*
-	 *	Charge it to the socket, dropping if the queue is full.
-	 */
-	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
-		goto drop;
-	nf_reset(skb);
-
-	if (up->encap_type) {
-		/*
-		 * This is an encapsulation socket so pass the skb to
-		 * the socket's udp_encap_rcv() hook. Otherwise, just
-		 * fall through and pass this up the UDP socket.
-		 * up->encap_rcv() returns the following value:
-		 * =0 if skb was successfully passed to the encap
-		 *    handler or was discarded by it.
-		 * >0 if skb should be passed on to UDP.
-		 * <0 if skb should be resubmitted as proto -N
-		 */
-
-		/* if we're overly short, let UDP handle it */
-		if (skb->len > sizeof(struct udphdr) &&
-		    up->encap_rcv != NULL) {
-			int ret;
-
-			ret = (*up->encap_rcv)(sk, skb);
-			if (ret <= 0) {
-				UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS,
-						 is_udplite);
-				return -ret;
-			}
-		}
-
-		/* FALLTHROUGH -- it's a UDP Packet */
-	}
-
-	/*
-	 * 	UDP-Lite specific tests, ignored on UDP sockets
-	 */
-	if ((is_udplite & UDPLITE_RECV_CC)  &&  UDP_SKB_CB(skb)->partial_cov) {
-
-		/*
-		 * MIB statistics other than incrementing the error count are
-		 * disabled for the following two types of errors: these depend
-		 * on the application settings, not on the functioning of the
-		 * protocol stack as such.
-		 *
-		 * RFC 3828 here recommends (sec 3.3): "There should also be a
-		 * way ... to ... at least let the receiving application block
-		 * delivery of packets with coverage values less than a value
-		 * provided by the application."
-		 */
-		if (up->pcrlen == 0) {          /* full coverage was set  */
-			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE: partial coverage "
-				"%d while full coverage %d requested\n",
-				UDP_SKB_CB(skb)->cscov, skb->len);
-			goto drop;
-		}
-		/* The next case involves violating the min. coverage requested
-		 * by the receiver. This is subtle: if receiver wants x and x is
-		 * greater than the buffersize/MTU then receiver will complain
-		 * that it wants x while sender emits packets of smaller size y.
-		 * Therefore the above ...()->partial_cov statement is essential.
-		 */
-		if (UDP_SKB_CB(skb)->cscov  <  up->pcrlen) {
-			LIMIT_NETDEBUG(KERN_WARNING
-				"UDPLITE: coverage %d too small, need min %d\n",
-				UDP_SKB_CB(skb)->cscov, up->pcrlen);
-			goto drop;
-		}
-	}
-
-	if (sk->sk_filter) {
-		if (udp_lib_checksum_complete(skb))
-			goto drop;
-	}
-
-	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
-		/* Note that an ENOMEM error is charged twice */
-		if (rc == -ENOMEM)
-			UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
-		goto drop;
-	}
-
-	return 0;
-
-drop:
-	UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite);
-	kfree_skb(skb);
-	return -1;
-}
-
-/*
- *	Multicasts and broadcasts go to each listener.
- *
- *	Note: called only from the BH handler context,
- *	so we don't need to lock the hashes.
- */
-static int __udp4_lib_mcast_deliver(struct sk_buff *skb,
-				    struct udphdr  *uh,
-				    __be32 saddr, __be32 daddr,
-				    struct hlist_head udptable[])
-{
-	struct sock *sk;
-	int dif;
-
-	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
-	dif = skb->dev->ifindex;
-	sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
-	if (sk) {
-		struct sock *sknext = NULL;
-
-		do {
-			struct sk_buff *skb1 = skb;
-
-			sknext = udp_v4_mcast_next(sk_next(sk), uh->dest, daddr,
-						   uh->source, saddr, dif);
-			if (sknext)
-				skb1 = skb_clone(skb, GFP_ATOMIC);
-
-			if (skb1) {
-				int ret = 0;
-
-				bh_lock_sock_nested(sk);
-				if (!sock_owned_by_user(sk))
-					ret = udp_queue_rcv_skb(sk, skb1);
-				else
-					sk_add_backlog(sk, skb1);
-				bh_unlock_sock(sk);
-
-				if (ret > 0)
-					/* we should probably re-process instead
-					 * of dropping packets here. */
-					kfree_skb(skb1);
-			}
-			sk = sknext;
-		} while (sknext);
-	} else
-		kfree_skb(skb);
-	read_unlock(&udp_hash_lock);
-	return 0;
-}
-
-/* Initialize UDP checksum. If exited with zero value (success),
- * CHECKSUM_UNNECESSARY means, that no more checks are required.
- * Otherwise, csum completion requires chacksumming packet body,
- * including udp header and folding it to skb->csum.
- */
-static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh,
-				 int proto)
-{
-	const struct iphdr *iph;
-	int err;
-
-	UDP_SKB_CB(skb)->partial_cov = 0;
-	UDP_SKB_CB(skb)->cscov = skb->len;
-
-	if (IS_PROTO_UDPLITE(proto)) {
-		err = udplite_checksum_init(skb, uh);
-		if (err)
-			return err;
-	}
-
-	iph = ip_hdr(skb);
-	if (uh->check == 0) {
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-	} else if (skb->ip_summed == CHECKSUM_COMPLETE) {
-	       if (!csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len,
-				      proto, skb->csum))
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-	}
-	if (!skb_csum_unnecessary(skb))
-		skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
-					       skb->len, proto, 0);
-	/* Probably, we should checksum udp header (it should be in cache
-	 * in any case) and data in tiny packets (< rx copybreak).
-	 */
-
-	return 0;
-}
-
-/*
- *	All we need to do is get the socket, and then do a checksum.
- */
-
-int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
-		   int proto)
-{
-	struct sock *sk;
-	struct udphdr *uh = udp_hdr(skb);
-	unsigned short ulen;
-	struct rtable *rt = skb->rtable;
-	__be32 saddr = ip_hdr(skb)->saddr;
-	__be32 daddr = ip_hdr(skb)->daddr;
-
-	/*
-	 *  Validate the packet.
-	 */
-	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
-		goto drop;		/* No space for header. */
-
-	ulen = ntohs(uh->len);
-	if (ulen > skb->len)
-		goto short_packet;
-
-	if (IS_PROTO_UDPLITE(proto)) {
-		/* UDP validates ulen. */
-		if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
-			goto short_packet;
-		uh = udp_hdr(skb);
-	}
-
-	if (udp4_csum_init(skb, uh, proto))
-		goto csum_error;
-
-	if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
-		return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable);
-
-	sk = __udp4_lib_lookup(skb->dev->nd_net, saddr, uh->source, daddr,
-			uh->dest, inet_iif(skb), udptable);
-
-	if (sk != NULL) {
-		int ret = 0;
-		bh_lock_sock_nested(sk);
-		if (!sock_owned_by_user(sk))
-			ret = udp_queue_rcv_skb(sk, skb);
-		else
-			sk_add_backlog(sk, skb);
-		bh_unlock_sock(sk);
-		sock_put(sk);
-
-		/* a return value > 0 means to resubmit the input, but
-		 * it wants the return to be -protocol, or 0
-		 */
-		if (ret > 0)
-			return -ret;
-		return 0;
-	}
-
-	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
-		goto drop;
-	nf_reset(skb);
-
-	/* No socket. Drop packet silently, if checksum is wrong */
-	if (udp_lib_checksum_complete(skb))
-		goto csum_error;
-
-	UDP_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto));
-	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
-
-	/*
-	 * Hmm.  We got an UDP packet to a port to which we
-	 * don't wanna listen.  Ignore it.
-	 */
-	kfree_skb(skb);
-	return 0;
-
-short_packet:
-	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n",
-		       IS_PROTO_UDPLITE(proto) ? "-Lite" : "",
-		       NIPQUAD(saddr),
-		       ntohs(uh->source),
-		       ulen,
-		       skb->len,
-		       NIPQUAD(daddr),
-		       ntohs(uh->dest));
-	goto drop;
-
-csum_error:
-	/*
-	 * RFC1122: OK.  Discards the bad packet silently (as far as
-	 * the network is concerned, anyway) as per 4.1.3.4 (MUST).
-	 */
-	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n",
-		       IS_PROTO_UDPLITE(proto) ? "-Lite" : "",
-		       NIPQUAD(saddr),
-		       ntohs(uh->source),
-		       NIPQUAD(daddr),
-		       ntohs(uh->dest),
-		       ulen);
-drop:
-	UDP_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto));
-	kfree_skb(skb);
-	return 0;
-}
-
-int udp_rcv(struct sk_buff *skb)
-{
-	return __udp4_lib_rcv(skb, udp_hash, IPPROTO_UDP);
-}
-
-int udp_destroy_sock(struct sock *sk)
-{
-	lock_sock(sk);
-	udp_flush_pending_frames(sk);
-	release_sock(sk);
-	return 0;
-}
-
-int udp_setsockopt(struct sock *sk, int level, int optname,
-		   char __user *optval, int optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
-					  udp_push_pending_frames);
-	return ip_setsockopt(sk, level, optname, optval, optlen);
-}
-
-#ifdef CONFIG_COMPAT
-int compat_udp_setsockopt(struct sock *sk, int level, int optname,
-			  char __user *optval, int optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
-					  udp_push_pending_frames);
-	return compat_ip_setsockopt(sk, level, optname, optval, optlen);
-}
-#endif
-
-int udp_getsockopt(struct sock *sk, int level, int optname,
-		   char __user *optval, int __user *optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
-	return ip_getsockopt(sk, level, optname, optval, optlen);
-}
-
-#ifdef CONFIG_COMPAT
-int compat_udp_getsockopt(struct sock *sk, int level, int optname,
-			  char __user *optval, int __user *optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
-	return compat_ip_getsockopt(sk, level, optname, optval, optlen);
-}
-#endif
-
-/* ------------------------------------------------------------------------ */
-DEFINE_PROTO_INUSE(udp)
-
-struct proto udp_prot = {
-	.name		   = "UDP",
-	.owner		   = THIS_MODULE,
-	.close		   = udp_lib_close,
-	.connect	   = ip4_datagram_connect,
-	.disconnect	   = udp_disconnect,
-	.ioctl		   = udp_ioctl,
-	.destroy	   = udp_destroy_sock,
-	.setsockopt	   = udp_setsockopt,
-	.getsockopt	   = udp_getsockopt,
-	.sendmsg	   = udp_sendmsg,
-	.recvmsg	   = udp_recvmsg,
-	.sendpage	   = udp_sendpage,
-	.backlog_rcv	   = udp_queue_rcv_skb,
-	.hash		   = udp_lib_hash,
-	.unhash		   = udp_lib_unhash,
-	.get_port	   = udp_v4_get_port,
-	.memory_allocated  = &udp_memory_allocated,
-	.sysctl_mem	   = sysctl_udp_mem,
-	.sysctl_wmem	   = &sysctl_udp_wmem_min,
-	.sysctl_rmem	   = &sysctl_udp_rmem_min,
-	.obj_size	   = sizeof(struct udp_sock),
-#ifdef CONFIG_COMPAT
-	.compat_setsockopt = compat_udp_setsockopt,
-	.compat_getsockopt = compat_udp_getsockopt,
-#endif
-	REF_PROTO_INUSE(udp)
-};
-
-/* ------------------------------------------------------------------------ */
-static void udp4_format_sock(struct sock *sp, char *tmpbuf, int bucket)
-{
-	struct inet_sock *inet = inet_sk(sp);
-	__be32 dest = inet->daddr;
-	__be32 src  = inet->rcv_saddr;
-	__u16 destp	  = ntohs(inet->dport);
-	__u16 srcp	  = ntohs(inet->sport);
-
-	sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
-		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p",
-		bucket, src, srcp, dest, destp, sp->sk_state,
-		atomic_read(&sp->sk_wmem_alloc),
-		atomic_read(&sp->sk_rmem_alloc),
-		0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp),
-		atomic_read(&sp->sk_refcnt), sp);
-}
-
-int udp4_seq_show(struct seq_file *seq, void *v)
-{
-	if (v == SEQ_START_TOKEN)
-		seq_printf(seq, "%-127s\n",
-			   "  sl  local_address rem_address   st tx_queue "
-			   "rx_queue tr tm->when retrnsmt   uid  timeout "
-			   "inode");
-	else {
-		char tmpbuf[129];
-		struct udp_iter_state *state = seq->private;
-
-		udp4_format_sock(v, tmpbuf, state->bucket);
-		seq_printf(seq, "%-127s\n", tmpbuf);
-	}
-	return 0;
-}
-
-/* ------------------------------------------------------------------------ */
-#ifdef CONFIG_PROC_FS
-static struct file_operations udp4_seq_fops;
-static struct udp_seq_afinfo udp4_seq_afinfo = {
-	.owner		= THIS_MODULE,
-	.name		= "udp",
-	.family		= AF_INET,
-	.hashtable	= udp_hash,
-	.seq_show	= udp4_seq_show,
-	.seq_fops	= &udp4_seq_fops,
-};
-
-int __init udp4_proc_init(void)
-{
-	return udp_proc_register(&udp4_seq_afinfo);
-}
-
-void udp4_proc_exit(void)
-{
-	udp_proc_unregister(&udp4_seq_afinfo);
-}
-#endif /* CONFIG_PROC_FS */
-
-EXPORT_SYMBOL(udp_prot);
-EXPORT_SYMBOL(udp_sendmsg);
-
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
new file mode 100644
index 000000000000..d49c6d68c8a9
--- /dev/null
+++ b/net/ipv4/udplite.c
@@ -0,0 +1,121 @@
+/*
+ *  UDPLITE     An implementation of the UDP-Lite protocol (RFC 3828).
+ *
+ *  Version:    $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $
+ *
+ *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
+ *
+ *  Changes:
+ *  Fixes:
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ */
+#include "udp_impl.h"
+DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics)	__read_mostly;
+
+struct hlist_head 	udplite_hash[UDP_HTABLE_SIZE];
+
+int udplite_get_port(struct sock *sk, unsigned short p,
+		     int (*c)(const struct sock *, const struct sock *))
+{
+	return  __udp_lib_get_port(sk, p, udplite_hash, c);
+}
+
+static int udplite_v4_get_port(struct sock *sk, unsigned short snum)
+{
+	return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal);
+}
+
+static int udplite_rcv(struct sk_buff *skb)
+{
+	return __udp4_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE);
+}
+
+static void udplite_err(struct sk_buff *skb, u32 info)
+{
+	__udp4_lib_err(skb, info, udplite_hash);
+}
+
+static	struct net_protocol udplite_protocol = {
+	.handler	= udplite_rcv,
+	.err_handler	= udplite_err,
+	.no_policy	= 1,
+};
+
+DEFINE_PROTO_INUSE(udplite)
+
+struct proto 	udplite_prot = {
+	.name		   = "UDP-Lite",
+	.owner		   = THIS_MODULE,
+	.close		   = udp_lib_close,
+	.connect	   = ip4_datagram_connect,
+	.disconnect	   = udp_disconnect,
+	.ioctl		   = udp_ioctl,
+	.init		   = udplite_sk_init,
+	.destroy	   = udp_destroy_sock,
+	.setsockopt	   = udp_setsockopt,
+	.getsockopt	   = udp_getsockopt,
+	.sendmsg	   = udp_sendmsg,
+	.recvmsg	   = udp_recvmsg,
+	.sendpage	   = udp_sendpage,
+	.backlog_rcv	   = udp_queue_rcv_skb,
+	.hash		   = udp_lib_hash,
+	.unhash		   = udp_lib_unhash,
+	.get_port	   = udplite_v4_get_port,
+	.obj_size	   = sizeof(struct udp_sock),
+#ifdef CONFIG_COMPAT
+	.compat_setsockopt = compat_udp_setsockopt,
+	.compat_getsockopt = compat_udp_getsockopt,
+#endif
+	REF_PROTO_INUSE(udplite)
+};
+
+static struct inet_protosw udplite4_protosw = {
+	.type		=  SOCK_DGRAM,
+	.protocol	=  IPPROTO_UDPLITE,
+	.prot		=  &udplite_prot,
+	.ops		=  &inet_dgram_ops,
+	.capability	= -1,
+	.no_check	=  0,		/* must checksum (RFC 3828) */
+	.flags		=  INET_PROTOSW_PERMANENT,
+};
+
+#ifdef CONFIG_PROC_FS
+static struct file_operations udplite4_seq_fops;
+static struct udp_seq_afinfo udplite4_seq_afinfo = {
+	.owner		= THIS_MODULE,
+	.name		= "udplite",
+	.family		= AF_INET,
+	.hashtable	= udplite_hash,
+	.seq_show	= udp4_seq_show,
+	.seq_fops	= &udplite4_seq_fops,
+};
+#endif
+
+void __init udplite4_register(void)
+{
+	if (proto_register(&udplite_prot, 1))
+		goto out_register_err;
+
+	if (inet_add_protocol(&udplite_protocol, IPPROTO_UDPLITE) < 0)
+		goto out_unregister_proto;
+
+	inet_register_protosw(&udplite4_protosw);
+
+#ifdef CONFIG_PROC_FS
+	if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */
+		printk(KERN_ERR "%s: Cannot register /proc!\n", __func__);
+#endif
+	return;
+
+out_unregister_proto:
+	proto_unregister(&udplite_prot);
+out_register_err:
+	printk(KERN_CRIT "%s: Cannot add UDP-Lite protocol.\n", __func__);
+}
+
+EXPORT_SYMBOL(udplite_hash);
+EXPORT_SYMBOL(udplite_prot);
+EXPORT_SYMBOL(udplite_get_port);
diff --git a/net/ipv4/udplite_ipv4.c b/net/ipv4/udplite_ipv4.c
deleted file mode 100644
index d49c6d68c8a9..000000000000
--- a/net/ipv4/udplite_ipv4.c
+++ /dev/null
@@ -1,121 +0,0 @@
-/*
- *  UDPLITE     An implementation of the UDP-Lite protocol (RFC 3828).
- *
- *  Version:    $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $
- *
- *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
- *
- *  Changes:
- *  Fixes:
- *		This program is free software; you can redistribute it and/or
- *		modify it under the terms of the GNU General Public License
- *		as published by the Free Software Foundation; either version
- *		2 of the License, or (at your option) any later version.
- */
-#include "udp_impl.h"
-DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics)	__read_mostly;
-
-struct hlist_head 	udplite_hash[UDP_HTABLE_SIZE];
-
-int udplite_get_port(struct sock *sk, unsigned short p,
-		     int (*c)(const struct sock *, const struct sock *))
-{
-	return  __udp_lib_get_port(sk, p, udplite_hash, c);
-}
-
-static int udplite_v4_get_port(struct sock *sk, unsigned short snum)
-{
-	return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal);
-}
-
-static int udplite_rcv(struct sk_buff *skb)
-{
-	return __udp4_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE);
-}
-
-static void udplite_err(struct sk_buff *skb, u32 info)
-{
-	__udp4_lib_err(skb, info, udplite_hash);
-}
-
-static	struct net_protocol udplite_protocol = {
-	.handler	= udplite_rcv,
-	.err_handler	= udplite_err,
-	.no_policy	= 1,
-};
-
-DEFINE_PROTO_INUSE(udplite)
-
-struct proto 	udplite_prot = {
-	.name		   = "UDP-Lite",
-	.owner		   = THIS_MODULE,
-	.close		   = udp_lib_close,
-	.connect	   = ip4_datagram_connect,
-	.disconnect	   = udp_disconnect,
-	.ioctl		   = udp_ioctl,
-	.init		   = udplite_sk_init,
-	.destroy	   = udp_destroy_sock,
-	.setsockopt	   = udp_setsockopt,
-	.getsockopt	   = udp_getsockopt,
-	.sendmsg	   = udp_sendmsg,
-	.recvmsg	   = udp_recvmsg,
-	.sendpage	   = udp_sendpage,
-	.backlog_rcv	   = udp_queue_rcv_skb,
-	.hash		   = udp_lib_hash,
-	.unhash		   = udp_lib_unhash,
-	.get_port	   = udplite_v4_get_port,
-	.obj_size	   = sizeof(struct udp_sock),
-#ifdef CONFIG_COMPAT
-	.compat_setsockopt = compat_udp_setsockopt,
-	.compat_getsockopt = compat_udp_getsockopt,
-#endif
-	REF_PROTO_INUSE(udplite)
-};
-
-static struct inet_protosw udplite4_protosw = {
-	.type		=  SOCK_DGRAM,
-	.protocol	=  IPPROTO_UDPLITE,
-	.prot		=  &udplite_prot,
-	.ops		=  &inet_dgram_ops,
-	.capability	= -1,
-	.no_check	=  0,		/* must checksum (RFC 3828) */
-	.flags		=  INET_PROTOSW_PERMANENT,
-};
-
-#ifdef CONFIG_PROC_FS
-static struct file_operations udplite4_seq_fops;
-static struct udp_seq_afinfo udplite4_seq_afinfo = {
-	.owner		= THIS_MODULE,
-	.name		= "udplite",
-	.family		= AF_INET,
-	.hashtable	= udplite_hash,
-	.seq_show	= udp4_seq_show,
-	.seq_fops	= &udplite4_seq_fops,
-};
-#endif
-
-void __init udplite4_register(void)
-{
-	if (proto_register(&udplite_prot, 1))
-		goto out_register_err;
-
-	if (inet_add_protocol(&udplite_protocol, IPPROTO_UDPLITE) < 0)
-		goto out_unregister_proto;
-
-	inet_register_protosw(&udplite4_protosw);
-
-#ifdef CONFIG_PROC_FS
-	if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */
-		printk(KERN_ERR "%s: Cannot register /proc!\n", __func__);
-#endif
-	return;
-
-out_unregister_proto:
-	proto_unregister(&udplite_prot);
-out_register_err:
-	printk(KERN_CRIT "%s: Cannot add UDP-Lite protocol.\n", __func__);
-}
-
-EXPORT_SYMBOL(udplite_hash);
-EXPORT_SYMBOL(udplite_prot);
-EXPORT_SYMBOL(udplite_get_port);
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index 107051f7c227..ae14617e607f 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -6,7 +6,7 @@ obj-$(CONFIG_IPV6) += ipv6.o
 
 ipv6-objs :=	af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \
 		addrlabel.o \
-		route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp_ipv6.o \
+		route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \
 		raw.o protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \
 		exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o
 
@@ -17,7 +17,6 @@ ipv6-$(CONFIG_NETFILTER) += netfilter.o
 ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o
 ipv6-$(CONFIG_PROC_FS) += proc.o
 ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o
-ipv6-$(CONFIG_IP_UDPLITE) += udplite_ipv6.o
 
 ipv6-objs += $(ipv6-y)
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index afe9276d0420..730a861b8f41 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -813,16 +813,12 @@ static int __init init_ipv6_mibs(void)
 		goto err_icmpmsg_mib;
 	if (snmp_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib)) < 0)
 		goto err_udp_mib;
-#ifdef CONFIG_IP_UDPLITE
 	if (snmp_mib_init((void **)udplite_stats_in6,
 			  sizeof (struct udp_mib)) < 0)
 		goto err_udplite_mib;
-#endif
 	return 0;
 
-#ifdef CONFIG_IP_UDPLITE
 err_udplite_mib:
-#endif
 	snmp_mib_free((void **)udp_stats_in6);
 err_udp_mib:
 	snmp_mib_free((void **)icmpv6msg_statistics);
@@ -841,9 +837,7 @@ static void cleanup_ipv6_mibs(void)
 	snmp_mib_free((void **)icmpv6_statistics);
 	snmp_mib_free((void **)icmpv6msg_statistics);
 	snmp_mib_free((void **)udp_stats_in6);
-#ifdef CONFIG_IP_UDPLITE
 	snmp_mib_free((void **)udplite_stats_in6);
-#endif
 }
 
 static int inet6_net_init(struct net *net)
@@ -888,11 +882,9 @@ static int __init inet6_init(void)
 	if (err)
 		goto out_unregister_tcp_proto;
 
-#ifdef CONFIG_IP_UDPLITE
 	err = proto_register(&udplitev6_prot, 1);
 	if (err)
 		goto out_unregister_udp_proto;
-#endif
 
 	err = proto_register(&rawv6_prot, 1);
 	if (err)
@@ -1063,10 +1055,8 @@ out_sock_register_fail:
 out_unregister_raw_proto:
 	proto_unregister(&rawv6_prot);
 out_unregister_udplite_proto:
-#ifdef CONFIG_IP_UDPLITE
 	proto_unregister(&udplitev6_prot);
 out_unregister_udp_proto:
-#endif
 	proto_unregister(&udpv6_prot);
 out_unregister_tcp_proto:
 	proto_unregister(&tcpv6_prot);
@@ -1085,9 +1075,7 @@ static void __exit inet6_exit(void)
 	ipv6_sysctl_unregister();
 #endif
 	udpv6_exit();
-#ifdef CONFIG_IP_UDPLITE
 	udplitev6_exit();
-#endif
 	tcpv6_exit();
 
 	/* Cleanup code parts. */
@@ -1117,9 +1105,7 @@ static void __exit inet6_exit(void)
 	unregister_pernet_subsys(&inet6_net_ops);
 	cleanup_ipv6_mibs();
 	proto_unregister(&rawv6_prot);
-#ifdef CONFIG_IP_UDPLITE
 	proto_unregister(&udplitev6_prot);
-#endif
 	proto_unregister(&udpv6_prot);
 	proto_unregister(&tcpv6_prot);
 }
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 3bbfdff698d2..5eea6fa506e5 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -127,9 +127,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 			struct sk_buff *pktopt;
 
 			if (sk->sk_protocol != IPPROTO_UDP &&
-#ifdef CONFIG_IP_UDPLITE
 			    sk->sk_protocol != IPPROTO_UDPLITE &&
-#endif
 			    sk->sk_protocol != IPPROTO_TCP)
 				break;
 
@@ -169,7 +167,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 			} else {
 				struct proto *prot = &udp_prot;
 
-				if (IS_PROTO_UDPLITE(sk->sk_protocol))
+				if (sk->sk_protocol == IPPROTO_UDPLITE)
 					prot = &udplite_prot;
 				local_bh_disable();
 				sock_prot_inuse_add(sk->sk_prot, -1);
@@ -734,9 +732,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
 	switch (optname) {
 	case IPV6_ADDRFORM:
 		if (sk->sk_protocol != IPPROTO_UDP &&
-#ifdef CONFIG_IP_UDPLITE
 		    sk->sk_protocol != IPPROTO_UDPLITE &&
-#endif
 		    sk->sk_protocol != IPPROTO_TCP)
 			return -EINVAL;
 		if (sk->sk_state != TCP_ESTABLISHED)
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 2453f2229ef7..8a5be290c710 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -39,10 +39,8 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v)
 		       sock_prot_inuse_get(&tcpv6_prot));
 	seq_printf(seq, "UDP6: inuse %d\n",
 		       sock_prot_inuse_get(&udpv6_prot));
-#ifdef CONFIG_IP_UDPLITE
 	seq_printf(seq, "UDPLITE6: inuse %d\n",
 			sock_prot_inuse_get(&udplitev6_prot));
-#endif
 	seq_printf(seq, "RAW6: inuse %d\n",
 		       sock_prot_inuse_get(&rawv6_prot));
 	seq_printf(seq, "FRAG6: inuse %d memory %d\n",
@@ -113,7 +111,6 @@ static struct snmp_mib snmp6_udp6_list[] = {
 	SNMP_MIB_SENTINEL
 };
 
-#ifdef CONFIG_IP_UDPLITE
 static struct snmp_mib snmp6_udplite6_list[] = {
 	SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS),
 	SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS),
@@ -121,7 +118,6 @@ static struct snmp_mib snmp6_udplite6_list[] = {
 	SNMP_MIB_ITEM("UdpLite6OutDatagrams", UDP_MIB_OUTDATAGRAMS),
 	SNMP_MIB_SENTINEL
 };
-#endif
 
 static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib)
 {
@@ -180,9 +176,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v)
 		snmp6_seq_show_item(seq, (void **)icmpv6_statistics, snmp6_icmp6_list);
 		snmp6_seq_show_icmpv6msg(seq, (void **)icmpv6msg_statistics);
 		snmp6_seq_show_item(seq, (void **)udp_stats_in6, snmp6_udp6_list);
-#ifdef CONFIG_IP_UDPLITE
 		snmp6_seq_show_item(seq, (void **)udplite_stats_in6, snmp6_udplite6_list);
-#endif
 	}
 	return 0;
 }
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
new file mode 100644
index 000000000000..53739de829db
--- /dev/null
+++ b/net/ipv6/udp.c
@@ -0,0 +1,1065 @@
+/*
+ *	UDP over IPv6
+ *	Linux INET6 implementation
+ *
+ *	Authors:
+ *	Pedro Roque		<roque@di.fc.ul.pt>
+ *
+ *	Based on linux/ipv4/udp.c
+ *
+ *	$Id: udp.c,v 1.65 2002/02/01 22:01:04 davem Exp $
+ *
+ *	Fixes:
+ *	Hideaki YOSHIFUJI	:	sin6_scope_id support
+ *	YOSHIFUJI Hideaki @USAGI and:	Support IPV6_V6ONLY socket option, which
+ *	Alexey Kuznetsov		allow both IPv4 and IPv6 sockets to bind
+ *					a single port at the same time.
+ *      Kazunori MIYAZAWA @USAGI:       change process style to use ip6_append_data
+ *      YOSHIFUJI Hideaki @USAGI:	convert /proc/net/udp6 to seq_file.
+ *
+ *	This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/sockios.h>
+#include <linux/net.h>
+#include <linux/in6.h>
+#include <linux/netdevice.h>
+#include <linux/if_arp.h>
+#include <linux/ipv6.h>
+#include <linux/icmpv6.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <asm/uaccess.h>
+
+#include <net/ndisc.h>
+#include <net/protocol.h>
+#include <net/transp_v6.h>
+#include <net/ip6_route.h>
+#include <net/raw.h>
+#include <net/tcp_states.h>
+#include <net/ip6_checksum.h>
+#include <net/xfrm.h>
+
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include "udp_impl.h"
+
+static inline int udp_v6_get_port(struct sock *sk, unsigned short snum)
+{
+	return udp_get_port(sk, snum, ipv6_rcv_saddr_equal);
+}
+
+static struct sock *__udp6_lib_lookup(struct net *net,
+				      struct in6_addr *saddr, __be16 sport,
+				      struct in6_addr *daddr, __be16 dport,
+				      int dif, struct hlist_head udptable[])
+{
+	struct sock *sk, *result = NULL;
+	struct hlist_node *node;
+	unsigned short hnum = ntohs(dport);
+	int badness = -1;
+
+	read_lock(&udp_hash_lock);
+	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
+		struct inet_sock *inet = inet_sk(sk);
+
+		if (sk->sk_net == net && sk->sk_hash == hnum &&
+				sk->sk_family == PF_INET6) {
+			struct ipv6_pinfo *np = inet6_sk(sk);
+			int score = 0;
+			if (inet->dport) {
+				if (inet->dport != sport)
+					continue;
+				score++;
+			}
+			if (!ipv6_addr_any(&np->rcv_saddr)) {
+				if (!ipv6_addr_equal(&np->rcv_saddr, daddr))
+					continue;
+				score++;
+			}
+			if (!ipv6_addr_any(&np->daddr)) {
+				if (!ipv6_addr_equal(&np->daddr, saddr))
+					continue;
+				score++;
+			}
+			if (sk->sk_bound_dev_if) {
+				if (sk->sk_bound_dev_if != dif)
+					continue;
+				score++;
+			}
+			if (score == 4) {
+				result = sk;
+				break;
+			} else if (score > badness) {
+				result = sk;
+				badness = score;
+			}
+		}
+	}
+	if (result)
+		sock_hold(result);
+	read_unlock(&udp_hash_lock);
+	return result;
+}
+
+/*
+ * 	This should be easy, if there is something there we
+ * 	return it, otherwise we block.
+ */
+
+int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk,
+		  struct msghdr *msg, size_t len,
+		  int noblock, int flags, int *addr_len)
+{
+	struct ipv6_pinfo *np = inet6_sk(sk);
+	struct inet_sock *inet = inet_sk(sk);
+	struct sk_buff *skb;
+	unsigned int ulen, copied;
+	int peeked;
+	int err;
+	int is_udplite = IS_UDPLITE(sk);
+
+	if (addr_len)
+		*addr_len=sizeof(struct sockaddr_in6);
+
+	if (flags & MSG_ERRQUEUE)
+		return ipv6_recv_error(sk, msg, len);
+
+try_again:
+	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
+				  &peeked, &err);
+	if (!skb)
+		goto out;
+
+	ulen = skb->len - sizeof(struct udphdr);
+	copied = len;
+	if (copied > ulen)
+		copied = ulen;
+	else if (copied < ulen)
+		msg->msg_flags |= MSG_TRUNC;
+
+	/*
+	 * If checksum is needed at all, try to do it while copying the
+	 * data.  If the data is truncated, or if we only want a partial
+	 * coverage checksum (UDP-Lite), do it before the copy.
+	 */
+
+	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
+		if (udp_lib_checksum_complete(skb))
+			goto csum_copy_err;
+	}
+
+	if (skb_csum_unnecessary(skb))
+		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
+					      msg->msg_iov, copied       );
+	else {
+		err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov);
+		if (err == -EINVAL)
+			goto csum_copy_err;
+	}
+	if (err)
+		goto out_free;
+
+	if (!peeked)
+		UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
+
+	sock_recv_timestamp(msg, sk, skb);
+
+	/* Copy the address. */
+	if (msg->msg_name) {
+		struct sockaddr_in6 *sin6;
+
+		sin6 = (struct sockaddr_in6 *) msg->msg_name;
+		sin6->sin6_family = AF_INET6;
+		sin6->sin6_port = udp_hdr(skb)->source;
+		sin6->sin6_flowinfo = 0;
+		sin6->sin6_scope_id = 0;
+
+		if (skb->protocol == htons(ETH_P_IP))
+			ipv6_addr_set(&sin6->sin6_addr, 0, 0,
+				      htonl(0xffff), ip_hdr(skb)->saddr);
+		else {
+			ipv6_addr_copy(&sin6->sin6_addr,
+				       &ipv6_hdr(skb)->saddr);
+			if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL)
+				sin6->sin6_scope_id = IP6CB(skb)->iif;
+		}
+
+	}
+	if (skb->protocol == htons(ETH_P_IP)) {
+		if (inet->cmsg_flags)
+			ip_cmsg_recv(msg, skb);
+	} else {
+		if (np->rxopt.all)
+			datagram_recv_ctl(sk, msg, skb);
+	}
+
+	err = copied;
+	if (flags & MSG_TRUNC)
+		err = ulen;
+
+out_free:
+	lock_sock(sk);
+	skb_free_datagram(sk, skb);
+	release_sock(sk);
+out:
+	return err;
+
+csum_copy_err:
+	lock_sock(sk);
+	if (!skb_kill_datagram(sk, skb, flags))
+		UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
+	release_sock(sk);
+
+	if (flags & MSG_DONTWAIT)
+		return -EAGAIN;
+	goto try_again;
+}
+
+void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
+		    int type, int code, int offset, __be32 info,
+		    struct hlist_head udptable[]                    )
+{
+	struct ipv6_pinfo *np;
+	struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data;
+	struct in6_addr *saddr = &hdr->saddr;
+	struct in6_addr *daddr = &hdr->daddr;
+	struct udphdr *uh = (struct udphdr*)(skb->data+offset);
+	struct sock *sk;
+	int err;
+
+	sk = __udp6_lib_lookup(skb->dev->nd_net, daddr, uh->dest,
+			       saddr, uh->source, inet6_iif(skb), udptable);
+	if (sk == NULL)
+		return;
+
+	np = inet6_sk(sk);
+
+	if (!icmpv6_err_convert(type, code, &err) && !np->recverr)
+		goto out;
+
+	if (sk->sk_state != TCP_ESTABLISHED && !np->recverr)
+		goto out;
+
+	if (np->recverr)
+		ipv6_icmp_error(sk, skb, err, uh->dest, ntohl(info), (u8 *)(uh+1));
+
+	sk->sk_err = err;
+	sk->sk_error_report(sk);
+out:
+	sock_put(sk);
+}
+
+static __inline__ void udpv6_err(struct sk_buff *skb,
+				 struct inet6_skb_parm *opt, int type,
+				 int code, int offset, __be32 info     )
+{
+	__udp6_lib_err(skb, opt, type, code, offset, info, udp_hash);
+}
+
+int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
+{
+	struct udp_sock *up = udp_sk(sk);
+	int rc;
+	int is_udplite = IS_UDPLITE(sk);
+
+	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
+		goto drop;
+
+	/*
+	 * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c).
+	 */
+	if ((is_udplite & UDPLITE_RECV_CC)  &&  UDP_SKB_CB(skb)->partial_cov) {
+
+		if (up->pcrlen == 0) {          /* full coverage was set  */
+			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: partial coverage"
+				" %d while full coverage %d requested\n",
+				UDP_SKB_CB(skb)->cscov, skb->len);
+			goto drop;
+		}
+		if (UDP_SKB_CB(skb)->cscov  <  up->pcrlen) {
+			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: coverage %d "
+						    "too small, need min %d\n",
+				       UDP_SKB_CB(skb)->cscov, up->pcrlen);
+			goto drop;
+		}
+	}
+
+	if (sk->sk_filter) {
+		if (udp_lib_checksum_complete(skb))
+			goto drop;
+	}
+
+	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
+		/* Note that an ENOMEM error is charged twice */
+		if (rc == -ENOMEM)
+			UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
+		goto drop;
+	}
+
+	return 0;
+drop:
+	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite);
+	kfree_skb(skb);
+	return -1;
+}
+
+static struct sock *udp_v6_mcast_next(struct sock *sk,
+				      __be16 loc_port, struct in6_addr *loc_addr,
+				      __be16 rmt_port, struct in6_addr *rmt_addr,
+				      int dif)
+{
+	struct hlist_node *node;
+	struct sock *s = sk;
+	unsigned short num = ntohs(loc_port);
+
+	sk_for_each_from(s, node) {
+		struct inet_sock *inet = inet_sk(s);
+
+		if (s->sk_hash == num && s->sk_family == PF_INET6) {
+			struct ipv6_pinfo *np = inet6_sk(s);
+			if (inet->dport) {
+				if (inet->dport != rmt_port)
+					continue;
+			}
+			if (!ipv6_addr_any(&np->daddr) &&
+			    !ipv6_addr_equal(&np->daddr, rmt_addr))
+				continue;
+
+			if (s->sk_bound_dev_if && s->sk_bound_dev_if != dif)
+				continue;
+
+			if (!ipv6_addr_any(&np->rcv_saddr)) {
+				if (!ipv6_addr_equal(&np->rcv_saddr, loc_addr))
+					continue;
+			}
+			if (!inet6_mc_check(s, loc_addr, rmt_addr))
+				continue;
+			return s;
+		}
+	}
+	return NULL;
+}
+
+/*
+ * Note: called only from the BH handler context,
+ * so we don't need to lock the hashes.
+ */
+static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr,
+			   struct in6_addr *daddr, struct hlist_head udptable[])
+{
+	struct sock *sk, *sk2;
+	const struct udphdr *uh = udp_hdr(skb);
+	int dif;
+
+	read_lock(&udp_hash_lock);
+	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
+	dif = inet6_iif(skb);
+	sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
+	if (!sk) {
+		kfree_skb(skb);
+		goto out;
+	}
+
+	sk2 = sk;
+	while ((sk2 = udp_v6_mcast_next(sk_next(sk2), uh->dest, daddr,
+					uh->source, saddr, dif))) {
+		struct sk_buff *buff = skb_clone(skb, GFP_ATOMIC);
+		if (buff) {
+			bh_lock_sock_nested(sk2);
+			if (!sock_owned_by_user(sk2))
+				udpv6_queue_rcv_skb(sk2, buff);
+			else
+				sk_add_backlog(sk2, buff);
+			bh_unlock_sock(sk2);
+		}
+	}
+	bh_lock_sock_nested(sk);
+	if (!sock_owned_by_user(sk))
+		udpv6_queue_rcv_skb(sk, skb);
+	else
+		sk_add_backlog(sk, skb);
+	bh_unlock_sock(sk);
+out:
+	read_unlock(&udp_hash_lock);
+	return 0;
+}
+
+static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh,
+				 int proto)
+{
+	int err;
+
+	UDP_SKB_CB(skb)->partial_cov = 0;
+	UDP_SKB_CB(skb)->cscov = skb->len;
+
+	if (proto == IPPROTO_UDPLITE) {
+		err = udplite_checksum_init(skb, uh);
+		if (err)
+			return err;
+	}
+
+	if (uh->check == 0) {
+		/* RFC 2460 section 8.1 says that we SHOULD log
+		   this error. Well, it is reasonable.
+		 */
+		LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n");
+		return 1;
+	}
+	if (skb->ip_summed == CHECKSUM_COMPLETE &&
+	    !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr,
+			     skb->len, proto, skb->csum))
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	if (!skb_csum_unnecessary(skb))
+		skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
+							 &ipv6_hdr(skb)->daddr,
+							 skb->len, proto, 0));
+
+	return 0;
+}
+
+int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
+		   int proto)
+{
+	struct sock *sk;
+	struct udphdr *uh;
+	struct net_device *dev = skb->dev;
+	struct in6_addr *saddr, *daddr;
+	u32 ulen = 0;
+
+	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
+		goto short_packet;
+
+	saddr = &ipv6_hdr(skb)->saddr;
+	daddr = &ipv6_hdr(skb)->daddr;
+	uh = udp_hdr(skb);
+
+	ulen = ntohs(uh->len);
+	if (ulen > skb->len)
+		goto short_packet;
+
+	if (proto == IPPROTO_UDP) {
+		/* UDP validates ulen. */
+
+		/* Check for jumbo payload */
+		if (ulen == 0)
+			ulen = skb->len;
+
+		if (ulen < sizeof(*uh))
+			goto short_packet;
+
+		if (ulen < skb->len) {
+			if (pskb_trim_rcsum(skb, ulen))
+				goto short_packet;
+			saddr = &ipv6_hdr(skb)->saddr;
+			daddr = &ipv6_hdr(skb)->daddr;
+			uh = udp_hdr(skb);
+		}
+	}
+
+	if (udp6_csum_init(skb, uh, proto))
+		goto discard;
+
+	/*
+	 *	Multicast receive code
+	 */
+	if (ipv6_addr_is_multicast(daddr))
+		return __udp6_lib_mcast_deliver(skb, saddr, daddr, udptable);
+
+	/* Unicast */
+
+	/*
+	 * check socket cache ... must talk to Alan about his plans
+	 * for sock caches... i'll skip this for now.
+	 */
+	sk = __udp6_lib_lookup(skb->dev->nd_net, saddr, uh->source,
+			       daddr, uh->dest, inet6_iif(skb), udptable);
+
+	if (sk == NULL) {
+		if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb))
+			goto discard;
+
+		if (udp_lib_checksum_complete(skb))
+			goto discard;
+		UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+
+		icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev);
+
+		kfree_skb(skb);
+		return 0;
+	}
+
+	/* deliver */
+
+	bh_lock_sock_nested(sk);
+	if (!sock_owned_by_user(sk))
+		udpv6_queue_rcv_skb(sk, skb);
+	else
+		sk_add_backlog(sk, skb);
+	bh_unlock_sock(sk);
+	sock_put(sk);
+	return 0;
+
+short_packet:
+	LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n",
+		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
+		       ulen, skb->len);
+
+discard:
+	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+	kfree_skb(skb);
+	return 0;
+}
+
+static __inline__ int udpv6_rcv(struct sk_buff *skb)
+{
+	return __udp6_lib_rcv(skb, udp_hash, IPPROTO_UDP);
+}
+
+/*
+ * Throw away all pending data and cancel the corking. Socket is locked.
+ */
+static void udp_v6_flush_pending_frames(struct sock *sk)
+{
+	struct udp_sock *up = udp_sk(sk);
+
+	if (up->pending) {
+		up->len = 0;
+		up->pending = 0;
+		ip6_flush_pending_frames(sk);
+	}
+}
+
+/*
+ *	Sending
+ */
+
+static int udp_v6_push_pending_frames(struct sock *sk)
+{
+	struct sk_buff *skb;
+	struct udphdr *uh;
+	struct udp_sock  *up = udp_sk(sk);
+	struct inet_sock *inet = inet_sk(sk);
+	struct flowi *fl = &inet->cork.fl;
+	int err = 0;
+	int is_udplite = IS_UDPLITE(sk);
+	__wsum csum = 0;
+
+	/* Grab the skbuff where UDP header space exists. */
+	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
+		goto out;
+
+	/*
+	 * Create a UDP header
+	 */
+	uh = udp_hdr(skb);
+	uh->source = fl->fl_ip_sport;
+	uh->dest = fl->fl_ip_dport;
+	uh->len = htons(up->len);
+	uh->check = 0;
+
+	if (is_udplite)
+		csum = udplite_csum_outgoing(sk, skb);
+	 else
+		csum = udp_csum_outgoing(sk, skb);
+
+	/* add protocol-dependent pseudo-header */
+	uh->check = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst,
+				    up->len, fl->proto, csum   );
+	if (uh->check == 0)
+		uh->check = CSUM_MANGLED_0;
+
+	err = ip6_push_pending_frames(sk);
+out:
+	up->len = 0;
+	up->pending = 0;
+	if (!err)
+		UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite);
+	return err;
+}
+
+int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk,
+		  struct msghdr *msg, size_t len)
+{
+	struct ipv6_txoptions opt_space;
+	struct udp_sock *up = udp_sk(sk);
+	struct inet_sock *inet = inet_sk(sk);
+	struct ipv6_pinfo *np = inet6_sk(sk);
+	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) msg->msg_name;
+	struct in6_addr *daddr, *final_p = NULL, final;
+	struct ipv6_txoptions *opt = NULL;
+	struct ip6_flowlabel *flowlabel = NULL;
+	struct flowi fl;
+	struct dst_entry *dst;
+	int addr_len = msg->msg_namelen;
+	int ulen = len;
+	int hlimit = -1;
+	int tclass = -1;
+	int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
+	int err;
+	int connected = 0;
+	int is_udplite = IS_UDPLITE(sk);
+	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
+
+	/* destination address check */
+	if (sin6) {
+		if (addr_len < offsetof(struct sockaddr, sa_data))
+			return -EINVAL;
+
+		switch (sin6->sin6_family) {
+		case AF_INET6:
+			if (addr_len < SIN6_LEN_RFC2133)
+				return -EINVAL;
+			daddr = &sin6->sin6_addr;
+			break;
+		case AF_INET:
+			goto do_udp_sendmsg;
+		case AF_UNSPEC:
+			msg->msg_name = sin6 = NULL;
+			msg->msg_namelen = addr_len = 0;
+			daddr = NULL;
+			break;
+		default:
+			return -EINVAL;
+		}
+	} else if (!up->pending) {
+		if (sk->sk_state != TCP_ESTABLISHED)
+			return -EDESTADDRREQ;
+		daddr = &np->daddr;
+	} else
+		daddr = NULL;
+
+	if (daddr) {
+		if (ipv6_addr_v4mapped(daddr)) {
+			struct sockaddr_in sin;
+			sin.sin_family = AF_INET;
+			sin.sin_port = sin6 ? sin6->sin6_port : inet->dport;
+			sin.sin_addr.s_addr = daddr->s6_addr32[3];
+			msg->msg_name = &sin;
+			msg->msg_namelen = sizeof(sin);
+do_udp_sendmsg:
+			if (__ipv6_only_sock(sk))
+				return -ENETUNREACH;
+			return udp_sendmsg(iocb, sk, msg, len);
+		}
+	}
+
+	if (up->pending == AF_INET)
+		return udp_sendmsg(iocb, sk, msg, len);
+
+	/* Rough check on arithmetic overflow,
+	   better check is made in ip6_append_data().
+	   */
+	if (len > INT_MAX - sizeof(struct udphdr))
+		return -EMSGSIZE;
+
+	if (up->pending) {
+		/*
+		 * There are pending frames.
+		 * The socket lock must be held while it's corked.
+		 */
+		lock_sock(sk);
+		if (likely(up->pending)) {
+			if (unlikely(up->pending != AF_INET6)) {
+				release_sock(sk);
+				return -EAFNOSUPPORT;
+			}
+			dst = NULL;
+			goto do_append_data;
+		}
+		release_sock(sk);
+	}
+	ulen += sizeof(struct udphdr);
+
+	memset(&fl, 0, sizeof(fl));
+
+	if (sin6) {
+		if (sin6->sin6_port == 0)
+			return -EINVAL;
+
+		fl.fl_ip_dport = sin6->sin6_port;
+		daddr = &sin6->sin6_addr;
+
+		if (np->sndflow) {
+			fl.fl6_flowlabel = sin6->sin6_flowinfo&IPV6_FLOWINFO_MASK;
+			if (fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) {
+				flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel);
+				if (flowlabel == NULL)
+					return -EINVAL;
+				daddr = &flowlabel->dst;
+			}
+		}
+
+		/*
+		 * Otherwise it will be difficult to maintain
+		 * sk->sk_dst_cache.
+		 */
+		if (sk->sk_state == TCP_ESTABLISHED &&
+		    ipv6_addr_equal(daddr, &np->daddr))
+			daddr = &np->daddr;
+
+		if (addr_len >= sizeof(struct sockaddr_in6) &&
+		    sin6->sin6_scope_id &&
+		    ipv6_addr_type(daddr)&IPV6_ADDR_LINKLOCAL)
+			fl.oif = sin6->sin6_scope_id;
+	} else {
+		if (sk->sk_state != TCP_ESTABLISHED)
+			return -EDESTADDRREQ;
+
+		fl.fl_ip_dport = inet->dport;
+		daddr = &np->daddr;
+		fl.fl6_flowlabel = np->flow_label;
+		connected = 1;
+	}
+
+	if (!fl.oif)
+		fl.oif = sk->sk_bound_dev_if;
+
+	if (msg->msg_controllen) {
+		opt = &opt_space;
+		memset(opt, 0, sizeof(struct ipv6_txoptions));
+		opt->tot_len = sizeof(*opt);
+
+		err = datagram_send_ctl(msg, &fl, opt, &hlimit, &tclass);
+		if (err < 0) {
+			fl6_sock_release(flowlabel);
+			return err;
+		}
+		if ((fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) && !flowlabel) {
+			flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel);
+			if (flowlabel == NULL)
+				return -EINVAL;
+		}
+		if (!(opt->opt_nflen|opt->opt_flen))
+			opt = NULL;
+		connected = 0;
+	}
+	if (opt == NULL)
+		opt = np->opt;
+	if (flowlabel)
+		opt = fl6_merge_options(&opt_space, flowlabel, opt);
+	opt = ipv6_fixup_options(&opt_space, opt);
+
+	fl.proto = sk->sk_protocol;
+	ipv6_addr_copy(&fl.fl6_dst, daddr);
+	if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr))
+		ipv6_addr_copy(&fl.fl6_src, &np->saddr);
+	fl.fl_ip_sport = inet->sport;
+
+	/* merge ip6_build_xmit from ip6_output */
+	if (opt && opt->srcrt) {
+		struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
+		ipv6_addr_copy(&final, &fl.fl6_dst);
+		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
+		final_p = &final;
+		connected = 0;
+	}
+
+	if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) {
+		fl.oif = np->mcast_oif;
+		connected = 0;
+	}
+
+	security_sk_classify_flow(sk, &fl);
+
+	err = ip6_sk_dst_lookup(sk, &dst, &fl);
+	if (err)
+		goto out;
+	if (final_p)
+		ipv6_addr_copy(&fl.fl6_dst, final_p);
+
+	if ((err = __xfrm_lookup(&dst, &fl, sk, XFRM_LOOKUP_WAIT)) < 0) {
+		if (err == -EREMOTE)
+			err = ip6_dst_blackhole(sk, &dst, &fl);
+		if (err < 0)
+			goto out;
+	}
+
+	if (hlimit < 0) {
+		if (ipv6_addr_is_multicast(&fl.fl6_dst))
+			hlimit = np->mcast_hops;
+		else
+			hlimit = np->hop_limit;
+		if (hlimit < 0)
+			hlimit = dst_metric(dst, RTAX_HOPLIMIT);
+		if (hlimit < 0)
+			hlimit = ipv6_get_hoplimit(dst->dev);
+	}
+
+	if (tclass < 0) {
+		tclass = np->tclass;
+		if (tclass < 0)
+			tclass = 0;
+	}
+
+	if (msg->msg_flags&MSG_CONFIRM)
+		goto do_confirm;
+back_from_confirm:
+
+	lock_sock(sk);
+	if (unlikely(up->pending)) {
+		/* The socket is already corked while preparing it. */
+		/* ... which is an evident application bug. --ANK */
+		release_sock(sk);
+
+		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n");
+		err = -EINVAL;
+		goto out;
+	}
+
+	up->pending = AF_INET6;
+
+do_append_data:
+	up->len += ulen;
+	getfrag  =  is_udplite ?  udplite_getfrag : ip_generic_getfrag;
+	err = ip6_append_data(sk, getfrag, msg->msg_iov, ulen,
+		sizeof(struct udphdr), hlimit, tclass, opt, &fl,
+		(struct rt6_info*)dst,
+		corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
+	if (err)
+		udp_v6_flush_pending_frames(sk);
+	else if (!corkreq)
+		err = udp_v6_push_pending_frames(sk);
+	else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
+		up->pending = 0;
+
+	if (dst) {
+		if (connected) {
+			ip6_dst_store(sk, dst,
+				      ipv6_addr_equal(&fl.fl6_dst, &np->daddr) ?
+				      &np->daddr : NULL,
+#ifdef CONFIG_IPV6_SUBTREES
+				      ipv6_addr_equal(&fl.fl6_src, &np->saddr) ?
+				      &np->saddr :
+#endif
+				      NULL);
+		} else {
+			dst_release(dst);
+		}
+	}
+
+	if (err > 0)
+		err = np->recverr ? net_xmit_errno(err) : 0;
+	release_sock(sk);
+out:
+	fl6_sock_release(flowlabel);
+	if (!err)
+		return len;
+	/*
+	 * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space.  Reporting
+	 * ENOBUFS might not be good (it's not tunable per se), but otherwise
+	 * we don't have a good statistic (IpOutDiscards but it can be too many
+	 * things).  We could add another new stat but at least for now that
+	 * seems like overkill.
+	 */
+	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
+		UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite);
+	}
+	return err;
+
+do_confirm:
+	dst_confirm(dst);
+	if (!(msg->msg_flags&MSG_PROBE) || len)
+		goto back_from_confirm;
+	err = 0;
+	goto out;
+}
+
+int udpv6_destroy_sock(struct sock *sk)
+{
+	lock_sock(sk);
+	udp_v6_flush_pending_frames(sk);
+	release_sock(sk);
+
+	inet6_destroy_sock(sk);
+
+	return 0;
+}
+
+/*
+ *	Socket option code for UDP
+ */
+int udpv6_setsockopt(struct sock *sk, int level, int optname,
+		     char __user *optval, int optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
+					  udp_v6_push_pending_frames);
+	return ipv6_setsockopt(sk, level, optname, optval, optlen);
+}
+
+#ifdef CONFIG_COMPAT
+int compat_udpv6_setsockopt(struct sock *sk, int level, int optname,
+			    char __user *optval, int optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
+					  udp_v6_push_pending_frames);
+	return compat_ipv6_setsockopt(sk, level, optname, optval, optlen);
+}
+#endif
+
+int udpv6_getsockopt(struct sock *sk, int level, int optname,
+		     char __user *optval, int __user *optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
+	return ipv6_getsockopt(sk, level, optname, optval, optlen);
+}
+
+#ifdef CONFIG_COMPAT
+int compat_udpv6_getsockopt(struct sock *sk, int level, int optname,
+			    char __user *optval, int __user *optlen)
+{
+	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
+		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
+	return compat_ipv6_getsockopt(sk, level, optname, optval, optlen);
+}
+#endif
+
+static struct inet6_protocol udpv6_protocol = {
+	.handler	=	udpv6_rcv,
+	.err_handler	=	udpv6_err,
+	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
+};
+
+/* ------------------------------------------------------------------------ */
+#ifdef CONFIG_PROC_FS
+
+static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket)
+{
+	struct inet_sock *inet = inet_sk(sp);
+	struct ipv6_pinfo *np = inet6_sk(sp);
+	struct in6_addr *dest, *src;
+	__u16 destp, srcp;
+
+	dest  = &np->daddr;
+	src   = &np->rcv_saddr;
+	destp = ntohs(inet->dport);
+	srcp  = ntohs(inet->sport);
+	seq_printf(seq,
+		   "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
+		   "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p\n",
+		   bucket,
+		   src->s6_addr32[0], src->s6_addr32[1],
+		   src->s6_addr32[2], src->s6_addr32[3], srcp,
+		   dest->s6_addr32[0], dest->s6_addr32[1],
+		   dest->s6_addr32[2], dest->s6_addr32[3], destp,
+		   sp->sk_state,
+		   atomic_read(&sp->sk_wmem_alloc),
+		   atomic_read(&sp->sk_rmem_alloc),
+		   0, 0L, 0,
+		   sock_i_uid(sp), 0,
+		   sock_i_ino(sp),
+		   atomic_read(&sp->sk_refcnt), sp);
+}
+
+int udp6_seq_show(struct seq_file *seq, void *v)
+{
+	if (v == SEQ_START_TOKEN)
+		seq_printf(seq,
+			   "  sl  "
+			   "local_address                         "
+			   "remote_address                        "
+			   "st tx_queue rx_queue tr tm->when retrnsmt"
+			   "   uid  timeout inode\n");
+	else
+		udp6_sock_seq_show(seq, v, ((struct udp_iter_state *)seq->private)->bucket);
+	return 0;
+}
+
+static struct file_operations udp6_seq_fops;
+static struct udp_seq_afinfo udp6_seq_afinfo = {
+	.owner		= THIS_MODULE,
+	.name		= "udp6",
+	.family		= AF_INET6,
+	.hashtable	= udp_hash,
+	.seq_show	= udp6_seq_show,
+	.seq_fops	= &udp6_seq_fops,
+};
+
+int __init udp6_proc_init(void)
+{
+	return udp_proc_register(&udp6_seq_afinfo);
+}
+
+void udp6_proc_exit(void) {
+	udp_proc_unregister(&udp6_seq_afinfo);
+}
+#endif /* CONFIG_PROC_FS */
+
+/* ------------------------------------------------------------------------ */
+
+DEFINE_PROTO_INUSE(udpv6)
+
+struct proto udpv6_prot = {
+	.name		   = "UDPv6",
+	.owner		   = THIS_MODULE,
+	.close		   = udp_lib_close,
+	.connect	   = ip6_datagram_connect,
+	.disconnect	   = udp_disconnect,
+	.ioctl		   = udp_ioctl,
+	.destroy	   = udpv6_destroy_sock,
+	.setsockopt	   = udpv6_setsockopt,
+	.getsockopt	   = udpv6_getsockopt,
+	.sendmsg	   = udpv6_sendmsg,
+	.recvmsg	   = udpv6_recvmsg,
+	.backlog_rcv	   = udpv6_queue_rcv_skb,
+	.hash		   = udp_lib_hash,
+	.unhash		   = udp_lib_unhash,
+	.get_port	   = udp_v6_get_port,
+	.memory_allocated  = &udp_memory_allocated,
+	.sysctl_mem	   = sysctl_udp_mem,
+	.sysctl_wmem	   = &sysctl_udp_wmem_min,
+	.sysctl_rmem	   = &sysctl_udp_rmem_min,
+	.obj_size	   = sizeof(struct udp6_sock),
+#ifdef CONFIG_COMPAT
+	.compat_setsockopt = compat_udpv6_setsockopt,
+	.compat_getsockopt = compat_udpv6_getsockopt,
+#endif
+	REF_PROTO_INUSE(udpv6)
+};
+
+static struct inet_protosw udpv6_protosw = {
+	.type =      SOCK_DGRAM,
+	.protocol =  IPPROTO_UDP,
+	.prot =      &udpv6_prot,
+	.ops =       &inet6_dgram_ops,
+	.capability =-1,
+	.no_check =  UDP_CSUM_DEFAULT,
+	.flags =     INET_PROTOSW_PERMANENT,
+};
+
+
+int __init udpv6_init(void)
+{
+	int ret;
+
+	ret = inet6_add_protocol(&udpv6_protocol, IPPROTO_UDP);
+	if (ret)
+		goto out;
+
+	ret = inet6_register_protosw(&udpv6_protosw);
+	if (ret)
+		goto out_udpv6_protocol;
+out:
+	return ret;
+
+out_udpv6_protocol:
+	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
+	goto out;
+}
+
+void udpv6_exit(void)
+{
+	inet6_unregister_protosw(&udpv6_protosw);
+	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
+}
diff --git a/net/ipv6/udp_ipv6.c b/net/ipv6/udp_ipv6.c
deleted file mode 100644
index 55feac7ba717..000000000000
--- a/net/ipv6/udp_ipv6.c
+++ /dev/null
@@ -1,1065 +0,0 @@
-/*
- *	UDP over IPv6
- *	Linux INET6 implementation
- *
- *	Authors:
- *	Pedro Roque		<roque@di.fc.ul.pt>
- *
- *	Based on linux/ipv4/udp.c
- *
- *	$Id: udp.c,v 1.65 2002/02/01 22:01:04 davem Exp $
- *
- *	Fixes:
- *	Hideaki YOSHIFUJI	:	sin6_scope_id support
- *	YOSHIFUJI Hideaki @USAGI and:	Support IPV6_V6ONLY socket option, which
- *	Alexey Kuznetsov		allow both IPv4 and IPv6 sockets to bind
- *					a single port at the same time.
- *      Kazunori MIYAZAWA @USAGI:       change process style to use ip6_append_data
- *      YOSHIFUJI Hideaki @USAGI:	convert /proc/net/udp6 to seq_file.
- *
- *	This program is free software; you can redistribute it and/or
- *      modify it under the terms of the GNU General Public License
- *      as published by the Free Software Foundation; either version
- *      2 of the License, or (at your option) any later version.
- */
-
-#include <linux/errno.h>
-#include <linux/types.h>
-#include <linux/socket.h>
-#include <linux/sockios.h>
-#include <linux/net.h>
-#include <linux/in6.h>
-#include <linux/netdevice.h>
-#include <linux/if_arp.h>
-#include <linux/ipv6.h>
-#include <linux/icmpv6.h>
-#include <linux/init.h>
-#include <linux/module.h>
-#include <linux/skbuff.h>
-#include <asm/uaccess.h>
-
-#include <net/ndisc.h>
-#include <net/protocol.h>
-#include <net/transp_v6.h>
-#include <net/ip6_route.h>
-#include <net/raw.h>
-#include <net/tcp_states.h>
-#include <net/ip6_checksum.h>
-#include <net/xfrm.h>
-
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
-#include "udp_impl.h"
-
-static inline int udp_v6_get_port(struct sock *sk, unsigned short snum)
-{
-	return udp_get_port(sk, snum, ipv6_rcv_saddr_equal);
-}
-
-static struct sock *__udp6_lib_lookup(struct net *net,
-				      struct in6_addr *saddr, __be16 sport,
-				      struct in6_addr *daddr, __be16 dport,
-				      int dif, struct hlist_head udptable[])
-{
-	struct sock *sk, *result = NULL;
-	struct hlist_node *node;
-	unsigned short hnum = ntohs(dport);
-	int badness = -1;
-
-	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
-		struct inet_sock *inet = inet_sk(sk);
-
-		if (sk->sk_net == net && sk->sk_hash == hnum &&
-				sk->sk_family == PF_INET6) {
-			struct ipv6_pinfo *np = inet6_sk(sk);
-			int score = 0;
-			if (inet->dport) {
-				if (inet->dport != sport)
-					continue;
-				score++;
-			}
-			if (!ipv6_addr_any(&np->rcv_saddr)) {
-				if (!ipv6_addr_equal(&np->rcv_saddr, daddr))
-					continue;
-				score++;
-			}
-			if (!ipv6_addr_any(&np->daddr)) {
-				if (!ipv6_addr_equal(&np->daddr, saddr))
-					continue;
-				score++;
-			}
-			if (sk->sk_bound_dev_if) {
-				if (sk->sk_bound_dev_if != dif)
-					continue;
-				score++;
-			}
-			if (score == 4) {
-				result = sk;
-				break;
-			} else if (score > badness) {
-				result = sk;
-				badness = score;
-			}
-		}
-	}
-	if (result)
-		sock_hold(result);
-	read_unlock(&udp_hash_lock);
-	return result;
-}
-
-/*
- * 	This should be easy, if there is something there we
- * 	return it, otherwise we block.
- */
-
-int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk,
-		  struct msghdr *msg, size_t len,
-		  int noblock, int flags, int *addr_len)
-{
-	struct ipv6_pinfo *np = inet6_sk(sk);
-	struct inet_sock *inet = inet_sk(sk);
-	struct sk_buff *skb;
-	unsigned int ulen, copied;
-	int peeked;
-	int err;
-	int is_udplite = IS_UDPLITE(sk);
-
-	if (addr_len)
-		*addr_len=sizeof(struct sockaddr_in6);
-
-	if (flags & MSG_ERRQUEUE)
-		return ipv6_recv_error(sk, msg, len);
-
-try_again:
-	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
-				  &peeked, &err);
-	if (!skb)
-		goto out;
-
-	ulen = skb->len - sizeof(struct udphdr);
-	copied = len;
-	if (copied > ulen)
-		copied = ulen;
-	else if (copied < ulen)
-		msg->msg_flags |= MSG_TRUNC;
-
-	/*
-	 * If checksum is needed at all, try to do it while copying the
-	 * data.  If the data is truncated, or if we only want a partial
-	 * coverage checksum (UDP-Lite), do it before the copy.
-	 */
-
-	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
-		if (udp_lib_checksum_complete(skb))
-			goto csum_copy_err;
-	}
-
-	if (skb_csum_unnecessary(skb))
-		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
-					      msg->msg_iov, copied       );
-	else {
-		err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov);
-		if (err == -EINVAL)
-			goto csum_copy_err;
-	}
-	if (err)
-		goto out_free;
-
-	if (!peeked)
-		UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite);
-
-	sock_recv_timestamp(msg, sk, skb);
-
-	/* Copy the address. */
-	if (msg->msg_name) {
-		struct sockaddr_in6 *sin6;
-
-		sin6 = (struct sockaddr_in6 *) msg->msg_name;
-		sin6->sin6_family = AF_INET6;
-		sin6->sin6_port = udp_hdr(skb)->source;
-		sin6->sin6_flowinfo = 0;
-		sin6->sin6_scope_id = 0;
-
-		if (skb->protocol == htons(ETH_P_IP))
-			ipv6_addr_set(&sin6->sin6_addr, 0, 0,
-				      htonl(0xffff), ip_hdr(skb)->saddr);
-		else {
-			ipv6_addr_copy(&sin6->sin6_addr,
-				       &ipv6_hdr(skb)->saddr);
-			if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL)
-				sin6->sin6_scope_id = IP6CB(skb)->iif;
-		}
-
-	}
-	if (skb->protocol == htons(ETH_P_IP)) {
-		if (inet->cmsg_flags)
-			ip_cmsg_recv(msg, skb);
-	} else {
-		if (np->rxopt.all)
-			datagram_recv_ctl(sk, msg, skb);
-	}
-
-	err = copied;
-	if (flags & MSG_TRUNC)
-		err = ulen;
-
-out_free:
-	lock_sock(sk);
-	skb_free_datagram(sk, skb);
-	release_sock(sk);
-out:
-	return err;
-
-csum_copy_err:
-	lock_sock(sk);
-	if (!skb_kill_datagram(sk, skb, flags))
-		UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite);
-	release_sock(sk);
-
-	if (flags & MSG_DONTWAIT)
-		return -EAGAIN;
-	goto try_again;
-}
-
-void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-		    int type, int code, int offset, __be32 info,
-		    struct hlist_head udptable[]                    )
-{
-	struct ipv6_pinfo *np;
-	struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data;
-	struct in6_addr *saddr = &hdr->saddr;
-	struct in6_addr *daddr = &hdr->daddr;
-	struct udphdr *uh = (struct udphdr*)(skb->data+offset);
-	struct sock *sk;
-	int err;
-
-	sk = __udp6_lib_lookup(skb->dev->nd_net, daddr, uh->dest,
-			       saddr, uh->source, inet6_iif(skb), udptable);
-	if (sk == NULL)
-		return;
-
-	np = inet6_sk(sk);
-
-	if (!icmpv6_err_convert(type, code, &err) && !np->recverr)
-		goto out;
-
-	if (sk->sk_state != TCP_ESTABLISHED && !np->recverr)
-		goto out;
-
-	if (np->recverr)
-		ipv6_icmp_error(sk, skb, err, uh->dest, ntohl(info), (u8 *)(uh+1));
-
-	sk->sk_err = err;
-	sk->sk_error_report(sk);
-out:
-	sock_put(sk);
-}
-
-static __inline__ void udpv6_err(struct sk_buff *skb,
-				 struct inet6_skb_parm *opt, int type,
-				 int code, int offset, __be32 info     )
-{
-	__udp6_lib_err(skb, opt, type, code, offset, info, udp_hash);
-}
-
-int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
-{
-	struct udp_sock *up = udp_sk(sk);
-	int rc;
-	int is_udplite = IS_UDPLITE(sk);
-
-	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
-		goto drop;
-
-	/*
-	 * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c).
-	 */
-	if ((is_udplite & UDPLITE_RECV_CC)  &&  UDP_SKB_CB(skb)->partial_cov) {
-
-		if (up->pcrlen == 0) {          /* full coverage was set  */
-			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: partial coverage"
-				" %d while full coverage %d requested\n",
-				UDP_SKB_CB(skb)->cscov, skb->len);
-			goto drop;
-		}
-		if (UDP_SKB_CB(skb)->cscov  <  up->pcrlen) {
-			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: coverage %d "
-						    "too small, need min %d\n",
-				       UDP_SKB_CB(skb)->cscov, up->pcrlen);
-			goto drop;
-		}
-	}
-
-	if (sk->sk_filter) {
-		if (udp_lib_checksum_complete(skb))
-			goto drop;
-	}
-
-	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
-		/* Note that an ENOMEM error is charged twice */
-		if (rc == -ENOMEM)
-			UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite);
-		goto drop;
-	}
-
-	return 0;
-drop:
-	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite);
-	kfree_skb(skb);
-	return -1;
-}
-
-static struct sock *udp_v6_mcast_next(struct sock *sk,
-				      __be16 loc_port, struct in6_addr *loc_addr,
-				      __be16 rmt_port, struct in6_addr *rmt_addr,
-				      int dif)
-{
-	struct hlist_node *node;
-	struct sock *s = sk;
-	unsigned short num = ntohs(loc_port);
-
-	sk_for_each_from(s, node) {
-		struct inet_sock *inet = inet_sk(s);
-
-		if (s->sk_hash == num && s->sk_family == PF_INET6) {
-			struct ipv6_pinfo *np = inet6_sk(s);
-			if (inet->dport) {
-				if (inet->dport != rmt_port)
-					continue;
-			}
-			if (!ipv6_addr_any(&np->daddr) &&
-			    !ipv6_addr_equal(&np->daddr, rmt_addr))
-				continue;
-
-			if (s->sk_bound_dev_if && s->sk_bound_dev_if != dif)
-				continue;
-
-			if (!ipv6_addr_any(&np->rcv_saddr)) {
-				if (!ipv6_addr_equal(&np->rcv_saddr, loc_addr))
-					continue;
-			}
-			if (!inet6_mc_check(s, loc_addr, rmt_addr))
-				continue;
-			return s;
-		}
-	}
-	return NULL;
-}
-
-/*
- * Note: called only from the BH handler context,
- * so we don't need to lock the hashes.
- */
-static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr,
-			   struct in6_addr *daddr, struct hlist_head udptable[])
-{
-	struct sock *sk, *sk2;
-	const struct udphdr *uh = udp_hdr(skb);
-	int dif;
-
-	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
-	dif = inet6_iif(skb);
-	sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
-	if (!sk) {
-		kfree_skb(skb);
-		goto out;
-	}
-
-	sk2 = sk;
-	while ((sk2 = udp_v6_mcast_next(sk_next(sk2), uh->dest, daddr,
-					uh->source, saddr, dif))) {
-		struct sk_buff *buff = skb_clone(skb, GFP_ATOMIC);
-		if (buff) {
-			bh_lock_sock_nested(sk2);
-			if (!sock_owned_by_user(sk2))
-				udpv6_queue_rcv_skb(sk2, buff);
-			else
-				sk_add_backlog(sk2, buff);
-			bh_unlock_sock(sk2);
-		}
-	}
-	bh_lock_sock_nested(sk);
-	if (!sock_owned_by_user(sk))
-		udpv6_queue_rcv_skb(sk, skb);
-	else
-		sk_add_backlog(sk, skb);
-	bh_unlock_sock(sk);
-out:
-	read_unlock(&udp_hash_lock);
-	return 0;
-}
-
-static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh,
-				 int proto)
-{
-	int err;
-
-	UDP_SKB_CB(skb)->partial_cov = 0;
-	UDP_SKB_CB(skb)->cscov = skb->len;
-
-	if (IS_PROTO_UDPLITE(proto)) {
-		err = udplite_checksum_init(skb, uh);
-		if (err)
-			return err;
-	}
-
-	if (uh->check == 0) {
-		/* RFC 2460 section 8.1 says that we SHOULD log
-		   this error. Well, it is reasonable.
-		 */
-		LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n");
-		return 1;
-	}
-	if (skb->ip_summed == CHECKSUM_COMPLETE &&
-	    !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr,
-			     skb->len, proto, skb->csum))
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-
-	if (!skb_csum_unnecessary(skb))
-		skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
-							 &ipv6_hdr(skb)->daddr,
-							 skb->len, proto, 0));
-
-	return 0;
-}
-
-int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
-		   int proto)
-{
-	struct sock *sk;
-	struct udphdr *uh;
-	struct net_device *dev = skb->dev;
-	struct in6_addr *saddr, *daddr;
-	u32 ulen = 0;
-
-	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
-		goto short_packet;
-
-	saddr = &ipv6_hdr(skb)->saddr;
-	daddr = &ipv6_hdr(skb)->daddr;
-	uh = udp_hdr(skb);
-
-	ulen = ntohs(uh->len);
-	if (ulen > skb->len)
-		goto short_packet;
-
-	if (proto == IPPROTO_UDP) {
-		/* UDP validates ulen. */
-
-		/* Check for jumbo payload */
-		if (ulen == 0)
-			ulen = skb->len;
-
-		if (ulen < sizeof(*uh))
-			goto short_packet;
-
-		if (ulen < skb->len) {
-			if (pskb_trim_rcsum(skb, ulen))
-				goto short_packet;
-			saddr = &ipv6_hdr(skb)->saddr;
-			daddr = &ipv6_hdr(skb)->daddr;
-			uh = udp_hdr(skb);
-		}
-	}
-
-	if (udp6_csum_init(skb, uh, proto))
-		goto discard;
-
-	/*
-	 *	Multicast receive code
-	 */
-	if (ipv6_addr_is_multicast(daddr))
-		return __udp6_lib_mcast_deliver(skb, saddr, daddr, udptable);
-
-	/* Unicast */
-
-	/*
-	 * check socket cache ... must talk to Alan about his plans
-	 * for sock caches... i'll skip this for now.
-	 */
-	sk = __udp6_lib_lookup(skb->dev->nd_net, saddr, uh->source,
-			       daddr, uh->dest, inet6_iif(skb), udptable);
-
-	if (sk == NULL) {
-		if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb))
-			goto discard;
-
-		if (udp_lib_checksum_complete(skb))
-			goto discard;
-		UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto));
-
-		icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev);
-
-		kfree_skb(skb);
-		return 0;
-	}
-
-	/* deliver */
-
-	bh_lock_sock_nested(sk);
-	if (!sock_owned_by_user(sk))
-		udpv6_queue_rcv_skb(sk, skb);
-	else
-		sk_add_backlog(sk, skb);
-	bh_unlock_sock(sk);
-	sock_put(sk);
-	return 0;
-
-short_packet:
-	LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n",
-		       IS_PROTO_UDPLITE(proto) ? "-Lite" : "",
-		       ulen, skb->len);
-
-discard:
-	UDP6_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto));
-	kfree_skb(skb);
-	return 0;
-}
-
-static __inline__ int udpv6_rcv(struct sk_buff *skb)
-{
-	return __udp6_lib_rcv(skb, udp_hash, IPPROTO_UDP);
-}
-
-/*
- * Throw away all pending data and cancel the corking. Socket is locked.
- */
-static void udp_v6_flush_pending_frames(struct sock *sk)
-{
-	struct udp_sock *up = udp_sk(sk);
-
-	if (up->pending) {
-		up->len = 0;
-		up->pending = 0;
-		ip6_flush_pending_frames(sk);
-	}
-}
-
-/*
- *	Sending
- */
-
-static int udp_v6_push_pending_frames(struct sock *sk)
-{
-	struct sk_buff *skb;
-	struct udphdr *uh;
-	struct udp_sock  *up = udp_sk(sk);
-	struct inet_sock *inet = inet_sk(sk);
-	struct flowi *fl = &inet->cork.fl;
-	int err = 0;
-	int is_udplite = IS_UDPLITE(sk);
-	__wsum csum = 0;
-
-	/* Grab the skbuff where UDP header space exists. */
-	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
-		goto out;
-
-	/*
-	 * Create a UDP header
-	 */
-	uh = udp_hdr(skb);
-	uh->source = fl->fl_ip_sport;
-	uh->dest = fl->fl_ip_dport;
-	uh->len = htons(up->len);
-	uh->check = 0;
-
-	if (is_udplite)
-		csum = udplite_csum_outgoing(sk, skb);
-	 else
-		csum = udp_csum_outgoing(sk, skb);
-
-	/* add protocol-dependent pseudo-header */
-	uh->check = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst,
-				    up->len, fl->proto, csum   );
-	if (uh->check == 0)
-		uh->check = CSUM_MANGLED_0;
-
-	err = ip6_push_pending_frames(sk);
-out:
-	up->len = 0;
-	up->pending = 0;
-	if (!err)
-		UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite);
-	return err;
-}
-
-int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk,
-		  struct msghdr *msg, size_t len)
-{
-	struct ipv6_txoptions opt_space;
-	struct udp_sock *up = udp_sk(sk);
-	struct inet_sock *inet = inet_sk(sk);
-	struct ipv6_pinfo *np = inet6_sk(sk);
-	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) msg->msg_name;
-	struct in6_addr *daddr, *final_p = NULL, final;
-	struct ipv6_txoptions *opt = NULL;
-	struct ip6_flowlabel *flowlabel = NULL;
-	struct flowi fl;
-	struct dst_entry *dst;
-	int addr_len = msg->msg_namelen;
-	int ulen = len;
-	int hlimit = -1;
-	int tclass = -1;
-	int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
-	int err;
-	int connected = 0;
-	int is_udplite = IS_UDPLITE(sk);
-	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
-
-	/* destination address check */
-	if (sin6) {
-		if (addr_len < offsetof(struct sockaddr, sa_data))
-			return -EINVAL;
-
-		switch (sin6->sin6_family) {
-		case AF_INET6:
-			if (addr_len < SIN6_LEN_RFC2133)
-				return -EINVAL;
-			daddr = &sin6->sin6_addr;
-			break;
-		case AF_INET:
-			goto do_udp_sendmsg;
-		case AF_UNSPEC:
-			msg->msg_name = sin6 = NULL;
-			msg->msg_namelen = addr_len = 0;
-			daddr = NULL;
-			break;
-		default:
-			return -EINVAL;
-		}
-	} else if (!up->pending) {
-		if (sk->sk_state != TCP_ESTABLISHED)
-			return -EDESTADDRREQ;
-		daddr = &np->daddr;
-	} else
-		daddr = NULL;
-
-	if (daddr) {
-		if (ipv6_addr_v4mapped(daddr)) {
-			struct sockaddr_in sin;
-			sin.sin_family = AF_INET;
-			sin.sin_port = sin6 ? sin6->sin6_port : inet->dport;
-			sin.sin_addr.s_addr = daddr->s6_addr32[3];
-			msg->msg_name = &sin;
-			msg->msg_namelen = sizeof(sin);
-do_udp_sendmsg:
-			if (__ipv6_only_sock(sk))
-				return -ENETUNREACH;
-			return udp_sendmsg(iocb, sk, msg, len);
-		}
-	}
-
-	if (up->pending == AF_INET)
-		return udp_sendmsg(iocb, sk, msg, len);
-
-	/* Rough check on arithmetic overflow,
-	   better check is made in ip6_append_data().
-	   */
-	if (len > INT_MAX - sizeof(struct udphdr))
-		return -EMSGSIZE;
-
-	if (up->pending) {
-		/*
-		 * There are pending frames.
-		 * The socket lock must be held while it's corked.
-		 */
-		lock_sock(sk);
-		if (likely(up->pending)) {
-			if (unlikely(up->pending != AF_INET6)) {
-				release_sock(sk);
-				return -EAFNOSUPPORT;
-			}
-			dst = NULL;
-			goto do_append_data;
-		}
-		release_sock(sk);
-	}
-	ulen += sizeof(struct udphdr);
-
-	memset(&fl, 0, sizeof(fl));
-
-	if (sin6) {
-		if (sin6->sin6_port == 0)
-			return -EINVAL;
-
-		fl.fl_ip_dport = sin6->sin6_port;
-		daddr = &sin6->sin6_addr;
-
-		if (np->sndflow) {
-			fl.fl6_flowlabel = sin6->sin6_flowinfo&IPV6_FLOWINFO_MASK;
-			if (fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) {
-				flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel);
-				if (flowlabel == NULL)
-					return -EINVAL;
-				daddr = &flowlabel->dst;
-			}
-		}
-
-		/*
-		 * Otherwise it will be difficult to maintain
-		 * sk->sk_dst_cache.
-		 */
-		if (sk->sk_state == TCP_ESTABLISHED &&
-		    ipv6_addr_equal(daddr, &np->daddr))
-			daddr = &np->daddr;
-
-		if (addr_len >= sizeof(struct sockaddr_in6) &&
-		    sin6->sin6_scope_id &&
-		    ipv6_addr_type(daddr)&IPV6_ADDR_LINKLOCAL)
-			fl.oif = sin6->sin6_scope_id;
-	} else {
-		if (sk->sk_state != TCP_ESTABLISHED)
-			return -EDESTADDRREQ;
-
-		fl.fl_ip_dport = inet->dport;
-		daddr = &np->daddr;
-		fl.fl6_flowlabel = np->flow_label;
-		connected = 1;
-	}
-
-	if (!fl.oif)
-		fl.oif = sk->sk_bound_dev_if;
-
-	if (msg->msg_controllen) {
-		opt = &opt_space;
-		memset(opt, 0, sizeof(struct ipv6_txoptions));
-		opt->tot_len = sizeof(*opt);
-
-		err = datagram_send_ctl(msg, &fl, opt, &hlimit, &tclass);
-		if (err < 0) {
-			fl6_sock_release(flowlabel);
-			return err;
-		}
-		if ((fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) && !flowlabel) {
-			flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel);
-			if (flowlabel == NULL)
-				return -EINVAL;
-		}
-		if (!(opt->opt_nflen|opt->opt_flen))
-			opt = NULL;
-		connected = 0;
-	}
-	if (opt == NULL)
-		opt = np->opt;
-	if (flowlabel)
-		opt = fl6_merge_options(&opt_space, flowlabel, opt);
-	opt = ipv6_fixup_options(&opt_space, opt);
-
-	fl.proto = sk->sk_protocol;
-	ipv6_addr_copy(&fl.fl6_dst, daddr);
-	if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr))
-		ipv6_addr_copy(&fl.fl6_src, &np->saddr);
-	fl.fl_ip_sport = inet->sport;
-
-	/* merge ip6_build_xmit from ip6_output */
-	if (opt && opt->srcrt) {
-		struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
-		ipv6_addr_copy(&final, &fl.fl6_dst);
-		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-		final_p = &final;
-		connected = 0;
-	}
-
-	if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) {
-		fl.oif = np->mcast_oif;
-		connected = 0;
-	}
-
-	security_sk_classify_flow(sk, &fl);
-
-	err = ip6_sk_dst_lookup(sk, &dst, &fl);
-	if (err)
-		goto out;
-	if (final_p)
-		ipv6_addr_copy(&fl.fl6_dst, final_p);
-
-	if ((err = __xfrm_lookup(&dst, &fl, sk, XFRM_LOOKUP_WAIT)) < 0) {
-		if (err == -EREMOTE)
-			err = ip6_dst_blackhole(sk, &dst, &fl);
-		if (err < 0)
-			goto out;
-	}
-
-	if (hlimit < 0) {
-		if (ipv6_addr_is_multicast(&fl.fl6_dst))
-			hlimit = np->mcast_hops;
-		else
-			hlimit = np->hop_limit;
-		if (hlimit < 0)
-			hlimit = dst_metric(dst, RTAX_HOPLIMIT);
-		if (hlimit < 0)
-			hlimit = ipv6_get_hoplimit(dst->dev);
-	}
-
-	if (tclass < 0) {
-		tclass = np->tclass;
-		if (tclass < 0)
-			tclass = 0;
-	}
-
-	if (msg->msg_flags&MSG_CONFIRM)
-		goto do_confirm;
-back_from_confirm:
-
-	lock_sock(sk);
-	if (unlikely(up->pending)) {
-		/* The socket is already corked while preparing it. */
-		/* ... which is an evident application bug. --ANK */
-		release_sock(sk);
-
-		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n");
-		err = -EINVAL;
-		goto out;
-	}
-
-	up->pending = AF_INET6;
-
-do_append_data:
-	up->len += ulen;
-	getfrag  =  is_udplite ?  udplite_getfrag : ip_generic_getfrag;
-	err = ip6_append_data(sk, getfrag, msg->msg_iov, ulen,
-		sizeof(struct udphdr), hlimit, tclass, opt, &fl,
-		(struct rt6_info*)dst,
-		corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
-	if (err)
-		udp_v6_flush_pending_frames(sk);
-	else if (!corkreq)
-		err = udp_v6_push_pending_frames(sk);
-	else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
-		up->pending = 0;
-
-	if (dst) {
-		if (connected) {
-			ip6_dst_store(sk, dst,
-				      ipv6_addr_equal(&fl.fl6_dst, &np->daddr) ?
-				      &np->daddr : NULL,
-#ifdef CONFIG_IPV6_SUBTREES
-				      ipv6_addr_equal(&fl.fl6_src, &np->saddr) ?
-				      &np->saddr :
-#endif
-				      NULL);
-		} else {
-			dst_release(dst);
-		}
-	}
-
-	if (err > 0)
-		err = np->recverr ? net_xmit_errno(err) : 0;
-	release_sock(sk);
-out:
-	fl6_sock_release(flowlabel);
-	if (!err)
-		return len;
-	/*
-	 * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space.  Reporting
-	 * ENOBUFS might not be good (it's not tunable per se), but otherwise
-	 * we don't have a good statistic (IpOutDiscards but it can be too many
-	 * things).  We could add another new stat but at least for now that
-	 * seems like overkill.
-	 */
-	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
-		UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite);
-	}
-	return err;
-
-do_confirm:
-	dst_confirm(dst);
-	if (!(msg->msg_flags&MSG_PROBE) || len)
-		goto back_from_confirm;
-	err = 0;
-	goto out;
-}
-
-int udpv6_destroy_sock(struct sock *sk)
-{
-	lock_sock(sk);
-	udp_v6_flush_pending_frames(sk);
-	release_sock(sk);
-
-	inet6_destroy_sock(sk);
-
-	return 0;
-}
-
-/*
- *	Socket option code for UDP
- */
-int udpv6_setsockopt(struct sock *sk, int level, int optname,
-		     char __user *optval, int optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
-					  udp_v6_push_pending_frames);
-	return ipv6_setsockopt(sk, level, optname, optval, optlen);
-}
-
-#ifdef CONFIG_COMPAT
-int compat_udpv6_setsockopt(struct sock *sk, int level, int optname,
-			    char __user *optval, int optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
-					  udp_v6_push_pending_frames);
-	return compat_ipv6_setsockopt(sk, level, optname, optval, optlen);
-}
-#endif
-
-int udpv6_getsockopt(struct sock *sk, int level, int optname,
-		     char __user *optval, int __user *optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
-	return ipv6_getsockopt(sk, level, optname, optval, optlen);
-}
-
-#ifdef CONFIG_COMPAT
-int compat_udpv6_getsockopt(struct sock *sk, int level, int optname,
-			    char __user *optval, int __user *optlen)
-{
-	if (IS_SOL_UDPFAMILY(level))
-		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
-	return compat_ipv6_getsockopt(sk, level, optname, optval, optlen);
-}
-#endif
-
-static struct inet6_protocol udpv6_protocol = {
-	.handler	=	udpv6_rcv,
-	.err_handler	=	udpv6_err,
-	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
-};
-
-/* ------------------------------------------------------------------------ */
-#ifdef CONFIG_PROC_FS
-
-static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket)
-{
-	struct inet_sock *inet = inet_sk(sp);
-	struct ipv6_pinfo *np = inet6_sk(sp);
-	struct in6_addr *dest, *src;
-	__u16 destp, srcp;
-
-	dest  = &np->daddr;
-	src   = &np->rcv_saddr;
-	destp = ntohs(inet->dport);
-	srcp  = ntohs(inet->sport);
-	seq_printf(seq,
-		   "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
-		   "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p\n",
-		   bucket,
-		   src->s6_addr32[0], src->s6_addr32[1],
-		   src->s6_addr32[2], src->s6_addr32[3], srcp,
-		   dest->s6_addr32[0], dest->s6_addr32[1],
-		   dest->s6_addr32[2], dest->s6_addr32[3], destp,
-		   sp->sk_state,
-		   atomic_read(&sp->sk_wmem_alloc),
-		   atomic_read(&sp->sk_rmem_alloc),
-		   0, 0L, 0,
-		   sock_i_uid(sp), 0,
-		   sock_i_ino(sp),
-		   atomic_read(&sp->sk_refcnt), sp);
-}
-
-int udp6_seq_show(struct seq_file *seq, void *v)
-{
-	if (v == SEQ_START_TOKEN)
-		seq_printf(seq,
-			   "  sl  "
-			   "local_address                         "
-			   "remote_address                        "
-			   "st tx_queue rx_queue tr tm->when retrnsmt"
-			   "   uid  timeout inode\n");
-	else
-		udp6_sock_seq_show(seq, v, ((struct udp_iter_state *)seq->private)->bucket);
-	return 0;
-}
-
-static struct file_operations udp6_seq_fops;
-static struct udp_seq_afinfo udp6_seq_afinfo = {
-	.owner		= THIS_MODULE,
-	.name		= "udp6",
-	.family		= AF_INET6,
-	.hashtable	= udp_hash,
-	.seq_show	= udp6_seq_show,
-	.seq_fops	= &udp6_seq_fops,
-};
-
-int __init udp6_proc_init(void)
-{
-	return udp_proc_register(&udp6_seq_afinfo);
-}
-
-void udp6_proc_exit(void) {
-	udp_proc_unregister(&udp6_seq_afinfo);
-}
-#endif /* CONFIG_PROC_FS */
-
-/* ------------------------------------------------------------------------ */
-
-DEFINE_PROTO_INUSE(udpv6)
-
-struct proto udpv6_prot = {
-	.name		   = "UDPv6",
-	.owner		   = THIS_MODULE,
-	.close		   = udp_lib_close,
-	.connect	   = ip6_datagram_connect,
-	.disconnect	   = udp_disconnect,
-	.ioctl		   = udp_ioctl,
-	.destroy	   = udpv6_destroy_sock,
-	.setsockopt	   = udpv6_setsockopt,
-	.getsockopt	   = udpv6_getsockopt,
-	.sendmsg	   = udpv6_sendmsg,
-	.recvmsg	   = udpv6_recvmsg,
-	.backlog_rcv	   = udpv6_queue_rcv_skb,
-	.hash		   = udp_lib_hash,
-	.unhash		   = udp_lib_unhash,
-	.get_port	   = udp_v6_get_port,
-	.memory_allocated  = &udp_memory_allocated,
-	.sysctl_mem	   = sysctl_udp_mem,
-	.sysctl_wmem	   = &sysctl_udp_wmem_min,
-	.sysctl_rmem	   = &sysctl_udp_rmem_min,
-	.obj_size	   = sizeof(struct udp6_sock),
-#ifdef CONFIG_COMPAT
-	.compat_setsockopt = compat_udpv6_setsockopt,
-	.compat_getsockopt = compat_udpv6_getsockopt,
-#endif
-	REF_PROTO_INUSE(udpv6)
-};
-
-static struct inet_protosw udpv6_protosw = {
-	.type =      SOCK_DGRAM,
-	.protocol =  IPPROTO_UDP,
-	.prot =      &udpv6_prot,
-	.ops =       &inet6_dgram_ops,
-	.capability =-1,
-	.no_check =  UDP_CSUM_DEFAULT,
-	.flags =     INET_PROTOSW_PERMANENT,
-};
-
-
-int __init udpv6_init(void)
-{
-	int ret;
-
-	ret = inet6_add_protocol(&udpv6_protocol, IPPROTO_UDP);
-	if (ret)
-		goto out;
-
-	ret = inet6_register_protosw(&udpv6_protosw);
-	if (ret)
-		goto out_udpv6_protocol;
-out:
-	return ret;
-
-out_udpv6_protocol:
-	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
-	goto out;
-}
-
-void udpv6_exit(void)
-{
-	inet6_unregister_protosw(&udpv6_protosw);
-	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
-}
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
new file mode 100644
index 000000000000..87d4202522ee
--- /dev/null
+++ b/net/ipv6/udplite.c
@@ -0,0 +1,125 @@
+/*
+ *  UDPLITEv6   An implementation of the UDP-Lite protocol over IPv6.
+ *              See also net/ipv4/udplite.c
+ *
+ *  Version:    $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $
+ *
+ *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
+ *
+ *  Changes:
+ *  Fixes:
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ */
+#include "udp_impl.h"
+
+DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6) __read_mostly;
+
+static int udplitev6_rcv(struct sk_buff *skb)
+{
+	return __udp6_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE);
+}
+
+static void udplitev6_err(struct sk_buff *skb,
+			  struct inet6_skb_parm *opt,
+			  int type, int code, int offset, __be32 info)
+{
+	__udp6_lib_err(skb, opt, type, code, offset, info, udplite_hash);
+}
+
+static struct inet6_protocol udplitev6_protocol = {
+	.handler	=	udplitev6_rcv,
+	.err_handler	=	udplitev6_err,
+	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
+};
+
+static int udplite_v6_get_port(struct sock *sk, unsigned short snum)
+{
+	return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal);
+}
+
+DEFINE_PROTO_INUSE(udplitev6)
+
+struct proto udplitev6_prot = {
+	.name		   = "UDPLITEv6",
+	.owner		   = THIS_MODULE,
+	.close		   = udp_lib_close,
+	.connect	   = ip6_datagram_connect,
+	.disconnect	   = udp_disconnect,
+	.ioctl		   = udp_ioctl,
+	.init		   = udplite_sk_init,
+	.destroy	   = udpv6_destroy_sock,
+	.setsockopt	   = udpv6_setsockopt,
+	.getsockopt	   = udpv6_getsockopt,
+	.sendmsg	   = udpv6_sendmsg,
+	.recvmsg	   = udpv6_recvmsg,
+	.backlog_rcv	   = udpv6_queue_rcv_skb,
+	.hash		   = udp_lib_hash,
+	.unhash		   = udp_lib_unhash,
+	.get_port	   = udplite_v6_get_port,
+	.obj_size	   = sizeof(struct udp6_sock),
+#ifdef CONFIG_COMPAT
+	.compat_setsockopt = compat_udpv6_setsockopt,
+	.compat_getsockopt = compat_udpv6_getsockopt,
+#endif
+	REF_PROTO_INUSE(udplitev6)
+};
+
+static struct inet_protosw udplite6_protosw = {
+	.type		= SOCK_DGRAM,
+	.protocol	= IPPROTO_UDPLITE,
+	.prot		= &udplitev6_prot,
+	.ops		= &inet6_dgram_ops,
+	.capability	= -1,
+	.no_check	= 0,
+	.flags		= INET_PROTOSW_PERMANENT,
+};
+
+int __init udplitev6_init(void)
+{
+	int ret;
+
+	ret = inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
+	if (ret)
+		goto out;
+
+	ret = inet6_register_protosw(&udplite6_protosw);
+	if (ret)
+		goto out_udplitev6_protocol;
+out:
+	return ret;
+
+out_udplitev6_protocol:
+	inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
+	goto out;
+}
+
+void udplitev6_exit(void)
+{
+	inet6_unregister_protosw(&udplite6_protosw);
+	inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
+}
+
+#ifdef CONFIG_PROC_FS
+static struct file_operations udplite6_seq_fops;
+static struct udp_seq_afinfo udplite6_seq_afinfo = {
+	.owner		= THIS_MODULE,
+	.name		= "udplite6",
+	.family		= AF_INET6,
+	.hashtable	= udplite_hash,
+	.seq_show	= udp6_seq_show,
+	.seq_fops	= &udplite6_seq_fops,
+};
+
+int __init udplite6_proc_init(void)
+{
+	return udp_proc_register(&udplite6_seq_afinfo);
+}
+
+void udplite6_proc_exit(void)
+{
+	udp_proc_unregister(&udplite6_seq_afinfo);
+}
+#endif
diff --git a/net/ipv6/udplite_ipv6.c b/net/ipv6/udplite_ipv6.c
deleted file mode 100644
index 87d4202522ee..000000000000
--- a/net/ipv6/udplite_ipv6.c
+++ /dev/null
@@ -1,125 +0,0 @@
-/*
- *  UDPLITEv6   An implementation of the UDP-Lite protocol over IPv6.
- *              See also net/ipv4/udplite.c
- *
- *  Version:    $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $
- *
- *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
- *
- *  Changes:
- *  Fixes:
- *		This program is free software; you can redistribute it and/or
- *		modify it under the terms of the GNU General Public License
- *		as published by the Free Software Foundation; either version
- *		2 of the License, or (at your option) any later version.
- */
-#include "udp_impl.h"
-
-DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6) __read_mostly;
-
-static int udplitev6_rcv(struct sk_buff *skb)
-{
-	return __udp6_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE);
-}
-
-static void udplitev6_err(struct sk_buff *skb,
-			  struct inet6_skb_parm *opt,
-			  int type, int code, int offset, __be32 info)
-{
-	__udp6_lib_err(skb, opt, type, code, offset, info, udplite_hash);
-}
-
-static struct inet6_protocol udplitev6_protocol = {
-	.handler	=	udplitev6_rcv,
-	.err_handler	=	udplitev6_err,
-	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
-};
-
-static int udplite_v6_get_port(struct sock *sk, unsigned short snum)
-{
-	return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal);
-}
-
-DEFINE_PROTO_INUSE(udplitev6)
-
-struct proto udplitev6_prot = {
-	.name		   = "UDPLITEv6",
-	.owner		   = THIS_MODULE,
-	.close		   = udp_lib_close,
-	.connect	   = ip6_datagram_connect,
-	.disconnect	   = udp_disconnect,
-	.ioctl		   = udp_ioctl,
-	.init		   = udplite_sk_init,
-	.destroy	   = udpv6_destroy_sock,
-	.setsockopt	   = udpv6_setsockopt,
-	.getsockopt	   = udpv6_getsockopt,
-	.sendmsg	   = udpv6_sendmsg,
-	.recvmsg	   = udpv6_recvmsg,
-	.backlog_rcv	   = udpv6_queue_rcv_skb,
-	.hash		   = udp_lib_hash,
-	.unhash		   = udp_lib_unhash,
-	.get_port	   = udplite_v6_get_port,
-	.obj_size	   = sizeof(struct udp6_sock),
-#ifdef CONFIG_COMPAT
-	.compat_setsockopt = compat_udpv6_setsockopt,
-	.compat_getsockopt = compat_udpv6_getsockopt,
-#endif
-	REF_PROTO_INUSE(udplitev6)
-};
-
-static struct inet_protosw udplite6_protosw = {
-	.type		= SOCK_DGRAM,
-	.protocol	= IPPROTO_UDPLITE,
-	.prot		= &udplitev6_prot,
-	.ops		= &inet6_dgram_ops,
-	.capability	= -1,
-	.no_check	= 0,
-	.flags		= INET_PROTOSW_PERMANENT,
-};
-
-int __init udplitev6_init(void)
-{
-	int ret;
-
-	ret = inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
-	if (ret)
-		goto out;
-
-	ret = inet6_register_protosw(&udplite6_protosw);
-	if (ret)
-		goto out_udplitev6_protocol;
-out:
-	return ret;
-
-out_udplitev6_protocol:
-	inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
-	goto out;
-}
-
-void udplitev6_exit(void)
-{
-	inet6_unregister_protosw(&udplite6_protosw);
-	inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
-}
-
-#ifdef CONFIG_PROC_FS
-static struct file_operations udplite6_seq_fops;
-static struct udp_seq_afinfo udplite6_seq_afinfo = {
-	.owner		= THIS_MODULE,
-	.name		= "udplite6",
-	.family		= AF_INET6,
-	.hashtable	= udplite_hash,
-	.seq_show	= udp6_seq_show,
-	.seq_fops	= &udplite6_seq_fops,
-};
-
-int __init udplite6_proc_init(void)
-{
-	return udp_proc_register(&udplite6_seq_afinfo);
-}
-
-void udplite6_proc_exit(void)
-{
-	udp_proc_unregister(&udplite6_seq_afinfo);
-}
-#endif
-- 
cgit v1.2.3


From 1762f7e88eb34f653b4a915be99a102e347dd45e Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 7 Mar 2008 11:15:34 -0800
Subject: [NETNS][IPV6] ndisc - make socket control per namespace

Make ndisc socket control per namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |  1 +
 net/ipv6/ndisc.c         | 77 +++++++++++++++++++++++++++++++++---------------
 2 files changed, 54 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 90e6e24df858..b773256d0d1d 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -51,5 +51,6 @@ struct netns_ipv6 {
 	struct fib_rules_ops    *fib6_rules_ops;
 #endif
 	struct sock		**icmp_sk;
+	struct sock             *ndisc_sk;
 };
 #endif
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index a539b9ea53fd..24e76ed98884 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -89,8 +89,6 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter_ipv6.h>
 
-static struct socket *ndisc_socket;
-
 static u32 ndisc_hash(const void *pkey, const struct net_device *dev);
 static int ndisc_constructor(struct neighbour *neigh);
 static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb);
@@ -449,7 +447,8 @@ static void __ndisc_send(struct net_device *dev,
 {
 	struct flowi fl;
 	struct dst_entry *dst;
-	struct sock *sk = ndisc_socket->sk;
+	struct net *net = dev->nd_net;
+	struct sock *sk = net->ipv6.ndisc_sk;
 	struct sk_buff *skb;
 	struct icmp6hdr *hdr;
 	struct inet6_dev *idev;
@@ -459,8 +458,7 @@ static void __ndisc_send(struct net_device *dev,
 
 	type = icmp6h->icmp6_type;
 
-	icmpv6_flow_init(ndisc_socket->sk, &fl, type,
-			 saddr, daddr, dev->ifindex);
+	icmpv6_flow_init(sk, &fl, type, saddr, daddr, dev->ifindex);
 
 	dst = icmp6_dst_alloc(dev, neigh, daddr);
 	if (!dst)
@@ -1394,13 +1392,14 @@ static void ndisc_redirect_rcv(struct sk_buff *skb)
 void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
 			 struct in6_addr *target)
 {
-	struct sock *sk = ndisc_socket->sk;
+	struct net_device *dev = skb->dev;
+	struct net *net = dev->nd_net;
+	struct sock *sk = net->ipv6.ndisc_sk;
 	int len = sizeof(struct icmp6hdr) + 2 * sizeof(struct in6_addr);
 	struct sk_buff *buff;
 	struct icmp6hdr *icmph;
 	struct in6_addr saddr_buf;
 	struct in6_addr *addrp;
-	struct net_device *dev;
 	struct rt6_info *rt;
 	struct dst_entry *dst;
 	struct inet6_dev *idev;
@@ -1411,8 +1410,6 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
 	int hlen;
 	u8 ha_buf[MAX_ADDR_LEN], *ha = NULL;
 
-	dev = skb->dev;
-
 	if (ipv6_get_lladdr(dev, &saddr_buf, IFA_F_TENTATIVE)) {
 		ND_PRINTK2(KERN_WARNING
 			   "ICMPv6 Redirect: no link-local address on %s\n",
@@ -1427,10 +1424,10 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
 		return;
 	}
 
-	icmpv6_flow_init(ndisc_socket->sk, &fl, NDISC_REDIRECT,
+	icmpv6_flow_init(sk, &fl, NDISC_REDIRECT,
 			 &saddr_buf, &ipv6_hdr(skb)->saddr, dev->ifindex);
 
-	dst = ip6_route_output(&init_net, NULL, &fl);
+	dst = ip6_route_output(net, NULL, &fl);
 	if (dst == NULL)
 		return;
 
@@ -1719,22 +1716,24 @@ static int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name,
 
 #endif
 
-int __init ndisc_init(void)
+static int ndisc_net_init(struct net *net)
 {
+	struct socket *sock;
 	struct ipv6_pinfo *np;
 	struct sock *sk;
 	int err;
 
-	err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &ndisc_socket);
+	err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &sock);
 	if (err < 0) {
 		ND_PRINTK0(KERN_ERR
 			   "ICMPv6 NDISC: Failed to initialize the control socket (err %d).\n",
 			   err);
-		ndisc_socket = NULL; /* For safety. */
 		return err;
 	}
 
-	sk = ndisc_socket->sk;
+	net->ipv6.ndisc_sk = sk = sock->sk;
+	sk_change_net(sk, net);
+
 	np = inet6_sk(sk);
 	sk->sk_allocation = GFP_ATOMIC;
 	np->hop_limit = 255;
@@ -1742,21 +1741,52 @@ int __init ndisc_init(void)
 	np->mc_loop = 0;
 	sk->sk_prot->unhash(sk);
 
+	return 0;
+}
+
+static void ndisc_net_exit(struct net *net)
+{
+	sk_release_kernel(net->ipv6.ndisc_sk);
+}
+
+static struct pernet_operations ndisc_net_ops = {
+	.init = ndisc_net_init,
+	.exit = ndisc_net_exit,
+};
+
+int __init ndisc_init(void)
+{
+	int err;
+
+	err = register_pernet_subsys(&ndisc_net_ops);
+	if (err)
+		return err;
 	/*
 	 * Initialize the neighbour table
 	 */
-
 	neigh_table_init(&nd_tbl);
 
 #ifdef CONFIG_SYSCTL
-	neigh_sysctl_register(NULL, &nd_tbl.parms, NET_IPV6, NET_IPV6_NEIGH,
-			      "ipv6",
-			      &ndisc_ifinfo_sysctl_change,
-			      &ndisc_ifinfo_sysctl_strategy);
+	err = neigh_sysctl_register(NULL, &nd_tbl.parms, NET_IPV6,
+				    NET_IPV6_NEIGH, "ipv6",
+				    &ndisc_ifinfo_sysctl_change,
+				    &ndisc_ifinfo_sysctl_strategy);
+	if (err)
+		goto out_unregister_pernet;
 #endif
+	err = register_netdevice_notifier(&ndisc_netdev_notifier);
+	if (err)
+		goto out_unregister_sysctl;
+out:
+	return err;
 
-	register_netdevice_notifier(&ndisc_netdev_notifier);
-	return 0;
+out_unregister_sysctl:
+#ifdef CONFIG_SYSCTL
+	neigh_sysctl_unregister(&nd_tbl.parms);
+out_unregister_pernet:
+#endif
+	unregister_pernet_subsys(&ndisc_net_ops);
+	goto out;
 }
 
 void ndisc_cleanup(void)
@@ -1766,6 +1796,5 @@ void ndisc_cleanup(void)
 	neigh_sysctl_unregister(&nd_tbl.parms);
 #endif
 	neigh_table_clear(&nd_tbl);
-	sock_release(ndisc_socket);
-	ndisc_socket = NULL; /* For safety. */
+	unregister_pernet_subsys(&ndisc_net_ops);
 }
-- 
cgit v1.2.3


From 93ec926b075d14bb7edcbc054e92199e9ad4ad1d Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 7 Mar 2008 11:16:02 -0800
Subject: [NETNS][IPV6] tcp6 - make socket control per namespace

Instead of having a tcp6_socket global to all the namespace, there is
tcp6 socket control per namespace. That is consistent with which
namespace sent a RST and allows to pass the socket to the underlying
function to retrieve the network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |  1 +
 net/ipv6/tcp_ipv6.c      | 43 +++++++++++++++++++++++++++++++++----------
 2 files changed, 34 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index b773256d0d1d..ddb9ccddcc4d 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -52,5 +52,6 @@ struct netns_ipv6 {
 #endif
 	struct sock		**icmp_sk;
 	struct sock             *ndisc_sk;
+	struct sock             *tcp_sk;
 };
 #endif
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index b47cce8fea44..7954b4e99c4f 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -69,9 +69,6 @@
 #include <linux/crypto.h>
 #include <linux/scatterlist.h>
 
-/* Socket used for sending RSTs and ACKs */
-static struct socket *tcp6_socket;
-
 static void	tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb);
 static void	tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req);
 static void	tcp_v6_send_check(struct sock *sk, int len,
@@ -1075,10 +1072,11 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 	 * Underlying function will use this to retrieve the network
 	 * namespace
 	 */
-	if (!ip6_dst_lookup(tcp6_socket->sk, &buff->dst, &fl)) {
+	if (!ip6_dst_lookup(init_net.ipv6.tcp_sk, &buff->dst, &fl)) {
 
 		if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
-			ip6_xmit(tcp6_socket->sk, buff, &fl, NULL, 0);
+			ip6_xmit(init_net.ipv6.tcp_sk,
+				 buff, &fl, NULL, 0);
 			TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
 			TCP_INC_STATS_BH(TCP_MIB_OUTRSTS);
 			return;
@@ -1175,9 +1173,10 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw,
 	fl.fl_ip_sport = t1->source;
 	security_skb_classify_flow(skb, &fl);
 
-	if (!ip6_dst_lookup(tcp6_socket->sk, &buff->dst, &fl)) {
+	if (!ip6_dst_lookup(init_net.ipv6.tcp_sk, &buff->dst, &fl)) {
 		if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
-			ip6_xmit(tcp6_socket->sk, buff, &fl, NULL, 0);
+			ip6_xmit(init_net.ipv6.tcp_sk,
+				 buff, &fl, NULL, 0);
 			TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
 			return;
 		}
@@ -2198,6 +2197,31 @@ static struct inet_protosw tcpv6_protosw = {
 				INET_PROTOSW_ICSK,
 };
 
+static int tcpv6_net_init(struct net *net)
+{
+	int err;
+	struct socket *sock;
+	struct sock *sk;
+
+	err = inet_csk_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP);
+	if (err)
+		return err;
+
+	net->ipv6.tcp_sk = sk = sock->sk;
+	sk_change_net(sk, net);
+	return err;
+}
+
+static void tcpv6_net_exit(struct net *net)
+{
+	sk_release_kernel(net->ipv6.tcp_sk);
+}
+
+static struct pernet_operations tcpv6_net_ops = {
+	.init = tcpv6_net_init,
+	.exit = tcpv6_net_exit,
+};
+
 int __init tcpv6_init(void)
 {
 	int ret;
@@ -2211,8 +2235,7 @@ int __init tcpv6_init(void)
 	if (ret)
 		goto out_tcpv6_protocol;
 
-	ret = inet_csk_ctl_sock_create(&tcp6_socket, PF_INET6,
-				       SOCK_RAW, IPPROTO_TCP);
+	ret = register_pernet_subsys(&tcpv6_net_ops);
 	if (ret)
 		goto out_tcpv6_protosw;
 out:
@@ -2227,7 +2250,7 @@ out_tcpv6_protosw:
 
 void tcpv6_exit(void)
 {
-	sock_release(tcp6_socket);
+	unregister_pernet_subsys(&tcpv6_net_ops);
 	inet6_unregister_protosw(&tcpv6_protosw);
 	inet6_del_protocol(&tcpv6_protocol, IPPROTO_TCP);
 }
-- 
cgit v1.2.3


From b8ad0cbc58f703972e9e37c4e2a8081dd7e6a551 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 7 Mar 2008 11:16:55 -0800
Subject: [NETNS][IPV6] mcast - handle several network namespace

This patch make use of the network namespace information at the right
places to handle the multicast for several network namespaces.  It
makes the socket control to be per namespace too.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv6.h |   1 +
 net/ipv6/mcast.c         | 116 +++++++++++++++++++++++++++++++----------------
 net/ipv6/udp.c           |   3 ++
 3 files changed, 82 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index ddb9ccddcc4d..ac053be6c256 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -53,5 +53,6 @@ struct netns_ipv6 {
 	struct sock		**icmp_sk;
 	struct sock             *ndisc_sk;
 	struct sock             *tcp_sk;
+	struct sock             *igmp_sk;
 };
 #endif
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 197ca390a15d..f2879056fab0 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -126,8 +126,6 @@ static struct in6_addr mld2_all_mcr = MLD2_ALL_MCR_INIT;
 /* Big mc list lock for all the sockets */
 static DEFINE_RWLOCK(ipv6_sk_mc_lock);
 
-static struct socket *igmp6_socket;
-
 int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr);
 
 static void igmp6_join_group(struct ifmcaddr6 *ma);
@@ -183,6 +181,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 	struct net_device *dev = NULL;
 	struct ipv6_mc_socklist *mc_lst;
 	struct ipv6_pinfo *np = inet6_sk(sk);
+	struct net *net = sk->sk_net;
 	int err;
 
 	if (!ipv6_addr_is_multicast(addr))
@@ -208,14 +207,14 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 
 	if (ifindex == 0) {
 		struct rt6_info *rt;
-		rt = rt6_lookup(&init_net, addr, NULL, 0, 0);
+		rt = rt6_lookup(net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
 			dst_release(&rt->u.dst);
 		}
 	} else
-		dev = dev_get_by_index(&init_net, ifindex);
+		dev = dev_get_by_index(net, ifindex);
 
 	if (dev == NULL) {
 		sock_kfree_s(sk, mc_lst, sizeof(*mc_lst));
@@ -256,6 +255,7 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6_mc_socklist *mc_lst, **lnk;
+	struct net *net = sk->sk_net;
 
 	write_lock_bh(&ipv6_sk_mc_lock);
 	for (lnk = &np->ipv6_mc_list; (mc_lst = *lnk) !=NULL ; lnk = &mc_lst->next) {
@@ -266,7 +266,8 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
 			*lnk = mc_lst->next;
 			write_unlock_bh(&ipv6_sk_mc_lock);
 
-			if ((dev = dev_get_by_index(&init_net, mc_lst->ifindex)) != NULL) {
+			dev = dev_get_by_index(net, mc_lst->ifindex);
+			if (dev != NULL) {
 				struct inet6_dev *idev = in6_dev_get(dev);
 
 				(void) ip6_mc_leave_src(sk, mc_lst, idev);
@@ -286,7 +287,9 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
 	return -EADDRNOTAVAIL;
 }
 
-static struct inet6_dev *ip6_mc_find_dev(struct in6_addr *group, int ifindex)
+static struct inet6_dev *ip6_mc_find_dev(struct net *net,
+					 struct in6_addr *group,
+					 int ifindex)
 {
 	struct net_device *dev = NULL;
 	struct inet6_dev *idev = NULL;
@@ -294,14 +297,14 @@ static struct inet6_dev *ip6_mc_find_dev(struct in6_addr *group, int ifindex)
 	if (ifindex == 0) {
 		struct rt6_info *rt;
 
-		rt = rt6_lookup(&init_net, group, NULL, 0, 0);
+		rt = rt6_lookup(net, group, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
 			dst_release(&rt->u.dst);
 		}
 	} else
-		dev = dev_get_by_index(&init_net, ifindex);
+		dev = dev_get_by_index(net, ifindex);
 
 	if (!dev)
 		return NULL;
@@ -324,6 +327,7 @@ void ipv6_sock_mc_close(struct sock *sk)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6_mc_socklist *mc_lst;
+	struct net *net = sk->sk_net;
 
 	write_lock_bh(&ipv6_sk_mc_lock);
 	while ((mc_lst = np->ipv6_mc_list) != NULL) {
@@ -332,7 +336,7 @@ void ipv6_sock_mc_close(struct sock *sk)
 		np->ipv6_mc_list = mc_lst->next;
 		write_unlock_bh(&ipv6_sk_mc_lock);
 
-		dev = dev_get_by_index(&init_net, mc_lst->ifindex);
+		dev = dev_get_by_index(net, mc_lst->ifindex);
 		if (dev) {
 			struct inet6_dev *idev = in6_dev_get(dev);
 
@@ -361,6 +365,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
 	struct inet6_dev *idev;
 	struct ipv6_pinfo *inet6 = inet6_sk(sk);
 	struct ip6_sf_socklist *psl;
+	struct net *net = sk->sk_net;
 	int i, j, rv;
 	int leavegroup = 0;
 	int pmclocked = 0;
@@ -376,7 +381,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
 	if (!ipv6_addr_is_multicast(group))
 		return -EINVAL;
 
-	idev = ip6_mc_find_dev(group, pgsr->gsr_interface);
+	idev = ip6_mc_find_dev(net, group, pgsr->gsr_interface);
 	if (!idev)
 		return -ENODEV;
 	dev = idev->dev;
@@ -500,6 +505,7 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf)
 	struct inet6_dev *idev;
 	struct ipv6_pinfo *inet6 = inet6_sk(sk);
 	struct ip6_sf_socklist *newpsl, *psl;
+	struct net *net = sk->sk_net;
 	int leavegroup = 0;
 	int i, err;
 
@@ -511,7 +517,7 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf)
 	    gsf->gf_fmode != MCAST_EXCLUDE)
 		return -EINVAL;
 
-	idev = ip6_mc_find_dev(group, gsf->gf_interface);
+	idev = ip6_mc_find_dev(net, group, gsf->gf_interface);
 
 	if (!idev)
 		return -ENODEV;
@@ -592,13 +598,14 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 	struct net_device *dev;
 	struct ipv6_pinfo *inet6 = inet6_sk(sk);
 	struct ip6_sf_socklist *psl;
+	struct net *net = sk->sk_net;
 
 	group = &((struct sockaddr_in6 *)&gsf->gf_group)->sin6_addr;
 
 	if (!ipv6_addr_is_multicast(group))
 		return -EINVAL;
 
-	idev = ip6_mc_find_dev(group, gsf->gf_interface);
+	idev = ip6_mc_find_dev(net, group, gsf->gf_interface);
 
 	if (!idev)
 		return -ENODEV;
@@ -1393,7 +1400,8 @@ mld_scount(struct ifmcaddr6 *pmc, int type, int gdeleted, int sdeleted)
 
 static struct sk_buff *mld_newpack(struct net_device *dev, int size)
 {
-	struct sock *sk = igmp6_socket->sk;
+	struct net *net = dev->nd_net;
+	struct sock *sk = net->ipv6.igmp_sk;
 	struct sk_buff *skb;
 	struct mld2_report *pmr;
 	struct in6_addr addr_buf;
@@ -1440,6 +1448,7 @@ static void mld_sendpack(struct sk_buff *skb)
 			      (struct mld2_report *)skb_transport_header(skb);
 	int payload_len, mldlen;
 	struct inet6_dev *idev = in6_dev_get(skb->dev);
+	struct net *net = skb->dev->nd_net;
 	int err;
 	struct flowi fl;
 
@@ -1459,7 +1468,7 @@ static void mld_sendpack(struct sk_buff *skb)
 		goto err_out;
 	}
 
-	icmpv6_flow_init(igmp6_socket->sk, &fl, ICMPV6_MLD2_REPORT,
+	icmpv6_flow_init(net->ipv6.igmp_sk, &fl, ICMPV6_MLD2_REPORT,
 			 &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr,
 			 skb->dev->ifindex);
 
@@ -1753,7 +1762,8 @@ static void mld_send_cr(struct inet6_dev *idev)
 
 static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 {
-	struct sock *sk = igmp6_socket->sk;
+	struct net *net = dev->nd_net;
+	struct sock *sk = net->ipv6.igmp_sk;
 	struct inet6_dev *idev;
 	struct sk_buff *skb;
 	struct icmp6hdr *hdr;
@@ -1824,7 +1834,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 		goto err_out;
 	}
 
-	icmpv6_flow_init(igmp6_socket->sk, &fl, type,
+	icmpv6_flow_init(sk, &fl, type,
 			 &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr,
 			 skb->dev->ifindex);
 
@@ -2334,6 +2344,7 @@ void ipv6_mc_destroy_dev(struct inet6_dev *idev)
 
 #ifdef CONFIG_PROC_FS
 struct igmp6_mc_iter_state {
+	struct seq_net_private p;
 	struct net_device *dev;
 	struct inet6_dev *idev;
 };
@@ -2344,9 +2355,10 @@ static inline struct ifmcaddr6 *igmp6_mc_get_first(struct seq_file *seq)
 {
 	struct ifmcaddr6 *im = NULL;
 	struct igmp6_mc_iter_state *state = igmp6_mc_seq_private(seq);
+	struct net *net = state->p.net;
 
 	state->idev = NULL;
-	for_each_netdev(&init_net, state->dev) {
+	for_each_netdev(net, state->dev) {
 		struct inet6_dev *idev;
 		idev = in6_dev_get(state->dev);
 		if (!idev)
@@ -2448,8 +2460,8 @@ static const struct seq_operations igmp6_mc_seq_ops = {
 
 static int igmp6_mc_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open_private(file, &igmp6_mc_seq_ops,
-			sizeof(struct igmp6_mc_iter_state));
+	return seq_open_net(inode, file, &igmp6_mc_seq_ops,
+			    sizeof(struct igmp6_mc_iter_state));
 }
 
 static const struct file_operations igmp6_mc_seq_fops = {
@@ -2457,10 +2469,11 @@ static const struct file_operations igmp6_mc_seq_fops = {
 	.open		=	igmp6_mc_seq_open,
 	.read		=	seq_read,
 	.llseek		=	seq_lseek,
-	.release	=	seq_release_private,
+	.release	=	seq_release_net,
 };
 
 struct igmp6_mcf_iter_state {
+	struct seq_net_private p;
 	struct net_device *dev;
 	struct inet6_dev *idev;
 	struct ifmcaddr6 *im;
@@ -2473,10 +2486,11 @@ static inline struct ip6_sf_list *igmp6_mcf_get_first(struct seq_file *seq)
 	struct ip6_sf_list *psf = NULL;
 	struct ifmcaddr6 *im = NULL;
 	struct igmp6_mcf_iter_state *state = igmp6_mcf_seq_private(seq);
+	struct net *net = state->p.net;
 
 	state->idev = NULL;
 	state->im = NULL;
-	for_each_netdev(&init_net, state->dev) {
+	for_each_netdev(net, state->dev) {
 		struct inet6_dev *idev;
 		idev = in6_dev_get(state->dev);
 		if (unlikely(idev == NULL))
@@ -2608,8 +2622,8 @@ static const struct seq_operations igmp6_mcf_seq_ops = {
 
 static int igmp6_mcf_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open_private(file, &igmp6_mcf_seq_ops,
-			sizeof(struct igmp6_mcf_iter_state));
+	return seq_open_net(inode, file, &igmp6_mcf_seq_ops,
+			    sizeof(struct igmp6_mcf_iter_state));
 }
 
 static const struct file_operations igmp6_mcf_seq_fops = {
@@ -2617,26 +2631,27 @@ static const struct file_operations igmp6_mcf_seq_fops = {
 	.open		=	igmp6_mcf_seq_open,
 	.read		=	seq_read,
 	.llseek		=	seq_lseek,
-	.release	=	seq_release_private,
+	.release	=	seq_release_net,
 };
 #endif
 
-int __init igmp6_init(void)
+static int igmp6_net_init(struct net *net)
 {
 	struct ipv6_pinfo *np;
+	struct socket *sock;
 	struct sock *sk;
 	int err;
 
-	err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &igmp6_socket);
+	err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &sock);
 	if (err < 0) {
 		printk(KERN_ERR
 		       "Failed to initialize the IGMP6 control socket (err %d).\n",
 		       err);
-		igmp6_socket = NULL; /* For safety. */
-		return err;
+		goto out;
 	}
 
-	sk = igmp6_socket->sk;
+	net->ipv6.igmp_sk = sk = sock->sk;
+	sk_change_net(sk, net);
 	sk->sk_allocation = GFP_ATOMIC;
 	sk->sk_prot->unhash(sk);
 
@@ -2644,20 +2659,45 @@ int __init igmp6_init(void)
 	np->hop_limit = 1;
 
 #ifdef CONFIG_PROC_FS
-	proc_net_fops_create(&init_net, "igmp6", S_IRUGO, &igmp6_mc_seq_fops);
-	proc_net_fops_create(&init_net, "mcfilter6", S_IRUGO, &igmp6_mcf_seq_fops);
+	err = -ENOMEM;
+	if (!proc_net_fops_create(net, "igmp6", S_IRUGO, &igmp6_mc_seq_fops))
+		goto out_sock_create;
+	if (!proc_net_fops_create(net, "mcfilter6", S_IRUGO,
+				  &igmp6_mcf_seq_fops)) {
+		proc_net_remove(net, "igmp6");
+		goto out_sock_create;
+	}
 #endif
 
-	return 0;
+	err = 0;
+out:
+	return err;
+
+out_sock_create:
+	sk_release_kernel(net->ipv6.igmp_sk);
+	goto out;
 }
 
-void igmp6_cleanup(void)
+static void igmp6_net_exit(struct net *net)
 {
-	sock_release(igmp6_socket);
-	igmp6_socket = NULL; /* for safety */
-
+	sk_release_kernel(net->ipv6.igmp_sk);
 #ifdef CONFIG_PROC_FS
-	proc_net_remove(&init_net, "mcfilter6");
-	proc_net_remove(&init_net, "igmp6");
+	proc_net_remove(net, "mcfilter6");
+	proc_net_remove(net, "igmp6");
 #endif
 }
+
+static struct pernet_operations igmp6_net_ops = {
+	.init = igmp6_net_init,
+	.exit = igmp6_net_exit,
+};
+
+int __init igmp6_init(void)
+{
+	return register_pernet_subsys(&igmp6_net_ops);
+}
+
+void igmp6_cleanup(void)
+{
+	unregister_pernet_subsys(&igmp6_net_ops);
+}
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 53739de829db..d6e311f6c8eb 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -323,6 +323,9 @@ static struct sock *udp_v6_mcast_next(struct sock *sk,
 	sk_for_each_from(s, node) {
 		struct inet_sock *inet = inet_sk(s);
 
+		if (s->sk_net != sk->sk_net)
+			continue;
+
 		if (s->sk_hash == num && s->sk_family == PF_INET6) {
 			struct ipv6_pinfo *np = inet6_sk(s);
 			if (inet->dport) {
-- 
cgit v1.2.3


From 11f4b1cec98ad95abda80dc20bdc3cecac145d77 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Tue, 4 Mar 2008 18:09:26 -0800
Subject: mac80211: adding mac80211_tx_control_flags and HT flags

This patch makes enum from the defines previously dwelled inside
ieee80211_tx_control for better readability.
The patch also addes HT flags, for 802.11n drivers:
- IEEE80211_TXCTL_OFDM_HT: request low-level driver to use HT OFDM rates
- IEEE80211_TXCTL_GREEN_FIELD: use green field protection
- IEEE80211_TXCTL_DUP_DATA: duplicate data on both 20 Mhz channels
- IEEE80211_TXCTL_40_MHZ_WIDTH: send this frame in 40Mhz width
- IEEE80211_TXCTL_SHORT_GI: send this frame with short guard interval

Tx command can be a combination of any of these flags, along with
bitrate represented by ieee80211_rate. this will allow legacy drivers to
switch easily to any 11n rate representation.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
CC: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 86 ++++++++++++++++++++++++++++++++------------------
 1 file changed, 55 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index d002b1c6e78e..cde1b4953534 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -205,6 +205,58 @@ struct ieee80211_bss_conf {
 	bool use_short_preamble;
 };
 
+/**
+ * enum mac80211_tx_control_flags - flags to describe Tx configuration for
+ * 				    the Tx frame
+ *
+ * These flags are used with the @flags member of &ieee80211_tx_control
+ *
+ * @IEEE80211_TXCTL_REQ_TX_STATUS: request TX status callback for this frame.
+ * @IEEE80211_TXCTL_DO_NOT_ENCRYPT: send this frame without encryption;
+ * 				    e.g., for EAPOL frame
+ * @IEEE80211_TXCTL_USE_RTS_CTS: use RTS-CTS before sending frame
+ * @IEEE80211_TXCTL_USE_CTS_PROTECT: use CTS protection for the frame (e.g.,
+ * 				     for combined 802.11g / 802.11b networks)
+ * @IEEE80211_TXCTL_NO_ACK: tell the low level not to wait for an ack
+ * @IEEE80211_TXCTL_RATE_CTRL_PROBE
+ * @EEE80211_TXCTL_CLEAR_PS_FILT: clear powersave filter
+ *                                 for destination station
+ * @IEEE80211_TXCTL_REQUEUE:
+ * @IEEE80211_TXCTL_FIRST_FRAGMENT: this is a first fragment of the frame
+ * @IEEE80211_TXCTL_LONG_RETRY_LIMIT: this frame should be send using the
+ * 				      through set_retry_limit configured long
+ * 				      retry value
+ * @IEEE80211_TXCTL_EAPOL_FRAME: internal to mac80211
+ * @IEEE80211_TXCTL_SEND_AFTER_DTIM: send this frame after DTIM beacon
+ * @IEEE80211_TXCTL_AMPDU: this frame should be sent as part of an A-MPDU
+ * @IEEE80211_TXCTL_OFDM_HT: this frame can be sent in HT OFDM rates
+ * @IEEE80211_TXCTL_GREEN_FIELD: use green field protection for this frame
+ * @IEEE80211_TXCTL_40_MHZ_WIDTH: send this frame using 40 Mhz channel width
+ * @IEEE80211_TXCTL_DUP_DATA: duplicate data frame on both 20 Mhz channels
+ * @IEEE80211_TXCTL_SHORT_GI: send this frame using short guard interval
+ */
+enum mac80211_tx_control_flags {
+	IEEE80211_TXCTL_REQ_TX_STATUS		= (1<<0),
+	IEEE80211_TXCTL_DO_NOT_ENCRYPT		= (1<<1),
+	IEEE80211_TXCTL_USE_RTS_CTS		= (1<<2),
+	IEEE80211_TXCTL_USE_CTS_PROTECT		= (1<<3),
+	IEEE80211_TXCTL_NO_ACK			= (1<<4),
+	IEEE80211_TXCTL_RATE_CTRL_PROBE		= (1<<5),
+	IEEE80211_TXCTL_CLEAR_PS_FILT		= (1<<6),
+	IEEE80211_TXCTL_REQUEUE			= (1<<7),
+	IEEE80211_TXCTL_FIRST_FRAGMENT		= (1<<8),
+	IEEE80211_TXCTL_SHORT_PREAMBLE		= (1<<9),
+	IEEE80211_TXCTL_LONG_RETRY_LIMIT	= (1<<10),
+	IEEE80211_TXCTL_EAPOL_FRAME		= (1<<11),
+	IEEE80211_TXCTL_SEND_AFTER_DTIM		= (1<<12),
+	IEEE80211_TXCTL_AMPDU			= (1<<13),
+	IEEE80211_TXCTL_OFDM_HT			= (1<<14),
+	IEEE80211_TXCTL_GREEN_FIELD		= (1<<15),
+	IEEE80211_TXCTL_40_MHZ_WIDTH		= (1<<16),
+	IEEE80211_TXCTL_DUP_DATA		= (1<<17),
+	IEEE80211_TXCTL_SHORT_GI		= (1<<18),
+};
+
 /* Transmit control fields. This data structure is passed to low-level driver
  * with each TX frame. The low-level driver is responsible for configuring
  * the hardware to use given values (depending on what is supported). */
@@ -219,42 +271,14 @@ struct ieee80211_tx_control {
 	/* retry rate for the last retries */
 	struct ieee80211_rate *alt_retry_rate;
 
-#define IEEE80211_TXCTL_REQ_TX_STATUS	(1<<0)/* request TX status callback for
-						* this frame */
-#define IEEE80211_TXCTL_DO_NOT_ENCRYPT	(1<<1) /* send this frame without
-						* encryption; e.g., for EAPOL
-						* frames */
-#define IEEE80211_TXCTL_USE_RTS_CTS	(1<<2) /* use RTS-CTS before sending
-						* frame */
-#define IEEE80211_TXCTL_USE_CTS_PROTECT	(1<<3) /* use CTS protection for the
-						* frame (e.g., for combined
-						* 802.11g / 802.11b networks) */
-#define IEEE80211_TXCTL_NO_ACK		(1<<4) /* tell the low level not to
-						* wait for an ack */
-#define IEEE80211_TXCTL_RATE_CTRL_PROBE	(1<<5)
-#define IEEE80211_TXCTL_CLEAR_PS_FILT	(1<<6) /* clear powersave filter
-						* for destination station */
-#define IEEE80211_TXCTL_REQUEUE		(1<<7)
-#define IEEE80211_TXCTL_FIRST_FRAGMENT	(1<<8) /* this is a first fragment of
-						* the frame */
-#define IEEE80211_TXCTL_SHORT_PREAMBLE	(1<<9)
-#define IEEE80211_TXCTL_LONG_RETRY_LIMIT (1<<10) /* this frame should be send
-						  * using the through
-						  * set_retry_limit configured
-						  * long retry value */
-#define IEEE80211_TXCTL_EAPOL_FRAME	(1<<11) /* internal to mac80211 */
-#define IEEE80211_TXCTL_SEND_AFTER_DTIM	(1<<12) /* send this frame after DTIM
-						 * beacon */
-#define IEEE80211_TXCTL_AMPDU		(1<<13) /* this frame should be sent
-						 * as part of an A-MPDU */
-	u32 flags;			       /* tx control flags defined
-						* above */
+	u32 flags;		/* tx control flags defined above */
 	u8 key_idx;		/* keyidx from hw->set_key(), undefined if
 				 * IEEE80211_TXCTL_DO_NOT_ENCRYPT is set */
 	u8 retry_limit;		/* 1 = only first attempt, 2 = one retry, ..
 				 * This could be used when set_retry_limit
 				 * is not implemented by the driver */
-	u8 antenna_sel_tx; 	/* 0 = default/diversity, 1 = Ant0, 2 = Ant1 */
+	u8 antenna_sel_tx; 	/* 0 = default/diversity, otherwise bit
+				 * position represents antenna number used */
 	u8 icv_len;		/* length of the ICV/MIC field in octets */
 	u8 iv_len;		/* length of the IV field in octets */
 	u8 queue;		/* hardware queue to use for this frame;
-- 
cgit v1.2.3


From 6c5ef8a7059e4f7adc37b337face8b0a8cbd4f48 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Wed, 5 Mar 2008 17:38:59 +0200
Subject: mac80211: document IEEE80211_TXCTL_OFDM_HT

This patch clarifies the use of IEEE80211_TXCTL_OFDM_HT flag.

Can by united with patch "mac80211: adding mac80211_tx_control
flags and HT flags"

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index cde1b4953534..5ab6a350ee6d 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -229,7 +229,11 @@ struct ieee80211_bss_conf {
  * @IEEE80211_TXCTL_EAPOL_FRAME: internal to mac80211
  * @IEEE80211_TXCTL_SEND_AFTER_DTIM: send this frame after DTIM beacon
  * @IEEE80211_TXCTL_AMPDU: this frame should be sent as part of an A-MPDU
- * @IEEE80211_TXCTL_OFDM_HT: this frame can be sent in HT OFDM rates
+ * @IEEE80211_TXCTL_OFDM_HT: this frame can be sent in HT OFDM rates. number
+ * 			     of streams when this flag is on can be extracted
+ *			     from antenna_sel_tx, so if 1 antenna is marked
+ *			     use SISO, 2 antennas marked use MIMO, n antennas
+ *			     marked use MIMO_n.
  * @IEEE80211_TXCTL_GREEN_FIELD: use green field protection for this frame
  * @IEEE80211_TXCTL_40_MHZ_WIDTH: send this frame using 40 Mhz channel width
  * @IEEE80211_TXCTL_DUP_DATA: duplicate data frame on both 20 Mhz channels
-- 
cgit v1.2.3


From c70f8f68676866d778564de337bec6b8734c3850 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Thu, 28 Feb 2008 16:47:50 +0800
Subject: [IA64] regset: 64-bit support

This is the 64-bit regset implementation under IA64. Basically register
read/write, which is derived from current ptrace register read/write.

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ptrace.c | 690 ++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-ia64/elf.h    |  24 ++
 2 files changed, 714 insertions(+)

(limited to 'include')

diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index ab784ec4319d..7136c7811efc 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -3,6 +3,9 @@
  *
  * Copyright (C) 1999-2005 Hewlett-Packard Co
  *	David Mosberger-Tang <davidm@hpl.hp.com>
+ * Copyright (C) 2006 Intel Co
+ *  2006-08-12	- IA64 Native Utrace implementation support added by
+ *	Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
  *
  * Derived from the x86 and Alpha versions.
  */
@@ -17,6 +20,8 @@
 #include <linux/security.h>
 #include <linux/audit.h>
 #include <linux/signal.h>
+#include <linux/regset.h>
+#include <linux/elf.h>
 
 #include <asm/pgtable.h>
 #include <asm/processor.h>
@@ -1626,3 +1631,688 @@ syscall_trace_leave (long arg0, long arg1, long arg2, long arg3,
 	if (test_thread_flag(TIF_RESTORE_RSE))
 		ia64_sync_krbs();
 }
+
+/* Utrace implementation starts here */
+struct regset_get {
+	void *kbuf;
+	void __user *ubuf;
+};
+
+struct regset_set {
+	const void *kbuf;
+	const void __user *ubuf;
+};
+
+struct regset_getset {
+	struct task_struct *target;
+	const struct user_regset *regset;
+	union {
+		struct regset_get get;
+		struct regset_set set;
+	} u;
+	unsigned int pos;
+	unsigned int count;
+	int ret;
+};
+
+static int
+access_elf_gpreg(struct task_struct *target, struct unw_frame_info *info,
+		unsigned long addr, unsigned long *data, int write_access)
+{
+	struct pt_regs *pt;
+	unsigned long *ptr = NULL;
+	int ret;
+	char nat = 0;
+
+	pt = task_pt_regs(target);
+	switch (addr) {
+	case ELF_GR_OFFSET(1):
+		ptr = &pt->r1;
+		break;
+	case ELF_GR_OFFSET(2):
+	case ELF_GR_OFFSET(3):
+		ptr = (void *)&pt->r2 + (addr - ELF_GR_OFFSET(2));
+		break;
+	case ELF_GR_OFFSET(4) ... ELF_GR_OFFSET(7):
+		if (write_access) {
+			/* read NaT bit first: */
+			unsigned long dummy;
+
+			ret = unw_get_gr(info, addr/8, &dummy, &nat);
+			if (ret < 0)
+				return ret;
+		}
+		return unw_access_gr(info, addr/8, data, &nat, write_access);
+	case ELF_GR_OFFSET(8) ... ELF_GR_OFFSET(11):
+		ptr = (void *)&pt->r8 + addr - ELF_GR_OFFSET(8);
+		break;
+	case ELF_GR_OFFSET(12):
+	case ELF_GR_OFFSET(13):
+		ptr = (void *)&pt->r12 + addr - ELF_GR_OFFSET(12);
+		break;
+	case ELF_GR_OFFSET(14):
+		ptr = &pt->r14;
+		break;
+	case ELF_GR_OFFSET(15):
+		ptr = &pt->r15;
+	}
+	if (write_access)
+		*ptr = *data;
+	else
+		*data = *ptr;
+	return 0;
+}
+
+static int
+access_elf_breg(struct task_struct *target, struct unw_frame_info *info,
+		unsigned long addr, unsigned long *data, int write_access)
+{
+	struct pt_regs *pt;
+	unsigned long *ptr = NULL;
+
+	pt = task_pt_regs(target);
+	switch (addr) {
+	case ELF_BR_OFFSET(0):
+		ptr = &pt->b0;
+		break;
+	case ELF_BR_OFFSET(1) ... ELF_BR_OFFSET(5):
+		return unw_access_br(info, (addr - ELF_BR_OFFSET(0))/8,
+				     data, write_access);
+	case ELF_BR_OFFSET(6):
+		ptr = &pt->b6;
+		break;
+	case ELF_BR_OFFSET(7):
+		ptr = &pt->b7;
+	}
+	if (write_access)
+		*ptr = *data;
+	else
+		*data = *ptr;
+	return 0;
+}
+
+static int
+access_elf_areg(struct task_struct *target, struct unw_frame_info *info,
+		unsigned long addr, unsigned long *data, int write_access)
+{
+	struct pt_regs *pt;
+	unsigned long cfm, urbs_end;
+	unsigned long *ptr = NULL;
+
+	pt = task_pt_regs(target);
+	if (addr >= ELF_AR_RSC_OFFSET && addr <= ELF_AR_SSD_OFFSET) {
+		switch (addr) {
+		case ELF_AR_RSC_OFFSET:
+			/* force PL3 */
+			if (write_access)
+				pt->ar_rsc = *data | (3 << 2);
+			else
+				*data = pt->ar_rsc;
+			return 0;
+		case ELF_AR_BSP_OFFSET:
+			/*
+			 * By convention, we use PT_AR_BSP to refer to
+			 * the end of the user-level backing store.
+			 * Use ia64_rse_skip_regs(PT_AR_BSP, -CFM.sof)
+			 * to get the real value of ar.bsp at the time
+			 * the kernel was entered.
+			 *
+			 * Furthermore, when changing the contents of
+			 * PT_AR_BSP (or PT_CFM) while the task is
+			 * blocked in a system call, convert the state
+			 * so that the non-system-call exit
+			 * path is used.  This ensures that the proper
+			 * state will be picked up when resuming
+			 * execution.  However, it *also* means that
+			 * once we write PT_AR_BSP/PT_CFM, it won't be
+			 * possible to modify the syscall arguments of
+			 * the pending system call any longer.  This
+			 * shouldn't be an issue because modifying
+			 * PT_AR_BSP/PT_CFM generally implies that
+			 * we're either abandoning the pending system
+			 * call or that we defer it's re-execution
+			 * (e.g., due to GDB doing an inferior
+			 * function call).
+			 */
+			urbs_end = ia64_get_user_rbs_end(target, pt, &cfm);
+			if (write_access) {
+				if (*data != urbs_end) {
+					if (in_syscall(pt))
+						convert_to_non_syscall(target,
+								       pt,
+								       cfm);
+					/*
+					 * Simulate user-level write
+					 * of ar.bsp:
+					 */
+					pt->loadrs = 0;
+					pt->ar_bspstore = *data;
+				}
+			} else
+				*data = urbs_end;
+			return 0;
+		case ELF_AR_BSPSTORE_OFFSET:
+			ptr = &pt->ar_bspstore;
+			break;
+		case ELF_AR_RNAT_OFFSET:
+			ptr = &pt->ar_rnat;
+			break;
+		case ELF_AR_CCV_OFFSET:
+			ptr = &pt->ar_ccv;
+			break;
+		case ELF_AR_UNAT_OFFSET:
+			ptr = &pt->ar_unat;
+			break;
+		case ELF_AR_FPSR_OFFSET:
+			ptr = &pt->ar_fpsr;
+			break;
+		case ELF_AR_PFS_OFFSET:
+			ptr = &pt->ar_pfs;
+			break;
+		case ELF_AR_LC_OFFSET:
+			return unw_access_ar(info, UNW_AR_LC, data,
+					     write_access);
+		case ELF_AR_EC_OFFSET:
+			return unw_access_ar(info, UNW_AR_EC, data,
+					     write_access);
+		case ELF_AR_CSD_OFFSET:
+			ptr = &pt->ar_csd;
+			break;
+		case ELF_AR_SSD_OFFSET:
+			ptr = &pt->ar_ssd;
+		}
+	} else if (addr >= ELF_CR_IIP_OFFSET && addr <= ELF_CR_IPSR_OFFSET) {
+		switch (addr) {
+		case ELF_CR_IIP_OFFSET:
+			ptr = &pt->cr_iip;
+			break;
+		case ELF_CFM_OFFSET:
+			urbs_end = ia64_get_user_rbs_end(target, pt, &cfm);
+			if (write_access) {
+				if (((cfm ^ *data) & PFM_MASK) != 0) {
+					if (in_syscall(pt))
+						convert_to_non_syscall(target,
+								       pt,
+								       cfm);
+					pt->cr_ifs = ((pt->cr_ifs & ~PFM_MASK)
+						      | (*data & PFM_MASK));
+				}
+			} else
+				*data = cfm;
+			return 0;
+		case ELF_CR_IPSR_OFFSET:
+			if (write_access) {
+				unsigned long tmp = *data;
+				/* psr.ri==3 is a reserved value: SDM 2:25 */
+				if ((tmp & IA64_PSR_RI) == IA64_PSR_RI)
+					tmp &= ~IA64_PSR_RI;
+				pt->cr_ipsr = ((tmp & IPSR_MASK)
+					       | (pt->cr_ipsr & ~IPSR_MASK));
+			} else
+				*data = (pt->cr_ipsr & IPSR_MASK);
+			return 0;
+		}
+	} else if (addr == ELF_NAT_OFFSET)
+		return access_nat_bits(target, pt, info,
+				       data, write_access);
+	else if (addr == ELF_PR_OFFSET)
+		ptr = &pt->pr;
+	else
+		return -1;
+
+	if (write_access)
+		*ptr = *data;
+	else
+		*data = *ptr;
+
+	return 0;
+}
+
+static int
+access_elf_reg(struct task_struct *target, struct unw_frame_info *info,
+		unsigned long addr, unsigned long *data, int write_access)
+{
+	if (addr >= ELF_GR_OFFSET(1) && addr <= ELF_GR_OFFSET(15))
+		return access_elf_gpreg(target, info, addr, data, write_access);
+	else if (addr >= ELF_BR_OFFSET(0) && addr <= ELF_BR_OFFSET(7))
+		return access_elf_breg(target, info, addr, data, write_access);
+	else
+		return access_elf_areg(target, info, addr, data, write_access);
+}
+
+void do_gpregs_get(struct unw_frame_info *info, void *arg)
+{
+	struct pt_regs *pt;
+	struct regset_getset *dst = arg;
+	elf_greg_t tmp[16];
+	unsigned int i, index, min_copy;
+
+	if (unw_unwind_to_user(info) < 0)
+		return;
+
+	/*
+	 * coredump format:
+	 *      r0-r31
+	 *      NaT bits (for r0-r31; bit N == 1 iff rN is a NaT)
+	 *      predicate registers (p0-p63)
+	 *      b0-b7
+	 *      ip cfm user-mask
+	 *      ar.rsc ar.bsp ar.bspstore ar.rnat
+	 *      ar.ccv ar.unat ar.fpsr ar.pfs ar.lc ar.ec
+	 */
+
+
+	/* Skip r0 */
+	if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(1)) {
+		dst->ret = user_regset_copyout_zero(&dst->pos, &dst->count,
+						      &dst->u.get.kbuf,
+						      &dst->u.get.ubuf,
+						      0, ELF_GR_OFFSET(1));
+		if (dst->ret || dst->count == 0)
+			return;
+	}
+
+	/* gr1 - gr15 */
+	if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(16)) {
+		index = (dst->pos - ELF_GR_OFFSET(1)) / sizeof(elf_greg_t);
+		min_copy = ELF_GR_OFFSET(16) > (dst->pos + dst->count) ?
+			 (dst->pos + dst->count) : ELF_GR_OFFSET(16);
+		for (i = dst->pos; i < min_copy; i += sizeof(elf_greg_t),
+				index++)
+			if (access_elf_reg(dst->target, info, i,
+						&tmp[index], 0) < 0) {
+				dst->ret = -EIO;
+				return;
+			}
+		dst->ret = user_regset_copyout(&dst->pos, &dst->count,
+				&dst->u.get.kbuf, &dst->u.get.ubuf, tmp,
+				ELF_GR_OFFSET(1), ELF_GR_OFFSET(16));
+		if (dst->ret || dst->count == 0)
+			return;
+	}
+
+	/* r16-r31 */
+	if (dst->count > 0 && dst->pos < ELF_NAT_OFFSET) {
+		pt = task_pt_regs(dst->target);
+		dst->ret = user_regset_copyout(&dst->pos, &dst->count,
+				&dst->u.get.kbuf, &dst->u.get.ubuf, &pt->r16,
+				ELF_GR_OFFSET(16), ELF_NAT_OFFSET);
+		if (dst->ret || dst->count == 0)
+			return;
+	}
+
+	/* nat, pr, b0 - b7 */
+	if (dst->count > 0 && dst->pos < ELF_CR_IIP_OFFSET) {
+		index = (dst->pos - ELF_NAT_OFFSET) / sizeof(elf_greg_t);
+		min_copy = ELF_CR_IIP_OFFSET > (dst->pos + dst->count) ?
+			 (dst->pos + dst->count) : ELF_CR_IIP_OFFSET;
+		for (i = dst->pos; i < min_copy; i += sizeof(elf_greg_t),
+				index++)
+			if (access_elf_reg(dst->target, info, i,
+						&tmp[index], 0) < 0) {
+				dst->ret = -EIO;
+				return;
+			}
+		dst->ret = user_regset_copyout(&dst->pos, &dst->count,
+				&dst->u.get.kbuf, &dst->u.get.ubuf, tmp,
+				ELF_NAT_OFFSET, ELF_CR_IIP_OFFSET);
+		if (dst->ret || dst->count == 0)
+			return;
+	}
+
+	/* ip cfm psr ar.rsc ar.bsp ar.bspstore ar.rnat
+	 * ar.ccv ar.unat ar.fpsr ar.pfs ar.lc ar.ec ar.csd ar.ssd
+	 */
+	if (dst->count > 0 && dst->pos < (ELF_AR_END_OFFSET)) {
+		index = (dst->pos - ELF_CR_IIP_OFFSET) / sizeof(elf_greg_t);
+		min_copy = ELF_AR_END_OFFSET > (dst->pos + dst->count) ?
+			 (dst->pos + dst->count) : ELF_AR_END_OFFSET;
+		for (i = dst->pos; i < min_copy; i += sizeof(elf_greg_t),
+				index++)
+			if (access_elf_reg(dst->target, info, i,
+						&tmp[index], 0) < 0) {
+				dst->ret = -EIO;
+				return;
+			}
+		dst->ret = user_regset_copyout(&dst->pos, &dst->count,
+				&dst->u.get.kbuf, &dst->u.get.ubuf, tmp,
+				ELF_CR_IIP_OFFSET, ELF_AR_END_OFFSET);
+	}
+}
+
+void do_gpregs_set(struct unw_frame_info *info, void *arg)
+{
+	struct pt_regs *pt;
+	struct regset_getset *dst = arg;
+	elf_greg_t tmp[16];
+	unsigned int i, index;
+
+	if (unw_unwind_to_user(info) < 0)
+		return;
+
+	/* Skip r0 */
+	if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(1)) {
+		dst->ret = user_regset_copyin_ignore(&dst->pos, &dst->count,
+						       &dst->u.set.kbuf,
+						       &dst->u.set.ubuf,
+						       0, ELF_GR_OFFSET(1));
+		if (dst->ret || dst->count == 0)
+			return;
+	}
+
+	/* gr1-gr15 */
+	if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(16)) {
+		i = dst->pos;
+		index = (dst->pos - ELF_GR_OFFSET(1)) / sizeof(elf_greg_t);
+		dst->ret = user_regset_copyin(&dst->pos, &dst->count,
+				&dst->u.set.kbuf, &dst->u.set.ubuf, tmp,
+				ELF_GR_OFFSET(1), ELF_GR_OFFSET(16));
+		if (dst->ret)
+			return;
+		for ( ; i < dst->pos; i += sizeof(elf_greg_t), index++)
+			if (access_elf_reg(dst->target, info, i,
+						&tmp[index], 1) < 0) {
+				dst->ret = -EIO;
+				return;
+			}
+		if (dst->count == 0)
+			return;
+	}
+
+	/* gr16-gr31 */
+	if (dst->count > 0 && dst->pos < ELF_NAT_OFFSET) {
+		pt = task_pt_regs(dst->target);
+		dst->ret = user_regset_copyin(&dst->pos, &dst->count,
+				&dst->u.set.kbuf, &dst->u.set.ubuf, &pt->r16,
+				ELF_GR_OFFSET(16), ELF_NAT_OFFSET);
+		if (dst->ret || dst->count == 0)
+			return;
+	}
+
+	/* nat, pr, b0 - b7 */
+	if (dst->count > 0 && dst->pos < ELF_CR_IIP_OFFSET) {
+		i = dst->pos;
+		index = (dst->pos - ELF_NAT_OFFSET) / sizeof(elf_greg_t);
+		dst->ret = user_regset_copyin(&dst->pos, &dst->count,
+				&dst->u.set.kbuf, &dst->u.set.ubuf, tmp,
+				ELF_NAT_OFFSET, ELF_CR_IIP_OFFSET);
+		if (dst->ret)
+			return;
+		for (; i < dst->pos; i += sizeof(elf_greg_t), index++)
+			if (access_elf_reg(dst->target, info, i,
+						&tmp[index], 1) < 0) {
+				dst->ret = -EIO;
+				return;
+			}
+		if (dst->count == 0)
+			return;
+	}
+
+	/* ip cfm psr ar.rsc ar.bsp ar.bspstore ar.rnat
+	 * ar.ccv ar.unat ar.fpsr ar.pfs ar.lc ar.ec ar.csd ar.ssd
+	 */
+	if (dst->count > 0 && dst->pos < (ELF_AR_END_OFFSET)) {
+		i = dst->pos;
+		index = (dst->pos - ELF_CR_IIP_OFFSET) / sizeof(elf_greg_t);
+		dst->ret = user_regset_copyin(&dst->pos, &dst->count,
+				&dst->u.set.kbuf, &dst->u.set.ubuf, tmp,
+				ELF_CR_IIP_OFFSET, ELF_AR_END_OFFSET);
+		if (dst->ret)
+			return;
+		for ( ; i < dst->pos; i += sizeof(elf_greg_t), index++)
+			if (access_elf_reg(dst->target, info, i,
+						&tmp[index], 1) < 0) {
+				dst->ret = -EIO;
+				return;
+			}
+	}
+}
+
+#define ELF_FP_OFFSET(i)	(i * sizeof(elf_fpreg_t))
+
+void do_fpregs_get(struct unw_frame_info *info, void *arg)
+{
+	struct regset_getset *dst = arg;
+	struct task_struct *task = dst->target;
+	elf_fpreg_t tmp[30];
+	int index, min_copy, i;
+
+	if (unw_unwind_to_user(info) < 0)
+		return;
+
+	/* Skip pos 0 and 1 */
+	if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(2)) {
+		dst->ret = user_regset_copyout_zero(&dst->pos, &dst->count,
+						      &dst->u.get.kbuf,
+						      &dst->u.get.ubuf,
+						      0, ELF_FP_OFFSET(2));
+		if (dst->count == 0 || dst->ret)
+			return;
+	}
+
+	/* fr2-fr31 */
+	if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(32)) {
+		index = (dst->pos - ELF_FP_OFFSET(2)) / sizeof(elf_fpreg_t);
+
+		min_copy = min(((unsigned int)ELF_FP_OFFSET(32)),
+				dst->pos + dst->count);
+		for (i = dst->pos; i < min_copy; i += sizeof(elf_fpreg_t),
+				index++)
+			if (unw_get_fr(info, i / sizeof(elf_fpreg_t),
+					 &tmp[index])) {
+				dst->ret = -EIO;
+				return;
+			}
+		dst->ret = user_regset_copyout(&dst->pos, &dst->count,
+				&dst->u.get.kbuf, &dst->u.get.ubuf, tmp,
+				ELF_FP_OFFSET(2), ELF_FP_OFFSET(32));
+		if (dst->count == 0 || dst->ret)
+			return;
+	}
+
+	/* fph */
+	if (dst->count > 0) {
+		ia64_flush_fph(dst->target);
+		if (task->thread.flags & IA64_THREAD_FPH_VALID)
+			dst->ret = user_regset_copyout(
+				&dst->pos, &dst->count,
+				&dst->u.get.kbuf, &dst->u.get.ubuf,
+				&dst->target->thread.fph,
+				ELF_FP_OFFSET(32), -1);
+		else
+			/* Zero fill instead.  */
+			dst->ret = user_regset_copyout_zero(
+				&dst->pos, &dst->count,
+				&dst->u.get.kbuf, &dst->u.get.ubuf,
+				ELF_FP_OFFSET(32), -1);
+	}
+}
+
+void do_fpregs_set(struct unw_frame_info *info, void *arg)
+{
+	struct regset_getset *dst = arg;
+	elf_fpreg_t fpreg, tmp[30];
+	int index, start, end;
+
+	if (unw_unwind_to_user(info) < 0)
+		return;
+
+	/* Skip pos 0 and 1 */
+	if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(2)) {
+		dst->ret = user_regset_copyin_ignore(&dst->pos, &dst->count,
+						       &dst->u.set.kbuf,
+						       &dst->u.set.ubuf,
+						       0, ELF_FP_OFFSET(2));
+		if (dst->count == 0 || dst->ret)
+			return;
+	}
+
+	/* fr2-fr31 */
+	if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(32)) {
+		start = dst->pos;
+		end = min(((unsigned int)ELF_FP_OFFSET(32)),
+			 dst->pos + dst->count);
+		dst->ret = user_regset_copyin(&dst->pos, &dst->count,
+				&dst->u.set.kbuf, &dst->u.set.ubuf, tmp,
+				ELF_FP_OFFSET(2), ELF_FP_OFFSET(32));
+		if (dst->ret)
+			return;
+
+		if (start & 0xF) { /* only write high part */
+			if (unw_get_fr(info, start / sizeof(elf_fpreg_t),
+					 &fpreg)) {
+				dst->ret = -EIO;
+				return;
+			}
+			tmp[start / sizeof(elf_fpreg_t) - 2].u.bits[0]
+				= fpreg.u.bits[0];
+			start &= ~0xFUL;
+		}
+		if (end & 0xF) { /* only write low part */
+			if (unw_get_fr(info, end / sizeof(elf_fpreg_t),
+					&fpreg)) {
+				dst->ret = -EIO;
+				return;
+			}
+			tmp[end / sizeof(elf_fpreg_t) - 2].u.bits[1]
+				= fpreg.u.bits[1];
+			end = (end + 0xF) & ~0xFUL;
+		}
+
+		for ( ;	start < end ; start += sizeof(elf_fpreg_t)) {
+			index = start / sizeof(elf_fpreg_t);
+			if (unw_set_fr(info, index, tmp[index - 2])) {
+				dst->ret = -EIO;
+				return;
+			}
+		}
+		if (dst->ret || dst->count == 0)
+			return;
+	}
+
+	/* fph */
+	if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(128)) {
+		ia64_sync_fph(dst->target);
+		dst->ret = user_regset_copyin(&dst->pos, &dst->count,
+						&dst->u.set.kbuf,
+						&dst->u.set.ubuf,
+						&dst->target->thread.fph,
+						ELF_FP_OFFSET(32), -1);
+	}
+}
+
+static int
+do_regset_call(void (*call)(struct unw_frame_info *, void *),
+	       struct task_struct *target,
+	       const struct user_regset *regset,
+	       unsigned int pos, unsigned int count,
+	       const void *kbuf, const void __user *ubuf)
+{
+	struct regset_getset info = { .target = target, .regset = regset,
+				 .pos = pos, .count = count,
+				 .u.set = { .kbuf = kbuf, .ubuf = ubuf },
+				 .ret = 0 };
+
+	if (target == current)
+		unw_init_running(call, &info);
+	else {
+		struct unw_frame_info ufi;
+		memset(&ufi, 0, sizeof(ufi));
+		unw_init_from_blocked_task(&ufi, target);
+		(*call)(&ufi, &info);
+	}
+
+	return info.ret;
+}
+
+static int
+gpregs_get(struct task_struct *target,
+	   const struct user_regset *regset,
+	   unsigned int pos, unsigned int count,
+	   void *kbuf, void __user *ubuf)
+{
+	return do_regset_call(do_gpregs_get, target, regset, pos, count,
+		kbuf, ubuf);
+}
+
+static int gpregs_set(struct task_struct *target,
+		const struct user_regset *regset,
+		unsigned int pos, unsigned int count,
+		const void *kbuf, const void __user *ubuf)
+{
+	return do_regset_call(do_gpregs_set, target, regset, pos, count,
+		kbuf, ubuf);
+}
+
+static void do_gpregs_writeback(struct unw_frame_info *info, void *arg)
+{
+	do_sync_rbs(info, ia64_sync_user_rbs);
+}
+
+/*
+ * This is called to write back the register backing store.
+ * ptrace does this before it stops, so that a tracer reading the user
+ * memory after the thread stops will get the current register data.
+ */
+static int
+gpregs_writeback(struct task_struct *target,
+		 const struct user_regset *regset,
+		 int now)
+{
+	if (test_and_set_tsk_thread_flag(target, TIF_RESTORE_RSE))
+		return 0;
+	tsk_set_notify_resume(target);
+	return do_regset_call(do_gpregs_writeback, target, regset, 0, 0,
+		NULL, NULL);
+}
+
+static int
+fpregs_active(struct task_struct *target, const struct user_regset *regset)
+{
+	return (target->thread.flags & IA64_THREAD_FPH_VALID) ? 128 : 32;
+}
+
+static int fpregs_get(struct task_struct *target,
+		const struct user_regset *regset,
+		unsigned int pos, unsigned int count,
+		void *kbuf, void __user *ubuf)
+{
+	return do_regset_call(do_fpregs_get, target, regset, pos, count,
+		kbuf, ubuf);
+}
+
+static int fpregs_set(struct task_struct *target,
+		const struct user_regset *regset,
+		unsigned int pos, unsigned int count,
+		const void *kbuf, const void __user *ubuf)
+{
+	return do_regset_call(do_fpregs_set, target, regset, pos, count,
+		kbuf, ubuf);
+}
+
+static const struct user_regset native_regsets[] = {
+	{
+		.core_note_type = NT_PRSTATUS,
+		.n = ELF_NGREG,
+		.size = sizeof(elf_greg_t), .align = sizeof(elf_greg_t),
+		.get = gpregs_get, .set = gpregs_set,
+		.writeback = gpregs_writeback
+	},
+	{
+		.core_note_type = NT_PRFPREG,
+		.n = ELF_NFPREG,
+		.size = sizeof(elf_fpreg_t), .align = sizeof(elf_fpreg_t),
+		.get = fpregs_get, .set = fpregs_set, .active = fpregs_active
+	},
+};
+
+static const struct user_regset_view user_ia64_view = {
+	.name = "ia64",
+	.e_machine = EM_IA_64,
+	.regsets = native_regsets, .n = ARRAY_SIZE(native_regsets)
+};
+
+const struct user_regset_view *task_user_regset_view(struct task_struct *tsk)
+{
+	return &user_ia64_view;
+}
diff --git a/include/asm-ia64/elf.h b/include/asm-ia64/elf.h
index f8e83eca67a2..064cf7dcea8e 100644
--- a/include/asm-ia64/elf.h
+++ b/include/asm-ia64/elf.h
@@ -154,6 +154,30 @@ extern void ia64_init_addr_space (void);
 #define ELF_NGREG	128	/* we really need just 72 but let's leave some headroom... */
 #define ELF_NFPREG	128	/* f0 and f1 could be omitted, but so what... */
 
+/* elf_gregset_t register offsets */
+#define ELF_GR_0_OFFSET     0
+#define ELF_NAT_OFFSET     (32 * sizeof(elf_greg_t))
+#define ELF_PR_OFFSET      (33 * sizeof(elf_greg_t))
+#define ELF_BR_0_OFFSET    (34 * sizeof(elf_greg_t))
+#define ELF_CR_IIP_OFFSET  (42 * sizeof(elf_greg_t))
+#define ELF_CFM_OFFSET     (43 * sizeof(elf_greg_t))
+#define ELF_CR_IPSR_OFFSET (44 * sizeof(elf_greg_t))
+#define ELF_GR_OFFSET(i)   (ELF_GR_0_OFFSET + i * sizeof(elf_greg_t))
+#define ELF_BR_OFFSET(i)   (ELF_BR_0_OFFSET + i * sizeof(elf_greg_t))
+#define ELF_AR_RSC_OFFSET  (45 * sizeof(elf_greg_t))
+#define ELF_AR_BSP_OFFSET  (46 * sizeof(elf_greg_t))
+#define ELF_AR_BSPSTORE_OFFSET (47 * sizeof(elf_greg_t))
+#define ELF_AR_RNAT_OFFSET (48 * sizeof(elf_greg_t))
+#define ELF_AR_CCV_OFFSET  (49 * sizeof(elf_greg_t))
+#define ELF_AR_UNAT_OFFSET (50 * sizeof(elf_greg_t))
+#define ELF_AR_FPSR_OFFSET (51 * sizeof(elf_greg_t))
+#define ELF_AR_PFS_OFFSET  (52 * sizeof(elf_greg_t))
+#define ELF_AR_LC_OFFSET   (53 * sizeof(elf_greg_t))
+#define ELF_AR_EC_OFFSET   (54 * sizeof(elf_greg_t))
+#define ELF_AR_CSD_OFFSET  (55 * sizeof(elf_greg_t))
+#define ELF_AR_SSD_OFFSET  (56 * sizeof(elf_greg_t))
+#define ELF_AR_END_OFFSET  (57 * sizeof(elf_greg_t))
+
 typedef unsigned long elf_fpxregset_t;
 
 typedef unsigned long elf_greg_t;
-- 
cgit v1.2.3


From 6cb53d7a6f40858181facde0f52587731d2e621f Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Thu, 28 Feb 2008 16:09:38 +0800
Subject: [IA64] use CORE_DUMP_USE_REGSET

After we have regset support, we can use CORE_DUMP_USE_REGSET.

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/process.c | 30 ------------------------------
 include/asm-ia64/elf.h     |  7 +------
 2 files changed, 1 insertion(+), 36 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index 49937a383b23..a5ea817cbcbf 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -625,42 +625,12 @@ do_dump_fpu (struct unw_frame_info *info, void *arg)
 	do_dump_task_fpu(current, info, arg);
 }
 
-int
-dump_task_regs(struct task_struct *task, elf_gregset_t *regs)
-{
-	struct unw_frame_info tcore_info;
-
-	if (current == task) {
-		unw_init_running(do_copy_regs, regs);
-	} else {
-		memset(&tcore_info, 0, sizeof(tcore_info));
-		unw_init_from_blocked_task(&tcore_info, task);
-		do_copy_task_regs(task, &tcore_info, regs);
-	}
-	return 1;
-}
-
 void
 ia64_elf_core_copy_regs (struct pt_regs *pt, elf_gregset_t dst)
 {
 	unw_init_running(do_copy_regs, dst);
 }
 
-int
-dump_task_fpu (struct task_struct *task, elf_fpregset_t *dst)
-{
-	struct unw_frame_info tcore_info;
-
-	if (current == task) {
-		unw_init_running(do_dump_fpu, dst);
-	} else {
-		memset(&tcore_info, 0, sizeof(tcore_info));
-		unw_init_from_blocked_task(&tcore_info, task);
-		do_dump_task_fpu(task, &tcore_info, dst);
-	}
-	return 1;
-}
-
 int
 dump_fpu (struct pt_regs *pt, elf_fpregset_t dst)
 {
diff --git a/include/asm-ia64/elf.h b/include/asm-ia64/elf.h
index 064cf7dcea8e..5e0c1a6bce8d 100644
--- a/include/asm-ia64/elf.h
+++ b/include/asm-ia64/elf.h
@@ -26,6 +26,7 @@
 #define ELF_ARCH	EM_IA_64
 
 #define USE_ELF_CORE_DUMP
+#define CORE_DUMP_USE_REGSET
 
 /* Least-significant four bits of ELF header's e_flags are OS-specific.  The bits are
    interpreted as follows by Linux: */
@@ -207,12 +208,6 @@ extern void ia64_elf_core_copy_regs (struct pt_regs *src, elf_gregset_t dst);
 
 struct task_struct;
 
-extern int dump_task_regs(struct task_struct *, elf_gregset_t *);
-extern int dump_task_fpu (struct task_struct *, elf_fpregset_t *);
-
-#define ELF_CORE_COPY_TASK_REGS(tsk, elf_gregs) dump_task_regs(tsk, elf_gregs)
-#define ELF_CORE_COPY_FPREGS(tsk, elf_fpregs) dump_task_fpu(tsk, elf_fpregs)
-
 #define GATE_EHDR	((const struct elfhdr *) GATE_ADDR)
 
 /* update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT entries changes */
-- 
cgit v1.2.3


From 7524d7d6de5d5d3f081de8cf5479819fad339661 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 4 Mar 2008 15:26:14 -0800
Subject: the scheduled ieee80211 softmac removal

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 Documentation/feature-removal-schedule.txt      |   8 -
 MAINTAINERS                                     |   6 -
 include/net/ieee80211softmac.h                  | 373 ----------------
 include/net/ieee80211softmac_wx.h               |  99 -----
 net/ieee80211/Kconfig                           |   1 -
 net/ieee80211/Makefile                          |   1 -
 net/ieee80211/softmac/Kconfig                   |  12 -
 net/ieee80211/softmac/Makefile                  |   9 -
 net/ieee80211/softmac/ieee80211softmac_assoc.c  | 489 --------------------
 net/ieee80211/softmac/ieee80211softmac_auth.c   | 413 -----------------
 net/ieee80211/softmac/ieee80211softmac_event.c  | 189 --------
 net/ieee80211/softmac/ieee80211softmac_io.c     | 488 --------------------
 net/ieee80211/softmac/ieee80211softmac_module.c | 568 ------------------------
 net/ieee80211/softmac/ieee80211softmac_priv.h   | 244 ----------
 net/ieee80211/softmac/ieee80211softmac_scan.c   | 254 -----------
 net/ieee80211/softmac/ieee80211softmac_wx.c     | 508 ---------------------
 16 files changed, 3662 deletions(-)
 delete mode 100644 include/net/ieee80211softmac.h
 delete mode 100644 include/net/ieee80211softmac_wx.h
 delete mode 100644 net/ieee80211/softmac/Kconfig
 delete mode 100644 net/ieee80211/softmac/Makefile
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_assoc.c
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_auth.c
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_event.c
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_io.c
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_module.c
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_priv.h
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_scan.c
 delete mode 100644 net/ieee80211/softmac/ieee80211softmac_wx.c

(limited to 'include')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index dade1d14f5b2..2cf4d7ad7b0b 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -230,14 +230,6 @@ Who:	Jean Delvare <khali@linux-fr.org>
 
 ---------------------------
 
-What:	ieee80211 softmac wireless networking component
-When:	2.6.26 (or after removal of bcm43xx and port of zd1211rw to mac80211)
-Files:	net/ieee80211/softmac
-Why:	No in-kernel drivers will depend on it any longer.
-Who:	John W. Linville <linville@tuxdriver.com>
-
----------------------------
-
 What:	rc80211-simple rate control algorithm for mac80211
 When:	2.6.26
 Files:	net/mac80211/rc80211-simple.c
diff --git a/MAINTAINERS b/MAINTAINERS
index 38b767611bbf..b9fc1f620f29 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3548,12 +3548,6 @@ M:	mhoffman@lightlink.com
 L:	lm-sensors@lm-sensors.org
 S:	Maintained
 
-SOFTMAC LAYER (IEEE 802.11)
-P:	Daniel Drake
-M:	dsd@gentoo.org
-L:	linux-wireless@vger.kernel.org
-S:	Obsolete
-
 SOFTWARE RAID (Multiple Disks) SUPPORT
 P:	Ingo Molnar
 M:	mingo@redhat.com
diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h
deleted file mode 100644
index 1ef6282fdded..000000000000
--- a/include/net/ieee80211softmac.h
+++ /dev/null
@@ -1,373 +0,0 @@
-/*
- * ieee80211softmac.h - public interface to the softmac
- *
- * Copyright (c) 2005 Johannes Berg <johannes@sipsolutions.net>
- *                    Joseph Jezak <josejx@gentoo.org>
- *                    Larry Finger <Larry.Finger@lwfinger.net>
- *                    Danny van Dyk <kugelfang@gentoo.org>
- *                    Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#ifndef IEEE80211SOFTMAC_H_
-#define IEEE80211SOFTMAC_H_
-
-#include <linux/kernel.h>
-#include <linux/spinlock.h>
-#include <linux/workqueue.h>
-#include <linux/list.h>
-#include <net/ieee80211.h>
-
-/* Once the API is considered more or less stable,
- * this should be incremented on API incompatible changes.
- */
-#define IEEE80211SOFTMAC_API	0
-
-#define IEEE80211SOFTMAC_MAX_RATES_LEN		8
-#define IEEE80211SOFTMAC_MAX_EX_RATES_LEN	255
-
-struct ieee80211softmac_ratesinfo {
-	u8 count;
-	u8 rates[IEEE80211SOFTMAC_MAX_RATES_LEN + IEEE80211SOFTMAC_MAX_EX_RATES_LEN];
-};
-
-/* internal structures */
-struct ieee80211softmac_network;
-struct ieee80211softmac_scaninfo;
-
-struct ieee80211softmac_essid {
-	u8 len;
-	char data[IW_ESSID_MAX_SIZE+1];
-};
-
-struct ieee80211softmac_wpa {
-	char *IE;
-	int IElen;
-	int IEbuflen;
-};
-
-/*
- * Information about association
- */
-struct ieee80211softmac_assoc_info {
-
-	struct mutex mutex;
-
-	/*
-	 * This is the requested ESSID. It is written
-	 * only by the WX handlers.
-	 *
-	 */
-	struct ieee80211softmac_essid req_essid;
-	/*
-	 * the ESSID of the network we're currently
-	 * associated (or trying) to. This is
-	 * updated to the network's actual ESSID
-	 * even if the requested ESSID was 'ANY'
-	 */
-	struct ieee80211softmac_essid associate_essid;
-	
-	/* BSSID we're trying to associate to */
-	char bssid[ETH_ALEN];
-	
-	/* some flags.
-	 * static_essid is valid if the essid is constant,
-	 * this is for use by the wx handlers only.
-	 *
-	 * associating is true, if the network has been
-	 * auth'ed on and we are in the process of associating.
-	 *
-	 * bssvalid is true if we found a matching network
-	 * and saved it's BSSID into the bssid above.
-	 *
-	 * bssfixed is used for SIOCSIWAP.
-	 */
-	u8 static_essid;
-	u8 short_preamble_available;
-	u8 associating;
-	u8 associated;
-	u8 assoc_wait;
-	u8 bssvalid;
-	u8 bssfixed;
-
-	/* Scan retries remaining */
-	int scan_retry;
-
-	struct delayed_work work;
-	struct delayed_work timeout;
-};
-
-struct ieee80211softmac_bss_info {
-	/* Rates supported by the network */
-	struct ieee80211softmac_ratesinfo supported_rates;
-
-	/* This indicates whether frames can currently be transmitted with
-	 * short preamble (only use this variable during TX at CCK rates) */
-	u8 short_preamble:1;
-
-	/* This indicates whether protection (e.g. self-CTS) should be used
-	 * when transmitting with OFDM modulation */
-	u8 use_protection:1;
-};
-
-enum {
-	IEEE80211SOFTMAC_AUTH_OPEN_REQUEST	= 1,
-	IEEE80211SOFTMAC_AUTH_OPEN_RESPONSE	= 2,
-};
-
-enum {
-	IEEE80211SOFTMAC_AUTH_SHARED_REQUEST	= 1,
-	IEEE80211SOFTMAC_AUTH_SHARED_CHALLENGE = 2,
-	IEEE80211SOFTMAC_AUTH_SHARED_RESPONSE	= 3,
-	IEEE80211SOFTMAC_AUTH_SHARED_PASS	= 4,
-};
-
-/* We should make these tunable
- * AUTH_TIMEOUT seems really long, but that's what it is in BSD */
-#define IEEE80211SOFTMAC_AUTH_TIMEOUT		(12 * HZ)
-#define IEEE80211SOFTMAC_AUTH_RETRY_LIMIT	5
-#define IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT	3
-
-struct ieee80211softmac_txrates {
-	/* The Bit-Rate to be used for multicast frames. */
-	u8 mcast_rate;
-
-	/* The Bit-Rate to be used for multicast management frames. */
-	u8 mgt_mcast_rate;
-
-	/* The Bit-Rate to be used for any other (normal) data packet. */
-	u8 default_rate;
-	/* The Bit-Rate to be used for default fallback
-	 * (If the device supports fallback and hardware-retry)
-	 */
-	u8 default_fallback;
-
-	/* This is the rate that the user asked for */
-	u8 user_rate;
-};
-
-/* Bits for txrates_change callback. */
-#define IEEE80211SOFTMAC_TXRATECHG_DEFAULT		(1 << 0) /* default_rate */
-#define IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK	(1 << 1) /* default_fallback */
-#define IEEE80211SOFTMAC_TXRATECHG_MCAST		(1 << 2) /* mcast_rate */
-#define IEEE80211SOFTMAC_TXRATECHG_MGT_MCAST		(1 << 3) /* mgt_mcast_rate */
-
-#define IEEE80211SOFTMAC_BSSINFOCHG_RATES		(1 << 0) /* supported_rates */
-#define IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE	(1 << 1) /* short_preamble */
-#define IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION		(1 << 2) /* use_protection */
-
-struct ieee80211softmac_device {
-	/* 802.11 structure for data stuff */
-	struct ieee80211_device *ieee;
-	struct net_device *dev;
-
-	/* only valid if associated, then holds the Association ID */
-	u16 association_id;
-	
-	/* the following methods are callbacks that the driver
-	 * using this framework has to assign
-	 */
-
-	/* always assign these */
-	void (*set_bssid_filter)(struct net_device *dev, const u8 *bssid);
-	void (*set_channel)(struct net_device *dev, u8 channel);
-
-	/* assign if you need it, informational only */
-	void (*link_change)(struct net_device *dev);
-
-	/* If the hardware can do scanning, assign _all_ three of these callbacks.
-	 * When the scan finishes, call ieee80211softmac_scan_finished().
-	 */
-	
-	/* when called, start_scan is guaranteed to not be called again
-	 * until you call ieee80211softmac_scan_finished.
-	 * Return 0 if scanning could start, error otherwise.
-	 * SOFTMAC AUTHORS: don't call this, use ieee80211softmac_start_scan */
-	int (*start_scan)(struct net_device *dev);
-	/* this should block until after ieee80211softmac_scan_finished was called
-	 * SOFTMAC AUTHORS: don't call this, use ieee80211softmac_wait_for_scan */
-	void (*wait_for_scan)(struct net_device *dev);
-	/* stop_scan aborts a scan, but is asynchronous.
-	 * if you want to wait for it too, use wait_for_scan
-	 * SOFTMAC AUTHORS: don't call this, use ieee80211softmac_stop_scan */
-	void (*stop_scan)(struct net_device *dev);
-
-	/* we'll need something about beacons here too, for AP or ad-hoc modes */
-
-	/* Transmission rates to be used by the driver.
-	 * The SoftMAC figures out the best possible rates.
-	 * The driver just needs to read them.
-	 */
-	struct ieee80211softmac_txrates txrates;
-
-	/* If the driver needs to do stuff on TX rate changes, assign this
-	 * callback. See IEEE80211SOFTMAC_TXRATECHG for change flags. */
-	void (*txrates_change)(struct net_device *dev,
-			       u32 changes);
-
-	/* If the driver needs to do stuff when BSS properties change, assign
-	 * this callback. see IEEE80211SOFTMAC_BSSINFOCHG for change flags. */
-	void (*bssinfo_change)(struct net_device *dev,
-			       u32 changes);
-
-	/* private stuff follows */
-	/* this lock protects this structure */
-	spinlock_t lock;
-
-	struct workqueue_struct *wq;
-
-	u8 running; /* SoftMAC started? */
-	u8 scanning;
-
-	struct ieee80211softmac_scaninfo *scaninfo;
-	struct ieee80211softmac_assoc_info associnfo;
-	struct ieee80211softmac_bss_info bssinfo;
-
-	struct list_head auth_queue;
-	struct list_head events;
-
- 	struct ieee80211softmac_ratesinfo ratesinfo;
-	int txrate_badness;
-	
-	/* WPA stuff */
-	struct ieee80211softmac_wpa wpa;
-
-	/* we need to keep a list of network structs we copied */
-	struct list_head network_list;
-
-	/* This must be the last item so that it points to the data
-	 * allocated beyond this structure by alloc_ieee80211 */
-	u8 priv[0];
-};
-
-extern void ieee80211softmac_scan_finished(struct ieee80211softmac_device *sm);
-
-static inline void * ieee80211softmac_priv(struct net_device *dev)
-{
-	return ((struct ieee80211softmac_device *)ieee80211_priv(dev))->priv;
-}
-
-extern struct net_device * alloc_ieee80211softmac(int sizeof_priv);
-extern void free_ieee80211softmac(struct net_device *dev);
-
-/* Call this function if you detect a lost TX fragment.
- * (If the device indicates failure of ACK RX, for example.)
- * It is wise to call this function if you are able to detect lost packets,
- * because it contributes to the TX Rates auto adjustment.
- */
-extern void ieee80211softmac_fragment_lost(struct net_device *dev,
-					   u16 wireless_sequence_number);
-/* Call this function before _start to tell the softmac what rates
- * the hw supports. The rates parameter is copied, so you can
- * free it right after calling this function. 
- * Note that the rates need to be sorted. */
-extern void ieee80211softmac_set_rates(struct net_device *dev, u8 count, u8 *rates);
-
-/* Finds the highest rate which is:
- *  1. Present in ri (optionally a basic rate)
- *  2. Supported by the device
- *  3. Less than or equal to the user-defined rate
- */
-extern u8 ieee80211softmac_highest_supported_rate(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_ratesinfo *ri, int basic_only);
-
-/* Helper function which advises you the rate at which a frame should be
- * transmitted at. */
-static inline u8 ieee80211softmac_suggest_txrate(struct ieee80211softmac_device *mac,
-						 int is_multicast,
-						 int is_mgt)
-{
-	struct ieee80211softmac_txrates *txrates = &mac->txrates;
-
-	if (!mac->associnfo.associated)
-		return txrates->mgt_mcast_rate;
-
-	/* We are associated, sending unicast frame */
-	if (!is_multicast)
-		return txrates->default_rate;
-
-	/* We are associated, sending multicast frame */
-	if (is_mgt)
-		return txrates->mgt_mcast_rate;
-	else
-		return txrates->mcast_rate;
-}
-
-/* Helper function which advises you when it is safe to transmit with short
- * preamble.
- * You should only call this function when transmitting at CCK rates. */
-static inline int ieee80211softmac_short_preamble_ok(struct ieee80211softmac_device *mac,
-						    int is_multicast,
-						    int is_mgt)
-{
-	return (is_multicast && is_mgt) ? 0 : mac->bssinfo.short_preamble;
-}
-
-/* Helper function which advises you whether protection (e.g. self-CTS) is
- * needed. 1 = protection needed, 0 = no protection needed
- * Only use this function when transmitting with OFDM modulation. */
-static inline int ieee80211softmac_protection_needed(struct ieee80211softmac_device *mac)
-{
-	return mac->bssinfo.use_protection;
-}
-
-/* Start the SoftMAC. Call this after you initialized the device
- * and it is ready to run.
- */
-extern void ieee80211softmac_start(struct net_device *dev);
-/* Stop the SoftMAC. Call this before you shutdown the device. */
-extern void ieee80211softmac_stop(struct net_device *dev);
-
-/*
- * Event system
- */
-
-/* valid event types */
-#define IEEE80211SOFTMAC_EVENT_ANY			-1 /*private use only*/
-#define IEEE80211SOFTMAC_EVENT_SCAN_FINISHED		0
-#define IEEE80211SOFTMAC_EVENT_ASSOCIATED		1
-#define IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED		2
-#define IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT	3
-#define IEEE80211SOFTMAC_EVENT_AUTHENTICATED		4
-#define IEEE80211SOFTMAC_EVENT_AUTH_FAILED		5
-#define IEEE80211SOFTMAC_EVENT_AUTH_TIMEOUT		6
-#define IEEE80211SOFTMAC_EVENT_ASSOCIATE_NET_NOT_FOUND	7
-#define IEEE80211SOFTMAC_EVENT_DISASSOCIATED		8
-/* keep this updated! */
-#define IEEE80211SOFTMAC_EVENT_LAST			8
-/*
- * If you want to be notified of certain events, you can call
- * ieee80211softmac_notify[_atomic] with
- * 	- event set to one of the constants below
- * 	- fun set to a function pointer of the appropriate type
- *	- context set to the context data you want passed
- * The return value is 0, or an error.
- */
-typedef void (*notify_function_ptr)(struct net_device *dev, int event_type, void *context);
-
-#define ieee80211softmac_notify(dev, event, fun, context) ieee80211softmac_notify_gfp(dev, event, fun, context, GFP_KERNEL);
-#define ieee80211softmac_notify_atomic(dev, event, fun, context) ieee80211softmac_notify_gfp(dev, event, fun, context, GFP_ATOMIC);
-
-extern int ieee80211softmac_notify_gfp(struct net_device *dev,
-	int event, notify_function_ptr fun, void *context, gfp_t gfp_mask);
-
-/* To clear pending work (for ifconfig down, etc.) */
-extern void
-ieee80211softmac_clear_pending_work(struct ieee80211softmac_device *sm);
-
-#endif /* IEEE80211SOFTMAC_H_ */
diff --git a/include/net/ieee80211softmac_wx.h b/include/net/ieee80211softmac_wx.h
deleted file mode 100644
index 4ee3ad57283f..000000000000
--- a/include/net/ieee80211softmac_wx.h
+++ /dev/null
@@ -1,99 +0,0 @@
-/*
- * This file contains the prototypes for the wireless extension
- * handlers that the softmac API provides. Include this file to
- * use the wx handlers, you can assign these directly.
- *
- * Copyright (c) 2005 Johannes Berg <johannes@sipsolutions.net>
- *                    Joseph Jezak <josejx@gentoo.org>
- *                    Larry Finger <Larry.Finger@lwfinger.net>
- *                    Danny van Dyk <kugelfang@gentoo.org>
- *                    Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#ifndef _IEEE80211SOFTMAC_WX_H
-#define _IEEE80211SOFTMAC_WX_H
-
-#include <net/ieee80211softmac.h>
-#include <net/iw_handler.h>
-
-extern int
-ieee80211softmac_wx_trigger_scan(struct net_device *net_dev,
-				 struct iw_request_info *info,
-				 union iwreq_data *data,
-				 char *extra);
-
-extern int
-ieee80211softmac_wx_get_scan_results(struct net_device *net_dev,
-				     struct iw_request_info *info,
-				     union iwreq_data *data,
-				     char *extra);
-
-extern int
-ieee80211softmac_wx_set_essid(struct net_device *net_dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *data,
-			      char *extra);
-
-extern int
-ieee80211softmac_wx_get_essid(struct net_device *net_dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *data,
-			      char *extra);
-
-extern int
-ieee80211softmac_wx_set_rate(struct net_device *net_dev,
-			     struct iw_request_info *info,
-			     union iwreq_data *data,
-			     char *extra);
-
-extern int
-ieee80211softmac_wx_get_rate(struct net_device *net_dev,
-			     struct iw_request_info *info,
-			     union iwreq_data *data,
-			     char *extra);
-
-extern int
-ieee80211softmac_wx_get_wap(struct net_device *net_dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *data,
-			      char *extra);
-
-extern int
-ieee80211softmac_wx_set_wap(struct net_device *net_dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *data,
-			      char *extra);
-
-extern int
-ieee80211softmac_wx_set_genie(struct net_device *dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *wrqu,
-			      char *extra);
-
-extern int
-ieee80211softmac_wx_get_genie(struct net_device *dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *wrqu,
-			      char *extra);
-extern int
-ieee80211softmac_wx_set_mlme(struct net_device *dev,
-			     struct iw_request_info *info,
-			     union iwreq_data *wrqu,
-			     char *extra);
-#endif /* _IEEE80211SOFTMAC_WX */
diff --git a/net/ieee80211/Kconfig b/net/ieee80211/Kconfig
index bd501046c9c0..94ed7d3cd9da 100644
--- a/net/ieee80211/Kconfig
+++ b/net/ieee80211/Kconfig
@@ -71,4 +71,3 @@ config IEEE80211_CRYPT_TKIP
 	This can be compiled as a module and it will be called
 	"ieee80211_crypt_tkip".
 
-source "net/ieee80211/softmac/Kconfig"
diff --git a/net/ieee80211/Makefile b/net/ieee80211/Makefile
index 796a7c76ee48..f988417121da 100644
--- a/net/ieee80211/Makefile
+++ b/net/ieee80211/Makefile
@@ -10,4 +10,3 @@ ieee80211-objs := \
 	ieee80211_wx.o \
 	ieee80211_geo.o
 
-obj-$(CONFIG_IEEE80211_SOFTMAC) += softmac/
diff --git a/net/ieee80211/softmac/Kconfig b/net/ieee80211/softmac/Kconfig
deleted file mode 100644
index 2811651cb134..000000000000
--- a/net/ieee80211/softmac/Kconfig
+++ /dev/null
@@ -1,12 +0,0 @@
-config IEEE80211_SOFTMAC
-	tristate "Software MAC add-on to the IEEE 802.11 networking stack"
-	depends on IEEE80211 && EXPERIMENTAL
-	select WIRELESS_EXT
-	select IEEE80211_CRYPT_WEP
-	---help---
-	This option enables the hardware independent software MAC addon
-	for the IEEE 802.11 networking stack.
-
-config IEEE80211_SOFTMAC_DEBUG
-	bool "Enable full debugging output"
-	depends on IEEE80211_SOFTMAC
diff --git a/net/ieee80211/softmac/Makefile b/net/ieee80211/softmac/Makefile
deleted file mode 100644
index bfcb391bb2c7..000000000000
--- a/net/ieee80211/softmac/Makefile
+++ /dev/null
@@ -1,9 +0,0 @@
-obj-$(CONFIG_IEEE80211_SOFTMAC) += ieee80211softmac.o
-ieee80211softmac-objs := \
-			ieee80211softmac_io.o \
-			ieee80211softmac_auth.o \
-			ieee80211softmac_module.o \
-			ieee80211softmac_scan.o \
-			ieee80211softmac_wx.o \
-			ieee80211softmac_assoc.o \
-			ieee80211softmac_event.o
diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c
deleted file mode 100644
index c4d122ddd72c..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_assoc.c
+++ /dev/null
@@ -1,489 +0,0 @@
-/*
- * This file contains the softmac's association logic.
- *
- * Copyright (c) 2005, 2006 Johannes Berg <johannes@sipsolutions.net>
- *                          Joseph Jezak <josejx@gentoo.org>
- *                          Larry Finger <Larry.Finger@lwfinger.net>
- *                          Danny van Dyk <kugelfang@gentoo.org>
- *                          Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#include "ieee80211softmac_priv.h"
-
-/*
- * Overview
- *
- * Before you can associate, you have to authenticate.
- *
- */
-
-/* Sends out an association request to the desired AP */
-static void
-ieee80211softmac_assoc(struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net)
-{
-	unsigned long flags;
-
-	/* Switch to correct channel for this network */
-	mac->set_channel(mac->dev, net->channel);
-
-	/* Send association request */
-	ieee80211softmac_send_mgt_frame(mac, net, IEEE80211_STYPE_ASSOC_REQ, 0);
-
-	dprintk(KERN_INFO PFX "sent association request!\n");
-
-	spin_lock_irqsave(&mac->lock, flags);
-	mac->associnfo.associated = 0; /* just to make sure */
-
-	/* Set a timer for timeout */
-	/* FIXME: make timeout configurable */
-	if (likely(mac->running))
-		queue_delayed_work(mac->wq, &mac->associnfo.timeout, 5 * HZ);
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
-
-void
-ieee80211softmac_assoc_timeout(struct work_struct *work)
-{
-	struct ieee80211softmac_device *mac =
-		container_of(work, struct ieee80211softmac_device,
-			     associnfo.timeout.work);
-	struct ieee80211softmac_network *n;
-
-	mutex_lock(&mac->associnfo.mutex);
-	/* we might race against ieee80211softmac_handle_assoc_response,
-	 * so make sure only one of us does something */
-	if (!mac->associnfo.associating)
-		goto out;
-	mac->associnfo.associating = 0;
-	mac->associnfo.bssvalid = 0;
-	mac->associnfo.associated = 0;
-
-	n = ieee80211softmac_get_network_by_bssid_locked(mac, mac->associnfo.bssid);
-
-	dprintk(KERN_INFO PFX "assoc request timed out!\n");
-	ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT, n);
-out:
-	mutex_unlock(&mac->associnfo.mutex);
-}
-
-void
-ieee80211softmac_disassoc(struct ieee80211softmac_device *mac)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	if (mac->associnfo.associating)
-		cancel_delayed_work(&mac->associnfo.timeout);
-
-	netif_carrier_off(mac->dev);
-
-	mac->associnfo.associated = 0;
-	mac->associnfo.bssvalid = 0;
-	mac->associnfo.associating = 0;
-	ieee80211softmac_init_bss(mac);
-	ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_DISASSOCIATED, NULL);
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
-
-/* Sends out a disassociation request to the desired AP */
-void
-ieee80211softmac_send_disassoc_req(struct ieee80211softmac_device *mac, u16 reason)
-{
-	struct ieee80211softmac_network *found;
-
-	if (mac->associnfo.bssvalid && mac->associnfo.associated) {
-		found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid);
-		if (found)
-			ieee80211softmac_send_mgt_frame(mac, found, IEEE80211_STYPE_DISASSOC, reason);
-	}
-
-	ieee80211softmac_disassoc(mac);
-}
-
-static inline int
-we_support_all_basic_rates(struct ieee80211softmac_device *mac, u8 *from, u8 from_len)
-{
-	int idx;
-	u8 rate;
-
-	for (idx = 0; idx < (from_len); idx++) {
-		rate = (from)[idx];
-		if (!(rate & IEEE80211_BASIC_RATE_MASK))
-			continue;
-		rate &= ~IEEE80211_BASIC_RATE_MASK;
-		if (!ieee80211softmac_ratesinfo_rate_supported(&mac->ratesinfo, rate))
-			return 0;
-	}
-	return 1;
-}
-
-static int
-network_matches_request(struct ieee80211softmac_device *mac, struct ieee80211_network *net)
-{
-	/* we cannot associate to networks whose name we don't know */
-	if (ieee80211_is_empty_essid(net->ssid, net->ssid_len))
-		return 0;
-	/* do not associate to a network whose BSSBasicRateSet we cannot support */
-	if (!we_support_all_basic_rates(mac, net->rates, net->rates_len))
-		return 0;
-	/* do we really need to check the ex rates? */
-	if (!we_support_all_basic_rates(mac, net->rates_ex, net->rates_ex_len))
-		return 0;
-
-	/* assume that users know what they're doing ...
-	 * (note we don't let them select a net we're incompatible with) */
-	if (mac->associnfo.bssfixed) {
-		return !memcmp(mac->associnfo.bssid, net->bssid, ETH_ALEN);
-	}
-
-	/* if 'ANY' network requested, take any that doesn't have privacy enabled */
-	if (mac->associnfo.req_essid.len == 0
-	    && !(net->capability & WLAN_CAPABILITY_PRIVACY))
-		return 1;
-	if (net->ssid_len != mac->associnfo.req_essid.len)
-		return 0;
-	if (!memcmp(net->ssid, mac->associnfo.req_essid.data, mac->associnfo.req_essid.len))
-		return 1;
-	return 0;
-}
-
-static void
-ieee80211softmac_assoc_notify_scan(struct net_device *dev, int event_type, void *context)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	ieee80211softmac_assoc_work(&mac->associnfo.work.work);
-}
-
-static void
-ieee80211softmac_assoc_notify_auth(struct net_device *dev, int event_type, void *context)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-
-	switch (event_type) {
-	case IEEE80211SOFTMAC_EVENT_AUTHENTICATED:
-		ieee80211softmac_assoc_work(&mac->associnfo.work.work);
-		break;
-	case IEEE80211SOFTMAC_EVENT_AUTH_FAILED:
-	case IEEE80211SOFTMAC_EVENT_AUTH_TIMEOUT:
-		ieee80211softmac_disassoc(mac);
-		break;
-	}
-}
-
-/* This function is called to handle userspace requests (asynchronously) */
-void
-ieee80211softmac_assoc_work(struct work_struct *work)
-{
-	struct ieee80211softmac_device *mac =
-		container_of(work, struct ieee80211softmac_device,
-			     associnfo.work.work);
-	struct ieee80211softmac_network *found = NULL;
-	struct ieee80211_network *net = NULL, *best = NULL;
-	int bssvalid;
-	unsigned long flags;
-
-	mutex_lock(&mac->associnfo.mutex);
-
-	if (!mac->associnfo.associating)
-		goto out;
-
-	/* ieee80211_disassoc might clear this */
-	bssvalid = mac->associnfo.bssvalid;
-
-	/* meh */
-	if (mac->associnfo.associated)
-		ieee80211softmac_send_disassoc_req(mac, WLAN_REASON_DISASSOC_STA_HAS_LEFT);
-
-	/* try to find the requested network in our list, if we found one already */
-	if (bssvalid || mac->associnfo.bssfixed)
-		found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid);
-
-	/* Search the ieee80211 networks for this network if we didn't find it by bssid,
-	 * but only if we've scanned at least once (to get a better list of networks to
-	 * select from). If we have not scanned before, the !found logic below will be
-	 * invoked and will scan. */
-	if (!found && (mac->associnfo.scan_retry < IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT))
-	{
-		s8 rssi = -128;	/* if I don't initialise, gcc emits an invalid warning
-				   because it cannot follow the best pointer logic. */
-		spin_lock_irqsave(&mac->ieee->lock, flags);
-		list_for_each_entry(net, &mac->ieee->network_list, list) {
-			/* we're supposed to find the network with
-			 * the best signal here, as we're asked to join
-			 * any network with a specific ESSID, and many
-			 * different ones could have that.
-			 *
-			 * I'll for now just go with the reported rssi.
-			 *
-			 * We also should take into account the rateset
-			 * here to find the best BSSID to try.
-			 */
-			if (network_matches_request(mac, net)) {
-				if (!best) {
-					best = net;
-					rssi = best->stats.rssi;
-					continue;
-				}
-				/* we already had a matching network, so
-				 * compare their properties to get the
-				 * better of the two ... (see above)
-				 */
-				if (rssi < net->stats.rssi) {
-					best = net;
-					rssi = best->stats.rssi;
-				}
-			}
-		}
-		/* if we unlock here, we might get interrupted and the `best'
-		 * pointer could go stale */
-		if (best) {
-			found = ieee80211softmac_create_network(mac, best);
-			/* if found is still NULL, then we got -ENOMEM somewhere */
-			if (found)
-				ieee80211softmac_add_network(mac, found);
-		}
-		spin_unlock_irqrestore(&mac->ieee->lock, flags);
-	}
-
-	if (!found) {
-		if (mac->associnfo.scan_retry > 0) {
-			mac->associnfo.scan_retry--;
-
-			/* We know of no such network. Let's scan.
-			 * NB: this also happens if we had no memory to copy the network info...
-			 * Maybe we can hope to have more memory after scanning finishes ;)
-			 */
-			dprintk(KERN_INFO PFX "Associate: Scanning for networks first.\n");
-			ieee80211softmac_notify(mac->dev, IEEE80211SOFTMAC_EVENT_SCAN_FINISHED, ieee80211softmac_assoc_notify_scan, NULL);
-			if (ieee80211softmac_start_scan(mac)) {
-				dprintk(KERN_INFO PFX "Associate: failed to initiate scan. Is device up?\n");
-			}
-			goto out;
-		} else {
-			mac->associnfo.associating = 0;
-			mac->associnfo.associated = 0;
-
-			dprintk(KERN_INFO PFX "Unable to find matching network after scan!\n");
-			/* reset the retry counter for the next user request since we
-			 * break out and don't reschedule ourselves after this point. */
-			mac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
-			ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_NET_NOT_FOUND, NULL);
-			goto out;
-		}
-	}
-
-	/* reset the retry counter for the next user request since we
-	 * now found a net and will try to associate to it, but not
-	 * schedule this function again. */
-	mac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
-	mac->associnfo.bssvalid = 1;
-	memcpy(mac->associnfo.bssid, found->bssid, ETH_ALEN);
-	/* copy the ESSID for displaying it */
-	mac->associnfo.associate_essid.len = found->essid.len;
-	memcpy(mac->associnfo.associate_essid.data, found->essid.data, IW_ESSID_MAX_SIZE + 1);
-
-	/* we found a network! authenticate (if necessary) and associate to it. */
-	if (found->authenticating) {
-		dprintk(KERN_INFO PFX "Already requested authentication, waiting...\n");
-		if(!mac->associnfo.assoc_wait) {
-			mac->associnfo.assoc_wait = 1;
-			ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL);
-		}
-		goto out;
-	}
-	if (!found->authenticated && !found->authenticating) {
-		/* This relies on the fact that _auth_req only queues the work,
-		 * otherwise adding the notification would be racy. */
-		if (!ieee80211softmac_auth_req(mac, found)) {
-			if(!mac->associnfo.assoc_wait) {
-				dprintk(KERN_INFO PFX "Cannot associate without being authenticated, requested authentication\n");
-				mac->associnfo.assoc_wait = 1;
-				ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL);
-			}
-		} else {
-			printkl(KERN_WARNING PFX "Not authenticated, but requesting authentication failed. Giving up to associate\n");
-			mac->associnfo.assoc_wait = 0;
-			ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED, found);
-		}
-		goto out;
-	}
-	/* finally! now we can start associating */
-	mac->associnfo.assoc_wait = 0;
-	ieee80211softmac_assoc(mac, found);
-
-out:
-	mutex_unlock(&mac->associnfo.mutex);
-}
-
-/* call this to do whatever is necessary when we're associated */
-static void
-ieee80211softmac_associated(struct ieee80211softmac_device *mac,
-	struct ieee80211_assoc_response * resp,
-	struct ieee80211softmac_network *net)
-{
-	u16 cap = le16_to_cpu(resp->capability);
-	u8 erp_value = net->erp_value;
-
-	mac->associnfo.associating = 0;
-	mac->bssinfo.supported_rates = net->supported_rates;
-	ieee80211softmac_recalc_txrates(mac);
-
-	mac->associnfo.associated = 1;
-
-	mac->associnfo.short_preamble_available =
-		(cap & WLAN_CAPABILITY_SHORT_PREAMBLE) != 0;
-	ieee80211softmac_process_erp(mac, erp_value);
-
-	if (mac->set_bssid_filter)
-		mac->set_bssid_filter(mac->dev, net->bssid);
-	memcpy(mac->ieee->bssid, net->bssid, ETH_ALEN);
-	netif_carrier_on(mac->dev);
-
-	mac->association_id = le16_to_cpup(&resp->aid);
-}
-
-/* received frame handling functions */
-int
-ieee80211softmac_handle_assoc_response(struct net_device * dev,
-				       struct ieee80211_assoc_response * resp,
-				       struct ieee80211_network * _ieee80211_network)
-{
-	/* NOTE: the network parameter has to be mostly ignored by
-	 *       this code because it is the ieee80211's pointer
-	 *       to the struct, not ours (we made a copy)
-	 */
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	u16 status = le16_to_cpup(&resp->status);
-	struct ieee80211softmac_network *network = NULL;
-	unsigned long flags;
-	DECLARE_MAC_BUF(mac2);
-
-	if (unlikely(!mac->running))
-		return -ENODEV;
-
-	spin_lock_irqsave(&mac->lock, flags);
-
-	if (!mac->associnfo.associating) {
-		/* we race against the timeout function, so make sure
-		 * only one of us can do work */
-		spin_unlock_irqrestore(&mac->lock, flags);
-		return 0;
-	}
-	network = ieee80211softmac_get_network_by_bssid_locked(mac, resp->header.addr3);
-
-	/* someone sending us things without us knowing him? Ignore. */
-	if (!network) {
-		dprintk(KERN_INFO PFX "Received unrequested assocation response from %s\n",
-			print_mac(mac2, resp->header.addr3));
-		spin_unlock_irqrestore(&mac->lock, flags);
-		return 0;
-	}
-
-	/* now that we know it was for us, we can cancel the timeout */
-	cancel_delayed_work(&mac->associnfo.timeout);
-
-	/* if the association response included an ERP IE, update our saved
-	 * copy */
-	if (_ieee80211_network->flags & NETWORK_HAS_ERP_VALUE)
-		network->erp_value = _ieee80211_network->erp_value;
-
-	switch (status) {
-		case 0:
-			dprintk(KERN_INFO PFX "associated!\n");
-			ieee80211softmac_associated(mac, resp, network);
-			ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATED, network);
-			break;
-		case WLAN_REASON_STA_REQ_ASSOC_WITHOUT_AUTH:
-			if (!network->auth_desynced_once) {
-				/* there seem to be a few rare cases where our view of
-				 * the world is obscured, or buggy APs that don't DEAUTH
-				 * us properly. So we handle that, but allow it only once.
-				 */
-				printkl(KERN_INFO PFX "We were not authenticated during association, retrying...\n");
-				network->authenticated = 0;
-				/* we don't want to do this more than once ... */
-				network->auth_desynced_once = 1;
-				queue_delayed_work(mac->wq, &mac->associnfo.work, 0);
-				break;
-			}
-		default:
-			dprintk(KERN_INFO PFX "associating failed (reason: 0x%x)!\n", status);
-			mac->associnfo.associating = 0;
-			mac->associnfo.bssvalid = 0;
-			mac->associnfo.associated = 0;
-			ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED, network);
-	}
-
-	spin_unlock_irqrestore(&mac->lock, flags);
-	return 0;
-}
-
-void
-ieee80211softmac_try_reassoc(struct ieee80211softmac_device *mac)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	mac->associnfo.associating = 1;
-	queue_delayed_work(mac->wq, &mac->associnfo.work, 0);
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
-
-int
-ieee80211softmac_handle_disassoc(struct net_device * dev,
-				 struct ieee80211_disassoc *disassoc)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-
-	if (unlikely(!mac->running))
-		return -ENODEV;
-
-	if (memcmp(disassoc->header.addr2, mac->associnfo.bssid, ETH_ALEN))
-		return 0;
-
-	if (memcmp(disassoc->header.addr1, mac->dev->dev_addr, ETH_ALEN))
-		return 0;
-
-	dprintk(KERN_INFO PFX "got disassoc frame\n");
-	ieee80211softmac_disassoc(mac);
-
-	ieee80211softmac_try_reassoc(mac);
-
-	return 0;
-}
-
-int
-ieee80211softmac_handle_reassoc_req(struct net_device * dev,
-				    struct ieee80211_reassoc_request * resp)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	struct ieee80211softmac_network *network;
-
-	if (unlikely(!mac->running))
-		return -ENODEV;
-
-	network = ieee80211softmac_get_network_by_bssid(mac, resp->header.addr3);
-	if (!network) {
-		dprintkl(KERN_INFO PFX "reassoc request from unknown network\n");
-		return 0;
-	}
-	queue_delayed_work(mac->wq, &mac->associnfo.work, 0);
-
-	return 0;
-}
diff --git a/net/ieee80211/softmac/ieee80211softmac_auth.c b/net/ieee80211/softmac/ieee80211softmac_auth.c
deleted file mode 100644
index 1a96c2572578..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_auth.c
+++ /dev/null
@@ -1,413 +0,0 @@
-/*
- * This file contains the softmac's authentication logic.
- *
- * Copyright (c) 2005, 2006 Johannes Berg <johannes@sipsolutions.net>
- *                          Joseph Jezak <josejx@gentoo.org>
- *                          Larry Finger <Larry.Finger@lwfinger.net>
- *                          Danny van Dyk <kugelfang@gentoo.org>
- *                          Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#include "ieee80211softmac_priv.h"
-
-static void ieee80211softmac_auth_queue(struct work_struct *work);
-
-/* Queues an auth request to the desired AP */
-int
-ieee80211softmac_auth_req(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net)
-{
-	struct ieee80211softmac_auth_queue_item *auth;
-	unsigned long flags;
-	DECLARE_MAC_BUF(mac2);
-
-	if (net->authenticating || net->authenticated)
-		return 0;
-	net->authenticating = 1;
-
-	/* Add the network if it's not already added */
-	ieee80211softmac_add_network(mac, net);
-
-	dprintk(KERN_NOTICE PFX "Queueing Authentication Request to %s\n", print_mac(mac2, net->bssid));
-	/* Queue the auth request */
-	auth = (struct ieee80211softmac_auth_queue_item *)
-		kmalloc(sizeof(struct ieee80211softmac_auth_queue_item), GFP_KERNEL);
-	if(auth == NULL)
-		return -ENOMEM;
-
-	auth->net = net;
-	auth->mac = mac;
-	auth->retry = IEEE80211SOFTMAC_AUTH_RETRY_LIMIT;
-	auth->state = IEEE80211SOFTMAC_AUTH_OPEN_REQUEST;
-	INIT_DELAYED_WORK(&auth->work, ieee80211softmac_auth_queue);
-
-	/* Lock (for list) */
-	spin_lock_irqsave(&mac->lock, flags);
-
-	/* add to list */
-	list_add_tail(&auth->list, &mac->auth_queue);
-	queue_delayed_work(mac->wq, &auth->work, 0);
-	spin_unlock_irqrestore(&mac->lock, flags);
-
-	return 0;
-}
-
-
-/* Sends an auth request to the desired AP and handles timeouts */
-static void
-ieee80211softmac_auth_queue(struct work_struct *work)
-{
-	struct ieee80211softmac_device *mac;
-	struct ieee80211softmac_auth_queue_item *auth;
-	struct ieee80211softmac_network *net;
-	unsigned long flags;
-	DECLARE_MAC_BUF(mac2);
-
-	auth = container_of(work, struct ieee80211softmac_auth_queue_item,
-			    work.work);
-	net = auth->net;
-	mac = auth->mac;
-
-	if(auth->retry > 0) {
-		/* Switch to correct channel for this network */
-		mac->set_channel(mac->dev, net->channel);
-
-		/* Lock and set flags */
-		spin_lock_irqsave(&mac->lock, flags);
-		if (unlikely(!mac->running)) {
-			/* Prevent reschedule on workqueue flush */
-			spin_unlock_irqrestore(&mac->lock, flags);
-			return;
-		}
-		net->authenticated = 0;
-		/* add a timeout call so we eventually give up waiting for an auth reply */
-		queue_delayed_work(mac->wq, &auth->work, IEEE80211SOFTMAC_AUTH_TIMEOUT);
-		auth->retry--;
-		spin_unlock_irqrestore(&mac->lock, flags);
-		if (ieee80211softmac_send_mgt_frame(mac, auth->net, IEEE80211_STYPE_AUTH, auth->state))
-			dprintk(KERN_NOTICE PFX "Sending Authentication Request to %s failed (this shouldn't happen, wait for the timeout).\n",
-				print_mac(mac2, net->bssid));
-		else
-			dprintk(KERN_NOTICE PFX "Sent Authentication Request to %s.\n", print_mac(mac2, net->bssid));
-		return;
-	}
-
-	printkl(KERN_WARNING PFX "Authentication timed out with %s\n", print_mac(mac2, net->bssid));
-	/* Remove this item from the queue */
-	spin_lock_irqsave(&mac->lock, flags);
-	net->authenticating = 0;
-	ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_AUTH_TIMEOUT, net);
-	cancel_delayed_work(&auth->work); /* just to make sure... */
-	list_del(&auth->list);
-	spin_unlock_irqrestore(&mac->lock, flags);
-	/* Free it */
-	kfree(auth);
-}
-
-/* Sends a response to an auth challenge (for shared key auth). */
-static void
-ieee80211softmac_auth_challenge_response(struct work_struct *work)
-{
-	struct ieee80211softmac_auth_queue_item *aq =
-		container_of(work, struct ieee80211softmac_auth_queue_item,
-			     work.work);
-
-	/* Send our response */
-	ieee80211softmac_send_mgt_frame(aq->mac, aq->net, IEEE80211_STYPE_AUTH, aq->state);
-}
-
-/* Handle the auth response from the AP
- * This should be registered with ieee80211 as handle_auth
- */
-int
-ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *auth)
-{
-
-	struct list_head *list_ptr;
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	struct ieee80211softmac_auth_queue_item *aq = NULL;
-	struct ieee80211softmac_network *net = NULL;
-	unsigned long flags;
-	u8 * data;
-	DECLARE_MAC_BUF(mac2);
-
-	if (unlikely(!mac->running))
-		return -ENODEV;
-
-	/* Find correct auth queue item */
-	spin_lock_irqsave(&mac->lock, flags);
-	list_for_each(list_ptr, &mac->auth_queue) {
-		aq = list_entry(list_ptr, struct ieee80211softmac_auth_queue_item, list);
-		net = aq->net;
-		if (!memcmp(net->bssid, auth->header.addr2, ETH_ALEN))
-			break;
-		else
-			aq = NULL;
-	}
-	spin_unlock_irqrestore(&mac->lock, flags);
-
-	/* Make sure that we've got an auth queue item for this request */
-	if(aq == NULL)
-	{
-		dprintkl(KERN_DEBUG PFX "Authentication response received from %s but no queue item exists.\n", print_mac(mac2, auth->header.addr2));
-		/* Error #? */
-		return -1;
-	}
-
-	/* Check for out of order authentication */
-	if(!net->authenticating)
-	{
-		dprintkl(KERN_DEBUG PFX "Authentication response received from %s but did not request authentication.\n",print_mac(mac2, auth->header.addr2));
-		return -1;
-	}
-
-	/* Parse the auth packet */
-	switch(le16_to_cpu(auth->algorithm)) {
-	case WLAN_AUTH_OPEN:
-		/* Check the status code of the response */
-
-		switch(le16_to_cpu(auth->status)) {
-		case WLAN_STATUS_SUCCESS:
-			/* Update the status to Authenticated */
-			spin_lock_irqsave(&mac->lock, flags);
-			net->authenticating = 0;
-			net->authenticated = 1;
-			spin_unlock_irqrestore(&mac->lock, flags);
-
-			/* Send event */
-			printkl(KERN_NOTICE PFX "Open Authentication completed with %s\n", print_mac(mac2, net->bssid));
-			ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_AUTHENTICATED, net);
-			break;
-		default:
-			/* Lock and reset flags */
-			spin_lock_irqsave(&mac->lock, flags);
-			net->authenticated = 0;
-			net->authenticating = 0;
-			spin_unlock_irqrestore(&mac->lock, flags);
-
-			printkl(KERN_NOTICE PFX "Open Authentication with %s failed, error code: %i\n",
-				print_mac(mac2, net->bssid), le16_to_cpup(&auth->status));
-			/* Count the error? */
-			break;
-		}
-		goto free_aq;
-		break;
-	case WLAN_AUTH_SHARED_KEY:
-		/* Figure out where we are in the process */
-		switch(le16_to_cpu(auth->transaction)) {
-		case IEEE80211SOFTMAC_AUTH_SHARED_CHALLENGE:
-			/* Check to make sure we have a challenge IE */
-			data = (u8 *)auth->info_element;
-			if (*data++ != MFIE_TYPE_CHALLENGE) {
-				printkl(KERN_NOTICE PFX "Shared Key Authentication failed due to a missing challenge.\n");
-				break;
-			}
-			/* Save the challenge */
-			spin_lock_irqsave(&mac->lock, flags);
-			net->challenge_len = *data++;
-			if (net->challenge_len > WLAN_AUTH_CHALLENGE_LEN)
-				net->challenge_len = WLAN_AUTH_CHALLENGE_LEN;
-			kfree(net->challenge);
-			net->challenge = kmemdup(data, net->challenge_len,
-						 GFP_ATOMIC);
-			if (net->challenge == NULL) {
-				printkl(KERN_NOTICE PFX "Shared Key "
-					"Authentication failed due to "
-					"memory shortage.\n");
-				spin_unlock_irqrestore(&mac->lock, flags);
-				break;
-			}
-			aq->state = IEEE80211SOFTMAC_AUTH_SHARED_RESPONSE;
-
-			/* We reuse the work struct from the auth request here.
-			 * It is safe to do so as each one is per-request, and
-			 * at this point (dealing with authentication response)
-			 * we have obviously already sent the initial auth
-			 * request. */
-			cancel_delayed_work(&aq->work);
-			INIT_DELAYED_WORK(&aq->work, &ieee80211softmac_auth_challenge_response);
-			queue_delayed_work(mac->wq, &aq->work, 0);
-			spin_unlock_irqrestore(&mac->lock, flags);
-			return 0;
-		case IEEE80211SOFTMAC_AUTH_SHARED_PASS:
-			kfree(net->challenge);
-			net->challenge = NULL;
-			net->challenge_len = 0;
-			/* Check the status code of the response */
-			switch(auth->status) {
-			case WLAN_STATUS_SUCCESS:
-				/* Update the status to Authenticated */
-				spin_lock_irqsave(&mac->lock, flags);
-				net->authenticating = 0;
-				net->authenticated = 1;
-				spin_unlock_irqrestore(&mac->lock, flags);
-				printkl(KERN_NOTICE PFX "Shared Key Authentication completed with %s\n",
-					print_mac(mac2, net->bssid));
-				ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_AUTHENTICATED, net);
-				break;
-			default:
-				printkl(KERN_NOTICE PFX "Shared Key Authentication with %s failed, error code: %i\n",
-					print_mac(mac2, net->bssid), le16_to_cpup(&auth->status));
-				/* Lock and reset flags */
-				spin_lock_irqsave(&mac->lock, flags);
-				net->authenticating = 0;
-				net->authenticated = 0;
-				spin_unlock_irqrestore(&mac->lock, flags);
-				/* Count the error? */
-				break;
-			}
-			goto free_aq;
-			break;
-		default:
-			printkl(KERN_WARNING PFX "Unhandled Authentication Step: %i\n", auth->transaction);
-			break;
-		}
-		goto free_aq;
-		break;
-	default:
-		/* ERROR */
-		goto free_aq;
-		break;
-	}
-	return 0;
-free_aq:
-	/* Cancel the timeout */
-	spin_lock_irqsave(&mac->lock, flags);
-	cancel_delayed_work(&aq->work);
-	/* Remove this item from the queue */
-	list_del(&aq->list);
-	spin_unlock_irqrestore(&mac->lock, flags);
-
-	/* Free it */
-	kfree(aq);
-	return 0;
-}
-
-/*
- * Handle deauthorization
- */
-static void
-ieee80211softmac_deauth_from_net(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net)
-{
-	struct ieee80211softmac_auth_queue_item *aq = NULL;
-	struct list_head *list_ptr;
-	unsigned long flags;
-
-	/* deauthentication implies disassociation */
-	ieee80211softmac_disassoc(mac);
-
-	/* Lock and reset status flags */
-	spin_lock_irqsave(&mac->lock, flags);
-	net->authenticating = 0;
-	net->authenticated = 0;
-
-	/* Find correct auth queue item, if it exists */
-	list_for_each(list_ptr, &mac->auth_queue) {
-		aq = list_entry(list_ptr, struct ieee80211softmac_auth_queue_item, list);
-		if (!memcmp(net->bssid, aq->net->bssid, ETH_ALEN))
-			break;
-		else
-			aq = NULL;
-	}
-
-	/* Cancel pending work */
-	if(aq != NULL)
-		/* Not entirely safe?  What about running work? */
-		cancel_delayed_work(&aq->work);
-
-	/* Free our network ref */
-	ieee80211softmac_del_network_locked(mac, net);
-	if(net->challenge != NULL)
-		kfree(net->challenge);
-	kfree(net);
-
-	/* can't transmit data right now... */
-	netif_carrier_off(mac->dev);
-	spin_unlock_irqrestore(&mac->lock, flags);
-
-	ieee80211softmac_try_reassoc(mac);
-}
-
-/*
- * Sends a deauth request to the desired AP
- */
-int
-ieee80211softmac_deauth_req(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net, int reason)
-{
-	int ret;
-
-	/* Make sure the network is authenticated */
-	if (!net->authenticated)
-	{
-		dprintkl(KERN_DEBUG PFX "Can't send deauthentication packet, network is not authenticated.\n");
-		/* Error okay? */
-		return -EPERM;
-	}
-
-	/* Send the de-auth packet */
-	if((ret = ieee80211softmac_send_mgt_frame(mac, net, IEEE80211_STYPE_DEAUTH, reason)))
-		return ret;
-
-	ieee80211softmac_deauth_from_net(mac, net);
-	return 0;
-}
-
-/*
- * This should be registered with ieee80211 as handle_deauth
- */
-int
-ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *deauth)
-{
-
-	struct ieee80211softmac_network *net = NULL;
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	DECLARE_MAC_BUF(mac2);
-
-	if (unlikely(!mac->running))
-		return -ENODEV;
-
-	if (!deauth) {
-		dprintk("deauth without deauth packet. eek!\n");
-		return 0;
-	}
-
-	net = ieee80211softmac_get_network_by_bssid(mac, deauth->header.addr2);
-
-	if (net == NULL) {
-		dprintkl(KERN_DEBUG PFX "Received deauthentication packet from %s, but that network is unknown.\n",
-			print_mac(mac2, deauth->header.addr2));
-		return 0;
-	}
-
-	/* Make sure the network is authenticated */
-	if(!net->authenticated)
-	{
-		dprintkl(KERN_DEBUG PFX "Can't perform deauthentication, network is not authenticated.\n");
-		/* Error okay? */
-		return -EPERM;
-	}
-
-	ieee80211softmac_deauth_from_net(mac, net);
-
-	/* let's try to re-associate */
-	queue_delayed_work(mac->wq, &mac->associnfo.work, 0);
-	return 0;
-}
diff --git a/net/ieee80211/softmac/ieee80211softmac_event.c b/net/ieee80211/softmac/ieee80211softmac_event.c
deleted file mode 100644
index 8cef05b60f16..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_event.c
+++ /dev/null
@@ -1,189 +0,0 @@
-/*
- * Event system
- * Also see comments in public header file and longer explanation below.
- *
- * Copyright (c) 2005, 2006 Johannes Berg <johannes@sipsolutions.net>
- *                          Joseph Jezak <josejx@gentoo.org>
- *                          Larry Finger <Larry.Finger@lwfinger.net>
- *                          Danny van Dyk <kugelfang@gentoo.org>
- *                          Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#include "ieee80211softmac_priv.h"
-
-/*
- * Each event has associated to it
- *  - an event type (see constants in public header)
- *  - an event context (see below)
- *  - the function to be called
- *  - a context (extra parameter to call the function with)
- *  - and the softmac struct
- *
- * The event context is private and can only be used from
- * within this module. Its meaning varies with the event
- * type:
- *  SCAN_FINISHED,
- *  DISASSOCIATED:	NULL
- *  ASSOCIATED,
- *  ASSOCIATE_FAILED,
- *  ASSOCIATE_TIMEOUT,
- *  AUTHENTICATED,
- *  AUTH_FAILED,
- *  AUTH_TIMEOUT:	a pointer to the network struct
- * ...
- * Code within this module can use the event context to be only
- * called when the event is true for that specific context
- * as per above table.
- * If the event context is NULL, then the notification is always called,
- * regardless of the event context. The event context is not passed to
- * the callback, it is assumed that the context suffices.
- *
- * You can also use the event context only by setting the event type
- * to -1 (private use only), in which case you'll be notified
- * whenever the event context matches.
- */
-
-static char *event_descriptions[IEEE80211SOFTMAC_EVENT_LAST+1] = {
-	NULL, /* scan finished */
-	NULL, /* associated */
-	"associating failed",
-	"associating timed out",
-	"authenticated",
-	"authenticating failed",
-	"authenticating timed out",
-	"associating failed because no suitable network was found",
-	NULL, /* disassociated */
-};
-
-
-static void
-ieee80211softmac_notify_callback(struct work_struct *work)
-{
-	struct ieee80211softmac_event *pevent =
-		container_of(work, struct ieee80211softmac_event, work.work);
-	struct ieee80211softmac_event event = *pevent;
-	kfree(pevent);
-
-	event.fun(event.mac->dev, event.event_type, event.context);
-}
-
-int
-ieee80211softmac_notify_internal(struct ieee80211softmac_device *mac,
-	int event, void *event_context, notify_function_ptr fun, void *context, gfp_t gfp_mask)
-{
-	struct ieee80211softmac_event *eventptr;
-	unsigned long flags;
-
-	if (event < -1 || event > IEEE80211SOFTMAC_EVENT_LAST)
-		return -ENOSYS;
-
-	if (!fun)
-		return -EINVAL;
-
-	eventptr = kmalloc(sizeof(struct ieee80211softmac_event), gfp_mask);
-	if (!eventptr)
-		return -ENOMEM;
-
-	eventptr->event_type = event;
-	INIT_DELAYED_WORK(&eventptr->work, ieee80211softmac_notify_callback);
-	eventptr->fun = fun;
-	eventptr->context = context;
-	eventptr->mac = mac;
-	eventptr->event_context = event_context;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	list_add(&eventptr->list, &mac->events);
-	spin_unlock_irqrestore(&mac->lock, flags);
-
-	return 0;
-}
-
-int
-ieee80211softmac_notify_gfp(struct net_device *dev,
-	int event, notify_function_ptr fun, void *context, gfp_t gfp_mask)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-
-	if (event < 0 || event > IEEE80211SOFTMAC_EVENT_LAST)
-		return -ENOSYS;
-
-	return ieee80211softmac_notify_internal(mac, event, NULL, fun, context, gfp_mask);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_notify_gfp);
-
-/* private -- calling all callbacks that were specified */
-void
-ieee80211softmac_call_events_locked(struct ieee80211softmac_device *mac, int event, void *event_ctx)
-{
-	struct ieee80211softmac_event *eventptr, *tmp;
-	struct ieee80211softmac_network *network;
-
-	if (event >= 0) {
-		union iwreq_data wrqu;
-		int we_event;
-		char *msg = NULL;
-
-		memset(&wrqu, '\0', sizeof (union iwreq_data));
-
-		switch(event) {
-		case IEEE80211SOFTMAC_EVENT_ASSOCIATED:
-			network = (struct ieee80211softmac_network *)event_ctx;
-			memcpy(wrqu.ap_addr.sa_data, &network->bssid[0], ETH_ALEN);
-			/* fall through */
-		case IEEE80211SOFTMAC_EVENT_DISASSOCIATED:
-			wrqu.ap_addr.sa_family = ARPHRD_ETHER;
-			we_event = SIOCGIWAP;
-			break;
-		case IEEE80211SOFTMAC_EVENT_SCAN_FINISHED:
-			we_event = SIOCGIWSCAN;
-			break;
-		default:
-			msg = event_descriptions[event];
-			if (!msg)
-				msg = "SOFTMAC EVENT BUG";
-			wrqu.data.length = strlen(msg);
-			we_event = IWEVCUSTOM;
-			break;
-		}
-		wireless_send_event(mac->dev, we_event, &wrqu, msg);
-	}
-
-	if (!list_empty(&mac->events))
-		list_for_each_entry_safe(eventptr, tmp, &mac->events, list) {
-			if ((eventptr->event_type == event || eventptr->event_type == -1)
-				&& (eventptr->event_context == NULL || eventptr->event_context == event_ctx)) {
-				list_del(&eventptr->list);
-				/* User may have subscribed to ANY event, so
-				 * we tell them which event triggered it. */
-				eventptr->event_type = event;
-				queue_delayed_work(mac->wq, &eventptr->work, 0);
-			}
-		}
-}
-
-void
-ieee80211softmac_call_events(struct ieee80211softmac_device *mac, int event, void *event_ctx)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	ieee80211softmac_call_events_locked(mac, event, event_ctx);
-
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
diff --git a/net/ieee80211/softmac/ieee80211softmac_io.c b/net/ieee80211/softmac/ieee80211softmac_io.c
deleted file mode 100644
index 73b4b13fbd8f..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_io.c
+++ /dev/null
@@ -1,488 +0,0 @@
-/*
- * Some parts based on code from net80211
- * Copyright (c) 2001 Atsushi Onoe
- * Copyright (c) 2002-2005 Sam Leffler, Errno Consulting
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- * 1. Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- * 2. Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in the
- *    documentation and/or other materials provided with the distribution.
- * 3. The name of the author may not be used to endorse or promote products
- *    derived from this software without specific prior written permission.
- *
- * Alternatively, this software may be distributed under the terms of the
- * GNU General Public License ("GPL") version 2 as published by the Free
- * Software Foundation.
- *
- * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
- * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
- * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
- * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
- * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
- * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
- * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- */
-
-#include "ieee80211softmac_priv.h"
-
-/* Helper functions for inserting data into the frames */
-
-/*
- * Adds an ESSID element to the frame
- *
- */
-static u8 *
-ieee80211softmac_add_essid(u8 *dst, struct ieee80211softmac_essid *essid)
-{
-	if (essid) {
-		*dst++ = MFIE_TYPE_SSID;
-		*dst++ = essid->len;
-		memcpy(dst, essid->data, essid->len);
-		return dst+essid->len;
-	} else {
-		*dst++ = MFIE_TYPE_SSID;
-		*dst++ = 0;
-		return dst;
-	}
-}
-
-/* Adds Supported Rates and if required Extended Rates Information Element
- * to the frame, ASSUMES WE HAVE A SORTED LIST OF RATES */
-static u8 *
-ieee80211softmac_frame_add_rates(u8 *dst, const struct ieee80211softmac_ratesinfo *r)
-{
-	int cck_len, ofdm_len;
-	*dst++ = MFIE_TYPE_RATES;
-
-	for(cck_len=0; ieee80211_is_cck_rate(r->rates[cck_len]) && (cck_len < r->count);cck_len++);
-
-	if(cck_len > IEEE80211SOFTMAC_MAX_RATES_LEN)
-		cck_len = IEEE80211SOFTMAC_MAX_RATES_LEN;
-	*dst++ = cck_len;
-	memcpy(dst, r->rates, cck_len);
-	dst += cck_len;
-
-	if(cck_len < r->count){
-		for (ofdm_len=0; ieee80211_is_ofdm_rate(r->rates[ofdm_len + cck_len]) && (ofdm_len + cck_len < r->count); ofdm_len++);
-		if (ofdm_len > 0) {
-			if (ofdm_len > IEEE80211SOFTMAC_MAX_EX_RATES_LEN)
-				ofdm_len = IEEE80211SOFTMAC_MAX_EX_RATES_LEN;
-			*dst++ = MFIE_TYPE_RATES_EX;
-			*dst++ = ofdm_len;
-			memcpy(dst, r->rates + cck_len, ofdm_len);
-			dst += ofdm_len;
-		}
-	}
-	return dst;
-}
-
-/* Allocate a management frame */
-static u8 *
-ieee80211softmac_alloc_mgt(u32 size)
-{
-	u8 * data;
-
-	/* Add the header and FCS to the size */
-	size = size + IEEE80211_3ADDR_LEN;
-	if(size > IEEE80211_DATA_LEN)
-		return NULL;
-	/* Allocate the frame */
-	data = kzalloc(size, GFP_ATOMIC);
-	return data;
-}
-
-/*
- * Add a 2 Address Header
- */
-static void
-ieee80211softmac_hdr_2addr(struct ieee80211softmac_device *mac,
-	struct ieee80211_hdr_2addr *header, u32 type, u8 *dest)
-{
-	/* Fill in the frame control flags */
-	header->frame_ctl = cpu_to_le16(type);
-	/* Control packets always have WEP turned off */
-	if(type > IEEE80211_STYPE_CFENDACK && type < IEEE80211_STYPE_PSPOLL)
-		header->frame_ctl |= mac->ieee->sec.level ? cpu_to_le16(IEEE80211_FCTL_PROTECTED) : 0;
-
-	/* Fill in the duration */
-	header->duration_id = 0;
-	/* FIXME: How do I find this?
-	 * calculate. But most drivers just fill in 0 (except if it's a station id of course) */
-
-	/* Fill in the Destination Address */
-	if(dest == NULL)
-		memset(header->addr1, 0xFF, ETH_ALEN);
-	else
-		memcpy(header->addr1, dest, ETH_ALEN);
-	/* Fill in the Source Address */
-	memcpy(header->addr2, mac->ieee->dev->dev_addr, ETH_ALEN);
-
-}
-
-
-/* Add a 3 Address Header */
-static void
-ieee80211softmac_hdr_3addr(struct ieee80211softmac_device *mac,
-	struct ieee80211_hdr_3addr *header, u32 type, u8 *dest, u8 *bssid)
-{
-	/* This is common with 2addr, so use that instead */
-	ieee80211softmac_hdr_2addr(mac, (struct ieee80211_hdr_2addr *)header, type, dest);
-
-	/* Fill in the BSS ID */
-	if(bssid == NULL)
-		memset(header->addr3, 0xFF, ETH_ALEN);
-	else
-		memcpy(header->addr3, bssid, ETH_ALEN);
-
-	/* Fill in the sequence # */
-	/* FIXME: I need to add this to the softmac struct
-	 * shouldn't the sequence number be in ieee80211? */
-}
-
-static __le16
-ieee80211softmac_capabilities(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net)
-{
-	__le16 capability = 0;
-
-	/* ESS and IBSS bits are set according to the current mode */
-	switch (mac->ieee->iw_mode) {
-	case IW_MODE_INFRA:
-		capability = cpu_to_le16(WLAN_CAPABILITY_ESS);
-		break;
-	case IW_MODE_ADHOC:
-		capability = cpu_to_le16(WLAN_CAPABILITY_IBSS);
-		break;
-	case IW_MODE_AUTO:
-		capability = cpu_to_le16(net->capabilities &
-			(WLAN_CAPABILITY_ESS|WLAN_CAPABILITY_IBSS));
-		break;
-	default:
-		/* bleh. we don't ever go to these modes */
-		printk(KERN_ERR PFX "invalid iw_mode!\n");
-		break;
-	}
-
-	/* CF Pollable / CF Poll Request */
-	/* Needs to be implemented, for now, the 0's == not supported */
-
-	/* Privacy Bit */
-	capability |= mac->ieee->sec.level ?
-		cpu_to_le16(WLAN_CAPABILITY_PRIVACY) : 0;
-
-	/* Short Preamble */
-	/* Always supported: we probably won't ever be powering devices which
-	 * dont support this... */
-	capability |= cpu_to_le16(WLAN_CAPABILITY_SHORT_PREAMBLE);
-
-	/* PBCC */
-	/* Not widely used */
-
-	/* Channel Agility */
-	/* Not widely used */
-
-	/* Short Slot */
-	/* Will be implemented later */
-
-	/* DSSS-OFDM */
-	/* Not widely used */
-
-	return capability;
-}
-
-/*****************************************************************************
- * Create Management packets
- *****************************************************************************/
-
-/* Creates an association request packet */
-static u32
-ieee80211softmac_assoc_req(struct ieee80211_assoc_request **pkt,
-	struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net)
-{
-	u8 *data;
-	(*pkt) = (struct ieee80211_assoc_request *)ieee80211softmac_alloc_mgt(
-		2 +		/* Capability Info */
-		2 +	 	/* Listen Interval */
-		/* SSID IE */
-		1 + 1 + IW_ESSID_MAX_SIZE +
-		/* Rates IE */
-		1 + 1 + IEEE80211SOFTMAC_MAX_RATES_LEN +
-		/* Extended Rates IE */
-		1 + 1 + IEEE80211SOFTMAC_MAX_EX_RATES_LEN +
-		/* WPA IE if present */
-		mac->wpa.IElen
-		/* Other IE's?  Optional?
-		 * Yeah, probably need an extra IE parameter -- lots of vendors like to
-		 * fill in their own IEs */
-	);
-	if (unlikely((*pkt) == NULL))
-		return 0;
-	ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_ASSOC_REQ, net->bssid, net->bssid);
-
-	/* Fill in the capabilities */
-	(*pkt)->capability = ieee80211softmac_capabilities(mac, net);
-
-	/* Fill in Listen Interval (?) */
-	(*pkt)->listen_interval = cpu_to_le16(10);
-
-	data = (u8 *)(*pkt)->info_element;
-	/* Add SSID */
-	data = ieee80211softmac_add_essid(data, &net->essid);
-	/* Add Rates */
-	data = ieee80211softmac_frame_add_rates(data, &mac->ratesinfo);
-	/* Add WPA IE */
-	if (mac->wpa.IElen && mac->wpa.IE) {
-		memcpy(data, mac->wpa.IE, mac->wpa.IElen);
-		data += mac->wpa.IElen;
-	}
-	/* Return the number of used bytes */
-	return (data - (u8*)(*pkt));
-}
-
-/* Create a reassociation request packet */
-static u32
-ieee80211softmac_reassoc_req(struct ieee80211_reassoc_request **pkt,
-	struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net)
-{
-	u8 *data;
-	(*pkt) = (struct ieee80211_reassoc_request *)ieee80211softmac_alloc_mgt(
-		2 +		/* Capability Info */
-		2 +	 	/* Listen Interval */
-		ETH_ALEN +	/* AP MAC */
-		/* SSID IE */
-		1 + 1 + IW_ESSID_MAX_SIZE +
-		/* Rates IE */
-		1 + 1 + IEEE80211SOFTMAC_MAX_RATES_LEN +
-		/* Extended Rates IE */
-		1 + 1 + IEEE80211SOFTMAC_MAX_EX_RATES_LEN
-		/* Other IE's? */
-	);
-	if (unlikely((*pkt) == NULL))
-		return 0;
-	ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_REASSOC_REQ, net->bssid, net->bssid);
-
-	/* Fill in the capabilities */
-	(*pkt)->capability = ieee80211softmac_capabilities(mac, net);
-
-	/* Fill in Listen Interval (?) */
-	(*pkt)->listen_interval = cpu_to_le16(10);
-	/* Fill in the current AP MAC */
-	memcpy((*pkt)->current_ap, mac->ieee->bssid, ETH_ALEN);
-
-	data = (u8 *)(*pkt)->info_element;
-	/* Add SSID */
-	data = ieee80211softmac_add_essid(data, &net->essid);
-	/* Add Rates */
-	data = ieee80211softmac_frame_add_rates(data, &mac->ratesinfo);
-	/* Return packet size */
-	return (data - (u8 *)(*pkt));
-}
-
-/* Create an authentication packet */
-static u32
-ieee80211softmac_auth(struct ieee80211_auth **pkt,
-	struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net,
-	u16 transaction, u16 status, int *encrypt_mpdu)
-{
-	u8 *data;
-	int auth_mode = mac->ieee->sec.auth_mode;
-	int is_shared_response = (auth_mode == WLAN_AUTH_SHARED_KEY
-		&& transaction == IEEE80211SOFTMAC_AUTH_SHARED_RESPONSE);
-
-	/* Allocate Packet */
-	(*pkt) = (struct ieee80211_auth *)ieee80211softmac_alloc_mgt(
-		2 +		/* Auth Algorithm */
-		2 +		/* Auth Transaction Seq */
-		2 +		/* Status Code */
-		 /* Challenge Text IE */
-		(is_shared_response ? 1 + 1 + net->challenge_len : 0)
-	);
-	if (unlikely((*pkt) == NULL))
-		return 0;
-	ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_AUTH, net->bssid, net->bssid);
-
-	/* Algorithm */
-	(*pkt)->algorithm = cpu_to_le16(auth_mode);
-	/* Transaction */
-	(*pkt)->transaction = cpu_to_le16(transaction);
-	/* Status */
-	(*pkt)->status = cpu_to_le16(status);
-
-	data = (u8 *)(*pkt)->info_element;
-	/* Challenge Text */
-	if (is_shared_response) {
-		*data = MFIE_TYPE_CHALLENGE;
-		data++;
-
-		/* Copy the challenge in */
-		*data = net->challenge_len;
-		data++;
-		memcpy(data, net->challenge, net->challenge_len);
-		data += net->challenge_len;
-
-		/* Make sure this frame gets encrypted with the shared key */
-		*encrypt_mpdu = 1;
-	} else
-		*encrypt_mpdu = 0;
-
-	/* Return the packet size */
-	return (data - (u8 *)(*pkt));
-}
-
-/* Create a disassocation or deauthentication packet */
-static u32
-ieee80211softmac_disassoc_deauth(struct ieee80211_disassoc **pkt,
-	struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net,
-	u16 type, u16 reason)
-{
-	/* Allocate Packet */
-	(*pkt) = (struct ieee80211_disassoc *)ieee80211softmac_alloc_mgt(2);
-	if (unlikely((*pkt) == NULL))
-		return 0;
-	ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), type, net->bssid, net->bssid);
-	/* Reason */
-	(*pkt)->reason = cpu_to_le16(reason);
-	/* Return the packet size */
-	return (2 + IEEE80211_3ADDR_LEN);
-}
-
-/* Create a probe request packet */
-static u32
-ieee80211softmac_probe_req(struct ieee80211_probe_request **pkt,
-	struct ieee80211softmac_device *mac, struct ieee80211softmac_essid *essid)
-{
-	u8 *data;
-	/* Allocate Packet */
-	(*pkt) = (struct ieee80211_probe_request *)ieee80211softmac_alloc_mgt(
-		/* SSID of requested network */
-		1 + 1 + IW_ESSID_MAX_SIZE +
-		/* Rates IE */
-		1 + 1 + IEEE80211SOFTMAC_MAX_RATES_LEN +
-		/* Extended Rates IE */
-		1 + 1 + IEEE80211SOFTMAC_MAX_EX_RATES_LEN
-	);
-	if (unlikely((*pkt) == NULL))
-		return 0;
-	ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_PROBE_REQ, NULL, NULL);
-
-	data = (u8 *)(*pkt)->info_element;
-	/* Add ESSID (can be NULL) */
-	data = ieee80211softmac_add_essid(data, essid);
-	/* Add Rates */
-	data = ieee80211softmac_frame_add_rates(data, &mac->ratesinfo);
-	/* Return packet size */
-	return (data - (u8 *)(*pkt));
-}
-
-/* Create a probe response packet */
-/* FIXME: Not complete */
-static u32
-ieee80211softmac_probe_resp(struct ieee80211_probe_response **pkt,
-	struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net)
-{
-	u8 *data;
-	/* Allocate Packet */
-	(*pkt) = (struct ieee80211_probe_response *)ieee80211softmac_alloc_mgt(
-		8 +		/* Timestamp */
-		2 +		/* Beacon Interval */
-		2 +		/* Capability Info */
-				/* SSID IE */
-		1 + 1 + IW_ESSID_MAX_SIZE +
-		7 + 		/* FH Parameter Set */
-		2 +		/* DS Parameter Set */
-		8 +		/* CF Parameter Set */
-		4 		/* IBSS Parameter Set */
-	);
-	if (unlikely((*pkt) == NULL))
-		return 0;
-	ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_PROBE_RESP, net->bssid, net->bssid);
-	data = (u8 *)(*pkt)->info_element;
-
-	/* Return the packet size */
-	return (data - (u8 *)(*pkt));
-}
-
-
-/* Sends a manangement packet
- * FIXME: document the use of the arg parameter
- * for _AUTH: (transaction #) | (status << 16)
- */
-int
-ieee80211softmac_send_mgt_frame(struct ieee80211softmac_device *mac,
-	void *ptrarg, u32 type, u32 arg)
-{
-	void *pkt = NULL;
-	u32 pkt_size = 0;
-	int encrypt_mpdu = 0;
-
-	switch(type) {
-	case IEEE80211_STYPE_ASSOC_REQ:
-		pkt_size = ieee80211softmac_assoc_req((struct ieee80211_assoc_request **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg);
-		break;
-	case IEEE80211_STYPE_REASSOC_REQ:
-		pkt_size = ieee80211softmac_reassoc_req((struct ieee80211_reassoc_request **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg);
-		break;
-	case IEEE80211_STYPE_AUTH:
-		pkt_size = ieee80211softmac_auth((struct ieee80211_auth **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg, (u16)(arg & 0xFFFF), (u16) (arg >> 16), &encrypt_mpdu);
-		break;
-	case IEEE80211_STYPE_DISASSOC:
-	case IEEE80211_STYPE_DEAUTH:
-		pkt_size = ieee80211softmac_disassoc_deauth((struct ieee80211_disassoc **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg, type, (u16)(arg & 0xFFFF));
-		break;
-	case IEEE80211_STYPE_PROBE_REQ:
-		pkt_size = ieee80211softmac_probe_req((struct ieee80211_probe_request **)(&pkt), mac, (struct ieee80211softmac_essid *)ptrarg);
-		break;
-	case IEEE80211_STYPE_PROBE_RESP:
-		pkt_size = ieee80211softmac_probe_resp((struct ieee80211_probe_response **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg);
-		break;
-	default:
-		printkl(KERN_DEBUG PFX "Unsupported Management Frame type: %i\n", type);
-		return -EINVAL;
-	};
-
-	if(pkt_size == 0 || pkt == NULL) {
-		printkl(KERN_DEBUG PFX "Error, packet is nonexistant or 0 length\n");
-		return -ENOMEM;
-	}
-
-	/* Send the packet to the ieee80211 layer for tx */
-	/* we defined softmac->mgmt_xmit for this. Should we keep it
-	 * as it is (that means we'd need to wrap this into a txb),
-	 * modify the prototype (so it matches this function),
-	 * or get rid of it alltogether?
-	 * Does this work for you now?
-	 */
-	ieee80211_tx_frame(mac->ieee, (struct ieee80211_hdr *)pkt,
-		IEEE80211_3ADDR_LEN, pkt_size, encrypt_mpdu);
-
-	kfree(pkt);
-	return 0;
-}
-
-/* Beacon handling */
-int ieee80211softmac_handle_beacon(struct net_device *dev,
-	struct ieee80211_beacon *beacon,
-	struct ieee80211_network *network)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-
-	/* This might race, but we don't really care and it's not worth
-	 * adding heavyweight locking in this fastpath.
-	 */
-	if (mac->associnfo.associated) {
-		if (memcmp(network->bssid, mac->associnfo.bssid, ETH_ALEN) == 0)
-			ieee80211softmac_process_erp(mac, network->erp_value);
-	}
-
-	return 0;
-}
-
diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c
deleted file mode 100644
index 07505ca859af..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_module.c
+++ /dev/null
@@ -1,568 +0,0 @@
-/*
- * Contains some basic softmac functions along with module registration code etc.
- *
- * Copyright (c) 2005, 2006 Johannes Berg <johannes@sipsolutions.net>
- *                          Joseph Jezak <josejx@gentoo.org>
- *                          Larry Finger <Larry.Finger@lwfinger.net>
- *                          Danny van Dyk <kugelfang@gentoo.org>
- *                          Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#include "ieee80211softmac_priv.h"
-#include <linux/sort.h>
-#include <linux/etherdevice.h>
-
-struct net_device *alloc_ieee80211softmac(int sizeof_priv)
-{
-	struct ieee80211softmac_device *softmac;
-	struct net_device *dev;
-
-	dev = alloc_ieee80211(sizeof(*softmac) + sizeof_priv);
-	if (!dev)
-		return NULL;
-	softmac = ieee80211_priv(dev);
-	softmac->wq = create_freezeable_workqueue("softmac");
-	if (!softmac->wq) {
-		free_ieee80211(dev);
-		return NULL;
-	}
-
-	softmac->dev = dev;
-	softmac->ieee = netdev_priv(dev);
-	spin_lock_init(&softmac->lock);
-
-	softmac->ieee->handle_auth = ieee80211softmac_auth_resp;
-	softmac->ieee->handle_deauth = ieee80211softmac_deauth_resp;
-	softmac->ieee->handle_assoc_response = ieee80211softmac_handle_assoc_response;
-	softmac->ieee->handle_reassoc_request = ieee80211softmac_handle_reassoc_req;
-	softmac->ieee->handle_disassoc = ieee80211softmac_handle_disassoc;
-	softmac->ieee->handle_beacon = ieee80211softmac_handle_beacon;
-	softmac->scaninfo = NULL;
-
-	softmac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
-
-	/* TODO: initialise all the other callbacks in the ieee struct
-	 *	 (once they're written)
-	 */
-
-	INIT_LIST_HEAD(&softmac->auth_queue);
-	INIT_LIST_HEAD(&softmac->network_list);
-	INIT_LIST_HEAD(&softmac->events);
-
-	mutex_init(&softmac->associnfo.mutex);
-	INIT_DELAYED_WORK(&softmac->associnfo.work, ieee80211softmac_assoc_work);
-	INIT_DELAYED_WORK(&softmac->associnfo.timeout, ieee80211softmac_assoc_timeout);
-	softmac->start_scan = ieee80211softmac_start_scan_implementation;
-	softmac->wait_for_scan = ieee80211softmac_wait_for_scan_implementation;
-	softmac->stop_scan = ieee80211softmac_stop_scan_implementation;
-
-	/* to start with, we can't send anything ... */
-	netif_carrier_off(dev);
-
-	return dev;
-}
-EXPORT_SYMBOL_GPL(alloc_ieee80211softmac);
-
-/* Clears the pending work queue items, stops all scans, etc. */
-void
-ieee80211softmac_clear_pending_work(struct ieee80211softmac_device *sm)
-{
-	unsigned long flags;
-	struct ieee80211softmac_event *eventptr, *eventtmp;
-	struct ieee80211softmac_auth_queue_item *authptr, *authtmp;
-	struct ieee80211softmac_network *netptr, *nettmp;
-
-	ieee80211softmac_stop_scan(sm);
-	ieee80211softmac_wait_for_scan(sm);
-
-	spin_lock_irqsave(&sm->lock, flags);
-	sm->running = 0;
-
-	/* Free all pending assoc work items */
-	cancel_delayed_work(&sm->associnfo.work);
-
-	/* Free all pending scan work items */
-	if(sm->scaninfo != NULL)
-		cancel_delayed_work(&sm->scaninfo->softmac_scan);
-
-	/* Free all pending auth work items */
-	list_for_each_entry(authptr, &sm->auth_queue, list)
-		cancel_delayed_work(&authptr->work);
-
-	/* delete all pending event calls and work items */
-	list_for_each_entry_safe(eventptr, eventtmp, &sm->events, list)
-		cancel_delayed_work(&eventptr->work);
-
-	spin_unlock_irqrestore(&sm->lock, flags);
-	flush_workqueue(sm->wq);
-
-	/* now we should be save and no longer need locking... */
-	spin_lock_irqsave(&sm->lock, flags);
-	/* Free all pending auth work items */
-	list_for_each_entry_safe(authptr, authtmp, &sm->auth_queue, list) {
-		list_del(&authptr->list);
-		kfree(authptr);
-	}
-
-	/* delete all pending event calls and work items */
-	list_for_each_entry_safe(eventptr, eventtmp, &sm->events, list) {
-		list_del(&eventptr->list);
-		kfree(eventptr);
-	}
-
-	/* Free all networks */
-	list_for_each_entry_safe(netptr, nettmp, &sm->network_list, list) {
-		ieee80211softmac_del_network_locked(sm, netptr);
-		if(netptr->challenge != NULL)
-			kfree(netptr->challenge);
-		kfree(netptr);
-	}
-
-	spin_unlock_irqrestore(&sm->lock, flags);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_clear_pending_work);
-
-void free_ieee80211softmac(struct net_device *dev)
-{
-	struct ieee80211softmac_device *sm = ieee80211_priv(dev);
-	ieee80211softmac_clear_pending_work(sm);
-	kfree(sm->scaninfo);
-	kfree(sm->wpa.IE);
-	destroy_workqueue(sm->wq);
-	free_ieee80211(dev);
-}
-EXPORT_SYMBOL_GPL(free_ieee80211softmac);
-
-static void ieee80211softmac_start_check_rates(struct ieee80211softmac_device *mac)
-{
-	struct ieee80211softmac_ratesinfo *ri = &mac->ratesinfo;
-	/* I took out the sorting check, we're seperating by modulation now. */
-	if (ri->count)
-		return;
-	/* otherwise assume we hav'em all! */
-	if (mac->ieee->modulation & IEEE80211_CCK_MODULATION) {
-		ri->rates[ri->count++] = IEEE80211_CCK_RATE_1MB;
-		ri->rates[ri->count++] = IEEE80211_CCK_RATE_2MB;
-		ri->rates[ri->count++] = IEEE80211_CCK_RATE_5MB;
-		ri->rates[ri->count++] = IEEE80211_CCK_RATE_11MB;
-	}
-	if (mac->ieee->modulation & IEEE80211_OFDM_MODULATION) {
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_6MB;
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_9MB;
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_12MB;
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_18MB;
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_24MB;
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_36MB;
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_48MB;
-		ri->rates[ri->count++] = IEEE80211_OFDM_RATE_54MB;
-	}
-}
-
-int ieee80211softmac_ratesinfo_rate_supported(struct ieee80211softmac_ratesinfo *ri, u8 rate)
-{
-	int search;
-	u8 search_rate;
-
-	for (search = 0; search < ri->count; search++) {
-		search_rate = ri->rates[search];
-		search_rate &= ~IEEE80211_BASIC_RATE_MASK;
-		if (rate == search_rate)
-			return 1;
-	}
-
-	return 0;
-}
-
-u8 ieee80211softmac_highest_supported_rate(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_ratesinfo *ri, int basic_only)
-{
-	u8 user_rate = mac->txrates.user_rate;
-	int i;
-
-	if (ri->count == 0)
-		return IEEE80211_CCK_RATE_1MB;
-
-	for (i = ri->count - 1; i >= 0; i--) {
-		u8 rate = ri->rates[i];
-		if (basic_only && !(rate & IEEE80211_BASIC_RATE_MASK))
-			continue;
-		rate &= ~IEEE80211_BASIC_RATE_MASK;
-		if (rate > user_rate)
-			continue;
-		if (ieee80211softmac_ratesinfo_rate_supported(&mac->ratesinfo, rate))
-			return rate;
-	}
-
-	/* If we haven't found a suitable rate by now, just trust the user */
-	return user_rate;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_highest_supported_rate);
-
-void ieee80211softmac_process_erp(struct ieee80211softmac_device *mac,
-	u8 erp_value)
-{
-	int use_protection;
-	int short_preamble;
-	u32 changes = 0;
-
-	/* Barker preamble mode */
-	short_preamble = ((erp_value & WLAN_ERP_BARKER_PREAMBLE) == 0
-			  && mac->associnfo.short_preamble_available) ? 1 : 0;
-
-	/* Protection needed? */
-	use_protection = (erp_value & WLAN_ERP_USE_PROTECTION) != 0;
-
-	if (mac->bssinfo.short_preamble != short_preamble) {
-		changes |= IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE;
-		mac->bssinfo.short_preamble = short_preamble;
-	}
-
-	if (mac->bssinfo.use_protection != use_protection) {
-		changes |= IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION;
-		mac->bssinfo.use_protection = use_protection;
-	}
-
-	if (mac->bssinfo_change && changes)
-		mac->bssinfo_change(mac->dev, changes);
-}
-
-void ieee80211softmac_recalc_txrates(struct ieee80211softmac_device *mac)
-{
-	struct ieee80211softmac_txrates *txrates = &mac->txrates;
-	u32 change = 0;
-
-	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
-	txrates->default_rate = ieee80211softmac_highest_supported_rate(mac, &mac->bssinfo.supported_rates, 0);
-
-	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK;
-	txrates->default_fallback = lower_rate(mac, txrates->default_rate);
-
-	change |= IEEE80211SOFTMAC_TXRATECHG_MCAST;
-	txrates->mcast_rate = ieee80211softmac_highest_supported_rate(mac, &mac->bssinfo.supported_rates, 1);
-
-	if (mac->txrates_change)
-		mac->txrates_change(mac->dev, change);
-
-}
-
-void ieee80211softmac_init_bss(struct ieee80211softmac_device *mac)
-{
-	struct ieee80211_device *ieee = mac->ieee;
-	u32 change = 0;
-	struct ieee80211softmac_txrates *txrates = &mac->txrates;
-	struct ieee80211softmac_bss_info *bssinfo = &mac->bssinfo;
-
-	/* TODO: We need some kind of state machine to lower the default rates
-	 *       if we loose too many packets.
-	 */
-	/* Change the default txrate to the highest possible value.
-	 * The txrate machine will lower it, if it is too high.
-	 */
-	if (ieee->modulation & IEEE80211_OFDM_MODULATION)
-		txrates->user_rate = IEEE80211_OFDM_RATE_24MB;
-	else
-		txrates->user_rate = IEEE80211_CCK_RATE_11MB;
-
-	txrates->default_rate = IEEE80211_CCK_RATE_1MB;
-	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
-
-	txrates->default_fallback = IEEE80211_CCK_RATE_1MB;
-	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK;
-
-	txrates->mcast_rate = IEEE80211_CCK_RATE_1MB;
-	change |= IEEE80211SOFTMAC_TXRATECHG_MCAST;
-
-	txrates->mgt_mcast_rate = IEEE80211_CCK_RATE_1MB;
-	change |= IEEE80211SOFTMAC_TXRATECHG_MGT_MCAST;
-
-	if (mac->txrates_change)
-		mac->txrates_change(mac->dev, change);
-
-	change = 0;
-
-	bssinfo->supported_rates.count = 0;
-	memset(bssinfo->supported_rates.rates, 0,
-		sizeof(bssinfo->supported_rates.rates));
-	change |= IEEE80211SOFTMAC_BSSINFOCHG_RATES;
-
-	bssinfo->short_preamble = 0;
-	change |= IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE;
-
-	bssinfo->use_protection = 0;
-	change |= IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION;
-
-	if (mac->bssinfo_change)
-		mac->bssinfo_change(mac->dev, change);
-
-	mac->running = 1;
-}
-
-void ieee80211softmac_start(struct net_device *dev)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-
-	ieee80211softmac_start_check_rates(mac);
-	ieee80211softmac_init_bss(mac);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_start);
-
-void ieee80211softmac_stop(struct net_device *dev)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-
-	ieee80211softmac_clear_pending_work(mac);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_stop);
-
-void ieee80211softmac_set_rates(struct net_device *dev, u8 count, u8 *rates)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	unsigned long flags;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	memcpy(mac->ratesinfo.rates, rates, count);
-	mac->ratesinfo.count = count;
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_set_rates);
-
-static u8 raise_rate(struct ieee80211softmac_device *mac, u8 rate)
-{
-	int i;
-	struct ieee80211softmac_ratesinfo *ri = &mac->ratesinfo;
-
-	for (i=0; i<ri->count-1; i++) {
-		if (ri->rates[i] == rate)
-			return ri->rates[i+1];
-	}
-	/* I guess we can't go any higher... */
-	return ri->rates[ri->count];
-}
-
-u8 ieee80211softmac_lower_rate_delta(struct ieee80211softmac_device *mac, u8 rate, int delta)
-{
-	int i;
-	struct ieee80211softmac_ratesinfo *ri = &mac->ratesinfo;
-
-	for (i=delta; i<ri->count; i++) {
-		if (ri->rates[i] == rate)
-			return ri->rates[i-delta];
-	}
-	/* I guess we can't go any lower... */
-	return ri->rates[0];
-}
-
-static void ieee80211softmac_add_txrates_badness(struct ieee80211softmac_device *mac,
-						 int amount)
-{
-	u8 default_rate = mac->txrates.default_rate;
-	u8 default_fallback = mac->txrates.default_fallback;
-	u32 changes = 0;
-
-	//TODO: This is highly experimental code.
-	//      Maybe the dynamic rate selection does not work
-	//      and it has to be removed again.
-
-printk("badness %d\n", mac->txrate_badness);
-	mac->txrate_badness += amount;
-	if (mac->txrate_badness <= -1000) {
-		/* Very small badness. Try a faster bitrate. */
-		default_rate = raise_rate(mac, default_rate);
-		changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
-		default_fallback = get_fallback_rate(mac, default_rate);
-		changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK;
-		mac->txrate_badness = 0;
-printk("Bitrate raised to %u\n", default_rate);
-	} else if (mac->txrate_badness >= 10000) {
-		/* Very high badness. Try a slower bitrate. */
-		default_rate = lower_rate(mac, default_rate);
-		changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
-		default_fallback = get_fallback_rate(mac, default_rate);
-		changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK;
-		mac->txrate_badness = 0;
-printk("Bitrate lowered to %u\n", default_rate);
-	}
-
-	mac->txrates.default_rate = default_rate;
-	mac->txrates.default_fallback = default_fallback;
-
-	if (changes && mac->txrates_change)
-		mac->txrates_change(mac->dev, changes);
-}
-
-void ieee80211softmac_fragment_lost(struct net_device *dev,
-				    u16 wl_seq)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	unsigned long flags;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	ieee80211softmac_add_txrates_badness(mac, 1000);
-	//TODO
-
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_fragment_lost);
-
-static int rate_cmp(const void *a_, const void *b_) {
-	u8 *a, *b;
-	a = (u8*)a_;
-	b = (u8*)b_;
-	return ((*a & ~IEEE80211_BASIC_RATE_MASK) - (*b & ~IEEE80211_BASIC_RATE_MASK));
-}
-
-/* Allocate a softmac network struct and fill it from a network */
-struct ieee80211softmac_network *
-ieee80211softmac_create_network(struct ieee80211softmac_device *mac,
-	struct ieee80211_network *net)
-{
-	struct ieee80211softmac_network *softnet;
-	softnet = kzalloc(sizeof(struct ieee80211softmac_network), GFP_ATOMIC);
-	if(softnet == NULL)
-		return NULL;
-	memcpy(softnet->bssid, net->bssid, ETH_ALEN);
-	softnet->channel = net->channel;
-	softnet->essid.len = net->ssid_len;
-	memcpy(softnet->essid.data, net->ssid, softnet->essid.len);
-
-	/* copy rates over */
-	softnet->supported_rates.count = net->rates_len;
-	memcpy(&softnet->supported_rates.rates[0], net->rates, net->rates_len);
-	memcpy(&softnet->supported_rates.rates[softnet->supported_rates.count], net->rates_ex, net->rates_ex_len);
-	softnet->supported_rates.count += net->rates_ex_len;
-	sort(softnet->supported_rates.rates, softnet->supported_rates.count, sizeof(softnet->supported_rates.rates[0]), rate_cmp, NULL);
-
-	/* we save the ERP value because it is needed at association time, and
-	 * many AP's do not include an ERP IE in the association response. */
-	softnet->erp_value = net->erp_value;
-
-	softnet->capabilities = net->capability;
-	return softnet;
-}
-
-
-/* Add a network to the list, while locked */
-void
-ieee80211softmac_add_network_locked(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *add_net)
-{
-	struct ieee80211softmac_network *softmac_net;
-
-	list_for_each_entry(softmac_net, &mac->network_list, list) {
-		if(!memcmp(softmac_net->bssid, add_net->bssid, ETH_ALEN))
-			return;
-	}
-	list_add(&(add_net->list), &mac->network_list);
-}
-
-/* Add a network to the list, with locking */
-void
-ieee80211softmac_add_network(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *add_net)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&mac->lock, flags);
-	ieee80211softmac_add_network_locked(mac, add_net);
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
-
-
-/* Delete a network from the list, while locked*/
-void
-ieee80211softmac_del_network_locked(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *del_net)
-{
-	list_del(&(del_net->list));
-}
-
-/* Delete a network from the list with locking */
-void
-ieee80211softmac_del_network(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *del_net)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&mac->lock, flags);
-	ieee80211softmac_del_network_locked(mac, del_net);
-	spin_unlock_irqrestore(&mac->lock, flags);
-}
-
-/* Get a network from the list by MAC while locked */
-struct ieee80211softmac_network *
-ieee80211softmac_get_network_by_bssid_locked(struct ieee80211softmac_device *mac,
-	u8 *bssid)
-{
-	struct ieee80211softmac_network *softmac_net;
-
-	list_for_each_entry(softmac_net, &mac->network_list, list) {
-		if(!memcmp(softmac_net->bssid, bssid, ETH_ALEN))
-			return softmac_net;
-	}
-	return NULL;
-}
-
-/* Get a network from the list by BSSID with locking */
-struct ieee80211softmac_network *
-ieee80211softmac_get_network_by_bssid(struct ieee80211softmac_device *mac,
-	u8 *bssid)
-{
-	unsigned long flags;
-	struct ieee80211softmac_network *softmac_net;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	softmac_net = ieee80211softmac_get_network_by_bssid_locked(mac, bssid);
-	spin_unlock_irqrestore(&mac->lock, flags);
-	return softmac_net;
-}
-
-/* Get a network from the list by ESSID while locked */
-struct ieee80211softmac_network *
-ieee80211softmac_get_network_by_essid_locked(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_essid *essid)
-{
-	struct ieee80211softmac_network *softmac_net;
-
-	list_for_each_entry(softmac_net, &mac->network_list, list) {
-		if (softmac_net->essid.len == essid->len &&
-			!memcmp(softmac_net->essid.data, essid->data, essid->len))
-			return softmac_net;
-	}
-	return NULL;
-}
-
-/* Get a network from the list by ESSID with locking */
-struct ieee80211softmac_network *
-ieee80211softmac_get_network_by_essid(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_essid *essid)
-{
-	unsigned long flags;
-	struct ieee80211softmac_network *softmac_net = NULL;
-
-	spin_lock_irqsave(&mac->lock, flags);
-	softmac_net = ieee80211softmac_get_network_by_essid_locked(mac, essid);
-	spin_unlock_irqrestore(&mac->lock, flags);
-	return softmac_net;
-}
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Johannes Berg");
-MODULE_AUTHOR("Joseph Jezak");
-MODULE_AUTHOR("Larry Finger");
-MODULE_AUTHOR("Danny van Dyk");
-MODULE_AUTHOR("Michael Buesch");
-MODULE_DESCRIPTION("802.11 software MAC");
diff --git a/net/ieee80211/softmac/ieee80211softmac_priv.h b/net/ieee80211/softmac/ieee80211softmac_priv.h
deleted file mode 100644
index f9232c8f6bdc..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_priv.h
+++ /dev/null
@@ -1,244 +0,0 @@
-/*
- * Internal softmac API definitions.
- *
- * Copyright (c) 2005, 2006 Johannes Berg <johannes@sipsolutions.net>
- *                          Joseph Jezak <josejx@gentoo.org>
- *                          Larry Finger <Larry.Finger@lwfinger.net>
- *                          Danny van Dyk <kugelfang@gentoo.org>
- *                          Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#ifndef IEEE80211SOFTMAC_PRIV_H_
-#define IEEE80211SOFTMAC_PRIV_H_
-
-#include <net/ieee80211softmac.h>
-#include <net/ieee80211softmac_wx.h>
-#include <linux/kernel.h>
-#include <linux/stringify.h>
-
-
-#define PFX				"SoftMAC: "
-
-#ifdef assert
-# undef assert
-#endif
-#ifdef CONFIG_IEEE80211_SOFTMAC_DEBUG
-#define assert(expr) \
-	do {										\
-		if (unlikely(!(expr))) {						\
-		printkl(KERN_ERR PFX "ASSERTION FAILED (%s) at: %s:%d:%s()\n", #expr,	\
-			__FILE__, __LINE__, __func__);				\
-		}									\
-	} while (0)
-#else
-#define assert(expr) do {} while (0)
-#endif
-
-/* rate limited printk(). */
-#ifdef printkl
-# undef printkl
-#endif
-#define printkl(f, x...)  do { if (printk_ratelimit()) printk(f ,##x); } while (0)
-/* rate limited printk() for debugging */
-#ifdef dprintkl
-# undef dprintkl
-#endif
-#ifdef CONFIG_IEEE80211_SOFTMAC_DEBUG
-# define dprintkl		printkl
-#else
-# define dprintkl(f, x...)	do { /* nothing */ } while (0)
-#endif
-
-/* debugging printk() */
-#ifdef dprintk
-# undef dprintk
-#endif
-#ifdef CONFIG_IEEE80211_SOFTMAC_DEBUG
-# define dprintk(f, x...)  do { printk(f ,##x); } while (0)
-#else
-# define dprintk(f, x...)  do { /* nothing */ } while (0)
-#endif
-
-/* private definitions and prototypes */
-
-/*** prototypes from _scan.c */
-void ieee80211softmac_scan(struct work_struct *work);
-/* for internal use if scanning is needed */
-int ieee80211softmac_start_scan(struct ieee80211softmac_device *mac);
-void ieee80211softmac_stop_scan(struct ieee80211softmac_device *mac);
-void ieee80211softmac_wait_for_scan(struct ieee80211softmac_device *mac);
-
-/* for use by _module.c to assign to the callbacks */
-int ieee80211softmac_start_scan_implementation(struct net_device *dev);
-void ieee80211softmac_stop_scan_implementation(struct net_device *dev);
-void ieee80211softmac_wait_for_scan_implementation(struct net_device *dev);
-
-/*** Network prototypes from _module.c */
-struct ieee80211softmac_network * ieee80211softmac_create_network(
-	struct ieee80211softmac_device *mac, struct ieee80211_network *net);
-void ieee80211softmac_add_network_locked(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net);
-void ieee80211softmac_add_network(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net);
-void ieee80211softmac_del_network_locked(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net);
-void ieee80211softmac_del_network(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_network *net);
-struct ieee80211softmac_network * ieee80211softmac_get_network_by_bssid_locked(
-	struct ieee80211softmac_device *mac, u8 *ea);
-struct ieee80211softmac_network * ieee80211softmac_get_network_by_bssid(
-	struct ieee80211softmac_device *mac, u8 *ea);
-struct ieee80211softmac_network * ieee80211softmac_get_network_by_ssid_locked(
-	struct ieee80211softmac_device *mac, u8 *ssid, u8 ssid_len);
-struct ieee80211softmac_network * ieee80211softmac_get_network_by_ssid(
-	struct ieee80211softmac_device *mac, u8 *ssid, u8 ssid_len);
-struct ieee80211softmac_network *
-ieee80211softmac_get_network_by_essid_locked(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_essid *essid);
-struct ieee80211softmac_network *
-ieee80211softmac_get_network_by_essid(struct ieee80211softmac_device *mac,
-	struct ieee80211softmac_essid *essid);
-
-/* Rates related */
-void ieee80211softmac_process_erp(struct ieee80211softmac_device *mac,
-	u8 erp_value);
-int ieee80211softmac_ratesinfo_rate_supported(struct ieee80211softmac_ratesinfo *ri, u8 rate);
-u8 ieee80211softmac_lower_rate_delta(struct ieee80211softmac_device *mac, u8 rate, int delta);
-void ieee80211softmac_init_bss(struct ieee80211softmac_device *mac);
-void ieee80211softmac_recalc_txrates(struct ieee80211softmac_device *mac);
-static inline u8 lower_rate(struct ieee80211softmac_device *mac, u8 rate) {
-	return ieee80211softmac_lower_rate_delta(mac, rate, 1);
-}
-
-static inline u8 get_fallback_rate(struct ieee80211softmac_device *mac, u8 rate)
-{
-	return ieee80211softmac_lower_rate_delta(mac, rate, 2);
-}
-
-
-/*** prototypes from _io.c */
-int ieee80211softmac_send_mgt_frame(struct ieee80211softmac_device *mac,
-	void* ptrarg, u32 type, u32 arg);
-int ieee80211softmac_handle_beacon(struct net_device *dev,
-	struct ieee80211_beacon *beacon,
-	struct ieee80211_network *network);
-
-/*** prototypes from _auth.c */
-/* do these have to go into the public header? */
-int ieee80211softmac_auth_req(struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net);
-int ieee80211softmac_deauth_req(struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net, int reason);
-
-/* for use by _module.c to assign to the callbacks */
-int ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *auth);
-int ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *deauth);
-
-/*** prototypes from _assoc.c */
-void ieee80211softmac_assoc_work(struct work_struct *work);
-int ieee80211softmac_handle_assoc_response(struct net_device * dev,
-					   struct ieee80211_assoc_response * resp,
-					   struct ieee80211_network * network);
-int ieee80211softmac_handle_disassoc(struct net_device * dev,
-				     struct ieee80211_disassoc * disassoc);
-int ieee80211softmac_handle_reassoc_req(struct net_device * dev,
-					struct ieee80211_reassoc_request * reassoc);
-void ieee80211softmac_assoc_timeout(struct work_struct *work);
-void ieee80211softmac_send_disassoc_req(struct ieee80211softmac_device *mac, u16 reason);
-void ieee80211softmac_disassoc(struct ieee80211softmac_device *mac);
-
-/* some helper functions */
-static inline int ieee80211softmac_scan_handlers_check_self(struct ieee80211softmac_device *sm)
-{
-	return (sm->start_scan == ieee80211softmac_start_scan_implementation) &&
-		(sm->stop_scan == ieee80211softmac_stop_scan_implementation) &&
-		(sm->wait_for_scan == ieee80211softmac_wait_for_scan_implementation);
-}
-
-static inline int ieee80211softmac_scan_sanity_check(struct ieee80211softmac_device *sm)
-{
-	return ((sm->start_scan != ieee80211softmac_start_scan_implementation) &&
-		(sm->stop_scan != ieee80211softmac_stop_scan_implementation) &&
-		(sm->wait_for_scan != ieee80211softmac_wait_for_scan_implementation)
-		) || ieee80211softmac_scan_handlers_check_self(sm);
-}
-
-#define IEEE80211SOFTMAC_PROBE_DELAY		HZ/50
-#define IEEE80211SOFTMAC_WORKQUEUE_NAME_LEN	(17 + IFNAMSIZ)
-
-struct ieee80211softmac_network {
-	struct list_head		list;	/* List */
-	/* Network information copied from ieee80211_network */
-	u8 bssid[ETH_ALEN];
-	u8 channel;
-	struct ieee80211softmac_essid essid;
-
-	struct ieee80211softmac_ratesinfo supported_rates;
-
-	/* SoftMAC specific */
-	u16 authenticating:1,			/* Status Flags */
-	    authenticated:1,
-	    auth_desynced_once:1;
-
-	u8 erp_value;				/* Saved ERP value */
-	u16 capabilities;			/* Capabilities bitfield */
-	u8 challenge_len;			/* Auth Challenge length */
-	char *challenge;			/* Challenge Text */
-};
-
-/* structure used to keep track of networks we're auth'ing to */
-struct ieee80211softmac_auth_queue_item {
-	struct list_head		list;	/* List head */
-	struct ieee80211softmac_network	*net;	/* Network to auth */
-	struct ieee80211softmac_device	*mac;	/* SoftMAC device */
-	u8 retry;				/* Retry limit */
-	u8 state;				/* Auth State */
-	struct delayed_work		work;	/* Work queue */
-};
-
-/* scanning information */
-struct ieee80211softmac_scaninfo {
-	u8 current_channel_idx,
-	   number_channels;
-	struct ieee80211_channel *channels;
-	u8 started:1,
-	   stop:1;
-	u8 skip_flags;
-	struct completion finished;
-	struct delayed_work softmac_scan;
-	struct ieee80211softmac_device *mac;
-};
-
-/* private event struct */
-struct ieee80211softmac_event {
-	struct list_head list;
-	int event_type;
-	void *event_context;
-	struct delayed_work work;
-	notify_function_ptr fun;
-	void *context;
-	struct ieee80211softmac_device *mac;
-};
-
-void ieee80211softmac_call_events(struct ieee80211softmac_device *mac, int event, void *event_context);
-void ieee80211softmac_call_events_locked(struct ieee80211softmac_device *mac, int event, void *event_context);
-int ieee80211softmac_notify_internal(struct ieee80211softmac_device *mac,
-	int event, void *event_context, notify_function_ptr fun, void *context, gfp_t gfp_mask);
-
-void ieee80211softmac_try_reassoc(struct ieee80211softmac_device *mac);
-
-#endif /* IEEE80211SOFTMAC_PRIV_H_ */
diff --git a/net/ieee80211/softmac/ieee80211softmac_scan.c b/net/ieee80211/softmac/ieee80211softmac_scan.c
deleted file mode 100644
index bfab8d7db88f..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_scan.c
+++ /dev/null
@@ -1,254 +0,0 @@
-/*
- * Scanning routines.
- *
- * These are not exported because they're assigned to the function pointers.
- *
- * Copyright (c) 2005, 2006 Johannes Berg <johannes@sipsolutions.net>
- *                          Joseph Jezak <josejx@gentoo.org>
- *                          Larry Finger <Larry.Finger@lwfinger.net>
- *                          Danny van Dyk <kugelfang@gentoo.org>
- *                          Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#include <linux/completion.h>
-#include "ieee80211softmac_priv.h"
-
-/* internal, use to trigger scanning if needed.
- * Returns -EBUSY if already scanning,
- * result of start_scan otherwise */
-int
-ieee80211softmac_start_scan(struct ieee80211softmac_device *sm)
-{
-	unsigned long flags;
-	int ret;
-
-	spin_lock_irqsave(&sm->lock, flags);
-	if (sm->scanning)
-	{
-		spin_unlock_irqrestore(&sm->lock, flags);
-		return -EINPROGRESS;
-	}
-	sm->scanning = 1;
-	spin_unlock_irqrestore(&sm->lock, flags);
-
-	ret = sm->start_scan(sm->dev);
-	if (ret) {
-		spin_lock_irqsave(&sm->lock, flags);
-		sm->scanning = 0;
-		spin_unlock_irqrestore(&sm->lock, flags);
-	}
-	return ret;
-}
-
-void
-ieee80211softmac_stop_scan(struct ieee80211softmac_device *sm)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&sm->lock, flags);
-
-	if (!sm->scanning) {
-		spin_unlock_irqrestore(&sm->lock, flags);
-		return;
-	}
-
-	spin_unlock_irqrestore(&sm->lock, flags);
-	sm->stop_scan(sm->dev);
-}
-
-void
-ieee80211softmac_wait_for_scan(struct ieee80211softmac_device *sm)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&sm->lock, flags);
-
-	if (!sm->scanning) {
-		spin_unlock_irqrestore(&sm->lock, flags);
-		return;
-	}
-
-	spin_unlock_irqrestore(&sm->lock, flags);
-	sm->wait_for_scan(sm->dev);
-}
-
-
-/* internal scanning implementation follows */
-void ieee80211softmac_scan(struct work_struct *work)
-{
-	int invalid_channel;
-	u8 current_channel_idx;
-	struct ieee80211softmac_scaninfo *si =
-		container_of(work, struct ieee80211softmac_scaninfo,
-			     softmac_scan.work);
-	struct ieee80211softmac_device *sm = si->mac;
-	unsigned long flags;
-
-	while (!(si->stop) && (si->current_channel_idx < si->number_channels)) {
-		current_channel_idx = si->current_channel_idx;
-		si->current_channel_idx++; /* go to the next channel */
-
-		invalid_channel = (si->skip_flags & si->channels[current_channel_idx].flags);
-
-		if (!invalid_channel) {
-			sm->set_channel(sm->dev, si->channels[current_channel_idx].channel);
-			// FIXME make this user configurable (active/passive)
-			if(ieee80211softmac_send_mgt_frame(sm, NULL, IEEE80211_STYPE_PROBE_REQ, 0))
-				printkl(KERN_DEBUG PFX "Sending Probe Request Failed\n");
-
-			/* also send directed management frame for the network we're looking for */
-			// TODO: is this if correct, or should we do this only if scanning from assoc request?
-			if (sm->associnfo.req_essid.len)
-				ieee80211softmac_send_mgt_frame(sm, &sm->associnfo.req_essid, IEEE80211_STYPE_PROBE_REQ, 0);
-
-			spin_lock_irqsave(&sm->lock, flags);
-			if (unlikely(!sm->running)) {
-				/* Prevent reschedule on workqueue flush */
-				spin_unlock_irqrestore(&sm->lock, flags);
-				break;
-			}
-			queue_delayed_work(sm->wq, &si->softmac_scan, IEEE80211SOFTMAC_PROBE_DELAY);
-			spin_unlock_irqrestore(&sm->lock, flags);
-			return;
-		} else {
-			dprintk(PFX "Not probing Channel %d (not allowed here)\n", si->channels[current_channel_idx].channel);
-		}
-	}
-
-	spin_lock_irqsave(&sm->lock, flags);
-	cancel_delayed_work(&si->softmac_scan);
-	si->started = 0;
-	spin_unlock_irqrestore(&sm->lock, flags);
-
-	dprintk(PFX "Scanning finished: scanned %d channels starting with channel %d\n",
-		     sm->scaninfo->number_channels, sm->scaninfo->channels[0].channel);
-	ieee80211softmac_scan_finished(sm);
-	complete_all(&sm->scaninfo->finished);
-}
-
-static inline struct ieee80211softmac_scaninfo *allocate_scaninfo(struct ieee80211softmac_device *mac)
-{
-	/* ugh. can we call this without having the spinlock held? */
-	struct ieee80211softmac_scaninfo *info = kmalloc(sizeof(struct ieee80211softmac_scaninfo), GFP_ATOMIC);
-	if (unlikely(!info))
-		return NULL;
-	INIT_DELAYED_WORK(&info->softmac_scan, ieee80211softmac_scan);
-	info->mac = mac;
-	init_completion(&info->finished);
-	return info;
-}
-
-int ieee80211softmac_start_scan_implementation(struct net_device *dev)
-{
-	struct ieee80211softmac_device *sm = ieee80211_priv(dev);
-	unsigned long flags;
-
-	if (!(dev->flags & IFF_UP))
-		return -ENODEV;
-
-	assert(ieee80211softmac_scan_handlers_check_self(sm));
-	if (!ieee80211softmac_scan_handlers_check_self(sm))
-		return -EINVAL;
-
-	spin_lock_irqsave(&sm->lock, flags);
-	/* it looks like we need to hold the lock here
-	 * to make sure we don't allocate two of these... */
-	if (unlikely(!sm->scaninfo))
-		sm->scaninfo = allocate_scaninfo(sm);
-	if (unlikely(!sm->scaninfo)) {
-		spin_unlock_irqrestore(&sm->lock, flags);
-		return -ENOMEM;
-	}
-
-	sm->scaninfo->skip_flags = IEEE80211_CH_INVALID;
-	if (0 /* not scanning in IEEE802.11b */)//TODO
-		sm->scaninfo->skip_flags |= IEEE80211_CH_B_ONLY;
-	if (0 /* IEEE802.11a */) {//TODO
-		sm->scaninfo->channels = sm->ieee->geo.a;
-		sm->scaninfo->number_channels = sm->ieee->geo.a_channels;
-	} else {
-		sm->scaninfo->channels = sm->ieee->geo.bg;
-		sm->scaninfo->number_channels = sm->ieee->geo.bg_channels;
-	}
-	sm->scaninfo->current_channel_idx = 0;
-	sm->scaninfo->started = 1;
-	sm->scaninfo->stop = 0;
-	INIT_COMPLETION(sm->scaninfo->finished);
-	queue_delayed_work(sm->wq, &sm->scaninfo->softmac_scan, 0);
-	spin_unlock_irqrestore(&sm->lock, flags);
-	return 0;
-}
-
-void ieee80211softmac_stop_scan_implementation(struct net_device *dev)
-{
-	struct ieee80211softmac_device *sm = ieee80211_priv(dev);
-	unsigned long flags;
-
-	assert(ieee80211softmac_scan_handlers_check_self(sm));
-	if (!ieee80211softmac_scan_handlers_check_self(sm))
-		return;
-
-	spin_lock_irqsave(&sm->lock, flags);
-	assert(sm->scaninfo != NULL);
-	if (sm->scaninfo) {
-		if (sm->scaninfo->started)
-			sm->scaninfo->stop = 1;
-		else
-			complete_all(&sm->scaninfo->finished);
-	}
-	spin_unlock_irqrestore(&sm->lock, flags);
-}
-
-void ieee80211softmac_wait_for_scan_implementation(struct net_device *dev)
-{
-	struct ieee80211softmac_device *sm = ieee80211_priv(dev);
-	unsigned long flags;
-
-	assert(ieee80211softmac_scan_handlers_check_self(sm));
-	if (!ieee80211softmac_scan_handlers_check_self(sm))
-		return;
-
-	spin_lock_irqsave(&sm->lock, flags);
-	if (!sm->scaninfo->started) {
-		spin_unlock_irqrestore(&sm->lock, flags);
-		return;
-	}
-	spin_unlock_irqrestore(&sm->lock, flags);
-	wait_for_completion(&sm->scaninfo->finished);
-}
-
-/* this is what drivers (that do scanning) call when they're done */
-void ieee80211softmac_scan_finished(struct ieee80211softmac_device *sm)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&sm->lock, flags);
-	sm->scanning = 0;
-	spin_unlock_irqrestore(&sm->lock, flags);
-
-	if (sm->associnfo.bssvalid) {
-		struct ieee80211softmac_network *net;
-
-		net = ieee80211softmac_get_network_by_bssid(sm, sm->associnfo.bssid);
-		if (net)
-			sm->set_channel(sm->dev, net->channel);
-	}
-	ieee80211softmac_call_events(sm, IEEE80211SOFTMAC_EVENT_SCAN_FINISHED, NULL);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_scan_finished);
diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c
deleted file mode 100644
index e01b59aedc54..000000000000
--- a/net/ieee80211/softmac/ieee80211softmac_wx.c
+++ /dev/null
@@ -1,508 +0,0 @@
-/*
- * This file contains our _wx handlers. Make sure you EXPORT_SYMBOL_GPL them
- *
- * Copyright (c) 2005, 2006 Johannes Berg <johannes@sipsolutions.net>
- *                          Joseph Jezak <josejx@gentoo.org>
- *                          Larry Finger <Larry.Finger@lwfinger.net>
- *                          Danny van Dyk <kugelfang@gentoo.org>
- *                          Michael Buesch <mbuesch@freenet.de>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
- *
- * The full GNU General Public License is included in this distribution in the
- * file called COPYING.
- */
-
-#include "ieee80211softmac_priv.h"
-
-#include <net/iw_handler.h>
-/* for is_broadcast_ether_addr and is_zero_ether_addr */
-#include <linux/etherdevice.h>
-
-int
-ieee80211softmac_wx_trigger_scan(struct net_device *net_dev,
-				 struct iw_request_info *info,
-				 union iwreq_data *data,
-				 char *extra)
-{
-	struct ieee80211softmac_device *sm = ieee80211_priv(net_dev);
-	return ieee80211softmac_start_scan(sm);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_trigger_scan);
-
-
-/* if we're still scanning, return -EAGAIN so that userspace tools
- * can get the complete scan results, otherwise return 0. */
-int
-ieee80211softmac_wx_get_scan_results(struct net_device *net_dev,
-				     struct iw_request_info *info,
-				     union iwreq_data *data,
-				     char *extra)
-{
-	unsigned long flags;
-	struct ieee80211softmac_device *sm = ieee80211_priv(net_dev);
-
-	spin_lock_irqsave(&sm->lock, flags);
-	if (sm->scanning) {
-		spin_unlock_irqrestore(&sm->lock, flags);
-		return -EAGAIN;
-	}
-	spin_unlock_irqrestore(&sm->lock, flags);
-	return ieee80211_wx_get_scan(sm->ieee, info, data, extra);
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_scan_results);
-
-int
-ieee80211softmac_wx_set_essid(struct net_device *net_dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *data,
-			      char *extra)
-{
-	struct ieee80211softmac_device *sm = ieee80211_priv(net_dev);
-	struct ieee80211softmac_auth_queue_item *authptr;
-	int length = 0;
-	DECLARE_MAC_BUF(mac);
-
-check_assoc_again:
-	mutex_lock(&sm->associnfo.mutex);
-	if((sm->associnfo.associating || sm->associnfo.associated) &&
-	   (data->essid.flags && data->essid.length)) {
-		dprintk(KERN_INFO PFX "Canceling existing associate request!\n");
-		/* Cancel assoc work */
-		cancel_delayed_work(&sm->associnfo.work);
-		/* We don't have to do this, but it's a little cleaner */
-		list_for_each_entry(authptr, &sm->auth_queue, list)
-			cancel_delayed_work(&authptr->work);
-		sm->associnfo.bssvalid = 0;
-		sm->associnfo.bssfixed = 0;
-		sm->associnfo.associating = 0;
-		sm->associnfo.associated = 0;
-		/* We must unlock to avoid deadlocks with the assoc workqueue
-		 * on the associnfo.mutex */
-		mutex_unlock(&sm->associnfo.mutex);
-		flush_workqueue(sm->wq);
-		/* Avoid race! Check assoc status again. Maybe someone started an
-		 * association while we flushed. */
-		goto check_assoc_again;
-	}
-
-	sm->associnfo.static_essid = 0;
-	sm->associnfo.assoc_wait = 0;
-
-	if (data->essid.flags && data->essid.length) {
-		length = min((int)data->essid.length, IW_ESSID_MAX_SIZE);
-		if (length) {
-			memcpy(sm->associnfo.req_essid.data, extra, length);
-			sm->associnfo.static_essid = 1;
-		}
-	}
-
-	/* set our requested ESSID length.
-	 * If applicable, we have already copied the data in */
-	sm->associnfo.req_essid.len = length;
-
-	sm->associnfo.associating = 1;
-	/* queue lower level code to do work (if necessary) */
-	queue_delayed_work(sm->wq, &sm->associnfo.work, 0);
-
-	mutex_unlock(&sm->associnfo.mutex);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_essid);
-
-int
-ieee80211softmac_wx_get_essid(struct net_device *net_dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *data,
-			      char *extra)
-{
-	struct ieee80211softmac_device *sm = ieee80211_priv(net_dev);
-
-	mutex_lock(&sm->associnfo.mutex);
-	/* If all fails, return ANY (empty) */
-	data->essid.length = 0;
-	data->essid.flags = 0;  /* active */
-
-	/* If we have a statically configured ESSID then return it */
-	if (sm->associnfo.static_essid) {
-		data->essid.length = sm->associnfo.req_essid.len;
-		data->essid.flags = 1;  /* active */
-		memcpy(extra, sm->associnfo.req_essid.data, sm->associnfo.req_essid.len);
-		dprintk(KERN_INFO PFX "Getting essid from req_essid\n");
-	} else if (sm->associnfo.associated || sm->associnfo.associating) {
-	/* If we're associating/associated, return that */
-		data->essid.length = sm->associnfo.associate_essid.len;
-		data->essid.flags = 1;  /* active */
-		memcpy(extra, sm->associnfo.associate_essid.data, sm->associnfo.associate_essid.len);
-		dprintk(KERN_INFO PFX "Getting essid from associate_essid\n");
-	}
-	mutex_unlock(&sm->associnfo.mutex);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_essid);
-
-int
-ieee80211softmac_wx_set_rate(struct net_device *net_dev,
-			     struct iw_request_info *info,
-			     union iwreq_data *data,
-			     char *extra)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(net_dev);
-	struct ieee80211_device *ieee = mac->ieee;
-	unsigned long flags;
-	s32 in_rate = data->bitrate.value;
-	u8 rate;
-	int is_ofdm = 0;
-	int err = -EINVAL;
-
-	if (in_rate == -1) {
-		if (ieee->modulation & IEEE80211_OFDM_MODULATION)
-			in_rate = 24000000;
-		else
-			in_rate = 11000000;
-	}
-
-	switch (in_rate) {
-	case 1000000:
-		rate = IEEE80211_CCK_RATE_1MB;
-		break;
-	case 2000000:
-		rate = IEEE80211_CCK_RATE_2MB;
-		break;
-	case 5500000:
-		rate = IEEE80211_CCK_RATE_5MB;
-		break;
-	case 11000000:
-		rate = IEEE80211_CCK_RATE_11MB;
-		break;
-	case 6000000:
-		rate = IEEE80211_OFDM_RATE_6MB;
-		is_ofdm = 1;
-		break;
-	case 9000000:
-		rate = IEEE80211_OFDM_RATE_9MB;
-		is_ofdm = 1;
-		break;
-	case 12000000:
-		rate = IEEE80211_OFDM_RATE_12MB;
-		is_ofdm = 1;
-		break;
-	case 18000000:
-		rate = IEEE80211_OFDM_RATE_18MB;
-		is_ofdm = 1;
-		break;
-	case 24000000:
-		rate = IEEE80211_OFDM_RATE_24MB;
-		is_ofdm = 1;
-		break;
-	case 36000000:
-		rate = IEEE80211_OFDM_RATE_36MB;
-		is_ofdm = 1;
-		break;
-	case 48000000:
-		rate = IEEE80211_OFDM_RATE_48MB;
-		is_ofdm = 1;
-		break;
-	case 54000000:
-		rate = IEEE80211_OFDM_RATE_54MB;
-		is_ofdm = 1;
-		break;
-	default:
-		goto out;
-	}
-
-	spin_lock_irqsave(&mac->lock, flags);
-
-	/* Check if correct modulation for this PHY. */
-	if (is_ofdm && !(ieee->modulation & IEEE80211_OFDM_MODULATION))
-		goto out_unlock;
-
-	mac->txrates.user_rate = rate;
-	ieee80211softmac_recalc_txrates(mac);
-	err = 0;
-
-out_unlock:
-	spin_unlock_irqrestore(&mac->lock, flags);
-out:
-	return err;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_rate);
-
-int
-ieee80211softmac_wx_get_rate(struct net_device *net_dev,
-			     struct iw_request_info *info,
-			     union iwreq_data *data,
-			     char *extra)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(net_dev);
-	unsigned long flags;
-	int err = -EINVAL;
-
-	spin_lock_irqsave(&mac->lock, flags);
-
-	if (unlikely(!mac->running)) {
-		err = -ENODEV;
-		goto out_unlock;
-	}
-
-	switch (mac->txrates.default_rate) {
-	case IEEE80211_CCK_RATE_1MB:
-		data->bitrate.value = 1000000;
-		break;
-	case IEEE80211_CCK_RATE_2MB:
-		data->bitrate.value = 2000000;
-		break;
-	case IEEE80211_CCK_RATE_5MB:
-		data->bitrate.value = 5500000;
-		break;
-	case IEEE80211_CCK_RATE_11MB:
-		data->bitrate.value = 11000000;
-		break;
-	case IEEE80211_OFDM_RATE_6MB:
-		data->bitrate.value = 6000000;
-		break;
-	case IEEE80211_OFDM_RATE_9MB:
-		data->bitrate.value = 9000000;
-		break;
-	case IEEE80211_OFDM_RATE_12MB:
-		data->bitrate.value = 12000000;
-		break;
-	case IEEE80211_OFDM_RATE_18MB:
-		data->bitrate.value = 18000000;
-		break;
-	case IEEE80211_OFDM_RATE_24MB:
-		data->bitrate.value = 24000000;
-		break;
-	case IEEE80211_OFDM_RATE_36MB:
-		data->bitrate.value = 36000000;
-		break;
-	case IEEE80211_OFDM_RATE_48MB:
-		data->bitrate.value = 48000000;
-		break;
-	case IEEE80211_OFDM_RATE_54MB:
-		data->bitrate.value = 54000000;
-		break;
-	default:
-		assert(0);
-		goto out_unlock;
-	}
-	err = 0;
-out_unlock:
-	spin_unlock_irqrestore(&mac->lock, flags);
-
-	return err;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_rate);
-
-int
-ieee80211softmac_wx_get_wap(struct net_device *net_dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *data,
-			    char *extra)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(net_dev);
-	int err = 0;
-
-	mutex_lock(&mac->associnfo.mutex);
-	if (mac->associnfo.bssvalid)
-		memcpy(data->ap_addr.sa_data, mac->associnfo.bssid, ETH_ALEN);
-	else
-		memset(data->ap_addr.sa_data, 0xff, ETH_ALEN);
-	data->ap_addr.sa_family = ARPHRD_ETHER;
-	mutex_unlock(&mac->associnfo.mutex);
-
-	return err;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_wap);
-
-int
-ieee80211softmac_wx_set_wap(struct net_device *net_dev,
-			    struct iw_request_info *info,
-			    union iwreq_data *data,
-			    char *extra)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(net_dev);
-
-	/* sanity check */
-	if (data->ap_addr.sa_family != ARPHRD_ETHER) {
-		return -EINVAL;
-	}
-
-	mutex_lock(&mac->associnfo.mutex);
-	if (is_broadcast_ether_addr(data->ap_addr.sa_data)) {
-		/* the bssid we have is not to be fixed any longer,
-		 * and we should reassociate to the best AP. */
-		mac->associnfo.bssfixed = 0;
-		/* force reassociation */
-		mac->associnfo.bssvalid = 0;
-		if (mac->associnfo.associated)
-			queue_delayed_work(mac->wq, &mac->associnfo.work, 0);
-	} else if (is_zero_ether_addr(data->ap_addr.sa_data)) {
-		/* the bssid we have is no longer fixed */
-		mac->associnfo.bssfixed = 0;
-	} else {
-		if (!memcmp(mac->associnfo.bssid, data->ap_addr.sa_data, ETH_ALEN)) {
-			if (mac->associnfo.associating || mac->associnfo.associated) {
-			/* bssid unchanged and associated or associating - just return */
-				goto out;
-			}
-		} else {
-			/* copy new value in data->ap_addr.sa_data to bssid */
-			memcpy(mac->associnfo.bssid, data->ap_addr.sa_data, ETH_ALEN);
-		}
-		/* tell the other code that this bssid should be used no matter what */
-		mac->associnfo.bssfixed = 1;
-		/* queue associate if new bssid or (old one again and not associated) */
-		queue_delayed_work(mac->wq, &mac->associnfo.work, 0);
-	}
-
- out:
-	mutex_unlock(&mac->associnfo.mutex);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_wap);
-
-int
-ieee80211softmac_wx_set_genie(struct net_device *dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *wrqu,
-			      char *extra)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	unsigned long flags;
-	int err = 0;
-	char *buf;
-	int i;
-
-	mutex_lock(&mac->associnfo.mutex);
-	spin_lock_irqsave(&mac->lock, flags);
-	/* bleh. shouldn't be locked for that kmalloc... */
-
-	if (wrqu->data.length) {
-		if ((wrqu->data.length < 2) || (extra[1]+2 != wrqu->data.length)) {
-			/* this is an IE, so the length must be
-			 * correct. Is it possible though that
-			 * more than one IE is passed in?
-			 */
-			err = -EINVAL;
-			goto out;
-		}
-		if (mac->wpa.IEbuflen <= wrqu->data.length) {
-			buf = kmalloc(wrqu->data.length, GFP_ATOMIC);
-			if (!buf) {
-				err = -ENOMEM;
-				goto out;
-			}
-			kfree(mac->wpa.IE);
-			mac->wpa.IE = buf;
-			mac->wpa.IEbuflen = wrqu->data.length;
-		}
-		memcpy(mac->wpa.IE, extra, wrqu->data.length);
-		dprintk(KERN_INFO PFX "generic IE set to ");
-		for (i=0;i<wrqu->data.length;i++)
-			dprintk("%.2x", (u8)mac->wpa.IE[i]);
-		dprintk("\n");
-		mac->wpa.IElen = wrqu->data.length;
-	} else {
-		kfree(mac->wpa.IE);
-		mac->wpa.IE = NULL;
-		mac->wpa.IElen = 0;
-		mac->wpa.IEbuflen = 0;
-	}
-
- out:
-	spin_unlock_irqrestore(&mac->lock, flags);
-	mutex_unlock(&mac->associnfo.mutex);
-
-	return err;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_genie);
-
-int
-ieee80211softmac_wx_get_genie(struct net_device *dev,
-			      struct iw_request_info *info,
-			      union iwreq_data *wrqu,
-			      char *extra)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	unsigned long flags;
-	int err = 0;
-	int space = wrqu->data.length;
-
-	mutex_lock(&mac->associnfo.mutex);
-	spin_lock_irqsave(&mac->lock, flags);
-
-	wrqu->data.length = 0;
-
-	if (mac->wpa.IE && mac->wpa.IElen) {
-		wrqu->data.length = mac->wpa.IElen;
-		if (mac->wpa.IElen <= space)
-			memcpy(extra, mac->wpa.IE, mac->wpa.IElen);
-		else
-			err = -E2BIG;
-	}
-	spin_unlock_irqrestore(&mac->lock, flags);
-	mutex_unlock(&mac->associnfo.mutex);
-
-	return err;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_genie);
-
-int
-ieee80211softmac_wx_set_mlme(struct net_device *dev,
-			     struct iw_request_info *info,
-			     union iwreq_data *wrqu,
-			     char *extra)
-{
-	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
-	struct iw_mlme *mlme = (struct iw_mlme *)extra;
-	u16 reason = mlme->reason_code;
-	struct ieee80211softmac_network *net;
-	int err = -EINVAL;
-
-	mutex_lock(&mac->associnfo.mutex);
-
-	if (memcmp(mac->associnfo.bssid, mlme->addr.sa_data, ETH_ALEN)) {
-		printk(KERN_DEBUG PFX "wx_set_mlme: requested operation on net we don't use\n");
-		goto out;
-	}
-
-	switch (mlme->cmd) {
-	case IW_MLME_DEAUTH:
-		net = ieee80211softmac_get_network_by_bssid_locked(mac, mlme->addr.sa_data);
-		if (!net) {
-			printk(KERN_DEBUG PFX "wx_set_mlme: we should know the net here...\n");
-			goto out;
-		}
-		err =  ieee80211softmac_deauth_req(mac, net, reason);
-		goto out;
-	case IW_MLME_DISASSOC:
-		ieee80211softmac_send_disassoc_req(mac, reason);
-		mac->associnfo.associated = 0;
-		mac->associnfo.associating = 0;
-		err = 0;
-		goto out;
-	default:
-		err = -EOPNOTSUPP;
-	}
-
-out:
-	mutex_unlock(&mac->associnfo.mutex);
-
-	return err;
-}
-EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_mlme);
-- 
cgit v1.2.3


From 068edceb7e73c05f77e204442ea8f86e238575da Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 9 Mar 2008 16:55:10 -0700
Subject: include/net/ieee80211.h - remove duplicate include

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index 285b2adfa648..529816bfbc52 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -183,7 +183,6 @@ const char *escape_essid(const char *essid, u8 essid_len);
 #define IEEE80211_DEBUG_RX(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_RX, f, ## a)
 #define IEEE80211_DEBUG_QOS(f, a...)  IEEE80211_DEBUG(IEEE80211_DL_QOS, f, ## a)
 #include <linux/netdevice.h>
-#include <linux/wireless.h>
 #include <linux/if_arp.h>	/* ARPHRD_ETHER */
 
 #ifndef WIRELESS_SPY
-- 
cgit v1.2.3


From e7ec2e3230633a858af1b0b359f6c4670dbeb997 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Mon, 10 Mar 2008 17:26:32 +0100
Subject: ssb: Add SPROM/invariants support for PCMCIA devices

This adds support for reading/writing the SPROM invariants
for PCMCIA based devices.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/Kconfig       |   6 +
 drivers/ssb/Makefile      |   1 +
 drivers/ssb/main.c        |  23 +-
 drivers/ssb/pci.c         | 113 +---------
 drivers/ssb/pcmcia.c      | 518 ++++++++++++++++++++++++++++++++++++++++------
 drivers/ssb/sprom.c       | 133 ++++++++++++
 drivers/ssb/ssb_private.h |  17 ++
 include/linux/ssb/ssb.h   |   4 +-
 8 files changed, 648 insertions(+), 167 deletions(-)
 create mode 100644 drivers/ssb/sprom.c

(limited to 'include')

diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig
index f69ef0ba2613..0f7cce2560d1 100644
--- a/drivers/ssb/Kconfig
+++ b/drivers/ssb/Kconfig
@@ -20,6 +20,10 @@ config SSB
 
 	  If unsure, say N.
 
+# Common SPROM support routines
+config SSB_SPROM
+	bool
+
 config SSB_PCIHOST_POSSIBLE
 	bool
 	depends on SSB && (PCI = y || PCI = SSB)
@@ -28,6 +32,7 @@ config SSB_PCIHOST_POSSIBLE
 config SSB_PCIHOST
 	bool "Support for SSB on PCI-bus host"
 	depends on SSB_PCIHOST_POSSIBLE
+	select SSB_SPROM
 	default y
 	help
 	  Support for a Sonics Silicon Backplane on top
@@ -48,6 +53,7 @@ config SSB_PCMCIAHOST_POSSIBLE
 config SSB_PCMCIAHOST
 	bool "Support for SSB on PCMCIA-bus host (EXPERIMENTAL)"
 	depends on SSB_PCMCIAHOST_POSSIBLE
+	select SSB_SPROM
 	help
 	  Support for a Sonics Silicon Backplane on top
 	  of a PCMCIA device.
diff --git a/drivers/ssb/Makefile b/drivers/ssb/Makefile
index 910f35e32fc9..6f255e9c5af9 100644
--- a/drivers/ssb/Makefile
+++ b/drivers/ssb/Makefile
@@ -1,6 +1,7 @@
 # core
 ssb-y					+= main.o scan.o
 ssb-$(CONFIG_SSB_EMBEDDED)		+= embedded.o
+ssb-$(CONFIG_SSB_SPROM)			+= sprom.o
 
 # host support
 ssb-$(CONFIG_SSB_PCIHOST)		+= pci.o pcihost_wrapper.o
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index 49d7bbb9bea7..e12371916444 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -69,6 +69,25 @@ found:
 }
 #endif /* CONFIG_SSB_PCIHOST */
 
+#ifdef CONFIG_SSB_PCMCIAHOST
+struct ssb_bus *ssb_pcmcia_dev_to_bus(struct pcmcia_device *pdev)
+{
+	struct ssb_bus *bus;
+
+	ssb_buses_lock();
+	list_for_each_entry(bus, &buses, list) {
+		if (bus->bustype == SSB_BUSTYPE_PCMCIA &&
+		    bus->host_pcmcia == pdev)
+			goto found;
+	}
+	bus = NULL;
+found:
+	ssb_buses_unlock();
+
+	return bus;
+}
+#endif /* CONFIG_SSB_PCMCIAHOST */
+
 int ssb_for_each_bus_call(unsigned long data,
 			  int (*func)(struct ssb_bus *bus, unsigned long data))
 {
@@ -398,7 +417,7 @@ void ssb_bus_unregister(struct ssb_bus *bus)
 	list_del(&bus->list);
 	ssb_buses_unlock();
 
-	/* ssb_pcmcia_exit(bus); */
+	ssb_pcmcia_exit(bus);
 	ssb_pci_exit(bus);
 	ssb_iounmap(bus);
 }
@@ -663,7 +682,7 @@ out:
 err_dequeue:
 	list_del(&bus->list);
 err_pcmcia_exit:
-/*	ssb_pcmcia_exit(bus); */
+	ssb_pcmcia_exit(bus);
 err_pci_exit:
 	ssb_pci_exit(bus);
 err_unmap:
diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index 1facc7620fc8..f1514b33cfae 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -227,7 +227,7 @@ static u8 ssb_sprom_crc(const u16 *sprom, u16 size)
 	return crc;
 }
 
-static int sprom_check_crc(const u16 *sprom, u16 size)
+static int sprom_check_crc(const u16 *sprom, size_t size)
 {
 	u8 crc;
 	u8 expected_crc;
@@ -242,12 +242,14 @@ static int sprom_check_crc(const u16 *sprom, u16 size)
 	return 0;
 }
 
-static void sprom_do_read(struct ssb_bus *bus, u16 *sprom)
+static int sprom_do_read(struct ssb_bus *bus, u16 *sprom)
 {
 	int i;
 
 	for (i = 0; i < bus->sprom_size; i++)
 		sprom[i] = ioread16(bus->mmio + SSB_SPROM_BASE + (i * 2));
+
+	return 0;
 }
 
 static int sprom_do_write(struct ssb_bus *bus, const u16 *sprom)
@@ -660,71 +662,18 @@ const struct ssb_bus_ops ssb_pci_ops = {
 	.write32	= ssb_pci_write32,
 };
 
-static int sprom2hex(const u16 *sprom, char *buf, size_t buf_len, u16 size)
-{
-	int i, pos = 0;
-
-	for (i = 0; i < size; i++)
-		pos += snprintf(buf + pos, buf_len - pos - 1,
-				"%04X", swab16(sprom[i]) & 0xFFFF);
-	pos += snprintf(buf + pos, buf_len - pos - 1, "\n");
-
-	return pos + 1;
-}
-
-static int hex2sprom(u16 *sprom, const char *dump, size_t len, u16 size)
-{
-	char tmp[5] = { 0 };
-	int cnt = 0;
-	unsigned long parsed;
-
-	if (len < size * 2)
-		return -EINVAL;
-
-	while (cnt < size) {
-		memcpy(tmp, dump, 4);
-		dump += 4;
-		parsed = simple_strtoul(tmp, NULL, 16);
-		sprom[cnt++] = swab16((u16)parsed);
-	}
-
-	return 0;
-}
-
 static ssize_t ssb_pci_attr_sprom_show(struct device *pcidev,
 				       struct device_attribute *attr,
 				       char *buf)
 {
 	struct pci_dev *pdev = container_of(pcidev, struct pci_dev, dev);
 	struct ssb_bus *bus;
-	u16 *sprom;
-	int err = -ENODEV;
-	ssize_t count = 0;
 
 	bus = ssb_pci_dev_to_bus(pdev);
 	if (!bus)
-		goto out;
-	err = -ENOMEM;
-	sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL);
-	if (!sprom)
-		goto out;
+		return -ENODEV;
 
-	/* Use interruptible locking, as the SPROM write might
-	 * be holding the lock for several seconds. So allow userspace
-	 * to cancel operation. */
-	err = -ERESTARTSYS;
-	if (mutex_lock_interruptible(&bus->pci_sprom_mutex))
-		goto out_kfree;
-	sprom_do_read(bus, sprom);
-	mutex_unlock(&bus->pci_sprom_mutex);
-
-	count = sprom2hex(sprom, buf, PAGE_SIZE, bus->sprom_size);
-	err = 0;
-
-out_kfree:
-	kfree(sprom);
-out:
-	return err ? err : count;
+	return ssb_attr_sprom_show(bus, buf, sprom_do_read);
 }
 
 static ssize_t ssb_pci_attr_sprom_store(struct device *pcidev,
@@ -733,55 +682,13 @@ static ssize_t ssb_pci_attr_sprom_store(struct device *pcidev,
 {
 	struct pci_dev *pdev = container_of(pcidev, struct pci_dev, dev);
 	struct ssb_bus *bus;
-	u16 *sprom;
-	int res = 0, err = -ENODEV;
 
 	bus = ssb_pci_dev_to_bus(pdev);
 	if (!bus)
-		goto out;
-	err = -ENOMEM;
-	sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL);
-	if (!sprom)
-		goto out;
-	err = hex2sprom(sprom, buf, count, bus->sprom_size);
-	if (err) {
-		err = -EINVAL;
-		goto out_kfree;
-	}
-	err = sprom_check_crc(sprom, bus->sprom_size);
-	if (err) {
-		err = -EINVAL;
-		goto out_kfree;
-	}
+		return -ENODEV;
 
-	/* Use interruptible locking, as the SPROM write might
-	 * be holding the lock for several seconds. So allow userspace
-	 * to cancel operation. */
-	err = -ERESTARTSYS;
-	if (mutex_lock_interruptible(&bus->pci_sprom_mutex))
-		goto out_kfree;
-	err = ssb_devices_freeze(bus);
-	if (err == -EOPNOTSUPP) {
-		ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze devices. "
-			   "No suspend support. Is CONFIG_PM enabled?\n");
-		goto out_unlock;
-	}
-	if (err) {
-		ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze all devices\n");
-		goto out_unlock;
-	}
-	res = sprom_do_write(bus, sprom);
-	err = ssb_devices_thaw(bus);
-	if (err)
-		ssb_printk(KERN_ERR PFX "SPROM write: Could not thaw all devices\n");
-out_unlock:
-	mutex_unlock(&bus->pci_sprom_mutex);
-out_kfree:
-	kfree(sprom);
-out:
-	if (res)
-		return res;
-	return err ? err : count;
+	return ssb_attr_sprom_store(bus, buf, count,
+				    sprom_check_crc, sprom_do_write);
 }
 
 static DEVICE_ATTR(ssb_sprom, 0600,
@@ -808,7 +715,7 @@ int ssb_pci_init(struct ssb_bus *bus)
 		return 0;
 
 	pdev = bus->host_pci;
-	mutex_init(&bus->pci_sprom_mutex);
+	mutex_init(&bus->sprom_mutex);
 	err = device_create_file(&pdev->dev, &dev_attr_ssb_sprom);
 	if (err)
 		goto out;
diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c
index 84b3a845a8a8..cd49f7c65531 100644
--- a/drivers/ssb/pcmcia.c
+++ b/drivers/ssb/pcmcia.c
@@ -3,7 +3,7 @@
  * PCMCIA-Hostbus related functions
  *
  * Copyright 2006 Johannes Berg <johannes@sipsolutions.net>
- * Copyright 2007 Michael Buesch <mb@bu3sch.de>
+ * Copyright 2007-2008 Michael Buesch <mb@bu3sch.de>
  *
  * Licensed under the GNU/GPL. See COPYING for details.
  */
@@ -11,6 +11,7 @@
 #include <linux/ssb/ssb.h>
 #include <linux/delay.h>
 #include <linux/io.h>
+#include <linux/etherdevice.h>
 
 #include <pcmcia/cs_types.h>
 #include <pcmcia/cs.h>
@@ -26,59 +27,132 @@
 #define SSB_VERBOSE_PCMCIACORESWITCH_DEBUG		0
 
 
+/* PCMCIA configuration registers */
+#define SSB_PCMCIA_CORECTL		0x00
+#define  SSB_PCMCIA_CORECTL_RESET	0x80 /* Core reset */
+#define  SSB_PCMCIA_CORECTL_IRQEN	0x04 /* IRQ enable */
+#define  SSB_PCMCIA_CORECTL_FUNCEN	0x01 /* Function enable */
+#define SSB_PCMCIA_CORECTL2		0x80
+#define SSB_PCMCIA_ADDRESS0		0x2E
+#define SSB_PCMCIA_ADDRESS1		0x30
+#define SSB_PCMCIA_ADDRESS2		0x32
+#define SSB_PCMCIA_MEMSEG		0x34
+#define SSB_PCMCIA_SPROMCTL		0x36
+#define  SSB_PCMCIA_SPROMCTL_IDLE	0
+#define  SSB_PCMCIA_SPROMCTL_WRITE	1
+#define  SSB_PCMCIA_SPROMCTL_READ	2
+#define  SSB_PCMCIA_SPROMCTL_WRITEEN	4
+#define  SSB_PCMCIA_SPROMCTL_WRITEDIS	7
+#define  SSB_PCMCIA_SPROMCTL_DONE	8
+#define SSB_PCMCIA_SPROM_DATALO		0x38
+#define SSB_PCMCIA_SPROM_DATAHI		0x3A
+#define SSB_PCMCIA_SPROM_ADDRLO		0x3C
+#define SSB_PCMCIA_SPROM_ADDRHI		0x3E
+
+/* Hardware invariants CIS tuples */
+#define SSB_PCMCIA_CIS			0x80
+#define  SSB_PCMCIA_CIS_ID		0x01
+#define  SSB_PCMCIA_CIS_BOARDREV	0x02
+#define  SSB_PCMCIA_CIS_PA		0x03
+#define   SSB_PCMCIA_CIS_PA_PA0B0_LO	0
+#define   SSB_PCMCIA_CIS_PA_PA0B0_HI	1
+#define   SSB_PCMCIA_CIS_PA_PA0B1_LO	2
+#define   SSB_PCMCIA_CIS_PA_PA0B1_HI	3
+#define   SSB_PCMCIA_CIS_PA_PA0B2_LO	4
+#define   SSB_PCMCIA_CIS_PA_PA0B2_HI	5
+#define   SSB_PCMCIA_CIS_PA_ITSSI	6
+#define   SSB_PCMCIA_CIS_PA_MAXPOW	7
+#define  SSB_PCMCIA_CIS_OEMNAME		0x04
+#define  SSB_PCMCIA_CIS_CCODE		0x05
+#define  SSB_PCMCIA_CIS_ANTENNA		0x06
+#define  SSB_PCMCIA_CIS_ANTGAIN		0x07
+#define  SSB_PCMCIA_CIS_BFLAGS		0x08
+#define  SSB_PCMCIA_CIS_LEDS		0x09
+
+/* PCMCIA SPROM size. */
+#define SSB_PCMCIA_SPROM_SIZE		256
+#define SSB_PCMCIA_SPROM_SIZE_BYTES	(SSB_PCMCIA_SPROM_SIZE * sizeof(u16))
+
+
+/* Write to a PCMCIA configuration register. */
+static int ssb_pcmcia_cfg_write(struct ssb_bus *bus, u8 offset, u8 value)
+{
+	conf_reg_t reg;
+	int res;
+
+	memset(&reg, 0, sizeof(reg));
+	reg.Offset = offset;
+	reg.Action = CS_WRITE;
+	reg.Value = value;
+	res = pcmcia_access_configuration_register(bus->host_pcmcia, &reg);
+	if (unlikely(res != CS_SUCCESS))
+		return -EBUSY;
+
+	return 0;
+}
+
+/* Read from a PCMCIA configuration register. */
+static int ssb_pcmcia_cfg_read(struct ssb_bus *bus, u8 offset, u8 *value)
+{
+	conf_reg_t reg;
+	int res;
+
+	memset(&reg, 0, sizeof(reg));
+	reg.Offset = offset;
+	reg.Action = CS_READ;
+	res = pcmcia_access_configuration_register(bus->host_pcmcia, &reg);
+	if (unlikely(res != CS_SUCCESS))
+		return -EBUSY;
+	*value = reg.Value;
+
+	return 0;
+}
+
 int ssb_pcmcia_switch_coreidx(struct ssb_bus *bus,
 			      u8 coreidx)
 {
-	struct pcmcia_device *pdev = bus->host_pcmcia;
 	int err;
 	int attempts = 0;
 	u32 cur_core;
-	conf_reg_t reg;
 	u32 addr;
 	u32 read_addr;
+	u8 val;
 
 	addr = (coreidx * SSB_CORE_SIZE) + SSB_ENUM_BASE;
 	while (1) {
-		reg.Action = CS_WRITE;
-		reg.Offset = 0x2E;
-		reg.Value = (addr & 0x0000F000) >> 12;
-		err = pcmcia_access_configuration_register(pdev, &reg);
-		if (err != CS_SUCCESS)
+		err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_ADDRESS0,
+					   (addr & 0x0000F000) >> 12);
+		if (err)
 			goto error;
-		reg.Offset = 0x30;
-		reg.Value = (addr & 0x00FF0000) >> 16;
-		err = pcmcia_access_configuration_register(pdev, &reg);
-		if (err != CS_SUCCESS)
+		err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_ADDRESS1,
+					   (addr & 0x00FF0000) >> 16);
+		if (err)
 			goto error;
-		reg.Offset = 0x32;
-		reg.Value = (addr & 0xFF000000) >> 24;
-		err = pcmcia_access_configuration_register(pdev, &reg);
-		if (err != CS_SUCCESS)
+		err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_ADDRESS2,
+					   (addr & 0xFF000000) >> 24);
+		if (err)
 			goto error;
 
 		read_addr = 0;
 
-		reg.Action = CS_READ;
-		reg.Offset = 0x2E;
-		err = pcmcia_access_configuration_register(pdev, &reg);
-		if (err != CS_SUCCESS)
+		err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_ADDRESS0, &val);
+		if (err)
 			goto error;
-		read_addr |= ((u32)(reg.Value & 0x0F)) << 12;
-		reg.Offset = 0x30;
-		err = pcmcia_access_configuration_register(pdev, &reg);
-		if (err != CS_SUCCESS)
+		read_addr |= ((u32)(val & 0x0F)) << 12;
+		err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_ADDRESS1, &val);
+		if (err)
 			goto error;
-		read_addr |= ((u32)reg.Value) << 16;
-		reg.Offset = 0x32;
-		err = pcmcia_access_configuration_register(pdev, &reg);
-		if (err != CS_SUCCESS)
+		read_addr |= ((u32)val) << 16;
+		err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_ADDRESS2, &val);
+		if (err)
 			goto error;
-		read_addr |= ((u32)reg.Value) << 24;
+		read_addr |= ((u32)val) << 24;
 
 		cur_core = (read_addr - SSB_ENUM_BASE) / SSB_CORE_SIZE;
 		if (cur_core == coreidx)
 			break;
 
+		err = -ETIMEDOUT;
 		if (attempts++ > SSB_BAR0_MAX_RETRIES)
 			goto error;
 		udelay(10);
@@ -87,7 +161,7 @@ int ssb_pcmcia_switch_coreidx(struct ssb_bus *bus,
 	return 0;
 error:
 	ssb_printk(KERN_ERR PFX "Failed to switch to core %u\n", coreidx);
-	return -ENODEV;
+	return err;
 }
 
 int ssb_pcmcia_switch_core(struct ssb_bus *bus,
@@ -112,27 +186,21 @@ int ssb_pcmcia_switch_core(struct ssb_bus *bus,
 int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg)
 {
 	int attempts = 0;
-	conf_reg_t reg;
-	int res;
+	int err;
+	u8 val;
 
 	SSB_WARN_ON((seg != 0) && (seg != 1));
-	reg.Offset = 0x34;
-	reg.Function = 0;
 	while (1) {
-		reg.Action = CS_WRITE;
-		reg.Value = seg;
-		res = pcmcia_access_configuration_register(bus->host_pcmcia, &reg);
-		if (unlikely(res != CS_SUCCESS))
+		err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_MEMSEG, seg);
+		if (err)
 			goto error;
-		reg.Value = 0xFF;
-		reg.Action = CS_READ;
-		res = pcmcia_access_configuration_register(bus->host_pcmcia, &reg);
-		if (unlikely(res != CS_SUCCESS))
+		err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_MEMSEG, &val);
+		if (err)
 			goto error;
-
-		if (reg.Value == seg)
+		if (val == seg)
 			break;
 
+		err = -ETIMEDOUT;
 		if (unlikely(attempts++ > SSB_BAR0_MAX_RETRIES))
 			goto error;
 		udelay(10);
@@ -142,7 +210,7 @@ int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg)
 	return 0;
 error:
 	ssb_printk(KERN_ERR PFX "Failed to switch pcmcia segment\n");
-	return -ENODEV;
+	return err;
 }
 
 static int select_core_and_segment(struct ssb_device *dev,
@@ -276,18 +344,344 @@ const struct ssb_bus_ops ssb_pcmcia_ops = {
 	.write32	= ssb_pcmcia_write32,
 };
 
-#include <linux/etherdevice.h>
+static int ssb_pcmcia_sprom_command(struct ssb_bus *bus, u8 command)
+{
+	unsigned int i;
+	int err;
+	u8 value;
+
+	err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROMCTL, command);
+	if (err)
+		return err;
+	for (i = 0; i < 1000; i++) {
+		err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_SPROMCTL, &value);
+		if (err)
+			return err;
+		if (value & SSB_PCMCIA_SPROMCTL_DONE)
+			return 0;
+		udelay(10);
+	}
+
+	return -ETIMEDOUT;
+}
+
+/* offset is the 16bit word offset */
+static int ssb_pcmcia_sprom_read(struct ssb_bus *bus, u16 offset, u16 *value)
+{
+	int err;
+	u8 lo, hi;
+
+	offset *= 2; /* Make byte offset */
+
+	err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRLO,
+				   (offset & 0x00FF));
+	if (err)
+		return err;
+	err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRHI,
+				   (offset & 0xFF00) >> 8);
+	if (err)
+		return err;
+	err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_READ);
+	if (err)
+		return err;
+	err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_SPROM_DATALO, &lo);
+	if (err)
+		return err;
+	err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_SPROM_DATAHI, &hi);
+	if (err)
+		return err;
+	*value = (lo | (((u16)hi) << 8));
+
+	return 0;
+}
+
+/* offset is the 16bit word offset */
+static int ssb_pcmcia_sprom_write(struct ssb_bus *bus, u16 offset, u16 value)
+{
+	int err;
+
+	offset *= 2; /* Make byte offset */
+
+	err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRLO,
+				   (offset & 0x00FF));
+	if (err)
+		return err;
+	err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRHI,
+				   (offset & 0xFF00) >> 8);
+	if (err)
+		return err;
+	err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_DATALO,
+				   (value & 0x00FF));
+	if (err)
+		return err;
+	err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_DATAHI,
+				   (value & 0xFF00) >> 8);
+	if (err)
+		return err;
+	err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_WRITE);
+	if (err)
+		return err;
+	msleep(20);
+
+	return 0;
+}
+
+/* Read the SPROM image. bufsize is in 16bit words. */
+static int ssb_pcmcia_sprom_read_all(struct ssb_bus *bus, u16 *sprom)
+{
+	int err, i;
+
+	for (i = 0; i < SSB_PCMCIA_SPROM_SIZE; i++) {
+		err = ssb_pcmcia_sprom_read(bus, i, &sprom[i]);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/* Write the SPROM image. size is in 16bit words. */
+static int ssb_pcmcia_sprom_write_all(struct ssb_bus *bus, const u16 *sprom)
+{
+	int i, err;
+	bool failed = 0;
+	size_t size = SSB_PCMCIA_SPROM_SIZE;
+
+	ssb_printk(KERN_NOTICE PFX
+		   "Writing SPROM. Do NOT turn off the power! "
+		   "Please stand by...\n");
+	err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_WRITEEN);
+	if (err) {
+		ssb_printk(KERN_NOTICE PFX
+			   "Could not enable SPROM write access.\n");
+		return -EBUSY;
+	}
+	ssb_printk(KERN_NOTICE PFX "[ 0%%");
+	msleep(500);
+	for (i = 0; i < size; i++) {
+		if (i == size / 4)
+			ssb_printk("25%%");
+		else if (i == size / 2)
+			ssb_printk("50%%");
+		else if (i == (size * 3) / 4)
+			ssb_printk("75%%");
+		else if (i % 2)
+			ssb_printk(".");
+		err = ssb_pcmcia_sprom_write(bus, i, sprom[i]);
+		if (err) {
+			ssb_printk("\n" KERN_NOTICE PFX
+				   "Failed to write to SPROM.\n");
+			failed = 1;
+			break;
+		}
+	}
+	err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_WRITEDIS);
+	if (err) {
+		ssb_printk("\n" KERN_NOTICE PFX
+			   "Could not disable SPROM write access.\n");
+		failed = 1;
+	}
+	msleep(500);
+	if (!failed) {
+		ssb_printk("100%% ]\n");
+		ssb_printk(KERN_NOTICE PFX "SPROM written.\n");
+	}
+
+	return failed ? -EBUSY : 0;
+}
+
+static int ssb_pcmcia_sprom_check_crc(const u16 *sprom, size_t size)
+{
+	//TODO
+	return 0;
+}
+
+#define GOTO_ERROR_ON(condition, description) do {	\
+	if (unlikely(condition)) {			\
+		error_description = description;	\
+		goto error;				\
+	}						\
+  } while (0)
+
 int ssb_pcmcia_get_invariants(struct ssb_bus *bus,
 			      struct ssb_init_invariants *iv)
 {
-	//TODO
-	random_ether_addr(iv->sprom.il0mac);
+	tuple_t tuple;
+	int res;
+	unsigned char buf[32];
+	struct ssb_sprom *sprom = &iv->sprom;
+	struct ssb_boardinfo *bi = &iv->boardinfo;
+	const char *error_description;
+
+	memset(sprom, 0xFF, sizeof(*sprom));
+	sprom->revision = 1;
+	sprom->boardflags_lo = 0;
+	sprom->boardflags_hi = 0;
+
+	/* First fetch the MAC address. */
+	memset(&tuple, 0, sizeof(tuple));
+	tuple.DesiredTuple = CISTPL_FUNCE;
+	tuple.TupleData = buf;
+	tuple.TupleDataMax = sizeof(buf);
+	res = pcmcia_get_first_tuple(bus->host_pcmcia, &tuple);
+	GOTO_ERROR_ON(res != CS_SUCCESS, "MAC first tpl");
+	res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple);
+	GOTO_ERROR_ON(res != CS_SUCCESS, "MAC first tpl data");
+	while (1) {
+		GOTO_ERROR_ON(tuple.TupleDataLen < 1, "MAC tpl < 1");
+		if (tuple.TupleData[0] == CISTPL_FUNCE_LAN_NODE_ID)
+			break;
+		res = pcmcia_get_next_tuple(bus->host_pcmcia, &tuple);
+		GOTO_ERROR_ON(res != CS_SUCCESS, "MAC next tpl");
+		res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple);
+		GOTO_ERROR_ON(res != CS_SUCCESS, "MAC next tpl data");
+	}
+	GOTO_ERROR_ON(tuple.TupleDataLen != ETH_ALEN + 2, "MAC tpl size");
+	memcpy(sprom->il0mac, &tuple.TupleData[2], ETH_ALEN);
+
+	/* Fetch the vendor specific tuples. */
+	memset(&tuple, 0, sizeof(tuple));
+	tuple.DesiredTuple = SSB_PCMCIA_CIS;
+	tuple.TupleData = buf;
+	tuple.TupleDataMax = sizeof(buf);
+	res = pcmcia_get_first_tuple(bus->host_pcmcia, &tuple);
+	GOTO_ERROR_ON(res != CS_SUCCESS, "VEN first tpl");
+	res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple);
+	GOTO_ERROR_ON(res != CS_SUCCESS, "VEN first tpl data");
+	while (1) {
+		GOTO_ERROR_ON(tuple.TupleDataLen < 1, "VEN tpl < 1");
+		switch (tuple.TupleData[0]) {
+		case SSB_PCMCIA_CIS_ID:
+			GOTO_ERROR_ON((tuple.TupleDataLen != 5) &&
+				      (tuple.TupleDataLen != 7),
+				      "id tpl size");
+			bi->vendor = tuple.TupleData[1] |
+			       ((u16)tuple.TupleData[2] << 8);
+			break;
+		case SSB_PCMCIA_CIS_BOARDREV:
+			GOTO_ERROR_ON(tuple.TupleDataLen != 2,
+				      "boardrev tpl size");
+			sprom->board_rev = tuple.TupleData[1];
+			break;
+		case SSB_PCMCIA_CIS_PA:
+			GOTO_ERROR_ON(tuple.TupleDataLen != 9,
+				      "pa tpl size");
+			sprom->pa0b0 = tuple.TupleData[1] |
+				 ((u16)tuple.TupleData[2] << 8);
+			sprom->pa0b1 = tuple.TupleData[3] |
+				 ((u16)tuple.TupleData[4] << 8);
+			sprom->pa0b2 = tuple.TupleData[5] |
+				 ((u16)tuple.TupleData[6] << 8);
+			sprom->itssi_a = tuple.TupleData[7];
+			sprom->itssi_bg = tuple.TupleData[7];
+			sprom->maxpwr_a = tuple.TupleData[8];
+			sprom->maxpwr_bg = tuple.TupleData[8];
+			break;
+		case SSB_PCMCIA_CIS_OEMNAME:
+			/* We ignore this. */
+			break;
+		case SSB_PCMCIA_CIS_CCODE:
+			GOTO_ERROR_ON(tuple.TupleDataLen != 2,
+				      "ccode tpl size");
+			sprom->country_code = tuple.TupleData[1];
+			break;
+		case SSB_PCMCIA_CIS_ANTENNA:
+			GOTO_ERROR_ON(tuple.TupleDataLen != 2,
+				      "ant tpl size");
+			sprom->ant_available_a = tuple.TupleData[1];
+			sprom->ant_available_bg = tuple.TupleData[1];
+			break;
+		case SSB_PCMCIA_CIS_ANTGAIN:
+			GOTO_ERROR_ON(tuple.TupleDataLen != 2,
+				      "antg tpl size");
+			sprom->antenna_gain.ghz24.a0 = tuple.TupleData[1];
+			sprom->antenna_gain.ghz24.a1 = tuple.TupleData[1];
+			sprom->antenna_gain.ghz24.a2 = tuple.TupleData[1];
+			sprom->antenna_gain.ghz24.a3 = tuple.TupleData[1];
+			sprom->antenna_gain.ghz5.a0 = tuple.TupleData[1];
+			sprom->antenna_gain.ghz5.a1 = tuple.TupleData[1];
+			sprom->antenna_gain.ghz5.a2 = tuple.TupleData[1];
+			sprom->antenna_gain.ghz5.a3 = tuple.TupleData[1];
+			break;
+		case SSB_PCMCIA_CIS_BFLAGS:
+			GOTO_ERROR_ON(tuple.TupleDataLen != 3,
+				      "bfl tpl size");
+			sprom->boardflags_lo = tuple.TupleData[1] |
+					 ((u16)tuple.TupleData[2] << 8);
+			break;
+		case SSB_PCMCIA_CIS_LEDS:
+			GOTO_ERROR_ON(tuple.TupleDataLen != 5,
+				      "leds tpl size");
+			sprom->gpio0 = tuple.TupleData[1];
+			sprom->gpio1 = tuple.TupleData[2];
+			sprom->gpio2 = tuple.TupleData[3];
+			sprom->gpio3 = tuple.TupleData[4];
+			break;
+		}
+		res = pcmcia_get_next_tuple(bus->host_pcmcia, &tuple);
+		if (res == CS_NO_MORE_ITEMS)
+			break;
+		GOTO_ERROR_ON(res != CS_SUCCESS, "VEN next tpl");
+		res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple);
+		GOTO_ERROR_ON(res != CS_SUCCESS, "VEN next tpl data");
+	}
+
 	return 0;
+error:
+	ssb_printk(KERN_ERR PFX
+		   "PCMCIA: Failed to fetch device invariants: %s\n",
+		   error_description);
+	return -ENODEV;
+}
+
+static ssize_t ssb_pcmcia_attr_sprom_show(struct device *pcmciadev,
+					  struct device_attribute *attr,
+					  char *buf)
+{
+	struct pcmcia_device *pdev =
+		container_of(pcmciadev, struct pcmcia_device, dev);
+	struct ssb_bus *bus;
+
+	bus = ssb_pcmcia_dev_to_bus(pdev);
+	if (!bus)
+		return -ENODEV;
+
+	return ssb_attr_sprom_show(bus, buf,
+				   ssb_pcmcia_sprom_read_all);
+}
+
+static ssize_t ssb_pcmcia_attr_sprom_store(struct device *pcmciadev,
+					   struct device_attribute *attr,
+					   const char *buf, size_t count)
+{
+	struct pcmcia_device *pdev =
+		container_of(pcmciadev, struct pcmcia_device, dev);
+	struct ssb_bus *bus;
+
+	bus = ssb_pcmcia_dev_to_bus(pdev);
+	if (!bus)
+		return -ENODEV;
+
+	return ssb_attr_sprom_store(bus, buf, count,
+				    ssb_pcmcia_sprom_check_crc,
+				    ssb_pcmcia_sprom_write_all);
+}
+
+static DEVICE_ATTR(ssb_sprom, 0600,
+		   ssb_pcmcia_attr_sprom_show,
+		   ssb_pcmcia_attr_sprom_store);
+
+void ssb_pcmcia_exit(struct ssb_bus *bus)
+{
+	if (bus->bustype != SSB_BUSTYPE_PCMCIA)
+		return;
+
+	device_remove_file(&bus->host_pcmcia->dev, &dev_attr_ssb_sprom);
 }
 
 int ssb_pcmcia_init(struct ssb_bus *bus)
 {
-	conf_reg_t reg;
+	u8 val, offset;
 	int err;
 
 	if (bus->bustype != SSB_BUSTYPE_PCMCIA)
@@ -298,22 +692,26 @@ int ssb_pcmcia_init(struct ssb_bus *bus)
 	ssb_pcmcia_switch_segment(bus, 0);
 
 	/* Init IRQ routing */
-	reg.Action = CS_READ;
-	reg.Function = 0;
 	if (bus->chip_id == 0x4306)
-		reg.Offset = 0x00;
+		offset = SSB_PCMCIA_CORECTL;
 	else
-		reg.Offset = 0x80;
-	err = pcmcia_access_configuration_register(bus->host_pcmcia, &reg);
-	if (err != CS_SUCCESS)
+		offset = SSB_PCMCIA_CORECTL2;
+	err = ssb_pcmcia_cfg_read(bus, offset, &val);
+	if (err)
 		goto error;
-	reg.Action = CS_WRITE;
-	reg.Value |= 0x04 | 0x01;
-	err = pcmcia_access_configuration_register(bus->host_pcmcia, &reg);
-	if (err != CS_SUCCESS)
+	val |= SSB_PCMCIA_CORECTL_IRQEN | SSB_PCMCIA_CORECTL_FUNCEN;
+	err = ssb_pcmcia_cfg_write(bus, offset, val);
+	if (err)
+		goto error;
+
+	bus->sprom_size = SSB_PCMCIA_SPROM_SIZE;
+	mutex_init(&bus->sprom_mutex);
+	err = device_create_file(&bus->host_pcmcia->dev, &dev_attr_ssb_sprom);
+	if (err)
 		goto error;
 
 	return 0;
 error:
-	return -ENODEV;
+	ssb_printk(KERN_ERR PFX "Failed to initialize PCMCIA host device\n");
+	return err;
 }
diff --git a/drivers/ssb/sprom.c b/drivers/ssb/sprom.c
new file mode 100644
index 000000000000..3668edb39315
--- /dev/null
+++ b/drivers/ssb/sprom.c
@@ -0,0 +1,133 @@
+/*
+ * Sonics Silicon Backplane
+ * Common SPROM support routines
+ *
+ * Copyright (C) 2005-2008 Michael Buesch <mb@bu3sch.de>
+ * Copyright (C) 2005 Martin Langer <martin-langer@gmx.de>
+ * Copyright (C) 2005 Stefano Brivio <st3@riseup.net>
+ * Copyright (C) 2005 Danny van Dyk <kugelfang@gentoo.org>
+ * Copyright (C) 2005 Andreas Jaggi <andreas.jaggi@waterwave.ch>
+ *
+ * Licensed under the GNU/GPL. See COPYING for details.
+ */
+
+#include "ssb_private.h"
+
+
+static int sprom2hex(const u16 *sprom, char *buf, size_t buf_len,
+		     size_t sprom_size_words)
+{
+	int i, pos = 0;
+
+	for (i = 0; i < sprom_size_words; i++)
+		pos += snprintf(buf + pos, buf_len - pos - 1,
+				"%04X", swab16(sprom[i]) & 0xFFFF);
+	pos += snprintf(buf + pos, buf_len - pos - 1, "\n");
+
+	return pos + 1;
+}
+
+static int hex2sprom(u16 *sprom, const char *dump, size_t len,
+		     size_t sprom_size_words)
+{
+	char tmp[5] = { 0 };
+	int cnt = 0;
+	unsigned long parsed;
+
+	if (len < sprom_size_words * 2)
+		return -EINVAL;
+
+	while (cnt < sprom_size_words) {
+		memcpy(tmp, dump, 4);
+		dump += 4;
+		parsed = simple_strtoul(tmp, NULL, 16);
+		sprom[cnt++] = swab16((u16)parsed);
+	}
+
+	return 0;
+}
+
+/* Common sprom device-attribute show-handler */
+ssize_t ssb_attr_sprom_show(struct ssb_bus *bus, char *buf,
+			    int (*sprom_read)(struct ssb_bus *bus, u16 *sprom))
+{
+	u16 *sprom;
+	int err = -ENOMEM;
+	ssize_t count = 0;
+	size_t sprom_size_words = bus->sprom_size;
+
+	sprom = kcalloc(sprom_size_words, sizeof(u16), GFP_KERNEL);
+	if (!sprom)
+		goto out;
+
+	/* Use interruptible locking, as the SPROM write might
+	 * be holding the lock for several seconds. So allow userspace
+	 * to cancel operation. */
+	err = -ERESTARTSYS;
+	if (mutex_lock_interruptible(&bus->sprom_mutex))
+		goto out_kfree;
+	err = sprom_read(bus, sprom);
+	mutex_unlock(&bus->sprom_mutex);
+
+	if (!err)
+		count = sprom2hex(sprom, buf, PAGE_SIZE, sprom_size_words);
+
+out_kfree:
+	kfree(sprom);
+out:
+	return err ? err : count;
+}
+
+/* Common sprom device-attribute store-handler */
+ssize_t ssb_attr_sprom_store(struct ssb_bus *bus,
+			     const char *buf, size_t count,
+			     int (*sprom_check_crc)(const u16 *sprom, size_t size),
+			     int (*sprom_write)(struct ssb_bus *bus, const u16 *sprom))
+{
+	u16 *sprom;
+	int res = 0, err = -ENOMEM;
+	size_t sprom_size_words = bus->sprom_size;
+
+	sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL);
+	if (!sprom)
+		goto out;
+	err = hex2sprom(sprom, buf, count, sprom_size_words);
+	if (err) {
+		err = -EINVAL;
+		goto out_kfree;
+	}
+	err = sprom_check_crc(sprom, sprom_size_words);
+	if (err) {
+		err = -EINVAL;
+		goto out_kfree;
+	}
+
+	/* Use interruptible locking, as the SPROM write might
+	 * be holding the lock for several seconds. So allow userspace
+	 * to cancel operation. */
+	err = -ERESTARTSYS;
+	if (mutex_lock_interruptible(&bus->sprom_mutex))
+		goto out_kfree;
+	err = ssb_devices_freeze(bus);
+	if (err == -EOPNOTSUPP) {
+		ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze devices. "
+			   "No suspend support. Is CONFIG_PM enabled?\n");
+		goto out_unlock;
+	}
+	if (err) {
+		ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze all devices\n");
+		goto out_unlock;
+	}
+	res = sprom_write(bus, sprom);
+	err = ssb_devices_thaw(bus);
+	if (err)
+		ssb_printk(KERN_ERR PFX "SPROM write: Could not thaw all devices\n");
+out_unlock:
+	mutex_unlock(&bus->sprom_mutex);
+out_kfree:
+	kfree(sprom);
+out:
+	if (res)
+		return res;
+	return err ? err : count;
+}
diff --git a/drivers/ssb/ssb_private.h b/drivers/ssb/ssb_private.h
index d03b20983b1e..a83bf7a4d80b 100644
--- a/drivers/ssb/ssb_private.h
+++ b/drivers/ssb/ssb_private.h
@@ -81,6 +81,7 @@ extern int ssb_pcmcia_switch_segment(struct ssb_bus *bus,
 				     u8 seg);
 extern int ssb_pcmcia_get_invariants(struct ssb_bus *bus,
 				     struct ssb_init_invariants *iv);
+extern void ssb_pcmcia_exit(struct ssb_bus *bus);
 extern int ssb_pcmcia_init(struct ssb_bus *bus);
 extern const struct ssb_bus_ops ssb_pcmcia_ops;
 #else /* CONFIG_SSB_PCMCIAHOST */
@@ -99,6 +100,9 @@ static inline int ssb_pcmcia_switch_segment(struct ssb_bus *bus,
 {
 	return 0;
 }
+static inline void ssb_pcmcia_exit(struct ssb_bus *bus)
+{
+}
 static inline int ssb_pcmcia_init(struct ssb_bus *bus)
 {
 	return 0;
@@ -113,6 +117,17 @@ extern int ssb_bus_scan(struct ssb_bus *bus,
 extern void ssb_iounmap(struct ssb_bus *ssb);
 
 
+/* sprom.c */
+extern
+ssize_t ssb_attr_sprom_show(struct ssb_bus *bus, char *buf,
+			    int (*sprom_read)(struct ssb_bus *bus, u16 *sprom));
+extern
+ssize_t ssb_attr_sprom_store(struct ssb_bus *bus,
+			     const char *buf, size_t count,
+			     int (*sprom_check_crc)(const u16 *sprom, size_t size),
+			     int (*sprom_write)(struct ssb_bus *bus, const u16 *sprom));
+
+
 /* core.c */
 extern u32 ssb_calc_clock_rate(u32 plltype, u32 n, u32 m);
 extern int ssb_devices_freeze(struct ssb_bus *bus);
@@ -120,6 +135,8 @@ extern int ssb_devices_thaw(struct ssb_bus *bus);
 extern struct ssb_bus *ssb_pci_dev_to_bus(struct pci_dev *pdev);
 int ssb_for_each_bus_call(unsigned long data,
 			  int (*func)(struct ssb_bus *bus, unsigned long data));
+extern struct ssb_bus *ssb_pcmcia_dev_to_bus(struct pcmcia_device *pdev);
+
 
 /* b43_pci_bridge.c */
 #ifdef CONFIG_SSB_B43_PCI_BRIDGE
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index b7c388972fcf..8644e03cf588 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -245,9 +245,9 @@ struct ssb_bus {
 	/* Pointer to the PCMCIA device (only if bustype == SSB_BUSTYPE_PCMCIA). */
 	struct pcmcia_device *host_pcmcia;
 
-#ifdef CONFIG_SSB_PCIHOST
+#ifdef CONFIG_SSB_SPROM
 	/* Mutex to protect the SPROM writing. */
-	struct mutex pci_sprom_mutex;
+	struct mutex sprom_mutex;
 #endif
 
 	/* ID information about the Chip. */
-- 
cgit v1.2.3


From 3e94794355724f77dc6cbb5ad956f7c72d8313a4 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Fri, 22 Feb 2008 19:55:15 +0900
Subject: smc91x: introduce platform data flags V2

This patch introduces struct smc91x_platdata and modifies the driver so
bus width is checked during run time using SMC_nBIT() instead of
SMC_CAN_USE_nBIT.

V2 keeps static configuration lean using SMC_DYNAMIC_BUS_CONFIG.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/smc91x.c   | 34 +++++++++++++++++++++++++-----
 drivers/net/smc91x.h   | 57 +++++++++++++++++++++++++++++++-------------------
 include/linux/smc91x.h | 13 ++++++++++++
 3 files changed, 77 insertions(+), 27 deletions(-)
 create mode 100644 include/linux/smc91x.h

(limited to 'include')

diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c
index d0ef80ae018a..97bdb2a43bc8 100644
--- a/drivers/net/smc91x.c
+++ b/drivers/net/smc91x.c
@@ -1997,6 +1997,8 @@ err_out:
 
 static int smc_enable_device(struct platform_device *pdev)
 {
+	struct net_device *ndev = platform_get_drvdata(pdev);
+	struct smc_local *lp = netdev_priv(ndev);
 	unsigned long flags;
 	unsigned char ecor, ecsr;
 	void __iomem *addr;
@@ -2039,7 +2041,7 @@ static int smc_enable_device(struct platform_device *pdev)
 	 * Set the appropriate byte/word mode.
 	 */
 	ecsr = readb(addr + (ECSR << SMC_IO_SHIFT)) & ~ECSR_IOIS8;
-	if (!SMC_CAN_USE_16BIT)
+	if (!SMC_16BIT(lp))
 		ecsr |= ECSR_IOIS8;
 	writeb(ecsr, addr + (ECSR << SMC_IO_SHIFT));
 	local_irq_restore(flags);
@@ -2124,10 +2126,11 @@ static void smc_release_datacs(struct platform_device *pdev, struct net_device *
  */
 static int smc_drv_probe(struct platform_device *pdev)
 {
+	struct smc91x_platdata *pd = pdev->dev.platform_data;
+	struct smc_local *lp;
 	struct net_device *ndev;
 	struct resource *res, *ires;
 	unsigned int __iomem *addr;
-	unsigned long irq_flags = SMC_IRQ_FLAGS;
 	int ret;
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-regs");
@@ -2152,6 +2155,27 @@ static int smc_drv_probe(struct platform_device *pdev)
 	}
 	SET_NETDEV_DEV(ndev, &pdev->dev);
 
+	/* get configuration from platform data, only allow use of
+	 * bus width if both SMC_CAN_USE_xxx and SMC91X_USE_xxx are set.
+	 */
+
+	lp = netdev_priv(ndev);
+	lp->cfg.irq_flags = SMC_IRQ_FLAGS;
+
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+	if (pd)
+		memcpy(&lp->cfg, pd, sizeof(lp->cfg));
+	else {
+		lp->cfg.flags = SMC91X_USE_8BIT;
+		lp->cfg.flags |= SMC91X_USE_16BIT;
+		lp->cfg.flags |= SMC91X_USE_32BIT;
+	}
+
+	lp->cfg.flags &= ~(SMC_CAN_USE_8BIT ? 0 : SMC91X_USE_8BIT);
+	lp->cfg.flags &= ~(SMC_CAN_USE_16BIT ? 0 : SMC91X_USE_16BIT);
+	lp->cfg.flags &= ~(SMC_CAN_USE_32BIT ? 0 : SMC91X_USE_32BIT);
+#endif
+
 	ndev->dma = (unsigned char)-1;
 
 	ires = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
@@ -2162,7 +2186,7 @@ static int smc_drv_probe(struct platform_device *pdev)
 
 	ndev->irq = ires->start;
 	if (SMC_IRQ_FLAGS == -1)
-		irq_flags = ires->flags & IRQF_TRIGGER_MASK;
+		lp->cfg.irq_flags = ires->flags & IRQF_TRIGGER_MASK;
 
 	ret = smc_request_attrib(pdev);
 	if (ret)
@@ -2170,6 +2194,7 @@ static int smc_drv_probe(struct platform_device *pdev)
 #if defined(CONFIG_SA1100_ASSABET)
 	NCR_0 |= NCR_ENET_OSC_EN;
 #endif
+	platform_set_drvdata(pdev, ndev);
 	ret = smc_enable_device(pdev);
 	if (ret)
 		goto out_release_attrib;
@@ -2188,8 +2213,7 @@ static int smc_drv_probe(struct platform_device *pdev)
 	}
 #endif
 
-	platform_set_drvdata(pdev, ndev);
-	ret = smc_probe(ndev, addr, irq_flags);
+	ret = smc_probe(ndev, addr, lp->cfg.irq_flags);
 	if (ret != 0)
 		goto out_iounmap;
 
diff --git a/drivers/net/smc91x.h b/drivers/net/smc91x.h
index 92ff9c42367e..e044b4de1397 100644
--- a/drivers/net/smc91x.h
+++ b/drivers/net/smc91x.h
@@ -34,6 +34,7 @@
 #ifndef _SMC91X_H_
 #define _SMC91X_H_
 
+#include <linux/smc91x.h>
 
 /*
  * Define your architecture specific bus configuration parameters here.
@@ -481,6 +482,7 @@ static inline void LPD7_SMC_outsw (unsigned char* a, int r,
 #define RPC_LSA_DEFAULT		RPC_LED_100_10
 #define RPC_LSB_DEFAULT		RPC_LED_TX_RX
 
+#define SMC_DYNAMIC_BUS_CONFIG
 #endif
 
 
@@ -526,8 +528,19 @@ struct smc_local {
 #endif
 	void __iomem *base;
 	void __iomem *datacs;
+
+	struct smc91x_platdata cfg;
 };
 
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+#define SMC_8BIT(p) (((p)->cfg.flags & SMC91X_USE_8BIT) && SMC_CAN_USE_8BIT)
+#define SMC_16BIT(p) (((p)->cfg.flags & SMC91X_USE_16BIT) && SMC_CAN_USE_16BIT)
+#define SMC_32BIT(p) (((p)->cfg.flags & SMC91X_USE_32BIT) && SMC_CAN_USE_32BIT)
+#else
+#define SMC_8BIT(p) SMC_CAN_USE_8BIT
+#define SMC_16BIT(p) SMC_CAN_USE_16BIT
+#define SMC_32BIT(p) SMC_CAN_USE_32BIT
+#endif
 
 #ifdef SMC_USE_PXA_DMA
 /*
@@ -1108,41 +1121,41 @@ static const char * chip_ids[ 16 ] =  {
  *
  * Enforce it on any 32-bit capable setup for now.
  */
-#define SMC_MUST_ALIGN_WRITE	SMC_CAN_USE_32BIT
+#define SMC_MUST_ALIGN_WRITE(lp)	SMC_32BIT(lp)
 
 #define SMC_GET_PN(lp)						\
-	(SMC_CAN_USE_8BIT	? (SMC_inb(ioaddr, PN_REG(lp)))	\
+	(SMC_8BIT(lp)	? (SMC_inb(ioaddr, PN_REG(lp)))	\
 				: (SMC_inw(ioaddr, PN_REG(lp)) & 0xFF))
 
 #define SMC_SET_PN(lp, x)						\
 	do {								\
-		if (SMC_MUST_ALIGN_WRITE)				\
+		if (SMC_MUST_ALIGN_WRITE(lp))				\
 			SMC_outl((x)<<16, ioaddr, SMC_REG(lp, 0, 2));	\
-		else if (SMC_CAN_USE_8BIT)				\
+		else if (SMC_8BIT(lp))				\
 			SMC_outb(x, ioaddr, PN_REG(lp));		\
 		else							\
 			SMC_outw(x, ioaddr, PN_REG(lp));		\
 	} while (0)
 
 #define SMC_GET_AR(lp)						\
-	(SMC_CAN_USE_8BIT	? (SMC_inb(ioaddr, AR_REG(lp)))	\
+	(SMC_8BIT(lp)	? (SMC_inb(ioaddr, AR_REG(lp)))	\
 				: (SMC_inw(ioaddr, PN_REG(lp)) >> 8))
 
 #define SMC_GET_TXFIFO(lp)						\
-	(SMC_CAN_USE_8BIT	? (SMC_inb(ioaddr, TXFIFO_REG(lp)))	\
+	(SMC_8BIT(lp)	? (SMC_inb(ioaddr, TXFIFO_REG(lp)))	\
 				: (SMC_inw(ioaddr, TXFIFO_REG(lp)) & 0xFF))
 
 #define SMC_GET_RXFIFO(lp)						\
-	(SMC_CAN_USE_8BIT	? (SMC_inb(ioaddr, RXFIFO_REG(lp)))	\
+	(SMC_8BIT(lp)	? (SMC_inb(ioaddr, RXFIFO_REG(lp)))	\
 				: (SMC_inw(ioaddr, TXFIFO_REG(lp)) >> 8))
 
 #define SMC_GET_INT(lp)						\
-	(SMC_CAN_USE_8BIT	? (SMC_inb(ioaddr, INT_REG(lp)))	\
+	(SMC_8BIT(lp)	? (SMC_inb(ioaddr, INT_REG(lp)))	\
 				: (SMC_inw(ioaddr, INT_REG(lp)) & 0xFF))
 
 #define SMC_ACK_INT(lp, x)						\
 	do {								\
-		if (SMC_CAN_USE_8BIT)					\
+		if (SMC_8BIT(lp))					\
 			SMC_outb(x, ioaddr, INT_REG(lp));		\
 		else {							\
 			unsigned long __flags;				\
@@ -1155,12 +1168,12 @@ static const char * chip_ids[ 16 ] =  {
 	} while (0)
 
 #define SMC_GET_INT_MASK(lp)						\
-	(SMC_CAN_USE_8BIT	? (SMC_inb(ioaddr, IM_REG(lp)))	\
+	(SMC_8BIT(lp)	? (SMC_inb(ioaddr, IM_REG(lp)))	\
 				: (SMC_inw(ioaddr, INT_REG(lp)) >> 8))
 
 #define SMC_SET_INT_MASK(lp, x)					\
 	do {								\
-		if (SMC_CAN_USE_8BIT)					\
+		if (SMC_8BIT(lp))					\
 			SMC_outb(x, ioaddr, IM_REG(lp));		\
 		else							\
 			SMC_outw((x) << 8, ioaddr, INT_REG(lp));	\
@@ -1170,7 +1183,7 @@ static const char * chip_ids[ 16 ] =  {
 
 #define SMC_SELECT_BANK(lp, x)					\
 	do {								\
-		if (SMC_MUST_ALIGN_WRITE)				\
+		if (SMC_MUST_ALIGN_WRITE(lp))				\
 			SMC_outl((x)<<16, ioaddr, 12<<SMC_IO_SHIFT);	\
 		else							\
 			SMC_outw(x, ioaddr, BANK_SELECT);		\
@@ -1208,7 +1221,7 @@ static const char * chip_ids[ 16 ] =  {
 
 #define SMC_SET_PTR(lp, x)						\
 	do {								\
-		if (SMC_MUST_ALIGN_WRITE)				\
+		if (SMC_MUST_ALIGN_WRITE(lp))				\
 			SMC_outl((x)<<16, ioaddr, SMC_REG(lp, 4, 2));	\
 		else							\
 			SMC_outw(x, ioaddr, PTR_REG(lp));		\
@@ -1226,7 +1239,7 @@ static const char * chip_ids[ 16 ] =  {
 
 #define SMC_SET_RPC(lp, x)						\
 	do {								\
-		if (SMC_MUST_ALIGN_WRITE)				\
+		if (SMC_MUST_ALIGN_WRITE(lp))				\
 			SMC_outl((x)<<16, ioaddr, SMC_REG(lp, 8, 0));	\
 		else							\
 			SMC_outw(x, ioaddr, RPC_REG(lp));		\
@@ -1267,7 +1280,7 @@ static const char * chip_ids[ 16 ] =  {
 
 #define SMC_PUT_PKT_HDR(lp, status, length)				\
 	do {								\
-		if (SMC_CAN_USE_32BIT)					\
+		if (SMC_32BIT(lp))					\
 			SMC_outl((status) | (length)<<16, ioaddr,	\
 				 DATA_REG(lp));			\
 		else {							\
@@ -1278,7 +1291,7 @@ static const char * chip_ids[ 16 ] =  {
 
 #define SMC_GET_PKT_HDR(lp, status, length)				\
 	do {								\
-		if (SMC_CAN_USE_32BIT) {				\
+		if (SMC_32BIT(lp)) {				\
 			unsigned int __val = SMC_inl(ioaddr, DATA_REG(lp)); \
 			(status) = __val & 0xffff;			\
 			(length) = __val >> 16;				\
@@ -1290,7 +1303,7 @@ static const char * chip_ids[ 16 ] =  {
 
 #define SMC_PUSH_DATA(lp, p, l)					\
 	do {								\
-		if (SMC_CAN_USE_32BIT) {				\
+		if (SMC_32BIT(lp)) {				\
 			void *__ptr = (p);				\
 			int __len = (l);				\
 			void __iomem *__ioaddr = ioaddr;		\
@@ -1308,15 +1321,15 @@ static const char * chip_ids[ 16 ] =  {
 				SMC_outw(*((u16 *)__ptr), ioaddr,	\
 					 DATA_REG(lp));		\
 			}						\
-		} else if (SMC_CAN_USE_16BIT)				\
+		} else if (SMC_16BIT(lp))				\
 			SMC_outsw(ioaddr, DATA_REG(lp), p, (l) >> 1);	\
-		else if (SMC_CAN_USE_8BIT)				\
+		else if (SMC_8BIT(lp))				\
 			SMC_outsb(ioaddr, DATA_REG(lp), p, l);	\
 	} while (0)
 
 #define SMC_PULL_DATA(lp, p, l)					\
 	do {								\
-		if (SMC_CAN_USE_32BIT) {				\
+		if (SMC_32BIT(lp)) {				\
 			void *__ptr = (p);				\
 			int __len = (l);				\
 			void __iomem *__ioaddr = ioaddr;		\
@@ -1343,9 +1356,9 @@ static const char * chip_ids[ 16 ] =  {
 				__ioaddr = lp->datacs;			\
 			__len += 2;					\
 			SMC_insl(__ioaddr, DATA_REG(lp), __ptr, __len>>2); \
-		} else if (SMC_CAN_USE_16BIT)				\
+		} else if (SMC_16BIT(lp))				\
 			SMC_insw(ioaddr, DATA_REG(lp), p, (l) >> 1);	\
-		else if (SMC_CAN_USE_8BIT)				\
+		else if (SMC_8BIT(lp))				\
 			SMC_insb(ioaddr, DATA_REG(lp), p, l);		\
 	} while (0)
 
diff --git a/include/linux/smc91x.h b/include/linux/smc91x.h
new file mode 100644
index 000000000000..8e0556b8781c
--- /dev/null
+++ b/include/linux/smc91x.h
@@ -0,0 +1,13 @@
+#ifndef __SMC91X_H__
+#define __SMC91X_H__
+
+#define SMC91X_USE_8BIT (1 << 0)
+#define SMC91X_USE_16BIT (1 << 1)
+#define SMC91X_USE_32BIT (1 << 2)
+
+struct smc91x_platdata {
+	unsigned long flags;
+	unsigned long irq_flags; /* IRQF_... */
+};
+
+#endif /* __SMC91X_H__ */
-- 
cgit v1.2.3


From 82cc1a7a56872056af0ead6c7d695aa223f36695 Mon Sep 17 00:00:00 2001
From: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Date: Fri, 21 Mar 2008 03:43:19 -0700
Subject: [NET]: Add per-connection option to set max TSO frame size

Update: My mailer ate one of Jarek's feedback mails...  Fixed the
parameter in netif_set_gso_max_size() to be u32, not u16.  Fixed the
whitespace issue due to a patch import botch.  Changed the types from
u32 to unsigned int to be more consistent with other variables in the
area.  Also brought the patch up to the latest net-2.6.26 tree.

Update: Made gso_max_size container 32 bits, not 16.  Moved the
location of gso_max_size within netdev to be less hotpath.  Made more
consistent names between the sock and netdev layers, and added a
define for the max GSO size.

Update: Respun for net-2.6.26 tree.

Update: changed max_gso_frame_size and sk_gso_max_size from signed to
unsigned - thanks Stephen!

This patch adds the ability for device drivers to control the size of
the TSO frames being sent to them, per TCP connection.  By setting the
netdevice's gso_max_size value, the socket layer will set the GSO
frame size based on that value.  This will propogate into the TCP
layer, and send TSO's of that size to the hardware.

This can be desirable to help tune the bursty nature of TSO on a
per-adapter basis, where one may have 1 GbE and 10 GbE devices
coexisting in a system, one running multiqueue and the other not, etc.

This can also be desirable for devices that cannot support full 64 KB
TSO's, but still want to benefit from some level of segmentation
offloading.

Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 10 ++++++++++
 include/net/sock.h        |  2 ++
 net/core/dev.c            |  1 +
 net/core/sock.c           |  6 ++++--
 net/ipv4/tcp_output.c     |  4 ++--
 5 files changed, 19 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a2f003239c85..ced61f87660e 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -724,6 +724,10 @@ struct net_device
 	/* rtnetlink link ops */
 	const struct rtnl_link_ops *rtnl_link_ops;
 
+	/* for setting kernel sock attribute on TCP connection setup */
+#define GSO_MAX_SIZE		65536
+	unsigned int		gso_max_size;
+
 	/* The TX queue control structures */
 	unsigned int			egress_subqueue_count;
 	struct net_device_subqueue	egress_subqueue[1];
@@ -1475,6 +1479,12 @@ static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
 		unlikely(skb->ip_summed != CHECKSUM_PARTIAL));
 }
 
+static inline void netif_set_gso_max_size(struct net_device *dev,
+					  unsigned int size)
+{
+	dev->gso_max_size = size;
+}
+
 /* On bonding slaves other than the currently active slave, suppress
  * duplicates except for 802.3ad ETH_P_SLOW, alb non-mcast/bcast, and
  * ARP on active-backup slaves with arp_validate enabled.
diff --git a/include/net/sock.h b/include/net/sock.h
index 39112e75411c..8358fff002eb 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -151,6 +151,7 @@ struct sock_common {
   *	@sk_no_check: %SO_NO_CHECK setting, wether or not checkup packets
   *	@sk_route_caps: route capabilities (e.g. %NETIF_F_TSO)
   *	@sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4)
+  *	@sk_gso_max_size: Maximum GSO segment size to build
   *	@sk_lingertime: %SO_LINGER l_linger setting
   *	@sk_backlog: always used with the per-socket spinlock held
   *	@sk_callback_lock: used with the callbacks in the end of this struct
@@ -237,6 +238,7 @@ struct sock {
 	gfp_t			sk_allocation;
 	int			sk_route_caps;
 	int			sk_gso_type;
+	unsigned int		sk_gso_max_size;
 	int			sk_rcvlowat;
 	unsigned long 		sk_flags;
 	unsigned long	        sk_lingertime;
diff --git a/net/core/dev.c b/net/core/dev.c
index fcdf03cf3b3f..f973e38b81af 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4021,6 +4021,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 	}
 
 	dev->egress_subqueue_count = queue_count;
+	dev->gso_max_size = GSO_MAX_SIZE;
 
 	dev->get_stats = internal_stats;
 	netpoll_netdev_init(dev);
diff --git a/net/core/sock.c b/net/core/sock.c
index bb5236aee643..b1a6ed4d33c1 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1095,10 +1095,12 @@ void sk_setup_caps(struct sock *sk, struct dst_entry *dst)
 	if (sk->sk_route_caps & NETIF_F_GSO)
 		sk->sk_route_caps |= NETIF_F_GSO_SOFTWARE;
 	if (sk_can_gso(sk)) {
-		if (dst->header_len)
+		if (dst->header_len) {
 			sk->sk_route_caps &= ~NETIF_F_GSO_MASK;
-		else
+		} else {
 			sk->sk_route_caps |= NETIF_F_SG | NETIF_F_HW_CSUM;
+			sk->sk_gso_max_size = dst->dev->gso_max_size;
+		}
 	}
 }
 EXPORT_SYMBOL_GPL(sk_setup_caps);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index b4e11d834c9f..a627616314ba 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -998,7 +998,7 @@ unsigned int tcp_current_mss(struct sock *sk, int large_allowed)
 	xmit_size_goal = mss_now;
 
 	if (doing_tso) {
-		xmit_size_goal = (65535 -
+		xmit_size_goal = ((sk->sk_gso_max_size - 1) -
 				  inet_csk(sk)->icsk_af_ops->net_header_len -
 				  inet_csk(sk)->icsk_ext_hdr_len -
 				  tp->tcp_header_len);
@@ -1282,7 +1282,7 @@ static int tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb)
 	limit = min(send_win, cong_win);
 
 	/* If a full-sized TSO skb can be sent, do it. */
-	if (limit >= 65536)
+	if (limit >= sk->sk_gso_max_size)
 		goto send_now;
 
 	if (sysctl_tcp_tso_win_divisor) {
-- 
cgit v1.2.3


From a91275eff43a527e1a25d6d034cbcd19ee323e64 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 21 Mar 2008 04:11:58 -0700
Subject: [NETNS][IPV6] udp - make proc handle the network namespace

This patch makes the common udp proc functions to take care of which
socket they should show taking into account the namespace it belongs.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h |  1 +
 net/ipv4/udp.c    | 32 ++++++++++++++++++++++++++++----
 2 files changed, 29 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index c6669c0a74c7..a1b33d667199 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -192,6 +192,7 @@ struct udp_seq_afinfo {
 };
 
 struct udp_iter_state {
+	struct net              *net;
 	sa_family_t		family;
 	struct hlist_head	*hashtable;
 	int			bucket;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 7ea1b67b6de1..049e92519616 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1512,10 +1512,13 @@ static struct sock *udp_get_first(struct seq_file *seq)
 {
 	struct sock *sk;
 	struct udp_iter_state *state = seq->private;
+	struct net *net = state->net;
 
 	for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) {
 		struct hlist_node *node;
 		sk_for_each(sk, node, state->hashtable + state->bucket) {
+			if (sk->sk_net != net)
+				continue;
 			if (sk->sk_family == state->family)
 				goto found;
 		}
@@ -1528,12 +1531,13 @@ found:
 static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk)
 {
 	struct udp_iter_state *state = seq->private;
+	struct net *net = state->net;
 
 	do {
 		sk = sk_next(sk);
 try_again:
 		;
-	} while (sk && sk->sk_family != state->family);
+	} while (sk && sk->sk_net != net && sk->sk_family != state->family);
 
 	if (!sk && ++state->bucket < UDP_HTABLE_SIZE) {
 		sk = sk_head(state->hashtable + state->bucket);
@@ -1582,31 +1586,51 @@ static int udp_seq_open(struct inode *inode, struct file *file)
 {
 	struct udp_seq_afinfo *afinfo = PDE(inode)->data;
 	struct seq_file *seq;
+	struct net *net;
 	int rc = -ENOMEM;
 	struct udp_iter_state *s = kzalloc(sizeof(*s), GFP_KERNEL);
 
 	if (!s)
 		goto out;
+
+	rc = -ENXIO;
+	net = get_proc_net(inode);
+	if (!net)
+		goto out_kfree;
+
 	s->family		= afinfo->family;
 	s->hashtable		= afinfo->hashtable;
 	s->seq_ops.start	= udp_seq_start;
 	s->seq_ops.next		= udp_seq_next;
 	s->seq_ops.show		= afinfo->seq_show;
 	s->seq_ops.stop		= udp_seq_stop;
+	s->net                  = net;
 
 	rc = seq_open(file, &s->seq_ops);
 	if (rc)
-		goto out_kfree;
+		goto out_put_net;
 
-	seq	     = file->private_data;
+	seq = file->private_data;
 	seq->private = s;
 out:
 	return rc;
+out_put_net:
+	put_net(net);
 out_kfree:
 	kfree(s);
 	goto out;
 }
 
+static int udp_seq_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq = file->private_data;
+	struct udp_iter_state *s = seq->private;
+
+	put_net(s->net);
+	seq_release_private(inode, file);
+	return 0;
+}
+
 /* ------------------------------------------------------------------------ */
 int udp_proc_register(struct udp_seq_afinfo *afinfo)
 {
@@ -1619,7 +1643,7 @@ int udp_proc_register(struct udp_seq_afinfo *afinfo)
 	afinfo->seq_fops->open		= udp_seq_open;
 	afinfo->seq_fops->read		= seq_read;
 	afinfo->seq_fops->llseek	= seq_lseek;
-	afinfo->seq_fops->release	= seq_release_private;
+	afinfo->seq_fops->release	= udp_seq_release;
 
 	p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops);
 	if (p)
-- 
cgit v1.2.3


From f40c8174d3c21bf178283f3ef3aa8c7bf238fdec Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 21 Mar 2008 04:13:54 -0700
Subject: [NETNS][IPV4] tcp - make proc handle the network namespaces

This patch, like udp proc, makes the proc functions to take care of
which namespace the socket belongs.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   |  1 +
 net/ipv4/tcp_ipv4.c | 44 +++++++++++++++++++++++++++++++++++---------
 2 files changed, 36 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 11119e33acfe..6b08dab1b1fa 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1328,6 +1328,7 @@ struct tcp_seq_afinfo {
 };
 
 struct tcp_iter_state {
+	struct net              *net;
 	sa_family_t		family;
 	enum tcp_seq_states	state;
 	struct sock		*syn_wait_sk;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a79e324638eb..f9b30dc3bd6c 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1948,6 +1948,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur)
 	struct hlist_node *node;
 	struct sock *sk = cur;
 	struct tcp_iter_state* st = seq->private;
+	struct net *net = st->net;
 
 	if (!sk) {
 		st->bucket = 0;
@@ -1964,7 +1965,8 @@ static void *listening_get_next(struct seq_file *seq, void *cur)
 		req = req->dl_next;
 		while (1) {
 			while (req) {
-				if (req->rsk_ops->family == st->family) {
+				if (req->rsk_ops->family == st->family &&
+				    req->sk->sk_net == net) {
 					cur = req;
 					goto out;
 				}
@@ -1988,7 +1990,7 @@ get_req:
 	}
 get_sk:
 	sk_for_each_from(sk, node) {
-		if (sk->sk_family == st->family) {
+		if (sk->sk_family == st->family && sk->sk_net == net) {
 			cur = sk;
 			goto out;
 		}
@@ -2027,6 +2029,7 @@ static void *listening_get_idx(struct seq_file *seq, loff_t *pos)
 static void *established_get_first(struct seq_file *seq)
 {
 	struct tcp_iter_state* st = seq->private;
+	struct net *net = st->net;
 	void *rc = NULL;
 
 	for (st->bucket = 0; st->bucket < tcp_hashinfo.ehash_size; ++st->bucket) {
@@ -2037,7 +2040,8 @@ static void *established_get_first(struct seq_file *seq)
 
 		read_lock_bh(lock);
 		sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) {
-			if (sk->sk_family != st->family) {
+			if (sk->sk_family != st->family ||
+			    sk->sk_net != net) {
 				continue;
 			}
 			rc = sk;
@@ -2046,7 +2050,8 @@ static void *established_get_first(struct seq_file *seq)
 		st->state = TCP_SEQ_STATE_TIME_WAIT;
 		inet_twsk_for_each(tw, node,
 				   &tcp_hashinfo.ehash[st->bucket].twchain) {
-			if (tw->tw_family != st->family) {
+			if (tw->tw_family != st->family &&
+			    tw->tw_net != net) {
 				continue;
 			}
 			rc = tw;
@@ -2065,6 +2070,7 @@ static void *established_get_next(struct seq_file *seq, void *cur)
 	struct inet_timewait_sock *tw;
 	struct hlist_node *node;
 	struct tcp_iter_state* st = seq->private;
+	struct net *net = st->net;
 
 	++st->num;
 
@@ -2072,7 +2078,7 @@ static void *established_get_next(struct seq_file *seq, void *cur)
 		tw = cur;
 		tw = tw_next(tw);
 get_tw:
-		while (tw && tw->tw_family != st->family) {
+		while (tw && tw->tw_family != st->family && tw->tw_net != net) {
 			tw = tw_next(tw);
 		}
 		if (tw) {
@@ -2093,7 +2099,7 @@ get_tw:
 		sk = sk_next(sk);
 
 	sk_for_each_from(sk, node) {
-		if (sk->sk_family == st->family)
+		if (sk->sk_family == st->family && sk->sk_net == net)
 			goto found;
 	}
 
@@ -2201,6 +2207,7 @@ static int tcp_seq_open(struct inode *inode, struct file *file)
 	struct tcp_seq_afinfo *afinfo = PDE(inode)->data;
 	struct seq_file *seq;
 	struct tcp_iter_state *s;
+	struct net *net;
 	int rc;
 
 	if (unlikely(afinfo == NULL))
@@ -2209,24 +2216,43 @@ static int tcp_seq_open(struct inode *inode, struct file *file)
 	s = kzalloc(sizeof(*s), GFP_KERNEL);
 	if (!s)
 		return -ENOMEM;
+
+	rc = -ENXIO;
+	net = get_proc_net(inode);
+	if (!net)
+		goto out_kfree;
+
 	s->family		= afinfo->family;
 	s->seq_ops.start	= tcp_seq_start;
 	s->seq_ops.next		= tcp_seq_next;
 	s->seq_ops.show		= afinfo->seq_show;
 	s->seq_ops.stop		= tcp_seq_stop;
+	s->net                  = net;
 
 	rc = seq_open(file, &s->seq_ops);
 	if (rc)
-		goto out_kfree;
-	seq	     = file->private_data;
+		goto out_put_net;
+	seq = file->private_data;
 	seq->private = s;
 out:
 	return rc;
+out_put_net:
+	put_net(net);
 out_kfree:
 	kfree(s);
 	goto out;
 }
 
+static int tcp_seq_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *seq = file->private_data;
+	struct tcp_iter_state *s = seq->private;
+
+	put_net(s->net);
+	seq_release_private(inode, file);
+	return 0;
+}
+
 int tcp_proc_register(struct tcp_seq_afinfo *afinfo)
 {
 	int rc = 0;
@@ -2238,7 +2264,7 @@ int tcp_proc_register(struct tcp_seq_afinfo *afinfo)
 	afinfo->seq_fops->open		= tcp_seq_open;
 	afinfo->seq_fops->read		= seq_read;
 	afinfo->seq_fops->llseek	= seq_lseek;
-	afinfo->seq_fops->release	= seq_release_private;
+	afinfo->seq_fops->release	= tcp_seq_release;
 
 	p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops);
 	if (p)
-- 
cgit v1.2.3


From 0c96d8c50bffb7f02690dd8a8cf1adb8e07e100f Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 21 Mar 2008 04:14:17 -0700
Subject: [NETNS][IPV6] udp6 - make proc per namespace

The proc init/exit functions take a new network namespace parameter in
order to register/unregister /proc/net/udp6 for a namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h  |  4 ++--
 include/net/udp.h   |  4 ++--
 net/ipv4/udp.c      | 12 ++++++------
 net/ipv4/udplite.c  |  2 +-
 net/ipv6/af_inet6.c | 19 ++++++++++++-------
 net/ipv6/udp.c      |  8 ++++----
 net/ipv6/udplite.c  |  4 ++--
 7 files changed, 29 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 8db06af1efbb..e01a563132e8 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -597,8 +597,8 @@ extern int  raw6_proc_init(void);
 extern void raw6_proc_exit(void);
 extern int  tcp6_proc_init(void);
 extern void tcp6_proc_exit(void);
-extern int  udp6_proc_init(void);
-extern void udp6_proc_exit(void);
+extern int  udp6_proc_init(struct net *net);
+extern void udp6_proc_exit(struct net *net);
 extern int  udplite6_proc_init(void);
 extern void udplite6_proc_exit(void);
 extern int  ipv6_misc_proc_init(void);
diff --git a/include/net/udp.h b/include/net/udp.h
index a1b33d667199..b4cbdce883d1 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -200,8 +200,8 @@ struct udp_iter_state {
 };
 
 #ifdef CONFIG_PROC_FS
-extern int udp_proc_register(struct udp_seq_afinfo *afinfo);
-extern void udp_proc_unregister(struct udp_seq_afinfo *afinfo);
+extern int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo);
+extern void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo);
 
 extern int  udp4_proc_init(void);
 extern void udp4_proc_exit(void);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 049e92519616..a98c43c0a89c 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1632,7 +1632,7 @@ static int udp_seq_release(struct inode *inode, struct file *file)
 }
 
 /* ------------------------------------------------------------------------ */
-int udp_proc_register(struct udp_seq_afinfo *afinfo)
+int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo)
 {
 	struct proc_dir_entry *p;
 	int rc = 0;
@@ -1645,7 +1645,7 @@ int udp_proc_register(struct udp_seq_afinfo *afinfo)
 	afinfo->seq_fops->llseek	= seq_lseek;
 	afinfo->seq_fops->release	= udp_seq_release;
 
-	p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops);
+	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops);
 	if (p)
 		p->data = afinfo;
 	else
@@ -1653,11 +1653,11 @@ int udp_proc_register(struct udp_seq_afinfo *afinfo)
 	return rc;
 }
 
-void udp_proc_unregister(struct udp_seq_afinfo *afinfo)
+void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo)
 {
 	if (!afinfo)
 		return;
-	proc_net_remove(&init_net, afinfo->name);
+	proc_net_remove(net, afinfo->name);
 	memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops));
 }
 
@@ -1709,12 +1709,12 @@ static struct udp_seq_afinfo udp4_seq_afinfo = {
 
 int __init udp4_proc_init(void)
 {
-	return udp_proc_register(&udp4_seq_afinfo);
+	return udp_proc_register(&init_net, &udp4_seq_afinfo);
 }
 
 void udp4_proc_exit(void)
 {
-	udp_proc_unregister(&udp4_seq_afinfo);
+	udp_proc_unregister(&init_net, &udp4_seq_afinfo);
 }
 #endif /* CONFIG_PROC_FS */
 
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index d49c6d68c8a9..2469b5104587 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -105,7 +105,7 @@ void __init udplite4_register(void)
 	inet_register_protosw(&udplite4_protosw);
 
 #ifdef CONFIG_PROC_FS
-	if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */
+	if (udp_proc_register(&init_net, &udplite4_seq_afinfo))
 		printk(KERN_ERR "%s: Cannot register /proc!\n", __func__);
 #endif
 	return;
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 730a861b8f41..e3e09147d134 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -842,6 +842,8 @@ static void cleanup_ipv6_mibs(void)
 
 static int inet6_net_init(struct net *net)
 {
+	int err = 0;
+
 	net->ipv6.sysctl.bindv6only = 0;
 	net->ipv6.sysctl.flush_delay = 0;
 	net->ipv6.sysctl.ip6_rt_max_size = 4096;
@@ -853,12 +855,20 @@ static int inet6_net_init(struct net *net)
 	net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40;
 	net->ipv6.sysctl.icmpv6_time = 1*HZ;
 
-	return 0;
+#ifdef CONFIG_PROC_FS
+	err = udp6_proc_init(net);
+	if (err)
+		goto out;
+out:
+#endif
+	return err;
 }
 
 static void inet6_net_exit(struct net *net)
 {
-	return;
+#ifdef CONFIG_PROC_FS
+	udp6_proc_exit(net);
+#endif
 }
 
 static struct pernet_operations inet6_net_ops = {
@@ -943,8 +953,6 @@ static int __init inet6_init(void)
 		goto proc_raw6_fail;
 	if (tcp6_proc_init())
 		goto proc_tcp6_fail;
-	if (udp6_proc_init())
-		goto proc_udp6_fail;
 	if (udplite6_proc_init())
 		goto proc_udplite6_fail;
 	if (ipv6_misc_proc_init())
@@ -1029,8 +1037,6 @@ proc_anycast6_fail:
 proc_misc6_fail:
 	udplite6_proc_exit();
 proc_udplite6_fail:
-	udp6_proc_exit();
-proc_udp6_fail:
 	tcp6_proc_exit();
 proc_tcp6_fail:
 	raw6_proc_exit();
@@ -1092,7 +1098,6 @@ static void __exit inet6_exit(void)
 	ac6_proc_exit();
 	ipv6_misc_proc_exit();
 	udplite6_proc_exit();
-	udp6_proc_exit();
 	tcp6_proc_exit();
 	raw6_proc_exit();
 #endif
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index d6e311f6c8eb..af619d48ba80 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -989,13 +989,13 @@ static struct udp_seq_afinfo udp6_seq_afinfo = {
 	.seq_fops	= &udp6_seq_fops,
 };
 
-int __init udp6_proc_init(void)
+int udp6_proc_init(struct net *net)
 {
-	return udp_proc_register(&udp6_seq_afinfo);
+	return udp_proc_register(net, &udp6_seq_afinfo);
 }
 
-void udp6_proc_exit(void) {
-	udp_proc_unregister(&udp6_seq_afinfo);
+void udp6_proc_exit(struct net *net) {
+	udp_proc_unregister(net, &udp6_seq_afinfo);
 }
 #endif /* CONFIG_PROC_FS */
 
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 87d4202522ee..815190be528c 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -115,11 +115,11 @@ static struct udp_seq_afinfo udplite6_seq_afinfo = {
 
 int __init udplite6_proc_init(void)
 {
-	return udp_proc_register(&udplite6_seq_afinfo);
+	return udp_proc_register(&init_net, &udplite6_seq_afinfo);
 }
 
 void udplite6_proc_exit(void)
 {
-	udp_proc_unregister(&udplite6_seq_afinfo);
+	udp_proc_unregister(&init_net, &udplite6_seq_afinfo);
 }
 #endif
-- 
cgit v1.2.3


From 6f8b13bcb3369a5df2e63acc422bed6098f5b8c4 Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Fri, 21 Mar 2008 04:14:45 -0700
Subject: [NETNS][IPV6] tcp6 - make proc per namespace

Make the proc for tcp6 to be per namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h  |  4 ++--
 include/net/tcp.h   |  4 ++--
 net/ipv4/tcp_ipv4.c | 12 ++++++------
 net/ipv6/af_inet6.c | 15 ++++++++++-----
 net/ipv6/tcp_ipv6.c |  8 ++++----
 5 files changed, 24 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index e01a563132e8..e82f1814d96b 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -595,8 +595,8 @@ extern int  ac6_proc_init(void);
 extern void ac6_proc_exit(void);
 extern int  raw6_proc_init(void);
 extern void raw6_proc_exit(void);
-extern int  tcp6_proc_init(void);
-extern void tcp6_proc_exit(void);
+extern int  tcp6_proc_init(struct net *net);
+extern void tcp6_proc_exit(struct net *net);
 extern int  udp6_proc_init(struct net *net);
 extern void udp6_proc_exit(struct net *net);
 extern int  udplite6_proc_init(void);
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 6b08dab1b1fa..847e1634e1f4 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1336,8 +1336,8 @@ struct tcp_iter_state {
 	struct seq_operations	seq_ops;
 };
 
-extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo);
-extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo);
+extern int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo);
+extern void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo);
 
 extern struct request_sock_ops tcp_request_sock_ops;
 extern struct request_sock_ops tcp6_request_sock_ops;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index f9b30dc3bd6c..744bc9d6cebc 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2253,7 +2253,7 @@ static int tcp_seq_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-int tcp_proc_register(struct tcp_seq_afinfo *afinfo)
+int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo)
 {
 	int rc = 0;
 	struct proc_dir_entry *p;
@@ -2266,7 +2266,7 @@ int tcp_proc_register(struct tcp_seq_afinfo *afinfo)
 	afinfo->seq_fops->llseek	= seq_lseek;
 	afinfo->seq_fops->release	= tcp_seq_release;
 
-	p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops);
+	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops);
 	if (p)
 		p->data = afinfo;
 	else
@@ -2274,11 +2274,11 @@ int tcp_proc_register(struct tcp_seq_afinfo *afinfo)
 	return rc;
 }
 
-void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo)
+void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo)
 {
 	if (!afinfo)
 		return;
-	proc_net_remove(&init_net, afinfo->name);
+	proc_net_remove(net, afinfo->name);
 	memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops));
 }
 
@@ -2419,12 +2419,12 @@ static struct tcp_seq_afinfo tcp4_seq_afinfo = {
 
 int __init tcp4_proc_init(void)
 {
-	return tcp_proc_register(&tcp4_seq_afinfo);
+	return tcp_proc_register(&init_net, &tcp4_seq_afinfo);
 }
 
 void tcp4_proc_exit(void)
 {
-	tcp_proc_unregister(&tcp4_seq_afinfo);
+	tcp_proc_unregister(&init_net, &tcp4_seq_afinfo);
 }
 #endif /* CONFIG_PROC_FS */
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e3e09147d134..f52bdaed8a1b 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -859,15 +859,25 @@ static int inet6_net_init(struct net *net)
 	err = udp6_proc_init(net);
 	if (err)
 		goto out;
+	err = tcp6_proc_init(net);
+	if (err)
+		goto proc_tcp6_fail;
 out:
 #endif
 	return err;
+
+#ifdef CONFIG_PROC_FS
+proc_tcp6_fail:
+	udp6_proc_exit(net);
+	goto out;
+#endif
 }
 
 static void inet6_net_exit(struct net *net)
 {
 #ifdef CONFIG_PROC_FS
 	udp6_proc_exit(net);
+	tcp6_proc_exit(net);
 #endif
 }
 
@@ -951,8 +961,6 @@ static int __init inet6_init(void)
 	err = -ENOMEM;
 	if (raw6_proc_init())
 		goto proc_raw6_fail;
-	if (tcp6_proc_init())
-		goto proc_tcp6_fail;
 	if (udplite6_proc_init())
 		goto proc_udplite6_fail;
 	if (ipv6_misc_proc_init())
@@ -1037,8 +1045,6 @@ proc_anycast6_fail:
 proc_misc6_fail:
 	udplite6_proc_exit();
 proc_udplite6_fail:
-	tcp6_proc_exit();
-proc_tcp6_fail:
 	raw6_proc_exit();
 proc_raw6_fail:
 #endif
@@ -1098,7 +1104,6 @@ static void __exit inet6_exit(void)
 	ac6_proc_exit();
 	ipv6_misc_proc_exit();
 	udplite6_proc_exit();
-	tcp6_proc_exit();
 	raw6_proc_exit();
 #endif
 	ipv6_netfilter_fini();
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index caf0cc1c00e1..56d0cea7d578 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2129,14 +2129,14 @@ static struct tcp_seq_afinfo tcp6_seq_afinfo = {
 	.seq_fops	= &tcp6_seq_fops,
 };
 
-int __init tcp6_proc_init(void)
+int tcp6_proc_init(struct net *net)
 {
-	return tcp_proc_register(&tcp6_seq_afinfo);
+	return tcp_proc_register(net, &tcp6_seq_afinfo);
 }
 
-void tcp6_proc_exit(void)
+void tcp6_proc_exit(struct net *net)
 {
-	tcp_proc_unregister(&tcp6_seq_afinfo);
+	tcp_proc_unregister(net, &tcp6_seq_afinfo);
 }
 #endif
 
-- 
cgit v1.2.3


From 4cd9029d25f6612302f82634620f107c65f790b1 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Fri, 21 Mar 2008 15:54:53 -0700
Subject: socket: SOCK_DEBUG type checking

Use the inline trick (same as pr_debug) to get checking of debug
statements even if no code is generated.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 8358fff002eb..b89680d2693b 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -70,7 +70,11 @@
 #define SOCK_DEBUG(sk, msg...) do { if ((sk) && sock_flag((sk), SOCK_DBG)) \
 					printk(KERN_DEBUG msg); } while (0)
 #else
-#define SOCK_DEBUG(sk, msg...) do { } while (0)
+/* Validate arguments and do nothing */
+static void inline int __attribute__ ((format (printf, 2, 3)))
+SOCK_DEBUG(struct sock *sk, const char *msg, ...)
+{
+}
 #endif
 
 /* This is the per-socket lock.  The spinlock provides a synchronization
-- 
cgit v1.2.3


From ec3c0982a2dd1e671bad8e9d26c28dcba0039d87 Mon Sep 17 00:00:00 2001
From: Patrick McManus <mcmanus@ducksong.com>
Date: Fri, 21 Mar 2008 16:33:01 -0700
Subject: [TCP]: TCP_DEFER_ACCEPT updates - process as established

Change TCP_DEFER_ACCEPT implementation so that it transitions a
connection to ESTABLISHED after handshake is complete instead of
leaving it in SYN-RECV until some data arrvies. Place connection in
accept queue when first data packet arrives from slow path.

Benefits:
  - established connection is now reset if it never makes it
   to the accept queue

 - diagnostic state of established matches with the packet traces
   showing completed handshake

 - TCP_DEFER_ACCEPT timeouts are expressed in seconds and can now be
   enforced with reasonable accuracy instead of rounding up to next
   exponential back-off of syn-ack retry.

Signed-off-by: Patrick McManus <mcmanus@ducksong.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h             |  7 +++++++
 include/net/request_sock.h      |  4 ++--
 include/net/tcp.h               |  1 +
 net/ipv4/inet_connection_sock.c | 11 +++-------
 net/ipv4/tcp.c                  | 18 +++++++---------
 net/ipv4/tcp_input.c            | 46 +++++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp_ipv4.c             |  8 +++++++
 net/ipv4/tcp_minisocks.c        | 32 +++++++++++++++++-----------
 net/ipv4/tcp_timer.c            |  5 +++++
 9 files changed, 99 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 08027f1d7f31..d96d9b122304 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -239,6 +239,11 @@ static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req)
 	return (struct tcp_request_sock *)req;
 }
 
+struct tcp_deferred_accept_info {
+	struct sock *listen_sk;
+	struct request_sock *request;
+};
+
 struct tcp_sock {
 	/* inet_connection_sock has to be the first member of tcp_sock */
 	struct inet_connection_sock	inet_conn;
@@ -374,6 +379,8 @@ struct tcp_sock {
 	unsigned int		keepalive_intvl;  /* time interval between keep alive probes */
 	int			linger2;
 
+	struct tcp_deferred_accept_info defer_tcp_accept;
+
 	unsigned long last_synq_overflow; 
 
 	u32	tso_deferred;
diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index 040780add355..0369f98e9f3a 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -115,8 +115,8 @@ struct request_sock_queue {
 	struct request_sock	*rskq_accept_head;
 	struct request_sock	*rskq_accept_tail;
 	rwlock_t		syn_wait_lock;
-	u8			rskq_defer_accept;
-	/* 3 bytes hole, try to pack */
+	u16			rskq_defer_accept;
+	/* 2 bytes hole, try to pack */
 	struct listen_sock	*listen_opt;
 };
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 847e1634e1f4..67cc3956d29c 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -139,6 +139,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo);
 #define MAX_TCP_KEEPINTVL	32767
 #define MAX_TCP_KEEPCNT		127
 #define MAX_TCP_SYNCNT		127
+#define MAX_TCP_ACCEPT_DEFERRED 65535
 
 #define TCP_SYNQ_INTERVAL	(HZ/5)	/* Period of SYNACK timer */
 
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 8a45be988709..cc1a1859a61b 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -414,8 +414,7 @@ void inet_csk_reqsk_queue_prune(struct sock *parent,
 	struct inet_connection_sock *icsk = inet_csk(parent);
 	struct request_sock_queue *queue = &icsk->icsk_accept_queue;
 	struct listen_sock *lopt = queue->listen_opt;
-	int max_retries = icsk->icsk_syn_retries ? : sysctl_tcp_synack_retries;
-	int thresh = max_retries;
+	int thresh = icsk->icsk_syn_retries ? : sysctl_tcp_synack_retries;
 	unsigned long now = jiffies;
 	struct request_sock **reqp, *req;
 	int i, budget;
@@ -451,9 +450,6 @@ void inet_csk_reqsk_queue_prune(struct sock *parent,
 		}
 	}
 
-	if (queue->rskq_defer_accept)
-		max_retries = queue->rskq_defer_accept;
-
 	budget = 2 * (lopt->nr_table_entries / (timeout / interval));
 	i = lopt->clock_hand;
 
@@ -461,9 +457,8 @@ void inet_csk_reqsk_queue_prune(struct sock *parent,
 		reqp=&lopt->syn_table[i];
 		while ((req = *reqp) != NULL) {
 			if (time_after_eq(now, req->expires)) {
-				if ((req->retrans < (inet_rsk(req)->acked ? max_retries : thresh)) &&
-				    (inet_rsk(req)->acked ||
-				     !req->rsk_ops->rtx_syn_ack(parent, req))) {
+				if (req->retrans < thresh &&
+				    !req->rsk_ops->rtx_syn_ack(parent, req)) {
 					unsigned long timeo;
 
 					if (req->retrans++ == 0)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 071e83a894ad..e0fbc25ca816 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2105,15 +2105,12 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 		break;
 
 	case TCP_DEFER_ACCEPT:
-		icsk->icsk_accept_queue.rskq_defer_accept = 0;
-		if (val > 0) {
-			/* Translate value in seconds to number of
-			 * retransmits */
-			while (icsk->icsk_accept_queue.rskq_defer_accept < 32 &&
-			       val > ((TCP_TIMEOUT_INIT / HZ) <<
-				       icsk->icsk_accept_queue.rskq_defer_accept))
-				icsk->icsk_accept_queue.rskq_defer_accept++;
-			icsk->icsk_accept_queue.rskq_defer_accept++;
+		if (val < 0) {
+			err = -EINVAL;
+		} else {
+			if (val > MAX_TCP_ACCEPT_DEFERRED)
+				val = MAX_TCP_ACCEPT_DEFERRED;
+			icsk->icsk_accept_queue.rskq_defer_accept = val;
 		}
 		break;
 
@@ -2295,8 +2292,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 			val = (val ? : sysctl_tcp_fin_timeout) / HZ;
 		break;
 	case TCP_DEFER_ACCEPT:
-		val = !icsk->icsk_accept_queue.rskq_defer_accept ? 0 :
-			((TCP_TIMEOUT_INIT / HZ) << (icsk->icsk_accept_queue.rskq_defer_accept - 1));
+		val = icsk->icsk_accept_queue.rskq_defer_accept;
 		break;
 	case TCP_WINDOW_CLAMP:
 		val = tp->window_clamp;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 9cf446427cc2..6e46b4c0f28c 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4451,6 +4451,49 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, struct tcphdr *th)
 	}
 }
 
+static int tcp_defer_accept_check(struct sock *sk)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	if (tp->defer_tcp_accept.request) {
+		int queued_data =  tp->rcv_nxt - tp->copied_seq;
+		int hasfin =  !skb_queue_empty(&sk->sk_receive_queue) ?
+			tcp_hdr((struct sk_buff *)
+				sk->sk_receive_queue.prev)->fin : 0;
+
+		if (queued_data && hasfin)
+			queued_data--;
+
+		if (queued_data &&
+		    tp->defer_tcp_accept.listen_sk->sk_state == TCP_LISTEN) {
+			if (sock_flag(sk, SOCK_KEEPOPEN)) {
+				inet_csk_reset_keepalive_timer(sk,
+							       keepalive_time_when(tp));
+			} else {
+				inet_csk_delete_keepalive_timer(sk);
+			}
+
+			inet_csk_reqsk_queue_add(
+				tp->defer_tcp_accept.listen_sk,
+				tp->defer_tcp_accept.request,
+				sk);
+
+			tp->defer_tcp_accept.listen_sk->sk_data_ready(
+				tp->defer_tcp_accept.listen_sk, 0);
+
+			sock_put(tp->defer_tcp_accept.listen_sk);
+			sock_put(sk);
+			tp->defer_tcp_accept.listen_sk = NULL;
+			tp->defer_tcp_accept.request = NULL;
+		} else if (hasfin ||
+			   tp->defer_tcp_accept.listen_sk->sk_state != TCP_LISTEN) {
+			tcp_reset(sk);
+			return -1;
+		}
+	}
+	return 0;
+}
+
 static int tcp_copy_to_iovec(struct sock *sk, struct sk_buff *skb, int hlen)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
@@ -4811,6 +4854,9 @@ step5:
 
 	tcp_data_snd_check(sk);
 	tcp_ack_snd_check(sk);
+
+	if (tcp_defer_accept_check(sk))
+		return -1;
 	return 0;
 
 csum_error:
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 0ba6e911c979..167a0f557531 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1920,6 +1920,14 @@ int tcp_v4_destroy_sock(struct sock *sk)
 		sk->sk_sndmsg_page = NULL;
 	}
 
+	if (tp->defer_tcp_accept.request) {
+		reqsk_free(tp->defer_tcp_accept.request);
+		sock_put(tp->defer_tcp_accept.listen_sk);
+		sock_put(sk);
+		tp->defer_tcp_accept.listen_sk = NULL;
+		tp->defer_tcp_accept.request = NULL;
+	}
+
 	atomic_dec(&tcp_sockets_allocated);
 
 	return 0;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 8245247a6ceb..019c8c16e5cc 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -571,10 +571,8 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
 	   does sequence test, SYN is truncated, and thus we consider
 	   it a bare ACK.
 
-	   If icsk->icsk_accept_queue.rskq_defer_accept, we silently drop this
-	   bare ACK.  Otherwise, we create an established connection.  Both
-	   ends (listening sockets) accept the new incoming connection and try
-	   to talk to each other. 8-)
+	   Both ends (listening sockets) accept the new incoming
+	   connection and try to talk to each other. 8-)
 
 	   Note: This case is both harmless, and rare.  Possibility is about the
 	   same as us discovering intelligent life on another plant tomorrow.
@@ -642,13 +640,6 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
 		if (!(flg & TCP_FLAG_ACK))
 			return NULL;
 
-		/* If TCP_DEFER_ACCEPT is set, drop bare ACK. */
-		if (inet_csk(sk)->icsk_accept_queue.rskq_defer_accept &&
-		    TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) {
-			inet_rsk(req)->acked = 1;
-			return NULL;
-		}
-
 		/* OK, ACK is valid, create big socket and
 		 * feed this segment to it. It will repeat all
 		 * the tests. THIS SEGMENT MUST MOVE SOCKET TO
@@ -687,7 +678,24 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb,
 		inet_csk_reqsk_queue_unlink(sk, req, prev);
 		inet_csk_reqsk_queue_removed(sk, req);
 
-		inet_csk_reqsk_queue_add(sk, req, child);
+		if (inet_csk(sk)->icsk_accept_queue.rskq_defer_accept &&
+		    TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) {
+
+			/* the accept queue handling is done is est recv slow
+			 * path so lets make sure to start there
+			 */
+			tcp_sk(child)->pred_flags = 0;
+			sock_hold(sk);
+			sock_hold(child);
+			tcp_sk(child)->defer_tcp_accept.listen_sk = sk;
+			tcp_sk(child)->defer_tcp_accept.request = req;
+
+			inet_csk_reset_keepalive_timer(child,
+						       inet_csk(sk)->icsk_accept_queue.rskq_defer_accept * HZ);
+		} else {
+			inet_csk_reqsk_queue_add(sk, req, child);
+		}
+
 		return child;
 
 	listen_overflow:
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 803d758a2b12..160d16f9f4fc 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -481,6 +481,11 @@ static void tcp_keepalive_timer (unsigned long data)
 		goto death;
 	}
 
+	if (tp->defer_tcp_accept.request && sk->sk_state == TCP_ESTABLISHED) {
+		tcp_send_active_reset(sk, GFP_ATOMIC);
+		goto death;
+	}
+
 	if (!sock_flag(sk, SOCK_KEEPOPEN) || sk->sk_state == TCP_CLOSE)
 		goto out;
 
-- 
cgit v1.2.3


From ef722495c8867aacc1db0675a6737e5cf1e72e07 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sat, 22 Mar 2008 16:35:29 -0700
Subject: [IPV4]: Remove unused ip_options->is_data.

ip_options->is_data is assigned only and never checked. The structure is
not a part of kernel interface to the userspace. So, it is safe to remove
this field.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h | 3 +--
 net/ipv4/cipso_ipv4.c   | 1 -
 net/ipv4/ip_options.c   | 5 -----
 3 files changed, 1 insertion(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 8660cb0fa0dd..b6db16d2766a 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -43,8 +43,7 @@ struct ip_options {
 	unsigned char	srr;
 	unsigned char	rr;
 	unsigned char	ts;
-	unsigned char	is_data:1,
-			is_strictroute:1,
+	unsigned char	is_strictroute:1,
 			srr_is_hit:1,
 			is_changed:1,
 			rr_needaddr:1,
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 8cd357f41283..4637ded3dba8 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -1800,7 +1800,6 @@ int cipso_v4_sock_setattr(struct sock *sk,
 	}
 	memcpy(opt->__data, buf, buf_len);
 	opt->optlen = opt_len;
-	opt->is_data = 1;
 	opt->cipso = sizeof(struct iphdr);
 	kfree(buf);
 	buf = NULL;
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 7e94bf850a07..2fa411381b11 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -45,7 +45,6 @@ void ip_options_build(struct sk_buff * skb, struct ip_options * opt,
 	memcpy(&(IPCB(skb)->opt), opt, sizeof(struct ip_options));
 	memcpy(iph+sizeof(struct iphdr), opt->__data, opt->optlen);
 	opt = &(IPCB(skb)->opt);
-	opt->is_data = 0;
 
 	if (opt->srr)
 		memcpy(iph+opt->srr+iph[opt->srr+1]-4, &daddr, 4);
@@ -95,8 +94,6 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb)
 
 	memset(dopt, 0, sizeof(struct ip_options));
 
-	dopt->is_data = 1;
-
 	sopt = &(IPCB(skb)->opt);
 
 	if (sopt->optlen == 0) {
@@ -265,7 +262,6 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb)
 		iph = skb_network_header(skb);
 		opt->optlen = ((struct iphdr *)iph)->ihl*4 - sizeof(struct iphdr);
 		optptr = iph + sizeof(struct iphdr);
-		opt->is_data = 0;
 	} else {
 		optptr = opt->__data;
 		iph = optptr - sizeof(struct iphdr);
@@ -519,7 +515,6 @@ static int ip_options_get_finish(struct ip_options **optp,
 	while (optlen & 3)
 		opt->__data[optlen++] = IPOPT_END;
 	opt->optlen = optlen;
-	opt->is_data = 1;
 	if (optlen && ip_options_compile(opt, NULL)) {
 		kfree(opt);
 		return -EINVAL;
-- 
cgit v1.2.3


From 39d8cda76cfb1178455f9d196b39e773878e6c05 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 22 Mar 2008 16:50:58 -0700
Subject: [SOCK]: Add udp_hash member to struct proto.

Inspired by the commit ab1e0a13 ([SOCK] proto: Add hashinfo member to
struct proto) from Arnaldo, I made similar thing for UDP/-Lite IPv4
and -v6 protocols.

The result is not that exciting, but it removes some levels of
indirection in udpxxx_get_port and saves some space in code and text.

The first step is to union existing hashinfo and new udp_hash on the
struct proto and give a name to this union, since future initialization
of tcpxxx_prot, dccp_vx_protinfo and udpxxx_protinfo will cause gcc
warning about inability to initialize anonymous member this way.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h   | 2 +-
 include/net/sock.h              | 5 ++++-
 net/dccp/ipv4.c                 | 2 +-
 net/dccp/ipv6.c                 | 2 +-
 net/ipv4/inet_connection_sock.c | 2 +-
 net/ipv4/inet_hashtables.c      | 8 ++++----
 net/ipv4/tcp_ipv4.c             | 2 +-
 net/ipv6/inet6_hashtables.c     | 2 +-
 net/ipv6/tcp_ipv6.c             | 2 +-
 9 files changed, 15 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 97dc35ad09be..d99c1ba2ece0 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -223,7 +223,7 @@ static inline int inet_sk_listen_hashfn(const struct sock *sk)
 /* Caller must disable local BH processing. */
 static inline void __inet_inherit_port(struct sock *sk, struct sock *child)
 {
-	struct inet_hashinfo *table = sk->sk_prot->hashinfo;
+	struct inet_hashinfo *table = sk->sk_prot->h.hashinfo;
 	const int bhash = inet_bhashfn(inet_sk(child)->num, table->bhash_size);
 	struct inet_bind_hashbucket *head = &table->bhash[bhash];
 	struct inet_bind_bucket *tb;
diff --git a/include/net/sock.h b/include/net/sock.h
index b89680d2693b..c3175c400b79 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -586,7 +586,10 @@ struct proto {
 	struct request_sock_ops	*rsk_prot;
 	struct timewait_sock_ops *twsk_prot;
 
-	struct inet_hashinfo	*hashinfo;
+	union {
+		struct inet_hashinfo	*hashinfo;
+		struct hlist_head	*udp_hash;
+	} h;
 
 	struct module		*owner;
 
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 17ad69e90e48..4ca8b0c93c80 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -941,7 +941,7 @@ static struct proto dccp_v4_prot = {
 	.obj_size		= sizeof(struct dccp_sock),
 	.rsk_prot		= &dccp_request_sock_ops,
 	.twsk_prot		= &dccp_timewait_sock_ops,
-	.hashinfo		= &dccp_hashinfo,
+	.h.hashinfo		= &dccp_hashinfo,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt	= compat_dccp_setsockopt,
 	.compat_getsockopt	= compat_dccp_getsockopt,
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 1a5e50b90677..2fec1af1a8c3 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -1126,7 +1126,7 @@ static struct proto dccp_v6_prot = {
 	.obj_size	   = sizeof(struct dccp6_sock),
 	.rsk_prot	   = &dccp6_request_sock_ops,
 	.twsk_prot	   = &dccp6_timewait_sock_ops,
-	.hashinfo	   = &dccp_hashinfo,
+	.h.hashinfo	   = &dccp_hashinfo,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_dccp_setsockopt,
 	.compat_getsockopt = compat_dccp_getsockopt,
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index cc1a1859a61b..f9c5c4def1ba 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -80,7 +80,7 @@ EXPORT_SYMBOL_GPL(inet_csk_bind_conflict);
  */
 int inet_csk_get_port(struct sock *sk, unsigned short snum)
 {
-	struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo;
+	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
 	struct inet_bind_hashbucket *head;
 	struct hlist_node *node;
 	struct inet_bind_bucket *tb;
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 1aba606f6bbb..8cd1ad9b9111 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -68,7 +68,7 @@ void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
  */
 static void __inet_put_port(struct sock *sk)
 {
-	struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo;
+	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
 	const int bhash = inet_bhashfn(inet_sk(sk)->num, hashinfo->bhash_size);
 	struct inet_bind_hashbucket *head = &hashinfo->bhash[bhash];
 	struct inet_bind_bucket *tb;
@@ -318,7 +318,7 @@ static inline u32 inet_sk_port_offset(const struct sock *sk)
 
 void __inet_hash_nolisten(struct sock *sk)
 {
-	struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo;
+	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
 	struct hlist_head *list;
 	rwlock_t *lock;
 	struct inet_ehash_bucket *head;
@@ -339,7 +339,7 @@ EXPORT_SYMBOL_GPL(__inet_hash_nolisten);
 
 static void __inet_hash(struct sock *sk)
 {
-	struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo;
+	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
 	struct hlist_head *list;
 	rwlock_t *lock;
 
@@ -372,7 +372,7 @@ EXPORT_SYMBOL_GPL(inet_hash);
 void inet_unhash(struct sock *sk)
 {
 	rwlock_t *lock;
-	struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo;
+	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
 
 	if (sk_unhashed(sk))
 		goto out;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 167a0f557531..1a47719a55ad 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2468,7 +2468,7 @@ struct proto tcp_prot = {
 	.obj_size		= sizeof(struct tcp_sock),
 	.twsk_prot		= &tcp_timewait_sock_ops,
 	.rsk_prot		= &tcp_request_sock_ops,
-	.hashinfo		= &tcp_hashinfo,
+	.h.hashinfo		= &tcp_hashinfo,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt	= compat_tcp_setsockopt,
 	.compat_getsockopt	= compat_tcp_getsockopt,
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 99fd25f7f005..c0c8d2d17682 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -24,7 +24,7 @@
 
 void __inet6_hash(struct sock *sk)
 {
-	struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo;
+	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
 	struct hlist_head *list;
 	rwlock_t *lock;
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 56d0cea7d578..8dd72966ff78 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2172,7 +2172,7 @@ struct proto tcpv6_prot = {
 	.obj_size		= sizeof(struct tcp6_sock),
 	.twsk_prot		= &tcp6_timewait_sock_ops,
 	.rsk_prot		= &tcp6_request_sock_ops,
-	.hashinfo		= &tcp_hashinfo,
+	.h.hashinfo		= &tcp_hashinfo,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt	= compat_tcp_setsockopt,
 	.compat_getsockopt	= compat_tcp_getsockopt,
-- 
cgit v1.2.3


From 6ba5a3c52da00015e739469e3b00cd6d0d4c5c67 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 22 Mar 2008 16:51:21 -0700
Subject: [UDP]: Make full use of proto.h.udp_hash innovation.

After this we have only udp_lib_get_port to get the port and two
stubs for ipv4 and ipv6. No difference in udp and udplite except
for initialized h.udp_hash member.

I tried to find a graceful way to drop the only difference between
udp_v4_get_port and udp_v6_get_port (i.e. the rcv_saddr comparison
routine), but adding one more callback on the struct proto didn't
appear such :( Maybe later.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h   |  2 ++
 net/ipv4/udp.c      | 22 ++++++++--------------
 net/ipv4/udp_impl.h |  6 +-----
 net/ipv4/udplite.c  | 15 ++-------------
 net/ipv6/udp.c      |  5 +++--
 net/ipv6/udp_impl.h |  2 ++
 net/ipv6/udplite.c  |  8 ++------
 7 files changed, 20 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index b4cbdce883d1..635940d374ab 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -125,6 +125,8 @@ static inline void udp_lib_close(struct sock *sk, long timeout)
 	sk_common_release(sk);
 }
 
+extern int	udp_lib_get_port(struct sock *sk, unsigned short snum,
+		int (*)(const struct sock*,const struct sock*));
 
 /* net/ipv4/udp.c */
 extern int	udp_get_port(struct sock *sk, unsigned short snum,
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index fa946829d1e8..8c1f5eaafcd2 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -143,18 +143,17 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
 }
 
 /**
- *  __udp_lib_get_port  -  UDP/-Lite port lookup for IPv4 and IPv6
+ *  udp_lib_get_port  -  UDP/-Lite port lookup for IPv4 and IPv6
  *
  *  @sk:          socket struct in question
  *  @snum:        port number to look up
- *  @udptable:    hash list table, must be of UDP_HTABLE_SIZE
  *  @saddr_comp:  AF-dependent comparison of bound local IP addresses
  */
-int __udp_lib_get_port(struct sock *sk, unsigned short snum,
-		       struct hlist_head udptable[],
+int udp_lib_get_port(struct sock *sk, unsigned short snum,
 		       int (*saddr_comp)(const struct sock *sk1,
 					 const struct sock *sk2 )    )
 {
+	struct hlist_head *udptable = sk->sk_prot->h.udp_hash;
 	struct hlist_node *node;
 	struct hlist_head *head;
 	struct sock *sk2;
@@ -240,13 +239,7 @@ fail:
 	return error;
 }
 
-int udp_get_port(struct sock *sk, unsigned short snum,
-			int (*scmp)(const struct sock *, const struct sock *))
-{
-	return  __udp_lib_get_port(sk, snum, udp_hash, scmp);
-}
-
-int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
+static int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
 {
 	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);
 
@@ -255,9 +248,9 @@ int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
 		   inet1->rcv_saddr == inet2->rcv_saddr      ));
 }
 
-static inline int udp_v4_get_port(struct sock *sk, unsigned short snum)
+int udp_v4_get_port(struct sock *sk, unsigned short snum)
 {
-	return udp_get_port(sk, snum, ipv4_rcv_saddr_equal);
+	return udp_lib_get_port(sk, snum, ipv4_rcv_saddr_equal);
 }
 
 /* UDP is nearly always wildcards out the wazoo, it makes no sense to try
@@ -1498,6 +1491,7 @@ struct proto udp_prot = {
 	.sysctl_wmem	   = &sysctl_udp_wmem_min,
 	.sysctl_rmem	   = &sysctl_udp_rmem_min,
 	.obj_size	   = sizeof(struct udp_sock),
+	.h.udp_hash	   = udp_hash,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_udp_setsockopt,
 	.compat_getsockopt = compat_udp_getsockopt,
@@ -1741,12 +1735,12 @@ EXPORT_SYMBOL(udp_disconnect);
 EXPORT_SYMBOL(udp_hash);
 EXPORT_SYMBOL(udp_hash_lock);
 EXPORT_SYMBOL(udp_ioctl);
-EXPORT_SYMBOL(udp_get_port);
 EXPORT_SYMBOL(udp_prot);
 EXPORT_SYMBOL(udp_sendmsg);
 EXPORT_SYMBOL(udp_lib_getsockopt);
 EXPORT_SYMBOL(udp_lib_setsockopt);
 EXPORT_SYMBOL(udp_poll);
+EXPORT_SYMBOL(udp_lib_get_port);
 
 #ifdef CONFIG_PROC_FS
 EXPORT_SYMBOL(udp_proc_register);
diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h
index 6c55828e41ba..7288bf7977fb 100644
--- a/net/ipv4/udp_impl.h
+++ b/net/ipv4/udp_impl.h
@@ -8,11 +8,7 @@
 extern int  	__udp4_lib_rcv(struct sk_buff *, struct hlist_head [], int );
 extern void 	__udp4_lib_err(struct sk_buff *, u32, struct hlist_head []);
 
-extern int	__udp_lib_get_port(struct sock *sk, unsigned short snum,
-				   struct hlist_head udptable[],
-				   int (*)(const struct sock*,const struct sock*));
-extern int	ipv4_rcv_saddr_equal(const struct sock *, const struct sock *);
-
+extern int	udp_v4_get_port(struct sock *sk, unsigned short snum);
 
 extern int	udp_setsockopt(struct sock *sk, int level, int optname,
 			       char __user *optval, int optlen);
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 2469b5104587..8d42e344b043 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -17,17 +17,6 @@ DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics)	__read_mostly;
 
 struct hlist_head 	udplite_hash[UDP_HTABLE_SIZE];
 
-int udplite_get_port(struct sock *sk, unsigned short p,
-		     int (*c)(const struct sock *, const struct sock *))
-{
-	return  __udp_lib_get_port(sk, p, udplite_hash, c);
-}
-
-static int udplite_v4_get_port(struct sock *sk, unsigned short snum)
-{
-	return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal);
-}
-
 static int udplite_rcv(struct sk_buff *skb)
 {
 	return __udp4_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE);
@@ -63,8 +52,9 @@ struct proto 	udplite_prot = {
 	.backlog_rcv	   = udp_queue_rcv_skb,
 	.hash		   = udp_lib_hash,
 	.unhash		   = udp_lib_unhash,
-	.get_port	   = udplite_v4_get_port,
+	.get_port	   = udp_v4_get_port,
 	.obj_size	   = sizeof(struct udp_sock),
+	.h.udp_hash	   = udplite_hash,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_udp_setsockopt,
 	.compat_getsockopt = compat_udp_getsockopt,
@@ -118,4 +108,3 @@ out_register_err:
 
 EXPORT_SYMBOL(udplite_hash);
 EXPORT_SYMBOL(udplite_prot);
-EXPORT_SYMBOL(udplite_get_port);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index af619d48ba80..5f5d1218c34e 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -51,9 +51,9 @@
 #include <linux/seq_file.h>
 #include "udp_impl.h"
 
-static inline int udp_v6_get_port(struct sock *sk, unsigned short snum)
+int udp_v6_get_port(struct sock *sk, unsigned short snum)
 {
-	return udp_get_port(sk, snum, ipv6_rcv_saddr_equal);
+	return udp_lib_get_port(sk, snum, ipv6_rcv_saddr_equal);
 }
 
 static struct sock *__udp6_lib_lookup(struct net *net,
@@ -1024,6 +1024,7 @@ struct proto udpv6_prot = {
 	.sysctl_wmem	   = &sysctl_udp_wmem_min,
 	.sysctl_rmem	   = &sysctl_udp_rmem_min,
 	.obj_size	   = sizeof(struct udp6_sock),
+	.h.udp_hash	   = udp_hash,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_udpv6_setsockopt,
 	.compat_getsockopt = compat_udpv6_getsockopt,
diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h
index 21be3a83e7bc..321b81a4d418 100644
--- a/net/ipv6/udp_impl.h
+++ b/net/ipv6/udp_impl.h
@@ -11,6 +11,8 @@ extern int  	__udp6_lib_rcv(struct sk_buff *, struct hlist_head [], int );
 extern void 	__udp6_lib_err(struct sk_buff *, struct inet6_skb_parm *,
 			       int , int , int , __be32 , struct hlist_head []);
 
+extern int	udp_v6_get_port(struct sock *sk, unsigned short snum);
+
 extern int	udpv6_getsockopt(struct sock *sk, int level, int optname,
 				 char __user *optval, int __user *optlen);
 extern int	udpv6_setsockopt(struct sock *sk, int level, int optname,
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 815190be528c..93e52e0d57f2 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -35,11 +35,6 @@ static struct inet6_protocol udplitev6_protocol = {
 	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
 };
 
-static int udplite_v6_get_port(struct sock *sk, unsigned short snum)
-{
-	return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal);
-}
-
 DEFINE_PROTO_INUSE(udplitev6)
 
 struct proto udplitev6_prot = {
@@ -58,8 +53,9 @@ struct proto udplitev6_prot = {
 	.backlog_rcv	   = udpv6_queue_rcv_skb,
 	.hash		   = udp_lib_hash,
 	.unhash		   = udp_lib_unhash,
-	.get_port	   = udplite_v6_get_port,
+	.get_port	   = udp_v6_get_port,
 	.obj_size	   = sizeof(struct udp6_sock),
+	.h.udp_hash	   = udplite_hash,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_udpv6_setsockopt,
 	.compat_getsockopt = compat_udpv6_getsockopt,
-- 
cgit v1.2.3


From fc8717baa8f52dd8d1b90df9008300ef3ec794ed Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sat, 22 Mar 2008 16:56:51 -0700
Subject: [RAW]: Add raw_hashinfo member on struct proto.

Sorry for the patch sequence confusion :| but I found that the similar
thing can be done for raw sockets easily too late.

Expand the proto.h union with the raw_hashinfo member and use it in
raw_prot and rawv6_prot. This allows to drop the protocol specific
versions of hash and unhash callbacks.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/raw.h  |  4 ++--
 include/net/sock.h |  2 ++
 net/ipv4/raw.c     | 22 ++++++++--------------
 net/ipv6/raw.c     | 16 +++-------------
 4 files changed, 15 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/net/raw.h b/include/net/raw.h
index 1828f81fe374..6c14a656357a 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -53,7 +53,7 @@ int raw_seq_open(struct inode *ino, struct file *file,
 
 #endif
 
-void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h);
-void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h);
+void raw_hash_sk(struct sock *sk);
+void raw_unhash_sk(struct sock *sk);
 
 #endif	/* _RAW_H */
diff --git a/include/net/sock.h b/include/net/sock.h
index c3175c400b79..b433b1ed203d 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -504,6 +504,7 @@ extern int sk_wait_data(struct sock *sk, long *timeo);
 struct request_sock_ops;
 struct timewait_sock_ops;
 struct inet_hashinfo;
+struct raw_hashinfo;
 
 /* Networking protocol blocks we attach to sockets.
  * socket layer -> transport layer interface
@@ -589,6 +590,7 @@ struct proto {
 	union {
 		struct inet_hashinfo	*hashinfo;
 		struct hlist_head	*udp_hash;
+		struct raw_hashinfo	*raw_hash;
 	} h;
 
 	struct module		*owner;
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index e10ef6c2202a..b433b485a883 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -84,8 +84,9 @@ static struct raw_hashinfo raw_v4_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(raw_v4_hashinfo.lock),
 };
 
-void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h)
+void raw_hash_sk(struct sock *sk)
 {
+	struct raw_hashinfo *h = sk->sk_prot->h.raw_hash;
 	struct hlist_head *head;
 
 	head = &h->ht[inet_sk(sk)->num & (RAW_HTABLE_SIZE - 1)];
@@ -97,8 +98,10 @@ void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h)
 }
 EXPORT_SYMBOL_GPL(raw_hash_sk);
 
-void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h)
+void raw_unhash_sk(struct sock *sk)
 {
+	struct raw_hashinfo *h = sk->sk_prot->h.raw_hash;
+
 	write_lock_bh(&h->lock);
 	if (sk_del_node_init(sk))
 		sock_prot_inuse_add(sk->sk_prot, -1);
@@ -106,16 +109,6 @@ void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h)
 }
 EXPORT_SYMBOL_GPL(raw_unhash_sk);
 
-static void raw_v4_hash(struct sock *sk)
-{
-	raw_hash_sk(sk, &raw_v4_hashinfo);
-}
-
-static void raw_v4_unhash(struct sock *sk)
-{
-	raw_unhash_sk(sk, &raw_v4_hashinfo);
-}
-
 static struct sock *__raw_v4_lookup(struct net *net, struct sock *sk,
 		unsigned short num, __be32 raddr, __be32 laddr, int dif)
 {
@@ -841,9 +834,10 @@ struct proto raw_prot = {
 	.recvmsg	   = raw_recvmsg,
 	.bind		   = raw_bind,
 	.backlog_rcv	   = raw_rcv_skb,
-	.hash		   = raw_v4_hash,
-	.unhash		   = raw_v4_unhash,
+	.hash		   = raw_hash_sk,
+	.unhash		   = raw_unhash_sk,
 	.obj_size	   = sizeof(struct raw_sock),
+	.h.raw_hash	   = &raw_v4_hashinfo,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_raw_setsockopt,
 	.compat_getsockopt = compat_raw_getsockopt,
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index d6afa0216759..a9e4235157a2 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -65,17 +65,6 @@ static struct raw_hashinfo raw_v6_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(raw_v6_hashinfo.lock),
 };
 
-static void raw_v6_hash(struct sock *sk)
-{
-	raw_hash_sk(sk, &raw_v6_hashinfo);
-}
-
-static void raw_v6_unhash(struct sock *sk)
-{
-	raw_unhash_sk(sk, &raw_v6_hashinfo);
-}
-
-
 static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk,
 		unsigned short num, struct in6_addr *loc_addr,
 		struct in6_addr *rmt_addr, int dif)
@@ -1201,9 +1190,10 @@ struct proto rawv6_prot = {
 	.recvmsg	   = rawv6_recvmsg,
 	.bind		   = rawv6_bind,
 	.backlog_rcv	   = rawv6_rcv_skb,
-	.hash		   = raw_v6_hash,
-	.unhash		   = raw_v6_unhash,
+	.hash		   = raw_hash_sk,
+	.unhash		   = raw_unhash_sk,
 	.obj_size	   = sizeof(struct raw6_sock),
+	.h.raw_hash	   = &raw_v6_hashinfo,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_rawv6_setsockopt,
 	.compat_getsockopt = compat_rawv6_getsockopt,
-- 
cgit v1.2.3


From 0098b7273e968fb9989a6e1e4e4c024cd081fe0d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Sat, 22 Mar 2008 17:18:47 -0700
Subject: [NET]: NPROTO is redundant; it's equal to AF_MAX/PF_MAX.

DaveM pointed out NPROTO exposed to userspace, so keep it around,
just make sure it stays in sync.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index c414d90e647b..71f7dd559285 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -19,6 +19,7 @@
 #define _LINUX_NET_H
 
 #include <linux/wait.h>
+#include <linux/socket.h>
 #include <asm/socket.h>
 
 struct poll_table_struct;
@@ -26,7 +27,7 @@ struct pipe_inode_info;
 struct inode;
 struct net;
 
-#define NPROTO		34		/* should be enough for now..	*/
+#define NPROTO		AF_MAX
 
 #define SYS_SOCKET	1		/* sys_socket(2)		*/
 #define SYS_BIND	2		/* sys_bind(2)			*/
-- 
cgit v1.2.3


From 7d164be8aa4392fe55474f4608547f2097e07c41 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 22:03:56 -0700
Subject: [NET]: include/net/route.h - remove duplicate include

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index eadad5901429..28dba925663c 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -34,7 +34,6 @@
 #include <linux/ip.h>
 #include <linux/cache.h>
 #include <linux/security.h>
-#include <net/sock.h>
 
 #ifndef __KERNEL__
 #warning This file is not supposed to be used outside of kernel.
-- 
cgit v1.2.3


From 414f69d8a6ff0b30e7ea5ce10534b19f851e172e Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 22:04:31 -0700
Subject: [NET]: include/linux/atalk.h - remove duplicate include

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/atalk.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/atalk.h b/include/linux/atalk.h
index ced8a1ed080c..e9ebac2e2ecc 100644
--- a/include/linux/atalk.h
+++ b/include/linux/atalk.h
@@ -85,8 +85,6 @@ static inline struct atalk_sock *at_sk(struct sock *sk)
 	return (struct atalk_sock *)sk;
 }
 
-#include <asm/byteorder.h>
-
 struct ddpehdr {
 	__be16	deh_len_hops;	/* lower 10 bits are length, next 4 - hops */
 	__be16	deh_sum;
-- 
cgit v1.2.3


From cc32e05416b4023a5466a2f66e3c02236a771c5b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 22:05:44 -0700
Subject: [NET]: include/linux/igmp.h - remove duplicate include

Removed duplicate #include <linux/skbuff.h>
Combined #ifdef __KERNEL__ blocks

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h | 38 +++++++++++++++++---------------------
 1 file changed, 17 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index f510e7e382a8..f5a1a0db2e8e 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -80,27 +80,6 @@ struct igmpv3_query {
 	__be32 srcs[0];
 };
 
-#ifdef __KERNEL__
-#include <linux/skbuff.h>
-
-static inline struct igmphdr *igmp_hdr(const struct sk_buff *skb)
-{
-	return (struct igmphdr *)skb_transport_header(skb);
-}
-
-static inline struct igmpv3_report *
-			igmpv3_report_hdr(const struct sk_buff *skb)
-{
-	return (struct igmpv3_report *)skb_transport_header(skb);
-}
-
-static inline struct igmpv3_query *
-			igmpv3_query_hdr(const struct sk_buff *skb)
-{
-	return (struct igmpv3_query *)skb_transport_header(skb);
-}
-#endif
-
 #define IGMP_HOST_MEMBERSHIP_QUERY	0x11	/* From RFC1112 */
 #define IGMP_HOST_MEMBERSHIP_REPORT	0x12	/* Ditto */
 #define IGMP_DVMRP			0x13	/* DVMRP routing */
@@ -151,6 +130,23 @@ static inline struct igmpv3_query *
 #include <linux/timer.h>
 #include <linux/in.h>
 
+static inline struct igmphdr *igmp_hdr(const struct sk_buff *skb)
+{
+	return (struct igmphdr *)skb_transport_header(skb);
+}
+
+static inline struct igmpv3_report *
+			igmpv3_report_hdr(const struct sk_buff *skb)
+{
+	return (struct igmpv3_report *)skb_transport_header(skb);
+}
+
+static inline struct igmpv3_query *
+			igmpv3_query_hdr(const struct sk_buff *skb)
+{
+	return (struct igmpv3_query *)skb_transport_header(skb);
+}
+
 extern int sysctl_igmp_max_memberships;
 extern int sysctl_igmp_max_msf;
 
-- 
cgit v1.2.3


From 310afe86af8ddd96a06b75aa61ef1af233f80e89 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 22:06:51 -0700
Subject: [NET]: include/linux/udp.h - remove duplicate include

Remove duplicate #include <linux/types.h>
Combine #ifdef __KERNEL__ blocks

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h | 18 +++++++-----------
 1 file changed, 7 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 1e7b7cb5703b..581ca2c14c52 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -26,15 +26,6 @@ struct udphdr {
 	__sum16	check;
 };
 
-#ifdef __KERNEL__
-#include <linux/skbuff.h>
-
-static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
-{
-	return (struct udphdr *)skb_transport_header(skb);
-}
-#endif
-
 /* UDP socket options */
 #define UDP_CORK	1	/* Never send partially complete segments */
 #define UDP_ENCAP	100	/* Set the socket to accept encapsulated packets */
@@ -45,9 +36,14 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 #define UDP_ENCAP_L2TPINUDP	3 /* rfc2661 */
 
 #ifdef __KERNEL__
-#include <linux/types.h>
-
 #include <net/inet_sock.h>
+#include <linux/skbuff.h>
+
+static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
+{
+	return (struct udphdr *)skb_transport_header(skb);
+}
+
 #define UDP_HTABLE_SIZE		128
 
 struct udp_sock {
-- 
cgit v1.2.3


From 2051f11fb86b0056fec440fe7e9fa8370d60a5c6 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sun, 23 Mar 2008 22:21:28 -0700
Subject: [TCP]: Shrink syncookie_secret by 8 byte.

the first u32 copied from syncookie_secret is overwritten by the
minute-counter four lines below.  After adjusting the destination
address, the size of syncookie_secret can be reduced accordingly.

AFAICS, the only other user of syncookie_secret[] is the ipv6
syncookie support.  Because ipv6 syncookies only grab 44 bytes from
syncookie_secret[], this shouldn't affect them in any way.

With fixes from Glenn Griffin.

Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Glenn Griffin <ggriffin.kernel@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h     | 2 +-
 net/ipv4/syncookies.c | 4 ++--
 net/ipv6/syncookies.c | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 67cc3956d29c..723b36851dde 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -436,7 +436,7 @@ extern int			tcp_disconnect(struct sock *sk, int flags);
 extern void			tcp_unhash(struct sock *sk);
 
 /* From syncookies.c */
-extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS];
+extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS];
 extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, 
 				    struct ip_options *opt);
 extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, 
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 4704f27f6c0b..abc752d45cf7 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -21,7 +21,7 @@
 
 extern int sysctl_tcp_syncookies;
 
-__u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS];
+__u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS];
 EXPORT_SYMBOL(syncookie_secret);
 
 static __init int init_syncookies(void)
@@ -41,7 +41,7 @@ static u32 cookie_hash(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport,
 {
 	__u32 *tmp = __get_cpu_var(cookie_scratch);
 
-	memcpy(tmp + 3, syncookie_secret[c], sizeof(syncookie_secret[c]));
+	memcpy(tmp + 4, syncookie_secret[c], sizeof(syncookie_secret[c]));
 	tmp[0] = (__force u32)saddr;
 	tmp[1] = (__force u32)daddr;
 	tmp[2] = ((__force u32)sport << 16) + (__force u32)dport;
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 827c5aa7524c..3a622e7abc02 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -22,7 +22,7 @@
 #include <net/tcp.h>
 
 extern int sysctl_tcp_syncookies;
-extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS];
+extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS];
 
 #define COOKIEBITS 24	/* Upper bits store count */
 #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1)
-- 
cgit v1.2.3


From 80445cfb28a6b093540582b68d9ae928bf34cfe7 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sun, 23 Mar 2008 22:47:08 -0700
Subject: [SCTP]: Remove redundant wrapper functions.

sctp_datamsg_free and sctp_datamsg_track are just aliases for
sctp_datamsg_put and sctp_chunk_hold, respectively.

Saves 32 Bytes on x86.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/structs.h |  2 --
 net/sctp/chunk.c           | 16 +---------------
 net/sctp/socket.c          |  4 ++--
 3 files changed, 3 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 8966599ddb9f..0ce0443c5b79 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -637,8 +637,6 @@ struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *,
 					    struct sctp_sndrcvinfo *,
 					    struct msghdr *, int len);
 void sctp_datamsg_put(struct sctp_datamsg *);
-void sctp_datamsg_free(struct sctp_datamsg *);
-void sctp_datamsg_track(struct sctp_chunk *);
 void sctp_chunk_fail(struct sctp_chunk *, int error);
 int sctp_chunk_abandoned(struct sctp_chunk *);
 
diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c
index e1f355080026..ed857643e3ff 100644
--- a/net/sctp/chunk.c
+++ b/net/sctp/chunk.c
@@ -136,20 +136,6 @@ void sctp_datamsg_put(struct sctp_datamsg *msg)
 		sctp_datamsg_destroy(msg);
 }
 
-/* Free a message.  Really just give up a reference, the
- * really free happens in sctp_datamsg_destroy().
- */
-void sctp_datamsg_free(struct sctp_datamsg *msg)
-{
-	sctp_datamsg_put(msg);
-}
-
-/* Hold on to all the fragments until all chunks have been sent. */
-void sctp_datamsg_track(struct sctp_chunk *chunk)
-{
-	sctp_chunk_hold(chunk);
-}
-
 /* Assign a chunk to this datamsg. */
 static void sctp_datamsg_assign(struct sctp_datamsg *msg, struct sctp_chunk *chunk)
 {
@@ -295,7 +281,7 @@ errout:
 		chunk = list_entry(pos, struct sctp_chunk, frag_list);
 		sctp_chunk_free(chunk);
 	}
-	sctp_datamsg_free(msg);
+	sctp_datamsg_put(msg);
 	return NULL;
 }
 
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index a3138a0fe2c5..76c747056dd7 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -1729,7 +1729,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk,
 	/* Now send the (possibly) fragmented message. */
 	list_for_each(pos, &datamsg->chunks) {
 		chunk = list_entry(pos, struct sctp_chunk, frag_list);
-		sctp_datamsg_track(chunk);
+		sctp_chunk_hold(chunk);
 
 		/* Do accounting for the write space.  */
 		sctp_set_owner_w(chunk);
@@ -1748,7 +1748,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk,
 		SCTP_DEBUG_PRINTK("We sent primitively.\n");
 	}
 
-	sctp_datamsg_free(datamsg);
+	sctp_datamsg_put(datamsg);
 	if (err)
 		goto out_free;
 	else
-- 
cgit v1.2.3


From 0e6bd4a1c6c3881c9ed82985ecb9824d4450c4ba Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 24 Mar 2008 15:29:23 -0700
Subject: [NETNS]: Add namespace parameter to ip_options_compile.

ip_options_compile uses inet_addr_type which requires a namespace. The
packet argument is optional, so parameter is the only way to obtain
it. Pass the init_net there for now.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h      | 3 ++-
 net/ipv4/ip_input.c   | 2 +-
 net/ipv4/ip_options.c | 7 ++++---
 3 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 9f50d4f1f157..bcc3afa3df36 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -347,7 +347,8 @@ extern int ip_forward(struct sk_buff *skb);
 extern void ip_options_build(struct sk_buff *skb, struct ip_options *opt, __be32 daddr, struct rtable *rt, int is_frag);
 extern int ip_options_echo(struct ip_options *dopt, struct sk_buff *skb);
 extern void ip_options_fragment(struct sk_buff *skb);
-extern int ip_options_compile(struct ip_options *opt, struct sk_buff *skb);
+extern int ip_options_compile(struct net *net,
+			      struct ip_options *opt, struct sk_buff *skb);
 extern int ip_options_get(struct ip_options **optp,
 			  unsigned char *data, int optlen);
 extern int ip_options_get_from_user(struct ip_options **optp,
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index e3a0c78fa7bf..f3a7a08a463f 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -286,7 +286,7 @@ static inline int ip_rcv_options(struct sk_buff *skb)
 	opt = &(IPCB(skb)->opt);
 	opt->optlen = iph->ihl*4 - sizeof(struct iphdr);
 
-	if (ip_options_compile(opt, skb)) {
+	if (ip_options_compile(&init_net, opt, skb)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
 		goto drop;
 	}
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index aeed4e5858ec..f0949b42e79e 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -248,7 +248,8 @@ void ip_options_fragment(struct sk_buff * skb)
  * If opt == NULL, then skb->data should point to IP header.
  */
 
-int ip_options_compile(struct ip_options * opt, struct sk_buff * skb)
+int ip_options_compile(struct net *net,
+		       struct ip_options * opt, struct sk_buff * skb)
 {
 	int l;
 	unsigned char * iph;
@@ -389,7 +390,7 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb)
 					{
 						__be32 addr;
 						memcpy(&addr, &optptr[optptr[2]-1], 4);
-						if (inet_addr_type(&init_net, addr) == RTN_UNICAST)
+						if (inet_addr_type(net, addr) == RTN_UNICAST)
 							break;
 						if (skb)
 							timeptr = (__be32*)&optptr[optptr[2]+3];
@@ -512,7 +513,7 @@ static int ip_options_get_finish(struct ip_options **optp,
 	while (optlen & 3)
 		opt->__data[optlen++] = IPOPT_END;
 	opt->optlen = optlen;
-	if (optlen && ip_options_compile(opt, NULL)) {
+	if (optlen && ip_options_compile(&init_net, opt, NULL)) {
 		kfree(opt);
 		return -EINVAL;
 	}
-- 
cgit v1.2.3


From f2c4802b3fdfb0d9596d932ca2af0ef6f8d60491 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 24 Mar 2008 15:29:55 -0700
Subject: [NETNS]: Add namespace parameter to ip_options_get(...).

Pass the init_net there for now.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h       |  4 ++--
 net/ipv4/ip_options.c  | 14 ++++++++------
 net/ipv4/ip_sockglue.c |  4 ++--
 3 files changed, 12 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index bcc3afa3df36..531270dc48a6 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -349,9 +349,9 @@ extern int ip_options_echo(struct ip_options *dopt, struct sk_buff *skb);
 extern void ip_options_fragment(struct sk_buff *skb);
 extern int ip_options_compile(struct net *net,
 			      struct ip_options *opt, struct sk_buff *skb);
-extern int ip_options_get(struct ip_options **optp,
+extern int ip_options_get(struct net *net, struct ip_options **optp,
 			  unsigned char *data, int optlen);
-extern int ip_options_get_from_user(struct ip_options **optp,
+extern int ip_options_get_from_user(struct net *net, struct ip_options **optp,
 				    unsigned char __user *data, int optlen);
 extern void ip_options_undo(struct ip_options * opt);
 extern void ip_forward_options(struct sk_buff *skb);
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index f0949b42e79e..59f7ddfb29bf 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -507,13 +507,13 @@ static struct ip_options *ip_options_get_alloc(const int optlen)
 		       GFP_KERNEL);
 }
 
-static int ip_options_get_finish(struct ip_options **optp,
+static int ip_options_get_finish(struct net *net, struct ip_options **optp,
 				 struct ip_options *opt, int optlen)
 {
 	while (optlen & 3)
 		opt->__data[optlen++] = IPOPT_END;
 	opt->optlen = optlen;
-	if (optlen && ip_options_compile(&init_net, opt, NULL)) {
+	if (optlen && ip_options_compile(net, opt, NULL)) {
 		kfree(opt);
 		return -EINVAL;
 	}
@@ -522,7 +522,8 @@ static int ip_options_get_finish(struct ip_options **optp,
 	return 0;
 }
 
-int ip_options_get_from_user(struct ip_options **optp, unsigned char __user *data, int optlen)
+int ip_options_get_from_user(struct net *net, struct ip_options **optp,
+			     unsigned char __user *data, int optlen)
 {
 	struct ip_options *opt = ip_options_get_alloc(optlen);
 
@@ -532,10 +533,11 @@ int ip_options_get_from_user(struct ip_options **optp, unsigned char __user *dat
 		kfree(opt);
 		return -EFAULT;
 	}
-	return ip_options_get_finish(optp, opt, optlen);
+	return ip_options_get_finish(net, optp, opt, optlen);
 }
 
-int ip_options_get(struct ip_options **optp, unsigned char *data, int optlen)
+int ip_options_get(struct net *net, struct ip_options **optp,
+		   unsigned char *data, int optlen)
 {
 	struct ip_options *opt = ip_options_get_alloc(optlen);
 
@@ -543,7 +545,7 @@ int ip_options_get(struct ip_options **optp, unsigned char *data, int optlen)
 		return -ENOMEM;
 	if (optlen)
 		memcpy(opt->__data, data, optlen);
-	return ip_options_get_finish(optp, opt, optlen);
+	return ip_options_get_finish(net, optp, opt, optlen);
 }
 
 void ip_forward_options(struct sk_buff *skb)
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index bb3cbe5ec36d..1b86a50269bc 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -176,7 +176,7 @@ int ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc)
 		switch (cmsg->cmsg_type) {
 		case IP_RETOPTS:
 			err = cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr));
-			err = ip_options_get(&ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40);
+			err = ip_options_get(&init_net, &ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40);
 			if (err)
 				return err;
 			break;
@@ -449,7 +449,7 @@ static int do_ip_setsockopt(struct sock *sk, int level,
 		struct ip_options * opt = NULL;
 		if (optlen > 40 || optlen < 0)
 			goto e_inval;
-		err = ip_options_get_from_user(&opt, optval, optlen);
+		err = ip_options_get_from_user(&init_net, &opt, optval, optlen);
 		if (err)
 			break;
 		if (inet->is_icsk) {
-- 
cgit v1.2.3


From 7a6adb92fe301c10ca4dbd0d9f2422f5880595e7 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 24 Mar 2008 15:30:27 -0700
Subject: [NETNS]: Add namespace parameter to ip_cmsg_send.

Pass the init_net there for now.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h       | 3 ++-
 net/ipv4/ip_sockglue.c | 4 ++--
 net/ipv4/raw.c         | 2 +-
 net/ipv4/udp.c         | 2 +-
 4 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 531270dc48a6..6d7bcd5e62d4 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -362,7 +362,8 @@ extern int ip_options_rcv_srr(struct sk_buff *skb);
  */
 
 extern void	ip_cmsg_recv(struct msghdr *msg, struct sk_buff *skb);
-extern int	ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc);
+extern int	ip_cmsg_send(struct net *net,
+			     struct msghdr *msg, struct ipcm_cookie *ipc);
 extern int	ip_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen);
 extern int	ip_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen);
 extern int	compat_ip_setsockopt(struct sock *sk, int level,
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 1b86a50269bc..0857f2d042cd 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -163,7 +163,7 @@ void ip_cmsg_recv(struct msghdr *msg, struct sk_buff *skb)
 		ip_cmsg_recv_security(msg, skb);
 }
 
-int ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc)
+int ip_cmsg_send(struct net *net, struct msghdr *msg, struct ipcm_cookie *ipc)
 {
 	int err;
 	struct cmsghdr *cmsg;
@@ -176,7 +176,7 @@ int ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc)
 		switch (cmsg->cmsg_type) {
 		case IP_RETOPTS:
 			err = cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr));
-			err = ip_options_get(&init_net, &ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40);
+			err = ip_options_get(net, &ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40);
 			if (err)
 				return err;
 			break;
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index b433b485a883..7d29a05bf887 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -499,7 +499,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	ipc.oif = sk->sk_bound_dev_if;
 
 	if (msg->msg_controllen) {
-		err = ip_cmsg_send(msg, &ipc);
+		err = ip_cmsg_send(&init_net, msg, &ipc);
 		if (err)
 			goto out;
 		if (ipc.opt)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 9cd48b0bb7d4..d142b87e9eb8 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -607,7 +607,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 
 	ipc.oif = sk->sk_bound_dev_if;
 	if (msg->msg_controllen) {
-		err = ip_cmsg_send(msg, &ipc);
+		err = ip_cmsg_send(&init_net, msg, &ipc);
 		if (err)
 			return err;
 		if (ipc.opt)
-- 
cgit v1.2.3


From f145049a06f470d0489f47cb83ff3ccb2a0de622 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Mon, 24 Mar 2008 15:33:00 -0700
Subject: [NETNS]: Drop packets in the non-initial namespace on the
 per/protocol basis.

IP layer now can handle multiple namespaces normally. So, process such
packets normally and drop them only if the transport layer is not
aware about namespaces.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/protocol.h | 3 ++-
 net/ipv4/ip_input.c    | 8 ++++----
 2 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/protocol.h b/include/net/protocol.h
index ad8c584233a6..8d024d7cb741 100644
--- a/include/net/protocol.h
+++ b/include/net/protocol.h
@@ -39,7 +39,8 @@ struct net_protocol {
 	int			(*gso_send_check)(struct sk_buff *skb);
 	struct sk_buff	       *(*gso_segment)(struct sk_buff *skb,
 					       int features);
-	int			no_policy;
+	unsigned int		no_policy:1,
+				netns_ok:1;
 };
 
 #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index eb1fa27dc0c4..2aeea5d15425 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -199,6 +199,8 @@ int ip_call_ra_chain(struct sk_buff *skb)
 
 static int ip_local_deliver_finish(struct sk_buff *skb)
 {
+	struct net *net = skb->dev->nd_net;
+
 	__skb_pull(skb, ip_hdrlen(skb));
 
 	/* Point into the IP datagram, just past the header. */
@@ -214,7 +216,8 @@ static int ip_local_deliver_finish(struct sk_buff *skb)
 		raw = raw_local_deliver(skb, protocol);
 
 		hash = protocol & (MAX_INET_PROTOS - 1);
-		if ((ipprot = rcu_dereference(inet_protos[hash])) != NULL) {
+		ipprot = rcu_dereference(inet_protos[hash]);
+		if (ipprot != NULL && (net == &init_net || ipprot->netns_ok)) {
 			int ret;
 
 			if (!ipprot->no_policy) {
@@ -375,9 +378,6 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	struct iphdr *iph;
 	u32 len;
 
-	if (dev->nd_net != &init_net)
-		goto drop;
-
 	/* When the interface is in promisc. mode, drop all the crap
 	 * that it receives, do not try to analyse it.
 	 */
-- 
cgit v1.2.3


From 9bb182a7007515239091b237fe7169b1328a61d3 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Fri, 22 Feb 2008 14:48:22 +0900
Subject: [XFRM] MIP6: Fix address keys for routing search.

Each MIPv6 XFRM state (DSTOPT/RH2) holds either destination or source
address to be mangled in the IPv6 header (that is "CoA").
On Inter-MN communication after both nodes binds each other,
they use route optimized traffic two MIPv6 states applied, and
both source and destination address in the IPv6 header
are replaced by the states respectively.
The packet format is correct, however, next-hop routing search
are not.
This patch fixes it by remembering address pairs for later states.

Based on patch from Masahide NAKAMURA <nakam@linux-ipv6.org>.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/xfrm.h     | 17 +++++++++++++++++
 net/xfrm/xfrm_policy.c | 49 ++++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 57 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index c435620dbb37..bed7d43932f6 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1045,6 +1045,23 @@ xfrm_address_t *xfrm_flowi_saddr(struct flowi *fl, unsigned short family)
 	return NULL;
 }
 
+static __inline__
+void xfrm_flowi_addr_get(struct flowi *fl,
+			 xfrm_address_t *saddr, xfrm_address_t *daddr,
+			 unsigned short family)
+{
+	switch(family) {
+	case AF_INET:
+		memcpy(&saddr->a4, &fl->fl4_src, sizeof(saddr->a4));
+		memcpy(&daddr->a4, &fl->fl4_dst, sizeof(daddr->a4));
+		break;
+	case AF_INET6:
+		ipv6_addr_copy((struct in6_addr *)&saddr->a6, &fl->fl6_src);
+		ipv6_addr_copy((struct in6_addr *)&daddr->a6, &fl->fl6_dst);
+		break;
+	}
+}
+
 static __inline__ int
 __xfrm4_state_addr_check(struct xfrm_state *x,
 			 xfrm_address_t *daddr, xfrm_address_t *saddr)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index bae94a8031a2..8e588f20c60c 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -97,25 +97,52 @@ int xfrm_selector_match(struct xfrm_selector *sel, struct flowi *fl,
 	return 0;
 }
 
+static inline struct dst_entry *__xfrm_dst_lookup(int tos,
+						  xfrm_address_t *saddr,
+						  xfrm_address_t *daddr,
+						  int family)
+{
+	struct xfrm_policy_afinfo *afinfo;
+	struct dst_entry *dst;
+
+	afinfo = xfrm_policy_get_afinfo(family);
+	if (unlikely(afinfo == NULL))
+		return ERR_PTR(-EAFNOSUPPORT);
+
+	dst = afinfo->dst_lookup(tos, saddr, daddr);
+
+	xfrm_policy_put_afinfo(afinfo);
+
+	return dst;
+}
+
 static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos,
+						xfrm_address_t *prev_saddr,
+						xfrm_address_t *prev_daddr,
 						int family)
 {
 	xfrm_address_t *saddr = &x->props.saddr;
 	xfrm_address_t *daddr = &x->id.daddr;
-	struct xfrm_policy_afinfo *afinfo;
 	struct dst_entry *dst;
 
-	if (x->type->flags & XFRM_TYPE_LOCAL_COADDR)
+	if (x->type->flags & XFRM_TYPE_LOCAL_COADDR) {
 		saddr = x->coaddr;
-	if (x->type->flags & XFRM_TYPE_REMOTE_COADDR)
+		daddr = prev_daddr;
+	}
+	if (x->type->flags & XFRM_TYPE_REMOTE_COADDR) {
+		saddr = prev_saddr;
 		daddr = x->coaddr;
+	}
 
-	afinfo = xfrm_policy_get_afinfo(family);
-	if (unlikely(afinfo == NULL))
-		return ERR_PTR(-EAFNOSUPPORT);
+	dst = __xfrm_dst_lookup(tos, saddr, daddr, family);
+
+	if (!IS_ERR(dst)) {
+		if (prev_saddr != saddr)
+			memcpy(prev_saddr, saddr,  sizeof(*prev_saddr));
+		if (prev_daddr != daddr)
+			memcpy(prev_daddr, daddr,  sizeof(*prev_daddr));
+	}
 
-	dst = afinfo->dst_lookup(tos, saddr, daddr);
-	xfrm_policy_put_afinfo(afinfo);
 	return dst;
 }
 
@@ -1354,6 +1381,9 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
 	int trailer_len = 0;
 	int tos;
 	int family = policy->selector.family;
+	xfrm_address_t saddr, daddr;
+
+	xfrm_flowi_addr_get(fl, &saddr, &daddr, family);
 
 	tos = xfrm_get_tos(fl, family);
 	err = tos;
@@ -1384,7 +1414,8 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
 
 		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
 			family = xfrm[i]->props.family;
-			dst = xfrm_dst_lookup(xfrm[i], tos, family);
+			dst = xfrm_dst_lookup(xfrm[i], tos, &saddr, &daddr,
+					      family);
 			err = PTR_ERR(dst);
 			if (IS_ERR(dst))
 				goto put_states;
-- 
cgit v1.2.3


From c8cdaf998df221b01134a051aba38c570105061b Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Mon, 10 Mar 2008 04:30:37 -0400
Subject: [IPV4,IPV6]: Share cork.rt between IPv4 and IPv6.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/ipv6.h    |  1 -
 include/net/inet_sock.h |  2 +-
 net/ipv4/ip_output.c    | 14 ++++++--------
 net/ipv6/ip6_output.c   | 12 ++++++------
 4 files changed, 13 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 4aaefc349a4b..2102d8b67c01 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -315,7 +315,6 @@ struct ipv6_pinfo {
 	struct sk_buff		*pktoptions;
 	struct {
 		struct ipv6_txoptions *opt;
-		struct rt6_info	*rt;
 		int hop_limit;
 		int tclass;
 	} cork;
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index b6db16d2766a..a42cd63d241a 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -136,7 +136,7 @@ struct inet_sock {
 		unsigned int		flags;
 		unsigned int		fragsize;
 		struct ip_options	*opt;
-		struct rtable		*rt;
+		struct dst_entry	*dst;
 		int			length; /* Total length of all frames */
 		__be32			addr;
 		struct flowi		fl;
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 349fae58c1a3..913266cd9902 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -825,7 +825,7 @@ int ip_append_data(struct sock *sk,
 		inet->cork.fragsize = mtu = inet->pmtudisc == IP_PMTUDISC_PROBE ?
 					    rt->u.dst.dev->mtu :
 					    dst_mtu(rt->u.dst.path);
-		inet->cork.rt = rt;
+		inet->cork.dst = &rt->u.dst;
 		inet->cork.length = 0;
 		sk->sk_sndmsg_page = NULL;
 		sk->sk_sndmsg_off = 0;
@@ -834,7 +834,7 @@ int ip_append_data(struct sock *sk,
 			transhdrlen += exthdrlen;
 		}
 	} else {
-		rt = inet->cork.rt;
+		rt = (struct rtable *)inet->cork.dst;
 		if (inet->cork.flags & IPCORK_OPT)
 			opt = inet->cork.opt;
 
@@ -1083,7 +1083,7 @@ ssize_t	ip_append_page(struct sock *sk, struct page *page,
 	if (skb_queue_empty(&sk->sk_write_queue))
 		return -EINVAL;
 
-	rt = inet->cork.rt;
+	rt = (struct rtable *)inet->cork.dst;
 	if (inet->cork.flags & IPCORK_OPT)
 		opt = inet->cork.opt;
 
@@ -1208,10 +1208,8 @@ static void ip_cork_release(struct inet_sock *inet)
 	inet->cork.flags &= ~IPCORK_OPT;
 	kfree(inet->cork.opt);
 	inet->cork.opt = NULL;
-	if (inet->cork.rt) {
-		ip_rt_put(inet->cork.rt);
-		inet->cork.rt = NULL;
-	}
+	dst_release(inet->cork.dst);
+	inet->cork.dst = NULL;
 }
 
 /*
@@ -1224,7 +1222,7 @@ int ip_push_pending_frames(struct sock *sk)
 	struct sk_buff **tail_skb;
 	struct inet_sock *inet = inet_sk(sk);
 	struct ip_options *opt = NULL;
-	struct rtable *rt = inet->cork.rt;
+	struct rtable *rt = (struct rtable *)inet->cork.dst;
 	struct iphdr *iph;
 	__be16 df = 0;
 	__u8 ttl;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 98762fde2b65..ed6482667a25 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1115,7 +1115,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 			/* need source address above miyazawa*/
 		}
 		dst_hold(&rt->u.dst);
-		np->cork.rt = rt;
+		inet->cork.dst = &rt->u.dst;
 		inet->cork.fl = *fl;
 		np->cork.hop_limit = hlimit;
 		np->cork.tclass = tclass;
@@ -1136,7 +1136,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 		length += exthdrlen;
 		transhdrlen += exthdrlen;
 	} else {
-		rt = np->cork.rt;
+		rt = (struct rt6_info *)inet->cork.dst;
 		fl = &inet->cork.fl;
 		if (inet->cork.flags & IPCORK_OPT)
 			opt = np->cork.opt;
@@ -1381,9 +1381,9 @@ static void ip6_cork_release(struct inet_sock *inet, struct ipv6_pinfo *np)
 	inet->cork.flags &= ~IPCORK_OPT;
 	kfree(np->cork.opt);
 	np->cork.opt = NULL;
-	if (np->cork.rt) {
-		dst_release(&np->cork.rt->u.dst);
-		np->cork.rt = NULL;
+	if (inet->cork.dst) {
+		dst_release(inet->cork.dst);
+		inet->cork.dst = NULL;
 		inet->cork.flags &= ~IPCORK_ALLFRAG;
 	}
 	memset(&inet->cork.fl, 0, sizeof(inet->cork.fl));
@@ -1398,7 +1398,7 @@ int ip6_push_pending_frames(struct sock *sk)
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6hdr *hdr;
 	struct ipv6_txoptions *opt = np->cork.opt;
-	struct rt6_info *rt = np->cork.rt;
+	struct rt6_info *rt = (struct rt6_info *)inet->cork.dst;
 	struct flowi *fl = &inet->cork.fl;
 	unsigned char proto = fl->proto;
 	int err = 0;
-- 
cgit v1.2.3


From 4725474584d6aa2f07b3d47442dfbc4f6544f65e Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Mon, 10 Mar 2008 04:41:33 -0400
Subject: [IPV6]: Convert cork.hop_limit and cork.tclass into u8 instead of
 int.

Values of those fields are always between 0 and 255 (inclusive),
so use u8 and save some memory on 32bit systems.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/ipv6.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 2102d8b67c01..9b59e37afad9 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -315,8 +315,8 @@ struct ipv6_pinfo {
 	struct sk_buff		*pktoptions;
 	struct {
 		struct ipv6_txoptions *opt;
-		int hop_limit;
-		int tclass;
+		u8 hop_limit;
+		u8 tclass;
 	} cork;
 };
 
-- 
cgit v1.2.3


From 6b75d0908185bf853b188afa6f269426f6554c5b Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Mon, 10 Mar 2008 06:00:30 -0400
Subject: [IPV6]: Optimize hop-limit determination.

Last part of hop-limit determination is always:
    hoplimit = dst_metric(dst, RTAX_HOPLIMIT);
    if (hoplimit < 0)
        hoplimit = ipv6_get_hoplimit(dst->dev).

Let's consolidate it as ip6_dst_hoplimit(dst).

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h   |  2 --
 include/net/ip6_route.h  |  2 ++
 net/ipv6/icmp.c          |  8 ++------
 net/ipv6/ip6_output.c    |  4 +---
 net/ipv6/ipv6_sockglue.c |  4 +---
 net/ipv6/raw.c           |  4 +---
 net/ipv6/route.c         | 20 +++++++++++---------
 net/ipv6/udp.c           |  4 +---
 8 files changed, 19 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 232da20e7171..edcb4bbaab7d 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -123,8 +123,6 @@ extern int ipv6_is_mld(struct sk_buff *skb, int nexthdr);
 
 extern void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len);
 
-extern int ipv6_get_hoplimit(struct net_device *dev);
-
 /*
  *	anycast prototypes (anycast.c)
  */
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 0e2895c8b270..5c3b67c86aef 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -88,6 +88,8 @@ extern struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 					   const struct in6_addr *addr,
 					   int anycast);
 
+extern int			ip6_dst_hoplimit(struct dst_entry *dst);
+
 /*
  *	support functions for ND
  *
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 6b5391ab8346..86332417b402 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -464,9 +464,7 @@ route_done:
 	else
 		hlimit = np->hop_limit;
 	if (hlimit < 0)
-		hlimit = dst_metric(dst, RTAX_HOPLIMIT);
-	if (hlimit < 0)
-		hlimit = ipv6_get_hoplimit(dst->dev);
+		hlimit = ip6_dst_hoplimit(dst);
 
 	tclass = np->tclass;
 	if (tclass < 0)
@@ -560,9 +558,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
 	else
 		hlimit = np->hop_limit;
 	if (hlimit < 0)
-		hlimit = dst_metric(dst, RTAX_HOPLIMIT);
-	if (hlimit < 0)
-		hlimit = ipv6_get_hoplimit(dst->dev);
+		hlimit = ip6_dst_hoplimit(dst);
 
 	tclass = np->tclass;
 	if (tclass < 0)
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index ed6482667a25..2a4f08c8a02d 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -237,9 +237,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl,
 	if (np)
 		hlimit = np->hop_limit;
 	if (hlimit < 0)
-		hlimit = dst_metric(dst, RTAX_HOPLIMIT);
-	if (hlimit < 0)
-		hlimit = ipv6_get_hoplimit(dst->dev);
+		hlimit = ip6_dst_hoplimit(dst);
 
 	tclass = -1;
 	if (np)
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index c11c76cab371..8e29fb1d1df6 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -904,9 +904,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
 		dst = sk_dst_get(sk);
 		if (dst) {
 			if (val < 0)
-				val = dst_metric(dst, RTAX_HOPLIMIT);
-			if (val < 0)
-				val = ipv6_get_hoplimit(dst->dev);
+				val = ip6_dst_hoplimit(dst);
 			dst_release(dst);
 		}
 		if (val < 0)
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index a9e4235157a2..548d0763f4d3 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -885,9 +885,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 		else
 			hlimit = np->hop_limit;
 		if (hlimit < 0)
-			hlimit = dst_metric(dst, RTAX_HOPLIMIT);
-		if (hlimit < 0)
-			hlimit = ipv6_get_hoplimit(dst->dev);
+			hlimit = ip6_dst_hoplimit(dst);
 	}
 
 	if (tclass < 0) {
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index a4b5aee0f68a..aa3f08718e44 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1034,15 +1034,17 @@ static int ipv6_get_mtu(struct net_device *dev)
 	return mtu;
 }
 
-int ipv6_get_hoplimit(struct net_device *dev)
-{
-	int hoplimit = ipv6_devconf.hop_limit;
-	struct inet6_dev *idev;
-
-	idev = in6_dev_get(dev);
-	if (idev) {
-		hoplimit = idev->cnf.hop_limit;
-		in6_dev_put(idev);
+int ip6_dst_hoplimit(struct dst_entry *dst)
+{
+	int hoplimit = dst_metric(dst, RTAX_HOPLIMIT);
+	if (hoplimit < 0) {
+		struct net_device *dev = dst->dev;
+		struct inet6_dev *idev = in6_dev_get(dev);
+		if (idev) {
+			hoplimit = idev->cnf.hop_limit;
+			in6_dev_put(idev);
+		} else
+			hoplimit = ipv6_devconf.hop_limit;
 	}
 	return hoplimit;
 }
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 5f5d1218c34e..593d3efadaf9 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -792,9 +792,7 @@ do_udp_sendmsg:
 		else
 			hlimit = np->hop_limit;
 		if (hlimit < 0)
-			hlimit = dst_metric(dst, RTAX_HOPLIMIT);
-		if (hlimit < 0)
-			hlimit = ipv6_get_hoplimit(dst->dev);
+			hlimit = ip6_dst_hoplimit(dst);
 	}
 
 	if (tclass < 0) {
-- 
cgit v1.2.3


From 1d5d236d309ab90fa6aedf712f586b3595721373 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Mon, 10 Mar 2008 10:56:55 -0400
Subject: [IPV6]: Use bitfields for hop_limit and mcast_hops.

Save some bits for future extensions.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/ipv6.h | 28 ++++++++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 9b59e37afad9..87ae4e389ce1 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -274,8 +274,29 @@ struct ipv6_pinfo {
 
 	__be32			flow_label;
 	__u32			frag_size;
-	__s16			hop_limit;
-	__s16			mcast_hops;
+
+	/*
+	 * Packed in 16bits.
+	 * Omit one shift by by putting the signed field at MSB.
+	 */
+#if defined(__BIG_ENDIAN_BITFIELD)
+	__s16			hop_limit:9;
+	__u16			__unused_1:7;
+#else
+	__u16			__unused_1:7;
+	__s16			hop_limit:9;
+#endif
+
+#if defined(__BIG_ENDIAN_BITFIELD)
+	/* Packed in 16bits. */
+	__s16			mcast_hops:9;
+	__u16			__unused_2:6,
+				mc_loop:1;
+#else
+	__u16			mc_loop:1,
+				__unused_2:6;
+	__s16			mcast_hops:9;
+#endif
 	int			mcast_oif;
 
 	/* pktoption flags */
@@ -298,8 +319,7 @@ struct ipv6_pinfo {
 	} rxopt;
 
 	/* sockopt flags */
-	__u8			mc_loop:1,
-	                        recverr:1,
+	__u8			recverr:1,
 	                        sndflow:1,
 				pmtudisc:2,
 				ipv6only:1;
-- 
cgit v1.2.3


From 7cbca67c073263c179f605bdbbdc565ab29d801d Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Tue, 25 Mar 2008 09:37:42 +0900
Subject: [IPV6]: Support Source Address Selection API (RFC5014).

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/in6.h      | 11 +++++++
 include/linux/ipv6.h     |  6 +++-
 include/net/addrconf.h   |  1 +
 include/net/ip6_route.h  |  9 ++++--
 net/ipv6/addrconf.c      | 17 +++++++++--
 net/ipv6/fib6_rules.c    | 12 +++++++-
 net/ipv6/ip6_output.c    |  4 ++-
 net/ipv6/ipv6_sockglue.c | 78 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/ndisc.c         |  4 ++-
 net/ipv6/route.c         | 11 ++++++-
 net/ipv6/xfrm6_policy.c  |  2 +-
 net/sctp/ipv6.c          |  4 ++-
 12 files changed, 146 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index 2a61c82af115..f674000c6c99 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -249,4 +249,15 @@ struct in6_flowlabel_req
  * IP6T_SO_GET_REVISION_TARGET	69
  */
 
+/* RFC5014: Source address selection */
+#define IPV6_ADDR_PREFERENCES	72
+
+#define IPV6_PREFER_SRC_TMP		0x0001
+#define IPV6_PREFER_SRC_PUBLIC		0x0002
+#define IPV6_PREFER_SRC_PUBTMP_DEFAULT	0x0100
+#define IPV6_PREFER_SRC_COA		0x0004
+#define IPV6_PREFER_SRC_HOME		0x0400
+#define IPV6_PREFER_SRC_CGA		0x0008
+#define IPV6_PREFER_SRC_NONCGA		0x0800
+
 #endif
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 87ae4e389ce1..c9ba0da16ce9 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -322,7 +322,11 @@ struct ipv6_pinfo {
 	__u8			recverr:1,
 	                        sndflow:1,
 				pmtudisc:2,
-				ipv6only:1;
+				ipv6only:1,
+				srcprefs:3;	/* 001: prefer temporary address
+						 * 010: prefer public address
+						 * 100: prefer care-of address
+						 */
 	__u8			tclass;
 
 	__u32			dst_cookie;
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index edcb4bbaab7d..c9276c72764d 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -78,6 +78,7 @@ extern struct inet6_ifaddr      *ipv6_get_ifaddr(struct net *net,
 
 extern int			ipv6_dev_get_saddr(struct net_device *dev, 
 					       struct in6_addr *daddr,
+					       unsigned int srcprefs,
 					       struct in6_addr *saddr);
 extern int			ipv6_get_lladdr(struct net_device *dev,
 						struct in6_addr *addr,
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 5c3b67c86aef..3ae6799c2b14 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -30,9 +30,12 @@ struct route_info {
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 
-#define RT6_LOOKUP_F_IFACE	0x1
-#define RT6_LOOKUP_F_REACHABLE	0x2
-#define RT6_LOOKUP_F_HAS_SADDR	0x4
+#define RT6_LOOKUP_F_IFACE		0x00000001
+#define RT6_LOOKUP_F_REACHABLE		0x00000002
+#define RT6_LOOKUP_F_HAS_SADDR		0x00000004
+#define RT6_LOOKUP_F_SRCPREF_TMP	0x00000008
+#define RT6_LOOKUP_F_SRCPREF_PUBLIC	0x00000010
+#define RT6_LOOKUP_F_SRCPREF_COA	0x00000020
 
 extern struct rt6_info	*ip6_null_entry;
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 787e90af166c..89954885dee1 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -909,6 +909,7 @@ struct ipv6_saddr_dst {
 	int ifindex;
 	int scope;
 	int label;
+	unsigned int prefs;
 };
 
 static inline int ipv6_saddr_preferred(int type)
@@ -984,9 +985,12 @@ static int ipv6_get_saddr_eval(struct ipv6_saddr_score *score,
 		break;
 #ifdef CONFIG_IPV6_MIP6
 	case IPV6_SADDR_RULE_HOA:
+	    {
 		/* Rule 4: Prefer home address */
-		ret = !!(score->ifa->flags & IFA_F_HOMEADDRESS);
+		int prefhome = !(dst->prefs & IPV6_PREFER_SRC_COA);
+		ret = !(score->ifa->flags & IFA_F_HOMEADDRESS) ^ prefhome;
 		break;
+	    }
 #endif
 	case IPV6_SADDR_RULE_OIF:
 		/* Rule 5: Prefer outgoing interface */
@@ -1000,11 +1004,16 @@ static int ipv6_get_saddr_eval(struct ipv6_saddr_score *score,
 		break;
 #ifdef CONFIG_IPV6_PRIVACY
 	case IPV6_SADDR_RULE_PRIVACY:
+	    {
 		/* Rule 7: Prefer public address
 		 * Note: prefer temprary address if use_tempaddr >= 2
 		 */
-		ret = (!(score->ifa->flags & IFA_F_TEMPORARY)) ^ (score->ifa->idev->cnf.use_tempaddr >= 2);
+		int preftmp = dst->prefs & (IPV6_PREFER_SRC_PUBLIC|IPV6_PREFER_SRC_TMP) ?
+				!!(dst->prefs & IPV6_PREFER_SRC_TMP) :
+				score->ifa->idev->cnf.use_tempaddr >= 2;
+		ret = (!(score->ifa->flags & IFA_F_TEMPORARY)) ^ preftmp;
 		break;
+	    }
 #endif
 	case IPV6_SADDR_RULE_ORCHID:
 		/* Rule 8-: Prefer ORCHID vs ORCHID or
@@ -1030,7 +1039,8 @@ out:
 }
 
 int ipv6_dev_get_saddr(struct net_device *dst_dev,
-		       struct in6_addr *daddr, struct in6_addr *saddr)
+		       struct in6_addr *daddr, unsigned int prefs,
+		       struct in6_addr *saddr)
 {
 	struct ipv6_saddr_score scores[2],
 				*score = &scores[0], *hiscore = &scores[1];
@@ -1044,6 +1054,7 @@ int ipv6_dev_get_saddr(struct net_device *dst_dev,
 	dst.ifindex = dst_dev ? dst_dev->ifindex : 0;
 	dst.scope = __ipv6_addr_src_scope(dst_type);
 	dst.label = ipv6_addr_label(daddr, dst_type, dst.ifindex);
+	dst.prefs = prefs;
 
 	hiscore->rule = -1;
 	hiscore->ifa = NULL;
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 55137408f054..e7a7fe26cebf 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -84,8 +84,18 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 		if ((rule->flags & FIB_RULE_FIND_SADDR) &&
 		    r->src.plen && !(flags & RT6_LOOKUP_F_HAS_SADDR)) {
 			struct in6_addr saddr;
+			unsigned int srcprefs = 0;
+
+			if (flags & RT6_LOOKUP_F_SRCPREF_TMP)
+				srcprefs |= IPV6_PREFER_SRC_TMP;
+			if (flags & RT6_LOOKUP_F_SRCPREF_PUBLIC)
+				srcprefs |= IPV6_PREFER_SRC_PUBLIC;
+			if (flags & RT6_LOOKUP_F_SRCPREF_COA)
+				srcprefs |= IPV6_PREFER_SRC_COA;
+
 			if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev,
-					       &flp->fl6_dst, &saddr))
+					       &flp->fl6_dst, srcprefs,
+					       &saddr))
 				goto again;
 			if (!ipv6_prefix_equal(&saddr, &r->src.addr,
 					       r->src.plen))
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 2a4f08c8a02d..d34aa61353bb 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -920,7 +920,9 @@ static int ip6_dst_lookup_tail(struct sock *sk,
 
 	if (ipv6_addr_any(&fl->fl6_src)) {
 		err = ipv6_dev_get_saddr(ip6_dst_idev(*dst)->dev,
-					 &fl->fl6_dst, &fl->fl6_src);
+					 &fl->fl6_dst,
+					 sk ? inet6_sk(sk)->srcprefs : 0,
+					 &fl->fl6_src);
 		if (err)
 			goto out_err_release;
 	}
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 8e29fb1d1df6..dc6695cc5767 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -617,7 +617,67 @@ done:
 		retv = xfrm_user_policy(sk, optname, optval, optlen);
 		break;
 
+	case IPV6_ADDR_PREFERENCES:
+	    {
+		unsigned int pref = 0;
+		unsigned int prefmask = ~0;
+
+		retv = -EINVAL;
+
+		/* check PUBLIC/TMP/PUBTMP_DEFAULT conflicts */
+		switch (val & (IPV6_PREFER_SRC_PUBLIC|
+			       IPV6_PREFER_SRC_TMP|
+			       IPV6_PREFER_SRC_PUBTMP_DEFAULT)) {
+		case IPV6_PREFER_SRC_PUBLIC:
+			pref |= IPV6_PREFER_SRC_PUBLIC;
+			break;
+		case IPV6_PREFER_SRC_TMP:
+			pref |= IPV6_PREFER_SRC_TMP;
+			break;
+		case IPV6_PREFER_SRC_PUBTMP_DEFAULT:
+			break;
+		case 0:
+			goto pref_skip_pubtmp;
+		default:
+			goto e_inval;
+		}
+
+		prefmask &= ~(IPV6_PREFER_SRC_PUBLIC|
+			      IPV6_PREFER_SRC_TMP);
+pref_skip_pubtmp:
+
+		/* check HOME/COA conflicts */
+		switch (val & (IPV6_PREFER_SRC_HOME|IPV6_PREFER_SRC_COA)) {
+		case IPV6_PREFER_SRC_HOME:
+			break;
+		case IPV6_PREFER_SRC_COA:
+			pref |= IPV6_PREFER_SRC_COA;
+		case 0:
+			goto pref_skip_coa;
+		default:
+			goto e_inval;
+		}
+
+		prefmask &= ~IPV6_PREFER_SRC_COA;
+pref_skip_coa:
+
+		/* check CGA/NONCGA conflicts */
+		switch (val & (IPV6_PREFER_SRC_CGA|IPV6_PREFER_SRC_NONCGA)) {
+		case IPV6_PREFER_SRC_CGA:
+		case IPV6_PREFER_SRC_NONCGA:
+		case 0:
+			break;
+		default:
+			goto e_inval;
+		}
+
+		np->srcprefs = (np->srcprefs & prefmask) | pref;
+		retv = 0;
+
+		break;
+	    }
 	}
+
 	release_sock(sk);
 
 	return retv;
@@ -932,6 +992,24 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
 		val = np->sndflow;
 		break;
 
+	case IPV6_ADDR_PREFERENCES:
+		val = 0;
+
+		if (np->srcprefs & IPV6_PREFER_SRC_TMP)
+			val |= IPV6_PREFER_SRC_TMP;
+		else if (np->srcprefs & IPV6_PREFER_SRC_PUBLIC)
+			val |= IPV6_PREFER_SRC_PUBLIC;
+		else {
+			/* XXX: should we return system default? */
+			val |= IPV6_PREFER_SRC_PUBTMP_DEFAULT;
+		}
+
+		if (np->srcprefs & IPV6_PREFER_SRC_COA)
+			val |= IPV6_PREFER_SRC_COA;
+		else
+			val |= IPV6_PREFER_SRC_HOME;
+		break;
+
 	default:
 		return -ENOPROTOOPT;
 	}
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index e7d8e74704c1..3f68a6eae7b2 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -546,7 +546,9 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
 			override = 0;
 		in6_ifa_put(ifp);
 	} else {
-		if (ipv6_dev_get_saddr(dev, daddr, &tmpaddr))
+		if (ipv6_dev_get_saddr(dev, daddr,
+				       inet6_sk(dev->nd_net->ipv6.ndisc_sk)->srcprefs,
+				       &tmpaddr))
 			return;
 		src_addr = &tmpaddr;
 	}
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index aa3f08718e44..06faa46920e1 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -782,6 +782,15 @@ struct dst_entry * ip6_route_output(struct net *net, struct sock *sk,
 
 	if (!ipv6_addr_any(&fl->fl6_src))
 		flags |= RT6_LOOKUP_F_HAS_SADDR;
+	else if (sk) {
+		unsigned int prefs = inet6_sk(sk)->srcprefs;
+		if (prefs & IPV6_PREFER_SRC_TMP)
+			flags |= RT6_LOOKUP_F_SRCPREF_TMP;
+		if (prefs & IPV6_PREFER_SRC_PUBLIC)
+			flags |= RT6_LOOKUP_F_SRCPREF_PUBLIC;
+		if (prefs & IPV6_PREFER_SRC_COA)
+			flags |= RT6_LOOKUP_F_SRCPREF_COA;
+	}
 
 	return fib6_rule_lookup(net, fl, flags, ip6_pol_route_output);
 }
@@ -2162,7 +2171,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 	else if (dst) {
 		struct in6_addr saddr_buf;
 		if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev,
-				       dst, &saddr_buf) == 0)
+				       dst, 0, &saddr_buf) == 0)
 			NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf);
 	}
 
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index e96dafdc7032..d92d1fceb8cf 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -58,7 +58,7 @@ static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
 		return -EHOSTUNREACH;
 
 	ipv6_dev_get_saddr(ip6_dst_idev(dst)->dev,
-			   (struct in6_addr *)&daddr->a6,
+			   (struct in6_addr *)&daddr->a6, 0,
 			   (struct in6_addr *)&saddr->a6);
 	dst_release(dst);
 	return 0;
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index 46c5b3c5cb99..dc71d0d83753 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -316,7 +316,9 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc,
 
 	if (!asoc) {
 		ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL,
-				   &daddr->v6.sin6_addr, &saddr->v6.sin6_addr);
+				   &daddr->v6.sin6_addr,
+				   inet6_sk(asoc->base.sk)->srcprefs,
+				   &saddr->v6.sin6_addr);
 		SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " NIP6_FMT "\n",
 				  NIP6(saddr->v6.sin6_addr));
 		return;
-- 
cgit v1.2.3


From c346dca10840a874240c78efe3f39acf4312a1f2 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Tue, 25 Mar 2008 21:47:49 +0900
Subject: [NET] NETNS: Omit net_device->nd_net without CONFIG_NET_NS.

Introduce per-net_device inlines: dev_net(), dev_net_set().
Without CONFIG_NET_NS, no namespace other than &init_net exists.
Let's explicitly define them to help compiler optimizations.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 arch/ia64/hp/sim/simeth.c           |  2 +-
 drivers/block/aoe/aoenet.c          |  2 +-
 drivers/net/bonding/bond_3ad.c      |  2 +-
 drivers/net/bonding/bond_alb.c      |  2 +-
 drivers/net/bonding/bond_main.c     |  6 ++---
 drivers/net/hamradio/bpqether.c     |  4 ++--
 drivers/net/loopback.c              |  2 +-
 drivers/net/macvlan.c               |  2 +-
 drivers/net/pppoe.c                 |  6 ++---
 drivers/net/veth.c                  |  2 +-
 drivers/net/via-velocity.c          |  2 +-
 drivers/net/wan/dlci.c              |  2 +-
 drivers/net/wan/hdlc.c              |  4 ++--
 drivers/net/wan/lapbether.c         |  4 ++--
 drivers/net/wan/syncppp.c           |  2 +-
 drivers/s390/net/qeth_l3_main.c     |  2 +-
 include/linux/inetdevice.h          |  6 ++---
 include/linux/netdevice.h           | 25 ++++++++++++++++++++-
 net/8021q/vlan.c                    |  2 +-
 net/8021q/vlan_dev.c                |  2 +-
 net/appletalk/aarp.c                |  4 ++--
 net/appletalk/ddp.c                 |  6 ++---
 net/atm/clip.c                      |  2 +-
 net/atm/mpc.c                       |  2 +-
 net/ax25/af_ax25.c                  |  2 +-
 net/ax25/ax25_in.c                  |  2 +-
 net/bridge/br_notify.c              |  2 +-
 net/bridge/br_stp_bpdu.c            |  2 +-
 net/can/af_can.c                    |  4 ++--
 net/can/bcm.c                       |  2 +-
 net/can/raw.c                       |  2 +-
 net/core/dev.c                      | 22 +++++++++----------
 net/core/dst.c                      |  2 +-
 net/core/fib_rules.c                |  2 +-
 net/core/neighbour.c                | 12 +++++-----
 net/core/pktgen.c                   |  2 +-
 net/core/rtnetlink.c                |  4 ++--
 net/decnet/af_decnet.c              |  2 +-
 net/decnet/dn_route.c               |  2 +-
 net/econet/af_econet.c              |  4 ++--
 net/ipv4/arp.c                      | 14 ++++++------
 net/ipv4/devinet.c                  | 10 ++++-----
 net/ipv4/fib_frontend.c             | 12 +++++-----
 net/ipv4/icmp.c                     |  8 +++----
 net/ipv4/igmp.c                     | 16 +++++++-------
 net/ipv4/ip_fragment.c              |  2 +-
 net/ipv4/ip_gre.c                   |  2 +-
 net/ipv4/ip_input.c                 |  6 ++---
 net/ipv4/ip_options.c               |  2 +-
 net/ipv4/ipconfig.c                 |  4 ++--
 net/ipv4/ipmr.c                     |  2 +-
 net/ipv4/netfilter/ip_queue.c       |  2 +-
 net/ipv4/netfilter/ipt_MASQUERADE.c |  2 +-
 net/ipv4/raw.c                      |  4 ++--
 net/ipv4/route.c                    | 28 +++++++++++------------
 net/ipv4/tcp_ipv4.c                 |  6 ++---
 net/ipv4/udp.c                      |  4 ++--
 net/ipv4/xfrm4_policy.c             |  2 +-
 net/ipv6/addrconf.c                 | 44 ++++++++++++++++++-------------------
 net/ipv6/icmp.c                     |  4 ++--
 net/ipv6/ip6_output.c               |  2 +-
 net/ipv6/mcast.c                    |  6 ++---
 net/ipv6/ndisc.c                    | 24 ++++++++++----------
 net/ipv6/netfilter/ip6_queue.c      |  2 +-
 net/ipv6/proc.c                     |  2 +-
 net/ipv6/raw.c                      |  4 ++--
 net/ipv6/reassembly.c               |  2 +-
 net/ipv6/route.c                    | 40 ++++++++++++++++-----------------
 net/ipv6/tcp_ipv6.c                 | 10 ++++-----
 net/ipv6/udp.c                      |  4 ++--
 net/ipv6/xfrm6_policy.c             |  2 +-
 net/ipx/af_ipx.c                    |  4 ++--
 net/irda/irlap_frame.c              |  2 +-
 net/llc/llc_input.c                 |  2 +-
 net/netfilter/core.c                |  2 +-
 net/netfilter/nfnetlink_queue.c     |  2 +-
 net/netlabel/netlabel_unlabeled.c   |  2 +-
 net/netrom/af_netrom.c              |  2 +-
 net/packet/af_packet.c              |  8 +++----
 net/rose/af_rose.c                  |  2 +-
 net/sctp/protocol.c                 |  2 +-
 net/tipc/eth_media.c                |  4 ++--
 net/wireless/wext.c                 |  2 +-
 net/x25/af_x25.c                    |  2 +-
 net/x25/x25_dev.c                   |  2 +-
 net/xfrm/xfrm_policy.c              |  4 ++--
 security/selinux/netif.c            |  2 +-
 87 files changed, 251 insertions(+), 228 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/hp/sim/simeth.c b/arch/ia64/hp/sim/simeth.c
index 969fe9f443c4..3d47839a0c48 100644
--- a/arch/ia64/hp/sim/simeth.c
+++ b/arch/ia64/hp/sim/simeth.c
@@ -294,7 +294,7 @@ simeth_device_event(struct notifier_block *this,unsigned long event, void *ptr)
 		return NOTIFY_DONE;
 	}
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if ( event != NETDEV_UP && event != NETDEV_DOWN ) return NOTIFY_DONE;
diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c
index 8460ef736d56..18d243c73eee 100644
--- a/drivers/block/aoe/aoenet.c
+++ b/drivers/block/aoe/aoenet.c
@@ -115,7 +115,7 @@ aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt,
 	struct aoe_hdr *h;
 	u32 n;
 
-	if (ifp->nd_net != &init_net)
+	if (dev_net(ifp) != &init_net)
 		goto exit;
 
 	skb = skb_share_check(skb, GFP_ATOMIC);
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index cb3c6faa7888..457d81f73e39 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -2429,7 +2429,7 @@ int bond_3ad_lacpdu_recv(struct sk_buff *skb, struct net_device *dev, struct pac
 	struct slave *slave = NULL;
 	int ret = NET_RX_DROP;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto out;
 
 	if (!(dev->flags & IFF_MASTER))
diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index b57bc9467dbe..b986dacf5d33 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -345,7 +345,7 @@ static int rlb_arp_recv(struct sk_buff *skb, struct net_device *bond_dev, struct
 	struct arp_pkt *arp = (struct arp_pkt *)skb->data;
 	int res = NET_RX_DROP;
 
-	if (bond_dev->nd_net != &init_net)
+	if (dev_net(bond_dev) != &init_net)
 		goto out;
 
 	if (!(bond_dev->flags & IFF_MASTER))
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 5fc9d8d58ece..ac688fcb27d7 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2629,7 +2629,7 @@ static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct pack
 	unsigned char *arp_ptr;
 	__be32 sip, tip;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto out;
 
 	if (!(dev->priv_flags & IFF_BONDING) || !(dev->flags & IFF_MASTER))
@@ -3470,7 +3470,7 @@ static int bond_netdev_event(struct notifier_block *this, unsigned long event, v
 {
 	struct net_device *event_dev = (struct net_device *)ptr;
 
-	if (event_dev->nd_net != &init_net)
+	if (dev_net(event_dev) != &init_net)
 		return NOTIFY_DONE;
 
 	dprintk("event_dev: %s, event: %lx\n",
@@ -3508,7 +3508,7 @@ static int bond_inetaddr_event(struct notifier_block *this, unsigned long event,
 	struct bonding *bond, *bond_next;
 	struct vlan_entry *vlan, *vlan_next;
 
-	if (ifa->ifa_dev->dev->nd_net != &init_net)
+	if (dev_net(ifa->ifa_dev->dev) != &init_net)
 		return NOTIFY_DONE;
 
 	list_for_each_entry_safe(bond, bond_next, &bond_dev_list, bond_list) {
diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c
index 5ddf8b0c34f9..5f4b4c6c9f76 100644
--- a/drivers/net/hamradio/bpqether.c
+++ b/drivers/net/hamradio/bpqether.c
@@ -172,7 +172,7 @@ static int bpq_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty
 	struct ethhdr *eth;
 	struct bpqdev *bpq;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
@@ -553,7 +553,7 @@ static int bpq_device_event(struct notifier_block *this,unsigned long event, voi
 {
 	struct net_device *dev = (struct net_device *)ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (!dev_is_ethdev(dev))
diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index f2a6e7132241..41b774baac4d 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -258,7 +258,7 @@ static __net_init int loopback_net_init(struct net *net)
 	if (!dev)
 		goto out;
 
-	dev->nd_net = net;
+	dev_net_set(dev, net);
 	err = register_netdev(dev);
 	if (err)
 		goto out_free_netdev;
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index f651a816b280..2056cfc624dc 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -402,7 +402,7 @@ static int macvlan_newlink(struct net_device *dev,
 	if (!tb[IFLA_LINK])
 		return -EINVAL;
 
-	lowerdev = __dev_get_by_index(dev->nd_net, nla_get_u32(tb[IFLA_LINK]));
+	lowerdev = __dev_get_by_index(dev_net(dev), nla_get_u32(tb[IFLA_LINK]));
 	if (lowerdev == NULL)
 		return -ENODEV;
 
diff --git a/drivers/net/pppoe.c b/drivers/net/pppoe.c
index ac0ac98b19cd..4fad4ddb3504 100644
--- a/drivers/net/pppoe.c
+++ b/drivers/net/pppoe.c
@@ -301,7 +301,7 @@ static int pppoe_device_event(struct notifier_block *this,
 {
 	struct net_device *dev = (struct net_device *) ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* Only look at sockets that are using this specific device. */
@@ -392,7 +392,7 @@ static int pppoe_rcv(struct sk_buff *skb,
 	if (!(skb = skb_share_check(skb, GFP_ATOMIC)))
 		goto out;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	if (!pskb_may_pull(skb, sizeof(struct pppoe_hdr)))
@@ -424,7 +424,7 @@ static int pppoe_disc_rcv(struct sk_buff *skb,
 	struct pppoe_hdr *ph;
 	struct pppox_sock *po;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto abort;
 
 	if (!pskb_may_pull(skb, sizeof(struct pppoe_hdr)))
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index e2ad98bee6e7..31cd817f33f9 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -375,7 +375,7 @@ static int veth_newlink(struct net_device *dev,
 	else
 		snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
 
-	peer = rtnl_create_link(dev->nd_net, ifname, &veth_link_ops, tbp);
+	peer = rtnl_create_link(dev_net(dev), ifname, &veth_link_ops, tbp);
 	if (IS_ERR(peer))
 		return PTR_ERR(peer);
 
diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c
index 1525e8a89844..ed1afaf683a4 100644
--- a/drivers/net/via-velocity.c
+++ b/drivers/net/via-velocity.c
@@ -3464,7 +3464,7 @@ static int velocity_netdev_event(struct notifier_block *nb, unsigned long notifi
 	struct velocity_info *vptr;
 	unsigned long flags;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	spin_lock_irqsave(&velocity_dev_list_lock, flags);
diff --git a/drivers/net/wan/dlci.c b/drivers/net/wan/dlci.c
index 96b232446c0b..b14242768fad 100644
--- a/drivers/net/wan/dlci.c
+++ b/drivers/net/wan/dlci.c
@@ -517,7 +517,7 @@ static int dlci_dev_event(struct notifier_block *unused,
 {
 	struct net_device *dev = (struct net_device *) ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event == NETDEV_UNREGISTER) {
diff --git a/drivers/net/wan/hdlc.c b/drivers/net/wan/hdlc.c
index 39951d0c34d6..9a83c9d5b8cf 100644
--- a/drivers/net/wan/hdlc.c
+++ b/drivers/net/wan/hdlc.c
@@ -68,7 +68,7 @@ static int hdlc_rcv(struct sk_buff *skb, struct net_device *dev,
 {
 	struct hdlc_device *hdlc = dev_to_hdlc(dev);
 
-	if (dev->nd_net != &init_net) {
+	if (dev_net(dev) != &init_net) {
 		kfree_skb(skb);
 		return 0;
 	}
@@ -105,7 +105,7 @@ static int hdlc_device_event(struct notifier_block *this, unsigned long event,
 	unsigned long flags;
 	int on;
  
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (dev->get_stats != hdlc_get_stats)
diff --git a/drivers/net/wan/lapbether.c b/drivers/net/wan/lapbether.c
index fb37b8095231..629c909e05f9 100644
--- a/drivers/net/wan/lapbether.c
+++ b/drivers/net/wan/lapbether.c
@@ -91,7 +91,7 @@ static int lapbeth_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	int len, err;
 	struct lapbethdev *lapbeth;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
@@ -393,7 +393,7 @@ static int lapbeth_device_event(struct notifier_block *this,
 	struct lapbethdev *lapbeth;
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (!dev_is_ethdev(dev))
diff --git a/drivers/net/wan/syncppp.c b/drivers/net/wan/syncppp.c
index 61e24b7a45a3..29b4b94e4947 100644
--- a/drivers/net/wan/syncppp.c
+++ b/drivers/net/wan/syncppp.c
@@ -1444,7 +1444,7 @@ static void sppp_print_bytes (u_char *p, u16 len)
 
 static int sppp_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *p, struct net_device *orig_dev)
 {
-	if (dev->nd_net != &init_net) {
+	if (dev_net(dev) != &init_net) {
 		kfree_skb(skb);
 		return 0;
 	}
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index a856cb47fc78..21c439046b3c 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -3250,7 +3250,7 @@ static int qeth_l3_ip_event(struct notifier_block *this,
 	struct qeth_ipaddr *addr;
 	struct qeth_card *card;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	QETH_DBF_TEXT(trace, 3, "ipevent");
diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index da05ab47ff2f..7009b0cdd06f 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -70,13 +70,13 @@ static inline void ipv4_devconf_setall(struct in_device *in_dev)
 	ipv4_devconf_set((in_dev), NET_IPV4_CONF_ ## attr, (val))
 
 #define IN_DEV_ANDCONF(in_dev, attr) \
-	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr) && \
+	(IPV4_DEVCONF_ALL(dev_net(in_dev->dev), attr) && \
 	 IN_DEV_CONF_GET((in_dev), attr))
 #define IN_DEV_ORCONF(in_dev, attr) \
-	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr) || \
+	(IPV4_DEVCONF_ALL(dev_net(in_dev->dev), attr) || \
 	 IN_DEV_CONF_GET((in_dev), attr))
 #define IN_DEV_MAXCONF(in_dev, attr) \
-	(max(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr), \
+	(max(IPV4_DEVCONF_ALL(dev_net(in_dev->dev), attr), \
 	     IN_DEV_CONF_GET((in_dev), attr)))
 
 #define IN_DEV_FORWARD(in_dev)		IN_DEV_CONF_GET((in_dev), FORWARDING)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ced61f87660e..d146be40f46c 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -708,8 +708,10 @@ struct net_device
 	void                    (*poll_controller)(struct net_device *dev);
 #endif
 
+#ifdef CONFIG_NET_NS
 	/* Network namespace this network device is inside */
 	struct net		*nd_net;
+#endif
 
 	/* bridge stuff */
 	struct net_bridge_port	*br_port;
@@ -737,6 +739,27 @@ struct net_device
 #define	NETDEV_ALIGN		32
 #define	NETDEV_ALIGN_CONST	(NETDEV_ALIGN - 1)
 
+/*
+ * Net namespace inlines
+ */
+static inline
+struct net *dev_net(const struct net_device *dev)
+{
+#ifdef CONFIG_NET_NS
+	return dev->nd_net;
+#else
+	return &init_net;
+#endif
+}
+
+static inline
+void dev_net_set(struct net_device *dev, const struct net *net)
+{
+#ifdef CONFIG_NET_NS
+	dev->nd_dev = net;
+#endif
+}
+
 /**
  *	netdev_priv - access network device private data
  *	@dev: network device
@@ -813,7 +836,7 @@ static inline struct net_device *next_net_device(struct net_device *dev)
 	struct list_head *lh;
 	struct net *net;
 
-	net = dev->nd_net;
+	net = dev_net(dev);
 	lh = dev->dev_list.next;
 	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
 }
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index dbc81b965096..c35dc230365c 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -382,7 +382,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 	int i, flgs;
 	struct net_device *vlandev;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (!grp)
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 1e5c9904571d..e536162b1ebc 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -153,7 +153,7 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 	struct net_device_stats *stats;
 	unsigned short vlan_TCI;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto err_free;
 
 	skb = skb_share_check(skb, GFP_ATOMIC);
diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c
index 61166f66479f..25aa37ce9430 100644
--- a/net/appletalk/aarp.c
+++ b/net/appletalk/aarp.c
@@ -333,7 +333,7 @@ static int aarp_device_event(struct notifier_block *this, unsigned long event,
 	struct net_device *dev = ptr;
 	int ct;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event == NETDEV_DOWN) {
@@ -716,7 +716,7 @@ static int aarp_rcv(struct sk_buff *skb, struct net_device *dev,
 	struct atalk_addr sa, *ma, da;
 	struct atalk_iface *ifa;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto out0;
 
 	/* We only do Ethernet SNAP AARP. */
diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c
index 3be55c8ca4ef..44cd42f7786b 100644
--- a/net/appletalk/ddp.c
+++ b/net/appletalk/ddp.c
@@ -648,7 +648,7 @@ static int ddp_device_event(struct notifier_block *this, unsigned long event,
 {
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event == NETDEV_DOWN)
@@ -1405,7 +1405,7 @@ static int atalk_rcv(struct sk_buff *skb, struct net_device *dev,
 	int origlen;
 	__u16 len_hops;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto freeit;
 
 	/* Don't mangle buffer if shared */
@@ -1493,7 +1493,7 @@ freeit:
 static int ltalk_rcv(struct sk_buff *skb, struct net_device *dev,
 		     struct packet_type *pt, struct net_device *orig_dev)
 {
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto freeit;
 
 	/* Expand any short form frames */
diff --git a/net/atm/clip.c b/net/atm/clip.c
index e82da6746723..6f8223ebf551 100644
--- a/net/atm/clip.c
+++ b/net/atm/clip.c
@@ -612,7 +612,7 @@ static int clip_device_event(struct notifier_block *this, unsigned long event,
 {
 	struct net_device *dev = arg;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event == NETDEV_UNREGISTER) {
diff --git a/net/atm/mpc.c b/net/atm/mpc.c
index 9c7f712fc7e9..9db332e7a6c0 100644
--- a/net/atm/mpc.c
+++ b/net/atm/mpc.c
@@ -964,7 +964,7 @@ static int mpoa_event_listener(struct notifier_block *mpoa_notifier, unsigned lo
 
 	dev = (struct net_device *)dev_ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (dev->name == NULL || strncmp(dev->name, "lec", 3))
diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c
index 48bfcc741f25..ee9dd83e7561 100644
--- a/net/ax25/af_ax25.c
+++ b/net/ax25/af_ax25.c
@@ -116,7 +116,7 @@ static int ax25_device_event(struct notifier_block *this, unsigned long event,
 {
 	struct net_device *dev = (struct net_device *)ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* Reject non AX.25 devices */
diff --git a/net/ax25/ax25_in.c b/net/ax25/ax25_in.c
index d1be080dcb25..33790a8efbc8 100644
--- a/net/ax25/ax25_in.c
+++ b/net/ax25/ax25_in.c
@@ -451,7 +451,7 @@ int ax25_kiss_rcv(struct sk_buff *skb, struct net_device *dev,
 	skb->sk = NULL;		/* Initially we don't know who it's for */
 	skb->destructor = NULL;	/* Who initializes this, dammit?! */
 
-	if (dev->nd_net != &init_net) {
+	if (dev_net(dev) != &init_net) {
 		kfree_skb(skb);
 		return 0;
 	}
diff --git a/net/bridge/br_notify.c b/net/bridge/br_notify.c
index 07ac3ae68d8f..00644a544e3c 100644
--- a/net/bridge/br_notify.c
+++ b/net/bridge/br_notify.c
@@ -37,7 +37,7 @@ static int br_device_event(struct notifier_block *unused, unsigned long event, v
 	struct net_bridge_port *p = dev->br_port;
 	struct net_bridge *br;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* not a port of a bridge */
diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c
index 0edbd2a1c3f3..8deab645ef75 100644
--- a/net/bridge/br_stp_bpdu.c
+++ b/net/bridge/br_stp_bpdu.c
@@ -142,7 +142,7 @@ int br_stp_rcv(struct sk_buff *skb, struct net_device *dev,
 	struct net_bridge *br;
 	const unsigned char *buf;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto err;
 
 	if (!p)
diff --git a/net/can/af_can.c b/net/can/af_can.c
index 36b9f22ed83a..2759b76f731c 100644
--- a/net/can/af_can.c
+++ b/net/can/af_can.c
@@ -599,7 +599,7 @@ static int can_rcv(struct sk_buff *skb, struct net_device *dev,
 	struct dev_rcv_lists *d;
 	int matches;
 
-	if (dev->type != ARPHRD_CAN || dev->nd_net != &init_net) {
+	if (dev->type != ARPHRD_CAN || dev_net(dev) != &init_net) {
 		kfree_skb(skb);
 		return 0;
 	}
@@ -710,7 +710,7 @@ static int can_notifier(struct notifier_block *nb, unsigned long msg,
 	struct net_device *dev = (struct net_device *)data;
 	struct dev_rcv_lists *d;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (dev->type != ARPHRD_CAN)
diff --git a/net/can/bcm.c b/net/can/bcm.c
index bd4282dae754..e9f99b2c6bc9 100644
--- a/net/can/bcm.c
+++ b/net/can/bcm.c
@@ -1285,7 +1285,7 @@ static int bcm_notifier(struct notifier_block *nb, unsigned long msg,
 	struct bcm_op *op;
 	int notify_enodev = 0;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (dev->type != ARPHRD_CAN)
diff --git a/net/can/raw.c b/net/can/raw.c
index 94cd7f27c444..ead50c7c0d40 100644
--- a/net/can/raw.c
+++ b/net/can/raw.c
@@ -210,7 +210,7 @@ static int raw_notifier(struct notifier_block *nb,
 	struct raw_sock *ro = container_of(nb, struct raw_sock, notifier);
 	struct sock *sk = &ro->sk;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (dev->type != ARPHRD_CAN)
diff --git a/net/core/dev.c b/net/core/dev.c
index aebd08606040..812534828914 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -216,7 +216,7 @@ static inline struct hlist_head *dev_index_hash(struct net *net, int ifindex)
 /* Device list insertion */
 static int list_netdevice(struct net_device *dev)
 {
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 
 	ASSERT_RTNL();
 
@@ -852,8 +852,8 @@ int dev_alloc_name(struct net_device *dev, const char *name)
 	struct net *net;
 	int ret;
 
-	BUG_ON(!dev->nd_net);
-	net = dev->nd_net;
+	BUG_ON(!dev_net(dev));
+	net = dev_net(dev);
 	ret = __dev_alloc_name(net, name, buf);
 	if (ret >= 0)
 		strlcpy(dev->name, buf, IFNAMSIZ);
@@ -877,9 +877,9 @@ int dev_change_name(struct net_device *dev, char *newname)
 	struct net *net;
 
 	ASSERT_RTNL();
-	BUG_ON(!dev->nd_net);
+	BUG_ON(!dev_net(dev));
 
-	net = dev->nd_net;
+	net = dev_net(dev);
 	if (dev->flags & IFF_UP)
 		return -EBUSY;
 
@@ -2615,7 +2615,7 @@ static int ptype_seq_show(struct seq_file *seq, void *v)
 
 	if (v == SEQ_START_TOKEN)
 		seq_puts(seq, "Type Device      Function\n");
-	else if (pt->dev == NULL || pt->dev->nd_net == seq_file_net(seq)) {
+	else if (pt->dev == NULL || dev_net(pt->dev) == seq_file_net(seq)) {
 		if (pt->type == htons(ETH_P_ALL))
 			seq_puts(seq, "ALL ");
 		else
@@ -3689,8 +3689,8 @@ int register_netdevice(struct net_device *dev)
 
 	/* When net_device's are persistent, this will be fatal. */
 	BUG_ON(dev->reg_state != NETREG_UNINITIALIZED);
-	BUG_ON(!dev->nd_net);
-	net = dev->nd_net;
+	BUG_ON(!dev_net(dev));
+	net = dev_net(dev);
 
 	spin_lock_init(&dev->queue_lock);
 	spin_lock_init(&dev->_xmit_lock);
@@ -4011,7 +4011,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 	dev = (struct net_device *)
 		(((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST);
 	dev->padded = (char *)dev - (char *)p;
-	dev->nd_net = &init_net;
+	dev_net_set(dev, &init_net);
 
 	if (sizeof_priv) {
 		dev->priv = ((char *)dev +
@@ -4136,7 +4136,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char
 
 	/* Get out if there is nothing todo */
 	err = 0;
-	if (dev->nd_net == net)
+	if (dev_net(dev) == net)
 		goto out;
 
 	/* Pick the destination device name, and ensure
@@ -4187,7 +4187,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char
 	dev_addr_discard(dev);
 
 	/* Actually switch the network namespace */
-	dev->nd_net = net;
+	dev_net_set(dev, net);
 
 	/* Assign the new device name */
 	if (destname != dev->name)
diff --git a/net/core/dst.c b/net/core/dst.c
index 3a01a819ba47..694cd2a3f6d2 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -279,7 +279,7 @@ static inline void dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 	if (!unregister) {
 		dst->input = dst->output = dst_discard;
 	} else {
-		dst->dev = dst->dev->nd_net->loopback_dev;
+		dst->dev = dev_net(dst->dev)->loopback_dev;
 		dev_hold(dst->dev);
 		dev_put(dev);
 		if (dst->neighbour && dst->neighbour->dev == dev) {
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 42ccaf5b8509..942be93a2eb0 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -618,7 +618,7 @@ static int fib_rules_event(struct notifier_block *this, unsigned long event,
 			    void *ptr)
 {
 	struct net_device *dev = ptr;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct fib_rules_ops *ops;
 
 	ASSERT_RTNL();
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 23c0a10c0c37..c978bd1cd659 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -388,7 +388,7 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net,
 	hash_val = tbl->hash(pkey, NULL);
 	for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) {
 		if (!memcmp(n->primary_key, pkey, key_len) &&
-		    (net == n->dev->nd_net)) {
+		    dev_net(n->dev) == net) {
 			neigh_hold(n);
 			NEIGH_CACHE_STAT_INC(tbl, hits);
 			break;
@@ -1298,7 +1298,7 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev,
 	struct neigh_parms *p, *ref;
 	struct net *net;
 
-	net = dev->nd_net;
+	net = dev_net(dev);
 	ref = lookup_neigh_params(tbl, net, 0);
 	if (!ref)
 		return NULL;
@@ -2050,7 +2050,7 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb,
 			s_idx = 0;
 		for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next) {
 			int lidx;
-			if (n->dev->nd_net != net)
+			if (dev_net(n->dev) != net)
 				continue;
 			lidx = idx++;
 			if (lidx < s_idx)
@@ -2155,7 +2155,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq)
 		n = tbl->hash_buckets[bucket];
 
 		while (n) {
-			if (n->dev->nd_net != net)
+			if (dev_net(n->dev) != net)
 				goto next;
 			if (state->neigh_sub_iter) {
 				loff_t fakep = 0;
@@ -2198,7 +2198,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 
 	while (1) {
 		while (n) {
-			if (n->dev->nd_net != net)
+			if (dev_net(n->dev) != net)
 				goto next;
 			if (state->neigh_sub_iter) {
 				void *v = state->neigh_sub_iter(state, n, pos);
@@ -2482,7 +2482,7 @@ static inline size_t neigh_nlmsg_size(void)
 
 static void __neigh_notify(struct neighbour *n, int type, int flags)
 {
-	struct net *net = n->dev->nd_net;
+	struct net *net = dev_net(n->dev);
 	struct sk_buff *skb;
 	int err = -ENOBUFS;
 
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 20e63b302ba6..a803b442234c 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -1874,7 +1874,7 @@ static int pktgen_device_event(struct notifier_block *unused,
 {
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* It is OK that we do not hold the group lock right now,
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 2bd9c5f7627d..09250a0800f6 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -972,7 +972,7 @@ struct net_device *rtnl_create_link(struct net *net, char *ifname,
 			goto err_free;
 	}
 
-	dev->nd_net = net;
+	dev_net_set(dev, net);
 	dev->rtnl_link_ops = ops;
 
 	if (tb[IFLA_MTU])
@@ -1198,7 +1198,7 @@ static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
 
 void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
 {
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct sk_buff *skb;
 	int err = -ENOBUFS;
 
diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index 23fd95a7ad15..3554fb3d251c 100644
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -2089,7 +2089,7 @@ static int dn_device_event(struct notifier_block *this, unsigned long event,
 {
 	struct net_device *dev = (struct net_device *)ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	switch(event) {
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 9dc0abb50eaf..0a46b6c10e51 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -580,7 +580,7 @@ int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type
 	struct dn_dev *dn = (struct dn_dev *)dev->dn_ptr;
 	unsigned char padlen = 0;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto dump_it;
 
 	if (dn == NULL)
diff --git a/net/econet/af_econet.c b/net/econet/af_econet.c
index bc0f6252613f..68d154480043 100644
--- a/net/econet/af_econet.c
+++ b/net/econet/af_econet.c
@@ -1064,7 +1064,7 @@ static int econet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet
 	struct sock *sk;
 	struct ec_device *edev = dev->ec_ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	if (skb->pkt_type == PACKET_OTHERHOST)
@@ -1121,7 +1121,7 @@ static int econet_notifier(struct notifier_block *this, unsigned long msg, void
 	struct net_device *dev = (struct net_device *)data;
 	struct ec_device *edev;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	switch (msg) {
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 832473e30b36..3ce2e137e7bc 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -242,7 +242,7 @@ static int arp_constructor(struct neighbour *neigh)
 		return -EINVAL;
 	}
 
-	neigh->type = inet_addr_type(dev->nd_net, addr);
+	neigh->type = inet_addr_type(dev_net(dev), addr);
 
 	parms = in_dev->arp_parms;
 	__neigh_parms_put(neigh->parms);
@@ -341,14 +341,14 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
 	switch (IN_DEV_ARP_ANNOUNCE(in_dev)) {
 	default:
 	case 0:		/* By default announce any local IP */
-		if (skb && inet_addr_type(dev->nd_net, ip_hdr(skb)->saddr) == RTN_LOCAL)
+		if (skb && inet_addr_type(dev_net(dev), ip_hdr(skb)->saddr) == RTN_LOCAL)
 			saddr = ip_hdr(skb)->saddr;
 		break;
 	case 1:		/* Restrict announcements of saddr in same subnet */
 		if (!skb)
 			break;
 		saddr = ip_hdr(skb)->saddr;
-		if (inet_addr_type(dev->nd_net, saddr) == RTN_LOCAL) {
+		if (inet_addr_type(dev_net(dev), saddr) == RTN_LOCAL) {
 			/* saddr should be known to target */
 			if (inet_addr_onlink(in_dev, target, saddr))
 				break;
@@ -424,7 +424,7 @@ static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev)
 	int flag = 0;
 	/*unsigned long now; */
 
-	if (ip_route_output_key(dev->nd_net, &rt, &fl) < 0)
+	if (ip_route_output_key(dev_net(dev), &rt, &fl) < 0)
 		return 1;
 	if (rt->u.dst.dev != dev) {
 		NET_INC_STATS_BH(LINUX_MIB_ARPFILTER);
@@ -477,7 +477,7 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb)
 
 	paddr = skb->rtable->rt_gateway;
 
-	if (arp_set_predefined(inet_addr_type(dev->nd_net, paddr), haddr, paddr, dev))
+	if (arp_set_predefined(inet_addr_type(dev_net(dev), paddr), haddr, paddr, dev))
 		return 0;
 
 	n = __neigh_lookup(&arp_tbl, &paddr, dev, 1);
@@ -709,7 +709,7 @@ static int arp_process(struct sk_buff *skb)
 	u16 dev_type = dev->type;
 	int addr_type;
 	struct neighbour *n;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 
 	/* arp_rcv below verifies the ARP header and verifies the device
 	 * is ARP'able.
@@ -858,7 +858,7 @@ static int arp_process(struct sk_buff *skb)
 
 	n = __neigh_lookup(&arp_tbl, &sip, dev, 0);
 
-	if (IPV4_DEVCONF_ALL(dev->nd_net, ARP_ACCEPT)) {
+	if (IPV4_DEVCONF_ALL(dev_net(dev), ARP_ACCEPT)) {
 		/* Unsolicited ARP is not accepted by default.
 		   It is possible, that this option should be enabled for some
 		   devices (strip is candidate)
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 4a10dbbbe0a1..823c724a8593 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -165,7 +165,7 @@ static struct in_device *inetdev_init(struct net_device *dev)
 	if (!in_dev)
 		goto out;
 	INIT_RCU_HEAD(&in_dev->rcu_head);
-	memcpy(&in_dev->cnf, dev->nd_net->ipv4.devconf_dflt,
+	memcpy(&in_dev->cnf, dev_net(dev)->ipv4.devconf_dflt,
 			sizeof(in_dev->cnf));
 	in_dev->cnf.sysctl = NULL;
 	in_dev->dev = dev;
@@ -872,7 +872,7 @@ __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope)
 {
 	__be32 addr = 0;
 	struct in_device *in_dev;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 
 	rcu_read_lock();
 	in_dev = __in_dev_get_rcu(dev);
@@ -974,7 +974,7 @@ __be32 inet_confirm_addr(struct in_device *in_dev,
 	if (scope != RT_SCOPE_LINK)
 		return confirm_addr_indev(in_dev, dst, local, scope);
 
-	net = in_dev->dev->nd_net;
+	net = dev_net(in_dev->dev);
 	read_lock(&dev_base_lock);
 	rcu_read_lock();
 	for_each_netdev(net, dev) {
@@ -1203,7 +1203,7 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa, struct nlmsghdr *nlh,
 	int err = -ENOBUFS;
 	struct net *net;
 
-	net = ifa->ifa_dev->dev->nd_net;
+	net = dev_net(ifa->ifa_dev->dev);
 	skb = nlmsg_new(inet_nlmsg_size(), GFP_KERNEL);
 	if (skb == NULL)
 		goto errout;
@@ -1517,7 +1517,7 @@ static void devinet_sysctl_register(struct in_device *idev)
 {
 	neigh_sysctl_register(idev->dev, idev->arp_parms, NET_IPV4,
 			NET_IPV4_NEIGH, "ipv4", NULL, NULL);
-	__devinet_sysctl_register(idev->dev->nd_net, idev->dev->name,
+	__devinet_sysctl_register(dev_net(idev->dev), idev->dev->name,
 			idev->dev->ifindex, &idev->cnf);
 }
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 86ff2711fc95..0e4b34b07cb5 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -257,7 +257,7 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
 	if (in_dev == NULL)
 		goto e_inval;
 
-	net = dev->nd_net;
+	net = dev_net(dev);
 	if (fib_lookup(net, &fl, &res))
 		goto last_resort;
 	if (res.type != RTN_UNICAST)
@@ -674,7 +674,7 @@ out:
 
 static void fib_magic(int cmd, int type, __be32 dst, int dst_len, struct in_ifaddr *ifa)
 {
-	struct net *net = ifa->ifa_dev->dev->nd_net;
+	struct net *net = dev_net(ifa->ifa_dev->dev);
 	struct fib_table *tb;
 	struct fib_config cfg = {
 		.fc_protocol = RTPROT_KERNEL,
@@ -801,15 +801,15 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
 		fib_magic(RTM_DELROUTE, RTN_LOCAL, ifa->ifa_local, 32, prim);
 
 		/* Check, that this local address finally disappeared. */
-		if (inet_addr_type(dev->nd_net, ifa->ifa_local) != RTN_LOCAL) {
+		if (inet_addr_type(dev_net(dev), ifa->ifa_local) != RTN_LOCAL) {
 			/* And the last, but not the least thing.
 			   We must flush stray FIB entries.
 
 			   First of all, we scan fib_info list searching
 			   for stray nexthop entries, then ignite fib_flush.
 			*/
-			if (fib_sync_down_addr(dev->nd_net, ifa->ifa_local))
-				fib_flush(dev->nd_net);
+			if (fib_sync_down_addr(dev_net(dev), ifa->ifa_local))
+				fib_flush(dev_net(dev));
 		}
 	}
 #undef LOCAL_OK
@@ -899,7 +899,7 @@ static void nl_fib_lookup_exit(struct net *net)
 static void fib_disable_ip(struct net_device *dev, int force)
 {
 	if (fib_sync_down_dev(dev, force))
-		fib_flush(dev->nd_net);
+		fib_flush(dev_net(dev));
 	rt_cache_flush(0);
 	arp_ifdown(dev);
 }
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index ff9a8e643fcc..f38f093ef751 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -351,7 +351,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
 	struct sock *sk;
 	struct sk_buff *skb;
 
-	sk = icmp_sk(rt->u.dst.dev->nd_net);
+	sk = icmp_sk(dev_net(rt->u.dst.dev));
 	if (ip_append_data(sk, icmp_glue_bits, icmp_param,
 			   icmp_param->data_len+icmp_param->head_len,
 			   icmp_param->head_len,
@@ -382,7 +382,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 {
 	struct ipcm_cookie ipc;
 	struct rtable *rt = skb->rtable;
-	struct net *net = rt->u.dst.dev->nd_net;
+	struct net *net = dev_net(rt->u.dst.dev);
 	struct sock *sk = icmp_sk(net);
 	struct inet_sock *inet = inet_sk(sk);
 	__be32 daddr;
@@ -447,7 +447,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 
 	if (!rt)
 		goto out;
-	net = rt->u.dst.dev->nd_net;
+	net = dev_net(rt->u.dst.dev);
 	sk = icmp_sk(net);
 
 	/*
@@ -677,7 +677,7 @@ static void icmp_unreach(struct sk_buff *skb)
 	u32 info = 0;
 	struct net *net;
 
-	net = skb->dst->dev->nd_net;
+	net = dev_net(skb->dst->dev);
 
 	/*
 	 *	Incomplete header ?
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 6a4ee8da6994..682f632bfb77 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -130,12 +130,12 @@
  */
 
 #define IGMP_V1_SEEN(in_dev) \
-	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, FORCE_IGMP_VERSION) == 1 || \
+	(IPV4_DEVCONF_ALL(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 1 || \
 	 IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 1 || \
 	 ((in_dev)->mr_v1_seen && \
 	  time_before(jiffies, (in_dev)->mr_v1_seen)))
 #define IGMP_V2_SEEN(in_dev) \
-	(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, FORCE_IGMP_VERSION) == 2 || \
+	(IPV4_DEVCONF_ALL(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 2 || \
 	 IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 2 || \
 	 ((in_dev)->mr_v2_seen && \
 	  time_before(jiffies, (in_dev)->mr_v2_seen)))
@@ -1198,7 +1198,7 @@ void ip_mc_inc_group(struct in_device *in_dev, __be32 addr)
 
 	ASSERT_RTNL();
 
-	if (in_dev->dev->nd_net != &init_net)
+	if (dev_net(in_dev->dev) != &init_net)
 		return;
 
 	for (im=in_dev->mc_list; im; im=im->next) {
@@ -1280,7 +1280,7 @@ void ip_mc_dec_group(struct in_device *in_dev, __be32 addr)
 
 	ASSERT_RTNL();
 
-	if (in_dev->dev->nd_net != &init_net)
+	if (dev_net(in_dev->dev) != &init_net)
 		return;
 
 	for (ip=&in_dev->mc_list; (i=*ip)!=NULL; ip=&i->next) {
@@ -1310,7 +1310,7 @@ void ip_mc_down(struct in_device *in_dev)
 
 	ASSERT_RTNL();
 
-	if (in_dev->dev->nd_net != &init_net)
+	if (dev_net(in_dev->dev) != &init_net)
 		return;
 
 	for (i=in_dev->mc_list; i; i=i->next)
@@ -1333,7 +1333,7 @@ void ip_mc_init_dev(struct in_device *in_dev)
 {
 	ASSERT_RTNL();
 
-	if (in_dev->dev->nd_net != &init_net)
+	if (dev_net(in_dev->dev) != &init_net)
 		return;
 
 	in_dev->mc_tomb = NULL;
@@ -1359,7 +1359,7 @@ void ip_mc_up(struct in_device *in_dev)
 
 	ASSERT_RTNL();
 
-	if (in_dev->dev->nd_net != &init_net)
+	if (dev_net(in_dev->dev) != &init_net)
 		return;
 
 	ip_mc_inc_group(in_dev, IGMP_ALL_HOSTS);
@@ -1378,7 +1378,7 @@ void ip_mc_destroy_dev(struct in_device *in_dev)
 
 	ASSERT_RTNL();
 
-	if (in_dev->dev->nd_net != &init_net)
+	if (dev_net(in_dev->dev) != &init_net)
 		return;
 
 	/* Deactivate timers */
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 8b448c4b9080..fcb60e76b234 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -571,7 +571,7 @@ int ip_defrag(struct sk_buff *skb, u32 user)
 
 	IP_INC_STATS_BH(IPSTATS_MIB_REASMREQDS);
 
-	net = skb->dev ? skb->dev->nd_net : skb->dst->dev->nd_net;
+	net = skb->dev ? dev_net(skb->dev) : dev_net(skb->dst->dev);
 	/* Start by cleaning up the memory. */
 	if (atomic_read(&net->ipv4.frags.mem) > net->ipv4.frags.high_thresh)
 		ip_evictor(net);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index f9ee84420cb3..50972b397a9a 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -1190,7 +1190,7 @@ static int ipgre_close(struct net_device *dev)
 	struct ip_tunnel *t = netdev_priv(dev);
 	if (ipv4_is_multicast(t->parms.iph.daddr) && t->mlink) {
 		struct in_device *in_dev;
-		in_dev = inetdev_by_index(dev->nd_net, t->mlink);
+		in_dev = inetdev_by_index(dev_net(dev), t->mlink);
 		if (in_dev) {
 			ip_mc_dec_group(in_dev, t->parms.iph.daddr);
 			in_dev_put(in_dev);
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 2aeea5d15425..26685c83a146 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -172,7 +172,7 @@ int ip_call_ra_chain(struct sk_buff *skb)
 		if (sk && inet_sk(sk)->num == protocol &&
 		    (!sk->sk_bound_dev_if ||
 		     sk->sk_bound_dev_if == dev->ifindex) &&
-		    sk->sk_net == dev->nd_net) {
+		    sk->sk_net == dev_net(dev)) {
 			if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) {
 				if (ip_defrag(skb, IP_DEFRAG_CALL_RA_CHAIN)) {
 					read_unlock(&ip_ra_lock);
@@ -199,7 +199,7 @@ int ip_call_ra_chain(struct sk_buff *skb)
 
 static int ip_local_deliver_finish(struct sk_buff *skb)
 {
-	struct net *net = skb->dev->nd_net;
+	struct net *net = dev_net(skb->dev);
 
 	__skb_pull(skb, ip_hdrlen(skb));
 
@@ -291,7 +291,7 @@ static inline int ip_rcv_options(struct sk_buff *skb)
 	opt = &(IPCB(skb)->opt);
 	opt->optlen = iph->ihl*4 - sizeof(struct iphdr);
 
-	if (ip_options_compile(dev->nd_net, opt, skb)) {
+	if (ip_options_compile(dev_net(dev), opt, skb)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
 		goto drop;
 	}
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 87cc1222c600..d107543d3f81 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -145,7 +145,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb)
 						__be32 addr;
 
 						memcpy(&addr, sptr+soffset-1, 4);
-						if (inet_addr_type(skb->dst->dev->nd_net, addr) != RTN_LOCAL) {
+						if (inet_addr_type(dev_net(skb->dst->dev), addr) != RTN_LOCAL) {
 							dopt->ts_needtime = 1;
 							soffset += 8;
 						}
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 96138b128de8..08e8fb60d315 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -434,7 +434,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 	unsigned char *sha, *tha;		/* s for "source", t for "target" */
 	struct ic_device *d;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
@@ -854,7 +854,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str
 	struct ic_device *d;
 	int len, ext_len;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	/* Perform verifications before taking the lock.  */
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 7d63d74ef62a..e54bc1364473 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1089,7 +1089,7 @@ static int ipmr_device_event(struct notifier_block *this, unsigned long event, v
 	struct vif_device *v;
 	int ct;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event != NETDEV_UNREGISTER)
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index fe05da41d6ba..500998a2dec1 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -481,7 +481,7 @@ ipq_rcv_dev_event(struct notifier_block *this,
 {
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* Drop any packets associated with the downed device */
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index c6817b18366a..84c26dd27d81 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -120,7 +120,7 @@ static int masq_device_event(struct notifier_block *this,
 {
 	const struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event == NETDEV_DOWN) {
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 3f68a937b602..8756d502a47f 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -168,7 +168,7 @@ static int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash)
 	if (hlist_empty(head))
 		goto out;
 
-	net = skb->dev->nd_net;
+	net = dev_net(skb->dev);
 	sk = __raw_v4_lookup(net, __sk_head(head), iph->protocol,
 			     iph->saddr, iph->daddr,
 			     skb->dev->ifindex);
@@ -276,7 +276,7 @@ void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info)
 	raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]);
 	if (raw_sk != NULL) {
 		iph = (struct iphdr *)skb->data;
-		net = skb->dev->nd_net;
+		net = dev_net(skb->dev);
 
 		while ((raw_sk = __raw_v4_lookup(net, raw_sk, protocol,
 						iph->daddr, iph->saddr,
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 2941ef21f203..7768d718e199 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -284,7 +284,7 @@ static struct rtable *rt_cache_get_first(struct rt_cache_iter_state *st)
 		rcu_read_lock_bh();
 		r = rcu_dereference(rt_hash_table[st->bucket].chain);
 		while (r) {
-			if (r->u.dst.dev->nd_net == st->p.net &&
+			if (dev_net(r->u.dst.dev) == st->p.net &&
 			    r->rt_genid == st->genid)
 				return r;
 			r = rcu_dereference(r->u.dst.rt_next);
@@ -312,7 +312,7 @@ static struct rtable *rt_cache_get_next(struct rt_cache_iter_state *st,
 					struct rtable *r)
 {
 	while ((r = __rt_cache_get_next(st, r)) != NULL) {
-		if (r->u.dst.dev->nd_net != st->p.net)
+		if (dev_net(r->u.dst.dev) != st->p.net)
 			continue;
 		if (r->rt_genid == st->genid)
 			break;
@@ -680,7 +680,7 @@ static inline int compare_keys(struct flowi *fl1, struct flowi *fl2)
 
 static inline int compare_netns(struct rtable *rt1, struct rtable *rt2)
 {
-	return rt1->u.dst.dev->nd_net == rt2->u.dst.dev->nd_net;
+	return dev_net(rt1->u.dst.dev) == dev_net(rt2->u.dst.dev);
 }
 
 /*
@@ -1164,7 +1164,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 	if (!in_dev)
 		return;
 
-	net = dev->nd_net;
+	net = dev_net(dev);
 	if (new_gw == old_gw || !IN_DEV_RX_REDIRECTS(in_dev)
 	    || ipv4_is_multicast(new_gw) || ipv4_is_lbcast(new_gw)
 	    || ipv4_is_zeronet(new_gw))
@@ -1195,7 +1195,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 				    rth->fl.oif != ikeys[k] ||
 				    rth->fl.iif != 0 ||
 				    rth->rt_genid != atomic_read(&rt_genid) ||
-				    rth->u.dst.dev->nd_net != net) {
+				    dev_net(rth->u.dst.dev) != net) {
 					rthp = &rth->u.dst.rt_next;
 					continue;
 				}
@@ -1454,7 +1454,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
 			    rth->rt_src  == iph->saddr &&
 			    rth->fl.iif == 0 &&
 			    !(dst_metric_locked(&rth->u.dst, RTAX_MTU)) &&
-			    rth->u.dst.dev->nd_net == net &&
+			    dev_net(rth->u.dst.dev) == net &&
 			    rth->rt_genid == atomic_read(&rt_genid)) {
 				unsigned short mtu = new_mtu;
 
@@ -1530,9 +1530,9 @@ static void ipv4_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 {
 	struct rtable *rt = (struct rtable *) dst;
 	struct in_device *idev = rt->idev;
-	if (dev != dev->nd_net->loopback_dev && idev && idev->dev == dev) {
+	if (dev != dev_net(dev)->loopback_dev && idev && idev->dev == dev) {
 		struct in_device *loopback_idev =
-			in_dev_get(dev->nd_net->loopback_dev);
+			in_dev_get(dev_net(dev)->loopback_dev);
 		if (loopback_idev) {
 			rt->idev = loopback_idev;
 			in_dev_put(idev);
@@ -1576,7 +1576,7 @@ void ip_rt_get_source(u8 *addr, struct rtable *rt)
 
 	if (rt->fl.iif == 0)
 		src = rt->rt_src;
-	else if (fib_lookup(rt->u.dst.dev->nd_net, &rt->fl, &res) == 0) {
+	else if (fib_lookup(dev_net(rt->u.dst.dev), &rt->fl, &res) == 0) {
 		src = FIB_RES_PREFSRC(res);
 		fib_res_put(&res);
 	} else
@@ -1900,7 +1900,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	__be32		spec_dst;
 	int		err = -EINVAL;
 	int		free_res = 0;
-	struct net    * net = dev->nd_net;
+	struct net    * net = dev_net(dev);
 
 	/* IP on this device is disabled. */
 
@@ -2071,7 +2071,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	int iif = dev->ifindex;
 	struct net *net;
 
-	net = dev->nd_net;
+	net = dev_net(dev);
 	tos &= IPTOS_RT_MASK;
 	hash = rt_hash(daddr, saddr, iif);
 
@@ -2084,7 +2084,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 		    rth->fl.oif == 0 &&
 		    rth->fl.mark == skb->mark &&
 		    rth->fl.fl4_tos == tos &&
-		    rth->u.dst.dev->nd_net == net &&
+		    dev_net(rth->u.dst.dev) == net &&
 		    rth->rt_genid == atomic_read(&rt_genid)) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(in_hit);
@@ -2486,7 +2486,7 @@ int __ip_route_output_key(struct net *net, struct rtable **rp,
 		    rth->fl.mark == flp->mark &&
 		    !((rth->fl.fl4_tos ^ flp->fl4_tos) &
 			    (IPTOS_RT_MASK | RTO_ONLINK)) &&
-		    rth->u.dst.dev->nd_net == net &&
+		    dev_net(rth->u.dst.dev) == net &&
 		    rth->rt_genid == atomic_read(&rt_genid)) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(out_hit);
@@ -2795,7 +2795,7 @@ int ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb)
 		rcu_read_lock_bh();
 		for (rt = rcu_dereference(rt_hash_table[h].chain), idx = 0; rt;
 		     rt = rcu_dereference(rt->u.dst.rt_next), idx++) {
-			if (rt->u.dst.dev->nd_net != net || idx < s_idx)
+			if (dev_net(rt->u.dst.dev) != net || idx < s_idx)
 				continue;
 			if (rt->rt_genid != atomic_read(&rt_genid))
 				continue;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 649d00a50cb1..28bece6f281b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -353,7 +353,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info)
 		return;
 	}
 
-	sk = inet_lookup(skb->dev->nd_net, &tcp_hashinfo, iph->daddr, th->dest,
+	sk = inet_lookup(dev_net(skb->dev), &tcp_hashinfo, iph->daddr, th->dest,
 			iph->saddr, th->source, inet_iif(skb));
 	if (!sk) {
 		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
@@ -1644,7 +1644,7 @@ int tcp_v4_rcv(struct sk_buff *skb)
 	TCP_SKB_CB(skb)->flags	 = iph->tos;
 	TCP_SKB_CB(skb)->sacked	 = 0;
 
-	sk = __inet_lookup(skb->dev->nd_net, &tcp_hashinfo, iph->saddr,
+	sk = __inet_lookup(dev_net(skb->dev), &tcp_hashinfo, iph->saddr,
 			th->source, iph->daddr, th->dest, inet_iif(skb));
 	if (!sk)
 		goto no_tcp_socket;
@@ -1718,7 +1718,7 @@ do_time_wait:
 	}
 	switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) {
 	case TCP_TW_SYN: {
-		struct sock *sk2 = inet_lookup_listener(skb->dev->nd_net,
+		struct sock *sk2 = inet_lookup_listener(dev_net(skb->dev),
 							&tcp_hashinfo,
 							iph->daddr, th->dest,
 							inet_iif(skb));
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index b37581dfd029..e2cd93481359 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -357,7 +357,7 @@ void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[])
 	int harderr;
 	int err;
 
-	sk = __udp4_lib_lookup(skb->dev->nd_net, iph->daddr, uh->dest,
+	sk = __udp4_lib_lookup(dev_net(skb->dev), iph->daddr, uh->dest,
 			iph->saddr, uh->source, skb->dev->ifindex, udptable);
 	if (sk == NULL) {
 		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
@@ -1181,7 +1181,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
 		return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable);
 
-	sk = __udp4_lib_lookup(skb->dev->nd_net, saddr, uh->source, daddr,
+	sk = __udp4_lib_lookup(dev_net(skb->dev), saddr, uh->source, daddr,
 			uh->dest, inet_iif(skb), udptable);
 
 	if (sk != NULL) {
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 10ed70491434..c63de0a72aba 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -221,7 +221,7 @@ static void xfrm4_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 	xdst = (struct xfrm_dst *)dst;
 	if (xdst->u.rt.idev->dev == dev) {
 		struct in_device *loopback_idev =
-			in_dev_get(dev->nd_net->loopback_dev);
+			in_dev_get(dev_net(dev)->loopback_dev);
 		BUG_ON(!loopback_idev);
 
 		do {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 89954885dee1..d1de9ec74261 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -335,7 +335,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 
 	rwlock_init(&ndev->lock);
 	ndev->dev = dev;
-	memcpy(&ndev->cnf, dev->nd_net->ipv6.devconf_dflt, sizeof(ndev->cnf));
+	memcpy(&ndev->cnf, dev_net(dev)->ipv6.devconf_dflt, sizeof(ndev->cnf));
 	ndev->cnf.mtu6 = dev->mtu;
 	ndev->cnf.sysctl = NULL;
 	ndev->nd_parms = neigh_parms_alloc(dev, &nd_tbl);
@@ -561,7 +561,7 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen,
 	write_lock(&addrconf_hash_lock);
 
 	/* Ignore adding duplicate addresses on an interface */
-	if (ipv6_chk_same_addr(idev->dev->nd_net, addr, idev->dev)) {
+	if (ipv6_chk_same_addr(dev_net(idev->dev), addr, idev->dev)) {
 		ADBG(("ipv6_add_addr: already assigned\n"));
 		err = -EEXIST;
 		goto out;
@@ -751,7 +751,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
 	if ((ifp->flags & IFA_F_PERMANENT) && onlink < 1) {
 		struct in6_addr prefix;
 		struct rt6_info *rt;
-		struct net *net = ifp->idev->dev->nd_net;
+		struct net *net = dev_net(ifp->idev->dev);
 		ipv6_addr_prefix(&prefix, &ifp->addr, ifp->prefix_len);
 		rt = rt6_lookup(net, &prefix, NULL, ifp->idev->dev->ifindex, 1);
 
@@ -1044,7 +1044,7 @@ int ipv6_dev_get_saddr(struct net_device *dst_dev,
 {
 	struct ipv6_saddr_score scores[2],
 				*score = &scores[0], *hiscore = &scores[1];
-	struct net *net = dst_dev->nd_net;
+	struct net *net = dev_net(dst_dev);
 	struct ipv6_saddr_dst dst;
 	struct net_device *dev;
 	int dst_type;
@@ -1217,7 +1217,7 @@ int ipv6_chk_addr(struct net *net, struct in6_addr *addr,
 
 	read_lock_bh(&addrconf_hash_lock);
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
-		if (ifp->idev->dev->nd_net != net)
+		if (dev_net(ifp->idev->dev) != net)
 			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr) &&
 		    !(ifp->flags&IFA_F_TENTATIVE)) {
@@ -1239,7 +1239,7 @@ int ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr,
 	u8 hash = ipv6_addr_hash(addr);
 
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
-		if (ifp->idev->dev->nd_net != net)
+		if (dev_net(ifp->idev->dev) != net)
 			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr)) {
 			if (dev == NULL || ifp->idev->dev == dev)
@@ -1257,7 +1257,7 @@ struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr,
 
 	read_lock_bh(&addrconf_hash_lock);
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
-		if (ifp->idev->dev->nd_net != net)
+		if (dev_net(ifp->idev->dev) != net)
 			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr)) {
 			if (dev == NULL || ifp->idev->dev == dev ||
@@ -1559,7 +1559,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev,
 		.fc_expires = expires,
 		.fc_dst_len = plen,
 		.fc_flags = RTF_UP | flags,
-		.fc_nlinfo.nl_net = dev->nd_net,
+		.fc_nlinfo.nl_net = dev_net(dev),
 	};
 
 	ipv6_addr_copy(&cfg.fc_dst, pfx);
@@ -1586,7 +1586,7 @@ static void addrconf_add_mroute(struct net_device *dev)
 		.fc_ifindex = dev->ifindex,
 		.fc_dst_len = 8,
 		.fc_flags = RTF_UP,
-		.fc_nlinfo.nl_net = dev->nd_net,
+		.fc_nlinfo.nl_net = dev_net(dev),
 	};
 
 	ipv6_addr_set(&cfg.fc_dst, htonl(0xFF000000), 0, 0, 0);
@@ -1603,7 +1603,7 @@ static void sit_route_add(struct net_device *dev)
 		.fc_ifindex = dev->ifindex,
 		.fc_dst_len = 96,
 		.fc_flags = RTF_UP | RTF_NONEXTHOP,
-		.fc_nlinfo.nl_net = dev->nd_net,
+		.fc_nlinfo.nl_net = dev_net(dev),
 	};
 
 	/* prefix length - 96 bits "::d.d.d.d" */
@@ -1704,7 +1704,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
 
 	if (pinfo->onlink) {
 		struct rt6_info *rt;
-		rt = rt6_lookup(dev->nd_net, &pinfo->prefix, NULL,
+		rt = rt6_lookup(dev_net(dev), &pinfo->prefix, NULL,
 				dev->ifindex, 1);
 
 		if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
@@ -1748,7 +1748,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
 
 ok:
 
-		ifp = ipv6_get_ifaddr(dev->nd_net, &addr, dev, 1);
+		ifp = ipv6_get_ifaddr(dev_net(dev), &addr, dev, 1);
 
 		if (ifp == NULL && valid_lft) {
 			int max_addresses = in6_dev->cnf.max_addresses;
@@ -2071,7 +2071,7 @@ static void sit_add_v4_addrs(struct inet6_dev *idev)
 	struct inet6_ifaddr * ifp;
 	struct in6_addr addr;
 	struct net_device *dev;
-	struct net *net = idev->dev->nd_net;
+	struct net *net = dev_net(idev->dev);
 	int scope;
 
 	ASSERT_RTNL();
@@ -2261,7 +2261,7 @@ ipv6_inherit_linklocal(struct inet6_dev *idev, struct net_device *link_dev)
 static void ip6_tnl_add_linklocal(struct inet6_dev *idev)
 {
 	struct net_device *link_dev;
-	struct net *net = idev->dev->nd_net;
+	struct net *net = dev_net(idev->dev);
 
 	/* first try to inherit the link-local address from the link device */
 	if (idev->dev->iflink &&
@@ -2442,7 +2442,7 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 {
 	struct inet6_dev *idev;
 	struct inet6_ifaddr *ifa, **bifa;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	int i;
 
 	ASSERT_RTNL();
@@ -2771,7 +2771,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq)
 	for (state->bucket = 0; state->bucket < IN6_ADDR_HSIZE; ++state->bucket) {
 		ifa = inet6_addr_lst[state->bucket];
 
-		while (ifa && ifa->idev->dev->nd_net != net)
+		while (ifa && dev_net(ifa->idev->dev) != net)
 			ifa = ifa->lst_next;
 		if (ifa)
 			break;
@@ -2787,7 +2787,7 @@ static struct inet6_ifaddr *if6_get_next(struct seq_file *seq, struct inet6_ifad
 	ifa = ifa->lst_next;
 try_again:
 	if (ifa) {
-		if (ifa->idev->dev->nd_net != net) {
+		if (dev_net(ifa->idev->dev) != net) {
 			ifa = ifa->lst_next;
 			goto try_again;
 		}
@@ -2905,7 +2905,7 @@ int ipv6_chk_home_addr(struct net *net, struct in6_addr *addr)
 	u8 hash = ipv6_addr_hash(addr);
 	read_lock_bh(&addrconf_hash_lock);
 	for (ifp = inet6_addr_lst[hash]; ifp; ifp = ifp->lst_next) {
-		if (ifp->idev->dev->nd_net != net)
+		if (dev_net(ifp->idev->dev) != net)
 			continue;
 		if (ipv6_addr_cmp(&ifp->addr, addr) == 0 &&
 		    (ifp->flags & IFA_F_HOMEADDRESS)) {
@@ -3469,7 +3469,7 @@ errout:
 static void inet6_ifa_notify(int event, struct inet6_ifaddr *ifa)
 {
 	struct sk_buff *skb;
-	struct net *net = ifa->idev->dev->nd_net;
+	struct net *net = dev_net(ifa->idev->dev);
 	int err = -ENOBUFS;
 
 	skb = nlmsg_new(inet6_ifaddr_msgsize(), GFP_ATOMIC);
@@ -3675,7 +3675,7 @@ cont:
 void inet6_ifinfo_notify(int event, struct inet6_dev *idev)
 {
 	struct sk_buff *skb;
-	struct net *net = idev->dev->nd_net;
+	struct net *net = dev_net(idev->dev);
 	int err = -ENOBUFS;
 
 	skb = nlmsg_new(inet6_if_nlmsg_size(), GFP_ATOMIC);
@@ -3745,7 +3745,7 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev,
 			 struct prefix_info *pinfo)
 {
 	struct sk_buff *skb;
-	struct net *net = idev->dev->nd_net;
+	struct net *net = dev_net(idev->dev);
 	int err = -ENOBUFS;
 
 	skb = nlmsg_new(inet6_prefix_nlmsg_size(), GFP_ATOMIC);
@@ -4157,7 +4157,7 @@ static void addrconf_sysctl_register(struct inet6_dev *idev)
 			      NET_IPV6_NEIGH, "ipv6",
 			      &ndisc_ifinfo_sysctl_change,
 			      NULL);
-	__addrconf_sysctl_register(idev->dev->nd_net, idev->dev->name,
+	__addrconf_sysctl_register(dev_net(idev->dev), idev->dev->name,
 			idev->dev->ifindex, idev, &idev->cnf);
 }
 
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 86332417b402..50857662e6b7 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -306,7 +306,7 @@ static inline void mip6_addr_swap(struct sk_buff *skb) {}
 void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 		 struct net_device *dev)
 {
-	struct net *net = skb->dev->nd_net;
+	struct net *net = dev_net(skb->dev);
 	struct inet6_dev *idev = NULL;
 	struct ipv6hdr *hdr = ipv6_hdr(skb);
 	struct sock *sk;
@@ -507,7 +507,7 @@ EXPORT_SYMBOL(icmpv6_send);
 
 static void icmpv6_echo_reply(struct sk_buff *skb)
 {
-	struct net *net = skb->dev->nd_net;
+	struct net *net = dev_net(skb->dev);
 	struct sock *sk;
 	struct inet6_dev *idev;
 	struct ipv6_pinfo *np;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index d34aa61353bb..556300f0eba5 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -402,7 +402,7 @@ int ip6_forward(struct sk_buff *skb)
 	struct dst_entry *dst = skb->dst;
 	struct ipv6hdr *hdr = ipv6_hdr(skb);
 	struct inet6_skb_parm *opt = IP6CB(skb);
-	struct net *net = dst->dev->nd_net;
+	struct net *net = dev_net(dst->dev);
 
 	if (ipv6_devconf.forwarding == 0)
 		goto error;
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 957ac7e9e929..0357de8e78c8 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -1400,7 +1400,7 @@ mld_scount(struct ifmcaddr6 *pmc, int type, int gdeleted, int sdeleted)
 
 static struct sk_buff *mld_newpack(struct net_device *dev, int size)
 {
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct sock *sk = net->ipv6.igmp_sk;
 	struct sk_buff *skb;
 	struct mld2_report *pmr;
@@ -1448,7 +1448,7 @@ static void mld_sendpack(struct sk_buff *skb)
 			      (struct mld2_report *)skb_transport_header(skb);
 	int payload_len, mldlen;
 	struct inet6_dev *idev = in6_dev_get(skb->dev);
-	struct net *net = skb->dev->nd_net;
+	struct net *net = dev_net(skb->dev);
 	int err;
 	struct flowi fl;
 
@@ -1762,7 +1762,7 @@ static void mld_send_cr(struct inet6_dev *idev)
 
 static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 {
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct sock *sk = net->ipv6.igmp_sk;
 	struct inet6_dev *idev;
 	struct sk_buff *skb;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 3f68a6eae7b2..79af57f586e8 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -447,7 +447,7 @@ static void __ndisc_send(struct net_device *dev,
 {
 	struct flowi fl;
 	struct dst_entry *dst;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct sock *sk = net->ipv6.ndisc_sk;
 	struct sk_buff *skb;
 	struct icmp6hdr *hdr;
@@ -539,7 +539,7 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
 	};
 
 	/* for anycast or proxy, solicited_addr != src_addr */
-	ifp = ipv6_get_ifaddr(dev->nd_net, solicited_addr, dev, 1);
+	ifp = ipv6_get_ifaddr(dev_net(dev), solicited_addr, dev, 1);
 	if (ifp) {
 		src_addr = solicited_addr;
 		if (ifp->flags & IFA_F_OPTIMISTIC)
@@ -547,7 +547,7 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
 		in6_ifa_put(ifp);
 	} else {
 		if (ipv6_dev_get_saddr(dev, daddr,
-				       inet6_sk(dev->nd_net->ipv6.ndisc_sk)->srcprefs,
+				       inet6_sk(dev_net(dev)->ipv6.ndisc_sk)->srcprefs,
 				       &tmpaddr))
 			return;
 		src_addr = &tmpaddr;
@@ -601,7 +601,7 @@ void ndisc_send_rs(struct net_device *dev, struct in6_addr *saddr,
 	 * suppress the inclusion of the sllao.
 	 */
 	if (send_sllao) {
-		struct inet6_ifaddr *ifp = ipv6_get_ifaddr(dev->nd_net, saddr,
+		struct inet6_ifaddr *ifp = ipv6_get_ifaddr(dev_net(dev), saddr,
 							   dev, 1);
 		if (ifp) {
 			if (ifp->flags & IFA_F_OPTIMISTIC)  {
@@ -639,7 +639,7 @@ static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb)
 	struct in6_addr *target = (struct in6_addr *)&neigh->primary_key;
 	int probes = atomic_read(&neigh->probes);
 
-	if (skb && ipv6_chk_addr(dev->nd_net, &ipv6_hdr(skb)->saddr, dev, 1))
+	if (skb && ipv6_chk_addr(dev_net(dev), &ipv6_hdr(skb)->saddr, dev, 1))
 		saddr = &ipv6_hdr(skb)->saddr;
 
 	if ((probes -= neigh->parms->ucast_probes) < 0) {
@@ -727,7 +727,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 
 	inc = ipv6_addr_is_multicast(daddr);
 
-	ifp = ipv6_get_ifaddr(dev->nd_net, &msg->target, dev, 1);
+	ifp = ipv6_get_ifaddr(dev_net(dev), &msg->target, dev, 1);
 	if (ifp) {
 
 		if (ifp->flags & (IFA_F_TENTATIVE|IFA_F_OPTIMISTIC)) {
@@ -776,7 +776,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 		if (ipv6_chk_acast_addr(dev, &msg->target) ||
 		    (idev->cnf.forwarding &&
 		     (ipv6_devconf.proxy_ndp || idev->cnf.proxy_ndp) &&
-		     (pneigh = pneigh_lookup(&nd_tbl, dev->nd_net,
+		     (pneigh = pneigh_lookup(&nd_tbl, dev_net(dev),
 					     &msg->target, dev, 0)) != NULL)) {
 			if (!(NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED) &&
 			    skb->pkt_type != PACKET_HOST &&
@@ -886,7 +886,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
 			return;
 		}
 	}
-	ifp = ipv6_get_ifaddr(dev->nd_net, &msg->target, dev, 1);
+	ifp = ipv6_get_ifaddr(dev_net(dev), &msg->target, dev, 1);
 	if (ifp) {
 		if (ifp->flags & IFA_F_TENTATIVE) {
 			addrconf_dad_failure(ifp);
@@ -918,7 +918,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
 		 */
 		if (lladdr && !memcmp(lladdr, dev->dev_addr, dev->addr_len) &&
 		    ipv6_devconf.forwarding && ipv6_devconf.proxy_ndp &&
-		    pneigh_lookup(&nd_tbl, dev->nd_net, &msg->target, dev, 0)) {
+		    pneigh_lookup(&nd_tbl, dev_net(dev), &msg->target, dev, 0)) {
 			/* XXX: idev->cnf.prixy_ndp */
 			goto out;
 		}
@@ -1008,7 +1008,7 @@ static void ndisc_ra_useropt(struct sk_buff *ra, struct nd_opt_hdr *opt)
 	struct sk_buff *skb;
 	struct nlmsghdr *nlh;
 	struct nduseroptmsg *ndmsg;
-	struct net *net = ra->dev->nd_net;
+	struct net *net = dev_net(ra->dev);
 	int err;
 	int base_size = NLMSG_ALIGN(sizeof(struct nduseroptmsg)
 				    + (opt->nd_opt_len << 3));
@@ -1395,7 +1395,7 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
 			 struct in6_addr *target)
 {
 	struct net_device *dev = skb->dev;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct sock *sk = net->ipv6.ndisc_sk;
 	int len = sizeof(struct icmp6hdr) + 2 * sizeof(struct in6_addr);
 	struct sk_buff *buff;
@@ -1597,7 +1597,7 @@ int ndisc_rcv(struct sk_buff *skb)
 static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, void *ptr)
 {
 	struct net_device *dev = ptr;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 
 	switch (event) {
 	case NETDEV_CHANGEADDR:
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index cc2f9afcf808..a6d30626b47c 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -484,7 +484,7 @@ ipq_rcv_dev_event(struct notifier_block *this,
 {
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* Drop any packets associated with the downed device */
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 8a5be290c710..364dc332532c 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -214,7 +214,7 @@ int snmp6_register_dev(struct inet6_dev *idev)
 	if (!idev || !idev->dev)
 		return -EINVAL;
 
-	if (idev->dev->nd_net != &init_net)
+	if (dev_net(idev->dev) != &init_net)
 		return 0;
 
 	if (!proc_net_devsnmp6)
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 548d0763f4d3..efb0047f6880 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -176,7 +176,7 @@ static int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
 	if (sk == NULL)
 		goto out;
 
-	net = skb->dev->nd_net;
+	net = dev_net(skb->dev);
 	sk = __raw_v6_lookup(net, sk, nexthdr, daddr, saddr, IP6CB(skb)->iif);
 
 	while (sk) {
@@ -363,7 +363,7 @@ void raw6_icmp_error(struct sk_buff *skb, int nexthdr,
 	if (sk != NULL) {
 		saddr = &ipv6_hdr(skb)->saddr;
 		daddr = &ipv6_hdr(skb)->daddr;
-		net = skb->dev->nd_net;
+		net = dev_net(skb->dev);
 
 		while ((sk = __raw_v6_lookup(net, sk, nexthdr, saddr, daddr,
 						IP6CB(skb)->iif))) {
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index f936d045a39d..4e1447634f36 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -600,7 +600,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb)
 		return 1;
 	}
 
-	net = skb->dev->nd_net;
+	net = dev_net(skb->dev);
 	if (atomic_read(&net->ipv6.frags.mem) > net->ipv6.frags.high_thresh)
 		ip6_evictor(net, ip6_dst_idev(skb->dst));
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 06faa46920e1..65053fba8c1a 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -208,7 +208,7 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 	struct rt6_info *rt = (struct rt6_info *)dst;
 	struct inet6_dev *idev = rt->rt6i_idev;
 	struct net_device *loopback_dev =
-		dev->nd_net->loopback_dev;
+		dev_net(dev)->loopback_dev;
 
 	if (dev != loopback_dev && idev != NULL && idev->dev == dev) {
 		struct inet6_dev *loopback_idev =
@@ -433,7 +433,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 	RT6_TRACE("%s() => %p\n",
 		  __func__, match);
 
-	net = rt0->rt6i_dev->nd_net;
+	net = dev_net(rt0->rt6i_dev);
 	return (match ? match : net->ipv6.ip6_null_entry);
 }
 
@@ -441,7 +441,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 		  struct in6_addr *gwaddr)
 {
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct route_info *rinfo = (struct route_info *) opt;
 	struct in6_addr prefix_buf, *prefix;
 	unsigned int pref;
@@ -607,7 +607,7 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info)
 int ip6_ins_rt(struct rt6_info *rt)
 {
 	struct nl_info info = {
-		.nl_net = rt->rt6i_dev->nd_net,
+		.nl_net = dev_net(rt->rt6i_dev),
 	};
 	return __ip6_ins_rt(rt, &info);
 }
@@ -745,7 +745,7 @@ static struct rt6_info *ip6_pol_route_input(struct net *net, struct fib6_table *
 void ip6_route_input(struct sk_buff *skb)
 {
 	struct ipv6hdr *iph = ipv6_hdr(skb);
-	struct net *net = skb->dev->nd_net;
+	struct net *net = dev_net(skb->dev);
 	int flags = RT6_LOOKUP_F_HAS_SADDR;
 	struct flowi fl = {
 		.iif = skb->dev->ifindex,
@@ -928,7 +928,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 {
 	struct rt6_info *rt;
 	struct inet6_dev *idev = in6_dev_get(dev);
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 
 	if (unlikely(idev == NULL))
 		return NULL;
@@ -1252,7 +1252,7 @@ install_route:
 	rt->rt6i_idev = idev;
 	rt->rt6i_table = table;
 
-	cfg->fc_nlinfo.nl_net = dev->nd_net;
+	cfg->fc_nlinfo.nl_net = dev_net(dev);
 
 	return __ip6_ins_rt(rt, &cfg->fc_nlinfo);
 
@@ -1270,7 +1270,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 {
 	int err;
 	struct fib6_table *table;
-	struct net *net = rt->rt6i_dev->nd_net;
+	struct net *net = dev_net(rt->rt6i_dev);
 
 	if (rt == net->ipv6.ip6_null_entry)
 		return -ENOENT;
@@ -1289,7 +1289,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 int ip6_del_rt(struct rt6_info *rt)
 {
 	struct nl_info info = {
-		.nl_net = rt->rt6i_dev->nd_net,
+		.nl_net = dev_net(rt->rt6i_dev),
 	};
 	return __ip6_del_rt(rt, &info);
 }
@@ -1401,7 +1401,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest,
 					   struct net_device *dev)
 {
 	int flags = RT6_LOOKUP_F_HAS_SADDR;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	struct ip6rd_flowi rdfl = {
 		.fl = {
 			.oif = dev->ifindex,
@@ -1428,7 +1428,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 {
 	struct rt6_info *rt, *nrt = NULL;
 	struct netevent_redirect netevent;
-	struct net *net = neigh->dev->nd_net;
+	struct net *net = dev_net(neigh->dev);
 
 	rt = ip6_route_redirect(dest, src, saddr, neigh->dev);
 
@@ -1477,7 +1477,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 	nrt->rt6i_nexthop = neigh_clone(neigh);
 	/* Reset pmtu, it may be better */
 	nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
-	nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(neigh->dev->nd_net,
+	nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dev_net(neigh->dev),
 							dst_mtu(&nrt->u.dst));
 
 	if (ip6_ins_rt(nrt))
@@ -1506,7 +1506,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 			struct net_device *dev, u32 pmtu)
 {
 	struct rt6_info *rt, *nrt;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 	int allfrag = 0;
 
 	rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0);
@@ -1583,7 +1583,7 @@ out:
 
 static struct rt6_info * ip6_rt_copy(struct rt6_info *ort)
 {
-	struct net *net = ort->rt6i_dev->nd_net;
+	struct net *net = dev_net(ort->rt6i_dev);
 	struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
 
 	if (rt) {
@@ -1682,7 +1682,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
 	struct rt6_info *rt;
 	struct fib6_table *table;
 
-	table = fib6_get_table(dev->nd_net, RT6_TABLE_DFLT);
+	table = fib6_get_table(dev_net(dev), RT6_TABLE_DFLT);
 	if (table == NULL)
 		return NULL;
 
@@ -1713,7 +1713,7 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 				  RTF_UP | RTF_EXPIRES | RTF_PREF(pref),
 		.fc_nlinfo.pid = 0,
 		.fc_nlinfo.nlh = NULL,
-		.fc_nlinfo.nl_net = dev->nd_net,
+		.fc_nlinfo.nl_net = dev_net(dev),
 	};
 
 	ipv6_addr_copy(&cfg.fc_gateway, gwaddr);
@@ -1862,7 +1862,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 				    const struct in6_addr *addr,
 				    int anycast)
 {
-	struct net *net = idev->dev->nd_net;
+	struct net *net = dev_net(idev->dev);
 	struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
 
 	if (rt == NULL)
@@ -1939,7 +1939,7 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg)
 {
 	struct rt6_mtu_change_arg *arg = (struct rt6_mtu_change_arg *) p_arg;
 	struct inet6_dev *idev;
-	struct net *net = arg->dev->nd_net;
+	struct net *net = dev_net(arg->dev);
 
 	/* In IPv6 pmtu discovery is not optional,
 	   so that RTAX_MTU lock cannot disable it.
@@ -1983,7 +1983,7 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu)
 		.mtu = mtu,
 	};
 
-	fib6_clean_all(dev->nd_net, rt6_mtu_change_route, 0, &arg);
+	fib6_clean_all(dev_net(dev), rt6_mtu_change_route, 0, &arg);
 }
 
 static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = {
@@ -2321,7 +2321,7 @@ static int ip6_route_dev_notify(struct notifier_block *this,
 				unsigned long event, void *data)
 {
 	struct net_device *dev = (struct net_device *)data;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 
 	if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) {
 		net->ipv6.ip6_null_entry->u.dst.dev = dev;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 8dd72966ff78..086deffff9c9 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -321,7 +321,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	struct tcp_sock *tp;
 	__u32 seq;
 
-	sk = inet6_lookup(skb->dev->nd_net, &tcp_hashinfo, &hdr->daddr,
+	sk = inet6_lookup(dev_net(skb->dev), &tcp_hashinfo, &hdr->daddr,
 			th->dest, &hdr->saddr, th->source, skb->dev->ifindex);
 
 	if (sk == NULL) {
@@ -988,7 +988,7 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 	struct tcphdr *th = tcp_hdr(skb), *t1;
 	struct sk_buff *buff;
 	struct flowi fl;
-	struct net *net = skb->dst->dev->nd_net;
+	struct net *net = dev_net(skb->dst->dev);
 	struct sock *ctl_sk = net->ipv6.tcp_sk;
 	unsigned int tot_len = sizeof(*th);
 #ifdef CONFIG_TCP_MD5SIG
@@ -1093,7 +1093,7 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw,
 	struct tcphdr *th = tcp_hdr(skb), *t1;
 	struct sk_buff *buff;
 	struct flowi fl;
-	struct net *net = skb->dev->nd_net;
+	struct net *net = dev_net(skb->dev);
 	struct sock *ctl_sk = net->ipv6.tcp_sk;
 	unsigned int tot_len = sizeof(struct tcphdr);
 	__be32 *topt;
@@ -1739,7 +1739,7 @@ static int tcp_v6_rcv(struct sk_buff *skb)
 	TCP_SKB_CB(skb)->flags = ipv6_get_dsfield(ipv6_hdr(skb));
 	TCP_SKB_CB(skb)->sacked = 0;
 
-	sk = __inet6_lookup(skb->dev->nd_net, &tcp_hashinfo,
+	sk = __inet6_lookup(dev_net(skb->dev), &tcp_hashinfo,
 			&ipv6_hdr(skb)->saddr, th->source,
 			&ipv6_hdr(skb)->daddr, ntohs(th->dest),
 			inet6_iif(skb));
@@ -1822,7 +1822,7 @@ do_time_wait:
 	{
 		struct sock *sk2;
 
-		sk2 = inet6_lookup_listener(skb->dev->nd_net, &tcp_hashinfo,
+		sk2 = inet6_lookup_listener(dev_net(skb->dev), &tcp_hashinfo,
 					    &ipv6_hdr(skb)->daddr,
 					    ntohs(th->dest), inet6_iif(skb));
 		if (sk2 != NULL) {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 593d3efadaf9..6683c04b427e 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -235,7 +235,7 @@ void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	struct sock *sk;
 	int err;
 
-	sk = __udp6_lib_lookup(skb->dev->nd_net, daddr, uh->dest,
+	sk = __udp6_lib_lookup(dev_net(skb->dev), daddr, uh->dest,
 			       saddr, uh->source, inet6_iif(skb), udptable);
 	if (sk == NULL)
 		return;
@@ -483,7 +483,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 	 * check socket cache ... must talk to Alan about his plans
 	 * for sock caches... i'll skip this for now.
 	 */
-	sk = __udp6_lib_lookup(skb->dev->nd_net, saddr, uh->source,
+	sk = __udp6_lib_lookup(dev_net(skb->dev), saddr, uh->source,
 			       daddr, uh->dest, inet6_iif(skb), udptable);
 
 	if (sk == NULL) {
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index d92d1fceb8cf..8f1e0543b3c4 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -247,7 +247,7 @@ static void xfrm6_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 	xdst = (struct xfrm_dst *)dst;
 	if (xdst->u.rt6.rt6i_idev->dev == dev) {
 		struct inet6_dev *loopback_idev =
-			in6_dev_get(dev->nd_net->loopback_dev);
+			in6_dev_get(dev_net(dev)->loopback_dev);
 		BUG_ON(!loopback_idev);
 
 		do {
diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c
index c76a9523091b..81ae8735f5e3 100644
--- a/net/ipx/af_ipx.c
+++ b/net/ipx/af_ipx.c
@@ -335,7 +335,7 @@ static int ipxitf_device_event(struct notifier_block *notifier,
 	struct net_device *dev = ptr;
 	struct ipx_interface *i, *tmp;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event != NETDEV_DOWN && event != NETDEV_UP)
@@ -1636,7 +1636,7 @@ static int ipx_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty
 	u16 ipx_pktsize;
 	int rc = 0;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	/* Not ours */
diff --git a/net/irda/irlap_frame.c b/net/irda/irlap_frame.c
index a38b231c8689..90894534f3cc 100644
--- a/net/irda/irlap_frame.c
+++ b/net/irda/irlap_frame.c
@@ -1326,7 +1326,7 @@ int irlap_driver_rcv(struct sk_buff *skb, struct net_device *dev,
 	int command;
 	__u8 control;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto out;
 
 	/* FIXME: should we get our own field? */
diff --git a/net/llc/llc_input.c b/net/llc/llc_input.c
index b9143d2a04e1..a69c5c427fe3 100644
--- a/net/llc/llc_input.c
+++ b/net/llc/llc_input.c
@@ -146,7 +146,7 @@ int llc_rcv(struct sk_buff *skb, struct net_device *dev,
 	int (*rcv)(struct sk_buff *, struct net_device *,
 		   struct packet_type *, struct net_device *);
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	/*
diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index ec05684c56d7..292fa28146fb 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -168,7 +168,7 @@ int nf_hook_slow(int pf, unsigned int hook, struct sk_buff *skb,
 #ifdef CONFIG_NET_NS
 	struct net *net;
 
-	net = indev == NULL ? outdev->nd_net : indev->nd_net;
+	net = indev == NULL ? dev_net(outdev) : dev_net(indev);
 	if (net != &init_net)
 		return 1;
 #endif
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index 012cb6910820..81fb048add88 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -557,7 +557,7 @@ nfqnl_rcv_dev_event(struct notifier_block *this,
 {
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* Drop any packets associated with the downed device */
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index 4478f2f6079d..a547c6320eb3 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -954,7 +954,7 @@ static int netlbl_unlhsh_netdev_handler(struct notifier_block *this,
 	struct net_device *dev = ptr;
 	struct netlbl_unlhsh_iface *iface = NULL;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	/* XXX - should this be a check for NETDEV_DOWN or _UNREGISTER? */
diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 972250c974f1..a270ebf9f765 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -106,7 +106,7 @@ static int nr_device_event(struct notifier_block *this, unsigned long event, voi
 {
 	struct net_device *dev = (struct net_device *)ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event != NETDEV_DOWN)
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index a56ed2120e07..baa290d3444a 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -263,7 +263,7 @@ static int packet_rcv_spkt(struct sk_buff *skb, struct net_device *dev,  struct
 	if (skb->pkt_type == PACKET_LOOPBACK)
 		goto out;
 
-	if (dev->nd_net != sk->sk_net)
+	if (dev_net(dev) != sk->sk_net)
 		goto out;
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
@@ -451,7 +451,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet
 	sk = pt->af_packet_priv;
 	po = pkt_sk(sk);
 
-	if (dev->nd_net != sk->sk_net)
+	if (dev_net(dev) != sk->sk_net)
 		goto drop;
 
 	skb->dev = dev;
@@ -568,7 +568,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	sk = pt->af_packet_priv;
 	po = pkt_sk(sk);
 
-	if (dev->nd_net != sk->sk_net)
+	if (dev_net(dev) != sk->sk_net)
 		goto drop;
 
 	if (dev->header_ops) {
@@ -1450,7 +1450,7 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void
 	struct sock *sk;
 	struct hlist_node *node;
 	struct net_device *dev = data;
-	struct net *net = dev->nd_net;
+	struct net *net = dev_net(dev);
 
 	read_lock(&net->packet.sklist_lock);
 	sk_for_each(sk, node, &net->packet.sklist) {
diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index 4a31a81059ab..1a7f143cf741 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c
@@ -197,7 +197,7 @@ static int rose_device_event(struct notifier_block *this, unsigned long event,
 {
 	struct net_device *dev = (struct net_device *)ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event != NETDEV_DOWN)
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index beea2fb18b15..2faa0d8839eb 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -630,7 +630,7 @@ static int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev,
 	struct sctp_sockaddr_entry *temp;
 	int found = 0;
 
-	if (ifa->ifa_dev->dev->nd_net != &init_net)
+	if (dev_net(ifa->ifa_dev->dev) != &init_net)
 		return NOTIFY_DONE;
 
 	switch (ev) {
diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c
index 3bbef2ab22ae..9cd35eec3e7f 100644
--- a/net/tipc/eth_media.c
+++ b/net/tipc/eth_media.c
@@ -101,7 +101,7 @@ static int recv_msg(struct sk_buff *buf, struct net_device *dev,
 	struct eth_bearer *eb_ptr = (struct eth_bearer *)pt->af_packet_priv;
 	u32 size;
 
-	if (dev->nd_net != &init_net) {
+	if (dev_net(dev) != &init_net) {
 		kfree_skb(buf);
 		return 0;
 	}
@@ -198,7 +198,7 @@ static int recv_notification(struct notifier_block *nb, unsigned long evt,
 	struct eth_bearer *eb_ptr = &eth_bearers[0];
 	struct eth_bearer *stop = &eth_bearers[MAX_ETH_BEARERS];
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	while ((eb_ptr->dev != dev)) {
diff --git a/net/wireless/wext.c b/net/wireless/wext.c
index 2c569b63e7d8..947188a5b937 100644
--- a/net/wireless/wext.c
+++ b/net/wireless/wext.c
@@ -1157,7 +1157,7 @@ static void rtmsg_iwinfo(struct net_device *dev, char *event, int event_len)
 	struct sk_buff *skb;
 	int err;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return;
 
 	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c
index 339ca4a8e89e..7a46ea73fe2d 100644
--- a/net/x25/af_x25.c
+++ b/net/x25/af_x25.c
@@ -191,7 +191,7 @@ static int x25_device_event(struct notifier_block *this, unsigned long event,
 	struct net_device *dev = ptr;
 	struct x25_neigh *nb;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (dev->type == ARPHRD_X25
diff --git a/net/x25/x25_dev.c b/net/x25/x25_dev.c
index f0679d283110..3ff206c0ae94 100644
--- a/net/x25/x25_dev.c
+++ b/net/x25/x25_dev.c
@@ -95,7 +95,7 @@ int x25_lapb_receive_frame(struct sk_buff *skb, struct net_device *dev,
 	struct sk_buff *nskb;
 	struct x25_neigh *nb;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		goto drop;
 
 	nskb = skb_copy(skb, GFP_ATOMIC);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 8e588f20c60c..15d73e47cc2c 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -2079,7 +2079,7 @@ static int stale_bundle(struct dst_entry *dst)
 void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev)
 {
 	while ((dst = dst->child) && dst->xfrm && dst->dev == dev) {
-		dst->dev = dev->nd_net->loopback_dev;
+		dst->dev = dev_net(dev)->loopback_dev;
 		dev_hold(dst->dev);
 		dev_put(dev);
 	}
@@ -2350,7 +2350,7 @@ static int xfrm_dev_event(struct notifier_block *this, unsigned long event, void
 {
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	switch (event) {
diff --git a/security/selinux/netif.c b/security/selinux/netif.c
index 013d3117a86b..9c8a82aa8baf 100644
--- a/security/selinux/netif.c
+++ b/security/selinux/netif.c
@@ -281,7 +281,7 @@ static int sel_netif_netdev_notifier_handler(struct notifier_block *this,
 {
 	struct net_device *dev = ptr;
 
-	if (dev->nd_net != &init_net)
+	if (dev_net(dev) != &init_net)
 		return NOTIFY_DONE;
 
 	if (event == NETDEV_DOWN)
-- 
cgit v1.2.3


From 3b1e0a655f8eba44ab1ee2a1068d169ccfb853b9 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 26 Mar 2008 02:26:21 +0900
Subject: [NET] NETNS: Omit sock->sk_net without CONFIG_NET_NS.

Introduce per-sock inlines: sock_net(), sock_net_set()
and per-inet_timewait_sock inlines: twsk_net(), twsk_net_set().
Without CONFIG_NET_NS, no namespace other than &init_net exists.
Let's explicitly define them to help compiler optimizations.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/ipv6.h             |  4 ++--
 include/net/inet_hashtables.h    |  8 ++++----
 include/net/inet_timewait_sock.h | 18 ++++++++++++++++++
 include/net/route.h              |  4 ++--
 include/net/sock.h               | 24 ++++++++++++++++++++++--
 net/atm/svc.c                    |  2 +-
 net/ax25/af_ax25.c               |  2 +-
 net/bluetooth/l2cap.c            |  2 +-
 net/bluetooth/rfcomm/sock.c      |  2 +-
 net/bluetooth/sco.c              |  2 +-
 net/bridge/br_netlink.c          |  4 ++--
 net/core/fib_rules.c             |  6 +++---
 net/core/neighbour.c             | 10 +++++-----
 net/core/rtnetlink.c             | 12 ++++++------
 net/core/sock.c                  | 10 +++++-----
 net/decnet/af_decnet.c           |  2 +-
 net/decnet/dn_dev.c              |  6 +++---
 net/decnet/dn_fib.c              |  4 ++--
 net/decnet/dn_route.c            |  4 ++--
 net/decnet/dn_table.c            |  2 +-
 net/ipv4/af_inet.c               |  6 +++---
 net/ipv4/devinet.c               |  6 +++---
 net/ipv4/fib_frontend.c          |  8 ++++----
 net/ipv4/fib_rules.c             |  2 +-
 net/ipv4/igmp.c                  | 14 +++++++-------
 net/ipv4/inet_connection_sock.c  |  4 ++--
 net/ipv4/inet_hashtables.c       |  8 ++++----
 net/ipv4/inet_timewait_sock.c    |  2 +-
 net/ipv4/ip_input.c              |  2 +-
 net/ipv4/ip_output.c             |  4 ++--
 net/ipv4/ip_sockglue.c           |  6 +++---
 net/ipv4/ipmr.c                  |  4 ++--
 net/ipv4/netfilter/arp_tables.c  | 16 ++++++++--------
 net/ipv4/netfilter/ip_tables.c   | 16 ++++++++--------
 net/ipv4/raw.c                   | 12 ++++++------
 net/ipv4/route.c                 |  4 ++--
 net/ipv4/tcp_ipv4.c              | 14 +++++++-------
 net/ipv4/udp.c                   | 16 ++++++++--------
 net/ipv6/addrconf.c              | 10 +++++-----
 net/ipv6/addrlabel.c             |  6 +++---
 net/ipv6/af_inet6.c              |  4 ++--
 net/ipv6/fib6_rules.c            |  2 +-
 net/ipv6/icmp.c                  |  2 +-
 net/ipv6/inet6_hashtables.c      |  4 ++--
 net/ipv6/ip6_fib.c               |  2 +-
 net/ipv6/ip6_output.c            |  2 +-
 net/ipv6/ipv6_sockglue.c         |  2 +-
 net/ipv6/mcast.c                 | 12 ++++++------
 net/ipv6/netfilter/ip6_tables.c  | 16 ++++++++--------
 net/ipv6/raw.c                   |  6 +++---
 net/ipv6/route.c                 |  4 ++--
 net/ipv6/tcp_ipv6.c              |  2 +-
 net/ipv6/udp.c                   |  4 ++--
 net/irda/af_irda.c               |  2 +-
 net/llc/llc_conn.c               |  2 +-
 net/netfilter/nf_sockopt.c       |  2 +-
 net/netlink/af_netlink.c         | 30 +++++++++++++++---------------
 net/netrom/af_netrom.c           |  2 +-
 net/packet/af_packet.c           | 28 ++++++++++++++--------------
 net/rose/af_rose.c               |  2 +-
 net/sched/act_api.c              |  4 ++--
 net/sched/cls_api.c              |  4 ++--
 net/sched/sch_api.c              | 10 +++++-----
 net/sctp/ipv6.c                  |  2 +-
 net/sctp/protocol.c              |  2 +-
 net/socket.c                     |  4 ++--
 net/tipc/socket.c                |  2 +-
 net/unix/af_unix.c               | 20 ++++++++++----------
 net/x25/af_x25.c                 |  2 +-
 69 files changed, 253 insertions(+), 215 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index c9ba0da16ce9..b90d3d461d4e 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -481,7 +481,7 @@ static inline struct raw6_sock *raw6_sk(const struct sock *sk)
 #endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */
 
 #define INET6_MATCH(__sk, __net, __hash, __saddr, __daddr, __ports, __dif)\
-	(((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net))	&& \
+	(((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net)	&& \
 	 ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports))  	&& \
 	 ((__sk)->sk_family		== AF_INET6)		&& \
 	 ipv6_addr_equal(&inet6_sk(__sk)->daddr, (__saddr))	&& \
@@ -489,7 +489,7 @@ static inline struct raw6_sock *raw6_sk(const struct sock *sk)
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 
 #define INET6_TW_MATCH(__sk, __net, __hash, __saddr, __daddr, __ports, __dif) \
-	(((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net))	&& \
+	(((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net)	&& \
 	 (*((__portpair *)&(inet_twsk(__sk)->tw_dport)) == (__ports))	&& \
 	 ((__sk)->sk_family	       == PF_INET6)			&& \
 	 (ipv6_addr_equal(&inet6_twsk(__sk)->tw_v6_daddr, (__saddr)))	&& \
diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index d99c1ba2ece0..5525227c5e92 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -314,25 +314,25 @@ typedef __u64 __bitwise __addrpair;
 				   ((__force __u64)(__be32)(__saddr)));
 #endif /* __BIG_ENDIAN */
 #define INET_MATCH(__sk, __net, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
-	(((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net))	&&	\
+	(((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net)	&&	\
 	 ((*((__addrpair *)&(inet_sk(__sk)->daddr))) == (__cookie))	&&	\
 	 ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports))	&&	\
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 #define INET_TW_MATCH(__sk, __net, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
-	(((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net))	&&	\
+	(((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net)	&&	\
 	 ((*((__addrpair *)&(inet_twsk(__sk)->tw_daddr))) == (__cookie)) &&	\
 	 ((*((__portpair *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) &&	\
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 #else /* 32-bit arch */
 #define INET_ADDR_COOKIE(__name, __saddr, __daddr)
 #define INET_MATCH(__sk, __net, __hash, __cookie, __saddr, __daddr, __ports, __dif)	\
-	(((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net))	&&	\
+	(((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net)	&&	\
 	 (inet_sk(__sk)->daddr		== (__saddr))		&&	\
 	 (inet_sk(__sk)->rcv_saddr	== (__daddr))		&&	\
 	 ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports))	&&	\
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 #define INET_TW_MATCH(__sk, __net, __hash,__cookie, __saddr, __daddr, __ports, __dif)	\
-	(((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net))	&&	\
+	(((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net)	&&	\
 	 (inet_twsk(__sk)->tw_daddr	== (__saddr))		&&	\
 	 (inet_twsk(__sk)->tw_rcv_saddr	== (__daddr))		&&	\
 	 ((*((__portpair *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) &&	\
diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 296547bfb0b7..07fe0d1a4f03 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -207,4 +207,22 @@ extern void inet_twsk_schedule(struct inet_timewait_sock *tw,
 			       const int timeo, const int timewait_len);
 extern void inet_twsk_deschedule(struct inet_timewait_sock *tw,
 				 struct inet_timewait_death_row *twdr);
+
+static inline
+struct net *twsk_net(const struct inet_timewait_sock *twsk)
+{
+#ifdef CONFIG_NET_NS
+	return twsk->tw_net;
+#else
+	return &init_net;
+#endif
+}
+
+static inline
+void twsk_net_set(struct inet_timewait_sock *twsk, const struct net *net)
+{
+#ifdef CONFIG_NET_NS
+	twsk->tw_net = net;
+#endif
+}
 #endif	/* _INET_TIMEWAIT_SOCK_ */
diff --git a/include/net/route.h b/include/net/route.h
index 28dba925663c..c6338802e8f1 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -160,7 +160,7 @@ static inline int ip_route_connect(struct rtable **rp, __be32 dst,
 					 .dport = dport } } };
 
 	int err;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	if (!dst || !src) {
 		err = __ip_route_output_key(net, rp, &fl);
 		if (err)
@@ -188,7 +188,7 @@ static inline int ip_route_newports(struct rtable **rp, u8 protocol,
 		ip_rt_put(*rp);
 		*rp = NULL;
 		security_sk_classify_flow(sk, &fl);
-		return ip_route_output_flow(sk->sk_net, rp, &fl, sk, 0);
+		return ip_route_output_flow(sock_net(sk), rp, &fl, sk, 0);
 	}
 	return 0;
 }
diff --git a/include/net/sock.h b/include/net/sock.h
index b433b1ed203d..7e0d4a0c4d12 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -126,7 +126,9 @@ struct sock_common {
 	atomic_t		skc_refcnt;
 	unsigned int		skc_hash;
 	struct proto		*skc_prot;
+#ifdef CONFIG_NET_NS
 	struct net	 	*skc_net;
+#endif
 };
 
 /**
@@ -1345,6 +1347,24 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_e
 }
 #endif
 
+static inline
+struct net *sock_net(const struct sock *sk)
+{
+#ifdef CONFIG_NET_NS
+	return sk->sk_net;
+#else
+	return &init_net;
+#endif
+}
+
+static inline
+void sock_net_set(struct sock *sk, const struct net *net)
+{
+#ifdef CONFIG_NET_NS
+	sk->sk_net = net;
+#endif
+}
+
 /*
  * Kernel sockets, f.e. rtnl or icmp_socket, are a part of a namespace.
  * They should not hold a referrence to a namespace in order to allow
@@ -1353,8 +1373,8 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_e
  */
 static inline void sk_change_net(struct sock *sk, struct net *net)
 {
-	put_net(sk->sk_net);
-	sk->sk_net = net;
+	put_net(sock_net(sk));
+	sock_net_set(sk, net);
 }
 
 extern void sock_enable_timestamp(struct sock *sk);
diff --git a/net/atm/svc.c b/net/atm/svc.c
index daf9a48a7db0..de1e4f2f3a43 100644
--- a/net/atm/svc.c
+++ b/net/atm/svc.c
@@ -326,7 +326,7 @@ static int svc_accept(struct socket *sock,struct socket *newsock,int flags)
 
 	lock_sock(sk);
 
-	error = svc_create(sk->sk_net, newsock,0);
+	error = svc_create(sock_net(sk), newsock,0);
 	if (error)
 		goto out;
 
diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c
index ee9dd83e7561..2712544cf0ca 100644
--- a/net/ax25/af_ax25.c
+++ b/net/ax25/af_ax25.c
@@ -869,7 +869,7 @@ struct sock *ax25_make_new(struct sock *osk, struct ax25_dev *ax25_dev)
 	struct sock *sk;
 	ax25_cb *ax25, *oax25;
 
-	sk = sk_alloc(osk->sk_net, PF_AX25, GFP_ATOMIC,	osk->sk_prot);
+	sk = sk_alloc(sock_net(osk), PF_AX25, GFP_ATOMIC,	osk->sk_prot);
 	if (sk == NULL)
 		return NULL;
 
diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c
index 34f8bf98bc05..6b995ac832f5 100644
--- a/net/bluetooth/l2cap.c
+++ b/net/bluetooth/l2cap.c
@@ -1499,7 +1499,7 @@ static inline int l2cap_connect_req(struct l2cap_conn *conn, struct l2cap_cmd_hd
 		goto response;
 	}
 
-	sk = l2cap_sock_alloc(parent->sk_net, NULL, BTPROTO_L2CAP, GFP_ATOMIC);
+	sk = l2cap_sock_alloc(sock_net(parent), NULL, BTPROTO_L2CAP, GFP_ATOMIC);
 	if (!sk)
 		goto response;
 
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index c46d51035e77..c103fa02893b 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -868,7 +868,7 @@ int rfcomm_connect_ind(struct rfcomm_session *s, u8 channel, struct rfcomm_dlc *
 		goto done;
 	}
 
-	sk = rfcomm_sock_alloc(parent->sk_net, NULL, BTPROTO_RFCOMM, GFP_ATOMIC);
+	sk = rfcomm_sock_alloc(sock_net(parent), NULL, BTPROTO_RFCOMM, GFP_ATOMIC);
 	if (!sk)
 		goto done;
 
diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c
index b91d3c81a73c..2a5953b4405d 100644
--- a/net/bluetooth/sco.c
+++ b/net/bluetooth/sco.c
@@ -803,7 +803,7 @@ static void sco_conn_ready(struct sco_conn *conn)
 
 		bh_lock_sock(parent);
 
-		sk = sco_sock_alloc(parent->sk_net, NULL, BTPROTO_SCO, GFP_ATOMIC);
+		sk = sco_sock_alloc(sock_net(parent), NULL, BTPROTO_SCO, GFP_ATOMIC);
 		if (!sk) {
 			bh_unlock_sock(parent);
 			goto done;
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index f5d69336d97b..f155e6ce8a21 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -108,7 +108,7 @@ errout:
  */
 static int br_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct net_device *dev;
 	int idx;
 
@@ -140,7 +140,7 @@ skip:
  */
 static int br_rtm_setlink(struct sk_buff *skb,  struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ifinfomsg *ifm;
 	struct nlattr *protinfo;
 	struct net_device *dev;
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 942be93a2eb0..540c07283e31 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -214,7 +214,7 @@ errout:
 
 static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct fib_rule_hdr *frh = nlmsg_data(nlh);
 	struct fib_rules_ops *ops = NULL;
 	struct fib_rule *rule, *r, *last = NULL;
@@ -352,7 +352,7 @@ errout:
 
 static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct fib_rule_hdr *frh = nlmsg_data(nlh);
 	struct fib_rules_ops *ops = NULL;
 	struct fib_rule *rule, *tmp;
@@ -534,7 +534,7 @@ skip:
 
 static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct fib_rules_ops *ops;
 	int idx = 0, family;
 
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index c978bd1cd659..065fbac7ecd3 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -1478,7 +1478,7 @@ int neigh_table_clear(struct neigh_table *tbl)
 
 static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ndmsg *ndm;
 	struct nlattr *dst_attr;
 	struct neigh_table *tbl;
@@ -1544,7 +1544,7 @@ out:
 
 static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ndmsg *ndm;
 	struct nlattr *tb[NDA_MAX+1];
 	struct neigh_table *tbl;
@@ -1812,7 +1812,7 @@ static const struct nla_policy nl_ntbl_parm_policy[NDTPA_MAX+1] = {
 
 static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct neigh_table *tbl;
 	struct ndtmsg *ndtmsg;
 	struct nlattr *tb[NDTA_MAX+1];
@@ -1937,7 +1937,7 @@ errout:
 
 static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int family, tidx, nidx = 0;
 	int tbl_skip = cb->args[0];
 	int neigh_skip = cb->args[1];
@@ -2037,7 +2037,7 @@ static void neigh_update_notify(struct neighbour *neigh)
 static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb,
 			    struct netlink_callback *cb)
 {
-	struct net * net = skb->sk->sk_net;
+	struct net * net = sock_net(skb->sk);
 	struct neighbour *n;
 	int rc, h, s_h = cb->args[1];
 	int idx, s_idx = idx = cb->args[2];
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 09250a0800f6..da99ac0871bf 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -662,7 +662,7 @@ nla_put_failure:
 
 static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int idx;
 	int s_idx = cb->args[0];
 	struct net_device *dev;
@@ -879,7 +879,7 @@ errout:
 
 static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ifinfomsg *ifm;
 	struct net_device *dev;
 	int err;
@@ -921,7 +921,7 @@ errout:
 
 static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	const struct rtnl_link_ops *ops;
 	struct net_device *dev;
 	struct ifinfomsg *ifm;
@@ -1000,7 +1000,7 @@ err:
 
 static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	const struct rtnl_link_ops *ops;
 	struct net_device *dev;
 	struct ifinfomsg *ifm;
@@ -1132,7 +1132,7 @@ replay:
 
 static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ifinfomsg *ifm;
 	struct nlattr *tb[IFLA_MAX+1];
 	struct net_device *dev = NULL;
@@ -1227,7 +1227,7 @@ static int rtattr_max;
 
 static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	rtnl_doit_func doit;
 	int sz_idx, kind;
 	int min_len;
diff --git a/net/core/sock.c b/net/core/sock.c
index b1a6ed4d33c1..3ee95060dbd0 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -372,7 +372,7 @@ static int sock_bindtodevice(struct sock *sk, char __user *optval, int optlen)
 {
 	int ret = -ENOPROTOOPT;
 #ifdef CONFIG_NETDEVICES
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	char devname[IFNAMSIZ];
 	int index;
 
@@ -958,7 +958,7 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority,
 		 */
 		sk->sk_prot = sk->sk_prot_creator = prot;
 		sock_lock_init(sk);
-		sk->sk_net = get_net(net);
+		sock_net_set(sk, get_net(net));
 	}
 
 	return sk;
@@ -983,7 +983,7 @@ void sk_free(struct sock *sk)
 		printk(KERN_DEBUG "%s: optmem leakage (%d bytes) detected.\n",
 		       __func__, atomic_read(&sk->sk_omem_alloc));
 
-	put_net(sk->sk_net);
+	put_net(sock_net(sk));
 	sk_prot_free(sk->sk_prot_creator, sk);
 }
 
@@ -1001,7 +1001,7 @@ void sk_release_kernel(struct sock *sk)
 
 	sock_hold(sk);
 	sock_release(sk->sk_socket);
-	sk->sk_net = get_net(&init_net);
+	sock_net_set(sk, get_net(&init_net));
 	sock_put(sk);
 }
 EXPORT_SYMBOL(sk_release_kernel);
@@ -1017,7 +1017,7 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority)
 		sock_copy(newsk, sk);
 
 		/* SANITY */
-		get_net(newsk->sk_net);
+		get_net(sock_net(newsk));
 		sk_node_init(&newsk->sk_node);
 		sock_lock_init(newsk);
 		bh_lock_sock(newsk);
diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index 3554fb3d251c..fc2efe899e91 100644
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -1094,7 +1094,7 @@ static int dn_accept(struct socket *sock, struct socket *newsock, int flags)
 
 	cb = DN_SKB_CB(skb);
 	sk->sk_ack_backlog--;
-	newsk = dn_alloc_sock(sk->sk_net, newsock, sk->sk_allocation);
+	newsk = dn_alloc_sock(sock_net(sk), newsock, sk->sk_allocation);
 	if (newsk == NULL) {
 		release_sock(sk);
 		kfree_skb(skb);
diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c
index 1bbfce5f7a2d..2f0ac3c3eb71 100644
--- a/net/decnet/dn_dev.c
+++ b/net/decnet/dn_dev.c
@@ -625,7 +625,7 @@ static const struct nla_policy dn_ifa_policy[IFA_MAX+1] = {
 
 static int dn_nl_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct nlattr *tb[IFA_MAX+1];
 	struct dn_dev *dn_db;
 	struct ifaddrmsg *ifm;
@@ -663,7 +663,7 @@ errout:
 
 static int dn_nl_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct nlattr *tb[IFA_MAX+1];
 	struct net_device *dev;
 	struct dn_dev *dn_db;
@@ -779,7 +779,7 @@ errout:
 
 static int dn_nl_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int idx, dn_idx = 0, skip_ndevs, skip_naddr;
 	struct net_device *dev;
 	struct dn_dev *dn_db;
diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c
index 4aa9a423e606..27ea2e9b080a 100644
--- a/net/decnet/dn_fib.c
+++ b/net/decnet/dn_fib.c
@@ -504,7 +504,7 @@ static int dn_fib_check_attr(struct rtmsg *r, struct rtattr **rta)
 
 static int dn_fib_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct dn_fib_table *tb;
 	struct rtattr **rta = arg;
 	struct rtmsg *r = NLMSG_DATA(nlh);
@@ -524,7 +524,7 @@ static int dn_fib_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *
 
 static int dn_fib_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct dn_fib_table *tb;
 	struct rtattr **rta = arg;
 	struct rtmsg *r = NLMSG_DATA(nlh);
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 0a46b6c10e51..2f665a516476 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -1512,7 +1512,7 @@ rtattr_failure:
  */
 static int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = in_skb->sk->sk_net;
+	struct net *net = sock_net(in_skb->sk);
 	struct rtattr **rta = arg;
 	struct rtmsg *rtm = NLMSG_DATA(nlh);
 	struct dn_route *rt = NULL;
@@ -1601,7 +1601,7 @@ out_free:
  */
 int dn_cache_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct dn_route *rt;
 	int h, s_h;
 	int idx, s_idx;
diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c
index e09d915dbd77..3a2830ac89c2 100644
--- a/net/decnet/dn_table.c
+++ b/net/decnet/dn_table.c
@@ -463,7 +463,7 @@ static int dn_fib_table_dump(struct dn_fib_table *tb, struct sk_buff *skb,
 
 int dn_fib_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	unsigned int h, s_h;
 	unsigned int e = 0, s_e;
 	struct dn_fib_table *tb;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 06cfb0bed631..5882a1316441 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -464,7 +464,7 @@ int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	if (addr_len < sizeof(struct sockaddr_in))
 		goto out;
 
-	chk_addr_ret = inet_addr_type(sk->sk_net, addr->sin_addr.s_addr);
+	chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr);
 
 	/* Not specified by any standard per-se, however it breaks too
 	 * many applications when removed.  It is unfortunate since
@@ -802,7 +802,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 {
 	struct sock *sk = sock->sk;
 	int err = 0;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	switch (cmd) {
 		case SIOCGSTAMP:
@@ -1132,7 +1132,7 @@ int inet_sk_rebuild_header(struct sock *sk)
 	};
 
 	security_sk_classify_flow(sk, &fl);
-	err = ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 0);
+	err = ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 0);
 }
 	if (!err)
 		sk_setup_caps(sk, &rt->u.dst);
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 823c724a8593..6848e4760f34 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -437,7 +437,7 @@ struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix,
 
 static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct nlattr *tb[IFA_MAX+1];
 	struct in_device *in_dev;
 	struct ifaddrmsg *ifm;
@@ -552,7 +552,7 @@ errout:
 
 static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct in_ifaddr *ifa;
 
 	ASSERT_RTNL();
@@ -1158,7 +1158,7 @@ nla_put_failure:
 
 static int inet_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int idx, ip_idx;
 	struct net_device *dev;
 	struct in_device *in_dev;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 0e4b34b07cb5..0f1557a4ac7a 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -583,7 +583,7 @@ errout:
 
 static int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct fib_config cfg;
 	struct fib_table *tb;
 	int err;
@@ -605,7 +605,7 @@ errout:
 
 static int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct fib_config cfg;
 	struct fib_table *tb;
 	int err;
@@ -627,7 +627,7 @@ errout:
 
 static int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	unsigned int h, s_h;
 	unsigned int e = 0, s_e;
 	struct fib_table *tb;
@@ -857,7 +857,7 @@ static void nl_fib_input(struct sk_buff *skb)
 	struct fib_table *tb;
 	u32 pid;
 
-	net = skb->sk->sk_net;
+	net = sock_net(skb->sk);
 	nlh = nlmsg_hdr(skb);
 	if (skb->len < NLMSG_SPACE(0) || skb->len < nlh->nlmsg_len ||
 	    nlh->nlmsg_len < NLMSG_LENGTH(sizeof(*frn)))
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 19274d01afa4..1fb56876be54 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -137,7 +137,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 			       struct nlmsghdr *nlh, struct fib_rule_hdr *frh,
 			       struct nlattr **tb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int err = -EINVAL;
 	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 682f632bfb77..6250f4239b61 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -1762,7 +1762,7 @@ int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr)
 	if (!ipv4_is_multicast(addr))
 		return -EINVAL;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return -EPROTONOSUPPORT;
 
 	rtnl_lock();
@@ -1833,7 +1833,7 @@ int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
 	u32 ifindex;
 	int ret = -EADDRNOTAVAIL;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return -EPROTONOSUPPORT;
 
 	rtnl_lock();
@@ -1881,7 +1881,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct
 	if (!ipv4_is_multicast(addr))
 		return -EINVAL;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return -EPROTONOSUPPORT;
 
 	rtnl_lock();
@@ -2017,7 +2017,7 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex)
 	    msf->imsf_fmode != MCAST_EXCLUDE)
 		return -EINVAL;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return -EPROTONOSUPPORT;
 
 	rtnl_lock();
@@ -2100,7 +2100,7 @@ int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf,
 	if (!ipv4_is_multicast(addr))
 		return -EINVAL;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return -EPROTONOSUPPORT;
 
 	rtnl_lock();
@@ -2165,7 +2165,7 @@ int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf,
 	if (!ipv4_is_multicast(addr))
 		return -EINVAL;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return -EPROTONOSUPPORT;
 
 	rtnl_lock();
@@ -2252,7 +2252,7 @@ void ip_mc_drop_socket(struct sock *sk)
 	if (inet->mc_list == NULL)
 		return;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return;
 
 	rtnl_lock();
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index d13c5f12bb32..a7fcaf205644 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -85,7 +85,7 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum)
 	struct hlist_node *node;
 	struct inet_bind_bucket *tb;
 	int ret;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	local_bh_disable();
 	if (!snum) {
@@ -333,7 +333,7 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 					 .dport = ireq->rmt_port } } };
 
 	security_req_classify_flow(req, &fl);
-	if (ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 0)) {
+	if (ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 0)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
 	}
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 8cd1ad9b9111..1064111e5b96 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -139,7 +139,7 @@ static struct sock *inet_lookup_listener_slow(struct net *net,
 	sk_for_each(sk, node, head) {
 		const struct inet_sock *inet = inet_sk(sk);
 
-		if (sk->sk_net == net && inet->num == hnum &&
+		if (sock_net(sk) == net && inet->num == hnum &&
 				!ipv6_only_sock(sk)) {
 			const __be32 rcv_saddr = inet->rcv_saddr;
 			int score = sk->sk_family == PF_INET ? 1 : 0;
@@ -182,7 +182,7 @@ struct sock *__inet_lookup_listener(struct net *net,
 		if (inet->num == hnum && !sk->sk_node.next &&
 		    (!inet->rcv_saddr || inet->rcv_saddr == daddr) &&
 		    (sk->sk_family == PF_INET || !ipv6_only_sock(sk)) &&
-		    !sk->sk_bound_dev_if && sk->sk_net == net)
+		    !sk->sk_bound_dev_if && sock_net(sk) == net)
 			goto sherry_cache;
 		sk = inet_lookup_listener_slow(net, head, daddr, hnum, dif);
 	}
@@ -254,7 +254,7 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row,
 	struct sock *sk2;
 	const struct hlist_node *node;
 	struct inet_timewait_sock *tw;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	prefetch(head->chain.first);
 	write_lock(lock);
@@ -406,7 +406,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row,
 	struct inet_bind_hashbucket *head;
 	struct inet_bind_bucket *tb;
 	int ret;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	if (!snum) {
 		int i, remaining, low, high, port;
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 717c411a5c6b..f12bc24de46f 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -124,7 +124,7 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk, const int stat
 		tw->tw_hash	    = sk->sk_hash;
 		tw->tw_ipv6only	    = 0;
 		tw->tw_prot	    = sk->sk_prot_creator;
-		tw->tw_net          = sk->sk_net;
+		twsk_net_set(tw, sock_net(sk));
 		atomic_set(&tw->tw_refcnt, 1);
 		inet_twsk_dead_node_init(tw);
 		__module_get(tw->tw_prot->owner);
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 26685c83a146..4be00959b748 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -172,7 +172,7 @@ int ip_call_ra_chain(struct sk_buff *skb)
 		if (sk && inet_sk(sk)->num == protocol &&
 		    (!sk->sk_bound_dev_if ||
 		     sk->sk_bound_dev_if == dev->ifindex) &&
-		    sk->sk_net == dev_net(dev)) {
+		    sock_net(sk) == dev_net(dev)) {
 			if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) {
 				if (ip_defrag(skb, IP_DEFRAG_CALL_RA_CHAIN)) {
 					read_unlock(&ip_ra_lock);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 913266cd9902..08349267ceb4 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -351,7 +351,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok)
 			 * itself out.
 			 */
 			security_sk_classify_flow(sk, &fl);
-			if (ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 0))
+			if (ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 0))
 				goto no_route;
 		}
 		sk_setup_caps(sk, &rt->u.dst);
@@ -1382,7 +1382,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 						 .dport = tcp_hdr(skb)->source } },
 				    .proto = sk->sk_protocol };
 		security_skb_classify_flow(skb, &fl);
-		if (ip_route_output_key(sk->sk_net, &rt, &fl))
+		if (ip_route_output_key(sock_net(sk), &rt, &fl))
 			return;
 	}
 
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index b854431047a4..d6e76f5229cc 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -449,7 +449,7 @@ static int do_ip_setsockopt(struct sock *sk, int level,
 		struct ip_options * opt = NULL;
 		if (optlen > 40 || optlen < 0)
 			goto e_inval;
-		err = ip_options_get_from_user(sk->sk_net, &opt,
+		err = ip_options_get_from_user(sock_net(sk), &opt,
 					       optval, optlen);
 		if (err)
 			break;
@@ -590,13 +590,13 @@ static int do_ip_setsockopt(struct sock *sk, int level,
 				err = 0;
 				break;
 			}
-			dev = ip_dev_find(sk->sk_net, mreq.imr_address.s_addr);
+			dev = ip_dev_find(sock_net(sk), mreq.imr_address.s_addr);
 			if (dev) {
 				mreq.imr_ifindex = dev->ifindex;
 				dev_put(dev);
 			}
 		} else
-			dev = __dev_get_by_index(sk->sk_net, mreq.imr_ifindex);
+			dev = __dev_get_by_index(sock_net(sk), mreq.imr_ifindex);
 
 
 		err = -EADDRNOTAVAIL;
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index e54bc1364473..11700a4dcd95 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -849,7 +849,7 @@ static void mrtsock_destruct(struct sock *sk)
 {
 	rtnl_lock();
 	if (sk == mroute_socket) {
-		IPV4_DEVCONF_ALL(sk->sk_net, MC_FORWARDING)--;
+		IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)--;
 
 		write_lock_bh(&mrt_lock);
 		mroute_socket=NULL;
@@ -898,7 +898,7 @@ int ip_mroute_setsockopt(struct sock *sk,int optname,char __user *optval,int opt
 			mroute_socket=sk;
 			write_unlock_bh(&mrt_lock);
 
-			IPV4_DEVCONF_ALL(sk->sk_net, MC_FORWARDING)++;
+			IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)++;
 		}
 		rtnl_unlock();
 		return ret;
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 756bc0e1a7c6..1563f29b5117 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -1496,11 +1496,11 @@ static int compat_do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user,
 
 	switch (cmd) {
 	case ARPT_SO_SET_REPLACE:
-		ret = compat_do_replace(sk->sk_net, user, len);
+		ret = compat_do_replace(sock_net(sk), user, len);
 		break;
 
 	case ARPT_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(sk->sk_net, user, len, 1);
+		ret = do_add_counters(sock_net(sk), user, len, 1);
 		break;
 
 	default:
@@ -1644,10 +1644,10 @@ static int compat_do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user,
 
 	switch (cmd) {
 	case ARPT_SO_GET_INFO:
-		ret = get_info(sk->sk_net, user, len, 1);
+		ret = get_info(sock_net(sk), user, len, 1);
 		break;
 	case ARPT_SO_GET_ENTRIES:
-		ret = compat_get_entries(sk->sk_net, user, len);
+		ret = compat_get_entries(sock_net(sk), user, len);
 		break;
 	default:
 		ret = do_arpt_get_ctl(sk, cmd, user, len);
@@ -1665,11 +1665,11 @@ static int do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned
 
 	switch (cmd) {
 	case ARPT_SO_SET_REPLACE:
-		ret = do_replace(sk->sk_net, user, len);
+		ret = do_replace(sock_net(sk), user, len);
 		break;
 
 	case ARPT_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(sk->sk_net, user, len, 0);
+		ret = do_add_counters(sock_net(sk), user, len, 0);
 		break;
 
 	default:
@@ -1689,11 +1689,11 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len
 
 	switch (cmd) {
 	case ARPT_SO_GET_INFO:
-		ret = get_info(sk->sk_net, user, len, 0);
+		ret = get_info(sock_net(sk), user, len, 0);
 		break;
 
 	case ARPT_SO_GET_ENTRIES:
-		ret = get_entries(sk->sk_net, user, len);
+		ret = get_entries(sock_net(sk), user, len);
 		break;
 
 	case ARPT_SO_GET_REVISION_TARGET: {
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 85a75e186b4b..a819d191e1aa 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -1852,11 +1852,11 @@ compat_do_ipt_set_ctl(struct sock *sk,	int cmd, void __user *user,
 
 	switch (cmd) {
 	case IPT_SO_SET_REPLACE:
-		ret = compat_do_replace(sk->sk_net, user, len);
+		ret = compat_do_replace(sock_net(sk), user, len);
 		break;
 
 	case IPT_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(sk->sk_net, user, len, 1);
+		ret = do_add_counters(sock_net(sk), user, len, 1);
 		break;
 
 	default:
@@ -1963,10 +1963,10 @@ compat_do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 
 	switch (cmd) {
 	case IPT_SO_GET_INFO:
-		ret = get_info(sk->sk_net, user, len, 1);
+		ret = get_info(sock_net(sk), user, len, 1);
 		break;
 	case IPT_SO_GET_ENTRIES:
-		ret = compat_get_entries(sk->sk_net, user, len);
+		ret = compat_get_entries(sock_net(sk), user, len);
 		break;
 	default:
 		ret = do_ipt_get_ctl(sk, cmd, user, len);
@@ -1985,11 +1985,11 @@ do_ipt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len)
 
 	switch (cmd) {
 	case IPT_SO_SET_REPLACE:
-		ret = do_replace(sk->sk_net, user, len);
+		ret = do_replace(sock_net(sk), user, len);
 		break;
 
 	case IPT_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(sk->sk_net, user, len, 0);
+		ret = do_add_counters(sock_net(sk), user, len, 0);
 		break;
 
 	default:
@@ -2010,11 +2010,11 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 
 	switch (cmd) {
 	case IPT_SO_GET_INFO:
-		ret = get_info(sk->sk_net, user, len, 0);
+		ret = get_info(sock_net(sk), user, len, 0);
 		break;
 
 	case IPT_SO_GET_ENTRIES:
-		ret = get_entries(sk->sk_net, user, len);
+		ret = get_entries(sock_net(sk), user, len);
 		break;
 
 	case IPT_SO_GET_REVISION_MATCH:
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 8756d502a47f..be19a4048d7c 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -117,7 +117,7 @@ static struct sock *__raw_v4_lookup(struct net *net, struct sock *sk,
 	sk_for_each_from(sk, node) {
 		struct inet_sock *inet = inet_sk(sk);
 
-		if (sk->sk_net == net && inet->num == num 		&&
+		if (sock_net(sk) == net && inet->num == num 		&&
 		    !(inet->daddr && inet->daddr != raddr) 		&&
 		    !(inet->rcv_saddr && inet->rcv_saddr != laddr)	&&
 		    !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif))
@@ -499,7 +499,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	ipc.oif = sk->sk_bound_dev_if;
 
 	if (msg->msg_controllen) {
-		err = ip_cmsg_send(sk->sk_net, msg, &ipc);
+		err = ip_cmsg_send(sock_net(sk), msg, &ipc);
 		if (err)
 			goto out;
 		if (ipc.opt)
@@ -553,7 +553,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		}
 
 		security_sk_classify_flow(sk, &fl);
-		err = ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 1);
+		err = ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 1);
 	}
 	if (err)
 		goto done;
@@ -620,7 +620,7 @@ static int raw_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 
 	if (sk->sk_state != TCP_CLOSE || addr_len < sizeof(struct sockaddr_in))
 		goto out;
-	chk_addr_ret = inet_addr_type(sk->sk_net, addr->sin_addr.s_addr);
+	chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr);
 	ret = -EADDRNOTAVAIL;
 	if (addr->sin_addr.s_addr && chk_addr_ret != RTN_LOCAL &&
 	    chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST)
@@ -856,7 +856,7 @@ static struct sock *raw_get_first(struct seq_file *seq)
 		struct hlist_node *node;
 
 		sk_for_each(sk, node, &state->h->ht[state->bucket])
-			if (sk->sk_net == state->p.net)
+			if (sock_net(sk) == state->p.net)
 				goto found;
 	}
 	sk = NULL;
@@ -872,7 +872,7 @@ static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk)
 		sk = sk_next(sk);
 try_again:
 		;
-	} while (sk && sk->sk_net != state->p.net);
+	} while (sk && sock_net(sk) != state->p.net);
 
 	if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
 		sk = sk_head(&state->h->ht[state->bucket]);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 7768d718e199..194f5cca3121 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2689,7 +2689,7 @@ nla_put_failure:
 
 static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = in_skb->sk->sk_net;
+	struct net *net = sock_net(in_skb->sk);
 	struct rtmsg *rtm;
 	struct nlattr *tb[RTA_MAX+1];
 	struct rtable *rt = NULL;
@@ -2785,7 +2785,7 @@ int ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb)
 	int idx, s_idx;
 	struct net *net;
 
-	net = skb->sk->sk_net;
+	net = sock_net(skb->sk);
 
 	s_h = cb->args[0];
 	if (s_h < 0)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 28bece6f281b..46847e600a46 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1486,7 +1486,7 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
 	if (req)
 		return tcp_check_req(sk, skb, req, prev);
 
-	nsk = inet_lookup_established(sk->sk_net, &tcp_hashinfo, iph->saddr,
+	nsk = inet_lookup_established(sock_net(sk), &tcp_hashinfo, iph->saddr,
 			th->source, iph->daddr, th->dest, inet_iif(skb));
 
 	if (nsk) {
@@ -1974,7 +1974,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur)
 		while (1) {
 			while (req) {
 				if (req->rsk_ops->family == st->family &&
-				    req->sk->sk_net == net) {
+				    sock_net(req->sk) == net) {
 					cur = req;
 					goto out;
 				}
@@ -1998,7 +1998,7 @@ get_req:
 	}
 get_sk:
 	sk_for_each_from(sk, node) {
-		if (sk->sk_family == st->family && sk->sk_net == net) {
+		if (sk->sk_family == st->family && sock_net(sk) == net) {
 			cur = sk;
 			goto out;
 		}
@@ -2049,7 +2049,7 @@ static void *established_get_first(struct seq_file *seq)
 		read_lock_bh(lock);
 		sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) {
 			if (sk->sk_family != st->family ||
-			    sk->sk_net != net) {
+			    sock_net(sk) != net) {
 				continue;
 			}
 			rc = sk;
@@ -2059,7 +2059,7 @@ static void *established_get_first(struct seq_file *seq)
 		inet_twsk_for_each(tw, node,
 				   &tcp_hashinfo.ehash[st->bucket].twchain) {
 			if (tw->tw_family != st->family ||
-			    tw->tw_net != net) {
+			    twsk_net(tw) != net) {
 				continue;
 			}
 			rc = tw;
@@ -2086,7 +2086,7 @@ static void *established_get_next(struct seq_file *seq, void *cur)
 		tw = cur;
 		tw = tw_next(tw);
 get_tw:
-		while (tw && (tw->tw_family != st->family || tw->tw_net != net)) {
+		while (tw && (tw->tw_family != st->family || twsk_net(tw) != net)) {
 			tw = tw_next(tw);
 		}
 		if (tw) {
@@ -2107,7 +2107,7 @@ get_tw:
 		sk = sk_next(sk);
 
 	sk_for_each_from(sk, node) {
-		if (sk->sk_family == st->family && sk->sk_net == net)
+		if (sk->sk_family == st->family && sock_net(sk) == net)
 			goto found;
 	}
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index e2cd93481359..76d52d37d6ac 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -137,7 +137,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
 	struct hlist_node *node;
 
 	sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)])
-		if (sk->sk_net == net && sk->sk_hash == num)
+		if (sock_net(sk) == net && sk->sk_hash == num)
 			return 1;
 	return 0;
 }
@@ -158,7 +158,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 	struct hlist_head *head;
 	struct sock *sk2;
 	int    error = 1;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	write_lock_bh(&udp_hash_lock);
 
@@ -218,7 +218,7 @@ gotit:
 		sk_for_each(sk2, node, head)
 			if (sk2->sk_hash == snum                             &&
 			    sk2 != sk                                        &&
-			    sk2->sk_net == net				     &&
+			    sock_net(sk2) == net				     &&
 			    (!sk2->sk_reuse        || !sk->sk_reuse)         &&
 			    (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if
 			     || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
@@ -269,7 +269,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
 	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
-		if (sk->sk_net == net && sk->sk_hash == hnum &&
+		if (sock_net(sk) == net && sk->sk_hash == hnum &&
 				!ipv6_only_sock(sk)) {
 			int score = (sk->sk_family == PF_INET ? 1 : 0);
 			if (inet->rcv_saddr) {
@@ -607,7 +607,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 
 	ipc.oif = sk->sk_bound_dev_if;
 	if (msg->msg_controllen) {
-		err = ip_cmsg_send(sk->sk_net, msg, &ipc);
+		err = ip_cmsg_send(sock_net(sk), msg, &ipc);
 		if (err)
 			return err;
 		if (ipc.opt)
@@ -656,7 +656,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 					       { .sport = inet->sport,
 						 .dport = dport } } };
 		security_sk_classify_flow(sk, &fl);
-		err = ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 1);
+		err = ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 1);
 		if (err) {
 			if (err == -ENETUNREACH)
 				IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
@@ -1511,7 +1511,7 @@ static struct sock *udp_get_first(struct seq_file *seq)
 	for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) {
 		struct hlist_node *node;
 		sk_for_each(sk, node, state->hashtable + state->bucket) {
-			if (sk->sk_net != net)
+			if (sock_net(sk) != net)
 				continue;
 			if (sk->sk_family == state->family)
 				goto found;
@@ -1531,7 +1531,7 @@ static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk)
 		sk = sk_next(sk);
 try_again:
 		;
-	} while (sk && (sk->sk_net != net || sk->sk_family != state->family));
+	} while (sk && (sock_net(sk) != net || sk->sk_family != state->family));
 
 	if (!sk && ++state->bucket < UDP_HTABLE_SIZE) {
 		sk = sk_head(state->hashtable + state->bucket);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index d1de9ec74261..f2c90f145cbb 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3054,7 +3054,7 @@ static const struct nla_policy ifa_ipv6_policy[IFA_MAX+1] = {
 static int
 inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ifaddrmsg *ifm;
 	struct nlattr *tb[IFA_MAX+1];
 	struct in6_addr *pfx;
@@ -3112,7 +3112,7 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, u8 ifa_flags,
 static int
 inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ifaddrmsg *ifm;
 	struct nlattr *tb[IFA_MAX+1];
 	struct in6_addr *pfx;
@@ -3322,7 +3322,7 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb,
 	struct inet6_ifaddr *ifa;
 	struct ifmcaddr6 *ifmca;
 	struct ifacaddr6 *ifaca;
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 
 	s_idx = cb->args[0];
 	s_ip_idx = ip_idx = cb->args[1];
@@ -3418,7 +3418,7 @@ static int inet6_dump_ifacaddr(struct sk_buff *skb, struct netlink_callback *cb)
 static int inet6_rtm_getaddr(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 			     void *arg)
 {
-	struct net *net = in_skb->sk->sk_net;
+	struct net *net = sock_net(in_skb->sk);
 	struct ifaddrmsg *ifm;
 	struct nlattr *tb[IFA_MAX+1];
 	struct in6_addr *addr = NULL;
@@ -3645,7 +3645,7 @@ nla_put_failure:
 
 static int inet6_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int idx, err;
 	int s_idx = cb->args[0];
 	struct net_device *dev;
diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c
index de371b5997fe..9bfa8846f262 100644
--- a/net/ipv6/addrlabel.c
+++ b/net/ipv6/addrlabel.c
@@ -364,7 +364,7 @@ static const struct nla_policy ifal_policy[IFAL_MAX+1] = {
 static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh,
 			     void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ifaddrlblmsg *ifal;
 	struct nlattr *tb[IFAL_MAX+1];
 	struct in6_addr *pfx;
@@ -452,7 +452,7 @@ static int ip6addrlbl_fill(struct sk_buff *skb,
 
 static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct ip6addrlbl_entry *p;
 	struct hlist_node *pos;
 	int idx = 0, s_idx = cb->args[0];
@@ -490,7 +490,7 @@ static inline int ip6addrlbl_msgsize(void)
 static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
 			  void *arg)
 {
-	struct net *net = in_skb->sk->sk_net;
+	struct net *net = sock_net(in_skb->sk);
 	struct ifaddrlblmsg *ifal;
 	struct nlattr *tb[IFAL_MAX+1];
 	struct in6_addr *addr;
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index f52bdaed8a1b..12f04e9d3e88 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -245,7 +245,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	struct sock *sk = sock->sk;
 	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	__be32 v4addr = 0;
 	unsigned short snum;
 	int addr_type = 0;
@@ -438,7 +438,7 @@ EXPORT_SYMBOL(inet6_getname);
 int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 {
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	switch(cmd)
 	{
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index e7a7fe26cebf..cac580749ebe 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -154,7 +154,7 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 			       struct nlattr **tb)
 {
 	int err = -EINVAL;
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct fib6_rule *rule6 = (struct fib6_rule *) rule;
 
 	if (rule->action == FR_ACT_TO_TBL) {
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 50857662e6b7..63309d10df3a 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -163,7 +163,7 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type,
 				     struct flowi *fl)
 {
 	struct dst_entry *dst;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	int res = 0;
 
 	/* Informational messages are not limited. */
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index c0c8d2d17682..21c467675412 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -105,7 +105,7 @@ struct sock *inet6_lookup_listener(struct net *net,
 
 	read_lock(&hashinfo->lhash_lock);
 	sk_for_each(sk, node, &hashinfo->listening_hash[inet_lhashfn(hnum)]) {
-		if (sk->sk_net == net && inet_sk(sk)->num == hnum &&
+		if (sock_net(sk) == net && inet_sk(sk)->num == hnum &&
 				sk->sk_family == PF_INET6) {
 			const struct ipv6_pinfo *np = inet6_sk(sk);
 
@@ -172,7 +172,7 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row,
 	struct sock *sk2;
 	const struct hlist_node *node;
 	struct inet_timewait_sock *tw;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	prefetch(head->chain.first);
 	write_lock(lock);
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index b0814b0082e7..b3f6e03c454c 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -346,7 +346,7 @@ end:
 
 static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	unsigned int h, s_h;
 	unsigned int e = 0, s_e;
 	struct rt6_rtnl_dump_arg arg;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 556300f0eba5..a8b4da25b0a7 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -910,7 +910,7 @@ static int ip6_dst_lookup_tail(struct sock *sk,
 			       struct dst_entry **dst, struct flowi *fl)
 {
 	int err;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	if (*dst == NULL)
 		*dst = ip6_route_output(net, sk, fl);
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index dc6695cc5767..d3d93d752e10 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -107,7 +107,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 		    char __user *optval, int optlen)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	int val, valbool;
 	int retv = -ENOPROTOOPT;
 
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 0357de8e78c8..20a3d8e2f6c6 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -181,7 +181,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 	struct net_device *dev = NULL;
 	struct ipv6_mc_socklist *mc_lst;
 	struct ipv6_pinfo *np = inet6_sk(sk);
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	int err;
 
 	if (!ipv6_addr_is_multicast(addr))
@@ -255,7 +255,7 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6_mc_socklist *mc_lst, **lnk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	write_lock_bh(&ipv6_sk_mc_lock);
 	for (lnk = &np->ipv6_mc_list; (mc_lst = *lnk) !=NULL ; lnk = &mc_lst->next) {
@@ -327,7 +327,7 @@ void ipv6_sock_mc_close(struct sock *sk)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6_mc_socklist *mc_lst;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	write_lock_bh(&ipv6_sk_mc_lock);
 	while ((mc_lst = np->ipv6_mc_list) != NULL) {
@@ -365,7 +365,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
 	struct inet6_dev *idev;
 	struct ipv6_pinfo *inet6 = inet6_sk(sk);
 	struct ip6_sf_socklist *psl;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	int i, j, rv;
 	int leavegroup = 0;
 	int pmclocked = 0;
@@ -505,7 +505,7 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf)
 	struct inet6_dev *idev;
 	struct ipv6_pinfo *inet6 = inet6_sk(sk);
 	struct ip6_sf_socklist *newpsl, *psl;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	int leavegroup = 0;
 	int i, err;
 
@@ -598,7 +598,7 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 	struct net_device *dev;
 	struct ipv6_pinfo *inet6 = inet6_sk(sk);
 	struct ip6_sf_socklist *psl;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 
 	group = &((struct sockaddr_in6 *)&gsf->gf_group)->sin6_addr;
 
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index af1ec7ba757c..70ef0d276cc0 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -1879,11 +1879,11 @@ compat_do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user,
 
 	switch (cmd) {
 	case IP6T_SO_SET_REPLACE:
-		ret = compat_do_replace(sk->sk_net, user, len);
+		ret = compat_do_replace(sock_net(sk), user, len);
 		break;
 
 	case IP6T_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(sk->sk_net, user, len, 1);
+		ret = do_add_counters(sock_net(sk), user, len, 1);
 		break;
 
 	default:
@@ -1990,10 +1990,10 @@ compat_do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 
 	switch (cmd) {
 	case IP6T_SO_GET_INFO:
-		ret = get_info(sk->sk_net, user, len, 1);
+		ret = get_info(sock_net(sk), user, len, 1);
 		break;
 	case IP6T_SO_GET_ENTRIES:
-		ret = compat_get_entries(sk->sk_net, user, len);
+		ret = compat_get_entries(sock_net(sk), user, len);
 		break;
 	default:
 		ret = do_ip6t_get_ctl(sk, cmd, user, len);
@@ -2012,11 +2012,11 @@ do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len)
 
 	switch (cmd) {
 	case IP6T_SO_SET_REPLACE:
-		ret = do_replace(sk->sk_net, user, len);
+		ret = do_replace(sock_net(sk), user, len);
 		break;
 
 	case IP6T_SO_SET_ADD_COUNTERS:
-		ret = do_add_counters(sk->sk_net, user, len, 0);
+		ret = do_add_counters(sock_net(sk), user, len, 0);
 		break;
 
 	default:
@@ -2037,11 +2037,11 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 
 	switch (cmd) {
 	case IP6T_SO_GET_INFO:
-		ret = get_info(sk->sk_net, user, len, 0);
+		ret = get_info(sock_net(sk), user, len, 0);
 		break;
 
 	case IP6T_SO_GET_ENTRIES:
-		ret = get_entries(sk->sk_net, user, len);
+		ret = get_entries(sock_net(sk), user, len);
 		break;
 
 	case IP6T_SO_GET_REVISION_MATCH:
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index efb0047f6880..12c7a1560977 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -76,7 +76,7 @@ static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk,
 		if (inet_sk(sk)->num == num) {
 			struct ipv6_pinfo *np = inet6_sk(sk);
 
-			if (sk->sk_net != net)
+			if (sock_net(sk) != net)
 				continue;
 
 			if (!ipv6_addr_any(&np->daddr) &&
@@ -280,7 +280,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 			if (!sk->sk_bound_dev_if)
 				goto out;
 
-			dev = dev_get_by_index(sk->sk_net, sk->sk_bound_dev_if);
+			dev = dev_get_by_index(sock_net(sk), sk->sk_bound_dev_if);
 			if (!dev) {
 				err = -ENODEV;
 				goto out;
@@ -293,7 +293,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 		v4addr = LOOPBACK4_IPV6;
 		if (!(addr_type & IPV6_ADDR_MULTICAST))	{
 			err = -EADDRNOTAVAIL;
-			if (!ipv6_chk_addr(sk->sk_net, &addr->sin6_addr,
+			if (!ipv6_chk_addr(sock_net(sk), &addr->sin6_addr,
 					   dev, 0)) {
 				if (dev)
 					dev_put(dev);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 65053fba8c1a..ac4428371432 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2020,7 +2020,7 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 	cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid;
 	cfg->fc_nlinfo.nlh = nlh;
-	cfg->fc_nlinfo.nl_net = skb->sk->sk_net;
+	cfg->fc_nlinfo.nl_net = sock_net(skb->sk);
 
 	if (tb[RTA_GATEWAY]) {
 		nla_memcpy(&cfg->fc_gateway, tb[RTA_GATEWAY], 16);
@@ -2216,7 +2216,7 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg)
 
 static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct net *net = in_skb->sk->sk_net;
+	struct net *net = sock_net(in_skb->sk);
 	struct nlattr *tb[RTA_MAX+1];
 	struct rt6_info *rt;
 	struct sk_buff *skb;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 086deffff9c9..323c7e06ef43 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1218,7 +1218,7 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb)
 	if (req)
 		return tcp_check_req(sk, skb, req, prev);
 
-	nsk = __inet6_lookup_established(sk->sk_net, &tcp_hashinfo,
+	nsk = __inet6_lookup_established(sock_net(sk), &tcp_hashinfo,
 			&ipv6_hdr(skb)->saddr, th->source,
 			&ipv6_hdr(skb)->daddr, ntohs(th->dest), inet6_iif(skb));
 
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 6683c04b427e..db266ff297e5 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -70,7 +70,7 @@ static struct sock *__udp6_lib_lookup(struct net *net,
 	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
-		if (sk->sk_net == net && sk->sk_hash == hnum &&
+		if (sock_net(sk) == net && sk->sk_hash == hnum &&
 				sk->sk_family == PF_INET6) {
 			struct ipv6_pinfo *np = inet6_sk(sk);
 			int score = 0;
@@ -323,7 +323,7 @@ static struct sock *udp_v6_mcast_next(struct sock *sk,
 	sk_for_each_from(s, node) {
 		struct inet_sock *inet = inet_sk(s);
 
-		if (s->sk_net != sk->sk_net)
+		if (sock_net(s) != sock_net(sk))
 			continue;
 
 		if (s->sk_hash == num && s->sk_family == PF_INET6) {
diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c
index 6f21a53cb3e7..ae54b20d0470 100644
--- a/net/irda/af_irda.c
+++ b/net/irda/af_irda.c
@@ -837,7 +837,7 @@ static int irda_accept(struct socket *sock, struct socket *newsock, int flags)
 
 	IRDA_DEBUG(2, "%s()\n", __func__);
 
-	err = irda_create(sk->sk_net, newsock, sk->sk_protocol);
+	err = irda_create(sock_net(sk), newsock, sk->sk_protocol);
 	if (err)
 		return err;
 
diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c
index 5ebfd93ff5e7..5c6d89c6d51d 100644
--- a/net/llc/llc_conn.c
+++ b/net/llc/llc_conn.c
@@ -700,7 +700,7 @@ static struct sock *llc_create_incoming_sock(struct sock *sk,
 					     struct llc_addr *saddr,
 					     struct llc_addr *daddr)
 {
-	struct sock *newsk = llc_sk_alloc(sk->sk_net, sk->sk_family, GFP_ATOMIC,
+	struct sock *newsk = llc_sk_alloc(sock_net(sk), sk->sk_family, GFP_ATOMIC,
 					  sk->sk_prot);
 	struct llc_sock *newllc, *llc = llc_sk(sk);
 
diff --git a/net/netfilter/nf_sockopt.c b/net/netfilter/nf_sockopt.c
index 3dd4b3c76d81..69d699f95f4c 100644
--- a/net/netfilter/nf_sockopt.c
+++ b/net/netfilter/nf_sockopt.c
@@ -65,7 +65,7 @@ static struct nf_sockopt_ops *nf_sockopt_find(struct sock *sk, int pf,
 {
 	struct nf_sockopt_ops *ops;
 
-	if (sk->sk_net != &init_net)
+	if (sock_net(sk) != &init_net)
 		return ERR_PTR(-ENOPROTOOPT);
 
 	if (mutex_lock_interruptible(&nf_sockopt_mutex) != 0)
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 86bd8660a8f2..712a7bff8560 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -228,7 +228,7 @@ static inline struct sock *netlink_lookup(struct net *net, int protocol,
 	read_lock(&nl_table_lock);
 	head = nl_pid_hashfn(hash, pid);
 	sk_for_each(sk, node, head) {
-		if ((sk->sk_net == net) && (nlk_sk(sk)->pid == pid)) {
+		if (sock_net(sk) == net && (nlk_sk(sk)->pid == pid)) {
 			sock_hold(sk);
 			goto found;
 		}
@@ -348,7 +348,7 @@ static int netlink_insert(struct sock *sk, struct net *net, u32 pid)
 	head = nl_pid_hashfn(hash, pid);
 	len = 0;
 	sk_for_each(osk, node, head) {
-		if ((osk->sk_net == net) && (nlk_sk(osk)->pid == pid))
+		if (sock_net(osk) == net && (nlk_sk(osk)->pid == pid))
 			break;
 		len++;
 	}
@@ -486,7 +486,7 @@ static int netlink_release(struct socket *sock)
 
 	if (nlk->pid && !nlk->subscriptions) {
 		struct netlink_notify n = {
-						.net = sk->sk_net,
+						.net = sock_net(sk),
 						.protocol = sk->sk_protocol,
 						.pid = nlk->pid,
 					  };
@@ -518,7 +518,7 @@ static int netlink_release(struct socket *sock)
 static int netlink_autobind(struct socket *sock)
 {
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	struct nl_pid_hash *hash = &nl_table[sk->sk_protocol].hash;
 	struct hlist_head *head;
 	struct sock *osk;
@@ -532,7 +532,7 @@ retry:
 	netlink_table_grab();
 	head = nl_pid_hashfn(hash, pid);
 	sk_for_each(osk, node, head) {
-		if ((osk->sk_net != net))
+		if (sock_net(osk) != net)
 			continue;
 		if (nlk_sk(osk)->pid == pid) {
 			/* Bind collision, search negative pid values. */
@@ -611,7 +611,7 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr,
 			int addr_len)
 {
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	struct netlink_sock *nlk = nlk_sk(sk);
 	struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr;
 	int err;
@@ -720,7 +720,7 @@ static struct sock *netlink_getsockbypid(struct sock *ssk, u32 pid)
 	struct sock *sock;
 	struct netlink_sock *nlk;
 
-	sock = netlink_lookup(ssk->sk_net, ssk->sk_protocol, pid);
+	sock = netlink_lookup(sock_net(ssk), ssk->sk_protocol, pid);
 	if (!sock)
 		return ERR_PTR(-ECONNREFUSED);
 
@@ -962,7 +962,7 @@ static inline int do_one_broadcast(struct sock *sk,
 	    !test_bit(p->group - 1, nlk->groups))
 		goto out;
 
-	if ((sk->sk_net != p->net))
+	if (sock_net(sk) != p->net)
 		goto out;
 
 	if (p->failure) {
@@ -1006,7 +1006,7 @@ out:
 int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 pid,
 		      u32 group, gfp_t allocation)
 {
-	struct net *net = ssk->sk_net;
+	struct net *net = sock_net(ssk);
 	struct netlink_broadcast_data info;
 	struct hlist_node *node;
 	struct sock *sk;
@@ -1064,7 +1064,7 @@ static inline int do_one_set_err(struct sock *sk,
 	if (sk == p->exclude_sk)
 		goto out;
 
-	if (sk->sk_net != p->exclude_sk->sk_net)
+	if (sock_net(sk) != sock_net(p->exclude_sk))
 		goto out;
 
 	if (nlk->pid == p->pid || p->group - 1 >= nlk->ngroups ||
@@ -1601,7 +1601,7 @@ int netlink_dump_start(struct sock *ssk, struct sk_buff *skb,
 	atomic_inc(&skb->users);
 	cb->skb = skb;
 
-	sk = netlink_lookup(ssk->sk_net, ssk->sk_protocol, NETLINK_CB(skb).pid);
+	sk = netlink_lookup(sock_net(ssk), ssk->sk_protocol, NETLINK_CB(skb).pid);
 	if (sk == NULL) {
 		netlink_destroy_callback(cb);
 		return -ECONNREFUSED;
@@ -1643,7 +1643,7 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err)
 	if (!skb) {
 		struct sock *sk;
 
-		sk = netlink_lookup(in_skb->sk->sk_net,
+		sk = netlink_lookup(sock_net(in_skb->sk),
 				    in_skb->sk->sk_protocol,
 				    NETLINK_CB(in_skb).pid);
 		if (sk) {
@@ -1758,7 +1758,7 @@ static struct sock *netlink_seq_socket_idx(struct seq_file *seq, loff_t pos)
 
 		for (j = 0; j <= hash->mask; j++) {
 			sk_for_each(s, node, &hash->table[j]) {
-				if (iter->p.net != s->sk_net)
+				if (sock_net(s) != iter->p.net)
 					continue;
 				if (off == pos) {
 					iter->link = i;
@@ -1794,7 +1794,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 	s = v;
 	do {
 		s = sk_next(s);
-	} while (s && (iter->p.net != s->sk_net));
+	} while (s && (sock_net(s) != iter->p.net));
 	if (s)
 		return s;
 
@@ -1806,7 +1806,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 
 		for (; j <= hash->mask; j++) {
 			s = sk_head(&hash->table[j]);
-			while (s && (iter->p.net != s->sk_net))
+			while (s && sock_net(s) != iter->p.net)
 				s = sk_next(s);
 			if (s) {
 				iter->link = i;
diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index a270ebf9f765..4bae8b998cab 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -466,7 +466,7 @@ static struct sock *nr_make_new(struct sock *osk)
 	if (osk->sk_type != SOCK_SEQPACKET)
 		return NULL;
 
-	sk = sk_alloc(osk->sk_net, PF_NETROM, GFP_ATOMIC, osk->sk_prot);
+	sk = sk_alloc(sock_net(osk), PF_NETROM, GFP_ATOMIC, osk->sk_prot);
 	if (sk == NULL)
 		return NULL;
 
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index baa290d3444a..25070240d4ae 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -263,7 +263,7 @@ static int packet_rcv_spkt(struct sk_buff *skb, struct net_device *dev,  struct
 	if (skb->pkt_type == PACKET_LOOPBACK)
 		goto out;
 
-	if (dev_net(dev) != sk->sk_net)
+	if (dev_net(dev) != sock_net(sk))
 		goto out;
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
@@ -337,7 +337,7 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock,
 	 */
 
 	saddr->spkt_device[13] = 0;
-	dev = dev_get_by_name(sk->sk_net, saddr->spkt_device);
+	dev = dev_get_by_name(sock_net(sk), saddr->spkt_device);
 	err = -ENODEV;
 	if (dev == NULL)
 		goto out_unlock;
@@ -451,7 +451,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet
 	sk = pt->af_packet_priv;
 	po = pkt_sk(sk);
 
-	if (dev_net(dev) != sk->sk_net)
+	if (dev_net(dev) != sock_net(sk))
 		goto drop;
 
 	skb->dev = dev;
@@ -568,7 +568,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	sk = pt->af_packet_priv;
 	po = pkt_sk(sk);
 
-	if (dev_net(dev) != sk->sk_net)
+	if (dev_net(dev) != sock_net(sk))
 		goto drop;
 
 	if (dev->header_ops) {
@@ -728,7 +728,7 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock,
 	}
 
 
-	dev = dev_get_by_index(sk->sk_net, ifindex);
+	dev = dev_get_by_index(sock_net(sk), ifindex);
 	err = -ENXIO;
 	if (dev == NULL)
 		goto out_unlock;
@@ -800,7 +800,7 @@ static int packet_release(struct socket *sock)
 	if (!sk)
 		return 0;
 
-	net = sk->sk_net;
+	net = sock_net(sk);
 	po = pkt_sk(sk);
 
 	write_lock_bh(&net->packet.sklist_lock);
@@ -914,7 +914,7 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr *uaddr, int add
 		return -EINVAL;
 	strlcpy(name,uaddr->sa_data,sizeof(name));
 
-	dev = dev_get_by_name(sk->sk_net, name);
+	dev = dev_get_by_name(sock_net(sk), name);
 	if (dev) {
 		err = packet_do_bind(sk, dev, pkt_sk(sk)->num);
 		dev_put(dev);
@@ -941,7 +941,7 @@ static int packet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len
 
 	if (sll->sll_ifindex) {
 		err = -ENODEV;
-		dev = dev_get_by_index(sk->sk_net, sll->sll_ifindex);
+		dev = dev_get_by_index(sock_net(sk), sll->sll_ifindex);
 		if (dev == NULL)
 			goto out;
 	}
@@ -1135,7 +1135,7 @@ static int packet_getname_spkt(struct socket *sock, struct sockaddr *uaddr,
 		return -EOPNOTSUPP;
 
 	uaddr->sa_family = AF_PACKET;
-	dev = dev_get_by_index(sk->sk_net, pkt_sk(sk)->ifindex);
+	dev = dev_get_by_index(sock_net(sk), pkt_sk(sk)->ifindex);
 	if (dev) {
 		strlcpy(uaddr->sa_data, dev->name, 15);
 		dev_put(dev);
@@ -1160,7 +1160,7 @@ static int packet_getname(struct socket *sock, struct sockaddr *uaddr,
 	sll->sll_family = AF_PACKET;
 	sll->sll_ifindex = po->ifindex;
 	sll->sll_protocol = po->num;
-	dev = dev_get_by_index(sk->sk_net, po->ifindex);
+	dev = dev_get_by_index(sock_net(sk), po->ifindex);
 	if (dev) {
 		sll->sll_hatype = dev->type;
 		sll->sll_halen = dev->addr_len;
@@ -1212,7 +1212,7 @@ static int packet_mc_add(struct sock *sk, struct packet_mreq_max *mreq)
 	rtnl_lock();
 
 	err = -ENODEV;
-	dev = __dev_get_by_index(sk->sk_net, mreq->mr_ifindex);
+	dev = __dev_get_by_index(sock_net(sk), mreq->mr_ifindex);
 	if (!dev)
 		goto done;
 
@@ -1266,7 +1266,7 @@ static int packet_mc_drop(struct sock *sk, struct packet_mreq_max *mreq)
 			if (--ml->count == 0) {
 				struct net_device *dev;
 				*mlp = ml->next;
-				dev = dev_get_by_index(sk->sk_net, ml->ifindex);
+				dev = dev_get_by_index(sock_net(sk), ml->ifindex);
 				if (dev) {
 					packet_dev_mc(dev, ml, -1);
 					dev_put(dev);
@@ -1294,7 +1294,7 @@ static void packet_flush_mclist(struct sock *sk)
 		struct net_device *dev;
 
 		po->mclist = ml->next;
-		if ((dev = dev_get_by_index(sk->sk_net, ml->ifindex)) != NULL) {
+		if ((dev = dev_get_by_index(sock_net(sk), ml->ifindex)) != NULL) {
 			packet_dev_mc(dev, ml, -1);
 			dev_put(dev);
 		}
@@ -1540,7 +1540,7 @@ static int packet_ioctl(struct socket *sock, unsigned int cmd,
 		case SIOCGIFDSTADDR:
 		case SIOCSIFDSTADDR:
 		case SIOCSIFFLAGS:
-			if (sk->sk_net != &init_net)
+			if (sock_net(sk) != &init_net)
 				return -ENOIOCTLCMD;
 			return inet_dgram_ops.ioctl(sock, cmd, arg);
 #endif
diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index 1a7f143cf741..92d85c38e4d2 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c
@@ -551,7 +551,7 @@ static struct sock *rose_make_new(struct sock *osk)
 	if (osk->sk_type != SOCK_SEQPACKET)
 		return NULL;
 
-	sk = sk_alloc(osk->sk_net, PF_ROSE, GFP_ATOMIC, &rose_proto);
+	sk = sk_alloc(sock_net(osk), PF_ROSE, GFP_ATOMIC, &rose_proto);
 	if (sk == NULL)
 		return NULL;
 
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 0b8eb235bc13..74e662cbb2c5 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -951,7 +951,7 @@ done:
 
 static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct nlattr *tca[TCA_ACT_MAX + 1];
 	u32 pid = skb ? NETLINK_CB(skb).pid : 0;
 	int ret = 0, ovr = 0;
@@ -1029,7 +1029,7 @@ find_dump_kind(struct nlmsghdr *n)
 static int
 tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct nlmsghdr *nlh;
 	unsigned char *b = skb_tail_pointer(skb);
 	struct nlattr *nest;
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 0fbedcabf111..1086df7478bc 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -118,7 +118,7 @@ static inline u32 tcf_auto_prio(struct tcf_proto *tp)
 
 static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct nlattr *tca[TCA_MAX + 1];
 	struct tcmsg *t;
 	u32 protocol;
@@ -389,7 +389,7 @@ static int tcf_node_dump(struct tcf_proto *tp, unsigned long n,
 
 static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int t;
 	int s_t;
 	struct net_device *dev;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 7e3c048ba9b1..15b91a9ee8e8 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -605,7 +605,7 @@ check_loop_fn(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w)
 
 static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct tcmsg *tcm = NLMSG_DATA(n);
 	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
@@ -674,7 +674,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 
 static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct tcmsg *tcm;
 	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
@@ -893,7 +893,7 @@ err_out:
 
 static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int idx, q_idx;
 	int s_idx, s_q_idx;
 	struct net_device *dev;
@@ -945,7 +945,7 @@ done:
 
 static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	struct tcmsg *tcm = NLMSG_DATA(n);
 	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
@@ -1139,7 +1139,7 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk
 
 static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct net *net = skb->sk->sk_net;
+	struct net *net = sock_net(skb->sk);
 	int t;
 	int s_t;
 	struct net_device *dev;
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index dc71d0d83753..036bfcc8d15b 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -636,7 +636,7 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk,
 	struct ipv6_pinfo *newnp, *np = inet6_sk(sk);
 	struct sctp6_sock *newsctp6sk;
 
-	newsk = sk_alloc(sk->sk_net, PF_INET6, GFP_KERNEL, sk->sk_prot);
+	newsk = sk_alloc(sock_net(sk), PF_INET6, GFP_KERNEL, sk->sk_prot);
 	if (!newsk)
 		goto out;
 
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index 2faa0d8839eb..5aea91137fbb 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -554,7 +554,7 @@ static struct sock *sctp_v4_create_accept_sk(struct sock *sk,
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct inet_sock *newinet;
-	struct sock *newsk = sk_alloc(sk->sk_net, PF_INET, GFP_KERNEL,
+	struct sock *newsk = sk_alloc(sock_net(sk), PF_INET, GFP_KERNEL,
 			sk->sk_prot);
 
 	if (!newsk)
diff --git a/net/socket.c b/net/socket.c
index 9d3fbfbc8535..79e5382fd110 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -857,7 +857,7 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 
 	sock = file->private_data;
 	sk = sock->sk;
-	net = sk->sk_net;
+	net = sock_net(sk);
 	if (cmd >= SIOCDEVPRIVATE && cmd <= (SIOCDEVPRIVATE + 15)) {
 		err = dev_ioctl(net, cmd, argp);
 	} else
@@ -1375,7 +1375,7 @@ asmlinkage long sys_listen(int fd, int backlog)
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (sock) {
-		somaxconn = sock->sk->sk_net->sysctl_somaxconn;
+		somaxconn = sock_net(sock->sk)->sysctl_somaxconn;
 		if ((unsigned)backlog > somaxconn)
 			backlog = somaxconn;
 
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 3220d5cb5b5d..ae45df060e3a 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -1375,7 +1375,7 @@ static int accept(struct socket *sock, struct socket *newsock, int flags)
 	}
 	buf = skb_peek(&sock->sk->sk_receive_queue);
 
-	res = tipc_create(sock->sk->sk_net, newsock, 0);
+	res = tipc_create(sock_net(sock->sk), newsock, 0);
 	if (!res) {
 		struct tipc_sock *new_tsock = tipc_sk(newsock->sk);
 		struct tipc_portid id;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index ae584356852c..cb9d0cb5f270 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -252,7 +252,7 @@ static struct sock *__unix_find_socket_byname(struct net *net,
 	sk_for_each(s, node, &unix_socket_table[hash ^ type]) {
 		struct unix_sock *u = unix_sk(s);
 
-		if (s->sk_net != net)
+		if (sock_net(s) != net)
 			continue;
 
 		if (u->addr->len == len &&
@@ -289,7 +289,7 @@ static struct sock *unix_find_socket_byinode(struct net *net, struct inode *i)
 		    &unix_socket_table[i->i_ino & (UNIX_HASH_SIZE - 1)]) {
 		struct dentry *dentry = unix_sk(s)->dentry;
 
-		if (s->sk_net != net)
+		if (sock_net(s) != net)
 			continue;
 
 		if(dentry && dentry->d_inode == i)
@@ -654,7 +654,7 @@ static int unix_release(struct socket *sock)
 static int unix_autobind(struct socket *sock)
 {
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	struct unix_sock *u = unix_sk(sk);
 	static u32 ordernum = 1;
 	struct unix_address * addr;
@@ -758,7 +758,7 @@ fail:
 static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 {
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	struct unix_sock *u = unix_sk(sk);
 	struct sockaddr_un *sunaddr=(struct sockaddr_un *)uaddr;
 	struct dentry * dentry = NULL;
@@ -899,7 +899,7 @@ static int unix_dgram_connect(struct socket *sock, struct sockaddr *addr,
 			      int alen, int flags)
 {
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	struct sockaddr_un *sunaddr=(struct sockaddr_un*)addr;
 	struct sock *other;
 	unsigned hash;
@@ -996,7 +996,7 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 {
 	struct sockaddr_un *sunaddr=(struct sockaddr_un *)uaddr;
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	struct unix_sock *u = unix_sk(sk), *newu, *otheru;
 	struct sock *newsk = NULL;
 	struct sock *other = NULL;
@@ -1025,7 +1025,7 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 	err = -ENOMEM;
 
 	/* create new sock for complete connection */
-	newsk = unix_create1(sk->sk_net, NULL);
+	newsk = unix_create1(sock_net(sk), NULL);
 	if (newsk == NULL)
 		goto out;
 
@@ -1312,7 +1312,7 @@ static int unix_dgram_sendmsg(struct kiocb *kiocb, struct socket *sock,
 {
 	struct sock_iocb *siocb = kiocb_to_siocb(kiocb);
 	struct sock *sk = sock->sk;
-	struct net *net = sk->sk_net;
+	struct net *net = sock_net(sk);
 	struct unix_sock *u = unix_sk(sk);
 	struct sockaddr_un *sunaddr=msg->msg_name;
 	struct sock *other = NULL;
@@ -2022,7 +2022,7 @@ static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos)
 	struct sock *s;
 
 	for (s = first_unix_socket(&iter->i); s; s = next_unix_socket(&iter->i, s)) {
-		if (s->sk_net != iter->p.net)
+		if (sock_net(s) != iter->p.net)
 			continue;
 		if (off == pos)
 			return s;
@@ -2050,7 +2050,7 @@ static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 		sk = first_unix_socket(&iter->i);
 	else
 		sk = next_unix_socket(&iter->i, sk);
-	while (sk && (sk->sk_net != iter->p.net))
+	while (sk && (sock_net(sk) != iter->p.net))
 		sk = next_unix_socket(&iter->i, sk);
 	return sk;
 }
diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c
index 7a46ea73fe2d..6ba67c523c16 100644
--- a/net/x25/af_x25.c
+++ b/net/x25/af_x25.c
@@ -549,7 +549,7 @@ static struct sock *x25_make_new(struct sock *osk)
 	if (osk->sk_type != SOCK_SEQPACKET)
 		goto out;
 
-	if ((sk = x25_alloc_socket(osk->sk_net)) == NULL)
+	if ((sk = x25_alloc_socket(sock_net(osk))) == NULL)
 		goto out;
 
 	x25 = x25_sk(sk);
-- 
cgit v1.2.3


From 1218854afa6f659be90b748cf1bc7badee954a35 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 26 Mar 2008 02:36:06 +0900
Subject: [NET] NETNS: Omit seq_net_private->net without CONFIG_NET_NS.

Without CONFIG_NET_NS, no namespace other than &init_net exists,
no need to store net in seq_net_private.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 fs/proc/proc_net.c       |  6 +++---
 include/linux/seq_file.h |  7 +++++++
 net/core/neighbour.c     |  8 ++++----
 net/ipv4/fib_hash.c      |  5 ++---
 net/ipv4/fib_trie.c      | 13 ++++++-------
 net/ipv4/raw.c           |  4 ++--
 net/ipv4/route.c         | 29 +++++++++++++++--------------
 net/ipv6/addrconf.c      |  4 ++--
 net/ipv6/mcast.c         |  4 ++--
 net/netfilter/x_tables.c |  4 ++--
 net/netlink/af_netlink.c |  6 +++---
 net/unix/af_unix.c       | 10 +++++-----
 12 files changed, 53 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index 4caa5f774fb7..13cd7835d0df 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -44,7 +44,9 @@ int seq_open_net(struct inode *ino, struct file *f,
 		put_net(net);
 		return -ENOMEM;
 	}
+#ifdef CONFIG_NET_NS
 	p->net = net;
+#endif
 	return 0;
 }
 EXPORT_SYMBOL_GPL(seq_open_net);
@@ -52,12 +54,10 @@ EXPORT_SYMBOL_GPL(seq_open_net);
 int seq_release_net(struct inode *ino, struct file *f)
 {
 	struct seq_file *seq;
-	struct seq_net_private *p;
 
 	seq = f->private_data;
-	p = seq->private;
 
-	put_net(p->net);
+	put_net(seq_file_net(seq));
 	seq_release_private(ino, f);
 	return 0;
 }
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 67c2563961f3..d870a8253769 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -5,6 +5,7 @@
 #include <linux/types.h>
 #include <linux/string.h>
 #include <linux/mutex.h>
+#include <net/net_namespace.h>
 
 struct seq_operations;
 struct file;
@@ -64,7 +65,9 @@ extern struct list_head *seq_list_next(void *v, struct list_head *head,
 
 struct net;
 struct seq_net_private {
+#ifdef CONFIG_NET_NS
 	struct net *net;
+#endif
 };
 
 int seq_open_net(struct inode *, struct file *,
@@ -72,7 +75,11 @@ int seq_open_net(struct inode *, struct file *,
 int seq_release_net(struct inode *, struct file *);
 static inline struct net *seq_file_net(struct seq_file *seq)
 {
+#ifdef CONFIG_NET_NS
 	return ((struct seq_net_private *)seq->private)->net;
+#else
+	return &init_net;
+#endif
 }
 
 #endif
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 065fbac7ecd3..b8d491fb4b42 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -2145,7 +2145,7 @@ EXPORT_SYMBOL(__neigh_for_each_release);
 static struct neighbour *neigh_get_first(struct seq_file *seq)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net *net = state->p.net;
+	struct net *net = seq_file_net(seq);
 	struct neigh_table *tbl = state->tbl;
 	struct neighbour *n = NULL;
 	int bucket = state->bucket;
@@ -2186,7 +2186,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 					loff_t *pos)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net *net = state->p.net;
+	struct net *net = seq_file_net(seq);
 	struct neigh_table *tbl = state->tbl;
 
 	if (state->neigh_sub_iter) {
@@ -2246,7 +2246,7 @@ static struct neighbour *neigh_get_idx(struct seq_file *seq, loff_t *pos)
 static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net * net = state->p.net;
+	struct net *net = seq_file_net(seq);
 	struct neigh_table *tbl = state->tbl;
 	struct pneigh_entry *pn = NULL;
 	int bucket = state->bucket;
@@ -2269,7 +2269,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
 					    loff_t *pos)
 {
 	struct neigh_seq_state *state = seq->private;
-	struct net * net = state->p.net;
+	struct net *net = seq_file_net(seq);
 	struct neigh_table *tbl = state->tbl;
 
 	pn = pn->next;
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 8d58d85dfac6..02088deb0461 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -821,7 +821,7 @@ static struct fib_alias *fib_get_first(struct seq_file *seq)
 	struct fib_table *main_table;
 	struct fn_hash *table;
 
-	main_table = fib_get_table(iter->p.net, RT_TABLE_MAIN);
+	main_table = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN);
 	table = (struct fn_hash *)main_table->tb_data;
 
 	iter->bucket    = 0;
@@ -959,11 +959,10 @@ static struct fib_alias *fib_get_idx(struct seq_file *seq, loff_t pos)
 static void *fib_seq_start(struct seq_file *seq, loff_t *pos)
 	__acquires(fib_hash_lock)
 {
-	struct fib_iter_state *iter = seq->private;
 	void *v = NULL;
 
 	read_lock(&fib_hash_lock);
-	if (fib_get_table(iter->p.net, RT_TABLE_MAIN))
+	if (fib_get_table(seq_file_net(seq), RT_TABLE_MAIN))
 		v = *pos ? fib_get_idx(seq, *pos - 1) : SEQ_START_TOKEN;
 	return v;
 }
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index ce6cb34e28e1..9e491e70e855 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2279,9 +2279,10 @@ static const struct file_operations fib_triestat_fops = {
 	.release = fib_triestat_seq_release,
 };
 
-static struct node *fib_trie_get_idx(struct fib_trie_iter *iter, loff_t pos)
+static struct node *fib_trie_get_idx(struct seq_file *seq, loff_t pos)
 {
-	struct net *net = iter->p.net;
+	struct fib_trie_iter *iter = seq->private;
+	struct net *net = seq_file_net(seq);
 	loff_t idx = 0;
 	unsigned int h;
 
@@ -2309,16 +2310,14 @@ static struct node *fib_trie_get_idx(struct fib_trie_iter *iter, loff_t pos)
 static void *fib_trie_seq_start(struct seq_file *seq, loff_t *pos)
 	__acquires(RCU)
 {
-	struct fib_trie_iter *iter = seq->private;
-
 	rcu_read_lock();
-	return fib_trie_get_idx(iter, *pos);
+	return fib_trie_get_idx(seq, *pos);
 }
 
 static void *fib_trie_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	struct fib_trie_iter *iter = seq->private;
-	struct net *net = iter->p.net;
+	struct net *net = seq_file_net(seq);
 	struct fib_table *tb = iter->tb;
 	struct hlist_node *tb_node;
 	unsigned int h;
@@ -2513,7 +2512,7 @@ static void *fib_route_seq_start(struct seq_file *seq, loff_t *pos)
 	struct fib_table *tb;
 
 	rcu_read_lock();
-	tb = fib_get_table(iter->p.net, RT_TABLE_MAIN);
+	tb = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN);
 	if (!tb)
 		return NULL;
 
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index be19a4048d7c..25dc8b38cac3 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -856,7 +856,7 @@ static struct sock *raw_get_first(struct seq_file *seq)
 		struct hlist_node *node;
 
 		sk_for_each(sk, node, &state->h->ht[state->bucket])
-			if (sock_net(sk) == state->p.net)
+			if (sock_net(sk) == seq_file_net(seq))
 				goto found;
 	}
 	sk = NULL;
@@ -872,7 +872,7 @@ static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk)
 		sk = sk_next(sk);
 try_again:
 		;
-	} while (sk && sock_net(sk) != state->p.net);
+	} while (sk && sock_net(sk) != seq_file_net(seq));
 
 	if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
 		sk = sk_head(&state->h->ht[state->bucket]);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 194f5cca3121..eab8d75e5222 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -276,15 +276,16 @@ struct rt_cache_iter_state {
 	int genid;
 };
 
-static struct rtable *rt_cache_get_first(struct rt_cache_iter_state *st)
+static struct rtable *rt_cache_get_first(struct seq_file *seq)
 {
+	struct rt_cache_iter_state *st = seq->private;
 	struct rtable *r = NULL;
 
 	for (st->bucket = rt_hash_mask; st->bucket >= 0; --st->bucket) {
 		rcu_read_lock_bh();
 		r = rcu_dereference(rt_hash_table[st->bucket].chain);
 		while (r) {
-			if (dev_net(r->u.dst.dev) == st->p.net &&
+			if (dev_net(r->u.dst.dev) == seq_file_net(seq) &&
 			    r->rt_genid == st->genid)
 				return r;
 			r = rcu_dereference(r->u.dst.rt_next);
@@ -294,9 +295,10 @@ static struct rtable *rt_cache_get_first(struct rt_cache_iter_state *st)
 	return r;
 }
 
-static struct rtable *__rt_cache_get_next(struct rt_cache_iter_state *st,
+static struct rtable *__rt_cache_get_next(struct seq_file *seq,
 					  struct rtable *r)
 {
+	struct rt_cache_iter_state *st = seq->private;
 	r = r->u.dst.rt_next;
 	while (!r) {
 		rcu_read_unlock_bh();
@@ -308,11 +310,12 @@ static struct rtable *__rt_cache_get_next(struct rt_cache_iter_state *st,
 	return rcu_dereference(r);
 }
 
-static struct rtable *rt_cache_get_next(struct rt_cache_iter_state *st,
+static struct rtable *rt_cache_get_next(struct seq_file *seq,
 					struct rtable *r)
 {
-	while ((r = __rt_cache_get_next(st, r)) != NULL) {
-		if (dev_net(r->u.dst.dev) != st->p.net)
+	struct rt_cache_iter_state *st = seq->private;
+	while ((r = __rt_cache_get_next(seq, r)) != NULL) {
+		if (dev_net(r->u.dst.dev) != seq_file_net(seq))
 			continue;
 		if (r->rt_genid == st->genid)
 			break;
@@ -320,12 +323,12 @@ static struct rtable *rt_cache_get_next(struct rt_cache_iter_state *st,
 	return r;
 }
 
-static struct rtable *rt_cache_get_idx(struct rt_cache_iter_state *st, loff_t pos)
+static struct rtable *rt_cache_get_idx(struct seq_file *seq, loff_t pos)
 {
-	struct rtable *r = rt_cache_get_first(st);
+	struct rtable *r = rt_cache_get_first(seq);
 
 	if (r)
-		while (pos && (r = rt_cache_get_next(st, r)))
+		while (pos && (r = rt_cache_get_next(seq, r)))
 			--pos;
 	return pos ? NULL : r;
 }
@@ -333,9 +336,8 @@ static struct rtable *rt_cache_get_idx(struct rt_cache_iter_state *st, loff_t po
 static void *rt_cache_seq_start(struct seq_file *seq, loff_t *pos)
 {
 	struct rt_cache_iter_state *st = seq->private;
-
 	if (*pos)
-		return rt_cache_get_idx(st, *pos - 1);
+		return rt_cache_get_idx(seq, *pos - 1);
 	st->genid = atomic_read(&rt_genid);
 	return SEQ_START_TOKEN;
 }
@@ -343,12 +345,11 @@ static void *rt_cache_seq_start(struct seq_file *seq, loff_t *pos)
 static void *rt_cache_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	struct rtable *r;
-	struct rt_cache_iter_state *st = seq->private;
 
 	if (v == SEQ_START_TOKEN)
-		r = rt_cache_get_first(st);
+		r = rt_cache_get_first(seq);
 	else
-		r = rt_cache_get_next(st, v);
+		r = rt_cache_get_next(seq, v);
 	++*pos;
 	return r;
 }
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index f2c90f145cbb..ac5d4f4b6312 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -2766,7 +2766,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq)
 {
 	struct inet6_ifaddr *ifa = NULL;
 	struct if6_iter_state *state = seq->private;
-	struct net *net = state->p.net;
+	struct net *net = seq_file_net(seq);
 
 	for (state->bucket = 0; state->bucket < IN6_ADDR_HSIZE; ++state->bucket) {
 		ifa = inet6_addr_lst[state->bucket];
@@ -2782,7 +2782,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq)
 static struct inet6_ifaddr *if6_get_next(struct seq_file *seq, struct inet6_ifaddr *ifa)
 {
 	struct if6_iter_state *state = seq->private;
-	struct net *net = state->p.net;
+	struct net *net = seq_file_net(seq);
 
 	ifa = ifa->lst_next;
 try_again:
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 20a3d8e2f6c6..d810cff818cf 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -2355,7 +2355,7 @@ static inline struct ifmcaddr6 *igmp6_mc_get_first(struct seq_file *seq)
 {
 	struct ifmcaddr6 *im = NULL;
 	struct igmp6_mc_iter_state *state = igmp6_mc_seq_private(seq);
-	struct net *net = state->p.net;
+	struct net *net = seq_file_net(seq);
 
 	state->idev = NULL;
 	for_each_netdev(net, state->dev) {
@@ -2486,7 +2486,7 @@ static inline struct ip6_sf_list *igmp6_mcf_get_first(struct seq_file *seq)
 	struct ip6_sf_list *psf = NULL;
 	struct ifmcaddr6 *im = NULL;
 	struct igmp6_mcf_iter_state *state = igmp6_mcf_seq_private(seq);
-	struct net *net = state->p.net;
+	struct net *net = seq_file_net(seq);
 
 	state->idev = NULL;
 	state->im = NULL;
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index a6792089fcf9..0bd95680a494 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -727,7 +727,7 @@ struct xt_names_priv {
 static void *xt_table_seq_start(struct seq_file *seq, loff_t *pos)
 {
 	struct xt_names_priv *priv = seq->private;
-	struct net *net = priv->p.net;
+	struct net *net = seq_file_net(seq);
 	int af = priv->af;
 
 	mutex_lock(&xt[af].mutex);
@@ -737,7 +737,7 @@ static void *xt_table_seq_start(struct seq_file *seq, loff_t *pos)
 static void *xt_table_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	struct xt_names_priv *priv = seq->private;
-	struct net *net = priv->p.net;
+	struct net *net = seq_file_net(seq);
 	int af = priv->af;
 
 	return seq_list_next(v, &net->xt.tables[af], pos);
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 712a7bff8560..1d16d95dfaaf 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1758,7 +1758,7 @@ static struct sock *netlink_seq_socket_idx(struct seq_file *seq, loff_t pos)
 
 		for (j = 0; j <= hash->mask; j++) {
 			sk_for_each(s, node, &hash->table[j]) {
-				if (sock_net(s) != iter->p.net)
+				if (sock_net(s) != seq_file_net(seq))
 					continue;
 				if (off == pos) {
 					iter->link = i;
@@ -1794,7 +1794,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 	s = v;
 	do {
 		s = sk_next(s);
-	} while (s && (sock_net(s) != iter->p.net));
+	} while (s && sock_net(s) != seq_file_net(seq));
 	if (s)
 		return s;
 
@@ -1806,7 +1806,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 
 		for (; j <= hash->mask; j++) {
 			s = sk_head(&hash->table[j]);
-			while (s && sock_net(s) != iter->p.net)
+			while (s && sock_net(s) != seq_file_net(seq))
 				s = sk_next(s);
 			if (s) {
 				iter->link = i;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index cb9d0cb5f270..4a4793051bcb 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -2016,13 +2016,14 @@ struct unix_iter_state {
 	struct seq_net_private p;
 	int i;
 };
-static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos)
+static struct sock *unix_seq_idx(struct seq_file *seq, loff_t pos)
 {
+	struct unix_iter_state *iter = seq->private;
 	loff_t off = 0;
 	struct sock *s;
 
 	for (s = first_unix_socket(&iter->i); s; s = next_unix_socket(&iter->i, s)) {
-		if (sock_net(s) != iter->p.net)
+		if (sock_net(s) != seq_file_net(seq))
 			continue;
 		if (off == pos)
 			return s;
@@ -2035,9 +2036,8 @@ static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos)
 static void *unix_seq_start(struct seq_file *seq, loff_t *pos)
 	__acquires(unix_table_lock)
 {
-	struct unix_iter_state *iter = seq->private;
 	spin_lock(&unix_table_lock);
-	return *pos ? unix_seq_idx(iter, *pos - 1) : ((void *) 1);
+	return *pos ? unix_seq_idx(seq, *pos - 1) : ((void *) 1);
 }
 
 static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos)
@@ -2050,7 +2050,7 @@ static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 		sk = first_unix_socket(&iter->i);
 	else
 		sk = next_unix_socket(&iter->i, sk);
-	while (sk && (sock_net(sk) != iter->p.net))
+	while (sk && (sock_net(sk) != seq_file_net(seq)))
 		sk = next_unix_socket(&iter->i, sk);
 	return sk;
 }
-- 
cgit v1.2.3


From 57da52c1e62c6c13875e97de6c69d3156f8416da Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 26 Mar 2008 03:49:59 +0900
Subject: [NET] NETNS: Omit neigh_parms->net and pneigh_entry->net without
 CONFIG_NET_NS.

Introduce neigh_parms/pneigh_entry inlines: neigh_parms_net(), pneigh_net().
Without CONFIG_NET_NS, no namespace other than &init_net exists.
Let's explicitly define them to help compiler optimizations.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/neighbour.h | 25 +++++++++++++++++++++++++
 net/core/neighbour.c    | 26 ++++++++++++++++----------
 2 files changed, 41 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 062281872064..8bec0d69b270 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -38,7 +38,9 @@ struct neighbour;
 
 struct neigh_parms
 {
+#ifdef CONFIG_NET_NS
 	struct net *net;
+#endif
 	struct net_device *dev;
 	struct neigh_parms *next;
 	int	(*neigh_setup)(struct neighbour *);
@@ -131,7 +133,9 @@ struct neigh_ops
 struct pneigh_entry
 {
 	struct pneigh_entry	*next;
+#ifdef CONFIG_NET_NS
 	struct net		*net;
+#endif
 	struct net_device	*dev;
 	u8			flags;
 	u8			key[0];
@@ -213,6 +217,17 @@ extern struct neighbour 	*neigh_event_ns(struct neigh_table *tbl,
 
 extern struct neigh_parms	*neigh_parms_alloc(struct net_device *dev, struct neigh_table *tbl);
 extern void			neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms);
+
+static inline
+struct net			*neigh_parms_net(const struct neigh_parms *parms)
+{
+#ifdef CONFIG_NET_NS
+	return parms->net;
+#else
+	return &init_net;
+#endif
+}
+
 extern unsigned long		neigh_rand_reach_time(unsigned long base);
 
 extern void			pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
@@ -220,6 +235,16 @@ extern void			pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
 extern struct pneigh_entry	*pneigh_lookup(struct neigh_table *tbl, struct net *net, const void *key, struct net_device *dev, int creat);
 extern int			pneigh_delete(struct neigh_table *tbl, struct net *net, const void *key, struct net_device *dev);
 
+static inline
+struct net			*pneigh_net(const struct pneigh_entry *pneigh)
+{
+#ifdef CONFIG_NET_NS
+	return pneigh->net;
+#else
+	return &init_net;
+#endif
+}
+
 extern void neigh_app_ns(struct neighbour *n);
 extern void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void *), void *cookie);
 extern void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct neighbour *));
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index b8d491fb4b42..de654ea8a944 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -483,7 +483,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl,
 
 	for (n = tbl->phash_buckets[hash_val]; n; n = n->next) {
 		if (!memcmp(n->key, pkey, key_len) &&
-		    (n->net == net) &&
+		    (pneigh_net(n) == net) &&
 		    (n->dev == dev || !n->dev)) {
 			read_unlock_bh(&tbl->lock);
 			goto out;
@@ -500,7 +500,9 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl,
 	if (!n)
 		goto out;
 
+#ifdef CONFIG_NET_NS
 	n->net = hold_net(net);
+#endif
 	memcpy(n->key, pkey, key_len);
 	n->dev = dev;
 	if (dev)
@@ -540,14 +542,14 @@ int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey,
 	for (np = &tbl->phash_buckets[hash_val]; (n = *np) != NULL;
 	     np = &n->next) {
 		if (!memcmp(n->key, pkey, key_len) && n->dev == dev &&
-		    (n->net == net)) {
+		    (pneigh_net(n) == net)) {
 			*np = n->next;
 			write_unlock_bh(&tbl->lock);
 			if (tbl->pdestructor)
 				tbl->pdestructor(n);
 			if (n->dev)
 				dev_put(n->dev);
-			release_net(n->net);
+			release_net(pneigh_net(n));
 			kfree(n);
 			return 0;
 		}
@@ -570,7 +572,7 @@ static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev)
 					tbl->pdestructor(n);
 				if (n->dev)
 					dev_put(n->dev);
-				release_net(n->net);
+				release_net(pneigh_net(n));
 				kfree(n);
 				continue;
 			}
@@ -1284,7 +1286,7 @@ static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl,
 	struct neigh_parms *p;
 
 	for (p = &tbl->parms; p; p = p->next) {
-		if ((p->dev && p->dev->ifindex == ifindex && p->net == net) ||
+		if ((p->dev && p->dev->ifindex == ifindex && neigh_parms_net(p) == net) ||
 		    (!p->dev && !ifindex))
 			return p;
 	}
@@ -1318,7 +1320,9 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev,
 
 		dev_hold(dev);
 		p->dev = dev;
+#ifdef CONFIG_NET_NS
 		p->net = hold_net(net);
+#endif
 		p->sysctl_table = NULL;
 		write_lock_bh(&tbl->lock);
 		p->next		= tbl->parms.next;
@@ -1360,7 +1364,7 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms)
 
 static void neigh_parms_destroy(struct neigh_parms *parms)
 {
-	release_net(parms->net);
+	release_net(neigh_parms_net(parms));
 	kfree(parms);
 }
 
@@ -1371,7 +1375,9 @@ void neigh_table_init_no_netlink(struct neigh_table *tbl)
 	unsigned long now = jiffies;
 	unsigned long phsize;
 
+#ifdef CONFIG_NET_NS
 	tbl->parms.net = &init_net;
+#endif
 	atomic_set(&tbl->parms.refcnt, 1);
 	INIT_RCU_HEAD(&tbl->parms.rcu_head);
 	tbl->parms.reachable_time =
@@ -1958,7 +1964,7 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 			break;
 
 		for (nidx = 0, p = tbl->parms.next; p; p = p->next) {
-			if (net != p->net)
+			if (net != neigh_parms_net(p))
 				continue;
 
 			if (nidx++ < neigh_skip)
@@ -2254,7 +2260,7 @@ static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
 	state->flags |= NEIGH_SEQ_IS_PNEIGH;
 	for (bucket = 0; bucket <= PNEIGH_HASHMASK; bucket++) {
 		pn = tbl->phash_buckets[bucket];
-		while (pn && (pn->net != net))
+		while (pn && (pneigh_net(pn) != net))
 			pn = pn->next;
 		if (pn)
 			break;
@@ -2277,7 +2283,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
 		if (++state->bucket > PNEIGH_HASHMASK)
 			break;
 		pn = tbl->phash_buckets[state->bucket];
-		while (pn && (pn->net != net))
+		while (pn && (pneigh_net(pn) != net))
 			pn = pn->next;
 		if (pn)
 			break;
@@ -2740,7 +2746,7 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p,
 	neigh_path[NEIGH_CTL_PATH_PROTO].ctl_name = p_id;
 
 	t->sysctl_header =
-		register_net_sysctl_table(p->net, neigh_path, t->neigh_vars);
+		register_net_sysctl_table(neigh_parms_net(p), neigh_path, t->neigh_vars);
 	if (!t->sysctl_header)
 		goto free_procname;
 
-- 
cgit v1.2.3


From 878628fbf2589eb24357e42027d5f54b1dafd3c8 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 26 Mar 2008 03:57:35 +0900
Subject: [NET] NETNS: Omit namespace comparision without CONFIG_NET_NS.

Introduce an inline net_eq() to compare two namespaces.
Without CONFIG_NET_NS, since no namespace other than &init_net
exists, it is always 1.

We do not need to convert 1) inline vs inline and
2) inline vs &init_net comparisons.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/net_namespace.h | 12 ++++++++++++
 net/core/dev.c              |  2 +-
 net/core/neighbour.c        | 18 +++++++++---------
 net/ipv4/inet_hashtables.c  |  4 ++--
 net/ipv4/raw.c              |  2 +-
 net/ipv4/route.c            | 10 +++++-----
 net/ipv4/tcp_ipv4.c         | 12 ++++++------
 net/ipv4/udp.c              | 10 +++++-----
 net/ipv6/addrconf.c         | 12 ++++++------
 net/ipv6/inet6_hashtables.c |  2 +-
 net/ipv6/raw.c              |  2 +-
 net/ipv6/udp.c              |  2 +-
 net/netlink/af_netlink.c    |  8 ++++----
 net/unix/af_unix.c          |  4 ++--
 14 files changed, 56 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 923f2b8b9096..f8f3d1a5fc35 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -118,6 +118,12 @@ static inline void release_net(struct net *net)
 {
 	atomic_dec(&net->use_count);
 }
+
+static inline
+int net_eq(const struct net *net1, const struct net *net2)
+{
+	return net1 == net2;
+}
 #else
 static inline struct net *get_net(struct net *net)
 {
@@ -141,6 +147,12 @@ static inline struct net *maybe_get_net(struct net *net)
 {
 	return net;
 }
+
+static inline
+int net_eq(const struct net *net1, const struct net *net2)
+{
+	return 1;
+}
 #endif
 
 #define for_each_net(VAR)				\
diff --git a/net/core/dev.c b/net/core/dev.c
index 812534828914..75c3f7f4edd5 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4136,7 +4136,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char
 
 	/* Get out if there is nothing todo */
 	err = 0;
-	if (dev_net(dev) == net)
+	if (net_eq(dev_net(dev), net))
 		goto out;
 
 	/* Pick the destination device name, and ensure
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index de654ea8a944..857915a12c15 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -388,7 +388,7 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net,
 	hash_val = tbl->hash(pkey, NULL);
 	for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) {
 		if (!memcmp(n->primary_key, pkey, key_len) &&
-		    dev_net(n->dev) == net) {
+		    net_eq(dev_net(n->dev), net)) {
 			neigh_hold(n);
 			NEIGH_CACHE_STAT_INC(tbl, hits);
 			break;
@@ -483,7 +483,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl,
 
 	for (n = tbl->phash_buckets[hash_val]; n; n = n->next) {
 		if (!memcmp(n->key, pkey, key_len) &&
-		    (pneigh_net(n) == net) &&
+		    net_eq(pneigh_net(n), net) &&
 		    (n->dev == dev || !n->dev)) {
 			read_unlock_bh(&tbl->lock);
 			goto out;
@@ -542,7 +542,7 @@ int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey,
 	for (np = &tbl->phash_buckets[hash_val]; (n = *np) != NULL;
 	     np = &n->next) {
 		if (!memcmp(n->key, pkey, key_len) && n->dev == dev &&
-		    (pneigh_net(n) == net)) {
+		    net_eq(pneigh_net(n), net)) {
 			*np = n->next;
 			write_unlock_bh(&tbl->lock);
 			if (tbl->pdestructor)
@@ -1286,7 +1286,7 @@ static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl,
 	struct neigh_parms *p;
 
 	for (p = &tbl->parms; p; p = p->next) {
-		if ((p->dev && p->dev->ifindex == ifindex && neigh_parms_net(p) == net) ||
+		if ((p->dev && p->dev->ifindex == ifindex && net_eq(neigh_parms_net(p), net)) ||
 		    (!p->dev && !ifindex))
 			return p;
 	}
@@ -1964,7 +1964,7 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 			break;
 
 		for (nidx = 0, p = tbl->parms.next; p; p = p->next) {
-			if (net != neigh_parms_net(p))
+			if (!net_eq(neigh_parms_net(p), net))
 				continue;
 
 			if (nidx++ < neigh_skip)
@@ -2161,7 +2161,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq)
 		n = tbl->hash_buckets[bucket];
 
 		while (n) {
-			if (dev_net(n->dev) != net)
+			if (!net_eq(dev_net(n->dev), net))
 				goto next;
 			if (state->neigh_sub_iter) {
 				loff_t fakep = 0;
@@ -2204,7 +2204,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 
 	while (1) {
 		while (n) {
-			if (dev_net(n->dev) != net)
+			if (!net_eq(dev_net(n->dev), net))
 				goto next;
 			if (state->neigh_sub_iter) {
 				void *v = state->neigh_sub_iter(state, n, pos);
@@ -2260,7 +2260,7 @@ static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
 	state->flags |= NEIGH_SEQ_IS_PNEIGH;
 	for (bucket = 0; bucket <= PNEIGH_HASHMASK; bucket++) {
 		pn = tbl->phash_buckets[bucket];
-		while (pn && (pneigh_net(pn) != net))
+		while (pn && !net_eq(pneigh_net(pn), net))
 			pn = pn->next;
 		if (pn)
 			break;
@@ -2283,7 +2283,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
 		if (++state->bucket > PNEIGH_HASHMASK)
 			break;
 		pn = tbl->phash_buckets[state->bucket];
-		while (pn && (pneigh_net(pn) != net))
+		while (pn && !net_eq(pneigh_net(pn), net))
 			pn = pn->next;
 		if (pn)
 			break;
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 1064111e5b96..1b6ff513c75d 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -139,7 +139,7 @@ static struct sock *inet_lookup_listener_slow(struct net *net,
 	sk_for_each(sk, node, head) {
 		const struct inet_sock *inet = inet_sk(sk);
 
-		if (sock_net(sk) == net && inet->num == hnum &&
+		if (net_eq(sock_net(sk), net) && inet->num == hnum &&
 				!ipv6_only_sock(sk)) {
 			const __be32 rcv_saddr = inet->rcv_saddr;
 			int score = sk->sk_family == PF_INET ? 1 : 0;
@@ -182,7 +182,7 @@ struct sock *__inet_lookup_listener(struct net *net,
 		if (inet->num == hnum && !sk->sk_node.next &&
 		    (!inet->rcv_saddr || inet->rcv_saddr == daddr) &&
 		    (sk->sk_family == PF_INET || !ipv6_only_sock(sk)) &&
-		    !sk->sk_bound_dev_if && sock_net(sk) == net)
+		    !sk->sk_bound_dev_if && net_eq(sock_net(sk), net))
 			goto sherry_cache;
 		sk = inet_lookup_listener_slow(net, head, daddr, hnum, dif);
 	}
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 25dc8b38cac3..d965f0a39c84 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -117,7 +117,7 @@ static struct sock *__raw_v4_lookup(struct net *net, struct sock *sk,
 	sk_for_each_from(sk, node) {
 		struct inet_sock *inet = inet_sk(sk);
 
-		if (sock_net(sk) == net && inet->num == num 		&&
+		if (net_eq(sock_net(sk), net) && inet->num == num	&&
 		    !(inet->daddr && inet->daddr != raddr) 		&&
 		    !(inet->rcv_saddr && inet->rcv_saddr != laddr)	&&
 		    !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif))
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index eab8d75e5222..230716c2dfe0 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1196,7 +1196,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 				    rth->fl.oif != ikeys[k] ||
 				    rth->fl.iif != 0 ||
 				    rth->rt_genid != atomic_read(&rt_genid) ||
-				    dev_net(rth->u.dst.dev) != net) {
+				    !net_eq(dev_net(rth->u.dst.dev), net)) {
 					rthp = &rth->u.dst.rt_next;
 					continue;
 				}
@@ -1455,7 +1455,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
 			    rth->rt_src  == iph->saddr &&
 			    rth->fl.iif == 0 &&
 			    !(dst_metric_locked(&rth->u.dst, RTAX_MTU)) &&
-			    dev_net(rth->u.dst.dev) == net &&
+			    net_eq(dev_net(rth->u.dst.dev), net) &&
 			    rth->rt_genid == atomic_read(&rt_genid)) {
 				unsigned short mtu = new_mtu;
 
@@ -2085,7 +2085,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 		    rth->fl.oif == 0 &&
 		    rth->fl.mark == skb->mark &&
 		    rth->fl.fl4_tos == tos &&
-		    dev_net(rth->u.dst.dev) == net &&
+		    net_eq(dev_net(rth->u.dst.dev), net) &&
 		    rth->rt_genid == atomic_read(&rt_genid)) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(in_hit);
@@ -2487,7 +2487,7 @@ int __ip_route_output_key(struct net *net, struct rtable **rp,
 		    rth->fl.mark == flp->mark &&
 		    !((rth->fl.fl4_tos ^ flp->fl4_tos) &
 			    (IPTOS_RT_MASK | RTO_ONLINK)) &&
-		    dev_net(rth->u.dst.dev) == net &&
+		    net_eq(dev_net(rth->u.dst.dev), net) &&
 		    rth->rt_genid == atomic_read(&rt_genid)) {
 			dst_use(&rth->u.dst, jiffies);
 			RT_CACHE_STAT_INC(out_hit);
@@ -2796,7 +2796,7 @@ int ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb)
 		rcu_read_lock_bh();
 		for (rt = rcu_dereference(rt_hash_table[h].chain), idx = 0; rt;
 		     rt = rcu_dereference(rt->u.dst.rt_next), idx++) {
-			if (dev_net(rt->u.dst.dev) != net || idx < s_idx)
+			if (!net_eq(dev_net(rt->u.dst.dev), net) || idx < s_idx)
 				continue;
 			if (rt->rt_genid != atomic_read(&rt_genid))
 				continue;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 46847e600a46..2a5881c81778 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1974,7 +1974,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur)
 		while (1) {
 			while (req) {
 				if (req->rsk_ops->family == st->family &&
-				    sock_net(req->sk) == net) {
+				    net_eq(sock_net(req->sk), net)) {
 					cur = req;
 					goto out;
 				}
@@ -1998,7 +1998,7 @@ get_req:
 	}
 get_sk:
 	sk_for_each_from(sk, node) {
-		if (sk->sk_family == st->family && sock_net(sk) == net) {
+		if (sk->sk_family == st->family && net_eq(sock_net(sk), net)) {
 			cur = sk;
 			goto out;
 		}
@@ -2049,7 +2049,7 @@ static void *established_get_first(struct seq_file *seq)
 		read_lock_bh(lock);
 		sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) {
 			if (sk->sk_family != st->family ||
-			    sock_net(sk) != net) {
+			    !net_eq(sock_net(sk), net)) {
 				continue;
 			}
 			rc = sk;
@@ -2059,7 +2059,7 @@ static void *established_get_first(struct seq_file *seq)
 		inet_twsk_for_each(tw, node,
 				   &tcp_hashinfo.ehash[st->bucket].twchain) {
 			if (tw->tw_family != st->family ||
-			    twsk_net(tw) != net) {
+			    !net_eq(twsk_net(tw), net)) {
 				continue;
 			}
 			rc = tw;
@@ -2086,7 +2086,7 @@ static void *established_get_next(struct seq_file *seq, void *cur)
 		tw = cur;
 		tw = tw_next(tw);
 get_tw:
-		while (tw && (tw->tw_family != st->family || twsk_net(tw) != net)) {
+		while (tw && (tw->tw_family != st->family || !net_eq(twsk_net(tw), net))) {
 			tw = tw_next(tw);
 		}
 		if (tw) {
@@ -2107,7 +2107,7 @@ get_tw:
 		sk = sk_next(sk);
 
 	sk_for_each_from(sk, node) {
-		if (sk->sk_family == st->family && sock_net(sk) == net)
+		if (sk->sk_family == st->family && net_eq(sock_net(sk), net))
 			goto found;
 	}
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 76d52d37d6ac..80007c79f12f 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -137,7 +137,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
 	struct hlist_node *node;
 
 	sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)])
-		if (sock_net(sk) == net && sk->sk_hash == num)
+		if (net_eq(sock_net(sk), net) && sk->sk_hash == num)
 			return 1;
 	return 0;
 }
@@ -218,7 +218,7 @@ gotit:
 		sk_for_each(sk2, node, head)
 			if (sk2->sk_hash == snum                             &&
 			    sk2 != sk                                        &&
-			    sock_net(sk2) == net				     &&
+			    net_eq(sock_net(sk2), net)			     &&
 			    (!sk2->sk_reuse        || !sk->sk_reuse)         &&
 			    (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if
 			     || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
@@ -269,7 +269,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
 	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
-		if (sock_net(sk) == net && sk->sk_hash == hnum &&
+		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
 				!ipv6_only_sock(sk)) {
 			int score = (sk->sk_family == PF_INET ? 1 : 0);
 			if (inet->rcv_saddr) {
@@ -1511,7 +1511,7 @@ static struct sock *udp_get_first(struct seq_file *seq)
 	for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) {
 		struct hlist_node *node;
 		sk_for_each(sk, node, state->hashtable + state->bucket) {
-			if (sock_net(sk) != net)
+			if (!net_eq(sock_net(sk), net))
 				continue;
 			if (sk->sk_family == state->family)
 				goto found;
@@ -1531,7 +1531,7 @@ static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk)
 		sk = sk_next(sk);
 try_again:
 		;
-	} while (sk && (sock_net(sk) != net || sk->sk_family != state->family));
+	} while (sk && (!net_eq(sock_net(sk), net) || sk->sk_family != state->family));
 
 	if (!sk && ++state->bucket < UDP_HTABLE_SIZE) {
 		sk = sk_head(state->hashtable + state->bucket);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index ac5d4f4b6312..5ab9973571ef 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1217,7 +1217,7 @@ int ipv6_chk_addr(struct net *net, struct in6_addr *addr,
 
 	read_lock_bh(&addrconf_hash_lock);
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
-		if (dev_net(ifp->idev->dev) != net)
+		if (!net_eq(dev_net(ifp->idev->dev), net))
 			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr) &&
 		    !(ifp->flags&IFA_F_TENTATIVE)) {
@@ -1239,7 +1239,7 @@ int ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr,
 	u8 hash = ipv6_addr_hash(addr);
 
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
-		if (dev_net(ifp->idev->dev) != net)
+		if (!net_eq(dev_net(ifp->idev->dev), net))
 			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr)) {
 			if (dev == NULL || ifp->idev->dev == dev)
@@ -1257,7 +1257,7 @@ struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr,
 
 	read_lock_bh(&addrconf_hash_lock);
 	for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) {
-		if (dev_net(ifp->idev->dev) != net)
+		if (!net_eq(dev_net(ifp->idev->dev), net))
 			continue;
 		if (ipv6_addr_equal(&ifp->addr, addr)) {
 			if (dev == NULL || ifp->idev->dev == dev ||
@@ -2771,7 +2771,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq)
 	for (state->bucket = 0; state->bucket < IN6_ADDR_HSIZE; ++state->bucket) {
 		ifa = inet6_addr_lst[state->bucket];
 
-		while (ifa && dev_net(ifa->idev->dev) != net)
+		while (ifa && !net_eq(dev_net(ifa->idev->dev), net))
 			ifa = ifa->lst_next;
 		if (ifa)
 			break;
@@ -2787,7 +2787,7 @@ static struct inet6_ifaddr *if6_get_next(struct seq_file *seq, struct inet6_ifad
 	ifa = ifa->lst_next;
 try_again:
 	if (ifa) {
-		if (dev_net(ifa->idev->dev) != net) {
+		if (!net_eq(dev_net(ifa->idev->dev), net)) {
 			ifa = ifa->lst_next;
 			goto try_again;
 		}
@@ -2905,7 +2905,7 @@ int ipv6_chk_home_addr(struct net *net, struct in6_addr *addr)
 	u8 hash = ipv6_addr_hash(addr);
 	read_lock_bh(&addrconf_hash_lock);
 	for (ifp = inet6_addr_lst[hash]; ifp; ifp = ifp->lst_next) {
-		if (dev_net(ifp->idev->dev) != net)
+		if (!net_eq(dev_net(ifp->idev->dev), net))
 			continue;
 		if (ipv6_addr_cmp(&ifp->addr, addr) == 0 &&
 		    (ifp->flags & IFA_F_HOMEADDRESS)) {
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 21c467675412..340c7d42b83a 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -105,7 +105,7 @@ struct sock *inet6_lookup_listener(struct net *net,
 
 	read_lock(&hashinfo->lhash_lock);
 	sk_for_each(sk, node, &hashinfo->listening_hash[inet_lhashfn(hnum)]) {
-		if (sock_net(sk) == net && inet_sk(sk)->num == hnum &&
+		if (net_eq(sock_net(sk), net) && inet_sk(sk)->num == hnum &&
 				sk->sk_family == PF_INET6) {
 			const struct ipv6_pinfo *np = inet6_sk(sk);
 
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 12c7a1560977..830da4603697 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -76,7 +76,7 @@ static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk,
 		if (inet_sk(sk)->num == num) {
 			struct ipv6_pinfo *np = inet6_sk(sk);
 
-			if (sock_net(sk) != net)
+			if (!net_eq(sock_net(sk), net))
 				continue;
 
 			if (!ipv6_addr_any(&np->daddr) &&
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index db266ff297e5..aacbc82ecf0f 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -70,7 +70,7 @@ static struct sock *__udp6_lib_lookup(struct net *net,
 	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
-		if (sock_net(sk) == net && sk->sk_hash == hnum &&
+		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
 				sk->sk_family == PF_INET6) {
 			struct ipv6_pinfo *np = inet6_sk(sk);
 			int score = 0;
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 1d16d95dfaaf..36f75d873898 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -228,7 +228,7 @@ static inline struct sock *netlink_lookup(struct net *net, int protocol,
 	read_lock(&nl_table_lock);
 	head = nl_pid_hashfn(hash, pid);
 	sk_for_each(sk, node, head) {
-		if (sock_net(sk) == net && (nlk_sk(sk)->pid == pid)) {
+		if (net_eq(sock_net(sk), net) && (nlk_sk(sk)->pid == pid)) {
 			sock_hold(sk);
 			goto found;
 		}
@@ -348,7 +348,7 @@ static int netlink_insert(struct sock *sk, struct net *net, u32 pid)
 	head = nl_pid_hashfn(hash, pid);
 	len = 0;
 	sk_for_each(osk, node, head) {
-		if (sock_net(osk) == net && (nlk_sk(osk)->pid == pid))
+		if (net_eq(sock_net(osk), net) && (nlk_sk(osk)->pid == pid))
 			break;
 		len++;
 	}
@@ -532,7 +532,7 @@ retry:
 	netlink_table_grab();
 	head = nl_pid_hashfn(hash, pid);
 	sk_for_each(osk, node, head) {
-		if (sock_net(osk) != net)
+		if (!net_eq(sock_net(osk), net))
 			continue;
 		if (nlk_sk(osk)->pid == pid) {
 			/* Bind collision, search negative pid values. */
@@ -962,7 +962,7 @@ static inline int do_one_broadcast(struct sock *sk,
 	    !test_bit(p->group - 1, nlk->groups))
 		goto out;
 
-	if (sock_net(sk) != p->net)
+	if (!net_eq(sock_net(sk), p->net))
 		goto out;
 
 	if (p->failure) {
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 4a4793051bcb..50bbf6bb1a22 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -252,7 +252,7 @@ static struct sock *__unix_find_socket_byname(struct net *net,
 	sk_for_each(s, node, &unix_socket_table[hash ^ type]) {
 		struct unix_sock *u = unix_sk(s);
 
-		if (sock_net(s) != net)
+		if (!net_eq(sock_net(s), net))
 			continue;
 
 		if (u->addr->len == len &&
@@ -289,7 +289,7 @@ static struct sock *unix_find_socket_byinode(struct net *net, struct inode *i)
 		    &unix_socket_table[i->i_ino & (UNIX_HASH_SIZE - 1)]) {
 		struct dentry *dentry = unix_sk(s)->dentry;
 
-		if (sock_net(s) != net)
+		if (!net_eq(sock_net(s), net))
 			continue;
 
 		if(dentry && dentry->d_inode == i)
-- 
cgit v1.2.3


From 5d2cdcd4e85c5187db30a6b29f79fbbe59f39f78 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 20 Mar 2008 15:06:41 +0200
Subject: mac80211: get a TKIP phase key from skb

This patch makes mac80211 able to compute a TKIP key from an skb.
The requested key can be a phase 1 or a phase 2 key.
This is useful for drivers who need to provide tkip key to their
HW to enable HW encryption.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 30 ++++++++++++++++++++++++++++
 net/mac80211/tkip.c    | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 83 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5ab6a350ee6d..2a134582fc16 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -643,6 +643,21 @@ enum sta_notify_cmd {
 	STA_NOTIFY_ADD, STA_NOTIFY_REMOVE
 };
 
+/**
+ * enum ieee80211_tkip_key_type - get tkip key
+ *
+ * Used by drivers which need to get a tkip key for skb. Some drivers need a
+ * phase 1 key, others need a phase 2 key. A single function allows the driver
+ * to get the key, this enum indicates what type of key is required.
+ *
+ * @IEEE80211_TKIP_P1_KEY: the driver needs a phase 1 key
+ * @IEEE80211_TKIP_P2_KEY: the driver needs a phase 2 key
+ */
+enum ieee80211_tkip_key_type {
+	IEEE80211_TKIP_P1_KEY,
+	IEEE80211_TKIP_P2_KEY,
+};
+
 /**
  * enum ieee80211_hw_flags - hardware flags
  *
@@ -1471,6 +1486,21 @@ int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb);
  */
 int ieee80211_get_hdrlen(u16 fc);
 
+/**
+ * ieee80211_get_tkip_key - get a TKIP rc4 for skb
+ *
+ * This function computes a TKIP rc4 key for an skb. It computes
+ * a phase 1 key if needed (iv16 wraps around). This function is to
+ * be used by drivers which can do HW encryption but need to compute
+ * to phase 1/2 key in SW.
+ *
+ * @keyconf: the parameter passed with the set key
+ * @skb: the skb for which the key is needed
+ * @rc4key: a buffer to which the key will be written
+ */
+void ieee80211_get_tkip_key(struct ieee80211_key_conf *keyconf,
+				struct sk_buff *skb,
+				enum ieee80211_tkip_key_type type, u8 *key);
 /**
  * ieee80211_wake_queue - wake specific queue
  * @hw: pointer as obtained from ieee80211_alloc_hw().
diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c
index 3abe194e4d55..5c36b2df3faf 100644
--- a/net/mac80211/tkip.c
+++ b/net/mac80211/tkip.c
@@ -214,6 +214,59 @@ void ieee80211_tkip_gen_rc4key(struct ieee80211_key *key, u8 *ta,
 			   key->u.tkip.iv16, rc4key);
 }
 
+void ieee80211_get_tkip_key(struct ieee80211_key_conf *keyconf,
+			struct sk_buff *skb, enum ieee80211_tkip_key_type type,
+			u8 *outkey)
+{
+	struct ieee80211_key *key = (struct ieee80211_key *)
+			container_of(keyconf, struct ieee80211_key, conf);
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
+	u8 *data = (u8 *) hdr;
+	u16 fc = le16_to_cpu(hdr->frame_control);
+	int hdr_len = ieee80211_get_hdrlen(fc);
+	u8 *ta = hdr->addr2;
+	u16 iv16;
+	u32 iv32;
+
+	iv16 = data[hdr_len] << 8;
+	iv16 += data[hdr_len + 2];
+	iv32 = data[hdr_len + 4] +
+		(data[hdr_len + 5] >> 8) +
+		(data[hdr_len + 6] >> 16) +
+		(data[hdr_len + 7] >> 24);
+
+#ifdef CONFIG_TKIP_DEBUG
+	printk(KERN_DEBUG "TKIP encrypt: iv16 = 0x%04x, iv32 = 0x%08x\n",
+			iv16, iv32);
+
+	if (iv32 != key->u.tkip.iv32) {
+		printk(KERN_DEBUG "skb: iv32 = 0x%08x key: iv32 = 0x%08x\n",
+			iv32, key->u.tkip.iv32);
+		printk(KERN_DEBUG "Wrap around of iv16 in the middle of a "
+			"fragmented packet\n");
+	}
+#endif /* CONFIG_TKIP_DEBUG */
+
+	/* Update the p1k only when the iv16 in the packet wraps around, this
+	 * might occur after the wrap around of iv16 in the key in case of
+	 * fragmented packets. */
+	if (iv16 == 0 || !key->u.tkip.tx_initialized) {
+		/* IV16 wrapped around - perform TKIP phase 1 */
+		tkip_mixing_phase1(ta, &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY],
+			iv32, key->u.tkip.p1k);
+		key->u.tkip.tx_initialized = 1;
+	}
+
+	if (type == IEEE80211_TKIP_P1_KEY) {
+		memcpy(outkey, key->u.tkip.p1k, sizeof(u16) * 5);
+		return;
+	}
+
+	tkip_mixing_phase2(key->u.tkip.p1k,
+		&key->conf.key[ALG_TKIP_TEMP_ENCR_KEY],	iv16, outkey);
+}
+EXPORT_SYMBOL(ieee80211_get_tkip_key);
+
 /* Encrypt packet payload with TKIP using @key. @pos is a pointer to the
  * beginning of the buffer containing payload. This payload must include
  * headroom of eight octets for IV and Ext. IV and taildroom of four octets
-- 
cgit v1.2.3


From 9ae4fda332df616ef47d5bb710c39681641d4303 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 20 Mar 2008 15:06:42 +0200
Subject: mac80211: allows driver to request a Phase 1 RX key

This patch makes mac80211 able to send a phase1 key for TKIP
decryption.
This is needed for drivers that don't do the rekeying by themselves
(i.e. iwlwifi). Upon IV16 wrap around, the packet is decrypted in SW,
if decryption is ok, mac80211 calls to update_tkip_key  with a new
phase 1 RX key.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 17 +++++++++++++++++
 net/mac80211/tkip.c    | 15 ++++++++++++++-
 net/mac80211/tkip.h    |  2 +-
 net/mac80211/wpa.c     |  2 +-
 4 files changed, 33 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 2a134582fc16..48428a6b9109 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -827,6 +827,16 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr)
  * parameter is guaranteed to be valid until another call to set_key()
  * removes it, but it can only be used as a cookie to differentiate
  * keys.
+ *
+ * In TKIP some HW need to be provided a phase 1 key, for RX decryption
+ * acceleration (i.e. iwlwifi). Those drivers should provide update_tkip_key
+ * handler.
+ * The update_tkip_key() call updates the driver with the new phase 1 key.
+ * This happens everytime the iv16 wraps around (every 65536 packets). The
+ * set_key() call will happen only once for each key (unless the AP did
+ * rekeying), it will not include a valid phase 1 key. The valid phase 1 key is
+ * provided by udpate_tkip_key only. The trigger that makes mac80211 call this
+ * handler is software decryption with wrap around of iv16.
  */
 
 /**
@@ -1003,6 +1013,10 @@ enum ieee80211_ampdu_mlme_action {
  *	and remove_interface calls, i.e. while the interface with the
  *	given local_address is enabled.
  *
+ * @update_tkip_key: See the section "Hardware crypto acceleration"
+ * 	This callback will be called in the context of Rx. Called for drivers
+ * 	which set IEEE80211_KEY_FLAG_TKIP_REQ_RX_P1_KEY.
+ *
  * @hw_scan: Ask the hardware to service the scan request, no need to start
  *	the scan state machine in stack. The scan must honour the channel
  *	configuration done by the regulatory agent in the wiphy's registered
@@ -1094,6 +1108,9 @@ struct ieee80211_ops {
 	int (*set_key)(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 		       const u8 *local_address, const u8 *address,
 		       struct ieee80211_key_conf *key);
+	void (*update_tkip_key)(struct ieee80211_hw *hw,
+			struct ieee80211_key_conf *conf, const u8 *address,
+			u32 iv32, u16 *phase1key);
 	int (*hw_scan)(struct ieee80211_hw *hw, u8 *ssid, size_t len);
 	int (*get_stats)(struct ieee80211_hw *hw,
 			 struct ieee80211_low_level_stats *stats);
diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c
index 5c36b2df3faf..45d59f19c29f 100644
--- a/net/mac80211/tkip.c
+++ b/net/mac80211/tkip.c
@@ -291,7 +291,7 @@ void ieee80211_tkip_encrypt_data(struct crypto_blkcipher *tfm,
 int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm,
 				struct ieee80211_key *key,
 				u8 *payload, size_t payload_len, u8 *ta,
-				int only_iv, int queue,
+				u8 *ra, int only_iv, int queue,
 				u32 *out_iv32, u16 *out_iv16)
 {
 	u32 iv32;
@@ -368,6 +368,19 @@ int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm,
 			printk("\n");
 		}
 #endif /* CONFIG_TKIP_DEBUG */
+		if (key->local->ops->update_tkip_key &&
+			key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) {
+			u8 bcast[ETH_ALEN] =
+				{0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
+			u8 *sta_addr = key->sta->addr;
+
+			if (is_multicast_ether_addr(ra))
+				sta_addr = bcast;
+
+			key->local->ops->update_tkip_key(
+				local_to_hw(key->local), &key->conf,
+				sta_addr, iv32, key->u.tkip.p1k_rx[queue]);
+		}
 	}
 
 	tkip_mixing_phase2(key->u.tkip.p1k_rx[queue],
diff --git a/net/mac80211/tkip.h b/net/mac80211/tkip.h
index 73d8ef2a93b0..ffaee3253e19 100644
--- a/net/mac80211/tkip.h
+++ b/net/mac80211/tkip.h
@@ -31,7 +31,7 @@ enum {
 int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm,
 				struct ieee80211_key *key,
 				u8 *payload, size_t payload_len, u8 *ta,
-				int only_iv, int queue,
+				u8 *ra, int only_iv, int queue,
 				u32 *out_iv32, u16 *out_iv16);
 
 #endif /* TKIP_H */
diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c
index df0b7341efc8..45709ada8fee 100644
--- a/net/mac80211/wpa.c
+++ b/net/mac80211/wpa.c
@@ -312,7 +312,7 @@ ieee80211_crypto_tkip_decrypt(struct ieee80211_rx_data *rx)
 	res = ieee80211_tkip_decrypt_data(rx->local->wep_rx_tfm,
 					  key, skb->data + hdrlen,
 					  skb->len - hdrlen, rx->sta->addr,
-					  hwaccel, rx->queue,
+					  hdr->addr1, hwaccel, rx->queue,
 					  &rx->tkip_iv32,
 					  &rx->tkip_iv16);
 	if (res != TKIP_DECRYPT_OK || wpa_test) {
-- 
cgit v1.2.3


From 906c730a2db950b7bce4ef17d65399acd791c360 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sun, 16 Mar 2008 18:34:33 +0100
Subject: wireless: add wiphy channel freq to channel struct lookup helper

Add ieee80211_get_channel() which gets you a channel struct for a
specific wiphy if that channel is present in that wiphy.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/wireless.h |  6 ++++++
 net/wireless/util.c    | 23 +++++++++++++++++++++++
 2 files changed, 29 insertions(+)

(limited to 'include')

diff --git a/include/net/wireless.h b/include/net/wireless.h
index c7f805ee5545..f4b77ab66bae 100644
--- a/include/net/wireless.h
+++ b/include/net/wireless.h
@@ -304,4 +304,10 @@ extern int ieee80211_channel_to_frequency(int chan);
  */
 extern int ieee80211_frequency_to_channel(int freq);
 
+/**
+ * ieee80211_get_channel - get channel struct from wiphy for specified frequency
+ */
+extern struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy,
+						       int freq);
+
 #endif /* __NET_WIRELESS_H */
diff --git a/net/wireless/util.c b/net/wireless/util.c
index 77336c22fcf2..f3e623df3515 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -33,6 +33,29 @@ int ieee80211_frequency_to_channel(int freq)
 }
 EXPORT_SYMBOL(ieee80211_frequency_to_channel);
 
+struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy,
+						int freq)
+{
+	enum ieee80211_band band;
+	struct ieee80211_supported_band *sband;
+	int i;
+
+	for (band = 0; band < IEEE80211_NUM_BANDS; band++) {
+		sband = wiphy->bands[band];
+
+		if (!sband)
+			continue;
+
+		for (i = 0; i < sband->n_channels; i++) {
+			if (sband->channels[i].center_freq == freq)
+				return &sband->channels[i];
+		}
+	}
+
+	return NULL;
+}
+EXPORT_SYMBOL(ieee80211_get_channel);
+
 static void set_mandatory_flags_band(struct ieee80211_supported_band *sband,
 				     enum ieee80211_band band)
 {
-- 
cgit v1.2.3


From ef27559b70bd5312dfcbeab3b9ab0296206413c4 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:07:38 -0700
Subject: [NETFILTER]: nf_conntrack: fix NF_CT_TUPLE_DUMP for IPv4

NF_CT_TUPLE_DUMP prints IPv4 addresses as IPv6, fix this and use printk
(guarded by #ifdef DEBUG) directly instead of pr_debug since the tuple
is usually printed at the end of line and we don't want to include a
log-level.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack_tuple.h | 38 ++++++++++++++++++++++++++----
 1 file changed, 33 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h
index e69ab2e87597..d9b53dd601ad 100644
--- a/include/net/netfilter/nf_conntrack_tuple.h
+++ b/include/net/netfilter/nf_conntrack_tuple.h
@@ -113,11 +113,39 @@ struct nf_conntrack_tuple_mask
 
 #ifdef __KERNEL__
 
-#define NF_CT_DUMP_TUPLE(tp)						     \
-pr_debug("tuple %p: %u %u " NIP6_FMT " %hu -> " NIP6_FMT " %hu\n",	     \
-	 (tp), (tp)->src.l3num, (tp)->dst.protonum,			     \
-	 NIP6(*(struct in6_addr *)(tp)->src.u3.all), ntohs((tp)->src.u.all), \
-	 NIP6(*(struct in6_addr *)(tp)->dst.u3.all), ntohs((tp)->dst.u.all))
+static inline void nf_ct_dump_tuple_ip(const struct nf_conntrack_tuple *t)
+{
+#ifdef DEBUG
+	printk("tuple %p: %u " NIPQUAD_FMT ":%hu -> " NIPQUAD_FMT ":%hu\n",
+	       t, t->dst.protonum,
+	       NIPQUAD(t->src.u3.ip), ntohs(t->src.u.all),
+	       NIPQUAD(t->dst.u3.ip), ntohs(t->dst.u.all));
+#endif
+}
+
+static inline void nf_ct_dump_tuple_ipv6(const struct nf_conntrack_tuple *t)
+{
+#ifdef DEBUG
+	printk("tuple %p: %u " NIP6_FMT " %hu -> " NIP6_FMT " %hu\n",
+	       t, t->dst.protonum,
+	       NIP6(*(struct in6_addr *)t->src.u3.all), ntohs(t->src.u.all),
+	       NIP6(*(struct in6_addr *)t->dst.u3.all), ntohs(t->dst.u.all));
+#endif
+}
+
+static inline void nf_ct_dump_tuple(const struct nf_conntrack_tuple *t)
+{
+	switch (t->src.l3num) {
+	case AF_INET:
+		nf_ct_dump_tuple_ip(t);
+		break;
+	case AF_INET6:
+		nf_ct_dump_tuple_ipv6(t);
+		break;
+	}
+}
+
+#define NF_CT_DUMP_TUPLE(tp)	nf_ct_dump_tuple(tp)
 
 /* If we're the first tuple, it's the original dir. */
 #define NF_CT_DIRECTION(h)						\
-- 
cgit v1.2.3


From 1d9d752259ab2e16b99f84b7e7bd4a30c884c289 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:07:58 -0700
Subject: [NETFILTER]: nf_conntrack_expect: constify nf_ct_expect_init
 arguments

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack_expect.h | 6 +++---
 net/netfilter/nf_conntrack_expect.c         | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index cb608a1b44e5..f1bdcb4f3f2a 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -75,9 +75,9 @@ void nf_ct_unexpect_related(struct nf_conntrack_expect *exp);
    nf_ct_expect_related.  You will have to call put afterwards. */
 struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me);
 void nf_ct_expect_init(struct nf_conntrack_expect *, int,
-		       union nf_inet_addr *,
-		       union nf_inet_addr *,
-		       u_int8_t, __be16 *, __be16 *);
+		       const union nf_inet_addr *,
+		       const union nf_inet_addr *,
+		       u_int8_t, const __be16 *, const __be16 *);
 void nf_ct_expect_put(struct nf_conntrack_expect *exp);
 int nf_ct_expect_related(struct nf_conntrack_expect *expect);
 
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index 684ec9c1ad38..740d94d30091 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -229,9 +229,9 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me)
 EXPORT_SYMBOL_GPL(nf_ct_expect_alloc);
 
 void nf_ct_expect_init(struct nf_conntrack_expect *exp, int family,
-		       union nf_inet_addr *saddr,
-		       union nf_inet_addr *daddr,
-		       u_int8_t proto, __be16 *src, __be16 *dst)
+		       const union nf_inet_addr *saddr,
+		       const union nf_inet_addr *daddr,
+		       u_int8_t proto, const __be16 *src, const __be16 *dst)
 {
 	int len;
 
-- 
cgit v1.2.3


From 359b9ab614aba71c2c3bc047efbd6d12dd4a2b9e Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:08:37 -0700
Subject: [NETFILTER]: nf_conntrack_expect: support inactive expectations

This is useful for the SIP helper and signalling expectations.
We don't want to create a full-blown expectation with a wildcard
as source based on a single UDP packet, but need to know the
final port anyways. With inactive expectations we can register
the expectation and reserve the tuple, but wait for confirmation
from the registrar before activating it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack_expect.h |  3 ++-
 net/netfilter/nf_conntrack_expect.c         | 25 +++++++++++++++++++++----
 2 files changed, 23 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index f1bdcb4f3f2a..47c28dd07896 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -53,7 +53,8 @@ struct nf_conntrack_expect
 	struct rcu_head rcu;
 };
 
-#define NF_CT_EXPECT_PERMANENT 0x1
+#define NF_CT_EXPECT_PERMANENT	0x1
+#define NF_CT_EXPECT_INACTIVE	0x2
 
 int nf_conntrack_expect_init(void);
 void nf_conntrack_expect_fini(void);
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index 4c05a588116f..882602f1c0ef 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -126,9 +126,21 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_find_get);
 struct nf_conntrack_expect *
 nf_ct_find_expectation(const struct nf_conntrack_tuple *tuple)
 {
-	struct nf_conntrack_expect *exp;
+	struct nf_conntrack_expect *i, *exp = NULL;
+	struct hlist_node *n;
+	unsigned int h;
+
+	if (!nf_ct_expect_count)
+		return NULL;
 
-	exp = __nf_ct_expect_find(tuple);
+	h = nf_ct_expect_dst_hash(tuple);
+	hlist_for_each_entry(i, n, &nf_ct_expect_hash[h], hnode) {
+		if (!(i->flags & NF_CT_EXPECT_INACTIVE) &&
+		    nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) {
+			exp = i;
+			break;
+		}
+	}
 	if (!exp)
 		return NULL;
 
@@ -460,6 +472,7 @@ static int exp_seq_show(struct seq_file *s, void *v)
 {
 	struct nf_conntrack_expect *expect;
 	struct hlist_node *n = v;
+	char *delim = "";
 
 	expect = hlist_entry(n, struct nf_conntrack_expect, hnode);
 
@@ -476,8 +489,12 @@ static int exp_seq_show(struct seq_file *s, void *v)
 		    __nf_ct_l4proto_find(expect->tuple.src.l3num,
 				       expect->tuple.dst.protonum));
 
-	if (expect->flags & NF_CT_EXPECT_PERMANENT)
-		seq_printf(s, "PERMANENT ");
+	if (expect->flags & NF_CT_EXPECT_PERMANENT) {
+		seq_printf(s, "PERMANENT");
+		delim = ",";
+	}
+	if (expect->flags & NF_CT_EXPECT_INACTIVE)
+		seq_printf(s, "%sINACTIVE", delim);
 
 	return seq_putc(s, '\n');
 }
-- 
cgit v1.2.3


From 6002f266b3e7f0acc2d5158cddbed41730b02e82 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:09:15 -0700
Subject: [NETFILTER]: nf_conntrack: introduce expectation classes and policies

Introduce expectation classes and policies. An expectation class
is used to distinguish different types of expectations by the
same helper (for example audio/video/t.120). The expectation
policy is used to hold the maximum number of expectations and
the initial timeout for each class.

The individual classes are isolated from each other, which means
that for example an audio expectation will only evict other audio
expectations.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netfilter/nf_conntrack.h        |  5 ++-
 include/net/netfilter/nf_conntrack_expect.h | 13 +++++-
 include/net/netfilter/nf_conntrack_helper.h |  5 +--
 net/ipv4/netfilter/nf_nat_snmp_basic.c      | 12 ++++--
 net/netfilter/nf_conntrack_amanda.c         | 14 +++---
 net/netfilter/nf_conntrack_expect.c         | 50 ++++++++++++++--------
 net/netfilter/nf_conntrack_ftp.c            | 10 +++--
 net/netfilter/nf_conntrack_h323_main.c      | 66 ++++++++++++++++++-----------
 net/netfilter/nf_conntrack_helper.c         |  3 +-
 net/netfilter/nf_conntrack_irc.c            | 10 +++--
 net/netfilter/nf_conntrack_netbios_ns.c     |  9 +++-
 net/netfilter/nf_conntrack_pptp.c           | 14 ++++--
 net/netfilter/nf_conntrack_sane.c           | 11 +++--
 net/netfilter/nf_conntrack_sip.c            | 10 +++--
 net/netfilter/nf_conntrack_tftp.c           | 11 +++--
 15 files changed, 166 insertions(+), 77 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 90b3e7f5df5f..922877133598 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -75,6 +75,9 @@ do {									\
 
 struct nf_conntrack_helper;
 
+/* Must be kept in sync with the classes defined by helpers */
+#define NF_CT_MAX_EXPECT_CLASSES	1
+
 /* nf_conn feature for connections that have a helper */
 struct nf_conn_help {
 	/* Helper. if any */
@@ -85,7 +88,7 @@ struct nf_conn_help {
 	struct hlist_head expectations;
 
 	/* Current number of expected connections */
-	unsigned int expecting;
+	u8 expecting[NF_CT_MAX_EXPECT_CLASSES];
 };
 
 
diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index 47c28dd07896..dfdf4b459475 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -41,6 +41,9 @@ struct nf_conntrack_expect
 	/* Flags */
 	unsigned int flags;
 
+	/* Expectation class */
+	unsigned int class;
+
 #ifdef CONFIG_NF_NAT_NEEDED
 	__be32 saved_ip;
 	/* This is the original per-proto part, used to map the
@@ -53,6 +56,14 @@ struct nf_conntrack_expect
 	struct rcu_head rcu;
 };
 
+struct nf_conntrack_expect_policy
+{
+	unsigned int	max_expected;
+	unsigned int	timeout;
+};
+
+#define NF_CT_EXPECT_CLASS_DEFAULT	0
+
 #define NF_CT_EXPECT_PERMANENT	0x1
 #define NF_CT_EXPECT_INACTIVE	0x2
 
@@ -75,7 +86,7 @@ void nf_ct_unexpect_related(struct nf_conntrack_expect *exp);
 /* Allocate space for an expectation: this is mandatory before calling
    nf_ct_expect_related.  You will have to call put afterwards. */
 struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me);
-void nf_ct_expect_init(struct nf_conntrack_expect *, int,
+void nf_ct_expect_init(struct nf_conntrack_expect *, unsigned int, int,
 		       const union nf_inet_addr *,
 		       const union nf_inet_addr *,
 		       u_int8_t, const __be16 *, const __be16 *);
diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h
index 4ca125e9b3ce..f8060ab5a083 100644
--- a/include/net/netfilter/nf_conntrack_helper.h
+++ b/include/net/netfilter/nf_conntrack_helper.h
@@ -20,9 +20,7 @@ struct nf_conntrack_helper
 
 	const char *name;		/* name of the module */
 	struct module *me;		/* pointer to self */
-	unsigned int max_expected;	/* Maximum number of concurrent 
-					 * expected connections */
-	unsigned int timeout;		/* timeout for expecteds */
+	const struct nf_conntrack_expect_policy *expect_policy;
 
 	/* Tuple of things we will help (compared against server response) */
 	struct nf_conntrack_tuple tuple;
@@ -37,6 +35,7 @@ struct nf_conntrack_helper
 	void (*destroy)(struct nf_conn *ct);
 
 	int (*to_nlattr)(struct sk_buff *skb, const struct nf_conn *ct);
+	unsigned int expect_class_max;
 };
 
 extern struct nf_conntrack_helper *
diff --git a/net/ipv4/netfilter/nf_nat_snmp_basic.c b/net/ipv4/netfilter/nf_nat_snmp_basic.c
index 540ce6ae887c..000e080bac5c 100644
--- a/net/ipv4/netfilter/nf_nat_snmp_basic.c
+++ b/net/ipv4/netfilter/nf_nat_snmp_basic.c
@@ -50,6 +50,7 @@
 #include <net/udp.h>
 
 #include <net/netfilter/nf_nat.h>
+#include <net/netfilter/nf_conntrack_expect.h>
 #include <net/netfilter/nf_conntrack_helper.h>
 #include <net/netfilter/nf_nat_helper.h>
 
@@ -1267,11 +1268,15 @@ static int help(struct sk_buff *skb, unsigned int protoff,
 	return ret;
 }
 
+static const struct nf_conntrack_expect_policy snmp_exp_policy = {
+	.max_expected	= 0,
+	.timeout	= 180,
+};
+
 static struct nf_conntrack_helper snmp_helper __read_mostly = {
-	.max_expected		= 0,
-	.timeout		= 180,
 	.me			= THIS_MODULE,
 	.help			= help,
+	.expect_policy		= &snmp_exp_policy,
 	.name			= "snmp",
 	.tuple.src.l3num	= AF_INET,
 	.tuple.src.u.udp.port	= __constant_htons(SNMP_PORT),
@@ -1279,10 +1284,9 @@ static struct nf_conntrack_helper snmp_helper __read_mostly = {
 };
 
 static struct nf_conntrack_helper snmp_trap_helper __read_mostly = {
-	.max_expected		= 0,
-	.timeout		= 180,
 	.me			= THIS_MODULE,
 	.help			= help,
+	.expect_policy		= &snmp_exp_policy,
 	.name			= "snmp_trap",
 	.tuple.src.l3num	= AF_INET,
 	.tuple.src.u.udp.port	= __constant_htons(SNMP_TRAP_PORT),
diff --git a/net/netfilter/nf_conntrack_amanda.c b/net/netfilter/nf_conntrack_amanda.c
index 7b8239c0cd5e..d14585a19b7d 100644
--- a/net/netfilter/nf_conntrack_amanda.c
+++ b/net/netfilter/nf_conntrack_amanda.c
@@ -148,7 +148,8 @@ static int amanda_help(struct sk_buff *skb,
 			goto out;
 		}
 		tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
-		nf_ct_expect_init(exp, family, &tuple->src.u3, &tuple->dst.u3,
+		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
+				  &tuple->src.u3, &tuple->dst.u3,
 				  IPPROTO_TCP, NULL, &port);
 
 		nf_nat_amanda = rcu_dereference(nf_nat_amanda_hook);
@@ -164,26 +165,29 @@ out:
 	return ret;
 }
 
+static const struct nf_conntrack_expect_policy amanda_exp_policy = {
+	.max_expected		= 3,
+	.timeout		= 180,
+};
+
 static struct nf_conntrack_helper amanda_helper[2] __read_mostly = {
 	{
 		.name			= "amanda",
-		.max_expected		= 3,
-		.timeout		= 180,
 		.me			= THIS_MODULE,
 		.help			= amanda_help,
 		.tuple.src.l3num	= AF_INET,
 		.tuple.src.u.udp.port	= __constant_htons(10080),
 		.tuple.dst.protonum	= IPPROTO_UDP,
+		.expect_policy		= &amanda_exp_policy,
 	},
 	{
 		.name			= "amanda",
-		.max_expected		= 3,
-		.timeout		= 180,
 		.me			= THIS_MODULE,
 		.help			= amanda_help,
 		.tuple.src.l3num	= AF_INET6,
 		.tuple.src.u.udp.port	= __constant_htons(10080),
 		.tuple.dst.protonum	= IPPROTO_UDP,
+		.expect_policy		= &amanda_exp_policy,
 	},
 };
 
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index 882602f1c0ef..e31beeb33b2b 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -54,7 +54,7 @@ void nf_ct_unlink_expect(struct nf_conntrack_expect *exp)
 	nf_ct_expect_count--;
 
 	hlist_del(&exp->lnode);
-	master_help->expecting--;
+	master_help->expecting[exp->class]--;
 	nf_ct_expect_put(exp);
 
 	NF_CT_STAT_INC(expect_delete);
@@ -171,7 +171,7 @@ void nf_ct_remove_expectations(struct nf_conn *ct)
 	struct hlist_node *n, *next;
 
 	/* Optimization: most connection never expect any others. */
-	if (!help || help->expecting == 0)
+	if (!help)
 		return;
 
 	hlist_for_each_entry_safe(exp, n, next, &help->expectations, lnode) {
@@ -205,7 +205,7 @@ static inline int expect_clash(const struct nf_conntrack_expect *a,
 static inline int expect_matches(const struct nf_conntrack_expect *a,
 				 const struct nf_conntrack_expect *b)
 {
-	return a->master == b->master
+	return a->master == b->master && a->class == b->class
 		&& nf_ct_tuple_equal(&a->tuple, &b->tuple)
 		&& nf_ct_tuple_mask_equal(&a->mask, &b->mask);
 }
@@ -240,7 +240,8 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me)
 }
 EXPORT_SYMBOL_GPL(nf_ct_expect_alloc);
 
-void nf_ct_expect_init(struct nf_conntrack_expect *exp, int family,
+void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class,
+		       int family,
 		       const union nf_inet_addr *saddr,
 		       const union nf_inet_addr *daddr,
 		       u_int8_t proto, const __be16 *src, const __be16 *dst)
@@ -253,6 +254,7 @@ void nf_ct_expect_init(struct nf_conntrack_expect *exp, int family,
 		len = 16;
 
 	exp->flags = 0;
+	exp->class = class;
 	exp->expectfn = NULL;
 	exp->helper = NULL;
 	exp->tuple.src.l3num = family;
@@ -309,19 +311,21 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_put);
 static void nf_ct_expect_insert(struct nf_conntrack_expect *exp)
 {
 	struct nf_conn_help *master_help = nfct_help(exp->master);
+	const struct nf_conntrack_expect_policy *p;
 	unsigned int h = nf_ct_expect_dst_hash(&exp->tuple);
 
 	atomic_inc(&exp->use);
 
 	hlist_add_head(&exp->lnode, &master_help->expectations);
-	master_help->expecting++;
+	master_help->expecting[exp->class]++;
 
 	hlist_add_head_rcu(&exp->hnode, &nf_ct_expect_hash[h]);
 	nf_ct_expect_count++;
 
 	setup_timer(&exp->timeout, nf_ct_expectation_timed_out,
 		    (unsigned long)exp);
-	exp->timeout.expires = jiffies + master_help->helper->timeout * HZ;
+	p = &master_help->helper->expect_policy[exp->class];
+	exp->timeout.expires = jiffies + p->timeout * HZ;
 	add_timer(&exp->timeout);
 
 	atomic_inc(&exp->use);
@@ -329,35 +333,41 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp)
 }
 
 /* Race with expectations being used means we could have none to find; OK. */
-static void evict_oldest_expect(struct nf_conn *master)
+static void evict_oldest_expect(struct nf_conn *master,
+				struct nf_conntrack_expect *new)
 {
 	struct nf_conn_help *master_help = nfct_help(master);
-	struct nf_conntrack_expect *exp = NULL;
+	struct nf_conntrack_expect *exp, *last = NULL;
 	struct hlist_node *n;
 
-	hlist_for_each_entry(exp, n, &master_help->expectations, lnode)
-		; /* nothing */
+	hlist_for_each_entry(exp, n, &master_help->expectations, lnode) {
+		if (exp->class == new->class)
+			last = exp;
+	}
 
-	if (exp && del_timer(&exp->timeout)) {
-		nf_ct_unlink_expect(exp);
-		nf_ct_expect_put(exp);
+	if (last && del_timer(&last->timeout)) {
+		nf_ct_unlink_expect(last);
+		nf_ct_expect_put(last);
 	}
 }
 
 static inline int refresh_timer(struct nf_conntrack_expect *i)
 {
 	struct nf_conn_help *master_help = nfct_help(i->master);
+	const struct nf_conntrack_expect_policy *p;
 
 	if (!del_timer(&i->timeout))
 		return 0;
 
-	i->timeout.expires = jiffies + master_help->helper->timeout*HZ;
+	p = &master_help->helper->expect_policy[i->class];
+	i->timeout.expires = jiffies + p->timeout * HZ;
 	add_timer(&i->timeout);
 	return 1;
 }
 
 int nf_ct_expect_related(struct nf_conntrack_expect *expect)
 {
+	const struct nf_conntrack_expect_policy *p;
 	struct nf_conntrack_expect *i;
 	struct nf_conn *master = expect->master;
 	struct nf_conn_help *master_help = nfct_help(master);
@@ -386,9 +396,15 @@ int nf_ct_expect_related(struct nf_conntrack_expect *expect)
 		}
 	}
 	/* Will be over limit? */
-	if (master_help->helper->max_expected &&
-	    master_help->expecting >= master_help->helper->max_expected)
-		evict_oldest_expect(master);
+	p = &master_help->helper->expect_policy[expect->class];
+	if (p->max_expected &&
+	    master_help->expecting[expect->class] >= p->max_expected) {
+		evict_oldest_expect(master, expect);
+		if (master_help->expecting[expect->class] >= p->max_expected) {
+			ret = -EMFILE;
+			goto out;
+		}
+	}
 
 	if (nf_ct_expect_count >= nf_ct_expect_max) {
 		if (net_ratelimit())
diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c
index 6770baf2e845..7eff876bb8bc 100644
--- a/net/netfilter/nf_conntrack_ftp.c
+++ b/net/netfilter/nf_conntrack_ftp.c
@@ -483,7 +483,7 @@ static int help(struct sk_buff *skb,
 		daddr = &cmd.u3;
 	}
 
-	nf_ct_expect_init(exp, cmd.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, cmd.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3, daddr,
 			  IPPROTO_TCP, NULL, &cmd.u.tcp.port);
 
@@ -517,6 +517,11 @@ out_update_nl:
 static struct nf_conntrack_helper ftp[MAX_PORTS][2] __read_mostly;
 static char ftp_names[MAX_PORTS][2][sizeof("ftp-65535")] __read_mostly;
 
+static const struct nf_conntrack_expect_policy ftp_exp_policy = {
+	.max_expected	= 1,
+	.timeout	= 5 * 60,
+};
+
 /* don't make this __exit, since it's called from __init ! */
 static void nf_conntrack_ftp_fini(void)
 {
@@ -556,8 +561,7 @@ static int __init nf_conntrack_ftp_init(void)
 		for (j = 0; j < 2; j++) {
 			ftp[i][j].tuple.src.u.tcp.port = htons(ports[i]);
 			ftp[i][j].tuple.dst.protonum = IPPROTO_TCP;
-			ftp[i][j].max_expected = 1;
-			ftp[i][j].timeout = 5 * 60;	/* 5 Minutes */
+			ftp[i][j].expect_policy = &ftp_exp_policy;
 			ftp[i][j].me = THIS_MODULE;
 			ftp[i][j].help = help;
 			tmpname = &ftp_names[i][j][0];
diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c
index 898f1922b5b8..505052d495cf 100644
--- a/net/netfilter/nf_conntrack_h323_main.c
+++ b/net/netfilter/nf_conntrack_h323_main.c
@@ -277,7 +277,8 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct,
 	/* Create expect for RTP */
 	if ((rtp_exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(rtp_exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(rtp_exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_UDP, NULL, &rtp_port);
@@ -287,7 +288,8 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct,
 		nf_ct_expect_put(rtp_exp);
 		return -1;
 	}
-	nf_ct_expect_init(rtcp_exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(rtcp_exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_UDP, NULL, &rtcp_port);
@@ -344,7 +346,8 @@ static int expect_t120(struct sk_buff *skb,
 	/* Create expect for T.120 connections */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_TCP, NULL, &port);
@@ -612,13 +615,17 @@ static int h245_help(struct sk_buff *skb, unsigned int protoff,
 }
 
 /****************************************************************************/
+static const struct nf_conntrack_expect_policy h245_exp_policy = {
+	.max_expected	= H323_RTP_CHANNEL_MAX * 4 + 2 /* T.120 */,
+	.timeout	= 240,
+};
+
 static struct nf_conntrack_helper nf_conntrack_helper_h245 __read_mostly = {
 	.name			= "H.245",
 	.me			= THIS_MODULE,
-	.max_expected		= H323_RTP_CHANNEL_MAX * 4 + 2 /* T.120 */,
-	.timeout		= 240,
 	.tuple.dst.protonum	= IPPROTO_UDP,
-	.help			= h245_help
+	.help			= h245_help,
+	.expect_policy		= &h245_exp_policy,
 };
 
 /****************************************************************************/
@@ -676,7 +683,8 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct,
 	/* Create expect for h245 connection */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_TCP, NULL, &port);
@@ -792,7 +800,8 @@ static int expect_callforwarding(struct sk_buff *skb,
 	/* Create expect for the second call leg */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_TCP, NULL, &port);
 	exp->helper = nf_conntrack_helper_q931;
@@ -1156,28 +1165,30 @@ static int q931_help(struct sk_buff *skb, unsigned int protoff,
 }
 
 /****************************************************************************/
+static const struct nf_conntrack_expect_policy q931_exp_policy = {
+	/* T.120 and H.245 */
+	.max_expected		= H323_RTP_CHANNEL_MAX * 4 + 4,
+	.timeout		= 240,
+};
+
 static struct nf_conntrack_helper nf_conntrack_helper_q931[] __read_mostly = {
 	{
 		.name			= "Q.931",
 		.me			= THIS_MODULE,
-					  /* T.120 and H.245 */
-		.max_expected		= H323_RTP_CHANNEL_MAX * 4 + 4,
-		.timeout		= 240,
 		.tuple.src.l3num	= AF_INET,
 		.tuple.src.u.tcp.port	= __constant_htons(Q931_PORT),
 		.tuple.dst.protonum	= IPPROTO_TCP,
-		.help			= q931_help
+		.help			= q931_help,
+		.expect_policy		= &q931_exp_policy,
 	},
 	{
 		.name			= "Q.931",
 		.me			= THIS_MODULE,
-					  /* T.120 and H.245 */
-		.max_expected		= H323_RTP_CHANNEL_MAX * 4 + 4,
-		.timeout		= 240,
 		.tuple.src.l3num	= AF_INET6,
 		.tuple.src.u.tcp.port	= __constant_htons(Q931_PORT),
 		.tuple.dst.protonum	= IPPROTO_TCP,
-		.help			= q931_help
+		.help			= q931_help,
+		.expect_policy		= &q931_exp_policy,
 	},
 };
 
@@ -1261,7 +1272,8 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct,
 	/* Create expect for Q.931 */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  gkrouted_only ? /* only accept calls from GK? */
 				&ct->tuplehash[!dir].tuple.src.u3 : NULL,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
@@ -1332,7 +1344,8 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct,
 	/* Need new expect */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_UDP, NULL, &port);
 	exp->helper = nf_conntrack_helper_ras;
@@ -1536,7 +1549,8 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct,
 	/* Need new expect */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_TCP, NULL, &port);
 	exp->flags = NF_CT_EXPECT_PERMANENT;
@@ -1589,7 +1603,8 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct,
 	/* Need new expect for call signal */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[!dir].tuple.src.l3num,
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_TCP, NULL, &port);
 	exp->flags = NF_CT_EXPECT_PERMANENT;
@@ -1728,26 +1743,29 @@ static int ras_help(struct sk_buff *skb, unsigned int protoff,
 }
 
 /****************************************************************************/
+static const struct nf_conntrack_expect_policy ras_exp_policy = {
+	.max_expected		= 32,
+	.timeout		= 240,
+};
+
 static struct nf_conntrack_helper nf_conntrack_helper_ras[] __read_mostly = {
 	{
 		.name			= "RAS",
 		.me			= THIS_MODULE,
-		.max_expected		= 32,
-		.timeout		= 240,
 		.tuple.src.l3num	= AF_INET,
 		.tuple.src.u.udp.port	= __constant_htons(RAS_PORT),
 		.tuple.dst.protonum	= IPPROTO_UDP,
 		.help			= ras_help,
+		.expect_policy		= &ras_exp_policy,
 	},
 	{
 		.name			= "RAS",
 		.me			= THIS_MODULE,
-		.max_expected		= 32,
-		.timeout		= 240,
 		.tuple.src.l3num	= AF_INET6,
 		.tuple.src.u.udp.port	= __constant_htons(RAS_PORT),
 		.tuple.dst.protonum	= IPPROTO_UDP,
 		.help			= ras_help,
+		.expect_policy		= &ras_exp_policy,
 	},
 };
 
diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c
index b1fd21cc1dbc..e350f56d43c9 100644
--- a/net/netfilter/nf_conntrack_helper.c
+++ b/net/netfilter/nf_conntrack_helper.c
@@ -110,7 +110,8 @@ int nf_conntrack_helper_register(struct nf_conntrack_helper *me)
 {
 	unsigned int h = helper_hash(&me->tuple);
 
-	BUG_ON(me->timeout == 0);
+	BUG_ON(me->expect_policy == NULL);
+	BUG_ON(me->expect_class_max >= NF_CT_MAX_EXPECT_CLASSES);
 
 	mutex_lock(&nf_ct_helper_mutex);
 	hlist_add_head_rcu(&me->hnode, &nf_ct_helper_hash[h]);
diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c
index c336b07a0d4c..02f21cbe5ae7 100644
--- a/net/netfilter/nf_conntrack_irc.c
+++ b/net/netfilter/nf_conntrack_irc.c
@@ -187,7 +187,8 @@ static int help(struct sk_buff *skb, unsigned int protoff,
 			}
 			tuple = &ct->tuplehash[!dir].tuple;
 			port = htons(dcc_port);
-			nf_ct_expect_init(exp, tuple->src.l3num,
+			nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+					  tuple->src.l3num,
 					  NULL, &tuple->dst.u3,
 					  IPPROTO_TCP, NULL, &port);
 
@@ -210,6 +211,7 @@ static int help(struct sk_buff *skb, unsigned int protoff,
 
 static struct nf_conntrack_helper irc[MAX_PORTS] __read_mostly;
 static char irc_names[MAX_PORTS][sizeof("irc-65535")] __read_mostly;
+static struct nf_conntrack_expect_policy irc_exp_policy;
 
 static void nf_conntrack_irc_fini(void);
 
@@ -223,6 +225,9 @@ static int __init nf_conntrack_irc_init(void)
 		return -EINVAL;
 	}
 
+	irc_exp_policy.max_expected = max_dcc_channels;
+	irc_exp_policy.timeout = dcc_timeout;
+
 	irc_buffer = kmalloc(65536, GFP_KERNEL);
 	if (!irc_buffer)
 		return -ENOMEM;
@@ -235,8 +240,7 @@ static int __init nf_conntrack_irc_init(void)
 		irc[i].tuple.src.l3num = AF_INET;
 		irc[i].tuple.src.u.tcp.port = htons(ports[i]);
 		irc[i].tuple.dst.protonum = IPPROTO_TCP;
-		irc[i].max_expected = max_dcc_channels;
-		irc[i].timeout = dcc_timeout;
+		irc[i].expect_policy = &irc_exp_policy;
 		irc[i].me = THIS_MODULE;
 		irc[i].help = help;
 
diff --git a/net/netfilter/nf_conntrack_netbios_ns.c b/net/netfilter/nf_conntrack_netbios_ns.c
index 60dedaded84e..08404e6755fb 100644
--- a/net/netfilter/nf_conntrack_netbios_ns.c
+++ b/net/netfilter/nf_conntrack_netbios_ns.c
@@ -86,6 +86,7 @@ static int help(struct sk_buff *skb, unsigned int protoff,
 
 	exp->expectfn             = NULL;
 	exp->flags                = NF_CT_EXPECT_PERMANENT;
+	exp->class		  = NF_CT_EXPECT_CLASS_DEFAULT;
 	exp->helper               = NULL;
 
 	nf_ct_expect_related(exp);
@@ -96,19 +97,23 @@ out:
 	return NF_ACCEPT;
 }
 
+static struct nf_conntrack_expect_policy exp_policy = {
+	.max_expected	= 1,
+};
+
 static struct nf_conntrack_helper helper __read_mostly = {
 	.name			= "netbios-ns",
 	.tuple.src.l3num	= AF_INET,
 	.tuple.src.u.udp.port	= __constant_htons(NMBD_PORT),
 	.tuple.dst.protonum	= IPPROTO_UDP,
-	.max_expected		= 1,
 	.me			= THIS_MODULE,
 	.help			= help,
+	.expect_policy		= &exp_policy,
 };
 
 static int __init nf_conntrack_netbios_ns_init(void)
 {
-	helper.timeout = timeout;
+	exp_policy.timeout = timeout;
 	return nf_conntrack_helper_register(&helper);
 }
 
diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c
index b5cb8e831230..8fd83470d1b3 100644
--- a/net/netfilter/nf_conntrack_pptp.c
+++ b/net/netfilter/nf_conntrack_pptp.c
@@ -208,7 +208,8 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid)
 
 	/* original direction, PNS->PAC */
 	dir = IP_CT_DIR_ORIGINAL;
-	nf_ct_expect_init(exp_orig, ct->tuplehash[dir].tuple.src.l3num,
+	nf_ct_expect_init(exp_orig, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[dir].tuple.src.l3num,
 			  &ct->tuplehash[dir].tuple.src.u3,
 			  &ct->tuplehash[dir].tuple.dst.u3,
 			  IPPROTO_GRE, &peer_callid, &callid);
@@ -216,7 +217,8 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid)
 
 	/* reply direction, PAC->PNS */
 	dir = IP_CT_DIR_REPLY;
-	nf_ct_expect_init(exp_reply, ct->tuplehash[dir].tuple.src.l3num,
+	nf_ct_expect_init(exp_reply, NF_CT_EXPECT_CLASS_DEFAULT,
+			  ct->tuplehash[dir].tuple.src.l3num,
 			  &ct->tuplehash[dir].tuple.src.u3,
 			  &ct->tuplehash[dir].tuple.dst.u3,
 			  IPPROTO_GRE, &callid, &peer_callid);
@@ -575,17 +577,21 @@ conntrack_pptp_help(struct sk_buff *skb, unsigned int protoff,
 	return ret;
 }
 
+static const struct nf_conntrack_expect_policy pptp_exp_policy = {
+	.max_expected	= 2,
+	.timeout	= 5 * 60,
+};
+
 /* control protocol helper */
 static struct nf_conntrack_helper pptp __read_mostly = {
 	.name			= "pptp",
 	.me			= THIS_MODULE,
-	.max_expected		= 2,
-	.timeout		= 5 * 60,
 	.tuple.src.l3num	= AF_INET,
 	.tuple.src.u.tcp.port	= __constant_htons(PPTP_CONTROL_PORT),
 	.tuple.dst.protonum	= IPPROTO_TCP,
 	.help			= conntrack_pptp_help,
 	.destroy		= pptp_destroy_siblings,
+	.expect_policy		= &pptp_exp_policy,
 };
 
 static int __init nf_conntrack_pptp_init(void)
diff --git a/net/netfilter/nf_conntrack_sane.c b/net/netfilter/nf_conntrack_sane.c
index a70051d741a7..7542e25eede3 100644
--- a/net/netfilter/nf_conntrack_sane.c
+++ b/net/netfilter/nf_conntrack_sane.c
@@ -143,7 +143,8 @@ static int help(struct sk_buff *skb,
 	}
 
 	tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
-	nf_ct_expect_init(exp, family, &tuple->src.u3, &tuple->dst.u3,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
+			  &tuple->src.u3, &tuple->dst.u3,
 			  IPPROTO_TCP, NULL, &reply->port);
 
 	pr_debug("nf_ct_sane: expect: ");
@@ -163,6 +164,11 @@ out:
 static struct nf_conntrack_helper sane[MAX_PORTS][2] __read_mostly;
 static char sane_names[MAX_PORTS][2][sizeof("sane-65535")] __read_mostly;
 
+static const struct nf_conntrack_expect_policy sane_exp_policy = {
+	.max_expected	= 1,
+	.timeout	= 5 * 60,
+};
+
 /* don't make this __exit, since it's called from __init ! */
 static void nf_conntrack_sane_fini(void)
 {
@@ -200,8 +206,7 @@ static int __init nf_conntrack_sane_init(void)
 		for (j = 0; j < 2; j++) {
 			sane[i][j].tuple.src.u.tcp.port = htons(ports[i]);
 			sane[i][j].tuple.dst.protonum = IPPROTO_TCP;
-			sane[i][j].max_expected = 1;
-			sane[i][j].timeout = 5 * 60;	/* 5 Minutes */
+			sane[i][j].expect_policy = &sane_exp_policy;
 			sane[i][j].me = THIS_MODULE;
 			sane[i][j].help = help;
 			tmpname = &sane_names[i][j][0];
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index c521c891d351..0021d5b60cec 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -380,7 +380,7 @@ static int set_expected_rtp(struct sk_buff *skb,
 	exp = nf_ct_expect_alloc(ct);
 	if (exp == NULL)
 		return NF_DROP;
-	nf_ct_expect_init(exp, family,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
 			  &ct->tuplehash[!dir].tuple.src.u3, addr,
 			  IPPROTO_UDP, NULL, &port);
 
@@ -476,6 +476,11 @@ out:
 static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly;
 static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly;
 
+static const struct nf_conntrack_expect_policy sip_exp_policy = {
+	.max_expected	= 2,
+	.timeout	= 3 * 60,
+};
+
 static void nf_conntrack_sip_fini(void)
 {
 	int i, j;
@@ -505,8 +510,7 @@ static int __init nf_conntrack_sip_init(void)
 		for (j = 0; j < 2; j++) {
 			sip[i][j].tuple.dst.protonum = IPPROTO_UDP;
 			sip[i][j].tuple.src.u.udp.port = htons(ports[i]);
-			sip[i][j].max_expected = 2;
-			sip[i][j].timeout = 3 * 60; /* 3 minutes */
+			sip[i][j].expect_policy = &sip_exp_policy;
 			sip[i][j].me = THIS_MODULE;
 			sip[i][j].help = sip_help;
 
diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c
index bd2e800f23cc..a28341b30f21 100644
--- a/net/netfilter/nf_conntrack_tftp.c
+++ b/net/netfilter/nf_conntrack_tftp.c
@@ -63,7 +63,8 @@ static int tftp_help(struct sk_buff *skb,
 		if (exp == NULL)
 			return NF_DROP;
 		tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-		nf_ct_expect_init(exp, family, &tuple->src.u3, &tuple->dst.u3,
+		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
+				  &tuple->src.u3, &tuple->dst.u3,
 				  IPPROTO_UDP, NULL, &tuple->dst.u.udp.port);
 
 		pr_debug("expect: ");
@@ -92,6 +93,11 @@ static int tftp_help(struct sk_buff *skb,
 static struct nf_conntrack_helper tftp[MAX_PORTS][2] __read_mostly;
 static char tftp_names[MAX_PORTS][2][sizeof("tftp-65535")] __read_mostly;
 
+static const struct nf_conntrack_expect_policy tftp_exp_policy = {
+	.max_expected	= 1,
+	.timeout	= 5 * 60,
+};
+
 static void nf_conntrack_tftp_fini(void)
 {
 	int i, j;
@@ -118,8 +124,7 @@ static int __init nf_conntrack_tftp_init(void)
 		for (j = 0; j < 2; j++) {
 			tftp[i][j].tuple.dst.protonum = IPPROTO_UDP;
 			tftp[i][j].tuple.src.u.udp.port = htons(ports[i]);
-			tftp[i][j].max_expected = 1;
-			tftp[i][j].timeout = 5 * 60; /* 5 minutes */
+			tftp[i][j].expect_policy = &tftp_exp_policy;
 			tftp[i][j].me = THIS_MODULE;
 			tftp[i][j].help = tftp_help;
 
-- 
cgit v1.2.3


From b8beedd25d3913d45b8330a08ab88fdf90eb54b8 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:09:33 -0700
Subject: [NETFILTER]: Add nf_inet_addr_cmp()

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h                  |  9 +++++++++
 include/net/netfilter/nf_conntrack_tuple.h | 15 +++------------
 2 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index f0680c2bee73..89e6c72ad295 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -61,6 +61,15 @@ union nf_inet_addr {
 #ifdef __KERNEL__
 #ifdef CONFIG_NETFILTER
 
+static inline int nf_inet_addr_cmp(const union nf_inet_addr *a1,
+				   const union nf_inet_addr *a2)
+{
+	return a1->all[0] == a2->all[0] &&
+	       a1->all[1] == a2->all[1] &&
+	       a1->all[2] == a2->all[2] &&
+	       a1->all[3] == a2->all[3];
+}
+
 extern void netfilter_init(void);
 
 /* Largest hook number + 1 */
diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h
index d9b53dd601ad..168c91754d89 100644
--- a/include/net/netfilter/nf_conntrack_tuple.h
+++ b/include/net/netfilter/nf_conntrack_tuple.h
@@ -163,10 +163,7 @@ struct nf_conntrack_tuple_hash
 static inline int __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1,
 					  const struct nf_conntrack_tuple *t2)
 { 
-	return (t1->src.u3.all[0] == t2->src.u3.all[0] &&
-		t1->src.u3.all[1] == t2->src.u3.all[1] &&
-		t1->src.u3.all[2] == t2->src.u3.all[2] &&
-		t1->src.u3.all[3] == t2->src.u3.all[3] &&
+	return (nf_inet_addr_cmp(&t1->src.u3, &t2->src.u3) &&
 		t1->src.u.all == t2->src.u.all &&
 		t1->src.l3num == t2->src.l3num);
 }
@@ -174,10 +171,7 @@ static inline int __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1,
 static inline int __nf_ct_tuple_dst_equal(const struct nf_conntrack_tuple *t1,
 					  const struct nf_conntrack_tuple *t2)
 {
-	return (t1->dst.u3.all[0] == t2->dst.u3.all[0] &&
-		t1->dst.u3.all[1] == t2->dst.u3.all[1] &&
-		t1->dst.u3.all[2] == t2->dst.u3.all[2] &&
-		t1->dst.u3.all[3] == t2->dst.u3.all[3] &&
+	return (nf_inet_addr_cmp(&t1->dst.u3, &t2->dst.u3) &&
 		t1->dst.u.all == t2->dst.u.all &&
 		t1->dst.protonum == t2->dst.protonum);
 }
@@ -192,10 +186,7 @@ static inline int nf_ct_tuple_equal(const struct nf_conntrack_tuple *t1,
 static inline int nf_ct_tuple_mask_equal(const struct nf_conntrack_tuple_mask *m1,
 					 const struct nf_conntrack_tuple_mask *m2)
 {
-	return (m1->src.u3.all[0] == m2->src.u3.all[0] &&
-		m1->src.u3.all[1] == m2->src.u3.all[1] &&
-		m1->src.u3.all[2] == m2->src.u3.all[2] &&
-		m1->src.u3.all[3] == m2->src.u3.all[3] &&
+	return (nf_inet_addr_cmp(&m1->src.u3, &m2->src.u3) &&
 		m1->src.u.all == m2->src.u.all);
 }
 
-- 
cgit v1.2.3


From 2a6cfb22ae002330d445f734668d9158db9e90de Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:16:54 -0700
Subject: [NETFILTER]: nf_conntrack_sip: adjust dptr and datalen after packet
 mangling

After mangling the packet, the pointer to the data and the length of the data
portion may change and need to be adjusted.

Use double data pointers and a pointer to the length everywhere and add a
helper function to the NAT helper for performing the adjustments.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  6 +-
 net/ipv4/netfilter/nf_nat_sip.c            | 91 ++++++++++++++++--------------
 net/netfilter/nf_conntrack_sip.c           | 14 +++--
 3 files changed, 60 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 8e5ce1ca7bfc..9d0dbfb26300 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -24,11 +24,13 @@ enum sip_header_pos {
 extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb,
 				       enum ip_conntrack_info ctinfo,
 				       struct nf_conn *ct,
-				       const char **dptr);
+				       const char **dptr,
+				       unsigned int *datalen);
 extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				       enum ip_conntrack_info ctinfo,
 				       struct nf_conntrack_expect *exp,
-				       const char *dptr);
+				       const char **dptr,
+				       unsigned int *datalen);
 
 extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr,
                            size_t dlen, unsigned int *matchoff,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index 84d8b4982cdf..e77122e65283 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -60,15 +60,35 @@ static void addr_map_init(const struct nf_conn *ct, struct addr_map *map)
 	}
 }
 
+static unsigned int mangle_packet(struct sk_buff *skb,
+				  const char **dptr, unsigned int *datalen,
+				  unsigned int matchoff, unsigned int matchlen,
+				  const char *buffer, unsigned int buflen)
+{
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+
+	if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo, matchoff, matchlen,
+				      buffer, buflen))
+		return 0;
+
+	/* Reload data pointer and adjust datalen value */
+	*dptr = skb->data + ip_hdrlen(skb) + sizeof(struct udphdr);
+	*datalen += buflen - matchlen;
+	return 1;
+}
+
 static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo,
-			struct nf_conn *ct, const char **dptr, size_t dlen,
+			struct nf_conn *ct,
+			const char **dptr, unsigned int *datalen,
 			enum sip_header_pos pos, struct addr_map *map)
 {
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	unsigned int matchlen, matchoff, addrlen;
 	char *addr;
 
-	if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0)
+	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
+			    pos) <= 0)
 		return 1;
 
 	if ((matchlen == map->addr[dir].srciplen ||
@@ -84,26 +104,19 @@ static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo,
 	} else
 		return 1;
 
-	if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo,
-				      matchoff, matchlen, addr, addrlen))
-		return 0;
-	*dptr = skb->data + ip_hdrlen(skb) + sizeof(struct udphdr);
-	return 1;
-
+	return mangle_packet(skb, dptr, datalen, matchoff, matchlen,
+			     addr, addrlen);
 }
 
 static unsigned int ip_nat_sip(struct sk_buff *skb,
 			       enum ip_conntrack_info ctinfo,
 			       struct nf_conn *ct,
-			       const char **dptr)
+			       const char **dptr, unsigned int *datalen)
 {
 	enum sip_header_pos pos;
 	struct addr_map map;
-	int dataoff, datalen;
 
-	dataoff = ip_hdrlen(skb) + sizeof(struct udphdr);
-	datalen = skb->len - dataoff;
-	if (datalen < sizeof("SIP/2.0") - 1)
+	if (*datalen < sizeof("SIP/2.0") - 1)
 		return NF_ACCEPT;
 
 	addr_map_init(ct, &map);
@@ -115,7 +128,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb,
 		 * The "userinfo" and "@" components of the SIP URI MUST NOT
 		 * be present.
 		 */
-		if (datalen >= sizeof("REGISTER") - 1 &&
+		if (*datalen >= sizeof("REGISTER") - 1 &&
 		    strncmp(*dptr, "REGISTER", sizeof("REGISTER") - 1) == 0)
 			pos = POS_REG_REQ_URI;
 		else
@@ -136,51 +149,45 @@ static unsigned int ip_nat_sip(struct sk_buff *skb,
 static unsigned int mangle_sip_packet(struct sk_buff *skb,
 				      enum ip_conntrack_info ctinfo,
 				      struct nf_conn *ct,
-				      const char **dptr, size_t dlen,
+				      const char **dptr, unsigned int *datalen,
 				      char *buffer, int bufflen,
 				      enum sip_header_pos pos)
 {
 	unsigned int matchlen, matchoff;
 
-	if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0)
-		return 0;
-
-	if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo,
-				      matchoff, matchlen, buffer, bufflen))
+	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
+			    pos) <= 0)
 		return 0;
 
-	/* We need to reload this. Thanks Patrick. */
-	*dptr = skb->data + ip_hdrlen(skb) + sizeof(struct udphdr);
-	return 1;
+	return mangle_packet(skb, dptr, datalen, matchoff, matchlen,
+			     buffer, bufflen);
 }
 
 static int mangle_content_len(struct sk_buff *skb,
 			      enum ip_conntrack_info ctinfo,
 			      struct nf_conn *ct,
-			      const char *dptr)
+			      const char **dptr, unsigned int *datalen)
 {
-	unsigned int dataoff, matchoff, matchlen;
+	unsigned int matchoff, matchlen;
 	char buffer[sizeof("65536")];
 	int bufflen;
 
-	dataoff = ip_hdrlen(skb) + sizeof(struct udphdr);
-
 	/* Get actual SDP length */
-	if (ct_sip_get_info(ct, dptr, skb->len - dataoff, &matchoff,
+	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff,
 			    &matchlen, POS_SDP_HEADER) > 0) {
 
 		/* since ct_sip_get_info() give us a pointer passing 'v='
 		   we need to add 2 bytes in this count. */
-		int c_len = skb->len - dataoff - matchoff + 2;
+		int c_len = *datalen - matchoff + 2;
 
 		/* Now, update SDP length */
-		if (ct_sip_get_info(ct, dptr, skb->len - dataoff, &matchoff,
+		if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff,
 				    &matchlen, POS_CONTENT) > 0) {
 
 			bufflen = sprintf(buffer, "%u", c_len);
-			return nf_nat_mangle_udp_packet(skb, ct, ctinfo,
-							matchoff, matchlen,
-							buffer, bufflen);
+			return mangle_packet(skb, dptr, datalen,
+					     matchoff, matchlen,
+					     buffer, bufflen);
 		}
 	}
 	return 0;
@@ -190,30 +197,28 @@ static unsigned int mangle_sdp(struct sk_buff *skb,
 			       enum ip_conntrack_info ctinfo,
 			       struct nf_conn *ct,
 			       __be32 newip, u_int16_t port,
-			       const char *dptr)
+			       const char **dptr, unsigned int *datalen)
 {
 	char buffer[sizeof("nnn.nnn.nnn.nnn")];
-	unsigned int dataoff, bufflen;
-
-	dataoff = ip_hdrlen(skb) + sizeof(struct udphdr);
+	unsigned int bufflen;
 
 	/* Mangle owner and contact info. */
 	bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip));
-	if (!mangle_sip_packet(skb, ctinfo, ct, &dptr, skb->len - dataoff,
+	if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen,
 			       buffer, bufflen, POS_OWNER_IP4))
 		return 0;
 
-	if (!mangle_sip_packet(skb, ctinfo, ct, &dptr, skb->len - dataoff,
+	if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen,
 			       buffer, bufflen, POS_CONNECTION_IP4))
 		return 0;
 
 	/* Mangle media port. */
 	bufflen = sprintf(buffer, "%u", port);
-	if (!mangle_sip_packet(skb, ctinfo, ct, &dptr, skb->len - dataoff,
+	if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen,
 			       buffer, bufflen, POS_MEDIA))
 		return 0;
 
-	return mangle_content_len(skb, ctinfo, ct, dptr);
+	return mangle_content_len(skb, ctinfo, ct, dptr, datalen);
 }
 
 static void ip_nat_sdp_expect(struct nf_conn *ct,
@@ -242,7 +247,7 @@ static void ip_nat_sdp_expect(struct nf_conn *ct,
 static unsigned int ip_nat_sdp(struct sk_buff *skb,
 			       enum ip_conntrack_info ctinfo,
 			       struct nf_conntrack_expect *exp,
-			       const char *dptr)
+			       const char **dptr, unsigned int *datalen)
 {
 	struct nf_conn *ct = exp->master;
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
@@ -275,7 +280,7 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb,
 	if (port == 0)
 		return NF_DROP;
 
-	if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr)) {
+	if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr, datalen)) {
 		nf_ct_unexpect_related(exp);
 		return NF_DROP;
 	}
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 016e1c1aafe4..fa0d5599ff24 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -39,13 +39,15 @@ MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session");
 unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb,
 				enum ip_conntrack_info ctinfo,
 				struct nf_conn *ct,
-				const char **dptr) __read_mostly;
+				const char **dptr,
+				unsigned int *datalen) __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sip_hook);
 
 unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				enum ip_conntrack_info ctinfo,
 				struct nf_conntrack_expect *exp,
-				const char *dptr) __read_mostly;
+				const char **dptr,
+				unsigned int *datalen) __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sdp_hook);
 
 static int digits_len(const struct nf_conn *, const char *, const char *, int *);
@@ -369,7 +371,7 @@ static int set_expected_rtp(struct sk_buff *skb,
 			    enum ip_conntrack_info ctinfo,
 			    union nf_inet_addr *addr,
 			    __be16 port,
-			    const char *dptr)
+			    const char **dptr, unsigned int *datalen)
 {
 	struct nf_conntrack_expect *exp;
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
@@ -386,7 +388,7 @@ static int set_expected_rtp(struct sk_buff *skb,
 
 	nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook);
 	if (nf_nat_sdp && ct->status & IPS_NAT_MASK)
-		ret = nf_nat_sdp(skb, ctinfo, exp, dptr);
+		ret = nf_nat_sdp(skb, ctinfo, exp, dptr, datalen);
 	else {
 		if (nf_ct_expect_related(exp) != 0)
 			ret = NF_DROP;
@@ -429,7 +431,7 @@ static int sip_help(struct sk_buff *skb,
 
 	nf_nat_sip = rcu_dereference(nf_nat_sip_hook);
 	if (nf_nat_sip && ct->status & IPS_NAT_MASK) {
-		if (!nf_nat_sip(skb, ctinfo, ct, &dptr)) {
+		if (!nf_nat_sip(skb, ctinfo, ct, &dptr, &datalen)) {
 			ret = NF_DROP;
 			goto out;
 		}
@@ -466,7 +468,7 @@ static int sip_help(struct sk_buff *skb,
 				goto out;
 			}
 			ret = set_expected_rtp(skb, ct, ctinfo, &addr,
-					       htons(port), dptr);
+					       htons(port), &dptr, &datalen);
 		}
 	}
 out:
-- 
cgit v1.2.3


From 212440a7d04a12ee13787afecc6c86c7fc4e6184 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:17:13 -0700
Subject: [NETFILTER]: nf_conntrack_sip: remove redundant function arguments

The conntrack reference and ctinfo can be derived from the packet.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  7 ++---
 net/ipv4/netfilter/nf_nat_sip.c            | 49 +++++++++++++++---------------
 net/netfilter/nf_conntrack_sip.c           | 24 ++++++---------
 3 files changed, 37 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 9d0dbfb26300..b94de3d60303 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -22,15 +22,12 @@ enum sip_header_pos {
 };
 
 extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb,
-				       enum ip_conntrack_info ctinfo,
-				       struct nf_conn *ct,
 				       const char **dptr,
 				       unsigned int *datalen);
 extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
-				       enum ip_conntrack_info ctinfo,
-				       struct nf_conntrack_expect *exp,
 				       const char **dptr,
-				       unsigned int *datalen);
+				       unsigned int *datalen,
+				       struct nf_conntrack_expect *exp);
 
 extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr,
                            size_t dlen, unsigned int *matchoff,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index e77122e65283..acaa7d4569fa 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -78,11 +78,12 @@ static unsigned int mangle_packet(struct sk_buff *skb,
 	return 1;
 }
 
-static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo,
-			struct nf_conn *ct,
+static int map_sip_addr(struct sk_buff *skb,
 			const char **dptr, unsigned int *datalen,
 			enum sip_header_pos pos, struct addr_map *map)
 {
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	unsigned int matchlen, matchoff, addrlen;
 	char *addr;
@@ -109,10 +110,10 @@ static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo,
 }
 
 static unsigned int ip_nat_sip(struct sk_buff *skb,
-			       enum ip_conntrack_info ctinfo,
-			       struct nf_conn *ct,
 			       const char **dptr, unsigned int *datalen)
 {
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	enum sip_header_pos pos;
 	struct addr_map map;
 
@@ -134,25 +135,25 @@ static unsigned int ip_nat_sip(struct sk_buff *skb,
 		else
 			pos = POS_REQ_URI;
 
-		if (!map_sip_addr(skb, ctinfo, ct, dptr, datalen, pos, &map))
+		if (!map_sip_addr(skb, dptr, datalen, pos, &map))
 			return NF_DROP;
 	}
 
-	if (!map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_FROM, &map) ||
-	    !map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_TO, &map) ||
-	    !map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_VIA, &map) ||
-	    !map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_CONTACT, &map))
+	if (!map_sip_addr(skb, dptr, datalen, POS_FROM, &map) ||
+	    !map_sip_addr(skb, dptr, datalen, POS_TO, &map) ||
+	    !map_sip_addr(skb, dptr, datalen, POS_VIA, &map) ||
+	    !map_sip_addr(skb, dptr, datalen, POS_CONTACT, &map))
 		return NF_DROP;
 	return NF_ACCEPT;
 }
 
 static unsigned int mangle_sip_packet(struct sk_buff *skb,
-				      enum ip_conntrack_info ctinfo,
-				      struct nf_conn *ct,
 				      const char **dptr, unsigned int *datalen,
 				      char *buffer, int bufflen,
 				      enum sip_header_pos pos)
 {
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	unsigned int matchlen, matchoff;
 
 	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
@@ -164,10 +165,10 @@ static unsigned int mangle_sip_packet(struct sk_buff *skb,
 }
 
 static int mangle_content_len(struct sk_buff *skb,
-			      enum ip_conntrack_info ctinfo,
-			      struct nf_conn *ct,
 			      const char **dptr, unsigned int *datalen)
 {
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	unsigned int matchoff, matchlen;
 	char buffer[sizeof("65536")];
 	int bufflen;
@@ -204,21 +205,21 @@ static unsigned int mangle_sdp(struct sk_buff *skb,
 
 	/* Mangle owner and contact info. */
 	bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip));
-	if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen,
-			       buffer, bufflen, POS_OWNER_IP4))
+	if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen,
+			       POS_OWNER_IP4))
 		return 0;
 
-	if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen,
-			       buffer, bufflen, POS_CONNECTION_IP4))
+	if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen,
+			       POS_CONNECTION_IP4))
 		return 0;
 
 	/* Mangle media port. */
 	bufflen = sprintf(buffer, "%u", port);
-	if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen,
-			       buffer, bufflen, POS_MEDIA))
+	if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen,
+			       POS_MEDIA))
 		return 0;
 
-	return mangle_content_len(skb, ctinfo, ct, dptr, datalen);
+	return mangle_content_len(skb, dptr, datalen);
 }
 
 static void ip_nat_sdp_expect(struct nf_conn *ct,
@@ -245,11 +246,11 @@ static void ip_nat_sdp_expect(struct nf_conn *ct,
 /* So, this packet has hit the connection tracking matching code.
    Mangle it, and change the expectation to match the new version. */
 static unsigned int ip_nat_sdp(struct sk_buff *skb,
-			       enum ip_conntrack_info ctinfo,
-			       struct nf_conntrack_expect *exp,
-			       const char **dptr, unsigned int *datalen)
+			       const char **dptr, unsigned int *datalen,
+			       struct nf_conntrack_expect *exp)
 {
-	struct nf_conn *ct = exp->master;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	__be32 newip;
 	u_int16_t port;
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index fa0d5599ff24..38e1e7a05334 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -37,17 +37,14 @@ module_param(sip_timeout, uint, 0600);
 MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session");
 
 unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb,
-				enum ip_conntrack_info ctinfo,
-				struct nf_conn *ct,
 				const char **dptr,
 				unsigned int *datalen) __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sip_hook);
 
 unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
-				enum ip_conntrack_info ctinfo,
-				struct nf_conntrack_expect *exp,
 				const char **dptr,
-				unsigned int *datalen) __read_mostly;
+				unsigned int *datalen,
+				struct nf_conntrack_expect *exp) __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sdp_hook);
 
 static int digits_len(const struct nf_conn *, const char *, const char *, int *);
@@ -367,13 +364,12 @@ int ct_sip_get_info(const struct nf_conn *ct,
 EXPORT_SYMBOL_GPL(ct_sip_get_info);
 
 static int set_expected_rtp(struct sk_buff *skb,
-			    struct nf_conn *ct,
-			    enum ip_conntrack_info ctinfo,
-			    union nf_inet_addr *addr,
-			    __be16 port,
-			    const char **dptr, unsigned int *datalen)
+			    const char **dptr, unsigned int *datalen,
+			    union nf_inet_addr *addr, __be16 port)
 {
 	struct nf_conntrack_expect *exp;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	int family = ct->tuplehash[!dir].tuple.src.l3num;
 	int ret;
@@ -388,7 +384,7 @@ static int set_expected_rtp(struct sk_buff *skb,
 
 	nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook);
 	if (nf_nat_sdp && ct->status & IPS_NAT_MASK)
-		ret = nf_nat_sdp(skb, ctinfo, exp, dptr, datalen);
+		ret = nf_nat_sdp(skb, dptr, datalen, exp);
 	else {
 		if (nf_ct_expect_related(exp) != 0)
 			ret = NF_DROP;
@@ -431,7 +427,7 @@ static int sip_help(struct sk_buff *skb,
 
 	nf_nat_sip = rcu_dereference(nf_nat_sip_hook);
 	if (nf_nat_sip && ct->status & IPS_NAT_MASK) {
-		if (!nf_nat_sip(skb, ctinfo, ct, &dptr, &datalen)) {
+		if (!nf_nat_sip(skb, &dptr, &datalen)) {
 			ret = NF_DROP;
 			goto out;
 		}
@@ -467,8 +463,8 @@ static int sip_help(struct sk_buff *skb,
 				ret = NF_DROP;
 				goto out;
 			}
-			ret = set_expected_rtp(skb, ct, ctinfo, &addr,
-					       htons(port), &dptr, &datalen);
+			ret = set_expected_rtp(skb, &dptr, &datalen,
+					       &addr, htons(port));
 		}
 	}
 out:
-- 
cgit v1.2.3


From 3e9b4600b4e71beaa9d943251bfe9c25f6a97b8c Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:17:55 -0700
Subject: [NETFILTER]: nf_conntrack_sip: add seperate SDP header parsing
 function

SDP and SIP headers are quite different, SIP can have continuation lines,
leading and trailing whitespace after the colon and is mostly case-insensitive
while SDP headers always begin on a new line and are followed by an equal
sign and the value, without any whitespace.

Introduce new SDP header parsing function and convert all users that used
the SIP header parsing function. This will allow to properly deal with the
special SIP cases in the SIP header parsing function later.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  49 +++++++--
 net/ipv4/netfilter/nf_nat_sip.c            |  69 ++++++-------
 net/netfilter/nf_conntrack_sip.c           | 159 +++++++++++++++++------------
 3 files changed, 169 insertions(+), 108 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index b94de3d60303..9131cbc9b9de 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -13,12 +13,42 @@ enum sip_header_pos {
 	POS_VIA,
 	POS_CONTACT,
 	POS_CONTENT,
-	POS_MEDIA,
-	POS_OWNER_IP4,
-	POS_CONNECTION_IP4,
-	POS_OWNER_IP6,
-	POS_CONNECTION_IP6,
-	POS_SDP_HEADER,
+};
+
+struct sip_header {
+	const char	*name;
+	const char	*cname;
+	const char	*search;
+	unsigned int	len;
+	unsigned int	clen;
+	unsigned int	slen;
+	int		(*match_len)(const struct nf_conn *ct,
+				     const char *dptr, const char *limit,
+				     int *shift);
+};
+
+#define __SIP_HDR(__name, __cname, __search, __match)			\
+{									\
+	.name		= (__name),					\
+	.len		= sizeof(__name) - 1,				\
+	.cname		= (__cname),					\
+	.clen		= (__cname) ? sizeof(__cname) - 1 : 0,		\
+	.search		= (__search),					\
+	.slen		= (__search) ? sizeof(__search) - 1 : 0,	\
+	.match_len	= (__match),					\
+}
+
+#define SDP_HDR(__name, __search, __match) \
+	__SIP_HDR(__name, NULL, __search, __match)
+
+enum sdp_header_types {
+	SDP_HDR_UNSPEC,
+	SDP_HDR_VERSION,
+	SDP_HDR_OWNER_IP4,
+	SDP_HDR_CONNECTION_IP4,
+	SDP_HDR_OWNER_IP6,
+	SDP_HDR_CONNECTION_IP6,
+	SDP_HDR_MEDIA,
 };
 
 extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb,
@@ -36,5 +66,12 @@ extern int ct_sip_lnlen(const char *line, const char *limit);
 extern const char *ct_sip_search(const char *needle, const char *haystack,
 				 size_t needle_len, size_t haystack_len,
 				 int case_sensitive);
+
+extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr,
+				 unsigned int dataoff, unsigned int datalen,
+				 enum sdp_header_types type,
+				 enum sdp_header_types term,
+				 unsigned int *matchoff, unsigned int *matchlen);
+
 #endif /* __KERNEL__ */
 #endif /* __NF_CONNTRACK_SIP_H__ */
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index dd1b2d86deee..aa8a4f492baf 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -147,51 +147,46 @@ static unsigned int ip_nat_sip(struct sk_buff *skb,
 	return NF_ACCEPT;
 }
 
-static unsigned int mangle_sip_packet(struct sk_buff *skb,
-				      const char **dptr, unsigned int *datalen,
-				      char *buffer, int bufflen,
-				      enum sip_header_pos pos)
+static int mangle_content_len(struct sk_buff *skb,
+			      const char **dptr, unsigned int *datalen)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
-	unsigned int matchlen, matchoff;
+	unsigned int matchoff, matchlen;
+	char buffer[sizeof("65536")];
+	int buflen, c_len;
 
+	/* Get actual SDP length */
+	if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen,
+				  SDP_HDR_VERSION, SDP_HDR_UNSPEC,
+				  &matchoff, &matchlen) <= 0)
+		return 0;
+	c_len = *datalen - matchoff + strlen("v=");
+
+	/* Now, update SDP length */
 	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
-			    pos) <= 0)
+			    POS_CONTENT) <= 0)
 		return 0;
 
+	buflen = sprintf(buffer, "%u", c_len);
 	return mangle_packet(skb, dptr, datalen, matchoff, matchlen,
-			     buffer, bufflen);
+			     buffer, buflen);
 }
 
-static int mangle_content_len(struct sk_buff *skb,
-			      const char **dptr, unsigned int *datalen)
+static unsigned mangle_sdp_packet(struct sk_buff *skb,
+				  const char **dptr, unsigned int *datalen,
+				  enum sdp_header_types type,
+				  char *buffer, int buflen)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
-	unsigned int matchoff, matchlen;
-	char buffer[sizeof("65536")];
-	int bufflen;
+	unsigned int matchlen, matchoff;
 
-	/* Get actual SDP length */
-	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff,
-			    &matchlen, POS_SDP_HEADER) > 0) {
-
-		/* since ct_sip_get_info() give us a pointer passing 'v='
-		   we need to add 2 bytes in this count. */
-		int c_len = *datalen - matchoff + 2;
-
-		/* Now, update SDP length */
-		if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff,
-				    &matchlen, POS_CONTENT) > 0) {
-
-			bufflen = sprintf(buffer, "%u", c_len);
-			return mangle_packet(skb, dptr, datalen,
-					     matchoff, matchlen,
-					     buffer, bufflen);
-		}
-	}
-	return 0;
+	if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, type, SDP_HDR_UNSPEC,
+				  &matchoff, &matchlen) <= 0)
+		return 0;
+	return mangle_packet(skb, dptr, datalen, matchoff, matchlen,
+			     buffer, buflen);
 }
 
 static unsigned int mangle_sdp(struct sk_buff *skb,
@@ -205,18 +200,18 @@ static unsigned int mangle_sdp(struct sk_buff *skb,
 
 	/* Mangle owner and contact info. */
 	bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip));
-	if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen,
-			       POS_OWNER_IP4))
+	if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_OWNER_IP4,
+			       buffer, bufflen))
 		return 0;
 
-	if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen,
-			       POS_CONNECTION_IP4))
+	if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_CONNECTION_IP4,
+			       buffer, bufflen))
 		return 0;
 
 	/* Mangle media port. */
 	bufflen = sprintf(buffer, "%u", port);
-	if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen,
-			       POS_MEDIA))
+	if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_MEDIA,
+			       buffer, bufflen))
 		return 0;
 
 	return mangle_content_len(skb, dptr, datalen);
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index cf19a7082a75..801fcb3c749f 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -124,66 +124,6 @@ static const struct sip_header_nfo ct_sip_hdrs[] = {
 		.ln_strlen	= sizeof(":") - 1,
 		.match_len	= skp_digits_len
 	},
-	[POS_MEDIA] = {		/* SDP media info */
-		.case_sensitive	= 1,
-		.lname		= "\nm=",
-		.lnlen		= sizeof("\nm=") - 1,
-		.sname		= "\rm=",
-		.snlen		= sizeof("\rm=") - 1,
-		.ln_str		= "audio ",
-		.ln_strlen	= sizeof("audio ") - 1,
-		.match_len	= digits_len
-	},
-	[POS_OWNER_IP4] = {	/* SDP owner address*/
-		.case_sensitive	= 1,
-		.lname		= "\no=",
-		.lnlen		= sizeof("\no=") - 1,
-		.sname		= "\ro=",
-		.snlen		= sizeof("\ro=") - 1,
-		.ln_str		= "IN IP4 ",
-		.ln_strlen	= sizeof("IN IP4 ") - 1,
-		.match_len	= epaddr_len
-	},
-	[POS_CONNECTION_IP4] = {/* SDP connection info */
-		.case_sensitive	= 1,
-		.lname		= "\nc=",
-		.lnlen		= sizeof("\nc=") - 1,
-		.sname		= "\rc=",
-		.snlen		= sizeof("\rc=") - 1,
-		.ln_str		= "IN IP4 ",
-		.ln_strlen	= sizeof("IN IP4 ") - 1,
-		.match_len	= epaddr_len
-	},
-	[POS_OWNER_IP6] = {	/* SDP owner address*/
-		.case_sensitive	= 1,
-		.lname		= "\no=",
-		.lnlen		= sizeof("\no=") - 1,
-		.sname		= "\ro=",
-		.snlen		= sizeof("\ro=") - 1,
-		.ln_str		= "IN IP6 ",
-		.ln_strlen	= sizeof("IN IP6 ") - 1,
-		.match_len	= epaddr_len
-	},
-	[POS_CONNECTION_IP6] = {/* SDP connection info */
-		.case_sensitive	= 1,
-		.lname		= "\nc=",
-		.lnlen		= sizeof("\nc=") - 1,
-		.sname		= "\rc=",
-		.snlen		= sizeof("\rc=") - 1,
-		.ln_str		= "IN IP6 ",
-		.ln_strlen	= sizeof("IN IP6 ") - 1,
-		.match_len	= epaddr_len
-	},
-	[POS_SDP_HEADER] = { 	/* SDP version header */
-		.case_sensitive	= 1,
-		.lname		= "\nv=",
-		.lnlen		= sizeof("\nv=") - 1,
-		.sname		= "\rv=",
-		.snlen		= sizeof("\rv=") - 1,
-		.ln_str		= "=",
-		.ln_strlen	= sizeof("=") - 1,
-		.match_len	= digits_len
-	}
 };
 
 /* get line length until first CR or LF seen. */
@@ -363,6 +303,92 @@ int ct_sip_get_info(const struct nf_conn *ct,
 }
 EXPORT_SYMBOL_GPL(ct_sip_get_info);
 
+/* SDP header parsing: a SDP session description contains an ordered set of
+ * headers, starting with a section containing general session parameters,
+ * optionally followed by multiple media descriptions.
+ *
+ * SDP headers always start at the beginning of a line. According to RFC 2327:
+ * "The sequence CRLF (0x0d0a) is used to end a record, although parsers should
+ * be tolerant and also accept records terminated with a single newline
+ * character". We handle both cases.
+ */
+static const struct sip_header ct_sdp_hdrs[] = {
+	[SDP_HDR_VERSION]		= SDP_HDR("v=", NULL, digits_len),
+	[SDP_HDR_OWNER_IP4]		= SDP_HDR("o=", "IN IP4 ", epaddr_len),
+	[SDP_HDR_CONNECTION_IP4]	= SDP_HDR("c=", "IN IP4 ", epaddr_len),
+	[SDP_HDR_OWNER_IP6]		= SDP_HDR("o=", "IN IP6 ", epaddr_len),
+	[SDP_HDR_CONNECTION_IP6]	= SDP_HDR("c=", "IN IP6 ", epaddr_len),
+	[SDP_HDR_MEDIA]			= SDP_HDR("m=", "audio ", digits_len),
+};
+
+/* Linear string search within SDP header values */
+static const char *ct_sdp_header_search(const char *dptr, const char *limit,
+					const char *needle, unsigned int len)
+{
+	for (limit -= len; dptr < limit; dptr++) {
+		if (*dptr == '\r' || *dptr == '\n')
+			break;
+		if (strncmp(dptr, needle, len) == 0)
+			return dptr;
+	}
+	return NULL;
+}
+
+/* Locate a SDP header (optionally a substring within the header value),
+ * optionally stopping at the first occurence of the term header, parse
+ * it and return the offset and length of the data we're interested in.
+ */
+int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr,
+			  unsigned int dataoff, unsigned int datalen,
+			  enum sdp_header_types type,
+			  enum sdp_header_types term,
+			  unsigned int *matchoff, unsigned int *matchlen)
+{
+	const struct sip_header *hdr = &ct_sdp_hdrs[type];
+	const struct sip_header *thdr = &ct_sdp_hdrs[term];
+	const char *start = dptr, *limit = dptr + datalen;
+	int shift = 0;
+
+	for (dptr += dataoff; dptr < limit; dptr++) {
+		/* Find beginning of line */
+		if (*dptr != '\r' && *dptr != '\n')
+			continue;
+		if (++dptr >= limit)
+			break;
+		if (*(dptr - 1) == '\r' && *dptr == '\n') {
+			if (++dptr >= limit)
+				break;
+		}
+
+		if (term != SDP_HDR_UNSPEC &&
+		    limit - dptr >= thdr->len &&
+		    strnicmp(dptr, thdr->name, thdr->len) == 0)
+			break;
+		else if (limit - dptr >= hdr->len &&
+			 strnicmp(dptr, hdr->name, hdr->len) == 0)
+			dptr += hdr->len;
+		else
+			continue;
+
+		*matchoff = dptr - start;
+		if (hdr->search) {
+			dptr = ct_sdp_header_search(dptr, limit, hdr->search,
+						    hdr->slen);
+			if (!dptr)
+				return -1;
+			dptr += hdr->slen;
+		}
+
+		*matchlen = hdr->match_len(ct, dptr, limit, &shift);
+		if (!*matchlen)
+			return -1;
+		*matchoff = dptr - start + shift;
+		return 1;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ct_sip_get_sdp_header);
+
 static int set_expected_rtp(struct sk_buff *skb,
 			    const char **dptr, unsigned int *datalen,
 			    union nf_inet_addr *addr, __be16 port)
@@ -408,7 +434,7 @@ static int sip_help(struct sk_buff *skb,
 	int ret = NF_ACCEPT;
 	unsigned int matchoff, matchlen;
 	u_int16_t port;
-	enum sip_header_pos pos;
+	enum sdp_header_types type;
 	typeof(nf_nat_sip_hook) nf_nat_sip;
 
 	/* No Data ? */
@@ -446,8 +472,10 @@ static int sip_help(struct sk_buff *skb,
 		goto out;
 	}
 	/* Get address and port from SDP packet. */
-	pos = family == AF_INET ? POS_CONNECTION_IP4 : POS_CONNECTION_IP6;
-	if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen, pos) > 0) {
+	type = family == AF_INET ? SDP_HDR_CONNECTION_IP4 :
+				   SDP_HDR_CONNECTION_IP6;
+	if (ct_sip_get_sdp_header(ct, dptr, 0, datalen, type, SDP_HDR_UNSPEC,
+				  &matchoff, &matchlen) > 0) {
 
 		/* We'll drop only if there are parse problems. */
 		if (!parse_addr(ct, dptr + matchoff, NULL, &addr,
@@ -455,8 +483,9 @@ static int sip_help(struct sk_buff *skb,
 			ret = NF_DROP;
 			goto out;
 		}
-		if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen,
-				    POS_MEDIA) > 0) {
+		if (ct_sip_get_sdp_header(ct, dptr, 0, datalen,
+					  SDP_HDR_MEDIA, SDP_HDR_UNSPEC,
+					  &matchoff, &matchlen) > 0) {
 
 			port = simple_strtoul(dptr + matchoff, NULL, 10);
 			if (port < 1024) {
-- 
cgit v1.2.3


From ac3677406d4e36e86b1eb5a453997a3b3e0c089a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:18:40 -0700
Subject: [NETFILTER]: nf_conntrack_sip: kill request URI "header" definitions

The request URI is not a header and needs to be treated differently than
real SIP headers. Add a seperate function for parsing it and get rid of
the POS_REQ_URI/POS_REG_REQ_URI definitions.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  5 ++-
 net/ipv4/netfilter/nf_nat_sip.c            | 46 +++++++++++----------
 net/netfilter/nf_conntrack_sip.c           | 64 +++++++++++++++++++++++-------
 3 files changed, 77 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 9131cbc9b9de..480b26f40ce4 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -6,8 +6,6 @@
 #define SIP_TIMEOUT	3600
 
 enum sip_header_pos {
-	POS_REG_REQ_URI,
-	POS_REQ_URI,
 	POS_FROM,
 	POS_TO,
 	POS_VIA,
@@ -59,6 +57,9 @@ extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				       unsigned int *datalen,
 				       struct nf_conntrack_expect *exp);
 
+extern int ct_sip_parse_request(const struct nf_conn *ct,
+				const char *dptr, unsigned int datalen,
+				unsigned int *matchoff, unsigned int *matchlen);
 extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr,
                            size_t dlen, unsigned int *matchoff,
                            unsigned int *matchlen, enum sip_header_pos pos);
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index aa8a4f492baf..60151b5901a5 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -78,20 +78,17 @@ static unsigned int mangle_packet(struct sk_buff *skb,
 	return 1;
 }
 
-static int map_sip_addr(struct sk_buff *skb,
-			const char **dptr, unsigned int *datalen,
-			enum sip_header_pos pos, struct addr_map *map)
+static int map_addr(struct sk_buff *skb,
+		    const char **dptr, unsigned int *datalen,
+		    unsigned int matchoff, unsigned int matchlen,
+		    struct addr_map *map)
 {
 	enum ip_conntrack_info ctinfo;
-	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+	struct nf_conn *ct __maybe_unused = nf_ct_get(skb, &ctinfo);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
-	unsigned int matchlen, matchoff, addrlen;
+	unsigned int addrlen;
 	char *addr;
 
-	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
-			    pos) <= 0)
-		return 1;
-
 	if ((matchlen == map->addr[dir].srciplen ||
 	     matchlen == map->addr[dir].srclen) &&
 	    strncmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) {
@@ -109,13 +106,27 @@ static int map_sip_addr(struct sk_buff *skb,
 			     addr, addrlen);
 }
 
+static int map_sip_addr(struct sk_buff *skb,
+			const char **dptr, unsigned int *datalen,
+			enum sip_header_pos pos, struct addr_map *map)
+{
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+	unsigned int matchlen, matchoff;
+
+	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
+			    pos) <= 0)
+		return 1;
+	return map_addr(skb, dptr, datalen, matchoff, matchlen, map);
+}
+
 static unsigned int ip_nat_sip(struct sk_buff *skb,
 			       const char **dptr, unsigned int *datalen)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
-	enum sip_header_pos pos;
 	struct addr_map map;
+	unsigned int matchoff, matchlen;
 
 	if (*datalen < strlen("SIP/2.0"))
 		return NF_ACCEPT;
@@ -124,18 +135,9 @@ static unsigned int ip_nat_sip(struct sk_buff *skb,
 
 	/* Basic rules: requests and responses. */
 	if (strnicmp(*dptr, "SIP/2.0", strlen("SIP/2.0")) != 0) {
-		/* 10.2: Constructing the REGISTER Request:
-		 *
-		 * The "userinfo" and "@" components of the SIP URI MUST NOT
-		 * be present.
-		 */
-		if (*datalen >= strlen("REGISTER") &&
-		    strnicmp(*dptr, "REGISTER", strlen("REGISTER")) == 0)
-			pos = POS_REG_REQ_URI;
-		else
-			pos = POS_REQ_URI;
-
-		if (!map_sip_addr(skb, dptr, datalen, pos, &map))
+		if (ct_sip_parse_request(ct, *dptr, *datalen,
+					 &matchoff, &matchlen) > 0 &&
+		    !map_addr(skb, dptr, datalen, matchoff, matchlen, &map))
 			return NF_DROP;
 	}
 
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 801fcb3c749f..bb4396155681 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -65,20 +65,6 @@ struct sip_header_nfo {
 };
 
 static const struct sip_header_nfo ct_sip_hdrs[] = {
-	[POS_REG_REQ_URI] = { 	/* SIP REGISTER request URI */
-		.lname		= "sip:",
-		.lnlen		= sizeof("sip:") - 1,
-		.ln_str		= ":",
-		.ln_strlen	= sizeof(":") - 1,
-		.match_len	= epaddr_len,
-	},
-	[POS_REQ_URI] = { 	/* SIP request URI */
-		.lname		= "sip:",
-		.lnlen		= sizeof("sip:") - 1,
-		.ln_str		= "@",
-		.ln_strlen	= sizeof("@") - 1,
-		.match_len	= epaddr_len,
-	},
 	[POS_FROM] = {		/* SIP From header */
 		.lname		= "From:",
 		.lnlen		= sizeof("From:") - 1,
@@ -164,6 +150,18 @@ const char *ct_sip_search(const char *needle, const char *haystack,
 }
 EXPORT_SYMBOL_GPL(ct_sip_search);
 
+static int string_len(const struct nf_conn *ct, const char *dptr,
+		      const char *limit, int *shift)
+{
+	int len = 0;
+
+	while (dptr < limit && isalpha(*dptr)) {
+		dptr++;
+		len++;
+	}
+	return len;
+}
+
 static int digits_len(const struct nf_conn *ct, const char *dptr,
 		      const char *limit, int *shift)
 {
@@ -258,6 +256,44 @@ static int skp_epaddr_len(const struct nf_conn *ct, const char *dptr,
 	return epaddr_len(ct, dptr, limit, shift);
 }
 
+/* Parse a SIP request line of the form:
+ *
+ * Request-Line = Method SP Request-URI SP SIP-Version CRLF
+ *
+ * and return the offset and length of the address contained in the Request-URI.
+ */
+int ct_sip_parse_request(const struct nf_conn *ct,
+			 const char *dptr, unsigned int datalen,
+			 unsigned int *matchoff, unsigned int *matchlen)
+{
+	const char *start = dptr, *limit = dptr + datalen;
+	unsigned int mlen;
+	int shift = 0;
+
+	/* Skip method and following whitespace */
+	mlen = string_len(ct, dptr, limit, NULL);
+	if (!mlen)
+		return 0;
+	dptr += mlen;
+	if (++dptr >= limit)
+		return 0;
+
+	/* Find SIP URI */
+	limit -= strlen("sip:");
+	for (; dptr < limit; dptr++) {
+		if (*dptr == '\r' || *dptr == '\n')
+			return -1;
+		if (strnicmp(dptr, "sip:", strlen("sip:")) == 0)
+			break;
+	}
+	*matchlen = skp_epaddr_len(ct, dptr, limit, &shift);
+	if (!*matchlen)
+		return 0;
+	*matchoff = dptr - start + shift;
+	return 1;
+}
+EXPORT_SYMBOL_GPL(ct_sip_parse_request);
+
 /* Returns 0 if not found, -1 error parsing. */
 int ct_sip_get_info(const struct nf_conn *ct,
 		    const char *dptr, size_t dlen,
-- 
cgit v1.2.3


From ea45f12a2766dae54e5426a23e8f4bafdbe2782e Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:18:57 -0700
Subject: [NETFILTER]: nf_conntrack_sip: parse SIP headers properly

Introduce new function for SIP header parsing that properly deals with
continuation lines and whitespace in headers and use it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  30 ++--
 net/ipv4/netfilter/nf_nat_sip.c            |  18 +-
 net/netfilter/nf_conntrack_sip.c           | 271 ++++++++++++++---------------
 3 files changed, 151 insertions(+), 168 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 480b26f40ce4..ccc701422963 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -5,14 +5,6 @@
 #define SIP_PORT	5060
 #define SIP_TIMEOUT	3600
 
-enum sip_header_pos {
-	POS_FROM,
-	POS_TO,
-	POS_VIA,
-	POS_CONTACT,
-	POS_CONTENT,
-};
-
 struct sip_header {
 	const char	*name;
 	const char	*cname;
@@ -36,9 +28,20 @@ struct sip_header {
 	.match_len	= (__match),					\
 }
 
+#define SIP_HDR(__name, __cname, __search, __match) \
+	__SIP_HDR(__name, __cname, __search, __match)
+
 #define SDP_HDR(__name, __search, __match) \
 	__SIP_HDR(__name, NULL, __search, __match)
 
+enum sip_header_types {
+	SIP_HDR_FROM,
+	SIP_HDR_TO,
+	SIP_HDR_CONTACT,
+	SIP_HDR_VIA,
+	SIP_HDR_CONTENT_LENGTH,
+};
+
 enum sdp_header_types {
 	SDP_HDR_UNSPEC,
 	SDP_HDR_VERSION,
@@ -60,13 +63,10 @@ extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 extern int ct_sip_parse_request(const struct nf_conn *ct,
 				const char *dptr, unsigned int datalen,
 				unsigned int *matchoff, unsigned int *matchlen);
-extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr,
-                           size_t dlen, unsigned int *matchoff,
-                           unsigned int *matchlen, enum sip_header_pos pos);
-extern int ct_sip_lnlen(const char *line, const char *limit);
-extern const char *ct_sip_search(const char *needle, const char *haystack,
-				 size_t needle_len, size_t haystack_len,
-				 int case_sensitive);
+extern int ct_sip_get_header(const struct nf_conn *ct, const char *dptr,
+			     unsigned int dataoff, unsigned int datalen,
+			     enum sip_header_types type,
+			     unsigned int *matchoff, unsigned int *matchlen);
 
 extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr,
 				 unsigned int dataoff, unsigned int datalen,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index 60151b5901a5..c13e43862361 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -108,14 +108,14 @@ static int map_addr(struct sk_buff *skb,
 
 static int map_sip_addr(struct sk_buff *skb,
 			const char **dptr, unsigned int *datalen,
-			enum sip_header_pos pos, struct addr_map *map)
+			enum sip_header_types type, struct addr_map *map)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	unsigned int matchlen, matchoff;
 
-	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
-			    pos) <= 0)
+	if (ct_sip_get_header(ct, *dptr, 0, *datalen, type,
+			      &matchoff, &matchlen) <= 0)
 		return 1;
 	return map_addr(skb, dptr, datalen, matchoff, matchlen, map);
 }
@@ -141,10 +141,10 @@ static unsigned int ip_nat_sip(struct sk_buff *skb,
 			return NF_DROP;
 	}
 
-	if (!map_sip_addr(skb, dptr, datalen, POS_FROM, &map) ||
-	    !map_sip_addr(skb, dptr, datalen, POS_TO, &map) ||
-	    !map_sip_addr(skb, dptr, datalen, POS_VIA, &map) ||
-	    !map_sip_addr(skb, dptr, datalen, POS_CONTACT, &map))
+	if (!map_sip_addr(skb, dptr, datalen, SIP_HDR_FROM, &map) ||
+	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_TO, &map) ||
+	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_VIA, &map) ||
+	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_CONTACT, &map))
 		return NF_DROP;
 	return NF_ACCEPT;
 }
@@ -166,8 +166,8 @@ static int mangle_content_len(struct sk_buff *skb,
 	c_len = *datalen - matchoff + strlen("v=");
 
 	/* Now, update SDP length */
-	if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen,
-			    POS_CONTENT) <= 0)
+	if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_CONTENT_LENGTH,
+			      &matchoff, &matchlen) <= 0)
 		return 0;
 
 	buflen = sprintf(buffer, "%u", c_len);
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index bb4396155681..cbc91598acee 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -47,109 +47,6 @@ unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				struct nf_conntrack_expect *exp) __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sdp_hook);
 
-static int digits_len(const struct nf_conn *, const char *, const char *, int *);
-static int epaddr_len(const struct nf_conn *, const char *, const char *, int *);
-static int skp_digits_len(const struct nf_conn *, const char *, const char *, int *);
-static int skp_epaddr_len(const struct nf_conn *, const char *, const char *, int *);
-
-struct sip_header_nfo {
-	const char	*lname;
-	const char	*sname;
-	const char	*ln_str;
-	size_t		lnlen;
-	size_t		snlen;
-	size_t		ln_strlen;
-	int		case_sensitive;
-	int		(*match_len)(const struct nf_conn *, const char *,
-				     const char *, int *);
-};
-
-static const struct sip_header_nfo ct_sip_hdrs[] = {
-	[POS_FROM] = {		/* SIP From header */
-		.lname		= "From:",
-		.lnlen		= sizeof("From:") - 1,
-		.sname		= "\r\nf:",
-		.snlen		= sizeof("\r\nf:") - 1,
-		.ln_str		= "sip:",
-		.ln_strlen	= sizeof("sip:") - 1,
-		.match_len	= skp_epaddr_len,
-	},
-	[POS_TO] = {		/* SIP To header */
-		.lname		= "To:",
-		.lnlen		= sizeof("To:") - 1,
-		.sname		= "\r\nt:",
-		.snlen		= sizeof("\r\nt:") - 1,
-		.ln_str		= "sip:",
-		.ln_strlen	= sizeof("sip:") - 1,
-		.match_len	= skp_epaddr_len
-	},
-	[POS_VIA] = { 		/* SIP Via header */
-		.lname		= "Via:",
-		.lnlen		= sizeof("Via:") - 1,
-		.sname		= "\r\nv:",
-		.snlen		= sizeof("\r\nv:") - 1, /* rfc3261 "\r\n" */
-		.ln_str		= "UDP ",
-		.ln_strlen	= sizeof("UDP ") - 1,
-		.match_len	= epaddr_len,
-	},
-	[POS_CONTACT] = { 	/* SIP Contact header */
-		.lname		= "Contact:",
-		.lnlen		= sizeof("Contact:") - 1,
-		.sname		= "\r\nm:",
-		.snlen		= sizeof("\r\nm:") - 1,
-		.ln_str		= "sip:",
-		.ln_strlen	= sizeof("sip:") - 1,
-		.match_len	= skp_epaddr_len
-	},
-	[POS_CONTENT] = { 	/* SIP Content length header */
-		.lname		= "Content-Length:",
-		.lnlen		= sizeof("Content-Length:") - 1,
-		.sname		= "\r\nl:",
-		.snlen		= sizeof("\r\nl:") - 1,
-		.ln_str		= ":",
-		.ln_strlen	= sizeof(":") - 1,
-		.match_len	= skp_digits_len
-	},
-};
-
-/* get line length until first CR or LF seen. */
-int ct_sip_lnlen(const char *line, const char *limit)
-{
-	const char *k = line;
-
-	while ((line < limit) && (*line == '\r' || *line == '\n'))
-		line++;
-
-	while (line < limit) {
-		if (*line == '\r' || *line == '\n')
-			break;
-		line++;
-	}
-	return line - k;
-}
-EXPORT_SYMBOL_GPL(ct_sip_lnlen);
-
-/* Linear string search, case sensitive. */
-const char *ct_sip_search(const char *needle, const char *haystack,
-			  size_t needle_len, size_t haystack_len,
-			  int case_sensitive)
-{
-	const char *limit = haystack + (haystack_len - needle_len);
-
-	while (haystack < limit) {
-		if (case_sensitive) {
-			if (strncmp(haystack, needle, needle_len) == 0)
-				return haystack;
-		} else {
-			if (strnicmp(haystack, needle, needle_len) == 0)
-				return haystack;
-		}
-		haystack++;
-	}
-	return NULL;
-}
-EXPORT_SYMBOL_GPL(ct_sip_search);
-
 static int string_len(const struct nf_conn *ct, const char *dptr,
 		      const char *limit, int *shift)
 {
@@ -173,16 +70,6 @@ static int digits_len(const struct nf_conn *ct, const char *dptr,
 	return len;
 }
 
-/* get digits length, skipping blank spaces. */
-static int skp_digits_len(const struct nf_conn *ct, const char *dptr,
-			  const char *limit, int *shift)
-{
-	for (; dptr < limit && *dptr == ' '; dptr++)
-		(*shift)++;
-
-	return digits_len(ct, dptr, limit, shift);
-}
-
 static int parse_addr(const struct nf_conn *ct, const char *cp,
                       const char **endp, union nf_inet_addr *addr,
                       const char *limit)
@@ -294,50 +181,146 @@ int ct_sip_parse_request(const struct nf_conn *ct,
 }
 EXPORT_SYMBOL_GPL(ct_sip_parse_request);
 
-/* Returns 0 if not found, -1 error parsing. */
-int ct_sip_get_info(const struct nf_conn *ct,
-		    const char *dptr, size_t dlen,
-		    unsigned int *matchoff,
-		    unsigned int *matchlen,
-		    enum sip_header_pos pos)
+/* SIP header parsing: SIP headers are located at the beginning of a line, but
+ * may span several lines, in which case the continuation lines begin with a
+ * whitespace character. RFC 2543 allows lines to be terminated with CR, LF or
+ * CRLF, RFC 3261 allows only CRLF, we support both.
+ *
+ * Headers are followed by (optionally) whitespace, a colon, again (optionally)
+ * whitespace and the values. Whitespace in this context means any amount of
+ * tabs, spaces and continuation lines, which are treated as a single whitespace
+ * character.
+ */
+static const struct sip_header ct_sip_hdrs[] = {
+	[SIP_HDR_FROM]			= SIP_HDR("From", "f", "sip:", skp_epaddr_len),
+	[SIP_HDR_TO]			= SIP_HDR("To", "t", "sip:", skp_epaddr_len),
+	[SIP_HDR_CONTACT]		= SIP_HDR("Contact", "m", "sip:", skp_epaddr_len),
+	[SIP_HDR_VIA]			= SIP_HDR("Via", "v", "UDP ", epaddr_len),
+	[SIP_HDR_CONTENT_LENGTH]	= SIP_HDR("Content-Length", "l", NULL, digits_len),
+};
+
+static const char *sip_follow_continuation(const char *dptr, const char *limit)
 {
-	const struct sip_header_nfo *hnfo = &ct_sip_hdrs[pos];
-	const char *limit, *aux, *k = dptr;
-	int shift = 0;
+	/* Walk past newline */
+	if (++dptr >= limit)
+		return NULL;
+
+	/* Skip '\n' in CR LF */
+	if (*(dptr - 1) == '\r' && *dptr == '\n') {
+		if (++dptr >= limit)
+			return NULL;
+	}
+
+	/* Continuation line? */
+	if (*dptr != ' ' && *dptr != '\t')
+		return NULL;
+
+	/* skip leading whitespace */
+	for (; dptr < limit; dptr++) {
+		if (*dptr != ' ' && *dptr != '\t')
+			break;
+	}
+	return dptr;
+}
+
+static const char *sip_skip_whitespace(const char *dptr, const char *limit)
+{
+	for (; dptr < limit; dptr++) {
+		if (*dptr == ' ')
+			continue;
+		if (*dptr != '\r' && *dptr != '\n')
+			break;
+		dptr = sip_follow_continuation(dptr, limit);
+		if (dptr == NULL)
+			return NULL;
+	}
+	return dptr;
+}
 
-	limit = dptr + (dlen - hnfo->lnlen);
+/* Search within a SIP header value, dealing with continuation lines */
+static const char *ct_sip_header_search(const char *dptr, const char *limit,
+					const char *needle, unsigned int len)
+{
+	for (limit -= len; dptr < limit; dptr++) {
+		if (*dptr == '\r' || *dptr == '\n') {
+			dptr = sip_follow_continuation(dptr, limit);
+			if (dptr == NULL)
+				break;
+			continue;
+		}
 
-	while (dptr < limit) {
-		if ((strncmp(dptr, hnfo->lname, hnfo->lnlen) != 0) &&
-		    (hnfo->sname == NULL ||
-		     strncmp(dptr, hnfo->sname, hnfo->snlen) != 0)) {
-			dptr++;
+		if (strnicmp(dptr, needle, len) == 0)
+			return dptr;
+	}
+	return NULL;
+}
+
+int ct_sip_get_header(const struct nf_conn *ct, const char *dptr,
+		      unsigned int dataoff, unsigned int datalen,
+		      enum sip_header_types type,
+		      unsigned int *matchoff, unsigned int *matchlen)
+{
+	const struct sip_header *hdr = &ct_sip_hdrs[type];
+	const char *start = dptr, *limit = dptr + datalen;
+	int shift = 0;
+
+	for (dptr += dataoff; dptr < limit; dptr++) {
+		/* Find beginning of line */
+		if (*dptr != '\r' && *dptr != '\n')
 			continue;
+		if (++dptr >= limit)
+			break;
+		if (*(dptr - 1) == '\r' && *dptr == '\n') {
+			if (++dptr >= limit)
+				break;
 		}
-		aux = ct_sip_search(hnfo->ln_str, dptr, hnfo->ln_strlen,
-				    ct_sip_lnlen(dptr, limit),
-				    hnfo->case_sensitive);
-		if (!aux) {
-			pr_debug("'%s' not found in '%s'.\n", hnfo->ln_str,
-				 hnfo->lname);
-			return -1;
+
+		/* Skip continuation lines */
+		if (*dptr == ' ' || *dptr == '\t')
+			continue;
+
+		/* Find header. Compact headers must be followed by a
+		 * non-alphabetic character to avoid mismatches. */
+		if (limit - dptr >= hdr->len &&
+		    strnicmp(dptr, hdr->name, hdr->len) == 0)
+			dptr += hdr->len;
+		else if (hdr->cname && limit - dptr >= hdr->clen + 1 &&
+			 strnicmp(dptr, hdr->cname, hdr->clen) == 0 &&
+			 !isalpha(*(dptr + hdr->clen + 1)))
+			dptr += hdr->clen;
+		else
+			continue;
+
+		/* Find and skip colon */
+		dptr = sip_skip_whitespace(dptr, limit);
+		if (dptr == NULL)
+			break;
+		if (*dptr != ':' || ++dptr >= limit)
+			break;
+
+		/* Skip whitespace after colon */
+		dptr = sip_skip_whitespace(dptr, limit);
+		if (dptr == NULL)
+			break;
+
+		*matchoff = dptr - start;
+		if (hdr->search) {
+			dptr = ct_sip_header_search(dptr, limit, hdr->search,
+						    hdr->slen);
+			if (!dptr)
+				return -1;
+			dptr += hdr->slen;
 		}
-		aux += hnfo->ln_strlen;
 
-		*matchlen = hnfo->match_len(ct, aux, limit, &shift);
+		*matchlen = hdr->match_len(ct, dptr, limit, &shift);
 		if (!*matchlen)
 			return -1;
-
-		*matchoff = (aux - k) + shift;
-
-		pr_debug("%s match succeeded! - len: %u\n", hnfo->lname,
-			 *matchlen);
+		*matchoff = dptr - start + shift;
 		return 1;
 	}
-	pr_debug("%s header not found.\n", hnfo->lname);
 	return 0;
 }
-EXPORT_SYMBOL_GPL(ct_sip_get_info);
+EXPORT_SYMBOL_GPL(ct_sip_get_header);
 
 /* SDP header parsing: a SDP session description contains an ordered set of
  * headers, starting with a section containing general session parameters,
-- 
cgit v1.2.3


From 05e3ced297fe755093140e7487e292fb7603316e Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:19:13 -0700
Subject: [NETFILTER]: nf_conntrack_sip: introduce SIP-URI parsing helper

Introduce a helper function to parse a SIP-URI in a header value, optionally
iterating through all headers of this kind.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |   5 ++
 net/netfilter/nf_conntrack_sip.c           | 107 +++++++++++++++++++++++++++++
 2 files changed, 112 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index ccc701422963..87bc6f79efc4 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -67,6 +67,11 @@ extern int ct_sip_get_header(const struct nf_conn *ct, const char *dptr,
 			     unsigned int dataoff, unsigned int datalen,
 			     enum sip_header_types type,
 			     unsigned int *matchoff, unsigned int *matchlen);
+extern int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr,
+				   unsigned int *dataoff, unsigned int datalen,
+				   enum sip_header_types type, int *in_header,
+				   unsigned int *matchoff, unsigned int *matchlen,
+				   union nf_inet_addr *addr, __be16 *port);
 
 extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr,
 				 unsigned int dataoff, unsigned int datalen,
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index cbc91598acee..a74d76a97312 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -190,6 +190,9 @@ EXPORT_SYMBOL_GPL(ct_sip_parse_request);
  * whitespace and the values. Whitespace in this context means any amount of
  * tabs, spaces and continuation lines, which are treated as a single whitespace
  * character.
+ *
+ * Some headers may appear multiple times. A comma seperated list of values is
+ * equivalent to multiple headers.
  */
 static const struct sip_header ct_sip_hdrs[] = {
 	[SIP_HDR_FROM]			= SIP_HDR("From", "f", "sip:", skp_epaddr_len),
@@ -322,6 +325,110 @@ int ct_sip_get_header(const struct nf_conn *ct, const char *dptr,
 }
 EXPORT_SYMBOL_GPL(ct_sip_get_header);
 
+/* Get next header field in a list of comma seperated values */
+static int ct_sip_next_header(const struct nf_conn *ct, const char *dptr,
+			      unsigned int dataoff, unsigned int datalen,
+			      enum sip_header_types type,
+			      unsigned int *matchoff, unsigned int *matchlen)
+{
+	const struct sip_header *hdr = &ct_sip_hdrs[type];
+	const char *start = dptr, *limit = dptr + datalen;
+	int shift = 0;
+
+	dptr += dataoff;
+
+	dptr = ct_sip_header_search(dptr, limit, ",", strlen(","));
+	if (!dptr)
+		return 0;
+
+	dptr = ct_sip_header_search(dptr, limit, hdr->search, hdr->slen);
+	if (!dptr)
+		return 0;
+	dptr += hdr->slen;
+
+	*matchoff = dptr - start;
+	*matchlen = hdr->match_len(ct, dptr, limit, &shift);
+	if (!*matchlen)
+		return -1;
+	*matchoff += shift;
+	return 1;
+}
+
+/* Walk through headers until a parsable one is found or no header of the
+ * given type is left. */
+static int ct_sip_walk_headers(const struct nf_conn *ct, const char *dptr,
+			       unsigned int dataoff, unsigned int datalen,
+			       enum sip_header_types type, int *in_header,
+			       unsigned int *matchoff, unsigned int *matchlen)
+{
+	int ret;
+
+	if (in_header && *in_header) {
+		while (1) {
+			ret = ct_sip_next_header(ct, dptr, dataoff, datalen,
+						 type, matchoff, matchlen);
+			if (ret > 0)
+				return ret;
+			if (ret == 0)
+				break;
+			dataoff += *matchoff;
+		}
+		*in_header = 0;
+	}
+
+	while (1) {
+		ret = ct_sip_get_header(ct, dptr, dataoff, datalen,
+					type, matchoff, matchlen);
+		if (ret > 0)
+			break;
+		if (ret == 0)
+			return ret;
+		dataoff += *matchoff;
+	}
+
+	if (in_header)
+		*in_header = 1;
+	return 1;
+}
+
+/* Locate a SIP header, parse the URI and return the offset and length of
+ * the address as well as the address and port themselves. A stream of
+ * headers can be parsed by handing in a non-NULL datalen and in_header
+ * pointer.
+ */
+int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr,
+			    unsigned int *dataoff, unsigned int datalen,
+			    enum sip_header_types type, int *in_header,
+			    unsigned int *matchoff, unsigned int *matchlen,
+			    union nf_inet_addr *addr, __be16 *port)
+{
+	const char *c, *limit = dptr + datalen;
+	unsigned int p;
+	int ret;
+
+	ret = ct_sip_walk_headers(ct, dptr, dataoff ? *dataoff : 0, datalen,
+				  type, in_header, matchoff, matchlen);
+	WARN_ON(ret < 0);
+	if (ret == 0)
+		return ret;
+
+	if (!parse_addr(ct, dptr + *matchoff, &c, addr, limit))
+		return -1;
+	if (*c == ':') {
+		c++;
+		p = simple_strtoul(c, (char **)&c, 10);
+		if (p < 1024 || p > 65535)
+			return -1;
+		*port = htons(p);
+	} else
+		*port = htons(SIP_PORT);
+
+	if (dataoff)
+		*dataoff = c - dptr;
+	return 1;
+}
+EXPORT_SYMBOL_GPL(ct_sip_parse_header_uri);
+
 /* SDP header parsing: a SDP session description contains an ordered set of
  * headers, starting with a section containing general session parameters,
  * optionally followed by multiple media descriptions.
-- 
cgit v1.2.3


From 624f8b7bba98c27a1464f5f858c4a861d5d3e8d7 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:19:30 -0700
Subject: [NETFILTER]: nf_nat_sip: get rid of text based header translation

Use the URI parsing helper to get the numerical addresses and get rid of the
text based header translation.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |   3 +-
 net/ipv4/netfilter/nf_nat_sip.c            | 100 +++++++++++------------------
 net/netfilter/nf_conntrack_sip.c           |  27 ++++++--
 3 files changed, 62 insertions(+), 68 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 87bc6f79efc4..68a0d6a41733 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -62,7 +62,8 @@ extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 
 extern int ct_sip_parse_request(const struct nf_conn *ct,
 				const char *dptr, unsigned int datalen,
-				unsigned int *matchoff, unsigned int *matchlen);
+				unsigned int *matchoff, unsigned int *matchlen,
+				union nf_inet_addr *addr, __be16 *port);
 extern int ct_sip_get_header(const struct nf_conn *ct, const char *dptr,
 			     unsigned int dataoff, unsigned int datalen,
 			     enum sip_header_types type,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index c13e43862361..5b4a5cd23f39 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -26,39 +26,6 @@ MODULE_AUTHOR("Christian Hentschel <chentschel@arnet.com.ar>");
 MODULE_DESCRIPTION("SIP NAT helper");
 MODULE_ALIAS("ip_nat_sip");
 
-struct addr_map {
-	struct {
-		char		src[sizeof("nnn.nnn.nnn.nnn:nnnnn")];
-		char		dst[sizeof("nnn.nnn.nnn.nnn:nnnnn")];
-		unsigned int	srclen, srciplen;
-		unsigned int	dstlen, dstiplen;
-	} addr[IP_CT_DIR_MAX];
-};
-
-static void addr_map_init(const struct nf_conn *ct, struct addr_map *map)
-{
-	const struct nf_conntrack_tuple *t;
-	enum ip_conntrack_dir dir;
-	unsigned int n;
-
-	for (dir = 0; dir < IP_CT_DIR_MAX; dir++) {
-		t = &ct->tuplehash[dir].tuple;
-
-		n = sprintf(map->addr[dir].src, "%u.%u.%u.%u",
-			    NIPQUAD(t->src.u3.ip));
-		map->addr[dir].srciplen = n;
-		n += sprintf(map->addr[dir].src + n, ":%u",
-			     ntohs(t->src.u.udp.port));
-		map->addr[dir].srclen = n;
-
-		n = sprintf(map->addr[dir].dst, "%u.%u.%u.%u",
-			    NIPQUAD(t->dst.u3.ip));
-		map->addr[dir].dstiplen = n;
-		n += sprintf(map->addr[dir].dst + n, ":%u",
-			     ntohs(t->dst.u.udp.port));
-		map->addr[dir].dstlen = n;
-	}
-}
 
 static unsigned int mangle_packet(struct sk_buff *skb,
 				  const char **dptr, unsigned int *datalen,
@@ -81,43 +48,51 @@ static unsigned int mangle_packet(struct sk_buff *skb,
 static int map_addr(struct sk_buff *skb,
 		    const char **dptr, unsigned int *datalen,
 		    unsigned int matchoff, unsigned int matchlen,
-		    struct addr_map *map)
+		    union nf_inet_addr *addr, __be16 port)
 {
 	enum ip_conntrack_info ctinfo;
-	struct nf_conn *ct __maybe_unused = nf_ct_get(skb, &ctinfo);
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
-	unsigned int addrlen;
-	char *addr;
-
-	if ((matchlen == map->addr[dir].srciplen ||
-	     matchlen == map->addr[dir].srclen) &&
-	    strncmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) {
-		addr    = map->addr[!dir].dst;
-		addrlen = map->addr[!dir].dstlen;
-	} else if ((matchlen == map->addr[dir].dstiplen ||
-		    matchlen == map->addr[dir].dstlen) &&
-		   strncmp(*dptr + matchoff, map->addr[dir].dst, matchlen) == 0) {
-		addr    = map->addr[!dir].src;
-		addrlen = map->addr[!dir].srclen;
+	char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")];
+	unsigned int buflen;
+	__be32 newaddr;
+	__be16 newport;
+
+	if (ct->tuplehash[dir].tuple.src.u3.ip == addr->ip &&
+	    ct->tuplehash[dir].tuple.src.u.udp.port == port) {
+		newaddr = ct->tuplehash[!dir].tuple.dst.u3.ip;
+		newport = ct->tuplehash[!dir].tuple.dst.u.udp.port;
+	} else if (ct->tuplehash[dir].tuple.dst.u3.ip == addr->ip &&
+		   ct->tuplehash[dir].tuple.dst.u.udp.port == port) {
+		newaddr = ct->tuplehash[!dir].tuple.src.u3.ip;
+		newport = ct->tuplehash[!dir].tuple.src.u.udp.port;
 	} else
 		return 1;
 
+	if (newaddr == addr->ip && newport == port)
+		return 1;
+
+	buflen = sprintf(buffer, "%u.%u.%u.%u:%u",
+			 NIPQUAD(newaddr), ntohs(newport));
+
 	return mangle_packet(skb, dptr, datalen, matchoff, matchlen,
-			     addr, addrlen);
+			     buffer, buflen);
 }
 
 static int map_sip_addr(struct sk_buff *skb,
 			const char **dptr, unsigned int *datalen,
-			enum sip_header_types type, struct addr_map *map)
+			enum sip_header_types type)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	unsigned int matchlen, matchoff;
+	union nf_inet_addr addr;
+	__be16 port;
 
-	if (ct_sip_get_header(ct, *dptr, 0, *datalen, type,
-			      &matchoff, &matchlen) <= 0)
+	if (ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen, type, NULL,
+				    &matchoff, &matchlen, &addr, &port) <= 0)
 		return 1;
-	return map_addr(skb, dptr, datalen, matchoff, matchlen, map);
+	return map_addr(skb, dptr, datalen, matchoff, matchlen, &addr, port);
 }
 
 static unsigned int ip_nat_sip(struct sk_buff *skb,
@@ -125,26 +100,27 @@ static unsigned int ip_nat_sip(struct sk_buff *skb,
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
-	struct addr_map map;
 	unsigned int matchoff, matchlen;
+	union nf_inet_addr addr;
+	__be16 port;
 
 	if (*datalen < strlen("SIP/2.0"))
 		return NF_ACCEPT;
 
-	addr_map_init(ct, &map);
-
 	/* Basic rules: requests and responses. */
 	if (strnicmp(*dptr, "SIP/2.0", strlen("SIP/2.0")) != 0) {
 		if (ct_sip_parse_request(ct, *dptr, *datalen,
-					 &matchoff, &matchlen) > 0 &&
-		    !map_addr(skb, dptr, datalen, matchoff, matchlen, &map))
+					 &matchoff, &matchlen,
+					 &addr, &port) > 0 &&
+		    !map_addr(skb, dptr, datalen, matchoff, matchlen,
+			      &addr, port))
 			return NF_DROP;
 	}
 
-	if (!map_sip_addr(skb, dptr, datalen, SIP_HDR_FROM, &map) ||
-	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_TO, &map) ||
-	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_VIA, &map) ||
-	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_CONTACT, &map))
+	if (!map_sip_addr(skb, dptr, datalen, SIP_HDR_FROM) ||
+	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_TO) ||
+	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_VIA) ||
+	    !map_sip_addr(skb, dptr, datalen, SIP_HDR_CONTACT))
 		return NF_DROP;
 	return NF_ACCEPT;
 }
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index a74d76a97312..f20fa2d94c0a 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -151,10 +151,12 @@ static int skp_epaddr_len(const struct nf_conn *ct, const char *dptr,
  */
 int ct_sip_parse_request(const struct nf_conn *ct,
 			 const char *dptr, unsigned int datalen,
-			 unsigned int *matchoff, unsigned int *matchlen)
+			 unsigned int *matchoff, unsigned int *matchlen,
+			 union nf_inet_addr *addr, __be16 *port)
 {
-	const char *start = dptr, *limit = dptr + datalen;
+	const char *start = dptr, *limit = dptr + datalen, *end;
 	unsigned int mlen;
+	unsigned int p;
 	int shift = 0;
 
 	/* Skip method and following whitespace */
@@ -173,10 +175,25 @@ int ct_sip_parse_request(const struct nf_conn *ct,
 		if (strnicmp(dptr, "sip:", strlen("sip:")) == 0)
 			break;
 	}
-	*matchlen = skp_epaddr_len(ct, dptr, limit, &shift);
-	if (!*matchlen)
+	if (!skp_epaddr_len(ct, dptr, limit, &shift))
 		return 0;
-	*matchoff = dptr - start + shift;
+	dptr += shift;
+
+	if (!parse_addr(ct, dptr, &end, addr, limit))
+		return -1;
+	if (end < limit && *end == ':') {
+		end++;
+		p = simple_strtoul(end, (char **)&end, 10);
+		if (p < 1024 || p > 65535)
+			return -1;
+		*port = htons(p);
+	} else
+		*port = htons(SIP_PORT);
+
+	if (end == dptr)
+		return 0;
+	*matchoff = dptr - start;
+	*matchlen = end - dptr;
 	return 1;
 }
 EXPORT_SYMBOL_GPL(ct_sip_parse_request);
-- 
cgit v1.2.3


From 30f33e6dee80c6ded917f978e4f377d1069d519d Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:22:20 -0700
Subject: [NETFILTER]: nf_conntrack_sip: support method specific
 request/response handling

Add support for per-method request/response handlers and perform SDP
parsing for INVITE/UPDATE requests and for all informational and
successful responses.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  20 ++++++
 net/netfilter/nf_conntrack_sip.c           | 107 ++++++++++++++++++++++++++---
 2 files changed, 117 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 68a0d6a41733..da93e80804c2 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -5,6 +5,25 @@
 #define SIP_PORT	5060
 #define SIP_TIMEOUT	3600
 
+struct sip_handler {
+	const char	*method;
+	unsigned int	len;
+	int		(*request)(struct sk_buff *skb,
+				   const char **dptr, unsigned int *datalen,
+				   unsigned int cseq);
+	int		(*response)(struct sk_buff *skb,
+				    const char **dptr, unsigned int *datalen,
+				    unsigned int cseq, unsigned int code);
+};
+
+#define SIP_HANDLER(__method, __request, __response)			\
+{									\
+	.method		= (__method),					\
+	.len		= sizeof(__method) - 1,				\
+	.request	= (__request),					\
+	.response	= (__response),					\
+}
+
 struct sip_header {
 	const char	*name;
 	const char	*cname;
@@ -35,6 +54,7 @@ struct sip_header {
 	__SIP_HDR(__name, NULL, __search, __match)
 
 enum sip_header_types {
+	SIP_HDR_CSEQ,
 	SIP_HDR_FROM,
 	SIP_HDR_TO,
 	SIP_HDR_CONTACT,
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 96bedb52bd4b..1be949febab7 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -212,6 +212,7 @@ EXPORT_SYMBOL_GPL(ct_sip_parse_request);
  * equivalent to multiple headers.
  */
 static const struct sip_header ct_sip_hdrs[] = {
+	[SIP_HDR_CSEQ]			= SIP_HDR("CSeq", NULL, NULL, digits_len),
 	[SIP_HDR_FROM]			= SIP_HDR("From", "f", "sip:", skp_epaddr_len),
 	[SIP_HDR_TO]			= SIP_HDR("To", "t", "sip:", skp_epaddr_len),
 	[SIP_HDR_CONTACT]		= SIP_HDR("Contact", "m", "sip:", skp_epaddr_len),
@@ -566,7 +567,8 @@ static int set_expected_rtp(struct sk_buff *skb,
 }
 
 static int process_sdp(struct sk_buff *skb,
-		       const char **dptr, unsigned int *datalen)
+		       const char **dptr, unsigned int *datalen,
+		       unsigned int cseq)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
@@ -600,6 +602,96 @@ static int process_sdp(struct sk_buff *skb,
 
 	return set_expected_rtp(skb, dptr, datalen, &addr, htons(port));
 }
+static int process_invite_response(struct sk_buff *skb,
+				   const char **dptr, unsigned int *datalen,
+				   unsigned int cseq, unsigned int code)
+{
+	if ((code >= 100 && code <= 199) ||
+	    (code >= 200 && code <= 299))
+		return process_sdp(skb, dptr, datalen, cseq);
+
+	return NF_ACCEPT;
+}
+
+static int process_update_response(struct sk_buff *skb,
+				   const char **dptr, unsigned int *datalen,
+				   unsigned int cseq, unsigned int code)
+{
+	if ((code >= 100 && code <= 199) ||
+	    (code >= 200 && code <= 299))
+		return process_sdp(skb, dptr, datalen, cseq);
+
+	return NF_ACCEPT;
+}
+
+static const struct sip_handler sip_handlers[] = {
+	SIP_HANDLER("INVITE", process_sdp, process_invite_response),
+	SIP_HANDLER("UPDATE", process_sdp, process_update_response),
+};
+
+static int process_sip_response(struct sk_buff *skb,
+				const char **dptr, unsigned int *datalen)
+{
+	static const struct sip_handler *handler;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+	unsigned int matchoff, matchlen;
+	unsigned int code, cseq, dataoff, i;
+
+	if (*datalen < strlen("SIP/2.0 200"))
+		return NF_ACCEPT;
+	code = simple_strtoul(*dptr + strlen("SIP/2.0 "), NULL, 10);
+	if (!code)
+		return NF_DROP;
+
+	if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_CSEQ,
+			      &matchoff, &matchlen) <= 0)
+		return NF_DROP;
+	cseq = simple_strtoul(*dptr + matchoff, NULL, 10);
+	if (!cseq)
+		return NF_DROP;
+	dataoff = matchoff + matchlen + 1;
+
+	for (i = 0; i < ARRAY_SIZE(sip_handlers); i++) {
+		handler = &sip_handlers[i];
+		if (handler->response == NULL)
+			continue;
+		if (*datalen < dataoff + handler->len ||
+		    strnicmp(*dptr + dataoff, handler->method, handler->len))
+			continue;
+		return handler->response(skb, dptr, datalen, cseq, code);
+	}
+	return NF_ACCEPT;
+}
+
+static int process_sip_request(struct sk_buff *skb,
+			       const char **dptr, unsigned int *datalen)
+{
+	static const struct sip_handler *handler;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+	unsigned int matchoff, matchlen;
+	unsigned int cseq, i;
+
+	for (i = 0; i < ARRAY_SIZE(sip_handlers); i++) {
+		handler = &sip_handlers[i];
+		if (handler->request == NULL)
+			continue;
+		if (*datalen < handler->len ||
+		    strnicmp(*dptr, handler->method, handler->len))
+			continue;
+
+		if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_CSEQ,
+				      &matchoff, &matchlen) <= 0)
+			return NF_DROP;
+		cseq = simple_strtoul(*dptr + matchoff, NULL, 10);
+		if (!cseq)
+			return NF_DROP;
+
+		return handler->request(skb, dptr, datalen, cseq);
+	}
+	return NF_ACCEPT;
+}
 
 static int sip_help(struct sk_buff *skb,
 		    unsigned int protoff,
@@ -634,15 +726,10 @@ static int sip_help(struct sk_buff *skb,
 	if (datalen < strlen("SIP/2.0 200"))
 		return NF_ACCEPT;
 
-	/* RTP info only in some SDP pkts */
-	if (strnicmp(dptr, "INVITE", strlen("INVITE")) != 0 &&
-	    strnicmp(dptr, "UPDATE", strlen("UPDATE")) != 0 &&
-	    strnicmp(dptr, "SIP/2.0 180", strlen("SIP/2.0 180")) != 0 &&
-	    strnicmp(dptr, "SIP/2.0 183", strlen("SIP/2.0 183")) != 0 &&
-	    strnicmp(dptr, "SIP/2.0 200", strlen("SIP/2.0 200")) != 0)
-		return NF_ACCEPT;
-
-	return process_sdp(skb, &dptr, &datalen);
+	if (strnicmp(dptr, "SIP/2.0 ", strlen("SIP/2.0 ")) != 0)
+		return process_sip_request(skb, &dptr, &datalen);
+	else
+		return process_sip_response(skb, &dptr, &datalen);
 }
 
 static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly;
-- 
cgit v1.2.3


From 2bbb21168a90c788e12fe722eb66f27e611e7df7 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:24:24 -0700
Subject: [NETFILTER]: nf_conntrack_sip: introduce URI and header parameter
 parsing helpers

Introduce URI and header parameter parsing helpers. These are needed
by the conntrack helper to parse expiration values in Contact: header
parameters and by the NAT helper to properly update the Via-header
rport=, received= and maddr= parameters.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h | 10 ++++++
 net/netfilter/nf_conntrack_sip.c           | 58 ++++++++++++++++++++++++++++++
 2 files changed, 68 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index da93e80804c2..87e402825dba 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -93,6 +93,16 @@ extern int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr,
 				   enum sip_header_types type, int *in_header,
 				   unsigned int *matchoff, unsigned int *matchlen,
 				   union nf_inet_addr *addr, __be16 *port);
+extern int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr,
+				      unsigned int dataoff, unsigned int datalen,
+				      const char *name,
+				      unsigned int *matchoff, unsigned int *matchlen,
+				      union nf_inet_addr *addr);
+extern int ct_sip_parse_numerical_param(const struct nf_conn *ct, const char *dptr,
+					unsigned int off, unsigned int datalen,
+					const char *name,
+					unsigned int *matchoff, unsigned int *matchen,
+					unsigned int *val);
 
 extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr,
 				 unsigned int dataoff, unsigned int datalen,
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 8e7e5b465ffb..126f30842d60 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -448,6 +448,64 @@ int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr,
 }
 EXPORT_SYMBOL_GPL(ct_sip_parse_header_uri);
 
+/* Parse address from header parameter and return address, offset and length */
+int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr,
+			       unsigned int dataoff, unsigned int datalen,
+			       const char *name,
+			       unsigned int *matchoff, unsigned int *matchlen,
+			       union nf_inet_addr *addr)
+{
+	const char *limit = dptr + datalen;
+	const char *start, *end;
+
+	limit = ct_sip_header_search(dptr + dataoff, limit, ",", strlen(","));
+	if (!limit)
+		limit = dptr + datalen;
+
+	start = ct_sip_header_search(dptr + dataoff, limit, name, strlen(name));
+	if (!start)
+		return 0;
+
+	start += strlen(name);
+	if (!parse_addr(ct, start, &end, addr, limit))
+		return 0;
+	*matchoff = start - dptr;
+	*matchlen = end - start;
+	return 1;
+}
+EXPORT_SYMBOL_GPL(ct_sip_parse_address_param);
+
+/* Parse numerical header parameter and return value, offset and length */
+int ct_sip_parse_numerical_param(const struct nf_conn *ct, const char *dptr,
+				 unsigned int dataoff, unsigned int datalen,
+				 const char *name,
+				 unsigned int *matchoff, unsigned int *matchlen,
+				 unsigned int *val)
+{
+	const char *limit = dptr + datalen;
+	const char *start;
+	char *end;
+
+	limit = ct_sip_header_search(dptr + dataoff, limit, ",", strlen(","));
+	if (!limit)
+		limit = dptr + datalen;
+
+	start = ct_sip_header_search(dptr + dataoff, limit, name, strlen(name));
+	if (!start)
+		return 0;
+
+	start += strlen(name);
+	*val = simple_strtoul(start, &end, 0);
+	if (start == end)
+		return 0;
+	if (matchoff && matchlen) {
+		*matchoff = start - dptr;
+		*matchlen = end - start;
+	}
+	return 1;
+}
+EXPORT_SYMBOL_GPL(ct_sip_parse_numerical_param);
+
 /* SDP header parsing: a SDP session description contains an ordered set of
  * headers, starting with a section containing general session parameters,
  * optionally followed by multiple media descriptions.
-- 
cgit v1.2.3


From 0f32a40fc91a9ebbbf66e826ac2a829ab37d9cf8 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:25:13 -0700
Subject: [NETFILTER]: nf_conntrack_sip: create signalling expectations

Create expectations for incoming signalling connections when seeing
a REGISTER request. This is needed when the registrar uses a
different source port number for signalling messages and for receiving
incoming calls from other endpoints than the registrar.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  18 +++
 include/net/netfilter/nf_conntrack.h       |   4 +-
 net/ipv4/netfilter/nf_nat_sip.c            | 111 +++++++++++---
 net/netfilter/nf_conntrack_sip.c           | 232 +++++++++++++++++++++++++++--
 4 files changed, 332 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 87e402825dba..7cc84ed0c5da 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -5,6 +5,17 @@
 #define SIP_PORT	5060
 #define SIP_TIMEOUT	3600
 
+struct nf_ct_sip_master {
+	unsigned int	register_cseq;
+};
+
+enum sip_expectation_classes {
+	SIP_EXPECT_SIGNALLING,
+	SIP_EXPECT_AUDIO,
+	__SIP_EXPECT_MAX
+};
+#define SIP_EXPECT_MAX	(__SIP_EXPECT_MAX - 1)
+
 struct sip_handler {
 	const char	*method;
 	unsigned int	len;
@@ -59,6 +70,7 @@ enum sip_header_types {
 	SIP_HDR_TO,
 	SIP_HDR_CONTACT,
 	SIP_HDR_VIA,
+	SIP_HDR_EXPIRES,
 	SIP_HDR_CONTENT_LENGTH,
 };
 
@@ -75,6 +87,12 @@ enum sdp_header_types {
 extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb,
 				       const char **dptr,
 				       unsigned int *datalen);
+extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb,
+					      const char **dptr,
+					      unsigned int *datalen,
+					      struct nf_conntrack_expect *exp,
+					      unsigned int matchoff,
+					      unsigned int matchlen);
 extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				       const char **dptr,
 				       unsigned int *datalen,
diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 922877133598..4a4f870d2a5e 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -46,6 +46,7 @@ union nf_conntrack_expect_proto {
 #include <linux/netfilter/nf_conntrack_pptp.h>
 #include <linux/netfilter/nf_conntrack_h323.h>
 #include <linux/netfilter/nf_conntrack_sane.h>
+#include <linux/netfilter/nf_conntrack_sip.h>
 
 /* per conntrack: application helper private data */
 union nf_conntrack_help {
@@ -54,6 +55,7 @@ union nf_conntrack_help {
 	struct nf_ct_pptp_master ct_pptp_info;
 	struct nf_ct_h323_master ct_h323_info;
 	struct nf_ct_sane_master ct_sane_info;
+	struct nf_ct_sip_master ct_sip_info;
 };
 
 #include <linux/types.h>
@@ -76,7 +78,7 @@ do {									\
 struct nf_conntrack_helper;
 
 /* Must be kept in sync with the classes defined by helpers */
-#define NF_CT_MAX_EXPECT_CLASSES	1
+#define NF_CT_MAX_EXPECT_CLASSES	2
 
 /* nf_conn feature for connections that have a helper */
 struct nf_conn_help {
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index b443618a857f..4b85e21a2a4a 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -205,6 +205,91 @@ next:
 	return NF_ACCEPT;
 }
 
+/* Handles expected signalling connections and media streams */
+static void ip_nat_sip_expected(struct nf_conn *ct,
+				struct nf_conntrack_expect *exp)
+{
+	struct nf_nat_range range;
+
+	/* This must be a fresh one. */
+	BUG_ON(ct->status & IPS_NAT_DONE_MASK);
+
+	/* For DST manip, map port here to where it's expected. */
+	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
+	range.min = range.max = exp->saved_proto;
+	range.min_ip = range.max_ip = exp->saved_ip;
+	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST);
+
+	/* Change src to where master sends to, but only if the connection
+	 * actually came from the same source. */
+	if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip ==
+	    ct->master->tuplehash[exp->dir].tuple.src.u3.ip) {
+		range.flags = IP_NAT_RANGE_MAP_IPS;
+		range.min_ip = range.max_ip
+			= ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip;
+		nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC);
+	}
+}
+
+static unsigned int ip_nat_sip_expect(struct sk_buff *skb,
+				      const char **dptr, unsigned int *datalen,
+				      struct nf_conntrack_expect *exp,
+				      unsigned int matchoff,
+				      unsigned int matchlen)
+{
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+	__be32 newip;
+	u_int16_t port;
+	char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")];
+	unsigned buflen;
+
+	/* Connection will come from reply */
+	if (ct->tuplehash[dir].tuple.src.u3.ip == ct->tuplehash[!dir].tuple.dst.u3.ip)
+		newip = exp->tuple.dst.u3.ip;
+	else
+		newip = ct->tuplehash[!dir].tuple.dst.u3.ip;
+
+	/* If the signalling port matches the connection's source port in the
+	 * original direction, try to use the destination port in the opposite
+	 * direction. */
+	if (exp->tuple.dst.u.udp.port ==
+	    ct->tuplehash[dir].tuple.src.u.udp.port)
+		port = ntohs(ct->tuplehash[!dir].tuple.dst.u.udp.port);
+	else
+		port = ntohs(exp->tuple.dst.u.udp.port);
+
+	exp->saved_ip = exp->tuple.dst.u3.ip;
+	exp->tuple.dst.u3.ip = newip;
+	exp->saved_proto.udp.port = exp->tuple.dst.u.udp.port;
+	exp->dir = !dir;
+	exp->expectfn = ip_nat_sip_expected;
+
+	for (; port != 0; port++) {
+		exp->tuple.dst.u.udp.port = htons(port);
+		if (nf_ct_expect_related(exp) == 0)
+			break;
+	}
+
+	if (port == 0)
+		return NF_DROP;
+
+	if (exp->tuple.dst.u3.ip != exp->saved_ip ||
+	    exp->tuple.dst.u.udp.port != exp->saved_proto.udp.port) {
+		buflen = sprintf(buffer, "%u.%u.%u.%u:%u",
+				 NIPQUAD(newip), port);
+		if (!mangle_packet(skb, dptr, datalen, matchoff, matchlen,
+				   buffer, buflen))
+			goto err;
+	}
+	return NF_ACCEPT;
+
+err:
+	nf_ct_unexpect_related(exp);
+	return NF_DROP;
+}
+
 static int mangle_content_len(struct sk_buff *skb,
 			      const char **dptr, unsigned int *datalen)
 {
@@ -275,27 +360,6 @@ static unsigned int mangle_sdp(struct sk_buff *skb,
 	return mangle_content_len(skb, dptr, datalen);
 }
 
-static void ip_nat_sdp_expect(struct nf_conn *ct,
-			      struct nf_conntrack_expect *exp)
-{
-	struct nf_nat_range range;
-
-	/* This must be a fresh one. */
-	BUG_ON(ct->status & IPS_NAT_DONE_MASK);
-
-	/* For DST manip, map port here to where it's expected. */
-	range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED);
-	range.min = range.max = exp->saved_proto;
-	range.min_ip = range.max_ip = exp->saved_ip;
-	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST);
-
-	/* Change src to where master sends to */
-	range.flags = IP_NAT_RANGE_MAP_IPS;
-	range.min_ip = range.max_ip
-		= ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip;
-	nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC);
-}
-
 /* So, this packet has hit the connection tracking matching code.
    Mangle it, and change the expectation to match the new version. */
 static unsigned int ip_nat_sdp(struct sk_buff *skb,
@@ -322,7 +386,7 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb,
 
 	/* When you see the packet, we need to NAT it the same as the
 	   this one. */
-	exp->expectfn = ip_nat_sdp_expect;
+	exp->expectfn = ip_nat_sip_expected;
 
 	/* Try to get same port: if not, try to change it. */
 	for (port = ntohs(exp->saved_proto.udp.port); port != 0; port++) {
@@ -344,6 +408,7 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb,
 static void __exit nf_nat_sip_fini(void)
 {
 	rcu_assign_pointer(nf_nat_sip_hook, NULL);
+	rcu_assign_pointer(nf_nat_sip_expect_hook, NULL);
 	rcu_assign_pointer(nf_nat_sdp_hook, NULL);
 	synchronize_rcu();
 }
@@ -351,8 +416,10 @@ static void __exit nf_nat_sip_fini(void)
 static int __init nf_nat_sip_init(void)
 {
 	BUG_ON(nf_nat_sip_hook != NULL);
+	BUG_ON(nf_nat_sip_expect_hook != NULL);
 	BUG_ON(nf_nat_sdp_hook != NULL);
 	rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip);
+	rcu_assign_pointer(nf_nat_sip_expect_hook, ip_nat_sip_expect);
 	rcu_assign_pointer(nf_nat_sdp_hook, ip_nat_sdp);
 	return 0;
 }
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 126f30842d60..043aa557e7a8 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -37,11 +37,24 @@ static unsigned int sip_timeout __read_mostly = SIP_TIMEOUT;
 module_param(sip_timeout, uint, 0600);
 MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session");
 
+static int sip_direct_signalling __read_mostly = 1;
+module_param(sip_direct_signalling, int, 0600);
+MODULE_PARM_DESC(sip_direct_signalling, "expect incoming calls from registrar "
+					"only (default 1)");
+
 unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb,
 				const char **dptr,
 				unsigned int *datalen) __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sip_hook);
 
+unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb,
+				       const char **dptr,
+				       unsigned int *datalen,
+				       struct nf_conntrack_expect *exp,
+				       unsigned int matchoff,
+				       unsigned int matchlen) __read_mostly;
+EXPORT_SYMBOL_GPL(nf_nat_sip_expect_hook);
+
 unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				const char **dptr,
 				unsigned int *datalen,
@@ -218,6 +231,7 @@ static const struct sip_header ct_sip_hdrs[] = {
 	[SIP_HDR_TO]			= SIP_HDR("To", "t", "sip:", skp_epaddr_len),
 	[SIP_HDR_CONTACT]		= SIP_HDR("Contact", "m", "sip:", skp_epaddr_len),
 	[SIP_HDR_VIA]			= SIP_HDR("Via", "v", "UDP ", epaddr_len),
+	[SIP_HDR_EXPIRES]		= SIP_HDR("Expires", NULL, NULL, digits_len),
 	[SIP_HDR_CONTENT_LENGTH]	= SIP_HDR("Content-Length", "l", NULL, digits_len),
 };
 
@@ -592,18 +606,50 @@ int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr,
 }
 EXPORT_SYMBOL_GPL(ct_sip_get_sdp_header);
 
-static void flush_expectations(struct nf_conn *ct)
+static int refresh_signalling_expectation(struct nf_conn *ct,
+					  union nf_inet_addr *addr,
+					  __be16 port,
+					  unsigned int expires)
 {
 	struct nf_conn_help *help = nfct_help(ct);
 	struct nf_conntrack_expect *exp;
 	struct hlist_node *n, *next;
+	int found = 0;
 
 	spin_lock_bh(&nf_conntrack_lock);
 	hlist_for_each_entry_safe(exp, n, next, &help->expectations, lnode) {
+		if (exp->class != SIP_EXPECT_SIGNALLING ||
+		    !nf_inet_addr_cmp(&exp->tuple.dst.u3, addr) ||
+		    exp->tuple.dst.u.udp.port != port)
+			continue;
+		if (!del_timer(&exp->timeout))
+			continue;
+		exp->flags &= ~NF_CT_EXPECT_INACTIVE;
+		exp->timeout.expires = jiffies + expires * HZ;
+		add_timer(&exp->timeout);
+		found = 1;
+		break;
+	}
+	spin_unlock_bh(&nf_conntrack_lock);
+	return found;
+}
+
+static void flush_expectations(struct nf_conn *ct, bool media)
+{
+	struct nf_conn_help *help = nfct_help(ct);
+	struct nf_conntrack_expect *exp;
+	struct hlist_node *n, *next;
+
+	spin_lock_bh(&nf_conntrack_lock);
+	hlist_for_each_entry_safe(exp, n, next, &help->expectations, lnode) {
+		if ((exp->class != SIP_EXPECT_SIGNALLING) ^ media)
+			continue;
 		if (!del_timer(&exp->timeout))
 			continue;
 		nf_ct_unlink_expect(exp);
 		nf_ct_expect_put(exp);
+		if (!media)
+			break;
 	}
 	spin_unlock_bh(&nf_conntrack_lock);
 }
@@ -623,7 +669,7 @@ static int set_expected_rtp(struct sk_buff *skb,
 	exp = nf_ct_expect_alloc(ct);
 	if (exp == NULL)
 		return NF_DROP;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
+	nf_ct_expect_init(exp, SIP_EXPECT_AUDIO, family,
 			  &ct->tuplehash[!dir].tuple.src.u3, addr,
 			  IPPROTO_UDP, NULL, &port);
 
@@ -688,7 +734,7 @@ static int process_invite_response(struct sk_buff *skb,
 	    (code >= 200 && code <= 299))
 		return process_sdp(skb, dptr, datalen, cseq);
 	else {
-		flush_expectations(ct);
+		flush_expectations(ct, true);
 		return NF_ACCEPT;
 	}
 }
@@ -704,7 +750,7 @@ static int process_update_response(struct sk_buff *skb,
 	    (code >= 200 && code <= 299))
 		return process_sdp(skb, dptr, datalen, cseq);
 	else {
-		flush_expectations(ct);
+		flush_expectations(ct, true);
 		return NF_ACCEPT;
 	}
 }
@@ -720,7 +766,7 @@ static int process_prack_response(struct sk_buff *skb,
 	    (code >= 200 && code <= 299))
 		return process_sdp(skb, dptr, datalen, cseq);
 	else {
-		flush_expectations(ct);
+		flush_expectations(ct, true);
 		return NF_ACCEPT;
 	}
 }
@@ -732,7 +778,165 @@ static int process_bye_request(struct sk_buff *skb,
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 
-	flush_expectations(ct);
+	flush_expectations(ct, true);
+	return NF_ACCEPT;
+}
+
+/* Parse a REGISTER request and create a permanent expectation for incoming
+ * signalling connections. The expectation is marked inactive and is activated
+ * when receiving a response indicating success from the registrar.
+ */
+static int process_register_request(struct sk_buff *skb,
+				    const char **dptr, unsigned int *datalen,
+				    unsigned int cseq)
+{
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+	struct nf_conn_help *help = nfct_help(ct);
+	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
+	unsigned int matchoff, matchlen;
+	struct nf_conntrack_expect *exp;
+	union nf_inet_addr *saddr, daddr;
+	__be16 port;
+	unsigned int expires = 0;
+	int ret;
+	typeof(nf_nat_sip_expect_hook) nf_nat_sip_expect;
+
+	/* Expected connections can not register again. */
+	if (ct->status & IPS_EXPECTED)
+		return NF_ACCEPT;
+
+	/* We must check the expiration time: a value of zero signals the
+	 * registrar to release the binding. We'll remove our expectation
+	 * when receiving the new bindings in the response, but we don't
+	 * want to create new ones.
+	 *
+	 * The expiration time may be contained in Expires: header, the
+	 * Contact: header parameters or the URI parameters.
+	 */
+	if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_EXPIRES,
+			      &matchoff, &matchlen) > 0)
+		expires = simple_strtoul(*dptr + matchoff, NULL, 10);
+
+	ret = ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen,
+				      SIP_HDR_CONTACT, NULL,
+				      &matchoff, &matchlen, &daddr, &port);
+	if (ret < 0)
+		return NF_DROP;
+	else if (ret == 0)
+		return NF_ACCEPT;
+
+	/* We don't support third-party registrations */
+	if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.src.u3, &daddr))
+		return NF_ACCEPT;
+
+	if (ct_sip_parse_numerical_param(ct, *dptr,
+					 matchoff + matchlen, *datalen,
+					 "expires=", NULL, NULL, &expires) < 0)
+		return NF_DROP;
+
+	if (expires == 0) {
+		ret = NF_ACCEPT;
+		goto store_cseq;
+	}
+
+	exp = nf_ct_expect_alloc(ct);
+	if (!exp)
+		return NF_DROP;
+
+	saddr = NULL;
+	if (sip_direct_signalling)
+		saddr = &ct->tuplehash[!dir].tuple.src.u3;
+
+	nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, family, saddr, &daddr,
+			  IPPROTO_UDP, NULL, &port);
+	exp->timeout.expires = sip_timeout * HZ;
+	exp->helper = nfct_help(ct)->helper;
+	exp->flags = NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE;
+
+	nf_nat_sip_expect = rcu_dereference(nf_nat_sip_expect_hook);
+	if (nf_nat_sip_expect && ct->status & IPS_NAT_MASK)
+		ret = nf_nat_sip_expect(skb, dptr, datalen, exp,
+					matchoff, matchlen);
+	else {
+		if (nf_ct_expect_related(exp) != 0)
+			ret = NF_DROP;
+		else
+			ret = NF_ACCEPT;
+	}
+	nf_ct_expect_put(exp);
+
+store_cseq:
+	if (ret == NF_ACCEPT)
+		help->help.ct_sip_info.register_cseq = cseq;
+	return ret;
+}
+
+static int process_register_response(struct sk_buff *skb,
+				     const char **dptr, unsigned int *datalen,
+				     unsigned int cseq, unsigned int code)
+{
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
+	struct nf_conn_help *help = nfct_help(ct);
+	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+	union nf_inet_addr addr;
+	__be16 port;
+	unsigned int matchoff, matchlen, dataoff = 0;
+	unsigned int expires = 0;
+	int in_contact = 0, ret;
+
+	/* According to RFC 3261, "UAs MUST NOT send a new registration until
+	 * they have received a final response from the registrar for the
+	 * previous one or the previous REGISTER request has timed out".
+	 *
+	 * However, some servers fail to detect retransmissions and send late
+	 * responses, so we store the sequence number of the last valid
+	 * request and compare it here.
+	 */
+	if (help->help.ct_sip_info.register_cseq != cseq)
+		return NF_ACCEPT;
+
+	if (code >= 100 && code <= 199)
+		return NF_ACCEPT;
+	if (code < 200 || code > 299)
+		goto flush;
+
+	if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_EXPIRES,
+			      &matchoff, &matchlen) > 0)
+		expires = simple_strtoul(*dptr + matchoff, NULL, 10);
+
+	while (1) {
+		unsigned int c_expires = expires;
+
+		ret = ct_sip_parse_header_uri(ct, *dptr, &dataoff, *datalen,
+					      SIP_HDR_CONTACT, &in_contact,
+					      &matchoff, &matchlen,
+					      &addr, &port);
+		if (ret < 0)
+			return NF_DROP;
+		else if (ret == 0)
+			break;
+
+		/* We don't support third-party registrations */
+		if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.dst.u3, &addr))
+			continue;
+
+		ret = ct_sip_parse_numerical_param(ct, *dptr,
+						   matchoff + matchlen,
+						   *datalen, "expires=",
+						   NULL, NULL, &c_expires);
+		if (ret < 0)
+			return NF_DROP;
+		if (c_expires == 0)
+			break;
+		if (refresh_signalling_expectation(ct, &addr, port, c_expires))
+			return NF_ACCEPT;
+	}
+
+flush:
+	flush_expectations(ct, false);
 	return NF_ACCEPT;
 }
 
@@ -742,6 +946,7 @@ static const struct sip_handler sip_handlers[] = {
 	SIP_HANDLER("ACK", process_sdp, NULL),
 	SIP_HANDLER("PRACK", process_sdp, process_prack_response),
 	SIP_HANDLER("BYE", process_bye_request, NULL),
+	SIP_HANDLER("REGISTER", process_register_request, process_register_response),
 };
 
 static int process_sip_response(struct sk_buff *skb,
@@ -853,9 +1058,15 @@ static int sip_help(struct sk_buff *skb,
 static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly;
 static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly;
 
-static const struct nf_conntrack_expect_policy sip_exp_policy = {
-	.max_expected	= 2,
-	.timeout	= 3 * 60,
+static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1] = {
+	[SIP_EXPECT_SIGNALLING] = {
+		.max_expected	= 1,
+		.timeout	= 3 * 60,
+	},
+	[SIP_EXPECT_AUDIO] = {
+		.max_expected	= IP_CT_DIR_MAX,
+		.timeout	= 3 * 60,
+	},
 };
 
 static void nf_conntrack_sip_fini(void)
@@ -887,7 +1098,8 @@ static int __init nf_conntrack_sip_init(void)
 		for (j = 0; j < 2; j++) {
 			sip[i][j].tuple.dst.protonum = IPPROTO_UDP;
 			sip[i][j].tuple.src.u.udp.port = htons(ports[i]);
-			sip[i][j].expect_policy = &sip_exp_policy;
+			sip[i][j].expect_policy = sip_exp_policy;
+			sip[i][j].expect_class_max = SIP_EXPECT_MAX;
 			sip[i][j].me = THIS_MODULE;
 			sip[i][j].help = sip_help;
 
-- 
cgit v1.2.3


From a9c1d35917c0c95c8f95a8e497fb91e301419693 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:25:49 -0700
Subject: [NETFILTER]: nf_conntrack_sip: create RTCP expectations

Create expectations for the RTCP connections in addition to RTP connections.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  3 +-
 net/ipv4/netfilter/nf_nat_sip.c            | 42 +++++++++++++---------
 net/netfilter/nf_conntrack_sip.c           | 58 +++++++++++++++++++-----------
 3 files changed, 66 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 7cc84ed0c5da..6ddf95f51fb5 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -96,7 +96,8 @@ extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb,
 extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				       const char **dptr,
 				       unsigned int *datalen,
-				       struct nf_conntrack_expect *exp);
+				       struct nf_conntrack_expect *rtp_exp,
+				       struct nf_conntrack_expect *rtcp_exp);
 
 extern int ct_sip_parse_request(const struct nf_conn *ct,
 				const char *dptr, unsigned int datalen,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index 4b85e21a2a4a..f73ab4883b75 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -364,7 +364,8 @@ static unsigned int mangle_sdp(struct sk_buff *skb,
    Mangle it, and change the expectation to match the new version. */
 static unsigned int ip_nat_sdp(struct sk_buff *skb,
 			       const char **dptr, unsigned int *datalen,
-			       struct nf_conntrack_expect *exp)
+			       struct nf_conntrack_expect *rtp_exp,
+			       struct nf_conntrack_expect *rtcp_exp)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
@@ -375,31 +376,40 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb,
 	/* Connection will come from reply */
 	if (ct->tuplehash[dir].tuple.src.u3.ip ==
 	    ct->tuplehash[!dir].tuple.dst.u3.ip)
-		newip = exp->tuple.dst.u3.ip;
+		newip = rtp_exp->tuple.dst.u3.ip;
 	else
 		newip = ct->tuplehash[!dir].tuple.dst.u3.ip;
 
-	exp->saved_ip = exp->tuple.dst.u3.ip;
-	exp->tuple.dst.u3.ip = newip;
-	exp->saved_proto.udp.port = exp->tuple.dst.u.udp.port;
-	exp->dir = !dir;
-
-	/* When you see the packet, we need to NAT it the same as the
-	   this one. */
-	exp->expectfn = ip_nat_sip_expected;
-
-	/* Try to get same port: if not, try to change it. */
-	for (port = ntohs(exp->saved_proto.udp.port); port != 0; port++) {
-		exp->tuple.dst.u.udp.port = htons(port);
-		if (nf_ct_expect_related(exp) == 0)
+	rtp_exp->saved_ip = rtp_exp->tuple.dst.u3.ip;
+	rtp_exp->tuple.dst.u3.ip = newip;
+	rtp_exp->saved_proto.udp.port = rtp_exp->tuple.dst.u.udp.port;
+	rtp_exp->dir = !dir;
+	rtp_exp->expectfn = ip_nat_sip_expected;
+
+	rtcp_exp->saved_ip = rtcp_exp->tuple.dst.u3.ip;
+	rtcp_exp->tuple.dst.u3.ip = newip;
+	rtcp_exp->saved_proto.udp.port = rtcp_exp->tuple.dst.u.udp.port;
+	rtcp_exp->dir = !dir;
+	rtcp_exp->expectfn = ip_nat_sip_expected;
+
+	/* Try to get same pair of ports: if not, try to change them. */
+	for (port = ntohs(rtp_exp->tuple.dst.u.udp.port);
+	     port != 0; port += 2) {
+		rtp_exp->tuple.dst.u.udp.port = htons(port);
+		if (nf_ct_expect_related(rtp_exp) != 0)
+			continue;
+		rtcp_exp->tuple.dst.u.udp.port = htons(port + 1);
+		if (nf_ct_expect_related(rtcp_exp) == 0)
 			break;
+		nf_ct_unexpect_related(rtp_exp);
 	}
 
 	if (port == 0)
 		return NF_DROP;
 
 	if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr, datalen)) {
-		nf_ct_unexpect_related(exp);
+		nf_ct_unexpect_related(rtp_exp);
+		nf_ct_unexpect_related(rtcp_exp);
 		return NF_DROP;
 	}
 	return NF_ACCEPT;
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 813aa8c67e4c..217262e23403 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -63,7 +63,9 @@ EXPORT_SYMBOL_GPL(nf_nat_sip_expect_hook);
 unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
 				const char **dptr,
 				unsigned int *datalen,
-				struct nf_conntrack_expect *exp) __read_mostly;
+				struct nf_conntrack_expect *rtp_exp,
+				struct nf_conntrack_expect *rtcp_exp)
+				__read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sdp_hook);
 
 static int string_len(const struct nf_conn *ct, const char *dptr,
@@ -659,18 +661,20 @@ static void flush_expectations(struct nf_conn *ct, bool media)
 	spin_unlock_bh(&nf_conntrack_lock);
 }
 
-static int set_expected_rtp(struct sk_buff *skb,
-			    const char **dptr, unsigned int *datalen,
-			    union nf_inet_addr *daddr, __be16 port)
+static int set_expected_rtp_rtcp(struct sk_buff *skb,
+				 const char **dptr, unsigned int *datalen,
+				 union nf_inet_addr *daddr, __be16 port)
 {
-	struct nf_conntrack_expect *exp;
+	struct nf_conntrack_expect *exp, *rtp_exp, *rtcp_exp;
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	union nf_inet_addr *saddr;
 	struct nf_conntrack_tuple tuple;
 	int family = ct->tuplehash[!dir].tuple.src.l3num;
-	int skip_expect = 0, ret;
+	int skip_expect = 0, ret = NF_DROP;
+	u_int16_t base_port;
+	__be16 rtp_port, rtcp_port;
 	typeof(nf_nat_sdp_hook) nf_nat_sdp;
 
 	saddr = NULL;
@@ -704,23 +708,37 @@ static int set_expected_rtp(struct sk_buff *skb,
 	if (skip_expect)
 		return NF_ACCEPT;
 
-	exp = nf_ct_expect_alloc(ct);
-	if (exp == NULL)
-		return NF_DROP;
-	nf_ct_expect_init(exp, SIP_EXPECT_AUDIO, family, saddr, daddr,
-			  IPPROTO_UDP, NULL, &port);
+	base_port = ntohs(tuple.dst.u.udp.port) & ~1;
+	rtp_port = htons(base_port);
+	rtcp_port = htons(base_port + 1);
+
+	rtp_exp = nf_ct_expect_alloc(ct);
+	if (rtp_exp == NULL)
+		goto err1;
+	nf_ct_expect_init(rtp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr,
+			  IPPROTO_UDP, NULL, &rtp_port);
+
+	rtcp_exp = nf_ct_expect_alloc(ct);
+	if (rtcp_exp == NULL)
+		goto err2;
+	nf_ct_expect_init(rtcp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr,
+			  IPPROTO_UDP, NULL, &rtcp_port);
 
 	nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook);
 	if (nf_nat_sdp && ct->status & IPS_NAT_MASK)
-		ret = nf_nat_sdp(skb, dptr, datalen, exp);
+		ret = nf_nat_sdp(skb, dptr, datalen, rtp_exp, rtcp_exp);
 	else {
-		if (nf_ct_expect_related(exp) != 0)
-			ret = NF_DROP;
-		else
-			ret = NF_ACCEPT;
+		if (nf_ct_expect_related(rtp_exp) == 0) {
+			if (nf_ct_expect_related(rtcp_exp) != 0)
+				nf_ct_unexpect_related(rtp_exp);
+			else
+				ret = NF_ACCEPT;
+		}
 	}
-	nf_ct_expect_put(exp);
-
+	nf_ct_expect_put(rtcp_exp);
+err2:
+	nf_ct_expect_put(rtp_exp);
+err1:
 	return ret;
 }
 
@@ -758,7 +776,7 @@ static int process_sdp(struct sk_buff *skb,
 	if (port < 1024 || port > 65535)
 		return NF_DROP;
 
-	return set_expected_rtp(skb, dptr, datalen, &addr, htons(port));
+	return set_expected_rtp_rtcp(skb, dptr, datalen, &addr, htons(port));
 }
 static int process_invite_response(struct sk_buff *skb,
 				   const char **dptr, unsigned int *datalen,
@@ -1101,7 +1119,7 @@ static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1
 		.timeout	= 3 * 60,
 	},
 	[SIP_EXPECT_AUDIO] = {
-		.max_expected	= IP_CT_DIR_MAX,
+		.max_expected	= 2 * IP_CT_DIR_MAX,
 		.timeout	= 3 * 60,
 	},
 };
-- 
cgit v1.2.3


From 4ab9e64e5e3c0516577818804aaf13a630d67bc9 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:26:08 -0700
Subject: [NETFILTER]: nf_nat_sip: split up SDP mangling

The SDP connection addresses may be contained in the payload multiple
times (in the session description and/or once per media description),
currently only the session description is properly updated. Split up
SDP mangling so the function setting up expectations only updates the
media port, update connection addresses from media descriptions while
parsing them and at the end update the session description when the
final addresses are known.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  25 ++++-
 net/ipv4/netfilter/nf_nat_sip.c            | 121 ++++++++++++++++--------
 net/netfilter/nf_conntrack_sip.c           | 142 +++++++++++++++++++++++------
 3 files changed, 219 insertions(+), 69 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 6ddf95f51fb5..eca3ad3f28dc 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -93,11 +93,26 @@ extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb,
 					      struct nf_conntrack_expect *exp,
 					      unsigned int matchoff,
 					      unsigned int matchlen);
-extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
-				       const char **dptr,
-				       unsigned int *datalen,
-				       struct nf_conntrack_expect *rtp_exp,
-				       struct nf_conntrack_expect *rtcp_exp);
+extern unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb,
+					    const char **dptr,
+					    unsigned int dataoff,
+					    unsigned int *datalen,
+					    enum sdp_header_types type,
+					    enum sdp_header_types term,
+					    const union nf_inet_addr *addr);
+extern unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb,
+					       const char **dptr,
+					       unsigned int dataoff,
+					       unsigned int *datalen,
+					       const union nf_inet_addr *addr);
+extern unsigned int (*nf_nat_sdp_media_hook)(struct sk_buff *skb,
+					     const char **dptr,
+					     unsigned int *datalen,
+					     struct nf_conntrack_expect *rtp_exp,
+					     struct nf_conntrack_expect *rtcp_exp,
+					     unsigned int mediaoff,
+					     unsigned int medialen,
+					     union nf_inet_addr *rtp_addr);
 
 extern int ct_sip_parse_request(const struct nf_conn *ct,
 				const char *dptr, unsigned int datalen,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index f73ab4883b75..4429069d9b42 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -316,45 +316,77 @@ static int mangle_content_len(struct sk_buff *skb,
 			     buffer, buflen);
 }
 
-static unsigned mangle_sdp_packet(struct sk_buff *skb,
-				  const char **dptr, unsigned int *datalen,
+static unsigned mangle_sdp_packet(struct sk_buff *skb, const char **dptr,
+				  unsigned int dataoff, unsigned int *datalen,
 				  enum sdp_header_types type,
+				  enum sdp_header_types term,
 				  char *buffer, int buflen)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	unsigned int matchlen, matchoff;
 
-	if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, type, SDP_HDR_UNSPEC,
+	if (ct_sip_get_sdp_header(ct, *dptr, dataoff, *datalen, type, term,
 				  &matchoff, &matchlen) <= 0)
 		return 0;
 	return mangle_packet(skb, dptr, datalen, matchoff, matchlen,
 			     buffer, buflen);
 }
 
-static unsigned int mangle_sdp(struct sk_buff *skb,
-			       enum ip_conntrack_info ctinfo,
-			       struct nf_conn *ct,
-			       __be32 newip, u_int16_t port,
-			       const char **dptr, unsigned int *datalen)
+static unsigned int ip_nat_sdp_addr(struct sk_buff *skb, const char **dptr,
+				    unsigned int dataoff,
+				    unsigned int *datalen,
+				    enum sdp_header_types type,
+				    enum sdp_header_types term,
+				    const union nf_inet_addr *addr)
 {
 	char buffer[sizeof("nnn.nnn.nnn.nnn")];
-	unsigned int bufflen;
+	unsigned int buflen;
 
-	/* Mangle owner and contact info. */
-	bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip));
-	if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_OWNER_IP4,
-			       buffer, bufflen))
+	buflen = sprintf(buffer, NIPQUAD_FMT, NIPQUAD(addr->ip));
+	if (!mangle_sdp_packet(skb, dptr, dataoff, datalen, type, term,
+			       buffer, buflen))
 		return 0;
 
-	if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_CONNECTION_IP4,
-			       buffer, bufflen))
+	return mangle_content_len(skb, dptr, datalen);
+}
+
+static unsigned int ip_nat_sdp_port(struct sk_buff *skb,
+				    const char **dptr,
+				    unsigned int *datalen,
+				    unsigned int matchoff,
+				    unsigned int matchlen,
+				    u_int16_t port)
+{
+	char buffer[sizeof("nnnnn")];
+	unsigned int buflen;
+
+	buflen = sprintf(buffer, "%u", port);
+	if (!mangle_packet(skb, dptr, datalen, matchoff, matchlen,
+			   buffer, buflen))
 		return 0;
 
-	/* Mangle media port. */
-	bufflen = sprintf(buffer, "%u", port);
-	if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_MEDIA,
-			       buffer, bufflen))
+	return mangle_content_len(skb, dptr, datalen);
+}
+
+static unsigned int ip_nat_sdp_session(struct sk_buff *skb, const char **dptr,
+				       unsigned int dataoff,
+				       unsigned int *datalen,
+				       const union nf_inet_addr *addr)
+{
+	char buffer[sizeof("nnn.nnn.nnn.nnn")];
+	unsigned int buflen;
+
+	/* Mangle session description owner and contact addresses */
+	buflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(addr->ip));
+	if (!mangle_sdp_packet(skb, dptr, dataoff, datalen,
+			       SDP_HDR_OWNER_IP4, SDP_HDR_MEDIA,
+			       buffer, buflen))
+		return 0;
+
+	if (!mangle_sdp_packet(skb, dptr, dataoff, datalen,
+			       SDP_HDR_CONNECTION_IP4, SDP_HDR_MEDIA,
+			       buffer, buflen))
 		return 0;
 
 	return mangle_content_len(skb, dptr, datalen);
@@ -362,32 +394,35 @@ static unsigned int mangle_sdp(struct sk_buff *skb,
 
 /* So, this packet has hit the connection tracking matching code.
    Mangle it, and change the expectation to match the new version. */
-static unsigned int ip_nat_sdp(struct sk_buff *skb,
-			       const char **dptr, unsigned int *datalen,
-			       struct nf_conntrack_expect *rtp_exp,
-			       struct nf_conntrack_expect *rtcp_exp)
+static unsigned int ip_nat_sdp_media(struct sk_buff *skb,
+				     const char **dptr,
+				     unsigned int *datalen,
+				     struct nf_conntrack_expect *rtp_exp,
+				     struct nf_conntrack_expect *rtcp_exp,
+				     unsigned int mediaoff,
+				     unsigned int medialen,
+				     union nf_inet_addr *rtp_addr)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
-	__be32 newip;
 	u_int16_t port;
 
 	/* Connection will come from reply */
 	if (ct->tuplehash[dir].tuple.src.u3.ip ==
 	    ct->tuplehash[!dir].tuple.dst.u3.ip)
-		newip = rtp_exp->tuple.dst.u3.ip;
+		rtp_addr->ip = rtp_exp->tuple.dst.u3.ip;
 	else
-		newip = ct->tuplehash[!dir].tuple.dst.u3.ip;
+		rtp_addr->ip = ct->tuplehash[!dir].tuple.dst.u3.ip;
 
 	rtp_exp->saved_ip = rtp_exp->tuple.dst.u3.ip;
-	rtp_exp->tuple.dst.u3.ip = newip;
+	rtp_exp->tuple.dst.u3.ip = rtp_addr->ip;
 	rtp_exp->saved_proto.udp.port = rtp_exp->tuple.dst.u.udp.port;
 	rtp_exp->dir = !dir;
 	rtp_exp->expectfn = ip_nat_sip_expected;
 
 	rtcp_exp->saved_ip = rtcp_exp->tuple.dst.u3.ip;
-	rtcp_exp->tuple.dst.u3.ip = newip;
+	rtcp_exp->tuple.dst.u3.ip = rtp_addr->ip;
 	rtcp_exp->saved_proto.udp.port = rtcp_exp->tuple.dst.u.udp.port;
 	rtcp_exp->dir = !dir;
 	rtcp_exp->expectfn = ip_nat_sip_expected;
@@ -405,21 +440,29 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb,
 	}
 
 	if (port == 0)
-		return NF_DROP;
+		goto err1;
+
+	/* Update media port. */
+	if (rtp_exp->tuple.dst.u.udp.port != rtp_exp->saved_proto.udp.port &&
+	    !ip_nat_sdp_port(skb, dptr, datalen, mediaoff, medialen, port))
+		goto err2;
 
-	if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr, datalen)) {
-		nf_ct_unexpect_related(rtp_exp);
-		nf_ct_unexpect_related(rtcp_exp);
-		return NF_DROP;
-	}
 	return NF_ACCEPT;
+
+err2:
+	nf_ct_unexpect_related(rtp_exp);
+	nf_ct_unexpect_related(rtcp_exp);
+err1:
+	return NF_DROP;
 }
 
 static void __exit nf_nat_sip_fini(void)
 {
 	rcu_assign_pointer(nf_nat_sip_hook, NULL);
 	rcu_assign_pointer(nf_nat_sip_expect_hook, NULL);
-	rcu_assign_pointer(nf_nat_sdp_hook, NULL);
+	rcu_assign_pointer(nf_nat_sdp_addr_hook, NULL);
+	rcu_assign_pointer(nf_nat_sdp_session_hook, NULL);
+	rcu_assign_pointer(nf_nat_sdp_media_hook, NULL);
 	synchronize_rcu();
 }
 
@@ -427,10 +470,14 @@ static int __init nf_nat_sip_init(void)
 {
 	BUG_ON(nf_nat_sip_hook != NULL);
 	BUG_ON(nf_nat_sip_expect_hook != NULL);
-	BUG_ON(nf_nat_sdp_hook != NULL);
+	BUG_ON(nf_nat_sdp_addr_hook != NULL);
+	BUG_ON(nf_nat_sdp_session_hook != NULL);
+	BUG_ON(nf_nat_sdp_media_hook != NULL);
 	rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip);
 	rcu_assign_pointer(nf_nat_sip_expect_hook, ip_nat_sip_expect);
-	rcu_assign_pointer(nf_nat_sdp_hook, ip_nat_sdp);
+	rcu_assign_pointer(nf_nat_sdp_addr_hook, ip_nat_sdp_addr);
+	rcu_assign_pointer(nf_nat_sdp_session_hook, ip_nat_sdp_session);
+	rcu_assign_pointer(nf_nat_sdp_media_hook, ip_nat_sdp_media);
 	return 0;
 }
 
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 217262e23403..f929add324f3 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -60,13 +60,34 @@ unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb,
 				       unsigned int matchlen) __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sip_expect_hook);
 
-unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb,
-				const char **dptr,
-				unsigned int *datalen,
-				struct nf_conntrack_expect *rtp_exp,
-				struct nf_conntrack_expect *rtcp_exp)
-				__read_mostly;
-EXPORT_SYMBOL_GPL(nf_nat_sdp_hook);
+unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb,
+				     const char **dptr,
+				     unsigned int dataoff,
+				     unsigned int *datalen,
+				     enum sdp_header_types type,
+				     enum sdp_header_types term,
+				     const union nf_inet_addr *addr)
+				     __read_mostly;
+EXPORT_SYMBOL_GPL(nf_nat_sdp_addr_hook);
+
+unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb,
+					const char **dptr,
+					unsigned int dataoff,
+					unsigned int *datalen,
+					const union nf_inet_addr *addr)
+					__read_mostly;
+EXPORT_SYMBOL_GPL(nf_nat_sdp_session_hook);
+
+unsigned int (*nf_nat_sdp_media_hook)(struct sk_buff *skb,
+				      const char **dptr,
+				      unsigned int *datalen,
+				      struct nf_conntrack_expect *rtp_exp,
+				      struct nf_conntrack_expect *rtcp_exp,
+				      unsigned int mediaoff,
+				      unsigned int medialen,
+				      union nf_inet_addr *rtp_addr)
+				      __read_mostly;
+EXPORT_SYMBOL_GPL(nf_nat_sdp_media_hook);
 
 static int string_len(const struct nf_conn *ct, const char *dptr,
 		      const char *limit, int *shift)
@@ -613,6 +634,26 @@ int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr,
 }
 EXPORT_SYMBOL_GPL(ct_sip_get_sdp_header);
 
+static int ct_sip_parse_sdp_addr(const struct nf_conn *ct, const char *dptr,
+				 unsigned int dataoff, unsigned int datalen,
+				 enum sdp_header_types type,
+				 enum sdp_header_types term,
+				 unsigned int *matchoff, unsigned int *matchlen,
+				 union nf_inet_addr *addr)
+{
+	int ret;
+
+	ret = ct_sip_get_sdp_header(ct, dptr, dataoff, datalen, type, term,
+				    matchoff, matchlen);
+	if (ret <= 0)
+		return ret;
+
+	if (!parse_addr(ct, dptr + *matchoff, NULL, addr,
+			dptr + *matchoff + *matchlen))
+		return -1;
+	return 1;
+}
+
 static int refresh_signalling_expectation(struct nf_conn *ct,
 					  union nf_inet_addr *addr,
 					  __be16 port,
@@ -663,7 +704,8 @@ static void flush_expectations(struct nf_conn *ct, bool media)
 
 static int set_expected_rtp_rtcp(struct sk_buff *skb,
 				 const char **dptr, unsigned int *datalen,
-				 union nf_inet_addr *daddr, __be16 port)
+				 union nf_inet_addr *daddr, __be16 port,
+				 unsigned int mediaoff, unsigned int medialen)
 {
 	struct nf_conntrack_expect *exp, *rtp_exp, *rtcp_exp;
 	enum ip_conntrack_info ctinfo;
@@ -675,7 +717,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	int skip_expect = 0, ret = NF_DROP;
 	u_int16_t base_port;
 	__be16 rtp_port, rtcp_port;
-	typeof(nf_nat_sdp_hook) nf_nat_sdp;
+	typeof(nf_nat_sdp_media_hook) nf_nat_sdp_media;
 
 	saddr = NULL;
 	if (sip_direct_media) {
@@ -724,9 +766,10 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	nf_ct_expect_init(rtcp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr,
 			  IPPROTO_UDP, NULL, &rtcp_port);
 
-	nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook);
-	if (nf_nat_sdp && ct->status & IPS_NAT_MASK)
-		ret = nf_nat_sdp(skb, dptr, datalen, rtp_exp, rtcp_exp);
+	nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook);
+	if (nf_nat_sdp_media && ct->status & IPS_NAT_MASK)
+		ret = nf_nat_sdp_media(skb, dptr, datalen, rtp_exp, rtcp_exp,
+				       mediaoff, medialen, daddr);
 	else {
 		if (nf_ct_expect_related(rtp_exp) == 0) {
 			if (nf_ct_expect_related(rtcp_exp) != 0)
@@ -750,33 +793,78 @@ static int process_sdp(struct sk_buff *skb,
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	unsigned int matchoff, matchlen;
-	union nf_inet_addr addr;
+	unsigned int mediaoff, medialen;
+	unsigned int sdpoff;
+	unsigned int caddr_len, maddr_len;
+	union nf_inet_addr caddr, maddr, rtp_addr;
 	unsigned int port;
-	enum sdp_header_types type;
+	enum sdp_header_types c_hdr;
+	int ret;
+	typeof(nf_nat_sdp_addr_hook) nf_nat_sdp_addr;
+	typeof(nf_nat_sdp_session_hook) nf_nat_sdp_session;
 
-	/* Get address and port from SDP packet. */
-	type = family == AF_INET ? SDP_HDR_CONNECTION_IP4 :
-				   SDP_HDR_CONNECTION_IP6;
+	c_hdr = family == AF_INET ? SDP_HDR_CONNECTION_IP4 :
+				    SDP_HDR_CONNECTION_IP6;
 
+	/* Find beginning of session description */
 	if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen,
-				  type, SDP_HDR_UNSPEC,
+				  SDP_HDR_VERSION, SDP_HDR_UNSPEC,
 				  &matchoff, &matchlen) <= 0)
 		return NF_ACCEPT;
-
-	/* We'll drop only if there are parse problems. */
-	if (!parse_addr(ct, *dptr + matchoff, NULL, &addr, *dptr + *datalen))
-		return NF_DROP;
-
-	if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen,
+	sdpoff = matchoff;
+
+	/* The connection information is contained in the session description
+	 * and/or once per media description. The first media description marks
+	 * the end of the session description. */
+	caddr_len = 0;
+	if (ct_sip_parse_sdp_addr(ct, *dptr, sdpoff, *datalen,
+				  c_hdr, SDP_HDR_MEDIA,
+				  &matchoff, &matchlen, &caddr) > 0)
+		caddr_len = matchlen;
+
+	if (ct_sip_get_sdp_header(ct, *dptr, sdpoff, *datalen,
 				  SDP_HDR_MEDIA, SDP_HDR_UNSPEC,
-				  &matchoff, &matchlen) <= 0)
+				  &mediaoff, &medialen) <= 0)
 		return NF_ACCEPT;
 
-	port = simple_strtoul(*dptr + matchoff, NULL, 10);
+	port = simple_strtoul(*dptr + mediaoff, NULL, 10);
 	if (port < 1024 || port > 65535)
 		return NF_DROP;
 
-	return set_expected_rtp_rtcp(skb, dptr, datalen, &addr, htons(port));
+	/* The media description overrides the session description. */
+	maddr_len = 0;
+	if (ct_sip_parse_sdp_addr(ct, *dptr, mediaoff, *datalen,
+				  c_hdr, SDP_HDR_MEDIA,
+				  &matchoff, &matchlen, &maddr) > 0) {
+		maddr_len = matchlen;
+		memcpy(&rtp_addr, &maddr, sizeof(rtp_addr));
+	} else if (caddr_len)
+		memcpy(&rtp_addr, &caddr, sizeof(rtp_addr));
+	else
+		return NF_DROP;
+
+	ret = set_expected_rtp_rtcp(skb, dptr, datalen, &rtp_addr, htons(port),
+				    mediaoff, medialen);
+	if (ret != NF_ACCEPT)
+		return ret;
+
+	/* Update media connection address if present */
+	if (maddr_len) {
+		nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook);
+		if (nf_nat_sdp_addr && ct->status & IPS_NAT_MASK) {
+			ret = nf_nat_sdp_addr(skb, dptr, mediaoff, datalen,
+					      c_hdr, SDP_HDR_MEDIA, &rtp_addr);
+			if (ret != NF_ACCEPT)
+				return ret;
+		}
+	}
+
+	/* Update session connection and owner addresses */
+	nf_nat_sdp_session = rcu_dereference(nf_nat_sdp_session_hook);
+	if (nf_nat_sdp_session && ct->status & IPS_NAT_MASK)
+		ret = nf_nat_sdp_session(skb, dptr, sdpoff, datalen, &rtp_addr);
+
+	return ret;
 }
 static int process_invite_response(struct sk_buff *skb,
 				   const char **dptr, unsigned int *datalen,
-- 
cgit v1.2.3


From 0d0ab0378d67517a4f4ae3497706c13d9dd24af1 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:26:24 -0700
Subject: [NETFILTER]: nf_conntrack_sip: support multiple media channels

Add support for multiple media channels and use it to create
expectations for video streams when present.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  14 ++++
 include/net/netfilter/nf_conntrack.h       |   2 +-
 net/netfilter/nf_conntrack_sip.c           | 121 +++++++++++++++++++++--------
 3 files changed, 105 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index eca3ad3f28dc..71fa3eb5f485 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -12,10 +12,24 @@ struct nf_ct_sip_master {
 enum sip_expectation_classes {
 	SIP_EXPECT_SIGNALLING,
 	SIP_EXPECT_AUDIO,
+	SIP_EXPECT_VIDEO,
 	__SIP_EXPECT_MAX
 };
 #define SIP_EXPECT_MAX	(__SIP_EXPECT_MAX - 1)
 
+struct sdp_media_type {
+	const char			*name;
+	unsigned int			len;
+	enum sip_expectation_classes	class;
+};
+
+#define SDP_MEDIA_TYPE(__name, __class)					\
+{									\
+	.name	= (__name),						\
+	.len	= sizeof(__name) - 1,					\
+	.class	= (__class),						\
+}
+
 struct sip_handler {
 	const char	*method;
 	unsigned int	len;
diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 4a4f870d2a5e..a3567a7a6d67 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -78,7 +78,7 @@ do {									\
 struct nf_conntrack_helper;
 
 /* Must be kept in sync with the classes defined by helpers */
-#define NF_CT_MAX_EXPECT_CLASSES	2
+#define NF_CT_MAX_EXPECT_CLASSES	3
 
 /* nf_conn feature for connections that have a helper */
 struct nf_conn_help {
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index f929add324f3..f40a525732d1 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -112,6 +112,21 @@ static int digits_len(const struct nf_conn *ct, const char *dptr,
 	return len;
 }
 
+/* get media type + port length */
+static int media_len(const struct nf_conn *ct, const char *dptr,
+		     const char *limit, int *shift)
+{
+	int len = string_len(ct, dptr, limit, shift);
+
+	dptr += len;
+	if (dptr >= limit || *dptr != ' ')
+		return 0;
+	len++;
+	dptr++;
+
+	return len + digits_len(ct, dptr, limit, shift);
+}
+
 static int parse_addr(const struct nf_conn *ct, const char *cp,
                       const char **endp, union nf_inet_addr *addr,
                       const char *limit)
@@ -563,7 +578,7 @@ static const struct sip_header ct_sdp_hdrs[] = {
 	[SDP_HDR_CONNECTION_IP4]	= SDP_HDR("c=", "IN IP4 ", epaddr_len),
 	[SDP_HDR_OWNER_IP6]		= SDP_HDR("o=", "IN IP6 ", epaddr_len),
 	[SDP_HDR_CONNECTION_IP6]	= SDP_HDR("c=", "IN IP6 ", epaddr_len),
-	[SDP_HDR_MEDIA]			= SDP_HDR("m=", "audio ", digits_len),
+	[SDP_HDR_MEDIA]			= SDP_HDR("m=", NULL, media_len),
 };
 
 /* Linear string search within SDP header values */
@@ -705,6 +720,7 @@ static void flush_expectations(struct nf_conn *ct, bool media)
 static int set_expected_rtp_rtcp(struct sk_buff *skb,
 				 const char **dptr, unsigned int *datalen,
 				 union nf_inet_addr *daddr, __be16 port,
+				 enum sip_expectation_classes class,
 				 unsigned int mediaoff, unsigned int medialen)
 {
 	struct nf_conntrack_expect *exp, *rtp_exp, *rtcp_exp;
@@ -743,7 +759,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	exp = __nf_ct_expect_find(&tuple);
 	if (exp && exp->master != ct &&
 	    nfct_help(exp->master)->helper == nfct_help(ct)->helper &&
-	    exp->class == SIP_EXPECT_AUDIO)
+	    exp->class == class)
 		skip_expect = 1;
 	rcu_read_unlock();
 
@@ -757,13 +773,13 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	rtp_exp = nf_ct_expect_alloc(ct);
 	if (rtp_exp == NULL)
 		goto err1;
-	nf_ct_expect_init(rtp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr,
+	nf_ct_expect_init(rtp_exp, class, family, saddr, daddr,
 			  IPPROTO_UDP, NULL, &rtp_port);
 
 	rtcp_exp = nf_ct_expect_alloc(ct);
 	if (rtcp_exp == NULL)
 		goto err2;
-	nf_ct_expect_init(rtcp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr,
+	nf_ct_expect_init(rtcp_exp, class, family, saddr, daddr,
 			  IPPROTO_UDP, NULL, &rtcp_port);
 
 	nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook);
@@ -785,6 +801,28 @@ err1:
 	return ret;
 }
 
+static const struct sdp_media_type sdp_media_types[] = {
+	SDP_MEDIA_TYPE("audio ", SIP_EXPECT_AUDIO),
+	SDP_MEDIA_TYPE("video ", SIP_EXPECT_VIDEO),
+};
+
+static const struct sdp_media_type *sdp_media_type(const char *dptr,
+						   unsigned int matchoff,
+						   unsigned int matchlen)
+{
+	const struct sdp_media_type *t;
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(sdp_media_types); i++) {
+		t = &sdp_media_types[i];
+		if (matchlen < t->len ||
+		    strncmp(dptr + matchoff, t->name, t->len))
+			continue;
+		return t;
+	}
+	return NULL;
+}
+
 static int process_sdp(struct sk_buff *skb,
 		       const char **dptr, unsigned int *datalen,
 		       unsigned int cseq)
@@ -796,13 +834,16 @@ static int process_sdp(struct sk_buff *skb,
 	unsigned int mediaoff, medialen;
 	unsigned int sdpoff;
 	unsigned int caddr_len, maddr_len;
+	unsigned int i;
 	union nf_inet_addr caddr, maddr, rtp_addr;
 	unsigned int port;
 	enum sdp_header_types c_hdr;
-	int ret;
+	const struct sdp_media_type *t;
+	int ret = NF_ACCEPT;
 	typeof(nf_nat_sdp_addr_hook) nf_nat_sdp_addr;
 	typeof(nf_nat_sdp_session_hook) nf_nat_sdp_session;
 
+	nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook);
 	c_hdr = family == AF_INET ? SDP_HDR_CONNECTION_IP4 :
 				    SDP_HDR_CONNECTION_IP6;
 
@@ -822,41 +863,55 @@ static int process_sdp(struct sk_buff *skb,
 				  &matchoff, &matchlen, &caddr) > 0)
 		caddr_len = matchlen;
 
-	if (ct_sip_get_sdp_header(ct, *dptr, sdpoff, *datalen,
-				  SDP_HDR_MEDIA, SDP_HDR_UNSPEC,
-				  &mediaoff, &medialen) <= 0)
-		return NF_ACCEPT;
+	mediaoff = sdpoff;
+	for (i = 0; i < ARRAY_SIZE(sdp_media_types); ) {
+		if (ct_sip_get_sdp_header(ct, *dptr, mediaoff, *datalen,
+					  SDP_HDR_MEDIA, SDP_HDR_UNSPEC,
+					  &mediaoff, &medialen) <= 0)
+			break;
 
-	port = simple_strtoul(*dptr + mediaoff, NULL, 10);
-	if (port < 1024 || port > 65535)
-		return NF_DROP;
+		/* Get media type and port number. A media port value of zero
+		 * indicates an inactive stream. */
+		t = sdp_media_type(*dptr, mediaoff, medialen);
+		if (!t) {
+			mediaoff += medialen;
+			continue;
+		}
+		mediaoff += t->len;
+		medialen -= t->len;
 
-	/* The media description overrides the session description. */
-	maddr_len = 0;
-	if (ct_sip_parse_sdp_addr(ct, *dptr, mediaoff, *datalen,
-				  c_hdr, SDP_HDR_MEDIA,
-				  &matchoff, &matchlen, &maddr) > 0) {
-		maddr_len = matchlen;
-		memcpy(&rtp_addr, &maddr, sizeof(rtp_addr));
-	} else if (caddr_len)
-		memcpy(&rtp_addr, &caddr, sizeof(rtp_addr));
-	else
-		return NF_DROP;
+		port = simple_strtoul(*dptr + mediaoff, NULL, 10);
+		if (port == 0)
+			continue;
+		if (port < 1024 || port > 65535)
+			return NF_DROP;
 
-	ret = set_expected_rtp_rtcp(skb, dptr, datalen, &rtp_addr, htons(port),
-				    mediaoff, medialen);
-	if (ret != NF_ACCEPT)
-		return ret;
+		/* The media description overrides the session description. */
+		maddr_len = 0;
+		if (ct_sip_parse_sdp_addr(ct, *dptr, mediaoff, *datalen,
+					  c_hdr, SDP_HDR_MEDIA,
+					  &matchoff, &matchlen, &maddr) > 0) {
+			maddr_len = matchlen;
+			memcpy(&rtp_addr, &maddr, sizeof(rtp_addr));
+		} else if (caddr_len)
+			memcpy(&rtp_addr, &caddr, sizeof(rtp_addr));
+		else
+			return NF_DROP;
+
+		ret = set_expected_rtp_rtcp(skb, dptr, datalen,
+					    &rtp_addr, htons(port), t->class,
+					    mediaoff, medialen);
+		if (ret != NF_ACCEPT)
+			return ret;
 
-	/* Update media connection address if present */
-	if (maddr_len) {
-		nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook);
-		if (nf_nat_sdp_addr && ct->status & IPS_NAT_MASK) {
+		/* Update media connection address if present */
+		if (maddr_len && nf_nat_sdp_addr && ct->status & IPS_NAT_MASK) {
 			ret = nf_nat_sdp_addr(skb, dptr, mediaoff, datalen,
 					      c_hdr, SDP_HDR_MEDIA, &rtp_addr);
 			if (ret != NF_ACCEPT)
 				return ret;
 		}
+		i++;
 	}
 
 	/* Update session connection and owner addresses */
@@ -1210,6 +1265,10 @@ static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1
 		.max_expected	= 2 * IP_CT_DIR_MAX,
 		.timeout	= 3 * 60,
 	},
+	[SIP_EXPECT_VIDEO] = {
+		.max_expected	= 2 * IP_CT_DIR_MAX,
+		.timeout	= 3 * 60,
+	},
 };
 
 static void nf_conntrack_sip_fini(void)
-- 
cgit v1.2.3


From c7f485abd618e0d249bdd1abdc586bd10fee1954 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 25 Mar 2008 20:26:43 -0700
Subject: [NETFILTER]: nf_conntrack_sip: RTP routing optimization

Optimize call routing between NATed endpoints: when an external
registrar sends a media description that contains an existing RTP
expectation from a different SNATed connection, the gatekeeper
is trying to route the call directly between the two endpoints.

We assume both endpoints can reach each other directly and
"un-NAT" the addresses, which makes the media stream go between
the two endpoints directly.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_sip.h |  6 +++
 net/ipv4/netfilter/nf_nat_sip.c            |  3 ++
 net/netfilter/nf_conntrack_sip.c           | 59 +++++++++++++++++++++++++-----
 3 files changed, 58 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index 71fa3eb5f485..5da04e586a3f 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -114,6 +114,12 @@ extern unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb,
 					    enum sdp_header_types type,
 					    enum sdp_header_types term,
 					    const union nf_inet_addr *addr);
+extern unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb,
+					    const char **dptr,
+					    unsigned int *datalen,
+					    unsigned int matchoff,
+					    unsigned int matchlen,
+					    u_int16_t port);
 extern unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb,
 					       const char **dptr,
 					       unsigned int dataoff,
diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c
index 4429069d9b42..bcddccddf768 100644
--- a/net/ipv4/netfilter/nf_nat_sip.c
+++ b/net/ipv4/netfilter/nf_nat_sip.c
@@ -461,6 +461,7 @@ static void __exit nf_nat_sip_fini(void)
 	rcu_assign_pointer(nf_nat_sip_hook, NULL);
 	rcu_assign_pointer(nf_nat_sip_expect_hook, NULL);
 	rcu_assign_pointer(nf_nat_sdp_addr_hook, NULL);
+	rcu_assign_pointer(nf_nat_sdp_port_hook, NULL);
 	rcu_assign_pointer(nf_nat_sdp_session_hook, NULL);
 	rcu_assign_pointer(nf_nat_sdp_media_hook, NULL);
 	synchronize_rcu();
@@ -471,11 +472,13 @@ static int __init nf_nat_sip_init(void)
 	BUG_ON(nf_nat_sip_hook != NULL);
 	BUG_ON(nf_nat_sip_expect_hook != NULL);
 	BUG_ON(nf_nat_sdp_addr_hook != NULL);
+	BUG_ON(nf_nat_sdp_port_hook != NULL);
 	BUG_ON(nf_nat_sdp_session_hook != NULL);
 	BUG_ON(nf_nat_sdp_media_hook != NULL);
 	rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip);
 	rcu_assign_pointer(nf_nat_sip_expect_hook, ip_nat_sip_expect);
 	rcu_assign_pointer(nf_nat_sdp_addr_hook, ip_nat_sdp_addr);
+	rcu_assign_pointer(nf_nat_sdp_port_hook, ip_nat_sdp_port);
 	rcu_assign_pointer(nf_nat_sdp_session_hook, ip_nat_sdp_session);
 	rcu_assign_pointer(nf_nat_sdp_media_hook, ip_nat_sdp_media);
 	return 0;
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index f40a525732d1..57de22c770a3 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -70,6 +70,14 @@ unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb,
 				     __read_mostly;
 EXPORT_SYMBOL_GPL(nf_nat_sdp_addr_hook);
 
+unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb,
+				     const char **dptr,
+				     unsigned int *datalen,
+				     unsigned int matchoff,
+				     unsigned int matchlen,
+				     u_int16_t port) __read_mostly;
+EXPORT_SYMBOL_GPL(nf_nat_sdp_port_hook);
+
 unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb,
 					const char **dptr,
 					unsigned int dataoff,
@@ -730,9 +738,10 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	union nf_inet_addr *saddr;
 	struct nf_conntrack_tuple tuple;
 	int family = ct->tuplehash[!dir].tuple.src.l3num;
-	int skip_expect = 0, ret = NF_DROP;
+	int direct_rtp = 0, skip_expect = 0, ret = NF_DROP;
 	u_int16_t base_port;
 	__be16 rtp_port, rtcp_port;
+	typeof(nf_nat_sdp_port_hook) nf_nat_sdp_port;
 	typeof(nf_nat_sdp_media_hook) nf_nat_sdp_media;
 
 	saddr = NULL;
@@ -746,6 +755,14 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	 * to register it since we can see the same media description multiple
 	 * times on different connections in case multiple endpoints receive
 	 * the same call.
+	 *
+	 * RTP optimization: if we find a matching media channel expectation
+	 * and both the expectation and this connection are SNATed, we assume
+	 * both sides can reach each other directly and use the final
+	 * destination address from the expectation. We still need to keep
+	 * the NATed expectations for media that might arrive from the
+	 * outside, and additionally need to expect the direct RTP stream
+	 * in case it passes through us even without NAT.
 	 */
 	memset(&tuple, 0, sizeof(tuple));
 	if (saddr)
@@ -756,20 +773,42 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	tuple.dst.u.udp.port	= port;
 
 	rcu_read_lock();
-	exp = __nf_ct_expect_find(&tuple);
-	if (exp && exp->master != ct &&
-	    nfct_help(exp->master)->helper == nfct_help(ct)->helper &&
-	    exp->class == class)
-		skip_expect = 1;
-	rcu_read_unlock();
+	do {
+		exp = __nf_ct_expect_find(&tuple);
 
-	if (skip_expect)
-		return NF_ACCEPT;
+		if (!exp || exp->master == ct ||
+		    nfct_help(exp->master)->helper != nfct_help(ct)->helper ||
+		    exp->class != class)
+			break;
+
+		if (exp->tuple.src.l3num == AF_INET && !direct_rtp &&
+		    (exp->saved_ip != exp->tuple.dst.u3.ip ||
+		     exp->saved_proto.udp.port != exp->tuple.dst.u.udp.port) &&
+		    ct->status & IPS_NAT_MASK) {
+			daddr->ip		= exp->saved_ip;
+			tuple.dst.u3.ip		= exp->saved_ip;
+			tuple.dst.u.udp.port	= exp->saved_proto.udp.port;
+			direct_rtp = 1;
+		} else
+			skip_expect = 1;
+	} while (!skip_expect);
+	rcu_read_unlock();
 
 	base_port = ntohs(tuple.dst.u.udp.port) & ~1;
 	rtp_port = htons(base_port);
 	rtcp_port = htons(base_port + 1);
 
+	if (direct_rtp) {
+		nf_nat_sdp_port = rcu_dereference(nf_nat_sdp_port_hook);
+		if (nf_nat_sdp_port &&
+		    !nf_nat_sdp_port(skb, dptr, datalen,
+				     mediaoff, medialen, ntohs(rtp_port)))
+			goto err1;
+	}
+
+	if (skip_expect)
+		return NF_ACCEPT;
+
 	rtp_exp = nf_ct_expect_alloc(ct);
 	if (rtp_exp == NULL)
 		goto err1;
@@ -783,7 +822,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 			  IPPROTO_UDP, NULL, &rtcp_port);
 
 	nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook);
-	if (nf_nat_sdp_media && ct->status & IPS_NAT_MASK)
+	if (nf_nat_sdp_media && ct->status & IPS_NAT_MASK && !direct_rtp)
 		ret = nf_nat_sdp_media(skb, dptr, datalen, rtp_exp, rtcp_exp,
 				       mediaoff, medialen, daddr);
 	else {
-- 
cgit v1.2.3


From 9c2f5746b9cd536f0007709196d85a7e7d0070fa Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 26 Mar 2008 00:47:14 -0700
Subject: [NETNS]: Compilation fix for include/linux/netdevice.h.

Commit commit c346dca10840a874240c78efe3f39acf4312a1f2
([NET] NETNS: Omit net_device->nd_net without CONFIG_NET_NS)
breaks compilation with CONFIG_NET_NS set.

Fix the typo.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index d146be40f46c..06ca84d71db4 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -756,7 +756,7 @@ static inline
 void dev_net_set(struct net_device *dev, const struct net *net)
 {
 #ifdef CONFIG_NET_NS
-	dev->nd_dev = net;
+	dev->nd_net = net;
 #endif
 }
 
-- 
cgit v1.2.3


From f5aa23fd49063745f85644dd7a9330acd706add6 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 26 Mar 2008 00:48:17 -0700
Subject: [NETNS]: Compilation warnings under CONFIG_NET_NS.

Recent commits from YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
have been introduced a several compilation warnings
'assignment discards qualifiers from pointer target type'
due to extra const modifier in the inline call parameters of
{dev|sock|twsk}_net_set.

Drop it.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h        | 2 +-
 include/net/inet_timewait_sock.h | 2 +-
 include/net/sock.h               | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 06ca84d71db4..15fa84a15c27 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -753,7 +753,7 @@ struct net *dev_net(const struct net_device *dev)
 }
 
 static inline
-void dev_net_set(struct net_device *dev, const struct net *net)
+void dev_net_set(struct net_device *dev, struct net *net)
 {
 #ifdef CONFIG_NET_NS
 	dev->nd_net = net;
diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 07fe0d1a4f03..95c660c9719b 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -219,7 +219,7 @@ struct net *twsk_net(const struct inet_timewait_sock *twsk)
 }
 
 static inline
-void twsk_net_set(struct inet_timewait_sock *twsk, const struct net *net)
+void twsk_net_set(struct inet_timewait_sock *twsk, struct net *net)
 {
 #ifdef CONFIG_NET_NS
 	twsk->tw_net = net;
diff --git a/include/net/sock.h b/include/net/sock.h
index 7e0d4a0c4d12..1c9d059223ee 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1358,7 +1358,7 @@ struct net *sock_net(const struct sock *sk)
 }
 
 static inline
-void sock_net_set(struct sock *sk, const struct net *net)
+void sock_net_set(struct sock *sk, struct net *net)
 {
 #ifdef CONFIG_NET_NS
 	sk->sk_net = net;
-- 
cgit v1.2.3


From a24022e1887978decaa28fb11d1ddff63e31497f Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 26 Mar 2008 01:55:37 -0700
Subject: [NETNS][ICMP]: Move ICMP sysctls on struct net.

Initialization is moved to icmp_sk_init, all the places, that
refer to them use init_net for now.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h         |  7 ------
 include/net/netns/ipv4.h   |  7 ++++++
 net/ipv4/icmp.c            | 60 ++++++++++++++++++++++++----------------------
 net/ipv4/sysctl_net_ipv4.c | 12 +++++-----
 4 files changed, 44 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index faba64db8ff0..dddb839ff4b5 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -65,11 +65,4 @@ static inline struct raw_sock *raw_sk(const struct sock *sk)
 	return (struct raw_sock *)sk;
 }
 
-extern int sysctl_icmp_echo_ignore_all;
-extern int sysctl_icmp_echo_ignore_broadcasts;
-extern int sysctl_icmp_ignore_bogus_error_responses;
-extern int sysctl_icmp_errors_use_inbound_ifaddr;
-extern int sysctl_icmp_ratelimit;
-extern int sysctl_icmp_ratemask;
-
 #endif	/* _ICMP_H */
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 504fde174525..e3de0ff8ea4b 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -35,5 +35,12 @@ struct netns_ipv4 {
 	struct xt_table		*iptable_raw;
 	struct xt_table		*arptable_filter;
 #endif
+
+	int sysctl_icmp_echo_ignore_all;
+	int sysctl_icmp_echo_ignore_broadcasts;
+	int sysctl_icmp_ignore_bogus_error_responses;
+	int sysctl_icmp_ratelimit;
+	int sysctl_icmp_ratemask;
+	int sysctl_icmp_errors_use_inbound_ifaddr;
 };
 #endif
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index f38f093ef751..958a38445120 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -188,29 +188,6 @@ struct icmp_err icmp_err_convert[] = {
 	},
 };
 
-/* Control parameters for ECHO replies. */
-int sysctl_icmp_echo_ignore_all __read_mostly;
-int sysctl_icmp_echo_ignore_broadcasts __read_mostly = 1;
-
-/* Control parameter - ignore bogus broadcast responses? */
-int sysctl_icmp_ignore_bogus_error_responses __read_mostly = 1;
-
-/*
- * 	Configurable global rate limit.
- *
- *	ratelimit defines tokens/packet consumed for dst->rate_token bucket
- *	ratemask defines which icmp types are ratelimited by setting
- * 	it's bit position.
- *
- *	default:
- *	dest unreachable (3), source quench (4),
- *	time exceeded (11), parameter problem (12)
- */
-
-int sysctl_icmp_ratelimit __read_mostly = 1 * HZ;
-int sysctl_icmp_ratemask __read_mostly = 0x1818;
-int sysctl_icmp_errors_use_inbound_ifaddr __read_mostly;
-
 /*
  *	ICMP control array. This specifies what to do with each ICMP.
  */
@@ -310,8 +287,8 @@ static inline int icmpv4_xrlim_allow(struct rtable *rt, int type, int code)
 		goto out;
 
 	/* Limit if icmp type is enabled in ratemask. */
-	if ((1 << type) & sysctl_icmp_ratemask)
-		rc = xrlim_allow(dst, sysctl_icmp_ratelimit);
+	if ((1 << type) & init_net.ipv4.sysctl_icmp_ratemask)
+		rc = xrlim_allow(dst, init_net.ipv4.sysctl_icmp_ratelimit);
 out:
 	return rc;
 }
@@ -523,7 +500,8 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 	if (!(rt->rt_flags & RTCF_LOCAL)) {
 		struct net_device *dev = NULL;
 
-		if (rt->fl.iif && sysctl_icmp_errors_use_inbound_ifaddr)
+		if (rt->fl.iif &&
+			init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr)
 			dev = dev_get_by_index(net, rt->fl.iif);
 
 		if (dev) {
@@ -745,7 +723,7 @@ static void icmp_unreach(struct sk_buff *skb)
 	 *	get the other vendor to fix their kit.
 	 */
 
-	if (!sysctl_icmp_ignore_bogus_error_responses &&
+	if (!init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses &&
 	    inet_addr_type(net, iph->daddr) == RTN_BROADCAST) {
 		if (net_ratelimit())
 			printk(KERN_WARNING "%u.%u.%u.%u sent an invalid ICMP "
@@ -840,7 +818,7 @@ out_err:
 
 static void icmp_echo(struct sk_buff *skb)
 {
-	if (!sysctl_icmp_echo_ignore_all) {
+	if (!init_net.ipv4.sysctl_icmp_echo_ignore_all) {
 		struct icmp_bxm icmp_param;
 
 		icmp_param.data.icmph	   = *icmp_hdr(skb);
@@ -1051,7 +1029,7 @@ int icmp_rcv(struct sk_buff *skb)
 		 */
 		if ((icmph->type == ICMP_ECHO ||
 		     icmph->type == ICMP_TIMESTAMP) &&
-		    sysctl_icmp_echo_ignore_broadcasts) {
+		    init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts) {
 			goto error;
 		}
 		if (icmph->type != ICMP_ECHO &&
@@ -1195,6 +1173,30 @@ int __net_init icmp_sk_init(struct net *net)
 		 */
 		sk->sk_prot->unhash(sk);
 	}
+
+	/* Control parameters for ECHO replies. */
+	net->ipv4.sysctl_icmp_echo_ignore_all = 0;
+	net->ipv4.sysctl_icmp_echo_ignore_broadcasts = 1;
+
+	/* Control parameter - ignore bogus broadcast responses? */
+	net->ipv4.sysctl_icmp_ignore_bogus_error_responses = 1;
+
+	/*
+	 * 	Configurable global rate limit.
+	 *
+	 *	ratelimit defines tokens/packet consumed for dst->rate_token
+	 *	bucket ratemask defines which icmp types are ratelimited by
+	 *	setting	it's bit position.
+	 *
+	 *	default:
+	 *	dest unreachable (3), source quench (4),
+	 *	time exceeded (11), parameter problem (12)
+	 */
+
+	net->ipv4.sysctl_icmp_ratelimit = 1 * HZ;
+	net->ipv4.sysctl_icmp_ratemask = 0x1818;
+	net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr = 0;
+
 	return 0;
 
 fail:
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index c2fca3024d2e..e9585c02f243 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -406,7 +406,7 @@ static struct ctl_table ipv4_table[] = {
 	{
 		.ctl_name	= NET_IPV4_ICMP_ECHO_IGNORE_ALL,
 		.procname	= "icmp_echo_ignore_all",
-		.data		= &sysctl_icmp_echo_ignore_all,
+		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_all,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -414,7 +414,7 @@ static struct ctl_table ipv4_table[] = {
 	{
 		.ctl_name	= NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS,
 		.procname	= "icmp_echo_ignore_broadcasts",
-		.data		= &sysctl_icmp_echo_ignore_broadcasts,
+		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -422,7 +422,7 @@ static struct ctl_table ipv4_table[] = {
 	{
 		.ctl_name	= NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES,
 		.procname	= "icmp_ignore_bogus_error_responses",
-		.data		= &sysctl_icmp_ignore_bogus_error_responses,
+		.data		= &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -430,7 +430,7 @@ static struct ctl_table ipv4_table[] = {
 	{
 		.ctl_name	= NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR,
 		.procname	= "icmp_errors_use_inbound_ifaddr",
-		.data		= &sysctl_icmp_errors_use_inbound_ifaddr,
+		.data		= &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -588,7 +588,7 @@ static struct ctl_table ipv4_table[] = {
 	{
 		.ctl_name	= NET_IPV4_ICMP_RATELIMIT,
 		.procname	= "icmp_ratelimit",
-		.data		= &sysctl_icmp_ratelimit,
+		.data		= &init_net.ipv4.sysctl_icmp_ratelimit,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -596,7 +596,7 @@ static struct ctl_table ipv4_table[] = {
 	{
 		.ctl_name	= NET_IPV4_ICMP_RATEMASK,
 		.procname	= "icmp_ratemask",
-		.data		= &sysctl_icmp_ratemask,
+		.data		= &init_net.ipv4.sysctl_icmp_ratemask,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
-- 
cgit v1.2.3


From 68528f09980a60c8df046d16336333cac4fc0c32 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 26 Mar 2008 01:56:24 -0700
Subject: [NETNS][ICMP]: Make ctl tables for ICMP sysctls per-net.

Add some flesh to ipv4_sysctl_init_net and ipv4_sysctl_exit_net,
i.e. copy the table, alter .data pointers and register it per-net.

Other ipv4_table's sysctls are now global, but this is going to
change once sysctl permissions patches migrate from -mm tree to
mainline in 2.6.26 merge window :)

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h   |   1 +
 net/ipv4/sysctl_net_ipv4.c | 138 +++++++++++++++++++++++++++++----------------
 2 files changed, 91 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index e3de0ff8ea4b..af685f71f4b5 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -17,6 +17,7 @@ struct netns_ipv4 {
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header	*forw_hdr;
 	struct ctl_table_header	*frags_hdr;
+	struct ctl_table_header	*ipv4_hdr;
 #endif
 	struct ipv4_devconf	*devconf_all;
 	struct ipv4_devconf	*devconf_dflt;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index e9585c02f243..c437f804ee38 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -403,38 +403,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= &ipv4_local_port_range,
 		.strategy	= &ipv4_sysctl_local_port_range,
 	},
-	{
-		.ctl_name	= NET_IPV4_ICMP_ECHO_IGNORE_ALL,
-		.procname	= "icmp_echo_ignore_all",
-		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_all,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS,
-		.procname	= "icmp_echo_ignore_broadcasts",
-		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES,
-		.procname	= "icmp_ignore_bogus_error_responses",
-		.data		= &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR,
-		.procname	= "icmp_errors_use_inbound_ifaddr",
-		.data		= &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
 	{
 		.ctl_name	= NET_IPV4_ROUTE,
 		.procname	= "route",
@@ -585,22 +553,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
-	{
-		.ctl_name	= NET_IPV4_ICMP_RATELIMIT,
-		.procname	= "icmp_ratelimit",
-		.data		= &init_net.ipv4.sysctl_icmp_ratelimit,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
-	{
-		.ctl_name	= NET_IPV4_ICMP_RATEMASK,
-		.procname	= "icmp_ratemask",
-		.data		= &init_net.ipv4.sysctl_icmp_ratemask,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
-	},
 	{
 		.ctl_name	= NET_TCP_TW_REUSE,
 		.procname	= "tcp_tw_reuse",
@@ -804,6 +756,58 @@ static struct ctl_table ipv4_table[] = {
 	{ .ctl_name = 0 }
 };
 
+static struct ctl_table ipv4_net_table[] = {
+	{
+		.ctl_name	= NET_IPV4_ICMP_ECHO_IGNORE_ALL,
+		.procname	= "icmp_echo_ignore_all",
+		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_all,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS,
+		.procname	= "icmp_echo_ignore_broadcasts",
+		.data		= &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES,
+		.procname	= "icmp_ignore_bogus_error_responses",
+		.data		= &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR,
+		.procname	= "icmp_errors_use_inbound_ifaddr",
+		.data		= &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV4_ICMP_RATELIMIT,
+		.procname	= "icmp_ratelimit",
+		.data		= &init_net.ipv4.sysctl_icmp_ratelimit,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= NET_IPV4_ICMP_RATEMASK,
+		.procname	= "icmp_ratemask",
+		.data		= &init_net.ipv4.sysctl_icmp_ratemask,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{ }
+};
+
 struct ctl_path net_ipv4_ctl_path[] = {
 	{ .procname = "net", .ctl_name = CTL_NET, },
 	{ .procname = "ipv4", .ctl_name = NET_IPV4, },
@@ -813,11 +817,49 @@ EXPORT_SYMBOL_GPL(net_ipv4_ctl_path);
 
 static __net_init int ipv4_sysctl_init_net(struct net *net)
 {
+	struct ctl_table *table;
+
+	table = ipv4_net_table;
+	if (net != &init_net) {
+		table = kmemdup(table, sizeof(ipv4_net_table), GFP_KERNEL);
+		if (table == NULL)
+			goto err_alloc;
+
+		table[0].data =
+			&net->ipv4.sysctl_icmp_echo_ignore_all;
+		table[1].data =
+			&net->ipv4.sysctl_icmp_echo_ignore_broadcasts;
+		table[2].data =
+			&net->ipv4.sysctl_icmp_ignore_bogus_error_responses;
+		table[3].data =
+			&net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr;
+		table[4].data =
+			&net->ipv4.sysctl_icmp_ratelimit;
+		table[5].data =
+			&net->ipv4.sysctl_icmp_ratemask;
+	}
+
+	net->ipv4.ipv4_hdr = register_net_sysctl_table(net,
+			net_ipv4_ctl_path, table);
+	if (net->ipv4.ipv4_hdr == NULL)
+		goto err_reg;
+
 	return 0;
+
+err_reg:
+	if (net != &init_net)
+		kfree(table);
+err_alloc:
+	return -ENOMEM;
 }
 
 static __net_exit void ipv4_sysctl_exit_net(struct net *net)
 {
+	struct ctl_table *table;
+
+	table = net->ipv4.ipv4_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->ipv4.ipv4_hdr);
+	kfree(table);
 }
 
 static __net_initdata struct pernet_operations ipv4_sysctl_ops = {
-- 
cgit v1.2.3


From 67727184f28c38d06013c6659560bb046c1d9f9c Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 26 Mar 2008 16:27:22 -0700
Subject: [VLAN]: Reduce memory consumed by vlan_groups

Currently each vlan_groupd contains 8 pointers on arrays with 512
pointers on struct net_device each  :)  Such a construction "in many
cases ... wastes memory".

My proposal is to allow for some of these arrays pointers be NULL,
meaning that there are no devices in it. When a new device is added
to the vlan_group, the appropriate array is allocated.

The check in vlan_group_get_device's is safe, since the pointer
vg->vlan_devices_arrays[x] can only switch from NULL to not-NULL.
The vlan_group_prealloc_vid() is guarded with rtnl lock and is
also safe.

I've checked (I hope that) all the places, that use these arrays
and found, that the register_vlan_dev is the only place, that can
put a vlan device on an empty vlan_group.

Rough calculations shows, that after the patch a setup with a
single vlan dev (or up to 512 vlans with sequential vids) will
occupy approximately 8 times less memory.

The question I have is - does this patch makes sense, or a totally
new structures are required to store the vlan_devs?

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/if_vlan.h |  2 +-
 net/8021q/vlan.c        | 36 +++++++++++++++++++++++-------------
 2 files changed, 24 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 79504b22a932..edd55af7ebd6 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -93,7 +93,7 @@ static inline struct net_device *vlan_group_get_device(struct vlan_group *vg,
 {
 	struct net_device **array;
 	array = vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN];
-	return array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN];
+	return array ? array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN] : NULL;
 }
 
 static inline void vlan_group_set_device(struct vlan_group *vg,
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index c35dc230365c..694be86e4490 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -106,29 +106,35 @@ static void vlan_group_free(struct vlan_group *grp)
 static struct vlan_group *vlan_group_alloc(int ifindex)
 {
 	struct vlan_group *grp;
-	unsigned int size;
-	unsigned int i;
 
 	grp = kzalloc(sizeof(struct vlan_group), GFP_KERNEL);
 	if (!grp)
 		return NULL;
 
-	size = sizeof(struct net_device *) * VLAN_GROUP_ARRAY_PART_LEN;
-
-	for (i = 0; i < VLAN_GROUP_ARRAY_SPLIT_PARTS; i++) {
-		grp->vlan_devices_arrays[i] = kzalloc(size, GFP_KERNEL);
-		if (!grp->vlan_devices_arrays[i])
-			goto err;
-	}
-
 	grp->real_dev_ifindex = ifindex;
 	hlist_add_head_rcu(&grp->hlist,
 			   &vlan_group_hash[vlan_grp_hashfn(ifindex)]);
 	return grp;
+}
 
-err:
-	vlan_group_free(grp);
-	return NULL;
+static int vlan_group_prealloc_vid(struct vlan_group *vg, int vid)
+{
+	struct net_device **array;
+	unsigned int size;
+
+	ASSERT_RTNL();
+
+	array = vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN];
+	if (array != NULL)
+		return 0;
+
+	size = sizeof(struct net_device *) * VLAN_GROUP_ARRAY_PART_LEN;
+	array = kzalloc(size, GFP_KERNEL);
+	if (array == NULL)
+		return -ENOBUFS;
+
+	vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN] = array;
+	return 0;
 }
 
 static void vlan_rcu_free(struct rcu_head *rcu)
@@ -247,6 +253,10 @@ int register_vlan_dev(struct net_device *dev)
 			return -ENOBUFS;
 	}
 
+	err = vlan_group_prealloc_vid(grp, vlan_id);
+	if (err < 0)
+		goto out_free_group;
+
 	err = register_netdevice(dev);
 	if (err < 0)
 		goto out_free_group;
-- 
cgit v1.2.3


From 6ab57e7e7fa316552d0f94eaebf1def1d49f18da Mon Sep 17 00:00:00 2001
From: Daniel Lezcano <dlezcano@fr.ibm.com>
Date: Wed, 26 Mar 2008 16:52:32 -0700
Subject: [NETNS][IPV6] anycast - handle several network namespace

Make use of the network namespace information to have this protocol to
handle several network namespace.

Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h |  3 ++-
 include/net/ipv6.h     | 17 ++++++-----------
 net/ipv6/af_inet6.c    | 12 ++++++------
 net/ipv6/anycast.c     | 38 ++++++++++++++++++++++----------------
 net/ipv6/ndisc.c       |  2 +-
 5 files changed, 37 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index c9276c72764d..d0c47c306046 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -134,7 +134,8 @@ extern int inet6_ac_check(struct sock *sk, struct in6_addr *addr, int ifindex);
 
 extern int ipv6_dev_ac_inc(struct net_device *dev, struct in6_addr *addr);
 extern int __ipv6_dev_ac_dec(struct inet6_dev *idev, struct in6_addr *addr);
-extern int ipv6_chk_acast_addr(struct net_device *dev, struct in6_addr *addr);
+extern int ipv6_chk_acast_addr(struct net *net, struct net_device *dev,
+			       struct in6_addr *addr);
 
 
 /* Device notifier */
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index e82f1814d96b..1c98e737dbd0 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -591,8 +591,8 @@ extern int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 			 int __user *optlen);
 
 #ifdef CONFIG_PROC_FS
-extern int  ac6_proc_init(void);
-extern void ac6_proc_exit(void);
+extern int  ac6_proc_init(struct net *net);
+extern void ac6_proc_exit(struct net *net);
 extern int  raw6_proc_init(void);
 extern void raw6_proc_exit(void);
 extern int  tcp6_proc_init(struct net *net);
@@ -607,15 +607,10 @@ extern int snmp6_register_dev(struct inet6_dev *idev);
 extern int snmp6_unregister_dev(struct inet6_dev *idev);
 
 #else
-static inline int snmp6_register_dev(struct inet6_dev *idev)
-{
-	return 0;
-}
-
-static inline int snmp6_unregister_dev(struct inet6_dev *idev)
-{
-	return 0;
-}
+static inline int ac6_proc_init(struct net *net) { return 0; }
+static inline void ac6_proc_exit(struct net *net) { }
+static inline int snmp6_register_dev(struct inet6_dev *idev) { return 0; }
+static inline int snmp6_unregister_dev(struct inet6_dev *idev) { return 0; }
 #endif
 
 #ifdef CONFIG_SYSCTL
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 12f04e9d3e88..1731b0abf7f5 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -862,11 +862,16 @@ static int inet6_net_init(struct net *net)
 	err = tcp6_proc_init(net);
 	if (err)
 		goto proc_tcp6_fail;
+	err = ac6_proc_init(net);
+	if (err)
+		goto proc_ac6_fail;
 out:
 #endif
 	return err;
 
 #ifdef CONFIG_PROC_FS
+proc_ac6_fail:
+	tcp6_proc_exit(net);
 proc_tcp6_fail:
 	udp6_proc_exit(net);
 	goto out;
@@ -878,6 +883,7 @@ static void inet6_net_exit(struct net *net)
 #ifdef CONFIG_PROC_FS
 	udp6_proc_exit(net);
 	tcp6_proc_exit(net);
+	ac6_proc_exit(net);
 #endif
 }
 
@@ -965,9 +971,6 @@ static int __init inet6_init(void)
 		goto proc_udplite6_fail;
 	if (ipv6_misc_proc_init())
 		goto proc_misc6_fail;
-
-	if (ac6_proc_init())
-		goto proc_anycast6_fail;
 	if (if6_proc_init())
 		goto proc_if6_fail;
 #endif
@@ -1039,8 +1042,6 @@ ip6_route_fail:
 #ifdef CONFIG_PROC_FS
 	if6_proc_exit();
 proc_if6_fail:
-	ac6_proc_exit();
-proc_anycast6_fail:
 	ipv6_misc_proc_exit();
 proc_misc6_fail:
 	udplite6_proc_exit();
@@ -1101,7 +1102,6 @@ static void __exit inet6_exit(void)
 
 	/* Cleanup code parts. */
 	if6_proc_exit();
-	ac6_proc_exit();
 	ipv6_misc_proc_exit();
 	udplite6_proc_exit();
 	raw6_proc_exit();
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index 96868b994b37..463bd95d6b13 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -82,6 +82,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 	struct net_device *dev = NULL;
 	struct inet6_dev *idev;
 	struct ipv6_ac_socklist *pac;
+	struct net *net = sock_net(sk);
 	int	ishost = !ipv6_devconf.forwarding;
 	int	err = 0;
 
@@ -89,7 +90,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 		return -EPERM;
 	if (ipv6_addr_is_multicast(addr))
 		return -EINVAL;
-	if (ipv6_chk_addr(&init_net, addr, NULL, 0))
+	if (ipv6_chk_addr(net, addr, NULL, 0))
 		return -EINVAL;
 
 	pac = sock_kmalloc(sk, sizeof(struct ipv6_ac_socklist), GFP_KERNEL);
@@ -101,7 +102,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 	if (ifindex == 0) {
 		struct rt6_info *rt;
 
-		rt = rt6_lookup(&init_net, addr, NULL, 0, 0);
+		rt = rt6_lookup(net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
@@ -112,10 +113,10 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 		} else {
 			/* router, no matching interface: just pick one */
 
-			dev = dev_get_by_flags(&init_net, IFF_UP, IFF_UP|IFF_LOOPBACK);
+			dev = dev_get_by_flags(net, IFF_UP, IFF_UP|IFF_LOOPBACK);
 		}
 	} else
-		dev = dev_get_by_index(&init_net, ifindex);
+		dev = dev_get_by_index(net, ifindex);
 
 	if (dev == NULL) {
 		err = -ENODEV;
@@ -176,6 +177,7 @@ int ipv6_sock_ac_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct net_device *dev;
 	struct ipv6_ac_socklist *pac, *prev_pac;
+	struct net *net = sock_net(sk);
 
 	write_lock_bh(&ipv6_sk_ac_lock);
 	prev_pac = NULL;
@@ -196,7 +198,7 @@ int ipv6_sock_ac_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
 
 	write_unlock_bh(&ipv6_sk_ac_lock);
 
-	dev = dev_get_by_index(&init_net, pac->acl_ifindex);
+	dev = dev_get_by_index(net, pac->acl_ifindex);
 	if (dev) {
 		ipv6_dev_ac_dec(dev, &pac->acl_addr);
 		dev_put(dev);
@@ -210,6 +212,7 @@ void ipv6_sock_ac_close(struct sock *sk)
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct net_device *dev = NULL;
 	struct ipv6_ac_socklist *pac;
+	struct net *net = sock_net(sk);
 	int	prev_index;
 
 	write_lock_bh(&ipv6_sk_ac_lock);
@@ -224,7 +227,7 @@ void ipv6_sock_ac_close(struct sock *sk)
 		if (pac->acl_ifindex != prev_index) {
 			if (dev)
 				dev_put(dev);
-			dev = dev_get_by_index(&init_net, pac->acl_ifindex);
+			dev = dev_get_by_index(net, pac->acl_ifindex);
 			prev_index = pac->acl_ifindex;
 		}
 		if (dev)
@@ -422,14 +425,15 @@ static int ipv6_chk_acast_dev(struct net_device *dev, struct in6_addr *addr)
 /*
  *	check if given interface (or any, if dev==0) has this anycast address
  */
-int ipv6_chk_acast_addr(struct net_device *dev, struct in6_addr *addr)
+int ipv6_chk_acast_addr(struct net *net, struct net_device *dev,
+			struct in6_addr *addr)
 {
 	int found = 0;
 
 	if (dev)
 		return ipv6_chk_acast_dev(dev, addr);
 	read_lock(&dev_base_lock);
-	for_each_netdev(&init_net, dev)
+	for_each_netdev(net, dev)
 		if (ipv6_chk_acast_dev(dev, addr)) {
 			found = 1;
 			break;
@@ -441,6 +445,7 @@ int ipv6_chk_acast_addr(struct net_device *dev, struct in6_addr *addr)
 
 #ifdef CONFIG_PROC_FS
 struct ac6_iter_state {
+	struct seq_net_private p;
 	struct net_device *dev;
 	struct inet6_dev *idev;
 };
@@ -451,9 +456,10 @@ static inline struct ifacaddr6 *ac6_get_first(struct seq_file *seq)
 {
 	struct ifacaddr6 *im = NULL;
 	struct ac6_iter_state *state = ac6_seq_private(seq);
+	struct net *net = seq_file_net(seq);
 
 	state->idev = NULL;
-	for_each_netdev(&init_net, state->dev) {
+	for_each_netdev(net, state->dev) {
 		struct inet6_dev *idev;
 		idev = in6_dev_get(state->dev);
 		if (!idev)
@@ -551,8 +557,8 @@ static const struct seq_operations ac6_seq_ops = {
 
 static int ac6_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open_private(file, &ac6_seq_ops,
-			sizeof(struct ac6_iter_state));
+	return seq_open_net(inode, file, &ac6_seq_ops,
+			    sizeof(struct ac6_iter_state));
 }
 
 static const struct file_operations ac6_seq_fops = {
@@ -560,20 +566,20 @@ static const struct file_operations ac6_seq_fops = {
 	.open		=	ac6_seq_open,
 	.read		=	seq_read,
 	.llseek		=	seq_lseek,
-	.release	=	seq_release_private,
+	.release	=	seq_release_net,
 };
 
-int __init ac6_proc_init(void)
+int ac6_proc_init(struct net *net)
 {
-	if (!proc_net_fops_create(&init_net, "anycast6", S_IRUGO, &ac6_seq_fops))
+	if (!proc_net_fops_create(net, "anycast6", S_IRUGO, &ac6_seq_fops))
 		return -ENOMEM;
 
 	return 0;
 }
 
-void ac6_proc_exit(void)
+void ac6_proc_exit(struct net *net)
 {
-	proc_net_remove(&init_net, "anycast6");
+	proc_net_remove(net, "anycast6");
 }
 #endif
 
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 79af57f586e8..b4d8e331432e 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -773,7 +773,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 			return;
 		}
 
-		if (ipv6_chk_acast_addr(dev, &msg->target) ||
+		if (ipv6_chk_acast_addr(dev_net(dev), dev, &msg->target) ||
 		    (idev->cnf.forwarding &&
 		     (ipv6_devconf.proxy_ndp || idev->cnf.proxy_ndp) &&
 		     (pneigh = pneigh_lookup(&nd_tbl, dev_net(dev),
-- 
cgit v1.2.3


From 60e8fbc4c53d3ef0cbffa393a9e7b77e2a1bae58 Mon Sep 17 00:00:00 2001
From: Benjamin Thery <benjamin.thery@bull.net>
Date: Wed, 26 Mar 2008 16:53:08 -0700
Subject: [NETNS][IPV6] flowlabels - make flowlabels per namespace

This patch introduces a new member, fl_net, in struct ip6_flowlabel.
This allows to create labels with the same value in different namespaces.

Signed-off-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h       |  1 +
 net/ipv6/ip6_flowlabel.c | 72 +++++++++++++++++++++++++++++++++++++-----------
 2 files changed, 57 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 1c98e737dbd0..296f61d84709 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -202,6 +202,7 @@ struct ip6_flowlabel
 	u32			owner;
 	unsigned long		lastuse;
 	unsigned long		expires;
+	struct net		*fl_net;
 };
 
 #define IPV6_FLOWINFO_MASK	__constant_htonl(0x0FFFFFFF)
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index 2b7d9ee98832..78d1d913e36c 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -62,23 +62,23 @@ static DEFINE_RWLOCK(ip6_fl_lock);
 static DEFINE_RWLOCK(ip6_sk_fl_lock);
 
 
-static __inline__ struct ip6_flowlabel * __fl_lookup(__be32 label)
+static inline struct ip6_flowlabel *__fl_lookup(struct net *net, __be32 label)
 {
 	struct ip6_flowlabel *fl;
 
 	for (fl=fl_ht[FL_HASH(label)]; fl; fl = fl->next) {
-		if (fl->label == label)
+		if (fl->label == label && fl->fl_net == net)
 			return fl;
 	}
 	return NULL;
 }
 
-static struct ip6_flowlabel * fl_lookup(__be32 label)
+static struct ip6_flowlabel *fl_lookup(struct net *net, __be32 label)
 {
 	struct ip6_flowlabel *fl;
 
 	read_lock_bh(&ip6_fl_lock);
-	fl = __fl_lookup(label);
+	fl = __fl_lookup(net, label);
 	if (fl)
 		atomic_inc(&fl->users);
 	read_unlock_bh(&ip6_fl_lock);
@@ -88,8 +88,10 @@ static struct ip6_flowlabel * fl_lookup(__be32 label)
 
 static void fl_free(struct ip6_flowlabel *fl)
 {
-	if (fl)
+	if (fl) {
+		release_net(fl->fl_net);
 		kfree(fl->opt);
+	}
 	kfree(fl);
 }
 
@@ -112,7 +114,6 @@ static void fl_release(struct ip6_flowlabel *fl)
 		    time_after(ip6_fl_gc_timer.expires, ttd))
 			mod_timer(&ip6_fl_gc_timer, ttd);
 	}
-
 	write_unlock_bh(&ip6_fl_lock);
 }
 
@@ -148,13 +149,34 @@ static void ip6_fl_gc(unsigned long dummy)
 	if (!sched && atomic_read(&fl_size))
 		sched = now + FL_MAX_LINGER;
 	if (sched) {
-		ip6_fl_gc_timer.expires = sched;
-		add_timer(&ip6_fl_gc_timer);
+		mod_timer(&ip6_fl_gc_timer, sched);
 	}
 	write_unlock(&ip6_fl_lock);
 }
 
-static struct ip6_flowlabel *fl_intern(struct ip6_flowlabel *fl, __be32 label)
+static void ip6_fl_purge(struct net *net)
+{
+	int i;
+
+	write_lock(&ip6_fl_lock);
+	for (i = 0; i <= FL_HASH_MASK; i++) {
+		struct ip6_flowlabel *fl, **flp;
+		flp = &fl_ht[i];
+		while ((fl = *flp) != NULL) {
+			if (fl->fl_net == net && atomic_read(&fl->users) == 0) {
+				*flp = fl->next;
+				fl_free(fl);
+				atomic_dec(&fl_size);
+				continue;
+			}
+			flp = &fl->next;
+		}
+	}
+	write_unlock(&ip6_fl_lock);
+}
+
+static struct ip6_flowlabel *fl_intern(struct net *net,
+				       struct ip6_flowlabel *fl, __be32 label)
 {
 	struct ip6_flowlabel *lfl;
 
@@ -165,7 +187,7 @@ static struct ip6_flowlabel *fl_intern(struct ip6_flowlabel *fl, __be32 label)
 		for (;;) {
 			fl->label = htonl(net_random())&IPV6_FLOWLABEL_MASK;
 			if (fl->label) {
-				lfl = __fl_lookup(fl->label);
+				lfl = __fl_lookup(net, fl->label);
 				if (lfl == NULL)
 					break;
 			}
@@ -179,7 +201,7 @@ static struct ip6_flowlabel *fl_intern(struct ip6_flowlabel *fl, __be32 label)
 		 * done in ipv6_flowlabel_opt - sock is locked, so new entry
 		 * with the same label can only appear on another sock
 		 */
-		lfl = __fl_lookup(fl->label);
+		lfl = __fl_lookup(net, fl->label);
 		if (lfl != NULL) {
 			atomic_inc(&lfl->users);
 			write_unlock_bh(&ip6_fl_lock);
@@ -298,7 +320,8 @@ static int fl6_renew(struct ip6_flowlabel *fl, unsigned long linger, unsigned lo
 }
 
 static struct ip6_flowlabel *
-fl_create(struct in6_flowlabel_req *freq, char __user *optval, int optlen, int *err_p)
+fl_create(struct net *net, struct in6_flowlabel_req *freq, char __user *optval,
+	  int optlen, int *err_p)
 {
 	struct ip6_flowlabel *fl;
 	int olen;
@@ -343,6 +366,7 @@ fl_create(struct in6_flowlabel_req *freq, char __user *optval, int optlen, int *
 		}
 	}
 
+	fl->fl_net = hold_net(net);
 	fl->expires = jiffies;
 	err = fl6_renew(fl, freq->flr_linger, freq->flr_expires);
 	if (err)
@@ -441,6 +465,7 @@ static inline void fl_link(struct ipv6_pinfo *np, struct ipv6_fl_socklist *sfl,
 int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
 {
 	int err;
+	struct net *net = sock_net(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct in6_flowlabel_req freq;
 	struct ipv6_fl_socklist *sfl1=NULL;
@@ -483,7 +508,7 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
 		read_unlock_bh(&ip6_sk_fl_lock);
 
 		if (freq.flr_share == IPV6_FL_S_NONE && capable(CAP_NET_ADMIN)) {
-			fl = fl_lookup(freq.flr_label);
+			fl = fl_lookup(net, freq.flr_label);
 			if (fl) {
 				err = fl6_renew(fl, freq.flr_linger, freq.flr_expires);
 				fl_release(fl);
@@ -496,7 +521,7 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
 		if (freq.flr_label & ~IPV6_FLOWLABEL_MASK)
 			return -EINVAL;
 
-		fl = fl_create(&freq, optval, optlen, &err);
+		fl = fl_create(net, &freq, optval, optlen, &err);
 		if (fl == NULL)
 			return err;
 		sfl1 = kmalloc(sizeof(*sfl1), GFP_KERNEL);
@@ -518,7 +543,7 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
 			read_unlock_bh(&ip6_sk_fl_lock);
 
 			if (fl1 == NULL)
-				fl1 = fl_lookup(freq.flr_label);
+				fl1 = fl_lookup(net, freq.flr_label);
 			if (fl1) {
 recheck:
 				err = -EEXIST;
@@ -559,7 +584,7 @@ release:
 		if (sfl1 == NULL || (err = mem_check(sk)) != 0)
 			goto done;
 
-		fl1 = fl_intern(fl, freq.flr_label);
+		fl1 = fl_intern(net, fl, freq.flr_label);
 		if (fl1 != NULL)
 			goto recheck;
 
@@ -717,13 +742,28 @@ static inline void ip6_flowlabel_proc_fini(struct net *net)
 }
 #endif
 
+static inline void ip6_flowlabel_net_exit(struct net *net)
+{
+	ip6_fl_purge(net);
+}
+
+static struct pernet_operations ip6_flowlabel_net_ops = {
+	.exit = ip6_flowlabel_net_exit,
+};
+
 int ip6_flowlabel_init(void)
 {
+	int err;
+
+	err = register_pernet_subsys(&ip6_flowlabel_net_ops);
+	if (err)
+		return err;
 	return ip6_flowlabel_proc_init(&init_net);
 }
 
 void ip6_flowlabel_cleanup(void)
 {
 	del_timer(&ip6_fl_gc_timer);
+	unregister_pernet_subsys(&ip6_flowlabel_net_ops);
 	ip6_flowlabel_proc_fini(&init_net);
 }
-- 
cgit v1.2.3


From 6c507cd0400cb51dd2ee251c1b8756b9375a1128 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Wed, 26 Mar 2008 14:14:55 +0100
Subject: cfg80211: don't export ieee80211_get_channel

This patch makes ieee80211_get_channel a static inline defined in
cfg80211's header file which simply calls __ieee80211_get_channel
to avoid symbol clashes with the ieee80211 code.

The problem was pointed out by David Miller, thanks!

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/wireless.h | 18 +++++++++++++++---
 net/wireless/util.c    |  6 +++---
 2 files changed, 18 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/wireless.h b/include/net/wireless.h
index f4b77ab66bae..667b4080d30f 100644
--- a/include/net/wireless.h
+++ b/include/net/wireless.h
@@ -304,10 +304,22 @@ extern int ieee80211_channel_to_frequency(int chan);
  */
 extern int ieee80211_frequency_to_channel(int freq);
 
+/*
+ * Name indirection necessary because the ieee80211 code also has
+ * a function named "ieee80211_get_channel", so if you include
+ * cfg80211's header file you get cfg80211's version, if you try
+ * to include both header files you'll (rightfully!) get a symbol
+ * clash.
+ */
+extern struct ieee80211_channel *__ieee80211_get_channel(struct wiphy *wiphy,
+							 int freq);
+
 /**
  * ieee80211_get_channel - get channel struct from wiphy for specified frequency
  */
-extern struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy,
-						       int freq);
-
+static inline struct ieee80211_channel *
+ieee80211_get_channel(struct wiphy *wiphy, int freq)
+{
+	return __ieee80211_get_channel(wiphy, freq);
+}
 #endif /* __NET_WIRELESS_H */
diff --git a/net/wireless/util.c b/net/wireless/util.c
index f3e623df3515..f54424693a38 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -33,8 +33,8 @@ int ieee80211_frequency_to_channel(int freq)
 }
 EXPORT_SYMBOL(ieee80211_frequency_to_channel);
 
-struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy,
-						int freq)
+struct ieee80211_channel *__ieee80211_get_channel(struct wiphy *wiphy,
+						  int freq)
 {
 	enum ieee80211_band band;
 	struct ieee80211_supported_band *sband;
@@ -54,7 +54,7 @@ struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy,
 
 	return NULL;
 }
-EXPORT_SYMBOL(ieee80211_get_channel);
+EXPORT_SYMBOL(__ieee80211_get_channel);
 
 static void set_mandatory_flags_band(struct ieee80211_supported_band *sband,
 				     enum ieee80211_band band)
-- 
cgit v1.2.3


From 0e5f8be1388093edc324a78ebf241170b258eba3 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 27 Mar 2008 14:25:53 -0700
Subject: [NETNS]: Compile NET /proc support only if CONFIG_NET is set.

This fix broken compilation for 'allnoconfig'. This was introduced by
Introduced by commit 1218854afa6f659be90b748cf1bc7badee954a35 ("[NET]
NETNS: Omit seq_net_private->net without CONFIG_NET_NS.")

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c       | 2 ++
 include/linux/seq_file.h | 2 ++
 2 files changed, 4 insertions(+)

(limited to 'include')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index 13cd7835d0df..7034facf8b8f 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -51,6 +51,7 @@ int seq_open_net(struct inode *ino, struct file *f,
 }
 EXPORT_SYMBOL_GPL(seq_open_net);
 
+#ifdef CONFIG_NET
 int seq_release_net(struct inode *ino, struct file *f)
 {
 	struct seq_file *seq;
@@ -218,3 +219,4 @@ int __init proc_net_init(void)
 
 	return register_pernet_subsys(&proc_net_ns_ops);
 }
+#endif /* CONFIG_NET */
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index d870a8253769..5da70c3f4417 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -63,6 +63,7 @@ extern struct list_head *seq_list_start_head(struct list_head *head,
 extern struct list_head *seq_list_next(void *v, struct list_head *head,
 		loff_t *ppos);
 
+#ifdef CONFIG_NET
 struct net;
 struct seq_net_private {
 #ifdef CONFIG_NET_NS
@@ -81,6 +82,7 @@ static inline struct net *seq_file_net(struct seq_file *seq)
 	return &init_net;
 #endif
 }
+#endif	/* CONFIG_NET */
 
 #endif
 #endif
-- 
cgit v1.2.3


From 09382bac667821017144de6b733a26f29a1c185b Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 27 Mar 2008 16:53:37 -0700
Subject: [PKT_SCHED]: Pass real namespace in net scheduler classifiers.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/pkt_cls.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index d349c66ef828..aa9e282db485 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -353,7 +353,7 @@ tcf_match_indev(struct sk_buff *skb, char *indev)
 	if (indev[0]) {
 		if  (!skb->iif)
 			return 0;
-		dev = __dev_get_by_index(&init_net, skb->iif);
+		dev = __dev_get_by_index(dev_net(skb->dev), skb->iif);
 		if (!dev || strcmp(indev, dev->name))
 			return 0;
 	}
-- 
cgit v1.2.3


From 4f95165d4bf6369d57052d60cc0266c569c6b077 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Thu, 27 Mar 2008 17:39:19 -0700
Subject: [IPV6]: Remove three unused method declarations in include/net/ipv6.h

This patch removes three unused method declarations in include/net/ipv6.h:
inet_getfrag_t(), ipv6_build_nfrag_opts() and ipv6_build_frag_opts().

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h | 17 -----------------
 1 file changed, 17 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 296f61d84709..5738c1c73ac1 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -250,15 +250,6 @@ int ip6_frag_mem(struct net *net);
 
 #define IPV6_FRAG_TIMEOUT	(60*HZ)		/* 60 seconds */
 
-/*
- *	Function prototype for build_xmit
- */
-
-typedef int		(*inet_getfrag_t) (const void *data,
-					   struct in6_addr *addr,
-					   char *,
-					   unsigned int, unsigned int);
-
 extern int __ipv6_addr_type(const struct in6_addr *addr);
 static inline int ipv6_addr_type(const struct in6_addr *addr)
 {
@@ -510,14 +501,6 @@ extern int			ip6_local_out(struct sk_buff *skb);
  *	Extension header (options) processing
  */
 
-extern u8 *			ipv6_build_nfrag_opts(struct sk_buff *skb,
-						      u8 *prev_hdr,
-						      struct ipv6_txoptions *opt,
-						      struct in6_addr *daddr,
-						      u32 jumbolen);
-extern u8 *			ipv6_build_frag_opts(struct sk_buff *skb,
-						     u8 *prev_hdr,
-						     struct ipv6_txoptions *opt);
 extern void 			ipv6_push_nfrag_opts(struct sk_buff *skb,
 						     struct ipv6_txoptions *opt,
 						     u8 *proto,
-- 
cgit v1.2.3


From 0dde3e16485dca16eb682dd59da1a598bf62e284 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 27 Mar 2008 17:43:41 -0700
Subject: [NET]: uninline skb_put, de-bloats a lot
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Allyesconfig (v2.6.24-mm1):

~500 files changed
...
 869 funcs, 198 +, 111003 -, diff: -110805 --- skb_put
  skb_put                       | +104

Without number of debug related CONFIGs (v2.6.25-rc2-mm1):

-60744  855 funcs, 861 +, 61605 -, diff: -60744 --- skb_put
  skb_put                       |  +57

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 21 +--------------------
 net/core/skbuff.c      | 21 +++++++++++++++++++++
 2 files changed, 22 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 7beb239d2ee0..f085955cb5a7 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -892,6 +892,7 @@ static inline void skb_set_tail_pointer(struct sk_buff *skb, const int offset)
 /*
  *	Add data to an sk_buff
  */
+extern unsigned char *skb_put(struct sk_buff *skb, unsigned int len);
 static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len)
 {
 	unsigned char *tmp = skb_tail_pointer(skb);
@@ -901,26 +902,6 @@ static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len)
 	return tmp;
 }
 
-/**
- *	skb_put - add data to a buffer
- *	@skb: buffer to use
- *	@len: amount of data to add
- *
- *	This function extends the used data area of the buffer. If this would
- *	exceed the total buffer size the kernel will panic. A pointer to the
- *	first byte of the extra data is returned.
- */
-static inline unsigned char *skb_put(struct sk_buff *skb, unsigned int len)
-{
-	unsigned char *tmp = skb_tail_pointer(skb);
-	SKB_LINEAR_ASSERT(skb);
-	skb->tail += len;
-	skb->len  += len;
-	if (unlikely(skb->tail > skb->end))
-		skb_over_panic(skb, len, current_text_addr());
-	return tmp;
-}
-
 static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len)
 {
 	skb->data -= len;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 0d0fd28a9041..3402eca768f8 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -857,6 +857,27 @@ free_skb:
 	return err;
 }
 
+/**
+ *	skb_put - add data to a buffer
+ *	@skb: buffer to use
+ *	@len: amount of data to add
+ *
+ *	This function extends the used data area of the buffer. If this would
+ *	exceed the total buffer size the kernel will panic. A pointer to the
+ *	first byte of the extra data is returned.
+ */
+unsigned char *skb_put(struct sk_buff *skb, unsigned int len)
+{
+	unsigned char *tmp = skb_tail_pointer(skb);
+	SKB_LINEAR_ASSERT(skb);
+	skb->tail += len;
+	skb->len  += len;
+	if (unlikely(skb->tail > skb->end))
+		skb_over_panic(skb, len, __builtin_return_address(0));
+	return tmp;
+}
+EXPORT_SYMBOL(skb_put);
+
 /* Trims skb to length len. It can change skb pointers.
  */
 
-- 
cgit v1.2.3


From 6be8ac2fdc5e69dec53913a42312a92dbfbd4907 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 27 Mar 2008 17:47:24 -0700
Subject: [NET]: uninline skb_pull, de-bloats a lot
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Allyesconfig (v2.6.24-mm1):

-28162  354 funcs, 3005 +, 31167 -, diff: -28162 --- skb_pull

Without number of debug related CONFIGs (v2.6.25-rc2-mm1):

-9697  338 funcs, 221 +, 9918 -, diff: -9697 --- skb_pull
skb_pull                      |  +44

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 16 +---------------
 net/core/skbuff.c      | 16 ++++++++++++++++
 2 files changed, 17 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index f085955cb5a7..6d6cde7b243c 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -927,6 +927,7 @@ static inline unsigned char *skb_push(struct sk_buff *skb, unsigned int len)
 	return skb->data;
 }
 
+extern unsigned char *skb_pull(struct sk_buff *skb, unsigned int len);
 static inline unsigned char *__skb_pull(struct sk_buff *skb, unsigned int len)
 {
 	skb->len -= len;
@@ -934,21 +935,6 @@ static inline unsigned char *__skb_pull(struct sk_buff *skb, unsigned int len)
 	return skb->data += len;
 }
 
-/**
- *	skb_pull - remove data from the start of a buffer
- *	@skb: buffer to use
- *	@len: amount of data to remove
- *
- *	This function removes data from the start of a buffer, returning
- *	the memory to the headroom. A pointer to the next data in the buffer
- *	is returned. Once the data has been pulled future pushes will overwrite
- *	the old data.
- */
-static inline unsigned char *skb_pull(struct sk_buff *skb, unsigned int len)
-{
-	return unlikely(len > skb->len) ? NULL : __skb_pull(skb, len);
-}
-
 extern unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta);
 
 static inline unsigned char *__pskb_pull(struct sk_buff *skb, unsigned int len)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 3402eca768f8..cf489b6329e8 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -878,6 +878,22 @@ unsigned char *skb_put(struct sk_buff *skb, unsigned int len)
 }
 EXPORT_SYMBOL(skb_put);
 
+/**
+ *	skb_pull - remove data from the start of a buffer
+ *	@skb: buffer to use
+ *	@len: amount of data to remove
+ *
+ *	This function removes data from the start of a buffer, returning
+ *	the memory to the headroom. A pointer to the next data in the buffer
+ *	is returned. Once the data has been pulled future pushes will overwrite
+ *	the old data.
+ */
+unsigned char *skb_pull(struct sk_buff *skb, unsigned int len)
+{
+	return unlikely(len > skb->len) ? NULL : __skb_pull(skb, len);
+}
+EXPORT_SYMBOL(skb_pull);
+
 /* Trims skb to length len. It can change skb pointers.
  */
 
-- 
cgit v1.2.3


From f58518e678e5eef430c8d5cdcc7cd28d285f1980 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 27 Mar 2008 17:51:31 -0700
Subject: [NET]: uninline dev_alloc_skb, de-bloats a lot
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Allyesconfig (v2.6.24-mm1):

-23668  392 funcs, 104 +, 23772 -, diff: -23668 --- dev_alloc_skb

Without many debug CONFIGs (v2.6.25-rc2-mm1):

-12178  382 funcs, 157 +, 12335 -, diff: -12178 --- dev_alloc_skb
dev_alloc_skb                 |  +37

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 17 +----------------
 net/core/skbuff.c      | 18 ++++++++++++++++++
 2 files changed, 19 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 6d6cde7b243c..01a11b0c0291 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1272,22 +1272,7 @@ static inline struct sk_buff *__dev_alloc_skb(unsigned int length,
 	return skb;
 }
 
-/**
- *	dev_alloc_skb - allocate an skbuff for receiving
- *	@length: length to allocate
- *
- *	Allocate a new &sk_buff and assign it a usage count of one. The
- *	buffer has unspecified headroom built in. Users should allocate
- *	the headroom they think they need without accounting for the
- *	built in space. The built in space is used for optimisations.
- *
- *	%NULL is returned if there is no free memory. Although this function
- *	allocates memory it can be called from an interrupt.
- */
-static inline struct sk_buff *dev_alloc_skb(unsigned int length)
-{
-	return __dev_alloc_skb(length, GFP_ATOMIC);
-}
+extern struct sk_buff *dev_alloc_skb(unsigned int length);
 
 extern struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
 		unsigned int length, gfp_t gfp_mask);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index cf489b6329e8..0daf5c0e5b8d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -263,6 +263,24 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
 	return skb;
 }
 
+/**
+ *	dev_alloc_skb - allocate an skbuff for receiving
+ *	@length: length to allocate
+ *
+ *	Allocate a new &sk_buff and assign it a usage count of one. The
+ *	buffer has unspecified headroom built in. Users should allocate
+ *	the headroom they think they need without accounting for the
+ *	built in space. The built in space is used for optimisations.
+ *
+ *	%NULL is returned if there is no free memory. Although this function
+ *	allocates memory it can be called from an interrupt.
+ */
+struct sk_buff *dev_alloc_skb(unsigned int length)
+{
+	return __dev_alloc_skb(length, GFP_ATOMIC);
+}
+EXPORT_SYMBOL(dev_alloc_skb);
+
 static void skb_drop_list(struct sk_buff **listp)
 {
 	struct sk_buff *list = *listp;
-- 
cgit v1.2.3


From c2aa270ad73d385bd6cdebf5d741bdf18a3e17ad Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 27 Mar 2008 17:52:40 -0700
Subject: [NET]: uninline skb_push, de-bloats a lot
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Allyesconfig (v2.6.24-mm1):

-21593  356 funcs, 2418 +, 24011 -, diff: -21593 --- skb_push

Without many debug related CONFIGs (v2.6.25-rc2-mm1):

-13890  341 funcs, 189 +, 14079 -, diff: -13890 --- skb_push
skb_push                      |  +46

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 19 +------------------
 net/core/skbuff.c      | 19 +++++++++++++++++++
 2 files changed, 20 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 01a11b0c0291..1baf4d43bb2d 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -902,6 +902,7 @@ static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len)
 	return tmp;
 }
 
+extern unsigned char *skb_push(struct sk_buff *skb, unsigned int len);
 static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len)
 {
 	skb->data -= len;
@@ -909,24 +910,6 @@ static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len)
 	return skb->data;
 }
 
-/**
- *	skb_push - add data to the start of a buffer
- *	@skb: buffer to use
- *	@len: amount of data to add
- *
- *	This function extends the used data area of the buffer at the buffer
- *	start. If this would exceed the total buffer headroom the kernel will
- *	panic. A pointer to the first byte of the extra data is returned.
- */
-static inline unsigned char *skb_push(struct sk_buff *skb, unsigned int len)
-{
-	skb->data -= len;
-	skb->len  += len;
-	if (unlikely(skb->data<skb->head))
-		skb_under_panic(skb, len, current_text_addr());
-	return skb->data;
-}
-
 extern unsigned char *skb_pull(struct sk_buff *skb, unsigned int len);
 static inline unsigned char *__skb_pull(struct sk_buff *skb, unsigned int len)
 {
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 0daf5c0e5b8d..a37127b5899c 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -896,6 +896,25 @@ unsigned char *skb_put(struct sk_buff *skb, unsigned int len)
 }
 EXPORT_SYMBOL(skb_put);
 
+/**
+ *	skb_push - add data to the start of a buffer
+ *	@skb: buffer to use
+ *	@len: amount of data to add
+ *
+ *	This function extends the used data area of the buffer at the buffer
+ *	start. If this would exceed the total buffer headroom the kernel will
+ *	panic. A pointer to the first byte of the extra data is returned.
+ */
+unsigned char *skb_push(struct sk_buff *skb, unsigned int len)
+{
+	skb->data -= len;
+	skb->len  += len;
+	if (unlikely(skb->data<skb->head))
+		skb_under_panic(skb, len, __builtin_return_address(0));
+	return skb->data;
+}
+EXPORT_SYMBOL(skb_push);
+
 /**
  *	skb_pull - remove data from the start of a buffer
  *	@skb: buffer to use
-- 
cgit v1.2.3


From 8d3308687f7f1eaa1bb5d202d14752d5f90068eb Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 27 Mar 2008 17:53:31 -0700
Subject: [NET]: uninline dst_release
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Codiff stats (allyesconfig, v2.6.24-mm1):
-16420  187 funcs, 103 +, 16523 -, diff: -16420 --- dst_release

Without number of debug related CONFIGs (v2.6.25-rc2-mm1):
-7257  186 funcs, 70 +, 7327 -, diff: -7257 --- dst_release
dst_release                   |  +40

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h | 10 +---------
 net/core/dst.c    | 10 ++++++++++
 2 files changed, 11 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index ae13370e8484..002500e631f5 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -163,15 +163,7 @@ struct dst_entry * dst_clone(struct dst_entry * dst)
 	return dst;
 }
 
-static inline
-void dst_release(struct dst_entry * dst)
-{
-	if (dst) {
-		WARN_ON(atomic_read(&dst->__refcnt) < 1);
-		smp_mb__before_atomic_dec();
-		atomic_dec(&dst->__refcnt);
-	}
-}
+extern void dst_release(struct dst_entry *dst);
 
 /* Children define the path of the packet through the
  * Linux networking.  Thus, destinations are stackable.
diff --git a/net/core/dst.c b/net/core/dst.c
index 694cd2a3f6d2..fe03266130b6 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -259,6 +259,16 @@ again:
 	return NULL;
 }
 
+void dst_release(struct dst_entry *dst)
+{
+	if (dst) {
+		WARN_ON(atomic_read(&dst->__refcnt) < 1);
+		smp_mb__before_atomic_dec();
+		atomic_dec(&dst->__refcnt);
+	}
+}
+EXPORT_SYMBOL(dst_release);
+
 /* Dirty hack. We did it in 2.2 (in __dst_free),
  * we have _very_ good reasons not to repeat
  * this mistake in 2.3, but we have no choice
-- 
cgit v1.2.3


From 419ae74ecc9494e58928a5c6652f4c072f3ca744 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 27 Mar 2008 17:54:01 -0700
Subject: [NET]: uninline skb_trim, de-bloats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Allyesconfig (v2.6.24-mm1):
-10976  209 funcs, 123 +, 11099 -, diff: -10976 --- skb_trim

Without number of debug related CONFIGs (v2.6.25-rc2-mm1):
-7360  192 funcs, 131 +, 7491 -, diff: -7360 --- skb_trim
skb_trim                      |  +42

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 16 +---------------
 net/core/skbuff.c      | 16 ++++++++++++++++
 2 files changed, 17 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 1baf4d43bb2d..ff72145d5d9e 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1158,21 +1158,7 @@ static inline void __skb_trim(struct sk_buff *skb, unsigned int len)
 	skb_set_tail_pointer(skb, len);
 }
 
-/**
- *	skb_trim - remove end from a buffer
- *	@skb: buffer to alter
- *	@len: new length
- *
- *	Cut the length of a buffer down by removing data from the tail. If
- *	the buffer is already under the length specified it is not modified.
- *	The skb must be linear.
- */
-static inline void skb_trim(struct sk_buff *skb, unsigned int len)
-{
-	if (skb->len > len)
-		__skb_trim(skb, len);
-}
-
+extern void skb_trim(struct sk_buff *skb, unsigned int len);
 
 static inline int __pskb_trim(struct sk_buff *skb, unsigned int len)
 {
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index a37127b5899c..86e5682728be 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -931,6 +931,22 @@ unsigned char *skb_pull(struct sk_buff *skb, unsigned int len)
 }
 EXPORT_SYMBOL(skb_pull);
 
+/**
+ *	skb_trim - remove end from a buffer
+ *	@skb: buffer to alter
+ *	@len: new length
+ *
+ *	Cut the length of a buffer down by removing data from the tail. If
+ *	the buffer is already under the length specified it is not modified.
+ *	The skb must be linear.
+ */
+void skb_trim(struct sk_buff *skb, unsigned int len)
+{
+	if (skb->len > len)
+		__skb_trim(skb, len);
+}
+EXPORT_SYMBOL(skb_trim);
+
 /* Trims skb to length len. It can change skb pointers.
  */
 
-- 
cgit v1.2.3


From bc09dff198e67a98a82c42000006b39f6d502031 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 27 Mar 2008 17:54:29 -0700
Subject: [SCTP]: Remove sctp_add_cmd_sf wrapper bloat
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

With a was number of callsites sctp_add_cmd_sf wrapper bloats
kernel by some amount. Due to unlikely tracking allyesconfig,
with the initial result were around ~7kB (thus caught my
attention) while a non-debug config produced only ~2.3kB effect.

I (ij) proposed first a patch to uninline it but Vlad responded
with a patch that removed the only sctp_add_cmd call which is
wrapped by sctp_add_cmd_sf (I wasn't sure if I could do that).
I did minor cleanup to Vlad's patch.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/command.h |  3 +--
 include/net/sctp/sm.h      |  8 --------
 net/sctp/command.c         | 10 ++--------
 net/sctp/sm_statefuns.c    |  8 ++------
 4 files changed, 5 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h
index 10ae2da6f93b..4263af857794 100644
--- a/include/net/sctp/command.h
+++ b/include/net/sctp/command.h
@@ -205,12 +205,11 @@ typedef struct {
 int sctp_init_cmd_seq(sctp_cmd_seq_t *seq);
 
 /* Add a command to an sctp_cmd_seq_t.
- * Return 0 if the command sequence is full.
  *
  * Use the SCTP_* constructors defined by SCTP_ARG_CONSTRUCTOR() above
  * to wrap data which goes in the obj argument.
  */
-int sctp_add_cmd(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj);
+void sctp_add_cmd_sf(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj);
 
 /* Return the next command structure in an sctp_cmd_seq.
  * Return NULL at the end of the sequence.
diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h
index ef9e7ed2c82e..24811732bdb2 100644
--- a/include/net/sctp/sm.h
+++ b/include/net/sctp/sm.h
@@ -385,14 +385,6 @@ static inline int ADDIP_SERIAL_gte(__u16 s, __u16 t)
 	return (((s) == (t)) || (((t) - (s)) & ADDIP_SERIAL_SIGN_BIT));
 }
 
-
-/* Run sctp_add_cmd() generating a BUG() if there is a failure.  */
-static inline void sctp_add_cmd_sf(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj)
-{
-	if (unlikely(!sctp_add_cmd(seq, verb, obj)))
-		BUG();
-}
-
 /* Check VTAG of the packet matches the sender's own tag. */
 static inline int
 sctp_vtag_verify(const struct sctp_chunk *chunk,
diff --git a/net/sctp/command.c b/net/sctp/command.c
index bb977330002a..c0044019db9e 100644
--- a/net/sctp/command.c
+++ b/net/sctp/command.c
@@ -52,18 +52,12 @@ int sctp_init_cmd_seq(sctp_cmd_seq_t *seq)
 /* Add a command to a sctp_cmd_seq_t.
  * Return 0 if the command sequence is full.
  */
-int sctp_add_cmd(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj)
+void sctp_add_cmd_sf(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj)
 {
-	if (seq->next_free_slot >= SCTP_MAX_NUM_COMMANDS)
-		goto fail;
+	BUG_ON(seq->next_free_slot >= SCTP_MAX_NUM_COMMANDS);
 
 	seq->cmds[seq->next_free_slot].verb = verb;
 	seq->cmds[seq->next_free_slot++].obj = obj;
-
-	return 1;
-
-fail:
-	return 0;
 }
 
 /* Return the next command structure in a sctp_cmd_seq.
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 6545b5fcbc73..b534dbef864f 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -3135,12 +3135,8 @@ sctp_disposition_t sctp_sf_operr_notify(const struct sctp_endpoint *ep,
 		if (!ev)
 			goto nomem;
 
-		if (!sctp_add_cmd(commands, SCTP_CMD_EVENT_ULP,
-				  SCTP_ULPEVENT(ev))) {
-			sctp_ulpevent_free(ev);
-			goto nomem;
-		}
-
+		sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP,
+				SCTP_ULPEVENT(ev));
 		sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_OPERR,
 				SCTP_CHUNK(chunk));
 	}
-- 
cgit v1.2.3


From 1567ca7eec7664b8be3b07755ac59dc1b1ec76cb Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 28 Mar 2008 15:53:11 -0700
Subject: [NET]: Protect device namespace inlines with CONFIG_NET

Include sites should not be bothered by whether
CONFIG_NET is set or not when trying to include
benign files like linux/etherdevice.h et al.

From a report by Stephen Rothwell.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 3b54f8a2c055..8576ca928dae 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -741,6 +741,7 @@ struct net_device
 #define	NETDEV_ALIGN		32
 #define	NETDEV_ALIGN_CONST	(NETDEV_ALIGN - 1)
 
+#ifdef CONFIG_NET
 /*
  * Net namespace inlines
  */
@@ -761,6 +762,7 @@ void dev_net_set(struct net_device *dev, struct net *net)
 	dev->nd_net = net;
 #endif
 }
+#endif
 
 /**
  *	netdev_priv - access network device private data
-- 
cgit v1.2.3


From be2ce06b4962658d807410e58f7c6b739dc6a0c3 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Fri, 28 Mar 2008 16:26:45 -0700
Subject: [IPV6]: Remove unused method declaration in include/net/addrconf.h.

This patches removes unused declaration of addrconf_forwarding_on() method
in include/net/addrconf.h.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index d0c47c306046..d89b0bc7ab75 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -185,7 +185,6 @@ static inline void in6_ifa_put(struct inet6_ifaddr *ifp)
 #define in6_ifa_hold(ifp)	atomic_inc(&(ifp)->refcnt)
 
 
-extern void			addrconf_forwarding_on(void);
 /*
  *	Hash function taken from net_alias.c
  */
-- 
cgit v1.2.3


From bc578a54f0fd489d0722303f9a52508495ccaf9a Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Fri, 28 Mar 2008 16:35:27 -0700
Subject: [NET]: Rename inet_frag.h identifiers COMPLETE, FIRST_IN, LAST_IN to
 INET_FRAG_*

On Fri, 2008-03-28 at 03:24 -0700, Andrew Morton wrote:
> they should all be renamed.

Done for include/net and net

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_frag.h                 |  6 +++---
 net/ipv4/inet_fragment.c                | 10 +++++-----
 net/ipv4/ip_fragment.c                  | 17 +++++++++--------
 net/ipv6/netfilter/nf_conntrack_reasm.c | 15 ++++++++-------
 net/ipv6/reassembly.c                   | 17 +++++++++--------
 5 files changed, 34 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 7374251b9787..e081eefd6f47 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -25,9 +25,9 @@ struct inet_frag_queue {
 	int			meat;
 	__u8			last_in;    /* first/last segment arrived? */
 
-#define COMPLETE		4
-#define FIRST_IN		2
-#define LAST_IN			1
+#define INET_FRAG_COMPLETE	4
+#define INET_FRAG_FIRST_IN	2
+#define INET_FRAG_LAST_IN	1
 };
 
 #define INETFRAGS_HASHSZ		64
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 724d69aed031..93170bfcc22e 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -104,10 +104,10 @@ void inet_frag_kill(struct inet_frag_queue *fq, struct inet_frags *f)
 	if (del_timer(&fq->timer))
 		atomic_dec(&fq->refcnt);
 
-	if (!(fq->last_in & COMPLETE)) {
+	if (!(fq->last_in & INET_FRAG_COMPLETE)) {
 		fq_unlink(fq, f);
 		atomic_dec(&fq->refcnt);
-		fq->last_in |= COMPLETE;
+		fq->last_in |= INET_FRAG_COMPLETE;
 	}
 }
 
@@ -131,7 +131,7 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f,
 	struct sk_buff *fp;
 	struct netns_frags *nf;
 
-	BUG_TRAP(q->last_in & COMPLETE);
+	BUG_TRAP(q->last_in & INET_FRAG_COMPLETE);
 	BUG_TRAP(del_timer(&q->timer) == 0);
 
 	/* Release all fragment data. */
@@ -174,7 +174,7 @@ int inet_frag_evictor(struct netns_frags *nf, struct inet_frags *f)
 		read_unlock(&f->lock);
 
 		spin_lock(&q->lock);
-		if (!(q->last_in & COMPLETE))
+		if (!(q->last_in & INET_FRAG_COMPLETE))
 			inet_frag_kill(q, f);
 		spin_unlock(&q->lock);
 
@@ -206,7 +206,7 @@ static struct inet_frag_queue *inet_frag_intern(struct netns_frags *nf,
 		if (qp->net == nf && f->match(qp, arg)) {
 			atomic_inc(&qp->refcnt);
 			write_unlock(&f->lock);
-			qp_in->last_in |= COMPLETE;
+			qp_in->last_in |= INET_FRAG_COMPLETE;
 			inet_frag_put(qp_in, f);
 			return qp;
 		}
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index fcb60e76b234..02ae470fe60c 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -194,7 +194,7 @@ static void ip_expire(unsigned long arg)
 
 	spin_lock(&qp->q.lock);
 
-	if (qp->q.last_in & COMPLETE)
+	if (qp->q.last_in & INET_FRAG_COMPLETE)
 		goto out;
 
 	ipq_kill(qp);
@@ -202,7 +202,7 @@ static void ip_expire(unsigned long arg)
 	IP_INC_STATS_BH(IPSTATS_MIB_REASMTIMEOUT);
 	IP_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
 
-	if ((qp->q.last_in&FIRST_IN) && qp->q.fragments != NULL) {
+	if ((qp->q.last_in & INET_FRAG_FIRST_IN) && qp->q.fragments != NULL) {
 		struct sk_buff *head = qp->q.fragments;
 		struct net *net;
 
@@ -301,7 +301,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 	int ihl, end;
 	int err = -ENOENT;
 
-	if (qp->q.last_in & COMPLETE)
+	if (qp->q.last_in & INET_FRAG_COMPLETE)
 		goto err;
 
 	if (!(IPCB(skb)->flags & IPSKB_FRAG_COMPLETE) &&
@@ -327,9 +327,9 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 		 * or have different end, the segment is corrrupted.
 		 */
 		if (end < qp->q.len ||
-		    ((qp->q.last_in & LAST_IN) && end != qp->q.len))
+		    ((qp->q.last_in & INET_FRAG_LAST_IN) && end != qp->q.len))
 			goto err;
-		qp->q.last_in |= LAST_IN;
+		qp->q.last_in |= INET_FRAG_LAST_IN;
 		qp->q.len = end;
 	} else {
 		if (end&7) {
@@ -339,7 +339,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 		}
 		if (end > qp->q.len) {
 			/* Some bits beyond end -> corruption. */
-			if (qp->q.last_in & LAST_IN)
+			if (qp->q.last_in & INET_FRAG_LAST_IN)
 				goto err;
 			qp->q.len = end;
 		}
@@ -438,9 +438,10 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 	qp->q.meat += skb->len;
 	atomic_add(skb->truesize, &qp->q.net->mem);
 	if (offset == 0)
-		qp->q.last_in |= FIRST_IN;
+		qp->q.last_in |= INET_FRAG_FIRST_IN;
 
-	if (qp->q.last_in == (FIRST_IN | LAST_IN) && qp->q.meat == qp->q.len)
+	if (qp->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) &&
+	    qp->q.meat == qp->q.len)
 		return ip_frag_reasm(qp, prev, dev);
 
 	write_lock(&ip4_frags.lock);
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 2a0d698b24d5..ad8066200f9d 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -183,7 +183,7 @@ static void nf_ct_frag6_expire(unsigned long data)
 
 	spin_lock(&fq->q.lock);
 
-	if (fq->q.last_in & COMPLETE)
+	if (fq->q.last_in & INET_FRAG_COMPLETE)
 		goto out;
 
 	fq_kill(fq);
@@ -225,7 +225,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
 	struct sk_buff *prev, *next;
 	int offset, end;
 
-	if (fq->q.last_in & COMPLETE) {
+	if (fq->q.last_in & INET_FRAG_COMPLETE) {
 		pr_debug("Allready completed\n");
 		goto err;
 	}
@@ -252,11 +252,11 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
 		 * or have different end, the segment is corrupted.
 		 */
 		if (end < fq->q.len ||
-		    ((fq->q.last_in & LAST_IN) && end != fq->q.len)) {
+		    ((fq->q.last_in & INET_FRAG_LAST_IN) && end != fq->q.len)) {
 			pr_debug("already received last fragment\n");
 			goto err;
 		}
-		fq->q.last_in |= LAST_IN;
+		fq->q.last_in |= INET_FRAG_LAST_IN;
 		fq->q.len = end;
 	} else {
 		/* Check if the fragment is rounded to 8 bytes.
@@ -271,7 +271,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
 		}
 		if (end > fq->q.len) {
 			/* Some bits beyond end -> corruption. */
-			if (fq->q.last_in & LAST_IN) {
+			if (fq->q.last_in & INET_FRAG_LAST_IN) {
 				pr_debug("last packet already reached.\n");
 				goto err;
 			}
@@ -383,7 +383,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
 	 */
 	if (offset == 0) {
 		fq->nhoffset = nhoff;
-		fq->q.last_in |= FIRST_IN;
+		fq->q.last_in |= INET_FRAG_FIRST_IN;
 	}
 	write_lock(&nf_frags.lock);
 	list_move_tail(&fq->q.lru_list, &nf_init_frags.lru_list);
@@ -645,7 +645,8 @@ struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb)
 		goto ret_orig;
 	}
 
-	if (fq->q.last_in == (FIRST_IN|LAST_IN) && fq->q.meat == fq->q.len) {
+	if (fq->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) &&
+	    fq->q.meat == fq->q.len) {
 		ret_skb = nf_ct_frag6_reasm(fq, dev);
 		if (ret_skb == NULL)
 			pr_debug("Can't reassemble fragmented packets\n");
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 4e1447634f36..7b247e3a16fe 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -202,7 +202,7 @@ static void ip6_frag_expire(unsigned long data)
 
 	spin_lock(&fq->q.lock);
 
-	if (fq->q.last_in & COMPLETE)
+	if (fq->q.last_in & INET_FRAG_COMPLETE)
 		goto out;
 
 	fq_kill(fq);
@@ -217,7 +217,7 @@ static void ip6_frag_expire(unsigned long data)
 	rcu_read_unlock();
 
 	/* Don't send error if the first segment did not arrive. */
-	if (!(fq->q.last_in&FIRST_IN) || !fq->q.fragments)
+	if (!(fq->q.last_in & INET_FRAG_FIRST_IN) || !fq->q.fragments)
 		goto out;
 
 	/*
@@ -265,7 +265,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 	struct net_device *dev;
 	int offset, end;
 
-	if (fq->q.last_in & COMPLETE)
+	if (fq->q.last_in & INET_FRAG_COMPLETE)
 		goto err;
 
 	offset = ntohs(fhdr->frag_off) & ~0x7;
@@ -294,9 +294,9 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 		 * or have different end, the segment is corrupted.
 		 */
 		if (end < fq->q.len ||
-		    ((fq->q.last_in & LAST_IN) && end != fq->q.len))
+		    ((fq->q.last_in & INET_FRAG_LAST_IN) && end != fq->q.len))
 			goto err;
-		fq->q.last_in |= LAST_IN;
+		fq->q.last_in |= INET_FRAG_LAST_IN;
 		fq->q.len = end;
 	} else {
 		/* Check if the fragment is rounded to 8 bytes.
@@ -314,7 +314,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 		}
 		if (end > fq->q.len) {
 			/* Some bits beyond end -> corruption. */
-			if (fq->q.last_in & LAST_IN)
+			if (fq->q.last_in & INET_FRAG_LAST_IN)
 				goto err;
 			fq->q.len = end;
 		}
@@ -417,10 +417,11 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 	 */
 	if (offset == 0) {
 		fq->nhoffset = nhoff;
-		fq->q.last_in |= FIRST_IN;
+		fq->q.last_in |= INET_FRAG_FIRST_IN;
 	}
 
-	if (fq->q.last_in == (FIRST_IN | LAST_IN) && fq->q.meat == fq->q.len)
+	if (fq->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) &&
+	    fq->q.meat == fq->q.len)
 		return ip6_frag_reasm(fq, prev, dev);
 
 	write_lock(&ip6_frags.lock);
-- 
cgit v1.2.3


From 13ff3d6fa4e6d8b6ee7c64245a0078e6a0e6f977 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 28 Mar 2008 16:38:17 -0700
Subject: [SOCK]: Enumerate struct proto-s to facilitate percpu inuse
 accounting (v2).

The inuse counters are going to become a per-cpu array.  Introduce an
index for this array on the struct proto.

To handle the case of proto register-unregister-register loop the
bitmap is used. All its bits manipulations are protected with
proto_list_lock and a sanity check for the bitmap being exhausted is
also added.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h |  1 +
 net/core/sock.c    | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 35 insertions(+)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 1c9d059223ee..abc6341f536f 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -562,6 +562,7 @@ struct proto {
 
 	/* Keeping track of sockets in use */
 #ifdef CONFIG_PROC_FS
+	unsigned int		inuse_idx;
 	struct pcounter		inuse;
 #endif
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 3ee95060dbd0..7d2c8add5f5a 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1940,6 +1940,38 @@ EXPORT_SYMBOL(sk_common_release);
 static DEFINE_RWLOCK(proto_list_lock);
 static LIST_HEAD(proto_list);
 
+#ifdef CONFIG_PROC_FS
+#define PROTO_INUSE_NR	64	/* should be enough for the first time */
+
+static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR);
+
+static void assign_proto_idx(struct proto *prot)
+{
+	prot->inuse_idx = find_first_zero_bit(proto_inuse_idx, PROTO_INUSE_NR);
+
+	if (unlikely(prot->inuse_idx == PROTO_INUSE_NR - 1)) {
+		printk(KERN_ERR "PROTO_INUSE_NR exhausted\n");
+		return;
+	}
+
+	set_bit(prot->inuse_idx, proto_inuse_idx);
+}
+
+static void release_proto_idx(struct proto *prot)
+{
+	if (prot->inuse_idx != PROTO_INUSE_NR - 1)
+		clear_bit(prot->inuse_idx, proto_inuse_idx);
+}
+#else
+static inline void assign_proto_idx(struct proto *prot)
+{
+}
+
+static inline void release_proto_idx(struct proto *prot)
+{
+}
+#endif
+
 int proto_register(struct proto *prot, int alloc_slab)
 {
 	char *request_sock_slab_name = NULL;
@@ -2000,6 +2032,7 @@ int proto_register(struct proto *prot, int alloc_slab)
 
 	write_lock(&proto_list_lock);
 	list_add(&prot->node, &proto_list);
+	assign_proto_idx(prot);
 	write_unlock(&proto_list_lock);
 	return 0;
 
@@ -2026,6 +2059,7 @@ EXPORT_SYMBOL(proto_register);
 void proto_unregister(struct proto *prot)
 {
 	write_lock(&proto_list_lock);
+	release_proto_idx(prot);
 	list_del(&prot->node);
 	write_unlock(&proto_list_lock);
 
-- 
cgit v1.2.3


From 1338d466d9c3f8a65cc6d83c629cd906f2a989f8 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 28 Mar 2008 16:38:43 -0700
Subject: [SOCK]: Introduce a percpu inuse counters array (v2).

And redirect sock_prot_inuse_add and _get to use one.

As far as the dereferences are concerned. Before the patch we made
1 dereference to proto->inuse.add call, the call itself and then
called the __get_cpu_var() on a static variable. After the patch we
make a direct call, then one dereference to proto->inuse_idx and
then the same __get_cpu_var() on a still static variable. So this
patch doesn't seem to produce performance penalty on SMP.

This is not per-net yet, but I will deliberately make NET_NS=y case
separated from NET_NS=n one, since it'll cost us one-or-two more
dereferences to get the struct net and the inuse counter.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 13 +++++--------
 net/core/sock.c    | 22 ++++++++++++++++++++++
 2 files changed, 27 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index abc6341f536f..ebf9552664b2 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -639,18 +639,15 @@ static inline void sk_refcnt_debug_release(const struct sock *sk)
 # define DEFINE_PROTO_INUSE(NAME) DEFINE_PCOUNTER(NAME)
 # define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse)
 /* Called with local bh disabled */
-static inline void sock_prot_inuse_add(struct proto *prot, int inc)
-{
-	pcounter_add(&prot->inuse, inc);
-}
+extern void sock_prot_inuse_add(struct proto *prot, int inc);
+
 static inline int sock_prot_inuse_init(struct proto *proto)
 {
 	return pcounter_alloc(&proto->inuse);
 }
-static inline int sock_prot_inuse_get(struct proto *proto)
-{
-	return pcounter_getval(&proto->inuse);
-}
+
+extern int sock_prot_inuse_get(struct proto *proto);
+
 static inline void sock_prot_inuse_free(struct proto *proto)
 {
 	pcounter_free(&proto->inuse);
diff --git a/net/core/sock.c b/net/core/sock.c
index 7d2c8add5f5a..174c64bc7a43 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1942,8 +1942,30 @@ static LIST_HEAD(proto_list);
 
 #ifdef CONFIG_PROC_FS
 #define PROTO_INUSE_NR	64	/* should be enough for the first time */
+struct prot_inuse {
+	int val[PROTO_INUSE_NR];
+};
 
 static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR);
+static DEFINE_PER_CPU(struct prot_inuse, prot_inuse);
+
+void sock_prot_inuse_add(struct proto *prot, int val)
+{
+	__get_cpu_var(prot_inuse).val[prot->inuse_idx] += val;
+}
+EXPORT_SYMBOL_GPL(sock_prot_inuse_add);
+
+int sock_prot_inuse_get(struct proto *prot)
+{
+	int cpu, idx = prot->inuse_idx;
+	int res = 0;
+
+	for_each_possible_cpu(cpu)
+		res += per_cpu(prot_inuse, cpu).val[idx];
+
+	return res >= 0 ? res : 0;
+}
+EXPORT_SYMBOL_GPL(sock_prot_inuse_get);
 
 static void assign_proto_idx(struct proto *prot)
 {
-- 
cgit v1.2.3


From 60e7663d462af3994f292cb3691ea4f7371a9220 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 28 Mar 2008 16:39:10 -0700
Subject: [SOCK]: Drop per-proto inuse init and fre functions (v2).

Constructive part of the set is finished here. We have to remove the
pcounter, so start with its init and free functions.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 18 ------------------
 net/core/sock.c    | 11 +----------
 2 files changed, 1 insertion(+), 28 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index ebf9552664b2..1f4294252dd7 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -640,31 +640,13 @@ static inline void sk_refcnt_debug_release(const struct sock *sk)
 # define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse)
 /* Called with local bh disabled */
 extern void sock_prot_inuse_add(struct proto *prot, int inc);
-
-static inline int sock_prot_inuse_init(struct proto *proto)
-{
-	return pcounter_alloc(&proto->inuse);
-}
-
 extern int sock_prot_inuse_get(struct proto *proto);
-
-static inline void sock_prot_inuse_free(struct proto *proto)
-{
-	pcounter_free(&proto->inuse);
-}
 #else
 # define DEFINE_PROTO_INUSE(NAME)
 # define REF_PROTO_INUSE(NAME)
 static void inline sock_prot_inuse_add(struct proto *prot, int inc)
 {
 }
-static int inline sock_prot_inuse_init(struct proto *proto)
-{
-	return 0;
-}
-static void inline sock_prot_inuse_free(struct proto *proto)
-{
-}
 #endif
 
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 174c64bc7a43..c1ae56eb96ec 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1999,11 +1999,6 @@ int proto_register(struct proto *prot, int alloc_slab)
 	char *request_sock_slab_name = NULL;
 	char *timewait_sock_slab_name;
 
-	if (sock_prot_inuse_init(prot) != 0) {
-		printk(KERN_CRIT "%s: Can't alloc inuse counters!\n", prot->name);
-		goto out;
-	}
-
 	if (alloc_slab) {
 		prot->slab = kmem_cache_create(prot->name, prot->obj_size, 0,
 					       SLAB_HWCACHE_ALIGN, NULL);
@@ -2011,7 +2006,7 @@ int proto_register(struct proto *prot, int alloc_slab)
 		if (prot->slab == NULL) {
 			printk(KERN_CRIT "%s: Can't create sock SLAB cache!\n",
 			       prot->name);
-			goto out_free_inuse;
+			goto out;
 		}
 
 		if (prot->rsk_prot != NULL) {
@@ -2070,8 +2065,6 @@ out_free_request_sock_slab_name:
 out_free_sock_slab:
 	kmem_cache_destroy(prot->slab);
 	prot->slab = NULL;
-out_free_inuse:
-	sock_prot_inuse_free(prot);
 out:
 	return -ENOBUFS;
 }
@@ -2085,8 +2078,6 @@ void proto_unregister(struct proto *prot)
 	list_del(&prot->node);
 	write_unlock(&proto_list_lock);
 
-	sock_prot_inuse_free(prot);
-
 	if (prot->slab != NULL) {
 		kmem_cache_destroy(prot->slab);
 		prot->slab = NULL;
-- 
cgit v1.2.3


From bdcde3d71a67e97f25e851f3ca97c9bb5ef03e7f Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 28 Mar 2008 16:39:33 -0700
Subject: [SOCK]: Drop inuse pcounter from struct proto (v2).

An uppercut - do not use the pcounter on struct proto.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h  | 6 ------
 net/dccp/ipv4.c     | 3 ---
 net/dccp/ipv6.c     | 3 ---
 net/ipv4/raw.c      | 3 ---
 net/ipv4/tcp_ipv4.c | 3 ---
 net/ipv4/udp.c      | 3 ---
 net/ipv4/udplite.c  | 3 ---
 net/ipv6/raw.c      | 3 ---
 net/ipv6/tcp_ipv6.c | 3 ---
 net/ipv6/udp.c      | 3 ---
 net/ipv6/udplite.c  | 3 ---
 net/sctp/socket.c   | 5 -----
 12 files changed, 41 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 1f4294252dd7..2a3344f666aa 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -47,7 +47,6 @@
 #include <linux/module.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
-#include <linux/pcounter.h>
 #include <linux/skbuff.h>	/* struct sk_buff */
 #include <linux/mm.h>
 #include <linux/security.h>
@@ -563,7 +562,6 @@ struct proto {
 	/* Keeping track of sockets in use */
 #ifdef CONFIG_PROC_FS
 	unsigned int		inuse_idx;
-	struct pcounter		inuse;
 #endif
 
 	/* Memory pressure */
@@ -636,14 +634,10 @@ static inline void sk_refcnt_debug_release(const struct sock *sk)
 
 
 #ifdef CONFIG_PROC_FS
-# define DEFINE_PROTO_INUSE(NAME) DEFINE_PCOUNTER(NAME)
-# define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse)
 /* Called with local bh disabled */
 extern void sock_prot_inuse_add(struct proto *prot, int inc);
 extern int sock_prot_inuse_get(struct proto *proto);
 #else
-# define DEFINE_PROTO_INUSE(NAME)
-# define REF_PROTO_INUSE(NAME)
 static void inline sock_prot_inuse_add(struct proto *prot, int inc)
 {
 }
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 4ca8b0c93c80..7d62f7eb6134 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -915,8 +915,6 @@ static struct timewait_sock_ops dccp_timewait_sock_ops = {
 	.twsk_obj_size	= sizeof(struct inet_timewait_sock),
 };
 
-DEFINE_PROTO_INUSE(dccp_v4)
-
 static struct proto dccp_v4_prot = {
 	.name			= "DCCP",
 	.owner			= THIS_MODULE,
@@ -946,7 +944,6 @@ static struct proto dccp_v4_prot = {
 	.compat_setsockopt	= compat_dccp_setsockopt,
 	.compat_getsockopt	= compat_dccp_getsockopt,
 #endif
-	REF_PROTO_INUSE(dccp_v4)
 };
 
 static struct net_protocol dccp_v4_protocol = {
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 2fec1af1a8c3..ea3f32648618 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -1100,8 +1100,6 @@ static struct timewait_sock_ops dccp6_timewait_sock_ops = {
 	.twsk_obj_size	= sizeof(struct dccp6_timewait_sock),
 };
 
-DEFINE_PROTO_INUSE(dccp_v6)
-
 static struct proto dccp_v6_prot = {
 	.name		   = "DCCPv6",
 	.owner		   = THIS_MODULE,
@@ -1131,7 +1129,6 @@ static struct proto dccp_v6_prot = {
 	.compat_setsockopt = compat_dccp_setsockopt,
 	.compat_getsockopt = compat_dccp_getsockopt,
 #endif
-	REF_PROTO_INUSE(dccp_v6)
 };
 
 static struct inet6_protocol dccp_v6_protocol = {
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index d965f0a39c84..349205048557 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -818,8 +818,6 @@ static int raw_ioctl(struct sock *sk, int cmd, unsigned long arg)
 	}
 }
 
-DEFINE_PROTO_INUSE(raw)
-
 struct proto raw_prot = {
 	.name		   = "RAW",
 	.owner		   = THIS_MODULE,
@@ -842,7 +840,6 @@ struct proto raw_prot = {
 	.compat_setsockopt = compat_raw_setsockopt,
 	.compat_getsockopt = compat_raw_getsockopt,
 #endif
-	REF_PROTO_INUSE(raw)
 };
 
 #ifdef CONFIG_PROC_FS
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 2a5881c81778..ef141b841d2f 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2451,8 +2451,6 @@ void tcp4_proc_exit(void)
 }
 #endif /* CONFIG_PROC_FS */
 
-DEFINE_PROTO_INUSE(tcp)
-
 struct proto tcp_prot = {
 	.name			= "TCP",
 	.owner			= THIS_MODULE,
@@ -2488,7 +2486,6 @@ struct proto tcp_prot = {
 	.compat_setsockopt	= compat_tcp_setsockopt,
 	.compat_getsockopt	= compat_tcp_getsockopt,
 #endif
-	REF_PROTO_INUSE(tcp)
 };
 
 void __init tcp_v4_init(void)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 80007c79f12f..0255f373b3e5 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1467,8 +1467,6 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait)
 
 }
 
-DEFINE_PROTO_INUSE(udp)
-
 struct proto udp_prot = {
 	.name		   = "UDP",
 	.owner		   = THIS_MODULE,
@@ -1496,7 +1494,6 @@ struct proto udp_prot = {
 	.compat_setsockopt = compat_udp_setsockopt,
 	.compat_getsockopt = compat_udp_getsockopt,
 #endif
-	REF_PROTO_INUSE(udp)
 };
 
 /* ------------------------------------------------------------------------ */
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 53f3ed496328..7c1e0271c0db 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -34,8 +34,6 @@ static	struct net_protocol udplite_protocol = {
 	.netns_ok	= 1,
 };
 
-DEFINE_PROTO_INUSE(udplite)
-
 struct proto 	udplite_prot = {
 	.name		   = "UDP-Lite",
 	.owner		   = THIS_MODULE,
@@ -60,7 +58,6 @@ struct proto 	udplite_prot = {
 	.compat_setsockopt = compat_udp_setsockopt,
 	.compat_getsockopt = compat_udp_getsockopt,
 #endif
-	REF_PROTO_INUSE(udplite)
 };
 
 static struct inet_protosw udplite4_protosw = {
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 830da4603697..c1e495f105ad 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -1171,8 +1171,6 @@ static int rawv6_init_sk(struct sock *sk)
 	return(0);
 }
 
-DEFINE_PROTO_INUSE(rawv6)
-
 struct proto rawv6_prot = {
 	.name		   = "RAWv6",
 	.owner		   = THIS_MODULE,
@@ -1196,7 +1194,6 @@ struct proto rawv6_prot = {
 	.compat_setsockopt = compat_rawv6_setsockopt,
 	.compat_getsockopt = compat_rawv6_getsockopt,
 #endif
-	REF_PROTO_INUSE(rawv6)
 };
 
 #ifdef CONFIG_PROC_FS
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 323c7e06ef43..6d851c3c3db9 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2140,8 +2140,6 @@ void tcp6_proc_exit(struct net *net)
 }
 #endif
 
-DEFINE_PROTO_INUSE(tcpv6)
-
 struct proto tcpv6_prot = {
 	.name			= "TCPv6",
 	.owner			= THIS_MODULE,
@@ -2177,7 +2175,6 @@ struct proto tcpv6_prot = {
 	.compat_setsockopt	= compat_tcp_setsockopt,
 	.compat_getsockopt	= compat_tcp_getsockopt,
 #endif
-	REF_PROTO_INUSE(tcpv6)
 };
 
 static struct inet6_protocol tcpv6_protocol = {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index aacbc82ecf0f..463ae4d448a3 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -999,8 +999,6 @@ void udp6_proc_exit(struct net *net) {
 
 /* ------------------------------------------------------------------------ */
 
-DEFINE_PROTO_INUSE(udpv6)
-
 struct proto udpv6_prot = {
 	.name		   = "UDPv6",
 	.owner		   = THIS_MODULE,
@@ -1027,7 +1025,6 @@ struct proto udpv6_prot = {
 	.compat_setsockopt = compat_udpv6_setsockopt,
 	.compat_getsockopt = compat_udpv6_getsockopt,
 #endif
-	REF_PROTO_INUSE(udpv6)
 };
 
 static struct inet_protosw udpv6_protosw = {
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 706c5c375a01..c5f5357d115d 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -35,8 +35,6 @@ static struct inet6_protocol udplitev6_protocol = {
 	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
 };
 
-DEFINE_PROTO_INUSE(udplitev6)
-
 struct proto udplitev6_prot = {
 	.name		   = "UDPLITEv6",
 	.owner		   = THIS_MODULE,
@@ -60,7 +58,6 @@ struct proto udplitev6_prot = {
 	.compat_setsockopt = compat_udpv6_setsockopt,
 	.compat_getsockopt = compat_udpv6_getsockopt,
 #endif
-	REF_PROTO_INUSE(udplitev6)
 };
 
 static struct inet_protosw udplite6_protosw = {
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 76c747056dd7..00ebd0610be2 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -6513,8 +6513,6 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk,
 }
 
 
-DEFINE_PROTO_INUSE(sctp)
-
 /* This proto struct describes the ULP interface for SCTP.  */
 struct proto sctp_prot = {
 	.name        =	"SCTP",
@@ -6544,11 +6542,9 @@ struct proto sctp_prot = {
 	.enter_memory_pressure = sctp_enter_memory_pressure,
 	.memory_allocated = &sctp_memory_allocated,
 	.sockets_allocated = &sctp_sockets_allocated,
-	REF_PROTO_INUSE(sctp)
 };
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-DEFINE_PROTO_INUSE(sctpv6)
 
 struct proto sctpv6_prot = {
 	.name		= "SCTPv6",
@@ -6578,6 +6574,5 @@ struct proto sctpv6_prot = {
 	.enter_memory_pressure = sctp_enter_memory_pressure,
 	.memory_allocated = &sctp_memory_allocated,
 	.sockets_allocated = &sctp_sockets_allocated,
-	REF_PROTO_INUSE(sctpv6)
 };
 #endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */
-- 
cgit v1.2.3


From 095d911201b0741e7f326d269a005dba55985acf Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 28 Mar 2008 16:39:58 -0700
Subject: [LIB]: Drop the pcounter itself.

The knock-out. The pcounter abstraction is not used any longer in the
kernel.

Not sure whether this should go via netdev tree, but as far as I
remember it was added via this one, and besides Eric thinks that
Andrew shouldn't mind this.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pcounter.h | 74 ------------------------------------------------
 lib/Makefile             |  1 -
 lib/pcounter.c           | 58 -------------------------------------
 3 files changed, 133 deletions(-)
 delete mode 100644 include/linux/pcounter.h
 delete mode 100644 lib/pcounter.c

(limited to 'include')

diff --git a/include/linux/pcounter.h b/include/linux/pcounter.h
deleted file mode 100644
index a82d9f2628ca..000000000000
--- a/include/linux/pcounter.h
+++ /dev/null
@@ -1,74 +0,0 @@
-#ifndef __LINUX_PCOUNTER_H
-#define __LINUX_PCOUNTER_H
-/*
- * Using a dynamic percpu 'int' variable has a cost :
- * 1) Extra dereference
- * Current per_cpu_ptr() implementation uses an array per 'percpu variable'.
- * 2) memory cost of NR_CPUS*(32+sizeof(void *)) instead of num_possible_cpus()*4
- *
- * This pcounter implementation is an abstraction to be able to use
- * either a static or a dynamic per cpu variable.
- * One dynamic per cpu variable gets a fast & cheap implementation, we can
- * change pcounter implementation too.
- */
-struct pcounter {
-#ifdef CONFIG_SMP
-	void		(*add)(struct pcounter *self, int inc);
-	int		(*getval)(const struct pcounter *self, int cpu);
-	int		*per_cpu_values;
-#else
-	int		val;
-#endif
-};
-
-#ifdef CONFIG_SMP
-#include <linux/percpu.h>
-
-#define DEFINE_PCOUNTER(NAME)						\
-static DEFINE_PER_CPU(int, NAME##_pcounter_values);			\
-static void NAME##_pcounter_add(struct pcounter *self, int val)		\
-{									\
-       __get_cpu_var(NAME##_pcounter_values) += val;			\
-}									\
-static int NAME##_pcounter_getval(const struct pcounter *self, int cpu)	\
-{									\
-	return per_cpu(NAME##_pcounter_values, cpu);			\
-}									\
-
-#define PCOUNTER_MEMBER_INITIALIZER(NAME, MEMBER)		\
-	MEMBER = {						\
-		.add	= NAME##_pcounter_add,			\
-		.getval = NAME##_pcounter_getval,		\
-	}
-
-
-static inline void pcounter_add(struct pcounter *self, int inc)
-{
-	self->add(self, inc);
-}
-
-extern int pcounter_getval(const struct pcounter *self);
-extern int pcounter_alloc(struct pcounter *self);
-extern void pcounter_free(struct pcounter *self);
-
-
-#else /* CONFIG_SMP */
-
-static inline void pcounter_add(struct pcounter *self, int inc)
-{
-	self->val += inc;
-}
-
-static inline int pcounter_getval(const struct pcounter *self)
-{
-	return self->val;
-}
-
-#define DEFINE_PCOUNTER(NAME)
-#define PCOUNTER_MEMBER_INITIALIZER(NAME, MEMBER)
-#define pcounter_alloc(self) 0
-#define pcounter_free(self)
-
-#endif /* CONFIG_SMP */
-
-#endif /* __LINUX_PCOUNTER_H */
diff --git a/lib/Makefile b/lib/Makefile
index 23de261a4c83..4d059d469554 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -61,7 +61,6 @@ obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
 obj-$(CONFIG_TEXTSEARCH_BM) += ts_bm.o
 obj-$(CONFIG_TEXTSEARCH_FSM) += ts_fsm.o
 obj-$(CONFIG_SMP) += percpu_counter.o
-obj-$(CONFIG_SMP) += pcounter.o
 obj-$(CONFIG_AUDIT_GENERIC) += audit.o
 
 obj-$(CONFIG_SWIOTLB) += swiotlb.o
diff --git a/lib/pcounter.c b/lib/pcounter.c
deleted file mode 100644
index 9b56807da93b..000000000000
--- a/lib/pcounter.c
+++ /dev/null
@@ -1,58 +0,0 @@
-/*
- * Define default pcounter functions
- * Note that often used pcounters use dedicated functions to get a speed increase.
- * (see DEFINE_PCOUNTER/REF_PCOUNTER_MEMBER)
- */
-
-#include <linux/module.h>
-#include <linux/pcounter.h>
-#include <linux/smp.h>
-#include <linux/cpumask.h>
-
-static void pcounter_dyn_add(struct pcounter *self, int inc)
-{
-	per_cpu_ptr(self->per_cpu_values, smp_processor_id())[0] += inc;
-}
-
-static int pcounter_dyn_getval(const struct pcounter *self, int cpu)
-{
-	return per_cpu_ptr(self->per_cpu_values, cpu)[0];
-}
-
-int pcounter_getval(const struct pcounter *self)
-{
-	int res = 0, cpu;
-
-	for_each_possible_cpu(cpu)
-		res += self->getval(self, cpu);
-
-	return res;
-}
-EXPORT_SYMBOL_GPL(pcounter_getval);
-
-int pcounter_alloc(struct pcounter *self)
-{
-	int rc = 0;
-	if (self->add == NULL) {
-		self->per_cpu_values = alloc_percpu(int);
-		if (self->per_cpu_values != NULL) {
-			self->add    = pcounter_dyn_add;
-			self->getval = pcounter_dyn_getval;
-		} else
-			rc = 1;
-	}
-	return rc;
-}
-EXPORT_SYMBOL_GPL(pcounter_alloc);
-
-void pcounter_free(struct pcounter *self)
-{
-	if (self->per_cpu_values != NULL) {
-		free_percpu(self->per_cpu_values);
-		self->per_cpu_values = NULL;
-		self->getval = NULL;
-		self->add = NULL;
-	}
-}
-EXPORT_SYMBOL_GPL(pcounter_free);
-
-- 
cgit v1.2.3


From 6f191efe48af62dd5f8917dd21d187f896cd6c81 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 28 Mar 2008 18:23:33 -0700
Subject: [UDP]: Replace struct net on udp_iter_state with seq_net_private.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h | 2 +-
 net/ipv4/udp.c    | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 635940d374ab..77af7d46d868 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -194,7 +194,7 @@ struct udp_seq_afinfo {
 };
 
 struct udp_iter_state {
-	struct net              *net;
+	struct seq_net_private  p;
 	sa_family_t		family;
 	struct hlist_head	*hashtable;
 	int			bucket;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 0255f373b3e5..d695097be2dc 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1503,7 +1503,7 @@ static struct sock *udp_get_first(struct seq_file *seq)
 {
 	struct sock *sk;
 	struct udp_iter_state *state = seq->private;
-	struct net *net = state->net;
+	struct net *net = seq_file_net(seq);
 
 	for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) {
 		struct hlist_node *node;
@@ -1522,7 +1522,7 @@ found:
 static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk)
 {
 	struct udp_iter_state *state = seq->private;
-	struct net *net = state->net;
+	struct net *net = seq_file_net(seq);
 
 	do {
 		sk = sk_next(sk);
@@ -1595,7 +1595,7 @@ static int udp_seq_open(struct inode *inode, struct file *file)
 	s->seq_ops.next		= udp_seq_next;
 	s->seq_ops.show		= afinfo->seq_show;
 	s->seq_ops.stop		= udp_seq_stop;
-	s->net                  = net;
+	s->p.net                = net;
 
 	rc = seq_open(file, &s->seq_ops);
 	if (rc)
@@ -1617,7 +1617,7 @@ static int udp_seq_release(struct inode *inode, struct file *file)
 	struct seq_file *seq = file->private_data;
 	struct udp_iter_state *s = seq->private;
 
-	put_net(s->net);
+	put_net(s->p.net);
 	seq_release_private(inode, file);
 	return 0;
 }
-- 
cgit v1.2.3


From dda61925f84d89e2f2a4597d6298a05a2bc05c20 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 28 Mar 2008 18:24:26 -0700
Subject: [UDP]: Move seq_ops from udp_iter_state to udp_seq_afinfo.

No need to create seq_operations for each instance of 'netstat'.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h  |  3 +--
 net/ipv4/udp.c     | 14 ++++++++------
 net/ipv4/udplite.c |  4 +++-
 net/ipv6/udp.c     |  4 +++-
 net/ipv6/udplite.c |  4 +++-
 5 files changed, 18 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 77af7d46d868..0079d17fd3a3 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -189,8 +189,8 @@ struct udp_seq_afinfo {
 	char			*name;
 	sa_family_t		family;
 	struct hlist_head	*hashtable;
-	int 			(*seq_show) (struct seq_file *m, void *v);
 	struct file_operations	*seq_fops;
+	struct seq_operations	seq_ops;
 };
 
 struct udp_iter_state {
@@ -198,7 +198,6 @@ struct udp_iter_state {
 	sa_family_t		family;
 	struct hlist_head	*hashtable;
 	int			bucket;
-	struct seq_operations	seq_ops;
 };
 
 #ifdef CONFIG_PROC_FS
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index eb0c64da2634..bae94b3eaa43 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1591,13 +1591,9 @@ static int udp_seq_open(struct inode *inode, struct file *file)
 
 	s->family		= afinfo->family;
 	s->hashtable		= afinfo->hashtable;
-	s->seq_ops.start	= udp_seq_start;
-	s->seq_ops.next		= udp_seq_next;
-	s->seq_ops.show		= afinfo->seq_show;
-	s->seq_ops.stop		= udp_seq_stop;
 	s->p.net                = net;
 
-	rc = seq_open(file, &s->seq_ops);
+	rc = seq_open(file, &afinfo->seq_ops);
 	if (rc)
 		goto out_put_net;
 
@@ -1634,6 +1630,10 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo)
 	afinfo->seq_fops->llseek	= seq_lseek;
 	afinfo->seq_fops->release	= udp_seq_release;
 
+	afinfo->seq_ops.start		= udp_seq_start;
+	afinfo->seq_ops.next		= udp_seq_next;
+	afinfo->seq_ops.stop		= udp_seq_stop;
+
 	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops);
 	if (p)
 		p->data = afinfo;
@@ -1690,8 +1690,10 @@ static struct udp_seq_afinfo udp4_seq_afinfo = {
 	.name		= "udp",
 	.family		= AF_INET,
 	.hashtable	= udp_hash,
-	.seq_show	= udp4_seq_show,
 	.seq_fops	= &udp4_seq_fops,
+	.seq_ops	= {
+		.show		= udp4_seq_show,
+	},
 };
 
 static int udp4_proc_init_net(struct net *net)
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 7c1e0271c0db..4add87553bce 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -77,8 +77,10 @@ static struct udp_seq_afinfo udplite4_seq_afinfo = {
 	.name		= "udplite",
 	.family		= AF_INET,
 	.hashtable	= udplite_hash,
-	.seq_show	= udp4_seq_show,
 	.seq_fops	= &udplite4_seq_fops,
+	.seq_ops	= {
+		.show		= udp4_seq_show,
+	},
 };
 
 static int udplite4_proc_init_net(struct net *net)
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 463ae4d448a3..e35921a46689 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -983,8 +983,10 @@ static struct udp_seq_afinfo udp6_seq_afinfo = {
 	.name		= "udp6",
 	.family		= AF_INET6,
 	.hashtable	= udp_hash,
-	.seq_show	= udp6_seq_show,
 	.seq_fops	= &udp6_seq_fops,
+	.seq_ops	= {
+		.show		= udp6_seq_show,
+	},
 };
 
 int udp6_proc_init(struct net *net)
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index c5f5357d115d..710366c3f0a6 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -102,8 +102,10 @@ static struct udp_seq_afinfo udplite6_seq_afinfo = {
 	.name		= "udplite6",
 	.family		= AF_INET6,
 	.hashtable	= udplite_hash,
-	.seq_show	= udp6_seq_show,
 	.seq_fops	= &udplite6_seq_fops,
+	.seq_ops	= {
+		.show		= udp6_seq_show,
+	},
 };
 
 static int udplite6_proc_init_net(struct net *net)
-- 
cgit v1.2.3


From 3ba9441bdf07370670a684e6d95dfc523476677f Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 28 Mar 2008 18:25:32 -0700
Subject: [UDP]: Place file operations directly into udp_seq_afinfo.

No need to have separate never-used variable.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h  |  2 +-
 net/ipv4/udp.c     | 15 ++++++---------
 net/ipv4/udplite.c |  2 --
 net/ipv6/udp.c     |  2 --
 net/ipv6/udplite.c |  2 --
 5 files changed, 7 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 0079d17fd3a3..5cf0e593dda7 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -189,7 +189,7 @@ struct udp_seq_afinfo {
 	char			*name;
 	sa_family_t		family;
 	struct hlist_head	*hashtable;
-	struct file_operations	*seq_fops;
+	struct file_operations	seq_fops;
 	struct seq_operations	seq_ops;
 };
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index e716d2d55e1e..a95dff8063a4 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1596,17 +1596,17 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo)
 	struct proc_dir_entry *p;
 	int rc = 0;
 
-	afinfo->seq_fops->owner		= afinfo->owner;
-	afinfo->seq_fops->open		= udp_seq_open;
-	afinfo->seq_fops->read		= seq_read;
-	afinfo->seq_fops->llseek	= seq_lseek;
-	afinfo->seq_fops->release	= seq_release_net;
+	afinfo->seq_fops.owner		= afinfo->owner;
+	afinfo->seq_fops.open		= udp_seq_open;
+	afinfo->seq_fops.read		= seq_read;
+	afinfo->seq_fops.llseek		= seq_lseek;
+	afinfo->seq_fops.release	= seq_release_net;
 
 	afinfo->seq_ops.start		= udp_seq_start;
 	afinfo->seq_ops.next		= udp_seq_next;
 	afinfo->seq_ops.stop		= udp_seq_stop;
 
-	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops);
+	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, &afinfo->seq_fops);
 	if (p)
 		p->data = afinfo;
 	else
@@ -1617,7 +1617,6 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo)
 void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo)
 {
 	proc_net_remove(net, afinfo->name);
-	memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops));
 }
 
 /* ------------------------------------------------------------------------ */
@@ -1656,13 +1655,11 @@ int udp4_seq_show(struct seq_file *seq, void *v)
 }
 
 /* ------------------------------------------------------------------------ */
-static struct file_operations udp4_seq_fops;
 static struct udp_seq_afinfo udp4_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "udp",
 	.family		= AF_INET,
 	.hashtable	= udp_hash,
-	.seq_fops	= &udp4_seq_fops,
 	.seq_ops	= {
 		.show		= udp4_seq_show,
 	},
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 4add87553bce..148ae72af111 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -71,13 +71,11 @@ static struct inet_protosw udplite4_protosw = {
 };
 
 #ifdef CONFIG_PROC_FS
-static struct file_operations udplite4_seq_fops;
 static struct udp_seq_afinfo udplite4_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "udplite",
 	.family		= AF_INET,
 	.hashtable	= udplite_hash,
-	.seq_fops	= &udplite4_seq_fops,
 	.seq_ops	= {
 		.show		= udp4_seq_show,
 	},
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index e35921a46689..ff8d53ecdd60 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -977,13 +977,11 @@ int udp6_seq_show(struct seq_file *seq, void *v)
 	return 0;
 }
 
-static struct file_operations udp6_seq_fops;
 static struct udp_seq_afinfo udp6_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "udp6",
 	.family		= AF_INET6,
 	.hashtable	= udp_hash,
-	.seq_fops	= &udp6_seq_fops,
 	.seq_ops	= {
 		.show		= udp6_seq_show,
 	},
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 710366c3f0a6..5adf651d0721 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -96,13 +96,11 @@ void udplitev6_exit(void)
 }
 
 #ifdef CONFIG_PROC_FS
-static struct file_operations udplite6_seq_fops;
 static struct udp_seq_afinfo udplite6_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "udplite6",
 	.family		= AF_INET6,
 	.hashtable	= udplite_hash,
-	.seq_fops	= &udplite6_seq_fops,
 	.seq_ops	= {
 		.show		= udp6_seq_show,
 	},
-- 
cgit v1.2.3


From 4ad96d39a2d74c1b2e400b602da2594f5098fc26 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Fri, 28 Mar 2008 18:25:53 -0700
Subject: [UDP]: Remove owner from udp_seq_afinfo.

Move it to udp_seq_afinfo->seq_fops as should be.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h  | 1 -
 net/ipv4/udp.c     | 5 +++--
 net/ipv4/udplite.c | 4 +++-
 net/ipv6/udp.c     | 4 +++-
 net/ipv6/udplite.c | 4 +++-
 5 files changed, 12 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 5cf0e593dda7..24a41fa31641 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -185,7 +185,6 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
 
 /* /proc */
 struct udp_seq_afinfo {
-	struct module		*owner;
 	char			*name;
 	sa_family_t		family;
 	struct hlist_head	*hashtable;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index a95dff8063a4..9143645f9a1b 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1596,7 +1596,6 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo)
 	struct proc_dir_entry *p;
 	int rc = 0;
 
-	afinfo->seq_fops.owner		= afinfo->owner;
 	afinfo->seq_fops.open		= udp_seq_open;
 	afinfo->seq_fops.read		= seq_read;
 	afinfo->seq_fops.llseek		= seq_lseek;
@@ -1656,10 +1655,12 @@ int udp4_seq_show(struct seq_file *seq, void *v)
 
 /* ------------------------------------------------------------------------ */
 static struct udp_seq_afinfo udp4_seq_afinfo = {
-	.owner		= THIS_MODULE,
 	.name		= "udp",
 	.family		= AF_INET,
 	.hashtable	= udp_hash,
+	.seq_fops	= {
+		.owner	=	THIS_MODULE,
+	},
 	.seq_ops	= {
 		.show		= udp4_seq_show,
 	},
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 148ae72af111..72ce26b6c4d3 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -72,10 +72,12 @@ static struct inet_protosw udplite4_protosw = {
 
 #ifdef CONFIG_PROC_FS
 static struct udp_seq_afinfo udplite4_seq_afinfo = {
-	.owner		= THIS_MODULE,
 	.name		= "udplite",
 	.family		= AF_INET,
 	.hashtable	= udplite_hash,
+	.seq_fops	= {
+		.owner	=	THIS_MODULE,
+	},
 	.seq_ops	= {
 		.show		= udp4_seq_show,
 	},
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index ff8d53ecdd60..30ef7dc5d403 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -978,10 +978,12 @@ int udp6_seq_show(struct seq_file *seq, void *v)
 }
 
 static struct udp_seq_afinfo udp6_seq_afinfo = {
-	.owner		= THIS_MODULE,
 	.name		= "udp6",
 	.family		= AF_INET6,
 	.hashtable	= udp_hash,
+	.seq_fops	= {
+		.owner	=	THIS_MODULE,
+	},
 	.seq_ops	= {
 		.show		= udp6_seq_show,
 	},
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 5adf651d0721..491efd00a866 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -97,10 +97,12 @@ void udplitev6_exit(void)
 
 #ifdef CONFIG_PROC_FS
 static struct udp_seq_afinfo udplite6_seq_afinfo = {
-	.owner		= THIS_MODULE,
 	.name		= "udplite6",
 	.family		= AF_INET6,
 	.hashtable	= udplite_hash,
+	.seq_fops	= {
+		.owner	=	THIS_MODULE,
+	},
 	.seq_ops	= {
 		.show		= udp6_seq_show,
 	},
-- 
cgit v1.2.3


From 9307b570a745da4f2d83195f5337927e98221bb2 Mon Sep 17 00:00:00 2001
From: "S.Caglar Onur" <caglar@pardus.org.tr>
Date: Fri, 28 Mar 2008 14:41:24 -0700
Subject: drivers/net/arcnet/arcnet.c: use time_* macros

The functions time_before, time_before_eq, time_after, and time_after_eq are
more robust for comparing jiffies against other values.

So use the time_after() macro, defined in linux/jiffies.h, which deals with
wrapping correctly.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: S.Caglar Onur <caglar@pardus.org.tr>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/arcnet/arcnet.c | 5 +++--
 include/linux/arcdevice.h   | 4 ++--
 2 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/net/arcnet/arcnet.c b/drivers/net/arcnet/arcnet.c
index c59c8067de99..bdc4c0bb56d9 100644
--- a/drivers/net/arcnet/arcnet.c
+++ b/drivers/net/arcnet/arcnet.c
@@ -940,7 +940,7 @@ irqreturn_t arcnet_interrupt(int irq, void *dev_id)
 
 			/* is the RECON info empty or old? */
 			if (!lp->first_recon || !lp->last_recon ||
-			    jiffies - lp->last_recon > HZ * 10) {
+			    time_after(jiffies, lp->last_recon + HZ * 10)) {
 				if (lp->network_down)
 					BUGMSG(D_NORMAL, "reconfiguration detected: cabling restored?\n");
 				lp->first_recon = lp->last_recon = jiffies;
@@ -974,7 +974,8 @@ irqreturn_t arcnet_interrupt(int irq, void *dev_id)
 					lp->num_recons = 1;
 				}
 			}
-		} else if (lp->network_down && jiffies - lp->last_recon > HZ * 10) {
+		} else if (lp->network_down &&
+				time_after(jiffies, lp->last_recon + HZ * 10)) {
 			if (lp->network_down)
 				BUGMSG(D_NORMAL, "cabling restored?\n");
 			lp->first_recon = lp->last_recon = 0;
diff --git a/include/linux/arcdevice.h b/include/linux/arcdevice.h
index fde675872c56..a1916078fd08 100644
--- a/include/linux/arcdevice.h
+++ b/include/linux/arcdevice.h
@@ -283,8 +283,8 @@ struct arcnet_local {
 	int next_buf, first_free_buf;
 
 	/* network "reconfiguration" handling */
-	time_t first_recon,	/* time of "first" RECON message to count */
-		last_recon;	/* time of most recent RECON */
+	unsigned long first_recon; /* time of "first" RECON message to count */
+	unsigned long last_recon;  /* time of most recent RECON */
 	int num_recons;		/* number of RECONs between first and last. */
 	bool network_down;	/* do we think the network is down? */
 
-- 
cgit v1.2.3


From 3edf8fa5ccf10688a9280b5cbca8ed3947c42866 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 31 Mar 2008 00:28:14 -0700
Subject: [NET]: Fix allnoconfig build on powerpc and avr32

As reported by Haavard Skinnemoen and Stephen Rothwell:

> allnoconfig fails with
>
> include/linux/netdevice.h:843: error: implicit declaration of function 'dev_net'
>
> which seems to be because the definition of dev_net is inside #ifdef
> CONFIG_NET, while next_net_device, which calls it, is not.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 8576ca928dae..993758f924be 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -827,6 +827,7 @@ struct packet_type {
 extern rwlock_t				dev_base_lock;		/* Device list lock */
 
 
+#ifdef CONFIG_NET
 #define for_each_netdev(net, d)		\
 		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
 #define for_each_netdev_safe(net, d, n)	\
@@ -850,6 +851,7 @@ static inline struct net_device *first_net_device(struct net *net)
 	return list_empty(&net->dev_base_head) ? NULL :
 		net_device_entry(net->dev_base_head.next);
 }
+#endif
 
 extern int 			netdev_boot_setup_check(struct net_device *dev);
 extern unsigned long		netdev_boot_base(const char *prefix, int unit);
-- 
cgit v1.2.3


From 58e9fee13e579df44922172dbe3c9e3ba3edf7a3 Mon Sep 17 00:00:00 2001
From: Benjamin Marzinski <bmarzins@redhat.com>
Date: Fri, 14 Mar 2008 13:52:52 -0500
Subject: [GFS2] Invalidate cache at correct point

GFS2 wasn't invalidating its cache before it called into the lock manager
with a request that could potentially drop a lock.  This was leaving a
window where the lock could be actually be held by another node, but the
file's page cache would still appear valid, causing coherency problems.
This patch moves the cache invalidation to before the lock manager call
when dropping a lock. It also adds the option to the lock_dlm lock
manager to not use conversion mode deadlock avoidance, which, on a
conversion from shared to exclusive, could internally drop the lock, and
then reacquire in. GFS2 now asks lock_dlm to not do this.  Instead, GFS2
manually drops the lock and reacquires it.

Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
---
 fs/gfs2/glock.c              | 35 +++++++++++++++++++++--------------
 fs/gfs2/incore.h             |  1 +
 fs/gfs2/locking/dlm/lock.c   |  3 ++-
 fs/gfs2/locking/dlm/thread.c | 10 +++++++++-
 fs/gfs2/ops_fstype.c         |  2 +-
 include/linux/lm_interface.h | 10 ++++++++++
 6 files changed, 44 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 63981e2fb835..d636b3e80f5d 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -764,7 +764,7 @@ static void state_change(struct gfs2_glock *gl, unsigned int new_state)
 static void drop_bh(struct gfs2_glock *gl, unsigned int ret)
 {
 	struct gfs2_sbd *sdp = gl->gl_sbd;
-	const struct gfs2_glock_operations *glops = gl->gl_ops;
+	struct gfs2_holder *gh = gl->gl_req_gh;
 
 	gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));
 	gfs2_assert_warn(sdp, list_empty(&gl->gl_holders));
@@ -772,8 +772,14 @@ static void drop_bh(struct gfs2_glock *gl, unsigned int ret)
 
 	state_change(gl, LM_ST_UNLOCKED);
 
-	if (glops->go_inval)
-		glops->go_inval(gl, DIO_METADATA);
+	if (test_and_clear_bit(GLF_CONV_DEADLK, &gl->gl_flags)) {
+		spin_lock(&gl->gl_spin);
+		gh->gh_error = 0;
+		spin_unlock(&gl->gl_spin);
+		gfs2_glock_xmote_th(gl, gl->gl_req_gh);
+		gfs2_glock_put(gl);
+		return;
+	}
 
 	spin_lock(&gl->gl_spin);
 	gfs2_demote_wake(gl);
@@ -794,7 +800,6 @@ static void xmote_bh(struct gfs2_glock *gl, unsigned int ret)
 	struct gfs2_sbd *sdp = gl->gl_sbd;
 	const struct gfs2_glock_operations *glops = gl->gl_ops;
 	struct gfs2_holder *gh = gl->gl_req_gh;
-	int prev_state = gl->gl_state;
 	int op_done = 1;
 
 	if (!gh && (ret & LM_OUT_ST_MASK) == LM_ST_UNLOCKED) {
@@ -808,16 +813,6 @@ static void xmote_bh(struct gfs2_glock *gl, unsigned int ret)
 
 	state_change(gl, ret & LM_OUT_ST_MASK);
 
-	if (prev_state != LM_ST_UNLOCKED && !(ret & LM_OUT_CACHEABLE)) {
-		if (glops->go_inval)
-			glops->go_inval(gl, DIO_METADATA);
-	} else if (gl->gl_state == LM_ST_DEFERRED) {
-		/* We might not want to do this here.
-		   Look at moving to the inode glops. */
-		if (glops->go_inval)
-			glops->go_inval(gl, 0);
-	}
-
 	/*  Deal with each possible exit condition  */
 
 	if (!gh) {
@@ -837,6 +832,14 @@ static void xmote_bh(struct gfs2_glock *gl, unsigned int ret)
 		}
 	} else {
 		spin_lock(&gl->gl_spin);
+		if (ret & LM_OUT_CONV_DEADLK) {
+			gh->gh_error = 0;
+			set_bit(GLF_CONV_DEADLK, &gl->gl_flags);
+			spin_unlock(&gl->gl_spin);
+			gfs2_glock_drop_th(gl);
+			gfs2_glock_put(gl);
+			return;
+		}
 		list_del_init(&gh->gh_list);
 		gh->gh_error = -EIO;
 		if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) 
@@ -910,6 +913,8 @@ static void gfs2_glock_xmote_th(struct gfs2_glock *gl, struct gfs2_holder *gh)
 
 	if (glops->go_xmote_th)
 		glops->go_xmote_th(gl);
+	if (state == LM_ST_DEFERRED && glops->go_inval)
+		glops->go_inval(gl, DIO_METADATA);
 
 	gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));
 	gfs2_assert_warn(sdp, list_empty(&gl->gl_holders));
@@ -952,6 +957,8 @@ static void gfs2_glock_drop_th(struct gfs2_glock *gl)
 
 	if (glops->go_xmote_th)
 		glops->go_xmote_th(gl);
+	if (glops->go_inval)
+		glops->go_inval(gl, DIO_METADATA);
 
 	gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));
 	gfs2_assert_warn(sdp, list_empty(&gl->gl_holders));
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 4ba2ea63119d..9c2c0b90b22a 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -167,6 +167,7 @@ enum {
 	GLF_DEMOTE_IN_PROGRESS	= 6,
 	GLF_LFLUSH		= 7,
 	GLF_WAITERS2		= 8,
+	GLF_CONV_DEADLK		= 9,
 };
 
 struct gfs2_glock {
diff --git a/fs/gfs2/locking/dlm/lock.c b/fs/gfs2/locking/dlm/lock.c
index 542a797ac89a..53a6ab3c0919 100644
--- a/fs/gfs2/locking/dlm/lock.c
+++ b/fs/gfs2/locking/dlm/lock.c
@@ -137,7 +137,8 @@ static inline unsigned int make_flags(struct gdlm_lock *lp,
 
 		/* Conversion deadlock avoidance by DLM */
 
-		if (!test_bit(LFL_FORCE_PROMOTE, &lp->flags) &&
+		if (!(lp->ls->fsflags & LM_MFLAG_CONV_NODROP) &&
+		    !test_bit(LFL_FORCE_PROMOTE, &lp->flags) &&
 		    !(lkf & DLM_LKF_NOQUEUE) &&
 		    cur > DLM_LOCK_NL && req > DLM_LOCK_NL && cur != req)
 			lkf |= DLM_LKF_CONVDEADLK;
diff --git a/fs/gfs2/locking/dlm/thread.c b/fs/gfs2/locking/dlm/thread.c
index 521694fc19d6..e53db6fd28ab 100644
--- a/fs/gfs2/locking/dlm/thread.c
+++ b/fs/gfs2/locking/dlm/thread.c
@@ -135,7 +135,15 @@ static void process_complete(struct gdlm_lock *lp)
 			 lp->lksb.sb_status, lp->lockname.ln_type,
 			 (unsigned long long)lp->lockname.ln_number,
 			 lp->flags);
-		return;
+		if (lp->lksb.sb_status == -EDEADLOCK &&
+		    lp->ls->fsflags & LM_MFLAG_CONV_NODROP) {
+			lp->req = lp->cur;
+			acb.lc_ret |= LM_OUT_CONV_DEADLK;
+			if (lp->cur == DLM_LOCK_IV)
+				lp->lksb.sb_lkid = 0;
+			goto out;
+		} else
+			return;
 	}
 
 	/*
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 5b518f73497a..ef9c6c4f80f6 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -723,7 +723,7 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)
 {
 	char *proto = sdp->sd_proto_name;
 	char *table = sdp->sd_table_name;
-	int flags = 0;
+	int flags = LM_MFLAG_CONV_NODROP;
 	int error;
 
 	if (sdp->sd_args.ar_spectator)
diff --git a/include/linux/lm_interface.h b/include/linux/lm_interface.h
index 1418fdc9ac02..f274997bc283 100644
--- a/include/linux/lm_interface.h
+++ b/include/linux/lm_interface.h
@@ -21,9 +21,15 @@ typedef void (*lm_callback_t) (void *ptr, unsigned int type, void *data);
  * modify the filesystem.  The lock module shouldn't assign a journal to the FS
  * mount.  It shouldn't send recovery callbacks to the FS mount.  If the node
  * dies or withdraws, all locks can be wiped immediately.
+ *
+ * LM_MFLAG_CONV_NODROP
+ * Do not allow the dlm to internally resolve conversion deadlocks by demoting
+ * the lock to unlocked and then reacquiring it in the requested mode. Instead,
+ * it should cancel the request and return LM_OUT_CONV_DEADLK.
  */
 
 #define LM_MFLAG_SPECTATOR	0x00000001
+#define LM_MFLAG_CONV_NODROP	0x00000002
 
 /*
  * lm_lockstruct flags
@@ -110,6 +116,9 @@ typedef void (*lm_callback_t) (void *ptr, unsigned int type, void *data);
  *
  * LM_OUT_ASYNC
  * The result of the request will be returned in an LM_CB_ASYNC callback.
+ *
+ * LM_OUT_CONV_DEADLK
+ * The lock request was canceled do to a conversion deadlock.
  */
 
 #define LM_OUT_ST_MASK		0x00000003
@@ -117,6 +126,7 @@ typedef void (*lm_callback_t) (void *ptr, unsigned int type, void *data);
 #define LM_OUT_CANCELED		0x00000008
 #define LM_OUT_ASYNC		0x00000080
 #define LM_OUT_ERROR		0x00000100
+#define LM_OUT_CONV_DEADLK	0x00000200
 
 /*
  * lm_callback_t types
-- 
cgit v1.2.3


From 8efa6e93cb2666dceafc4844057fdcb9aa324fb7 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 31 Mar 2008 19:41:14 -0700
Subject: [NETNS]: Introduce a netns_core structure.

There's already some stuff on the struct net, that should better
be folded into netns_core structure. I'm making the per-proto inuse
counter be per-net also, which is also a candidate for this, so
introduce this structure and populate it a bit.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  6 ++----
 include/net/netns/core.h    | 13 +++++++++++++
 net/core/sysctl_net_core.c  | 12 ++++++------
 net/socket.c                |  2 +-
 4 files changed, 22 insertions(+), 11 deletions(-)
 create mode 100644 include/net/netns/core.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index f8f3d1a5fc35..c01d45fe08c3 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -8,6 +8,7 @@
 #include <linux/workqueue.h>
 #include <linux/list.h>
 
+#include <net/netns/core.h>
 #include <net/netns/unix.h>
 #include <net/netns/packet.h>
 #include <net/netns/ipv4.h>
@@ -46,10 +47,7 @@ struct net {
 
 	struct sock 		*rtnl;			/* rtnetlink socket */
 
-	/* core sysctls */
-	struct ctl_table_header	*sysctl_core_hdr;
-	int			sysctl_somaxconn;
-
+	struct netns_core	core;
 	struct netns_packet	packet;
 	struct netns_unix	unx;
 	struct netns_ipv4	ipv4;
diff --git a/include/net/netns/core.h b/include/net/netns/core.h
new file mode 100644
index 000000000000..0e8c0f8435d4
--- /dev/null
+++ b/include/net/netns/core.h
@@ -0,0 +1,13 @@
+#ifndef __NETNS_CORE_H__
+#define __NETNS_CORE_H__
+
+struct ctl_table_header;
+
+struct netns_core {
+	/* core sysctls */
+	struct ctl_table_header	*sysctl_hdr;
+
+	int	sysctl_somaxconn;
+};
+
+#endif
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 130338f83ae5..5fc801057244 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -127,7 +127,7 @@ static struct ctl_table net_core_table[] = {
 	{
 		.ctl_name	= NET_CORE_SOMAXCONN,
 		.procname	= "somaxconn",
-		.data		= &init_net.sysctl_somaxconn,
+		.data		= &init_net.core.sysctl_somaxconn,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
@@ -161,7 +161,7 @@ static __net_init int sysctl_core_net_init(struct net *net)
 {
 	struct ctl_table *tbl, *tmp;
 
-	net->sysctl_somaxconn = SOMAXCONN;
+	net->core.sysctl_somaxconn = SOMAXCONN;
 
 	tbl = net_core_table;
 	if (net != &init_net) {
@@ -178,9 +178,9 @@ static __net_init int sysctl_core_net_init(struct net *net)
 		}
 	}
 
-	net->sysctl_core_hdr = register_net_sysctl_table(net,
+	net->core.sysctl_hdr = register_net_sysctl_table(net,
 			net_core_path, tbl);
-	if (net->sysctl_core_hdr == NULL)
+	if (net->core.sysctl_hdr == NULL)
 		goto err_reg;
 
 	return 0;
@@ -196,8 +196,8 @@ static __net_exit void sysctl_core_net_exit(struct net *net)
 {
 	struct ctl_table *tbl;
 
-	tbl = net->sysctl_core_hdr->ctl_table_arg;
-	unregister_net_sysctl_table(net->sysctl_core_hdr);
+	tbl = net->core.sysctl_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->core.sysctl_hdr);
 	BUG_ON(tbl == net_core_table);
 	kfree(tbl);
 }
diff --git a/net/socket.c b/net/socket.c
index 79e5382fd110..9b5c917f8a6b 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1375,7 +1375,7 @@ asmlinkage long sys_listen(int fd, int backlog)
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (sock) {
-		somaxconn = sock_net(sock->sk)->sysctl_somaxconn;
+		somaxconn = sock_net(sock->sk)->core.sysctl_somaxconn;
 		if ((unsigned)backlog > somaxconn)
 			backlog = somaxconn;
 
-- 
cgit v1.2.3


From c29a0bc4dfc4d833eb702b1929cec96a3eeb9f7a Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 31 Mar 2008 19:41:46 -0700
Subject: [SOCK][NETNS]: Add a struct net argument to sock_prot_inuse_add and
 _get.

This counter is about to become per-proto-and-per-net, so we'll need
two arguments to determine which cell in this "table" to work with.

All the places, but proc already pass proper net to it - proc will be
tuned a bit later.

Some indentation with spaces in proc files is done to keep the file
coding style consistent.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h            |  7 ++++---
 include/net/udp.h             |  2 +-
 net/core/sock.c               |  4 ++--
 net/ipv4/inet_hashtables.c    |  8 ++++----
 net/ipv4/inet_timewait_sock.c |  2 +-
 net/ipv4/proc.c               | 11 +++++++----
 net/ipv4/raw.c                |  4 ++--
 net/ipv4/udp.c                |  2 +-
 net/ipv6/inet6_hashtables.c   |  4 ++--
 net/ipv6/ipv6_sockglue.c      | 10 ++++++----
 net/ipv6/proc.c               |  8 ++++----
 11 files changed, 34 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 2a3344f666aa..f4fdd101c9a2 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -635,10 +635,11 @@ static inline void sk_refcnt_debug_release(const struct sock *sk)
 
 #ifdef CONFIG_PROC_FS
 /* Called with local bh disabled */
-extern void sock_prot_inuse_add(struct proto *prot, int inc);
-extern int sock_prot_inuse_get(struct proto *proto);
+extern void sock_prot_inuse_add(struct net *net, struct proto *prot, int inc);
+extern int sock_prot_inuse_get(struct net *net, struct proto *proto);
 #else
-static void inline sock_prot_inuse_add(struct proto *prot, int inc)
+static void inline sock_prot_inuse_add(struct net *net, struct proto *prot,
+		int inc)
 {
 }
 #endif
diff --git a/include/net/udp.h b/include/net/udp.h
index 24a41fa31641..3e55a99b0ba3 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -115,7 +115,7 @@ static inline void udp_lib_unhash(struct sock *sk)
 	write_lock_bh(&udp_hash_lock);
 	if (sk_del_node_init(sk)) {
 		inet_sk(sk)->num = 0;
-		sock_prot_inuse_add(sk->sk_prot, -1);
+		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
 	}
 	write_unlock_bh(&udp_hash_lock);
 }
diff --git a/net/core/sock.c b/net/core/sock.c
index c1ae56eb96ec..6f36ab91bb59 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1949,13 +1949,13 @@ struct prot_inuse {
 static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR);
 static DEFINE_PER_CPU(struct prot_inuse, prot_inuse);
 
-void sock_prot_inuse_add(struct proto *prot, int val)
+void sock_prot_inuse_add(struct net *net, struct proto *prot, int val)
 {
 	__get_cpu_var(prot_inuse).val[prot->inuse_idx] += val;
 }
 EXPORT_SYMBOL_GPL(sock_prot_inuse_add);
 
-int sock_prot_inuse_get(struct proto *prot)
+int sock_prot_inuse_get(struct net *net, struct proto *prot)
 {
 	int cpu, idx = prot->inuse_idx;
 	int res = 0;
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 1b6ff513c75d..32ca2f8b581c 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -288,7 +288,7 @@ unique:
 	sk->sk_hash = hash;
 	BUG_TRAP(sk_unhashed(sk));
 	__sk_add_node(sk, &head->chain);
-	sock_prot_inuse_add(sk->sk_prot, 1);
+	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock(lock);
 
 	if (twp) {
@@ -332,7 +332,7 @@ void __inet_hash_nolisten(struct sock *sk)
 
 	write_lock(lock);
 	__sk_add_node(sk, list);
-	sock_prot_inuse_add(sk->sk_prot, 1);
+	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock(lock);
 }
 EXPORT_SYMBOL_GPL(__inet_hash_nolisten);
@@ -354,7 +354,7 @@ static void __inet_hash(struct sock *sk)
 
 	inet_listen_wlock(hashinfo);
 	__sk_add_node(sk, list);
-	sock_prot_inuse_add(sk->sk_prot, 1);
+	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock(lock);
 	wake_up(&hashinfo->lhash_wait);
 }
@@ -387,7 +387,7 @@ void inet_unhash(struct sock *sk)
 	}
 
 	if (__sk_del_node_init(sk))
-		sock_prot_inuse_add(sk->sk_prot, -1);
+		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
 	write_unlock_bh(lock);
 out:
 	if (sk->sk_state == TCP_LISTEN)
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index f12bc24de46f..a74137866fbc 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -91,7 +91,7 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
 
 	/* Step 2: Remove SK from established hash. */
 	if (__sk_del_node_init(sk))
-		sock_prot_inuse_add(sk->sk_prot, -1);
+		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
 
 	/* Step 3: Hash TW into TIMEWAIT chain. */
 	inet_twsk_add_node(tw, &ehead->twchain);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index d63474c6b400..8156c26f9337 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -53,14 +53,17 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 {
 	socket_seq_show(seq);
 	seq_printf(seq, "TCP: inuse %d orphan %d tw %d alloc %d mem %d\n",
-		   sock_prot_inuse_get(&tcp_prot),
+		   sock_prot_inuse_get(&init_net, &tcp_prot),
 		   atomic_read(&tcp_orphan_count),
 		   tcp_death_row.tw_count, atomic_read(&tcp_sockets_allocated),
 		   atomic_read(&tcp_memory_allocated));
-	seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse_get(&udp_prot),
+	seq_printf(seq, "UDP: inuse %d mem %d\n",
+		   sock_prot_inuse_get(&init_net, &udp_prot),
 		   atomic_read(&udp_memory_allocated));
-	seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot));
-	seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot));
+	seq_printf(seq, "UDPLITE: inuse %d\n",
+		   sock_prot_inuse_get(&init_net, &udplite_prot));
+	seq_printf(seq, "RAW: inuse %d\n",
+		   sock_prot_inuse_get(&init_net, &raw_prot));
 	seq_printf(seq,  "FRAG: inuse %d memory %d\n",
 			ip_frag_nqueues(&init_net), ip_frag_mem(&init_net));
 	return 0;
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 349205048557..11d7f753a820 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -93,7 +93,7 @@ void raw_hash_sk(struct sock *sk)
 
 	write_lock_bh(&h->lock);
 	sk_add_node(sk, head);
-	sock_prot_inuse_add(sk->sk_prot, 1);
+	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock_bh(&h->lock);
 }
 EXPORT_SYMBOL_GPL(raw_hash_sk);
@@ -104,7 +104,7 @@ void raw_unhash_sk(struct sock *sk)
 
 	write_lock_bh(&h->lock);
 	if (sk_del_node_init(sk))
-		sock_prot_inuse_add(sk->sk_prot, -1);
+		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
 	write_unlock_bh(&h->lock);
 }
 EXPORT_SYMBOL_GPL(raw_unhash_sk);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 9143645f9a1b..03bd70697481 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -231,7 +231,7 @@ gotit:
 	if (sk_unhashed(sk)) {
 		head = &udptable[snum & (UDP_HTABLE_SIZE - 1)];
 		sk_add_node(sk, head);
-		sock_prot_inuse_add(sk->sk_prot, 1);
+		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	}
 	error = 0;
 fail:
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 340c7d42b83a..580014aea4d6 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -43,7 +43,7 @@ void __inet6_hash(struct sock *sk)
 	}
 
 	__sk_add_node(sk, list);
-	sock_prot_inuse_add(sk->sk_prot, 1);
+	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock(lock);
 }
 EXPORT_SYMBOL(__inet6_hash);
@@ -204,7 +204,7 @@ unique:
 	BUG_TRAP(sk_unhashed(sk));
 	__sk_add_node(sk, &head->chain);
 	sk->sk_hash = hash;
-	sock_prot_inuse_add(sk->sk_prot, 1);
+	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock(lock);
 
 	if (twp != NULL) {
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index d3d93d752e10..4195ac92345e 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -155,10 +155,11 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 
 			if (sk->sk_protocol == IPPROTO_TCP) {
 				struct inet_connection_sock *icsk = inet_csk(sk);
+				struct net *net = sock_net(sk);
 
 				local_bh_disable();
-				sock_prot_inuse_add(sk->sk_prot, -1);
-				sock_prot_inuse_add(&tcp_prot, 1);
+				sock_prot_inuse_add(net, sk->sk_prot, -1);
+				sock_prot_inuse_add(net, &tcp_prot, 1);
 				local_bh_enable();
 				sk->sk_prot = &tcp_prot;
 				icsk->icsk_af_ops = &ipv4_specific;
@@ -167,12 +168,13 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 				tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
 			} else {
 				struct proto *prot = &udp_prot;
+				struct net *net = sock_net(sk);
 
 				if (sk->sk_protocol == IPPROTO_UDPLITE)
 					prot = &udplite_prot;
 				local_bh_disable();
-				sock_prot_inuse_add(sk->sk_prot, -1);
-				sock_prot_inuse_add(prot, 1);
+				sock_prot_inuse_add(net, sk->sk_prot, -1);
+				sock_prot_inuse_add(net, prot, 1);
 				local_bh_enable();
 				sk->sk_prot = prot;
 				sk->sk_socket->ops = &inet_dgram_ops;
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 364dc332532c..4b9d5a905725 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -36,13 +36,13 @@ static struct proc_dir_entry *proc_net_devsnmp6;
 static int sockstat6_seq_show(struct seq_file *seq, void *v)
 {
 	seq_printf(seq, "TCP6: inuse %d\n",
-		       sock_prot_inuse_get(&tcpv6_prot));
+		       sock_prot_inuse_get(&init_net, &tcpv6_prot));
 	seq_printf(seq, "UDP6: inuse %d\n",
-		       sock_prot_inuse_get(&udpv6_prot));
+		       sock_prot_inuse_get(&init_net, &udpv6_prot));
 	seq_printf(seq, "UDPLITE6: inuse %d\n",
-			sock_prot_inuse_get(&udplitev6_prot));
+			sock_prot_inuse_get(&init_net, &udplitev6_prot));
 	seq_printf(seq, "RAW6: inuse %d\n",
-		       sock_prot_inuse_get(&rawv6_prot));
+		       sock_prot_inuse_get(&init_net, &rawv6_prot));
 	seq_printf(seq, "FRAG6: inuse %d memory %d\n",
 		       ip6_frag_nqueues(&init_net), ip6_frag_mem(&init_net));
 	return 0;
-- 
cgit v1.2.3


From 70ee115942be6ce52ff10e5e813fb4da82cdb25a Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 31 Mar 2008 19:42:16 -0700
Subject: [SOCK][NETNS]: Add the percpu prot_inuse counter in the struct net.

Such an accounting would cost us two more dereferences to get the
percpu variable from the struct net, so I make sock_prot_inuse_get
and _add calls work differently depending on CONFIG_NET_NS - without
it old optimized routines are used.

The per-cpu counter for init_net is prepared in core_initcall, so
that even af_inet, that starts as fs_initcall, will already have the
init_net prepared.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/core.h |  3 +++
 net/core/sock.c          | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 51 insertions(+)

(limited to 'include')

diff --git a/include/net/netns/core.h b/include/net/netns/core.h
index 0e8c0f8435d4..24d4be76bbd1 100644
--- a/include/net/netns/core.h
+++ b/include/net/netns/core.h
@@ -2,12 +2,15 @@
 #define __NETNS_CORE_H__
 
 struct ctl_table_header;
+struct prot_inuse;
 
 struct netns_core {
 	/* core sysctls */
 	struct ctl_table_header	*sysctl_hdr;
 
 	int	sysctl_somaxconn;
+
+	struct prot_inuse	*inuse;
 };
 
 #endif
diff --git a/net/core/sock.c b/net/core/sock.c
index 6f36ab91bb59..83e11f7260ff 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1947,6 +1947,53 @@ struct prot_inuse {
 };
 
 static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR);
+
+#ifdef CONFIG_NET_NS
+void sock_prot_inuse_add(struct net *net, struct proto *prot, int val)
+{
+	int cpu = smp_processor_id();
+	per_cpu_ptr(net->core.inuse, cpu)->val[prot->inuse_idx] += val;
+}
+EXPORT_SYMBOL_GPL(sock_prot_inuse_add);
+
+int sock_prot_inuse_get(struct net *net, struct proto *prot)
+{
+	int cpu, idx = prot->inuse_idx;
+	int res = 0;
+
+	for_each_possible_cpu(cpu)
+		res += per_cpu_ptr(net->core.inuse, cpu)->val[idx];
+
+	return res >= 0 ? res : 0;
+}
+EXPORT_SYMBOL_GPL(sock_prot_inuse_get);
+
+static int sock_inuse_init_net(struct net *net)
+{
+	net->core.inuse = alloc_percpu(struct prot_inuse);
+	return net->core.inuse ? 0 : -ENOMEM;
+}
+
+static void sock_inuse_exit_net(struct net *net)
+{
+	free_percpu(net->core.inuse);
+}
+
+static struct pernet_operations net_inuse_ops = {
+	.init = sock_inuse_init_net,
+	.exit = sock_inuse_exit_net,
+};
+
+static __init int net_inuse_init(void)
+{
+	if (register_pernet_subsys(&net_inuse_ops))
+		panic("Cannot initialize net inuse counters");
+
+	return 0;
+}
+
+core_initcall(net_inuse_init);
+#else
 static DEFINE_PER_CPU(struct prot_inuse, prot_inuse);
 
 void sock_prot_inuse_add(struct net *net, struct proto *prot, int val)
@@ -1966,6 +2013,7 @@ int sock_prot_inuse_get(struct net *net, struct proto *prot)
 	return res >= 0 ? res : 0;
 }
 EXPORT_SYMBOL_GPL(sock_prot_inuse_get);
+#endif
 
 static void assign_proto_idx(struct proto *prot)
 {
-- 
cgit v1.2.3


From 225c0a0107735597100dc4133cd88c5ed10d9e63 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 2 Apr 2008 00:09:29 -0700
Subject: [NETNS]: Merge ifdef CONFIG_NET in include/net/net_namespace.h.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index c01d45fe08c3..4a37037b1d17 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -59,25 +59,27 @@ struct net {
 #endif
 };
 
+
 #ifdef CONFIG_NET
 /* Init's network namespace */
 extern struct net init_net;
 #define INIT_NET_NS(net_ns) .net_ns = &init_net,
-#else
-#define INIT_NET_NS(net_ns)
-#endif
 
-extern struct list_head net_namespace_list;
-
-#ifdef CONFIG_NET
 extern struct net *copy_net_ns(unsigned long flags, struct net *net_ns);
-#else
+
+#else /* CONFIG_NET */
+
+#define INIT_NET_NS(net_ns)
+
 static inline struct net *copy_net_ns(unsigned long flags, struct net *net_ns)
 {
 	/* There is nothing to copy so this is a noop */
 	return net_ns;
 }
-#endif
+#endif /* CONFIG_NET */
+
+
+extern struct list_head net_namespace_list;
 
 #ifdef CONFIG_NET_NS
 extern void __put_net(struct net *net);
-- 
cgit v1.2.3


From c0f39322c335412339dec16ebfd2a05ceba5ebcf Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 2 Apr 2008 00:10:28 -0700
Subject: [NETNS]: Do not include net/net_namespace.h from seq_file.h

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/seq_file.h     | 22 ----------------------
 include/linux/seq_file_net.h | 27 +++++++++++++++++++++++++++
 include/net/net_namespace.h  |  2 ++
 3 files changed, 29 insertions(+), 22 deletions(-)
 create mode 100644 include/linux/seq_file_net.h

(limited to 'include')

diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 5da70c3f4417..1da1e6208a0a 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -5,7 +5,6 @@
 #include <linux/types.h>
 #include <linux/string.h>
 #include <linux/mutex.h>
-#include <net/net_namespace.h>
 
 struct seq_operations;
 struct file;
@@ -63,26 +62,5 @@ extern struct list_head *seq_list_start_head(struct list_head *head,
 extern struct list_head *seq_list_next(void *v, struct list_head *head,
 		loff_t *ppos);
 
-#ifdef CONFIG_NET
-struct net;
-struct seq_net_private {
-#ifdef CONFIG_NET_NS
-	struct net *net;
-#endif
-};
-
-int seq_open_net(struct inode *, struct file *,
-		 const struct seq_operations *, int);
-int seq_release_net(struct inode *, struct file *);
-static inline struct net *seq_file_net(struct seq_file *seq)
-{
-#ifdef CONFIG_NET_NS
-	return ((struct seq_net_private *)seq->private)->net;
-#else
-	return &init_net;
-#endif
-}
-#endif	/* CONFIG_NET */
-
 #endif
 #endif
diff --git a/include/linux/seq_file_net.h b/include/linux/seq_file_net.h
new file mode 100644
index 000000000000..4ac52542a563
--- /dev/null
+++ b/include/linux/seq_file_net.h
@@ -0,0 +1,27 @@
+#ifndef __SEQ_FILE_NET_H__
+#define __SEQ_FILE_NET_H__
+
+#include <linux/seq_file.h>
+
+struct net;
+extern struct net init_net;
+
+struct seq_net_private {
+#ifdef CONFIG_NET_NS
+	struct net *net;
+#endif
+};
+
+int seq_open_net(struct inode *, struct file *,
+		 const struct seq_operations *, int);
+int seq_release_net(struct inode *, struct file *);
+static inline struct net *seq_file_net(struct seq_file *seq)
+{
+#ifdef CONFIG_NET_NS
+	return ((struct seq_net_private *)seq->private)->net;
+#else
+	return &init_net;
+#endif
+}
+
+#endif
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 4a37037b1d17..6c9a48a46685 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -61,6 +61,8 @@ struct net {
 
 
 #ifdef CONFIG_NET
+#include <linux/seq_file_net.h>
+
 /* Init's network namespace */
 extern struct net init_net;
 #define INIT_NET_NS(net_ns) .net_ns = &init_net,
-- 
cgit v1.2.3


From fadf6bf06069138f8e97c9a963be38348ba2708b Mon Sep 17 00:00:00 2001
From: "Templin, Fred L" <Fred.L.Templin@boeing.com>
Date: Tue, 11 Mar 2008 18:35:59 -0400
Subject: [IPV6] SIT: Add PRL management for ISATAP.

This patch updates the Linux the Intra-Site Automatic Tunnel Addressing
Protocol (ISATAP) implementation. It places the ISATAP potential router
list (PRL) in the kernel and adds three new private ioctls for PRL
management.

[Add several changes of structure name, constant names etc. - yoshfuji]

Signed-off-by: Fred L. Templin <fred.l.templin@boeing.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/if_tunnel.h |  18 ++++-
 include/linux/skbuff.h    |   3 +-
 include/net/ipip.h        |   7 ++
 include/net/ndisc.h       |   9 +++
 net/ipv6/ndisc.c          |  24 ++++++
 net/ipv6/route.c          |   2 -
 net/ipv6/sit.c            | 186 +++++++++++++++++++++++++++++++++++-----------
 7 files changed, 199 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h
index 228eb4eb3129..f20c224d544c 100644
--- a/include/linux/if_tunnel.h
+++ b/include/linux/if_tunnel.h
@@ -7,6 +7,9 @@
 #define SIOCADDTUNNEL   (SIOCDEVPRIVATE + 1)
 #define SIOCDELTUNNEL   (SIOCDEVPRIVATE + 2)
 #define SIOCCHGTUNNEL   (SIOCDEVPRIVATE + 3)
+#define SIOCADDPRL      (SIOCDEVPRIVATE + 5)
+#define SIOCDELPRL      (SIOCDEVPRIVATE + 6)
+#define SIOCCHGPRL      (SIOCDEVPRIVATE + 7)
 
 #define GRE_CSUM	__constant_htons(0x8000)
 #define GRE_ROUTING	__constant_htons(0x4000)
@@ -17,9 +20,6 @@
 #define GRE_FLAGS	__constant_htons(0x00F8)
 #define GRE_VERSION	__constant_htons(0x0007)
 
-/* i_flags values for SIT mode */
-#define	SIT_ISATAP	0x0001
-
 struct ip_tunnel_parm
 {
 	char			name[IFNAMSIZ];
@@ -31,4 +31,16 @@ struct ip_tunnel_parm
 	struct iphdr		iph;
 };
 
+/* SIT-mode i_flags */
+#define	SIT_ISATAP	0x0001
+
+struct ip_tunnel_prl {
+	__be32			addr;
+	__u16			flags;
+	__u16			__reserved;
+};
+
+/* PRL flags */
+#define	PRL_DEFAULT		0x0001
+
 #endif /* _IF_TUNNEL_H_ */
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index ff72145d5d9e..e10e55c9b081 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -313,7 +313,8 @@ struct sk_buff {
 	__u16			tc_verd;	/* traffic control verdict */
 #endif
 #endif
-	/* 2 byte hole */
+	__u8			ndisc_nodetype:2;
+	/* 14 bit hole */
 
 #ifdef CONFIG_NET_DMA
 	dma_cookie_t		dma_cookie;
diff --git a/include/net/ipip.h b/include/net/ipip.h
index 549e132bca9c..205536a014e8 100644
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -24,6 +24,13 @@ struct ip_tunnel
 	int			mlink;
 
 	struct ip_tunnel_parm	parms;
+	struct ip_tunnel_prl_entry	*prl;		/* potential router list */
+};
+
+struct ip_tunnel_prl_entry
+{
+	struct ip_tunnel_prl_entry	*next;
+	struct ip_tunnel_prl		entry;
 };
 
 #define IPTUNNEL_XMIT() do {						\
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index 5aedf324de66..9f2bae68d28c 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -11,6 +11,15 @@
 #define NDISC_NEIGHBOUR_ADVERTISEMENT	136
 #define NDISC_REDIRECT			137
 
+/*
+ * Router type: cross-layer information from link-layer to
+ * IPv6 layer reported by certain link types (e.g., RFC4214).
+ */
+#define NDISC_NODETYPE_UNSPEC		0	/* unspecified (default) */
+#define NDISC_NODETYPE_HOST		1	/* host or unauthorized router */
+#define NDISC_NODETYPE_NODEFAULT	2	/* non-default router */
+#define NDISC_NODETYPE_DEFAULT		3	/* default router */
+
 /*
  *	ndisc options
  */
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 510aa747a404..53b546019fd5 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1092,6 +1092,12 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 		return;
 	}
 
+	if (skb->ndisc_nodetype == NDISC_NODETYPE_HOST) {
+		ND_PRINTK2(KERN_WARNING
+			   "ICMPv6 RA: from host or unauthorized router\n");
+		return;
+	}
+
 	/*
 	 *	set the RA_RECV flag in the interface
 	 */
@@ -1115,6 +1121,10 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 		return;
 	}
 
+	/* skip link-specific parameters from interior routers */
+	if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT)
+		goto skip_linkparms;
+
 	if (in6_dev->if_flags & IF_RS_SENT) {
 		/*
 		 *	flag that an RA was received after an RS was sent
@@ -1229,6 +1239,8 @@ skip_defrtr:
 		}
 	}
 
+skip_linkparms:
+
 	/*
 	 *	Process options.
 	 */
@@ -1268,6 +1280,10 @@ skip_defrtr:
 	}
 #endif
 
+	/* skip link-specific ndopts from interior routers */
+	if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT)
+		goto out;
+
 	if (in6_dev->cnf.accept_ra_pinfo && ndopts.nd_opts_pi) {
 		struct nd_opt_hdr *p;
 		for (p = ndopts.nd_opts_pi;
@@ -1331,6 +1347,14 @@ static void ndisc_redirect_rcv(struct sk_buff *skb)
 	int optlen;
 	u8 *lladdr = NULL;
 
+	switch (skb->ndisc_nodetype) {
+	case NDISC_NODETYPE_HOST:
+	case NDISC_NODETYPE_NODEFAULT:
+		ND_PRINTK2(KERN_WARNING
+			   "ICMPv6 Redirect: from host or unauthorized router\n");
+		return;
+	}
+
 	if (!(ipv6_addr_type(&ipv6_hdr(skb)->saddr) & IPV6_ADDR_LINKLOCAL)) {
 		ND_PRINTK2(KERN_WARNING
 			   "ICMPv6 Redirect: source address is not link-local.\n");
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index cd82b6db35ff..f17b2f61891e 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1699,8 +1699,6 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
 	return rt;
 }
 
-EXPORT_SYMBOL(rt6_get_dflt_router);
-
 struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 				     struct net_device *dev,
 				     unsigned int pref)
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 1b8196c8d145..4786419ade0e 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -16,7 +16,7 @@
  *	Changes:
  * Roger Venning <r.venning@telstra.com>:	6to4 support
  * Nate Thompson <nate@thebog.net>:		6to4 support
- * Fred L. Templin <fltemplin@acm.org>:		isatap support
+ * Fred Templin <fred.l.templin@boeing.com>:	isatap support
  */
 
 #include <linux/module.h>
@@ -197,6 +197,119 @@ failed:
 	return NULL;
 }
 
+static struct ip_tunnel_prl_entry *
+ipip6_tunnel_locate_prl(struct ip_tunnel *t, __be32 addr)
+{
+	struct ip_tunnel_prl_entry *p = (struct ip_tunnel_prl_entry *)NULL;
+
+	for (p = t->prl; p; p = p->next)
+		if (p->entry.addr == addr)
+			break;
+	return p;
+
+}
+
+static int
+ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg)
+{
+	struct ip_tunnel_prl_entry *p;
+
+	for (p = t->prl; p; p = p->next) {
+		if (p->entry.addr == a->addr) {
+			if (chg) {
+				p->entry = *a;
+				return 0;
+			}
+			return -EEXIST;
+		}
+	}
+
+	if (chg)
+		return -ENXIO;
+
+	p = kzalloc(sizeof(struct ip_tunnel_prl_entry), GFP_KERNEL);
+	if (!p)
+		return -ENOBUFS;
+
+	p->entry = *a;
+	p->next = t->prl;
+	t->prl = p;
+	return 0;
+}
+
+static int
+ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
+{
+	struct ip_tunnel_prl_entry *x, **p;
+
+	if (a) {
+		for (p = &t->prl; *p; p = &(*p)->next) {
+			if ((*p)->entry.addr == a->addr) {
+				x = *p;
+				*p = x->next;
+				kfree(x);
+				return 0;
+			}
+		}
+		return -ENXIO;
+	} else {
+		while (t->prl) {
+			x = t->prl;
+			t->prl = t->prl->next;
+			kfree(x);
+		}
+	}
+	return 0;
+}
+
+/* copied directly from anycast.c */
+static int
+ipip6_onlink(struct in6_addr *addr, struct net_device *dev)
+{
+	struct inet6_dev	*idev;
+	struct inet6_ifaddr	*ifa;
+	int	onlink;
+
+	onlink = 0;
+	rcu_read_lock();
+	idev = __in6_dev_get(dev);
+	if (idev) {
+		read_lock_bh(&idev->lock);
+		for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) {
+			onlink = ipv6_prefix_equal(addr, &ifa->addr,
+						   ifa->prefix_len);
+			if (onlink)
+				break;
+		}
+		read_unlock_bh(&idev->lock);
+	}
+	rcu_read_unlock();
+	return onlink;
+}
+
+static int
+isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t)
+{
+	struct ip_tunnel_prl_entry *p = ipip6_tunnel_locate_prl(t, iph->saddr);
+	int ok = 1;
+
+	if (p) {
+		if (p->entry.flags & PRL_DEFAULT)
+			skb->ndisc_nodetype = NDISC_NODETYPE_DEFAULT;
+		else
+			skb->ndisc_nodetype = NDISC_NODETYPE_NODEFAULT;
+	} else {
+		struct in6_addr *addr6 = &ipv6_hdr(skb)->saddr;
+		if (ipv6_addr_is_isatap(addr6) &&
+		    (addr6->s6_addr32[3] == iph->saddr) &&
+		    ipip6_onlink(addr6, t->dev))
+			skb->ndisc_nodetype = NDISC_NODETYPE_HOST;
+		else
+			ok = 0;
+	}
+	return ok;
+}
+
 static void ipip6_tunnel_uninit(struct net_device *dev)
 {
 	if (dev == ipip6_fb_tunnel_dev) {
@@ -206,6 +319,7 @@ static void ipip6_tunnel_uninit(struct net_device *dev)
 		dev_put(dev);
 	} else {
 		ipip6_tunnel_unlink(netdev_priv(dev));
+		ipip6_tunnel_del_prl(netdev_priv(dev), 0);
 		dev_put(dev);
 	}
 }
@@ -365,48 +479,6 @@ static inline void ipip6_ecn_decapsulate(struct iphdr *iph, struct sk_buff *skb)
 		IP6_ECN_set_ce(ipv6_hdr(skb));
 }
 
-/* ISATAP (RFC4214) - check source address */
-static int
-isatap_srcok(struct sk_buff *skb, struct iphdr *iph, struct net_device *dev)
-{
-	struct neighbour *neigh;
-	struct dst_entry *dst;
-	struct rt6_info *rt;
-	struct flowi fl;
-	struct in6_addr *addr6;
-	struct in6_addr rtr;
-	struct ipv6hdr *iph6;
-	int ok = 0;
-
-	/* from onlink default router */
-	ipv6_addr_set(&rtr,  htonl(0xFE800000), 0, 0, 0);
-	ipv6_isatap_eui64(rtr.s6_addr + 8, iph->saddr);
-	if ((rt = rt6_get_dflt_router(&rtr, dev))) {
-		dst_release(&rt->u.dst);
-		return 1;
-	}
-
-	iph6 = ipv6_hdr(skb);
-	memset(&fl, 0, sizeof(fl));
-	fl.proto = iph6->nexthdr;
-	ipv6_addr_copy(&fl.fl6_dst, &iph6->saddr);
-	fl.oif = dev->ifindex;
-	security_skb_classify_flow(skb, &fl);
-
-	dst = ip6_route_output(&init_net, NULL, &fl);
-	if (!dst->error && (dst->dev == dev) && (neigh = dst->neighbour)) {
-
-		addr6 = (struct in6_addr*)&neigh->primary_key;
-
-		/* from correct previous hop */
-		if (ipv6_addr_is_isatap(addr6) &&
-		    (addr6->s6_addr32[3] == iph->saddr))
-			ok = 1;
-	}
-	dst_release(dst);
-	return ok;
-}
-
 static int ipip6_rcv(struct sk_buff *skb)
 {
 	struct iphdr *iph;
@@ -427,7 +499,7 @@ static int ipip6_rcv(struct sk_buff *skb)
 		skb->pkt_type = PACKET_HOST;
 
 		if ((tunnel->dev->priv_flags & IFF_ISATAP) &&
-		    !isatap_srcok(skb, iph, tunnel->dev)) {
+		    !isatap_chksrc(skb, iph, tunnel)) {
 			tunnel->stat.rx_errors++;
 			read_unlock(&ipip6_lock);
 			kfree_skb(skb);
@@ -707,6 +779,7 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 {
 	int err = 0;
 	struct ip_tunnel_parm p;
+	struct ip_tunnel_prl prl;
 	struct ip_tunnel *t;
 
 	switch (cmd) {
@@ -806,6 +879,31 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 		err = 0;
 		break;
 
+	case SIOCADDPRL:
+	case SIOCDELPRL:
+	case SIOCCHGPRL:
+		err = -EPERM;
+		if (!capable(CAP_NET_ADMIN))
+			goto done;
+		err = -EINVAL;
+		if (dev == ipip6_fb_tunnel_dev)
+			goto done;
+		err = -EFAULT;
+		if (copy_from_user(&prl, ifr->ifr_ifru.ifru_data, sizeof(prl)))
+			goto done;
+		err = -ENOENT;
+		if (!(t = netdev_priv(dev)))
+			goto done;
+
+		ipip6_tunnel_unlink(t);
+		if (cmd == SIOCDELPRL)
+			err = ipip6_tunnel_del_prl(t, &prl);
+		else
+			err = ipip6_tunnel_add_prl(t, &prl, cmd == SIOCCHGPRL);
+		ipip6_tunnel_link(t);
+		netdev_state_change(dev);
+		break;
+
 	default:
 		err = -EINVAL;
 	}
-- 
cgit v1.2.3


From 300aaeeaab5f447fcf40e911afe96df3de28f0db Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Mon, 24 Mar 2008 18:28:39 +0900
Subject: [IPV6] SIT: Add SIOCGETPRL ioctl to get/dump PRL.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/if_tunnel.h |  4 ++
 include/net/ipip.h        |  5 ++-
 net/ipv6/sit.c            | 96 ++++++++++++++++++++++++++++++++++++++++++-----
 3 files changed, 95 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h
index f20c224d544c..f1fbe9c930d7 100644
--- a/include/linux/if_tunnel.h
+++ b/include/linux/if_tunnel.h
@@ -7,6 +7,7 @@
 #define SIOCADDTUNNEL   (SIOCDEVPRIVATE + 1)
 #define SIOCDELTUNNEL   (SIOCDEVPRIVATE + 2)
 #define SIOCCHGTUNNEL   (SIOCDEVPRIVATE + 3)
+#define SIOCGETPRL      (SIOCDEVPRIVATE + 4)
 #define SIOCADDPRL      (SIOCDEVPRIVATE + 5)
 #define SIOCDELPRL      (SIOCDEVPRIVATE + 6)
 #define SIOCCHGPRL      (SIOCDEVPRIVATE + 7)
@@ -38,6 +39,9 @@ struct ip_tunnel_prl {
 	__be32			addr;
 	__u16			flags;
 	__u16			__reserved;
+	__u32			datalen;
+	__u32			__reserved2;
+	void __user		*data;
 };
 
 /* PRL flags */
diff --git a/include/net/ipip.h b/include/net/ipip.h
index 205536a014e8..633ed4def8e3 100644
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -24,13 +24,16 @@ struct ip_tunnel
 	int			mlink;
 
 	struct ip_tunnel_parm	parms;
+
 	struct ip_tunnel_prl_entry	*prl;		/* potential router list */
+	unsigned int			prl_count;	/* # of entries in PRL */
 };
 
 struct ip_tunnel_prl_entry
 {
 	struct ip_tunnel_prl_entry	*next;
-	struct ip_tunnel_prl		entry;
+	__be32				addr;
+	u16				flags;
 };
 
 #define IPTUNNEL_XMIT() do {						\
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 84c1ed246afb..08a483a8de50 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -203,12 +203,73 @@ __ipip6_tunnel_locate_prl(struct ip_tunnel *t, __be32 addr)
 	struct ip_tunnel_prl_entry *p = (struct ip_tunnel_prl_entry *)NULL;
 
 	for (p = t->prl; p; p = p->next)
-		if (p->entry.addr == addr)
+		if (p->addr == addr)
 			break;
 	return p;
 
 }
 
+static int ipip6_tunnel_get_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
+{
+	struct ip_tunnel_prl *kp;
+	struct ip_tunnel_prl_entry *prl;
+	unsigned int cmax, c = 0, ca, len;
+	int ret = 0;
+
+	cmax = a->datalen / sizeof(*a);
+	if (cmax > 1 && a->addr != htonl(INADDR_ANY))
+		cmax = 1;
+
+	/* For simple GET or for root users,
+	 * we try harder to allocate.
+	 */
+	kp = (cmax <= 1 || capable(CAP_NET_ADMIN)) ?
+		kcalloc(cmax, sizeof(*kp), GFP_KERNEL) :
+		NULL;
+
+	read_lock(&ipip6_lock);
+
+	ca = t->prl_count < cmax ? t->prl_count : cmax;
+
+	if (!kp) {
+		/* We don't try hard to allocate much memory for
+		 * non-root users.
+		 * For root users, retry allocating enough memory for
+		 * the answer.
+		 */
+		kp = kcalloc(ca, sizeof(*kp), GFP_ATOMIC);
+		if (!kp) {
+			ret = -ENOMEM;
+			goto out;
+		}
+	}
+
+	c = 0;
+	for (prl = t->prl; prl; prl = prl->next) {
+		if (c > cmax)
+			break;
+		if (a->addr != htonl(INADDR_ANY) && prl->addr != a->addr)
+			continue;
+		kp[c].addr = prl->addr;
+		kp[c].flags = prl->flags;
+		c++;
+		if (a->addr != htonl(INADDR_ANY))
+			break;
+	}
+out:
+	read_unlock(&ipip6_lock);
+
+	len = sizeof(*kp) * c;
+	ret = len ? copy_to_user(a->data, kp, len) : 0;
+
+	kfree(kp);
+	if (ret)
+		return -EFAULT;
+
+	a->datalen = len;
+	return 0;
+}
+
 static int
 ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg)
 {
@@ -221,7 +282,7 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg)
 	write_lock(&ipip6_lock);
 
 	for (p = t->prl; p; p = p->next) {
-		if (p->entry.addr == a->addr) {
+		if (p->addr == a->addr) {
 			if (chg)
 				goto update;
 			err = -EEXIST;
@@ -242,8 +303,10 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg)
 
 	p->next = t->prl;
 	t->prl = p;
+	t->prl_count++;
 update:
-	p->entry = *a;
+	p->addr = a->addr;
+	p->flags = a->flags;
 out:
 	write_unlock(&ipip6_lock);
 	return err;
@@ -259,10 +322,11 @@ ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
 
 	if (a && a->addr != htonl(INADDR_ANY)) {
 		for (p = &t->prl; *p; p = &(*p)->next) {
-			if ((*p)->entry.addr == a->addr) {
+			if ((*p)->addr == a->addr) {
 				x = *p;
 				*p = x->next;
 				kfree(x);
+				t->prl_count--;
 				goto out;
 			}
 		}
@@ -272,6 +336,7 @@ ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
 			x = t->prl;
 			t->prl = t->prl->next;
 			kfree(x);
+			t->prl_count--;
 		}
 	}
 out:
@@ -313,7 +378,7 @@ isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t)
 	read_lock(&ipip6_lock);
 	p = __ipip6_tunnel_locate_prl(t, iph->saddr);
 	if (p) {
-		if (p->entry.flags & PRL_DEFAULT)
+		if (p->flags & PRL_DEFAULT)
 			skb->ndisc_nodetype = NDISC_NODETYPE_DEFAULT;
 		else
 			skb->ndisc_nodetype = NDISC_NODETYPE_NODEFAULT;
@@ -899,11 +964,12 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 		err = 0;
 		break;
 
+	case SIOCGETPRL:
 	case SIOCADDPRL:
 	case SIOCDELPRL:
 	case SIOCCHGPRL:
 		err = -EPERM;
-		if (!capable(CAP_NET_ADMIN))
+		if (cmd != SIOCGETPRL && !capable(CAP_NET_ADMIN))
 			goto done;
 		err = -EINVAL;
 		if (dev == ipip6_fb_tunnel_dev)
@@ -915,11 +981,23 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 		if (!(t = netdev_priv(dev)))
 			goto done;
 
-		if (cmd == SIOCDELPRL)
+		switch (cmd) {
+		case SIOCGETPRL:
+			err = ipip6_tunnel_get_prl(t, &prl);
+			if (!err && copy_to_user(ifr->ifr_ifru.ifru_data,
+						 &prl, sizeof(prl)))
+				err = -EFAULT;
+			break;
+		case SIOCDELPRL:
 			err = ipip6_tunnel_del_prl(t, &prl);
-		else
+			break;
+		case SIOCADDPRL:
+		case SIOCCHGPRL:
 			err = ipip6_tunnel_add_prl(t, &prl, cmd == SIOCCHGPRL);
-		netdev_state_change(dev);
+			break;
+		}
+		if (cmd != SIOCGETPRL)
+			netdev_state_change(dev);
 		break;
 
 	default:
-- 
cgit v1.2.3


From 52eeeb8481d705e61e2e9aae974e7799a93783e9 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 15 Mar 2008 22:54:23 -0400
Subject: [IPV6]: Unify ip6_onlink() and ipip6_onlink().

Both are identical, let's create ipv6_chk_prefix() and use it
in both places.
---
 include/net/addrconf.h |  4 ++++
 net/ipv6/addrconf.c    | 25 +++++++++++++++++++++++++
 net/ipv6/anycast.c     | 25 +------------------------
 net/ipv6/sit.c         | 27 +--------------------------
 4 files changed, 31 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index d89b0bc7ab75..bdcc863a60a4 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -71,6 +71,10 @@ extern int			ipv6_chk_addr(struct net *net,
 extern int			ipv6_chk_home_addr(struct net *net,
 						   struct in6_addr *addr);
 #endif
+
+extern int			ipv6_chk_prefix(struct in6_addr *addr,
+						struct net_device *dev);
+
 extern struct inet6_ifaddr      *ipv6_get_ifaddr(struct net *net,
 						 struct in6_addr *addr,
 						 struct net_device *dev,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 5ab9973571ef..c17f8c0b933e 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1249,6 +1249,31 @@ int ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr,
 	return ifp != NULL;
 }
 
+int ipv6_chk_prefix(struct in6_addr *addr, struct net_device *dev)
+{
+	struct inet6_dev *idev;
+	struct inet6_ifaddr *ifa;
+	int	onlink;
+
+	onlink = 0;
+	rcu_read_lock();
+	idev = __in6_dev_get(dev);
+	if (idev) {
+		read_lock_bh(&idev->lock);
+		for (ifa = idev->addr_list; ifa; ifa = ifa->if_next) {
+			onlink = ipv6_prefix_equal(addr, &ifa->addr,
+						   ifa->prefix_len);
+			if (onlink)
+				break;
+		}
+		read_unlock_bh(&idev->lock);
+	}
+	rcu_read_unlock();
+	return onlink;
+}
+
+EXPORT_SYMBOL(ipv6_chk_prefix);
+
 struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr,
 				     struct net_device *dev, int strict)
 {
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index 463bd95d6b13..36e817492095 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -48,29 +48,6 @@ static int ipv6_dev_ac_dec(struct net_device *dev, struct in6_addr *addr);
 /* Big ac list lock for all the sockets */
 static DEFINE_RWLOCK(ipv6_sk_ac_lock);
 
-static int
-ip6_onlink(struct in6_addr *addr, struct net_device *dev)
-{
-	struct inet6_dev	*idev;
-	struct inet6_ifaddr	*ifa;
-	int	onlink;
-
-	onlink = 0;
-	rcu_read_lock();
-	idev = __in6_dev_get(dev);
-	if (idev) {
-		read_lock_bh(&idev->lock);
-		for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) {
-			onlink = ipv6_prefix_equal(addr, &ifa->addr,
-						   ifa->prefix_len);
-			if (onlink)
-				break;
-		}
-		read_unlock_bh(&idev->lock);
-	}
-	rcu_read_unlock();
-	return onlink;
-}
 
 /*
  *	socket join an anycast group
@@ -142,7 +119,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 	 * This obviates the need for propagating anycast routes while
 	 * still allowing some non-router anycast participation.
 	 */
-	if (!ip6_onlink(addr, dev)) {
+	if (!ipv6_chk_prefix(addr, dev)) {
 		if (ishost)
 			err = -EADDRNOTAVAIL;
 		if (err)
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 08a483a8de50..cc16fe07bbff 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -344,31 +344,6 @@ out:
 	return 0;
 }
 
-/* copied directly from anycast.c */
-static int
-ipip6_onlink(struct in6_addr *addr, struct net_device *dev)
-{
-	struct inet6_dev	*idev;
-	struct inet6_ifaddr	*ifa;
-	int	onlink;
-
-	onlink = 0;
-	rcu_read_lock();
-	idev = __in6_dev_get(dev);
-	if (idev) {
-		read_lock_bh(&idev->lock);
-		for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) {
-			onlink = ipv6_prefix_equal(addr, &ifa->addr,
-						   ifa->prefix_len);
-			if (onlink)
-				break;
-		}
-		read_unlock_bh(&idev->lock);
-	}
-	rcu_read_unlock();
-	return onlink;
-}
-
 static int
 isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t)
 {
@@ -386,7 +361,7 @@ isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t)
 		struct in6_addr *addr6 = &ipv6_hdr(skb)->saddr;
 		if (ipv6_addr_is_isatap(addr6) &&
 		    (addr6->s6_addr32[3] == iph->saddr) &&
-		    ipip6_onlink(addr6, t->dev))
+		    ipv6_chk_prefix(addr6, t->dev))
 			skb->ndisc_nodetype = NDISC_NODETYPE_HOST;
 		else
 			ok = 0;
-- 
cgit v1.2.3


From de357cc01334a468e4d5b7ba66a17b0d3ca9d63e Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 15 Mar 2008 23:59:18 -0400
Subject: [IPV6] NDISC: Don't rely on node-type hint from L2 unless required.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/skbuff.h |  2 ++
 net/ipv6/Kconfig       |  4 ++++
 net/ipv6/ndisc.c       | 10 ++++++++++
 3 files changed, 16 insertions(+)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index e10e55c9b081..e517701c25ba 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -313,7 +313,9 @@ struct sk_buff {
 	__u16			tc_verd;	/* traffic control verdict */
 #endif
 #endif
+#ifdef CONFIG_IPV6_NDISC_NODETYPE
 	__u8			ndisc_nodetype:2;
+#endif
 	/* 14 bit hole */
 
 #ifdef CONFIG_NET_DMA
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 47263e45bacb..7d2e7f0941ac 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -168,6 +168,7 @@ config IPV6_SIT
 	tristate "IPv6: IPv6-in-IPv4 tunnel (SIT driver)"
 	depends on IPV6
 	select INET_TUNNEL
+	select IPV6_NDISC_NODETYPE
 	default y
 	---help---
 	  Tunneling means encapsulating data of one protocol type within
@@ -178,6 +179,9 @@ config IPV6_SIT
 
 	  Saying M here will produce a module called sit.ko. If unsure, say Y.
 
+config IPV6_NDISC_NODETYPE
+	bool
+
 config IPV6_TUNNEL
 	tristate "IPv6: IP-in-IPv6 tunnel (RFC2473)"
 	select INET6_TUNNEL
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 16273e11e53d..c400b874097a 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1092,11 +1092,13 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 		return;
 	}
 
+#ifdef CONFIG_IPV6_NDISC_NODETYPE
 	if (skb->ndisc_nodetype == NDISC_NODETYPE_HOST) {
 		ND_PRINTK2(KERN_WARNING
 			   "ICMPv6 RA: from host or unauthorized router\n");
 		return;
 	}
+#endif
 
 	/*
 	 *	set the RA_RECV flag in the interface
@@ -1121,9 +1123,11 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 		return;
 	}
 
+#ifdef CONFIG_IPV6_NDISC_NODETYPE
 	/* skip link-specific parameters from interior routers */
 	if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT)
 		goto skip_linkparms;
+#endif
 
 	if (in6_dev->if_flags & IF_RS_SENT) {
 		/*
@@ -1239,7 +1243,9 @@ skip_defrtr:
 		}
 	}
 
+#ifdef CONFIG_IPV6_NDISC_NODETYPE
 skip_linkparms:
+#endif
 
 	/*
 	 *	Process options.
@@ -1286,9 +1292,11 @@ skip_linkparms:
 	}
 #endif
 
+#ifdef CONFIG_IPV6_NDISC_NODETYPE
 	/* skip link-specific ndopts from interior routers */
 	if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT)
 		goto out;
+#endif
 
 	if (in6_dev->cnf.accept_ra_pinfo && ndopts.nd_opts_pi) {
 		struct nd_opt_hdr *p;
@@ -1353,6 +1361,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb)
 	int optlen;
 	u8 *lladdr = NULL;
 
+#ifdef CONFIG_IPV6_NDISC_NODETYPE
 	switch (skb->ndisc_nodetype) {
 	case NDISC_NODETYPE_HOST:
 	case NDISC_NODETYPE_NODEFAULT:
@@ -1360,6 +1369,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb)
 			   "ICMPv6 Redirect: from host or unauthorized router\n");
 		return;
 	}
+#endif
 
 	if (!(ipv6_addr_type(&ipv6_hdr(skb)->saddr) & IPV6_ADDR_LINKLOCAL)) {
 		ND_PRINTK2(KERN_WARNING
-- 
cgit v1.2.3


From 96651896b8d9ad4244a1c3ed9691faa3e382f503 Mon Sep 17 00:00:00 2001
From: Xiantao Zhang <xiantao.zhang@intel.com>
Date: Thu, 3 Apr 2008 11:02:58 -0700
Subject: [IA64] Add API for allocating Dynamic TR resource.

Dynamic TR resource should be managed in the uniform way.
Add two interfaces for kernel:
ia64_itr_entry: Allocate a (pair of) TR for caller.
ia64_ptr_entry: Purge a (pair of ) TR by caller.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Anthony Xu <anthony.xu@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/mca.c     |  49 ++++++++++++
 arch/ia64/kernel/mca_asm.S |   5 ++
 arch/ia64/mm/tlb.c         | 196 +++++++++++++++++++++++++++++++++++++++++++++
 include/asm-ia64/kregs.h   |   3 +
 include/asm-ia64/tlb.h     |  26 ++++++
 5 files changed, 279 insertions(+)

(limited to 'include')

diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index 6c18221dba36..607006a6a976 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -97,6 +97,7 @@
 
 #include <asm/irq.h>
 #include <asm/hw_irq.h>
+#include <asm/tlb.h>
 
 #include "mca_drv.h"
 #include "entry.h"
@@ -112,6 +113,7 @@ DEFINE_PER_CPU(u64, ia64_mca_data); /* == __per_cpu_mca[smp_processor_id()] */
 DEFINE_PER_CPU(u64, ia64_mca_per_cpu_pte); /* PTE to map per-CPU area */
 DEFINE_PER_CPU(u64, ia64_mca_pal_pte);	    /* PTE to map PAL code */
 DEFINE_PER_CPU(u64, ia64_mca_pal_base);    /* vaddr PAL code granule */
+DEFINE_PER_CPU(u64, ia64_mca_tr_reload);   /* Flag for TR reload */
 
 unsigned long __per_cpu_mca[NR_CPUS];
 
@@ -1182,6 +1184,49 @@ all_in:
 	return;
 }
 
+/*  mca_insert_tr
+ *
+ *  Switch rid when TR reload and needed!
+ *  iord: 1: itr, 2: itr;
+ *
+*/
+static void mca_insert_tr(u64 iord)
+{
+
+	int i;
+	u64 old_rr;
+	struct ia64_tr_entry *p;
+	unsigned long psr;
+	int cpu = smp_processor_id();
+
+	psr = ia64_clear_ic();
+	for (i = IA64_TR_ALLOC_BASE; i < IA64_TR_ALLOC_MAX; i++) {
+		p = &__per_cpu_idtrs[cpu][iord-1][i];
+		if (p->pte & 0x1) {
+			old_rr = ia64_get_rr(p->ifa);
+			if (old_rr != p->rr) {
+				ia64_set_rr(p->ifa, p->rr);
+				ia64_srlz_d();
+			}
+			ia64_ptr(iord, p->ifa, p->itir >> 2);
+			ia64_srlz_i();
+			if (iord & 0x1) {
+				ia64_itr(0x1, i, p->ifa, p->pte, p->itir >> 2);
+				ia64_srlz_i();
+			}
+			if (iord & 0x2) {
+				ia64_itr(0x2, i, p->ifa, p->pte, p->itir >> 2);
+				ia64_srlz_i();
+			}
+			if (old_rr != p->rr) {
+				ia64_set_rr(p->ifa, old_rr);
+				ia64_srlz_d();
+			}
+		}
+	}
+	ia64_set_psr(psr);
+}
+
 /*
  * ia64_mca_handler
  *
@@ -1271,6 +1316,10 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw,
 		monarch_cpu = -1;
 #endif
 	}
+	if (__get_cpu_var(ia64_mca_tr_reload)) {
+		mca_insert_tr(0x1); /*Reload dynamic itrs*/
+		mca_insert_tr(0x2); /*Reload dynamic itrs*/
+	}
 	if (notify_die(DIE_MCA_MONARCH_LEAVE, "MCA", regs, (long)&nd, 0, recover)
 			== NOTIFY_STOP)
 		ia64_mca_spin(__func__);
diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S
index 8bc7d259e0c6..a06d46548ff9 100644
--- a/arch/ia64/kernel/mca_asm.S
+++ b/arch/ia64/kernel/mca_asm.S
@@ -219,8 +219,13 @@ ia64_reload_tr:
 	mov r20=IA64_TR_CURRENT_STACK
 	;;
 	itr.d dtr[r20]=r16
+	GET_THIS_PADDR(r2, ia64_mca_tr_reload)
+	mov r18 = 1
 	;;
 	srlz.d
+	;;
+	st8 [r2] =r18
+	;;
 
 done_tlb_purge_and_reload:
 
diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c
index 655da240d13c..3d8903f936a5 100644
--- a/arch/ia64/mm/tlb.c
+++ b/arch/ia64/mm/tlb.c
@@ -26,6 +26,8 @@
 #include <asm/pal.h>
 #include <asm/tlbflush.h>
 #include <asm/dma.h>
+#include <asm/processor.h>
+#include <asm/tlb.h>
 
 static struct {
 	unsigned long mask;	/* mask of supported purge page-sizes */
@@ -39,6 +41,10 @@ struct ia64_ctx ia64_ctx = {
 };
 
 DEFINE_PER_CPU(u8, ia64_need_tlb_flush);
+DEFINE_PER_CPU(u8, ia64_tr_num);  /*Number of TR slots in current processor*/
+DEFINE_PER_CPU(u8, ia64_tr_used); /*Max Slot number used by kernel*/
+
+struct ia64_tr_entry __per_cpu_idtrs[NR_CPUS][2][IA64_TR_ALLOC_MAX];
 
 /*
  * Initializes the ia64_ctx.bitmap array based on max_ctx+1.
@@ -190,6 +196,9 @@ ia64_tlb_init (void)
 	ia64_ptce_info_t uninitialized_var(ptce_info); /* GCC be quiet */
 	unsigned long tr_pgbits;
 	long status;
+	pal_vm_info_1_u_t vm_info_1;
+	pal_vm_info_2_u_t vm_info_2;
+	int cpu = smp_processor_id();
 
 	if ((status = ia64_pal_vm_page_size(&tr_pgbits, &purge.mask)) != 0) {
 		printk(KERN_ERR "PAL_VM_PAGE_SIZE failed with status=%ld; "
@@ -206,4 +215,191 @@ ia64_tlb_init (void)
 	local_cpu_data->ptce_stride[1] = ptce_info.stride[1];
 
 	local_flush_tlb_all();	/* nuke left overs from bootstrapping... */
+	status = ia64_pal_vm_summary(&vm_info_1, &vm_info_2);
+
+	if (status) {
+		printk(KERN_ERR "ia64_pal_vm_summary=%ld\n", status);
+		per_cpu(ia64_tr_num, cpu) = 8;
+		return;
+	}
+	per_cpu(ia64_tr_num, cpu) = vm_info_1.pal_vm_info_1_s.max_itr_entry+1;
+	if (per_cpu(ia64_tr_num, cpu) >
+				(vm_info_1.pal_vm_info_1_s.max_dtr_entry+1))
+		per_cpu(ia64_tr_num, cpu) =
+				vm_info_1.pal_vm_info_1_s.max_dtr_entry+1;
+	if (per_cpu(ia64_tr_num, cpu) > IA64_TR_ALLOC_MAX) {
+		per_cpu(ia64_tr_num, cpu) = IA64_TR_ALLOC_MAX;
+		printk(KERN_DEBUG "TR register number exceeds IA64_TR_ALLOC_MAX!"
+			"IA64_TR_ALLOC_MAX should be extended\n");
+	}
+}
+
+/*
+ * is_tr_overlap
+ *
+ * Check overlap with inserted TRs.
+ */
+static int is_tr_overlap(struct ia64_tr_entry *p, u64 va, u64 log_size)
+{
+	u64 tr_log_size;
+	u64 tr_end;
+	u64 va_rr = ia64_get_rr(va);
+	u64 va_rid = RR_TO_RID(va_rr);
+	u64 va_end = va + (1<<log_size) - 1;
+
+	if (va_rid != RR_TO_RID(p->rr))
+		return 0;
+	tr_log_size = (p->itir & 0xff) >> 2;
+	tr_end = p->ifa + (1<<tr_log_size) - 1;
+
+	if (va > tr_end || p->ifa > va_end)
+		return 0;
+	return 1;
+
+}
+
+/*
+ * ia64_insert_tr in virtual mode. Allocate a TR slot
+ *
+ * target_mask : 0x1 : itr, 0x2 : dtr, 0x3 : idtr
+ *
+ * va 	: virtual address.
+ * pte 	: pte entries inserted.
+ * log_size: range to be covered.
+ *
+ * Return value:  <0 :  error No.
+ *
+ *		  >=0 : slot number allocated for TR.
+ * Must be called with preemption disabled.
+ */
+int ia64_itr_entry(u64 target_mask, u64 va, u64 pte, u64 log_size)
+{
+	int i, r;
+	unsigned long psr;
+	struct ia64_tr_entry *p;
+	int cpu = smp_processor_id();
+
+	r = -EINVAL;
+	/*Check overlap with existing TR entries*/
+	if (target_mask & 0x1) {
+		p = &__per_cpu_idtrs[cpu][0][0];
+		for (i = IA64_TR_ALLOC_BASE; i <= per_cpu(ia64_tr_used, cpu);
+								i++, p++) {
+			if (p->pte & 0x1)
+				if (is_tr_overlap(p, va, log_size)) {
+					printk(KERN_DEBUG "Overlapped Entry"
+						"Inserted for TR Reigster!!\n");
+					goto out;
+			}
+		}
+	}
+	if (target_mask & 0x2) {
+		p = &__per_cpu_idtrs[cpu][1][0];
+		for (i = IA64_TR_ALLOC_BASE; i <= per_cpu(ia64_tr_used, cpu);
+								i++, p++) {
+			if (p->pte & 0x1)
+				if (is_tr_overlap(p, va, log_size)) {
+					printk(KERN_DEBUG "Overlapped Entry"
+						"Inserted for TR Reigster!!\n");
+					goto out;
+				}
+		}
+	}
+
+	for (i = IA64_TR_ALLOC_BASE; i < per_cpu(ia64_tr_num, cpu); i++) {
+		switch (target_mask & 0x3) {
+		case 1:
+			if (!(__per_cpu_idtrs[cpu][0][i].pte & 0x1))
+				goto found;
+			continue;
+		case 2:
+			if (!(__per_cpu_idtrs[cpu][1][i].pte & 0x1))
+				goto found;
+			continue;
+		case 3:
+			if (!(__per_cpu_idtrs[cpu][0][i].pte & 0x1) &&
+				!(__per_cpu_idtrs[cpu][1][i].pte & 0x1))
+				goto found;
+			continue;
+		default:
+			r = -EINVAL;
+			goto out;
+		}
+	}
+found:
+	if (i >= per_cpu(ia64_tr_num, cpu))
+		return -EBUSY;
+
+	/*Record tr info for mca hander use!*/
+	if (i > per_cpu(ia64_tr_used, cpu))
+		per_cpu(ia64_tr_used, cpu) = i;
+
+	psr = ia64_clear_ic();
+	if (target_mask & 0x1) {
+		ia64_itr(0x1, i, va, pte, log_size);
+		ia64_srlz_i();
+		p = &__per_cpu_idtrs[cpu][0][i];
+		p->ifa = va;
+		p->pte = pte;
+		p->itir = log_size << 2;
+		p->rr = ia64_get_rr(va);
+	}
+	if (target_mask & 0x2) {
+		ia64_itr(0x2, i, va, pte, log_size);
+		ia64_srlz_i();
+		p = &__per_cpu_idtrs[cpu][1][i];
+		p->ifa = va;
+		p->pte = pte;
+		p->itir = log_size << 2;
+		p->rr = ia64_get_rr(va);
+	}
+	ia64_set_psr(psr);
+	r = i;
+out:
+	return r;
+}
+EXPORT_SYMBOL_GPL(ia64_itr_entry);
+
+/*
+ * ia64_purge_tr
+ *
+ * target_mask: 0x1: purge itr, 0x2 : purge dtr, 0x3 purge idtr.
+ * slot: slot number to be freed.
+ *
+ * Must be called with preemption disabled.
+ */
+void ia64_ptr_entry(u64 target_mask, int slot)
+{
+	int cpu = smp_processor_id();
+	int i;
+	struct ia64_tr_entry *p;
+
+	if (slot < IA64_TR_ALLOC_BASE || slot >= per_cpu(ia64_tr_num, cpu))
+		return;
+
+	if (target_mask & 0x1) {
+		p = &__per_cpu_idtrs[cpu][0][slot];
+		if ((p->pte&0x1) && is_tr_overlap(p, p->ifa, p->itir>>2)) {
+			p->pte = 0;
+			ia64_ptr(0x1, p->ifa, p->itir>>2);
+			ia64_srlz_i();
+		}
+	}
+
+	if (target_mask & 0x2) {
+		p = &__per_cpu_idtrs[cpu][1][slot];
+		if ((p->pte & 0x1) && is_tr_overlap(p, p->ifa, p->itir>>2)) {
+			p->pte = 0;
+			ia64_ptr(0x2, p->ifa, p->itir>>2);
+			ia64_srlz_i();
+		}
+	}
+
+	for (i = per_cpu(ia64_tr_used, cpu); i >= IA64_TR_ALLOC_BASE; i--) {
+		if ((__per_cpu_idtrs[cpu][0][i].pte & 0x1) ||
+				(__per_cpu_idtrs[cpu][1][i].pte & 0x1))
+			break;
+	}
+	per_cpu(ia64_tr_used, cpu) = i;
 }
+EXPORT_SYMBOL_GPL(ia64_ptr_entry);
diff --git a/include/asm-ia64/kregs.h b/include/asm-ia64/kregs.h
index 7e55a584975c..aefcdfee7f23 100644
--- a/include/asm-ia64/kregs.h
+++ b/include/asm-ia64/kregs.h
@@ -31,6 +31,9 @@
 #define IA64_TR_PALCODE		1	/* itr1: maps PALcode as required by EFI */
 #define IA64_TR_CURRENT_STACK	1	/* dtr1: maps kernel's memory- & register-stacks */
 
+#define IA64_TR_ALLOC_BASE	2 	/* itr&dtr: Base of dynamic TR resource*/
+#define IA64_TR_ALLOC_MAX	32 	/* Max number for dynamic use*/
+
 /* Processor status register bits: */
 #define IA64_PSR_BE_BIT		1
 #define IA64_PSR_UP_BIT		2
diff --git a/include/asm-ia64/tlb.h b/include/asm-ia64/tlb.h
index 26edcb750f9f..20d8a39680c2 100644
--- a/include/asm-ia64/tlb.h
+++ b/include/asm-ia64/tlb.h
@@ -64,6 +64,32 @@ struct mmu_gather {
 	struct page 		*pages[FREE_PTE_NR];
 };
 
+struct ia64_tr_entry {
+	u64 ifa;
+	u64 itir;
+	u64 pte;
+	u64 rr;
+}; /*Record for tr entry!*/
+
+extern int ia64_itr_entry(u64 target_mask, u64 va, u64 pte, u64 log_size);
+extern void ia64_ptr_entry(u64 target_mask, int slot);
+
+extern struct ia64_tr_entry __per_cpu_idtrs[NR_CPUS][2][IA64_TR_ALLOC_MAX];
+
+/*
+ region register macros
+*/
+#define RR_TO_VE(val)   (((val) >> 0) & 0x0000000000000001)
+#define RR_VE(val)	(((val) & 0x0000000000000001) << 0)
+#define RR_VE_MASK	0x0000000000000001L
+#define RR_VE_SHIFT	0
+#define RR_TO_PS(val)	(((val) >> 2) & 0x000000000000003f)
+#define RR_PS(val)	(((val) & 0x000000000000003f) << 2)
+#define RR_PS_MASK	0x00000000000000fcL
+#define RR_PS_SHIFT	2
+#define RR_RID_MASK	0x00000000ffffff00L
+#define RR_TO_RID(val) 	((val >> 8) & 0xffffff)
+
 /* Users of the generic TLB shootdown code must declare this storage space. */
 DECLARE_PER_CPU(struct mmu_gather, mmu_gathers);
 
-- 
cgit v1.2.3


From 31a6b11fed6ceec07ec4bdfefae56b8252d450cf Mon Sep 17 00:00:00 2001
From: Xiantao Zhang <xiantao.zhang@intel.com>
Date: Thu, 3 Apr 2008 11:39:43 -0700
Subject: [IA64] Implement smp_call_function_mask for ia64

This interface provides more flexible functionality for smp
infrastructure ... e.g. KVM frequently needs to operate on
a subset of cpus.

Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/smp.c | 82 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-ia64/smp.h |  3 ++
 2 files changed, 85 insertions(+)

(limited to 'include')

diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c
index 4e446aa5f4ac..9a9d4c489330 100644
--- a/arch/ia64/kernel/smp.c
+++ b/arch/ia64/kernel/smp.c
@@ -209,6 +209,19 @@ send_IPI_allbutself (int op)
 	}
 }
 
+/*
+ * Called with preemption disabled.
+ */
+static inline void
+send_IPI_mask(cpumask_t mask, int op)
+{
+	unsigned int cpu;
+
+	for_each_cpu_mask(cpu, mask) {
+			send_IPI_single(cpu, op);
+	}
+}
+
 /*
  * Called with preemption disabled.
  */
@@ -401,6 +414,75 @@ smp_call_function_single (int cpuid, void (*func) (void *info), void *info, int
 }
 EXPORT_SYMBOL(smp_call_function_single);
 
+/**
+ * smp_call_function_mask(): Run a function on a set of other CPUs.
+ * <mask>	The set of cpus to run on.  Must not include the current cpu.
+ * <func> 	The function to run. This must be fast and non-blocking.
+ * <info>	An arbitrary pointer to pass to the function.
+ * <wait>	If true, wait (atomically) until function
+ *		has completed on other CPUs.
+ *
+ * Returns 0 on success, else a negative status code.
+ *
+ * If @wait is true, then returns once @func has returned; otherwise
+ * it returns just before the target cpu calls @func.
+ *
+ * You must not call this function with disabled interrupts or from a
+ * hardware interrupt handler or from a bottom half handler.
+ */
+int smp_call_function_mask(cpumask_t mask,
+			   void (*func)(void *), void *info,
+			   int wait)
+{
+	struct call_data_struct data;
+	cpumask_t allbutself;
+	int cpus;
+
+	spin_lock(&call_lock);
+	allbutself = cpu_online_map;
+	cpu_clear(smp_processor_id(), allbutself);
+
+	cpus_and(mask, mask, allbutself);
+	cpus = cpus_weight(mask);
+	if (!cpus) {
+		spin_unlock(&call_lock);
+		return 0;
+	}
+
+	/* Can deadlock when called with interrupts disabled */
+	WARN_ON(irqs_disabled());
+
+	data.func = func;
+	data.info = info;
+	atomic_set(&data.started, 0);
+	data.wait = wait;
+	if (wait)
+		atomic_set(&data.finished, 0);
+
+	call_data = &data;
+	mb(); /* ensure store to call_data precedes setting of IPI_CALL_FUNC*/
+
+	/* Send a message to other CPUs */
+	if (cpus_equal(mask, allbutself))
+		send_IPI_allbutself(IPI_CALL_FUNC);
+	else
+		send_IPI_mask(mask, IPI_CALL_FUNC);
+
+	/* Wait for response */
+	while (atomic_read(&data.started) != cpus)
+		cpu_relax();
+
+	if (wait)
+		while (atomic_read(&data.finished) != cpus)
+			cpu_relax();
+	call_data = NULL;
+
+	spin_unlock(&call_lock);
+	return 0;
+
+}
+EXPORT_SYMBOL(smp_call_function_mask);
+
 /*
  * this function sends a 'generic call function' IPI to all other CPUs
  * in the system.
diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h
index 4fa733dd417a..ec5f355fb7e3 100644
--- a/include/asm-ia64/smp.h
+++ b/include/asm-ia64/smp.h
@@ -38,6 +38,9 @@ ia64_get_lid (void)
 	return lid.f.id << 8 | lid.f.eid;
 }
 
+extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
+				  void *info, int wait);
+
 #define hard_smp_processor_id()		ia64_get_lid()
 
 #ifdef CONFIG_SMP
-- 
cgit v1.2.3


From a4aa834a9165150252c5cd953faab4de29d51b87 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 3 Apr 2008 13:04:33 -0700
Subject: [NETNS]: Declare init_net even without CONFIG_NET defined.

This does not look good, but there is no other choice. The compilation
without CONFIG_NET is broken and can not be fixed with ease.

After that there is no need for the following commits:
1567ca7eec7664b8be3b07755ac59dc1b1ec76cb
3edf8fa5ccf10688a9280b5cbca8ed3947c42866
2d38f9a4f8d2ebdc799f03eecf82345825495711

Revert them.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h   | 4 ----
 include/net/net_namespace.h | 3 ++-
 lib/kobject_uevent.c        | 2 --
 3 files changed, 2 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c36c76caf20b..8b17ed40dea2 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -741,7 +741,6 @@ struct net_device
 #define	NETDEV_ALIGN		32
 #define	NETDEV_ALIGN_CONST	(NETDEV_ALIGN - 1)
 
-#ifdef CONFIG_NET
 /*
  * Net namespace inlines
  */
@@ -762,7 +761,6 @@ void dev_net_set(struct net_device *dev, struct net *net)
 	dev->nd_net = net;
 #endif
 }
-#endif
 
 /**
  *	netdev_priv - access network device private data
@@ -827,7 +825,6 @@ struct packet_type {
 extern rwlock_t				dev_base_lock;		/* Device list lock */
 
 
-#ifdef CONFIG_NET
 #define for_each_netdev(net, d)		\
 		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
 #define for_each_netdev_safe(net, d, n)	\
@@ -851,7 +848,6 @@ static inline struct net_device *first_net_device(struct net *net)
 	return list_empty(&net->dev_base_head) ? NULL :
 		net_device_entry(net->dev_base_head.next);
 }
-#endif
 
 extern int 			netdev_boot_setup_check(struct net_device *dev);
 extern unsigned long		netdev_boot_base(const char *prefix, int unit);
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 6c9a48a46685..0ab62ed2fdef 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -60,11 +60,12 @@ struct net {
 };
 
 
-#ifdef CONFIG_NET
 #include <linux/seq_file_net.h>
 
 /* Init's network namespace */
 extern struct net init_net;
+
+#ifdef CONFIG_NET
 #define INIT_NET_NS(net_ns) .net_ns = &init_net,
 
 extern struct net *copy_net_ns(unsigned long flags, struct net *net_ns);
diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c
index 0d56dad319ad..b06185ed1895 100644
--- a/lib/kobject_uevent.c
+++ b/lib/kobject_uevent.c
@@ -19,12 +19,10 @@
 #include <linux/kobject.h>
 #include <linux/module.h>
 
-#ifdef CONFIG_NET
 #include <linux/socket.h>
 #include <linux/skbuff.h>
 #include <linux/netlink.h>
 #include <net/sock.h>
-#endif
 
 
 u64 uevent_seqnum;
-- 
cgit v1.2.3


From 3d58b5fa8e4c461ab09afdacd3d1754fccca06ad Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 3 Apr 2008 14:22:32 -0700
Subject: [INET]: Rename inet_csk_ctl_sock_create to inet_ctl_sock_create.

This call is nothing common with INET connection sockets code. It
simply creates an unhashes kernel sockets for protocol messages.

Move the new call into af_inet.c after the rename.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_common.h          |  5 +++++
 include/net/inet_connection_sock.h |  5 -----
 net/dccp/ipv4.c                    |  4 ++--
 net/dccp/ipv6.c                    |  4 ++--
 net/ipv4/af_inet.c                 | 19 +++++++++++++++++++
 net/ipv4/inet_connection_sock.c    | 19 -------------------
 net/ipv4/tcp_ipv4.c                |  4 ++--
 net/ipv6/tcp_ipv6.c                |  3 ++-
 8 files changed, 32 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_common.h b/include/net/inet_common.h
index 38d5a1e9980d..d6238bdefbaf 100644
--- a/include/net/inet_common.h
+++ b/include/net/inet_common.h
@@ -39,6 +39,11 @@ extern int			inet_getname(struct socket *sock,
 extern int			inet_ioctl(struct socket *sock, 
 					   unsigned int cmd, unsigned long arg);
 
+extern int			inet_ctl_sock_create(struct socket **sock,
+						     unsigned short family,
+						     unsigned short type,
+						     unsigned char protocol);
+
 #endif
 
 
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index f00f0573627b..2ff545a56fb5 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -327,11 +327,6 @@ extern void inet_csk_listen_stop(struct sock *sk);
 
 extern void inet_csk_addr2sockaddr(struct sock *sk, struct sockaddr *uaddr);
 
-extern int inet_csk_ctl_sock_create(struct socket **sock,
-				    unsigned short family,
-				    unsigned short type,
-				    unsigned char protocol);
-
 extern int inet_csk_compat_getsockopt(struct sock *sk, int level, int optname,
 				      char __user *optval, int __user *optlen);
 extern int inet_csk_compat_setsockopt(struct sock *sk, int level, int optname,
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 6d8f684a7a49..feb3fa5b7141 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -1003,8 +1003,8 @@ static int __init dccp_v4_init(void)
 
 	inet_register_protosw(&dccp_v4_protosw);
 
-	err = inet_csk_ctl_sock_create(&socket, PF_INET,
-				       SOCK_DCCP, IPPROTO_DCCP);
+	err = inet_ctl_sock_create(&socket, PF_INET,
+				   SOCK_DCCP, IPPROTO_DCCP);
 	if (err)
 		goto out_unregister_protosw;
 	dccp_v4_ctl_sk = socket->sk;
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index c5d9d1be56f3..5690fbd3bf68 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -1185,8 +1185,8 @@ static int __init dccp_v6_init(void)
 
 	inet6_register_protosw(&dccp_v6_protosw);
 
-	err = inet_csk_ctl_sock_create(&socket, PF_INET6,
-				       SOCK_DCCP, IPPROTO_DCCP);
+	err = inet_ctl_sock_create(&socket, PF_INET6,
+				   SOCK_DCCP, IPPROTO_DCCP);
 	if (err != 0)
 		goto out_unregister_protosw;
 	dccp_v6_ctl_sk = socket->sk;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 5882a1316441..7ab0bd64c9d1 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1250,6 +1250,25 @@ out:
 	return segs;
 }
 
+int inet_ctl_sock_create(struct socket **sock, unsigned short family,
+			 unsigned short type, unsigned char protocol)
+{
+	int rc = sock_create_kern(family, type, protocol, sock);
+
+	if (rc == 0) {
+		(*sock)->sk->sk_allocation = GFP_ATOMIC;
+		inet_sk((*sock)->sk)->uc_ttl = -1;
+		/*
+		 * Unhash it so that IP input processing does not even see it,
+		 * we do not wish this socket to see incoming packets.
+		 */
+		(*sock)->sk->sk_prot->unhash((*sock)->sk);
+	}
+	return rc;
+}
+
+EXPORT_SYMBOL_GPL(inet_ctl_sock_create);
+
 unsigned long snmp_fold_field(void *mib[], int offt)
 {
 	unsigned long res = 0;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index a7fcaf205644..ee55678a987d 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -651,25 +651,6 @@ void inet_csk_addr2sockaddr(struct sock *sk, struct sockaddr *uaddr)
 
 EXPORT_SYMBOL_GPL(inet_csk_addr2sockaddr);
 
-int inet_csk_ctl_sock_create(struct socket **sock, unsigned short family,
-			     unsigned short type, unsigned char protocol)
-{
-	int rc = sock_create_kern(family, type, protocol, sock);
-
-	if (rc == 0) {
-		(*sock)->sk->sk_allocation = GFP_ATOMIC;
-		inet_sk((*sock)->sk)->uc_ttl = -1;
-		/*
-		 * Unhash it so that IP input processing does not even see it,
-		 * we do not wish this socket to see incoming packets.
-		 */
-		(*sock)->sk->sk_prot->unhash((*sock)->sk);
-	}
-	return rc;
-}
-
-EXPORT_SYMBOL_GPL(inet_csk_ctl_sock_create);
-
 #ifdef CONFIG_COMPAT
 int inet_csk_compat_getsockopt(struct sock *sk, int level, int optname,
 			       char __user *optval, int __user *optlen)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 1d77f37d7708..edf5a37bb5c3 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2491,8 +2491,8 @@ struct proto tcp_prot = {
 void __init tcp_v4_init(void)
 {
 	struct socket *__tcp_socket;
-	if (inet_csk_ctl_sock_create(&__tcp_socket, PF_INET, SOCK_RAW,
-				     IPPROTO_TCP) < 0)
+	if (inet_ctl_sock_create(&__tcp_socket, PF_INET, SOCK_RAW,
+				 IPPROTO_TCP) < 0)
 		panic("Failed to create the TCP control socket.\n");
 	tcp_sock = __tcp_socket->sk;
 }
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 6d851c3c3db9..d98222fba041 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -60,6 +60,7 @@
 #include <net/dsfield.h>
 #include <net/timewait_sock.h>
 #include <net/netdma.h>
+#include <net/inet_common.h>
 
 #include <asm/uaccess.h>
 
@@ -2202,7 +2203,7 @@ static int tcpv6_net_init(struct net *net)
 	struct socket *sock;
 	struct sock *sk;
 
-	err = inet_csk_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP);
+	err = inet_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP);
 	if (err)
 		return err;
 
-- 
cgit v1.2.3


From eee4fe4ded6e9c196168aee8f9787771f4df9c90 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 3 Apr 2008 14:27:58 -0700
Subject: [INET]: Let inet_ctl_sock_create return sock rather than socket.

All upper protocol layers are already use sock internally.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_common.h |  2 +-
 net/dccp/ipv4.c           |  4 +---
 net/dccp/ipv6.c           |  4 +---
 net/ipv4/af_inet.c        | 12 +++++++-----
 net/ipv4/tcp_ipv4.c       |  4 +---
 net/ipv6/tcp_ipv6.c       |  5 ++---
 net/sctp/protocol.c       |  4 +---
 7 files changed, 14 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_common.h b/include/net/inet_common.h
index d6238bdefbaf..4bfcf3f3555f 100644
--- a/include/net/inet_common.h
+++ b/include/net/inet_common.h
@@ -39,7 +39,7 @@ extern int			inet_getname(struct socket *sock,
 extern int			inet_ioctl(struct socket *sock, 
 					   unsigned int cmd, unsigned long arg);
 
-extern int			inet_ctl_sock_create(struct socket **sock,
+extern int			inet_ctl_sock_create(struct sock **sk,
 						     unsigned short family,
 						     unsigned short type,
 						     unsigned char protocol);
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index feb3fa5b7141..5669c895c873 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -991,7 +991,6 @@ static struct inet_protosw dccp_v4_protosw = {
 
 static int __init dccp_v4_init(void)
 {
-	struct socket *socket;
 	int err = proto_register(&dccp_v4_prot, 1);
 
 	if (err != 0)
@@ -1003,11 +1002,10 @@ static int __init dccp_v4_init(void)
 
 	inet_register_protosw(&dccp_v4_protosw);
 
-	err = inet_ctl_sock_create(&socket, PF_INET,
+	err = inet_ctl_sock_create(&dccp_v4_ctl_sk, PF_INET,
 				   SOCK_DCCP, IPPROTO_DCCP);
 	if (err)
 		goto out_unregister_protosw;
-	dccp_v4_ctl_sk = socket->sk;
 out:
 	return err;
 out_unregister_protosw:
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 5690fbd3bf68..cf598bfc6a18 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -1173,7 +1173,6 @@ static struct inet_protosw dccp_v6_protosw = {
 
 static int __init dccp_v6_init(void)
 {
-	struct socket *socket;
 	int err = proto_register(&dccp_v6_prot, 1);
 
 	if (err != 0)
@@ -1185,11 +1184,10 @@ static int __init dccp_v6_init(void)
 
 	inet6_register_protosw(&dccp_v6_protosw);
 
-	err = inet_ctl_sock_create(&socket, PF_INET6,
+	err = inet_ctl_sock_create(&dccp_v6_ctl_sk, PF_INET6,
 				   SOCK_DCCP, IPPROTO_DCCP);
 	if (err != 0)
 		goto out_unregister_protosw;
-	dccp_v6_ctl_sk = socket->sk;
 out:
 	return err;
 out_unregister_protosw:
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 7ab0bd64c9d1..cad664bf3f2e 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1250,19 +1250,21 @@ out:
 	return segs;
 }
 
-int inet_ctl_sock_create(struct socket **sock, unsigned short family,
+int inet_ctl_sock_create(struct sock **sk, unsigned short family,
 			 unsigned short type, unsigned char protocol)
 {
-	int rc = sock_create_kern(family, type, protocol, sock);
+	struct socket *sock;
+	int rc = sock_create_kern(family, type, protocol, &sock);
 
 	if (rc == 0) {
-		(*sock)->sk->sk_allocation = GFP_ATOMIC;
-		inet_sk((*sock)->sk)->uc_ttl = -1;
+		*sk = sock->sk;
+		(*sk)->sk_allocation = GFP_ATOMIC;
+		inet_sk(*sk)->uc_ttl = -1;
 		/*
 		 * Unhash it so that IP input processing does not even see it,
 		 * we do not wish this socket to see incoming packets.
 		 */
-		(*sock)->sk->sk_prot->unhash((*sock)->sk);
+		(*sk)->sk_prot->unhash(*sk);
 	}
 	return rc;
 }
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index edf5a37bb5c3..cfe5df76e14b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2490,11 +2490,9 @@ struct proto tcp_prot = {
 
 void __init tcp_v4_init(void)
 {
-	struct socket *__tcp_socket;
-	if (inet_ctl_sock_create(&__tcp_socket, PF_INET, SOCK_RAW,
+	if (inet_ctl_sock_create(&tcp_sock, PF_INET, SOCK_RAW,
 				 IPPROTO_TCP) < 0)
 		panic("Failed to create the TCP control socket.\n");
-	tcp_sock = __tcp_socket->sk;
 }
 
 EXPORT_SYMBOL(ipv4_specific);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index d98222fba041..2882cc51669e 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2200,14 +2200,13 @@ static struct inet_protosw tcpv6_protosw = {
 static int tcpv6_net_init(struct net *net)
 {
 	int err;
-	struct socket *sock;
 	struct sock *sk;
 
-	err = inet_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP);
+	err = inet_ctl_sock_create(&sk, PF_INET6, SOCK_RAW, IPPROTO_TCP);
 	if (err)
 		return err;
 
-	net->ipv6.tcp_sk = sk = sock->sk;
+	net->ipv6.tcp_sk = sk;
 	sk_change_net(sk, net);
 	return err;
 }
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index ac0833c19450..3c08d334d4a8 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -674,21 +674,19 @@ static int sctp_ctl_sock_init(void)
 {
 	int err;
 	sa_family_t family;
-	struct socket *socket;
 
 	if (sctp_get_pf_specific(PF_INET6))
 		family = PF_INET6;
 	else
 		family = PF_INET;
 
-	err = inet_ctl_sock_create(&socket, family,
+	err = inet_ctl_sock_create(&sctp_ctl_sock, family,
 				   SOCK_SEQPACKET, IPPROTO_SCTP);
 	if (err < 0) {
 		printk(KERN_ERR
 		       "SCTP: Failed to create the SCTP control socket.\n");
 		return err;
 	}
-	sctp_ctl_sock = socket->sk;
 	return 0;
 }
 
-- 
cgit v1.2.3


From 5677242f432102dea9e6eceec1dc089e2f709ca4 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 3 Apr 2008 14:28:30 -0700
Subject: [NETNS]: Inet control socket should not hold a namespace.

This is a generic requirement, so make inet_ctl_sock_create namespace
aware and create a inet_ctl_sock_destroy wrapper around
sk_release_kernel.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_common.h |  8 +++++++-
 net/dccp/ipv4.c           |  4 ++--
 net/dccp/ipv6.c           |  4 ++--
 net/ipv4/af_inet.c        |  5 ++++-
 net/ipv4/tcp_ipv4.c       |  2 +-
 net/ipv6/tcp_ipv6.c       | 14 +++-----------
 net/sctp/protocol.c       |  6 +++---
 7 files changed, 22 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_common.h b/include/net/inet_common.h
index 4bfcf3f3555f..18c773286b91 100644
--- a/include/net/inet_common.h
+++ b/include/net/inet_common.h
@@ -42,7 +42,13 @@ extern int			inet_ioctl(struct socket *sock,
 extern int			inet_ctl_sock_create(struct sock **sk,
 						     unsigned short family,
 						     unsigned short type,
-						     unsigned char protocol);
+						     unsigned char protocol,
+						     struct net *net);
+
+static inline void inet_ctl_sock_destroy(struct sock *sk)
+{
+	sk_release_kernel(sk);
+}
 
 #endif
 
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 5669c895c873..b12803bcba56 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -1003,7 +1003,7 @@ static int __init dccp_v4_init(void)
 	inet_register_protosw(&dccp_v4_protosw);
 
 	err = inet_ctl_sock_create(&dccp_v4_ctl_sk, PF_INET,
-				   SOCK_DCCP, IPPROTO_DCCP);
+				   SOCK_DCCP, IPPROTO_DCCP, &init_net);
 	if (err)
 		goto out_unregister_protosw;
 out:
@@ -1018,7 +1018,7 @@ out_proto_unregister:
 
 static void __exit dccp_v4_exit(void)
 {
-	sock_release(dccp_v4_ctl_sk->sk_socket);
+	inet_ctl_sock_destroy(dccp_v4_ctl_sk);
 	inet_unregister_protosw(&dccp_v4_protosw);
 	inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP);
 	proto_unregister(&dccp_v4_prot);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index cf598bfc6a18..94d749e6d494 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -1185,7 +1185,7 @@ static int __init dccp_v6_init(void)
 	inet6_register_protosw(&dccp_v6_protosw);
 
 	err = inet_ctl_sock_create(&dccp_v6_ctl_sk, PF_INET6,
-				   SOCK_DCCP, IPPROTO_DCCP);
+				   SOCK_DCCP, IPPROTO_DCCP, &init_net);
 	if (err != 0)
 		goto out_unregister_protosw;
 out:
@@ -1200,7 +1200,7 @@ out_unregister_proto:
 
 static void __exit dccp_v6_exit(void)
 {
-	sock_release(dccp_v6_ctl_sk->sk_socket);
+	inet_ctl_sock_destroy(dccp_v6_ctl_sk);
 	inet6_del_protocol(&dccp_v6_protocol, IPPROTO_DCCP);
 	inet6_unregister_protosw(&dccp_v6_protosw);
 	proto_unregister(&dccp_v6_prot);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index cad664bf3f2e..cf766ad15776 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1251,7 +1251,8 @@ out:
 }
 
 int inet_ctl_sock_create(struct sock **sk, unsigned short family,
-			 unsigned short type, unsigned char protocol)
+			 unsigned short type, unsigned char protocol,
+			 struct net *net)
 {
 	struct socket *sock;
 	int rc = sock_create_kern(family, type, protocol, &sock);
@@ -1265,6 +1266,8 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family,
 		 * we do not wish this socket to see incoming packets.
 		 */
 		(*sk)->sk_prot->unhash(*sk);
+
+		sk_change_net(*sk, net);
 	}
 	return rc;
 }
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index cfe5df76e14b..dc8c3dc75fe5 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2491,7 +2491,7 @@ struct proto tcp_prot = {
 void __init tcp_v4_init(void)
 {
 	if (inet_ctl_sock_create(&tcp_sock, PF_INET, SOCK_RAW,
-				 IPPROTO_TCP) < 0)
+				 IPPROTO_TCP, &init_net) < 0)
 		panic("Failed to create the TCP control socket.\n");
 }
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 2882cc51669e..378cc4002a76 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2199,21 +2199,13 @@ static struct inet_protosw tcpv6_protosw = {
 
 static int tcpv6_net_init(struct net *net)
 {
-	int err;
-	struct sock *sk;
-
-	err = inet_ctl_sock_create(&sk, PF_INET6, SOCK_RAW, IPPROTO_TCP);
-	if (err)
-		return err;
-
-	net->ipv6.tcp_sk = sk;
-	sk_change_net(sk, net);
-	return err;
+	return inet_ctl_sock_create(&net->ipv6.tcp_sk, PF_INET6,
+				    SOCK_RAW, IPPROTO_TCP, net);
 }
 
 static void tcpv6_net_exit(struct net *net)
 {
-	sk_release_kernel(net->ipv6.tcp_sk);
+	inet_ctl_sock_destroy(net->ipv6.tcp_sk);
 }
 
 static struct pernet_operations tcpv6_net_ops = {
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index 3c08d334d4a8..067c8a1658d6 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -681,7 +681,7 @@ static int sctp_ctl_sock_init(void)
 		family = PF_INET;
 
 	err = inet_ctl_sock_create(&sctp_ctl_sock, family,
-				   SOCK_SEQPACKET, IPPROTO_SCTP);
+				   SOCK_SEQPACKET, IPPROTO_SCTP, &init_net);
 	if (err < 0) {
 		printk(KERN_ERR
 		       "SCTP: Failed to create the SCTP control socket.\n");
@@ -1284,7 +1284,7 @@ err_v6_add_protocol:
 	sctp_v6_del_protocol();
 err_add_protocol:
 	sctp_v4_del_protocol();
-	sock_release(sctp_ctl_sock->sk_socket);
+	inet_ctl_sock_destroy(sctp_ctl_sock);
 err_ctl_sock_init:
 	sctp_v6_protosw_exit();
 err_v6_protosw_init:
@@ -1328,7 +1328,7 @@ SCTP_STATIC __exit void sctp_exit(void)
 	sctp_v4_del_protocol();
 
 	/* Free the control endpoint.  */
-	sock_release(sctp_ctl_sock->sk_socket);
+	inet_ctl_sock_destroy(sctp_ctl_sock);
 
 	/* Free protosw registrations */
 	sctp_v6_protosw_exit();
-- 
cgit v1.2.3


From 046ee902357adc046d041441956ec7eeb30c77c4 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 3 Apr 2008 14:31:33 -0700
Subject: [NETNS]: Create tcp control socket in the each namespace.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/ipv4.h |  1 +
 net/ipv4/tcp_ipv4.c      | 21 +++++++++++++++++++--
 2 files changed, 20 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index af685f71f4b5..34ee348a2cf2 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -28,6 +28,7 @@ struct netns_ipv4 {
 	struct sock		*fibnl;
 
 	struct sock		**icmp_sk;
+	struct sock		*tcp_sock;
 
 	struct netns_frags	frags;
 #ifdef CONFIG_NETFILTER
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index dc8c3dc75fe5..1d4a77acdc0b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2488,11 +2488,28 @@ struct proto tcp_prot = {
 #endif
 };
 
+
+static int __net_init tcp_sk_init(struct net *net)
+{
+	return inet_ctl_sock_create(&net->ipv4.tcp_sock,
+				    PF_INET, SOCK_RAW, IPPROTO_TCP, net);
+}
+
+static void __net_exit tcp_sk_exit(struct net *net)
+{
+	inet_ctl_sock_destroy(net->ipv4.tcp_sock);
+}
+
+static struct pernet_operations __net_initdata tcp_sk_ops = {
+       .init = tcp_sk_init,
+       .exit = tcp_sk_exit,
+};
+
 void __init tcp_v4_init(void)
 {
-	if (inet_ctl_sock_create(&tcp_sock, PF_INET, SOCK_RAW,
-				 IPPROTO_TCP, &init_net) < 0)
+	if (register_pernet_device(&tcp_sk_ops))
 		panic("Failed to create the TCP control socket.\n");
+	tcp_sock = init_net.ipv4.tcp_sock;
 }
 
 EXPORT_SYMBOL(ipv4_specific);
-- 
cgit v1.2.3


From 2e8046271f68198dd37451017c1a4a2432e4ec68 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 3 Apr 2008 09:22:09 +0900
Subject: [IPV4] MROUTE: Move PIM definitions to <linux/pim.h>.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/Kbuild   |  1 +
 include/linux/mroute.h | 22 +---------------------
 include/linux/pim.h    | 29 +++++++++++++++++++++++++++++
 3 files changed, 31 insertions(+), 21 deletions(-)
 create mode 100644 include/linux/pim.h

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 9cdd12a9e843..84736acb4b99 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -289,6 +289,7 @@ unifdef-y += parport.h
 unifdef-y += patchkey.h
 unifdef-y += pci.h
 unifdef-y += personality.h
+unifdef-y += pim.h
 unifdef-y += pktcdvd.h
 unifdef-y += pmu.h
 unifdef-y += poll.h
diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index 35a8277ec1bd..c41b4217ae3b 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -3,6 +3,7 @@
 
 #include <linux/sockios.h>
 #include <linux/in.h>
+#include <linux/pim.h>
 
 /*
  *	Based on the MROUTING 3.5 defines primarily to keep
@@ -210,27 +211,6 @@ struct mfc_cache
 #define IGMPMSG_WHOLEPKT	3		/* For PIM Register processing */
 
 #ifdef __KERNEL__
-
-#define PIM_V1_VERSION		__constant_htonl(0x10000000)
-#define PIM_V1_REGISTER		1
-
-#define PIM_VERSION		2
-#define PIM_REGISTER		1
-
-#define PIM_NULL_REGISTER	__constant_htonl(0x40000000)
-
-/* PIMv2 register message header layout (ietf-draft-idmr-pimvsm-v2-00.ps */
-
-struct pimreghdr
-{
-	__u8	type;
-	__u8	reserved;
-	__be16	csum;
-	__be32	flags;
-};
-
-extern int pim_rcv_v1(struct sk_buff *);
-
 struct rtmsg;
 extern int ipmr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait);
 #endif
diff --git a/include/linux/pim.h b/include/linux/pim.h
new file mode 100644
index 000000000000..6f689dc85503
--- /dev/null
+++ b/include/linux/pim.h
@@ -0,0 +1,29 @@
+#ifndef __LINUX_PIM_H
+#define __LINUX_PIM_H
+
+#include <asm/byteorder.h>
+
+/* Message types - V1 */
+#define PIM_V1_VERSION		__constant_htonl(0x10000000)
+#define PIM_V1_REGISTER		1
+
+/* Message types - V2 */
+#define PIM_VERSION		2
+#define PIM_REGISTER		1
+
+#if defined(__KERNEL__)
+#define PIM_NULL_REGISTER	__constant_htonl(0x40000000)
+
+/* PIMv2 register message header layout (ietf-draft-idmr-pimvsm-v2-00.ps */
+struct pimreghdr
+{
+	__u8	type;
+	__u8	reserved;
+	__be16	csum;
+	__be32	flags;
+};
+
+struct sk_buff;
+extern int pim_rcv_v1(struct sk_buff *);
+#endif
+#endif
-- 
cgit v1.2.3


From 80a9492a33dd7d852465625022d56ff76d62174d Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 3 Apr 2008 09:22:52 +0900
Subject: [IPV4] MROUTE: Adjust include files for user-space.

<linux/mroute.h> needs <linux/types.h>.
Avoid including <linux/in.h> in user-space, which conflicts with
standard <netinet/in.h>.
Add basic struct and constant in <linux/pim.h>.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/mroute.h |  3 +++
 include/linux/pim.h    | 16 ++++++++++++++++
 2 files changed, 19 insertions(+)

(limited to 'include')

diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index c41b4217ae3b..de4decfa1bfc 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -2,7 +2,10 @@
 #define __LINUX_MROUTE_H
 
 #include <linux/sockios.h>
+#include <linux/types.h>
+#ifdef __KERNEL__
 #include <linux/in.h>
+#endif
 #include <linux/pim.h>
 
 /*
diff --git a/include/linux/pim.h b/include/linux/pim.h
index 6f689dc85503..236ffd317394 100644
--- a/include/linux/pim.h
+++ b/include/linux/pim.h
@@ -3,6 +3,22 @@
 
 #include <asm/byteorder.h>
 
+#ifndef __KERNEL__
+struct pim {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u8	pim_type:4,		/* PIM message type */
+		pim_ver:4;		/* PIM version */
+#elif defined(__BIG_ENDIAN_BITFIELD)
+	__u8	pim_ver:4;		/* PIM version */
+		pim_type:4;		/* PIM message type */
+#endif
+	__u8	pim_rsv;		/* Reserved */
+	__be16	pim_cksum;		/* Checksum */
+};
+
+#define PIM_MINLEN		8
+#endif
+
 /* Message types - V1 */
 #define PIM_V1_VERSION		__constant_htonl(0x10000000)
 #define PIM_V1_REGISTER		1
-- 
cgit v1.2.3


From 2046b94e7c4fce92eb8165c2c36c6478f4927178 Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu@intel.com>
Date: Fri, 4 Apr 2008 11:05:59 -0700
Subject: [IA64] Multiple outstanding ptc.g instruction support

According to SDM2.2, Itanium supports multiple outstanding ptc.g instructions.
But current kernel function ia64_global_tlb_purge() uses a spinlock to serialize
ptc.g instructions issued by multiple processors. This serialization might have
scalability issue on a big SMP machine where many processors could purge TLB
in parallel.

The patch fixes this problem by issuing multiple ptc.g instructions in
ia64_global_tlb_purge(). It also adds support for the "PALO" table to get
a platform view of the max number of outstanding ptc.g instructions (which
may be different from the processor view found from PAL_VM_SUMMARY).

PALO specification can be found at: http://www.dig64.org/home/DIG64_PALO_R1_0.pdf

spinaphore implementation by Matthew Wilcox.

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/efi.c      |  46 ++++++++++++++++
 arch/ia64/kernel/setup.c    |   6 ++-
 arch/ia64/mm/tlb.c          | 125 ++++++++++++++++++++++++++++++++++++++------
 include/asm-ia64/sal.h      |  17 ++++++
 include/asm-ia64/tlbflush.h |   1 +
 5 files changed, 178 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c
index 728d7247a1a6..003cd09b0732 100644
--- a/arch/ia64/kernel/efi.c
+++ b/arch/ia64/kernel/efi.c
@@ -37,6 +37,7 @@
 #include <asm/pgtable.h>
 #include <asm/processor.h>
 #include <asm/mca.h>
+#include <asm/tlbflush.h>
 
 #define EFI_DEBUG	0
 
@@ -403,6 +404,41 @@ efi_get_pal_addr (void)
 	return NULL;
 }
 
+
+static u8 __init palo_checksum(u8 *buffer, u32 length)
+{
+	u8 sum = 0;
+	u8 *end = buffer + length;
+
+	while (buffer < end)
+		sum = (u8) (sum + *(buffer++));
+
+	return sum;
+}
+
+/*
+ * Parse and handle PALO table which is published at:
+ * http://www.dig64.org/home/DIG64_PALO_R1_0.pdf
+ */
+static void __init handle_palo(unsigned long palo_phys)
+{
+	struct palo_table *palo = __va(palo_phys);
+	u8  checksum;
+
+	if (strncmp(palo->signature, PALO_SIG, sizeof(PALO_SIG) - 1)) {
+		printk(KERN_INFO "PALO signature incorrect.\n");
+		return;
+	}
+
+	checksum = palo_checksum((u8 *)palo, palo->length);
+	if (checksum) {
+		printk(KERN_INFO "PALO checksum incorrect.\n");
+		return;
+	}
+
+	setup_ptcg_sem(palo->max_tlb_purges, 1);
+}
+
 void
 efi_map_pal_code (void)
 {
@@ -432,6 +468,7 @@ efi_init (void)
 	u64 efi_desc_size;
 	char *cp, vendor[100] = "unknown";
 	int i;
+	unsigned long palo_phys;
 
 	/*
 	 * It's too early to be able to use the standard kernel command line
@@ -496,6 +533,8 @@ efi_init (void)
 	efi.hcdp       = EFI_INVALID_TABLE_ADDR;
 	efi.uga        = EFI_INVALID_TABLE_ADDR;
 
+	palo_phys      = EFI_INVALID_TABLE_ADDR;
+
 	for (i = 0; i < (int) efi.systab->nr_tables; i++) {
 		if (efi_guidcmp(config_tables[i].guid, MPS_TABLE_GUID) == 0) {
 			efi.mps = config_tables[i].table;
@@ -515,10 +554,17 @@ efi_init (void)
 		} else if (efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID) == 0) {
 			efi.hcdp = config_tables[i].table;
 			printk(" HCDP=0x%lx", config_tables[i].table);
+		} else if (efi_guidcmp(config_tables[i].guid,
+			 PROCESSOR_ABSTRACTION_LAYER_OVERWRITE_GUID) == 0) {
+			palo_phys = config_tables[i].table;
+			printk(" PALO=0x%lx", config_tables[i].table);
 		}
 	}
 	printk("\n");
 
+	if (palo_phys != EFI_INVALID_TABLE_ADDR)
+		handle_palo(palo_phys);
+
 	runtime = __va(efi.systab->runtime);
 	efi.get_time = phys_get_time;
 	efi.set_time = phys_set_time;
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 4aa9eaea76c3..1cbd26340d87 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -59,6 +59,7 @@
 #include <asm/setup.h>
 #include <asm/smp.h>
 #include <asm/system.h>
+#include <asm/tlbflush.h>
 #include <asm/unistd.h>
 #include <asm/hpsim.h>
 
@@ -946,9 +947,10 @@ cpu_init (void)
 #endif
 
 	/* set ia64_ctx.max_rid to the maximum RID that is supported by all CPUs: */
-	if (ia64_pal_vm_summary(NULL, &vmi) == 0)
+	if (ia64_pal_vm_summary(NULL, &vmi) == 0) {
 		max_ctx = (1U << (vmi.pal_vm_info_2_s.rid_size - 3)) - 1;
-	else {
+		setup_ptcg_sem(vmi.pal_vm_info_2_s.max_purges, 0);
+	} else {
 		printk(KERN_WARNING "cpu_init: PAL VM summary failed, assuming 18 RID bits\n");
 		max_ctx = (1U << 15) - 1;	/* use architected minimum */
 	}
diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c
index 655da240d13c..d41d6076ed03 100644
--- a/arch/ia64/mm/tlb.c
+++ b/arch/ia64/mm/tlb.c
@@ -11,6 +11,9 @@
  * Rohit Seth <rohit.seth@intel.com>
  * Ken Chen <kenneth.w.chen@intel.com>
  * Christophe de Dinechin <ddd@hp.com>: Avoid ptc.e on memory allocation
+ * Copyright (C) 2007 Intel Corp
+ *	Fenghua Yu <fenghua.yu@intel.com>
+ *	Add multiple ptc.g/ptc.ga instruction support in global tlb purge.
  */
 #include <linux/module.h>
 #include <linux/init.h>
@@ -26,6 +29,7 @@
 #include <asm/pal.h>
 #include <asm/tlbflush.h>
 #include <asm/dma.h>
+#include <asm/sal.h>
 
 static struct {
 	unsigned long mask;	/* mask of supported purge page-sizes */
@@ -84,14 +88,104 @@ wrap_mmu_context (struct mm_struct *mm)
 	local_flush_tlb_all();
 }
 
+/*
+ * Implement "spinaphores" ... like counting semaphores, but they
+ * spin instead of sleeping.  If there are ever any other users for
+ * this primitive it can be moved up to a spinaphore.h header.
+ */
+struct spinaphore {
+	atomic_t	cur;
+};
+
+static inline void spinaphore_init(struct spinaphore *ss, int val)
+{
+	atomic_set(&ss->cur, val);
+}
+
+static inline void down_spin(struct spinaphore *ss)
+{
+	while (unlikely(!atomic_add_unless(&ss->cur, -1, 0)))
+		while (atomic_read(&ss->cur) == 0)
+			cpu_relax();
+}
+
+static inline void up_spin(struct spinaphore *ss)
+{
+	atomic_add(1, &ss->cur);
+}
+
+static struct spinaphore ptcg_sem;
+static u16 nptcg = 1;
+static int need_ptcg_sem = 1;
+static int toolatetochangeptcgsem = 0;
+
+/*
+ * Maximum number of simultaneous ptc.g purges in the system can
+ * be defined by PAL_VM_SUMMARY (in which case we should take
+ * the smallest value for any cpu in the system) or by the PAL
+ * override table (in which case we should ignore the value from
+ * PAL_VM_SUMMARY).
+ *
+ * Complicating the logic here is the fact that num_possible_cpus()
+ * isn't fully setup until we start bringing cpus online.
+ */
+void
+setup_ptcg_sem(int max_purges, int from_palo)
+{
+	static int have_palo;
+	static int firstcpu = 1;
+
+	if (toolatetochangeptcgsem) {
+		BUG_ON(max_purges < nptcg);
+		return;
+	}
+
+	if (from_palo) {
+		have_palo = 1;
+
+		/* In PALO max_purges == 0 really means it! */
+		if (max_purges == 0)
+			panic("Whoa! Platform does not support global TLB purges.\n");
+		nptcg = max_purges;
+		if (nptcg == PALO_MAX_TLB_PURGES) {
+			need_ptcg_sem = 0;
+			return;
+		}
+		goto resetsema;
+	}
+	if (have_palo) {
+		if (nptcg != PALO_MAX_TLB_PURGES)
+			need_ptcg_sem = (num_possible_cpus() > nptcg);
+		return;
+	}
+
+	/* In PAL_VM_SUMMARY max_purges == 0 actually means 1 */
+	if (max_purges == 0) max_purges = 1;
+
+	if (firstcpu) {
+		nptcg = max_purges;
+		firstcpu = 0;
+	}
+	if (max_purges < nptcg)
+		nptcg = max_purges;
+	if (nptcg == PAL_MAX_PURGES) {
+		need_ptcg_sem = 0;
+		return;
+	} else
+		need_ptcg_sem = (num_possible_cpus() > nptcg);
+
+resetsema:
+	spinaphore_init(&ptcg_sem, max_purges);
+}
+
 void
 ia64_global_tlb_purge (struct mm_struct *mm, unsigned long start,
 		       unsigned long end, unsigned long nbits)
 {
-	static DEFINE_SPINLOCK(ptcg_lock);
-
 	struct mm_struct *active_mm = current->active_mm;
 
+	toolatetochangeptcgsem = 1;
+
 	if (mm != active_mm) {
 		/* Restore region IDs for mm */
 		if (mm && active_mm) {
@@ -102,19 +196,20 @@ ia64_global_tlb_purge (struct mm_struct *mm, unsigned long start,
 		}
 	}
 
-	/* HW requires global serialization of ptc.ga.  */
-	spin_lock(&ptcg_lock);
-	{
-		do {
-			/*
-			 * Flush ALAT entries also.
-			 */
-			ia64_ptcga(start, (nbits<<2));
-			ia64_srlz_i();
-			start += (1UL << nbits);
-		} while (start < end);
-	}
-	spin_unlock(&ptcg_lock);
+	if (need_ptcg_sem)
+		down_spin(&ptcg_sem);
+
+	do {
+		/*
+		 * Flush ALAT entries also.
+		 */
+		ia64_ptcga(start, (nbits << 2));
+		ia64_srlz_i();
+		start += (1UL << nbits);
+	} while (start < end);
+
+	if (need_ptcg_sem)
+		up_spin(&ptcg_sem);
 
         if (mm != active_mm) {
                 activate_context(active_mm);
diff --git a/include/asm-ia64/sal.h b/include/asm-ia64/sal.h
index f4904db3b057..3cd637a2c051 100644
--- a/include/asm-ia64/sal.h
+++ b/include/asm-ia64/sal.h
@@ -296,6 +296,9 @@ enum {
     EFI_GUID(0xe429faf8, 0x3cb7, 0x11d4, 0xbc, 0xa7, 0x0, 0x80, 0xc7, 0x3c, 0x88, 0x81)
 #define SAL_PLAT_BUS_ERR_SECT_GUID  \
     EFI_GUID(0xe429faf9, 0x3cb7, 0x11d4, 0xbc, 0xa7, 0x0, 0x80, 0xc7, 0x3c, 0x88, 0x81)
+#define PROCESSOR_ABSTRACTION_LAYER_OVERWRITE_GUID \
+    EFI_GUID(0x6cb0a200, 0x893a, 0x11da, 0x96, 0xd2, 0x0, 0x10, 0x83, 0xff, \
+		0xca, 0x4d)
 
 #define MAX_CACHE_ERRORS	6
 #define MAX_TLB_ERRORS		6
@@ -879,6 +882,20 @@ extern void ia64_jump_to_sal(struct sal_to_os_boot *);
 
 extern void ia64_sal_handler_init(void *entry_point, void *gpval);
 
+#define PALO_MAX_TLB_PURGES	0xFFFF
+#define PALO_SIG	"PALO"
+
+struct palo_table {
+	u8  signature[4];	/* Should be "PALO" */
+	u32 length;
+	u8  minor_revision;
+	u8  major_revision;
+	u8  checksum;
+	u8  reserved1[5];
+	u16 max_tlb_purges;
+	u8  reserved2[6];
+};
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* _ASM_IA64_SAL_H */
diff --git a/include/asm-ia64/tlbflush.h b/include/asm-ia64/tlbflush.h
index 7774a1cac0cc..3be25dfed164 100644
--- a/include/asm-ia64/tlbflush.h
+++ b/include/asm-ia64/tlbflush.h
@@ -17,6 +17,7 @@
  * Now for some TLB flushing routines.  This is the kind of stuff that
  * can be very expensive, so try to avoid them whenever possible.
  */
+extern void setup_ptcg_sem(int max_purges, int from_palo);
 
 /*
  * Flush everything (kernel mapping may also have changed due to
-- 
cgit v1.2.3


From a6c75b86ce9f01db4ea9912877b526c2dc4d2f0a Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu@intel.com>
Date: Fri, 14 Mar 2008 13:57:08 -0700
Subject: [IA64] Kernel parameter for max number of concurrent global TLB
 purges

The patch defines kernel parameter "nptcg=". The parameter overrides max number
of concurrent global TLB purges which is reported from either PAL_VM_SUMMARY or
SAL PALO.

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 Documentation/kernel-parameters.txt |  4 ++++
 arch/ia64/kernel/efi.c              |  2 +-
 arch/ia64/kernel/setup.c            |  2 +-
 arch/ia64/mm/tlb.c                  | 46 +++++++++++++++++++++++++++++++++----
 include/asm-ia64/sal.h              |  4 ++++
 5 files changed, 51 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 4cd1a5da80a4..be92e6165722 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1335,6 +1335,10 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	nowb		[ARM]
 
+	nptcg=		[IA64] Override max number of concurrent global TLB
+			purges which is reported from either PAL_VM_SUMMARY or
+			SAL PALO.
+
 	numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
 			one of ['zone', 'node', 'default'] can be specified
 			This can be set from sysctl after boot.
diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c
index 003cd09b0732..d45f215bc8fc 100644
--- a/arch/ia64/kernel/efi.c
+++ b/arch/ia64/kernel/efi.c
@@ -436,7 +436,7 @@ static void __init handle_palo(unsigned long palo_phys)
 		return;
 	}
 
-	setup_ptcg_sem(palo->max_tlb_purges, 1);
+	setup_ptcg_sem(palo->max_tlb_purges, NPTCG_FROM_PALO);
 }
 
 void
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 1cbd26340d87..f798c0769d36 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -949,7 +949,7 @@ cpu_init (void)
 	/* set ia64_ctx.max_rid to the maximum RID that is supported by all CPUs: */
 	if (ia64_pal_vm_summary(NULL, &vmi) == 0) {
 		max_ctx = (1U << (vmi.pal_vm_info_2_s.rid_size - 3)) - 1;
-		setup_ptcg_sem(vmi.pal_vm_info_2_s.max_purges, 0);
+		setup_ptcg_sem(vmi.pal_vm_info_2_s.max_purges, NPTCG_FROM_PAL);
 	} else {
 		printk(KERN_WARNING "cpu_init: PAL VM summary failed, assuming 18 RID bits\n");
 		max_ctx = (1U << 15) - 1;	/* use architected minimum */
diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c
index d41d6076ed03..1a8948fd0029 100644
--- a/arch/ia64/mm/tlb.c
+++ b/arch/ia64/mm/tlb.c
@@ -119,6 +119,27 @@ static u16 nptcg = 1;
 static int need_ptcg_sem = 1;
 static int toolatetochangeptcgsem = 0;
 
+/*
+ * Kernel parameter "nptcg=" overrides max number of concurrent global TLB
+ * purges which is reported from either PAL or SAL PALO.
+ *
+ * We don't have sanity checking for nptcg value. It's the user's responsibility
+ * for valid nptcg value on the platform. Otherwise, kernel may hang in some
+ * cases.
+ */
+static int __init
+set_nptcg(char *str)
+{
+	int value = 0;
+
+	get_option(&str, &value);
+	setup_ptcg_sem(value, NPTCG_FROM_KERNEL_PARAMETER);
+
+	return 1;
+}
+
+__setup("nptcg=", set_nptcg);
+
 /*
  * Maximum number of simultaneous ptc.g purges in the system can
  * be defined by PAL_VM_SUMMARY (in which case we should take
@@ -126,13 +147,18 @@ static int toolatetochangeptcgsem = 0;
  * override table (in which case we should ignore the value from
  * PAL_VM_SUMMARY).
  *
+ * Kernel parameter "nptcg=" overrides maximum number of simultanesous ptc.g
+ * purges defined in either PAL_VM_SUMMARY or PAL override table. In this case,
+ * we should ignore the value from either PAL_VM_SUMMARY or PAL override table.
+ *
  * Complicating the logic here is the fact that num_possible_cpus()
  * isn't fully setup until we start bringing cpus online.
  */
 void
-setup_ptcg_sem(int max_purges, int from_palo)
+setup_ptcg_sem(int max_purges, int nptcg_from)
 {
-	static int have_palo;
+	static int kp_override;
+	static int palo_override;
 	static int firstcpu = 1;
 
 	if (toolatetochangeptcgsem) {
@@ -140,8 +166,18 @@ setup_ptcg_sem(int max_purges, int from_palo)
 		return;
 	}
 
-	if (from_palo) {
-		have_palo = 1;
+	if (nptcg_from == NPTCG_FROM_KERNEL_PARAMETER) {
+		kp_override = 1;
+		nptcg = max_purges;
+		goto resetsema;
+	}
+	if (kp_override) {
+		need_ptcg_sem = num_possible_cpus() > nptcg;
+		return;
+	}
+
+	if (nptcg_from == NPTCG_FROM_PALO) {
+		palo_override = 1;
 
 		/* In PALO max_purges == 0 really means it! */
 		if (max_purges == 0)
@@ -153,7 +189,7 @@ setup_ptcg_sem(int max_purges, int from_palo)
 		}
 		goto resetsema;
 	}
-	if (have_palo) {
+	if (palo_override) {
 		if (nptcg != PALO_MAX_TLB_PURGES)
 			need_ptcg_sem = (num_possible_cpus() > nptcg);
 		return;
diff --git a/include/asm-ia64/sal.h b/include/asm-ia64/sal.h
index 3cd637a2c051..89594b442f83 100644
--- a/include/asm-ia64/sal.h
+++ b/include/asm-ia64/sal.h
@@ -896,6 +896,10 @@ struct palo_table {
 	u8  reserved2[6];
 };
 
+#define NPTCG_FROM_PAL			0
+#define NPTCG_FROM_PALO			1
+#define NPTCG_FROM_KERNEL_PARAMETER	2
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* _ASM_IA64_SAL_H */
-- 
cgit v1.2.3


From 7bc570c8b4f75ddb3fd5dbeb38127cdc4acbcc9c Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 3 Apr 2008 09:22:53 +0900
Subject: [IPV6] MROUTE: Support multicast forwarding.

Based on ancient patch by Mickael Hoerdt
<hoerdt@clarinet.u-strasbg.fr>, which is available at
<http://www-r2.u-strasbg.fr/~hoerdt/dev/linux_ipv6_mforwarding/patch-linux-ipv6-mforwarding-0.1a>.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/Kbuild     |    1 +
 include/linux/ipv6.h     |    5 +
 include/linux/mroute6.h  |  227 ++++++++
 net/ipv6/Kconfig         |    7 +
 net/ipv6/Makefile        |    2 +
 net/ipv6/addrconf.c      |   15 +-
 net/ipv6/af_inet6.c      |    6 +
 net/ipv6/ip6_input.c     |   87 ++-
 net/ipv6/ip6_output.c    |    6 +-
 net/ipv6/ip6mr.c         | 1384 ++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/ipv6_sockglue.c |    7 +
 net/ipv6/raw.c           |    7 +-
 net/ipv6/route.c         |   30 +-
 13 files changed, 1754 insertions(+), 30 deletions(-)
 create mode 100644 include/linux/mroute6.h
 create mode 100644 net/ipv6/ip6mr.c

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 84736acb4b99..29ab9b95d376 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -261,6 +261,7 @@ unifdef-y += mempolicy.h
 unifdef-y += mii.h
 unifdef-y += mman.h
 unifdef-y += mroute.h
+unifdef-y += mroute6.h
 unifdef-y += msdos_fs.h
 unifdef-y += msg.h
 unifdef-y += nbd.h
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index b90d3d461d4e..f53e4764fc05 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -159,6 +159,9 @@ struct ipv6_devconf {
 	__s32		accept_source_route;
 #ifdef CONFIG_IPV6_OPTIMISTIC_DAD
 	__s32		optimistic_dad;
+#endif
+#ifdef CONFIG_IPV6_MROUTE
+	__s32		mc_forwarding;
 #endif
 	void		*sysctl;
 };
@@ -190,6 +193,7 @@ enum {
 	DEVCONF_PROXY_NDP,
 	DEVCONF_OPTIMISTIC_DAD,
 	DEVCONF_ACCEPT_SOURCE_ROUTE,
+	DEVCONF_MC_FORWARDING,
 	DEVCONF_MAX
 };
 
@@ -230,6 +234,7 @@ struct inet6_skb_parm {
 #endif
 
 #define IP6SKB_XFRM_TRANSFORMED	1
+#define IP6SKB_FORWARDED	2
 };
 
 #define IP6CB(skb)	((struct inet6_skb_parm*)((skb)->cb))
diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
new file mode 100644
index 000000000000..b92190304e0b
--- /dev/null
+++ b/include/linux/mroute6.h
@@ -0,0 +1,227 @@
+#ifndef __LINUX_MROUTE6_H
+#define __LINUX_MROUTE6_H
+
+#include <linux/types.h>
+#include <linux/sockios.h>
+
+/*
+ *	Based on the MROUTING 3.5 defines primarily to keep
+ *	source compatibility with BSD.
+ *
+ *	See the pim6sd code for the original history.
+ *
+ *      Protocol Independent Multicast (PIM) data structures included
+ *      Carlos Picoto (cap@di.fc.ul.pt)
+ *
+ */
+
+#define MRT6_BASE	200
+#define MRT6_INIT	(MRT6_BASE)	/* Activate the kernel mroute code 	*/
+#define MRT6_DONE	(MRT6_BASE+1)	/* Shutdown the kernel mroute		*/
+#define MRT6_ADD_MIF	(MRT6_BASE+2)	/* Add a virtual interface		*/
+#define MRT6_DEL_MIF	(MRT6_BASE+3)	/* Delete a virtual interface		*/
+#define MRT6_ADD_MFC	(MRT6_BASE+4)	/* Add a multicast forwarding entry	*/
+#define MRT6_DEL_MFC	(MRT6_BASE+5)	/* Delete a multicast forwarding entry	*/
+#define MRT6_VERSION	(MRT6_BASE+6)	/* Get the kernel multicast version	*/
+
+#define SIOCGETMIFCNT_IN6	SIOCPROTOPRIVATE	/* IP protocol privates */
+#define SIOCGETSGCNT_IN6	(SIOCPROTOPRIVATE+1)
+#define SIOCGETRPF	(SIOCPROTOPRIVATE+2)
+
+#define MAXMIFS		32
+typedef unsigned long mifbitmap_t;	/* User mode code depends on this lot */
+typedef unsigned short mifi_t;
+#define ALL_MIFS	((mifi_t)(-1))
+
+#ifndef IF_SETSIZE
+#define IF_SETSIZE	256
+#endif
+
+typedef	__u32		if_mask;
+#define NIFBITS (sizeof(if_mask) * 8)        /* bits per mask */
+
+#if !defined(__KERNEL__) && !defined(DIV_ROUND_UP)
+#define	DIV_ROUND_UP(x,y)	(((x) + ((y) - 1)) / (y))
+#endif
+
+typedef struct if_set {
+	if_mask ifs_bits[DIV_ROUND_UP(IF_SETSIZE, NIFBITS)];
+} if_set;
+
+#define IF_SET(n, p)    ((p)->ifs_bits[(n)/NIFBITS] |= (1 << ((n) % NIFBITS)))
+#define IF_CLR(n, p)    ((p)->ifs_bits[(n)/NIFBITS] &= ~(1 << ((n) % NIFBITS)))
+#define IF_ISSET(n, p)  ((p)->ifs_bits[(n)/NIFBITS] & (1 << ((n) % NIFBITS)))
+#define IF_COPY(f, t)   bcopy(f, t, sizeof(*(f)))
+#define IF_ZERO(p)      bzero(p, sizeof(*(p)))
+
+/*
+ *	Passed by mrouted for an MRT_ADD_MIF - again we use the
+ *	mrouted 3.6 structures for compatibility
+ */
+
+struct mif6ctl {
+	mifi_t	mif6c_mifi;		/* Index of MIF */
+	unsigned char mif6c_flags;	/* MIFF_ flags */
+	unsigned char vifc_threshold;	/* ttl limit */
+	u_short	 mif6c_pifi;		/* the index of the physical IF */
+	unsigned int vifc_rate_limit;	/* Rate limiter values (NI) */
+};
+
+#define MIFF_REGISTER	0x1	/* register vif	*/
+
+/*
+ *	Cache manipulation structures for mrouted and PIMd
+ */
+
+struct mf6cctl
+{
+	struct sockaddr_in6 mf6cc_origin;		/* Origin of mcast	*/
+	struct sockaddr_in6 mf6cc_mcastgrp;		/* Group in question	*/
+	mifi_t	mf6cc_parent;			/* Where it arrived	*/
+	struct if_set mf6cc_ifset;		/* Where it is going */
+};
+
+/*
+ *	Group count retrieval for pim6sd
+ */
+
+struct sioc_sg_req6
+{
+	struct sockaddr_in6 src;
+	struct sockaddr_in6 grp;
+	unsigned long pktcnt;
+	unsigned long bytecnt;
+	unsigned long wrong_if;
+};
+
+/*
+ *	To get vif packet counts
+ */
+
+struct sioc_mif_req6
+{
+	mifi_t	mifi;		/* Which iface */
+	unsigned long icount;	/* In packets */
+	unsigned long ocount;	/* Out packets */
+	unsigned long ibytes;	/* In bytes */
+	unsigned long obytes;	/* Out bytes */
+};
+
+/*
+ *	That's all usermode folks
+ */
+
+#ifdef __KERNEL__
+
+#include <linux/skbuff.h>	/* for struct sk_buff_head */
+
+struct net_device;
+struct inet6_dev *ipv6_find_idev(struct net_device *dev);
+
+#ifdef CONFIG_IPV6_MROUTE
+static inline int ip6_mroute_opt(int opt)
+{
+	return (opt >= MRT6_BASE) && (opt <= MRT6_BASE + 10);
+}
+#else
+static inline int ip6_mroute_opt(int opt)
+{
+	return 0;
+}
+#endif
+
+struct sock;
+
+extern int ip6_mroute_setsockopt(struct sock *, int, char __user *, int);
+extern int ip6_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
+extern int ip6_mr_input(struct sk_buff *skb);
+extern int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg);
+extern void ip6_mr_init(void);
+
+struct mif_device
+{
+	struct net_device 	*dev;			/* Device we are using */
+	unsigned long	bytes_in,bytes_out;
+	unsigned long	pkt_in,pkt_out;		/* Statistics 			*/
+	unsigned long	rate_limit;		/* Traffic shaping (NI) 	*/
+	unsigned char	threshold;		/* TTL threshold 		*/
+	unsigned short	flags;			/* Control flags 		*/
+	int		link;			/* Physical interface index	*/
+};
+
+#define VIFF_STATIC 0x8000
+
+struct mfc6_cache
+{
+	struct mfc6_cache *next;		/* Next entry on cache line 	*/
+	struct in6_addr mf6c_mcastgrp;			/* Group the entry belongs to 	*/
+	struct in6_addr mf6c_origin;			/* Source of packet 		*/
+	mifi_t mf6c_parent;			/* Source interface		*/
+	int mfc_flags;				/* Flags on line		*/
+
+	union {
+		struct {
+			unsigned long expires;
+			struct sk_buff_head unresolved;	/* Unresolved buffers		*/
+		} unres;
+		struct {
+			unsigned long last_assert;
+			int minvif;
+			int maxvif;
+			unsigned long bytes;
+			unsigned long pkt;
+			unsigned long wrong_if;
+			unsigned char ttls[MAXMIFS];	/* TTL thresholds		*/
+		} res;
+	} mfc_un;
+};
+
+#define MFC_STATIC		1
+#define MFC_NOTIFY		2
+
+#define MFC6_LINES		64
+
+#define MFC6_HASH(a, g) (((__force u32)(a)->s6_addr32[0] ^ \
+			  (__force u32)(a)->s6_addr32[1] ^ \
+			  (__force u32)(a)->s6_addr32[2] ^ \
+			  (__force u32)(a)->s6_addr32[3] ^ \
+			  (__force u32)(g)->s6_addr32[0] ^ \
+			  (__force u32)(g)->s6_addr32[1] ^ \
+			  (__force u32)(g)->s6_addr32[2] ^ \
+			  (__force u32)(g)->s6_addr32[3]) % MFC6_LINES)
+
+#define MFC_ASSERT_THRESH (3*HZ)		/* Maximal freq. of asserts */
+
+#endif
+
+#ifdef __KERNEL__
+struct rtmsg;
+extern int ip6mr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait);
+
+#ifdef CONFIG_IPV6_MROUTE
+extern struct sock *mroute6_socket;
+extern int ip6mr_sk_done(struct sock *sk);
+#else
+#define mroute6_socket NULL
+static inline int ip6mr_sk_done(struct sock *sk) { return 0; }
+#endif
+#endif
+
+/*
+ * Structure used to communicate from kernel to multicast router.
+ * We'll overlay the structure onto an MLD header (not an IPv6 heder like igmpmsg{}
+ * used for IPv4 implementation). This is because this structure will be passed via an
+ * IPv6 raw socket, on wich an application will only receiver the payload i.e the data after
+ * the IPv6 header and all the extension headers. (See section 3 of RFC 3542)
+ */
+
+struct mrt6msg {
+#define MRT6MSG_NOCACHE		1
+	__u8		im6_mbz;		/* must be zero		   */
+	__u8		im6_msgtype;		/* what type of message    */
+	__u16		im6_mif;		/* mif rec'd on		   */
+	__u32		im6_pad;		/* padding for 64 bit arch */
+	struct in6_addr	im6_src, im6_dst;
+};
+
+#endif
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 47263e45bacb..9a2ea81e499f 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -209,3 +209,10 @@ config IPV6_SUBTREES
 
 	  If unsure, say N.
 
+config IPV6_MROUTE
+	bool "IPv6: multicast routing (EXPERIMENTAL)"
+	depends on IPV6 && EXPERIMENTAL
+	---help---
+	  Experimental support for IPv6 multicast forwarding.
+	  If unsure, say N.
+
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index ae14617e607f..686934acfac1 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -11,6 +11,8 @@ ipv6-objs :=	af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \
 		exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o
 
 ipv6-$(CONFIG_SYSCTL) = sysctl_net_ipv6.o
+ipv6-$(CONFIG_IPV6_MROUTE) += ip6mr.o
+
 ipv6-$(CONFIG_XFRM) += xfrm6_policy.o xfrm6_state.o xfrm6_input.o \
 	xfrm6_output.o
 ipv6-$(CONFIG_NETFILTER) += netfilter.o
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 11037615bc73..dbc51af69017 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -412,7 +412,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 	return ndev;
 }
 
-static struct inet6_dev * ipv6_find_idev(struct net_device *dev)
+struct inet6_dev * ipv6_find_idev(struct net_device *dev)
 {
 	struct inet6_dev *idev;
 
@@ -3547,6 +3547,9 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 #ifdef CONFIG_IPV6_OPTIMISTIC_DAD
 	array[DEVCONF_OPTIMISTIC_DAD] = cnf->optimistic_dad;
 #endif
+#ifdef CONFIG_IPV6_MROUTE
+	array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding;
+#endif
 }
 
 static inline size_t inet6_if_nlmsg_size(void)
@@ -4094,6 +4097,16 @@ static struct addrconf_sysctl_table
 			.proc_handler   =       &proc_dointvec,
 
 		},
+#endif
+#ifdef CONFIG_IPV6_MROUTE
+		{
+			.ctl_name	=	CTL_UNNUMBERED,
+			.procname	=	"mc_forwarding",
+			.data		=	&ipv6_devconf.mc_forwarding,
+			.maxlen		=	sizeof(int),
+			.mode		=	0644,
+			.proc_handler	=	&proc_dointvec,
+		},
 #endif
 		{
 			.ctl_name	=	0,	/* sentinel */
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 1731b0abf7f5..3c6aafb02183 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -61,6 +61,9 @@
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
+#ifdef CONFIG_IPV6_MROUTE
+#include <linux/mroute6.h>
+#endif
 
 MODULE_AUTHOR("Cast of dozens");
 MODULE_DESCRIPTION("IPv6 protocol stack for Linux");
@@ -953,6 +956,9 @@ static int __init inet6_init(void)
 	err = icmpv6_init();
 	if (err)
 		goto icmp_fail;
+#ifdef CONFIG_IPV6_MROUTE
+	ip6_mr_init();
+#endif
 	err = ndisc_init();
 	if (err)
 		goto ndisc_fail;
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 43a617e2268b..09a3201e408a 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -29,6 +29,7 @@
 #include <linux/netdevice.h>
 #include <linux/in6.h>
 #include <linux/icmpv6.h>
+#include <linux/mroute6.h>
 
 #include <linux/netfilter.h>
 #include <linux/netfilter_ipv6.h>
@@ -236,36 +237,84 @@ int ip6_mc_input(struct sk_buff *skb)
 	hdr = ipv6_hdr(skb);
 	deliver = ipv6_chk_mcast_addr(skb->dev, &hdr->daddr, NULL);
 
+#ifdef CONFIG_IPV6_MROUTE
 	/*
-	 *	IPv6 multicast router mode isnt currently supported.
+	 *      IPv6 multicast router mode is now supported ;)
 	 */
-#if 0
-	if (ipv6_config.multicast_route) {
-		int addr_type;
-
-		addr_type = ipv6_addr_type(&hdr->daddr);
-
-		if (!(addr_type & (IPV6_ADDR_LOOPBACK | IPV6_ADDR_LINKLOCAL))) {
-			struct sk_buff *skb2;
-			struct dst_entry *dst;
+	if (ipv6_devconf.mc_forwarding &&
+	    likely(!(IP6CB(skb)->flags & IP6SKB_FORWARDED))) {
+		/*
+		 * Okay, we try to forward - split and duplicate
+		 * packets.
+		 */
+		struct sk_buff *skb2;
+		struct inet6_skb_parm *opt = IP6CB(skb);
+
+		/* Check for MLD */
+		if (unlikely(opt->ra)) {
+			/* Check if this is a mld message */
+			u8 *ptr = skb_network_header(skb) + opt->ra;
+			struct icmp6hdr *icmp6;
+			u8 nexthdr = hdr->nexthdr;
+			int offset;
+
+			/* Check if the value of Router Alert
+			 * is for MLD (0x0000).
+			 */
+			if ((ptr[2] | ptr[3]) == 0) {
+				if (!ipv6_ext_hdr(nexthdr)) {
+					/* BUG */
+					goto discard;
+				}
+				offset = ipv6_skip_exthdr(skb, sizeof(*hdr),
+							  &nexthdr);
+				if (offset < 0)
+					goto discard;
+
+				if (nexthdr != IPPROTO_ICMPV6)
+					goto discard;
+
+				if (!pskb_may_pull(skb, (skb_network_header(skb) +
+						   offset + 1 - skb->data)))
+					goto discard;
+
+				icmp6 = (struct icmp6hdr *)(skb_network_header(skb) + offset);
+
+				switch (icmp6->icmp6_type) {
+				case ICMPV6_MGM_QUERY:
+				case ICMPV6_MGM_REPORT:
+				case ICMPV6_MGM_REDUCTION:
+				case ICMPV6_MLD2_REPORT:
+					break;
+				default:
+					/* Bogus */
+					goto discard;
+				}
+				deliver = 1;
+				goto out;
+			}
+			/* unknown RA - process it normally */
+		}
 
-			dst = skb->dst;
+		if (deliver)
+			skb2 = skb_clone(skb, GFP_ATOMIC);
+		else {
+			skb2 = skb;
+			skb = NULL;
+		}
 
-			if (deliver) {
-				skb2 = skb_clone(skb, GFP_ATOMIC);
-				dst_output(skb2);
-			} else {
-				dst_output(skb);
-				return 0;
-			}
+		if (skb2) {
+			skb2->dev = skb2->dst->dev;
+			ip6_mr_input(skb2);
 		}
 	}
 #endif
-
+out:
 	if (likely(deliver)) {
 		ip6_input(skb);
 		return 0;
 	}
+discard:
 	/* discard */
 	kfree_skb(skb);
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index a8b4da25b0a7..c0dbe549cc42 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -55,6 +55,7 @@
 #include <net/icmp.h>
 #include <net/xfrm.h>
 #include <net/checksum.h>
+#include <linux/mroute6.h>
 
 static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *));
 
@@ -137,8 +138,9 @@ static int ip6_output2(struct sk_buff *skb)
 		struct inet6_dev *idev = ip6_dst_idev(skb->dst);
 
 		if (!(dev->flags & IFF_LOOPBACK) && (!np || np->mc_loop) &&
-		    ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr,
-					&ipv6_hdr(skb)->saddr)) {
+		    ((mroute6_socket && !(IP6CB(skb)->flags & IP6SKB_FORWARDED)) ||
+		     ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr,
+					 &ipv6_hdr(skb)->saddr))) {
 			struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
 
 			/* Do not check for IFF_ALLMULTI; multicast routing
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
new file mode 100644
index 000000000000..1bdf3c177d58
--- /dev/null
+++ b/net/ipv6/ip6mr.c
@@ -0,0 +1,1384 @@
+/*
+ *	Linux IPv6 multicast routing support for BSD pim6sd
+ *	Based on net/ipv4/ipmr.c.
+ *
+ *	(c) 2004 Mickael Hoerdt, <hoerdt@clarinet.u-strasbg.fr>
+ *		LSIIT Laboratory, Strasbourg, France
+ *	(c) 2004 Jean-Philippe Andriot, <jean-philippe.andriot@6WIND.com>
+ *		6WIND, Paris, France
+ *	Copyright (C)2007,2008 USAGI/WIDE Project
+ *		YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ */
+
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/timer.h>
+#include <linux/mm.h>
+#include <linux/kernel.h>
+#include <linux/fcntl.h>
+#include <linux/stat.h>
+#include <linux/socket.h>
+#include <linux/in.h>
+#include <linux/inet.h>
+#include <linux/netdevice.h>
+#include <linux/inetdevice.h>
+#include <linux/igmp.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/mroute.h>
+#include <linux/init.h>
+#include <net/ip.h>
+#include <net/protocol.h>
+#include <linux/skbuff.h>
+#include <net/sock.h>
+#include <net/icmp.h>
+#include <net/udp.h>
+#include <net/raw.h>
+#include <net/route.h>
+#include <linux/notifier.h>
+#include <linux/if_arp.h>
+#include <linux/netfilter_ipv4.h>
+#include <net/ipip.h>
+#include <net/checksum.h>
+#include <net/netlink.h>
+
+#include <net/ipv6.h>
+#include <net/ip6_route.h>
+#include <linux/mroute6.h>
+#include <net/addrconf.h>
+#include <linux/netfilter_ipv6.h>
+
+struct sock *mroute6_socket;
+
+
+/* Big lock, protecting vif table, mrt cache and mroute socket state.
+   Note that the changes are semaphored via rtnl_lock.
+ */
+
+static DEFINE_RWLOCK(mrt_lock);
+
+/*
+ *	Multicast router control variables
+ */
+
+static struct mif_device vif6_table[MAXMIFS];		/* Devices 		*/
+static int maxvif;
+
+#define MIF_EXISTS(idx) (vif6_table[idx].dev != NULL)
+
+static struct mfc6_cache *mfc6_cache_array[MFC_LINES];	/* Forwarding cache	*/
+
+static struct mfc6_cache *mfc_unres_queue;		/* Queue of unresolved entries */
+static atomic_t cache_resolve_queue_len;		/* Size of unresolved	*/
+
+/* Special spinlock for queue of unresolved entries */
+static DEFINE_SPINLOCK(mfc_unres_lock);
+
+/* We return to original Alan's scheme. Hash table of resolved
+   entries is changed only in process context and protected
+   with weak lock mrt_lock. Queue of unresolved entries is protected
+   with strong spinlock mfc_unres_lock.
+
+   In this case data path is free of exclusive locks at all.
+ */
+
+static struct kmem_cache *mrt_cachep __read_mostly;
+
+static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache);
+static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert);
+static int ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm);
+
+static struct timer_list ipmr_expire_timer;
+
+
+#ifdef CONFIG_PROC_FS
+
+struct ipmr_mfc_iter {
+	struct mfc6_cache **cache;
+	int ct;
+};
+
+
+static struct mfc6_cache *ipmr_mfc_seq_idx(struct ipmr_mfc_iter *it, loff_t pos)
+{
+	struct mfc6_cache *mfc;
+
+	it->cache = mfc6_cache_array;
+	read_lock(&mrt_lock);
+	for (it->ct = 0; it->ct < ARRAY_SIZE(mfc6_cache_array); it->ct++)
+		for (mfc = mfc6_cache_array[it->ct]; mfc; mfc = mfc->next)
+			if (pos-- == 0)
+				return mfc;
+	read_unlock(&mrt_lock);
+
+	it->cache = &mfc_unres_queue;
+	spin_lock_bh(&mfc_unres_lock);
+	for (mfc = mfc_unres_queue; mfc; mfc = mfc->next)
+		if (pos-- == 0)
+			return mfc;
+	spin_unlock_bh(&mfc_unres_lock);
+
+	it->cache = NULL;
+	return NULL;
+}
+
+
+
+
+/*
+ *	The /proc interfaces to multicast routing /proc/ip6_mr_cache /proc/ip6_mr_vif
+ */
+
+struct ipmr_vif_iter {
+	int ct;
+};
+
+static struct mif_device *ip6mr_vif_seq_idx(struct ipmr_vif_iter *iter,
+					    loff_t pos)
+{
+	for (iter->ct = 0; iter->ct < maxvif; ++iter->ct) {
+		if (!MIF_EXISTS(iter->ct))
+			continue;
+		if (pos-- == 0)
+			return &vif6_table[iter->ct];
+	}
+	return NULL;
+}
+
+static void *ip6mr_vif_seq_start(struct seq_file *seq, loff_t *pos)
+	__acquires(mrt_lock)
+{
+	read_lock(&mrt_lock);
+	return (*pos ? ip6mr_vif_seq_idx(seq->private, *pos - 1)
+		: SEQ_START_TOKEN);
+}
+
+static void *ip6mr_vif_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	struct ipmr_vif_iter *iter = seq->private;
+
+	++*pos;
+	if (v == SEQ_START_TOKEN)
+		return ip6mr_vif_seq_idx(iter, 0);
+
+	while (++iter->ct < maxvif) {
+		if (!MIF_EXISTS(iter->ct))
+			continue;
+		return &vif6_table[iter->ct];
+	}
+	return NULL;
+}
+
+static void ip6mr_vif_seq_stop(struct seq_file *seq, void *v)
+	__releases(mrt_lock)
+{
+	read_unlock(&mrt_lock);
+}
+
+static int ip6mr_vif_seq_show(struct seq_file *seq, void *v)
+{
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(seq,
+			 "Interface      BytesIn  PktsIn  BytesOut PktsOut Flags\n");
+	} else {
+		const struct mif_device *vif = v;
+		const char *name = vif->dev ? vif->dev->name : "none";
+
+		seq_printf(seq,
+			   "%2Zd %-10s %8ld %7ld  %8ld %7ld %05X\n",
+			   vif - vif6_table,
+			   name, vif->bytes_in, vif->pkt_in,
+			   vif->bytes_out, vif->pkt_out,
+			   vif->flags);
+	}
+	return 0;
+}
+
+static struct seq_operations ip6mr_vif_seq_ops = {
+	.start = ip6mr_vif_seq_start,
+	.next  = ip6mr_vif_seq_next,
+	.stop  = ip6mr_vif_seq_stop,
+	.show  = ip6mr_vif_seq_show,
+};
+
+static int ip6mr_vif_open(struct inode *inode, struct file *file)
+{
+	return seq_open_private(file, &ip6mr_vif_seq_ops,
+				sizeof(struct ipmr_vif_iter));
+}
+
+static struct file_operations ip6mr_vif_fops = {
+	.owner	 = THIS_MODULE,
+	.open    = ip6mr_vif_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = seq_release,
+};
+
+static void *ipmr_mfc_seq_start(struct seq_file *seq, loff_t *pos)
+{
+	return (*pos ? ipmr_mfc_seq_idx(seq->private, *pos - 1)
+		: SEQ_START_TOKEN);
+}
+
+static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	struct mfc6_cache *mfc = v;
+	struct ipmr_mfc_iter *it = seq->private;
+
+	++*pos;
+
+	if (v == SEQ_START_TOKEN)
+		return ipmr_mfc_seq_idx(seq->private, 0);
+
+	if (mfc->next)
+		return mfc->next;
+
+	if (it->cache == &mfc_unres_queue)
+		goto end_of_list;
+
+	BUG_ON(it->cache != mfc6_cache_array);
+
+	while (++it->ct < ARRAY_SIZE(mfc6_cache_array)) {
+		mfc = mfc6_cache_array[it->ct];
+		if (mfc)
+			return mfc;
+	}
+
+	/* exhausted cache_array, show unresolved */
+	read_unlock(&mrt_lock);
+	it->cache = &mfc_unres_queue;
+	it->ct = 0;
+
+	spin_lock_bh(&mfc_unres_lock);
+	mfc = mfc_unres_queue;
+	if (mfc)
+		return mfc;
+
+ end_of_list:
+	spin_unlock_bh(&mfc_unres_lock);
+	it->cache = NULL;
+
+	return NULL;
+}
+
+static void ipmr_mfc_seq_stop(struct seq_file *seq, void *v)
+{
+	struct ipmr_mfc_iter *it = seq->private;
+
+	if (it->cache == &mfc_unres_queue)
+		spin_unlock_bh(&mfc_unres_lock);
+	else if (it->cache == mfc6_cache_array)
+		read_unlock(&mrt_lock);
+}
+
+static int ipmr_mfc_seq_show(struct seq_file *seq, void *v)
+{
+	int n;
+
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(seq,
+			 "Group                            "
+			 "Origin                           "
+			 "Iif      Pkts  Bytes     Wrong  Oifs\n");
+	} else {
+		const struct mfc6_cache *mfc = v;
+		const struct ipmr_mfc_iter *it = seq->private;
+
+		seq_printf(seq,
+			   NIP6_FMT " " NIP6_FMT " %-3d %8ld %8ld %8ld",
+			   NIP6(mfc->mf6c_mcastgrp), NIP6(mfc->mf6c_origin),
+			   mfc->mf6c_parent,
+			   mfc->mfc_un.res.pkt,
+			   mfc->mfc_un.res.bytes,
+			   mfc->mfc_un.res.wrong_if);
+
+		if (it->cache != &mfc_unres_queue) {
+			for (n = mfc->mfc_un.res.minvif;
+			     n < mfc->mfc_un.res.maxvif; n++) {
+				if (MIF_EXISTS(n) &&
+				    mfc->mfc_un.res.ttls[n] < 255)
+					seq_printf(seq,
+						   " %2d:%-3d",
+						   n, mfc->mfc_un.res.ttls[n]);
+			}
+		}
+		seq_putc(seq, '\n');
+	}
+	return 0;
+}
+
+static struct seq_operations ipmr_mfc_seq_ops = {
+	.start = ipmr_mfc_seq_start,
+	.next  = ipmr_mfc_seq_next,
+	.stop  = ipmr_mfc_seq_stop,
+	.show  = ipmr_mfc_seq_show,
+};
+
+static int ipmr_mfc_open(struct inode *inode, struct file *file)
+{
+	return seq_open_private(file, &ipmr_mfc_seq_ops,
+				sizeof(struct ipmr_mfc_iter));
+}
+
+static struct file_operations ip6mr_mfc_fops = {
+	.owner	 = THIS_MODULE,
+	.open    = ipmr_mfc_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = seq_release,
+};
+#endif
+
+/*
+ *	Delete a VIF entry
+ */
+
+static int mif6_delete(int vifi)
+{
+	struct mif_device *v;
+	struct net_device *dev;
+	if (vifi < 0 || vifi >= maxvif)
+		return -EADDRNOTAVAIL;
+
+	v = &vif6_table[vifi];
+
+	write_lock_bh(&mrt_lock);
+	dev = v->dev;
+	v->dev = NULL;
+
+	if (!dev) {
+		write_unlock_bh(&mrt_lock);
+		return -EADDRNOTAVAIL;
+	}
+
+	if (vifi + 1 == maxvif) {
+		int tmp;
+		for (tmp = vifi - 1; tmp >= 0; tmp--) {
+			if (MIF_EXISTS(tmp))
+				break;
+		}
+		maxvif = tmp + 1;
+	}
+
+	write_unlock_bh(&mrt_lock);
+
+	dev_set_allmulti(dev, -1);
+
+	if (v->flags & MIFF_REGISTER)
+		unregister_netdevice(dev);
+
+	dev_put(dev);
+	return 0;
+}
+
+/* Destroy an unresolved cache entry, killing queued skbs
+   and reporting error to netlink readers.
+ */
+
+static void ip6mr_destroy_unres(struct mfc6_cache *c)
+{
+	struct sk_buff *skb;
+
+	atomic_dec(&cache_resolve_queue_len);
+
+	while((skb = skb_dequeue(&c->mfc_un.unres.unresolved)) != NULL) {
+		if (ipv6_hdr(skb)->version == 0) {
+			struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct ipv6hdr));
+			nlh->nlmsg_type = NLMSG_ERROR;
+			nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nlmsgerr));
+			skb_trim(skb, nlh->nlmsg_len);
+			((struct nlmsgerr *)NLMSG_DATA(nlh))->error = -ETIMEDOUT;
+			rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid);
+		} else
+			kfree_skb(skb);
+	}
+
+	kmem_cache_free(mrt_cachep, c);
+}
+
+
+/* Single timer process for all the unresolved queue. */
+
+static void ipmr_do_expire_process(unsigned long dummy)
+{
+	unsigned long now = jiffies;
+	unsigned long expires = 10 * HZ;
+	struct mfc6_cache *c, **cp;
+
+	cp = &mfc_unres_queue;
+
+	while ((c = *cp) != NULL) {
+		if (time_after(c->mfc_un.unres.expires, now)) {
+			/* not yet... */
+			unsigned long interval = c->mfc_un.unres.expires - now;
+			if (interval < expires)
+				expires = interval;
+			cp = &c->next;
+			continue;
+		}
+
+		*cp = c->next;
+		ip6mr_destroy_unres(c);
+	}
+
+	if (atomic_read(&cache_resolve_queue_len))
+		mod_timer(&ipmr_expire_timer, jiffies + expires);
+}
+
+static void ipmr_expire_process(unsigned long dummy)
+{
+	if (!spin_trylock(&mfc_unres_lock)) {
+		mod_timer(&ipmr_expire_timer, jiffies + 1);
+		return;
+	}
+
+	if (atomic_read(&cache_resolve_queue_len))
+		ipmr_do_expire_process(dummy);
+
+	spin_unlock(&mfc_unres_lock);
+}
+
+/* Fill oifs list. It is called under write locked mrt_lock. */
+
+static void ip6mr_update_thresholds(struct mfc6_cache *cache, unsigned char *ttls)
+{
+	int vifi;
+
+	cache->mfc_un.res.minvif = MAXVIFS;
+	cache->mfc_un.res.maxvif = 0;
+	memset(cache->mfc_un.res.ttls, 255, MAXVIFS);
+
+	for (vifi = 0; vifi < maxvif; vifi++) {
+		if (MIF_EXISTS(vifi) && ttls[vifi] && ttls[vifi] < 255) {
+			cache->mfc_un.res.ttls[vifi] = ttls[vifi];
+			if (cache->mfc_un.res.minvif > vifi)
+				cache->mfc_un.res.minvif = vifi;
+			if (cache->mfc_un.res.maxvif <= vifi)
+				cache->mfc_un.res.maxvif = vifi + 1;
+		}
+	}
+}
+
+static int mif6_add(struct mif6ctl *vifc, int mrtsock)
+{
+	int vifi = vifc->mif6c_mifi;
+	struct mif_device *v = &vif6_table[vifi];
+	struct net_device *dev;
+
+	/* Is vif busy ? */
+	if (MIF_EXISTS(vifi))
+		return -EADDRINUSE;
+
+	switch (vifc->mif6c_flags) {
+	case 0:
+		dev = dev_get_by_index(&init_net, vifc->mif6c_pifi);
+		if (!dev)
+			return -EADDRNOTAVAIL;
+		dev_put(dev);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	dev_set_allmulti(dev, 1);
+
+	/*
+	 *	Fill in the VIF structures
+	 */
+	v->rate_limit = vifc->vifc_rate_limit;
+	v->flags = vifc->mif6c_flags;
+	if (!mrtsock)
+		v->flags |= VIFF_STATIC;
+	v->threshold = vifc->vifc_threshold;
+	v->bytes_in = 0;
+	v->bytes_out = 0;
+	v->pkt_in = 0;
+	v->pkt_out = 0;
+	v->link = dev->ifindex;
+	if (v->flags & MIFF_REGISTER)
+		v->link = dev->iflink;
+
+	/* And finish update writing critical data */
+	write_lock_bh(&mrt_lock);
+	dev_hold(dev);
+	v->dev = dev;
+	if (vifi + 1 > maxvif)
+		maxvif = vifi + 1;
+	write_unlock_bh(&mrt_lock);
+	return 0;
+}
+
+static struct mfc6_cache *ip6mr_cache_find(struct in6_addr *origin, struct in6_addr *mcastgrp)
+{
+	int line = MFC6_HASH(mcastgrp, origin);
+	struct mfc6_cache *c;
+
+	for (c = mfc6_cache_array[line]; c; c = c->next) {
+		if (ipv6_addr_equal(&c->mf6c_origin, origin) &&
+		    ipv6_addr_equal(&c->mf6c_mcastgrp, mcastgrp))
+			break;
+	}
+	return c;
+}
+
+/*
+ *	Allocate a multicast cache entry
+ */
+static struct mfc6_cache *ip6mr_cache_alloc(void)
+{
+	struct mfc6_cache *c = kmem_cache_alloc(mrt_cachep, GFP_KERNEL);
+	if (c == NULL)
+		return NULL;
+	memset(c, 0, sizeof(*c));
+	c->mfc_un.res.minvif = MAXVIFS;
+	return c;
+}
+
+static struct mfc6_cache *ip6mr_cache_alloc_unres(void)
+{
+	struct mfc6_cache *c = kmem_cache_alloc(mrt_cachep, GFP_ATOMIC);
+	if (c == NULL)
+		return NULL;
+	memset(c, 0, sizeof(*c));
+	skb_queue_head_init(&c->mfc_un.unres.unresolved);
+	c->mfc_un.unres.expires = jiffies + 10 * HZ;
+	return c;
+}
+
+/*
+ *	A cache entry has gone into a resolved state from queued
+ */
+
+static void ip6mr_cache_resolve(struct mfc6_cache *uc, struct mfc6_cache *c)
+{
+	struct sk_buff *skb;
+
+	/*
+	 *	Play the pending entries through our router
+	 */
+
+	while((skb = __skb_dequeue(&uc->mfc_un.unres.unresolved))) {
+		if (ipv6_hdr(skb)->version == 0) {
+			int err;
+			struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct ipv6hdr));
+
+			if (ip6mr_fill_mroute(skb, c, NLMSG_DATA(nlh)) > 0) {
+				nlh->nlmsg_len = skb->tail - (u8 *)nlh;
+			} else {
+				nlh->nlmsg_type = NLMSG_ERROR;
+				nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nlmsgerr));
+				skb_trim(skb, nlh->nlmsg_len);
+				((struct nlmsgerr *)NLMSG_DATA(nlh))->error = -EMSGSIZE;
+			}
+			err = rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid);
+		} else
+			ip6_mr_forward(skb, c);
+	}
+}
+
+/*
+ *	Bounce a cache query up to pim6sd. We could use netlink for this but pim6sd
+ *	expects the following bizarre scheme.
+ *
+ *	Called under mrt_lock.
+ */
+
+static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert)
+{
+	struct sk_buff *skb;
+	struct mrt6msg *msg;
+	int ret;
+
+	skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC);
+
+	if (!skb)
+		return -ENOBUFS;
+
+	/* I suppose that internal messages
+	 * do not require checksums */
+
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	/*
+	 *	Copy the IP header
+	 */
+
+	skb_put(skb, sizeof(struct ipv6hdr));
+	skb_reset_network_header(skb);
+	skb_copy_to_linear_data(skb, ipv6_hdr(pkt), sizeof(struct ipv6hdr));
+
+	/*
+	 *	Add our header
+	 */
+	skb_put(skb, sizeof(*msg));
+	skb_reset_transport_header(skb);
+	msg = (struct mrt6msg *)skb_transport_header(skb);
+
+	msg->im6_mbz = 0;
+	msg->im6_msgtype = assert;
+	msg->im6_mif = vifi;
+	msg->im6_pad = 0;
+	ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr);
+	ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr);
+
+	skb->dst = dst_clone(pkt->dst);
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	skb_pull(skb, sizeof(struct ipv6hdr));
+
+	if (mroute6_socket == NULL) {
+		kfree_skb(skb);
+		return -EINVAL;
+	}
+
+	/*
+	 *	Deliver to user space multicast routing algorithms
+	 */
+	if ((ret = sock_queue_rcv_skb(mroute6_socket, skb)) < 0) {
+		if (net_ratelimit())
+			printk(KERN_WARNING "mroute6: pending queue full, dropping entries.\n");
+		kfree_skb(skb);
+	}
+
+	return ret;
+}
+
+/*
+ *	Queue a packet for resolution. It gets locked cache entry!
+ */
+
+static int
+ip6mr_cache_unresolved(vifi_t vifi, struct sk_buff *skb)
+{
+	int err;
+	struct mfc6_cache *c;
+
+	spin_lock_bh(&mfc_unres_lock);
+	for (c = mfc_unres_queue; c; c = c->next) {
+		if (ipv6_addr_equal(&c->mf6c_mcastgrp, &ipv6_hdr(skb)->daddr) &&
+		    ipv6_addr_equal(&c->mf6c_origin, &ipv6_hdr(skb)->saddr))
+			break;
+	}
+
+	if (c == NULL) {
+		/*
+		 *	Create a new entry if allowable
+		 */
+
+		if (atomic_read(&cache_resolve_queue_len) >= 10 ||
+		    (c = ip6mr_cache_alloc_unres()) == NULL) {
+			spin_unlock_bh(&mfc_unres_lock);
+
+			kfree_skb(skb);
+			return -ENOBUFS;
+		}
+
+		/*
+		 *	Fill in the new cache entry
+		 */
+		c->mf6c_parent = -1;
+		c->mf6c_origin = ipv6_hdr(skb)->saddr;
+		c->mf6c_mcastgrp = ipv6_hdr(skb)->daddr;
+
+		/*
+		 *	Reflect first query at pim6sd
+		 */
+		if ((err = ip6mr_cache_report(skb, vifi, MRT6MSG_NOCACHE)) < 0) {
+			/* If the report failed throw the cache entry
+			   out - Brad Parker
+			 */
+			spin_unlock_bh(&mfc_unres_lock);
+
+			kmem_cache_free(mrt_cachep, c);
+			kfree_skb(skb);
+			return err;
+		}
+
+		atomic_inc(&cache_resolve_queue_len);
+		c->next = mfc_unres_queue;
+		mfc_unres_queue = c;
+
+		ipmr_do_expire_process(1);
+	}
+
+	/*
+	 *	See if we can append the packet
+	 */
+	if (c->mfc_un.unres.unresolved.qlen > 3) {
+		kfree_skb(skb);
+		err = -ENOBUFS;
+	} else {
+		skb_queue_tail(&c->mfc_un.unres.unresolved, skb);
+		err = 0;
+	}
+
+	spin_unlock_bh(&mfc_unres_lock);
+	return err;
+}
+
+/*
+ *	MFC6 cache manipulation by user space
+ */
+
+static int ip6mr_mfc_delete(struct mf6cctl *mfc)
+{
+	int line;
+	struct mfc6_cache *c, **cp;
+
+	line = MFC6_HASH(&mfc->mf6cc_mcastgrp.sin6_addr, &mfc->mf6cc_origin.sin6_addr);
+
+	for (cp = &mfc6_cache_array[line]; (c = *cp) != NULL; cp = &c->next) {
+		if (ipv6_addr_equal(&c->mf6c_origin, &mfc->mf6cc_origin.sin6_addr) &&
+		    ipv6_addr_equal(&c->mf6c_mcastgrp, &mfc->mf6cc_mcastgrp.sin6_addr)) {
+			write_lock_bh(&mrt_lock);
+			*cp = c->next;
+			write_unlock_bh(&mrt_lock);
+
+			kmem_cache_free(mrt_cachep, c);
+			return 0;
+		}
+	}
+	return -ENOENT;
+}
+
+static int ip6mr_device_event(struct notifier_block *this,
+			      unsigned long event, void *ptr)
+{
+	struct net_device *dev = ptr;
+	struct mif_device *v;
+	int ct;
+
+	if (dev_net(dev) != &init_net)
+		return NOTIFY_DONE;
+
+	if (event != NETDEV_UNREGISTER)
+		return NOTIFY_DONE;
+
+	v = &vif6_table[0];
+	for (ct = 0; ct < maxvif; ct++, v++) {
+		if (v->dev == dev)
+			mif6_delete(ct);
+	}
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block ip6_mr_notifier = {
+	.notifier_call = ip6mr_device_event
+};
+
+/*
+ *	Setup for IP multicast routing
+ */
+
+void __init ip6_mr_init(void)
+{
+	mrt_cachep = kmem_cache_create("ip6_mrt_cache",
+				       sizeof(struct mfc6_cache),
+				       0, SLAB_HWCACHE_ALIGN,
+				       NULL);
+	if (!mrt_cachep)
+		panic("cannot allocate ip6_mrt_cache");
+
+	setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0);
+	register_netdevice_notifier(&ip6_mr_notifier);
+#ifdef CONFIG_PROC_FS
+	proc_net_fops_create(&init_net, "ip6_mr_vif", 0, &ip6mr_vif_fops);
+	proc_net_fops_create(&init_net, "ip6_mr_cache", 0, &ip6mr_mfc_fops);
+#endif
+}
+
+
+static int ip6mr_mfc_add(struct mf6cctl *mfc, int mrtsock)
+{
+	int line;
+	struct mfc6_cache *uc, *c, **cp;
+	unsigned char ttls[MAXVIFS];
+	int i;
+
+	memset(ttls, 255, MAXVIFS);
+	for (i = 0; i < MAXVIFS; i++) {
+		if (IF_ISSET(i, &mfc->mf6cc_ifset))
+			ttls[i] = 1;
+
+	}
+
+	line = MFC6_HASH(&mfc->mf6cc_mcastgrp.sin6_addr, &mfc->mf6cc_origin.sin6_addr);
+
+	for (cp = &mfc6_cache_array[line]; (c = *cp) != NULL; cp = &c->next) {
+		if (ipv6_addr_equal(&c->mf6c_origin, &mfc->mf6cc_origin.sin6_addr) &&
+		    ipv6_addr_equal(&c->mf6c_mcastgrp, &mfc->mf6cc_mcastgrp.sin6_addr))
+			break;
+	}
+
+	if (c != NULL) {
+		write_lock_bh(&mrt_lock);
+		c->mf6c_parent = mfc->mf6cc_parent;
+		ip6mr_update_thresholds(c, ttls);
+		if (!mrtsock)
+			c->mfc_flags |= MFC_STATIC;
+		write_unlock_bh(&mrt_lock);
+		return 0;
+	}
+
+	if (!ipv6_addr_is_multicast(&mfc->mf6cc_mcastgrp.sin6_addr))
+		return -EINVAL;
+
+	c = ip6mr_cache_alloc();
+	if (c == NULL)
+		return -ENOMEM;
+
+	c->mf6c_origin = mfc->mf6cc_origin.sin6_addr;
+	c->mf6c_mcastgrp = mfc->mf6cc_mcastgrp.sin6_addr;
+	c->mf6c_parent = mfc->mf6cc_parent;
+	ip6mr_update_thresholds(c, ttls);
+	if (!mrtsock)
+		c->mfc_flags |= MFC_STATIC;
+
+	write_lock_bh(&mrt_lock);
+	c->next = mfc6_cache_array[line];
+	mfc6_cache_array[line] = c;
+	write_unlock_bh(&mrt_lock);
+
+	/*
+	 *	Check to see if we resolved a queued list. If so we
+	 *	need to send on the frames and tidy up.
+	 */
+	spin_lock_bh(&mfc_unres_lock);
+	for (cp = &mfc_unres_queue; (uc = *cp) != NULL;
+	     cp = &uc->next) {
+		if (ipv6_addr_equal(&uc->mf6c_origin, &c->mf6c_origin) &&
+		    ipv6_addr_equal(&uc->mf6c_mcastgrp, &c->mf6c_mcastgrp)) {
+			*cp = uc->next;
+			if (atomic_dec_and_test(&cache_resolve_queue_len))
+				del_timer(&ipmr_expire_timer);
+			break;
+		}
+	}
+	spin_unlock_bh(&mfc_unres_lock);
+
+	if (uc) {
+		ip6mr_cache_resolve(uc, c);
+		kmem_cache_free(mrt_cachep, uc);
+	}
+	return 0;
+}
+
+/*
+ *	Close the multicast socket, and clear the vif tables etc
+ */
+
+static void mroute_clean_tables(struct sock *sk)
+{
+	int i;
+
+	/*
+	 *	Shut down all active vif entries
+	 */
+	for (i = 0; i < maxvif; i++) {
+		if (!(vif6_table[i].flags & VIFF_STATIC))
+			mif6_delete(i);
+	}
+
+	/*
+	 *	Wipe the cache
+	 */
+	for (i = 0; i < ARRAY_SIZE(mfc6_cache_array); i++) {
+		struct mfc6_cache *c, **cp;
+
+		cp = &mfc6_cache_array[i];
+		while ((c = *cp) != NULL) {
+			if (c->mfc_flags & MFC_STATIC) {
+				cp = &c->next;
+				continue;
+			}
+			write_lock_bh(&mrt_lock);
+			*cp = c->next;
+			write_unlock_bh(&mrt_lock);
+
+			kmem_cache_free(mrt_cachep, c);
+		}
+	}
+
+	if (atomic_read(&cache_resolve_queue_len) != 0) {
+		struct mfc6_cache *c;
+
+		spin_lock_bh(&mfc_unres_lock);
+		while (mfc_unres_queue != NULL) {
+			c = mfc_unres_queue;
+			mfc_unres_queue = c->next;
+			spin_unlock_bh(&mfc_unres_lock);
+
+			ip6mr_destroy_unres(c);
+
+			spin_lock_bh(&mfc_unres_lock);
+		}
+		spin_unlock_bh(&mfc_unres_lock);
+	}
+}
+
+static int ip6mr_sk_init(struct sock *sk)
+{
+	int err = 0;
+
+	rtnl_lock();
+	write_lock_bh(&mrt_lock);
+	if (likely(mroute6_socket == NULL))
+		mroute6_socket = sk;
+	else
+		err = -EADDRINUSE;
+	write_unlock_bh(&mrt_lock);
+
+	rtnl_unlock();
+
+	return err;
+}
+
+int ip6mr_sk_done(struct sock *sk)
+{
+	int err = 0;
+
+	rtnl_lock();
+	if (sk == mroute6_socket) {
+		write_lock_bh(&mrt_lock);
+		mroute6_socket = NULL;
+		write_unlock_bh(&mrt_lock);
+
+		mroute_clean_tables(sk);
+	} else
+		err = -EACCES;
+	rtnl_unlock();
+
+	return err;
+}
+
+/*
+ *	Socket options and virtual interface manipulation. The whole
+ *	virtual interface system is a complete heap, but unfortunately
+ *	that's how BSD mrouted happens to think. Maybe one day with a proper
+ *	MOSPF/PIM router set up we can clean this up.
+ */
+
+int ip6_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int optlen)
+{
+	int ret;
+	struct mif6ctl vif;
+	struct mf6cctl mfc;
+	mifi_t mifi;
+
+	if (optname != MRT6_INIT) {
+		if (sk != mroute6_socket && !capable(CAP_NET_ADMIN))
+			return -EACCES;
+	}
+
+	switch (optname) {
+	case MRT6_INIT:
+		if (sk->sk_type != SOCK_RAW ||
+		    inet_sk(sk)->num != IPPROTO_ICMPV6)
+			return -EOPNOTSUPP;
+		if (optlen < sizeof(int))
+			return -EINVAL;
+
+		return ip6mr_sk_init(sk);
+
+	case MRT6_DONE:
+		return ip6mr_sk_done(sk);
+
+	case MRT6_ADD_MIF:
+		if (optlen < sizeof(vif))
+			return -EINVAL;
+		if (copy_from_user(&vif, optval, sizeof(vif)))
+			return -EFAULT;
+		if (vif.mif6c_mifi >= MAXVIFS)
+			return -ENFILE;
+		rtnl_lock();
+		ret = mif6_add(&vif, sk == mroute6_socket);
+		rtnl_unlock();
+		return ret;
+
+	case MRT6_DEL_MIF:
+		if (optlen < sizeof(mifi_t))
+			return -EINVAL;
+		if (copy_from_user(&mifi, optval, sizeof(mifi_t)))
+			return -EFAULT;
+		rtnl_lock();
+		ret = mif6_delete(mifi);
+		rtnl_unlock();
+		return ret;
+
+	/*
+	 *	Manipulate the forwarding caches. These live
+	 *	in a sort of kernel/user symbiosis.
+	 */
+	case MRT6_ADD_MFC:
+	case MRT6_DEL_MFC:
+		if (optlen < sizeof(mfc))
+			return -EINVAL;
+		if (copy_from_user(&mfc, optval, sizeof(mfc)))
+			return -EFAULT;
+		rtnl_lock();
+		if (optname == MRT6_DEL_MFC)
+			ret = ip6mr_mfc_delete(&mfc);
+		else
+			ret = ip6mr_mfc_add(&mfc, sk == mroute6_socket);
+		rtnl_unlock();
+		return ret;
+
+	/*
+	 *	Spurious command, or MRT_VERSION which you cannot
+	 *	set.
+	 */
+	default:
+		return -ENOPROTOOPT;
+	}
+}
+
+/*
+ *	Getsock opt support for the multicast routing system.
+ */
+
+int ip6_mroute_getsockopt(struct sock *sk, int optname, char __user *optval,
+			  int __user *optlen)
+{
+	int olr;
+	int val;
+
+	switch (optname) {
+	case MRT6_VERSION:
+		val = 0x0305;
+		break;
+	default:
+		return -ENOPROTOOPT;
+	}
+
+	if (get_user(olr, optlen))
+		return -EFAULT;
+
+	olr = min_t(int, olr, sizeof(int));
+	if (olr < 0)
+		return -EINVAL;
+
+	if (put_user(olr, optlen))
+		return -EFAULT;
+	if (copy_to_user(optval, &val, olr))
+		return -EFAULT;
+	return 0;
+}
+
+/*
+ *	The IP multicast ioctl support routines.
+ */
+
+int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg)
+{
+	struct sioc_sg_req6 sr;
+	struct sioc_mif_req6 vr;
+	struct mif_device *vif;
+	struct mfc6_cache *c;
+
+	switch (cmd) {
+	case SIOCGETMIFCNT_IN6:
+		if (copy_from_user(&vr, arg, sizeof(vr)))
+			return -EFAULT;
+		if (vr.mifi >= maxvif)
+			return -EINVAL;
+		read_lock(&mrt_lock);
+		vif = &vif6_table[vr.mifi];
+		if (MIF_EXISTS(vr.mifi)) {
+			vr.icount = vif->pkt_in;
+			vr.ocount = vif->pkt_out;
+			vr.ibytes = vif->bytes_in;
+			vr.obytes = vif->bytes_out;
+			read_unlock(&mrt_lock);
+
+			if (copy_to_user(arg, &vr, sizeof(vr)))
+				return -EFAULT;
+			return 0;
+		}
+		read_unlock(&mrt_lock);
+		return -EADDRNOTAVAIL;
+	case SIOCGETSGCNT_IN6:
+		if (copy_from_user(&sr, arg, sizeof(sr)))
+			return -EFAULT;
+
+		read_lock(&mrt_lock);
+		c = ip6mr_cache_find(&sr.src.sin6_addr, &sr.grp.sin6_addr);
+		if (c) {
+			sr.pktcnt = c->mfc_un.res.pkt;
+			sr.bytecnt = c->mfc_un.res.bytes;
+			sr.wrong_if = c->mfc_un.res.wrong_if;
+			read_unlock(&mrt_lock);
+
+			if (copy_to_user(arg, &sr, sizeof(sr)))
+				return -EFAULT;
+			return 0;
+		}
+		read_unlock(&mrt_lock);
+		return -EADDRNOTAVAIL;
+	default:
+		return -ENOIOCTLCMD;
+	}
+}
+
+
+static inline int ip6mr_forward2_finish(struct sk_buff *skb)
+{
+	/* XXX stats */
+	return dst_output(skb);
+}
+
+/*
+ *	Processing handlers for ip6mr_forward
+ */
+
+static int ip6mr_forward2(struct sk_buff *skb, struct mfc6_cache *c, int vifi)
+{
+	struct ipv6hdr *ipv6h;
+	struct mif_device *vif = &vif6_table[vifi];
+	struct net_device *dev;
+	struct dst_entry *dst;
+	struct flowi fl;
+
+	if (vif->dev == NULL)
+		goto out_free;
+
+	ipv6h = ipv6_hdr(skb);
+
+	fl = (struct flowi) {
+		.oif = vif->link,
+		.nl_u = { .ip6_u =
+				{ .daddr = ipv6h->daddr, }
+		}
+	};
+
+	dst = ip6_route_output(&init_net, NULL, &fl);
+	if (!dst)
+		goto out_free;
+
+	dst_release(skb->dst);
+	skb->dst = dst;
+
+	/*
+	 * RFC1584 teaches, that DVMRP/PIM router must deliver packets locally
+	 * not only before forwarding, but after forwarding on all output
+	 * interfaces. It is clear, if mrouter runs a multicasting
+	 * program, it should receive packets not depending to what interface
+	 * program is joined.
+	 * If we will not make it, the program will have to join on all
+	 * interfaces. On the other hand, multihoming host (or router, but
+	 * not mrouter) cannot join to more than one interface - it will
+	 * result in receiving multiple packets.
+	 */
+	dev = vif->dev;
+	skb->dev = dev;
+	vif->pkt_out++;
+	vif->bytes_out += skb->len;
+
+	/* We are about to write */
+	/* XXX: extension headers? */
+	if (skb_cow(skb, sizeof(*ipv6h) + LL_RESERVED_SPACE(dev)))
+		goto out_free;
+
+	ipv6h = ipv6_hdr(skb);
+	ipv6h->hop_limit--;
+
+	IP6CB(skb)->flags |= IP6SKB_FORWARDED;
+
+	return NF_HOOK(PF_INET6, NF_INET_FORWARD, skb, skb->dev, dev,
+		       ip6mr_forward2_finish);
+
+out_free:
+	kfree_skb(skb);
+	return 0;
+}
+
+static int ip6mr_find_vif(struct net_device *dev)
+{
+	int ct;
+	for (ct = maxvif - 1; ct >= 0; ct--) {
+		if (vif6_table[ct].dev == dev)
+			break;
+	}
+	return ct;
+}
+
+static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache)
+{
+	int psend = -1;
+	int vif, ct;
+
+	vif = cache->mf6c_parent;
+	cache->mfc_un.res.pkt++;
+	cache->mfc_un.res.bytes += skb->len;
+
+	vif6_table[vif].pkt_in++;
+	vif6_table[vif].bytes_in += skb->len;
+
+	/*
+	 *	Forward the frame
+	 */
+	for (ct = cache->mfc_un.res.maxvif - 1; ct >= cache->mfc_un.res.minvif; ct--) {
+		if (ipv6_hdr(skb)->hop_limit > cache->mfc_un.res.ttls[ct]) {
+			if (psend != -1) {
+				struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
+				if (skb2)
+					ip6mr_forward2(skb2, cache, psend);
+			}
+			psend = ct;
+		}
+	}
+	if (psend != -1) {
+		ip6mr_forward2(skb, cache, psend);
+		return 0;
+	}
+
+	kfree_skb(skb);
+	return 0;
+}
+
+
+/*
+ *	Multicast packets for forwarding arrive here
+ */
+
+int ip6_mr_input(struct sk_buff *skb)
+{
+	struct mfc6_cache *cache;
+
+	read_lock(&mrt_lock);
+	cache = ip6mr_cache_find(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr);
+
+	/*
+	 *	No usable cache entry
+	 */
+	if (cache == NULL) {
+		int vif;
+
+		vif = ip6mr_find_vif(skb->dev);
+		if (vif >= 0) {
+			int err = ip6mr_cache_unresolved(vif, skb);
+			read_unlock(&mrt_lock);
+
+			return err;
+		}
+		read_unlock(&mrt_lock);
+		kfree_skb(skb);
+		return -ENODEV;
+	}
+
+	ip6_mr_forward(skb, cache);
+
+	read_unlock(&mrt_lock);
+
+	return 0;
+}
+
+
+static int
+ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm)
+{
+	int ct;
+	struct rtnexthop *nhp;
+	struct net_device *dev = vif6_table[c->mf6c_parent].dev;
+	u8 *b = skb->tail;
+	struct rtattr *mp_head;
+
+	if (dev)
+		RTA_PUT(skb, RTA_IIF, 4, &dev->ifindex);
+
+	mp_head = (struct rtattr *)skb_put(skb, RTA_LENGTH(0));
+
+	for (ct = c->mfc_un.res.minvif; ct < c->mfc_un.res.maxvif; ct++) {
+		if (c->mfc_un.res.ttls[ct] < 255) {
+			if (skb_tailroom(skb) < RTA_ALIGN(RTA_ALIGN(sizeof(*nhp)) + 4))
+				goto rtattr_failure;
+			nhp = (struct rtnexthop *)skb_put(skb, RTA_ALIGN(sizeof(*nhp)));
+			nhp->rtnh_flags = 0;
+			nhp->rtnh_hops = c->mfc_un.res.ttls[ct];
+			nhp->rtnh_ifindex = vif6_table[ct].dev->ifindex;
+			nhp->rtnh_len = sizeof(*nhp);
+		}
+	}
+	mp_head->rta_type = RTA_MULTIPATH;
+	mp_head->rta_len = skb->tail - (u8 *)mp_head;
+	rtm->rtm_type = RTN_MULTICAST;
+	return 1;
+
+rtattr_failure:
+	nlmsg_trim(skb, b);
+	return -EMSGSIZE;
+}
+
+int ip6mr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait)
+{
+	int err;
+	struct mfc6_cache *cache;
+	struct rt6_info *rt = (struct rt6_info *)skb->dst;
+
+	read_lock(&mrt_lock);
+	cache = ip6mr_cache_find(&rt->rt6i_src.addr, &rt->rt6i_dst.addr);
+
+	if (!cache) {
+		struct sk_buff *skb2;
+		struct ipv6hdr *iph;
+		struct net_device *dev;
+		int vif;
+
+		if (nowait) {
+			read_unlock(&mrt_lock);
+			return -EAGAIN;
+		}
+
+		dev = skb->dev;
+		if (dev == NULL || (vif = ip6mr_find_vif(dev)) < 0) {
+			read_unlock(&mrt_lock);
+			return -ENODEV;
+		}
+
+		/* really correct? */
+		skb2 = alloc_skb(sizeof(struct ipv6hdr), GFP_ATOMIC);
+		if (!skb2) {
+			read_unlock(&mrt_lock);
+			return -ENOMEM;
+		}
+
+		skb_reset_transport_header(skb2);
+
+		skb_put(skb2, sizeof(struct ipv6hdr));
+		skb_reset_network_header(skb2);
+
+		iph = ipv6_hdr(skb2);
+		iph->version = 0;
+		iph->priority = 0;
+		iph->flow_lbl[0] = 0;
+		iph->flow_lbl[1] = 0;
+		iph->flow_lbl[2] = 0;
+		iph->payload_len = 0;
+		iph->nexthdr = IPPROTO_NONE;
+		iph->hop_limit = 0;
+		ipv6_addr_copy(&iph->saddr, &rt->rt6i_src.addr);
+		ipv6_addr_copy(&iph->daddr, &rt->rt6i_dst.addr);
+
+		err = ip6mr_cache_unresolved(vif, skb2);
+		read_unlock(&mrt_lock);
+
+		return err;
+	}
+
+	if (!nowait && (rtm->rtm_flags&RTM_F_NOTIFY))
+		cache->mfc_flags |= MFC_NOTIFY;
+
+	err = ip6mr_fill_mroute(skb, cache, rtm);
+	read_unlock(&mrt_lock);
+	return err;
+}
+
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 4195ac92345e..99624109c010 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -33,6 +33,7 @@
 #include <linux/sockios.h>
 #include <linux/net.h>
 #include <linux/in6.h>
+#include <linux/mroute6.h>
 #include <linux/netdevice.h>
 #include <linux/if_arp.h>
 #include <linux/init.h>
@@ -118,6 +119,9 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 
 	valbool = (val!=0);
 
+	if (ip6_mroute_opt(optname))
+		return ip6_mroute_setsockopt(sk, optname, optval, optlen);
+
 	lock_sock(sk);
 
 	switch (optname) {
@@ -790,6 +794,9 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
 	int len;
 	int val;
 
+	if (ip6_mroute_opt(optname))
+		return ip6_mroute_getsockopt(sk, optname, optval, optlen);
+
 	if (get_user(len, optlen))
 		return -EFAULT;
 	switch (optname) {
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index aae6cedf1709..088b80b4ce74 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -53,6 +53,7 @@
 #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
 #include <net/mip6.h>
 #endif
+#include <linux/mroute6.h>
 
 #include <net/raw.h>
 #include <net/rawv6.h>
@@ -1135,7 +1136,11 @@ static int rawv6_ioctl(struct sock *sk, int cmd, unsigned long arg)
 		}
 
 		default:
+#ifdef CONFIG_IPV6_MROUTE
+			return ip6mr_ioctl(sk, cmd, (void __user *)arg);
+#else
 			return -ENOIOCTLCMD;
+#endif
 	}
 }
 
@@ -1143,7 +1148,7 @@ static void rawv6_close(struct sock *sk, long timeout)
 {
 	if (inet_sk(sk)->num == IPPROTO_RAW)
 		ip6_ra_control(sk, -1, NULL);
-
+	ip6mr_sk_done(sk);
 	sk_common_release(sk);
 }
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index cd82b6db35ff..3c314d5f46c6 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -36,6 +36,7 @@
 #include <linux/route.h>
 #include <linux/netdevice.h>
 #include <linux/in6.h>
+#include <linux/mroute6.h>
 #include <linux/init.h>
 #include <linux/if_arp.h>
 #include <linux/proc_fs.h>
@@ -2106,7 +2107,7 @@ static inline size_t rt6_nlmsg_size(void)
 static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 			 struct in6_addr *dst, struct in6_addr *src,
 			 int iif, int type, u32 pid, u32 seq,
-			 int prefix, unsigned int flags)
+			 int prefix, int nowait, unsigned int flags)
 {
 	struct rtmsg *rtm;
 	struct nlmsghdr *nlh;
@@ -2166,9 +2167,24 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 	} else if (rtm->rtm_src_len)
 		NLA_PUT(skb, RTA_SRC, 16, &rt->rt6i_src.addr);
 #endif
-	if (iif)
-		NLA_PUT_U32(skb, RTA_IIF, iif);
-	else if (dst) {
+	if (iif) {
+#ifdef CONFIG_IPV6_MROUTE
+		if (ipv6_addr_is_multicast(&rt->rt6i_dst.addr)) {
+			int err = ip6mr_get_route(skb, rtm, nowait);
+			if (err <= 0) {
+				if (!nowait) {
+					if (err == 0)
+						return 0;
+					goto nla_put_failure;
+				} else {
+					if (err == -EMSGSIZE)
+						goto nla_put_failure;
+				}
+			}
+		} else
+#endif
+			NLA_PUT_U32(skb, RTA_IIF, iif);
+	} else if (dst) {
 		struct in6_addr saddr_buf;
 		if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev,
 				       dst, 0, &saddr_buf) == 0)
@@ -2211,7 +2227,7 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg)
 
 	return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE,
 		     NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq,
-		     prefix, NLM_F_MULTI);
+		     prefix, 0, NLM_F_MULTI);
 }
 
 static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
@@ -2277,7 +2293,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 
 	err = rt6_fill_node(skb, rt, &fl.fl6_dst, &fl.fl6_src, iif,
 			    RTM_NEWROUTE, NETLINK_CB(in_skb).pid,
-			    nlh->nlmsg_seq, 0, 0);
+			    nlh->nlmsg_seq, 0, 0, 0);
 	if (err < 0) {
 		kfree_skb(skb);
 		goto errout;
@@ -2303,7 +2319,7 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
 		goto errout;
 
 	err = rt6_fill_node(skb, rt, NULL, NULL, 0,
-				event, info->pid, seq, 0, 0);
+				event, info->pid, seq, 0, 0, 0);
 	if (err < 0) {
 		/* -EMSGSIZE implies BUG in rt6_nlmsg_size() */
 		WARN_ON(err == -EMSGSIZE);
-- 
cgit v1.2.3


From 14fb64e1f449ef6666f1c3a3fa4e13aec669b98d Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 3 Apr 2008 09:22:54 +0900
Subject: [IPV6] MROUTE: Support PIM-SM (SSM).

Based on ancient patch by Mickael Hoerdt
<hoerdt@clarinet.u-strasbg.fr>, which is available at
<http://www-r2.u-strasbg.fr/~hoerdt/dev/linux_ipv6_mforwarding/patch-linux-ipv6-mforwarding-0.1a>.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/mroute6.h |   4 +
 net/ipv6/Kconfig        |   7 ++
 net/ipv6/ip6mr.c        | 275 +++++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 285 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
index b92190304e0b..f6469fb90840 100644
--- a/include/linux/mroute6.h
+++ b/include/linux/mroute6.h
@@ -23,6 +23,8 @@
 #define MRT6_ADD_MFC	(MRT6_BASE+4)	/* Add a multicast forwarding entry	*/
 #define MRT6_DEL_MFC	(MRT6_BASE+5)	/* Delete a multicast forwarding entry	*/
 #define MRT6_VERSION	(MRT6_BASE+6)	/* Get the kernel multicast version	*/
+#define MRT6_ASSERT	(MRT6_BASE+7)	/* Activate PIM assert mode		*/
+#define MRT6_PIM	(MRT6_BASE+8)	/* enable PIM code	*/
 
 #define SIOCGETMIFCNT_IN6	SIOCPROTOPRIVATE	/* IP protocol privates */
 #define SIOCGETSGCNT_IN6	(SIOCPROTOPRIVATE+1)
@@ -217,6 +219,8 @@ static inline int ip6mr_sk_done(struct sock *sk) { return 0; }
 
 struct mrt6msg {
 #define MRT6MSG_NOCACHE		1
+#define MRT6MSG_WRONGMIF	2
+#define MRT6MSG_WHOLEPKT	3		/* used for use level encap */
 	__u8		im6_mbz;		/* must be zero		   */
 	__u8		im6_msgtype;		/* what type of message    */
 	__u16		im6_mif;		/* mif rec'd on		   */
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 9a2ea81e499f..82f987b4ef84 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -216,3 +216,10 @@ config IPV6_MROUTE
 	  Experimental support for IPv6 multicast forwarding.
 	  If unsure, say N.
 
+config IPV6_PIMSM_V2
+	bool "IPv6: PIM-SM version 2 support (EXPERIMENTAL)"
+	depends on IPV6_MROUTE
+	---help---
+	  Support for IPv6 PIM multicast routing protocol PIM-SMv2.
+	  If unsure, say N.
+
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 1bdf3c177d58..2b70774be61f 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -54,6 +54,7 @@
 #include <net/ipv6.h>
 #include <net/ip6_route.h>
 #include <linux/mroute6.h>
+#include <linux/pim.h>
 #include <net/addrconf.h>
 #include <linux/netfilter_ipv6.h>
 
@@ -75,6 +76,13 @@ static int maxvif;
 
 #define MIF_EXISTS(idx) (vif6_table[idx].dev != NULL)
 
+static int mroute_do_assert;				/* Set in PIM assert	*/
+#ifdef CONFIG_IPV6_PIMSM_V2
+static int mroute_do_pim;
+#else
+#define mroute_do_pim 0
+#endif
+
 static struct mfc6_cache *mfc6_cache_array[MFC_LINES];	/* Forwarding cache	*/
 
 static struct mfc6_cache *mfc_unres_queue;		/* Queue of unresolved entries */
@@ -97,6 +105,10 @@ static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache);
 static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert);
 static int ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm);
 
+#ifdef CONFIG_IPV6_PIMSM_V2
+static struct inet6_protocol pim6_protocol;
+#endif
+
 static struct timer_list ipmr_expire_timer;
 
 
@@ -339,6 +351,132 @@ static struct file_operations ip6mr_mfc_fops = {
 };
 #endif
 
+#ifdef CONFIG_IPV6_PIMSM_V2
+static int reg_vif_num = -1;
+
+static int pim6_rcv(struct sk_buff *skb)
+{
+	struct pimreghdr *pim;
+	struct ipv6hdr   *encap;
+	struct net_device  *reg_dev = NULL;
+
+	if (!pskb_may_pull(skb, sizeof(*pim) + sizeof(*encap)))
+		goto drop;
+
+	pim = (struct pimreghdr *)skb_transport_header(skb);
+	if (pim->type != ((PIM_VERSION << 4) | PIM_REGISTER) ||
+	    (pim->flags & PIM_NULL_REGISTER) ||
+	    (ip_compute_csum((void *)pim, sizeof(*pim)) != 0 &&
+	     (u16)csum_fold(skb_checksum(skb, 0, skb->len, 0))))
+		goto drop;
+
+	/* check if the inner packet is destined to mcast group */
+	encap = (struct ipv6hdr *)(skb_transport_header(skb) +
+				   sizeof(*pim));
+
+	if (!ipv6_addr_is_multicast(&encap->daddr) ||
+	    encap->payload_len == 0 ||
+	    ntohs(encap->payload_len) + sizeof(*pim) > skb->len)
+		goto drop;
+
+	read_lock(&mrt_lock);
+	if (reg_vif_num >= 0)
+		reg_dev = vif6_table[reg_vif_num].dev;
+	if (reg_dev)
+		dev_hold(reg_dev);
+	read_unlock(&mrt_lock);
+
+	if (reg_dev == NULL)
+		goto drop;
+
+	skb->mac_header = skb->network_header;
+	skb_pull(skb, (u8 *)encap - skb->data);
+	skb_reset_network_header(skb);
+	skb->dev = reg_dev;
+	skb->protocol = htons(ETH_P_IP);
+	skb->ip_summed = 0;
+	skb->pkt_type = PACKET_HOST;
+	dst_release(skb->dst);
+	((struct net_device_stats *)netdev_priv(reg_dev))->rx_bytes += skb->len;
+	((struct net_device_stats *)netdev_priv(reg_dev))->rx_packets++;
+	skb->dst = NULL;
+	nf_reset(skb);
+	netif_rx(skb);
+	dev_put(reg_dev);
+	return 0;
+ drop:
+	kfree_skb(skb);
+	return 0;
+}
+
+static struct inet6_protocol pim6_protocol = {
+	.handler	=	pim6_rcv,
+};
+
+/* Service routines creating virtual interfaces: PIMREG */
+
+static int reg_vif_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	read_lock(&mrt_lock);
+	((struct net_device_stats *)netdev_priv(dev))->tx_bytes += skb->len;
+	((struct net_device_stats *)netdev_priv(dev))->tx_packets++;
+	ip6mr_cache_report(skb, reg_vif_num, MRT6MSG_WHOLEPKT);
+	read_unlock(&mrt_lock);
+	kfree_skb(skb);
+	return 0;
+}
+
+static struct net_device_stats *reg_vif_get_stats(struct net_device *dev)
+{
+	return (struct net_device_stats *)netdev_priv(dev);
+}
+
+static void reg_vif_setup(struct net_device *dev)
+{
+	dev->type		= ARPHRD_PIMREG;
+	dev->mtu		= 1500 - sizeof(struct ipv6hdr) - 8;
+	dev->flags		= IFF_NOARP;
+	dev->hard_start_xmit	= reg_vif_xmit;
+	dev->get_stats		= reg_vif_get_stats;
+	dev->destructor		= free_netdev;
+}
+
+static struct net_device *ip6mr_reg_vif(void)
+{
+	struct net_device *dev;
+	struct inet6_dev *in_dev;
+
+	dev = alloc_netdev(sizeof(struct net_device_stats), "pim6reg",
+			   reg_vif_setup);
+
+	if (dev == NULL)
+		return NULL;
+
+	if (register_netdevice(dev)) {
+		free_netdev(dev);
+		return NULL;
+	}
+	dev->iflink = 0;
+
+	in_dev = ipv6_find_idev(dev);
+	if (!in_dev)
+		goto failure;
+
+	if (dev_open(dev))
+		goto failure;
+
+	return dev;
+
+failure:
+	/* allow the register to be completed before unregistering. */
+	rtnl_unlock();
+	rtnl_lock();
+
+	unregister_netdevice(dev);
+	return NULL;
+}
+#endif
+
 /*
  *	Delete a VIF entry
  */
@@ -361,6 +499,11 @@ static int mif6_delete(int vifi)
 		return -EADDRNOTAVAIL;
 	}
 
+#ifdef CONFIG_IPV6_PIMSM_V2
+	if (vifi == reg_vif_num)
+		reg_vif_num = -1;
+#endif
+
 	if (vifi + 1 == maxvif) {
 		int tmp;
 		for (tmp = vifi - 1; tmp >= 0; tmp--) {
@@ -480,6 +623,19 @@ static int mif6_add(struct mif6ctl *vifc, int mrtsock)
 		return -EADDRINUSE;
 
 	switch (vifc->mif6c_flags) {
+#ifdef CONFIG_IPV6_PIMSM_V2
+	case MIFF_REGISTER:
+		/*
+		 * Special Purpose VIF in PIM
+		 * All the packets will be sent to the daemon
+		 */
+		if (reg_vif_num >= 0)
+			return -EADDRINUSE;
+		dev = ip6mr_reg_vif();
+		if (!dev)
+			return -ENOBUFS;
+		break;
+#endif
 	case 0:
 		dev = dev_get_by_index(&init_net, vifc->mif6c_pifi);
 		if (!dev)
@@ -512,6 +668,10 @@ static int mif6_add(struct mif6ctl *vifc, int mrtsock)
 	write_lock_bh(&mrt_lock);
 	dev_hold(dev);
 	v->dev = dev;
+#ifdef CONFIG_IPV6_PIMSM_V2
+	if (v->flags & MIFF_REGISTER)
+		reg_vif_num = vifi;
+#endif
 	if (vifi + 1 > maxvif)
 		maxvif = vifi + 1;
 	write_unlock_bh(&mrt_lock);
@@ -599,7 +759,13 @@ static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert)
 	struct mrt6msg *msg;
 	int ret;
 
-	skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC);
+#ifdef CONFIG_IPV6_PIMSM_V2
+	if (assert == MRT6MSG_WHOLEPKT)
+		skb = skb_realloc_headroom(pkt, -skb_network_offset(pkt)
+						+sizeof(*msg));
+	else
+#endif
+		skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC);
 
 	if (!skb)
 		return -ENOBUFS;
@@ -609,6 +775,29 @@ static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert)
 
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 
+#ifdef CONFIG_IPV6_PIMSM_V2
+	if (assert == MRT6MSG_WHOLEPKT) {
+		/* Ugly, but we have no choice with this interface.
+		   Duplicate old header, fix length etc.
+		   And all this only to mangle msg->im6_msgtype and
+		   to set msg->im6_mbz to "mbz" :-)
+		 */
+		skb_push(skb, -skb_network_offset(pkt));
+
+		skb_push(skb, sizeof(*msg));
+		skb_reset_transport_header(skb);
+		msg = (struct mrt6msg *)skb_transport_header(skb);
+		msg->im6_mbz = 0;
+		msg->im6_msgtype = MRT6MSG_WHOLEPKT;
+		msg->im6_mif = reg_vif_num;
+		msg->im6_pad = 0;
+		ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr);
+		ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr);
+
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	} else
+#endif
+	{
 	/*
 	 *	Copy the IP header
 	 */
@@ -635,6 +824,7 @@ static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert)
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 
 	skb_pull(skb, sizeof(struct ipv6hdr));
+	}
 
 	if (mroute6_socket == NULL) {
 		kfree_skb(skb);
@@ -1033,6 +1223,44 @@ int ip6_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int
 		rtnl_unlock();
 		return ret;
 
+	/*
+	 *	Control PIM assert (to activate pim will activate assert)
+	 */
+	case MRT6_ASSERT:
+	{
+		int v;
+		if (get_user(v, (int __user *)optval))
+			return -EFAULT;
+		mroute_do_assert = !!v;
+		return 0;
+	}
+
+#ifdef CONFIG_IPV6_PIMSM_V2
+	case MRT6_PIM:
+	{
+		int v, ret;
+		if (get_user(v, (int __user *)optval))
+			return -EFAULT;
+		v = !!v;
+		rtnl_lock();
+		ret = 0;
+		if (v != mroute_do_pim) {
+			mroute_do_pim = v;
+			mroute_do_assert = v;
+			if (mroute_do_pim)
+				ret = inet6_add_protocol(&pim6_protocol,
+							 IPPROTO_PIM);
+			else
+				ret = inet6_del_protocol(&pim6_protocol,
+							 IPPROTO_PIM);
+			if (ret < 0)
+				ret = -EAGAIN;
+		}
+		rtnl_unlock();
+		return ret;
+	}
+
+#endif
 	/*
 	 *	Spurious command, or MRT_VERSION which you cannot
 	 *	set.
@@ -1056,6 +1284,14 @@ int ip6_mroute_getsockopt(struct sock *sk, int optname, char __user *optval,
 	case MRT6_VERSION:
 		val = 0x0305;
 		break;
+#ifdef CONFIG_IPV6_PIMSM_V2
+	case MRT6_PIM:
+		val = mroute_do_pim;
+		break;
+#endif
+	case MRT6_ASSERT:
+		val = mroute_do_assert;
+		break;
 	default:
 		return -ENOPROTOOPT;
 	}
@@ -1151,6 +1387,18 @@ static int ip6mr_forward2(struct sk_buff *skb, struct mfc6_cache *c, int vifi)
 	if (vif->dev == NULL)
 		goto out_free;
 
+#ifdef CONFIG_IPV6_PIMSM_V2
+	if (vif->flags & MIFF_REGISTER) {
+		vif->pkt_out++;
+		vif->bytes_out += skb->len;
+		((struct net_device_stats *)netdev_priv(vif->dev))->tx_bytes += skb->len;
+		((struct net_device_stats *)netdev_priv(vif->dev))->tx_packets++;
+		ip6mr_cache_report(skb, vifi, MRT6MSG_WHOLEPKT);
+		kfree_skb(skb);
+		return 0;
+	}
+#endif
+
 	ipv6h = ipv6_hdr(skb);
 
 	fl = (struct flowi) {
@@ -1220,6 +1468,30 @@ static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache)
 	cache->mfc_un.res.pkt++;
 	cache->mfc_un.res.bytes += skb->len;
 
+	/*
+	 * Wrong interface: drop packet and (maybe) send PIM assert.
+	 */
+	if (vif6_table[vif].dev != skb->dev) {
+		int true_vifi;
+
+		cache->mfc_un.res.wrong_if++;
+		true_vifi = ip6mr_find_vif(skb->dev);
+
+		if (true_vifi >= 0 && mroute_do_assert &&
+		    /* pimsm uses asserts, when switching from RPT to SPT,
+		       so that we cannot check that packet arrived on an oif.
+		       It is bad, but otherwise we would need to move pretty
+		       large chunk of pimd to kernel. Ough... --ANK
+		     */
+		    (mroute_do_pim || cache->mfc_un.res.ttls[true_vifi] < 255) &&
+		    time_after(jiffies,
+			       cache->mfc_un.res.last_assert + MFC_ASSERT_THRESH)) {
+			cache->mfc_un.res.last_assert = jiffies;
+			ip6mr_cache_report(skb, true_vifi, MRT6MSG_WRONGMIF);
+		}
+		goto dont_forward;
+	}
+
 	vif6_table[vif].pkt_in++;
 	vif6_table[vif].bytes_in += skb->len;
 
@@ -1241,6 +1513,7 @@ static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache)
 		return 0;
 	}
 
+dont_forward:
 	kfree_skb(skb);
 	return 0;
 }
-- 
cgit v1.2.3


From 12802d058a003048104fe405a8d283b94ac50801 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 3 Apr 2008 09:22:56 +0900
Subject: [IPV6]: Comment MRT6_xxx sockopts in include/linux/in6.h.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/in6.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index f674000c6c99..e6aa8de2b939 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -260,4 +260,19 @@ struct in6_flowlabel_req
 #define IPV6_PREFER_SRC_CGA		0x0008
 #define IPV6_PREFER_SRC_NONCGA		0x0800
 
+/*
+ * Multicast Routing:
+ * see include/linux/mroute6.h.
+ *
+ * MRT6_INIT			200
+ * MRT6_DONE			201
+ * MRT6_ADD_MIF			202
+ * MRT6_DEL_MIF			203
+ * MRT6_ADD_MFC			204
+ * MRT6_DEL_MFC			205
+ * MRT6_VERSION			206
+ * MRT6_ASSERT			207
+ * MRT6_PIM			208
+ * (reserved)			209
+ */
 #endif
-- 
cgit v1.2.3


From 45e6cdf41437c72ed79cee64dc69e7f740511e50 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <djwong@us.ibm.com>
Date: Tue, 19 Feb 2008 10:49:40 -0800
Subject: [SCSI] libsas: Provide a transport-level facility to request SAS
 addrs

Provide a facility to use the request_firmware() interface to get a SAS
address from userspace.  This can be used by SAS LLDDs that cannot
obtain the address from the host adapter.

Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libsas/sas_scsi_host.c | 41 +++++++++++++++++++++++++++++++++++++
 include/scsi/libsas.h               |  2 ++
 2 files changed, 43 insertions(+)

(limited to 'include')

diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
index 1f8241563c6c..601ec5b6a7f6 100644
--- a/drivers/scsi/libsas/sas_scsi_host.c
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -24,6 +24,8 @@
  */
 
 #include <linux/kthread.h>
+#include <linux/firmware.h>
+#include <linux/ctype.h>
 
 #include "sas_internal.h"
 
@@ -1064,6 +1066,45 @@ void sas_target_destroy(struct scsi_target *starget)
 	return;
 }
 
+static void sas_parse_addr(u8 *sas_addr, const char *p)
+{
+	int i;
+	for (i = 0; i < SAS_ADDR_SIZE; i++) {
+		u8 h, l;
+		if (!*p)
+			break;
+		h = isdigit(*p) ? *p-'0' : toupper(*p)-'A'+10;
+		p++;
+		l = isdigit(*p) ? *p-'0' : toupper(*p)-'A'+10;
+		p++;
+		sas_addr[i] = (h<<4) | l;
+	}
+}
+
+#define SAS_STRING_ADDR_SIZE	16
+
+int sas_request_addr(struct Scsi_Host *shost, u8 *addr)
+{
+	int res;
+	const struct firmware *fw;
+
+	res = request_firmware(&fw, "sas_addr", &shost->shost_gendev);
+	if (res)
+		return res;
+
+	if (fw->size < SAS_STRING_ADDR_SIZE) {
+		res = -ENODEV;
+		goto out;
+	}
+
+	sas_parse_addr(addr, fw->data);
+
+out:
+	release_firmware(fw);
+	return res;
+}
+EXPORT_SYMBOL_GPL(sas_request_addr);
+
 EXPORT_SYMBOL_GPL(sas_queuecommand);
 EXPORT_SYMBOL_GPL(sas_target_alloc);
 EXPORT_SYMBOL_GPL(sas_slave_configure);
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index 39e1cac24bb7..98724ba65a79 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -677,4 +677,6 @@ extern void sas_ssp_task_response(struct device *dev, struct sas_task *task,
 				  struct ssp_response_iu *iu);
 struct sas_phy *sas_find_local_phy(struct domain_device *dev);
 
+int sas_request_addr(struct Scsi_Host *shost, u8 *addr);
+
 #endif /* _SASLIB_H_ */
-- 
cgit v1.2.3


From 40f6b36c6243462fb95d0343237331c423494b03 Mon Sep 17 00:00:00 2001
From: Kai Makisara <Kai.Makisara@kolumbus.fi>
Date: Sun, 24 Feb 2008 22:23:24 +0200
Subject: [SCSI] st: add option to use SILI in variable block reads

Add new option MT_ST_SILI to enable setting the SILI bit in reads in variable
block mode. If SILI is set, reading a block shorter than the byte count does
not result in CHECK CONDITION. The length of the block is determined using the
residual count from the HBA. Avoiding the REQUEST SENSE command for every
block speeds up some real applications considerably.

Signed-off-by: Kai Makisara <kai.makisara@kolumbus.fi>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 Documentation/scsi/st.txt |  7 ++++++-
 drivers/scsi/st.c         | 40 ++++++++++++++++++++++++++++++++++++----
 drivers/scsi/st.h         |  3 +++
 drivers/scsi/st_options.h |  6 +++++-
 include/linux/mtio.h      |  1 +
 5 files changed, 51 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/Documentation/scsi/st.txt b/Documentation/scsi/st.txt
index b7be95b5bd24..38f81188def0 100644
--- a/Documentation/scsi/st.txt
+++ b/Documentation/scsi/st.txt
@@ -2,7 +2,7 @@ This file contains brief information about the SCSI tape driver.
 The driver is currently maintained by Kai Mäkisara (email
 Kai.Makisara@kolumbus.fi)
 
-Last modified: Mon Mar  7 21:14:44 2005 by kai.makisara
+Last modified: Thu Feb 21 21:54:16 2008 by kai.makisara
 
 
 BASICS
@@ -372,6 +372,11 @@ MTSETDRVBUFFER
 	     MT_ST_SYSV sets the SYSV semantics (mode)
 	     MT_ST_NOWAIT enables immediate mode (i.e., don't wait for
 	        the command to finish) for some commands (e.g., rewind)
+	     MT_ST_SILI enables setting the SILI bit in SCSI commands when
+		reading in variable block mode to enhance performance when
+		reading blocks shorter than the byte count; set this only
+		if you are sure that the drive supports SILI and the HBA
+		correctly returns transfer residuals
 	     MT_ST_DEBUGGING debugging (global; debugging must be
 		compiled into the driver)
 	MT_ST_SETBOOLEANS
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index 0a52d9d2da2c..a4361a8c6ac6 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -17,7 +17,7 @@
    Last modified: 18-JAN-1998 Richard Gooch <rgooch@atnf.csiro.au> Devfs support
  */
 
-static const char *verstr = "20080221";
+static const char *verstr = "20080224";
 
 #include <linux/module.h>
 
@@ -183,6 +183,7 @@ static int modes_defined;
 
 static struct st_buffer *new_tape_buffer(int, int, int);
 static int enlarge_buffer(struct st_buffer *, int, int);
+static void clear_buffer(struct st_buffer *);
 static void normalize_buffer(struct st_buffer *);
 static int append_to_buffer(const char __user *, struct st_buffer *, int);
 static int from_buffer(struct st_buffer *, char __user *, int);
@@ -442,6 +443,7 @@ static void st_sleep_done(void *data, char *sense, int result, int resid)
 
 	memcpy(SRpnt->sense, sense, SCSI_SENSE_BUFFERSIZE);
 	(STp->buffer)->cmdstat.midlevel_result = SRpnt->result = result;
+	(STp->buffer)->cmdstat.residual = resid;
 	DEB( STp->write_pending = 0; )
 
 	if (SRpnt->waiting)
@@ -1159,6 +1161,7 @@ static int st_open(struct inode *inode, struct file *filp)
 		goto err_out;
 	}
 
+	(STp->buffer)->cleared = 0;
 	(STp->buffer)->writing = 0;
 	(STp->buffer)->syscall_result = 0;
 
@@ -1432,8 +1435,14 @@ static int setup_buffering(struct scsi_tape *STp, const char __user *buf,
 		if (STp->block_size)
 			bufsize = STp->block_size > st_fixed_buffer_size ?
 				STp->block_size : st_fixed_buffer_size;
-		else
+		else {
 			bufsize = count;
+			/* Make sure that data from previous user is not leaked even if
+			   HBA does not return correct residual */
+			if (is_read && STp->sili && !STbp->cleared)
+				clear_buffer(STbp);
+		}
+
 		if (bufsize > STbp->buffer_size &&
 		    !enlarge_buffer(STbp, bufsize, STp->restr_dma)) {
 			printk(KERN_WARNING "%s: Can't allocate %d byte tape buffer.\n",
@@ -1783,6 +1792,8 @@ static long read_tape(struct scsi_tape *STp, long count,
 	memset(cmd, 0, MAX_COMMAND_SIZE);
 	cmd[0] = READ_6;
 	cmd[1] = (STp->block_size != 0);
+	if (!cmd[1] && STp->sili)
+		cmd[1] |= 2;
 	cmd[2] = blks >> 16;
 	cmd[3] = blks >> 8;
 	cmd[4] = blks;
@@ -1911,8 +1922,11 @@ static long read_tape(struct scsi_tape *STp, long count,
 
 	}
 	/* End of error handling */ 
-	else			/* Read successful */
+	else {			/* Read successful */
 		STbp->buffer_bytes = bytes;
+		if (STp->sili) /* In fixed block mode residual is always zero here */
+			STbp->buffer_bytes -= STp->buffer->cmdstat.residual;
+	}
 
 	if (STps->drv_block >= 0) {
 		if (STp->block_size == 0)
@@ -2090,7 +2104,8 @@ static void st_log_options(struct scsi_tape * STp, struct st_modedef * STm, char
 		       name, STm->defaults_for_writes, STp->omit_blklims, STp->can_partitions,
 		       STp->scsi2_logical);
 		printk(KERN_INFO
-		       "%s:    sysv: %d nowait: %d\n", name, STm->sysv, STp->immediate);
+		       "%s:    sysv: %d nowait: %d sili: %d\n", name, STm->sysv, STp->immediate,
+			STp->sili);
 		printk(KERN_INFO "%s:    debugging: %d\n",
 		       name, debugging);
 	}
@@ -2133,6 +2148,7 @@ static int st_set_options(struct scsi_tape *STp, long options)
 		STp->scsi2_logical = (options & MT_ST_SCSI2LOGICAL) != 0;
 		STp->immediate = (options & MT_ST_NOWAIT) != 0;
 		STm->sysv = (options & MT_ST_SYSV) != 0;
+		STp->sili = (options & MT_ST_SILI) != 0;
 		DEB( debugging = (options & MT_ST_DEBUGGING) != 0;
 		     st_log_options(STp, STm, name); )
 	} else if (code == MT_ST_SETBOOLEANS || code == MT_ST_CLEARBOOLEANS) {
@@ -2164,6 +2180,8 @@ static int st_set_options(struct scsi_tape *STp, long options)
 			STp->immediate = value;
 		if ((options & MT_ST_SYSV) != 0)
 			STm->sysv = value;
+		if ((options & MT_ST_SILI) != 0)
+			STp->sili = value;
                 DEB(
 		if ((options & MT_ST_DEBUGGING) != 0)
 			debugging = value;
@@ -3655,6 +3673,8 @@ static int enlarge_buffer(struct st_buffer * STbuffer, int new_size, int need_dm
 		STbuffer->frp_segs += 1;
 		got += b_size;
 		STbuffer->buffer_size = got;
+		if (STbuffer->cleared)
+			memset(page_address(STbuffer->frp[segs].page), 0, b_size);
 		segs++;
 	}
 	STbuffer->b_data = page_address(STbuffer->frp[0].page);
@@ -3663,6 +3683,17 @@ static int enlarge_buffer(struct st_buffer * STbuffer, int new_size, int need_dm
 }
 
 
+/* Make sure that no data from previous user is in the internal buffer */
+static void clear_buffer(struct st_buffer * st_bp)
+{
+	int i;
+
+	for (i=0; i < st_bp->frp_segs; i++)
+		memset(page_address(st_bp->frp[i].page), 0, st_bp->frp[i].length);
+	st_bp->cleared = 1;
+}
+
+
 /* Release the extra buffer */
 static void normalize_buffer(struct st_buffer * STbuffer)
 {
@@ -3987,6 +4018,7 @@ static int st_probe(struct device *dev)
 	tpnt->two_fm = ST_TWO_FM;
 	tpnt->fast_mteom = ST_FAST_MTEOM;
 	tpnt->scsi2_logical = ST_SCSI2LOGICAL;
+	tpnt->sili = ST_SILI;
 	tpnt->immediate = ST_NOWAIT;
 	tpnt->default_drvbuffer = 0xff;		/* No forced buffering */
 	tpnt->partition = 0;
diff --git a/drivers/scsi/st.h b/drivers/scsi/st.h
index 5931726fcf93..b92712f95931 100644
--- a/drivers/scsi/st.h
+++ b/drivers/scsi/st.h
@@ -12,6 +12,7 @@ struct st_cmdstatus {
 	int midlevel_result;
 	struct scsi_sense_hdr sense_hdr;
 	int have_sense;
+	int residual;
 	u64 uremainder64;
 	u8 flags;
 	u8 remainder_valid;
@@ -34,6 +35,7 @@ struct st_request {
 struct st_buffer {
 	unsigned char dma;	/* DMA-able buffer */
 	unsigned char do_dio;   /* direct i/o set up? */
+	unsigned char cleared;  /* internal buffer cleared after open? */
 	int buffer_size;
 	int buffer_blocks;
 	int buffer_bytes;
@@ -122,6 +124,7 @@ struct scsi_tape {
 	unsigned char try_dio_now;		/* try direct i/o before next close? */
 	unsigned char c_algo;			/* compression algorithm */
 	unsigned char pos_unknown;			/* after reset position unknown */
+	unsigned char sili;			/* use SILI when reading in variable b mode */
 	int tape_type;
 	int long_timeout;	/* timeout for commands known to take long time */
 
diff --git a/drivers/scsi/st_options.h b/drivers/scsi/st_options.h
index b6b5c9c37677..d2f947935554 100644
--- a/drivers/scsi/st_options.h
+++ b/drivers/scsi/st_options.h
@@ -3,7 +3,7 @@
 
    Copyright 1995-2003 Kai Makisara.
 
-   Last modified: Mon Apr  7 22:49:18 2003 by makisara
+   Last modified: Thu Feb 21 21:47:07 2008 by kai.makisara
 */
 
 #ifndef _ST_OPTIONS_H
@@ -94,6 +94,10 @@
    The default is BSD semantics. */
 #define ST_SYSV 0
 
+/* If ST_SILI is non-zero, the SILI bit is set when reading in variable block
+   mode and the block size is determined using the residual returned by the HBA. */
+#define ST_SILI 0
+
 /* Time to wait for the drive to become ready if blocking open */
 #define ST_BLOCK_SECONDS     120
 
diff --git a/include/linux/mtio.h b/include/linux/mtio.h
index 6f8d2d45a8fb..ef01d6aa5934 100644
--- a/include/linux/mtio.h
+++ b/include/linux/mtio.h
@@ -192,6 +192,7 @@ struct	mtpos {
 #define MT_ST_SCSI2LOGICAL      0x800
 #define MT_ST_SYSV              0x1000
 #define MT_ST_NOWAIT            0x2000
+#define MT_ST_SILI		0x4000
 
 /* The mode parameters to be controlled. Parameter chosen with bits 20-28 */
 #define MT_ST_CLEAR_DEFAULT	0xfffff
-- 
cgit v1.2.3


From 30bd7df8ced23eefec87a5cda96dc99b002ed9da Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Fri, 29 Feb 2008 18:25:19 -0600
Subject: [SCSI] scsi_error: add target reset handler

The problem is that serveral drivers are sending a target reset from the
device reset handler, and if we have multiple devices a target reset gets
sent for each device when only one would be sufficient. And if we do a target
reset it affects all the commands on the target so the device reset handler
code only cleaning up one devices's commands makes programming the driver a
little more difficult than it should be.

This patch adds a target reset handler, which drivers can use to send
a target reset. If successful it cleans up the commands for a devices
accessed through that starget.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_error.c | 122 +++++++++++++++++++++++++++++++++++++++-------
 include/scsi/scsi_eh.h    |   1 +
 include/scsi/scsi_host.h  |   1 +
 3 files changed, 106 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 045a0868fc7b..1221d2ca0c64 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -524,6 +524,41 @@ static int scsi_try_bus_reset(struct scsi_cmnd *scmd)
 	return rtn;
 }
 
+static void __scsi_report_device_reset(struct scsi_device *sdev, void *data)
+{
+	sdev->was_reset = 1;
+	sdev->expecting_cc_ua = 1;
+}
+
+/**
+ * scsi_try_target_reset - Ask host to perform a target reset
+ * @scmd:	SCSI cmd used to send a target reset
+ *
+ * Notes:
+ *    There is no timeout for this operation.  if this operation is
+ *    unreliable for a given host, then the host itself needs to put a
+ *    timer on it, and set the host back to a consistent state prior to
+ *    returning.
+ */
+static int scsi_try_target_reset(struct scsi_cmnd *scmd)
+{
+	unsigned long flags;
+	int rtn;
+
+	if (!scmd->device->host->hostt->eh_target_reset_handler)
+		return FAILED;
+
+	rtn = scmd->device->host->hostt->eh_target_reset_handler(scmd);
+	if (rtn == SUCCESS) {
+		spin_lock_irqsave(scmd->device->host->host_lock, flags);
+		__starget_for_each_device(scsi_target(scmd->device), NULL,
+					  __scsi_report_device_reset);
+		spin_unlock_irqrestore(scmd->device->host->host_lock, flags);
+	}
+
+	return rtn;
+}
+
 /**
  * scsi_try_bus_device_reset - Ask host to perform a BDR on a dev
  * @scmd:	SCSI cmd used to send BDR
@@ -542,11 +577,8 @@ static int scsi_try_bus_device_reset(struct scsi_cmnd *scmd)
 		return FAILED;
 
 	rtn = scmd->device->host->hostt->eh_device_reset_handler(scmd);
-	if (rtn == SUCCESS) {
-		scmd->device->was_reset = 1;
-		scmd->device->expecting_cc_ua = 1;
-	}
-
+	if (rtn == SUCCESS)
+		__scsi_report_device_reset(scmd->device, NULL);
 	return rtn;
 }
 
@@ -584,8 +616,9 @@ static void scsi_abort_eh_cmnd(struct scsi_cmnd *scmd)
 {
 	if (__scsi_try_to_abort_cmd(scmd) != SUCCESS)
 		if (scsi_try_bus_device_reset(scmd) != SUCCESS)
-			if (scsi_try_bus_reset(scmd) != SUCCESS)
-				scsi_try_host_reset(scmd);
+			if (scsi_try_target_reset(scmd) != SUCCESS)
+				if (scsi_try_bus_reset(scmd) != SUCCESS)
+					scsi_try_host_reset(scmd);
 }
 
 /**
@@ -1059,6 +1092,56 @@ static int scsi_eh_bus_device_reset(struct Scsi_Host *shost,
 	return list_empty(work_q);
 }
 
+/**
+ * scsi_eh_target_reset - send target reset if needed
+ * @shost:	scsi host being recovered.
+ * @work_q:     &list_head for pending commands.
+ * @done_q:	&list_head for processed commands.
+ *
+ * Notes:
+ *    Try a target reset.
+ */
+static int scsi_eh_target_reset(struct Scsi_Host *shost,
+				struct list_head *work_q,
+				struct list_head *done_q)
+{
+	struct scsi_cmnd *scmd, *tgtr_scmd, *next;
+	unsigned int id;
+	int rtn;
+
+	for (id = 0; id <= shost->max_id; id++) {
+		tgtr_scmd = NULL;
+		list_for_each_entry(scmd, work_q, eh_entry) {
+			if (id == scmd_id(scmd)) {
+				tgtr_scmd = scmd;
+				break;
+			}
+		}
+		if (!tgtr_scmd)
+			continue;
+
+		SCSI_LOG_ERROR_RECOVERY(3, printk("%s: Sending target reset "
+						  "to target %d\n",
+						  current->comm, id));
+		rtn = scsi_try_target_reset(tgtr_scmd);
+		if (rtn == SUCCESS) {
+			list_for_each_entry_safe(scmd, next, work_q, eh_entry) {
+				if (id == scmd_id(scmd))
+					if (!scsi_device_online(scmd->device) ||
+					    !scsi_eh_tur(tgtr_scmd))
+						scsi_eh_finish_cmd(scmd,
+								   done_q);
+			}
+		} else
+			SCSI_LOG_ERROR_RECOVERY(3, printk("%s: Target reset"
+							  " failed target: "
+							  "%d\n",
+							  current->comm, id));
+	}
+
+	return list_empty(work_q);
+}
+
 /**
  * scsi_eh_bus_reset - send a bus reset 
  * @shost:	&scsi host being recovered.
@@ -1447,9 +1530,11 @@ void scsi_eh_ready_devs(struct Scsi_Host *shost,
 {
 	if (!scsi_eh_stu(shost, work_q, done_q))
 		if (!scsi_eh_bus_device_reset(shost, work_q, done_q))
-			if (!scsi_eh_bus_reset(shost, work_q, done_q))
-				if (!scsi_eh_host_reset(work_q, done_q))
-					scsi_eh_offline_sdevs(work_q, done_q);
+			if (!scsi_eh_target_reset(shost, work_q, done_q))
+				if (!scsi_eh_bus_reset(shost, work_q, done_q))
+					if (!scsi_eh_host_reset(work_q, done_q))
+						scsi_eh_offline_sdevs(work_q,
+								      done_q);
 }
 EXPORT_SYMBOL_GPL(scsi_eh_ready_devs);
 
@@ -1619,10 +1704,8 @@ void scsi_report_bus_reset(struct Scsi_Host *shost, int channel)
 	struct scsi_device *sdev;
 
 	__shost_for_each_device(sdev, shost) {
-		if (channel == sdev_channel(sdev)) {
-			sdev->was_reset = 1;
-			sdev->expecting_cc_ua = 1;
-		}
+		if (channel == sdev_channel(sdev))
+			__scsi_report_device_reset(sdev, NULL);
 	}
 }
 EXPORT_SYMBOL(scsi_report_bus_reset);
@@ -1655,10 +1738,8 @@ void scsi_report_device_reset(struct Scsi_Host *shost, int channel, int target)
 
 	__shost_for_each_device(sdev, shost) {
 		if (channel == sdev_channel(sdev) &&
-		    target == sdev_id(sdev)) {
-			sdev->was_reset = 1;
-			sdev->expecting_cc_ua = 1;
-		}
+		    target == sdev_id(sdev))
+			__scsi_report_device_reset(sdev, NULL);
 	}
 }
 EXPORT_SYMBOL(scsi_report_device_reset);
@@ -1714,6 +1795,11 @@ scsi_reset_provider(struct scsi_device *dev, int flag)
 		if (rtn == SUCCESS)
 			break;
 		/* FALLTHROUGH */
+	case SCSI_TRY_RESET_TARGET:
+		rtn = scsi_try_target_reset(scmd);
+		if (rtn == SUCCESS)
+			break;
+		/* FALLTHROUGH */
 	case SCSI_TRY_RESET_BUS:
 		rtn = scsi_try_bus_reset(scmd);
 		if (rtn == SUCCESS)
diff --git a/include/scsi/scsi_eh.h b/include/scsi/scsi_eh.h
index 25071d5d9bf8..37a7614f62f4 100644
--- a/include/scsi/scsi_eh.h
+++ b/include/scsi/scsi_eh.h
@@ -64,6 +64,7 @@ extern int scsi_get_sense_info_fld(const u8 * sense_buffer, int sb_len,
 #define SCSI_TRY_RESET_DEVICE	1
 #define SCSI_TRY_RESET_BUS	2
 #define SCSI_TRY_RESET_HOST	3
+#define SCSI_TRY_RESET_TARGET	4
 
 extern int scsi_reset_provider(struct scsi_device *, int);
 
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index 530ff4c553f8..49132862bfaa 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -172,6 +172,7 @@ struct scsi_host_template {
 	 */
 	int (* eh_abort_handler)(struct scsi_cmnd *);
 	int (* eh_device_reset_handler)(struct scsi_cmnd *);
+	int (* eh_target_reset_handler)(struct scsi_cmnd *);
 	int (* eh_bus_reset_handler)(struct scsi_cmnd *);
 	int (* eh_host_reset_handler)(struct scsi_cmnd *);
 
-- 
cgit v1.2.3


From b1adaf65ba0398c9a1adc8f3a274533165a4df61 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Tue, 18 Mar 2008 00:15:03 +0900
Subject: [SCSI] block: add sg buffer copy helper functions

This patch adds new three helper functions to copy data between an SG
list and a linear buffer.

- sg_copy_from_buffer copies data from linear buffer to an SG list

- sg_copy_to_buffer copies data from an SG list to a linear buffer

When the APIs copy data from a linear buffer to an SG list,
flush_kernel_dcache_page is called. It's not necessary for everyone
but it's a no-op on most architectures and in general the API is not
used in performance critical path.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 include/linux/scatterlist.h |   5 +++
 lib/scatterlist.c           | 102 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 107 insertions(+)

(limited to 'include')

diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h
index a3d567a974e8..71fc81360048 100644
--- a/include/linux/scatterlist.h
+++ b/include/linux/scatterlist.h
@@ -213,6 +213,11 @@ int __sg_alloc_table(struct sg_table *, unsigned int, unsigned int, gfp_t,
 		     sg_alloc_fn *);
 int sg_alloc_table(struct sg_table *, unsigned int, gfp_t);
 
+size_t sg_copy_from_buffer(struct scatterlist *sgl, unsigned int nents,
+			   void *buf, size_t buflen);
+size_t sg_copy_to_buffer(struct scatterlist *sgl, unsigned int nents,
+			 void *buf, size_t buflen);
+
 /*
  * Maximum number of entries that will be allocated in one piece, if
  * a list larger than this is required then chaining will be utilized.
diff --git a/lib/scatterlist.c b/lib/scatterlist.c
index acca4901046c..b80c21100d78 100644
--- a/lib/scatterlist.c
+++ b/lib/scatterlist.c
@@ -8,6 +8,7 @@
  */
 #include <linux/module.h>
 #include <linux/scatterlist.h>
+#include <linux/highmem.h>
 
 /**
  * sg_next - return the next scatterlist entry in a list
@@ -292,3 +293,104 @@ int sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask)
 	return ret;
 }
 EXPORT_SYMBOL(sg_alloc_table);
+
+/**
+ * sg_copy_buffer - Copy data between a linear buffer and an SG list
+ * @sgl:		 The SG list
+ * @nents:		 Number of SG entries
+ * @buf:		 Where to copy from
+ * @buflen:		 The number of bytes to copy
+ * @to_buffer: 		 transfer direction (non zero == from an sg list to a
+ * 			 buffer, 0 == from a buffer to an sg list
+ *
+ * Returns the number of copied bytes.
+ *
+ **/
+static size_t sg_copy_buffer(struct scatterlist *sgl, unsigned int nents,
+			     void *buf, size_t buflen, int to_buffer)
+{
+	struct scatterlist *sg;
+	size_t buf_off = 0;
+	int i;
+
+	WARN_ON(!irqs_disabled());
+
+	for_each_sg(sgl, sg, nents, i) {
+		struct page *page;
+		int n = 0;
+		unsigned int sg_off = sg->offset;
+		unsigned int sg_copy = sg->length;
+
+		if (sg_copy > buflen)
+			sg_copy = buflen;
+		buflen -= sg_copy;
+
+		while (sg_copy > 0) {
+			unsigned int page_copy;
+			void *p;
+
+			page_copy = PAGE_SIZE - sg_off;
+			if (page_copy > sg_copy)
+				page_copy = sg_copy;
+
+			page = nth_page(sg_page(sg), n);
+			p = kmap_atomic(page, KM_BIO_SRC_IRQ);
+
+			if (to_buffer)
+				memcpy(buf + buf_off, p + sg_off, page_copy);
+			else {
+				memcpy(p + sg_off, buf + buf_off, page_copy);
+				flush_kernel_dcache_page(page);
+			}
+
+			kunmap_atomic(p, KM_BIO_SRC_IRQ);
+
+			buf_off += page_copy;
+			sg_off += page_copy;
+			if (sg_off == PAGE_SIZE) {
+				sg_off = 0;
+				n++;
+			}
+			sg_copy -= page_copy;
+		}
+
+		if (!buflen)
+			break;
+	}
+
+	return buf_off;
+}
+
+/**
+ * sg_copy_from_buffer - Copy from a linear buffer to an SG list
+ * @sgl:		 The SG list
+ * @nents:		 Number of SG entries
+ * @buf:		 Where to copy from
+ * @buflen:		 The number of bytes to copy
+ *
+ * Returns the number of copied bytes.
+ *
+ **/
+size_t sg_copy_from_buffer(struct scatterlist *sgl, unsigned int nents,
+			   void *buf, size_t buflen)
+{
+	return sg_copy_buffer(sgl, nents, buf, buflen, 0);
+}
+EXPORT_SYMBOL(sg_copy_from_buffer);
+
+/**
+ * sg_copy_to_buffer - Copy from an SG list to a linear buffer
+ * @sgl:		 The SG list
+ * @nents:		 Number of SG entries
+ * @buf:		 Where to copy to
+ * @buflen:		 The number of bytes to copy
+ *
+ * Returns the number of copied bytes.
+ *
+ **/
+size_t sg_copy_to_buffer(struct scatterlist *sgl, unsigned int nents,
+			 void *buf, size_t buflen)
+{
+	return sg_copy_buffer(sgl, nents, buf, buflen, 1);
+}
+EXPORT_SYMBOL(sg_copy_to_buffer);
-- 
cgit v1.2.3


From 9ac16b616ab117dab3fce9790368d3b58ca441ef Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Sun, 9 Mar 2008 13:44:29 +0900
Subject: [SCSI] scsi: add wrapper functions for sg buffer copy helper
 functions

LLDs need to copies data between the SG table in struct scsi_cmnd and
liner buffer. So they use the helper functions like

sg_copy_from_buffer(scsi_sglist(sc), scsi_sg_count(sc), buf, buflen)
sg_copy_to_buffer(scsi_sglist(sc), scsi_sg_count(sc), buf, buflen)

This patch just adds wrapper functions:

scsi_sg_copy_from_buffer(sc, buf, buflen)
scsi_sg_copy_to_buffer(sc, buf, buflen)

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 include/scsi/scsi_cmnd.h | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

(limited to 'include')

diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index de28aab820b0..b260be6d0d73 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -175,4 +175,18 @@ static inline struct scsi_data_buffer *scsi_out(struct scsi_cmnd *cmd)
 	return &cmd->sdb;
 }
 
+static inline int scsi_sg_copy_from_buffer(struct scsi_cmnd *cmd,
+					   void *buf, int buflen)
+{
+	return sg_copy_from_buffer(scsi_sglist(cmd), scsi_sg_count(cmd),
+				   buf, buflen);
+}
+
+static inline int scsi_sg_copy_to_buffer(struct scsi_cmnd *cmd,
+					 void *buf, int buflen)
+{
+	return sg_copy_to_buffer(scsi_sglist(cmd), scsi_sg_count(cmd),
+				 buf, buflen);
+}
+
 #endif /* _SCSI_SCSI_CMND_H */
-- 
cgit v1.2.3


From 1c353f7d616a4ef04b5e73fe7a2184baa039f06f Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Thu, 13 Mar 2008 11:19:36 -0500
Subject: [SCSI] export command allocation and freeing functions independently
 of the host

This is needed by things like USB storage that want to set up static
commands for later use at start of day.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi.c      | 149 +++++++++++++++++++++++++++++++++++------------
 include/scsi/scsi_cmnd.h |   3 +
 2 files changed, 115 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 2cf9a625f227..f6980bd9d8f9 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -330,30 +330,16 @@ void scsi_put_command(struct scsi_cmnd *cmd)
 }
 EXPORT_SYMBOL(scsi_put_command);
 
-/**
- * scsi_setup_command_freelist - Setup the command freelist for a scsi host.
- * @shost: host to allocate the freelist for.
- *
- * Description: The command freelist protects against system-wide out of memory
- * deadlock by preallocating one SCSI command structure for each host, so the
- * system can always write to a swap file on a device associated with that host.
- *
- * Returns:	Nothing.
- */
-int scsi_setup_command_freelist(struct Scsi_Host *shost)
+static struct scsi_host_cmd_pool *scsi_get_host_cmd_pool(gfp_t gfp_mask)
 {
-	struct scsi_host_cmd_pool *pool;
-	struct scsi_cmnd *cmd;
-
-	spin_lock_init(&shost->free_list_lock);
-	INIT_LIST_HEAD(&shost->free_list);
-
+	struct scsi_host_cmd_pool *retval = NULL, *pool;
 	/*
 	 * Select a command slab for this host and create it if not
 	 * yet existent.
 	 */
 	mutex_lock(&host_cmd_pool_mutex);
-	pool = (shost->unchecked_isa_dma ? &scsi_cmd_dma_pool : &scsi_cmd_pool);
+	pool = (gfp_mask & __GFP_DMA) ? &scsi_cmd_dma_pool :
+		&scsi_cmd_pool;
 	if (!pool->users) {
 		pool->cmd_slab = kmem_cache_create(pool->cmd_name,
 						   sizeof(struct scsi_cmnd), 0,
@@ -371,28 +357,122 @@ int scsi_setup_command_freelist(struct Scsi_Host *shost)
 	}
 
 	pool->users++;
-	shost->cmd_pool = pool;
+	retval = pool;
+ fail:
 	mutex_unlock(&host_cmd_pool_mutex);
+	return retval;
+}
+
+static void scsi_put_host_cmd_pool(gfp_t gfp_mask)
+{
+	struct scsi_host_cmd_pool *pool;
 
+	mutex_lock(&host_cmd_pool_mutex);
+	pool = (gfp_mask & __GFP_DMA) ? &scsi_cmd_dma_pool :
+		&scsi_cmd_pool;
 	/*
-	 * Get one backup command for this host.
+	 * This may happen if a driver has a mismatched get and put
+	 * of the command pool; the driver should be implicated in
+	 * the stack trace
 	 */
-	cmd = scsi_pool_alloc_command(shost->cmd_pool, GFP_KERNEL);
-	if (!cmd)
-		goto fail2;
+	BUG_ON(pool->users == 0);
 
-	list_add(&cmd->list, &shost->free_list);
-	return 0;
-
- fail2:
-	mutex_lock(&host_cmd_pool_mutex);
 	if (!--pool->users) {
 		kmem_cache_destroy(pool->cmd_slab);
 		kmem_cache_destroy(pool->sense_slab);
 	}
- fail:
 	mutex_unlock(&host_cmd_pool_mutex);
-	return -ENOMEM;
+}
+
+/**
+ * scsi_allocate_command - get a fully allocated SCSI command
+ * @gfp_mask:	allocation mask
+ *
+ * This function is for use outside of the normal host based pools.
+ * It allocates the relevant command and takes an additional reference
+ * on the pool it used.  This function *must* be paired with
+ * scsi_free_command which also has the identical mask, otherwise the
+ * free pool counts will eventually go wrong and you'll trigger a bug.
+ *
+ * This function should *only* be used by drivers that need a static
+ * command allocation at start of day for internal functions.
+ */
+struct scsi_cmnd *scsi_allocate_command(gfp_t gfp_mask)
+{
+	struct scsi_host_cmd_pool *pool = scsi_get_host_cmd_pool(gfp_mask);
+
+	if (!pool)
+		return NULL;
+
+	return scsi_pool_alloc_command(pool, gfp_mask);
+}
+EXPORT_SYMBOL(scsi_allocate_command);
+
+/**
+ * scsi_free_command - free a command allocated by scsi_allocate_command
+ * @gfp_mask:	mask used in the original allocation
+ * @cmd:	command to free
+ *
+ * Note: using the original allocation mask is vital because that's
+ * what determines which command pool we use to free the command.  Any
+ * mismatch will cause the system to BUG eventually.
+ */
+void scsi_free_command(gfp_t gfp_mask, struct scsi_cmnd *cmd)
+{
+	struct scsi_host_cmd_pool *pool = scsi_get_host_cmd_pool(gfp_mask);
+
+	/*
+	 * this could trigger if the mask to scsi_allocate_command
+	 * doesn't match this mask.  Otherwise we're guaranteed that this
+	 * succeeds because scsi_allocate_command must have taken a reference
+	 * on the pool
+	 */
+	BUG_ON(!pool);
+
+	scsi_pool_free_command(pool, cmd);
+	/*
+	 * scsi_put_host_cmd_pool is called twice; once to release the
+	 * reference we took above, and once to release the reference
+	 * originally taken by scsi_allocate_command
+	 */
+	scsi_put_host_cmd_pool(gfp_mask);
+	scsi_put_host_cmd_pool(gfp_mask);
+}
+EXPORT_SYMBOL(scsi_free_command);
+
+/**
+ * scsi_setup_command_freelist - Setup the command freelist for a scsi host.
+ * @shost: host to allocate the freelist for.
+ *
+ * Description: The command freelist protects against system-wide out of memory
+ * deadlock by preallocating one SCSI command structure for each host, so the
+ * system can always write to a swap file on a device associated with that host.
+ *
+ * Returns:	Nothing.
+ */
+int scsi_setup_command_freelist(struct Scsi_Host *shost)
+{
+	struct scsi_cmnd *cmd;
+	const gfp_t gfp_mask = shost->unchecked_isa_dma ? GFP_DMA : GFP_KERNEL;
+
+	spin_lock_init(&shost->free_list_lock);
+	INIT_LIST_HEAD(&shost->free_list);
+
+	shost->cmd_pool = scsi_get_host_cmd_pool(gfp_mask);
+
+	if (!shost->cmd_pool)
+		return -ENOMEM;
+
+	/*
+	 * Get one backup command for this host.
+	 */
+	cmd = scsi_pool_alloc_command(shost->cmd_pool, gfp_mask);
+	if (!cmd) {
+		scsi_put_host_cmd_pool(gfp_mask);
+		return -ENOMEM;
+	}
+	list_add(&cmd->list, &shost->free_list);
+	return 0;
 }
 
 /**
@@ -408,13 +488,8 @@ void scsi_destroy_command_freelist(struct Scsi_Host *shost)
 		list_del_init(&cmd->list);
 		scsi_pool_free_command(shost->cmd_pool, cmd);
 	}
-
-	mutex_lock(&host_cmd_pool_mutex);
-	if (!--shost->cmd_pool->users) {
-		kmem_cache_destroy(shost->cmd_pool->cmd_slab);
-		kmem_cache_destroy(shost->cmd_pool->sense_slab);
-	}
-	mutex_unlock(&host_cmd_pool_mutex);
+	shost->cmd_pool = NULL;
+	scsi_put_host_cmd_pool(shost->unchecked_isa_dma ? GFP_DMA : GFP_KERNEL);
 }
 
 #ifdef CONFIG_SCSI_LOGGING
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index b260be6d0d73..8d20e60a94b7 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -130,6 +130,9 @@ extern void scsi_release_buffers(struct scsi_cmnd *cmd);
 extern int scsi_dma_map(struct scsi_cmnd *cmd);
 extern void scsi_dma_unmap(struct scsi_cmnd *cmd);
 
+struct scsi_cmnd *scsi_allocate_command(gfp_t gfp_mask);
+void scsi_free_command(gfp_t gfp_mask, struct scsi_cmnd *cmd);
+
 static inline unsigned scsi_sg_count(struct scsi_cmnd *cmd)
 {
 	return cmd->sdb.table.nents;
-- 
cgit v1.2.3


From 3bc6a26192d2548397a3e721d786cf8345ee54e1 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Tue, 25 Mar 2008 09:26:49 +0900
Subject: [SCSI] add scsi_build_sense_buffer helper function

This adds scsi_build_sense_buffer, a simple helper function to build
sense data in a buffer.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_error.c | 28 ++++++++++++++++++++++++++++
 include/scsi/scsi_eh.h    |  4 +++-
 2 files changed, 31 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 1221d2ca0c64..221f31e36d26 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1993,3 +1993,31 @@ int scsi_get_sense_info_fld(const u8 * sense_buffer, int sb_len,
 	}
 }
 EXPORT_SYMBOL(scsi_get_sense_info_fld);
+
+/**
+ * scsi_build_sense_buffer - build sense data in a buffer
+ * @desc:	Sense format (non zero == descriptor format,
+ * 		0 == fixed format)
+ * @buf:	Where to build sense data
+ * @key:	Sense key
+ * @asc:	Additional sense code
+ * @ascq:	Additional sense code qualifier
+ *
+ **/
+void scsi_build_sense_buffer(int desc, u8 *buf, u8 key, u8 asc, u8 ascq)
+{
+	if (desc) {
+		buf[0] = 0x72;	/* descriptor, current */
+		buf[1] = key;
+		buf[2] = asc;
+		buf[3] = ascq;
+		buf[7] = 0;
+	} else {
+		buf[0] = 0x70;	/* fixed, current */
+		buf[2] = key;
+		buf[7] = 0xa;
+		buf[12] = asc;
+		buf[13] = ascq;
+	}
+}
+EXPORT_SYMBOL(scsi_build_sense_buffer);
diff --git a/include/scsi/scsi_eh.h b/include/scsi/scsi_eh.h
index 37a7614f62f4..d3a133b4a072 100644
--- a/include/scsi/scsi_eh.h
+++ b/include/scsi/scsi_eh.h
@@ -57,7 +57,9 @@ extern const u8 * scsi_sense_desc_find(const u8 * sense_buffer, int sb_len,
 
 extern int scsi_get_sense_info_fld(const u8 * sense_buffer, int sb_len,
 				   u64 * info_out);
- 
+
+extern void scsi_build_sense_buffer(int desc, u8 *buf, u8 key, u8 asc, u8 ascq);
+
 /*
  * Reset request from external source
  */
-- 
cgit v1.2.3


From 2f3edc6936e3f6be3f1df1e89c141ae028fa605e Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Wed, 2 Apr 2008 10:05:48 -0500
Subject: [SCSI] transport_class: BUG if we can't release the attribute
 container

Every current transport class calls transport_container_release but
ignores the return value.  This is catastrophic if it returns an error
because the containers are part of a global list and the next action of
almost every transport class is to free the memory used by the
container.

Fix this by making transport_container_release a void, but making it BUG
if attribute_container_release returns an error ... this catches the
root cause of a system panic much earlier.  If we don't do this, we get
an eventual BUG when the attribute container list notices the corruption
caused by the freed memory it's still referencing.

Also made attribute_container_release __must_check as a reminder.

Cc: Greg KH <greg@kroah.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/base/transport_class.c      | 3 ++-
 drivers/scsi/raid_class.c           | 2 +-
 include/linux/attribute_container.h | 2 +-
 include/linux/transport_class.h     | 5 +++--
 4 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/base/transport_class.c b/drivers/base/transport_class.c
index 40bca48abc12..cabd0edf2156 100644
--- a/drivers/base/transport_class.c
+++ b/drivers/base/transport_class.c
@@ -108,7 +108,8 @@ EXPORT_SYMBOL_GPL(anon_transport_class_register);
  */
 void anon_transport_class_unregister(struct anon_transport_class *atc)
 {
-	attribute_container_unregister(&atc->container);
+	if (unlikely(attribute_container_unregister(&atc->container)))
+		BUG();
 }
 EXPORT_SYMBOL_GPL(anon_transport_class_unregister);
 
diff --git a/drivers/scsi/raid_class.c b/drivers/scsi/raid_class.c
index 86e13183c9ba..52182a744ba6 100644
--- a/drivers/scsi/raid_class.c
+++ b/drivers/scsi/raid_class.c
@@ -289,7 +289,7 @@ raid_class_release(struct raid_template *r)
 {
 	struct raid_internal *i = to_raid_internal(r);
 
-	attribute_container_unregister(&i->r.raid_attrs.ac);
+	BUG_ON(attribute_container_unregister(&i->r.raid_attrs.ac));
 
 	kfree(i);
 }
diff --git a/include/linux/attribute_container.h b/include/linux/attribute_container.h
index f5582332af04..574b201b99d8 100644
--- a/include/linux/attribute_container.h
+++ b/include/linux/attribute_container.h
@@ -37,7 +37,7 @@ attribute_container_set_no_classdevs(struct attribute_container *atc)
 }
 
 int attribute_container_register(struct attribute_container *cont);
-int attribute_container_unregister(struct attribute_container *cont);
+int __must_check attribute_container_unregister(struct attribute_container *cont);
 void attribute_container_create_device(struct device *dev,
 				       int (*fn)(struct attribute_container *,
 						 struct device *,
diff --git a/include/linux/transport_class.h b/include/linux/transport_class.h
index 1d6cc22e5f42..6696cf79c4f7 100644
--- a/include/linux/transport_class.h
+++ b/include/linux/transport_class.h
@@ -86,9 +86,10 @@ static inline int transport_container_register(struct transport_container *tc)
 	return attribute_container_register(&tc->ac);
 }
 
-static inline int transport_container_unregister(struct transport_container *tc)
+static inline void transport_container_unregister(struct transport_container *tc)
 {
-	return attribute_container_unregister(&tc->ac);
+	if (unlikely(attribute_container_unregister(&tc->ac)))
+		BUG();
 }
 
 int transport_class_register(struct transport_class *);
-- 
cgit v1.2.3


From 79bc14813cd7e1b75d2e4cbbc17043261cf4bcdc Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Thu, 3 Apr 2008 09:04:31 -0500
Subject: [SCSI] libsas: fix missing inlines in header file
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Two functions in include/scsi/sas_ata.h don't have static inlines
leading to problems if they're built in:

On Thu, 2008-04-03 at 14:06 +0200, Toralf Förster wrote:
> drivers/scsi/mvsas.o: In function `sas_ata_init_host_and_port':
> mvsas.c:(.text+0x0): multiple definition of `sas_ata_init_host_and_port'
> drivers/scsi/libsas/built-in.o:(.text+0x37f4): first defined here
> drivers/scsi/mvsas.o: In function `sas_ata_task_abort':
> mvsas.c:(.text+0x7): multiple definition of `sas_ata_task_abort'
> drivers/scsi/libsas/built-in.o:(.text+0x37fb): first defined here
> make[2]: *** [drivers/scsi/built-in.o] Error 1
> make[1]: *** [drivers/scsi] Error 2
> make: *** [drivers] Error 2

Add the correct static inline modifiers.

Tested-by: Toralf Förster <toralf.foerster@gmx.de>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 include/scsi/sas_ata.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/scsi/sas_ata.h b/include/scsi/sas_ata.h
index dd5edc915417..c583193ae929 100644
--- a/include/scsi/sas_ata.h
+++ b/include/scsi/sas_ata.h
@@ -47,12 +47,12 @@ static inline int dev_is_sata(struct domain_device *dev)
 {
 	return 0;
 }
-int sas_ata_init_host_and_port(struct domain_device *found_dev,
+static inline int sas_ata_init_host_and_port(struct domain_device *found_dev,
 			       struct scsi_target *starget)
 {
 	return 0;
 }
-void sas_ata_task_abort(struct sas_task *task)
+static inline void sas_ata_task_abort(struct sas_task *task)
 {
 }
 #endif
-- 
cgit v1.2.3


From 38668c059f5202f5fd9612391f9aa1b38a97241b Mon Sep 17 00:00:00 2001
From: Tomas Winkler <tomas.winkler@intel.com>
Date: Fri, 28 Mar 2008 16:33:32 -0700
Subject: mac80211: eliminate conf_ht

This patch eliminates the use of conf_ht, replacing it with
bss_info_changed.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h       | 10 +++++-
 net/mac80211/ieee80211.c     | 73 ++++++++++++++++++++++++++------------------
 net/mac80211/ieee80211_i.h   |  6 ++--
 net/mac80211/ieee80211_sta.c | 33 +++++++++++---------
 4 files changed, 75 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 48428a6b9109..5174eaa89500 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -178,11 +178,13 @@ struct ieee80211_low_level_stats {
  *	also implies a change in the AID.
  * @BSS_CHANGED_ERP_CTS_PROT: CTS protection changed
  * @BSS_CHANGED_ERP_PREAMBLE: preamble changed
+ * @BSS_CHANGED_HT: 802.11n parameters changed
  */
 enum ieee80211_bss_change {
 	BSS_CHANGED_ASSOC		= 1<<0,
 	BSS_CHANGED_ERP_CTS_PROT	= 1<<1,
 	BSS_CHANGED_ERP_PREAMBLE	= 1<<2,
+	BSS_CHANGED_HT                  = 1<<4,
 };
 
 /**
@@ -195,6 +197,9 @@ enum ieee80211_bss_change {
  * @aid: association ID number, valid only when @assoc is true
  * @use_cts_prot: use CTS protection
  * @use_short_preamble: use 802.11b short preamble
+ * @assoc_ht: association in HT mode
+ * @ht_conf: ht capabilities
+ * @ht_bss_conf: ht extended capabilities
  */
 struct ieee80211_bss_conf {
 	/* association related data */
@@ -203,6 +208,10 @@ struct ieee80211_bss_conf {
 	/* erp related data */
 	bool use_cts_prot;
 	bool use_short_preamble;
+	/* ht related data */
+	bool assoc_ht;
+	struct ieee80211_ht_info *ht_conf;
+	struct ieee80211_ht_bss_info *ht_bss_conf;
 };
 
 /**
@@ -1132,7 +1141,6 @@ struct ieee80211_ops {
 			     struct sk_buff *skb,
 			     struct ieee80211_tx_control *control);
 	int (*tx_last_beacon)(struct ieee80211_hw *hw);
-	int (*conf_ht)(struct ieee80211_hw *hw, struct ieee80211_conf *conf);
 	int (*ampdu_action)(struct ieee80211_hw *hw,
 			    enum ieee80211_ampdu_mlme_action action,
 			    const u8 *addr, u16 tid, u16 *ssn);
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index aaa5480e204a..5d30dd463f22 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -1046,54 +1046,69 @@ int ieee80211_hw_config(struct ieee80211_local *local)
 }
 
 /**
- * ieee80211_hw_config_ht should be used only after legacy configuration
- * has been determined, as ht configuration depends upon the hardware's
- * HT abilities for a _specific_ band.
+ * ieee80211_handle_ht should be used only after legacy configuration
+ * has been determined namely band, as ht configuration depends upon
+ * the hardware's HT abilities for a _specific_ band.
  */
-int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
+u32 ieee80211_handle_ht(struct ieee80211_local *local, int enable_ht,
 			   struct ieee80211_ht_info *req_ht_cap,
 			   struct ieee80211_ht_bss_info *req_bss_cap)
 {
 	struct ieee80211_conf *conf = &local->hw.conf;
 	struct ieee80211_supported_band *sband;
+	struct ieee80211_ht_info ht_conf;
+	struct ieee80211_ht_bss_info ht_bss_conf;
 	int i;
+	u32 changed = 0;
 
 	sband = local->hw.wiphy->bands[conf->channel->band];
 
 	/* HT is not supported */
 	if (!sband->ht_info.ht_supported) {
 		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
-		return -EOPNOTSUPP;
+		return 0;
 	}
 
-	/* disable HT */
-	if (!enable_ht) {
-		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
-	} else {
+	memset(&ht_conf, 0, sizeof(struct ieee80211_ht_info));
+	memset(&ht_bss_conf, 0, sizeof(struct ieee80211_ht_bss_info));
+
+	if (enable_ht) {
+		if (!(conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE))
+			changed |= BSS_CHANGED_HT;
+
 		conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE;
-		conf->ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap;
-		conf->ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
-		conf->ht_conf.cap |=
-			sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
-		conf->ht_bss_conf.primary_channel =
-			req_bss_cap->primary_channel;
-		conf->ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
-		conf->ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
+		ht_conf.ht_supported = 1;
+
+		ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap;
+		ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
+		ht_conf.cap |= sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
+
 		for (i = 0; i < SUPP_MCS_SET_LEN; i++)
-			conf->ht_conf.supp_mcs_set[i] =
-				sband->ht_info.supp_mcs_set[i] &
-				  req_ht_cap->supp_mcs_set[i];
-
-		/* In STA mode, this gives us indication
-		 * to the AP's mode of operation */
-		conf->ht_conf.ht_supported = 1;
-		conf->ht_conf.ampdu_factor = req_ht_cap->ampdu_factor;
-		conf->ht_conf.ampdu_density = req_ht_cap->ampdu_density;
+			ht_conf.supp_mcs_set[i] =
+					sband->ht_info.supp_mcs_set[i] &
+					req_ht_cap->supp_mcs_set[i];
+
+		ht_bss_conf.primary_channel = req_bss_cap->primary_channel;
+		ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
+		ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
+
+		ht_conf.ampdu_factor = req_ht_cap->ampdu_factor;
+		ht_conf.ampdu_density = req_ht_cap->ampdu_density;
+
+		/* if bss configuration changed store the new one */
+		if (memcmp(&conf->ht_conf, &ht_conf, sizeof(ht_conf)) ||
+		    memcmp(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf))) {
+			changed |= BSS_CHANGED_HT;
+			memcpy(&conf->ht_conf, &ht_conf, sizeof(ht_conf));
+			memcpy(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf));
+		}
+	} else {
+		if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE)
+			changed |= BSS_CHANGED_HT;
+		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
 	}
 
-	local->ops->conf_ht(local_to_hw(local), &local->hw.conf);
-
-	return 0;
+	return changed;
 }
 
 void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 0997a0f96203..377c448a0556 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -865,9 +865,9 @@ int ieee80211_if_config(struct net_device *dev);
 int ieee80211_if_config_beacon(struct net_device *dev);
 void ieee80211_tx_set_protected(struct ieee80211_tx_data *tx);
 void ieee80211_if_setup(struct net_device *dev);
-int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht,
-			   struct ieee80211_ht_info *req_ht_cap,
-			   struct ieee80211_ht_bss_info *req_bss_cap);
+u32 ieee80211_handle_ht(struct ieee80211_local *local, int enable_ht,
+			struct ieee80211_ht_info *req_ht_cap,
+			struct ieee80211_ht_bss_info *req_bss_cap);
 
 /* ieee80211_ioctl.c */
 extern const struct iw_handler_def ieee80211_iw_handler_def;
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index 1ee07f0b02e2..0e936233dac5 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -493,6 +493,7 @@ static void ieee80211_set_associated(struct net_device *dev,
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct ieee80211_local *local = sdata->local;
+	struct ieee80211_conf *conf = &local_to_hw(local)->conf;
 	union iwreq_data wrqu;
 	u32 changed = BSS_CHANGED_ASSOC;
 
@@ -505,7 +506,7 @@ static void ieee80211_set_associated(struct net_device *dev,
 			return;
 
 		bss = ieee80211_rx_bss_get(dev, ifsta->bssid,
-					   local->hw.conf.channel->center_freq,
+					   conf->channel->center_freq,
 					   ifsta->ssid, ifsta->ssid_len);
 		if (bss) {
 			if (bss->has_erp_value)
@@ -514,6 +515,13 @@ static void ieee80211_set_associated(struct net_device *dev,
 			ieee80211_rx_bss_put(dev, bss);
 		}
 
+		if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) {
+			changed |= BSS_CHANGED_HT;
+			sdata->bss_conf.assoc_ht = 1;
+			sdata->bss_conf.ht_conf = &conf->ht_conf;
+			sdata->bss_conf.ht_bss_conf = &conf->ht_bss_conf;
+		}
+
 		netif_carrier_on(dev);
 		ifsta->flags |= IEEE80211_STA_PREV_BSSID_SET;
 		memcpy(ifsta->prev_bssid, sdata->u.sta.bssid, ETH_ALEN);
@@ -524,6 +532,11 @@ static void ieee80211_set_associated(struct net_device *dev,
 		ifsta->flags &= ~IEEE80211_STA_ASSOCIATED;
 		netif_carrier_off(dev);
 		ieee80211_reset_erp_info(dev);
+
+		sdata->bss_conf.assoc_ht = 0;
+		sdata->bss_conf.ht_conf = NULL;
+		sdata->bss_conf.ht_bss_conf = NULL;
+
 		memset(wrqu.ap_addr.sa_data, 0, ETH_ALEN);
 	}
 	wrqu.ap_addr.sa_family = ARPHRD_ETHER;
@@ -1999,17 +2012,15 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 	else
 		sdata->flags &= ~IEEE80211_SDATA_OPERATING_GMODE;
 
-	if (elems.ht_cap_elem && elems.ht_info_elem && elems.wmm_param &&
-	    local->ops->conf_ht) {
+	if (elems.ht_cap_elem && elems.ht_info_elem && elems.wmm_param) {
 		struct ieee80211_ht_bss_info bss_info;
-
 		ieee80211_ht_cap_ie_to_ht_info(
 				(struct ieee80211_ht_cap *)
 				elems.ht_cap_elem, &sta->ht_info);
 		ieee80211_ht_addt_info_ie_to_ht_bss_info(
 				(struct ieee80211_ht_addt_info *)
 				elems.ht_info_elem, &bss_info);
-		ieee80211_hw_config_ht(local, 1, &sta->ht_info, &bss_info);
+		ieee80211_handle_ht(local, 1, &sta->ht_info, &bss_info);
 	}
 
 	rate_control_rate_init(sta, local);
@@ -2760,20 +2771,14 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 		changed |= ieee80211_handle_erp_ie(sdata, elems.erp_info[0]);
 
 	if (elems.ht_cap_elem && elems.ht_info_elem &&
-	    elems.wmm_param && local->ops->conf_ht &&
-	    conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) {
+	    elems.wmm_param && conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) {
 		struct ieee80211_ht_bss_info bss_info;
 
 		ieee80211_ht_addt_info_ie_to_ht_bss_info(
 				(struct ieee80211_ht_addt_info *)
 				elems.ht_info_elem, &bss_info);
-		/* check if AP changed bss inforamation */
-		if ((conf->ht_bss_conf.primary_channel !=
-		     bss_info.primary_channel) ||
-		    (conf->ht_bss_conf.bss_cap != bss_info.bss_cap) ||
-		    (conf->ht_bss_conf.bss_op_mode != bss_info.bss_op_mode))
-			ieee80211_hw_config_ht(local, 1, &conf->ht_conf,
-						&bss_info);
+		changed |= ieee80211_handle_ht(local, 1, &conf->ht_conf,
+					       &bss_info);
 	}
 
 	if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) {
-- 
cgit v1.2.3


From 21c0cbe760ca6b5d4c6927c3ec1352a843a8c11c Mon Sep 17 00:00:00 2001
From: Tomas Winkler <tomas.winkler@intel.com>
Date: Fri, 28 Mar 2008 16:33:34 -0700
Subject: mac80211: add association capabilty and timing info into bss_conf

This patch adds assocation capability, timestamp (tsf) and beacon interval
to bss_conf. This is required for successful assocation of iwlwifi drivers

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h       | 6 ++++++
 net/mac80211/ieee80211_sta.c | 9 ++++++++-
 2 files changed, 14 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5174eaa89500..01b32152b89e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -197,6 +197,9 @@ enum ieee80211_bss_change {
  * @aid: association ID number, valid only when @assoc is true
  * @use_cts_prot: use CTS protection
  * @use_short_preamble: use 802.11b short preamble
+ * @timestamp: beacon timestamp
+ * @beacon_int: beacon interval
+ * @assoc_capability: capabbilities taken from assoc resp
  * @assoc_ht: association in HT mode
  * @ht_conf: ht capabilities
  * @ht_bss_conf: ht extended capabilities
@@ -208,6 +211,9 @@ struct ieee80211_bss_conf {
 	/* erp related data */
 	bool use_cts_prot;
 	bool use_short_preamble;
+	u16 beacon_int;
+	u16 assoc_capability;
+	u64 timestamp;
 	/* ht related data */
 	bool assoc_ht;
 	struct ieee80211_ht_info *ht_conf;
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index 0e936233dac5..cb119d3b6353 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -509,9 +509,14 @@ static void ieee80211_set_associated(struct net_device *dev,
 					   conf->channel->center_freq,
 					   ifsta->ssid, ifsta->ssid_len);
 		if (bss) {
+			/* set timing information */
+			sdata->bss_conf.beacon_int = bss->beacon_int;
+			sdata->bss_conf.timestamp = bss->timestamp;
+
 			if (bss->has_erp_value)
 				changed |= ieee80211_handle_erp_ie(
 						sdata, bss->erp_value);
+
 			ieee80211_rx_bss_put(dev, bss);
 		}
 
@@ -2033,8 +2038,10 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata,
 	} else
 		rcu_read_unlock();
 
-	/* set AID, ieee80211_set_associated() will tell the driver */
+	/* set AID and assoc capability,
+	 * ieee80211_set_associated() will tell the driver */
 	bss_conf->aid = aid;
+	bss_conf->assoc_capability = capab_info;
 	ieee80211_set_associated(dev, ifsta, 1);
 
 	ieee80211_associated(dev, ifsta);
-- 
cgit v1.2.3


From 8fe2b65a18e49bfde56a59ed4ab3fc7aa0c2f325 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Sun, 30 Mar 2008 00:10:50 +0100
Subject: ssb: Turn suspend/resume upside down

Turn the SSB bus suspend mechanism upside down.
Instead of deciding by an internal reference count when to suspend/resume,
let the parent bus call us in their suspend/resume routine.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/b43/pcmcia.c         | 10 ++--
 drivers/ssb/driver_chipcommon.c           |  2 +-
 drivers/ssb/main.c                        | 81 +++++++++++++------------------
 drivers/ssb/pcihost_wrapper.c             | 10 ++++
 drivers/ssb/pcmcia.c                      | 34 +++++++++----
 drivers/ssb/ssb_private.h                 |  5 ++
 include/linux/ssb/ssb.h                   | 10 ++--
 include/linux/ssb/ssb_driver_chipcommon.h |  3 +-
 8 files changed, 88 insertions(+), 67 deletions(-)

(limited to 'include')

diff --git a/drivers/net/wireless/b43/pcmcia.c b/drivers/net/wireless/b43/pcmcia.c
index 371e4a119511..b8aa16307f79 100644
--- a/drivers/net/wireless/b43/pcmcia.c
+++ b/drivers/net/wireless/b43/pcmcia.c
@@ -43,14 +43,16 @@ MODULE_DEVICE_TABLE(pcmcia, b43_pcmcia_tbl);
 #ifdef CONFIG_PM
 static int b43_pcmcia_suspend(struct pcmcia_device *dev)
 {
-	//TODO
-	return 0;
+	struct ssb_bus *ssb = dev->priv;
+
+	return ssb_bus_suspend(ssb);
 }
 
 static int b43_pcmcia_resume(struct pcmcia_device *dev)
 {
-	//TODO
-	return 0;
+	struct ssb_bus *ssb = dev->priv;
+
+	return ssb_bus_resume(ssb);
 }
 #else /* CONFIG_PM */
 # define b43_pcmcia_suspend		NULL
diff --git a/drivers/ssb/driver_chipcommon.c b/drivers/ssb/driver_chipcommon.c
index 45b672a69003..571f4fd55236 100644
--- a/drivers/ssb/driver_chipcommon.c
+++ b/drivers/ssb/driver_chipcommon.c
@@ -251,7 +251,7 @@ void ssb_chipcommon_init(struct ssb_chipcommon *cc)
 	calc_fast_powerup_delay(cc);
 }
 
-void ssb_chipco_suspend(struct ssb_chipcommon *cc, pm_message_t state)
+void ssb_chipco_suspend(struct ssb_chipcommon *cc)
 {
 	if (!cc->dev)
 		return;
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index 2fcfd73b3b6e..c0cbdba07aee 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -120,35 +120,12 @@ static void ssb_device_put(struct ssb_device *dev)
 		put_device(dev->dev);
 }
 
-static int ssb_bus_resume(struct ssb_bus *bus)
-{
-	int err;
-
-	ssb_pci_xtal(bus, SSB_GPIO_XTAL | SSB_GPIO_PLL, 1);
-	err = ssb_pcmcia_init(bus);
-	if (err) {
-		/* No need to disable XTAL, as we don't have one on PCMCIA. */
-		return err;
-	}
-	ssb_chipco_resume(&bus->chipco);
-
-	return 0;
-}
-
 static int ssb_device_resume(struct device *dev)
 {
 	struct ssb_device *ssb_dev = dev_to_ssb_dev(dev);
 	struct ssb_driver *ssb_drv;
-	struct ssb_bus *bus;
 	int err = 0;
 
-	bus = ssb_dev->bus;
-	if (bus->suspend_cnt == bus->nr_devices) {
-		err = ssb_bus_resume(bus);
-		if (err)
-			return err;
-	}
-	bus->suspend_cnt--;
 	if (dev->driver) {
 		ssb_drv = drv_to_ssb_drv(dev->driver);
 		if (ssb_drv && ssb_drv->resume)
@@ -160,27 +137,10 @@ out:
 	return err;
 }
 
-static void ssb_bus_suspend(struct ssb_bus *bus, pm_message_t state)
-{
-	ssb_chipco_suspend(&bus->chipco, state);
-	ssb_pci_xtal(bus, SSB_GPIO_XTAL | SSB_GPIO_PLL, 0);
-
-	/* Reset HW state information in memory, so that HW is
-	 * completely reinitialized on resume. */
-	bus->mapped_device = NULL;
-#ifdef CONFIG_SSB_DRIVER_PCICORE
-	bus->pcicore.setup_done = 0;
-#endif
-#ifdef CONFIG_SSB_DEBUG
-	bus->powered_up = 0;
-#endif
-}
-
 static int ssb_device_suspend(struct device *dev, pm_message_t state)
 {
 	struct ssb_device *ssb_dev = dev_to_ssb_dev(dev);
 	struct ssb_driver *ssb_drv;
-	struct ssb_bus *bus;
 	int err = 0;
 
 	if (dev->driver) {
@@ -190,17 +150,44 @@ static int ssb_device_suspend(struct device *dev, pm_message_t state)
 		if (err)
 			goto out;
 	}
+out:
+	return err;
+}
+
+int ssb_bus_resume(struct ssb_bus *bus)
+{
+	int err;
 
-	bus = ssb_dev->bus;
-	bus->suspend_cnt++;
-	if (bus->suspend_cnt == bus->nr_devices) {
-		/* All devices suspended. Shutdown the bus. */
-		ssb_bus_suspend(bus, state);
+	/* Reset HW state information in memory, so that HW is
+	 * completely reinitialized. */
+	bus->mapped_device = NULL;
+#ifdef CONFIG_SSB_DRIVER_PCICORE
+	bus->pcicore.setup_done = 0;
+#endif
+
+	err = ssb_bus_powerup(bus, 0);
+	if (err)
+		return err;
+	err = ssb_pcmcia_hardware_setup(bus);
+	if (err) {
+		ssb_bus_may_powerdown(bus);
+		return err;
 	}
+	ssb_chipco_resume(&bus->chipco);
+	ssb_bus_may_powerdown(bus);
 
-out:
-	return err;
+	return 0;
+}
+EXPORT_SYMBOL(ssb_bus_resume);
+
+int ssb_bus_suspend(struct ssb_bus *bus)
+{
+	ssb_chipco_suspend(&bus->chipco);
+	ssb_pci_xtal(bus, SSB_GPIO_XTAL | SSB_GPIO_PLL, 0);
+
+	return 0;
 }
+EXPORT_SYMBOL(ssb_bus_suspend);
 
 #ifdef CONFIG_SSB_PCIHOST
 int ssb_devices_freeze(struct ssb_bus *bus)
diff --git a/drivers/ssb/pcihost_wrapper.c b/drivers/ssb/pcihost_wrapper.c
index 82a10abef640..e82db4aaa050 100644
--- a/drivers/ssb/pcihost_wrapper.c
+++ b/drivers/ssb/pcihost_wrapper.c
@@ -18,6 +18,12 @@
 #ifdef CONFIG_PM
 static int ssb_pcihost_suspend(struct pci_dev *dev, pm_message_t state)
 {
+	struct ssb_bus *ssb = pci_get_drvdata(dev);
+	int err;
+
+	err = ssb_bus_suspend(ssb);
+	if (err)
+		return err;
 	pci_save_state(dev);
 	pci_disable_device(dev);
 	pci_set_power_state(dev, pci_choose_state(dev, state));
@@ -27,6 +33,7 @@ static int ssb_pcihost_suspend(struct pci_dev *dev, pm_message_t state)
 
 static int ssb_pcihost_resume(struct pci_dev *dev)
 {
+	struct ssb_bus *ssb = pci_get_drvdata(dev);
 	int err;
 
 	pci_set_power_state(dev, 0);
@@ -34,6 +41,9 @@ static int ssb_pcihost_resume(struct pci_dev *dev)
 	if (err)
 		return err;
 	pci_restore_state(dev);
+	err = ssb_bus_resume(ssb);
+	if (err)
+		return err;
 
 	return 0;
 }
diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c
index d674cef7210d..dcaf2412bea7 100644
--- a/drivers/ssb/pcmcia.c
+++ b/drivers/ssb/pcmcia.c
@@ -684,6 +684,29 @@ static int ssb_pcmcia_cor_setup(struct ssb_bus *bus, u8 cor)
 	return 0;
 }
 
+/* Initialize the PCMCIA hardware. This is called on Init and Resume. */
+int ssb_pcmcia_hardware_setup(struct ssb_bus *bus)
+{
+	int err;
+
+	if (bus->bustype != SSB_BUSTYPE_PCMCIA)
+		return 0;
+
+	/* Switch segment to a known state and sync
+	 * bus->mapped_pcmcia_seg with hardware state. */
+	ssb_pcmcia_switch_segment(bus, 0);
+	/* Init the COR register. */
+	err = ssb_pcmcia_cor_setup(bus, CISREG_COR);
+	if (err)
+		return err;
+	/* Some cards also need this register to get poked. */
+	err = ssb_pcmcia_cor_setup(bus, CISREG_COR + 0x80);
+	if (err)
+		return err;
+
+	return 0;
+}
+
 void ssb_pcmcia_exit(struct ssb_bus *bus)
 {
 	if (bus->bustype != SSB_BUSTYPE_PCMCIA)
@@ -699,16 +722,7 @@ int ssb_pcmcia_init(struct ssb_bus *bus)
 	if (bus->bustype != SSB_BUSTYPE_PCMCIA)
 		return 0;
 
-	/* Switch segment to a known state and sync
-	 * bus->mapped_pcmcia_seg with hardware state. */
-	ssb_pcmcia_switch_segment(bus, 0);
-
-	/* Init the COR register. */
-	err = ssb_pcmcia_cor_setup(bus, CISREG_COR);
-	if (err)
-		goto error;
-	/* Some cards also need this register to get poked. */
-	err = ssb_pcmcia_cor_setup(bus, CISREG_COR + 0x80);
+	err = ssb_pcmcia_hardware_setup(bus);
 	if (err)
 		goto error;
 
diff --git a/drivers/ssb/ssb_private.h b/drivers/ssb/ssb_private.h
index a83bf7a4d80b..ebc32d8fe15f 100644
--- a/drivers/ssb/ssb_private.h
+++ b/drivers/ssb/ssb_private.h
@@ -81,6 +81,7 @@ extern int ssb_pcmcia_switch_segment(struct ssb_bus *bus,
 				     u8 seg);
 extern int ssb_pcmcia_get_invariants(struct ssb_bus *bus,
 				     struct ssb_init_invariants *iv);
+extern int ssb_pcmcia_hardware_setup(struct ssb_bus *bus);
 extern void ssb_pcmcia_exit(struct ssb_bus *bus);
 extern int ssb_pcmcia_init(struct ssb_bus *bus);
 extern const struct ssb_bus_ops ssb_pcmcia_ops;
@@ -100,6 +101,10 @@ static inline int ssb_pcmcia_switch_segment(struct ssb_bus *bus,
 {
 	return 0;
 }
+static inline int ssb_pcmcia_hardware_setup(struct ssb_bus *bus)
+{
+	return 0;
+}
 static inline void ssb_pcmcia_exit(struct ssb_bus *bus)
 {
 }
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 8644e03cf588..a8ca396f810a 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -260,9 +260,6 @@ struct ssb_bus {
 	struct ssb_device devices[SSB_MAX_NR_CORES];
 	u8 nr_devices;
 
-	/* Reference count. Number of suspended devices. */
-	u8 suspend_cnt;
-
 	/* Software ID number for this bus. */
 	unsigned int busnumber;
 
@@ -334,6 +331,13 @@ extern int ssb_bus_pcmciabus_register(struct ssb_bus *bus,
 
 extern void ssb_bus_unregister(struct ssb_bus *bus);
 
+/* Suspend a SSB bus.
+ * Call this from the parent bus suspend routine. */
+extern int ssb_bus_suspend(struct ssb_bus *bus);
+/* Resume a SSB bus.
+ * Call this from the parent bus resume routine. */
+extern int ssb_bus_resume(struct ssb_bus *bus);
+
 extern u32 ssb_clockspeed(struct ssb_bus *bus);
 
 /* Is the device enabled in hardware? */
diff --git a/include/linux/ssb/ssb_driver_chipcommon.h b/include/linux/ssb/ssb_driver_chipcommon.h
index b548a54ff1f5..7d7e03dcf77c 100644
--- a/include/linux/ssb/ssb_driver_chipcommon.h
+++ b/include/linux/ssb/ssb_driver_chipcommon.h
@@ -367,8 +367,7 @@ static inline bool ssb_chipco_available(struct ssb_chipcommon *cc)
 
 extern void ssb_chipcommon_init(struct ssb_chipcommon *cc);
 
-#include <linux/pm.h>
-extern void ssb_chipco_suspend(struct ssb_chipcommon *cc, pm_message_t state);
+extern void ssb_chipco_suspend(struct ssb_chipcommon *cc);
 extern void ssb_chipco_resume(struct ssb_chipcommon *cc);
 
 extern void ssb_chipco_get_clockcpu(struct ssb_chipcommon *cc,
-- 
cgit v1.2.3


From fff7710937f755099209357e5b5740d42a2c9f97 Mon Sep 17 00:00:00 2001
From: Chr <chunkeey@web.de>
Date: Tue, 1 Apr 2008 21:45:18 +0200
Subject: mac80211: add station aid into ieee80211_tx_control

This patch is necessary for the upcoming Accesspoint patch for p54.

Signed-off-by: Christian Lamparter <chunkeey@web.de>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h | 1 +
 net/mac80211/tx.c      | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 01b32152b89e..999f970da6ba 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -302,6 +302,7 @@ struct ieee80211_tx_control {
 	u8 iv_len;		/* length of the IV field in octets */
 	u8 queue;		/* hardware queue to use for this frame;
 				 * 0 = highest, hw->queues-1 = lowest */
+	u16 aid;		/* Station AID */
 	int type;	/* internal */
 };
 
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 80f4343a3007..ea3fa0f91906 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -741,6 +741,7 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx)
 	}
 
 	if (tx->sta) {
+		control->aid = tx->sta->aid;
 		tx->sta->tx_packets++;
 		tx->sta->tx_fragments++;
 		tx->sta->tx_bytes += tx->skb->len;
-- 
cgit v1.2.3


From d625a29ba649a4df6027520ffc378f23c0e6883e Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Wed, 2 Apr 2008 19:46:56 +0200
Subject: ssb: Add support for block-I/O

This adds support for block based I/O to SSB.
This is needed in order to efficiently support PIO data
transfers to the card.
The block-I/O support is only compiled, if it's selected by the
weird driver that needs it. So there's no overhead for sane devices.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/ssb/Kconfig     |   5 ++
 drivers/ssb/main.c      | 102 +++++++++++++++++++++++++++++++++++++++++
 drivers/ssb/pci.c       |  70 ++++++++++++++++++++++++++++
 drivers/ssb/pcmcia.c    | 119 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/ssb/ssb.h |  19 ++++++++
 5 files changed, 315 insertions(+)

(limited to 'include')

diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig
index 0f7cce2560d1..cd845b8acd17 100644
--- a/drivers/ssb/Kconfig
+++ b/drivers/ssb/Kconfig
@@ -24,6 +24,11 @@ config SSB
 config SSB_SPROM
 	bool
 
+# Support for Block-I/O. SELECT this from the driver that needs it.
+config SSB_BLOCKIO
+	bool
+	depends on SSB
+
 config SSB_PCIHOST_POSSIBLE
 	bool
 	depends on SSB && (PCI = y || PCI = SSB)
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index 3e58db7ef608..19ddd2bd1057 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -555,6 +555,55 @@ static u32 ssb_ssb_read32(struct ssb_device *dev, u16 offset)
 	return readl(bus->mmio + offset);
 }
 
+#ifdef CONFIG_SSB_BLOCKIO
+static void ssb_ssb_block_read(struct ssb_device *dev, void *buffer,
+			       size_t count, u16 offset, u8 reg_width)
+{
+	struct ssb_bus *bus = dev->bus;
+	void __iomem *addr;
+
+	offset += dev->core_index * SSB_CORE_SIZE;
+	addr = bus->mmio + offset;
+
+	switch (reg_width) {
+	case sizeof(u8): {
+		u8 *buf = buffer;
+
+		while (count) {
+			*buf = __raw_readb(addr);
+			buf++;
+			count--;
+		}
+		break;
+	}
+	case sizeof(u16): {
+		__le16 *buf = buffer;
+
+		SSB_WARN_ON(count & 1);
+		while (count) {
+			*buf = (__force __le16)__raw_readw(addr);
+			buf++;
+			count -= 2;
+		}
+		break;
+	}
+	case sizeof(u32): {
+		__le32 *buf = buffer;
+
+		SSB_WARN_ON(count & 3);
+		while (count) {
+			*buf = (__force __le32)__raw_readl(addr);
+			buf++;
+			count -= 4;
+		}
+		break;
+	}
+	default:
+		SSB_WARN_ON(1);
+	}
+}
+#endif /* CONFIG_SSB_BLOCKIO */
+
 static void ssb_ssb_write8(struct ssb_device *dev, u16 offset, u8 value)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -579,6 +628,55 @@ static void ssb_ssb_write32(struct ssb_device *dev, u16 offset, u32 value)
 	writel(value, bus->mmio + offset);
 }
 
+#ifdef CONFIG_SSB_BLOCKIO
+static void ssb_ssb_block_write(struct ssb_device *dev, const void *buffer,
+				size_t count, u16 offset, u8 reg_width)
+{
+	struct ssb_bus *bus = dev->bus;
+	void __iomem *addr;
+
+	offset += dev->core_index * SSB_CORE_SIZE;
+	addr = bus->mmio + offset;
+
+	switch (reg_width) {
+	case sizeof(u8): {
+		const u8 *buf = buffer;
+
+		while (count) {
+			__raw_writeb(*buf, addr);
+			buf++;
+			count--;
+		}
+		break;
+	}
+	case sizeof(u16): {
+		const __le16 *buf = buffer;
+
+		SSB_WARN_ON(count & 1);
+		while (count) {
+			__raw_writew((__force u16)(*buf), addr);
+			buf++;
+			count -= 2;
+		}
+		break;
+	}
+	case sizeof(u32): {
+		const __le32 *buf = buffer;
+
+		SSB_WARN_ON(count & 3);
+		while (count) {
+			__raw_writel((__force u32)(*buf), addr);
+			buf++;
+			count -= 4;
+		}
+		break;
+	}
+	default:
+		SSB_WARN_ON(1);
+	}
+}
+#endif /* CONFIG_SSB_BLOCKIO */
+
 /* Ops for the plain SSB bus without a host-device (no PCI or PCMCIA). */
 static const struct ssb_bus_ops ssb_ssb_ops = {
 	.read8		= ssb_ssb_read8,
@@ -587,6 +685,10 @@ static const struct ssb_bus_ops ssb_ssb_ops = {
 	.write8		= ssb_ssb_write8,
 	.write16	= ssb_ssb_write16,
 	.write32	= ssb_ssb_write32,
+#ifdef CONFIG_SSB_BLOCKIO
+	.block_read	= ssb_ssb_block_read,
+	.block_write	= ssb_ssb_block_write,
+#endif
 };
 
 static int ssb_fetch_invariants(struct ssb_bus *bus,
diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index f1514b33cfae..904b1a8d0885 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -613,6 +613,41 @@ static u32 ssb_pci_read32(struct ssb_device *dev, u16 offset)
 	return ioread32(bus->mmio + offset);
 }
 
+#ifdef CONFIG_SSB_BLOCKIO
+static void ssb_pci_block_read(struct ssb_device *dev, void *buffer,
+			       size_t count, u16 offset, u8 reg_width)
+{
+	struct ssb_bus *bus = dev->bus;
+	void __iomem *addr = bus->mmio + offset;
+
+	if (unlikely(ssb_pci_assert_buspower(bus)))
+		goto error;
+	if (unlikely(bus->mapped_device != dev)) {
+		if (unlikely(ssb_pci_switch_core(bus, dev)))
+			goto error;
+	}
+	switch (reg_width) {
+	case sizeof(u8):
+		ioread8_rep(addr, buffer, count);
+		break;
+	case sizeof(u16):
+		SSB_WARN_ON(count & 1);
+		ioread16_rep(addr, buffer, count >> 1);
+		break;
+	case sizeof(u32):
+		SSB_WARN_ON(count & 3);
+		ioread32_rep(addr, buffer, count >> 2);
+		break;
+	default:
+		SSB_WARN_ON(1);
+	}
+
+	return;
+error:
+	memset(buffer, 0xFF, count);
+}
+#endif /* CONFIG_SSB_BLOCKIO */
+
 static void ssb_pci_write8(struct ssb_device *dev, u16 offset, u8 value)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -652,6 +687,37 @@ static void ssb_pci_write32(struct ssb_device *dev, u16 offset, u32 value)
 	iowrite32(value, bus->mmio + offset);
 }
 
+#ifdef CONFIG_SSB_BLOCKIO
+static void ssb_pci_block_write(struct ssb_device *dev, const void *buffer,
+				size_t count, u16 offset, u8 reg_width)
+{
+	struct ssb_bus *bus = dev->bus;
+	void __iomem *addr = bus->mmio + offset;
+
+	if (unlikely(ssb_pci_assert_buspower(bus)))
+		return;
+	if (unlikely(bus->mapped_device != dev)) {
+		if (unlikely(ssb_pci_switch_core(bus, dev)))
+			return;
+	}
+	switch (reg_width) {
+	case sizeof(u8):
+		iowrite8_rep(addr, buffer, count);
+		break;
+	case sizeof(u16):
+		SSB_WARN_ON(count & 1);
+		iowrite16_rep(addr, buffer, count >> 1);
+		break;
+	case sizeof(u32):
+		SSB_WARN_ON(count & 3);
+		iowrite32_rep(addr, buffer, count >> 2);
+		break;
+	default:
+		SSB_WARN_ON(1);
+	}
+}
+#endif /* CONFIG_SSB_BLOCKIO */
+
 /* Not "static", as it's used in main.c */
 const struct ssb_bus_ops ssb_pci_ops = {
 	.read8		= ssb_pci_read8,
@@ -660,6 +726,10 @@ const struct ssb_bus_ops ssb_pci_ops = {
 	.write8		= ssb_pci_write8,
 	.write16	= ssb_pci_write16,
 	.write32	= ssb_pci_write32,
+#ifdef CONFIG_SSB_BLOCKIO
+	.block_read	= ssb_pci_block_read,
+	.block_write	= ssb_pci_block_write,
+#endif
 };
 
 static ssize_t ssb_pci_attr_sprom_show(struct device *pcidev,
diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c
index dcaf2412bea7..24c2a46c1476 100644
--- a/drivers/ssb/pcmcia.c
+++ b/drivers/ssb/pcmcia.c
@@ -285,6 +285,64 @@ static u32 ssb_pcmcia_read32(struct ssb_device *dev, u16 offset)
 	return (lo | (hi << 16));
 }
 
+#ifdef CONFIG_SSB_BLOCKIO
+static void ssb_pcmcia_block_read(struct ssb_device *dev, void *buffer,
+				  size_t count, u16 offset, u8 reg_width)
+{
+	struct ssb_bus *bus = dev->bus;
+	unsigned long flags;
+	void __iomem *addr = bus->mmio + offset;
+	int err;
+
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (unlikely(err)) {
+		memset(buffer, 0xFF, count);
+		goto unlock;
+	}
+	switch (reg_width) {
+	case sizeof(u8): {
+		u8 *buf = buffer;
+
+		while (count) {
+			*buf = __raw_readb(addr);
+			buf++;
+			count--;
+		}
+		break;
+	}
+	case sizeof(u16): {
+		__le16 *buf = buffer;
+
+		SSB_WARN_ON(count & 1);
+		while (count) {
+			*buf = (__force __le16)__raw_readw(addr);
+			buf++;
+			count -= 2;
+		}
+		break;
+	}
+	case sizeof(u32): {
+		__le16 *buf = buffer;
+
+		SSB_WARN_ON(count & 3);
+		while (count) {
+			*buf = (__force __le16)__raw_readw(addr);
+			buf++;
+			*buf = (__force __le16)__raw_readw(addr + 2);
+			buf++;
+			count -= 4;
+		}
+		break;
+	}
+	default:
+		SSB_WARN_ON(1);
+	}
+unlock:
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
+}
+#endif /* CONFIG_SSB_BLOCKIO */
+
 static void ssb_pcmcia_write8(struct ssb_device *dev, u16 offset, u8 value)
 {
 	struct ssb_bus *bus = dev->bus;
@@ -329,6 +387,63 @@ static void ssb_pcmcia_write32(struct ssb_device *dev, u16 offset, u32 value)
 	spin_unlock_irqrestore(&bus->bar_lock, flags);
 }
 
+#ifdef CONFIG_SSB_BLOCKIO
+static void ssb_pcmcia_block_write(struct ssb_device *dev, const void *buffer,
+				   size_t count, u16 offset, u8 reg_width)
+{
+	struct ssb_bus *bus = dev->bus;
+	unsigned long flags;
+	void __iomem *addr = bus->mmio + offset;
+	int err;
+
+	spin_lock_irqsave(&bus->bar_lock, flags);
+	err = select_core_and_segment(dev, &offset);
+	if (unlikely(err))
+		goto unlock;
+	switch (reg_width) {
+	case sizeof(u8): {
+		const u8 *buf = buffer;
+
+		while (count) {
+			__raw_writeb(*buf, addr);
+			buf++;
+			count--;
+		}
+		break;
+	}
+	case sizeof(u16): {
+		const __le16 *buf = buffer;
+
+		SSB_WARN_ON(count & 1);
+		while (count) {
+			__raw_writew((__force u16)(*buf), addr);
+			buf++;
+			count -= 2;
+		}
+		break;
+	}
+	case sizeof(u32): {
+		const __le16 *buf = buffer;
+
+		SSB_WARN_ON(count & 3);
+		while (count) {
+			__raw_writew((__force u16)(*buf), addr);
+			buf++;
+			__raw_writew((__force u16)(*buf), addr + 2);
+			buf++;
+			count -= 4;
+		}
+		break;
+	}
+	default:
+		SSB_WARN_ON(1);
+	}
+unlock:
+	mmiowb();
+	spin_unlock_irqrestore(&bus->bar_lock, flags);
+}
+#endif /* CONFIG_SSB_BLOCKIO */
+
 /* Not "static", as it's used in main.c */
 const struct ssb_bus_ops ssb_pcmcia_ops = {
 	.read8		= ssb_pcmcia_read8,
@@ -337,6 +452,10 @@ const struct ssb_bus_ops ssb_pcmcia_ops = {
 	.write8		= ssb_pcmcia_write8,
 	.write16	= ssb_pcmcia_write16,
 	.write32	= ssb_pcmcia_write32,
+#ifdef CONFIG_SSB_BLOCKIO
+	.block_read	= ssb_pcmcia_block_read,
+	.block_write	= ssb_pcmcia_block_write,
+#endif
 };
 
 static int ssb_pcmcia_sprom_command(struct ssb_bus *bus, u8 command)
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index a8ca396f810a..9f95afd0e9e3 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -78,6 +78,12 @@ struct ssb_bus_ops {
 	void (*write8)(struct ssb_device *dev, u16 offset, u8 value);
 	void (*write16)(struct ssb_device *dev, u16 offset, u16 value);
 	void (*write32)(struct ssb_device *dev, u16 offset, u32 value);
+#ifdef CONFIG_SSB_BLOCKIO
+	void (*block_read)(struct ssb_device *dev, void *buffer,
+			   size_t count, u16 offset, u8 reg_width);
+	void (*block_write)(struct ssb_device *dev, const void *buffer,
+			    size_t count, u16 offset, u8 reg_width);
+#endif
 };
 
 
@@ -374,6 +380,19 @@ static inline void ssb_write32(struct ssb_device *dev, u16 offset, u32 value)
 {
 	dev->ops->write32(dev, offset, value);
 }
+#ifdef CONFIG_SSB_BLOCKIO
+static inline void ssb_block_read(struct ssb_device *dev, void *buffer,
+				  size_t count, u16 offset, u8 reg_width)
+{
+	dev->ops->block_read(dev, buffer, count, offset, reg_width);
+}
+
+static inline void ssb_block_write(struct ssb_device *dev, const void *buffer,
+				   size_t count, u16 offset, u8 reg_width)
+{
+	dev->ops->block_write(dev, buffer, count, offset, reg_width);
+}
+#endif /* CONFIG_SSB_BLOCKIO */
 
 
 /* Translation (routing) bits that need to be ORed to DMA
-- 
cgit v1.2.3


From 84363e6e07f17f8cc580065260907ee3f0520485 Mon Sep 17 00:00:00 2001
From: Mohamed Abbas <mabbas@linux.intel.com>
Date: Fri, 4 Apr 2008 16:59:58 -0700
Subject: mac80211: notify mac from low level driver (iwlwifi)

Add new API to MAC80211 to allow low level driver to
notify MAC with driver status.

Signed-off-by: Mohamed Abbas <mabbas@linux.intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  1 +
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  1 +
 include/net/mac80211.h                      | 19 +++++++++++++++++++
 net/mac80211/ieee80211_sta.c                | 23 +++++++++++++++++++++++
 4 files changed, 44 insertions(+)

(limited to 'include')

diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 5e51cfcda39f..29a9ecdcbf35 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -5886,6 +5886,7 @@ static void iwl3945_alive_start(struct iwl3945_priv *priv)
 	if (priv->error_recovering)
 		iwl3945_error_recovery(priv);
 
+	ieee80211_notify_mac(priv->hw, IEEE80211_NOTIFY_RE_ASSOC);
 	return;
 
  restart:
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index b043871d53f1..06e44dad5f02 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -5714,6 +5714,7 @@ static void iwl4965_alive_start(struct iwl_priv *priv)
 		iwl4965_error_recovery(priv);
 
 	iwlcore_low_level_notify(priv, IWLCORE_START_EVT);
+	ieee80211_notify_mac(priv->hw, IEEE80211_NOTIFY_RE_ASSOC);
 	return;
 
  restart:
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 999f970da6ba..079e7bd86c90 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -73,6 +73,14 @@
  * not do so then mac80211 may add this under certain circumstances.
  */
 
+/**
+ * enum ieee80211_notification_type - Low level driver notification
+ * @IEEE80211_NOTIFY_RE_ASSOC: start the re-association sequence
+ */
+enum ieee80211_notification_types {
+	IEEE80211_NOTIFY_RE_ASSOC,
+};
+
 /**
  * struct ieee80211_ht_bss_info - describing BSS's HT characteristics
  *
@@ -1678,4 +1686,15 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid);
 void ieee80211_stop_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra,
 				     u16 tid);
 
+/**
+ * ieee80211_notify_mac - low level driver notification
+ * @hw: pointer as obtained from ieee80211_alloc_hw().
+ * @notification_types: enum ieee80211_notification_types
+ *
+ * This function must be called by low level driver to inform mac80211 of
+ * low level driver status change or force mac80211 to re-assoc for low
+ * level driver internal error that require re-assoc.
+ */
+void ieee80211_notify_mac(struct ieee80211_hw *hw,
+			  enum ieee80211_notification_types  notif_type);
 #endif /* MAC80211_H */
diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c
index 9e30333aa81e..89481c919cb6 100644
--- a/net/mac80211/ieee80211_sta.c
+++ b/net/mac80211/ieee80211_sta.c
@@ -4225,3 +4225,26 @@ int ieee80211_sta_disassociate(struct net_device *dev, u16 reason)
 	ieee80211_set_disassoc(dev, ifsta, 0);
 	return 0;
 }
+
+void ieee80211_notify_mac(struct ieee80211_hw *hw,
+			  enum ieee80211_notification_types  notif_type)
+{
+	struct ieee80211_local *local = hw_to_local(hw);
+	struct ieee80211_sub_if_data *sdata;
+
+	switch (notif_type) {
+	case IEEE80211_NOTIFY_RE_ASSOC:
+		rcu_read_lock();
+		list_for_each_entry_rcu(sdata, &local->interfaces, list) {
+
+			if (sdata->vif.type == IEEE80211_IF_TYPE_STA) {
+				ieee80211_sta_req_auth(sdata->dev,
+						       &sdata->u.sta);
+			}
+
+		}
+		rcu_read_unlock();
+		break;
+	}
+}
+EXPORT_SYMBOL(ieee80211_notify_mac);
-- 
cgit v1.2.3


From 2c6e6db41f01b6b4eb98809350827c9678996698 Mon Sep 17 00:00:00 2001
From: "holt@sgi.com" <holt@sgi.com>
Date: Thu, 3 Apr 2008 15:17:13 -0500
Subject: [IA64] Minimize per_cpu reservations.

This attached patch significantly shrinks boot memory allocation on ia64.
It does this by not allocating per_cpu areas for cpus that can never
exist.

In the case where acpi does not have any numa node description of the
cpus, I defaulted to assigning the first 32 round-robin on the known
nodes..  For the !CONFIG_ACPI  I used for_each_possible_cpu().

Signed-off-by: Robin Holt <holt@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/acpi.c  |  4 +++-
 arch/ia64/kernel/numa.c  |  2 +-
 arch/ia64/kernel/setup.c |  2 ++
 arch/ia64/mm/discontig.c | 12 ++++--------
 arch/ia64/mm/numa.c      |  4 +++-
 include/asm-ia64/acpi.h  | 33 +++++++++++++++++++++++++++++++++
 include/asm-ia64/numa.h  |  2 ++
 7 files changed, 48 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index 78f28d825f30..c7467f863c7a 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -423,6 +423,7 @@ static u32 __devinitdata pxm_flag[PXM_FLAG_LEN];
 #define pxm_bit_set(bit)	(set_bit(bit,(void *)pxm_flag))
 #define pxm_bit_test(bit)	(test_bit(bit,(void *)pxm_flag))
 static struct acpi_table_slit __initdata *slit_table;
+cpumask_t early_cpu_possible_map = CPU_MASK_NONE;
 
 static int get_processor_proximity_domain(struct acpi_srat_cpu_affinity *pa)
 {
@@ -482,6 +483,7 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 	    (pa->apic_id << 8) | (pa->local_sapic_eid);
 	/* nid should be overridden as logical node id later */
 	node_cpuid[srat_num_cpus].nid = pxm;
+	cpu_set(srat_num_cpus, early_cpu_possible_map);
 	srat_num_cpus++;
 }
 
@@ -559,7 +561,7 @@ void __init acpi_numa_arch_fixup(void)
 	}
 
 	/* set logical node id in cpu structure */
-	for (i = 0; i < srat_num_cpus; i++)
+	for_each_possible_early_cpu(i)
 		node_cpuid[i].nid = pxm_to_node(node_cpuid[i].nid);
 
 	printk(KERN_INFO "Number of logical nodes in system = %d\n",
diff --git a/arch/ia64/kernel/numa.c b/arch/ia64/kernel/numa.c
index a78b45f5fe2f..c93420c97409 100644
--- a/arch/ia64/kernel/numa.c
+++ b/arch/ia64/kernel/numa.c
@@ -73,7 +73,7 @@ void __init build_cpu_to_node_map(void)
 	for(node=0; node < MAX_NUMNODES; node++)
 		cpus_clear(node_to_cpu_mask[node]);
 
-	for(cpu = 0; cpu < NR_CPUS; ++cpu) {
+	for_each_possible_early_cpu(cpu) {
 		node = -1;
 		for (i = 0; i < NR_CPUS; ++i)
 			if (cpu_physical_id(cpu) == node_cpuid[i].phys_id) {
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 4aa9eaea76c3..6206541f9e87 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -493,6 +493,8 @@ setup_arch (char **cmdline_p)
 	acpi_table_init();
 # ifdef CONFIG_ACPI_NUMA
 	acpi_numa_init();
+	per_cpu_scan_finalize((cpus_weight(early_cpu_possible_map) == 0 ?
+		32 : cpus_weight(early_cpu_possible_map)), additional_cpus);
 # endif
 #else
 # ifdef CONFIG_SMP
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 06c540a29467..6136a4c6df11 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -104,7 +104,7 @@ static int __meminit early_nr_cpus_node(int node)
 {
 	int cpu, n = 0;
 
-	for (cpu = 0; cpu < NR_CPUS; cpu++)
+	for_each_possible_early_cpu(cpu)
 		if (node == node_cpuid[cpu].nid)
 			n++;
 
@@ -143,7 +143,7 @@ static void *per_cpu_node_setup(void *cpu_data, int node)
 #ifdef CONFIG_SMP
 	int cpu;
 
-	for (cpu = 0; cpu < NR_CPUS; cpu++) {
+	for_each_possible_early_cpu(cpu) {
 		if (node == node_cpuid[cpu].nid) {
 			memcpy(__va(cpu_data), __phys_per_cpu_start,
 			       __per_cpu_end - __per_cpu_start);
@@ -346,7 +346,7 @@ static void __init initialize_pernode_data(void)
 
 #ifdef CONFIG_SMP
 	/* Set the node_data pointer for each per-cpu struct */
-	for (cpu = 0; cpu < NR_CPUS; cpu++) {
+	for_each_possible_early_cpu(cpu) {
 		node = node_cpuid[cpu].nid;
 		per_cpu(cpu_info, cpu).node_data = mem_data[node].node_data;
 	}
@@ -494,13 +494,9 @@ void __cpuinit *per_cpu_init(void)
 	int cpu;
 	static int first_time = 1;
 
-
-	if (smp_processor_id() != 0)
-		return __per_cpu_start + __per_cpu_offset[smp_processor_id()];
-
 	if (first_time) {
 		first_time = 0;
-		for (cpu = 0; cpu < NR_CPUS; cpu++)
+		for_each_possible_early_cpu(cpu)
 			per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu];
 	}
 
diff --git a/arch/ia64/mm/numa.c b/arch/ia64/mm/numa.c
index 7807fc5c0422..b73bf1838e57 100644
--- a/arch/ia64/mm/numa.c
+++ b/arch/ia64/mm/numa.c
@@ -27,7 +27,9 @@
  */
 int num_node_memblks;
 struct node_memblk_s node_memblk[NR_NODE_MEMBLKS];
-struct node_cpuid_s node_cpuid[NR_CPUS];
+struct node_cpuid_s node_cpuid[NR_CPUS] =
+	{ [0 ... NR_CPUS-1] = { .phys_id = 0, .nid = NUMA_NO_NODE } };
+
 /*
  * This is a matrix with "distances" between nodes, they should be
  * proportional to the memory access latency ratios.
diff --git a/include/asm-ia64/acpi.h b/include/asm-ia64/acpi.h
index cd1cc39b5599..fcfad326f4c7 100644
--- a/include/asm-ia64/acpi.h
+++ b/include/asm-ia64/acpi.h
@@ -35,6 +35,7 @@
 #include <linux/init.h>
 #include <linux/numa.h>
 #include <asm/system.h>
+#include <asm/numa.h>
 
 #define COMPILER_DEPENDENT_INT64	long
 #define COMPILER_DEPENDENT_UINT64	unsigned long
@@ -115,7 +116,11 @@ extern unsigned int is_cpu_cpei_target(unsigned int cpu);
 extern void set_cpei_target_cpu(unsigned int cpu);
 extern unsigned int get_cpei_target_cpu(void);
 extern void prefill_possible_map(void);
+#ifdef CONFIG_ACPI_HOTPLUG_CPU
 extern int additional_cpus;
+#else
+#define additional_cpus 0
+#endif
 
 #ifdef CONFIG_ACPI_NUMA
 #if MAX_NUMNODES > 256
@@ -129,6 +134,34 @@ extern int __initdata nid_to_pxm_map[MAX_NUMNODES];
 
 #define acpi_unlazy_tlb(x)
 
+#ifdef CONFIG_ACPI_NUMA
+extern cpumask_t early_cpu_possible_map;
+#define for_each_possible_early_cpu(cpu)  \
+	for_each_cpu_mask((cpu), early_cpu_possible_map)
+
+static inline void per_cpu_scan_finalize(int min_cpus, int reserve_cpus)
+{
+	int low_cpu, high_cpu;
+	int cpu;
+	int next_nid = 0;
+
+	low_cpu = cpus_weight(early_cpu_possible_map);
+
+	high_cpu = max(low_cpu, min_cpus);
+	high_cpu = min(high_cpu + reserve_cpus, NR_CPUS);
+
+	for (cpu = low_cpu; cpu < high_cpu; cpu++) {
+		cpu_set(cpu, early_cpu_possible_map);
+		if (node_cpuid[cpu].nid == NUMA_NO_NODE) {
+			node_cpuid[cpu].nid = next_nid;
+			next_nid++;
+			if (next_nid >= num_online_nodes())
+				next_nid = 0;
+		}
+	}
+}
+#endif /* CONFIG_ACPI_NUMA */
+
 #endif /*__KERNEL__*/
 
 #endif /*_ASM_ACPI_H*/
diff --git a/include/asm-ia64/numa.h b/include/asm-ia64/numa.h
index 6a8a27cfae3e..3499ff57bf42 100644
--- a/include/asm-ia64/numa.h
+++ b/include/asm-ia64/numa.h
@@ -22,6 +22,8 @@
 
 #include <asm/mmzone.h>
 
+#define NUMA_NO_NODE	-1
+
 extern u16 cpu_to_node_map[NR_CPUS] __cacheline_aligned;
 extern cpumask_t node_to_cpu_mask[MAX_NUMNODES] __cacheline_aligned;
 extern pg_data_t *pgdat_list[MAX_NUMNODES];
-- 
cgit v1.2.3


From e4b05d4097eb6dab08bda86a72f6fdfdd9816395 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 9 Apr 2008 12:26:10 +0900
Subject: [IA64] pgd_offset() constfication.

when compile 2.6.25-rc8-mm1, below warning happend.
because walk_page_range pass argument as "const struct mm*",
but pgd_offset() receive as "struct mm*".

  CC      mm/pagewalk.o
mm/pagewalk.c: In function 'walk_page_range':
mm/pagewalk.c:111: warning: passing argument 1 of 'pgd_offset' discards qualifiers from pointer target type

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/pgtable.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-ia64/pgtable.h b/include/asm-ia64/pgtable.h
index e6204f14f614..ed70862ea247 100644
--- a/include/asm-ia64/pgtable.h
+++ b/include/asm-ia64/pgtable.h
@@ -371,7 +371,7 @@ pgd_index (unsigned long address)
 /* The offset in the 1-level directory is given by the 3 region bits
    (61..63) and the level-1 bits.  */
 static inline pgd_t*
-pgd_offset (struct mm_struct *mm, unsigned long address)
+pgd_offset (const struct mm_struct *mm, unsigned long address)
 {
 	return mm->pgd + pgd_index(address);
 }
-- 
cgit v1.2.3


From 34e1ceb1881ec895ad9b1b52d073f414f3aa87a9 Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Fri, 28 Mar 2008 14:27:02 -0700
Subject: [IA64] kprobes: kprobe-booster for ia64

Add kprobe-booster support on ia64.

Kprobe-booster improves the performance of kprobes by eliminating single-step,
where possible.  Currently, kprobe-booster is implemented on x86 and x86-64.
This is an ia64 port.

On ia64, kprobe-booster executes a copied bundle directly, instead of single
stepping.  Bundles which have B or X unit and which may cause an exception
(including break) are not executed directly.  And also, to prevent hitting
break exceptions on the copied bundle, only the hindmost kprobe is executed
directly if several kprobes share a bundle and are placed in different slots.
Note: set_brl_inst() is used for preparing an instruction buffer(it does not
modify any active code), so it does not need any atomic operation.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: bibo,mao <bibo.mao@intel.com>
Cc: Rusty Lynch <rusty.lynch@intel.com>
Cc: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/kprobes.c | 133 ++++++++++++++++++++++++++++++++++++---------
 include/asm-ia64/kprobes.h |   7 ++-
 2 files changed, 113 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 8d9a446a0d17..233434f4f88f 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -78,6 +78,20 @@ static enum instruction_type bundle_encoding[32][3] = {
   { u, u, u },  			/* 1F */
 };
 
+/* Insert a long branch code */
+static void __kprobes set_brl_inst(void *from, void *to)
+{
+	s64 rel = ((s64) to - (s64) from) >> 4;
+	bundle_t *brl;
+	brl = (bundle_t *) ((u64) from & ~0xf);
+	brl->quad0.template = 0x05;	/* [MLX](stop) */
+	brl->quad0.slot0 = NOP_M_INST;	/* nop.m 0x0 */
+	brl->quad0.slot1_p0 = ((rel >> 20) & 0x7fffffffff) << 2;
+	brl->quad1.slot1_p1 = (((rel >> 20) & 0x7fffffffff) << 2) >> (64 - 46);
+	/* brl.cond.sptk.many.clr rel<<4 (qp=0) */
+	brl->quad1.slot2 = BRL_INST(rel >> 59, rel & 0xfffff);
+}
+
 /*
  * In this function we check to see if the instruction
  * is IP relative instruction and update the kprobe
@@ -496,6 +510,77 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 	regs->b0 = ((struct fnptr *)kretprobe_trampoline)->ip;
 }
 
+/* Check the instruction in the slot is break */
+static int __kprobes __is_ia64_break_inst(bundle_t *bundle, uint slot)
+{
+	unsigned int major_opcode;
+	unsigned int template = bundle->quad0.template;
+	unsigned long kprobe_inst;
+
+	/* Move to slot 2, if bundle is MLX type and kprobe slot is 1 */
+	if (slot == 1 && bundle_encoding[template][1] == L)
+		slot++;
+
+	/* Get Kprobe probe instruction at given slot*/
+	get_kprobe_inst(bundle, slot, &kprobe_inst, &major_opcode);
+
+	/* For break instruction,
+	 * Bits 37:40 Major opcode to be zero
+	 * Bits 27:32 X6 to be zero
+	 * Bits 32:35 X3 to be zero
+	 */
+	if (major_opcode || ((kprobe_inst >> 27) & 0x1FF)) {
+		/* Not a break instruction */
+		return 0;
+	}
+
+	/* Is a break instruction */
+	return 1;
+}
+
+/*
+ * In this function, we check whether the target bundle modifies IP or
+ * it triggers an exception. If so, it cannot be boostable.
+ */
+static int __kprobes can_boost(bundle_t *bundle, uint slot,
+			       unsigned long bundle_addr)
+{
+	unsigned int template = bundle->quad0.template;
+
+	do {
+		if (search_exception_tables(bundle_addr + slot) ||
+		    __is_ia64_break_inst(bundle, slot))
+			return 0;	/* exception may occur in this bundle*/
+	} while ((++slot) < 3);
+	template &= 0x1e;
+	if (template >= 0x10 /* including B unit */ ||
+	    template == 0x04 /* including X unit */ ||
+	    template == 0x06) /* undefined */
+		return 0;
+
+	return 1;
+}
+
+/* Prepare long jump bundle and disables other boosters if need */
+static void __kprobes prepare_booster(struct kprobe *p)
+{
+	unsigned long addr = (unsigned long)p->addr & ~0xFULL;
+	unsigned int slot = (unsigned long)p->addr & 0xf;
+	struct kprobe *other_kp;
+
+	if (can_boost(&p->ainsn.insn[0].bundle, slot, addr)) {
+		set_brl_inst(&p->ainsn.insn[1].bundle, (bundle_t *)addr + 1);
+		p->ainsn.inst_flag |= INST_FLAG_BOOSTABLE;
+	}
+
+	/* disables boosters in previous slots */
+	for (; addr < (unsigned long)p->addr; addr++) {
+		other_kp = get_kprobe((void *)addr);
+		if (other_kp)
+			other_kp->ainsn.inst_flag &= ~INST_FLAG_BOOSTABLE;
+	}
+}
+
 int __kprobes arch_prepare_kprobe(struct kprobe *p)
 {
 	unsigned long addr = (unsigned long) p->addr;
@@ -530,6 +615,8 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 
 	prepare_break_inst(template, slot, major_opcode, kprobe_inst, p, qp);
 
+	prepare_booster(p);
+
 	return 0;
 }
 
@@ -543,7 +630,9 @@ void __kprobes arch_arm_kprobe(struct kprobe *p)
 	src = &p->opcode.bundle;
 
 	flush_icache_range((unsigned long)p->ainsn.insn,
-			(unsigned long)p->ainsn.insn + sizeof(kprobe_opcode_t));
+			   (unsigned long)p->ainsn.insn +
+			   sizeof(kprobe_opcode_t) * MAX_INSN_SIZE);
+
 	switch (p->ainsn.slot) {
 		case 0:
 			dest->quad0.slot0 = src->quad0.slot0;
@@ -584,13 +673,13 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
 	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, 0);
+	free_insn_slot(p->ainsn.insn, p->ainsn.inst_flag & INST_FLAG_BOOSTABLE);
 	mutex_unlock(&kprobe_mutex);
 }
 /*
  * We are resuming execution after a single step fault, so the pt_regs
  * structure reflects the register state after we executed the instruction
- * located in the kprobe (p->ainsn.insn.bundle).  We still need to adjust
+ * located in the kprobe (p->ainsn.insn->bundle).  We still need to adjust
  * the ip to point back to the original stack address. To set the IP address
  * to original stack address, handle the case where we need to fixup the
  * relative IP address and/or fixup branch register.
@@ -607,7 +696,7 @@ static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
 	if (slot == 1 && bundle_encoding[template][1] == L)
 		slot = 2;
 
-	if (p->ainsn.inst_flag) {
+	if (p->ainsn.inst_flag & ~INST_FLAG_BOOSTABLE) {
 
 		if (p->ainsn.inst_flag & INST_FLAG_FIX_RELATIVE_IP_ADDR) {
 			/* Fix relative IP address */
@@ -686,33 +775,12 @@ static void __kprobes prepare_ss(struct kprobe *p, struct pt_regs *regs)
 static int __kprobes is_ia64_break_inst(struct pt_regs *regs)
 {
 	unsigned int slot = ia64_psr(regs)->ri;
-	unsigned int template, major_opcode;
-	unsigned long kprobe_inst;
 	unsigned long *kprobe_addr = (unsigned long *)regs->cr_iip;
 	bundle_t bundle;
 
 	memcpy(&bundle, kprobe_addr, sizeof(bundle_t));
-	template = bundle.quad0.template;
-
-	/* Move to slot 2, if bundle is MLX type and kprobe slot is 1 */
-	if (slot == 1 && bundle_encoding[template][1] == L)
-		slot++;
 
-	/* Get Kprobe probe instruction at given slot*/
-	get_kprobe_inst(&bundle, slot, &kprobe_inst, &major_opcode);
-
-	/* For break instruction,
-	 * Bits 37:40 Major opcode to be zero
-	 * Bits 27:32 X6 to be zero
-	 * Bits 32:35 X3 to be zero
-	 */
-	if (major_opcode || ((kprobe_inst >> 27) & 0x1FF) ) {
-		/* Not a break instruction */
-		return 0;
-	}
-
-	/* Is a break instruction */
-	return 1;
+	return __is_ia64_break_inst(&bundle, slot);
 }
 
 static int __kprobes pre_kprobes_handler(struct die_args *args)
@@ -802,6 +870,19 @@ static int __kprobes pre_kprobes_handler(struct die_args *args)
 		return 1;
 
 ss_probe:
+#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM)
+	if (p->ainsn.inst_flag == INST_FLAG_BOOSTABLE && !p->post_handler) {
+		/* Boost up -- we can execute copied instructions directly */
+		ia64_psr(regs)->ri = p->ainsn.slot;
+		regs->cr_iip = (unsigned long)&p->ainsn.insn->bundle & ~0xFULL;
+		/* turn single stepping off */
+		ia64_psr(regs)->ss = 0;
+
+		reset_current_kprobe();
+		preempt_enable_no_resched();
+		return 1;
+	}
+#endif
 	prepare_ss(p, regs);
 	kcb->kprobe_status = KPROBE_HIT_SS;
 	return 1;
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index d03bf9ff68e3..ef71b57fc2f4 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -30,8 +30,12 @@
 #include <asm/break.h>
 
 #define __ARCH_WANT_KPROBES_INSN_SLOT
-#define MAX_INSN_SIZE   1
+#define MAX_INSN_SIZE   2	/* last half is for kprobe-booster */
 #define BREAK_INST	(long)(__IA64_BREAK_KPROBE << 6)
+#define NOP_M_INST	(long)(1<<27)
+#define BRL_INST(i1, i2) ((long)((0xcL << 37) |	/* brl */ \
+				(0x1L << 12) |	/* many */ \
+				(((i1) & 1) << 36) | ((i2) << 13))) /* imm */
 
 typedef union cmp_inst {
 	struct {
@@ -112,6 +116,7 @@ struct arch_specific_insn {
  #define INST_FLAG_FIX_RELATIVE_IP_ADDR		1
  #define INST_FLAG_FIX_BRANCH_REG		2
  #define INST_FLAG_BREAK_INST			4
+ #define INST_FLAG_BOOSTABLE			8
  	unsigned long inst_flag;
  	unsigned short target_br_reg;
 	unsigned short slot;
-- 
cgit v1.2.3


From c19b2930df0621500913c005c06978bd8933110b Mon Sep 17 00:00:00 2001
From: Russ Anderson <rja@sgi.com>
Date: Fri, 29 Feb 2008 17:14:44 -0600
Subject: [IA64] Itanium Spec updates
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Updates based on the "Intel® Itanium® Architecture Software Developer's Manual
Specification Update October 2007".

http://download.intel.com/design/itanium/specupdt/24869911.pdf

Signed-off-by: Russ Anderson <rja@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/pal.h | 72 +++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 63 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h
index 8a695d3407d2..67b02901ead4 100644
--- a/include/asm-ia64/pal.h
+++ b/include/asm-ia64/pal.h
@@ -13,6 +13,7 @@
  * Copyright (C) 1999 VA Linux Systems
  * Copyright (C) 1999 Walt Drummond <drummond@valinux.com>
  * Copyright (C) 1999 Srinivasa Prasad Thirumalachar <sprasad@sprasad.engr.sgi.com>
+ * Copyright (C) 2008 Silicon Graphics, Inc. (SGI)
  *
  * 99/10/01	davidm	Make sure we pass zero for reserved parameters.
  * 00/03/07	davidm	Updated pal_cache_flush() to be in sync with PAL v2.6.
@@ -73,6 +74,8 @@
 #define PAL_CACHE_SHARED_INFO	43	/* returns information on caches shared by logical processor */
 #define PAL_GET_HW_POLICY	48	/* Get current hardware resource sharing policy */
 #define PAL_SET_HW_POLICY	49	/* Set current hardware resource sharing policy */
+#define PAL_VP_INFO		50	/* Information about virtual processor features */
+#define PAL_MC_HW_TRACKING	51	/* Hardware tracking status */
 
 #define PAL_COPY_PAL		256	/* relocate PAL procedures and PAL PMI */
 #define PAL_HALT_INFO		257	/* return the low power capabilities of processor */
@@ -504,7 +507,8 @@ typedef struct pal_cache_check_info_s {
 			wiv		: 1,	/* Way field valid */
 			reserved2	: 1,
 			dp		: 1,	/* Data poisoned on MBE */
-			reserved3	: 8,
+			reserved3	: 6,
+			hlth		: 2,	/* Health indicator */
 
 			index		: 20,	/* Cache line index */
 			reserved4	: 2,
@@ -542,7 +546,9 @@ typedef struct pal_tlb_check_info_s {
 			dtc		: 1,	/* Fail in data TC */
 			itc		: 1,	/* Fail in inst. TC */
 			op		: 4,	/* Cache operation */
-			reserved3	: 30,
+			reserved3	: 6,
+			hlth		: 2,	/* Health indicator */
+			reserved4	: 22,
 
 			is		: 1,	/* instruction set (1 == ia32) */
 			iv		: 1,	/* instruction set field valid */
@@ -633,7 +639,8 @@ typedef struct pal_uarch_check_info_s {
 			way		: 6,	/* Way of structure */
 			wv		: 1,	/* way valid */
 			xv		: 1,	/* index valid */
-			reserved1	: 8,
+			reserved1	: 6,
+			hlth		: 2,	/* Health indicator */
 			index		: 8,	/* Index or set of the uarch
 						 * structure that failed.
 						 */
@@ -1213,14 +1220,12 @@ ia64_pal_mc_drain (void)
 
 /* Return the machine check dynamic processor state */
 static inline s64
-ia64_pal_mc_dynamic_state (u64 offset, u64 *size, u64 *pds)
+ia64_pal_mc_dynamic_state (u64 info_type, u64 dy_buffer, u64 *size)
 {
 	struct ia64_pal_retval iprv;
-	PAL_CALL(iprv, PAL_MC_DYNAMIC_STATE, offset, 0, 0);
+	PAL_CALL(iprv, PAL_MC_DYNAMIC_STATE, info_type, dy_buffer, 0);
 	if (size)
 		*size = iprv.v0;
-	if (pds)
-		*pds = iprv.v1;
 	return iprv.status;
 }
 
@@ -1281,15 +1286,41 @@ ia64_pal_mc_expected (u64 expected, u64 *previous)
 	return iprv.status;
 }
 
+typedef union pal_hw_tracking_u {
+	u64			pht_data;
+	struct {
+		u64		itc	:4,	/* Instruction cache tracking */
+				dct	:4,	/* Date cache tracking */
+				itt	:4,	/* Instruction TLB tracking */
+				ddt	:4,	/* Data TLB tracking */
+				reserved:48;
+	} pal_hw_tracking_s;
+} pal_hw_tracking_u_t;
+
+/*
+ * Hardware tracking status.
+ */
+static inline s64
+ia64_pal_mc_hw_tracking (u64 *status)
+{
+	struct ia64_pal_retval iprv;
+	PAL_CALL(iprv, PAL_MC_HW_TRACKING, 0, 0, 0);
+	if (status)
+		*status = iprv.v0;
+	return iprv.status;
+}
+
 /* Register a platform dependent location with PAL to which it can save
  * minimal processor state in the event of a machine check or initialization
  * event.
  */
 static inline s64
-ia64_pal_mc_register_mem (u64 physical_addr)
+ia64_pal_mc_register_mem (u64 physical_addr, u64 size, u64 *req_size)
 {
 	struct ia64_pal_retval iprv;
-	PAL_CALL(iprv, PAL_MC_REGISTER_MEM, physical_addr, 0, 0);
+	PAL_CALL(iprv, PAL_MC_REGISTER_MEM, physical_addr, size, 0);
+	if (req_size)
+		*req_size = iprv.v0;
 	return iprv.status;
 }
 
@@ -1631,6 +1662,29 @@ ia64_pal_vm_summary (pal_vm_info_1_u_t *vm_info_1, pal_vm_info_2_u_t *vm_info_2)
 	return iprv.status;
 }
 
+typedef union pal_vp_info_u {
+	u64			pvi_val;
+	struct {
+		u64		index:		48,	/* virtual feature set info */
+				vmm_id:		16;	/* feature set id */
+	} pal_vp_info_s;
+} pal_vp_info_u_t;
+
+/*
+ * Returns infomation about virtual processor features
+ */
+static inline s64
+ia64_pal_vp_info (u64 feature_set, u64 vp_buffer, u64 *vp_info, u64 *vmm_id)
+{
+	struct ia64_pal_retval iprv;
+	PAL_CALL(iprv, PAL_VP_INFO, feature_set, vp_buffer, 0);
+	if (vp_info)
+		*vp_info = iprv.v0;
+	if (vmm_id)
+		*vmm_id = iprv.v1;
+	return iprv.status;
+}
+
 typedef union pal_itr_valid_u {
 	u64			piv_val;
 	struct {
-- 
cgit v1.2.3


From b715631fad3ed320b85d386a84a6fb0b3f86b0b9 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Thu, 10 Apr 2008 01:33:47 -0700
Subject: socket: sk_filter minor cleanups

Some minor style cleanups:
  * Move __KERNEL__ definitions to one place in filter.h
  * Use const for sk_filter_len
  * Line wrapping
  * Put EXPORT_SYMBOL next to function definition

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/filter.h | 31 +++++++++++++++----------------
 net/core/filter.c      |  5 ++---
 2 files changed, 17 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/linux/filter.h b/include/linux/filter.h
index ddfa0372a3b7..bfc5d319b946 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -37,21 +37,6 @@ struct sock_fprog	/* Required for SO_ATTACH_FILTER. */
 	struct sock_filter __user *filter;
 };
 
-#ifdef __KERNEL__
-struct sk_filter
-{
-	atomic_t		refcnt;
-	unsigned int         	len;	/* Number of filter blocks */
-	struct rcu_head		rcu;
-	struct sock_filter     	insns[0];
-};
-
-static inline unsigned int sk_filter_len(struct sk_filter *fp)
-{
-	return fp->len*sizeof(struct sock_filter) + sizeof(*fp);
-}
-#endif
-
 /*
  * Instruction classes
  */
@@ -141,10 +126,24 @@ static inline unsigned int sk_filter_len(struct sk_filter *fp)
 #define SKF_LL_OFF    (-0x200000)
 
 #ifdef __KERNEL__
+struct sk_filter
+{
+	atomic_t		refcnt;
+	unsigned int         	len;	/* Number of filter blocks */
+	struct rcu_head		rcu;
+	struct sock_filter     	insns[0];
+};
+
+static inline unsigned int sk_filter_len(const struct sk_filter *fp)
+{
+	return fp->len * sizeof(struct sock_filter) + sizeof(*fp);
+}
+
 struct sk_buff;
 struct sock;
 
-extern unsigned int sk_run_filter(struct sk_buff *skb, struct sock_filter *filter, int flen);
+extern unsigned int sk_run_filter(struct sk_buff *skb,
+				  struct sock_filter *filter, int flen);
 extern int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk);
 extern int sk_detach_filter(struct sock *sk);
 extern int sk_chk_filter(struct sock_filter *filter, int flen);
diff --git a/net/core/filter.c b/net/core/filter.c
index e0a06942c025..85a5febab567 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -275,6 +275,7 @@ load_b:
 
 	return 0;
 }
+EXPORT_SYMBOL(sk_run_filter);
 
 /**
  *	sk_chk_filter - verify socket filter code
@@ -385,6 +386,7 @@ int sk_chk_filter(struct sock_filter *filter, int flen)
 
 	return (BPF_CLASS(filter[flen - 1].code) == BPF_RET) ? 0 : -EINVAL;
 }
+EXPORT_SYMBOL(sk_chk_filter);
 
 /**
  * 	sk_filter_rcu_release: Release a socket filter by rcu_head
@@ -467,6 +469,3 @@ int sk_detach_filter(struct sock *sk)
 	rcu_read_unlock_bh();
 	return ret;
 }
-
-EXPORT_SYMBOL(sk_chk_filter);
-EXPORT_SYMBOL(sk_run_filter);
-- 
cgit v1.2.3


From 43db6d65e0ef943a361cb91f8baa49132009227b Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Thu, 10 Apr 2008 01:43:09 -0700
Subject: socket: sk_filter deinline

The sk_filter function is too big to be inlined. This saves 2296 bytes
of text on allyesconfig.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/filter.h |  1 +
 include/net/sock.h     | 35 -----------------------------------
 net/core/filter.c      | 35 +++++++++++++++++++++++++++++++++++
 3 files changed, 36 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/linux/filter.h b/include/linux/filter.h
index bfc5d319b946..673e5677ebcc 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -142,6 +142,7 @@ static inline unsigned int sk_filter_len(const struct sk_filter *fp)
 struct sk_buff;
 struct sock;
 
+extern int sk_filter(struct sock *sk, struct sk_buff *skb);
 extern unsigned int sk_run_filter(struct sk_buff *skb,
 				  struct sock_filter *filter, int flen);
 extern int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk);
diff --git a/include/net/sock.h b/include/net/sock.h
index f4fdd101c9a2..09255eae93e9 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -927,41 +927,6 @@ extern void sk_common_release(struct sock *sk);
 /* Initialise core socket variables */
 extern void sock_init_data(struct socket *sock, struct sock *sk);
 
-/**
- *	sk_filter - run a packet through a socket filter
- *	@sk: sock associated with &sk_buff
- *	@skb: buffer to filter
- *	@needlock: set to 1 if the sock is not locked by caller.
- *
- * Run the filter code and then cut skb->data to correct size returned by
- * sk_run_filter. If pkt_len is 0 we toss packet. If skb->len is smaller
- * than pkt_len we keep whole skb->data. This is the socket level
- * wrapper to sk_run_filter. It returns 0 if the packet should
- * be accepted or -EPERM if the packet should be tossed.
- *
- */
-
-static inline int sk_filter(struct sock *sk, struct sk_buff *skb)
-{
-	int err;
-	struct sk_filter *filter;
-	
-	err = security_sock_rcv_skb(sk, skb);
-	if (err)
-		return err;
-	
-	rcu_read_lock_bh();
-	filter = rcu_dereference(sk->sk_filter);
-	if (filter) {
-		unsigned int pkt_len = sk_run_filter(skb, filter->insns,
-				filter->len);
-		err = pkt_len ? pskb_trim(skb, pkt_len) : -EPERM;
-	}
- 	rcu_read_unlock_bh();
-
-	return err;
-}
-
 /**
  *	sk_filter_release: Release a socket filter
  *	@sk: socket
diff --git a/net/core/filter.c b/net/core/filter.c
index 85a5febab567..bbb53c69857c 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -63,6 +63,41 @@ static inline void *load_pointer(struct sk_buff *skb, int k,
 	}
 }
 
+/**
+ *	sk_filter - run a packet through a socket filter
+ *	@sk: sock associated with &sk_buff
+ *	@skb: buffer to filter
+ *	@needlock: set to 1 if the sock is not locked by caller.
+ *
+ * Run the filter code and then cut skb->data to correct size returned by
+ * sk_run_filter. If pkt_len is 0 we toss packet. If skb->len is smaller
+ * than pkt_len we keep whole skb->data. This is the socket level
+ * wrapper to sk_run_filter. It returns 0 if the packet should
+ * be accepted or -EPERM if the packet should be tossed.
+ *
+ */
+int sk_filter(struct sock *sk, struct sk_buff *skb)
+{
+	int err;
+	struct sk_filter *filter;
+
+	err = security_sock_rcv_skb(sk, skb);
+	if (err)
+		return err;
+
+	rcu_read_lock_bh();
+	filter = rcu_dereference(sk->sk_filter);
+	if (filter) {
+		unsigned int pkt_len = sk_run_filter(skb, filter->insns,
+				filter->len);
+		err = pkt_len ? pskb_trim(skb, pkt_len) : -EPERM;
+	}
+	rcu_read_unlock_bh();
+
+	return err;
+}
+EXPORT_SYMBOL(sk_filter);
+
 /**
  *	sk_run_filter - run a filter on a socket
  *	@skb: buffer to run the filter on
-- 
cgit v1.2.3


From 3cccd6078413e9707f0ef3652b4e6e9cb84e9fa0 Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Thu, 10 Apr 2008 02:01:21 -0700
Subject: [IPV6] Remove three method declarations in include/net/ndisc.h.

This patch removes two unused method declarations in
include/net/ndisc.h: ndisc_forwarding_on(void) and
ndisc_forwarding_off(void);

Also igmp6_cleanup(void) appears twice in this header, so one
igmp6_cleanup(void) declaration is removed.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ndisc.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index 9f2bae68d28c..16424236fe2f 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -102,9 +102,6 @@ extern void			ndisc_send_rs(struct net_device *dev,
 					      struct in6_addr *saddr,
 					      struct in6_addr *daddr);
 
-extern void			ndisc_forwarding_on(void);
-extern void			ndisc_forwarding_off(void);
-
 extern void			ndisc_send_redirect(struct sk_buff *skb,
 						    struct neighbour *neigh,
 						    struct in6_addr *target);
@@ -124,7 +121,6 @@ extern int			igmp6_event_query(struct sk_buff *skb);
 
 extern int			igmp6_event_report(struct sk_buff *skb);
 
-extern void			igmp6_cleanup(void);
 
 #ifdef CONFIG_SYSCTL
 extern int 			ndisc_ifinfo_sysctl_change(struct ctl_table *ctl,
-- 
cgit v1.2.3


From 4738c1db1593687713869fa69e733eebc7b0d6d8 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 10 Apr 2008 02:02:28 -0700
Subject: [SKFILTER]: Add SKF_ADF_NLATTR instruction

SKF_ADF_NLATTR searches for a netlink attribute, which avoids manually
parsing and walking attributes. It takes the offset at which to start
searching in the 'A' register and the attribute type in the 'X' register
and returns the offset in the 'A' register. When the attribute is not
found it returns zero.

A top-level attribute can be located using a filter like this
(example for nfnetlink, using struct nfgenmsg):

	...
	{
		/* A = offset of first attribute */
		.code	= BPF_LD | BPF_IMM,
		.k	= sizeof(struct nlmsghdr) + sizeof(struct nfgenmsg)
	},
	{
		/* X = CTA_PROTOINFO */
		.code	= BPF_LDX | BPF_IMM,
		.k	= CTA_PROTOINFO,
	},
	{
		/* A = netlink attribute offset */
		.code	= BPF_LD | BPF_B | BPF_ABS,
		.k	= SKF_AD_OFF + SKF_AD_NLATTR
	},
	{
		/* Exit if not found */
		.code   = BPF_JMP | BPF_JEQ | BPF_K,
		.k	= 0,
		.jt	= <error>
	},
	...

A nested attribute below the CTA_PROTOINFO attribute would then
be parsed like this:

	...
	{
		/* A += sizeof(struct nlattr) */
		.code	= BPF_ALU | BPF_ADD | BPF_K,
		.k	= sizeof(struct nlattr),
	},
	{
		/* X = CTA_PROTOINFO_TCP */
		.code	= BPF_LDX | BPF_IMM,
		.k	= CTA_PROTOINFO_TCP,
	},
	{
		/* A = netlink attribute offset */
		.code	= BPF_LD | BPF_B | BPF_ABS,
		.k	= SKF_AD_OFF + SKF_AD_NLATTR
	},
	...

The data of an attribute can be loaded into 'A' like this:

	...
	{
		/* X = A (attribute offset) */
		.code	= BPF_MISC | BPF_TAX,
	},
	{
		/* A = skb->data[X + k] */
		.code 	= BPF_LD | BPF_B | BPF_IND,
		.k	= sizeof(struct nlattr),
	},
	...

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/filter.h |  3 ++-
 net/core/filter.c      | 17 +++++++++++++++++
 2 files changed, 19 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 673e5677ebcc..b6ea9aa9e853 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -121,7 +121,8 @@ struct sock_fprog	/* Required for SO_ATTACH_FILTER. */
 #define SKF_AD_PROTOCOL 0
 #define SKF_AD_PKTTYPE 	4
 #define SKF_AD_IFINDEX 	8
-#define SKF_AD_MAX 	12
+#define SKF_AD_NLATTR	12
+#define SKF_AD_MAX 	16
 #define SKF_NET_OFF   (-0x100000)
 #define SKF_LL_OFF    (-0x200000)
 
diff --git a/net/core/filter.c b/net/core/filter.c
index bbb53c69857c..f5f3cf603064 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -27,6 +27,7 @@
 #include <linux/if_packet.h>
 #include <net/ip.h>
 #include <net/protocol.h>
+#include <net/netlink.h>
 #include <linux/skbuff.h>
 #include <net/sock.h>
 #include <linux/errno.h>
@@ -303,6 +304,22 @@ load_b:
 		case SKF_AD_IFINDEX:
 			A = skb->dev->ifindex;
 			continue;
+		case SKF_AD_NLATTR: {
+			struct nlattr *nla;
+
+			if (skb_is_nonlinear(skb))
+				return 0;
+			if (A > skb->len - sizeof(struct nlattr))
+				return 0;
+
+			nla = nla_find((struct nlattr *)&skb->data[A],
+				       skb->len - A, X);
+			if (nla)
+				A = (void *)nla - (void *)skb->data;
+			else
+				A = 0;
+			continue;
+		}
 		default:
 			return 0;
 		}
-- 
cgit v1.2.3


From 5c06f510a25153ff79e8c2dca312b732a367c5bb Mon Sep 17 00:00:00 2001
From: Rami Rosen <ramirose@gmail.com>
Date: Thu, 10 Apr 2008 02:31:20 -0700
Subject: [IPV6]: Remove unused declarations in include/net/ip6_route.h.

1) Standlaone ip6_null_entry is no longer needed as it is replaced by
   the ip6_null_entry member of ipv6 (instance of struct netns_ipv6) in
   struct net (as a result of Network Namespaces patches).


2) These 3 methods from this same header are not defined anywhere:
   ip6_rt_addr_add(), ip6_rt_addr_del(), rt6_sndmsg()

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h | 15 ---------------
 1 file changed, 15 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 3ae6799c2b14..9080076ce0e5 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -37,7 +37,6 @@ struct route_info {
 #define RT6_LOOKUP_F_SRCPREF_PUBLIC	0x00000010
 #define RT6_LOOKUP_F_SRCPREF_COA	0x00000020
 
-extern struct rt6_info	*ip6_null_entry;
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 extern struct rt6_info	*ip6_prohibit_entry;
@@ -61,20 +60,6 @@ extern int			ip6_route_add(struct fib6_config *cfg);
 extern int			ip6_ins_rt(struct rt6_info *);
 extern int			ip6_del_rt(struct rt6_info *);
 
-extern int			ip6_rt_addr_add(struct in6_addr *addr,
-						struct net_device *dev,
-						int anycast);
-
-extern int			ip6_rt_addr_del(struct in6_addr *addr,
-						struct net_device *dev);
-
-extern void			rt6_sndmsg(int type, struct in6_addr *dst,
-					   struct in6_addr *src,
-					   struct in6_addr *gw,
-					   struct net_device *dev, 
-					   int dstlen, int srclen,
-					   int metric, __u32 flags);
-
 extern struct rt6_info		*rt6_lookup(struct net *net,
 					    struct in6_addr *daddr,
 					    struct in6_addr *saddr,
-- 
cgit v1.2.3


From 4dfc2817025965a2fc78a18c50f540736a6b5c24 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 10 Apr 2008 03:12:40 -0700
Subject: [Syncookies]: Add support for TCP options via timestamps.

Allow the use of SACK and window scaling when syncookies are used
and the client supports tcp timestamps. Options are encoded into
the timestamp sent in the syn-ack and restored from the timestamp
echo when the ack is received.

Based on earlier work by Glenn Griffin.
This patch avoids increasing the size of structs by encoding TCP
options into the least significant bits of the timestamp and
by not using any 'timestamp offset'.

The downside is that the timestamp sent in the packet after the synack
will increase by several seconds.

changes since v1:
 don't duplicate timestamp echo decoding function, put it into ipv4/syncookie.c
 and have ipv6/syncookies.c use it.
 Feedback from Glenn Griffin: fix line indented with spaces, kill redundant if ()

Reviewed-by: Hagen Paul Pfeifer <hagen@jauu.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/request_sock.h |  2 +-
 include/net/tcp.h          |  4 +++
 net/ipv4/syncookies.c      | 89 +++++++++++++++++++++++++++++++++++++++++++---
 net/ipv4/tcp_ipv4.c        |  5 ++-
 net/ipv4/tcp_output.c      |  6 +++-
 net/ipv6/syncookies.c      | 20 ++++++++---
 net/ipv6/tcp_ipv6.c        |  5 ++-
 7 files changed, 114 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index 0369f98e9f3a..b220b5f624de 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -45,7 +45,7 @@ struct request_sock {
 	struct request_sock		*dl_next; /* Must be first member! */
 	u16				mss;
 	u8				retrans;
-	u8				__pad;
+	u8				cookie_ts; /* syncookie: encode tcpopts in timestamp */
 	/* The following two fields can be easily recomputed I think -AK */
 	u32				window_clamp; /* window clamp at creation time */
 	u32				rcv_wnd;	  /* rcv_wnd offered first time */
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 723b36851dde..7b41bb962b9c 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -442,6 +442,9 @@ extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, 
 				     __u16 *mss);
 
+extern __u32 cookie_init_timestamp(struct request_sock *req);
+extern void cookie_check_timestamp(struct tcp_options_received *tcp_opt);
+
 /* From net/ipv6/syncookies.c */
 extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb);
 extern __u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb,
@@ -956,6 +959,7 @@ static inline void tcp_openreq_init(struct request_sock *req,
 	struct inet_request_sock *ireq = inet_rsk(req);
 
 	req->rcv_wnd = 0;		/* So that tcp_send_synack() knows! */
+	req->cookie_ts = 0;
 	tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq;
 	req->mss = rx_opt->mss_clamp;
 	req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0;
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index abc752d45cf7..73ba98921d64 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -19,6 +19,10 @@
 #include <linux/kernel.h>
 #include <net/tcp.h>
 
+/* Timestamps: lowest 9 bits store TCP options */
+#define TSBITS 9
+#define TSMASK (((__u32)1 << TSBITS) - 1)
+
 extern int sysctl_tcp_syncookies;
 
 __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS];
@@ -51,6 +55,39 @@ static u32 cookie_hash(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport,
 	return tmp[17];
 }
 
+
+/*
+ * when syncookies are in effect and tcp timestamps are enabled we encode
+ * tcp options in the lowest 9 bits of the timestamp value that will be
+ * sent in the syn-ack.
+ * Since subsequent timestamps use the normal tcp_time_stamp value, we
+ * must make sure that the resulting initial timestamp is <= tcp_time_stamp.
+ */
+__u32 cookie_init_timestamp(struct request_sock *req)
+{
+	struct inet_request_sock *ireq;
+	u32 ts, ts_now = tcp_time_stamp;
+	u32 options = 0;
+
+	ireq = inet_rsk(req);
+	if (ireq->wscale_ok) {
+		options = ireq->snd_wscale;
+		options |= ireq->rcv_wscale << 4;
+	}
+	options |= ireq->sack_ok << 8;
+
+	ts = ts_now & ~TSMASK;
+	ts |= options;
+	if (ts > ts_now) {
+		ts >>= TSBITS;
+		ts--;
+		ts <<= TSBITS;
+		ts |= options;
+	}
+	return ts;
+}
+
+
 static __u32 secure_tcp_syn_cookie(__be32 saddr, __be32 daddr, __be16 sport,
 				   __be16 dport, __u32 sseq, __u32 count,
 				   __u32 data)
@@ -185,6 +222,35 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb,
 	return child;
 }
 
+
+/*
+ * when syncookies are in effect and tcp timestamps are enabled we stored
+ * additional tcp options in the timestamp.
+ * This extracts these options from the timestamp echo.
+ *
+ * The lowest 4 bits are for snd_wscale
+ * The next 4 lsb are for rcv_wscale
+ * The next lsb is for sack_ok
+ */
+void cookie_check_timestamp(struct tcp_options_received *tcp_opt)
+{
+	/* echoed timestamp, 9 lowest bits contain options */
+	u32 options = tcp_opt->rcv_tsecr & TSMASK;
+
+	tcp_opt->snd_wscale = options & 0xf;
+	options >>= 4;
+	tcp_opt->rcv_wscale = options & 0xf;
+
+	tcp_opt->sack_ok = (options >> 4) & 0x1;
+
+	if (tcp_opt->sack_ok)
+		tcp_sack_reset(tcp_opt);
+
+	if (tcp_opt->snd_wscale || tcp_opt->rcv_wscale)
+		tcp_opt->wscale_ok = 1;
+}
+EXPORT_SYMBOL(cookie_check_timestamp);
+
 struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 			     struct ip_options *opt)
 {
@@ -198,6 +264,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 	int mss;
 	struct rtable *rt;
 	__u8 rcv_wscale;
+	struct tcp_options_received tcp_opt;
 
 	if (!sysctl_tcp_syncookies || !th->ack)
 		goto out;
@@ -210,6 +277,13 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 
 	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV);
 
+	/* check for timestamp cookie support */
+	memset(&tcp_opt, 0, sizeof(tcp_opt));
+	tcp_parse_options(skb, &tcp_opt, 0);
+
+	if (tcp_opt.saw_tstamp)
+		cookie_check_timestamp(&tcp_opt);
+
 	ret = NULL;
 	req = reqsk_alloc(&tcp_request_sock_ops); /* for safety */
 	if (!req)
@@ -228,6 +302,12 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 	ireq->loc_addr		= ip_hdr(skb)->daddr;
 	ireq->rmt_addr		= ip_hdr(skb)->saddr;
 	ireq->opt		= NULL;
+	ireq->snd_wscale	= tcp_opt.snd_wscale;
+	ireq->rcv_wscale	= tcp_opt.rcv_wscale;
+	ireq->sack_ok		= tcp_opt.sack_ok;
+	ireq->wscale_ok		= tcp_opt.wscale_ok;
+	ireq->tstamp_ok		= tcp_opt.saw_tstamp;
+	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
 
 	/* We throwed the options of the initial SYN away, so we hope
 	 * the ACK carries the same options again (see RFC1122 4.2.3.8)
@@ -242,8 +322,6 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 		}
 	}
 
-	ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0;
-	ireq->wscale_ok	 = ireq->sack_ok = 0;
 	req->expires	= 0UL;
 	req->retrans	= 0;
 
@@ -272,11 +350,12 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 	}
 
 	/* Try to redo what tcp_v4_send_synack did. */
-	req->window_clamp = dst_metric(&rt->u.dst, RTAX_WINDOW);
+	req->window_clamp = tp->window_clamp ? :dst_metric(&rt->u.dst, RTAX_WINDOW);
+
 	tcp_select_initial_window(tcp_full_space(sk), req->mss,
 				  &req->rcv_wnd, &req->window_clamp,
-				  0, &rcv_wscale);
-	/* BTW win scale with syncookies is 0 by definition */
+				  ireq->wscale_ok, &rcv_wscale);
+
 	ireq->rcv_wscale  = rcv_wscale;
 
 	ret = get_cookie_sock(sk, skb, req, &rt->u.dst);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index df89a566a5a1..52e3ae603ca9 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1299,10 +1299,8 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	tcp_parse_options(skb, &tmp_opt, 0);
 
-	if (want_cookie) {
+	if (want_cookie && !tmp_opt.saw_tstamp)
 		tcp_clear_options(&tmp_opt);
-		tmp_opt.saw_tstamp = 0;
-	}
 
 	if (tmp_opt.saw_tstamp && !tmp_opt.rcv_tsval) {
 		/* Some OSes (unknown ones, but I see them on web server, which
@@ -1330,6 +1328,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	if (want_cookie) {
 #ifdef CONFIG_SYN_COOKIES
 		syn_flood_warning(skb);
+		req->cookie_ts = tmp_opt.tstamp_ok;
 #endif
 		isn = cookie_v4_init_sequence(sk, skb, &req->mss);
 	} else if (!isn) {
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index a627616314ba..76b3653e9b4c 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2233,7 +2233,11 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 
 	/* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */
 	th->window = htons(min(req->rcv_wnd, 65535U));
-
+#ifdef CONFIG_SYN_COOKIES
+	if (unlikely(req->cookie_ts))
+		TCP_SKB_CB(skb)->when = cookie_init_timestamp(req);
+	else
+#endif
 	TCP_SKB_CB(skb)->when = tcp_time_stamp;
 	tcp_syn_build_options((__be32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok,
 			      ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale,
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 3a622e7abc02..938ce4ecde55 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -170,6 +170,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	int mss;
 	struct dst_entry *dst;
 	__u8 rcv_wscale;
+	struct tcp_options_received tcp_opt;
 
 	if (!sysctl_tcp_syncookies || !th->ack)
 		goto out;
@@ -182,6 +183,13 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 
 	NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV);
 
+	/* check for timestamp cookie support */
+	memset(&tcp_opt, 0, sizeof(tcp_opt));
+	tcp_parse_options(skb, &tcp_opt, 0);
+
+	if (tcp_opt.saw_tstamp)
+		cookie_check_timestamp(&tcp_opt);
+
 	ret = NULL;
 	req = inet6_reqsk_alloc(&tcp6_request_sock_ops);
 	if (!req)
@@ -216,8 +224,12 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 
 	req->expires = 0UL;
 	req->retrans = 0;
-	ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0;
-	ireq->wscale_ok = ireq->sack_ok = 0;
+	ireq->snd_wscale	= tcp_opt.snd_wscale;
+	ireq->rcv_wscale	= tcp_opt.rcv_wscale;
+	ireq->sack_ok		= tcp_opt.sack_ok;
+	ireq->wscale_ok		= tcp_opt.wscale_ok;
+	ireq->tstamp_ok		= tcp_opt.saw_tstamp;
+	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
 	treq->rcv_isn = ntohl(th->seq) - 1;
 	treq->snt_isn = cookie;
 
@@ -253,10 +265,10 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 			goto out;
 	}
 
-	req->window_clamp = dst_metric(dst, RTAX_WINDOW);
+	req->window_clamp = tp->window_clamp ? :dst_metric(dst, RTAX_WINDOW);
 	tcp_select_initial_window(tcp_full_space(sk), req->mss,
 				  &req->rcv_wnd, &req->window_clamp,
-				  0, &rcv_wscale);
+				  ireq->wscale_ok, &rcv_wscale);
 
 	ireq->rcv_wscale = rcv_wscale;
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 378cc4002a76..8ebf6de29562 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1290,10 +1290,8 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	tcp_parse_options(skb, &tmp_opt, 0);
 
-	if (want_cookie) {
+	if (want_cookie && !tmp_opt.saw_tstamp)
 		tcp_clear_options(&tmp_opt);
-		tmp_opt.saw_tstamp = 0;
-	}
 
 	tmp_opt.tstamp_ok = tmp_opt.saw_tstamp;
 	tcp_openreq_init(req, &tmp_opt, skb);
@@ -1307,6 +1305,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	if (want_cookie) {
 		isn = cookie_v6_init_sequence(sk, skb, &req->mss);
+		req->cookie_ts = tmp_opt.tstamp_ok;
 	} else if (!isn) {
 		if (ipv6_opt_accepted(sk, skb) ||
 		    np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo ||
-- 
cgit v1.2.3


From 98075d245a5bc4aeebc2e9f16fa8b089a5c200ac Mon Sep 17 00:00:00 2001
From: Zoltan Menyhart <Zoltan.Menyhart@bull.net>
Date: Fri, 11 Apr 2008 15:21:35 -0700
Subject: [IA64] Fix NUMA configuration issue

There is a NUMA memory configuration issue in 2.6.24:

A 2-node machine of ours has got the following memory layout:

Node 0:	0 - 2 Gbytes
Node 0:	4 - 8 Gbytes
Node 1:	8 - 16 Gbytes
Node 0:	16 - 18 Gbytes

"efi_memmap_init()" merges the three last ranges into one.

"register_active_ranges()" is called as follows:

efi_memmap_walk(register_active_ranges, NULL);

i.e. once for the 4 - 18 Gbytes range. It picks up the node
number from the start address, and registers all the memory for
the node #0.

"register_active_ranges()" should be called as follows to
make sure there is no merged address range at its entry:

efi_memmap_walk(filter_memory, register_active_ranges);

"filter_memory()" is similar to "filter_rsvd_memory()",
but the reserved memory ranges are not filtered out.

Signed-off-by: Zoltan Menyhart <Zoltan.Menyhart@bull.net>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/setup.c   | 23 +++++++++++++++++++++++
 arch/ia64/mm/contig.c      |  2 +-
 arch/ia64/mm/discontig.c   |  2 +-
 arch/ia64/mm/init.c        |  6 ++----
 include/asm-ia64/meminit.h |  3 ++-
 5 files changed, 29 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 4aa9eaea76c3..c85b7dd6ef33 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -176,6 +176,29 @@ filter_rsvd_memory (unsigned long start, unsigned long end, void *arg)
 	return 0;
 }
 
+/*
+ * Similar to "filter_rsvd_memory()", but the reserved memory ranges
+ * are not filtered out.
+ */
+int __init
+filter_memory(unsigned long start, unsigned long end, void *arg)
+{
+	void (*func)(unsigned long, unsigned long, int);
+
+#if IGNORE_PFN0
+	if (start == PAGE_OFFSET) {
+		printk(KERN_WARNING "warning: skipping physical page 0\n");
+		start += PAGE_SIZE;
+		if (start >= end)
+			return 0;
+	}
+#endif
+	func = arg;
+	if (start < end)
+		call_pernode_memory(__pa(start), end - start, func);
+	return 0;
+}
+
 static void __init
 sort_regions (struct rsvd_region *rsvd_region, int max)
 {
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index 0479661fa41a..798bf9835a51 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -253,7 +253,7 @@ paging_init (void)
 	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
-	efi_memmap_walk(register_active_ranges, NULL);
+	efi_memmap_walk(filter_memory, register_active_ranges);
 	efi_memmap_walk(find_largest_hole, (u64 *)&max_gap);
 	if (max_gap < LARGE_GAP) {
 		vmem_map = (struct page *) 0;
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index ffee1ea00bb0..96d5fbfa44a4 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -444,7 +444,7 @@ void __init find_memory(void)
 			mem_data[node].min_pfn = ~0UL;
 		}
 
-	efi_memmap_walk(register_active_ranges, NULL);
+	efi_memmap_walk(filter_memory, register_active_ranges);
 
 	/*
 	 * Initialize the boot memory maps in reverse order since that's
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index da05893294b5..5c1de53c8c1c 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -547,12 +547,10 @@ find_largest_hole (u64 start, u64 end, void *arg)
 #endif /* CONFIG_VIRTUAL_MEM_MAP */
 
 int __init
-register_active_ranges(u64 start, u64 end, void *arg)
+register_active_ranges(u64 start, u64 len, int nid)
 {
-	int nid = paddr_to_nid(__pa(start));
+	u64 end = start + len;
 
-	if (nid < 0)
-		nid = 0;
 #ifdef CONFIG_KEXEC
 	if (start > crashk_res.start && start < crashk_res.end)
 		start = crashk_res.end;
diff --git a/include/asm-ia64/meminit.h b/include/asm-ia64/meminit.h
index f93308f54b61..7245a5781594 100644
--- a/include/asm-ia64/meminit.h
+++ b/include/asm-ia64/meminit.h
@@ -35,6 +35,7 @@ extern void find_memory (void);
 extern void reserve_memory (void);
 extern void find_initrd (void);
 extern int filter_rsvd_memory (unsigned long start, unsigned long end, void *arg);
+extern int filter_memory (unsigned long start, unsigned long end, void *arg);
 extern unsigned long efi_memmap_init(unsigned long *s, unsigned long *e);
 extern int find_max_min_low_pfn (unsigned long , unsigned long, void *);
 
@@ -56,7 +57,7 @@ extern int reserve_elfcorehdr(unsigned long *start, unsigned long *end);
 
 #define IGNORE_PFN0	1	/* XXX fix me: ignore pfn 0 until TLB miss handler is updated... */
 
-extern int register_active_ranges(u64 start, u64 end, void *arg);
+extern int register_active_ranges(u64 start, u64 len, int nid);
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
 # define LARGE_GAP	0x40000000 /* Use virtual mem map if hole is > than this */
-- 
cgit v1.2.3


From fed85383ac34d82e96f227ce49ce68117cec23a0 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Fri, 11 Apr 2008 20:17:55 +0900
Subject: [IPV6]: Use XOR and OR rather than mutiple ands for ipv6 address
 comparisons.

ipv6_addr_equal(), ipv6_addr_v4mapped(),
ipv6_addr_is_ll_all_{nodes,routers}(),
ipv6_masked_addr_cmp()

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h | 14 ++++++--------
 include/net/ipv6.h     | 22 ++++++++++------------
 2 files changed, 16 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index bdcc863a60a4..1dc9d03372d9 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -240,18 +240,16 @@ static inline int ipv6_addr_is_multicast(const struct in6_addr *addr)
 
 static inline int ipv6_addr_is_ll_all_nodes(const struct in6_addr *addr)
 {
-	return (addr->s6_addr32[0] == htonl(0xff020000) &&
-		addr->s6_addr32[1] == 0 &&
-		addr->s6_addr32[2] == 0 &&
-		addr->s6_addr32[3] == htonl(0x00000001));
+	return (((addr->s6_addr32[0] ^ htonl(0xff020000)) |
+		addr->s6_addr32[1] | addr->s6_addr32[2] |
+		(addr->s6_addr32[3] ^ htonl(0x00000001))) == 0);
 }
 
 static inline int ipv6_addr_is_ll_all_routers(const struct in6_addr *addr)
 {
-	return (addr->s6_addr32[0] == htonl(0xff020000) &&
-		addr->s6_addr32[1] == 0 &&
-		addr->s6_addr32[2] == 0 &&
-		addr->s6_addr32[3] == htonl(0x00000002));
+	return (((addr->s6_addr32[0] ^ htonl(0xff020000)) |
+		addr->s6_addr32[1] | addr->s6_addr32[2] |
+		(addr->s6_addr32[3] ^ htonl(0x00000002))) == 0);
 }
 
 static inline int ipv6_isatap_eui64(u8 *eui, __be32 addr)
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 5738c1c73ac1..a0c285b6311e 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -280,12 +280,10 @@ static inline int
 ipv6_masked_addr_cmp(const struct in6_addr *a1, const struct in6_addr *m,
 		     const struct in6_addr *a2)
 {
-	unsigned int i;
-
-	for (i = 0; i < 4; i++)
-		if ((a1->s6_addr32[i] ^ a2->s6_addr32[i]) & m->s6_addr32[i])
-			return 1;
-	return 0;
+	return (!!(((a1->s6_addr32[0] ^ a2->s6_addr32[0]) & m->s6_addr32[0]) |
+		   ((a1->s6_addr32[1] ^ a2->s6_addr32[1]) & m->s6_addr32[1]) |
+		   ((a1->s6_addr32[2] ^ a2->s6_addr32[2]) & m->s6_addr32[2]) |
+		   ((a1->s6_addr32[3] ^ a2->s6_addr32[3]) & m->s6_addr32[3])));
 }
 
 static inline void ipv6_addr_copy(struct in6_addr *a1, const struct in6_addr *a2)
@@ -320,10 +318,10 @@ static inline void ipv6_addr_set(struct in6_addr *addr,
 static inline int ipv6_addr_equal(const struct in6_addr *a1,
 				  const struct in6_addr *a2)
 {
-	return (a1->s6_addr32[0] == a2->s6_addr32[0] &&
-		a1->s6_addr32[1] == a2->s6_addr32[1] &&
-		a1->s6_addr32[2] == a2->s6_addr32[2] &&
-		a1->s6_addr32[3] == a2->s6_addr32[3]);
+	return (((a1->s6_addr32[0] ^ a2->s6_addr32[0]) |
+		 (a1->s6_addr32[1] ^ a2->s6_addr32[1]) |
+		 (a1->s6_addr32[2] ^ a2->s6_addr32[2]) |
+		 (a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0);
 }
 
 static inline int __ipv6_prefix_equal(const __be32 *a1, const __be32 *a2,
@@ -371,8 +369,8 @@ static inline int ipv6_addr_any(const struct in6_addr *a)
 
 static inline int ipv6_addr_v4mapped(const struct in6_addr *a)
 {
-	return ((a->s6_addr32[0] | a->s6_addr32[1]) == 0 &&
-		 a->s6_addr32[2] == htonl(0x0000ffff));
+	return ((a->s6_addr32[0] | a->s6_addr32[1] |
+		 (a->s6_addr32[2] ^ htonl(0x0000ffff))) == 0);
 }
 
 /*
-- 
cgit v1.2.3


From 3eb84f49290461e2b83d6e8ee1f3f0e504340c8b Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 10 Apr 2008 15:42:08 +0900
Subject: [IPV6] ADDRCONF: Uninline ipv6_addr_hash().

The function is only used in net/ipv6/addrconf.c.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h | 19 -------------------
 net/ipv6/addrconf.c    | 19 +++++++++++++++++++
 2 files changed, 19 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 1dc9d03372d9..1ba4e5bf5801 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -189,25 +189,6 @@ static inline void in6_ifa_put(struct inet6_ifaddr *ifp)
 #define in6_ifa_hold(ifp)	atomic_inc(&(ifp)->refcnt)
 
 
-/*
- *	Hash function taken from net_alias.c
- */
-
-static __inline__ u8 ipv6_addr_hash(const struct in6_addr *addr)
-{	
-	__u32 word;
-
-	/* 
-	 * We perform the hash function over the last 64 bits of the address
-	 * This will include the IEEE address token on links that support it.
-	 */
-
-	word = (__force u32)(addr->s6_addr32[2] ^ addr->s6_addr32[3]);
-	word ^= (word >> 16);
-	word ^= (word >> 8);
-
-	return ((word ^ (word >> 4)) & 0x0f);
-}
 
 /*
  *	compute link-local solicited-node multicast address
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index ed9938014cf9..b17fafc56241 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -541,6 +541,25 @@ ipv6_link_dev_addr(struct inet6_dev *idev, struct inet6_ifaddr *ifp)
 	*ifap = ifp;
 }
 
+/*
+ *	Hash function taken from net_alias.c
+ */
+static u8 ipv6_addr_hash(const struct in6_addr *addr)
+{
+	__u32 word;
+
+	/*
+	 * We perform the hash function over the last 64 bits of the address
+	 * This will include the IEEE address token on links that support it.
+	 */
+
+	word = (__force u32)(addr->s6_addr32[2] ^ addr->s6_addr32[3]);
+	word ^= (word >> 16);
+	word ^= (word >> 8);
+
+	return ((word ^ (word >> 4)) & 0x0f);
+}
+
 /* On success it returns ifp with increased reference count */
 
 static struct inet6_ifaddr *
-- 
cgit v1.2.3


From dfd982baff01c18e3e1717c97fdac79c28f105ce Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 10 Apr 2008 15:42:09 +0900
Subject: [IPV6] ADDRCONF: Uninline ipv6_isatap_eui64().

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h | 15 +--------------
 net/ipv6/addrconf.c    | 26 ++++++++++++++++++++++++--
 2 files changed, 25 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 1ba4e5bf5801..8317c1bcd86c 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -233,20 +233,7 @@ static inline int ipv6_addr_is_ll_all_routers(const struct in6_addr *addr)
 		(addr->s6_addr32[3] ^ htonl(0x00000002))) == 0);
 }
 
-static inline int ipv6_isatap_eui64(u8 *eui, __be32 addr)
-{
-	eui[0] = (ipv4_is_zeronet(addr) || ipv4_is_private_10(addr) ||
-		  ipv4_is_loopback(addr) || ipv4_is_linklocal_169(addr) ||
-		  ipv4_is_private_172(addr) || ipv4_is_test_192(addr) ||
-		  ipv4_is_anycast_6to4(addr) || ipv4_is_private_192(addr) ||
-		  ipv4_is_test_198(addr) || ipv4_is_multicast(addr) ||
-		  ipv4_is_lbcast(addr)) ? 0x00 : 0x02;
-	eui[1] = 0;
-	eui[2] = 0x5E;
-	eui[3] = 0xFE;
-	memcpy (eui+4, &addr, 4);
-	return 0;
-}
+extern int __ipv6_isatap_ifid(u8 *eui, __be32 addr);
 
 static inline int ipv6_addr_is_isatap(const struct in6_addr *addr)
 {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index b17fafc56241..d15f3e095e7e 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1494,6 +1494,29 @@ static int addrconf_ifid_infiniband(u8 *eui, struct net_device *dev)
 	return 0;
 }
 
+int __ipv6_isatap_ifid(u8 *eui, __be32 addr)
+{
+	eui[0] = (ipv4_is_zeronet(addr) || ipv4_is_private_10(addr) ||
+		  ipv4_is_loopback(addr) || ipv4_is_linklocal_169(addr) ||
+		  ipv4_is_private_172(addr) || ipv4_is_test_192(addr) ||
+		  ipv4_is_anycast_6to4(addr) || ipv4_is_private_192(addr) ||
+		  ipv4_is_test_198(addr) || ipv4_is_multicast(addr) ||
+		  ipv4_is_lbcast(addr)) ? 0x00 : 0x02;
+	eui[1] = 0;
+	eui[2] = 0x5E;
+	eui[3] = 0xFE;
+	memcpy(eui + 4, &addr, 4);
+	return 0;
+}
+EXPORT_SYMBOL(__ipv6_isatap_ifid);
+
+static int addrconf_ifid_sit(u8 *eui, struct net_device *dev)
+{
+	if (dev->priv_flags & IFF_ISATAP)
+		return __ipv6_isatap_ifid(eui, *(__be32 *)dev->dev_addr);
+	return -1;
+}
+
 static int ipv6_generate_eui64(u8 *eui, struct net_device *dev)
 {
 	switch (dev->type) {
@@ -1506,8 +1529,7 @@ static int ipv6_generate_eui64(u8 *eui, struct net_device *dev)
 	case ARPHRD_INFINIBAND:
 		return addrconf_ifid_infiniband(eui, dev);
 	case ARPHRD_SIT:
-		if (dev->priv_flags & IFF_ISATAP)
-			return ipv6_isatap_eui64(eui, *(__be32 *)dev->dev_addr);
+		return addrconf_ifid_sit(eui, dev);
 	}
 	return -1;
 }
-- 
cgit v1.2.3


From 9acd9f3ae92d0dc0ca7504fb48c1040e8bbc39fe Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 10 Apr 2008 15:42:10 +0900
Subject: [IPV6]: Make address arguments const.

- net/ipv6/addrconf.c:
	ipv6_get_ifaddr(), ipv6_dev_get_saddr()
- net/ipv6/mcast.c:
	ipv6_sock_mc_join(), ipv6_sock_mc_drop(),
	inet6_mc_check(),
	ipv6_dev_mc_inc(), __ipv6_dev_mc_dec(), ipv6_dev_mc_dec(),
	ipv6_chk_mcast_addr()
- net/ipv6/route.c:
	rt6_lookup(), icmp6_dst_alloc()
- net/ipv6/ip6_output.c:
	ip6_nd_hdr()
- net/ipv6/ndisc.c:
	ndisc_send_ns(), ndisc_send_rs(), ndisc_send_redirect(),
	ndisc_get_neigh(), __ndisc_send()

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/addrconf.h  | 28 +++++++++++++++-------------
 include/net/ip6_route.h |  6 +++---
 include/net/ipv6.h      |  4 ++--
 include/net/ndisc.h     | 14 +++++++-------
 net/ipv6/addrconf.c     |  6 +++---
 net/ipv6/ip6_output.c   |  2 +-
 net/ipv6/mcast.c        | 20 +++++++++-----------
 net/ipv6/ndisc.c        | 22 ++++++++++++----------
 net/ipv6/route.c        |  6 +++---
 9 files changed, 55 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 8317c1bcd86c..92af23d66eb9 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -76,12 +76,12 @@ extern int			ipv6_chk_prefix(struct in6_addr *addr,
 						struct net_device *dev);
 
 extern struct inet6_ifaddr      *ipv6_get_ifaddr(struct net *net,
-						 struct in6_addr *addr,
+						 const struct in6_addr *addr,
 						 struct net_device *dev,
 						 int strict);
 
 extern int			ipv6_dev_get_saddr(struct net_device *dev, 
-					       struct in6_addr *daddr,
+					       const struct in6_addr *daddr,
 					       unsigned int srcprefs,
 					       struct in6_addr *saddr);
 extern int			ipv6_get_lladdr(struct net_device *dev,
@@ -105,25 +105,27 @@ extern u32			ipv6_addr_label(const struct in6_addr *addr,
 /*
  *	multicast prototypes (mcast.c)
  */
-extern int ipv6_sock_mc_join(struct sock *sk, int ifindex, 
-		  struct in6_addr *addr);
-extern int ipv6_sock_mc_drop(struct sock *sk, int ifindex, 
-		  struct in6_addr *addr);
+extern int ipv6_sock_mc_join(struct sock *sk, int ifindex,
+			     const struct in6_addr *addr);
+extern int ipv6_sock_mc_drop(struct sock *sk, int ifindex,
+			     const struct in6_addr *addr);
 extern void ipv6_sock_mc_close(struct sock *sk);
-extern int inet6_mc_check(struct sock *sk, struct in6_addr *mc_addr,
-		struct in6_addr *src_addr);
+extern int inet6_mc_check(struct sock *sk,
+			  const struct in6_addr *mc_addr,
+			  const struct in6_addr *src_addr);
 
-extern int ipv6_dev_mc_inc(struct net_device *dev, struct in6_addr *addr);
-extern int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr);
-extern int ipv6_dev_mc_dec(struct net_device *dev, struct in6_addr *addr);
+extern int ipv6_dev_mc_inc(struct net_device *dev, const struct in6_addr *addr);
+extern int __ipv6_dev_mc_dec(struct inet6_dev *idev, const struct in6_addr *addr);
+extern int ipv6_dev_mc_dec(struct net_device *dev, const struct in6_addr *addr);
 extern void ipv6_mc_up(struct inet6_dev *idev);
 extern void ipv6_mc_down(struct inet6_dev *idev);
 extern void ipv6_mc_init_dev(struct inet6_dev *idev);
 extern void ipv6_mc_destroy_dev(struct inet6_dev *idev);
 extern void addrconf_dad_failure(struct inet6_ifaddr *ifp);
 
-extern int ipv6_chk_mcast_addr(struct net_device *dev, struct in6_addr *group,
-		struct in6_addr *src_addr);
+extern int ipv6_chk_mcast_addr(struct net_device *dev,
+			       const struct in6_addr *group,
+			       const struct in6_addr *src_addr);
 extern int ipv6_is_mld(struct sk_buff *skb, int nexthdr);
 
 extern void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len);
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 9080076ce0e5..9313491e3dad 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -61,13 +61,13 @@ extern int			ip6_ins_rt(struct rt6_info *);
 extern int			ip6_del_rt(struct rt6_info *);
 
 extern struct rt6_info		*rt6_lookup(struct net *net,
-					    struct in6_addr *daddr,
-					    struct in6_addr *saddr,
+					    const struct in6_addr *daddr,
+					    const struct in6_addr *saddr,
 					    int oif, int flags);
 
 extern struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 					 struct neighbour *neigh,
-					 struct in6_addr *addr);
+					 const struct in6_addr *addr);
 extern int icmp6_dst_gc(int *more);
 
 extern void fib6_force_start_gc(struct net *net);
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index a0c285b6311e..49c48983019f 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -451,8 +451,8 @@ extern int			ip6_xmit(struct sock *sk,
 extern int			ip6_nd_hdr(struct sock *sk,
 					   struct sk_buff *skb,
 					   struct net_device *dev,
-					   struct in6_addr *saddr,
-					   struct in6_addr *daddr,
+					   const struct in6_addr *saddr,
+					   const struct in6_addr *daddr,
 					   int proto, int len);
 
 extern int			ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr);
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index 16424236fe2f..9c451ff2f4f4 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -94,17 +94,17 @@ extern int			ndisc_rcv(struct sk_buff *skb);
 
 extern void			ndisc_send_ns(struct net_device *dev,
 					      struct neighbour *neigh,
-					      struct in6_addr *solicit,
-					      struct in6_addr *daddr,
-					      struct in6_addr *saddr);
+					      const struct in6_addr *solicit,
+					      const struct in6_addr *daddr,
+					      const struct in6_addr *saddr);
 
 extern void			ndisc_send_rs(struct net_device *dev,
-					      struct in6_addr *saddr,
-					      struct in6_addr *daddr);
+					      const struct in6_addr *saddr,
+					      const struct in6_addr *daddr);
 
 extern void			ndisc_send_redirect(struct sk_buff *skb,
 						    struct neighbour *neigh,
-						    struct in6_addr *target);
+						    const struct in6_addr *target);
 
 extern int			ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int dir);
 
@@ -134,7 +134,7 @@ extern int 			ndisc_ifinfo_sysctl_change(struct ctl_table *ctl,
 extern void 			inet6_ifinfo_notify(int event,
 						    struct inet6_dev *idev);
 
-static inline struct neighbour * ndisc_get_neigh(struct net_device *dev, struct in6_addr *addr)
+static inline struct neighbour * ndisc_get_neigh(struct net_device *dev, const struct in6_addr *addr)
 {
 
 	if (dev)
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index d15f3e095e7e..4048c2b73b0b 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -940,7 +940,7 @@ struct ipv6_saddr_score {
 };
 
 struct ipv6_saddr_dst {
-	struct in6_addr *addr;
+	const struct in6_addr *addr;
 	int ifindex;
 	int scope;
 	int label;
@@ -1074,7 +1074,7 @@ out:
 }
 
 int ipv6_dev_get_saddr(struct net_device *dst_dev,
-		       struct in6_addr *daddr, unsigned int prefs,
+		       const struct in6_addr *daddr, unsigned int prefs,
 		       struct in6_addr *saddr)
 {
 	struct ipv6_saddr_score scores[2],
@@ -1309,7 +1309,7 @@ int ipv6_chk_prefix(struct in6_addr *addr, struct net_device *dev)
 
 EXPORT_SYMBOL(ipv6_chk_prefix);
 
-struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr,
+struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, const struct in6_addr *addr,
 				     struct net_device *dev, int strict)
 {
 	struct inet6_ifaddr * ifp;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index c0dbe549cc42..0af2e055f883 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -286,7 +286,7 @@ EXPORT_SYMBOL(ip6_xmit);
  */
 
 int ip6_nd_hdr(struct sock *sk, struct sk_buff *skb, struct net_device *dev,
-	       struct in6_addr *saddr, struct in6_addr *daddr,
+	       const struct in6_addr *saddr, const struct in6_addr *daddr,
 	       int proto, int len)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 2e6a53f3cc38..0a0132a1c443 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -127,8 +127,6 @@ static struct in6_addr mld2_all_mcr = MLD2_ALL_MCR_INIT;
 /* Big mc list lock for all the sockets */
 static DEFINE_RWLOCK(ipv6_sk_mc_lock);
 
-int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr);
-
 static void igmp6_join_group(struct ifmcaddr6 *ma);
 static void igmp6_leave_group(struct ifmcaddr6 *ma);
 static void igmp6_timer_handler(unsigned long data);
@@ -177,7 +175,7 @@ int sysctl_mld_max_msf __read_mostly = IPV6_MLD_MAX_MSF;
  *	socket join on multicast group
  */
 
-int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr)
+int ipv6_sock_mc_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
 {
 	struct net_device *dev = NULL;
 	struct ipv6_mc_socklist *mc_lst;
@@ -252,7 +250,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 /*
  *	socket leave on multicast group
  */
-int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
+int ipv6_sock_mc_drop(struct sock *sk, int ifindex, const struct in6_addr *addr)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6_mc_socklist *mc_lst, **lnk;
@@ -664,8 +662,8 @@ done:
 	return err;
 }
 
-int inet6_mc_check(struct sock *sk, struct in6_addr *mc_addr,
-	struct in6_addr *src_addr)
+int inet6_mc_check(struct sock *sk, const struct in6_addr *mc_addr,
+		   const struct in6_addr *src_addr)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6_mc_socklist *mc;
@@ -871,7 +869,7 @@ static void mld_clear_delrec(struct inet6_dev *idev)
 /*
  *	device multicast group inc (add if not found)
  */
-int ipv6_dev_mc_inc(struct net_device *dev, struct in6_addr *addr)
+int ipv6_dev_mc_inc(struct net_device *dev, const struct in6_addr *addr)
 {
 	struct ifmcaddr6 *mc;
 	struct inet6_dev *idev;
@@ -942,7 +940,7 @@ int ipv6_dev_mc_inc(struct net_device *dev, struct in6_addr *addr)
 /*
  *	device multicast group del
  */
-int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr)
+int __ipv6_dev_mc_dec(struct inet6_dev *idev, const struct in6_addr *addr)
 {
 	struct ifmcaddr6 *ma, **map;
 
@@ -967,7 +965,7 @@ int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr)
 	return -ENOENT;
 }
 
-int ipv6_dev_mc_dec(struct net_device *dev, struct in6_addr *addr)
+int ipv6_dev_mc_dec(struct net_device *dev, const struct in6_addr *addr)
 {
 	struct inet6_dev *idev = in6_dev_get(dev);
 	int err;
@@ -1012,8 +1010,8 @@ int ipv6_is_mld(struct sk_buff *skb, int nexthdr)
 /*
  *	check if the interface/address pair is valid
  */
-int ipv6_chk_mcast_addr(struct net_device *dev, struct in6_addr *group,
-	struct in6_addr *src_addr)
+int ipv6_chk_mcast_addr(struct net_device *dev, const struct in6_addr *group,
+			const struct in6_addr *src_addr)
 {
 	struct inet6_dev *idev;
 	struct ifmcaddr6 *mc;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index b3295d82fece..5b9ad5e2f56d 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -442,8 +442,9 @@ static void pndisc_destructor(struct pneigh_entry *n)
  */
 static void __ndisc_send(struct net_device *dev,
 			 struct neighbour *neigh,
-			 struct in6_addr *daddr, struct in6_addr *saddr,
-			 struct icmp6hdr *icmp6h, struct in6_addr *target,
+			 const struct in6_addr *daddr,
+			 const struct in6_addr *saddr,
+			 struct icmp6hdr *icmp6h, const struct in6_addr *target,
 			 int llinfo)
 {
 	struct flowi fl;
@@ -529,12 +530,13 @@ static void __ndisc_send(struct net_device *dev,
 }
 
 static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
-		   struct in6_addr *daddr, struct in6_addr *solicited_addr,
-		   int router, int solicited, int override, int inc_opt)
+			  const struct in6_addr *daddr,
+			  const struct in6_addr *solicited_addr,
+			  int router, int solicited, int override, int inc_opt)
 {
 	struct in6_addr tmpaddr;
 	struct inet6_ifaddr *ifp;
-	struct in6_addr *src_addr;
+	const struct in6_addr *src_addr;
 	struct icmp6hdr icmp6h = {
 		.icmp6_type = NDISC_NEIGHBOUR_ADVERTISEMENT,
 	};
@@ -564,8 +566,8 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
 }
 
 void ndisc_send_ns(struct net_device *dev, struct neighbour *neigh,
-		   struct in6_addr *solicit,
-		   struct in6_addr *daddr, struct in6_addr *saddr)
+		   const struct in6_addr *solicit,
+		   const struct in6_addr *daddr, const struct in6_addr *saddr)
 {
 	struct in6_addr addr_buf;
 	struct icmp6hdr icmp6h = {
@@ -584,8 +586,8 @@ void ndisc_send_ns(struct net_device *dev, struct neighbour *neigh,
 		     !ipv6_addr_any(saddr) ? ND_OPT_SOURCE_LL_ADDR : 0);
 }
 
-void ndisc_send_rs(struct net_device *dev, struct in6_addr *saddr,
-		   struct in6_addr *daddr)
+void ndisc_send_rs(struct net_device *dev, const struct in6_addr *saddr,
+		   const struct in6_addr *daddr)
 {
 	struct icmp6hdr icmp6h = {
 		.icmp6_type = NDISC_ROUTER_SOLICITATION,
@@ -1447,7 +1449,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb)
 }
 
 void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh,
-			 struct in6_addr *target)
+			 const struct in6_addr *target)
 {
 	struct net_device *dev = skb->dev;
 	struct net *net = dev_net(dev);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 269b76093288..6293cb91ed1d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -556,8 +556,8 @@ out:
 
 }
 
-struct rt6_info *rt6_lookup(struct net *net, struct in6_addr *daddr,
-			    struct in6_addr *saddr, int oif, int strict)
+struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr,
+			    const struct in6_addr *saddr, int oif, int strict)
 {
 	struct flowi fl = {
 		.oif = oif,
@@ -925,7 +925,7 @@ static DEFINE_SPINLOCK(icmp6_dst_lock);
 
 struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 				  struct neighbour *neigh,
-				  struct in6_addr *addr)
+				  const struct in6_addr *addr)
 {
 	struct rt6_info *rt;
 	struct inet6_dev *idev = in6_dev_get(dev);
-- 
cgit v1.2.3


From f3ee4010e84452aa133e5163e6cfabc52b194e94 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 10 Apr 2008 15:42:11 +0900
Subject: [IPV6]: Define constants for link-local multicast addresses.

- Define link-local all-node / all-router multicast addresses.
- Remove ipv6_addr_all_nodes() and ipv6_addr_all_routers().

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/in6.h    |  8 ++++++++
 include/net/addrconf.h | 11 -----------
 net/ipv6/addrconf.c    | 25 +++++++------------------
 net/ipv6/mcast.c       | 23 ++++++++---------------
 net/ipv6/ndisc.c       |  5 +----
 5 files changed, 24 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index e6aa8de2b939..bc492048c349 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -48,6 +48,14 @@ extern const struct in6_addr in6addr_any;
 #define IN6ADDR_ANY_INIT { { { 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } } }
 extern const struct in6_addr in6addr_loopback;
 #define IN6ADDR_LOOPBACK_INIT { { { 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1 } } }
+#ifdef __KERNEL__
+extern const struct in6_addr in6addr_linklocal_allnodes;
+#define IN6ADDR_LINKLOCAL_ALLNODES_INIT	\
+		{ { { 0xff,2,0,0,0,0,0,0,0,0,0,0,0,0,0,1 } } }
+extern const struct in6_addr in6addr_linklocal_allrouters;
+#define IN6ADDR_LINKLOCAL_ALLROUTERS_INIT \
+		{ { { 0xff,2,0,0,0,0,0,0,0,0,0,0,0,0,0,2 } } }
+#endif
 
 struct sockaddr_in6 {
 	unsigned short int	sin6_family;    /* AF_INET6 */
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 92af23d66eb9..0a2f0372df31 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -205,17 +205,6 @@ static inline void addrconf_addr_solict_mult(const struct in6_addr *addr,
 		      htonl(0xFF000000) | addr->s6_addr32[3]);
 }
 
-
-static inline void ipv6_addr_all_nodes(struct in6_addr *addr)
-{
-	ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x1));
-}
-
-static inline void ipv6_addr_all_routers(struct in6_addr *addr)
-{
-	ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x2));
-}
-
 static inline int ipv6_addr_is_multicast(const struct in6_addr *addr)
 {
 	return (addr->s6_addr32[0] & htonl(0xFF000000)) == htonl(0xFF000000);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 4048c2b73b0b..7df04d294924 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -222,6 +222,8 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 /* IPv6 Wildcard Address and Loopback Address defined by RFC2553 */
 const struct in6_addr in6addr_any = IN6ADDR_ANY_INIT;
 const struct in6_addr in6addr_loopback = IN6ADDR_LOOPBACK_INIT;
+const struct in6_addr in6addr_linklocal_allnodes = IN6ADDR_LINKLOCAL_ALLNODES_INIT;
+const struct in6_addr in6addr_linklocal_allrouters = IN6ADDR_LINKLOCAL_ALLROUTERS_INIT;
 
 /* Check if a valid qdisc is available */
 static inline int addrconf_qdisc_ok(struct net_device *dev)
@@ -321,7 +323,6 @@ EXPORT_SYMBOL(in6_dev_finish_destroy);
 static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 {
 	struct inet6_dev *ndev;
-	struct in6_addr maddr;
 
 	ASSERT_RTNL();
 
@@ -406,8 +407,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 	rcu_assign_pointer(dev->ip6_ptr, ndev);
 
 	/* Join all-node multicast group */
-	ipv6_addr_all_nodes(&maddr);
-	ipv6_dev_mc_inc(dev, &maddr);
+	ipv6_dev_mc_inc(dev, &in6addr_linklocal_allnodes);
 
 	return ndev;
 }
@@ -433,18 +433,15 @@ static void dev_forward_change(struct inet6_dev *idev)
 {
 	struct net_device *dev;
 	struct inet6_ifaddr *ifa;
-	struct in6_addr addr;
 
 	if (!idev)
 		return;
 	dev = idev->dev;
 	if (dev && (dev->flags & IFF_MULTICAST)) {
-		ipv6_addr_all_routers(&addr);
-
 		if (idev->cnf.forwarding)
-			ipv6_dev_mc_inc(dev, &addr);
+			ipv6_dev_mc_inc(dev, &in6addr_linklocal_allrouters);
 		else
-			ipv6_dev_mc_dec(dev, &addr);
+			ipv6_dev_mc_dec(dev, &in6addr_linklocal_allrouters);
 	}
 	for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) {
 		if (ifa->flags&IFA_F_TENTATIVE)
@@ -2654,8 +2651,6 @@ static void addrconf_rs_timer(unsigned long data)
 
 	spin_lock(&ifp->lock);
 	if (ifp->probes++ < ifp->idev->cnf.rtr_solicits) {
-		struct in6_addr all_routers;
-
 		/* The wait after the last probe can be shorter */
 		addrconf_mod_timer(ifp, AC_RS,
 				   (ifp->probes == ifp->idev->cnf.rtr_solicits) ?
@@ -2663,9 +2658,7 @@ static void addrconf_rs_timer(unsigned long data)
 				   ifp->idev->cnf.rtr_solicit_interval);
 		spin_unlock(&ifp->lock);
 
-		ipv6_addr_all_routers(&all_routers);
-
-		ndisc_send_rs(ifp->idev->dev, &ifp->addr, &all_routers);
+		ndisc_send_rs(ifp->idev->dev, &ifp->addr, &in6addr_linklocal_allrouters);
 	} else {
 		spin_unlock(&ifp->lock);
 		/*
@@ -2806,16 +2799,12 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp)
 	    ifp->idev->cnf.rtr_solicits > 0 &&
 	    (dev->flags&IFF_LOOPBACK) == 0 &&
 	    (ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL)) {
-		struct in6_addr all_routers;
-
-		ipv6_addr_all_routers(&all_routers);
-
 		/*
 		 *	If a host as already performed a random delay
 		 *	[...] as part of DAD [...] there is no need
 		 *	to delay again before sending the first RS
 		 */
-		ndisc_send_rs(ifp->idev->dev, &ifp->addr, &all_routers);
+		ndisc_send_rs(ifp->idev->dev, &ifp->addr, &in6addr_linklocal_allrouters);
 
 		spin_lock_bh(&ifp->lock);
 		ifp->probes = 1;
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 0a0132a1c443..c2dc2e2b6c07 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -1766,10 +1766,9 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 	struct inet6_dev *idev;
 	struct sk_buff *skb;
 	struct icmp6hdr *hdr;
-	struct in6_addr *snd_addr;
+	const struct in6_addr *snd_addr;
 	struct in6_addr *addrp;
 	struct in6_addr addr_buf;
-	struct in6_addr all_routers;
 	int err, len, payload_len, full_len;
 	u8 ra[8] = { IPPROTO_ICMPV6, 0,
 		     IPV6_TLV_ROUTERALERT, 2, 0, 0,
@@ -1780,11 +1779,10 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 	IP6_INC_STATS(__in6_dev_get(dev),
 		      IPSTATS_MIB_OUTREQUESTS);
 	rcu_read_unlock();
-	snd_addr = addr;
-	if (type == ICMPV6_MGM_REDUCTION) {
-		snd_addr = &all_routers;
-		ipv6_addr_all_routers(&all_routers);
-	}
+	if (type == ICMPV6_MGM_REDUCTION)
+		snd_addr = &in6addr_linklocal_allrouters;
+	else
+		snd_addr = addr;
 
 	len = sizeof(struct icmp6hdr) + sizeof(struct in6_addr);
 	payload_len = len + sizeof(ra);
@@ -2309,24 +2307,19 @@ void ipv6_mc_init_dev(struct inet6_dev *idev)
 void ipv6_mc_destroy_dev(struct inet6_dev *idev)
 {
 	struct ifmcaddr6 *i;
-	struct in6_addr maddr;
 
 	/* Deactivate timers */
 	ipv6_mc_down(idev);
 
 	/* Delete all-nodes address. */
-	ipv6_addr_all_nodes(&maddr);
-
 	/* We cannot call ipv6_dev_mc_dec() directly, our caller in
 	 * addrconf.c has NULL'd out dev->ip6_ptr so in6_dev_get() will
 	 * fail.
 	 */
-	__ipv6_dev_mc_dec(idev, &maddr);
+	__ipv6_dev_mc_dec(idev, &in6addr_linklocal_allnodes);
 
-	if (idev->cnf.forwarding) {
-		ipv6_addr_all_routers(&maddr);
-		__ipv6_dev_mc_dec(idev, &maddr);
-	}
+	if (idev->cnf.forwarding)
+		__ipv6_dev_mc_dec(idev, &in6addr_linklocal_allrouters);
 
 	write_lock_bh(&idev->lock);
 	while ((i = idev->mc_list) != NULL) {
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 5b9ad5e2f56d..2c74885f8355 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -818,10 +818,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 		is_router = !!idev->cnf.forwarding;
 
 	if (dad) {
-		struct in6_addr maddr;
-
-		ipv6_addr_all_nodes(&maddr);
-		ndisc_send_na(dev, NULL, &maddr, &msg->target,
+		ndisc_send_na(dev, NULL, &in6addr_linklocal_allnodes, &msg->target,
 			      is_router, 0, (ifp != NULL), 1);
 		goto out;
 	}
-- 
cgit v1.2.3


From 7f1eced8b0a001c4d5a8cfa5ac7b5cbc89fedab8 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 10 Apr 2008 15:42:12 +0900
Subject: [IPV6] MIP6: Use our standard definitions for paddings.

MIP6_OPT_PAD_X are actually for paddings in destination
option header.  Replace them with our standard IPV6_TLV_PADX.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/mip6.h | 3 ---
 net/ipv6/mip6.c    | 4 ++--
 2 files changed, 2 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/mip6.h b/include/net/mip6.h
index 63272610a24a..a83ad1982a90 100644
--- a/include/net/mip6.h
+++ b/include/net/mip6.h
@@ -28,9 +28,6 @@
 #include <linux/skbuff.h>
 #include <net/sock.h>
 
-#define MIP6_OPT_PAD_1	0
-#define MIP6_OPT_PAD_N	1
-
 /*
  * Mobility Header
  */
diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c
index 42403c626c27..ad1cc5bbf977 100644
--- a/net/ipv6/mip6.c
+++ b/net/ipv6/mip6.c
@@ -44,9 +44,9 @@ static inline void *mip6_padn(__u8 *data, __u8 padlen)
 	if (!data)
 		return NULL;
 	if (padlen == 1) {
-		data[0] = MIP6_OPT_PAD_1;
+		data[0] = IPV6_TLV_PAD0;
 	} else if (padlen > 1) {
-		data[0] = MIP6_OPT_PAD_N;
+		data[0] = IPV6_TLV_PADN;
 		data[1] = padlen - 2;
 		if (padlen > 2)
 			memset(data+2, 0, data[1]);
-- 
cgit v1.2.3


From f4ad85ca3ef8a1ede76c5020a28a8f4057b4d24f Mon Sep 17 00:00:00 2001
From: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Date: Sat, 12 Apr 2008 18:39:34 -0700
Subject: [SCTP]: Fix protocol violation when receiving an error lenght
 INIT-ACK

When receiving an error length INIT-ACK during COOKIE-WAIT,
a 0-vtag ABORT will be responsed. This action violates the
protocol apparently. This patch achieves the following things.
1 If the INIT-ACK contains all the fixed parameters, use init-tag
  recorded from INIT-ACK as vtag.
2 If the INIT-ACK doesn't contain all the fixed parameters,
  just reflect its vtag.

Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/command.h |  1 +
 net/sctp/outqueue.c        |  3 +++
 net/sctp/sm_sideeffect.c   |  3 +++
 net/sctp/sm_statefuns.c    | 18 ++++++++++++++++++
 4 files changed, 25 insertions(+)

(limited to 'include')

diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h
index 10ae2da6f93b..35b1e83fb96a 100644
--- a/include/net/sctp/command.h
+++ b/include/net/sctp/command.h
@@ -104,6 +104,7 @@ typedef enum {
 	SCTP_CMD_ADAPTATION_IND, /* generate and send adaptation event */
 	SCTP_CMD_ASSOC_SHKEY,    /* generate the association shared keys */
 	SCTP_CMD_T1_RETRAN,	 /* Mark for retransmission after T1 timeout  */
+	SCTP_CMD_UPDATE_INITTAG, /* Update peer inittag */
 	SCTP_CMD_LAST
 } sctp_verb_t;
 
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 1bb3c5c35d2a..c0714469233c 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -793,6 +793,9 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 			break;
 
 		case SCTP_CID_ABORT:
+			if (sctp_test_T_bit(chunk)) {
+				packet->vtag = asoc->c.my_vtag;
+			}
 		case SCTP_CID_SACK:
 		case SCTP_CID_HEARTBEAT:
 		case SCTP_CID_HEARTBEAT_ACK:
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index 28eb38eb6083..a4763fd24fd8 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -1536,6 +1536,9 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 			error = sctp_auth_asoc_init_active_key(asoc,
 						GFP_ATOMIC);
 			break;
+		case SCTP_CMD_UPDATE_INITTAG:
+			asoc->peer.i.init_tag = cmd->obj.u32;
+			break;
 
 		default:
 			printk(KERN_WARNING "Impossible command: %u, %p\n",
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index f2ed6473feef..3ef97499df0d 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -4144,6 +4144,24 @@ static sctp_disposition_t sctp_sf_abort_violation(
 		goto nomem;
 
 	if (asoc) {
+		/* Treat INIT-ACK as a special case during COOKIE-WAIT. */
+		if (chunk->chunk_hdr->type == SCTP_CID_INIT_ACK &&
+		    !asoc->peer.i.init_tag) {
+			sctp_initack_chunk_t *initack;
+
+			initack = (sctp_initack_chunk_t *)chunk->chunk_hdr;
+			if (!sctp_chunk_length_valid(chunk,
+						     sizeof(sctp_initack_chunk_t)))
+				abort->chunk_hdr->flags |= SCTP_CHUNK_FLAG_T;
+			else {
+				unsigned int inittag;
+
+				inittag = ntohl(initack->init_hdr.init_tag);
+				sctp_add_cmd_sf(commands, SCTP_CMD_UPDATE_INITTAG,
+						SCTP_U32(inittag));
+			}
+		}
+
 		sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort));
 		SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS);
 
-- 
cgit v1.2.3


From ab38fb04c9f8928cfaf6f4966633d783419906a1 Mon Sep 17 00:00:00 2001
From: Vlad Yasevich <vladislav.yasevich@hp.com>
Date: Sat, 12 Apr 2008 18:40:06 -0700
Subject: [SCTP]: Fix compiler warning about const qualifiers

Fix 3 warnings about discarding const qualifiers:

net/sctp/ulpevent.c:862: warning: passing argument 1 of 'sctp_event2skb' discards qualifiers from pointer target type
net/sctp/sm_statefuns.c:4393: warning: passing argument 1 of 'SCTP_ASOC' discards qualifiers from pointer target type
net/sctp/socket.c:5874: warning: passing argument 1 of 'cmsg_nxthdr' discards qualifiers from pointer target type

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/ulpevent.h | 2 +-
 net/sctp/sm_statefuns.c     | 5 +++--
 net/sctp/socket.c           | 5 +++--
 net/sctp/ulpevent.c         | 2 +-
 4 files changed, 8 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/ulpevent.h b/include/net/sctp/ulpevent.h
index 9bcfc12275e8..7ea12e8e6676 100644
--- a/include/net/sctp/ulpevent.h
+++ b/include/net/sctp/ulpevent.h
@@ -67,7 +67,7 @@ struct sctp_ulpevent {
 };
 
 /* Retrieve the skb this event sits inside of. */
-static inline struct sk_buff *sctp_event2skb(struct sctp_ulpevent *ev)
+static inline struct sk_buff *sctp_event2skb(const struct sctp_ulpevent *ev)
 {
 	return container_of((void *)ev, struct sk_buff, cb);
 }
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 3ef97499df0d..07194c2a32df 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -4367,6 +4367,7 @@ sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep,
 				       sctp_cmd_seq_t *commands)
 {
 	struct sctp_chunk *repl;
+	struct sctp_association* my_asoc;
 
 	/* The comment below says that we enter COOKIE-WAIT AFTER
 	 * sending the INIT, but that doesn't actually work in our
@@ -4390,8 +4391,8 @@ sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep,
 	/* Cast away the const modifier, as we want to just
 	 * rerun it through as a sideffect.
 	 */
-	sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC,
-			SCTP_ASOC((struct sctp_association *) asoc));
+	my_asoc = (struct sctp_association *)asoc;
+	sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC(my_asoc));
 
 	/* Choose transport for INIT. */
 	sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT,
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index d994d822900d..998e63a31311 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -5868,11 +5868,12 @@ SCTP_STATIC int sctp_msghdr_parse(const struct msghdr *msg,
 				  sctp_cmsgs_t *cmsgs)
 {
 	struct cmsghdr *cmsg;
+	struct msghdr *my_msg = (struct msghdr *)msg;
 
 	for (cmsg = CMSG_FIRSTHDR(msg);
 	     cmsg != NULL;
-	     cmsg = CMSG_NXTHDR((struct msghdr*)msg, cmsg)) {
-		if (!CMSG_OK(msg, cmsg))
+	     cmsg = CMSG_NXTHDR(my_msg, cmsg)) {
+		if (!CMSG_OK(my_msg, cmsg))
 			return -EINVAL;
 
 		/* Should we parse this header or ignore?  */
diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c
index b43f1f110f87..ce6cda6b6994 100644
--- a/net/sctp/ulpevent.c
+++ b/net/sctp/ulpevent.c
@@ -859,7 +859,7 @@ __u16 sctp_ulpevent_get_notification_type(const struct sctp_ulpevent *event)
 	union sctp_notification *notification;
 	struct sk_buff *skb;
 
-	skb = sctp_event2skb((struct sctp_ulpevent *)event);
+	skb = sctp_event2skb(event);
 	notification = (union sctp_notification *) skb->data;
 	return notification->sn_header.sn_type;
 }
-- 
cgit v1.2.3


From 14daa02139dcb3193b2b0250c0720a23ef610c49 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Sat, 12 Apr 2008 18:48:58 -0700
Subject: net: make struct tun_struct private to tun.c

There's no reason for this to be in the header, and it just hurts
recompile time.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyanskiy <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 33 +++++++++++++++++++++++++++++++++
 include/linux/Kbuild   |  2 +-
 include/linux/if_tun.h | 39 ---------------------------------------
 3 files changed, 34 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 7b816a032957..970ec4793442 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -67,9 +67,42 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 
+/* Uncomment to enable debugging */
+/* #define TUN_DEBUG 1 */
+
 #ifdef TUN_DEBUG
 static int debug;
+
+#define DBG  if(tun->debug)printk
+#define DBG1 if(debug==2)printk
+#else
+#define DBG( a... )
+#define DBG1( a... )
+#endif
+
+struct tun_struct {
+	struct list_head        list;
+	unsigned long 		flags;
+	int			attached;
+	uid_t			owner;
+	gid_t			group;
+
+	wait_queue_head_t	read_wait;
+	struct sk_buff_head	readq;
+
+	struct net_device	*dev;
+
+	struct fasync_struct    *fasync;
+
+	unsigned long if_flags;
+	u8 dev_addr[ETH_ALEN];
+	u32 chr_filter[2];
+	u32 net_filter[2];
+
+#ifdef TUN_DEBUG
+	int debug;
 #endif
+};
 
 /* Network device part of the driver */
 
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 9cdd12a9e843..cedbbd806bf6 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -86,6 +86,7 @@ header-y += if_plip.h
 header-y += if_ppp.h
 header-y += if_slip.h
 header-y += if_strip.h
+header-y += if_tun.h
 header-y += if_tunnel.h
 header-y += in6.h
 header-y += in_route.h
@@ -229,7 +230,6 @@ unifdef-y += if_link.h
 unifdef-y += if_pppol2tp.h
 unifdef-y += if_pppox.h
 unifdef-y += if_tr.h
-unifdef-y += if_tun.h
 unifdef-y += if_vlan.h
 unifdef-y += if_wanpipe.h
 unifdef-y += igmp.h
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 72f1c5f47be3..8c71fe2fb1f5 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -18,47 +18,8 @@
 #ifndef __IF_TUN_H
 #define __IF_TUN_H
 
-/* Uncomment to enable debugging */
-/* #define TUN_DEBUG 1 */
-
 #include <linux/types.h>
 
-#ifdef __KERNEL__
-
-#ifdef TUN_DEBUG
-#define DBG  if(tun->debug)printk
-#define DBG1 if(debug==2)printk
-#else
-#define DBG( a... )
-#define DBG1( a... )
-#endif
-
-struct tun_struct {
-	struct list_head        list;
-	unsigned long 		flags;
-	int			attached;
-	uid_t			owner;
-	gid_t			group;
-
-	wait_queue_head_t	read_wait;
-	struct sk_buff_head	readq;
-
-	struct net_device	*dev;
-
-	struct fasync_struct    *fasync;
-
-	unsigned long if_flags;
-	u8 dev_addr[ETH_ALEN];
-	u32 chr_filter[2];
-	u32 net_filter[2];
-
-#ifdef TUN_DEBUG	
-	int debug;
-#endif  
-};
-
-#endif /* __KERNEL__ */
-
 /* Read queue size */
 #define TUN_READQ_SIZE	500
 
-- 
cgit v1.2.3


From 00447872a643787411c2c0cb1df6169dda8b0c47 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Sat, 12 Apr 2008 19:06:42 -0700
Subject: NetLabel: Allow passing the LSM domain as a shared pointer

Smack doesn't have the need to create a private copy of the LSM "domain" when
setting NetLabel security attributes like SELinux, however, the current
NetLabel code requires a private copy of the LSM "domain".  This patches fixes
that by letting the LSM determine how it wants to pass the domain value.

 * NETLBL_SECATTR_DOMAIN_CPY
   The current behavior, NetLabel assumes that the domain value is a copy and
   frees it when done

 * NETLBL_SECATTR_DOMAIN
   New, Smack-friendly behavior, NetLabel assumes that the domain value is a
   reference to a string managed by the LSM and does not free it when done

Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlabel.h         | 14 ++++++++++----
 security/selinux/ss/services.c |  2 +-
 security/smack/smack_lsm.c     |  2 +-
 3 files changed, 12 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index 0ca67d73c7ad..5e53a85b5ca1 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -162,7 +162,7 @@ struct netlbl_lsm_secattr_catmap {
 
 /**
  * struct netlbl_lsm_secattr - NetLabel LSM security attributes
- * @flags: indicate which attributes are contained in this structure
+ * @flags: indicate structure attributes, see NETLBL_SECATTR_*
  * @type: indicate the NLTYPE of the attributes
  * @domain: the NetLabel LSM domain
  * @cache: NetLabel LSM specific cache
@@ -180,17 +180,22 @@ struct netlbl_lsm_secattr_catmap {
  * NetLabel itself when returning security attributes to the LSM.
  *
  */
+struct netlbl_lsm_secattr {
+	u32 flags;
+	/* bitmap values for 'flags' */
 #define NETLBL_SECATTR_NONE             0x00000000
 #define NETLBL_SECATTR_DOMAIN           0x00000001
+#define NETLBL_SECATTR_DOMAIN_CPY       (NETLBL_SECATTR_DOMAIN | \
+					 NETLBL_SECATTR_FREE_DOMAIN)
 #define NETLBL_SECATTR_CACHE            0x00000002
 #define NETLBL_SECATTR_MLS_LVL          0x00000004
 #define NETLBL_SECATTR_MLS_CAT          0x00000008
 #define NETLBL_SECATTR_SECID            0x00000010
+	/* bitmap meta-values for 'flags' */
+#define NETLBL_SECATTR_FREE_DOMAIN      0x01000000
 #define NETLBL_SECATTR_CACHEABLE        (NETLBL_SECATTR_MLS_LVL | \
 					 NETLBL_SECATTR_MLS_CAT | \
 					 NETLBL_SECATTR_SECID)
-struct netlbl_lsm_secattr {
-	u32 flags;
 	u32 type;
 	char *domain;
 	struct netlbl_lsm_cache *cache;
@@ -303,7 +308,8 @@ static inline void netlbl_secattr_init(struct netlbl_lsm_secattr *secattr)
  */
 static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr)
 {
-	kfree(secattr->domain);
+	if (secattr->flags & NETLBL_SECATTR_FREE_DOMAIN)
+		kfree(secattr->domain);
 	if (secattr->flags & NETLBL_SECATTR_CACHE)
 		netlbl_secattr_cache_free(secattr->cache);
 	if (secattr->flags & NETLBL_SECATTR_MLS_CAT)
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index f37418601215..47295acd09c9 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2649,7 +2649,7 @@ int security_netlbl_sid_to_secattr(u32 sid, struct netlbl_lsm_secattr *secattr)
 		goto netlbl_sid_to_secattr_failure;
 	secattr->domain = kstrdup(policydb.p_type_val_to_name[ctx->type - 1],
 				  GFP_ATOMIC);
-	secattr->flags |= NETLBL_SECATTR_DOMAIN;
+	secattr->flags |= NETLBL_SECATTR_DOMAIN_CPY;
 	mls_export_netlbl_lvl(ctx, secattr);
 	rc = mls_export_netlbl_cat(ctx, secattr);
 	if (rc != 0)
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 732ba27923c4..e2d6f7cd9254 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -1275,7 +1275,7 @@ static void smack_to_secattr(char *smack, struct netlbl_lsm_secattr *nlsp)
 
 	switch (smack_net_nltype) {
 	case NETLBL_NLTYPE_CIPSOV4:
-		nlsp->domain = kstrdup(smack, GFP_ATOMIC);
+		nlsp->domain = smack;
 		nlsp->flags = NETLBL_SECATTR_DOMAIN | NETLBL_SECATTR_MLS_LVL;
 
 		rc = smack_to_cipso(smack, &cipso);
-- 
cgit v1.2.3


From 03e1ad7b5d871d4189b1da3125c2f12d1b5f7d0b Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Sat, 12 Apr 2008 19:07:52 -0700
Subject: LSM: Make the Labeled IPsec hooks more stack friendly

The xfrm_get_policy() and xfrm_add_pol_expire() put some rather large structs
on the stack to work around the LSM API.  This patch attempts to fix that
problem by changing the LSM API to require only the relevant "security"
pointers instead of the entire SPD entry; we do this for all of the
security_xfrm_policy*() functions to keep things consistent.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/security.h        | 48 ++++++++++++++++++++---------------------
 net/key/af_key.c                | 23 ++++++++++----------
 net/xfrm/xfrm_policy.c          | 24 +++++++++++++--------
 net/xfrm/xfrm_user.c            | 33 ++++++++++++++--------------
 security/dummy.c                | 14 ++++++------
 security/security.c             | 21 +++++++++---------
 security/selinux/include/xfrm.h | 13 +++++------
 security/selinux/xfrm.c         | 39 ++++++++++++++-------------------
 8 files changed, 109 insertions(+), 106 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index c673dfd4dffc..f5eb9ff47ac5 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -910,24 +910,24 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  * Security hooks for XFRM operations.
  *
  * @xfrm_policy_alloc_security:
- *	@xp contains the xfrm_policy being added to Security Policy Database
- *	used by the XFRM system.
+ *	@ctxp is a pointer to the xfrm_sec_ctx being added to Security Policy
+ *	Database used by the XFRM system.
  *	@sec_ctx contains the security context information being provided by
  *	the user-level policy update program (e.g., setkey).
  *	Allocate a security structure to the xp->security field; the security
  *	field is initialized to NULL when the xfrm_policy is allocated.
  *	Return 0 if operation was successful (memory to allocate, legal context)
  * @xfrm_policy_clone_security:
- *	@old contains an existing xfrm_policy in the SPD.
- *	@new contains a new xfrm_policy being cloned from old.
- *	Allocate a security structure to the new->security field
- *	that contains the information from the old->security field.
+ *	@old_ctx contains an existing xfrm_sec_ctx.
+ *	@new_ctxp contains a new xfrm_sec_ctx being cloned from old.
+ *	Allocate a security structure in new_ctxp that contains the
+ *	information from the old_ctx structure.
  *	Return 0 if operation was successful (memory to allocate).
  * @xfrm_policy_free_security:
- *	@xp contains the xfrm_policy
+ *	@ctx contains the xfrm_sec_ctx
  *	Deallocate xp->security.
  * @xfrm_policy_delete_security:
- *	@xp contains the xfrm_policy.
+ *	@ctx contains the xfrm_sec_ctx.
  *	Authorize deletion of xp->security.
  * @xfrm_state_alloc_security:
  *	@x contains the xfrm_state being added to the Security Association
@@ -947,7 +947,7 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@x contains the xfrm_state.
  *	Authorize deletion of x->security.
  * @xfrm_policy_lookup:
- *	@xp contains the xfrm_policy for which the access control is being
+ *	@ctx contains the xfrm_sec_ctx for which the access control is being
  *	checked.
  *	@fl_secid contains the flow security label that is used to authorize
  *	access to the policy xp.
@@ -1454,17 +1454,17 @@ struct security_operations {
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
-	int (*xfrm_policy_alloc_security) (struct xfrm_policy *xp,
+	int (*xfrm_policy_alloc_security) (struct xfrm_sec_ctx **ctxp,
 			struct xfrm_user_sec_ctx *sec_ctx);
-	int (*xfrm_policy_clone_security) (struct xfrm_policy *old, struct xfrm_policy *new);
-	void (*xfrm_policy_free_security) (struct xfrm_policy *xp);
-	int (*xfrm_policy_delete_security) (struct xfrm_policy *xp);
+	int (*xfrm_policy_clone_security) (struct xfrm_sec_ctx *old_ctx, struct xfrm_sec_ctx **new_ctx);
+	void (*xfrm_policy_free_security) (struct xfrm_sec_ctx *ctx);
+	int (*xfrm_policy_delete_security) (struct xfrm_sec_ctx *ctx);
 	int (*xfrm_state_alloc_security) (struct xfrm_state *x,
 		struct xfrm_user_sec_ctx *sec_ctx,
 		u32 secid);
 	void (*xfrm_state_free_security) (struct xfrm_state *x);
 	int (*xfrm_state_delete_security) (struct xfrm_state *x);
-	int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 fl_secid, u8 dir);
+	int (*xfrm_policy_lookup)(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir);
 	int (*xfrm_state_pol_flow_match)(struct xfrm_state *x,
 			struct xfrm_policy *xp, struct flowi *fl);
 	int (*xfrm_decode_session)(struct sk_buff *skb, u32 *secid, int ckall);
@@ -2562,16 +2562,16 @@ static inline void security_inet_conn_established(struct sock *sk,
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
 
-int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx);
-int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new);
-void security_xfrm_policy_free(struct xfrm_policy *xp);
-int security_xfrm_policy_delete(struct xfrm_policy *xp);
+int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx);
+int security_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx, struct xfrm_sec_ctx **new_ctxp);
+void security_xfrm_policy_free(struct xfrm_sec_ctx *ctx);
+int security_xfrm_policy_delete(struct xfrm_sec_ctx *ctx);
 int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx);
 int security_xfrm_state_alloc_acquire(struct xfrm_state *x,
 				      struct xfrm_sec_ctx *polsec, u32 secid);
 int security_xfrm_state_delete(struct xfrm_state *x);
 void security_xfrm_state_free(struct xfrm_state *x);
-int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir);
+int security_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir);
 int security_xfrm_state_pol_flow_match(struct xfrm_state *x,
 				       struct xfrm_policy *xp, struct flowi *fl);
 int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid);
@@ -2579,21 +2579,21 @@ void security_skb_classify_flow(struct sk_buff *skb, struct flowi *fl);
 
 #else	/* CONFIG_SECURITY_NETWORK_XFRM */
 
-static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx)
+static inline int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx)
 {
 	return 0;
 }
 
-static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new)
+static inline int security_xfrm_policy_clone(struct xfrm_sec_ctx *old, struct xfrm_sec_ctx **new_ctxp)
 {
 	return 0;
 }
 
-static inline void security_xfrm_policy_free(struct xfrm_policy *xp)
+static inline void security_xfrm_policy_free(struct xfrm_sec_ctx *ctx)
 {
 }
 
-static inline int security_xfrm_policy_delete(struct xfrm_policy *xp)
+static inline int security_xfrm_policy_delete(struct xfrm_sec_ctx *ctx)
 {
 	return 0;
 }
@@ -2619,7 +2619,7 @@ static inline int security_xfrm_state_delete(struct xfrm_state *x)
 	return 0;
 }
 
-static inline int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir)
+static inline int security_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir)
 {
 	return 0;
 }
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 6db58924368a..1fb0fe42a72e 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -2292,7 +2292,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h
 			goto out;
 		}
 
-		err = security_xfrm_policy_alloc(xp, uctx);
+		err = security_xfrm_policy_alloc(&xp->security, uctx);
 		kfree(uctx);
 
 		if (err)
@@ -2352,10 +2352,11 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg
 	int err;
 	struct sadb_address *sa;
 	struct sadb_x_policy *pol;
-	struct xfrm_policy *xp, tmp;
+	struct xfrm_policy *xp;
 	struct xfrm_selector sel;
 	struct km_event c;
 	struct sadb_x_sec_ctx *sec_ctx;
+	struct xfrm_sec_ctx *pol_ctx;
 
 	if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1],
 				     ext_hdrs[SADB_EXT_ADDRESS_DST-1]) ||
@@ -2385,25 +2386,23 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg
 		sel.dport_mask = htons(0xffff);
 
 	sec_ctx = (struct sadb_x_sec_ctx *) ext_hdrs[SADB_X_EXT_SEC_CTX-1];
-	memset(&tmp, 0, sizeof(struct xfrm_policy));
-
 	if (sec_ctx != NULL) {
 		struct xfrm_user_sec_ctx *uctx = pfkey_sadb2xfrm_user_sec_ctx(sec_ctx);
 
 		if (!uctx)
 			return -ENOMEM;
 
-		err = security_xfrm_policy_alloc(&tmp, uctx);
+		err = security_xfrm_policy_alloc(&pol_ctx, uctx);
 		kfree(uctx);
-
 		if (err)
 			return err;
-	}
-
-	xp = xfrm_policy_bysel_ctx(XFRM_POLICY_TYPE_MAIN, pol->sadb_x_policy_dir-1,
-				   &sel, tmp.security, 1, &err);
-	security_xfrm_policy_free(&tmp);
+	} else
+		pol_ctx = NULL;
 
+	xp = xfrm_policy_bysel_ctx(XFRM_POLICY_TYPE_MAIN,
+				   pol->sadb_x_policy_dir - 1, &sel, pol_ctx,
+				   1, &err);
+	security_xfrm_policy_free(pol_ctx);
 	if (xp == NULL)
 		return -ENOENT;
 
@@ -3298,7 +3297,7 @@ static struct xfrm_policy *pfkey_compile_policy(struct sock *sk, int opt,
 		if ((*dir = verify_sec_ctx_len(p)))
 			goto out;
 		uctx = pfkey_sadb2xfrm_user_sec_ctx(sec_ctx);
-		*dir = security_xfrm_policy_alloc(xp, uctx);
+		*dir = security_xfrm_policy_alloc(&xp->security, uctx);
 		kfree(uctx);
 
 		if (*dir)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 15d73e47cc2c..ab4d0e598a2c 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -263,7 +263,7 @@ void xfrm_policy_destroy(struct xfrm_policy *policy)
 	list_del(&policy->bytype);
 	write_unlock_bh(&xfrm_policy_lock);
 
-	security_xfrm_policy_free(policy);
+	security_xfrm_policy_free(policy->security);
 	kfree(policy);
 }
 EXPORT_SYMBOL(xfrm_policy_destroy);
@@ -676,7 +676,8 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir,
 		    xfrm_sec_ctx_match(ctx, pol->security)) {
 			xfrm_pol_hold(pol);
 			if (delete) {
-				*err = security_xfrm_policy_delete(pol);
+				*err = security_xfrm_policy_delete(
+								pol->security);
 				if (*err) {
 					write_unlock_bh(&xfrm_policy_lock);
 					return pol;
@@ -718,7 +719,8 @@ struct xfrm_policy *xfrm_policy_byid(u8 type, int dir, u32 id, int delete,
 		if (pol->type == type && pol->index == id) {
 			xfrm_pol_hold(pol);
 			if (delete) {
-				*err = security_xfrm_policy_delete(pol);
+				*err = security_xfrm_policy_delete(
+								pol->security);
 				if (*err) {
 					write_unlock_bh(&xfrm_policy_lock);
 					return pol;
@@ -756,7 +758,7 @@ xfrm_policy_flush_secctx_check(u8 type, struct xfrm_audit *audit_info)
 				     &xfrm_policy_inexact[dir], bydst) {
 			if (pol->type != type)
 				continue;
-			err = security_xfrm_policy_delete(pol);
+			err = security_xfrm_policy_delete(pol->security);
 			if (err) {
 				xfrm_audit_policy_delete(pol, 0,
 							 audit_info->loginuid,
@@ -770,7 +772,8 @@ xfrm_policy_flush_secctx_check(u8 type, struct xfrm_audit *audit_info)
 					     bydst) {
 				if (pol->type != type)
 					continue;
-				err = security_xfrm_policy_delete(pol);
+				err = security_xfrm_policy_delete(
+								pol->security);
 				if (err) {
 					xfrm_audit_policy_delete(pol, 0,
 							audit_info->loginuid,
@@ -931,7 +934,8 @@ static int xfrm_policy_match(struct xfrm_policy *pol, struct flowi *fl,
 
 	match = xfrm_selector_match(sel, fl, family);
 	if (match)
-		ret = security_xfrm_policy_lookup(pol, fl->secid, dir);
+		ret = security_xfrm_policy_lookup(pol->security, fl->secid,
+						  dir);
 
 	return ret;
 }
@@ -1048,8 +1052,9 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc
 		int err = 0;
 
 		if (match) {
-			err = security_xfrm_policy_lookup(pol, fl->secid,
-					policy_to_flow_dir(dir));
+			err = security_xfrm_policy_lookup(pol->security,
+						      fl->secid,
+						      policy_to_flow_dir(dir));
 			if (!err)
 				xfrm_pol_hold(pol);
 			else if (err == -ESRCH)
@@ -1138,7 +1143,8 @@ static struct xfrm_policy *clone_policy(struct xfrm_policy *old, int dir)
 
 	if (newp) {
 		newp->selector = old->selector;
-		if (security_xfrm_policy_clone(old, newp)) {
+		if (security_xfrm_policy_clone(old->security,
+					       &newp->security)) {
 			kfree(newp);
 			return NULL;  /* ENOMEM */
 		}
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 5578c909fcf6..ecf9d67daef5 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -959,7 +959,7 @@ static int copy_from_user_sec_ctx(struct xfrm_policy *pol, struct nlattr **attrs
 		return 0;
 
 	uctx = nla_data(rt);
-	return security_xfrm_policy_alloc(pol, uctx);
+	return security_xfrm_policy_alloc(&pol->security, uctx);
 }
 
 static void copy_templates(struct xfrm_policy *xp, struct xfrm_user_tmpl *ut,
@@ -1143,7 +1143,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh,
 			      NETLINK_CB(skb).sid);
 
 	if (err) {
-		security_xfrm_policy_free(xp);
+		security_xfrm_policy_free(xp->security);
 		kfree(xp);
 		return err;
 	}
@@ -1337,22 +1337,23 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh,
 		xp = xfrm_policy_byid(type, p->dir, p->index, delete, &err);
 	else {
 		struct nlattr *rt = attrs[XFRMA_SEC_CTX];
-		struct xfrm_policy tmp;
+		struct xfrm_sec_ctx *ctx;
 
 		err = verify_sec_ctx_len(attrs);
 		if (err)
 			return err;
 
-		memset(&tmp, 0, sizeof(struct xfrm_policy));
 		if (rt) {
 			struct xfrm_user_sec_ctx *uctx = nla_data(rt);
 
-			if ((err = security_xfrm_policy_alloc(&tmp, uctx)))
+			err = security_xfrm_policy_alloc(&ctx, uctx);
+			if (err)
 				return err;
-		}
-		xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, tmp.security,
+		} else
+			ctx = NULL;
+		xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, ctx,
 					   delete, &err);
-		security_xfrm_policy_free(&tmp);
+		security_xfrm_policy_free(ctx);
 	}
 	if (xp == NULL)
 		return -ENOENT;
@@ -1572,26 +1573,26 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh,
 		xp = xfrm_policy_byid(type, p->dir, p->index, 0, &err);
 	else {
 		struct nlattr *rt = attrs[XFRMA_SEC_CTX];
-		struct xfrm_policy tmp;
+		struct xfrm_sec_ctx *ctx;
 
 		err = verify_sec_ctx_len(attrs);
 		if (err)
 			return err;
 
-		memset(&tmp, 0, sizeof(struct xfrm_policy));
 		if (rt) {
 			struct xfrm_user_sec_ctx *uctx = nla_data(rt);
 
-			if ((err = security_xfrm_policy_alloc(&tmp, uctx)))
+			err = security_xfrm_policy_alloc(&ctx, uctx);
+			if (err)
 				return err;
-		}
-		xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, tmp.security,
-					   0, &err);
-		security_xfrm_policy_free(&tmp);
+		} else
+			ctx = NULL;
+		xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, ctx, 0, &err);
+		security_xfrm_policy_free(ctx);
 	}
-
 	if (xp == NULL)
 		return -ENOENT;
+
 	read_lock(&xp->lock);
 	if (xp->dead) {
 		read_unlock(&xp->lock);
diff --git a/security/dummy.c b/security/dummy.c
index 78d8f92310a4..480366f9c41d 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -876,22 +876,23 @@ static inline void dummy_req_classify_flow(const struct request_sock *req,
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
-static int dummy_xfrm_policy_alloc_security(struct xfrm_policy *xp,
-		struct xfrm_user_sec_ctx *sec_ctx)
+static int dummy_xfrm_policy_alloc_security(struct xfrm_sec_ctx **ctxp,
+					    struct xfrm_user_sec_ctx *sec_ctx)
 {
 	return 0;
 }
 
-static inline int dummy_xfrm_policy_clone_security(struct xfrm_policy *old, struct xfrm_policy *new)
+static inline int dummy_xfrm_policy_clone_security(struct xfrm_sec_ctx *old_ctx,
+					   struct xfrm_sec_ctx **new_ctxp)
 {
 	return 0;
 }
 
-static void dummy_xfrm_policy_free_security(struct xfrm_policy *xp)
+static void dummy_xfrm_policy_free_security(struct xfrm_sec_ctx *ctx)
 {
 }
 
-static int dummy_xfrm_policy_delete_security(struct xfrm_policy *xp)
+static int dummy_xfrm_policy_delete_security(struct xfrm_sec_ctx *ctx)
 {
 	return 0;
 }
@@ -911,7 +912,8 @@ static int dummy_xfrm_state_delete_security(struct xfrm_state *x)
 	return 0;
 }
 
-static int dummy_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir)
+static int dummy_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx,
+				    u32 sk_sid, u8 dir)
 {
 	return 0;
 }
diff --git a/security/security.c b/security/security.c
index b1387a6b416d..c9ff7d18c2f4 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1014,26 +1014,27 @@ void security_inet_conn_established(struct sock *sk,
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
 
-int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx)
+int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx)
 {
-	return security_ops->xfrm_policy_alloc_security(xp, sec_ctx);
+	return security_ops->xfrm_policy_alloc_security(ctxp, sec_ctx);
 }
 EXPORT_SYMBOL(security_xfrm_policy_alloc);
 
-int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new)
+int security_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx,
+			      struct xfrm_sec_ctx **new_ctxp)
 {
-	return security_ops->xfrm_policy_clone_security(old, new);
+	return security_ops->xfrm_policy_clone_security(old_ctx, new_ctxp);
 }
 
-void security_xfrm_policy_free(struct xfrm_policy *xp)
+void security_xfrm_policy_free(struct xfrm_sec_ctx *ctx)
 {
-	security_ops->xfrm_policy_free_security(xp);
+	security_ops->xfrm_policy_free_security(ctx);
 }
 EXPORT_SYMBOL(security_xfrm_policy_free);
 
-int security_xfrm_policy_delete(struct xfrm_policy *xp)
+int security_xfrm_policy_delete(struct xfrm_sec_ctx *ctx)
 {
-	return security_ops->xfrm_policy_delete_security(xp);
+	return security_ops->xfrm_policy_delete_security(ctx);
 }
 
 int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx)
@@ -1065,9 +1066,9 @@ void security_xfrm_state_free(struct xfrm_state *x)
 	security_ops->xfrm_state_free_security(x);
 }
 
-int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir)
+int security_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir)
 {
-	return security_ops->xfrm_policy_lookup(xp, fl_secid, dir);
+	return security_ops->xfrm_policy_lookup(ctx, fl_secid, dir);
 }
 
 int security_xfrm_state_pol_flow_match(struct xfrm_state *x,
diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h
index 36b0510efa7b..289e24b39e3e 100644
--- a/security/selinux/include/xfrm.h
+++ b/security/selinux/include/xfrm.h
@@ -7,16 +7,17 @@
 #ifndef _SELINUX_XFRM_H_
 #define _SELINUX_XFRM_H_
 
-int selinux_xfrm_policy_alloc(struct xfrm_policy *xp,
-		struct xfrm_user_sec_ctx *sec_ctx);
-int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new);
-void selinux_xfrm_policy_free(struct xfrm_policy *xp);
-int selinux_xfrm_policy_delete(struct xfrm_policy *xp);
+int selinux_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp,
+			      struct xfrm_user_sec_ctx *sec_ctx);
+int selinux_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx,
+			      struct xfrm_sec_ctx **new_ctxp);
+void selinux_xfrm_policy_free(struct xfrm_sec_ctx *ctx);
+int selinux_xfrm_policy_delete(struct xfrm_sec_ctx *ctx);
 int selinux_xfrm_state_alloc(struct xfrm_state *x,
 	struct xfrm_user_sec_ctx *sec_ctx, u32 secid);
 void selinux_xfrm_state_free(struct xfrm_state *x);
 int selinux_xfrm_state_delete(struct xfrm_state *x);
-int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir);
+int selinux_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir);
 int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x,
 			struct xfrm_policy *xp, struct flowi *fl);
 
diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c
index 7e158205d081..874d17c83c61 100644
--- a/security/selinux/xfrm.c
+++ b/security/selinux/xfrm.c
@@ -77,20 +77,18 @@ static inline int selinux_authorizable_xfrm(struct xfrm_state *x)
  * LSM hook implementation that authorizes that a flow can use
  * a xfrm policy rule.
  */
-int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir)
+int selinux_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir)
 {
 	int rc;
 	u32 sel_sid;
-	struct xfrm_sec_ctx *ctx;
 
 	/* Context sid is either set to label or ANY_ASSOC */
-	if ((ctx = xp->security)) {
+	if (ctx) {
 		if (!selinux_authorizable_ctx(ctx))
 			return -EINVAL;
 
 		sel_sid = ctx->ctx_sid;
-	}
-	else
+	} else
 		/*
 		 * All flows should be treated as polmatch'ing an
 		 * otherwise applicable "non-labeled" policy. This
@@ -103,7 +101,7 @@ int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir)
 			  NULL);
 
 	if (rc == -EACCES)
-		rc = -ESRCH;
+		return -ESRCH;
 
 	return rc;
 }
@@ -287,15 +285,14 @@ out2:
  * LSM hook implementation that allocs and transfers uctx spec to
  * xfrm_policy.
  */
-int selinux_xfrm_policy_alloc(struct xfrm_policy *xp,
-		struct xfrm_user_sec_ctx *uctx)
+int selinux_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp,
+			      struct xfrm_user_sec_ctx *uctx)
 {
 	int err;
 
-	BUG_ON(!xp);
 	BUG_ON(!uctx);
 
-	err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, 0);
+	err = selinux_xfrm_sec_ctx_alloc(ctxp, uctx, 0);
 	if (err == 0)
 		atomic_inc(&selinux_xfrm_refcount);
 
@@ -307,32 +304,29 @@ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp,
  * LSM hook implementation that copies security data structure from old to
  * new for policy cloning.
  */
-int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new)
+int selinux_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx,
+			      struct xfrm_sec_ctx **new_ctxp)
 {
-	struct xfrm_sec_ctx *old_ctx, *new_ctx;
-
-	old_ctx = old->security;
+	struct xfrm_sec_ctx *new_ctx;
 
 	if (old_ctx) {
-		new_ctx = new->security = kmalloc(sizeof(*new_ctx) +
-						  old_ctx->ctx_len,
-						  GFP_KERNEL);
-
+		new_ctx = kmalloc(sizeof(*old_ctx) + old_ctx->ctx_len,
+				  GFP_KERNEL);
 		if (!new_ctx)
 			return -ENOMEM;
 
 		memcpy(new_ctx, old_ctx, sizeof(*new_ctx));
 		memcpy(new_ctx->ctx_str, old_ctx->ctx_str, new_ctx->ctx_len);
+		*new_ctxp = new_ctx;
 	}
 	return 0;
 }
 
 /*
- * LSM hook implementation that frees xfrm_policy security information.
+ * LSM hook implementation that frees xfrm_sec_ctx security information.
  */
-void selinux_xfrm_policy_free(struct xfrm_policy *xp)
+void selinux_xfrm_policy_free(struct xfrm_sec_ctx *ctx)
 {
-	struct xfrm_sec_ctx *ctx = xp->security;
 	if (ctx)
 		kfree(ctx);
 }
@@ -340,10 +334,9 @@ void selinux_xfrm_policy_free(struct xfrm_policy *xp)
 /*
  * LSM hook implementation that authorizes deletion of labeled policies.
  */
-int selinux_xfrm_policy_delete(struct xfrm_policy *xp)
+int selinux_xfrm_policy_delete(struct xfrm_sec_ctx *ctx)
 {
 	struct task_security_struct *tsec = current->security;
-	struct xfrm_sec_ctx *ctx = xp->security;
 	int rc = 0;
 
 	if (ctx) {
-- 
cgit v1.2.3


From a4146b1b2c6ba995db08b1a2aef5af1b17b151e6 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sun, 13 Apr 2008 22:11:14 -0700
Subject: [TCP]: Replace struct net on tcp_iter_state with seq_net_private.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   |  2 +-
 net/ipv4/tcp_ipv4.c | 11 +++++------
 2 files changed, 6 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 7b41bb962b9c..f5b61e0f01f2 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1333,7 +1333,7 @@ struct tcp_seq_afinfo {
 };
 
 struct tcp_iter_state {
-	struct net              *net;
+	struct seq_net_private	p;
 	sa_family_t		family;
 	enum tcp_seq_states	state;
 	struct sock		*syn_wait_sk;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 52e3ae603ca9..86148cdfb21f 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1954,7 +1954,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur)
 	struct hlist_node *node;
 	struct sock *sk = cur;
 	struct tcp_iter_state* st = seq->private;
-	struct net *net = st->net;
+	struct net *net = seq_file_net(seq);
 
 	if (!sk) {
 		st->bucket = 0;
@@ -2035,7 +2035,7 @@ static void *listening_get_idx(struct seq_file *seq, loff_t *pos)
 static void *established_get_first(struct seq_file *seq)
 {
 	struct tcp_iter_state* st = seq->private;
-	struct net *net = st->net;
+	struct net *net = seq_file_net(seq);
 	void *rc = NULL;
 
 	for (st->bucket = 0; st->bucket < tcp_hashinfo.ehash_size; ++st->bucket) {
@@ -2076,7 +2076,7 @@ static void *established_get_next(struct seq_file *seq, void *cur)
 	struct inet_timewait_sock *tw;
 	struct hlist_node *node;
 	struct tcp_iter_state* st = seq->private;
-	struct net *net = st->net;
+	struct net *net = seq_file_net(seq);
 
 	++st->num;
 
@@ -2233,7 +2233,7 @@ static int tcp_seq_open(struct inode *inode, struct file *file)
 	s->seq_ops.next		= tcp_seq_next;
 	s->seq_ops.show		= afinfo->seq_show;
 	s->seq_ops.stop		= tcp_seq_stop;
-	s->net                  = net;
+	s->p.net                = net;
 
 	rc = seq_open(file, &s->seq_ops);
 	if (rc)
@@ -2252,9 +2252,8 @@ out_kfree:
 static int tcp_seq_release(struct inode *inode, struct file *file)
 {
 	struct seq_file *seq = file->private_data;
-	struct tcp_iter_state *s = seq->private;
 
-	put_net(s->net);
+	put_net(seq_file_net(seq));
 	seq_release_private(inode, file);
 	return 0;
 }
-- 
cgit v1.2.3


From 9427c4b36b8fe652df1d7c89eae678948e1f4b32 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sun, 13 Apr 2008 22:12:13 -0700
Subject: [TCP]: Move seq_ops from tcp_iter_state to tcp_seq_afinfo.

No need to create seq_operations for each instance of 'netstat'.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   |  3 +--
 net/ipv4/tcp_ipv4.c | 14 ++++++++------
 net/ipv6/tcp_ipv6.c |  4 +++-
 3 files changed, 12 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index f5b61e0f01f2..2c9a650412c0 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1328,8 +1328,8 @@ struct tcp_seq_afinfo {
 	struct module		*owner;
 	char			*name;
 	sa_family_t		family;
-	int			(*seq_show) (struct seq_file *m, void *v);
 	struct file_operations	*seq_fops;
+	struct seq_operations	seq_ops;
 };
 
 struct tcp_iter_state {
@@ -1338,7 +1338,6 @@ struct tcp_iter_state {
 	enum tcp_seq_states	state;
 	struct sock		*syn_wait_sk;
 	int			bucket, sbucket, num, uid;
-	struct seq_operations	seq_ops;
 };
 
 extern int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 978d9db1df34..c50dd1793643 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2229,13 +2229,9 @@ static int tcp_seq_open(struct inode *inode, struct file *file)
 		goto out_kfree;
 
 	s->family		= afinfo->family;
-	s->seq_ops.start	= tcp_seq_start;
-	s->seq_ops.next		= tcp_seq_next;
-	s->seq_ops.show		= afinfo->seq_show;
-	s->seq_ops.stop		= tcp_seq_stop;
 	s->p.net                = net;
 
-	rc = seq_open(file, &s->seq_ops);
+	rc = seq_open(file, &afinfo->seq_ops);
 	if (rc)
 		goto out_put_net;
 	seq = file->private_data;
@@ -2269,6 +2265,10 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo)
 	afinfo->seq_fops->llseek	= seq_lseek;
 	afinfo->seq_fops->release	= tcp_seq_release;
 
+	afinfo->seq_ops.start		= tcp_seq_start;
+	afinfo->seq_ops.next		= tcp_seq_next;
+	afinfo->seq_ops.stop		= tcp_seq_stop;
+
 	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops);
 	if (p)
 		p->data = afinfo;
@@ -2414,8 +2414,10 @@ static struct tcp_seq_afinfo tcp4_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "tcp",
 	.family		= AF_INET,
-	.seq_show	= tcp4_seq_show,
 	.seq_fops	= &tcp4_seq_fops,
+	.seq_ops	= {
+		.show		= tcp4_seq_show,
+	},
 };
 
 static int tcp4_proc_init_net(struct net *net)
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 80eab71e77ff..8bf59ee51cdb 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2125,8 +2125,10 @@ static struct tcp_seq_afinfo tcp6_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "tcp6",
 	.family		= AF_INET6,
-	.seq_show	= tcp6_seq_show,
 	.seq_fops	= &tcp6_seq_fops,
+	.seq_ops	= {
+		.show		= tcp6_seq_show,
+	},
 };
 
 int tcp6_proc_init(struct net *net)
-- 
cgit v1.2.3


From 68fcadd16c371d5e0698ba366f33a4f990ce83ce Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sun, 13 Apr 2008 22:13:30 -0700
Subject: [TCP]: Place file operations directly into tcp_seq_afinfo.

No need to have separate never-used variable.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   |  2 +-
 net/ipv4/tcp_ipv4.c | 15 ++++++---------
 net/ipv6/tcp_ipv6.c |  2 --
 3 files changed, 7 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 2c9a650412c0..93479ebf3e33 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1328,7 +1328,7 @@ struct tcp_seq_afinfo {
 	struct module		*owner;
 	char			*name;
 	sa_family_t		family;
-	struct file_operations	*seq_fops;
+	struct file_operations	seq_fops;
 	struct seq_operations	seq_ops;
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 840346b390d7..3696c83aec19 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2232,17 +2232,17 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo)
 	int rc = 0;
 	struct proc_dir_entry *p;
 
-	afinfo->seq_fops->owner		= afinfo->owner;
-	afinfo->seq_fops->open		= tcp_seq_open;
-	afinfo->seq_fops->read		= seq_read;
-	afinfo->seq_fops->llseek	= seq_lseek;
-	afinfo->seq_fops->release	= seq_release_net;
+	afinfo->seq_fops.owner		= afinfo->owner;
+	afinfo->seq_fops.open		= tcp_seq_open;
+	afinfo->seq_fops.read		= seq_read;
+	afinfo->seq_fops.llseek		= seq_lseek;
+	afinfo->seq_fops.release	= seq_release_net;
 
 	afinfo->seq_ops.start		= tcp_seq_start;
 	afinfo->seq_ops.next		= tcp_seq_next;
 	afinfo->seq_ops.stop		= tcp_seq_stop;
 
-	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops);
+	p = proc_net_fops_create(net, afinfo->name, S_IRUGO, &afinfo->seq_fops);
 	if (p)
 		p->data = afinfo;
 	else
@@ -2253,7 +2253,6 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo)
 void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo)
 {
 	proc_net_remove(net, afinfo->name);
-	memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops));
 }
 
 static void get_openreq4(struct sock *sk, struct request_sock *req,
@@ -2382,12 +2381,10 @@ out:
 	return 0;
 }
 
-static struct file_operations tcp4_seq_fops;
 static struct tcp_seq_afinfo tcp4_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "tcp",
 	.family		= AF_INET,
-	.seq_fops	= &tcp4_seq_fops,
 	.seq_ops	= {
 		.show		= tcp4_seq_show,
 	},
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 8bf59ee51cdb..e33a3dc7a000 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2120,12 +2120,10 @@ out:
 	return 0;
 }
 
-static struct file_operations tcp6_seq_fops;
 static struct tcp_seq_afinfo tcp6_seq_afinfo = {
 	.owner		= THIS_MODULE,
 	.name		= "tcp6",
 	.family		= AF_INET6,
-	.seq_fops	= &tcp6_seq_fops,
 	.seq_ops	= {
 		.show		= tcp6_seq_show,
 	},
-- 
cgit v1.2.3


From 5f4472c5a640c9671ca5becaebdfd6e651482176 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Sun, 13 Apr 2008 22:13:53 -0700
Subject: [TCP]: Remove owner from tcp_seq_afinfo.

Move it to tcp_seq_afinfo->seq_fops as should be.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   | 1 -
 net/ipv4/tcp_ipv4.c | 5 +++--
 net/ipv6/tcp_ipv6.c | 4 +++-
 3 files changed, 6 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 93479ebf3e33..58d82822414d 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1325,7 +1325,6 @@ enum tcp_seq_states {
 };
 
 struct tcp_seq_afinfo {
-	struct module		*owner;
 	char			*name;
 	sa_family_t		family;
 	struct file_operations	seq_fops;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3696c83aec19..02519730e0d5 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2232,7 +2232,6 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo)
 	int rc = 0;
 	struct proc_dir_entry *p;
 
-	afinfo->seq_fops.owner		= afinfo->owner;
 	afinfo->seq_fops.open		= tcp_seq_open;
 	afinfo->seq_fops.read		= seq_read;
 	afinfo->seq_fops.llseek		= seq_lseek;
@@ -2382,9 +2381,11 @@ out:
 }
 
 static struct tcp_seq_afinfo tcp4_seq_afinfo = {
-	.owner		= THIS_MODULE,
 	.name		= "tcp",
 	.family		= AF_INET,
+	.seq_fops	= {
+		.owner		= THIS_MODULE,
+	},
 	.seq_ops	= {
 		.show		= tcp4_seq_show,
 	},
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index e33a3dc7a000..231c4dddfb8c 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2121,9 +2121,11 @@ out:
 }
 
 static struct tcp_seq_afinfo tcp6_seq_afinfo = {
-	.owner		= THIS_MODULE,
 	.name		= "tcp6",
 	.family		= AF_INET6,
+	.seq_fops	= {
+		.owner		= THIS_MODULE,
+	},
 	.seq_ops	= {
 		.show		= tcp6_seq_show,
 	},
-- 
cgit v1.2.3


From 67019cc9ee3f4868c8e5e493b2873c4722306019 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sun, 13 Apr 2008 22:28:42 -0700
Subject: [NETNS]: Add an empty netns_dccp structure on struct net.

According to the overall struct net design, it will be
filled with DCCP-related members.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h | 4 ++++
 include/net/netns/dccp.h    | 7 +++++++
 2 files changed, 11 insertions(+)
 create mode 100644 include/net/netns/dccp.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 0ab62ed2fdef..e2aee2689abe 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -13,6 +13,7 @@
 #include <net/netns/packet.h>
 #include <net/netns/ipv4.h>
 #include <net/netns/ipv6.h>
+#include <net/netns/dccp.h>
 #include <net/netns/x_tables.h>
 
 struct proc_dir_entry;
@@ -54,6 +55,9 @@ struct net {
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 	struct netns_ipv6	ipv6;
 #endif
+#if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE)
+	struct netns_dccp	dccp;
+#endif
 #ifdef CONFIG_NETFILTER
 	struct netns_xt		xt;
 #endif
diff --git a/include/net/netns/dccp.h b/include/net/netns/dccp.h
new file mode 100644
index 000000000000..056e6adc70b6
--- /dev/null
+++ b/include/net/netns/dccp.h
@@ -0,0 +1,7 @@
+#ifndef __NETNS_DCCP_H__
+#define __NETNS_DCCP_H__
+
+struct netns_dccp {
+};
+
+#endif
-- 
cgit v1.2.3


From 7b1cffa8c90269dc3dc721d084d1e0d742d87c31 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sun, 13 Apr 2008 22:29:37 -0700
Subject: [NETNS][DCCPV4]: Move the dccp_v4_ctl_sk on the struct net.

And replace all its usage with init_net's socket.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/dccp.h |  3 +++
 net/dccp/ipv4.c          | 20 ++++++++++----------
 2 files changed, 13 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/dccp.h b/include/net/netns/dccp.h
index 056e6adc70b6..e4f16faa898a 100644
--- a/include/net/netns/dccp.h
+++ b/include/net/netns/dccp.h
@@ -1,7 +1,10 @@
 #ifndef __NETNS_DCCP_H__
 #define __NETNS_DCCP_H__
 
+struct sock;
+
 struct netns_dccp {
+	struct sock *v4_ctl_sk;
 };
 
 #endif
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 10bba039c4b8..85931dce744c 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -32,11 +32,10 @@
 #include "feat.h"
 
 /*
- * This is the global socket data structure used for responding to
+ * The dccp_ctl_sk is the global socket data structure used for responding to
  * the Out-of-the-blue (OOTB) packets. A control sock will be created
  * for this socket at the initialization time.
  */
-static struct sock *dccp_v4_ctl_sk;
 
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
@@ -506,6 +505,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 	const struct iphdr *rxiph;
 	struct sk_buff *skb;
 	struct dst_entry *dst;
+	struct sock *ctl_sk = init_net.dccp.v4_ctl_sk;
 
 	/* Never send a reset in response to a reset. */
 	if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
@@ -514,11 +514,11 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 	if (rxskb->rtable->rt_type != RTN_LOCAL)
 		return;
 
-	dst = dccp_v4_route_skb(dccp_v4_ctl_sk, rxskb);
+	dst = dccp_v4_route_skb(ctl_sk, rxskb);
 	if (dst == NULL)
 		return;
 
-	skb = dccp_ctl_make_reset(dccp_v4_ctl_sk, rxskb);
+	skb = dccp_ctl_make_reset(ctl_sk, rxskb);
 	if (skb == NULL)
 		goto out;
 
@@ -527,10 +527,10 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 								 rxiph->daddr);
 	skb->dst = dst_clone(dst);
 
-	bh_lock_sock(dccp_v4_ctl_sk);
-	err = ip_build_and_send_pkt(skb, dccp_v4_ctl_sk,
+	bh_lock_sock(ctl_sk);
+	err = ip_build_and_send_pkt(skb, ctl_sk,
 				    rxiph->daddr, rxiph->saddr, NULL);
-	bh_unlock_sock(dccp_v4_ctl_sk);
+	bh_unlock_sock(ctl_sk);
 
 	if (net_xmit_eval(err) == 0) {
 		DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS);
@@ -1016,7 +1016,7 @@ static int __init dccp_v4_init(void)
 
 	inet_register_protosw(&dccp_v4_protosw);
 
-	err = inet_ctl_sock_create(&dccp_v4_ctl_sk, PF_INET,
+	err = inet_ctl_sock_create(&init_net.dccp.v4_ctl_sk, PF_INET,
 				   SOCK_DCCP, IPPROTO_DCCP, &init_net);
 	if (err)
 		goto out_unregister_protosw;
@@ -1027,7 +1027,7 @@ static int __init dccp_v4_init(void)
 out:
 	return err;
 out_destroy_ctl_sock:
-	inet_ctl_sock_destroy(dccp_v4_ctl_sk);
+	inet_ctl_sock_destroy(init_net.dccp.v4_ctl_sk);
 out_unregister_protosw:
 	inet_unregister_protosw(&dccp_v4_protosw);
 	inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP);
@@ -1039,7 +1039,7 @@ out_proto_unregister:
 static void __exit dccp_v4_exit(void)
 {
 	unregister_pernet_subsys(&dccp_v4_ops);
-	inet_ctl_sock_destroy(dccp_v4_ctl_sk);
+	inet_ctl_sock_destroy(init_net.dccp.v4_ctl_sk);
 	inet_unregister_protosw(&dccp_v4_protosw);
 	inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP);
 	proto_unregister(&dccp_v4_prot);
-- 
cgit v1.2.3


From 0204774191d3e7bc69e3ae6bbf328b635607505a Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Sun, 13 Apr 2008 22:32:25 -0700
Subject: [NETNS][DCCPV6]: Move the dccp_v6_ctl_sk on the struct net.

And replace all its usage with init_net's socket.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netns/dccp.h |  1 +
 net/dccp/ipv6.c          | 16 ++++++++--------
 2 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/netns/dccp.h b/include/net/netns/dccp.h
index e4f16faa898a..98d2a7ce1f71 100644
--- a/include/net/netns/dccp.h
+++ b/include/net/netns/dccp.h
@@ -5,6 +5,7 @@ struct sock;
 
 struct netns_dccp {
 	struct sock *v4_ctl_sk;
+	struct sock *v6_ctl_sk;
 };
 
 #endif
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 2ca52913f4be..109dab3550f2 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -33,8 +33,7 @@
 #include "ipv6.h"
 #include "feat.h"
 
-/* Socket used for sending RSTs and ACKs */
-static struct sock *dccp_v6_ctl_sk;
+/* dccp_v6_ctl_sk is used for sending RSTs and ACKs */
 
 static struct inet_connection_sock_af_ops dccp_ipv6_mapped;
 static struct inet_connection_sock_af_ops dccp_ipv6_af_ops;
@@ -296,6 +295,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 	struct ipv6hdr *rxip6h;
 	struct sk_buff *skb;
 	struct flowi fl;
+	struct sock *ctl_sk = init_net.dccp.v6_ctl_sk;
 
 	if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
 		return;
@@ -303,7 +303,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 	if (!ipv6_unicast_destination(rxskb))
 		return;
 
-	skb = dccp_ctl_make_reset(dccp_v6_ctl_sk, rxskb);
+	skb = dccp_ctl_make_reset(ctl_sk, rxskb);
 	if (skb == NULL)
 		return;
 
@@ -322,9 +322,9 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
 	security_skb_classify_flow(rxskb, &fl);
 
 	/* sk = NULL, but it is safe for now. RST socket required. */
-	if (!ip6_dst_lookup(dccp_v6_ctl_sk, &skb->dst, &fl)) {
+	if (!ip6_dst_lookup(ctl_sk, &skb->dst, &fl)) {
 		if (xfrm_lookup(&skb->dst, &fl, NULL, 0) >= 0) {
-			ip6_xmit(dccp_v6_ctl_sk, skb, &fl, NULL, 0);
+			ip6_xmit(ctl_sk, skb, &fl, NULL, 0);
 			DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS);
 			DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS);
 			return;
@@ -1198,7 +1198,7 @@ static int __init dccp_v6_init(void)
 
 	inet6_register_protosw(&dccp_v6_protosw);
 
-	err = inet_ctl_sock_create(&dccp_v6_ctl_sk, PF_INET6,
+	err = inet_ctl_sock_create(&init_net.dccp.v6_ctl_sk, PF_INET6,
 				   SOCK_DCCP, IPPROTO_DCCP, &init_net);
 	if (err != 0)
 		goto out_unregister_protosw;
@@ -1210,7 +1210,7 @@ out:
 	return err;
 
 out_destroy_ctl_sock:
-	inet_ctl_sock_destroy(dccp_v6_ctl_sk);
+	inet_ctl_sock_destroy(init_net.dccp.v6_ctl_sk);
 out_unregister_protosw:
 	inet6_del_protocol(&dccp_v6_protocol, IPPROTO_DCCP);
 	inet6_unregister_protosw(&dccp_v6_protosw);
@@ -1222,7 +1222,7 @@ out_unregister_proto:
 static void __exit dccp_v6_exit(void)
 {
 	unregister_pernet_subsys(&dccp_v6_ops);
-	inet_ctl_sock_destroy(dccp_v6_ctl_sk);
+	inet_ctl_sock_destroy(init_net.dccp.v6_ctl_sk);
 	inet6_del_protocol(&dccp_v6_protocol, IPPROTO_DCCP);
 	inet6_unregister_protosw(&dccp_v6_protosw);
 	proto_unregister(&dccp_v6_prot);
-- 
cgit v1.2.3


From cee8947338d46bccece54c752bf6cd4043035f05 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sun, 13 Apr 2008 23:21:16 -0700
Subject: [IPV6] MROUTE: Do not call ipv6_find_idev() directly.

Since NETDEV_REGISTER notifier chain is responsible for creating
inet6_dev{}, we do not need to call ipv6_find_idev() directly here.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/mroute6.h | 3 ---
 net/ipv6/addrconf.c     | 2 +-
 net/ipv6/ip6mr.c        | 5 -----
 3 files changed, 1 insertion(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
index f6469fb90840..e7989593142b 100644
--- a/include/linux/mroute6.h
+++ b/include/linux/mroute6.h
@@ -117,9 +117,6 @@ struct sioc_mif_req6
 
 #include <linux/skbuff.h>	/* for struct sk_buff_head */
 
-struct net_device;
-struct inet6_dev *ipv6_find_idev(struct net_device *dev);
-
 #ifdef CONFIG_IPV6_MROUTE
 static inline int ip6_mroute_opt(int opt)
 {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index e93fa62089f8..9d49ed2578d7 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -412,7 +412,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 	return ndev;
 }
 
-struct inet6_dev * ipv6_find_idev(struct net_device *dev)
+static struct inet6_dev * ipv6_find_idev(struct net_device *dev)
 {
 	struct inet6_dev *idev;
 
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 94ede696da2a..6e2e3c957a31 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -435,7 +435,6 @@ static void reg_vif_setup(struct net_device *dev)
 static struct net_device *ip6mr_reg_vif(void)
 {
 	struct net_device *dev;
-	struct inet6_dev *in_dev;
 
 	dev = alloc_netdev(sizeof(struct net_device_stats), "pim6reg",
 			   reg_vif_setup);
@@ -449,10 +448,6 @@ static struct net_device *ip6mr_reg_vif(void)
 	}
 	dev->iflink = 0;
 
-	in_dev = ipv6_find_idev(dev);
-	if (!in_dev)
-		goto failure;
-
 	if (dev_open(dev))
 		goto failure;
 
-- 
cgit v1.2.3


From 7cd636fe9ce5de0051c11283911baca4291868c8 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sun, 13 Apr 2008 23:33:52 -0700
Subject: [IPV6]: IPv6 extension header structures need to be packed.

struct ipv6_opt_hdr is the common structure for IPv6 extension
headers, and it is common to increment the pointer to get
the real content.  On the other hand, since the structure
consists only of 1-byte next-header field and 1-byte length
field, size of that structure depends on architecture; 2 or 4.
Add "packed" attribute to get 2.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 4aaefc349a4b..134c8e5cf07c 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -53,7 +53,7 @@ struct ipv6_opt_hdr {
 	/* 
 	 * TLV encoded option data follows.
 	 */
-};
+} __attribute__ ((packed));	/* required for some archs */
 
 #define ipv6_destopt_hdr ipv6_opt_hdr
 #define ipv6_hopopt_hdr  ipv6_opt_hdr
-- 
cgit v1.2.3


From e9df2e8fd8fbc95c57dbd1d33dada66c4627b44c Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sun, 13 Apr 2008 23:40:51 -0700
Subject: [IPV6]: Use appropriate sock tclass setting for routing lookup.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_ecn.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h
index ba33db053854..7040a782c656 100644
--- a/include/net/inet_ecn.h
+++ b/include/net/inet_ecn.h
@@ -47,7 +47,7 @@ static inline __u8 INET_ECN_encapsulate(__u8 outer, __u8 inner)
     } while (0)
 
 #define	IP6_ECN_flow_xmit(sk, label) do {				\
-	if (INET_ECN_is_capable(inet_sk(sk)->tos))			\
+	if (INET_ECN_is_capable(inet6_sk(sk)->tclass))			\
 		(label) |= htonl(INET_ECN_ECT_0 << 20);			\
     } while (0)
 
-- 
cgit v1.2.3


From f525c06d12b72cddb085df7f6f348c3c5a39b3ce Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Mon, 14 Apr 2008 00:04:12 -0700
Subject: [SKB]: __skb_dequeue = skb_peek + __skb_unlink

By rearranging the order of declarations, __skb_dequeue() is expressed in terms of

 * skb_peek() and
 * __skb_unlink(),

thus in effect mirroring the analogue implementation of __skb_dequeue_tail().

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 47 ++++++++++++++++-------------------------------
 1 file changed, 16 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index e517701c25ba..c2116200580a 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -734,35 +734,6 @@ static inline void __skb_queue_tail(struct sk_buff_head *list,
 	next->prev  = prev->next = newsk;
 }
 
-
-/**
- *	__skb_dequeue - remove from the head of the queue
- *	@list: list to dequeue from
- *
- *	Remove the head of the list. This function does not take any locks
- *	so must be used with appropriate locks held only. The head item is
- *	returned or %NULL if the list is empty.
- */
-extern struct sk_buff *skb_dequeue(struct sk_buff_head *list);
-static inline struct sk_buff *__skb_dequeue(struct sk_buff_head *list)
-{
-	struct sk_buff *next, *prev, *result;
-
-	prev = (struct sk_buff *) list;
-	next = prev->next;
-	result = NULL;
-	if (next != prev) {
-		result	     = next;
-		next	     = next->next;
-		list->qlen--;
-		next->prev   = prev;
-		prev->next   = next;
-		result->next = result->prev = NULL;
-	}
-	return result;
-}
-
-
 /*
  *	Insert a packet on a list.
  */
@@ -803,8 +774,22 @@ static inline void __skb_unlink(struct sk_buff *skb, struct sk_buff_head *list)
 	prev->next = next;
 }
 
-
-/* XXX: more streamlined implementation */
+/**
+ *	__skb_dequeue - remove from the head of the queue
+ *	@list: list to dequeue from
+ *
+ *	Remove the head of the list. This function does not take any locks
+ *	so must be used with appropriate locks held only. The head item is
+ *	returned or %NULL if the list is empty.
+ */
+extern struct sk_buff *skb_dequeue(struct sk_buff_head *list);
+static inline struct sk_buff *__skb_dequeue(struct sk_buff_head *list)
+{
+	struct sk_buff *skb = skb_peek(list);
+	if (skb)
+		__skb_unlink(skb, list);
+	return skb;
+}
 
 /**
  *	__skb_dequeue_tail - remove from the tail of the queue
-- 
cgit v1.2.3


From bf299275882624b1908521ee8074df85160e9679 Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Mon, 14 Apr 2008 00:04:51 -0700
Subject: [SKB]: __skb_queue_after(prev) = __skb_insert(prev, prev->next)

By reordering, __skb_queue_after() is expressed in terms of __skb_insert().

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 34 ++++++++++++----------------------
 1 file changed, 12 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index c2116200580a..bb107ab675fc 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -663,11 +663,21 @@ static inline void skb_queue_head_init_class(struct sk_buff_head *list,
 }
 
 /*
- *	Insert an sk_buff at the start of a list.
+ *	Insert an sk_buff on a list.
  *
  *	The "__skb_xxxx()" functions are the non-atomic ones that
  *	can only be called with interrupts disabled.
  */
+extern void        skb_insert(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list);
+static inline void __skb_insert(struct sk_buff *newsk,
+				struct sk_buff *prev, struct sk_buff *next,
+				struct sk_buff_head *list)
+{
+	newsk->next = next;
+	newsk->prev = prev;
+	next->prev  = prev->next = newsk;
+	list->qlen++;
+}
 
 /**
  *	__skb_queue_after - queue a buffer at the list head
@@ -684,13 +694,7 @@ static inline void __skb_queue_after(struct sk_buff_head *list,
 				     struct sk_buff *prev,
 				     struct sk_buff *newsk)
 {
-	struct sk_buff *next;
-	list->qlen++;
-
-	next = prev->next;
-	newsk->next = next;
-	newsk->prev = prev;
-	next->prev  = prev->next = newsk;
+	__skb_insert(newsk, prev, prev->next, list);
 }
 
 /**
@@ -734,20 +738,6 @@ static inline void __skb_queue_tail(struct sk_buff_head *list,
 	next->prev  = prev->next = newsk;
 }
 
-/*
- *	Insert a packet on a list.
- */
-extern void        skb_insert(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list);
-static inline void __skb_insert(struct sk_buff *newsk,
-				struct sk_buff *prev, struct sk_buff *next,
-				struct sk_buff_head *list)
-{
-	newsk->next = next;
-	newsk->prev = prev;
-	next->prev  = prev->next = newsk;
-	list->qlen++;
-}
-
 /*
  *	Place a packet after a given packet in a list.
  */
-- 
cgit v1.2.3


From 7de6c033367ab86f39c7723392caf73325cbf286 Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Mon, 14 Apr 2008 00:05:09 -0700
Subject: [SKB]: __skb_append = __skb_queue_after

This expresses __skb_append in terms of __skb_queue_after, exploiting that

  __skb_append(old, new, list) = __skb_queue_after(list, old, new).

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 12 +++---------
 include/net/tcp.h      |  2 +-
 net/core/skbuff.c      |  2 +-
 net/ipv4/tcp_input.c   |  2 +-
 4 files changed, 6 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index bb107ab675fc..83c851846829 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -697,6 +697,9 @@ static inline void __skb_queue_after(struct sk_buff_head *list,
 	__skb_insert(newsk, prev, prev->next, list);
 }
 
+extern void skb_append(struct sk_buff *old, struct sk_buff *newsk,
+		       struct sk_buff_head *list);
+
 /**
  *	__skb_queue_head - queue a buffer at the list head
  *	@list: list to use
@@ -738,15 +741,6 @@ static inline void __skb_queue_tail(struct sk_buff_head *list,
 	next->prev  = prev->next = newsk;
 }
 
-/*
- *	Place a packet after a given packet in a list.
- */
-extern void	   skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list);
-static inline void __skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list)
-{
-	__skb_insert(newsk, old, old->next, list);
-}
-
 /*
  * remove sk_buff from list. _Must_ be called atomically, and with
  * the list known..
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 58d82822414d..2ab350eca02e 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1247,7 +1247,7 @@ static inline void tcp_insert_write_queue_after(struct sk_buff *skb,
 						struct sk_buff *buff,
 						struct sock *sk)
 {
-	__skb_append(skb, buff, &sk->sk_write_queue);
+	__skb_queue_after(&sk->sk_write_queue, skb, buff);
 }
 
 /* Insert skb between prev and next on the write queue of sk.  */
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index e4259215607f..4cd12d99b12e 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1860,7 +1860,7 @@ void skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head
 	unsigned long flags;
 
 	spin_lock_irqsave(&list->lock, flags);
-	__skb_append(old, newsk, list);
+	__skb_queue_after(list, old, newsk);
 	spin_unlock_irqrestore(&list->lock, flags);
 }
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 6e46b4c0f28c..743611956045 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3968,7 +3968,7 @@ drop:
 		u32 end_seq = TCP_SKB_CB(skb)->end_seq;
 
 		if (seq == TCP_SKB_CB(skb1)->end_seq) {
-			__skb_append(skb1, skb, &tp->out_of_order_queue);
+			__skb_queue_after(&tp->out_of_order_queue, skb1, skb);
 
 			if (!tp->rx_opt.num_sacks ||
 			    tp->selective_acks[0].end_seq != seq)
-- 
cgit v1.2.3


From f5572855ec492334d8c3ec0e0e86c31865d5cf07 Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Mon, 14 Apr 2008 00:05:28 -0700
Subject: [SKB]: __skb_queue_tail = __skb_insert before

This expresses __skb_queue_tail() in terms of __skb_insert(),
using __skb_insert_before() as auxiliary function.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 83c851846829..11fd9f2c4093 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -700,6 +700,13 @@ static inline void __skb_queue_after(struct sk_buff_head *list,
 extern void skb_append(struct sk_buff *old, struct sk_buff *newsk,
 		       struct sk_buff_head *list);
 
+static inline void __skb_queue_before(struct sk_buff_head *list,
+				      struct sk_buff *next,
+				      struct sk_buff *newsk)
+{
+	__skb_insert(newsk, next->prev, next, list);
+}
+
 /**
  *	__skb_queue_head - queue a buffer at the list head
  *	@list: list to use
@@ -731,14 +738,7 @@ extern void skb_queue_tail(struct sk_buff_head *list, struct sk_buff *newsk);
 static inline void __skb_queue_tail(struct sk_buff_head *list,
 				   struct sk_buff *newsk)
 {
-	struct sk_buff *prev, *next;
-
-	list->qlen++;
-	next = (struct sk_buff *)list;
-	prev = next->prev;
-	newsk->next = next;
-	newsk->prev = prev;
-	next->prev  = prev->next = newsk;
+	__skb_queue_before(list, (struct sk_buff *)list, newsk);
 }
 
 /*
-- 
cgit v1.2.3


From 666953df353194bef76086fa3f126241cbac3e3a Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@sw.ru>
Date: Mon, 14 Apr 2008 09:56:02 +0200
Subject: [NETFILTER]: ip_tables: per-netns FILTER/MANGLE/RAW tables for real

Commit 9335f047fe61587ec82ff12fbb1220bcfdd32006 aka
"[NETFILTER]: ip_tables: per-netns FILTER, MANGLE, RAW"
added per-netns _view_ of iptables rules. They were shown to user, but
ignored by filtering code. Now that it's possible to at least ping loopback,
per-netns tables can affect filtering decisions.

netns is taken in case of
	PRE_ROUTING, LOCAL_IN -- from in device,
	POST_ROUTING, LOCAL_OUT -- from out device,
	FORWARD -- from in device which should be equal to out device's netns.
		   This code is relatively new, so BUG_ON was plugged.

Wrappers were added to a) keep code the same from CONFIG_NET_NS=n users
(overwhelming majority), b) consolidate code in one place -- similar
changes will be done in ipv6 and arp netfilter code.

Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter.h           | 54 ++++++++++++++++++++++++++++++++++++-
 net/ipv4/netfilter/iptable_filter.c | 19 ++++++++++---
 net/ipv4/netfilter/iptable_mangle.c | 49 ++++++++++++++++++++++++++++-----
 net/ipv4/netfilter/iptable_raw.c    |  6 +++--
 4 files changed, 115 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 89e6c72ad295..66bc52060fd6 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -6,11 +6,13 @@
 #include <linux/types.h>
 #include <linux/skbuff.h>
 #include <linux/net.h>
+#include <linux/netdevice.h>
 #include <linux/if.h>
 #include <linux/in.h>
 #include <linux/in6.h>
 #include <linux/wait.h>
 #include <linux/list.h>
+#include <net/net_namespace.h>
 #endif
 #include <linux/compiler.h>
 
@@ -76,7 +78,6 @@ extern void netfilter_init(void);
 #define NF_MAX_HOOKS 8
 
 struct sk_buff;
-struct net_device;
 
 typedef unsigned int nf_hookfn(unsigned int hooknum,
 			       struct sk_buff *skb,
@@ -320,5 +321,56 @@ extern void (*nf_ct_destroy)(struct nf_conntrack *);
 static inline void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) {}
 #endif
 
+static inline struct net *nf_pre_routing_net(const struct net_device *in,
+					     const struct net_device *out)
+{
+#ifdef CONFIG_NET_NS
+	return in->nd_net;
+#else
+	return &init_net;
+#endif
+}
+
+static inline struct net *nf_local_in_net(const struct net_device *in,
+					  const struct net_device *out)
+{
+#ifdef CONFIG_NET_NS
+	return in->nd_net;
+#else
+	return &init_net;
+#endif
+}
+
+static inline struct net *nf_forward_net(const struct net_device *in,
+					 const struct net_device *out)
+{
+#ifdef CONFIG_NET_NS
+	BUG_ON(in->nd_net != out->nd_net);
+	return in->nd_net;
+#else
+	return &init_net;
+#endif
+}
+
+static inline struct net *nf_local_out_net(const struct net_device *in,
+					   const struct net_device *out)
+{
+#ifdef CONFIG_NET_NS
+	return out->nd_net;
+#else
+	return &init_net;
+#endif
+}
+
+static inline struct net *nf_post_routing_net(const struct net_device *in,
+					      const struct net_device *out)
+{
+#ifdef CONFIG_NET_NS
+	return out->nd_net;
+#else
+	return &init_net;
+#endif
+}
+
 #endif /*__KERNEL__*/
 #endif /*__LINUX_NETFILTER_H*/
diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c
index 69f3d7e6e96f..7fcf60adbbed 100644
--- a/net/ipv4/netfilter/iptable_filter.c
+++ b/net/ipv4/netfilter/iptable_filter.c
@@ -62,6 +62,17 @@ static struct xt_table packet_filter = {
 };
 
 /* The work comes in here from netfilter.c. */
+static unsigned int
+ipt_local_in_hook(unsigned int hook,
+		  struct sk_buff *skb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_in_net(in, out)->ipv4.iptable_filter);
+}
+
 static unsigned int
 ipt_hook(unsigned int hook,
 	 struct sk_buff *skb,
@@ -69,7 +80,8 @@ ipt_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_filter);
+	return ipt_do_table(skb, hook, in, out,
+			    nf_forward_net(in, out)->ipv4.iptable_filter);
 }
 
 static unsigned int
@@ -88,12 +100,13 @@ ipt_local_out_hook(unsigned int hook,
 		return NF_ACCEPT;
 	}
 
-	return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_filter);
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_out_net(in, out)->ipv4.iptable_filter);
 }
 
 static struct nf_hook_ops ipt_ops[] __read_mostly = {
 	{
-		.hook		= ipt_hook,
+		.hook		= ipt_local_in_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
 		.hooknum	= NF_INET_LOCAL_IN,
diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c
index c55a210853a7..ba827035b691 100644
--- a/net/ipv4/netfilter/iptable_mangle.c
+++ b/net/ipv4/netfilter/iptable_mangle.c
@@ -74,13 +74,47 @@ static struct xt_table packet_mangler = {
 
 /* The work comes in here from netfilter.c. */
 static unsigned int
-ipt_route_hook(unsigned int hook,
+ipt_pre_routing_hook(unsigned int hook,
+		     struct sk_buff *skb,
+		     const struct net_device *in,
+		     const struct net_device *out,
+		     int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_pre_routing_net(in, out)->ipv4.iptable_mangle);
+}
+
+static unsigned int
+ipt_post_routing_hook(unsigned int hook,
+		      struct sk_buff *skb,
+		      const struct net_device *in,
+		      const struct net_device *out,
+		      int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_post_routing_net(in, out)->ipv4.iptable_mangle);
+}
+
+static unsigned int
+ipt_local_in_hook(unsigned int hook,
+		  struct sk_buff *skb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_in_net(in, out)->ipv4.iptable_mangle);
+}
+
+static unsigned int
+ipt_forward_hook(unsigned int hook,
 	 struct sk_buff *skb,
 	 const struct net_device *in,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_mangle);
+	return ipt_do_table(skb, hook, in, out,
+			    nf_forward_net(in, out)->ipv4.iptable_mangle);
 }
 
 static unsigned int
@@ -112,7 +146,8 @@ ipt_local_hook(unsigned int hook,
 	daddr = iph->daddr;
 	tos = iph->tos;
 
-	ret = ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_mangle);
+	ret = ipt_do_table(skb, hook, in, out,
+			   nf_local_out_net(in, out)->ipv4.iptable_mangle);
 	/* Reroute for ANY change. */
 	if (ret != NF_DROP && ret != NF_STOLEN && ret != NF_QUEUE) {
 		iph = ip_hdr(skb);
@@ -130,21 +165,21 @@ ipt_local_hook(unsigned int hook,
 
 static struct nf_hook_ops ipt_ops[] __read_mostly = {
 	{
-		.hook		= ipt_route_hook,
+		.hook		= ipt_pre_routing_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
 		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 	{
-		.hook		= ipt_route_hook,
+		.hook		= ipt_local_in_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 	{
-		.hook		= ipt_route_hook,
+		.hook		= ipt_forward_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
 		.hooknum	= NF_INET_FORWARD,
@@ -158,7 +193,7 @@ static struct nf_hook_ops ipt_ops[] __read_mostly = {
 		.priority	= NF_IP_PRI_MANGLE,
 	},
 	{
-		.hook		= ipt_route_hook,
+		.hook		= ipt_post_routing_hook,
 		.owner		= THIS_MODULE,
 		.pf		= PF_INET,
 		.hooknum	= NF_INET_POST_ROUTING,
diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c
index e41fe8ca4e1c..4b689742d58b 100644
--- a/net/ipv4/netfilter/iptable_raw.c
+++ b/net/ipv4/netfilter/iptable_raw.c
@@ -52,7 +52,8 @@ ipt_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_raw);
+	return ipt_do_table(skb, hook, in, out,
+			    nf_pre_routing_net(in, out)->ipv4.iptable_raw);
 }
 
 static unsigned int
@@ -70,7 +71,8 @@ ipt_local_hook(unsigned int hook,
 			       "packet.\n");
 		return NF_ACCEPT;
 	}
-	return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_raw);
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_out_net(in, out)->ipv4.iptable_raw);
 }
 
 /* 'raw' is the very first table. */
-- 
cgit v1.2.3


From b9f61b160336da5eaaacb0cb41ebe32169e3bde5 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 09:56:04 +0200
Subject: [NETFILTER]: xt_sctp: simplify xt_sctp.h

The use of xt_sctp.h flagged up -Wshadow warnings in userspace, which
prompted me to look at it and clean it up. Basic operations have been
directly replaced by library calls (memcpy, memset is both available
in the kernel and userspace, and usually faster than a self-made
loop). The is_set and is_clear functions now use a processing time
shortcut, too.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/xt_sctp.h | 84 ++++++++++++++++-----------------------
 1 file changed, 35 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/xt_sctp.h b/include/linux/netfilter/xt_sctp.h
index dd5a4fd4cfd3..32000ba6ecef 100644
--- a/include/linux/netfilter/xt_sctp.h
+++ b/include/linux/netfilter/xt_sctp.h
@@ -37,68 +37,54 @@ struct xt_sctp_info {
 
 #define SCTP_CHUNKMAP_SET(chunkmap, type) 		\
 	do { 						\
-		chunkmap[type / bytes(u_int32_t)] |= 	\
+		(chunkmap)[type / bytes(u_int32_t)] |= 	\
 			1 << (type % bytes(u_int32_t));	\
 	} while (0)
 
 #define SCTP_CHUNKMAP_CLEAR(chunkmap, type)		 	\
 	do {							\
-		chunkmap[type / bytes(u_int32_t)] &= 		\
+		(chunkmap)[type / bytes(u_int32_t)] &= 		\
 			~(1 << (type % bytes(u_int32_t)));	\
 	} while (0)
 
 #define SCTP_CHUNKMAP_IS_SET(chunkmap, type) 			\
 ({								\
-	(chunkmap[type / bytes (u_int32_t)] & 			\
+	((chunkmap)[type / bytes (u_int32_t)] & 		\
 		(1 << (type % bytes (u_int32_t)))) ? 1: 0;	\
 })
 
-#define SCTP_CHUNKMAP_RESET(chunkmap) 				\
-	do {							\
-		int i; 						\
-		for (i = 0; i < ARRAY_SIZE(chunkmap); i++)	\
-			chunkmap[i] = 0;			\
-	} while (0)
-
-#define SCTP_CHUNKMAP_SET_ALL(chunkmap) 			\
-	do {							\
-		int i; 						\
-		for (i = 0; i < ARRAY_SIZE(chunkmap); i++) 	\
-			chunkmap[i] = ~0;			\
-	} while (0)
-
-#define SCTP_CHUNKMAP_COPY(destmap, srcmap) 			\
-	do {							\
-		int i; 						\
-		for (i = 0; i < ARRAY_SIZE(srcmap); i++) 	\
-			destmap[i] = srcmap[i];			\
-	} while (0)
-
-#define SCTP_CHUNKMAP_IS_CLEAR(chunkmap) 		\
-({							\
-	int i; 						\
-	int flag = 1;					\
-	for (i = 0; i < ARRAY_SIZE(chunkmap); i++) {	\
-		if (chunkmap[i]) {			\
-			flag = 0;			\
-			break;				\
-		}					\
-	}						\
-        flag;						\
-})
-
-#define SCTP_CHUNKMAP_IS_ALL_SET(chunkmap) 		\
-({							\
-	int i; 						\
-	int flag = 1;					\
-	for (i = 0; i < ARRAY_SIZE(chunkmap); i++) {	\
-		if (chunkmap[i] != ~0) {		\
-			flag = 0;			\
-				break;			\
-		}					\
-	}						\
-        flag;						\
-})
+#define SCTP_CHUNKMAP_RESET(chunkmap) \
+	memset((chunkmap), 0, sizeof(chunkmap))
+
+#define SCTP_CHUNKMAP_SET_ALL(chunkmap) \
+	memset((chunkmap), ~0U, sizeof(chunkmap))
+
+#define SCTP_CHUNKMAP_COPY(destmap, srcmap) \
+	memcpy((destmap), (srcmap), sizeof(srcmap))
+
+#define SCTP_CHUNKMAP_IS_CLEAR(chunkmap) \
+	__sctp_chunkmap_is_clear((chunkmap), ARRAY_SIZE(chunkmap))
+static inline bool
+__sctp_chunkmap_is_clear(const u_int32_t *chunkmap, unsigned int n)
+{
+	unsigned int i;
+	for (i = 0; i < n; ++i)
+		if (chunkmap[i])
+			return false;
+	return true;
+}
+
+#define SCTP_CHUNKMAP_IS_ALL_SET(chunkmap) \
+	__sctp_chunkmap_is_all_set((chunkmap), ARRAY_SIZE(chunkmap))
+static inline bool
+__sctp_chunkmap_is_all_set(const u_int32_t *chunkmap, unsigned int n)
+{
+	unsigned int i;
+	for (i = 0; i < n; ++i)
+		if (chunkmap[i] != ~0U)
+			return false;
+	return true;
+}
 
 #endif /* _XT_SCTP_H_ */
 
-- 
cgit v1.2.3


From 5452e425adfdfc4647b618e303f73d48f2405b0e Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:35 +0200
Subject: [NETFILTER]: annotate {arp,ip,ip6,x}tables with const

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/x_tables.h |  4 ++--
 net/ipv4/netfilter/arp_tables.c    | 31 ++++++++++++++++---------------
 net/ipv4/netfilter/arpt_mangle.c   |  2 +-
 net/ipv4/netfilter/ip_tables.c     | 31 ++++++++++++++++---------------
 net/ipv6/netfilter/ip6_tables.c    | 29 +++++++++++++++--------------
 net/netfilter/x_tables.c           | 18 +++++++++---------
 6 files changed, 59 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index b2c62cc618f5..2326296b6f25 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -430,13 +430,13 @@ extern int xt_compat_add_offset(int af, unsigned int offset, short delta);
 extern void xt_compat_flush_offsets(int af);
 extern short xt_compat_calc_jump(int af, unsigned int offset);
 
-extern int xt_compat_match_offset(struct xt_match *match);
+extern int xt_compat_match_offset(const struct xt_match *match);
 extern int xt_compat_match_from_user(struct xt_entry_match *m,
 				     void **dstptr, unsigned int *size);
 extern int xt_compat_match_to_user(struct xt_entry_match *m,
 				   void __user **dstptr, unsigned int *size);
 
-extern int xt_compat_target_offset(struct xt_target *target);
+extern int xt_compat_target_offset(const struct xt_target *target);
 extern void xt_compat_target_from_user(struct xt_entry_target *t,
 				       void **dstptr, unsigned int *size);
 extern int xt_compat_target_to_user(struct xt_entry_target *t,
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 1563f29b5117..10cc442330c3 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -59,7 +59,7 @@ do {								\
 #endif
 
 static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap,
-				      char *hdr_addr, int len)
+				      const char *hdr_addr, int len)
 {
 	int i, ret;
 
@@ -80,8 +80,8 @@ static inline int arp_packet_match(const struct arphdr *arphdr,
 				   const char *outdev,
 				   const struct arpt_arp *arpinfo)
 {
-	char *arpptr = (char *)(arphdr + 1);
-	char *src_devaddr, *tgt_devaddr;
+	const char *arpptr = (char *)(arphdr + 1);
+	const char *src_devaddr, *tgt_devaddr;
 	__be32 src_ipaddr, tgt_ipaddr;
 	int i, ret;
 
@@ -226,12 +226,12 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 {
 	static const char nulldevname[IFNAMSIZ];
 	unsigned int verdict = NF_DROP;
-	struct arphdr *arp;
+	const struct arphdr *arp;
 	bool hotdrop = false;
 	struct arpt_entry *e, *back;
 	const char *indev, *outdev;
 	void *table_base;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 
 	if (!pskb_may_pull(skb, arp_hdr_len(skb->dev)))
 		return NF_DROP;
@@ -352,7 +352,7 @@ static int mark_source_chains(struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			struct arpt_standard_target *t
+			const struct arpt_standard_target *t
 				= (void *)arpt_get_target(e);
 			int visited = e->comefrom & (1 << hook);
 
@@ -437,7 +437,7 @@ static int mark_source_chains(struct xt_table_info *newinfo,
 
 static inline int check_entry(struct arpt_entry *e, const char *name)
 {
-	struct arpt_entry_target *t;
+	const struct arpt_entry_target *t;
 
 	if (!arp_checkentry(&e->arp)) {
 		duprintf("arp_tables: arp check failed %p %s.\n", e, name);
@@ -710,7 +710,7 @@ static inline struct xt_counters *alloc_counters(struct arpt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	 * (other than comefrom, which userspace doesn't care
@@ -737,7 +737,7 @@ static int copy_entries_to_user(unsigned int total_size,
 	unsigned int off, num;
 	struct arpt_entry *e;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 	int ret = 0;
 	void *loc_cpu_entry;
 
@@ -872,7 +872,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 				    "arptable_%s", name);
 	if (t && !IS_ERR(t)) {
 		struct arpt_getinfo info;
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
 
 #ifdef CONFIG_COMPAT
 		if (compat) {
@@ -927,7 +927,8 @@ static int get_entries(struct net *net, struct arpt_get_entries __user *uptr,
 
 	t = xt_find_table_lock(net, NF_ARP, get.name);
 	if (t && !IS_ERR(t)) {
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
+
 		duprintf("t->private->number = %u\n",
 			 private->number);
 		if (get.size == private->size)
@@ -1087,11 +1088,11 @@ static int do_add_counters(struct net *net, void __user *user, unsigned int len,
 	struct xt_counters_info tmp;
 	struct xt_counters *paddc;
 	unsigned int num_counters;
-	char *name;
+	const char *name;
 	int size;
 	void *ptmp;
 	struct arpt_table *t;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 	int ret = 0;
 	void *loc_cpu_entry;
 #ifdef CONFIG_COMPAT
@@ -1558,7 +1559,7 @@ static int compat_copy_entries_to_user(unsigned int total_size,
 				       void __user *userptr)
 {
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 	void __user *pos;
 	unsigned int size;
 	int ret = 0;
@@ -1609,7 +1610,7 @@ static int compat_get_entries(struct net *net,
 	xt_compat_lock(NF_ARP);
 	t = xt_find_table_lock(net, NF_ARP, get.name);
 	if (t && !IS_ERR(t)) {
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
 		struct xt_table_info info;
 
 		duprintf("t->private->number = %u\n", private->number);
diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c
index 3f4222b0a803..3e732c827fc2 100644
--- a/net/ipv4/netfilter/arpt_mangle.c
+++ b/net/ipv4/netfilter/arpt_mangle.c
@@ -15,7 +15,7 @@ target(struct sk_buff *skb,
        const void *targinfo)
 {
 	const struct arpt_mangle *mangle = targinfo;
-	struct arphdr *arp;
+	const struct arphdr *arp;
 	unsigned char *arpptr;
 	int pln, hln;
 
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index a819d191e1aa..aa124b50cb4a 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -296,7 +296,7 @@ static void trace_packet(struct sk_buff *skb,
 			 struct ipt_entry *e)
 {
 	void *table_base;
-	struct ipt_entry *root;
+	const struct ipt_entry *root;
 	char *hookname, *chainname, *comment;
 	unsigned int rulenum = 0;
 
@@ -327,7 +327,7 @@ ipt_do_table(struct sk_buff *skb,
 {
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	u_int16_t offset;
-	struct iphdr *ip;
+	const struct iphdr *ip;
 	u_int16_t datalen;
 	bool hotdrop = false;
 	/* Initializing verdict to NF_DROP keeps gcc happy. */
@@ -926,7 +926,7 @@ static struct xt_counters * alloc_counters(struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	   (other than comefrom, which userspace doesn't care
@@ -953,9 +953,9 @@ copy_entries_to_user(unsigned int total_size,
 	unsigned int off, num;
 	struct ipt_entry *e;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 	int ret = 0;
-	void *loc_cpu_entry;
+	const void *loc_cpu_entry;
 
 	counters = alloc_counters(table);
 	if (IS_ERR(counters))
@@ -975,8 +975,8 @@ copy_entries_to_user(unsigned int total_size,
 	/* ... then go back and fix counters and names */
 	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
 		unsigned int i;
-		struct ipt_entry_match *m;
-		struct ipt_entry_target *t;
+		const struct ipt_entry_match *m;
+		const struct ipt_entry_target *t;
 
 		e = (struct ipt_entry *)(loc_cpu_entry + off);
 		if (copy_to_user(userptr + off
@@ -1116,7 +1116,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 				    "iptable_%s", name);
 	if (t && !IS_ERR(t)) {
 		struct ipt_getinfo info;
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
 
 #ifdef CONFIG_COMPAT
 		if (compat) {
@@ -1172,7 +1172,7 @@ get_entries(struct net *net, struct ipt_get_entries __user *uptr, int *len)
 
 	t = xt_find_table_lock(net, AF_INET, get.name);
 	if (t && !IS_ERR(t)) {
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
 		duprintf("t->private->number = %u\n", private->number);
 		if (get.size == private->size)
 			ret = copy_entries_to_user(private->size,
@@ -1337,11 +1337,11 @@ do_add_counters(struct net *net, void __user *user, unsigned int len, int compat
 	struct xt_counters_info tmp;
 	struct xt_counters *paddc;
 	unsigned int num_counters;
-	char *name;
+	const char *name;
 	int size;
 	void *ptmp;
 	struct xt_table *t;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 	int ret = 0;
 	void *loc_cpu_entry;
 #ifdef CONFIG_COMPAT
@@ -1878,11 +1878,11 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table,
 			    void __user *userptr)
 {
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 	void __user *pos;
 	unsigned int size;
 	int ret = 0;
-	void *loc_cpu_entry;
+	const void *loc_cpu_entry;
 	unsigned int i = 0;
 
 	counters = alloc_counters(table);
@@ -1929,7 +1929,7 @@ compat_get_entries(struct net *net, struct compat_ipt_get_entries __user *uptr,
 	xt_compat_lock(AF_INET);
 	t = xt_find_table_lock(net, AF_INET, get.name);
 	if (t && !IS_ERR(t)) {
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
 		struct xt_table_info info;
 		duprintf("t->private->number = %u\n", private->number);
 		ret = compat_table_info(private, &info);
@@ -2130,7 +2130,8 @@ icmp_match(const struct sk_buff *skb,
 	   unsigned int protoff,
 	   bool *hotdrop)
 {
-	struct icmphdr _icmph, *ic;
+	const struct icmphdr *ic;
+	struct icmphdr _icmph;
 	const struct ipt_icmp *icmpinfo = matchinfo;
 
 	/* Must not be a fragment. */
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 70ef0d276cc0..782183f63366 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -325,7 +325,7 @@ static void trace_packet(struct sk_buff *skb,
 			 struct ip6t_entry *e)
 {
 	void *table_base;
-	struct ip6t_entry *root;
+	const struct ip6t_entry *root;
 	char *hookname, *chainname, *comment;
 	unsigned int rulenum = 0;
 
@@ -952,7 +952,7 @@ static struct xt_counters *alloc_counters(struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 
 	/* We need atomic snapshot of counters: rest doesn't change
 	   (other than comefrom, which userspace doesn't care
@@ -979,9 +979,9 @@ copy_entries_to_user(unsigned int total_size,
 	unsigned int off, num;
 	struct ip6t_entry *e;
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 	int ret = 0;
-	void *loc_cpu_entry;
+	const void *loc_cpu_entry;
 
 	counters = alloc_counters(table);
 	if (IS_ERR(counters))
@@ -1001,8 +1001,8 @@ copy_entries_to_user(unsigned int total_size,
 	/* ... then go back and fix counters and names */
 	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
 		unsigned int i;
-		struct ip6t_entry_match *m;
-		struct ip6t_entry_target *t;
+		const struct ip6t_entry_match *m;
+		const struct ip6t_entry_target *t;
 
 		e = (struct ip6t_entry *)(loc_cpu_entry + off);
 		if (copy_to_user(userptr + off
@@ -1142,7 +1142,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat)
 				    "ip6table_%s", name);
 	if (t && !IS_ERR(t)) {
 		struct ip6t_getinfo info;
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
 
 #ifdef CONFIG_COMPAT
 		if (compat) {
@@ -1225,7 +1225,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks,
 	struct xt_table *t;
 	struct xt_table_info *oldinfo;
 	struct xt_counters *counters;
-	void *loc_cpu_old_entry;
+	const void *loc_cpu_old_entry;
 
 	ret = 0;
 	counters = vmalloc_node(num_counters * sizeof(struct xt_counters),
@@ -1369,9 +1369,9 @@ do_add_counters(struct net *net, void __user *user, unsigned int len,
 	int size;
 	void *ptmp;
 	struct xt_table *t;
-	struct xt_table_info *private;
+	const struct xt_table_info *private;
 	int ret = 0;
-	void *loc_cpu_entry;
+	const void *loc_cpu_entry;
 #ifdef CONFIG_COMPAT
 	struct compat_xt_counters_info compat_tmp;
 
@@ -1905,11 +1905,11 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table,
 			    void __user *userptr)
 {
 	struct xt_counters *counters;
-	struct xt_table_info *private = table->private;
+	const struct xt_table_info *private = table->private;
 	void __user *pos;
 	unsigned int size;
 	int ret = 0;
-	void *loc_cpu_entry;
+	const void *loc_cpu_entry;
 	unsigned int i = 0;
 
 	counters = alloc_counters(table);
@@ -1956,7 +1956,7 @@ compat_get_entries(struct net *net, struct compat_ip6t_get_entries __user *uptr,
 	xt_compat_lock(AF_INET6);
 	t = xt_find_table_lock(net, AF_INET6, get.name);
 	if (t && !IS_ERR(t)) {
-		struct xt_table_info *private = t->private;
+		const struct xt_table_info *private = t->private;
 		struct xt_table_info info;
 		duprintf("t->private->number = %u\n", private->number);
 		ret = compat_table_info(private, &info);
@@ -2155,7 +2155,8 @@ icmp6_match(const struct sk_buff *skb,
 	   unsigned int protoff,
 	   bool *hotdrop)
 {
-	struct icmp6hdr _icmph, *ic;
+	const struct icmp6hdr *ic;
+	struct icmp6hdr _icmph;
 	const struct ip6t_icmp *icmpinfo = matchinfo;
 
 	/* Must not be a fragment. */
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index 0bd95680a494..f52f7f810ac4 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -58,7 +58,7 @@ static struct xt_af *xt;
 #define duprintf(format, args...)
 #endif
 
-static const char *xt_prefix[NPROTO] = {
+static const char *const xt_prefix[NPROTO] = {
 	[AF_INET]	= "ip",
 	[AF_INET6]	= "ip6",
 	[NF_ARP]	= "arp",
@@ -248,7 +248,7 @@ EXPORT_SYMBOL_GPL(xt_request_find_target);
 
 static int match_revfn(int af, const char *name, u8 revision, int *bestp)
 {
-	struct xt_match *m;
+	const struct xt_match *m;
 	int have_rev = 0;
 
 	list_for_each_entry(m, &xt[af].match, list) {
@@ -264,7 +264,7 @@ static int match_revfn(int af, const char *name, u8 revision, int *bestp)
 
 static int target_revfn(int af, const char *name, u8 revision, int *bestp)
 {
-	struct xt_target *t;
+	const struct xt_target *t;
 	int have_rev = 0;
 
 	list_for_each_entry(t, &xt[af].target, list) {
@@ -385,7 +385,7 @@ short xt_compat_calc_jump(int af, unsigned int offset)
 }
 EXPORT_SYMBOL_GPL(xt_compat_calc_jump);
 
-int xt_compat_match_offset(struct xt_match *match)
+int xt_compat_match_offset(const struct xt_match *match)
 {
 	u_int16_t csize = match->compatsize ? : match->matchsize;
 	return XT_ALIGN(match->matchsize) - COMPAT_XT_ALIGN(csize);
@@ -395,7 +395,7 @@ EXPORT_SYMBOL_GPL(xt_compat_match_offset);
 int xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr,
 			      unsigned int *size)
 {
-	struct xt_match *match = m->u.kernel.match;
+	const struct xt_match *match = m->u.kernel.match;
 	struct compat_xt_entry_match *cm = (struct compat_xt_entry_match *)m;
 	int pad, off = xt_compat_match_offset(match);
 	u_int16_t msize = cm->u.user.match_size;
@@ -422,7 +422,7 @@ EXPORT_SYMBOL_GPL(xt_compat_match_from_user);
 int xt_compat_match_to_user(struct xt_entry_match *m, void __user **dstptr,
 			    unsigned int *size)
 {
-	struct xt_match *match = m->u.kernel.match;
+	const struct xt_match *match = m->u.kernel.match;
 	struct compat_xt_entry_match __user *cm = *dstptr;
 	int off = xt_compat_match_offset(match);
 	u_int16_t msize = m->u.user.match_size - off;
@@ -479,7 +479,7 @@ int xt_check_target(const struct xt_target *target, unsigned short family,
 EXPORT_SYMBOL_GPL(xt_check_target);
 
 #ifdef CONFIG_COMPAT
-int xt_compat_target_offset(struct xt_target *target)
+int xt_compat_target_offset(const struct xt_target *target)
 {
 	u_int16_t csize = target->compatsize ? : target->targetsize;
 	return XT_ALIGN(target->targetsize) - COMPAT_XT_ALIGN(csize);
@@ -489,7 +489,7 @@ EXPORT_SYMBOL_GPL(xt_compat_target_offset);
 void xt_compat_target_from_user(struct xt_entry_target *t, void **dstptr,
 				unsigned int *size)
 {
-	struct xt_target *target = t->u.kernel.target;
+	const struct xt_target *target = t->u.kernel.target;
 	struct compat_xt_entry_target *ct = (struct compat_xt_entry_target *)t;
 	int pad, off = xt_compat_target_offset(target);
 	u_int16_t tsize = ct->u.user.target_size;
@@ -515,7 +515,7 @@ EXPORT_SYMBOL_GPL(xt_compat_target_from_user);
 int xt_compat_target_to_user(struct xt_entry_target *t, void __user **dstptr,
 			     unsigned int *size)
 {
-	struct xt_target *target = t->u.kernel.target;
+	const struct xt_target *target = t->u.kernel.target;
 	struct compat_xt_entry_target __user *ct = *dstptr;
 	int off = xt_compat_target_offset(target);
 	u_int16_t tsize = t->u.user.target_size - off;
-- 
cgit v1.2.3


From 4abff0775d5e4feb20b21371e1c63a1b30fc2140 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:43 +0200
Subject: [NETFILTER]: remove arpt_table indirection macro

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter_arp/arp_tables.h | 11 +++++------
 net/ipv4/netfilter/arp_tables.c          | 27 +++++++++++++--------------
 net/ipv4/netfilter/arptable_filter.c     |  2 +-
 3 files changed, 19 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index db223ca92c8b..102c4134a713 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -24,7 +24,6 @@
 #define ARPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN
 #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
 #define arpt_target xt_target
-#define arpt_table xt_table
 
 #define ARPT_DEV_ADDR_LEN_MAX 16
 
@@ -271,15 +270,15 @@ struct arpt_error
  	xt_register_target(tgt); })
 #define arpt_unregister_target(tgt) xt_unregister_target(tgt)
 
-extern struct arpt_table *arpt_register_table(struct net *net,
-					      struct arpt_table *table,
-					      const struct arpt_replace *repl);
-extern void arpt_unregister_table(struct arpt_table *table);
+extern struct xt_table *arpt_register_table(struct net *net,
+					    struct xt_table *table,
+					    const struct arpt_replace *repl);
+extern void arpt_unregister_table(struct xt_table *table);
 extern unsigned int arpt_do_table(struct sk_buff *skb,
 				  unsigned int hook,
 				  const struct net_device *in,
 				  const struct net_device *out,
-				  struct arpt_table *table);
+				  struct xt_table *table);
 
 #define ARPT_ALIGN(s) XT_ALIGN(s)
 
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 10cc442330c3..34c42c831b18 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -222,7 +222,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 			   unsigned int hook,
 			   const struct net_device *in,
 			   const struct net_device *out,
-			   struct arpt_table *table)
+			   struct xt_table *table)
 {
 	static const char nulldevname[IFNAMSIZ];
 	unsigned int verdict = NF_DROP;
@@ -706,7 +706,7 @@ static void get_counters(const struct xt_table_info *t,
 	}
 }
 
-static inline struct xt_counters *alloc_counters(struct arpt_table *table)
+static inline struct xt_counters *alloc_counters(struct xt_table *table)
 {
 	unsigned int countersize;
 	struct xt_counters *counters;
@@ -731,13 +731,13 @@ static inline struct xt_counters *alloc_counters(struct arpt_table *table)
 }
 
 static int copy_entries_to_user(unsigned int total_size,
-				struct arpt_table *table,
+				struct xt_table *table,
 				void __user *userptr)
 {
 	unsigned int off, num;
 	struct arpt_entry *e;
 	struct xt_counters *counters;
-	const struct xt_table_info *private = table->private;
+	struct xt_table_info *private = table->private;
 	int ret = 0;
 	void *loc_cpu_entry;
 
@@ -851,7 +851,7 @@ static int compat_table_info(const struct xt_table_info *info,
 static int get_info(struct net *net, void __user *user, int *len, int compat)
 {
 	char name[ARPT_TABLE_MAXNAMELEN];
-	struct arpt_table *t;
+	struct xt_table *t;
 	int ret;
 
 	if (*len != sizeof(struct arpt_getinfo)) {
@@ -911,7 +911,7 @@ static int get_entries(struct net *net, struct arpt_get_entries __user *uptr,
 {
 	int ret;
 	struct arpt_get_entries get;
-	struct arpt_table *t;
+	struct xt_table *t;
 
 	if (*len < sizeof(get)) {
 		duprintf("get_entries: %u < %Zu\n", *len, sizeof(get));
@@ -954,7 +954,7 @@ static int __do_replace(struct net *net, const char *name,
 			void __user *counters_ptr)
 {
 	int ret;
-	struct arpt_table *t;
+	struct xt_table *t;
 	struct xt_table_info *oldinfo;
 	struct xt_counters *counters;
 	void *loc_cpu_old_entry;
@@ -1091,7 +1091,7 @@ static int do_add_counters(struct net *net, void __user *user, unsigned int len,
 	const char *name;
 	int size;
 	void *ptmp;
-	struct arpt_table *t;
+	struct xt_table *t;
 	const struct xt_table_info *private;
 	int ret = 0;
 	void *loc_cpu_entry;
@@ -1555,7 +1555,7 @@ out:
 }
 
 static int compat_copy_entries_to_user(unsigned int total_size,
-				       struct arpt_table *table,
+				       struct xt_table *table,
 				       void __user *userptr)
 {
 	struct xt_counters *counters;
@@ -1593,7 +1593,7 @@ static int compat_get_entries(struct net *net,
 {
 	int ret;
 	struct compat_arpt_get_entries get;
-	struct arpt_table *t;
+	struct xt_table *t;
 
 	if (*len < sizeof(get)) {
 		duprintf("compat_get_entries: %u < %zu\n", *len, sizeof(get));
@@ -1723,9 +1723,8 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len
 	return ret;
 }
 
-struct arpt_table *arpt_register_table(struct net *net,
-				       struct arpt_table *table,
-				       const struct arpt_replace *repl)
+struct xt_table *arpt_register_table(struct net *net, struct xt_table *table,
+				     const struct arpt_replace *repl)
 {
 	int ret;
 	struct xt_table_info *newinfo;
@@ -1767,7 +1766,7 @@ out:
 	return ERR_PTR(ret);
 }
 
-void arpt_unregister_table(struct arpt_table *table)
+void arpt_unregister_table(struct xt_table *table)
 {
 	struct xt_table_info *private;
 	void *loc_cpu_entry;
diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c
index 629e4951a9b1..9f6526c87757 100644
--- a/net/ipv4/netfilter/arptable_filter.c
+++ b/net/ipv4/netfilter/arptable_filter.c
@@ -45,7 +45,7 @@ static struct
 	.term = ARPT_ERROR_INIT,
 };
 
-static struct arpt_table packet_filter = {
+static struct xt_table packet_filter = {
 	.name		= "filter",
 	.valid_hooks	= FILTER_VALID_HOOKS,
 	.lock		= __RW_LOCK_UNLOCKED(packet_filter.lock),
-- 
cgit v1.2.3


From 95eea855af69bfd54a7b73546190e76046ca2e07 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:43 +0200
Subject: [NETFILTER]: remove arpt_target indirection macro

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter_arp/arp_tables.h | 1 -
 net/ipv4/netfilter/arp_tables.c          | 8 ++++----
 net/ipv4/netfilter/arpt_mangle.c         | 2 +-
 3 files changed, 5 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 102c4134a713..782b83e5bdb9 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -23,7 +23,6 @@
 
 #define ARPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN
 #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
-#define arpt_target xt_target
 
 #define ARPT_DEV_ADDR_LEN_MAX 16
 
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 34c42c831b18..d55f3b42eba5 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -457,7 +457,7 @@ static inline int check_entry(struct arpt_entry *e, const char *name)
 static inline int check_target(struct arpt_entry *e, const char *name)
 {
 	struct arpt_entry_target *t;
-	struct arpt_target *target;
+	struct xt_target *target;
 	int ret;
 
 	t = arpt_get_target(e);
@@ -480,7 +480,7 @@ find_check_entry(struct arpt_entry *e, const char *name, unsigned int size,
 		 unsigned int *i)
 {
 	struct arpt_entry_target *t;
-	struct arpt_target *target;
+	struct xt_target *target;
 	int ret;
 
 	ret = check_entry(e, name);
@@ -1784,7 +1784,7 @@ void arpt_unregister_table(struct xt_table *table)
 }
 
 /* The built-in targets: standard (NULL) and error. */
-static struct arpt_target arpt_standard_target __read_mostly = {
+static struct xt_target arpt_standard_target __read_mostly = {
 	.name		= ARPT_STANDARD_TARGET,
 	.targetsize	= sizeof(int),
 	.family		= NF_ARP,
@@ -1795,7 +1795,7 @@ static struct arpt_target arpt_standard_target __read_mostly = {
 #endif
 };
 
-static struct arpt_target arpt_error_target __read_mostly = {
+static struct xt_target arpt_error_target __read_mostly = {
 	.name		= ARPT_ERROR_TARGET,
 	.target		= arpt_error,
 	.targetsize	= ARPT_FUNCTION_MAXNAMELEN,
diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c
index 3e732c827fc2..f9c102ab891b 100644
--- a/net/ipv4/netfilter/arpt_mangle.c
+++ b/net/ipv4/netfilter/arpt_mangle.c
@@ -73,7 +73,7 @@ checkentry(const char *tablename, const void *e, const struct xt_target *target,
 	return true;
 }
 
-static struct arpt_target arpt_mangle_reg __read_mostly = {
+static struct xt_target arpt_mangle_reg __read_mostly = {
 	.name		= "mangle",
 	.target		= target,
 	.targetsize	= sizeof(struct arpt_mangle),
-- 
cgit v1.2.3


From 3bb0362d2f53fa54a17b88c96b43fc093e47699b Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:44 +0200
Subject: [NETFILTER]: remove arpt_(un)register_target indirection macros

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter_arp/arp_tables.h | 5 -----
 net/ipv4/netfilter/arpt_mangle.c         | 8 +++-----
 2 files changed, 3 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 782b83e5bdb9..dd9c97f2d436 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -264,11 +264,6 @@ struct arpt_error
 	.target.errorname = "ERROR",					       \
 }
 
-#define arpt_register_target(tgt) 	\
-({	(tgt)->family = NF_ARP;		\
- 	xt_register_target(tgt); })
-#define arpt_unregister_target(tgt) xt_unregister_target(tgt)
-
 extern struct xt_table *arpt_register_table(struct net *net,
 					    struct xt_table *table,
 					    const struct arpt_replace *repl);
diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c
index f9c102ab891b..a385959d2655 100644
--- a/net/ipv4/netfilter/arpt_mangle.c
+++ b/net/ipv4/netfilter/arpt_mangle.c
@@ -75,6 +75,7 @@ checkentry(const char *tablename, const void *e, const struct xt_target *target,
 
 static struct xt_target arpt_mangle_reg __read_mostly = {
 	.name		= "mangle",
+	.family		= NF_ARP,
 	.target		= target,
 	.targetsize	= sizeof(struct arpt_mangle),
 	.checkentry	= checkentry,
@@ -83,15 +84,12 @@ static struct xt_target arpt_mangle_reg __read_mostly = {
 
 static int __init arpt_mangle_init(void)
 {
-	if (arpt_register_target(&arpt_mangle_reg))
-		return -EINVAL;
-
-	return 0;
+	return xt_register_target(&arpt_mangle_reg);
 }
 
 static void __exit arpt_mangle_fini(void)
 {
-	arpt_unregister_target(&arpt_mangle_reg);
+	xt_unregister_target(&arpt_mangle_reg);
 }
 
 module_init(arpt_mangle_init);
-- 
cgit v1.2.3


From 937e0dfd87a8b7946a17161664500fba93eb13fd Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 20 Mar 2008 15:15:47 +0100
Subject: [NETFILTER]: nf_nat: add helpers for common NAT protocol operations

Add generic ->in_range and ->unique_tuple ops to avoid duplicating them
again and again for future NAT modules and save a few bytes of text:

net/ipv4/netfilter/nf_nat_proto_tcp.c:
  tcp_in_range     |  -62 (removed)
  tcp_unique_tuple | -259 # 271 -> 12, # inlines: 1 -> 0, size inlines: 7 -> 0
 2 functions changed, 321 bytes removed

net/ipv4/netfilter/nf_nat_proto_udp.c:
  udp_in_range     |  -62 (removed)
  udp_unique_tuple | -259 # 271 -> 12, # inlines: 1 -> 0, size inlines: 7 -> 0
 2 functions changed, 321 bytes removed

net/ipv4/netfilter/nf_nat_proto_gre.c:
  gre_in_range |  -62 (removed)
 1 function changed, 62 bytes removed

vmlinux:
 5 functions changed, 704 bytes removed

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_nat_protocol.h  | 11 +++++
 net/ipv4/netfilter/Makefile              |  2 +-
 net/ipv4/netfilter/nf_nat_proto_common.c | 85 ++++++++++++++++++++++++++++++++
 net/ipv4/netfilter/nf_nat_proto_gre.c    | 20 +-------
 net/ipv4/netfilter/nf_nat_proto_tcp.c    | 65 ++----------------------
 net/ipv4/netfilter/nf_nat_proto_udp.c    | 64 ++----------------------
 6 files changed, 106 insertions(+), 141 deletions(-)
 create mode 100644 net/ipv4/netfilter/nf_nat_proto_common.c

(limited to 'include')

diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h
index 4aa0edbb5b96..fa06f6d0de54 100644
--- a/include/net/netfilter/nf_nat_protocol.h
+++ b/include/net/netfilter/nf_nat_protocol.h
@@ -62,6 +62,17 @@ extern int init_protocols(void) __init;
 extern void cleanup_protocols(void);
 extern const struct nf_nat_protocol *find_nat_proto(u_int16_t protonum);
 
+extern int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple,
+				 enum nf_nat_manip_type maniptype,
+				 const union nf_conntrack_man_proto *min,
+				 const union nf_conntrack_man_proto *max);
+
+extern int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
+				     const struct nf_nat_range *range,
+				     enum nf_nat_manip_type maniptype,
+				     const struct nf_conn *ct,
+				     u_int16_t *rover);
+
 extern int nf_nat_port_range_to_nlattr(struct sk_buff *skb,
 				       const struct nf_nat_range *range);
 extern int nf_nat_port_nlattr_to_range(struct nlattr *tb[],
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index 0c7dc78a62e9..e73d0eb9994a 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -10,7 +10,7 @@ nf_conntrack_ipv4-objs	+= nf_conntrack_l3proto_ipv4_compat.o
 endif
 endif
 
-nf_nat-objs		:= nf_nat_core.o nf_nat_helper.o nf_nat_proto_unknown.o nf_nat_proto_tcp.o nf_nat_proto_udp.o nf_nat_proto_icmp.o
+nf_nat-objs		:= nf_nat_core.o nf_nat_helper.o nf_nat_proto_unknown.o nf_nat_proto_common.o nf_nat_proto_tcp.o nf_nat_proto_udp.o nf_nat_proto_icmp.o
 iptable_nat-objs	:= nf_nat_rule.o nf_nat_standalone.o
 
 # connection tracking
diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c
new file mode 100644
index 000000000000..a124213fb9da
--- /dev/null
+++ b/net/ipv4/netfilter/nf_nat_proto_common.c
@@ -0,0 +1,85 @@
+/* (C) 1999-2001 Paul `Rusty' Russell
+ * (C) 2002-2006 Netfilter Core Team <coreteam@netfilter.org>
+ * (C) 2008 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/types.h>
+#include <linux/random.h>
+#include <linux/ip.h>
+
+#include <linux/netfilter.h>
+#include <net/netfilter/nf_nat.h>
+#include <net/netfilter/nf_nat_core.h>
+#include <net/netfilter/nf_nat_rule.h>
+#include <net/netfilter/nf_nat_protocol.h>
+
+int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple,
+			  enum nf_nat_manip_type maniptype,
+			  const union nf_conntrack_man_proto *min,
+			  const union nf_conntrack_man_proto *max)
+{
+	__be16 port;
+
+	if (maniptype == IP_NAT_MANIP_SRC)
+		port = tuple->src.u.all;
+	else
+		port = tuple->dst.u.all;
+
+	return ntohs(port) >= ntohs(min->all) &&
+	       ntohs(port) <= ntohs(max->all);
+}
+EXPORT_SYMBOL_GPL(nf_nat_proto_in_range);
+
+int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
+			      const struct nf_nat_range *range,
+			      enum nf_nat_manip_type maniptype,
+			      const struct nf_conn *ct,
+			      u_int16_t *rover)
+{
+	unsigned int range_size, min, i;
+	__be16 *portptr;
+
+	if (maniptype == IP_NAT_MANIP_SRC)
+		portptr = &tuple->src.u.all;
+	else
+		portptr = &tuple->dst.u.all;
+
+	/* If no range specified... */
+	if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) {
+		/* If it's dst rewrite, can't change port */
+		if (maniptype == IP_NAT_MANIP_DST)
+			return 0;
+
+		if (ntohs(*portptr) < 1024) {
+			/* Loose convention: >> 512 is credential passing */
+			if (ntohs(*portptr) < 512) {
+				min = 1;
+				range_size = 511 - min + 1;
+			} else {
+				min = 600;
+				range_size = 1023 - min + 1;
+			}
+		} else {
+			min = 1024;
+			range_size = 65535 - 1024 + 1;
+		}
+	} else {
+		min = ntohs(range->min.all);
+		range_size = ntohs(range->max.all) - min + 1;
+	}
+
+	if (range->flags & IP_NAT_RANGE_PROTO_RANDOM)
+		*rover = net_random();
+
+	for (i = 0; i < range_size; i++, (*rover)++) {
+		*portptr = htons(min + *rover % range_size);
+		if (!nf_nat_used_tuple(tuple, ct))
+			return 1;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(nf_nat_proto_unique_tuple);
diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c
index a1e4da16da2e..87af63d9e692 100644
--- a/net/ipv4/netfilter/nf_nat_proto_gre.c
+++ b/net/ipv4/netfilter/nf_nat_proto_gre.c
@@ -36,24 +36,6 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Harald Welte <laforge@gnumonks.org>");
 MODULE_DESCRIPTION("Netfilter NAT protocol helper module for GRE");
 
-/* is key in given range between min and max */
-static int
-gre_in_range(const struct nf_conntrack_tuple *tuple,
-	     enum nf_nat_manip_type maniptype,
-	     const union nf_conntrack_man_proto *min,
-	     const union nf_conntrack_man_proto *max)
-{
-	__be16 key;
-
-	if (maniptype == IP_NAT_MANIP_SRC)
-		key = tuple->src.u.gre.key;
-	else
-		key = tuple->dst.u.gre.key;
-
-	return ntohs(key) >= ntohs(min->gre.key) &&
-	       ntohs(key) <= ntohs(max->gre.key);
-}
-
 /* generate unique tuple ... */
 static int
 gre_unique_tuple(struct nf_conntrack_tuple *tuple,
@@ -140,7 +122,7 @@ static const struct nf_nat_protocol gre = {
 	.protonum		= IPPROTO_GRE,
 	.me			= THIS_MODULE,
 	.manip_pkt		= gre_manip_pkt,
-	.in_range		= gre_in_range,
+	.in_range		= nf_nat_proto_in_range,
 	.unique_tuple		= gre_unique_tuple,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 	.range_to_nlattr	= nf_nat_port_range_to_nlattr,
diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c
index ffd5d1589eca..f8c498fc24fd 100644
--- a/net/ipv4/netfilter/nf_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c
@@ -8,7 +8,6 @@
 
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/random.h>
 #include <linux/ip.h>
 #include <linux/tcp.h>
 
@@ -19,22 +18,7 @@
 #include <net/netfilter/nf_nat_protocol.h>
 #include <net/netfilter/nf_nat_core.h>
 
-static int
-tcp_in_range(const struct nf_conntrack_tuple *tuple,
-	     enum nf_nat_manip_type maniptype,
-	     const union nf_conntrack_man_proto *min,
-	     const union nf_conntrack_man_proto *max)
-{
-	__be16 port;
-
-	if (maniptype == IP_NAT_MANIP_SRC)
-		port = tuple->src.u.tcp.port;
-	else
-		port = tuple->dst.u.tcp.port;
-
-	return ntohs(port) >= ntohs(min->tcp.port) &&
-	       ntohs(port) <= ntohs(max->tcp.port);
-}
+static u_int16_t tcp_port_rover;
 
 static int
 tcp_unique_tuple(struct nf_conntrack_tuple *tuple,
@@ -42,49 +26,8 @@ tcp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		 enum nf_nat_manip_type maniptype,
 		 const struct nf_conn *ct)
 {
-	static u_int16_t port;
-	__be16 *portptr;
-	unsigned int range_size, min, i;
-
-	if (maniptype == IP_NAT_MANIP_SRC)
-		portptr = &tuple->src.u.tcp.port;
-	else
-		portptr = &tuple->dst.u.tcp.port;
-
-	/* If no range specified... */
-	if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) {
-		/* If it's dst rewrite, can't change port */
-		if (maniptype == IP_NAT_MANIP_DST)
-			return 0;
-
-		/* Map privileged onto privileged. */
-		if (ntohs(*portptr) < 1024) {
-			/* Loose convention: >> 512 is credential passing */
-			if (ntohs(*portptr)<512) {
-				min = 1;
-				range_size = 511 - min + 1;
-			} else {
-				min = 600;
-				range_size = 1023 - min + 1;
-			}
-		} else {
-			min = 1024;
-			range_size = 65535 - 1024 + 1;
-		}
-	} else {
-		min = ntohs(range->min.tcp.port);
-		range_size = ntohs(range->max.tcp.port) - min + 1;
-	}
-
-	if (range->flags & IP_NAT_RANGE_PROTO_RANDOM)
-		port =  net_random();
-
-	for (i = 0; i < range_size; i++, port++) {
-		*portptr = htons(min + port % range_size);
-		if (!nf_nat_used_tuple(tuple, ct))
-			return 1;
-	}
-	return 0;
+	return nf_nat_proto_unique_tuple(tuple, range, maniptype, ct,
+					 &tcp_port_rover);
 }
 
 static int
@@ -142,7 +85,7 @@ const struct nf_nat_protocol nf_nat_protocol_tcp = {
 	.protonum		= IPPROTO_TCP,
 	.me			= THIS_MODULE,
 	.manip_pkt		= tcp_manip_pkt,
-	.in_range		= tcp_in_range,
+	.in_range		= nf_nat_proto_in_range,
 	.unique_tuple		= tcp_unique_tuple,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 	.range_to_nlattr	= nf_nat_port_range_to_nlattr,
diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c
index 4b8f49910ff2..a182f5ac3177 100644
--- a/net/ipv4/netfilter/nf_nat_proto_udp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_udp.c
@@ -8,7 +8,6 @@
 
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/random.h>
 #include <linux/ip.h>
 #include <linux/udp.h>
 
@@ -18,22 +17,7 @@
 #include <net/netfilter/nf_nat_rule.h>
 #include <net/netfilter/nf_nat_protocol.h>
 
-static int
-udp_in_range(const struct nf_conntrack_tuple *tuple,
-	     enum nf_nat_manip_type maniptype,
-	     const union nf_conntrack_man_proto *min,
-	     const union nf_conntrack_man_proto *max)
-{
-	__be16 port;
-
-	if (maniptype == IP_NAT_MANIP_SRC)
-		port = tuple->src.u.udp.port;
-	else
-		port = tuple->dst.u.udp.port;
-
-	return ntohs(port) >= ntohs(min->udp.port) &&
-	       ntohs(port) <= ntohs(max->udp.port);
-}
+static u_int16_t udp_port_rover;
 
 static int
 udp_unique_tuple(struct nf_conntrack_tuple *tuple,
@@ -41,48 +25,8 @@ udp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		 enum nf_nat_manip_type maniptype,
 		 const struct nf_conn *ct)
 {
-	static u_int16_t port;
-	__be16 *portptr;
-	unsigned int range_size, min, i;
-
-	if (maniptype == IP_NAT_MANIP_SRC)
-		portptr = &tuple->src.u.udp.port;
-	else
-		portptr = &tuple->dst.u.udp.port;
-
-	/* If no range specified... */
-	if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) {
-		/* If it's dst rewrite, can't change port */
-		if (maniptype == IP_NAT_MANIP_DST)
-			return 0;
-
-		if (ntohs(*portptr) < 1024) {
-			/* Loose convention: >> 512 is credential passing */
-			if (ntohs(*portptr)<512) {
-				min = 1;
-				range_size = 511 - min + 1;
-			} else {
-				min = 600;
-				range_size = 1023 - min + 1;
-			}
-		} else {
-			min = 1024;
-			range_size = 65535 - 1024 + 1;
-		}
-	} else {
-		min = ntohs(range->min.udp.port);
-		range_size = ntohs(range->max.udp.port) - min + 1;
-	}
-
-	if (range->flags & IP_NAT_RANGE_PROTO_RANDOM)
-		port = net_random();
-
-	for (i = 0; i < range_size; i++, port++) {
-		*portptr = htons(min + port % range_size);
-		if (!nf_nat_used_tuple(tuple, ct))
-			return 1;
-	}
-	return 0;
+	return nf_nat_proto_unique_tuple(tuple, range, maniptype, ct,
+					 &udp_port_rover);
 }
 
 static int
@@ -132,7 +76,7 @@ const struct nf_nat_protocol nf_nat_protocol_udp = {
 	.protonum		= IPPROTO_UDP,
 	.me			= THIS_MODULE,
 	.manip_pkt		= udp_manip_pkt,
-	.in_range		= udp_in_range,
+	.in_range		= nf_nat_proto_in_range,
 	.unique_tuple		= udp_unique_tuple,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
 	.range_to_nlattr	= nf_nat_port_range_to_nlattr,
-- 
cgit v1.2.3


From 535b57c7c1524125444aa1b874332f6ff1608ef5 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Apr 2008 11:15:47 +0200
Subject: [NETFILTER]: nf_nat: move NAT ctnetlink helpers to
 nf_nat_proto_common

Move to nf_nat_proto_common and rename to nf_nat_proto_... since they're
also used by protocols that don't have port numbers.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_nat_protocol.h  |  8 +++----
 net/ipv4/netfilter/nf_nat_core.c         | 40 --------------------------------
 net/ipv4/netfilter/nf_nat_proto_common.c | 38 ++++++++++++++++++++++++++++++
 net/ipv4/netfilter/nf_nat_proto_gre.c    |  4 ++--
 net/ipv4/netfilter/nf_nat_proto_icmp.c   |  4 ++--
 net/ipv4/netfilter/nf_nat_proto_tcp.c    |  4 ++--
 net/ipv4/netfilter/nf_nat_proto_udp.c    |  4 ++--
 7 files changed, 50 insertions(+), 52 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h
index fa06f6d0de54..8ce227624eac 100644
--- a/include/net/netfilter/nf_nat_protocol.h
+++ b/include/net/netfilter/nf_nat_protocol.h
@@ -73,9 +73,9 @@ extern int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
 				     const struct nf_conn *ct,
 				     u_int16_t *rover);
 
-extern int nf_nat_port_range_to_nlattr(struct sk_buff *skb,
-				       const struct nf_nat_range *range);
-extern int nf_nat_port_nlattr_to_range(struct nlattr *tb[],
-				       struct nf_nat_range *range);
+extern int nf_nat_proto_range_to_nlattr(struct sk_buff *skb,
+					const struct nf_nat_range *range);
+extern int nf_nat_proto_nlattr_to_range(struct nlattr *tb[],
+					struct nf_nat_range *range);
 
 #endif /*_NF_NAT_PROTO_H*/
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index 9c8aa8df2d5e..9320c7ac5729 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -544,46 +544,6 @@ void nf_nat_protocol_unregister(const struct nf_nat_protocol *proto)
 }
 EXPORT_SYMBOL(nf_nat_protocol_unregister);
 
-#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
-int
-nf_nat_port_range_to_nlattr(struct sk_buff *skb,
-			    const struct nf_nat_range *range)
-{
-	NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MIN, range->min.tcp.port);
-	NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MAX, range->max.tcp.port);
-
-	return 0;
-
-nla_put_failure:
-	return -1;
-}
-EXPORT_SYMBOL_GPL(nf_nat_port_nlattr_to_range);
-
-int
-nf_nat_port_nlattr_to_range(struct nlattr *tb[], struct nf_nat_range *range)
-{
-	int ret = 0;
-
-	/* we have to return whether we actually parsed something or not */
-
-	if (tb[CTA_PROTONAT_PORT_MIN]) {
-		ret = 1;
-		range->min.tcp.port = nla_get_be16(tb[CTA_PROTONAT_PORT_MIN]);
-	}
-
-	if (!tb[CTA_PROTONAT_PORT_MAX]) {
-		if (ret)
-			range->max.tcp.port = range->min.tcp.port;
-	} else {
-		ret = 1;
-		range->max.tcp.port = nla_get_be16(tb[CTA_PROTONAT_PORT_MAX]);
-	}
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(nf_nat_port_range_to_nlattr);
-#endif
-
 /* Noone using conntrack by the time this called. */
 static void nf_nat_cleanup_conntrack(struct nf_conn *ct)
 {
diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c
index 871ab0eb325d..ef4dc3988925 100644
--- a/net/ipv4/netfilter/nf_nat_proto_common.c
+++ b/net/ipv4/netfilter/nf_nat_proto_common.c
@@ -88,3 +88,41 @@ int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
 	return 0;
 }
 EXPORT_SYMBOL_GPL(nf_nat_proto_unique_tuple);
+
+#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+int nf_nat_proto_range_to_nlattr(struct sk_buff *skb,
+				 const struct nf_nat_range *range)
+{
+	NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MIN, range->min.all);
+	NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MAX, range->max.all);
+	return 0;
+
+nla_put_failure:
+	return -1;
+}
+EXPORT_SYMBOL_GPL(nf_nat_proto_nlattr_to_range);
+
+int nf_nat_proto_nlattr_to_range(struct nlattr *tb[],
+				 struct nf_nat_range *range)
+{
+	int ret = 0;
+
+	/* we have to return whether we actually parsed something or not */
+
+	if (tb[CTA_PROTONAT_PORT_MIN]) {
+		ret = 1;
+		range->min.all = nla_get_be16(tb[CTA_PROTONAT_PORT_MIN]);
+	}
+
+	if (!tb[CTA_PROTONAT_PORT_MAX]) {
+		if (ret)
+			range->max.all = range->min.all;
+	} else {
+		ret = 1;
+		range->max.all = nla_get_be16(tb[CTA_PROTONAT_PORT_MAX]);
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(nf_nat_proto_range_to_nlattr);
+#endif
diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c
index 87af63d9e692..71b0935ee1c8 100644
--- a/net/ipv4/netfilter/nf_nat_proto_gre.c
+++ b/net/ipv4/netfilter/nf_nat_proto_gre.c
@@ -125,8 +125,8 @@ static const struct nf_nat_protocol gre = {
 	.in_range		= nf_nat_proto_in_range,
 	.unique_tuple		= gre_unique_tuple,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
-	.range_to_nlattr	= nf_nat_port_range_to_nlattr,
-	.nlattr_to_range	= nf_nat_port_nlattr_to_range,
+	.range_to_nlattr	= nf_nat_proto_range_to_nlattr,
+	.nlattr_to_range	= nf_nat_proto_nlattr_to_range,
 #endif
 };
 
diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c
index 03a02969aa57..ca601f84c4dc 100644
--- a/net/ipv4/netfilter/nf_nat_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c
@@ -79,7 +79,7 @@ const struct nf_nat_protocol nf_nat_protocol_icmp = {
 	.in_range		= icmp_in_range,
 	.unique_tuple		= icmp_unique_tuple,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
-	.range_to_nlattr	= nf_nat_port_range_to_nlattr,
-	.nlattr_to_range	= nf_nat_port_nlattr_to_range,
+	.range_to_nlattr	= nf_nat_proto_range_to_nlattr,
+	.nlattr_to_range	= nf_nat_proto_nlattr_to_range,
 #endif
 };
diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c
index f8c498fc24fd..1d73a11f55d9 100644
--- a/net/ipv4/netfilter/nf_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c
@@ -88,7 +88,7 @@ const struct nf_nat_protocol nf_nat_protocol_tcp = {
 	.in_range		= nf_nat_proto_in_range,
 	.unique_tuple		= tcp_unique_tuple,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
-	.range_to_nlattr	= nf_nat_port_range_to_nlattr,
-	.nlattr_to_range	= nf_nat_port_nlattr_to_range,
+	.range_to_nlattr	= nf_nat_proto_range_to_nlattr,
+	.nlattr_to_range	= nf_nat_proto_nlattr_to_range,
 #endif
 };
diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c
index a182f5ac3177..f36ce552a161 100644
--- a/net/ipv4/netfilter/nf_nat_proto_udp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_udp.c
@@ -79,7 +79,7 @@ const struct nf_nat_protocol nf_nat_protocol_udp = {
 	.in_range		= nf_nat_proto_in_range,
 	.unique_tuple		= udp_unique_tuple,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
-	.range_to_nlattr	= nf_nat_port_range_to_nlattr,
-	.nlattr_to_range	= nf_nat_port_nlattr_to_range,
+	.range_to_nlattr	= nf_nat_proto_range_to_nlattr,
+	.nlattr_to_range	= nf_nat_proto_nlattr_to_range,
 #endif
 };
-- 
cgit v1.2.3


From 2d2d84c40e19a7fce51ba1f124ecde105104192d Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 20 Mar 2008 15:15:49 +0100
Subject: [NETFILTER]: nf_nat: remove unused name from struct nf_nat_protocol

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_nat_protocol.h   | 3 ---
 net/ipv4/netfilter/nf_nat_proto_gre.c     | 1 -
 net/ipv4/netfilter/nf_nat_proto_icmp.c    | 1 -
 net/ipv4/netfilter/nf_nat_proto_tcp.c     | 1 -
 net/ipv4/netfilter/nf_nat_proto_udp.c     | 1 -
 net/ipv4/netfilter/nf_nat_proto_unknown.c | 1 -
 6 files changed, 8 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h
index 8ce227624eac..fba94a2028d5 100644
--- a/include/net/netfilter/nf_nat_protocol.h
+++ b/include/net/netfilter/nf_nat_protocol.h
@@ -8,9 +8,6 @@ struct nf_nat_range;
 
 struct nf_nat_protocol
 {
-	/* Protocol name */
-	const char *name;
-
 	/* Protocol number. */
 	unsigned int protonum;
 
diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c
index 71b0935ee1c8..84bb7854137a 100644
--- a/net/ipv4/netfilter/nf_nat_proto_gre.c
+++ b/net/ipv4/netfilter/nf_nat_proto_gre.c
@@ -118,7 +118,6 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff,
 }
 
 static const struct nf_nat_protocol gre = {
-	.name			= "GRE",
 	.protonum		= IPPROTO_GRE,
 	.me			= THIS_MODULE,
 	.manip_pkt		= gre_manip_pkt,
diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c
index ca601f84c4dc..ab3a0ec2a2d1 100644
--- a/net/ipv4/netfilter/nf_nat_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c
@@ -72,7 +72,6 @@ icmp_manip_pkt(struct sk_buff *skb,
 }
 
 const struct nf_nat_protocol nf_nat_protocol_icmp = {
-	.name			= "ICMP",
 	.protonum		= IPPROTO_ICMP,
 	.me			= THIS_MODULE,
 	.manip_pkt		= icmp_manip_pkt,
diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c
index 1d73a11f55d9..5d4c8a0e89c0 100644
--- a/net/ipv4/netfilter/nf_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c
@@ -81,7 +81,6 @@ tcp_manip_pkt(struct sk_buff *skb,
 }
 
 const struct nf_nat_protocol nf_nat_protocol_tcp = {
-	.name			= "TCP",
 	.protonum		= IPPROTO_TCP,
 	.me			= THIS_MODULE,
 	.manip_pkt		= tcp_manip_pkt,
diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c
index f36ce552a161..74a7e7b63465 100644
--- a/net/ipv4/netfilter/nf_nat_proto_udp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_udp.c
@@ -72,7 +72,6 @@ udp_manip_pkt(struct sk_buff *skb,
 }
 
 const struct nf_nat_protocol nf_nat_protocol_udp = {
-	.name			= "UDP",
 	.protonum		= IPPROTO_UDP,
 	.me			= THIS_MODULE,
 	.manip_pkt		= udp_manip_pkt,
diff --git a/net/ipv4/netfilter/nf_nat_proto_unknown.c b/net/ipv4/netfilter/nf_nat_proto_unknown.c
index a26efeb073cb..cda21ff0e4cf 100644
--- a/net/ipv4/netfilter/nf_nat_proto_unknown.c
+++ b/net/ipv4/netfilter/nf_nat_proto_unknown.c
@@ -46,7 +46,6 @@ unknown_manip_pkt(struct sk_buff *skb,
 }
 
 const struct nf_nat_protocol nf_nat_unknown_protocol = {
-	.name			= "unknown",
 	/* .me isn't set: getting a ref to this cannot fail. */
 	.manip_pkt		= unknown_manip_pkt,
 	.in_range		= unknown_in_range,
-- 
cgit v1.2.3


From d63a650736f566a1f9e9434725d2089597c0d2cc Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 20 Mar 2008 15:15:53 +0100
Subject: [NETFILTER]: Add partial checksum validation helper

Move the UDP-Lite conntrack checksum validation to a generic helper
similar to nf_checksum() and make it fall back to nf_checksum()
in case the full packet is to be checksummed and hardware checksums
are available. This is to be used by DCCP conntrack, which also
needs to verify partial checksums.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter.h                  | 22 ++++++++++++++++
 net/ipv4/netfilter.c                       | 37 +++++++++++++++++++++-----
 net/ipv6/netfilter.c                       | 42 +++++++++++++++++++++++++-----
 net/netfilter/nf_conntrack_proto_udplite.c | 33 +++++------------------
 4 files changed, 94 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 66bc52060fd6..e4c66593b5c6 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -234,6 +234,11 @@ struct nf_afinfo {
 	unsigned short	family;
 	__sum16		(*checksum)(struct sk_buff *skb, unsigned int hook,
 				    unsigned int dataoff, u_int8_t protocol);
+	__sum16		(*checksum_partial)(struct sk_buff *skb,
+					    unsigned int hook,
+					    unsigned int dataoff,
+					    unsigned int len,
+					    u_int8_t protocol);
 	int		(*route)(struct dst_entry **dst, struct flowi *fl);
 	void		(*saveroute)(const struct sk_buff *skb,
 				     struct nf_queue_entry *entry);
@@ -263,6 +268,23 @@ nf_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff,
 	return csum;
 }
 
+static inline __sum16
+nf_checksum_partial(struct sk_buff *skb, unsigned int hook,
+		    unsigned int dataoff, unsigned int len,
+		    u_int8_t protocol, unsigned short family)
+{
+	const struct nf_afinfo *afinfo;
+	__sum16 csum = 0;
+
+	rcu_read_lock();
+	afinfo = nf_get_afinfo(family);
+	if (afinfo)
+		csum = afinfo->checksum_partial(skb, hook, dataoff, len,
+						protocol);
+	rcu_read_unlock();
+	return csum;
+}
+
 extern int nf_register_afinfo(const struct nf_afinfo *afinfo);
 extern void nf_unregister_afinfo(const struct nf_afinfo *afinfo);
 
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 9a904c6c0dc8..f8edacdf991d 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -182,21 +182,44 @@ __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
 	}
 	return csum;
 }
-
 EXPORT_SYMBOL(nf_ip_checksum);
 
+static __sum16 nf_ip_checksum_partial(struct sk_buff *skb, unsigned int hook,
+				      unsigned int dataoff, unsigned int len,
+				      u_int8_t protocol)
+{
+	const struct iphdr *iph = ip_hdr(skb);
+	__sum16 csum = 0;
+
+	switch (skb->ip_summed) {
+	case CHECKSUM_COMPLETE:
+		if (len == skb->len - dataoff)
+			return nf_ip_checksum(skb, hook, dataoff, protocol);
+		/* fall through */
+	case CHECKSUM_NONE:
+		skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, protocol,
+					       skb->len - dataoff, 0);
+		skb->ip_summed = CHECKSUM_NONE;
+		csum = __skb_checksum_complete_head(skb, dataoff + len);
+		if (!csum)
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+	}
+	return csum;
+}
+
 static int nf_ip_route(struct dst_entry **dst, struct flowi *fl)
 {
 	return ip_route_output_key(&init_net, (struct rtable **)dst, fl);
 }
 
 static const struct nf_afinfo nf_ip_afinfo = {
-	.family		= AF_INET,
-	.checksum	= nf_ip_checksum,
-	.route		= nf_ip_route,
-	.saveroute	= nf_ip_saveroute,
-	.reroute	= nf_ip_reroute,
-	.route_key_size	= sizeof(struct ip_rt_info),
+	.family			= AF_INET,
+	.checksum		= nf_ip_checksum,
+	.checksum_partial	= nf_ip_checksum_partial,
+	.route			= nf_ip_route,
+	.saveroute		= nf_ip_saveroute,
+	.reroute		= nf_ip_reroute,
+	.route_key_size		= sizeof(struct ip_rt_info),
 };
 
 static int ipv4_netfilter_init(void)
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index aed51bcc66b4..8c6c5e71f210 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -121,16 +121,44 @@ __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
 	}
 	return csum;
 }
-
 EXPORT_SYMBOL(nf_ip6_checksum);
 
+static __sum16 nf_ip6_checksum_partial(struct sk_buff *skb, unsigned int hook,
+				       unsigned int dataoff, unsigned int len,
+				       u_int8_t protocol)
+{
+	struct ipv6hdr *ip6h = ipv6_hdr(skb);
+	__wsum hsum;
+	__sum16 csum = 0;
+
+	switch (skb->ip_summed) {
+	case CHECKSUM_COMPLETE:
+		if (len == skb->len - dataoff)
+			return nf_ip6_checksum(skb, hook, dataoff, protocol);
+		/* fall through */
+	case CHECKSUM_NONE:
+		hsum = skb_checksum(skb, 0, dataoff, 0);
+		skb->csum = ~csum_unfold(csum_ipv6_magic(&ip6h->saddr,
+							 &ip6h->daddr,
+							 skb->len - dataoff,
+							 protocol,
+							 csum_sub(0, hsum)));
+		skb->ip_summed = CHECKSUM_NONE;
+		csum = __skb_checksum_complete_head(skb, dataoff + len);
+		if (!csum)
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+	}
+	return csum;
+};
+
 static const struct nf_afinfo nf_ip6_afinfo = {
-	.family		= AF_INET6,
-	.checksum	= nf_ip6_checksum,
-	.route		= nf_ip6_route,
-	.saveroute	= nf_ip6_saveroute,
-	.reroute	= nf_ip6_reroute,
-	.route_key_size	= sizeof(struct ip6_rt_info),
+	.family			= AF_INET6,
+	.checksum		= nf_ip6_checksum,
+	.checksum_partial	= nf_ip6_checksum_partial,
+	.route			= nf_ip6_route,
+	.saveroute		= nf_ip6_saveroute,
+	.reroute		= nf_ip6_reroute,
+	.route_key_size		= sizeof(struct ip6_rt_info),
 };
 
 int __init ipv6_netfilter_init(void)
diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c
index 9dd03c7aeac6..c3eaee6afffd 100644
--- a/net/netfilter/nf_conntrack_proto_udplite.c
+++ b/net/netfilter/nf_conntrack_proto_udplite.c
@@ -127,32 +127,13 @@ static int udplite_error(struct sk_buff *skb, unsigned int dataoff,
 	}
 
 	/* Checksum invalid? Ignore. */
-	if (nf_conntrack_checksum && !skb_csum_unnecessary(skb) &&
-	    hooknum == NF_INET_PRE_ROUTING) {
-		if (pf == PF_INET) {
-			struct iphdr *iph = ip_hdr(skb);
-
-			skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
-						       udplen, IPPROTO_UDPLITE, 0);
-		} else {
-			struct ipv6hdr *ipv6h = ipv6_hdr(skb);
-			__wsum hsum = skb_checksum(skb, 0, dataoff, 0);
-
-			skb->csum = ~csum_unfold(
-				csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
-						udplen, IPPROTO_UDPLITE,
-						csum_sub(0, hsum)));
-		}
-
-		skb->ip_summed = CHECKSUM_NONE;
-		if (__skb_checksum_complete_head(skb, dataoff + cscov)) {
-			if (LOG_INVALID(IPPROTO_UDPLITE))
-				nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
-					      "nf_ct_udplite: bad UDPLite "
-					      "checksum ");
-			return -NF_ACCEPT;
-		}
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING &&
+	    nf_checksum_partial(skb, hooknum, dataoff, cscov, IPPROTO_UDP,
+	    			pf)) {
+		if (LOG_INVALID(IPPROTO_UDPLITE))
+			nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
+				      "nf_ct_udplite: bad UDPLite checksum ");
+		return -NF_ACCEPT;
 	}
 
 	return NF_ACCEPT;
-- 
cgit v1.2.3


From 2bc780499aa33311ec0f3e42624dfaa7be0ade5e Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 20 Mar 2008 15:15:55 +0100
Subject: [NETFILTER]: nf_conntrack: add DCCP protocol support

Add DCCP conntrack helper. Thanks to Gerrit Renker <gerrit@erg.abdn.ac.uk>
for review and testing.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/nf_conntrack_dccp.h   |  40 ++
 include/linux/netfilter/nfnetlink_conntrack.h |   8 +
 include/net/netfilter/nf_conntrack.h          |   2 +
 include/net/netfilter/nf_conntrack_tuple.h    |   6 +
 net/netfilter/Kconfig                         |  10 +
 net/netfilter/Makefile                        |   1 +
 net/netfilter/nf_conntrack_proto_dccp.c       | 816 ++++++++++++++++++++++++++
 7 files changed, 883 insertions(+)
 create mode 100644 include/linux/netfilter/nf_conntrack_dccp.h
 create mode 100644 net/netfilter/nf_conntrack_proto_dccp.c

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_dccp.h b/include/linux/netfilter/nf_conntrack_dccp.h
new file mode 100644
index 000000000000..40dcc82058d1
--- /dev/null
+++ b/include/linux/netfilter/nf_conntrack_dccp.h
@@ -0,0 +1,40 @@
+#ifndef _NF_CONNTRACK_DCCP_H
+#define _NF_CONNTRACK_DCCP_H
+
+/* Exposed to userspace over nfnetlink */
+enum ct_dccp_states {
+	CT_DCCP_NONE,
+	CT_DCCP_REQUEST,
+	CT_DCCP_RESPOND,
+	CT_DCCP_PARTOPEN,
+	CT_DCCP_OPEN,
+	CT_DCCP_CLOSEREQ,
+	CT_DCCP_CLOSING,
+	CT_DCCP_TIMEWAIT,
+	CT_DCCP_IGNORE,
+	CT_DCCP_INVALID,
+	__CT_DCCP_MAX
+};
+#define CT_DCCP_MAX		(__CT_DCCP_MAX - 1)
+
+enum ct_dccp_roles {
+	CT_DCCP_ROLE_CLIENT,
+	CT_DCCP_ROLE_SERVER,
+	__CT_DCCP_ROLE_MAX
+};
+#define CT_DCCP_ROLE_MAX	(__CT_DCCP_ROLE_MAX - 1)
+
+#ifdef __KERNEL__
+#include <net/netfilter/nf_conntrack_tuple.h>
+
+struct nf_ct_dccp {
+	u_int8_t	role[IP_CT_DIR_MAX];
+	u_int8_t	state;
+	u_int8_t	last_pkt;
+	u_int8_t	last_dir;
+	u_int64_t	handshake_seq;
+};
+
+#endif /* __KERNEL__ */
+
+#endif /* _NF_CONNTRACK_DCCP_H */
diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index e3e1533aba2d..0a383ac083cb 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -80,6 +80,7 @@ enum ctattr_l4proto {
 enum ctattr_protoinfo {
 	CTA_PROTOINFO_UNSPEC,
 	CTA_PROTOINFO_TCP,
+	CTA_PROTOINFO_DCCP,
 	__CTA_PROTOINFO_MAX
 };
 #define CTA_PROTOINFO_MAX (__CTA_PROTOINFO_MAX - 1)
@@ -95,6 +96,13 @@ enum ctattr_protoinfo_tcp {
 };
 #define CTA_PROTOINFO_TCP_MAX (__CTA_PROTOINFO_TCP_MAX - 1)
 
+enum ctattr_protoinfo_dccp {
+	CTA_PROTOINFO_DCCP_UNSPEC,
+	CTA_PROTOINFO_DCCP_STATE,
+	__CTA_PROTOINFO_DCCP_MAX,
+};
+#define CTA_PROTOINFO_DCCP_MAX (__CTA_PROTOINFO_DCCP_MAX - 1)
+
 enum ctattr_counters {
 	CTA_COUNTERS_UNSPEC,
 	CTA_COUNTERS_PACKETS,		/* old 64bit counters */
diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index a3567a7a6d67..bb9fc852e973 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -20,6 +20,7 @@
 #include <asm/atomic.h>
 
 #include <linux/netfilter/nf_conntrack_tcp.h>
+#include <linux/netfilter/nf_conntrack_dccp.h>
 #include <linux/netfilter/nf_conntrack_sctp.h>
 #include <linux/netfilter/nf_conntrack_proto_gre.h>
 #include <net/netfilter/ipv4/nf_conntrack_icmp.h>
@@ -30,6 +31,7 @@
 /* per conntrack: protocol private data */
 union nf_conntrack_proto {
 	/* insert conntrack proto private data here */
+	struct nf_ct_dccp dccp;
 	struct ip_ct_sctp sctp;
 	struct ip_ct_tcp tcp;
 	struct ip_ct_icmp icmp;
diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h
index 168c91754d89..bdeec3461384 100644
--- a/include/net/netfilter/nf_conntrack_tuple.h
+++ b/include/net/netfilter/nf_conntrack_tuple.h
@@ -39,6 +39,9 @@ union nf_conntrack_man_proto
 	struct {
 		__be16 id;
 	} icmp;
+	struct {
+		__be16 port;
+	} dccp;
 	struct {
 		__be16 port;
 	} sctp;
@@ -77,6 +80,9 @@ struct nf_conntrack_tuple
 			struct {
 				u_int8_t type, code;
 			} icmp;
+			struct {
+				__be16 port;
+			} dccp;
 			struct {
 				__be16 port;
 			} sctp;
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index daf5b881064d..c1fc0f1a641c 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -86,6 +86,16 @@ config NF_CONNTRACK_EVENTS
 
 	  If unsure, say `N'.
 
+config NF_CT_PROTO_DCCP
+	tristate 'DCCP protocol connection tracking support (EXPERIMENTAL)'
+	depends on EXPERIMENTAL && NF_CONNTRACK
+	depends on NETFILTER_ADVANCED
+	help
+	  With this option enabled, the layer 3 independent connection
+	  tracking code will be able to do state tracking on DCCP connections.
+
+	  If unsure, say 'N'.
+
 config NF_CT_PROTO_GRE
 	tristate
 	depends on NF_CONNTRACK
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index ea7508387f95..5c4b183f6422 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_NETFILTER_NETLINK_LOG) += nfnetlink_log.o
 obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o
 
 # SCTP protocol connection tracking
+obj-$(CONFIG_NF_CT_PROTO_DCCP) += nf_conntrack_proto_dccp.o
 obj-$(CONFIG_NF_CT_PROTO_GRE) += nf_conntrack_proto_gre.o
 obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o
 obj-$(CONFIG_NF_CT_PROTO_UDPLITE) += nf_conntrack_proto_udplite.o
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
new file mode 100644
index 000000000000..db88c5bcc5fd
--- /dev/null
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -0,0 +1,816 @@
+/*
+ * DCCP connection tracking protocol helper
+ *
+ * Copyright (c) 2005, 2006, 2008 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/sysctl.h>
+#include <linux/spinlock.h>
+#include <linux/skbuff.h>
+#include <linux/dccp.h>
+
+#include <linux/netfilter/nfnetlink_conntrack.h>
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_l4proto.h>
+#include <net/netfilter/nf_log.h>
+
+static DEFINE_RWLOCK(dccp_lock);
+
+static int nf_ct_dccp_loose __read_mostly = 1;
+
+/* Timeouts are based on values from RFC4340:
+ *
+ * - REQUEST:
+ *
+ *   8.1.2. Client Request
+ *
+ *   A client MAY give up on its DCCP-Requests after some time
+ *   (3 minutes, for example).
+ *
+ * - RESPOND:
+ *
+ *   8.1.3. Server Response
+ *
+ *   It MAY also leave the RESPOND state for CLOSED after a timeout of
+ *   not less than 4MSL (8 minutes);
+ *
+ * - PARTOPEN:
+ *
+ *   8.1.5. Handshake Completion
+ *
+ *   If the client remains in PARTOPEN for more than 4MSL (8 minutes),
+ *   it SHOULD reset the connection with Reset Code 2, "Aborted".
+ *
+ * - OPEN:
+ *
+ *   The DCCP timestamp overflows after 11.9 hours. If the connection
+ *   stays idle this long the sequence number won't be recognized
+ *   as valid anymore.
+ *
+ * - CLOSEREQ/CLOSING:
+ *
+ *   8.3. Termination
+ *
+ *   The retransmission timer should initially be set to go off in two
+ *   round-trip times and should back off to not less than once every
+ *   64 seconds ...
+ *
+ * - TIMEWAIT:
+ *
+ *   4.3. States
+ *
+ *   A server or client socket remains in this state for 2MSL (4 minutes)
+ *   after the connection has been town down, ...
+ */
+
+#define DCCP_MSL (2 * 60 * HZ)
+
+static unsigned int dccp_timeout[CT_DCCP_MAX + 1] __read_mostly = {
+	[CT_DCCP_REQUEST]	= 2 * DCCP_MSL,
+	[CT_DCCP_RESPOND]	= 4 * DCCP_MSL,
+	[CT_DCCP_PARTOPEN]	= 4 * DCCP_MSL,
+	[CT_DCCP_OPEN]		= 12 * 3600 * HZ,
+	[CT_DCCP_CLOSEREQ]	= 64 * HZ,
+	[CT_DCCP_CLOSING]	= 64 * HZ,
+	[CT_DCCP_TIMEWAIT]	= 2 * DCCP_MSL,
+};
+
+static const char * const dccp_state_names[] = {
+	[CT_DCCP_NONE]		= "NONE",
+	[CT_DCCP_REQUEST]	= "REQUEST",
+	[CT_DCCP_RESPOND]	= "RESPOND",
+	[CT_DCCP_PARTOPEN]	= "PARTOPEN",
+	[CT_DCCP_OPEN]		= "OPEN",
+	[CT_DCCP_CLOSEREQ]	= "CLOSEREQ",
+	[CT_DCCP_CLOSING]	= "CLOSING",
+	[CT_DCCP_TIMEWAIT]	= "TIMEWAIT",
+	[CT_DCCP_IGNORE]	= "IGNORE",
+	[CT_DCCP_INVALID]	= "INVALID",
+};
+
+#define sNO	CT_DCCP_NONE
+#define sRQ	CT_DCCP_REQUEST
+#define sRS	CT_DCCP_RESPOND
+#define sPO	CT_DCCP_PARTOPEN
+#define sOP	CT_DCCP_OPEN
+#define sCR	CT_DCCP_CLOSEREQ
+#define sCG	CT_DCCP_CLOSING
+#define sTW	CT_DCCP_TIMEWAIT
+#define sIG	CT_DCCP_IGNORE
+#define sIV	CT_DCCP_INVALID
+
+/*
+ * DCCP state transistion table
+ *
+ * The assumption is the same as for TCP tracking:
+ *
+ * We are the man in the middle. All the packets go through us but might
+ * get lost in transit to the destination. It is assumed that the destination
+ * can't receive segments we haven't seen.
+ *
+ * The following states exist:
+ *
+ * NONE:	Initial state, expecting Request
+ * REQUEST:	Request seen, waiting for Response from server
+ * RESPOND:	Response from server seen, waiting for Ack from client
+ * PARTOPEN:	Ack after Response seen, waiting for packet other than Response,
+ * 		Reset or Sync from server
+ * OPEN:	Packet other than Response, Reset or Sync seen
+ * CLOSEREQ:	CloseReq from server seen, expecting Close from client
+ * CLOSING:	Close seen, expecting Reset
+ * TIMEWAIT:	Reset seen
+ * IGNORE:	Not determinable whether packet is valid
+ *
+ * Some states exist only on one side of the connection: REQUEST, RESPOND,
+ * PARTOPEN, CLOSEREQ. For the other side these states are equivalent to
+ * the one it was in before.
+ *
+ * Packets are marked as ignored (sIG) if we don't know if they're valid
+ * (for example a reincarnation of a connection we didn't notice is dead
+ * already) and the server may send back a connection closing Reset or a
+ * Response. They're also used for Sync/SyncAck packets, which we don't
+ * care about.
+ */
+static const u_int8_t
+dccp_state_table[CT_DCCP_ROLE_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = {
+	[CT_DCCP_ROLE_CLIENT] = {
+		[DCCP_PKT_REQUEST] = {
+		/*
+		 * sNO -> sRQ		Regular Request
+		 * sRQ -> sRQ		Retransmitted Request or reincarnation
+		 * sRS -> sRS		Retransmitted Request (apparently Response
+		 * 			got lost after we saw it) or reincarnation
+		 * sPO -> sIG		Ignore, conntrack might be out of sync
+		 * sOP -> sIG		Ignore, conntrack might be out of sync
+		 * sCR -> sIG		Ignore, conntrack might be out of sync
+		 * sCG -> sIG		Ignore, conntrack might be out of sync
+		 * sTW -> sRQ		Reincarnation
+		 *
+		 *	sNO, sRQ, sRS, sPO. sOP, sCR, sCG, sTW, */
+			sRQ, sRQ, sRS, sIG, sIG, sIG, sIG, sRQ,
+		},
+		[DCCP_PKT_RESPONSE] = {
+		/*
+		 * sNO -> sIV		Invalid
+		 * sRQ -> sIG		Ignore, might be response to ignored Request
+		 * sRS -> sIG		Ignore, might be response to ignored Request
+		 * sPO -> sIG		Ignore, might be response to ignored Request
+		 * sOP -> sIG		Ignore, might be response to ignored Request
+		 * sCR -> sIG		Ignore, might be response to ignored Request
+		 * sCG -> sIG		Ignore, might be response to ignored Request
+		 * sTW -> sIV		Invalid, reincarnation in reverse direction
+		 *			goes through sRQ
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIG, sIG, sIG, sIG, sIG, sIG, sIV,
+		},
+		[DCCP_PKT_ACK] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sPO		Ack for Response, move to PARTOPEN (8.1.5.)
+		 * sPO -> sPO		Retransmitted Ack for Response, remain in PARTOPEN
+		 * sOP -> sOP		Regular ACK, remain in OPEN
+		 * sCR -> sCR		Ack in CLOSEREQ MAY be processed (8.3.)
+		 * sCG -> sCG		Ack in CLOSING MAY be processed (8.3.)
+		 * sTW -> sIV
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sPO, sPO, sOP, sCR, sCG, sIV
+		},
+		[DCCP_PKT_DATA] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sIV		No connection
+		 * sPO -> sIV		MUST use DataAck in PARTOPEN state (8.1.5.)
+		 * sOP -> sOP		Regular Data packet
+		 * sCR -> sCR		Data in CLOSEREQ MAY be processed (8.3.)
+		 * sCG -> sCG		Data in CLOSING MAY be processed (8.3.)
+		 * sTW -> sIV
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sIV, sOP, sCR, sCG, sIV,
+		},
+		[DCCP_PKT_DATAACK] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sPO		Ack for Response, move to PARTOPEN (8.1.5.)
+		 * sPO -> sPO		Remain in PARTOPEN state
+		 * sOP -> sOP		Regular DataAck packet in OPEN state
+		 * sCR -> sCR		DataAck in CLOSEREQ MAY be processed (8.3.)
+		 * sCG -> sCG		DataAck in CLOSING MAY be processed (8.3.)
+		 * sTW -> sIV
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sPO, sPO, sOP, sCR, sCG, sIV
+		},
+		[DCCP_PKT_CLOSEREQ] = {
+		/*
+		 * CLOSEREQ may only be sent by the server.
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV
+		},
+		[DCCP_PKT_CLOSE] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sIV		No connection
+		 * sPO -> sCG		Client-initiated close
+		 * sOP -> sCG		Client-initiated close
+		 * sCR -> sCG		Close in response to CloseReq (8.3.)
+		 * sCG -> sCG		Retransmit
+		 * sTW -> sIV		Late retransmit, already in TIME_WAIT
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sCG, sCG, sCG, sIV, sIV
+		},
+		[DCCP_PKT_RESET] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sTW		Sync received or timeout, SHOULD send Reset (8.1.1.)
+		 * sRS -> sTW		Response received without Request
+		 * sPO -> sTW		Timeout, SHOULD send Reset (8.1.5.)
+		 * sOP -> sTW		Connection reset
+		 * sCR -> sTW		Connection reset
+		 * sCG -> sTW		Connection reset
+		 * sTW -> sIG		Ignore (don't refresh timer)
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sTW, sTW, sTW, sTW, sTW, sTW, sIG
+		},
+		[DCCP_PKT_SYNC] = {
+		/*
+		 * We currently ignore Sync packets
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG,
+		},
+		[DCCP_PKT_SYNCACK] = {
+		/*
+		 * We currently ignore SyncAck packets
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG,
+		},
+	},
+	[CT_DCCP_ROLE_SERVER] = {
+		[DCCP_PKT_REQUEST] = {
+		/*
+		 * sNO -> sIV		Invalid
+		 * sRQ -> sIG		Ignore, conntrack might be out of sync
+		 * sRS -> sIG		Ignore, conntrack might be out of sync
+		 * sPO -> sIG		Ignore, conntrack might be out of sync
+		 * sOP -> sIG		Ignore, conntrack might be out of sync
+		 * sCR -> sIG		Ignore, conntrack might be out of sync
+		 * sCG -> sIG		Ignore, conntrack might be out of sync
+		 * sTW -> sRQ		Reincarnation, must reverse roles
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIG, sIG, sIG, sIG, sIG, sIG, sRQ
+		},
+		[DCCP_PKT_RESPONSE] = {
+		/*
+		 * sNO -> sIV		Response without Request
+		 * sRQ -> sRS		Response to clients Request
+		 * sRS -> sRS		Retransmitted Response (8.1.3. SHOULD NOT)
+		 * sPO -> sIG		Response to an ignored Request or late retransmit
+		 * sOP -> sIG		Ignore, might be response to ignored Request
+		 * sCR -> sIG		Ignore, might be response to ignored Request
+		 * sCG -> sIG		Ignore, might be response to ignored Request
+		 * sTW -> sIV		Invalid, Request from client in sTW moves to sRQ
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sRS, sRS, sIG, sIG, sIG, sIG, sIV
+		},
+		[DCCP_PKT_ACK] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sIV		No connection
+		 * sPO -> sOP		Enter OPEN state (8.1.5.)
+		 * sOP -> sOP		Regular Ack in OPEN state
+		 * sCR -> sIV		Waiting for Close from client
+		 * sCG -> sCG		Ack in CLOSING MAY be processed (8.3.)
+		 * sTW -> sIV
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sOP, sOP, sIV, sCG, sIV
+		},
+		[DCCP_PKT_DATA] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sIV		No connection
+		 * sPO -> sOP		Enter OPEN state (8.1.5.)
+		 * sOP -> sOP		Regular Data packet in OPEN state
+		 * sCR -> sIV		Waiting for Close from client
+		 * sCG -> sCG		Data in CLOSING MAY be processed (8.3.)
+		 * sTW -> sIV
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sOP, sOP, sIV, sCG, sIV
+		},
+		[DCCP_PKT_DATAACK] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sIV		No connection
+		 * sPO -> sOP		Enter OPEN state (8.1.5.)
+		 * sOP -> sOP		Regular DataAck in OPEN state
+		 * sCR -> sIV		Waiting for Close from client
+		 * sCG -> sCG		Data in CLOSING MAY be processed (8.3.)
+		 * sTW -> sIV
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sOP, sOP, sIV, sCG, sIV
+		},
+		[DCCP_PKT_CLOSEREQ] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sIV		No connection
+		 * sPO -> sOP -> sCR	Move directly to CLOSEREQ (8.1.5.)
+		 * sOP -> sCR		CloseReq in OPEN state
+		 * sCR -> sCR		Retransmit
+		 * sCG -> sCR		Simultaneous close, client sends another Close
+		 * sTW -> sIV		Already closed
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sCR, sCR, sCR, sCR, sIV
+		},
+		[DCCP_PKT_CLOSE] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sIV		No connection
+		 * sRS -> sIV		No connection
+		 * sPO -> sOP -> sCG	Move direcly to CLOSING
+		 * sOP -> sCG		Move to CLOSING
+		 * sCR -> sIV		Close after CloseReq is invalid
+		 * sCG -> sCG		Retransmit
+		 * sTW -> sIV		Already closed
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIV, sIV, sIV, sCG, sCG, sIV, sCG, sIV
+		},
+		[DCCP_PKT_RESET] = {
+		/*
+		 * sNO -> sIV		No connection
+		 * sRQ -> sTW		Reset in response to Request
+		 * sRS -> sTW		Timeout, SHOULD send Reset (8.1.3.)
+		 * sPO -> sTW		Timeout, SHOULD send Reset (8.1.3.)
+		 * sOP -> sTW
+		 * sCR -> sTW
+		 * sCG -> sTW
+		 * sTW -> sIG		Ignore (don't refresh timer)
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW, sTW */
+			sIV, sTW, sTW, sTW, sTW, sTW, sTW, sTW, sIG
+		},
+		[DCCP_PKT_SYNC] = {
+		/*
+		 * We currently ignore Sync packets
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG,
+		},
+		[DCCP_PKT_SYNCACK] = {
+		/*
+		 * We currently ignore SyncAck packets
+		 *
+		 *	sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
+			sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG,
+		},
+	},
+};
+
+static int dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
+			     struct nf_conntrack_tuple *tuple)
+{
+	struct dccp_hdr _hdr, *dh;
+
+	dh = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr);
+	if (dh == NULL)
+		return 0;
+
+	tuple->src.u.dccp.port = dh->dccph_sport;
+	tuple->dst.u.dccp.port = dh->dccph_dport;
+	return 1;
+}
+
+static int dccp_invert_tuple(struct nf_conntrack_tuple *inv,
+			     const struct nf_conntrack_tuple *tuple)
+{
+	inv->src.u.dccp.port = tuple->dst.u.dccp.port;
+	inv->dst.u.dccp.port = tuple->src.u.dccp.port;
+	return 1;
+}
+
+static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
+		    unsigned int dataoff)
+{
+	int pf = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
+	struct dccp_hdr _dh, *dh;
+	const char *msg;
+	u_int8_t state;
+
+	dh = skb_header_pointer(skb, dataoff, sizeof(_dh), &dh);
+	BUG_ON(dh == NULL);
+
+	state = dccp_state_table[CT_DCCP_ROLE_CLIENT][dh->dccph_type][CT_DCCP_NONE];
+	switch (state) {
+	default:
+		if (nf_ct_dccp_loose == 0) {
+			msg = "nf_ct_dccp: not picking up existing connection ";
+			goto out_invalid;
+		}
+	case CT_DCCP_REQUEST:
+		break;
+	case CT_DCCP_INVALID:
+		msg = "nf_ct_dccp: invalid state transition ";
+		goto out_invalid;
+	}
+
+	ct->proto.dccp.role[IP_CT_DIR_ORIGINAL] = CT_DCCP_ROLE_CLIENT;
+	ct->proto.dccp.role[IP_CT_DIR_REPLY] = CT_DCCP_ROLE_SERVER;
+	ct->proto.dccp.state = CT_DCCP_NONE;
+	return 1;
+
+out_invalid:
+	if (LOG_INVALID(IPPROTO_DCCP))
+		nf_log_packet(pf, 0, skb, NULL, NULL, NULL, msg);
+	return 0;
+}
+
+static u64 dccp_ack_seq(const struct dccp_hdr *dh)
+{
+	const struct dccp_hdr_ack_bits *dhack;
+
+	dhack = (void *)dh + __dccp_basic_hdr_len(dh);
+	return ((u64)ntohs(dhack->dccph_ack_nr_high) << 32) +
+		     ntohl(dhack->dccph_ack_nr_low);
+}
+
+static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb,
+		       unsigned int dataoff, enum ip_conntrack_info ctinfo,
+		       int pf, unsigned int hooknum)
+{
+	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+	struct dccp_hdr _dh, *dh;
+	u_int8_t type, old_state, new_state;
+	enum ct_dccp_roles role;
+
+	dh = skb_header_pointer(skb, dataoff, sizeof(_dh), &dh);
+	BUG_ON(dh == NULL);
+	type = dh->dccph_type;
+
+	if (type == DCCP_PKT_RESET &&
+	    !test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
+		/* Tear down connection immediately if only reply is a RESET */
+		if (del_timer(&ct->timeout))
+			ct->timeout.function((unsigned long)ct);
+		return NF_ACCEPT;
+	}
+
+	write_lock_bh(&dccp_lock);
+
+	role = ct->proto.dccp.role[dir];
+	old_state = ct->proto.dccp.state;
+	new_state = dccp_state_table[role][type][old_state];
+
+	switch (new_state) {
+	case CT_DCCP_REQUEST:
+		if (old_state == CT_DCCP_TIMEWAIT &&
+		    role == CT_DCCP_ROLE_SERVER) {
+			/* Reincarnation in the reverse direction: reopen and
+			 * reverse client/server roles. */
+			ct->proto.dccp.role[dir] = CT_DCCP_ROLE_CLIENT;
+			ct->proto.dccp.role[!dir] = CT_DCCP_ROLE_SERVER;
+		}
+		break;
+	case CT_DCCP_RESPOND:
+		if (old_state == CT_DCCP_REQUEST)
+			ct->proto.dccp.handshake_seq = dccp_hdr_seq(dh);
+		break;
+	case CT_DCCP_PARTOPEN:
+		if (old_state == CT_DCCP_RESPOND &&
+		    type == DCCP_PKT_ACK &&
+		    dccp_ack_seq(dh) == ct->proto.dccp.handshake_seq)
+			set_bit(IPS_ASSURED_BIT, &ct->status);
+		break;
+	case CT_DCCP_IGNORE:
+		/*
+		 * Connection tracking might be out of sync, so we ignore
+		 * packets that might establish a new connection and resync
+		 * if the server responds with a valid Response.
+		 */
+		if (ct->proto.dccp.last_dir == !dir &&
+		    ct->proto.dccp.last_pkt == DCCP_PKT_REQUEST &&
+		    type == DCCP_PKT_RESPONSE) {
+			ct->proto.dccp.role[!dir] = CT_DCCP_ROLE_CLIENT;
+			ct->proto.dccp.role[dir] = CT_DCCP_ROLE_SERVER;
+			ct->proto.dccp.handshake_seq = dccp_hdr_seq(dh);
+			new_state = CT_DCCP_RESPOND;
+			break;
+		}
+		ct->proto.dccp.last_dir = dir;
+		ct->proto.dccp.last_pkt = type;
+
+		write_unlock_bh(&dccp_lock);
+		if (LOG_INVALID(IPPROTO_DCCP))
+			nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
+				      "nf_ct_dccp: invalid packet ignored ");
+		return NF_ACCEPT;
+	case CT_DCCP_INVALID:
+		write_unlock_bh(&dccp_lock);
+		if (LOG_INVALID(IPPROTO_DCCP))
+			nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
+				      "nf_ct_dccp: invalid state transition ");
+		return -NF_ACCEPT;
+	}
+
+	ct->proto.dccp.last_dir = dir;
+	ct->proto.dccp.last_pkt = type;
+	ct->proto.dccp.state = new_state;
+	write_unlock_bh(&dccp_lock);
+	nf_ct_refresh_acct(ct, ctinfo, skb, dccp_timeout[new_state]);
+
+	return NF_ACCEPT;
+}
+
+static int dccp_error(struct sk_buff *skb, unsigned int dataoff,
+		      enum ip_conntrack_info *ctinfo, int pf,
+		      unsigned int hooknum)
+{
+	struct dccp_hdr _dh, *dh;
+	unsigned int dccp_len = skb->len - dataoff;
+	unsigned int cscov;
+	const char *msg;
+
+	dh = skb_header_pointer(skb, dataoff, sizeof(_dh), &dh);
+	if (dh == NULL) {
+		msg = "nf_ct_dccp: short packet ";
+		goto out_invalid;
+	}
+
+	if (dh->dccph_doff * 4 < sizeof(struct dccp_hdr) ||
+	    dh->dccph_doff * 4 > dccp_len) {
+		msg = "nf_ct_dccp: truncated/malformed packet ";
+		goto out_invalid;
+	}
+
+	cscov = dccp_len;
+	if (dh->dccph_cscov) {
+		cscov = (dh->dccph_cscov - 1) * 4;
+		if (cscov > dccp_len) {
+			msg = "nf_ct_dccp: bad checksum coverage ";
+			goto out_invalid;
+		}
+	}
+
+	if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING &&
+	    nf_checksum_partial(skb, hooknum, dataoff, cscov, IPPROTO_DCCP,
+				pf)) {
+		msg = "nf_ct_dccp: bad checksum ";
+		goto out_invalid;
+	}
+
+	if (dh->dccph_type >= DCCP_PKT_INVALID) {
+		msg = "nf_ct_dccp: reserved packet type ";
+		goto out_invalid;
+	}
+
+	return NF_ACCEPT;
+
+out_invalid:
+	if (LOG_INVALID(IPPROTO_DCCP))
+		nf_log_packet(pf, 0, skb, NULL, NULL, NULL, msg);
+	return -NF_ACCEPT;
+}
+
+static int dccp_print_tuple(struct seq_file *s,
+			    const struct nf_conntrack_tuple *tuple)
+{
+	return seq_printf(s, "sport=%hu dport=%hu ",
+			  ntohs(tuple->src.u.dccp.port),
+			  ntohs(tuple->dst.u.dccp.port));
+}
+
+static int dccp_print_conntrack(struct seq_file *s, const struct nf_conn *ct)
+{
+	return seq_printf(s, "%s ", dccp_state_names[ct->proto.dccp.state]);
+}
+
+#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+static int dccp_to_nlattr(struct sk_buff *skb, struct nlattr *nla,
+			  const struct nf_conn *ct)
+{
+	struct nlattr *nest_parms;
+
+	read_lock_bh(&dccp_lock);
+	nest_parms = nla_nest_start(skb, CTA_PROTOINFO_DCCP | NLA_F_NESTED);
+	if (!nest_parms)
+		goto nla_put_failure;
+	NLA_PUT_U8(skb, CTA_PROTOINFO_DCCP_STATE, ct->proto.dccp.state);
+	nla_nest_end(skb, nest_parms);
+	read_unlock_bh(&dccp_lock);
+	return 0;
+
+nla_put_failure:
+	read_unlock_bh(&dccp_lock);
+	return -1;
+}
+
+static const struct nla_policy dccp_nla_policy[CTA_PROTOINFO_DCCP_MAX + 1] = {
+	[CTA_PROTOINFO_DCCP_STATE]	= { .type = NLA_U8 },
+};
+
+static int nlattr_to_dccp(struct nlattr *cda[], struct nf_conn *ct)
+{
+	struct nlattr *attr = cda[CTA_PROTOINFO_DCCP];
+	struct nlattr *tb[CTA_PROTOINFO_DCCP_MAX + 1];
+	int err;
+
+	if (!attr)
+		return 0;
+
+	err = nla_parse_nested(tb, CTA_PROTOINFO_DCCP_MAX, attr,
+			       dccp_nla_policy);
+	if (err < 0)
+		return err;
+
+	if (!tb[CTA_PROTOINFO_DCCP_STATE] ||
+	    nla_get_u8(tb[CTA_PROTOINFO_DCCP_STATE]) >= CT_DCCP_IGNORE)
+		return -EINVAL;
+
+	write_lock_bh(&dccp_lock);
+	ct->proto.dccp.state = nla_get_u8(tb[CTA_PROTOINFO_DCCP_STATE]);
+	write_unlock_bh(&dccp_lock);
+	return 0;
+}
+#endif
+
+#ifdef CONFIG_SYSCTL
+static unsigned int dccp_sysctl_table_users;
+static struct ctl_table_header *dccp_sysctl_header;
+static ctl_table dccp_sysctl_table[] = {
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_timeout_request",
+		.data		= &dccp_timeout[CT_DCCP_REQUEST],
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_timeout_respond",
+		.data		= &dccp_timeout[CT_DCCP_RESPOND],
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_timeout_partopen",
+		.data		= &dccp_timeout[CT_DCCP_PARTOPEN],
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_timeout_open",
+		.data		= &dccp_timeout[CT_DCCP_OPEN],
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_timeout_closereq",
+		.data		= &dccp_timeout[CT_DCCP_CLOSEREQ],
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_timeout_closing",
+		.data		= &dccp_timeout[CT_DCCP_CLOSING],
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_timeout_timewait",
+		.data		= &dccp_timeout[CT_DCCP_TIMEWAIT],
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_dccp_loose",
+		.data		= &nf_ct_dccp_loose,
+		.maxlen		= sizeof(nf_ct_dccp_loose),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
+		.ctl_name	= 0,
+	}
+};
+#endif /* CONFIG_SYSCTL */
+
+static struct nf_conntrack_l4proto dccp_proto4 __read_mostly = {
+	.l3proto		= AF_INET,
+	.l4proto		= IPPROTO_DCCP,
+	.name			= "dccp",
+	.pkt_to_tuple		= dccp_pkt_to_tuple,
+	.invert_tuple		= dccp_invert_tuple,
+	.new			= dccp_new,
+	.packet			= dccp_packet,
+	.error			= dccp_error,
+	.print_tuple		= dccp_print_tuple,
+	.print_conntrack	= dccp_print_conntrack,
+#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+	.to_nlattr		= dccp_to_nlattr,
+	.from_nlattr		= nlattr_to_dccp,
+	.tuple_to_nlattr	= nf_ct_port_tuple_to_nlattr,
+	.nlattr_to_tuple	= nf_ct_port_nlattr_to_tuple,
+	.nla_policy		= nf_ct_port_nla_policy,
+#endif
+#ifdef CONFIG_SYSCTL
+	.ctl_table_users	= &dccp_sysctl_table_users,
+	.ctl_table_header	= &dccp_sysctl_header,
+	.ctl_table		= dccp_sysctl_table,
+#endif
+};
+
+static struct nf_conntrack_l4proto dccp_proto6 __read_mostly = {
+	.l3proto		= AF_INET6,
+	.l4proto		= IPPROTO_DCCP,
+	.name			= "dccp",
+	.pkt_to_tuple		= dccp_pkt_to_tuple,
+	.invert_tuple		= dccp_invert_tuple,
+	.new			= dccp_new,
+	.packet			= dccp_packet,
+	.error			= dccp_error,
+	.print_tuple		= dccp_print_tuple,
+	.print_conntrack	= dccp_print_conntrack,
+#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+	.to_nlattr		= dccp_to_nlattr,
+	.from_nlattr		= nlattr_to_dccp,
+	.tuple_to_nlattr	= nf_ct_port_tuple_to_nlattr,
+	.nlattr_to_tuple	= nf_ct_port_nlattr_to_tuple,
+	.nla_policy		= nf_ct_port_nla_policy,
+#endif
+#ifdef CONFIG_SYSCTL
+	.ctl_table_users	= &dccp_sysctl_table_users,
+	.ctl_table_header	= &dccp_sysctl_header,
+	.ctl_table		= dccp_sysctl_table,
+#endif
+};
+
+static int __init nf_conntrack_proto_dccp_init(void)
+{
+	int err;
+
+	err = nf_conntrack_l4proto_register(&dccp_proto4);
+	if (err < 0)
+		goto err1;
+
+	err = nf_conntrack_l4proto_register(&dccp_proto6);
+	if (err < 0)
+		goto err2;
+	return 0;
+
+err2:
+	nf_conntrack_l4proto_unregister(&dccp_proto4);
+err1:
+	return err;
+}
+
+static void __exit nf_conntrack_proto_dccp_fini(void)
+{
+	nf_conntrack_l4proto_unregister(&dccp_proto6);
+	nf_conntrack_l4proto_unregister(&dccp_proto4);
+}
+
+module_init(nf_conntrack_proto_dccp_init);
+module_exit(nf_conntrack_proto_dccp_fini);
+
+MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
+MODULE_DESCRIPTION("DCCP connection tracking protocol helper");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3


From 8c87238b726e543f8af4bdb4296020a328df4744 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Apr 2008 11:15:51 +0200
Subject: [NETFILTER]: nf_nat: don't add NAT extension for confirmed conntracks

Adding extensions to confirmed conntracks is not allowed to avoid races
on reallocation. Don't setup NAT for confirmed conntracks in case NAT
module is loaded late.

The has one side-effect, the connections existing before the NAT module
was loaded won't enter the bysource hash. The only case where this actually
makes a difference is in case of SNAT to a multirange where the IP before
NAT is also part of the range. Since old connections don't enter the
bysource hash the first new connection from the IP will have a new address
selected. This shouldn't matter at all.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_nat_rule.h    |  3 ---
 net/ipv4/netfilter/nf_nat_rule.c       | 19 -------------------
 net/ipv4/netfilter/nf_nat_standalone.c |  8 ++++----
 3 files changed, 4 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_nat_rule.h b/include/net/netfilter/nf_nat_rule.h
index 75d1825031d7..e4a18ae361c6 100644
--- a/include/net/netfilter/nf_nat_rule.h
+++ b/include/net/netfilter/nf_nat_rule.h
@@ -14,7 +14,4 @@ extern int nf_nat_rule_find(struct sk_buff *skb,
 
 extern unsigned int
 alloc_null_binding(struct nf_conn *ct, unsigned int hooknum);
-
-extern unsigned int
-alloc_null_binding_confirmed(struct nf_conn *ct, unsigned int hooknum);
 #endif /* _NF_NAT_RULE_H */
diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c
index ebe0c7903ae9..e8b4d0d4439e 100644
--- a/net/ipv4/netfilter/nf_nat_rule.c
+++ b/net/ipv4/netfilter/nf_nat_rule.c
@@ -188,25 +188,6 @@ alloc_null_binding(struct nf_conn *ct, unsigned int hooknum)
 	return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum));
 }
 
-unsigned int
-alloc_null_binding_confirmed(struct nf_conn *ct, unsigned int hooknum)
-{
-	__be32 ip
-		= (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC
-		   ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip
-		   : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip);
-	__be16 all
-		= (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC
-		   ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.all
-		   : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u.all);
-	struct nf_nat_range range
-		= { IP_NAT_RANGE_MAP_IPS, ip, ip, { all }, { all } };
-
-	pr_debug("Allocating NULL binding for confirmed %p (%u.%u.%u.%u)\n",
-		 ct, NIPQUAD(ip));
-	return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum));
-}
-
 int nf_nat_rule_find(struct sk_buff *skb,
 		     unsigned int hooknum,
 		     const struct net_device *in,
diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c
index c362f672755a..a366b5865b9c 100644
--- a/net/ipv4/netfilter/nf_nat_standalone.c
+++ b/net/ipv4/netfilter/nf_nat_standalone.c
@@ -102,6 +102,9 @@ nf_nat_fn(unsigned int hooknum,
 
 	nat = nfct_nat(ct);
 	if (!nat) {
+		/* NAT module was loaded late. */
+		if (nf_ct_is_confirmed(ct))
+			return NF_ACCEPT;
 		nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC);
 		if (nat == NULL) {
 			pr_debug("failed to add NAT extension\n");
@@ -127,10 +130,7 @@ nf_nat_fn(unsigned int hooknum,
 		if (!nf_nat_initialized(ct, maniptype)) {
 			unsigned int ret;
 
-			if (unlikely(nf_ct_is_confirmed(ct)))
-				/* NAT module was loaded late */
-				ret = alloc_null_binding_confirmed(ct, hooknum);
-			else if (hooknum == NF_INET_LOCAL_IN)
+			if (hooknum == NF_INET_LOCAL_IN)
 				/* LOCAL_IN hook doesn't have a chain!  */
 				ret = alloc_null_binding(ct, hooknum);
 			else
-- 
cgit v1.2.3


From 55871d04793d9c069ee277b1e98794b88d92ed80 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Apr 2008 11:15:51 +0200
Subject: [NETFILTER]: nf_conntrack_extend: warn on confirmed conntracks

New extensions may only be added to unconfirmed conntracks to avoid races
when reallocating the storage.

Also change NF_CT_ASSERT to use WARN_ON to get backtraces.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack.h | 9 +--------
 net/netfilter/nf_conntrack_extend.c  | 3 +++
 2 files changed, 4 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index bb9fc852e973..4a0496aa32d5 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -65,14 +65,7 @@ union nf_conntrack_help {
 #include <linux/timer.h>
 
 #ifdef CONFIG_NETFILTER_DEBUG
-#define NF_CT_ASSERT(x)							\
-do {									\
-	if (!(x))							\
-		/* Wooah!  I'm tripping my conntrack in a frenzy of	\
-		   netplay... */					\
-		printk("NF_CT_ASSERT: %s:%i(%s)\n",			\
-		       __FILE__, __LINE__, __FUNCTION__);		\
-} while(0)
+#define NF_CT_ASSERT(x)		WARN_ON(!(x))
 #else
 #define NF_CT_ASSERT(x)
 #endif
diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c
index 2bd9963b5b3e..bcc19fa4ed1e 100644
--- a/net/netfilter/nf_conntrack_extend.c
+++ b/net/netfilter/nf_conntrack_extend.c
@@ -71,6 +71,9 @@ void *__nf_ct_ext_add(struct nf_conn *ct, enum nf_ct_ext_id id, gfp_t gfp)
 	int i, newlen, newoff;
 	struct nf_ct_ext_type *t;
 
+	/* Conntrack must not be confirmed to avoid races on reallocation. */
+	NF_CT_ASSERT(!nf_ct_is_confirmed(ct));
+
 	if (!ct->ext)
 		return nf_ct_ext_create(&ct->ext, id, gfp);
 
-- 
cgit v1.2.3


From dd13b010368f85dfa59364ba87bfe8ae930b2832 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Apr 2008 11:15:52 +0200
Subject: [NETFILTER]: nf_nat: kill helper and seq_adjust hooks

Connection tracking helpers (specifically FTP) need to be called
before NAT sequence numbers adjustments are performed to be able
to compare them against previously seen ones. We've introduced
two new hooks around 2.6.11 to maintain this ordering when NAT
modules were changed to get called from conntrack helpers directly.

The cost of netfilter hooks is quite high and sequence number
adjustments are only rarely needed however. Add a RCU-protected
sequence number adjustment function pointer and call it from
IPv4 conntrack after calling the helper.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter_ipv4.h                 |  2 -
 include/net/netfilter/nf_nat_helper.h          |  3 ++
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 56 +++++++++++++-------------
 net/ipv4/netfilter/nf_nat_core.c               |  5 +++
 net/ipv4/netfilter/nf_nat_helper.c             |  1 -
 net/ipv4/netfilter/nf_nat_standalone.c         | 35 ----------------
 6 files changed, 35 insertions(+), 67 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h
index 9a10092e358c..650318b0c405 100644
--- a/include/linux/netfilter_ipv4.h
+++ b/include/linux/netfilter_ipv4.h
@@ -62,8 +62,6 @@ enum nf_ip_hook_priorities {
 	NF_IP_PRI_FILTER = 0,
 	NF_IP_PRI_NAT_SRC = 100,
 	NF_IP_PRI_SELINUX_LAST = 225,
-	NF_IP_PRI_CONNTRACK_HELPER = INT_MAX - 2,
-	NF_IP_PRI_NAT_SEQ_ADJUST = INT_MAX - 1,
 	NF_IP_PRI_CONNTRACK_CONFIRM = INT_MAX,
 	NF_IP_PRI_LAST = INT_MAX,
 };
diff --git a/include/net/netfilter/nf_nat_helper.h b/include/net/netfilter/nf_nat_helper.h
index 58dd22687949..237a961f40e1 100644
--- a/include/net/netfilter/nf_nat_helper.h
+++ b/include/net/netfilter/nf_nat_helper.h
@@ -24,6 +24,9 @@ extern int nf_nat_mangle_udp_packet(struct sk_buff *skb,
 extern int nf_nat_seq_adjust(struct sk_buff *skb,
 			     struct nf_conn *ct,
 			     enum ip_conntrack_info ctinfo);
+extern int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb,
+				     struct nf_conn *ct,
+				     enum ip_conntrack_info ctinfo);
 
 /* Setup NAT on this expected conntrack so it follows master, but goes
  * to port ct->master->saved_proto. */
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
index a65b845c5f15..41e79613eb0a 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
@@ -23,6 +23,12 @@
 #include <net/netfilter/nf_conntrack_l3proto.h>
 #include <net/netfilter/nf_conntrack_core.h>
 #include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
+#include <net/netfilter/nf_nat_helper.h>
+
+int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb,
+			      struct nf_conn *ct,
+			      enum ip_conntrack_info ctinfo);
+EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook);
 
 static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
 			     struct nf_conntrack_tuple *tuple)
@@ -100,36 +106,42 @@ static unsigned int ipv4_confirm(unsigned int hooknum,
 				 const struct net_device *in,
 				 const struct net_device *out,
 				 int (*okfn)(struct sk_buff *))
-{
-	/* We've seen it coming out the other side: confirm it */
-	return nf_conntrack_confirm(skb);
-}
-
-static unsigned int ipv4_conntrack_help(unsigned int hooknum,
-				      struct sk_buff *skb,
-				      const struct net_device *in,
-				      const struct net_device *out,
-				      int (*okfn)(struct sk_buff *))
 {
 	struct nf_conn *ct;
 	enum ip_conntrack_info ctinfo;
 	const struct nf_conn_help *help;
 	const struct nf_conntrack_helper *helper;
+	unsigned int ret;
 
 	/* This is where we call the helper: as the packet goes out. */
 	ct = nf_ct_get(skb, &ctinfo);
 	if (!ct || ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY)
-		return NF_ACCEPT;
+		goto out;
 
 	help = nfct_help(ct);
 	if (!help)
-		return NF_ACCEPT;
+		goto out;
+
 	/* rcu_read_lock()ed by nf_hook_slow */
 	helper = rcu_dereference(help->helper);
 	if (!helper)
-		return NF_ACCEPT;
-	return helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
-			    ct, ctinfo);
+		goto out;
+
+	ret = helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
+			   ct, ctinfo);
+	if (ret != NF_ACCEPT)
+		return ret;
+
+	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) {
+		typeof(nf_nat_seq_adjust_hook) seq_adjust;
+
+		seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook);
+		if (!seq_adjust || !seq_adjust(skb, ct, ctinfo))
+			return NF_DROP;
+	}
+out:
+	/* We've seen it coming out the other side: confirm it */
+	return nf_conntrack_confirm(skb);
 }
 
 static unsigned int ipv4_conntrack_defrag(unsigned int hooknum,
@@ -210,20 +222,6 @@ static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = {
 		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_CONNTRACK,
 	},
-	{
-		.hook		= ipv4_conntrack_help,
-		.owner		= THIS_MODULE,
-		.pf		= PF_INET,
-		.hooknum	= NF_INET_POST_ROUTING,
-		.priority	= NF_IP_PRI_CONNTRACK_HELPER,
-	},
-	{
-		.hook		= ipv4_conntrack_help,
-		.owner		= THIS_MODULE,
-		.pf		= PF_INET,
-		.hooknum	= NF_INET_LOCAL_IN,
-		.priority	= NF_IP_PRI_CONNTRACK_HELPER,
-	},
 	{
 		.hook		= ipv4_confirm,
 		.owner		= THIS_MODULE,
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index 9320c7ac5729..25c3efe4207e 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -618,6 +618,9 @@ static int __init nf_nat_init(void)
 	nf_conntrack_untracked.status |= IPS_NAT_DONE_MASK;
 
 	l3proto = nf_ct_l3proto_find_get((u_int16_t)AF_INET);
+
+	BUG_ON(nf_nat_seq_adjust_hook != NULL);
+	rcu_assign_pointer(nf_nat_seq_adjust_hook, nf_nat_seq_adjust);
 	return 0;
 
  cleanup_extend:
@@ -644,6 +647,8 @@ static void __exit nf_nat_cleanup(void)
 	nf_ct_free_hashtable(bysource, nf_nat_vmalloced, nf_nat_htable_size);
 	nf_ct_l3proto_put(l3proto);
 	nf_ct_extend_unregister(&nat_extend);
+	rcu_assign_pointer(nf_nat_seq_adjust_hook, NULL);
+	synchronize_net();
 }
 
 MODULE_LICENSE("GPL");
diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c
index 2fca727aa8ba..11976ea29884 100644
--- a/net/ipv4/netfilter/nf_nat_helper.c
+++ b/net/ipv4/netfilter/nf_nat_helper.c
@@ -416,7 +416,6 @@ nf_nat_seq_adjust(struct sk_buff *skb,
 
 	return 1;
 }
-EXPORT_SYMBOL(nf_nat_seq_adjust);
 
 /* Setup NAT on this expected conntrack so it follows master. */
 /* If we fail to get a free NAT slot, we'll get dropped on confirm */
diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c
index a366b5865b9c..b7dd695691a0 100644
--- a/net/ipv4/netfilter/nf_nat_standalone.c
+++ b/net/ipv4/netfilter/nf_nat_standalone.c
@@ -245,25 +245,6 @@ nf_nat_local_fn(unsigned int hooknum,
 	return ret;
 }
 
-static unsigned int
-nf_nat_adjust(unsigned int hooknum,
-	      struct sk_buff *skb,
-	      const struct net_device *in,
-	      const struct net_device *out,
-	      int (*okfn)(struct sk_buff *))
-{
-	struct nf_conn *ct;
-	enum ip_conntrack_info ctinfo;
-
-	ct = nf_ct_get(skb, &ctinfo);
-	if (ct && test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) {
-		pr_debug("nf_nat_standalone: adjusting sequence number\n");
-		if (!nf_nat_seq_adjust(skb, ct, ctinfo))
-			return NF_DROP;
-	}
-	return NF_ACCEPT;
-}
-
 /* We must be after connection tracking and before packet filtering. */
 
 static struct nf_hook_ops nf_nat_ops[] __read_mostly = {
@@ -283,14 +264,6 @@ static struct nf_hook_ops nf_nat_ops[] __read_mostly = {
 		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP_PRI_NAT_SRC,
 	},
-	/* After conntrack, adjust sequence number */
-	{
-		.hook		= nf_nat_adjust,
-		.owner		= THIS_MODULE,
-		.pf		= PF_INET,
-		.hooknum	= NF_INET_POST_ROUTING,
-		.priority	= NF_IP_PRI_NAT_SEQ_ADJUST,
-	},
 	/* Before packet filtering, change destination */
 	{
 		.hook		= nf_nat_local_fn,
@@ -307,14 +280,6 @@ static struct nf_hook_ops nf_nat_ops[] __read_mostly = {
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_NAT_SRC,
 	},
-	/* After conntrack, adjust sequence number */
-	{
-		.hook		= nf_nat_adjust,
-		.owner		= THIS_MODULE,
-		.pf		= PF_INET,
-		.hooknum	= NF_INET_LOCAL_IN,
-		.priority	= NF_IP_PRI_NAT_SEQ_ADJUST,
-	},
 };
 
 static int __init nf_nat_standalone_init(void)
-- 
cgit v1.2.3


From 5e8fbe2ac8a3f1e34e7004c5750ef59bf9304f82 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Apr 2008 11:15:52 +0200
Subject: [NETFILTER]: nf_conntrack: add tuplehash l3num/protonum accessors

Add accessors for l3num and protonum and get rid of some overly long
expressions.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack.h               | 10 ++++++
 .../netfilter/nf_conntrack_l3proto_ipv4_compat.c   | 13 +++-----
 net/netfilter/nf_conntrack_amanda.c                |  4 +--
 net/netfilter/nf_conntrack_core.c                  |  3 +-
 net/netfilter/nf_conntrack_ftp.c                   |  4 +--
 net/netfilter/nf_conntrack_h323_main.c             | 39 ++++++++--------------
 net/netfilter/nf_conntrack_netlink.c               | 19 ++++-------
 net/netfilter/nf_conntrack_pptp.c                  |  4 +--
 net/netfilter/nf_conntrack_proto.c                 |  9 ++---
 net/netfilter/nf_conntrack_proto_dccp.c            |  3 +-
 net/netfilter/nf_conntrack_sane.c                  |  3 +-
 net/netfilter/nf_conntrack_sip.c                   | 20 +++++------
 net/netfilter/nf_conntrack_standalone.c            | 15 +++------
 net/netfilter/nf_conntrack_tftp.c                  |  4 +--
 net/netfilter/xt_connlimit.c                       |  4 +--
 net/netfilter/xt_conntrack.c                       |  4 +--
 16 files changed, 64 insertions(+), 94 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 4a0496aa32d5..26e6a6e2b5a2 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -140,6 +140,16 @@ nf_ct_tuplehash_to_ctrack(const struct nf_conntrack_tuple_hash *hash)
 			    tuplehash[hash->tuple.dst.dir]);
 }
 
+static inline u_int16_t nf_ct_l3num(const struct nf_conn *ct)
+{
+	return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
+}
+
+static inline u_int8_t nf_ct_protonum(const struct nf_conn *ct)
+{
+	return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum;
+}
+
 /* get master conntrack via master expectation */
 #define master_ct(conntr) (conntr->master)
 
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
index e60b885d2dcd..40a46d482490 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
@@ -106,21 +106,16 @@ static int ct_seq_show(struct seq_file *s, void *v)
 	/* we only want to print DIR_ORIGINAL */
 	if (NF_CT_DIRECTION(hash))
 		return 0;
-	if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num != AF_INET)
+	if (nf_ct_l3num(ct) != AF_INET)
 		return 0;
 
-	l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL]
-				       .tuple.src.l3num);
+	l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct));
 	NF_CT_ASSERT(l3proto);
-	l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL]
-				       .tuple.src.l3num,
-				       ct->tuplehash[IP_CT_DIR_ORIGINAL]
-				       .tuple.dst.protonum);
+	l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct));
 	NF_CT_ASSERT(l4proto);
 
 	if (seq_printf(s, "%-8s %u %ld ",
-		      l4proto->name,
-		      ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum,
+		      l4proto->name, nf_ct_protonum(ct),
 		      timer_pending(&ct->timeout)
 		      ? (long)(ct->timeout.expires - jiffies)/HZ : 0) != 0)
 		return -ENOSPC;
diff --git a/net/netfilter/nf_conntrack_amanda.c b/net/netfilter/nf_conntrack_amanda.c
index ddfac99cbe63..38aedeeaf4e1 100644
--- a/net/netfilter/nf_conntrack_amanda.c
+++ b/net/netfilter/nf_conntrack_amanda.c
@@ -91,7 +91,6 @@ static int amanda_help(struct sk_buff *skb,
 	char pbuf[sizeof("65535")], *tmp;
 	u_int16_t len;
 	__be16 port;
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	int ret = NF_ACCEPT;
 	typeof(nf_nat_amanda_hook) nf_nat_amanda;
 
@@ -148,7 +147,8 @@ static int amanda_help(struct sk_buff *skb,
 			goto out;
 		}
 		tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
-		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
+		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+				  nf_ct_l3num(ct),
 				  &tuple->src.u3, &tuple->dst.u3,
 				  IPPROTO_TCP, NULL, &port);
 
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index b77eb56a87e3..21ab0c3846ac 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -194,8 +194,7 @@ destroy_conntrack(struct nf_conntrack *nfct)
 	 * destroy_conntrack() MUST NOT be called with a write lock
 	 * to nf_conntrack_lock!!! -HW */
 	rcu_read_lock();
-	l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num,
-				       ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum);
+	l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct));
 	if (l4proto && l4proto->destroy)
 		l4proto->destroy(ct);
 
diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c
index 87ca39b353bb..bb20672fe036 100644
--- a/net/netfilter/nf_conntrack_ftp.c
+++ b/net/netfilter/nf_conntrack_ftp.c
@@ -406,7 +406,7 @@ static int help(struct sk_buff *skb,
 
 	/* Initialize IP/IPv6 addr to expected address (it's not mentioned
 	   in EPSV responses) */
-	cmd.l3num = ct->tuplehash[dir].tuple.src.l3num;
+	cmd.l3num = nf_ct_l3num(ct);
 	memcpy(cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all,
 	       sizeof(cmd.u3.all));
 
@@ -453,7 +453,7 @@ static int help(struct sk_buff *skb,
 	daddr = &ct->tuplehash[!dir].tuple.dst.u3;
 
 	/* Update the ftp info */
-	if ((cmd.l3num == ct->tuplehash[dir].tuple.src.l3num) &&
+	if ((cmd.l3num == nf_ct_l3num(ct)) &&
 	    memcmp(&cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all,
 		     sizeof(cmd.u3.all))) {
 		/* Enrico Scholz's passive FTP to partially RNAT'd ftp
diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c
index 505052d495cf..c3f87094de43 100644
--- a/net/netfilter/nf_conntrack_h323_main.c
+++ b/net/netfilter/nf_conntrack_h323_main.c
@@ -218,7 +218,6 @@ static int get_h245_addr(struct nf_conn *ct, const unsigned char *data,
 			 union nf_inet_addr *addr, __be16 *port)
 {
 	const unsigned char *p;
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	int len;
 
 	if (taddr->choice != eH245_TransportAddress_unicastAddress)
@@ -226,13 +225,13 @@ static int get_h245_addr(struct nf_conn *ct, const unsigned char *data,
 
 	switch (taddr->unicastAddress.choice) {
 	case eUnicastAddress_iPAddress:
-		if (family != AF_INET)
+		if (nf_ct_l3num(ct) != AF_INET)
 			return 0;
 		p = data + taddr->unicastAddress.iPAddress.network;
 		len = 4;
 		break;
 	case eUnicastAddress_iP6Address:
-		if (family != AF_INET6)
+		if (nf_ct_l3num(ct) != AF_INET6)
 			return 0;
 		p = data + taddr->unicastAddress.iP6Address.network;
 		len = 16;
@@ -277,8 +276,7 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct,
 	/* Create expect for RTP */
 	if ((rtp_exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(rtp_exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(rtp_exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_UDP, NULL, &rtp_port);
@@ -288,8 +286,7 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct,
 		nf_ct_expect_put(rtp_exp);
 		return -1;
 	}
-	nf_ct_expect_init(rtcp_exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(rtcp_exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_UDP, NULL, &rtcp_port);
@@ -346,8 +343,7 @@ static int expect_t120(struct sk_buff *skb,
 	/* Create expect for T.120 connections */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_TCP, NULL, &port);
@@ -634,18 +630,17 @@ int get_h225_addr(struct nf_conn *ct, unsigned char *data,
 		  union nf_inet_addr *addr, __be16 *port)
 {
 	const unsigned char *p;
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	int len;
 
 	switch (taddr->choice) {
 	case eTransportAddress_ipAddress:
-		if (family != AF_INET)
+		if (nf_ct_l3num(ct) != AF_INET)
 			return 0;
 		p = data + taddr->ipAddress.ip;
 		len = 4;
 		break;
 	case eTransportAddress_ip6Address:
-		if (family != AF_INET6)
+		if (nf_ct_l3num(ct) != AF_INET6)
 			return 0;
 		p = data + taddr->ip6Address.ip;
 		len = 16;
@@ -683,8 +678,7 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct,
 	/* Create expect for h245 connection */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
 			  IPPROTO_TCP, NULL, &port);
@@ -792,7 +786,7 @@ static int expect_callforwarding(struct sk_buff *skb,
 	 * we don't need to track the second call */
 	if (callforward_filter &&
 	    callforward_do_filter(&addr, &ct->tuplehash[!dir].tuple.src.u3,
-				  ct->tuplehash[!dir].tuple.src.l3num)) {
+				  nf_ct_l3num(ct))) {
 		pr_debug("nf_ct_q931: Call Forwarding not tracked\n");
 		return 0;
 	}
@@ -800,8 +794,7 @@ static int expect_callforwarding(struct sk_buff *skb,
 	/* Create expect for the second call leg */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_TCP, NULL, &port);
 	exp->helper = nf_conntrack_helper_q931;
@@ -1272,8 +1265,7 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct,
 	/* Create expect for Q.931 */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  gkrouted_only ? /* only accept calls from GK? */
 				&ct->tuplehash[!dir].tuple.src.u3 : NULL,
 			  &ct->tuplehash[!dir].tuple.dst.u3,
@@ -1344,8 +1336,7 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct,
 	/* Need new expect */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_UDP, NULL, &port);
 	exp->helper = nf_conntrack_helper_ras;
@@ -1549,8 +1540,7 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct,
 	/* Need new expect */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_TCP, NULL, &port);
 	exp->flags = NF_CT_EXPECT_PERMANENT;
@@ -1603,8 +1593,7 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct,
 	/* Need new expect for call signal */
 	if ((exp = nf_ct_expect_alloc(ct)) == NULL)
 		return -1;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[!dir].tuple.src.l3num,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &ct->tuplehash[!dir].tuple.src.u3, &addr,
 			  IPPROTO_TCP, NULL, &port);
 	exp->flags = NF_CT_EXPECT_PERMANENT;
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 79d5ecde0ddc..16774ecd1c4e 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -145,10 +145,11 @@ nla_put_failure:
 static inline int
 ctnetlink_dump_protoinfo(struct sk_buff *skb, const struct nf_conn *ct)
 {
-	struct nf_conntrack_l4proto *l4proto = nf_ct_l4proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum);
+	struct nf_conntrack_l4proto *l4proto;
 	struct nlattr *nest_proto;
 	int ret;
 
+	l4proto = nf_ct_l4proto_find_get(nf_ct_l3num(ct), nf_ct_protonum(ct));
 	if (!l4proto->to_nlattr) {
 		nf_ct_l4proto_put(l4proto);
 		return 0;
@@ -368,8 +369,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq,
 	nfmsg  = NLMSG_DATA(nlh);
 
 	nlh->nlmsg_flags    = (nowait && pid) ? NLM_F_MULTI : 0;
-	nfmsg->nfgen_family =
-		ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
+	nfmsg->nfgen_family = nf_ct_l3num(ct);
 	nfmsg->version      = NFNETLINK_V0;
 	nfmsg->res_id	    = 0;
 
@@ -454,7 +454,7 @@ static int ctnetlink_conntrack_event(struct notifier_block *this,
 	nfmsg = NLMSG_DATA(nlh);
 
 	nlh->nlmsg_flags    = flags;
-	nfmsg->nfgen_family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
+	nfmsg->nfgen_family = nf_ct_l3num(ct);
 	nfmsg->version	= NFNETLINK_V0;
 	nfmsg->res_id	= 0;
 
@@ -535,8 +535,6 @@ static int ctnetlink_done(struct netlink_callback *cb)
 	return 0;
 }
 
-#define L3PROTO(ct) (ct)->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num
-
 static int
 ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb)
 {
@@ -558,7 +556,7 @@ restart:
 			/* Dump entries of a given L3 protocol number.
 			 * If it is not specified, ie. l3proto == 0,
 			 * then dump everything. */
-			if (l3proto && L3PROTO(ct) != l3proto)
+			if (l3proto && nf_ct_l3num(ct) != l3proto)
 				continue;
 			if (cb->args[1]) {
 				if (ct != last)
@@ -704,7 +702,7 @@ static int nfnetlink_parse_nat_proto(struct nlattr *attr,
 	if (err < 0)
 		return err;
 
-	npt = nf_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum);
+	npt = nf_nat_proto_find_get(nf_ct_protonum(ct));
 	if (npt->nlattr_to_range)
 		err = npt->nlattr_to_range(tb, range);
 	nf_nat_proto_put(npt);
@@ -1001,14 +999,11 @@ ctnetlink_change_protoinfo(struct nf_conn *ct, struct nlattr *cda[])
 {
 	struct nlattr *tb[CTA_PROTOINFO_MAX+1], *attr = cda[CTA_PROTOINFO];
 	struct nf_conntrack_l4proto *l4proto;
-	u_int16_t npt = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum;
-	u_int16_t l3num = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	int err = 0;
 
 	nla_parse_nested(tb, CTA_PROTOINFO_MAX, attr, NULL);
 
-	l4proto = nf_ct_l4proto_find_get(l3num, npt);
-
+	l4proto = nf_ct_l4proto_find_get(nf_ct_l3num(ct), nf_ct_protonum(ct));
 	if (l4proto->from_nlattr)
 		err = l4proto->from_nlattr(tb, ct);
 	nf_ct_l4proto_put(l4proto);
diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c
index 8fd83470d1b3..4793cc078789 100644
--- a/net/netfilter/nf_conntrack_pptp.c
+++ b/net/netfilter/nf_conntrack_pptp.c
@@ -209,7 +209,7 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid)
 	/* original direction, PNS->PAC */
 	dir = IP_CT_DIR_ORIGINAL;
 	nf_ct_expect_init(exp_orig, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[dir].tuple.src.l3num,
+			  nf_ct_l3num(ct),
 			  &ct->tuplehash[dir].tuple.src.u3,
 			  &ct->tuplehash[dir].tuple.dst.u3,
 			  IPPROTO_GRE, &peer_callid, &callid);
@@ -218,7 +218,7 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid)
 	/* reply direction, PAC->PNS */
 	dir = IP_CT_DIR_REPLY;
 	nf_ct_expect_init(exp_reply, NF_CT_EXPECT_CLASS_DEFAULT,
-			  ct->tuplehash[dir].tuple.src.l3num,
+			  nf_ct_l3num(ct),
 			  &ct->tuplehash[dir].tuple.src.u3,
 			  &ct->tuplehash[dir].tuple.dst.u3,
 			  IPPROTO_GRE, &callid, &peer_callid);
diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c
index 8595b5946acf..a49fc932629b 100644
--- a/net/netfilter/nf_conntrack_proto.c
+++ b/net/netfilter/nf_conntrack_proto.c
@@ -146,18 +146,15 @@ EXPORT_SYMBOL_GPL(nf_ct_l3proto_module_put);
 
 static int kill_l3proto(struct nf_conn *i, void *data)
 {
-	return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num ==
-			((struct nf_conntrack_l3proto *)data)->l3proto);
+	return nf_ct_l3num(i) == ((struct nf_conntrack_l3proto *)data)->l3proto;
 }
 
 static int kill_l4proto(struct nf_conn *i, void *data)
 {
 	struct nf_conntrack_l4proto *l4proto;
 	l4proto = (struct nf_conntrack_l4proto *)data;
-	return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum ==
-			l4proto->l4proto) &&
-	       (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num ==
-			l4proto->l3proto);
+	return nf_ct_protonum(i) == l4proto->l4proto &&
+	       nf_ct_l3num(i) == l4proto->l3proto;
 }
 
 static int nf_ct_l3proto_register_sysctl(struct nf_conntrack_l3proto *l3proto)
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index db88c5bcc5fd..9376dcd394bd 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -418,7 +418,6 @@ static int dccp_invert_tuple(struct nf_conntrack_tuple *inv,
 static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
 		    unsigned int dataoff)
 {
-	int pf = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	struct dccp_hdr _dh, *dh;
 	const char *msg;
 	u_int8_t state;
@@ -447,7 +446,7 @@ static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
 
 out_invalid:
 	if (LOG_INVALID(IPPROTO_DCCP))
-		nf_log_packet(pf, 0, skb, NULL, NULL, NULL, msg);
+		nf_log_packet(nf_ct_l3num(ct), 0, skb, NULL, NULL, NULL, msg);
 	return 0;
 }
 
diff --git a/net/netfilter/nf_conntrack_sane.c b/net/netfilter/nf_conntrack_sane.c
index 7542e25eede3..c3d5e84dcc9b 100644
--- a/net/netfilter/nf_conntrack_sane.c
+++ b/net/netfilter/nf_conntrack_sane.c
@@ -72,7 +72,6 @@ static int help(struct sk_buff *skb,
 	struct nf_conntrack_tuple *tuple;
 	struct sane_request *req;
 	struct sane_reply_net_start *reply;
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 
 	ct_sane_info = &nfct_help(ct)->help.ct_sane_info;
 	/* Until there's been traffic both ways, don't look in packets. */
@@ -143,7 +142,7 @@ static int help(struct sk_buff *skb,
 	}
 
 	tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
-	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
 			  &tuple->src.u3, &tuple->dst.u3,
 			  IPPROTO_TCP, NULL, &reply->port);
 
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index f3915f8724f9..65b3ba57a3b7 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -142,11 +142,10 @@ static int parse_addr(const struct nf_conn *ct, const char *cp,
                       const char *limit)
 {
 	const char *end;
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	int ret = 0;
 
 	memset(addr, 0, sizeof(*addr));
-	switch (family) {
+	switch (nf_ct_l3num(ct)) {
 	case AF_INET:
 		ret = in4_pton(cp, limit - cp, (u8 *)&addr->ip, -1, &end);
 		break;
@@ -740,7 +739,6 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	union nf_inet_addr *saddr;
 	struct nf_conntrack_tuple tuple;
-	int family = ct->tuplehash[!dir].tuple.src.l3num;
 	int direct_rtp = 0, skip_expect = 0, ret = NF_DROP;
 	u_int16_t base_port;
 	__be16 rtp_port, rtcp_port;
@@ -770,7 +768,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	memset(&tuple, 0, sizeof(tuple));
 	if (saddr)
 		tuple.src.u3 = *saddr;
-	tuple.src.l3num		= family;
+	tuple.src.l3num		= nf_ct_l3num(ct);
 	tuple.dst.protonum	= IPPROTO_UDP;
 	tuple.dst.u3		= *daddr;
 	tuple.dst.u.udp.port	= port;
@@ -815,13 +813,13 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb,
 	rtp_exp = nf_ct_expect_alloc(ct);
 	if (rtp_exp == NULL)
 		goto err1;
-	nf_ct_expect_init(rtp_exp, class, family, saddr, daddr,
+	nf_ct_expect_init(rtp_exp, class, nf_ct_l3num(ct), saddr, daddr,
 			  IPPROTO_UDP, NULL, &rtp_port);
 
 	rtcp_exp = nf_ct_expect_alloc(ct);
 	if (rtcp_exp == NULL)
 		goto err2;
-	nf_ct_expect_init(rtcp_exp, class, family, saddr, daddr,
+	nf_ct_expect_init(rtcp_exp, class, nf_ct_l3num(ct), saddr, daddr,
 			  IPPROTO_UDP, NULL, &rtcp_port);
 
 	nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook);
@@ -871,7 +869,6 @@ static int process_sdp(struct sk_buff *skb,
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	unsigned int matchoff, matchlen;
 	unsigned int mediaoff, medialen;
 	unsigned int sdpoff;
@@ -886,8 +883,8 @@ static int process_sdp(struct sk_buff *skb,
 	typeof(nf_nat_sdp_session_hook) nf_nat_sdp_session;
 
 	nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook);
-	c_hdr = family == AF_INET ? SDP_HDR_CONNECTION_IP4 :
-				    SDP_HDR_CONNECTION_IP6;
+	c_hdr = nf_ct_l3num(ct) == AF_INET ? SDP_HDR_CONNECTION_IP4 :
+					     SDP_HDR_CONNECTION_IP6;
 
 	/* Find beginning of session description */
 	if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen,
@@ -1034,7 +1031,6 @@ static int process_register_request(struct sk_buff *skb,
 	struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
 	struct nf_conn_help *help = nfct_help(ct);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	unsigned int matchoff, matchlen;
 	struct nf_conntrack_expect *exp;
 	union nf_inet_addr *saddr, daddr;
@@ -1089,8 +1085,8 @@ static int process_register_request(struct sk_buff *skb,
 	if (sip_direct_signalling)
 		saddr = &ct->tuplehash[!dir].tuple.src.u3;
 
-	nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, family, saddr, &daddr,
-			  IPPROTO_UDP, NULL, &port);
+	nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, nf_ct_l3num(ct),
+			  saddr, &daddr, IPPROTO_UDP, NULL, &port);
 	exp->timeout.expires = sip_timeout * HZ;
 	exp->helper = nfct_help(ct)->helper;
 	exp->flags = NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE;
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index 01d1f7e178f3..b59871f6bdda 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -127,21 +127,14 @@ static int ct_seq_show(struct seq_file *s, void *v)
 	if (NF_CT_DIRECTION(hash))
 		return 0;
 
-	l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL]
-				       .tuple.src.l3num);
-
+	l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct));
 	NF_CT_ASSERT(l3proto);
-	l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL]
-				   .tuple.src.l3num,
-				   ct->tuplehash[IP_CT_DIR_ORIGINAL]
-				   .tuple.dst.protonum);
+	l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct));
 	NF_CT_ASSERT(l4proto);
 
 	if (seq_printf(s, "%-8s %u %-8s %u %ld ",
-		       l3proto->name,
-		       ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num,
-		       l4proto->name,
-		       ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum,
+		       l3proto->name, nf_ct_l3num(ct),
+		       l4proto->name, nf_ct_protonum(ct),
 		       timer_pending(&ct->timeout)
 		       ? (long)(ct->timeout.expires - jiffies)/HZ : 0) != 0)
 		return -ENOSPC;
diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c
index a28341b30f21..ea5ff49d77bc 100644
--- a/net/netfilter/nf_conntrack_tftp.c
+++ b/net/netfilter/nf_conntrack_tftp.c
@@ -44,7 +44,6 @@ static int tftp_help(struct sk_buff *skb,
 	struct nf_conntrack_expect *exp;
 	struct nf_conntrack_tuple *tuple;
 	unsigned int ret = NF_ACCEPT;
-	int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num;
 	typeof(nf_nat_tftp_hook) nf_nat_tftp;
 
 	tfh = skb_header_pointer(skb, protoff + sizeof(struct udphdr),
@@ -63,7 +62,8 @@ static int tftp_help(struct sk_buff *skb,
 		if (exp == NULL)
 			return NF_DROP;
 		tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family,
+		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
+				  nf_ct_l3num(ct),
 				  &tuple->src.u3, &tuple->dst.u3,
 				  IPPROTO_UDP, NULL, &tuple->dst.u.udp.port);
 
diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c
index 0ca9fe9da203..2e89a00df92c 100644
--- a/net/netfilter/xt_connlimit.c
+++ b/net/netfilter/xt_connlimit.c
@@ -72,9 +72,7 @@ connlimit_iphash6(const union nf_inet_addr *addr,
 
 static inline bool already_closed(const struct nf_conn *conn)
 {
-	u_int16_t proto = conn->tuplehash[0].tuple.dst.protonum;
-
-	if (proto == IPPROTO_TCP)
+	if (nf_ct_protonum(conn) == IPPROTO_TCP)
 		return conn->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT;
 	else
 		return 0;
diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c
index 0c50b2894055..d61412f58ef7 100644
--- a/net/netfilter/xt_conntrack.c
+++ b/net/netfilter/xt_conntrack.c
@@ -65,7 +65,7 @@ conntrack_mt_v0(const struct sk_buff *skb, const struct net_device *in,
 	}
 
 	if (sinfo->flags & XT_CONNTRACK_PROTO &&
-	    FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum !=
+	    FWINV(nf_ct_protonum(ct) !=
 		  sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum,
 		  XT_CONNTRACK_PROTO))
 		return false;
@@ -174,7 +174,7 @@ ct_proto_port_check(const struct xt_conntrack_mtinfo1 *info,
 
 	tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
 	if ((info->match_flags & XT_CONNTRACK_PROTO) &&
-	    (tuple->dst.protonum == info->l4proto) ^
+	    (nf_ct_protonum(ct) == info->l4proto) ^
 	    !(info->invert_flags & XT_CONNTRACK_PROTO))
 		return false;
 
-- 
cgit v1.2.3


From 9dbae7917899d78a094aceeb5062cd76efa89052 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:52 +0200
Subject: [NETFILTER]: Remove unused callbacks in nf_conntrack_l3proto

These functions are never called.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack_l3proto.h | 11 -----------
 1 file changed, 11 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h
index b886e3ae6cad..db0825803b16 100644
--- a/include/net/netfilter/nf_conntrack_l3proto.h
+++ b/include/net/netfilter/nf_conntrack_l3proto.h
@@ -42,17 +42,6 @@ struct nf_conntrack_l3proto
 	int (*print_tuple)(struct seq_file *s,
 			   const struct nf_conntrack_tuple *);
 
-	/* Returns verdict for packet, or -1 for invalid. */
-	int (*packet)(struct nf_conn *ct,
-		      const struct sk_buff *skb,
-		      enum ip_conntrack_info ctinfo);
-
-	/*
-	 * Called when a new connection for this protocol found;
-	 * returns TRUE if it's OK.  If so, packet() called next.
-	 */
-	int (*new)(struct nf_conn *ct, const struct sk_buff *skb);
-
 	/*
 	 * Called before tracking. 
 	 *	*dataoff: offset of protocol header (TCP, UDP,...) in skb
-- 
cgit v1.2.3


From 8ce8439a31f723f3aa28adf27fe8797a5678dde1 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:52 +0200
Subject: [NETFILTER]: nf_conntrack: use bool type in struct
 nf_conntrack_l3proto

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack_l3proto.h   |  8 ++++----
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 14 +++++++-------
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 14 +++++++-------
 net/netfilter/nf_conntrack_l3proto_generic.c   | 12 ++++++------
 4 files changed, 24 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h
index db0825803b16..0378676c3dd8 100644
--- a/include/net/netfilter/nf_conntrack_l3proto.h
+++ b/include/net/netfilter/nf_conntrack_l3proto.h
@@ -28,15 +28,15 @@ struct nf_conntrack_l3proto
 	 * Try to fill in the third arg: nhoff is offset of l3 proto
          * hdr.  Return true if possible.
 	 */
-	int (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int nhoff,
-			    struct nf_conntrack_tuple *tuple);
+	bool (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int nhoff,
+			     struct nf_conntrack_tuple *tuple);
 
 	/*
 	 * Invert the per-proto part of the tuple: ie. turn xmit into reply.
 	 * Some packets can't be inverted: return 0 in that case.
 	 */
-	int (*invert_tuple)(struct nf_conntrack_tuple *inverse,
-			    const struct nf_conntrack_tuple *orig);
+	bool (*invert_tuple)(struct nf_conntrack_tuple *inverse,
+			     const struct nf_conntrack_tuple *orig);
 
 	/* Print out the per-protocol part of the tuple. */
 	int (*print_tuple)(struct seq_file *s,
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
index 41e79613eb0a..cacb9cb27dab 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
@@ -30,29 +30,29 @@ int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb,
 			      enum ip_conntrack_info ctinfo);
 EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook);
 
-static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
-			     struct nf_conntrack_tuple *tuple)
+static bool ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
+			      struct nf_conntrack_tuple *tuple)
 {
 	const __be32 *ap;
 	__be32 _addrs[2];
 	ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr),
 				sizeof(u_int32_t) * 2, _addrs);
 	if (ap == NULL)
-		return 0;
+		return false;
 
 	tuple->src.u3.ip = ap[0];
 	tuple->dst.u3.ip = ap[1];
 
-	return 1;
+	return true;
 }
 
-static int ipv4_invert_tuple(struct nf_conntrack_tuple *tuple,
-			   const struct nf_conntrack_tuple *orig)
+static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple,
+			      const struct nf_conntrack_tuple *orig)
 {
 	tuple->src.u3.ip = orig->dst.u3.ip;
 	tuple->dst.u3.ip = orig->src.u3.ip;
 
-	return 1;
+	return true;
 }
 
 static int ipv4_print_tuple(struct seq_file *s,
diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
index 3717bdf34f6e..85050c072abd 100644
--- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
@@ -27,8 +27,8 @@
 #include <net/netfilter/nf_conntrack_l3proto.h>
 #include <net/netfilter/nf_conntrack_core.h>
 
-static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
-			     struct nf_conntrack_tuple *tuple)
+static bool ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
+			      struct nf_conntrack_tuple *tuple)
 {
 	const u_int32_t *ap;
 	u_int32_t _addrs[8];
@@ -36,21 +36,21 @@ static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
 	ap = skb_header_pointer(skb, nhoff + offsetof(struct ipv6hdr, saddr),
 				sizeof(_addrs), _addrs);
 	if (ap == NULL)
-		return 0;
+		return false;
 
 	memcpy(tuple->src.u3.ip6, ap, sizeof(tuple->src.u3.ip6));
 	memcpy(tuple->dst.u3.ip6, ap + 4, sizeof(tuple->dst.u3.ip6));
 
-	return 1;
+	return true;
 }
 
-static int ipv6_invert_tuple(struct nf_conntrack_tuple *tuple,
-			     const struct nf_conntrack_tuple *orig)
+static bool ipv6_invert_tuple(struct nf_conntrack_tuple *tuple,
+			      const struct nf_conntrack_tuple *orig)
 {
 	memcpy(tuple->src.u3.ip6, orig->dst.u3.ip6, sizeof(tuple->src.u3.ip6));
 	memcpy(tuple->dst.u3.ip6, orig->src.u3.ip6, sizeof(tuple->dst.u3.ip6));
 
-	return 1;
+	return true;
 }
 
 static int ipv6_print_tuple(struct seq_file *s,
diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c
index 8e914e5ffea8..e7eb807fe07d 100644
--- a/net/netfilter/nf_conntrack_l3proto_generic.c
+++ b/net/netfilter/nf_conntrack_l3proto_generic.c
@@ -31,22 +31,22 @@
 #include <net/netfilter/nf_conntrack_core.h>
 #include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
 
-static int generic_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
-				struct nf_conntrack_tuple *tuple)
+static bool generic_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
+				 struct nf_conntrack_tuple *tuple)
 {
 	memset(&tuple->src.u3, 0, sizeof(tuple->src.u3));
 	memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3));
 
-	return 1;
+	return true;
 }
 
-static int generic_invert_tuple(struct nf_conntrack_tuple *tuple,
-			   const struct nf_conntrack_tuple *orig)
+static bool generic_invert_tuple(struct nf_conntrack_tuple *tuple,
+				 const struct nf_conntrack_tuple *orig)
 {
 	memset(&tuple->src.u3, 0, sizeof(tuple->src.u3));
 	memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3));
 
-	return 1;
+	return true;
 }
 
 static int generic_print_tuple(struct seq_file *s,
-- 
cgit v1.2.3


From 09f263cd39751cada63dec2dccc71e67c00bc38c Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:53 +0200
Subject: [NETFILTER]: nf_conntrack: use bool type in struct
 nf_conntrack_l4proto

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack_l4proto.h   | 13 +++----
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   | 25 ++++++-------
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 27 +++++++------
 net/netfilter/nf_conntrack_proto_dccp.c        | 22 +++++------
 net/netfilter/nf_conntrack_proto_generic.c     | 20 +++++-----
 net/netfilter/nf_conntrack_proto_gre.c         | 25 ++++++-------
 net/netfilter/nf_conntrack_proto_sctp.c        | 33 ++++++++--------
 net/netfilter/nf_conntrack_proto_tcp.c         | 52 +++++++++++++-------------
 net/netfilter/nf_conntrack_proto_udp.c         | 18 ++++-----
 net/netfilter/nf_conntrack_proto_udplite.c     | 22 +++++------
 10 files changed, 125 insertions(+), 132 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index efc16eccddb1..723df9d1cc35 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -25,15 +25,14 @@ struct nf_conntrack_l4proto
 
 	/* Try to fill in the third arg: dataoff is offset past network protocol
            hdr.  Return true if possible. */
-	int (*pkt_to_tuple)(const struct sk_buff *skb,
-			    unsigned int dataoff,
-			    struct nf_conntrack_tuple *tuple);
+	bool (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int dataoff,
+			     struct nf_conntrack_tuple *tuple);
 
 	/* Invert the per-proto part of the tuple: ie. turn xmit into reply.
 	 * Some packets can't be inverted: return 0 in that case.
 	 */
-	int (*invert_tuple)(struct nf_conntrack_tuple *inverse,
-			    const struct nf_conntrack_tuple *orig);
+	bool (*invert_tuple)(struct nf_conntrack_tuple *inverse,
+			     const struct nf_conntrack_tuple *orig);
 
 	/* Returns verdict for packet, or -1 for invalid. */
 	int (*packet)(struct nf_conn *ct,
@@ -45,8 +44,8 @@ struct nf_conntrack_l4proto
 
 	/* Called when a new connection for this protocol found;
 	 * returns TRUE if it's OK.  If so, packet() called next. */
-	int (*new)(struct nf_conn *ct, const struct sk_buff *skb,
-		   unsigned int dataoff);
+	bool (*new)(struct nf_conn *ct, const struct sk_buff *skb,
+		    unsigned int dataoff);
 
 	/* Called when a conntrack entry is destroyed */
 	void (*destroy)(struct nf_conn *ct);
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index 6873fddb3529..193a845fe7f8 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -22,22 +22,21 @@
 
 static unsigned long nf_ct_icmp_timeout __read_mostly = 30*HZ;
 
-static int icmp_pkt_to_tuple(const struct sk_buff *skb,
-			     unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple)
+static bool icmp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
+			      struct nf_conntrack_tuple *tuple)
 {
 	const struct icmphdr *hp;
 	struct icmphdr _hdr;
 
 	hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr);
 	if (hp == NULL)
-		return 0;
+		return false;
 
 	tuple->dst.u.icmp.type = hp->type;
 	tuple->src.u.icmp.id = hp->un.echo.id;
 	tuple->dst.u.icmp.code = hp->code;
 
-	return 1;
+	return true;
 }
 
 /* Add 1; spaces filled with 0. */
@@ -52,17 +51,17 @@ static const u_int8_t invmap[] = {
 	[ICMP_ADDRESSREPLY] = ICMP_ADDRESS + 1
 };
 
-static int icmp_invert_tuple(struct nf_conntrack_tuple *tuple,
-			     const struct nf_conntrack_tuple *orig)
+static bool icmp_invert_tuple(struct nf_conntrack_tuple *tuple,
+			      const struct nf_conntrack_tuple *orig)
 {
 	if (orig->dst.u.icmp.type >= sizeof(invmap)
 	    || !invmap[orig->dst.u.icmp.type])
-		return 0;
+		return false;
 
 	tuple->src.u.icmp.id = orig->src.u.icmp.id;
 	tuple->dst.u.icmp.type = invmap[orig->dst.u.icmp.type] - 1;
 	tuple->dst.u.icmp.code = orig->dst.u.icmp.code;
-	return 1;
+	return true;
 }
 
 /* Print out the per-protocol part of the tuple. */
@@ -101,8 +100,8 @@ static int icmp_packet(struct nf_conn *ct,
 }
 
 /* Called when a new connection for this protocol found. */
-static int icmp_new(struct nf_conn *ct,
-		    const struct sk_buff *skb, unsigned int dataoff)
+static bool icmp_new(struct nf_conn *ct, const struct sk_buff *skb,
+		     unsigned int dataoff)
 {
 	static const u_int8_t valid_new[] = {
 		[ICMP_ECHO] = 1,
@@ -117,10 +116,10 @@ static int icmp_new(struct nf_conn *ct,
 		pr_debug("icmp: can't create new conn with type %u\n",
 			 ct->tuplehash[0].tuple.dst.u.icmp.type);
 		NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple);
-		return 0;
+		return false;
 	}
 	atomic_set(&ct->proto.icmp.count, 0);
-	return 1;
+	return true;
 }
 
 /* Returns conntrack if it dealt with ICMP, and filled in skb fields */
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index 0897d0f4c4a2..9ad40e0e17fc 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -28,21 +28,21 @@
 
 static unsigned long nf_ct_icmpv6_timeout __read_mostly = 30*HZ;
 
-static int icmpv6_pkt_to_tuple(const struct sk_buff *skb,
-			       unsigned int dataoff,
-			       struct nf_conntrack_tuple *tuple)
+static bool icmpv6_pkt_to_tuple(const struct sk_buff *skb,
+				unsigned int dataoff,
+				struct nf_conntrack_tuple *tuple)
 {
 	const struct icmp6hdr *hp;
 	struct icmp6hdr _hdr;
 
 	hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr);
 	if (hp == NULL)
-		return 0;
+		return false;
 	tuple->dst.u.icmp.type = hp->icmp6_type;
 	tuple->src.u.icmp.id = hp->icmp6_identifier;
 	tuple->dst.u.icmp.code = hp->icmp6_code;
 
-	return 1;
+	return true;
 }
 
 /* Add 1; spaces filled with 0. */
@@ -53,17 +53,17 @@ static const u_int8_t invmap[] = {
 	[ICMPV6_NI_REPLY - 128]		= ICMPV6_NI_REPLY +1
 };
 
-static int icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple,
-			       const struct nf_conntrack_tuple *orig)
+static bool icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple,
+				const struct nf_conntrack_tuple *orig)
 {
 	int type = orig->dst.u.icmp.type - 128;
 	if (type < 0 || type >= sizeof(invmap) || !invmap[type])
-		return 0;
+		return false;
 
 	tuple->src.u.icmp.id   = orig->src.u.icmp.id;
 	tuple->dst.u.icmp.type = invmap[type] - 1;
 	tuple->dst.u.icmp.code = orig->dst.u.icmp.code;
-	return 1;
+	return true;
 }
 
 /* Print out the per-protocol part of the tuple. */
@@ -102,9 +102,8 @@ static int icmpv6_packet(struct nf_conn *ct,
 }
 
 /* Called when a new connection for this protocol found. */
-static int icmpv6_new(struct nf_conn *ct,
-		      const struct sk_buff *skb,
-		      unsigned int dataoff)
+static bool icmpv6_new(struct nf_conn *ct, const struct sk_buff *skb,
+		       unsigned int dataoff)
 {
 	static const u_int8_t valid_new[] = {
 		[ICMPV6_ECHO_REQUEST - 128] = 1,
@@ -117,10 +116,10 @@ static int icmpv6_new(struct nf_conn *ct,
 		pr_debug("icmpv6: can't create new conn with type %u\n",
 			 type + 128);
 		NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple);
-		return 0;
+		return false;
 	}
 	atomic_set(&ct->proto.icmp.count, 0);
-	return 1;
+	return true;
 }
 
 static int
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index 9376dcd394bd..afb4a1861d2c 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -393,30 +393,30 @@ dccp_state_table[CT_DCCP_ROLE_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] =
 	},
 };
 
-static int dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple)
+static bool dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
+			      struct nf_conntrack_tuple *tuple)
 {
 	struct dccp_hdr _hdr, *dh;
 
 	dh = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr);
 	if (dh == NULL)
-		return 0;
+		return false;
 
 	tuple->src.u.dccp.port = dh->dccph_sport;
 	tuple->dst.u.dccp.port = dh->dccph_dport;
-	return 1;
+	return true;
 }
 
-static int dccp_invert_tuple(struct nf_conntrack_tuple *inv,
-			     const struct nf_conntrack_tuple *tuple)
+static bool dccp_invert_tuple(struct nf_conntrack_tuple *inv,
+			      const struct nf_conntrack_tuple *tuple)
 {
 	inv->src.u.dccp.port = tuple->dst.u.dccp.port;
 	inv->dst.u.dccp.port = tuple->src.u.dccp.port;
-	return 1;
+	return true;
 }
 
-static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
-		    unsigned int dataoff)
+static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
+		     unsigned int dataoff)
 {
 	struct dccp_hdr _dh, *dh;
 	const char *msg;
@@ -442,12 +442,12 @@ static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
 	ct->proto.dccp.role[IP_CT_DIR_ORIGINAL] = CT_DCCP_ROLE_CLIENT;
 	ct->proto.dccp.role[IP_CT_DIR_REPLY] = CT_DCCP_ROLE_SERVER;
 	ct->proto.dccp.state = CT_DCCP_NONE;
-	return 1;
+	return true;
 
 out_invalid:
 	if (LOG_INVALID(IPPROTO_DCCP))
 		nf_log_packet(nf_ct_l3num(ct), 0, skb, NULL, NULL, NULL, msg);
-	return 0;
+	return false;
 }
 
 static u64 dccp_ack_seq(const struct dccp_hdr *dh)
diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c
index 55458915575f..e31b0e7bd0b1 100644
--- a/net/netfilter/nf_conntrack_proto_generic.c
+++ b/net/netfilter/nf_conntrack_proto_generic.c
@@ -14,23 +14,23 @@
 
 static unsigned int nf_ct_generic_timeout __read_mostly = 600*HZ;
 
-static int generic_pkt_to_tuple(const struct sk_buff *skb,
-				unsigned int dataoff,
-				struct nf_conntrack_tuple *tuple)
+static bool generic_pkt_to_tuple(const struct sk_buff *skb,
+				 unsigned int dataoff,
+				 struct nf_conntrack_tuple *tuple)
 {
 	tuple->src.u.all = 0;
 	tuple->dst.u.all = 0;
 
-	return 1;
+	return true;
 }
 
-static int generic_invert_tuple(struct nf_conntrack_tuple *tuple,
-				const struct nf_conntrack_tuple *orig)
+static bool generic_invert_tuple(struct nf_conntrack_tuple *tuple,
+				 const struct nf_conntrack_tuple *orig)
 {
 	tuple->src.u.all = 0;
 	tuple->dst.u.all = 0;
 
-	return 1;
+	return true;
 }
 
 /* Print out the per-protocol part of the tuple. */
@@ -53,10 +53,10 @@ static int packet(struct nf_conn *ct,
 }
 
 /* Called when a new connection for this protocol found. */
-static int new(struct nf_conn *ct, const struct sk_buff *skb,
-	       unsigned int dataoff)
+static bool new(struct nf_conn *ct, const struct sk_buff *skb,
+		unsigned int dataoff)
 {
-	return 1;
+	return true;
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c
index e10024a1b666..7d37a2ea67b2 100644
--- a/net/netfilter/nf_conntrack_proto_gre.c
+++ b/net/netfilter/nf_conntrack_proto_gre.c
@@ -148,18 +148,17 @@ EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_destroy);
 /* PUBLIC CONNTRACK PROTO HELPER FUNCTIONS */
 
 /* invert gre part of tuple */
-static int gre_invert_tuple(struct nf_conntrack_tuple *tuple,
-			    const struct nf_conntrack_tuple *orig)
+static bool gre_invert_tuple(struct nf_conntrack_tuple *tuple,
+			     const struct nf_conntrack_tuple *orig)
 {
 	tuple->dst.u.gre.key = orig->src.u.gre.key;
 	tuple->src.u.gre.key = orig->dst.u.gre.key;
-	return 1;
+	return true;
 }
 
 /* gre hdr info to tuple */
-static int gre_pkt_to_tuple(const struct sk_buff *skb,
-			   unsigned int dataoff,
-			   struct nf_conntrack_tuple *tuple)
+static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
+			     struct nf_conntrack_tuple *tuple)
 {
 	const struct gre_hdr_pptp *pgrehdr;
 	struct gre_hdr_pptp _pgrehdr;
@@ -173,24 +172,24 @@ static int gre_pkt_to_tuple(const struct sk_buff *skb,
 		/* try to behave like "nf_conntrack_proto_generic" */
 		tuple->src.u.all = 0;
 		tuple->dst.u.all = 0;
-		return 1;
+		return true;
 	}
 
 	/* PPTP header is variable length, only need up to the call_id field */
 	pgrehdr = skb_header_pointer(skb, dataoff, 8, &_pgrehdr);
 	if (!pgrehdr)
-		return 1;
+		return true;
 
 	if (ntohs(grehdr->protocol) != GRE_PROTOCOL_PPTP) {
 		pr_debug("GRE_VERSION_PPTP but unknown proto\n");
-		return 0;
+		return false;
 	}
 
 	tuple->dst.u.gre.key = pgrehdr->call_id;
 	srckey = gre_keymap_lookup(tuple);
 	tuple->src.u.gre.key = srckey;
 
-	return 1;
+	return true;
 }
 
 /* print gre part of tuple */
@@ -235,8 +234,8 @@ static int gre_packet(struct nf_conn *ct,
 }
 
 /* Called when a new connection for this protocol found. */
-static int gre_new(struct nf_conn *ct, const struct sk_buff *skb,
-		   unsigned int dataoff)
+static bool gre_new(struct nf_conn *ct, const struct sk_buff *skb,
+		    unsigned int dataoff)
 {
 	pr_debug(": ");
 	NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
@@ -246,7 +245,7 @@ static int gre_new(struct nf_conn *ct, const struct sk_buff *skb,
 	ct->proto.gre.stream_timeout = GRE_STREAM_TIMEOUT;
 	ct->proto.gre.timeout = GRE_TIMEOUT;
 
-	return 1;
+	return true;
 }
 
 /* Called when a conntrack entry has already been removed from the hashes
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index f9a08370dbb3..2d47351f70dc 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -130,28 +130,27 @@ static const u8 sctp_conntracks[2][9][SCTP_CONNTRACK_MAX] = {
 	}
 };
 
-static int sctp_pkt_to_tuple(const struct sk_buff *skb,
-			     unsigned int dataoff,
-			     struct nf_conntrack_tuple *tuple)
+static bool sctp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
+			      struct nf_conntrack_tuple *tuple)
 {
 	sctp_sctphdr_t _hdr, *hp;
 
 	/* Actually only need first 8 bytes. */
 	hp = skb_header_pointer(skb, dataoff, 8, &_hdr);
 	if (hp == NULL)
-		return 0;
+		return false;
 
 	tuple->src.u.sctp.port = hp->source;
 	tuple->dst.u.sctp.port = hp->dest;
-	return 1;
+	return true;
 }
 
-static int sctp_invert_tuple(struct nf_conntrack_tuple *tuple,
-			     const struct nf_conntrack_tuple *orig)
+static bool sctp_invert_tuple(struct nf_conntrack_tuple *tuple,
+			      const struct nf_conntrack_tuple *orig)
 {
 	tuple->src.u.sctp.port = orig->dst.u.sctp.port;
 	tuple->dst.u.sctp.port = orig->src.u.sctp.port;
-	return 1;
+	return true;
 }
 
 /* Print out the per-protocol part of the tuple. */
@@ -390,8 +389,8 @@ out:
 }
 
 /* Called when a new connection for this protocol found. */
-static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
-		    unsigned int dataoff)
+static bool sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
+		     unsigned int dataoff)
 {
 	enum sctp_conntrack new_state;
 	sctp_sctphdr_t _sctph, *sh;
@@ -401,16 +400,16 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 
 	sh = skb_header_pointer(skb, dataoff, sizeof(_sctph), &_sctph);
 	if (sh == NULL)
-		return 0;
+		return false;
 
 	if (do_basic_checks(ct, skb, dataoff, map) != 0)
-		return 0;
+		return false;
 
 	/* If an OOTB packet has any of these chunks discard (Sec 8.4) */
 	if (test_bit(SCTP_CID_ABORT, map) ||
 	    test_bit(SCTP_CID_SHUTDOWN_COMPLETE, map) ||
 	    test_bit(SCTP_CID_COOKIE_ACK, map))
-		return 0;
+		return false;
 
 	new_state = SCTP_CONNTRACK_MAX;
 	for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) {
@@ -422,7 +421,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 		if (new_state == SCTP_CONNTRACK_NONE ||
 		    new_state == SCTP_CONNTRACK_MAX) {
 			pr_debug("nf_conntrack_sctp: invalid new deleting.\n");
-			return 0;
+			return false;
 		}
 
 		/* Copy the vtag into the state info */
@@ -433,7 +432,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 				ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t),
 							sizeof(_inithdr), &_inithdr);
 				if (ih == NULL)
-					return 0;
+					return false;
 
 				pr_debug("Setting vtag %x for new conn\n",
 					 ih->init_tag);
@@ -442,7 +441,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 								ih->init_tag;
 			} else {
 				/* Sec 8.5.1 (A) */
-				return 0;
+				return false;
 			}
 		}
 		/* If it is a shutdown ack OOTB packet, we expect a return
@@ -456,7 +455,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 		ct->proto.sctp.state = new_state;
 	}
 
-	return 1;
+	return true;
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 57831c75fa9f..73a8b32db7be 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -257,9 +257,8 @@ static const u8 tcp_conntracks[2][6][TCP_CONNTRACK_MAX] = {
 	}
 };
 
-static int tcp_pkt_to_tuple(const struct sk_buff *skb,
-			    unsigned int dataoff,
-			    struct nf_conntrack_tuple *tuple)
+static bool tcp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
+			     struct nf_conntrack_tuple *tuple)
 {
 	const struct tcphdr *hp;
 	struct tcphdr _hdr;
@@ -267,20 +266,20 @@ static int tcp_pkt_to_tuple(const struct sk_buff *skb,
 	/* Actually only need first 8 bytes. */
 	hp = skb_header_pointer(skb, dataoff, 8, &_hdr);
 	if (hp == NULL)
-		return 0;
+		return false;
 
 	tuple->src.u.tcp.port = hp->source;
 	tuple->dst.u.tcp.port = hp->dest;
 
-	return 1;
+	return true;
 }
 
-static int tcp_invert_tuple(struct nf_conntrack_tuple *tuple,
-			    const struct nf_conntrack_tuple *orig)
+static bool tcp_invert_tuple(struct nf_conntrack_tuple *tuple,
+			     const struct nf_conntrack_tuple *orig)
 {
 	tuple->src.u.tcp.port = orig->dst.u.tcp.port;
 	tuple->dst.u.tcp.port = orig->src.u.tcp.port;
-	return 1;
+	return true;
 }
 
 /* Print out the per-protocol part of the tuple. */
@@ -478,20 +477,20 @@ static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff,
 	}
 }
 
-static int tcp_in_window(const struct nf_conn *ct,
-			 struct ip_ct_tcp *state,
-			 enum ip_conntrack_dir dir,
-			 unsigned int index,
-			 const struct sk_buff *skb,
-			 unsigned int dataoff,
-			 const struct tcphdr *tcph,
-			 int pf)
+static bool tcp_in_window(const struct nf_conn *ct,
+			  struct ip_ct_tcp *state,
+			  enum ip_conntrack_dir dir,
+			  unsigned int index,
+			  const struct sk_buff *skb,
+			  unsigned int dataoff,
+			  const struct tcphdr *tcph,
+			  int pf)
 {
 	struct ip_ct_tcp_state *sender = &state->seen[dir];
 	struct ip_ct_tcp_state *receiver = &state->seen[!dir];
 	const struct nf_conntrack_tuple *tuple = &ct->tuplehash[dir].tuple;
 	__u32 seq, ack, sack, end, win, swin;
-	int res;
+	bool res;
 
 	/*
 	 * Get the required data from the packet.
@@ -657,12 +656,12 @@ static int tcp_in_window(const struct nf_conn *ct,
 				state->retrans = 0;
 			}
 		}
-		res = 1;
+		res = true;
 	} else {
-		res = 0;
+		res = false;
 		if (sender->flags & IP_CT_TCP_FLAG_BE_LIBERAL ||
 		    nf_ct_tcp_be_liberal)
-			res = 1;
+			res = true;
 		if (!res && LOG_INVALID(IPPROTO_TCP))
 			nf_log_packet(pf, 0, skb, NULL, NULL, NULL,
 			"nf_ct_tcp: %s ",
@@ -676,7 +675,7 @@ static int tcp_in_window(const struct nf_conn *ct,
 			: "SEQ is over the upper bound (over the window of the receiver)");
 	}
 
-	pr_debug("tcp_in_window: res=%i sender end=%u maxend=%u maxwin=%u "
+	pr_debug("tcp_in_window: res=%u sender end=%u maxend=%u maxwin=%u "
 		 "receiver end=%u maxend=%u maxwin=%u\n",
 		 res, sender->td_end, sender->td_maxend, sender->td_maxwin,
 		 receiver->td_end, receiver->td_maxend, receiver->td_maxwin);
@@ -982,9 +981,8 @@ static int tcp_packet(struct nf_conn *ct,
 }
 
 /* Called when a new connection for this protocol found. */
-static int tcp_new(struct nf_conn *ct,
-		   const struct sk_buff *skb,
-		   unsigned int dataoff)
+static bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb,
+		    unsigned int dataoff)
 {
 	enum tcp_conntrack new_state;
 	const struct tcphdr *th;
@@ -1003,7 +1001,7 @@ static int tcp_new(struct nf_conn *ct,
 	/* Invalid: delete conntrack */
 	if (new_state >= TCP_CONNTRACK_MAX) {
 		pr_debug("nf_ct_tcp: invalid new deleting.\n");
-		return 0;
+		return false;
 	}
 
 	if (new_state == TCP_CONNTRACK_SYN_SENT) {
@@ -1021,7 +1019,7 @@ static int tcp_new(struct nf_conn *ct,
 		ct->proto.tcp.seen[1].flags = 0;
 	} else if (nf_ct_tcp_loose == 0) {
 		/* Don't try to pick up connections. */
-		return 0;
+		return false;
 	} else {
 		/*
 		 * We are in the middle of a connection,
@@ -1061,7 +1059,7 @@ static int tcp_new(struct nf_conn *ct,
 		 sender->td_scale,
 		 receiver->td_end, receiver->td_maxend, receiver->td_maxwin,
 		 receiver->td_scale);
-	return 1;
+	return true;
 }
 
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c
index b8a35cc06416..8b21762e65de 100644
--- a/net/netfilter/nf_conntrack_proto_udp.c
+++ b/net/netfilter/nf_conntrack_proto_udp.c
@@ -26,7 +26,7 @@
 static unsigned int nf_ct_udp_timeout __read_mostly = 30*HZ;
 static unsigned int nf_ct_udp_timeout_stream __read_mostly = 180*HZ;
 
-static int udp_pkt_to_tuple(const struct sk_buff *skb,
+static bool udp_pkt_to_tuple(const struct sk_buff *skb,
 			     unsigned int dataoff,
 			     struct nf_conntrack_tuple *tuple)
 {
@@ -36,20 +36,20 @@ static int udp_pkt_to_tuple(const struct sk_buff *skb,
 	/* Actually only need first 8 bytes. */
 	hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr);
 	if (hp == NULL)
-		return 0;
+		return false;
 
 	tuple->src.u.udp.port = hp->source;
 	tuple->dst.u.udp.port = hp->dest;
 
-	return 1;
+	return true;
 }
 
-static int udp_invert_tuple(struct nf_conntrack_tuple *tuple,
-			    const struct nf_conntrack_tuple *orig)
+static bool udp_invert_tuple(struct nf_conntrack_tuple *tuple,
+			     const struct nf_conntrack_tuple *orig)
 {
 	tuple->src.u.udp.port = orig->dst.u.udp.port;
 	tuple->dst.u.udp.port = orig->src.u.udp.port;
-	return 1;
+	return true;
 }
 
 /* Print out the per-protocol part of the tuple. */
@@ -83,10 +83,10 @@ static int udp_packet(struct nf_conn *ct,
 }
 
 /* Called when a new connection for this protocol found. */
-static int udp_new(struct nf_conn *ct, const struct sk_buff *skb,
-		   unsigned int dataoff)
+static bool udp_new(struct nf_conn *ct, const struct sk_buff *skb,
+		    unsigned int dataoff)
 {
-	return 1;
+	return true;
 }
 
 static int udp_error(struct sk_buff *skb, unsigned int dataoff,
diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c
index c3eaee6afffd..1fa62f3c24f1 100644
--- a/net/netfilter/nf_conntrack_proto_udplite.c
+++ b/net/netfilter/nf_conntrack_proto_udplite.c
@@ -27,28 +27,28 @@
 static unsigned int nf_ct_udplite_timeout __read_mostly = 30*HZ;
 static unsigned int nf_ct_udplite_timeout_stream __read_mostly = 180*HZ;
 
-static int udplite_pkt_to_tuple(const struct sk_buff *skb,
-				unsigned int dataoff,
-				struct nf_conntrack_tuple *tuple)
+static bool udplite_pkt_to_tuple(const struct sk_buff *skb,
+				 unsigned int dataoff,
+				 struct nf_conntrack_tuple *tuple)
 {
 	const struct udphdr *hp;
 	struct udphdr _hdr;
 
 	hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr);
 	if (hp == NULL)
-		return 0;
+		return false;
 
 	tuple->src.u.udp.port = hp->source;
 	tuple->dst.u.udp.port = hp->dest;
-	return 1;
+	return true;
 }
 
-static int udplite_invert_tuple(struct nf_conntrack_tuple *tuple,
-				const struct nf_conntrack_tuple *orig)
+static bool udplite_invert_tuple(struct nf_conntrack_tuple *tuple,
+				 const struct nf_conntrack_tuple *orig)
 {
 	tuple->src.u.udp.port = orig->dst.u.udp.port;
 	tuple->dst.u.udp.port = orig->src.u.udp.port;
-	return 1;
+	return true;
 }
 
 /* Print out the per-protocol part of the tuple. */
@@ -83,10 +83,10 @@ static int udplite_packet(struct nf_conn *ct,
 }
 
 /* Called when a new connection for this protocol found. */
-static int udplite_new(struct nf_conn *ct, const struct sk_buff *skb,
-		       unsigned int dataoff)
+static bool udplite_new(struct nf_conn *ct, const struct sk_buff *skb,
+			unsigned int dataoff)
 {
-	return 1;
+	return true;
 }
 
 static int udplite_error(struct sk_buff *skb, unsigned int dataoff,
-- 
cgit v1.2.3


From 5f2b4c9006fc667c4614f0b079efab3721f68316 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:53 +0200
Subject: [NETFILTER]: nf_conntrack: use bool type in struct
 nf_conntrack_tuple.h

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack.h       | 11 ++++-----
 include/net/netfilter/nf_conntrack_core.h  |  4 +--
 include/net/netfilter/nf_conntrack_tuple.h | 39 ++++++++++++++++--------------
 net/netfilter/nf_conntrack_core.c          | 22 ++++++++---------
 4 files changed, 38 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 26e6a6e2b5a2..2dbd6c015b94 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -194,12 +194,11 @@ extern void nf_conntrack_hash_insert(struct nf_conn *ct);
 
 extern void nf_conntrack_flush(void);
 
-extern int nf_ct_get_tuplepr(const struct sk_buff *skb,
-			     unsigned int nhoff,
-			     u_int16_t l3num,
-			     struct nf_conntrack_tuple *tuple);
-extern int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse,
-				const struct nf_conntrack_tuple *orig);
+extern bool nf_ct_get_tuplepr(const struct sk_buff *skb,
+			      unsigned int nhoff, u_int16_t l3num,
+			      struct nf_conntrack_tuple *tuple);
+extern bool nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse,
+				 const struct nf_conntrack_tuple *orig);
 
 extern void __nf_ct_refresh_acct(struct nf_conn *ct,
 				 enum ip_conntrack_info ctinfo,
diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h
index 9ee26469c759..a81771210934 100644
--- a/include/net/netfilter/nf_conntrack_core.h
+++ b/include/net/netfilter/nf_conntrack_core.h
@@ -30,7 +30,7 @@ extern void nf_conntrack_cleanup(void);
 extern int nf_conntrack_proto_init(void);
 extern void nf_conntrack_proto_fini(void);
 
-extern int
+extern bool
 nf_ct_get_tuple(const struct sk_buff *skb,
 		unsigned int nhoff,
 		unsigned int dataoff,
@@ -40,7 +40,7 @@ nf_ct_get_tuple(const struct sk_buff *skb,
 		const struct nf_conntrack_l3proto *l3proto,
 		const struct nf_conntrack_l4proto *l4proto);
 
-extern int
+extern bool
 nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse,
 		   const struct nf_conntrack_tuple *orig,
 		   const struct nf_conntrack_l3proto *l3proto,
diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h
index bdeec3461384..2722b13ecd6c 100644
--- a/include/net/netfilter/nf_conntrack_tuple.h
+++ b/include/net/netfilter/nf_conntrack_tuple.h
@@ -166,61 +166,64 @@ struct nf_conntrack_tuple_hash
 
 #endif /* __KERNEL__ */
 
-static inline int __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1,
-					  const struct nf_conntrack_tuple *t2)
+static inline bool __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1,
+					   const struct nf_conntrack_tuple *t2)
 { 
 	return (nf_inet_addr_cmp(&t1->src.u3, &t2->src.u3) &&
 		t1->src.u.all == t2->src.u.all &&
 		t1->src.l3num == t2->src.l3num);
 }
 
-static inline int __nf_ct_tuple_dst_equal(const struct nf_conntrack_tuple *t1,
-					  const struct nf_conntrack_tuple *t2)
+static inline bool __nf_ct_tuple_dst_equal(const struct nf_conntrack_tuple *t1,
+					   const struct nf_conntrack_tuple *t2)
 {
 	return (nf_inet_addr_cmp(&t1->dst.u3, &t2->dst.u3) &&
 		t1->dst.u.all == t2->dst.u.all &&
 		t1->dst.protonum == t2->dst.protonum);
 }
 
-static inline int nf_ct_tuple_equal(const struct nf_conntrack_tuple *t1,
-				    const struct nf_conntrack_tuple *t2)
+static inline bool nf_ct_tuple_equal(const struct nf_conntrack_tuple *t1,
+				     const struct nf_conntrack_tuple *t2)
 {
 	return __nf_ct_tuple_src_equal(t1, t2) &&
 	       __nf_ct_tuple_dst_equal(t1, t2);
 }
 
-static inline int nf_ct_tuple_mask_equal(const struct nf_conntrack_tuple_mask *m1,
-					 const struct nf_conntrack_tuple_mask *m2)
+static inline bool
+nf_ct_tuple_mask_equal(const struct nf_conntrack_tuple_mask *m1,
+		       const struct nf_conntrack_tuple_mask *m2)
 {
 	return (nf_inet_addr_cmp(&m1->src.u3, &m2->src.u3) &&
 		m1->src.u.all == m2->src.u.all);
 }
 
-static inline int nf_ct_tuple_src_mask_cmp(const struct nf_conntrack_tuple *t1,
-					   const struct nf_conntrack_tuple *t2,
-					   const struct nf_conntrack_tuple_mask *mask)
+static inline bool
+nf_ct_tuple_src_mask_cmp(const struct nf_conntrack_tuple *t1,
+			 const struct nf_conntrack_tuple *t2,
+			 const struct nf_conntrack_tuple_mask *mask)
 {
 	int count;
 
 	for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++) {
 		if ((t1->src.u3.all[count] ^ t2->src.u3.all[count]) &
 		    mask->src.u3.all[count])
-			return 0;
+			return false;
 	}
 
 	if ((t1->src.u.all ^ t2->src.u.all) & mask->src.u.all)
-		return 0;
+		return false;
 
 	if (t1->src.l3num != t2->src.l3num ||
 	    t1->dst.protonum != t2->dst.protonum)
-		return 0;
+		return false;
 
-	return 1;
+	return true;
 }
 
-static inline int nf_ct_tuple_mask_cmp(const struct nf_conntrack_tuple *t,
-				       const struct nf_conntrack_tuple *tuple,
-				       const struct nf_conntrack_tuple_mask *mask)
+static inline bool
+nf_ct_tuple_mask_cmp(const struct nf_conntrack_tuple *t,
+		     const struct nf_conntrack_tuple *tuple,
+		     const struct nf_conntrack_tuple_mask *mask)
 {
 	return nf_ct_tuple_src_mask_cmp(t, tuple, mask) &&
 	       __nf_ct_tuple_dst_equal(t, tuple);
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 21ab0c3846ac..a3fe9db412d3 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -94,7 +94,7 @@ static inline u_int32_t hash_conntrack(const struct nf_conntrack_tuple *tuple)
 				nf_conntrack_hash_rnd);
 }
 
-int
+bool
 nf_ct_get_tuple(const struct sk_buff *skb,
 		unsigned int nhoff,
 		unsigned int dataoff,
@@ -108,7 +108,7 @@ nf_ct_get_tuple(const struct sk_buff *skb,
 
 	tuple->src.l3num = l3num;
 	if (l3proto->pkt_to_tuple(skb, nhoff, tuple) == 0)
-		return 0;
+		return false;
 
 	tuple->dst.protonum = protonum;
 	tuple->dst.dir = IP_CT_DIR_ORIGINAL;
@@ -117,10 +117,8 @@ nf_ct_get_tuple(const struct sk_buff *skb,
 }
 EXPORT_SYMBOL_GPL(nf_ct_get_tuple);
 
-int nf_ct_get_tuplepr(const struct sk_buff *skb,
-		      unsigned int nhoff,
-		      u_int16_t l3num,
-		      struct nf_conntrack_tuple *tuple)
+bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff,
+		       u_int16_t l3num, struct nf_conntrack_tuple *tuple)
 {
 	struct nf_conntrack_l3proto *l3proto;
 	struct nf_conntrack_l4proto *l4proto;
@@ -134,7 +132,7 @@ int nf_ct_get_tuplepr(const struct sk_buff *skb,
 	ret = l3proto->get_l4proto(skb, nhoff, &protoff, &protonum);
 	if (ret != NF_ACCEPT) {
 		rcu_read_unlock();
-		return 0;
+		return false;
 	}
 
 	l4proto = __nf_ct_l4proto_find(l3num, protonum);
@@ -147,7 +145,7 @@ int nf_ct_get_tuplepr(const struct sk_buff *skb,
 }
 EXPORT_SYMBOL_GPL(nf_ct_get_tuplepr);
 
-int
+bool
 nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse,
 		   const struct nf_conntrack_tuple *orig,
 		   const struct nf_conntrack_l3proto *l3proto,
@@ -157,7 +155,7 @@ nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse,
 
 	inverse->src.l3num = orig->src.l3num;
 	if (l3proto->invert_tuple(inverse, orig) == 0)
-		return 0;
+		return false;
 
 	inverse->dst.dir = !orig->dst.dir;
 
@@ -738,10 +736,10 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(nf_conntrack_in);
 
-int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse,
-			 const struct nf_conntrack_tuple *orig)
+bool nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse,
+			  const struct nf_conntrack_tuple *orig)
 {
-	int ret;
+	bool ret;
 
 	rcu_read_lock();
 	ret = nf_ct_invert_tuple(inverse, orig,
-- 
cgit v1.2.3


From f2ea825f483d5d78754ae813b6db63f8b74e9343 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:53 +0200
Subject: [NETFILTER]: nf_nat: use bool type in nf_nat_proto

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_nat_protocol.h   | 42 +++++++++++++++----------------
 net/ipv4/netfilter/nf_nat_core.c          |  8 +++---
 net/ipv4/netfilter/nf_nat_proto_common.c  | 24 +++++++++---------
 net/ipv4/netfilter/nf_nat_proto_dccp.c    | 10 ++++----
 net/ipv4/netfilter/nf_nat_proto_gre.c     | 18 ++++++-------
 net/ipv4/netfilter/nf_nat_proto_icmp.c    | 14 +++++------
 net/ipv4/netfilter/nf_nat_proto_sctp.c    |  8 +++---
 net/ipv4/netfilter/nf_nat_proto_tcp.c     | 10 ++++----
 net/ipv4/netfilter/nf_nat_proto_udp.c     |  8 +++---
 net/ipv4/netfilter/nf_nat_proto_udplite.c |  8 +++---
 net/ipv4/netfilter/nf_nat_proto_unknown.c | 24 +++++++++---------
 11 files changed, 87 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h
index fba94a2028d5..f3662c4394ef 100644
--- a/include/net/netfilter/nf_nat_protocol.h
+++ b/include/net/netfilter/nf_nat_protocol.h
@@ -15,25 +15,25 @@ struct nf_nat_protocol
 
 	/* Translate a packet to the target according to manip type.
 	   Return true if succeeded. */
-	int (*manip_pkt)(struct sk_buff *skb,
-			 unsigned int iphdroff,
-			 const struct nf_conntrack_tuple *tuple,
-			 enum nf_nat_manip_type maniptype);
+	bool (*manip_pkt)(struct sk_buff *skb,
+			  unsigned int iphdroff,
+			  const struct nf_conntrack_tuple *tuple,
+			  enum nf_nat_manip_type maniptype);
 
 	/* Is the manipable part of the tuple between min and max incl? */
-	int (*in_range)(const struct nf_conntrack_tuple *tuple,
-			enum nf_nat_manip_type maniptype,
-			const union nf_conntrack_man_proto *min,
-			const union nf_conntrack_man_proto *max);
+	bool (*in_range)(const struct nf_conntrack_tuple *tuple,
+			 enum nf_nat_manip_type maniptype,
+			 const union nf_conntrack_man_proto *min,
+			 const union nf_conntrack_man_proto *max);
 
 	/* Alter the per-proto part of the tuple (depending on
 	   maniptype), to give a unique tuple in the given range if
 	   possible; return false if not.  Per-protocol part of tuple
 	   is initialized to the incoming packet. */
-	int (*unique_tuple)(struct nf_conntrack_tuple *tuple,
-			    const struct nf_nat_range *range,
-			    enum nf_nat_manip_type maniptype,
-			    const struct nf_conn *ct);
+	bool (*unique_tuple)(struct nf_conntrack_tuple *tuple,
+			     const struct nf_nat_range *range,
+			     enum nf_nat_manip_type maniptype,
+			     const struct nf_conn *ct);
 
 	int (*range_to_nlattr)(struct sk_buff *skb,
 			       const struct nf_nat_range *range);
@@ -59,16 +59,16 @@ extern int init_protocols(void) __init;
 extern void cleanup_protocols(void);
 extern const struct nf_nat_protocol *find_nat_proto(u_int16_t protonum);
 
-extern int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple,
-				 enum nf_nat_manip_type maniptype,
-				 const union nf_conntrack_man_proto *min,
-				 const union nf_conntrack_man_proto *max);
+extern bool nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple,
+				  enum nf_nat_manip_type maniptype,
+				  const union nf_conntrack_man_proto *min,
+				  const union nf_conntrack_man_proto *max);
 
-extern int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
-				     const struct nf_nat_range *range,
-				     enum nf_nat_manip_type maniptype,
-				     const struct nf_conn *ct,
-				     u_int16_t *rover);
+extern bool nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
+				      const struct nf_nat_range *range,
+				      enum nf_nat_manip_type maniptype,
+				      const struct nf_conn *ct,
+				      u_int16_t *rover);
 
 extern int nf_nat_proto_range_to_nlattr(struct sk_buff *skb,
 					const struct nf_nat_range *range);
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index 25c3efe4207e..07a2fbc59622 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -349,7 +349,7 @@ nf_nat_setup_info(struct nf_conn *ct,
 EXPORT_SYMBOL(nf_nat_setup_info);
 
 /* Returns true if succeeded. */
-static int
+static bool
 manip_pkt(u_int16_t proto,
 	  struct sk_buff *skb,
 	  unsigned int iphdroff,
@@ -360,7 +360,7 @@ manip_pkt(u_int16_t proto,
 	const struct nf_nat_protocol *p;
 
 	if (!skb_make_writable(skb, iphdroff + sizeof(*iph)))
-		return 0;
+		return false;
 
 	iph = (void *)skb->data + iphdroff;
 
@@ -369,7 +369,7 @@ manip_pkt(u_int16_t proto,
 	/* rcu_read_lock()ed by nf_hook_slow */
 	p = __nf_nat_proto_find(proto);
 	if (!p->manip_pkt(skb, iphdroff, target, maniptype))
-		return 0;
+		return false;
 
 	iph = (void *)skb->data + iphdroff;
 
@@ -380,7 +380,7 @@ manip_pkt(u_int16_t proto,
 		csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip);
 		iph->daddr = target->dst.u3.ip;
 	}
-	return 1;
+	return true;
 }
 
 /* Do packet manipulations according to nf_nat_setup_info. */
diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c
index 4904b86265e1..91537f11273f 100644
--- a/net/ipv4/netfilter/nf_nat_proto_common.c
+++ b/net/ipv4/netfilter/nf_nat_proto_common.c
@@ -17,10 +17,10 @@
 #include <net/netfilter/nf_nat_rule.h>
 #include <net/netfilter/nf_nat_protocol.h>
 
-int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple,
-			  enum nf_nat_manip_type maniptype,
-			  const union nf_conntrack_man_proto *min,
-			  const union nf_conntrack_man_proto *max)
+bool nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple,
+			   enum nf_nat_manip_type maniptype,
+			   const union nf_conntrack_man_proto *min,
+			   const union nf_conntrack_man_proto *max)
 {
 	__be16 port;
 
@@ -34,11 +34,11 @@ int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple,
 }
 EXPORT_SYMBOL_GPL(nf_nat_proto_in_range);
 
-int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
-			      const struct nf_nat_range *range,
-			      enum nf_nat_manip_type maniptype,
-			      const struct nf_conn *ct,
-			      u_int16_t *rover)
+bool nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
+			       const struct nf_nat_range *range,
+			       enum nf_nat_manip_type maniptype,
+			       const struct nf_conn *ct,
+			       u_int16_t *rover)
 {
 	unsigned int range_size, min, i;
 	__be16 *portptr;
@@ -53,7 +53,7 @@ int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
 	if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) {
 		/* If it's dst rewrite, can't change port */
 		if (maniptype == IP_NAT_MANIP_DST)
-			return 0;
+			return false;
 
 		if (ntohs(*portptr) < 1024) {
 			/* Loose convention: >> 512 is credential passing */
@@ -83,9 +83,9 @@ int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple,
 			continue;
 		if (!(range->flags & IP_NAT_RANGE_PROTO_RANDOM))
 			*rover = off;
-		return 1;
+		return true;
 	}
-	return 0;
+	return false;
 }
 EXPORT_SYMBOL_GPL(nf_nat_proto_unique_tuple);
 
diff --git a/net/ipv4/netfilter/nf_nat_proto_dccp.c b/net/ipv4/netfilter/nf_nat_proto_dccp.c
index 12b51b38442e..f78eb26e9a20 100644
--- a/net/ipv4/netfilter/nf_nat_proto_dccp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_dccp.c
@@ -22,7 +22,7 @@
 
 static u_int16_t dccp_port_rover;
 
-static int
+static bool
 dccp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		  const struct nf_nat_range *range,
 		  enum nf_nat_manip_type maniptype,
@@ -32,7 +32,7 @@ dccp_unique_tuple(struct nf_conntrack_tuple *tuple,
 					 &dccp_port_rover);
 }
 
-static int
+static bool
 dccp_manip_pkt(struct sk_buff *skb,
 	       unsigned int iphdroff,
 	       const struct nf_conntrack_tuple *tuple,
@@ -49,7 +49,7 @@ dccp_manip_pkt(struct sk_buff *skb,
 		hdrsize = sizeof(struct dccp_hdr);
 
 	if (!skb_make_writable(skb, hdroff + hdrsize))
-		return 0;
+		return false;
 
 	iph = (struct iphdr *)(skb->data + iphdroff);
 	hdr = (struct dccp_hdr *)(skb->data + hdroff);
@@ -70,12 +70,12 @@ dccp_manip_pkt(struct sk_buff *skb,
 	*portptr = newport;
 
 	if (hdrsize < sizeof(*hdr))
-		return 1;
+		return true;
 
 	inet_proto_csum_replace4(&hdr->dccph_checksum, skb, oldip, newip, 1);
 	inet_proto_csum_replace2(&hdr->dccph_checksum, skb, oldport, newport,
 				 0);
-	return 1;
+	return true;
 }
 
 static const struct nf_nat_protocol nf_nat_protocol_dccp = {
diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c
index 84bb7854137a..4c4af5a6d6c8 100644
--- a/net/ipv4/netfilter/nf_nat_proto_gre.c
+++ b/net/ipv4/netfilter/nf_nat_proto_gre.c
@@ -37,7 +37,7 @@ MODULE_AUTHOR("Harald Welte <laforge@gnumonks.org>");
 MODULE_DESCRIPTION("Netfilter NAT protocol helper module for GRE");
 
 /* generate unique tuple ... */
-static int
+static bool
 gre_unique_tuple(struct nf_conntrack_tuple *tuple,
 		 const struct nf_nat_range *range,
 		 enum nf_nat_manip_type maniptype,
@@ -50,7 +50,7 @@ gre_unique_tuple(struct nf_conntrack_tuple *tuple,
 	/* If there is no master conntrack we are not PPTP,
 	   do not change tuples */
 	if (!ct->master)
-		return 0;
+		return false;
 
 	if (maniptype == IP_NAT_MANIP_SRC)
 		keyptr = &tuple->src.u.gre.key;
@@ -71,15 +71,15 @@ gre_unique_tuple(struct nf_conntrack_tuple *tuple,
 	for (i = 0; i < range_size; i++, key++) {
 		*keyptr = htons(min + key % range_size);
 		if (!nf_nat_used_tuple(tuple, ct))
-			return 1;
+			return true;
 	}
 
 	pr_debug("%p: no NAT mapping\n", ct);
-	return 0;
+	return false;
 }
 
 /* manipulate a GRE packet according to maniptype */
-static int
+static bool
 gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff,
 	      const struct nf_conntrack_tuple *tuple,
 	      enum nf_nat_manip_type maniptype)
@@ -92,7 +92,7 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff,
 	/* pgreh includes two optional 32bit fields which are not required
 	 * to be there.  That's where the magic '8' comes from */
 	if (!skb_make_writable(skb, hdroff + sizeof(*pgreh) - 8))
-		return 0;
+		return false;
 
 	greh = (void *)skb->data + hdroff;
 	pgreh = (struct gre_hdr_pptp *)greh;
@@ -100,7 +100,7 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff,
 	/* we only have destination manip of a packet, since 'source key'
 	 * is not present in the packet itself */
 	if (maniptype != IP_NAT_MANIP_DST)
-		return 1;
+		return true;
 	switch (greh->version) {
 	case GRE_VERSION_1701:
 		/* We do not currently NAT any GREv0 packets.
@@ -112,9 +112,9 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff,
 		break;
 	default:
 		pr_debug("can't nat unknown GRE version\n");
-		return 0;
+		return false;
 	}
-	return 1;
+	return true;
 }
 
 static const struct nf_nat_protocol gre = {
diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c
index ab3a0ec2a2d1..19a8b0b07d8e 100644
--- a/net/ipv4/netfilter/nf_nat_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c
@@ -17,7 +17,7 @@
 #include <net/netfilter/nf_nat_rule.h>
 #include <net/netfilter/nf_nat_protocol.h>
 
-static int
+static bool
 icmp_in_range(const struct nf_conntrack_tuple *tuple,
 	      enum nf_nat_manip_type maniptype,
 	      const union nf_conntrack_man_proto *min,
@@ -27,7 +27,7 @@ icmp_in_range(const struct nf_conntrack_tuple *tuple,
 	       ntohs(tuple->src.u.icmp.id) <= ntohs(max->icmp.id);
 }
 
-static int
+static bool
 icmp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		  const struct nf_nat_range *range,
 		  enum nf_nat_manip_type maniptype,
@@ -46,12 +46,12 @@ icmp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		tuple->src.u.icmp.id = htons(ntohs(range->min.icmp.id) +
 					     (id % range_size));
 		if (!nf_nat_used_tuple(tuple, ct))
-			return 1;
+			return true;
 	}
-	return 0;
+	return false;
 }
 
-static int
+static bool
 icmp_manip_pkt(struct sk_buff *skb,
 	       unsigned int iphdroff,
 	       const struct nf_conntrack_tuple *tuple,
@@ -62,13 +62,13 @@ icmp_manip_pkt(struct sk_buff *skb,
 	unsigned int hdroff = iphdroff + iph->ihl*4;
 
 	if (!skb_make_writable(skb, hdroff + sizeof(*hdr)))
-		return 0;
+		return false;
 
 	hdr = (struct icmphdr *)(skb->data + hdroff);
 	inet_proto_csum_replace2(&hdr->checksum, skb,
 				 hdr->un.echo.id, tuple->src.u.icmp.id, 0);
 	hdr->un.echo.id = tuple->src.u.icmp.id;
-	return 1;
+	return true;
 }
 
 const struct nf_nat_protocol nf_nat_protocol_icmp = {
diff --git a/net/ipv4/netfilter/nf_nat_proto_sctp.c b/net/ipv4/netfilter/nf_nat_proto_sctp.c
index 3d3faa9d5f6d..82e4c0e286b8 100644
--- a/net/ipv4/netfilter/nf_nat_proto_sctp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_sctp.c
@@ -16,7 +16,7 @@
 
 static u_int16_t nf_sctp_port_rover;
 
-static int
+static bool
 sctp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		  const struct nf_nat_range *range,
 		  enum nf_nat_manip_type maniptype,
@@ -26,7 +26,7 @@ sctp_unique_tuple(struct nf_conntrack_tuple *tuple,
 					 &nf_sctp_port_rover);
 }
 
-static int
+static bool
 sctp_manip_pkt(struct sk_buff *skb,
 	       unsigned int iphdroff,
 	       const struct nf_conntrack_tuple *tuple,
@@ -39,7 +39,7 @@ sctp_manip_pkt(struct sk_buff *skb,
 	u32 crc32;
 
 	if (!skb_make_writable(skb, hdroff + sizeof(*hdr)))
-		return 0;
+		return false;
 
 	iph = (struct iphdr *)(skb->data + iphdroff);
 	hdr = (struct sctphdr *)(skb->data + hdroff);
@@ -63,7 +63,7 @@ sctp_manip_pkt(struct sk_buff *skb,
 	crc32 = sctp_end_cksum(crc32);
 	hdr->checksum = htonl(crc32);
 
-	return 1;
+	return true;
 }
 
 static const struct nf_nat_protocol nf_nat_protocol_sctp = {
diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c
index 5d4c8a0e89c0..399e2cfa263b 100644
--- a/net/ipv4/netfilter/nf_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c
@@ -20,7 +20,7 @@
 
 static u_int16_t tcp_port_rover;
 
-static int
+static bool
 tcp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		 const struct nf_nat_range *range,
 		 enum nf_nat_manip_type maniptype,
@@ -30,7 +30,7 @@ tcp_unique_tuple(struct nf_conntrack_tuple *tuple,
 					 &tcp_port_rover);
 }
 
-static int
+static bool
 tcp_manip_pkt(struct sk_buff *skb,
 	      unsigned int iphdroff,
 	      const struct nf_conntrack_tuple *tuple,
@@ -50,7 +50,7 @@ tcp_manip_pkt(struct sk_buff *skb,
 		hdrsize = sizeof(struct tcphdr);
 
 	if (!skb_make_writable(skb, hdroff + hdrsize))
-		return 0;
+		return false;
 
 	iph = (struct iphdr *)(skb->data + iphdroff);
 	hdr = (struct tcphdr *)(skb->data + hdroff);
@@ -73,11 +73,11 @@ tcp_manip_pkt(struct sk_buff *skb,
 	*portptr = newport;
 
 	if (hdrsize < sizeof(*hdr))
-		return 1;
+		return true;
 
 	inet_proto_csum_replace4(&hdr->check, skb, oldip, newip, 1);
 	inet_proto_csum_replace2(&hdr->check, skb, oldport, newport, 0);
-	return 1;
+	return true;
 }
 
 const struct nf_nat_protocol nf_nat_protocol_tcp = {
diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c
index 74a7e7b63465..9e61c79492e4 100644
--- a/net/ipv4/netfilter/nf_nat_proto_udp.c
+++ b/net/ipv4/netfilter/nf_nat_proto_udp.c
@@ -19,7 +19,7 @@
 
 static u_int16_t udp_port_rover;
 
-static int
+static bool
 udp_unique_tuple(struct nf_conntrack_tuple *tuple,
 		 const struct nf_nat_range *range,
 		 enum nf_nat_manip_type maniptype,
@@ -29,7 +29,7 @@ udp_unique_tuple(struct nf_conntrack_tuple *tuple,
 					 &udp_port_rover);
 }
 
-static int
+static bool
 udp_manip_pkt(struct sk_buff *skb,
 	      unsigned int iphdroff,
 	      const struct nf_conntrack_tuple *tuple,
@@ -42,7 +42,7 @@ udp_manip_pkt(struct sk_buff *skb,
 	__be16 *portptr, newport;
 
 	if (!skb_make_writable(skb, hdroff + sizeof(*hdr)))
-		return 0;
+		return false;
 
 	iph = (struct iphdr *)(skb->data + iphdroff);
 	hdr = (struct udphdr *)(skb->data + hdroff);
@@ -68,7 +68,7 @@ udp_manip_pkt(struct sk_buff *skb,
 			hdr->check = CSUM_MANGLED_0;
 	}
 	*portptr = newport;
-	return 1;
+	return true;
 }
 
 const struct nf_nat_protocol nf_nat_protocol_udp = {
diff --git a/net/ipv4/netfilter/nf_nat_proto_udplite.c b/net/ipv4/netfilter/nf_nat_proto_udplite.c
index b29346d0e7ab..440a229bbd87 100644
--- a/net/ipv4/netfilter/nf_nat_proto_udplite.c
+++ b/net/ipv4/netfilter/nf_nat_proto_udplite.c
@@ -18,7 +18,7 @@
 
 static u_int16_t udplite_port_rover;
 
-static int
+static bool
 udplite_unique_tuple(struct nf_conntrack_tuple *tuple,
 		     const struct nf_nat_range *range,
 		     enum nf_nat_manip_type maniptype,
@@ -28,7 +28,7 @@ udplite_unique_tuple(struct nf_conntrack_tuple *tuple,
 					 &udplite_port_rover);
 }
 
-static int
+static bool
 udplite_manip_pkt(struct sk_buff *skb,
 		  unsigned int iphdroff,
 		  const struct nf_conntrack_tuple *tuple,
@@ -41,7 +41,7 @@ udplite_manip_pkt(struct sk_buff *skb,
 	__be16 *portptr, newport;
 
 	if (!skb_make_writable(skb, hdroff + sizeof(*hdr)))
-		return 0;
+		return false;
 
 	iph = (struct iphdr *)(skb->data + iphdroff);
 	hdr = (struct udphdr *)(skb->data + hdroff);
@@ -66,7 +66,7 @@ udplite_manip_pkt(struct sk_buff *skb,
 		hdr->check = CSUM_MANGLED_0;
 
 	*portptr = newport;
-	return 1;
+	return true;
 }
 
 static const struct nf_nat_protocol nf_nat_protocol_udplite = {
diff --git a/net/ipv4/netfilter/nf_nat_proto_unknown.c b/net/ipv4/netfilter/nf_nat_proto_unknown.c
index cda21ff0e4cf..14381c62acea 100644
--- a/net/ipv4/netfilter/nf_nat_proto_unknown.c
+++ b/net/ipv4/netfilter/nf_nat_proto_unknown.c
@@ -18,31 +18,31 @@
 #include <net/netfilter/nf_nat_rule.h>
 #include <net/netfilter/nf_nat_protocol.h>
 
-static int unknown_in_range(const struct nf_conntrack_tuple *tuple,
-			    enum nf_nat_manip_type manip_type,
-			    const union nf_conntrack_man_proto *min,
-			    const union nf_conntrack_man_proto *max)
+static bool unknown_in_range(const struct nf_conntrack_tuple *tuple,
+			     enum nf_nat_manip_type manip_type,
+			     const union nf_conntrack_man_proto *min,
+			     const union nf_conntrack_man_proto *max)
 {
-	return 1;
+	return true;
 }
 
-static int unknown_unique_tuple(struct nf_conntrack_tuple *tuple,
-				const struct nf_nat_range *range,
-				enum nf_nat_manip_type maniptype,
-				const struct nf_conn *ct)
+static bool unknown_unique_tuple(struct nf_conntrack_tuple *tuple,
+				 const struct nf_nat_range *range,
+				 enum nf_nat_manip_type maniptype,
+				 const struct nf_conn *ct)
 {
 	/* Sorry: we can't help you; if it's not unique, we can't frob
 	   anything. */
-	return 0;
+	return false;
 }
 
-static int
+static bool
 unknown_manip_pkt(struct sk_buff *skb,
 		  unsigned int iphdroff,
 		  const struct nf_conntrack_tuple *tuple,
 		  enum nf_nat_manip_type maniptype)
 {
-	return 1;
+	return true;
 }
 
 const struct nf_nat_protocol nf_nat_unknown_protocol = {
-- 
cgit v1.2.3


From 3c9fba656a185cf56872a325e5594d9b4d4168ec Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@computergmbh.de>
Date: Mon, 14 Apr 2008 11:15:54 +0200
Subject: [NETFILTER]: nf_conntrack: replace NF_CT_DUMP_TUPLE macro indrection
 by function call

Directly call IPv4 and IPv6 variants where the address family is
easily known.

Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack_tuple.h     |  2 --
 net/ipv4/netfilter/ipt_CLUSTERIP.c             |  2 +-
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  2 +-
 net/ipv4/netfilter/nf_nat_pptp.c               |  2 +-
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  2 +-
 net/netfilter/nf_conntrack_core.c              |  2 +-
 net/netfilter/nf_conntrack_h323_main.c         | 26 +++++++++++++-------------
 net/netfilter/nf_conntrack_pptp.c              |  4 ++--
 net/netfilter/nf_conntrack_proto_gre.c         |  6 +++---
 net/netfilter/nf_conntrack_proto_tcp.c         |  6 +++---
 net/netfilter/nf_conntrack_sane.c              |  2 +-
 net/netfilter/nf_conntrack_tftp.c              |  6 +++---
 12 files changed, 30 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h
index 2722b13ecd6c..1bb7087833d3 100644
--- a/include/net/netfilter/nf_conntrack_tuple.h
+++ b/include/net/netfilter/nf_conntrack_tuple.h
@@ -151,8 +151,6 @@ static inline void nf_ct_dump_tuple(const struct nf_conntrack_tuple *t)
 	}
 }
 
-#define NF_CT_DUMP_TUPLE(tp)	nf_ct_dump_tuple(tp)
-
 /* If we're the first tuple, it's the original dir. */
 #define NF_CT_DIRECTION(h)						\
 	((enum ip_conntrack_dir)(h)->tuple.dst.dir)
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index 2510d4fcdb54..c1f970cb7221 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -331,7 +331,7 @@ clusterip_tg(struct sk_buff *skb, const struct net_device *in,
 	}
 
 #ifdef DEBUG
-	NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
+	nf_ct_dump_tuple_ip(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
 #endif
 	pr_debug("hash=%u ct_hash=%u ", hash, ct->mark);
 	if (!clusterip_responsible(cipinfo->config, hash)) {
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index 193a845fe7f8..78ab19accace 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -115,7 +115,7 @@ static bool icmp_new(struct nf_conn *ct, const struct sk_buff *skb,
 		/* Can't create a new ICMP `conn' with this. */
 		pr_debug("icmp: can't create new conn with type %u\n",
 			 ct->tuplehash[0].tuple.dst.u.icmp.type);
-		NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple);
+		nf_ct_dump_tuple_ip(&ct->tuplehash[0].tuple);
 		return false;
 	}
 	atomic_set(&ct->proto.icmp.count, 0);
diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c
index 3a1e6d6afc0a..da3d91a5ef5c 100644
--- a/net/ipv4/netfilter/nf_nat_pptp.c
+++ b/net/ipv4/netfilter/nf_nat_pptp.c
@@ -72,7 +72,7 @@ static void pptp_nat_expected(struct nf_conn *ct,
 	}
 
 	pr_debug("trying to unexpect other dir: ");
-	NF_CT_DUMP_TUPLE(&t);
+	nf_ct_dump_tuple_ip(&t);
 	other_exp = nf_ct_expect_find_get(&t);
 	if (other_exp) {
 		nf_ct_unexpect_related(other_exp);
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index 9ad40e0e17fc..ee713b03e9ec 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -115,7 +115,7 @@ static bool icmpv6_new(struct nf_conn *ct, const struct sk_buff *skb,
 		/* Can't create a new ICMPv6 `conn' with this. */
 		pr_debug("icmpv6: can't create new conn with type %u\n",
 			 type + 128);
-		NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple);
+		nf_ct_dump_tuple_ipv6(&ct->tuplehash[0].tuple);
 		return false;
 	}
 	atomic_set(&ct->proto.icmp.count, 0);
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index a3fe9db412d3..351237399e27 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -763,7 +763,7 @@ void nf_conntrack_alter_reply(struct nf_conn *ct,
 	NF_CT_ASSERT(!nf_ct_is_confirmed(ct));
 
 	pr_debug("Altering reply tuple of %p to ", ct);
-	NF_CT_DUMP_TUPLE(newreply);
+	nf_ct_dump_tuple(newreply);
 
 	ct->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply;
 	if (ct->master || (help && help->expecting != 0))
diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c
index c3f87094de43..95da1a24aab7 100644
--- a/net/netfilter/nf_conntrack_h323_main.c
+++ b/net/netfilter/nf_conntrack_h323_main.c
@@ -303,9 +303,9 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct,
 		if (nf_ct_expect_related(rtp_exp) == 0) {
 			if (nf_ct_expect_related(rtcp_exp) == 0) {
 				pr_debug("nf_ct_h323: expect RTP ");
-				NF_CT_DUMP_TUPLE(&rtp_exp->tuple);
+				nf_ct_dump_tuple(&rtp_exp->tuple);
 				pr_debug("nf_ct_h323: expect RTCP ");
-				NF_CT_DUMP_TUPLE(&rtcp_exp->tuple);
+				nf_ct_dump_tuple(&rtcp_exp->tuple);
 			} else {
 				nf_ct_unexpect_related(rtp_exp);
 				ret = -1;
@@ -360,7 +360,7 @@ static int expect_t120(struct sk_buff *skb,
 	} else {		/* Conntrack only */
 		if (nf_ct_expect_related(exp) == 0) {
 			pr_debug("nf_ct_h323: expect T.120 ");
-			NF_CT_DUMP_TUPLE(&exp->tuple);
+			nf_ct_dump_tuple(&exp->tuple);
 		} else
 			ret = -1;
 	}
@@ -582,7 +582,7 @@ static int h245_help(struct sk_buff *skb, unsigned int protoff,
 	while (get_tpkt_data(skb, protoff, ct, ctinfo,
 			     &data, &datalen, &dataoff)) {
 		pr_debug("nf_ct_h245: TPKT len=%d ", datalen);
-		NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple);
+		nf_ct_dump_tuple(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple);
 
 		/* Decode H.245 signal */
 		ret = DecodeMultimediaSystemControlMessage(data, datalen,
@@ -695,7 +695,7 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct,
 	} else {		/* Conntrack only */
 		if (nf_ct_expect_related(exp) == 0) {
 			pr_debug("nf_ct_q931: expect H.245 ");
-			NF_CT_DUMP_TUPLE(&exp->tuple);
+			nf_ct_dump_tuple(&exp->tuple);
 		} else
 			ret = -1;
 	}
@@ -810,7 +810,7 @@ static int expect_callforwarding(struct sk_buff *skb,
 	} else {		/* Conntrack only */
 		if (nf_ct_expect_related(exp) == 0) {
 			pr_debug("nf_ct_q931: expect Call Forwarding ");
-			NF_CT_DUMP_TUPLE(&exp->tuple);
+			nf_ct_dump_tuple(&exp->tuple);
 		} else
 			ret = -1;
 	}
@@ -1130,7 +1130,7 @@ static int q931_help(struct sk_buff *skb, unsigned int protoff,
 	while (get_tpkt_data(skb, protoff, ct, ctinfo,
 			     &data, &datalen, &dataoff)) {
 		pr_debug("nf_ct_q931: TPKT len=%d ", datalen);
-		NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple);
+		nf_ct_dump_tuple(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple);
 
 		/* Decode Q.931 signal */
 		ret = DecodeQ931(data, datalen, &q931);
@@ -1279,7 +1279,7 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct,
 	} else {		/* Conntrack only */
 		if (nf_ct_expect_related(exp) == 0) {
 			pr_debug("nf_ct_ras: expect Q.931 ");
-			NF_CT_DUMP_TUPLE(&exp->tuple);
+			nf_ct_dump_tuple(&exp->tuple);
 
 			/* Save port for looking up expect in processing RCF */
 			info->sig_port[dir] = port;
@@ -1343,7 +1343,7 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct,
 
 	if (nf_ct_expect_related(exp) == 0) {
 		pr_debug("nf_ct_ras: expect RAS ");
-		NF_CT_DUMP_TUPLE(&exp->tuple);
+		nf_ct_dump_tuple(&exp->tuple);
 	} else
 		ret = -1;
 
@@ -1427,7 +1427,7 @@ static int process_rcf(struct sk_buff *skb, struct nf_conn *ct,
 			pr_debug("nf_ct_ras: set Q.931 expect "
 				 "timeout to %u seconds for",
 				 info->timeout);
-			NF_CT_DUMP_TUPLE(&exp->tuple);
+			nf_ct_dump_tuple(&exp->tuple);
 			set_expect_timeout(exp, info->timeout);
 		}
 		spin_unlock_bh(&nf_conntrack_lock);
@@ -1548,7 +1548,7 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct,
 
 	if (nf_ct_expect_related(exp) == 0) {
 		pr_debug("nf_ct_ras: expect Q.931 ");
-		NF_CT_DUMP_TUPLE(&exp->tuple);
+		nf_ct_dump_tuple(&exp->tuple);
 	} else
 		ret = -1;
 
@@ -1601,7 +1601,7 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct,
 
 	if (nf_ct_expect_related(exp) == 0) {
 		pr_debug("nf_ct_ras: expect Q.931 ");
-		NF_CT_DUMP_TUPLE(&exp->tuple);
+		nf_ct_dump_tuple(&exp->tuple);
 	} else
 		ret = -1;
 
@@ -1705,7 +1705,7 @@ static int ras_help(struct sk_buff *skb, unsigned int protoff,
 	if (data == NULL)
 		goto accept;
 	pr_debug("nf_ct_ras: RAS message len=%d ", datalen);
-	NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple);
+	nf_ct_dump_tuple(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple);
 
 	/* Decode RAS message */
 	ret = DecodeRasMessage(data, datalen, &ras);
diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c
index 4793cc078789..97e54b0e43a3 100644
--- a/net/netfilter/nf_conntrack_pptp.c
+++ b/net/netfilter/nf_conntrack_pptp.c
@@ -119,7 +119,7 @@ static void pptp_expectfn(struct nf_conn *ct,
 		/* obviously this tuple inversion only works until you do NAT */
 		nf_ct_invert_tuplepr(&inv_t, &exp->tuple);
 		pr_debug("trying to unexpect other dir: ");
-		NF_CT_DUMP_TUPLE(&inv_t);
+		nf_ct_dump_tuple(&inv_t);
 
 		exp_other = nf_ct_expect_find_get(&inv_t);
 		if (exp_other) {
@@ -141,7 +141,7 @@ static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t)
 	struct nf_conn *sibling;
 
 	pr_debug("trying to timeout ct or exp for tuple ");
-	NF_CT_DUMP_TUPLE(t);
+	nf_ct_dump_tuple(t);
 
 	h = nf_conntrack_find_get(t);
 	if (h)  {
diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c
index 7d37a2ea67b2..654a4f7f12c6 100644
--- a/net/netfilter/nf_conntrack_proto_gre.c
+++ b/net/netfilter/nf_conntrack_proto_gre.c
@@ -82,7 +82,7 @@ static __be16 gre_keymap_lookup(struct nf_conntrack_tuple *t)
 	read_unlock_bh(&nf_ct_gre_lock);
 
 	pr_debug("lookup src key 0x%x for ", key);
-	NF_CT_DUMP_TUPLE(t);
+	nf_ct_dump_tuple(t);
 
 	return key;
 }
@@ -113,7 +113,7 @@ int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir,
 	*kmp = km;
 
 	pr_debug("adding new entry %p: ", km);
-	NF_CT_DUMP_TUPLE(&km->tuple);
+	nf_ct_dump_tuple(&km->tuple);
 
 	write_lock_bh(&nf_ct_gre_lock);
 	list_add_tail(&km->list, &gre_keymap_list);
@@ -238,7 +238,7 @@ static bool gre_new(struct nf_conn *ct, const struct sk_buff *skb,
 		    unsigned int dataoff)
 {
 	pr_debug(": ");
-	NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
+	nf_ct_dump_tuple(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
 
 	/* initialize to sane value.  Ideally a conntrack helper
 	 * (e.g. in case of pptp) is increasing them */
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 73a8b32db7be..ba94004fe323 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -505,7 +505,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
 
 	pr_debug("tcp_in_window: START\n");
 	pr_debug("tcp_in_window: ");
-	NF_CT_DUMP_TUPLE(tuple);
+	nf_ct_dump_tuple(tuple);
 	pr_debug("seq=%u ack=%u sack=%u win=%u end=%u\n",
 		 seq, ack, sack, win, end);
 	pr_debug("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i "
@@ -592,7 +592,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
 		seq = end = sender->td_end;
 
 	pr_debug("tcp_in_window: ");
-	NF_CT_DUMP_TUPLE(tuple);
+	nf_ct_dump_tuple(tuple);
 	pr_debug("seq=%u ack=%u sack =%u win=%u end=%u\n",
 		 seq, ack, sack, win, end);
 	pr_debug("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i "
@@ -936,7 +936,7 @@ static int tcp_packet(struct nf_conn *ct,
 	ct->proto.tcp.last_dir = dir;
 
 	pr_debug("tcp_conntracks: ");
-	NF_CT_DUMP_TUPLE(tuple);
+	nf_ct_dump_tuple(tuple);
 	pr_debug("syn=%i ack=%i fin=%i rst=%i old=%i new=%i\n",
 		 (th->syn ? 1 : 0), (th->ack ? 1 : 0),
 		 (th->fin ? 1 : 0), (th->rst ? 1 : 0),
diff --git a/net/netfilter/nf_conntrack_sane.c b/net/netfilter/nf_conntrack_sane.c
index c3d5e84dcc9b..a94294b2b23c 100644
--- a/net/netfilter/nf_conntrack_sane.c
+++ b/net/netfilter/nf_conntrack_sane.c
@@ -147,7 +147,7 @@ static int help(struct sk_buff *skb,
 			  IPPROTO_TCP, NULL, &reply->port);
 
 	pr_debug("nf_ct_sane: expect: ");
-	NF_CT_DUMP_TUPLE(&exp->tuple);
+	nf_ct_dump_tuple(&exp->tuple);
 
 	/* Can't expect this?  Best to drop packet now. */
 	if (nf_ct_expect_related(exp) != 0)
diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c
index ea5ff49d77bc..f57f6e7a71ee 100644
--- a/net/netfilter/nf_conntrack_tftp.c
+++ b/net/netfilter/nf_conntrack_tftp.c
@@ -55,8 +55,8 @@ static int tftp_help(struct sk_buff *skb,
 	case TFTP_OPCODE_READ:
 	case TFTP_OPCODE_WRITE:
 		/* RRQ and WRQ works the same way */
-		NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
-		NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
+		nf_ct_dump_tuple(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
+		nf_ct_dump_tuple(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
 
 		exp = nf_ct_expect_alloc(ct);
 		if (exp == NULL)
@@ -68,7 +68,7 @@ static int tftp_help(struct sk_buff *skb,
 				  IPPROTO_UDP, NULL, &tuple->dst.u.udp.port);
 
 		pr_debug("expect: ");
-		NF_CT_DUMP_TUPLE(&exp->tuple);
+		nf_ct_dump_tuple(&exp->tuple);
 
 		nf_nat_tftp = rcu_dereference(nf_nat_tftp_hook);
 		if (nf_nat_tftp && ct->status & IPS_NAT_MASK)
-- 
cgit v1.2.3


From e7bfd0a1a6c8f82977253dab19be9d9979c1ec1b Mon Sep 17 00:00:00 2001
From: Peter Warasin <peter@endian.com>
Date: Mon, 14 Apr 2008 11:15:54 +0200
Subject: [NETFILTER]: bridge: add ebt_nflog watcher

This patch adds the ebtables nflog watcher to the kernel in order to
allow ebtables log through the nfnetlink_log backend.

Signed-off-by: Peter Warasin <peter@endian.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter_bridge/ebt_nflog.h | 21 +++++++++
 net/bridge/netfilter/Kconfig               | 14 ++++++
 net/bridge/netfilter/Makefile              |  1 +
 net/bridge/netfilter/ebt_nflog.c           | 74 ++++++++++++++++++++++++++++++
 4 files changed, 110 insertions(+)
 create mode 100644 include/linux/netfilter_bridge/ebt_nflog.h
 create mode 100644 net/bridge/netfilter/ebt_nflog.c

(limited to 'include')

diff --git a/include/linux/netfilter_bridge/ebt_nflog.h b/include/linux/netfilter_bridge/ebt_nflog.h
new file mode 100644
index 000000000000..052817849b83
--- /dev/null
+++ b/include/linux/netfilter_bridge/ebt_nflog.h
@@ -0,0 +1,21 @@
+#ifndef __LINUX_BRIDGE_EBT_NFLOG_H
+#define __LINUX_BRIDGE_EBT_NFLOG_H
+
+#define EBT_NFLOG_MASK 0x0
+
+#define EBT_NFLOG_PREFIX_SIZE 64
+#define EBT_NFLOG_WATCHER "nflog"
+
+#define EBT_NFLOG_DEFAULT_GROUP		0x1
+#define EBT_NFLOG_DEFAULT_THRESHOLD	1
+
+struct ebt_nflog_info {
+	u_int32_t len;
+	u_int16_t group;
+	u_int16_t threshold;
+	u_int16_t flags;
+	u_int16_t pad;
+	char prefix[EBT_NFLOG_PREFIX_SIZE];
+};
+
+#endif				/* __LINUX_BRIDGE_EBT_NFLOG_H */
diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig
index 4a3e2bf892c7..7beeefa0f9c0 100644
--- a/net/bridge/netfilter/Kconfig
+++ b/net/bridge/netfilter/Kconfig
@@ -212,4 +212,18 @@ config BRIDGE_EBT_ULOG
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config BRIDGE_EBT_NFLOG
+	tristate "ebt: nflog support"
+	depends on BRIDGE_NF_EBTABLES
+	help
+	  This option enables the nflog watcher, which allows to LOG
+	  messages through the netfilter logging API, which can use
+	  either the old LOG target, the old ULOG target or nfnetlink_log
+	  as backend.
+
+	  This option adds the ulog watcher, that you can use in any rule
+	  in any ebtables table.
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 endmenu
diff --git a/net/bridge/netfilter/Makefile b/net/bridge/netfilter/Makefile
index 905087e0d485..83715d73a503 100644
--- a/net/bridge/netfilter/Makefile
+++ b/net/bridge/netfilter/Makefile
@@ -30,3 +30,4 @@ obj-$(CONFIG_BRIDGE_EBT_SNAT) += ebt_snat.o
 # watchers
 obj-$(CONFIG_BRIDGE_EBT_LOG) += ebt_log.o
 obj-$(CONFIG_BRIDGE_EBT_ULOG) += ebt_ulog.o
+obj-$(CONFIG_BRIDGE_EBT_NFLOG) += ebt_nflog.o
diff --git a/net/bridge/netfilter/ebt_nflog.c b/net/bridge/netfilter/ebt_nflog.c
new file mode 100644
index 000000000000..8e799aa9e560
--- /dev/null
+++ b/net/bridge/netfilter/ebt_nflog.c
@@ -0,0 +1,74 @@
+/*
+ * ebt_nflog
+ *
+ *	Author:
+ *	Peter Warasin <peter@endian.com>
+ *
+ *  February, 2008
+ *
+ * Based on:
+ *  xt_NFLOG.c, (C) 2006 by Patrick McHardy <kaber@trash.net>
+ *  ebt_ulog.c, (C) 2004 by Bart De Schuymer <bdschuym@pandora.be>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/netfilter_bridge/ebtables.h>
+#include <linux/netfilter_bridge/ebt_nflog.h>
+#include <net/netfilter/nf_log.h>
+
+static void ebt_nflog(const struct sk_buff *skb,
+		      unsigned int hooknr,
+		      const struct net_device *in,
+		      const struct net_device *out,
+		      const void *data, unsigned int datalen)
+{
+	struct ebt_nflog_info *info = (struct ebt_nflog_info *)data;
+	struct nf_loginfo li;
+
+	li.type = NF_LOG_TYPE_ULOG;
+	li.u.ulog.copy_len = info->len;
+	li.u.ulog.group = info->group;
+	li.u.ulog.qthreshold = info->threshold;
+
+	nf_log_packet(PF_BRIDGE, hooknr, skb, in, out, &li, "%s", info->prefix);
+}
+
+static int ebt_nflog_check(const char *tablename,
+			   unsigned int hookmask,
+			   const struct ebt_entry *e,
+			   void *data, unsigned int datalen)
+{
+	struct ebt_nflog_info *info = (struct ebt_nflog_info *)data;
+
+	if (datalen != EBT_ALIGN(sizeof(struct ebt_nflog_info)))
+		return -EINVAL;
+	if (info->flags & ~EBT_NFLOG_MASK)
+		return -EINVAL;
+	info->prefix[EBT_NFLOG_PREFIX_SIZE - 1] = '\0';
+	return 0;
+}
+
+static struct ebt_watcher nflog __read_mostly = {
+	.name = EBT_NFLOG_WATCHER,
+	.watcher = ebt_nflog,
+	.check = ebt_nflog_check,
+	.me = THIS_MODULE,
+};
+
+static int __init ebt_nflog_init(void)
+{
+	return ebt_register_watcher(&nflog);
+}
+
+static void __exit ebt_nflog_fini(void)
+{
+	ebt_unregister_watcher(&nflog);
+}
+
+module_init(ebt_nflog_init);
+module_exit(ebt_nflog_fini);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Peter Warasin <peter@endian.com>");
+MODULE_DESCRIPTION("ebtables NFLOG netfilter logging module");
-- 
cgit v1.2.3


From 0f389ec63077521166f071e1e970aed36147fd45 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 14 Apr 2008 18:53:02 +0300
Subject: slub: No need for per node slab counters if !SLUB_DEBUG

The per node counters are used mainly for showing data through the sysfs API.
If that API is not compiled in then there is no point in keeping track of this
data. Disable counters for the number of slabs and the number of total slabs
if !SLUB_DEBUG. Incrementing the per node counters is also accessing a
potentially contended cacheline so this could actually be a performance
benefit to embedded systems.

SLABINFO support is also affected. It now must depends on SLUB_DEBUG (which
is on by default).

Patch also avoids a check for a NULL kmem_cache_node pointer in new_slab()
if the system is not compiled with NUMA support.

[penberg@cs.helsinki.fi: fix oops and move ->nr_slabs into CONFIG_SLUB_DEBUG]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
---
 include/linux/slub_def.h |  2 +-
 init/Kconfig             |  2 +-
 mm/slub.c                | 51 +++++++++++++++++++++++++++++++++++++-----------
 3 files changed, 42 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/slub_def.h b/include/linux/slub_def.h
index b00c1c73eb0a..79d59c937fac 100644
--- a/include/linux/slub_def.h
+++ b/include/linux/slub_def.h
@@ -45,9 +45,9 @@ struct kmem_cache_cpu {
 struct kmem_cache_node {
 	spinlock_t list_lock;	/* Protect partial list and nr_partial */
 	unsigned long nr_partial;
-	atomic_long_t nr_slabs;
 	struct list_head partial;
 #ifdef CONFIG_SLUB_DEBUG
+	atomic_long_t nr_slabs;
 	struct list_head full;
 #endif
 };
diff --git a/init/Kconfig b/init/Kconfig
index a97924bc5b8d..7fccf09bb95a 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -763,7 +763,7 @@ endmenu		# General setup
 config SLABINFO
 	bool
 	depends on PROC_FS
-	depends on SLAB || SLUB
+	depends on SLAB || SLUB_DEBUG
 	default y
 
 config RT_MUTEXES
diff --git a/mm/slub.c b/mm/slub.c
index f924cffb29e7..7f8aaa291a4e 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -837,6 +837,35 @@ static void remove_full(struct kmem_cache *s, struct page *page)
 	spin_unlock(&n->list_lock);
 }
 
+/* Tracking of the number of slabs for debugging purposes */
+static inline unsigned long slabs_node(struct kmem_cache *s, int node)
+{
+	struct kmem_cache_node *n = get_node(s, node);
+
+	return atomic_long_read(&n->nr_slabs);
+}
+
+static inline void inc_slabs_node(struct kmem_cache *s, int node)
+{
+	struct kmem_cache_node *n = get_node(s, node);
+
+	/*
+	 * May be called early in order to allocate a slab for the
+	 * kmem_cache_node structure. Solve the chicken-egg
+	 * dilemma by deferring the increment of the count during
+	 * bootstrap (see early_kmem_cache_node_alloc).
+	 */
+	if (!NUMA_BUILD || n)
+		atomic_long_inc(&n->nr_slabs);
+}
+static inline void dec_slabs_node(struct kmem_cache *s, int node)
+{
+	struct kmem_cache_node *n = get_node(s, node);
+
+	atomic_long_dec(&n->nr_slabs);
+}
+
+/* Object debug checks for alloc/free paths */
 static void setup_object_debug(struct kmem_cache *s, struct page *page,
 								void *object)
 {
@@ -1028,6 +1057,11 @@ static inline unsigned long kmem_cache_flags(unsigned long objsize,
 	return flags;
 }
 #define slub_debug 0
+
+static inline unsigned long slabs_node(struct kmem_cache *s, int node)
+							{ return 0; }
+static inline void inc_slabs_node(struct kmem_cache *s, int node) {}
+static inline void dec_slabs_node(struct kmem_cache *s, int node) {}
 #endif
 /*
  * Slab allocation and freeing
@@ -1066,7 +1100,6 @@ static void setup_object(struct kmem_cache *s, struct page *page,
 static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
 {
 	struct page *page;
-	struct kmem_cache_node *n;
 	void *start;
 	void *last;
 	void *p;
@@ -1078,9 +1111,7 @@ static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
 	if (!page)
 		goto out;
 
-	n = get_node(s, page_to_nid(page));
-	if (n)
-		atomic_long_inc(&n->nr_slabs);
+	inc_slabs_node(s, page_to_nid(page));
 	page->slab = s;
 	page->flags |= 1 << PG_slab;
 	if (s->flags & (SLAB_DEBUG_FREE | SLAB_RED_ZONE | SLAB_POISON |
@@ -1153,9 +1184,7 @@ static void free_slab(struct kmem_cache *s, struct page *page)
 
 static void discard_slab(struct kmem_cache *s, struct page *page)
 {
-	struct kmem_cache_node *n = get_node(s, page_to_nid(page));
-
-	atomic_long_dec(&n->nr_slabs);
+	dec_slabs_node(s, page_to_nid(page));
 	free_slab(s, page);
 }
 
@@ -1894,10 +1923,10 @@ static void init_kmem_cache_cpu(struct kmem_cache *s,
 static void init_kmem_cache_node(struct kmem_cache_node *n)
 {
 	n->nr_partial = 0;
-	atomic_long_set(&n->nr_slabs, 0);
 	spin_lock_init(&n->list_lock);
 	INIT_LIST_HEAD(&n->partial);
 #ifdef CONFIG_SLUB_DEBUG
+	atomic_long_set(&n->nr_slabs, 0);
 	INIT_LIST_HEAD(&n->full);
 #endif
 }
@@ -2066,7 +2095,7 @@ static struct kmem_cache_node *early_kmem_cache_node_alloc(gfp_t gfpflags,
 	init_tracking(kmalloc_caches, n);
 #endif
 	init_kmem_cache_node(n);
-	atomic_long_inc(&n->nr_slabs);
+	inc_slabs_node(kmalloc_caches, node);
 
 	/*
 	 * lockdep requires consistent irq usage for each lock
@@ -2379,7 +2408,7 @@ static inline int kmem_cache_close(struct kmem_cache *s)
 		struct kmem_cache_node *n = get_node(s, node);
 
 		n->nr_partial -= free_list(s, n, &n->partial);
-		if (atomic_long_read(&n->nr_slabs))
+		if (slabs_node(s, node))
 			return 1;
 	}
 	free_kmem_cache_nodes(s);
@@ -2801,7 +2830,7 @@ static void slab_mem_offline_callback(void *arg)
 			 * and offline_pages() function shoudn't call this
 			 * callback. So, we must fail.
 			 */
-			BUG_ON(atomic_long_read(&n->nr_slabs));
+			BUG_ON(slabs_node(s, offline_node));
 
 			s->node[offline_node] = NULL;
 			kmem_cache_free(kmalloc_caches, n);
-- 
cgit v1.2.3


From 0c3141e910eaaa0b617e2f26c69b266d1cd1f035 Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@windriver.com>
Date: Tue, 15 Apr 2008 00:22:02 -0700
Subject: [TIPC]: Overhaul of socket locking logic

This patch modifies TIPC's socket code to follow the same approach
used by other protocols.  This change eliminates the need for a
mutex in the TIPC-specific portion of the socket protocol data
structure -- in its place, the standard Linux socket backlog queue
and associated locking routines are utilized.  These changes fix
a long-standing receive queue bug on SMP systems, and also enable
individual read and write threads to utilize a socket without
unnecessarily interfering with each other.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tipc/tipc_port.h |    6 +
 net/tipc/port.c              |   31 +-
 net/tipc/socket.c            | 1002 ++++++++++++++++++++++++------------------
 3 files changed, 608 insertions(+), 431 deletions(-)

(limited to 'include')

diff --git a/include/net/tipc/tipc_port.h b/include/net/tipc/tipc_port.h
index c9b36b77a0b9..11105bcc4457 100644
--- a/include/net/tipc/tipc_port.h
+++ b/include/net/tipc/tipc_port.h
@@ -96,6 +96,12 @@ struct tipc_port *tipc_get_port(const u32 ref);
 
 void *tipc_get_handle(const u32 ref);
 
+/*
+ * The following routines require that the port be locked on entry
+ */
+
+int tipc_disconnect_port(struct tipc_port *tp_ptr);
+
 
 #endif
 
diff --git a/net/tipc/port.c b/net/tipc/port.c
index e2646a96935d..2f5806410c64 100644
--- a/net/tipc/port.c
+++ b/net/tipc/port.c
@@ -1240,6 +1240,28 @@ exit:
 	return res;
 }
 
+/**
+ * tipc_disconnect_port - disconnect port from peer
+ *
+ * Port must be locked.
+ */
+
+int tipc_disconnect_port(struct tipc_port *tp_ptr)
+{
+	int res;
+
+	if (tp_ptr->connected) {
+		tp_ptr->connected = 0;
+		/* let timer expire on it's own to avoid deadlock! */
+		tipc_nodesub_unsubscribe(
+			&((struct port *)tp_ptr)->subscription);
+		res = TIPC_OK;
+	} else {
+		res = -ENOTCONN;
+	}
+	return res;
+}
+
 /*
  * tipc_disconnect(): Disconnect port form peer.
  *                    This is a node local operation.
@@ -1248,17 +1270,12 @@ exit:
 int tipc_disconnect(u32 ref)
 {
 	struct port *p_ptr;
-	int res = -ENOTCONN;
+	int res;
 
 	p_ptr = tipc_port_lock(ref);
 	if (!p_ptr)
 		return -EINVAL;
-	if (p_ptr->publ.connected) {
-		p_ptr->publ.connected = 0;
-		/* let timer expire on it's own to avoid deadlock! */
-		tipc_nodesub_unsubscribe(&p_ptr->subscription);
-		res = TIPC_OK;
-	}
+	res = tipc_disconnect_port((struct tipc_port *)p_ptr);
 	tipc_port_unlock(p_ptr);
 	return res;
 }
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index df1960384364..05853159536a 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -43,7 +43,6 @@
 #include <linux/slab.h>
 #include <linux/poll.h>
 #include <linux/fcntl.h>
-#include <linux/mutex.h>
 #include <asm/string.h>
 #include <asm/atomic.h>
 #include <net/sock.h>
@@ -64,11 +63,12 @@
 struct tipc_sock {
 	struct sock sk;
 	struct tipc_port *p;
-	struct mutex lock;
 };
 
-#define tipc_sk(sk) ((struct tipc_sock*)sk)
+#define tipc_sk(sk) ((struct tipc_sock *)(sk))
+#define tipc_sk_port(sk) ((struct tipc_port *)(tipc_sk(sk)->p))
 
+static int backlog_rcv(struct sock *sk, struct sk_buff *skb);
 static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf);
 static void wakeupdispatch(struct tipc_port *tport);
 
@@ -82,55 +82,115 @@ static int sockets_enabled = 0;
 
 static atomic_t tipc_queue_size = ATOMIC_INIT(0);
 
-
 /*
- * sock_lock(): Lock a port/socket pair. lock_sock() can
- * not be used here, since the same lock must protect ports
- * with non-socket interfaces.
- * See net.c for description of locking policy.
+ * Revised TIPC socket locking policy:
+ *
+ * Most socket operations take the standard socket lock when they start
+ * and hold it until they finish (or until they need to sleep).  Acquiring
+ * this lock grants the owner exclusive access to the fields of the socket
+ * data structures, with the exception of the backlog queue.  A few socket
+ * operations can be done without taking the socket lock because they only
+ * read socket information that never changes during the life of the socket.
+ *
+ * Socket operations may acquire the lock for the associated TIPC port if they
+ * need to perform an operation on the port.  If any routine needs to acquire
+ * both the socket lock and the port lock it must take the socket lock first
+ * to avoid the risk of deadlock.
+ *
+ * The dispatcher handling incoming messages cannot grab the socket lock in
+ * the standard fashion, since invoked it runs at the BH level and cannot block.
+ * Instead, it checks to see if the socket lock is currently owned by someone,
+ * and either handles the message itself or adds it to the socket's backlog
+ * queue; in the latter case the queued message is processed once the process
+ * owning the socket lock releases it.
+ *
+ * NOTE: Releasing the socket lock while an operation is sleeping overcomes
+ * the problem of a blocked socket operation preventing any other operations
+ * from occurring.  However, applications must be careful if they have
+ * multiple threads trying to send (or receive) on the same socket, as these
+ * operations might interfere with each other.  For example, doing a connect
+ * and a receive at the same time might allow the receive to consume the
+ * ACK message meant for the connect.  While additional work could be done
+ * to try and overcome this, it doesn't seem to be worthwhile at the present.
+ *
+ * NOTE: Releasing the socket lock while an operation is sleeping also ensures
+ * that another operation that must be performed in a non-blocking manner is
+ * not delayed for very long because the lock has already been taken.
+ *
+ * NOTE: This code assumes that certain fields of a port/socket pair are
+ * constant over its lifetime; such fields can be examined without taking
+ * the socket lock and/or port lock, and do not need to be re-read even
+ * after resuming processing after waiting.  These fields include:
+ *   - socket type
+ *   - pointer to socket sk structure (aka tipc_sock structure)
+ *   - pointer to port structure
+ *   - port reference
+ */
+
+/**
+ * advance_rx_queue - discard first buffer in socket receive queue
+ *
+ * Caller must hold socket lock
  */
-static void sock_lock(struct tipc_sock* tsock)
+
+static void advance_rx_queue(struct sock *sk)
 {
-	spin_lock_bh(tsock->p->lock);
+	buf_discard(__skb_dequeue(&sk->sk_receive_queue));
+	atomic_dec(&tipc_queue_size);
 }
 
-/*
- * sock_unlock(): Unlock a port/socket pair
+/**
+ * discard_rx_queue - discard all buffers in socket receive queue
+ *
+ * Caller must hold socket lock
  */
-static void sock_unlock(struct tipc_sock* tsock)
+
+static void discard_rx_queue(struct sock *sk)
 {
-	spin_unlock_bh(tsock->p->lock);
+	struct sk_buff *buf;
+
+	while ((buf = __skb_dequeue(&sk->sk_receive_queue))) {
+		atomic_dec(&tipc_queue_size);
+		buf_discard(buf);
+	}
 }
 
 /**
- * advance_queue - discard first buffer in queue
- * @tsock: TIPC socket
+ * reject_rx_queue - reject all buffers in socket receive queue
+ *
+ * Caller must hold socket lock
  */
 
-static void advance_queue(struct tipc_sock *tsock)
+static void reject_rx_queue(struct sock *sk)
 {
-	sock_lock(tsock);
-	buf_discard(skb_dequeue(&tsock->sk.sk_receive_queue));
-	sock_unlock(tsock);
-	atomic_dec(&tipc_queue_size);
+	struct sk_buff *buf;
+
+	while ((buf = __skb_dequeue(&sk->sk_receive_queue))) {
+		tipc_reject_msg(buf, TIPC_ERR_NO_PORT);
+		atomic_dec(&tipc_queue_size);
+	}
 }
 
 /**
  * tipc_create - create a TIPC socket
+ * @net: network namespace (must be default network)
  * @sock: pre-allocated socket structure
  * @protocol: protocol indicator (must be 0)
  *
- * This routine creates and attaches a 'struct sock' to the 'struct socket',
- * then create and attaches a TIPC port to the 'struct sock' part.
+ * This routine creates additional data structures used by the TIPC socket,
+ * initializes them, and links them together.
  *
  * Returns 0 on success, errno otherwise
  */
+
 static int tipc_create(struct net *net, struct socket *sock, int protocol)
 {
-	struct tipc_sock *tsock;
-	struct tipc_port *port;
+	const struct proto_ops *ops;
+	socket_state state;
 	struct sock *sk;
-	u32 ref;
+	u32 portref;
+
+	/* Validate arguments */
 
 	if (net != &init_net)
 		return -EAFNOSUPPORT;
@@ -138,53 +198,56 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 	if (unlikely(protocol != 0))
 		return -EPROTONOSUPPORT;
 
-	ref = tipc_createport_raw(NULL, &dispatch, &wakeupdispatch, TIPC_LOW_IMPORTANCE);
-	if (unlikely(!ref))
-		return -ENOMEM;
-
-	sock->state = SS_UNCONNECTED;
-
 	switch (sock->type) {
 	case SOCK_STREAM:
-		sock->ops = &stream_ops;
+		ops = &stream_ops;
+		state = SS_UNCONNECTED;
 		break;
 	case SOCK_SEQPACKET:
-		sock->ops = &packet_ops;
+		ops = &packet_ops;
+		state = SS_UNCONNECTED;
 		break;
 	case SOCK_DGRAM:
-		tipc_set_portunreliable(ref, 1);
-		/* fall through */
 	case SOCK_RDM:
-		tipc_set_portunreturnable(ref, 1);
-		sock->ops = &msg_ops;
-		sock->state = SS_READY;
+		ops = &msg_ops;
+		state = SS_READY;
 		break;
 	default:
-		tipc_deleteport(ref);
 		return -EPROTOTYPE;
 	}
 
+	/* Allocate socket's protocol area */
+
 	sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto);
-	if (!sk) {
-		tipc_deleteport(ref);
+	if (sk == NULL)
 		return -ENOMEM;
-	}
 
-	sock_init_data(sock, sk);
-	sk->sk_rcvtimeo = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT);
+	/* Allocate TIPC port for socket to use */
 
-	tsock = tipc_sk(sk);
-	port = tipc_get_port(ref);
+	portref = tipc_createport_raw(sk, &dispatch, &wakeupdispatch,
+				      TIPC_LOW_IMPORTANCE);
+	if (unlikely(portref == 0)) {
+		sk_free(sk);
+		return -ENOMEM;
+	}
 
-	tsock->p = port;
-	port->usr_handle = tsock;
+	/* Finish initializing socket data structures */
 
-	mutex_init(&tsock->lock);
+	sock->ops = ops;
+	sock->state = state;
 
-	dbg("sock_create: %x\n",tsock);
+	sock_init_data(sock, sk);
+	sk->sk_rcvtimeo = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT);
+	sk->sk_backlog_rcv = backlog_rcv;
+	tipc_sk(sk)->p = tipc_get_port(portref);
 
-	atomic_inc(&tipc_user_count);
+	if (sock->state == SS_READY) {
+		tipc_set_portunreturnable(portref, 1);
+		if (sock->type == SOCK_DGRAM)
+			tipc_set_portunreliable(portref, 1);
+	}
 
+	atomic_inc(&tipc_user_count);
 	return 0;
 }
 
@@ -207,52 +270,62 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol)
 
 static int release(struct socket *sock)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
 	struct sock *sk = sock->sk;
-	int res = TIPC_OK;
+	struct tipc_port *tport;
 	struct sk_buff *buf;
+	int res;
 
-	dbg("sock_delete: %x\n",tsock);
-	if (!tsock)
-		return 0;
-	mutex_lock(&tsock->lock);
-	if (!sock->sk) {
-		mutex_unlock(&tsock->lock);
+	/*
+	 * Exit if socket isn't fully initialized (occurs when a failed accept()
+	 * releases a pre-allocated child socket that was never used)
+	 */
+
+	if (sk == NULL)
 		return 0;
-	}
 
-	/* Reject unreceived messages, unless no longer connected */
+	tport = tipc_sk_port(sk);
+	lock_sock(sk);
+
+	/*
+	 * Reject all unreceived messages, except on an active connection
+	 * (which disconnects locally & sends a 'FIN+' to peer)
+	 */
 
 	while (sock->state != SS_DISCONNECTING) {
-		sock_lock(tsock);
-		buf = skb_dequeue(&sk->sk_receive_queue);
-		if (!buf)
-			tsock->p->usr_handle = NULL;
-		sock_unlock(tsock);
-		if (!buf)
+		buf = __skb_dequeue(&sk->sk_receive_queue);
+		if (buf == NULL)
 			break;
+		atomic_dec(&tipc_queue_size);
 		if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf)))
 			buf_discard(buf);
-		else
+		else {
+			if ((sock->state == SS_CONNECTING) ||
+			    (sock->state == SS_CONNECTED)) {
+				sock->state = SS_DISCONNECTING;
+				tipc_disconnect(tport->ref);
+			}
 			tipc_reject_msg(buf, TIPC_ERR_NO_PORT);
-		atomic_dec(&tipc_queue_size);
+		}
 	}
 
-	/* Delete TIPC port */
+	/*
+	 * Delete TIPC port; this ensures no more messages are queued
+	 * (also disconnects an active connection & sends a 'FIN-' to peer)
+	 */
 
-	res = tipc_deleteport(tsock->p->ref);
-	sock->sk = NULL;
+	res = tipc_deleteport(tport->ref);
 
-	/* Discard any remaining messages */
+	/* Discard any remaining (connection-based) messages in receive queue */
 
-	while ((buf = skb_dequeue(&sk->sk_receive_queue))) {
-		buf_discard(buf);
-		atomic_dec(&tipc_queue_size);
-	}
+	discard_rx_queue(sk);
 
-	mutex_unlock(&tsock->lock);
+	/* Reject any messages that accumulated in backlog queue */
+
+	sock->state = SS_DISCONNECTING;
+	release_sock(sk);
 
 	sock_put(sk);
+	sock->sk = NULL;
 
 	atomic_dec(&tipc_user_count);
 	return res;
@@ -269,47 +342,32 @@ static int release(struct socket *sock)
  * (i.e. a socket address length of 0) unbinds all names from the socket.
  *
  * Returns 0 on success, errno otherwise
+ *
+ * NOTE: This routine doesn't need to take the socket lock since it doesn't
+ *       access any non-constant socket information.
  */
 
 static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
 	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
-	int res;
+	u32 portref = tipc_sk_port(sock->sk)->ref;
 
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
+	if (unlikely(!uaddr_len))
+		return tipc_withdraw(portref, 0, NULL);
 
-	if (unlikely(!uaddr_len)) {
-		res = tipc_withdraw(tsock->p->ref, 0, NULL);
-		goto exit;
-	}
-
-	if (uaddr_len < sizeof(struct sockaddr_tipc)) {
-		res = -EINVAL;
-		goto exit;
-	}
+	if (uaddr_len < sizeof(struct sockaddr_tipc))
+		return -EINVAL;
+	if (addr->family != AF_TIPC)
+		return -EAFNOSUPPORT;
 
-	if (addr->family != AF_TIPC) {
-		res = -EAFNOSUPPORT;
-		goto exit;
-	}
 	if (addr->addrtype == TIPC_ADDR_NAME)
 		addr->addr.nameseq.upper = addr->addr.nameseq.lower;
-	else if (addr->addrtype != TIPC_ADDR_NAMESEQ) {
-		res = -EAFNOSUPPORT;
-		goto exit;
-	}
+	else if (addr->addrtype != TIPC_ADDR_NAMESEQ)
+		return -EAFNOSUPPORT;
 
-	if (addr->scope > 0)
-		res = tipc_publish(tsock->p->ref, addr->scope,
-				   &addr->addr.nameseq);
-	else
-		res = tipc_withdraw(tsock->p->ref, -addr->scope,
-				    &addr->addr.nameseq);
-exit:
-	mutex_unlock(&tsock->lock);
-	return res;
+	return (addr->scope > 0) ?
+		tipc_publish(portref, addr->scope, &addr->addr.nameseq) :
+		tipc_withdraw(portref, -addr->scope, &addr->addr.nameseq);
 }
 
 /**
@@ -320,30 +378,33 @@ exit:
  * @peer: 0 to obtain socket name, 1 to obtain peer socket name
  *
  * Returns 0 on success, errno otherwise
+ *
+ * NOTE: This routine doesn't need to take the socket lock since it doesn't
+ *       access any non-constant socket information.
  */
 
 static int get_name(struct socket *sock, struct sockaddr *uaddr,
 		    int *uaddr_len, int peer)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
 	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
+	u32 portref = tipc_sk_port(sock->sk)->ref;
 	u32 res;
 
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
+	if (peer) {
+		res = tipc_peer(portref, &addr->addr.id);
+		if (res)
+			return res;
+	} else {
+		tipc_ownidentity(portref, &addr->addr.id);
+	}
 
 	*uaddr_len = sizeof(*addr);
 	addr->addrtype = TIPC_ADDR_ID;
 	addr->family = AF_TIPC;
 	addr->scope = 0;
-	if (peer)
-		res = tipc_peer(tsock->p->ref, &addr->addr.id);
-	else
-		res = tipc_ownidentity(tsock->p->ref, &addr->addr.id);
 	addr->addr.name.domain = 0;
 
-	mutex_unlock(&tsock->lock);
-	return res;
+	return 0;
 }
 
 /**
@@ -414,7 +475,6 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m)
 		return 0;
 	if (likely(dest->addr.name.name.type == TIPC_TOP_SRV))
 		return 0;
-
 	if (likely(dest->addr.name.name.type != TIPC_CFG_SRV))
 		return -EACCES;
 
@@ -428,7 +488,7 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m)
 
 /**
  * send_msg - send message in connectionless manner
- * @iocb: (unused)
+ * @iocb: if NULL, indicates that socket lock is already held
  * @sock: socket structure
  * @m: message to send
  * @total_len: length of message
@@ -444,9 +504,9 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m)
 static int send_msg(struct kiocb *iocb, struct socket *sock,
 		    struct msghdr *m, size_t total_len)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name;
-	struct sk_buff *buf;
 	int needs_conn;
 	int res = -EINVAL;
 
@@ -456,48 +516,46 @@ static int send_msg(struct kiocb *iocb, struct socket *sock,
 		     (dest->family != AF_TIPC)))
 		return -EINVAL;
 
+	if (iocb)
+		lock_sock(sk);
+
 	needs_conn = (sock->state != SS_READY);
 	if (unlikely(needs_conn)) {
-		if (sock->state == SS_LISTENING)
-			return -EPIPE;
-		if (sock->state != SS_UNCONNECTED)
-			return -EISCONN;
-		if ((tsock->p->published) ||
-		    ((sock->type == SOCK_STREAM) && (total_len != 0)))
-			return -EOPNOTSUPP;
+		if (sock->state == SS_LISTENING) {
+			res = -EPIPE;
+			goto exit;
+		}
+		if (sock->state != SS_UNCONNECTED) {
+			res = -EISCONN;
+			goto exit;
+		}
+		if ((tport->published) ||
+		    ((sock->type == SOCK_STREAM) && (total_len != 0))) {
+			res = -EOPNOTSUPP;
+			goto exit;
+		}
 		if (dest->addrtype == TIPC_ADDR_NAME) {
-			tsock->p->conn_type = dest->addr.name.name.type;
-			tsock->p->conn_instance = dest->addr.name.name.instance;
+			tport->conn_type = dest->addr.name.name.type;
+			tport->conn_instance = dest->addr.name.name.instance;
 		}
-	}
-
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
-
-	if (needs_conn) {
 
 		/* Abort any pending connection attempts (very unlikely) */
 
-		while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) {
-			tipc_reject_msg(buf, TIPC_ERR_NO_PORT);
-			atomic_dec(&tipc_queue_size);
-		}
-
-		sock->state = SS_CONNECTING;
+		reject_rx_queue(sk);
 	}
 
 	do {
 		if (dest->addrtype == TIPC_ADDR_NAME) {
 			if ((res = dest_name_check(dest, m)))
-				goto exit;
-			res = tipc_send2name(tsock->p->ref,
+				break;
+			res = tipc_send2name(tport->ref,
 					     &dest->addr.name.name,
 					     dest->addr.name.domain,
 					     m->msg_iovlen,
 					     m->msg_iov);
 		}
 		else if (dest->addrtype == TIPC_ADDR_ID) {
-			res = tipc_send2port(tsock->p->ref,
+			res = tipc_send2port(tport->ref,
 					     &dest->addr.id,
 					     m->msg_iovlen,
 					     m->msg_iov);
@@ -505,36 +563,43 @@ static int send_msg(struct kiocb *iocb, struct socket *sock,
 		else if (dest->addrtype == TIPC_ADDR_MCAST) {
 			if (needs_conn) {
 				res = -EOPNOTSUPP;
-				goto exit;
+				break;
 			}
 			if ((res = dest_name_check(dest, m)))
-				goto exit;
-			res = tipc_multicast(tsock->p->ref,
+				break;
+			res = tipc_multicast(tport->ref,
 					     &dest->addr.nameseq,
 					     0,
 					     m->msg_iovlen,
 					     m->msg_iov);
 		}
 		if (likely(res != -ELINKCONG)) {
-exit:
-			mutex_unlock(&tsock->lock);
-			return res;
+			if (needs_conn && (res >= 0)) {
+				sock->state = SS_CONNECTING;
+			}
+			break;
 		}
 		if (m->msg_flags & MSG_DONTWAIT) {
 			res = -EWOULDBLOCK;
-			goto exit;
-		}
-		if (wait_event_interruptible(*sock->sk->sk_sleep,
-					     !tsock->p->congested)) {
-		    res = -ERESTARTSYS;
-		    goto exit;
+			break;
 		}
+		release_sock(sk);
+		res = wait_event_interruptible(*sk->sk_sleep,
+					       !tport->congested);
+		lock_sock(sk);
+		if (res)
+			break;
 	} while (1);
+
+exit:
+	if (iocb)
+		release_sock(sk);
+	return res;
 }
 
 /**
  * send_packet - send a connection-oriented message
- * @iocb: (unused)
+ * @iocb: if NULL, indicates that socket lock is already held
  * @sock: socket structure
  * @m: message to send
  * @total_len: length of message
@@ -547,7 +612,8 @@ exit:
 static int send_packet(struct kiocb *iocb, struct socket *sock,
 		       struct msghdr *m, size_t total_len)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name;
 	int res;
 
@@ -556,9 +622,8 @@ static int send_packet(struct kiocb *iocb, struct socket *sock,
 	if (unlikely(dest))
 		return send_msg(iocb, sock, m, total_len);
 
-	if (mutex_lock_interruptible(&tsock->lock)) {
-		return -ERESTARTSYS;
-	}
+	if (iocb)
+		lock_sock(sk);
 
 	do {
 		if (unlikely(sock->state != SS_CONNECTED)) {
@@ -566,25 +631,28 @@ static int send_packet(struct kiocb *iocb, struct socket *sock,
 				res = -EPIPE;
 			else
 				res = -ENOTCONN;
-			goto exit;
+			break;
 		}
 
-		res = tipc_send(tsock->p->ref, m->msg_iovlen, m->msg_iov);
+		res = tipc_send(tport->ref, m->msg_iovlen, m->msg_iov);
 		if (likely(res != -ELINKCONG)) {
-exit:
-			mutex_unlock(&tsock->lock);
-			return res;
+			break;
 		}
 		if (m->msg_flags & MSG_DONTWAIT) {
 			res = -EWOULDBLOCK;
-			goto exit;
-		}
-		if (wait_event_interruptible(*sock->sk->sk_sleep,
-					     !tsock->p->congested)) {
-		    res = -ERESTARTSYS;
-		    goto exit;
+			break;
 		}
+		release_sock(sk);
+		res = wait_event_interruptible(*sk->sk_sleep,
+			(!tport->congested || !tport->connected));
+		lock_sock(sk);
+		if (res)
+			break;
 	} while (1);
+
+	if (iocb)
+		release_sock(sk);
+	return res;
 }
 
 /**
@@ -600,11 +668,11 @@ exit:
  * or errno if no data sent
  */
 
-
 static int send_stream(struct kiocb *iocb, struct socket *sock,
 		       struct msghdr *m, size_t total_len)
 {
-	struct tipc_port *tport;
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	struct msghdr my_msg;
 	struct iovec my_iov;
 	struct iovec *curr_iov;
@@ -616,19 +684,27 @@ static int send_stream(struct kiocb *iocb, struct socket *sock,
 	int bytes_sent;
 	int res;
 
+	lock_sock(sk);
+
 	/* Handle special cases where there is no connection */
 
 	if (unlikely(sock->state != SS_CONNECTED)) {
-		if (sock->state == SS_UNCONNECTED)
-			return send_packet(iocb, sock, m, total_len);
-		else if (sock->state == SS_DISCONNECTING)
-			return -EPIPE;
-		else
-			return -ENOTCONN;
+		if (sock->state == SS_UNCONNECTED) {
+			res = send_packet(NULL, sock, m, total_len);
+			goto exit;
+		} else if (sock->state == SS_DISCONNECTING) {
+			res = -EPIPE;
+			goto exit;
+		} else {
+			res = -ENOTCONN;
+			goto exit;
+		}
 	}
 
-	if (unlikely(m->msg_name))
-		return -EISCONN;
+	if (unlikely(m->msg_name)) {
+		res = -EISCONN;
+		goto exit;
+	}
 
 	/*
 	 * Send each iovec entry using one or more messages
@@ -646,7 +722,6 @@ static int send_stream(struct kiocb *iocb, struct socket *sock,
 	my_msg.msg_name = NULL;
 	bytes_sent = 0;
 
-	tport = tipc_sk(sock->sk)->p;
 	hdr_size = msg_hdr_sz(&tport->phdr);
 
 	while (curr_iovlen--) {
@@ -661,10 +736,10 @@ static int send_stream(struct kiocb *iocb, struct socket *sock,
 				bytes_to_send = curr_left;
 			my_iov.iov_base = curr_start;
 			my_iov.iov_len = bytes_to_send;
-			if ((res = send_packet(iocb, sock, &my_msg, 0)) < 0) {
-				if (bytes_sent != 0)
+			if ((res = send_packet(NULL, sock, &my_msg, 0)) < 0) {
+				if (bytes_sent)
 					res = bytes_sent;
-				return res;
+				goto exit;
 			}
 			curr_left -= bytes_to_send;
 			curr_start += bytes_to_send;
@@ -673,22 +748,23 @@ static int send_stream(struct kiocb *iocb, struct socket *sock,
 
 		curr_iov++;
 	}
-
-	return bytes_sent;
+	res = bytes_sent;
+exit:
+	release_sock(sk);
+	return res;
 }
 
 /**
  * auto_connect - complete connection setup to a remote port
  * @sock: socket structure
- * @tsock: TIPC-specific socket structure
  * @msg: peer's response message
  *
  * Returns 0 on success, errno otherwise
  */
 
-static int auto_connect(struct socket *sock, struct tipc_sock *tsock,
-			struct tipc_msg *msg)
+static int auto_connect(struct socket *sock, struct tipc_msg *msg)
 {
+	struct tipc_port *tport = tipc_sk_port(sock->sk);
 	struct tipc_portid peer;
 
 	if (msg_errcode(msg)) {
@@ -698,8 +774,8 @@ static int auto_connect(struct socket *sock, struct tipc_sock *tsock,
 
 	peer.ref = msg_origport(msg);
 	peer.node = msg_orignode(msg);
-	tipc_connect2port(tsock->p->ref, &peer);
-	tipc_set_portimportance(tsock->p->ref, msg_importance(msg));
+	tipc_connect2port(tport->ref, &peer);
+	tipc_set_portimportance(tport->ref, msg_importance(msg));
 	sock->state = SS_CONNECTED;
 	return 0;
 }
@@ -812,62 +888,54 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg,
 static int recv_msg(struct kiocb *iocb, struct socket *sock,
 		    struct msghdr *m, size_t buf_len, int flags)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	struct sk_buff *buf;
 	struct tipc_msg *msg;
-	unsigned int q_len;
 	unsigned int sz;
 	u32 err;
 	int res;
 
-	/* Currently doesn't support receiving into multiple iovec entries */
+	/* Catch invalid receive requests */
 
 	if (m->msg_iovlen != 1)
-		return -EOPNOTSUPP;
-
-	/* Catch invalid receive attempts */
+		return -EOPNOTSUPP;   /* Don't do multiple iovec entries yet */
 
 	if (unlikely(!buf_len))
 		return -EINVAL;
 
-	if (sock->type == SOCK_SEQPACKET) {
-		if (unlikely(sock->state == SS_UNCONNECTED))
-			return -ENOTCONN;
-		if (unlikely((sock->state == SS_DISCONNECTING) &&
-			     (skb_queue_len(&sock->sk->sk_receive_queue) == 0)))
-			return -ENOTCONN;
-	}
+	lock_sock(sk);
 
-	/* Look for a message in receive queue; wait if necessary */
-
-	if (unlikely(mutex_lock_interruptible(&tsock->lock)))
-		return -ERESTARTSYS;
-
-restart:
-	if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) &&
-		     (flags & MSG_DONTWAIT))) {
-		res = -EWOULDBLOCK;
+	if (unlikely(sock->state == SS_UNCONNECTED)) {
+		res = -ENOTCONN;
 		goto exit;
 	}
 
-	if ((res = wait_event_interruptible(
-		*sock->sk->sk_sleep,
-		((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) ||
-		 (sock->state == SS_DISCONNECTING))) )) {
-		goto exit;
-	}
+restart:
 
-	/* Catch attempt to receive on an already terminated connection */
-	/* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */
+	/* Look for a message in receive queue; wait if necessary */
 
-	if (!q_len) {
-		res = -ENOTCONN;
-		goto exit;
+	while (skb_queue_empty(&sk->sk_receive_queue)) {
+		if (sock->state == SS_DISCONNECTING) {
+			res = -ENOTCONN;
+			goto exit;
+		}
+		if (flags & MSG_DONTWAIT) {
+			res = -EWOULDBLOCK;
+			goto exit;
+		}
+		release_sock(sk);
+		res = wait_event_interruptible(*sk->sk_sleep,
+			(!skb_queue_empty(&sk->sk_receive_queue) ||
+			 (sock->state == SS_DISCONNECTING)));
+		lock_sock(sk);
+		if (res)
+			goto exit;
 	}
 
-	/* Get access to first message in receive queue */
+	/* Look at first message in receive queue */
 
-	buf = skb_peek(&sock->sk->sk_receive_queue);
+	buf = skb_peek(&sk->sk_receive_queue);
 	msg = buf_msg(buf);
 	sz = msg_data_sz(msg);
 	err = msg_errcode(msg);
@@ -875,14 +943,15 @@ restart:
 	/* Complete connection setup for an implied connect */
 
 	if (unlikely(sock->state == SS_CONNECTING)) {
-		if ((res = auto_connect(sock, tsock, msg)))
+		res = auto_connect(sock, msg);
+		if (res)
 			goto exit;
 	}
 
 	/* Discard an empty non-errored message & try again */
 
 	if ((!sz) && (!err)) {
-		advance_queue(tsock);
+		advance_rx_queue(sk);
 		goto restart;
 	}
 
@@ -892,7 +961,8 @@ restart:
 
 	/* Capture ancillary data (optional) */
 
-	if ((res = anc_data_recv(m, msg, tsock->p)))
+	res = anc_data_recv(m, msg, tport);
+	if (res)
 		goto exit;
 
 	/* Capture message data (if valid) & compute return value (always) */
@@ -920,12 +990,12 @@ restart:
 
 	if (likely(!(flags & MSG_PEEK))) {
 		if ((sock->state != SS_READY) &&
-		    (++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
-			tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked);
-		advance_queue(tsock);
+		    (++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
+			tipc_acknowledge(tport->ref, tport->conn_unacked);
+		advance_rx_queue(sk);
 	}
 exit:
-	mutex_unlock(&tsock->lock);
+	release_sock(sk);
 	return res;
 }
 
@@ -945,10 +1015,10 @@ exit:
 static int recv_stream(struct kiocb *iocb, struct socket *sock,
 		       struct msghdr *m, size_t buf_len, int flags)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	struct sk_buff *buf;
 	struct tipc_msg *msg;
-	unsigned int q_len;
 	unsigned int sz;
 	int sz_to_copy;
 	int sz_copied = 0;
@@ -956,54 +1026,49 @@ static int recv_stream(struct kiocb *iocb, struct socket *sock,
 	char __user *crs = m->msg_iov->iov_base;
 	unsigned char *buf_crs;
 	u32 err;
-	int res;
+	int res = 0;
 
-	/* Currently doesn't support receiving into multiple iovec entries */
+	/* Catch invalid receive attempts */
 
 	if (m->msg_iovlen != 1)
-		return -EOPNOTSUPP;
-
-	/* Catch invalid receive attempts */
+		return -EOPNOTSUPP;   /* Don't do multiple iovec entries yet */
 
 	if (unlikely(!buf_len))
 		return -EINVAL;
 
-	if (unlikely(sock->state == SS_DISCONNECTING)) {
-		if (skb_queue_len(&sock->sk->sk_receive_queue) == 0)
-			return -ENOTCONN;
-	} else if (unlikely(sock->state != SS_CONNECTED))
-		return -ENOTCONN;
-
-	/* Look for a message in receive queue; wait if necessary */
-
-	if (unlikely(mutex_lock_interruptible(&tsock->lock)))
-		return -ERESTARTSYS;
+	lock_sock(sk);
 
-restart:
-	if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) &&
-		     (flags & MSG_DONTWAIT))) {
-		res = -EWOULDBLOCK;
+	if (unlikely((sock->state == SS_UNCONNECTED) ||
+		     (sock->state == SS_CONNECTING))) {
+		res = -ENOTCONN;
 		goto exit;
 	}
 
-	if ((res = wait_event_interruptible(
-		*sock->sk->sk_sleep,
-		((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) ||
-		 (sock->state == SS_DISCONNECTING))) )) {
-		goto exit;
-	}
+restart:
 
-	/* Catch attempt to receive on an already terminated connection */
-	/* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */
+	/* Look for a message in receive queue; wait if necessary */
 
-	if (!q_len) {
-		res = -ENOTCONN;
-		goto exit;
+	while (skb_queue_empty(&sk->sk_receive_queue)) {
+		if (sock->state == SS_DISCONNECTING) {
+			res = -ENOTCONN;
+			goto exit;
+		}
+		if (flags & MSG_DONTWAIT) {
+			res = -EWOULDBLOCK;
+			goto exit;
+		}
+		release_sock(sk);
+		res = wait_event_interruptible(*sk->sk_sleep,
+			(!skb_queue_empty(&sk->sk_receive_queue) ||
+			 (sock->state == SS_DISCONNECTING)));
+		lock_sock(sk);
+		if (res)
+			goto exit;
 	}
 
-	/* Get access to first message in receive queue */
+	/* Look at first message in receive queue */
 
-	buf = skb_peek(&sock->sk->sk_receive_queue);
+	buf = skb_peek(&sk->sk_receive_queue);
 	msg = buf_msg(buf);
 	sz = msg_data_sz(msg);
 	err = msg_errcode(msg);
@@ -1011,7 +1076,7 @@ restart:
 	/* Discard an empty non-errored message & try again */
 
 	if ((!sz) && (!err)) {
-		advance_queue(tsock);
+		advance_rx_queue(sk);
 		goto restart;
 	}
 
@@ -1019,7 +1084,8 @@ restart:
 
 	if (sz_copied == 0) {
 		set_orig_addr(m, msg);
-		if ((res = anc_data_recv(m, msg, tsock->p)))
+		res = anc_data_recv(m, msg, tport);
+		if (res)
 			goto exit;
 	}
 
@@ -1057,9 +1123,9 @@ restart:
 	/* Consume received message (optional) */
 
 	if (likely(!(flags & MSG_PEEK))) {
-		if (unlikely(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
-			tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked);
-		advance_queue(tsock);
+		if (unlikely(++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
+			tipc_acknowledge(tport->ref, tport->conn_unacked);
+		advance_rx_queue(sk);
 	}
 
 	/* Loop around if more data is required */
@@ -1074,7 +1140,7 @@ restart:
 		goto restart;
 
 exit:
-	mutex_unlock(&tsock->lock);
+	release_sock(sk);
 	return sz_copied ? sz_copied : res;
 }
 
@@ -1108,37 +1174,24 @@ static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base)
 }
 
 /**
- * async_disconnect - wrapper function used to disconnect port
- * @portref: TIPC port reference (passed as pointer-sized value)
- */
-
-static void async_disconnect(unsigned long portref)
-{
-	tipc_disconnect((u32)portref);
-}
-
-/**
- * dispatch - handle arriving message
- * @tport: TIPC port that received message
+ * filter_rcv - validate incoming message
+ * @sk: socket
  * @buf: message
  *
- * Called with port locked.  Must not take socket lock to avoid deadlock risk.
+ * Enqueues message on receive queue if acceptable; optionally handles
+ * disconnect indication for a connected socket.
+ *
+ * Called with socket lock already taken; port lock may also be taken.
  *
  * Returns TIPC error status code (TIPC_OK if message is not to be rejected)
  */
 
-static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf)
+static u32 filter_rcv(struct sock *sk, struct sk_buff *buf)
 {
+	struct socket *sock = sk->sk_socket;
 	struct tipc_msg *msg = buf_msg(buf);
-	struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle;
-	struct socket *sock;
 	u32 recv_q_len;
 
-	/* Reject message if socket is closing */
-
-	if (!tsock)
-		return TIPC_ERR_NO_PORT;
-
 	/* Reject message if it is wrong sort of message for socket */
 
 	/*
@@ -1146,7 +1199,7 @@ static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf)
 	 * "NO PORT" ISN'T REALLY THE RIGHT ERROR CODE, AND THERE MAY
 	 * BE SECURITY IMPLICATIONS INHERENT IN REJECTING INVALID TRAFFIC
 	 */
-	sock = tsock->sk.sk_socket;
+
 	if (sock->state == SS_READY) {
 		if (msg_connected(msg)) {
 			msg_dbg(msg, "dispatch filter 1\n");
@@ -1194,45 +1247,98 @@ static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf)
 		if (rx_queue_full(msg, recv_q_len, OVERLOAD_LIMIT_BASE))
 			return TIPC_ERR_OVERLOAD;
 	}
-	recv_q_len = skb_queue_len(&tsock->sk.sk_receive_queue);
+	recv_q_len = skb_queue_len(&sk->sk_receive_queue);
 	if (unlikely(recv_q_len >= (OVERLOAD_LIMIT_BASE / 2))) {
 		if (rx_queue_full(msg, recv_q_len, OVERLOAD_LIMIT_BASE / 2))
 			return TIPC_ERR_OVERLOAD;
 	}
 
+	/* Enqueue message (finally!) */
+
+	msg_dbg(msg, "<DISP<: ");
+	TIPC_SKB_CB(buf)->handle = msg_data(msg);
+	atomic_inc(&tipc_queue_size);
+	__skb_queue_tail(&sk->sk_receive_queue, buf);
+
 	/* Initiate connection termination for an incoming 'FIN' */
 
 	if (unlikely(msg_errcode(msg) && (sock->state == SS_CONNECTED))) {
 		sock->state = SS_DISCONNECTING;
-		/* Note: Use signal since port lock is already taken! */
-		tipc_k_signal((Handler)async_disconnect, tport->ref);
+		tipc_disconnect_port(tipc_sk_port(sk));
 	}
 
-	/* Enqueue message (finally!) */
+	if (waitqueue_active(sk->sk_sleep))
+		wake_up_interruptible(sk->sk_sleep);
+	return TIPC_OK;
+}
 
-	msg_dbg(msg,"<DISP<: ");
-	TIPC_SKB_CB(buf)->handle = msg_data(msg);
-	atomic_inc(&tipc_queue_size);
-	skb_queue_tail(&sock->sk->sk_receive_queue, buf);
+/**
+ * backlog_rcv - handle incoming message from backlog queue
+ * @sk: socket
+ * @buf: message
+ *
+ * Caller must hold socket lock, but not port lock.
+ *
+ * Returns 0
+ */
 
-	if (waitqueue_active(sock->sk->sk_sleep))
-		wake_up_interruptible(sock->sk->sk_sleep);
-	return TIPC_OK;
+static int backlog_rcv(struct sock *sk, struct sk_buff *buf)
+{
+	u32 res;
+
+	res = filter_rcv(sk, buf);
+	if (res)
+		tipc_reject_msg(buf, res);
+	return 0;
+}
+
+/**
+ * dispatch - handle incoming message
+ * @tport: TIPC port that received message
+ * @buf: message
+ *
+ * Called with port lock already taken.
+ *
+ * Returns TIPC error status code (TIPC_OK if message is not to be rejected)
+ */
+
+static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf)
+{
+	struct sock *sk = (struct sock *)tport->usr_handle;
+	u32 res;
+
+	/*
+	 * Process message if socket is unlocked; otherwise add to backlog queue
+	 *
+	 * This code is based on sk_receive_skb(), but must be distinct from it
+	 * since a TIPC-specific filter/reject mechanism is utilized
+	 */
+
+	bh_lock_sock(sk);
+	if (!sock_owned_by_user(sk)) {
+		res = filter_rcv(sk, buf);
+	} else {
+		sk_add_backlog(sk, buf);
+		res = TIPC_OK;
+	}
+	bh_unlock_sock(sk);
+
+	return res;
 }
 
 /**
  * wakeupdispatch - wake up port after congestion
  * @tport: port to wakeup
  *
- * Called with port lock on.
+ * Called with port lock already taken.
  */
 
 static void wakeupdispatch(struct tipc_port *tport)
 {
-	struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle;
+	struct sock *sk = (struct sock *)tport->usr_handle;
 
-	if (waitqueue_active(tsock->sk.sk_sleep))
-		wake_up_interruptible(tsock->sk.sk_sleep);
+	if (waitqueue_active(sk->sk_sleep))
+		wake_up_interruptible(sk->sk_sleep);
 }
 
 /**
@@ -1240,7 +1346,7 @@ static void wakeupdispatch(struct tipc_port *tport)
  * @sock: socket structure
  * @dest: socket address for destination port
  * @destlen: size of socket address data structure
- * @flags: (unused)
+ * @flags: file-related flags associated with socket
  *
  * Returns 0 on success, errno otherwise
  */
@@ -1248,31 +1354,43 @@ static void wakeupdispatch(struct tipc_port *tport)
 static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 		   int flags)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
 	struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest;
 	struct msghdr m = {NULL,};
 	struct sk_buff *buf;
 	struct tipc_msg *msg;
 	int res;
 
+	lock_sock(sk);
+
 	/* For now, TIPC does not allow use of connect() with DGRAM/RDM types */
 
-	if (sock->state == SS_READY)
-		return -EOPNOTSUPP;
+	if (sock->state == SS_READY) {
+		res = -EOPNOTSUPP;
+		goto exit;
+	}
 
 	/* For now, TIPC does not support the non-blocking form of connect() */
 
-	if (flags & O_NONBLOCK)
-		return -EWOULDBLOCK;
+	if (flags & O_NONBLOCK) {
+		res = -EWOULDBLOCK;
+		goto exit;
+	}
 
 	/* Issue Posix-compliant error code if socket is in the wrong state */
 
-	if (sock->state == SS_LISTENING)
-		return -EOPNOTSUPP;
-	if (sock->state == SS_CONNECTING)
-		return -EALREADY;
-	if (sock->state != SS_UNCONNECTED)
-		return -EISCONN;
+	if (sock->state == SS_LISTENING) {
+		res = -EOPNOTSUPP;
+		goto exit;
+	}
+	if (sock->state == SS_CONNECTING) {
+		res = -EALREADY;
+		goto exit;
+	}
+	if (sock->state != SS_UNCONNECTED) {
+		res = -EISCONN;
+		goto exit;
+	}
 
 	/*
 	 * Reject connection attempt using multicast address
@@ -1281,8 +1399,14 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 	 *       so there's no need to do it here
 	 */
 
-	if (dst->addrtype == TIPC_ADDR_MCAST)
-		return -EINVAL;
+	if (dst->addrtype == TIPC_ADDR_MCAST) {
+		res = -EINVAL;
+		goto exit;
+	}
+
+	/* Reject any messages already in receive queue (very unlikely) */
+
+	reject_rx_queue(sk);
 
 	/* Send a 'SYN-' to destination */
 
@@ -1290,25 +1414,33 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 	m.msg_namelen = destlen;
 	res = send_msg(NULL, sock, &m, 0);
 	if (res < 0) {
-		sock->state = SS_DISCONNECTING;
-		return res;
+		goto exit;
 	}
 
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
+	/* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */
 
-	/* Wait for destination's 'ACK' response */
+	release_sock(sk);
+	res = wait_event_interruptible_timeout(*sk->sk_sleep,
+			(!skb_queue_empty(&sk->sk_receive_queue) ||
+			(sock->state != SS_CONNECTING)),
+			sk->sk_rcvtimeo);
+	lock_sock(sk);
 
-	res = wait_event_interruptible_timeout(*sock->sk->sk_sleep,
-			skb_queue_len(&sock->sk->sk_receive_queue),
-			sock->sk->sk_rcvtimeo);
-	buf = skb_peek(&sock->sk->sk_receive_queue);
 	if (res > 0) {
-		msg = buf_msg(buf);
-		res = auto_connect(sock, tsock, msg);
-		if (!res) {
-			if (!msg_data_sz(msg))
-				advance_queue(tsock);
+		buf = skb_peek(&sk->sk_receive_queue);
+		if (buf != NULL) {
+			msg = buf_msg(buf);
+			res = auto_connect(sock, msg);
+			if (!res) {
+				if (!msg_data_sz(msg))
+					advance_rx_queue(sk);
+			}
+		} else {
+			if (sock->state == SS_CONNECTED) {
+				res = -EISCONN;
+			} else {
+				res = -ECONNREFUSED;
+			}
 		}
 	} else {
 		if (res == 0)
@@ -1318,7 +1450,8 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 		sock->state = SS_DISCONNECTING;
 	}
 
-	mutex_unlock(&tsock->lock);
+exit:
+	release_sock(sk);
 	return res;
 }
 
@@ -1332,14 +1465,22 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 
 static int listen(struct socket *sock, int len)
 {
-	/* REQUIRES SOCKET LOCKING OF SOME SORT? */
+	struct sock *sk = sock->sk;
+	int res;
+
+	lock_sock(sk);
 
 	if (sock->state == SS_READY)
-		return -EOPNOTSUPP;
-	if (sock->state != SS_UNCONNECTED)
-		return -EINVAL;
-	sock->state = SS_LISTENING;
-	return 0;
+		res = -EOPNOTSUPP;
+	else if (sock->state != SS_UNCONNECTED)
+		res = -EINVAL;
+	else {
+		sock->state = SS_LISTENING;
+		res = 0;
+	}
+
+	release_sock(sk);
+	return res;
 }
 
 /**
@@ -1351,50 +1492,69 @@ static int listen(struct socket *sock, int len)
  * Returns 0 on success, errno otherwise
  */
 
-static int accept(struct socket *sock, struct socket *newsock, int flags)
+static int accept(struct socket *sock, struct socket *new_sock, int flags)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
 	struct sk_buff *buf;
-	int res = -EFAULT;
-
-	if (sock->state == SS_READY)
-		return -EOPNOTSUPP;
-	if (sock->state != SS_LISTENING)
-		return -EINVAL;
-
-	if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) &&
-		     (flags & O_NONBLOCK)))
-		return -EWOULDBLOCK;
+	int res;
 
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
+	lock_sock(sk);
 
-	if (wait_event_interruptible(*sock->sk->sk_sleep,
-				     skb_queue_len(&sock->sk->sk_receive_queue))) {
-		res = -ERESTARTSYS;
+	if (sock->state == SS_READY) {
+		res = -EOPNOTSUPP;
+		goto exit;
+	}
+	if (sock->state != SS_LISTENING) {
+		res = -EINVAL;
 		goto exit;
 	}
-	buf = skb_peek(&sock->sk->sk_receive_queue);
 
-	res = tipc_create(sock_net(sock->sk), newsock, 0);
+	while (skb_queue_empty(&sk->sk_receive_queue)) {
+		if (flags & O_NONBLOCK) {
+			res = -EWOULDBLOCK;
+			goto exit;
+		}
+		release_sock(sk);
+		res = wait_event_interruptible(*sk->sk_sleep,
+				(!skb_queue_empty(&sk->sk_receive_queue)));
+		lock_sock(sk);
+		if (res)
+			goto exit;
+	}
+
+	buf = skb_peek(&sk->sk_receive_queue);
+
+	res = tipc_create(sock_net(sock->sk), new_sock, 0);
 	if (!res) {
-		struct tipc_sock *new_tsock = tipc_sk(newsock->sk);
+		struct sock *new_sk = new_sock->sk;
+		struct tipc_port *new_tport = tipc_sk_port(new_sk);
+		u32 new_ref = new_tport->ref;
 		struct tipc_portid id;
 		struct tipc_msg *msg = buf_msg(buf);
-		u32 new_ref = new_tsock->p->ref;
+
+		lock_sock(new_sk);
+
+		/*
+		 * Reject any stray messages received by new socket
+		 * before the socket lock was taken (very, very unlikely)
+		 */
+
+		reject_rx_queue(new_sk);
+
+		/* Connect new socket to it's peer */
 
 		id.ref = msg_origport(msg);
 		id.node = msg_orignode(msg);
 		tipc_connect2port(new_ref, &id);
-		newsock->state = SS_CONNECTED;
+		new_sock->state = SS_CONNECTED;
 
 		tipc_set_portimportance(new_ref, msg_importance(msg));
 		if (msg_named(msg)) {
-			new_tsock->p->conn_type = msg_nametype(msg);
-			new_tsock->p->conn_instance = msg_nameinst(msg);
+			new_tport->conn_type = msg_nametype(msg);
+			new_tport->conn_instance = msg_nameinst(msg);
 		}
 
-	       /*
+		/*
 		 * Respond to 'SYN-' by discarding it & returning 'ACK'-.
 		 * Respond to 'SYN+' by queuing it on new socket.
 		 */
@@ -1403,17 +1563,16 @@ static int accept(struct socket *sock, struct socket *newsock, int flags)
 		if (!msg_data_sz(msg)) {
 			struct msghdr m = {NULL,};
 
-			send_packet(NULL, newsock, &m, 0);
-			advance_queue(tsock);
+			advance_rx_queue(sk);
+			send_packet(NULL, new_sock, &m, 0);
 		} else {
-			sock_lock(tsock);
-			skb_dequeue(&sock->sk->sk_receive_queue);
-			sock_unlock(tsock);
-			skb_queue_head(&newsock->sk->sk_receive_queue, buf);
+			__skb_dequeue(&sk->sk_receive_queue);
+			__skb_queue_head(&new_sk->sk_receive_queue, buf);
 		}
+		release_sock(new_sk);
 	}
 exit:
-	mutex_unlock(&tsock->lock);
+	release_sock(sk);
 	return res;
 }
 
@@ -1429,54 +1588,46 @@ exit:
 
 static int shutdown(struct socket *sock, int how)
 {
-	struct tipc_sock* tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	struct sk_buff *buf;
 	int res;
 
 	if (how != SHUT_RDWR)
 		return -EINVAL;
 
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
-
-	sock_lock(tsock);
+	lock_sock(sk);
 
 	switch (sock->state) {
+	case SS_CONNECTING:
 	case SS_CONNECTED:
 
-		/* Send 'FIN+' or 'FIN-' message to peer */
-
-		sock_unlock(tsock);
+		/* Disconnect and send a 'FIN+' or 'FIN-' message to peer */
 restart:
-		if ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) {
+		buf = __skb_dequeue(&sk->sk_receive_queue);
+		if (buf) {
 			atomic_dec(&tipc_queue_size);
 			if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) {
 				buf_discard(buf);
 				goto restart;
 			}
+			tipc_disconnect(tport->ref);
 			tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN);
+		} else {
+			tipc_shutdown(tport->ref);
 		}
-		else {
-			tipc_shutdown(tsock->p->ref);
-		}
-		sock_lock(tsock);
+
+		sock->state = SS_DISCONNECTING;
 
 		/* fall through */
 
 	case SS_DISCONNECTING:
 
-		/* Discard any unreceived messages */
-
-		while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) {
-			atomic_dec(&tipc_queue_size);
-			buf_discard(buf);
-		}
-		tsock->p->conn_unacked = 0;
+		/* Discard any unreceived messages; wake up sleeping tasks */
 
-		/* fall through */
-
-	case SS_CONNECTING:
-		sock->state = SS_DISCONNECTING;
+		discard_rx_queue(sk);
+		if (waitqueue_active(sk->sk_sleep))
+			wake_up_interruptible(sk->sk_sleep);
 		res = 0;
 		break;
 
@@ -1484,9 +1635,7 @@ restart:
 		res = -ENOTCONN;
 	}
 
-	sock_unlock(tsock);
-
-	mutex_unlock(&tsock->lock);
+	release_sock(sk);
 	return res;
 }
 
@@ -1507,7 +1656,8 @@ restart:
 static int setsockopt(struct socket *sock,
 		      int lvl, int opt, char __user *ov, int ol)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	u32 value;
 	int res;
 
@@ -1520,30 +1670,31 @@ static int setsockopt(struct socket *sock,
 	if ((res = get_user(value, (u32 __user *)ov)))
 		return res;
 
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
+	lock_sock(sk);
 
 	switch (opt) {
 	case TIPC_IMPORTANCE:
-		res = tipc_set_portimportance(tsock->p->ref, value);
+		res = tipc_set_portimportance(tport->ref, value);
 		break;
 	case TIPC_SRC_DROPPABLE:
 		if (sock->type != SOCK_STREAM)
-			res = tipc_set_portunreliable(tsock->p->ref, value);
+			res = tipc_set_portunreliable(tport->ref, value);
 		else
 			res = -ENOPROTOOPT;
 		break;
 	case TIPC_DEST_DROPPABLE:
-		res = tipc_set_portunreturnable(tsock->p->ref, value);
+		res = tipc_set_portunreturnable(tport->ref, value);
 		break;
 	case TIPC_CONN_TIMEOUT:
-		sock->sk->sk_rcvtimeo = msecs_to_jiffies(value);
+		sk->sk_rcvtimeo = msecs_to_jiffies(value);
+		/* no need to set "res", since already 0 at this point */
 		break;
 	default:
 		res = -EINVAL;
 	}
 
-	mutex_unlock(&tsock->lock);
+	release_sock(sk);
+
 	return res;
 }
 
@@ -1564,7 +1715,8 @@ static int setsockopt(struct socket *sock,
 static int getsockopt(struct socket *sock,
 		      int lvl, int opt, char __user *ov, int __user *ol)
 {
-	struct tipc_sock *tsock = tipc_sk(sock->sk);
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
 	int len;
 	u32 value;
 	int res;
@@ -1576,26 +1728,28 @@ static int getsockopt(struct socket *sock,
 	if ((res = get_user(len, ol)))
 		return res;
 
-	if (mutex_lock_interruptible(&tsock->lock))
-		return -ERESTARTSYS;
+	lock_sock(sk);
 
 	switch (opt) {
 	case TIPC_IMPORTANCE:
-		res = tipc_portimportance(tsock->p->ref, &value);
+		res = tipc_portimportance(tport->ref, &value);
 		break;
 	case TIPC_SRC_DROPPABLE:
-		res = tipc_portunreliable(tsock->p->ref, &value);
+		res = tipc_portunreliable(tport->ref, &value);
 		break;
 	case TIPC_DEST_DROPPABLE:
-		res = tipc_portunreturnable(tsock->p->ref, &value);
+		res = tipc_portunreturnable(tport->ref, &value);
 		break;
 	case TIPC_CONN_TIMEOUT:
-		value = jiffies_to_msecs(sock->sk->sk_rcvtimeo);
+		value = jiffies_to_msecs(sk->sk_rcvtimeo);
+		/* no need to set "res", since already 0 at this point */
 		break;
 	default:
 		res = -EINVAL;
 	}
 
+	release_sock(sk);
+
 	if (res) {
 		/* "get" failed */
 	}
@@ -1609,7 +1763,6 @@ static int getsockopt(struct socket *sock,
 		res = put_user(sizeof(value), ol);
 	}
 
-	mutex_unlock(&tsock->lock);
 	return res;
 }
 
@@ -1722,6 +1875,7 @@ int tipc_socket_init(void)
 /**
  * tipc_socket_stop - stop TIPC socket interface
  */
+
 void tipc_socket_stop(void)
 {
 	if (!sockets_enabled)
-- 
cgit v1.2.3


From d237e5c7ce82b6c7e1aa50348520a0c5a47274fa Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 15 Apr 2008 09:25:33 +0200
Subject: io context: increment task attachment count in ioc_task_link()

Thanks to Nikanth Karthikesan <knikanth@suse.de> for reporting this.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/iocontext.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h
index cac4b364cd40..2b7a1187cb29 100644
--- a/include/linux/iocontext.h
+++ b/include/linux/iocontext.h
@@ -91,8 +91,10 @@ static inline struct io_context *ioc_task_link(struct io_context *ioc)
 	 * if ref count is zero, don't allow sharing (ioc is going away, it's
 	 * a race).
 	 */
-	if (ioc && atomic_inc_not_zero(&ioc->refcount))
+	if (ioc && atomic_inc_not_zero(&ioc->refcount)) {
+		atomic_inc(&ioc->nr_tasks);
 		return ioc;
+	}
 
 	return NULL;
 }
-- 
cgit v1.2.3


From 99971e70fdc1862e120f3319fc0a4dba8c728acf Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 15 Apr 2008 00:27:58 -0700
Subject: [WANPIPE]: Forgotten bits of Sangoma drivers removal.

Robert P. J. Day spotted that my removal of the Sangoma drivers missed
a few bits.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/00-INDEX       |   2 -
 Documentation/networking/wan-router.txt | 621 --------------------------------
 include/linux/Kbuild                    |   1 -
 include/linux/if_wanpipe.h              | 124 -------
 4 files changed, 748 deletions(-)
 delete mode 100644 Documentation/networking/wan-router.txt
 delete mode 100644 include/linux/if_wanpipe.h

(limited to 'include')

diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX
index c485ee028bd9..1634c6dcecae 100644
--- a/Documentation/networking/00-INDEX
+++ b/Documentation/networking/00-INDEX
@@ -100,8 +100,6 @@ tuntap.txt
 	- TUN/TAP device driver, allowing user space Rx/Tx of packets.
 vortex.txt
 	- info on using 3Com Vortex (3c590, 3c592, 3c595, 3c597) Ethernet cards.
-wan-router.txt
-	- WAN router documentation
 wavelan.txt
 	- AT&T GIS (nee NCR) WaveLAN card: An Ethernet-like radio transceiver
 x25.txt
diff --git a/Documentation/networking/wan-router.txt b/Documentation/networking/wan-router.txt
deleted file mode 100644
index bc2ab419a74a..000000000000
--- a/Documentation/networking/wan-router.txt
+++ /dev/null
@@ -1,621 +0,0 @@
-------------------------------------------------------------------------------
-Linux WAN Router Utilities Package
-------------------------------------------------------------------------------
-Version 2.2.1 
-Mar 28, 2001
-Author: Nenad Corbic <ncorbic@sangoma.com>
-Copyright (c) 1995-2001 Sangoma Technologies Inc.
-------------------------------------------------------------------------------
-
-INTRODUCTION
-
-Wide Area Networks (WANs) are used to interconnect Local Area Networks (LANs)
-and/or stand-alone hosts over vast distances with data transfer rates
-significantly higher than those achievable with commonly used dial-up
-connections.
-
-Usually an external device called `WAN router' sitting on your local network
-or connected to your machine's serial port provides physical connection to
-WAN.  Although router's job may be as simple as taking your local network
-traffic, converting it to WAN format and piping it through the WAN link, these
-devices are notoriously expensive, with prices as much as 2 - 5 times higher
-then the price of a typical PC box.
-
-Alternatively, considering robustness and multitasking capabilities of Linux,
-an internal router can be built (most routers use some sort of stripped down
-Unix-like operating system anyway). With a number of relatively inexpensive WAN
-interface cards available on the market, a perfectly usable router can be
-built for less than half a price of an external router.  Yet a Linux box
-acting as a router can still be used for other purposes, such as fire-walling,
-running FTP, WWW or DNS server, etc.
-
-This kernel module introduces the notion of a WAN Link Driver (WLD) to Linux
-operating system and provides generic hardware-independent services for such
-drivers.  Why can existing Linux network device interface not be used for
-this purpose?  Well, it can.  However, there are a few key differences between
-a typical network interface (e.g. Ethernet) and a WAN link.
-
-Many WAN protocols, such as X.25 and frame relay, allow for multiple logical
-connections (known as `virtual circuits' in X.25 terminology) over a single
-physical link.  Each such virtual circuit may (and almost always does) lead
-to a different geographical location and, therefore, different network.  As a
-result, it is the virtual circuit, not the physical link, that represents a
-route and, therefore, a network interface in Linux terms.
-
-To further complicate things, virtual circuits are usually volatile in nature
-(excluding so called `permanent' virtual circuits or PVCs).  With almost no
-time required to set up and tear down a virtual circuit, it is highly desirable
-to implement on-demand connections in order to minimize network charges.  So
-unlike a typical network driver, the WAN driver must be able to handle multiple
-network interfaces and cope as multiple virtual circuits come into existence
-and go away dynamically.
- 
-Last, but not least, WAN configuration is much more complex than that of say
-Ethernet and may well amount to several dozens of parameters.  Some of them
-are "link-wide"  while others are virtual circuit-specific.  The same holds
-true for WAN statistics which is by far more extensive and extremely useful
-when troubleshooting WAN connections.  Extending the ifconfig utility to suit
-these needs may be possible, but does not seem quite reasonable.  Therefore, a
-WAN configuration utility and corresponding application programmer's interface
-is needed for this purpose.
-
-Most of these problems are taken care of by this module.  Its goal is to
-provide a user with more-or-less standard look and feel for all WAN devices and
-assist a WAN device driver writer by providing common services, such as:
-
- o User-level interface via /proc file system
- o Centralized configuration
- o Device management (setup, shutdown, etc.)
- o Network interface management (dynamic creation/destruction)
- o Protocol encapsulation/decapsulation
-
-To ba able to use the Linux WAN Router you will also need a WAN Tools package
-available from
-
-	ftp.sangoma.com/pub/linux/current_wanpipe/wanpipe-X.Y.Z.tgz
-
-where vX.Y.Z represent the wanpipe version number.
-
-For technical questions and/or comments please e-mail to ncorbic@sangoma.com.
-For general inquiries please contact Sangoma Technologies Inc. by
-
-	Hotline:	1-800-388-2475	(USA and Canada, toll free)
-	Phone:		(905) 474-1990  ext: 106
-	Fax:		(905) 474-9223
-	E-mail:		dm@sangoma.com	(David Mandelstam)
-	WWW:		http://www.sangoma.com
-
-
-INSTALLATION
-
-Please read the WanpipeForLinux.pdf manual on how to 
-install the WANPIPE tools and drivers properly. 
-
-
-After installing wanpipe package: /usr/local/wanrouter/doc. 
-On the ftp.sangoma.com : /linux/current_wanpipe/doc
-
-
-COPYRIGHT AND LICENSING INFORMATION
-
-This program is free software; you can redistribute it and/or modify it under
-the terms of the GNU General Public License as published by the Free Software
-Foundation; either version 2, or (at your option) any later version.
-
-This program is distributed in the hope that it will be useful, but WITHOUT
-ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
-FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License along with
-this program; if not, write to the Free Software Foundation, Inc., 675 Mass
-Ave, Cambridge, MA 02139, USA.
-
-
-
-ACKNOWLEDGEMENTS
-
-This product is based on the WANPIPE(tm) Multiprotocol WAN Router developed
-by Sangoma Technologies Inc. for Linux 2.0.x and 2.2.x.  Success of the WANPIPE
-together with the next major release of Linux kernel in summer 1996 commanded
-adequate changes to the WANPIPE code to take full advantage of new Linux
-features.
-
-Instead of continuing developing proprietary interface tied to Sangoma WAN
-cards, we decided to separate all hardware-independent code into a separate
-module and defined two levels of interfaces - one for user-level applications
-and another for kernel-level WAN drivers.  WANPIPE is now implemented as a
-WAN driver compliant with the WAN Link Driver interface.  Also a general
-purpose WAN configuration utility and a set of shell scripts was developed to 
-support WAN router at the user level.
-
-Many useful ideas concerning hardware-independent interface implementation
-were given by Mike McLagan <mike.mclagan@linux.org> and his implementation
-of the Frame Relay router and drivers for Sangoma cards (dlci/sdla).
-
-With the new implementation of the APIs being incorporated into the WANPIPE,
-a special thank goes to Alan Cox in providing insight into BSD sockets.
-
-Special thanks to all the WANPIPE users who performed field-testing, reported
-bugs and made valuable comments and suggestions that help us to improve this
-product.
-
-
-
-NEW IN THIS RELEASE
-
-	o Updated the WANCFG utility
-		Calls the pppconfig to configure the PPPD
-		for async connections.
-
-	o Added the PPPCONFIG utility
-		Used to configure the PPPD daemon for the
-		WANPIPE Async PPP and standard serial port.
-		The wancfg calls the pppconfig to configure
-		the pppd.
-
-	o Fixed the PCI autodetect feature.  
-		The SLOT 0 was used as an autodetect option
-		however, some high end PC's slot numbers start
-		from 0. 
-
-	o This release has been tested with the new backupd
-	  daemon release.
-	
-
-PRODUCT COMPONENTS AND RELATED FILES
-
-/etc: (or user defined)
-	wanpipe1.conf	default router configuration file
-
-/lib/modules/X.Y.Z/misc:
-	wanrouter.o	router kernel loadable module
-	af_wanpipe.o	wanpipe api socket module
-
-/lib/modules/X.Y.Z/net:
-	sdladrv.o	Sangoma SDLA support module
-	wanpipe.o	Sangoma WANPIPE(tm) driver module
-
-/proc/net/wanrouter
-	Config		reads current router configuration
-	Status		reads current router status
-	{name}		reads WAN driver statistics
-
-/usr/sbin:
-	wanrouter	wanrouter start-up script
-	wanconfig	wanrouter configuration utility
-	sdladump	WANPIPE adapter memory dump utility
-        fpipemon        Monitor for Frame Relay
-        cpipemon        Monitor for Cisco HDLC
-	ppipemon 	Monitor for PPP
-	xpipemon 	Monitor for X25
-	wpkbdmon        WANPIPE keyboard led monitor/debugger
-
-/usr/local/wanrouter:
-	README		this file
-	COPYING		GNU General Public License
-	Setup		installation script
-	Filelist	distribution definition file
-	wanrouter.rc	meta-configuration file 
-			(used by the Setup and wanrouter script)
-
-/usr/local/wanrouter/doc:
-	wanpipeForLinux.pdf 	WAN Router User's Manual
-
-/usr/local/wanrouter/patches:
-	wanrouter-v2213.gz  	patch for Linux kernels 2.2.11 up to 2.2.13.
-	wanrouter-v2214.gz	patch for Linux kernel 2.2.14. 
-	wanrouter-v2215.gz	patch for Linux kernels 2.2.15 to 2.2.17.
-	wanrouter-v2218.gz	patch for Linux kernels 2.2.18 and up.
-	wanrouter-v240.gz	patch for Linux kernel 2.4.0.  
-	wanrouter-v242.gz	patch for Linux kernel 2.4.2 and up.
-	wanrouter-v2034.gz	patch for Linux kernel 2.0.34
-	wanrouter-v2036.gz 	patch for Linux kernel 2.0.36 and up. 
-
-/usr/local/wanrouter/patches/kdrivers:
-	Sources of the latest WANPIPE device drivers.
-	These are used to UPGRADE the linux kernel to the newest
-	version if the kernel source has already been patched with
-	WANPIPE drivers.
-
-/usr/local/wanrouter/samples:
-	interface	sample interface configuration file
-	wanpipe1.cpri 	CHDLC primary port
-     	wanpipe2.csec 	CHDLC secondary port
-     	wanpipe1.fr   	Frame Relay protocol
-     	wanpipe1.ppp  	PPP protocol ) 
-	wanpipe1.asy	CHDLC ASYNC protocol
-	wanpipe1.x25	X25 protocol
-	wanpipe1.stty	Sync TTY driver (Used by Kernel PPPD daemon)
-	wanpipe1.atty	Async TTY driver (Used by Kernel PPPD daemon)
-	wanrouter.rc	sample meta-configuration file
-
-/usr/local/wanrouter/util:
-	*		wan-tools utilities source code
-
-/usr/local/wanrouter/api/x25:
-	*		x25 api sample programs.
-/usr/local/wanrouter/api/chdlc:
-	*		chdlc api sample programs.
-/usr/local/wanrouter/api/fr:
-	*		fr api sample programs.
-/usr/local/wanrouter/config/wancfg:
-	wancfg		WANPIPE GUI configuration program.
-                        Creates wanpipe#.conf files. 
-/usr/local/wanrouter/config/cfgft1:
-	cfgft1		GUI CSU/DSU configuration program.
-
-/usr/include/linux:
-	wanrouter.h	router API definitions
-	wanpipe.h	WANPIPE API definitions
-	sdladrv.h	SDLA support module API definitions
-	sdlasfm.h	SDLA firmware module definitions
-	if_wanpipe.h	WANPIPE Socket definitions
-	sdlapci.h	WANPIPE PCI definitions
-	
-
-/usr/src/linux/net/wanrouter:
-	*		wanrouter source code
-
-/var/log:
-	wanrouter	wanrouter start-up log (created by the Setup script)
-
-/var/lock:  (or /var/lock/subsys for RedHat)
-	wanrouter	wanrouter lock file (created by the Setup script)
-
-/usr/local/wanrouter/firmware:
-	fr514.sfm	Frame relay firmware for Sangoma S508/S514 card
-	cdual514.sfm	Dual Port Cisco HDLC firmware for Sangoma S508/S514 card
-	ppp514.sfm      PPP Firmware for Sangoma S508 and S514 cards
-	x25_508.sfm	X25 Firmware for Sangoma S508 card.
-
-
-REVISION HISTORY
-
-1.0.0	December 31, 1996	Initial version
-
-1.0.1	January 30, 1997	Status and statistics can be read via /proc
-				filesystem entries.
-
-1.0.2   April 30, 1997          Added UDP management via monitors.
-
-1.0.3	June 3, 1997		UDP management for multiple boards using Frame
-				Relay and PPP
-				Enabled continuous transmission of Configure 
-				Request Packet for PPP (for 508 only)
-				Connection Timeout for PPP changed from 900 to 0
-				Flow Control Problem fixed for Frame Relay
-
-1.0.4	July 10, 1997		S508/FT1 monitoring capability in fpipemon and
-				ppipemon utilities.
-				Configurable TTL for UDP packets.
-				Multicast and Broadcast IP source addresses are
-				silently discarded.
-
-1.0.5	July 28, 1997		Configurable T391,T392,N391,N392,N393 for Frame
-				Relay in router.conf.
-				Configurable Memory Address through router.conf 
-				for Frame Relay, PPP and X.25. (commenting this
- 				out enables auto-detection).
-				Fixed freeing up received buffers using kfree()
- 				for Frame Relay and X.25.
-				Protect sdla_peek() by calling save_flags(),
-				cli() and restore_flags().
-				Changed number of Trace elements from 32 to 20
-				Added DLCI specific data monitoring in FPIPEMON. 
-2.0.0	Nov 07, 1997		Implemented protection of RACE conditions by 
-				critical flags for FRAME RELAY and PPP.
-				DLCI List interrupt mode implemented.
-				IPX support in FRAME RELAY and PPP.
-				IPX Server Support (MARS)
-				More driver specific stats included in FPIPEMON
-				and PIPEMON.
-
-2.0.1	Nov 28, 1997		Bug Fixes for version 2.0.0.
-				Protection of "enable_irq()" while 
-				"disable_irq()" has been enabled from any other
-				routine (for Frame Relay, PPP and X25).
-				Added additional Stats for Fpipemon and Ppipemon
-				Improved Load Sharing for multiple boards
-
-2.0.2	Dec 09, 1997		Support for PAP and CHAP for ppp has been
-				implemented.
-
-2.0.3	Aug 15, 1998		New release supporting Cisco HDLC, CIR for Frame
-				relay, Dynamic IP assignment for PPP and Inverse
-				Arp support for Frame-relay.  Man Pages are 
-				included for better support and a new utility
-				for configuring FT1 cards.
-
-2.0.4	Dec 09, 1998	        Dual Port support for Cisco HDLC.
-				Support for HDLC (LAPB) API.
-				Supports BiSync Streaming code for S502E 
-				and S503 cards.
-				Support for Streaming HDLC API.
-				Provides a BSD socket interface for 
-				creating applications using BiSync
-   				streaming.        
-
-2.0.5   Aug 04, 1999 		CHDLC initialization bug fix.
-				PPP interrupt driven driver: 
-  				Fix to the PPP line hangup problem.
-				New PPP firmware
-				Added comments to the startup SYSTEM ERROR messages
-				Xpipemon debugging application for the X25 protocol
-				New USER_MANUAL.txt
-				Fixed the odd boundary 4byte writes to the board.
-				BiSync Streaming code has been taken out.  
-				 Available as a patch.
-				Streaming HDLC API has been taken out.  
-				 Available as a patch.                 
-
-2.0.6   Aug 17, 1999		Increased debugging in statup scripts
-				Fixed installation bugs from 2.0.5
-				Kernel patch works for both 2.2.10 and 2.2.11 kernels.
-				There is no functional difference between the two packages         
-
-2.0.7   Aug 26, 1999		o  Merged X25API code into WANPIPE.
-				o  Fixed a memory leak for X25API
-				o  Updated the X25API code for 2.2.X kernels.
-				o  Improved NEM handling.   
-
-2.1.0	Oct 25, 1999		o New code for S514 PCI Card
-				o New CHDLC and Frame Relay drivers
-				o PPP and X25 are not supported in this release    
-
-2.1.1	Nov 30, 1999		o PPP support for S514 PCI Cards
-
-2.1.3   Apr 06, 2000		o Socket based x25api 
-				o Socket based chdlc api
-				o Socket based fr api
-				o Dual Port Receive only CHDLC support.
-				o Asynchronous CHDLC support (Secondary Port)
-				o cfgft1 GUI csu/dsu configurator
-				o wancfg GUI configuration file 
-				  configurator.
-				o Architectural directory changes.
-
-beta-2.1.4 Jul 2000		o Dynamic interface configuration:
-					Network interfaces reflect the state
-					of protocol layer.  If the protocol becomes
-					disconnected, driver will bring down
-					the interface.  Once the protocol reconnects
-					the interface will be brought up. 
-					
-					Note: This option is turned off by default.
-
-				o Dynamic wanrouter setup using 'wanconfig':
-					wanconfig utility can be used to
-					shutdown,restart,start or reconfigure 
-					a virtual circuit dynamically.
-				     
-					Frame Relay:  Each DLCI can be: 
-						      created,stopped,restarted and reconfigured
-						      dynamically using wanconfig.
-					
-						      ex: wanconfig card wanpipe1 dev wp1_fr16 up
-				  
-				o Wanrouter startup via command line arguments:
-					wanconfig also supports wanrouter startup via command line
-					arguments.  Thus, there is no need to create a wanpipe#.conf
-					configuration file.  
-
-				o Socket based x25api update/bug fixes.
-					Added support for LCN numbers greater than 255.
-					Option to pass up modem messages.
-					Provided a PCI IRQ check, so a single S514
-					card is guaranteed to have a non-sharing interrupt.
-
-				o Fixes to the wancfg utility.
-				o New FT1 debugging support via *pipemon utilities.
-				o Frame Relay ARP support Enabled.
-
-beta3-2.1.4 Jul 2000		o X25 M_BIT Problem fix.
-				o Added the Multi-Port PPP
-				  Updated utilities for the Multi-Port PPP.
-
-2.1.4	Aut 2000
-				o In X25API:
-					Maximum packet an application can send
-					to the driver has been extended to 4096 bytes.
-
-					Fixed the x25 startup bug. Enable 
-					communications only after all interfaces
-					come up.  HIGH SVC/PVC is used to calculate
-					the number of channels.
-					Enable protocol only after all interfaces
-					are enabled.
-
-				o Added an extra state to the FT1 config, kernel module.
-				o Updated the pipemon debuggers.
-
-				o Blocked the Multi-Port PPP from running on kernels
-				  2.2.16 or greater, due to syncppp kernel module
-				  change. 
-	  
-beta1-2.1.5 	Nov 15 2000
-				o Fixed the MultiPort PPP Support for kernels 2.2.16 and above.
-				  2.2.X kernels only
-
-				o Secured the driver UDP debugging calls
-					- All illegal network debugging calls are reported to
-					  the log.
-					- Defined a set of allowed commands, all other denied.
-					
-				o Cpipemon
-					- Added set FT1 commands to the cpipemon. Thus CSU/DSU
-					  configuration can be performed using cpipemon.
-					  All systems that cannot run cfgft1 GUI utility should
-					  use cpipemon to configure the on board CSU/DSU.
-
-
-				o Keyboard Led Monitor/Debugger
-					- A new utility /usr/sbin/wpkbdmon uses keyboard leds
-					  to convey operational statistic information of the 
-					  Sangoma WANPIPE cards.
-					NUM_LOCK    = Line State  (On=connected,    Off=disconnected)
-					CAPS_LOCK   = Tx data     (On=transmitting, Off=no tx data)
-					SCROLL_LOCK = Rx data     (On=receiving,    Off=no rx data
-					
-				o Hardware probe on module load and dynamic device allocation
-					- During WANPIPE module load, all Sangoma cards are probed
-					  and found information is printed in the /var/log/messages.
-					- If no cards are found, the module load fails.
-					- Appropriate number of devices are dynamically loaded 
-					  based on the number of Sangoma cards found.
-
-					  Note: The kernel configuration option 
-						CONFIG_WANPIPE_CARDS has been taken out.
-					
-				o Fixed the Frame Relay and Chdlc network interfaces so they are
-				  compatible with libpcap libraries.  Meaning, tcpdump, snort,
-				  ethereal, and all other packet sniffers and debuggers work on
-				  all WANPIPE network interfaces.
-					- Set the network interface encoding type to ARPHRD_PPP.
-					  This tell the sniffers that data obtained from the
-					  network interface is in pure IP format.
-				  Fix for 2.2.X kernels only.
-				
-				o True interface encoding option for Frame Relay and CHDLC
-					- The above fix sets the network interface encoding
-					  type to ARPHRD_PPP, however some customers use
-					  the encoding interface type to determine the
-					  protocol running.  Therefore, the TURE ENCODING
-					  option will set the interface type back to the
-					  original value.  
-
-					  NOTE: If this option is used with Frame Relay and CHDLC
-						libpcap library support will be broken.  
-						i.e. tcpdump will not work.
-					Fix for 2.2.x Kernels only.
-						
-				o Ethernet Bridgind over Frame Relay
-					- The Frame Relay bridging has been developed by 
-					  Kristian Hoffmann and Mark Wells.  
-					- The Linux kernel bridge is used to send ethernet 
-					  data over the frame relay links.
-					For 2.2.X Kernels only.
-
-				o Added extensive 2.0.X support. Most new features of
-				  2.1.5 for protocols Frame Relay, PPP and CHDLC are
-				  supported under 2.0.X kernels. 
-
-beta1-2.2.0 	Dec 30 2000
-				o Updated drivers for 2.4.X kernels.
-				o Updated drivers for SMP support.
-				o X25API is now able to share PCI interrupts.
-				o Took out a general polling routine that was used
-				  only by X25API. 
-				o Added appropriate locks to the dynamic reconfiguration
-				  code.
-				o Fixed a bug in the keyboard debug monitor.
-
-beta2-2.2.0	Jan 8 2001
-				o Patches for 2.4.0 kernel
-				o Patches for 2.2.18 kernel
-				o Minor updates to PPP and CHLDC drivers.
-				  Note: No functional difference.
-
-beta3-2.2.9	Jan 10 2001
-				o I missed the 2.2.18 kernel patches in beta2-2.2.0
-				  release.  They are included in this release.
-
-Stable Release
-2.2.0		Feb 01 2001
-				o Bug fix in wancfg GUI configurator.
-					The edit function didn't work properly.
-
-
-bata1-2.2.1	Feb 09 2001
-			o WANPIPE TTY Driver emulation. 
-			  Two modes of operation Sync and Async.
-				Sync: Using the PPPD daemon, kernel SyncPPP layer
-				      and the Wanpipe sync TTY driver: a PPP protocol 
-				      connection can be established via Sangoma adapter, over
-				      a T1 leased line.
-			
-				      The 2.4.0 kernel PPP layer supports MULTILINK
-				      protocol, that can be used to bundle any number of Sangoma
-				      adapters (T1 lines) into one, under a single IP address.
-				      Thus, efficiently obtaining multiple T1 throughput. 
-
-				      NOTE: The remote side must also implement MULTILINK PPP
-					    protocol.
-
-				Async:Using the PPPD daemon, kernel AsyncPPP layer
-				      and the WANPIPE async TTY driver: a PPP protocol
-				      connection can be established via Sangoma adapter and
-				      a modem, over a telephone line.
-
-				      Thus, the WANPIPE async TTY driver simulates a serial
-				      TTY driver that would normally be used to interface the 
-				      MODEM to the linux kernel.
-				
-			o WANPIPE PPP Backup Utility
-				This utility will monitor the state of the PPP T1 line.
-				In case of failure, a dial up connection will be established
-				via pppd daemon, ether via a serial tty driver (serial port), 
-				or a WANPIPE async TTY driver (in case serial port is unavailable).
-				
-				Furthermore, while in dial up mode, the primary PPP T1 link
-				will be monitored for signs of life.  
-
-				If the PPP T1 link comes back to life, the dial up connection
-				will be shutdown and T1 line re-established.
-			
-
-			o New Setup installation script.
-				Option to UPGRADE device drivers if the kernel source has
-				already been patched with WANPIPE.
-
-				Option to COMPILE WANPIPE modules against the currently 
-				running kernel, thus no need for manual kernel and module
-				re-compilation.
-			
-			o Updates and Bug Fixes to wancfg utility.
-
-bata2-2.2.1	Feb 20 2001
-
-			o Bug fixes to the CHDLC device drivers.
-				The driver had compilation problems under kernels
-				2.2.14 or lower.
-
-			o Bug fixes to the Setup installation script.
-				The device drivers compilation options didn't work
-				properly.
-
-			o Update to the wpbackupd daemon.  
-				Optimized the cross-over times, between the primary
-				link and the backup dialup.
-
-beta3-2.2.1	Mar 02 2001
-			o Patches for 2.4.2 kernel.
-
-			o Bug fixes to util/ make files.
-			o Bug fixes to the Setup installation script.
-
-			o Took out the backupd support and made it into
-			  as separate package.
-			  
-beta4-2.2.1     Mar 12 2001
-
-		o Fix to the Frame Relay Device driver.
-			IPSAC sends a packet of zero length
-			header to the frame relay driver.  The
-			driver tries to push its own 2 byte header
-			into the packet, which causes the driver to
-			crash.
-
-		o Fix the WANPIPE re-configuration code.
-			Bug was found by trying to run  the cfgft1 while the
-			interface was already running.  
-
-		o Updates to cfgft1.
-			Writes a wanpipe#.cfgft1 configuration file
-			once the CSU/DSU is configured. This file can
-			holds the current CSU/DSU configuration.
-
-
-
->>>>>> END OF README <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
-
-
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index e56b739d8e23..b3d9ccde0c27 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -231,7 +231,6 @@ unifdef-y += if_pppol2tp.h
 unifdef-y += if_pppox.h
 unifdef-y += if_tr.h
 unifdef-y += if_vlan.h
-unifdef-y += if_wanpipe.h
 unifdef-y += igmp.h
 unifdef-y += inet_diag.h
 unifdef-y += in.h
diff --git a/include/linux/if_wanpipe.h b/include/linux/if_wanpipe.h
deleted file mode 100644
index e594ca6069e5..000000000000
--- a/include/linux/if_wanpipe.h
+++ /dev/null
@@ -1,124 +0,0 @@
-/*****************************************************************************
-* if_wanpipe.h	Header file for the Sangoma AF_WANPIPE Socket 	
-*
-* Author: 	Nenad Corbic 	
-*
-* Copyright:	(c) 2000 Sangoma Technologies Inc.
-*
-*		This program is free software; you can redistribute it and/or
-*		modify it under the terms of the GNU General Public License
-*		as published by the Free Software Foundation; either version
-*		2 of the License, or (at your option) any later version.
-* ============================================================================
-*
-* Jan 28, 2000	Nenad Corbic 	Initial Version
-*
-*****************************************************************************/
-
-#ifndef __LINUX_IF_WAN_PACKET_H
-#define __LINUX_IF_WAN_PACKET_H
-
-struct wan_sockaddr_ll
-{
-	unsigned short	sll_family;
-	unsigned short	sll_protocol;
-	int		sll_ifindex;
-	unsigned short	sll_hatype;
-	unsigned char	sll_pkttype;
-	unsigned char	sll_halen;
-	unsigned char	sll_addr[8];
-	unsigned char   sll_device[14];
-	unsigned char 	sll_card[14];
-};
-
-typedef struct 
-{
-	unsigned char free;
-	unsigned char state_sk;
-	int rcvbuf;
-	int sndbuf;
-	int rmem;
-	int wmem;
-	int sk_count;
-	unsigned char bound;
-	char name[14];
-	unsigned char d_state;
-	unsigned char svc;
-	unsigned short lcn;
-	unsigned char mbox;
-	unsigned char cmd_busy;
-	unsigned char command;
-	unsigned poll;
-	unsigned poll_cnt;
-	int rblock;	
-} wan_debug_hdr_t;
-
-#define MAX_NUM_DEBUG  10
-#define X25_PROT       0x16
-#define PVC_PROT       0x17	
-
-typedef struct
-{
-	wan_debug_hdr_t debug[MAX_NUM_DEBUG];
-}wan_debug_t;
-
-#define	SIOC_WANPIPE_GET_CALL_DATA	(SIOCPROTOPRIVATE + 0)
-#define	SIOC_WANPIPE_SET_CALL_DATA	(SIOCPROTOPRIVATE + 1)
-#define SIOC_WANPIPE_ACCEPT_CALL	(SIOCPROTOPRIVATE + 2)
-#define SIOC_WANPIPE_CLEAR_CALL	        (SIOCPROTOPRIVATE + 3)
-#define SIOC_WANPIPE_RESET_CALL	        (SIOCPROTOPRIVATE + 4)
-#define SIOC_WANPIPE_DEBUG	        (SIOCPROTOPRIVATE + 5)
-#define SIOC_WANPIPE_SET_NONBLOCK	(SIOCPROTOPRIVATE + 6)
-#define SIOC_WANPIPE_CHECK_TX		(SIOCPROTOPRIVATE + 7)
-#define SIOC_WANPIPE_SOCK_STATE		(SIOCPROTOPRIVATE + 8)
-
-/* Packet types */
-
-#define WAN_PACKET_HOST		0		/* To us		*/
-#define WAN_PACKET_BROADCAST	1		/* To all		*/
-#define WAN_PACKET_MULTICAST	2		/* To group		*/
-#define WAN_PACKET_OTHERHOST	3		/* To someone else 	*/
-#define WAN_PACKET_OUTGOING		4		/* Outgoing of any type */
-/* These ones are invisible by user level */
-#define WAN_PACKET_LOOPBACK		5		/* MC/BRD frame looped back */
-#define WAN_PACKET_FASTROUTE	6		/* Fastrouted frame	*/
-
-
-/* X25 specific */
-#define WAN_PACKET_DATA 	7
-#define WAN_PACKET_CMD 		8
-#define WAN_PACKET_ASYNC	9
-#define WAN_PACKET_ERR	       10
-
-/* Packet socket options */
-
-#define WAN_PACKET_ADD_MEMBERSHIP		1
-#define WAN_PACKET_DROP_MEMBERSHIP		2
-
-#define WAN_PACKET_MR_MULTICAST	0
-#define WAN_PACKET_MR_PROMISC	1
-#define WAN_PACKET_MR_ALLMULTI	2
-
-#ifdef __KERNEL__
-
-/* Private wanpipe socket structures. */
-struct wanpipe_opt
-{
-	void   *mbox;		/* Mail box  */
-	void   *card; 		/* Card bouded to */
-	struct net_device *dev;	/* Bounded device */
-	unsigned short lcn;	/* Binded LCN */
-	unsigned char  svc;	/* 0=pvc, 1=svc */
-	unsigned char  timer;   /* flag for delayed transmit*/	
-	struct timer_list tx_timer;
-	unsigned poll_cnt;
-	unsigned char force;	/* Used to force sock release */
-	atomic_t packet_sent;   
-	unsigned short num; 
-};
-
-#define wp_sk(__sk) ((struct wanpipe_opt *)(__sk)->sk_protinfo)
-
-#endif
-
-#endif
-- 
cgit v1.2.3


From 7ef3abd2104232a35f259dad6a213310edc7c9fe Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 15 Apr 2008 00:29:24 -0700
Subject: [IRDA]: Remove irlan_eth_send_gratuitous_arp()

Even kernel 2.2.26 (sic) already contains the
  #undef CONFIG_IRLAN_SEND_GRATUITOUS_ARP
with the comment "but for some reason the machine crashes if you use DHCP".

Either someone finally looks into this or it's simply time to remove
this dead code.

Reported-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/irda/irlan_eth.h  |  1 -
 net/irda/irlan/irlan_common.c | 10 ----------
 net/irda/irlan/irlan_eth.c    | 33 ---------------------------------
 3 files changed, 44 deletions(-)

(limited to 'include')

diff --git a/include/net/irda/irlan_eth.h b/include/net/irda/irlan_eth.h
index 0062347600b9..de5c81691f33 100644
--- a/include/net/irda/irlan_eth.h
+++ b/include/net/irda/irlan_eth.h
@@ -29,5 +29,4 @@ struct net_device *alloc_irlandev(const char *name);
 int  irlan_eth_receive(void *instance, void *sap, struct sk_buff *skb);
 
 void irlan_eth_flow_indication( void *instance, void *sap, LOCAL_FLOW flow);
-void irlan_eth_send_gratuitous_arp(struct net_device *dev);
 #endif
diff --git a/net/irda/irlan/irlan_common.c b/net/irda/irlan/irlan_common.c
index 75bb6a9dcaaa..9a1cd87e7142 100644
--- a/net/irda/irlan/irlan_common.c
+++ b/net/irda/irlan/irlan_common.c
@@ -54,13 +54,6 @@
 #include <net/irda/irlan_filter.h>
 
 
-/*
- * Send gratuitous ARP when connected to a new AP or not. May be a clever
- * thing to do, but for some reason the machine crashes if you use DHCP. So
- * lets not use it by default.
- */
-#undef CONFIG_IRLAN_SEND_GRATUITOUS_ARP
-
 /* extern char sysctl_devname[]; */
 
 /*
@@ -393,9 +386,6 @@ static void irlan_connect_confirm(void *instance, void *sap,
 	/* Ready to transfer Ethernet frames */
 	netif_start_queue(self->dev);
 	self->disconnect_reason = 0; /* Clear reason */
-#ifdef CONFIG_IRLAN_SEND_GRATUITOUS_ARP
-	irlan_eth_send_gratuitous_arp(&self->dev);
-#endif
 	wake_up_interruptible(&self->open_wait);
 }
 
diff --git a/net/irda/irlan/irlan_eth.c b/net/irda/irlan/irlan_eth.c
index 7a6b14ab1e7f..05112be99569 100644
--- a/net/irda/irlan/irlan_eth.c
+++ b/net/irda/irlan/irlan_eth.c
@@ -288,39 +288,6 @@ void irlan_eth_flow_indication(void *instance, void *sap, LOCAL_FLOW flow)
 	}
 }
 
-/*
- * Function irlan_etc_send_gratuitous_arp (dev)
- *
- *    Send gratuitous ARP to announce that we have changed
- *    hardware address, so that all peers updates their ARP tables
- */
-void irlan_eth_send_gratuitous_arp(struct net_device *dev)
-{
-#ifdef CONFIG_INET
-	struct in_device *in_dev;
-
-	/*
-	 * When we get a new MAC address do a gratuitous ARP. This
-	 * is useful if we have changed access points on the same
-	 * subnet.
-	 */
-	IRDA_DEBUG(4, "IrLAN: Sending gratuitous ARP\n");
-	rcu_read_lock();
-	in_dev = __in_dev_get_rcu(dev);
-	if (in_dev == NULL)
-		goto out;
-	if (in_dev->ifa_list)
-
-	arp_send(ARPOP_REQUEST, ETH_P_ARP,
-		 in_dev->ifa_list->ifa_address,
-		 dev,
-		 in_dev->ifa_list->ifa_address,
-		 NULL, dev->dev_addr, NULL);
-out:
-	rcu_read_unlock();
-#endif /* CONFIG_INET */
-}
-
 /*
  * Function set_multicast_list (dev)
  *
-- 
cgit v1.2.3


From 31efdf0530b6351b0658d35a602a0f2d6bc2ed6f Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 15 Apr 2008 00:30:16 -0700
Subject: [ISDN] include/linux/isdn.h: remove dead code

This patch remove the usage of a nonexisting kconfig variable.

Reported-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/isdn.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/isdn.h b/include/linux/isdn.h
index 9cb2855bb170..44cd663c53b6 100644
--- a/include/linux/isdn.h
+++ b/include/linux/isdn.h
@@ -16,14 +16,8 @@
 
 #include <linux/ioctl.h>
 
-#ifdef CONFIG_COBALT_MICRO_SERVER
-/* Save memory */
-#define ISDN_MAX_DRIVERS    2
-#define ISDN_MAX_CHANNELS   8
-#else
 #define ISDN_MAX_DRIVERS    32
 #define ISDN_MAX_CHANNELS   64
-#endif
 
 /* New ioctl-codes */
 #define IIOCNETAIF  _IO('I',1)
-- 
cgit v1.2.3


From c93cf61fd1d5378134f9b06703f7078067542e00 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 15 Apr 2008 00:35:23 -0700
Subject: [NETNS]: The net-subsys IDs generator.

To make some per-net generic pointers, we need some way to address
them, i.e. - IDs. This is simple IDA-based IDs generator for pernet
subsystems.

Addressing questions about potential checkpoint/restart problems:
these IDs are "lite-offsets" within the net structure and are by no
means supposed to be exported to the userspace.

Since it will be used in the nearest future by devices only (tun,
vlan, tunnels, bridge, etc), I make it resemble the functionality
of register_pernet_device().

The new ids is stored in the *id pointer _before_ calling the init
callback to make this id available in this callback.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  2 ++
 net/core/net_namespace.c    | 38 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 40 insertions(+)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index e2aee2689abe..72fad1a0956b 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -185,6 +185,8 @@ extern int register_pernet_subsys(struct pernet_operations *);
 extern void unregister_pernet_subsys(struct pernet_operations *);
 extern int register_pernet_device(struct pernet_operations *);
 extern void unregister_pernet_device(struct pernet_operations *);
+extern int register_pernet_gen_device(int *id, struct pernet_operations *);
+extern void unregister_pernet_gen_device(int id, struct pernet_operations *);
 
 struct ctl_path;
 struct ctl_table;
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 7b660834a4c2..2197d51aef3b 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -5,6 +5,7 @@
 #include <linux/list.h>
 #include <linux/delay.h>
 #include <linux/sched.h>
+#include <linux/idr.h>
 #include <net/net_namespace.h>
 
 /*
@@ -253,6 +254,8 @@ static void unregister_pernet_operations(struct pernet_operations *ops)
 }
 #endif
 
+static DEFINE_IDA(net_generic_ids);
+
 /**
  *      register_pernet_subsys - register a network namespace subsystem
  *	@ops:  pernet operations structure for the subsystem
@@ -330,6 +333,30 @@ int register_pernet_device(struct pernet_operations *ops)
 }
 EXPORT_SYMBOL_GPL(register_pernet_device);
 
+int register_pernet_gen_device(int *id, struct pernet_operations *ops)
+{
+	int error;
+	mutex_lock(&net_mutex);
+again:
+	error = ida_get_new_above(&net_generic_ids, 1, id);
+	if (error) {
+		if (error == -EAGAIN) {
+			ida_pre_get(&net_generic_ids, GFP_KERNEL);
+			goto again;
+		}
+		goto out;
+	}
+	error = register_pernet_operations(&pernet_list, ops);
+	if (error)
+		ida_remove(&net_generic_ids, *id);
+	else if (first_device == &pernet_list)
+		first_device = &ops->list;
+out:
+	mutex_unlock(&net_mutex);
+	return error;
+}
+EXPORT_SYMBOL_GPL(register_pernet_gen_device);
+
 /**
  *      unregister_pernet_device - unregister a network namespace netdevice
  *	@ops: pernet operations structure to manipulate
@@ -348,3 +375,14 @@ void unregister_pernet_device(struct pernet_operations *ops)
 	mutex_unlock(&net_mutex);
 }
 EXPORT_SYMBOL_GPL(unregister_pernet_device);
+
+void unregister_pernet_gen_device(int id, struct pernet_operations *ops)
+{
+	mutex_lock(&net_mutex);
+	if (&ops->list == first_device)
+		first_device = first_device->next;
+	unregister_pernet_operations(ops);
+	ida_remove(&net_generic_ids, id);
+	mutex_unlock(&net_mutex);
+}
+EXPORT_SYMBOL_GPL(unregister_pernet_gen_device);
-- 
cgit v1.2.3


From dec827d174d7f76c457238800183ca864a639365 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Tue, 15 Apr 2008 00:36:08 -0700
Subject: [NETNS]: The generic per-net pointers.

Add the elastic array of void * pointer to the struct net.
The access rules are simple:

 1. register the ops with register_pernet_gen_device to get
    the id of your private pointer
 2. call net_assign_generic() to put the private data on the
    struct net (most preferably this should be done in the
    ->init callback of the ops registered)
 3. do not store any private reference on the net_generic array;
 4. do not change this pointer while the net is alive;
 5. use the net_generic() to get the pointer.

When adding a new pointer, I copy the old array, replace it
with a new one and schedule the old for kfree after an RCU
grace period.

Since the net_generic explores the net->gen array inside rcu
read section and once set the net->gen->ptr[x] pointer never
changes, this grants us a safe access to generic pointers.

Quoting Paul: "... RCU is protecting -only- the net_generic
structure that net_generic() is traversing, and the [pointer]
returned by net_generic() is protected by a reference counter
in the upper-level struct net."

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h |  2 ++
 include/net/netns/generic.h | 49 +++++++++++++++++++++++++++++++++++
 net/core/net_namespace.c    | 62 +++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 113 insertions(+)
 create mode 100644 include/net/netns/generic.h

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 72fad1a0956b..f880b0f9f107 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -20,6 +20,7 @@ struct proc_dir_entry;
 struct net_device;
 struct sock;
 struct ctl_table_header;
+struct net_generic;
 
 struct net {
 	atomic_t		count;		/* To decided when the network
@@ -61,6 +62,7 @@ struct net {
 #ifdef CONFIG_NETFILTER
 	struct netns_xt		xt;
 #endif
+	struct net_generic	*gen;
 };
 
 
diff --git a/include/net/netns/generic.h b/include/net/netns/generic.h
new file mode 100644
index 000000000000..0c04fd2a700b
--- /dev/null
+++ b/include/net/netns/generic.h
@@ -0,0 +1,49 @@
+/*
+ * generic net pointers
+ */
+
+#ifndef __NET_GENERIC_H__
+#define __NET_GENERIC_H__
+
+#include <linux/rcupdate.h>
+
+/*
+ * Generic net pointers are to be used by modules to put some private
+ * stuff on the struct net without explicit struct net modification
+ *
+ * The rules are simple:
+ * 1. register the ops with register_pernet_gen_device to get the id
+ *    of your private pointer;
+ * 2. call net_assign_generic() to put the private data on the struct
+ *    net (most preferably this should be done in the ->init callback
+ *    of the ops registered);
+ * 3. do not change this pointer while the net is alive;
+ * 4. do not try to have any private reference on the net_generic object.
+ *
+ * After accomplishing all of the above, the private pointer can be
+ * accessed with the net_generic() call.
+ */
+
+struct net_generic {
+	unsigned int len;
+	struct rcu_head rcu;
+
+	void *ptr[0];
+};
+
+static inline void *net_generic(struct net *net, int id)
+{
+	struct net_generic *ng;
+	void *ptr;
+
+	rcu_read_lock();
+	ng = rcu_dereference(net->gen);
+	BUG_ON(id == 0 || id > ng->len);
+	ptr = ng->ptr[id - 1];
+	rcu_read_unlock();
+
+	return ptr;
+}
+
+extern int net_assign_generic(struct net *net, int id, void *data);
+#endif
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 2197d51aef3b..763674e1e593 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -7,6 +7,7 @@
 #include <linux/sched.h>
 #include <linux/idr.h>
 #include <net/net_namespace.h>
+#include <net/netns/generic.h>
 
 /*
  *	Our network namespace constructor/destructor lists
@@ -21,6 +22,8 @@ LIST_HEAD(net_namespace_list);
 struct net init_net;
 EXPORT_SYMBOL(init_net);
 
+#define INITIAL_NET_GEN_PTRS	13 /* +1 for len +2 for rcu_head */
+
 /*
  * setup_net runs the initializers for the network namespace object.
  */
@@ -29,10 +32,21 @@ static __net_init int setup_net(struct net *net)
 	/* Must be called with net_mutex held */
 	struct pernet_operations *ops;
 	int error;
+	struct net_generic *ng;
 
 	atomic_set(&net->count, 1);
 	atomic_set(&net->use_count, 0);
 
+	error = -ENOMEM;
+	ng = kzalloc(sizeof(struct net_generic) +
+			INITIAL_NET_GEN_PTRS * sizeof(void *), GFP_KERNEL);
+	if (ng == NULL)
+		goto out;
+
+	ng->len = INITIAL_NET_GEN_PTRS;
+	INIT_RCU_HEAD(&ng->rcu);
+	rcu_assign_pointer(net->gen, ng);
+
 	error = 0;
 	list_for_each_entry(ops, &pernet_list, list) {
 		if (ops->init) {
@@ -54,6 +68,7 @@ out_undo:
 	}
 
 	rcu_barrier();
+	kfree(ng);
 	goto out;
 }
 
@@ -386,3 +401,50 @@ void unregister_pernet_gen_device(int id, struct pernet_operations *ops)
 	mutex_unlock(&net_mutex);
 }
 EXPORT_SYMBOL_GPL(unregister_pernet_gen_device);
+
+static void net_generic_release(struct rcu_head *rcu)
+{
+	struct net_generic *ng;
+
+	ng = container_of(rcu, struct net_generic, rcu);
+	kfree(ng);
+}
+
+int net_assign_generic(struct net *net, int id, void *data)
+{
+	struct net_generic *ng, *old_ng;
+
+	BUG_ON(!mutex_is_locked(&net_mutex));
+	BUG_ON(id == 0);
+
+	ng = old_ng = net->gen;
+	if (old_ng->len >= id)
+		goto assign;
+
+	ng = kzalloc(sizeof(struct net_generic) +
+			id * sizeof(void *), GFP_KERNEL);
+	if (ng == NULL)
+		return -ENOMEM;
+
+	/*
+	 * Some synchronisation notes:
+	 *
+	 * The net_generic explores the net->gen array inside rcu
+	 * read section. Besides once set the net->gen->ptr[x]
+	 * pointer never changes (see rules in netns/generic.h).
+	 *
+	 * That said, we simply duplicate this array and schedule
+	 * the old copy for kfree after a grace period.
+	 */
+
+	ng->len = id;
+	INIT_RCU_HEAD(&ng->rcu);
+	memcpy(&ng->ptr, &old_ng->ptr, old_ng->len);
+
+	rcu_assign_pointer(net->gen, ng);
+	call_rcu(&old_ng->rcu, net_generic_release);
+assign:
+	ng->ptr[id - 1] = data;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(net_assign_generic);
-- 
cgit v1.2.3


From 4ac58469f13028e1eb97f8bc7b0fca5072591d8d Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Fri, 11 Apr 2008 11:59:00 +0200
Subject: ssb: Fix usage of struct device used for DMAing

This fixes DMA on architectures where DMA is nontrivial, like PPC64.
We must use the host-device's (PCI) struct device for any DMA
operation instead of the SSB device. For this we add a new
struct device pointer to the SSB device structure that will always
point to the right device for DMAing.

Without this patch b43 and b44 drivers won't work on complex-DMA
architectures, that for example need dev->archdata for DMA operations.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/b44.c              | 52 +++++++++++++++++++++---------------------
 drivers/net/wireless/b43/dma.c | 27 +++++++++++-----------
 drivers/ssb/main.c             | 14 +++++++-----
 include/linux/ssb/ssb.h        |  4 ++++
 4 files changed, 52 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index 25f1337cd02c..59dce6aa0865 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -148,7 +148,7 @@ static inline void b44_sync_dma_desc_for_device(struct ssb_device *sdev,
 						unsigned long offset,
 						enum dma_data_direction dir)
 {
-	dma_sync_single_range_for_device(sdev->dev, dma_base,
+	dma_sync_single_range_for_device(sdev->dma_dev, dma_base,
 					 offset & dma_desc_align_mask,
 					 dma_desc_sync_size, dir);
 }
@@ -158,7 +158,7 @@ static inline void b44_sync_dma_desc_for_cpu(struct ssb_device *sdev,
 					     unsigned long offset,
 					     enum dma_data_direction dir)
 {
-	dma_sync_single_range_for_cpu(sdev->dev, dma_base,
+	dma_sync_single_range_for_cpu(sdev->dma_dev, dma_base,
 				      offset & dma_desc_align_mask,
 				      dma_desc_sync_size, dir);
 }
@@ -613,7 +613,7 @@ static void b44_tx(struct b44 *bp)
 
 		BUG_ON(skb == NULL);
 
-		dma_unmap_single(bp->sdev->dev,
+		dma_unmap_single(bp->sdev->dma_dev,
 				 rp->mapping,
 				 skb->len,
 				 DMA_TO_DEVICE);
@@ -653,7 +653,7 @@ static int b44_alloc_rx_skb(struct b44 *bp, int src_idx, u32 dest_idx_unmasked)
 	if (skb == NULL)
 		return -ENOMEM;
 
-	mapping = dma_map_single(bp->sdev->dev, skb->data,
+	mapping = dma_map_single(bp->sdev->dma_dev, skb->data,
 				 RX_PKT_BUF_SZ,
 				 DMA_FROM_DEVICE);
 
@@ -663,19 +663,19 @@ static int b44_alloc_rx_skb(struct b44 *bp, int src_idx, u32 dest_idx_unmasked)
 		mapping + RX_PKT_BUF_SZ > DMA_30BIT_MASK) {
 		/* Sigh... */
 		if (!dma_mapping_error(mapping))
-			dma_unmap_single(bp->sdev->dev, mapping,
+			dma_unmap_single(bp->sdev->dma_dev, mapping,
 					RX_PKT_BUF_SZ, DMA_FROM_DEVICE);
 		dev_kfree_skb_any(skb);
 		skb = __netdev_alloc_skb(bp->dev, RX_PKT_BUF_SZ, GFP_ATOMIC|GFP_DMA);
 		if (skb == NULL)
 			return -ENOMEM;
-		mapping = dma_map_single(bp->sdev->dev, skb->data,
+		mapping = dma_map_single(bp->sdev->dma_dev, skb->data,
 					 RX_PKT_BUF_SZ,
 					 DMA_FROM_DEVICE);
 		if (dma_mapping_error(mapping) ||
 			mapping + RX_PKT_BUF_SZ > DMA_30BIT_MASK) {
 			if (!dma_mapping_error(mapping))
-				dma_unmap_single(bp->sdev->dev, mapping, RX_PKT_BUF_SZ,DMA_FROM_DEVICE);
+				dma_unmap_single(bp->sdev->dma_dev, mapping, RX_PKT_BUF_SZ,DMA_FROM_DEVICE);
 			dev_kfree_skb_any(skb);
 			return -ENOMEM;
 		}
@@ -750,7 +750,7 @@ static void b44_recycle_rx(struct b44 *bp, int src_idx, u32 dest_idx_unmasked)
 					     dest_idx * sizeof(dest_desc),
 					     DMA_BIDIRECTIONAL);
 
-	dma_sync_single_for_device(bp->sdev->dev, le32_to_cpu(src_desc->addr),
+	dma_sync_single_for_device(bp->sdev->dma_dev, le32_to_cpu(src_desc->addr),
 				   RX_PKT_BUF_SZ,
 				   DMA_FROM_DEVICE);
 }
@@ -772,7 +772,7 @@ static int b44_rx(struct b44 *bp, int budget)
 		struct rx_header *rh;
 		u16 len;
 
-		dma_sync_single_for_cpu(bp->sdev->dev, map,
+		dma_sync_single_for_cpu(bp->sdev->dma_dev, map,
 					    RX_PKT_BUF_SZ,
 					    DMA_FROM_DEVICE);
 		rh = (struct rx_header *) skb->data;
@@ -806,7 +806,7 @@ static int b44_rx(struct b44 *bp, int budget)
 			skb_size = b44_alloc_rx_skb(bp, cons, bp->rx_prod);
 			if (skb_size < 0)
 				goto drop_it;
-			dma_unmap_single(bp->sdev->dev, map,
+			dma_unmap_single(bp->sdev->dma_dev, map,
 					 skb_size, DMA_FROM_DEVICE);
 			/* Leave out rx_header */
                 	skb_put(skb, len + RX_PKT_OFFSET);
@@ -966,24 +966,24 @@ static int b44_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto err_out;
 	}
 
-	mapping = dma_map_single(bp->sdev->dev, skb->data, len, DMA_TO_DEVICE);
+	mapping = dma_map_single(bp->sdev->dma_dev, skb->data, len, DMA_TO_DEVICE);
 	if (dma_mapping_error(mapping) || mapping + len > DMA_30BIT_MASK) {
 		struct sk_buff *bounce_skb;
 
 		/* Chip can't handle DMA to/from >1GB, use bounce buffer */
 		if (!dma_mapping_error(mapping))
-			dma_unmap_single(bp->sdev->dev, mapping, len,
+			dma_unmap_single(bp->sdev->dma_dev, mapping, len,
 					DMA_TO_DEVICE);
 
 		bounce_skb = __dev_alloc_skb(len, GFP_ATOMIC | GFP_DMA);
 		if (!bounce_skb)
 			goto err_out;
 
-		mapping = dma_map_single(bp->sdev->dev, bounce_skb->data,
+		mapping = dma_map_single(bp->sdev->dma_dev, bounce_skb->data,
 					 len, DMA_TO_DEVICE);
 		if (dma_mapping_error(mapping) || mapping + len > DMA_30BIT_MASK) {
 			if (!dma_mapping_error(mapping))
-				dma_unmap_single(bp->sdev->dev, mapping,
+				dma_unmap_single(bp->sdev->dma_dev, mapping,
 					 len, DMA_TO_DEVICE);
 			dev_kfree_skb_any(bounce_skb);
 			goto err_out;
@@ -1082,7 +1082,7 @@ static void b44_free_rings(struct b44 *bp)
 
 		if (rp->skb == NULL)
 			continue;
-		dma_unmap_single(bp->sdev->dev, rp->mapping, RX_PKT_BUF_SZ,
+		dma_unmap_single(bp->sdev->dma_dev, rp->mapping, RX_PKT_BUF_SZ,
 					DMA_FROM_DEVICE);
 		dev_kfree_skb_any(rp->skb);
 		rp->skb = NULL;
@@ -1094,7 +1094,7 @@ static void b44_free_rings(struct b44 *bp)
 
 		if (rp->skb == NULL)
 			continue;
-		dma_unmap_single(bp->sdev->dev, rp->mapping, rp->skb->len,
+		dma_unmap_single(bp->sdev->dma_dev, rp->mapping, rp->skb->len,
 					DMA_TO_DEVICE);
 		dev_kfree_skb_any(rp->skb);
 		rp->skb = NULL;
@@ -1117,12 +1117,12 @@ static void b44_init_rings(struct b44 *bp)
 	memset(bp->tx_ring, 0, B44_TX_RING_BYTES);
 
 	if (bp->flags & B44_FLAG_RX_RING_HACK)
-		dma_sync_single_for_device(bp->sdev->dev, bp->rx_ring_dma,
+		dma_sync_single_for_device(bp->sdev->dma_dev, bp->rx_ring_dma,
 			                  DMA_TABLE_BYTES,
 			                  DMA_BIDIRECTIONAL);
 
 	if (bp->flags & B44_FLAG_TX_RING_HACK)
-		dma_sync_single_for_device(bp->sdev->dev, bp->tx_ring_dma,
+		dma_sync_single_for_device(bp->sdev->dma_dev, bp->tx_ring_dma,
 			                  DMA_TABLE_BYTES,
 			                  DMA_TO_DEVICE);
 
@@ -1144,24 +1144,24 @@ static void b44_free_consistent(struct b44 *bp)
 	bp->tx_buffers = NULL;
 	if (bp->rx_ring) {
 		if (bp->flags & B44_FLAG_RX_RING_HACK) {
-			dma_unmap_single(bp->sdev->dev, bp->rx_ring_dma,
+			dma_unmap_single(bp->sdev->dma_dev, bp->rx_ring_dma,
 					DMA_TABLE_BYTES,
 					DMA_BIDIRECTIONAL);
 			kfree(bp->rx_ring);
 		} else
-			dma_free_coherent(bp->sdev->dev, DMA_TABLE_BYTES,
+			dma_free_coherent(bp->sdev->dma_dev, DMA_TABLE_BYTES,
 					    bp->rx_ring, bp->rx_ring_dma);
 		bp->rx_ring = NULL;
 		bp->flags &= ~B44_FLAG_RX_RING_HACK;
 	}
 	if (bp->tx_ring) {
 		if (bp->flags & B44_FLAG_TX_RING_HACK) {
-			dma_unmap_single(bp->sdev->dev, bp->tx_ring_dma,
+			dma_unmap_single(bp->sdev->dma_dev, bp->tx_ring_dma,
 					DMA_TABLE_BYTES,
 					DMA_TO_DEVICE);
 			kfree(bp->tx_ring);
 		} else
-			dma_free_coherent(bp->sdev->dev, DMA_TABLE_BYTES,
+			dma_free_coherent(bp->sdev->dma_dev, DMA_TABLE_BYTES,
 					    bp->tx_ring, bp->tx_ring_dma);
 		bp->tx_ring = NULL;
 		bp->flags &= ~B44_FLAG_TX_RING_HACK;
@@ -1187,7 +1187,7 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		goto out_err;
 
 	size = DMA_TABLE_BYTES;
-	bp->rx_ring = dma_alloc_coherent(bp->sdev->dev, size, &bp->rx_ring_dma, gfp);
+	bp->rx_ring = dma_alloc_coherent(bp->sdev->dma_dev, size, &bp->rx_ring_dma, gfp);
 	if (!bp->rx_ring) {
 		/* Allocation may have failed due to pci_alloc_consistent
 		   insisting on use of GFP_DMA, which is more restrictive
@@ -1199,7 +1199,7 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		if (!rx_ring)
 			goto out_err;
 
-		rx_ring_dma = dma_map_single(bp->sdev->dev, rx_ring,
+		rx_ring_dma = dma_map_single(bp->sdev->dma_dev, rx_ring,
 			                    DMA_TABLE_BYTES,
 			                    DMA_BIDIRECTIONAL);
 
@@ -1214,7 +1214,7 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		bp->flags |= B44_FLAG_RX_RING_HACK;
 	}
 
-	bp->tx_ring = dma_alloc_coherent(bp->sdev->dev, size, &bp->tx_ring_dma, gfp);
+	bp->tx_ring = dma_alloc_coherent(bp->sdev->dma_dev, size, &bp->tx_ring_dma, gfp);
 	if (!bp->tx_ring) {
 		/* Allocation may have failed due to dma_alloc_coherent
 		   insisting on use of GFP_DMA, which is more restrictive
@@ -1226,7 +1226,7 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		if (!tx_ring)
 			goto out_err;
 
-		tx_ring_dma = dma_map_single(bp->sdev->dev, tx_ring,
+		tx_ring_dma = dma_map_single(bp->sdev->dma_dev, tx_ring,
 			                    DMA_TABLE_BYTES,
 			                    DMA_TO_DEVICE);
 
diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c
index 948eb1fe916b..48e912487b16 100644
--- a/drivers/net/wireless/b43/dma.c
+++ b/drivers/net/wireless/b43/dma.c
@@ -373,10 +373,10 @@ static inline
 	dma_addr_t dmaaddr;
 
 	if (tx) {
-		dmaaddr = dma_map_single(ring->dev->dev->dev,
+		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
 					 buf, len, DMA_TO_DEVICE);
 	} else {
-		dmaaddr = dma_map_single(ring->dev->dev->dev,
+		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
 					 buf, len, DMA_FROM_DEVICE);
 	}
 
@@ -388,9 +388,10 @@ static inline
 			  dma_addr_t addr, size_t len, int tx)
 {
 	if (tx) {
-		dma_unmap_single(ring->dev->dev->dev, addr, len, DMA_TO_DEVICE);
+		dma_unmap_single(ring->dev->dev->dma_dev,
+				 addr, len, DMA_TO_DEVICE);
 	} else {
-		dma_unmap_single(ring->dev->dev->dev,
+		dma_unmap_single(ring->dev->dev->dma_dev,
 				 addr, len, DMA_FROM_DEVICE);
 	}
 }
@@ -400,7 +401,7 @@ static inline
 				 dma_addr_t addr, size_t len)
 {
 	B43_WARN_ON(ring->tx);
-	dma_sync_single_for_cpu(ring->dev->dev->dev,
+	dma_sync_single_for_cpu(ring->dev->dev->dma_dev,
 				addr, len, DMA_FROM_DEVICE);
 }
 
@@ -409,7 +410,7 @@ static inline
 				    dma_addr_t addr, size_t len)
 {
 	B43_WARN_ON(ring->tx);
-	dma_sync_single_for_device(ring->dev->dev->dev,
+	dma_sync_single_for_device(ring->dev->dev->dma_dev,
 				   addr, len, DMA_FROM_DEVICE);
 }
 
@@ -425,7 +426,7 @@ static inline
 
 static int alloc_ringmemory(struct b43_dmaring *ring)
 {
-	struct device *dev = ring->dev->dev->dev;
+	struct device *dma_dev = ring->dev->dev->dma_dev;
 	gfp_t flags = GFP_KERNEL;
 
 	/* The specs call for 4K buffers for 30- and 32-bit DMA with 4K
@@ -439,7 +440,7 @@ static int alloc_ringmemory(struct b43_dmaring *ring)
 	 */
 	if (ring->type == B43_DMA_64BIT)
 		flags |= GFP_DMA;
-	ring->descbase = dma_alloc_coherent(dev, B43_DMA_RINGMEMSIZE,
+	ring->descbase = dma_alloc_coherent(dma_dev, B43_DMA_RINGMEMSIZE,
 					    &(ring->dmabase), flags);
 	if (!ring->descbase) {
 		b43err(ring->dev->wl, "DMA ringmemory allocation failed\n");
@@ -452,9 +453,9 @@ static int alloc_ringmemory(struct b43_dmaring *ring)
 
 static void free_ringmemory(struct b43_dmaring *ring)
 {
-	struct device *dev = ring->dev->dev->dev;
+	struct device *dma_dev = ring->dev->dev->dma_dev;
 
-	dma_free_coherent(dev, B43_DMA_RINGMEMSIZE,
+	dma_free_coherent(dma_dev, B43_DMA_RINGMEMSIZE,
 			  ring->descbase, ring->dmabase);
 }
 
@@ -854,7 +855,7 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			goto err_kfree_meta;
 
 		/* test for ability to dma to txhdr_cache */
-		dma_test = dma_map_single(dev->dev->dev,
+		dma_test = dma_map_single(dev->dev->dma_dev,
 					  ring->txhdr_cache,
 					  b43_txhdr_size(dev),
 					  DMA_TO_DEVICE);
@@ -869,7 +870,7 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			if (!ring->txhdr_cache)
 				goto err_kfree_meta;
 
-			dma_test = dma_map_single(dev->dev->dev,
+			dma_test = dma_map_single(dev->dev->dma_dev,
 						  ring->txhdr_cache,
 						  b43_txhdr_size(dev),
 						  DMA_TO_DEVICE);
@@ -883,7 +884,7 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			}
 		}
 
-		dma_unmap_single(dev->dev->dev,
+		dma_unmap_single(dev->dev->dma_dev,
 				 dma_test, b43_txhdr_size(dev),
 				 DMA_TO_DEVICE);
 	}
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index 72017bf2e577..8003a9e55ac4 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -436,15 +436,18 @@ static int ssb_devices_register(struct ssb_bus *bus)
 #ifdef CONFIG_SSB_PCIHOST
 			sdev->irq = bus->host_pci->irq;
 			dev->parent = &bus->host_pci->dev;
+			sdev->dma_dev = &bus->host_pci->dev;
 #endif
 			break;
 		case SSB_BUSTYPE_PCMCIA:
 #ifdef CONFIG_SSB_PCMCIAHOST
 			sdev->irq = bus->host_pcmcia->irq.AssignedIRQ;
 			dev->parent = &bus->host_pcmcia->dev;
+			sdev->dma_dev = &bus->host_pcmcia->dev;
 #endif
 			break;
 		case SSB_BUSTYPE_SSB:
+			sdev->dma_dev = dev;
 			break;
 		}
 
@@ -1018,15 +1021,14 @@ EXPORT_SYMBOL(ssb_dma_translation);
 
 int ssb_dma_set_mask(struct ssb_device *ssb_dev, u64 mask)
 {
-	struct device *dev = ssb_dev->dev;
+	struct device *dma_dev = ssb_dev->dma_dev;
 
 #ifdef CONFIG_SSB_PCIHOST
-	if (ssb_dev->bus->bustype == SSB_BUSTYPE_PCI &&
-	    !dma_supported(dev, mask))
-		return -EIO;
+	if (ssb_dev->bus->bustype == SSB_BUSTYPE_PCI)
+		return dma_set_mask(dma_dev, mask);
 #endif
-	dev->coherent_dma_mask = mask;
-	dev->dma_mask = &dev->coherent_dma_mask;
+	dma_dev->coherent_dma_mask = mask;
+	dma_dev->dma_mask = &dma_dev->coherent_dma_mask;
 
 	return 0;
 }
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 20add65215af..db53defde5ee 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -129,6 +129,10 @@ struct ssb_device {
 	const struct ssb_bus_ops *ops;
 
 	struct device *dev;
+	/* Pointer to the device that has to be used for
+	 * any DMA related operation. */
+	struct device *dma_dev;
+
 	struct ssb_bus *bus;
 	struct ssb_device_id id;
 
-- 
cgit v1.2.3


From 1f4deba80a0d9fa92832684e683335b742a530e8 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Tue, 15 Apr 2008 14:34:38 -0700
Subject: uml: compile error fix

This patch fixes this error:

In file included from /home/wangcong/projects/linux-2.6/arch/um/kernel/smp.c:9:
include2/asm/tlb.h: In function `tlb_remove_page':
include2/asm/tlb.h:101: error: implicit declaration of function `page_cache_release'

And since including <linux/pagemap.h> in <linux/swap.h> will break sparc,
we add this #include in uml's own header.

Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: WANG Cong <wangcong@zeuux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-um/tlb.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-um/tlb.h b/include/asm-um/tlb.h
index 39fc475df6c9..5240fa1c5e08 100644
--- a/include/asm-um/tlb.h
+++ b/include/asm-um/tlb.h
@@ -1,6 +1,7 @@
 #ifndef __UM_TLB_H
 #define __UM_TLB_H
 
+#include <linux/pagemap.h>
 #include <linux/swap.h>
 #include <asm/percpu.h>
 #include <asm/pgalloc.h>
-- 
cgit v1.2.3


From 335e92e8a515420bd47a6b0f01cb9a206c0ed6e4 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Tue, 15 Apr 2008 14:34:43 -0700
Subject: vfs: fix possible deadlock in ext2, ext3, ext4 when using xattrs

mb_cache_entry_alloc() was allocating cache entries with GFP_KERNEL.  But
filesystems are calling this function while holding xattr_sem so possible
recursion into the fs violates locking ordering of xattr_sem and transaction
start / i_mutex for ext2-4.  Change mb_cache_entry_alloc() so that filesystems
can specify desired gfp mask and use GFP_NOFS from all of them.

Signed-off-by: Jan Kara <jack@suse.cz>
Reported-by: Dave Jones <davej@redhat.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext2/xattr.c         | 2 +-
 fs/ext3/xattr.c         | 2 +-
 fs/ext4/xattr.c         | 2 +-
 fs/mbcache.c            | 4 ++--
 include/linux/mbcache.h | 2 +-
 5 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index 3e8683dbb13f..a99d46f3b26e 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -835,7 +835,7 @@ ext2_xattr_cache_insert(struct buffer_head *bh)
 	struct mb_cache_entry *ce;
 	int error;
 
-	ce = mb_cache_entry_alloc(ext2_xattr_cache);
+	ce = mb_cache_entry_alloc(ext2_xattr_cache, GFP_NOFS);
 	if (!ce)
 		return -ENOMEM;
 	error = mb_cache_entry_insert(ce, bh->b_bdev, bh->b_blocknr, &hash);
diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c
index a6ea4d6a8bb2..42856541e9a5 100644
--- a/fs/ext3/xattr.c
+++ b/fs/ext3/xattr.c
@@ -1126,7 +1126,7 @@ ext3_xattr_cache_insert(struct buffer_head *bh)
 	struct mb_cache_entry *ce;
 	int error;
 
-	ce = mb_cache_entry_alloc(ext3_xattr_cache);
+	ce = mb_cache_entry_alloc(ext3_xattr_cache, GFP_NOFS);
 	if (!ce) {
 		ea_bdebug(bh, "out of memory");
 		return;
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index d7962139c010..e9054c1c7d93 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -1386,7 +1386,7 @@ ext4_xattr_cache_insert(struct buffer_head *bh)
 	struct mb_cache_entry *ce;
 	int error;
 
-	ce = mb_cache_entry_alloc(ext4_xattr_cache);
+	ce = mb_cache_entry_alloc(ext4_xattr_cache, GFP_NOFS);
 	if (!ce) {
 		ea_bdebug(bh, "out of memory");
 		return;
diff --git a/fs/mbcache.c b/fs/mbcache.c
index eb31b73e7d69..ec88ff3d04a9 100644
--- a/fs/mbcache.c
+++ b/fs/mbcache.c
@@ -399,11 +399,11 @@ mb_cache_destroy(struct mb_cache *cache)
  * if no more memory was available.
  */
 struct mb_cache_entry *
-mb_cache_entry_alloc(struct mb_cache *cache)
+mb_cache_entry_alloc(struct mb_cache *cache, gfp_t gfp_flags)
 {
 	struct mb_cache_entry *ce;
 
-	ce = kmem_cache_alloc(cache->c_entry_cache, GFP_KERNEL);
+	ce = kmem_cache_alloc(cache->c_entry_cache, gfp_flags);
 	if (ce) {
 		atomic_inc(&cache->c_entry_count);
 		INIT_LIST_HEAD(&ce->e_lru_list);
diff --git a/include/linux/mbcache.h b/include/linux/mbcache.h
index 99e044b4efc6..a09b84e4fdb4 100644
--- a/include/linux/mbcache.h
+++ b/include/linux/mbcache.h
@@ -34,7 +34,7 @@ void mb_cache_destroy(struct mb_cache *);
 
 /* Functions on cache entries */
 
-struct mb_cache_entry *mb_cache_entry_alloc(struct mb_cache *);
+struct mb_cache_entry *mb_cache_entry_alloc(struct mb_cache *, gfp_t);
 int mb_cache_entry_insert(struct mb_cache_entry *, struct block_device *,
 			  sector_t, unsigned int[]);
 void mb_cache_entry_release(struct mb_cache_entry *);
-- 
cgit v1.2.3


From d1e7780638a9192f15caf590e0081bf915fdef71 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 15 Apr 2008 14:34:46 -0700
Subject: spi: spi_s3c24xx must initialize num_chipselect

The SPI core now expects num_chipselect to be set correctly as due to added
checks on the chip being selected before an transfer is allowed.  This patch
adds a num_cs field to the platform data which needs to be set correctly
before adding the SPI platform device.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/spi_s3c24xx.c          | 4 ++++
 include/asm-arm/arch-s3c2410/spi.h | 1 +
 2 files changed, 5 insertions(+)

(limited to 'include')

diff --git a/drivers/spi/spi_s3c24xx.c b/drivers/spi/spi_s3c24xx.c
index 80719ef54365..b7476b888197 100644
--- a/drivers/spi/spi_s3c24xx.c
+++ b/drivers/spi/spi_s3c24xx.c
@@ -267,6 +267,10 @@ static int __init s3c24xx_spi_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, hw);
 	init_completion(&hw->done);
 
+	/* setup the master state. */
+
+	master->num_chipselect = hw->pdata->num_cs;
+
 	/* setup the state for the bitbang driver */
 
 	hw->bitbang.master         = hw->master;
diff --git a/include/asm-arm/arch-s3c2410/spi.h b/include/asm-arm/arch-s3c2410/spi.h
index 7ca0ed97a6d0..352d33860b63 100644
--- a/include/asm-arm/arch-s3c2410/spi.h
+++ b/include/asm-arm/arch-s3c2410/spi.h
@@ -15,6 +15,7 @@
 
 struct s3c2410_spi_info {
 	unsigned long		 pin_cs;	/* simple gpio cs */
+	unsigned int		 num_cs;	/* total chipselects */
 
 	void (*set_cs)(struct s3c2410_spi_info *spi, int cs, int pol);
 };
-- 
cgit v1.2.3


From 5f1a3f2ac4ddf966cd0555dc445c6df65709c580 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Tue, 15 Apr 2008 14:34:47 -0700
Subject: acpi thermal trip points increased to 12

The THERMAL_MAX_TRIPS value is set to 10.  It is too few for the Compaq AP550
machine which has 12 trip points.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: Len Brown <lenb@kernel.org>
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/thermal/thermal.c | 4 ++++
 include/linux/thermal.h   | 2 +-
 2 files changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/thermal/thermal.c b/drivers/thermal/thermal.c
index 8b86e53ccf7a..7f79bbf652d7 100644
--- a/drivers/thermal/thermal.c
+++ b/drivers/thermal/thermal.c
@@ -196,6 +196,10 @@ static struct device_attribute trip_point_attrs[] = {
 	__ATTR(trip_point_8_temp, 0444, trip_point_temp_show, NULL),
 	__ATTR(trip_point_9_type, 0444, trip_point_type_show, NULL),
 	__ATTR(trip_point_9_temp, 0444, trip_point_temp_show, NULL),
+	__ATTR(trip_point_10_type, 0444, trip_point_type_show, NULL),
+	__ATTR(trip_point_10_temp, 0444, trip_point_temp_show, NULL),
+	__ATTR(trip_point_11_type, 0444, trip_point_type_show, NULL),
+	__ATTR(trip_point_11_temp, 0444, trip_point_temp_show, NULL),
 };
 
 #define TRIP_POINT_ATTR_ADD(_dev, _index, result)     \
diff --git a/include/linux/thermal.h b/include/linux/thermal.h
index 818ca1cf0b6d..90c1c191ea69 100644
--- a/include/linux/thermal.h
+++ b/include/linux/thermal.h
@@ -50,7 +50,7 @@ struct thermal_cooling_device_ops {
 };
 
 #define THERMAL_TRIPS_NONE -1
-#define THERMAL_MAX_TRIPS 10
+#define THERMAL_MAX_TRIPS 12
 #define THERMAL_NAME_LENGTH 20
 struct thermal_cooling_device {
 	int id;
-- 
cgit v1.2.3


From 669f87baab90183e13b95480aecf8d7bac92ca3c Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Apr 2008 00:46:52 -0700
Subject: [RTNL]: Introduce the rtnl_kill_links helper.

This one is responsible for calling ->dellink on each net
device found in net to help with vlan net_exit hook in the
nearest future.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/rtnetlink.h |  1 +
 net/core/rtnetlink.c    | 29 +++++++++++++++++++++--------
 2 files changed, 22 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h
index 793863e09c69..3c1895e54b7f 100644
--- a/include/net/rtnetlink.h
+++ b/include/net/rtnetlink.h
@@ -74,6 +74,7 @@ struct rtnl_link_ops {
 
 extern int	__rtnl_link_register(struct rtnl_link_ops *ops);
 extern void	__rtnl_link_unregister(struct rtnl_link_ops *ops);
+extern void	rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops);
 
 extern int	rtnl_link_register(struct rtnl_link_ops *ops);
 extern void	rtnl_link_unregister(struct rtnl_link_ops *ops);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index edc6dbfe48f2..bc39e417694a 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -269,6 +269,26 @@ int rtnl_link_register(struct rtnl_link_ops *ops)
 
 EXPORT_SYMBOL_GPL(rtnl_link_register);
 
+static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops)
+{
+	struct net_device *dev;
+restart:
+	for_each_netdev(net, dev) {
+		if (dev->rtnl_link_ops == ops) {
+			ops->dellink(dev);
+			goto restart;
+		}
+	}
+}
+
+void rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops)
+{
+	rtnl_lock();
+	__rtnl_kill_links(net, ops);
+	rtnl_unlock();
+}
+EXPORT_SYMBOL_GPL(rtnl_kill_links);
+
 /**
  * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink.
  * @ops: struct rtnl_link_ops * to unregister
@@ -277,17 +297,10 @@ EXPORT_SYMBOL_GPL(rtnl_link_register);
  */
 void __rtnl_link_unregister(struct rtnl_link_ops *ops)
 {
-	struct net_device *dev;
 	struct net *net;
 
 	for_each_net(net) {
-restart:
-		for_each_netdev(net, dev) {
-			if (dev->rtnl_link_ops == ops) {
-				ops->dellink(dev);
-				goto restart;
-			}
-		}
+		__rtnl_kill_links(net, ops);
 	}
 	list_del(&ops->list);
 }
-- 
cgit v1.2.3


From a9fde2607895667823e9d1172fc193087125ef68 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Wed, 16 Apr 2008 00:48:04 -0700
Subject: [VLAN]: Tag vlan_group_device with net device, not ifindex.

Currently vlan group is searched using one key - the ifindex.
We'll have to lookup the vlan_group by two keys - ifindex and
net. Turning the vlan_group lookup key to struct net_device
pointer will make this process easier.

Besides, this will eliminate one more place in the networking,
that assumes that indexes are unique in the kernel.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h |  4 +++-
 net/8021q/vlan.c        | 22 +++++++++++-----------
 2 files changed, 14 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index edd55af7ebd6..15ace02b7b24 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -81,7 +81,9 @@ extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *));
 #define VLAN_GROUP_ARRAY_PART_LEN     (VLAN_GROUP_ARRAY_LEN/VLAN_GROUP_ARRAY_SPLIT_PARTS)
 
 struct vlan_group {
-	int real_dev_ifindex; /* The ifindex of the ethernet(like) device the vlan is attached to. */
+	struct net_device	*real_dev; /* The ethernet(like) device
+					    * the vlan is attached to.
+					    */
 	unsigned int		nr_vlans;
 	struct hlist_node	hlist;	/* linked list */
 	struct net_device **vlan_devices_arrays[VLAN_GROUP_ARRAY_SPLIT_PARTS];
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 5975ec3be7f3..cf8d810a130d 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -65,14 +65,14 @@ static inline unsigned int vlan_grp_hashfn(unsigned int idx)
 }
 
 /* Must be invoked with RCU read lock (no preempt) */
-static struct vlan_group *__vlan_find_group(int real_dev_ifindex)
+static struct vlan_group *__vlan_find_group(struct net_device *real_dev)
 {
 	struct vlan_group *grp;
 	struct hlist_node *n;
-	int hash = vlan_grp_hashfn(real_dev_ifindex);
+	int hash = vlan_grp_hashfn(real_dev->ifindex);
 
 	hlist_for_each_entry_rcu(grp, n, &vlan_group_hash[hash], hlist) {
-		if (grp->real_dev_ifindex == real_dev_ifindex)
+		if (grp->real_dev == real_dev)
 			return grp;
 	}
 
@@ -86,7 +86,7 @@ static struct vlan_group *__vlan_find_group(int real_dev_ifindex)
 struct net_device *__find_vlan_dev(struct net_device *real_dev,
 				   unsigned short VID)
 {
-	struct vlan_group *grp = __vlan_find_group(real_dev->ifindex);
+	struct vlan_group *grp = __vlan_find_group(real_dev);
 
 	if (grp)
 		return vlan_group_get_device(grp, VID);
@@ -103,7 +103,7 @@ static void vlan_group_free(struct vlan_group *grp)
 	kfree(grp);
 }
 
-static struct vlan_group *vlan_group_alloc(int ifindex)
+static struct vlan_group *vlan_group_alloc(struct net_device *real_dev)
 {
 	struct vlan_group *grp;
 
@@ -111,9 +111,9 @@ static struct vlan_group *vlan_group_alloc(int ifindex)
 	if (!grp)
 		return NULL;
 
-	grp->real_dev_ifindex = ifindex;
+	grp->real_dev = real_dev;
 	hlist_add_head_rcu(&grp->hlist,
-			   &vlan_group_hash[vlan_grp_hashfn(ifindex)]);
+			&vlan_group_hash[vlan_grp_hashfn(real_dev->ifindex)]);
 	return grp;
 }
 
@@ -151,7 +151,7 @@ void unregister_vlan_dev(struct net_device *dev)
 
 	ASSERT_RTNL();
 
-	grp = __vlan_find_group(real_dev->ifindex);
+	grp = __vlan_find_group(real_dev);
 	BUG_ON(!grp);
 
 	vlan_proc_rem_dev(dev);
@@ -246,9 +246,9 @@ int register_vlan_dev(struct net_device *dev)
 	struct vlan_group *grp, *ngrp = NULL;
 	int err;
 
-	grp = __vlan_find_group(real_dev->ifindex);
+	grp = __vlan_find_group(real_dev);
 	if (!grp) {
-		ngrp = grp = vlan_group_alloc(real_dev->ifindex);
+		ngrp = grp = vlan_group_alloc(real_dev);
 		if (!grp)
 			return -ENOBUFS;
 	}
@@ -412,7 +412,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 		goto out;
 	}
 
-	grp = __vlan_find_group(dev->ifindex);
+	grp = __vlan_find_group(dev);
 	if (!grp)
 		goto out;
 
-- 
cgit v1.2.3


From 5d1e4468a7705db7c1415a65fd16f07113afc1b2 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 16 Apr 2008 01:58:04 -0700
Subject: [NETNS]: Make netns refconting debug like a socket one.

Make release_net/hold_net noop for performance-hungry people. This is a debug
staff and should be used in the debug mode only.

Add check for net != NULL in hold/release calls. This will be required
later on.

[ Added minor simplifications suggested by Brian Haley. -DaveM ]

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/net_namespace.h | 40 ++++++++++++++++++++++++----------------
 net/core/net_namespace.c    |  4 ++++
 2 files changed, 28 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index f880b0f9f107..aa540e6be502 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -26,9 +26,11 @@ struct net {
 	atomic_t		count;		/* To decided when the network
 						 *  namespace should be freed.
 						 */
+#ifdef NETNS_REFCNT_DEBUG
 	atomic_t		use_count;	/* To track references we
 						 * destroy on demand
 						 */
+#endif
 	struct list_head	list;		/* list of network namespaces */
 	struct work_struct	work;		/* work struct for freeing */
 
@@ -117,17 +119,6 @@ static inline void put_net(struct net *net)
 		__put_net(net);
 }
 
-static inline struct net *hold_net(struct net *net)
-{
-	atomic_inc(&net->use_count);
-	return net;
-}
-
-static inline void release_net(struct net *net)
-{
-	atomic_dec(&net->use_count);
-}
-
 static inline
 int net_eq(const struct net *net1, const struct net *net2)
 {
@@ -143,27 +134,44 @@ static inline void put_net(struct net *net)
 {
 }
 
+static inline struct net *maybe_get_net(struct net *net)
+{
+	return net;
+}
+
+static inline
+int net_eq(const struct net *net1, const struct net *net2)
+{
+	return 1;
+}
+#endif
+
+
+#ifdef NETNS_REFCNT_DEBUG
 static inline struct net *hold_net(struct net *net)
 {
+	if (net)
+		atomic_inc(&net->use_count);
 	return net;
 }
 
 static inline void release_net(struct net *net)
 {
+	if (net)
+		atomic_dec(&net->use_count);
 }
-
-static inline struct net *maybe_get_net(struct net *net)
+#else
+static inline struct net *hold_net(struct net *net)
 {
 	return net;
 }
 
-static inline
-int net_eq(const struct net *net1, const struct net *net2)
+static inline void release_net(struct net *net)
 {
-	return 1;
 }
 #endif
 
+
 #define for_each_net(VAR)				\
 	list_for_each_entry(VAR, &net_namespace_list, list)
 
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 763674e1e593..72b4c184dd84 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -35,7 +35,9 @@ static __net_init int setup_net(struct net *net)
 	struct net_generic *ng;
 
 	atomic_set(&net->count, 1);
+#ifdef NETNS_REFCNT_DEBUG
 	atomic_set(&net->use_count, 0);
+#endif
 
 	error = -ENOMEM;
 	ng = kzalloc(sizeof(struct net_generic) +
@@ -86,11 +88,13 @@ static void net_free(struct net *net)
 	if (!net)
 		return;
 
+#ifdef NETNS_REFCNT_DEBUG
 	if (unlikely(atomic_read(&net->use_count) != 0)) {
 		printk(KERN_EMERG "network namespace not free! Usage: %d\n",
 			atomic_read(&net->use_count));
 		return;
 	}
+#endif
 
 	kmem_cache_free(net_cachep, net);
 }
-- 
cgit v1.2.3


From 65a18ec58e5e6186103f62f720acea94dfb26f4e Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 16 Apr 2008 01:59:46 -0700
Subject: [NETNS]: Add netns refcnt debug for kernel sockets.

Protocol control sockets and netlink kernel sockets should not prevent the
namespace stop request. They are initialized and disposed in a special way by
sk_change_net/sk_release_kernel.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sock.h | 2 +-
 net/core/sock.c    | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/sock.h b/include/net/sock.h
index 09255eae93e9..dc42b44c2aa1 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1314,7 +1314,7 @@ void sock_net_set(struct sock *sk, struct net *net)
 static inline void sk_change_net(struct sock *sk, struct net *net)
 {
 	put_net(sock_net(sk));
-	sock_net_set(sk, net);
+	sock_net_set(sk, hold_net(net));
 }
 
 extern void sock_enable_timestamp(struct sock *sk);
diff --git a/net/core/sock.c b/net/core/sock.c
index c0ecbdcf75d8..54c836a2216b 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1001,6 +1001,7 @@ void sk_release_kernel(struct sock *sk)
 
 	sock_hold(sk);
 	sock_release(sk->sk_socket);
+	release_net(sock_net(sk));
 	sock_net_set(sk, get_net(&init_net));
 	sock_put(sk);
 }
-- 
cgit v1.2.3


From 3661a910836a509be65afc3c1e512d900e1280f9 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 16 Apr 2008 02:01:56 -0700
Subject: [NETNS]: Add netns refcnt debug to fib rules.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/fib_rules.h | 1 +
 net/core/fib_rules.c    | 5 +++--
 2 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 34349f9f4331..a5c6ccc5bb19 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -87,6 +87,7 @@ static inline void fib_rule_get(struct fib_rule *rule)
 static inline void fib_rule_put_rcu(struct rcu_head *head)
 {
 	struct fib_rule *rule = container_of(head, struct fib_rule, rcu);
+	release_net(rule->fr_net);
 	kfree(rule);
 }
 
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 540c07283e31..e3e9ab0f74e3 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -29,7 +29,7 @@ int fib_default_rule_add(struct fib_rules_ops *ops,
 	r->pref = pref;
 	r->table = table;
 	r->flags = flags;
-	r->fr_net = ops->fro_net;
+	r->fr_net = hold_net(ops->fro_net);
 
 	/* The lock is not required here, the list in unreacheable
 	 * at the moment this function is called */
@@ -243,7 +243,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 		err = -ENOMEM;
 		goto errout;
 	}
-	rule->fr_net = net;
+	rule->fr_net = hold_net(net);
 
 	if (tb[FRA_PRIORITY])
 		rule->pref = nla_get_u32(tb[FRA_PRIORITY]);
@@ -344,6 +344,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	return 0;
 
 errout_free:
+	release_net(rule->fr_net);
 	kfree(rule);
 errout:
 	rules_ops_put(ops);
-- 
cgit v1.2.3


From f3005d7f4abe03ad41af33b1548602cd086d86a2 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Wed, 16 Apr 2008 02:02:18 -0700
Subject: [NETNS]: Add netns refcnt debug for network devices.

dev_set_net is called for
- just allocated devices
- devices moving from one namespace to another
release_net has proper check inside to distinguish these cases.

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 3 ++-
 net/core/dev.c            | 2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 8b17ed40dea2..7c1d4466583b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -758,7 +758,8 @@ static inline
 void dev_net_set(struct net_device *dev, struct net *net)
 {
 #ifdef CONFIG_NET_NS
-	dev->nd_net = net;
+	release_net(dev->nd_net);
+	dev->nd_net = hold_net(net);
 #endif
 }
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 7aa01125287e..77530e9a34fc 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4042,6 +4042,8 @@ EXPORT_SYMBOL(alloc_netdev_mq);
  */
 void free_netdev(struct net_device *dev)
 {
+	release_net(dev_net(dev));
+
 	/*  Compatibility with error handling in drivers */
 	if (dev->reg_state == NETREG_UNINITIALIZED) {
 		kfree((char *)dev - dev->padded);
-- 
cgit v1.2.3


From dd9e0dda66ba38a2ddd1405ac279894260dc5c36 Mon Sep 17 00:00:00 2001
From: John Heffner <johnwheffner@gmail.com>
Date: Tue, 15 Apr 2008 15:26:39 -0700
Subject: [TCP]: Increase the max_burst threshold from 3 to tp->reordering.

This change is necessary to allow cwnd to grow during persistent
reordering.  Cwnd moderation is applied when in the disorder state
and an ack that fills the hole comes in.  If the hole was greater
than 3 packets, but less than tp->reordering, cwnd will shrink when
it should not have.

Signed-off-by: John Heffner <jheffner@napa.(none)>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 2c14edf7b076..633147cb6bbc 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -787,11 +787,14 @@ extern void tcp_enter_cwr(struct sock *sk, const int set_ssthresh);
 extern __u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst);
 
 /* Slow start with delack produces 3 packets of burst, so that
- * it is safe "de facto".
+ * it is safe "de facto".  This will be the default - same as
+ * the default reordering threshold - but if reordering increases,
+ * we must be able to allow cwnd to burst at least this much in order
+ * to not pull it back when holes are filled.
  */
 static __inline__ __u32 tcp_max_burst(const struct tcp_sock *tp)
 {
-	return 3;
+	return tp->reordering;
 }
 
 /* Returns end sequence number of the receiver's advertised window */
-- 
cgit v1.2.3


From d18ef29f34eb33099d387a327abe139f3915a829 Mon Sep 17 00:00:00 2001
From: Reinette Chatre <reinette.chatre@intel.com>
Date: Wed, 9 Apr 2008 16:56:15 -0700
Subject: mac80211: no BSS changes to driver from beacons processed during
 scanning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is no need to send BSS changes to driver from beacons processed
during scanning. We are more interested in beacons from an AP with which
we are associated - these will still be used to send updates to driver as
the beacons are received without scanning.

This change·removes the requirement that bss_info_changed needs to be atomic.
The beacons received during scanning are processed from a tasklet, but if we
do not call bss_info_changed for these beacons there is no need for it to be
atomic. This function (bss_info_changed) is called either from workqueue or
ioctl in all other instances.

Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Acked-by: Tomas Winkler <tomas.winkler@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/mac80211.h |  3 +--
 net/mac80211/mlme.c    | 16 +++++++++++-----
 2 files changed, 12 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 079e7bd86c90..4a80d74975e8 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1020,8 +1020,7 @@ enum ieee80211_ampdu_mlme_action {
  *	level driver (e.g. assoc/disassoc status, erp parameters).
  *	This function should not be used if no BSS has been set, unless
  *	for association indication. The @changed parameter indicates which
- *	of the bss parameters has changed when a call is made. This callback
- *	has to be atomic.
+ *	of the bss parameters has changed when a call is made.
  *
  * @configure_filter: Configure the device's RX filter.
  *	See the section "Frame filtering" for more information.
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 83e8b497e6db..e3f2cb086588 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2797,6 +2797,17 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 
 	ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems);
 
+	if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) {
+		ieee80211_sta_wmm_params(dev, ifsta, elems.wmm_param,
+					 elems.wmm_param_len);
+	}
+
+	/* Do not send changes to driver if we are scanning. This removes
+	 * requirement that driver's bss_info_changed function needs to be
+	 * atomic. */
+	if (local->sta_sw_scanning || local->sta_hw_scanning)
+		return;
+
 	if (elems.erp_info && elems.erp_info_len >= 1)
 		changed |= ieee80211_handle_erp_ie(sdata, elems.erp_info[0]);
 	else {
@@ -2816,11 +2827,6 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev,
 					       &bss_info);
 	}
 
-	if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) {
-		ieee80211_sta_wmm_params(dev, ifsta, elems.wmm_param,
-					 elems.wmm_param_len);
-	}
-
 	ieee80211_bss_info_change_notify(sdata, changed);
 }
 
-- 
cgit v1.2.3


From 09a77441f27f51f5f9878c90c7ed67e303197a2a Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Date: Thu, 17 Apr 2008 01:14:33 +0200
Subject: Au1200: kill IDE driver function prototypes

Fix these warnings emitted when compiling drivers/ide/mips/au1xxx-ide.c:

include/asm/mach-au1x00/au1xxx_ide.h:137: warning: 'auide_tune_drive' declared
`static' but never defined
include/asm/mach-au1x00/au1xxx_ide.h:138: warning: 'auide_tune_chipset' declared
 `static' but never defined

by wiping out the whole "function prototyping" section from the header file
<asm-mips/mach-au1x00/au1xxx_ide.h> as it mostly declared functions that are
already dead in the IDE driver; move the only useful prototype into the driver.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/mips/au1xxx-ide.c             |  2 ++
 include/asm-mips/mach-au1x00/au1xxx_ide.h | 18 ------------------
 2 files changed, 2 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 0f4bf5d72835..da91060c320f 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -56,6 +56,8 @@
 static _auide_hwif auide_hwif;
 static int dbdma_init_done;
 
+static int auide_ddma_init(_auide_hwif *auide);
+
 #if defined(CONFIG_BLK_DEV_IDE_AU1XXX_PIO_DBDMA)
 
 void auide_insw(unsigned long port, void *addr, u32 count)
diff --git a/include/asm-mips/mach-au1x00/au1xxx_ide.h b/include/asm-mips/mach-au1x00/au1xxx_ide.h
index e4fe26c160ba..89655c0cdcd6 100644
--- a/include/asm-mips/mach-au1x00/au1xxx_ide.h
+++ b/include/asm-mips/mach-au1x00/au1xxx_ide.h
@@ -122,24 +122,6 @@ static const struct drive_list_entry dma_black_list [] = {
 };
 #endif
 
-/* function prototyping */
-u8 auide_inb(unsigned long port);
-u16 auide_inw(unsigned long port);
-u32 auide_inl(unsigned long port);
-void auide_insw(unsigned long port, void *addr, u32 count);
-void auide_insl(unsigned long port, void *addr, u32 count);
-void auide_outb(u8 addr, unsigned long port);
-void auide_outbsync(ide_drive_t *drive, u8 addr, unsigned long port);
-void auide_outw(u16 addr, unsigned long port);
-void auide_outl(u32 addr, unsigned long port);
-void auide_outsw(unsigned long port, void *addr, u32 count);
-void auide_outsl(unsigned long port, void *addr, u32 count);
-static void auide_tune_drive(ide_drive_t *drive, byte pio);
-static int auide_tune_chipset(ide_drive_t *drive, u8 speed);
-static int auide_ddma_init( _auide_hwif *auide );
-static void auide_setup_ports(hw_regs_t *hw, _auide_hwif *ahwif);
-int __init auide_probe(void);
-
 /*******************************************************************************
 * PIO Mode timing calculation :                                                *
 *                                                                              *
-- 
cgit v1.2.3


From b4dcaea36b0376456c97698deba0089d2d67cbe7 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Date: Thu, 17 Apr 2008 01:14:33 +0200
Subject: Pb1200/DBAu1200: fix bad IDE resource size

The header files for the Pb1200/DBAu1200 boards have wrong definition for the
IDE interface's decoded range length -- it should be 512 bytes according to
what the IDE driver does.  In addition, the IDE platform device claims 1 byte
too many for its memory resource -- fix the platform code and the IDE driver
in accordance.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 arch/mips/au1000/common/platform.c    | 2 +-
 drivers/ide/mips/au1xxx-ide.c         | 7 ++++---
 include/asm-mips/mach-db1x00/db1200.h | 4 ++--
 include/asm-mips/mach-pb1x00/pb1200.h | 4 ++--
 4 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/mips/au1000/common/platform.c b/arch/mips/au1000/common/platform.c
index 841904cdef4d..39d681265297 100644
--- a/arch/mips/au1000/common/platform.c
+++ b/arch/mips/au1000/common/platform.c
@@ -189,7 +189,7 @@ static struct resource au1200_lcd_resources[] = {
 static struct resource au1200_ide0_resources[] = {
 	[0] = {
 		.start		= AU1XXX_ATA_PHYS_ADDR,
-		.end 		= AU1XXX_ATA_PHYS_ADDR + AU1XXX_ATA_PHYS_LEN,
+		.end 		= AU1XXX_ATA_PHYS_ADDR + AU1XXX_ATA_PHYS_LEN - 1,
 		.flags		= IORESOURCE_MEM,
 	},
 	[1] = {
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index f84411b6609a..85c016bdfd38 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -586,13 +586,14 @@ static int au_ide_probe(struct device *dev)
 		goto out;
 	}
 
-	if (!request_mem_region (res->start, res->end-res->start, pdev->name)) {
+	if (!request_mem_region(res->start, res->end - res->start + 1,
+				pdev->name)) {
 		pr_debug("%s: request_mem_region failed\n", DRV_NAME);
 		ret =  -EBUSY;
 		goto out;
 	}
 
-	ahwif->regbase = (u32)ioremap(res->start, res->end-res->start);
+	ahwif->regbase = (u32)ioremap(res->start, res->end - res->start + 1);
 	if (ahwif->regbase == 0) {
 		ret = -ENOMEM;
 		goto out;
@@ -677,7 +678,7 @@ static int au_ide_remove(struct device *dev)
 	iounmap((void *)ahwif->regbase);
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	release_mem_region(res->start, res->end - res->start);
+	release_mem_region(res->start, res->end - res->start + 1);
 
 	return 0;
 }
diff --git a/include/asm-mips/mach-db1x00/db1200.h b/include/asm-mips/mach-db1x00/db1200.h
index a6bdac61ab49..d2e28e64932e 100644
--- a/include/asm-mips/mach-db1x00/db1200.h
+++ b/include/asm-mips/mach-db1x00/db1200.h
@@ -173,8 +173,8 @@ static BCSR * const bcsr = (BCSR *)BCSR_KSEG1_ADDR;
 #define AU1XXX_SMC91111_IRQ			DB1200_ETH_INT
 
 #define AU1XXX_ATA_PHYS_ADDR		(0x18800000)
-#define AU1XXX_ATA_PHYS_LEN			(0x100)
-#define AU1XXX_ATA_REG_OFFSET	(5)
+#define AU1XXX_ATA_REG_OFFSET		(5)
+#define AU1XXX_ATA_PHYS_LEN		(16 << AU1XXX_ATA_REG_OFFSET)
 #define AU1XXX_ATA_INT			DB1200_IDE_INT
 #define AU1XXX_ATA_DDMA_REQ		DSCR_CMD0_DMA_REQ1;
 #define AU1XXX_ATA_RQSIZE		128
diff --git a/include/asm-mips/mach-pb1x00/pb1200.h b/include/asm-mips/mach-pb1x00/pb1200.h
index 72213e3d02c7..edaa489b58f1 100644
--- a/include/asm-mips/mach-pb1x00/pb1200.h
+++ b/include/asm-mips/mach-pb1x00/pb1200.h
@@ -186,8 +186,8 @@ static BCSR * const bcsr = (BCSR *)BCSR_KSEG1_ADDR;
 #define AU1XXX_SMC91111_IRQ			PB1200_ETH_INT
 
 #define AU1XXX_ATA_PHYS_ADDR		(0x0C800000)
-#define AU1XXX_ATA_PHYS_LEN			(0x100)
-#define AU1XXX_ATA_REG_OFFSET	(5)
+#define AU1XXX_ATA_REG_OFFSET		(5)
+#define AU1XXX_ATA_PHYS_LEN		(16 << AU1XXX_ATA_REG_OFFSET)
 #define AU1XXX_ATA_INT			PB1200_IDE_INT
 #define AU1XXX_ATA_DDMA_REQ		DSCR_CMD0_DMA_REQ1;
 #define AU1XXX_ATA_RQSIZE		128
-- 
cgit v1.2.3


From 9d9326d3bc0ea9a8bbe40bf3e5e66c7b9858caa0 Mon Sep 17 00:00:00 2001
From: Andy Fleming <afleming@freescale.com>
Date: Wed, 9 Apr 2008 19:38:13 -0500
Subject: phy: Change mii_bus id field to a string

Having the id field be an int was making more complex bus topologies
excessively difficult.  For now, just convert it to a string, and
change all instances of "bus->id = val" to
snprintf(id, MII_BUS_ID_LEN, "%x", val).

Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 arch/powerpc/platforms/82xx/ep8248e.c     |  2 +-
 arch/powerpc/platforms/pasemi/gpio_mdio.c |  2 +-
 arch/powerpc/sysdev/fsl_soc.c             |  5 +++--
 drivers/net/au1000_eth.c                  |  6 +++---
 drivers/net/bfin_mac.c                    |  2 +-
 drivers/net/cpmac.c                       |  5 ++---
 drivers/net/fec_mpc52xx.c                 |  2 +-
 drivers/net/fec_mpc52xx_phy.c             |  2 +-
 drivers/net/fs_enet/fs_enet-main.c        |  4 ++--
 drivers/net/fs_enet/mii-bitbang.c         |  4 ++--
 drivers/net/fs_enet/mii-fec.c             |  4 ++--
 drivers/net/gianfar_mii.c                 |  2 +-
 drivers/net/macb.c                        |  2 +-
 drivers/net/pasemi_mac.c                  |  2 +-
 drivers/net/phy/fixed.c                   |  2 +-
 drivers/net/sb1250-mac.c                  |  2 +-
 drivers/net/ucc_geth.c                    |  2 +-
 drivers/net/ucc_geth.h                    |  2 +-
 drivers/net/ucc_geth_mii.c                |  2 +-
 include/linux/fsl_devices.h               |  2 +-
 include/linux/phy.h                       | 12 ++++++++----
 21 files changed, 36 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/82xx/ep8248e.c b/arch/powerpc/platforms/82xx/ep8248e.c
index ba93d8ae9b0c..d5770fdf7f09 100644
--- a/arch/powerpc/platforms/82xx/ep8248e.c
+++ b/arch/powerpc/platforms/82xx/ep8248e.c
@@ -138,7 +138,7 @@ static int __devinit ep8248e_mdio_probe(struct of_device *ofdev,
 
 	bus->name = "ep8248e-mdio-bitbang";
 	bus->dev = &ofdev->dev;
-	bus->id = res.start;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start);
 
 	return mdiobus_register(bus);
 }
diff --git a/arch/powerpc/platforms/pasemi/gpio_mdio.c b/arch/powerpc/platforms/pasemi/gpio_mdio.c
index b46542990cf8..ab6955412ba4 100644
--- a/arch/powerpc/platforms/pasemi/gpio_mdio.c
+++ b/arch/powerpc/platforms/pasemi/gpio_mdio.c
@@ -241,7 +241,7 @@ static int __devinit gpio_mdio_probe(struct of_device *ofdev,
 	new_bus->reset = &gpio_mdio_reset;
 
 	prop = of_get_property(np, "reg", NULL);
-	new_bus->id = *prop;
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", *prop);
 	new_bus->priv = priv;
 
 	new_bus->phy_mask = 0;
diff --git a/arch/powerpc/sysdev/fsl_soc.c b/arch/powerpc/sysdev/fsl_soc.c
index 2c5388ce902a..3581416905ea 100644
--- a/arch/powerpc/sysdev/fsl_soc.c
+++ b/arch/powerpc/sysdev/fsl_soc.c
@@ -341,7 +341,7 @@ static int __init gfar_of_init(void)
 				goto unreg;
 			}
 
-			gfar_data.bus_id = 0;
+			snprintf(gfar_data.bus_id, MII_BUS_ID_SIZE, "0");
 			gfar_data.phy_id = fixed_link[0];
 		} else {
 			phy = of_find_node_by_phandle(*ph);
@@ -362,7 +362,8 @@ static int __init gfar_of_init(void)
 			}
 
 			gfar_data.phy_id = *id;
-			gfar_data.bus_id = res.start;
+			snprintf(gfar_data.bus_id, MII_BUS_ID_SIZE, "%x",
+					res.start);
 
 			of_node_put(phy);
 			of_node_put(mdio);
diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c
index 504b7ce2747d..3634b5fd7919 100644
--- a/drivers/net/au1000_eth.c
+++ b/drivers/net/au1000_eth.c
@@ -701,7 +701,7 @@ static struct net_device * au1000_probe(int port_num)
 	aup->mii_bus.write = mdiobus_write;
 	aup->mii_bus.reset = mdiobus_reset;
 	aup->mii_bus.name = "au1000_eth_mii";
-	aup->mii_bus.id = aup->mac_id;
+	snprintf(aup->mii_bus.id, MII_BUS_ID_SIZE, "%x", aup->mac_id);
 	aup->mii_bus.irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL);
 	for(i = 0; i < PHY_MAX_ADDR; ++i)
 		aup->mii_bus.irq[i] = PHY_POLL;
@@ -709,11 +709,11 @@ static struct net_device * au1000_probe(int port_num)
 	/* if known, set corresponding PHY IRQs */
 #if defined(AU1XXX_PHY_STATIC_CONFIG)
 # if defined(AU1XXX_PHY0_IRQ)
-	if (AU1XXX_PHY0_BUSID == aup->mii_bus.id)
+	if (AU1XXX_PHY0_BUSID == aup->mac_id)
 		aup->mii_bus.irq[AU1XXX_PHY0_ADDR] = AU1XXX_PHY0_IRQ;
 # endif
 # if defined(AU1XXX_PHY1_IRQ)
-	if (AU1XXX_PHY1_BUSID == aup->mii_bus.id)
+	if (AU1XXX_PHY1_BUSID == aup->mac_id)
 		aup->mii_bus.irq[AU1XXX_PHY1_ADDR] = AU1XXX_PHY1_IRQ;
 # endif
 #endif
diff --git a/drivers/net/bfin_mac.c b/drivers/net/bfin_mac.c
index 26b2dd5016cd..717dcc1aa1e9 100644
--- a/drivers/net/bfin_mac.c
+++ b/drivers/net/bfin_mac.c
@@ -969,7 +969,7 @@ static int __init bf537mac_probe(struct net_device *dev)
 	lp->mii_bus.write = mdiobus_write;
 	lp->mii_bus.reset = mdiobus_reset;
 	lp->mii_bus.name = "bfin_mac_mdio";
-	lp->mii_bus.id = 0;
+	snprintf(lp->mii_bus.id, MII_BUS_ID_SIZE, "0");
 	lp->mii_bus.irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL);
 	for (i = 0; i < PHY_MAX_ADDR; ++i)
 		lp->mii_bus.irq[i] = PHY_POLL;
diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index c85194f2cd2d..9da7ff437031 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -987,7 +987,7 @@ static int external_switch;
 static int __devinit cpmac_probe(struct platform_device *pdev)
 {
 	int rc, phy_id, i;
-	int mdio_bus_id = cpmac_mii.id;
+	char *mdio_bus_id = "0";
 	struct resource *mem;
 	struct cpmac_priv *priv;
 	struct net_device *dev;
@@ -1008,8 +1008,6 @@ static int __devinit cpmac_probe(struct platform_device *pdev)
 		if (external_switch || dumb_switch) {
 			struct fixed_phy_status status = {};
 
-			mdio_bus_id = 0;
-
 			/*
 			 * FIXME: this should be in the platform code!
 			 * Since there is not platform code at all (that is,
@@ -1143,6 +1141,7 @@ int __devinit cpmac_init(void)
 	}
 
 	cpmac_mii.phy_mask = ~(mask | 0x80000000);
+	snprintf(cpmac_mii.id, MII_BUS_ID_SIZE, "0");
 
 	res = mdiobus_register(&cpmac_mii);
 	if (res)
diff --git a/drivers/net/fec_mpc52xx.c b/drivers/net/fec_mpc52xx.c
index 58b71e60204e..43b5f30743c2 100644
--- a/drivers/net/fec_mpc52xx.c
+++ b/drivers/net/fec_mpc52xx.c
@@ -198,7 +198,7 @@ static int mpc52xx_fec_init_phy(struct net_device *dev)
 	struct phy_device *phydev;
 	char phy_id[BUS_ID_SIZE];
 
-	snprintf(phy_id, BUS_ID_SIZE, PHY_ID_FMT,
+	snprintf(phy_id, BUS_ID_SIZE, "%x:%02x",
 			(unsigned int)dev->base_addr, priv->phy_addr);
 
 	priv->link = PHY_DOWN;
diff --git a/drivers/net/fec_mpc52xx_phy.c b/drivers/net/fec_mpc52xx_phy.c
index 6a3ac4ea97e9..956836fc5ec0 100644
--- a/drivers/net/fec_mpc52xx_phy.c
+++ b/drivers/net/fec_mpc52xx_phy.c
@@ -124,7 +124,7 @@ static int mpc52xx_fec_mdio_probe(struct of_device *of, const struct of_device_i
 		goto out_free;
 	}
 
-	bus->id = res.start;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start);
 	bus->priv = priv;
 
 	bus->dev = dev;
diff --git a/drivers/net/fs_enet/fs_enet-main.c b/drivers/net/fs_enet/fs_enet-main.c
index 940e2041ba38..67b4b0728fce 100644
--- a/drivers/net/fs_enet/fs_enet-main.c
+++ b/drivers/net/fs_enet/fs_enet-main.c
@@ -1178,7 +1178,7 @@ static int __devinit find_phy(struct device_node *np,
 
 	data  = of_get_property(np, "fixed-link", NULL);
 	if (data) {
-		snprintf(fpi->bus_id, 16, PHY_ID_FMT, 0, *data);
+		snprintf(fpi->bus_id, 16, "%x:%02x", 0, *data);
 		return 0;
 	}
 
@@ -1202,7 +1202,7 @@ static int __devinit find_phy(struct device_node *np,
 	if (!data || len != 4)
 		goto out_put_mdio;
 
-	snprintf(fpi->bus_id, 16, PHY_ID_FMT, res.start, *data);
+	snprintf(fpi->bus_id, 16, "%x:%02x", res.start, *data);
 
 out_put_mdio:
 	of_node_put(mdionode);
diff --git a/drivers/net/fs_enet/mii-bitbang.c b/drivers/net/fs_enet/mii-bitbang.c
index b8e4a736a130..1620030cd33c 100644
--- a/drivers/net/fs_enet/mii-bitbang.c
+++ b/drivers/net/fs_enet/mii-bitbang.c
@@ -130,7 +130,7 @@ static int __devinit fs_mii_bitbang_init(struct mii_bus *bus,
 	 * we get is an int, and the odds of multiple bitbang mdio buses
 	 * is low enough that it's not worth going too crazy.
 	 */
-	bus->id = res.start;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start);
 
 	data = of_get_property(np, "fsl,mdio-pin", &len);
 	if (!data || len != 4)
@@ -307,7 +307,7 @@ static int __devinit fs_enet_mdio_probe(struct device *dev)
 		return -ENOMEM;
 
 	new_bus->name = "BB MII Bus",
-	new_bus->id = pdev->id;
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id);
 
 	new_bus->phy_mask = ~0x9;
 	pdata = (struct fs_mii_bb_platform_info *)pdev->dev.platform_data;
diff --git a/drivers/net/fs_enet/mii-fec.c b/drivers/net/fs_enet/mii-fec.c
index a89cf15090b8..ba75efc9f5b5 100644
--- a/drivers/net/fs_enet/mii-fec.c
+++ b/drivers/net/fs_enet/mii-fec.c
@@ -196,7 +196,7 @@ static int __devinit fs_enet_mdio_probe(struct of_device *ofdev,
 	if (ret)
 		return ret;
 
-	new_bus->id = res.start;
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", res.start);
 
 	fec->fecp = ioremap(res.start, res.end - res.start + 1);
 	if (!fec->fecp)
@@ -309,7 +309,7 @@ static int __devinit fs_enet_fec_mdio_probe(struct device *dev)
 	new_bus->read = &fs_enet_fec_mii_read,
 	new_bus->write = &fs_enet_fec_mii_write,
 	new_bus->reset = &fs_enet_fec_mii_reset,
-	new_bus->id = pdev->id;
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id);
 
 	pdata = (struct fs_mii_fec_platform_info *)pdev->dev.platform_data;
 
diff --git a/drivers/net/gianfar_mii.c b/drivers/net/gianfar_mii.c
index 24327629bf03..b8898927236a 100644
--- a/drivers/net/gianfar_mii.c
+++ b/drivers/net/gianfar_mii.c
@@ -173,7 +173,7 @@ int gfar_mdio_probe(struct device *dev)
 	new_bus->read = &gfar_mdio_read,
 	new_bus->write = &gfar_mdio_write,
 	new_bus->reset = &gfar_mdio_reset,
-	new_bus->id = pdev->id;
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id);
 
 	pdata = (struct gianfar_mdio_data *)pdev->dev.platform_data;
 
diff --git a/drivers/net/macb.c b/drivers/net/macb.c
index 489c7c3b90d9..d513bb8a4902 100644
--- a/drivers/net/macb.c
+++ b/drivers/net/macb.c
@@ -246,7 +246,7 @@ static int macb_mii_init(struct macb *bp)
 	bp->mii_bus.read = &macb_mdio_read;
 	bp->mii_bus.write = &macb_mdio_write;
 	bp->mii_bus.reset = &macb_mdio_reset;
-	bp->mii_bus.id = bp->pdev->id;
+	snprintf(bp->mii_bus.id, MII_BUS_ID_SIZE, "%x", bp->pdev->id);
 	bp->mii_bus.priv = bp;
 	bp->mii_bus.dev = &bp->dev->dev;
 	pdata = bp->pdev->dev.platform_data;
diff --git a/drivers/net/pasemi_mac.c b/drivers/net/pasemi_mac.c
index 2e39e0285d8f..bcd7f9814ed8 100644
--- a/drivers/net/pasemi_mac.c
+++ b/drivers/net/pasemi_mac.c
@@ -1012,7 +1012,7 @@ static int pasemi_mac_phy_init(struct net_device *dev)
 		goto err;
 
 	phy_id = *prop;
-	snprintf(mac->phy_id, BUS_ID_SIZE, PHY_ID_FMT, (int)r.start, phy_id);
+	snprintf(mac->phy_id, BUS_ID_SIZE, "%x:%02x", (int)r.start, phy_id);
 
 	of_node_put(phy_dn);
 
diff --git a/drivers/net/phy/fixed.c b/drivers/net/phy/fixed.c
index ca9b040f9ad9..4e07956a483b 100644
--- a/drivers/net/phy/fixed.c
+++ b/drivers/net/phy/fixed.c
@@ -213,7 +213,7 @@ static int __init fixed_mdio_bus_init(void)
 		goto err_pdev;
 	}
 
-	fmb->mii_bus.id = 0;
+	snprintf(fmb->mii_bus.id, MII_BUS_ID_SIZE, "0");
 	fmb->mii_bus.name = "Fixed MDIO Bus";
 	fmb->mii_bus.dev = &pdev->dev;
 	fmb->mii_bus.read = &fixed_mdio_read;
diff --git a/drivers/net/sb1250-mac.c b/drivers/net/sb1250-mac.c
index 7b53d658e337..888b7dec9866 100644
--- a/drivers/net/sb1250-mac.c
+++ b/drivers/net/sb1250-mac.c
@@ -2374,7 +2374,7 @@ static int sbmac_init(struct platform_device *pldev, long long base)
 	       dev->name, base, print_mac(mac, eaddr));
 
 	sc->mii_bus.name = sbmac_mdio_string;
-	sc->mii_bus.id = idx;
+	snprintf(sc->mii_bus.id, MII_BUS_ID_SIZE, "%x", idx);
 	sc->mii_bus.priv = sc;
 	sc->mii_bus.read = sbmac_mii_read;
 	sc->mii_bus.write = sbmac_mii_write;
diff --git a/drivers/net/ucc_geth.c b/drivers/net/ucc_geth.c
index 0ee4c168e4c0..29a4d650e8a8 100644
--- a/drivers/net/ucc_geth.c
+++ b/drivers/net/ucc_geth.c
@@ -3954,7 +3954,7 @@ static int ucc_geth_probe(struct of_device* ofdev, const struct of_device_id *ma
 		if (err)
 			return -1;
 
-		ug_info->mdio_bus = res.start;
+		snprintf(ug_info->mdio_bus, MII_BUS_ID_SIZE, "%x", res.start);
 	}
 
 	/* get the phy interface type, or default to MII */
diff --git a/drivers/net/ucc_geth.h b/drivers/net/ucc_geth.h
index 4fb95b3af948..9f8b7580a3a4 100644
--- a/drivers/net/ucc_geth.h
+++ b/drivers/net/ucc_geth.h
@@ -1156,7 +1156,7 @@ struct ucc_geth_info {
 	u16 pausePeriod;
 	u16 extensionField;
 	u8 phy_address;
-	u32 mdio_bus;
+	char mdio_bus[MII_BUS_ID_SIZE];
 	u8 weightfactor[NUM_TX_QUEUES];
 	u8 interruptcoalescingmaxvalue[NUM_RX_QUEUES];
 	u8 l2qt[UCC_GETH_VLAN_PRIORITY_MAX];
diff --git a/drivers/net/ucc_geth_mii.c b/drivers/net/ucc_geth_mii.c
index c69e654d539f..e4d3f330bac3 100644
--- a/drivers/net/ucc_geth_mii.c
+++ b/drivers/net/ucc_geth_mii.c
@@ -157,7 +157,7 @@ static int uec_mdio_probe(struct of_device *ofdev, const struct of_device_id *ma
 	if (err)
 		goto reg_map_fail;
 
-	new_bus->id = res.start;
+	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", res.start);
 
 	new_bus->irq = kmalloc(32 * sizeof(int), GFP_KERNEL);
 
diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h
index 1831b196c70a..2cad5c67397e 100644
--- a/include/linux/fsl_devices.h
+++ b/include/linux/fsl_devices.h
@@ -50,7 +50,7 @@ struct gianfar_platform_data {
 	u32	device_flags;
 	/* board specific information */
 	u32	board_flags;
-	u32	bus_id;
+	char	bus_id[MII_BUS_ID_SIZE];
 	u32	phy_id;
 	u8	mac_addr[6];
 	phy_interface_t interface;
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 5e43ae751412..6509f377bb10 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -63,8 +63,6 @@ typedef enum {
 	PHY_INTERFACE_MODE_RTBI
 } phy_interface_t;
 
-#define MII_BUS_MAX 4
-
 
 #define PHY_INIT_TIMEOUT	100000
 #define PHY_STATE_TIME		1
@@ -74,13 +72,19 @@ typedef enum {
 #define PHY_MAX_ADDR	32
 
 /* Used when trying to connect to a specific phy (mii bus id:phy device id) */
-#define PHY_ID_FMT "%x:%02x"
+#define PHY_ID_FMT "%s:%02x"
+
+/*
+ * Need to be a little smaller than phydev->dev.bus_id to leave room
+ * for the ":%02x"
+ */
+#define MII_BUS_ID_SIZE	(BUS_ID_SIZE - 3)
 
 /* The Bus class for PHYs.  Devices which provide access to
  * PHYs should register using this structure */
 struct mii_bus {
 	const char *name;
-	int id;
+	char id[MII_BUS_ID_SIZE];
 	void *priv;
 	int (*read)(struct mii_bus *bus, int phy_id, int regnum);
 	int (*write)(struct mii_bus *bus, int phy_id, int regnum, u16 val);
-- 
cgit v1.2.3


From c5e38a949bfa11d10f73927fbf4fe66b73bc3001 Mon Sep 17 00:00:00 2001
From: Andy Fleming <afleming@freescale.com>
Date: Wed, 9 Apr 2008 19:38:27 -0500
Subject: phy: Clean up header style

Multi-line comments weren't all CodingStyle compliant

Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/phy.h | 48 ++++++++++++++++++++++++++++++++----------------
 1 file changed, 32 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/phy.h b/include/linux/phy.h
index 6509f377bb10..2d838448415c 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -39,7 +39,8 @@
 				 SUPPORTED_1000baseT_Half | \
 				 SUPPORTED_1000baseT_Full)
 
-/* Set phydev->irq to PHY_POLL if interrupts are not supported,
+/*
+ * Set phydev->irq to PHY_POLL if interrupts are not supported,
  * or not desired for this PHY.  Set to PHY_IGNORE_INTERRUPT if
  * the attached driver handles the interrupt
  */
@@ -80,8 +81,10 @@ typedef enum {
  */
 #define MII_BUS_ID_SIZE	(BUS_ID_SIZE - 3)
 
-/* The Bus class for PHYs.  Devices which provide access to
- * PHYs should register using this structure */
+/*
+ * The Bus class for PHYs.  Devices which provide access to
+ * PHYs should register using this structure
+ */
 struct mii_bus {
 	const char *name;
 	char id[MII_BUS_ID_SIZE];
@@ -90,8 +93,10 @@ struct mii_bus {
 	int (*write)(struct mii_bus *bus, int phy_id, int regnum, u16 val);
 	int (*reset)(struct mii_bus *bus);
 
-	/* A lock to ensure that only one thing can read/write
-	 * the MDIO bus at a time */
+	/*
+	 * A lock to ensure that only one thing can read/write
+	 * the MDIO bus at a time
+	 */
 	struct mutex mdio_lock;
 
 	struct device *dev;
@@ -102,8 +107,10 @@ struct mii_bus {
 	/* Phy addresses to be ignored when probing */
 	u32 phy_mask;
 
-	/* Pointer to an array of interrupts, each PHY's
-	 * interrupt at the index matching its address */
+	/*
+	 * Pointer to an array of interrupts, each PHY's
+	 * interrupt at the index matching its address
+	 */
 	int *irq;
 };
 
@@ -255,7 +262,8 @@ struct phy_device {
 	/* Bus address of the PHY (0-32) */
 	int addr;
 
-	/* forced speed & duplex (no autoneg)
+	/*
+	 * forced speed & duplex (no autoneg)
 	 * partner speed & duplex & pause (autoneg)
 	 */
 	int speed;
@@ -278,8 +286,10 @@ struct phy_device {
 
 	int link_timeout;
 
-	/* Interrupt number for this PHY
-	 * -1 means no interrupt */
+	/*
+	 * Interrupt number for this PHY
+	 * -1 means no interrupt
+	 */
 	int irq;
 
 	/* private data pointer */
@@ -329,22 +339,28 @@ struct phy_driver {
 	u32 features;
 	u32 flags;
 
-	/* Called to initialize the PHY,
-	 * including after a reset */
+	/*
+	 * Called to initialize the PHY,
+	 * including after a reset
+	 */
 	int (*config_init)(struct phy_device *phydev);
 
-	/* Called during discovery.  Used to set
-	 * up device-specific structures, if any */
+	/*
+	 * Called during discovery.  Used to set
+	 * up device-specific structures, if any
+	 */
 	int (*probe)(struct phy_device *phydev);
 
 	/* PHY Power Management */
 	int (*suspend)(struct phy_device *phydev);
 	int (*resume)(struct phy_device *phydev);
 
-	/* Configures the advertisement and resets
+	/*
+	 * Configures the advertisement and resets
 	 * autonegotiation if phydev->autoneg is on,
 	 * forces the speed to the current settings in phydev
-	 * if phydev->autoneg is off */
+	 * if phydev->autoneg is off
+	 */
 	int (*config_aneg)(struct phy_device *phydev);
 
 	/* Determines the negotiated speed and duplex */
-- 
cgit v1.2.3


From b3d636b0d1b2eb870a55ae196b8f3838e1399554 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Wed, 16 Apr 2008 21:01:06 -0700
Subject: IB: Make struct ib_uobject.id a signed int

IDR IDs are signed, so struct ib_uobject.id should be signed.  This
avoids some sparse pointer signedness warnings.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 include/rdma/ib_verbs.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 701e7b40560a..40ff51244d19 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -730,7 +730,7 @@ struct ib_uobject {
 	struct ib_ucontext     *context;	/* associated user context */
 	void		       *object;		/* containing object */
 	struct list_head	list;		/* link to context's list */
-	u32			id;		/* index into kernel idr */
+	int			id;		/* index into kernel idr */
 	struct kref		ref;
 	struct rw_semaphore	mutex;		/* protects .live */
 	int			live;
-- 
cgit v1.2.3


From 37608eea86a358ac6a18df0af55d4f77d08a1f30 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Wed, 16 Apr 2008 21:01:08 -0700
Subject: mlx4_core: Fix confusion between mlx4_event and mlx4_dev_event enums

The struct mlx4_interface.event() method was supposed to get an enum
mlx4_dev_event, but the driver code was actually passing in the
hardware enum mlx4_event values.  Fix up the callers of
mlx4_dispatch_event() so that they pass in the right type of value,
and fix up the event method in mlx4_ib so that it can handle the enum
mlx4_dev_event values.

This eliminates the need for the subtype parameter to the event
method, so remove it.

This also fixes the sparse warning

    drivers/net/mlx4/intf.c:127:48: warning: mixing different enum types
    drivers/net/mlx4/intf.c:127:48:     int enum mlx4_event  versus
    drivers/net/mlx4/intf.c:127:48:     int enum mlx4_dev_event

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/main.c | 14 ++++++++------
 drivers/net/mlx4/catas.c          |  2 +-
 drivers/net/mlx4/eq.c             |  5 ++++-
 drivers/net/mlx4/intf.c           |  8 ++------
 drivers/net/mlx4/mlx4.h           |  4 ++--
 include/linux/mlx4/driver.h       |  3 +--
 6 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 96a39b5c9254..d5512011999c 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -675,18 +675,20 @@ static void mlx4_ib_remove(struct mlx4_dev *dev, void *ibdev_ptr)
 }
 
 static void mlx4_ib_event(struct mlx4_dev *dev, void *ibdev_ptr,
-			  enum mlx4_dev_event event, int subtype,
-			  int port)
+			  enum mlx4_dev_event event, int port)
 {
 	struct ib_event ibev;
 
 	switch (event) {
-	case MLX4_EVENT_TYPE_PORT_CHANGE:
-		ibev.event = subtype == MLX4_PORT_CHANGE_SUBTYPE_ACTIVE ?
-			IB_EVENT_PORT_ACTIVE : IB_EVENT_PORT_ERR;
+	case MLX4_DEV_EVENT_PORT_UP:
+		ibev.event = IB_EVENT_PORT_ACTIVE;
 		break;
 
-	case MLX4_EVENT_TYPE_LOCAL_CATAS_ERROR:
+	case MLX4_DEV_EVENT_PORT_DOWN:
+		ibev.event = IB_EVENT_PORT_ERR;
+		break;
+
+	case MLX4_DEV_EVENT_CATASTROPHIC_ERROR:
 		ibev.event = IB_EVENT_DEVICE_FATAL;
 		break;
 
diff --git a/drivers/net/mlx4/catas.c b/drivers/net/mlx4/catas.c
index 6b32ec94b3a8..aa9528779044 100644
--- a/drivers/net/mlx4/catas.c
+++ b/drivers/net/mlx4/catas.c
@@ -69,7 +69,7 @@ static void poll_catas(unsigned long dev_ptr)
 	if (readl(priv->catas_err.map)) {
 		dump_err_buf(dev);
 
-		mlx4_dispatch_event(dev, MLX4_EVENT_TYPE_LOCAL_CATAS_ERROR, 0, 0);
+		mlx4_dispatch_event(dev, MLX4_DEV_EVENT_CATASTROPHIC_ERROR, 0);
 
 		if (internal_err_reset) {
 			spin_lock(&catas_lock);
diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c
index 9c36c2034030..e141a1513f07 100644
--- a/drivers/net/mlx4/eq.c
+++ b/drivers/net/mlx4/eq.c
@@ -202,7 +202,10 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
 			break;
 
 		case MLX4_EVENT_TYPE_PORT_CHANGE:
-			mlx4_dispatch_event(dev, eqe->type, eqe->subtype,
+			mlx4_dispatch_event(dev,
+					    eqe->subtype == MLX4_PORT_CHANGE_SUBTYPE_ACTIVE ?
+					    MLX4_DEV_EVENT_PORT_UP :
+					    MLX4_DEV_EVENT_PORT_DOWN,
 					    be32_to_cpu(eqe->event.port_change.port) >> 28);
 			break;
 
diff --git a/drivers/net/mlx4/intf.c b/drivers/net/mlx4/intf.c
index be5d9e90ccf2..4a6c4d526f1b 100644
--- a/drivers/net/mlx4/intf.c
+++ b/drivers/net/mlx4/intf.c
@@ -30,8 +30,6 @@
  * SOFTWARE.
  */
 
-#include <linux/mlx4/driver.h>
-
 #include "mlx4.h"
 
 struct mlx4_device_context {
@@ -113,8 +111,7 @@ void mlx4_unregister_interface(struct mlx4_interface *intf)
 }
 EXPORT_SYMBOL_GPL(mlx4_unregister_interface);
 
-void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_event type,
-			 int subtype, int port)
+void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int port)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_device_context *dev_ctx;
@@ -124,8 +121,7 @@ void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_event type,
 
 	list_for_each_entry(dev_ctx, &priv->ctx_list, list)
 		if (dev_ctx->intf->event)
-			dev_ctx->intf->event(dev, dev_ctx->context, type,
-					     subtype, port);
+			dev_ctx->intf->event(dev, dev_ctx->context, type, port);
 
 	spin_unlock_irqrestore(&priv->ctx_lock, flags);
 }
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index 53a1cdddfc13..73336810e652 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -42,6 +42,7 @@
 #include <linux/timer.h>
 
 #include <linux/mlx4/device.h>
+#include <linux/mlx4/driver.h>
 #include <linux/mlx4/doorbell.h>
 
 #define DRV_NAME	"mlx4_core"
@@ -313,8 +314,7 @@ void mlx4_catas_cleanup(void);
 int mlx4_restart_one(struct pci_dev *pdev);
 int mlx4_register_device(struct mlx4_dev *dev);
 void mlx4_unregister_device(struct mlx4_dev *dev);
-void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_event type,
-			 int subtype, int port);
+void mlx4_dispatch_event(struct mlx4_dev *dev, enum mlx4_dev_event type, int port);
 
 struct mlx4_dev_cap;
 struct mlx4_init_hca_param;
diff --git a/include/linux/mlx4/driver.h b/include/linux/mlx4/driver.h
index 1b835ca49df1..53c5fdb6eac4 100644
--- a/include/linux/mlx4/driver.h
+++ b/include/linux/mlx4/driver.h
@@ -48,8 +48,7 @@ struct mlx4_interface {
 	void *			(*add)	 (struct mlx4_dev *dev);
 	void			(*remove)(struct mlx4_dev *dev, void *context);
 	void			(*event) (struct mlx4_dev *dev, void *context,
-					  enum mlx4_dev_event event, int subtype,
-					  int port);
+					  enum mlx4_dev_event event, int port);
 	struct list_head	list;
 };
 
-- 
cgit v1.2.3


From 8ff095ec4bce7be943beff3b330562e2f0e42167 Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@dev.mellanox.co.il>
Date: Wed, 16 Apr 2008 21:01:10 -0700
Subject: IB/mlx4: Add IPoIB checksum offload support

ConnectX devices support checksum generation and verification of TCP
and UDP packets for UD IPoIB messages.  This patch checks if the HCA
supports this and sets the IB_DEVICE_UD_IP_CSUM capability flag if it
does.  It implements support for handling the IB_SEND_IP_CSUM send
flag and setting the csum_ok field in receive work completions.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Ali Ayub <ali@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/cq.c   | 16 ++++++++++++++++
 drivers/infiniband/hw/mlx4/main.c |  2 ++
 drivers/infiniband/hw/mlx4/qp.c   |  3 +++
 drivers/net/mlx4/fw.c             |  4 ++++
 include/linux/mlx4/cq.h           | 14 ++++++++++++--
 include/linux/mlx4/qp.h           | 10 ++++++----
 6 files changed, 43 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 7360bbafbe84..d2e32b03e2f7 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -297,6 +297,20 @@ static void mlx4_ib_handle_error_cqe(struct mlx4_err_cqe *cqe,
 	wc->vendor_err = cqe->vendor_err_syndrome;
 }
 
+static int mlx4_ib_ipoib_csum_ok(__be32 status, __be16 checksum)
+{
+	return ((status & cpu_to_be32(MLX4_CQE_IPOIB_STATUS_IPV4	|
+				      MLX4_CQE_IPOIB_STATUS_IPV4F	|
+				      MLX4_CQE_IPOIB_STATUS_IPV4OPT	|
+				      MLX4_CQE_IPOIB_STATUS_IPV6	|
+				      MLX4_CQE_IPOIB_STATUS_IPOK)) ==
+		cpu_to_be32(MLX4_CQE_IPOIB_STATUS_IPV4	|
+			    MLX4_CQE_IPOIB_STATUS_IPOK))		&&
+		(status & cpu_to_be32(MLX4_CQE_IPOIB_STATUS_UDP	|
+				      MLX4_CQE_IPOIB_STATUS_TCP))	&&
+		checksum == cpu_to_be16(0xffff);
+}
+
 static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq,
 			    struct mlx4_ib_qp **cur_qp,
 			    struct ib_wc *wc)
@@ -434,6 +448,8 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq,
 		wc->dlid_path_bits = (g_mlpath_rqpn >> 24) & 0x7f;
 		wc->wc_flags	  |= g_mlpath_rqpn & 0x80000000 ? IB_WC_GRH : 0;
 		wc->pkey_index     = be32_to_cpu(cqe->immed_rss_invalid) & 0x7f;
+		wc->csum_ok	   = mlx4_ib_ipoib_csum_ok(cqe->ipoib_status,
+							   cqe->checksum);
 	}
 
 	return 0;
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index d5512011999c..6ea4746c2e9b 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -99,6 +99,8 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
 		props->device_cap_flags |= IB_DEVICE_AUTO_PATH_MIG;
 	if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_UD_AV_PORT)
 		props->device_cap_flags |= IB_DEVICE_UD_AV_PORT_ENFORCE;
+	if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_IPOIB_CSUM)
+		props->device_cap_flags |= IB_DEVICE_UD_IP_CSUM;
 
 	props->vendor_id	   = be32_to_cpup((__be32 *) (out_mad->data + 36)) &
 		0xffffff;
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index ac965ab28845..31b2b5b230bd 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1436,6 +1436,9 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			 cpu_to_be32(MLX4_WQE_CTRL_CQ_UPDATE) : 0) |
 			(wr->send_flags & IB_SEND_SOLICITED ?
 			 cpu_to_be32(MLX4_WQE_CTRL_SOLICITED) : 0) |
+			((wr->send_flags & IB_SEND_IP_CSUM) ?
+			 cpu_to_be32(MLX4_WQE_CTRL_IP_CSUM |
+				     MLX4_WQE_CTRL_TCP_UDP_CSUM) : 0) |
 			qp->sq_signal_bits;
 
 		if (wr->opcode == IB_WR_SEND_WITH_IMM ||
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 61dc4951d6b0..f494c3e8bce3 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -696,6 +696,10 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 	/* Check port for UD address vector: */
 	*(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1);
 
+	/* Enable IPoIB checksumming if we can: */
+	if (dev->caps.flags & MLX4_DEV_CAP_FLAG_IPOIB_CSUM)
+		*(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 3);
+
 	/* QPC/EEC/CQC/EQC/RDMARC attributes */
 
 	MLX4_PUT(inbox, param->qpc_base,      INIT_HCA_QPC_BASE_OFFSET);
diff --git a/include/linux/mlx4/cq.h b/include/linux/mlx4/cq.h
index 0181e0a57cbf..1243ebace561 100644
--- a/include/linux/mlx4/cq.h
+++ b/include/linux/mlx4/cq.h
@@ -45,11 +45,11 @@ struct mlx4_cqe {
 	u8			sl;
 	u8			reserved1;
 	__be16			rlid;
-	u32			reserved2;
+	__be32			ipoib_status;
 	__be32			byte_cnt;
 	__be16			wqe_index;
 	__be16			checksum;
-	u8			reserved3[3];
+	u8			reserved2[3];
 	u8			owner_sr_opcode;
 };
 
@@ -85,6 +85,16 @@ enum {
 	MLX4_CQE_SYNDROME_REMOTE_ABORTED_ERR		= 0x22,
 };
 
+enum {
+	MLX4_CQE_IPOIB_STATUS_IPV4			= 1 << 22,
+	MLX4_CQE_IPOIB_STATUS_IPV4F			= 1 << 23,
+	MLX4_CQE_IPOIB_STATUS_IPV6			= 1 << 24,
+	MLX4_CQE_IPOIB_STATUS_IPV4OPT			= 1 << 25,
+	MLX4_CQE_IPOIB_STATUS_TCP			= 1 << 26,
+	MLX4_CQE_IPOIB_STATUS_UDP			= 1 << 27,
+	MLX4_CQE_IPOIB_STATUS_IPOK			= 1 << 28,
+};
+
 static inline void mlx4_cq_arm(struct mlx4_cq *cq, u32 cmd,
 			       void __iomem *uar_page,
 			       spinlock_t *doorbell_lock)
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 09a2230923f2..31f9eb3ccbb3 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -158,10 +158,12 @@ struct mlx4_qp_context {
 #define MLX4_FW_VER_WQE_CTRL_NEC mlx4_fw_ver(2, 2, 232)
 
 enum {
-	MLX4_WQE_CTRL_NEC	= 1 << 29,
-	MLX4_WQE_CTRL_FENCE	= 1 << 6,
-	MLX4_WQE_CTRL_CQ_UPDATE	= 3 << 2,
-	MLX4_WQE_CTRL_SOLICITED	= 1 << 1,
+	MLX4_WQE_CTRL_NEC		= 1 << 29,
+	MLX4_WQE_CTRL_FENCE		= 1 << 6,
+	MLX4_WQE_CTRL_CQ_UPDATE		= 3 << 2,
+	MLX4_WQE_CTRL_SOLICITED		= 1 << 1,
+	MLX4_WQE_CTRL_IP_CSUM		= 1 << 4,
+	MLX4_WQE_CTRL_TCP_UDP_CSUM	= 1 << 5,
 };
 
 struct mlx4_wqe_ctrl_seg {
-- 
cgit v1.2.3


From b846f25aa2a353355aec5202fe4dbdc6674dfc64 Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@dev.mellanox.co.il>
Date: Wed, 16 Apr 2008 21:09:27 -0700
Subject: IB/core: Add creation flags to struct ib_qp_init_attr

Add a create_flags member to struct ib_qp_init_attr that will allow a
kernel verbs consumer to create a pass special flags when creating a QP.
Add a flag value for telling low-level drivers that a QP will be used
for IPoIB UD LSO.  The create_flags member will also be useful for XRC
and ehca low-latency QP support.

Since no create_flags handling is implemented yet, add code to all
low-level drivers to return -EINVAL if create_flags is non-zero.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/uverbs_cmd.c         | 1 +
 drivers/infiniband/hw/amso1100/c2_provider.c | 3 +++
 drivers/infiniband/hw/ehca/ehca_qp.c         | 3 +++
 drivers/infiniband/hw/ipath/ipath_qp.c       | 5 +++++
 drivers/infiniband/hw/mlx4/qp.c              | 3 +++
 drivers/infiniband/hw/mthca/mthca_provider.c | 3 +++
 drivers/infiniband/hw/nes/nes_verbs.c        | 3 +++
 include/rdma/ib_verbs.h                      | 5 +++++
 8 files changed, 26 insertions(+)

(limited to 'include')

diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 495c803fb11d..9e98cec6230f 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -1065,6 +1065,7 @@ ssize_t ib_uverbs_create_qp(struct ib_uverbs_file *file,
 	attr.srq           = srq;
 	attr.sq_sig_type   = cmd.sq_sig_all ? IB_SIGNAL_ALL_WR : IB_SIGNAL_REQ_WR;
 	attr.qp_type       = cmd.qp_type;
+	attr.create_flags  = 0;
 
 	attr.cap.max_send_wr     = cmd.max_send_wr;
 	attr.cap.max_recv_wr     = cmd.max_recv_wr;
diff --git a/drivers/infiniband/hw/amso1100/c2_provider.c b/drivers/infiniband/hw/amso1100/c2_provider.c
index 21fb46d39319..e10d27a6e145 100644
--- a/drivers/infiniband/hw/amso1100/c2_provider.c
+++ b/drivers/infiniband/hw/amso1100/c2_provider.c
@@ -245,6 +245,9 @@ static struct ib_qp *c2_create_qp(struct ib_pd *pd,
 
 	pr_debug("%s:%u\n", __func__, __LINE__);
 
+	if (init_attr->create_flags)
+		return ERR_PTR(-EINVAL);
+
 	switch (init_attr->qp_type) {
 	case IB_QPT_RC:
 		qp = kzalloc(sizeof(*qp), GFP_KERNEL);
diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c
index a9fd419855cb..3eb14a52cbf2 100644
--- a/drivers/infiniband/hw/ehca/ehca_qp.c
+++ b/drivers/infiniband/hw/ehca/ehca_qp.c
@@ -421,6 +421,9 @@ static struct ehca_qp *internal_create_qp(
 	u32 swqe_size = 0, rwqe_size = 0, ib_qp_num;
 	unsigned long flags;
 
+	if (init_attr->create_flags)
+		return ERR_PTR(-EINVAL);
+
 	memset(&parms, 0, sizeof(parms));
 	qp_type = init_attr->qp_type;
 
diff --git a/drivers/infiniband/hw/ipath/ipath_qp.c b/drivers/infiniband/hw/ipath/ipath_qp.c
index 6a4a5e3b78ba..812b42c500e1 100644
--- a/drivers/infiniband/hw/ipath/ipath_qp.c
+++ b/drivers/infiniband/hw/ipath/ipath_qp.c
@@ -747,6 +747,11 @@ struct ib_qp *ipath_create_qp(struct ib_pd *ibpd,
 	size_t sz;
 	struct ib_qp *ret;
 
+	if (init_attr->create_flags) {
+		ret = ERR_PTR(-EINVAL);
+		goto bail;
+	}
+
 	if (init_attr->cap.max_send_sge > ib_ipath_max_sges ||
 	    init_attr->cap.max_send_wr > ib_ipath_max_qp_wrs) {
 		ret = ERR_PTR(-EINVAL);
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 31b2b5b230bd..320c25fa74b1 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -673,6 +673,9 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 	struct mlx4_ib_qp *qp;
 	int err;
 
+	if (init_attr->create_flags)
+		return ERR_PTR(-EINVAL);
+
 	switch (init_attr->qp_type) {
 	case IB_QPT_RC:
 	case IB_QPT_UC:
diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c
index ee9bc1456226..81b257e18bb6 100644
--- a/drivers/infiniband/hw/mthca/mthca_provider.c
+++ b/drivers/infiniband/hw/mthca/mthca_provider.c
@@ -540,6 +540,9 @@ static struct ib_qp *mthca_create_qp(struct ib_pd *pd,
 	struct mthca_qp *qp;
 	int err;
 
+	if (init_attr->create_flags)
+		return ERR_PTR(-EINVAL);
+
 	switch (init_attr->qp_type) {
 	case IB_QPT_RC:
 	case IB_QPT_UC:
diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c
index 46b3b8ebcc14..7c27420c2240 100644
--- a/drivers/infiniband/hw/nes/nes_verbs.c
+++ b/drivers/infiniband/hw/nes/nes_verbs.c
@@ -1252,6 +1252,9 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd,
 	u8 rq_encoded_size;
 	/* int counter; */
 
+	if (init_attr->create_flags)
+		return ERR_PTR(-EINVAL);
+
 	atomic_inc(&qps_created);
 	switch (init_attr->qp_type) {
 		case IB_QPT_RC:
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 40ff51244d19..c3299be41c6f 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -495,6 +495,10 @@ enum ib_qp_type {
 	IB_QPT_RAW_ETY
 };
 
+enum ib_qp_create_flags {
+	IB_QP_CREATE_IPOIB_UD_LSO	= 1 << 0,
+};
+
 struct ib_qp_init_attr {
 	void                  (*event_handler)(struct ib_event *, void *);
 	void		       *qp_context;
@@ -504,6 +508,7 @@ struct ib_qp_init_attr {
 	struct ib_qp_cap	cap;
 	enum ib_sig_type	sq_sig_type;
 	enum ib_qp_type		qp_type;
+	enum ib_qp_create_flags	create_flags;
 	u8			port_num; /* special QP types only */
 };
 
-- 
cgit v1.2.3


From c93570f23a98c633570397aedc6d1808f5d5846a Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@dev.mellanox.co.il>
Date: Wed, 16 Apr 2008 21:09:27 -0700
Subject: IB/core: Add IPoIB UD LSO support

LSO (large send offload) allows the networking stack to pass SKBs with
data size larger than the MTU to the IPoIB driver and have the HCA HW
fragment the data to multiple MSS-sized packets.  Add a device
capability flag IB_DEVICE_UD_TSO for devices that can perform TCP
segmentation offload, a new send work request opcode IB_WR_LSO,
header, hlen and mss fields for the work request structure, and a new
IB_WC_LSO completion type.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 include/rdma/ib_verbs.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index c3299be41c6f..66928e9cab19 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -104,6 +104,7 @@ enum ib_device_cap_flags {
 	 * IPoIB driver may set NETIF_F_IP_CSUM for datagram mode.
 	 */
 	IB_DEVICE_UD_IP_CSUM		= (1<<18),
+	IB_DEVICE_UD_TSO		= (1<<19),
 };
 
 enum ib_atomic_cap {
@@ -411,6 +412,7 @@ enum ib_wc_opcode {
 	IB_WC_COMP_SWAP,
 	IB_WC_FETCH_ADD,
 	IB_WC_BIND_MW,
+	IB_WC_LSO,
 /*
  * Set value of IB_WC_RECV so consumers can test if a completion is a
  * receive by testing (opcode & IB_WC_RECV).
@@ -622,7 +624,8 @@ enum ib_wr_opcode {
 	IB_WR_SEND_WITH_IMM,
 	IB_WR_RDMA_READ,
 	IB_WR_ATOMIC_CMP_AND_SWP,
-	IB_WR_ATOMIC_FETCH_AND_ADD
+	IB_WR_ATOMIC_FETCH_AND_ADD,
+	IB_WR_LSO
 };
 
 enum ib_send_flags {
@@ -660,6 +663,9 @@ struct ib_send_wr {
 		} atomic;
 		struct {
 			struct ib_ah *ah;
+			void   *header;
+			int     hlen;
+			int     mss;
 			u32	remote_qpn;
 			u32	remote_qkey;
 			u16	pkey_index; /* valid for GSI only */
-- 
cgit v1.2.3


From b832be1e4007f4a54954ec68bd865ff05d6babca Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@dev.mellanox.co.il>
Date: Wed, 16 Apr 2008 21:09:27 -0700
Subject: IB/mlx4: Add IPoIB LSO support

Add TSO support to the mlx4_ib driver.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/cq.c      |  3 ++
 drivers/infiniband/hw/mlx4/main.c    |  2 +
 drivers/infiniband/hw/mlx4/mlx4_ib.h |  5 +++
 drivers/infiniband/hw/mlx4/qp.c      | 72 +++++++++++++++++++++++++++++++-----
 drivers/net/mlx4/fw.c                |  9 +++++
 drivers/net/mlx4/fw.h                |  1 +
 drivers/net/mlx4/main.c              |  1 +
 include/linux/mlx4/device.h          |  1 +
 include/linux/mlx4/qp.h              |  5 +++
 9 files changed, 90 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index d2e32b03e2f7..7d70af7952b0 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -420,6 +420,9 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq,
 		case MLX4_OPCODE_BIND_MW:
 			wc->opcode    = IB_WC_BIND_MW;
 			break;
+		case MLX4_OPCODE_LSO:
+			wc->opcode    = IB_WC_LSO;
+			break;
 		}
 	} else {
 		wc->byte_len = be32_to_cpu(cqe->byte_cnt);
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 6ea4746c2e9b..e9330a0d6c03 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -101,6 +101,8 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
 		props->device_cap_flags |= IB_DEVICE_UD_AV_PORT_ENFORCE;
 	if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_IPOIB_CSUM)
 		props->device_cap_flags |= IB_DEVICE_UD_IP_CSUM;
+	if (dev->dev->caps.max_gso_sz)
+		props->device_cap_flags |= IB_DEVICE_UD_TSO;
 
 	props->vendor_id	   = be32_to_cpup((__be32 *) (out_mad->data + 36)) &
 		0xffffff;
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index 3726e451a327..3f8bd0a37b96 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -110,6 +110,10 @@ struct mlx4_ib_wq {
 	unsigned		tail;
 };
 
+enum mlx4_ib_qp_flags {
+	MLX4_IB_QP_LSO		= 1 << 0
+};
+
 struct mlx4_ib_qp {
 	struct ib_qp		ibqp;
 	struct mlx4_qp		mqp;
@@ -129,6 +133,7 @@ struct mlx4_ib_qp {
 	struct mlx4_mtt		mtt;
 	int			buf_size;
 	struct mutex		mutex;
+	u32			flags;
 	u8			port;
 	u8			alt_port;
 	u8			atomic_rd_en;
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 320c25fa74b1..2ba243084089 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -71,6 +71,7 @@ enum {
 
 static const __be32 mlx4_ib_opcode[] = {
 	[IB_WR_SEND]			= __constant_cpu_to_be32(MLX4_OPCODE_SEND),
+	[IB_WR_LSO]			= __constant_cpu_to_be32(MLX4_OPCODE_LSO),
 	[IB_WR_SEND_WITH_IMM]		= __constant_cpu_to_be32(MLX4_OPCODE_SEND_IMM),
 	[IB_WR_RDMA_WRITE]		= __constant_cpu_to_be32(MLX4_OPCODE_RDMA_WRITE),
 	[IB_WR_RDMA_WRITE_WITH_IMM]	= __constant_cpu_to_be32(MLX4_OPCODE_RDMA_WRITE_IMM),
@@ -242,7 +243,7 @@ static void mlx4_ib_qp_event(struct mlx4_qp *qp, enum mlx4_event type)
 	}
 }
 
-static int send_wqe_overhead(enum ib_qp_type type)
+static int send_wqe_overhead(enum ib_qp_type type, u32 flags)
 {
 	/*
 	 * UD WQEs must have a datagram segment.
@@ -253,7 +254,8 @@ static int send_wqe_overhead(enum ib_qp_type type)
 	switch (type) {
 	case IB_QPT_UD:
 		return sizeof (struct mlx4_wqe_ctrl_seg) +
-			sizeof (struct mlx4_wqe_datagram_seg);
+			sizeof (struct mlx4_wqe_datagram_seg) +
+			((flags & MLX4_IB_QP_LSO) ? 64 : 0);
 	case IB_QPT_UC:
 		return sizeof (struct mlx4_wqe_ctrl_seg) +
 			sizeof (struct mlx4_wqe_raddr_seg);
@@ -315,7 +317,7 @@ static int set_kernel_sq_size(struct mlx4_ib_dev *dev, struct ib_qp_cap *cap,
 	/* Sanity check SQ size before proceeding */
 	if (cap->max_send_wr	 > dev->dev->caps.max_wqes  ||
 	    cap->max_send_sge	 > dev->dev->caps.max_sq_sg ||
-	    cap->max_inline_data + send_wqe_overhead(type) +
+	    cap->max_inline_data + send_wqe_overhead(type, qp->flags) +
 	    sizeof (struct mlx4_wqe_inline_seg) > dev->dev->caps.max_sq_desc_sz)
 		return -EINVAL;
 
@@ -329,7 +331,7 @@ static int set_kernel_sq_size(struct mlx4_ib_dev *dev, struct ib_qp_cap *cap,
 
 	s = max(cap->max_send_sge * sizeof (struct mlx4_wqe_data_seg),
 		cap->max_inline_data + sizeof (struct mlx4_wqe_inline_seg)) +
-		send_wqe_overhead(type);
+		send_wqe_overhead(type, qp->flags);
 
 	/*
 	 * Hermon supports shrinking WQEs, such that a single work
@@ -394,7 +396,8 @@ static int set_kernel_sq_size(struct mlx4_ib_dev *dev, struct ib_qp_cap *cap,
 	}
 
 	qp->sq.max_gs = ((qp->sq_max_wqes_per_wr << qp->sq.wqe_shift) -
-			 send_wqe_overhead(type)) / sizeof (struct mlx4_wqe_data_seg);
+			 send_wqe_overhead(type, qp->flags)) /
+		sizeof (struct mlx4_wqe_data_seg);
 
 	qp->buf_size = (qp->rq.wqe_cnt << qp->rq.wqe_shift) +
 		(qp->sq.wqe_cnt << qp->sq.wqe_shift);
@@ -503,6 +506,9 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 	} else {
 		qp->sq_no_prefetch = 0;
 
+		if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO)
+			qp->flags |= MLX4_IB_QP_LSO;
+
 		err = set_kernel_sq_size(dev, &init_attr->cap, init_attr->qp_type, qp);
 		if (err)
 			goto err;
@@ -673,7 +679,11 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 	struct mlx4_ib_qp *qp;
 	int err;
 
-	if (init_attr->create_flags)
+	/* We only support LSO, and only for kernel UD QPs. */
+	if (init_attr->create_flags & ~IB_QP_CREATE_IPOIB_UD_LSO)
+		return ERR_PTR(-EINVAL);
+	if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO &&
+	    (pd->uobject || init_attr->qp_type != IB_QPT_UD))
 		return ERR_PTR(-EINVAL);
 
 	switch (init_attr->qp_type) {
@@ -879,10 +889,15 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp,
 		}
 	}
 
-	if (ibqp->qp_type == IB_QPT_GSI || ibqp->qp_type == IB_QPT_SMI ||
-	    ibqp->qp_type == IB_QPT_UD)
+	if (ibqp->qp_type == IB_QPT_GSI || ibqp->qp_type == IB_QPT_SMI)
 		context->mtu_msgmax = (IB_MTU_4096 << 5) | 11;
-	else if (attr_mask & IB_QP_PATH_MTU) {
+	else if (ibqp->qp_type == IB_QPT_UD) {
+		if (qp->flags & MLX4_IB_QP_LSO)
+			context->mtu_msgmax = (IB_MTU_4096 << 5) |
+					      ilog2(dev->dev->caps.max_gso_sz);
+		else
+			context->mtu_msgmax = (IB_MTU_4096 << 5) | 11;
+	} else if (attr_mask & IB_QP_PATH_MTU) {
 		if (attr->path_mtu < IB_MTU_256 || attr->path_mtu > IB_MTU_4096) {
 			printk(KERN_ERR "path MTU (%u) is invalid\n",
 			       attr->path_mtu);
@@ -1399,6 +1414,34 @@ static void __set_data_seg(struct mlx4_wqe_data_seg *dseg, struct ib_sge *sg)
 	dseg->addr       = cpu_to_be64(sg->addr);
 }
 
+static int build_lso_seg(struct mlx4_lso_seg *wqe, struct ib_send_wr *wr,
+			 struct mlx4_ib_qp *qp, unsigned *lso_seg_len)
+{
+	unsigned halign = ALIGN(sizeof *wqe + wr->wr.ud.hlen, 16);
+
+	/*
+	 * This is a temporary limitation and will be removed in
+	 * a forthcoming FW release:
+	 */
+	if (unlikely(halign > 64))
+		return -EINVAL;
+
+	if (unlikely(!(qp->flags & MLX4_IB_QP_LSO) &&
+		     wr->num_sge > qp->sq.max_gs - (halign >> 4)))
+		return -EINVAL;
+
+	memcpy(wqe->header, wr->wr.ud.header, wr->wr.ud.hlen);
+
+	/* make sure LSO header is written before overwriting stamping */
+	wmb();
+
+	wqe->mss_hdr_size = cpu_to_be32((wr->wr.ud.mss - wr->wr.ud.hlen) << 16 |
+					wr->wr.ud.hlen);
+
+	*lso_seg_len = halign;
+	return 0;
+}
+
 int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 		      struct ib_send_wr **bad_wr)
 {
@@ -1412,6 +1455,7 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 	unsigned ind;
 	int uninitialized_var(stamp);
 	int uninitialized_var(size);
+	unsigned seglen;
 	int i;
 
 	spin_lock_irqsave(&qp->sq.lock, flags);
@@ -1490,6 +1534,16 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			set_datagram_seg(wqe, wr);
 			wqe  += sizeof (struct mlx4_wqe_datagram_seg);
 			size += sizeof (struct mlx4_wqe_datagram_seg) / 16;
+
+			if (wr->opcode == IB_WR_LSO) {
+				err = build_lso_seg(wqe, wr, qp, &seglen);
+				if (unlikely(err)) {
+					*bad_wr = wr;
+					goto out;
+				}
+				wqe  += seglen;
+				size += seglen / 16;
+			}
 			break;
 
 		case IB_QPT_SMI:
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index f494c3e8bce3..d82f2751d2c7 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -133,6 +133,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 #define QUERY_DEV_CAP_MAX_AV_OFFSET		0x27
 #define QUERY_DEV_CAP_MAX_REQ_QP_OFFSET		0x29
 #define QUERY_DEV_CAP_MAX_RES_QP_OFFSET		0x2b
+#define QUERY_DEV_CAP_MAX_GSO_OFFSET		0x2d
 #define QUERY_DEV_CAP_MAX_RDMA_OFFSET		0x2f
 #define QUERY_DEV_CAP_RSZ_SRQ_OFFSET		0x33
 #define QUERY_DEV_CAP_ACK_DELAY_OFFSET		0x35
@@ -215,6 +216,13 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev_cap->max_requester_per_qp = 1 << (field & 0x3f);
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_MAX_RES_QP_OFFSET);
 	dev_cap->max_responder_per_qp = 1 << (field & 0x3f);
+	MLX4_GET(field, outbox, QUERY_DEV_CAP_MAX_GSO_OFFSET);
+	field &= 0x1f;
+	if (!field)
+		dev_cap->max_gso_sz = 0;
+	else
+		dev_cap->max_gso_sz = 1 << field;
+
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_MAX_RDMA_OFFSET);
 	dev_cap->max_rdma_global = 1 << (field & 0x3f);
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_ACK_DELAY_OFFSET);
@@ -377,6 +385,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		 dev_cap->max_sq_desc_sz, dev_cap->max_sq_sg);
 	mlx4_dbg(dev, "Max RQ desc size: %d, max RQ S/G: %d\n",
 		 dev_cap->max_rq_desc_sz, dev_cap->max_rq_sg);
+	mlx4_dbg(dev, "Max GSO size: %d\n", dev_cap->max_gso_sz);
 
 	dump_dev_cap_flags(dev, dev_cap->flags);
 
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index e16dec890413..306cb9b0242d 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -96,6 +96,7 @@ struct mlx4_dev_cap {
 	u8  bmme_flags;
 	u32 reserved_lkey;
 	u64 max_icm_sz;
+	int max_gso_sz;
 };
 
 struct mlx4_adapter {
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 08bfc130a33e..7cfbe75114d1 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -159,6 +159,7 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.page_size_cap	     = ~(u32) (dev_cap->min_page_sz - 1);
 	dev->caps.flags		     = dev_cap->flags;
 	dev->caps.stat_rate_support  = dev_cap->stat_rate_support;
+	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 
 	return 0;
 }
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 6cdf813cd478..ff7df1a2222f 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -186,6 +186,7 @@ struct mlx4_caps {
 	u32			flags;
 	u16			stat_rate_support;
 	u8			port_width_cap[MLX4_MAX_PORTS + 1];
+	int			max_gso_sz;
 };
 
 struct mlx4_buf_list {
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 31f9eb3ccbb3..a5e43febee4f 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -219,6 +219,11 @@ struct mlx4_wqe_datagram_seg {
 	__be32			reservd[2];
 };
 
+struct mlx4_lso_seg {
+	__be32			mss_hdr_size;
+	__be32			header[0];
+};
+
 struct mlx4_wqe_bind_seg {
 	__be32			flags1;
 	__be32			flags2;
-- 
cgit v1.2.3


From 0f39cf3d54e67a705773fd0ec56ca3dcd3e9272f Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Wed, 16 Apr 2008 21:09:32 -0700
Subject: IB/core: Add support for "send with invalidate" work requests

Add a new IB_WR_SEND_WITH_INV send opcode that can be used to mark a
"send with invalidate" work request as defined in the iWARP verbs and
the InfiniBand base memory management extensions.  Also put "imm_data"
and a new "invalidate_rkey" member in a new "ex" union in struct
ib_send_wr. The invalidate_rkey member can be used to pass in an
R_Key/STag to be invalidated.  Add this new union to struct
ib_uverbs_send_wr.  Add code to copy the invalidate_rkey field in
ib_uverbs_post_send().

Fix up low-level drivers to deal with the change to struct ib_send_wr,
and just remove the imm_data initialization from net/sunrpc/xprtrdma/,
since that code never does any send with immediate operations.

Also, move the existing IB_DEVICE_SEND_W_INV flag to a new bit, since
the iWARP drivers currently in the tree set the bit.  The amso1100
driver at least will silently fail to honor the IB_SEND_INVALIDATE bit
if passed in as part of userspace send requests (since it does not
implement kernel bypass work request queueing).  Remove the flag from
all existing drivers that set it until we know which ones are OK.

The values chosen for the new flag is not consecutive to avoid clashing
with flags defined in the XRC patches, which are not merged yet but
which are already in use and are likely to be merged soon.

This resurrects a patch sent long ago by Mikkel Hagen <mhagen@iol.unh.edu>.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/uverbs_cmd.c        | 13 +++++++++++--
 drivers/infiniband/hw/amso1100/c2_rnic.c    |  2 +-
 drivers/infiniband/hw/cxgb3/iwch_provider.c |  3 +--
 drivers/infiniband/hw/cxgb3/iwch_qp.c       |  4 ++--
 drivers/infiniband/hw/ehca/ehca_reqs.c      |  2 +-
 drivers/infiniband/hw/ipath/ipath_rc.c      |  8 ++++----
 drivers/infiniband/hw/ipath/ipath_ruc.c     |  4 ++--
 drivers/infiniband/hw/ipath/ipath_uc.c      |  8 ++++----
 drivers/infiniband/hw/ipath/ipath_ud.c      |  4 ++--
 drivers/infiniband/hw/mlx4/qp.c             |  4 ++--
 drivers/infiniband/hw/mthca/mthca_qp.c      |  6 +++---
 drivers/infiniband/hw/nes/nes_hw.c          |  2 +-
 include/rdma/ib_user_verbs.h                |  5 ++++-
 include/rdma/ib_verbs.h                     | 11 ++++++++---
 net/sunrpc/xprtrdma/verbs.c                 |  1 -
 15 files changed, 46 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 9e98cec6230f..2c3bff5fe867 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -1463,7 +1463,6 @@ ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file,
 		next->num_sge    = user_wr->num_sge;
 		next->opcode     = user_wr->opcode;
 		next->send_flags = user_wr->send_flags;
-		next->imm_data   = (__be32 __force) user_wr->imm_data;
 
 		if (is_ud) {
 			next->wr.ud.ah = idr_read_ah(user_wr->wr.ud.ah,
@@ -1476,14 +1475,24 @@ ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file,
 			next->wr.ud.remote_qkey = user_wr->wr.ud.remote_qkey;
 		} else {
 			switch (next->opcode) {
-			case IB_WR_RDMA_WRITE:
 			case IB_WR_RDMA_WRITE_WITH_IMM:
+				next->ex.imm_data =
+					(__be32 __force) user_wr->ex.imm_data;
+			case IB_WR_RDMA_WRITE:
 			case IB_WR_RDMA_READ:
 				next->wr.rdma.remote_addr =
 					user_wr->wr.rdma.remote_addr;
 				next->wr.rdma.rkey        =
 					user_wr->wr.rdma.rkey;
 				break;
+			case IB_WR_SEND_WITH_IMM:
+				next->ex.imm_data =
+					(__be32 __force) user_wr->ex.imm_data;
+				break;
+			case IB_WR_SEND_WITH_INV:
+				next->ex.invalidate_rkey =
+					user_wr->ex.invalidate_rkey;
+				break;
 			case IB_WR_ATOMIC_CMP_AND_SWP:
 			case IB_WR_ATOMIC_FETCH_AND_ADD:
 				next->wr.atomic.remote_addr =
diff --git a/drivers/infiniband/hw/amso1100/c2_rnic.c b/drivers/infiniband/hw/amso1100/c2_rnic.c
index 7a625524e0c5..b1441aeb60c2 100644
--- a/drivers/infiniband/hw/amso1100/c2_rnic.c
+++ b/drivers/infiniband/hw/amso1100/c2_rnic.c
@@ -455,7 +455,7 @@ int __devinit c2_rnic_init(struct c2_dev *c2dev)
 	     IB_DEVICE_CURR_QP_STATE_MOD |
 	     IB_DEVICE_SYS_IMAGE_GUID |
 	     IB_DEVICE_ZERO_STAG |
-	     IB_DEVICE_SEND_W_INV | IB_DEVICE_MEM_WINDOW);
+	     IB_DEVICE_MEM_WINDOW);
 
 	/* Allocate the qptr_array */
 	c2dev->qptr_array = vmalloc(C2_MAX_CQS * sizeof(void *));
diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c
index 50e1f2a16e0c..ca7265443c05 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_provider.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c
@@ -1109,8 +1109,7 @@ int iwch_register_device(struct iwch_dev *dev)
 	memcpy(&dev->ibdev.node_guid, dev->rdev.t3cdev_p->lldev->dev_addr, 6);
 	dev->ibdev.owner = THIS_MODULE;
 	dev->device_cap_flags =
-	    (IB_DEVICE_ZERO_STAG |
-	     IB_DEVICE_SEND_W_INV | IB_DEVICE_MEM_WINDOW);
+	    (IB_DEVICE_ZERO_STAG | IB_DEVICE_MEM_WINDOW);
 
 	dev->ibdev.uverbs_cmd_mask =
 	    (1ull << IB_USER_VERBS_CMD_GET_CONTEXT) |
diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c
index bc5d9b0813e5..8891c3b0a3d5 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_qp.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c
@@ -72,7 +72,7 @@ static int iwch_build_rdma_send(union t3_wr *wqe, struct ib_send_wr *wr,
 	wqe->send.reserved[2] = 0;
 	if (wr->opcode == IB_WR_SEND_WITH_IMM) {
 		plen = 4;
-		wqe->send.sgl[0].stag = wr->imm_data;
+		wqe->send.sgl[0].stag = wr->ex.imm_data;
 		wqe->send.sgl[0].len = __constant_cpu_to_be32(0);
 		wqe->send.num_sgle = __constant_cpu_to_be32(0);
 		*flit_cnt = 5;
@@ -112,7 +112,7 @@ static int iwch_build_rdma_write(union t3_wr *wqe, struct ib_send_wr *wr,
 
 	if (wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM) {
 		plen = 4;
-		wqe->write.sgl[0].stag = wr->imm_data;
+		wqe->write.sgl[0].stag = wr->ex.imm_data;
 		wqe->write.sgl[0].len = __constant_cpu_to_be32(0);
 		wqe->write.num_sgle = __constant_cpu_to_be32(0);
 		*flit_cnt = 6;
diff --git a/drivers/infiniband/hw/ehca/ehca_reqs.c b/drivers/infiniband/hw/ehca/ehca_reqs.c
index 2ce8cffb8664..a20bbf466188 100644
--- a/drivers/infiniband/hw/ehca/ehca_reqs.c
+++ b/drivers/infiniband/hw/ehca/ehca_reqs.c
@@ -188,7 +188,7 @@ static inline int ehca_write_swqe(struct ehca_qp *qp,
 	if (send_wr->opcode == IB_WR_SEND_WITH_IMM ||
 	    send_wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM) {
 		/* this might not work as long as HW does not support it */
-		wqe_p->immediate_data = be32_to_cpu(send_wr->imm_data);
+		wqe_p->immediate_data = be32_to_cpu(send_wr->ex.imm_data);
 		wqe_p->wr_flag |= WQE_WRFLAG_IMM_DATA_PRESENT;
 	}
 
diff --git a/drivers/infiniband/hw/ipath/ipath_rc.c b/drivers/infiniband/hw/ipath/ipath_rc.c
index 467981905bbe..c405dfba5531 100644
--- a/drivers/infiniband/hw/ipath/ipath_rc.c
+++ b/drivers/infiniband/hw/ipath/ipath_rc.c
@@ -308,7 +308,7 @@ int ipath_make_rc_req(struct ipath_qp *qp)
 			else {
 				qp->s_state = OP(SEND_ONLY_WITH_IMMEDIATE);
 				/* Immediate data comes after the BTH */
-				ohdr->u.imm_data = wqe->wr.imm_data;
+				ohdr->u.imm_data = wqe->wr.ex.imm_data;
 				hwords += 1;
 			}
 			if (wqe->wr.send_flags & IB_SEND_SOLICITED)
@@ -346,7 +346,7 @@ int ipath_make_rc_req(struct ipath_qp *qp)
 				qp->s_state =
 					OP(RDMA_WRITE_ONLY_WITH_IMMEDIATE);
 				/* Immediate data comes after RETH */
-				ohdr->u.rc.imm_data = wqe->wr.imm_data;
+				ohdr->u.rc.imm_data = wqe->wr.ex.imm_data;
 				hwords += 1;
 				if (wqe->wr.send_flags & IB_SEND_SOLICITED)
 					bth0 |= 1 << 23;
@@ -490,7 +490,7 @@ int ipath_make_rc_req(struct ipath_qp *qp)
 		else {
 			qp->s_state = OP(SEND_LAST_WITH_IMMEDIATE);
 			/* Immediate data comes after the BTH */
-			ohdr->u.imm_data = wqe->wr.imm_data;
+			ohdr->u.imm_data = wqe->wr.ex.imm_data;
 			hwords += 1;
 		}
 		if (wqe->wr.send_flags & IB_SEND_SOLICITED)
@@ -526,7 +526,7 @@ int ipath_make_rc_req(struct ipath_qp *qp)
 		else {
 			qp->s_state = OP(RDMA_WRITE_LAST_WITH_IMMEDIATE);
 			/* Immediate data comes after the BTH */
-			ohdr->u.imm_data = wqe->wr.imm_data;
+			ohdr->u.imm_data = wqe->wr.ex.imm_data;
 			hwords += 1;
 			if (wqe->wr.send_flags & IB_SEND_SOLICITED)
 				bth0 |= 1 << 23;
diff --git a/drivers/infiniband/hw/ipath/ipath_ruc.c b/drivers/infiniband/hw/ipath/ipath_ruc.c
index bcaa2914e341..8ac5c1d82ccd 100644
--- a/drivers/infiniband/hw/ipath/ipath_ruc.c
+++ b/drivers/infiniband/hw/ipath/ipath_ruc.c
@@ -310,7 +310,7 @@ again:
 	switch (wqe->wr.opcode) {
 	case IB_WR_SEND_WITH_IMM:
 		wc.wc_flags = IB_WC_WITH_IMM;
-		wc.imm_data = wqe->wr.imm_data;
+		wc.imm_data = wqe->wr.ex.imm_data;
 		/* FALLTHROUGH */
 	case IB_WR_SEND:
 		if (!ipath_get_rwqe(qp, 0)) {
@@ -339,7 +339,7 @@ again:
 			goto err;
 		}
 		wc.wc_flags = IB_WC_WITH_IMM;
-		wc.imm_data = wqe->wr.imm_data;
+		wc.imm_data = wqe->wr.ex.imm_data;
 		if (!ipath_get_rwqe(qp, 1))
 			goto rnr_nak;
 		/* FALLTHROUGH */
diff --git a/drivers/infiniband/hw/ipath/ipath_uc.c b/drivers/infiniband/hw/ipath/ipath_uc.c
index 2dd8de20d221..bfe8926b5514 100644
--- a/drivers/infiniband/hw/ipath/ipath_uc.c
+++ b/drivers/infiniband/hw/ipath/ipath_uc.c
@@ -94,7 +94,7 @@ int ipath_make_uc_req(struct ipath_qp *qp)
 				qp->s_state =
 					OP(SEND_ONLY_WITH_IMMEDIATE);
 				/* Immediate data comes after the BTH */
-				ohdr->u.imm_data = wqe->wr.imm_data;
+				ohdr->u.imm_data = wqe->wr.ex.imm_data;
 				hwords += 1;
 			}
 			if (wqe->wr.send_flags & IB_SEND_SOLICITED)
@@ -123,7 +123,7 @@ int ipath_make_uc_req(struct ipath_qp *qp)
 				qp->s_state =
 					OP(RDMA_WRITE_ONLY_WITH_IMMEDIATE);
 				/* Immediate data comes after the RETH */
-				ohdr->u.rc.imm_data = wqe->wr.imm_data;
+				ohdr->u.rc.imm_data = wqe->wr.ex.imm_data;
 				hwords += 1;
 				if (wqe->wr.send_flags & IB_SEND_SOLICITED)
 					bth0 |= 1 << 23;
@@ -152,7 +152,7 @@ int ipath_make_uc_req(struct ipath_qp *qp)
 		else {
 			qp->s_state = OP(SEND_LAST_WITH_IMMEDIATE);
 			/* Immediate data comes after the BTH */
-			ohdr->u.imm_data = wqe->wr.imm_data;
+			ohdr->u.imm_data = wqe->wr.ex.imm_data;
 			hwords += 1;
 		}
 		if (wqe->wr.send_flags & IB_SEND_SOLICITED)
@@ -177,7 +177,7 @@ int ipath_make_uc_req(struct ipath_qp *qp)
 			qp->s_state =
 				OP(RDMA_WRITE_LAST_WITH_IMMEDIATE);
 			/* Immediate data comes after the BTH */
-			ohdr->u.imm_data = wqe->wr.imm_data;
+			ohdr->u.imm_data = wqe->wr.ex.imm_data;
 			hwords += 1;
 			if (wqe->wr.send_flags & IB_SEND_SOLICITED)
 				bth0 |= 1 << 23;
diff --git a/drivers/infiniband/hw/ipath/ipath_ud.c b/drivers/infiniband/hw/ipath/ipath_ud.c
index 918f52070653..8b6a261c89e3 100644
--- a/drivers/infiniband/hw/ipath/ipath_ud.c
+++ b/drivers/infiniband/hw/ipath/ipath_ud.c
@@ -95,7 +95,7 @@ static void ipath_ud_loopback(struct ipath_qp *sqp, struct ipath_swqe *swqe)
 
 	if (swqe->wr.opcode == IB_WR_SEND_WITH_IMM) {
 		wc.wc_flags = IB_WC_WITH_IMM;
-		wc.imm_data = swqe->wr.imm_data;
+		wc.imm_data = swqe->wr.ex.imm_data;
 	} else {
 		wc.wc_flags = 0;
 		wc.imm_data = 0;
@@ -327,7 +327,7 @@ int ipath_make_ud_req(struct ipath_qp *qp)
 	}
 	if (wqe->wr.opcode == IB_WR_SEND_WITH_IMM) {
 		qp->s_hdrwords++;
-		ohdr->u.ud.imm_data = wqe->wr.imm_data;
+		ohdr->u.ud.imm_data = wqe->wr.ex.imm_data;
 		bth0 = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE << 24;
 	} else
 		bth0 = IB_OPCODE_UD_SEND_ONLY << 24;
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index f5210c17e312..38e651a67589 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1249,7 +1249,7 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_send_wr *wr,
 	case IB_WR_SEND_WITH_IMM:
 		sqp->ud_header.bth.opcode	 = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE;
 		sqp->ud_header.immediate_present = 1;
-		sqp->ud_header.immediate_data    = wr->imm_data;
+		sqp->ud_header.immediate_data    = wr->ex.imm_data;
 		break;
 	default:
 		return -EINVAL;
@@ -1492,7 +1492,7 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 
 		if (wr->opcode == IB_WR_SEND_WITH_IMM ||
 		    wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM)
-			ctrl->imm = wr->imm_data;
+			ctrl->imm = wr->ex.imm_data;
 		else
 			ctrl->imm = 0;
 
diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c
index 8433897624bc..b3fd6b05d79d 100644
--- a/drivers/infiniband/hw/mthca/mthca_qp.c
+++ b/drivers/infiniband/hw/mthca/mthca_qp.c
@@ -1532,7 +1532,7 @@ static int build_mlx_header(struct mthca_dev *dev, struct mthca_sqp *sqp,
 	case IB_WR_SEND_WITH_IMM:
 		sqp->ud_header.bth.opcode = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE;
 		sqp->ud_header.immediate_present = 1;
-		sqp->ud_header.immediate_data = wr->imm_data;
+		sqp->ud_header.immediate_data = wr->ex.imm_data;
 		break;
 	default:
 		return -EINVAL;
@@ -1679,7 +1679,7 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			cpu_to_be32(1);
 		if (wr->opcode == IB_WR_SEND_WITH_IMM ||
 		    wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM)
-			((struct mthca_next_seg *) wqe)->imm = wr->imm_data;
+			((struct mthca_next_seg *) wqe)->imm = wr->ex.imm_data;
 
 		wqe += sizeof (struct mthca_next_seg);
 		size = sizeof (struct mthca_next_seg) / 16;
@@ -2020,7 +2020,7 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			cpu_to_be32(1);
 		if (wr->opcode == IB_WR_SEND_WITH_IMM ||
 		    wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM)
-			((struct mthca_next_seg *) wqe)->imm = wr->imm_data;
+			((struct mthca_next_seg *) wqe)->imm = wr->ex.imm_data;
 
 		wqe += sizeof (struct mthca_next_seg);
 		size = sizeof (struct mthca_next_seg) / 16;
diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c
index 134189d77ed3..aa53aab91bf8 100644
--- a/drivers/infiniband/hw/nes/nes_hw.c
+++ b/drivers/infiniband/hw/nes/nes_hw.c
@@ -393,7 +393,7 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) {
 	nesadapter->base_pd = 1;
 
 	nesadapter->device_cap_flags =
-			IB_DEVICE_ZERO_STAG | IB_DEVICE_SEND_W_INV | IB_DEVICE_MEM_WINDOW;
+		IB_DEVICE_ZERO_STAG | IB_DEVICE_MEM_WINDOW;
 
 	nesadapter->allocated_qps = (unsigned long *)&(((unsigned char *)nesadapter)
 			[(sizeof(struct nes_adapter)+(sizeof(unsigned long)-1))&(~(sizeof(unsigned long)-1))]);
diff --git a/include/rdma/ib_user_verbs.h b/include/rdma/ib_user_verbs.h
index 64a721fcbc1c..8d65bf0a625b 100644
--- a/include/rdma/ib_user_verbs.h
+++ b/include/rdma/ib_user_verbs.h
@@ -533,7 +533,10 @@ struct ib_uverbs_send_wr {
 	__u32 num_sge;
 	__u32 opcode;
 	__u32 send_flags;
-	__u32 imm_data;
+	union {
+		__u32 imm_data;
+		__u32 invalidate_rkey;
+	} ex;
 	union {
 		struct {
 			__u64 remote_addr;
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 66928e9cab19..c48f6af5ef9a 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -94,7 +94,7 @@ enum ib_device_cap_flags {
 	IB_DEVICE_SRQ_RESIZE		= (1<<13),
 	IB_DEVICE_N_NOTIFY_CQ		= (1<<14),
 	IB_DEVICE_ZERO_STAG		= (1<<15),
-	IB_DEVICE_SEND_W_INV		= (1<<16),
+	IB_DEVICE_RESERVED		= (1<<16), /* old SEND_W_INV */
 	IB_DEVICE_MEM_WINDOW		= (1<<17),
 	/*
 	 * Devices should set IB_DEVICE_UD_IP_SUM if they support
@@ -105,6 +105,7 @@ enum ib_device_cap_flags {
 	 */
 	IB_DEVICE_UD_IP_CSUM		= (1<<18),
 	IB_DEVICE_UD_TSO		= (1<<19),
+	IB_DEVICE_SEND_W_INV		= (1<<21),
 };
 
 enum ib_atomic_cap {
@@ -625,7 +626,8 @@ enum ib_wr_opcode {
 	IB_WR_RDMA_READ,
 	IB_WR_ATOMIC_CMP_AND_SWP,
 	IB_WR_ATOMIC_FETCH_AND_ADD,
-	IB_WR_LSO
+	IB_WR_LSO,
+	IB_WR_SEND_WITH_INV,
 };
 
 enum ib_send_flags {
@@ -649,7 +651,10 @@ struct ib_send_wr {
 	int			num_sge;
 	enum ib_wr_opcode	opcode;
 	int			send_flags;
-	__be32			imm_data;
+	union {
+		__be32		imm_data;
+		u32		invalidate_rkey;
+	} ex;
 	union {
 		struct {
 			u64	remote_addr;
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index ffbf22a1d2ca..8ea283ecc522 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1573,7 +1573,6 @@ rpcrdma_ep_post(struct rpcrdma_ia *ia,
 	send_wr.sg_list = req->rl_send_iov;
 	send_wr.num_sge = req->rl_niovs;
 	send_wr.opcode = IB_WR_SEND;
-	send_wr.imm_data = 0;
 	if (send_wr.num_sge == 4)	/* no need to sync any pad (constant) */
 		ib_dma_sync_single_for_device(ia->ri_id->device,
 			req->rl_send_iov[3].addr, req->rl_send_iov[3].length,
-- 
cgit v1.2.3


From 2dd5716227878d5950988514a2cbabf72f7fc888 Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@dev.mellanox.co.il>
Date: Wed, 16 Apr 2008 21:09:33 -0700
Subject: IB/core: Add support for modify CQ

Add support for modifying CQ parameters for controlling event
generation moderation.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/verbs.c |  7 +++++++
 include/rdma/ib_verbs.h         | 11 +++++++++++
 2 files changed, 18 insertions(+)

(limited to 'include')

diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 8ffb5f2c21f7..05042089de6e 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -630,6 +630,13 @@ struct ib_cq *ib_create_cq(struct ib_device *device,
 }
 EXPORT_SYMBOL(ib_create_cq);
 
+int ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period)
+{
+	return cq->device->modify_cq ?
+		cq->device->modify_cq(cq, cq_count, cq_period) : -ENOSYS;
+}
+EXPORT_SYMBOL(ib_modify_cq);
+
 int ib_destroy_cq(struct ib_cq *cq)
 {
 	if (atomic_read(&cq->usecnt))
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index c48f6af5ef9a..95bf4bac44cb 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -987,6 +987,8 @@ struct ib_device {
 						int comp_vector,
 						struct ib_ucontext *context,
 						struct ib_udata *udata);
+	int                        (*modify_cq)(struct ib_cq *cq, u16 cq_count,
+						u16 cq_period);
 	int                        (*destroy_cq)(struct ib_cq *cq);
 	int                        (*resize_cq)(struct ib_cq *cq, int cqe,
 						struct ib_udata *udata);
@@ -1391,6 +1393,15 @@ struct ib_cq *ib_create_cq(struct ib_device *device,
  */
 int ib_resize_cq(struct ib_cq *cq, int cqe);
 
+/**
+ * ib_modify_cq - Modifies moderation params of the CQ
+ * @cq: The CQ to modify.
+ * @cq_count: number of CQEs that will trigger an event
+ * @cq_period: max period of time in usec before triggering an event
+ *
+ */
+int ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period);
+
 /**
  * ib_destroy_cq - Destroys the specified CQ.
  * @cq: The CQ to destroy.
-- 
cgit v1.2.3


From 3fdcb97f0b8d8a29117dc36acd0b15965d2a2160 Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@dev.mellanox.co.il>
Date: Wed, 16 Apr 2008 21:09:33 -0700
Subject: IB/mlx4: Add support for modifying CQ moderation parameters

Signed-off-by: Eli Cohen <eli@mellnaox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/cq.c      |  8 +++++++
 drivers/infiniband/hw/mlx4/main.c    |  1 +
 drivers/infiniband/hw/mlx4/mlx4_ib.h |  1 +
 drivers/net/mlx4/cq.c                | 44 ++++++++++++++++++++++++++++++------
 include/linux/mlx4/cmd.h             |  2 +-
 include/linux/mlx4/cq.h              |  3 +++
 6 files changed, 51 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 7d70af7952b0..e4fb64b118e3 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -85,6 +85,14 @@ static struct mlx4_cqe *next_cqe_sw(struct mlx4_ib_cq *cq)
 	return get_sw_cqe(cq, cq->mcq.cons_index);
 }
 
+int mlx4_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period)
+{
+	struct mlx4_ib_cq *mcq = to_mcq(cq);
+	struct mlx4_ib_dev *dev = to_mdev(cq->device);
+
+	return mlx4_cq_modify(dev->dev, &mcq->mcq, cq_count, cq_period);
+}
+
 struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector,
 				struct ib_ucontext *context,
 				struct ib_udata *udata)
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index e9330a0d6c03..76dd45c764b4 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -609,6 +609,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	ibdev->ib_dev.post_send		= mlx4_ib_post_send;
 	ibdev->ib_dev.post_recv		= mlx4_ib_post_recv;
 	ibdev->ib_dev.create_cq		= mlx4_ib_create_cq;
+	ibdev->ib_dev.modify_cq		= mlx4_ib_modify_cq;
 	ibdev->ib_dev.destroy_cq	= mlx4_ib_destroy_cq;
 	ibdev->ib_dev.poll_cq		= mlx4_ib_poll_cq;
 	ibdev->ib_dev.req_notify_cq	= mlx4_ib_arm_cq;
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index 3f8bd0a37b96..ef8ad96e1432 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -254,6 +254,7 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 				  struct ib_udata *udata);
 int mlx4_ib_dereg_mr(struct ib_mr *mr);
 
+int mlx4_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period);
 struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector,
 				struct ib_ucontext *context,
 				struct ib_udata *udata);
diff --git a/drivers/net/mlx4/cq.c b/drivers/net/mlx4/cq.c
index d4441fee3d80..8c314341434f 100644
--- a/drivers/net/mlx4/cq.c
+++ b/drivers/net/mlx4/cq.c
@@ -38,6 +38,7 @@
 #include <linux/hardirq.h>
 
 #include <linux/mlx4/cmd.h>
+#include <linux/mlx4/cq.h>
 
 #include "mlx4.h"
 #include "icm.h"
@@ -47,21 +48,19 @@ struct mlx4_cq_context {
 	u16			reserved1[3];
 	__be16			page_offset;
 	__be32			logsize_usrpage;
-	u8			reserved2;
-	u8			cq_period;
-	u8			reserved3;
-	u8			cq_max_count;
-	u8			reserved4[3];
+	__be16			cq_period;
+	__be16			cq_max_count;
+	u8			reserved2[3];
 	u8			comp_eqn;
 	u8			log_page_size;
-	u8			reserved5[2];
+	u8			reserved3[2];
 	u8			mtt_base_addr_h;
 	__be32			mtt_base_addr_l;
 	__be32			last_notified_index;
 	__be32			solicit_producer_index;
 	__be32			consumer_index;
 	__be32			producer_index;
-	u32			reserved6[2];
+	u32			reserved4[2];
 	__be64			db_rec_addr;
 };
 
@@ -121,6 +120,13 @@ static int mlx4_SW2HW_CQ(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
 			MLX4_CMD_TIME_CLASS_A);
 }
 
+static int mlx4_MODIFY_CQ(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
+			 int cq_num, u32 opmod)
+{
+	return mlx4_cmd(dev, mailbox->dma, cq_num, opmod, MLX4_CMD_MODIFY_CQ,
+			MLX4_CMD_TIME_CLASS_A);
+}
+
 static int mlx4_HW2SW_CQ(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
 			 int cq_num)
 {
@@ -129,6 +135,30 @@ static int mlx4_HW2SW_CQ(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
 			    MLX4_CMD_TIME_CLASS_A);
 }
 
+int mlx4_cq_modify(struct mlx4_dev *dev, struct mlx4_cq *cq,
+		   u16 count, u16 period)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_cq_context *cq_context;
+	int err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	cq_context = mailbox->buf;
+	memset(cq_context, 0, sizeof *cq_context);
+
+	cq_context->cq_max_count = cpu_to_be16(count);
+	cq_context->cq_period    = cpu_to_be16(period);
+
+	err = mlx4_MODIFY_CQ(dev, mailbox, cq->cqn, 1);
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_cq_modify);
+
 int mlx4_cq_alloc(struct mlx4_dev *dev, int nent, struct mlx4_mtt *mtt,
 		  struct mlx4_uar *uar, u64 db_rec, struct mlx4_cq *cq)
 {
diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h
index 7d1eaa97de13..77323a72dd3c 100644
--- a/include/linux/mlx4/cmd.h
+++ b/include/linux/mlx4/cmd.h
@@ -81,7 +81,7 @@ enum {
 	MLX4_CMD_SW2HW_CQ	 = 0x16,
 	MLX4_CMD_HW2SW_CQ	 = 0x17,
 	MLX4_CMD_QUERY_CQ	 = 0x18,
-	MLX4_CMD_RESIZE_CQ	 = 0x2c,
+	MLX4_CMD_MODIFY_CQ	 = 0x2c,
 
 	/* SRQ commands */
 	MLX4_CMD_SW2HW_SRQ	 = 0x35,
diff --git a/include/linux/mlx4/cq.h b/include/linux/mlx4/cq.h
index 1243ebace561..f7c3511c594b 100644
--- a/include/linux/mlx4/cq.h
+++ b/include/linux/mlx4/cq.h
@@ -130,4 +130,7 @@ enum {
 	MLX4_CQ_DB_REQ_NOT		= 2 << 24
 };
 
+int mlx4_cq_modify(struct mlx4_dev *dev, struct mlx4_cq *cq,
+		   u16 count, u16 period);
+
 #endif /* MLX4_CQ_H */
-- 
cgit v1.2.3


From bbf8eed1a0f8949f7385146624f736f829992a70 Mon Sep 17 00:00:00 2001
From: Vladimir Sokolovsky <vlad@dev.mellanox.co.il>
Date: Wed, 16 Apr 2008 21:09:33 -0700
Subject: IB/mlx4: Add support for resizing CQs

Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/cq.c      | 292 +++++++++++++++++++++++++++++++----
 drivers/infiniband/hw/mlx4/main.c    |   2 +
 drivers/infiniband/hw/mlx4/mlx4_ib.h |   9 ++
 drivers/net/mlx4/cq.c                |  28 ++++
 include/linux/mlx4/cq.h              |   2 +
 5 files changed, 300 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index e4fb64b118e3..3557e7edc9b6 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -93,6 +93,74 @@ int mlx4_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period)
 	return mlx4_cq_modify(dev->dev, &mcq->mcq, cq_count, cq_period);
 }
 
+static int mlx4_ib_alloc_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf *buf, int nent)
+{
+	int err;
+
+	err = mlx4_buf_alloc(dev->dev, nent * sizeof(struct mlx4_cqe),
+			     PAGE_SIZE * 2, &buf->buf);
+
+	if (err)
+		goto out;
+
+	err = mlx4_mtt_init(dev->dev, buf->buf.npages, buf->buf.page_shift,
+				    &buf->mtt);
+	if (err)
+		goto err_buf;
+
+	err = mlx4_buf_write_mtt(dev->dev, &buf->mtt, &buf->buf);
+	if (err)
+		goto err_mtt;
+
+	return 0;
+
+err_mtt:
+	mlx4_mtt_cleanup(dev->dev, &buf->mtt);
+
+err_buf:
+	mlx4_buf_free(dev->dev, nent * sizeof(struct mlx4_cqe),
+			      &buf->buf);
+
+out:
+	return err;
+}
+
+static void mlx4_ib_free_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf *buf, int cqe)
+{
+	mlx4_buf_free(dev->dev, (cqe + 1) * sizeof(struct mlx4_cqe), &buf->buf);
+}
+
+static int mlx4_ib_get_cq_umem(struct mlx4_ib_dev *dev, struct ib_ucontext *context,
+			       struct mlx4_ib_cq_buf *buf, struct ib_umem **umem,
+			       u64 buf_addr, int cqe)
+{
+	int err;
+
+	*umem = ib_umem_get(context, buf_addr, cqe * sizeof (struct mlx4_cqe),
+			    IB_ACCESS_LOCAL_WRITE);
+	if (IS_ERR(*umem))
+		return PTR_ERR(*umem);
+
+	err = mlx4_mtt_init(dev->dev, ib_umem_page_count(*umem),
+			    ilog2((*umem)->page_size), &buf->mtt);
+	if (err)
+		goto err_buf;
+
+	err = mlx4_ib_umem_write_mtt(dev, &buf->mtt, *umem);
+	if (err)
+		goto err_mtt;
+
+	return 0;
+
+err_mtt:
+	mlx4_mtt_cleanup(dev->dev, &buf->mtt);
+
+err_buf:
+	ib_umem_release(*umem);
+
+	return err;
+}
+
 struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector,
 				struct ib_ucontext *context,
 				struct ib_udata *udata)
@@ -100,7 +168,6 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector
 	struct mlx4_ib_dev *dev = to_mdev(ibdev);
 	struct mlx4_ib_cq *cq;
 	struct mlx4_uar *uar;
-	int buf_size;
 	int err;
 
 	if (entries < 1 || entries > dev->dev->caps.max_cqes)
@@ -112,8 +179,10 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector
 
 	entries      = roundup_pow_of_two(entries + 1);
 	cq->ibcq.cqe = entries - 1;
-	buf_size     = entries * sizeof (struct mlx4_cqe);
+	mutex_init(&cq->resize_mutex);
 	spin_lock_init(&cq->lock);
+	cq->resize_buf = NULL;
+	cq->resize_umem = NULL;
 
 	if (context) {
 		struct mlx4_ib_create_cq ucmd;
@@ -123,21 +192,10 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector
 			goto err_cq;
 		}
 
-		cq->umem = ib_umem_get(context, ucmd.buf_addr, buf_size,
-				       IB_ACCESS_LOCAL_WRITE);
-		if (IS_ERR(cq->umem)) {
-			err = PTR_ERR(cq->umem);
-			goto err_cq;
-		}
-
-		err = mlx4_mtt_init(dev->dev, ib_umem_page_count(cq->umem),
-				    ilog2(cq->umem->page_size), &cq->buf.mtt);
+		err = mlx4_ib_get_cq_umem(dev, context, &cq->buf, &cq->umem,
+					  ucmd.buf_addr, entries);
 		if (err)
-			goto err_buf;
-
-		err = mlx4_ib_umem_write_mtt(dev, &cq->buf.mtt, cq->umem);
-		if (err)
-			goto err_mtt;
+			goto err_cq;
 
 		err = mlx4_ib_db_map_user(to_mucontext(context), ucmd.db_addr,
 					  &cq->db);
@@ -155,19 +213,9 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector
 		*cq->mcq.set_ci_db = 0;
 		*cq->mcq.arm_db    = 0;
 
-		if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2, &cq->buf.buf)) {
-			err = -ENOMEM;
-			goto err_db;
-		}
-
-		err = mlx4_mtt_init(dev->dev, cq->buf.buf.npages, cq->buf.buf.page_shift,
-				    &cq->buf.mtt);
+		err = mlx4_ib_alloc_cq_buf(dev, &cq->buf, entries);
 		if (err)
-			goto err_buf;
-
-		err = mlx4_buf_write_mtt(dev->dev, &cq->buf.mtt, &cq->buf.buf);
-		if (err)
-			goto err_mtt;
+			goto err_db;
 
 		uar = &dev->priv_uar;
 	}
@@ -195,12 +243,10 @@ err_dbmap:
 err_mtt:
 	mlx4_mtt_cleanup(dev->dev, &cq->buf.mtt);
 
-err_buf:
 	if (context)
 		ib_umem_release(cq->umem);
 	else
-		mlx4_buf_free(dev->dev, entries * sizeof (struct mlx4_cqe),
-			      &cq->buf.buf);
+		mlx4_ib_free_cq_buf(dev, &cq->buf, entries);
 
 err_db:
 	if (!context)
@@ -212,6 +258,170 @@ err_cq:
 	return ERR_PTR(err);
 }
 
+static int mlx4_alloc_resize_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq *cq,
+				  int entries)
+{
+	int err;
+
+	if (cq->resize_buf)
+		return -EBUSY;
+
+	cq->resize_buf = kmalloc(sizeof *cq->resize_buf, GFP_ATOMIC);
+	if (!cq->resize_buf)
+		return -ENOMEM;
+
+	err = mlx4_ib_alloc_cq_buf(dev, &cq->resize_buf->buf, entries);
+	if (err) {
+		kfree(cq->resize_buf);
+		cq->resize_buf = NULL;
+		return err;
+	}
+
+	cq->resize_buf->cqe = entries - 1;
+
+	return 0;
+}
+
+static int mlx4_alloc_resize_umem(struct mlx4_ib_dev *dev, struct mlx4_ib_cq *cq,
+				   int entries, struct ib_udata *udata)
+{
+	struct mlx4_ib_resize_cq ucmd;
+	int err;
+
+	if (cq->resize_umem)
+		return -EBUSY;
+
+	if (ib_copy_from_udata(&ucmd, udata, sizeof ucmd))
+		return -EFAULT;
+
+	cq->resize_buf = kmalloc(sizeof *cq->resize_buf, GFP_ATOMIC);
+	if (!cq->resize_buf)
+		return -ENOMEM;
+
+	err = mlx4_ib_get_cq_umem(dev, cq->umem->context, &cq->resize_buf->buf,
+				  &cq->resize_umem, ucmd.buf_addr, entries);
+	if (err) {
+		kfree(cq->resize_buf);
+		cq->resize_buf = NULL;
+		return err;
+	}
+
+	cq->resize_buf->cqe = entries - 1;
+
+	return 0;
+}
+
+static int mlx4_ib_get_outstanding_cqes(struct mlx4_ib_cq *cq)
+{
+	u32 i;
+
+	i = cq->mcq.cons_index;
+	while (get_sw_cqe(cq, i & cq->ibcq.cqe))
+		++i;
+
+	return i - cq->mcq.cons_index;
+}
+
+static void mlx4_ib_cq_resize_copy_cqes(struct mlx4_ib_cq *cq)
+{
+	struct mlx4_cqe *cqe;
+	int i;
+
+	i = cq->mcq.cons_index;
+	cqe = get_cqe(cq, i & cq->ibcq.cqe);
+	while ((cqe->owner_sr_opcode & MLX4_CQE_OPCODE_MASK) != MLX4_CQE_OPCODE_RESIZE) {
+		memcpy(get_cqe_from_buf(&cq->resize_buf->buf,
+					(i + 1) & cq->resize_buf->cqe),
+			get_cqe(cq, i & cq->ibcq.cqe), sizeof(struct mlx4_cqe));
+		cqe = get_cqe(cq, ++i & cq->ibcq.cqe);
+	}
+	++cq->mcq.cons_index;
+}
+
+int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
+{
+	struct mlx4_ib_dev *dev = to_mdev(ibcq->device);
+	struct mlx4_ib_cq *cq = to_mcq(ibcq);
+	int outst_cqe;
+	int err;
+
+	mutex_lock(&cq->resize_mutex);
+
+	if (entries < 1 || entries > dev->dev->caps.max_cqes) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	entries = roundup_pow_of_two(entries + 1);
+	if (entries == ibcq->cqe + 1) {
+		err = 0;
+		goto out;
+	}
+
+	if (ibcq->uobject) {
+		err = mlx4_alloc_resize_umem(dev, cq, entries, udata);
+		if (err)
+			goto out;
+	} else {
+		/* Can't be smaller then the number of outstanding CQEs */
+		outst_cqe = mlx4_ib_get_outstanding_cqes(cq);
+		if (entries < outst_cqe + 1) {
+			err = 0;
+			goto out;
+		}
+
+		err = mlx4_alloc_resize_buf(dev, cq, entries);
+		if (err)
+			goto out;
+	}
+
+	err = mlx4_cq_resize(dev->dev, &cq->mcq, entries, &cq->resize_buf->buf.mtt);
+	if (err)
+		goto err_buf;
+
+	if (ibcq->uobject) {
+		cq->buf      = cq->resize_buf->buf;
+		cq->ibcq.cqe = cq->resize_buf->cqe;
+		ib_umem_release(cq->umem);
+		cq->umem     = cq->resize_umem;
+
+		kfree(cq->resize_buf);
+		cq->resize_buf = NULL;
+		cq->resize_umem = NULL;
+	} else {
+		spin_lock_irq(&cq->lock);
+		if (cq->resize_buf) {
+			mlx4_ib_cq_resize_copy_cqes(cq);
+			mlx4_ib_free_cq_buf(dev, &cq->buf, cq->ibcq.cqe);
+			cq->buf      = cq->resize_buf->buf;
+			cq->ibcq.cqe = cq->resize_buf->cqe;
+
+			kfree(cq->resize_buf);
+			cq->resize_buf = NULL;
+		}
+		spin_unlock_irq(&cq->lock);
+	}
+
+	goto out;
+
+err_buf:
+	if (!ibcq->uobject)
+		mlx4_ib_free_cq_buf(dev, &cq->resize_buf->buf,
+				    cq->resize_buf->cqe);
+
+	kfree(cq->resize_buf);
+	cq->resize_buf = NULL;
+
+	if (cq->resize_umem) {
+		ib_umem_release(cq->resize_umem);
+		cq->resize_umem = NULL;
+	}
+
+out:
+	mutex_unlock(&cq->resize_mutex);
+	return err;
+}
+
 int mlx4_ib_destroy_cq(struct ib_cq *cq)
 {
 	struct mlx4_ib_dev *dev = to_mdev(cq->device);
@@ -224,8 +434,7 @@ int mlx4_ib_destroy_cq(struct ib_cq *cq)
 		mlx4_ib_db_unmap_user(to_mucontext(cq->uobject->context), &mcq->db);
 		ib_umem_release(mcq->umem);
 	} else {
-		mlx4_buf_free(dev->dev, (cq->cqe + 1) * sizeof (struct mlx4_cqe),
-			      &mcq->buf.buf);
+		mlx4_ib_free_cq_buf(dev, &mcq->buf, cq->cqe + 1);
 		mlx4_ib_db_free(dev, &mcq->db);
 	}
 
@@ -332,6 +541,7 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq,
 	u32 g_mlpath_rqpn;
 	u16 wqe_ctr;
 
+repoll:
 	cqe = next_cqe_sw(cq);
 	if (!cqe)
 		return -EAGAIN;
@@ -354,6 +564,22 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq,
 		return -EINVAL;
 	}
 
+	/* Resize CQ in progress */
+	if (unlikely((cqe->owner_sr_opcode & MLX4_CQE_OPCODE_MASK) == MLX4_CQE_OPCODE_RESIZE)) {
+		if (cq->resize_buf) {
+			struct mlx4_ib_dev *dev = to_mdev(cq->ibcq.device);
+
+			mlx4_ib_free_cq_buf(dev, &cq->buf, cq->ibcq.cqe);
+			cq->buf      = cq->resize_buf->buf;
+			cq->ibcq.cqe = cq->resize_buf->cqe;
+
+			kfree(cq->resize_buf);
+			cq->resize_buf = NULL;
+		}
+
+		goto repoll;
+	}
+
 	if (!*cur_qp ||
 	    (be32_to_cpu(cqe->my_qpn) & 0xffffff) != (*cur_qp)->mqp.qpn) {
 		/*
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 76dd45c764b4..57885cd61937 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -571,6 +571,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 		(1ull << IB_USER_VERBS_CMD_DEREG_MR)		|
 		(1ull << IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL)	|
 		(1ull << IB_USER_VERBS_CMD_CREATE_CQ)		|
+		(1ull << IB_USER_VERBS_CMD_RESIZE_CQ)		|
 		(1ull << IB_USER_VERBS_CMD_DESTROY_CQ)		|
 		(1ull << IB_USER_VERBS_CMD_CREATE_QP)		|
 		(1ull << IB_USER_VERBS_CMD_MODIFY_QP)		|
@@ -610,6 +611,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	ibdev->ib_dev.post_recv		= mlx4_ib_post_recv;
 	ibdev->ib_dev.create_cq		= mlx4_ib_create_cq;
 	ibdev->ib_dev.modify_cq		= mlx4_ib_modify_cq;
+	ibdev->ib_dev.resize_cq		= mlx4_ib_resize_cq;
 	ibdev->ib_dev.destroy_cq	= mlx4_ib_destroy_cq;
 	ibdev->ib_dev.poll_cq		= mlx4_ib_poll_cq;
 	ibdev->ib_dev.req_notify_cq	= mlx4_ib_arm_cq;
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index ef8ad96e1432..9e637323c155 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -78,13 +78,21 @@ struct mlx4_ib_cq_buf {
 	struct mlx4_mtt		mtt;
 };
 
+struct mlx4_ib_cq_resize {
+	struct mlx4_ib_cq_buf	buf;
+	int			cqe;
+};
+
 struct mlx4_ib_cq {
 	struct ib_cq		ibcq;
 	struct mlx4_cq		mcq;
 	struct mlx4_ib_cq_buf	buf;
+	struct mlx4_ib_cq_resize *resize_buf;
 	struct mlx4_ib_db	db;
 	spinlock_t		lock;
+	struct mutex		resize_mutex;
 	struct ib_umem	       *umem;
+	struct ib_umem	       *resize_umem;
 };
 
 struct mlx4_ib_mr {
@@ -255,6 +263,7 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 int mlx4_ib_dereg_mr(struct ib_mr *mr);
 
 int mlx4_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period);
+int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata);
 struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector,
 				struct ib_ucontext *context,
 				struct ib_udata *udata);
diff --git a/drivers/net/mlx4/cq.c b/drivers/net/mlx4/cq.c
index 8c314341434f..caa5bcf54e35 100644
--- a/drivers/net/mlx4/cq.c
+++ b/drivers/net/mlx4/cq.c
@@ -159,6 +159,34 @@ int mlx4_cq_modify(struct mlx4_dev *dev, struct mlx4_cq *cq,
 }
 EXPORT_SYMBOL_GPL(mlx4_cq_modify);
 
+int mlx4_cq_resize(struct mlx4_dev *dev, struct mlx4_cq *cq,
+		   int entries, struct mlx4_mtt *mtt)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_cq_context *cq_context;
+	u64 mtt_addr;
+	int err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	cq_context = mailbox->buf;
+	memset(cq_context, 0, sizeof *cq_context);
+
+	cq_context->logsize_usrpage = cpu_to_be32(ilog2(entries) << 24);
+	cq_context->log_page_size   = mtt->page_shift - 12;
+	mtt_addr = mlx4_mtt_addr(dev, mtt);
+	cq_context->mtt_base_addr_h = mtt_addr >> 32;
+	cq_context->mtt_base_addr_l = cpu_to_be32(mtt_addr & 0xffffffff);
+
+	err = mlx4_MODIFY_CQ(dev, mailbox, cq->cqn, 1);
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_cq_resize);
+
 int mlx4_cq_alloc(struct mlx4_dev *dev, int nent, struct mlx4_mtt *mtt,
 		  struct mlx4_uar *uar, u64 db_rec, struct mlx4_cq *cq)
 {
diff --git a/include/linux/mlx4/cq.h b/include/linux/mlx4/cq.h
index f7c3511c594b..071cf96cf01f 100644
--- a/include/linux/mlx4/cq.h
+++ b/include/linux/mlx4/cq.h
@@ -132,5 +132,7 @@ enum {
 
 int mlx4_cq_modify(struct mlx4_dev *dev, struct mlx4_cq *cq,
 		   u16 count, u16 period);
+int mlx4_cq_resize(struct mlx4_dev *dev, struct mlx4_cq *cq,
+		   int entries, struct mlx4_mtt *mtt);
 
 #endif /* MLX4_CQ_H */
-- 
cgit v1.2.3


From 374b8f45f1d5cb17f45ba1d7c74ce8cc9e2f1407 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Thu, 17 Apr 2008 07:45:58 +0200
Subject: [S390] allnoconfig build error.

Fix the following link error with allnoconfig:

vmem.c:(.text+0x175c): undefined reference to `smp_ptlb_all'
vmem.c:(.text+0x1b24): undefined reference to `smp_ptlb_all'
fork.c:(.text+0x4190): undefined reference to `smp_ptlb_all'
: undefined reference to `smp_ptlb_all'
: undefined reference to `smp_ptlb_all'
mm/built-in.o:: more undefined references to `smp_ptlb_all' follow
make[1]: *** [.tmp_vmlinux1] Error 1
make: *** [sub-make] Error 2

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 include/asm-s390/tlbflush.h | 33 +++++++++++++++++++++------------
 1 file changed, 21 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/tlbflush.h b/include/asm-s390/tlbflush.h
index 35fb4f9127b2..de723470c5d4 100644
--- a/include/asm-s390/tlbflush.h
+++ b/include/asm-s390/tlbflush.h
@@ -13,6 +13,7 @@ static inline void __tlb_flush_local(void)
 	asm volatile("ptlb" : : : "memory");
 }
 
+#ifdef CONFIG_SMP
 /*
  * Flush all tlb entries on all cpus.
  */
@@ -39,6 +40,25 @@ static inline void __tlb_flush_global(void)
 		: : "d" (reg2), "d" (reg3), "d" (reg4), "m" (dummy) : "cc" );
 }
 
+static inline void __tlb_flush_full(struct mm_struct *mm)
+{
+	cpumask_t local_cpumask;
+
+	preempt_disable();
+	/*
+	 * If the process only ran on the local cpu, do a local flush.
+	 */
+	local_cpumask = cpumask_of_cpu(smp_processor_id());
+	if (cpus_equal(mm->cpu_vm_mask, local_cpumask))
+		__tlb_flush_local();
+	else
+		__tlb_flush_global();
+	preempt_enable();
+}
+#else
+#define __tlb_flush_full(mm)	__tlb_flush_local()
+#endif
+
 /*
  * Flush all tlb entries of a page table on all cpus.
  */
@@ -51,8 +71,6 @@ static inline void __tlb_flush_idte(unsigned long asce)
 
 static inline void __tlb_flush_mm(struct mm_struct * mm)
 {
-	cpumask_t local_cpumask;
-
 	if (unlikely(cpus_empty(mm->cpu_vm_mask)))
 		return;
 	/*
@@ -69,16 +87,7 @@ static inline void __tlb_flush_mm(struct mm_struct * mm)
 				 mm->context.asce_bits);
 		return;
 	}
-	preempt_disable();
-	/*
-	 * If the process only ran on the local cpu, do a local flush.
-	 */
-	local_cpumask = cpumask_of_cpu(smp_processor_id());
-	if (cpus_equal(mm->cpu_vm_mask, local_cpumask))
-		__tlb_flush_local();
-	else
-		__tlb_flush_global();
-	preempt_enable();
+	__tlb_flush_full(mm);
 }
 
 static inline void __tlb_flush_mm_cond(struct mm_struct * mm)
-- 
cgit v1.2.3


From aa24f7f08baca5aa9201901131cbdd0b14deceb6 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:09 +0200
Subject: [S390] KVM preparation: split sysinfo definitions for kvm use

drivers/s390/sysinfo.c uses the store system information intruction to query
the system about information of the machine, the LPAR and additional
hypervisors. KVM has to implement the host part for this instruction.

To avoid code duplication, this patch splits the common definitions from
sysinfo.c into a separate header file include/asm-s390/sysinfo.h for KVM use.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 drivers/s390/sysinfo.c     | 100 +---------------------------------------
 include/asm-s390/sysinfo.h | 111 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 112 insertions(+), 99 deletions(-)
 create mode 100644 include/asm-s390/sysinfo.h

(limited to 'include')

diff --git a/drivers/s390/sysinfo.c b/drivers/s390/sysinfo.c
index 291ff6235fe2..43fffa3b099d 100644
--- a/drivers/s390/sysinfo.c
+++ b/drivers/s390/sysinfo.c
@@ -11,111 +11,13 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <asm/ebcdic.h>
+#include <asm/sysinfo.h>
 
 /* Sigh, math-emu. Don't ask. */
 #include <asm/sfp-util.h>
 #include <math-emu/soft-fp.h>
 #include <math-emu/single.h>
 
-struct sysinfo_1_1_1 {
-	char reserved_0[32];
-	char manufacturer[16];
-	char type[4];
-	char reserved_1[12];
-	char model_capacity[16];
-	char sequence[16];
-	char plant[4];
-	char model[16];
-};
-
-struct sysinfo_1_2_1 {
-	char reserved_0[80];
-	char sequence[16];
-	char plant[4];
-	char reserved_1[2];
-	unsigned short cpu_address;
-};
-
-struct sysinfo_1_2_2 {
-	char format;
-	char reserved_0[1];
-	unsigned short acc_offset;
-	char reserved_1[24];
-	unsigned int secondary_capability;
-	unsigned int capability;
-	unsigned short cpus_total;
-	unsigned short cpus_configured;
-	unsigned short cpus_standby;
-	unsigned short cpus_reserved;
-	unsigned short adjustment[0];
-};
-
-struct sysinfo_1_2_2_extension {
-	unsigned int alt_capability;
-	unsigned short alt_adjustment[0];
-};
-
-struct sysinfo_2_2_1 {
-	char reserved_0[80];
-	char sequence[16];
-	char plant[4];
-	unsigned short cpu_id;
-	unsigned short cpu_address;
-};
-
-struct sysinfo_2_2_2 {
-	char reserved_0[32];
-	unsigned short lpar_number;
-	char reserved_1;
-	unsigned char characteristics;
-	unsigned short cpus_total;
-	unsigned short cpus_configured;
-	unsigned short cpus_standby;
-	unsigned short cpus_reserved;
-	char name[8];
-	unsigned int caf;
-	char reserved_2[16];
-	unsigned short cpus_dedicated;
-	unsigned short cpus_shared;
-};
-
-#define LPAR_CHAR_DEDICATED	(1 << 7)
-#define LPAR_CHAR_SHARED	(1 << 6)
-#define LPAR_CHAR_LIMITED	(1 << 5)
-
-struct sysinfo_3_2_2 {
-	char reserved_0[31];
-	unsigned char count;
-	struct {
-		char reserved_0[4];
-		unsigned short cpus_total;
-		unsigned short cpus_configured;
-		unsigned short cpus_standby;
-		unsigned short cpus_reserved;
-		char name[8];
-		unsigned int caf;
-		char cpi[16];
-		char reserved_1[24];
-
-	} vm[8];
-};
-
-static inline int stsi(void *sysinfo, int fc, int sel1, int sel2)
-{
-	register int r0 asm("0") = (fc << 28) | sel1;
-	register int r1 asm("1") = sel2;
-
-	asm volatile(
-		"   stsi 0(%2)\n"
-		"0: jz   2f\n"
-		"1: lhi  %0,%3\n"
-		"2:\n"
-		EX_TABLE(0b,1b)
-		: "+d" (r0) : "d" (r1), "a" (sysinfo), "K" (-ENOSYS)
-		: "cc", "memory" );
-	return r0;
-}
-
 static inline int stsi_0(void)
 {
 	int rc = stsi (NULL, 0, 0, 0);
diff --git a/include/asm-s390/sysinfo.h b/include/asm-s390/sysinfo.h
new file mode 100644
index 000000000000..014f2a24664e
--- /dev/null
+++ b/include/asm-s390/sysinfo.h
@@ -0,0 +1,111 @@
+/*
+ * definition for store system information stsi
+ *
+ * Copyright IBM Corp. 2001,2008
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ *
+ *    Author(s): Ulrich Weigand <weigand@de.ibm.com>
+ *		 Christian Borntraeger <borntraeger@de.ibm.com>
+ */
+
+struct sysinfo_1_1_1 {
+	char reserved_0[32];
+	char manufacturer[16];
+	char type[4];
+	char reserved_1[12];
+	char model_capacity[16];
+	char sequence[16];
+	char plant[4];
+	char model[16];
+};
+
+struct sysinfo_1_2_1 {
+	char reserved_0[80];
+	char sequence[16];
+	char plant[4];
+	char reserved_1[2];
+	unsigned short cpu_address;
+};
+
+struct sysinfo_1_2_2 {
+	char format;
+	char reserved_0[1];
+	unsigned short acc_offset;
+	char reserved_1[24];
+	unsigned int secondary_capability;
+	unsigned int capability;
+	unsigned short cpus_total;
+	unsigned short cpus_configured;
+	unsigned short cpus_standby;
+	unsigned short cpus_reserved;
+	unsigned short adjustment[0];
+};
+
+struct sysinfo_1_2_2_extension {
+	unsigned int alt_capability;
+	unsigned short alt_adjustment[0];
+};
+
+struct sysinfo_2_2_1 {
+	char reserved_0[80];
+	char sequence[16];
+	char plant[4];
+	unsigned short cpu_id;
+	unsigned short cpu_address;
+};
+
+struct sysinfo_2_2_2 {
+	char reserved_0[32];
+	unsigned short lpar_number;
+	char reserved_1;
+	unsigned char characteristics;
+	unsigned short cpus_total;
+	unsigned short cpus_configured;
+	unsigned short cpus_standby;
+	unsigned short cpus_reserved;
+	char name[8];
+	unsigned int caf;
+	char reserved_2[16];
+	unsigned short cpus_dedicated;
+	unsigned short cpus_shared;
+};
+
+#define LPAR_CHAR_DEDICATED	(1 << 7)
+#define LPAR_CHAR_SHARED	(1 << 6)
+#define LPAR_CHAR_LIMITED	(1 << 5)
+
+struct sysinfo_3_2_2 {
+	char reserved_0[31];
+	unsigned char count;
+	struct {
+		char reserved_0[4];
+		unsigned short cpus_total;
+		unsigned short cpus_configured;
+		unsigned short cpus_standby;
+		unsigned short cpus_reserved;
+		char name[8];
+		unsigned int caf;
+		char cpi[16];
+		char reserved_1[24];
+
+	} vm[8];
+};
+
+static inline int stsi(void *sysinfo, int fc, int sel1, int sel2)
+{
+	register int r0 asm("0") = (fc << 28) | sel1;
+	register int r1 asm("1") = sel2;
+
+	asm volatile(
+		"   stsi 0(%2)\n"
+		"0: jz   2f\n"
+		"1: lhi  %0,%3\n"
+		"2:\n"
+		EX_TABLE(0b, 1b)
+		: "+d" (r0) : "d" (r1), "a" (sysinfo), "K" (-ENOSYS)
+		: "cc", "memory");
+	return r0;
+}
-- 
cgit v1.2.3


From cbce70e687bf9c7968d63f058b4c3d2e90008ce2 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:10 +0200
Subject: [S390] Add new fields for System z10 to /proc/sysinfo

Add permanent and temporary model capacity and the corresponding
capacity value fields for the three capacity identifiers to the
output of /proc/sysinfo.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 drivers/s390/sysinfo.c     | 16 ++++++++++++++--
 include/asm-s390/sysinfo.h |  5 +++++
 2 files changed, 19 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/s390/sysinfo.c b/drivers/s390/sysinfo.c
index 43fffa3b099d..c3e4ab07b9cc 100644
--- a/drivers/s390/sysinfo.c
+++ b/drivers/s390/sysinfo.c
@@ -35,6 +35,8 @@ static int stsi_1_1_1(struct sysinfo_1_1_1 *info, char *page, int len)
 	EBCASC(info->sequence, sizeof(info->sequence));
 	EBCASC(info->plant, sizeof(info->plant));
 	EBCASC(info->model_capacity, sizeof(info->model_capacity));
+	EBCASC(info->model_perm_cap, sizeof(info->model_perm_cap));
+	EBCASC(info->model_temp_cap, sizeof(info->model_temp_cap));
 	len += sprintf(page + len, "Manufacturer:         %-16.16s\n",
 		       info->manufacturer);
 	len += sprintf(page + len, "Type:                 %-4.4s\n",
@@ -57,8 +59,18 @@ static int stsi_1_1_1(struct sysinfo_1_1_1 *info, char *page, int len)
 		       info->sequence);
 	len += sprintf(page + len, "Plant:                %-4.4s\n",
 		       info->plant);
-	len += sprintf(page + len, "Model Capacity:       %-16.16s\n",
-		       info->model_capacity);
+	len += sprintf(page + len, "Model Capacity:       %-16.16s %08u\n",
+		       info->model_capacity, *(u32 *) info->model_cap_rating);
+	if (info->model_perm_cap[0] != '\0')
+		len += sprintf(page + len,
+			       "Model Perm. Capacity: %-16.16s %08u\n",
+			       info->model_perm_cap,
+			       *(u32 *) info->model_perm_cap_rating);
+	if (info->model_temp_cap[0] != '\0')
+		len += sprintf(page + len,
+			       "Model Temp. Capacity: %-16.16s %08u\n",
+			       info->model_temp_cap,
+			       *(u32 *) info->model_temp_cap_rating);
 	return len;
 }
 
diff --git a/include/asm-s390/sysinfo.h b/include/asm-s390/sysinfo.h
index 014f2a24664e..abe10ae15e46 100644
--- a/include/asm-s390/sysinfo.h
+++ b/include/asm-s390/sysinfo.h
@@ -20,6 +20,11 @@ struct sysinfo_1_1_1 {
 	char sequence[16];
 	char plant[4];
 	char model[16];
+	char model_perm_cap[16];
+	char model_temp_cap[16];
+	char model_cap_rating[4];
+	char model_perm_cap_rating[4];
+	char model_temp_cap_rating[4];
 };
 
 struct sysinfo_1_2_1 {
-- 
cgit v1.2.3


From 7b758389a29cb7f456ec2d27b7a08cb3cc4e1f1c Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:11 +0200
Subject: [S390] Export stfle.

Make stfle visible so other code can call this.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/kernel/setup.c  | 11 +++++++++--
 include/asm-s390/system.h |  2 ++
 2 files changed, 11 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 290e504061a3..d682ff135a1c 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -687,7 +687,7 @@ static __init unsigned int stfl(void)
 	return S390_lowcore.stfl_fac_list;
 }
 
-static __init int stfle(unsigned long long *list, int doublewords)
+static int __init __stfle(unsigned long long *list, int doublewords)
 {
 	typedef struct { unsigned long long _[doublewords]; } addrtype;
 	register unsigned long __nr asm("0") = doublewords - 1;
@@ -697,6 +697,13 @@ static __init int stfle(unsigned long long *list, int doublewords)
 	return __nr + 1;
 }
 
+int __init stfle(unsigned long long *list, int doublewords)
+{
+	if (!(stfl() & (1UL << 24)))
+		return -EOPNOTSUPP;
+	return __stfle(list, doublewords);
+}
+
 /*
  * Setup hardware capabilities.
  */
@@ -741,7 +748,7 @@ static void __init setup_hwcaps(void)
 	 *   HWCAP_S390_DFP bit 6.
 	 */
 	if ((elf_hwcap & (1UL << 2)) &&
-	    stfle(&facility_list_extended, 1) > 0) {
+	    __stfle(&facility_list_extended, 1) > 0) {
 		if (facility_list_extended & (1ULL << (64 - 43)))
 			elf_hwcap |= 1UL << 6;
 	}
diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h
index 15aba30601a3..92098df4d6e3 100644
--- a/include/asm-s390/system.h
+++ b/include/asm-s390/system.h
@@ -406,6 +406,8 @@ __set_psw_mask(unsigned long mask)
 #define local_mcck_enable()  __set_psw_mask(psw_kernel_bits)
 #define local_mcck_disable() __set_psw_mask(psw_kernel_bits & ~PSW_MASK_MCHECK)
 
+int stfle(unsigned long long *list, int doublewords);
+
 #ifdef CONFIG_SMP
 
 extern void smp_ctl_set_bit(int cr, int bit);
-- 
cgit v1.2.3


From dbd70fb499952d0ba282f0159dafacfc31d50313 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:12 +0200
Subject: [S390] cpu topology support for s390.

Add s390 backend so we can give the scheduler some hints about the
cpu topology.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/Kconfig           |   4 +
 arch/s390/defconfig         |   1 +
 arch/s390/kernel/Makefile   |   2 +-
 arch/s390/kernel/setup.c    |   2 +
 arch/s390/kernel/smp.c      |   4 +-
 arch/s390/kernel/topology.c | 272 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-s390/smp.h      |   2 +
 include/asm-s390/topology.h |  14 +++
 8 files changed, 297 insertions(+), 4 deletions(-)
 create mode 100644 arch/s390/kernel/topology.c

(limited to 'include')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 1831833c430e..17cbc25c0c41 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -3,6 +3,10 @@
 # see Documentation/kbuild/kconfig-language.txt.
 #
 
+config SCHED_MC
+	def_bool y
+	depends on SMP
+
 config MMU
 	def_bool y
 
diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index 62f6b5a606dd..dcc3ec2ef643 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -3,6 +3,7 @@
 # Linux kernel version: 2.6.25-rc4
 # Wed Mar  5 11:22:59 2008
 #
+CONFIG_SCHED_MC=y
 CONFIG_MMU=y
 CONFIG_ZONE_DMA=y
 CONFIG_LOCKDEP_SUPPORT=y
diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index 4d3e38392cb1..ef2b2470c25c 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -19,7 +19,7 @@ obj-y	+= $(if $(CONFIG_64BIT),reipl64.o,reipl.o)
 extra-y				+= head.o init_task.o vmlinux.lds
 
 obj-$(CONFIG_MODULES)		+= s390_ksyms.o module.o
-obj-$(CONFIG_SMP)		+= smp.o
+obj-$(CONFIG_SMP)		+= smp.o topology.o
 
 obj-$(CONFIG_AUDIT)		+= audit.o
 compat-obj-$(CONFIG_AUDIT)	+= compat_audit.o
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index d682ff135a1c..22040d087d8a 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -39,6 +39,7 @@
 #include <linux/pfn.h>
 #include <linux/ctype.h>
 #include <linux/reboot.h>
+#include <linux/topology.h>
 
 #include <asm/ipl.h>
 #include <asm/uaccess.h>
@@ -830,6 +831,7 @@ setup_arch(char **cmdline_p)
 
         cpu_init();
         __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr;
+	s390_init_cpu_topology();
 
 	/*
 	 * Setup capabilities (ELF_HWCAP & ELF_PLATFORM).
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 8f894d380a62..5448aa87fa23 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -67,9 +67,7 @@ enum s390_cpu_state {
 	CPU_STATE_CONFIGURED,
 };
 
-#ifdef CONFIG_HOTPLUG_CPU
-static DEFINE_MUTEX(smp_cpu_state_mutex);
-#endif
+DEFINE_MUTEX(smp_cpu_state_mutex);
 static int smp_cpu_state[NR_CPUS];
 
 static DEFINE_PER_CPU(struct cpu, cpu_devices);
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
new file mode 100644
index 000000000000..369dc1c3bd10
--- /dev/null
+++ b/arch/s390/kernel/topology.c
@@ -0,0 +1,272 @@
+/*
+ *  arch/s390/kernel/topology.c
+ *
+ *    Copyright IBM Corp. 2007
+ *    Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/bootmem.h>
+#include <linux/sched.h>
+#include <linux/workqueue.h>
+#include <linux/cpu.h>
+#include <linux/smp.h>
+#include <asm/delay.h>
+#include <asm/s390_ext.h>
+#include <asm/sysinfo.h>
+
+#define CPU_BITS 64
+
+struct tl_cpu {
+	unsigned char reserved[6];
+	unsigned short origin;
+	unsigned long mask[CPU_BITS / BITS_PER_LONG];
+};
+
+struct tl_container {
+	unsigned char reserved[8];
+};
+
+union tl_entry {
+	unsigned char nl;
+	struct tl_cpu cpu;
+	struct tl_container container;
+};
+
+#define NR_MAG 6
+
+struct tl_info {
+	unsigned char reserved0[2];
+	unsigned short length;
+	unsigned char mag[NR_MAG];
+	unsigned char reserved1;
+	unsigned char mnest;
+	unsigned char reserved2[4];
+	union tl_entry tle[0];
+};
+
+struct core_info {
+	struct core_info *next;
+	cpumask_t mask;
+};
+
+static void topology_work_fn(struct work_struct *work);
+static struct tl_info *tl_info;
+static struct core_info core_info;
+static int machine_has_topology;
+static int machine_has_topology_irq;
+static struct timer_list topology_timer;
+static void set_topology_timer(void);
+static DECLARE_WORK(topology_work, topology_work_fn);
+
+cpumask_t cpu_coregroup_map(unsigned int cpu)
+{
+	struct core_info *core = &core_info;
+	cpumask_t mask;
+
+	cpus_clear(mask);
+	if (!machine_has_topology)
+		return cpu_present_map;
+	mutex_lock(&smp_cpu_state_mutex);
+	while (core) {
+		if (cpu_isset(cpu, core->mask)) {
+			mask = core->mask;
+			break;
+		}
+		core = core->next;
+	}
+	mutex_unlock(&smp_cpu_state_mutex);
+	if (cpus_empty(mask))
+		mask = cpumask_of_cpu(cpu);
+	return mask;
+}
+
+static void add_cpus_to_core(struct tl_cpu *tl_cpu, struct core_info *core)
+{
+	unsigned int cpu;
+
+	for (cpu = find_first_bit(&tl_cpu->mask[0], CPU_BITS);
+	     cpu < CPU_BITS;
+	     cpu = find_next_bit(&tl_cpu->mask[0], CPU_BITS, cpu + 1))
+	{
+		unsigned int rcpu, lcpu;
+
+		rcpu = CPU_BITS - 1 - cpu + tl_cpu->origin;
+		for_each_present_cpu(lcpu) {
+			if (__cpu_logical_map[lcpu] == rcpu)
+				cpu_set(lcpu, core->mask);
+		}
+	}
+}
+
+static void clear_cores(void)
+{
+	struct core_info *core = &core_info;
+
+	while (core) {
+		cpus_clear(core->mask);
+		core = core->next;
+	}
+}
+
+static union tl_entry *next_tle(union tl_entry *tle)
+{
+	if (tle->nl)
+		return (union tl_entry *)((struct tl_container *)tle + 1);
+	else
+		return (union tl_entry *)((struct tl_cpu *)tle + 1);
+}
+
+static void tl_to_cores(struct tl_info *info)
+{
+	union tl_entry *tle, *end;
+	struct core_info *core = &core_info;
+
+	mutex_lock(&smp_cpu_state_mutex);
+	clear_cores();
+	tle = (union tl_entry *)&info->tle;
+	end = (union tl_entry *)((unsigned long)info + info->length);
+	while (tle < end) {
+		switch (tle->nl) {
+		case 5:
+		case 4:
+		case 3:
+		case 2:
+			break;
+		case 1:
+			core = core->next;
+			break;
+		case 0:
+			add_cpus_to_core(&tle->cpu, core);
+			break;
+		default:
+			clear_cores();
+			machine_has_topology = 0;
+			return;
+		}
+		tle = next_tle(tle);
+	}
+	mutex_unlock(&smp_cpu_state_mutex);
+}
+
+static int ptf(void)
+{
+	int rc;
+
+	asm volatile(
+		"	.insn	rre,0xb9a20000,%1,%1\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		: "=d" (rc)
+		: "d" (2UL)  : "cc");
+	return rc;
+}
+
+void arch_update_cpu_topology(void)
+{
+	struct tl_info *info = tl_info;
+	struct sys_device *sysdev;
+	int cpu;
+
+	if (!machine_has_topology)
+		return;
+	ptf();
+	stsi(info, 15, 1, 2);
+	tl_to_cores(info);
+	for_each_online_cpu(cpu) {
+		sysdev = get_cpu_sysdev(cpu);
+		kobject_uevent(&sysdev->kobj, KOBJ_CHANGE);
+	}
+}
+
+static void topology_work_fn(struct work_struct *work)
+{
+	arch_reinit_sched_domains();
+}
+
+static void topology_timer_fn(unsigned long ignored)
+{
+	if (ptf())
+		schedule_work(&topology_work);
+	set_topology_timer();
+}
+
+static void set_topology_timer(void)
+{
+	topology_timer.function = topology_timer_fn;
+	topology_timer.data = 0;
+	topology_timer.expires = jiffies + 60 * HZ;
+	add_timer(&topology_timer);
+}
+
+static void topology_interrupt(__u16 code)
+{
+	schedule_work(&topology_work);
+}
+
+static int __init init_topology_update(void)
+{
+	int rc;
+
+	if (!machine_has_topology)
+		return 0;
+	init_timer(&topology_timer);
+	if (machine_has_topology_irq) {
+		rc = register_external_interrupt(0x2005, topology_interrupt);
+		if (rc)
+			return rc;
+		ctl_set_bit(0, 8);
+	}
+	else
+		set_topology_timer();
+	return 0;
+}
+__initcall(init_topology_update);
+
+void __init s390_init_cpu_topology(void)
+{
+	unsigned long long facility_bits;
+	struct tl_info *info;
+	struct core_info *core;
+	int nr_cores;
+	int i;
+
+	if (stfle(&facility_bits, 1) <= 0)
+		return;
+	if (!(facility_bits & (1ULL << 52)) || !(facility_bits & (1ULL << 61)))
+		return;
+	machine_has_topology = 1;
+
+	if (facility_bits & (1ULL << 51))
+		machine_has_topology_irq = 1;
+
+	tl_info = alloc_bootmem_pages(PAGE_SIZE);
+	if (!tl_info)
+		goto error;
+	info = tl_info;
+	stsi(info, 15, 1, 2);
+
+	nr_cores = info->mag[NR_MAG - 2];
+	for (i = 0; i < info->mnest - 2; i++)
+		nr_cores *= info->mag[NR_MAG - 3 - i];
+
+	printk(KERN_INFO "CPU topology:");
+	for (i = 0; i < NR_MAG; i++)
+		printk(" %d", info->mag[i]);
+	printk(" / %d\n", info->mnest);
+
+	core = &core_info;
+	for (i = 0; i < nr_cores; i++) {
+		core->next = alloc_bootmem(sizeof(struct core_info));
+		core = core->next;
+		if (!core)
+			goto error;
+	}
+	return;
+error:
+	machine_has_topology = 0;
+	machine_has_topology_irq = 0;
+}
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index c7b74326a527..a464a66c65d3 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -90,6 +90,8 @@ extern void __cpu_die (unsigned int cpu);
 extern void cpu_die (void) __attribute__ ((noreturn));
 extern int __cpu_up (unsigned int cpu);
 
+extern struct mutex smp_cpu_state_mutex;
+
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 	void *info, int wait);
 #endif
diff --git a/include/asm-s390/topology.h b/include/asm-s390/topology.h
index 613aa64019da..3b28a09a2108 100644
--- a/include/asm-s390/topology.h
+++ b/include/asm-s390/topology.h
@@ -1,6 +1,20 @@
 #ifndef _ASM_S390_TOPOLOGY_H
 #define _ASM_S390_TOPOLOGY_H
 
+#include <linux/cpumask.h>
+
+#define mc_capable()	(1)
+
+cpumask_t cpu_coregroup_map(unsigned int cpu);
+
+#ifdef CONFIG_SMP
+void s390_init_cpu_topology(void);
+#else
+static inline void s390_init_cpu_topology(void)
+{
+};
+#endif
+
 #include <asm-generic/topology.h>
 
 #endif /* _ASM_S390_TOPOLOGY_H */
-- 
cgit v1.2.3


From c10fde0d9e2112c25052a8742e893ec5965c0007 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:13 +0200
Subject: [S390] Vertical cpu management.

If vertical cpu polarization is active then the hypervisor will
dispatch certain cpus for a longer time than other cpus for maximum
performance. For example if a guest would have three virtual cpus,
each of them with a share of 33 percent, then in case of vertical
cpu polarization all of the processing time would be combined to a
single cpu which would run all the time, while the other two cpus
would get nearly no cpu time.

There are three different types of vertical cpus: high, medium and
low. Low cpus hardly get any real cpu time, while high cpus get a
full real cpu. Medium cpus get something in between.

In order to switch between the two possible modes (default is
horizontal) a 0 for horizontal polarization or a 1 for vertical
polarization must be written to the dispatching sysfs attribute:

/sys/devices/system/cpu/dispatching

The polarization of each single cpu can be figured out by the
polarization sysfs attribute of each cpu:

/sys/devices/system/cpu/cpuX/polarization

horizontal, vertical:high, vertical:medium, vertical:low or unknown.

When switching polarization the polarization attribute may contain
the value unknown until the configuration change is done and the
kernel has figured out the new polarization of each cpu.

Note that running a system with different types of vertical cpus may
result in significant performance regressions. If possible only one
type of vertical cpus should be used. All other cpus should be
offlined.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/kernel/smp.c      | 83 +++++++++++++++++++++++++++++++++++++++++++--
 arch/s390/kernel/topology.c | 72 +++++++++++++++++++++++++++++++--------
 include/asm-s390/smp.h      |  1 +
 include/asm-s390/topology.h |  9 +++++
 4 files changed, 148 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 5448aa87fa23..d1e8e8a3fb66 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -68,7 +68,9 @@ enum s390_cpu_state {
 };
 
 DEFINE_MUTEX(smp_cpu_state_mutex);
+int smp_cpu_polarization[NR_CPUS];
 static int smp_cpu_state[NR_CPUS];
+static int cpu_management;
 
 static DEFINE_PER_CPU(struct cpu, cpu_devices);
 DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
@@ -454,6 +456,7 @@ static int smp_rescan_cpus_sigp(cpumask_t avail)
 		if (cpu_known(cpu_id))
 			continue;
 		__cpu_logical_map[logical_cpu] = cpu_id;
+		smp_cpu_polarization[logical_cpu] = POLARIZATION_UNKNWN;
 		if (!cpu_stopped(logical_cpu))
 			continue;
 		cpu_set(logical_cpu, cpu_present_map);
@@ -487,6 +490,7 @@ static int smp_rescan_cpus_sclp(cpumask_t avail)
 		if (cpu_known(cpu_id))
 			continue;
 		__cpu_logical_map[logical_cpu] = cpu_id;
+		smp_cpu_polarization[logical_cpu] = POLARIZATION_UNKNWN;
 		cpu_set(logical_cpu, cpu_present_map);
 		if (cpu >= info->configured)
 			smp_cpu_state[logical_cpu] = CPU_STATE_STANDBY;
@@ -844,6 +848,7 @@ void __init smp_prepare_boot_cpu(void)
 	S390_lowcore.percpu_offset = __per_cpu_offset[0];
 	current_set[0] = current;
 	smp_cpu_state[0] = CPU_STATE_CONFIGURED;
+	smp_cpu_polarization[0] = POLARIZATION_UNKNWN;
 	spin_lock_init(&(&__get_cpu_var(s390_idle))->lock);
 }
 
@@ -895,15 +900,19 @@ static ssize_t cpu_configure_store(struct sys_device *dev, const char *buf,
 	case 0:
 		if (smp_cpu_state[cpu] == CPU_STATE_CONFIGURED) {
 			rc = sclp_cpu_deconfigure(__cpu_logical_map[cpu]);
-			if (!rc)
+			if (!rc) {
 				smp_cpu_state[cpu] = CPU_STATE_STANDBY;
+				smp_cpu_polarization[cpu] = POLARIZATION_UNKNWN;
+			}
 		}
 		break;
 	case 1:
 		if (smp_cpu_state[cpu] == CPU_STATE_STANDBY) {
 			rc = sclp_cpu_configure(__cpu_logical_map[cpu]);
-			if (!rc)
+			if (!rc) {
 				smp_cpu_state[cpu] = CPU_STATE_CONFIGURED;
+				smp_cpu_polarization[cpu] = POLARIZATION_UNKNWN;
+			}
 		}
 		break;
 	default:
@@ -917,6 +926,34 @@ out:
 static SYSDEV_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store);
 #endif /* CONFIG_HOTPLUG_CPU */
 
+static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf)
+{
+	int cpu = dev->id;
+	ssize_t count;
+
+	mutex_lock(&smp_cpu_state_mutex);
+	switch (smp_cpu_polarization[cpu]) {
+	case POLARIZATION_HRZ:
+		count = sprintf(buf, "horizontal\n");
+		break;
+	case POLARIZATION_VL:
+		count = sprintf(buf, "vertical:low\n");
+		break;
+	case POLARIZATION_VM:
+		count = sprintf(buf, "vertical:medium\n");
+		break;
+	case POLARIZATION_VH:
+		count = sprintf(buf, "vertical:high\n");
+		break;
+	default:
+		count = sprintf(buf, "unknown\n");
+		break;
+	}
+	mutex_unlock(&smp_cpu_state_mutex);
+	return count;
+}
+static SYSDEV_ATTR(polarization, 0444, cpu_polarization_show, NULL);
+
 static ssize_t show_cpu_address(struct sys_device *dev, char *buf)
 {
 	return sprintf(buf, "%d\n", __cpu_logical_map[dev->id]);
@@ -929,6 +966,7 @@ static struct attribute *cpu_common_attrs[] = {
 	&attr_configure.attr,
 #endif
 	&attr_address.attr,
+	&attr_polarization.attr,
 	NULL,
 };
 
@@ -1073,11 +1111,48 @@ static ssize_t __ref rescan_store(struct sys_device *dev,
 out:
 	put_online_cpus();
 	mutex_unlock(&smp_cpu_state_mutex);
+	if (!cpus_empty(newcpus))
+		topology_schedule_update();
 	return rc ? rc : count;
 }
 static SYSDEV_ATTR(rescan, 0200, NULL, rescan_store);
 #endif /* CONFIG_HOTPLUG_CPU */
 
+static ssize_t dispatching_show(struct sys_device *dev, char *buf)
+{
+	ssize_t count;
+
+	mutex_lock(&smp_cpu_state_mutex);
+	count = sprintf(buf, "%d\n", cpu_management);
+	mutex_unlock(&smp_cpu_state_mutex);
+	return count;
+}
+
+static ssize_t dispatching_store(struct sys_device *dev, const char *buf,
+				 size_t count)
+{
+	int val, rc;
+	char delim;
+
+	if (sscanf(buf, "%d %c", &val, &delim) != 1)
+		return -EINVAL;
+	if (val != 0 && val != 1)
+		return -EINVAL;
+	rc = 0;
+	mutex_lock(&smp_cpu_state_mutex);
+	get_online_cpus();
+	if (cpu_management == val)
+		goto out;
+	rc = topology_set_cpu_management(val);
+	if (!rc)
+		cpu_management = val;
+out:
+	put_online_cpus();
+	mutex_unlock(&smp_cpu_state_mutex);
+	return rc ? rc : count;
+}
+static SYSDEV_ATTR(dispatching, 0644, dispatching_show, dispatching_store);
+
 static int __init topology_init(void)
 {
 	int cpu;
@@ -1091,6 +1166,10 @@ static int __init topology_init(void)
 	if (rc)
 		return rc;
 #endif
+	rc = sysfs_create_file(&cpu_sysdev_class.kset.kobj,
+			       &attr_dispatching.attr);
+	if (rc)
+		return rc;
 	for_each_present_cpu(cpu) {
 		rc = smp_add_present_cpu(cpu);
 		if (rc)
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 369dc1c3bd10..12b39b3d9c38 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -1,6 +1,4 @@
 /*
- *  arch/s390/kernel/topology.c
- *
  *    Copyright IBM Corp. 2007
  *    Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
  */
@@ -19,9 +17,17 @@
 #include <asm/sysinfo.h>
 
 #define CPU_BITS 64
+#define NR_MAG 6
+
+#define PTF_HORIZONTAL	(0UL)
+#define PTF_VERTICAL	(1UL)
+#define PTF_CHECK	(2UL)
 
 struct tl_cpu {
-	unsigned char reserved[6];
+	unsigned char reserved0[4];
+	unsigned char :6;
+	unsigned char pp:2;
+	unsigned char reserved1;
 	unsigned short origin;
 	unsigned long mask[CPU_BITS / BITS_PER_LONG];
 };
@@ -36,8 +42,6 @@ union tl_entry {
 	struct tl_container container;
 };
 
-#define NR_MAG 6
-
 struct tl_info {
 	unsigned char reserved0[2];
 	unsigned short length;
@@ -96,8 +100,10 @@ static void add_cpus_to_core(struct tl_cpu *tl_cpu, struct core_info *core)
 
 		rcpu = CPU_BITS - 1 - cpu + tl_cpu->origin;
 		for_each_present_cpu(lcpu) {
-			if (__cpu_logical_map[lcpu] == rcpu)
+			if (__cpu_logical_map[lcpu] == rcpu) {
 				cpu_set(lcpu, core->mask);
+				smp_cpu_polarization[lcpu] = tl_cpu->pp;
+			}
 		}
 	}
 }
@@ -127,7 +133,7 @@ static void tl_to_cores(struct tl_info *info)
 
 	mutex_lock(&smp_cpu_state_mutex);
 	clear_cores();
-	tle = (union tl_entry *)&info->tle;
+	tle = info->tle;
 	end = (union tl_entry *)((unsigned long)info + info->length);
 	while (tle < end) {
 		switch (tle->nl) {
@@ -152,7 +158,17 @@ static void tl_to_cores(struct tl_info *info)
 	mutex_unlock(&smp_cpu_state_mutex);
 }
 
-static int ptf(void)
+static void topology_update_polarization_simple(void)
+{
+	int cpu;
+
+	mutex_lock(&smp_cpu_state_mutex);
+	for_each_present_cpu(cpu)
+		smp_cpu_polarization[cpu] = POLARIZATION_HRZ;
+	mutex_unlock(&smp_cpu_state_mutex);
+}
+
+static int ptf(unsigned long fc)
 {
 	int rc;
 
@@ -161,7 +177,25 @@ static int ptf(void)
 		"	ipm	%0\n"
 		"	srl	%0,28\n"
 		: "=d" (rc)
-		: "d" (2UL)  : "cc");
+		: "d" (fc)  : "cc");
+	return rc;
+}
+
+int topology_set_cpu_management(int fc)
+{
+	int cpu;
+	int rc;
+
+	if (!machine_has_topology)
+		return -EOPNOTSUPP;
+	if (fc)
+		rc = ptf(PTF_VERTICAL);
+	else
+		rc = ptf(PTF_HORIZONTAL);
+	if (rc)
+		return -EBUSY;
+	for_each_present_cpu(cpu)
+		smp_cpu_polarization[cpu] = POLARIZATION_UNKNWN;
 	return rc;
 }
 
@@ -171,9 +205,10 @@ void arch_update_cpu_topology(void)
 	struct sys_device *sysdev;
 	int cpu;
 
-	if (!machine_has_topology)
+	if (!machine_has_topology) {
+		topology_update_polarization_simple();
 		return;
-	ptf();
+	}
 	stsi(info, 15, 1, 2);
 	tl_to_cores(info);
 	for_each_online_cpu(cpu) {
@@ -187,10 +222,15 @@ static void topology_work_fn(struct work_struct *work)
 	arch_reinit_sched_domains();
 }
 
+void topology_schedule_update(void)
+{
+	schedule_work(&topology_work);
+}
+
 static void topology_timer_fn(unsigned long ignored)
 {
-	if (ptf())
-		schedule_work(&topology_work);
+	if (ptf(PTF_CHECK))
+		topology_schedule_update();
 	set_topology_timer();
 }
 
@@ -211,9 +251,11 @@ static int __init init_topology_update(void)
 {
 	int rc;
 
-	if (!machine_has_topology)
+	if (!machine_has_topology) {
+		topology_update_polarization_simple();
 		return 0;
-	init_timer(&topology_timer);
+	}
+	init_timer_deferrable(&topology_timer);
 	if (machine_has_topology_irq) {
 		rc = register_external_interrupt(0x2005, topology_interrupt);
 		if (rc)
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index a464a66c65d3..6f3821a6a902 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -91,6 +91,7 @@ extern void cpu_die (void) __attribute__ ((noreturn));
 extern int __cpu_up (unsigned int cpu);
 
 extern struct mutex smp_cpu_state_mutex;
+extern int smp_cpu_polarization[];
 
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 	void *info, int wait);
diff --git a/include/asm-s390/topology.h b/include/asm-s390/topology.h
index 3b28a09a2108..8e97b06f298a 100644
--- a/include/asm-s390/topology.h
+++ b/include/asm-s390/topology.h
@@ -7,6 +7,15 @@
 
 cpumask_t cpu_coregroup_map(unsigned int cpu);
 
+int topology_set_cpu_management(int fc);
+void topology_schedule_update(void);
+
+#define POLARIZATION_UNKNWN	(-1)
+#define POLARIZATION_HRZ	(0)
+#define POLARIZATION_VL		(1)
+#define POLARIZATION_VM		(2)
+#define POLARIZATION_VH		(3)
+
 #ifdef CONFIG_SMP
 void s390_init_cpu_topology(void);
 #else
-- 
cgit v1.2.3


From c0015f91d8414f55d2debfe9984a04b98b48f087 Mon Sep 17 00:00:00 2001
From: Jan Glauber <jan.glauber@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:16 +0200
Subject: [S390] switch sched_clock to store-clock-extended.

Add get_clock_xt to read an 8 byte clock value using store clock
extended (STCKE) and use get_clock_xt for sched_clock. STCKE should
be faster than STCK on newer machines.

Signed-off-by: Jan Glauber <jan.glauber@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/kernel/time.c  |  2 +-
 include/asm-s390/timex.h | 12 +++++++-----
 2 files changed, 8 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index cb232c155360..925f9dc0b0a0 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -66,7 +66,7 @@ static u64 xtime_cc;
  */
 unsigned long long sched_clock(void)
 {
-	return ((get_clock() - jiffies_timer_cc) * 125) >> 9;
+	return ((get_clock_xt() - jiffies_timer_cc) * 125) >> 9;
 }
 
 /*
diff --git a/include/asm-s390/timex.h b/include/asm-s390/timex.h
index 98229db24314..6dd7eecbb8e4 100644
--- a/include/asm-s390/timex.h
+++ b/include/asm-s390/timex.h
@@ -62,16 +62,18 @@ static inline unsigned long long get_clock (void)
 	return clk;
 }
 
-static inline void get_clock_extended(void *dest)
+static inline unsigned long long get_clock_xt(void)
 {
-	typedef struct { unsigned long long clk[2]; } __clock_t;
+	unsigned char clk[16];
 
 #if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
-	asm volatile("stcke %0" : "=Q" (*((__clock_t *)dest)) : : "cc");
+	asm volatile("stcke %0" : "=Q" (clk) : : "cc");
 #else /* __GNUC__ */
-	asm volatile("stcke 0(%1)" : "=m" (*((__clock_t *)dest))
-				   : "a" ((__clock_t *)dest) : "cc");
+	asm volatile("stcke 0(%1)" : "=m" (clk)
+				   : "a" (clk) : "cc");
 #endif /* __GNUC__ */
+
+	return *((unsigned long long *)&clk[1]);
 }
 
 static inline cycles_t get_cycles(void)
-- 
cgit v1.2.3


From 9637c3f318374e2fcc37e354f9782a705b517387 Mon Sep 17 00:00:00 2001
From: Michael Holzheu <holzheu@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:18 +0200
Subject: [S390] Add debug_register_mode() function to debug feature API

The new function supports setting of permissions for the debugfs files
created by the debug feature. In addition to that, the function provides
uid and gid as parameters for future use. Currently only root is allowed
for uid and gid.

Signed-off-by: Michael Holzheu <holzheu@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 Documentation/s390/s390dbf.txt | 21 +++++++++++++++++
 arch/s390/kernel/debug.c       | 51 +++++++++++++++++++++++++++++++-----------
 include/asm-s390/debug.h       |  5 +++++
 3 files changed, 64 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/Documentation/s390/s390dbf.txt b/Documentation/s390/s390dbf.txt
index 0eb7c58916de..e05420973698 100644
--- a/Documentation/s390/s390dbf.txt
+++ b/Documentation/s390/s390dbf.txt
@@ -115,6 +115,27 @@ Return Value: Handle for generated debug area
 Description:  Allocates memory for a debug log     
               Must not be called within an interrupt handler 
 
+----------------------------------------------------------------------------
+debug_info_t *debug_register_mode(char *name, int pages, int nr_areas,
+				  int buf_size, mode_t mode, uid_t uid,
+				  gid_t gid);
+
+Parameter:    name:	   Name of debug log (e.g. used for debugfs entry)
+	      pages:	   Number of pages, which will be allocated per area
+	      nr_areas:    Number of debug areas
+	      buf_size:    Size of data area in each debug entry
+	      mode:	   File mode for debugfs files. E.g. S_IRWXUGO
+	      uid:	   User ID for debugfs files. Currently only 0 is
+			   supported.
+	      gid:	   Group ID for debugfs files. Currently only 0 is
+			   supported.
+
+Return Value: Handle for generated debug area
+	      NULL if register failed
+
+Description:  Allocates memory for a debug log
+	      Must not be called within an interrupt handler
+
 ---------------------------------------------------------------------------
 void debug_unregister (debug_info_t * id);
 
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 1b2f5ce45320..95a46bc008b7 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -73,7 +73,7 @@ static ssize_t debug_input(struct file *file, const char __user *user_buf,
 static int debug_open(struct inode *inode, struct file *file);
 static int debug_close(struct inode *inode, struct file *file);
 static debug_info_t*  debug_info_create(char *name, int pages_per_area,
-			int nr_areas, int buf_size);
+			int nr_areas, int buf_size, mode_t mode);
 static void debug_info_get(debug_info_t *);
 static void debug_info_put(debug_info_t *);
 static int debug_prolog_level_fn(debug_info_t * id,
@@ -327,7 +327,8 @@ debug_info_free(debug_info_t* db_info){
  */
 
 static debug_info_t*
-debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size)
+debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size,
+		  mode_t mode)
 {
 	debug_info_t* rc;
 
@@ -336,6 +337,8 @@ debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size)
         if(!rc) 
 		goto out;
 
+	rc->mode = mode & ~S_IFMT;
+
 	/* create root directory */
         rc->debugfs_root_entry = debugfs_create_dir(rc->name,
 					debug_debugfs_root_entry);
@@ -676,23 +679,30 @@ debug_close(struct inode *inode, struct file *file)
 }
 
 /*
- * debug_register:
- * - creates and initializes debug area for the caller
- * - returns handle for debug area
+ * debug_register_mode:
+ * - Creates and initializes debug area for the caller
+ *   The mode parameter allows to specify access rights for the s390dbf files
+ * - Returns handle for debug area
  */
 
-debug_info_t*
-debug_register (char *name, int pages_per_area, int nr_areas, int buf_size)
+debug_info_t *debug_register_mode(char *name, int pages_per_area, int nr_areas,
+				  int buf_size, mode_t mode, uid_t uid,
+				  gid_t gid)
 {
 	debug_info_t *rc = NULL;
 
+	/* Since debugfs currently does not support uid/gid other than root, */
+	/* we do not allow gid/uid != 0 until we get support for that. */
+	if ((uid != 0) || (gid != 0))
+		printk(KERN_WARNING "debug: Warning - Currently only uid/gid "
+		       "= 0 are supported. Using root as owner now!");
 	if (!initialized)
 		BUG();
 	mutex_lock(&debug_mutex);
 
         /* create new debug_info */
 
-	rc = debug_info_create(name, pages_per_area, nr_areas, buf_size);
+	rc = debug_info_create(name, pages_per_area, nr_areas, buf_size, mode);
 	if(!rc) 
 		goto out;
 	debug_register_view(rc, &debug_level_view);
@@ -705,6 +715,20 @@ out:
 	mutex_unlock(&debug_mutex);
 	return rc;
 }
+EXPORT_SYMBOL(debug_register_mode);
+
+/*
+ * debug_register:
+ * - creates and initializes debug area for the caller
+ * - returns handle for debug area
+ */
+
+debug_info_t *debug_register(char *name, int pages_per_area, int nr_areas,
+			     int buf_size)
+{
+	return debug_register_mode(name, pages_per_area, nr_areas, buf_size,
+				   S_IRUSR | S_IWUSR, 0, 0);
+}
 
 /*
  * debug_unregister:
@@ -1073,15 +1097,16 @@ debug_register_view(debug_info_t * id, struct debug_view *view)
 	int rc = 0;
 	int i;
 	unsigned long flags;
-	mode_t mode = S_IFREG;
+	mode_t mode;
 	struct dentry *pde;
 
 	if (!id)
 		goto out;
-	if (view->prolog_proc || view->format_proc || view->header_proc)
-		mode |= S_IRUSR;
-	if (view->input_proc)
-		mode |= S_IWUSR;
+	mode = (id->mode | S_IFREG) & ~S_IXUGO;
+	if (!(view->prolog_proc || view->format_proc || view->header_proc))
+		mode &= ~(S_IRUSR | S_IRGRP | S_IROTH);
+	if (!view->input_proc)
+		mode &= ~(S_IWUSR | S_IWGRP | S_IWOTH);
 	pde = debugfs_create_file(view->name, mode, id->debugfs_root_entry,
 				id , &debug_file_ops);
 	if (!pde){
diff --git a/include/asm-s390/debug.h b/include/asm-s390/debug.h
index c00dd2b3dc50..335baf4fc64f 100644
--- a/include/asm-s390/debug.h
+++ b/include/asm-s390/debug.h
@@ -73,6 +73,7 @@ typedef struct debug_info {
 	struct dentry* debugfs_entries[DEBUG_MAX_VIEWS];
 	struct debug_view* views[DEBUG_MAX_VIEWS];	
 	char name[DEBUG_MAX_NAME_LEN];
+	mode_t mode;
 } debug_info_t;
 
 typedef int (debug_header_proc_t) (debug_info_t* id,
@@ -122,6 +123,10 @@ debug_entry_t* debug_exception_common(debug_info_t* id, int level,
 debug_info_t* debug_register(char* name, int pages, int nr_areas,
                              int buf_size);
 
+debug_info_t *debug_register_mode(char *name, int pages, int nr_areas,
+				  int buf_size, mode_t mode, uid_t uid,
+				  gid_t gid);
+
 void debug_unregister(debug_info_t* id);
 
 void debug_set_level(debug_info_t* id, int new_level);
-- 
cgit v1.2.3


From 43ca5c3a1cefdaa09231d64485b8f676118bf1e0 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:23 +0200
Subject: [S390] Convert monitor calls to function calls.

Remove the program check generating monitor calls and use function
calls instead. Theres is no real advantage in using monitor calls,
but they do make debugging harder, because of all the program checks
it generates.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/kernel/process.c  | 70 +++++++++++++++++++--------------------------
 arch/s390/kernel/s390_ext.c |  4 +--
 arch/s390/kernel/smp.c      |  1 -
 arch/s390/kernel/traps.c    |  2 --
 drivers/s390/cio/cio.c      |  3 +-
 include/asm-s390/cpu.h      |  8 ++++++
 6 files changed, 42 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index ce203154d8ce..eb768ce88672 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -76,6 +76,7 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
  * Need to know about CPUs going idle?
  */
 static ATOMIC_NOTIFIER_HEAD(idle_chain);
+DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
 
 int register_idle_notifier(struct notifier_block *nb)
 {
@@ -89,9 +90,33 @@ int unregister_idle_notifier(struct notifier_block *nb)
 }
 EXPORT_SYMBOL(unregister_idle_notifier);
 
-void do_monitor_call(struct pt_regs *regs, long interruption_code)
+static int s390_idle_enter(void)
+{
+	struct s390_idle_data *idle;
+	int nr_calls = 0;
+	void *hcpu;
+	int rc;
+
+	hcpu = (void *)(long)smp_processor_id();
+	rc = __atomic_notifier_call_chain(&idle_chain, S390_CPU_IDLE, hcpu, -1,
+					  &nr_calls);
+	if (rc == NOTIFY_BAD) {
+		nr_calls--;
+		__atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE,
+					     hcpu, nr_calls, NULL);
+		return rc;
+	}
+	idle = &__get_cpu_var(s390_idle);
+	spin_lock(&idle->lock);
+	idle->idle_count++;
+	idle->in_idle = 1;
+	idle->idle_enter = get_clock();
+	spin_unlock(&idle->lock);
+	return NOTIFY_OK;
+}
+
+void s390_idle_leave(void)
 {
-#ifdef CONFIG_SMP
 	struct s390_idle_data *idle;
 
 	idle = &__get_cpu_var(s390_idle);
@@ -99,10 +124,6 @@ void do_monitor_call(struct pt_regs *regs, long interruption_code)
 	idle->idle_time += get_clock() - idle->idle_enter;
 	idle->in_idle = 0;
 	spin_unlock(&idle->lock);
-#endif
-	/* disable monitor call class 0 */
-	__ctl_clear_bit(8, 15);
-
 	atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE,
 				   (void *)(long) smp_processor_id());
 }
@@ -113,61 +134,30 @@ extern void s390_handle_mcck(void);
  */
 static void default_idle(void)
 {
-	int cpu, rc;
-	int nr_calls = 0;
-	void *hcpu;
-#ifdef CONFIG_SMP
-	struct s390_idle_data *idle;
-#endif
-
 	/* CPU is going idle. */
-	cpu = smp_processor_id();
-	hcpu = (void *)(long)cpu;
 	local_irq_disable();
 	if (need_resched()) {
 		local_irq_enable();
 		return;
 	}
-
-	rc = __atomic_notifier_call_chain(&idle_chain, S390_CPU_IDLE, hcpu, -1,
-					  &nr_calls);
-	if (rc == NOTIFY_BAD) {
-		nr_calls--;
-		__atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE,
-					     hcpu, nr_calls, NULL);
+	if (s390_idle_enter() == NOTIFY_BAD) {
 		local_irq_enable();
 		return;
 	}
-
-	/* enable monitor call class 0 */
-	__ctl_set_bit(8, 15);
-
 #ifdef CONFIG_HOTPLUG_CPU
-	if (cpu_is_offline(cpu)) {
+	if (cpu_is_offline(smp_processor_id())) {
 		preempt_enable_no_resched();
 		cpu_die();
 	}
 #endif
-
 	local_mcck_disable();
 	if (test_thread_flag(TIF_MCCK_PENDING)) {
 		local_mcck_enable();
-		/* disable monitor call class 0 */
-		__ctl_clear_bit(8, 15);
-		atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE,
-					   hcpu);
+		s390_idle_leave();
 		local_irq_enable();
 		s390_handle_mcck();
 		return;
 	}
-#ifdef CONFIG_SMP
-	idle = &__get_cpu_var(s390_idle);
-	spin_lock(&idle->lock);
-	idle->idle_count++;
-	idle->in_idle = 1;
-	idle->idle_enter = get_clock();
-	spin_unlock(&idle->lock);
-#endif
 	trace_hardirqs_on();
 	/* Wait for external, I/O or machine check interrupt. */
 	__load_psw_mask(psw_kernel_bits | PSW_MASK_WAIT |
diff --git a/arch/s390/kernel/s390_ext.c b/arch/s390/kernel/s390_ext.c
index acf93dba7727..3a8772d3baea 100644
--- a/arch/s390/kernel/s390_ext.c
+++ b/arch/s390/kernel/s390_ext.c
@@ -13,7 +13,7 @@
 #include <linux/errno.h>
 #include <linux/kernel_stat.h>
 #include <linux/interrupt.h>
-
+#include <asm/cpu.h>
 #include <asm/lowcore.h>
 #include <asm/s390_ext.h>
 #include <asm/irq_regs.h>
@@ -119,7 +119,7 @@ void do_extint(struct pt_regs *regs, unsigned short code)
 
 	old_regs = set_irq_regs(regs);
 	irq_enter();
-	asm volatile ("mc 0,0");
+	s390_idle_check();
 	if (S390_lowcore.int_clock >= S390_lowcore.jiffy_timer)
 		/**
 		 * Make sure that the i/o interrupt did not "overtake"
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index d1e8e8a3fb66..5a445b1b1217 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -73,7 +73,6 @@ static int smp_cpu_state[NR_CPUS];
 static int cpu_management;
 
 static DEFINE_PER_CPU(struct cpu, cpu_devices);
-DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
 
 static void smp_ext_bitcall(int, ec_bit_sig);
 
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 60f728aeaf12..9452a205629b 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -59,7 +59,6 @@ int sysctl_userprocess_debug = 0;
 
 extern pgm_check_handler_t do_protection_exception;
 extern pgm_check_handler_t do_dat_exception;
-extern pgm_check_handler_t do_monitor_call;
 extern pgm_check_handler_t do_asce_exception;
 
 #define stack_pointer ({ void **sp; asm("la %0,0(15)" : "=&d" (sp)); sp; })
@@ -739,6 +738,5 @@ void __init trap_init(void)
         pgm_check_table[0x15] = &operand_exception;
         pgm_check_table[0x1C] = &space_switch_exception;
         pgm_check_table[0x1D] = &hfp_sqrt_exception;
-	pgm_check_table[0x40] = &do_monitor_call;
 	pfault_irq_init();
 }
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 60590a12d529..6dbe9488d3f9 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -24,6 +24,7 @@
 #include <asm/ipl.h>
 #include <asm/chpid.h>
 #include <asm/airq.h>
+#include <asm/cpu.h>
 #include "cio.h"
 #include "css.h"
 #include "chsc.h"
@@ -649,7 +650,7 @@ do_IRQ (struct pt_regs *regs)
 
 	old_regs = set_irq_regs(regs);
 	irq_enter();
-	asm volatile ("mc 0,0");
+	s390_idle_check();
 	if (S390_lowcore.int_clock >= S390_lowcore.jiffy_timer)
 		/**
 		 * Make sure that the i/o interrupt did not "overtake"
diff --git a/include/asm-s390/cpu.h b/include/asm-s390/cpu.h
index 352dde194f3c..e5a6a9ba3adf 100644
--- a/include/asm-s390/cpu.h
+++ b/include/asm-s390/cpu.h
@@ -22,4 +22,12 @@ struct s390_idle_data {
 
 DECLARE_PER_CPU(struct s390_idle_data, s390_idle);
 
+void s390_idle_leave(void);
+
+static inline void s390_idle_check(void)
+{
+	if ((&__get_cpu_var(s390_idle))->in_idle)
+		s390_idle_leave();
+}
+
 #endif /* _ASM_S390_CPU_H_ */
-- 
cgit v1.2.3


From d7b906897e9caae452947e33674df0a2d6f7e10f Mon Sep 17 00:00:00 2001
From: Russell King <rmk+lkml@arm.linux.org.uk>
Date: Thu, 17 Apr 2008 07:46:24 +0200
Subject: [S390] genirq/clockevents: move irq affinity prototypes/inlines to
 interrupt.h

> Generic code is not supposed to include irq.h. Replace this include
> by linux/hardirq.h instead and add/replace an include of linux/irq.h
> in asm header files where necessary.
> This change should only matter for architectures that make use of
> GENERIC_CLOCKEVENTS.
> Architectures in question are mips, x86, arm, sh, powerpc, uml and sparc64.
>
> I did some cross compile tests for mips, x86_64, arm, powerpc and sparc64.
> This patch fixes also build breakages caused by the include replacement in
> tick-common.h.

I generally dislike adding optional linux/* includes in asm/* includes -
I'm nervous about this causing include loops.

However, there's a separate point to be discussed here.

That is, what interfaces are expected of every architecture in the kernel.
If generic code wants to be able to set the affinity of interrupts, then
that needs to become part of the interfaces listed in linux/interrupt.h
rather than linux/irq.h.

So what I suggest is this approach instead (against Linus' tree of a
couple of days ago) - we move irq_set_affinity() and irq_can_set_affinity()
to linux/interrupt.h, change the linux/irq.h includes to linux/interrupt.h
and include asm/irq_regs.h where needed (asm/irq_regs.h is supposed to be
rarely used include since not much touches the stacked parent context
registers.)

Build tested on ARM PXA family kernels and ARM's Realview platform
kernels which both use genirq.

[ tglx@linutronix.de: add GENERIC_HARDIRQ dependencies ]

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 include/linux/interrupt.h    | 19 +++++++++++++++++++
 include/linux/irq.h          | 10 ----------
 kernel/time/tick-broadcast.c |  2 +-
 kernel/time/tick-common.c    |  4 +++-
 kernel/time/tick-oneshot.c   |  2 +-
 5 files changed, 24 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index f8ab4ce70564..b5fef13148bd 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -102,6 +102,25 @@ extern void disable_irq_nosync(unsigned int irq);
 extern void disable_irq(unsigned int irq);
 extern void enable_irq(unsigned int irq);
 
+#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
+
+extern int irq_set_affinity(unsigned int irq, cpumask_t cpumask);
+extern int irq_can_set_affinity(unsigned int irq);
+
+#else /* CONFIG_SMP */
+
+static inline int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
+{
+	return -EINVAL;
+}
+
+static inline int irq_can_set_affinity(unsigned int irq)
+{
+	return 0;
+}
+
+#endif /* CONFIG_SMP && CONFIG_GENERIC_HARDIRQS */
+
 #ifdef CONFIG_GENERIC_HARDIRQS
 /*
  * Special lockdep variants of irq disabling/enabling.
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 176e5e790a44..1883a85625dd 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -228,21 +228,11 @@ static inline void set_pending_irq(unsigned int irq, cpumask_t mask)
 
 #endif /* CONFIG_GENERIC_PENDING_IRQ */
 
-extern int irq_set_affinity(unsigned int irq, cpumask_t cpumask);
-extern int irq_can_set_affinity(unsigned int irq);
-
 #else /* CONFIG_SMP */
 
 #define move_native_irq(x)
 #define move_masked_irq(x)
 
-static inline int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
-{
-	return -EINVAL;
-}
-
-static inline int irq_can_set_affinity(unsigned int irq) { return 0; }
-
 #endif /* CONFIG_SMP */
 
 #ifdef CONFIG_IRQBALANCE
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index e1bd50cbbf5d..fdfa0c745bb6 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -14,7 +14,7 @@
 #include <linux/cpu.h>
 #include <linux/err.h>
 #include <linux/hrtimer.h>
-#include <linux/irq.h>
+#include <linux/interrupt.h>
 #include <linux/percpu.h>
 #include <linux/profile.h>
 #include <linux/sched.h>
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 1bea399a9ef0..4f3886562b8c 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -14,12 +14,14 @@
 #include <linux/cpu.h>
 #include <linux/err.h>
 #include <linux/hrtimer.h>
-#include <linux/irq.h>
+#include <linux/interrupt.h>
 #include <linux/percpu.h>
 #include <linux/profile.h>
 #include <linux/sched.h>
 #include <linux/tick.h>
 
+#include <asm/irq_regs.h>
+
 #include "tick-internal.h"
 
 /*
diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c
index 0258d3115d54..450c04935b66 100644
--- a/kernel/time/tick-oneshot.c
+++ b/kernel/time/tick-oneshot.c
@@ -14,7 +14,7 @@
 #include <linux/cpu.h>
 #include <linux/err.h>
 #include <linux/hrtimer.h>
-#include <linux/irq.h>
+#include <linux/interrupt.h>
 #include <linux/percpu.h>
 #include <linux/profile.h>
 #include <linux/sched.h>
-- 
cgit v1.2.3


From 5a62b192196af9a798e2f2f4c6a1324e7edf2f4b Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:25 +0200
Subject: [S390] Convert s390 to GENERIC_CLOCKEVENTS.

This way we get rid of s390's NO_IDLE_HZ and use the generic dynticks
variant instead. In addition we get high resolution timers for free.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/Kconfig           |  24 +----
 arch/s390/kernel/process.c  |   4 +-
 arch/s390/kernel/s390_ext.c |   9 +-
 arch/s390/kernel/setup.c    |   2 +-
 arch/s390/kernel/time.c     | 256 +++++++++++++-------------------------------
 arch/s390/lib/delay.c       |  14 +--
 drivers/s390/cio/cio.c      |   9 +-
 include/asm-s390/hardirq.h  |   2 +-
 include/asm-s390/lowcore.h  |   6 +-
 9 files changed, 98 insertions(+), 228 deletions(-)

(limited to 'include')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index da6ea64cc34d..f6a68e178fc5 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -43,6 +43,9 @@ config GENERIC_HWEIGHT
 config GENERIC_TIME
 	def_bool y
 
+config GENERIC_CLOCKEVENTS
+	def_bool y
+
 config GENERIC_BUG
 	bool
 	depends on BUG
@@ -73,6 +76,8 @@ menu "Base setup"
 
 comment "Processor type and features"
 
+source "kernel/time/Kconfig"
+
 config 64BIT
 	bool "64 bit kernel"
 	help
@@ -487,25 +492,6 @@ config APPLDATA_NET_SUM
 
 source kernel/Kconfig.hz
 
-config NO_IDLE_HZ
-	bool "No HZ timer ticks in idle"
-	help
-	  Switches the regular HZ timer off when the system is going idle.
-	  This helps z/VM to detect that the Linux system is idle. VM can
-	  then "swap-out" this guest which reduces memory usage. It also
-	  reduces the overhead of idle systems.
-
-	  The HZ timer can be switched on/off via /proc/sys/kernel/hz_timer.
-	  hz_timer=0 means HZ timer is disabled. hz_timer=1 means HZ
-	  timer is active.
-
-config NO_IDLE_HZ_INIT
-	bool "HZ timer in idle off by default"
-	depends on NO_IDLE_HZ
-	help
-	  The HZ timer is switched off in idle by default. That means the
-	  HZ timer is already disabled at boot time.
-
 config S390_HYPFS_FS
 	bool "s390 hypervisor file system support"
 	select SYS_HYPERVISOR
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index eb768ce88672..df033249f6b1 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -36,6 +36,7 @@
 #include <linux/module.h>
 #include <linux/notifier.h>
 #include <linux/utsname.h>
+#include <linux/tick.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/system.h>
@@ -167,9 +168,10 @@ static void default_idle(void)
 void cpu_idle(void)
 {
 	for (;;) {
+		tick_nohz_stop_sched_tick();
 		while (!need_resched())
 			default_idle();
-
+		tick_nohz_restart_sched_tick();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
diff --git a/arch/s390/kernel/s390_ext.c b/arch/s390/kernel/s390_ext.c
index 3a8772d3baea..947d8c74403b 100644
--- a/arch/s390/kernel/s390_ext.c
+++ b/arch/s390/kernel/s390_ext.c
@@ -120,12 +120,9 @@ void do_extint(struct pt_regs *regs, unsigned short code)
 	old_regs = set_irq_regs(regs);
 	irq_enter();
 	s390_idle_check();
-	if (S390_lowcore.int_clock >= S390_lowcore.jiffy_timer)
-		/**
-		 * Make sure that the i/o interrupt did not "overtake"
-		 * the last HZ timer interrupt.
-		 */
-		account_ticks(S390_lowcore.int_clock);
+	if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator)
+		/* Serve timer interrupts first. */
+		clock_comparator_work();
 	kstat_cpu(smp_processor_id()).irqs[EXTERNAL_INTERRUPT]++;
         index = ext_hash(code);
 	for (p = ext_int_hash[index]; p; p = p->next) {
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 22040d087d8a..7141147e6b63 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -428,7 +428,7 @@ setup_lowcore(void)
 	lc->io_new_psw.mask = psw_kernel_bits;
 	lc->io_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) io_int_handler;
 	lc->ipl_device = S390_lowcore.ipl_device;
-	lc->jiffy_timer = -1LL;
+	lc->clock_comparator = -1ULL;
 	lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE;
 	lc->async_stack = (unsigned long)
 		__alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE;
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 925f9dc0b0a0..17c4de9e1b6b 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -30,7 +30,7 @@
 #include <linux/timex.h>
 #include <linux/notifier.h>
 #include <linux/clocksource.h>
-
+#include <linux/clockchips.h>
 #include <asm/uaccess.h>
 #include <asm/delay.h>
 #include <asm/s390_ext.h>
@@ -57,9 +57,9 @@
 
 static ext_int_info_t ext_int_info_cc;
 static ext_int_info_t ext_int_etr_cc;
-static u64 init_timer_cc;
 static u64 jiffies_timer_cc;
-static u64 xtime_cc;
+
+static DEFINE_PER_CPU(struct clock_event_device, comparators);
 
 /*
  * Scheduler clock - returns current time in nanosec units.
@@ -95,162 +95,40 @@ void tod_to_timeval(__u64 todval, struct timespec *xtime)
 #define s390_do_profile()	do { ; } while(0)
 #endif /* CONFIG_PROFILING */
 
-/*
- * Advance the per cpu tick counter up to the time given with the
- * "time" argument. The per cpu update consists of accounting
- * the virtual cpu time, calling update_process_times and calling
- * the profiling hook. If xtime is before time it is advanced as well.
- */
-void account_ticks(u64 time)
+void clock_comparator_work(void)
 {
-	__u32 ticks;
-	__u64 tmp;
-
-	/* Calculate how many ticks have passed. */
-	if (time < S390_lowcore.jiffy_timer)
-		return;
-	tmp = time - S390_lowcore.jiffy_timer;
-	if (tmp >= 2*CLK_TICKS_PER_JIFFY) {  /* more than two ticks ? */
-		ticks = __div(tmp, CLK_TICKS_PER_JIFFY) + 1;
-		S390_lowcore.jiffy_timer +=
-			CLK_TICKS_PER_JIFFY * (__u64) ticks;
-	} else if (tmp >= CLK_TICKS_PER_JIFFY) {
-		ticks = 2;
-		S390_lowcore.jiffy_timer += 2*CLK_TICKS_PER_JIFFY;
-	} else {
-		ticks = 1;
-		S390_lowcore.jiffy_timer += CLK_TICKS_PER_JIFFY;
-	}
-
-#ifdef CONFIG_SMP
-	/*
-	 * Do not rely on the boot cpu to do the calls to do_timer.
-	 * Spread it over all cpus instead.
-	 */
-	write_seqlock(&xtime_lock);
-	if (S390_lowcore.jiffy_timer > xtime_cc) {
-		__u32 xticks;
-		tmp = S390_lowcore.jiffy_timer - xtime_cc;
-		if (tmp >= 2*CLK_TICKS_PER_JIFFY) {
-			xticks = __div(tmp, CLK_TICKS_PER_JIFFY);
-			xtime_cc += (__u64) xticks * CLK_TICKS_PER_JIFFY;
-		} else {
-			xticks = 1;
-			xtime_cc += CLK_TICKS_PER_JIFFY;
-		}
-		do_timer(xticks);
-	}
-	write_sequnlock(&xtime_lock);
-#else
-	do_timer(ticks);
-#endif
-
-	while (ticks--)
-		update_process_times(user_mode(get_irq_regs()));
+	struct clock_event_device *cd;
 
+	S390_lowcore.clock_comparator = -1ULL;
+	set_clock_comparator(S390_lowcore.clock_comparator);
+	cd = &__get_cpu_var(comparators);
+	cd->event_handler(cd);
 	s390_do_profile();
 }
 
-#ifdef CONFIG_NO_IDLE_HZ
-
-#ifdef CONFIG_NO_IDLE_HZ_INIT
-int sysctl_hz_timer = 0;
-#else
-int sysctl_hz_timer = 1;
-#endif
-
-/*
- * Stop the HZ tick on the current CPU.
- * Only cpu_idle may call this function.
- */
-static void stop_hz_timer(void)
-{
-	unsigned long flags;
-	unsigned long seq, next;
-	__u64 timer, todval;
-	int cpu = smp_processor_id();
-
-	if (sysctl_hz_timer != 0)
-		return;
-
-	cpu_set(cpu, nohz_cpu_mask);
-
-	/*
-	 * Leave the clock comparator set up for the next timer
-	 * tick if either rcu or a softirq is pending.
-	 */
-	if (rcu_needs_cpu(cpu) || local_softirq_pending()) {
-		cpu_clear(cpu, nohz_cpu_mask);
-		return;
-	}
-
-	/*
-	 * This cpu is going really idle. Set up the clock comparator
-	 * for the next event.
-	 */
-	next = next_timer_interrupt();
-	do {
-		seq = read_seqbegin_irqsave(&xtime_lock, flags);
-		timer = ((__u64) next) - ((__u64) jiffies) + jiffies_64;
-	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
-	todval = -1ULL;
-	/* Be careful about overflows. */
-	if (timer < (-1ULL / CLK_TICKS_PER_JIFFY)) {
-		timer = jiffies_timer_cc + timer * CLK_TICKS_PER_JIFFY;
-		if (timer >= jiffies_timer_cc)
-			todval = timer;
-	}
-	set_clock_comparator(todval);
-}
-
 /*
- * Start the HZ tick on the current CPU.
- * Only cpu_idle may call this function.
+ * Fixup the clock comparator.
  */
-static void start_hz_timer(void)
+static void fixup_clock_comparator(unsigned long long delta)
 {
-	if (!cpu_isset(smp_processor_id(), nohz_cpu_mask))
+	/* If nobody is waiting there's nothing to fix. */
+	if (S390_lowcore.clock_comparator == -1ULL)
 		return;
-	account_ticks(get_clock());
-	set_clock_comparator(S390_lowcore.jiffy_timer + CPU_DEVIATION);
-	cpu_clear(smp_processor_id(), nohz_cpu_mask);
-}
-
-static int nohz_idle_notify(struct notifier_block *self,
-			    unsigned long action, void *hcpu)
-{
-	switch (action) {
-	case S390_CPU_IDLE:
-		stop_hz_timer();
-		break;
-	case S390_CPU_NOT_IDLE:
-		start_hz_timer();
-		break;
-	}
-	return NOTIFY_OK;
+	S390_lowcore.clock_comparator += delta;
+	set_clock_comparator(S390_lowcore.clock_comparator);
 }
 
-static struct notifier_block nohz_idle_nb = {
-	.notifier_call = nohz_idle_notify,
-};
-
-static void __init nohz_init(void)
+static int s390_next_event(unsigned long delta,
+			   struct clock_event_device *evt)
 {
-	if (register_idle_notifier(&nohz_idle_nb))
-		panic("Couldn't register idle notifier");
+	S390_lowcore.clock_comparator = get_clock() + delta;
+	set_clock_comparator(S390_lowcore.clock_comparator);
+	return 0;
 }
 
-#endif
-
-/*
- * Set up per cpu jiffy timer and set the clock comparator.
- */
-static void setup_jiffy_timer(void)
+static void s390_set_mode(enum clock_event_mode mode,
+			  struct clock_event_device *evt)
 {
-	/* Set up clock comparator to next jiffy. */
-	S390_lowcore.jiffy_timer =
-		jiffies_timer_cc + (jiffies_64 + 1) * CLK_TICKS_PER_JIFFY;
-	set_clock_comparator(S390_lowcore.jiffy_timer + CPU_DEVIATION);
 }
 
 /*
@@ -259,7 +137,26 @@ static void setup_jiffy_timer(void)
  */
 void init_cpu_timer(void)
 {
-	setup_jiffy_timer();
+	struct clock_event_device *cd;
+	int cpu;
+
+	S390_lowcore.clock_comparator = -1ULL;
+	set_clock_comparator(S390_lowcore.clock_comparator);
+
+	cpu = smp_processor_id();
+	cd = &per_cpu(comparators, cpu);
+	cd->name		= "comparator";
+	cd->features		= CLOCK_EVT_FEAT_ONESHOT;
+	cd->mult		= 16777;
+	cd->shift		= 12;
+	cd->min_delta_ns	= 1;
+	cd->max_delta_ns	= LONG_MAX;
+	cd->rating		= 400;
+	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->set_next_event	= s390_next_event;
+	cd->set_mode		= s390_set_mode;
+
+	clockevents_register_device(cd);
 
 	/* Enable clock comparator timer interrupt. */
 	__ctl_set_bit(0,11);
@@ -270,8 +167,6 @@ void init_cpu_timer(void)
 
 static void clock_comparator_interrupt(__u16 code)
 {
-	/* set clock comparator for next tick */
-	set_clock_comparator(S390_lowcore.jiffy_timer + CPU_DEVIATION);
 }
 
 static void etr_reset(void);
@@ -316,8 +211,9 @@ static struct clocksource clocksource_tod = {
  */
 void __init time_init(void)
 {
+	u64 init_timer_cc;
+
 	init_timer_cc = reset_tod_clock();
-	xtime_cc = init_timer_cc + CLK_TICKS_PER_JIFFY;
 	jiffies_timer_cc = init_timer_cc - jiffies_64 * CLK_TICKS_PER_JIFFY;
 
 	/* set xtime */
@@ -342,10 +238,6 @@ void __init time_init(void)
 	/* Enable TOD clock interrupts on the boot cpu. */
 	init_cpu_timer();
 
-#ifdef CONFIG_NO_IDLE_HZ
-	nohz_init();
-#endif
-
 #ifdef CONFIG_VIRT_TIMER
 	vtime_init();
 #endif
@@ -699,53 +591,49 @@ static int etr_aib_follows(struct etr_aib *a1, struct etr_aib *a2, int p)
 }
 
 /*
- * The time is "clock". xtime is what we think the time is.
+ * The time is "clock". old is what we think the time is.
  * Adjust the value by a multiple of jiffies and add the delta to ntp.
  * "delay" is an approximation how long the synchronization took. If
  * the time correction is positive, then "delay" is subtracted from
  * the time difference and only the remaining part is passed to ntp.
  */
-static void etr_adjust_time(unsigned long long clock, unsigned long long delay)
+static unsigned long long etr_adjust_time(unsigned long long old,
+					  unsigned long long clock,
+					  unsigned long long delay)
 {
 	unsigned long long delta, ticks;
 	struct timex adjust;
 
-	/*
-	 * We don't have to take the xtime lock because the cpu
-	 * executing etr_adjust_time is running disabled in
-	 * tasklet context and all other cpus are looping in
-	 * etr_sync_cpu_start.
-	 */
-	if (clock > xtime_cc) {
+	if (clock > old) {
 		/* It is later than we thought. */
-		delta = ticks = clock - xtime_cc;
+		delta = ticks = clock - old;
 		delta = ticks = (delta < delay) ? 0 : delta - delay;
 		delta -= do_div(ticks, CLK_TICKS_PER_JIFFY);
-		init_timer_cc = init_timer_cc + delta;
-		jiffies_timer_cc = jiffies_timer_cc + delta;
-		xtime_cc = xtime_cc + delta;
 		adjust.offset = ticks * (1000000 / HZ);
 	} else {
 		/* It is earlier than we thought. */
-		delta = ticks = xtime_cc - clock;
+		delta = ticks = old - clock;
 		delta -= do_div(ticks, CLK_TICKS_PER_JIFFY);
-		init_timer_cc = init_timer_cc - delta;
-		jiffies_timer_cc = jiffies_timer_cc - delta;
-		xtime_cc = xtime_cc - delta;
+		delta = -delta;
 		adjust.offset = -ticks * (1000000 / HZ);
 	}
+	jiffies_timer_cc += delta;
 	if (adjust.offset != 0) {
 		printk(KERN_NOTICE "etr: time adjusted by %li micro-seconds\n",
 		       adjust.offset);
 		adjust.modes = ADJ_OFFSET_SINGLESHOT;
 		do_adjtimex(&adjust);
 	}
+	return delta;
 }
 
+static struct {
+	int in_sync;
+	unsigned long long fixup_cc;
+} etr_sync;
+
 static void etr_sync_cpu_start(void *dummy)
 {
-	int *in_sync = dummy;
-
 	etr_enable_sync_clock();
 	/*
 	 * This looks like a busy wait loop but it isn't. etr_sync_cpus
@@ -753,7 +641,7 @@ static void etr_sync_cpu_start(void *dummy)
 	 * __udelay will stop the cpu on an enabled wait psw until the
 	 * TOD is running again.
 	 */
-	while (*in_sync == 0) {
+	while (etr_sync.in_sync == 0) {
 		__udelay(1);
 		/*
 		 * A different cpu changes *in_sync. Therefore use
@@ -761,14 +649,14 @@ static void etr_sync_cpu_start(void *dummy)
 		 */
 		barrier();
 	}
-	if (*in_sync != 1)
+	if (etr_sync.in_sync != 1)
 		/* Didn't work. Clear per-cpu in sync bit again. */
 		etr_disable_sync_clock(NULL);
 	/*
 	 * This round of TOD syncing is done. Set the clock comparator
 	 * to the next tick and let the processor continue.
 	 */
-	setup_jiffy_timer();
+	fixup_clock_comparator(etr_sync.fixup_cc);
 }
 
 static void etr_sync_cpu_end(void *dummy)
@@ -783,8 +671,8 @@ static void etr_sync_cpu_end(void *dummy)
 static int etr_sync_clock(struct etr_aib *aib, int port)
 {
 	struct etr_aib *sync_port;
-	unsigned long long clock, delay;
-	int in_sync, follows;
+	unsigned long long clock, old_clock, delay, delta;
+	int follows;
 	int rc;
 
 	/* Check if the current aib is adjacent to the sync port aib. */
@@ -799,9 +687,9 @@ static int etr_sync_clock(struct etr_aib *aib, int port)
 	 * successfully synced the clock. smp_call_function will
 	 * return after all other cpus are in etr_sync_cpu_start.
 	 */
-	in_sync = 0;
+	memset(&etr_sync, 0, sizeof(etr_sync));
 	preempt_disable();
-	smp_call_function(etr_sync_cpu_start,&in_sync,0,0);
+	smp_call_function(etr_sync_cpu_start, NULL, 0, 0);
 	local_irq_disable();
 	etr_enable_sync_clock();
 
@@ -809,6 +697,7 @@ static int etr_sync_clock(struct etr_aib *aib, int port)
 	__ctl_set_bit(14, 21);
 	__ctl_set_bit(0, 29);
 	clock = ((unsigned long long) (aib->edf2.etv + 1)) << 32;
+	old_clock = get_clock();
 	if (set_clock(clock) == 0) {
 		__udelay(1);	/* Wait for the clock to start. */
 		__ctl_clear_bit(0, 29);
@@ -817,16 +706,17 @@ static int etr_sync_clock(struct etr_aib *aib, int port)
 		/* Adjust Linux timing variables. */
 		delay = (unsigned long long)
 			(aib->edf2.etv - sync_port->edf2.etv) << 32;
-		etr_adjust_time(clock, delay);
-		setup_jiffy_timer();
+		delta = etr_adjust_time(old_clock, clock, delay);
+		etr_sync.fixup_cc = delta;
+		fixup_clock_comparator(delta);
 		/* Verify that the clock is properly set. */
 		if (!etr_aib_follows(sync_port, aib, port)) {
 			/* Didn't work. */
 			etr_disable_sync_clock(NULL);
-			in_sync = -EAGAIN;
+			etr_sync.in_sync = -EAGAIN;
 			rc = -EAGAIN;
 		} else {
-			in_sync = 1;
+			etr_sync.in_sync = 1;
 			rc = 0;
 		}
 	} else {
@@ -834,7 +724,7 @@ static int etr_sync_clock(struct etr_aib *aib, int port)
 		__ctl_clear_bit(0, 29);
 		__ctl_clear_bit(14, 21);
 		etr_disable_sync_clock(NULL);
-		in_sync = -EAGAIN;
+		etr_sync.in_sync = -EAGAIN;
 		rc = -EAGAIN;
 	}
 	local_irq_enable();
diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c
index 70f2a862b670..eae21a8ac72d 100644
--- a/arch/s390/lib/delay.c
+++ b/arch/s390/lib/delay.c
@@ -34,7 +34,7 @@ void __delay(unsigned long loops)
  */
 void __udelay(unsigned long usecs)
 {
-	u64 end, time, jiffy_timer = 0;
+	u64 end, time, old_cc = 0;
 	unsigned long flags, cr0, mask, dummy;
 	int irq_context;
 
@@ -43,8 +43,8 @@ void __udelay(unsigned long usecs)
 		local_bh_disable();
 	local_irq_save(flags);
 	if (raw_irqs_disabled_flags(flags)) {
-		jiffy_timer = S390_lowcore.jiffy_timer;
-		S390_lowcore.jiffy_timer = -1ULL - (4096 << 12);
+		old_cc = S390_lowcore.clock_comparator;
+		S390_lowcore.clock_comparator = -1ULL;
 		__ctl_store(cr0, 0, 0);
 		dummy = (cr0 & 0xffff00e0) | 0x00000800;
 		__ctl_load(dummy , 0, 0);
@@ -55,8 +55,8 @@ void __udelay(unsigned long usecs)
 
 	end = get_clock() + ((u64) usecs << 12);
 	do {
-		time = end < S390_lowcore.jiffy_timer ?
-			end : S390_lowcore.jiffy_timer;
+		time = end < S390_lowcore.clock_comparator ?
+			end : S390_lowcore.clock_comparator;
 		set_clock_comparator(time);
 		trace_hardirqs_on();
 		__load_psw_mask(mask);
@@ -65,10 +65,10 @@ void __udelay(unsigned long usecs)
 
 	if (raw_irqs_disabled_flags(flags)) {
 		__ctl_load(cr0, 0, 0);
-		S390_lowcore.jiffy_timer = jiffy_timer;
+		S390_lowcore.clock_comparator = old_cc;
 	}
 	if (!irq_context)
 		_local_bh_enable();
-	set_clock_comparator(S390_lowcore.jiffy_timer);
+	set_clock_comparator(S390_lowcore.clock_comparator);
 	local_irq_restore(flags);
 }
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 6dbe9488d3f9..41db3cc653f5 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -651,12 +651,9 @@ do_IRQ (struct pt_regs *regs)
 	old_regs = set_irq_regs(regs);
 	irq_enter();
 	s390_idle_check();
-	if (S390_lowcore.int_clock >= S390_lowcore.jiffy_timer)
-		/**
-		 * Make sure that the i/o interrupt did not "overtake"
-		 * the last HZ timer interrupt.
-		 */
-		account_ticks(S390_lowcore.int_clock);
+	if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator)
+		/* Serve timer interrupts first. */
+		clock_comparator_work();
 	/*
 	 * Get interrupt information from lowcore
 	 */
diff --git a/include/asm-s390/hardirq.h b/include/asm-s390/hardirq.h
index 31beb18cb3d1..4b7cb964ff35 100644
--- a/include/asm-s390/hardirq.h
+++ b/include/asm-s390/hardirq.h
@@ -32,6 +32,6 @@ typedef struct {
 
 #define HARDIRQ_BITS	8
 
-extern void account_ticks(u64 time);
+void clock_comparator_work(void);
 
 #endif /* __ASM_HARDIRQ_H */
diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h
index 801a6fd35b5b..6baa51b8683b 100644
--- a/include/asm-s390/lowcore.h
+++ b/include/asm-s390/lowcore.h
@@ -80,7 +80,6 @@
 #define __LC_CPUID                      0xC60
 #define __LC_CPUADDR                    0xC68
 #define __LC_IPLDEV                     0xC7C
-#define __LC_JIFFY_TIMER		0xC80
 #define __LC_CURRENT			0xC90
 #define __LC_INT_CLOCK			0xC98
 #else /* __s390x__ */
@@ -103,7 +102,6 @@
 #define __LC_CPUID			0xD80
 #define __LC_CPUADDR			0xD88
 #define __LC_IPLDEV                     0xDB8
-#define __LC_JIFFY_TIMER		0xDC0
 #define __LC_CURRENT			0xDD8
 #define __LC_INT_CLOCK			0xDE8
 #endif /* __s390x__ */
@@ -276,7 +274,7 @@ struct _lowcore
 	/* entry.S sensitive area end */
 
         /* SMP info area: defined by DJB */
-        __u64        jiffy_timer;              /* 0xc80 */
+	__u64	     clock_comparator;	       /* 0xc80 */
 	__u32        ext_call_fast;            /* 0xc88 */
 	__u32        percpu_offset;            /* 0xc8c */
 	__u32        current_task;	       /* 0xc90 */
@@ -368,7 +366,7 @@ struct _lowcore
 	/* entry.S sensitive area end */
 
         /* SMP info area: defined by DJB */
-        __u64        jiffy_timer;              /* 0xdc0 */
+	__u64	     clock_comparator;	       /* 0xdc0 */
 	__u64        ext_call_fast;            /* 0xdc8 */
 	__u64        percpu_offset;            /* 0xdd0 */
 	__u64        current_task;	       /* 0xdd8 */
-- 
cgit v1.2.3


From a806170e29c5468b1d641a22518243bdf1b8d58b Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:26 +0200
Subject: [S390] Fix a lot of sparse warnings.

Most noteable part of this commit is the new local header file entry.h
which contains all the function declarations of functions that get only
called from asm code or are arch internal. That way we can avoid extern
declarations in C files.
This is more or less the same that was done for sparc64.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/kernel/compat_linux.h  | 73 ++++++++++++++++++++++++++++++++++++++++
 arch/s390/kernel/compat_signal.c |  1 +
 arch/s390/kernel/debug.c         |  2 +-
 arch/s390/kernel/early.c         |  1 +
 arch/s390/kernel/entry.h         | 60 +++++++++++++++++++++++++++++++++
 arch/s390/kernel/ipl.c           |  2 +-
 arch/s390/kernel/kprobes.c       |  2 +-
 arch/s390/kernel/process.c       |  2 ++
 arch/s390/kernel/ptrace.c        |  1 +
 arch/s390/kernel/s390_ext.c      |  1 +
 arch/s390/kernel/signal.c        |  6 +---
 arch/s390/kernel/smp.c           |  3 +-
 arch/s390/kernel/sys_s390.c      |  2 +-
 arch/s390/kernel/time.c          |  1 +
 arch/s390/kernel/traps.c         |  5 +--
 arch/s390/mm/fault.c             | 13 ++-----
 drivers/s390/block/dasd.c        |  5 ++-
 drivers/s390/block/dasd_alias.c  | 49 ++++++++++++---------------
 drivers/s390/cio/cio.c           | 18 ++++++----
 drivers/s390/cio/cio.h           |  1 +
 drivers/s390/cio/device.c        |  1 -
 drivers/s390/cio/device.h        |  1 +
 drivers/s390/s390mach.h          |  4 +++
 include/asm-s390/cio.h           |  4 +++
 include/asm-s390/timex.h         |  1 +
 include/asm-s390/tlbflush.h      |  3 +-
 26 files changed, 198 insertions(+), 64 deletions(-)
 create mode 100644 arch/s390/kernel/entry.h

(limited to 'include')

diff --git a/arch/s390/kernel/compat_linux.h b/arch/s390/kernel/compat_linux.h
index e89f8c0c42a0..20723a062017 100644
--- a/arch/s390/kernel/compat_linux.h
+++ b/arch/s390/kernel/compat_linux.h
@@ -162,4 +162,77 @@ struct ucontext32 {
 	compat_sigset_t		uc_sigmask;	/* mask last for extensibility */
 };
 
+struct __sysctl_args32;
+struct stat64_emu31;
+struct mmap_arg_struct_emu31;
+struct fadvise64_64_args;
+struct old_sigaction32;
+struct old_sigaction32;
+
+long sys32_chown16(const char __user * filename, u16 user, u16 group);
+long sys32_lchown16(const char __user * filename, u16 user, u16 group);
+long sys32_fchown16(unsigned int fd, u16 user, u16 group);
+long sys32_setregid16(u16 rgid, u16 egid);
+long sys32_setgid16(u16 gid);
+long sys32_setreuid16(u16 ruid, u16 euid);
+long sys32_setuid16(u16 uid);
+long sys32_setresuid16(u16 ruid, u16 euid, u16 suid);
+long sys32_getresuid16(u16 __user *ruid, u16 __user *euid, u16 __user *suid);
+long sys32_setresgid16(u16 rgid, u16 egid, u16 sgid);
+long sys32_getresgid16(u16 __user *rgid, u16 __user *egid, u16 __user *sgid);
+long sys32_setfsuid16(u16 uid);
+long sys32_setfsgid16(u16 gid);
+long sys32_getgroups16(int gidsetsize, u16 __user *grouplist);
+long sys32_setgroups16(int gidsetsize, u16 __user *grouplist);
+long sys32_getuid16(void);
+long sys32_geteuid16(void);
+long sys32_getgid16(void);
+long sys32_getegid16(void);
+long sys32_ipc(u32 call, int first, int second, int third, u32 ptr);
+long sys32_truncate64(const char __user * path, unsigned long high,
+		      unsigned long low);
+long sys32_ftruncate64(unsigned int fd, unsigned long high, unsigned long low);
+long sys32_sched_rr_get_interval(compat_pid_t pid,
+				 struct compat_timespec __user *interval);
+long sys32_rt_sigprocmask(int how, compat_sigset_t __user *set,
+			  compat_sigset_t __user *oset, size_t sigsetsize);
+long sys32_rt_sigpending(compat_sigset_t __user *set, size_t sigsetsize);
+long sys32_rt_sigqueueinfo(int pid, int sig, compat_siginfo_t __user *uinfo);
+long sys32_execve(void);
+long sys32_init_module(void __user *umod, unsigned long len,
+		       const char __user *uargs);
+long sys32_delete_module(const char __user *name_user, unsigned int flags);
+long sys32_gettimeofday(struct compat_timeval __user *tv,
+			struct timezone __user *tz);
+long sys32_settimeofday(struct compat_timeval __user *tv,
+			struct timezone __user *tz);
+long sys32_pause(void);
+long sys32_pread64(unsigned int fd, char __user *ubuf, size_t count,
+		   u32 poshi, u32 poslo);
+long sys32_pwrite64(unsigned int fd, const char __user *ubuf,
+		    size_t count, u32 poshi, u32 poslo);
+compat_ssize_t sys32_readahead(int fd, u32 offhi, u32 offlo, s32 count);
+long sys32_sendfile(int out_fd, int in_fd, compat_off_t __user *offset,
+		    size_t count);
+long sys32_sendfile64(int out_fd, int in_fd, compat_loff_t __user *offset,
+		      s32 count);
+long sys32_sysctl(struct __sysctl_args32 __user *args);
+long sys32_stat64(char __user * filename, struct stat64_emu31 __user * statbuf);
+long sys32_lstat64(char __user * filename,
+		   struct stat64_emu31 __user * statbuf);
+long sys32_fstat64(unsigned long fd, struct stat64_emu31 __user * statbuf);
+long sys32_fstatat64(unsigned int dfd, char __user *filename,
+		     struct stat64_emu31 __user* statbuf, int flag);
+unsigned long old32_mmap(struct mmap_arg_struct_emu31 __user *arg);
+long sys32_mmap2(struct mmap_arg_struct_emu31 __user *arg);
+long sys32_read(unsigned int fd, char __user * buf, size_t count);
+long sys32_write(unsigned int fd, char __user * buf, size_t count);
+long sys32_clone(void);
+long sys32_fadvise64(int fd, loff_t offset, size_t len, int advise);
+long sys32_fadvise64_64(struct fadvise64_64_args __user *args);
+long sys32_sigaction(int sig, const struct old_sigaction32 __user *act,
+		     struct old_sigaction32 __user *oact);
+long sys32_rt_sigaction(int sig, const struct sigaction32 __user *act,
+			struct sigaction32 __user *oact, size_t sigsetsize);
+long sys32_sigaltstack(const stack_t32 __user *uss, stack_t32 __user *uoss);
 #endif /* _ASM_S390X_S390_H */
diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c
index ae2f2d313930..c7f02e777af2 100644
--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -29,6 +29,7 @@
 #include <asm/lowcore.h>
 #include "compat_linux.h"
 #include "compat_ptrace.h"
+#include "entry.h"
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 95a46bc008b7..1e7d4ac7068b 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -157,7 +157,7 @@ struct debug_view debug_sprintf_view = {
 };
 
 /* used by dump analysis tools to determine version of debug feature */
-unsigned int debug_feature_version = __DEBUG_FEATURE_VERSION;
+static unsigned int __used debug_feature_version = __DEBUG_FEATURE_VERSION;
 
 /* static globals */
 
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 01832c440636..540a67f979b6 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -21,6 +21,7 @@
 #include <asm/setup.h>
 #include <asm/cpcmd.h>
 #include <asm/sclp.h>
+#include "entry.h"
 
 /*
  * Create a Kernel NSS if the SAVESYS= parameter is defined
diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h
new file mode 100644
index 000000000000..6b1896345eda
--- /dev/null
+++ b/arch/s390/kernel/entry.h
@@ -0,0 +1,60 @@
+#ifndef _ENTRY_H
+#define _ENTRY_H
+
+#include <linux/types.h>
+#include <linux/signal.h>
+#include <asm/ptrace.h>
+
+typedef void pgm_check_handler_t(struct pt_regs *, long);
+extern pgm_check_handler_t *pgm_check_table[128];
+pgm_check_handler_t do_protection_exception;
+pgm_check_handler_t do_dat_exception;
+
+extern int sysctl_userprocess_debug;
+
+void do_single_step(struct pt_regs *regs);
+void syscall_trace(struct pt_regs *regs, int entryexit);
+void kernel_stack_overflow(struct pt_regs * regs);
+void do_signal(struct pt_regs *regs);
+int handle_signal32(unsigned long sig, struct k_sigaction *ka,
+		    siginfo_t *info, sigset_t *oldset, struct pt_regs *regs);
+
+void do_extint(struct pt_regs *regs, unsigned short code);
+int __cpuinit start_secondary(void *cpuvoid);
+void __init startup_init(void);
+void die(const char * str, struct pt_regs * regs, long err);
+
+struct new_utsname;
+struct mmap_arg_struct;
+struct fadvise64_64_args;
+struct old_sigaction;
+struct sel_arg_struct;
+
+long sys_pipe(unsigned long __user *fildes);
+long sys_mmap2(struct mmap_arg_struct __user  *arg);
+long old_mmap(struct mmap_arg_struct __user *arg);
+long sys_ipc(uint call, int first, unsigned long second,
+	     unsigned long third, void __user *ptr);
+long s390x_newuname(struct new_utsname __user *name);
+long s390x_personality(unsigned long personality);
+long s390_fadvise64(int fd, u32 offset_high, u32 offset_low,
+		    size_t len, int advice);
+long s390_fadvise64_64(struct fadvise64_64_args __user *args);
+long s390_fallocate(int fd, int mode, loff_t offset, u32 len_high, u32 len_low);
+long sys_fork(void);
+long sys_clone(void);
+long sys_vfork(void);
+void execve_tail(void);
+long sys_execve(void);
+int sys_sigsuspend(int history0, int history1, old_sigset_t mask);
+long sys_sigaction(int sig, const struct old_sigaction __user *act,
+		   struct old_sigaction __user *oact);
+long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss);
+long sys_sigreturn(void);
+long sys_rt_sigreturn(void);
+long sys32_sigreturn(void);
+long sys32_rt_sigreturn(void);
+long old_select(struct sel_arg_struct __user *arg);
+long sys_ptrace(long request, long pid, long addr, long data);
+
+#endif /* _ENTRY_H */
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 375232c46c7a..532542447d66 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -655,7 +655,7 @@ static struct kobj_attribute reipl_type_attr =
 
 static struct kset *reipl_kset;
 
-void reipl_run(struct shutdown_trigger *trigger)
+static void reipl_run(struct shutdown_trigger *trigger)
 {
 	struct ccw_dev_id devid;
 	static char buf[100];
diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index c5549a206284..ed04d1372d5d 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -360,7 +360,7 @@ no_kprobe:
  *	- When the probed function returns, this probe
  *		causes the handlers to fire
  */
-void kretprobe_trampoline_holder(void)
+static void __used kretprobe_trampoline_holder(void)
 {
 	asm volatile(".global kretprobe_trampoline\n"
 		     "kretprobe_trampoline: bcr 0,0\n");
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index df033249f6b1..dbefd0db395f 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -37,6 +37,7 @@
 #include <linux/notifier.h>
 #include <linux/utsname.h>
 #include <linux/tick.h>
+#include <linux/elfcore.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/system.h>
@@ -45,6 +46,7 @@
 #include <asm/irq.h>
 #include <asm/timer.h>
 #include <asm/cpu.h>
+#include "entry.h"
 
 asmlinkage void ret_from_fork(void) asm ("ret_from_fork");
 
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index 6e036bae9875..58a064296987 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -41,6 +41,7 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
+#include "entry.h"
 
 #ifdef CONFIG_COMPAT
 #include "compat_ptrace.h"
diff --git a/arch/s390/kernel/s390_ext.c b/arch/s390/kernel/s390_ext.c
index 947d8c74403b..e019b419efc6 100644
--- a/arch/s390/kernel/s390_ext.c
+++ b/arch/s390/kernel/s390_ext.c
@@ -18,6 +18,7 @@
 #include <asm/s390_ext.h>
 #include <asm/irq_regs.h>
 #include <asm/irq.h>
+#include "entry.h"
 
 /*
  * ext_int_hash[index] is the start of the list for all external interrupts
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index 8c92191949c2..b97682040215 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -27,6 +27,7 @@
 #include <asm/ucontext.h>
 #include <asm/uaccess.h>
 #include <asm/lowcore.h>
+#include "entry.h"
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
@@ -484,11 +485,6 @@ void do_signal(struct pt_regs *regs)
 		int ret;
 #ifdef CONFIG_COMPAT
 		if (test_thread_flag(TIF_31BIT)) {
-			extern int handle_signal32(unsigned long sig,
-						   struct k_sigaction *ka,
-						   siginfo_t *info,
-						   sigset_t *oldset,
-						   struct pt_regs *regs);
 			ret = handle_signal32(signr, &ka, &info, oldset, regs);
 	        }
 		else
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 5a445b1b1217..0dfa988c1b26 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -44,6 +44,7 @@
 #include <asm/lowcore.h>
 #include <asm/sclp.h>
 #include <asm/cpu.h>
+#include "entry.h"
 
 /*
  * An array with a pointer the lowcore of every CPU.
@@ -297,7 +298,7 @@ static void smp_ext_bitcall(int cpu, ec_bit_sig sig)
 /*
  * this function sends a 'purge tlb' signal to another CPU.
  */
-void smp_ptlb_callback(void *info)
+static void smp_ptlb_callback(void *info)
 {
 	__tlb_flush_local();
 }
diff --git a/arch/s390/kernel/sys_s390.c b/arch/s390/kernel/sys_s390.c
index fefee99f28aa..988d0d64c2c8 100644
--- a/arch/s390/kernel/sys_s390.c
+++ b/arch/s390/kernel/sys_s390.c
@@ -29,8 +29,8 @@
 #include <linux/personality.h>
 #include <linux/unistd.h>
 #include <linux/ipc.h>
-
 #include <asm/uaccess.h>
+#include "entry.h"
 
 /*
  * sys_pipe() is the normal C calling standard for creating
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 17c4de9e1b6b..7aec676fefd5 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -39,6 +39,7 @@
 #include <asm/irq_regs.h>
 #include <asm/timer.h>
 #include <asm/etr.h>
+#include <asm/cio.h>
 
 /* change this if you have some constant time drift */
 #define USECS_PER_JIFFY     ((unsigned long) 1000000/HZ)
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 9452a205629b..b3524134f213 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -42,11 +42,8 @@
 #include <asm/s390_ext.h>
 #include <asm/lowcore.h>
 #include <asm/debug.h>
+#include "entry.h"
 
-/* Called from entry.S only */
-extern void handle_per_exception(struct pt_regs *regs);
-
-typedef void pgm_check_handler_t(struct pt_regs *, long);
 pgm_check_handler_t *pgm_check_table[128];
 
 #ifdef CONFIG_SYSCTL
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index a9c635f01db8..2650f46001d0 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -28,11 +28,11 @@
 #include <linux/hardirq.h>
 #include <linux/kprobes.h>
 #include <linux/uaccess.h>
-
 #include <asm/system.h>
 #include <asm/pgtable.h>
 #include <asm/s390_ext.h>
 #include <asm/mmu_context.h>
+#include "../kernel/entry.h"
 
 #ifndef CONFIG_64BIT
 #define __FAIL_ADDR_MASK 0x7ffff000
@@ -50,8 +50,6 @@
 extern int sysctl_userprocess_debug;
 #endif
 
-extern void die(const char *,struct pt_regs *,long);
-
 #ifdef CONFIG_KPROBES
 static inline int notify_page_fault(struct pt_regs *regs, long err)
 {
@@ -245,11 +243,6 @@ static void do_sigbus(struct pt_regs *regs, unsigned long error_code,
 }
 
 #ifdef CONFIG_S390_EXEC_PROTECT
-extern long sys_sigreturn(void);
-extern long sys_rt_sigreturn(void);
-extern long sys32_sigreturn(void);
-extern long sys32_rt_sigreturn(void);
-
 static int signal_return(struct mm_struct *mm, struct pt_regs *regs,
 			 unsigned long address, unsigned long error_code)
 {
@@ -424,7 +417,7 @@ no_context:
 }
 
 void __kprobes do_protection_exception(struct pt_regs *regs,
-				       unsigned long error_code)
+				       long error_code)
 {
 	/* Protection exception is supressing, decrement psw address. */
 	regs->psw.addr -= (error_code >> 16);
@@ -440,7 +433,7 @@ void __kprobes do_protection_exception(struct pt_regs *regs,
 	do_exception(regs, 4, 1);
 }
 
-void __kprobes do_dat_exception(struct pt_regs *regs, unsigned long error_code)
+void __kprobes do_dat_exception(struct pt_regs *regs, long error_code)
 {
 	do_exception(regs, error_code & 0xff, 0);
 }
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index bb72e0a5b0e0..ac6d4d3218b3 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -2299,9 +2299,8 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
 	 * in the other openers.
 	 */
 	if (device->block) {
-		struct dasd_block *block = device->block;
-		max_count = block->bdev ? 0 : -1;
-		open_count = (int) atomic_read(&block->open_count);
+		max_count = device->block->bdev ? 0 : -1;
+		open_count = atomic_read(&device->block->open_count);
 		if (open_count > max_count) {
 			if (open_count > 0)
 				printk(KERN_WARNING "Can't offline dasd "
diff --git a/drivers/s390/block/dasd_alias.c b/drivers/s390/block/dasd_alias.c
index 3a40bee9d358..2d8df0b30538 100644
--- a/drivers/s390/block/dasd_alias.c
+++ b/drivers/s390/block/dasd_alias.c
@@ -745,6 +745,19 @@ static void flush_all_alias_devices_on_lcu(struct alias_lcu *lcu)
 	spin_unlock_irqrestore(&lcu->lock, flags);
 }
 
+static void __stop_device_on_lcu(struct dasd_device *device,
+				 struct dasd_device *pos)
+{
+	/* If pos == device then device is already locked! */
+	if (pos == device) {
+		pos->stopped |= DASD_STOPPED_SU;
+		return;
+	}
+	spin_lock(get_ccwdev_lock(pos->cdev));
+	pos->stopped |= DASD_STOPPED_SU;
+	spin_unlock(get_ccwdev_lock(pos->cdev));
+}
+
 /*
  * This function is called in interrupt context, so the
  * cdev lock for device is already locked!
@@ -755,35 +768,15 @@ static void _stop_all_devices_on_lcu(struct alias_lcu *lcu,
 	struct alias_pav_group *pavgroup;
 	struct dasd_device *pos;
 
-	list_for_each_entry(pos, &lcu->active_devices, alias_list) {
-		if (pos != device)
-			spin_lock(get_ccwdev_lock(pos->cdev));
-		pos->stopped |= DASD_STOPPED_SU;
-		if (pos != device)
-			spin_unlock(get_ccwdev_lock(pos->cdev));
-	}
-	list_for_each_entry(pos, &lcu->inactive_devices, alias_list) {
-		if (pos != device)
-			spin_lock(get_ccwdev_lock(pos->cdev));
-		pos->stopped |= DASD_STOPPED_SU;
-		if (pos != device)
-			spin_unlock(get_ccwdev_lock(pos->cdev));
-	}
+	list_for_each_entry(pos, &lcu->active_devices, alias_list)
+		__stop_device_on_lcu(device, pos);
+	list_for_each_entry(pos, &lcu->inactive_devices, alias_list)
+		__stop_device_on_lcu(device, pos);
 	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
-		list_for_each_entry(pos, &pavgroup->baselist, alias_list) {
-			if (pos != device)
-				spin_lock(get_ccwdev_lock(pos->cdev));
-			pos->stopped |= DASD_STOPPED_SU;
-			if (pos != device)
-				spin_unlock(get_ccwdev_lock(pos->cdev));
-		}
-		list_for_each_entry(pos, &pavgroup->aliaslist, alias_list) {
-			if (pos != device)
-				spin_lock(get_ccwdev_lock(pos->cdev));
-			pos->stopped |= DASD_STOPPED_SU;
-			if (pos != device)
-				spin_unlock(get_ccwdev_lock(pos->cdev));
-		}
+		list_for_each_entry(pos, &pavgroup->baselist, alias_list)
+			__stop_device_on_lcu(device, pos);
+		list_for_each_entry(pos, &pavgroup->aliaslist, alias_list)
+			__stop_device_on_lcu(device, pos);
 	}
 }
 
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 41db3cc653f5..23ffcc4768a7 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -670,10 +670,14 @@ do_IRQ (struct pt_regs *regs)
 			continue;
 		}
 		sch = (struct subchannel *)(unsigned long)tpi_info->intparm;
-		if (sch)
-			spin_lock(sch->lock);
+		if (!sch) {
+			/* Clear pending interrupt condition. */
+			tsch(tpi_info->schid, irb);
+			continue;
+		}
+		spin_lock(sch->lock);
 		/* Store interrupt response block to lowcore. */
-		if (tsch (tpi_info->schid, irb) == 0 && sch) {
+		if (tsch(tpi_info->schid, irb) == 0) {
 			/* Keep subchannel information word up to date. */
 			memcpy (&sch->schib.scsw, &irb->scsw,
 				sizeof (irb->scsw));
@@ -681,8 +685,7 @@ do_IRQ (struct pt_regs *regs)
 			if (sch->driver && sch->driver->irq)
 				sch->driver->irq(sch);
 		}
-		if (sch)
-			spin_unlock(sch->lock);
+		spin_unlock(sch->lock);
 		/*
 		 * Are more interrupts pending?
 		 * If so, the tpi instruction will update the lowcore
@@ -708,8 +711,9 @@ void *cio_get_console_priv(void)
 /*
  * busy wait for the next interrupt on the console
  */
-void
-wait_cons_dev (void)
+void wait_cons_dev(void)
+	__releases(console_subchannel.lock)
+	__acquires(console_subchannel.lock)
 {
 	unsigned long cr6      __attribute__ ((aligned (8)));
 	unsigned long save_cr6 __attribute__ ((aligned (8)));
diff --git a/drivers/s390/cio/cio.h b/drivers/s390/cio/cio.h
index 52afa4c784de..08f2235c5a6f 100644
--- a/drivers/s390/cio/cio.h
+++ b/drivers/s390/cio/cio.h
@@ -100,6 +100,7 @@ extern int cio_modify (struct subchannel *);
 
 int cio_create_sch_lock(struct subchannel *);
 void do_adapter_IO(void);
+void do_IRQ(struct pt_regs *);
 
 /* Use with care. */
 #ifdef CONFIG_CCW_CONSOLE
diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c
index fec004f62bcf..e0c7adb8958e 100644
--- a/drivers/s390/cio/device.c
+++ b/drivers/s390/cio/device.c
@@ -577,7 +577,6 @@ static DEVICE_ATTR(devtype, 0444, devtype_show, NULL);
 static DEVICE_ATTR(cutype, 0444, cutype_show, NULL);
 static DEVICE_ATTR(modalias, 0444, modalias_show, NULL);
 static DEVICE_ATTR(online, 0644, online_show, online_store);
-extern struct device_attribute dev_attr_cmb_enable;
 static DEVICE_ATTR(availability, 0444, available_show, NULL);
 
 static struct attribute * subch_attrs[] = {
diff --git a/drivers/s390/cio/device.h b/drivers/s390/cio/device.h
index d40a2ffaa000..cb08092be39f 100644
--- a/drivers/s390/cio/device.h
+++ b/drivers/s390/cio/device.h
@@ -127,4 +127,5 @@ extern struct bus_type ccw_bus_type;
 void retry_set_schib(struct ccw_device *cdev);
 void cmf_retry_copy_block(struct ccw_device *);
 int cmf_reenable(struct ccw_device *);
+extern struct device_attribute dev_attr_cmb_enable;
 #endif
diff --git a/drivers/s390/s390mach.h b/drivers/s390/s390mach.h
index d3ca4281a494..ca681f9b67fc 100644
--- a/drivers/s390/s390mach.h
+++ b/drivers/s390/s390mach.h
@@ -105,4 +105,8 @@ static inline int stcrw(struct crw *pcrw )
 #define ED_ETR_SYNC	12	/* External damage ETR sync check */
 #define ED_ETR_SWITCH	13	/* External damage ETR switch to local */
 
+struct pt_regs;
+
+void s390_handle_mcck(void);
+void s390_do_machine_check(struct pt_regs *regs);
 #endif /* __s390mach */
diff --git a/include/asm-s390/cio.h b/include/asm-s390/cio.h
index 123b557c3ff4..0818ecd30ca6 100644
--- a/include/asm-s390/cio.h
+++ b/include/asm-s390/cio.h
@@ -397,6 +397,10 @@ struct cio_iplinfo {
 
 extern int cio_get_iplinfo(struct cio_iplinfo *iplinfo);
 
+/* Function from drivers/s390/cio/chsc.c */
+int chsc_sstpc(void *page, unsigned int op, u16 ctrl);
+int chsc_sstpi(void *page, void *result, size_t size);
+
 #endif
 
 #endif
diff --git a/include/asm-s390/timex.h b/include/asm-s390/timex.h
index 6dd7eecbb8e4..d744c3d62de5 100644
--- a/include/asm-s390/timex.h
+++ b/include/asm-s390/timex.h
@@ -83,5 +83,6 @@ static inline cycles_t get_cycles(void)
 
 int get_sync_clock(unsigned long long *clock);
 void init_cpu_timer(void);
+unsigned long long monotonic_clock(void);
 
 #endif
diff --git a/include/asm-s390/tlbflush.h b/include/asm-s390/tlbflush.h
index de723470c5d4..9e57a93d7de1 100644
--- a/include/asm-s390/tlbflush.h
+++ b/include/asm-s390/tlbflush.h
@@ -17,9 +17,10 @@ static inline void __tlb_flush_local(void)
 /*
  * Flush all tlb entries on all cpus.
  */
+void smp_ptlb_all(void);
+
 static inline void __tlb_flush_global(void)
 {
-	extern void smp_ptlb_all(void);
 	register unsigned long reg2 asm("2");
 	register unsigned long reg3 asm("3");
 	register unsigned long reg4 asm("4");
-- 
cgit v1.2.3


From 1a5debaaace41f1e91014332e6eedde4499e5638 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:29 +0200
Subject: [S390] lowcore: Change type of lowcores softirq_pending to __u32.

As noted by akpm:

> kernel/time/tick-sched.c: In function 'tick_nohz_stop_sched_tick':
> kernel/time/tick-sched.c:229: warning: format '%02x' expects type 'unsigned int', but argument 2 has type '__u64'
>
> I don't think the architecture's local_softirq_pending() should return u64.
> This is the sort of thing which should be consistent across architectures.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 include/asm-s390/lowcore.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h
index 6baa51b8683b..007c8a4941b2 100644
--- a/include/asm-s390/lowcore.h
+++ b/include/asm-s390/lowcore.h
@@ -370,7 +370,8 @@ struct _lowcore
 	__u64        ext_call_fast;            /* 0xdc8 */
 	__u64        percpu_offset;            /* 0xdd0 */
 	__u64        current_task;	       /* 0xdd8 */
-	__u64        softirq_pending;	       /* 0xde0 */
+	__u32	     softirq_pending;	       /* 0xde0 */
+	__u32	     pad_0x0de4;	       /* 0xde4 */
 	__u64        int_clock;                /* 0xde8 */
         __u8         pad12[0xe00-0xdf0];       /* 0xdf0 */
 
-- 
cgit v1.2.3


From 9e74a6b8983c2653dd2a6f51e634efa281e95d59 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:30 +0200
Subject: [S390] kernel: show last breaking-event-address on oops

Newer s390 models have a breaking-event-address-recording register.
Each time an instruction causes a break in the sequential instruction
execution, the address is saved in that hardware register. On a program
interrupt the address is copied to the lowcore address 272-279, which
makes it software accessible.

This patch changes the program check handler and the stack overflow
checker to copy the value into the pt_regs argument.
The oops output is enhanced to show the last known breaking address.
It might give additional information if the stack trace is corrupted.

The feature is only available on 64 bit.

The new oops output looks like:

[---------snip----------]
Modules linked in: vmcp sunrpc qeth_l2 dm_mod qeth ccwgroup
CPU: 2 Not tainted 2.6.24zlive-host #8
Process modprobe (pid: 4788, task: 00000000bf3d8718, ksp: 00000000b2b0b8e0)
Krnl PSW : 0704200180000000 000003e000020028 (vmcp_init+0x28/0xe4 [vmcp])
           R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:0 CC:2 PM:0 EA:3
Krnl GPRS: 0000000004000002 000003e000020000 0000000000000000 0000000000000001
           000000000015734c ffffffffffffffff 000003e0000b3b00 0000000000000000
           000003e00007ca30 00000000b5bb5d40 00000000b5bb5800 000003e0000b3b00
           000003e0000a2000 00000000003ecf50 00000000b2b0bd50 00000000b2b0bcb0
Krnl Code: 000003e000020018: c0c000040ff4       larl    %r12,3e0000a2000
           000003e00002001e: e3e0f0000024       stg     %r14,0(%r15)
           000003e000020024: a7f40001           brc     15,3e000020026
          >000003e000020028: e310c0100004       lg      %r1,16(%r12)
           000003e00002002e: c020000413dc       larl    %r2,3e0000a27e6
           000003e000020034: c0a00004aee6       larl    %r10,3e0000b5e00
           000003e00002003a: a7490001           lghi    %r4,1
           000003e00002003e: a75900f0           lghi    %r5,240
Call Trace:
([<000000000014b300>] blocking_notifier_call_chain+0x2c/0x40)
 [<000000000015735c>] sys_init_module+0x19d8/0x1b08
 [<0000000000110afc>] sysc_noemu+0x10/0x16
 [<000002000011cda2>] 0x2000011cda2
Last Breaking-Event-Address:
 [<000003e000020024>] vmcp_init+0x24/0xe4 [vmcp]
[---------snip----------]

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/kernel/entry64.S   |  2 ++
 arch/s390/kernel/process.c   |  1 +
 arch/s390/kernel/traps.c     | 10 +++++++++-
 include/asm-s390/lowcore.h   |  2 ++
 include/asm-s390/processor.h |  7 +++++++
 5 files changed, 21 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index efde6e178f6c..cd959c0b2e16 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -475,6 +475,7 @@ pgm_check_handler:
 pgm_no_vtime:
 #endif
 	lg	%r9,__LC_THREAD_INFO	# load pointer to thread_info struct
+	mvc	SP_ARGS(8,%r15),__LC_LAST_BREAK
 	TRACE_IRQS_OFF
 	lgf	%r3,__LC_PGM_ILC	# load program interruption code
 	lghi	%r8,0x7f
@@ -847,6 +848,7 @@ stack_overflow:
 	je	0f
 	la	%r1,__LC_SAVE_AREA+32
 0:	mvc	SP_R12(32,%r15),0(%r1)	# move %r12-%r15 to stack
+	mvc	SP_ARGS(8,%r15),__LC_LAST_BREAK
 	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) # clear back chain
 	la	%r2,SP_PTREGS(%r15)	# load pt_regs
 	jg	kernel_stack_overflow
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index dbefd0db395f..c1aff194141d 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -195,6 +195,7 @@ void show_regs(struct pt_regs *regs)
 	/* Show stack backtrace if pt_regs is from kernel mode */
 	if (!(regs->psw.mask & PSW_MASK_PSTATE))
 		show_trace(NULL, (unsigned long *) regs->gprs[15]);
+	show_last_breaking_event(regs);
 }
 
 extern void kernel_thread_starter(void);
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index b3524134f213..57b607b61100 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -134,7 +134,6 @@ void show_trace(struct task_struct *task, unsigned long *stack)
 	else
 		__show_trace(sp, S390_lowcore.thread_info,
 			     S390_lowcore.thread_info + THREAD_SIZE);
-	printk("\n");
 	if (!task)
 		task = current;
 	debug_show_held_locks(task);
@@ -162,6 +161,15 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 	show_trace(task, sp);
 }
 
+#ifdef CONFIG_64BIT
+void show_last_breaking_event(struct pt_regs *regs)
+{
+	printk("Last Breaking-Event-Address:\n");
+	printk(" [<%016lx>] ", regs->args[0] & PSW_ADDR_INSN);
+	print_symbol("%s\n", regs->args[0] & PSW_ADDR_INSN);
+}
+#endif
+
 /*
  * The architecture-independent dump_stack generator
  */
diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h
index 007c8a4941b2..5de3efb31445 100644
--- a/include/asm-s390/lowcore.h
+++ b/include/asm-s390/lowcore.h
@@ -56,6 +56,8 @@
 #define __LC_IO_INT_WORD                0x0C0
 #define __LC_MCCK_CODE                  0x0E8
 
+#define __LC_LAST_BREAK 		0x110
+
 #define __LC_RETURN_PSW                 0x200
 
 #define __LC_SAVE_AREA                  0xC00
diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index 51d88912aa20..8eaf343a12a8 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -175,6 +175,13 @@ extern void task_show_regs(struct seq_file *m, struct task_struct *task);
 extern void show_registers(struct pt_regs *regs);
 extern void show_code(struct pt_regs *regs);
 extern void show_trace(struct task_struct *task, unsigned long *sp);
+#ifdef CONFIG_64BIT
+extern void show_last_breaking_event(struct pt_regs *regs);
+#else
+static inline void show_last_breaking_event(struct pt_regs *regs)
+{
+}
+#endif
 
 unsigned long get_wchan(struct task_struct *p);
 #define task_pt_regs(tsk) ((struct pt_regs *) \
-- 
cgit v1.2.3


From ca68305bf3c76c4a7cd1c77d5423219f39164df8 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Thu, 17 Apr 2008 07:46:31 +0200
Subject: [S390] Remove code duplication from monreader / dcssblk.

Move the function that prints the segment warning messages found in the
monreader driver and the dcssblk driver to the extmem base code.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/mm/extmem.c         | 67 +++++++++++++++++++++++++++++++++----------
 drivers/s390/block/dcssblk.c  | 53 +---------------------------------
 drivers/s390/char/monreader.c | 54 ++--------------------------------
 include/asm-s390/extmem.h     | 11 +++----
 4 files changed, 61 insertions(+), 124 deletions(-)

(limited to 'include')

diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c
index 880b0ebf894b..ed2af0a3303b 100644
--- a/arch/s390/mm/extmem.c
+++ b/arch/s390/mm/extmem.c
@@ -289,22 +289,8 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long
 
 	rc = add_shared_memory(seg->start_addr, seg->end - seg->start_addr + 1);
 
-	switch (rc) {
-	case 0:
-		break;
-	case -ENOSPC:
-		PRINT_WARN("segment_load: not loading segment %s - overlaps "
-			   "storage/segment\n", name);
-		goto out_free;
-	case -ERANGE:
-		PRINT_WARN("segment_load: not loading segment %s - exceeds "
-			   "kernel mapping range\n", name);
-		goto out_free;
-	default:
-		PRINT_WARN("segment_load: not loading segment %s (rc: %d)\n",
-			   name, rc);
+	if (rc)
 		goto out_free;
-	}
 
 	seg->res = kzalloc(sizeof(struct resource), GFP_KERNEL);
 	if (seg->res == NULL) {
@@ -582,8 +568,59 @@ out:
 	mutex_unlock(&dcss_lock);
 }
 
+/*
+ * print appropriate error message for segment_load()/segment_type()
+ * return code
+ */
+void segment_warning(int rc, char *seg_name)
+{
+	switch (rc) {
+	case -ENOENT:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "does not exist\n", seg_name);
+		break;
+	case -ENOSYS:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "not running on VM\n", seg_name);
+		break;
+	case -EIO:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "hardware error\n", seg_name);
+		break;
+	case -ENOTSUPP:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "is a multi-part segment\n", seg_name);
+		break;
+	case -ENOSPC:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "overlaps with storage\n", seg_name);
+		break;
+	case -EBUSY:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "overlaps with already loaded dcss\n", seg_name);
+		break;
+	case -EPERM:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "already loaded in incompatible mode\n", seg_name);
+		break;
+	case -ENOMEM:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "out of memory\n", seg_name);
+		break;
+	case -ERANGE:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "exceeds kernel mapping range\n", seg_name);
+		break;
+	default:
+		PRINT_WARN("cannot load/query segment %s, "
+			   "return value %i\n", seg_name, rc);
+		break;
+	}
+}
+
 EXPORT_SYMBOL(segment_load);
 EXPORT_SYMBOL(segment_unload);
 EXPORT_SYMBOL(segment_save);
 EXPORT_SYMBOL(segment_type);
 EXPORT_SYMBOL(segment_modify_shared);
+EXPORT_SYMBOL(segment_warning);
diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c
index e6c94dbfdeaa..04787eab1016 100644
--- a/drivers/s390/block/dcssblk.c
+++ b/drivers/s390/block/dcssblk.c
@@ -142,57 +142,6 @@ dcssblk_get_device_by_name(char *name)
 	return NULL;
 }
 
-/*
- * print appropriate error message for segment_load()/segment_type()
- * return code
- */
-static void
-dcssblk_segment_warn(int rc, char* seg_name)
-{
-	switch (rc) {
-	case -ENOENT:
-		PRINT_WARN("cannot load/query segment %s, does not exist\n",
-			   seg_name);
-		break;
-	case -ENOSYS:
-		PRINT_WARN("cannot load/query segment %s, not running on VM\n",
-			   seg_name);
-		break;
-	case -EIO:
-		PRINT_WARN("cannot load/query segment %s, hardware error\n",
-			   seg_name);
-		break;
-	case -ENOTSUPP:
-		PRINT_WARN("cannot load/query segment %s, is a multi-part "
-			   "segment\n", seg_name);
-		break;
-	case -ENOSPC:
-		PRINT_WARN("cannot load/query segment %s, overlaps with "
-			   "storage\n", seg_name);
-		break;
-	case -EBUSY:
-		PRINT_WARN("cannot load/query segment %s, overlaps with "
-			   "already loaded dcss\n", seg_name);
-		break;
-	case -EPERM:
-		PRINT_WARN("cannot load/query segment %s, already loaded in "
-			   "incompatible mode\n", seg_name);
-		break;
-	case -ENOMEM:
-		PRINT_WARN("cannot load/query segment %s, out of memory\n",
-			   seg_name);
-		break;
-	case -ERANGE:
-		PRINT_WARN("cannot load/query segment %s, exceeds kernel "
-			   "mapping range\n", seg_name);
-		break;
-	default:
-		PRINT_WARN("cannot load/query segment %s, return value %i\n",
-			   seg_name, rc);
-		break;
-	}
-}
-
 static void dcssblk_unregister_callback(struct device *dev)
 {
 	device_unregister(dev);
@@ -423,7 +372,7 @@ dcssblk_add_store(struct device *dev, struct device_attribute *attr, const char
 	rc = segment_load(local_buf, SEGMENT_SHARED,
 				&dev_info->start, &dev_info->end);
 	if (rc < 0) {
-		dcssblk_segment_warn(rc, dev_info->segment_name);
+		segment_warning(rc, dev_info->segment_name);
 		goto dealloc_gendisk;
 	}
 	seg_byte_size = (dev_info->end - dev_info->start + 1);
diff --git a/drivers/s390/char/monreader.c b/drivers/s390/char/monreader.c
index 67009bfa093e..1e1f50655bbf 100644
--- a/drivers/s390/char/monreader.c
+++ b/drivers/s390/char/monreader.c
@@ -111,56 +111,6 @@ static void dcss_mkname(char *ascii_name, char *ebcdic_name)
 	ASCEBC(ebcdic_name, 8);
 }
 
-/*
- * print appropriate error message for segment_load()/segment_type()
- * return code
- */
-static void mon_segment_warn(int rc, char* seg_name)
-{
-	switch (rc) {
-	case -ENOENT:
-		P_WARNING("cannot load/query segment %s, does not exist\n",
-			  seg_name);
-		break;
-	case -ENOSYS:
-		P_WARNING("cannot load/query segment %s, not running on VM\n",
-			  seg_name);
-		break;
-	case -EIO:
-		P_WARNING("cannot load/query segment %s, hardware error\n",
-			  seg_name);
-		break;
-	case -ENOTSUPP:
-		P_WARNING("cannot load/query segment %s, is a multi-part "
-			  "segment\n", seg_name);
-		break;
-	case -ENOSPC:
-		P_WARNING("cannot load/query segment %s, overlaps with "
-			  "storage\n", seg_name);
-		break;
-	case -EBUSY:
-		P_WARNING("cannot load/query segment %s, overlaps with "
-			  "already loaded dcss\n", seg_name);
-		break;
-	case -EPERM:
-		P_WARNING("cannot load/query segment %s, already loaded in "
-			  "incompatible mode\n", seg_name);
-		break;
-	case -ENOMEM:
-		P_WARNING("cannot load/query segment %s, out of memory\n",
-			  seg_name);
-		break;
-	case -ERANGE:
-		P_WARNING("cannot load/query segment %s, exceeds kernel "
-			  "mapping range\n", seg_name);
-		break;
-	default:
-		P_WARNING("cannot load/query segment %s, return value %i\n",
-			  seg_name, rc);
-		break;
-	}
-}
-
 static inline unsigned long mon_mca_start(struct mon_msg *monmsg)
 {
 	return *(u32 *) &monmsg->msg.rmmsg;
@@ -585,7 +535,7 @@ static int __init mon_init(void)
 
 	rc = segment_type(mon_dcss_name);
 	if (rc < 0) {
-		mon_segment_warn(rc, mon_dcss_name);
+		segment_warning(rc, mon_dcss_name);
 		goto out_iucv;
 	}
 	if (rc != SEG_TYPE_SC) {
@@ -598,7 +548,7 @@ static int __init mon_init(void)
 	rc = segment_load(mon_dcss_name, SEGMENT_SHARED,
 			  &mon_dcss_start, &mon_dcss_end);
 	if (rc < 0) {
-		mon_segment_warn(rc, mon_dcss_name);
+		segment_warning(rc, mon_dcss_name);
 		rc = -EINVAL;
 		goto out_iucv;
 	}
diff --git a/include/asm-s390/extmem.h b/include/asm-s390/extmem.h
index c8802c934b74..33837d756184 100644
--- a/include/asm-s390/extmem.h
+++ b/include/asm-s390/extmem.h
@@ -22,11 +22,12 @@
 #define SEGMENT_SHARED 0
 #define SEGMENT_EXCLUSIVE 1
 
-extern int segment_load (char *name,int segtype,unsigned long *addr,unsigned long *length);
-extern void segment_unload(char *name);
-extern void segment_save(char *name);
-extern int segment_type (char* name);
-extern int segment_modify_shared (char *name, int do_nonshared);
+int segment_load (char *name, int segtype, unsigned long *addr, unsigned long *length);
+void segment_unload(char *name);
+void segment_save(char *name);
+int segment_type (char* name);
+int segment_modify_shared (char *name, int do_nonshared);
+void segment_warning(int rc, char *seg_name);
 
 #endif
 #endif
-- 
cgit v1.2.3


From a332d86d3c262cddd3de0bfa90e1910de60b4f95 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 10 Feb 2008 09:04:12 +0100
Subject: hrtimer: add nanosleep specific restart_block member

The back and forth typecasting of restart_block->args is horrible. We
added a separate union member for futex already. Do the same for
nanosleep.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/thread_info.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
index 421323e5a2d6..accd7bad35b0 100644
--- a/include/linux/thread_info.h
+++ b/include/linux/thread_info.h
@@ -9,6 +9,9 @@
 
 #include <linux/types.h>
 
+struct timespec;
+struct compat_timespec;
+
 /*
  * System call restart block.
  */
@@ -26,6 +29,15 @@ struct restart_block {
 			u32 bitset;
 			u64 time;
 		} futex;
+		/* For nanosleep */
+		struct {
+			clockid_t index;
+			struct timespec __user *rmtp;
+#ifdef CONFIG_COMPAT
+			struct compat_timespec __user *compat_rmtp;
+#endif
+			u64 expires;
+		} nanosleep;
 	};
 };
 
-- 
cgit v1.2.3


From 8e60e05fdc7344415fa69a3883b11f65db967b47 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 4 Apr 2008 20:54:10 +0200
Subject: hrtimers: simplify lockdep handling

In order to avoid the false positive from lockdep, each per-cpu base->lock has
the separate lock class and migrate_hrtimers() uses double_spin_lock().

This is overcomplicated: except for migrate_hrtimers() we never take 2 locks
at once, and migrate_hrtimers() can use spin_lock_nested().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/hrtimer.h | 2 --
 kernel/hrtimer.c        | 9 ++++-----
 2 files changed, 4 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 1ad56a7b2f74..56f3236da829 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -173,7 +173,6 @@ struct hrtimer_clock_base {
  * struct hrtimer_cpu_base - the per cpu clock bases
  * @lock:		lock protecting the base and associated clock bases
  *			and timers
- * @lock_key:		the lock_class_key for use with lockdep
  * @clock_base:		array of clock bases for this cpu
  * @curr_timer:		the timer which is executing a callback right now
  * @expires_next:	absolute time of the next event which was scheduled
@@ -189,7 +188,6 @@ struct hrtimer_clock_base {
  */
 struct hrtimer_cpu_base {
 	spinlock_t			lock;
-	struct lock_class_key		lock_key;
 	struct hrtimer_clock_base	clock_base[HRTIMER_MAX_CLOCK_BASES];
 	struct list_head		cb_pending;
 #ifdef CONFIG_HIGH_RES_TIMERS
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 911e87d0440d..c642ef75069f 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -1424,7 +1424,6 @@ static void __cpuinit init_hrtimers_cpu(int cpu)
 	int i;
 
 	spin_lock_init(&cpu_base->lock);
-	lockdep_set_class(&cpu_base->lock, &cpu_base->lock_key);
 
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++)
 		cpu_base->clock_base[i].cpu_base = cpu_base;
@@ -1465,16 +1464,16 @@ static void migrate_hrtimers(int cpu)
 	tick_cancel_sched_timer(cpu);
 
 	local_irq_disable();
-	double_spin_lock(&new_base->lock, &old_base->lock,
-			 smp_processor_id() < cpu);
+	spin_lock(&new_base->lock);
+	spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING);
 
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) {
 		migrate_hrtimer_list(&old_base->clock_base[i],
 				     &new_base->clock_base[i]);
 	}
 
-	double_spin_unlock(&new_base->lock, &old_base->lock,
-			   smp_processor_id() < cpu);
+	spin_unlock(&old_base->lock);
+	spin_unlock(&new_base->lock);
 	local_irq_enable();
 	put_cpu_var(hrtimer_bases);
 }
-- 
cgit v1.2.3


From 3f3eafc921e2378954c28cfd0eb10910449f4c11 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 4 Apr 2008 20:54:10 +0200
Subject: locking: remove unused double_spin_lock()

double_spin_lock() has no callers, and it can't be used without additional
lockdep annotations, remove it.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/spinlock.h | 37 -------------------------------------
 1 file changed, 37 deletions(-)

(limited to 'include')

diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index 1129ee0a7180..d311a090fae7 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -295,43 +295,6 @@ do {						\
 	1 : ({ local_irq_restore(flags); 0; }); \
 })
 
-/*
- * Locks two spinlocks l1 and l2.
- * l1_first indicates if spinlock l1 should be taken first.
- */
-static inline void double_spin_lock(spinlock_t *l1, spinlock_t *l2,
-				    bool l1_first)
-	__acquires(l1)
-	__acquires(l2)
-{
-	if (l1_first) {
-		spin_lock(l1);
-		spin_lock(l2);
-	} else {
-		spin_lock(l2);
-		spin_lock(l1);
-	}
-}
-
-/*
- * Unlocks two spinlocks l1 and l2.
- * l1_taken_first indicates if spinlock l1 was taken first and therefore
- * should be released after spinlock l2.
- */
-static inline void double_spin_unlock(spinlock_t *l1, spinlock_t *l2,
-				      bool l1_taken_first)
-	__releases(l1)
-	__releases(l2)
-{
-	if (l1_taken_first) {
-		spin_unlock(l2);
-		spin_unlock(l1);
-	} else {
-		spin_unlock(l1);
-		spin_unlock(l2);
-	}
-}
-
 /*
  * Pull the atomic_t declaration:
  * (asm-mips/atomic.h needs above definitions)
-- 
cgit v1.2.3


From 8b91de2e58318d1168bc13d164478c1a7217a63a Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Tue, 26 Feb 2008 09:53:20 -0500
Subject: Fix quota.h includes

quota.h currently relies on asm/semaphore.h (through some chain; it
doesn't actually include semaphore.h itself) to include wait.h.  As
well as being bad practice to rely on an implicit include, subsequent
patches will break this.  While I'm in this file, add atomic.h and
list.h, and sort the list of includes.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 include/linux/quota.h | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/quota.h b/include/linux/quota.h
index 6e0393a5b2ea..eb560d031acd 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -160,14 +160,18 @@ enum {
 
 
 #ifdef __KERNEL__
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
+#include <linux/list.h>
 #include <linux/mutex.h>
+#include <linux/rwsem.h>
+#include <linux/spinlock.h>
+#include <linux/wait.h>
 
 #include <linux/dqblk_xfs.h>
 #include <linux/dqblk_v1.h>
 #include <linux/dqblk_v2.h>
 
+#include <asm/atomic.h>
+
 extern spinlock_t dq_data_lock;
 
 /* Maximal numbers of writes for quota operation (insert/delete/update)
-- 
cgit v1.2.3


From 64ac24e738823161693bf791f87adc802cf529ff Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Fri, 7 Mar 2008 21:55:58 -0500
Subject: Generic semaphore implementation

Semaphores are no longer performance-critical, so a generic C
implementation is better for maintainability, debuggability and
extensibility.  Thanks to Peter Zijlstra for fixing the lockdep
warning.  Thanks to Harvey Harrison for pointing out that the
unlikely() was unnecessary.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
---
 arch/alpha/kernel/Makefile               |   2 +-
 arch/alpha/kernel/alpha_ksyms.c          |   9 --
 arch/alpha/kernel/semaphore.c            | 224 ---------------------------
 arch/arm/kernel/Makefile                 |   2 +-
 arch/arm/kernel/semaphore.c              | 221 ---------------------------
 arch/avr32/kernel/Makefile               |   2 +-
 arch/avr32/kernel/semaphore.c            | 148 ------------------
 arch/blackfin/Kconfig                    |   4 -
 arch/blackfin/kernel/bfin_ksyms.c        |   5 -
 arch/cris/kernel/Makefile                |   3 +-
 arch/cris/kernel/crisksyms.c             |   7 -
 arch/cris/kernel/semaphore.c             | 129 ----------------
 arch/frv/kernel/Makefile                 |   2 +-
 arch/frv/kernel/frv_ksyms.c              |   1 -
 arch/frv/kernel/semaphore.c              | 155 -------------------
 arch/h8300/kernel/Makefile               |   2 +-
 arch/h8300/kernel/h8300_ksyms.c          |   1 -
 arch/h8300/kernel/semaphore.c            | 132 ----------------
 arch/ia64/kernel/Makefile                |   2 +-
 arch/ia64/kernel/ia64_ksyms.c            |   6 -
 arch/ia64/kernel/semaphore.c             | 165 --------------------
 arch/m32r/kernel/Makefile                |   2 +-
 arch/m32r/kernel/m32r_ksyms.c            |   5 -
 arch/m32r/kernel/semaphore.c             | 185 ----------------------
 arch/m68k/kernel/Makefile                |   2 +-
 arch/m68k/kernel/m68k_ksyms.c            |   6 -
 arch/m68k/kernel/semaphore.c             | 132 ----------------
 arch/m68k/lib/Makefile                   |   2 +-
 arch/m68k/lib/semaphore.S                |  53 -------
 arch/m68knommu/kernel/Makefile           |   2 +-
 arch/m68knommu/kernel/m68k_ksyms.c       |   6 -
 arch/m68knommu/kernel/semaphore.c        | 133 ----------------
 arch/m68knommu/lib/Makefile              |   2 +-
 arch/m68knommu/lib/semaphore.S           |  66 --------
 arch/mips/kernel/Makefile                |   2 +-
 arch/mips/kernel/semaphore.c             | 168 --------------------
 arch/mn10300/kernel/Makefile             |   2 +-
 arch/mn10300/kernel/semaphore.c          | 149 ------------------
 arch/parisc/kernel/Makefile              |   2 +-
 arch/parisc/kernel/parisc_ksyms.c        |   5 -
 arch/parisc/kernel/semaphore.c           | 102 -------------
 arch/powerpc/kernel/Makefile             |   2 +-
 arch/powerpc/kernel/ppc_ksyms.c          |   1 -
 arch/powerpc/kernel/semaphore.c          | 135 ----------------
 arch/ppc/kernel/semaphore.c              | 131 ----------------
 arch/s390/kernel/Makefile                |   2 +-
 arch/s390/kernel/s390_ksyms.c            |   7 -
 arch/s390/kernel/semaphore.c             | 108 -------------
 arch/sh/kernel/Makefile_32               |   2 +-
 arch/sh/kernel/Makefile_64               |   2 +-
 arch/sh/kernel/semaphore.c               | 139 -----------------
 arch/sh/kernel/sh_ksyms_32.c             |   7 -
 arch/sh/kernel/sh_ksyms_64.c             |   4 -
 arch/sparc/kernel/Makefile               |   2 +-
 arch/sparc/kernel/semaphore.c            | 155 -------------------
 arch/sparc/kernel/sparc_ksyms.c          |   5 -
 arch/sparc64/kernel/Makefile             |   2 +-
 arch/sparc64/kernel/semaphore.c          | 254 -------------------------------
 arch/sparc64/kernel/sparc64_ksyms.c      |   6 -
 arch/um/Kconfig.i386                     |   4 -
 arch/um/Kconfig.x86_64                   |   4 -
 arch/um/sys-i386/ksyms.c                 |  12 --
 arch/um/sys-ppc/Makefile                 |   8 +-
 arch/um/sys-x86_64/ksyms.c               |  13 +-
 arch/v850/kernel/Makefile                |   2 +-
 arch/v850/kernel/semaphore.c             | 166 --------------------
 arch/v850/kernel/v850_ksyms.c            |   7 -
 arch/x86/Kconfig                         |   3 -
 arch/x86/kernel/i386_ksyms_32.c          |   5 -
 arch/x86/kernel/x8664_ksyms_64.c         |   6 -
 arch/x86/lib/semaphore_32.S              |  83 ----------
 arch/x86/lib/thunk_64.S                  |   5 -
 arch/xtensa/kernel/Makefile              |   2 +-
 arch/xtensa/kernel/semaphore.c           | 226 ---------------------------
 arch/xtensa/kernel/xtensa_ksyms.c        |   9 --
 include/asm-alpha/semaphore.h            | 150 +-----------------
 include/asm-arm/semaphore-helper.h       |  84 ----------
 include/asm-arm/semaphore.h              |  99 +-----------
 include/asm-avr32/semaphore.h            | 109 +------------
 include/asm-blackfin/semaphore-helper.h  |  82 ----------
 include/asm-blackfin/semaphore.h         | 106 +------------
 include/asm-cris/semaphore-helper.h      |  78 ----------
 include/asm-cris/semaphore.h             | 134 +---------------
 include/asm-frv/semaphore.h              | 156 +------------------
 include/asm-h8300/semaphore-helper.h     |  85 -----------
 include/asm-h8300/semaphore.h            | 191 +----------------------
 include/asm-ia64/semaphore.h             | 100 +-----------
 include/asm-m32r/semaphore.h             | 145 +-----------------
 include/asm-m68k/semaphore-helper.h      | 142 -----------------
 include/asm-m68k/semaphore.h             | 164 +-------------------
 include/asm-m68knommu/semaphore-helper.h |  82 ----------
 include/asm-m68knommu/semaphore.h        | 154 +------------------
 include/asm-mips/semaphore.h             | 109 +------------
 include/asm-mn10300/semaphore.h          | 170 +--------------------
 include/asm-parisc/semaphore-helper.h    |  89 -----------
 include/asm-parisc/semaphore.h           | 146 +-----------------
 include/asm-powerpc/semaphore.h          |  95 +-----------
 include/asm-s390/semaphore.h             | 108 +------------
 include/asm-sh/semaphore-helper.h        |  89 -----------
 include/asm-sh/semaphore.h               | 116 +-------------
 include/asm-sparc/semaphore.h            | 193 +----------------------
 include/asm-sparc64/semaphore.h          |  54 +------
 include/asm-um/semaphore.h               |   7 +-
 include/asm-v850/semaphore.h             |  85 +----------
 include/asm-x86/semaphore.h              |   6 +-
 include/asm-x86/semaphore_32.h           | 175 ---------------------
 include/asm-x86/semaphore_64.h           | 180 ----------------------
 include/asm-xtensa/semaphore.h           | 100 +-----------
 include/linux/semaphore.h                |  77 ++++++++++
 kernel/Makefile                          |   2 +-
 kernel/semaphore.c                       | 187 +++++++++++++++++++++++
 lib/Makefile                             |   1 -
 lib/semaphore-sleepers.c                 | 176 ---------------------
 113 files changed, 314 insertions(+), 7679 deletions(-)
 delete mode 100644 arch/alpha/kernel/semaphore.c
 delete mode 100644 arch/arm/kernel/semaphore.c
 delete mode 100644 arch/avr32/kernel/semaphore.c
 delete mode 100644 arch/cris/kernel/semaphore.c
 delete mode 100644 arch/frv/kernel/semaphore.c
 delete mode 100644 arch/h8300/kernel/semaphore.c
 delete mode 100644 arch/ia64/kernel/semaphore.c
 delete mode 100644 arch/m32r/kernel/semaphore.c
 delete mode 100644 arch/m68k/kernel/semaphore.c
 delete mode 100644 arch/m68k/lib/semaphore.S
 delete mode 100644 arch/m68knommu/kernel/semaphore.c
 delete mode 100644 arch/m68knommu/lib/semaphore.S
 delete mode 100644 arch/mips/kernel/semaphore.c
 delete mode 100644 arch/mn10300/kernel/semaphore.c
 delete mode 100644 arch/parisc/kernel/semaphore.c
 delete mode 100644 arch/powerpc/kernel/semaphore.c
 delete mode 100644 arch/ppc/kernel/semaphore.c
 delete mode 100644 arch/s390/kernel/semaphore.c
 delete mode 100644 arch/sh/kernel/semaphore.c
 delete mode 100644 arch/sparc/kernel/semaphore.c
 delete mode 100644 arch/sparc64/kernel/semaphore.c
 delete mode 100644 arch/v850/kernel/semaphore.c
 delete mode 100644 arch/xtensa/kernel/semaphore.c
 delete mode 100644 include/asm-arm/semaphore-helper.h
 delete mode 100644 include/asm-blackfin/semaphore-helper.h
 delete mode 100644 include/asm-cris/semaphore-helper.h
 delete mode 100644 include/asm-h8300/semaphore-helper.h
 delete mode 100644 include/asm-m68k/semaphore-helper.h
 delete mode 100644 include/asm-m68knommu/semaphore-helper.h
 delete mode 100644 include/asm-parisc/semaphore-helper.h
 delete mode 100644 include/asm-sh/semaphore-helper.h
 delete mode 100644 include/asm-x86/semaphore_32.h
 delete mode 100644 include/asm-x86/semaphore_64.h
 create mode 100644 include/linux/semaphore.h
 create mode 100644 kernel/semaphore.c
 delete mode 100644 lib/semaphore-sleepers.c

(limited to 'include')

diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index dccf05245d4d..ac706c1d7ada 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -7,7 +7,7 @@ EXTRA_AFLAGS	:= $(KBUILD_CFLAGS)
 EXTRA_CFLAGS	:= -Werror -Wno-sign-compare
 
 obj-y    := entry.o traps.o process.o init_task.o osf_sys.o irq.o \
-	    irq_alpha.o signal.o setup.o ptrace.o time.o semaphore.o \
+	    irq_alpha.o signal.o setup.o ptrace.o time.o \
 	    alpha_ksyms.o systbls.o err_common.o io.o
 
 obj-$(CONFIG_VGA_HOSE)	+= console.o
diff --git a/arch/alpha/kernel/alpha_ksyms.c b/arch/alpha/kernel/alpha_ksyms.c
index e9762a33b043..d96e742d4dc2 100644
--- a/arch/alpha/kernel/alpha_ksyms.c
+++ b/arch/alpha/kernel/alpha_ksyms.c
@@ -77,15 +77,6 @@ EXPORT_SYMBOL(__do_clear_user);
 EXPORT_SYMBOL(__strncpy_from_user);
 EXPORT_SYMBOL(__strnlen_user);
 
-/* Semaphore helper functions.  */
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_failed_interruptible);
-EXPORT_SYMBOL(__up_wakeup);
-EXPORT_SYMBOL(down);
-EXPORT_SYMBOL(down_interruptible);
-EXPORT_SYMBOL(down_trylock);
-EXPORT_SYMBOL(up);
-
 /* 
  * SMP-specific symbols.
  */
diff --git a/arch/alpha/kernel/semaphore.c b/arch/alpha/kernel/semaphore.c
deleted file mode 100644
index 8d2982aa1b8d..000000000000
--- a/arch/alpha/kernel/semaphore.c
+++ /dev/null
@@ -1,224 +0,0 @@
-/*
- * Alpha semaphore implementation.
- *
- * (C) Copyright 1996 Linus Torvalds
- * (C) Copyright 1999, 2000 Richard Henderson
- */
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/init.h>
-
-/*
- * This is basically the PPC semaphore scheme ported to use
- * the Alpha ll/sc sequences, so see the PPC code for
- * credits.
- */
-
-/*
- * Atomically update sem->count.
- * This does the equivalent of the following:
- *
- *	old_count = sem->count;
- *	tmp = MAX(old_count, 0) + incr;
- *	sem->count = tmp;
- *	return old_count;
- */
-static inline int __sem_update_count(struct semaphore *sem, int incr)
-{
-	long old_count, tmp = 0;
-
-	__asm__ __volatile__(
-	"1:	ldl_l	%0,%2\n"
-	"	cmovgt	%0,%0,%1\n"
-	"	addl	%1,%3,%1\n"
-	"	stl_c	%1,%2\n"
-	"	beq	%1,2f\n"
-	"	mb\n"
-	".subsection 2\n"
-	"2:	br	1b\n"
-	".previous"
-	: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
-	: "Ir" (incr), "1" (tmp), "m" (sem->count));
-
-	return old_count;
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- */
-
-void __sched
-__down_failed(struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): down failed(%p)\n",
-	       tsk->comm, task_pid_nr(tsk), sem);
-#endif
-
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	wmb();
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	/*
-	 * Try to get the semaphore.  If the count is > 0, then we've
-	 * got the semaphore; we decrement count and exit the loop.
-	 * If the count is 0 or negative, we set it to -1, indicating
-	 * that we are asleep, and then sleep.
-	 */
-	while (__sem_update_count(sem, -1) <= 0) {
-		schedule();
-		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
-
-	/*
-	 * If there are any more sleepers, wake one of them up so
-	 * that it can either get the semaphore, or set count to -1
-	 * indicating that there are still processes sleeping.
-	 */
-	wake_up(&sem->wait);
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): down acquired(%p)\n",
-	       tsk->comm, task_pid_nr(tsk), sem);
-#endif
-}
-
-int __sched
-__down_failed_interruptible(struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	long ret = 0;
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): down failed(%p)\n",
-	       tsk->comm, task_pid_nr(tsk), sem);
-#endif
-
-	tsk->state = TASK_INTERRUPTIBLE;
-	wmb();
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	while (__sem_update_count(sem, -1) <= 0) {
-		if (signal_pending(current)) {
-			/*
-			 * A signal is pending - give up trying.
-			 * Set sem->count to 0 if it is negative,
-			 * since we are no longer sleeping.
-			 */
-			__sem_update_count(sem, 0);
-			ret = -EINTR;
-			break;
-		}
-		schedule();
-		set_task_state(tsk, TASK_INTERRUPTIBLE);
-	}
-
-	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
-	wake_up(&sem->wait);
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): down %s(%p)\n",
-	       current->comm, task_pid_nr(current),
-	       (ret < 0 ? "interrupted" : "acquired"), sem);
-#endif
-	return ret;
-}
-
-void
-__up_wakeup(struct semaphore *sem)
-{
-	/*
-	 * Note that we incremented count in up() before we came here,
-	 * but that was ineffective since the result was <= 0, and
-	 * any negative value of count is equivalent to 0.
-	 * This ends up setting count to 1, unless count is now > 0
-	 * (i.e. because some other cpu has called up() in the meantime),
-	 * in which case we just increment count.
-	 */
-	__sem_update_count(sem, 1);
-	wake_up(&sem->wait);
-}
-
-void __sched
-down(struct semaphore *sem)
-{
-#ifdef WAITQUEUE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): down(%p) <count=%d> from %p\n",
-	       current->comm, task_pid_nr(current), sem,
-	       atomic_read(&sem->count), __builtin_return_address(0));
-#endif
-	__down(sem);
-}
-
-int __sched
-down_interruptible(struct semaphore *sem)
-{
-#ifdef WAITQUEUE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): down(%p) <count=%d> from %p\n",
-	       current->comm, task_pid_nr(current), sem,
-	       atomic_read(&sem->count), __builtin_return_address(0));
-#endif
-	return __down_interruptible(sem);
-}
-
-int
-down_trylock(struct semaphore *sem)
-{
-	int ret;
-
-#ifdef WAITQUEUE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	ret = __down_trylock(sem);
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): down_trylock %s from %p\n",
-	       current->comm, task_pid_nr(current),
-	       ret ? "failed" : "acquired",
-	       __builtin_return_address(0));
-#endif
-
-	return ret;
-}
-
-void
-up(struct semaphore *sem)
-{
-#ifdef WAITQUEUE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	printk("%s(%d): up(%p) <count=%d> from %p\n",
-	       current->comm, task_pid_nr(current), sem,
-	       atomic_read(&sem->count), __builtin_return_address(0));
-#endif
-	__up(sem);
-}
diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile
index 00d44c6fbfe9..6235f72a14f0 100644
--- a/arch/arm/kernel/Makefile
+++ b/arch/arm/kernel/Makefile
@@ -7,7 +7,7 @@ AFLAGS_head.o := -DTEXT_OFFSET=$(TEXT_OFFSET)
 # Object file lists.
 
 obj-y		:= compat.o entry-armv.o entry-common.o irq.o \
-		   process.o ptrace.o semaphore.o setup.o signal.o \
+		   process.o ptrace.o setup.o signal.o \
 		   sys_arm.o stacktrace.o time.o traps.o
 
 obj-$(CONFIG_ISA_DMA_API)	+= dma.o
diff --git a/arch/arm/kernel/semaphore.c b/arch/arm/kernel/semaphore.c
deleted file mode 100644
index 981fe5c6ccbe..000000000000
--- a/arch/arm/kernel/semaphore.c
+++ /dev/null
@@ -1,221 +0,0 @@
-/*
- *  ARM semaphore implementation, taken from
- *
- *  i386 semaphore implementation.
- *
- *  (C) Copyright 1999 Linus Torvalds
- *
- *  Modified for ARM by Russell King
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/module.h>
-#include <linux/sched.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-
-#include <asm/semaphore.h>
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to acquire the semaphore, while the "sleeping"
- * variable is a count of such acquires.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * "sleeping" and the contention routine ordering is
- * protected by the semaphore spinlock.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-
-/*
- * Logic:
- *  - only on a boundary condition do we need to care. When we go
- *    from a negative count to a non-negative, we wake people up.
- *  - when we go from a non-negative count to a negative do we
- *    (a) synchronize with the "sleeper" count and (b) make sure
- *    that we're on the wakeup list before we synchronize so that
- *    we cannot lose wakeup events.
- */
-
-void __up(struct semaphore *sem)
-{
-	wake_up(&sem->wait);
-}
-
-static DEFINE_SPINLOCK(semaphore_lock);
-
-void __sched __down(struct semaphore * sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	spin_lock_irq(&semaphore_lock);
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irq(&semaphore_lock);
-
-		schedule();
-		tsk->state = TASK_UNINTERRUPTIBLE;
-		spin_lock_irq(&semaphore_lock);
-	}
-	spin_unlock_irq(&semaphore_lock);
-	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
-	wake_up(&sem->wait);
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	tsk->state = TASK_INTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	spin_lock_irq(&semaphore_lock);
-	sem->sleepers ++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * With signals pending, this turns into
-		 * the trylock failure case - we won't be
-		 * sleeping, and we* can't get the lock as
-		 * it has contention. Just correct the count
-		 * and exit.
-		 */
-		if (signal_pending(current)) {
-			retval = -EINTR;
-			sem->sleepers = 0;
-			atomic_add(sleepers, &sem->count);
-			break;
-		}
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock. The
-		 * "-1" is because we're still hoping to get
-		 * the lock.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irq(&semaphore_lock);
-
-		schedule();
-		tsk->state = TASK_INTERRUPTIBLE;
-		spin_lock_irq(&semaphore_lock);
-	}
-	spin_unlock_irq(&semaphore_lock);
-	tsk->state = TASK_RUNNING;
-	remove_wait_queue(&sem->wait, &wait);
-	wake_up(&sem->wait);
-	return retval;
-}
-
-/*
- * Trylock failed - make sure we correct for
- * having decremented the count.
- *
- * We could have done the trylock with a
- * single "cmpxchg" without failure cases,
- * but then it wouldn't work on a 386.
- */
-int __down_trylock(struct semaphore * sem)
-{
-	int sleepers;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_lock, flags);
-	sleepers = sem->sleepers + 1;
-	sem->sleepers = 0;
-
-	/*
-	 * Add "everybody else" and us into it. They aren't
-	 * playing, because we own the spinlock.
-	 */
-	if (!atomic_add_negative(sleepers, &sem->count))
-		wake_up(&sem->wait);
-
-	spin_unlock_irqrestore(&semaphore_lock, flags);
-	return 1;
-}
-
-/*
- * The semaphore operations have a special calling sequence that
- * allow us to do a simpler in-line version of them. These routines
- * need to convert that sequence back into the C sequence when
- * there is contention on the semaphore.
- *
- * ip contains the semaphore pointer on entry. Save the C-clobbered
- * registers (r0 to r3 and lr), but not ip, as we use it as a return
- * value in some cases..
- * To remain AAPCS compliant (64-bit stack align) we save r4 as well.
- */
-asm("	.section .sched.text,\"ax\",%progbits	\n\
-	.align	5				\n\
-	.globl	__down_failed			\n\
-__down_failed:					\n\
-	stmfd	sp!, {r0 - r4, lr}		\n\
-	mov	r0, ip				\n\
-	bl	__down				\n\
-	ldmfd	sp!, {r0 - r4, pc}		\n\
-						\n\
-	.align	5				\n\
-	.globl	__down_interruptible_failed	\n\
-__down_interruptible_failed:			\n\
-	stmfd	sp!, {r0 - r4, lr}		\n\
-	mov	r0, ip				\n\
-	bl	__down_interruptible		\n\
-	mov	ip, r0				\n\
-	ldmfd	sp!, {r0 - r4, pc}		\n\
-						\n\
-	.align	5				\n\
-	.globl	__down_trylock_failed		\n\
-__down_trylock_failed:				\n\
-	stmfd	sp!, {r0 - r4, lr}		\n\
-	mov	r0, ip				\n\
-	bl	__down_trylock			\n\
-	mov	ip, r0				\n\
-	ldmfd	sp!, {r0 - r4, pc}		\n\
-						\n\
-	.align	5				\n\
-	.globl	__up_wakeup			\n\
-__up_wakeup:					\n\
-	stmfd	sp!, {r0 - r4, lr}		\n\
-	mov	r0, ip				\n\
-	bl	__up				\n\
-	ldmfd	sp!, {r0 - r4, pc}		\n\
-	");
-
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_interruptible_failed);
-EXPORT_SYMBOL(__down_trylock_failed);
-EXPORT_SYMBOL(__up_wakeup);
diff --git a/arch/avr32/kernel/Makefile b/arch/avr32/kernel/Makefile
index e4b6d122b033..18229d0d1861 100644
--- a/arch/avr32/kernel/Makefile
+++ b/arch/avr32/kernel/Makefile
@@ -6,7 +6,7 @@ extra-y				:= head.o vmlinux.lds
 
 obj-$(CONFIG_SUBARCH_AVR32B)	+= entry-avr32b.o
 obj-y				+= syscall_table.o syscall-stubs.o irq.o
-obj-y				+= setup.o traps.o semaphore.o ocd.o ptrace.o
+obj-y				+= setup.o traps.o ocd.o ptrace.o
 obj-y				+= signal.o sys_avr32.o process.o time.o
 obj-y				+= init_task.o switch_to.o cpu.o
 obj-$(CONFIG_MODULES)		+= module.o avr32_ksyms.o
diff --git a/arch/avr32/kernel/semaphore.c b/arch/avr32/kernel/semaphore.c
deleted file mode 100644
index 1e2705a05016..000000000000
--- a/arch/avr32/kernel/semaphore.c
+++ /dev/null
@@ -1,148 +0,0 @@
-/*
- * AVR32 sempahore implementation.
- *
- * Copyright (C) 2004-2006 Atmel Corporation
- *
- * Based on linux/arch/i386/kernel/semaphore.c
- *  Copyright (C) 1999 Linus Torvalds
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/sched.h>
-#include <linux/errno.h>
-#include <linux/module.h>
-
-#include <asm/semaphore.h>
-#include <asm/atomic.h>
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to acquire the semaphore, while the "sleeping"
- * variable is a count of such acquires.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * "sleeping" and the contention routine ordering is protected
- * by the spinlock in the semaphore's waitqueue head.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-
-/*
- * Logic:
- *  - only on a boundary condition do we need to care. When we go
- *    from a negative count to a non-negative, we wake people up.
- *  - when we go from a non-negative count to a negative do we
- *    (a) synchronize with the "sleeper" count and (b) make sure
- *    that we're on the wakeup list before we synchronize so that
- *    we cannot lose wakeup events.
- */
-
-void __up(struct semaphore *sem)
-{
-	wake_up(&sem->wait);
-}
-EXPORT_SYMBOL(__up);
-
-void __sched __down(struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-        DECLARE_WAITQUEUE(wait, tsk);
-        unsigned long flags;
-
-        tsk->state = TASK_UNINTERRUPTIBLE;
-        spin_lock_irqsave(&sem->wait.lock, flags);
-        add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-        sem->sleepers++;
-        for (;;) {
-                int sleepers = sem->sleepers;
-
-                /*
-                 * Add "everybody else" into it. They aren't
-                 * playing, because we own the spinlock in
-                 * the wait_queue_head.
-                 */
-                if (atomic_add_return(sleepers - 1, &sem->count) >= 0) {
-                        sem->sleepers = 0;
-                        break;
-                }
-                sem->sleepers = 1;      /* us - see -1 above */
-                spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-                schedule();
-
-                spin_lock_irqsave(&sem->wait.lock, flags);
-                tsk->state = TASK_UNINTERRUPTIBLE;
-        }
-        remove_wait_queue_locked(&sem->wait, &wait);
-        wake_up_locked(&sem->wait);
-        spin_unlock_irqrestore(&sem->wait.lock, flags);
-        tsk->state = TASK_RUNNING;
-}
-EXPORT_SYMBOL(__down);
-
-int __sched __down_interruptible(struct semaphore *sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-        DECLARE_WAITQUEUE(wait, tsk);
-        unsigned long flags;
-
-        tsk->state = TASK_INTERRUPTIBLE;
-        spin_lock_irqsave(&sem->wait.lock, flags);
-        add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-        sem->sleepers++;
-        for (;;) {
-                int sleepers = sem->sleepers;
-
-		/*
-		 * With signals pending, this turns into the trylock
-		 * failure case - we won't be sleeping, and we can't
-		 * get the lock as it has contention. Just correct the
-		 * count and exit.
-		 */
-		if (signal_pending(current)) {
-			retval = -EINTR;
-			sem->sleepers = 0;
-			atomic_add(sleepers, &sem->count);
-			break;
-		}
-
-                /*
-                 * Add "everybody else" into it. They aren't
-                 * playing, because we own the spinlock in
-                 * the wait_queue_head.
-                 */
-                if (atomic_add_return(sleepers - 1, &sem->count) >= 0) {
-                        sem->sleepers = 0;
-                        break;
-                }
-                sem->sleepers = 1;      /* us - see -1 above */
-                spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-                schedule();
-
-                spin_lock_irqsave(&sem->wait.lock, flags);
-                tsk->state = TASK_INTERRUPTIBLE;
-        }
-        remove_wait_queue_locked(&sem->wait, &wait);
-        wake_up_locked(&sem->wait);
-        spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-        tsk->state = TASK_RUNNING;
-	return retval;
-}
-EXPORT_SYMBOL(__down_interruptible);
diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 589c6aca4803..2dd1f300a5cf 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -31,10 +31,6 @@ config ZONE_DMA
 	bool
 	default y
 
-config SEMAPHORE_SLEEPERS
-	bool
-	default y
-
 config GENERIC_FIND_NEXT_BIT
 	bool
 	default y
diff --git a/arch/blackfin/kernel/bfin_ksyms.c b/arch/blackfin/kernel/bfin_ksyms.c
index 0bfbb269e350..053edff6c0d8 100644
--- a/arch/blackfin/kernel/bfin_ksyms.c
+++ b/arch/blackfin/kernel/bfin_ksyms.c
@@ -42,11 +42,6 @@ EXPORT_SYMBOL(ip_fast_csum);
 
 EXPORT_SYMBOL(kernel_thread);
 
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_trylock);
-EXPORT_SYMBOL(__down_interruptible);
-
 EXPORT_SYMBOL(is_in_rom);
 EXPORT_SYMBOL(bfin_return_from_exception);
 
diff --git a/arch/cris/kernel/Makefile b/arch/cris/kernel/Makefile
index c8e8ea570989..ee7bcd4d20b2 100644
--- a/arch/cris/kernel/Makefile
+++ b/arch/cris/kernel/Makefile
@@ -5,8 +5,7 @@
 
 extra-y	:= vmlinux.lds
 
-obj-y   := process.o traps.o irq.o ptrace.o setup.o \
-	   time.o sys_cris.o semaphore.o
+obj-y   := process.o traps.o irq.o ptrace.o setup.o time.o sys_cris.o
 
 obj-$(CONFIG_MODULES)    += crisksyms.o
 obj-$(CONFIG_MODULES)	 += module.o
diff --git a/arch/cris/kernel/crisksyms.c b/arch/cris/kernel/crisksyms.c
index 62f0e752915a..7ac000f6a888 100644
--- a/arch/cris/kernel/crisksyms.c
+++ b/arch/cris/kernel/crisksyms.c
@@ -9,7 +9,6 @@
 #include <linux/string.h>
 #include <linux/tty.h>
 
-#include <asm/semaphore.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
 #include <asm/checksum.h>
@@ -49,12 +48,6 @@ EXPORT_SYMBOL(__negdi2);
 EXPORT_SYMBOL(__ioremap);
 EXPORT_SYMBOL(iounmap);
 
-/* Semaphore functions */
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_interruptible);
-EXPORT_SYMBOL(__down_trylock);
-
 /* Userspace access functions */
 EXPORT_SYMBOL(__copy_user_zeroing);
 EXPORT_SYMBOL(__copy_user);
diff --git a/arch/cris/kernel/semaphore.c b/arch/cris/kernel/semaphore.c
deleted file mode 100644
index f137a439041f..000000000000
--- a/arch/cris/kernel/semaphore.c
+++ /dev/null
@@ -1,129 +0,0 @@
-/*
- *  Generic semaphore code. Buyer beware. Do your own
- * specific changes in <asm/semaphore-helper.h>
- */
-
-#include <linux/sched.h>
-#include <asm/semaphore-helper.h>
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to sleep, while the "waking" variable is
- * incremented when the "up()" code goes to wake up waiting
- * processes.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * waking_non_zero() (from asm/semaphore.h) must execute
- * atomically.
- *
- * When __up() is called, the count was negative before
- * incrementing it, and we need to wake up somebody.
- *
- * This routine adds one to the count of processes that need to
- * wake up and exit.  ALL waiting processes actually wake up but
- * only the one that gets to the "waking" field first will gate
- * through and acquire the semaphore.  The others will go back
- * to sleep.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-void __up(struct semaphore *sem)
-{
-	wake_one_more(sem);
-	wake_up(&sem->wait);
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from __down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from __down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- *
- */
-
-#define DOWN_VAR				\
-	struct task_struct *tsk = current;	\
-	wait_queue_t wait;			\
-	init_waitqueue_entry(&wait, tsk);
-
-#define DOWN_HEAD(task_state)						\
-									\
-									\
-	tsk->state = (task_state);					\
-	add_wait_queue(&sem->wait, &wait);				\
-									\
-	/*								\
-	 * Ok, we're set up.  sem->count is known to be less than zero	\
-	 * so we must wait.						\
-	 *								\
-	 * We can let go the lock for purposes of waiting.		\
-	 * We re-acquire it after awaking so as to protect		\
-	 * all semaphore operations.					\
-	 *								\
-	 * If "up()" is called before we call waking_non_zero() then	\
-	 * we will catch it right away.  If it is called later then	\
-	 * we will have to go through a wakeup cycle to catch it.	\
-	 *								\
-	 * Multiple waiters contend for the semaphore lock to see	\
-	 * who gets to gate through and who has to wait some more.	\
-	 */								\
-	for (;;) {
-
-#define DOWN_TAIL(task_state)			\
-		tsk->state = (task_state);	\
-	}					\
-	tsk->state = TASK_RUNNING;		\
-	remove_wait_queue(&sem->wait, &wait);
-
-void __sched __down(struct semaphore * sem)
-{
-	DOWN_VAR
-	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
-	if (waking_non_zero(sem))
-		break;
-	schedule();
-	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-	DOWN_VAR
-	DOWN_HEAD(TASK_INTERRUPTIBLE)
-
-	ret = waking_non_zero_interruptible(sem, tsk);
-	if (ret)
-	{
-		if (ret == 1)
-			/* ret != 0 only if we get interrupted -arca */
-			ret = 0;
-		break;
-	}
-	schedule();
-	DOWN_TAIL(TASK_INTERRUPTIBLE)
-	return ret;
-}
-
-int __down_trylock(struct semaphore * sem)
-{
-	return waking_non_zero_trylock(sem);
-}
diff --git a/arch/frv/kernel/Makefile b/arch/frv/kernel/Makefile
index e8f73ed28b52..c36f70b6699a 100644
--- a/arch/frv/kernel/Makefile
+++ b/arch/frv/kernel/Makefile
@@ -9,7 +9,7 @@ extra-y:= head.o init_task.o vmlinux.lds
 
 obj-y := $(heads-y) entry.o entry-table.o break.o switch_to.o kernel_thread.o \
 	 kernel_execve.o process.o traps.o ptrace.o signal.o dma.o \
-	 sys_frv.o time.o semaphore.o setup.o frv_ksyms.o \
+	 sys_frv.o time.o setup.o frv_ksyms.o \
 	 debug-stub.o irq.o sleep.o uaccess.o
 
 obj-$(CONFIG_GDBSTUB)		+= gdb-stub.o gdb-io.o
diff --git a/arch/frv/kernel/frv_ksyms.c b/arch/frv/kernel/frv_ksyms.c
index f772704b3d28..0316b3c50eff 100644
--- a/arch/frv/kernel/frv_ksyms.c
+++ b/arch/frv/kernel/frv_ksyms.c
@@ -12,7 +12,6 @@
 #include <asm/pgalloc.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/semaphore.h>
 #include <asm/checksum.h>
 #include <asm/hardirq.h>
 #include <asm/cacheflush.h>
diff --git a/arch/frv/kernel/semaphore.c b/arch/frv/kernel/semaphore.c
deleted file mode 100644
index 7ee3a147b471..000000000000
--- a/arch/frv/kernel/semaphore.c
+++ /dev/null
@@ -1,155 +0,0 @@
-/* semaphore.c: FR-V semaphores
- *
- * Copyright (C) 2003 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- * - Derived from lib/rwsem-spinlock.c
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-
-#include <linux/sched.h>
-#include <linux/module.h>
-#include <asm/semaphore.h>
-
-struct sem_waiter {
-	struct list_head	list;
-	struct task_struct	*task;
-};
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-void semtrace(struct semaphore *sem, const char *str)
-{
-	if (sem->debug)
-		printk("[%d] %s({%d,%d})\n",
-		       current->pid,
-		       str,
-		       sem->counter,
-		       list_empty(&sem->wait_list) ? 0 : 1);
-}
-#else
-#define semtrace(SEM,STR) do { } while(0)
-#endif
-
-/*
- * wait for a token to be granted from a semaphore
- * - entered with lock held and interrupts disabled
- */
-void __down(struct semaphore *sem, unsigned long flags)
-{
-	struct task_struct *tsk = current;
-	struct sem_waiter waiter;
-
-	semtrace(sem, "Entering __down");
-
-	/* set up my own style of waitqueue */
-	waiter.task = tsk;
-	get_task_struct(tsk);
-
-	list_add_tail(&waiter.list, &sem->wait_list);
-
-	/* we don't need to touch the semaphore struct anymore */
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-
-	/* wait to be given the semaphore */
-	set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-
-	for (;;) {
-		if (list_empty(&waiter.list))
-			break;
-		schedule();
-		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	}
-
-	tsk->state = TASK_RUNNING;
-	semtrace(sem, "Leaving __down");
-}
-
-EXPORT_SYMBOL(__down);
-
-/*
- * interruptibly wait for a token to be granted from a semaphore
- * - entered with lock held and interrupts disabled
- */
-int __down_interruptible(struct semaphore *sem, unsigned long flags)
-{
-	struct task_struct *tsk = current;
-	struct sem_waiter waiter;
-	int ret;
-
-	semtrace(sem,"Entering __down_interruptible");
-
-	/* set up my own style of waitqueue */
-	waiter.task = tsk;
-	get_task_struct(tsk);
-
-	list_add_tail(&waiter.list, &sem->wait_list);
-
-	/* we don't need to touch the semaphore struct anymore */
-	set_task_state(tsk, TASK_INTERRUPTIBLE);
-
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-
-	/* wait to be given the semaphore */
-	ret = 0;
-	for (;;) {
-		if (list_empty(&waiter.list))
-			break;
-		if (unlikely(signal_pending(current)))
-			goto interrupted;
-		schedule();
-		set_task_state(tsk, TASK_INTERRUPTIBLE);
-	}
-
- out:
-	tsk->state = TASK_RUNNING;
-	semtrace(sem, "Leaving __down_interruptible");
-	return ret;
-
- interrupted:
-	spin_lock_irqsave(&sem->wait_lock, flags);
-
-	if (!list_empty(&waiter.list)) {
-		list_del(&waiter.list);
-		ret = -EINTR;
-	}
-
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-	if (ret == -EINTR)
-		put_task_struct(current);
-	goto out;
-}
-
-EXPORT_SYMBOL(__down_interruptible);
-
-/*
- * release a single token back to a semaphore
- * - entered with lock held and interrupts disabled
- */
-void __up(struct semaphore *sem)
-{
-	struct task_struct *tsk;
-	struct sem_waiter *waiter;
-
-	semtrace(sem,"Entering __up");
-
-	/* grant the token to the process at the front of the queue */
-	waiter = list_entry(sem->wait_list.next, struct sem_waiter, list);
-
-	/* We must be careful not to touch 'waiter' after we set ->task = NULL.
-	 * It is allocated on the waiter's stack and may become invalid at
-	 * any time after that point (due to a wakeup from another source).
-	 */
-	list_del_init(&waiter->list);
-	tsk = waiter->task;
-	mb();
-	waiter->task = NULL;
-	wake_up_process(tsk);
-	put_task_struct(tsk);
-
-	semtrace(sem,"Leaving __up");
-}
-
-EXPORT_SYMBOL(__up);
diff --git a/arch/h8300/kernel/Makefile b/arch/h8300/kernel/Makefile
index 874f6aefee65..6c248c3c5c3b 100644
--- a/arch/h8300/kernel/Makefile
+++ b/arch/h8300/kernel/Makefile
@@ -5,7 +5,7 @@
 extra-y := vmlinux.lds
 
 obj-y := process.o traps.o ptrace.o irq.o \
-	 sys_h8300.o time.o semaphore.o signal.o \
+	 sys_h8300.o time.o signal.o \
          setup.o gpio.o init_task.o syscalls.o \
 	 entry.o
 
diff --git a/arch/h8300/kernel/h8300_ksyms.c b/arch/h8300/kernel/h8300_ksyms.c
index d1b15267ac81..6866bd9c7fb4 100644
--- a/arch/h8300/kernel/h8300_ksyms.c
+++ b/arch/h8300/kernel/h8300_ksyms.c
@@ -12,7 +12,6 @@
 #include <asm/pgalloc.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/semaphore.h>
 #include <asm/checksum.h>
 #include <asm/current.h>
 #include <asm/gpio.h>
diff --git a/arch/h8300/kernel/semaphore.c b/arch/h8300/kernel/semaphore.c
deleted file mode 100644
index d12cbbfe6ebd..000000000000
--- a/arch/h8300/kernel/semaphore.c
+++ /dev/null
@@ -1,132 +0,0 @@
-/*
- *  Generic semaphore code. Buyer beware. Do your own
- * specific changes in <asm/semaphore-helper.h>
- */
-
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <asm/semaphore-helper.h>
-
-#ifndef CONFIG_RMW_INSNS
-spinlock_t semaphore_wake_lock;
-#endif
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to sleep, while the "waking" variable is
- * incremented when the "up()" code goes to wake up waiting
- * processes.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * waking_non_zero() (from asm/semaphore.h) must execute
- * atomically.
- *
- * When __up() is called, the count was negative before
- * incrementing it, and we need to wake up somebody.
- *
- * This routine adds one to the count of processes that need to
- * wake up and exit.  ALL waiting processes actually wake up but
- * only the one that gets to the "waking" field first will gate
- * through and acquire the semaphore.  The others will go back
- * to sleep.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-void __up(struct semaphore *sem)
-{
-	wake_one_more(sem);
-	wake_up(&sem->wait);
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from __down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from __down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- *
- */
-
-
-#define DOWN_HEAD(task_state)						\
-									\
-									\
-	current->state = (task_state);					\
-	add_wait_queue(&sem->wait, &wait);				\
-									\
-	/*								\
-	 * Ok, we're set up.  sem->count is known to be less than zero	\
-	 * so we must wait.						\
-	 *								\
-	 * We can let go the lock for purposes of waiting.		\
-	 * We re-acquire it after awaking so as to protect		\
-	 * all semaphore operations.					\
-	 *								\
-	 * If "up()" is called before we call waking_non_zero() then	\
-	 * we will catch it right away.  If it is called later then	\
-	 * we will have to go through a wakeup cycle to catch it.	\
-	 *								\
-	 * Multiple waiters contend for the semaphore lock to see	\
-	 * who gets to gate through and who has to wait some more.	\
-	 */								\
-	for (;;) {
-
-#define DOWN_TAIL(task_state)			\
-		current->state = (task_state);	\
-	}					\
-	current->state = TASK_RUNNING;		\
-	remove_wait_queue(&sem->wait, &wait);
-
-void __sched __down(struct semaphore * sem)
-{
-	DECLARE_WAITQUEUE(wait, current);
-
-	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
-	if (waking_non_zero(sem))
-		break;
-	schedule();
-	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	int ret = 0;
-
-	DOWN_HEAD(TASK_INTERRUPTIBLE)
-
-	ret = waking_non_zero_interruptible(sem, current);
-	if (ret)
-	{
-		if (ret == 1)
-			/* ret != 0 only if we get interrupted -arca */
-			ret = 0;
-		break;
-	}
-	schedule();
-	DOWN_TAIL(TASK_INTERRUPTIBLE)
-	return ret;
-}
-
-int __down_trylock(struct semaphore * sem)
-{
-	return waking_non_zero_trylock(sem);
-}
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index 33e5a598672d..13fd10e8699e 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -6,7 +6,7 @@ extra-y	:= head.o init_task.o vmlinux.lds
 
 obj-y := acpi.o entry.o efi.o efi_stub.o gate-data.o fsys.o ia64_ksyms.o irq.o irq_ia64.o	\
 	 irq_lsapic.o ivt.o machvec.o pal.o patch.o process.o perfmon.o ptrace.o sal.o		\
-	 salinfo.o semaphore.o setup.o signal.o sys_ia64.o time.o traps.o unaligned.o \
+	 salinfo.o setup.o signal.o sys_ia64.o time.o traps.o unaligned.o \
 	 unwind.o mca.o mca_asm.o topology.o
 
 obj-$(CONFIG_IA64_BRL_EMU)	+= brl_emu.o
diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index 8e7193d55528..6da1f20d7372 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -19,12 +19,6 @@ EXPORT_SYMBOL_GPL(empty_zero_page);
 EXPORT_SYMBOL(ip_fast_csum);		/* hand-coded assembly */
 EXPORT_SYMBOL(csum_ipv6_magic);
 
-#include <asm/semaphore.h>
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_interruptible);
-EXPORT_SYMBOL(__down_trylock);
-EXPORT_SYMBOL(__up);
-
 #include <asm/page.h>
 EXPORT_SYMBOL(clear_page);
 
diff --git a/arch/ia64/kernel/semaphore.c b/arch/ia64/kernel/semaphore.c
deleted file mode 100644
index 2724ef3fbae2..000000000000
--- a/arch/ia64/kernel/semaphore.c
+++ /dev/null
@@ -1,165 +0,0 @@
-/*
- * IA-64 semaphore implementation (derived from x86 version).
- *
- * Copyright (C) 1999-2000, 2002 Hewlett-Packard Co
- *	David Mosberger-Tang <davidm@hpl.hp.com>
- */
-
-/*
- * Semaphores are implemented using a two-way counter: The "count"
- * variable is decremented for each process that tries to acquire the
- * semaphore, while the "sleepers" variable is a count of such
- * acquires.
- *
- * Notably, the inline "up()" and "down()" functions can efficiently
- * test if they need to do any extra work (up needs to do something
- * only if count was negative before the increment operation.
- *
- * "sleeping" and the contention routine ordering is protected
- * by the spinlock in the semaphore's waitqueue head.
- *
- * Note that these functions are only called when there is contention
- * on the lock, and as such all this is the "non-critical" part of the
- * whole semaphore business. The critical part is the inline stuff in
- * <asm/semaphore.h> where we want to avoid any extra jumps and calls.
- */
-#include <linux/sched.h>
-#include <linux/init.h>
-
-#include <asm/errno.h>
-#include <asm/semaphore.h>
-
-/*
- * Logic:
- *  - Only on a boundary condition do we need to care. When we go
- *    from a negative count to a non-negative, we wake people up.
- *  - When we go from a non-negative count to a negative do we
- *    (a) synchronize with the "sleepers" count and (b) make sure
- *    that we're on the wakeup list before we synchronize so that
- *    we cannot lose wakeup events.
- */
-
-void
-__up (struct semaphore *sem)
-{
-	wake_up(&sem->wait);
-}
-
-void __sched __down (struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	unsigned long flags;
-
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock in
-		 * the wait_queue_head.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-		schedule();
-
-		spin_lock_irqsave(&sem->wait.lock, flags);
-		tsk->state = TASK_UNINTERRUPTIBLE;
-	}
-	remove_wait_queue_locked(&sem->wait, &wait);
-	wake_up_locked(&sem->wait);
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-	tsk->state = TASK_RUNNING;
-}
-
-int __sched __down_interruptible (struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	unsigned long flags;
-
-	tsk->state = TASK_INTERRUPTIBLE;
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-	sem->sleepers ++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * With signals pending, this turns into
-		 * the trylock failure case - we won't be
-		 * sleeping, and we* can't get the lock as
-		 * it has contention. Just correct the count
-		 * and exit.
-		 */
-		if (signal_pending(current)) {
-			retval = -EINTR;
-			sem->sleepers = 0;
-			atomic_add(sleepers, &sem->count);
-			break;
-		}
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock in
-		 * wait_queue_head. The "-1" is because we're
-		 * still hoping to get the semaphore.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-		schedule();
-
-		spin_lock_irqsave(&sem->wait.lock, flags);
-		tsk->state = TASK_INTERRUPTIBLE;
-	}
-	remove_wait_queue_locked(&sem->wait, &wait);
-	wake_up_locked(&sem->wait);
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-	tsk->state = TASK_RUNNING;
-	return retval;
-}
-
-/*
- * Trylock failed - make sure we correct for having decremented the
- * count.
- */
-int
-__down_trylock (struct semaphore *sem)
-{
-	unsigned long flags;
-	int sleepers;
-
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	sleepers = sem->sleepers + 1;
-	sem->sleepers = 0;
-
-	/*
-	 * Add "everybody else" and us into it. They aren't
-	 * playing, because we own the spinlock in the
-	 * wait_queue_head.
-	 */
-	if (!atomic_add_negative(sleepers, &sem->count)) {
-		wake_up_locked(&sem->wait);
-	}
-
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-	return 1;
-}
diff --git a/arch/m32r/kernel/Makefile b/arch/m32r/kernel/Makefile
index e97e26e87c9e..09200d4886e3 100644
--- a/arch/m32r/kernel/Makefile
+++ b/arch/m32r/kernel/Makefile
@@ -5,7 +5,7 @@
 extra-y	:= head.o init_task.o vmlinux.lds
 
 obj-y	:= process.o entry.o traps.o align.o irq.o setup.o time.o \
-	m32r_ksyms.o sys_m32r.o semaphore.o signal.o ptrace.o
+	m32r_ksyms.o sys_m32r.o signal.o ptrace.o
 
 obj-$(CONFIG_SMP)		+= smp.o smpboot.o
 obj-$(CONFIG_MODULES)		+= module.o
diff --git a/arch/m32r/kernel/m32r_ksyms.c b/arch/m32r/kernel/m32r_ksyms.c
index 41a4c95e06d6..e6709fe950ba 100644
--- a/arch/m32r/kernel/m32r_ksyms.c
+++ b/arch/m32r/kernel/m32r_ksyms.c
@@ -7,7 +7,6 @@
 #include <linux/interrupt.h>
 #include <linux/string.h>
 
-#include <asm/semaphore.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
 #include <asm/checksum.h>
@@ -22,10 +21,6 @@ EXPORT_SYMBOL(dump_fpu);
 EXPORT_SYMBOL(__ioremap);
 EXPORT_SYMBOL(iounmap);
 EXPORT_SYMBOL(kernel_thread);
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_interruptible);
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down_trylock);
 
 /* Networking helper routines. */
 /* Delay loops */
diff --git a/arch/m32r/kernel/semaphore.c b/arch/m32r/kernel/semaphore.c
deleted file mode 100644
index 940c2d37cfd1..000000000000
--- a/arch/m32r/kernel/semaphore.c
+++ /dev/null
@@ -1,185 +0,0 @@
-/*
- *  linux/arch/m32r/semaphore.c
- *    orig : i386 2.6.4
- *
- *  M32R semaphore implementation.
- *
- *	Copyright (c) 2002 - 2004 Hitoshi Yamamoto
- */
-
-/*
- * i386 semaphore implementation.
- *
- * (C) Copyright 1999 Linus Torvalds
- *
- * Portions Copyright 1999 Red Hat, Inc.
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- * rw semaphores implemented November 1999 by Benjamin LaHaise <bcrl@kvack.org>
- */
-#include <linux/sched.h>
-#include <linux/err.h>
-#include <linux/init.h>
-#include <asm/semaphore.h>
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to acquire the semaphore, while the "sleeping"
- * variable is a count of such acquires.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * "sleeping" and the contention routine ordering is protected
- * by the spinlock in the semaphore's waitqueue head.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-
-/*
- * Logic:
- *  - only on a boundary condition do we need to care. When we go
- *    from a negative count to a non-negative, we wake people up.
- *  - when we go from a non-negative count to a negative do we
- *    (a) synchronize with the "sleeper" count and (b) make sure
- *    that we're on the wakeup list before we synchronize so that
- *    we cannot lose wakeup events.
- */
-
-asmlinkage void __up(struct semaphore *sem)
-{
-	wake_up(&sem->wait);
-}
-
-asmlinkage void __sched __down(struct semaphore * sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	unsigned long flags;
-
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock in
-		 * the wait_queue_head.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-		schedule();
-
-		spin_lock_irqsave(&sem->wait.lock, flags);
-		tsk->state = TASK_UNINTERRUPTIBLE;
-	}
-	remove_wait_queue_locked(&sem->wait, &wait);
-	wake_up_locked(&sem->wait);
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-	tsk->state = TASK_RUNNING;
-}
-
-asmlinkage int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	unsigned long flags;
-
-	tsk->state = TASK_INTERRUPTIBLE;
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * With signals pending, this turns into
-		 * the trylock failure case - we won't be
-		 * sleeping, and we* can't get the lock as
-		 * it has contention. Just correct the count
-		 * and exit.
-		 */
-		if (signal_pending(current)) {
-			retval = -EINTR;
-			sem->sleepers = 0;
-			atomic_add(sleepers, &sem->count);
-			break;
-		}
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock in
-		 * wait_queue_head. The "-1" is because we're
-		 * still hoping to get the semaphore.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-		schedule();
-
-		spin_lock_irqsave(&sem->wait.lock, flags);
-		tsk->state = TASK_INTERRUPTIBLE;
-	}
-	remove_wait_queue_locked(&sem->wait, &wait);
-	wake_up_locked(&sem->wait);
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-	tsk->state = TASK_RUNNING;
-	return retval;
-}
-
-/*
- * Trylock failed - make sure we correct for
- * having decremented the count.
- *
- * We could have done the trylock with a
- * single "cmpxchg" without failure cases,
- * but then it wouldn't work on a 386.
- */
-asmlinkage int __down_trylock(struct semaphore * sem)
-{
-	int sleepers;
-	unsigned long flags;
-
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	sleepers = sem->sleepers + 1;
-	sem->sleepers = 0;
-
-	/*
-	 * Add "everybody else" and us into it. They aren't
-	 * playing, because we own the spinlock in the
-	 * wait_queue_head.
-	 */
-	if (!atomic_add_negative(sleepers, &sem->count)) {
-		wake_up_locked(&sem->wait);
-	}
-
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-	return 1;
-}
diff --git a/arch/m68k/kernel/Makefile b/arch/m68k/kernel/Makefile
index a806208c7fb5..7a62a718143b 100644
--- a/arch/m68k/kernel/Makefile
+++ b/arch/m68k/kernel/Makefile
@@ -10,7 +10,7 @@ endif
 extra-y	+= vmlinux.lds
 
 obj-y	:= entry.o process.o traps.o ints.o signal.o ptrace.o module.o \
-	   sys_m68k.o time.o semaphore.o setup.o m68k_ksyms.o devres.o
+	   sys_m68k.o time.o setup.o m68k_ksyms.o devres.o
 
 devres-y = ../../../kernel/irq/devres.o
 
diff --git a/arch/m68k/kernel/m68k_ksyms.c b/arch/m68k/kernel/m68k_ksyms.c
index 6fc69c74fe2e..d900e77e5363 100644
--- a/arch/m68k/kernel/m68k_ksyms.c
+++ b/arch/m68k/kernel/m68k_ksyms.c
@@ -1,5 +1,4 @@
 #include <linux/module.h>
-#include <asm/semaphore.h>
 
 asmlinkage long long __ashldi3 (long long, int);
 asmlinkage long long __ashrdi3 (long long, int);
@@ -15,8 +14,3 @@ EXPORT_SYMBOL(__ashrdi3);
 EXPORT_SYMBOL(__lshrdi3);
 EXPORT_SYMBOL(__muldi3);
 
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_failed_interruptible);
-EXPORT_SYMBOL(__down_failed_trylock);
-EXPORT_SYMBOL(__up_wakeup);
-
diff --git a/arch/m68k/kernel/semaphore.c b/arch/m68k/kernel/semaphore.c
deleted file mode 100644
index d12cbbfe6ebd..000000000000
--- a/arch/m68k/kernel/semaphore.c
+++ /dev/null
@@ -1,132 +0,0 @@
-/*
- *  Generic semaphore code. Buyer beware. Do your own
- * specific changes in <asm/semaphore-helper.h>
- */
-
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <asm/semaphore-helper.h>
-
-#ifndef CONFIG_RMW_INSNS
-spinlock_t semaphore_wake_lock;
-#endif
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to sleep, while the "waking" variable is
- * incremented when the "up()" code goes to wake up waiting
- * processes.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * waking_non_zero() (from asm/semaphore.h) must execute
- * atomically.
- *
- * When __up() is called, the count was negative before
- * incrementing it, and we need to wake up somebody.
- *
- * This routine adds one to the count of processes that need to
- * wake up and exit.  ALL waiting processes actually wake up but
- * only the one that gets to the "waking" field first will gate
- * through and acquire the semaphore.  The others will go back
- * to sleep.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-void __up(struct semaphore *sem)
-{
-	wake_one_more(sem);
-	wake_up(&sem->wait);
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from __down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from __down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- *
- */
-
-
-#define DOWN_HEAD(task_state)						\
-									\
-									\
-	current->state = (task_state);					\
-	add_wait_queue(&sem->wait, &wait);				\
-									\
-	/*								\
-	 * Ok, we're set up.  sem->count is known to be less than zero	\
-	 * so we must wait.						\
-	 *								\
-	 * We can let go the lock for purposes of waiting.		\
-	 * We re-acquire it after awaking so as to protect		\
-	 * all semaphore operations.					\
-	 *								\
-	 * If "up()" is called before we call waking_non_zero() then	\
-	 * we will catch it right away.  If it is called later then	\
-	 * we will have to go through a wakeup cycle to catch it.	\
-	 *								\
-	 * Multiple waiters contend for the semaphore lock to see	\
-	 * who gets to gate through and who has to wait some more.	\
-	 */								\
-	for (;;) {
-
-#define DOWN_TAIL(task_state)			\
-		current->state = (task_state);	\
-	}					\
-	current->state = TASK_RUNNING;		\
-	remove_wait_queue(&sem->wait, &wait);
-
-void __sched __down(struct semaphore * sem)
-{
-	DECLARE_WAITQUEUE(wait, current);
-
-	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
-	if (waking_non_zero(sem))
-		break;
-	schedule();
-	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	int ret = 0;
-
-	DOWN_HEAD(TASK_INTERRUPTIBLE)
-
-	ret = waking_non_zero_interruptible(sem, current);
-	if (ret)
-	{
-		if (ret == 1)
-			/* ret != 0 only if we get interrupted -arca */
-			ret = 0;
-		break;
-	}
-	schedule();
-	DOWN_TAIL(TASK_INTERRUPTIBLE)
-	return ret;
-}
-
-int __down_trylock(struct semaphore * sem)
-{
-	return waking_non_zero_trylock(sem);
-}
diff --git a/arch/m68k/lib/Makefile b/arch/m68k/lib/Makefile
index 6bbf19f96007..a18af095cd7c 100644
--- a/arch/m68k/lib/Makefile
+++ b/arch/m68k/lib/Makefile
@@ -5,4 +5,4 @@
 EXTRA_AFLAGS := -traditional
 
 lib-y	:= ashldi3.o ashrdi3.o lshrdi3.o muldi3.o \
-	   checksum.o string.o semaphore.o uaccess.o
+	   checksum.o string.o uaccess.o
diff --git a/arch/m68k/lib/semaphore.S b/arch/m68k/lib/semaphore.S
deleted file mode 100644
index 0215624c1602..000000000000
--- a/arch/m68k/lib/semaphore.S
+++ /dev/null
@@ -1,53 +0,0 @@
-/*
- *  linux/arch/m68k/lib/semaphore.S
- *
- *  Copyright (C) 1996  Linus Torvalds
- *
- *  m68k version by Andreas Schwab
- */
-
-#include <linux/linkage.h>
-#include <asm/semaphore.h>
-
-/*
- * The semaphore operations have a special calling sequence that
- * allow us to do a simpler in-line version of them. These routines
- * need to convert that sequence back into the C sequence when
- * there is contention on the semaphore.
- */
-ENTRY(__down_failed)
-	moveml %a0/%d0/%d1,-(%sp)
-	movel %a1,-(%sp)
-	jbsr __down
-	movel (%sp)+,%a1
-	moveml (%sp)+,%a0/%d0/%d1
-	rts
-
-ENTRY(__down_failed_interruptible)
-	movel %a0,-(%sp)
-	movel %d1,-(%sp)
-	movel %a1,-(%sp)
-	jbsr __down_interruptible
-	movel (%sp)+,%a1
-	movel (%sp)+,%d1
-	movel (%sp)+,%a0
-	rts
-
-ENTRY(__down_failed_trylock)
-	movel %a0,-(%sp)
-	movel %d1,-(%sp)
-	movel %a1,-(%sp)
-	jbsr __down_trylock
-	movel (%sp)+,%a1
-	movel (%sp)+,%d1
-	movel (%sp)+,%a0
-	rts
-
-ENTRY(__up_wakeup)
-	moveml %a0/%d0/%d1,-(%sp)
-	movel %a1,-(%sp)
-	jbsr __up
-	movel (%sp)+,%a1
-	moveml (%sp)+,%a0/%d0/%d1
-	rts
-
diff --git a/arch/m68knommu/kernel/Makefile b/arch/m68knommu/kernel/Makefile
index 1524b39ad63f..f0eab3dedb5a 100644
--- a/arch/m68knommu/kernel/Makefile
+++ b/arch/m68knommu/kernel/Makefile
@@ -5,7 +5,7 @@
 extra-y := vmlinux.lds
 
 obj-y += dma.o entry.o init_task.o irq.o m68k_ksyms.o process.o ptrace.o \
-	 semaphore.o setup.o signal.o syscalltable.o sys_m68k.o time.o traps.o
+	 setup.o signal.o syscalltable.o sys_m68k.o time.o traps.o
 
 obj-$(CONFIG_MODULES)	+= module.o
 obj-$(CONFIG_COMEMPCI)	+= comempci.o
diff --git a/arch/m68knommu/kernel/m68k_ksyms.c b/arch/m68knommu/kernel/m68k_ksyms.c
index 53fad1490282..39fe0a7aec32 100644
--- a/arch/m68knommu/kernel/m68k_ksyms.c
+++ b/arch/m68knommu/kernel/m68k_ksyms.c
@@ -13,7 +13,6 @@
 #include <asm/pgalloc.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/semaphore.h>
 #include <asm/checksum.h>
 #include <asm/current.h>
 
@@ -39,11 +38,6 @@ EXPORT_SYMBOL(csum_partial_copy_nocheck);
 EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(memset);
 
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_failed_interruptible);
-EXPORT_SYMBOL(__down_failed_trylock);
-EXPORT_SYMBOL(__up_wakeup);
-
 /*
  * libgcc functions - functions that are used internally by the
  * compiler...  (prototypes are not correct though, but that
diff --git a/arch/m68knommu/kernel/semaphore.c b/arch/m68knommu/kernel/semaphore.c
deleted file mode 100644
index bce2bc7d87c6..000000000000
--- a/arch/m68knommu/kernel/semaphore.c
+++ /dev/null
@@ -1,133 +0,0 @@
-/*
- *  Generic semaphore code. Buyer beware. Do your own
- * specific changes in <asm/semaphore-helper.h>
- */
-
-#include <linux/sched.h>
-#include <linux/err.h>
-#include <linux/init.h>
-#include <asm/semaphore-helper.h>
-
-#ifndef CONFIG_RMW_INSNS
-spinlock_t semaphore_wake_lock;
-#endif
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to sleep, while the "waking" variable is
- * incremented when the "up()" code goes to wake up waiting
- * processes.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * waking_non_zero() (from asm/semaphore.h) must execute
- * atomically.
- *
- * When __up() is called, the count was negative before
- * incrementing it, and we need to wake up somebody.
- *
- * This routine adds one to the count of processes that need to
- * wake up and exit.  ALL waiting processes actually wake up but
- * only the one that gets to the "waking" field first will gate
- * through and acquire the semaphore.  The others will go back
- * to sleep.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-void __up(struct semaphore *sem)
-{
-	wake_one_more(sem);
-	wake_up(&sem->wait);
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from __down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from __down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- *
- */
-
-
-#define DOWN_HEAD(task_state)						\
-									\
-									\
-	current->state = (task_state);					\
-	add_wait_queue(&sem->wait, &wait);				\
-									\
-	/*								\
-	 * Ok, we're set up.  sem->count is known to be less than zero	\
-	 * so we must wait.						\
-	 *								\
-	 * We can let go the lock for purposes of waiting.		\
-	 * We re-acquire it after awaking so as to protect		\
-	 * all semaphore operations.					\
-	 *								\
-	 * If "up()" is called before we call waking_non_zero() then	\
-	 * we will catch it right away.  If it is called later then	\
-	 * we will have to go through a wakeup cycle to catch it.	\
-	 *								\
-	 * Multiple waiters contend for the semaphore lock to see	\
-	 * who gets to gate through and who has to wait some more.	\
-	 */								\
-	for (;;) {
-
-#define DOWN_TAIL(task_state)			\
-		current->state = (task_state);	\
-	}					\
-	current->state = TASK_RUNNING;		\
-	remove_wait_queue(&sem->wait, &wait);
-
-void __sched __down(struct semaphore * sem)
-{
-	DECLARE_WAITQUEUE(wait, current);
-
-	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
-	if (waking_non_zero(sem))
-		break;
-	schedule();
-	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	int ret = 0;
-
-	DOWN_HEAD(TASK_INTERRUPTIBLE)
-
-	ret = waking_non_zero_interruptible(sem, current);
-	if (ret)
-	{
-		if (ret == 1)
-			/* ret != 0 only if we get interrupted -arca */
-			ret = 0;
-		break;
-	}
-	schedule();
-	DOWN_TAIL(TASK_INTERRUPTIBLE)
-	return ret;
-}
-
-int __down_trylock(struct semaphore * sem)
-{
-	return waking_non_zero_trylock(sem);
-}
diff --git a/arch/m68knommu/lib/Makefile b/arch/m68knommu/lib/Makefile
index e051a7913987..d94d709665aa 100644
--- a/arch/m68knommu/lib/Makefile
+++ b/arch/m68knommu/lib/Makefile
@@ -4,4 +4,4 @@
 
 lib-y	:= ashldi3.o ashrdi3.o lshrdi3.o \
 	   muldi3.o mulsi3.o divsi3.o udivsi3.o modsi3.o umodsi3.o \
-	   checksum.o semaphore.o memcpy.o memset.o delay.o
+	   checksum.o memcpy.o memset.o delay.o
diff --git a/arch/m68knommu/lib/semaphore.S b/arch/m68knommu/lib/semaphore.S
deleted file mode 100644
index 87c746034376..000000000000
--- a/arch/m68knommu/lib/semaphore.S
+++ /dev/null
@@ -1,66 +0,0 @@
-/*
- *  linux/arch/m68k/lib/semaphore.S
- *
- *  Copyright (C) 1996  Linus Torvalds
- *
- *  m68k version by Andreas Schwab
- *
- *  MAR/1999 -- modified to support ColdFire (gerg@snapgear.com)
- */
-
-#include <linux/linkage.h>
-#include <asm/semaphore.h>
-
-/*
- * "down_failed" is called with the eventual return address
- * in %a0, and the address of the semaphore in %a1. We need
- * to increment the number of waiters on the semaphore,
- * call "__down()", and then eventually return to try again.
- */
-ENTRY(__down_failed)
-#ifdef CONFIG_COLDFIRE
-	subl #12,%sp
-	moveml %a0/%d0/%d1,(%sp)
-#else
-	moveml %a0/%d0/%d1,-(%sp)
-#endif
-	movel %a1,-(%sp)
-	jbsr __down
-	movel (%sp)+,%a1
-	movel (%sp)+,%d0
-	movel (%sp)+,%d1
-	rts
-
-ENTRY(__down_failed_interruptible)
-	movel %a0,-(%sp)
-	movel %d1,-(%sp)
-	movel %a1,-(%sp)
-	jbsr __down_interruptible
-	movel (%sp)+,%a1
-	movel (%sp)+,%d1
-	rts
-
-ENTRY(__up_wakeup)
-#ifdef CONFIG_COLDFIRE
-	subl #12,%sp
-	moveml %a0/%d0/%d1,(%sp)
-#else
-	moveml %a0/%d0/%d1,-(%sp)
-#endif
-	movel %a1,-(%sp)
-	jbsr __up
-	movel (%sp)+,%a1
-	movel (%sp)+,%d0
-	movel (%sp)+,%d1
-	rts
-
-ENTRY(__down_failed_trylock)
-	movel %a0,-(%sp)
-	movel %d1,-(%sp)
-	movel %a1,-(%sp)
-	jbsr __down_trylock
-	movel (%sp)+,%a1
-	movel (%sp)+,%d1
-	movel (%sp)+,%a0
-	rts
-
diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile
index 9e78e1a4ca17..6fcdb6fda2e2 100644
--- a/arch/mips/kernel/Makefile
+++ b/arch/mips/kernel/Makefile
@@ -5,7 +5,7 @@
 extra-y		:= head.o init_task.o vmlinux.lds
 
 obj-y		+= cpu-probe.o branch.o entry.o genex.o irq.o process.o \
-		   ptrace.o reset.o semaphore.o setup.o signal.o syscall.o \
+		   ptrace.o reset.o setup.o signal.o syscall.o \
 		   time.o topology.o traps.o unaligned.o
 
 obj-$(CONFIG_CEVT_BCM1480)	+= cevt-bcm1480.o
diff --git a/arch/mips/kernel/semaphore.c b/arch/mips/kernel/semaphore.c
deleted file mode 100644
index 1265358cdca1..000000000000
--- a/arch/mips/kernel/semaphore.c
+++ /dev/null
@@ -1,168 +0,0 @@
-/*
- * MIPS-specific semaphore code.
- *
- * Copyright (C) 1999 Cort Dougan <cort@cs.nmt.edu>
- * Copyright (C) 2004 Ralf Baechle <ralf@linux-mips.org>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- *
- * April 2001 - Reworked by Paul Mackerras <paulus@samba.org>
- * to eliminate the SMP races in the old version between the updates
- * of `count' and `waking'.  Now we use negative `count' values to
- * indicate that some process(es) are waiting for the semaphore.
- */
-
-#include <linux/module.h>
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <asm/atomic.h>
-#include <asm/cpu-features.h>
-#include <asm/errno.h>
-#include <asm/semaphore.h>
-#include <asm/war.h>
-/*
- * Atomically update sem->count.
- * This does the equivalent of the following:
- *
- *	old_count = sem->count;
- *	tmp = MAX(old_count, 0) + incr;
- *	sem->count = tmp;
- *	return old_count;
- *
- * On machines without lld/scd we need a spinlock to make the manipulation of
- * sem->count and sem->waking atomic.  Scalability isn't an issue because
- * this lock is used on UP only so it's just an empty variable.
- */
-static inline int __sem_update_count(struct semaphore *sem, int incr)
-{
-	int old_count, tmp;
-
-	if (cpu_has_llsc && R10000_LLSC_WAR) {
-		__asm__ __volatile__(
-		"	.set	mips3					\n"
-		"1:	ll	%0, %2		# __sem_update_count	\n"
-		"	sra	%1, %0, 31				\n"
-		"	not	%1					\n"
-		"	and	%1, %0, %1				\n"
-		"	addu	%1, %1, %3				\n"
-		"	sc	%1, %2					\n"
-		"	beqzl	%1, 1b					\n"
-		"	.set	mips0					\n"
-		: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
-		: "r" (incr), "m" (sem->count));
-	} else if (cpu_has_llsc) {
-		__asm__ __volatile__(
-		"	.set	mips3					\n"
-		"1:	ll	%0, %2		# __sem_update_count	\n"
-		"	sra	%1, %0, 31				\n"
-		"	not	%1					\n"
-		"	and	%1, %0, %1				\n"
-		"	addu	%1, %1, %3				\n"
-		"	sc	%1, %2					\n"
-		"	beqz	%1, 1b					\n"
-		"	.set	mips0					\n"
-		: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
-		: "r" (incr), "m" (sem->count));
-	} else {
-		static DEFINE_SPINLOCK(semaphore_lock);
-		unsigned long flags;
-
-		spin_lock_irqsave(&semaphore_lock, flags);
-		old_count = atomic_read(&sem->count);
-		tmp = max_t(int, old_count, 0) + incr;
-		atomic_set(&sem->count, tmp);
-		spin_unlock_irqrestore(&semaphore_lock, flags);
-	}
-
-	return old_count;
-}
-
-void __up(struct semaphore *sem)
-{
-	/*
-	 * Note that we incremented count in up() before we came here,
-	 * but that was ineffective since the result was <= 0, and
-	 * any negative value of count is equivalent to 0.
-	 * This ends up setting count to 1, unless count is now > 0
-	 * (i.e. because some other cpu has called up() in the meantime),
-	 * in which case we just increment count.
-	 */
-	__sem_update_count(sem, 1);
-	wake_up(&sem->wait);
-}
-
-EXPORT_SYMBOL(__up);
-
-/*
- * Note that when we come in to __down or __down_interruptible,
- * we have already decremented count, but that decrement was
- * ineffective since the result was < 0, and any negative value
- * of count is equivalent to 0.
- * Thus it is only when we decrement count from some value > 0
- * that we have actually got the semaphore.
- */
-void __sched __down(struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	__set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	/*
-	 * Try to get the semaphore.  If the count is > 0, then we've
-	 * got the semaphore; we decrement count and exit the loop.
-	 * If the count is 0 or negative, we set it to -1, indicating
-	 * that we are asleep, and then sleep.
-	 */
-	while (__sem_update_count(sem, -1) <= 0) {
-		schedule();
-		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	__set_task_state(tsk, TASK_RUNNING);
-
-	/*
-	 * If there are any more sleepers, wake one of them up so
-	 * that it can either get the semaphore, or set count to -1
-	 * indicating that there are still processes sleeping.
-	 */
-	wake_up(&sem->wait);
-}
-
-EXPORT_SYMBOL(__down);
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	__set_task_state(tsk, TASK_INTERRUPTIBLE);
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	while (__sem_update_count(sem, -1) <= 0) {
-		if (signal_pending(current)) {
-			/*
-			 * A signal is pending - give up trying.
-			 * Set sem->count to 0 if it is negative,
-			 * since we are no longer sleeping.
-			 */
-			__sem_update_count(sem, 0);
-			retval = -EINTR;
-			break;
-		}
-		schedule();
-		set_task_state(tsk, TASK_INTERRUPTIBLE);
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	__set_task_state(tsk, TASK_RUNNING);
-
-	wake_up(&sem->wait);
-	return retval;
-}
-
-EXPORT_SYMBOL(__down_interruptible);
diff --git a/arch/mn10300/kernel/Makefile b/arch/mn10300/kernel/Makefile
index ef07c956170a..23f2ab67574c 100644
--- a/arch/mn10300/kernel/Makefile
+++ b/arch/mn10300/kernel/Makefile
@@ -3,7 +3,7 @@
 #
 extra-y := head.o init_task.o vmlinux.lds
 
-obj-y   := process.o semaphore.o signal.o entry.o fpu.o traps.o irq.o \
+obj-y   := process.o signal.o entry.o fpu.o traps.o irq.o \
 	   ptrace.o setup.o time.o sys_mn10300.o io.o kthread.o \
 	   switch_to.o mn10300_ksyms.o kernel_execve.o
 
diff --git a/arch/mn10300/kernel/semaphore.c b/arch/mn10300/kernel/semaphore.c
deleted file mode 100644
index 9153c4039fd2..000000000000
--- a/arch/mn10300/kernel/semaphore.c
+++ /dev/null
@@ -1,149 +0,0 @@
-/* MN10300 Semaphore implementation
- *
- * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-#include <linux/sched.h>
-#include <linux/module.h>
-#include <asm/semaphore.h>
-
-struct sem_waiter {
-	struct list_head	list;
-	struct task_struct	*task;
-};
-
-#if SEMAPHORE_DEBUG
-void semtrace(struct semaphore *sem, const char *str)
-{
-	if (sem->debug)
-		printk(KERN_DEBUG "[%d] %s({%d,%d})\n",
-		       current->pid,
-		       str,
-		       atomic_read(&sem->count),
-		       list_empty(&sem->wait_list) ? 0 : 1);
-}
-#else
-#define semtrace(SEM, STR) do { } while (0)
-#endif
-
-/*
- * wait for a token to be granted from a semaphore
- * - entered with lock held and interrupts disabled
- */
-void __down(struct semaphore *sem, unsigned long flags)
-{
-	struct task_struct *tsk = current;
-	struct sem_waiter waiter;
-
-	semtrace(sem, "Entering __down");
-
-	/* set up my own style of waitqueue */
-	waiter.task = tsk;
-	get_task_struct(tsk);
-
-	list_add_tail(&waiter.list, &sem->wait_list);
-
-	/* we don't need to touch the semaphore struct anymore */
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-
-	/* wait to be given the semaphore */
-	set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-
-	for (;;) {
-		if (!waiter.task)
-			break;
-		schedule();
-		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	}
-
-	tsk->state = TASK_RUNNING;
-	semtrace(sem, "Leaving __down");
-}
-EXPORT_SYMBOL(__down);
-
-/*
- * interruptibly wait for a token to be granted from a semaphore
- * - entered with lock held and interrupts disabled
- */
-int __down_interruptible(struct semaphore *sem, unsigned long flags)
-{
-	struct task_struct *tsk = current;
-	struct sem_waiter waiter;
-	int ret;
-
-	semtrace(sem, "Entering __down_interruptible");
-
-	/* set up my own style of waitqueue */
-	waiter.task = tsk;
-	get_task_struct(tsk);
-
-	list_add_tail(&waiter.list, &sem->wait_list);
-
-	/* we don't need to touch the semaphore struct anymore */
-	set_task_state(tsk, TASK_INTERRUPTIBLE);
-
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-
-	/* wait to be given the semaphore */
-	ret = 0;
-	for (;;) {
-		if (!waiter.task)
-			break;
-		if (unlikely(signal_pending(current)))
-			goto interrupted;
-		schedule();
-		set_task_state(tsk, TASK_INTERRUPTIBLE);
-	}
-
- out:
-	tsk->state = TASK_RUNNING;
-	semtrace(sem, "Leaving __down_interruptible");
-	return ret;
-
- interrupted:
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	list_del(&waiter.list);
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-
-	ret = 0;
-	if (!waiter.task) {
-		put_task_struct(current);
-		ret = -EINTR;
-	}
-	goto out;
-}
-EXPORT_SYMBOL(__down_interruptible);
-
-/*
- * release a single token back to a semaphore
- * - entered with lock held and interrupts disabled
- */
-void __up(struct semaphore *sem)
-{
-	struct task_struct *tsk;
-	struct sem_waiter *waiter;
-
-	semtrace(sem, "Entering __up");
-
-	/* grant the token to the process at the front of the queue */
-	waiter = list_entry(sem->wait_list.next, struct sem_waiter, list);
-
-	/* We must be careful not to touch 'waiter' after we set ->task = NULL.
-	 * It is an allocated on the waiter's stack and may become invalid at
-	 * any time after that point (due to a wakeup from another source).
-	 */
-	list_del_init(&waiter->list);
-	tsk = waiter->task;
-	smp_mb();
-	waiter->task = NULL;
-	wake_up_process(tsk);
-	put_task_struct(tsk);
-
-	semtrace(sem, "Leaving __up");
-}
-EXPORT_SYMBOL(__up);
diff --git a/arch/parisc/kernel/Makefile b/arch/parisc/kernel/Makefile
index 27827bc3717e..1f6585a56f97 100644
--- a/arch/parisc/kernel/Makefile
+++ b/arch/parisc/kernel/Makefile
@@ -9,7 +9,7 @@ AFLAGS_pacache.o := -traditional
 
 obj-y	     	:= cache.o pacache.o setup.o traps.o time.o irq.o \
 		   pa7300lc.o syscall.o entry.o sys_parisc.o firmware.o \
-		   ptrace.o hardware.o inventory.o drivers.o semaphore.o \
+		   ptrace.o hardware.o inventory.o drivers.o \
 		   signal.o hpmc.o real2.o parisc_ksyms.o unaligned.o \
 		   process.o processor.o pdc_cons.o pdc_chassis.o unwind.o \
 		   topology.o
diff --git a/arch/parisc/kernel/parisc_ksyms.c b/arch/parisc/kernel/parisc_ksyms.c
index 7aca704e96f0..5b7fc4aa044d 100644
--- a/arch/parisc/kernel/parisc_ksyms.c
+++ b/arch/parisc/kernel/parisc_ksyms.c
@@ -69,11 +69,6 @@ EXPORT_SYMBOL(memcpy_toio);
 EXPORT_SYMBOL(memcpy_fromio);
 EXPORT_SYMBOL(memset_io);
 
-#include <asm/semaphore.h>
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down_interruptible);
-EXPORT_SYMBOL(__down);
-
 extern void $$divI(void);
 extern void $$divU(void);
 extern void $$remI(void);
diff --git a/arch/parisc/kernel/semaphore.c b/arch/parisc/kernel/semaphore.c
deleted file mode 100644
index ee806bcc3726..000000000000
--- a/arch/parisc/kernel/semaphore.c
+++ /dev/null
@@ -1,102 +0,0 @@
-/*
- * Semaphore implementation Copyright (c) 2001 Matthew Wilcox, Hewlett-Packard
- */
-
-#include <linux/sched.h>
-#include <linux/spinlock.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-
-/*
- * Semaphores are complex as we wish to avoid using two variables.
- * `count' has multiple roles, depending on its value.  If it is positive
- * or zero, there are no waiters.  The functions here will never be
- * called; see <asm/semaphore.h>
- *
- * When count is -1 it indicates there is at least one task waiting
- * for the semaphore.
- *
- * When count is less than that, there are '- count - 1' wakeups
- * pending.  ie if it has value -3, there are 2 wakeups pending.
- *
- * Note that these functions are only called when there is contention
- * on the lock, and as such all this is the "non-critical" part of the
- * whole semaphore business. The critical part is the inline stuff in
- * <asm/semaphore.h> where we want to avoid any extra jumps and calls.
- */
-void __up(struct semaphore *sem)
-{
-	sem->count--;
-	wake_up(&sem->wait);
-}
-
-#define wakers(count) (-1 - count)
-
-#define DOWN_HEAD							\
-	int ret = 0;							\
-	DECLARE_WAITQUEUE(wait, current);				\
-									\
-	/* Note that someone is waiting */				\
-	if (sem->count == 0)						\
-		sem->count = -1;					\
-									\
-	/* protected by the sentry still -- use unlocked version */	\
-	wait.flags = WQ_FLAG_EXCLUSIVE;					\
-	__add_wait_queue_tail(&sem->wait, &wait);			\
- lost_race:								\
-	spin_unlock_irq(&sem->sentry);					\
-
-#define DOWN_TAIL							\
-	spin_lock_irq(&sem->sentry);					\
-	if (wakers(sem->count) == 0 && ret == 0)			\
-		goto lost_race;	/* Someone stole our wakeup */		\
-	__remove_wait_queue(&sem->wait, &wait);				\
-	current->state = TASK_RUNNING;					\
-	if (!waitqueue_active(&sem->wait) && (sem->count < 0))		\
-		sem->count = wakers(sem->count);
-
-#define UPDATE_COUNT							\
-	sem->count += (sem->count < 0) ? 1 : - 1;
-	
-
-void __sched __down(struct semaphore * sem)
-{
-	DOWN_HEAD
-
-	for(;;) {
-		set_task_state(current, TASK_UNINTERRUPTIBLE);
-		/* we can _read_ this without the sentry */
-		if (sem->count != -1)
-			break;
- 		schedule();
- 	}
-
-	DOWN_TAIL
-	UPDATE_COUNT
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	DOWN_HEAD
-
-	for(;;) {
-		set_task_state(current, TASK_INTERRUPTIBLE);
-		/* we can _read_ this without the sentry */
-		if (sem->count != -1)
-			break;
-
-		if (signal_pending(current)) {
-			ret = -EINTR;
-			break;
-		}
-		schedule();
-	}
-
-	DOWN_TAIL
-
-	if (!ret) {
-		UPDATE_COUNT
-	}
-
-	return ret;
-}
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index c1baf9d5903f..b9dbfff9afe9 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -12,7 +12,7 @@ CFLAGS_prom_init.o      += -fPIC
 CFLAGS_btext.o		+= -fPIC
 endif
 
-obj-y				:= semaphore.o cputable.o ptrace.o syscalls.o \
+obj-y				:= cputable.o ptrace.o syscalls.o \
 				   irq.o align.o signal_32.o pmc.o vdso.o \
 				   init_task.o process.o systbl.o idle.o \
 				   signal.o
diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c
index 9c98424277a8..65d14e6ddc3c 100644
--- a/arch/powerpc/kernel/ppc_ksyms.c
+++ b/arch/powerpc/kernel/ppc_ksyms.c
@@ -15,7 +15,6 @@
 #include <linux/bitops.h>
 
 #include <asm/page.h>
-#include <asm/semaphore.h>
 #include <asm/processor.h>
 #include <asm/cacheflush.h>
 #include <asm/uaccess.h>
diff --git a/arch/powerpc/kernel/semaphore.c b/arch/powerpc/kernel/semaphore.c
deleted file mode 100644
index 2f8c3c951394..000000000000
--- a/arch/powerpc/kernel/semaphore.c
+++ /dev/null
@@ -1,135 +0,0 @@
-/*
- * PowerPC-specific semaphore code.
- *
- * Copyright (C) 1999 Cort Dougan <cort@cs.nmt.edu>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- *
- * April 2001 - Reworked by Paul Mackerras <paulus@samba.org>
- * to eliminate the SMP races in the old version between the updates
- * of `count' and `waking'.  Now we use negative `count' values to
- * indicate that some process(es) are waiting for the semaphore.
- */
-
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <linux/module.h>
-
-#include <asm/atomic.h>
-#include <asm/semaphore.h>
-#include <asm/errno.h>
-
-/*
- * Atomically update sem->count.
- * This does the equivalent of the following:
- *
- *	old_count = sem->count;
- *	tmp = MAX(old_count, 0) + incr;
- *	sem->count = tmp;
- *	return old_count;
- */
-static inline int __sem_update_count(struct semaphore *sem, int incr)
-{
-	int old_count, tmp;
-
-	__asm__ __volatile__("\n"
-"1:	lwarx	%0,0,%3\n"
-"	srawi	%1,%0,31\n"
-"	andc	%1,%0,%1\n"
-"	add	%1,%1,%4\n"
-	PPC405_ERR77(0,%3)
-"	stwcx.	%1,0,%3\n"
-"	bne	1b"
-	: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
-	: "r" (&sem->count), "r" (incr), "m" (sem->count)
-	: "cc");
-
-	return old_count;
-}
-
-void __up(struct semaphore *sem)
-{
-	/*
-	 * Note that we incremented count in up() before we came here,
-	 * but that was ineffective since the result was <= 0, and
-	 * any negative value of count is equivalent to 0.
-	 * This ends up setting count to 1, unless count is now > 0
-	 * (i.e. because some other cpu has called up() in the meantime),
-	 * in which case we just increment count.
-	 */
-	__sem_update_count(sem, 1);
-	wake_up(&sem->wait);
-}
-EXPORT_SYMBOL(__up);
-
-/*
- * Note that when we come in to __down or __down_interruptible,
- * we have already decremented count, but that decrement was
- * ineffective since the result was < 0, and any negative value
- * of count is equivalent to 0.
- * Thus it is only when we decrement count from some value > 0
- * that we have actually got the semaphore.
- */
-void __sched __down(struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	__set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	/*
-	 * Try to get the semaphore.  If the count is > 0, then we've
-	 * got the semaphore; we decrement count and exit the loop.
-	 * If the count is 0 or negative, we set it to -1, indicating
-	 * that we are asleep, and then sleep.
-	 */
-	while (__sem_update_count(sem, -1) <= 0) {
-		schedule();
-		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	__set_task_state(tsk, TASK_RUNNING);
-
-	/*
-	 * If there are any more sleepers, wake one of them up so
-	 * that it can either get the semaphore, or set count to -1
-	 * indicating that there are still processes sleeping.
-	 */
-	wake_up(&sem->wait);
-}
-EXPORT_SYMBOL(__down);
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	__set_task_state(tsk, TASK_INTERRUPTIBLE);
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	while (__sem_update_count(sem, -1) <= 0) {
-		if (signal_pending(current)) {
-			/*
-			 * A signal is pending - give up trying.
-			 * Set sem->count to 0 if it is negative,
-			 * since we are no longer sleeping.
-			 */
-			__sem_update_count(sem, 0);
-			retval = -EINTR;
-			break;
-		}
-		schedule();
-		set_task_state(tsk, TASK_INTERRUPTIBLE);
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	__set_task_state(tsk, TASK_RUNNING);
-
-	wake_up(&sem->wait);
-	return retval;
-}
-EXPORT_SYMBOL(__down_interruptible);
diff --git a/arch/ppc/kernel/semaphore.c b/arch/ppc/kernel/semaphore.c
deleted file mode 100644
index 2fe429b27c14..000000000000
--- a/arch/ppc/kernel/semaphore.c
+++ /dev/null
@@ -1,131 +0,0 @@
-/*
- * PowerPC-specific semaphore code.
- *
- * Copyright (C) 1999 Cort Dougan <cort@cs.nmt.edu>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- *
- * April 2001 - Reworked by Paul Mackerras <paulus@samba.org>
- * to eliminate the SMP races in the old version between the updates
- * of `count' and `waking'.  Now we use negative `count' values to
- * indicate that some process(es) are waiting for the semaphore.
- */
-
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <asm/atomic.h>
-#include <asm/semaphore.h>
-#include <asm/errno.h>
-
-/*
- * Atomically update sem->count.
- * This does the equivalent of the following:
- *
- *	old_count = sem->count;
- *	tmp = MAX(old_count, 0) + incr;
- *	sem->count = tmp;
- *	return old_count;
- */
-static inline int __sem_update_count(struct semaphore *sem, int incr)
-{
-	int old_count, tmp;
-
-	__asm__ __volatile__("\n"
-"1:	lwarx	%0,0,%3\n"
-"	srawi	%1,%0,31\n"
-"	andc	%1,%0,%1\n"
-"	add	%1,%1,%4\n"
-	PPC405_ERR77(0,%3)
-"	stwcx.	%1,0,%3\n"
-"	bne	1b"
-	: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
-	: "r" (&sem->count), "r" (incr), "m" (sem->count)
-	: "cc");
-
-	return old_count;
-}
-
-void __up(struct semaphore *sem)
-{
-	/*
-	 * Note that we incremented count in up() before we came here,
-	 * but that was ineffective since the result was <= 0, and
-	 * any negative value of count is equivalent to 0.
-	 * This ends up setting count to 1, unless count is now > 0
-	 * (i.e. because some other cpu has called up() in the meantime),
-	 * in which case we just increment count.
-	 */
-	__sem_update_count(sem, 1);
-	wake_up(&sem->wait);
-}
-
-/*
- * Note that when we come in to __down or __down_interruptible,
- * we have already decremented count, but that decrement was
- * ineffective since the result was < 0, and any negative value
- * of count is equivalent to 0.
- * Thus it is only when we decrement count from some value > 0
- * that we have actually got the semaphore.
- */
-void __sched __down(struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-	smp_wmb();
-
-	/*
-	 * Try to get the semaphore.  If the count is > 0, then we've
-	 * got the semaphore; we decrement count and exit the loop.
-	 * If the count is 0 or negative, we set it to -1, indicating
-	 * that we are asleep, and then sleep.
-	 */
-	while (__sem_update_count(sem, -1) <= 0) {
-		schedule();
-		tsk->state = TASK_UNINTERRUPTIBLE;
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
-
-	/*
-	 * If there are any more sleepers, wake one of them up so
-	 * that it can either get the semaphore, or set count to -1
-	 * indicating that there are still processes sleeping.
-	 */
-	wake_up(&sem->wait);
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	tsk->state = TASK_INTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-	smp_wmb();
-
-	while (__sem_update_count(sem, -1) <= 0) {
-		if (signal_pending(current)) {
-			/*
-			 * A signal is pending - give up trying.
-			 * Set sem->count to 0 if it is negative,
-			 * since we are no longer sleeping.
-			 */
-			__sem_update_count(sem, 0);
-			retval = -EINTR;
-			break;
-		}
-		schedule();
-		tsk->state = TASK_INTERRUPTIBLE;
-	}
-	tsk->state = TASK_RUNNING;
-	remove_wait_queue(&sem->wait, &wait);
-	wake_up(&sem->wait);
-	return retval;
-}
diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index 4d3e38392cb1..ce144b67f060 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -11,7 +11,7 @@ CFLAGS_smp.o	:= -Wno-nonnull
 
 obj-y	:=  bitmap.o traps.o time.o process.o base.o early.o \
             setup.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \
-	    semaphore.o s390_ext.o debug.o irq.o ipl.o dis.o diag.o
+	    s390_ext.o debug.o irq.o ipl.o dis.o diag.o
 
 obj-y	+= $(if $(CONFIG_64BIT),entry64.o,entry.o)
 obj-y	+= $(if $(CONFIG_64BIT),reipl64.o,reipl.o)
diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c
index 7234c737f825..48238a114ce9 100644
--- a/arch/s390/kernel/s390_ksyms.c
+++ b/arch/s390/kernel/s390_ksyms.c
@@ -26,13 +26,6 @@ EXPORT_SYMBOL(_ni_bitmap);
 EXPORT_SYMBOL(_zb_findmap);
 EXPORT_SYMBOL(_sb_findmap);
 
-/*
- * semaphore ops
- */
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_interruptible);
-
 /*
  * binfmt_elf loader 
  */
diff --git a/arch/s390/kernel/semaphore.c b/arch/s390/kernel/semaphore.c
deleted file mode 100644
index 191303f6c1d8..000000000000
--- a/arch/s390/kernel/semaphore.c
+++ /dev/null
@@ -1,108 +0,0 @@
-/*
- *  linux/arch/s390/kernel/semaphore.c
- *
- *  S390 version
- *    Copyright (C) 1998-2000 IBM Corporation
- *    Author(s): Martin Schwidefsky
- *
- *  Derived from "linux/arch/i386/kernel/semaphore.c
- *    Copyright (C) 1999, Linus Torvalds
- *
- */
-#include <linux/sched.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-
-#include <asm/semaphore.h>
-
-/*
- * Atomically update sem->count. Equivalent to:
- *   old_val = sem->count.counter;
- *   new_val = ((old_val >= 0) ? old_val : 0) + incr;
- *   sem->count.counter = new_val;
- *   return old_val;
- */
-static inline int __sem_update_count(struct semaphore *sem, int incr)
-{
-	int old_val, new_val;
-
-	asm volatile(
-		"	l	%0,0(%3)\n"
-		"0:	ltr	%1,%0\n"
-		"	jhe	1f\n"
-		"	lhi	%1,0\n"
-		"1:	ar	%1,%4\n"
-		"	cs	%0,%1,0(%3)\n"
-		"	jl	0b\n"
-		: "=&d" (old_val), "=&d" (new_val), "=m" (sem->count)
-		: "a" (&sem->count), "d" (incr), "m" (sem->count)
-		: "cc");
-	return old_val;
-}
-
-/*
- * The inline function up() incremented count but the result
- * was <= 0. This indicates that some process is waiting on
- * the semaphore. The semaphore is free and we'll wake the
- * first sleeping process, so we set count to 1 unless some
- * other cpu has called up in the meantime in which case
- * we just increment count by 1.
- */
-void __up(struct semaphore *sem)
-{
-	__sem_update_count(sem, 1);
-	wake_up(&sem->wait);
-}
-
-/*
- * The inline function down() decremented count and the result
- * was < 0. The wait loop will atomically test and update the
- * semaphore counter following the rules:
- *   count > 0: decrement count, wake up queue and exit.
- *   count <= 0: set count to -1, go to sleep.
- */
-void __sched __down(struct semaphore * sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	__set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	add_wait_queue_exclusive(&sem->wait, &wait);
-	while (__sem_update_count(sem, -1) <= 0) {
-		schedule();
-		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	__set_task_state(tsk, TASK_RUNNING);
-	wake_up(&sem->wait);
-}
-
-/*
- * Same as __down() with an additional test for signals.
- * If a signal is pending the count is updated as follows:
- *   count > 0: wake up queue and exit.
- *   count <= 0: set count to 0, wake up queue and exit.
- */
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	__set_task_state(tsk, TASK_INTERRUPTIBLE);
-	add_wait_queue_exclusive(&sem->wait, &wait);
-	while (__sem_update_count(sem, -1) <= 0) {
-		if (signal_pending(current)) {
-			__sem_update_count(sem, 0);
-			retval = -EINTR;
-			break;
-		}
-		schedule();
-		set_task_state(tsk, TASK_INTERRUPTIBLE);
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	__set_task_state(tsk, TASK_RUNNING);
-	wake_up(&sem->wait);
-	return retval;
-}
-
diff --git a/arch/sh/kernel/Makefile_32 b/arch/sh/kernel/Makefile_32
index 62bf373266f7..4bbdce36b92b 100644
--- a/arch/sh/kernel/Makefile_32
+++ b/arch/sh/kernel/Makefile_32
@@ -5,7 +5,7 @@
 extra-y	:= head_32.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_32.o \
-	   ptrace_32.o semaphore.o setup.o signal_32.o sys_sh.o sys_sh32.o \
+	   ptrace_32.o setup.o signal_32.o sys_sh.o sys_sh32.o \
 	   syscalls_32.o time_32.o topology.o traps.o traps_32.o
 
 obj-y				+= cpu/ timers/
diff --git a/arch/sh/kernel/Makefile_64 b/arch/sh/kernel/Makefile_64
index e01283d49cbf..6edf53b93d94 100644
--- a/arch/sh/kernel/Makefile_64
+++ b/arch/sh/kernel/Makefile_64
@@ -1,7 +1,7 @@
 extra-y	:= head_64.o init_task.o vmlinux.lds
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_64.o \
-	   ptrace_64.o semaphore.o setup.o signal_64.o sys_sh.o sys_sh64.o \
+	   ptrace_64.o setup.o signal_64.o sys_sh.o sys_sh64.o \
 	   syscalls_64.o time_64.o topology.o traps.o traps_64.o
 
 obj-y				+= cpu/ timers/
diff --git a/arch/sh/kernel/semaphore.c b/arch/sh/kernel/semaphore.c
deleted file mode 100644
index 184119eeae56..000000000000
--- a/arch/sh/kernel/semaphore.c
+++ /dev/null
@@ -1,139 +0,0 @@
-/*
- * Just taken from alpha implementation.
- * This can't work well, perhaps.
- */
-/*
- *  Generic semaphore code. Buyer beware. Do your own
- * specific changes in <asm/semaphore-helper.h>
- */
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/wait.h>
-#include <linux/init.h>
-#include <asm/semaphore.h>
-#include <asm/semaphore-helper.h>
-
-DEFINE_SPINLOCK(semaphore_wake_lock);
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to sleep, while the "waking" variable is
- * incremented when the "up()" code goes to wake up waiting
- * processes.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * waking_non_zero() (from asm/semaphore.h) must execute
- * atomically.
- *
- * When __up() is called, the count was negative before
- * incrementing it, and we need to wake up somebody.
- *
- * This routine adds one to the count of processes that need to
- * wake up and exit.  ALL waiting processes actually wake up but
- * only the one that gets to the "waking" field first will gate
- * through and acquire the semaphore.  The others will go back
- * to sleep.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-void __up(struct semaphore *sem)
-{
-	wake_one_more(sem);
-	wake_up(&sem->wait);
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from __down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from __down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- *
- */
-
-#define DOWN_VAR				\
-	struct task_struct *tsk = current;	\
-	wait_queue_t wait;			\
-	init_waitqueue_entry(&wait, tsk);
-
-#define DOWN_HEAD(task_state)						\
-									\
-									\
-	tsk->state = (task_state);					\
-	add_wait_queue(&sem->wait, &wait);				\
-									\
-	/*								\
-	 * Ok, we're set up.  sem->count is known to be less than zero	\
-	 * so we must wait.						\
-	 *								\
-	 * We can let go the lock for purposes of waiting.		\
-	 * We re-acquire it after awaking so as to protect		\
-	 * all semaphore operations.					\
-	 *								\
-	 * If "up()" is called before we call waking_non_zero() then	\
-	 * we will catch it right away.  If it is called later then	\
-	 * we will have to go through a wakeup cycle to catch it.	\
-	 *								\
-	 * Multiple waiters contend for the semaphore lock to see	\
-	 * who gets to gate through and who has to wait some more.	\
-	 */								\
-	for (;;) {
-
-#define DOWN_TAIL(task_state)			\
-		tsk->state = (task_state);	\
-	}					\
-	tsk->state = TASK_RUNNING;		\
-	remove_wait_queue(&sem->wait, &wait);
-
-void __sched __down(struct semaphore * sem)
-{
-	DOWN_VAR
-	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
-	if (waking_non_zero(sem))
-		break;
-	schedule();
-	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-	DOWN_VAR
-	DOWN_HEAD(TASK_INTERRUPTIBLE)
-
-	ret = waking_non_zero_interruptible(sem, tsk);
-	if (ret)
-	{
-		if (ret == 1)
-			/* ret != 0 only if we get interrupted -arca */
-			ret = 0;
-		break;
-	}
-	schedule();
-	DOWN_TAIL(TASK_INTERRUPTIBLE)
-	return ret;
-}
-
-int __down_trylock(struct semaphore * sem)
-{
-	return waking_non_zero_trylock(sem);
-}
diff --git a/arch/sh/kernel/sh_ksyms_32.c b/arch/sh/kernel/sh_ksyms_32.c
index 45bb333fd9ec..6d405462cee8 100644
--- a/arch/sh/kernel/sh_ksyms_32.c
+++ b/arch/sh/kernel/sh_ksyms_32.c
@@ -9,7 +9,6 @@
 #include <linux/pci.h>
 #include <linux/irq.h>
 #include <asm/sections.h>
-#include <asm/semaphore.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
 #include <asm/checksum.h>
@@ -48,12 +47,6 @@ EXPORT_SYMBOL(__copy_user);
 EXPORT_SYMBOL(get_vm_area);
 #endif
 
-/* semaphore exports */
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_interruptible);
-EXPORT_SYMBOL(__down_trylock);
-
 EXPORT_SYMBOL(__udelay);
 EXPORT_SYMBOL(__ndelay);
 EXPORT_SYMBOL(__const_udelay);
diff --git a/arch/sh/kernel/sh_ksyms_64.c b/arch/sh/kernel/sh_ksyms_64.c
index b6410ce4bd1d..a310c9707f03 100644
--- a/arch/sh/kernel/sh_ksyms_64.c
+++ b/arch/sh/kernel/sh_ksyms_64.c
@@ -16,7 +16,6 @@
 #include <linux/in6.h>
 #include <linux/interrupt.h>
 #include <linux/screen_info.h>
-#include <asm/semaphore.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
 #include <asm/checksum.h>
@@ -37,9 +36,6 @@ EXPORT_SYMBOL(csum_partial_copy_nocheck);
 EXPORT_SYMBOL(screen_info);
 #endif
 
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_trylock);
-EXPORT_SYMBOL(__up);
 EXPORT_SYMBOL(__put_user_asm_l);
 EXPORT_SYMBOL(__get_user_asm_l);
 EXPORT_SYMBOL(copy_page);
diff --git a/arch/sparc/kernel/Makefile b/arch/sparc/kernel/Makefile
index bf1b15d3f6f5..2712bb166f6f 100644
--- a/arch/sparc/kernel/Makefile
+++ b/arch/sparc/kernel/Makefile
@@ -12,7 +12,7 @@ obj-y    := entry.o wof.o wuf.o etrap.o rtrap.o traps.o $(IRQ_OBJS) \
 	    sys_sparc.o sunos_asm.o systbls.o \
 	    time.o windows.o cpu.o devices.o sclow.o \
 	    tadpole.o tick14.o ptrace.o sys_solaris.o \
-	    unaligned.o una_asm.o muldiv.o semaphore.o \
+	    unaligned.o una_asm.o muldiv.o \
 	    prom.o of_device.o devres.o
 
 devres-y = ../../../kernel/irq/devres.o
diff --git a/arch/sparc/kernel/semaphore.c b/arch/sparc/kernel/semaphore.c
deleted file mode 100644
index 0c37c1a7cd7e..000000000000
--- a/arch/sparc/kernel/semaphore.c
+++ /dev/null
@@ -1,155 +0,0 @@
-/* $Id: semaphore.c,v 1.7 2001/04/18 21:06:05 davem Exp $ */
-
-/* sparc32 semaphore implementation, based on i386 version */
-
-#include <linux/sched.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-
-#include <asm/semaphore.h>
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to acquire the semaphore, while the "sleeping"
- * variable is a count of such acquires.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * "sleeping" and the contention routine ordering is
- * protected by the semaphore spinlock.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-
-/*
- * Logic:
- *  - only on a boundary condition do we need to care. When we go
- *    from a negative count to a non-negative, we wake people up.
- *  - when we go from a non-negative count to a negative do we
- *    (a) synchronize with the "sleeper" count and (b) make sure
- *    that we're on the wakeup list before we synchronize so that
- *    we cannot lose wakeup events.
- */
-
-void __up(struct semaphore *sem)
-{
-	wake_up(&sem->wait);
-}
-
-static DEFINE_SPINLOCK(semaphore_lock);
-
-void __sched __down(struct semaphore * sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	spin_lock_irq(&semaphore_lock);
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock.
-		 */
-		if (!atomic24_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irq(&semaphore_lock);
-
-		schedule();
-		tsk->state = TASK_UNINTERRUPTIBLE;
-		spin_lock_irq(&semaphore_lock);
-	}
-	spin_unlock_irq(&semaphore_lock);
-	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
-	wake_up(&sem->wait);
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	tsk->state = TASK_INTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	spin_lock_irq(&semaphore_lock);
-	sem->sleepers ++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * With signals pending, this turns into
-		 * the trylock failure case - we won't be
-		 * sleeping, and we* can't get the lock as
-		 * it has contention. Just correct the count
-		 * and exit.
-		 */
-		if (signal_pending(current)) {
-			retval = -EINTR;
-			sem->sleepers = 0;
-			atomic24_add(sleepers, &sem->count);
-			break;
-		}
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock. The
-		 * "-1" is because we're still hoping to get
-		 * the lock.
-		 */
-		if (!atomic24_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irq(&semaphore_lock);
-
-		schedule();
-		tsk->state = TASK_INTERRUPTIBLE;
-		spin_lock_irq(&semaphore_lock);
-	}
-	spin_unlock_irq(&semaphore_lock);
-	tsk->state = TASK_RUNNING;
-	remove_wait_queue(&sem->wait, &wait);
-	wake_up(&sem->wait);
-	return retval;
-}
-
-/*
- * Trylock failed - make sure we correct for
- * having decremented the count.
- */
-int __down_trylock(struct semaphore * sem)
-{
-	int sleepers;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_lock, flags);
-	sleepers = sem->sleepers + 1;
-	sem->sleepers = 0;
-
-	/*
-	 * Add "everybody else" and us into it. They aren't
-	 * playing, because we own the spinlock.
-	 */
-	if (!atomic24_add_negative(sleepers, &sem->count))
-		wake_up(&sem->wait);
-
-	spin_unlock_irqrestore(&semaphore_lock, flags);
-	return 1;
-}
diff --git a/arch/sparc/kernel/sparc_ksyms.c b/arch/sparc/kernel/sparc_ksyms.c
index c1025e551650..97b1de0e9094 100644
--- a/arch/sparc/kernel/sparc_ksyms.c
+++ b/arch/sparc/kernel/sparc_ksyms.c
@@ -107,11 +107,6 @@ EXPORT_SYMBOL(___rw_read_try);
 EXPORT_SYMBOL(___rw_read_exit);
 EXPORT_SYMBOL(___rw_write_enter);
 #endif
-/* semaphores */
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_trylock);
-EXPORT_SYMBOL(__down_interruptible);
 
 EXPORT_SYMBOL(sparc_valid_addr_bitmap);
 EXPORT_SYMBOL(phys_base);
diff --git a/arch/sparc64/kernel/Makefile b/arch/sparc64/kernel/Makefile
index 1bf5b187de49..459462e80a12 100644
--- a/arch/sparc64/kernel/Makefile
+++ b/arch/sparc64/kernel/Makefile
@@ -10,7 +10,7 @@ extra-y		:= head.o init_task.o vmlinux.lds
 obj-y		:= process.o setup.o cpu.o idprom.o \
 		   traps.o auxio.o una_asm.o sysfs.o iommu.o \
 		   irq.o ptrace.o time.o sys_sparc.o signal.o \
-		   unaligned.o central.o pci.o starfire.o semaphore.o \
+		   unaligned.o central.o pci.o starfire.o \
 		   power.o sbus.o sparc64_ksyms.o chmc.o \
 		   visemul.o prom.o of_device.o hvapi.o sstate.o mdesc.o
 
diff --git a/arch/sparc64/kernel/semaphore.c b/arch/sparc64/kernel/semaphore.c
deleted file mode 100644
index 9974a6899551..000000000000
--- a/arch/sparc64/kernel/semaphore.c
+++ /dev/null
@@ -1,254 +0,0 @@
-/* semaphore.c: Sparc64 semaphore implementation.
- *
- * This is basically the PPC semaphore scheme ported to use
- * the sparc64 atomic instructions, so see the PPC code for
- * credits.
- */
-
-#include <linux/sched.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-
-/*
- * Atomically update sem->count.
- * This does the equivalent of the following:
- *
- *	old_count = sem->count;
- *	tmp = MAX(old_count, 0) + incr;
- *	sem->count = tmp;
- *	return old_count;
- */
-static inline int __sem_update_count(struct semaphore *sem, int incr)
-{
-	int old_count, tmp;
-
-	__asm__ __volatile__("\n"
-"	! __sem_update_count old_count(%0) tmp(%1) incr(%4) &sem->count(%3)\n"
-"1:	ldsw	[%3], %0\n"
-"	mov	%0, %1\n"
-"	cmp	%0, 0\n"
-"	movl	%%icc, 0, %1\n"
-"	add	%1, %4, %1\n"
-"	cas	[%3], %0, %1\n"
-"	cmp	%0, %1\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	bne,pn	%%icc, 1b\n"
-"	 nop\n"
-	: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
-	: "r" (&sem->count), "r" (incr), "m" (sem->count)
-	: "cc");
-
-	return old_count;
-}
-
-static void __up(struct semaphore *sem)
-{
-	__sem_update_count(sem, 1);
-	wake_up(&sem->wait);
-}
-
-void up(struct semaphore *sem)
-{
-	/* This atomically does:
-	 * 	old_val = sem->count;
-	 *	new_val = sem->count + 1;
-	 *	sem->count = new_val;
-	 *	if (old_val < 0)
-	 *		__up(sem);
-	 *
-	 * The (old_val < 0) test is equivalent to
-	 * the more straightforward (new_val <= 0),
-	 * but it is easier to test the former because
-	 * of how the CAS instruction works.
-	 */
-
-	__asm__ __volatile__("\n"
-"	! up sem(%0)\n"
-"	membar	#StoreLoad | #LoadLoad\n"
-"1:	lduw	[%0], %%g1\n"
-"	add	%%g1, 1, %%g7\n"
-"	cas	[%0], %%g1, %%g7\n"
-"	cmp	%%g1, %%g7\n"
-"	bne,pn	%%icc, 1b\n"
-"	 addcc	%%g7, 1, %%g0\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	ble,pn	%%icc, 3f\n"
-"	 nop\n"
-"2:\n"
-"	.subsection 2\n"
-"3:	mov	%0, %%g1\n"
-"	save	%%sp, -160, %%sp\n"
-"	call	%1\n"
-"	 mov	%%g1, %%o0\n"
-"	ba,pt	%%xcc, 2b\n"
-"	 restore\n"
-"	.previous\n"
-	: : "r" (sem), "i" (__up)
-	: "g1", "g2", "g3", "g7", "memory", "cc");
-}
-
-static void __sched __down(struct semaphore * sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	while (__sem_update_count(sem, -1) <= 0) {
-		schedule();
-		tsk->state = TASK_UNINTERRUPTIBLE;
-	}
-	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
-
-	wake_up(&sem->wait);
-}
-
-void __sched down(struct semaphore *sem)
-{
-	might_sleep();
-	/* This atomically does:
-	 * 	old_val = sem->count;
-	 *	new_val = sem->count - 1;
-	 *	sem->count = new_val;
-	 *	if (old_val < 1)
-	 *		__down(sem);
-	 *
-	 * The (old_val < 1) test is equivalent to
-	 * the more straightforward (new_val < 0),
-	 * but it is easier to test the former because
-	 * of how the CAS instruction works.
-	 */
-
-	__asm__ __volatile__("\n"
-"	! down sem(%0)\n"
-"1:	lduw	[%0], %%g1\n"
-"	sub	%%g1, 1, %%g7\n"
-"	cas	[%0], %%g1, %%g7\n"
-"	cmp	%%g1, %%g7\n"
-"	bne,pn	%%icc, 1b\n"
-"	 cmp	%%g7, 1\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	bl,pn	%%icc, 3f\n"
-"	 nop\n"
-"2:\n"
-"	.subsection 2\n"
-"3:	mov	%0, %%g1\n"
-"	save	%%sp, -160, %%sp\n"
-"	call	%1\n"
-"	 mov	%%g1, %%o0\n"
-"	ba,pt	%%xcc, 2b\n"
-"	 restore\n"
-"	.previous\n"
-	: : "r" (sem), "i" (__down)
-	: "g1", "g2", "g3", "g7", "memory", "cc");
-}
-
-int down_trylock(struct semaphore *sem)
-{
-	int ret;
-
-	/* This atomically does:
-	 * 	old_val = sem->count;
-	 *	new_val = sem->count - 1;
-	 *	if (old_val < 1) {
-	 *		ret = 1;
-	 *	} else {
-	 *		sem->count = new_val;
-	 *		ret = 0;
-	 *	}
-	 *
-	 * The (old_val < 1) test is equivalent to
-	 * the more straightforward (new_val < 0),
-	 * but it is easier to test the former because
-	 * of how the CAS instruction works.
-	 */
-
-	__asm__ __volatile__("\n"
-"	! down_trylock sem(%1) ret(%0)\n"
-"1:	lduw	[%1], %%g1\n"
-"	sub	%%g1, 1, %%g7\n"
-"	cmp	%%g1, 1\n"
-"	bl,pn	%%icc, 2f\n"
-"	 mov	1, %0\n"
-"	cas	[%1], %%g1, %%g7\n"
-"	cmp	%%g1, %%g7\n"
-"	bne,pn	%%icc, 1b\n"
-"	 mov	0, %0\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"2:\n"
-	: "=&r" (ret)
-	: "r" (sem)
-	: "g1", "g7", "memory", "cc");
-
-	return ret;
-}
-
-static int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-
-	tsk->state = TASK_INTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	while (__sem_update_count(sem, -1) <= 0) {
-		if (signal_pending(current)) {
-			__sem_update_count(sem, 0);
-			retval = -EINTR;
-			break;
-		}
-		schedule();
-		tsk->state = TASK_INTERRUPTIBLE;
-	}
-	tsk->state = TASK_RUNNING;
-	remove_wait_queue(&sem->wait, &wait);
-	wake_up(&sem->wait);
-	return retval;
-}
-
-int __sched down_interruptible(struct semaphore *sem)
-{
-	int ret = 0;
-	
-	might_sleep();
-	/* This atomically does:
-	 * 	old_val = sem->count;
-	 *	new_val = sem->count - 1;
-	 *	sem->count = new_val;
-	 *	if (old_val < 1)
-	 *		ret = __down_interruptible(sem);
-	 *
-	 * The (old_val < 1) test is equivalent to
-	 * the more straightforward (new_val < 0),
-	 * but it is easier to test the former because
-	 * of how the CAS instruction works.
-	 */
-
-	__asm__ __volatile__("\n"
-"	! down_interruptible sem(%2) ret(%0)\n"
-"1:	lduw	[%2], %%g1\n"
-"	sub	%%g1, 1, %%g7\n"
-"	cas	[%2], %%g1, %%g7\n"
-"	cmp	%%g1, %%g7\n"
-"	bne,pn	%%icc, 1b\n"
-"	 cmp	%%g7, 1\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	bl,pn	%%icc, 3f\n"
-"	 nop\n"
-"2:\n"
-"	.subsection 2\n"
-"3:	mov	%2, %%g1\n"
-"	save	%%sp, -160, %%sp\n"
-"	call	%3\n"
-"	 mov	%%g1, %%o0\n"
-"	ba,pt	%%xcc, 2b\n"
-"	 restore\n"
-"	.previous\n"
-	: "=r" (ret)
-	: "0" (ret), "r" (sem), "i" (__down_interruptible)
-	: "g1", "g2", "g3", "g7", "memory", "cc");
-	return ret;
-}
diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c
index 51fa773f38c9..051b8d9cb989 100644
--- a/arch/sparc64/kernel/sparc64_ksyms.c
+++ b/arch/sparc64/kernel/sparc64_ksyms.c
@@ -130,12 +130,6 @@ EXPORT_SYMBOL(_mcount);
 
 EXPORT_SYMBOL(sparc64_get_clock_tick);
 
-/* semaphores */
-EXPORT_SYMBOL(down);
-EXPORT_SYMBOL(down_trylock);
-EXPORT_SYMBOL(down_interruptible);
-EXPORT_SYMBOL(up);
-
 /* RW semaphores */
 EXPORT_SYMBOL(__down_read);
 EXPORT_SYMBOL(__down_read_trylock);
diff --git a/arch/um/Kconfig.i386 b/arch/um/Kconfig.i386
index 3cd8a04d66d8..e09edfa560da 100644
--- a/arch/um/Kconfig.i386
+++ b/arch/um/Kconfig.i386
@@ -19,10 +19,6 @@ config 64BIT
 	bool
 	default n
 
-config SEMAPHORE_SLEEPERS
-	bool
-	default y
-
 config 3_LEVEL_PGTABLES
 	bool "Three-level pagetables (EXPERIMENTAL)"
 	default n
diff --git a/arch/um/Kconfig.x86_64 b/arch/um/Kconfig.x86_64
index 6533b349f061..3fbe69e359ed 100644
--- a/arch/um/Kconfig.x86_64
+++ b/arch/um/Kconfig.x86_64
@@ -11,10 +11,6 @@ config RWSEM_GENERIC_SPINLOCK
 	bool
 	default y
 
-config SEMAPHORE_SLEEPERS
-	bool
-	default y
-
 config 3_LEVEL_PGTABLES
        bool
        default y
diff --git a/arch/um/sys-i386/ksyms.c b/arch/um/sys-i386/ksyms.c
index 2a1eac1859ce..bfbefd30db8f 100644
--- a/arch/um/sys-i386/ksyms.c
+++ b/arch/um/sys-i386/ksyms.c
@@ -1,17 +1,5 @@
 #include "linux/module.h"
-#include "linux/in6.h"
-#include "linux/rwsem.h"
-#include "asm/byteorder.h"
-#include "asm/delay.h"
-#include "asm/semaphore.h"
-#include "asm/uaccess.h"
 #include "asm/checksum.h"
-#include "asm/errno.h"
-
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_failed_interruptible);
-EXPORT_SYMBOL(__down_failed_trylock);
-EXPORT_SYMBOL(__up_wakeup);
 
 /* Networking helper routines. */
 EXPORT_SYMBOL(csum_partial);
diff --git a/arch/um/sys-ppc/Makefile b/arch/um/sys-ppc/Makefile
index 08901526e893..b8bc844fd2c4 100644
--- a/arch/um/sys-ppc/Makefile
+++ b/arch/um/sys-ppc/Makefile
@@ -3,7 +3,7 @@ OBJ = built-in.o
 .S.o:
 	$(CC) $(KBUILD_AFLAGS) -D__ASSEMBLY__ -D__UM_PPC__ -c $< -o $*.o
 
-OBJS = ptrace.o sigcontext.o semaphore.o checksum.o miscthings.o misc.o \
+OBJS = ptrace.o sigcontext.o checksum.o miscthings.o misc.o \
 	ptrace_user.o sysrq.o
 
 EXTRA_AFLAGS := -DCONFIG_PPC32 -I. -I$(srctree)/arch/ppc/kernel
@@ -20,10 +20,6 @@ ptrace_user.o: ptrace_user.c
 sigcontext.o: sigcontext.c
 	$(CC) $(USER_CFLAGS) $(EXTRA_CFLAGS) -c -o $@ $<
 
-semaphore.c:
-	rm -f $@
-	ln -s $(srctree)/arch/ppc/kernel/$@ $@
-
 checksum.S:
 	rm -f $@
 	ln -s $(srctree)/arch/ppc/lib/$@ $@
@@ -66,4 +62,4 @@ misc.o: misc.S ppc_defs.h
 	$(CC) $(EXTRA_AFLAGS) $(KBUILD_AFLAGS) -D__ASSEMBLY__ -D__UM_PPC__ -c $< -o $*.o
 	rm -f asm
 
-clean-files := $(OBJS) ppc_defs.h checksum.S semaphore.c mk_defs.c
+clean-files := $(OBJS) ppc_defs.h checksum.S mk_defs.c
diff --git a/arch/um/sys-x86_64/ksyms.c b/arch/um/sys-x86_64/ksyms.c
index 12c593607c59..4d7d1a812d8f 100644
--- a/arch/um/sys-x86_64/ksyms.c
+++ b/arch/um/sys-x86_64/ksyms.c
@@ -1,16 +1,5 @@
 #include "linux/module.h"
-#include "linux/in6.h"
-#include "linux/rwsem.h"
-#include "asm/byteorder.h"
-#include "asm/semaphore.h"
-#include "asm/uaccess.h"
-#include "asm/checksum.h"
-#include "asm/errno.h"
-
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_failed_interruptible);
-EXPORT_SYMBOL(__down_failed_trylock);
-EXPORT_SYMBOL(__up_wakeup);
+#include "asm/string.h"
 
 /*XXX: we need them because they would be exported by x86_64 */
 EXPORT_SYMBOL(__memcpy);
diff --git a/arch/v850/kernel/Makefile b/arch/v850/kernel/Makefile
index 3930482bddc4..da5889c53576 100644
--- a/arch/v850/kernel/Makefile
+++ b/arch/v850/kernel/Makefile
@@ -11,7 +11,7 @@
 
 extra-y := head.o init_task.o vmlinux.lds
 
-obj-y += intv.o entry.o process.o syscalls.o time.o semaphore.o setup.o \
+obj-y += intv.o entry.o process.o syscalls.o time.o setup.o \
 	 signal.o irq.o mach.o ptrace.o bug.o
 obj-$(CONFIG_MODULES)		+= module.o v850_ksyms.o
 # chip-specific code
diff --git a/arch/v850/kernel/semaphore.c b/arch/v850/kernel/semaphore.c
deleted file mode 100644
index fc89fd661c99..000000000000
--- a/arch/v850/kernel/semaphore.c
+++ /dev/null
@@ -1,166 +0,0 @@
-/*
- * arch/v850/kernel/semaphore.c -- Semaphore support
- *
- *  Copyright (C) 1998-2000  IBM Corporation
- *  Copyright (C) 1999  Linus Torvalds
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * This file is a copy of the s390 version, arch/s390/kernel/semaphore.c
- *    Author(s): Martin Schwidefsky
- * which was derived from the i386 version, linux/arch/i386/kernel/semaphore.c
- */
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/init.h>
-
-#include <asm/semaphore.h>
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to acquire the semaphore, while the "sleeping"
- * variable is a count of such acquires.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * "sleeping" and the contention routine ordering is
- * protected by the semaphore spinlock.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-
-/*
- * Logic:
- *  - only on a boundary condition do we need to care. When we go
- *    from a negative count to a non-negative, we wake people up.
- *  - when we go from a non-negative count to a negative do we
- *    (a) synchronize with the "sleeper" count and (b) make sure
- *    that we're on the wakeup list before we synchronize so that
- *    we cannot lose wakeup events.
- */
-
-void __up(struct semaphore *sem)
-{
-	wake_up(&sem->wait);
-}
-
-static DEFINE_SPINLOCK(semaphore_lock);
-
-void __sched __down(struct semaphore * sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	spin_lock_irq(&semaphore_lock);
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irq(&semaphore_lock);
-
-		schedule();
-		tsk->state = TASK_UNINTERRUPTIBLE;
-		spin_lock_irq(&semaphore_lock);
-	}
-	spin_unlock_irq(&semaphore_lock);
-	remove_wait_queue(&sem->wait, &wait);
-	tsk->state = TASK_RUNNING;
-	wake_up(&sem->wait);
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	tsk->state = TASK_INTERRUPTIBLE;
-	add_wait_queue_exclusive(&sem->wait, &wait);
-
-	spin_lock_irq(&semaphore_lock);
-	sem->sleepers ++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * With signals pending, this turns into
-		 * the trylock failure case - we won't be
-		 * sleeping, and we* can't get the lock as
-		 * it has contention. Just correct the count
-		 * and exit.
-		 */
-		if (signal_pending(current)) {
-			retval = -EINTR;
-			sem->sleepers = 0;
-			atomic_add(sleepers, &sem->count);
-			break;
-		}
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock. The
-		 * "-1" is because we're still hoping to get
-		 * the lock.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irq(&semaphore_lock);
-
-		schedule();
-		tsk->state = TASK_INTERRUPTIBLE;
-		spin_lock_irq(&semaphore_lock);
-	}
-	spin_unlock_irq(&semaphore_lock);
-	tsk->state = TASK_RUNNING;
-	remove_wait_queue(&sem->wait, &wait);
-	wake_up(&sem->wait);
-	return retval;
-}
-
-/*
- * Trylock failed - make sure we correct for
- * having decremented the count.
- */
-int __down_trylock(struct semaphore * sem)
-{
-        unsigned long flags;
-	int sleepers;
-
-	spin_lock_irqsave(&semaphore_lock, flags);
-	sleepers = sem->sleepers + 1;
-	sem->sleepers = 0;
-
-	/*
-	 * Add "everybody else" and us into it. They aren't
-	 * playing, because we own the spinlock.
-	 */
-	if (!atomic_add_negative(sleepers, &sem->count))
-		wake_up(&sem->wait);
-
-	spin_unlock_irqrestore(&semaphore_lock, flags);
-	return 1;
-}
diff --git a/arch/v850/kernel/v850_ksyms.c b/arch/v850/kernel/v850_ksyms.c
index 93575fdc874d..8d386a5dbc4a 100644
--- a/arch/v850/kernel/v850_ksyms.c
+++ b/arch/v850/kernel/v850_ksyms.c
@@ -11,7 +11,6 @@
 #include <asm/pgalloc.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/semaphore.h>
 #include <asm/checksum.h>
 #include <asm/current.h>
 
@@ -34,12 +33,6 @@ EXPORT_SYMBOL (memset);
 EXPORT_SYMBOL (memcpy);
 EXPORT_SYMBOL (memmove);
 
-/* semaphores */
-EXPORT_SYMBOL (__down);
-EXPORT_SYMBOL (__down_interruptible);
-EXPORT_SYMBOL (__down_trylock);
-EXPORT_SYMBOL (__up);
-
 /*
  * libgcc functions - functions that are used internally by the
  * compiler...  (prototypes are not correct though, but that
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6c70fed0f9a0..e4b38861ea52 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -53,9 +53,6 @@ config STACKTRACE_SUPPORT
 config HAVE_LATENCYTOP_SUPPORT
 	def_bool y
 
-config SEMAPHORE_SLEEPERS
-	def_bool y
-
 config FAST_CMPXCHG_LOCAL
 	bool
 	default y
diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c
index 061627806a2d..deb43785e923 100644
--- a/arch/x86/kernel/i386_ksyms_32.c
+++ b/arch/x86/kernel/i386_ksyms_32.c
@@ -1,13 +1,8 @@
 #include <linux/module.h>
-#include <asm/semaphore.h>
 #include <asm/checksum.h>
 #include <asm/desc.h>
 #include <asm/pgtable.h>
 
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_failed_interruptible);
-EXPORT_SYMBOL(__down_failed_trylock);
-EXPORT_SYMBOL(__up_wakeup);
 /* Networking helper routines. */
 EXPORT_SYMBOL(csum_partial_copy_generic);
 
diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c
index a66e9c1a0537..95a993e18165 100644
--- a/arch/x86/kernel/x8664_ksyms_64.c
+++ b/arch/x86/kernel/x8664_ksyms_64.c
@@ -4,7 +4,6 @@
 #include <linux/module.h>
 #include <linux/smp.h>
 
-#include <asm/semaphore.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -12,11 +11,6 @@
 
 EXPORT_SYMBOL(kernel_thread);
 
-EXPORT_SYMBOL(__down_failed);
-EXPORT_SYMBOL(__down_failed_interruptible);
-EXPORT_SYMBOL(__down_failed_trylock);
-EXPORT_SYMBOL(__up_wakeup);
-
 EXPORT_SYMBOL(__get_user_1);
 EXPORT_SYMBOL(__get_user_2);
 EXPORT_SYMBOL(__get_user_4);
diff --git a/arch/x86/lib/semaphore_32.S b/arch/x86/lib/semaphore_32.S
index 3899bd37fdf0..648fe4741782 100644
--- a/arch/x86/lib/semaphore_32.S
+++ b/arch/x86/lib/semaphore_32.S
@@ -30,89 +30,6 @@
  * value or just clobbered..
  */
 	.section .sched.text, "ax"
-ENTRY(__down_failed)
-	CFI_STARTPROC
-	FRAME
-	pushl %edx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET edx,0
-	pushl %ecx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET ecx,0
-	call __down
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE ecx
-	popl %edx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE edx
-	ENDFRAME
-	ret
-	CFI_ENDPROC
-	ENDPROC(__down_failed)
-
-ENTRY(__down_failed_interruptible)
-	CFI_STARTPROC
-	FRAME
-	pushl %edx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET edx,0
-	pushl %ecx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET ecx,0
-	call __down_interruptible
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE ecx
-	popl %edx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE edx
-	ENDFRAME
-	ret
-	CFI_ENDPROC
-	ENDPROC(__down_failed_interruptible)
-
-ENTRY(__down_failed_trylock)
-	CFI_STARTPROC
-	FRAME
-	pushl %edx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET edx,0
-	pushl %ecx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET ecx,0
-	call __down_trylock
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE ecx
-	popl %edx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE edx
-	ENDFRAME
-	ret
-	CFI_ENDPROC
-	ENDPROC(__down_failed_trylock)
-
-ENTRY(__up_wakeup)
-	CFI_STARTPROC
-	FRAME
-	pushl %edx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET edx,0
-	pushl %ecx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET ecx,0
-	call __up
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE ecx
-	popl %edx
-	CFI_ADJUST_CFA_OFFSET -4
-	CFI_RESTORE edx
-	ENDFRAME
-	ret
-	CFI_ENDPROC
-	ENDPROC(__up_wakeup)
 
 /*
  * rw spinlock fallbacks
diff --git a/arch/x86/lib/thunk_64.S b/arch/x86/lib/thunk_64.S
index 8b92d428ab02..e009251d4e9f 100644
--- a/arch/x86/lib/thunk_64.S
+++ b/arch/x86/lib/thunk_64.S
@@ -41,11 +41,6 @@
 	thunk rwsem_downgrade_thunk,rwsem_downgrade_wake
 #endif	
 	
-	thunk __down_failed,__down
-	thunk_retrax __down_failed_interruptible,__down_interruptible
-	thunk_retrax __down_failed_trylock,__down_trylock
-	thunk __up_wakeup,__up
-
 #ifdef CONFIG_TRACE_IRQFLAGS
 	thunk trace_hardirqs_on_thunk,trace_hardirqs_on
 	thunk trace_hardirqs_off_thunk,trace_hardirqs_off
diff --git a/arch/xtensa/kernel/Makefile b/arch/xtensa/kernel/Makefile
index f582d6a24ec2..7419dbccf027 100644
--- a/arch/xtensa/kernel/Makefile
+++ b/arch/xtensa/kernel/Makefile
@@ -5,7 +5,7 @@
 extra-y := head.o vmlinux.lds
 
 
-obj-y := align.o entry.o irq.o coprocessor.o process.o ptrace.o semaphore.o  \
+obj-y := align.o entry.o irq.o coprocessor.o process.o ptrace.o \
 	 setup.o signal.o syscall.o time.o traps.o vectors.o platform.o  \
 	 pci-dma.o init_task.o io.o
 
diff --git a/arch/xtensa/kernel/semaphore.c b/arch/xtensa/kernel/semaphore.c
deleted file mode 100644
index 995c6410ae10..000000000000
--- a/arch/xtensa/kernel/semaphore.c
+++ /dev/null
@@ -1,226 +0,0 @@
-/*
- * arch/xtensa/kernel/semaphore.c
- *
- * Generic semaphore code. Buyer beware. Do your own specific changes
- * in <asm/semaphore-helper.h>
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2001 - 2005 Tensilica Inc.
- *
- * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
- * Chris Zankel	<chris@zankel.net>
- * Marc Gauthier<marc@tensilica.com, marc@alumni.uwaterloo.ca>
- * Kevin Chea
- */
-
-#include <linux/sched.h>
-#include <linux/wait.h>
-#include <linux/init.h>
-#include <asm/semaphore.h>
-#include <asm/errno.h>
-
-/*
- * These two _must_ execute atomically wrt each other.
- */
-
-static __inline__ void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc((atomic_t *)&sem->sleepers);
-}
-
-static __inline__ int waking_non_zero(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- *
- * We must undo the sem->count down_interruptible() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-
-static __inline__ int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- *
- * We must undo the sem->count down_trylock() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-
-static __inline__ int waking_non_zero_trylock(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 1;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers <= 0)
-		atomic_inc(&sem->count);
-	else {
-		sem->sleepers--;
-		ret = 0;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-DEFINE_SPINLOCK(semaphore_wake_lock);
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to sleep, while the "waking" variable is
- * incremented when the "up()" code goes to wake up waiting
- * processes.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * waking_non_zero() (from asm/semaphore.h) must execute
- * atomically.
- *
- * When __up() is called, the count was negative before
- * incrementing it, and we need to wake up somebody.
- *
- * This routine adds one to the count of processes that need to
- * wake up and exit.  ALL waiting processes actually wake up but
- * only the one that gets to the "waking" field first will gate
- * through and acquire the semaphore.  The others will go back
- * to sleep.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-
-void __up(struct semaphore *sem)
-{
-	wake_one_more(sem);
-	wake_up(&sem->wait);
-}
-
-/*
- * Perform the "down" function.  Return zero for semaphore acquired,
- * return negative for signalled out of the function.
- *
- * If called from __down, the return is ignored and the wait loop is
- * not interruptible.  This means that a task waiting on a semaphore
- * using "down()" cannot be killed until someone does an "up()" on
- * the semaphore.
- *
- * If called from __down_interruptible, the return value gets checked
- * upon return.  If the return value is negative then the task continues
- * with the negative value in the return register (it can be tested by
- * the caller).
- *
- * Either form may be used in conjunction with "up()".
- *
- */
-
-#define DOWN_VAR				\
-	struct task_struct *tsk = current;	\
-	wait_queue_t wait;			\
-	init_waitqueue_entry(&wait, tsk);
-
-#define DOWN_HEAD(task_state)						\
-									\
-									\
-	tsk->state = (task_state);					\
-	add_wait_queue(&sem->wait, &wait);				\
-									\
-	/*								\
-	 * Ok, we're set up.  sem->count is known to be less than zero	\
-	 * so we must wait.						\
-	 *								\
-	 * We can let go the lock for purposes of waiting.		\
-	 * We re-acquire it after awaking so as to protect		\
-	 * all semaphore operations.					\
-	 *								\
-	 * If "up()" is called before we call waking_non_zero() then	\
-	 * we will catch it right away.  If it is called later then	\
-	 * we will have to go through a wakeup cycle to catch it.	\
-	 *								\
-	 * Multiple waiters contend for the semaphore lock to see	\
-	 * who gets to gate through and who has to wait some more.	\
-	 */								\
-	for (;;) {
-
-#define DOWN_TAIL(task_state)			\
-		tsk->state = (task_state);	\
-	}					\
-	tsk->state = TASK_RUNNING;		\
-	remove_wait_queue(&sem->wait, &wait);
-
-void __sched __down(struct semaphore * sem)
-{
-	DOWN_VAR
-	DOWN_HEAD(TASK_UNINTERRUPTIBLE)
-	if (waking_non_zero(sem))
-		break;
-	schedule();
-	DOWN_TAIL(TASK_UNINTERRUPTIBLE)
-}
-
-int __sched __down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-	DOWN_VAR
-	DOWN_HEAD(TASK_INTERRUPTIBLE)
-
-	ret = waking_non_zero_interruptible(sem, tsk);
-	if (ret)
-	{
-		if (ret == 1)
-			/* ret != 0 only if we get interrupted -arca */
-			ret = 0;
-		break;
-	}
-	schedule();
-	DOWN_TAIL(TASK_INTERRUPTIBLE)
-	return ret;
-}
-
-int __down_trylock(struct semaphore * sem)
-{
-	return waking_non_zero_trylock(sem);
-}
diff --git a/arch/xtensa/kernel/xtensa_ksyms.c b/arch/xtensa/kernel/xtensa_ksyms.c
index 60dbdb43fb4c..6e52cdd6166f 100644
--- a/arch/xtensa/kernel/xtensa_ksyms.c
+++ b/arch/xtensa/kernel/xtensa_ksyms.c
@@ -26,7 +26,6 @@
 #include <asm/io.h>
 #include <asm/page.h>
 #include <asm/pgalloc.h>
-#include <asm/semaphore.h>
 #ifdef CONFIG_BLK_DEV_FD
 #include <asm/floppy.h>
 #endif
@@ -71,14 +70,6 @@ EXPORT_SYMBOL(__umodsi3);
 EXPORT_SYMBOL(__udivdi3);
 EXPORT_SYMBOL(__umoddi3);
 
-/*
- * Semaphore operations
- */
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_interruptible);
-EXPORT_SYMBOL(__down_trylock);
-EXPORT_SYMBOL(__up);
-
 #ifdef CONFIG_NET
 /*
  * Networking support
diff --git a/include/asm-alpha/semaphore.h b/include/asm-alpha/semaphore.h
index f1e9278a9fe2..d9b2034ed1d2 100644
--- a/include/asm-alpha/semaphore.h
+++ b/include/asm-alpha/semaphore.h
@@ -1,149 +1 @@
-#ifndef _ALPHA_SEMAPHORE_H
-#define _ALPHA_SEMAPHORE_H
-
-/*
- * SMP- and interrupt-safe semaphores..
- *
- * (C) Copyright 1996 Linus Torvalds
- * (C) Copyright 1996, 2000 Richard Henderson
- */
-
-#include <asm/current.h>
-#include <asm/system.h>
-#include <asm/atomic.h>
-#include <linux/compiler.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	atomic_t count;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)			\
-{								\
-	.count	= ATOMIC_INIT(n),				\
-  	.wait	= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait),	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count)		\
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name)		__DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init(struct semaphore *sem, int val)
-{
-	/*
-	 * Logically, 
-	 *   *sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
-	 * except that gcc produces better initializing by parts yet.
-	 */
-
-	atomic_set(&sem->count, val);
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void down(struct semaphore *);
-extern void __down_failed(struct semaphore *);
-extern int  down_interruptible(struct semaphore *);
-extern int  __down_failed_interruptible(struct semaphore *);
-extern int  down_trylock(struct semaphore *);
-extern void up(struct semaphore *);
-extern void __up_wakeup(struct semaphore *);
-
-/*
- * Hidden out of line code is fun, but extremely messy.  Rely on newer
- * compilers to do a respectable job with this.  The contention cases
- * are handled out of line in arch/alpha/kernel/semaphore.c.
- */
-
-static inline void __down(struct semaphore *sem)
-{
-	long count;
-	might_sleep();
-	count = atomic_dec_return(&sem->count);
-	if (unlikely(count < 0))
-		__down_failed(sem);
-}
-
-static inline int __down_interruptible(struct semaphore *sem)
-{
-	long count;
-	might_sleep();
-	count = atomic_dec_return(&sem->count);
-	if (unlikely(count < 0))
-		return __down_failed_interruptible(sem);
-	return 0;
-}
-
-/*
- * down_trylock returns 0 on success, 1 if we failed to get the lock.
- */
-
-static inline int __down_trylock(struct semaphore *sem)
-{
-	long ret;
-
-	/* "Equivalent" C:
-
-	   do {
-		ret = ldl_l;
-		--ret;
-		if (ret < 0)
-			break;
-		ret = stl_c = ret;
-	   } while (ret == 0);
-	*/
-	__asm__ __volatile__(
-		"1:	ldl_l	%0,%1\n"
-		"	subl	%0,1,%0\n"
-		"	blt	%0,2f\n"
-		"	stl_c	%0,%1\n"
-		"	beq	%0,3f\n"
-		"	mb\n"
-		"2:\n"
-		".subsection 2\n"
-		"3:	br	1b\n"
-		".previous"
-		: "=&r" (ret), "=m" (sem->count)
-		: "m" (sem->count));
-
-	return ret < 0;
-}
-
-static inline void __up(struct semaphore *sem)
-{
-	if (unlikely(atomic_inc_return(&sem->count) <= 0))
-		__up_wakeup(sem);
-}
-
-#if !defined(CONFIG_DEBUG_SEMAPHORE)
-extern inline void down(struct semaphore *sem)
-{
-	__down(sem);
-}
-extern inline int down_interruptible(struct semaphore *sem)
-{
-	return __down_interruptible(sem);
-}
-extern inline int down_trylock(struct semaphore *sem)
-{
-	return __down_trylock(sem);
-}
-extern inline void up(struct semaphore *sem)
-{
-	__up(sem);
-}
-#endif
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-arm/semaphore-helper.h b/include/asm-arm/semaphore-helper.h
deleted file mode 100644
index 1d7f1987edb9..000000000000
--- a/include/asm-arm/semaphore-helper.h
+++ /dev/null
@@ -1,84 +0,0 @@
-#ifndef ASMARM_SEMAPHORE_HELPER_H
-#define ASMARM_SEMAPHORE_HELPER_H
-
-/*
- * These two _must_ execute atomically wrt each other.
- */
-static inline void wake_one_more(struct semaphore * sem)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (atomic_read(&sem->count) <= 0)
-		sem->waking++;
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-}
-
-static inline int waking_non_zero(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->waking > 0) {
-		sem->waking--;
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking non zero interruptible
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- *
- * We must undo the sem->count down_interruptible() increment while we are
- * protected by the spinlock in order to make this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static inline int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->waking > 0) {
-		sem->waking--;
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;	
-}
-
-/*
- * waking_non_zero_try_lock:
- *	1	failed to lock
- *	0	got the lock
- *
- * We must undo the sem->count down_interruptible() increment while we are
- * protected by the spinlock in order to make this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static inline int waking_non_zero_trylock(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 1;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->waking <= 0)
-		atomic_inc(&sem->count);
-	else {
-		sem->waking--;
-		ret = 0;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-#endif
diff --git a/include/asm-arm/semaphore.h b/include/asm-arm/semaphore.h
index 1c8b441f89e3..d9b2034ed1d2 100644
--- a/include/asm-arm/semaphore.h
+++ b/include/asm-arm/semaphore.h
@@ -1,98 +1 @@
-/*
- * linux/include/asm-arm/semaphore.h
- */
-#ifndef __ASM_ARM_SEMAPHORE_H
-#define __ASM_ARM_SEMAPHORE_H
-
-#include <linux/linkage.h>
-#include <linux/spinlock.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-#include <asm/atomic.h>
-#include <asm/locks.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INIT(name, cnt)				\
-{								\
-	.count	= ATOMIC_INIT(cnt),				\
-	.wait	= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait),	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count)	\
-	struct semaphore name = __SEMAPHORE_INIT(name,count)
-
-#define DECLARE_MUTEX(name)		__DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init(struct semaphore *sem, int val)
-{
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX(struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED(struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-/*
- * special register calling convention
- */
-asmlinkage void __down_failed(void);
-asmlinkage int  __down_interruptible_failed(void);
-asmlinkage int  __down_trylock_failed(void);
-asmlinkage void __up_wakeup(void);
-
-extern void __down(struct semaphore * sem);
-extern int  __down_interruptible(struct semaphore * sem);
-extern int  __down_trylock(struct semaphore * sem);
-extern void __up(struct semaphore * sem);
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "__down" is the actual routine that waits...
- */
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	__down_op(sem, __down_failed);
-}
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "__down_interruptible" is the actual routine that waits...
- */
-static inline int down_interruptible (struct semaphore * sem)
-{
-	might_sleep();
-	return __down_op_ret(sem, __down_interruptible_failed);
-}
-
-static inline int down_trylock(struct semaphore *sem)
-{
-	return __down_op_ret(sem, __down_trylock_failed);
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore * sem)
-{
-	__up_op(sem, __up_wakeup);
-}
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-avr32/semaphore.h b/include/asm-avr32/semaphore.h
index feaf1d453386..d9b2034ed1d2 100644
--- a/include/asm-avr32/semaphore.h
+++ b/include/asm-avr32/semaphore.h
@@ -1,108 +1 @@
-/*
- * SMP- and interrupt-safe semaphores.
- *
- * Copyright (C) 2006 Atmel Corporation
- *
- * Based on include/asm-i386/semaphore.h
- *   Copyright (C) 1996 Linus Torvalds
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#ifndef __ASM_AVR32_SEMAPHORE_H
-#define __ASM_AVR32_SEMAPHORE_H
-
-#include <linux/linkage.h>
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-void __down(struct semaphore * sem);
-int  __down_interruptible(struct semaphore * sem);
-void __up(struct semaphore * sem);
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "__down_failed" is a special asm handler that calls the C
- * routine that actually waits. See arch/i386/kernel/semaphore.c
- */
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	if (unlikely(atomic_dec_return (&sem->count) < 0))
-		__down (sem);
-}
-
-/*
- * Interruptible try to acquire a semaphore.  If we obtained
- * it, return zero.  If we were interrupted, returns -EINTR
- */
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-
-	might_sleep();
-	if (unlikely(atomic_dec_return (&sem->count) < 0))
-		ret = __down_interruptible (sem);
-	return ret;
-}
-
-/*
- * Non-blockingly attempt to down() a semaphore.
- * Returns zero if we acquired it
- */
-static inline int down_trylock(struct semaphore * sem)
-{
-	return atomic_dec_if_positive(&sem->count) < 0;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore * sem)
-{
-	if (unlikely(atomic_inc_return (&sem->count) <= 0))
-		__up (sem);
-}
-
-#endif /*__ASM_AVR32_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-blackfin/semaphore-helper.h b/include/asm-blackfin/semaphore-helper.h
deleted file mode 100644
index 9082b0dc3eb5..000000000000
--- a/include/asm-blackfin/semaphore-helper.h
+++ /dev/null
@@ -1,82 +0,0 @@
-/* Based on M68K version,	Lineo Inc.	May 2001 */
-
-#ifndef _BFIN_SEMAPHORE_HELPER_H
-#define _BFIN_SEMAPHORE_HELPER_H
-
-/*
- * SMP- and interrupt-safe semaphores helper functions.
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- */
-
-#include <asm/errno.h>
-
-/*
- * These two _must_ execute atomically wrt each other.
- */
-static inline void wake_one_more(struct semaphore *sem)
-{
-	atomic_inc(&sem->waking);
-}
-
-static inline int waking_non_zero(struct semaphore *sem)
-{
-	int ret;
-	unsigned long flags = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 0;
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- */
-static inline int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	int ret = 0;
-	unsigned long flags = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- */
-static inline int waking_non_zero_trylock(struct semaphore *sem)
-{
-	int ret = 1;
-	unsigned long flags = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 0;
-	} else
-		atomic_inc(&sem->count);
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-#endif				/* _BFIN_SEMAPHORE_HELPER_H */
diff --git a/include/asm-blackfin/semaphore.h b/include/asm-blackfin/semaphore.h
index 533f90fb2e4e..d9b2034ed1d2 100644
--- a/include/asm-blackfin/semaphore.h
+++ b/include/asm-blackfin/semaphore.h
@@ -1,105 +1 @@
-#ifndef _BFIN_SEMAPHORE_H
-#define _BFIN_SEMAPHORE_H
-
-#ifndef __ASSEMBLY__
-
-#include <linux/linkage.h>
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-#include <asm/atomic.h>
-
-/*
- * Interrupt-safe semaphores..
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * BFIN version by akbar hussain Lineo Inc  April 2001
- *
- */
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init(struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER(*sem, val);
-}
-
-static inline void init_MUTEX(struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED(struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down(struct semaphore *sem);
-asmlinkage int __down_interruptible(struct semaphore *sem);
-asmlinkage int __down_trylock(struct semaphore *sem);
-asmlinkage void __up(struct semaphore *sem);
-
-extern spinlock_t semaphore_wake_lock;
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "down_failed" is a special asm handler that calls the C
- * routine that actually waits.
- */
-static inline void down(struct semaphore *sem)
-{
-	might_sleep();
-	if (atomic_dec_return(&sem->count) < 0)
-		__down(sem);
-}
-
-static inline int down_interruptible(struct semaphore *sem)
-{
-	int ret = 0;
-
-	might_sleep();
-	if (atomic_dec_return(&sem->count) < 0)
-		ret = __down_interruptible(sem);
-	return (ret);
-}
-
-static inline int down_trylock(struct semaphore *sem)
-{
-	int ret = 0;
-
-	if (atomic_dec_return(&sem->count) < 0)
-		ret = __down_trylock(sem);
-	return ret;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore *sem)
-{
-	if (atomic_inc_return(&sem->count) <= 0)
-		__up(sem);
-}
-
-#endif				/* __ASSEMBLY__ */
-#endif				/* _BFIN_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-cris/semaphore-helper.h b/include/asm-cris/semaphore-helper.h
deleted file mode 100644
index 27bfeca1b981..000000000000
--- a/include/asm-cris/semaphore-helper.h
+++ /dev/null
@@ -1,78 +0,0 @@
-/* $Id: semaphore-helper.h,v 1.3 2001/03/26 15:00:33 orjanf Exp $
- *
- * SMP- and interrupt-safe semaphores helper functions. Generic versions, no
- * optimizations whatsoever... 
- *
- */
-
-#ifndef _ASM_SEMAPHORE_HELPER_H
-#define _ASM_SEMAPHORE_HELPER_H
-
-#include <asm/atomic.h>
-#include <linux/errno.h>
-
-#define read(a) ((a)->counter)
-#define inc(a) (((a)->counter)++)
-#define dec(a) (((a)->counter)--)
-
-#define count_inc(a) ((*(a))++)
-
-/*
- * These two _must_ execute atomically wrt each other.
- */
-static inline void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc(&sem->waking);
-}
-
-static inline int waking_non_zero(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	local_irq_save(flags);
-	if (read(&sem->waking) > 0) {
-		dec(&sem->waking);
-		ret = 1;
-	}
-	local_irq_restore(flags);
-	return ret;
-}
-
-static inline int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	int ret = 0;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	if (read(&sem->waking) > 0) {
-		dec(&sem->waking);
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		inc(&sem->count);
-		ret = -EINTR;
-	}
-	local_irq_restore(flags);
-	return ret;
-}
-
-static inline int waking_non_zero_trylock(struct semaphore *sem)
-{
-        int ret = 1;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	if (read(&sem->waking) <= 0)
-		inc(&sem->count);
-	else {
-		dec(&sem->waking);
-		ret = 0;
-	}
-	local_irq_restore(flags);
-	return ret;
-}
-
-#endif /* _ASM_SEMAPHORE_HELPER_H */
-
-
diff --git a/include/asm-cris/semaphore.h b/include/asm-cris/semaphore.h
index 31a4ac448195..d9b2034ed1d2 100644
--- a/include/asm-cris/semaphore.h
+++ b/include/asm-cris/semaphore.h
@@ -1,133 +1 @@
-/* $Id: semaphore.h,v 1.3 2001/05/08 13:54:09 bjornw Exp $ */
-
-/* On the i386 these are coded in asm, perhaps we should as well. Later.. */
-
-#ifndef _CRIS_SEMAPHORE_H
-#define _CRIS_SEMAPHORE_H
-
-#define RW_LOCK_BIAS             0x01000000
-
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-/*
- * CRIS semaphores, implemented in C-only so far. 
- */
-
-struct semaphore {
-	atomic_t count;
-	atomic_t waking;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.waking		= ATOMIC_INIT(0),				\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)    \
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-        struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init(struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-        sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-        sema_init(sem, 0);
-}
-
-extern void __down(struct semaphore * sem);
-extern int __down_interruptible(struct semaphore * sem);
-extern int __down_trylock(struct semaphore * sem);
-extern void __up(struct semaphore * sem);
-
-/* notice - we probably can do cli/sti here instead of saving */
-
-static inline void down(struct semaphore * sem)
-{
-	unsigned long flags;
-	int failed;
-
-	might_sleep();
-
-	/* atomically decrement the semaphores count, and if its negative, we wait */
-	cris_atomic_save(sem, flags);
-	failed = --(sem->count.counter) < 0;
-	cris_atomic_restore(sem, flags);
-	if(failed) {
-		__down(sem);
-	}
-}
-
-/*
- * This version waits in interruptible state so that the waiting
- * process can be killed.  The down_interruptible routine
- * returns negative for signalled and zero for semaphore acquired.
- */
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	unsigned long flags;
-	int failed;
-
-	might_sleep();
-
-	/* atomically decrement the semaphores count, and if its negative, we wait */
-	cris_atomic_save(sem, flags);
-	failed = --(sem->count.counter) < 0;
-	cris_atomic_restore(sem, flags);
-	if(failed)
-		failed = __down_interruptible(sem);
-	return(failed);
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	unsigned long flags;
-	int failed;
-
-	cris_atomic_save(sem, flags);
-	failed = --(sem->count.counter) < 0;
-	cris_atomic_restore(sem, flags);
-	if(failed)
-		failed = __down_trylock(sem);
-	return(failed);
-
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore * sem)
-{  
-	unsigned long flags;
-	int wakeup;
-
-	/* atomically increment the semaphores count, and if it was negative, we wake people */
-	cris_atomic_save(sem, flags);
-	wakeup = ++(sem->count.counter) <= 0;
-	cris_atomic_restore(sem, flags);
-	if(wakeup) {
-		__up(sem);
-	}
-}
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-frv/semaphore.h b/include/asm-frv/semaphore.h
index d7aaa1911a1a..d9b2034ed1d2 100644
--- a/include/asm-frv/semaphore.h
+++ b/include/asm-frv/semaphore.h
@@ -1,155 +1 @@
-/* semaphore.h: semaphores for the FR-V
- *
- * Copyright (C) 2003 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-#ifndef _ASM_SEMAPHORE_H
-#define _ASM_SEMAPHORE_H
-
-#define RW_LOCK_BIAS		 0x01000000
-
-#ifndef __ASSEMBLY__
-
-#include <linux/linkage.h>
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-
-/*
- * the semaphore definition
- * - if counter is >0 then there are tokens available on the semaphore for down to collect
- * - if counter is <=0 then there are no spare tokens, and anyone that wants one must wait
- * - if wait_list is not empty, then there are processes waiting for the semaphore
- */
-struct semaphore {
-	unsigned		counter;
-	spinlock_t		wait_lock;
-	struct list_head	wait_list;
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	unsigned		__magic;
-#endif
-};
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-# define __SEM_DEBUG_INIT(name) , (long)&(name).__magic
-#else
-# define __SEM_DEBUG_INIT(name)
-#endif
-
-
-#define __SEMAPHORE_INITIALIZER(name,count) \
-{ count, SPIN_LOCK_UNLOCKED, LIST_HEAD_INIT((name).wait_list) __SEM_DEBUG_INIT(name) }
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore) __SEMAPHORE_INITIALIZER(*sem, val);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void __down(struct semaphore *sem, unsigned long flags);
-extern int  __down_interruptible(struct semaphore *sem, unsigned long flags);
-extern void __up(struct semaphore *sem);
-
-static inline void down(struct semaphore *sem)
-{
-	unsigned long flags;
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	if (likely(sem->counter > 0)) {
-		sem->counter--;
-		spin_unlock_irqrestore(&sem->wait_lock, flags);
-	}
-	else {
-		__down(sem, flags);
-	}
-}
-
-static inline int down_interruptible(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 0;
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	if (likely(sem->counter > 0)) {
-		sem->counter--;
-		spin_unlock_irqrestore(&sem->wait_lock, flags);
-	}
-	else {
-		ret = __down_interruptible(sem, flags);
-	}
-	return ret;
-}
-
-/*
- * non-blockingly attempt to down() a semaphore.
- * - returns zero if we acquired it
- */
-static inline int down_trylock(struct semaphore *sem)
-{
-	unsigned long flags;
-	int success = 0;
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	if (sem->counter > 0) {
-		sem->counter--;
-		success = 1;
-	}
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-	return !success;
-}
-
-static inline void up(struct semaphore *sem)
-{
-	unsigned long flags;
-
-#ifdef CONFIG_DEBUG_SEMAPHORE
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	if (!list_empty(&sem->wait_list))
-		__up(sem);
-	else
-		sem->counter++;
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-}
-
-static inline int sem_getcount(struct semaphore *sem)
-{
-	return sem->counter;
-}
-
-#endif /* __ASSEMBLY__ */
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-h8300/semaphore-helper.h b/include/asm-h8300/semaphore-helper.h
deleted file mode 100644
index 4fea36be5fd8..000000000000
--- a/include/asm-h8300/semaphore-helper.h
+++ /dev/null
@@ -1,85 +0,0 @@
-#ifndef _H8300_SEMAPHORE_HELPER_H
-#define _H8300_SEMAPHORE_HELPER_H
-
-/*
- * SMP- and interrupt-safe semaphores helper functions.
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * based on
- * m68k version by Andreas Schwab
- */
-
-#include <linux/errno.h>
-
-/*
- * These two _must_ execute atomically wrt each other.
- */
-static inline void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc((atomic_t *)&sem->sleepers);
-}
-
-static inline int waking_non_zero(struct semaphore *sem)
-{
-	int ret;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 0;
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- */
-static inline int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	int ret;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 0;
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- */
-static inline int waking_non_zero_trylock(struct semaphore *sem)
-{
-	int ret;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 1;
-	if (sem->sleepers <= 0)
-		atomic_inc(&sem->count);
-	else {
-		sem->sleepers--;
-		ret = 0;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-#endif
diff --git a/include/asm-h8300/semaphore.h b/include/asm-h8300/semaphore.h
index f3ffff83ff09..d9b2034ed1d2 100644
--- a/include/asm-h8300/semaphore.h
+++ b/include/asm-h8300/semaphore.h
@@ -1,190 +1 @@
-#ifndef _H8300_SEMAPHORE_H
-#define _H8300_SEMAPHORE_H
-
-#define RW_LOCK_BIAS		 0x01000000
-
-#ifndef __ASSEMBLY__
-
-#include <linux/linkage.h>
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-/*
- * Interrupt-safe semaphores..
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * H8/300 version by Yoshinori Sato
- */
-
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER(*sem, val);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down_failed(void /* special register calling convention */);
-asmlinkage int  __down_failed_interruptible(void  /* params in registers */);
-asmlinkage int  __down_failed_trylock(void  /* params in registers */);
-asmlinkage void __up_wakeup(void /* special register calling convention */);
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-extern spinlock_t semaphore_wake_lock;
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "down_failed" is a special asm handler that calls the C
- * routine that actually waits. See arch/m68k/lib/semaphore.S
- */
-static inline void down(struct semaphore * sem)
-{
-	register atomic_t *count asm("er0");
-
-	might_sleep();
-
-	count = &(sem->count);
-	__asm__ __volatile__(
-		"stc ccr,r3l\n\t"
-		"orc #0x80,ccr\n\t"
-		"mov.l %2, er1\n\t"
-		"dec.l #1,er1\n\t"
-		"mov.l er1,%0\n\t"
-		"bpl 1f\n\t"
-		"ldc r3l,ccr\n\t"
-		"mov.l %1,er0\n\t"
-		"jsr @___down\n\t"
-		"bra 2f\n"
-		"1:\n\t"
-		"ldc r3l,ccr\n"
-		"2:"
-		: "=m"(*count)
-		: "g"(sem),"m"(*count)
-		: "cc",  "er1", "er2", "er3");
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	register atomic_t *count asm("er0");
-
-	might_sleep();
-
-	count = &(sem->count);
-	__asm__ __volatile__(
-		"stc ccr,r1l\n\t"
-		"orc #0x80,ccr\n\t"
-		"mov.l %3, er2\n\t"
-		"dec.l #1,er2\n\t"
-		"mov.l er2,%1\n\t"
-		"bpl 1f\n\t"
-		"ldc r1l,ccr\n\t"
-		"mov.l %2,er0\n\t"
-		"jsr @___down_interruptible\n\t"
-		"bra 2f\n"
-		"1:\n\t"
-		"ldc r1l,ccr\n\t"
-		"sub.l %0,%0\n\t"
-		"2:\n\t"
-		: "=r" (count),"=m" (*count)
-		: "g"(sem),"m"(*count)
-		: "cc", "er1", "er2", "er3");
-	return (int)count;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	register atomic_t *count asm("er0");
-
-	count = &(sem->count);
-	__asm__ __volatile__(
-		"stc ccr,r3l\n\t"
-		"orc #0x80,ccr\n\t"
-		"mov.l %3,er2\n\t"
-		"dec.l #1,er2\n\t"
-		"mov.l er2,%0\n\t"
-		"bpl 1f\n\t"
-		"ldc r3l,ccr\n\t"
-		"jmp @3f\n\t"
-		LOCK_SECTION_START(".align 2\n\t")
-		"3:\n\t"
-		"mov.l %2,er0\n\t"
-		"jsr @___down_trylock\n\t"
-		"jmp @2f\n\t"
-		LOCK_SECTION_END
-		"1:\n\t"
-		"ldc r3l,ccr\n\t"
-		"sub.l %1,%1\n"
-		"2:"
-		: "=m" (*count),"=r"(count)
-		: "g"(sem),"m"(*count)
-		: "cc", "er1","er2", "er3");
-	return (int)count;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore * sem)
-{
-	register atomic_t *count asm("er0");
-
-	count = &(sem->count);
-	__asm__ __volatile__(
-		"stc ccr,r3l\n\t"
-		"orc #0x80,ccr\n\t"
-		"mov.l %2,er1\n\t"
-		"inc.l #1,er1\n\t"
-		"mov.l er1,%0\n\t"
-		"ldc r3l,ccr\n\t"
-		"sub.l er2,er2\n\t"
-		"cmp.l er2,er1\n\t"
-		"bgt 1f\n\t"
-		"mov.l %1,er0\n\t"
-		"jsr @___up\n"
-		"1:"
-		: "=m"(*count)
-		: "g"(sem),"m"(*count)
-		: "cc", "er1", "er2", "er3");
-}
-
-#endif /* __ASSEMBLY__ */
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-ia64/semaphore.h b/include/asm-ia64/semaphore.h
index d8393d11288d..d9b2034ed1d2 100644
--- a/include/asm-ia64/semaphore.h
+++ b/include/asm-ia64/semaphore.h
@@ -1,99 +1 @@
-#ifndef _ASM_IA64_SEMAPHORE_H
-#define _ASM_IA64_SEMAPHORE_H
-
-/*
- * Copyright (C) 1998-2000 Hewlett-Packard Co
- * Copyright (C) 1998-2000 David Mosberger-Tang <davidm@hpl.hp.com>
- */
-
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-#include <asm/atomic.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count)					\
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name, count)
-
-#define DECLARE_MUTEX(name)		__DECLARE_SEMAPHORE_GENERIC(name, 1)
-
-static inline void
-sema_init (struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore) __SEMAPHORE_INITIALIZER(*sem, val);
-}
-
-static inline void
-init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void
-init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void __down (struct semaphore * sem);
-extern int  __down_interruptible (struct semaphore * sem);
-extern int  __down_trylock (struct semaphore * sem);
-extern void __up (struct semaphore * sem);
-
-/*
- * Atomically decrement the semaphore's count.  If it goes negative,
- * block the calling thread in the TASK_UNINTERRUPTIBLE state.
- */
-static inline void
-down (struct semaphore *sem)
-{
-	might_sleep();
-	if (ia64_fetchadd(-1, &sem->count.counter, acq) < 1)
-		__down(sem);
-}
-
-/*
- * Atomically decrement the semaphore's count.  If it goes negative,
- * block the calling thread in the TASK_INTERRUPTIBLE state.
- */
-static inline int
-down_interruptible (struct semaphore * sem)
-{
-	int ret = 0;
-
-	might_sleep();
-	if (ia64_fetchadd(-1, &sem->count.counter, acq) < 1)
-		ret = __down_interruptible(sem);
-	return ret;
-}
-
-static inline int
-down_trylock (struct semaphore *sem)
-{
-	int ret = 0;
-
-	if (ia64_fetchadd(-1, &sem->count.counter, acq) < 1)
-		ret = __down_trylock(sem);
-	return ret;
-}
-
-static inline void
-up (struct semaphore * sem)
-{
-	if (ia64_fetchadd(1, &sem->count.counter, rel) <= -1)
-		__up(sem);
-}
-
-#endif /* _ASM_IA64_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-m32r/semaphore.h b/include/asm-m32r/semaphore.h
index b5bf95a6f2b4..d9b2034ed1d2 100644
--- a/include/asm-m32r/semaphore.h
+++ b/include/asm-m32r/semaphore.h
@@ -1,144 +1 @@
-#ifndef _ASM_M32R_SEMAPHORE_H
-#define _ASM_M32R_SEMAPHORE_H
-
-#include <linux/linkage.h>
-
-#ifdef __KERNEL__
-
-/*
- * SMP- and interrupt-safe semaphores..
- *
- * Copyright (C) 1996  Linus Torvalds
- * Copyright (C) 2004, 2006  Hirokazu Takata <takata at linux-m32r.org>
- */
-
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-#include <asm/assembler.h>
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-/*
- *	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
- *
- * i'd rather use the more flexible initialization above, but sadly
- * GCC 2.7.2.3 emits a bogus warning. EGCS doesnt. Oh well.
- */
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down_failed(void /* special register calling convention */);
-asmlinkage int  __down_failed_interruptible(void  /* params in registers */);
-asmlinkage int  __down_failed_trylock(void  /* params in registers */);
-asmlinkage void __up_wakeup(void /* special register calling convention */);
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-/*
- * Atomically decrement the semaphore's count.  If it goes negative,
- * block the calling thread in the TASK_UNINTERRUPTIBLE state.
- */
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	if (unlikely(atomic_dec_return(&sem->count) < 0))
-		__down(sem);
-}
-
-/*
- * Interruptible try to acquire a semaphore.  If we obtained
- * it, return zero.  If we were interrupted, returns -EINTR
- */
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int result = 0;
-
-	might_sleep();
-	if (unlikely(atomic_dec_return(&sem->count) < 0))
-		result = __down_interruptible(sem);
-
-	return result;
-}
-
-/*
- * Non-blockingly attempt to down() a semaphore.
- * Returns zero if we acquired it
- */
-static inline int down_trylock(struct semaphore * sem)
-{
-	unsigned long flags;
-	long count;
-	int result = 0;
-
-	local_irq_save(flags);
-	__asm__ __volatile__ (
-		"# down_trylock			\n\t"
-		DCACHE_CLEAR("%0", "r4", "%1")
-		M32R_LOCK" %0, @%1;		\n\t"
-		"addi	%0, #-1;		\n\t"
-		M32R_UNLOCK" %0, @%1;		\n\t"
-		: "=&r" (count)
-		: "r" (&sem->count)
-		: "memory"
-#ifdef CONFIG_CHIP_M32700_TS1
-		, "r4"
-#endif	/* CONFIG_CHIP_M32700_TS1 */
-	);
-	local_irq_restore(flags);
-
-	if (unlikely(count < 0))
-		result = __down_trylock(sem);
-
-	return result;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore * sem)
-{
-	if (unlikely(atomic_inc_return(&sem->count) <= 0))
-		__up(sem);
-}
-
-#endif  /* __KERNEL__ */
-
-#endif  /* _ASM_M32R_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-m68k/semaphore-helper.h b/include/asm-m68k/semaphore-helper.h
deleted file mode 100644
index eef30ba0b499..000000000000
--- a/include/asm-m68k/semaphore-helper.h
+++ /dev/null
@@ -1,142 +0,0 @@
-#ifndef _M68K_SEMAPHORE_HELPER_H
-#define _M68K_SEMAPHORE_HELPER_H
-
-/*
- * SMP- and interrupt-safe semaphores helper functions.
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * m68k version by Andreas Schwab
- */
-
-#include <linux/errno.h>
-
-/*
- * These two _must_ execute atomically wrt each other.
- */
-static inline void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc(&sem->waking);
-}
-
-#ifndef CONFIG_RMW_INSNS
-extern spinlock_t semaphore_wake_lock;
-#endif
-
-static inline int waking_non_zero(struct semaphore *sem)
-{
-	int ret;
-#ifndef CONFIG_RMW_INSNS
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 0;
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-#else
-	int tmp1, tmp2;
-
-	__asm__ __volatile__
-	  ("1:	movel	%1,%2\n"
-	   "    jle	2f\n"
-	   "	subql	#1,%2\n"
-	   "	casl	%1,%2,%3\n"
-	   "	jne	1b\n"
-	   "	moveq	#1,%0\n"
-	   "2:"
-	   : "=d" (ret), "=d" (tmp1), "=d" (tmp2)
-	   : "m" (sem->waking), "0" (0), "1" (sem->waking));
-#endif
-
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- */
-static inline int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	int ret;
-#ifndef CONFIG_RMW_INSNS
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 0;
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-#else
-	int tmp1, tmp2;
-
-	__asm__ __volatile__
-	  ("1:	movel	%1,%2\n"
-	   "	jle	2f\n"
-	   "	subql	#1,%2\n"
-	   "	casl	%1,%2,%3\n"
-	   "	jne	1b\n"
-	   "	moveq	#1,%0\n"
-	   "	jra	%a4\n"
-	   "2:"
-	   : "=d" (ret), "=d" (tmp1), "=d" (tmp2)
-	   : "m" (sem->waking), "i" (&&next), "0" (0), "1" (sem->waking));
-	if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-next:
-#endif
-
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- */
-static inline int waking_non_zero_trylock(struct semaphore *sem)
-{
-	int ret;
-#ifndef CONFIG_RMW_INSNS
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 1;
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 0;
-	} else
-		atomic_inc(&sem->count);
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-#else
-	int tmp1, tmp2;
-
-	__asm__ __volatile__
-	  ("1:	movel	%1,%2\n"
-	   "    jle	2f\n"
-	   "	subql	#1,%2\n"
-	   "	casl	%1,%2,%3\n"
-	   "	jne	1b\n"
-	   "	moveq	#0,%0\n"
-	   "2:"
-	   : "=d" (ret), "=d" (tmp1), "=d" (tmp2)
-	   : "m" (sem->waking), "0" (1), "1" (sem->waking));
-	if (ret)
-		atomic_inc(&sem->count);
-#endif
-	return ret;
-}
-
-#endif
diff --git a/include/asm-m68k/semaphore.h b/include/asm-m68k/semaphore.h
index 64d6b119bb0a..d9b2034ed1d2 100644
--- a/include/asm-m68k/semaphore.h
+++ b/include/asm-m68k/semaphore.h
@@ -1,163 +1 @@
-#ifndef _M68K_SEMAPHORE_H
-#define _M68K_SEMAPHORE_H
-
-#define RW_LOCK_BIAS		 0x01000000
-
-#ifndef __ASSEMBLY__
-
-#include <linux/linkage.h>
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-#include <linux/stringify.h>
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-/*
- * Interrupt-safe semaphores..
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * m68k version by Andreas Schwab
- */
-
-
-struct semaphore {
-	atomic_t count;
-	atomic_t waking;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.waking		= ATOMIC_INIT(0),				\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init(struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER(*sem, val);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down_failed(void /* special register calling convention */);
-asmlinkage int  __down_failed_interruptible(void  /* params in registers */);
-asmlinkage int  __down_failed_trylock(void  /* params in registers */);
-asmlinkage void __up_wakeup(void /* special register calling convention */);
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "down_failed" is a special asm handler that calls the C
- * routine that actually waits. See arch/m68k/lib/semaphore.S
- */
-static inline void down(struct semaphore *sem)
-{
-	register struct semaphore *sem1 __asm__ ("%a1") = sem;
-
-	might_sleep();
-	__asm__ __volatile__(
-		"| atomic down operation\n\t"
-		"subql #1,%0@\n\t"
-		"jmi 2f\n\t"
-		"1:\n"
-		LOCK_SECTION_START(".even\n\t")
-		"2:\tpea 1b\n\t"
-		"jbra __down_failed\n"
-		LOCK_SECTION_END
-		: /* no outputs */
-		: "a" (sem1)
-		: "memory");
-}
-
-static inline int down_interruptible(struct semaphore *sem)
-{
-	register struct semaphore *sem1 __asm__ ("%a1") = sem;
-	register int result __asm__ ("%d0");
-
-	might_sleep();
-	__asm__ __volatile__(
-		"| atomic interruptible down operation\n\t"
-		"subql #1,%1@\n\t"
-		"jmi 2f\n\t"
-		"clrl %0\n"
-		"1:\n"
-		LOCK_SECTION_START(".even\n\t")
-		"2:\tpea 1b\n\t"
-		"jbra __down_failed_interruptible\n"
-		LOCK_SECTION_END
-		: "=d" (result)
-		: "a" (sem1)
-		: "memory");
-	return result;
-}
-
-static inline int down_trylock(struct semaphore *sem)
-{
-	register struct semaphore *sem1 __asm__ ("%a1") = sem;
-	register int result __asm__ ("%d0");
-
-	__asm__ __volatile__(
-		"| atomic down trylock operation\n\t"
-		"subql #1,%1@\n\t"
-		"jmi 2f\n\t"
-		"clrl %0\n"
-		"1:\n"
-		LOCK_SECTION_START(".even\n\t")
-		"2:\tpea 1b\n\t"
-		"jbra __down_failed_trylock\n"
-		LOCK_SECTION_END
-		: "=d" (result)
-		: "a" (sem1)
-		: "memory");
-	return result;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore *sem)
-{
-	register struct semaphore *sem1 __asm__ ("%a1") = sem;
-
-	__asm__ __volatile__(
-		"| atomic up operation\n\t"
-		"addql #1,%0@\n\t"
-		"jle 2f\n"
-		"1:\n"
-		LOCK_SECTION_START(".even\n\t")
-		"2:\t"
-		"pea 1b\n\t"
-		"jbra __up_wakeup\n"
-		LOCK_SECTION_END
-		: /* no outputs */
-		: "a" (sem1)
-		: "memory");
-}
-
-#endif /* __ASSEMBLY__ */
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-m68knommu/semaphore-helper.h b/include/asm-m68knommu/semaphore-helper.h
deleted file mode 100644
index 43da7bc483c7..000000000000
--- a/include/asm-m68knommu/semaphore-helper.h
+++ /dev/null
@@ -1,82 +0,0 @@
-#ifndef _M68K_SEMAPHORE_HELPER_H
-#define _M68K_SEMAPHORE_HELPER_H
-
-/*
- * SMP- and interrupt-safe semaphores helper functions.
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * m68k version by Andreas Schwab
- */
-
-
-/*
- * These two _must_ execute atomically wrt each other.
- */
-static inline void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc(&sem->waking);
-}
-
-static inline int waking_non_zero(struct semaphore *sem)
-{
-	int ret;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 0;
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- */
-static inline int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	int ret;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 0;
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- */
-static inline int waking_non_zero_trylock(struct semaphore *sem)
-{
-	int ret;
-	unsigned long flags;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	ret = 1;
-	if (atomic_read(&sem->waking) > 0) {
-		atomic_dec(&sem->waking);
-		ret = 0;
-	} else
-		atomic_inc(&sem->count);
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-#endif
diff --git a/include/asm-m68knommu/semaphore.h b/include/asm-m68knommu/semaphore.h
index 5779eb6c0689..d9b2034ed1d2 100644
--- a/include/asm-m68knommu/semaphore.h
+++ b/include/asm-m68knommu/semaphore.h
@@ -1,153 +1 @@
-#ifndef _M68K_SEMAPHORE_H
-#define _M68K_SEMAPHORE_H
-
-#define RW_LOCK_BIAS		 0x01000000
-
-#ifndef __ASSEMBLY__
-
-#include <linux/linkage.h>
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-/*
- * Interrupt-safe semaphores..
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * m68k version by Andreas Schwab
- */
-
-
-struct semaphore {
-	atomic_t count;
-	atomic_t waking;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.waking		= ATOMIC_INIT(0),				\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER(*sem, val);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down_failed(void /* special register calling convention */);
-asmlinkage int  __down_failed_interruptible(void  /* params in registers */);
-asmlinkage int  __down_failed_trylock(void  /* params in registers */);
-asmlinkage void __up_wakeup(void /* special register calling convention */);
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-extern spinlock_t semaphore_wake_lock;
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "down_failed" is a special asm handler that calls the C
- * routine that actually waits. See arch/m68k/lib/semaphore.S
- */
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	__asm__ __volatile__(
-		"| atomic down operation\n\t"
-		"movel	%0, %%a1\n\t"
-		"lea	%%pc@(1f), %%a0\n\t"
-		"subql	#1, %%a1@\n\t"
-		"jmi __down_failed\n"
-		"1:"
-		: /* no outputs */
-		: "g" (sem)
-		: "cc", "%a0", "%a1", "memory");
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret;
-
-	might_sleep();
-	__asm__ __volatile__(
-		"| atomic down operation\n\t"
-		"movel	%1, %%a1\n\t"
-		"lea	%%pc@(1f), %%a0\n\t"
-		"subql	#1, %%a1@\n\t"
-		"jmi __down_failed_interruptible\n\t"
-		"clrl	%%d0\n"
-		"1: movel	%%d0, %0\n"
-		: "=d" (ret)
-		: "g" (sem)
-		: "cc", "%d0", "%a0", "%a1", "memory");
-	return(ret);
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	register struct semaphore *sem1 __asm__ ("%a1") = sem;
-	register int result __asm__ ("%d0");
-
-	__asm__ __volatile__(
-		"| atomic down trylock operation\n\t"
-		"subql #1,%1@\n\t"
-		"jmi 2f\n\t"
-		"clrl %0\n"
-		"1:\n"
-		".section .text.lock,\"ax\"\n"
-		".even\n"
-		"2:\tpea 1b\n\t"
-		"jbra __down_failed_trylock\n"
-		".previous"
-		: "=d" (result)
-		: "a" (sem1)
-		: "memory");
-	return result;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore * sem)
-{
-	__asm__ __volatile__(
-		"| atomic up operation\n\t"
-		"movel	%0, %%a1\n\t"
-		"lea	%%pc@(1f), %%a0\n\t"
-		"addql	#1, %%a1@\n\t"
-		"jle __up_wakeup\n"
-		"1:"
-		: /* no outputs */
-		: "g" (sem)
-		: "cc", "%a0", "%a1", "memory");
-}
-
-#endif /* __ASSEMBLY__ */
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-mips/semaphore.h b/include/asm-mips/semaphore.h
index fdf8042b784b..d9b2034ed1d2 100644
--- a/include/asm-mips/semaphore.h
+++ b/include/asm-mips/semaphore.h
@@ -1,108 +1 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1996  Linus Torvalds
- * Copyright (C) 1998, 99, 2000, 01, 04  Ralf Baechle
- * Copyright (C) 1999, 2000, 01  Silicon Graphics, Inc.
- * Copyright (C) 2000, 01 MIPS Technologies, Inc.
- *
- * In all honesty, little of the old MIPS code left - the PPC64 variant was
- * just looking nice and portable so I ripped it.  Credits to whoever wrote
- * it.
- */
-#ifndef __ASM_SEMAPHORE_H
-#define __ASM_SEMAPHORE_H
-
-/*
- * Remove spinlock-based RW semaphores; RW semaphore definitions are
- * now in rwsem.h and we use the generic lib/rwsem.c implementation.
- * Rework semaphores to use atomic_dec_if_positive.
- * -- Paul Mackerras (paulus@samba.org)
- */
-
-#ifdef __KERNEL__
-
-#include <asm/atomic.h>
-#include <asm/system.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	/*
-	 * Note that any negative value of count is equivalent to 0,
-	 * but additionally indicates that some process(es) might be
-	 * sleeping on `wait'.
-	 */
-	atomic_t count;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name, count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name, count)
-
-#define DECLARE_MUTEX(name)		__DECLARE_SEMAPHORE_GENERIC(name, 1)
-
-static inline void sema_init(struct semaphore *sem, int val)
-{
-	atomic_set(&sem->count, val);
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX(struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED(struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void __down(struct semaphore * sem);
-extern int  __down_interruptible(struct semaphore * sem);
-extern void __up(struct semaphore * sem);
-
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-
-	/*
-	 * Try to get the semaphore, take the slow path if we fail.
-	 */
-	if (unlikely(atomic_dec_return(&sem->count) < 0))
-		__down(sem);
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-
-	might_sleep();
-
-	if (unlikely(atomic_dec_return(&sem->count) < 0))
-		ret = __down_interruptible(sem);
-	return ret;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	return atomic_dec_if_positive(&sem->count) < 0;
-}
-
-static inline void up(struct semaphore * sem)
-{
-	if (unlikely(atomic_inc_return(&sem->count) <= 0))
-		__up(sem);
-}
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-mn10300/semaphore.h b/include/asm-mn10300/semaphore.h
index 5a9e1ad0b253..d9b2034ed1d2 100644
--- a/include/asm-mn10300/semaphore.h
+++ b/include/asm-mn10300/semaphore.h
@@ -1,169 +1 @@
-/* MN10300 Semaphores
- *
- * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-#ifndef _ASM_SEMAPHORE_H
-#define _ASM_SEMAPHORE_H
-
-#ifndef __ASSEMBLY__
-
-#include <linux/linkage.h>
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-
-#define SEMAPHORE_DEBUG		0
-
-/*
- * the semaphore definition
- * - if count is >0 then there are tokens available on the semaphore for down
- *   to collect
- * - if count is <=0 then there are no spare tokens, and anyone that wants one
- *   must wait
- * - if wait_list is not empty, then there are processes waiting for the
- *   semaphore
- */
-struct semaphore {
-	atomic_t		count;		/* it's not really atomic, it's
-						 * just that certain modules
-						 * expect to be able to access
-						 * it directly */
-	spinlock_t		wait_lock;
-	struct list_head	wait_list;
-#if SEMAPHORE_DEBUG
-	unsigned		__magic;
-#endif
-};
-
-#if SEMAPHORE_DEBUG
-# define __SEM_DEBUG_INIT(name) , (long)&(name).__magic
-#else
-# define __SEM_DEBUG_INIT(name)
-#endif
-
-
-#define __SEMAPHORE_INITIALIZER(name, init_count)			\
-{									\
-	.count		= ATOMIC_INIT(init_count),			\
-	.wait_lock	= __SPIN_LOCK_UNLOCKED((name).wait_lock),	\
-	.wait_list	= LIST_HEAD_INIT((name).wait_list)		\
-	__SEM_DEBUG_INIT(name)						\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name, count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name, 1)
-#define DECLARE_MUTEX_LOCKED(name) __DECLARE_SEMAPHORE_GENERIC(name, 0)
-
-static inline void sema_init(struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore) __SEMAPHORE_INITIALIZER(*sem, val);
-}
-
-static inline void init_MUTEX(struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED(struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void __down(struct semaphore *sem, unsigned long flags);
-extern int  __down_interruptible(struct semaphore *sem, unsigned long flags);
-extern void __up(struct semaphore *sem);
-
-static inline void down(struct semaphore *sem)
-{
-	unsigned long flags;
-	int count;
-
-#if SEMAPHORE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	count = atomic_read(&sem->count);
-	if (likely(count > 0)) {
-		atomic_set(&sem->count, count - 1);
-		spin_unlock_irqrestore(&sem->wait_lock, flags);
-	} else {
-		__down(sem, flags);
-	}
-}
-
-static inline int down_interruptible(struct semaphore *sem)
-{
-	unsigned long flags;
-	int count, ret = 0;
-
-#if SEMAPHORE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	count = atomic_read(&sem->count);
-	if (likely(count > 0)) {
-		atomic_set(&sem->count, count - 1);
-		spin_unlock_irqrestore(&sem->wait_lock, flags);
-	} else {
-		ret = __down_interruptible(sem, flags);
-	}
-	return ret;
-}
-
-/*
- * non-blockingly attempt to down() a semaphore.
- * - returns zero if we acquired it
- */
-static inline int down_trylock(struct semaphore *sem)
-{
-	unsigned long flags;
-	int count, success = 0;
-
-#if SEMAPHORE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	count = atomic_read(&sem->count);
-	if (likely(count > 0)) {
-		atomic_set(&sem->count, count - 1);
-		success = 1;
-	}
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-	return !success;
-}
-
-static inline void up(struct semaphore *sem)
-{
-	unsigned long flags;
-
-#if SEMAPHORE_DEBUG
-	CHECK_MAGIC(sem->__magic);
-#endif
-
-	spin_lock_irqsave(&sem->wait_lock, flags);
-	if (!list_empty(&sem->wait_list))
-		__up(sem);
-	else
-		atomic_set(&sem->count, atomic_read(&sem->count) + 1);
-	spin_unlock_irqrestore(&sem->wait_lock, flags);
-}
-
-static inline int sem_getcount(struct semaphore *sem)
-{
-	return atomic_read(&sem->count);
-}
-
-#endif /* __ASSEMBLY__ */
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-parisc/semaphore-helper.h b/include/asm-parisc/semaphore-helper.h
deleted file mode 100644
index 387f7c1277a2..000000000000
--- a/include/asm-parisc/semaphore-helper.h
+++ /dev/null
@@ -1,89 +0,0 @@
-#ifndef _ASM_PARISC_SEMAPHORE_HELPER_H
-#define _ASM_PARISC_SEMAPHORE_HELPER_H
-
-/*
- * SMP- and interrupt-safe semaphores helper functions.
- *
- * (C) Copyright 1996 Linus Torvalds
- * (C) Copyright 1999 Andrea Arcangeli
- */
-
-/*
- * These two _must_ execute atomically wrt each other.
- *
- * This is trivially done with load_locked/store_cond,
- * which we have.  Let the rest of the losers suck eggs.
- */
-static __inline__ void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc((atomic_t *)&sem->waking);
-}
-
-static __inline__ int waking_non_zero(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->waking > 0) {
-		sem->waking--;
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- *
- * We must undo the sem->count down_interruptible() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static __inline__ int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->waking > 0) {
-		sem->waking--;
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- *
- * We must undo the sem->count down_trylock() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static __inline__ int waking_non_zero_trylock(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 1;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->waking <= 0)
-		atomic_inc(&sem->count);
-	else {
-		sem->waking--;
-		ret = 0;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-#endif /* _ASM_PARISC_SEMAPHORE_HELPER_H */
diff --git a/include/asm-parisc/semaphore.h b/include/asm-parisc/semaphore.h
index a16271cdc748..d9b2034ed1d2 100644
--- a/include/asm-parisc/semaphore.h
+++ b/include/asm-parisc/semaphore.h
@@ -1,145 +1 @@
-/*    SMP- and interrupt-safe semaphores.
- *    PA-RISC version by Matthew Wilcox
- *
- *    Linux/PA-RISC Project (http://www.parisc-linux.org/)
- *    Copyright (C) 1996 Linus Torvalds
- *    Copyright (C) 1999-2001 Matthew Wilcox < willy at debian d0T org >
- *    Copyright (C) 2000 Grant Grundler < grundler a debian org >
- *
- *    This program is free software; you can redistribute it and/or modify
- *    it under the terms of the GNU General Public License as published by
- *    the Free Software Foundation; either version 2 of the License, or
- *    (at your option) any later version.
- *
- *    This program is distributed in the hope that it will be useful,
- *    but WITHOUT ANY WARRANTY; without even the implied warranty of
- *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *    GNU General Public License for more details.
- *
- *    You should have received a copy of the GNU General Public License
- *    along with this program; if not, write to the Free Software
- *    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
- */
-
-#ifndef _ASM_PARISC_SEMAPHORE_H
-#define _ASM_PARISC_SEMAPHORE_H
-
-#include <linux/spinlock.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-#include <asm/system.h>
-
-/*
- * The `count' is initialised to the number of people who are allowed to
- * take the lock.  (Normally we want a mutex, so this is `1').  if
- * `count' is positive, the lock can be taken.  if it's 0, no-one is
- * waiting on it.  if it's -1, at least one task is waiting.
- */
-struct semaphore {
-	spinlock_t	sentry;
-	int		count;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.sentry		= SPIN_LOCK_UNLOCKED,				\
-	.count		= n,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-static inline int sem_getcount(struct semaphore *sem)
-{
-	return sem->count;
-}
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-/* Semaphores can be `tried' from irq context.  So we have to disable
- * interrupts while we're messing with the semaphore.  Sorry.
- */
-
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	spin_lock_irq(&sem->sentry);
-	if (sem->count > 0) {
-		sem->count--;
-	} else {
-		__down(sem);
-	}
-	spin_unlock_irq(&sem->sentry);
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-	might_sleep();
-	spin_lock_irq(&sem->sentry);
-	if (sem->count > 0) {
-		sem->count--;
-	} else {
-		ret = __down_interruptible(sem);
-	}
-	spin_unlock_irq(&sem->sentry);
-	return ret;
-}
-
-/*
- * down_trylock returns 0 on success, 1 if we failed to get the lock.
- * May not sleep, but must preserve irq state
- */
-static inline int down_trylock(struct semaphore * sem)
-{
-	unsigned long flags;
-	int count;
-
-	spin_lock_irqsave(&sem->sentry, flags);
-	count = sem->count - 1;
-	if (count >= 0)
-		sem->count = count;
-	spin_unlock_irqrestore(&sem->sentry, flags);
-	return (count < 0);
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- */
-static inline void up(struct semaphore * sem)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&sem->sentry, flags);
-	if (sem->count < 0) {
-		__up(sem);
-	} else {
-		sem->count++;
-	}
-	spin_unlock_irqrestore(&sem->sentry, flags);
-}
-
-#endif /* _ASM_PARISC_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-powerpc/semaphore.h b/include/asm-powerpc/semaphore.h
index 48dd32e07749..d9b2034ed1d2 100644
--- a/include/asm-powerpc/semaphore.h
+++ b/include/asm-powerpc/semaphore.h
@@ -1,94 +1 @@
-#ifndef _ASM_POWERPC_SEMAPHORE_H
-#define _ASM_POWERPC_SEMAPHORE_H
-
-/*
- * Remove spinlock-based RW semaphores; RW semaphore definitions are
- * now in rwsem.h and we use the generic lib/rwsem.c implementation.
- * Rework semaphores to use atomic_dec_if_positive.
- * -- Paul Mackerras (paulus@samba.org)
- */
-
-#ifdef __KERNEL__
-
-#include <asm/atomic.h>
-#include <asm/system.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	/*
-	 * Note that any negative value of count is equivalent to 0,
-	 * but additionally indicates that some process(es) might be
-	 * sleeping on `wait'.
-	 */
-	atomic_t count;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name, count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name)		__DECLARE_SEMAPHORE_GENERIC(name, 1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	atomic_set(&sem->count, val);
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void __down(struct semaphore * sem);
-extern int  __down_interruptible(struct semaphore * sem);
-extern void __up(struct semaphore * sem);
-
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-
-	/*
-	 * Try to get the semaphore, take the slow path if we fail.
-	 */
-	if (unlikely(atomic_dec_return(&sem->count) < 0))
-		__down(sem);
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-
-	might_sleep();
-
-	if (unlikely(atomic_dec_return(&sem->count) < 0))
-		ret = __down_interruptible(sem);
-	return ret;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	return atomic_dec_if_positive(&sem->count) < 0;
-}
-
-static inline void up(struct semaphore * sem)
-{
-	if (unlikely(atomic_inc_return(&sem->count) <= 0))
-		__up(sem);
-}
-
-#endif /* __KERNEL__ */
-
-#endif /* _ASM_POWERPC_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-s390/semaphore.h b/include/asm-s390/semaphore.h
index 0e7001ad8392..d9b2034ed1d2 100644
--- a/include/asm-s390/semaphore.h
+++ b/include/asm-s390/semaphore.h
@@ -1,107 +1 @@
-/*
- *  include/asm-s390/semaphore.h
- *
- *  S390 version
- *    Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
- *
- *  Derived from "include/asm-i386/semaphore.h"
- *    (C) Copyright 1996 Linus Torvalds
- */
-
-#ifndef _S390_SEMAPHORE_H
-#define _S390_SEMAPHORE_H
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	/*
-	 * Note that any negative value of count is equivalent to 0,
-	 * but additionally indicates that some process(es) might be
-	 * sleeping on `wait'.
-	 */
-	atomic_t count;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name,count) \
-	{ ATOMIC_INIT(count), __WAIT_QUEUE_HEAD_INITIALIZER((name).wait) }
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	atomic_set(&sem->count, val);
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	if (atomic_dec_return(&sem->count) < 0)
-		__down(sem);
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-
-	might_sleep();
-	if (atomic_dec_return(&sem->count) < 0)
-		ret = __down_interruptible(sem);
-	return ret;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	int old_val, new_val;
-
-	/*
-	 * This inline assembly atomically implements the equivalent
-	 * to the following C code:
-	 *   old_val = sem->count.counter;
-	 *   if ((new_val = old_val) > 0)
-	 *       sem->count.counter = --new_val;
-	 * In the ppc code this is called atomic_dec_if_positive.
-	 */
-	asm volatile(
-		"	l	%0,0(%3)\n"
-		"0:	ltr	%1,%0\n"
-		"	jle	1f\n"
-		"	ahi	%1,-1\n"
-		"	cs	%0,%1,0(%3)\n"
-		"	jl	0b\n"
-		"1:"
-		: "=&d" (old_val), "=&d" (new_val), "=m" (sem->count.counter)
-		: "a" (&sem->count.counter), "m" (sem->count.counter)
-		: "cc", "memory");
-	return old_val <= 0;
-}
-
-static inline void up(struct semaphore * sem)
-{
-	if (atomic_inc_return(&sem->count) <= 0)
-		__up(sem);
-}
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-sh/semaphore-helper.h b/include/asm-sh/semaphore-helper.h
deleted file mode 100644
index bd8230c369ca..000000000000
--- a/include/asm-sh/semaphore-helper.h
+++ /dev/null
@@ -1,89 +0,0 @@
-#ifndef __ASM_SH_SEMAPHORE_HELPER_H
-#define __ASM_SH_SEMAPHORE_HELPER_H
-
-/*
- * SMP- and interrupt-safe semaphores helper functions.
- *
- * (C) Copyright 1996 Linus Torvalds
- * (C) Copyright 1999 Andrea Arcangeli
- */
-
-/*
- * These two _must_ execute atomically wrt each other.
- *
- * This is trivially done with load_locked/store_cond,
- * which we have.  Let the rest of the losers suck eggs.
- */
-static __inline__ void wake_one_more(struct semaphore * sem)
-{
-	atomic_inc((atomic_t *)&sem->sleepers);
-}
-
-static __inline__ int waking_non_zero(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_interruptible:
- *	1	got the lock
- *	0	go to sleep
- *	-EINTR	interrupted
- *
- * We must undo the sem->count down_interruptible() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static __inline__ int waking_non_zero_interruptible(struct semaphore *sem,
-						struct task_struct *tsk)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers > 0) {
-		sem->sleepers--;
-		ret = 1;
-	} else if (signal_pending(tsk)) {
-		atomic_inc(&sem->count);
-		ret = -EINTR;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-/*
- * waking_non_zero_trylock:
- *	1	failed to lock
- *	0	got the lock
- *
- * We must undo the sem->count down_trylock() increment while we are
- * protected by the spinlock in order to make atomic this atomic_inc() with the
- * atomic_read() in wake_one_more(), otherwise we can race. -arca
- */
-static __inline__ int waking_non_zero_trylock(struct semaphore *sem)
-{
-	unsigned long flags;
-	int ret = 1;
-
-	spin_lock_irqsave(&semaphore_wake_lock, flags);
-	if (sem->sleepers <= 0)
-		atomic_inc(&sem->count);
-	else {
-		sem->sleepers--;
-		ret = 0;
-	}
-	spin_unlock_irqrestore(&semaphore_wake_lock, flags);
-	return ret;
-}
-
-#endif /* __ASM_SH_SEMAPHORE_HELPER_H */
diff --git a/include/asm-sh/semaphore.h b/include/asm-sh/semaphore.h
index 9e5a37c4dce2..d9b2034ed1d2 100644
--- a/include/asm-sh/semaphore.h
+++ b/include/asm-sh/semaphore.h
@@ -1,115 +1 @@
-#ifndef __ASM_SH_SEMAPHORE_H
-#define __ASM_SH_SEMAPHORE_H
-
-#include <linux/linkage.h>
-
-#ifdef __KERNEL__
-/*
- * SMP- and interrupt-safe semaphores.
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * SuperH verison by Niibe Yutaka
- *  (Currently no asm implementation but generic C code...)
- */
-
-#include <linux/spinlock.h>
-#include <linux/rwsem.h>
-#include <linux/wait.h>
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-/*
- *	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
- *
- * i'd rather use the more flexible initialization above, but sadly
- * GCC 2.7.2.3 emits a bogus warning. EGCS doesn't. Oh well.
- */
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-#if 0
-asmlinkage void __down_failed(void /* special register calling convention */);
-asmlinkage int  __down_failed_interruptible(void  /* params in registers */);
-asmlinkage int  __down_failed_trylock(void  /* params in registers */);
-asmlinkage void __up_wakeup(void /* special register calling convention */);
-#endif
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-extern spinlock_t semaphore_wake_lock;
-
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	if (atomic_dec_return(&sem->count) < 0)
-		__down(sem);
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-
-	might_sleep();
-	if (atomic_dec_return(&sem->count) < 0)
-		ret = __down_interruptible(sem);
-	return ret;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	int ret = 0;
-
-	if (atomic_dec_return(&sem->count) < 0)
-		ret = __down_trylock(sem);
-	return ret;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- */
-static inline void up(struct semaphore * sem)
-{
-	if (atomic_inc_return(&sem->count) <= 0)
-		__up(sem);
-}
-
-#endif
-#endif /* __ASM_SH_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/asm-sparc/semaphore.h b/include/asm-sparc/semaphore.h
index 8018f9f4d497..d9b2034ed1d2 100644
--- a/include/asm-sparc/semaphore.h
+++ b/include/asm-sparc/semaphore.h
@@ -1,192 +1 @@
-#ifndef _SPARC_SEMAPHORE_H
-#define _SPARC_SEMAPHORE_H
-
-/* Dinky, good for nothing, just barely irq safe, Sparc semaphores. */
-
-#ifdef __KERNEL__
-
-#include <asm/atomic.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	atomic24_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC24_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	atomic24_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void __down(struct semaphore * sem);
-extern int __down_interruptible(struct semaphore * sem);
-extern int __down_trylock(struct semaphore * sem);
-extern void __up(struct semaphore * sem);
-
-static inline void down(struct semaphore * sem)
-{
-	register volatile int *ptr asm("g1");
-	register int increment asm("g2");
-
-	might_sleep();
-
-	ptr = &(sem->count.counter);
-	increment = 1;
-
-	__asm__ __volatile__(
-	"mov	%%o7, %%g4\n\t"
-	"call	___atomic24_sub\n\t"
-	" add	%%o7, 8, %%o7\n\t"
-	"tst	%%g2\n\t"
-	"bl	2f\n\t"
-	" nop\n"
-	"1:\n\t"
-	".subsection 2\n"
-	"2:\n\t"
-	"save	%%sp, -64, %%sp\n\t"
-	"mov	%%g1, %%l1\n\t"
-	"mov	%%g5, %%l5\n\t"
-	"call	%3\n\t"
-	" mov	%%g1, %%o0\n\t"
-	"mov	%%l1, %%g1\n\t"
-	"ba	1b\n\t"
-	" restore %%l5, %%g0, %%g5\n\t"
-	".previous\n"
-	: "=&r" (increment)
-	: "0" (increment), "r" (ptr), "i" (__down)
-	: "g3", "g4", "g7", "memory", "cc");
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	register volatile int *ptr asm("g1");
-	register int increment asm("g2");
-
-	might_sleep();
-
-	ptr = &(sem->count.counter);
-	increment = 1;
-
-	__asm__ __volatile__(
-	"mov	%%o7, %%g4\n\t"
-	"call	___atomic24_sub\n\t"
-	" add	%%o7, 8, %%o7\n\t"
-	"tst	%%g2\n\t"
-	"bl	2f\n\t"
-	" clr	%%g2\n"
-	"1:\n\t"
-	".subsection 2\n"
-	"2:\n\t"
-	"save	%%sp, -64, %%sp\n\t"
-	"mov	%%g1, %%l1\n\t"
-	"mov	%%g5, %%l5\n\t"
-	"call	%3\n\t"
-	" mov	%%g1, %%o0\n\t"
-	"mov	%%l1, %%g1\n\t"
-	"mov	%%l5, %%g5\n\t"
-	"ba	1b\n\t"
-	" restore %%o0, %%g0, %%g2\n\t"
-	".previous\n"
-	: "=&r" (increment)
-	: "0" (increment), "r" (ptr), "i" (__down_interruptible)
-	: "g3", "g4", "g7", "memory", "cc");
-
-	return increment;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	register volatile int *ptr asm("g1");
-	register int increment asm("g2");
-
-	ptr = &(sem->count.counter);
-	increment = 1;
-
-	__asm__ __volatile__(
-	"mov	%%o7, %%g4\n\t"
-	"call	___atomic24_sub\n\t"
-	" add	%%o7, 8, %%o7\n\t"
-	"tst	%%g2\n\t"
-	"bl	2f\n\t"
-	" clr	%%g2\n"
-	"1:\n\t"
-	".subsection 2\n"
-	"2:\n\t"
-	"save	%%sp, -64, %%sp\n\t"
-	"mov	%%g1, %%l1\n\t"
-	"mov	%%g5, %%l5\n\t"
-	"call	%3\n\t"
-	" mov	%%g1, %%o0\n\t"
-	"mov	%%l1, %%g1\n\t"
-	"mov	%%l5, %%g5\n\t"
-	"ba	1b\n\t"
-	" restore %%o0, %%g0, %%g2\n\t"
-	".previous\n"
-	: "=&r" (increment)
-	: "0" (increment), "r" (ptr), "i" (__down_trylock)
-	: "g3", "g4", "g7", "memory", "cc");
-
-	return increment;
-}
-
-static inline void up(struct semaphore * sem)
-{
-	register volatile int *ptr asm("g1");
-	register int increment asm("g2");
-
-	ptr = &(sem->count.counter);
-	increment = 1;
-
-	__asm__ __volatile__(
-	"mov	%%o7, %%g4\n\t"
-	"call	___atomic24_add\n\t"
-	" add	%%o7, 8, %%o7\n\t"
-	"tst	%%g2\n\t"
-	"ble	2f\n\t"
-	" nop\n"
-	"1:\n\t"
-	".subsection 2\n"
-	"2:\n\t"
-	"save	%%sp, -64, %%sp\n\t"
-	"mov	%%g1, %%l1\n\t"
-	"mov	%%g5, %%l5\n\t"
-	"call	%3\n\t"
-	" mov	%%g1, %%o0\n\t"
-	"mov	%%l1, %%g1\n\t"
-	"ba	1b\n\t"
-	" restore %%l5, %%g0, %%g5\n\t"
-	".previous\n"
-	: "=&r" (increment)
-	: "0" (increment), "r" (ptr), "i" (__up)
-	: "g3", "g4", "g7", "memory", "cc");
-}	
-
-#endif /* __KERNEL__ */
-
-#endif /* !(_SPARC_SEMAPHORE_H) */
+#include <linux/semaphore.h>
diff --git a/include/asm-sparc64/semaphore.h b/include/asm-sparc64/semaphore.h
index 7f7c0c4e024f..d9b2034ed1d2 100644
--- a/include/asm-sparc64/semaphore.h
+++ b/include/asm-sparc64/semaphore.h
@@ -1,53 +1 @@
-#ifndef _SPARC64_SEMAPHORE_H
-#define _SPARC64_SEMAPHORE_H
-
-/* These are actually reasonable on the V9.
- *
- * See asm-ppc/semaphore.h for implementation commentary,
- * only sparc64 specific issues are commented here.
- */
-#ifdef __KERNEL__
-
-#include <asm/atomic.h>
-#include <asm/system.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	atomic_t count;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, count) \
-	{ ATOMIC_INIT(count), \
-	  __WAIT_QUEUE_HEAD_INITIALIZER((name).wait) }
-
-#define __DECLARE_SEMAPHORE_GENERIC(name, count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name)		__DECLARE_SEMAPHORE_GENERIC(name, 1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	atomic_set(&sem->count, val);
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern void up(struct semaphore *sem);
-extern void down(struct semaphore *sem);
-extern int down_trylock(struct semaphore *sem);
-extern int down_interruptible(struct semaphore *sem);
-
-#endif /* __KERNEL__ */
-
-#endif /* !(_SPARC64_SEMAPHORE_H) */
+#include <linux/semaphore.h>
diff --git a/include/asm-um/semaphore.h b/include/asm-um/semaphore.h
index ff13c34de421..d9b2034ed1d2 100644
--- a/include/asm-um/semaphore.h
+++ b/include/asm-um/semaphore.h
@@ -1,6 +1 @@
-#ifndef __UM_SEMAPHORE_H
-#define __UM_SEMAPHORE_H
-
-#include "asm/arch/semaphore.h"
-
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-v850/semaphore.h b/include/asm-v850/semaphore.h
index 10ed0ccf37df..d9b2034ed1d2 100644
--- a/include/asm-v850/semaphore.h
+++ b/include/asm-v850/semaphore.h
@@ -1,84 +1 @@
-#ifndef __V850_SEMAPHORE_H__
-#define __V850_SEMAPHORE_H__
-
-#include <linux/linkage.h>
-#include <linux/spinlock.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-#include <asm/atomic.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name,count)				      \
-	{ ATOMIC_INIT (count), 0,					      \
-	  __WAIT_QUEUE_HEAD_INITIALIZER ((name).wait) }
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count)	\
-	struct semaphore name = __SEMAPHORE_INITIALIZER (name,count)
-
-#define DECLARE_MUTEX(name)		__DECLARE_SEMAPHORE_GENERIC (name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init (sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init (sem, 0);
-}
-
-/*
- * special register calling convention
- */
-asmlinkage void __down_failed (void);
-asmlinkage int  __down_interruptible_failed (void);
-asmlinkage int  __down_trylock_failed (void);
-asmlinkage void __up_wakeup (void);
-
-extern void __down (struct semaphore * sem);
-extern int  __down_interruptible (struct semaphore * sem);
-extern int  __down_trylock (struct semaphore * sem);
-extern void __up (struct semaphore * sem);
-
-static inline void down (struct semaphore * sem)
-{
-	might_sleep();
-	if (atomic_dec_return (&sem->count) < 0)
-		__down (sem);
-}
-
-static inline int down_interruptible (struct semaphore * sem)
-{
-	int ret = 0;
-	might_sleep();
-	if (atomic_dec_return (&sem->count) < 0)
-		ret = __down_interruptible (sem);
-	return ret;
-}
-
-static inline int down_trylock (struct semaphore *sem)
-{
-	int ret = 0;
-	if (atomic_dec_return (&sem->count) < 0)
-		ret = __down_trylock (sem);
-	return ret;
-}
-
-static inline void up (struct semaphore * sem)
-{
-	if (atomic_inc_return (&sem->count) <= 0)
-		__up (sem);
-}
-
-#endif /* __V850_SEMAPHORE_H__ */
+#include <linux/semaphore.h>
diff --git a/include/asm-x86/semaphore.h b/include/asm-x86/semaphore.h
index 572c0b67a6b0..d9b2034ed1d2 100644
--- a/include/asm-x86/semaphore.h
+++ b/include/asm-x86/semaphore.h
@@ -1,5 +1 @@
-#ifdef CONFIG_X86_32
-# include "semaphore_32.h"
-#else
-# include "semaphore_64.h"
-#endif
+#include <linux/semaphore.h>
diff --git a/include/asm-x86/semaphore_32.h b/include/asm-x86/semaphore_32.h
deleted file mode 100644
index ac96d3804d0c..000000000000
--- a/include/asm-x86/semaphore_32.h
+++ /dev/null
@@ -1,175 +0,0 @@
-#ifndef _I386_SEMAPHORE_H
-#define _I386_SEMAPHORE_H
-
-#include <linux/linkage.h>
-
-#ifdef __KERNEL__
-
-/*
- * SMP- and interrupt-safe semaphores..
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * Modified 1996-12-23 by Dave Grothe <dave@gcom.com> to fix bugs in
- *                     the original code and to make semaphore waits
- *                     interruptible so that processes waiting on
- *                     semaphores can be killed.
- * Modified 1999-02-14 by Andrea Arcangeli, split the sched.c helper
- *		       functions in asm/sempahore-helper.h while fixing a
- *		       potential and subtle race discovered by Ulrich Schmid
- *		       in down_interruptible(). Since I started to play here I
- *		       also implemented the `trylock' semaphore operation.
- *          1999-07-02 Artur Skawina <skawina@geocities.com>
- *                     Optimized "0(ecx)" -> "(ecx)" (the assembler does not
- *                     do this). Changed calling sequences from push/jmp to
- *                     traditional call/ret.
- * Modified 2001-01-01 Andreas Franck <afranck@gmx.de>
- *		       Some hacks to ensure compatibility with recent
- *		       GCC snapshots, to avoid stack corruption when compiling
- *		       with -fomit-frame-pointer. It's not sure if this will
- *		       be fixed in GCC, as our previous implementation was a
- *		       bit dubious.
- *
- * If you would like to see an analysis of this implementation, please
- * ftp to gcom.com and download the file
- * /pub/linux/src/semaphore/semaphore-2.0.24.tar.gz.
- *
- */
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-/*
- *	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
- *
- * i'd rather use the more flexible initialization above, but sadly
- * GCC 2.7.2.3 emits a bogus warning. EGCS doesn't. Oh well.
- */
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-extern asmregparm void __down_failed(atomic_t *count_ptr);
-extern asmregparm int  __down_failed_interruptible(atomic_t *count_ptr);
-extern asmregparm int  __down_failed_trylock(atomic_t *count_ptr);
-extern asmregparm void __up_wakeup(atomic_t *count_ptr);
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "__down_failed" is a special asm handler that calls the C
- * routine that actually waits. See arch/i386/kernel/semaphore.c
- */
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-	__asm__ __volatile__(
-		"# atomic down operation\n\t"
-		LOCK_PREFIX "decl %0\n\t"     /* --sem->count */
-		"jns 2f\n"
-		"\tlea %0,%%eax\n\t"
-		"call __down_failed\n"
-		"2:"
-		:"+m" (sem->count)
-		:
-		:"memory","ax");
-}
-
-/*
- * Interruptible try to acquire a semaphore.  If we obtained
- * it, return zero.  If we were interrupted, returns -EINTR
- */
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int result;
-
-	might_sleep();
-	__asm__ __volatile__(
-		"# atomic interruptible down operation\n\t"
-		"xorl %0,%0\n\t"
-		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"jns 2f\n\t"
-		"lea %1,%%eax\n\t"
-		"call __down_failed_interruptible\n"
-		"2:"
-		:"=&a" (result), "+m" (sem->count)
-		:
-		:"memory");
-	return result;
-}
-
-/*
- * Non-blockingly attempt to down() a semaphore.
- * Returns zero if we acquired it
- */
-static inline int down_trylock(struct semaphore * sem)
-{
-	int result;
-
-	__asm__ __volatile__(
-		"# atomic interruptible down operation\n\t"
-		"xorl %0,%0\n\t"
-		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"jns 2f\n\t"
-		"lea %1,%%eax\n\t"
-		"call __down_failed_trylock\n\t"
-		"2:\n"
-		:"=&a" (result), "+m" (sem->count)
-		:
-		:"memory");
-	return result;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- */
-static inline void up(struct semaphore * sem)
-{
-	__asm__ __volatile__(
-		"# atomic up operation\n\t"
-		LOCK_PREFIX "incl %0\n\t"     /* ++sem->count */
-		"jg 1f\n\t"
-		"lea %0,%%eax\n\t"
-		"call __up_wakeup\n"
-		"1:"
-		:"+m" (sem->count)
-		:
-		:"memory","ax");
-}
-
-#endif
-#endif
diff --git a/include/asm-x86/semaphore_64.h b/include/asm-x86/semaphore_64.h
deleted file mode 100644
index 79694306bf7d..000000000000
--- a/include/asm-x86/semaphore_64.h
+++ /dev/null
@@ -1,180 +0,0 @@
-#ifndef _X86_64_SEMAPHORE_H
-#define _X86_64_SEMAPHORE_H
-
-#include <linux/linkage.h>
-
-#ifdef __KERNEL__
-
-/*
- * SMP- and interrupt-safe semaphores..
- *
- * (C) Copyright 1996 Linus Torvalds
- *
- * Modified 1996-12-23 by Dave Grothe <dave@gcom.com> to fix bugs in
- *                     the original code and to make semaphore waits
- *                     interruptible so that processes waiting on
- *                     semaphores can be killed.
- * Modified 1999-02-14 by Andrea Arcangeli, split the sched.c helper
- *		       functions in asm/sempahore-helper.h while fixing a
- *		       potential and subtle race discovered by Ulrich Schmid
- *		       in down_interruptible(). Since I started to play here I
- *		       also implemented the `trylock' semaphore operation.
- *          1999-07-02 Artur Skawina <skawina@geocities.com>
- *                     Optimized "0(ecx)" -> "(ecx)" (the assembler does not
- *                     do this). Changed calling sequences from push/jmp to
- *                     traditional call/ret.
- * Modified 2001-01-01 Andreas Franck <afranck@gmx.de>
- *		       Some hacks to ensure compatibility with recent
- *		       GCC snapshots, to avoid stack corruption when compiling
- *		       with -fomit-frame-pointer. It's not sure if this will
- *		       be fixed in GCC, as our previous implementation was a
- *		       bit dubious.
- *
- * If you would like to see an analysis of this implementation, please
- * ftp to gcom.com and download the file
- * /pub/linux/src/semaphore/semaphore-2.0.24.tar.gz.
- *
- */
-
-#include <asm/system.h>
-#include <asm/atomic.h>
-#include <asm/rwlock.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-#include <linux/stringify.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name, n)				\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-/*
- *	*sem = (struct semaphore)__SEMAPHORE_INITIALIZER((*sem),val);
- *
- * i'd rather use the more flexible initialization above, but sadly
- * GCC 2.7.2.3 emits a bogus warning. EGCS doesn't. Oh well.
- */
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down_failed(void /* special register calling convention */);
-asmlinkage int  __down_failed_interruptible(void  /* params in registers */);
-asmlinkage int  __down_failed_trylock(void  /* params in registers */);
-asmlinkage void __up_wakeup(void /* special register calling convention */);
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-/*
- * This is ugly, but we want the default case to fall through.
- * "__down_failed" is a special asm handler that calls the C
- * routine that actually waits. See arch/x86_64/kernel/semaphore.c
- */
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-
-	__asm__ __volatile__(
-		"# atomic down operation\n\t"
-		LOCK_PREFIX "decl %0\n\t"     /* --sem->count */
-		"jns 1f\n\t"
-		"call __down_failed\n"
-		"1:"
-		:"=m" (sem->count)
-		:"D" (sem)
-		:"memory");
-}
-
-/*
- * Interruptible try to acquire a semaphore.  If we obtained
- * it, return zero.  If we were interrupted, returns -EINTR
- */
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int result;
-
-	might_sleep();
-
-	__asm__ __volatile__(
-		"# atomic interruptible down operation\n\t"
-		"xorl %0,%0\n\t"
-		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"jns 2f\n\t"
-		"call __down_failed_interruptible\n"
-		"2:\n"
-		:"=&a" (result), "=m" (sem->count)
-		:"D" (sem)
-		:"memory");
-	return result;
-}
-
-/*
- * Non-blockingly attempt to down() a semaphore.
- * Returns zero if we acquired it
- */
-static inline int down_trylock(struct semaphore * sem)
-{
-	int result;
-
-	__asm__ __volatile__(
-		"# atomic interruptible down operation\n\t"
-		"xorl %0,%0\n\t"
-		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"jns 2f\n\t"
-		"call __down_failed_trylock\n\t"
-		"2:\n"
-		:"=&a" (result), "=m" (sem->count)
-		:"D" (sem)
-		:"memory","cc");
-	return result;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
- */
-static inline void up(struct semaphore * sem)
-{
-	__asm__ __volatile__(
-		"# atomic up operation\n\t"
-		LOCK_PREFIX "incl %0\n\t"     /* ++sem->count */
-		"jg 1f\n\t"
-		"call __up_wakeup\n"
-		"1:"
-		:"=m" (sem->count)
-		:"D" (sem)
-		:"memory");
-}
-#endif /* __KERNEL__ */
-#endif
diff --git a/include/asm-xtensa/semaphore.h b/include/asm-xtensa/semaphore.h
index 3e04167cd9dc..d9b2034ed1d2 100644
--- a/include/asm-xtensa/semaphore.h
+++ b/include/asm-xtensa/semaphore.h
@@ -1,99 +1 @@
-/*
- * linux/include/asm-xtensa/semaphore.h
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2001 - 2005 Tensilica Inc.
- */
-
-#ifndef _XTENSA_SEMAPHORE_H
-#define _XTENSA_SEMAPHORE_H
-
-#include <asm/atomic.h>
-#include <asm/system.h>
-#include <linux/wait.h>
-#include <linux/rwsem.h>
-
-struct semaphore {
-	atomic_t count;
-	int sleepers;
-	wait_queue_head_t wait;
-};
-
-#define __SEMAPHORE_INITIALIZER(name,n)					\
-{									\
-	.count		= ATOMIC_INIT(n),				\
-	.sleepers	= 0,						\
-	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
-}
-
-#define __DECLARE_SEMAPHORE_GENERIC(name,count) 			\
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
-
-#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
-
-static inline void sema_init (struct semaphore *sem, int val)
-{
-	atomic_set(&sem->count, val);
-	sem->sleepers = 0;
-	init_waitqueue_head(&sem->wait);
-}
-
-static inline void init_MUTEX (struct semaphore *sem)
-{
-	sema_init(sem, 1);
-}
-
-static inline void init_MUTEX_LOCKED (struct semaphore *sem)
-{
-	sema_init(sem, 0);
-}
-
-asmlinkage void __down(struct semaphore * sem);
-asmlinkage int  __down_interruptible(struct semaphore * sem);
-asmlinkage int  __down_trylock(struct semaphore * sem);
-asmlinkage void __up(struct semaphore * sem);
-
-extern spinlock_t semaphore_wake_lock;
-
-static inline void down(struct semaphore * sem)
-{
-	might_sleep();
-
-	if (atomic_sub_return(1, &sem->count) < 0)
-		__down(sem);
-}
-
-static inline int down_interruptible(struct semaphore * sem)
-{
-	int ret = 0;
-
-	might_sleep();
-
-	if (atomic_sub_return(1, &sem->count) < 0)
-		ret = __down_interruptible(sem);
-	return ret;
-}
-
-static inline int down_trylock(struct semaphore * sem)
-{
-	int ret = 0;
-
-	if (atomic_sub_return(1, &sem->count) < 0)
-		ret = __down_trylock(sem);
-	return ret;
-}
-
-/*
- * Note! This is subtle. We jump to wake people up only if
- * the semaphore was negative (== somebody was waiting on it).
- */
-static inline void up(struct semaphore * sem)
-{
-	if (atomic_add_return(1, &sem->count) <= 0)
-		__up(sem);
-}
-
-#endif /* _XTENSA_SEMAPHORE_H */
+#include <linux/semaphore.h>
diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
new file mode 100644
index 000000000000..b3c691b089b2
--- /dev/null
+++ b/include/linux/semaphore.h
@@ -0,0 +1,77 @@
+/*
+ * Copyright (c) 2008 Intel Corporation
+ * Author: Matthew Wilcox <willy@linux.intel.com>
+ *
+ * Distributed under the terms of the GNU GPL, version 2
+ *
+ * Counting semaphores allow up to <n> tasks to acquire the semaphore
+ * simultaneously.
+ */
+#ifndef __LINUX_SEMAPHORE_H
+#define __LINUX_SEMAPHORE_H
+
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+/*
+ * The spinlock controls access to the other members of the semaphore.
+ * 'count' is decremented by every task which calls down*() and incremented
+ * by every call to up().  Thus, if it is positive, it indicates how many
+ * more tasks may acquire the lock.  If it is negative, it indicates how
+ * many tasks are waiting for the lock.  Tasks waiting for the lock are
+ * kept on the wait_list.
+ */
+struct semaphore {
+	spinlock_t		lock;
+	int			count;
+	struct list_head	wait_list;
+};
+
+#define __SEMAPHORE_INITIALIZER(name, n)				\
+{									\
+	.lock		= __SPIN_LOCK_UNLOCKED((name).lock),		\
+	.count		= n,						\
+	.wait_list	= LIST_HEAD_INIT((name).wait_list),		\
+}
+
+#define __DECLARE_SEMAPHORE_GENERIC(name, count) \
+	struct semaphore name = __SEMAPHORE_INITIALIZER(name, count)
+
+#define DECLARE_MUTEX(name)	__DECLARE_SEMAPHORE_GENERIC(name, 1)
+
+static inline void sema_init(struct semaphore *sem, int val)
+{
+	static struct lock_class_key __key;
+	*sem = (struct semaphore) __SEMAPHORE_INITIALIZER(*sem, val);
+	lockdep_init_map(&sem->lock.dep_map, "semaphore->lock", &__key, 0);
+}
+
+#define init_MUTEX(sem)		sema_init(sem, 1)
+#define init_MUTEX_LOCKED(sem)	sema_init(sem, 0)
+
+/*
+ * Attempt to acquire the semaphore.  If another task is already holding the
+ * semaphore, sleep until the semaphore is released.
+ */
+extern void down(struct semaphore *sem);
+
+/*
+ * As down(), except the sleep may be interrupted by a signal.  If it is,
+ * this function will return -EINTR.
+ */
+extern int __must_check down_interruptible(struct semaphore *sem);
+
+/*
+ * As down(), except this function will not sleep.  It will return 0 if it
+ * acquired the semaphore and 1 if the semaphore was contended.  This
+ * function may be called from any context, including interrupt and softirq.
+ */
+extern int __must_check down_trylock(struct semaphore *sem);
+
+/*
+ * Release the semaphore.  Unlike mutexes, up() may be called from any
+ * context and even by tasks which have never called down().
+ */
+extern void up(struct semaphore *sem);
+
+#endif /* __LINUX_SEMAPHORE_H */
diff --git a/kernel/Makefile b/kernel/Makefile
index 6c584c55a6e9..f45c69e69688 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -8,7 +8,7 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
 	    signal.o sys.o kmod.o workqueue.o pid.o \
 	    rcupdate.o extable.o params.o posix-timers.o \
 	    kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
-	    hrtimer.o rwsem.o nsproxy.o srcu.o \
+	    hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \
 	    notifier.o ksysfs.o pm_qos_params.o
 
 obj-$(CONFIG_SYSCTL) += sysctl_check.o
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
new file mode 100644
index 000000000000..d5a72702f261
--- /dev/null
+++ b/kernel/semaphore.c
@@ -0,0 +1,187 @@
+/*
+ * Copyright (c) 2008 Intel Corporation
+ * Author: Matthew Wilcox <willy@linux.intel.com>
+ *
+ * Distributed under the terms of the GNU GPL, version 2
+ */
+
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/semaphore.h>
+#include <linux/spinlock.h>
+
+/*
+ * Some notes on the implementation:
+ *
+ * down_trylock() and up() can be called from interrupt context.
+ * So we have to disable interrupts when taking the lock.
+ *
+ * The ->count variable, if positive, defines how many more tasks can
+ * acquire the semaphore.  If negative, it represents how many tasks are
+ * waiting on the semaphore (*).  If zero, no tasks are waiting, and no more
+ * tasks can acquire the semaphore.
+ *
+ * (*) Except for the window between one task calling up() and the task
+ * sleeping in a __down_common() waking up.  In order to avoid a third task
+ * coming in and stealing the second task's wakeup, we leave the ->count
+ * negative.  If we have a more complex situation, the ->count may become
+ * zero or negative (eg a semaphore with count = 2, three tasks attempt to
+ * acquire it, one sleeps, two finish and call up(), the second task to call
+ * up() notices that the list is empty and just increments count).
+ */
+
+static noinline void __down(struct semaphore *sem);
+static noinline int __down_interruptible(struct semaphore *sem);
+static noinline void __up(struct semaphore *sem);
+
+void down(struct semaphore *sem)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sem->lock, flags);
+	if (unlikely(sem->count-- <= 0))
+		__down(sem);
+	spin_unlock_irqrestore(&sem->lock, flags);
+}
+EXPORT_SYMBOL(down);
+
+int down_interruptible(struct semaphore *sem)
+{
+	unsigned long flags;
+	int result = 0;
+
+	spin_lock_irqsave(&sem->lock, flags);
+	if (unlikely(sem->count-- <= 0))
+		result = __down_interruptible(sem);
+	spin_unlock_irqrestore(&sem->lock, flags);
+
+	return result;
+}
+EXPORT_SYMBOL(down_interruptible);
+
+/**
+ * down_trylock - try to acquire the semaphore, without waiting
+ * @sem: the semaphore to be acquired
+ *
+ * Try to acquire the semaphore atomically.  Returns 0 if the mutex has
+ * been acquired successfully and 1 if it is contended.
+ *
+ * NOTE: This return value is inverted from both spin_trylock and
+ * mutex_trylock!  Be careful about this when converting code.
+ *
+ * Unlike mutex_trylock, this function can be used from interrupt context,
+ * and the semaphore can be released by any task or interrupt.
+ */
+int down_trylock(struct semaphore *sem)
+{
+	unsigned long flags;
+	int count;
+
+	spin_lock_irqsave(&sem->lock, flags);
+	count = sem->count - 1;
+	if (likely(count >= 0))
+		sem->count = count;
+	spin_unlock_irqrestore(&sem->lock, flags);
+
+	return (count < 0);
+}
+EXPORT_SYMBOL(down_trylock);
+
+void up(struct semaphore *sem)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sem->lock, flags);
+	if (likely(sem->count >= 0))
+		sem->count++;
+	else
+		__up(sem);
+	spin_unlock_irqrestore(&sem->lock, flags);
+}
+EXPORT_SYMBOL(up);
+
+/* Functions for the contended case */
+
+struct semaphore_waiter {
+	struct list_head list;
+	struct task_struct *task;
+	int up;
+};
+
+/*
+ * Wake up a process waiting on a semaphore.  We need to call this from both
+ * __up and __down_common as it's possible to race a task into the semaphore
+ * if it comes in at just the right time between two tasks calling up() and
+ * a third task waking up.  This function assumes the wait_list is already
+ * checked for being non-empty.
+ */
+static noinline void __sched __up_down_common(struct semaphore *sem)
+{
+	struct semaphore_waiter *waiter = list_first_entry(&sem->wait_list,
+						struct semaphore_waiter, list);
+	list_del(&waiter->list);
+	waiter->up = 1;
+	wake_up_process(waiter->task);
+}
+
+/*
+ * Because this function is inlined, the 'state' parameter will be constant,
+ * and thus optimised away by the compiler.
+ */
+static inline int __sched __down_common(struct semaphore *sem, long state)
+{
+	int result = 0;
+	struct task_struct *task = current;
+	struct semaphore_waiter waiter;
+
+	list_add_tail(&waiter.list, &sem->wait_list);
+	waiter.task = task;
+	waiter.up = 0;
+
+	for (;;) {
+		if (state == TASK_INTERRUPTIBLE && signal_pending(task))
+			goto interrupted;
+		__set_task_state(task, state);
+		spin_unlock_irq(&sem->lock);
+		schedule();
+		spin_lock_irq(&sem->lock);
+		if (waiter.up)
+			goto woken;
+	}
+
+ interrupted:
+	list_del(&waiter.list);
+	result = -EINTR;
+ woken:
+	/*
+	 * Account for the process which woke us up.  For the case where
+	 * we're interrupted, we need to increment the count on our own
+	 * behalf.  I don't believe we can hit the case where the
+	 * sem->count hits zero, *and* there's a second task sleeping,
+	 * but it doesn't hurt, that's not a commonly exercised path and
+	 * it's not a performance path either.
+	 */
+	if (unlikely((++sem->count >= 0) && !list_empty(&sem->wait_list)))
+		__up_down_common(sem);
+	return result;
+}
+
+static noinline void __sched __down(struct semaphore *sem)
+{
+	__down_common(sem, TASK_UNINTERRUPTIBLE);
+}
+
+static noinline int __sched __down_interruptible(struct semaphore *sem)
+{
+	return __down_common(sem, TASK_INTERRUPTIBLE);
+}
+
+static noinline void __sched __up(struct semaphore *sem)
+{
+	if (unlikely(list_empty(&sem->wait_list)))
+		sem->count++;
+	else
+		__up_down_common(sem);
+}
diff --git a/lib/Makefile b/lib/Makefile
index 23de261a4c83..28dba90d5020 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -29,7 +29,6 @@ obj-$(CONFIG_DEBUG_LOCKING_API_SELFTESTS) += locking-selftest.o
 obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock_debug.o
 lib-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o
 lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o
-lib-$(CONFIG_SEMAPHORE_SLEEPERS) += semaphore-sleepers.o
 lib-$(CONFIG_GENERIC_FIND_NEXT_BIT) += find_next_bit.o
 obj-$(CONFIG_GENERIC_HWEIGHT) += hweight.o
 obj-$(CONFIG_LOCK_KERNEL) += kernel_lock.o
diff --git a/lib/semaphore-sleepers.c b/lib/semaphore-sleepers.c
deleted file mode 100644
index 0198782cdacb..000000000000
--- a/lib/semaphore-sleepers.c
+++ /dev/null
@@ -1,176 +0,0 @@
-/*
- * i386 and x86-64 semaphore implementation.
- *
- * (C) Copyright 1999 Linus Torvalds
- *
- * Portions Copyright 1999 Red Hat, Inc.
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- * rw semaphores implemented November 1999 by Benjamin LaHaise <bcrl@kvack.org>
- */
-#include <linux/sched.h>
-#include <linux/err.h>
-#include <linux/init.h>
-#include <asm/semaphore.h>
-
-/*
- * Semaphores are implemented using a two-way counter:
- * The "count" variable is decremented for each process
- * that tries to acquire the semaphore, while the "sleeping"
- * variable is a count of such acquires.
- *
- * Notably, the inline "up()" and "down()" functions can
- * efficiently test if they need to do any extra work (up
- * needs to do something only if count was negative before
- * the increment operation.
- *
- * "sleeping" and the contention routine ordering is protected
- * by the spinlock in the semaphore's waitqueue head.
- *
- * Note that these functions are only called when there is
- * contention on the lock, and as such all this is the
- * "non-critical" part of the whole semaphore business. The
- * critical part is the inline stuff in <asm/semaphore.h>
- * where we want to avoid any extra jumps and calls.
- */
-
-/*
- * Logic:
- *  - only on a boundary condition do we need to care. When we go
- *    from a negative count to a non-negative, we wake people up.
- *  - when we go from a non-negative count to a negative do we
- *    (a) synchronize with the "sleeper" count and (b) make sure
- *    that we're on the wakeup list before we synchronize so that
- *    we cannot lose wakeup events.
- */
-
-void __up(struct semaphore *sem)
-{
-	wake_up(&sem->wait);
-}
-
-void __sched __down(struct semaphore *sem)
-{
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	unsigned long flags;
-
-	tsk->state = TASK_UNINTERRUPTIBLE;
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock in
-		 * the wait_queue_head.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-		schedule();
-
-		spin_lock_irqsave(&sem->wait.lock, flags);
-		tsk->state = TASK_UNINTERRUPTIBLE;
-	}
-	remove_wait_queue_locked(&sem->wait, &wait);
-	wake_up_locked(&sem->wait);
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-	tsk->state = TASK_RUNNING;
-}
-
-int __sched __down_interruptible(struct semaphore *sem)
-{
-	int retval = 0;
-	struct task_struct *tsk = current;
-	DECLARE_WAITQUEUE(wait, tsk);
-	unsigned long flags;
-
-	tsk->state = TASK_INTERRUPTIBLE;
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	add_wait_queue_exclusive_locked(&sem->wait, &wait);
-
-	sem->sleepers++;
-	for (;;) {
-		int sleepers = sem->sleepers;
-
-		/*
-		 * With signals pending, this turns into
-		 * the trylock failure case - we won't be
-		 * sleeping, and we* can't get the lock as
-		 * it has contention. Just correct the count
-		 * and exit.
-		 */
-		if (signal_pending(current)) {
-			retval = -EINTR;
-			sem->sleepers = 0;
-			atomic_add(sleepers, &sem->count);
-			break;
-		}
-
-		/*
-		 * Add "everybody else" into it. They aren't
-		 * playing, because we own the spinlock in
-		 * wait_queue_head. The "-1" is because we're
-		 * still hoping to get the semaphore.
-		 */
-		if (!atomic_add_negative(sleepers - 1, &sem->count)) {
-			sem->sleepers = 0;
-			break;
-		}
-		sem->sleepers = 1;	/* us - see -1 above */
-		spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-		schedule();
-
-		spin_lock_irqsave(&sem->wait.lock, flags);
-		tsk->state = TASK_INTERRUPTIBLE;
-	}
-	remove_wait_queue_locked(&sem->wait, &wait);
-	wake_up_locked(&sem->wait);
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-
-	tsk->state = TASK_RUNNING;
-	return retval;
-}
-
-/*
- * Trylock failed - make sure we correct for
- * having decremented the count.
- *
- * We could have done the trylock with a
- * single "cmpxchg" without failure cases,
- * but then it wouldn't work on a 386.
- */
-int __down_trylock(struct semaphore *sem)
-{
-	int sleepers;
-	unsigned long flags;
-
-	spin_lock_irqsave(&sem->wait.lock, flags);
-	sleepers = sem->sleepers + 1;
-	sem->sleepers = 0;
-
-	/*
-	 * Add "everybody else" and us into it. They aren't
-	 * playing, because we own the spinlock in the
-	 * wait_queue_head.
-	 */
-	if (!atomic_add_negative(sleepers, &sem->count)) {
-		wake_up_locked(&sem->wait);
-	}
-
-	spin_unlock_irqrestore(&sem->wait.lock, flags);
-	return 1;
-}
-- 
cgit v1.2.3


From f06d96865861c3dd01520f47e2e61c899db1631f Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Fri, 14 Mar 2008 13:19:33 -0400
Subject: Introduce down_killable()

down_killable() is the functional counterpart of mutex_lock_killable.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 include/linux/semaphore.h |  6 ++++++
 kernel/semaphore.c        | 22 ++++++++++++++++++++++
 2 files changed, 28 insertions(+)

(limited to 'include')

diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
index b3c691b089b2..88f2a28cc0f1 100644
--- a/include/linux/semaphore.h
+++ b/include/linux/semaphore.h
@@ -61,6 +61,12 @@ extern void down(struct semaphore *sem);
  */
 extern int __must_check down_interruptible(struct semaphore *sem);
 
+/*
+ * As down_interruptible(), except the sleep may only be interrupted by
+ * signals which are fatal to this process.
+ */
+extern int __must_check down_killable(struct semaphore *sem);
+
 /*
  * As down(), except this function will not sleep.  It will return 0 if it
  * acquired the semaphore and 1 if the semaphore was contended.  This
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index d5a72702f261..2da2aed950f3 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -34,6 +34,7 @@
 
 static noinline void __down(struct semaphore *sem);
 static noinline int __down_interruptible(struct semaphore *sem);
+static noinline int __down_killable(struct semaphore *sem);
 static noinline void __up(struct semaphore *sem);
 
 void down(struct semaphore *sem)
@@ -61,6 +62,20 @@ int down_interruptible(struct semaphore *sem)
 }
 EXPORT_SYMBOL(down_interruptible);
 
+int down_killable(struct semaphore *sem)
+{
+	unsigned long flags;
+	int result = 0;
+
+	spin_lock_irqsave(&sem->lock, flags);
+	if (unlikely(sem->count-- <= 0))
+		result = __down_killable(sem);
+	spin_unlock_irqrestore(&sem->lock, flags);
+
+	return result;
+}
+EXPORT_SYMBOL(down_killable);
+
 /**
  * down_trylock - try to acquire the semaphore, without waiting
  * @sem: the semaphore to be acquired
@@ -143,6 +158,8 @@ static inline int __sched __down_common(struct semaphore *sem, long state)
 	for (;;) {
 		if (state == TASK_INTERRUPTIBLE && signal_pending(task))
 			goto interrupted;
+		if (state == TASK_KILLABLE && fatal_signal_pending(task))
+			goto interrupted;
 		__set_task_state(task, state);
 		spin_unlock_irq(&sem->lock);
 		schedule();
@@ -178,6 +195,11 @@ static noinline int __sched __down_interruptible(struct semaphore *sem)
 	return __down_common(sem, TASK_INTERRUPTIBLE);
 }
 
+static noinline int __sched __down_killable(struct semaphore *sem)
+{
+	return __down_common(sem, TASK_KILLABLE);
+}
+
 static noinline void __sched __up(struct semaphore *sem)
 {
 	if (unlikely(list_empty(&sem->wait_list)))
-- 
cgit v1.2.3


From f1241c87a16c4fe9f4f51d6ed3589f031c505e8d Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Fri, 14 Mar 2008 13:43:13 -0400
Subject: Add down_timeout and change ACPI to use it

ACPI currently emulates a timeout for semaphores with calls to
down_trylock and sleep.  This produces horrible behaviour in terms of
fairness and excessive wakeups.  Now that we have a unified semaphore
implementation, adding a real down_trylock is almost trivial.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 drivers/acpi/osl.c        | 89 +++++++++++------------------------------------
 include/linux/semaphore.h |  6 ++++
 kernel/semaphore.c        | 42 ++++++++++++++++++----
 3 files changed, 62 insertions(+), 75 deletions(-)

(limited to 'include')

diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index a697fb6cf050..a498a6cc68fe 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -4,6 +4,8 @@
  *  Copyright (C) 2000       Andrew Henroid
  *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
  *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ *  Copyright (c) 2008 Intel Corporation
+ *   Author: Matthew Wilcox <willy@linux.intel.com>
  *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  *
@@ -37,15 +39,18 @@
 #include <linux/workqueue.h>
 #include <linux/nmi.h>
 #include <linux/acpi.h>
-#include <acpi/acpi.h>
-#include <asm/io.h>
-#include <acpi/acpi_bus.h>
-#include <acpi/processor.h>
-#include <asm/uaccess.h>
-
 #include <linux/efi.h>
 #include <linux/ioport.h>
 #include <linux/list.h>
+#include <linux/jiffies.h>
+#include <linux/semaphore.h>
+
+#include <asm/io.h>
+#include <asm/uaccess.h>
+
+#include <acpi/acpi.h>
+#include <acpi/acpi_bus.h>
+#include <acpi/processor.h>
 
 #define _COMPONENT		ACPI_OS_SERVICES
 ACPI_MODULE_NAME("osl");
@@ -764,7 +769,6 @@ acpi_os_create_semaphore(u32 max_units, u32 initial_units, acpi_handle * handle)
 {
 	struct semaphore *sem = NULL;
 
-
 	sem = acpi_os_allocate(sizeof(struct semaphore));
 	if (!sem)
 		return AE_NO_MEMORY;
@@ -791,12 +795,12 @@ acpi_status acpi_os_delete_semaphore(acpi_handle handle)
 {
 	struct semaphore *sem = (struct semaphore *)handle;
 
-
 	if (!sem)
 		return AE_BAD_PARAMETER;
 
 	ACPI_DEBUG_PRINT((ACPI_DB_MUTEX, "Deleting semaphore[%p].\n", handle));
 
+	BUG_ON(!list_empty(&sem->wait_list));
 	kfree(sem);
 	sem = NULL;
 
@@ -804,21 +808,15 @@ acpi_status acpi_os_delete_semaphore(acpi_handle handle)
 }
 
 /*
- * TODO: The kernel doesn't have a 'down_timeout' function -- had to
- * improvise.  The process is to sleep for one scheduler quantum
- * until the semaphore becomes available.  Downside is that this
- * may result in starvation for timeout-based waits when there's
- * lots of semaphore activity.
- *
  * TODO: Support for units > 1?
  */
 acpi_status acpi_os_wait_semaphore(acpi_handle handle, u32 units, u16 timeout)
 {
 	acpi_status status = AE_OK;
 	struct semaphore *sem = (struct semaphore *)handle;
+	long jiffies;
 	int ret = 0;
 
-
 	if (!sem || (units < 1))
 		return AE_BAD_PARAMETER;
 
@@ -828,58 +826,14 @@ acpi_status acpi_os_wait_semaphore(acpi_handle handle, u32 units, u16 timeout)
 	ACPI_DEBUG_PRINT((ACPI_DB_MUTEX, "Waiting for semaphore[%p|%d|%d]\n",
 			  handle, units, timeout));
 
-	/*
-	 * This can be called during resume with interrupts off.
-	 * Like boot-time, we should be single threaded and will
-	 * always get the lock if we try -- timeout or not.
-	 * If this doesn't succeed, then we will oops courtesy of
-	 * might_sleep() in down().
-	 */
-	if (!down_trylock(sem))
-		return AE_OK;
-
-	switch (timeout) {
-		/*
-		 * No Wait:
-		 * --------
-		 * A zero timeout value indicates that we shouldn't wait - just
-		 * acquire the semaphore if available otherwise return AE_TIME
-		 * (a.k.a. 'would block').
-		 */
-	case 0:
-		if (down_trylock(sem))
-			status = AE_TIME;
-		break;
-
-		/*
-		 * Wait Indefinitely:
-		 * ------------------
-		 */
-	case ACPI_WAIT_FOREVER:
-		down(sem);
-		break;
-
-		/*
-		 * Wait w/ Timeout:
-		 * ----------------
-		 */
-	default:
-		// TODO: A better timeout algorithm?
-		{
-			int i = 0;
-			static const int quantum_ms = 1000 / HZ;
-
-			ret = down_trylock(sem);
-			for (i = timeout; (i > 0 && ret != 0); i -= quantum_ms) {
-				schedule_timeout_interruptible(1);
-				ret = down_trylock(sem);
-			}
-
-			if (ret != 0)
-				status = AE_TIME;
-		}
-		break;
-	}
+	if (timeout == ACPI_WAIT_FOREVER)
+		jiffies = MAX_SCHEDULE_TIMEOUT;
+	else
+		jiffies = msecs_to_jiffies(timeout);
+	
+	ret = down_timeout(sem, jiffies);
+	if (ret)
+		status = AE_TIME;
 
 	if (ACPI_FAILURE(status)) {
 		ACPI_DEBUG_PRINT((ACPI_DB_MUTEX,
@@ -902,7 +856,6 @@ acpi_status acpi_os_signal_semaphore(acpi_handle handle, u32 units)
 {
 	struct semaphore *sem = (struct semaphore *)handle;
 
-
 	if (!sem || (units < 1))
 		return AE_BAD_PARAMETER;
 
diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
index 88f2a28cc0f1..a107aebd9148 100644
--- a/include/linux/semaphore.h
+++ b/include/linux/semaphore.h
@@ -74,6 +74,12 @@ extern int __must_check down_killable(struct semaphore *sem);
  */
 extern int __must_check down_trylock(struct semaphore *sem);
 
+/*
+ * As down(), except this function will return -ETIME if it fails to
+ * acquire the semaphore within the specified number of jiffies.
+ */
+extern int __must_check down_timeout(struct semaphore *sem, long jiffies);
+
 /*
  * Release the semaphore.  Unlike mutexes, up() may be called from any
  * context and even by tasks which have never called down().
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index 2da2aed950f3..5a12a8558982 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -35,6 +35,7 @@
 static noinline void __down(struct semaphore *sem);
 static noinline int __down_interruptible(struct semaphore *sem);
 static noinline int __down_killable(struct semaphore *sem);
+static noinline int __down_timeout(struct semaphore *sem, long jiffies);
 static noinline void __up(struct semaphore *sem);
 
 void down(struct semaphore *sem)
@@ -104,6 +105,20 @@ int down_trylock(struct semaphore *sem)
 }
 EXPORT_SYMBOL(down_trylock);
 
+int down_timeout(struct semaphore *sem, long jiffies)
+{
+	unsigned long flags;
+	int result = 0;
+
+	spin_lock_irqsave(&sem->lock, flags);
+	if (unlikely(sem->count-- <= 0))
+		result = __down_timeout(sem, jiffies);
+	spin_unlock_irqrestore(&sem->lock, flags);
+
+	return result;
+}
+EXPORT_SYMBOL(down_timeout);
+
 void up(struct semaphore *sem)
 {
 	unsigned long flags;
@@ -142,10 +157,12 @@ static noinline void __sched __up_down_common(struct semaphore *sem)
 }
 
 /*
- * Because this function is inlined, the 'state' parameter will be constant,
- * and thus optimised away by the compiler.
+ * Because this function is inlined, the 'state' parameter will be
+ * constant, and thus optimised away by the compiler.  Likewise the
+ * 'timeout' parameter for the cases without timeouts.
  */
-static inline int __sched __down_common(struct semaphore *sem, long state)
+static inline int __sched __down_common(struct semaphore *sem, long state,
+								long timeout)
 {
 	int result = 0;
 	struct task_struct *task = current;
@@ -160,14 +177,20 @@ static inline int __sched __down_common(struct semaphore *sem, long state)
 			goto interrupted;
 		if (state == TASK_KILLABLE && fatal_signal_pending(task))
 			goto interrupted;
+		if (timeout <= 0)
+			goto timed_out;
 		__set_task_state(task, state);
 		spin_unlock_irq(&sem->lock);
-		schedule();
+		timeout = schedule_timeout(timeout);
 		spin_lock_irq(&sem->lock);
 		if (waiter.up)
 			goto woken;
 	}
 
+ timed_out:
+	list_del(&waiter.list);
+	result = -ETIME;
+	goto woken;
  interrupted:
 	list_del(&waiter.list);
 	result = -EINTR;
@@ -187,17 +210,22 @@ static inline int __sched __down_common(struct semaphore *sem, long state)
 
 static noinline void __sched __down(struct semaphore *sem)
 {
-	__down_common(sem, TASK_UNINTERRUPTIBLE);
+	__down_common(sem, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
 }
 
 static noinline int __sched __down_interruptible(struct semaphore *sem)
 {
-	return __down_common(sem, TASK_INTERRUPTIBLE);
+	return __down_common(sem, TASK_INTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
 }
 
 static noinline int __sched __down_killable(struct semaphore *sem)
 {
-	return __down_common(sem, TASK_KILLABLE);
+	return __down_common(sem, TASK_KILLABLE, MAX_SCHEDULE_TIMEOUT);
+}
+
+static noinline int __sched __down_timeout(struct semaphore *sem, long jiffies)
+{
+	return __down_common(sem, TASK_UNINTERRUPTIBLE, jiffies);
 }
 
 static noinline void __sched __up(struct semaphore *sem)
-- 
cgit v1.2.3


From b17170b2fac96705db3188f093f89e8e838418e4 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Fri, 14 Mar 2008 14:35:22 -0400
Subject: Simplify semaphore implementation

By removing the negative values of 'count' and relying on the wait_list to
indicate whether we have any waiters, we can simplify the implementation
by removing the protection against an unlikely race condition.  Thanks to
David Howells for his suggestions.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 include/linux/semaphore.h |  9 ++----
 kernel/semaphore.c        | 78 +++++++++++++++--------------------------------
 2 files changed, 27 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
index a107aebd9148..a7125daaff90 100644
--- a/include/linux/semaphore.h
+++ b/include/linux/semaphore.h
@@ -15,15 +15,12 @@
 
 /*
  * The spinlock controls access to the other members of the semaphore.
- * 'count' is decremented by every task which calls down*() and incremented
- * by every call to up().  Thus, if it is positive, it indicates how many
- * more tasks may acquire the lock.  If it is negative, it indicates how
- * many tasks are waiting for the lock.  Tasks waiting for the lock are
- * kept on the wait_list.
+ * 'count' represents how many more tasks can acquire this semaphore.
+ * Tasks waiting for the lock are kept on the wait_list.
  */
 struct semaphore {
 	spinlock_t		lock;
-	int			count;
+	unsigned int		count;
 	struct list_head	wait_list;
 };
 
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index 5a12a8558982..bef977b16966 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -18,18 +18,8 @@
  * down_trylock() and up() can be called from interrupt context.
  * So we have to disable interrupts when taking the lock.
  *
- * The ->count variable, if positive, defines how many more tasks can
- * acquire the semaphore.  If negative, it represents how many tasks are
- * waiting on the semaphore (*).  If zero, no tasks are waiting, and no more
- * tasks can acquire the semaphore.
- *
- * (*) Except for the window between one task calling up() and the task
- * sleeping in a __down_common() waking up.  In order to avoid a third task
- * coming in and stealing the second task's wakeup, we leave the ->count
- * negative.  If we have a more complex situation, the ->count may become
- * zero or negative (eg a semaphore with count = 2, three tasks attempt to
- * acquire it, one sleeps, two finish and call up(), the second task to call
- * up() notices that the list is empty and just increments count).
+ * The ->count variable defines how many more tasks can acquire the
+ * semaphore.  If it's zero, there may be tasks waiting on the list.
  */
 
 static noinline void __down(struct semaphore *sem);
@@ -43,7 +33,9 @@ void down(struct semaphore *sem)
 	unsigned long flags;
 
 	spin_lock_irqsave(&sem->lock, flags);
-	if (unlikely(sem->count-- <= 0))
+	if (likely(sem->count > 0))
+		sem->count--;
+	else
 		__down(sem);
 	spin_unlock_irqrestore(&sem->lock, flags);
 }
@@ -55,7 +47,9 @@ int down_interruptible(struct semaphore *sem)
 	int result = 0;
 
 	spin_lock_irqsave(&sem->lock, flags);
-	if (unlikely(sem->count-- <= 0))
+	if (likely(sem->count > 0))
+		sem->count--;
+	else
 		result = __down_interruptible(sem);
 	spin_unlock_irqrestore(&sem->lock, flags);
 
@@ -69,7 +63,9 @@ int down_killable(struct semaphore *sem)
 	int result = 0;
 
 	spin_lock_irqsave(&sem->lock, flags);
-	if (unlikely(sem->count-- <= 0))
+	if (likely(sem->count > 0))
+		sem->count--;
+	else
 		result = __down_killable(sem);
 	spin_unlock_irqrestore(&sem->lock, flags);
 
@@ -111,7 +107,9 @@ int down_timeout(struct semaphore *sem, long jiffies)
 	int result = 0;
 
 	spin_lock_irqsave(&sem->lock, flags);
-	if (unlikely(sem->count-- <= 0))
+	if (likely(sem->count > 0))
+		sem->count--;
+	else
 		result = __down_timeout(sem, jiffies);
 	spin_unlock_irqrestore(&sem->lock, flags);
 
@@ -124,7 +122,7 @@ void up(struct semaphore *sem)
 	unsigned long flags;
 
 	spin_lock_irqsave(&sem->lock, flags);
-	if (likely(sem->count >= 0))
+	if (likely(list_empty(&sem->wait_list)))
 		sem->count++;
 	else
 		__up(sem);
@@ -140,22 +138,6 @@ struct semaphore_waiter {
 	int up;
 };
 
-/*
- * Wake up a process waiting on a semaphore.  We need to call this from both
- * __up and __down_common as it's possible to race a task into the semaphore
- * if it comes in at just the right time between two tasks calling up() and
- * a third task waking up.  This function assumes the wait_list is already
- * checked for being non-empty.
- */
-static noinline void __sched __up_down_common(struct semaphore *sem)
-{
-	struct semaphore_waiter *waiter = list_first_entry(&sem->wait_list,
-						struct semaphore_waiter, list);
-	list_del(&waiter->list);
-	waiter->up = 1;
-	wake_up_process(waiter->task);
-}
-
 /*
  * Because this function is inlined, the 'state' parameter will be
  * constant, and thus optimised away by the compiler.  Likewise the
@@ -164,7 +146,6 @@ static noinline void __sched __up_down_common(struct semaphore *sem)
 static inline int __sched __down_common(struct semaphore *sem, long state,
 								long timeout)
 {
-	int result = 0;
 	struct task_struct *task = current;
 	struct semaphore_waiter waiter;
 
@@ -184,28 +165,16 @@ static inline int __sched __down_common(struct semaphore *sem, long state,
 		timeout = schedule_timeout(timeout);
 		spin_lock_irq(&sem->lock);
 		if (waiter.up)
-			goto woken;
+			return 0;
 	}
 
  timed_out:
 	list_del(&waiter.list);
-	result = -ETIME;
-	goto woken;
+	return -ETIME;
+
  interrupted:
 	list_del(&waiter.list);
-	result = -EINTR;
- woken:
-	/*
-	 * Account for the process which woke us up.  For the case where
-	 * we're interrupted, we need to increment the count on our own
-	 * behalf.  I don't believe we can hit the case where the
-	 * sem->count hits zero, *and* there's a second task sleeping,
-	 * but it doesn't hurt, that's not a commonly exercised path and
-	 * it's not a performance path either.
-	 */
-	if (unlikely((++sem->count >= 0) && !list_empty(&sem->wait_list)))
-		__up_down_common(sem);
-	return result;
+	return -EINTR;
 }
 
 static noinline void __sched __down(struct semaphore *sem)
@@ -230,8 +199,9 @@ static noinline int __sched __down_timeout(struct semaphore *sem, long jiffies)
 
 static noinline void __sched __up(struct semaphore *sem)
 {
-	if (unlikely(list_empty(&sem->wait_list)))
-		sem->count++;
-	else
-		__up_down_common(sem);
+	struct semaphore_waiter *waiter = list_first_entry(&sem->wait_list,
+						struct semaphore_waiter, list);
+	list_del(&waiter->list);
+	waiter->up = 1;
+	wake_up_process(waiter->task);
 }
-- 
cgit v1.2.3


From 714493cd5468f42ca3c4f730a9c17c203abd5059 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Fri, 11 Apr 2008 15:23:52 -0400
Subject: Improve semaphore documentation

Move documentation from semaphore.h to semaphore.c as requested by
Andrew Morton.  Also reformat to kernel-doc style and add some more
notes about the implementation.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 include/linux/semaphore.h | 39 ++---------------------
 kernel/semaphore.c        | 79 ++++++++++++++++++++++++++++++++++++++++-------
 2 files changed, 70 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
index a7125daaff90..9cae64b00d6b 100644
--- a/include/linux/semaphore.h
+++ b/include/linux/semaphore.h
@@ -4,8 +4,7 @@
  *
  * Distributed under the terms of the GNU GPL, version 2
  *
- * Counting semaphores allow up to <n> tasks to acquire the semaphore
- * simultaneously.
+ * Please see kernel/semaphore.c for documentation of these functions
  */
 #ifndef __LINUX_SEMAPHORE_H
 #define __LINUX_SEMAPHORE_H
@@ -13,11 +12,7 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 
-/*
- * The spinlock controls access to the other members of the semaphore.
- * 'count' represents how many more tasks can acquire this semaphore.
- * Tasks waiting for the lock are kept on the wait_list.
- */
+/* Please don't access any members of this structure directly */
 struct semaphore {
 	spinlock_t		lock;
 	unsigned int		count;
@@ -46,41 +41,11 @@ static inline void sema_init(struct semaphore *sem, int val)
 #define init_MUTEX(sem)		sema_init(sem, 1)
 #define init_MUTEX_LOCKED(sem)	sema_init(sem, 0)
 
-/*
- * Attempt to acquire the semaphore.  If another task is already holding the
- * semaphore, sleep until the semaphore is released.
- */
 extern void down(struct semaphore *sem);
-
-/*
- * As down(), except the sleep may be interrupted by a signal.  If it is,
- * this function will return -EINTR.
- */
 extern int __must_check down_interruptible(struct semaphore *sem);
-
-/*
- * As down_interruptible(), except the sleep may only be interrupted by
- * signals which are fatal to this process.
- */
 extern int __must_check down_killable(struct semaphore *sem);
-
-/*
- * As down(), except this function will not sleep.  It will return 0 if it
- * acquired the semaphore and 1 if the semaphore was contended.  This
- * function may be called from any context, including interrupt and softirq.
- */
 extern int __must_check down_trylock(struct semaphore *sem);
-
-/*
- * As down(), except this function will return -ETIME if it fails to
- * acquire the semaphore within the specified number of jiffies.
- */
 extern int __must_check down_timeout(struct semaphore *sem, long jiffies);
-
-/*
- * Release the semaphore.  Unlike mutexes, up() may be called from any
- * context and even by tasks which have never called down().
- */
 extern void up(struct semaphore *sem);
 
 #endif /* __LINUX_SEMAPHORE_H */
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index bef977b16966..5c2942e768cd 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -3,6 +3,26 @@
  * Author: Matthew Wilcox <willy@linux.intel.com>
  *
  * Distributed under the terms of the GNU GPL, version 2
+ *
+ * This file implements counting semaphores.
+ * A counting semaphore may be acquired 'n' times before sleeping.
+ * See mutex.c for single-acquisition sleeping locks which enforce
+ * rules which allow code to be debugged more easily.
+ */
+
+/*
+ * Some notes on the implementation:
+ *
+ * The spinlock controls access to the other members of the semaphore.
+ * down_trylock() and up() can be called from interrupt context, so we
+ * have to disable interrupts when taking the lock.  It turns out various
+ * parts of the kernel expect to be able to use down() on a semaphore in
+ * interrupt context when they know it will succeed, so we have to use
+ * irqsave variants for down(), down_interruptible() and down_killable()
+ * too.
+ *
+ * The ->count variable represents how many more tasks can acquire this
+ * semaphore.  If it's zero, there may be tasks waiting on the wait_list.
  */
 
 #include <linux/compiler.h>
@@ -12,22 +32,23 @@
 #include <linux/semaphore.h>
 #include <linux/spinlock.h>
 
-/*
- * Some notes on the implementation:
- *
- * down_trylock() and up() can be called from interrupt context.
- * So we have to disable interrupts when taking the lock.
- *
- * The ->count variable defines how many more tasks can acquire the
- * semaphore.  If it's zero, there may be tasks waiting on the list.
- */
-
 static noinline void __down(struct semaphore *sem);
 static noinline int __down_interruptible(struct semaphore *sem);
 static noinline int __down_killable(struct semaphore *sem);
 static noinline int __down_timeout(struct semaphore *sem, long jiffies);
 static noinline void __up(struct semaphore *sem);
 
+/**
+ * down - acquire the semaphore
+ * @sem: the semaphore to be acquired
+ *
+ * Acquires the semaphore.  If no more tasks are allowed to acquire the
+ * semaphore, calling this function will put the task to sleep until the
+ * semaphore is released.
+ *
+ * Use of this function is deprecated, please use down_interruptible() or
+ * down_killable() instead.
+ */
 void down(struct semaphore *sem)
 {
 	unsigned long flags;
@@ -41,6 +62,15 @@ void down(struct semaphore *sem)
 }
 EXPORT_SYMBOL(down);
 
+/**
+ * down_interruptible - acquire the semaphore unless interrupted
+ * @sem: the semaphore to be acquired
+ *
+ * Attempts to acquire the semaphore.  If no more tasks are allowed to
+ * acquire the semaphore, calling this function will put the task to sleep.
+ * If the sleep is interrupted by a signal, this function will return -EINTR.
+ * If the semaphore is successfully acquired, this function returns 0.
+ */
 int down_interruptible(struct semaphore *sem)
 {
 	unsigned long flags;
@@ -57,6 +87,16 @@ int down_interruptible(struct semaphore *sem)
 }
 EXPORT_SYMBOL(down_interruptible);
 
+/**
+ * down_killable - acquire the semaphore unless killed
+ * @sem: the semaphore to be acquired
+ *
+ * Attempts to acquire the semaphore.  If no more tasks are allowed to
+ * acquire the semaphore, calling this function will put the task to sleep.
+ * If the sleep is interrupted by a fatal signal, this function will return
+ * -EINTR.  If the semaphore is successfully acquired, this function returns
+ * 0.
+ */
 int down_killable(struct semaphore *sem)
 {
 	unsigned long flags;
@@ -78,7 +118,7 @@ EXPORT_SYMBOL(down_killable);
  * @sem: the semaphore to be acquired
  *
  * Try to acquire the semaphore atomically.  Returns 0 if the mutex has
- * been acquired successfully and 1 if it is contended.
+ * been acquired successfully or 1 if it it cannot be acquired.
  *
  * NOTE: This return value is inverted from both spin_trylock and
  * mutex_trylock!  Be careful about this when converting code.
@@ -101,6 +141,16 @@ int down_trylock(struct semaphore *sem)
 }
 EXPORT_SYMBOL(down_trylock);
 
+/**
+ * down_timeout - acquire the semaphore within a specified time
+ * @sem: the semaphore to be acquired
+ * @jiffies: how long to wait before failing
+ *
+ * Attempts to acquire the semaphore.  If no more tasks are allowed to
+ * acquire the semaphore, calling this function will put the task to sleep.
+ * If the semaphore is not released within the specified number of jiffies,
+ * this function returns -ETIME.  It returns 0 if the semaphore was acquired.
+ */
 int down_timeout(struct semaphore *sem, long jiffies)
 {
 	unsigned long flags;
@@ -117,6 +167,13 @@ int down_timeout(struct semaphore *sem, long jiffies)
 }
 EXPORT_SYMBOL(down_timeout);
 
+/**
+ * up - release the semaphore
+ * @sem: the semaphore to release
+ *
+ * Release the semaphore.  Unlike mutexes, up() may be called from any
+ * context and even by tasks which have never called down().
+ */
 void up(struct semaphore *sem)
 {
 	unsigned long flags;
-- 
cgit v1.2.3


From 85eb69a16aab5a394ce043c2131319eae35e6493 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 21 Feb 2008 12:50:51 +0100
Subject: x86: increase the kernel text limit to 512 MB

people sometimes do crazy stuff like building really large static
arrays into their kernels or building allyesconfig kernels. Give
more space to the kernel and push modules up a bit: kernel has
512 MB and modules have 1.5 GB.

Should be enough for a few years ;-)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/head_64.S    | 4 ++--
 include/asm-x86/page_64.h    | 4 ++--
 include/asm-x86/pgtable_64.h | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index a007454133a3..017216916dff 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -383,12 +383,12 @@ NEXT_PAGE(level2_ident_pgt)
 
 NEXT_PAGE(level2_kernel_pgt)
 	/*
-	 * 128 MB kernel mapping. We spend a full page on this pagetable
+	 * 512 MB kernel mapping. We spend a full page on this pagetable
 	 * anyway.
 	 *
 	 * The kernel code+data+bss must not be bigger than that.
 	 *
-	 * (NOTE: at +128MB starts the module area, see MODULES_VADDR.
+	 * (NOTE: at +512MB starts the module area, see MODULES_VADDR.
 	 *  If you want to increase this then increase MODULES_VADDR
 	 *  too.)
 	 */
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index 143546073b95..aee05c616e05 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -48,10 +48,10 @@
 #define __VIRTUAL_MASK_SHIFT	48
 
 /*
- * Kernel image size is limited to 128 MB (see level2_kernel_pgt in
+ * Kernel image size is limited to 512 MB (see level2_kernel_pgt in
  * arch/x86/kernel/head_64.S), and it is mapped here:
  */
-#define KERNEL_IMAGE_SIZE	(128*1024*1024)
+#define KERNEL_IMAGE_SIZE	(512*1024*1024)
 #define KERNEL_IMAGE_START	_AC(0xffffffff80000000, UL)
 
 #ifndef __ASSEMBLY__
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 0a0b77bc736a..01d2359e7a34 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -140,7 +140,7 @@ static inline void native_pgd_clear(pgd_t * pgd)
 #define VMALLOC_START    _AC(0xffffc20000000000, UL)
 #define VMALLOC_END      _AC(0xffffe1ffffffffff, UL)
 #define VMEMMAP_START	 _AC(0xffffe20000000000, UL)
-#define MODULES_VADDR    _AC(0xffffffff88000000, UL)
+#define MODULES_VADDR    _AC(0xffffffffa0000000, UL)
 #define MODULES_END      _AC(0xfffffffffff00000, UL)
 #define MODULES_LEN   (MODULES_END - MODULES_VADDR)
 
-- 
cgit v1.2.3


From 00d1c5e05736f947687be27706bda01cec104e57 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 17 Apr 2008 17:40:45 +0200
Subject: x86: add gbpages switches

These new controls toggle experimental support for a new CPU feature,
the straightforward extension of largepages from the pmd level to the
pud level, which allows 1GB (kernel) TLBs instead of 2MB TLBs.

Turn it off by default, as this code has not been tested well enough yet.

Use the CONFIG_DIRECT_GBPAGES=y .config option or gbpages on the
boot line can be used to enable it. If enabled in the .config then
nogbpages boot option disables it.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/x86_64/boot-options.txt |  5 +++++
 arch/x86/Kconfig.debug                | 12 ++++++++++++
 arch/x86/mm/init_64.c                 | 20 ++++++++++++++++++++
 include/asm-x86/pgtable_64.h          |  2 ++
 4 files changed, 39 insertions(+)

(limited to 'include')

diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt
index 34abae4e9442..b0c7b6c4abda 100644
--- a/Documentation/x86_64/boot-options.txt
+++ b/Documentation/x86_64/boot-options.txt
@@ -307,3 +307,8 @@ Debugging
 			stuck (default)
 
 Miscellaneous
+
+	nogbpages
+		Do not use GB pages for kernel direct mappings.
+	gbpages
+		Use GB pages for kernel direct mappings.
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 7ce8e7025661..f4413c04e687 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -76,6 +76,18 @@ config DEBUG_RODATA
 	  data. This is recommended so that we can catch kernel bugs sooner.
 	  If in doubt, say "Y".
 
+config DIRECT_GBPAGES
+	bool "Enable gbpages-mapped kernel pagetables"
+	depends on DEBUG_KERNEL && EXPERIMENTAL && X86_64
+	help
+	  Enable gigabyte pages support (if the CPU supports it). This can
+	  improve the kernel's performance a tiny bit by reducing TLB
+	  pressure.
+
+	  This is experimental code.
+
+	  If in doubt, say "N".
+
 config DEBUG_RODATA_TEST
 	bool "Testcase for the DEBUG_RODATA feature"
 	depends on DEBUG_RODATA
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index a02a14f0f324..6e7d5a42a09a 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -54,6 +54,26 @@ static unsigned long dma_reserve __initdata;
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
+int direct_gbpages __meminitdata
+#ifdef CONFIG_DIRECT_GBPAGES
+				= 1
+#endif
+;
+
+static int __init parse_direct_gbpages_off(char *arg)
+{
+	direct_gbpages = 0;
+	return 0;
+}
+early_param("nogbpages", parse_direct_gbpages_off);
+
+static int __init parse_direct_gbpages_on(char *arg)
+{
+	direct_gbpages = 1;
+	return 0;
+}
+early_param("gbpages", parse_direct_gbpages_on);
+
 /*
  * NOTE: pagetable_init alloc all the fixmap pagetables contiguous on the
  * physical space so we can cache the place of the first one and move
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 01d2359e7a34..6ef09914acbe 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -239,6 +239,8 @@ static inline int pud_large(pud_t pte)
 
 #define update_mmu_cache(vma,address,pte) do { } while (0)
 
+extern int direct_gbpages;
+
 /* Encode and de-code a swap entry */
 #define __swp_type(x)			(((x).val >> 1) & 0x3f)
 #define __swp_offset(x)			((x).val >> 8)
-- 
cgit v1.2.3


From 48c508b364324c35018284328b5b92c51d2b30e0 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Thu, 17 Apr 2008 17:40:45 +0200
Subject: x86: clean up find_e820_area(), 64-bit

Change size to unsigned long, becase caller and user all used unsigned long.
Also make bad_addr take an alignment parameter.

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/e820_64.c | 22 ++++++++++++----------
 include/asm-x86/e820_64.h |  2 +-
 2 files changed, 13 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index 9be697126013..a8694a35352b 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -95,7 +95,8 @@ void __init early_res_to_bootmem(void)
 }
 
 /* Check for already reserved areas */
-static inline int bad_addr(unsigned long *addrp, unsigned long size)
+static inline int
+bad_addr(unsigned long *addrp, unsigned long size, unsigned long align)
 {
 	int i;
 	unsigned long addr = *addrp, last;
@@ -105,7 +106,7 @@ again:
 	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
 		struct early_res *r = &early_res[i];
 		if (last >= r->start && addr < r->end) {
-			*addrp = addr = r->end;
+			*addrp = addr = round_up(r->end, align);
 			changed = 1;
 			goto again;
 		}
@@ -174,26 +175,27 @@ int __init e820_all_mapped(unsigned long start, unsigned long end,
  * Find a free area with specified alignment in a specific range.
  */
 unsigned long __init find_e820_area(unsigned long start, unsigned long end,
-				    unsigned size, unsigned long align)
+				    unsigned long size, unsigned long align)
 {
 	int i;
-	unsigned long mask = ~(align - 1);
 
 	for (i = 0; i < e820.nr_map; i++) {
 		struct e820entry *ei = &e820.map[i];
-		unsigned long addr = ei->addr, last;
+		unsigned long addr, last;
+		unsigned long ei_last;
 
 		if (ei->type != E820_RAM)
 			continue;
+		addr = round_up(ei->addr, align);
+		ei_last = ei->addr + ei->size;
 		if (addr < start)
-			addr = start;
-		if (addr > ei->addr + ei->size)
+			addr = round_up(start, align);
+		if (addr > ei_last)
 			continue;
-		while (bad_addr(&addr, size) && addr+size <= ei->addr+ei->size)
+		while (bad_addr(&addr, size, align) && addr+size <= ei_last)
 			;
-		addr = (addr + align - 1) & mask;
 		last = addr + size;
-		if (last > ei->addr + ei->size)
+		if (last > ei_last)
 			continue;
 		if (last > end)
 			continue;
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index 22ede73ae724..9e06c6eb4e27 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -15,7 +15,7 @@
 
 #ifndef __ASSEMBLY__
 extern unsigned long find_e820_area(unsigned long start, unsigned long end, 
-				    unsigned size, unsigned long align);
+				    unsigned long size, unsigned long align);
 extern void add_memory_region(unsigned long start, unsigned long size, 
 			      int type);
 extern void update_memory_range(u64 start, u64 size, unsigned old_type,
-- 
cgit v1.2.3


From 92bc2056855b3250bf6fd5849f05f88d85839efa Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 8 Feb 2008 12:09:56 -0800
Subject: x86: change most X86_32 pt_regs members to unsigned long

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_32.c |  5 ++---
 arch/x86/kernel/signal_32.c  |  4 ++--
 include/asm-x86/ptrace.h     | 28 ++++++++++++++--------------
 3 files changed, 18 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 43930e73f657..2cd89b8a7050 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -332,7 +332,7 @@ void __show_registers(struct pt_regs *regs, int all)
 			init_utsname()->version);
 
 	printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n",
-			0xffff & regs->cs, regs->ip, regs->flags,
+			(u16)regs->cs, regs->ip, regs->flags,
 			smp_processor_id());
 	print_symbol("EIP is at %s\n", regs->ip);
 
@@ -341,8 +341,7 @@ void __show_registers(struct pt_regs *regs, int all)
 	printk("ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n",
 		regs->si, regs->di, regs->bp, sp);
 	printk(" DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n",
-	       regs->ds & 0xffff, regs->es & 0xffff,
-	       regs->fs & 0xffff, gs, ss);
+	       (u16)regs->ds, (u16)regs->es, (u16)regs->fs, gs, ss);
 
 	if (!all)
 		return;
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 011c62fa563c..34fc0416b135 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -393,8 +393,8 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 	regs->sp = (unsigned long) frame;
 	regs->ip = (unsigned long) ka->sa.sa_handler;
 	regs->ax = (unsigned long) sig;
-	regs->dx = (unsigned long) 0;
-	regs->cx = (unsigned long) 0;
+	regs->dx = 0;
+	regs->cx = 0;
 
 	regs->ds = __USER_DS;
 	regs->es = __USER_DS;
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index d9e04b46a440..708337a36727 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -36,23 +36,23 @@ struct pt_regs {
 #else /* __KERNEL__ */
 
 struct pt_regs {
-	long bx;
-	long cx;
-	long dx;
-	long si;
-	long di;
-	long bp;
+	unsigned long bx;
+	unsigned long cx;
+	unsigned long dx;
+	unsigned long si;
+	unsigned long di;
+	unsigned long bp;
 	long ax;
-	int  ds;
-	int  es;
-	int  fs;
+	unsigned long ds;
+	unsigned long es;
+	unsigned long fs;
 	/* int  gs; */
 	long orig_ax;
-	long ip;
-	int  cs;
-	long flags;
-	long sp;
-	int  ss;
+	unsigned long ip;
+	unsigned long cs;
+	unsigned long flags;
+	unsigned long sp;
+	unsigned long ss;
 };
 
 #include <asm/vm86.h>
-- 
cgit v1.2.3


From 9902a702c76f904be0057f8647dda9d6f89d4847 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 8 Feb 2008 12:09:57 -0800
Subject: x86: make X86_32 pt_regs members unsigned long

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/signal_32.c | 4 ++--
 include/asm-x86/ptrace.h    | 9 ++++++---
 2 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 34fc0416b135..b1e97e80a57c 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -527,7 +527,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	int ret;
 
 	/* Are we from a system call? */
-	if (regs->orig_ax >= 0) {
+	if ((long)regs->orig_ax >= 0) {
 		/* If so, check system call restarting.. */
 		switch (regs->ax) {
 		        case -ERESTART_RESTARTBLOCK:
@@ -625,7 +625,7 @@ static void do_signal(struct pt_regs *regs)
 	}
 
 	/* Did we come from a system call? */
-	if (regs->orig_ax >= 0) {
+	if ((long)regs->orig_ax >= 0) {
 		/* Restart the system call - no handlers present */
 		switch (regs->ax) {
 		case -ERESTARTNOHAND:
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index 708337a36727..bc442461ac64 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -42,12 +42,12 @@ struct pt_regs {
 	unsigned long si;
 	unsigned long di;
 	unsigned long bp;
-	long ax;
+	unsigned long ax;
 	unsigned long ds;
 	unsigned long es;
 	unsigned long fs;
 	/* int  gs; */
-	long orig_ax;
+	unsigned long orig_ax;
 	unsigned long ip;
 	unsigned long cs;
 	unsigned long flags;
@@ -145,7 +145,10 @@ extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int erro
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
 #endif
 
-#define regs_return_value(regs) ((regs)->ax)
+static inline unsigned long regs_return_value(struct pt_regs *regs)
+{
+	return regs->ax;
+}
 
 /*
  * user_mode_vm(regs) determines whether a register set came from user mode.
-- 
cgit v1.2.3


From 04adf11435a5187383c35017a94b55701984243b Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Sat, 16 Feb 2008 23:02:03 -0800
Subject: x86: remove never used nodenumer in pda

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/numa_64.c | 2 --
 include/asm-x86/pda.h | 1 -
 2 files changed, 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 16b82ad34b96..18267a02e67a 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -548,8 +548,6 @@ void __cpuinit numa_set_node(int cpu, int node)
 {
 	int *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
 
-	cpu_pda(cpu)->nodenumber = node;
-
 	if(cpu_to_node_map)
 		cpu_to_node_map[cpu] = node;
 	else if(per_cpu_offset(cpu))
diff --git a/include/asm-x86/pda.h b/include/asm-x86/pda.h
index c0305bff0f19..d9dc209c24ad 100644
--- a/include/asm-x86/pda.h
+++ b/include/asm-x86/pda.h
@@ -22,7 +22,6 @@ struct x8664_pda {
 					   offset 40!!! */
 #endif
 	char *irqstackptr;
-	unsigned int nodenumber;	/* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
 	short mmu_state;
-- 
cgit v1.2.3


From 2785c8d052278228cc3806233c09295088f83d42 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 11 Feb 2008 17:16:03 -0200
Subject: x86: call vsmp_init explicitly

It becomes to early for ioremap, so we use early_ioremap

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalemp.com>
Acked-by: Shai Fultheim <shai@scalemp.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/setup_64.c |  4 ++++
 arch/x86/kernel/vsmp_64.c  | 11 +++++------
 include/asm-x86/setup.h    |  4 ++++
 3 files changed, 13 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index f4f7ecfb898c..e67925674eae 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -345,6 +345,10 @@ void __init setup_arch(char **cmdline_p)
 	if (efi_enabled)
 		efi_init();
 
+#ifdef	CONFIG_PARAVIRT
+	vsmp_init();
+#endif
+
 	dmi_scan_machine();
 
 	io_delay_init();
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index 976691726de4..fdf9fba6ba9c 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -16,10 +16,10 @@
 #include <asm/pci-direct.h>
 #include <asm/io.h>
 
-static void __init vsmp_init(void)
+void __init vsmp_init(void)
 {
 	void *address;
-	unsigned int cap, ctl;
+	unsigned int cap, ctl, cfg;
 
 	if (!early_pci_allowed())
 		return;
@@ -32,7 +32,8 @@ static void __init vsmp_init(void)
 		return;
 
 	/* set vSMP magic bits to indicate vSMP capable kernel */
-	address = ioremap(read_pci_config(0, 0x1f, 0, PCI_BASE_ADDRESS_0), 8);
+	cfg = read_pci_config(0, 0x1f, 0, PCI_BASE_ADDRESS_0);
+	address = early_ioremap(cfg, 8);
 	cap = readl(address);
 	ctl = readl(address + 4);
 	printk(KERN_INFO "vSMP CTL: capabilities:0x%08x  control:0x%08x\n",
@@ -45,8 +46,6 @@ static void __init vsmp_init(void)
 		printk(KERN_INFO "vSMP CTL: control set to:0x%08x\n", ctl);
 	}
 
-	iounmap(address);
+	early_iounmap(address, 8);
 	return;
 }
-
-core_initcall(vsmp_init);
diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h
index 071e054abd82..f745de211191 100644
--- a/include/asm-x86/setup.h
+++ b/include/asm-x86/setup.h
@@ -4,6 +4,10 @@
 #define COMMAND_LINE_SIZE 2048
 
 #ifndef __ASSEMBLY__
+
+/* Interrupt control for vSMPowered x86_64 systems */
+void vsmp_init(void);
+
 char *machine_specific_memory_setup(void);
 #ifndef CONFIG_PARAVIRT
 #define paravirt_post_allocator_init()	do {} while (0)
-- 
cgit v1.2.3


From bc0a733facbbde6c464e3ba5e165607fe4824cca Mon Sep 17 00:00:00 2001
From: "David P. Reed" <dpreed@reed.com>
Date: Mon, 18 Feb 2008 13:58:34 -0500
Subject: x86: define outb_pic and inb_pic to stop using outb_p and inb_p

x86: define outb_pic and inb_pic to stop using outb_p and inb_p

The delay between io port accesses to the PIC is now defined using outb_pic
and inb_pic.  This fix provides the next step, using udelay(2) to define the
*PIC specific* timing requirements, rather than on bus-oriented timing, which
is not well calibrated.

Again, the primary reason for fixing this is to use proper delay strategy,
and in particular to fix crashes that can result from using port 80 writes
on machines that have resources on port 80, such as the ENE chips used by Quanta
in latops it designs and sells to, e.g. HP.

Signed-off-by: David P. Reed <dpreed@reed.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/i8259.h | 27 +++++++++++++++++++++++++--
 1 file changed, 25 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/i8259.h b/include/asm-x86/i8259.h
index 67c319e0efc7..e2650f21ca8d 100644
--- a/include/asm-x86/i8259.h
+++ b/include/asm-x86/i8259.h
@@ -1,6 +1,8 @@
 #ifndef __ASM_I8259_H__
 #define __ASM_I8259_H__
 
+#include <linux/delay.h>
+
 extern unsigned int cached_irq_mask;
 
 #define __byte(x,y)		(((unsigned char *) &(y))[x])
@@ -29,7 +31,28 @@ extern void enable_8259A_irq(unsigned int irq);
 extern void disable_8259A_irq(unsigned int irq);
 extern unsigned int startup_8259A_irq(unsigned int irq);
 
-#define inb_pic		inb_p
-#define outb_pic	outb_p
+/* the PIC may need a careful delay on some platforms, hence specific calls */
+static inline unsigned char inb_pic(unsigned int port)
+{
+	unsigned char value = inb(port);
+
+	/*
+	 * delay for some accesses to PIC on motherboard or in chipset
+	 * must be at least one microsecond, so be safe here:
+	 */
+	udelay(2);
+
+	return value;
+}
+
+static inline void outb_pic(unsigned char value, unsigned int port)
+{
+	outb(value, port);
+	/*
+	 * delay for some accesses to PIC on motherboard or in chipset
+	 * must be at least one microsecond, so be safe here:
+	 */
+	udelay(2);
+}
 
 #endif	/* __ASM_I8259_H__ */
-- 
cgit v1.2.3


From 4d46a89e7c867718020b2d5fd8f9e775293304be Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 21 Feb 2008 04:24:40 +0100
Subject: x86: clean up include/asm-x86/processor.h

basic style cleanup to flush out years of neglect:

 - consistent indentation
 - whitespace fixes
 - consistent comments

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor.h | 638 +++++++++++++++++++++++++-------------------
 1 file changed, 358 insertions(+), 280 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 45a2f0ab33d0..43d2cc829a94 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -24,6 +24,7 @@ struct mm_struct;
 #include <asm/msr.h>
 #include <asm/desc_defs.h>
 #include <asm/nops.h>
+
 #include <linux/personality.h>
 #include <linux/cpumask.h>
 #include <linux/cache.h>
@@ -37,16 +38,18 @@ struct mm_struct;
 static inline void *current_text_addr(void)
 {
 	void *pc;
-	asm volatile("mov $1f,%0\n1:":"=r" (pc));
+
+	asm volatile("mov $1f, %0; 1:":"=r" (pc));
+
 	return pc;
 }
 
 #ifdef CONFIG_X86_VSMP
-#define ARCH_MIN_TASKALIGN	(1 << INTERNODE_CACHE_SHIFT)
-#define ARCH_MIN_MMSTRUCT_ALIGN	(1 << INTERNODE_CACHE_SHIFT)
+# define ARCH_MIN_TASKALIGN		(1 << INTERNODE_CACHE_SHIFT)
+# define ARCH_MIN_MMSTRUCT_ALIGN	(1 << INTERNODE_CACHE_SHIFT)
 #else
-#define ARCH_MIN_TASKALIGN	16
-#define ARCH_MIN_MMSTRUCT_ALIGN	0
+# define ARCH_MIN_TASKALIGN		16
+# define ARCH_MIN_MMSTRUCT_ALIGN	0
 #endif
 
 /*
@@ -56,69 +59,81 @@ static inline void *current_text_addr(void)
  */
 
 struct cpuinfo_x86 {
-	__u8	x86;		/* CPU family */
-	__u8	x86_vendor;	/* CPU vendor */
-	__u8	x86_model;
-	__u8	x86_mask;
+	__u8			x86;		/* CPU family */
+	__u8			x86_vendor;	/* CPU vendor */
+	__u8			x86_model;
+	__u8			x86_mask;
 #ifdef CONFIG_X86_32
-	char	wp_works_ok;	/* It doesn't on 386's */
-	char	hlt_works_ok;	/* Problems on some 486Dx4's and old 386's */
-	char	hard_math;
-	char	rfu;
-	char	fdiv_bug;
-	char	f00f_bug;
-	char	coma_bug;
-	char	pad0;
+	char			wp_works_ok;	/* It doesn't on 386's */
+
+	/* Problems on some 486Dx4's and old 386's: */
+	char			hlt_works_ok;
+	char			hard_math;
+	char			rfu;
+	char			fdiv_bug;
+	char			f00f_bug;
+	char			coma_bug;
+	char			pad0;
 #else
-	/* number of 4K pages in DTLB/ITLB combined(in pages)*/
-	int     x86_tlbsize;
-	__u8    x86_virt_bits, x86_phys_bits;
-	/* cpuid returned core id bits */
-	__u8    x86_coreid_bits;
-	/* Max extended CPUID function supported */
-	__u32   extended_cpuid_level;
+	/* Number of 4K pages in DTLB/ITLB combined(in pages): */
+	int			 x86_tlbsize;
+	__u8			x86_virt_bits;
+	__u8			x86_phys_bits;
+	/* CPUID returned core id bits: */
+	__u8			x86_coreid_bits;
+	/* Max extended CPUID function supported: */
+	__u32			extended_cpuid_level;
 #endif
-	int	cpuid_level;	/* Maximum supported CPUID level, -1=no CPUID */
-	__u32	x86_capability[NCAPINTS];
-	char	x86_vendor_id[16];
-	char	x86_model_id[64];
-	int 	x86_cache_size;  /* in KB - valid for CPUS which support this
-				    call  */
-	int 	x86_cache_alignment;	/* In bytes */
-	int	x86_power;
-	unsigned long loops_per_jiffy;
+	/* Maximum supported CPUID level, -1=no CPUID: */
+	int			cpuid_level;
+	__u32			x86_capability[NCAPINTS];
+	char			x86_vendor_id[16];
+	char			x86_model_id[64];
+	/* in KB - valid for CPUS which support this call: */
+	int			x86_cache_size;
+	int			x86_cache_alignment;	/* In bytes */
+	int			x86_power;
+	unsigned long		loops_per_jiffy;
 #ifdef CONFIG_SMP
-	cpumask_t llc_shared_map;	/* cpus sharing the last level cache */
+	/* cpus sharing the last level cache: */
+	cpumask_t		llc_shared_map;
 #endif
-	u16 x86_max_cores;		/* cpuid returned max cores value */
-	u16 apicid;
-	u16 x86_clflush_size;
+	/* cpuid returned max cores value: */
+	u16			 x86_max_cores;
+	u16			apicid;
+	u16			x86_clflush_size;
 #ifdef CONFIG_SMP
-	u16 booted_cores;		/* number of cores as seen by OS */
-	u16 phys_proc_id; 		/* Physical processor id. */
-	u16 cpu_core_id;  		/* Core id */
-	u16 cpu_index;			/* index into per_cpu list */
+	/* number of cores as seen by the OS: */
+	u16			booted_cores;
+	/* Physical processor id: */
+	u16			phys_proc_id;
+	/* Core id: */
+	u16			cpu_core_id;
+	/* Index into per_cpu list: */
+	u16			cpu_index;
 #endif
 } __attribute__((__aligned__(SMP_CACHE_BYTES)));
 
-#define X86_VENDOR_INTEL 0
-#define X86_VENDOR_CYRIX 1
-#define X86_VENDOR_AMD 2
-#define X86_VENDOR_UMC 3
-#define X86_VENDOR_NEXGEN 4
-#define X86_VENDOR_CENTAUR 5
-#define X86_VENDOR_TRANSMETA 7
-#define X86_VENDOR_NSC 8
-#define X86_VENDOR_NUM 9
-#define X86_VENDOR_UNKNOWN 0xff
+#define X86_VENDOR_INTEL	0
+#define X86_VENDOR_CYRIX	1
+#define X86_VENDOR_AMD		2
+#define X86_VENDOR_UMC		3
+#define X86_VENDOR_NEXGEN	4
+#define X86_VENDOR_CENTAUR	5
+#define X86_VENDOR_TRANSMETA	7
+#define X86_VENDOR_NSC		8
+#define X86_VENDOR_NUM		9
+
+#define X86_VENDOR_UNKNOWN	0xff
 
 /*
  * capabilities of CPUs
  */
-extern struct cpuinfo_x86 boot_cpu_data;
-extern struct cpuinfo_x86 new_cpu_data;
-extern struct tss_struct doublefault_tss;
-extern __u32 cleared_cpu_caps[NCAPINTS];
+extern struct cpuinfo_x86	boot_cpu_data;
+extern struct cpuinfo_x86	new_cpu_data;
+
+extern struct tss_struct	doublefault_tss;
+extern __u32			cleared_cpu_caps[NCAPINTS];
 
 #ifdef CONFIG_SMP
 DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
@@ -129,7 +144,9 @@ DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
 #define current_cpu_data	boot_cpu_data
 #endif
 
-void cpu_detect(struct cpuinfo_x86 *c);
+#define cache_line_size()	(boot_cpu_data.x86_cache_alignment)
+
+extern void cpu_detect(struct cpuinfo_x86 *c);
 
 extern void identify_cpu(struct cpuinfo_x86 *);
 extern void identify_boot_cpu(void);
@@ -146,7 +163,7 @@ static inline void detect_ht(struct cpuinfo_x86 *c) {}
 #endif
 
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
-					 unsigned int *ecx, unsigned int *edx)
+				unsigned int *ecx, unsigned int *edx)
 {
 	/* ecx is often an input as well as an output. */
 	__asm__("cpuid"
@@ -165,54 +182,67 @@ static inline void load_cr3(pgd_t *pgdir)
 #ifdef CONFIG_X86_32
 /* This is the TSS defined by the hardware. */
 struct x86_hw_tss {
-	unsigned short	back_link, __blh;
-	unsigned long	sp0;
-	unsigned short	ss0, __ss0h;
-	unsigned long	sp1;
-	unsigned short	ss1, __ss1h;	/* ss1 caches MSR_IA32_SYSENTER_CS */
-	unsigned long	sp2;
-	unsigned short	ss2, __ss2h;
-	unsigned long	__cr3;
-	unsigned long	ip;
-	unsigned long	flags;
-	unsigned long	ax, cx, dx, bx;
-	unsigned long	sp, bp, si, di;
-	unsigned short	es, __esh;
-	unsigned short	cs, __csh;
-	unsigned short	ss, __ssh;
-	unsigned short	ds, __dsh;
-	unsigned short	fs, __fsh;
-	unsigned short	gs, __gsh;
-	unsigned short	ldt, __ldth;
-	unsigned short	trace, io_bitmap_base;
+	unsigned short		back_link, __blh;
+	unsigned long		sp0;
+	unsigned short		ss0, __ss0h;
+	unsigned long		sp1;
+	/* ss1 caches MSR_IA32_SYSENTER_CS: */
+	unsigned short		ss1, __ss1h;
+	unsigned long		sp2;
+	unsigned short		ss2, __ss2h;
+	unsigned long		__cr3;
+	unsigned long		ip;
+	unsigned long		flags;
+	unsigned long		ax;
+	unsigned long		cx;
+	unsigned long		dx;
+	unsigned long		bx;
+	unsigned long		sp;
+	unsigned long		bp;
+	unsigned long		si;
+	unsigned long		di;
+	unsigned short		es, __esh;
+	unsigned short		cs, __csh;
+	unsigned short		ss, __ssh;
+	unsigned short		ds, __dsh;
+	unsigned short		fs, __fsh;
+	unsigned short		gs, __gsh;
+	unsigned short		ldt, __ldth;
+	unsigned short		trace;
+	unsigned short		io_bitmap_base;
+
 } __attribute__((packed));
 #else
 struct x86_hw_tss {
-	u32 reserved1;
-	u64 sp0;
-	u64 sp1;
-	u64 sp2;
-	u64 reserved2;
-	u64 ist[7];
-	u32 reserved3;
-	u32 reserved4;
-	u16 reserved5;
-	u16 io_bitmap_base;
+	u32			reserved1;
+	u64			sp0;
+	u64			sp1;
+	u64			sp2;
+	u64			reserved2;
+	u64			ist[7];
+	u32			reserved3;
+	u32			reserved4;
+	u16			reserved5;
+	u16			io_bitmap_base;
+
 } __attribute__((packed)) ____cacheline_aligned;
 #endif
 
 /*
- * Size of io_bitmap.
+ * IO-bitmap sizes:
  */
-#define IO_BITMAP_BITS  65536
-#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
-#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
-#define IO_BITMAP_OFFSET offsetof(struct tss_struct, io_bitmap)
-#define INVALID_IO_BITMAP_OFFSET 0x8000
-#define INVALID_IO_BITMAP_OFFSET_LAZY 0x9000
+#define IO_BITMAP_BITS			65536
+#define IO_BITMAP_BYTES			(IO_BITMAP_BITS/8)
+#define IO_BITMAP_LONGS			(IO_BITMAP_BYTES/sizeof(long))
+#define IO_BITMAP_OFFSET		offsetof(struct tss_struct, io_bitmap)
+#define INVALID_IO_BITMAP_OFFSET	0x8000
+#define INVALID_IO_BITMAP_OFFSET_LAZY	0x9000
 
 struct tss_struct {
-	struct x86_hw_tss x86_tss;
+	/*
+	 * The hardware state:
+	 */
+	struct x86_hw_tss	x86_tss;
 
 	/*
 	 * The extra 1 is there because the CPU will access an
@@ -220,48 +250,54 @@ struct tss_struct {
 	 * bitmap. The extra byte must be all 1 bits, and must
 	 * be within the limit.
 	 */
-	unsigned long	io_bitmap[IO_BITMAP_LONGS + 1];
+	unsigned long		io_bitmap[IO_BITMAP_LONGS + 1];
 	/*
 	 * Cache the current maximum and the last task that used the bitmap:
 	 */
-	unsigned long io_bitmap_max;
-	struct thread_struct *io_bitmap_owner;
+	unsigned long		io_bitmap_max;
+	struct thread_struct	*io_bitmap_owner;
+
 	/*
-	 * pads the TSS to be cacheline-aligned (size is 0x100)
+	 * Pad the TSS to be cacheline-aligned (size is 0x100):
 	 */
-	unsigned long __cacheline_filler[35];
+	unsigned long		__cacheline_filler[35];
 	/*
-	 * .. and then another 0x100 bytes for emergency kernel stack
+	 * .. and then another 0x100 bytes for the emergency kernel stack:
 	 */
-	unsigned long stack[64];
+	unsigned long		stack[64];
+
 } __attribute__((packed));
 
 DECLARE_PER_CPU(struct tss_struct, init_tss);
 
-/* Save the original ist values for checking stack pointers during debugging */
+/*
+ * Save the original ist values for checking stack pointers during debugging
+ */
 struct orig_ist {
-	unsigned long ist[7];
+	unsigned long		ist[7];
 };
 
 #define	MXCSR_DEFAULT		0x1f80
 
 struct i387_fsave_struct {
-	u32	cwd;
-	u32	swd;
-	u32	twd;
-	u32	fip;
-	u32	fcs;
-	u32	foo;
-	u32	fos;
-	u32	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
-	u32	status;		/* software status information */
+	u32			cwd;
+	u32			swd;
+	u32			twd;
+	u32			fip;
+	u32			fcs;
+	u32			foo;
+	u32			fos;
+	/* 8*10 bytes for each FP-reg = 80 bytes: */
+	u32			st_space[20];
+	/* Software status information: */
+	u32			status;
 };
 
 struct i387_fxsave_struct {
-	u16	cwd;
-	u16	swd;
-	u16	twd;
-	u16	fop;
+	u16			cwd;
+	u16			swd;
+	u16			twd;
+	u16			fop;
 	union {
 		struct {
 			u64	rip;
@@ -274,31 +310,40 @@ struct i387_fxsave_struct {
 			u32	fos;
 		};
 	};
-	u32	mxcsr;
-	u32	mxcsr_mask;
-	u32	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
-	u32	xmm_space[64];	/* 16*16 bytes for each XMM-reg = 256 bytes */
-	u32	padding[24];
+	u32			mxcsr;
+	u32			mxcsr_mask;
+	/* 8*16 bytes for each FP-reg = 128 bytes: */
+	u32			st_space[32];
+	/* 16*16 bytes for each XMM-reg = 256 bytes: */
+	u32			xmm_space[64];
+	u32			padding[24];
+
 } __attribute__((aligned(16)));
 
 struct i387_soft_struct {
-	u32	cwd;
-	u32	swd;
-	u32	twd;
-	u32	fip;
-	u32	fcs;
-	u32	foo;
-	u32	fos;
-	u32	st_space[20];	/* 8*10 bytes for each FP-reg = 80 bytes */
-	u8	ftop, changed, lookahead, no_update, rm, alimit;
-	struct info	*info;
-	u32	entry_eip;
+	u32			cwd;
+	u32			swd;
+	u32			twd;
+	u32			fip;
+	u32			fcs;
+	u32			foo;
+	u32			fos;
+	/* 8*10 bytes for each FP-reg = 80 bytes: */
+	u32			st_space[20];
+	u8			ftop;
+	u8			changed;
+	u8			lookahead;
+	u8			no_update;
+	u8			rm;
+	u8			alimit;
+	struct info		*info;
+	u32			entry_eip;
 };
 
 union i387_union {
 	struct i387_fsave_struct	fsave;
 	struct i387_fxsave_struct	fxsave;
-	struct i387_soft_struct 	soft;
+	struct i387_soft_struct		soft;
 };
 
 #ifdef CONFIG_X86_32
@@ -313,42 +358,50 @@ extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
 extern unsigned short num_cache_leaves;
 
 struct thread_struct {
-/* cached TLS descriptors. */
-	struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
-	unsigned long	sp0;
-	unsigned long	sp;
+	/* Cached TLS descriptors: */
+	struct desc_struct	tls_array[GDT_ENTRY_TLS_ENTRIES];
+	unsigned long		sp0;
+	unsigned long		sp;
 #ifdef CONFIG_X86_32
-	unsigned long	sysenter_cs;
+	unsigned long		sysenter_cs;
 #else
-	unsigned long 	usersp;	/* Copy from PDA */
-	unsigned short	es, ds, fsindex, gsindex;
+	unsigned long		usersp;	/* Copy from PDA */
+	unsigned short		es;
+	unsigned short		ds;
+	unsigned short		fsindex;
+	unsigned short		gsindex;
 #endif
-	unsigned long	ip;
-	unsigned long	fs;
-	unsigned long	gs;
-/* Hardware debugging registers */
-	unsigned long	debugreg0;
-	unsigned long	debugreg1;
-	unsigned long	debugreg2;
-	unsigned long	debugreg3;
-	unsigned long	debugreg6;
-	unsigned long	debugreg7;
-/* fault info */
-	unsigned long	cr2, trap_no, error_code;
-/* floating point info */
+	unsigned long		ip;
+	unsigned long		fs;
+	unsigned long		gs;
+	/* Hardware debugging registers: */
+	unsigned long		debugreg0;
+	unsigned long		debugreg1;
+	unsigned long		debugreg2;
+	unsigned long		debugreg3;
+	unsigned long		debugreg6;
+	unsigned long		debugreg7;
+	/* Fault info: */
+	unsigned long		cr2;
+	unsigned long		trap_no;
+	unsigned long		error_code;
+	/* Floating point info: */
 	union i387_union	i387 __attribute__((aligned(16)));;
 #ifdef CONFIG_X86_32
-/* virtual 86 mode info */
+	/* Virtual 86 mode info */
 	struct vm86_struct __user *vm86_info;
 	unsigned long		screen_bitmap;
-	unsigned long		v86flags, v86mask, saved_sp0;
-	unsigned int		saved_fs, saved_gs;
+	unsigned long		v86flags;
+	unsigned long		v86mask;
+	unsigned long		saved_sp0;
+	unsigned int		saved_fs;
+	unsigned int		saved_gs;
 #endif
-/* IO permissions */
-	unsigned long	*io_bitmap_ptr;
-	unsigned long	iopl;
-/* max allowed port in the bitmap, in bytes: */
-	unsigned io_bitmap_max;
+	/* IO permissions: */
+	unsigned long		*io_bitmap_ptr;
+	unsigned long		iopl;
+	/* Max allowed port in the bitmap, in bytes: */
+	unsigned		io_bitmap_max;
 /* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
 	unsigned long	debugctlmsr;
 /* Debug Store - if not 0 points to a DS Save Area configuration;
@@ -358,7 +411,7 @@ struct thread_struct {
 
 static inline unsigned long native_get_debugreg(int regno)
 {
-	unsigned long val = 0; 	/* Damn you, gcc! */
+	unsigned long val = 0;	/* Damn you, gcc! */
 
 	switch (regno) {
 	case 0:
@@ -383,22 +436,22 @@ static inline void native_set_debugreg(int regno, unsigned long value)
 {
 	switch (regno) {
 	case 0:
-		asm("mov %0,%%db0"	: /* no output */ :"r" (value));
+		asm("mov %0, %%db0"	::"r" (value));
 		break;
 	case 1:
-		asm("mov %0,%%db1"	: /* no output */ :"r" (value));
+		asm("mov %0, %%db1"	::"r" (value));
 		break;
 	case 2:
-		asm("mov %0,%%db2"	: /* no output */ :"r" (value));
+		asm("mov %0, %%db2"	::"r" (value));
 		break;
 	case 3:
-		asm("mov %0,%%db3"	: /* no output */ :"r" (value));
+		asm("mov %0, %%db3"	::"r" (value));
 		break;
 	case 6:
-		asm("mov %0,%%db6"	: /* no output */ :"r" (value));
+		asm("mov %0, %%db6"	::"r" (value));
 		break;
 	case 7:
-		asm("mov %0,%%db7"	: /* no output */ :"r" (value));
+		asm("mov %0, %%db7"	::"r" (value));
 		break;
 	default:
 		BUG();
@@ -412,6 +465,7 @@ static inline void native_set_iopl_mask(unsigned mask)
 {
 #ifdef CONFIG_X86_32
 	unsigned int reg;
+
 	__asm__ __volatile__ ("pushfl;"
 			      "popl %0;"
 			      "andl %1, %0;"
@@ -423,12 +477,12 @@ static inline void native_set_iopl_mask(unsigned mask)
 #endif
 }
 
-static inline void native_load_sp0(struct tss_struct *tss,
-				   struct thread_struct *thread)
+static inline void
+native_load_sp0(struct tss_struct *tss, struct thread_struct *thread)
 {
 	tss->x86_tss.sp0 = thread->sp0;
 #ifdef CONFIG_X86_32
-	/* Only happens when SEP is enabled, no need to test "SEP"arately */
+	/* Only happens when SEP is enabled, no need to test "SEP"arately: */
 	if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) {
 		tss->x86_tss.ss1 = thread->sysenter_cs;
 		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
@@ -446,8 +500,8 @@ static inline void native_swapgs(void)
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
-#define __cpuid native_cpuid
-#define paravirt_enabled() 0
+#define __cpuid			native_cpuid
+#define paravirt_enabled()	0
 
 /*
  * These special macros can be used to get or set a debugging register
@@ -457,8 +511,8 @@ static inline void native_swapgs(void)
 #define set_debugreg(value, register)				\
 	native_set_debugreg(register, value)
 
-static inline void load_sp0(struct tss_struct *tss,
-			    struct thread_struct *thread)
+static inline void
+load_sp0(struct tss_struct *tss, struct thread_struct *thread)
 {
 	native_load_sp0(tss, thread);
 }
@@ -473,11 +527,12 @@ static inline void load_sp0(struct tss_struct *tss,
  * enable), so that any CPU's that boot up
  * after us can get the correct flags.
  */
-extern unsigned long mmu_cr4_features;
+extern unsigned long		mmu_cr4_features;
 
 static inline void set_in_cr4(unsigned long mask)
 {
 	unsigned cr4;
+
 	mmu_cr4_features |= mask;
 	cr4 = read_cr4();
 	cr4 |= mask;
@@ -487,6 +542,7 @@ static inline void set_in_cr4(unsigned long mask)
 static inline void clear_in_cr4(unsigned long mask)
 {
 	unsigned cr4;
+
 	mmu_cr4_features &= ~mask;
 	cr4 = read_cr4();
 	cr4 &= ~mask;
@@ -494,42 +550,42 @@ static inline void clear_in_cr4(unsigned long mask)
 }
 
 struct microcode_header {
-	unsigned int hdrver;
-	unsigned int rev;
-	unsigned int date;
-	unsigned int sig;
-	unsigned int cksum;
-	unsigned int ldrver;
-	unsigned int pf;
-	unsigned int datasize;
-	unsigned int totalsize;
-	unsigned int reserved[3];
+	unsigned int		hdrver;
+	unsigned int		rev;
+	unsigned int		date;
+	unsigned int		sig;
+	unsigned int		cksum;
+	unsigned int		ldrver;
+	unsigned int		pf;
+	unsigned int		datasize;
+	unsigned int		totalsize;
+	unsigned int		reserved[3];
 };
 
 struct microcode {
-	struct microcode_header hdr;
-	unsigned int bits[0];
+	struct microcode_header	hdr;
+	unsigned int		bits[0];
 };
 
-typedef struct microcode microcode_t;
-typedef struct microcode_header microcode_header_t;
+typedef struct microcode	microcode_t;
+typedef struct microcode_header	microcode_header_t;
 
 /* microcode format is extended from prescott processors */
 struct extended_signature {
-	unsigned int sig;
-	unsigned int pf;
-	unsigned int cksum;
+	unsigned int		sig;
+	unsigned int		pf;
+	unsigned int		cksum;
 };
 
 struct extended_sigtable {
-	unsigned int count;
-	unsigned int cksum;
-	unsigned int reserved[3];
+	unsigned int		count;
+	unsigned int		cksum;
+	unsigned int		reserved[3];
 	struct extended_signature sigs[0];
 };
 
 typedef struct {
-	unsigned long seg;
+	unsigned long		seg;
 } mm_segment_t;
 
 
@@ -541,7 +597,7 @@ extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
 /* Free all resources held by a thread. */
 extern void release_thread(struct task_struct *);
 
-/* Prepare to copy thread state - unlazy all lazy status */
+/* Prepare to copy thread state - unlazy all lazy state */
 extern void prepare_to_copy(struct task_struct *tsk);
 
 unsigned long get_wchan(struct task_struct *p);
@@ -578,118 +634,131 @@ static inline unsigned int cpuid_eax(unsigned int op)
 	unsigned int eax, ebx, ecx, edx;
 
 	cpuid(op, &eax, &ebx, &ecx, &edx);
+
 	return eax;
 }
+
 static inline unsigned int cpuid_ebx(unsigned int op)
 {
 	unsigned int eax, ebx, ecx, edx;
 
 	cpuid(op, &eax, &ebx, &ecx, &edx);
+
 	return ebx;
 }
+
 static inline unsigned int cpuid_ecx(unsigned int op)
 {
 	unsigned int eax, ebx, ecx, edx;
 
 	cpuid(op, &eax, &ebx, &ecx, &edx);
+
 	return ecx;
 }
+
 static inline unsigned int cpuid_edx(unsigned int op)
 {
 	unsigned int eax, ebx, ecx, edx;
 
 	cpuid(op, &eax, &ebx, &ecx, &edx);
+
 	return edx;
 }
 
 /* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
 static inline void rep_nop(void)
 {
-	__asm__ __volatile__("rep;nop": : :"memory");
+	__asm__ __volatile__("rep; nop" ::: "memory");
 }
 
-/* Stop speculative execution */
+static inline void cpu_relax(void)
+{
+	rep_nop();
+}
+
+/* Stop speculative execution: */
 static inline void sync_core(void)
 {
 	int tmp;
+
 	asm volatile("cpuid" : "=a" (tmp) : "0" (1)
 					  : "ebx", "ecx", "edx", "memory");
 }
 
-#define cpu_relax()   rep_nop()
-
-static inline void __monitor(const void *eax, unsigned long ecx,
-		unsigned long edx)
+static inline void
+__monitor(const void *eax, unsigned long ecx, unsigned long edx)
 {
-	/* "monitor %eax,%ecx,%edx;" */
+	/* "monitor %eax, %ecx, %edx;" */
 	asm volatile(
-		".byte 0x0f,0x01,0xc8;"
-		: :"a" (eax), "c" (ecx), "d"(edx));
+		".byte 0x0f, 0x01, 0xc8;"
+		:: "a" (eax), "c" (ecx), "d"(edx));
 }
 
 static inline void __mwait(unsigned long eax, unsigned long ecx)
 {
-	/* "mwait %eax,%ecx;" */
+	/* "mwait %eax, %ecx;" */
 	asm volatile(
-		".byte 0x0f,0x01,0xc9;"
-		: :"a" (eax), "c" (ecx));
+		".byte 0x0f, 0x01, 0xc9;"
+		:: "a" (eax), "c" (ecx));
 }
 
 static inline void __sti_mwait(unsigned long eax, unsigned long ecx)
 {
-	/* "mwait %eax,%ecx;" */
+	/* "mwait %eax, %ecx;" */
 	asm volatile(
-		"sti; .byte 0x0f,0x01,0xc9;"
-		: :"a" (eax), "c" (ecx));
+		"sti; .byte 0x0f, 0x01, 0xc9;"
+		:: "a" (eax), "c" (ecx));
 }
 
 extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx);
 
-extern int force_mwait;
+extern int			force_mwait;
 
 extern void select_idle_routine(const struct cpuinfo_x86 *c);
 
-extern unsigned long boot_option_idle_override;
+extern unsigned long		boot_option_idle_override;
 
 extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
 /* Defined in head.S */
-extern struct desc_ptr early_gdt_descr;
+extern struct desc_ptr		early_gdt_descr;
 
 extern void cpu_set_gdt(int);
 extern void switch_to_new_gdt(void);
 extern void cpu_init(void);
 extern void init_gdt(int cpu);
 
-/* from system description table in BIOS.  Mostly for MCA use, but
- * others may find it useful. */
-extern unsigned int machine_id;
-extern unsigned int machine_submodel_id;
-extern unsigned int BIOS_revision;
+/*
+ * from system description table in BIOS. Mostly for MCA use, but
+ * others may find it useful:
+ */
+extern unsigned int		machine_id;
+extern unsigned int		machine_submodel_id;
+extern unsigned int		BIOS_revision;
 
-/* Boot loader type from the setup header */
-extern int bootloader_type;
+/* Boot loader type from the setup header: */
+extern int			bootloader_type;
 
-extern char ignore_fpu_irq;
-#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
+extern char			ignore_fpu_irq;
 
 #define HAVE_ARCH_PICK_MMAP_LAYOUT 1
 #define ARCH_HAS_PREFETCHW
 #define ARCH_HAS_SPINLOCK_PREFETCH
 
 #ifdef CONFIG_X86_32
-#define BASE_PREFETCH	ASM_NOP4
-#define ARCH_HAS_PREFETCH
+# define BASE_PREFETCH		ASM_NOP4
+# define ARCH_HAS_PREFETCH
 #else
-#define BASE_PREFETCH	"prefetcht0 (%1)"
+# define BASE_PREFETCH		"prefetcht0 (%1)"
 #endif
 
-/* Prefetch instructions for Pentium III and AMD Athlon */
-/* It's not worth to care about 3dnow! prefetches for the K6
-   because they are microcoded there and very slow.
-   However we don't do prefetches for pre XP Athlons currently
-   That should be fixed. */
+/*
+ * Prefetch instructions for Pentium III (+) and AMD Athlon (+)
+ *
+ * It's not worth to care about 3dnow prefetches for the K6
+ * because they are microcoded there and very slow.
+ */
 static inline void prefetch(const void *x)
 {
 	alternative_input(BASE_PREFETCH,
@@ -698,8 +767,11 @@ static inline void prefetch(const void *x)
 			  "r" (x));
 }
 
-/* 3dnow! prefetch to get an exclusive cache line. Useful for
-   spinlocks to avoid one state transition in the cache coherency protocol. */
+/*
+ * 3dnow prefetch to get an exclusive cache line.
+ * Useful for spinlocks to avoid one state transition in the
+ * cache coherency protocol:
+ */
 static inline void prefetchw(const void *x)
 {
 	alternative_input(BASE_PREFETCH,
@@ -708,21 +780,25 @@ static inline void prefetchw(const void *x)
 			  "r" (x));
 }
 
-#define spin_lock_prefetch(x)	prefetchw(x)
+static inline void spin_lock_prefetch(const void *x)
+{
+	prefetchw(x);
+}
+
 #ifdef CONFIG_X86_32
 /*
  * User space process size: 3GB (default).
  */
-#define TASK_SIZE	(PAGE_OFFSET)
-#define STACK_TOP	TASK_SIZE
-#define STACK_TOP_MAX	STACK_TOP
-
-#define INIT_THREAD  {							\
-	.sp0 = sizeof(init_stack) + (long)&init_stack,			\
-	.vm86_info = NULL,						\
-	.sysenter_cs = __KERNEL_CS,					\
-	.io_bitmap_ptr = NULL,						\
-	.fs = __KERNEL_PERCPU,						\
+#define TASK_SIZE		PAGE_OFFSET
+#define STACK_TOP		TASK_SIZE
+#define STACK_TOP_MAX		STACK_TOP
+
+#define INIT_THREAD  {							  \
+	.sp0			= sizeof(init_stack) + (long)&init_stack, \
+	.vm86_info		= NULL,					  \
+	.sysenter_cs		= __KERNEL_CS,				  \
+	.io_bitmap_ptr		= NULL,					  \
+	.fs			= __KERNEL_PERCPU,			  \
 }
 
 /*
@@ -731,26 +807,27 @@ static inline void prefetchw(const void *x)
  * permission bitmap. The extra byte must be all 1 bits, and must
  * be within the limit.
  */
-#define INIT_TSS  {							\
-	.x86_tss = {							\
+#define INIT_TSS  {							  \
+	.x86_tss = {							  \
 		.sp0		= sizeof(init_stack) + (long)&init_stack, \
-		.ss0		= __KERNEL_DS,				\
-		.ss1		= __KERNEL_CS,				\
-		.io_bitmap_base	= INVALID_IO_BITMAP_OFFSET,		\
-	 },								\
-	.io_bitmap	= { [0 ... IO_BITMAP_LONGS] = ~0 },		\
+		.ss0		= __KERNEL_DS,				  \
+		.ss1		= __KERNEL_CS,				  \
+		.io_bitmap_base	= INVALID_IO_BITMAP_OFFSET,		  \
+	 },								  \
+	.io_bitmap		= { [0 ... IO_BITMAP_LONGS] = ~0 },	  \
 }
 
-#define start_thread(regs, new_eip, new_esp) do {		\
+#define start_thread(regs, new_eip, new_esp)			\
+do {								\
 	__asm__("movl %0,%%gs": :"r" (0));			\
-	regs->fs = 0;						\
+	regs->fs		= 0;				\
 	set_fs(USER_DS);					\
-	regs->ds = __USER_DS;					\
-	regs->es = __USER_DS;					\
-	regs->ss = __USER_DS;					\
-	regs->cs = __USER_CS;					\
-	regs->ip = new_eip;					\
-	regs->sp = new_esp;					\
+	regs->ds		= __USER_DS;			\
+	regs->es		= __USER_DS;			\
+	regs->ss		= __USER_DS;			\
+	regs->cs		= __USER_CS;			\
+	regs->ip		= new_eip;			\
+	regs->sp		= new_esp;			\
 } while (0)
 
 
@@ -780,24 +857,24 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
        __regs__ - 1;                                                   \
 })
 
-#define KSTK_ESP(task) (task_pt_regs(task)->sp)
+#define KSTK_ESP(task)		(task_pt_regs(task)->sp)
 
 #else
 /*
  * User space process size. 47bits minus one guard page.
  */
-#define TASK_SIZE64	(0x800000000000UL - 4096)
+#define TASK_SIZE64		(0x800000000000UL - 4096)
 
 /* This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
  */
-#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? \
-			   0xc0000000 : 0xFFFFe000)
+#define IA32_PAGE_OFFSET	((current->personality & ADDR_LIMIT_3GB) ? \
+					0xc0000000 : 0xFFFFe000)
 
-#define TASK_SIZE 		(test_thread_flag(TIF_IA32) ? \
-				 IA32_PAGE_OFFSET : TASK_SIZE64)
-#define TASK_SIZE_OF(child) 	((test_tsk_thread_flag(child, TIF_IA32)) ? \
-				  IA32_PAGE_OFFSET : TASK_SIZE64)
+#define TASK_SIZE		(test_thread_flag(TIF_IA32) ? \
+					IA32_PAGE_OFFSET : TASK_SIZE64)
+#define TASK_SIZE_OF(child)	((test_tsk_thread_flag(child, TIF_IA32)) ? \
+					IA32_PAGE_OFFSET : TASK_SIZE64)
 
 #define STACK_TOP		TASK_SIZE
 #define STACK_TOP_MAX		TASK_SIZE64
@@ -813,12 +890,12 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 #define start_thread(regs, new_rip, new_rsp) do { 			     \
 	asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0));  \
 	load_gs_index(0);						     \
-	(regs)->ip = (new_rip);						     \
-	(regs)->sp = (new_rsp);						     \
+	(regs)->ip		= (new_rip);				     \
+	(regs)->sp		= (new_rsp);				     \
 	write_pda(oldrsp, (new_rsp));					     \
-	(regs)->cs = __USER_CS;						     \
-	(regs)->ss = __USER_DS;						     \
-	(regs)->flags = 0x200;						     \
+	(regs)->cs		= __USER_CS;				     \
+	(regs)->ss		= __USER_DS;				     \
+	(regs)->flags		= 0x200;				     \
 	set_fs(USER_DS);						     \
 } while (0)
 
@@ -826,17 +903,18 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
  * Return saved PC of a blocked thread.
  * What is this good for? it will be always the scheduler or ret_from_fork.
  */
-#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8))
+#define thread_saved_pc(t)	(*(unsigned long *)((t)->thread.sp - 8))
 
-#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1)
-#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
+#define task_pt_regs(tsk)	((struct pt_regs *)(tsk)->thread.sp0 - 1)
+#define KSTK_ESP(tsk)		-1 /* sorry. doesn't work for syscall. */
 #endif /* CONFIG_X86_64 */
 
-/* This decides where the kernel will search for a free chunk of vm
+/*
+ * This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
  */
 #define TASK_UNMAPPED_BASE	(PAGE_ALIGN(TASK_SIZE / 3))
 
-#define KSTK_EIP(task) (task_pt_regs(task)->ip)
+#define KSTK_EIP(task)		(task_pt_regs(task)->ip)
 
 #endif
-- 
cgit v1.2.3


From 513ad84bf60d96a6998bca10ed07c3d340449be8 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 21 Feb 2008 05:18:40 +0100
Subject: x86: de-macro start_thread()

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/process_32.c | 15 +++++++++++++++
 arch/x86/kernel/process_64.c | 15 +++++++++++++++
 include/asm-x86/processor.h  | 29 +++--------------------------
 3 files changed, 33 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 2cd89b8a7050..9230ce060d09 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -512,6 +512,21 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 	return err;
 }
 
+void
+start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
+{
+	__asm__("movl %0, %%gs" :: "r"(0));
+	regs->fs		= 0;
+	set_fs(USER_DS);
+	regs->ds		= __USER_DS;
+	regs->es		= __USER_DS;
+	regs->ss		= __USER_DS;
+	regs->cs		= __USER_CS;
+	regs->ip		= new_ip;
+	regs->sp		= new_sp;
+}
+EXPORT_SYMBOL_GPL(start_thread);
+
 #ifdef CONFIG_SECCOMP
 static void hard_disable_TSC(void)
 {
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 46c4c546b499..1ffce14cff6e 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -528,6 +528,21 @@ out:
 	return err;
 }
 
+void
+start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
+{
+	asm volatile("movl %0, %%fs; movl %0, %%es; movl %0, %%ds" :: "r"(0));
+	load_gs_index(0);
+	regs->ip		= new_ip;
+	regs->sp		= new_sp;
+	write_pda(oldrsp, new_sp);
+	regs->cs		= __USER_CS;
+	regs->ss		= __USER_DS;
+	regs->flags		= 0x200;
+	set_fs(USER_DS);
+}
+EXPORT_SYMBOL_GPL(start_thread);
+
 /*
  * This special macro can be used to load a debugging register
  */
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 43d2cc829a94..9054734589fe 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -817,20 +817,6 @@ static inline void spin_lock_prefetch(const void *x)
 	.io_bitmap		= { [0 ... IO_BITMAP_LONGS] = ~0 },	  \
 }
 
-#define start_thread(regs, new_eip, new_esp)			\
-do {								\
-	__asm__("movl %0,%%gs": :"r" (0));			\
-	regs->fs		= 0;				\
-	set_fs(USER_DS);					\
-	regs->ds		= __USER_DS;			\
-	regs->es		= __USER_DS;			\
-	regs->ss		= __USER_DS;			\
-	regs->cs		= __USER_CS;			\
-	regs->ip		= new_eip;			\
-	regs->sp		= new_esp;			\
-} while (0)
-
-
 extern unsigned long thread_saved_pc(struct task_struct *tsk);
 
 #define THREAD_SIZE_LONGS      (THREAD_SIZE/sizeof(unsigned long))
@@ -887,18 +873,6 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 	.x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
 }
 
-#define start_thread(regs, new_rip, new_rsp) do { 			     \
-	asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0));  \
-	load_gs_index(0);						     \
-	(regs)->ip		= (new_rip);				     \
-	(regs)->sp		= (new_rsp);				     \
-	write_pda(oldrsp, (new_rsp));					     \
-	(regs)->cs		= __USER_CS;				     \
-	(regs)->ss		= __USER_DS;				     \
-	(regs)->flags		= 0x200;				     \
-	set_fs(USER_DS);						     \
-} while (0)
-
 /*
  * Return saved PC of a blocked thread.
  * What is this good for? it will be always the scheduler or ret_from_fork.
@@ -909,6 +883,9 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 #define KSTK_ESP(tsk)		-1 /* sorry. doesn't work for syscall. */
 #endif /* CONFIG_X86_64 */
 
+extern void start_thread(struct pt_regs *regs, unsigned long new_ip,
+					       unsigned long new_sp);
+
 /*
  * This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
-- 
cgit v1.2.3


From 3def3d6ddf43dbe20c00c3cbc38dfacc8586998f Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Fri, 22 Feb 2008 17:07:16 -0800
Subject: x86: clean up e820_reserve_resources on 64-bit

e820_resource_resources could use insert_resource instead of request_resource
also move code_resource, data_resource, bss_resource, and crashk_res
out of e820_reserve_resources.

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/e820_64.c  | 19 ++-----------------
 arch/x86/kernel/setup_64.c |  8 +++++++-
 include/asm-x86/e820_64.h  |  3 +--
 3 files changed, 10 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index a8694a35352b..8b914a833ac6 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -229,8 +229,7 @@ unsigned long __init e820_end_of_ram(void)
 /*
  * Mark e820 reserved areas as busy for the resource manager.
  */
-void __init e820_reserve_resources(struct resource *code_resource,
-		struct resource *data_resource, struct resource *bss_resource)
+void __init e820_reserve_resources(void)
 {
 	int i;
 	for (i = 0; i < e820.nr_map; i++) {
@@ -245,21 +244,7 @@ void __init e820_reserve_resources(struct resource *code_resource,
 		res->start = e820.map[i].addr;
 		res->end = res->start + e820.map[i].size - 1;
 		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
-		request_resource(&iomem_resource, res);
-		if (e820.map[i].type == E820_RAM) {
-			/*
-			 * We don't know which RAM region contains kernel data,
-			 * so we try it repeatedly and let the resource manager
-			 * test it.
-			 */
-			request_resource(res, code_resource);
-			request_resource(res, data_resource);
-			request_resource(res, bss_resource);
-#ifdef CONFIG_KEXEC
-			if (crashk_res.start != crashk_res.end)
-				request_resource(res, &crashk_res);
-#endif
-		}
+		insert_resource(&iomem_resource, res);
 	}
 }
 
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 187f084b9491..e3cb3ea96ca1 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -248,6 +248,7 @@ static void __init reserve_crashkernel(void)
 				(unsigned long)(total_mem >> 20));
 		crashk_res.start = crash_base;
 		crashk_res.end   = crash_base + crash_size - 1;
+		insert_resource(&iomem_resource, &crashk_res);
 	}
 }
 #else
@@ -322,6 +323,11 @@ void __init setup_arch(char **cmdline_p)
 
 	finish_e820_parsing();
 
+	/* after parse_early_param, so could debug it */
+	insert_resource(&iomem_resource, &code_resource);
+	insert_resource(&iomem_resource, &data_resource);
+	insert_resource(&iomem_resource, &bss_resource);
+
 	early_gart_iommu_check();
 
 	e820_register_active_regions(0, 0, -1UL);
@@ -454,7 +460,7 @@ void __init setup_arch(char **cmdline_p)
 	/*
 	 * We trust e820 completely. No explicit ROM probing in memory.
 	 */
-	e820_reserve_resources(&code_resource, &data_resource, &bss_resource);
+	e820_reserve_resources();
 	e820_mark_nosave_regions();
 
 	/* request I/O space for devices used on all i[345]86 PCs */
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index 9e06c6eb4e27..ef653a403e0b 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -23,8 +23,7 @@ extern void update_memory_range(u64 start, u64 size, unsigned old_type,
 extern void setup_memory_region(void);
 extern void contig_e820_setup(void); 
 extern unsigned long e820_end_of_ram(void);
-extern void e820_reserve_resources(struct resource *code_resource,
-		struct resource *data_resource, struct resource *bss_resource);
+extern void e820_reserve_resources(void);
 extern void e820_mark_nosave_regions(void);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
-- 
cgit v1.2.3


From f8fffa458368ed3d57385698f775880db629bd1a Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Sun, 24 Feb 2008 21:36:28 -0800
Subject: x86: apic_is_clustered_box for vsmp

quad core 8 socket system will have apic id lifting.the apic id range could
be [4, 0x23]. and apic_is_clustered_box will think that need to three clusters
and that is larger than 2. So it is treated as a clustered_box.

and will get:

   Marking TSC unstable due to TSCs unsynchronized

even if the CPUs have X86_FEATURE_CONSTANT_TSC set.

this quick fix will check if the cpu is from AMD.

but vsmp still needs that checking...

this patch is fix to make sure that vsmp not to be passed.

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c |  4 ++--
 arch/x86/kernel/vsmp_64.c | 27 +++++++++++++++++++++------
 include/asm-x86/apic.h    |  5 +++++
 3 files changed, 28 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index ac2405ed504d..f6eb01d8923a 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -1182,9 +1182,9 @@ __cpuinit int apic_is_clustered_box(void)
 	 * there is not this kind of box with AMD CPU yet.
 	 * Some AMD box with quadcore cpu and 8 sockets apicid
 	 * will be [4, 0x23] or [8, 0x27] could be thought to
-	 * have three apic_clusters. So go out early.
+	 * vsmp box still need checking...
 	 */
-	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
+	if (!is_vsmp_box() && (boot_cpu_data.x86_vendor == X86_VENDOR_AMD))
 		return 0;
 
 	bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr;
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index 54202b1805da..a00961d42e75 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -72,19 +72,34 @@ static unsigned __init vsmp_patch(u8 type, u16 clobbers, void *ibuf,
 
 }
 
+static int vsmp = -1;
+
+int is_vsmp_box(void)
+{
+	if (vsmp != -1)
+		return vsmp;
+
+	vsmp = 0;
+	if (!early_pci_allowed())
+		return vsmp;
+
+	/* Check if we are running on a ScaleMP vSMP box */
+	if (read_pci_config(0, 0x1f, 0, PCI_VENDOR_ID) ==
+	     (PCI_VENDOR_ID_SCALEMP || (PCI_DEVICE_ID_SCALEMP_VSMP_CTL << 16)))
+		vsmp = 1;
+
+	return vsmp;
+}
+
 void __init vsmp_init(void)
 {
 	void *address;
 	unsigned int cap, ctl, cfg;
 
-	if (!early_pci_allowed())
+	if (!is_vsmp_box())
 		return;
 
-	/* Check if we are running on a ScaleMP vSMP box */
-	if ((read_pci_config_16(0, 0x1f, 0, PCI_VENDOR_ID) !=
-	     PCI_VENDOR_ID_SCALEMP) ||
-	    (read_pci_config_16(0, 0x1f, 0, PCI_DEVICE_ID) !=
-	     PCI_DEVICE_ID_SCALEMP_VSMP_CTL))
+	if (!early_pci_allowed())
 		return;
 
 	/* If we are, use the distinguished irq functions */
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index bcfc07fd3661..f0321a427e16 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -51,12 +51,17 @@ extern unsigned boot_cpu_id;
  */
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
+extern int is_vsmp_box(void);
 #else
 #define apic_write native_apic_write
 #define apic_write_atomic native_apic_write_atomic
 #define apic_read native_apic_read
 #define setup_boot_clock setup_boot_APIC_clock
 #define setup_secondary_clock setup_secondary_APIC_clock
+static int inline is_vsmp_box(void)
+{
+	return 0;
+}
 #endif
 
 static inline void native_apic_write(unsigned long reg, u32 v)
-- 
cgit v1.2.3


From 0f8d2b926d15a68eac9c19edfdcb58a5d80b2960 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Tue, 26 Feb 2008 08:34:21 +0100
Subject: x86: clean up cpu capabilities accesses

introduce test_cpu_cap() for raw access to the real CPU
capabilities as they are present in x86_capability.

(cpu_has() will shortcut certain tests during build-time)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/cpufeature.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 065e92966c7c..1e3102eeb823 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -120,6 +120,9 @@
 extern const char * const x86_cap_flags[NCAPINTS*32];
 extern const char * const x86_power_flags[32];
 
+#define test_cpu_cap(c, bit)						\
+	 test_bit(bit, (unsigned long *)((c)->x86_capability))
+
 #define cpu_has(c, bit)							\
 	(__builtin_constant_p(bit) &&					\
 	 ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0)) ||	\
@@ -131,7 +134,8 @@ extern const char * const x86_power_flags[32];
 	   (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) ||	\
 	   (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) )	\
 	  ? 1 :								\
-	 test_bit(bit, (unsigned long *)((c)->x86_capability)))
+	 test_cpu_cap(c, bit))
+
 #define boot_cpu_has(bit)	cpu_has(&boot_cpu_data, bit)
 
 #define set_cpu_cap(c, bit)	set_bit(bit, (unsigned long *)((c)->x86_capability))
-- 
cgit v1.2.3


From 78a9909aab54123c7c471022389b36972e13b48e Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 17 Apr 2008 17:40:51 +0200
Subject: x86, tracing: add notrace to asm-x86/linkage.h

notrace signals that a function should not be traced. Most of the
time this is used by tracers to annotate code that cannot be
traced - it's in a volatile state (such as in user vdso context
or NMI context) or it's in the tracer internals.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/linkage.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h
index c048353f4b85..64e444f8e85b 100644
--- a/include/asm-x86/linkage.h
+++ b/include/asm-x86/linkage.h
@@ -1,6 +1,9 @@
 #ifndef __ASM_LINKAGE_H
 #define __ASM_LINKAGE_H
 
+#undef notrace
+#define notrace __attribute__((no_instrument_function))
+
 #ifdef CONFIG_X86_64
 #define __ALIGN .p2align 4,,15
 #define __ALIGN_STR ".p2align 4,,15"
-- 
cgit v1.2.3


From 40869cd038a0ecb867a7227aba46806224e4d11d Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 3 Mar 2008 13:55:32 +0100
Subject: x86: redo cded932b75ab0a5f9181e

redo commit cded932b75ab0a5f9181e.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pgtable_32.h | 4 +++-
 include/asm-x86/pgtable_64.h | 6 ++++--
 2 files changed, 7 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 4e6a0fca0b47..997c36c6b4d7 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -90,7 +90,9 @@ extern unsigned long pg0[];
 /* To avoid harmful races, pmd_none(x) should check only the lower when PAE */
 #define pmd_none(x)	(!(unsigned long)pmd_val(x))
 #define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
-#define	pmd_bad(x)	((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
+#define	pmd_bad(x)	((pmd_val(x) \
+			  & ~(PAGE_MASK | _PAGE_USER | _PAGE_PSE | _PAGE_NX)) \
+			 != _KERNPG_TABLE)
 
 
 #define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 6ef09914acbe..0a5081c98ae1 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -153,12 +153,14 @@ static inline unsigned long pgd_bad(pgd_t pgd)
 
 static inline unsigned long pud_bad(pud_t pud)
 {
-	return pud_val(pud) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER);
+	return pud_val(pud) &
+		~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER | _PAGE_PSE | _PAGE_NX);
 }
 
 static inline unsigned long pmd_bad(pmd_t pmd)
 {
-	return pmd_val(pmd) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER);
+	return pmd_val(pmd) &
+		~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER | _PAGE_PSE | _PAGE_NX);
 }
 
 #define pte_none(x)	(!pte_val(x))
-- 
cgit v1.2.3


From 9fc34113f6880b215cbea4e7017fc818700384c2 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 3 Mar 2008 09:53:17 +0100
Subject: x86: debug pmd_bad()

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/pgtable_32.c     | 7 +++++++
 include/asm-x86/pgtable_32.h | 6 +++++-
 2 files changed, 12 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index 2f9e9afcb9f4..76e4f4d26272 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -381,3 +381,10 @@ void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd)
 }
 
 #endif
+
+int pmd_bad(pmd_t pmd)
+{
+	WARN_ON_ONCE(pmd_bad_v1(pmd) != pmd_bad_v2(pmd));
+
+	return pmd_bad_v1(pmd);
+}
diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 997c36c6b4d7..1e2c0d839528 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -90,7 +90,11 @@ extern unsigned long pg0[];
 /* To avoid harmful races, pmd_none(x) should check only the lower when PAE */
 #define pmd_none(x)	(!(unsigned long)pmd_val(x))
 #define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
-#define	pmd_bad(x)	((pmd_val(x) \
+
+extern int pmd_bad(pmd_t pmd);
+
+#define pmd_bad_v1(x)	((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
+#define	pmd_bad_v2(x)	((pmd_val(x) \
 			  & ~(PAGE_MASK | _PAGE_USER | _PAGE_PSE | _PAGE_NX)) \
 			 != _KERNPG_TABLE)
 
-- 
cgit v1.2.3


From 0d7a1819e97ef89be5bcbb4b724acb9f6c873c97 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Mon, 3 Mar 2008 12:49:09 +0100
Subject: x86: wmb() confusion in system.h

Comment says wmb is a nop, but it is implemented as lock addl
below... Should it be compiled to nop if we know we are running on
"good" Intel cpu?

At least remove confusing comment for now.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/system.h | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 9cff02ffe6c2..428d9471497f 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -296,16 +296,7 @@ void default_idle(void);
  */
 #ifdef CONFIG_X86_32
 /*
- * For now, "wmb()" doesn't actually do anything, as all
- * Intel CPU's follow what Intel calls a *Processor Order*,
- * in which all writes are seen in the program order even
- * outside the CPU.
- *
- * I expect future Intel CPU's to have a weaker ordering,
- * but I'd also expect them to finally get their act together
- * and add some real memory barriers if so.
- *
- * Some non intel clones support out of order store. wmb() ceases to be a
+ * Some non-Intel clones support out of order store. wmb() ceases to be a
  * nop for these.
  */
 #define mb() alternative("lock; addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
-- 
cgit v1.2.3


From 23b55bd9f33a1812a664e548803db34c9bec56e8 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 5 Mar 2008 10:24:37 +0100
Subject: x86: clean up switch_to()

Make the code more readable and more hackable:

 - use symbolic asm parameters
 - use readable indentation
 - add comments that explains the details

No code changed:

kernel/sched.o:

   text	   data	    bss	    dec	    hex	filename
  28626	    684	   2640	  31950	   7cce	sched.o.before
  28626	    684	   2640	  31950	   7cce	sched.o.after

md5:
   2823d406c18b781975cdb2e7cfea0059  sched.o.before.asm
   2823d406c18b781975cdb2e7cfea0059  sched.o.after.asm

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/system.h | 46 +++++++++++++++++++++++++++++++---------------
 1 file changed, 31 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 428d9471497f..299ae9605cb1 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -27,22 +27,38 @@ struct task_struct *__switch_to(struct task_struct *prev,
  * Saving eflags is important. It switches not only IOPL between tasks,
  * it also protects other tasks from NT leaking through sysenter etc.
  */
-#define switch_to(prev, next, last) do {				\
+#define switch_to(prev, next, last)					\
+do {									\
 	unsigned long esi, edi;						\
-	asm volatile("pushfl\n\t"		/* Save flags */	\
-		     "pushl %%ebp\n\t"					\
-		     "movl %%esp,%0\n\t"	/* save ESP */		\
-		     "movl %5,%%esp\n\t"	/* restore ESP */	\
-		     "movl $1f,%1\n\t"		/* save EIP */		\
-		     "pushl %6\n\t"		/* restore EIP */	\
-		     "jmp __switch_to\n"				\
-		     "1:\t"						\
-		     "popl %%ebp\n\t"					\
-		     "popfl"						\
-		     :"=m" (prev->thread.sp), "=m" (prev->thread.ip),	\
-		      "=a" (last), "=S" (esi), "=D" (edi)		\
-		     :"m" (next->thread.sp), "m" (next->thread.ip),	\
-		      "2" (prev), "d" (next));				\
+									\
+	asm volatile(							\
+		"pushfl			\n\t"	/* save    flags */	\
+		"pushl %%ebp		\n\t"	/* save    EBP   */	\
+		"movl %%esp,%[prev_sp]	\n\t"	/* save    ESP   */	\
+		"movl %[next_sp],%%esp	\n\t"	/* restore ESP   */	\
+		"movl $1f,%[prev_ip]	\n\t"	/* save    EIP   */	\
+		"pushl %[next_ip]	\n\t"	/* restore EIP   */	\
+		"jmp __switch_to	\n"	/* regparm call  */	\
+		"1:			\t"				\
+		"popl %%ebp		\n\t"	/* restore EBP   */	\
+		"popfl			\n"	/* restore flags */	\
+									\
+		  /* output parameters */				\
+		: [prev_sp] "=m" (prev->thread.sp),			\
+		  [prev_ip] "=m" (prev->thread.ip),			\
+			    "=a" (last),				\
+									\
+		  /* clobbered output registers: */			\
+		  "=S" (esi), "=D" (edi)				\
+									\
+		  /* input parameters: */				\
+		: [next_sp]  "m" (next->thread.sp),			\
+		  [next_ip]  "m" (next->thread.ip),			\
+									\
+		  /* regparm parameters for __switch_to(): */		\
+		  [prev]     "a" (prev),				\
+		  [next]     "d" (next)					\
+	);								\
 } while (0)
 
 /*
-- 
cgit v1.2.3


From 8b6451fe5cf78909f28d3762f77df060c8603cd0 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 5 Mar 2008 10:46:38 +0100
Subject: x86: fix switch_to() clobbers

Liu Pingfan noticed that switch_to() clobbers more registers than its
asm constraints specify.

We get away with this due to luck mostly - schedule()
by its nature only has 'local' state which gets reloaded
automatically. Fix it nevertheless, we could hit this anytime.

it turns out that with the extra constraints gcc manages to make
schedule() even more compact:

   text	   data	    bss	    dec	    hex	filename
  28626	    684	   2640	  31950	   7cce	sched.o.before
  28613	    684	   2640	  31937	   7cc1	sched.o.after

Reported-by: Liu Pingfan <kernelfans@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/system.h | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 299ae9605cb1..33b0017156a7 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -29,7 +29,14 @@ struct task_struct *__switch_to(struct task_struct *prev,
  */
 #define switch_to(prev, next, last)					\
 do {									\
-	unsigned long esi, edi;						\
+	/*								\
+	 * Context-switching clobbers all registers, so we clobber	\
+	 * them explicitly, via unused output variables.		\
+	 * (EAX and EBP is not listed because EBP is saved/restored	\
+	 * explicitly for wchan access and EAX is the return value of	\
+	 * __switch_to())						\
+	 */								\
+	unsigned long ebx, ecx, edx, esi, edi;				\
 									\
 	asm volatile(							\
 		"pushfl			\n\t"	/* save    flags */	\
@@ -49,6 +56,7 @@ do {									\
 			    "=a" (last),				\
 									\
 		  /* clobbered output registers: */			\
+		  "=b" (ebx), "=c" (ecx), "=d" (edx),			\
 		  "=S" (esi), "=D" (edi)				\
 									\
 		  /* input parameters: */				\
-- 
cgit v1.2.3


From c27cfeffad436816ecd500b8dc94acf348182b13 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:29 -0300
Subject: x86: commonize smp.h

this is the first step of integrating smp.h between x86_64
and i386

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index f2e8319a6b0b..f250d1c3f8a0 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -1,5 +1,12 @@
+#ifndef _ASM_X86_SMP_H_
+#define _ASM_X86_SMP_H_
+#ifndef __ASSEMBLY__
+
 #ifdef CONFIG_X86_32
 # include "smp_32.h"
 #else
 # include "smp_64.h"
 #endif
+
+#endif /* __ASSEMBLY__ */
+#endif
-- 
cgit v1.2.3


From 639acb16e6b93342a786c01425cf8eb8ebbb1351 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:30 -0300
Subject: x86: merge extern function definitions

move extern function definitions that are the same between smp_{32,64}.h
to smp.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 3 +++
 include/asm-x86/smp_32.h | 4 ----
 include/asm-x86/smp_64.h | 4 ----
 3 files changed, 3 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index f250d1c3f8a0..ad7b99dda0dc 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -8,5 +8,8 @@
 # include "smp_64.h"
 #endif
 
+extern void smp_alloc_memory(void);
+extern void lock_ipi_call_lock(void);
+extern void unlock_ipi_call_lock(void);
 #endif /* __ASSEMBLY__ */
 #endif
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 56152e312287..27812258ac6d 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -22,10 +22,6 @@ extern cpumask_t cpu_callin_map;
 extern int smp_num_siblings;
 extern unsigned int num_processors;
 
-extern void smp_alloc_memory(void);
-extern void lock_ipi_call_lock(void);
-extern void unlock_ipi_call_lock(void);
-
 extern void (*mtrr_hook) (void);
 extern void zap_low_mappings (void);
 
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index e0a75519ad21..2c21df289da4 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -19,10 +19,6 @@ extern cpumask_t cpu_initialized;
 extern int smp_num_siblings;
 extern unsigned int num_processors;
 
-extern void smp_alloc_memory(void);
-extern void lock_ipi_call_lock(void);
-extern void unlock_ipi_call_lock(void);
-
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
-- 
cgit v1.2.3


From 53ebef4961c7d5347b4fa2b878258ccd11fc9663 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:31 -0300
Subject: x86: merge extern variables definitions

move extern definitions that are the same between smp_{32,64}.h
to smp.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 6 ++++++
 include/asm-x86/smp_32.h | 4 ----
 include/asm-x86/smp_64.h | 4 ----
 3 files changed, 6 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index ad7b99dda0dc..c130a87c956d 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -1,6 +1,12 @@
 #ifndef _ASM_X86_SMP_H_
 #define _ASM_X86_SMP_H_
 #ifndef __ASSEMBLY__
+#include <linux/cpumask.h>
+
+extern cpumask_t cpu_callout_map;
+
+extern int smp_num_siblings;
+extern unsigned int num_processors;
 
 #ifdef CONFIG_X86_32
 # include "smp_32.h"
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 27812258ac6d..9a4057d94367 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -16,12 +16,8 @@
 # endif
 #endif
 
-extern cpumask_t cpu_callout_map;
 extern cpumask_t cpu_callin_map;
 
-extern int smp_num_siblings;
-extern unsigned int num_processors;
-
 extern void (*mtrr_hook) (void);
 extern void zap_low_mappings (void);
 
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 2c21df289da4..284f701f2a8b 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -13,12 +13,8 @@
 #include <asm/pda.h>
 #include <asm/thread_info.h>
 
-extern cpumask_t cpu_callout_map;
 extern cpumask_t cpu_initialized;
 
-extern int smp_num_siblings;
-extern unsigned int num_processors;
-
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
-- 
cgit v1.2.3


From 16694024d6d6fa84dfcf5400b53afe1e75cebf0d Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:32 -0300
Subject: x86: define smp_ops in common header

x86_64 will benefit from it
Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 14 ++++++++++++++
 include/asm-x86/smp_32.h | 14 --------------
 2 files changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index c130a87c956d..d11b92b56353 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -8,6 +8,20 @@ extern cpumask_t cpu_callout_map;
 extern int smp_num_siblings;
 extern unsigned int num_processors;
 
+struct smp_ops {
+	void (*smp_prepare_boot_cpu)(void);
+	void (*smp_prepare_cpus)(unsigned max_cpus);
+	int (*cpu_up)(unsigned cpu);
+	void (*smp_cpus_done)(unsigned max_cpus);
+
+	void (*smp_send_stop)(void);
+	void (*smp_send_reschedule)(int cpu);
+	int (*smp_call_function_mask)(cpumask_t mask,
+				      void (*func)(void *info), void *info,
+				      int wait);
+};
+
+
 #ifdef CONFIG_X86_32
 # include "smp_32.h"
 #else
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 9a4057d94367..72faad6509c7 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -38,20 +38,6 @@ extern void remove_siblinginfo(int cpu);
 /* Globals due to paravirt */
 extern void set_cpu_sibling_map(int cpu);
 
-struct smp_ops
-{
-	void (*smp_prepare_boot_cpu)(void);
-	void (*smp_prepare_cpus)(unsigned max_cpus);
-	int (*cpu_up)(unsigned cpu);
-	void (*smp_cpus_done)(unsigned max_cpus);
-
-	void (*smp_send_stop)(void);
-	void (*smp_send_reschedule)(int cpu);
-	int (*smp_call_function_mask)(cpumask_t mask,
-				      void (*func)(void *info), void *info,
-				      int wait);
-};
-
 #ifdef CONFIG_SMP
 extern struct smp_ops smp_ops;
 
-- 
cgit v1.2.3


From c76cb36846da6d5d6fb2951968869faa4fd1001d Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:33 -0300
Subject: x86: move smp_ops extern declaration to common header

the smp_ops symbol is temporarily defined in smp_64.c, but it will soon
be unified

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c | 2 ++
 include/asm-x86/smp.h    | 3 +++
 include/asm-x86/smp_32.h | 2 --
 3 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 2fd74b06db67..80dba12b56af 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -528,3 +528,5 @@ asmlinkage void smp_call_function_interrupt(void)
 	}
 }
 
+struct smp_ops smp_ops;
+EXPORT_SYMBOL_GPL(smp_ops);
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index d11b92b56353..ee98beeb7511 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -21,6 +21,9 @@ struct smp_ops {
 				      int wait);
 };
 
+#ifdef CONFIG_SMP
+extern struct smp_ops smp_ops;
+#endif
 
 #ifdef CONFIG_X86_32
 # include "smp_32.h"
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 72faad6509c7..74755e8ffc77 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -39,8 +39,6 @@ extern void remove_siblinginfo(int cpu);
 extern void set_cpu_sibling_map(int cpu);
 
 #ifdef CONFIG_SMP
-extern struct smp_ops smp_ops;
-
 static inline void smp_prepare_boot_cpu(void)
 {
 	smp_ops.smp_prepare_boot_cpu();
-- 
cgit v1.2.3


From 8678969e60d80527d96d2af0011e72c87c9c1fe5 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:34 -0300
Subject: x86: merge smp_send_reschedule

function definition is moved to common header, x86_64 version is now called
native_smp_send_reschedule

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c | 7 +++++--
 include/asm-x86/smp.h    | 5 +++++
 include/asm-x86/smp_32.h | 4 ----
 include/asm-x86/smp_64.h | 2 --
 4 files changed, 10 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 80dba12b56af..fd1816123496 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -290,8 +290,9 @@ void flush_tlb_all(void)
  * anything. Worst case is that we lose a reschedule ...
  */
 
-void smp_send_reschedule(int cpu)
+static void native_smp_send_reschedule(int cpu)
 {
+	WARN_ON(cpu_is_offline(cpu));
 	send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
 }
 
@@ -528,5 +529,7 @@ asmlinkage void smp_call_function_interrupt(void)
 	}
 }
 
-struct smp_ops smp_ops;
+struct smp_ops smp_ops = {
+	.smp_send_reschedule = native_smp_send_reschedule,
+};
 EXPORT_SYMBOL_GPL(smp_ops);
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index ee98beeb7511..28f33c03d793 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -23,6 +23,11 @@ struct smp_ops {
 
 #ifdef CONFIG_SMP
 extern struct smp_ops smp_ops;
+
+static inline void smp_send_reschedule(int cpu)
+{
+	smp_ops.smp_send_reschedule(cpu);
+}
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 74755e8ffc77..c60a3dd3e802 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -60,10 +60,6 @@ static inline void smp_send_stop(void)
 {
 	smp_ops.smp_send_stop();
 }
-static inline void smp_send_reschedule(int cpu)
-{
-	smp_ops.smp_send_reschedule(cpu);
-}
 static inline int smp_call_function_mask(cpumask_t mask,
 					 void (*func) (void *info), void *info,
 					 int wait)
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 284f701f2a8b..b9204584aa06 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -65,8 +65,6 @@ static inline int num_booting_cpus(void)
 	return cpus_weight(cpu_callout_map);
 }
 
-extern void smp_send_reschedule(int cpu);
-
 #else /* CONFIG_SMP */
 
 extern unsigned int boot_cpu_id;
-- 
cgit v1.2.3


From 64b1a21e0924dca7ea3b7cf4287fa719c8ba7fc5 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:35 -0300
Subject: x86: unify smp_call_function_mask

definition is moved to common header, x86_64 function name
now is native_smp_call_function_mask

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c | 7 ++++---
 include/asm-x86/smp.h    | 7 +++++++
 include/asm-x86/smp_32.h | 6 ------
 3 files changed, 11 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index fd1816123496..225b765db5a2 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -386,9 +386,9 @@ static int __smp_call_function_mask(cpumask_t mask,
  * You must not call this function with disabled interrupts or from a
  * hardware interrupt handler or from a bottom half handler.
  */
-int smp_call_function_mask(cpumask_t mask,
-			   void (*func)(void *), void *info,
-			   int wait)
+int native_smp_call_function_mask(cpumask_t mask,
+				  void (*func)(void *), void *info,
+				  int wait)
 {
 	int ret;
 
@@ -531,5 +531,6 @@ asmlinkage void smp_call_function_interrupt(void)
 
 struct smp_ops smp_ops = {
 	.smp_send_reschedule = native_smp_send_reschedule,
+	.smp_call_function_mask = native_smp_call_function_mask,
 };
 EXPORT_SYMBOL_GPL(smp_ops);
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 28f33c03d793..d9782f4f469e 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -28,6 +28,13 @@ static inline void smp_send_reschedule(int cpu)
 {
 	smp_ops.smp_send_reschedule(cpu);
 }
+
+static inline int smp_call_function_mask(cpumask_t mask,
+					 void (*func) (void *info), void *info,
+					 int wait)
+{
+	return smp_ops.smp_call_function_mask(mask, func, info, wait);
+}
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index c60a3dd3e802..d9337ee8c2fc 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -60,12 +60,6 @@ static inline void smp_send_stop(void)
 {
 	smp_ops.smp_send_stop();
 }
-static inline int smp_call_function_mask(cpumask_t mask,
-					 void (*func) (void *info), void *info,
-					 int wait)
-{
-	return smp_ops.smp_call_function_mask(mask, func, info, wait);
-}
 
 void native_smp_prepare_boot_cpu(void);
 void native_smp_prepare_cpus(unsigned int max_cpus);
-- 
cgit v1.2.3


From 71d195492a6e0b22135a7156af1b41c0f99a116b Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:36 -0300
Subject: x86: unify __cpu_up.

function definition is moved to common header. x86_64 version
is now called native_cpu_up

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c     | 1 +
 arch/x86/kernel/smpboot_64.c | 2 +-
 include/asm-x86/smp.h        | 7 +++++++
 include/asm-x86/smp_32.h     | 5 -----
 4 files changed, 9 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 225b765db5a2..7cc20a3c6c19 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -532,5 +532,6 @@ asmlinkage void smp_call_function_interrupt(void)
 struct smp_ops smp_ops = {
 	.smp_send_reschedule = native_smp_send_reschedule,
 	.smp_call_function_mask = native_smp_call_function_mask,
+	.cpu_up = native_cpu_up,
 };
 EXPORT_SYMBOL_GPL(smp_ops);
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 0880f2c388a9..e381fe7792c4 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -929,7 +929,7 @@ void __init smp_prepare_boot_cpu(void)
 /*
  * Entry point to boot a CPU.
  */
-int __cpuinit __cpu_up(unsigned int cpu)
+int __cpuinit native_cpu_up(unsigned int cpu)
 {
 	int apicid = cpu_present_to_apicid(cpu);
 	unsigned long flags;
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index d9782f4f469e..7dd71410fe7d 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -24,6 +24,11 @@ struct smp_ops {
 #ifdef CONFIG_SMP
 extern struct smp_ops smp_ops;
 
+static inline int __cpu_up(unsigned int cpu)
+{
+	return smp_ops.cpu_up(cpu);
+}
+
 static inline void smp_send_reschedule(int cpu)
 {
 	smp_ops.smp_send_reschedule(cpu);
@@ -35,6 +40,8 @@ static inline int smp_call_function_mask(cpumask_t mask,
 {
 	return smp_ops.smp_call_function_mask(mask, func, info, wait);
 }
+
+int native_cpu_up(unsigned int cpunum);
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index d9337ee8c2fc..a7fab8e1517a 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -47,10 +47,6 @@ static inline void smp_prepare_cpus(unsigned int max_cpus)
 {
 	smp_ops.smp_prepare_cpus(max_cpus);
 }
-static inline int __cpu_up(unsigned int cpu)
-{
-	return smp_ops.cpu_up(cpu);
-}
 static inline void smp_cpus_done(unsigned int max_cpus)
 {
 	smp_ops.smp_cpus_done(max_cpus);
@@ -63,7 +59,6 @@ static inline void smp_send_stop(void)
 
 void native_smp_prepare_boot_cpu(void);
 void native_smp_prepare_cpus(unsigned int max_cpus);
-int native_cpu_up(unsigned int cpunum);
 void native_smp_cpus_done(unsigned int max_cpus);
 
 #ifndef CONFIG_PARAVIRT
-- 
cgit v1.2.3


From 1e3fac83da056f26bcb96e13967c157de55bf2ef Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:37 -0300
Subject: x86: unify prepare_boot_cpu

definition is moved to common header. x86_64 version is now called
native_prepare_boot_cpu

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c     | 1 +
 arch/x86/kernel/smpboot_64.c | 2 +-
 include/asm-x86/smp.h        | 6 ++++++
 include/asm-x86/smp_32.h     | 5 -----
 4 files changed, 8 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 7cc20a3c6c19..05116c1ddb6f 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -530,6 +530,7 @@ asmlinkage void smp_call_function_interrupt(void)
 }
 
 struct smp_ops smp_ops = {
+	.smp_prepare_boot_cpu = native_smp_prepare_boot_cpu,
 	.smp_send_reschedule = native_smp_send_reschedule,
 	.smp_call_function_mask = native_smp_call_function_mask,
 	.cpu_up = native_cpu_up,
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index e381fe7792c4..47e654cdc92d 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -918,7 +918,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 /*
  * Early setup to make printk work.
  */
-void __init smp_prepare_boot_cpu(void)
+void __init native_smp_prepare_boot_cpu(void)
 {
 	int me = smp_processor_id();
 	/* already set me in cpu_online_map in boot_cpu_init() */
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 7dd71410fe7d..a2d69a1bec24 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -24,6 +24,11 @@ struct smp_ops {
 #ifdef CONFIG_SMP
 extern struct smp_ops smp_ops;
 
+static inline void smp_prepare_boot_cpu(void)
+{
+	smp_ops.smp_prepare_boot_cpu();
+}
+
 static inline int __cpu_up(unsigned int cpu)
 {
 	return smp_ops.cpu_up(cpu);
@@ -41,6 +46,7 @@ static inline int smp_call_function_mask(cpumask_t mask,
 	return smp_ops.smp_call_function_mask(mask, func, info, wait);
 }
 
+void native_smp_prepare_boot_cpu(void);
 int native_cpu_up(unsigned int cpunum);
 #endif
 
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index a7fab8e1517a..2c7ecfaa8234 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -39,10 +39,6 @@ extern void remove_siblinginfo(int cpu);
 extern void set_cpu_sibling_map(int cpu);
 
 #ifdef CONFIG_SMP
-static inline void smp_prepare_boot_cpu(void)
-{
-	smp_ops.smp_prepare_boot_cpu();
-}
 static inline void smp_prepare_cpus(unsigned int max_cpus)
 {
 	smp_ops.smp_prepare_cpus(max_cpus);
@@ -57,7 +53,6 @@ static inline void smp_send_stop(void)
 	smp_ops.smp_send_stop();
 }
 
-void native_smp_prepare_boot_cpu(void);
 void native_smp_prepare_cpus(unsigned int max_cpus);
 void native_smp_cpus_done(unsigned int max_cpus);
 
-- 
cgit v1.2.3


From 7557da67208f6ed3a1073594b7597bf20c9eb63a Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:38 -0300
Subject: x86: unify smp_prepare_cpus

definition is moved to common header. x86_64 version is now called
native_smp_prepare_cpus

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c     | 1 +
 arch/x86/kernel/smpboot_64.c | 2 +-
 include/asm-x86/smp.h        | 6 ++++++
 include/asm-x86/smp_32.h     | 5 -----
 4 files changed, 8 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 05116c1ddb6f..c520374be1a8 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -531,6 +531,7 @@ asmlinkage void smp_call_function_interrupt(void)
 
 struct smp_ops smp_ops = {
 	.smp_prepare_boot_cpu = native_smp_prepare_boot_cpu,
+	.smp_prepare_cpus = native_smp_prepare_cpus,
 	.smp_send_reschedule = native_smp_send_reschedule,
 	.smp_call_function_mask = native_smp_call_function_mask,
 	.cpu_up = native_cpu_up,
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 47e654cdc92d..b106983050b0 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -867,7 +867,7 @@ static void __init smp_cpu_index_default(void)
  * Prepare for SMP bootup.  The MP table or ACPI has been read
  * earlier.  Just do some sanity checking here and enable APIC mode.
  */
-void __init smp_prepare_cpus(unsigned int max_cpus)
+void __init native_smp_prepare_cpus(unsigned int max_cpus)
 {
 	nmi_watchdog_default();
 	smp_cpu_index_default();
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index a2d69a1bec24..31bd99ddd8c2 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -29,6 +29,11 @@ static inline void smp_prepare_boot_cpu(void)
 	smp_ops.smp_prepare_boot_cpu();
 }
 
+static inline void smp_prepare_cpus(unsigned int max_cpus)
+{
+	smp_ops.smp_prepare_cpus(max_cpus);
+}
+
 static inline int __cpu_up(unsigned int cpu)
 {
 	return smp_ops.cpu_up(cpu);
@@ -47,6 +52,7 @@ static inline int smp_call_function_mask(cpumask_t mask,
 }
 
 void native_smp_prepare_boot_cpu(void);
+void native_smp_prepare_cpus(unsigned int max_cpus);
 int native_cpu_up(unsigned int cpunum);
 #endif
 
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 2c7ecfaa8234..50785389680b 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -39,10 +39,6 @@ extern void remove_siblinginfo(int cpu);
 extern void set_cpu_sibling_map(int cpu);
 
 #ifdef CONFIG_SMP
-static inline void smp_prepare_cpus(unsigned int max_cpus)
-{
-	smp_ops.smp_prepare_cpus(max_cpus);
-}
 static inline void smp_cpus_done(unsigned int max_cpus)
 {
 	smp_ops.smp_cpus_done(max_cpus);
@@ -53,7 +49,6 @@ static inline void smp_send_stop(void)
 	smp_ops.smp_send_stop();
 }
 
-void native_smp_prepare_cpus(unsigned int max_cpus);
 void native_smp_cpus_done(unsigned int max_cpus);
 
 #ifndef CONFIG_PARAVIRT
-- 
cgit v1.2.3


From c559764923dacef301116a248695856e6eb96e48 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:39 -0300
Subject: x86: unify smp_cpus_done

definition is moved to common header. x86_64 version is now called
native_smp_cpus_done

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c     | 2 ++
 arch/x86/kernel/smpboot_64.c | 2 +-
 include/asm-x86/smp.h        | 6 ++++++
 include/asm-x86/smp_32.h     | 7 -------
 4 files changed, 9 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index c520374be1a8..275101ab4b6d 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -532,6 +532,8 @@ asmlinkage void smp_call_function_interrupt(void)
 struct smp_ops smp_ops = {
 	.smp_prepare_boot_cpu = native_smp_prepare_boot_cpu,
 	.smp_prepare_cpus = native_smp_prepare_cpus,
+	.smp_cpus_done = native_smp_cpus_done,
+
 	.smp_send_reschedule = native_smp_send_reschedule,
 	.smp_call_function_mask = native_smp_call_function_mask,
 	.cpu_up = native_cpu_up,
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index b106983050b0..fd0d3a93b995 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -987,7 +987,7 @@ int __cpuinit native_cpu_up(unsigned int cpu)
 /*
  * Finish the SMP boot.
  */
-void __init smp_cpus_done(unsigned int max_cpus)
+void __init native_smp_cpus_done(unsigned int max_cpus)
 {
 	smp_cleanup_boot();
 	setup_ioapic_dest();
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 31bd99ddd8c2..9620165d3b78 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -34,6 +34,11 @@ static inline void smp_prepare_cpus(unsigned int max_cpus)
 	smp_ops.smp_prepare_cpus(max_cpus);
 }
 
+static inline void smp_cpus_done(unsigned int max_cpus)
+{
+	smp_ops.smp_cpus_done(max_cpus);
+}
+
 static inline int __cpu_up(unsigned int cpu)
 {
 	return smp_ops.cpu_up(cpu);
@@ -53,6 +58,7 @@ static inline int smp_call_function_mask(cpumask_t mask,
 
 void native_smp_prepare_boot_cpu(void);
 void native_smp_prepare_cpus(unsigned int max_cpus);
+void native_smp_cpus_done(unsigned int max_cpus);
 int native_cpu_up(unsigned int cpunum);
 #endif
 
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 50785389680b..bc90a4ed3235 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -39,18 +39,11 @@ extern void remove_siblinginfo(int cpu);
 extern void set_cpu_sibling_map(int cpu);
 
 #ifdef CONFIG_SMP
-static inline void smp_cpus_done(unsigned int max_cpus)
-{
-	smp_ops.smp_cpus_done(max_cpus);
-}
-
 static inline void smp_send_stop(void)
 {
 	smp_ops.smp_send_stop();
 }
 
-void native_smp_cpus_done(unsigned int max_cpus);
-
 #ifndef CONFIG_PARAVIRT
 #define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
 #endif
-- 
cgit v1.2.3


From 93b016f8f393c1f8c27e8c4df06ad1420fac65f5 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:40 -0300
Subject: x86: move disabled_cpus to common header

disabled_cpus is (up to now) a x86_64-only contruction.
But it's extern declaration can be moved to common header anyway

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 3 +++
 include/asm-x86/smp_64.h | 1 -
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 9620165d3b78..be8a511fdb1d 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -2,6 +2,7 @@
 #define _ASM_X86_SMP_H_
 #ifndef __ASSEMBLY__
 #include <linux/cpumask.h>
+#include <linux/init.h>
 
 extern cpumask_t cpu_callout_map;
 
@@ -60,6 +61,8 @@ void native_smp_prepare_boot_cpu(void);
 void native_smp_prepare_cpus(unsigned int max_cpus);
 void native_smp_cpus_done(unsigned int max_cpus);
 int native_cpu_up(unsigned int cpunum);
+
+extern unsigned disabled_cpus;
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index b9204584aa06..c7a00caa6ec6 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -44,7 +44,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
 extern void prefill_possible_map(void);
-extern unsigned __cpuinitdata disabled_cpus;
 
 #define raw_smp_processor_id()	read_pda(cpunumber)
 #define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
-- 
cgit v1.2.3


From 68a1c3f8cd893f5c3c1396fec5be7d8acac4fc93 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:42 -0300
Subject: x86: move prefill_possible_map to common file

this patches moves prefill_possible_map() to smpboot.c
Right now it is x86_64-specific, but nothing intrinsically
prevents it to be used by i386

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/Makefile     |  2 +-
 arch/x86/kernel/smpboot.c    | 53 ++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/smpboot_64.c | 51 ------------------------------------------
 include/asm-x86/smp.h        |  1 +
 include/asm-x86/smp_64.h     |  1 -
 5 files changed, 55 insertions(+), 53 deletions(-)
 create mode 100644 arch/x86/kernel/smpboot.c

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index df10327182d4..4c68bfc6df1d 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -46,7 +46,7 @@ obj-$(CONFIG_MICROCODE)		+= microcode.o
 obj-$(CONFIG_PCI)		+= early-quirks.o
 apm-y				:= apm_32.o
 obj-$(CONFIG_APM)		+= apm.o
-obj-$(CONFIG_X86_SMP)		+= smp_$(BITS).o smpboot_$(BITS).o tsc_sync.o
+obj-$(CONFIG_X86_SMP)		+= smp_$(BITS).o smpboot_$(BITS).o smpboot.o tsc_sync.o
 obj-$(CONFIG_X86_32_SMP)	+= smpcommon_32.o
 obj-$(CONFIG_X86_64_SMP)	+= smp_64.o smpboot_64.o tsc_sync.o
 obj-$(CONFIG_X86_TRAMPOLINE)	+= trampoline_$(BITS).o
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
new file mode 100644
index 000000000000..bffe10861390
--- /dev/null
+++ b/arch/x86/kernel/smpboot.c
@@ -0,0 +1,53 @@
+#include <linux/init.h>
+#include <linux/smp.h>
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+int additional_cpus __initdata = -1;
+
+static __init int setup_additional_cpus(char *s)
+{
+	return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL;
+}
+early_param("additional_cpus", setup_additional_cpus);
+
+/*
+ * cpu_possible_map should be static, it cannot change as cpu's
+ * are onlined, or offlined. The reason is per-cpu data-structures
+ * are allocated by some modules at init time, and dont expect to
+ * do this dynamically on cpu arrival/departure.
+ * cpu_present_map on the other hand can change dynamically.
+ * In case when cpu_hotplug is not compiled, then we resort to current
+ * behaviour, which is cpu_possible == cpu_present.
+ * - Ashok Raj
+ *
+ * Three ways to find out the number of additional hotplug CPUs:
+ * - If the BIOS specified disabled CPUs in ACPI/mptables use that.
+ * - The user can overwrite it with additional_cpus=NUM
+ * - Otherwise don't reserve additional CPUs.
+ * We do this because additional CPUs waste a lot of memory.
+ * -AK
+ */
+__init void prefill_possible_map(void)
+{
+	int i;
+	int possible;
+
+	if (additional_cpus == -1) {
+		if (disabled_cpus > 0)
+			additional_cpus = disabled_cpus;
+		else
+			additional_cpus = 0;
+	}
+	possible = num_processors + additional_cpus;
+	if (possible > NR_CPUS)
+		possible = NR_CPUS;
+
+	printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n",
+		possible, max_t(int, possible - num_processors, 0));
+
+	for (i = 0; i < possible; i++)
+		cpu_set(i, cpu_possible_map);
+}
+#endif
+
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index fd0d3a93b995..953b0ff72b65 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -749,51 +749,6 @@ static __init void disable_smp(void)
 	cpu_set(0, per_cpu(cpu_core_map, 0));
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
-
-int additional_cpus __initdata = -1;
-
-/*
- * cpu_possible_map should be static, it cannot change as cpu's
- * are onlined, or offlined. The reason is per-cpu data-structures
- * are allocated by some modules at init time, and dont expect to
- * do this dynamically on cpu arrival/departure.
- * cpu_present_map on the other hand can change dynamically.
- * In case when cpu_hotplug is not compiled, then we resort to current
- * behaviour, which is cpu_possible == cpu_present.
- * - Ashok Raj
- *
- * Three ways to find out the number of additional hotplug CPUs:
- * - If the BIOS specified disabled CPUs in ACPI/mptables use that.
- * - The user can overwrite it with additional_cpus=NUM
- * - Otherwise don't reserve additional CPUs.
- * We do this because additional CPUs waste a lot of memory.
- * -AK
- */
-__init void prefill_possible_map(void)
-{
-	int i;
-	int possible;
-
- 	if (additional_cpus == -1) {
- 		if (disabled_cpus > 0)
- 			additional_cpus = disabled_cpus;
- 		else
-			additional_cpus = 0;
- 	}
-	possible = num_processors + additional_cpus;
-	if (possible > NR_CPUS) 
-		possible = NR_CPUS;
-
-	printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n",
-		possible,
-	        max_t(int, possible - num_processors, 0));
-
-	for (i = 0; i < possible; i++)
-		cpu_set(i, cpu_possible_map);
-}
-#endif
-
 /*
  * Various sanity checks.
  */
@@ -1087,12 +1042,6 @@ void __cpu_die(unsigned int cpu)
  	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
 }
 
-static __init int setup_additional_cpus(char *s)
-{
-	return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL;
-}
-early_param("additional_cpus", setup_additional_cpus);
-
 #else /* ... !CONFIG_HOTPLUG_CPU */
 
 int __cpu_disable(void)
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index be8a511fdb1d..28cb1f8bb47e 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -63,6 +63,7 @@ void native_smp_cpus_done(unsigned int max_cpus);
 int native_cpu_up(unsigned int cpunum);
 
 extern unsigned disabled_cpus;
+extern void prefill_possible_map(void);
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index c7a00caa6ec6..e5bc1be70827 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -43,7 +43,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
-extern void prefill_possible_map(void);
 
 #define raw_smp_processor_id()	read_pda(cpunumber)
 #define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
-- 
cgit v1.2.3


From 3d3f487c58ef1ece714af280b29411960908149c Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:48 -0300
Subject: x86: provide hlt_works function.

In x86_64, hlt always work. in i386, we'll query the cpuinfo associated
with this cpu

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor.h | 9 +++++++++
 1 file changed, 9 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 9054734589fe..8bec23c15527 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -144,6 +144,15 @@ DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
 #define current_cpu_data	boot_cpu_data
 #endif
 
+static inline int hlt_works(int cpu)
+{
+#ifdef CONFIG_X86_32
+	return cpu_data(cpu).hlt_works_ok;
+#else
+	return 1;
+#endif
+}
+
 #define cache_line_size()	(boot_cpu_data.x86_cache_alignment)
 
 extern void cpu_detect(struct cpuinfo_x86 *c);
-- 
cgit v1.2.3


From 377d698426b8c685fb6d48fe89694fe4ce3aa1f8 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:51 -0300
Subject: x86: unify smp_send_stop

function definition is moved to common header.
x86_64 version is now called native_smp_send_stop

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smp_64.c | 3 ++-
 include/asm-x86/smp.h    | 5 +++++
 include/asm-x86/smp_32.h | 5 -----
 3 files changed, 7 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smp_64.c b/arch/x86/kernel/smp_64.c
index 4e1e2bce969c..ad11ef0c3fae 100644
--- a/arch/x86/kernel/smp_64.c
+++ b/arch/x86/kernel/smp_64.c
@@ -421,7 +421,7 @@ static void stop_this_cpu(void *dummy)
 	for (;;);
 }
 
-void smp_send_stop(void)
+void native_smp_send_stop(void)
 {
 	int nolock;
 	unsigned long flags;
@@ -482,6 +482,7 @@ struct smp_ops smp_ops = {
 	.smp_prepare_cpus = native_smp_prepare_cpus,
 	.smp_cpus_done = native_smp_cpus_done,
 
+	.smp_send_stop = native_smp_send_stop,
 	.smp_send_reschedule = native_smp_send_reschedule,
 	.smp_call_function_mask = native_smp_call_function_mask,
 	.cpu_up = native_cpu_up,
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 28cb1f8bb47e..2ab8ed4e99e0 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -25,6 +25,11 @@ struct smp_ops {
 #ifdef CONFIG_SMP
 extern struct smp_ops smp_ops;
 
+static inline void smp_send_stop(void)
+{
+	smp_ops.smp_send_stop();
+}
+
 static inline void smp_prepare_boot_cpu(void)
 {
 	smp_ops.smp_prepare_boot_cpu();
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index bc90a4ed3235..41b58e0bc75b 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -39,11 +39,6 @@ extern void remove_siblinginfo(int cpu);
 extern void set_cpu_sibling_map(int cpu);
 
 #ifdef CONFIG_SMP
-static inline void smp_send_stop(void)
-{
-	smp_ops.smp_send_stop();
-}
-
 #ifndef CONFIG_PARAVIRT
 #define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
 #endif
-- 
cgit v1.2.3


From fe6762030ca3728d3e24b556676114a6a64a97be Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:56 -0300
Subject: x86: remove cpu_llc_id from processor.h

it is already defined in smp.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 8bec23c15527..e49e5e69ebb0 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -355,9 +355,7 @@ union i387_union {
 	struct i387_soft_struct		soft;
 };
 
-#ifdef CONFIG_X86_32
-DECLARE_PER_CPU(u8, cpu_llc_id);
-#else
+#ifdef CONFIG_X86_64
 DECLARE_PER_CPU(struct orig_ist, orig_ist);
 #endif
 
-- 
cgit v1.2.3


From 5382e89670399f9db8a58b3c6f850fa4a94f6cca Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:57 -0300
Subject: x86: adjust types in smpcommon_32.c

so they can have the same type as x86_64

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_32.c | 6 +++---
 include/asm-x86/smp_32.h     | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 579b9b740c7c..5a446f079b33 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -67,7 +67,7 @@ int smp_num_siblings = 1;
 EXPORT_SYMBOL(smp_num_siblings);
 
 /* Last level cache ID of each logical CPU */
-DEFINE_PER_CPU(u8, cpu_llc_id) = BAD_APICID;
+DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
 
 /* representing HT siblings of each logical CPU */
 DEFINE_PER_CPU(cpumask_t, cpu_sibling_map);
@@ -92,10 +92,10 @@ DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
 EXPORT_PER_CPU_SYMBOL(cpu_info);
 
 /* which logical CPU number maps to which CPU (physical APIC ID) */
-u8 x86_cpu_to_apicid_init[NR_CPUS] __initdata =
+u16 x86_cpu_to_apicid_init[NR_CPUS] __initdata =
 			{ [0 ... NR_CPUS-1] = BAD_APICID };
 void *x86_cpu_to_apicid_early_ptr;
-DEFINE_PER_CPU(u8, x86_cpu_to_apicid) = BAD_APICID;
+DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
 u8 apicid_2_node[MAX_APICID];
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 41b58e0bc75b..29f61952ea8c 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -21,13 +21,13 @@ extern cpumask_t cpu_callin_map;
 extern void (*mtrr_hook) (void);
 extern void zap_low_mappings (void);
 
-extern u8 __initdata x86_cpu_to_apicid_init[];
+extern u16 __initdata x86_cpu_to_apicid_init[];
 extern void *x86_cpu_to_apicid_early_ptr;
 
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 DECLARE_PER_CPU(cpumask_t, cpu_core_map);
-DECLARE_PER_CPU(u8, cpu_llc_id);
-DECLARE_PER_CPU(u8, x86_cpu_to_apicid);
+DECLARE_PER_CPU(u16, cpu_llc_id);
+DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
 
 #ifdef CONFIG_HOTPLUG_CPU
 extern void cpu_exit_clear(void);
-- 
cgit v1.2.3


From a355352b97901d987f54ea7c7d7161eb51a3799c Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:58 -0300
Subject: x86: move equal types to common file

move definitions that are now equal in type from
smpboot_{32,64}.c to smpboot.c

cpu_callin_map is put temporarily in smp_64.h (already
exists in smp_32.h), and will soon be merged.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot.c    | 28 ++++++++++++++++++++++++++++
 arch/x86/kernel/smpboot_32.c | 27 ---------------------------
 arch/x86/kernel/smpboot_64.c | 33 ---------------------------------
 include/asm-x86/smp_64.h     |  1 +
 4 files changed, 29 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index bffe10861390..40a3b56952ef 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -1,6 +1,34 @@
 #include <linux/init.h>
 #include <linux/smp.h>
+#include <linux/module.h>
 
+/* Number of siblings per CPU package */
+int smp_num_siblings = 1;
+EXPORT_SYMBOL(smp_num_siblings);
+
+/* Last level cache ID of each logical CPU */
+DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
+
+/* bitmap of online cpus */
+cpumask_t cpu_online_map __read_mostly;
+EXPORT_SYMBOL(cpu_online_map);
+
+cpumask_t cpu_callin_map;
+cpumask_t cpu_callout_map;
+cpumask_t cpu_possible_map;
+EXPORT_SYMBOL(cpu_possible_map);
+
+/* representing HT siblings of each logical CPU */
+DEFINE_PER_CPU(cpumask_t, cpu_sibling_map);
+EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
+
+/* representing HT and core siblings of each logical CPU */
+DEFINE_PER_CPU(cpumask_t, cpu_core_map);
+EXPORT_PER_CPU_SYMBOL(cpu_core_map);
+
+/* Per CPU bogomips and other parameters */
+DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
+EXPORT_PER_CPU_SYMBOL(cpu_info);
 #ifdef CONFIG_HOTPLUG_CPU
 
 int additional_cpus __initdata = -1;
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 5a446f079b33..0fbc98163b4e 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -62,35 +62,8 @@
 /* Set if we find a B stepping CPU */
 static int __cpuinitdata smp_b_stepping;
 
-/* Number of siblings per CPU package */
-int smp_num_siblings = 1;
-EXPORT_SYMBOL(smp_num_siblings);
-
-/* Last level cache ID of each logical CPU */
-DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
-
-/* representing HT siblings of each logical CPU */
-DEFINE_PER_CPU(cpumask_t, cpu_sibling_map);
-EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
-
-/* representing HT and core siblings of each logical CPU */
-DEFINE_PER_CPU(cpumask_t, cpu_core_map);
-EXPORT_PER_CPU_SYMBOL(cpu_core_map);
-
-/* bitmap of online cpus */
-cpumask_t cpu_online_map __read_mostly;
-EXPORT_SYMBOL(cpu_online_map);
-
-cpumask_t cpu_callin_map;
-cpumask_t cpu_callout_map;
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
 static cpumask_t smp_commenced_mask;
 
-/* Per CPU bogomips and other parameters */
-DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
-EXPORT_PER_CPU_SYMBOL(cpu_info);
-
 /* which logical CPU number maps to which CPU (physical APIC ID) */
 u16 x86_cpu_to_apicid_init[NR_CPUS] __initdata =
 			{ [0 ... NR_CPUS-1] = BAD_APICID };
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 953b0ff72b65..c51279f05316 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -60,42 +60,9 @@
 #include <asm/hw_irq.h>
 #include <asm/numa.h>
 
-/* Number of siblings per CPU package */
-int smp_num_siblings = 1;
-EXPORT_SYMBOL(smp_num_siblings);
-
-/* Last level cache ID of each logical CPU */
-DEFINE_PER_CPU(u16, cpu_llc_id) = BAD_APICID;
-
-/* Bitmask of currently online CPUs */
-cpumask_t cpu_online_map __read_mostly;
-
-EXPORT_SYMBOL(cpu_online_map);
-
-/*
- * Private maps to synchronize booting between AP and BP.
- * Probably not needed anymore, but it makes for easier debugging. -AK
- */
-cpumask_t cpu_callin_map;
-cpumask_t cpu_callout_map;
-cpumask_t cpu_possible_map;
-EXPORT_SYMBOL(cpu_possible_map);
-
-/* Per CPU bogomips and other parameters */
-DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
-EXPORT_PER_CPU_SYMBOL(cpu_info);
-
 /* Set when the idlers are all forked */
 int smp_threads_ready;
 
-/* representing HT siblings of each logical CPU */
-DEFINE_PER_CPU(cpumask_t, cpu_sibling_map);
-EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
-
-/* representing HT and core siblings of each logical CPU */
-DEFINE_PER_CPU(cpumask_t, cpu_core_map);
-EXPORT_PER_CPU_SYMBOL(cpu_core_map);
-
 /*
  * Trampoline 80x86 program as an array.
  */
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index e5bc1be70827..1ecf8134bdc9 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -14,6 +14,7 @@
 #include <asm/thread_info.h>
 
 extern cpumask_t cpu_initialized;
+extern cpumask_t cpu_callin_map;
 
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
-- 
cgit v1.2.3


From 1452207689b3c0dd2ffed40735289a3a4a8c0c7c Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:12:59 -0300
Subject: x86: make set_cpu_sibling_map nonstatic

And move its extern definition to smp.h, the common header

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_64.c | 2 +-
 include/asm-x86/smp.h        | 3 +++
 include/asm-x86/smp_32.h     | 3 ---
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index c51279f05316..1e8f00a1d624 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -228,7 +228,7 @@ cpumask_t cpu_coregroup_map(int cpu)
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
 
-static inline void set_cpu_sibling_map(int cpu)
+void __cpuinit set_cpu_sibling_map(int cpu)
 {
 	int i;
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 2ab8ed4e99e0..1b4481aeb5c8 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -22,6 +22,9 @@ struct smp_ops {
 				      int wait);
 };
 
+/* Globals due to paravirt */
+extern void set_cpu_sibling_map(int cpu);
+
 #ifdef CONFIG_SMP
 extern struct smp_ops smp_ops;
 
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 29f61952ea8c..76247a947a5f 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -35,9 +35,6 @@ extern void cpu_uninit(void);
 extern void remove_siblinginfo(int cpu);
 #endif
 
-/* Globals due to paravirt */
-extern void set_cpu_sibling_map(int cpu);
-
 #ifdef CONFIG_SMP
 #ifndef CONFIG_PARAVIRT
 #define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
-- 
cgit v1.2.3


From 1dbb4726faebe9e64a1e9cf40e3b39fffa065a65 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:13:01 -0300
Subject: x86: move hotplug related extern definitions to smp.h

definitions that are inside CONFIG_HOTPLUG_CPU in
the arch-specific smp*.h files are moved to common
header

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 6 ++++++
 include/asm-x86/smp_32.h | 6 ------
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 1b4481aeb5c8..c800b815d378 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -80,6 +80,12 @@ extern void prefill_possible_map(void);
 # include "smp_64.h"
 #endif
 
+#ifdef CONFIG_HOTPLUG_CPU
+extern void cpu_exit_clear(void);
+extern void cpu_uninit(void);
+extern void remove_siblinginfo(int cpu);
+#endif
+
 extern void smp_alloc_memory(void);
 extern void lock_ipi_call_lock(void);
 extern void unlock_ipi_call_lock(void);
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 76247a947a5f..0b2513468870 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -29,12 +29,6 @@ DECLARE_PER_CPU(cpumask_t, cpu_core_map);
 DECLARE_PER_CPU(u16, cpu_llc_id);
 DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
 
-#ifdef CONFIG_HOTPLUG_CPU
-extern void cpu_exit_clear(void);
-extern void cpu_uninit(void);
-extern void remove_siblinginfo(int cpu);
-#endif
-
 #ifdef CONFIG_SMP
 #ifndef CONFIG_PARAVIRT
 #define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
-- 
cgit v1.2.3


From 69c18c15d39c4622c6e2f97e5db4d8c9c43adaaa Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:13:07 -0300
Subject: x86: merge __cpu_disable and cpu_die

They are now equal, and are moved to a common file

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot.c    | 85 ++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/smpboot_32.c | 67 ----------------------------------
 arch/x86/kernel/smpboot_64.c | 79 ----------------------------------------
 include/asm-x86/smp.h        |  3 ++
 include/asm-x86/smp_32.h     |  3 --
 include/asm-x86/smp_64.h     |  3 --
 6 files changed, 88 insertions(+), 152 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 644e60969f90..c35cd319d1ed 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -2,6 +2,13 @@
 #include <linux/smp.h>
 #include <linux/module.h>
 #include <linux/sched.h>
+#include <linux/percpu.h>
+
+#include <asm/nmi.h>
+#include <asm/irq.h>
+#include <asm/smp.h>
+#include <asm/cpu.h>
+#include <asm/numa.h>
 
 /* Number of siblings per CPU package */
 int smp_num_siblings = 1;
@@ -181,5 +188,83 @@ __init void prefill_possible_map(void)
 	for (i = 0; i < possible; i++)
 		cpu_set(i, cpu_possible_map);
 }
+
+static void __ref remove_cpu_from_maps(int cpu)
+{
+	cpu_clear(cpu, cpu_online_map);
+#ifdef CONFIG_X86_64
+	cpu_clear(cpu, cpu_callout_map);
+	cpu_clear(cpu, cpu_callin_map);
+	/* was set by cpu_init() */
+	clear_bit(cpu, (unsigned long *)&cpu_initialized);
+	clear_node_cpumask(cpu);
+#endif
+}
+
+int __cpu_disable(void)
+{
+	int cpu = smp_processor_id();
+
+	/*
+	 * Perhaps use cpufreq to drop frequency, but that could go
+	 * into generic code.
+	 *
+	 * We won't take down the boot processor on i386 due to some
+	 * interrupts only being able to be serviced by the BSP.
+	 * Especially so if we're not using an IOAPIC	-zwane
+	 */
+	if (cpu == 0)
+		return -EBUSY;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		stop_apic_nmi_watchdog(NULL);
+	clear_local_APIC();
+
+	/*
+	 * HACK:
+	 * Allow any queued timer interrupts to get serviced
+	 * This is only a temporary solution until we cleanup
+	 * fixup_irqs as we do for IA64.
+	 */
+	local_irq_enable();
+	mdelay(1);
+
+	local_irq_disable();
+	remove_siblinginfo(cpu);
+
+	/* It's now safe to remove this processor from the online map */
+	remove_cpu_from_maps(cpu);
+	fixup_irqs(cpu_online_map);
+	return 0;
+}
+
+void __cpu_die(unsigned int cpu)
+{
+	/* We don't do anything here: idle task is faking death itself. */
+	unsigned int i;
+
+	for (i = 0; i < 10; i++) {
+		/* They ack this in play_dead by setting CPU_DEAD */
+		if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
+			printk(KERN_INFO "CPU %d is now offline\n", cpu);
+			if (1 == num_online_cpus())
+				alternatives_smp_switch(0);
+			return;
+		}
+		msleep(100);
+	}
+	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
+}
+#else /* ... !CONFIG_HOTPLUG_CPU */
+int __cpu_disable(void)
+{
+	return -ENOSYS;
+}
+
+void __cpu_die(unsigned int cpu)
+{
+	/* We said "no" in __cpu_disable */
+	BUG();
+}
 #endif
 
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 3d21c663aa76..00b1b59cd560 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -1040,73 +1040,6 @@ void __init native_smp_prepare_boot_cpu(void)
 	__get_cpu_var(cpu_state) = CPU_ONLINE;
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
-static void __ref remove_cpu_from_maps(int cpu)
-{
-	cpu_clear(cpu, cpu_online_map);
-}
-
-int __cpu_disable(void)
-{
-	cpumask_t map = cpu_online_map;
-	int cpu = smp_processor_id();
-
-	/*
-	 * Perhaps use cpufreq to drop frequency, but that could go
-	 * into generic code.
- 	 *
-	 * We won't take down the boot processor on i386 due to some
-	 * interrupts only being able to be serviced by the BSP.
-	 * Especially so if we're not using an IOAPIC	-zwane
-	 */
-	if (cpu == 0)
-		return -EBUSY;
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-		stop_apic_nmi_watchdog(NULL);
-	clear_local_APIC();
-	/* Allow any queued timer interrupts to get serviced */
-	local_irq_enable();
-	mdelay(1);
-	local_irq_disable();
-
-	remove_siblinginfo(cpu);
-
-	remove_cpu_from_maps(cpu);
-	fixup_irqs(map);
-
-	return 0;
-}
-
-void __cpu_die(unsigned int cpu)
-{
-	/* We don't do anything here: idle task is faking death itself. */
-	unsigned int i;
-
-	for (i = 0; i < 10; i++) {
-		/* They ack this in play_dead by setting CPU_DEAD */
-		if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
-			printk ("CPU %d is now offline\n", cpu);
-			if (1 == num_online_cpus())
-				alternatives_smp_switch(0);
-			return;
-		}
-		msleep(100);
-	}
- 	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
-}
-#else /* ... !CONFIG_HOTPLUG_CPU */
-int __cpu_disable(void)
-{
-	return -ENOSYS;
-}
-
-void __cpu_die(unsigned int cpu)
-{
-	/* We said "no" in __cpu_disable */
-	BUG();
-}
-#endif /* CONFIG_HOTPLUG_CPU */
-
 int __cpuinit native_cpu_up(unsigned int cpu)
 {
 	unsigned long flags;
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 6509d3c1b3df..0c67e5ae9c9d 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -836,82 +836,3 @@ void __init native_smp_cpus_done(unsigned int max_cpus)
 	setup_ioapic_dest();
 	check_nmi_watchdog();
 }
-
-#ifdef CONFIG_HOTPLUG_CPU
-static void __ref remove_cpu_from_maps(int cpu)
-{
-	cpu_clear(cpu, cpu_online_map);
-	cpu_clear(cpu, cpu_callout_map);
-	cpu_clear(cpu, cpu_callin_map);
-	clear_bit(cpu, (unsigned long *)&cpu_initialized); /* was set by cpu_init() */
-	clear_node_cpumask(cpu);
-}
-
-int __cpu_disable(void)
-{
-	int cpu = smp_processor_id();
-
-	/*
-	 * Perhaps use cpufreq to drop frequency, but that could go
-	 * into generic code.
- 	 *
-	 * We won't take down the boot processor on i386 due to some
-	 * interrupts only being able to be serviced by the BSP.
-	 * Especially so if we're not using an IOAPIC	-zwane
-	 */
-	if (cpu == 0)
-		return -EBUSY;
-
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-		stop_apic_nmi_watchdog(NULL);
-	clear_local_APIC();
-
-	/*
-	 * HACK:
-	 * Allow any queued timer interrupts to get serviced
-	 * This is only a temporary solution until we cleanup
-	 * fixup_irqs as we do for IA64.
-	 */
-	local_irq_enable();
-	mdelay(1);
-
-	local_irq_disable();
-	remove_siblinginfo(cpu);
-
-	/* It's now safe to remove this processor from the online map */
-	remove_cpu_from_maps(cpu);
-	fixup_irqs(cpu_online_map);
-	return 0;
-}
-
-void __cpu_die(unsigned int cpu)
-{
-	/* We don't do anything here: idle task is faking death itself. */
-	unsigned int i;
-
-	for (i = 0; i < 10; i++) {
-		/* They ack this in play_dead by setting CPU_DEAD */
-		if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
-			printk ("CPU %d is now offline\n", cpu);
-			if (1 == num_online_cpus())
-				alternatives_smp_switch(0);
-			return;
-		}
-		msleep(100);
-	}
- 	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
-}
-
-#else /* ... !CONFIG_HOTPLUG_CPU */
-
-int __cpu_disable(void)
-{
-	return -ENOSYS;
-}
-
-void __cpu_die(unsigned int cpu)
-{
-	/* We said "no" in __cpu_disable */
-	BUG();
-}
-#endif /* CONFIG_HOTPLUG_CPU */
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index c800b815d378..27d9f6595232 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -70,6 +70,9 @@ void native_smp_prepare_cpus(unsigned int max_cpus);
 void native_smp_cpus_done(unsigned int max_cpus);
 int native_cpu_up(unsigned int cpunum);
 
+extern int __cpu_disable(void);
+extern void __cpu_die(unsigned int cpu);
+
 extern unsigned disabled_cpus;
 extern void prefill_possible_map(void);
 #endif
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 0b2513468870..4fec2feb6ac8 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -34,9 +34,6 @@ DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
 #define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
 #endif
 
-extern int __cpu_disable(void);
-extern void __cpu_die(unsigned int cpu);
-
 /*
  * This function is needed by all SMP systems. It must _always_ be valid
  * from the initial startup. We map APIC_BASE very early in page_setup(),
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 1ecf8134bdc9..d554d7d57327 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -42,9 +42,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 
 #define SMP_TRAMPOLINE_BASE 0x6000
 
-extern int __cpu_disable(void);
-extern void __cpu_die(unsigned int cpu);
-
 #define raw_smp_processor_id()	read_pda(cpunumber)
 #define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
 
-- 
cgit v1.2.3


From 420688293927a590d092ec76ef97c2565ae21aff Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:13:09 -0300
Subject: x86: move trampoline arrays extern definition to smp.h

In here, they can serve both architectures

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_32.c | 6 ------
 arch/x86/kernel/smpboot_64.c | 7 -------
 include/asm-x86/smp.h        | 6 ++++++
 3 files changed, 6 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 3236e843a9ad..a21f25418b3e 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -73,12 +73,6 @@ EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
 u8 apicid_2_node[MAX_APICID];
 
-/*
- * Trampoline 80x86 program as an array.
- */
-
-extern const unsigned char trampoline_data [];
-extern const unsigned char trampoline_end  [];
 static unsigned char *trampoline_base;
 
 static void map_cpu_to_logical_apicid(void);
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 0c67e5ae9c9d..2cc1b8b0601c 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -63,13 +63,6 @@
 /* Set when the idlers are all forked */
 int smp_threads_ready;
 
-/*
- * Trampoline 80x86 program as an array.
- */
-
-extern const unsigned char trampoline_data[];
-extern const unsigned char trampoline_end[];
-
 /* State of each CPU */
 DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 27d9f6595232..b2a1697e4700 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -9,6 +9,12 @@ extern cpumask_t cpu_callout_map;
 extern int smp_num_siblings;
 extern unsigned int num_processors;
 
+/*
+ * Trampoline 80x86 program as an array.
+ */
+extern const unsigned char trampoline_data [];
+extern const unsigned char trampoline_end  [];
+
 struct smp_ops {
 	void (*smp_prepare_boot_cpu)(void);
 	void (*smp_prepare_cpus)(unsigned max_cpus);
-- 
cgit v1.2.3


From 91718e8d13c23bfe0aa6fa6b730c5c33ee9771bf Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Mon, 3 Mar 2008 14:13:12 -0300
Subject: x86: unify setup_trampoline

setup_trampoline() looks very similar between architectures, and this
patch unifies them. The i386 version allocates bootmem memory, while
the x86_64 version uses a fixed address.

In this patch, we initialize the global trampoline_base to the x86_64 version,
and i386 allocation can later override it.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot.c    | 33 +++++++++++++++++++++++++++++++++
 arch/x86/kernel/smpboot_32.c | 29 -----------------------------
 arch/x86/kernel/smpboot_64.c | 14 --------------
 include/asm-x86/smp.h        |  4 ++++
 include/asm-x86/smp_64.h     |  2 --
 5 files changed, 37 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 34c31178041b..b13b9d55f9ce 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -3,6 +3,7 @@
 #include <linux/module.h>
 #include <linux/sched.h>
 #include <linux/percpu.h>
+#include <linux/bootmem.h>
 
 #include <asm/nmi.h>
 #include <asm/irq.h>
@@ -38,6 +39,9 @@ EXPORT_PER_CPU_SYMBOL(cpu_core_map);
 DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
 EXPORT_PER_CPU_SYMBOL(cpu_info);
 
+/* ready for x86_64, no harm for x86, since it will overwrite after alloc */
+unsigned char *trampoline_base = __va(SMP_TRAMPOLINE_BASE);
+
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
 
@@ -117,6 +121,35 @@ cpumask_t cpu_coregroup_map(int cpu)
 		return c->llc_shared_map;
 }
 
+/*
+ * Currently trivial. Write the real->protected mode
+ * bootstrap into the page concerned. The caller
+ * has made sure it's suitably aligned.
+ */
+
+unsigned long __cpuinit setup_trampoline(void)
+{
+	memcpy(trampoline_base, trampoline_data,
+	       trampoline_end - trampoline_data);
+	return virt_to_phys(trampoline_base);
+}
+
+#ifdef CONFIG_X86_32
+/*
+ * We are called very early to get the low memory for the
+ * SMP bootup trampoline page.
+ */
+void __init smp_alloc_memory(void)
+{
+	trampoline_base = alloc_bootmem_low_pages(PAGE_SIZE);
+	/*
+	 * Has to be in very low memory so we can execute
+	 * real-mode AP code.
+	 */
+	if (__pa(trampoline_base) >= 0x9F000)
+		BUG();
+}
+#endif
 
 #ifdef CONFIG_HOTPLUG_CPU
 void remove_siblinginfo(int cpu)
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index a21f25418b3e..ee826594aa03 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -73,40 +73,11 @@ EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
 u8 apicid_2_node[MAX_APICID];
 
-static unsigned char *trampoline_base;
-
 static void map_cpu_to_logical_apicid(void);
 
 /* State of each CPU. */
 DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
-/*
- * Currently trivial. Write the real->protected mode
- * bootstrap into the page concerned. The caller
- * has made sure it's suitably aligned.
- */
-
-static unsigned long __cpuinit setup_trampoline(void)
-{
-	memcpy(trampoline_base, trampoline_data, trampoline_end - trampoline_data);
-	return virt_to_phys(trampoline_base);
-}
-
-/*
- * We are called very early to get the low memory for the
- * SMP bootup trampoline page.
- */
-void __init smp_alloc_memory(void)
-{
-	trampoline_base = alloc_bootmem_low_pages(PAGE_SIZE);
-	/*
-	 * Has to be in very low memory so we can execute
-	 * real-mode AP code.
-	 */
-	if (__pa(trampoline_base) >= 0x9F000)
-		BUG();
-}
-
 /*
  * The bootstrap kernel entry code has set these up. Save them for
  * a given CPU
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 2cc1b8b0601c..9f4935e70e72 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -85,20 +85,6 @@ struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
 #define set_idle_for_cpu(x,p)   (idle_thread_array[(x)] = (p))
 #endif
 
-
-/*
- * Currently trivial. Write the real->protected mode
- * bootstrap into the page concerned. The caller
- * has made sure it's suitably aligned.
- */
-
-static unsigned long __cpuinit setup_trampoline(void)
-{
-	void *tramp = __va(SMP_TRAMPOLINE_BASE); 
-	memcpy(tramp, trampoline_data, trampoline_end - trampoline_data);
-	return virt_to_phys(tramp);
-}
-
 /*
  * The bootstrap kernel entry code has set these up. Save them for
  * a given CPU
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index b2a1697e4700..513c8571a4a0 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -14,6 +14,7 @@ extern unsigned int num_processors;
  */
 extern const unsigned char trampoline_data [];
 extern const unsigned char trampoline_end  [];
+extern unsigned char *trampoline_base;
 
 struct smp_ops {
 	void (*smp_prepare_boot_cpu)(void);
@@ -81,6 +82,9 @@ extern void __cpu_die(unsigned int cpu);
 
 extern unsigned disabled_cpus;
 extern void prefill_possible_map(void);
+
+#define SMP_TRAMPOLINE_BASE 0x6000
+extern unsigned long setup_trampoline(void);
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index d554d7d57327..394c78524331 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -40,8 +40,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 
 #ifdef CONFIG_SMP
 
-#define SMP_TRAMPOLINE_BASE 0x6000
-
 #define raw_smp_processor_id()	read_pda(cpunumber)
 #define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
 
-- 
cgit v1.2.3


From ca9cda2f7b53da619fabde4c0c1bd5f61039bd5b Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 5 Mar 2008 15:15:42 +0100
Subject: x86: add comments to processor.h

add comments to the FPU structures of processor.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor.h | 51 +++++++++++++++++++++++++--------------------
 1 file changed, 28 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index e49e5e69ebb0..1f9501a38493 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -289,42 +289,47 @@ struct orig_ist {
 #define	MXCSR_DEFAULT		0x1f80
 
 struct i387_fsave_struct {
-	u32			cwd;
-	u32			swd;
-	u32			twd;
-	u32			fip;
-	u32			fcs;
-	u32			foo;
-	u32			fos;
-	/* 8*10 bytes for each FP-reg = 80 bytes: */
+	u32			cwd;	/* FPU Control Word		*/
+	u32			swd;	/* FPU Status Word		*/
+	u32			twd;	/* FPU Tag Word			*/
+	u32			fip;	/* FPU IP Offset		*/
+	u32			fcs;	/* FPU IP Selector		*/
+	u32			foo;	/* FPU Operand Pointer Offset	*/
+	u32			fos;	/* FPU Operand Pointer Selector	*/
+
+	/* 8*10 bytes for each FP-reg = 80 bytes:			*/
 	u32			st_space[20];
-	/* Software status information: */
+
+	/* Software status information [not touched by FSAVE ]:		*/
 	u32			status;
 };
 
 struct i387_fxsave_struct {
-	u16			cwd;
-	u16			swd;
-	u16			twd;
-	u16			fop;
+	u16			cwd; /* Control Word			*/
+	u16			swd; /* Status Word			*/
+	u16			twd; /* Tag Word			*/
+	u16			fop; /* Last Instruction Opcode		*/
 	union {
 		struct {
-			u64	rip;
-			u64	rdp;
+			u64	rip; /* Instruction Pointer		*/
+			u64	rdp; /* Data Pointer			*/
 		};
 		struct {
-			u32	fip;
-			u32	fcs;
-			u32	foo;
-			u32	fos;
+			u32	fip; /* FPU IP Offset			*/
+			u32	fcs; /* FPU IP Selector			*/
+			u32	foo; /* FPU Operand Offset		*/
+			u32	fos; /* FPU Operand Selector		*/
 		};
 	};
-	u32			mxcsr;
-	u32			mxcsr_mask;
-	/* 8*16 bytes for each FP-reg = 128 bytes: */
+	u32			mxcsr;		/* MXCSR Register State */
+	u32			mxcsr_mask;	/* MXCSR Mask		*/
+
+	/* 8*16 bytes for each FP-reg = 128 bytes:			*/
 	u32			st_space[32];
-	/* 16*16 bytes for each XMM-reg = 256 bytes: */
+
+	/* 16*16 bytes for each XMM-reg = 256 bytes:			*/
 	u32			xmm_space[64];
+
 	u32			padding[24];
 
 } __attribute__((aligned(16)));
-- 
cgit v1.2.3


From 16281a998d7340a5bee4078f6f9a26c47208eb86 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 4 Mar 2008 16:46:27 -0800
Subject: x86: include/asm-x86/mutex_32.h - use angle brackets for include

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mutex_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/mutex_32.h b/include/asm-x86/mutex_32.h
index bbeefb96ddfd..9a6b3da25914 100644
--- a/include/asm-x86/mutex_32.h
+++ b/include/asm-x86/mutex_32.h
@@ -9,7 +9,7 @@
 #ifndef _ASM_MUTEX_H
 #define _ASM_MUTEX_H
 
-#include "asm/alternative.h"
+#include <asm/alternative.h>
 
 /**
  *  __mutex_fastpath_lock - try to take the lock by moving the count
-- 
cgit v1.2.3


From eee6dd15723639f9270e4c561a0c82e8e18bd587 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 6 Mar 2008 10:39:07 +0100
Subject: x86: move extern declaration to vdso.h

Before:
   total: 0 errors, 3 warnings, 685 lines checked
After:
   total: 0 errors, 1 warnings, 678 lines checked

No code changed:

arch/x86/kernel/signal_32.o:

   text	   data	    bss	    dec	    hex	filename
   5333	      0	      4	   5337	   14d9	signal_32.o.before
   5333	      0	      4	   5337	   14d9	signal_32.o.after

md5:
   c279e98012a2808e90cfa2a7787e42a4  signal_32.o.before.asm
   c279e98012a2808e90cfa2a7787e42a4  signal_32.o.after.asm

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/signal_32.c | 7 -------
 include/asm-x86/vdso.h      | 7 +++++++
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 3da3ffa39e9a..ba168e5743be 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -342,13 +342,6 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size)
 	return (void __user *) sp;
 }
 
-/*
- * These symbols are defined with the addresses in the vsyscall page.
- * See vsyscall-sigreturn.S.
- */
-extern void __user __kernel_sigreturn;
-extern void __user __kernel_rt_sigreturn;
-
 static int
 setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 	    struct pt_regs *regs)
diff --git a/include/asm-x86/vdso.h b/include/asm-x86/vdso.h
index 629bcb6e8e45..9bb86899abf0 100644
--- a/include/asm-x86/vdso.h
+++ b/include/asm-x86/vdso.h
@@ -25,4 +25,11 @@ extern const char VDSO32_PRELINK[];
 	(void *) (VDSO32_##name - VDSO32_PRELINK + (unsigned long) (base)); })
 #endif
 
+/*
+ * These symbols are defined with the addresses in the vsyscall page.
+ * See vsyscall-sigreturn.S.
+ */
+extern void __user __kernel_sigreturn;
+extern void __user __kernel_rt_sigreturn;
+
 #endif	/* asm-x86/vdso.h */
-- 
cgit v1.2.3


From e587cadd8f47e202a30712e2906a65a0606d5865 Mon Sep 17 00:00:00 2001
From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Date: Thu, 6 Mar 2008 08:48:49 -0500
Subject: x86: enhance DEBUG_RODATA support - alternatives

Fix a memcpy that should be a text_poke (in apply_alternatives).

Use kernel_wp_save/kernel_wp_restore in text_poke to support DEBUG_RODATA
correctly and so the CPU HOTPLUG special case can be removed.

Add text_poke_early, for alternatives and paravirt boot-time and module load
time patching.

Changelog:

- Fix text_set and text_poke alignment check (mixed up bitwise and and or)
- Remove text_set
- Export add_nops, so it can be used by others.
- Document text_poke_early.
- Remove clflush, since it breaks some VIA architectures and is not strictly
  necessary.
- Add kerneldoc to text_poke and text_poke_early.
- Create a second vmap instead of using the WP bit to support Xen and VMI.
- Move local_irq disable within text_poke and text_poke_early to be able to
  be sleepable in these functions.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: Andi Kleen <andi@firstfloor.org>
CC: pageexec@freemail.hu
CC: H. Peter Anvin <hpa@zytor.com>
CC: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/alternative.c | 88 ++++++++++++++++++++++++++++++++-----------
 include/asm-x86/alternative.h | 23 ++++++++++-
 2 files changed, 87 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index e2d30b8e08a2..0c92ad4d257a 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -11,6 +11,8 @@
 #include <asm/mce.h>
 #include <asm/nmi.h>
 #include <asm/vsyscall.h>
+#include <asm/cacheflush.h>
+#include <asm/io.h>
 
 #define MAX_PATCH_LEN (255-1)
 
@@ -177,7 +179,7 @@ static const unsigned char*const * find_nop_table(void)
 #endif /* CONFIG_X86_64 */
 
 /* Use this to add nops to a buffer, then text_poke the whole buffer. */
-static void add_nops(void *insns, unsigned int len)
+void add_nops(void *insns, unsigned int len)
 {
 	const unsigned char *const *noptable = find_nop_table();
 
@@ -190,6 +192,7 @@ static void add_nops(void *insns, unsigned int len)
 		len -= noplen;
 	}
 }
+EXPORT_SYMBOL_GPL(add_nops);
 
 extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
 extern u8 *__smp_locks[], *__smp_locks_end[];
@@ -223,7 +226,7 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
 		memcpy(insnbuf, a->replacement, a->replacementlen);
 		add_nops(insnbuf + a->replacementlen,
 			 a->instrlen - a->replacementlen);
-		text_poke(instr, insnbuf, a->instrlen);
+		text_poke_early(instr, insnbuf, a->instrlen);
 	}
 }
 
@@ -284,7 +287,6 @@ void alternatives_smp_module_add(struct module *mod, char *name,
 				 void *text,  void *text_end)
 {
 	struct smp_alt_module *smp;
-	unsigned long flags;
 
 	if (noreplace_smp)
 		return;
@@ -310,39 +312,37 @@ void alternatives_smp_module_add(struct module *mod, char *name,
 		__func__, smp->locks, smp->locks_end,
 		smp->text, smp->text_end, smp->name);
 
-	spin_lock_irqsave(&smp_alt, flags);
+	spin_lock(&smp_alt);
 	list_add_tail(&smp->next, &smp_alt_modules);
 	if (boot_cpu_has(X86_FEATURE_UP))
 		alternatives_smp_unlock(smp->locks, smp->locks_end,
 					smp->text, smp->text_end);
-	spin_unlock_irqrestore(&smp_alt, flags);
+	spin_unlock(&smp_alt);
 }
 
 void alternatives_smp_module_del(struct module *mod)
 {
 	struct smp_alt_module *item;
-	unsigned long flags;
 
 	if (smp_alt_once || noreplace_smp)
 		return;
 
-	spin_lock_irqsave(&smp_alt, flags);
+	spin_lock(&smp_alt);
 	list_for_each_entry(item, &smp_alt_modules, next) {
 		if (mod != item->mod)
 			continue;
 		list_del(&item->next);
-		spin_unlock_irqrestore(&smp_alt, flags);
+		spin_unlock(&smp_alt);
 		DPRINTK("%s: %s\n", __func__, item->name);
 		kfree(item);
 		return;
 	}
-	spin_unlock_irqrestore(&smp_alt, flags);
+	spin_unlock(&smp_alt);
 }
 
 void alternatives_smp_switch(int smp)
 {
 	struct smp_alt_module *mod;
-	unsigned long flags;
 
 #ifdef CONFIG_LOCKDEP
 	/*
@@ -359,7 +359,7 @@ void alternatives_smp_switch(int smp)
 		return;
 	BUG_ON(!smp && (num_online_cpus() > 1));
 
-	spin_lock_irqsave(&smp_alt, flags);
+	spin_lock(&smp_alt);
 
 	/*
 	 * Avoid unnecessary switches because it forces JIT based VMs to
@@ -383,7 +383,7 @@ void alternatives_smp_switch(int smp)
 						mod->text, mod->text_end);
 	}
 	smp_mode = smp;
-	spin_unlock_irqrestore(&smp_alt, flags);
+	spin_unlock(&smp_alt);
 }
 
 #endif
@@ -411,7 +411,7 @@ void apply_paravirt(struct paravirt_patch_site *start,
 
 		/* Pad the rest with nops */
 		add_nops(insnbuf + used, p->len - used);
-		text_poke(p->instr, insnbuf, p->len);
+		text_poke_early(p->instr, insnbuf, p->len);
 	}
 }
 extern struct paravirt_patch_site __start_parainstructions[],
@@ -420,8 +420,6 @@ extern struct paravirt_patch_site __start_parainstructions[],
 
 void __init alternative_instructions(void)
 {
-	unsigned long flags;
-
 	/* The patching is not fully atomic, so try to avoid local interruptions
 	   that might execute the to be patched code.
 	   Other CPUs are not running. */
@@ -430,7 +428,6 @@ void __init alternative_instructions(void)
 	stop_mce();
 #endif
 
-	local_irq_save(flags);
 	apply_alternatives(__alt_instructions, __alt_instructions_end);
 
 	/* switch to patch-once-at-boottime-only mode and free the
@@ -462,7 +459,6 @@ void __init alternative_instructions(void)
 	}
 #endif
  	apply_paravirt(__parainstructions, __parainstructions_end);
-	local_irq_restore(flags);
 
 	if (smp_alt_once)
 		free_init_pages("SMP alternatives",
@@ -475,18 +471,64 @@ void __init alternative_instructions(void)
 #endif
 }
 
-/*
- * Warning:
+/**
+ * text_poke_early - Update instructions on a live kernel at boot time
+ * @addr: address to modify
+ * @opcode: source of the copy
+ * @len: length to copy
+ *
  * When you use this code to patch more than one byte of an instruction
  * you need to make sure that other CPUs cannot execute this code in parallel.
- * Also no thread must be currently preempted in the middle of these instructions.
- * And on the local CPU you need to be protected again NMI or MCE handlers
- * seeing an inconsistent instruction while you patch.
+ * Also no thread must be currently preempted in the middle of these
+ * instructions. And on the local CPU you need to be protected again NMI or MCE
+ * handlers seeing an inconsistent instruction while you patch.
  */
-void __kprobes text_poke(void *addr, unsigned char *opcode, int len)
+void *text_poke_early(void *addr, const void *opcode, size_t len)
 {
+	unsigned long flags;
+	local_irq_save(flags);
 	memcpy(addr, opcode, len);
+	local_irq_restore(flags);
+	sync_core();
+	/* Could also do a CLFLUSH here to speed up CPU recovery; but
+	   that causes hangs on some VIA CPUs. */
+	return addr;
+}
+
+/**
+ * text_poke - Update instructions on a live kernel
+ * @addr: address to modify
+ * @opcode: source of the copy
+ * @len: length to copy
+ *
+ * Only atomic text poke/set should be allowed when not doing early patching.
+ * It means the size must be writable atomically and the address must be aligned
+ * in a way that permits an atomic write. It also makes sure we fit on a single
+ * page.
+ */
+void *__kprobes text_poke(void *addr, const void *opcode, size_t len)
+{
+	unsigned long flags;
+	char *vaddr;
+	int nr_pages = 2;
+
+	BUG_ON(len > sizeof(long));
+	BUG_ON((((long)addr + len - 1) & ~(sizeof(long) - 1))
+		- ((long)addr & ~(sizeof(long) - 1)));
+	{
+		struct page *pages[2] = { virt_to_page(addr),
+			virt_to_page(addr + PAGE_SIZE) };
+		if (!pages[1])
+			nr_pages = 1;
+		vaddr = vmap(pages, nr_pages, VM_MAP, PAGE_KERNEL);
+		BUG_ON(!vaddr);
+		local_irq_save(flags);
+		memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len);
+		local_irq_restore(flags);
+		vunmap(vaddr);
+	}
 	sync_core();
 	/* Could also do a CLFLUSH here to speed up CPU recovery; but
 	   that causes hangs on some VIA CPUs. */
+	return addr;
 }
diff --git a/include/asm-x86/alternative.h b/include/asm-x86/alternative.h
index d8bacf3c4b08..d26416b5722c 100644
--- a/include/asm-x86/alternative.h
+++ b/include/asm-x86/alternative.h
@@ -156,6 +156,27 @@ apply_paravirt(struct paravirt_patch_site *start,
 #define __parainstructions_end	NULL
 #endif
 
-extern void text_poke(void *addr, unsigned char *opcode, int len);
+extern void add_nops(void *insns, unsigned int len);
+
+/*
+ * Clear and restore the kernel write-protection flag on the local CPU.
+ * Allows the kernel to edit read-only pages.
+ * Side-effect: any interrupt handler running between save and restore will have
+ * the ability to write to read-only pages.
+ *
+ * Warning:
+ * Code patching in the UP case is safe if NMIs and MCE handlers are stopped and
+ * no thread can be preempted in the instructions being modified (no iret to an
+ * invalid instruction possible) or if the instructions are changed from a
+ * consistent state to another consistent state atomically.
+ * More care must be taken when modifying code in the SMP case because of
+ * Intel's errata.
+ * On the local CPU you need to be protected again NMI or MCE handlers seeing an
+ * inconsistent instruction while you patch.
+ * The _early version expects the memory to already be RW.
+ */
+
+extern void *text_poke(void *addr, const void *opcode, size_t len);
+extern void *text_poke_early(void *addr, const void *opcode, size_t len);
 
 #endif /* _ASM_X86_ALTERNATIVE_H */
-- 
cgit v1.2.3


From 459cce726730ca0ac93701e53aa1d0d055ce9e90 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 6 Mar 2008 18:38:52 +0100
Subject: x86: remove mach_reboot.h

all reboot details are handled in reboot.c and quirks are handled
via reboot_fixups_32.c.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mach-default/mach_reboot.h | 61 ------------------------------
 1 file changed, 61 deletions(-)
 delete mode 100644 include/asm-x86/mach-default/mach_reboot.h

(limited to 'include')

diff --git a/include/asm-x86/mach-default/mach_reboot.h b/include/asm-x86/mach-default/mach_reboot.h
deleted file mode 100644
index 6adee6a97dec..000000000000
--- a/include/asm-x86/mach-default/mach_reboot.h
+++ /dev/null
@@ -1,61 +0,0 @@
-/*
- *  arch/i386/mach-generic/mach_reboot.h
- *
- *  Machine specific reboot functions for generic.
- *  Split out from reboot.c by Osamu Tomita <tomita@cinet.co.jp>
- */
-#ifndef _MACH_REBOOT_H
-#define _MACH_REBOOT_H
-
-static inline void kb_wait(void)
-{
-	int i;
-
-	for (i = 0; i < 0x10000; i++)
-		if ((inb_p(0x64) & 0x02) == 0)
-			break;
-}
-
-static inline void mach_reboot(void)
-{
-	int i;
-
-	/* old method, works on most machines */
-	for (i = 0; i < 10; i++) {
-		kb_wait();
-		udelay(50);
-		outb(0xfe, 0x64);	/* pulse reset low */
-		udelay(50);
-	}
-
-	/* New method: sets the "System flag" which, when set, indicates
-	 * successful completion of the keyboard controller self-test (Basic
-	 * Assurance Test, BAT).  This is needed for some machines with no
-	 * keyboard plugged in.  This read-modify-write sequence sets only the
-	 * system flag
-	 */
-	for (i = 0; i < 10; i++) {
-		int cmd;
-
-		outb(0x20, 0x64);	/* read Controller Command Byte */
-		udelay(50);
-		kb_wait();
-		udelay(50);
-		cmd = inb(0x60);
-		udelay(50);
-		kb_wait();
-		udelay(50);
-		outb(0x60, 0x64);	/* write Controller Command Byte */
-		udelay(50);
-		kb_wait();
-		udelay(50);
-		outb(cmd | 0x14, 0x60); /* set "System flag" and "Keyboard Disabled" */
-		udelay(50);
-		kb_wait();
-		udelay(50);
-		outb(0xfe, 0x64);	/* pulse reset low */
-		udelay(50);
-	}
-}
-
-#endif /* !_MACH_REBOOT_H */
-- 
cgit v1.2.3


From 01aaea1afbcdb7c49fe4a567ebe3e295db9f720d Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Thu, 6 Mar 2008 13:46:39 -0800
Subject: x86: introduce initial apicid

store initial_apicid from early identify. it is could be different from
phys_proc_id later.

also print it out in /proc/cpuinfo.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/common.c | 13 ++++++-------
 arch/x86/kernel/cpu/proc.c   |  3 +++
 arch/x86/kernel/setup_64.c   | 11 ++++++-----
 include/asm-x86/processor.h  |  1 +
 4 files changed, 16 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 57a46c36fa23..0dd87b8d6707 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -369,10 +369,12 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
 			if (c->x86 >= 0x6)
 				c->x86_model += ((tfms >> 16) & 0xF) << 4;
 			c->x86_mask = tfms & 15;
+			c->initial_apicid = (ebx >> 24) & 0xFF;
 #ifdef CONFIG_X86_HT
-			c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0);
+			c->apicid = phys_pkg_id(c->initial_apicid, 0);
+			c->phys_proc_id = c->initial_apicid;
 #else
-			c->apicid = (ebx >> 24) & 0xFF;
+			c->apicid = c->initial_apicid;
 #endif
 			if (test_cpu_cap(c, X86_FEATURE_CLFLSH))
 				c->x86_clflush_size = ((ebx >> 8) & 0xff) * 8;
@@ -395,9 +397,6 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
 		init_scattered_cpuid_features(c);
 	}
 
-#ifdef CONFIG_X86_HT
-	c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff;
-#endif
 }
 
 static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
@@ -554,7 +553,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 		}
 
 		index_msb = get_count_order(smp_num_siblings);
-		c->phys_proc_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb);
+		c->phys_proc_id = phys_pkg_id(c->initial_apicid, index_msb);
 
 		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
 		       c->phys_proc_id);
@@ -565,7 +564,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 
 		core_bits = get_count_order(c->x86_max_cores);
 
-		c->cpu_core_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) &
+		c->cpu_core_id = phys_pkg_id(c->initial_apicid, index_msb) &
 					       ((1 << core_bits) - 1);
 
 		if (c->x86_max_cores > 1)
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index e8e58c09625d..0978a4a39418 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -19,6 +19,8 @@ static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c,
 			   cpus_weight(per_cpu(cpu_core_map, cpu)));
 		seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id);
 		seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
+		seq_printf(m, "apicid\t\t: %d\n", c->apicid);
+		seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid);
 	}
 #endif
 }
@@ -60,6 +62,7 @@ static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c,
 		seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id);
 		seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
 		seq_printf(m, "apicid\t\t: %d\n", c->apicid);
+		seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid);
 	}
 #endif
 }
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index f303c70dd688..13fe525bf065 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -562,9 +562,9 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
 	bits = c->x86_coreid_bits;
 
 	/* Low order bits define the core id (index of core in socket) */
-	c->cpu_core_id = c->phys_proc_id & ((1 << bits)-1);
-	/* Convert the APIC ID into the socket ID */
-	c->phys_proc_id = (c->apicid - boot_cpu_id) >> bits;
+	c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
+	/* Convert the initial APIC ID into the socket ID */
+	c->phys_proc_id = c->initial_apicid >> bits;
 
 #ifdef CONFIG_NUMA
 	node = c->phys_proc_id;
@@ -581,7 +581,7 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
 		   If that doesn't result in a usable node fall back to the
 		   path for the previous case.  */
 
-		int ht_nodeid = apicid - boot_cpu_id;
+		int ht_nodeid = c->initial_apicid;
 
 		if (ht_nodeid >= 0 &&
 		    apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
@@ -936,8 +936,9 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
 		c->x86 = 4;
 	}
 
+	c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff;
 #ifdef CONFIG_SMP
-	c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff;
+	c->phys_proc_id = c->initial_apicid;
 #endif
 	/* AMD-defined flags: level 0x80000001 */
 	xlvl = cpuid_eax(0x80000000);
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 1f9501a38493..d590da88c0df 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -101,6 +101,7 @@ struct cpuinfo_x86 {
 	/* cpuid returned max cores value: */
 	u16			 x86_max_cores;
 	u16			apicid;
+	u16			initial_apicid;
 	u16			x86_clflush_size;
 #ifdef CONFIG_SMP
 	/* number of cores as seen by the OS: */
-- 
cgit v1.2.3


From e40c0fe6b0b5dd16aec3c0dad311d36b19d78fd9 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 9 Mar 2008 12:35:00 -0700
Subject: x86: cleanup duplicate includes

Signed-off-by: Joe Perches <joe@perches.com>

 arch/x86/kernel/reboot.c      |    1 -
 include/asm-x86/elf.h         |    5 ++---
 include/asm-x86/posix_types.h |    8 +-------
 include/asm-x86/processor.h   |    3 +--
 include/asm-x86/unistd.h      |    8 +-------
 5 files changed, 5 insertions(+), 20 deletions(-)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/reboot.c      | 1 -
 include/asm-x86/elf.h         | 5 ++---
 include/asm-x86/posix_types.h | 8 +-------
 include/asm-x86/processor.h   | 3 +--
 include/asm-x86/unistd.h      | 8 +-------
 5 files changed, 5 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 484c4a80d38a..66cd4afc1e57 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -1,5 +1,4 @@
 #include <linux/module.h>
-#include <linux/init.h>
 #include <linux/reboot.h>
 #include <linux/init.h>
 #include <linux/pm.h>
diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index fb62f9941e38..77325646e2f4 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -82,8 +82,9 @@ extern unsigned int vdso_enabled;
 #define elf_check_arch_ia32(x) \
 	(((x)->e_machine == EM_386) || ((x)->e_machine == EM_486))
 
-#ifdef CONFIG_X86_32
 #include <asm/processor.h>
+
+#ifdef CONFIG_X86_32
 #include <asm/system.h>		/* for savesegment */
 #include <asm/desc.h>
 
@@ -135,8 +136,6 @@ extern unsigned int vdso_enabled;
 
 #else /* CONFIG_X86_32 */
 
-#include <asm/processor.h>
-
 /*
  * This is used to ensure we don't load something for the wrong architecture.
  */
diff --git a/include/asm-x86/posix_types.h b/include/asm-x86/posix_types.h
index bb7133dc155d..fe312a5ba204 100644
--- a/include/asm-x86/posix_types.h
+++ b/include/asm-x86/posix_types.h
@@ -1,11 +1,5 @@
 #ifdef __KERNEL__
-# ifdef CONFIG_X86_32
-#  include "posix_types_32.h"
-# else
-#  include "posix_types_64.h"
-# endif
-#else
-# ifdef __i386__
+# if defined(CONFIG_X86_32) || defined(__i386__)
 #  include "posix_types_32.h"
 # else
 #  include "posix_types_64.h"
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index d590da88c0df..cc0268395ea2 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -3,8 +3,7 @@
 
 #include <asm/processor-flags.h>
 
-/* migration helpers, for KVM - will be removed in 2.6.25: */
-#include <asm/vm86.h>
+/* migration helper, for KVM - will be removed in 2.6.25: */
 #define Xgt_desc_struct	desc_ptr
 
 /* Forward declaration, a strange C thing */
diff --git a/include/asm-x86/unistd.h b/include/asm-x86/unistd.h
index 2a58ed3e51d8..effc7ad8e12f 100644
--- a/include/asm-x86/unistd.h
+++ b/include/asm-x86/unistd.h
@@ -1,11 +1,5 @@
 #ifdef __KERNEL__
-# ifdef CONFIG_X86_32
-#  include "unistd_32.h"
-# else
-#  include "unistd_64.h"
-# endif
-#else
-# ifdef __i386__
+# if defined(CONFIG_X86_32) || defined(__i386__)
 #  include "unistd_32.h"
 # else
 #  include "unistd_64.h"
-- 
cgit v1.2.3


From 86975101e46ec93be972d8f46715aa6273102545 Mon Sep 17 00:00:00 2001
From: stephane eranian <eranian@googlemail.com>
Date: Fri, 7 Mar 2008 13:05:27 -0800
Subject: x86: add cpu_has_arch_perfmon

adds cpu_has_arch_perfmon to test presence of architectural perfmon on
Intel x86 processor

Signed-off-by: Stephane Eranian <eranian@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/cpufeature.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 1e3102eeb823..90feb6f2562c 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -185,6 +185,7 @@ extern const char * const x86_power_flags[32];
 #define cpu_has_clflush		boot_cpu_has(X86_FEATURE_CLFLSH)
 #define cpu_has_bts		boot_cpu_has(X86_FEATURE_BTS)
 #define cpu_has_gbpages		boot_cpu_has(X86_FEATURE_GBPAGES)
+#define cpu_has_arch_perfmon	boot_cpu_has(X86_FEATURE_ARCH_PERFMON)
 
 #if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64)
 # define cpu_has_invlpg		1
-- 
cgit v1.2.3


From 12db648c1518b2627cc983199a97ec6f5d6a1de2 Mon Sep 17 00:00:00 2001
From: stephane eranian <eranian@googlemail.com>
Date: Fri, 7 Mar 2008 13:05:39 -0800
Subject: x86: add AMD Northbridge MSR definition

adds AMD Northbridge config MSR definition

Signed-off-by: Stephane Eranian <eranian@gmail.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/msr-index.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h
index fae118a25278..3ed97144c07b 100644
--- a/include/asm-x86/msr-index.h
+++ b/include/asm-x86/msr-index.h
@@ -83,6 +83,7 @@
 /* AMD64 MSRs. Not complete. See the architecture manual for a more
    complete list. */
 
+#define MSR_AMD64_NB_CFG		0xc001001f
 #define MSR_AMD64_IBSFETCHCTL		0xc0011030
 #define MSR_AMD64_IBSFETCHLINAD		0xc0011031
 #define MSR_AMD64_IBSFETCHPHYSAD	0xc0011032
-- 
cgit v1.2.3


From 5b0e508415a83989fe704b4718a1a214bc333ca7 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Mon, 10 Mar 2008 13:11:17 +0000
Subject: x86: prevent unconditional writes to DebugCtl MSR

Otherwise, enabling (or better, subsequent disabling) of single
stepping would cause a kernel oops on CPUs not having this MSR.

The patch could have been added a conditional to the MSR write in
user_disable_single_step(), but centralizing the updates seems safer
and (looking forward) better manageable.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Markus Metzger <markus.t.metzger@intel.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/kprobes.c    | 4 ++--
 arch/x86/kernel/process_32.c | 4 ++--
 arch/x86/kernel/process_64.c | 4 ++--
 arch/x86/kernel/step.c       | 2 +-
 include/asm-x86/processor.h  | 9 +++++++++
 5 files changed, 16 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 34a591283f5d..1e3de7db9ad5 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -410,13 +410,13 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 static void __kprobes clear_btf(void)
 {
 	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, 0);
+		update_debugctlmsr(0);
 }
 
 static void __kprobes restore_btf(void)
 {
 	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, current->thread.debugctlmsr);
+		update_debugctlmsr(current->thread.debugctlmsr);
 }
 
 static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 9230ce060d09..ec05fb750dfc 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -564,12 +564,12 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		/* we clear debugctl to make sure DS
 		 * is not in use when we change it */
 		debugctl = 0;
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, 0);
+		update_debugctlmsr(0);
 		wrmsr(MSR_IA32_DS_AREA, next->ds_area_msr, 0);
 	}
 
 	if (next->debugctlmsr != debugctl)
-		wrmsr(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr, 0);
+		update_debugctlmsr(next->debugctlmsr);
 
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
 		set_debugreg(next->debugreg0, 0);
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 1ffce14cff6e..4f40272474dd 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -563,12 +563,12 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 		/* we clear debugctl to make sure DS
 		 * is not in use when we change it */
 		debugctl = 0;
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, 0);
+		update_debugctlmsr(0);
 		wrmsrl(MSR_IA32_DS_AREA, next->ds_area_msr);
 	}
 
 	if (next->debugctlmsr != debugctl)
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, next->debugctlmsr);
+		update_debugctlmsr(next->debugctlmsr);
 
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
 		loaddebug(next, 0);
diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index 071ff4798236..92c20fee6781 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -148,7 +148,7 @@ static void write_debugctlmsr(struct task_struct *child, unsigned long val)
 	if (child != current)
 		return;
 
-	wrmsrl(MSR_IA32_DEBUGCTLMSR, val);
+	update_debugctlmsr(val);
 }
 
 /*
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index cc0268395ea2..40227c9bf51b 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -741,6 +741,15 @@ extern void switch_to_new_gdt(void);
 extern void cpu_init(void);
 extern void init_gdt(int cpu);
 
+static inline void update_debugctlmsr(unsigned long debugctlmsr)
+{
+#ifndef CONFIG_X86_DEBUGCTLMSR
+	if (boot_cpu_data.x86 < 6)
+		return;
+#endif
+	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr);
+}
+
 /*
  * from system description table in BIOS. Mostly for MCA use, but
  * others may find it useful:
-- 
cgit v1.2.3


From 5d570cbbf25a62e9c077f5b351fb142dbfc67288 Mon Sep 17 00:00:00 2001
From: Mikael Pettersson <mikpe@it.uu.se>
Date: Tue, 11 Mar 2008 16:43:31 +0100
Subject: x86: correct/clarify comment in nops.h

<asm-x86/nops.h> describes certain multibyte instructions as
"generic" nops when in fact they aren't nops at all in 64-bit
mode (missing REX.W causing truncation of a register).

Update the comment to state that K8 or P6 style nops should be
used in 64-bit mode. This matches what the alternatives code does.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/nops.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/nops.h b/include/asm-x86/nops.h
index b3930ae539b3..ad0bedd10b89 100644
--- a/include/asm-x86/nops.h
+++ b/include/asm-x86/nops.h
@@ -5,6 +5,8 @@
 
 /* generic versions from gas
    1: nop
+   the following instructions are NOT nops in 64-bit mode,
+   for 64-bit mode use K8 or P6 nops instead
    2: movl %esi,%esi
    3: leal 0x00(%esi),%esi
    4: leal 0x00(,%esi,1),%esi
-- 
cgit v1.2.3


From 6079d2d5d11122eb52721f0f3c828952a490e6c1 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Tue, 11 Mar 2008 19:45:48 +0300
Subject: x86: move quad_local_to_mp_bus_id to numa.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c              | 1 -
 arch/x86/pci/numa.c                       | 2 ++
 include/asm-x86/mach-numaq/mach_mpparse.h | 2 ++
 include/asm-x86/mpspec.h                  | 1 -
 4 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index b2aded3fbfec..2abc9392799a 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -45,7 +45,6 @@ int apic_version [MAX_APICS];
 int mp_bus_id_to_type [MAX_MP_BUSSES];
 int mp_bus_id_to_node [MAX_MP_BUSSES];
 int mp_bus_id_to_local [MAX_MP_BUSSES];
-int quad_local_to_mp_bus_id [NR_CPUS/4][4];
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
 static int mp_current_pci_id;
 
diff --git a/arch/x86/pci/numa.c b/arch/x86/pci/numa.c
index 55270c26237c..3248d3ee388f 100644
--- a/arch/x86/pci/numa.c
+++ b/arch/x86/pci/numa.c
@@ -13,6 +13,8 @@
 
 #define BUS2QUAD(global) (mp_bus_id_to_node[global])
 #define BUS2LOCAL(global) (mp_bus_id_to_local[global])
+
+int quad_local_to_mp_bus_id [NR_CPUS/4][4];
 #define QUADLOCAL2BUS(quad,local) (quad_local_to_mp_bus_id[quad][local])
 
 extern void *xquad_portio;    /* Where the IO area was mapped */
diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h
index 51bbac8fc0c2..cd21f289e6b7 100644
--- a/include/asm-x86/mach-numaq/mach_mpparse.h
+++ b/include/asm-x86/mach-numaq/mach_mpparse.h
@@ -12,6 +12,8 @@ static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
 	printk("Bus #%d is %s (node %d)\n", m->mpc_busid, name, quad);
 }
 
+extern int quad_local_to_mp_bus_id[NR_CPUS/4][4];
+
 static inline void mpc_oem_pci_bus(struct mpc_config_bus *m, 
 				struct mpc_config_translation *translation)
 {
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 781ad74ab9e9..dbd63f8d4750 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -9,7 +9,6 @@
 extern int mp_bus_id_to_type[MAX_MP_BUSSES];
 extern int mp_bus_id_to_node[MAX_MP_BUSSES];
 extern int mp_bus_id_to_local[MAX_MP_BUSSES];
-extern int quad_local_to_mp_bus_id[NR_CPUS/4][4];
 
 extern unsigned int def_to_bigsmp;
 extern int apic_version[MAX_APICS];
-- 
cgit v1.2.3


From 8643f9d02a7bb9db74634b4c062d8e70ce7c59b9 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <Yinghai.Lu@Sun.COM>
Date: Tue, 19 Feb 2008 03:21:06 -0800
Subject: x86: get boot_cpu_id as early for k8_scan_nodes

When acpi=off or there is no SRAT defined, apicid_to_node is got from K8
Northbridge PCI configuration space in k8_scan_nodes() in
arch/x86_64/mm/k8toplogy.c.

The problem is that it assumes bsp apic id is 0 at that point.

For four socket system with Quad core cpus installed, all cpus apic id
is offset by 4, and bsp apic id is 4.

For eight socket system with dual core cpus installed, all cpus apic id
is offset by 2, and bsp apic id is 2.

We need get boot_cpu_id --- bsp apic id, before k8_scan_nodes by called.

So create early_acpi_boot_init and early_get_smp_config for get boot_cpu_id.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c    | 24 ++++++++++++
 arch/x86/kernel/mpparse_64.c | 89 ++++++++++++++++++++++++++++++++------------
 include/asm-x86/apic.h       |  1 +
 include/asm-x86/mpspec.h     |  3 ++
 4 files changed, 94 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index f6eb01d8923a..8a475793f736 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -861,6 +861,30 @@ static int __init detect_init_APIC(void)
 	return 0;
 }
 
+void __init early_init_lapic_mapping(void)
+{
+	unsigned long apic_phys;
+
+	/*
+	 * If no local APIC can be found then go out
+	 * : it means there is no mpatable and MADT
+	 */
+	if (!smp_found_config)
+		return;
+
+	apic_phys = mp_lapic_addr;
+
+	set_fixmap_nocache(FIX_APIC_BASE, apic_phys);
+	apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n",
+				 APIC_BASE, apic_phys);
+
+	/*
+	 * Fetch the APIC ID of the BSP in case we have a
+	 * default configuration (or the MP table is broken).
+	 */
+	boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
+}
+
 /**
  * init_apic_mappings - initialize APIC mappings
  */
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 2a1f7881c75b..529b1c22077e 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -224,8 +224,7 @@ static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
 /*
  * Read/parse the MPC
  */
-
-static int __init smp_read_mpc(struct mp_config_table *mpc)
+static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early)
 {
 	char str[16];
 	int count=sizeof(*mpc);
@@ -266,6 +265,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 	if (!acpi_lapic)
 		mp_lapic_addr = mpc->mpc_lapic;
 
+	if (early)
+		return 1;
+
 	/*
 	 *	Now process the configuration blocks.
 	 */
@@ -477,27 +479,38 @@ static struct intel_mp_floating *mpf_found;
 /*
  * Scan the memory blocks for an SMP configuration block.
  */
-void __init get_smp_config (void)
+static void __init __get_smp_config(unsigned early)
 {
 	struct intel_mp_floating *mpf = mpf_found;
 
+	if (acpi_lapic && early)
+		return;
 	/*
- 	 * ACPI supports both logical (e.g. Hyper-Threading) and physical 
- 	 * processors, where MPS only supports physical.
- 	 */
- 	if (acpi_lapic && acpi_ioapic) {
- 		printk(KERN_INFO "Using ACPI (MADT) for SMP configuration information\n");
- 		return;
-	}
- 	else if (acpi_lapic)
- 		printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n");
+	 * ACPI supports both logical (e.g. Hyper-Threading) and physical
+	 * processors, where MPS only supports physical.
+	 */
+	if (acpi_lapic && acpi_ioapic) {
+		printk(KERN_INFO "Using ACPI (MADT) for SMP configuration "
+			"information\n");
+		return;
+	} else if (acpi_lapic)
+		printk(KERN_INFO "Using ACPI for processor (LAPIC) "
+			"configuration information\n");
 
-	printk("Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification);
+	printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n",
+		mpf->mpf_specification);
 
 	/*
 	 * Now see if we need to read further.
 	 */
 	if (mpf->mpf_feature1 != 0) {
+		if (early) {
+			/*
+			 * local APIC has default address
+			 */
+			mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
+			return;
+		}
 
 		printk(KERN_INFO "Default MP configuration #%d\n", mpf->mpf_feature1);
 		construct_default_ISA_mptable(mpf->mpf_feature1);
@@ -508,12 +521,15 @@ void __init get_smp_config (void)
 		 * Read the physical hardware table.  Anything here will
 		 * override the defaults.
 		 */
-		if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr))) {
+		if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr), early)) {
 			smp_found_config = 0;
 			printk(KERN_ERR "BIOS bug, MP table errors detected!...\n");
 			printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");
 			return;
 		}
+
+		if (early)
+			return;
 		/*
 		 * If there are no explicit MP IRQ entries, then we are
 		 * broken.  We set up most of the low 16 IO-APIC pins to
@@ -535,13 +551,25 @@ void __init get_smp_config (void)
 	} else
 		BUG();
 
-	printk(KERN_INFO "Processors: %d\n", num_processors);
+	if (!early)
+		printk(KERN_INFO "Processors: %d\n", num_processors);
 	/*
 	 * Only use the first configuration found.
 	 */
 }
 
-static int __init smp_scan_config (unsigned long base, unsigned long length)
+void __init early_get_smp_config(void)
+{
+	__get_smp_config(1);
+}
+
+void __init get_smp_config(void)
+{
+	__get_smp_config(0);
+}
+
+static int __init smp_scan_config(unsigned long base, unsigned long length,
+				  unsigned reserve)
 {
 	extern void __bad_mpf_size(void); 
 	unsigned int *bp = phys_to_virt(base);
@@ -560,10 +588,15 @@ static int __init smp_scan_config (unsigned long base, unsigned long length)
 				|| (mpf->mpf_specification == 4)) ) {
 
 			smp_found_config = 1;
+			mpf_found = mpf;
+
+			if (!reserve)
+				return 1;
+
 			reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE);
 			if (mpf->mpf_physptr)
-				reserve_bootmem_generic(mpf->mpf_physptr, PAGE_SIZE);
-			mpf_found = mpf;
+				reserve_bootmem_generic(mpf->mpf_physptr,
+							PAGE_SIZE);
 			return 1;
 		}
 		bp += 4;
@@ -572,7 +605,7 @@ static int __init smp_scan_config (unsigned long base, unsigned long length)
 	return 0;
 }
 
-void __init find_smp_config(void)
+static void __init __find_smp_config(unsigned reserve)
 {
 	unsigned int address;
 
@@ -584,9 +617,9 @@ void __init find_smp_config(void)
 	 * 2) Scan the top 1K of base RAM
 	 * 3) Scan the 64K of bios
 	 */
-	if (smp_scan_config(0x0,0x400) ||
-		smp_scan_config(639*0x400,0x400) ||
-			smp_scan_config(0xF0000,0x10000))
+	if (smp_scan_config(0x0, 0x400, reserve) ||
+		smp_scan_config(639*0x400, 0x400, reserve) ||
+			smp_scan_config(0xF0000, 0x10000, reserve))
 		return;
 	/*
 	 * If it is an SMP machine we should know now.
@@ -603,13 +636,23 @@ void __init find_smp_config(void)
 
 	address = *(unsigned short *)phys_to_virt(0x40E);
 	address <<= 4;
-	if (smp_scan_config(address, 0x1000))
+	if (smp_scan_config(address, 0x1000, reserve))
 		return;
 
 	/* If we have come this far, we did not find an MP table  */
 	 printk(KERN_INFO "No mptable found.\n");
 }
 
+void __init early_find_smp_config(void)
+{
+	__find_smp_config(0);
+}
+
+void __init find_smp_config(void)
+{
+	__find_smp_config(1);
+}
+
 /* --------------------------------------------------------------------------
                             ACPI-based MP Configuration
    -------------------------------------------------------------------------- */
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index f0321a427e16..db5f7501aed6 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -129,6 +129,7 @@ extern void enable_NMI_through_LVT0(void);
  */
 #ifdef CONFIG_X86_64
 extern void setup_apic_routing(void);
+extern void early_init_lapic_mapping(void);
 #endif
 
 extern u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask);
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index dbd63f8d4750..982550bef2cd 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -25,6 +25,9 @@ extern int pic_mode;
 
 extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 
+extern void early_find_smp_config(void);
+extern void early_get_smp_config(void);
+
 #endif
 
 extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
-- 
cgit v1.2.3


From a6333c3ccbdc0ae001cff6ee1d3633942ef763f4 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Thu, 20 Mar 2008 14:54:09 +0300
Subject: x86: add mp_bus_not_pci bitmap to mpparse_32.c

Signed-off: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c | 4 ++++
 include/asm-x86/mpspec.h     | 3 +--
 2 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 2abc9392799a..c13092db578e 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -43,6 +43,7 @@ unsigned int __cpuinitdata maxcpus = NR_CPUS;
  */
 int apic_version [MAX_APICS];
 int mp_bus_id_to_type [MAX_MP_BUSSES];
+DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 int mp_bus_id_to_node [MAX_MP_BUSSES];
 int mp_bus_id_to_local [MAX_MP_BUSSES];
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
@@ -240,12 +241,14 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 	}
 #endif
 
+	set_bit(m->mpc_busid, mp_bus_not_pci);
 	if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA)-1) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
 	} else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA)-1) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
 	} else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI)-1) == 0) {
 		mpc_oem_pci_bus(m, translation_table[mpc_record]);
+		clear_bit(m->mpc_busid, mp_bus_not_pci);
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
 		mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
 		mp_current_pci_id++;
@@ -984,6 +987,7 @@ void __init mp_config_acpi_legacy_irqs (void)
 	 * Fabricate the legacy ISA bus (bus #31).
 	 */
 	mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
+	set_bit(MP_ISA_BUS, mp_bus_not_pci);
 	Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
 
 	/*
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 982550bef2cd..75df88e0a270 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -23,13 +23,12 @@ extern int pic_mode;
 /* Each PCI slot may be a combo card with its own bus.  4 IRQ pins per slot. */
 #define MAX_IRQ_SOURCES (MAX_MP_BUSSES * 4)
 
-extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
-
 extern void early_find_smp_config(void);
 extern void early_get_smp_config(void);
 
 #endif
 
+extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
 
 extern unsigned int boot_cpu_physical_apicid;
-- 
cgit v1.2.3


From c0a282c251181aa423d4831719613b8286b5b839 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Thu, 20 Mar 2008 14:55:02 +0300
Subject: x86: make mp_bus_id_to_type optional

[ mingo@elte.hu: fix boot regression. ]

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/io_apic_32.c | 4 ++++
 arch/x86/kernel/mpparse_32.c | 8 +++++++-
 include/asm-x86/mpspec.h     | 6 +++++-
 3 files changed, 16 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 97ac6104d63a..0d70acd3b134 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -915,6 +915,7 @@ void __init setup_ioapic_dest(void)
 }
 #endif
 
+#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
 /*
  * EISA Edge/Level control register, ELCR
  */
@@ -928,6 +929,7 @@ static int EISA_ELCR(unsigned int irq)
 			"Broken MPtable reports ISA irq %d\n", irq);
 	return 0;
 }
+#endif
 
 /* ISA interrupts are always polarity zero edge triggered,
  * when listed as conforming in the MP table. */
@@ -1013,6 +1015,7 @@ static int MPBIOS_trigger(int idx)
 			trigger = test_bit(bus, mp_bus_not_pci)?
 					default_ISA_trigger(idx):
 					default_PCI_trigger(idx);
+#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
 			switch (mp_bus_id_to_type[bus])
 			{
 				case MP_BUS_ISA: /* ISA pin */
@@ -1042,6 +1045,7 @@ static int MPBIOS_trigger(int idx)
 					break;
 				}
 			}
+#endif
 			break;
 		}
 		case 1: /* edge */
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 13adcc40a00e..8795122db9bc 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -42,7 +42,9 @@ unsigned int __cpuinitdata maxcpus = NR_CPUS;
  * MP-table.
  */
 int apic_version [MAX_APICS];
+#if defined (CONFIG_MCA) || defined (CONFIG_EISA)
 int mp_bus_id_to_type [MAX_MP_BUSSES];
+#endif
 DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 int mp_bus_id_to_node [MAX_MP_BUSSES];
 int mp_bus_id_to_local [MAX_MP_BUSSES];
@@ -245,9 +247,10 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 	if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI)-1) == 0) {
 		mpc_oem_pci_bus(m, translation_table[mpc_record]);
 		clear_bit(m->mpc_busid, mp_bus_not_pci);
-		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
 		mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
 		mp_current_pci_id++;
+#if defined(CONFIG_EISA) || defined (CONFIG_MCA)
+		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
 	} else if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA)-1) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
 	} else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA)-1) == 0) {
@@ -256,6 +259,7 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA;
 	} else {
 		printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str);
+#endif
 	}
 }
 
@@ -983,10 +987,12 @@ void __init mp_config_acpi_legacy_irqs (void)
 	int i = 0;
 	int ioapic = -1;
 
+#if defined (CONFIG_MCA) || defined (CONFIG_EISA)
 	/* 
 	 * Fabricate the legacy ISA bus (bus #31).
 	 */
 	mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
+#endif
 	set_bit(MP_ISA_BUS, mp_bus_not_pci);
 	Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
 
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 75df88e0a270..5fa8b8bb7d00 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -6,7 +6,6 @@
 #ifdef CONFIG_X86_32
 #include <mach_mpspec.h>
 
-extern int mp_bus_id_to_type[MAX_MP_BUSSES];
 extern int mp_bus_id_to_node[MAX_MP_BUSSES];
 extern int mp_bus_id_to_local[MAX_MP_BUSSES];
 
@@ -28,7 +27,12 @@ extern void early_get_smp_config(void);
 
 #endif
 
+#if defined(CONFIG_MCA) || defined(CONFIG_EISA)
+extern int mp_bus_id_to_type[MAX_MP_BUSSES];
+#endif
+
 extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
+
 extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
 
 extern unsigned int boot_cpu_physical_apicid;
-- 
cgit v1.2.3


From e129cb490e842753b43af7aae136935fc0928dc8 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Tue, 11 Mar 2008 22:55:42 +0300
Subject: x86: move mp_bus_id_to_local to numa.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c              | 1 -
 arch/x86/pci/numa.c                       | 2 ++
 include/asm-x86/mach-numaq/mach_mpparse.h | 2 ++
 include/asm-x86/mpspec.h                  | 1 -
 4 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 8795122db9bc..ae385b427841 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -47,7 +47,6 @@ int mp_bus_id_to_type [MAX_MP_BUSSES];
 #endif
 DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 int mp_bus_id_to_node [MAX_MP_BUSSES];
-int mp_bus_id_to_local [MAX_MP_BUSSES];
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
 static int mp_current_pci_id;
 
diff --git a/arch/x86/pci/numa.c b/arch/x86/pci/numa.c
index 3248d3ee388f..e765558f7e1b 100644
--- a/arch/x86/pci/numa.c
+++ b/arch/x86/pci/numa.c
@@ -12,6 +12,8 @@
 #define XQUAD_PORTIO_QUAD 0x40000  /* 256k per quad. */
 
 #define BUS2QUAD(global) (mp_bus_id_to_node[global])
+
+int mp_bus_id_to_local[MAX_MP_BUSSES];
 #define BUS2LOCAL(global) (mp_bus_id_to_local[global])
 
 int quad_local_to_mp_bus_id [NR_CPUS/4][4];
diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h
index cd21f289e6b7..0917b95c42ac 100644
--- a/include/asm-x86/mach-numaq/mach_mpparse.h
+++ b/include/asm-x86/mach-numaq/mach_mpparse.h
@@ -1,6 +1,8 @@
 #ifndef __ASM_MACH_MPPARSE_H
 #define __ASM_MACH_MPPARSE_H
 
+extern int mp_bus_id_to_local[MAX_MP_BUSSES];
+
 static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name, 
 				struct mpc_config_translation *translation)
 {
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 5fa8b8bb7d00..1c52bd8711f6 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -7,7 +7,6 @@
 #include <mach_mpspec.h>
 
 extern int mp_bus_id_to_node[MAX_MP_BUSSES];
-extern int mp_bus_id_to_local[MAX_MP_BUSSES];
 
 extern unsigned int def_to_bigsmp;
 extern int apic_version[MAX_APICS];
-- 
cgit v1.2.3


From 037cab07e9515149fecc2274775807f06ea6b036 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Tue, 11 Mar 2008 22:55:48 +0300
Subject: x86: move mp_bus_id_to_node to numa.c

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c              | 1 -
 arch/x86/kernel/summit_32.c               | 2 ++
 arch/x86/pci/numa.c                       | 1 +
 include/asm-x86/mach-numaq/mach_mpparse.h | 1 +
 include/asm-x86/mpspec.h                  | 2 --
 5 files changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index ae385b427841..1b1fd6e920e6 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -46,7 +46,6 @@ int apic_version [MAX_APICS];
 int mp_bus_id_to_type [MAX_MP_BUSSES];
 #endif
 DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
-int mp_bus_id_to_node [MAX_MP_BUSSES];
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
 static int mp_current_pci_id;
 
diff --git a/arch/x86/kernel/summit_32.c b/arch/x86/kernel/summit_32.c
index 30f04c3e68e4..6878a9c2df5d 100644
--- a/arch/x86/kernel/summit_32.c
+++ b/arch/x86/kernel/summit_32.c
@@ -35,6 +35,8 @@ static struct rio_table_hdr *rio_table_hdr __initdata;
 static struct scal_detail   *scal_devs[MAX_NUMNODES] __initdata;
 static struct rio_detail    *rio_devs[MAX_NUMNODES*4] __initdata;
 
+static int mp_bus_id_to_node[MAX_MP_BUSSES] __initdata;
+
 static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
 {
 	int twister = 0, node = 0;
diff --git a/arch/x86/pci/numa.c b/arch/x86/pci/numa.c
index e765558f7e1b..9e30378364ed 100644
--- a/arch/x86/pci/numa.c
+++ b/arch/x86/pci/numa.c
@@ -11,6 +11,7 @@
 #define XQUAD_PORTIO_BASE 0xfe400000
 #define XQUAD_PORTIO_QUAD 0x40000  /* 256k per quad. */
 
+int mp_bus_id_to_node[MAX_MP_BUSSES];
 #define BUS2QUAD(global) (mp_bus_id_to_node[global])
 
 int mp_bus_id_to_local[MAX_MP_BUSSES];
diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h
index 0917b95c42ac..254993ae04c7 100644
--- a/include/asm-x86/mach-numaq/mach_mpparse.h
+++ b/include/asm-x86/mach-numaq/mach_mpparse.h
@@ -2,6 +2,7 @@
 #define __ASM_MACH_MPPARSE_H
 
 extern int mp_bus_id_to_local[MAX_MP_BUSSES];
+extern int mp_bus_id_to_node[MAX_MP_BUSSES];
 
 static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name, 
 				struct mpc_config_translation *translation)
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 1c52bd8711f6..99da0a59b436 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -6,8 +6,6 @@
 #ifdef CONFIG_X86_32
 #include <mach_mpspec.h>
 
-extern int mp_bus_id_to_node[MAX_MP_BUSSES];
-
 extern unsigned int def_to_bigsmp;
 extern int apic_version[MAX_APICS];
 extern u8 apicid_2_node[];
-- 
cgit v1.2.3


From d0173aeac4f7fa90a63319b817bd207bdc0ac87e Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:24:59 -0300
Subject: x86: use start_ipi_hook in x86_64

It is used to match i386. The definition for the non-paravirt
case is moved to smp.h instead of smp_32.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_64.c | 8 ++++++++
 include/asm-x86/smp.h        | 3 +++
 include/asm-x86/smp_32.h     | 4 ----
 3 files changed, 11 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 57ebe6c04305..13ab1123d1d2 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -344,6 +344,14 @@ static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int sta
 
 	num_starts = 2;
 
+	/*
+	 * Paravirt / VMI wants a startup IPI hook here to set up the
+	 * target processor state.
+	 */
+	startup_ipi_hook(phys_apicid, (unsigned long) start_secondary,
+			(unsigned long) init_rsp);
+
+
 	/*
 	 * Run STARTUP IPI loop.
 	 */
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 513c8571a4a0..4dc271b43767 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -33,6 +33,9 @@ struct smp_ops {
 extern void set_cpu_sibling_map(int cpu);
 
 #ifdef CONFIG_SMP
+#ifndef CONFIG_PARAVIRT
+#define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
+#endif
 extern struct smp_ops smp_ops;
 
 static inline void smp_send_stop(void)
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 4fec2feb6ac8..76740def6092 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -30,10 +30,6 @@ DECLARE_PER_CPU(u16, cpu_llc_id);
 DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
 
 #ifdef CONFIG_SMP
-#ifndef CONFIG_PARAVIRT
-#define startup_ipi_hook(phys_apicid, start_eip, start_esp) do { } while (0)
-#endif
-
 /*
  * This function is needed by all SMP systems. It must _always_ be valid
  * from the initial startup. We map APIC_BASE very early in page_setup(),
-- 
cgit v1.2.3


From 1d89a7f072d4f76f0538edfb474d527066ee7838 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:05 -0300
Subject: x86: merge smp_store_cpu_info

now that it is the same between arches, put it into smpboot.c

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot.c    | 77 ++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/smpboot_32.c | 71 +---------------------------------------
 arch/x86/kernel/smpboot_64.c | 15 ---------
 include/asm-x86/smp.h        |  2 ++
 include/asm-x86/smp_32.h     |  2 --
 5 files changed, 80 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index b13b9d55f9ce..a157a5245923 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -45,6 +45,83 @@ unsigned char *trampoline_base = __va(SMP_TRAMPOLINE_BASE);
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
 
+#ifdef CONFIG_X86_32
+/* Set if we find a B stepping CPU */
+int __cpuinitdata smp_b_stepping;
+#endif
+
+static void __cpuinit smp_apply_quirks(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_32
+	/*
+	 * Mask B, Pentium, but not Pentium MMX
+	 */
+	if (c->x86_vendor == X86_VENDOR_INTEL &&
+	    c->x86 == 5 &&
+	    c->x86_mask >= 1 && c->x86_mask <= 4 &&
+	    c->x86_model <= 3)
+		/*
+		 * Remember we have B step Pentia with bugs
+		 */
+		smp_b_stepping = 1;
+
+	/*
+	 * Certain Athlons might work (for various values of 'work') in SMP
+	 * but they are not certified as MP capable.
+	 */
+	if ((c->x86_vendor == X86_VENDOR_AMD) && (c->x86 == 6)) {
+
+		if (num_possible_cpus() == 1)
+			goto valid_k7;
+
+		/* Athlon 660/661 is valid. */
+		if ((c->x86_model == 6) && ((c->x86_mask == 0) ||
+		    (c->x86_mask == 1)))
+			goto valid_k7;
+
+		/* Duron 670 is valid */
+		if ((c->x86_model == 7) && (c->x86_mask == 0))
+			goto valid_k7;
+
+		/*
+		 * Athlon 662, Duron 671, and Athlon >model 7 have capability
+		 * bit. It's worth noting that the A5 stepping (662) of some
+		 * Athlon XP's have the MP bit set.
+		 * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for
+		 * more.
+		 */
+		if (((c->x86_model == 6) && (c->x86_mask >= 2)) ||
+		    ((c->x86_model == 7) && (c->x86_mask >= 1)) ||
+		     (c->x86_model > 7))
+			if (cpu_has_mp)
+				goto valid_k7;
+
+		/* If we get here, not a certified SMP capable AMD system. */
+		add_taint(TAINT_UNSAFE_SMP);
+	}
+
+valid_k7:
+	;
+#endif
+}
+
+/*
+ * The bootstrap kernel entry code has set these up. Save them for
+ * a given CPU
+ */
+
+void __cpuinit smp_store_cpu_info(int id)
+{
+	struct cpuinfo_x86 *c = &cpu_data(id);
+
+	*c = boot_cpu_data;
+	c->cpu_index = id;
+	if (id != 0)
+		identify_secondary_cpu(c);
+	smp_apply_quirks(c);
+}
+
+
 void __cpuinit set_cpu_sibling_map(int cpu)
 {
 	int i;
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index e05006416d8c..0bfb31e13540 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -59,8 +59,7 @@
 #include <asm/vmi.h>
 #include <asm/mtrr.h>
 
-/* Set if we find a B stepping CPU */
-static int __cpuinitdata smp_b_stepping;
+extern int smp_b_stepping;
 
 static cpumask_t smp_commenced_mask;
 
@@ -78,74 +77,6 @@ static void map_cpu_to_logical_apicid(void);
 /* State of each CPU. */
 DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
-static void __cpuinit smp_apply_quirks(struct cpuinfo_x86 *c)
-{
-	/*
-	 * Mask B, Pentium, but not Pentium MMX
-	 */
-	if (c->x86_vendor == X86_VENDOR_INTEL &&
-	    c->x86 == 5 &&
-	    c->x86_mask >= 1 && c->x86_mask <= 4 &&
-	    c->x86_model <= 3)
-		/*
-		 * Remember we have B step Pentia with bugs
-		 */
-		smp_b_stepping = 1;
-
-	/*
-	 * Certain Athlons might work (for various values of 'work') in SMP
-	 * but they are not certified as MP capable.
-	 */
-	if ((c->x86_vendor == X86_VENDOR_AMD) && (c->x86 == 6)) {
-
-		if (num_possible_cpus() == 1)
-			goto valid_k7;
-
-		/* Athlon 660/661 is valid. */	
-		if ((c->x86_model==6) && ((c->x86_mask==0) || (c->x86_mask==1)))
-			goto valid_k7;
-
-		/* Duron 670 is valid */
-		if ((c->x86_model==7) && (c->x86_mask==0))
-			goto valid_k7;
-
-		/*
-		 * Athlon 662, Duron 671, and Athlon >model 7 have capability bit.
-		 * It's worth noting that the A5 stepping (662) of some Athlon XP's
-		 * have the MP bit set.
-		 * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for more.
-		 */
-		if (((c->x86_model==6) && (c->x86_mask>=2)) ||
-		    ((c->x86_model==7) && (c->x86_mask>=1)) ||
-		     (c->x86_model> 7))
-			if (cpu_has_mp)
-				goto valid_k7;
-
-		/* If we get here, it's not a certified SMP capable AMD system. */
-		add_taint(TAINT_UNSAFE_SMP);
-	}
-
-valid_k7:
-	;
-
-}
-
-/*
- * The bootstrap kernel entry code has set these up. Save them for
- * a given CPU
- */
-
-void __cpuinit smp_store_cpu_info(int id)
-{
-	struct cpuinfo_x86 *c = &cpu_data(id);
-
-	*c = boot_cpu_data;
-	c->cpu_index = id;
-	if (id != 0)
-		identify_secondary_cpu(c);
-	smp_apply_quirks(c);
-}
-
 static atomic_t init_deasserted;
 
 static void __cpuinit smp_callin(void)
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index f84e30da521a..c213345ca2f5 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -85,21 +85,6 @@ struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
 #define set_idle_for_cpu(x,p)   (idle_thread_array[(x)] = (p))
 #endif
 
-/*
- * The bootstrap kernel entry code has set these up. Save them for
- * a given CPU
- */
-
-static void __cpuinit smp_store_cpu_info(int id)
-{
-	struct cpuinfo_x86 *c = &cpu_data(id);
-
-	*c = boot_cpu_data;
-	c->cpu_index = id;
-	if (id != 0)
-		identify_secondary_cpu(c);
-}
-
 static inline void wait_for_init_deassert(atomic_t *deassert)
 {
 	while (!atomic_read(deassert))
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 4dc271b43767..b4c5143d7f8d 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -88,6 +88,8 @@ extern void prefill_possible_map(void);
 
 #define SMP_TRAMPOLINE_BASE 0x6000
 extern unsigned long setup_trampoline(void);
+
+void smp_store_cpu_info(int id);
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 76740def6092..51624abda43a 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -42,8 +42,6 @@ DECLARE_PER_CPU(int, cpu_number);
 
 extern int safe_smp_processor_id(void);
 
-void __cpuinit smp_store_cpu_info(int id);
-
 /* We don't mark CPUs online until __cpu_up(), so we need another measure */
 static inline int num_booting_cpus(void)
 {
-- 
cgit v1.2.3


From ac56ef61a1f65aaf1cb31dca2a407322c5f0a4dd Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:10 -0300
Subject: x86: provide APIC_INTEGRATED definition for x86_64

it is always integrated, so define as 1.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/apicdef.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index 550af7a6f88e..b0c6b285c916 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -22,7 +22,11 @@
 #define		APIC_LVR_MASK		0xFF00FF
 #define		GET_APIC_VERSION(x)	((x)&0xFFu)
 #define		GET_APIC_MAXLVT(x)	(((x)>>16)&0xFFu)
-#define		APIC_INTEGRATED(x)	((x)&0xF0u)
+#ifdef CONFIG_X86_32
+#  define	APIC_INTEGRATED(x)	((x)&0xF0u)
+#else
+#  define	APIC_INTEGRATED(x)	(1)
+#endif
 #define		APIC_XAPIC(x)		((x) >= 0x14)
 #define	APIC_TASKPRI	0x80
 #define		APIC_TPRI_MASK		0xFFu
-- 
cgit v1.2.3


From 7e1efc0cde2a266fc31932ea7aed4bb20f524544 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:18 -0300
Subject: x86: unify extern masks declaration

take them off smp_{32,64}.h and move to smp.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 12 ++++++++++++
 include/asm-x86/smp_32.h |  8 --------
 include/asm-x86/smp_64.h | 11 -----------
 3 files changed, 12 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index b4c5143d7f8d..d02e6eacee39 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -3,12 +3,24 @@
 #ifndef __ASSEMBLY__
 #include <linux/cpumask.h>
 #include <linux/init.h>
+#include <asm/percpu.h>
 
 extern cpumask_t cpu_callout_map;
 
 extern int smp_num_siblings;
 extern unsigned int num_processors;
 
+extern u16 x86_cpu_to_apicid_init[];
+extern u16 x86_bios_cpu_apicid_init[];
+extern void *x86_cpu_to_apicid_early_ptr;
+extern void *x86_bios_cpu_apicid_early_ptr;
+
+DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
+DECLARE_PER_CPU(cpumask_t, cpu_core_map);
+DECLARE_PER_CPU(u16, cpu_llc_id);
+DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
+DECLARE_PER_CPU(u16, x86_bios_cpu_apicid);
+
 /*
  * Trampoline 80x86 program as an array.
  */
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 51624abda43a..478f5564630f 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -21,14 +21,6 @@ extern cpumask_t cpu_callin_map;
 extern void (*mtrr_hook) (void);
 extern void zap_low_mappings (void);
 
-extern u16 __initdata x86_cpu_to_apicid_init[];
-extern void *x86_cpu_to_apicid_early_ptr;
-
-DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
-DECLARE_PER_CPU(cpumask_t, cpu_core_map);
-DECLARE_PER_CPU(u16, cpu_llc_id);
-DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
-
 #ifdef CONFIG_SMP
 /*
  * This function is needed by all SMP systems. It must _always_ be valid
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 394c78524331..1b3c0f1de9a9 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -19,17 +19,6 @@ extern cpumask_t cpu_callin_map;
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
-extern u16 __initdata x86_cpu_to_apicid_init[];
-extern u16 __initdata x86_bios_cpu_apicid_init[];
-extern void *x86_cpu_to_apicid_early_ptr;
-extern void *x86_bios_cpu_apicid_early_ptr;
-
-DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
-DECLARE_PER_CPU(cpumask_t, cpu_core_map);
-DECLARE_PER_CPU(u16, cpu_llc_id);
-DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
-DECLARE_PER_CPU(u16, x86_bios_cpu_apicid);
-
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
 	if (cpu_present(mps_cpu))
-- 
cgit v1.2.3


From cbe879fc6c77b5751a91167654b75a39421d0f3f Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:19 -0300
Subject: x86: define bios to apicid mapping

This mapping already exists in x86_64, just provide it for
i386

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_32.c            | 6 ++++++
 include/asm-x86/mach-bigsmp/mach_apic.h | 7 ++-----
 include/asm-x86/mach-es7000/mach_apic.h | 8 +++-----
 include/asm-x86/mach-summit/mach_apic.h | 3 +--
 4 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 0e86ccc90d82..92a5df6190b5 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -70,6 +70,12 @@ void *x86_cpu_to_apicid_early_ptr;
 DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
+u16 x86_bios_cpu_apicid_init[NR_CPUS] __initdata
+				= { [0 ... NR_CPUS-1] = BAD_APICID };
+void *x86_bios_cpu_apicid_early_ptr;
+DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID;
+EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid);
+
 u8 apicid_2_node[MAX_APICID];
 
 static void map_cpu_to_logical_apicid(void);
diff --git a/include/asm-x86/mach-bigsmp/mach_apic.h b/include/asm-x86/mach-bigsmp/mach_apic.h
index 6df235e8ea91..0d55b1f6d56b 100644
--- a/include/asm-x86/mach-bigsmp/mach_apic.h
+++ b/include/asm-x86/mach-bigsmp/mach_apic.h
@@ -1,10 +1,7 @@
 #ifndef __ASM_MACH_APIC_H
 #define __ASM_MACH_APIC_H
 
-
-extern u8 bios_cpu_apicid[];
-
-#define xapic_phys_to_log_apicid(cpu) (bios_cpu_apicid[cpu])
+#define xapic_phys_to_log_apicid(cpu) (per_cpu(x86_bios_cpu_apicid, cpu))
 #define esr_disable (1)
 
 static inline int apic_id_registered(void)
@@ -90,7 +87,7 @@ static inline int apicid_to_node(int logical_apicid)
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
 	if (mps_cpu < NR_CPUS)
-		return (int) bios_cpu_apicid[mps_cpu];
+		return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu);
 
 	return BAD_APICID;
 }
diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/mach-es7000/mach_apic.h
index d23011fdf454..04cba9f1e375 100644
--- a/include/asm-x86/mach-es7000/mach_apic.h
+++ b/include/asm-x86/mach-es7000/mach_apic.h
@@ -1,9 +1,7 @@
 #ifndef __ASM_MACH_APIC_H
 #define __ASM_MACH_APIC_H
 
-extern u8 bios_cpu_apicid[];
-
-#define xapic_phys_to_log_apicid(cpu) (bios_cpu_apicid[cpu])
+#define xapic_phys_to_log_apicid(cpu) per_cpu(x86_bios_cpu_apicid, cpu)
 #define esr_disable (1)
 
 static inline int apic_id_registered(void)
@@ -80,7 +78,7 @@ extern void enable_apic_mode(void);
 extern int apic_version [MAX_APICS];
 static inline void setup_apic_routing(void)
 {
-	int apic = bios_cpu_apicid[smp_processor_id()];
+	int apic = per_cpu(x86_bios_cpu_apicid, smp_processor_id());
 	printk("Enabling APIC mode:  %s.  Using %d I/O APICs, target cpus %lx\n",
 		(apic_version[apic] == 0x14) ? 
 		"Physical Cluster" : "Logical Cluster", nr_ioapics, cpus_addr(TARGET_CPUS)[0]);
@@ -102,7 +100,7 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 	if (!mps_cpu)
 		return boot_cpu_physical_apicid;
 	else if (mps_cpu < NR_CPUS)
-		return (int) bios_cpu_apicid[mps_cpu];
+		return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu);
 	else
 		return BAD_APICID;
 }
diff --git a/include/asm-x86/mach-summit/mach_apic.h b/include/asm-x86/mach-summit/mach_apic.h
index 062c97f6100b..91d7641cddc9 100644
--- a/include/asm-x86/mach-summit/mach_apic.h
+++ b/include/asm-x86/mach-summit/mach_apic.h
@@ -40,7 +40,6 @@ static inline unsigned long check_apicid_present(int bit)
 
 #define apicid_cluster(apicid) ((apicid) & XAPIC_DEST_CLUSTER_MASK)
 
-extern u8 bios_cpu_apicid[];
 extern u8 cpu_2_logical_apicid[];
 
 static inline void init_apic_ldr(void)
@@ -110,7 +109,7 @@ static inline int cpu_to_logical_apicid(int cpu)
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
 	if (mps_cpu < NR_CPUS)
-		return (int)bios_cpu_apicid[mps_cpu];
+		return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu);
 	else
 		return BAD_APICID;
 }
-- 
cgit v1.2.3


From 04d1dd20f64f2b41baf5c01f57c574ca942ab4eb Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:21 -0300
Subject: x86: make node to apic mapping declarations unconditional

Instead of declaring them inside of X86_64 ifdef, do it
unconditionally

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/topology.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index 8af05a93f097..dada89e5b152 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -32,15 +32,15 @@
 /* Mappings between logical cpu number and node number */
 #ifdef CONFIG_X86_32
 extern int cpu_to_node_map[];
-
 #else
-DECLARE_PER_CPU(int, x86_cpu_to_node_map);
-extern int x86_cpu_to_node_map_init[];
-extern void *x86_cpu_to_node_map_early_ptr;
 /* Returns the number of the current Node. */
 #define numa_node_id()		(early_cpu_to_node(raw_smp_processor_id()))
 #endif
 
+DECLARE_PER_CPU(int, x86_cpu_to_node_map);
+extern int x86_cpu_to_node_map_init[];
+extern void *x86_cpu_to_node_map_early_ptr;
+
 extern cpumask_t node_to_cpumask_map[];
 
 #define NUMA_NO_NODE	(-1)
-- 
cgit v1.2.3


From fbac7fcbadc54cc5d374873a2e60e924a056d198 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:22 -0300
Subject: x86: fix alloc_bootmem_pages_node macro

missing a semicolon

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmzone_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/mmzone_32.h b/include/asm-x86/mmzone_32.h
index 274a59566c45..b9f5be2f603b 100644
--- a/include/asm-x86/mmzone_32.h
+++ b/include/asm-x86/mmzone_32.h
@@ -129,7 +129,7 @@ static inline int pfn_valid(int pfn)
 	struct pglist_data  __maybe_unused			\
 				*__alloc_bootmem_node__pgdat = (pgdat);	\
 	__alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE,		\
-						__pa(MAX_DMA_ADDRESS))	\
+						__pa(MAX_DMA_ADDRESS));	\
 })
 #define alloc_bootmem_low_pages_node(pgdat, x)				\
 ({									\
-- 
cgit v1.2.3


From e32ede19ac64b5cd896e6d28aa51d34887791ab2 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:35 -0300
Subject: x86: wipe get_nmi_reason out of nmi_64.h

use mach_traps when it is supposed to be used.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/nmi_64.c   | 2 ++
 arch/x86/kernel/traps_64.c | 2 ++
 include/asm-x86/nmi_64.h   | 2 --
 3 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c
index 9a4fde74bee1..11f9130ac513 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -26,6 +26,8 @@
 #include <asm/proto.h>
 #include <asm/mce.h>
 
+#include <mach_traps.h>
+
 int unknown_nmi_panic;
 int nmi_watchdog_enabled;
 int panic_on_unrecovered_nmi;
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index 045466681911..33292ac814f4 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -33,6 +33,8 @@
 #include <linux/kdebug.h>
 #include <linux/utsname.h>
 
+#include <mach_traps.h>
+
 #if defined(CONFIG_EDAC)
 #include <linux/edac.h>
 #endif
diff --git a/include/asm-x86/nmi_64.h b/include/asm-x86/nmi_64.h
index 2eeb74e5f3ff..94a5b19e3620 100644
--- a/include/asm-x86/nmi_64.h
+++ b/include/asm-x86/nmi_64.h
@@ -36,8 +36,6 @@ static inline void unset_nmi_pm_callback(struct pm_dev * dev)
 extern void default_do_nmi(struct pt_regs *);
 extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
 
-#define get_nmi_reason() inb(0x61)
-
 extern int unknown_nmi_panic;
 extern int nmi_watchdog_enabled;
 
-- 
cgit v1.2.3


From 6d60cd5359e261cad1e519e77ca733c05c2f8025 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:36 -0300
Subject: x86: unify nmi_32.h and nmi_64.h

Two more files goes away. nmi_64.h and nmi_32.h gives birth
to nmi.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/nmi_32.c |  2 +-
 include/asm-x86/nmi.h    | 92 ++++++++++++++++++++++++++++++++++++++++++++++--
 include/asm-x86/nmi_32.h | 61 --------------------------------
 include/asm-x86/nmi_64.h | 88 ---------------------------------------------
 4 files changed, 90 insertions(+), 153 deletions(-)
 delete mode 100644 include/asm-x86/nmi_32.h
 delete mode 100644 include/asm-x86/nmi_64.h

(limited to 'include')

diff --git a/arch/x86/kernel/nmi_32.c b/arch/x86/kernel/nmi_32.c
index 6a0aa7038685..167385155765 100644
--- a/arch/x86/kernel/nmi_32.c
+++ b/arch/x86/kernel/nmi_32.c
@@ -67,7 +67,7 @@ static __init void nmi_cpu_busy(void *data)
 }
 #endif
 
-static int __init check_nmi_watchdog(void)
+int __init check_nmi_watchdog(void)
 {
 	unsigned int *prev_nmi_count;
 	int cpu;
diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h
index 53ccac14cead..2b9419903476 100644
--- a/include/asm-x86/nmi.h
+++ b/include/asm-x86/nmi.h
@@ -1,5 +1,91 @@
-#ifdef CONFIG_X86_32
-# include "nmi_32.h"
+#ifndef _ASM_X86_NMI_H_
+#define _ASM_X86_NMI_H_
+
+#include <linux/pm.h>
+#include <asm/irq.h>
+#include <asm/io.h>
+
+#ifdef ARCH_HAS_NMI_WATCHDOG
+
+/**
+ * do_nmi_callback
+ *
+ * Check to see if a callback exists and execute it.  Return 1
+ * if the handler exists and was handled successfully.
+ */
+int do_nmi_callback(struct pt_regs *regs, int cpu);
+
+#ifdef CONFIG_PM
+
+/** Replace the PM callback routine for NMI. */
+struct pm_dev *set_nmi_pm_callback(pm_callback callback);
+
+/** Unset the PM callback routine back to the default. */
+void unset_nmi_pm_callback(struct pm_dev *dev);
+
 #else
-# include "nmi_64.h"
+
+static inline struct pm_dev *set_nmi_pm_callback(pm_callback callback)
+{
+	return 0;
+}
+
+static inline void unset_nmi_pm_callback(struct pm_dev *dev)
+{
+}
+
+#endif /* CONFIG_PM */
+
+#ifdef CONFIG_X86_64
+extern void default_do_nmi(struct pt_regs *);
+extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
+#endif
+
+extern int check_nmi_watchdog(void);
+extern int nmi_watchdog_enabled;
+extern int unknown_nmi_panic;
+extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
+extern int avail_to_resrv_perfctr_nmi(unsigned int);
+extern int reserve_perfctr_nmi(unsigned int);
+extern void release_perfctr_nmi(unsigned int);
+extern int reserve_evntsel_nmi(unsigned int);
+extern void release_evntsel_nmi(unsigned int);
+extern void nmi_watchdog_default(void);
+
+extern void setup_apic_nmi_watchdog(void *);
+extern void stop_apic_nmi_watchdog(void *);
+extern void disable_timer_nmi_watchdog(void);
+extern void enable_timer_nmi_watchdog(void);
+extern int nmi_watchdog_tick(struct pt_regs *regs, unsigned reason);
+
+extern atomic_t nmi_active;
+extern unsigned int nmi_watchdog;
+#define NMI_DISABLED    -1
+#define NMI_NONE	0
+#define NMI_IO_APIC	1
+#define NMI_LOCAL_APIC	2
+#define NMI_INVALID	3
+#define NMI_DEFAULT	NMI_DISABLED
+
+struct ctl_table;
+struct file;
+extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
+			void __user *, size_t *, loff_t *);
+extern int unknown_nmi_panic;
+
+void __trigger_all_cpu_backtrace(void);
+#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace()
+
+#endif
+
+void lapic_watchdog_stop(void);
+int lapic_watchdog_init(unsigned nmi_hz);
+int lapic_wd_event(unsigned nmi_hz);
+unsigned lapic_adjust_nmi_hz(unsigned hz);
+int lapic_watchdog_ok(void);
+void disable_lapic_nmi_watchdog(void);
+void enable_lapic_nmi_watchdog(void);
+void stop_nmi(void);
+void restart_nmi(void);
+
 #endif
diff --git a/include/asm-x86/nmi_32.h b/include/asm-x86/nmi_32.h
deleted file mode 100644
index 7206c7e8a388..000000000000
--- a/include/asm-x86/nmi_32.h
+++ /dev/null
@@ -1,61 +0,0 @@
-#ifndef ASM_NMI_H
-#define ASM_NMI_H
-
-#include <linux/pm.h>
-#include <asm/irq.h>
-
-#ifdef ARCH_HAS_NMI_WATCHDOG
-
-/**
- * do_nmi_callback
- *
- * Check to see if a callback exists and execute it.  Return 1
- * if the handler exists and was handled successfully.
- */
-int do_nmi_callback(struct pt_regs *regs, int cpu);
-
-extern int nmi_watchdog_enabled;
-extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
-extern int avail_to_resrv_perfctr_nmi(unsigned int);
-extern int reserve_perfctr_nmi(unsigned int);
-extern void release_perfctr_nmi(unsigned int);
-extern int reserve_evntsel_nmi(unsigned int);
-extern void release_evntsel_nmi(unsigned int);
-
-extern void setup_apic_nmi_watchdog (void *);
-extern void stop_apic_nmi_watchdog (void *);
-extern void disable_timer_nmi_watchdog(void);
-extern void enable_timer_nmi_watchdog(void);
-extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
-
-extern atomic_t nmi_active;
-extern unsigned int nmi_watchdog;
-#define NMI_DISABLED    -1
-#define NMI_NONE	0
-#define NMI_IO_APIC	1
-#define NMI_LOCAL_APIC	2
-#define NMI_INVALID	3
-#define NMI_DEFAULT	NMI_DISABLED
-
-struct ctl_table;
-struct file;
-extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
-			void __user *, size_t *, loff_t *);
-extern int unknown_nmi_panic;
-
-void __trigger_all_cpu_backtrace(void);
-#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace()
-
-#endif
-
-void lapic_watchdog_stop(void);
-int lapic_watchdog_init(unsigned nmi_hz);
-int lapic_wd_event(unsigned nmi_hz);
-unsigned lapic_adjust_nmi_hz(unsigned hz);
-int lapic_watchdog_ok(void);
-void disable_lapic_nmi_watchdog(void);
-void enable_lapic_nmi_watchdog(void);
-void stop_nmi(void);
-void restart_nmi(void);
-
-#endif /* ASM_NMI_H */
diff --git a/include/asm-x86/nmi_64.h b/include/asm-x86/nmi_64.h
deleted file mode 100644
index 94a5b19e3620..000000000000
--- a/include/asm-x86/nmi_64.h
+++ /dev/null
@@ -1,88 +0,0 @@
-#ifndef ASM_NMI_H
-#define ASM_NMI_H
-
-#include <linux/pm.h>
-#include <asm/io.h>
- 
-/**
- * do_nmi_callback
- *
- * Check to see if a callback exists and execute it.  Return 1
- * if the handler exists and was handled successfully.
- */
-int do_nmi_callback(struct pt_regs *regs, int cpu);
-
-#ifdef CONFIG_PM
- 
-/** Replace the PM callback routine for NMI. */
-struct pm_dev * set_nmi_pm_callback(pm_callback callback);
-
-/** Unset the PM callback routine back to the default. */
-void unset_nmi_pm_callback(struct pm_dev * dev);
-
-#else
-
-static inline struct pm_dev * set_nmi_pm_callback(pm_callback callback)
-{
-	return 0;
-} 
- 
-static inline void unset_nmi_pm_callback(struct pm_dev * dev)
-{
-}
-
-#endif /* CONFIG_PM */
- 
-extern void default_do_nmi(struct pt_regs *);
-extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
-
-extern int unknown_nmi_panic;
-extern int nmi_watchdog_enabled;
-
-extern int check_nmi_watchdog(void);
-extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
-extern int avail_to_resrv_perfctr_nmi(unsigned int);
-extern int reserve_perfctr_nmi(unsigned int);
-extern void release_perfctr_nmi(unsigned int);
-extern int reserve_evntsel_nmi(unsigned int);
-extern void release_evntsel_nmi(unsigned int);
-
-extern void setup_apic_nmi_watchdog (void *);
-extern void stop_apic_nmi_watchdog (void *);
-extern void disable_timer_nmi_watchdog(void);
-extern void enable_timer_nmi_watchdog(void);
-extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
-
-extern void nmi_watchdog_default(void);
-
-extern atomic_t nmi_active;
-extern unsigned int nmi_watchdog;
-#define NMI_DISABLED    -1
-#define NMI_NONE	0
-#define NMI_IO_APIC	1
-#define NMI_LOCAL_APIC	2
-#define NMI_INVALID	3
-#define NMI_DEFAULT	NMI_DISABLED
-
-struct ctl_table;
-struct file;
-extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
-			void __user *, size_t *, loff_t *);
-
-extern int unknown_nmi_panic;
-
-void __trigger_all_cpu_backtrace(void);
-#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace()
-
-
-void lapic_watchdog_stop(void);
-int lapic_watchdog_init(unsigned nmi_hz);
-int lapic_wd_event(unsigned nmi_hz);
-unsigned lapic_adjust_nmi_hz(unsigned hz);
-int lapic_watchdog_ok(void);
-void disable_lapic_nmi_watchdog(void);
-void enable_lapic_nmi_watchdog(void);
-void stop_nmi(void);
-void restart_nmi(void);
-
-#endif /* ASM_NMI_H */
-- 
cgit v1.2.3


From 50e440aa5323860d9e5960143b720e461ed0c582 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:38 -0300
Subject: x86: call nmi_watchdog_default in i386

this does not exist, so it will be an empty macro

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_32.c | 1 +
 include/asm-x86/nmi.h        | 4 +++-
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 1f3aff4caaf7..a35055361b85 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -876,6 +876,7 @@ static void __init smp_boot_cpus(unsigned int max_cpus)
    who understands all this stuff should rewrite it properly. --RR 15/Jul/02 */
 void __init native_smp_prepare_cpus(unsigned int max_cpus)
 {
+	nmi_watchdog_default();
 	smp_commenced_mask = cpumask_of_cpu(0);
 	cpu_callin_map = cpumask_of_cpu(0);
 	mb();
diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h
index 2b9419903476..1e363021e72f 100644
--- a/include/asm-x86/nmi.h
+++ b/include/asm-x86/nmi.h
@@ -39,6 +39,9 @@ static inline void unset_nmi_pm_callback(struct pm_dev *dev)
 #ifdef CONFIG_X86_64
 extern void default_do_nmi(struct pt_regs *);
 extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
+extern void nmi_watchdog_default(void);
+#else
+#define nmi_watchdog_default() do {} while (0)
 #endif
 
 extern int check_nmi_watchdog(void);
@@ -50,7 +53,6 @@ extern int reserve_perfctr_nmi(unsigned int);
 extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
-extern void nmi_watchdog_default(void);
 
 extern void setup_apic_nmi_watchdog(void *);
 extern void stop_apic_nmi_watchdog(void *);
-- 
cgit v1.2.3


From f6bc40290964b5fcb48c226ccafa4b7536d62663 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:53 -0300
Subject: x86: include mach_apic.h in smpboot_64.c and smpboot.c

After the inclusion, a lot of files needs fixing for conflicts,
some of them in the headers themselves, to accomodate for both
i386 and x86_64 versions.

[ mingo@elte.hu: build fix ]

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/boot.c                 |  4 ++++
 arch/x86/kernel/mpparse_64.c                |  2 ++
 arch/x86/kernel/smpboot.c                   |  2 ++
 arch/x86/kernel/smpboot_64.c                |  1 +
 arch/x86/vdso/Makefile                      |  2 +-
 include/asm-x86/apic.h                      |  1 -
 include/asm-x86/apicdef.h                   |  6 ------
 include/asm-x86/mach-default/mach_apic.h    | 11 +++++++++++
 include/asm-x86/mach-default/mach_apicdef.h |  5 +++++
 include/asm-x86/smp_64.h                    |  9 +--------
 10 files changed, 27 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 2cdc9de9371d..956b60f3ebd5 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -40,6 +40,10 @@
 #include <asm/io.h>
 #include <asm/mpspec.h>
 
+#ifdef CONFIG_X86_LOCAL_APIC
+# include <mach_apic.h>
+#endif
+
 static int __initdata acpi_force = 0;
 
 #ifdef	CONFIG_ACPI
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 529b1c22077e..03ef1a8b53e8 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -30,6 +30,8 @@
 #include <asm/proto.h>
 #include <asm/acpi.h>
 
+#include <mach_apic.h>
+
 /* Have we found an MP table */
 int smp_found_config;
 unsigned int __cpuinitdata maxcpus = NR_CPUS;
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 6978f1bf6533..253be86a88e4 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -11,6 +11,8 @@
 #include <asm/cpu.h>
 #include <asm/numa.h>
 
+#include <mach_apic.h>
+
 /* Number of siblings per CPU package */
 int smp_num_siblings = 1;
 EXPORT_SYMBOL(smp_num_siblings);
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 7d1b4cb380db..8a59fa80f883 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -61,6 +61,7 @@
 #include <asm/numa.h>
 
 #include <mach_wakecpu.h>
+#include <mach_apic.h>
 #include <smpboot_hooks.h>
 
 /* Set when the idlers are all forked */
diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile
index 0a8f4742ef51..17a6b057856b 100644
--- a/arch/x86/vdso/Makefile
+++ b/arch/x86/vdso/Makefile
@@ -39,7 +39,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE
 
 CFL := $(PROFILING) -mcmodel=small -fPIC -g0 -O2 -fasynchronous-unwind-tables -m64
 
-$(vobjs): KBUILD_CFLAGS = $(CFL)
+$(vobjs): KBUILD_CFLAGS += $(CFL)
 
 targets += vdso-syms.lds
 obj-$(VDSO64-y)			+= vdso-syms.lds
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index db5f7501aed6..424e1dfe13c9 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -128,7 +128,6 @@ extern void enable_NMI_through_LVT0(void);
  * On 32bit this is mach-xxx local
  */
 #ifdef CONFIG_X86_64
-extern void setup_apic_routing(void);
 extern void early_init_lapic_mapping(void);
 #endif
 
diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index b0c6b285c916..674a2280e21e 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -12,12 +12,6 @@
 
 #define	APIC_ID		0x20
 
-#ifdef CONFIG_X86_64
-# define	APIC_ID_MASK		(0xFFu<<24)
-# define	GET_APIC_ID(x)		(((x)>>24)&0xFFu)
-# define	SET_APIC_ID(x)		(((x)<<24))
-#endif
-
 #define	APIC_LVR	0x30
 #define		APIC_LVR_MASK		0xFF00FF
 #define		GET_APIC_VERSION(x)	((x)&0xFFu)
diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h
index e3c2c1012c1c..e081bdccde2b 100644
--- a/include/asm-x86/mach-default/mach_apic.h
+++ b/include/asm-x86/mach-default/mach_apic.h
@@ -54,21 +54,27 @@ static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map)
 	return phys_map;
 }
 
+#ifdef CONFIG_X86_64
+extern void setup_apic_routing(void);
+#else
 static inline void setup_apic_routing(void)
 {
 	printk("Enabling APIC mode:  %s.  Using %d I/O APICs\n",
 					"Flat", nr_ioapics);
 }
+#endif
 
 static inline int multi_timer_check(int apic, int irq)
 {
 	return 0;
 }
 
+#ifdef CONFIG_X86_32
 static inline int apicid_to_node(int logical_apicid)
 {
 	return 0;
 }
+#endif
 
 /* Mapping from cpu number to logical apicid */
 static inline int cpu_to_logical_apicid(int cpu)
@@ -78,8 +84,13 @@ static inline int cpu_to_logical_apicid(int cpu)
 
 static inline int cpu_present_to_apicid(int mps_cpu)
 {
+#ifdef CONFIG_X86_64
+	if (cpu_present(mps_cpu))
+		return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu);
+#else
 	if (mps_cpu < get_physical_broadcast())
 		return  mps_cpu;
+#endif
 	else
 		return BAD_APICID;
 }
diff --git a/include/asm-x86/mach-default/mach_apicdef.h b/include/asm-x86/mach-default/mach_apicdef.h
index ae9841319094..7b78275e6d33 100644
--- a/include/asm-x86/mach-default/mach_apicdef.h
+++ b/include/asm-x86/mach-default/mach_apicdef.h
@@ -3,7 +3,12 @@
 
 #include <asm/apic.h>
 
+#ifdef CONFIG_X86_64
+#define	APIC_ID_MASK		(0xFFu<<24)
+#define	SET_APIC_ID(x)		(((x)<<24))
+#else
 #define		APIC_ID_MASK		(0xF<<24)
+#endif
 
 static inline unsigned get_apic_id(unsigned long x) 
 { 
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 1b3c0f1de9a9..be870a4881fc 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -19,14 +19,6 @@ extern cpumask_t cpu_callin_map;
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
-static inline int cpu_present_to_apicid(int mps_cpu)
-{
-	if (cpu_present(mps_cpu))
-		return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu);
-	else
-		return BAD_APICID;
-}
-
 #ifdef CONFIG_SMP
 
 #define raw_smp_processor_id()	read_pda(cpunumber)
@@ -64,6 +56,7 @@ static __inline int logical_smp_processor_id(void)
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
 }
 
+#include <mach_apicdef.h>
 static inline int hard_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
-- 
cgit v1.2.3


From 9d97d0da71ad6c7ceb76b4e29b02bed1ee9d4cd2 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:57 -0300
Subject: x86: move stack_start to smp.h

voyager would conflict with it, but the types are ultimately
compatible. So remove the extern definition from voyager_smp.c
in favour of the common one

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_32.c        | 6 ------
 arch/x86/mach-voyager/voyager_smp.c | 7 -------
 include/asm-x86/smp.h               | 7 +++++++
 3 files changed, 7 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index 1eb7b73b45a3..ae25927f08c1 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -257,12 +257,6 @@ void __devinit initialize_secondary(void)
 		:"m" (current->thread.sp),"m" (current->thread.ip));
 }
 
-/* Static state in head.S used to set up a CPU */
-extern struct {
-	void * sp;
-	unsigned short ss;
-} stack_start;
-
 static inline void __inquire_remote_apic(int apicid)
 {
 	unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 };
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 11b806c7ed94..4397235c2e30 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -520,13 +520,6 @@ static void __init do_boot_cpu(__u8 cpu)
 	    & ~(voyager_extended_vic_processors
 		& voyager_allowed_boot_processors);
 
-	/* This is an area in head.S which was used to set up the
-	 * initial kernel stack.  We need to alter this to give the
-	 * booting CPU a new stack (taken from its idle process) */
-	extern struct {
-		__u8 *sp;
-		unsigned short ss;
-	} stack_start;
 	/* This is the format of the CPI IDT gate (in real mode) which
 	 * we're hijacking to boot the CPU */
 	union IDTFormat {
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index d02e6eacee39..78ef16dd194b 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -28,6 +28,13 @@ extern const unsigned char trampoline_data [];
 extern const unsigned char trampoline_end  [];
 extern unsigned char *trampoline_base;
 
+/* Static state in head.S used to set up a CPU */
+extern struct {
+	void *sp;
+	unsigned short ss;
+} stack_start;
+
+
 struct smp_ops {
 	void (*smp_prepare_boot_cpu)(void);
 	void (*smp_prepare_cpus)(unsigned max_cpus);
-- 
cgit v1.2.3


From c70dcb74309cedfa64f0060f4a84792e873ceb53 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:58 -0300
Subject: x86: change boot_cpu_id to boot_cpu_physical_apicid

This is to match i386. The former name was cuter,
but the current is more meaningful and more general,
since cpu_id can be a logical id.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c    | 13 +++++++------
 arch/x86/kernel/mpparse_64.c | 12 ++++++------
 arch/x86/kernel/smpboot_64.c | 18 ++++++++++--------
 include/asm-x86/apic.h       |  1 -
 include/asm-x86/smp.h        |  3 +++
 include/asm-x86/smp_32.h     |  5 -----
 include/asm-x86/smp_64.h     |  4 ----
 7 files changed, 26 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 8a475793f736..868ec1deb19a 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -431,7 +431,8 @@ void __cpuinit check_boot_apic_timer_broadcast(void)
 	lapic_clockevent.features |= CLOCK_EVT_FEAT_DUMMY;
 
 	local_irq_enable();
-	clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE, &boot_cpu_id);
+	clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE,
+			   &boot_cpu_physical_apicid);
 	local_irq_disable();
 }
 
@@ -857,7 +858,7 @@ static int __init detect_init_APIC(void)
 	}
 
 	mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
-	boot_cpu_id = 0;
+	boot_cpu_physical_apicid = 0;
 	return 0;
 }
 
@@ -882,7 +883,7 @@ void __init early_init_lapic_mapping(void)
 	 * Fetch the APIC ID of the BSP in case we have a
 	 * default configuration (or the MP table is broken).
 	 */
-	boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
+	boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
 }
 
 /**
@@ -909,7 +910,7 @@ void __init init_apic_mappings(void)
 	 * Fetch the APIC ID of the BSP in case we have a
 	 * default configuration (or the MP table is broken).
 	 */
-	boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
+	boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
 }
 
 /*
@@ -930,8 +931,8 @@ int __init APIC_init_uniprocessor(void)
 
 	verify_local_APIC();
 
-	phys_cpu_present_map = physid_mask_of_physid(boot_cpu_id);
-	apic_write(APIC_ID, SET_APIC_ID(boot_cpu_id));
+	phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid);
+	apic_write(APIC_ID, SET_APIC_ID(boot_cpu_physical_apicid));
 
 	setup_local_APIC();
 
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 03ef1a8b53e8..20a345dd425b 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -59,8 +59,8 @@ unsigned long mp_lapic_addr = 0;
 
 
 /* Processor that is doing the boot up */
-unsigned int boot_cpu_id = -1U;
-EXPORT_SYMBOL(boot_cpu_id);
+unsigned int boot_cpu_physical_apicid = -1U;
+EXPORT_SYMBOL(boot_cpu_physical_apicid);
 
 /* Internal processor count */
 unsigned int num_processors;
@@ -107,7 +107,7 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
 	}
 	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
 		bootup_cpu = " (Bootup-CPU)";
-		boot_cpu_id = m->mpc_apicid;
+		boot_cpu_physical_apicid = m->mpc_apicid;
 	}
 
 	printk(KERN_INFO "Processor #%d%s\n", m->mpc_apicid, bootup_cpu);
@@ -665,8 +665,8 @@ void __init mp_register_lapic_address(u64 address)
 {
 	mp_lapic_addr = (unsigned long) address;
 	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
-	if (boot_cpu_id == -1U)
-		boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
+	if (boot_cpu_physical_apicid == -1U)
+		boot_cpu_physical_apicid  = GET_APIC_ID(apic_read(APIC_ID));
 }
 
 void __cpuinit mp_register_lapic (u8 id, u8 enabled)
@@ -674,7 +674,7 @@ void __cpuinit mp_register_lapic (u8 id, u8 enabled)
 	struct mpc_config_processor processor;
 	int			boot_cpu = 0;
 	
-	if (id == boot_cpu_id)
+	if (id == boot_cpu_physical_apicid)
 		boot_cpu = 1;
 
 	processor.mpc_type = MP_PROCESSOR;
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 7ec96218a97e..420ae4a33548 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -602,7 +602,8 @@ static __init void disable_smp(void)
 	cpu_present_map = cpumask_of_cpu(0);
 	cpu_possible_map = cpumask_of_cpu(0);
 	if (smp_found_config)
-		phys_cpu_present_map = physid_mask_of_physid(boot_cpu_id);
+		phys_cpu_present_map =
+				physid_mask_of_physid(boot_cpu_physical_apicid);
 	else
 		phys_cpu_present_map = physid_mask_of_physid(0);
 	cpu_set(0, per_cpu(cpu_sibling_map, 0));
@@ -637,9 +638,10 @@ static int __init smp_sanity_check(unsigned max_cpus)
 	 * Should not be necessary because the MP table should list the boot
 	 * CPU too, but we do it for the sake of robustness anyway.
 	 */
-	if (!physid_isset(boot_cpu_id, phys_cpu_present_map)) {
-		printk(KERN_NOTICE "weird, boot CPU (#%d) not listed by the BIOS.\n",
-								 boot_cpu_id);
+	if (!physid_isset(boot_cpu_physical_apicid, phys_cpu_present_map)) {
+		printk(KERN_NOTICE
+			"weird, boot CPU (#%d) not listed by the BIOS.\n",
+			boot_cpu_physical_apicid);
 		physid_set(hard_smp_processor_id(), phys_cpu_present_map);
 	}
 
@@ -648,7 +650,7 @@ static int __init smp_sanity_check(unsigned max_cpus)
 	 */
 	if (!cpu_has_apic) {
 		printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n",
-			boot_cpu_id);
+			boot_cpu_physical_apicid);
 		printk(KERN_ERR "... forcing use of dummy APIC emulation. (tell your hw vendor)\n");
 		nr_ioapics = 0;
 		return -1;
@@ -709,9 +711,9 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
 		enable_IO_APIC();
 	end_local_APIC_setup();
 
-	if (GET_APIC_ID(apic_read(APIC_ID)) != boot_cpu_id) {
+	if (GET_APIC_ID(apic_read(APIC_ID)) != boot_cpu_physical_apicid) {
 		panic("Boot APIC ID in local APIC unexpected (%d vs %d)",
-		      GET_APIC_ID(apic_read(APIC_ID)), boot_cpu_id);
+		     GET_APIC_ID(apic_read(APIC_ID)), boot_cpu_physical_apicid);
 		/* Or can we switch back to PIC here? */
 	}
 
@@ -756,7 +758,7 @@ int __cpuinit native_cpu_up(unsigned int cpu)
 
 	Dprintk("++++++++++++++++++++=_---CPU UP  %u\n", cpu);
 
-	if (apicid == BAD_APICID || apicid == boot_cpu_id ||
+	if (apicid == BAD_APICID || apicid == boot_cpu_physical_apicid ||
 	    !physid_isset(apicid, phys_cpu_present_map)) {
 		printk("__cpu_up: bad cpu %d\n", cpu);
 		return -EINVAL;
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 424e1dfe13c9..b804238489a1 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -44,7 +44,6 @@ extern int apic_runs_main_timer;
 extern int ioapic_force;
 extern int disable_apic;
 extern int disable_apic_timer;
-extern unsigned boot_cpu_id;
 
 /*
  * Basic functions accessing APICs.
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 78ef16dd194b..2ad2f4ffe498 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -109,6 +109,9 @@ extern void prefill_possible_map(void);
 extern unsigned long setup_trampoline(void);
 
 void smp_store_cpu_info(int id);
+#define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
+#else
+#define cpu_physical_id(cpu)		boot_cpu_physical_apicid
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 478f5564630f..f861d0415171 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -30,8 +30,6 @@ extern void zap_low_mappings (void);
 DECLARE_PER_CPU(int, cpu_number);
 #define raw_smp_processor_id() (x86_read_percpu(cpu_number))
 
-#define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
-
 extern int safe_smp_processor_id(void);
 
 /* We don't mark CPUs online until __cpu_up(), so we need another measure */
@@ -41,10 +39,7 @@ static inline int num_booting_cpus(void)
 }
 
 #else /* CONFIG_SMP */
-
 #define safe_smp_processor_id()		0
-#define cpu_physical_id(cpu)		boot_cpu_physical_apicid
-
 #endif /* !CONFIG_SMP */
 
 #ifdef CONFIG_X86_LOCAL_APIC
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index be870a4881fc..fd709cbba4d1 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -22,7 +22,6 @@ extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 #ifdef CONFIG_SMP
 
 #define raw_smp_processor_id()	read_pda(cpunumber)
-#define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
 
 #define stack_smp_processor_id()					\
 	({								\
@@ -41,9 +40,6 @@ static inline int num_booting_cpus(void)
 }
 
 #else /* CONFIG_SMP */
-
-extern unsigned int boot_cpu_id;
-#define cpu_physical_id(cpu)	boot_cpu_id
 #define stack_smp_processor_id() 0
 
 #endif /* !CONFIG_SMP */
-- 
cgit v1.2.3


From cb3c8b9003f15efa4a750a32d2d602d40cc45d5a Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:25:59 -0300
Subject: x86: integrate do_boot_cpu

This is a very large patch, because it depends on a lot
of auxiliary static functions. But they all have been modified
to the point that they're sufficiently close now. So they're just
merged in smpboot.c

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot.c    | 588 +++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/smpboot_32.c | 532 +--------------------------------------
 arch/x86/kernel/smpboot_64.c | 515 +------------------------------------
 include/asm-x86/smp.h        |   3 +
 4 files changed, 594 insertions(+), 1044 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 5bff87e99898..69c17965f48d 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -4,14 +4,42 @@
 #include <linux/sched.h>
 #include <linux/percpu.h>
 #include <linux/bootmem.h>
+#include <linux/err.h>
+#include <linux/nmi.h>
 
+#include <asm/desc.h>
 #include <asm/nmi.h>
 #include <asm/irq.h>
 #include <asm/smp.h>
 #include <asm/cpu.h>
 #include <asm/numa.h>
+#include <asm/pgtable.h>
+#include <asm/tlbflush.h>
+#include <asm/mtrr.h>
+#include <asm/nmi.h>
+#include <linux/mc146818rtc.h>
 
 #include <mach_apic.h>
+#include <mach_wakecpu.h>
+#include <smpboot_hooks.h>
+
+/* Store all idle threads, this can be reused instead of creating
+* a new thread. Also avoids complicated thread destroy functionality
+* for idle threads.
+*/
+#ifdef CONFIG_HOTPLUG_CPU
+/*
+ * Needed only for CONFIG_HOTPLUG_CPU because __cpuinitdata is
+ * removed after init for !CONFIG_HOTPLUG_CPU.
+ */
+static DEFINE_PER_CPU(struct task_struct *, idle_thread_array);
+#define get_idle_for_cpu(x)      (per_cpu(idle_thread_array, x))
+#define set_idle_for_cpu(x, p)   (per_cpu(idle_thread_array, x) = (p))
+#else
+struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
+#define get_idle_for_cpu(x)      (idle_thread_array[(x)])
+#define set_idle_for_cpu(x, p)   (idle_thread_array[(x)] = (p))
+#endif
 
 /* Number of siblings per CPU package */
 int smp_num_siblings = 1;
@@ -41,6 +69,8 @@ EXPORT_PER_CPU_SYMBOL(cpu_core_map);
 DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
 EXPORT_PER_CPU_SYMBOL(cpu_info);
 
+static atomic_t init_deasserted;
+
 /* ready for x86_64, no harm for x86, since it will overwrite after alloc */
 unsigned char *trampoline_base = __va(SMP_TRAMPOLINE_BASE);
 
@@ -110,6 +140,96 @@ void unmap_cpu_to_logical_apicid(int cpu)
 #define map_cpu_to_logical_apicid()  do {} while (0)
 #endif
 
+/*
+ * Report back to the Boot Processor.
+ * Running on AP.
+ */
+void __cpuinit smp_callin(void)
+{
+	int cpuid, phys_id;
+	unsigned long timeout;
+
+	/*
+	 * If waken up by an INIT in an 82489DX configuration
+	 * we may get here before an INIT-deassert IPI reaches
+	 * our local APIC.  We have to wait for the IPI or we'll
+	 * lock up on an APIC access.
+	 */
+	wait_for_init_deassert(&init_deasserted);
+
+	/*
+	 * (This works even if the APIC is not enabled.)
+	 */
+	phys_id = GET_APIC_ID(apic_read(APIC_ID));
+	cpuid = smp_processor_id();
+	if (cpu_isset(cpuid, cpu_callin_map)) {
+		panic("%s: phys CPU#%d, CPU#%d already present??\n", __func__,
+					phys_id, cpuid);
+	}
+	Dprintk("CPU#%d (phys ID: %d) waiting for CALLOUT\n", cpuid, phys_id);
+
+	/*
+	 * STARTUP IPIs are fragile beasts as they might sometimes
+	 * trigger some glue motherboard logic. Complete APIC bus
+	 * silence for 1 second, this overestimates the time the
+	 * boot CPU is spending to send the up to 2 STARTUP IPIs
+	 * by a factor of two. This should be enough.
+	 */
+
+	/*
+	 * Waiting 2s total for startup (udelay is not yet working)
+	 */
+	timeout = jiffies + 2*HZ;
+	while (time_before(jiffies, timeout)) {
+		/*
+		 * Has the boot CPU finished it's STARTUP sequence?
+		 */
+		if (cpu_isset(cpuid, cpu_callout_map))
+			break;
+		cpu_relax();
+	}
+
+	if (!time_before(jiffies, timeout)) {
+		panic("%s: CPU%d started up but did not get a callout!\n",
+		      __func__, cpuid);
+	}
+
+	/*
+	 * the boot CPU has finished the init stage and is spinning
+	 * on callin_map until we finish. We are free to set up this
+	 * CPU, first the APIC. (this is probably redundant on most
+	 * boards)
+	 */
+
+	Dprintk("CALLIN, before setup_local_APIC().\n");
+	smp_callin_clear_local_apic();
+	setup_local_APIC();
+	end_local_APIC_setup();
+	map_cpu_to_logical_apicid();
+
+	/*
+	 * Get our bogomips.
+	 *
+	 * Need to enable IRQs because it can take longer and then
+	 * the NMI watchdog might kill us.
+	 */
+	local_irq_enable();
+	calibrate_delay();
+	local_irq_disable();
+	Dprintk("Stack at about %p\n", &cpuid);
+
+	/*
+	 * Save our processor parameters
+	 */
+	smp_store_cpu_info(cpuid);
+
+	/*
+	 * Allow the master to continue.
+	 */
+	cpu_set(cpuid, cpu_callin_map);
+}
+
+
 static void __cpuinit smp_apply_quirks(struct cpuinfo_x86 *c)
 {
 #ifdef CONFIG_X86_32
@@ -327,6 +447,474 @@ void impress_friends(void)
 	Dprintk("Before bogocount - setting activated=1.\n");
 }
 
+static inline void __inquire_remote_apic(int apicid)
+{
+	unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 };
+	char *names[] = { "ID", "VERSION", "SPIV" };
+	int timeout;
+	u32 status;
+
+	printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid);
+
+	for (i = 0; i < ARRAY_SIZE(regs); i++) {
+		printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]);
+
+		/*
+		 * Wait for idle.
+		 */
+		status = safe_apic_wait_icr_idle();
+		if (status)
+			printk(KERN_CONT
+			       "a previous APIC delivery may have failed\n");
+
+		apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(apicid));
+		apic_write_around(APIC_ICR, APIC_DM_REMRD | regs[i]);
+
+		timeout = 0;
+		do {
+			udelay(100);
+			status = apic_read(APIC_ICR) & APIC_ICR_RR_MASK;
+		} while (status == APIC_ICR_RR_INPROG && timeout++ < 1000);
+
+		switch (status) {
+		case APIC_ICR_RR_VALID:
+			status = apic_read(APIC_RRR);
+			printk(KERN_CONT "%08x\n", status);
+			break;
+		default:
+			printk(KERN_CONT "failed\n");
+		}
+	}
+}
+
+#ifdef WAKE_SECONDARY_VIA_NMI
+/*
+ * Poke the other CPU in the eye via NMI to wake it up. Remember that the normal
+ * INIT, INIT, STARTUP sequence will reset the chip hard for us, and this
+ * won't ... remember to clear down the APIC, etc later.
+ */
+static int __devinit
+wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
+{
+	unsigned long send_status, accept_status = 0;
+	int maxlvt;
+
+	/* Target chip */
+	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(logical_apicid));
+
+	/* Boot on the stack */
+	/* Kick the second */
+	apic_write_around(APIC_ICR, APIC_DM_NMI | APIC_DEST_LOGICAL);
+
+	Dprintk("Waiting for send to finish...\n");
+	send_status = safe_apic_wait_icr_idle();
+
+	/*
+	 * Give the other CPU some time to accept the IPI.
+	 */
+	udelay(200);
+	/*
+	 * Due to the Pentium erratum 3AP.
+	 */
+	maxlvt = lapic_get_maxlvt();
+	if (maxlvt > 3) {
+		apic_read_around(APIC_SPIV);
+		apic_write(APIC_ESR, 0);
+	}
+	accept_status = (apic_read(APIC_ESR) & 0xEF);
+	Dprintk("NMI sent.\n");
+
+	if (send_status)
+		printk(KERN_ERR "APIC never delivered???\n");
+	if (accept_status)
+		printk(KERN_ERR "APIC delivery error (%lx).\n", accept_status);
+
+	return (send_status | accept_status);
+}
+#endif	/* WAKE_SECONDARY_VIA_NMI */
+
+extern void start_secondary(void *unused);
+#ifdef WAKE_SECONDARY_VIA_INIT
+static int __devinit
+wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
+{
+	unsigned long send_status, accept_status = 0;
+	int maxlvt, num_starts, j;
+
+	/*
+	 * Be paranoid about clearing APIC errors.
+	 */
+	if (APIC_INTEGRATED(apic_version[phys_apicid])) {
+		apic_read_around(APIC_SPIV);
+		apic_write(APIC_ESR, 0);
+		apic_read(APIC_ESR);
+	}
+
+	Dprintk("Asserting INIT.\n");
+
+	/*
+	 * Turn INIT on target chip
+	 */
+	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
+
+	/*
+	 * Send IPI
+	 */
+	apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_INT_ASSERT
+				| APIC_DM_INIT);
+
+	Dprintk("Waiting for send to finish...\n");
+	send_status = safe_apic_wait_icr_idle();
+
+	mdelay(10);
+
+	Dprintk("Deasserting INIT.\n");
+
+	/* Target chip */
+	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
+
+	/* Send IPI */
+	apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_DM_INIT);
+
+	Dprintk("Waiting for send to finish...\n");
+	send_status = safe_apic_wait_icr_idle();
+
+	mb();
+	atomic_set(&init_deasserted, 1);
+
+	/*
+	 * Should we send STARTUP IPIs ?
+	 *
+	 * Determine this based on the APIC version.
+	 * If we don't have an integrated APIC, don't send the STARTUP IPIs.
+	 */
+	if (APIC_INTEGRATED(apic_version[phys_apicid]))
+		num_starts = 2;
+	else
+		num_starts = 0;
+
+	/*
+	 * Paravirt / VMI wants a startup IPI hook here to set up the
+	 * target processor state.
+	 */
+	startup_ipi_hook(phys_apicid, (unsigned long) start_secondary,
+#ifdef CONFIG_X86_64
+			 (unsigned long)init_rsp);
+#else
+			 (unsigned long)stack_start.sp);
+#endif
+
+	/*
+	 * Run STARTUP IPI loop.
+	 */
+	Dprintk("#startup loops: %d.\n", num_starts);
+
+	maxlvt = lapic_get_maxlvt();
+
+	for (j = 1; j <= num_starts; j++) {
+		Dprintk("Sending STARTUP #%d.\n", j);
+		apic_read_around(APIC_SPIV);
+		apic_write(APIC_ESR, 0);
+		apic_read(APIC_ESR);
+		Dprintk("After apic_write.\n");
+
+		/*
+		 * STARTUP IPI
+		 */
+
+		/* Target chip */
+		apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
+
+		/* Boot on the stack */
+		/* Kick the second */
+		apic_write_around(APIC_ICR, APIC_DM_STARTUP
+					| (start_eip >> 12));
+
+		/*
+		 * Give the other CPU some time to accept the IPI.
+		 */
+		udelay(300);
+
+		Dprintk("Startup point 1.\n");
+
+		Dprintk("Waiting for send to finish...\n");
+		send_status = safe_apic_wait_icr_idle();
+
+		/*
+		 * Give the other CPU some time to accept the IPI.
+		 */
+		udelay(200);
+		/*
+		 * Due to the Pentium erratum 3AP.
+		 */
+		if (maxlvt > 3) {
+			apic_read_around(APIC_SPIV);
+			apic_write(APIC_ESR, 0);
+		}
+		accept_status = (apic_read(APIC_ESR) & 0xEF);
+		if (send_status || accept_status)
+			break;
+	}
+	Dprintk("After Startup.\n");
+
+	if (send_status)
+		printk(KERN_ERR "APIC never delivered???\n");
+	if (accept_status)
+		printk(KERN_ERR "APIC delivery error (%lx).\n", accept_status);
+
+	return (send_status | accept_status);
+}
+#endif	/* WAKE_SECONDARY_VIA_INIT */
+
+struct create_idle {
+	struct work_struct work;
+	struct task_struct *idle;
+	struct completion done;
+	int cpu;
+};
+
+static void __cpuinit do_fork_idle(struct work_struct *work)
+{
+	struct create_idle *c_idle =
+		container_of(work, struct create_idle, work);
+
+	c_idle->idle = fork_idle(c_idle->cpu);
+	complete(&c_idle->done);
+}
+
+static int __cpuinit do_boot_cpu(int apicid, int cpu)
+/*
+ * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
+ * (ie clustered apic addressing mode), this is a LOGICAL apic ID.
+ * Returns zero if CPU booted OK, else error code from wakeup_secondary_cpu.
+ */
+{
+	unsigned long boot_error = 0;
+	int timeout;
+	unsigned long start_ip;
+	unsigned short nmi_high = 0, nmi_low = 0;
+	struct create_idle c_idle = {
+		.cpu = cpu,
+		.done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done),
+	};
+	INIT_WORK(&c_idle.work, do_fork_idle);
+#ifdef CONFIG_X86_64
+	/* allocate memory for gdts of secondary cpus. Hotplug is considered */
+	if (!cpu_gdt_descr[cpu].address &&
+		!(cpu_gdt_descr[cpu].address = get_zeroed_page(GFP_KERNEL))) {
+		printk(KERN_ERR "Failed to allocate GDT for CPU %d\n", cpu);
+		return -1;
+	}
+
+	/* Allocate node local memory for AP pdas */
+	if (cpu_pda(cpu) == &boot_cpu_pda[cpu]) {
+		struct x8664_pda *newpda, *pda;
+		int node = cpu_to_node(cpu);
+		pda = cpu_pda(cpu);
+		newpda = kmalloc_node(sizeof(struct x8664_pda), GFP_ATOMIC,
+				      node);
+		if (newpda) {
+			memcpy(newpda, pda, sizeof(struct x8664_pda));
+			cpu_pda(cpu) = newpda;
+		} else
+			printk(KERN_ERR
+		"Could not allocate node local PDA for CPU %d on node %d\n",
+				cpu, node);
+	}
+#endif
+
+	alternatives_smp_switch(1);
+
+	c_idle.idle = get_idle_for_cpu(cpu);
+
+	/*
+	 * We can't use kernel_thread since we must avoid to
+	 * reschedule the child.
+	 */
+	if (c_idle.idle) {
+		c_idle.idle->thread.sp = (unsigned long) (((struct pt_regs *)
+			(THREAD_SIZE +  task_stack_page(c_idle.idle))) - 1);
+		init_idle(c_idle.idle, cpu);
+		goto do_rest;
+	}
+
+	if (!keventd_up() || current_is_keventd())
+		c_idle.work.func(&c_idle.work);
+	else {
+		schedule_work(&c_idle.work);
+		wait_for_completion(&c_idle.done);
+	}
+
+	if (IS_ERR(c_idle.idle)) {
+		printk("failed fork for CPU %d\n", cpu);
+		return PTR_ERR(c_idle.idle);
+	}
+
+	set_idle_for_cpu(cpu, c_idle.idle);
+do_rest:
+#ifdef CONFIG_X86_32
+	per_cpu(current_task, cpu) = c_idle.idle;
+	init_gdt(cpu);
+	early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu);
+	c_idle.idle->thread.ip = (unsigned long) start_secondary;
+	/* Stack for startup_32 can be just as for start_secondary onwards */
+	stack_start.sp = (void *) c_idle.idle->thread.sp;
+	irq_ctx_init(cpu);
+#else
+	cpu_pda(cpu)->pcurrent = c_idle.idle;
+	init_rsp = c_idle.idle->thread.sp;
+	load_sp0(&per_cpu(init_tss, cpu), &c_idle.idle->thread);
+	initial_code = (unsigned long)start_secondary;
+	clear_tsk_thread_flag(c_idle.idle, TIF_FORK);
+#endif
+
+	/* start_ip had better be page-aligned! */
+	start_ip = setup_trampoline();
+
+	/* So we see what's up   */
+	printk(KERN_INFO "Booting processor %d/%d ip %lx\n",
+			  cpu, apicid, start_ip);
+
+	/*
+	 * This grunge runs the startup process for
+	 * the targeted processor.
+	 */
+
+	atomic_set(&init_deasserted, 0);
+
+	Dprintk("Setting warm reset code and vector.\n");
+
+	store_NMI_vector(&nmi_high, &nmi_low);
+
+	smpboot_setup_warm_reset_vector(start_ip);
+	/*
+	 * Be paranoid about clearing APIC errors.
+	 */
+	apic_write(APIC_ESR, 0);
+	apic_read(APIC_ESR);
+
+
+	/*
+	 * Starting actual IPI sequence...
+	 */
+	boot_error = wakeup_secondary_cpu(apicid, start_ip);
+
+	if (!boot_error) {
+		/*
+		 * allow APs to start initializing.
+		 */
+		Dprintk("Before Callout %d.\n", cpu);
+		cpu_set(cpu, cpu_callout_map);
+		Dprintk("After Callout %d.\n", cpu);
+
+		/*
+		 * Wait 5s total for a response
+		 */
+		for (timeout = 0; timeout < 50000; timeout++) {
+			if (cpu_isset(cpu, cpu_callin_map))
+				break;	/* It has booted */
+			udelay(100);
+		}
+
+		if (cpu_isset(cpu, cpu_callin_map)) {
+			/* number CPUs logically, starting from 1 (BSP is 0) */
+			Dprintk("OK.\n");
+			printk(KERN_INFO "CPU%d: ", cpu);
+			print_cpu_info(&cpu_data(cpu));
+			Dprintk("CPU has booted.\n");
+		} else {
+			boot_error = 1;
+			if (*((volatile unsigned char *)trampoline_base)
+					== 0xA5)
+				/* trampoline started but...? */
+				printk(KERN_ERR "Stuck ??\n");
+			else
+				/* trampoline code not run */
+				printk(KERN_ERR "Not responding.\n");
+			inquire_remote_apic(apicid);
+		}
+	}
+
+	if (boot_error) {
+		/* Try to put things back the way they were before ... */
+		unmap_cpu_to_logical_apicid(cpu);
+#ifdef CONFIG_X86_64
+		clear_node_cpumask(cpu); /* was set by numa_add_cpu */
+#endif
+		cpu_clear(cpu, cpu_callout_map); /* was set by do_boot_cpu() */
+		cpu_clear(cpu, cpu_initialized); /* was set by cpu_init() */
+		cpu_clear(cpu, cpu_possible_map);
+		cpu_clear(cpu, cpu_present_map);
+		per_cpu(x86_cpu_to_apicid, cpu) = BAD_APICID;
+	}
+
+	/* mark "stuck" area as not stuck */
+	*((volatile unsigned long *)trampoline_base) = 0;
+
+	return boot_error;
+}
+
+int __cpuinit native_cpu_up(unsigned int cpu)
+{
+	int apicid = cpu_present_to_apicid(cpu);
+	unsigned long flags;
+	int err;
+
+	WARN_ON(irqs_disabled());
+
+	Dprintk("++++++++++++++++++++=_---CPU UP  %u\n", cpu);
+
+	if (apicid == BAD_APICID || apicid == boot_cpu_physical_apicid ||
+	    !physid_isset(apicid, phys_cpu_present_map)) {
+		printk(KERN_ERR "%s: bad cpu %d\n", __func__, cpu);
+		return -EINVAL;
+	}
+
+	/*
+	 * Already booted CPU?
+	 */
+	if (cpu_isset(cpu, cpu_callin_map)) {
+		Dprintk("do_boot_cpu %d Already started\n", cpu);
+		return -ENOSYS;
+	}
+
+	/*
+	 * Save current MTRR state in case it was changed since early boot
+	 * (e.g. by the ACPI SMI) to initialize new CPUs with MTRRs in sync:
+	 */
+	mtrr_save_state();
+
+	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
+
+#ifdef CONFIG_X86_32
+	/* init low mem mapping */
+	clone_pgd_range(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS,
+			min_t(unsigned long, KERNEL_PGD_PTRS, USER_PGD_PTRS));
+	flush_tlb_all();
+#endif
+
+	err = do_boot_cpu(apicid, cpu);
+	if (err < 0) {
+		Dprintk("do_boot_cpu failed %d\n", err);
+		return err;
+	}
+
+	/*
+	 * Check TSC synchronization with the AP (keep irqs disabled
+	 * while doing so):
+	 */
+	local_irq_save(flags);
+	check_tsc_sync_source(cpu);
+	local_irq_restore(flags);
+
+	while (!cpu_isset(cpu, cpu_online_map)) {
+		cpu_relax();
+		touch_nmi_watchdog();
+	}
+
+	return 0;
+}
+
 #ifdef CONFIG_HOTPLUG_CPU
 void remove_siblinginfo(int cpu)
 {
diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index ae25927f08c1..e82eeb2fdfef 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -80,114 +80,12 @@ extern void unmap_cpu_to_logical_apicid(int cpu);
 /* State of each CPU. */
 DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
-/* Store all idle threads, this can be reused instead of creating
-* a new thread. Also avoids complicated thread destroy functionality
-* for idle threads.
-*/
-#ifdef CONFIG_HOTPLUG_CPU
-/*
- * Needed only for CONFIG_HOTPLUG_CPU because __cpuinitdata is
- * removed after init for !CONFIG_HOTPLUG_CPU.
- */
-static DEFINE_PER_CPU(struct task_struct *, idle_thread_array);
-#define get_idle_for_cpu(x)      (per_cpu(idle_thread_array, x))
-#define set_idle_for_cpu(x, p)   (per_cpu(idle_thread_array, x) = (p))
-#else
-struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
-#define get_idle_for_cpu(x)      (idle_thread_array[(x)])
-#define set_idle_for_cpu(x, p)   (idle_thread_array[(x)] = (p))
-#endif
-
-static atomic_t init_deasserted;
-
-static void __cpuinit smp_callin(void)
-{
-	int cpuid, phys_id;
-	unsigned long timeout;
-
-	/*
-	 * If waken up by an INIT in an 82489DX configuration
-	 * we may get here before an INIT-deassert IPI reaches
-	 * our local APIC.  We have to wait for the IPI or we'll
-	 * lock up on an APIC access.
-	 */
-	wait_for_init_deassert(&init_deasserted);
-
-	/*
-	 * (This works even if the APIC is not enabled.)
-	 */
-	phys_id = GET_APIC_ID(apic_read(APIC_ID));
-	cpuid = smp_processor_id();
-	if (cpu_isset(cpuid, cpu_callin_map)) {
-		printk("huh, phys CPU#%d, CPU#%d already present??\n",
-					phys_id, cpuid);
-		BUG();
-	}
-	Dprintk("CPU#%d (phys ID: %d) waiting for CALLOUT\n", cpuid, phys_id);
-
-	/*
-	 * STARTUP IPIs are fragile beasts as they might sometimes
-	 * trigger some glue motherboard logic. Complete APIC bus
-	 * silence for 1 second, this overestimates the time the
-	 * boot CPU is spending to send the up to 2 STARTUP IPIs
-	 * by a factor of two. This should be enough.
-	 */
-
-	/*
-	 * Waiting 2s total for startup (udelay is not yet working)
-	 */
-	timeout = jiffies + 2*HZ;
-	while (time_before(jiffies, timeout)) {
-		/*
-		 * Has the boot CPU finished it's STARTUP sequence?
-		 */
-		if (cpu_isset(cpuid, cpu_callout_map))
-			break;
-		cpu_relax();
-	}
-
-	if (!time_before(jiffies, timeout)) {
-		printk("BUG: CPU%d started up but did not get a callout!\n",
-			cpuid);
-		BUG();
-	}
-
-	/*
-	 * the boot CPU has finished the init stage and is spinning
-	 * on callin_map until we finish. We are free to set up this
-	 * CPU, first the APIC. (this is probably redundant on most
-	 * boards)
-	 */
-
-	Dprintk("CALLIN, before setup_local_APIC().\n");
-	smp_callin_clear_local_apic();
-	setup_local_APIC();
-	end_local_APIC_setup();
-	map_cpu_to_logical_apicid();
-
-	/*
-	 * Get our bogomips.
-	 */
-	local_irq_enable();
-	calibrate_delay();
-	local_irq_disable();
-	Dprintk("Stack at about %p\n",&cpuid);
-
-	/*
-	 * Save our processor parameters
-	 */
-	smp_store_cpu_info(cpuid);
-
-	/*
-	 * Allow the master to continue.
-	 */
-	cpu_set(cpuid, cpu_callin_map);
-}
+extern void smp_callin(void);
 
 /*
  * Activate a secondary processor.
  */
-static void __cpuinit start_secondary(void *unused)
+void __cpuinit start_secondary(void *unused)
 {
 	/*
 	 * Don't put *anything* before cpu_init(), SMP booting is too
@@ -257,373 +155,6 @@ void __devinit initialize_secondary(void)
 		:"m" (current->thread.sp),"m" (current->thread.ip));
 }
 
-static inline void __inquire_remote_apic(int apicid)
-{
-	unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 };
-	char *names[] = { "ID", "VERSION", "SPIV" };
-	int timeout;
-	u32 status;
-
-	printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid);
-
-	for (i = 0; i < ARRAY_SIZE(regs); i++) {
-		printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]);
-
-		/*
-		 * Wait for idle.
-		 */
-		status = safe_apic_wait_icr_idle();
-		if (status)
-			printk(KERN_CONT
-			       "a previous APIC delivery may have failed\n");
-
-		apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(apicid));
-		apic_write_around(APIC_ICR, APIC_DM_REMRD | regs[i]);
-
-		timeout = 0;
-		do {
-			udelay(100);
-			status = apic_read(APIC_ICR) & APIC_ICR_RR_MASK;
-		} while (status == APIC_ICR_RR_INPROG && timeout++ < 1000);
-
-		switch (status) {
-		case APIC_ICR_RR_VALID:
-			status = apic_read(APIC_RRR);
-			printk(KERN_CONT "%08x\n", status);
-			break;
-		default:
-			printk(KERN_CONT "failed\n");
-		}
-	}
-}
-
-#ifdef WAKE_SECONDARY_VIA_NMI
-/* 
- * Poke the other CPU in the eye via NMI to wake it up. Remember that the normal
- * INIT, INIT, STARTUP sequence will reset the chip hard for us, and this
- * won't ... remember to clear down the APIC, etc later.
- */
-static int __devinit
-wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
-{
-	unsigned long send_status, accept_status = 0;
-	int maxlvt;
-
-	/* Target chip */
-	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(logical_apicid));
-
-	/* Boot on the stack */
-	/* Kick the second */
-	apic_write_around(APIC_ICR, APIC_DM_NMI | APIC_DEST_LOGICAL);
-
-	Dprintk("Waiting for send to finish...\n");
-	send_status = safe_apic_wait_icr_idle();
-
-	/*
-	 * Give the other CPU some time to accept the IPI.
-	 */
-	udelay(200);
-	/*
-	 * Due to the Pentium erratum 3AP.
-	 */
-	maxlvt = lapic_get_maxlvt();
-	if (maxlvt > 3) {
-		apic_read_around(APIC_SPIV);
-		apic_write(APIC_ESR, 0);
-	}
-	accept_status = (apic_read(APIC_ESR) & 0xEF);
-	Dprintk("NMI sent.\n");
-
-	if (send_status)
-		printk("APIC never delivered???\n");
-	if (accept_status)
-		printk("APIC delivery error (%lx).\n", accept_status);
-
-	return (send_status | accept_status);
-}
-#endif	/* WAKE_SECONDARY_VIA_NMI */
-
-#ifdef WAKE_SECONDARY_VIA_INIT
-static int __devinit
-wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
-{
-	unsigned long send_status, accept_status = 0;
-	int maxlvt, num_starts, j;
-
-	/*
-	 * Be paranoid about clearing APIC errors.
-	 */
-	if (APIC_INTEGRATED(apic_version[phys_apicid])) {
-		apic_read_around(APIC_SPIV);
-		apic_write(APIC_ESR, 0);
-		apic_read(APIC_ESR);
-	}
-
-	Dprintk("Asserting INIT.\n");
-
-	/*
-	 * Turn INIT on target chip
-	 */
-	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
-	/*
-	 * Send IPI
-	 */
-	apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_INT_ASSERT
-				| APIC_DM_INIT);
-
-	Dprintk("Waiting for send to finish...\n");
-	send_status = safe_apic_wait_icr_idle();
-
-	mdelay(10);
-
-	Dprintk("Deasserting INIT.\n");
-
-	/* Target chip */
-	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
-	/* Send IPI */
-	apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_DM_INIT);
-
-	Dprintk("Waiting for send to finish...\n");
-	send_status = safe_apic_wait_icr_idle();
-
-	mb();
-	atomic_set(&init_deasserted, 1);
-
-	/*
-	 * Should we send STARTUP IPIs ?
-	 *
-	 * Determine this based on the APIC version.
-	 * If we don't have an integrated APIC, don't send the STARTUP IPIs.
-	 */
-	if (APIC_INTEGRATED(apic_version[phys_apicid]))
-		num_starts = 2;
-	else
-		num_starts = 0;
-
-	/*
-	 * Paravirt / VMI wants a startup IPI hook here to set up the
-	 * target processor state.
-	 */
-	startup_ipi_hook(phys_apicid, (unsigned long) start_secondary,
-		         (unsigned long) stack_start.sp);
-
-	/*
-	 * Run STARTUP IPI loop.
-	 */
-	Dprintk("#startup loops: %d.\n", num_starts);
-
-	maxlvt = lapic_get_maxlvt();
-
-	for (j = 1; j <= num_starts; j++) {
-		Dprintk("Sending STARTUP #%d.\n",j);
-		apic_read_around(APIC_SPIV);
-		apic_write(APIC_ESR, 0);
-		apic_read(APIC_ESR);
-		Dprintk("After apic_write.\n");
-
-		/*
-		 * STARTUP IPI
-		 */
-
-		/* Target chip */
-		apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
-		/* Boot on the stack */
-		/* Kick the second */
-		apic_write_around(APIC_ICR, APIC_DM_STARTUP
-					| (start_eip >> 12));
-
-		/*
-		 * Give the other CPU some time to accept the IPI.
-		 */
-		udelay(300);
-
-		Dprintk("Startup point 1.\n");
-
-		Dprintk("Waiting for send to finish...\n");
-		send_status = safe_apic_wait_icr_idle();
-
-		/*
-		 * Give the other CPU some time to accept the IPI.
-		 */
-		udelay(200);
-		/*
-		 * Due to the Pentium erratum 3AP.
-		 */
-		if (maxlvt > 3) {
-			apic_read_around(APIC_SPIV);
-			apic_write(APIC_ESR, 0);
-		}
-		accept_status = (apic_read(APIC_ESR) & 0xEF);
-		if (send_status || accept_status)
-			break;
-	}
-	Dprintk("After Startup.\n");
-
-	if (send_status)
-		printk("APIC never delivered???\n");
-	if (accept_status)
-		printk("APIC delivery error (%lx).\n", accept_status);
-
-	return (send_status | accept_status);
-}
-#endif	/* WAKE_SECONDARY_VIA_INIT */
-
-extern cpumask_t cpu_initialized;
-
-struct create_idle {
-	struct work_struct work;
-	struct task_struct *idle;
-	struct completion done;
-	int cpu;
-};
-
-static void __cpuinit do_fork_idle(struct work_struct *work)
-{
-	struct create_idle *c_idle =
-		container_of(work, struct create_idle, work);
-
-	c_idle->idle = fork_idle(c_idle->cpu);
-	complete(&c_idle->done);
-}
-static int __cpuinit do_boot_cpu(int apicid, int cpu)
-/*
- * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
- * (ie clustered apic addressing mode), this is a LOGICAL apic ID.
- * Returns zero if CPU booted OK, else error code from wakeup_secondary_cpu.
- */
-{
-	unsigned long boot_error = 0;
-	int timeout;
-	unsigned long start_eip;
-	unsigned short nmi_high = 0, nmi_low = 0;
-	struct create_idle c_idle = {
-		.cpu = cpu,
-		.done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done),
-	};
-	INIT_WORK(&c_idle.work, do_fork_idle);
-
-	alternatives_smp_switch(1);
-
-	c_idle.idle = get_idle_for_cpu(cpu);
-
-	/*
-	 * We can't use kernel_thread since we must avoid to
-	 * reschedule the child.
-	 */
-	if (c_idle.idle) {
-		c_idle.idle->thread.sp = (unsigned long) (((struct pt_regs *)
-			(THREAD_SIZE +  task_stack_page(c_idle.idle))) - 1);
-		init_idle(c_idle.idle, cpu);
-		goto do_rest;
-	}
-
-	if (!keventd_up() || current_is_keventd())
-		c_idle.work.func(&c_idle.work);
-	else {
-		schedule_work(&c_idle.work);
-		wait_for_completion(&c_idle.done);
-	}
-
-	if (IS_ERR(c_idle.idle)) {
-		printk(KERN_ERR "failed fork for CPU %d\n", cpu);
-		return PTR_ERR(c_idle.idle);
-	}
-
-	set_idle_for_cpu(cpu, c_idle.idle);
-do_rest:
-	per_cpu(current_task, cpu) = c_idle.idle;
-	init_gdt(cpu);
-	early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu);
-
-	c_idle.idle->thread.ip = (unsigned long) start_secondary;
-	/* start_eip had better be page-aligned! */
-	start_eip = setup_trampoline();
-
-	/* So we see what's up   */
-	printk("Booting processor %d/%d ip %lx\n", cpu, apicid, start_eip);
-	/* Stack for startup_32 can be just as for start_secondary onwards */
-	stack_start.sp = (void *) c_idle.idle->thread.sp;
-
-	irq_ctx_init(cpu);
-
-	/*
-	 * This grunge runs the startup process for
-	 * the targeted processor.
-	 */
-
-	atomic_set(&init_deasserted, 0);
-
-	Dprintk("Setting warm reset code and vector.\n");
-
-	store_NMI_vector(&nmi_high, &nmi_low);
-
-	smpboot_setup_warm_reset_vector(start_eip);
-	/*
-	 * Be paranoid about clearing APIC errors.
-	 */
-	apic_write(APIC_ESR, 0);
-	apic_read(APIC_ESR);
-
-
-	/*
-	 * Starting actual IPI sequence...
-	 */
-	boot_error = wakeup_secondary_cpu(apicid, start_eip);
-
-	if (!boot_error) {
-		/*
-		 * allow APs to start initializing.
-		 */
-		Dprintk("Before Callout %d.\n", cpu);
-		cpu_set(cpu, cpu_callout_map);
-		Dprintk("After Callout %d.\n", cpu);
-
-		/*
-		 * Wait 5s total for a response
-		 */
-		for (timeout = 0; timeout < 50000; timeout++) {
-			if (cpu_isset(cpu, cpu_callin_map))
-				break;	/* It has booted */
-			udelay(100);
-		}
-
-		if (cpu_isset(cpu, cpu_callin_map)) {
-			/* number CPUs logically, starting from 1 (BSP is 0) */
-			Dprintk("OK.\n");
-			printk("CPU%d: ", cpu);
-			print_cpu_info(&cpu_data(cpu));
-			Dprintk("CPU has booted.\n");
-		} else {
-			boot_error= 1;
-			if (*((volatile unsigned char *)trampoline_base)
-					== 0xA5)
-				/* trampoline started but...? */
-				printk("Stuck ??\n");
-			else
-				/* trampoline code not run */
-				printk("Not responding.\n");
-			inquire_remote_apic(apicid);
-		}
-	}
-
-	if (boot_error) {
-		/* Try to put things back the way they were before ... */
-		unmap_cpu_to_logical_apicid(cpu);
-		cpu_clear(cpu, cpu_callout_map); /* was set here (do_boot_cpu()) */
-		cpu_clear(cpu, cpu_initialized); /* was set by cpu_init() */
-		cpu_clear(cpu, cpu_possible_map);
-		per_cpu(x86_cpu_to_apicid, cpu) = BAD_APICID;
-	}
-
-	/* mark "stuck" area as not stuck */
-	*((volatile unsigned long *)trampoline_base) = 0;
-
-	return boot_error;
-}
-
 #ifdef CONFIG_HOTPLUG_CPU
 void cpu_exit_clear(void)
 {
@@ -774,65 +305,6 @@ void __init native_smp_prepare_boot_cpu(void)
 	__get_cpu_var(cpu_state) = CPU_ONLINE;
 }
 
-int __cpuinit native_cpu_up(unsigned int cpu)
-{
-	int apicid = cpu_present_to_apicid(cpu);
-	unsigned long flags;
-	int err;
-
-	WARN_ON(irqs_disabled());
-
-	Dprintk("++++++++++++++++++++=_---CPU UP  %u\n", cpu);
-
-	if (apicid == BAD_APICID || apicid == boot_cpu_physical_apicid ||
-	    !physid_isset(apicid, phys_cpu_present_map)) {
-		printk(KERN_ERR "%s: bad cpu %d\n", __func__, cpu);
-		return -EINVAL;
-	}
-
-	/*
-	 * Already booted CPU?
-	 */
-	if (cpu_isset(cpu, cpu_callin_map)) {
-		Dprintk("do_boot_cpu %d Already started\n", cpu);
-		return -ENOSYS;
-	}
-
-	/*
-	 * Save current MTRR state in case it was changed since early boot
-	 * (e.g. by the ACPI SMI) to initialize new CPUs with MTRRs in sync:
-	 */
-	mtrr_save_state();
-
-	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
-
-	/* init low mem mapping */
-	clone_pgd_range(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS,
-			min_t(unsigned long, KERNEL_PGD_PTRS, USER_PGD_PTRS));
-	flush_tlb_all();
-
-	err = do_boot_cpu(apicid, cpu);
-	if (err < 0) {
-		Dprintk("do_boot_cpu failed %d\n", err);
-		return err;
-	}
-
-	/*
-	 * Check TSC synchronization with the AP (keep irqs disabled
-	 * while doing so):
-	 */
-	local_irq_save(flags);
-	check_tsc_sync_source(cpu);
-	local_irq_restore(flags);
-
-	while (!cpu_isset(cpu, cpu_online_map)) {
-		cpu_relax();
-		touch_nmi_watchdog();
-	}
-
-	return 0;
-}
-
 extern void impress_friends(void);
 extern void smp_checks(void);
 
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index 420ae4a33548..71f13b15bd89 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -71,119 +71,7 @@ int smp_threads_ready;
 /* State of each CPU */
 DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
-/*
- * Store all idle threads, this can be reused instead of creating
- * a new thread. Also avoids complicated thread destroy functionality
- * for idle threads.
- */
-#ifdef CONFIG_HOTPLUG_CPU
-/*
- * Needed only for CONFIG_HOTPLUG_CPU because __cpuinitdata is
- * removed after init for !CONFIG_HOTPLUG_CPU.
- */
-static DEFINE_PER_CPU(struct task_struct *, idle_thread_array);
-#define get_idle_for_cpu(x)     (per_cpu(idle_thread_array, x))
-#define set_idle_for_cpu(x,p)   (per_cpu(idle_thread_array, x) = (p))
-#else
-struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
-#define get_idle_for_cpu(x)     (idle_thread_array[(x)])
-#define set_idle_for_cpu(x,p)   (idle_thread_array[(x)] = (p))
-#endif
-
-static atomic_t init_deasserted __cpuinitdata;
-
-#define smp_callin_clear_local_apic() do {} while (0)
-#define map_cpu_to_logical_apicid() do {} while (0)
-
-/*
- * Report back to the Boot Processor.
- * Running on AP.
- */
-void __cpuinit smp_callin(void)
-{
-	int cpuid, phys_id;
-	unsigned long timeout;
-
-	/*
-	 * If waken up by an INIT in an 82489DX configuration
-	 * we may get here before an INIT-deassert IPI reaches
-	 * our local APIC.  We have to wait for the IPI or we'll
-	 * lock up on an APIC access.
-	 */
-	wait_for_init_deassert(&init_deasserted);
-
-	/*
-	 * (This works even if the APIC is not enabled.)
-	 */
-	phys_id = GET_APIC_ID(apic_read(APIC_ID));
-	cpuid = smp_processor_id();
-	if (cpu_isset(cpuid, cpu_callin_map)) {
-		panic("smp_callin: phys CPU#%d, CPU#%d already present??\n",
-					phys_id, cpuid);
-	}
-	Dprintk("CPU#%d (phys ID: %d) waiting for CALLOUT\n", cpuid, phys_id);
-
-	/*
-	 * STARTUP IPIs are fragile beasts as they might sometimes
-	 * trigger some glue motherboard logic. Complete APIC bus
-	 * silence for 1 second, this overestimates the time the
-	 * boot CPU is spending to send the up to 2 STARTUP IPIs
-	 * by a factor of two. This should be enough.
-	 */
-
-	/*
-	 * Waiting 2s total for startup (udelay is not yet working)
-	 */
-	timeout = jiffies + 2*HZ;
-	while (time_before(jiffies, timeout)) {
-		/*
-		 * Has the boot CPU finished it's STARTUP sequence?
-		 */
-		if (cpu_isset(cpuid, cpu_callout_map))
-			break;
-		cpu_relax();
-	}
-
-	if (!time_before(jiffies, timeout)) {
-		panic("smp_callin: CPU%d started up but did not get a callout!\n",
-			cpuid);
-	}
-
-	/*
-	 * the boot CPU has finished the init stage and is spinning
-	 * on callin_map until we finish. We are free to set up this
-	 * CPU, first the APIC. (this is probably redundant on most
-	 * boards)
-	 */
-
-	Dprintk("CALLIN, before setup_local_APIC().\n");
-	smp_callin_clear_local_apic();
-	setup_local_APIC();
-	end_local_APIC_setup();
-	map_cpu_to_logical_apicid();
-
-	/*
-	 * Get our bogomips.
- 	 *
- 	 * Need to enable IRQs because it can take longer and then
-	 * the NMI watchdog might kill us.
-	 */
-	local_irq_enable();
-	calibrate_delay();
-	local_irq_disable();
-	Dprintk("Stack at about %p\n",&cpuid);
-
-	/*
-	 * Save our processor parameters
-	 */
- 	smp_store_cpu_info(cpuid);
-
-	/*
-	 * Allow the master to continue.
-	 */
-	cpu_set(cpuid, cpu_callin_map);
-}
-
+extern void smp_callin(void);
 /*
  * Setup code on secondary processor (after comming out of the trampoline)
  */
@@ -246,349 +134,6 @@ void __cpuinit start_secondary(void)
 	cpu_idle();
 }
 
-extern volatile unsigned long init_rsp;
-extern void (*initial_code)(void);
-
-#ifdef APIC_DEBUG
-static void __inquire_remote_apic(int apicid)
-{
-	unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 };
-	char *names[] = { "ID", "VERSION", "SPIV" };
-	int timeout;
-	u32 status;
-
-	printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid);
-
-	for (i = 0; i < ARRAY_SIZE(regs); i++) {
-		printk(KERN_INFO "... APIC #%d %s: ", apicid, names[i]);
-
-		/*
-		 * Wait for idle.
-		 */
-		status = safe_apic_wait_icr_idle();
-		if (status)
-			printk(KERN_CONT
-			       "a previous APIC delivery may have failed\n");
-
-		apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(apicid));
-		apic_write_around(APIC_ICR, APIC_DM_REMRD | regs[i]);
-
-		timeout = 0;
-		do {
-			udelay(100);
-			status = apic_read(APIC_ICR) & APIC_ICR_RR_MASK;
-		} while (status == APIC_ICR_RR_INPROG && timeout++ < 1000);
-
-		switch (status) {
-		case APIC_ICR_RR_VALID:
-			status = apic_read(APIC_RRR);
-			printk(KERN_CONT "%08x\n", status);
-			break;
-		default:
-			printk(KERN_CONT "failed\n");
-		}
-	}
-}
-#endif
-
-/*
- * Kick the secondary to wake up.
- */
-static int __cpuinit wakeup_secondary_cpu(int phys_apicid,
-					  unsigned int start_rip)
-{
-	unsigned long send_status, accept_status = 0;
-	int maxlvt, num_starts, j;
-
-	/*
-	 * Be paranoid about clearing APIC errors.
-	 */
-	if (APIC_INTEGRATED(apic_version[phys_apicid])) {
-		apic_read_around(APIC_SPIV);
-		apic_write(APIC_ESR, 0);
-		apic_read(APIC_ESR);
-	}
-
-	Dprintk("Asserting INIT.\n");
-
-	/*
-	 * Turn INIT on target chip
-	 */
-	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
-	/*
-	 * Send IPI
-	 */
-	apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_INT_ASSERT
-				| APIC_DM_INIT);
-
-	Dprintk("Waiting for send to finish...\n");
-	send_status = safe_apic_wait_icr_idle();
-
-	mdelay(10);
-
-	Dprintk("Deasserting INIT.\n");
-
-	/* Target chip */
-	apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
-	/* Send IPI */
-	apic_write_around(APIC_ICR, APIC_INT_LEVELTRIG | APIC_DM_INIT);
-
-	Dprintk("Waiting for send to finish...\n");
-	send_status = safe_apic_wait_icr_idle();
-
-	mb();
-	atomic_set(&init_deasserted, 1);
-
-	if (APIC_INTEGRATED(apic_version[phys_apicid]))
-		num_starts = 2;
-	else
-		num_starts = 0;
-
-	/*
-	 * Paravirt / VMI wants a startup IPI hook here to set up the
-	 * target processor state.
-	 */
-	startup_ipi_hook(phys_apicid, (unsigned long) start_secondary,
-			(unsigned long) init_rsp);
-
-
-	/*
-	 * Run STARTUP IPI loop.
-	 */
-	Dprintk("#startup loops: %d.\n", num_starts);
-
-	maxlvt = lapic_get_maxlvt();
-
-	for (j = 1; j <= num_starts; j++) {
-		Dprintk("Sending STARTUP #%d.\n",j);
-		apic_read_around(APIC_SPIV);
-		apic_write(APIC_ESR, 0);
-		apic_read(APIC_ESR);
-		Dprintk("After apic_write.\n");
-
-		/*
-		 * STARTUP IPI
-		 */
-
-		/* Target chip */
-		apic_write_around(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
-		/* Boot on the stack */
-		/* Kick the second */
-		apic_write_around(APIC_ICR, APIC_DM_STARTUP | (start_rip>>12));
-
-		/*
-		 * Give the other CPU some time to accept the IPI.
-		 */
-		udelay(300);
-
-		Dprintk("Startup point 1.\n");
-
-		Dprintk("Waiting for send to finish...\n");
-		send_status = safe_apic_wait_icr_idle();
-
-		/*
-		 * Give the other CPU some time to accept the IPI.
-		 */
-		udelay(200);
-		/*
-		 * Due to the Pentium erratum 3AP.
-		 */
-		if (maxlvt > 3) {
-			apic_read_around(APIC_SPIV);
-			apic_write(APIC_ESR, 0);
-		}
-		accept_status = (apic_read(APIC_ESR) & 0xEF);
-		if (send_status || accept_status)
-			break;
-	}
-	Dprintk("After Startup.\n");
-
-	if (send_status)
-		printk(KERN_ERR "APIC never delivered???\n");
-	if (accept_status)
-		printk(KERN_ERR "APIC delivery error (%lx).\n", accept_status);
-
-	return (send_status | accept_status);
-}
-
-struct create_idle {
-	struct work_struct work;
-	struct task_struct *idle;
-	struct completion done;
-	int cpu;
-};
-
-static void __cpuinit do_fork_idle(struct work_struct *work)
-{
-	struct create_idle *c_idle =
-		container_of(work, struct create_idle, work);
-
-	c_idle->idle = fork_idle(c_idle->cpu);
-	complete(&c_idle->done);
-}
-
-/*
- * Boot one CPU.
- */
-static int __cpuinit do_boot_cpu(int cpu, int apicid)
-{
-	unsigned long boot_error = 0;
-	int timeout;
-	unsigned long start_rip;
-	struct create_idle c_idle = {
-		.cpu = cpu,
-		.done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done),
-	};
-	INIT_WORK(&c_idle.work, do_fork_idle);
-
-	/* allocate memory for gdts of secondary cpus. Hotplug is considered */
-	if (!cpu_gdt_descr[cpu].address &&
-		!(cpu_gdt_descr[cpu].address = get_zeroed_page(GFP_KERNEL))) {
-		printk(KERN_ERR "Failed to allocate GDT for CPU %d\n", cpu);
-		return -1;
-	}
-
-	/* Allocate node local memory for AP pdas */
-	if (cpu_pda(cpu) == &boot_cpu_pda[cpu]) {
-		struct x8664_pda *newpda, *pda;
-		int node = cpu_to_node(cpu);
-		pda = cpu_pda(cpu);
-		newpda = kmalloc_node(sizeof (struct x8664_pda), GFP_ATOMIC,
-				      node);
-		if (newpda) {
-			memcpy(newpda, pda, sizeof (struct x8664_pda));
-			cpu_pda(cpu) = newpda;
-		} else
-			printk(KERN_ERR
-		"Could not allocate node local PDA for CPU %d on node %d\n",
-				cpu, node);
-	}
-
-	alternatives_smp_switch(1);
-
-	c_idle.idle = get_idle_for_cpu(cpu);
-
-	if (c_idle.idle) {
-		c_idle.idle->thread.sp = (unsigned long) (((struct pt_regs *)
-			(THREAD_SIZE +  task_stack_page(c_idle.idle))) - 1);
-		init_idle(c_idle.idle, cpu);
-		goto do_rest;
-	}
-
-	/*
-	 * During cold boot process, keventd thread is not spun up yet.
-	 * When we do cpu hot-add, we create idle threads on the fly, we should
-	 * not acquire any attributes from the calling context. Hence the clean
-	 * way to create kernel_threads() is to do that from keventd().
-	 * We do the current_is_keventd() due to the fact that ACPI notifier
-	 * was also queuing to keventd() and when the caller is already running
-	 * in context of keventd(), we would end up with locking up the keventd
-	 * thread.
-	 */
-	if (!keventd_up() || current_is_keventd())
-		c_idle.work.func(&c_idle.work);
-	else {
-		schedule_work(&c_idle.work);
-		wait_for_completion(&c_idle.done);
-	}
-
-	if (IS_ERR(c_idle.idle)) {
-		printk("failed fork for CPU %d\n", cpu);
-		return PTR_ERR(c_idle.idle);
-	}
-
-	set_idle_for_cpu(cpu, c_idle.idle);
-
-do_rest:
-
-	cpu_pda(cpu)->pcurrent = c_idle.idle;
-
-	start_rip = setup_trampoline();
-
-	init_rsp = c_idle.idle->thread.sp;
-	load_sp0(&per_cpu(init_tss, cpu), &c_idle.idle->thread);
-	initial_code = start_secondary;
-	clear_tsk_thread_flag(c_idle.idle, TIF_FORK);
-
-	printk(KERN_INFO "Booting processor %d/%d APIC 0x%x\n", cpu,
-		cpus_weight(cpu_present_map),
-		apicid);
-
-	/*
-	 * This grunge runs the startup process for
-	 * the targeted processor.
-	 */
-
-	atomic_set(&init_deasserted, 0);
-
-	Dprintk("Setting warm reset code and vector.\n");
-
-	smpboot_setup_warm_reset_vector(start_rip);
-	/*
-	 * Be paranoid about clearing APIC errors.
-	 */
-	apic_write(APIC_ESR, 0);
-	apic_read(APIC_ESR);
-
-	/*
-	 * Starting actual IPI sequence...
-	 */
-	boot_error = wakeup_secondary_cpu(apicid, start_rip);
-
-	if (!boot_error) {
-		/*
-		 * allow APs to start initializing.
-		 */
-		Dprintk("Before Callout %d.\n", cpu);
-		cpu_set(cpu, cpu_callout_map);
-		Dprintk("After Callout %d.\n", cpu);
-
-		/*
-		 * Wait 5s total for a response
-		 */
-		for (timeout = 0; timeout < 50000; timeout++) {
-			if (cpu_isset(cpu, cpu_callin_map))
-				break;	/* It has booted */
-			udelay(100);
-		}
-
-		if (cpu_isset(cpu, cpu_callin_map)) {
-			/* number CPUs logically, starting from 1 (BSP is 0) */
-			Dprintk("CPU has booted.\n");
-			printk(KERN_INFO "CPU%d: ", cpu);
-			print_cpu_info(&cpu_data(cpu));
-		} else {
-			boot_error = 1;
-			if (*((volatile unsigned char *)trampoline_base)
-					== 0xA5)
-				/* trampoline started but...? */
-				printk("Stuck ??\n");
-			else
-				/* trampoline code not run */
-				printk("Not responding.\n");
-#ifdef APIC_DEBUG
-			inquire_remote_apic(apicid);
-#endif
-		}
-	}
-	if (boot_error) {
-		cpu_clear(cpu, cpu_callout_map); /* was set here (do_boot_cpu()) */
-		clear_bit(cpu, (unsigned long *)&cpu_initialized); /* was set by cpu_init() */
-		clear_node_cpumask(cpu); /* was set by numa_add_cpu */
-		cpu_clear(cpu, cpu_present_map);
-		cpu_clear(cpu, cpu_possible_map);
-		per_cpu(x86_cpu_to_apicid, cpu) = BAD_APICID;
-	}
-
-	/* mark "stuck" area as not stuck */
-	*((volatile unsigned long *)trampoline_base) = 0;
-
-	return boot_error;
-}
-
 cycles_t cacheflush_time;
 unsigned long cache_decay_ticks;
 
@@ -745,64 +290,6 @@ void __init native_smp_prepare_boot_cpu(void)
 	per_cpu(cpu_state, me) = CPU_ONLINE;
 }
 
-/*
- * Entry point to boot a CPU.
- */
-int __cpuinit native_cpu_up(unsigned int cpu)
-{
-	int apicid = cpu_present_to_apicid(cpu);
-	unsigned long flags;
-	int err;
-
-	WARN_ON(irqs_disabled());
-
-	Dprintk("++++++++++++++++++++=_---CPU UP  %u\n", cpu);
-
-	if (apicid == BAD_APICID || apicid == boot_cpu_physical_apicid ||
-	    !physid_isset(apicid, phys_cpu_present_map)) {
-		printk("__cpu_up: bad cpu %d\n", cpu);
-		return -EINVAL;
-	}
-
-	/*
-	 * Already booted CPU?
-	 */
- 	if (cpu_isset(cpu, cpu_callin_map)) {
-		Dprintk("do_boot_cpu %d Already started\n", cpu);
- 		return -ENOSYS;
-	}
-
-	/*
-	 * Save current MTRR state in case it was changed since early boot
-	 * (e.g. by the ACPI SMI) to initialize new CPUs with MTRRs in sync:
-	 */
-	mtrr_save_state();
-
-	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
-	/* Boot it! */
-	err = do_boot_cpu(cpu, apicid);
-	if (err < 0) {
-		Dprintk("do_boot_cpu failed %d\n", err);
-		return err;
-	}
-
-	/* Unleash the CPU! */
-	Dprintk("waiting for cpu %d\n", cpu);
-
-	/*
-  	 * Make sure and check TSC sync:
- 	 */
-	local_irq_save(flags);
-	check_tsc_sync_source(cpu);
-	local_irq_restore(flags);
-
-	while (!cpu_isset(cpu, cpu_online_map))
-		cpu_relax();
-	err = 0;
-
-	return err;
-}
-
 extern void impress_friends(void);
 extern void smp_checks(void);
 
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 2ad2f4ffe498..ef26911dc22a 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -9,6 +9,7 @@ extern cpumask_t cpu_callout_map;
 
 extern int smp_num_siblings;
 extern unsigned int num_processors;
+extern cpumask_t cpu_initialized;
 
 extern u16 x86_cpu_to_apicid_init[];
 extern u16 x86_bios_cpu_apicid_init[];
@@ -34,6 +35,8 @@ extern struct {
 	unsigned short ss;
 } stack_start;
 
+extern unsigned long init_rsp;
+extern unsigned long initial_code;
 
 struct smp_ops {
 	void (*smp_prepare_boot_cpu)(void);
-- 
cgit v1.2.3


From 3fa7b3487a1317f7d3be3043dbea316ca75abed5 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:26:06 -0300
Subject: x86: assign nr_ioapics = 0 in smpboot_hooks.h

change smpboot_setup_io_apic() by to match x86_64 behaviour

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mach-default/smpboot_hooks.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/mach-default/smpboot_hooks.h b/include/asm-x86/mach-default/smpboot_hooks.h
index 7f45f6311059..8e1c6c06d057 100644
--- a/include/asm-x86/mach-default/smpboot_hooks.h
+++ b/include/asm-x86/mach-default/smpboot_hooks.h
@@ -41,4 +41,6 @@ static inline void smpboot_setup_io_apic(void)
 	 */
 	if (!skip_ioapic_setup && nr_ioapics)
 		setup_IO_APIC();
+	else
+		nr_ioapics = 0;
 }
-- 
cgit v1.2.3


From 9f3734f631267d2f36008833b62670ca342ac000 Mon Sep 17 00:00:00 2001
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Date: Wed, 19 Mar 2008 14:26:10 -0300
Subject: x86: introduce smpboot_clear_io_apic

x86_64 has two nr_ioapics = 0 statements. In 32-bit, it can be done
too. We do it through the smpboot_clear_io_apic() inline function,
to cope with subarchitectures (visws) that does not compile mpparse in

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/smpboot_32.c                 | 2 ++
 arch/x86/kernel/smpboot_64.c                 | 4 ++--
 include/asm-x86/mach-default/smpboot_hooks.h | 5 +++++
 include/asm-x86/mach-visws/smpboot_hooks.h   | 4 ++++
 4 files changed, 13 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/smpboot_32.c b/arch/x86/kernel/smpboot_32.c
index ae23b603978c..5a0f57f35191 100644
--- a/arch/x86/kernel/smpboot_32.c
+++ b/arch/x86/kernel/smpboot_32.c
@@ -155,6 +155,7 @@ static int __init smp_sanity_check(unsigned max_cpus)
 		printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n",
 			boot_cpu_physical_apicid);
 		printk(KERN_ERR "... forcing use of dummy APIC emulation. (tell your hw vendor)\n");
+		smpboot_clear_io_apic();
 		return -1;
 	}
 
@@ -173,6 +174,7 @@ static int __init smp_sanity_check(unsigned max_cpus)
 			setup_local_APIC();
 			end_local_APIC_setup();
 		}
+		smpboot_clear_io_apic();
 		return -1;
 	}
 	return 0;
diff --git a/arch/x86/kernel/smpboot_64.c b/arch/x86/kernel/smpboot_64.c
index c66fb15b0131..775244545ffa 100644
--- a/arch/x86/kernel/smpboot_64.c
+++ b/arch/x86/kernel/smpboot_64.c
@@ -134,7 +134,7 @@ static int __init smp_sanity_check(unsigned max_cpus)
 		printk(KERN_ERR "BIOS bug, local APIC #%d not detected!...\n",
 			boot_cpu_physical_apicid);
 		printk(KERN_ERR "... forcing use of dummy APIC emulation. (tell your hw vendor)\n");
-		nr_ioapics = 0;
+		smpboot_clear_io_apic();
 		return -1;
 	}
 
@@ -145,7 +145,7 @@ static int __init smp_sanity_check(unsigned max_cpus)
 	 */
 	if (!max_cpus) {
 		printk(KERN_INFO "SMP mode deactivated, forcing use of dummy APIC emulation.\n");
-		nr_ioapics = 0;
+		smpboot_clear_io_apic();
 		return -1;
 	}
 
diff --git a/include/asm-x86/mach-default/smpboot_hooks.h b/include/asm-x86/mach-default/smpboot_hooks.h
index 8e1c6c06d057..3ff2c5bff93a 100644
--- a/include/asm-x86/mach-default/smpboot_hooks.h
+++ b/include/asm-x86/mach-default/smpboot_hooks.h
@@ -44,3 +44,8 @@ static inline void smpboot_setup_io_apic(void)
 	else
 		nr_ioapics = 0;
 }
+
+static inline void smpboot_clear_io_apic(void)
+{
+	nr_ioapics = 0;
+}
diff --git a/include/asm-x86/mach-visws/smpboot_hooks.h b/include/asm-x86/mach-visws/smpboot_hooks.h
index d926471fa359..c9b83e395a2e 100644
--- a/include/asm-x86/mach-visws/smpboot_hooks.h
+++ b/include/asm-x86/mach-visws/smpboot_hooks.h
@@ -22,3 +22,7 @@ static inline void smpboot_restore_warm_reset_vector(void)
 static inline void smpboot_setup_io_apic(void)
 {
 }
+
+static inline void smpboot_clear_io_apic(void)
+{
+}
-- 
cgit v1.2.3


From ce3fe6b2bfded4f5d931c5f2f9325dc2e3fd3a74 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Mon, 17 Mar 2008 22:08:17 +0300
Subject: x86: use get_bios_ebda in mpparse_64.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c             |  2 +-
 arch/x86/kernel/mpparse_64.c             | 11 ++++-------
 arch/x86/kernel/setup_32.c               |  2 +-
 arch/x86/mm/discontig_32.c               |  2 +-
 include/asm-x86/bios_ebda.h              | 15 +++++++++++++++
 include/asm-x86/mach-default/bios_ebda.h | 15 ---------------
 6 files changed, 22 insertions(+), 25 deletions(-)
 create mode 100644 include/asm-x86/bios_ebda.h
 delete mode 100644 include/asm-x86/mach-default/bios_ebda.h

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 838e4974e1ce..a2162644cb4e 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -27,11 +27,11 @@
 #include <asm/mtrr.h>
 #include <asm/mpspec.h>
 #include <asm/io_apic.h>
+#include <asm/bios_ebda.h>
 
 #include <mach_apic.h>
 #include <mach_apicdef.h>
 #include <mach_mpparse.h>
-#include <bios_ebda.h>
 
 /* Have we found an MP table */
 int smp_found_config;
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 269fd46df42c..fb74135f9d0e 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -29,6 +29,7 @@
 #include <asm/io_apic.h>
 #include <asm/proto.h>
 #include <asm/acpi.h>
+#include <asm/bios_ebda.h>
 
 #include <mach_apic.h>
 
@@ -641,13 +642,9 @@ static void __init __find_smp_config(unsigned reserve)
 	 * should be fixed.
 	 */
 
-	address = *(unsigned short *)phys_to_virt(0x40E);
-	address <<= 4;
-	if (smp_scan_config(address, 0x1000, reserve))
-		return;
-
-	/* If we have come this far, we did not find an MP table  */
-	printk(KERN_INFO "No mptable found.\n");
+	address = get_bios_ebda();
+	if (address)
+		smp_scan_config(address, 0x1000, reserve);
 }
 
 void __init early_find_smp_config(void)
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index eb97bcfe0f6f..58f3c1fbc5c3 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -62,7 +62,7 @@
 #include <asm/io.h>
 #include <asm/vmi.h>
 #include <setup_arch.h>
-#include <bios_ebda.h>
+#include <asm/bios_ebda.h>
 #include <asm/cacheflush.h>
 #include <asm/processor.h>
 
diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c
index 8e25e06ff730..eba0bbede7a6 100644
--- a/arch/x86/mm/discontig_32.c
+++ b/arch/x86/mm/discontig_32.c
@@ -37,7 +37,7 @@
 #include <asm/e820.h>
 #include <asm/setup.h>
 #include <asm/mmzone.h>
-#include <bios_ebda.h>
+#include <asm/bios_ebda.h>
 
 struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL(node_data);
diff --git a/include/asm-x86/bios_ebda.h b/include/asm-x86/bios_ebda.h
new file mode 100644
index 000000000000..9cbd9a668af8
--- /dev/null
+++ b/include/asm-x86/bios_ebda.h
@@ -0,0 +1,15 @@
+#ifndef _MACH_BIOS_EBDA_H
+#define _MACH_BIOS_EBDA_H
+
+/*
+ * there is a real-mode segmented pointer pointing to the
+ * 4K EBDA area at 0x40E.
+ */
+static inline unsigned int get_bios_ebda(void)
+{
+	unsigned int address = *(unsigned short *)phys_to_virt(0x40E);
+	address <<= 4;
+	return address;	/* 0 means none */
+}
+
+#endif /* _MACH_BIOS_EBDA_H */
diff --git a/include/asm-x86/mach-default/bios_ebda.h b/include/asm-x86/mach-default/bios_ebda.h
deleted file mode 100644
index 9cbd9a668af8..000000000000
--- a/include/asm-x86/mach-default/bios_ebda.h
+++ /dev/null
@@ -1,15 +0,0 @@
-#ifndef _MACH_BIOS_EBDA_H
-#define _MACH_BIOS_EBDA_H
-
-/*
- * there is a real-mode segmented pointer pointing to the
- * 4K EBDA area at 0x40E.
- */
-static inline unsigned int get_bios_ebda(void)
-{
-	unsigned int address = *(unsigned short *)phys_to_virt(0x40E);
-	address <<= 4;
-	return address;	/* 0 means none */
-}
-
-#endif /* _MACH_BIOS_EBDA_H */
-- 
cgit v1.2.3


From 4655c7deca112bea86ca00f616f19c3717f687aa Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Mon, 17 Mar 2008 22:08:36 +0300
Subject: x86: remove mpc_apic_id()

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c             |  9 +++++++++
 include/asm-x86/genapic_32.h             |  3 ---
 include/asm-x86/mach-bigsmp/mach_apic.h  | 11 -----------
 include/asm-x86/mach-default/mach_apic.h | 11 -----------
 include/asm-x86/mach-es7000/mach_apic.h  | 10 ----------
 include/asm-x86/mach-generic/mach_apic.h |  1 -
 include/asm-x86/mach-summit/mach_apic.h  | 11 -----------
 7 files changed, 9 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index a2162644cb4e..072fcc462399 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -118,7 +118,16 @@ static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 		return;
 	}
 
+#ifdef CONFIG_X86_NUMAQ
 	apicid = mpc_apic_id(m, translation_table[mpc_record]);
+#else
+	Dprintk("Processor #%d %u:%u APIC version %d\n",
+		m->mpc_apicid,
+		(m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
+		(m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
+		m->mpc_apicver);
+	apicid = m->mpc_apicid;
+#endif
 
 	if (m->mpc_featureflag&(1<<0))
 		Dprintk("    Floating point unit present.\n");
diff --git a/include/asm-x86/genapic_32.h b/include/asm-x86/genapic_32.h
index 33e3ffe1766c..5d024400ddde 100644
--- a/include/asm-x86/genapic_32.h
+++ b/include/asm-x86/genapic_32.h
@@ -42,8 +42,6 @@ struct genapic {
 	int (*cpu_to_logical_apicid)(int cpu);
 	int (*cpu_present_to_apicid)(int mps_cpu);
 	physid_mask_t (*apicid_to_cpu_present)(int phys_apicid);
-	int (*mpc_apic_id)(struct mpc_config_processor *m, 
-			   struct mpc_config_translation *t); 
 	void (*setup_portio_remap)(void); 
 	int (*check_phys_apicid_present)(int boot_cpu_physical_apicid);
 	void (*enable_apic_mode)(void);
@@ -105,7 +103,6 @@ struct genapic {
 	APICFUNC(cpu_to_logical_apicid) \
 	APICFUNC(cpu_present_to_apicid) \
 	APICFUNC(apicid_to_cpu_present) \
-	APICFUNC(mpc_apic_id) \
 	APICFUNC(setup_portio_remap) \
 	APICFUNC(check_phys_apicid_present) \
 	APICFUNC(mpc_oem_bus_info) \
diff --git a/include/asm-x86/mach-bigsmp/mach_apic.h b/include/asm-x86/mach-bigsmp/mach_apic.h
index 0d55b1f6d56b..8327907c79bf 100644
--- a/include/asm-x86/mach-bigsmp/mach_apic.h
+++ b/include/asm-x86/mach-bigsmp/mach_apic.h
@@ -106,17 +106,6 @@ static inline int cpu_to_logical_apicid(int cpu)
 	return cpu_physical_id(cpu);
 }
 
-static inline int mpc_apic_id(struct mpc_config_processor *m,
-			      struct mpc_config_translation *translation_record)
-{
-	printk("Processor #%d %u:%u APIC version %d\n",
-	       m->mpc_apicid,
-	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-	       m->mpc_apicver);
-	return m->mpc_apicid;
-}
-
 static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map)
 {
 	/* For clustered we don't have a good way to do this yet - hack */
diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h
index e081bdccde2b..13900e8cc1ab 100644
--- a/include/asm-x86/mach-default/mach_apic.h
+++ b/include/asm-x86/mach-default/mach_apic.h
@@ -100,17 +100,6 @@ static inline physid_mask_t apicid_to_cpu_present(int phys_apicid)
 	return physid_mask_of_physid(phys_apicid);
 }
 
-static inline int mpc_apic_id(struct mpc_config_processor *m,
-			      struct mpc_config_translation *translation_record)
-{
-	printk("Processor #%d %u:%u APIC version %d\n",
-	       m->mpc_apicid,
-	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-	       m->mpc_apicver);
-	return m->mpc_apicid;
-}
-
 static inline void setup_portio_remap(void)
 {
 }
diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/mach-es7000/mach_apic.h
index 04cba9f1e375..0137b6e142cc 100644
--- a/include/asm-x86/mach-es7000/mach_apic.h
+++ b/include/asm-x86/mach-es7000/mach_apic.h
@@ -127,16 +127,6 @@ static inline int cpu_to_logical_apicid(int cpu)
 #endif
 }
 
-static inline int mpc_apic_id(struct mpc_config_processor *m, struct mpc_config_translation *unused)
-{
-	printk("Processor #%d %u:%u APIC version %d\n",
-	       m->mpc_apicid,
-	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-	       m->mpc_apicver);
-	return (m->mpc_apicid);
-}
-
 static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map)
 {
 	/* For clustered we don't have a good way to do this yet - hack */
diff --git a/include/asm-x86/mach-generic/mach_apic.h b/include/asm-x86/mach-generic/mach_apic.h
index a236e7021528..6eff343e1233 100644
--- a/include/asm-x86/mach-generic/mach_apic.h
+++ b/include/asm-x86/mach-generic/mach_apic.h
@@ -19,7 +19,6 @@
 #define cpu_to_logical_apicid (genapic->cpu_to_logical_apicid) 
 #define cpu_present_to_apicid (genapic->cpu_present_to_apicid)
 #define apicid_to_cpu_present (genapic->apicid_to_cpu_present)
-#define mpc_apic_id (genapic->mpc_apic_id) 
 #define setup_portio_remap (genapic->setup_portio_remap)
 #define check_apicid_present (genapic->check_apicid_present)
 #define check_phys_apicid_present (genapic->check_phys_apicid_present)
diff --git a/include/asm-x86/mach-summit/mach_apic.h b/include/asm-x86/mach-summit/mach_apic.h
index 91d7641cddc9..1f76c2e70232 100644
--- a/include/asm-x86/mach-summit/mach_apic.h
+++ b/include/asm-x86/mach-summit/mach_apic.h
@@ -125,17 +125,6 @@ static inline physid_mask_t apicid_to_cpu_present(int apicid)
 	return physid_mask_of_physid(0);
 }
 
-static inline int mpc_apic_id(struct mpc_config_processor *m,
-			      struct mpc_config_translation *translation_record)
-{
-	printk("Processor #%d %u:%u APIC version %d\n",
-	       m->mpc_apicid,
-	       (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
-	       (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
-	       m->mpc_apicver);
-	return m->mpc_apicid;
-}
-
 static inline void setup_portio_remap(void)
 {
 }
-- 
cgit v1.2.3


From d285e338899a4ff662a17b22d3bb0e48bb1465d4 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Mon, 17 Mar 2008 22:08:42 +0300
Subject: x86: remove mpc_oem_pci_bus()

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c                |  2 ++
 arch/x86/pci/numa.c                         |  8 ++++++++
 include/asm-x86/genapic_32.h                |  3 ---
 include/asm-x86/mach-default/mach_mpparse.h |  5 -----
 include/asm-x86/mach-es7000/mach_mpparse.h  |  5 -----
 include/asm-x86/mach-generic/mach_mpparse.h |  1 -
 include/asm-x86/mach-numaq/mach_mpparse.h   | 12 ++----------
 include/asm-x86/mach-summit/mach_mpparse.h  |  5 -----
 8 files changed, 12 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 072fcc462399..621eac569550 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -278,7 +278,9 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 
 	set_bit(m->mpc_busid, mp_bus_not_pci);
 	if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI)-1) == 0) {
+#ifdef CONFIG_X86_NUMAQ
 		mpc_oem_pci_bus(m, translation_table[mpc_record]);
+#endif
 		clear_bit(m->mpc_busid, mp_bus_not_pci);
 		mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
 		mp_current_pci_id++;
diff --git a/arch/x86/pci/numa.c b/arch/x86/pci/numa.c
index 79d0a98b9d03..a98ae0e57272 100644
--- a/arch/x86/pci/numa.c
+++ b/arch/x86/pci/numa.c
@@ -19,6 +19,14 @@ int mp_bus_id_to_local[MAX_MP_BUSSES];
 
 int quad_local_to_mp_bus_id [NR_CPUS/4][4];
 #define QUADLOCAL2BUS(quad,local) (quad_local_to_mp_bus_id[quad][local])
+void mpc_oem_pci_bus(struct mpc_config_bus *m,
+	struct mpc_config_translation *translation)
+{
+	int quad = translation->trans_quad;
+	int local = translation->trans_local;
+
+	quad_local_to_mp_bus_id[quad][local] = m->mpc_busid;
+}
 
 /* Where the IO area was mapped on multiquad, always 0 otherwise */
 void *xquad_portio;
diff --git a/include/asm-x86/genapic_32.h b/include/asm-x86/genapic_32.h
index 5d024400ddde..5a1ae0164aa0 100644
--- a/include/asm-x86/genapic_32.h
+++ b/include/asm-x86/genapic_32.h
@@ -50,8 +50,6 @@ struct genapic {
 	/* mpparse */
 	void (*mpc_oem_bus_info)(struct mpc_config_bus *, char *, 
 				 struct mpc_config_translation *);
-	void (*mpc_oem_pci_bus)(struct mpc_config_bus *, 
-				struct mpc_config_translation *); 
 
 	/* When one of the next two hooks returns 1 the genapic
 	   is switched to this. Essentially they are additional probe 
@@ -106,7 +104,6 @@ struct genapic {
 	APICFUNC(setup_portio_remap) \
 	APICFUNC(check_phys_apicid_present) \
 	APICFUNC(mpc_oem_bus_info) \
-	APICFUNC(mpc_oem_pci_bus) \
 	APICFUNC(mps_oem_check) \
 	APICFUNC(get_apic_id) \
 	.apic_id_mask = APIC_ID_MASK, \
diff --git a/include/asm-x86/mach-default/mach_mpparse.h b/include/asm-x86/mach-default/mach_mpparse.h
index 1d3832482580..679393c26dc6 100644
--- a/include/asm-x86/mach-default/mach_mpparse.h
+++ b/include/asm-x86/mach-default/mach_mpparse.h
@@ -7,11 +7,6 @@ static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
 //	Dprintk("Bus #%d is %s\n", m->mpc_busid, name);
 }
 
-static inline void mpc_oem_pci_bus(struct mpc_config_bus *m, 
-				struct mpc_config_translation *translation)
-{
-}
-
 static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, 
 		char *productid)
 {
diff --git a/include/asm-x86/mach-es7000/mach_mpparse.h b/include/asm-x86/mach-es7000/mach_mpparse.h
index 52ee75cd0fe1..3afa1b010818 100644
--- a/include/asm-x86/mach-es7000/mach_mpparse.h
+++ b/include/asm-x86/mach-es7000/mach_mpparse.h
@@ -9,11 +9,6 @@ static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
 	Dprintk("Bus #%d is %s\n", m->mpc_busid, name);
 }
 
-static inline void mpc_oem_pci_bus(struct mpc_config_bus *m,
-				struct mpc_config_translation *translation)
-{
-}
-
 extern int parse_unisys_oem (char *oemptr);
 extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
 extern void setup_unisys(void);
diff --git a/include/asm-x86/mach-generic/mach_mpparse.h b/include/asm-x86/mach-generic/mach_mpparse.h
index dbd9fce54f4d..2a6937150334 100644
--- a/include/asm-x86/mach-generic/mach_mpparse.h
+++ b/include/asm-x86/mach-generic/mach_mpparse.h
@@ -4,7 +4,6 @@
 #include <asm/genapic.h>
 
 #define mpc_oem_bus_info (genapic->mpc_oem_bus_info)
-#define mpc_oem_pci_bus (genapic->mpc_oem_pci_bus)
 
 int mps_oem_check(struct mp_config_table *mpc, char *oem, char *productid); 
 int acpi_madt_oem_check(char *oem_id, char *oem_table_id); 
diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h
index 254993ae04c7..4a5d8a8b01a2 100644
--- a/include/asm-x86/mach-numaq/mach_mpparse.h
+++ b/include/asm-x86/mach-numaq/mach_mpparse.h
@@ -15,16 +15,8 @@ static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
 	printk("Bus #%d is %s (node %d)\n", m->mpc_busid, name, quad);
 }
 
-extern int quad_local_to_mp_bus_id[NR_CPUS/4][4];
-
-static inline void mpc_oem_pci_bus(struct mpc_config_bus *m, 
-				struct mpc_config_translation *translation)
-{
-	int quad = translation->trans_quad;
-	int local = translation->trans_local;
-
-	quad_local_to_mp_bus_id[quad][local] = m->mpc_busid;
-}
+extern void mpc_oem_pci_bus(struct mpc_config_bus *m,
+	struct mpc_config_translation *translation);
 
 /* Hook from generic ACPI tables.c */
 static inline void acpi_madt_oem_check(char *oem_id, char *oem_table_id)
diff --git a/include/asm-x86/mach-summit/mach_mpparse.h b/include/asm-x86/mach-summit/mach_mpparse.h
index c2520539d934..e1af489a8809 100644
--- a/include/asm-x86/mach-summit/mach_mpparse.h
+++ b/include/asm-x86/mach-summit/mach_mpparse.h
@@ -18,11 +18,6 @@ static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
 	Dprintk("Bus #%d is %s\n", m->mpc_busid, name);
 }
 
-static inline void mpc_oem_pci_bus(struct mpc_config_bus *m, 
-				struct mpc_config_translation *translation)
-{
-}
-
 static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, 
 		char *productid)
 {
-- 
cgit v1.2.3


From 0ec153af4dec8944e6da558093914a3bce4c76f9 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Mon, 17 Mar 2008 22:08:48 +0300
Subject: x86: remove mpc_oem_bus_info()

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c                |  4 ++++
 arch/x86/pci/numa.c                         | 12 ++++++++++++
 include/asm-x86/genapic_32.h                |  4 ----
 include/asm-x86/mach-default/mach_mpparse.h |  6 ------
 include/asm-x86/mach-es7000/mach_mpparse.h  |  6 ------
 include/asm-x86/mach-generic/mach_mpparse.h |  4 ----
 include/asm-x86/mach-numaq/mach_mpparse.h   | 16 ++--------------
 include/asm-x86/mach-summit/mach_mpparse.h  |  6 ------
 8 files changed, 18 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 621eac569550..febd69dbbee9 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -265,7 +265,11 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 	memcpy(str, m->mpc_bustype, 6);
 	str[6] = 0;
 
+#ifdef CONFIG_X86_NUMAQ
 	mpc_oem_bus_info(m, str, translation_table[mpc_record]);
+#else
+	Dprintk("Bus #%d is %s\n", m->mpc_busid, str);
+#endif
 
 #if MAX_MP_BUSSES < 256
 	if (m->mpc_busid >= MAX_MP_BUSSES) {
diff --git a/arch/x86/pci/numa.c b/arch/x86/pci/numa.c
index a98ae0e57272..d9afbae5092b 100644
--- a/arch/x86/pci/numa.c
+++ b/arch/x86/pci/numa.c
@@ -17,6 +17,18 @@ int mp_bus_id_to_node[MAX_MP_BUSSES];
 int mp_bus_id_to_local[MAX_MP_BUSSES];
 #define BUS2LOCAL(global) (mp_bus_id_to_local[global])
 
+void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
+	struct mpc_config_translation *translation)
+{
+	int quad = translation->trans_quad;
+	int local = translation->trans_local;
+
+	mp_bus_id_to_node[m->mpc_busid] = quad;
+	mp_bus_id_to_local[m->mpc_busid] = local;
+	printk(KERN_INFO "Bus #%d is %s (node %d)\n",
+	       m->mpc_busid, name, quad);
+}
+
 int quad_local_to_mp_bus_id [NR_CPUS/4][4];
 #define QUADLOCAL2BUS(quad,local) (quad_local_to_mp_bus_id[quad][local])
 void mpc_oem_pci_bus(struct mpc_config_bus *m,
diff --git a/include/asm-x86/genapic_32.h b/include/asm-x86/genapic_32.h
index 5a1ae0164aa0..5a1b68ac3ca9 100644
--- a/include/asm-x86/genapic_32.h
+++ b/include/asm-x86/genapic_32.h
@@ -48,9 +48,6 @@ struct genapic {
 	u32 (*phys_pkg_id)(u32 cpuid_apic, int index_msb);
 
 	/* mpparse */
-	void (*mpc_oem_bus_info)(struct mpc_config_bus *, char *, 
-				 struct mpc_config_translation *);
-
 	/* When one of the next two hooks returns 1 the genapic
 	   is switched to this. Essentially they are additional probe 
 	   functions. */
@@ -103,7 +100,6 @@ struct genapic {
 	APICFUNC(apicid_to_cpu_present) \
 	APICFUNC(setup_portio_remap) \
 	APICFUNC(check_phys_apicid_present) \
-	APICFUNC(mpc_oem_bus_info) \
 	APICFUNC(mps_oem_check) \
 	APICFUNC(get_apic_id) \
 	.apic_id_mask = APIC_ID_MASK, \
diff --git a/include/asm-x86/mach-default/mach_mpparse.h b/include/asm-x86/mach-default/mach_mpparse.h
index 679393c26dc6..d14108505bb8 100644
--- a/include/asm-x86/mach-default/mach_mpparse.h
+++ b/include/asm-x86/mach-default/mach_mpparse.h
@@ -1,12 +1,6 @@
 #ifndef __ASM_MACH_MPPARSE_H
 #define __ASM_MACH_MPPARSE_H
 
-static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name, 
-				struct mpc_config_translation *translation)
-{
-//	Dprintk("Bus #%d is %s\n", m->mpc_busid, name);
-}
-
 static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, 
 		char *productid)
 {
diff --git a/include/asm-x86/mach-es7000/mach_mpparse.h b/include/asm-x86/mach-es7000/mach_mpparse.h
index 3afa1b010818..ef26d3523625 100644
--- a/include/asm-x86/mach-es7000/mach_mpparse.h
+++ b/include/asm-x86/mach-es7000/mach_mpparse.h
@@ -3,12 +3,6 @@
 
 #include <linux/acpi.h>
 
-static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
-				struct mpc_config_translation *translation)
-{
-	Dprintk("Bus #%d is %s\n", m->mpc_busid, name);
-}
-
 extern int parse_unisys_oem (char *oemptr);
 extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
 extern void setup_unisys(void);
diff --git a/include/asm-x86/mach-generic/mach_mpparse.h b/include/asm-x86/mach-generic/mach_mpparse.h
index 2a6937150334..0d0b5ba2e9d1 100644
--- a/include/asm-x86/mach-generic/mach_mpparse.h
+++ b/include/asm-x86/mach-generic/mach_mpparse.h
@@ -1,10 +1,6 @@
 #ifndef _MACH_MPPARSE_H
 #define _MACH_MPPARSE_H 1
 
-#include <asm/genapic.h>
-
-#define mpc_oem_bus_info (genapic->mpc_oem_bus_info)
-
 int mps_oem_check(struct mp_config_table *mpc, char *oem, char *productid); 
 int acpi_madt_oem_check(char *oem_id, char *oem_table_id); 
 
diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h
index 4a5d8a8b01a2..459b12401187 100644
--- a/include/asm-x86/mach-numaq/mach_mpparse.h
+++ b/include/asm-x86/mach-numaq/mach_mpparse.h
@@ -1,20 +1,8 @@
 #ifndef __ASM_MACH_MPPARSE_H
 #define __ASM_MACH_MPPARSE_H
 
-extern int mp_bus_id_to_local[MAX_MP_BUSSES];
-extern int mp_bus_id_to_node[MAX_MP_BUSSES];
-
-static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name, 
-				struct mpc_config_translation *translation)
-{
-	int quad = translation->trans_quad;
-	int local = translation->trans_local;
-
-	mp_bus_id_to_node[m->mpc_busid] = quad;
-	mp_bus_id_to_local[m->mpc_busid] = local;
-	printk("Bus #%d is %s (node %d)\n", m->mpc_busid, name, quad);
-}
-
+extern void mpc_oem_bus_info(struct mpc_config_bus *m, char *name,
+			     struct mpc_config_translation *translation);
 extern void mpc_oem_pci_bus(struct mpc_config_bus *m,
 	struct mpc_config_translation *translation);
 
diff --git a/include/asm-x86/mach-summit/mach_mpparse.h b/include/asm-x86/mach-summit/mach_mpparse.h
index e1af489a8809..fdf591701339 100644
--- a/include/asm-x86/mach-summit/mach_mpparse.h
+++ b/include/asm-x86/mach-summit/mach_mpparse.h
@@ -12,12 +12,6 @@ extern void setup_summit(void);
 #define setup_summit()	{}
 #endif
 
-static inline void mpc_oem_bus_info(struct mpc_config_bus *m, char *name, 
-				struct mpc_config_translation *translation)
-{
-	Dprintk("Bus #%d is %s\n", m->mpc_busid, name);
-}
-
 static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, 
 		char *productid)
 {
-- 
cgit v1.2.3


From 864205062f1c752c80077be8ec2b15c81f4a6525 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Mon, 17 Mar 2008 22:08:55 +0300
Subject: x86: make struct mpc_config_translation NUMAQ-only

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c           |  6 ++++++
 include/asm-x86/genapic_32.h           |  1 -
 include/asm-x86/mach-numaq/mach_apic.h | 10 ++++++++++
 include/asm-x86/mpspec_def.h           | 11 -----------
 4 files changed, 16 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index febd69dbbee9..15dd87110298 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -98,6 +98,7 @@ static int __init mpf_checksum(unsigned char *mp, int len)
 	return sum & 0xFF;
 }
 
+#ifdef CONFIG_X86_NUMAQ
 /*
  * Have to match translation table entries to main table entries by counter
  * hence the mpc_record variable .... can't see a less disgusting way of
@@ -106,6 +107,7 @@ static int __init mpf_checksum(unsigned char *mp, int len)
 
 static int mpc_record; 
 static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __cpuinitdata;
+#endif
 
 static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 {
@@ -475,7 +477,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 	/*
 	 *	Now process the configuration blocks.
 	 */
+#ifdef CONFIG_X86_NUMAQ
 	mpc_record = 0;
+#endif
 	while (count < mpc->mpc_length) {
 		switch(*mpt) {
 			case MP_PROCESSOR:
@@ -532,7 +536,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 				break;
 			}
 		}
+#ifdef CONFIG_X86_NUMAQ
 		++mpc_record;
+#endif
 	}
 	setup_apic_routing();
 	if (!num_processors)
diff --git a/include/asm-x86/genapic_32.h b/include/asm-x86/genapic_32.h
index 5a1b68ac3ca9..b501ae7809ba 100644
--- a/include/asm-x86/genapic_32.h
+++ b/include/asm-x86/genapic_32.h
@@ -14,7 +14,6 @@
  * Copyright 2003 Andi Kleen, SuSE Labs.
  */
 
-struct mpc_config_translation;
 struct mpc_config_bus;
 struct mp_config_table;
 struct mpc_config_processor;
diff --git a/include/asm-x86/mach-numaq/mach_apic.h b/include/asm-x86/mach-numaq/mach_apic.h
index 3b637fac890b..75a56e5afbe7 100644
--- a/include/asm-x86/mach-numaq/mach_apic.h
+++ b/include/asm-x86/mach-numaq/mach_apic.h
@@ -95,6 +95,16 @@ static inline physid_mask_t apicid_to_cpu_present(int logical_apicid)
 	return physid_mask_of_physid(cpu + 4*node);
 }
 
+struct mpc_config_translation {
+	unsigned char mpc_type;
+	unsigned char trans_len;
+	unsigned char trans_type;
+	unsigned char trans_quad;
+	unsigned char trans_global;
+	unsigned char trans_local;
+	unsigned short trans_reserved;
+};
+
 static inline int mpc_apic_id(struct mpc_config_processor *m, 
 			struct mpc_config_translation *translation_record)
 {
diff --git a/include/asm-x86/mpspec_def.h b/include/asm-x86/mpspec_def.h
index 3504617fe648..1f35691b4f7c 100644
--- a/include/asm-x86/mpspec_def.h
+++ b/include/asm-x86/mpspec_def.h
@@ -166,17 +166,6 @@ struct mp_config_oemtable
 	char  mpc_oem[8];
 };
 
-struct mpc_config_translation
-{
-	unsigned char mpc_type;
-	unsigned char trans_len;
-	unsigned char trans_type;
-	unsigned char trans_quad;
-	unsigned char trans_global;
-	unsigned char trans_local;
-	unsigned short trans_reserved;
-};
-
 /*
  *	Default configurations
  *
-- 
cgit v1.2.3


From aa7d8e25eca5deb33eb08013bc78a80514349b40 Mon Sep 17 00:00:00 2001
From: Ravikiran G Thirumalai <kiran@scalex86.org>
Date: Thu, 20 Mar 2008 00:41:16 -0700
Subject: x86: fix build breakage when PCI is define and PARAVIRT is not

- Fix the the build breakage when PARAVIRT is defined
  but PCI is not
  This fixes problem reported at:
	http://marc.info/?l=linux-kernel&m=120525966600698&w=2
- Make is_vsmp_box() available even when PARAVIRT is not defined.
  This is needed to determine if tsc's are reliable as a time source
  even when PARAVIRT is not defined.
- split vsmp_init to use is_vsmp_box() and set_vsmp_pv_ops()
  set_vsmp_pv_ops will do nothing if PCI is not enabled in the config.

Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/Makefile   |  2 +-
 arch/x86/kernel/setup_64.c |  2 --
 arch/x86/kernel/vsmp_64.c  | 72 +++++++++++++++++++++++++++-------------------
 include/asm-x86/apic.h     |  7 ++---
 4 files changed, 46 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index d5a05a0cef62..1fe841a86f7e 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -60,7 +60,7 @@ obj-$(CONFIG_KEXEC)		+= relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_$(BITS).o
 obj-$(CONFIG_X86_NUMAQ)		+= numaq_32.o
 obj-$(CONFIG_X86_SUMMIT_NUMA)	+= summit_32.o
-obj-$(CONFIG_PARAVIRT)		+= vsmp_64.o
+obj-y				+= vsmp_64.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_MODULES)		+= module_$(BITS).o
 obj-$(CONFIG_ACPI_SRAT) 	+= srat_32.o
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 990724143c43..20034493b5ad 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -351,9 +351,7 @@ void __init setup_arch(char **cmdline_p)
 	if (efi_enabled)
 		efi_init();
 
-#ifdef	CONFIG_PARAVIRT
 	vsmp_init();
-#endif
 
 	dmi_scan_machine();
 
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index eb25584c54c3..4a790a5f61b7 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -19,6 +19,7 @@
 #include <asm/io.h>
 #include <asm/paravirt.h>
 
+#if defined CONFIG_PCI && defined CONFIG_PARAVIRT
 /*
  * Interrupt control on vSMPowered systems:
  * ~AC is a shadow of IF.  If IF is 'on' AC should be 'off'
@@ -72,39 +73,11 @@ static unsigned __init vsmp_patch(u8 type, u16 clobbers, void *ibuf,
 
 }
 
-static int vsmp = -1;
-
-int is_vsmp_box(void)
-{
-	if (vsmp != -1)
-		return vsmp;
-
-	vsmp = 0;
-	if (!early_pci_allowed())
-		return vsmp;
-
-	/* Check if we are running on a ScaleMP vSMP box */
-	if ((read_pci_config_16(0, 0x1f, 0, PCI_VENDOR_ID) ==
-	     PCI_VENDOR_ID_SCALEMP) &&
-	    (read_pci_config_16(0, 0x1f, 0, PCI_DEVICE_ID) ==
-	    PCI_DEVICE_ID_SCALEMP_VSMP_CTL))
-		vsmp = 1;
-
-	return vsmp;
-}
-
-void __init vsmp_init(void)
+static void __init set_vsmp_pv_ops(void)
 {
 	void *address;
 	unsigned int cap, ctl, cfg;
 
-	if (!is_vsmp_box())
-		return;
-
-	if (!early_pci_allowed())
-		return;
-
-	/* If we are, use the distinguished irq functions */
 	pv_irq_ops.irq_disable = vsmp_irq_disable;
 	pv_irq_ops.irq_enable  = vsmp_irq_enable;
 	pv_irq_ops.save_fl  = vsmp_save_fl;
@@ -127,5 +100,46 @@ void __init vsmp_init(void)
 	}
 
 	early_iounmap(address, 8);
+}
+#else
+static void __init set_vsmp_pv_ops(void)
+{
+}
+#endif
+
+#ifdef CONFIG_PCI
+static int vsmp = -1;
+
+int is_vsmp_box(void)
+{
+	if (vsmp != -1)
+		return vsmp;
+
+	vsmp = 0;
+	if (!early_pci_allowed())
+		return vsmp;
+
+	/* Check if we are running on a ScaleMP vSMP box */
+	if ((read_pci_config_16(0, 0x1f, 0, PCI_VENDOR_ID) ==
+	     PCI_VENDOR_ID_SCALEMP) &&
+	    (read_pci_config_16(0, 0x1f, 0, PCI_DEVICE_ID) ==
+	    PCI_DEVICE_ID_SCALEMP_VSMP_CTL))
+		vsmp = 1;
+
+	return vsmp;
+}
+#else
+int is_vsmp_box(void)
+{
+	return 0;
+}
+#endif
+
+void __init vsmp_init(void)
+{
+	if (!is_vsmp_box())
+		return;
+
+	set_vsmp_pv_ops();
 	return;
 }
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index b804238489a1..b11d6524fb13 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -50,19 +50,16 @@ extern int disable_apic_timer;
  */
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
-extern int is_vsmp_box(void);
 #else
 #define apic_write native_apic_write
 #define apic_write_atomic native_apic_write_atomic
 #define apic_read native_apic_read
 #define setup_boot_clock setup_boot_APIC_clock
 #define setup_secondary_clock setup_secondary_APIC_clock
-static int inline is_vsmp_box(void)
-{
-	return 0;
-}
 #endif
 
+extern int is_vsmp_box(void);
+
 static inline void native_apic_write(unsigned long reg, u32 v)
 {
 	*((volatile u32 *)(APIC_BASE + reg)) = v;
-- 
cgit v1.2.3


From ede1389f8ab4f3a1343e567133fa9720a054a3aa Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Mon, 17 Mar 2008 22:29:32 +0200
Subject: x86: remove the write-only timer_uses_ioapic_pin_0

This patch removes the write-only timer_uses_ioapic_pin_0
(gsi can't be <= 15 in the line of it's fake usage in mpparse_32.c).

Spotted by the GNU C compiler.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/io_apic_32.c | 5 -----
 arch/x86/kernel/mpparse_32.c | 3 +--
 include/asm-x86/io_apic.h    | 1 -
 3 files changed, 1 insertion(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 0d70acd3b134..742fab45e1c6 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -2114,8 +2114,6 @@ static inline void unlock_ExtINT_logic(void)
 	ioapic_write_entry(apic, pin, entry0);
 }
 
-int timer_uses_ioapic_pin_0;
-
 /*
  * This code may look a bit paranoid, but it's supposed to cooperate with
  * a wide range of boards and BIOS bugs.  Fortunately only the timer IRQ
@@ -2155,9 +2153,6 @@ static inline void __init check_timer(void)
 	pin2  = ioapic_i8259.pin;
 	apic2 = ioapic_i8259.apic;
 
-	if (pin1 == 0)
-		timer_uses_ioapic_pin_0 = 1;
-
 	printk(KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
 		vector, apic1, pin1, apic2, pin2);
 
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 15dd87110298..15265ee11f89 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -1176,8 +1176,7 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity)
 			 * So test for this condition, and if necessary, avoid
 			 * the pin collision.
 			 */
-			if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0))
-				gsi = pci_irq++;
+			gsi = pci_irq++;
 			/*
 			 * Don't assign IRQ used by ACPI SCI
 			 */
diff --git a/include/asm-x86/io_apic.h b/include/asm-x86/io_apic.h
index 0f5b3fef0b08..095e8e30a345 100644
--- a/include/asm-x86/io_apic.h
+++ b/include/asm-x86/io_apic.h
@@ -146,7 +146,6 @@ extern int io_apic_get_version(int ioapic);
 extern int io_apic_get_redir_entries(int ioapic);
 extern int io_apic_set_pci_routing(int ioapic, int pin, int irq,
 				   int edge_level, int active_high_low);
-extern int timer_uses_ioapic_pin_0;
 #endif /* CONFIG_ACPI */
 
 extern int (*ioapic_renumber_irq)(int ioapic, int irq);
-- 
cgit v1.2.3


From f7d909d5475bb27d261389a3902860e086b0d4c9 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Fri, 14 Mar 2008 07:56:32 +0000
Subject: x86: simplify sync_test_bit()

There really is no need for a redundant implementation here, just keep
the alternative name for allowing consumers to use consistent naming.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/sync_bitops.h | 21 +--------------------
 1 file changed, 1 insertion(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/sync_bitops.h b/include/asm-x86/sync_bitops.h
index 6b775c905666..bc249f40e0ee 100644
--- a/include/asm-x86/sync_bitops.h
+++ b/include/asm-x86/sync_bitops.h
@@ -123,26 +123,7 @@ static inline int sync_test_and_change_bit(int nr, volatile unsigned long* addr)
 	return oldbit;
 }
 
-static __always_inline int sync_constant_test_bit(int nr, const volatile unsigned long *addr)
-{
-	return ((1UL << (nr & 31)) &
-		(((const volatile unsigned int *)addr)[nr >> 5])) != 0;
-}
-
-static inline int sync_var_test_bit(int nr, const volatile unsigned long * addr)
-{
-	int oldbit;
-
-	__asm__ __volatile__("btl %2,%1\n\tsbbl %0,%0"
-			     :"=r" (oldbit)
-			     :"m" (ADDR),"Ir" (nr));
-	return oldbit;
-}
-
-#define sync_test_bit(nr,addr)			\
-	(__builtin_constant_p(nr) ?		\
-	 sync_constant_test_bit((nr),(addr)) :	\
-	 sync_var_test_bit((nr),(addr)))
+#define sync_test_bit test_bit
 
 #undef ADDR
 
-- 
cgit v1.2.3


From a5ae1c372dc5bbaee905bcede524d7180d22b362 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@gmail.com>
Date: Thu, 13 Mar 2008 19:44:56 +0300
Subject: x86: processor.h - use PAGE_SIZE instead of numeric value

This patch replaces numeric constant with an appropriate macro

Also 0x800000000000UL is changed to bit shifting which is complement
to the code comment (thanks hpa for notice)

Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 40227c9bf51b..b0dece41dbc4 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -870,7 +870,7 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 /*
  * User space process size. 47bits minus one guard page.
  */
-#define TASK_SIZE64		(0x800000000000UL - 4096)
+#define TASK_SIZE64	((1UL << 47) - PAGE_SIZE)
 
 /* This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
-- 
cgit v1.2.3


From 272b9cad6e7a2f61b13cfcd7dde0010e02e9376e Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel.send@gmail.com>
Date: Thu, 20 Mar 2008 23:58:33 -0700
Subject: x86: early memtest to find bad ram

do simple memtest after init_memory_mapping

use find_e820_area_size to find all ram range that is not reserved.

and do some simple bits test to find some bad ram.

if find some bad ram, use reserve_early to exclude that range.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/e820_64.c |  70 +++++++++++++++++++++++++++++-
 arch/x86/mm/init_64.c     | 106 +++++++++++++++++++++++++++++++++++++++++++++-
 include/asm-x86/e820_64.h |   3 ++
 3 files changed, 177 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index 4a0953857cb2..4509757844eb 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -114,6 +114,40 @@ again:
 	return changed;
 }
 
+/* Check for already reserved areas */
+static inline int
+bad_addr_size(unsigned long *addrp, unsigned long *sizep, unsigned long align)
+{
+	int i;
+	unsigned long addr = *addrp, last;
+	unsigned long size = *sizep;
+	int changed = 0;
+again:
+	last = addr + size;
+	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+		struct early_res *r = &early_res[i];
+		if (last > r->start && addr < r->start) {
+			size = r->start - addr;
+			changed = 1;
+			goto again;
+		}
+		if (last > r->end && addr < r->end) {
+			addr = round_up(r->end, align);
+			size = last - addr;
+			changed = 1;
+			goto again;
+		}
+		if (last <= r->end && addr >= r->start) {
+			(*sizep)++;
+			return 0;
+		}
+	}
+	if (changed) {
+		*addrp = addr;
+		*sizep = size;
+	}
+	return changed;
+}
 /*
  * This function checks if any part of the range <start,end> is mapped
  * with type.
@@ -190,7 +224,7 @@ unsigned long __init find_e820_area(unsigned long start, unsigned long end,
 		ei_last = ei->addr + ei->size;
 		if (addr < start)
 			addr = round_up(start, align);
-		if (addr > ei_last)
+		if (addr >= ei_last)
 			continue;
 		while (bad_addr(&addr, size, align) && addr+size <= ei_last)
 			;
@@ -204,6 +238,40 @@ unsigned long __init find_e820_area(unsigned long start, unsigned long end,
 	return -1UL;
 }
 
+/*
+ * Find next free range after *start
+ */
+unsigned long __init find_e820_area_size(unsigned long start, unsigned long *sizep, unsigned long align)
+{
+	int i;
+
+	for (i = 0; i < e820.nr_map; i++) {
+		struct e820entry *ei = &e820.map[i];
+		unsigned long addr, last;
+		unsigned long ei_last;
+
+		if (ei->type != E820_RAM)
+			continue;
+		addr = round_up(ei->addr, align);
+		ei_last = ei->addr + ei->size;
+//		printk(KERN_DEBUG "find_e820_area_size : e820 %d [%llx, %lx]\n", i, ei->addr, ei_last);
+		if (addr < start)
+			addr = round_up(start, align);
+//		printk(KERN_DEBUG "find_e820_area_size : 0 [%lx, %lx]\n", addr, ei_last);
+		if (addr >= ei_last)
+			continue;
+		*sizep = ei_last - addr;
+		while (bad_addr_size(&addr, sizep, align) && addr+ *sizep <= ei_last)
+			;
+		last = addr + *sizep;
+//		printk(KERN_DEBUG "find_e820_area_size : 1 [%lx, %lx]\n", addr, last);
+		if (last > ei_last)
+			continue;
+		return addr;
+	}
+	return -1UL;
+
+}
 /*
  * Find the highest page frame number we have available
  */
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 255e51feb157..52f54ee4559f 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -427,6 +427,106 @@ static void __init init_gbpages(void)
 		direct_gbpages = 0;
 }
 
+static void __init memtest(unsigned long start_phys, unsigned long size, unsigned pattern)
+{
+	unsigned long i;
+	unsigned long *start;
+	unsigned long start_bad;
+	unsigned long last_bad;
+	unsigned long val;
+	unsigned long start_phys_aligned;
+	unsigned long count;
+	unsigned long incr;
+
+	switch (pattern) {
+	case 0:
+		val = 0UL;
+		break;
+	case 1:
+		val = -1UL;
+		break;
+	case 2:
+		val = 0x5555555555555555UL;
+		break;
+	case 3:
+		val = 0xaaaaaaaaaaaaaaaaUL;
+		break;
+	default:
+		return;
+	}
+
+	incr = sizeof(unsigned long);
+	start_phys_aligned = ALIGN(start_phys, incr);
+	count = (size - (start_phys_aligned - start_phys))/incr;
+	start = __va(start_phys_aligned);
+	start_bad = 0;
+	last_bad = 0;
+
+	for (i = 0; i < count; i++)
+		start[i] = val;
+	for (i = 0; i < count; i++, start++, start_phys_aligned += incr) {
+		if (*start != val) {
+			if (start_phys_aligned == last_bad + incr) {
+				last_bad += incr;
+			} else {
+				if (start_bad) {
+					printk(KERN_INFO "  %016lxx bad mem addr %016lx - %016lx reserved\n",
+						val, start_bad, last_bad + incr);
+					reserve_early(start_bad, last_bad - start_bad, "BAD RAM");
+				}
+				start_bad = last_bad = start_phys_aligned;
+			}
+		}
+	}
+	if (start_bad) {
+		printk(KERN_INFO "  %016lx bad mem addr %016lx - %016lx reserved\n",
+			val, start_bad, last_bad + incr);
+		reserve_early(start_bad, last_bad - start_bad, "BAD RAM");
+	}
+
+}
+
+static int __initdata memtest_pattern;
+static int __init parse_memtest(char *arg)
+{
+	if (arg)
+		memtest_pattern = simple_strtoul(arg, NULL, 0) + 1;
+	return 0;
+}
+
+early_param("memtest", parse_memtest);
+
+static void __init early_memtest(unsigned long start, unsigned long end)
+{
+	unsigned long t_start, t_size;
+	unsigned pattern;
+
+	if (memtest_pattern)
+		printk(KERN_INFO "early_memtest: pattern num %d", memtest_pattern);
+	for (pattern = 0; pattern < memtest_pattern; pattern++) {
+		t_start = start;
+		t_size = 0;
+		while (t_start < end) {
+			t_start = find_e820_area_size(t_start, &t_size, 1);
+
+			/* done ? */
+			if (t_start >= end)
+				break;
+			if (t_start + t_size > end)
+				t_size = end - t_start;
+
+			printk(KERN_CONT "\n  %016lx - %016lx pattern %d",
+				t_start, t_start + t_size, pattern);
+
+			memtest(t_start, t_size, pattern);
+
+			t_start += t_size;
+		}
+	}
+	if (memtest_pattern)
+		printk(KERN_CONT "\n");
+}
+
 /*
  * Setup the direct mapping of the physical memory at PAGE_OFFSET.
  * This runs before bootmem is initialized and gets pages directly from
@@ -435,8 +535,9 @@ static void __init init_gbpages(void)
 void __init_refok init_memory_mapping(unsigned long start, unsigned long end)
 {
 	unsigned long next;
+	unsigned long start_phys = start, end_phys = end;
 
-	pr_debug("init_memory_mapping\n");
+	printk(KERN_INFO "init_memory_mapping\n");
 
 	/*
 	 * Find space for the kernel direct mapping tables.
@@ -479,6 +580,9 @@ void __init_refok init_memory_mapping(unsigned long start, unsigned long end)
 	if (!after_bootmem)
 		reserve_early(table_start << PAGE_SHIFT,
 				 table_end << PAGE_SHIFT, "PGTABLE");
+
+	if (!after_bootmem)
+		early_memtest(start_phys, end_phys);
 }
 
 #ifndef CONFIG_NUMA
diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index ef653a403e0b..d38820b31c1c 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -16,6 +16,9 @@
 #ifndef __ASSEMBLY__
 extern unsigned long find_e820_area(unsigned long start, unsigned long end, 
 				    unsigned long size, unsigned long align);
+extern unsigned long find_e820_area_size(unsigned long start,
+					 unsigned long *sizep,
+					 unsigned long align);
 extern void add_memory_region(unsigned long start, unsigned long size, 
 			      int type);
 extern void update_memory_range(u64 start, u64 size, unsigned old_type,
-- 
cgit v1.2.3


From 2e5d9c857d4e6c9e7b7d8c8c86a68a7842d213d6 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Tue, 18 Mar 2008 17:00:14 -0700
Subject: x86: PAT infrastructure patch

Sets up pat_init() infrastructure.

PAT MSR has following setting.
	PAT
	|PCD
	||PWT
	|||
	000 WB		_PAGE_CACHE_WB
	001 WC		_PAGE_CACHE_WC
	010 UC-		_PAGE_CACHE_UC_MINUS
	011 UC		_PAGE_CACHE_UC

We are effectively changing WT from boot time setting to WC.
UC_MINUS is used to provide backward compatibility to existing /dev/mem
users(X).

reserve_memtype and free_memtype are new interfaces for maintaining alias-free
mapping. It is currently implemented in a simple way with a linked list and
not optimized. reserve and free tracks the effective memory type, as a result
of PAT and MTRR setting rather than what is actually requested in PAT.

pat_init piggy backs on mtrr_init as the rules for setting both pat and mtrr
are same.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/Kconfig                   |  15 ++
 arch/x86/kernel/cpu/mtrr/generic.c | 120 +++++++++++
 arch/x86/mm/Makefile               |   3 +-
 arch/x86/mm/pageattr.c             |   4 +-
 arch/x86/mm/pat.c                  | 402 +++++++++++++++++++++++++++++++++++++
 include/asm-x86/cpufeature.h       |   1 +
 include/asm-x86/msr-index.h        |   2 +
 include/asm-x86/mtrr.h             |   2 +
 include/asm-x86/pat.h              |  16 ++
 include/asm-x86/pgtable.h          |   6 +
 10 files changed, 568 insertions(+), 3 deletions(-)
 create mode 100644 arch/x86/mm/pat.c
 create mode 100644 include/asm-x86/pat.h

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index fd27048087b8..5b46756e4c7c 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1009,6 +1009,21 @@ config MTRR
 
 	  See <file:Documentation/mtrr.txt> for more information.
 
+config X86_PAT
+	def_bool y
+	prompt "x86 PAT support"
+	depends on MTRR && NONPROMISC_DEVMEM
+	help
+	  Use PAT attributes to setup page level cache control.
+	---help---
+	  PATs are the modern equivalents of MTRRs and are much more
+	  flexible than MTRRs.
+
+	  Say N here if you see bootup problems (boot crash, boot hang,
+	  spontaneous reboots) or a non-working Xorg.
+
+	  If unsure, say Y.
+
 config EFI
 	def_bool n
 	prompt "EFI runtime service support"
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 3e18db4cefee..011e07e99cd1 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -11,6 +11,7 @@
 #include <asm/cpufeature.h>
 #include <asm/processor-flags.h>
 #include <asm/tlbflush.h>
+#include <asm/pat.h>
 #include "mtrr.h"
 
 struct mtrr_state {
@@ -35,6 +36,7 @@ static struct fixed_range_block fixed_range_blocks[] = {
 
 static unsigned long smp_changes_mask;
 static struct mtrr_state mtrr_state = {};
+static int mtrr_state_set;
 
 #undef MODULE_PARAM_PREFIX
 #define MODULE_PARAM_PREFIX "mtrr."
@@ -42,6 +44,106 @@ static struct mtrr_state mtrr_state = {};
 static int mtrr_show;
 module_param_named(show, mtrr_show, bool, 0);
 
+/*
+ * Returns the effective MTRR type for the region
+ * Error returns:
+ * - 0xFE - when the range is "not entirely covered" by _any_ var range MTRR
+ * - 0xFF - when MTRR is not enabled
+ */
+u8 mtrr_type_lookup(u64 start, u64 end)
+{
+	int i;
+	u64 base, mask;
+	u8 prev_match, curr_match;
+
+	if (!mtrr_state_set)
+		return 0xFF;
+
+	if (!mtrr_state.enabled)
+		return 0xFF;
+
+	/* Make end inclusive end, instead of exclusive */
+	end--;
+
+	/* Look in fixed ranges. Just return the type as per start */
+	if (mtrr_state.have_fixed && (start < 0x100000)) {
+		int idx;
+
+		if (start < 0x80000) {
+			idx = 0;
+			idx += (start >> 16);
+			return mtrr_state.fixed_ranges[idx];
+		} else if (start < 0xC0000) {
+			idx = 1 * 8;
+			idx += ((start - 0x80000) >> 14);
+			return mtrr_state.fixed_ranges[idx];
+		} else if (start < 0x1000000) {
+			idx = 3 * 8;
+			idx += ((start - 0xC0000) >> 12);
+			return mtrr_state.fixed_ranges[idx];
+		}
+	}
+
+	/*
+	 * Look in variable ranges
+	 * Look of multiple ranges matching this address and pick type
+	 * as per MTRR precedence
+	 */
+	if (!mtrr_state.enabled & 2) {
+		return mtrr_state.def_type;
+	}
+
+	prev_match = 0xFF;
+	for (i = 0; i < num_var_ranges; ++i) {
+		unsigned short start_state, end_state;
+
+		if (!(mtrr_state.var_ranges[i].mask_lo & (1 << 11)))
+			continue;
+
+		base = (((u64)mtrr_state.var_ranges[i].base_hi) << 32) +
+		       (mtrr_state.var_ranges[i].base_lo & PAGE_MASK);
+		mask = (((u64)mtrr_state.var_ranges[i].mask_hi) << 32) +
+		       (mtrr_state.var_ranges[i].mask_lo & PAGE_MASK);
+
+		start_state = ((start & mask) == (base & mask));
+		end_state = ((end & mask) == (base & mask));
+		if (start_state != end_state)
+			return 0xFE;
+
+		if ((start & mask) != (base & mask)) {
+			continue;
+		}
+
+		curr_match = mtrr_state.var_ranges[i].base_lo & 0xff;
+		if (prev_match == 0xFF) {
+			prev_match = curr_match;
+			continue;
+		}
+
+		if (prev_match == MTRR_TYPE_UNCACHABLE ||
+		    curr_match == MTRR_TYPE_UNCACHABLE) {
+			return MTRR_TYPE_UNCACHABLE;
+		}
+
+		if ((prev_match == MTRR_TYPE_WRBACK &&
+		     curr_match == MTRR_TYPE_WRTHROUGH) ||
+		    (prev_match == MTRR_TYPE_WRTHROUGH &&
+		     curr_match == MTRR_TYPE_WRBACK)) {
+			prev_match = MTRR_TYPE_WRTHROUGH;
+			curr_match = MTRR_TYPE_WRTHROUGH;
+		}
+
+		if (prev_match != curr_match) {
+			return MTRR_TYPE_UNCACHABLE;
+		}
+	}
+
+	if (prev_match != 0xFF)
+		return prev_match;
+
+	return mtrr_state.def_type;
+}
+
 /*  Get the MSR pair relating to a var range  */
 static void
 get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr)
@@ -79,12 +181,16 @@ static void print_fixed(unsigned base, unsigned step, const mtrr_type*types)
 			base, base + step - 1, mtrr_attrib_to_str(*types));
 }
 
+static void prepare_set(void);
+static void post_set(void);
+
 /*  Grab all of the MTRR state for this CPU into *state  */
 void __init get_mtrr_state(void)
 {
 	unsigned int i;
 	struct mtrr_var_range *vrs;
 	unsigned lo, dummy;
+	unsigned long flags;
 
 	vrs = mtrr_state.var_ranges;
 
@@ -131,6 +237,17 @@ void __init get_mtrr_state(void)
 				printk(KERN_INFO "MTRR %u disabled\n", i);
 		}
 	}
+	mtrr_state_set = 1;
+
+	/* PAT setup for BP. We need to go through sync steps here */
+	local_irq_save(flags);
+	prepare_set();
+
+	pat_init();
+
+	post_set();
+	local_irq_restore(flags);
+
 }
 
 /*  Some BIOS's are fucked and don't set all MTRRs the same!  */
@@ -397,6 +514,9 @@ static void generic_set_all(void)
 	/* Actually set the state */
 	mask = set_mtrr_state();
 
+	/* also set PAT */
+	pat_init();
+
 	post_set();
 	local_irq_restore(flags);
 
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index 9ab9889863f0..20941d2954e2 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -1,4 +1,5 @@
-obj-y	:=  init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o
+obj-y	:=  init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
+	    pat.o
 
 obj-$(CONFIG_X86_32)		+= pgtable_32.o
 
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 6cdfc0fd68be..f7d5ca170c22 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -773,14 +773,14 @@ static inline int change_page_attr_clear(unsigned long addr, int numpages,
 int set_memory_uc(unsigned long addr, int numpages)
 {
 	return change_page_attr_set(addr, numpages,
-				    __pgprot(_PAGE_PCD));
+				    __pgprot(_PAGE_CACHE_UC));
 }
 EXPORT_SYMBOL(set_memory_uc);
 
 int set_memory_wb(unsigned long addr, int numpages)
 {
 	return change_page_attr_clear(addr, numpages,
-				      __pgprot(_PAGE_PCD | _PAGE_PWT));
+				      __pgprot(_PAGE_CACHE_MASK));
 }
 EXPORT_SYMBOL(set_memory_wb);
 
diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
new file mode 100644
index 000000000000..7cc71d868483
--- /dev/null
+++ b/arch/x86/mm/pat.c
@@ -0,0 +1,402 @@
+/*
+ * Handle caching attributes in page tables (PAT)
+ *
+ * Authors: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
+ *          Suresh B Siddha <suresh.b.siddha@intel.com>
+ *
+ * Loosely based on earlier PAT patchset from Eric Biederman and Andi Kleen.
+ */
+
+#include <linux/mm.h>
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/fs.h>
+
+#include <asm/msr.h>
+#include <asm/tlbflush.h>
+#include <asm/processor.h>
+#include <asm/pgtable.h>
+#include <asm/pat.h>
+#include <asm/e820.h>
+#include <asm/cacheflush.h>
+#include <asm/fcntl.h>
+#include <asm/mtrr.h>
+
+int pat_wc_enabled = 1;
+
+static u64 __read_mostly boot_pat_state;
+
+static int nopat(char *str)
+{
+	pat_wc_enabled = 0;
+	printk(KERN_INFO "x86: PAT support disabled.\n");
+
+	return 0;
+}
+early_param("nopat", nopat);
+
+static int pat_known_cpu(void)
+{
+	if (!pat_wc_enabled)
+		return 0;
+
+	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
+	    (boot_cpu_data.x86 == 0xF ||
+	     (boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model >= 15))) {
+		if (cpu_has_pat) {
+			return 1;
+		}
+	}
+
+	pat_wc_enabled = 0;
+	printk(KERN_INFO "CPU and/or kernel does not support PAT.\n");
+	return 0;
+}
+
+enum {
+	PAT_UC = 0,		/* uncached */
+	PAT_WC = 1,		/* Write combining */
+	PAT_WT = 4,		/* Write Through */
+	PAT_WP = 5,		/* Write Protected */
+	PAT_WB = 6,		/* Write Back (default) */
+	PAT_UC_MINUS = 7,	/* UC, but can be overriden by MTRR */
+};
+
+#define PAT(x,y)	((u64)PAT_ ## y << ((x)*8))
+
+void pat_init(void)
+{
+	u64 pat;
+
+#ifndef CONFIG_X86_PAT
+	nopat(NULL);
+#endif
+
+	/* Boot CPU enables PAT based on CPU feature */
+	if (!smp_processor_id() && !pat_known_cpu())
+		return;
+
+	/* APs enable PAT iff boot CPU has enabled it before */
+	if (smp_processor_id() && !pat_wc_enabled)
+		return;
+
+	/* Set PWT to Write-Combining. All other bits stay the same */
+	/*
+	 * PTE encoding used in Linux:
+	 *      PAT
+	 *      |PCD
+	 *      ||PWT
+	 *      |||
+	 *      000 WB		_PAGE_CACHE_WB
+	 *      001 WC		_PAGE_CACHE_WC
+	 *      010 UC-		_PAGE_CACHE_UC_MINUS
+	 *      011 UC		_PAGE_CACHE_UC
+	 * PAT bit unused
+	 */
+	pat = PAT(0,WB) | PAT(1,WC) | PAT(2,UC_MINUS) | PAT(3,UC) |
+	      PAT(4,WB) | PAT(5,WC) | PAT(6,UC_MINUS) | PAT(7,UC);
+
+	/* Boot CPU check */
+	if (!smp_processor_id()) {
+		rdmsrl(MSR_IA32_CR_PAT, boot_pat_state);
+	}
+
+	wrmsrl(MSR_IA32_CR_PAT, pat);
+	printk(KERN_INFO "x86 PAT enabled: cpu %d, old 0x%Lx, new 0x%Lx\n",
+	       smp_processor_id(), boot_pat_state, pat);
+}
+
+#undef PAT
+
+static char *cattr_name(unsigned long flags)
+{
+	switch (flags & _PAGE_CACHE_MASK) {
+		case _PAGE_CACHE_UC:		return "uncached";
+		case _PAGE_CACHE_UC_MINUS:	return "uncached-minus";
+		case _PAGE_CACHE_WB:		return "write-back";
+		case _PAGE_CACHE_WC:		return "write-combining";
+		default:			return "broken";
+	}
+}
+
+/*
+ * The global memtype list keeps track of memory type for specific
+ * physical memory areas. Conflicting memory types in different
+ * mappings can cause CPU cache corruption. To avoid this we keep track.
+ *
+ * The list is sorted based on starting address and can contain multiple
+ * entries for each address (this allows reference counting for overlapping
+ * areas). All the aliases have the same cache attributes of course.
+ * Zero attributes are represented as holes.
+ *
+ * Currently the data structure is a list because the number of mappings
+ * are expected to be relatively small. If this should be a problem
+ * it could be changed to a rbtree or similar.
+ *
+ * memtype_lock protects the whole list.
+ */
+
+struct memtype {
+	u64 start;
+	u64 end;
+	unsigned long type;
+	struct list_head nd;
+};
+
+static LIST_HEAD(memtype_list);
+static DEFINE_SPINLOCK(memtype_lock); 	/* protects memtype list */
+
+/*
+ * Does intersection of PAT memory type and MTRR memory type and returns
+ * the resulting memory type as PAT understands it.
+ * (Type in pat and mtrr will not have same value)
+ * The intersection is based on "Effective Memory Type" tables in IA-32
+ * SDM vol 3a
+ */
+static int pat_x_mtrr_type(u64 start, u64 end, unsigned long prot,
+				unsigned long *ret_prot)
+{
+	unsigned long pat_type;
+	u8 mtrr_type;
+
+	mtrr_type = mtrr_type_lookup(start, end);
+	if (mtrr_type == 0xFF) {		/* MTRR not enabled */
+		*ret_prot = prot;
+		return 0;
+	}
+	if (mtrr_type == 0xFE) {		/* MTRR match error */
+		*ret_prot = _PAGE_CACHE_UC;
+		return -1;
+	}
+	if (mtrr_type != MTRR_TYPE_UNCACHABLE &&
+	    mtrr_type != MTRR_TYPE_WRBACK &&
+	    mtrr_type != MTRR_TYPE_WRCOMB) {	/* MTRR type unhandled */
+		*ret_prot = _PAGE_CACHE_UC;
+		return -1;
+	}
+
+	pat_type = prot & _PAGE_CACHE_MASK;
+	prot &= (~_PAGE_CACHE_MASK);
+
+	/* Currently doing intersection by hand. Optimize it later. */
+	if (pat_type == _PAGE_CACHE_WC) {
+		*ret_prot = prot | _PAGE_CACHE_WC;
+	} else if (pat_type == _PAGE_CACHE_UC_MINUS) {
+		*ret_prot = prot | _PAGE_CACHE_UC_MINUS;
+	} else if (pat_type == _PAGE_CACHE_UC ||
+	           mtrr_type == MTRR_TYPE_UNCACHABLE) {
+		*ret_prot = prot | _PAGE_CACHE_UC;
+	} else if (mtrr_type == MTRR_TYPE_WRCOMB) {
+		*ret_prot = prot | _PAGE_CACHE_WC;
+	} else {
+		*ret_prot = prot | _PAGE_CACHE_WB;
+	}
+
+	return 0;
+}
+
+int reserve_memtype(u64 start, u64 end, unsigned long req_type,
+			unsigned long *ret_type)
+{
+	struct memtype *new_entry = NULL;
+	struct memtype *parse;
+	unsigned long actual_type;
+	int err = 0;
+
+	/* Only track when pat_wc_enabled */
+	if (!pat_wc_enabled) {
+		if (ret_type)
+			*ret_type = req_type;
+
+		return 0;
+	}
+
+	/* Low ISA region is always mapped WB in page table. No need to track */
+	if (start >= ISA_START_ADDRESS && (end - 1) <= ISA_END_ADDRESS) {
+		if (ret_type)
+			*ret_type = _PAGE_CACHE_WB;
+
+		return 0;
+	}
+
+	req_type &= _PAGE_CACHE_MASK;
+	err = pat_x_mtrr_type(start, end, req_type, &actual_type);
+	if (err) {
+		if (ret_type)
+			*ret_type = actual_type;
+
+		return -EINVAL;
+	}
+
+	new_entry  = kmalloc(sizeof(struct memtype), GFP_KERNEL);
+	if (!new_entry)
+		return -ENOMEM;
+
+	new_entry->start = start;
+	new_entry->end = end;
+	new_entry->type = actual_type;
+
+	if (ret_type)
+		*ret_type = actual_type;
+
+	spin_lock(&memtype_lock);
+
+	/* Search for existing mapping that overlaps the current range */
+	list_for_each_entry(parse, &memtype_list, nd) {
+		struct memtype *saved_ptr;
+
+		if (parse->start >= end) {
+			list_add(&new_entry->nd, parse->nd.prev);
+			new_entry = NULL;
+			break;
+		}
+
+		if (start <= parse->start && end >= parse->start) {
+			if (actual_type != parse->type && ret_type) {
+				actual_type = parse->type;
+				*ret_type = actual_type;
+				new_entry->type = actual_type;
+			}
+
+			if (actual_type != parse->type) {
+				printk(
+		KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
+					current->comm, current->pid,
+					start, end,
+					cattr_name(actual_type),
+					cattr_name(parse->type));
+				err = -EBUSY;
+				break;
+			}
+
+			saved_ptr = parse;
+			/*
+			 * Check to see whether the request overlaps more
+			 * than one entry in the list
+			 */
+			list_for_each_entry_continue(parse, &memtype_list, nd) {
+				if (end <= parse->start) {
+					break;
+				}
+
+				if (actual_type != parse->type) {
+					printk(
+		KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
+						current->comm, current->pid,
+						start, end,
+						cattr_name(actual_type),
+						cattr_name(parse->type));
+					err = -EBUSY;
+					break;
+				}
+			}
+
+			if (err) {
+				break;
+			}
+
+			/* No conflict. Go ahead and add this new entry */
+			list_add(&new_entry->nd, saved_ptr->nd.prev);
+			new_entry = NULL;
+			break;
+		}
+
+		if (start < parse->end) {
+			if (actual_type != parse->type && ret_type) {
+				actual_type = parse->type;
+				*ret_type = actual_type;
+				new_entry->type = actual_type;
+			}
+
+			if (actual_type != parse->type) {
+				printk(
+		KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
+					current->comm, current->pid,
+					start, end,
+					cattr_name(actual_type),
+					cattr_name(parse->type));
+				err = -EBUSY;
+				break;
+			}
+
+			saved_ptr = parse;
+			/*
+			 * Check to see whether the request overlaps more
+			 * than one entry in the list
+			 */
+			list_for_each_entry_continue(parse, &memtype_list, nd) {
+				if (end <= parse->start) {
+					break;
+				}
+
+				if (actual_type != parse->type) {
+					printk(
+		KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
+						current->comm, current->pid,
+						start, end,
+						cattr_name(actual_type),
+						cattr_name(parse->type));
+					err = -EBUSY;
+					break;
+				}
+			}
+
+			if (err) {
+				break;
+			}
+
+			/* No conflict. Go ahead and add this new entry */
+			list_add(&new_entry->nd, &saved_ptr->nd);
+			new_entry = NULL;
+			break;
+		}
+	}
+
+	if (err) {
+		kfree(new_entry);
+		spin_unlock(&memtype_lock);
+		return err;
+	}
+
+	if (new_entry) {
+		/* No conflict. Not yet added to the list. Add to the tail */
+		list_add_tail(&new_entry->nd, &memtype_list);
+	}
+
+	spin_unlock(&memtype_lock);
+	return err;
+}
+
+int free_memtype(u64 start, u64 end)
+{
+	struct memtype *ml;
+	int err = -EINVAL;
+
+	/* Only track when pat_wc_enabled */
+	if (!pat_wc_enabled) {
+		return 0;
+	}
+
+	/* Low ISA region is always mapped WB. No need to track */
+	if (start >= ISA_START_ADDRESS && end <= ISA_END_ADDRESS) {
+		return 0;
+	}
+
+	spin_lock(&memtype_lock);
+	list_for_each_entry(ml, &memtype_list, nd) {
+		if (ml->start == start && ml->end == end) {
+			list_del(&ml->nd);
+			kfree(ml);
+			err = 0;
+			break;
+		}
+	}
+	spin_unlock(&memtype_lock);
+
+	if (err) {
+		printk(KERN_DEBUG "%s:%d freeing invalid memtype %Lx-%Lx\n",
+			current->comm, current->pid, start, end);
+	}
+	return err;
+}
+
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 90feb6f2562c..0d609c837a41 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -186,6 +186,7 @@ extern const char * const x86_power_flags[32];
 #define cpu_has_bts		boot_cpu_has(X86_FEATURE_BTS)
 #define cpu_has_gbpages		boot_cpu_has(X86_FEATURE_GBPAGES)
 #define cpu_has_arch_perfmon	boot_cpu_has(X86_FEATURE_ARCH_PERFMON)
+#define cpu_has_pat		boot_cpu_has(X86_FEATURE_PAT)
 
 #if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64)
 # define cpu_has_invlpg		1
diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h
index 3ed97144c07b..af4e07f661b8 100644
--- a/include/asm-x86/msr-index.h
+++ b/include/asm-x86/msr-index.h
@@ -57,6 +57,8 @@
 #define MSR_MTRRfix4K_F8000		0x0000026f
 #define MSR_MTRRdefType			0x000002ff
 
+#define MSR_IA32_CR_PAT			0x00000277
+
 #define MSR_IA32_DEBUGCTLMSR		0x000001d9
 #define MSR_IA32_LASTBRANCHFROMIP	0x000001db
 #define MSR_IA32_LASTBRANCHTOIP		0x000001dc
diff --git a/include/asm-x86/mtrr.h b/include/asm-x86/mtrr.h
index 319d065800be..968794af93f9 100644
--- a/include/asm-x86/mtrr.h
+++ b/include/asm-x86/mtrr.h
@@ -84,6 +84,8 @@ struct mtrr_gentry
 
 #ifdef __KERNEL__
 
+extern u8 mtrr_type_lookup(u64 addr, u64 end);
+
 /*  The following functions are for use by other drivers  */
 # ifdef CONFIG_MTRR
 extern void mtrr_save_fixed_ranges(void *);
diff --git a/include/asm-x86/pat.h b/include/asm-x86/pat.h
new file mode 100644
index 000000000000..8b822b5a1786
--- /dev/null
+++ b/include/asm-x86/pat.h
@@ -0,0 +1,16 @@
+
+#ifndef _ASM_PAT_H
+#define _ASM_PAT_H 1
+
+#include <linux/types.h>
+
+extern int pat_wc_enabled;
+
+extern void pat_init(void);
+
+extern int reserve_memtype(u64 start, u64 end,
+		unsigned long req_type, unsigned long *ret_type);
+extern int free_memtype(u64 start, u64 end);
+
+#endif
+
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 9cf472aeb9ce..ca6deb3de7c0 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -57,6 +57,12 @@
 
 #define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
 
+#define _PAGE_CACHE_MASK	(_PAGE_PCD | _PAGE_PWT)
+#define _PAGE_CACHE_WB		(0)
+#define _PAGE_CACHE_WC		(_PAGE_PWT)
+#define _PAGE_CACHE_UC_MINUS	(_PAGE_PCD)
+#define _PAGE_CACHE_UC		(_PAGE_PCD | _PAGE_PWT)
+
 #define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
 #define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
 
-- 
cgit v1.2.3


From 3a96ce8cac808fbed5493adc5c605bced28e2ca1 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Tue, 18 Mar 2008 17:00:16 -0700
Subject: x86: PAT make ioremap_change_attr non-static

Make ioremap_change_attr() non-static and use prot_val in place of ioremap_mode.
This interface is used in subsequent PAT patches.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/ioremap.c | 29 ++++++++++++-----------------
 include/asm-x86/io.h  |  3 +++
 2 files changed, 15 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index df95d1d6b4df..2ac09a5822cb 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -20,11 +20,6 @@
 #include <asm/tlbflush.h>
 #include <asm/pgalloc.h>
 
-enum ioremap_mode {
-	IOR_MODE_UNCACHED,
-	IOR_MODE_CACHED,
-};
-
 #ifdef CONFIG_X86_64
 
 unsigned long __phys_addr(unsigned long x)
@@ -90,18 +85,18 @@ int page_is_ram(unsigned long pagenr)
  * Fix up the linear direct mapping of the kernel to avoid cache attribute
  * conflicts.
  */
-static int ioremap_change_attr(unsigned long vaddr, unsigned long size,
-			       enum ioremap_mode mode)
+int ioremap_change_attr(unsigned long vaddr, unsigned long size,
+			       unsigned long prot_val)
 {
 	unsigned long nrpages = size >> PAGE_SHIFT;
 	int err;
 
-	switch (mode) {
-	case IOR_MODE_UNCACHED:
+	switch (prot_val) {
+	case _PAGE_CACHE_UC:
 	default:
 		err = set_memory_uc(vaddr, nrpages);
 		break;
-	case IOR_MODE_CACHED:
+	case _PAGE_CACHE_WB:
 		err = set_memory_wb(vaddr, nrpages);
 		break;
 	}
@@ -119,7 +114,7 @@ static int ioremap_change_attr(unsigned long vaddr, unsigned long size,
  * caller shouldn't need to know that small detail.
  */
 static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
-			       enum ioremap_mode mode)
+			       unsigned long prot_val)
 {
 	unsigned long pfn, offset, last_addr, vaddr;
 	struct vm_struct *area;
@@ -156,12 +151,12 @@ static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
 		WARN_ON_ONCE(is_ram);
 	}
 
-	switch (mode) {
-	case IOR_MODE_UNCACHED:
+	switch (prot_val) {
+	case _PAGE_CACHE_UC:
 	default:
 		prot = PAGE_KERNEL_NOCACHE;
 		break;
-	case IOR_MODE_CACHED:
+	case _PAGE_CACHE_WB:
 		prot = PAGE_KERNEL;
 		break;
 	}
@@ -186,7 +181,7 @@ static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
 		return NULL;
 	}
 
-	if (ioremap_change_attr(vaddr, size, mode) < 0) {
+	if (ioremap_change_attr(vaddr, size, prot_val) < 0) {
 		vunmap(area->addr);
 		return NULL;
 	}
@@ -217,13 +212,13 @@ static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
  */
 void __iomem *ioremap_nocache(resource_size_t phys_addr, unsigned long size)
 {
-	return __ioremap(phys_addr, size, IOR_MODE_UNCACHED);
+	return __ioremap(phys_addr, size, _PAGE_CACHE_UC);
 }
 EXPORT_SYMBOL(ioremap_nocache);
 
 void __iomem *ioremap_cache(resource_size_t phys_addr, unsigned long size)
 {
-	return __ioremap(phys_addr, size, IOR_MODE_CACHED);
+	return __ioremap(phys_addr, size, _PAGE_CACHE_WB);
 }
 EXPORT_SYMBOL(ioremap_cache);
 
diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h
index 5a58b176dd61..6fa150fa68fa 100644
--- a/include/asm-x86/io.h
+++ b/include/asm-x86/io.h
@@ -3,3 +3,6 @@
 #else
 # include "io_64.h"
 #endif
+extern int ioremap_change_attr(unsigned long vaddr, unsigned long size,
+				unsigned long prot_val);
+
-- 
cgit v1.2.3


From 1219333dfdd488e85f08cf07881b8bc63cf92f21 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Tue, 18 Mar 2008 17:00:18 -0700
Subject: x86: PAT use reserve free memtype in set_memory_uc

Use reserve_memtype and free_memtype interfaces in set_memory_uc/set_memory_wb
interfaces to avoid aliasing.
Usage model of set_memory_uc and set_memory_wb is for RAM memory and users
will first call set_memory_uc and call set_memory_wb after use to reset the
attribute.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/ioremap.c        |  4 ++--
 arch/x86/mm/pageattr.c       | 21 +++++++++++++++++++--
 include/asm-x86/cacheflush.h |  2 ++
 3 files changed, 23 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 20c01f2b2e11..0cdb7f11ce49 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -95,10 +95,10 @@ int ioremap_change_attr(unsigned long vaddr, unsigned long size,
 	switch (prot_val) {
 	case _PAGE_CACHE_UC:
 	default:
-		err = set_memory_uc(vaddr, nrpages);
+		err = _set_memory_uc(vaddr, nrpages);
 		break;
 	case _PAGE_CACHE_WB:
-		err = set_memory_wb(vaddr, nrpages);
+		err = _set_memory_wb(vaddr, nrpages);
 		break;
 	}
 
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index f7d5ca170c22..938df5e8402b 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -19,6 +19,7 @@
 #include <asm/uaccess.h>
 #include <asm/pgalloc.h>
 #include <asm/proto.h>
+#include <asm/pat.h>
 
 /*
  * The current flushing context - we pass it instead of 5 arguments:
@@ -770,18 +771,34 @@ static inline int change_page_attr_clear(unsigned long addr, int numpages,
 	return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask);
 }
 
-int set_memory_uc(unsigned long addr, int numpages)
+int _set_memory_uc(unsigned long addr, int numpages)
 {
 	return change_page_attr_set(addr, numpages,
 				    __pgprot(_PAGE_CACHE_UC));
 }
+
+int set_memory_uc(unsigned long addr, int numpages)
+{
+	if (reserve_memtype(addr, addr + numpages * PAGE_SIZE,
+	                    _PAGE_CACHE_UC, NULL))
+		return -EINVAL;
+
+	return _set_memory_uc(addr, numpages);
+}
 EXPORT_SYMBOL(set_memory_uc);
 
-int set_memory_wb(unsigned long addr, int numpages)
+int _set_memory_wb(unsigned long addr, int numpages)
 {
 	return change_page_attr_clear(addr, numpages,
 				      __pgprot(_PAGE_CACHE_MASK));
 }
+
+int set_memory_wb(unsigned long addr, int numpages)
+{
+	free_memtype(addr, addr + numpages * PAGE_SIZE);
+
+	return _set_memory_wb(addr, numpages);
+}
 EXPORT_SYMBOL(set_memory_wb);
 
 int set_memory_x(unsigned long addr, int numpages)
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 5396c212d8c0..5676fba10a09 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -34,6 +34,8 @@ int set_pages_nx(struct page *page, int numpages);
 int set_pages_ro(struct page *page, int numpages);
 int set_pages_rw(struct page *page, int numpages);
 
+int _set_memory_uc(unsigned long addr, int numpages);
+int _set_memory_wb(unsigned long addr, int numpages);
 int set_memory_uc(unsigned long addr, int numpages);
 int set_memory_wb(unsigned long addr, int numpages);
 int set_memory_x(unsigned long addr, int numpages);
-- 
cgit v1.2.3


From ef354af4629e5cc76a3f64fc46d452f2b56d5a59 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Tue, 18 Mar 2008 17:00:23 -0700
Subject: x86: PAT add set_memory_wc() interface

Add a set_memory_wc interface(), similar to set_memory_uc interface.
Callers has to call set_memory_uc, set_memory_wb and
set_memory_wc, set_memory_wb as pairs.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/pageattr.c       | 19 +++++++++++++++++++
 include/asm-x86/cacheflush.h |  2 ++
 2 files changed, 21 insertions(+)

(limited to 'include')

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 938df5e8402b..270cab2e6030 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -787,6 +787,25 @@ int set_memory_uc(unsigned long addr, int numpages)
 }
 EXPORT_SYMBOL(set_memory_uc);
 
+int _set_memory_wc(unsigned long addr, int numpages)
+{
+	return change_page_attr_set(addr, numpages,
+				    __pgprot(_PAGE_CACHE_WC));
+}
+
+int set_memory_wc(unsigned long addr, int numpages)
+{
+	if (!pat_wc_enabled)
+		return set_memory_uc(addr, numpages);
+
+	if (reserve_memtype(addr, addr + numpages * PAGE_SIZE,
+		_PAGE_CACHE_WC, NULL))
+		return -EINVAL;
+
+	return _set_memory_wc(addr, numpages);
+}
+EXPORT_SYMBOL(set_memory_wc);
+
 int _set_memory_wb(unsigned long addr, int numpages)
 {
 	return change_page_attr_clear(addr, numpages,
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 5676fba10a09..90437d3f7610 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -35,8 +35,10 @@ int set_pages_ro(struct page *page, int numpages);
 int set_pages_rw(struct page *page, int numpages);
 
 int _set_memory_uc(unsigned long addr, int numpages);
+int _set_memory_wc(unsigned long addr, int numpages);
 int _set_memory_wb(unsigned long addr, int numpages);
 int set_memory_uc(unsigned long addr, int numpages);
+int set_memory_wc(unsigned long addr, int numpages);
 int set_memory_wb(unsigned long addr, int numpages);
 int set_memory_x(unsigned long addr, int numpages);
 int set_memory_nx(unsigned long addr, int numpages);
-- 
cgit v1.2.3


From b310f381d220b2c6e3fab16e8c6e4ca13eea75b2 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Tue, 18 Mar 2008 17:00:24 -0700
Subject: x86: PAT add ioremap_wc() interface

Introduce ioremap_wc for wc remap.

(generic wrapper is in a later patch)

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/ioremap.c     | 30 ++++++++++++++++++++++++++++++
 include/asm-x86/io.h      |  3 +++
 include/asm-x86/pgtable.h |  2 ++
 3 files changed, 35 insertions(+)

(limited to 'include')

diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 0cdb7f11ce49..51cd3956c564 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -97,6 +97,9 @@ int ioremap_change_attr(unsigned long vaddr, unsigned long size,
 	default:
 		err = _set_memory_uc(vaddr, nrpages);
 		break;
+	case _PAGE_CACHE_WC:
+		err = _set_memory_wc(vaddr, nrpages);
+		break;
 	case _PAGE_CACHE_WB:
 		err = _set_memory_wb(vaddr, nrpages);
 		break;
@@ -166,8 +169,13 @@ static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
 		 * Do not fallback to certain memory types with certain
 		 * requested type:
 		 * - request is uncached, return cannot be write-back
+		 * - request is uncached, return cannot be write-combine
+		 * - request is write-combine, return cannot be write-back
 		 */
 		if ((prot_val == _PAGE_CACHE_UC &&
+		     (new_prot_val == _PAGE_CACHE_WB ||
+		      new_prot_val == _PAGE_CACHE_WC)) ||
+		    (prot_val == _PAGE_CACHE_WC &&
 		     new_prot_val == _PAGE_CACHE_WB)) {
 			free_memtype(phys_addr, phys_addr + size);
 			return NULL;
@@ -180,6 +188,9 @@ static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
 	default:
 		prot = PAGE_KERNEL_NOCACHE;
 		break;
+	case _PAGE_CACHE_WC:
+		prot = PAGE_KERNEL_WC;
+		break;
 	case _PAGE_CACHE_WB:
 		prot = PAGE_KERNEL;
 		break;
@@ -235,6 +246,25 @@ void __iomem *ioremap_nocache(resource_size_t phys_addr, unsigned long size)
 }
 EXPORT_SYMBOL(ioremap_nocache);
 
+/**
+ * ioremap_wc	-	map memory into CPU space write combined
+ * @offset:	bus address of the memory
+ * @size:	size of the resource to map
+ *
+ * This version of ioremap ensures that the memory is marked write combining.
+ * Write combining allows faster writes to some hardware devices.
+ *
+ * Must be freed with iounmap.
+ */
+void __iomem *ioremap_wc(unsigned long phys_addr, unsigned long size)
+{
+	if (pat_wc_enabled)
+		return __ioremap(phys_addr, size, _PAGE_CACHE_WC);
+	else
+		return ioremap_nocache(phys_addr, size);
+}
+EXPORT_SYMBOL(ioremap_wc);
+
 void __iomem *ioremap_cache(resource_size_t phys_addr, unsigned long size)
 {
 	return __ioremap(phys_addr, size, _PAGE_CACHE_WB);
diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h
index 6fa150fa68fa..599cad3505c1 100644
--- a/include/asm-x86/io.h
+++ b/include/asm-x86/io.h
@@ -1,3 +1,5 @@
+#define ARCH_HAS_IOREMAP_WC
+
 #ifdef CONFIG_X86_32
 # include "io_32.h"
 #else
@@ -5,4 +7,5 @@
 #endif
 extern int ioremap_change_attr(unsigned long vaddr, unsigned long size,
 				unsigned long prot_val);
+extern void __iomem * ioremap_wc(unsigned long offset, unsigned long size);
 
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index ca6deb3de7c0..e814cfe96af2 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -90,6 +90,7 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 #define __PAGE_KERNEL_RO		(__PAGE_KERNEL & ~_PAGE_RW)
 #define __PAGE_KERNEL_RX		(__PAGE_KERNEL_EXEC & ~_PAGE_RW)
 #define __PAGE_KERNEL_EXEC_NOCACHE	(__PAGE_KERNEL_EXEC | _PAGE_PCD | _PAGE_PWT)
+#define __PAGE_KERNEL_WC		(__PAGE_KERNEL | _PAGE_CACHE_WC)
 #define __PAGE_KERNEL_NOCACHE		(__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT)
 #define __PAGE_KERNEL_UC_MINUS		(__PAGE_KERNEL | _PAGE_PCD)
 #define __PAGE_KERNEL_VSYSCALL		(__PAGE_KERNEL_RX | _PAGE_USER)
@@ -107,6 +108,7 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
 #define PAGE_KERNEL_RO			MAKE_GLOBAL(__PAGE_KERNEL_RO)
 #define PAGE_KERNEL_EXEC		MAKE_GLOBAL(__PAGE_KERNEL_EXEC)
 #define PAGE_KERNEL_RX			MAKE_GLOBAL(__PAGE_KERNEL_RX)
+#define PAGE_KERNEL_WC			MAKE_GLOBAL(__PAGE_KERNEL_WC)
 #define PAGE_KERNEL_NOCACHE		MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE)
 #define PAGE_KERNEL_UC_MINUS		MAKE_GLOBAL(__PAGE_KERNEL_UC_MINUS)
 #define PAGE_KERNEL_EXEC_NOCACHE	MAKE_GLOBAL(__PAGE_KERNEL_EXEC_NOCACHE)
-- 
cgit v1.2.3


From 52783fa8d6b847857fdd86df418e09c026d816f5 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 21 Mar 2008 15:42:28 +0100
Subject: x86: PAT fix

build fix for !CONFIG_MTRR.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mtrr.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mtrr.h b/include/asm-x86/mtrr.h
index 968794af93f9..ee172296e056 100644
--- a/include/asm-x86/mtrr.h
+++ b/include/asm-x86/mtrr.h
@@ -84,10 +84,9 @@ struct mtrr_gentry
 
 #ifdef __KERNEL__
 
-extern u8 mtrr_type_lookup(u64 addr, u64 end);
-
 /*  The following functions are for use by other drivers  */
 # ifdef CONFIG_MTRR
+extern u8 mtrr_type_lookup(u64 addr, u64 end);
 extern void mtrr_save_fixed_ranges(void *);
 extern void mtrr_save_state(void);
 extern int mtrr_add (unsigned long base, unsigned long size,
@@ -101,6 +100,13 @@ extern void mtrr_ap_init(void);
 extern void mtrr_bp_init(void);
 extern int mtrr_trim_uncached_memory(unsigned long end_pfn);
 #  else
+static inline u8 mtrr_type_lookup(u64 addr, u64 end)
+{
+	/*
+	 * Return no-MTRRs:
+	 */
+	return 0xff;
+}
 #define mtrr_save_fixed_ranges(arg) do {} while (0)
 #define mtrr_save_state() do {} while (0)
 static __inline__ int mtrr_add (unsigned long base, unsigned long size,
-- 
cgit v1.2.3


From 35605a1027ac630f85a1b95684f7e86b82498cd6 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel.send@gmail.com>
Date: Mon, 24 Mar 2008 16:02:01 -0700
Subject: x86: enable PAT for amd k8 and fam10h

make known_pat_cpu to think amd k8 and fam10h is ok too.

also make tom2 below to be WRBACK

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/generic.c | 17 +++++++++++++++++
 arch/x86/kernel/cpu/mtrr/main.c    |  2 +-
 arch/x86/mm/pat.c                  |  6 ++++++
 include/asm-x86/mtrr.h             |  1 +
 4 files changed, 25 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 011e07e99cd1..74ec2ea4ed3e 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -37,6 +37,7 @@ static struct fixed_range_block fixed_range_blocks[] = {
 static unsigned long smp_changes_mask;
 static struct mtrr_state mtrr_state = {};
 static int mtrr_state_set;
+static u64 tom2;
 
 #undef MODULE_PARAM_PREFIX
 #define MODULE_PARAM_PREFIX "mtrr."
@@ -138,6 +139,11 @@ u8 mtrr_type_lookup(u64 start, u64 end)
 		}
 	}
 
+	if (tom2) {
+		if (start >= (1ULL<<32) && (end < tom2))
+			return MTRR_TYPE_WRBACK;
+	}
+
 	if (prev_match != 0xFF)
 		return prev_match;
 
@@ -206,6 +212,15 @@ void __init get_mtrr_state(void)
 	mtrr_state.def_type = (lo & 0xff);
 	mtrr_state.enabled = (lo & 0xc00) >> 10;
 
+	if (amd_special_default_mtrr()) {
+		unsigned lo, hi;
+		/* TOP_MEM2 */
+		rdmsr(MSR_K8_TOP_MEM2, lo, hi);
+		tom2 = hi;
+		tom2 <<= 32;
+		tom2 |= lo;
+		tom2 &= 0xffffff8000000ULL;
+	}
 	if (mtrr_show) {
 		int high_width;
 
@@ -236,6 +251,8 @@ void __init get_mtrr_state(void)
 			else
 				printk(KERN_INFO "MTRR %u disabled\n", i);
 		}
+		if (tom2)
+			printk(KERN_INFO "TOM2: %016lx aka %ldM\n", tom2, tom2>>20);
 	}
 	mtrr_state_set = 1;
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index a6450b3ae759..6a1e278d9323 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -627,7 +627,7 @@ early_param("disable_mtrr_trim", disable_mtrr_trim_setup);
 #define Tom2Enabled (1U << 21)
 #define Tom2ForceMemTypeWB (1U << 22)
 
-static __init int amd_special_default_mtrr(void)
+int __init amd_special_default_mtrr(void)
 {
 	u32 l, h;
 
diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
index 57a2af36d6ee..0648a2225b09 100644
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -47,6 +47,12 @@ static int pat_known_cpu(void)
 			return 1;
 		}
 	}
+	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
+	     boot_cpu_data.x86 >= 0xf && boot_cpu_data.x86 <= 0x11) {
+		if (cpu_has_pat) {
+			return 1;
+		}
+	}
 
 	pat_wc_enabled = 0;
 	printk(KERN_INFO "CPU and/or kernel does not support PAT.\n");
diff --git a/include/asm-x86/mtrr.h b/include/asm-x86/mtrr.h
index ee172296e056..d3d26625aeaf 100644
--- a/include/asm-x86/mtrr.h
+++ b/include/asm-x86/mtrr.h
@@ -99,6 +99,7 @@ extern void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi);
 extern void mtrr_ap_init(void);
 extern void mtrr_bp_init(void);
 extern int mtrr_trim_uncached_memory(unsigned long end_pfn);
+extern int amd_special_default_mtrr(void);
 #  else
 static inline u8 mtrr_type_lookup(u64 addr, u64 end)
 {
-- 
cgit v1.2.3


From 709f744f18ebc3a810d29c8d5502bf20c3cecc70 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Thu, 13 Mar 2008 09:08:51 +0000
Subject: x86: bitops asm constraint fixes

This (simplified) piece of code didn't behave as expected due to
incorrect constraints in some of the bitops functions, when
X86_FEATURE_xxx is referring to other than the first long:

int test(struct cpuinfo_x86 *c) {
	if (cpu_has(c, X86_FEATURE_xxx))
		clear_cpu_cap(c, X86_FEATURE_xxx);
	return cpu_has(c, X86_FEATURE_xxx);
}

I'd really like understand, though, what the policy of (not) having a
"memory" clobber in these operations is - currently, this appears to
be totally inconsistent. Also, many comments of the non-atomic
functions say those may also be re-ordered - this contradicts the use
of "asm volatile" in there, which again I'd like to understand.

As much as all of these, using 'int' for the 'nr' parameter and
'void *' for the 'addr' one is in conflict with
Documentation/atomic_ops.txt, especially because bt{,c,r,s} indeed
take the bit index as signed (which hence would really need special
precaution) and access the full 32 bits (if 'unsigned long' was used
properly here, 64 bits for x86-64) pointed at, so invalid uses like
referencing a 'char' array cannot currently be caught.

Finally, the code with and without this patch relies heavily on the
-fno-strict-aliasing compiler switch and I'm not certain this really
is a good idea.

In the light of all of this I'm sending this as RFC, as fixing the
above might warrant a much bigger patch...

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/bitops.h | 43 ++++++++++++++++++++++++-------------------
 1 file changed, 24 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bitops.h b/include/asm-x86/bitops.h
index 1a23ce1a5697..7a76555b676a 100644
--- a/include/asm-x86/bitops.h
+++ b/include/asm-x86/bitops.h
@@ -24,9 +24,12 @@
 /* Technically wrong, but this avoids compilation errors on some gcc
    versions. */
 #define ADDR "=m" (*(volatile long *) addr)
+#define BIT_ADDR "=m" (((volatile int *) addr)[nr >> 5])
 #else
 #define ADDR "+m" (*(volatile long *) addr)
+#define BIT_ADDR "+m" (((volatile int *) addr)[nr >> 5])
 #endif
+#define BASE_ADDR "m" (*(volatile int *) addr)
 
 /**
  * set_bit - Atomically set a bit in memory
@@ -79,9 +82,8 @@ static inline void __set_bit(int nr, volatile void *addr)
  */
 static inline void clear_bit(int nr, volatile void *addr)
 {
-	asm volatile(LOCK_PREFIX "btr %1,%0"
-		     : ADDR
-		     : "Ir" (nr));
+	asm volatile(LOCK_PREFIX "btr %1,%2"
+		     : BIT_ADDR : "Ir" (nr), BASE_ADDR);
 }
 
 /*
@@ -100,7 +102,7 @@ static inline void clear_bit_unlock(unsigned nr, volatile void *addr)
 
 static inline void __clear_bit(int nr, volatile void *addr)
 {
-	asm volatile("btr %1,%0" : ADDR : "Ir" (nr));
+	asm volatile("btr %1,%2" : BIT_ADDR : "Ir" (nr), BASE_ADDR);
 }
 
 /*
@@ -135,7 +137,7 @@ static inline void __clear_bit_unlock(unsigned nr, volatile void *addr)
  */
 static inline void __change_bit(int nr, volatile void *addr)
 {
-	asm volatile("btc %1,%0" : ADDR : "Ir" (nr));
+	asm volatile("btc %1,%2" : BIT_ADDR : "Ir" (nr), BASE_ADDR);
 }
 
 /**
@@ -149,8 +151,8 @@ static inline void __change_bit(int nr, volatile void *addr)
  */
 static inline void change_bit(int nr, volatile void *addr)
 {
-	asm volatile(LOCK_PREFIX "btc %1,%0"
-		     : ADDR : "Ir" (nr));
+	asm volatile(LOCK_PREFIX "btc %1,%2"
+		     : BIT_ADDR : "Ir" (nr), BASE_ADDR);
 }
 
 /**
@@ -198,10 +200,10 @@ static inline int __test_and_set_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
-	asm("bts %2,%1\n\t"
-	    "sbb %0,%0"
-	    : "=r" (oldbit), ADDR
-	    : "Ir" (nr));
+	asm volatile("bts %2,%3\n\t"
+		     "sbb %0,%0"
+		     : "=r" (oldbit), BIT_ADDR
+		     : "Ir" (nr), BASE_ADDR);
 	return oldbit;
 }
 
@@ -238,10 +240,10 @@ static inline int __test_and_clear_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
-	asm volatile("btr %2,%1\n\t"
+	asm volatile("btr %2,%3\n\t"
 		     "sbb %0,%0"
-		     : "=r" (oldbit), ADDR
-		     : "Ir" (nr));
+		     : "=r" (oldbit), BIT_ADDR
+		     : "Ir" (nr), BASE_ADDR);
 	return oldbit;
 }
 
@@ -250,10 +252,10 @@ static inline int __test_and_change_bit(int nr, volatile void *addr)
 {
 	int oldbit;
 
-	asm volatile("btc %2,%1\n\t"
+	asm volatile("btc %2,%3\n\t"
 		     "sbb %0,%0"
-		     : "=r" (oldbit), ADDR
-		     : "Ir" (nr) : "memory");
+		     : "=r" (oldbit), BIT_ADDR
+		     : "Ir" (nr), BASE_ADDR);
 
 	return oldbit;
 }
@@ -288,10 +290,11 @@ static inline int variable_test_bit(int nr, volatile const void *addr)
 {
 	int oldbit;
 
-	asm volatile("bt %2,%1\n\t"
+	asm volatile("bt %2,%3\n\t"
 		     "sbb %0,%0"
 		     : "=r" (oldbit)
-		     : "m" (*(unsigned long *)addr), "Ir" (nr));
+		     : "m" (((volatile const int *)addr)[nr >> 5]),
+		       "Ir" (nr), BASE_ADDR);
 
 	return oldbit;
 }
@@ -310,6 +313,8 @@ static int test_bit(int nr, const volatile unsigned long *addr);
 	 constant_test_bit((nr),(addr)) :	\
 	 variable_test_bit((nr),(addr)))
 
+#undef BASE_ADDR
+#undef BIT_ADDR
 #undef ADDR
 
 #ifdef CONFIG_X86_32
-- 
cgit v1.2.3


From 8434e73d9ec2aeaa86389e362b960ffba5edd9c9 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 21 Mar 2008 08:31:57 -0500
Subject: x86: increase max physical memory size of 64-bit

Increase the maximum physical address size of x86_64 system
to 44-bits. This is in preparation for future chips that
support larger physical memory sizes.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/sparsemem.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/sparsemem.h b/include/asm-x86/sparsemem.h
index fa58cd55411a..fa684f353aa7 100644
--- a/include/asm-x86/sparsemem.h
+++ b/include/asm-x86/sparsemem.h
@@ -26,8 +26,8 @@
 # endif
 #else /* CONFIG_X86_32 */
 # define SECTION_SIZE_BITS	27 /* matt - 128 is convenient right now */
-# define MAX_PHYSADDR_BITS	40
-# define MAX_PHYSMEM_BITS	40
+# define MAX_PHYSADDR_BITS	44
+# define MAX_PHYSMEM_BITS	44
 #endif
 
 #endif /* CONFIG_SPARSEMEM */
-- 
cgit v1.2.3


From 9b967106da0357ef8b08847dce35584a04134f20 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Fri, 21 Mar 2008 15:14:07 -0700
Subject: x86: fix smpboot integration

>  yhlu@mpk:~/xx/xx/kernel/x86/linux-2.6> git-bisect bad
>  d1c707188ad646c8094cac9afb1738e7d0196ff2 is first bad commit
>  commit d1c707188ad646c8094cac9afb1738e7d0196ff2
>  Author: Glauber de Oliveira Costa <gcosta@redhat.com>
>  Date:   Wed Mar 19 14:25:53 2008 -0300
>
>     x86: include mach_apic.h in smpboot_64.c and smpboot.c
>
>     After the inclusion, a lot of files needs fixing for conflicts,
>     some of them in the headers themselves, to accomodate for both
>     i386 and x86_64 versions.
>
>     [ mingo@elte.hu: build fix ]
>
>     Signed-off-by: Glauber Costa <gcosta@redhat.com>
>     Signed-off-by: Ingo Molnar <mingo@elte.hu>
>
>  :040000 040000 19f574e64bb8003bbe984f3a8c1315db969dfdcd
>  6ffe96588c77bc936705599fa110107856201115 M      arch
>  :040000 040000 61269347ad4f384ed85cc87c4f2d004ed94492ac
>  8f5c713da25579a3cdf63db3d4c2f795261d0521 M      include
>  yhlu@mpk:~/xx/xx/kernel/x86/linux-2.6>
>

attached patch fixes that.
---
 include/asm-x86/mach-default/mach_apicdef.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mach-default/mach_apicdef.h b/include/asm-x86/mach-default/mach_apicdef.h
index 7b78275e6d33..e4b29ba37de6 100644
--- a/include/asm-x86/mach-default/mach_apicdef.h
+++ b/include/asm-x86/mach-default/mach_apicdef.h
@@ -5,13 +5,12 @@
 
 #ifdef CONFIG_X86_64
 #define	APIC_ID_MASK		(0xFFu<<24)
+#define GET_APIC_ID(x)          (((x)>>24)&0xFFu)
 #define	SET_APIC_ID(x)		(((x)<<24))
 #else
 #define		APIC_ID_MASK		(0xF<<24)
-#endif
-
 static inline unsigned get_apic_id(unsigned long x) 
-{ 
+{
 	unsigned int ver = GET_APIC_VERSION(apic_read(APIC_LVR));
 	if (APIC_XAPIC(ver))
 		return (((x)>>24)&0xFF);
@@ -20,5 +19,6 @@ static inline unsigned get_apic_id(unsigned long x)
 } 
 
 #define		GET_APIC_ID(x)	get_apic_id(x)
+#endif
 
 #endif
-- 
cgit v1.2.3


From d3463c5a66147bdd21b5865ea29fdca50ea28f7f Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 4 Apr 2008 13:31:15 +0200
Subject: undo "x86: fix breakage of vSMP irq operations"

revert:

  "x86: fix breakage of vSMP irq operations"

the irqflags.h unification will solve this in a cleaner way.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/irqflags.h | 29 -----------------------------
 1 file changed, 29 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h
index 0e2292483b35..92021c1ffa3a 100644
--- a/include/asm-x86/irqflags.h
+++ b/include/asm-x86/irqflags.h
@@ -70,26 +70,6 @@ static inline void raw_local_irq_restore(unsigned long flags)
 	native_restore_fl(flags);
 }
 
-#ifdef CONFIG_X86_VSMP
-
-/*
- * Interrupt control for the VSMP architecture:
- */
-
-static inline void raw_local_irq_disable(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-	raw_local_irq_restore((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC);
-}
-
-static inline void raw_local_irq_enable(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-	raw_local_irq_restore((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC));
-}
-
-#else
-
 static inline void raw_local_irq_disable(void)
 {
 	native_irq_disable();
@@ -100,8 +80,6 @@ static inline void raw_local_irq_enable(void)
 	native_irq_enable();
 }
 
-#endif
-
 /*
  * Used in the idle loop; sti takes one instruction cycle
  * to complete:
@@ -159,17 +137,10 @@ static inline unsigned long __raw_local_irq_save(void)
 #define raw_local_irq_save(flags) \
 		do { (flags) = __raw_local_irq_save(); } while (0)
 
-#ifdef CONFIG_X86_VSMP
-static inline int raw_irqs_disabled_flags(unsigned long flags)
-{
-	return !(flags & X86_EFLAGS_IF) || (flags & X86_EFLAGS_AC);
-}
-#else
 static inline int raw_irqs_disabled_flags(unsigned long flags)
 {
 	return !(flags & X86_EFLAGS_IF);
 }
-#endif
 
 static inline int raw_irqs_disabled(void)
 {
-- 
cgit v1.2.3


From 8dbeeb24e4d64a8a4bb49d54a6e4d987b9d3b6fc Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:36 -0700
Subject: include/asm-x86/acpi.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/acpi.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h
index 7a72d6aa50be..14411c9de46f 100644
--- a/include/asm-x86/acpi.h
+++ b/include/asm-x86/acpi.h
@@ -67,16 +67,16 @@ int __acpi_release_global_lock(unsigned int *lock);
  */
 #define ACPI_DIV_64_BY_32(n_hi, n_lo, d32, q32, r32) \
 	asm("divl %2;"				     \
-	    :"=a"(q32), "=d"(r32)		     \
-	    :"r"(d32),				     \
+	    : "=a"(q32), "=d"(r32)		     \
+	    : "r"(d32),				     \
 	     "0"(n_lo), "1"(n_hi))
 
 
 #define ACPI_SHIFT_RIGHT_64(n_hi, n_lo) \
 	asm("shrl   $1,%2	;"	\
 	    "rcrl   $1,%3;"		\
-	    :"=r"(n_hi), "=r"(n_lo)	\
-	    :"0"(n_hi), "1"(n_lo))
+	    : "=r"(n_hi), "=r"(n_lo)	\
+	    : "0"(n_hi), "1"(n_lo))
 
 #ifdef CONFIG_ACPI
 extern int acpi_lapic;
-- 
cgit v1.2.3


From 2ac1ea7ccd81f0383c6525e495a31d18fcac92db Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:37 -0700
Subject: include/asm-x86/alternative.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/alternative.h | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/alternative.h b/include/asm-x86/alternative.h
index d26416b5722c..1f6a9ca10126 100644
--- a/include/asm-x86/alternative.h
+++ b/include/asm-x86/alternative.h
@@ -66,8 +66,8 @@ extern void alternatives_smp_module_del(struct module *mod);
 extern void alternatives_smp_switch(int smp);
 #else
 static inline void alternatives_smp_module_add(struct module *mod, char *name,
-					void *locks, void *locks_end,
-					void *text, void *text_end) {}
+					       void *locks, void *locks_end,
+					       void *text, void *text_end) {}
 static inline void alternatives_smp_module_del(struct module *mod) {}
 static inline void alternatives_smp_switch(int smp) {}
 #endif	/* CONFIG_SMP */
@@ -148,9 +148,8 @@ struct paravirt_patch_site;
 void apply_paravirt(struct paravirt_patch_site *start,
 		    struct paravirt_patch_site *end);
 #else
-static inline void
-apply_paravirt(struct paravirt_patch_site *start,
-	       struct paravirt_patch_site *end)
+static inline void apply_paravirt(struct paravirt_patch_site *start,
+				  struct paravirt_patch_site *end)
 {}
 #define __parainstructions	NULL
 #define __parainstructions_end	NULL
-- 
cgit v1.2.3


From b03aa8c6eb2d743ef4658b494dc1780e7740efc0 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:38 -0700
Subject: include/asm-x86/a.out-core.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/a.out-core.h | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/a.out-core.h b/include/asm-x86/a.out-core.h
index d2b6e11d3e97..714207a1c387 100644
--- a/include/asm-x86/a.out-core.h
+++ b/include/asm-x86/a.out-core.h
@@ -29,8 +29,9 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump)
 	dump->magic = CMAGIC;
 	dump->start_code = 0;
 	dump->start_stack = regs->sp & ~(PAGE_SIZE - 1);
-	dump->u_tsize = ((unsigned long) current->mm->end_code) >> PAGE_SHIFT;
-	dump->u_dsize = ((unsigned long) (current->mm->brk + (PAGE_SIZE-1))) >> PAGE_SHIFT;
+	dump->u_tsize = ((unsigned long)current->mm->end_code) >> PAGE_SHIFT;
+	dump->u_dsize = ((unsigned long)(current->mm->brk + (PAGE_SIZE - 1)))
+			>> PAGE_SHIFT;
 	dump->u_dsize -= dump->u_tsize;
 	dump->u_ssize = 0;
 	dump->u_debugreg[0] = current->thread.debugreg0;
@@ -43,7 +44,8 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump)
 	dump->u_debugreg[7] = current->thread.debugreg7;
 
 	if (dump->start_stack < TASK_SIZE)
-		dump->u_ssize = ((unsigned long) (TASK_SIZE - dump->start_stack)) >> PAGE_SHIFT;
+		dump->u_ssize = ((unsigned long)(TASK_SIZE - dump->start_stack))
+				>> PAGE_SHIFT;
 
 	dump->regs.bx = regs->bx;
 	dump->regs.cx = regs->cx;
@@ -55,7 +57,7 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump)
 	dump->regs.ds = (u16)regs->ds;
 	dump->regs.es = (u16)regs->es;
 	dump->regs.fs = (u16)regs->fs;
-	savesegment(gs,gs);
+	savesegment(gs, gs);
 	dump->regs.orig_ax = regs->orig_ax;
 	dump->regs.ip = regs->ip;
 	dump->regs.cs = (u16)regs->cs;
@@ -63,7 +65,7 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump)
 	dump->regs.sp = regs->sp;
 	dump->regs.ss = (u16)regs->ss;
 
-	dump->u_fpvalid = dump_fpu (regs, &dump->i387);
+	dump->u_fpvalid = dump_fpu(regs, &dump->i387);
 }
 
 #endif /* CONFIG_X86_32 */
-- 
cgit v1.2.3


From 79a4a961ef92744dd8876963690a8beb56758457 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:39 -0700
Subject: include/asm-x86/apicdef.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/apicdef.h | 50 +++++++++++++++++++++++------------------------
 1 file changed, 25 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index 674a2280e21e..8b244683431b 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -14,10 +14,10 @@
 
 #define	APIC_LVR	0x30
 #define		APIC_LVR_MASK		0xFF00FF
-#define		GET_APIC_VERSION(x)	((x)&0xFFu)
-#define		GET_APIC_MAXLVT(x)	(((x)>>16)&0xFFu)
+#define		GET_APIC_VERSION(x)	((x) & 0xFFu)
+#define		GET_APIC_MAXLVT(x)	(((x) >> 16) & 0xFFu)
 #ifdef CONFIG_X86_32
-#  define	APIC_INTEGRATED(x)	((x)&0xF0u)
+#  define	APIC_INTEGRATED(x)	((x) & 0xF0u)
 #else
 #  define	APIC_INTEGRATED(x)	(1)
 #endif
@@ -31,16 +31,16 @@
 #define		APIC_EIO_ACK		0x0
 #define	APIC_RRR	0xC0
 #define	APIC_LDR	0xD0
-#define		APIC_LDR_MASK		(0xFFu<<24)
-#define		GET_APIC_LOGICAL_ID(x)	(((x)>>24)&0xFFu)
-#define		SET_APIC_LOGICAL_ID(x)	(((x)<<24))
+#define		APIC_LDR_MASK		(0xFFu << 24)
+#define		GET_APIC_LOGICAL_ID(x)	(((x) >> 24) & 0xFFu)
+#define		SET_APIC_LOGICAL_ID(x)	(((x) << 24))
 #define		APIC_ALL_CPUS		0xFFu
 #define	APIC_DFR	0xE0
 #define		APIC_DFR_CLUSTER		0x0FFFFFFFul
 #define		APIC_DFR_FLAT			0xFFFFFFFFul
 #define	APIC_SPIV	0xF0
-#define		APIC_SPIV_FOCUS_DISABLED	(1<<9)
-#define		APIC_SPIV_APIC_ENABLED		(1<<8)
+#define		APIC_SPIV_FOCUS_DISABLED	(1 << 9)
+#define		APIC_SPIV_APIC_ENABLED		(1 << 8)
 #define	APIC_ISR	0x100
 #define	APIC_ISR_NR     0x8     /* Number of 32 bit ISR registers. */
 #define	APIC_TMR	0x180
@@ -76,27 +76,27 @@
 #define		APIC_DM_EXTINT		0x00700
 #define		APIC_VECTOR_MASK	0x000FF
 #define	APIC_ICR2	0x310
-#define		GET_APIC_DEST_FIELD(x)	(((x)>>24)&0xFF)
-#define		SET_APIC_DEST_FIELD(x)	((x)<<24)
+#define		GET_APIC_DEST_FIELD(x)	(((x) >> 24) & 0xFF)
+#define		SET_APIC_DEST_FIELD(x)	((x) << 24)
 #define	APIC_LVTT	0x320
 #define	APIC_LVTTHMR	0x330
 #define	APIC_LVTPC	0x340
 #define	APIC_LVT0	0x350
-#define		APIC_LVT_TIMER_BASE_MASK	(0x3<<18)
-#define		GET_APIC_TIMER_BASE(x)		(((x)>>18)&0x3)
-#define		SET_APIC_TIMER_BASE(x)		(((x)<<18))
+#define		APIC_LVT_TIMER_BASE_MASK	(0x3 << 18)
+#define		GET_APIC_TIMER_BASE(x)		(((x) >> 18) & 0x3)
+#define		SET_APIC_TIMER_BASE(x)		(((x) << 18))
 #define		APIC_TIMER_BASE_CLKIN		0x0
 #define		APIC_TIMER_BASE_TMBASE		0x1
 #define		APIC_TIMER_BASE_DIV		0x2
-#define		APIC_LVT_TIMER_PERIODIC		(1<<17)
-#define		APIC_LVT_MASKED			(1<<16)
-#define		APIC_LVT_LEVEL_TRIGGER		(1<<15)
-#define		APIC_LVT_REMOTE_IRR		(1<<14)
-#define		APIC_INPUT_POLARITY		(1<<13)
-#define		APIC_SEND_PENDING		(1<<12)
+#define		APIC_LVT_TIMER_PERIODIC		(1 << 17)
+#define		APIC_LVT_MASKED			(1 << 16)
+#define		APIC_LVT_LEVEL_TRIGGER		(1 << 15)
+#define		APIC_LVT_REMOTE_IRR		(1 << 14)
+#define		APIC_INPUT_POLARITY		(1 << 13)
+#define		APIC_SEND_PENDING		(1 << 12)
 #define		APIC_MODE_MASK			0x700
-#define		GET_APIC_DELIVERY_MODE(x)	(((x)>>8)&0x7)
-#define		SET_APIC_DELIVERY_MODE(x, y)	(((x)&~0x700)|((y)<<8))
+#define		GET_APIC_DELIVERY_MODE(x)	(((x) >> 8) & 0x7)
+#define		SET_APIC_DELIVERY_MODE(x, y)	(((x) & ~0x700) | ((y) << 8))
 #define			APIC_MODE_FIXED		0x0
 #define			APIC_MODE_NMI		0x4
 #define			APIC_MODE_EXTINT	0x7
@@ -105,7 +105,7 @@
 #define	APIC_TMICT	0x380
 #define	APIC_TMCCT	0x390
 #define	APIC_TDCR	0x3E0
-#define		APIC_TDR_DIV_TMBASE	(1<<2)
+#define		APIC_TDR_DIV_TMBASE	(1 << 2)
 #define		APIC_TDR_DIV_1		0xB
 #define		APIC_TDR_DIV_2		0x0
 #define		APIC_TDR_DIV_4		0x1
@@ -115,14 +115,14 @@
 #define		APIC_TDR_DIV_64		0x9
 #define		APIC_TDR_DIV_128	0xA
 #define	APIC_EILVT0     0x500
-#define		APIC_EILVT_NR_AMD_K8	1	/* Number of extended interrupts */
+#define		APIC_EILVT_NR_AMD_K8	1	/* # of extended interrupts */
 #define		APIC_EILVT_NR_AMD_10H	4
-#define		APIC_EILVT_LVTOFF(x)	(((x)>>4)&0xF)
+#define		APIC_EILVT_LVTOFF(x)	(((x) >> 4) & 0xF)
 #define		APIC_EILVT_MSG_FIX	0x0
 #define		APIC_EILVT_MSG_SMI	0x2
 #define		APIC_EILVT_MSG_NMI	0x4
 #define		APIC_EILVT_MSG_EXT	0x7
-#define		APIC_EILVT_MASKED	(1<<16)
+#define		APIC_EILVT_MASKED	(1 << 16)
 #define	APIC_EILVT1     0x510
 #define	APIC_EILVT2     0x520
 #define	APIC_EILVT3     0x530
-- 
cgit v1.2.3


From 3c311febfa8cc240e2922931d7403a6bb7f3fa1b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:40 -0700
Subject: include/asm-x86/apic.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/apic.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index b11d6524fb13..be9639a9a186 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -67,7 +67,7 @@ static inline void native_apic_write(unsigned long reg, u32 v)
 
 static inline void native_apic_write_atomic(unsigned long reg, u32 v)
 {
-	(void) xchg((u32*)(APIC_BASE + reg), v);
+	(void)xchg((u32 *)(APIC_BASE + reg), v);
 }
 
 static inline u32 native_apic_read(unsigned long reg)
-- 
cgit v1.2.3


From 78ff12eec42a4141d22dac4fdab04994384f6385 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:41 -0700
Subject: include/asm-x86/atomic_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/atomic_32.h | 143 +++++++++++++++++++++-----------------------
 1 file changed, 68 insertions(+), 75 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/atomic_32.h b/include/asm-x86/atomic_32.h
index 437aac801711..21a4825148c0 100644
--- a/include/asm-x86/atomic_32.h
+++ b/include/asm-x86/atomic_32.h
@@ -15,138 +15,133 @@
  * on us. We need to use _exactly_ the address the user gave us,
  * not some alias that contains the same information.
  */
-typedef struct { int counter; } atomic_t;
+typedef struct {
+	int counter;
+} atomic_t;
 
 #define ATOMIC_INIT(i)	{ (i) }
 
 /**
  * atomic_read - read atomic variable
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically reads the value of @v.
- */ 
+ */
 #define atomic_read(v)		((v)->counter)
 
 /**
  * atomic_set - set atomic variable
  * @v: pointer of type atomic_t
  * @i: required value
- * 
+ *
  * Atomically sets the value of @v to @i.
- */ 
-#define atomic_set(v,i)		(((v)->counter) = (i))
+ */
+#define atomic_set(v, i)	(((v)->counter) = (i))
 
 /**
  * atomic_add - add integer to atomic variable
  * @i: integer value to add
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically adds @i to @v.
  */
-static __inline__ void atomic_add(int i, atomic_t *v)
+static inline void atomic_add(int i, atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "addl %1,%0"
-		:"+m" (v->counter)
-		:"ir" (i));
+	asm volatile(LOCK_PREFIX "addl %1,%0"
+		     : "+m" (v->counter)
+		     : "ir" (i));
 }
 
 /**
  * atomic_sub - subtract integer from atomic variable
  * @i: integer value to subtract
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically subtracts @i from @v.
  */
-static __inline__ void atomic_sub(int i, atomic_t *v)
+static inline void atomic_sub(int i, atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "subl %1,%0"
-		:"+m" (v->counter)
-		:"ir" (i));
+	asm volatile(LOCK_PREFIX "subl %1,%0"
+		     : "+m" (v->counter)
+		     : "ir" (i));
 }
 
 /**
  * atomic_sub_and_test - subtract value from variable and test result
  * @i: integer value to subtract
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically subtracts @i from @v and returns
  * true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int atomic_sub_and_test(int i, atomic_t *v)
+static inline int atomic_sub_and_test(int i, atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "subl %2,%0; sete %1"
-		:"+m" (v->counter), "=qm" (c)
-		:"ir" (i) : "memory");
+	asm volatile(LOCK_PREFIX "subl %2,%0; sete %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : "ir" (i) : "memory");
 	return c;
 }
 
 /**
  * atomic_inc - increment atomic variable
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically increments @v by 1.
- */ 
-static __inline__ void atomic_inc(atomic_t *v)
+ */
+static inline void atomic_inc(atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "incl %0"
-		:"+m" (v->counter));
+	asm volatile(LOCK_PREFIX "incl %0"
+		     : "+m" (v->counter));
 }
 
 /**
  * atomic_dec - decrement atomic variable
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically decrements @v by 1.
- */ 
-static __inline__ void atomic_dec(atomic_t *v)
+ */
+static inline void atomic_dec(atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "decl %0"
-		:"+m" (v->counter));
+	asm volatile(LOCK_PREFIX "decl %0"
+		     : "+m" (v->counter));
 }
 
 /**
  * atomic_dec_and_test - decrement and test
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically decrements @v by 1 and
  * returns true if the result is 0, or false for all other
  * cases.
- */ 
-static __inline__ int atomic_dec_and_test(atomic_t *v)
+ */
+static inline int atomic_dec_and_test(atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "decl %0; sete %1"
-		:"+m" (v->counter), "=qm" (c)
-		: : "memory");
+	asm volatile(LOCK_PREFIX "decl %0; sete %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : : "memory");
 	return c != 0;
 }
 
 /**
- * atomic_inc_and_test - increment and test 
+ * atomic_inc_and_test - increment and test
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically increments @v by 1
  * and returns true if the result is zero, or false for all
  * other cases.
- */ 
-static __inline__ int atomic_inc_and_test(atomic_t *v)
+ */
+static inline int atomic_inc_and_test(atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "incl %0; sete %1"
-		:"+m" (v->counter), "=qm" (c)
-		: : "memory");
+	asm volatile(LOCK_PREFIX "incl %0; sete %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : : "memory");
 	return c != 0;
 }
 
@@ -154,19 +149,18 @@ static __inline__ int atomic_inc_and_test(atomic_t *v)
  * atomic_add_negative - add and test if negative
  * @v: pointer of type atomic_t
  * @i: integer value to add
- * 
+ *
  * Atomically adds @i to @v and returns true
  * if the result is negative, or false when
  * result is greater than or equal to zero.
- */ 
-static __inline__ int atomic_add_negative(int i, atomic_t *v)
+ */
+static inline int atomic_add_negative(int i, atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "addl %2,%0; sets %1"
-		:"+m" (v->counter), "=qm" (c)
-		:"ir" (i) : "memory");
+	asm volatile(LOCK_PREFIX "addl %2,%0; sets %1"
+		     : "+m" (v->counter), "=qm" (c)
+		     : "ir" (i) : "memory");
 	return c;
 }
 
@@ -177,20 +171,19 @@ static __inline__ int atomic_add_negative(int i, atomic_t *v)
  *
  * Atomically adds @i to @v and returns @i + @v
  */
-static __inline__ int atomic_add_return(int i, atomic_t *v)
+static inline int atomic_add_return(int i, atomic_t *v)
 {
 	int __i;
 #ifdef CONFIG_M386
 	unsigned long flags;
-	if(unlikely(boot_cpu_data.x86 <= 3))
+	if (unlikely(boot_cpu_data.x86 <= 3))
 		goto no_xadd;
 #endif
 	/* Modern 486+ processor */
 	__i = i;
-	__asm__ __volatile__(
-		LOCK_PREFIX "xaddl %0, %1"
-		:"+r" (i), "+m" (v->counter)
-		: : "memory");
+	asm volatile(LOCK_PREFIX "xaddl %0, %1"
+		     : "+r" (i), "+m" (v->counter)
+		     : : "memory");
 	return i + __i;
 
 #ifdef CONFIG_M386
@@ -210,9 +203,9 @@ no_xadd: /* Legacy 386 processor */
  *
  * Atomically subtracts @i from @v and returns @v - @i
  */
-static __inline__ int atomic_sub_return(int i, atomic_t *v)
+static inline int atomic_sub_return(int i, atomic_t *v)
 {
-	return atomic_add_return(-i,v);
+	return atomic_add_return(-i, v);
 }
 
 #define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), (old), (new)))
@@ -227,7 +220,7 @@ static __inline__ int atomic_sub_return(int i, atomic_t *v)
  * Atomically adds @a to @v, so long as @v was not already @u.
  * Returns non-zero if @v was not @u, and zero otherwise.
  */
-static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
+static inline int atomic_add_unless(atomic_t *v, int a, int u)
 {
 	int c, old;
 	c = atomic_read(v);
@@ -244,17 +237,17 @@ static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
 
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
-#define atomic_inc_return(v)  (atomic_add_return(1,v))
-#define atomic_dec_return(v)  (atomic_sub_return(1,v))
+#define atomic_inc_return(v)  (atomic_add_return(1, v))
+#define atomic_dec_return(v)  (atomic_sub_return(1, v))
 
 /* These are x86-specific, used by some header files */
-#define atomic_clear_mask(mask, addr) \
-__asm__ __volatile__(LOCK_PREFIX "andl %0,%1" \
-: : "r" (~(mask)),"m" (*addr) : "memory")
+#define atomic_clear_mask(mask, addr)				\
+	asm volatile(LOCK_PREFIX "andl %0,%1"			\
+		     : : "r" (~(mask)), "m" (*(addr)) : "memory")
 
-#define atomic_set_mask(mask, addr) \
-__asm__ __volatile__(LOCK_PREFIX "orl %0,%1" \
-: : "r" (mask),"m" (*(addr)) : "memory")
+#define atomic_set_mask(mask, addr)				\
+	asm volatile(LOCK_PREFIX "orl %0,%1"				\
+		     : : "r" (mask), "m" (*(addr)) : "memory")
 
 /* Atomic operations are already serializing on x86 */
 #define smp_mb__before_atomic_dec()	barrier()
-- 
cgit v1.2.3


From 7edb3cd6cbadb686864b8f180232e4dc69d959e8 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:42 -0700
Subject: include/asm-x86/atomic_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/atomic_64.h | 251 +++++++++++++++++++++-----------------------
 1 file changed, 119 insertions(+), 132 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/atomic_64.h b/include/asm-x86/atomic_64.h
index 2d20a7a19f62..3e0cd7d38335 100644
--- a/include/asm-x86/atomic_64.h
+++ b/include/asm-x86/atomic_64.h
@@ -22,140 +22,135 @@
  * on us. We need to use _exactly_ the address the user gave us,
  * not some alias that contains the same information.
  */
-typedef struct { int counter; } atomic_t;
+typedef struct {
+	int counter;
+} atomic_t;
 
 #define ATOMIC_INIT(i)	{ (i) }
 
 /**
  * atomic_read - read atomic variable
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically reads the value of @v.
- */ 
+ */
 #define atomic_read(v)		((v)->counter)
 
 /**
  * atomic_set - set atomic variable
  * @v: pointer of type atomic_t
  * @i: required value
- * 
+ *
  * Atomically sets the value of @v to @i.
- */ 
-#define atomic_set(v,i)		(((v)->counter) = (i))
+ */
+#define atomic_set(v, i)		(((v)->counter) = (i))
 
 /**
  * atomic_add - add integer to atomic variable
  * @i: integer value to add
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically adds @i to @v.
  */
-static __inline__ void atomic_add(int i, atomic_t *v)
+static inline void atomic_add(int i, atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "addl %1,%0"
-		:"=m" (v->counter)
-		:"ir" (i), "m" (v->counter));
+	asm volatile(LOCK_PREFIX "addl %1,%0"
+		     : "=m" (v->counter)
+		     : "ir" (i), "m" (v->counter));
 }
 
 /**
  * atomic_sub - subtract the atomic variable
  * @i: integer value to subtract
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically subtracts @i from @v.
  */
-static __inline__ void atomic_sub(int i, atomic_t *v)
+static inline void atomic_sub(int i, atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "subl %1,%0"
-		:"=m" (v->counter)
-		:"ir" (i), "m" (v->counter));
+	asm volatile(LOCK_PREFIX "subl %1,%0"
+		     : "=m" (v->counter)
+		     : "ir" (i), "m" (v->counter));
 }
 
 /**
  * atomic_sub_and_test - subtract value from variable and test result
  * @i: integer value to subtract
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically subtracts @i from @v and returns
  * true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int atomic_sub_and_test(int i, atomic_t *v)
+static inline int atomic_sub_and_test(int i, atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "subl %2,%0; sete %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"ir" (i), "m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "subl %2,%0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "ir" (i), "m" (v->counter) : "memory");
 	return c;
 }
 
 /**
  * atomic_inc - increment atomic variable
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically increments @v by 1.
- */ 
-static __inline__ void atomic_inc(atomic_t *v)
+ */
+static inline void atomic_inc(atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "incl %0"
-		:"=m" (v->counter)
-		:"m" (v->counter));
+	asm volatile(LOCK_PREFIX "incl %0"
+		     : "=m" (v->counter)
+		     : "m" (v->counter));
 }
 
 /**
  * atomic_dec - decrement atomic variable
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically decrements @v by 1.
- */ 
-static __inline__ void atomic_dec(atomic_t *v)
+ */
+static inline void atomic_dec(atomic_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "decl %0"
-		:"=m" (v->counter)
-		:"m" (v->counter));
+	asm volatile(LOCK_PREFIX "decl %0"
+		     : "=m" (v->counter)
+		     : "m" (v->counter));
 }
 
 /**
  * atomic_dec_and_test - decrement and test
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically decrements @v by 1 and
  * returns true if the result is 0, or false for all other
  * cases.
- */ 
-static __inline__ int atomic_dec_and_test(atomic_t *v)
+ */
+static inline int atomic_dec_and_test(atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "decl %0; sete %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "decl %0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "m" (v->counter) : "memory");
 	return c != 0;
 }
 
 /**
- * atomic_inc_and_test - increment and test 
+ * atomic_inc_and_test - increment and test
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically increments @v by 1
  * and returns true if the result is zero, or false for all
  * other cases.
- */ 
-static __inline__ int atomic_inc_and_test(atomic_t *v)
+ */
+static inline int atomic_inc_and_test(atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "incl %0; sete %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "incl %0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "m" (v->counter) : "memory");
 	return c != 0;
 }
 
@@ -163,19 +158,18 @@ static __inline__ int atomic_inc_and_test(atomic_t *v)
  * atomic_add_negative - add and test if negative
  * @i: integer value to add
  * @v: pointer of type atomic_t
- * 
+ *
  * Atomically adds @i to @v and returns true
  * if the result is negative, or false when
  * result is greater than or equal to zero.
- */ 
-static __inline__ int atomic_add_negative(int i, atomic_t *v)
+ */
+static inline int atomic_add_negative(int i, atomic_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "addl %2,%0; sets %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"ir" (i), "m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "addl %2,%0; sets %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "ir" (i), "m" (v->counter) : "memory");
 	return c;
 }
 
@@ -186,27 +180,28 @@ static __inline__ int atomic_add_negative(int i, atomic_t *v)
  *
  * Atomically adds @i to @v and returns @i + @v
  */
-static __inline__ int atomic_add_return(int i, atomic_t *v)
+static inline int atomic_add_return(int i, atomic_t *v)
 {
 	int __i = i;
-	__asm__ __volatile__(
-		LOCK_PREFIX "xaddl %0, %1"
-		:"+r" (i), "+m" (v->counter)
-		: : "memory");
+	asm volatile(LOCK_PREFIX "xaddl %0, %1"
+		     : "+r" (i), "+m" (v->counter)
+		     : : "memory");
 	return i + __i;
 }
 
-static __inline__ int atomic_sub_return(int i, atomic_t *v)
+static inline int atomic_sub_return(int i, atomic_t *v)
 {
-	return atomic_add_return(-i,v);
+	return atomic_add_return(-i, v);
 }
 
-#define atomic_inc_return(v)  (atomic_add_return(1,v))
-#define atomic_dec_return(v)  (atomic_sub_return(1,v))
+#define atomic_inc_return(v)  (atomic_add_return(1, v))
+#define atomic_dec_return(v)  (atomic_sub_return(1, v))
 
 /* An 64bit atomic type */
 
-typedef struct { long counter; } atomic64_t;
+typedef struct {
+	long counter;
+} atomic64_t;
 
 #define ATOMIC64_INIT(i)	{ (i) }
 
@@ -226,7 +221,7 @@ typedef struct { long counter; } atomic64_t;
  *
  * Atomically sets the value of @v to @i.
  */
-#define atomic64_set(v,i)		(((v)->counter) = (i))
+#define atomic64_set(v, i)		(((v)->counter) = (i))
 
 /**
  * atomic64_add - add integer to atomic64 variable
@@ -235,12 +230,11 @@ typedef struct { long counter; } atomic64_t;
  *
  * Atomically adds @i to @v.
  */
-static __inline__ void atomic64_add(long i, atomic64_t *v)
+static inline void atomic64_add(long i, atomic64_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "addq %1,%0"
-		:"=m" (v->counter)
-		:"ir" (i), "m" (v->counter));
+	asm volatile(LOCK_PREFIX "addq %1,%0"
+		     : "=m" (v->counter)
+		     : "ir" (i), "m" (v->counter));
 }
 
 /**
@@ -250,12 +244,11 @@ static __inline__ void atomic64_add(long i, atomic64_t *v)
  *
  * Atomically subtracts @i from @v.
  */
-static __inline__ void atomic64_sub(long i, atomic64_t *v)
+static inline void atomic64_sub(long i, atomic64_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "subq %1,%0"
-		:"=m" (v->counter)
-		:"ir" (i), "m" (v->counter));
+	asm volatile(LOCK_PREFIX "subq %1,%0"
+		     : "=m" (v->counter)
+		     : "ir" (i), "m" (v->counter));
 }
 
 /**
@@ -267,14 +260,13 @@ static __inline__ void atomic64_sub(long i, atomic64_t *v)
  * true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int atomic64_sub_and_test(long i, atomic64_t *v)
+static inline int atomic64_sub_and_test(long i, atomic64_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "subq %2,%0; sete %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"ir" (i), "m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "subq %2,%0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "ir" (i), "m" (v->counter) : "memory");
 	return c;
 }
 
@@ -284,12 +276,11 @@ static __inline__ int atomic64_sub_and_test(long i, atomic64_t *v)
  *
  * Atomically increments @v by 1.
  */
-static __inline__ void atomic64_inc(atomic64_t *v)
+static inline void atomic64_inc(atomic64_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "incq %0"
-		:"=m" (v->counter)
-		:"m" (v->counter));
+	asm volatile(LOCK_PREFIX "incq %0"
+		     : "=m" (v->counter)
+		     : "m" (v->counter));
 }
 
 /**
@@ -298,12 +289,11 @@ static __inline__ void atomic64_inc(atomic64_t *v)
  *
  * Atomically decrements @v by 1.
  */
-static __inline__ void atomic64_dec(atomic64_t *v)
+static inline void atomic64_dec(atomic64_t *v)
 {
-	__asm__ __volatile__(
-		LOCK_PREFIX "decq %0"
-		:"=m" (v->counter)
-		:"m" (v->counter));
+	asm volatile(LOCK_PREFIX "decq %0"
+		     : "=m" (v->counter)
+		     : "m" (v->counter));
 }
 
 /**
@@ -314,14 +304,13 @@ static __inline__ void atomic64_dec(atomic64_t *v)
  * returns true if the result is 0, or false for all other
  * cases.
  */
-static __inline__ int atomic64_dec_and_test(atomic64_t *v)
+static inline int atomic64_dec_and_test(atomic64_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "decq %0; sete %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "decq %0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "m" (v->counter) : "memory");
 	return c != 0;
 }
 
@@ -333,14 +322,13 @@ static __inline__ int atomic64_dec_and_test(atomic64_t *v)
  * and returns true if the result is zero, or false for all
  * other cases.
  */
-static __inline__ int atomic64_inc_and_test(atomic64_t *v)
+static inline int atomic64_inc_and_test(atomic64_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "incq %0; sete %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "incq %0; sete %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "m" (v->counter) : "memory");
 	return c != 0;
 }
 
@@ -353,14 +341,13 @@ static __inline__ int atomic64_inc_and_test(atomic64_t *v)
  * if the result is negative, or false when
  * result is greater than or equal to zero.
  */
-static __inline__ int atomic64_add_negative(long i, atomic64_t *v)
+static inline int atomic64_add_negative(long i, atomic64_t *v)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		LOCK_PREFIX "addq %2,%0; sets %1"
-		:"=m" (v->counter), "=qm" (c)
-		:"ir" (i), "m" (v->counter) : "memory");
+	asm volatile(LOCK_PREFIX "addq %2,%0; sets %1"
+		     : "=m" (v->counter), "=qm" (c)
+		     : "ir" (i), "m" (v->counter) : "memory");
 	return c;
 }
 
@@ -371,29 +358,28 @@ static __inline__ int atomic64_add_negative(long i, atomic64_t *v)
  *
  * Atomically adds @i to @v and returns @i + @v
  */
-static __inline__ long atomic64_add_return(long i, atomic64_t *v)
+static inline long atomic64_add_return(long i, atomic64_t *v)
 {
 	long __i = i;
-	__asm__ __volatile__(
-		LOCK_PREFIX "xaddq %0, %1;"
-		:"+r" (i), "+m" (v->counter)
-		: : "memory");
+	asm volatile(LOCK_PREFIX "xaddq %0, %1;"
+		     : "+r" (i), "+m" (v->counter)
+		     : : "memory");
 	return i + __i;
 }
 
-static __inline__ long atomic64_sub_return(long i, atomic64_t *v)
+static inline long atomic64_sub_return(long i, atomic64_t *v)
 {
-	return atomic64_add_return(-i,v);
+	return atomic64_add_return(-i, v);
 }
 
-#define atomic64_inc_return(v)  (atomic64_add_return(1,v))
-#define atomic64_dec_return(v)  (atomic64_sub_return(1,v))
+#define atomic64_inc_return(v)  (atomic64_add_return(1, (v)))
+#define atomic64_dec_return(v)  (atomic64_sub_return(1, (v)))
 
-#define atomic64_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), old, new))
+#define atomic64_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), (old), (new)))
 #define atomic64_xchg(v, new) (xchg(&((v)->counter), new))
 
-#define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), old, new))
-#define atomic_xchg(v, new) (xchg(&((v)->counter), new))
+#define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), (old), (new)))
+#define atomic_xchg(v, new) (xchg(&((v)->counter), (new)))
 
 /**
  * atomic_add_unless - add unless the number is a given value
@@ -404,7 +390,7 @@ static __inline__ long atomic64_sub_return(long i, atomic64_t *v)
  * Atomically adds @a to @v, so long as it was not @u.
  * Returns non-zero if @v was not @u, and zero otherwise.
  */
-static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
+static inline int atomic_add_unless(atomic_t *v, int a, int u)
 {
 	int c, old;
 	c = atomic_read(v);
@@ -430,7 +416,7 @@ static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
  * Atomically adds @a to @v, so long as it was not @u.
  * Returns non-zero if @v was not @u, and zero otherwise.
  */
-static __inline__ int atomic64_add_unless(atomic64_t *v, long a, long u)
+static inline int atomic64_add_unless(atomic64_t *v, long a, long u)
 {
 	long c, old;
 	c = atomic64_read(v);
@@ -448,13 +434,14 @@ static __inline__ int atomic64_add_unless(atomic64_t *v, long a, long u)
 #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
 
 /* These are x86-specific, used by some header files */
-#define atomic_clear_mask(mask, addr) \
-__asm__ __volatile__(LOCK_PREFIX "andl %0,%1" \
-: : "r" (~(mask)),"m" (*addr) : "memory")
-
-#define atomic_set_mask(mask, addr) \
-__asm__ __volatile__(LOCK_PREFIX "orl %0,%1" \
-: : "r" ((unsigned)mask),"m" (*(addr)) : "memory")
+#define atomic_clear_mask(mask, addr)					\
+	asm volatile(LOCK_PREFIX "andl %0,%1"				\
+		     : : "r" (~(mask)), "m" (*(addr)) : "memory")
+
+#define atomic_set_mask(mask, addr)					\
+	asm volatile(LOCK_PREFIX "orl %0,%1"				\
+		     : : "r" ((unsigned)(mask)), "m" (*(addr))		\
+		     : "memory")
 
 /* Atomic operations are already serializing on x86 */
 #define smp_mb__before_atomic_dec()	barrier()
-- 
cgit v1.2.3


From fd591acd0f0b2382e8c0a839be9974e7c019bdb6 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:43 -0700
Subject: include/asm-x86/bitops_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/bitops_32.h | 40 +++++++++++++++++++++-------------------
 1 file changed, 21 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bitops_32.h b/include/asm-x86/bitops_32.h
index e4d75fcf9c03..2513a81f82aa 100644
--- a/include/asm-x86/bitops_32.h
+++ b/include/asm-x86/bitops_32.h
@@ -20,20 +20,22 @@ static inline int find_first_zero_bit(const unsigned long *addr, unsigned size)
 
 	if (!size)
 		return 0;
-	/* This looks at memory. Mark it volatile to tell gcc not to move it around */
-	__asm__ __volatile__(
-		"movl $-1,%%eax\n\t"
-		"xorl %%edx,%%edx\n\t"
-		"repe; scasl\n\t"
-		"je 1f\n\t"
-		"xorl -4(%%edi),%%eax\n\t"
-		"subl $4,%%edi\n\t"
-		"bsfl %%eax,%%edx\n"
-		"1:\tsubl %%ebx,%%edi\n\t"
-		"shll $3,%%edi\n\t"
-		"addl %%edi,%%edx"
-		:"=d" (res), "=&c" (d0), "=&D" (d1), "=&a" (d2)
-		:"1" ((size + 31) >> 5), "2" (addr), "b" (addr) : "memory");
+	/* This looks at memory.
+	 * Mark it volatile to tell gcc not to move it around
+	 */
+	asm volatile("movl $-1,%%eax\n\t"
+		     "xorl %%edx,%%edx\n\t"
+		     "repe; scasl\n\t"
+		     "je 1f\n\t"
+		     "xorl -4(%%edi),%%eax\n\t"
+		     "subl $4,%%edi\n\t"
+		     "bsfl %%eax,%%edx\n"
+		     "1:\tsubl %%ebx,%%edi\n\t"
+		     "shll $3,%%edi\n\t"
+		     "addl %%edi,%%edx"
+		     : "=d" (res), "=&c" (d0), "=&D" (d1), "=&a" (d2)
+		     : "1" ((size + 31) >> 5), "2" (addr),
+		       "b" (addr) : "memory");
 	return res;
 }
 
@@ -75,7 +77,7 @@ static inline unsigned find_first_bit(const unsigned long *addr, unsigned size)
 		unsigned long val = *addr++;
 		if (val)
 			return __ffs(val) + x;
-		x += (sizeof(*addr)<<3);
+		x += sizeof(*addr) << 3;
 	}
 	return x;
 }
@@ -152,10 +154,10 @@ static inline int fls(int x)
 
 #include <asm-generic/bitops/ext2-non-atomic.h>
 
-#define ext2_set_bit_atomic(lock, nr, addr) \
-		test_and_set_bit((nr), (unsigned long *)addr)
-#define ext2_clear_bit_atomic(lock, nr, addr) \
-		test_and_clear_bit((nr), (unsigned long *)addr)
+#define ext2_set_bit_atomic(lock, nr, addr)			\
+	test_and_set_bit((nr), (unsigned long *)(addr))
+#define ext2_clear_bit_atomic(lock, nr, addr)			\
+	test_and_clear_bit((nr), (unsigned long *)(addr))
 
 #include <asm-generic/bitops/minix.h>
 
-- 
cgit v1.2.3


From 49f74946f008add0b22723244976a32b365de06f Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:44 -0700
Subject: include/asm-x86/bitops_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/bitops_64.h | 40 ++++++++++++++++++++--------------------
 1 file changed, 20 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bitops_64.h b/include/asm-x86/bitops_64.h
index aaf15194d536..365f8207ea59 100644
--- a/include/asm-x86/bitops_64.h
+++ b/include/asm-x86/bitops_64.h
@@ -17,35 +17,35 @@ static inline long __scanbit(unsigned long val, unsigned long max)
 	return val;
 }
 
-#define find_first_bit(addr,size) \
-((__builtin_constant_p(size) && (size) <= BITS_PER_LONG ? \
-  (__scanbit(*(unsigned long *)addr,(size))) : \
-  find_first_bit(addr,size)))
-
 #define find_next_bit(addr,size,off) \
 ((__builtin_constant_p(size) && (size) <= BITS_PER_LONG ? 	  \
   ((off) + (__scanbit((*(unsigned long *)addr) >> (off),(size)-(off)))) : \
 	find_next_bit(addr,size,off)))
 
-#define find_first_zero_bit(addr,size) \
-((__builtin_constant_p(size) && (size) <= BITS_PER_LONG ? \
-  (__scanbit(~*(unsigned long *)addr,(size))) : \
-  	find_first_zero_bit(addr,size)))
-	
 #define find_next_zero_bit(addr,size,off) \
 ((__builtin_constant_p(size) && (size) <= BITS_PER_LONG ? 	  \
   ((off)+(__scanbit(~(((*(unsigned long *)addr)) >> (off)),(size)-(off)))) : \
 	find_next_zero_bit(addr,size,off)))
 
-static inline void set_bit_string(unsigned long *bitmap, unsigned long i, 
-				  int len) 
-{ 
-	unsigned long end = i + len; 
+#define find_first_bit(addr, size)					\
+	((__builtin_constant_p((size)) && (size) <= BITS_PER_LONG	\
+	  ? (__scanbit(*(unsigned long *)(addr), (size)))		\
+	  : find_first_bit((addr), (size))))
+
+#define find_first_zero_bit(addr, size)					\
+	((__builtin_constant_p((size)) && (size) <= BITS_PER_LONG	\
+	  ? (__scanbit(~*(unsigned long *)(addr), (size)))		\
+	  : find_first_zero_bit((addr), (size))))
+
+static inline void set_bit_string(unsigned long *bitmap, unsigned long i,
+				  int len)
+{
+	unsigned long end = i + len;
 	while (i < end) {
-		__set_bit(i, bitmap); 
+		__set_bit(i, bitmap);
 		i++;
 	}
-} 
+}
 
 /**
  * ffz - find first zero in word.
@@ -150,10 +150,10 @@ static inline int fls(int x)
 
 #include <asm-generic/bitops/ext2-non-atomic.h>
 
-#define ext2_set_bit_atomic(lock,nr,addr) \
-	        test_and_set_bit((nr),(unsigned long*)addr)
-#define ext2_clear_bit_atomic(lock,nr,addr) \
-	        test_and_clear_bit((nr),(unsigned long*)addr)
+#define ext2_set_bit_atomic(lock, nr, addr)			\
+	test_and_set_bit((nr), (unsigned long *)(addr))
+#define ext2_clear_bit_atomic(lock, nr, addr)			\
+	test_and_clear_bit((nr), (unsigned long *)(addr))
 
 #include <asm-generic/bitops/minix.h>
 
-- 
cgit v1.2.3


From 286275c90f148562b973b1e1f39f9689e6676dc4 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:45 -0700
Subject: include/asm-x86/bitops.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/bitops.h | 37 +++++++++++++------------------------
 1 file changed, 13 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bitops.h b/include/asm-x86/bitops.h
index 7a76555b676a..1ae7b270a1ef 100644
--- a/include/asm-x86/bitops.h
+++ b/include/asm-x86/bitops.h
@@ -23,13 +23,13 @@
 #if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 1)
 /* Technically wrong, but this avoids compilation errors on some gcc
    versions. */
-#define ADDR "=m" (*(volatile long *) addr)
-#define BIT_ADDR "=m" (((volatile int *) addr)[nr >> 5])
+#define ADDR "=m" (*(volatile long *)addr)
+#define BIT_ADDR "=m" (((volatile int *)addr)[nr >> 5])
 #else
 #define ADDR "+m" (*(volatile long *) addr)
-#define BIT_ADDR "+m" (((volatile int *) addr)[nr >> 5])
+#define BIT_ADDR "+m" (((volatile int *)addr)[nr >> 5])
 #endif
-#define BASE_ADDR "m" (*(volatile int *) addr)
+#define BASE_ADDR "m" (*(volatile int *)addr)
 
 /**
  * set_bit - Atomically set a bit in memory
@@ -48,9 +48,7 @@
  */
 static inline void set_bit(int nr, volatile void *addr)
 {
-	asm volatile(LOCK_PREFIX "bts %1,%0"
-		     : ADDR
-		     : "Ir" (nr) : "memory");
+	asm volatile(LOCK_PREFIX "bts %1,%0" : ADDR : "Ir" (nr) : "memory");
 }
 
 /**
@@ -82,8 +80,7 @@ static inline void __set_bit(int nr, volatile void *addr)
  */
 static inline void clear_bit(int nr, volatile void *addr)
 {
-	asm volatile(LOCK_PREFIX "btr %1,%2"
-		     : BIT_ADDR : "Ir" (nr), BASE_ADDR);
+	asm volatile(LOCK_PREFIX "btr %1,%2" : BIT_ADDR : "Ir" (nr), BASE_ADDR);
 }
 
 /*
@@ -151,8 +148,7 @@ static inline void __change_bit(int nr, volatile void *addr)
  */
 static inline void change_bit(int nr, volatile void *addr)
 {
-	asm volatile(LOCK_PREFIX "btc %1,%2"
-		     : BIT_ADDR : "Ir" (nr), BASE_ADDR);
+	asm volatile(LOCK_PREFIX "btc %1,%2" : BIT_ADDR : "Ir" (nr), BASE_ADDR);
 }
 
 /**
@@ -168,9 +164,7 @@ static inline int test_and_set_bit(int nr, volatile void *addr)
 	int oldbit;
 
 	asm volatile(LOCK_PREFIX "bts %2,%1\n\t"
-		     "sbb %0,%0"
-		     : "=r" (oldbit), ADDR
-		     : "Ir" (nr) : "memory");
+		     "sbb %0,%0" : "=r" (oldbit), ADDR : "Ir" (nr) : "memory");
 
 	return oldbit;
 }
@@ -202,8 +196,7 @@ static inline int __test_and_set_bit(int nr, volatile void *addr)
 
 	asm volatile("bts %2,%3\n\t"
 		     "sbb %0,%0"
-		     : "=r" (oldbit), BIT_ADDR
-		     : "Ir" (nr), BASE_ADDR);
+		     : "=r" (oldbit), BIT_ADDR : "Ir" (nr), BASE_ADDR);
 	return oldbit;
 }
 
@@ -221,8 +214,7 @@ static inline int test_and_clear_bit(int nr, volatile void *addr)
 
 	asm volatile(LOCK_PREFIX "btr %2,%1\n\t"
 		     "sbb %0,%0"
-		     : "=r" (oldbit), ADDR
-		     : "Ir" (nr) : "memory");
+		     : "=r" (oldbit), ADDR : "Ir" (nr) : "memory");
 
 	return oldbit;
 }
@@ -242,8 +234,7 @@ static inline int __test_and_clear_bit(int nr, volatile void *addr)
 
 	asm volatile("btr %2,%3\n\t"
 		     "sbb %0,%0"
-		     : "=r" (oldbit), BIT_ADDR
-		     : "Ir" (nr), BASE_ADDR);
+		     : "=r" (oldbit), BIT_ADDR : "Ir" (nr), BASE_ADDR);
 	return oldbit;
 }
 
@@ -254,8 +245,7 @@ static inline int __test_and_change_bit(int nr, volatile void *addr)
 
 	asm volatile("btc %2,%3\n\t"
 		     "sbb %0,%0"
-		     : "=r" (oldbit), BIT_ADDR
-		     : "Ir" (nr), BASE_ADDR);
+		     : "=r" (oldbit), BIT_ADDR : "Ir" (nr), BASE_ADDR);
 
 	return oldbit;
 }
@@ -274,8 +264,7 @@ static inline int test_and_change_bit(int nr, volatile void *addr)
 
 	asm volatile(LOCK_PREFIX "btc %2,%1\n\t"
 		     "sbb %0,%0"
-		     : "=r" (oldbit), ADDR
-		     : "Ir" (nr) : "memory");
+		     : "=r" (oldbit), ADDR : "Ir" (nr) : "memory");
 
 	return oldbit;
 }
-- 
cgit v1.2.3


From 86d8a08616ecbc510323bfca591816a5709c6e54 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:46 -0700
Subject: include/asm-x86/bug.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/bug.h | 34 +++++++++++++++++-----------------
 1 file changed, 17 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/bug.h b/include/asm-x86/bug.h
index 8d477a201392..b69aa64b82a4 100644
--- a/include/asm-x86/bug.h
+++ b/include/asm-x86/bug.h
@@ -12,25 +12,25 @@
 # define __BUG_C0	"2:\t.quad 1b, %c0\n"
 #endif
 
-#define BUG()								\
-	do {								\
-		asm volatile("1:\tud2\n"				\
-			     ".pushsection __bug_table,\"a\"\n"		\
-			     __BUG_C0					\
-			     "\t.word %c1, 0\n"				\
-			     "\t.org 2b+%c2\n"				\
-			     ".popsection"				\
-			     : : "i" (__FILE__), "i" (__LINE__),	\
-			     "i" (sizeof(struct bug_entry)));		\
-		for(;;) ;						\
-	} while(0)
+#define BUG()							\
+do {								\
+	asm volatile("1:\tud2\n"				\
+		     ".pushsection __bug_table,\"a\"\n"		\
+		     __BUG_C0					\
+		     "\t.word %c1, 0\n"				\
+		     "\t.org 2b+%c2\n"				\
+		     ".popsection"				\
+		     : : "i" (__FILE__), "i" (__LINE__),	\
+		     "i" (sizeof(struct bug_entry)));		\
+	for (;;) ;						\
+} while (0)
 
 #else
-#define BUG()								\
-	do {								\
-		asm volatile("ud2");					\
-		for(;;) ;						\
-	} while(0)
+#define BUG()							\
+do {								\
+	asm volatile("ud2");					\
+	for (;;) ;						\
+} while (0)
 #endif
 
 #endif /* !CONFIG_BUG */
-- 
cgit v1.2.3


From 346050952cac11b25a98c7e1743412b416827314 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:47 -0700
Subject: include/asm-x86/byteorder.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/byteorder.h | 39 ++++++++++++++++++++++++---------------
 1 file changed, 24 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/byteorder.h b/include/asm-x86/byteorder.h
index fe2f2e5d51ba..e02ae2d89acf 100644
--- a/include/asm-x86/byteorder.h
+++ b/include/asm-x86/byteorder.h
@@ -8,50 +8,59 @@
 
 #ifdef __i386__
 
-static __inline__ __attribute_const__ __u32 ___arch__swab32(__u32 x)
+static inline __attribute_const__ __u32 ___arch__swab32(__u32 x)
 {
 #ifdef CONFIG_X86_BSWAP
-	__asm__("bswap %0" : "=r" (x) : "0" (x));
+	asm("bswap %0" : "=r" (x) : "0" (x));
 #else
-	__asm__("xchgb %b0,%h0\n\t"	/* swap lower bytes	*/
-		"rorl $16,%0\n\t"	/* swap words		*/
-		"xchgb %b0,%h0"		/* swap higher bytes	*/
-		:"=q" (x)
-		: "0" (x));
+	asm("xchgb %b0,%h0\n\t"	/* swap lower bytes	*/
+	    "rorl $16,%0\n\t"	/* swap words		*/
+	    "xchgb %b0,%h0"	/* swap higher bytes	*/
+	    : "=q" (x)
+	    : "0" (x));
 #endif
 	return x;
 }
 
-static __inline__ __attribute_const__ __u64 ___arch__swab64(__u64 val)
+static inline __attribute_const__ __u64 ___arch__swab64(__u64 val)
 {
 	union {
-		struct { __u32 a,b; } s;
+		struct {
+			__u32 a;
+			__u32 b;
+		} s;
 		__u64 u;
 	} v;
 	v.u = val;
 #ifdef CONFIG_X86_BSWAP
-	__asm__("bswapl %0 ; bswapl %1 ; xchgl %0,%1"
+	asm("bswapl %0 ; bswapl %1 ; xchgl %0,%1"
 	    : "=r" (v.s.a), "=r" (v.s.b)
 	    : "0" (v.s.a), "1" (v.s.b));
 #else
 	v.s.a = ___arch__swab32(v.s.a);
 	v.s.b = ___arch__swab32(v.s.b);
-	__asm__("xchgl %0,%1" : "=r" (v.s.a), "=r" (v.s.b) : "0" (v.s.a), "1" (v.s.b));
+	asm("xchgl %0,%1"
+	    : "=r" (v.s.a), "=r" (v.s.b)
+	    : "0" (v.s.a), "1" (v.s.b));
 #endif
 	return v.u;
 }
 
 #else /* __i386__ */
 
-static __inline__ __attribute_const__ __u64 ___arch__swab64(__u64 x)
+static inline __attribute_const__ __u64 ___arch__swab64(__u64 x)
 {
-	__asm__("bswapq %0" : "=r" (x) : "0" (x));
+	asm("bswapq %0"
+	    : "=r" (x)
+	    : "0" (x));
 	return x;
 }
 
-static __inline__ __attribute_const__ __u32 ___arch__swab32(__u32 x)
+static inline __attribute_const__ __u32 ___arch__swab32(__u32 x)
 {
-	__asm__("bswapl %0" : "=r" (x) : "0" (x));
+	asm("bswapl %0"
+	    : "=r" (x)
+	    : "0" (x));
 	return x;
 }
 
-- 
cgit v1.2.3


From 3f61b19a9f60a0469a19113be8bb3b3623006de2 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:48 -0700
Subject: include/asm-x86/cacheflush.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/cacheflush.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 90437d3f7610..7ab5b520b7bd 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -14,18 +14,18 @@
 #define flush_dcache_mmap_lock(mapping)		do { } while (0)
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 #define flush_icache_range(start, end)		do { } while (0)
-#define flush_icache_page(vma,pg)		do { } while (0)
-#define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
+#define flush_icache_page(vma, pg)		do { } while (0)
+#define flush_icache_user_range(vma, pg, adr, len)	do { } while (0)
 #define flush_cache_vmap(start, end)		do { } while (0)
 #define flush_cache_vunmap(start, end)		do { } while (0)
 
-#define copy_to_user_page(vma, page, vaddr, dst, src, len) \
-	memcpy(dst, src, len)
-#define copy_from_user_page(vma, page, vaddr, dst, src, len) \
-	memcpy(dst, src, len)
+#define copy_to_user_page(vma, page, vaddr, dst, src, len)	\
+	memcpy((dst), (src), (len))
+#define copy_from_user_page(vma, page, vaddr, dst, src, len)	\
+	memcpy((dst), (src), (len))
 
 int __deprecated_for_modules change_page_attr(struct page *page, int numpages,
-								pgprot_t prot);
+					      pgprot_t prot);
 
 int set_pages_uc(struct page *page, int numpages);
 int set_pages_wb(struct page *page, int numpages);
-- 
cgit v1.2.3


From 0883e91ae209f4ada4db9b383026df77351c1320 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:49 -0700
Subject: include/asm-x86/checksum_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/checksum_32.h | 152 +++++++++++++++++++++---------------------
 1 file changed, 75 insertions(+), 77 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/checksum_32.h b/include/asm-x86/checksum_32.h
index 75194abbe8ee..52bbb0d8c4c1 100644
--- a/include/asm-x86/checksum_32.h
+++ b/include/asm-x86/checksum_32.h
@@ -28,7 +28,8 @@ asmlinkage __wsum csum_partial(const void *buff, int len, __wsum sum);
  */
 
 asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst,
-						  int len, __wsum sum, int *src_err_ptr, int *dst_err_ptr);
+					    int len, __wsum sum,
+					    int *src_err_ptr, int *dst_err_ptr);
 
 /*
  *	Note: when you get a NULL pointer exception here this means someone
@@ -37,20 +38,20 @@ asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst,
  *	If you use these functions directly please don't forget the
  *	access_ok().
  */
-static __inline__
-__wsum csum_partial_copy_nocheck (const void *src, void *dst,
-					int len, __wsum sum)
+static inline __wsum csum_partial_copy_nocheck(const void *src, void *dst,
+					       int len, __wsum sum)
 {
-	return csum_partial_copy_generic ( src, dst, len, sum, NULL, NULL);
+	return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL);
 }
 
-static __inline__
-__wsum csum_partial_copy_from_user(const void __user *src, void *dst,
-						int len, __wsum sum, int *err_ptr)
+static inline __wsum csum_partial_copy_from_user(const void __user *src,
+						 void *dst,
+						 int len, __wsum sum,
+						 int *err_ptr)
 {
 	might_sleep();
 	return csum_partial_copy_generic((__force void *)src, dst,
-					len, sum, err_ptr, NULL);
+					 len, sum, err_ptr, NULL);
 }
 
 /*
@@ -64,30 +65,29 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 {
 	unsigned int sum;
 
-	__asm__ __volatile__(
-	    "movl (%1), %0	;\n"
-	    "subl $4, %2	;\n"
-	    "jbe 2f		;\n"
-	    "addl 4(%1), %0	;\n"
-	    "adcl 8(%1), %0	;\n"
-	    "adcl 12(%1), %0	;\n"
-"1:	    adcl 16(%1), %0	;\n"
-	    "lea 4(%1), %1	;\n"
-	    "decl %2		;\n"
-	    "jne 1b		;\n"
-	    "adcl $0, %0	;\n"
-	    "movl %0, %2	;\n"
-	    "shrl $16, %0	;\n"
-	    "addw %w2, %w0	;\n"
-	    "adcl $0, %0	;\n"
-	    "notl %0		;\n"
-"2:				;\n"
+	asm volatile("movl (%1), %0	;\n"
+		     "subl $4, %2	;\n"
+		     "jbe 2f		;\n"
+		     "addl 4(%1), %0	;\n"
+		     "adcl 8(%1), %0	;\n"
+		     "adcl 12(%1), %0;\n"
+		     "1:	adcl 16(%1), %0	;\n"
+		     "lea 4(%1), %1	;\n"
+		     "decl %2	;\n"
+		     "jne 1b		;\n"
+		     "adcl $0, %0	;\n"
+		     "movl %0, %2	;\n"
+		     "shrl $16, %0	;\n"
+		     "addw %w2, %w0	;\n"
+		     "adcl $0, %0	;\n"
+		     "notl %0	;\n"
+		     "2:		;\n"
 	/* Since the input registers which are loaded with iph and ihl
 	   are modified, we must also specify them as outputs, or gcc
 	   will assume they contain their original values. */
-	: "=r" (sum), "=r" (iph), "=r" (ihl)
-	: "1" (iph), "2" (ihl)
-	: "memory");
+		     : "=r" (sum), "=r" (iph), "=r" (ihl)
+		     : "1" (iph), "2" (ihl)
+		     : "memory");
 	return (__force __sum16)sum;
 }
 
@@ -97,29 +97,27 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 
 static inline __sum16 csum_fold(__wsum sum)
 {
-	__asm__(
-		"addl %1, %0		;\n"
-		"adcl $0xffff, %0	;\n"
-		: "=r" (sum)
-		: "r" ((__force u32)sum << 16),
-		  "0" ((__force u32)sum & 0xffff0000)
-	);
+	asm("addl %1, %0		;\n"
+	    "adcl $0xffff, %0	;\n"
+	    : "=r" (sum)
+	    : "r" ((__force u32)sum << 16),
+	      "0" ((__force u32)sum & 0xffff0000));
 	return (__force __sum16)(~(__force u32)sum >> 16);
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					   unsigned short len,
-					   unsigned short proto,
-					   __wsum sum)
+					unsigned short len,
+					unsigned short proto,
+					__wsum sum)
 {
-    __asm__(
-	"addl %1, %0	;\n"
-	"adcl %2, %0	;\n"
-	"adcl %3, %0	;\n"
-	"adcl $0, %0	;\n"
-	: "=r" (sum)
-	: "g" (daddr), "g"(saddr), "g"((len + proto) << 8), "0"(sum));
-    return sum;
+	asm("addl %1, %0	;\n"
+	    "adcl %2, %0	;\n"
+	    "adcl %3, %0	;\n"
+	    "adcl $0, %0	;\n"
+	    : "=r" (sum)
+	    : "g" (daddr), "g"(saddr),
+	      "g" ((len + proto) << 8), "0" (sum));
+	return sum;
 }
 
 /*
@@ -127,11 +125,11 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					unsigned short len,
+					unsigned short proto,
+					__wsum sum)
 {
-	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
 
 /*
@@ -141,30 +139,29 @@ static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
 
 static inline __sum16 ip_compute_csum(const void *buff, int len)
 {
-    return csum_fold (csum_partial(buff, len, 0));
+    return csum_fold(csum_partial(buff, len, 0));
 }
 
 #define _HAVE_ARCH_IPV6_CSUM
-static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
-					  const struct in6_addr *daddr,
-					  __u32 len, unsigned short proto,
-					  __wsum sum)
+static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
+				      const struct in6_addr *daddr,
+				      __u32 len, unsigned short proto,
+				      __wsum sum)
 {
-	__asm__(
-		"addl 0(%1), %0		;\n"
-		"adcl 4(%1), %0		;\n"
-		"adcl 8(%1), %0		;\n"
-		"adcl 12(%1), %0	;\n"
-		"adcl 0(%2), %0		;\n"
-		"adcl 4(%2), %0		;\n"
-		"adcl 8(%2), %0		;\n"
-		"adcl 12(%2), %0	;\n"
-		"adcl %3, %0		;\n"
-		"adcl %4, %0		;\n"
-		"adcl $0, %0		;\n"
-		: "=&r" (sum)
-		: "r" (saddr), "r" (daddr),
-		  "r"(htonl(len)), "r"(htonl(proto)), "0"(sum));
+	asm("addl 0(%1), %0	;\n"
+	    "adcl 4(%1), %0	;\n"
+	    "adcl 8(%1), %0	;\n"
+	    "adcl 12(%1), %0	;\n"
+	    "adcl 0(%2), %0	;\n"
+	    "adcl 4(%2), %0	;\n"
+	    "adcl 8(%2), %0	;\n"
+	    "adcl 12(%2), %0	;\n"
+	    "adcl %3, %0	;\n"
+	    "adcl %4, %0	;\n"
+	    "adcl $0, %0	;\n"
+	    : "=&r" (sum)
+	    : "r" (saddr), "r" (daddr),
+	      "r" (htonl(len)), "r" (htonl(proto)), "0" (sum));
 
 	return csum_fold(sum);
 }
@@ -173,14 +170,15 @@ static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
  *	Copy and checksum to user
  */
 #define HAVE_CSUM_COPY_USER
-static __inline__ __wsum csum_and_copy_to_user(const void *src,
-						     void __user *dst,
-						     int len, __wsum sum,
-						     int *err_ptr)
+static inline __wsum csum_and_copy_to_user(const void *src,
+					   void __user *dst,
+					   int len, __wsum sum,
+					   int *err_ptr)
 {
 	might_sleep();
 	if (access_ok(VERIFY_WRITE, dst, len))
-		return csum_partial_copy_generic(src, (__force void *)dst, len, sum, NULL, err_ptr);
+		return csum_partial_copy_generic(src, (__force void *)dst,
+						 len, sum, NULL, err_ptr);
 
 	if (len)
 		*err_ptr = -EFAULT;
-- 
cgit v1.2.3


From 3d3c6e10036dcbbe9fe7d69911f5638faecfbaeb Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:50 -0700
Subject: include/asm-x86/checksum_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/checksum_64.h | 118 ++++++++++++++++++++----------------------
 1 file changed, 57 insertions(+), 61 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/checksum_64.h b/include/asm-x86/checksum_64.h
index e5f79997decc..8bd861cc5267 100644
--- a/include/asm-x86/checksum_64.h
+++ b/include/asm-x86/checksum_64.h
@@ -1,33 +1,31 @@
 #ifndef _X86_64_CHECKSUM_H
 #define _X86_64_CHECKSUM_H
 
-/* 
- * Checksums for x86-64 
- * Copyright 2002 by Andi Kleen, SuSE Labs 
+/*
+ * Checksums for x86-64
+ * Copyright 2002 by Andi Kleen, SuSE Labs
  * with some code from asm-x86/checksum.h
- */ 
+ */
 
 #include <linux/compiler.h>
 #include <asm/uaccess.h>
 #include <asm/byteorder.h>
 
-/** 
+/**
  * csum_fold - Fold and invert a 32bit checksum.
  * sum: 32bit unfolded sum
- * 
+ *
  * Fold a 32bit running checksum to 16bit and invert it. This is usually
  * the last step before putting a checksum into a packet.
  * Make sure not to mix with 64bit checksums.
  */
 static inline __sum16 csum_fold(__wsum sum)
 {
-	__asm__(
-		"  addl %1,%0\n"
-		"  adcl $0xffff,%0"
-		: "=r" (sum)
-		: "r" ((__force u32)sum << 16),
-		  "0" ((__force u32)sum & 0xffff0000)
-	);
+	asm("  addl %1,%0\n"
+	    "  adcl $0xffff,%0"
+	    : "=r" (sum)
+	    : "r" ((__force u32)sum << 16),
+	      "0" ((__force u32)sum & 0xffff0000));
 	return (__force __sum16)(~(__force u32)sum >> 16);
 }
 
@@ -43,46 +41,46 @@ static inline __sum16 csum_fold(__wsum sum)
  * ip_fast_csum - Compute the IPv4 header checksum efficiently.
  * iph: ipv4 header
  * ihl: length of header / 4
- */ 
+ */
 static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 {
 	unsigned int sum;
 
-	asm(	"  movl (%1), %0\n"
-		"  subl $4, %2\n"
-		"  jbe 2f\n"
-		"  addl 4(%1), %0\n"
-		"  adcl 8(%1), %0\n"
-		"  adcl 12(%1), %0\n"
-		"1: adcl 16(%1), %0\n"
-		"  lea 4(%1), %1\n"
-		"  decl %2\n"
-		"  jne	1b\n"
-		"  adcl $0, %0\n"
-		"  movl %0, %2\n"
-		"  shrl $16, %0\n"
-		"  addw %w2, %w0\n"
-		"  adcl $0, %0\n"
-		"  notl %0\n"
-		"2:"
+	asm("  movl (%1), %0\n"
+	    "  subl $4, %2\n"
+	    "  jbe 2f\n"
+	    "  addl 4(%1), %0\n"
+	    "  adcl 8(%1), %0\n"
+	    "  adcl 12(%1), %0\n"
+	    "1: adcl 16(%1), %0\n"
+	    "  lea 4(%1), %1\n"
+	    "  decl %2\n"
+	    "  jne	1b\n"
+	    "  adcl $0, %0\n"
+	    "  movl %0, %2\n"
+	    "  shrl $16, %0\n"
+	    "  addw %w2, %w0\n"
+	    "  adcl $0, %0\n"
+	    "  notl %0\n"
+	    "2:"
 	/* Since the input registers which are loaded with iph and ihl
 	   are modified, we must also specify them as outputs, or gcc
 	   will assume they contain their original values. */
-	: "=r" (sum), "=r" (iph), "=r" (ihl)
-	: "1" (iph), "2" (ihl)
-	: "memory");
+	    : "=r" (sum), "=r" (iph), "=r" (ihl)
+	    : "1" (iph), "2" (ihl)
+	    : "memory");
 	return (__force __sum16)sum;
 }
 
-/** 
+/**
  * csum_tcpup_nofold - Compute an IPv4 pseudo header checksum.
  * @saddr: source address
  * @daddr: destination address
  * @len: length of packet
  * @proto: ip protocol of packet
- * @sum: initial sum to be added in (32bit unfolded) 
- * 
- * Returns the pseudo header checksum the input data. Result is 
+ * @sum: initial sum to be added in (32bit unfolded)
+ *
+ * Returns the pseudo header checksum the input data. Result is
  * 32bit unfolded.
  */
 static inline __wsum
@@ -93,32 +91,32 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
 	    "  adcl %2, %0\n"
 	    "  adcl %3, %0\n"
 	    "  adcl $0, %0\n"
-		: "=r" (sum)
+	    : "=r" (sum)
 	    : "g" (daddr), "g" (saddr),
 	      "g" ((len + proto)<<8), "0" (sum));
-    return sum;
+	return sum;
 }
 
 
-/** 
+/**
  * csum_tcpup_magic - Compute an IPv4 pseudo header checksum.
  * @saddr: source address
  * @daddr: destination address
  * @len: length of packet
  * @proto: ip protocol of packet
- * @sum: initial sum to be added in (32bit unfolded) 
- * 
+ * @sum: initial sum to be added in (32bit unfolded)
+ *
  * Returns the 16bit pseudo header checksum the input data already
  * complemented and ready to be filled in.
  */
-static inline __sum16
-csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-		  unsigned short len, unsigned short proto, __wsum sum)
+static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+					unsigned short len,
+					unsigned short proto, __wsum sum)
 {
-	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
 
-/** 
+/**
  * csum_partial - Compute an internet checksum.
  * @buff: buffer to be checksummed
  * @len: length of buffer.
@@ -127,7 +125,7 @@ csum_tcpudp_magic(__be32 saddr, __be32 daddr,
  * Returns the 32bit unfolded internet checksum of the buffer.
  * Before filling it in it needs to be csum_fold()'ed.
  * buff should be aligned to a 64bit boundary if possible.
- */ 
+ */
 extern __wsum csum_partial(const void *buff, int len, __wsum sum);
 
 #define  _HAVE_ARCH_COPY_AND_CSUM_FROM_USER 1
@@ -136,23 +134,22 @@ extern __wsum csum_partial(const void *buff, int len, __wsum sum);
 
 /* Do not call this directly. Use the wrappers below */
 extern __wsum csum_partial_copy_generic(const void *src, const void *dst,
-					       int len,
-					       __wsum sum,
-					       int *src_err_ptr, int *dst_err_ptr);
+					int len, __wsum sum,
+					int *src_err_ptr, int *dst_err_ptr);
 
 
 extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst,
-				       int len, __wsum isum, int *errp);
+					  int len, __wsum isum, int *errp);
 extern __wsum csum_partial_copy_to_user(const void *src, void __user *dst,
-				      int len, __wsum isum, int *errp);
-extern __wsum csum_partial_copy_nocheck(const void *src, void *dst, int len,
-					      __wsum sum);
+					int len, __wsum isum, int *errp);
+extern __wsum csum_partial_copy_nocheck(const void *src, void *dst,
+					int len, __wsum sum);
 
 /* Old names. To be removed. */
 #define csum_and_copy_to_user csum_partial_copy_to_user
 #define csum_and_copy_from_user csum_partial_copy_from_user
 
-/** 
+/**
  * ip_compute_csum - Compute an 16bit IP checksum.
  * @buff: buffer address.
  * @len: length of buffer.
@@ -170,7 +167,7 @@ extern __sum16 ip_compute_csum(const void *buff, int len);
  * @proto: protocol of packet
  * @sum: initial sum (32bit unfolded) to be added in
  *
- * Computes an IPv6 pseudo header checksum. This sum is added the checksum 
+ * Computes an IPv6 pseudo header checksum. This sum is added the checksum
  * into UDP/TCP packets and contains some link layer information.
  * Returns the unfolded 32bit checksum.
  */
@@ -185,11 +182,10 @@ csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr,
 static inline unsigned add32_with_carry(unsigned a, unsigned b)
 {
 	asm("addl %2,%0\n\t"
-	    "adcl $0,%0" 
-	    : "=r" (a) 
+	    "adcl $0,%0"
+	    : "=r" (a)
 	    : "0" (a), "r" (b));
 	return a;
 }
 
 #endif
-
-- 
cgit v1.2.3


From 8121019cad7bfe61f8f626a85427aca66dfe0f1e Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:51 -0700
Subject: include/asm-x86/cmpxchg_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/cmpxchg_32.h | 253 ++++++++++++++++++++++---------------------
 1 file changed, 132 insertions(+), 121 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/cmpxchg_32.h b/include/asm-x86/cmpxchg_32.h
index 959fad00dff5..bf5a69d1329e 100644
--- a/include/asm-x86/cmpxchg_32.h
+++ b/include/asm-x86/cmpxchg_32.h
@@ -8,9 +8,12 @@
  *       you need to test for the feature in boot_cpu_data.
  */
 
-#define xchg(ptr,v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr))))
+#define xchg(ptr, v)							\
+	((__typeof__(*(ptr)))__xchg((unsigned long)(v), (ptr), sizeof(*(ptr))))
 
-struct __xchg_dummy { unsigned long a[100]; };
+struct __xchg_dummy {
+	unsigned long a[100];
+};
 #define __xg(x) ((struct __xchg_dummy *)(x))
 
 /*
@@ -27,72 +30,74 @@ struct __xchg_dummy { unsigned long a[100]; };
  * of the instruction set reference 24319102.pdf. We need
  * the reader side to see the coherent 64bit value.
  */
-static inline void __set_64bit (unsigned long long * ptr,
-		unsigned int low, unsigned int high)
+static inline void __set_64bit(unsigned long long *ptr,
+			       unsigned int low, unsigned int high)
 {
-	__asm__ __volatile__ (
-		"\n1:\t"
-		"movl (%0), %%eax\n\t"
-		"movl 4(%0), %%edx\n\t"
-		LOCK_PREFIX "cmpxchg8b (%0)\n\t"
-		"jnz 1b"
-		: /* no outputs */
-		:	"D"(ptr),
-			"b"(low),
-			"c"(high)
-		:	"ax","dx","memory");
+	asm volatile("\n1:\t"
+		     "movl (%0), %%eax\n\t"
+		     "movl 4(%0), %%edx\n\t"
+		     LOCK_PREFIX "cmpxchg8b (%0)\n\t"
+		     "jnz 1b"
+		     : /* no outputs */
+		     : "D"(ptr),
+		       "b"(low),
+		       "c"(high)
+		     : "ax", "dx", "memory");
 }
 
-static inline void __set_64bit_constant (unsigned long long *ptr,
-						 unsigned long long value)
+static inline void __set_64bit_constant(unsigned long long *ptr,
+					unsigned long long value)
 {
-	__set_64bit(ptr,(unsigned int)(value), (unsigned int)((value)>>32ULL));
+	__set_64bit(ptr, (unsigned int)value, (unsigned int)(value >> 32));
 }
-#define ll_low(x)	*(((unsigned int*)&(x))+0)
-#define ll_high(x)	*(((unsigned int*)&(x))+1)
 
-static inline void __set_64bit_var (unsigned long long *ptr,
-			 unsigned long long value)
+#define ll_low(x)	*(((unsigned int *)&(x)) + 0)
+#define ll_high(x)	*(((unsigned int *)&(x)) + 1)
+
+static inline void __set_64bit_var(unsigned long long *ptr,
+				   unsigned long long value)
 {
-	__set_64bit(ptr,ll_low(value), ll_high(value));
+	__set_64bit(ptr, ll_low(value), ll_high(value));
 }
 
-#define set_64bit(ptr,value) \
-(__builtin_constant_p(value) ? \
- __set_64bit_constant(ptr, value) : \
- __set_64bit_var(ptr, value) )
+#define set_64bit(ptr, value)			\
+	(__builtin_constant_p((value))		\
+	 ? __set_64bit_constant((ptr), (value))	\
+	 : __set_64bit_var((ptr), (value)))
 
-#define _set_64bit(ptr,value) \
-(__builtin_constant_p(value) ? \
- __set_64bit(ptr, (unsigned int)(value), (unsigned int)((value)>>32ULL) ) : \
- __set_64bit(ptr, ll_low(value), ll_high(value)) )
+#define _set_64bit(ptr, value)						\
+	(__builtin_constant_p(value)					\
+	 ? __set_64bit(ptr, (unsigned int)(value),			\
+		       (unsigned int)((value) >> 32))			\
+	 : __set_64bit(ptr, ll_low((value)), ll_high((value))))
 
 /*
  * Note: no "lock" prefix even on SMP: xchg always implies lock anyway
  * Note 2: xchg has side effect, so that attribute volatile is necessary,
  *	  but generally the primitive is invalid, *ptr is output argument. --ANK
  */
-static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
+static inline unsigned long __xchg(unsigned long x, volatile void *ptr,
+				   int size)
 {
 	switch (size) {
-		case 1:
-			__asm__ __volatile__("xchgb %b0,%1"
-				:"=q" (x)
-				:"m" (*__xg(ptr)), "0" (x)
-				:"memory");
-			break;
-		case 2:
-			__asm__ __volatile__("xchgw %w0,%1"
-				:"=r" (x)
-				:"m" (*__xg(ptr)), "0" (x)
-				:"memory");
-			break;
-		case 4:
-			__asm__ __volatile__("xchgl %0,%1"
-				:"=r" (x)
-				:"m" (*__xg(ptr)), "0" (x)
-				:"memory");
-			break;
+	case 1:
+		asm volatile("xchgb %b0,%1"
+			     : "=q" (x)
+			     : "m" (*__xg(ptr)), "0" (x)
+			     : "memory");
+		break;
+	case 2:
+		asm volatile("xchgw %w0,%1"
+			     : "=r" (x)
+			     : "m" (*__xg(ptr)), "0" (x)
+			     : "memory");
+		break;
+	case 4:
+		asm volatile("xchgl %0,%1"
+			     : "=r" (x)
+			     : "m" (*__xg(ptr)), "0" (x)
+			     : "memory");
+		break;
 	}
 	return x;
 }
@@ -105,24 +110,27 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz
 
 #ifdef CONFIG_X86_CMPXCHG
 #define __HAVE_ARCH_CMPXCHG 1
-#define cmpxchg(ptr, o, n)						     \
-	((__typeof__(*(ptr)))__cmpxchg((ptr), (unsigned long)(o),	     \
-					(unsigned long)(n), sizeof(*(ptr))))
-#define sync_cmpxchg(ptr, o, n)						     \
-	((__typeof__(*(ptr)))__sync_cmpxchg((ptr), (unsigned long)(o),	     \
-					(unsigned long)(n), sizeof(*(ptr))))
-#define cmpxchg_local(ptr, o, n)					     \
-	((__typeof__(*(ptr)))__cmpxchg_local((ptr), (unsigned long)(o),	     \
-					(unsigned long)(n), sizeof(*(ptr))))
+#define cmpxchg(ptr, o, n)						\
+	((__typeof__(*(ptr)))__cmpxchg((ptr), (unsigned long)(o),	\
+				       (unsigned long)(n),		\
+				       sizeof(*(ptr))))
+#define sync_cmpxchg(ptr, o, n)						\
+	((__typeof__(*(ptr)))__sync_cmpxchg((ptr), (unsigned long)(o),	\
+					    (unsigned long)(n),		\
+					    sizeof(*(ptr))))
+#define cmpxchg_local(ptr, o, n)					\
+	((__typeof__(*(ptr)))__cmpxchg_local((ptr), (unsigned long)(o),	\
+					     (unsigned long)(n),	\
+					     sizeof(*(ptr))))
 #endif
 
 #ifdef CONFIG_X86_CMPXCHG64
-#define cmpxchg64(ptr, o, n)						      \
-	((__typeof__(*(ptr)))__cmpxchg64((ptr), (unsigned long long)(o),      \
-					(unsigned long long)(n)))
-#define cmpxchg64_local(ptr, o, n)					      \
-	((__typeof__(*(ptr)))__cmpxchg64_local((ptr), (unsigned long long)(o),\
-					(unsigned long long)(n)))
+#define cmpxchg64(ptr, o, n)						\
+	((__typeof__(*(ptr)))__cmpxchg64((ptr), (unsigned long long)(o), \
+					 (unsigned long long)(n)))
+#define cmpxchg64_local(ptr, o, n)					\
+	((__typeof__(*(ptr)))__cmpxchg64_local((ptr), (unsigned long long)(o), \
+					       (unsigned long long)(n)))
 #endif
 
 static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
@@ -131,22 +139,22 @@ static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		__asm__ __volatile__(LOCK_PREFIX "cmpxchgb %b1,%2"
-				     : "=a"(prev)
-				     : "q"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile(LOCK_PREFIX "cmpxchgb %b1,%2"
+			     : "=a"(prev)
+			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 2:
-		__asm__ __volatile__(LOCK_PREFIX "cmpxchgw %w1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile(LOCK_PREFIX "cmpxchgw %w1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 4:
-		__asm__ __volatile__(LOCK_PREFIX "cmpxchgl %1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile(LOCK_PREFIX "cmpxchgl %1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	}
 	return old;
@@ -158,85 +166,88 @@ static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
  * isn't.
  */
 static inline unsigned long __sync_cmpxchg(volatile void *ptr,
-					    unsigned long old,
-					    unsigned long new, int size)
+					   unsigned long old,
+					   unsigned long new, int size)
 {
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		__asm__ __volatile__("lock; cmpxchgb %b1,%2"
-				     : "=a"(prev)
-				     : "q"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("lock; cmpxchgb %b1,%2"
+			     : "=a"(prev)
+			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 2:
-		__asm__ __volatile__("lock; cmpxchgw %w1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("lock; cmpxchgw %w1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 4:
-		__asm__ __volatile__("lock; cmpxchgl %1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("lock; cmpxchgl %1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	}
 	return old;
 }
 
 static inline unsigned long __cmpxchg_local(volatile void *ptr,
-			unsigned long old, unsigned long new, int size)
+					    unsigned long old,
+					    unsigned long new, int size)
 {
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		__asm__ __volatile__("cmpxchgb %b1,%2"
-				     : "=a"(prev)
-				     : "q"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("cmpxchgb %b1,%2"
+			     : "=a"(prev)
+			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 2:
-		__asm__ __volatile__("cmpxchgw %w1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("cmpxchgw %w1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 4:
-		__asm__ __volatile__("cmpxchgl %1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("cmpxchgl %1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	}
 	return old;
 }
 
 static inline unsigned long long __cmpxchg64(volatile void *ptr,
-			unsigned long long old, unsigned long long new)
+					     unsigned long long old,
+					     unsigned long long new)
 {
 	unsigned long long prev;
-	__asm__ __volatile__(LOCK_PREFIX "cmpxchg8b %3"
-			     : "=A"(prev)
-			     : "b"((unsigned long)new),
-			       "c"((unsigned long)(new >> 32)),
-			       "m"(*__xg(ptr)),
-			       "0"(old)
-			     : "memory");
+	asm volatile(LOCK_PREFIX "cmpxchg8b %3"
+		     : "=A"(prev)
+		     : "b"((unsigned long)new),
+		       "c"((unsigned long)(new >> 32)),
+		       "m"(*__xg(ptr)),
+		       "0"(old)
+		     : "memory");
 	return prev;
 }
 
 static inline unsigned long long __cmpxchg64_local(volatile void *ptr,
-			unsigned long long old, unsigned long long new)
+						   unsigned long long old,
+						   unsigned long long new)
 {
 	unsigned long long prev;
-	__asm__ __volatile__("cmpxchg8b %3"
-			     : "=A"(prev)
-			     : "b"((unsigned long)new),
-			       "c"((unsigned long)(new >> 32)),
-			       "m"(*__xg(ptr)),
-			       "0"(old)
-			     : "memory");
+	asm volatile("cmpxchg8b %3"
+		     : "=A"(prev)
+		     : "b"((unsigned long)new),
+		       "c"((unsigned long)(new >> 32)),
+		       "m"(*__xg(ptr)),
+		       "0"(old)
+		     : "memory");
 	return prev;
 }
 
@@ -252,7 +263,7 @@ extern unsigned long cmpxchg_386_u16(volatile void *, u16, u16);
 extern unsigned long cmpxchg_386_u32(volatile void *, u32, u32);
 
 static inline unsigned long cmpxchg_386(volatile void *ptr, unsigned long old,
-				      unsigned long new, int size)
+					unsigned long new, int size)
 {
 	switch (size) {
 	case 1:
-- 
cgit v1.2.3


From e52da357a15db9e12b96b4e40dffe6b9e54bb976 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:52 -0700
Subject: include/asm-x86/cmpxchg_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/cmpxchg_64.h | 134 ++++++++++++++++++++++---------------------
 1 file changed, 69 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/cmpxchg_64.h b/include/asm-x86/cmpxchg_64.h
index 56f5b41e071c..d9b26b9a28cf 100644
--- a/include/asm-x86/cmpxchg_64.h
+++ b/include/asm-x86/cmpxchg_64.h
@@ -3,7 +3,8 @@
 
 #include <asm/alternative.h> /* Provides LOCK_PREFIX */
 
-#define xchg(ptr,v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr))))
+#define xchg(ptr, v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v), \
+						 (ptr), sizeof(*(ptr))))
 
 #define __xg(x) ((volatile long *)(x))
 
@@ -19,33 +20,34 @@ static inline void set_64bit(volatile unsigned long *ptr, unsigned long val)
  * Note 2: xchg has side effect, so that attribute volatile is necessary,
  *	  but generally the primitive is invalid, *ptr is output argument. --ANK
  */
-static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
+static inline unsigned long __xchg(unsigned long x, volatile void *ptr,
+				   int size)
 {
 	switch (size) {
-		case 1:
-			__asm__ __volatile__("xchgb %b0,%1"
-				:"=q" (x)
-				:"m" (*__xg(ptr)), "0" (x)
-				:"memory");
-			break;
-		case 2:
-			__asm__ __volatile__("xchgw %w0,%1"
-				:"=r" (x)
-				:"m" (*__xg(ptr)), "0" (x)
-				:"memory");
-			break;
-		case 4:
-			__asm__ __volatile__("xchgl %k0,%1"
-				:"=r" (x)
-				:"m" (*__xg(ptr)), "0" (x)
-				:"memory");
-			break;
-		case 8:
-			__asm__ __volatile__("xchgq %0,%1"
-				:"=r" (x)
-				:"m" (*__xg(ptr)), "0" (x)
-				:"memory");
-			break;
+	case 1:
+		asm volatile("xchgb %b0,%1"
+			     : "=q" (x)
+			     : "m" (*__xg(ptr)), "0" (x)
+			     : "memory");
+		break;
+	case 2:
+		asm volatile("xchgw %w0,%1"
+			     : "=r" (x)
+			     : "m" (*__xg(ptr)), "0" (x)
+			     : "memory");
+		break;
+	case 4:
+		asm volatile("xchgl %k0,%1"
+			     : "=r" (x)
+			     : "m" (*__xg(ptr)), "0" (x)
+			     : "memory");
+		break;
+	case 8:
+		asm volatile("xchgq %0,%1"
+			     : "=r" (x)
+			     : "m" (*__xg(ptr)), "0" (x)
+			     : "memory");
+		break;
 	}
 	return x;
 }
@@ -64,61 +66,62 @@ static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		__asm__ __volatile__(LOCK_PREFIX "cmpxchgb %b1,%2"
-				     : "=a"(prev)
-				     : "q"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile(LOCK_PREFIX "cmpxchgb %b1,%2"
+			     : "=a"(prev)
+			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 2:
-		__asm__ __volatile__(LOCK_PREFIX "cmpxchgw %w1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile(LOCK_PREFIX "cmpxchgw %w1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 4:
-		__asm__ __volatile__(LOCK_PREFIX "cmpxchgl %k1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile(LOCK_PREFIX "cmpxchgl %k1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 8:
-		__asm__ __volatile__(LOCK_PREFIX "cmpxchgq %1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile(LOCK_PREFIX "cmpxchgq %1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	}
 	return old;
 }
 
 static inline unsigned long __cmpxchg_local(volatile void *ptr,
-			unsigned long old, unsigned long new, int size)
+					    unsigned long old,
+					    unsigned long new, int size)
 {
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		__asm__ __volatile__("cmpxchgb %b1,%2"
-				     : "=a"(prev)
-				     : "q"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("cmpxchgb %b1,%2"
+			     : "=a"(prev)
+			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 2:
-		__asm__ __volatile__("cmpxchgw %w1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("cmpxchgw %w1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 4:
-		__asm__ __volatile__("cmpxchgl %k1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("cmpxchgl %k1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	case 8:
-		__asm__ __volatile__("cmpxchgq %1,%2"
-				     : "=a"(prev)
-				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
-				     : "memory");
+		asm volatile("cmpxchgq %1,%2"
+			     : "=a"(prev)
+			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+			     : "memory");
 		return prev;
 	}
 	return old;
@@ -126,19 +129,20 @@ static inline unsigned long __cmpxchg_local(volatile void *ptr,
 
 #define cmpxchg(ptr, o, n)						\
 	((__typeof__(*(ptr)))__cmpxchg((ptr), (unsigned long)(o),	\
-					(unsigned long)(n), sizeof(*(ptr))))
+				       (unsigned long)(n), sizeof(*(ptr))))
 #define cmpxchg64(ptr, o, n)						\
-  ({									\
+({									\
 	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
 	cmpxchg((ptr), (o), (n));					\
-  })
+})
 #define cmpxchg_local(ptr, o, n)					\
 	((__typeof__(*(ptr)))__cmpxchg_local((ptr), (unsigned long)(o),	\
-					(unsigned long)(n), sizeof(*(ptr))))
+					     (unsigned long)(n),	\
+					     sizeof(*(ptr))))
 #define cmpxchg64_local(ptr, o, n)					\
-  ({									\
+({									\
 	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
 	cmpxchg_local((ptr), (o), (n));					\
-  })
+})
 
 #endif
-- 
cgit v1.2.3


From c96a6d41aa033c03b43e5a09448bddcc971797b1 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:53 -0700
Subject: include/asm-x86/compat.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/compat.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/compat.h b/include/asm-x86/compat.h
index d3e8f3e87ee8..1793ac317a30 100644
--- a/include/asm-x86/compat.h
+++ b/include/asm-x86/compat.h
@@ -204,7 +204,7 @@ static inline compat_uptr_t ptr_to_compat(void __user *uptr)
 	return (u32)(unsigned long)uptr;
 }
 
-static __inline__ void __user *compat_alloc_user_space(long len)
+static inline void __user *compat_alloc_user_space(long len)
 {
 	struct pt_regs *regs = task_pt_regs(current);
 	return (void __user *)regs->sp - len;
-- 
cgit v1.2.3


From 83ecfdd0a0a065e55172889dc660c5db3c14633a Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:55 -0700
Subject: include/asm-x86/current_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/current_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/current_32.h b/include/asm-x86/current_32.h
index d35248539912..5af9bdb97a16 100644
--- a/include/asm-x86/current_32.h
+++ b/include/asm-x86/current_32.h
@@ -11,7 +11,7 @@ static __always_inline struct task_struct *get_current(void)
 {
 	return x86_read_percpu(current_task);
 }
- 
+
 #define current get_current()
 
 #endif /* !(_I386_CURRENT_H) */
-- 
cgit v1.2.3


From 2cade899df7d9cc9e977a24898b0bd5a6f4a8dfd Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:56 -0700
Subject: include/asm-x86/current_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/current_64.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/current_64.h b/include/asm-x86/current_64.h
index bc8adecee66d..2d368ede2fc1 100644
--- a/include/asm-x86/current_64.h
+++ b/include/asm-x86/current_64.h
@@ -1,23 +1,23 @@
 #ifndef _X86_64_CURRENT_H
 #define _X86_64_CURRENT_H
 
-#if !defined(__ASSEMBLY__) 
+#if !defined(__ASSEMBLY__)
 struct task_struct;
 
 #include <asm/pda.h>
 
-static inline struct task_struct *get_current(void) 
-{ 
-	struct task_struct *t = read_pda(pcurrent); 
+static inline struct task_struct *get_current(void)
+{
+	struct task_struct *t = read_pda(pcurrent);
 	return t;
-} 
+}
 
 #define current get_current()
 
 #else
 
 #ifndef ASM_OFFSET_H
-#include <asm/asm-offsets.h> 
+#include <asm/asm-offsets.h>
 #endif
 
 #define GET_CURRENT(reg) movq %gs:(pda_pcurrent),reg
-- 
cgit v1.2.3


From b77619001e3acd228209bfdd864734cfdddb3d92 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:57 -0700
Subject: include/asm-x86/desc_defs.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/desc_defs.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc_defs.h b/include/asm-x86/desc_defs.h
index e33f078b3e54..eccb4ea1f918 100644
--- a/include/asm-x86/desc_defs.h
+++ b/include/asm-x86/desc_defs.h
@@ -18,17 +18,19 @@
  * incrementally. We keep the signature as a struct, rather than an union,
  * so we can get rid of it transparently in the future -- glommer
  */
-// 8 byte segment descriptor
+/* 8 byte segment descriptor */
 struct desc_struct {
 	union {
-		struct { unsigned int a, b; };
+		struct {
+			unsigned int a;
+			unsigned int b;
+		};
 		struct {
 			u16 limit0;
 			u16 base0;
 			unsigned base1: 8, type: 4, s: 1, dpl: 2, p: 1;
 			unsigned limit: 4, avl: 1, l: 1, d: 1, g: 1, base2: 8;
 		};
-
 	};
 } __attribute__((packed));
 
@@ -39,7 +41,7 @@ enum {
 	GATE_TASK = 0x5,
 };
 
-// 16byte gate
+/* 16byte gate */
 struct gate_struct64 {
 	u16 offset_low;
 	u16 segment;
@@ -56,10 +58,10 @@ struct gate_struct64 {
 enum {
 	DESC_TSS = 0x9,
 	DESC_LDT = 0x2,
-	DESCTYPE_S =	0x10,	/* !system */
+	DESCTYPE_S = 0x10,	/* !system */
 };
 
-// LDT or TSS descriptor in the GDT. 16 bytes.
+/* LDT or TSS descriptor in the GDT. 16 bytes. */
 struct ldttss_desc64 {
 	u16 limit0;
 	u16 base0;
@@ -84,7 +86,6 @@ struct desc_ptr {
 	unsigned long address;
 } __attribute__((packed)) ;
 
-
 #endif /* !__ASSEMBLY__ */
 
 #endif
-- 
cgit v1.2.3


From c1773a167defdec919b1dbfc9100054c44fd0ff5 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:58 -0700
Subject: include/asm-x86/desc.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/desc.h | 61 +++++++++++++++++++++++++-------------------------
 1 file changed, 31 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index 5b6a05d3a771..268a012bcd79 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -62,8 +62,8 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 }
 
 static inline void pack_gate(gate_desc *gate, unsigned char type,
-       unsigned long base, unsigned dpl, unsigned flags, unsigned short seg)
-
+			     unsigned long base, unsigned dpl, unsigned flags,
+			     unsigned short seg)
 {
 	gate->a = (seg << 16) | (base & 0xffff);
 	gate->b = (base & 0xffff0000) |
@@ -84,22 +84,23 @@ static inline int desc_empty(const void *ptr)
 #define load_TR_desc() native_load_tr_desc()
 #define load_gdt(dtr) native_load_gdt(dtr)
 #define load_idt(dtr) native_load_idt(dtr)
-#define load_tr(tr) __asm__ __volatile("ltr %0"::"m" (tr))
-#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"m" (ldt))
+#define load_tr(tr) asm volatile("ltr %0"::"m" (tr))
+#define load_ldt(ldt) asm volatile("lldt %0"::"m" (ldt))
 
 #define store_gdt(dtr) native_store_gdt(dtr)
 #define store_idt(dtr) native_store_idt(dtr)
 #define store_tr(tr) (tr = native_store_tr())
-#define store_ldt(ldt) __asm__ ("sldt %0":"=m" (ldt))
+#define store_ldt(ldt) asm("sldt %0":"=m" (ldt))
 
 #define load_TLS(t, cpu) native_load_tls(t, cpu)
 #define set_ldt native_set_ldt
 
-#define write_ldt_entry(dt, entry, desc) \
-				native_write_ldt_entry(dt, entry, desc)
-#define write_gdt_entry(dt, entry, desc, type) \
-				native_write_gdt_entry(dt, entry, desc, type)
-#define write_idt_entry(dt, entry, g) native_write_idt_entry(dt, entry, g)
+#define write_ldt_entry(dt, entry, desc)	\
+	native_write_ldt_entry(dt, entry, desc)
+#define write_gdt_entry(dt, entry, desc, type)		\
+	native_write_gdt_entry(dt, entry, desc, type)
+#define write_idt_entry(dt, entry, g)		\
+	native_write_idt_entry(dt, entry, g)
 #endif
 
 static inline void native_write_idt_entry(gate_desc *idt, int entry,
@@ -138,8 +139,8 @@ static inline void pack_descriptor(struct desc_struct *desc, unsigned long base,
 {
 	desc->a = ((base & 0xffff) << 16) | (limit & 0xffff);
 	desc->b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
-		  (limit & 0x000f0000) | ((type & 0xff) << 8) |
-		  ((flags & 0xf) << 20);
+		(limit & 0x000f0000) | ((type & 0xff) << 8) |
+		((flags & 0xf) << 20);
 	desc->p = 1;
 }
 
@@ -159,7 +160,6 @@ static inline void set_tssldt_descriptor(void *d, unsigned long addr,
 	desc->base2 = (PTR_MIDDLE(addr) >> 8) & 0xFF;
 	desc->base3 = PTR_HIGH(addr);
 #else
-
 	pack_descriptor((struct desc_struct *)d, addr, size, 0x80 | type, 0);
 #endif
 }
@@ -177,7 +177,8 @@ static inline void __set_tss_desc(unsigned cpu, unsigned int entry, void *addr)
 	 * last valid byte
 	 */
 	set_tssldt_descriptor(&tss, (unsigned long)addr, DESC_TSS,
-		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
+			      IO_BITMAP_OFFSET + IO_BITMAP_BYTES +
+			      sizeof(unsigned long) - 1);
 	write_gdt_entry(d, entry, &tss, DESC_TSS);
 }
 
@@ -186,7 +187,7 @@ static inline void __set_tss_desc(unsigned cpu, unsigned int entry, void *addr)
 static inline void native_set_ldt(const void *addr, unsigned int entries)
 {
 	if (likely(entries == 0))
-		__asm__ __volatile__("lldt %w0"::"q" (0));
+		asm volatile("lldt %w0"::"q" (0));
 	else {
 		unsigned cpu = smp_processor_id();
 		ldt_desc ldt;
@@ -195,7 +196,7 @@ static inline void native_set_ldt(const void *addr, unsigned int entries)
 				      DESC_LDT, entries * sizeof(ldt) - 1);
 		write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT,
 				&ldt, DESC_LDT);
-		__asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
+		asm volatile("lldt %w0"::"q" (GDT_ENTRY_LDT*8));
 	}
 }
 
@@ -240,15 +241,15 @@ static inline void native_load_tls(struct thread_struct *t, unsigned int cpu)
 		gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i];
 }
 
-#define _LDT_empty(info) (\
-	(info)->base_addr	== 0	&& \
-	(info)->limit		== 0	&& \
-	(info)->contents	== 0	&& \
-	(info)->read_exec_only	== 1	&& \
-	(info)->seg_32bit	== 0	&& \
-	(info)->limit_in_pages	== 0	&& \
-	(info)->seg_not_present	== 1	&& \
-	(info)->useable		== 0)
+#define _LDT_empty(info)				\
+	((info)->base_addr		== 0	&&	\
+	 (info)->limit			== 0	&&	\
+	 (info)->contents		== 0	&&	\
+	 (info)->read_exec_only		== 1	&&	\
+	 (info)->seg_32bit		== 0	&&	\
+	 (info)->limit_in_pages		== 0	&&	\
+	 (info)->seg_not_present	== 1	&&	\
+	 (info)->useable		== 0)
 
 #ifdef CONFIG_X86_64
 #define LDT_empty(info) (_LDT_empty(info) && ((info)->lm == 0))
@@ -287,7 +288,7 @@ static inline unsigned long get_desc_limit(const struct desc_struct *desc)
 }
 
 static inline void _set_gate(int gate, unsigned type, void *addr,
-			      unsigned dpl, unsigned ist, unsigned seg)
+			     unsigned dpl, unsigned ist, unsigned seg)
 {
 	gate_desc s;
 	pack_gate(&s, type, (unsigned long)addr, dpl, ist, seg);
@@ -370,10 +371,10 @@ static inline void set_system_gate_ist(int n, void *addr, unsigned ist)
  *    Will read the base address of GDT_ENTRY_ESPFIX_SS and put it into %eax.
  */
 #define GET_DESC_BASE(idx, gdt, base, lo_w, lo_b, hi_b) \
-	movb idx*8+4(gdt), lo_b; \
-	movb idx*8+7(gdt), hi_b; \
-	shll $16, base; \
-	movw idx*8+2(gdt), lo_w;
+	movb idx * 8 + 4(gdt), lo_b;			\
+	movb idx * 8 + 7(gdt), hi_b;			\
+	shll $16, base;					\
+	movw idx * 8 + 2(gdt), lo_w;
 
 
 #endif /* __ASSEMBLY__ */
-- 
cgit v1.2.3


From 925a09b27fc73593eb705e50e6db86c4a22091db Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:01:59 -0700
Subject: include/asm-x86/div64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/div64.h | 35 ++++++++++++++++++-----------------
 1 file changed, 18 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/div64.h b/include/asm-x86/div64.h
index e98d16e7a37a..0dbf8bf3ef0a 100644
--- a/include/asm-x86/div64.h
+++ b/include/asm-x86/div64.h
@@ -17,18 +17,20 @@
  * This ends up being the most efficient "calling
  * convention" on x86.
  */
-#define do_div(n,base) ({ \
-	unsigned long __upper, __low, __high, __mod, __base; \
-	__base = (base); \
-	asm("":"=a" (__low), "=d" (__high):"A" (n)); \
-	__upper = __high; \
-	if (__high) { \
-		__upper = __high % (__base); \
-		__high = __high / (__base); \
-	} \
-	asm("divl %2":"=a" (__low), "=d" (__mod):"rm" (__base), "0" (__low), "1" (__upper)); \
-	asm("":"=A" (n):"a" (__low),"d" (__high)); \
-	__mod; \
+#define do_div(n, base)						\
+({								\
+	unsigned long __upper, __low, __high, __mod, __base;	\
+	__base = (base);					\
+	asm("":"=a" (__low), "=d" (__high) : "A" (n));		\
+	__upper = __high;					\
+	if (__high) {						\
+		__upper = __high % (__base);			\
+		__high = __high / (__base);			\
+	}							\
+	asm("divl %2":"=a" (__low), "=d" (__mod)		\
+	    : "rm" (__base), "0" (__low), "1" (__upper));	\
+	asm("":"=A" (n) : "a" (__low), "d" (__high));		\
+	__mod;							\
 })
 
 /*
@@ -37,14 +39,13 @@
  *
  * Warning, this will do an exception if X overflows.
  */
-#define div_long_long_rem(a,b,c) div_ll_X_l_rem(a,b,c)
+#define div_long_long_rem(a, b, c) div_ll_X_l_rem(a, b, c)
 
-static inline long
-div_ll_X_l_rem(long long divs, long div, long *rem)
+static inline long div_ll_X_l_rem(long long divs, long div, long *rem)
 {
 	long dum2;
-      __asm__("divl %2":"=a"(dum2), "=d"(*rem)
-      :	"rm"(div), "A"(divs));
+	asm("divl %2":"=a"(dum2), "=d"(*rem)
+	    : "rm"(div), "A"(divs));
 
 	return dum2;
 
-- 
cgit v1.2.3


From 113cbebac7c88ce78a48801e159108ce6c9d1fb3 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:00 -0700
Subject: include/asm-x86/dma.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/dma.h | 45 ++++++++++++++++++++++-----------------------
 1 file changed, 22 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/dma.h b/include/asm-x86/dma.h
index e9733ce89880..ca1098a7e580 100644
--- a/include/asm-x86/dma.h
+++ b/include/asm-x86/dma.h
@@ -12,7 +12,6 @@
 #include <asm/io.h>		/* need byte IO */
 #include <linux/delay.h>
 
-
 #ifdef HAVE_REALLY_SLOW_DMA_CONTROLLER
 #define dma_outb	outb_p
 #else
@@ -74,15 +73,15 @@
 #ifdef CONFIG_X86_32
 
 /* The maximum address that we can perform a DMA transfer to on this platform */
-#define MAX_DMA_ADDRESS      (PAGE_OFFSET+0x1000000)
+#define MAX_DMA_ADDRESS      (PAGE_OFFSET + 0x1000000)
 
 #else
 
 /* 16MB ISA DMA zone */
-#define MAX_DMA_PFN   ((16*1024*1024) >> PAGE_SHIFT)
+#define MAX_DMA_PFN   ((16 * 1024 * 1024) >> PAGE_SHIFT)
 
 /* 4GB broken PCI/AGP hardware bus master zone */
-#define MAX_DMA32_PFN ((4UL*1024*1024*1024) >> PAGE_SHIFT)
+#define MAX_DMA32_PFN ((4UL * 1024 * 1024 * 1024) >> PAGE_SHIFT)
 
 /* Compat define for old dma zone */
 #define MAX_DMA_ADDRESS ((unsigned long)__va(MAX_DMA_PFN << PAGE_SHIFT))
@@ -154,20 +153,20 @@
 
 extern spinlock_t  dma_spin_lock;
 
-static __inline__ unsigned long claim_dma_lock(void)
+static inline unsigned long claim_dma_lock(void)
 {
 	unsigned long flags;
 	spin_lock_irqsave(&dma_spin_lock, flags);
 	return flags;
 }
 
-static __inline__ void release_dma_lock(unsigned long flags)
+static inline void release_dma_lock(unsigned long flags)
 {
 	spin_unlock_irqrestore(&dma_spin_lock, flags);
 }
 
 /* enable/disable a specific DMA channel */
-static __inline__ void enable_dma(unsigned int dmanr)
+static inline void enable_dma(unsigned int dmanr)
 {
 	if (dmanr <= 3)
 		dma_outb(dmanr, DMA1_MASK_REG);
@@ -175,7 +174,7 @@ static __inline__ void enable_dma(unsigned int dmanr)
 		dma_outb(dmanr & 3, DMA2_MASK_REG);
 }
 
-static __inline__ void disable_dma(unsigned int dmanr)
+static inline void disable_dma(unsigned int dmanr)
 {
 	if (dmanr <= 3)
 		dma_outb(dmanr | 4, DMA1_MASK_REG);
@@ -190,7 +189,7 @@ static __inline__ void disable_dma(unsigned int dmanr)
  * --- In order to do that, the DMA routines below should ---
  * --- only be used while holding the DMA lock ! ---
  */
-static __inline__ void clear_dma_ff(unsigned int dmanr)
+static inline void clear_dma_ff(unsigned int dmanr)
 {
 	if (dmanr <= 3)
 		dma_outb(0, DMA1_CLEAR_FF_REG);
@@ -199,7 +198,7 @@ static __inline__ void clear_dma_ff(unsigned int dmanr)
 }
 
 /* set mode (above) for a specific DMA channel */
-static __inline__ void set_dma_mode(unsigned int dmanr, char mode)
+static inline void set_dma_mode(unsigned int dmanr, char mode)
 {
 	if (dmanr <= 3)
 		dma_outb(mode | dmanr, DMA1_MODE_REG);
@@ -212,7 +211,7 @@ static __inline__ void set_dma_mode(unsigned int dmanr, char mode)
  * the lower 16 bits of the DMA current address register, but a 64k boundary
  * may have been crossed.
  */
-static __inline__ void set_dma_page(unsigned int dmanr, char pagenr)
+static inline void set_dma_page(unsigned int dmanr, char pagenr)
 {
 	switch (dmanr) {
 	case 0:
@@ -243,15 +242,15 @@ static __inline__ void set_dma_page(unsigned int dmanr, char pagenr)
 /* Set transfer address & page bits for specific DMA channel.
  * Assumes dma flipflop is clear.
  */
-static __inline__ void set_dma_addr(unsigned int dmanr, unsigned int a)
+static inline void set_dma_addr(unsigned int dmanr, unsigned int a)
 {
 	set_dma_page(dmanr, a>>16);
 	if (dmanr <= 3)  {
 		dma_outb(a & 0xff, ((dmanr & 3) << 1) + IO_DMA1_BASE);
 		dma_outb((a >> 8) & 0xff, ((dmanr & 3) << 1) + IO_DMA1_BASE);
 	}  else  {
-	    dma_outb((a >> 1) & 0xff, ((dmanr & 3) << 2) + IO_DMA2_BASE);
-	    dma_outb((a >> 9) & 0xff, ((dmanr & 3) << 2) + IO_DMA2_BASE);
+		dma_outb((a >> 1) & 0xff, ((dmanr & 3) << 2) + IO_DMA2_BASE);
+		dma_outb((a >> 9) & 0xff, ((dmanr & 3) << 2) + IO_DMA2_BASE);
 	}
 }
 
@@ -264,18 +263,18 @@ static __inline__ void set_dma_addr(unsigned int dmanr, unsigned int a)
  * Assumes dma flip-flop is clear.
  * NOTE 2: "count" represents _bytes_ and must be even for channels 5-7.
  */
-static __inline__ void set_dma_count(unsigned int dmanr, unsigned int count)
+static inline void set_dma_count(unsigned int dmanr, unsigned int count)
 {
 	count--;
 	if (dmanr <= 3)  {
-	    dma_outb(count & 0xff, ((dmanr & 3) << 1) + 1 + IO_DMA1_BASE);
-	    dma_outb((count >> 8) & 0xff,
-		     ((dmanr & 3) << 1) + 1 + IO_DMA1_BASE);
+		dma_outb(count & 0xff, ((dmanr & 3) << 1) + 1 + IO_DMA1_BASE);
+		dma_outb((count >> 8) & 0xff,
+			 ((dmanr & 3) << 1) + 1 + IO_DMA1_BASE);
 	} else {
-	    dma_outb((count >> 1) & 0xff,
-		     ((dmanr & 3) << 2) + 2 + IO_DMA2_BASE);
-	    dma_outb((count >> 9) & 0xff,
-		     ((dmanr & 3) << 2) + 2 + IO_DMA2_BASE);
+		dma_outb((count >> 1) & 0xff,
+			 ((dmanr & 3) << 2) + 2 + IO_DMA2_BASE);
+		dma_outb((count >> 9) & 0xff,
+			 ((dmanr & 3) << 2) + 2 + IO_DMA2_BASE);
 	}
 }
 
@@ -288,7 +287,7 @@ static __inline__ void set_dma_count(unsigned int dmanr, unsigned int count)
  *
  * Assumes DMA flip-flop is clear.
  */
-static __inline__ int get_dma_residue(unsigned int dmanr)
+static inline int get_dma_residue(unsigned int dmanr)
 {
 	unsigned int io_port;
 	/* using short to get 16-bit wrap around */
-- 
cgit v1.2.3


From c884534f7e74e017f96f81efc31f26ef2b15572e Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:03 -0700
Subject: include/asm-x86/dwarf2_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/dwarf2_64.h | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/dwarf2_64.h b/include/asm-x86/dwarf2_64.h
index eedc08526b0b..c950519a264d 100644
--- a/include/asm-x86/dwarf2_64.h
+++ b/include/asm-x86/dwarf2_64.h
@@ -1,16 +1,15 @@
 #ifndef _DWARF2_H
 #define _DWARF2_H 1
 
-
 #ifndef __ASSEMBLY__
 #warning "asm/dwarf2.h should be only included in pure assembly files"
 #endif
 
-/* 
+/*
    Macros for dwarf2 CFI unwind table entries.
-   See "as.info" for details on these pseudo ops. Unfortunately 
-   they are only supported in very new binutils, so define them 
-   away for older version. 
+   See "as.info" for details on these pseudo ops. Unfortunately
+   they are only supported in very new binutils, so define them
+   away for older version.
  */
 
 #ifdef CONFIG_AS_CFI
-- 
cgit v1.2.3


From 1257d6e0408ba21d72dc3db9ef69d7287c74507b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:04 -0700
Subject: include/asm-x86/e820_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/e820_32.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/e820_32.h b/include/asm-x86/e820_32.h
index e7207a6de3e0..43b1a8bd4b34 100644
--- a/include/asm-x86/e820_32.h
+++ b/include/asm-x86/e820_32.h
@@ -34,8 +34,8 @@ extern void e820_register_memory(void);
 extern void limit_regions(unsigned long long size);
 extern void print_memory_map(char *who);
 extern void init_iomem_resources(struct resource *code_resource,
-			    struct resource *data_resource,
-			    struct resource *bss_resource);
+				 struct resource *data_resource,
+				 struct resource *bss_resource);
 
 #if defined(CONFIG_PM) && defined(CONFIG_HIBERNATION)
 extern void e820_mark_nosave_regions(void);
-- 
cgit v1.2.3


From 86bbc83526a4543db0a4ffed8e4cf679eddd534c Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:05 -0700
Subject: include/asm-x86/e820_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/e820_64.h | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/e820_64.h b/include/asm-x86/e820_64.h
index d38820b31c1c..f478c57eb060 100644
--- a/include/asm-x86/e820_64.h
+++ b/include/asm-x86/e820_64.h
@@ -14,22 +14,24 @@
 #include <linux/ioport.h>
 
 #ifndef __ASSEMBLY__
-extern unsigned long find_e820_area(unsigned long start, unsigned long end, 
+extern unsigned long find_e820_area(unsigned long start, unsigned long end,
 				    unsigned long size, unsigned long align);
 extern unsigned long find_e820_area_size(unsigned long start,
 					 unsigned long *sizep,
 					 unsigned long align);
-extern void add_memory_region(unsigned long start, unsigned long size, 
+extern void add_memory_region(unsigned long start, unsigned long size,
 			      int type);
 extern void update_memory_range(u64 start, u64 size, unsigned old_type,
 				unsigned new_type);
 extern void setup_memory_region(void);
-extern void contig_e820_setup(void); 
+extern void contig_e820_setup(void);
 extern unsigned long e820_end_of_ram(void);
 extern void e820_reserve_resources(void);
 extern void e820_mark_nosave_regions(void);
-extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
-extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
+extern int e820_any_mapped(unsigned long start, unsigned long end,
+			   unsigned type);
+extern int e820_all_mapped(unsigned long start, unsigned long end,
+			   unsigned type);
 extern int e820_any_non_reserved(unsigned long start, unsigned long end);
 extern int is_memory_any_valid(unsigned long start, unsigned long end);
 extern int e820_all_non_reserved(unsigned long start, unsigned long end);
@@ -37,8 +39,8 @@ extern int is_memory_all_valid(unsigned long start, unsigned long end);
 extern unsigned long e820_hole_size(unsigned long start, unsigned long end);
 
 extern void e820_setup_gap(void);
-extern void e820_register_active_regions(int nid,
-				unsigned long start_pfn, unsigned long end_pfn);
+extern void e820_register_active_regions(int nid, unsigned long start_pfn,
+					 unsigned long end_pfn);
 
 extern void finish_e820_parsing(void);
 
-- 
cgit v1.2.3


From 451dd9835898d4cc3c0ee8f9e4883807b760eb02 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:06 -0700
Subject: include/asm-x86/edac.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/edac.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/edac.h b/include/asm-x86/edac.h
index cf3200a745ad..a8088f63a30e 100644
--- a/include/asm-x86/edac.h
+++ b/include/asm-x86/edac.h
@@ -3,7 +3,7 @@
 
 /* ECC atomic, DMA, SMP and interrupt safe scrub function */
 
-static __inline__ void atomic_scrub(void *va, u32 size)
+static inline void atomic_scrub(void *va, u32 size)
 {
 	u32 i, *virt_addr = va;
 
@@ -12,7 +12,7 @@ static __inline__ void atomic_scrub(void *va, u32 size)
 	 * are interrupt, DMA and SMP safe.
 	 */
 	for (i = 0; i < size / 4; i++, virt_addr++)
-		__asm__ __volatile__("lock; addl $0, %0"::"m"(*virt_addr));
+		asm volatile("lock; addl $0, %0"::"m" (*virt_addr));
 }
 
 #endif
-- 
cgit v1.2.3


From 0c6f6abf6e63227b1e6e22e412c3f63c0163a520 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:07 -0700
Subject: include/asm-x86/efi.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/efi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/efi.h b/include/asm-x86/efi.h
index ea9734b74aca..d53004b855cc 100644
--- a/include/asm-x86/efi.h
+++ b/include/asm-x86/efi.h
@@ -20,7 +20,7 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...);
  */
 
 #define efi_call_virt(f, args...) \
-     ((efi_##f##_t __attribute__((regparm(0)))*)efi.systab->runtime->f)(args)
+	((efi_##f##_t __attribute__((regparm(0)))*)efi.systab->runtime->f)(args)
 
 #define efi_call_virt0(f)		efi_call_virt(f)
 #define efi_call_virt1(f, a1)		efi_call_virt(f, a1)
-- 
cgit v1.2.3


From 486386f6c1b5144ccc8eb2f28def1712e1dd6c3d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:08 -0700
Subject: include/asm-x86/elf.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/elf.h | 107 ++++++++++++++++++++++++++++----------------------
 1 file changed, 60 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/elf.h b/include/asm-x86/elf.h
index 77325646e2f4..8f232dc5b5fe 100644
--- a/include/asm-x86/elf.h
+++ b/include/asm-x86/elf.h
@@ -11,7 +11,7 @@
 
 typedef unsigned long elf_greg_t;
 
-#define ELF_NGREG (sizeof (struct user_regs_struct) / sizeof(elf_greg_t))
+#define ELF_NGREG (sizeof(struct user_regs_struct) / sizeof(elf_greg_t))
 typedef elf_greg_t elf_gregset_t[ELF_NGREG];
 
 typedef struct user_i387_struct elf_fpregset_t;
@@ -100,10 +100,11 @@ extern unsigned int vdso_enabled;
    We might as well make sure everything else is cleared too (except for %esp),
    just to make things more deterministic.
  */
-#define ELF_PLAT_INIT(_r, load_addr)	do { \
-	_r->bx = 0; _r->cx = 0; _r->dx = 0; \
-	_r->si = 0; _r->di = 0; _r->bp = 0; \
-	_r->ax = 0; \
+#define ELF_PLAT_INIT(_r, load_addr)		\
+	do {					\
+	_r->bx = 0; _r->cx = 0; _r->dx = 0;	\
+	_r->si = 0; _r->di = 0; _r->bp = 0;	\
+	_r->ax = 0;				\
 } while (0)
 
 /*
@@ -111,24 +112,25 @@ extern unsigned int vdso_enabled;
  * now struct_user_regs, they are different)
  */
 
-#define ELF_CORE_COPY_REGS(pr_reg, regs) do {		\
-	pr_reg[0] = regs->bx;				\
-	pr_reg[1] = regs->cx;				\
-	pr_reg[2] = regs->dx;				\
-	pr_reg[3] = regs->si;				\
-	pr_reg[4] = regs->di;				\
-	pr_reg[5] = regs->bp;				\
-	pr_reg[6] = regs->ax;				\
-	pr_reg[7] = regs->ds & 0xffff;			\
-	pr_reg[8] = regs->es & 0xffff;			\
-	pr_reg[9] = regs->fs & 0xffff;			\
-	savesegment(gs, pr_reg[10]);			\
-	pr_reg[11] = regs->orig_ax;			\
-	pr_reg[12] = regs->ip;				\
-	pr_reg[13] = regs->cs & 0xffff;			\
-	pr_reg[14] = regs->flags;			\
-	pr_reg[15] = regs->sp;				\
-	pr_reg[16] = regs->ss & 0xffff;			\
+#define ELF_CORE_COPY_REGS(pr_reg, regs)	\
+do {						\
+	pr_reg[0] = regs->bx;			\
+	pr_reg[1] = regs->cx;			\
+	pr_reg[2] = regs->dx;			\
+	pr_reg[3] = regs->si;			\
+	pr_reg[4] = regs->di;			\
+	pr_reg[5] = regs->bp;			\
+	pr_reg[6] = regs->ax;			\
+	pr_reg[7] = regs->ds & 0xffff;		\
+	pr_reg[8] = regs->es & 0xffff;		\
+	pr_reg[9] = regs->fs & 0xffff;		\
+	savesegment(gs, pr_reg[10]);		\
+	pr_reg[11] = regs->orig_ax;		\
+	pr_reg[12] = regs->ip;			\
+	pr_reg[13] = regs->cs & 0xffff;		\
+	pr_reg[14] = regs->flags;		\
+	pr_reg[15] = regs->sp;			\
+	pr_reg[16] = regs->ss & 0xffff;		\
 } while (0);
 
 #define ELF_PLATFORM	(utsname()->machine)
@@ -139,7 +141,7 @@ extern unsigned int vdso_enabled;
 /*
  * This is used to ensure we don't load something for the wrong architecture.
  */
-#define elf_check_arch(x) \
+#define elf_check_arch(x)			\
 	((x)->e_machine == EM_X86_64)
 
 #define compat_elf_check_arch(x)	elf_check_arch_ia32(x)
@@ -168,24 +170,30 @@ static inline void elf_common_init(struct thread_struct *t,
 	t->ds = t->es = ds;
 }
 
-#define ELF_PLAT_INIT(_r, load_addr)	do {		  \
-	elf_common_init(&current->thread, _r, 0);	  \
-	clear_thread_flag(TIF_IA32);			  \
+#define ELF_PLAT_INIT(_r, load_addr)			\
+do {							\
+	elf_common_init(&current->thread, _r, 0);	\
+	clear_thread_flag(TIF_IA32);			\
 } while (0)
 
-#define	COMPAT_ELF_PLAT_INIT(regs, load_addr)	\
+#define	COMPAT_ELF_PLAT_INIT(regs, load_addr)		\
 	elf_common_init(&current->thread, regs, __USER_DS)
-#define	compat_start_thread(regs, ip, sp)	do {		\
-		start_ia32_thread(regs, ip, sp);		\
-		set_fs(USER_DS);				\
-	} while (0)
-#define COMPAT_SET_PERSONALITY(ex, ibcs2)	do {		\
-		if (test_thread_flag(TIF_IA32))			\
-			clear_thread_flag(TIF_ABI_PENDING);	\
-		else						\
-			set_thread_flag(TIF_ABI_PENDING);	\
-		current->personality |= force_personality32;	\
-	} while (0)
+
+#define	compat_start_thread(regs, ip, sp)		\
+do {							\
+	start_ia32_thread(regs, ip, sp);		\
+	set_fs(USER_DS);				\
+} while (0)
+
+#define COMPAT_SET_PERSONALITY(ex, ibcs2)		\
+do {							\
+	if (test_thread_flag(TIF_IA32))			\
+		clear_thread_flag(TIF_ABI_PENDING);	\
+	else						\
+		set_thread_flag(TIF_ABI_PENDING);	\
+	current->personality |= force_personality32;	\
+} while (0)
+
 #define COMPAT_ELF_PLATFORM			("i686")
 
 /*
@@ -194,7 +202,8 @@ static inline void elf_common_init(struct thread_struct *t,
  * getting dumped.
  */
 
-#define ELF_CORE_COPY_REGS(pr_reg, regs)  do {			\
+#define ELF_CORE_COPY_REGS(pr_reg, regs)			\
+do {								\
 	unsigned v;						\
 	(pr_reg)[0] = (regs)->r15;				\
 	(pr_reg)[1] = (regs)->r14;				\
@@ -268,10 +277,12 @@ extern int force_personality32;
 
 struct task_struct;
 
-#define	ARCH_DLINFO_IA32(vdso_enabled) \
-do if (vdso_enabled) {							\
+#define	ARCH_DLINFO_IA32(vdso_enabled)					\
+do {									\
+	if (vdso_enabled) {						\
 		NEW_AUX_ENT(AT_SYSINFO,	VDSO_ENTRY);			\
 		NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_CURRENT_BASE);	\
+	}								\
 } while (0)
 
 #ifdef CONFIG_X86_32
@@ -289,9 +300,11 @@ do if (vdso_enabled) {							\
 /* 1GB for 64bit, 8MB for 32bit */
 #define STACK_RND_MASK (test_thread_flag(TIF_IA32) ? 0x7ff : 0x3fffff)
 
-#define ARCH_DLINFO						\
-do if (vdso_enabled) {						\
-	NEW_AUX_ENT(AT_SYSINFO_EHDR,(unsigned long)current->mm->context.vdso);\
+#define ARCH_DLINFO							\
+do {									\
+	if (vdso_enabled)						\
+		NEW_AUX_ENT(AT_SYSINFO_EHDR,				\
+			    (unsigned long)current->mm->context.vdso);	\
 } while (0)
 
 #define AT_SYSINFO		32
@@ -304,8 +317,8 @@ do if (vdso_enabled) {						\
 
 #define VDSO_CURRENT_BASE	((unsigned long)current->mm->context.vdso)
 
-#define VDSO_ENTRY \
-	((unsigned long) VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall))
+#define VDSO_ENTRY							\
+	((unsigned long)VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall))
 
 struct linux_binprm;
 
-- 
cgit v1.2.3


From cb7d0617b776b91a0643b8c5988f236414129b7d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:09 -0700
Subject: include/asm-x86/fixmap_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/fixmap_32.h | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h
index a7404d50686b..eb1665125c44 100644
--- a/include/asm-x86/fixmap_32.h
+++ b/include/asm-x86/fixmap_32.h
@@ -99,8 +99,7 @@ enum fixed_addresses {
 	 */
 #define NR_FIX_BTMAPS		64
 #define FIX_BTMAPS_NESTING	4
-	FIX_BTMAP_END =
-		__end_of_permanent_fixed_addresses + 512 -
+	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 512 -
 			(__end_of_permanent_fixed_addresses & 511),
 	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
 	FIX_WP_TEST,
@@ -110,20 +109,20 @@ enum fixed_addresses {
 	__end_of_fixed_addresses
 };
 
-extern void __set_fixmap (enum fixed_addresses idx,
-					unsigned long phys, pgprot_t flags);
+extern void __set_fixmap(enum fixed_addresses idx,
+			 unsigned long phys, pgprot_t flags);
 extern void reserve_top_address(unsigned long reserve);
 
-#define set_fixmap(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL)
+#define set_fixmap(idx, phys)				\
+	__set_fixmap(idx, phys, PAGE_KERNEL)
 /*
  * Some hardware wants to get fixmapped without caching.
  */
-#define set_fixmap_nocache(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
+#define set_fixmap_nocache(idx, phys)			\
+	__set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
 
-#define clear_fixmap(idx) \
-		__set_fixmap(idx, 0, __pgprot(0))
+#define clear_fixmap(idx)			\
+	__set_fixmap(idx, 0, __pgprot(0))
 
 #define FIXADDR_TOP	((unsigned long)__FIXADDR_TOP)
 
@@ -156,7 +155,7 @@ static __always_inline unsigned long fix_to_virt(const unsigned int idx)
 	if (idx >= __end_of_fixed_addresses)
 		__this_fixmap_does_not_exist();
 
-        return __fix_to_virt(idx);
+	return __fix_to_virt(idx);
 }
 
 static inline unsigned long virt_to_fix(const unsigned long vaddr)
-- 
cgit v1.2.3


From dca13fb0fc89d4214090a0ae14aeeaa50c42a6c4 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:10 -0700
Subject: include/asm-x86/fixmap_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/fixmap_64.h | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h
index 70ddb21e6458..f3d76858c0e6 100644
--- a/include/asm-x86/fixmap_64.h
+++ b/include/asm-x86/fixmap_64.h
@@ -34,32 +34,34 @@
 
 enum fixed_addresses {
 	VSYSCALL_LAST_PAGE,
-	VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
+	VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE
+			    + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
 	VSYSCALL_HPET,
 	FIX_DBGP_BASE,
 	FIX_EARLYCON_MEM_BASE,
 	FIX_HPET_BASE,
 	FIX_APIC_BASE,	/* local (CPU) APIC) -- required for SMP or not */
 	FIX_IO_APIC_BASE_0,
-	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
+	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS - 1,
 	FIX_EFI_IO_MAP_LAST_PAGE,
-	FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE+MAX_EFI_IO_PAGES-1,
+	FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE
+				  + MAX_EFI_IO_PAGES - 1,
 #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
 	FIX_OHCI1394_BASE,
 #endif
 	__end_of_fixed_addresses
 };
 
-extern void __set_fixmap (enum fixed_addresses idx,
-					unsigned long phys, pgprot_t flags);
+extern void __set_fixmap(enum fixed_addresses idx,
+			 unsigned long phys, pgprot_t flags);
 
-#define set_fixmap(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL)
+#define set_fixmap(idx, phys)			\
+	__set_fixmap(idx, phys, PAGE_KERNEL)
 /*
  * Some hardware wants to get fixmapped without caching.
  */
-#define set_fixmap_nocache(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
+#define set_fixmap_nocache(idx, phys)			\
+	__set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
 
 #define FIXADDR_TOP	(VSYSCALL_END-PAGE_SIZE)
 #define FIXADDR_SIZE	(__end_of_fixed_addresses << PAGE_SHIFT)
-- 
cgit v1.2.3


From 4637bc07c85621b0c10320da8cf3b34de83efa0f Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:11 -0700
Subject: include/asm-x86/floppy.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/floppy.h | 87 +++++++++++++++++++++++++-----------------------
 1 file changed, 45 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/floppy.h b/include/asm-x86/floppy.h
index a48d7153c097..438b3033a250 100644
--- a/include/asm-x86/floppy.h
+++ b/include/asm-x86/floppy.h
@@ -20,20 +20,21 @@
  * driver otherwise. It doesn't matter much for performance anyway, as most
  * floppy accesses go through the track buffer.
  */
-#define _CROSS_64KB(a,s,vdma) \
-(!(vdma) && ((unsigned long)(a)/K_64 != ((unsigned long)(a) + (s) - 1) / K_64))
+#define _CROSS_64KB(a, s, vdma)						\
+	(!(vdma) &&							\
+	 ((unsigned long)(a)/K_64 != ((unsigned long)(a) + (s) - 1) / K_64))
 
-#define CROSS_64KB(a,s) _CROSS_64KB(a,s,use_virtual_dma & 1)
+#define CROSS_64KB(a, s) _CROSS_64KB(a, s, use_virtual_dma & 1)
 
 
-#define SW fd_routine[use_virtual_dma&1]
+#define SW fd_routine[use_virtual_dma & 1]
 #define CSW fd_routine[can_use_virtual_dma & 1]
 
 
 #define fd_inb(port)		inb_p(port)
-#define fd_outb(value,port)	outb_p(value,port)
+#define fd_outb(value, port)	outb_p(value, port)
 
-#define fd_request_dma()	CSW._request_dma(FLOPPY_DMA,"floppy")
+#define fd_request_dma()	CSW._request_dma(FLOPPY_DMA, "floppy")
 #define fd_free_dma()		CSW._free_dma(FLOPPY_DMA)
 #define fd_enable_irq()		enable_irq(FLOPPY_IRQ)
 #define fd_disable_irq()	disable_irq(FLOPPY_IRQ)
@@ -57,15 +58,15 @@ static irqreturn_t floppy_hardint(int irq, void *dev_id)
 #undef TRACE_FLPY_INT
 
 #ifdef TRACE_FLPY_INT
-	static int calls=0;
-	static int bytes=0;
-	static int dma_wait=0;
+	static int calls;
+	static int bytes;
+	static int dma_wait;
 #endif
 	if (!doing_pdma)
 		return floppy_interrupt(irq, dev_id);
 
 #ifdef TRACE_FLPY_INT
-	if(!calls)
+	if (!calls)
 		bytes = virtual_dma_count;
 #endif
 
@@ -74,42 +75,42 @@ static irqreturn_t floppy_hardint(int irq, void *dev_id)
 		register char *lptr;
 
 		st = 1;
-		for(lcount=virtual_dma_count, lptr=virtual_dma_addr;
-		    lcount; lcount--, lptr++) {
-			st=inb(virtual_dma_port+4) & 0xa0 ;
-			if(st != 0xa0)
+		for (lcount = virtual_dma_count, lptr = virtual_dma_addr;
+		     lcount; lcount--, lptr++) {
+			st = inb(virtual_dma_port + 4) & 0xa0;
+			if (st != 0xa0)
 				break;
-			if(virtual_dma_mode)
-				outb_p(*lptr, virtual_dma_port+5);
+			if (virtual_dma_mode)
+				outb_p(*lptr, virtual_dma_port + 5);
 			else
-				*lptr = inb_p(virtual_dma_port+5);
+				*lptr = inb_p(virtual_dma_port + 5);
 		}
 		virtual_dma_count = lcount;
 		virtual_dma_addr = lptr;
-		st = inb(virtual_dma_port+4);
+		st = inb(virtual_dma_port + 4);
 	}
 
 #ifdef TRACE_FLPY_INT
 	calls++;
 #endif
-	if(st == 0x20)
+	if (st == 0x20)
 		return IRQ_HANDLED;
-	if(!(st & 0x20)) {
+	if (!(st & 0x20)) {
 		virtual_dma_residue += virtual_dma_count;
-		virtual_dma_count=0;
+		virtual_dma_count = 0;
 #ifdef TRACE_FLPY_INT
 		printk("count=%x, residue=%x calls=%d bytes=%d dma_wait=%d\n",
 		       virtual_dma_count, virtual_dma_residue, calls, bytes,
 		       dma_wait);
 		calls = 0;
-		dma_wait=0;
+		dma_wait = 0;
 #endif
 		doing_pdma = 0;
 		floppy_interrupt(irq, dev_id);
 		return IRQ_HANDLED;
 	}
 #ifdef TRACE_FLPY_INT
-	if(!virtual_dma_count)
+	if (!virtual_dma_count)
 		dma_wait++;
 #endif
 	return IRQ_HANDLED;
@@ -117,14 +118,14 @@ static irqreturn_t floppy_hardint(int irq, void *dev_id)
 
 static void fd_disable_dma(void)
 {
-	if(! (can_use_virtual_dma & 1))
+	if (!(can_use_virtual_dma & 1))
 		disable_dma(FLOPPY_DMA);
 	doing_pdma = 0;
 	virtual_dma_residue += virtual_dma_count;
-	virtual_dma_count=0;
+	virtual_dma_count = 0;
 }
 
-static int vdma_request_dma(unsigned int dmanr, const char * device_id)
+static int vdma_request_dma(unsigned int dmanr, const char *device_id)
 {
 	return 0;
 }
@@ -142,7 +143,7 @@ static int vdma_get_dma_residue(unsigned int dummy)
 
 static int fd_request_irq(void)
 {
-	if(can_use_virtual_dma)
+	if (can_use_virtual_dma)
 		return request_irq(FLOPPY_IRQ, floppy_hardint,
 				   IRQF_DISABLED, "floppy", NULL);
 	else
@@ -152,13 +153,13 @@ static int fd_request_irq(void)
 
 static unsigned long dma_mem_alloc(unsigned long size)
 {
-	return __get_dma_pages(GFP_KERNEL|__GFP_NORETRY,get_order(size));
+	return __get_dma_pages(GFP_KERNEL|__GFP_NORETRY, get_order(size));
 }
 
 
 static unsigned long vdma_mem_alloc(unsigned long size)
 {
-	return (unsigned long) vmalloc(size);
+	return (unsigned long)vmalloc(size);
 
 }
 
@@ -166,7 +167,7 @@ static unsigned long vdma_mem_alloc(unsigned long size)
 
 static void _fd_dma_mem_free(unsigned long addr, unsigned long size)
 {
-	if((unsigned long) addr >= (unsigned long) high_memory)
+	if ((unsigned long)addr >= (unsigned long)high_memory)
 		vfree((void *)addr);
 	else
 		free_pages(addr, get_order(size));
@@ -176,10 +177,10 @@ static void _fd_dma_mem_free(unsigned long addr, unsigned long size)
 
 static void _fd_chose_dma_mode(char *addr, unsigned long size)
 {
-	if(can_use_virtual_dma == 2) {
-		if((unsigned long) addr >= (unsigned long) high_memory ||
-		   isa_virt_to_bus(addr) >= 0x1000000 ||
-		   _CROSS_64KB(addr, size, 0))
+	if (can_use_virtual_dma == 2) {
+		if ((unsigned long)addr >= (unsigned long)high_memory ||
+		    isa_virt_to_bus(addr) >= 0x1000000 ||
+		    _CROSS_64KB(addr, size, 0))
 			use_virtual_dma = 1;
 		else
 			use_virtual_dma = 0;
@@ -195,7 +196,7 @@ static int vdma_dma_setup(char *addr, unsigned long size, int mode, int io)
 {
 	doing_pdma = 1;
 	virtual_dma_port = io;
-	virtual_dma_mode = (mode  == DMA_MODE_WRITE);
+	virtual_dma_mode = (mode == DMA_MODE_WRITE);
 	virtual_dma_addr = addr;
 	virtual_dma_count = size;
 	virtual_dma_residue = 0;
@@ -213,18 +214,18 @@ static int hard_dma_setup(char *addr, unsigned long size, int mode, int io)
 	/* actual, physical DMA */
 	doing_pdma = 0;
 	clear_dma_ff(FLOPPY_DMA);
-	set_dma_mode(FLOPPY_DMA,mode);
-	set_dma_addr(FLOPPY_DMA,isa_virt_to_bus(addr));
-	set_dma_count(FLOPPY_DMA,size);
+	set_dma_mode(FLOPPY_DMA, mode);
+	set_dma_addr(FLOPPY_DMA, isa_virt_to_bus(addr));
+	set_dma_count(FLOPPY_DMA, size);
 	enable_dma(FLOPPY_DMA);
 	return 0;
 }
 
 static struct fd_routine_l {
-	int (*_request_dma)(unsigned int dmanr, const char * device_id);
+	int (*_request_dma)(unsigned int dmanr, const char *device_id);
 	void (*_free_dma)(unsigned int dmanr);
 	int (*_get_dma_residue)(unsigned int dummy);
-	unsigned long (*_dma_mem_alloc) (unsigned long size);
+	unsigned long (*_dma_mem_alloc)(unsigned long size);
 	int (*_dma_setup)(char *addr, unsigned long size, int mode, int io);
 } fd_routine[] = {
 	{
@@ -252,7 +253,8 @@ static int FDC2 = -1;
  * is needed to prevent corrupted CMOS RAM in case "insmod floppy"
  * coincides with another rtc CMOS user.		Paul G.
  */
-#define FLOPPY0_TYPE	({				\
+#define FLOPPY0_TYPE					\
+({							\
 	unsigned long flags;				\
 	unsigned char val;				\
 	spin_lock_irqsave(&rtc_lock, flags);		\
@@ -261,7 +263,8 @@ static int FDC2 = -1;
 	val;						\
 })
 
-#define FLOPPY1_TYPE	({				\
+#define FLOPPY1_TYPE					\
+({							\
 	unsigned long flags;				\
 	unsigned char val;				\
 	spin_lock_irqsave(&rtc_lock, flags);		\
-- 
cgit v1.2.3


From 9407913fc16dde0e632b9639557422c6a792469d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:12 -0700
Subject: include/asm-x86/futex.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/futex.h | 101 ++++++++++++++++++++++++++----------------------
 1 file changed, 55 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/futex.h b/include/asm-x86/futex.h
index c9952ea9f698..ac0fbf24d722 100644
--- a/include/asm-x86/futex.h
+++ b/include/asm-x86/futex.h
@@ -12,35 +12,32 @@
 #include <asm/uaccess.h>
 
 #define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg)	\
-  __asm__ __volatile(						\
-"1:	" insn "\n"						\
-"2:	.section .fixup,\"ax\"\n				\
-3:	mov	%3, %1\n					\
-	jmp	2b\n						\
-	.previous\n"						\
-	_ASM_EXTABLE(1b,3b)					\
-	: "=r" (oldval), "=r" (ret), "+m" (*uaddr)		\
-	: "i" (-EFAULT), "0" (oparg), "1" (0))
+	asm volatile("1:\t" insn "\n"				\
+		     "2:\t.section .fixup,\"ax\"\n"		\
+		     "3:\tmov\t%3, %1\n"			\
+		     "\tjmp\t2b\n"				\
+		     "\t.previous\n"				\
+		     _ASM_EXTABLE(1b, 3b)			\
+		     : "=r" (oldval), "=r" (ret), "+m" (*uaddr)	\
+		     : "i" (-EFAULT), "0" (oparg), "1" (0))
 
 #define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg)	\
-  __asm__ __volatile(						\
-"1:	movl	%2, %0\n					\
-	movl	%0, %3\n"					\
-	insn "\n"						\
-"2:	lock; cmpxchgl %3, %2\n					\
-	jnz	1b\n						\
-3:	.section .fixup,\"ax\"\n				\
-4:	mov	%5, %1\n					\
-	jmp	3b\n						\
-	.previous\n"						\
-	_ASM_EXTABLE(1b,4b)					\
-	_ASM_EXTABLE(2b,4b)					\
-	: "=&a" (oldval), "=&r" (ret), "+m" (*uaddr),		\
-	  "=&r" (tem)						\
-	: "r" (oparg), "i" (-EFAULT), "1" (0))
-
-static inline int
-futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
+	asm volatile("1:\tmovl	%2, %0\n"			\
+		     "\tmovl\t%0, %3\n"				\
+		     "\t" insn "\n"				\
+		     "2:\tlock; cmpxchgl %3, %2\n"		\
+		     "\tjnz\t1b\n"				\
+		     "3:\t.section .fixup,\"ax\"\n"		\
+		     "4:\tmov\t%5, %1\n"			\
+		     "\tjmp\t3b\n"				\
+		     "\t.previous\n"				\
+		     _ASM_EXTABLE(1b, 4b)			\
+		     _ASM_EXTABLE(2b, 4b)			\
+		     : "=&a" (oldval), "=&r" (ret),		\
+		       "+m" (*uaddr), "=&r" (tem)		\
+		     : "r" (oparg), "i" (-EFAULT), "1" (0))
+
+static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
 {
 	int op = (encoded_op >> 28) & 7;
 	int cmp = (encoded_op >> 24) & 15;
@@ -87,20 +84,33 @@ futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
 
 	if (!ret) {
 		switch (cmp) {
-		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
-		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
-		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
-		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
-		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
-		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
-		default: ret = -ENOSYS;
+		case FUTEX_OP_CMP_EQ:
+			ret = (oldval == cmparg);
+			break;
+		case FUTEX_OP_CMP_NE:
+			ret = (oldval != cmparg);
+			break;
+		case FUTEX_OP_CMP_LT:
+			ret = (oldval < cmparg);
+			break;
+		case FUTEX_OP_CMP_GE:
+			ret = (oldval >= cmparg);
+			break;
+		case FUTEX_OP_CMP_LE:
+			ret = (oldval <= cmparg);
+			break;
+		case FUTEX_OP_CMP_GT:
+			ret = (oldval > cmparg);
+			break;
+		default:
+			ret = -ENOSYS;
 		}
 	}
 	return ret;
 }
 
-static inline int
-futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
+static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval,
+						int newval)
 {
 
 #if defined(CONFIG_X86_32) && !defined(CONFIG_X86_BSWAP)
@@ -112,16 +122,15 @@ futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
 	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
 		return -EFAULT;
 
-	__asm__ __volatile__(
-		"1:	lock; cmpxchgl %3, %1			\n"
-		"2:	.section .fixup, \"ax\"			\n"
-		"3:	mov     %2, %0				\n"
-		"	jmp     2b				\n"
-		"	.previous				\n"
-		_ASM_EXTABLE(1b,3b)
-		: "=a" (oldval), "+m" (*uaddr)
-		: "i" (-EFAULT), "r" (newval), "0" (oldval)
-		: "memory"
+	asm volatile("1:\tlock; cmpxchgl %3, %1\n"
+		     "2:\t.section .fixup, \"ax\"\n"
+		     "3:\tmov     %2, %0\n"
+		     "\tjmp     2b\n"
+		     "\t.previous\n"
+		     _ASM_EXTABLE(1b, 3b)
+		     : "=a" (oldval), "+m" (*uaddr)
+		     : "i" (-EFAULT), "r" (newval), "0" (oldval)
+		     : "memory"
 	);
 
 	return oldval;
-- 
cgit v1.2.3


From eee28c251ed8ef2c6b66f0e08e9467a8cc5cf886 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:13 -0700
Subject: include/asm-x86/genapic_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/genapic_32.h | 89 ++++++++++++++++++++++----------------------
 1 file changed, 45 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/genapic_32.h b/include/asm-x86/genapic_32.h
index b501ae7809ba..5fa893dce729 100644
--- a/include/asm-x86/genapic_32.h
+++ b/include/asm-x86/genapic_32.h
@@ -18,18 +18,18 @@ struct mpc_config_bus;
 struct mp_config_table;
 struct mpc_config_processor;
 
-struct genapic { 
-	char *name; 
-	int (*probe)(void); 
+struct genapic {
+	char *name;
+	int (*probe)(void);
 
 	int (*apic_id_registered)(void);
 	cpumask_t (*target_cpus)(void);
 	int int_delivery_mode;
-	int int_dest_mode; 
+	int int_dest_mode;
 	int ESR_DISABLE;
 	int apic_destination_logical;
 	unsigned long (*check_apicid_used)(physid_mask_t bitmap, int apicid);
-	unsigned long (*check_apicid_present)(int apicid); 
+	unsigned long (*check_apicid_present)(int apicid);
 	int no_balance_irq;
 	int no_ioapic_check;
 	void (*init_apic_ldr)(void);
@@ -37,21 +37,21 @@ struct genapic {
 
 	void (*setup_apic_routing)(void);
 	int (*multi_timer_check)(int apic, int irq);
-	int (*apicid_to_node)(int logical_apicid); 
+	int (*apicid_to_node)(int logical_apicid);
 	int (*cpu_to_logical_apicid)(int cpu);
 	int (*cpu_present_to_apicid)(int mps_cpu);
 	physid_mask_t (*apicid_to_cpu_present)(int phys_apicid);
-	void (*setup_portio_remap)(void); 
+	void (*setup_portio_remap)(void);
 	int (*check_phys_apicid_present)(int boot_cpu_physical_apicid);
 	void (*enable_apic_mode)(void);
 	u32 (*phys_pkg_id)(u32 cpuid_apic, int index_msb);
 
 	/* mpparse */
 	/* When one of the next two hooks returns 1 the genapic
-	   is switched to this. Essentially they are additional probe 
+	   is switched to this. Essentially they are additional probe
 	   functions. */
-	int (*mps_oem_check)(struct mp_config_table *mpc, char *oem, 
-			      char *productid);
+	int (*mps_oem_check)(struct mp_config_table *mpc, char *oem,
+			     char *productid);
 	int (*acpi_madt_oem_check)(char *oem_id, char *oem_table_id);
 
 	unsigned (*get_apic_id)(unsigned long x);
@@ -64,7 +64,7 @@ struct genapic {
 	void (*send_IPI_allbutself)(int vector);
 	void (*send_IPI_all)(int vector);
 #endif
-}; 
+};
 
 #define APICFUNC(x) .x = x,
 
@@ -77,39 +77,40 @@ struct genapic {
 #define IPIFUNC(x)
 #endif
 
-#define APIC_INIT(aname, aprobe) { \
-	.name = aname, \
-	.probe = aprobe, \
-	.int_delivery_mode = INT_DELIVERY_MODE, \
-	.int_dest_mode = INT_DEST_MODE, \
-	.no_balance_irq = NO_BALANCE_IRQ, \
-	.ESR_DISABLE = esr_disable, \
-	.apic_destination_logical = APIC_DEST_LOGICAL, \
-	APICFUNC(apic_id_registered) \
-	APICFUNC(target_cpus) \
-	APICFUNC(check_apicid_used) \
-	APICFUNC(check_apicid_present) \
-	APICFUNC(init_apic_ldr) \
-	APICFUNC(ioapic_phys_id_map) \
-	APICFUNC(setup_apic_routing) \
-	APICFUNC(multi_timer_check) \
-	APICFUNC(apicid_to_node) \
-	APICFUNC(cpu_to_logical_apicid) \
-	APICFUNC(cpu_present_to_apicid) \
-	APICFUNC(apicid_to_cpu_present) \
-	APICFUNC(setup_portio_remap) \
-	APICFUNC(check_phys_apicid_present) \
-	APICFUNC(mps_oem_check) \
-	APICFUNC(get_apic_id) \
-	.apic_id_mask = APIC_ID_MASK, \
-	APICFUNC(cpu_mask_to_apicid) \
-	APICFUNC(acpi_madt_oem_check) \
-	IPIFUNC(send_IPI_mask) \
-	IPIFUNC(send_IPI_allbutself) \
-	IPIFUNC(send_IPI_all) \
-	APICFUNC(enable_apic_mode) \
-	APICFUNC(phys_pkg_id) \
-	}
+#define APIC_INIT(aname, aprobe)			\
+{							\
+	.name = aname,					\
+	.probe = aprobe,				\
+	.int_delivery_mode = INT_DELIVERY_MODE,		\
+	.int_dest_mode = INT_DEST_MODE,			\
+	.no_balance_irq = NO_BALANCE_IRQ,		\
+	.ESR_DISABLE = esr_disable,			\
+	.apic_destination_logical = APIC_DEST_LOGICAL,	\
+	APICFUNC(apic_id_registered)			\
+	APICFUNC(target_cpus)				\
+	APICFUNC(check_apicid_used)			\
+	APICFUNC(check_apicid_present)			\
+	APICFUNC(init_apic_ldr)				\
+	APICFUNC(ioapic_phys_id_map)			\
+	APICFUNC(setup_apic_routing)			\
+	APICFUNC(multi_timer_check)			\
+	APICFUNC(apicid_to_node)			\
+	APICFUNC(cpu_to_logical_apicid)			\
+	APICFUNC(cpu_present_to_apicid)			\
+	APICFUNC(apicid_to_cpu_present)			\
+	APICFUNC(setup_portio_remap)			\
+	APICFUNC(check_phys_apicid_present)		\
+	APICFUNC(mps_oem_check)				\
+	APICFUNC(get_apic_id)				\
+	.apic_id_mask = APIC_ID_MASK,			\
+	APICFUNC(cpu_mask_to_apicid)			\
+	APICFUNC(acpi_madt_oem_check)			\
+	IPIFUNC(send_IPI_mask)				\
+	IPIFUNC(send_IPI_allbutself)			\
+	IPIFUNC(send_IPI_all)				\
+	APICFUNC(enable_apic_mode)			\
+	APICFUNC(phys_pkg_id)				\
+}
 
 extern struct genapic *genapic;
 
-- 
cgit v1.2.3


From 6394d982a96a4b8f7973dea1d9d5a99dbe1847f7 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:14 -0700
Subject: include/asm-x86/geode.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/geode.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/geode.h b/include/asm-x86/geode.h
index 9e7280092a48..9870cc1f2f8f 100644
--- a/include/asm-x86/geode.h
+++ b/include/asm-x86/geode.h
@@ -167,7 +167,7 @@ static inline int is_geode(void)
 /* MFGPTs */
 
 #define MFGPT_MAX_TIMERS	8
-#define MFGPT_TIMER_ANY		-1
+#define MFGPT_TIMER_ANY		(-1)
 
 #define MFGPT_DOMAIN_WORKING	1
 #define MFGPT_DOMAIN_STANDBY	2
-- 
cgit v1.2.3


From b9c2fcf0eb6190f942312a6e04f91a80efb91b72 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:15 -0700
Subject: include/asm-x86/highmem.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/highmem.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/highmem.h b/include/asm-x86/highmem.h
index 479767c9195f..e153f3b44774 100644
--- a/include/asm-x86/highmem.h
+++ b/include/asm-x86/highmem.h
@@ -8,7 +8,7 @@
  *		      Gerhard.Wichert@pdb.siemens.de
  *
  *
- * Redesigned the x86 32-bit VM architecture to deal with 
+ * Redesigned the x86 32-bit VM architecture to deal with
  * up to 16 Terabyte physical memory. With current x86 CPUs
  * we now support up to 64 Gigabytes physical RAM.
  *
-- 
cgit v1.2.3


From dbdab9f6308645a4462e81c607d6282bcb27141d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:16 -0700
Subject: include/asm-x86/hw_irq_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/hw_irq_64.h | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/hw_irq_64.h b/include/asm-x86/hw_irq_64.h
index 312a58d6dac6..0062ef390f67 100644
--- a/include/asm-x86/hw_irq_64.h
+++ b/include/asm-x86/hw_irq_64.h
@@ -36,7 +36,7 @@
  * cleanup after irq migration.
  */
 #define IRQ_MOVE_CLEANUP_VECTOR	FIRST_EXTERNAL_VECTOR
- 
+
 /*
  * Vectors 0x30-0x3f are used for ISA interrupts.
  */
@@ -159,13 +159,12 @@ extern atomic_t irq_mis_count;
  *	SMP has a few special interrupts for IPI messages
  */
 
-#define BUILD_IRQ(nr) \
-asmlinkage void IRQ_NAME(nr); \
-__asm__( \
-"\n.p2align\n" \
-"IRQ" #nr "_interrupt:\n\t" \
-	"push $~(" #nr ") ; " \
-	"jmp common_interrupt");
+#define BUILD_IRQ(nr)				\
+	asmlinkage void IRQ_NAME(nr);		\
+	asm("\n.p2align\n"			\
+	    "IRQ" #nr "_interrupt:\n\t"		\
+	    "push $~(" #nr ") ; "		\
+	    "jmp common_interrupt");
 
 #define platform_legacy_irq(irq)	((irq) < 16)
 
-- 
cgit v1.2.3


From e574b023cf03f796727ea654acfc16a0df805086 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:17 -0700
Subject: include/asm-x86/hypertransport.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/hypertransport.h | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/hypertransport.h b/include/asm-x86/hypertransport.h
index c16c6ff4bdd7..d2bbd238b3e1 100644
--- a/include/asm-x86/hypertransport.h
+++ b/include/asm-x86/hypertransport.h
@@ -8,12 +8,14 @@
 #define HT_IRQ_LOW_BASE			0xf8000000
 
 #define HT_IRQ_LOW_VECTOR_SHIFT		16
-#define  HT_IRQ_LOW_VECTOR_MASK		0x00ff0000
-#define  HT_IRQ_LOW_VECTOR(v)		(((v) << HT_IRQ_LOW_VECTOR_SHIFT) & HT_IRQ_LOW_VECTOR_MASK)
+#define HT_IRQ_LOW_VECTOR_MASK		0x00ff0000
+#define HT_IRQ_LOW_VECTOR(v)						\
+	(((v) << HT_IRQ_LOW_VECTOR_SHIFT) & HT_IRQ_LOW_VECTOR_MASK)
 
 #define HT_IRQ_LOW_DEST_ID_SHIFT	8
-#define  HT_IRQ_LOW_DEST_ID_MASK	0x0000ff00
-#define  HT_IRQ_LOW_DEST_ID(v)		(((v) << HT_IRQ_LOW_DEST_ID_SHIFT) & HT_IRQ_LOW_DEST_ID_MASK)
+#define HT_IRQ_LOW_DEST_ID_MASK		0x0000ff00
+#define HT_IRQ_LOW_DEST_ID(v)						\
+	(((v) << HT_IRQ_LOW_DEST_ID_SHIFT) & HT_IRQ_LOW_DEST_ID_MASK)
 
 #define HT_IRQ_LOW_DM_PHYSICAL		0x0000000
 #define HT_IRQ_LOW_DM_LOGICAL		0x0000040
@@ -36,7 +38,8 @@
 
 
 #define HT_IRQ_HIGH_DEST_ID_SHIFT	0
-#define  HT_IRQ_HIGH_DEST_ID_MASK	0x00ffffff
-#define  HT_IRQ_HIGH_DEST_ID(v)		((((v) >> 8) << HT_IRQ_HIGH_DEST_ID_SHIFT) & HT_IRQ_HIGH_DEST_ID_MASK)
+#define HT_IRQ_HIGH_DEST_ID_MASK	0x00ffffff
+#define HT_IRQ_HIGH_DEST_ID(v)						\
+	((((v) >> 8) << HT_IRQ_HIGH_DEST_ID_SHIFT) & HT_IRQ_HIGH_DEST_ID_MASK)
 
 #endif /* ASM_HYPERTRANSPORT_H */
-- 
cgit v1.2.3


From affe66374ca16572fbb22cefe267d6072c49ce9d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:18 -0700
Subject: include/asm-x86/i387.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/i387.h | 23 +++++++++++++----------
 1 file changed, 13 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h
index f377b76b2f34..54522b814f1c 100644
--- a/include/asm-x86/i387.h
+++ b/include/asm-x86/i387.h
@@ -41,7 +41,7 @@ static inline void tolerant_fwait(void)
 {
 	asm volatile("1: fwait\n"
 		     "2:\n"
-		     _ASM_EXTABLE(1b,2b));
+		     _ASM_EXTABLE(1b, 2b));
 }
 
 static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
@@ -54,7 +54,7 @@ static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
 		     "3:  movl $-1,%[err]\n"
 		     "    jmp  2b\n"
 		     ".previous\n"
-		     _ASM_EXTABLE(1b,3b)
+		     _ASM_EXTABLE(1b, 3b)
 		     : [err] "=r" (err)
 #if 0 /* See comment in __save_init_fpu() below. */
 		     : [fx] "r" (fx), "m" (*fx), "0" (0));
@@ -76,11 +76,11 @@ static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
 static inline void clear_fpu_state(struct i387_fxsave_struct *fx)
 {
 	if (unlikely(fx->swd & X87_FSW_ES))
-		 asm volatile("fnclex");
+		asm volatile("fnclex");
 	alternative_input(ASM_NOP8 ASM_NOP2,
-		     "    emms\n"		/* clear stack tags */
-		     "    fildl %%gs:0",	/* load to clear state */
-		     X86_FEATURE_FXSAVE_LEAK);
+			  "    emms\n"		/* clear stack tags */
+			  "    fildl %%gs:0",	/* load to clear state */
+			  X86_FEATURE_FXSAVE_LEAK);
 }
 
 static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
@@ -93,14 +93,15 @@ static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
 		     "3:  movl $-1,%[err]\n"
 		     "    jmp  2b\n"
 		     ".previous\n"
-		     _ASM_EXTABLE(1b,3b)
+		     _ASM_EXTABLE(1b, 3b)
 		     : [err] "=r" (err), "=m" (*fx)
 #if 0 /* See comment in __fxsave_clear() below. */
 		     : [fx] "r" (fx), "0" (0));
 #else
 		     : [fx] "cdaSDb" (fx), "0" (0));
 #endif
-	if (unlikely(err) && __clear_user(fx, sizeof(struct i387_fxsave_struct)))
+	if (unlikely(err) &&
+	    __clear_user(fx, sizeof(struct i387_fxsave_struct)))
 		err = -EFAULT;
 	/* No need to clear here because the caller clears USED_MATH */
 	return err;
@@ -156,8 +157,10 @@ static inline int save_i387(struct _fpstate __user *buf)
 		return 0;
 	clear_used_math(); /* trigger finit */
 	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		err = save_i387_checking((struct i387_fxsave_struct __user *)buf);
-		if (err) return err;
+		err = save_i387_checking((struct i387_fxsave_struct __user *)
+					 buf);
+		if (err)
+			return err;
 		task_thread_info(tsk)->status &= ~TS_USEDFPU;
 		stts();
 	} else {
-- 
cgit v1.2.3


From ace4fdbb2fa842fc557bc0ae5909d790eb335ff8 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:19 -0700
Subject: include/asm-x86/i8259.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/i8259.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/i8259.h b/include/asm-x86/i8259.h
index e2650f21ca8d..45d4df3e51e6 100644
--- a/include/asm-x86/i8259.h
+++ b/include/asm-x86/i8259.h
@@ -5,7 +5,7 @@
 
 extern unsigned int cached_irq_mask;
 
-#define __byte(x,y)		(((unsigned char *) &(y))[x])
+#define __byte(x, y)		(((unsigned char *)&(y))[x])
 #define cached_master_mask	(__byte(0, cached_irq_mask))
 #define cached_slave_mask	(__byte(1, cached_irq_mask))
 
-- 
cgit v1.2.3


From 9bd73425142a610ae72c5e2b89e036e5214bb6ca Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:20 -0700
Subject: include/asm-x86/ia32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/ia32.h | 62 ++++++++++++++++++++++++--------------------------
 1 file changed, 30 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ia32.h b/include/asm-x86/ia32.h
index aa9733206e29..55d3abe5276f 100644
--- a/include/asm-x86/ia32.h
+++ b/include/asm-x86/ia32.h
@@ -14,19 +14,19 @@
 
 /* signal.h */
 struct sigaction32 {
-       unsigned int  sa_handler;	/* Really a pointer, but need to deal 
-					     with 32 bits */
-       unsigned int sa_flags;
-       unsigned int sa_restorer;	/* Another 32 bit pointer */
-       compat_sigset_t sa_mask;		/* A 32 bit mask */
+	unsigned int  sa_handler;	/* Really a pointer, but need to deal
+					   with 32 bits */
+	unsigned int sa_flags;
+	unsigned int sa_restorer;	/* Another 32 bit pointer */
+	compat_sigset_t sa_mask;	/* A 32 bit mask */
 };
 
 struct old_sigaction32 {
-       unsigned int  sa_handler;	/* Really a pointer, but need to deal 
-					     with 32 bits */
-       compat_old_sigset_t sa_mask;		/* A 32 bit mask */
-       unsigned int sa_flags;
-       unsigned int sa_restorer;	/* Another 32 bit pointer */
+	unsigned int  sa_handler;	/* Really a pointer, but need to deal
+					   with 32 bits */
+	compat_old_sigset_t sa_mask;	/* A 32 bit mask */
+	unsigned int sa_flags;
+	unsigned int sa_restorer;	/* Another 32 bit pointer */
 };
 
 typedef struct sigaltstack_ia32 {
@@ -65,7 +65,7 @@ struct stat64 {
 	long long		st_size;
 	unsigned int		st_blksize;
 
-	long long		st_blocks;/* Number 512-byte blocks allocated. */
+	long long		st_blocks;/* Number 512-byte blocks allocated */
 
 	unsigned 		st_atime;
 	unsigned 		st_atime_nsec;
@@ -77,13 +77,13 @@ struct stat64 {
 	unsigned long long	st_ino;
 } __attribute__((packed));
 
-typedef struct compat_siginfo{
+typedef struct compat_siginfo {
 	int si_signo;
 	int si_errno;
 	int si_code;
 
 	union {
-		int _pad[((128/sizeof(int)) - 3)];
+		int _pad[((128 / sizeof(int)) - 3)];
 
 		/* kill() */
 		struct {
@@ -129,28 +129,26 @@ typedef struct compat_siginfo{
 	} _sifields;
 } compat_siginfo_t;
 
-struct sigframe32
-{
-        u32 pretcode;
-        int sig;
-        struct sigcontext_ia32 sc;
-        struct _fpstate_ia32 fpstate;
-        unsigned int extramask[_COMPAT_NSIG_WORDS-1];
+struct sigframe32 {
+	u32 pretcode;
+	int sig;
+	struct sigcontext_ia32 sc;
+	struct _fpstate_ia32 fpstate;
+	unsigned int extramask[_COMPAT_NSIG_WORDS-1];
 };
 
-struct rt_sigframe32
-{
-        u32 pretcode;
-        int sig;
-        u32 pinfo;
-        u32 puc;
-        compat_siginfo_t info;
-        struct ucontext_ia32 uc;
-        struct _fpstate_ia32 fpstate;
+struct rt_sigframe32 {
+	u32 pretcode;
+	int sig;
+	u32 pinfo;
+	u32 puc;
+	compat_siginfo_t info;
+	struct ucontext_ia32 uc;
+	struct _fpstate_ia32 fpstate;
 };
 
 struct ustat32 {
-	__u32	f_tfree;
+	__u32			f_tfree;
 	compat_ino_t		f_tinode;
 	char			f_fname[6];
 	char			f_fpack[6];
@@ -168,5 +166,5 @@ extern void ia32_pick_mmap_layout(struct mm_struct *mm);
 #endif
 
 #endif /* !CONFIG_IA32_SUPPORT */
- 
-#endif 
+
+#endif
-- 
cgit v1.2.3


From dcd215c9089c1203eed6f84dde2eb2cc0bbf9501 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:22 -0700
Subject: include/asm-x86/io_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/io_32.h | 137 +++++++++++++++++++++++++++++-------------------
 1 file changed, 83 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index d4d8fbd9378c..509045f5fda2 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -65,14 +65,14 @@
  *
  *	The returned physical address is the physical (CPU) mapping for
  *	the memory address given. It is only valid to use this function on
- *	addresses directly mapped or allocated via kmalloc. 
+ *	addresses directly mapped or allocated via kmalloc.
  *
  *	This function does not give bus mappings for DMA transfers. In
  *	almost all conceivable cases a device driver should not be using
  *	this function
  */
- 
-static inline unsigned long virt_to_phys(volatile void * address)
+
+static inline unsigned long virt_to_phys(volatile void *address)
 {
 	return __pa(address);
 }
@@ -90,7 +90,7 @@ static inline unsigned long virt_to_phys(volatile void * address)
  *	this function
  */
 
-static inline void * phys_to_virt(unsigned long address)
+static inline void *phys_to_virt(unsigned long address)
 {
 	return __va(address);
 }
@@ -169,16 +169,19 @@ extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
 
 static inline unsigned char readb(const volatile void __iomem *addr)
 {
-	return *(volatile unsigned char __force *) addr;
+	return *(volatile unsigned char __force *)addr;
 }
+
 static inline unsigned short readw(const volatile void __iomem *addr)
 {
-	return *(volatile unsigned short __force *) addr;
+	return *(volatile unsigned short __force *)addr;
 }
+
 static inline unsigned int readl(const volatile void __iomem *addr)
 {
 	return *(volatile unsigned int __force *) addr;
 }
+
 #define readb_relaxed(addr) readb(addr)
 #define readw_relaxed(addr) readw(addr)
 #define readl_relaxed(addr) readl(addr)
@@ -188,15 +191,17 @@ static inline unsigned int readl(const volatile void __iomem *addr)
 
 static inline void writeb(unsigned char b, volatile void __iomem *addr)
 {
-	*(volatile unsigned char __force *) addr = b;
+	*(volatile unsigned char __force *)addr = b;
 }
+
 static inline void writew(unsigned short b, volatile void __iomem *addr)
 {
-	*(volatile unsigned short __force *) addr = b;
+	*(volatile unsigned short __force *)addr = b;
 }
+
 static inline void writel(unsigned int b, volatile void __iomem *addr)
 {
-	*(volatile unsigned int __force *) addr = b;
+	*(volatile unsigned int __force *)addr = b;
 }
 #define __raw_writeb writeb
 #define __raw_writew writew
@@ -239,12 +244,12 @@ memcpy_toio(volatile void __iomem *dst, const void *src, int count)
  *	1. Out of order aware processors
  *	2. Accidentally out of order processors (PPro errata #51)
  */
- 
+
 #if defined(CONFIG_X86_OOSTORE) || defined(CONFIG_X86_PPRO_FENCE)
 
 static inline void flush_write_buffers(void)
 {
-	__asm__ __volatile__ ("lock; addl $0,0(%%esp)": : :"memory");
+	asm volatile("lock; addl $0,0(%%esp)": : :"memory");
 }
 
 #else
@@ -264,7 +269,8 @@ extern void io_delay_init(void);
 #include <asm/paravirt.h>
 #else
 
-static inline void slow_down_io(void) {
+static inline void slow_down_io(void)
+{
 	native_io_delay();
 #ifdef REALLY_SLOW_IO
 	native_io_delay();
@@ -275,51 +281,74 @@ static inline void slow_down_io(void) {
 
 #endif
 
-#define __BUILDIO(bwl,bw,type) \
-static inline void out##bwl(unsigned type value, int port) { \
-	out##bwl##_local(value, port); \
-} \
-static inline unsigned type in##bwl(int port) { \
-	return in##bwl##_local(port); \
+#define __BUILDIO(bwl, bw, type)				\
+static inline void out##bwl(unsigned type value, int port)	\
+{								\
+	out##bwl##_local(value, port);				\
+}								\
+								\
+static inline unsigned type in##bwl(int port)			\
+{								\
+	return in##bwl##_local(port);				\
 }
 
-#define BUILDIO(bwl,bw,type) \
-static inline void out##bwl##_local(unsigned type value, int port) { \
-	__asm__ __volatile__("out" #bwl " %" #bw "0, %w1" : : "a"(value), "Nd"(port)); \
-} \
-static inline unsigned type in##bwl##_local(int port) { \
-	unsigned type value; \
-	__asm__ __volatile__("in" #bwl " %w1, %" #bw "0" : "=a"(value) : "Nd"(port)); \
-	return value; \
-} \
-static inline void out##bwl##_local_p(unsigned type value, int port) { \
-	out##bwl##_local(value, port); \
-	slow_down_io(); \
-} \
-static inline unsigned type in##bwl##_local_p(int port) { \
-	unsigned type value = in##bwl##_local(port); \
-	slow_down_io(); \
-	return value; \
-} \
-__BUILDIO(bwl,bw,type) \
-static inline void out##bwl##_p(unsigned type value, int port) { \
-	out##bwl(value, port); \
-	slow_down_io(); \
-} \
-static inline unsigned type in##bwl##_p(int port) { \
-	unsigned type value = in##bwl(port); \
-	slow_down_io(); \
-	return value; \
-} \
-static inline void outs##bwl(int port, const void *addr, unsigned long count) { \
-	__asm__ __volatile__("rep; outs" #bwl : "+S"(addr), "+c"(count) : "d"(port)); \
-} \
-static inline void ins##bwl(int port, void *addr, unsigned long count) { \
-	__asm__ __volatile__("rep; ins" #bwl : "+D"(addr), "+c"(count) : "d"(port)); \
+#define BUILDIO(bwl, bw, type)						\
+static inline void out##bwl##_local(unsigned type value, int port)	\
+{									\
+	asm volatile("out" #bwl " %" #bw "0, %w1"		\
+		     : : "a"(value), "Nd"(port));			\
+}									\
+									\
+static inline unsigned type in##bwl##_local(int port)			\
+{									\
+	unsigned type value;						\
+	asm volatile("in" #bwl " %w1, %" #bw "0"		\
+		     : "=a"(value) : "Nd"(port));			\
+	return value;							\
+}									\
+									\
+static inline void out##bwl##_local_p(unsigned type value, int port)	\
+{									\
+	out##bwl##_local(value, port);					\
+	slow_down_io();							\
+}									\
+									\
+static inline unsigned type in##bwl##_local_p(int port)			\
+{									\
+	unsigned type value = in##bwl##_local(port);			\
+	slow_down_io();							\
+	return value;							\
+}									\
+									\
+__BUILDIO(bwl, bw, type)						\
+									\
+static inline void out##bwl##_p(unsigned type value, int port)		\
+{									\
+	out##bwl(value, port);						\
+	slow_down_io();							\
+}									\
+									\
+static inline unsigned type in##bwl##_p(int port)			\
+{									\
+	unsigned type value = in##bwl(port);				\
+	slow_down_io();							\
+	return value;							\
+}									\
+									\
+static inline void outs##bwl(int port, const void *addr, unsigned long count) \
+{									\
+	asm volatile("rep; outs" #bwl					\
+		     : "+S"(addr), "+c"(count) : "d"(port));		\
+}									\
+									\
+static inline void ins##bwl(int port, void *addr, unsigned long count)	\
+{									\
+	asm volatile("rep; ins" #bwl					\
+		     : "+D"(addr), "+c"(count) : "d"(port));		\
 }
 
-BUILDIO(b,b,char)
-BUILDIO(w,w,short)
-BUILDIO(l,,int)
+BUILDIO(b, b, char)
+BUILDIO(w, w, short)
+BUILDIO(l, , int)
 
 #endif
-- 
cgit v1.2.3


From 126f5f35a36b5734026db58263fc1d60e4ae466f Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:23 -0700
Subject: include/asm-x86/io_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/io_64.h | 110 ++++++++++++++++++++++++++++++------------------
 1 file changed, 68 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index db0be2011a3c..c2f5eef47b88 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -58,60 +58,75 @@ static inline void slow_down_io(void)
 /*
  * Talk about misusing macros..
  */
-#define __OUT1(s,x) \
+#define __OUT1(s, x)							\
 static inline void out##s(unsigned x value, unsigned short port) {
 
-#define __OUT2(s,s1,s2) \
-__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1"
+#define __OUT2(s, s1, s2)				\
+asm volatile ("out" #s " %" s1 "0,%" s2 "1"
 
 #ifndef REALLY_SLOW_IO
 #define REALLY_SLOW_IO
 #define UNSET_REALLY_SLOW_IO
 #endif
 
-#define __OUT(s,s1,x) \
-__OUT1(s,x) __OUT2(s,s1,"w") : : "a" (value), "Nd" (port)); } \
-__OUT1(s##_p, x) __OUT2(s, s1, "w") : : "a" (value), "Nd" (port)); \
-		slow_down_io(); }
+#define __OUT(s, s1, x)							\
+	__OUT1(s, x) __OUT2(s, s1, "w") : : "a" (value), "Nd" (port));	\
+	}								\
+	__OUT1(s##_p, x) __OUT2(s, s1, "w") : : "a" (value), "Nd" (port)); \
+	slow_down_io();							\
+}
 
-#define __IN1(s) \
-static inline RETURN_TYPE in##s(unsigned short port) { RETURN_TYPE _v;
+#define __IN1(s)							\
+static inline RETURN_TYPE in##s(unsigned short port)			\
+{									\
+	RETURN_TYPE _v;
 
-#define __IN2(s,s1,s2) \
-__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0"
+#define __IN2(s, s1, s2)						\
+	asm volatile ("in" #s " %" s2 "1,%" s1 "0"
 
-#define __IN(s,s1,i...) \
-__IN1(s) __IN2(s, s1, "w") : "=a" (_v) : "Nd" (port), ##i); return _v; } \
-__IN1(s##_p) __IN2(s, s1, "w") : "=a" (_v) : "Nd" (port), ##i);	  \
-				slow_down_io(); return _v; }
+#define __IN(s, s1, i...)						\
+	__IN1(s) __IN2(s, s1, "w") : "=a" (_v) : "Nd" (port), ##i);	\
+	return _v;							\
+	}								\
+	__IN1(s##_p) __IN2(s, s1, "w") : "=a" (_v) : "Nd" (port), ##i);	\
+	slow_down_io(); \
+	return _v; }
 
 #ifdef UNSET_REALLY_SLOW_IO
 #undef REALLY_SLOW_IO
 #endif
 
-#define __INS(s) \
-static inline void ins##s(unsigned short port, void * addr, unsigned long count) \
-{ __asm__ __volatile__ ("rep ; ins" #s \
-: "=D" (addr), "=c" (count) : "d" (port),"0" (addr),"1" (count)); }
+#define __INS(s)							\
+static inline void ins##s(unsigned short port, void *addr,		\
+			  unsigned long count)				\
+{									\
+	asm volatile ("rep ; ins" #s					\
+		      : "=D" (addr), "=c" (count)			\
+		      : "d" (port), "0" (addr), "1" (count));		\
+}
 
-#define __OUTS(s) \
-static inline void outs##s(unsigned short port, const void * addr, unsigned long count) \
-{ __asm__ __volatile__ ("rep ; outs" #s \
-: "=S" (addr), "=c" (count) : "d" (port),"0" (addr),"1" (count)); }
+#define __OUTS(s)							\
+static inline void outs##s(unsigned short port, const void *addr,	\
+			   unsigned long count)				\
+{									\
+	asm volatile ("rep ; outs" #s					\
+		      : "=S" (addr), "=c" (count)			\
+		      : "d" (port), "0" (addr), "1" (count));		\
+}
 
 #define RETURN_TYPE unsigned char
-__IN(b,"")
+__IN(b, "")
 #undef RETURN_TYPE
 #define RETURN_TYPE unsigned short
-__IN(w,"")
+__IN(w, "")
 #undef RETURN_TYPE
 #define RETURN_TYPE unsigned int
-__IN(l,"")
+__IN(l, "")
 #undef RETURN_TYPE
 
-__OUT(b,"b",char)
-__OUT(w,"w",short)
-__OUT(l,,int)
+__OUT(b, "b", char)
+__OUT(w, "w", short)
+__OUT(l, , int)
 
 __INS(b)
 __INS(w)
@@ -132,12 +147,12 @@ __OUTS(l)
  * Change virtual addresses to physical addresses and vv.
  * These are pretty trivial
  */
-static inline unsigned long virt_to_phys(volatile void * address)
+static inline unsigned long virt_to_phys(volatile void *address)
 {
 	return __pa(address);
 }
 
-static inline void * phys_to_virt(unsigned long address)
+static inline void *phys_to_virt(unsigned long address)
 {
 	return __va(address);
 }
@@ -200,18 +215,22 @@ static inline __u8 __readb(const volatile void __iomem *addr)
 {
 	return *(__force volatile __u8 *)addr;
 }
+
 static inline __u16 __readw(const volatile void __iomem *addr)
 {
 	return *(__force volatile __u16 *)addr;
 }
+
 static __always_inline __u32 __readl(const volatile void __iomem *addr)
 {
 	return *(__force volatile __u32 *)addr;
 }
+
 static inline __u64 __readq(const volatile void __iomem *addr)
 {
 	return *(__force volatile __u64 *)addr;
 }
+
 #define readb(x) __readb(x)
 #define readw(x) __readw(x)
 #define readl(x) __readl(x)
@@ -231,37 +250,44 @@ static inline void __writel(__u32 b, volatile void __iomem *addr)
 {
 	*(__force volatile __u32 *)addr = b;
 }
+
 static inline void __writeq(__u64 b, volatile void __iomem *addr)
 {
 	*(__force volatile __u64 *)addr = b;
 }
+
 static inline void __writeb(__u8 b, volatile void __iomem *addr)
 {
 	*(__force volatile __u8 *)addr = b;
 }
+
 static inline void __writew(__u16 b, volatile void __iomem *addr)
 {
 	*(__force volatile __u16 *)addr = b;
 }
-#define writeq(val,addr) __writeq((val),(addr))
-#define writel(val,addr) __writel((val),(addr))
-#define writew(val,addr) __writew((val),(addr))
-#define writeb(val,addr) __writeb((val),(addr))
+
+#define writeq(val, addr) __writeq((val), (addr))
+#define writel(val, addr) __writel((val), (addr))
+#define writew(val, addr) __writew((val), (addr))
+#define writeb(val, addr) __writeb((val), (addr))
 #define __raw_writeb writeb
 #define __raw_writew writew
 #define __raw_writel writel
 #define __raw_writeq writeq
 
-void __memcpy_fromio(void*,unsigned long,unsigned);
-void __memcpy_toio(unsigned long,const void*,unsigned);
+void __memcpy_fromio(void *, unsigned long, unsigned);
+void __memcpy_toio(unsigned long, const void *, unsigned);
 
-static inline void memcpy_fromio(void *to, const volatile void __iomem *from, unsigned len)
+static inline void memcpy_fromio(void *to, const volatile void __iomem *from,
+				 unsigned len)
 {
-	__memcpy_fromio(to,(unsigned long)from,len);
+	__memcpy_fromio(to, (unsigned long)from, len);
 }
-static inline void memcpy_toio(volatile void __iomem *to, const void *from, unsigned len)
+
+static inline void memcpy_toio(volatile void __iomem *to, const void *from,
+			       unsigned len)
 {
-	__memcpy_toio((unsigned long)to,from,len);
+	__memcpy_toio((unsigned long)to, from, len);
 }
 
 void memset_io(volatile void __iomem *a, int b, size_t c);
@@ -276,7 +302,7 @@ void memset_io(volatile void __iomem *a, int b, size_t c);
  */
 #define __ISA_IO_base ((char __iomem *)(PAGE_OFFSET))
 
-#define flush_write_buffers() 
+#define flush_write_buffers()
 
 extern int iommu_bio_merge;
 #define BIO_VMERGE_BOUNDARY iommu_bio_merge
-- 
cgit v1.2.3


From 88b4f6e98f79de31f49036da5bb1482872d512b5 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:24 -0700
Subject: include/asm-x86/ioctls.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/ioctls.h | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ioctls.h b/include/asm-x86/ioctls.h
index 93c894dc5154..c0c338bd4068 100644
--- a/include/asm-x86/ioctls.h
+++ b/include/asm-x86/ioctls.h
@@ -47,12 +47,13 @@
 #define TIOCSBRK	0x5427  /* BSD compatibility */
 #define TIOCCBRK	0x5428  /* BSD compatibility */
 #define TIOCGSID	0x5429  /* Return the session ID of FD */
-#define TCGETS2		_IOR('T',0x2A, struct termios2)
-#define TCSETS2		_IOW('T',0x2B, struct termios2)
-#define TCSETSW2	_IOW('T',0x2C, struct termios2)
-#define TCSETSF2	_IOW('T',0x2D, struct termios2)
-#define TIOCGPTN	_IOR('T',0x30, unsigned int) /* Get Pty Number (of pty-mux device) */
-#define TIOCSPTLCK	_IOW('T',0x31, int)  /* Lock/unlock Pty */
+#define TCGETS2		_IOR('T', 0x2A, struct termios2)
+#define TCSETS2		_IOW('T', 0x2B, struct termios2)
+#define TCSETSW2	_IOW('T', 0x2C, struct termios2)
+#define TCSETSF2	_IOW('T', 0x2D, struct termios2)
+#define TIOCGPTN	_IOR('T', 0x30, unsigned int)
+				/* Get Pty Number (of pty-mux device) */
+#define TIOCSPTLCK	_IOW('T', 0x31, int)  /* Lock/unlock Pty */
 
 #define FIONCLEX	0x5450
 #define FIOCLEX		0x5451
-- 
cgit v1.2.3


From 1774a5bed33d85f250657f397b359b9837390a90 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:25 -0700
Subject: include/asm-x86/io.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/io.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h
index 599cad3505c1..7b292d386713 100644
--- a/include/asm-x86/io.h
+++ b/include/asm-x86/io.h
@@ -7,5 +7,5 @@
 #endif
 extern int ioremap_change_attr(unsigned long vaddr, unsigned long size,
 				unsigned long prot_val);
-extern void __iomem * ioremap_wc(unsigned long offset, unsigned long size);
+extern void __iomem *ioremap_wc(unsigned long offset, unsigned long size);
 
-- 
cgit v1.2.3


From 5ca22aaad9ea078306537b2ef6fb788d931e8502 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:26 -0700
Subject: include/asm-x86/ipcbuf.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/ipcbuf.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ipcbuf.h b/include/asm-x86/ipcbuf.h
index 2adf8b39a40b..ee678fd51594 100644
--- a/include/asm-x86/ipcbuf.h
+++ b/include/asm-x86/ipcbuf.h
@@ -11,8 +11,7 @@
  * - 2 miscellaneous 32-bit values
  */
 
-struct ipc64_perm
-{
+struct ipc64_perm {
 	__kernel_key_t		key;
 	__kernel_uid32_t	uid;
 	__kernel_gid32_t	gid;
-- 
cgit v1.2.3


From 061b3d90bc3490af41c6756189b59713cd9ecaee Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:27 -0700
Subject: include/asm-x86/ipi.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/ipi.h | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ipi.h b/include/asm-x86/ipi.h
index 6d011bd6067d..ecc80f341f37 100644
--- a/include/asm-x86/ipi.h
+++ b/include/asm-x86/ipi.h
@@ -27,7 +27,8 @@
  * We use 'broadcast', CPU->CPU IPIs and self-IPIs too.
  */
 
-static inline unsigned int __prepare_ICR (unsigned int shortcut, int vector, unsigned int dest)
+static inline unsigned int __prepare_ICR(unsigned int shortcut, int vector,
+					 unsigned int dest)
 {
 	unsigned int icr = shortcut | dest;
 
@@ -42,12 +43,13 @@ static inline unsigned int __prepare_ICR (unsigned int shortcut, int vector, uns
 	return icr;
 }
 
-static inline int __prepare_ICR2 (unsigned int mask)
+static inline int __prepare_ICR2(unsigned int mask)
 {
 	return SET_APIC_DEST_FIELD(mask);
 }
 
-static inline void __send_IPI_shortcut(unsigned int shortcut, int vector, unsigned int dest)
+static inline void __send_IPI_shortcut(unsigned int shortcut, int vector,
+				       unsigned int dest)
 {
 	/*
 	 * Subtle. In the case of the 'never do double writes' workaround
@@ -78,7 +80,8 @@ static inline void __send_IPI_shortcut(unsigned int shortcut, int vector, unsign
  * This is used to send an IPI with no shorthand notation (the destination is
  * specified in bits 56 to 63 of the ICR).
  */
-static inline void __send_IPI_dest_field(unsigned int mask, int vector, unsigned int dest)
+static inline void __send_IPI_dest_field(unsigned int mask, int vector,
+					 unsigned int dest)
 {
 	unsigned long cfg;
 
-- 
cgit v1.2.3


From f4964d2ac51fcc8cd56975139422cdbb3a1e66dc Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:28 -0700
Subject: include/asm-x86/irq_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/irq_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/irq_32.h b/include/asm-x86/irq_32.h
index aca9c96e8e6b..0b79f3185243 100644
--- a/include/asm-x86/irq_32.h
+++ b/include/asm-x86/irq_32.h
@@ -15,7 +15,7 @@
 #include "irq_vectors.h"
 #include <asm/thread_info.h>
 
-static __inline__ int irq_canonicalize(int irq)
+static inline int irq_canonicalize(int irq)
 {
 	return ((irq == 2) ? 9 : irq);
 }
-- 
cgit v1.2.3


From 3ff3522497d7ad383df9770e812fbaf75d19b214 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:29 -0700
Subject: include/asm-x86/irq_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/irq_64.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/irq_64.h b/include/asm-x86/irq_64.h
index 5006c6e75656..083d35a62c94 100644
--- a/include/asm-x86/irq_64.h
+++ b/include/asm-x86/irq_64.h
@@ -31,10 +31,10 @@
 
 #define FIRST_SYSTEM_VECTOR	0xef   /* duplicated in hw_irq.h */
 
-#define NR_IRQS (NR_VECTORS + (32 *NR_CPUS))
+#define NR_IRQS (NR_VECTORS + (32 * NR_CPUS))
 #define NR_IRQ_VECTORS NR_IRQS
 
-static __inline__ int irq_canonicalize(int irq)
+static inline int irq_canonicalize(int irq)
 {
 	return ((irq == 2) ? 9 : irq);
 }
-- 
cgit v1.2.3


From cf7f7191cf20011e47243b594e433275a6db811b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:30 -0700
Subject: include/asm-x86/irqflags.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/irqflags.h | 30 +++++++++++++-----------------
 1 file changed, 13 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h
index 92021c1ffa3a..c242527f970e 100644
--- a/include/asm-x86/irqflags.h
+++ b/include/asm-x86/irqflags.h
@@ -12,25 +12,21 @@ static inline unsigned long native_save_fl(void)
 {
 	unsigned long flags;
 
-	__asm__ __volatile__(
-		"# __raw_save_flags\n\t"
-		"pushf ; pop %0"
-		: "=g" (flags)
-		: /* no input */
-		: "memory"
-	);
+	asm volatile("# __raw_save_flags\n\t"
+		     "pushf ; pop %0"
+		     : "=g" (flags)
+		     : /* no input */
+		     : "memory");
 
 	return flags;
 }
 
 static inline void native_restore_fl(unsigned long flags)
 {
-	__asm__ __volatile__(
-		"push %0 ; popf"
-		: /* no output */
-		:"g" (flags)
-		:"memory", "cc"
-	);
+	asm volatile("push %0 ; popf"
+		     : /* no output */
+		     :"g" (flags)
+		     :"memory", "cc");
 }
 
 static inline void native_irq_disable(void)
@@ -131,11 +127,11 @@ static inline unsigned long __raw_local_irq_save(void)
 #endif /* CONFIG_PARAVIRT */
 
 #ifndef __ASSEMBLY__
-#define raw_local_save_flags(flags) \
-		do { (flags) = __raw_local_save_flags(); } while (0)
+#define raw_local_save_flags(flags)				\
+	do { (flags) = __raw_local_save_flags(); } while (0)
 
-#define raw_local_irq_save(flags) \
-		do { (flags) = __raw_local_irq_save(); } while (0)
+#define raw_local_irq_save(flags)				\
+	do { (flags) = __raw_local_irq_save(); } while (0)
 
 static inline int raw_irqs_disabled_flags(unsigned long flags)
 {
-- 
cgit v1.2.3


From f461f1372cef1853534df2115f9ff5b7fbfc6958 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:31 -0700
Subject: include/asm-x86/kdebug.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/kdebug.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index 99dcbafa1511..0c4175390dab 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -23,12 +23,12 @@ enum die_val {
 };
 
 extern void printk_address(unsigned long address, int reliable);
-extern void die(const char *,struct pt_regs *,long);
+extern void die(const char *, struct pt_regs *,long);
 extern int __must_check __die(const char *, struct pt_regs *, long);
 extern void show_registers(struct pt_regs *regs);
 extern void __show_registers(struct pt_regs *, int all);
 extern void show_trace(struct task_struct *t, struct pt_regs *regs,
-			unsigned long *sp, unsigned long bp);
+		       unsigned long *sp, unsigned long bp);
 extern void __show_regs(struct pt_regs *regs);
 extern void show_regs(struct pt_regs *regs);
 extern unsigned long oops_begin(void);
-- 
cgit v1.2.3


From b69a3f9dc0bbdbf9278ac5bc8d4b6347c11a701b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:32 -0700
Subject: include/asm-x86/kexec.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/kexec.h | 71 ++++++++++++++++++++++++-------------------------
 1 file changed, 35 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kexec.h b/include/asm-x86/kexec.h
index c90d3c77afc2..8f855a15f64d 100644
--- a/include/asm-x86/kexec.h
+++ b/include/asm-x86/kexec.h
@@ -94,10 +94,9 @@ static inline void crash_fixup_ss_esp(struct pt_regs *newregs,
 {
 #ifdef CONFIG_X86_32
 	newregs->sp = (unsigned long)&(oldregs->sp);
-	__asm__ __volatile__(
-			"xorl %%eax, %%eax\n\t"
-			"movw %%ss, %%ax\n\t"
-			:"=a"(newregs->ss));
+	asm volatile("xorl %%eax, %%eax\n\t"
+		     "movw %%ss, %%ax\n\t"
+		     :"=a"(newregs->ss));
 #endif
 }
 
@@ -114,39 +113,39 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
 		crash_fixup_ss_esp(newregs, oldregs);
 	} else {
 #ifdef CONFIG_X86_32
-		__asm__ __volatile__("movl %%ebx,%0" : "=m"(newregs->bx));
-		__asm__ __volatile__("movl %%ecx,%0" : "=m"(newregs->cx));
-		__asm__ __volatile__("movl %%edx,%0" : "=m"(newregs->dx));
-		__asm__ __volatile__("movl %%esi,%0" : "=m"(newregs->si));
-		__asm__ __volatile__("movl %%edi,%0" : "=m"(newregs->di));
-		__asm__ __volatile__("movl %%ebp,%0" : "=m"(newregs->bp));
-		__asm__ __volatile__("movl %%eax,%0" : "=m"(newregs->ax));
-		__asm__ __volatile__("movl %%esp,%0" : "=m"(newregs->sp));
-		__asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
-		__asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
-		__asm__ __volatile__("movl %%ds, %%eax;" :"=a"(newregs->ds));
-		__asm__ __volatile__("movl %%es, %%eax;" :"=a"(newregs->es));
-		__asm__ __volatile__("pushfl; popl %0" :"=m"(newregs->flags));
+		asm volatile("movl %%ebx,%0" : "=m"(newregs->bx));
+		asm volatile("movl %%ecx,%0" : "=m"(newregs->cx));
+		asm volatile("movl %%edx,%0" : "=m"(newregs->dx));
+		asm volatile("movl %%esi,%0" : "=m"(newregs->si));
+		asm volatile("movl %%edi,%0" : "=m"(newregs->di));
+		asm volatile("movl %%ebp,%0" : "=m"(newregs->bp));
+		asm volatile("movl %%eax,%0" : "=m"(newregs->ax));
+		asm volatile("movl %%esp,%0" : "=m"(newregs->sp));
+		asm volatile("movl %%ss, %%eax;" :"=a"(newregs->ss));
+		asm volatile("movl %%cs, %%eax;" :"=a"(newregs->cs));
+		asm volatile("movl %%ds, %%eax;" :"=a"(newregs->ds));
+		asm volatile("movl %%es, %%eax;" :"=a"(newregs->es));
+		asm volatile("pushfl; popl %0" :"=m"(newregs->flags));
 #else
-		__asm__ __volatile__("movq %%rbx,%0" : "=m"(newregs->bx));
-		__asm__ __volatile__("movq %%rcx,%0" : "=m"(newregs->cx));
-		__asm__ __volatile__("movq %%rdx,%0" : "=m"(newregs->dx));
-		__asm__ __volatile__("movq %%rsi,%0" : "=m"(newregs->si));
-		__asm__ __volatile__("movq %%rdi,%0" : "=m"(newregs->di));
-		__asm__ __volatile__("movq %%rbp,%0" : "=m"(newregs->bp));
-		__asm__ __volatile__("movq %%rax,%0" : "=m"(newregs->ax));
-		__asm__ __volatile__("movq %%rsp,%0" : "=m"(newregs->sp));
-		__asm__ __volatile__("movq %%r8,%0" : "=m"(newregs->r8));
-		__asm__ __volatile__("movq %%r9,%0" : "=m"(newregs->r9));
-		__asm__ __volatile__("movq %%r10,%0" : "=m"(newregs->r10));
-		__asm__ __volatile__("movq %%r11,%0" : "=m"(newregs->r11));
-		__asm__ __volatile__("movq %%r12,%0" : "=m"(newregs->r12));
-		__asm__ __volatile__("movq %%r13,%0" : "=m"(newregs->r13));
-		__asm__ __volatile__("movq %%r14,%0" : "=m"(newregs->r14));
-		__asm__ __volatile__("movq %%r15,%0" : "=m"(newregs->r15));
-		__asm__ __volatile__("movl %%ss, %%eax;" :"=a"(newregs->ss));
-		__asm__ __volatile__("movl %%cs, %%eax;" :"=a"(newregs->cs));
-		__asm__ __volatile__("pushfq; popq %0" :"=m"(newregs->flags));
+		asm volatile("movq %%rbx,%0" : "=m"(newregs->bx));
+		asm volatile("movq %%rcx,%0" : "=m"(newregs->cx));
+		asm volatile("movq %%rdx,%0" : "=m"(newregs->dx));
+		asm volatile("movq %%rsi,%0" : "=m"(newregs->si));
+		asm volatile("movq %%rdi,%0" : "=m"(newregs->di));
+		asm volatile("movq %%rbp,%0" : "=m"(newregs->bp));
+		asm volatile("movq %%rax,%0" : "=m"(newregs->ax));
+		asm volatile("movq %%rsp,%0" : "=m"(newregs->sp));
+		asm volatile("movq %%r8,%0" : "=m"(newregs->r8));
+		asm volatile("movq %%r9,%0" : "=m"(newregs->r9));
+		asm volatile("movq %%r10,%0" : "=m"(newregs->r10));
+		asm volatile("movq %%r11,%0" : "=m"(newregs->r11));
+		asm volatile("movq %%r12,%0" : "=m"(newregs->r12));
+		asm volatile("movq %%r13,%0" : "=m"(newregs->r13));
+		asm volatile("movq %%r14,%0" : "=m"(newregs->r14));
+		asm volatile("movq %%r15,%0" : "=m"(newregs->r15));
+		asm volatile("movl %%ss, %%eax;" :"=a"(newregs->ss));
+		asm volatile("movl %%cs, %%eax;" :"=a"(newregs->cs));
+		asm volatile("pushfq; popq %0" :"=m"(newregs->flags));
 #endif
 		newregs->ip = (unsigned long)current_text_addr();
 	}
-- 
cgit v1.2.3


From 864dfa13bed444d2c6d76d584e81713405d60082 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:33 -0700
Subject: include/asm-x86/kprobes.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/kprobes.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h
index 61ad7b5d142e..54980b0b3892 100644
--- a/include/asm-x86/kprobes.h
+++ b/include/asm-x86/kprobes.h
@@ -35,12 +35,12 @@ typedef u8 kprobe_opcode_t;
 #define RELATIVEJUMP_INSTRUCTION 0xe9
 #define MAX_INSN_SIZE 16
 #define MAX_STACK_SIZE 64
-#define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
-	(((unsigned long)current_thread_info()) + THREAD_SIZE \
-	 - (unsigned long)(ADDR))) \
-	? (MAX_STACK_SIZE) \
-	: (((unsigned long)current_thread_info()) + THREAD_SIZE \
-	   - (unsigned long)(ADDR)))
+#define MIN_STACK_SIZE(ADDR)					       \
+	(((MAX_STACK_SIZE) < (((unsigned long)current_thread_info()) + \
+			      THREAD_SIZE - (unsigned long)(ADDR)))    \
+	 ? (MAX_STACK_SIZE)					       \
+	 : (((unsigned long)current_thread_info()) +		       \
+	    THREAD_SIZE - (unsigned long)(ADDR)))
 
 #define flush_insn_slot(p)	do { } while (0)
 
-- 
cgit v1.2.3


From 7d76b4d3767324fa44ac73032d6eeb428e46f004 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:34 -0700
Subject: include/asm-x86/kvm_host.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/kvm_host.h | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 4702b04b979a..68ee390b2844 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -22,15 +22,16 @@
 
 #define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1)
 #define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD))
-#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS|0xFFFFFF0000000000ULL)
+#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS |	\
+				  0xFFFFFF0000000000ULL)
 
-#define KVM_GUEST_CR0_MASK \
+#define KVM_GUEST_CR0_MASK				   \
 	(X86_CR0_PG | X86_CR0_PE | X86_CR0_WP | X86_CR0_NE \
 	 | X86_CR0_NW | X86_CR0_CD)
-#define KVM_VM_CR0_ALWAYS_ON \
+#define KVM_VM_CR0_ALWAYS_ON						\
 	(X86_CR0_PG | X86_CR0_PE | X86_CR0_WP | X86_CR0_NE | X86_CR0_TS \
 	 | X86_CR0_MP)
-#define KVM_GUEST_CR4_MASK \
+#define KVM_GUEST_CR4_MASK						\
 	(X86_CR4_VME | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE | X86_CR4_VMXE)
 #define KVM_PMODE_VM_CR4_ALWAYS_ON (X86_CR4_PAE | X86_CR4_VMXE)
 #define KVM_RMODE_VM_CR4_ALWAYS_ON (X86_CR4_VME | X86_CR4_PAE | X86_CR4_VMXE)
@@ -133,12 +134,12 @@ struct kvm_pte_chain {
 union kvm_mmu_page_role {
 	unsigned word;
 	struct {
-		unsigned glevels : 4;
-		unsigned level : 4;
-		unsigned quadrant : 2;
-		unsigned pad_for_nice_hex_output : 6;
-		unsigned metaphysical : 1;
-		unsigned access : 3;
+		unsigned glevels:4;
+		unsigned level:4;
+		unsigned quadrant:2;
+		unsigned pad_for_nice_hex_output:6;
+		unsigned metaphysical:1;
+		unsigned access:3;
 	};
 };
 
@@ -606,6 +607,7 @@ static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
 #define TSS_BASE_SIZE 0x68
 #define TSS_IOPB_SIZE (65536 / 8)
 #define TSS_REDIRECTION_SIZE (256 / 8)
-#define RMODE_TSS_SIZE (TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
+#define RMODE_TSS_SIZE							\
+	(TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
 
 #endif
-- 
cgit v1.2.3


From 0c7825e64d377409deaaf73cd6311da9df310db3 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:35 -0700
Subject: include/asm-x86/kvm_x86_emulate.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/kvm_x86_emulate.h | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/kvm_x86_emulate.h b/include/asm-x86/kvm_x86_emulate.h
index 7db91b9bdcd4..d6337f941c98 100644
--- a/include/asm-x86/kvm_x86_emulate.h
+++ b/include/asm-x86/kvm_x86_emulate.h
@@ -68,10 +68,10 @@ struct x86_emulate_ops {
 	 *  @val:   [OUT] Value read from memory, zero-extended to 'u_long'.
 	 *  @bytes: [IN ] Number of bytes to read from memory.
 	 */
-	int (*read_emulated) (unsigned long addr,
-			      void *val,
-			      unsigned int bytes,
-			      struct kvm_vcpu *vcpu);
+	int (*read_emulated)(unsigned long addr,
+			     void *val,
+			     unsigned int bytes,
+			     struct kvm_vcpu *vcpu);
 
 	/*
 	 * write_emulated: Read bytes from emulated/special memory area.
@@ -80,10 +80,10 @@ struct x86_emulate_ops {
 	 *                required).
 	 *  @bytes: [IN ] Number of bytes to write to memory.
 	 */
-	int (*write_emulated) (unsigned long addr,
-			       const void *val,
-			       unsigned int bytes,
-			       struct kvm_vcpu *vcpu);
+	int (*write_emulated)(unsigned long addr,
+			      const void *val,
+			      unsigned int bytes,
+			      struct kvm_vcpu *vcpu);
 
 	/*
 	 * cmpxchg_emulated: Emulate an atomic (LOCKed) CMPXCHG operation on an
@@ -93,11 +93,11 @@ struct x86_emulate_ops {
 	 *  @new:   [IN ] Value to write to @addr.
 	 *  @bytes: [IN ] Number of bytes to access using CMPXCHG.
 	 */
-	int (*cmpxchg_emulated) (unsigned long addr,
-				 const void *old,
-				 const void *new,
-				 unsigned int bytes,
-				 struct kvm_vcpu *vcpu);
+	int (*cmpxchg_emulated)(unsigned long addr,
+				const void *old,
+				const void *new,
+				unsigned int bytes,
+				struct kvm_vcpu *vcpu);
 
 };
 
@@ -143,7 +143,7 @@ struct x86_emulate_ctxt {
 	/* Register state before/after emulation. */
 	struct kvm_vcpu *vcpu;
 
-	/* Linear faulting address (if emulating a page-faulting instruction). */
+	/* Linear faulting address (if emulating a page-faulting instruction) */
 	unsigned long eflags;
 
 	/* Emulated execution mode, represented by an X86EMUL_MODE value. */
-- 
cgit v1.2.3


From fd1ea0c25ae00e2ac55881af55c3206664dd59a8 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:36 -0700
Subject: include/asm-x86/lguest_hcall.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/lguest_hcall.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/lguest_hcall.h b/include/asm-x86/lguest_hcall.h
index f239e7069cab..a3241f28e34a 100644
--- a/include/asm-x86/lguest_hcall.h
+++ b/include/asm-x86/lguest_hcall.h
@@ -46,7 +46,7 @@ hcall(unsigned long call,
 {
 	/* "int" is the Intel instruction to trigger a trap. */
 	asm volatile("int $" __stringify(LGUEST_TRAP_ENTRY)
-		       /* The call in %eax (aka "a") might be overwritten */
+		     /* The call in %eax (aka "a") might be overwritten */
 		     : "=a"(call)
 		       /* The arguments are in %eax, %edx, %ebx & %ecx */
 		     : "a"(call), "d"(arg1), "b"(arg2), "c"(arg3)
@@ -62,8 +62,7 @@ hcall(unsigned long call,
 #define LGUEST_IRQS (NR_IRQS < 32 ? NR_IRQS: 32)
 
 #define LHCALL_RING_SIZE 64
-struct hcall_args
-{
+struct hcall_args {
 	/* These map directly onto eax, ebx, ecx, edx in struct lguest_regs */
 	unsigned long arg0, arg2, arg3, arg1;
 };
-- 
cgit v1.2.3


From fb444c7b25420d57ce5e31cab486f734705bd278 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:37 -0700
Subject: include/asm-x86/lguest.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/lguest.h | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/lguest.h b/include/asm-x86/lguest.h
index 9b17571e9bc3..be4a7247fa2b 100644
--- a/include/asm-x86/lguest.h
+++ b/include/asm-x86/lguest.h
@@ -34,8 +34,7 @@ extern const char lgstart_iret[], lgend_iret[];
 extern void lguest_iret(void);
 extern void lguest_init(void);
 
-struct lguest_regs
-{
+struct lguest_regs {
 	/* Manually saved part. */
 	unsigned long eax, ebx, ecx, edx;
 	unsigned long esi, edi, ebp;
@@ -51,8 +50,7 @@ struct lguest_regs
 };
 
 /* This is a guest-specific page (mapped ro) into the guest. */
-struct lguest_ro_state
-{
+struct lguest_ro_state {
 	/* Host information we need to restore when we switch back. */
 	u32 host_cr3;
 	struct desc_ptr host_idt_desc;
@@ -67,8 +65,7 @@ struct lguest_ro_state
 	struct desc_struct guest_gdt[GDT_ENTRIES];
 };
 
-struct lg_cpu_arch
-{
+struct lg_cpu_arch {
 	/* The GDT entries copied into lguest_ro_state when running. */
 	struct desc_struct gdt[GDT_ENTRIES];
 
@@ -85,7 +82,7 @@ static inline void lguest_set_ts(void)
 
 	cr0 = read_cr0();
 	if (!(cr0 & 8))
-		write_cr0(cr0|8);
+		write_cr0(cr0 | 8);
 }
 
 /* Full 4G segment descriptors, suitable for CS and DS. */
-- 
cgit v1.2.3


From 69cde6512c3a0227878869f9ba8a02cdc72fc253 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:39 -0700
Subject: include/asm-x86/local.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/local.h | 105 +++++++++++++++++++++++-------------------------
 1 file changed, 50 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/local.h b/include/asm-x86/local.h
index f852c62b3319..330a72496abd 100644
--- a/include/asm-x86/local.h
+++ b/include/asm-x86/local.h
@@ -18,32 +18,28 @@ typedef struct {
 
 static inline void local_inc(local_t *l)
 {
-	__asm__ __volatile__(
-		_ASM_INC "%0"
-		:"+m" (l->a.counter));
+	asm volatile(_ASM_INC "%0"
+		     : "+m" (l->a.counter));
 }
 
 static inline void local_dec(local_t *l)
 {
-	__asm__ __volatile__(
-		_ASM_DEC "%0"
-		:"+m" (l->a.counter));
+	asm volatile(_ASM_DEC "%0"
+		     : "+m" (l->a.counter));
 }
 
 static inline void local_add(long i, local_t *l)
 {
-	__asm__ __volatile__(
-		_ASM_ADD "%1,%0"
-		:"+m" (l->a.counter)
-		:"ir" (i));
+	asm volatile(_ASM_ADD "%1,%0"
+		     : "+m" (l->a.counter)
+		     : "ir" (i));
 }
 
 static inline void local_sub(long i, local_t *l)
 {
-	__asm__ __volatile__(
-		_ASM_SUB "%1,%0"
-		:"+m" (l->a.counter)
-		:"ir" (i));
+	asm volatile(_ASM_SUB "%1,%0"
+		     : "+m" (l->a.counter)
+		     : "ir" (i));
 }
 
 /**
@@ -59,10 +55,9 @@ static inline int local_sub_and_test(long i, local_t *l)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		_ASM_SUB "%2,%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		:"ir" (i) : "memory");
+	asm volatile(_ASM_SUB "%2,%0; sete %1"
+		     : "+m" (l->a.counter), "=qm" (c)
+		     : "ir" (i) : "memory");
 	return c;
 }
 
@@ -78,10 +73,9 @@ static inline int local_dec_and_test(local_t *l)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		_ASM_DEC "%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		: : "memory");
+	asm volatile(_ASM_DEC "%0; sete %1"
+		     : "+m" (l->a.counter), "=qm" (c)
+		     : : "memory");
 	return c != 0;
 }
 
@@ -97,10 +91,9 @@ static inline int local_inc_and_test(local_t *l)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		_ASM_INC "%0; sete %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		: : "memory");
+	asm volatile(_ASM_INC "%0; sete %1"
+		     : "+m" (l->a.counter), "=qm" (c)
+		     : : "memory");
 	return c != 0;
 }
 
@@ -117,10 +110,9 @@ static inline int local_add_negative(long i, local_t *l)
 {
 	unsigned char c;
 
-	__asm__ __volatile__(
-		_ASM_ADD "%2,%0; sets %1"
-		:"+m" (l->a.counter), "=qm" (c)
-		:"ir" (i) : "memory");
+	asm volatile(_ASM_ADD "%2,%0; sets %1"
+		     : "+m" (l->a.counter), "=qm" (c)
+		     : "ir" (i) : "memory");
 	return c;
 }
 
@@ -141,10 +133,9 @@ static inline long local_add_return(long i, local_t *l)
 #endif
 	/* Modern 486+ processor */
 	__i = i;
-	__asm__ __volatile__(
-		_ASM_XADD "%0, %1;"
-		:"+r" (i), "+m" (l->a.counter)
-		: : "memory");
+	asm volatile(_ASM_XADD "%0, %1;"
+		     : "+r" (i), "+m" (l->a.counter)
+		     : : "memory");
 	return i + __i;
 
 #ifdef CONFIG_M386
@@ -182,11 +173,11 @@ static inline long local_sub_return(long i, local_t *l)
 #define local_add_unless(l, a, u)				\
 ({								\
 	long c, old;						\
-	c = local_read(l);					\
+	c = local_read((l));					\
 	for (;;) {						\
 		if (unlikely(c == (u)))				\
 			break;					\
-		old = local_cmpxchg((l), c, c + (a));	\
+		old = local_cmpxchg((l), c, c + (a));		\
 		if (likely(old == c))				\
 			break;					\
 		c = old;					\
@@ -214,26 +205,30 @@ static inline long local_sub_return(long i, local_t *l)
 
 /* Need to disable preemption for the cpu local counters otherwise we could
    still access a variable of a previous CPU in a non atomic way. */
-#define cpu_local_wrap_v(l)	 	\
-	({ local_t res__;		\
-	   preempt_disable(); 		\
-	   res__ = (l);			\
-	   preempt_enable();		\
-	   res__; })
+#define cpu_local_wrap_v(l)		\
+({					\
+	local_t res__;			\
+	preempt_disable(); 		\
+	res__ = (l);			\
+	preempt_enable();		\
+	res__;				\
+})
 #define cpu_local_wrap(l)		\
-	({ preempt_disable();		\
-	   l;				\
-	   preempt_enable(); })		\
-
-#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var(l)))
-#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var(l), (i)))
-#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var(l)))
-#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var(l)))
-#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var(l)))
-#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var(l)))
-
-#define __cpu_local_inc(l)	cpu_local_inc(l)
-#define __cpu_local_dec(l)	cpu_local_dec(l)
+({					\
+	preempt_disable();		\
+	(l);				\
+	preempt_enable();		\
+})					\
+
+#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var((l))))
+#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var((l)), (i)))
+#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var((l))))
+#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var((l))))
+#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var((l))))
+#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var((l))))
+
+#define __cpu_local_inc(l)	cpu_local_inc((l))
+#define __cpu_local_dec(l)	cpu_local_dec((l))
 #define __cpu_local_add(i, l)	cpu_local_add((i), (l))
 #define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
 
-- 
cgit v1.2.3


From 933a44155caeb4ff5b58fcf755e3381ae37e72d4 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:40 -0700
Subject: include/asm-x86/mc146818rtc.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mc146818rtc.h | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mc146818rtc.h b/include/asm-x86/mc146818rtc.h
index cdd9f965835a..daf1ccde77af 100644
--- a/include/asm-x86/mc146818rtc.h
+++ b/include/asm-x86/mc146818rtc.h
@@ -42,7 +42,7 @@ extern volatile unsigned long cmos_lock;
 static inline void lock_cmos(unsigned char reg)
 {
 	unsigned long new;
-	new = ((smp_processor_id()+1) << 8) | reg;
+	new = ((smp_processor_id() + 1) << 8) | reg;
 	for (;;) {
 		if (cmos_lock) {
 			cpu_relax();
@@ -57,22 +57,26 @@ static inline void unlock_cmos(void)
 {
 	cmos_lock = 0;
 }
+
 static inline int do_i_have_lock_cmos(void)
 {
-	return (cmos_lock >> 8) == (smp_processor_id()+1);
+	return (cmos_lock >> 8) == (smp_processor_id() + 1);
 }
+
 static inline unsigned char current_lock_cmos_reg(void)
 {
 	return cmos_lock & 0xff;
 }
-#define lock_cmos_prefix(reg) \
+
+#define lock_cmos_prefix(reg)			\
 	do {					\
 		unsigned long cmos_flags;	\
 		local_irq_save(cmos_flags);	\
 		lock_cmos(reg)
-#define lock_cmos_suffix(reg) \
-		unlock_cmos();			\
-		local_irq_restore(cmos_flags);	\
+
+#define lock_cmos_suffix(reg)			\
+	unlock_cmos();				\
+	local_irq_restore(cmos_flags);		\
 	} while (0)
 #else
 #define lock_cmos_prefix(reg) do {} while (0)
-- 
cgit v1.2.3


From 77d511ed744f7febcd2a3444f63e6c54ac32d0c3 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:41 -0700
Subject: include/asm-x86/mca_dma.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mca_dma.h | 34 +++++++++++++++++-----------------
 1 file changed, 17 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mca_dma.h b/include/asm-x86/mca_dma.h
index fbb1f3b71279..c3dca6edc6b1 100644
--- a/include/asm-x86/mca_dma.h
+++ b/include/asm-x86/mca_dma.h
@@ -12,18 +12,18 @@
  *   count by 2 when using 16-bit dma; that is not handled by these functions.
  *
  * Ramen Noodles are yummy.
- * 
- *  1998 Tymm Twillman <tymm@computer.org>  
+ *
+ *  1998 Tymm Twillman <tymm@computer.org>
  */
 
 /*
- * Registers that are used by the DMA controller; FN is the function register 
+ * Registers that are used by the DMA controller; FN is the function register
  *   (tell the controller what to do) and EXE is the execution register (how
  *   to do it)
  */
 
 #define MCA_DMA_REG_FN  0x18
-#define MCA_DMA_REG_EXE 0x1A 
+#define MCA_DMA_REG_EXE 0x1A
 
 /*
  * Functions that the DMA controller can do
@@ -43,9 +43,9 @@
 
 /*
  * Modes (used by setting MCA_DMA_FN_MODE in the function register)
- * 
+ *
  * Note that the MODE_READ is read from memory (write to device), and
- *   MODE_WRITE is vice-versa.  
+ *   MODE_WRITE is vice-versa.
  */
 
 #define MCA_DMA_MODE_XFER  0x04  /* read by default */
@@ -63,7 +63,7 @@
  *	IRQ context.
  */
 
-static __inline__ void mca_enable_dma(unsigned int dmanr)
+static inline void mca_enable_dma(unsigned int dmanr)
 {
 	outb(MCA_DMA_FN_RESET_MASK | dmanr, MCA_DMA_REG_FN);
 }
@@ -76,7 +76,7 @@ static __inline__ void mca_enable_dma(unsigned int dmanr)
  *	IRQ context.
  */
 
-static __inline__ void mca_disable_dma(unsigned int dmanr)
+static inline void mca_disable_dma(unsigned int dmanr)
 {
 	outb(MCA_DMA_FN_MASK | dmanr, MCA_DMA_REG_FN);
 }
@@ -87,10 +87,10 @@ static __inline__ void mca_disable_dma(unsigned int dmanr)
  *	@a: 24bit bus address
  *
  *	Load the address register in the DMA controller. This has a 24bit
- *	limitation (16Mb). 
+ *	limitation (16Mb).
  */
 
-static __inline__ void mca_set_dma_addr(unsigned int dmanr, unsigned int a)
+static inline void mca_set_dma_addr(unsigned int dmanr, unsigned int a)
 {
 	outb(MCA_DMA_FN_SET_ADDR | dmanr, MCA_DMA_REG_FN);
 	outb(a & 0xff, MCA_DMA_REG_EXE);
@@ -106,14 +106,14 @@ static __inline__ void mca_set_dma_addr(unsigned int dmanr, unsigned int a)
  *	limitation (16Mb). The return is a bus address.
  */
 
-static __inline__ unsigned int mca_get_dma_addr(unsigned int dmanr)
+static inline unsigned int mca_get_dma_addr(unsigned int dmanr)
 {
 	unsigned int addr;
 
 	outb(MCA_DMA_FN_GET_ADDR | dmanr, MCA_DMA_REG_FN);
 	addr = inb(MCA_DMA_REG_EXE);
 	addr |= inb(MCA_DMA_REG_EXE) << 8;
-	addr |= inb(MCA_DMA_REG_EXE) << 16;  
+	addr |= inb(MCA_DMA_REG_EXE) << 16;
 
 	return addr;
 }
@@ -127,7 +127,7 @@ static __inline__ unsigned int mca_get_dma_addr(unsigned int dmanr)
  *	Setting a count of zero will not do what you expect.
  */
 
-static __inline__ void mca_set_dma_count(unsigned int dmanr, unsigned int count)
+static inline void mca_set_dma_count(unsigned int dmanr, unsigned int count)
 {
 	count--;  /* transfers one more than count -- correct for this */
 
@@ -144,7 +144,7 @@ static __inline__ void mca_set_dma_count(unsigned int dmanr, unsigned int count)
  *	on this DMA channel.
  */
 
-static __inline__ unsigned int mca_get_dma_residue(unsigned int dmanr)
+static inline unsigned int mca_get_dma_residue(unsigned int dmanr)
 {
 	unsigned short count;
 
@@ -164,12 +164,12 @@ static __inline__ unsigned int mca_get_dma_residue(unsigned int dmanr)
  *	with an I/O port target.
  */
 
-static __inline__ void mca_set_dma_io(unsigned int dmanr, unsigned int io_addr)
+static inline void mca_set_dma_io(unsigned int dmanr, unsigned int io_addr)
 {
 	/*
 	 * DMA from a port address -- set the io address
 	 */
-	
+
 	outb(MCA_DMA_FN_SET_IO | dmanr, MCA_DMA_REG_FN);
 	outb(io_addr & 0xff, MCA_DMA_REG_EXE);
 	outb((io_addr >>  8) & 0xff, MCA_DMA_REG_EXE);
@@ -192,7 +192,7 @@ static __inline__ void mca_set_dma_io(unsigned int dmanr, unsigned int io_addr)
  *	%MCA_DMA_MODE_16 to do 16bit transfers.
  */
 
-static __inline__ void mca_set_dma_mode(unsigned int dmanr, unsigned int mode)
+static inline void mca_set_dma_mode(unsigned int dmanr, unsigned int mode)
 {
 	outb(MCA_DMA_FN_SET_MODE | dmanr, MCA_DMA_REG_FN);
 	outb(mode, MCA_DMA_REG_EXE);
-- 
cgit v1.2.3


From 55464da94a845e057ffb94a9fc7be1aa86ffcd89 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:42 -0700
Subject: include/asm-x86/mmu_context_32.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmu_context_32.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mmu_context_32.h b/include/asm-x86/mmu_context_32.h
index 8198d1cca1f3..9756ae0f1dd3 100644
--- a/include/asm-x86/mmu_context_32.h
+++ b/include/asm-x86/mmu_context_32.h
@@ -62,7 +62,7 @@ static inline void switch_mm(struct mm_struct *prev,
 		BUG_ON(per_cpu(cpu_tlbstate, cpu).active_mm != next);
 
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
-			/* We were in lazy tlb mode and leave_mm disabled 
+			/* We were in lazy tlb mode and leave_mm disabled
 			 * tlb flush IPI delivery. We must reload %cr3.
 			 */
 			load_cr3(next->pgd);
@@ -75,10 +75,10 @@ static inline void switch_mm(struct mm_struct *prev,
 #define deactivate_mm(tsk, mm)			\
 	asm("movl %0,%%gs": :"r" (0));
 
-#define activate_mm(prev, next)				\
-	do {						\
-		paravirt_activate_mm(prev, next);	\
-		switch_mm((prev),(next),NULL);		\
-	} while(0);
+#define activate_mm(prev, next)			\
+do {						\
+	paravirt_activate_mm((prev), (next));	\
+	switch_mm((prev), (next), NULL);	\
+} while (0);
 
 #endif
-- 
cgit v1.2.3


From c4fe760efde84e52168a81bf125f25ba2f118b51 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:43 -0700
Subject: include/asm-x86/mmu_context_64.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmu_context_64.h | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mmu_context_64.h b/include/asm-x86/mmu_context_64.h
index ad6dc821ef9e..ca44c71e7fb3 100644
--- a/include/asm-x86/mmu_context_64.h
+++ b/include/asm-x86/mmu_context_64.h
@@ -20,12 +20,12 @@ void destroy_context(struct mm_struct *mm);
 static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {
 #ifdef CONFIG_SMP
-	if (read_pda(mmu_state) == TLBSTATE_OK) 
+	if (read_pda(mmu_state) == TLBSTATE_OK)
 		write_pda(mmu_state, TLBSTATE_LAZY);
 #endif
 }
 
-static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, 
+static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 			     struct task_struct *tsk)
 {
 	unsigned cpu = smp_processor_id();
@@ -39,7 +39,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 		cpu_set(cpu, next->cpu_vm_mask);
 		load_cr3(next->pgd);
 
-		if (unlikely(next->context.ldt != prev->context.ldt)) 
+		if (unlikely(next->context.ldt != prev->context.ldt))
 			load_LDT_nolock(&next->context);
 	}
 #ifdef CONFIG_SMP
@@ -48,7 +48,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 		if (read_pda(active_mm) != next)
 			BUG();
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
-			/* We were in lazy tlb mode and leave_mm disabled 
+			/* We were in lazy tlb mode and leave_mm disabled
 			 * tlb flush IPI delivery. We must reload CR3
 			 * to make sure to use no freed page tables.
 			 */
@@ -59,13 +59,14 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 #endif
 }
 
-#define deactivate_mm(tsk,mm)	do { \
-	load_gs_index(0); \
-	asm volatile("movl %0,%%fs"::"r"(0));  \
-} while(0)
+#define deactivate_mm(tsk, mm)			\
+do {						\
+	load_gs_index(0);			\
+	asm volatile("movl %0,%%fs"::"r"(0));	\
+} while (0)
 
-#define activate_mm(prev, next) \
-	switch_mm((prev),(next),NULL)
+#define activate_mm(prev, next)			\
+	switch_mm((prev), (next), NULL)
 
 
 #endif
-- 
cgit v1.2.3


From 710d0e9cce8bfcfa821c5929c1d4dbf30a660ce6 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:44 -0700
Subject: include/asm-x86/mmu.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmu.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mmu.h b/include/asm-x86/mmu.h
index efa962c38897..00e88679e11f 100644
--- a/include/asm-x86/mmu.h
+++ b/include/asm-x86/mmu.h
@@ -10,10 +10,10 @@
  *
  * cpu_vm_mask is used to optimize ldt flushing.
  */
-typedef struct { 
+typedef struct {
 	void *ldt;
 #ifdef CONFIG_X86_64
-	rwlock_t ldtlock; 
+	rwlock_t ldtlock;
 #endif
 	int size;
 	struct mutex lock;
-- 
cgit v1.2.3


From f2334076890bbe3cddca2c053684653c614e9b48 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:45 -0700
Subject: include/asm-x86/mmx.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmx.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/mmx.h b/include/asm-x86/mmx.h
index 46b71da99869..940881218ff8 100644
--- a/include/asm-x86/mmx.h
+++ b/include/asm-x86/mmx.h
@@ -6,7 +6,7 @@
  */
 
 #include <linux/types.h>
- 
+
 extern void *_mmx_memcpy(void *to, const void *from, size_t size);
 extern void mmx_clear_page(void *page);
 extern void mmx_copy_page(void *to, void *from);
-- 
cgit v1.2.3


From 7491d33d9ab042f3fdb9ec00054f69737dcd180f Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:46 -0700
Subject: include/asm-x86/mmzone_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmzone_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/mmzone_32.h b/include/asm-x86/mmzone_32.h
index b9f5be2f603b..cb2cad0b65a7 100644
--- a/include/asm-x86/mmzone_32.h
+++ b/include/asm-x86/mmzone_32.h
@@ -18,7 +18,7 @@ extern struct pglist_data *node_data[];
 	#include <asm/srat.h>
 #endif
 
-extern int get_memcfg_numa_flat(void );
+extern int get_memcfg_numa_flat(void);
 /*
  * This allows any one NUMA architecture to be compiled
  * for, and still fall back to the flat function if it
-- 
cgit v1.2.3


From 60e9bfd1bd5be433e56b050e025d3d5c91c967a8 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:47 -0700
Subject: include/asm-x86/mmzone_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mmzone_64.h | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h
index ebaf9663aa8a..594bd0dc1d08 100644
--- a/include/asm-x86/mmzone_64.h
+++ b/include/asm-x86/mmzone_64.h
@@ -7,7 +7,7 @@
 
 #ifdef CONFIG_NUMA
 
-#define VIRTUAL_BUG_ON(x) 
+#define VIRTUAL_BUG_ON(x)
 
 #include <asm/smp.h>
 
@@ -16,7 +16,7 @@ struct memnode {
 	int shift;
 	unsigned int mapsize;
 	s16 *map;
-	s16 embedded_map[64-8];
+	s16 embedded_map[64 - 8];
 } ____cacheline_aligned; /* total size = 128 bytes */
 extern struct memnode memnode;
 #define memnode_shift memnode.shift
@@ -25,27 +25,27 @@ extern struct memnode memnode;
 
 extern struct pglist_data *node_data[];
 
-static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) 
-{ 
-	unsigned nid; 
+static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
+{
+	unsigned nid;
 	VIRTUAL_BUG_ON(!memnodemap);
 	VIRTUAL_BUG_ON((addr >> memnode_shift) >= memnodemapsize);
-	nid = memnodemap[addr >> memnode_shift]; 
-	VIRTUAL_BUG_ON(nid >= MAX_NUMNODES || !node_data[nid]); 
-	return nid; 
-} 
+	nid = memnodemap[addr >> memnode_shift];
+	VIRTUAL_BUG_ON(nid >= MAX_NUMNODES || !node_data[nid]);
+	return nid;
+}
 
 #define NODE_DATA(nid)		(node_data[nid])
 
 #define node_start_pfn(nid)	(NODE_DATA(nid)->node_start_pfn)
-#define node_end_pfn(nid)       (NODE_DATA(nid)->node_start_pfn + \
+#define node_end_pfn(nid)       (NODE_DATA(nid)->node_start_pfn +	\
 				 NODE_DATA(nid)->node_spanned_pages)
 
 extern int early_pfn_to_nid(unsigned long pfn);
 
 #ifdef CONFIG_NUMA_EMU
-#define FAKE_NODE_MIN_SIZE	(64*1024*1024)
-#define FAKE_NODE_MIN_HASH_MASK	(~(FAKE_NODE_MIN_SIZE - 1uL))
+#define FAKE_NODE_MIN_SIZE	(64 * 1024 * 1024)
+#define FAKE_NODE_MIN_HASH_MASK	(~(FAKE_NODE_MIN_SIZE - 1UL))
 #endif
 
 #endif
-- 
cgit v1.2.3


From 8f08403e61a86c3179642239184aff3a5f636be1 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:48 -0700
Subject: include/asm-x86/mpspec_def.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mpspec_def.h | 35 +++++++++++++----------------------
 1 file changed, 13 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mpspec_def.h b/include/asm-x86/mpspec_def.h
index 1f35691b4f7c..dc6ef85e3624 100644
--- a/include/asm-x86/mpspec_def.h
+++ b/include/asm-x86/mpspec_def.h
@@ -11,7 +11,7 @@
  * information is.
  */
 
-#define SMP_MAGIC_IDENT	(('_'<<24)|('P'<<16)|('M'<<8)|'_')
+#define SMP_MAGIC_IDENT	(('_'<<24) | ('P'<<16) | ('M'<<8) | '_')
 
 #ifdef CONFIG_X86_32
 # define MAX_MPC_ENTRY 1024
@@ -23,8 +23,7 @@
 # define MAX_APICS 255
 #endif
 
-struct intel_mp_floating
-{
+struct intel_mp_floating {
 	char mpf_signature[4];		/* "_MP_"			*/
 	unsigned int mpf_physptr;	/* Configuration table address	*/
 	unsigned char mpf_length;	/* Our length (paragraphs)	*/
@@ -39,14 +38,13 @@ struct intel_mp_floating
 
 #define MPC_SIGNATURE "PCMP"
 
-struct mp_config_table
-{
+struct mp_config_table {
 	char mpc_signature[4];
 	unsigned short mpc_length;	/* Size of table */
-	char  mpc_spec;			/* 0x01 */
-	char  mpc_checksum;
-	char  mpc_oem[8];
-	char  mpc_productid[12];
+	char mpc_spec;			/* 0x01 */
+	char mpc_checksum;
+	char mpc_oem[8];
+	char mpc_productid[12];
 	unsigned int mpc_oemptr;	/* 0 if not present */
 	unsigned short mpc_oemsize;	/* 0 if not present */
 	unsigned short mpc_oemcount;
@@ -71,8 +69,7 @@ struct mp_config_table
 #define CPU_MODEL_MASK		0x00F0
 #define CPU_FAMILY_MASK		0x0F00
 
-struct mpc_config_processor
-{
+struct mpc_config_processor {
 	unsigned char mpc_type;
 	unsigned char mpc_apicid;	/* Local APIC number */
 	unsigned char mpc_apicver;	/* Its versions */
@@ -82,8 +79,7 @@ struct mpc_config_processor
 	unsigned int mpc_reserved[2];
 };
 
-struct mpc_config_bus
-{
+struct mpc_config_bus {
 	unsigned char mpc_type;
 	unsigned char mpc_busid;
 	unsigned char mpc_bustype[6];
@@ -111,8 +107,7 @@ struct mpc_config_bus
 
 #define MPC_APIC_USABLE		0x01
 
-struct mpc_config_ioapic
-{
+struct mpc_config_ioapic {
 	unsigned char mpc_type;
 	unsigned char mpc_apicid;
 	unsigned char mpc_apicver;
@@ -120,8 +115,7 @@ struct mpc_config_ioapic
 	unsigned int mpc_apicaddr;
 };
 
-struct mpc_config_intsrc
-{
+struct mpc_config_intsrc {
 	unsigned char mpc_type;
 	unsigned char mpc_irqtype;
 	unsigned short mpc_irqflag;
@@ -144,8 +138,7 @@ enum mp_irq_source_types {
 
 #define MP_APIC_ALL	0xFF
 
-struct mpc_config_lintsrc
-{
+struct mpc_config_lintsrc {
 	unsigned char mpc_type;
 	unsigned char mpc_irqtype;
 	unsigned short mpc_irqflag;
@@ -157,8 +150,7 @@ struct mpc_config_lintsrc
 
 #define MPC_OEM_SIGNATURE "_OEM"
 
-struct mp_config_oemtable
-{
+struct mp_config_oemtable {
 	char oem_signature[4];
 	unsigned short oem_length;	/* Size of table */
 	char  oem_rev;			/* 0x01 */
@@ -185,4 +177,3 @@ enum mp_bustype {
 	MP_BUS_MCA,
 };
 #endif
-
-- 
cgit v1.2.3


From 30971e17ee484f72e081826a0bf3e489ef3b4c30 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:49 -0700
Subject: include/asm-x86/mpspec.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mpspec.h | 23 +++++++++++------------
 1 file changed, 11 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 99da0a59b436..eccbc581ec84 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -55,8 +55,7 @@ extern int mp_register_gsi(u32 gsi, int edge_level, int active_high_low);
 
 #define PHYSID_ARRAY_SIZE	BITS_TO_LONGS(MAX_APICS)
 
-struct physid_mask
-{
+struct physid_mask {
 	unsigned long mask[PHYSID_ARRAY_SIZE];
 };
 
@@ -65,34 +64,34 @@ typedef struct physid_mask physid_mask_t;
 #define physid_set(physid, map)			set_bit(physid, (map).mask)
 #define physid_clear(physid, map)		clear_bit(physid, (map).mask)
 #define physid_isset(physid, map)		test_bit(physid, (map).mask)
-#define physid_test_and_set(physid, map) \
+#define physid_test_and_set(physid, map)			\
 	test_and_set_bit(physid, (map).mask)
 
-#define physids_and(dst, src1, src2) \
+#define physids_and(dst, src1, src2)					\
 	bitmap_and((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
 
-#define physids_or(dst, src1, src2) \
+#define physids_or(dst, src1, src2)					\
 	bitmap_or((dst).mask, (src1).mask, (src2).mask, MAX_APICS)
 
-#define physids_clear(map) \
+#define physids_clear(map)					\
 	bitmap_zero((map).mask, MAX_APICS)
 
-#define physids_complement(dst, src) \
+#define physids_complement(dst, src)				\
 	bitmap_complement((dst).mask, (src).mask, MAX_APICS)
 
-#define physids_empty(map) \
+#define physids_empty(map)					\
 	bitmap_empty((map).mask, MAX_APICS)
 
-#define physids_equal(map1, map2) \
+#define physids_equal(map1, map2)				\
 	bitmap_equal((map1).mask, (map2).mask, MAX_APICS)
 
-#define physids_weight(map) \
+#define physids_weight(map)					\
 	bitmap_weight((map).mask, MAX_APICS)
 
-#define physids_shift_right(d, s, n) \
+#define physids_shift_right(d, s, n)				\
 	bitmap_shift_right((d).mask, (s).mask, n, MAX_APICS)
 
-#define physids_shift_left(d, s, n) \
+#define physids_shift_left(d, s, n)				\
 	bitmap_shift_left((d).mask, (s).mask, n, MAX_APICS)
 
 #define physids_coerce(map)			((map).mask[0])
-- 
cgit v1.2.3


From 934902b474bdb235a273985ad4c61eb136afe11d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:50 -0700
Subject: include/asm-x86/msidef.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/msidef.h | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msidef.h b/include/asm-x86/msidef.h
index 5b8acddb70fb..296f29ce426d 100644
--- a/include/asm-x86/msidef.h
+++ b/include/asm-x86/msidef.h
@@ -11,7 +11,8 @@
 
 #define MSI_DATA_VECTOR_SHIFT		0
 #define  MSI_DATA_VECTOR_MASK		0x000000ff
-#define	 MSI_DATA_VECTOR(v)		(((v) << MSI_DATA_VECTOR_SHIFT) & MSI_DATA_VECTOR_MASK)
+#define	 MSI_DATA_VECTOR(v)		(((v) << MSI_DATA_VECTOR_SHIFT) & \
+					 MSI_DATA_VECTOR_MASK)
 
 #define MSI_DATA_DELIVERY_MODE_SHIFT	8
 #define  MSI_DATA_DELIVERY_FIXED	(0 << MSI_DATA_DELIVERY_MODE_SHIFT)
@@ -37,11 +38,14 @@
 #define	 MSI_ADDR_DEST_MODE_LOGICAL	(1 << MSI_ADDR_DEST_MODE_SHIFT)
 
 #define MSI_ADDR_REDIRECTION_SHIFT	3
-#define  MSI_ADDR_REDIRECTION_CPU	(0 << MSI_ADDR_REDIRECTION_SHIFT) /* dedicated cpu */
-#define  MSI_ADDR_REDIRECTION_LOWPRI	(1 << MSI_ADDR_REDIRECTION_SHIFT) /* lowest priority */
+#define  MSI_ADDR_REDIRECTION_CPU	(0 << MSI_ADDR_REDIRECTION_SHIFT)
+					/* dedicated cpu */
+#define  MSI_ADDR_REDIRECTION_LOWPRI	(1 << MSI_ADDR_REDIRECTION_SHIFT)
+					/* lowest priority */
 
 #define MSI_ADDR_DEST_ID_SHIFT		12
 #define	 MSI_ADDR_DEST_ID_MASK		0x00ffff0
-#define  MSI_ADDR_DEST_ID(dest)		(((dest) << MSI_ADDR_DEST_ID_SHIFT) & MSI_ADDR_DEST_ID_MASK)
+#define  MSI_ADDR_DEST_ID(dest)		(((dest) << MSI_ADDR_DEST_ID_SHIFT) & \
+					 MSI_ADDR_DEST_ID_MASK)
 
 #endif /* ASM_MSIDEF_H */
-- 
cgit v1.2.3


From abb0ade013507c93a9a0b263bbb7b0327d7c38db Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:51 -0700
Subject: include/asm-x86/msr.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/msr.h | 81 ++++++++++++++++++++++++++-------------------------
 1 file changed, 42 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 3ca29ebebbb1..2c698a2e81f9 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -16,8 +16,8 @@
 static inline unsigned long long native_read_tscp(unsigned int *aux)
 {
 	unsigned long low, high;
-	asm volatile (".byte 0x0f,0x01,0xf9"
-		      : "=a" (low), "=d" (high), "=c" (*aux));
+	asm volatile(".byte 0x0f,0x01,0xf9"
+		     : "=a" (low), "=d" (high), "=c" (*aux));
 	return low | ((u64)high >> 32);
 }
 
@@ -29,7 +29,7 @@ static inline unsigned long long native_read_tscp(unsigned int *aux)
  */
 #ifdef CONFIG_X86_64
 #define DECLARE_ARGS(val, low, high)	unsigned low, high
-#define EAX_EDX_VAL(val, low, high)	(low | ((u64)(high) << 32))
+#define EAX_EDX_VAL(val, low, high)	((low) | ((u64)(high) << 32))
 #define EAX_EDX_ARGS(val, low, high)	"a" (low), "d" (high)
 #define EAX_EDX_RET(val, low, high)	"=a" (low), "=d" (high)
 #else
@@ -57,7 +57,7 @@ static inline unsigned long long native_read_msr_safe(unsigned int msr,
 		     ".section .fixup,\"ax\"\n\t"
 		     "3:  mov %3,%0 ; jmp 1b\n\t"
 		     ".previous\n\t"
-		     _ASM_EXTABLE(2b,3b)
+		     _ASM_EXTABLE(2b, 3b)
 		     : "=r" (*err), EAX_EDX_RET(val, low, high)
 		     : "c" (msr), "i" (-EFAULT));
 	return EAX_EDX_VAL(val, low, high);
@@ -78,10 +78,10 @@ static inline int native_write_msr_safe(unsigned int msr,
 		     ".section .fixup,\"ax\"\n\t"
 		     "3:  mov %4,%0 ; jmp 1b\n\t"
 		     ".previous\n\t"
-		     _ASM_EXTABLE(2b,3b)
+		     _ASM_EXTABLE(2b, 3b)
 		     : "=a" (err)
 		     : "c" (msr), "0" (low), "d" (high),
-		       "i" (-EFAULT));
+		     "i" (-EFAULT));
 	return err;
 }
 
@@ -116,23 +116,23 @@ static inline unsigned long long native_read_pmc(int counter)
  * pointer indirection), this allows gcc to optimize better
  */
 
-#define rdmsr(msr,val1,val2)						\
-	do {								\
-		u64 __val = native_read_msr(msr);			\
-		(val1) = (u32)__val;					\
-		(val2) = (u32)(__val >> 32);				\
-	} while(0)
+#define rdmsr(msr, val1, val2)					\
+do {								\
+	u64 __val = native_read_msr((msr));			\
+	(val1) = (u32)__val;					\
+	(val2) = (u32)(__val >> 32);				\
+} while (0)
 
 static inline void wrmsr(unsigned msr, unsigned low, unsigned high)
 {
 	native_write_msr(msr, low, high);
 }
 
-#define rdmsrl(msr,val)							\
-	((val) = native_read_msr(msr))
+#define rdmsrl(msr, val)			\
+	((val) = native_read_msr((msr)))
 
 #define wrmsrl(msr, val)						\
-	native_write_msr(msr, (u32)((u64)(val)), (u32)((u64)(val) >> 32))
+	native_write_msr((msr), (u32)((u64)(val)), (u32)((u64)(val) >> 32))
 
 /* wrmsr with exception handling */
 static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
@@ -141,14 +141,14 @@ static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
 }
 
 /* rdmsr with exception handling */
-#define rdmsr_safe(msr,p1,p2)						\
-	({								\
-		int __err;						\
-		u64 __val = native_read_msr_safe(msr, &__err);		\
-		(*p1) = (u32)__val;					\
-		(*p2) = (u32)(__val >> 32);				\
-		__err;							\
-	})
+#define rdmsr_safe(msr, p1, p2)					\
+({								\
+	int __err;						\
+	u64 __val = native_read_msr_safe((msr), &__err);	\
+	(*p1) = (u32)__val;					\
+	(*p2) = (u32)(__val >> 32);				\
+	__err;							\
+})
 
 #define rdtscl(low)						\
 	((low) = (u32)native_read_tsc())
@@ -156,35 +156,37 @@ static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
 #define rdtscll(val)						\
 	((val) = native_read_tsc())
 
-#define rdpmc(counter,low,high)					\
-	do {							\
-		u64 _l = native_read_pmc(counter);		\
-		(low)  = (u32)_l;				\
-		(high) = (u32)(_l >> 32);			\
-	} while(0)
+#define rdpmc(counter, low, high)			\
+do {							\
+	u64 _l = native_read_pmc((counter));		\
+	(low)  = (u32)_l;				\
+	(high) = (u32)(_l >> 32);			\
+} while (0)
 
-#define rdtscp(low, high, aux)						\
-       do {                                                            \
-		unsigned long long _val = native_read_tscp(&(aux));     \
-		(low) = (u32)_val;                                      \
-		(high) = (u32)(_val >> 32);                             \
-       } while (0)
+#define rdtscp(low, high, aux)					\
+do {                                                            \
+	unsigned long long _val = native_read_tscp(&(aux));     \
+	(low) = (u32)_val;                                      \
+	(high) = (u32)(_val >> 32);                             \
+} while (0)
 
 #define rdtscpll(val, aux) (val) = native_read_tscp(&(aux))
 
 #endif	/* !CONFIG_PARAVIRT */
 
 
-#define checking_wrmsrl(msr,val) wrmsr_safe(msr,(u32)(val),(u32)((val)>>32))
+#define checking_wrmsrl(msr, val) wrmsr_safe((msr), (u32)(val),		\
+					     (u32)((val) >> 32))
 
-#define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
+#define write_tsc(val1, val2) wrmsr(0x10, (val1), (val2))
 
-#define write_rdtscp_aux(val) wrmsr(0xc0000103, val, 0)
+#define write_rdtscp_aux(val) wrmsr(0xc0000103, (val), 0)
 
 #ifdef CONFIG_SMP
 void rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
 void wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
 int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
+
 int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
 #else  /*  CONFIG_SMP  */
 static inline void rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h)
@@ -195,7 +197,8 @@ static inline void wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h)
 {
 	wrmsr(msr_no, l, h);
 }
-static inline int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h)
+static inline int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no,
+				    u32 *l, u32 *h)
 {
 	return rdmsr_safe(msr_no, l, h);
 }
-- 
cgit v1.2.3


From 9969b4405469e12070c560ff27dbe587470fc945 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:52 -0700
Subject: include/asm-x86/mtrr.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mtrr.h | 64 +++++++++++++++++++++++---------------------------
 1 file changed, 30 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mtrr.h b/include/asm-x86/mtrr.h
index d3d26625aeaf..a69a01a51729 100644
--- a/include/asm-x86/mtrr.h
+++ b/include/asm-x86/mtrr.h
@@ -28,8 +28,7 @@
 
 #define	MTRR_IOCTL_BASE	'M'
 
-struct mtrr_sentry
-{
+struct mtrr_sentry {
     unsigned long base;    /*  Base address     */
     unsigned int size;    /*  Size of region   */
     unsigned int type;     /*  Type of region   */
@@ -41,8 +40,7 @@ struct mtrr_sentry
    will break. */
 
 #ifdef __i386__
-struct mtrr_gentry
-{
+struct mtrr_gentry {
     unsigned int regnum;   /*  Register number  */
     unsigned long base;    /*  Base address     */
     unsigned int size;    /*  Size of region   */
@@ -51,8 +49,7 @@ struct mtrr_gentry
 
 #else /* __i386__ */
 
-struct mtrr_gentry
-{
+struct mtrr_gentry {
     unsigned long base;    /*  Base address     */
     unsigned int size;    /*  Size of region   */
     unsigned int regnum;   /*  Register number  */
@@ -89,12 +86,12 @@ struct mtrr_gentry
 extern u8 mtrr_type_lookup(u64 addr, u64 end);
 extern void mtrr_save_fixed_ranges(void *);
 extern void mtrr_save_state(void);
-extern int mtrr_add (unsigned long base, unsigned long size,
-		     unsigned int type, bool increment);
-extern int mtrr_add_page (unsigned long base, unsigned long size,
-		     unsigned int type, bool increment);
-extern int mtrr_del (int reg, unsigned long base, unsigned long size);
-extern int mtrr_del_page (int reg, unsigned long base, unsigned long size);
+extern int mtrr_add(unsigned long base, unsigned long size,
+		    unsigned int type, bool increment);
+extern int mtrr_add_page(unsigned long base, unsigned long size,
+			 unsigned int type, bool increment);
+extern int mtrr_del(int reg, unsigned long base, unsigned long size);
+extern int mtrr_del_page(int reg, unsigned long base, unsigned long size);
 extern void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi);
 extern void mtrr_ap_init(void);
 extern void mtrr_bp_init(void);
@@ -110,23 +107,21 @@ static inline u8 mtrr_type_lookup(u64 addr, u64 end)
 }
 #define mtrr_save_fixed_ranges(arg) do {} while (0)
 #define mtrr_save_state() do {} while (0)
-static __inline__ int mtrr_add (unsigned long base, unsigned long size,
-				unsigned int type, bool increment)
+static inline int mtrr_add(unsigned long base, unsigned long size,
+			   unsigned int type, bool increment)
 {
     return -ENODEV;
 }
-static __inline__ int mtrr_add_page (unsigned long base, unsigned long size,
+static inline int mtrr_add_page(unsigned long base, unsigned long size,
 				unsigned int type, bool increment)
 {
     return -ENODEV;
 }
-static __inline__ int mtrr_del (int reg, unsigned long base,
-				unsigned long size)
+static inline int mtrr_del(int reg, unsigned long base, unsigned long size)
 {
     return -ENODEV;
 }
-static __inline__ int mtrr_del_page (int reg, unsigned long base,
-				unsigned long size)
+static inline int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 {
     return -ENODEV;
 }
@@ -134,7 +129,9 @@ static inline int mtrr_trim_uncached_memory(unsigned long end_pfn)
 {
 	return 0;
 }
-static __inline__ void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi) {;}
+static inline void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi)
+{
+}
 
 #define mtrr_ap_init() do {} while (0)
 #define mtrr_bp_init() do {} while (0)
@@ -143,15 +140,13 @@ static __inline__ void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi) {;}
 #ifdef CONFIG_COMPAT
 #include <linux/compat.h>
 
-struct mtrr_sentry32
-{
+struct mtrr_sentry32 {
     compat_ulong_t base;    /*  Base address     */
     compat_uint_t size;    /*  Size of region   */
     compat_uint_t type;     /*  Type of region   */
 };
 
-struct mtrr_gentry32
-{
+struct mtrr_gentry32 {
     compat_ulong_t regnum;   /*  Register number  */
     compat_uint_t base;    /*  Base address     */
     compat_uint_t size;    /*  Size of region   */
@@ -160,16 +155,17 @@ struct mtrr_gentry32
 
 #define MTRR_IOCTL_BASE 'M'
 
-#define MTRRIOC32_ADD_ENTRY        _IOW(MTRR_IOCTL_BASE,  0, struct mtrr_sentry32)
-#define MTRRIOC32_SET_ENTRY        _IOW(MTRR_IOCTL_BASE,  1, struct mtrr_sentry32)
-#define MTRRIOC32_DEL_ENTRY        _IOW(MTRR_IOCTL_BASE,  2, struct mtrr_sentry32)
-#define MTRRIOC32_GET_ENTRY        _IOWR(MTRR_IOCTL_BASE, 3, struct mtrr_gentry32)
-#define MTRRIOC32_KILL_ENTRY       _IOW(MTRR_IOCTL_BASE,  4, struct mtrr_sentry32)
-#define MTRRIOC32_ADD_PAGE_ENTRY   _IOW(MTRR_IOCTL_BASE,  5, struct mtrr_sentry32)
-#define MTRRIOC32_SET_PAGE_ENTRY   _IOW(MTRR_IOCTL_BASE,  6, struct mtrr_sentry32)
-#define MTRRIOC32_DEL_PAGE_ENTRY   _IOW(MTRR_IOCTL_BASE,  7, struct mtrr_sentry32)
-#define MTRRIOC32_GET_PAGE_ENTRY   _IOWR(MTRR_IOCTL_BASE, 8, struct mtrr_gentry32)
-#define MTRRIOC32_KILL_PAGE_ENTRY  _IOW(MTRR_IOCTL_BASE,  9, struct mtrr_sentry32)
+#define MTRRIOC32_ADD_ENTRY      _IOW(MTRR_IOCTL_BASE,  0, struct mtrr_sentry32)
+#define MTRRIOC32_SET_ENTRY      _IOW(MTRR_IOCTL_BASE,  1, struct mtrr_sentry32)
+#define MTRRIOC32_DEL_ENTRY      _IOW(MTRR_IOCTL_BASE,  2, struct mtrr_sentry32)
+#define MTRRIOC32_GET_ENTRY      _IOWR(MTRR_IOCTL_BASE, 3, struct mtrr_gentry32)
+#define MTRRIOC32_KILL_ENTRY     _IOW(MTRR_IOCTL_BASE,  4, struct mtrr_sentry32)
+#define MTRRIOC32_ADD_PAGE_ENTRY _IOW(MTRR_IOCTL_BASE,  5, struct mtrr_sentry32)
+#define MTRRIOC32_SET_PAGE_ENTRY _IOW(MTRR_IOCTL_BASE,  6, struct mtrr_sentry32)
+#define MTRRIOC32_DEL_PAGE_ENTRY _IOW(MTRR_IOCTL_BASE,  7, struct mtrr_sentry32)
+#define MTRRIOC32_GET_PAGE_ENTRY _IOWR(MTRR_IOCTL_BASE, 8, struct mtrr_gentry32)
+#define MTRRIOC32_KILL_PAGE_ENTRY		\
+				 _IOW(MTRR_IOCTL_BASE,  9, struct mtrr_sentry32)
 #endif /* CONFIG_COMPAT */
 
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From b2347fad517f61553e03135db60def2392d9c2bc Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:53 -0700
Subject: include/asm-x86/mutex_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mutex_32.h | 64 ++++++++++++++++++++++------------------------
 1 file changed, 30 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mutex_32.h b/include/asm-x86/mutex_32.h
index 9a6b3da25914..73e928ef5f03 100644
--- a/include/asm-x86/mutex_32.h
+++ b/include/asm-x86/mutex_32.h
@@ -21,22 +21,20 @@
  * wasn't 1 originally. This function MUST leave the value lower than 1
  * even when the "1" assertion wasn't true.
  */
-#define __mutex_fastpath_lock(count, fail_fn)				\
-do {									\
-	unsigned int dummy;						\
-									\
-	typecheck(atomic_t *, count);					\
+#define __mutex_fastpath_lock(count, fail_fn)			\
+do {								\
+	unsigned int dummy;					\
+								\
+	typecheck(atomic_t *, count);				\
 	typecheck_fn(void (*)(atomic_t *), fail_fn);		\
-									\
-	__asm__ __volatile__(						\
-		LOCK_PREFIX "   decl (%%eax)	\n"			\
-			"   jns 1f		\n"			\
-			"   call "#fail_fn"	\n"			\
-			"1:			\n"			\
-									\
-		:"=a" (dummy)						\
-		: "a" (count)						\
-		: "memory", "ecx", "edx");				\
+								\
+	asm volatile(LOCK_PREFIX "   decl (%%eax)\n"		\
+		     "   jns 1f	\n"				\
+		     "   call " #fail_fn "\n"			\
+		     "1:\n"					\
+		     : "=a" (dummy)				\
+		     : "a" (count)				\
+		     : "memory", "ecx", "edx");			\
 } while (0)
 
 
@@ -50,8 +48,8 @@ do {									\
  * wasn't 1 originally. This function returns 0 if the fastpath succeeds,
  * or anything the slow path function returns
  */
-static inline int
-__mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *))
+static inline int __mutex_fastpath_lock_retval(atomic_t *count,
+					       int (*fail_fn)(atomic_t *))
 {
 	if (unlikely(atomic_dec_return(count) < 0))
 		return fail_fn(count);
@@ -72,22 +70,20 @@ __mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *))
  * __mutex_slowpath_needs_to_unlock() macro needs to return 1, it needs
  * to return 0 otherwise.
  */
-#define __mutex_fastpath_unlock(count, fail_fn)				\
-do {									\
-	unsigned int dummy;						\
-									\
-	typecheck(atomic_t *, count);					\
+#define __mutex_fastpath_unlock(count, fail_fn)			\
+do {								\
+	unsigned int dummy;					\
+								\
+	typecheck(atomic_t *, count);				\
 	typecheck_fn(void (*)(atomic_t *), fail_fn);		\
-									\
-	__asm__ __volatile__(						\
-		LOCK_PREFIX "   incl (%%eax)	\n"			\
-			"   jg	1f		\n"			\
-			"   call "#fail_fn"	\n"			\
-			"1:			\n"			\
-									\
-		:"=a" (dummy)						\
-		: "a" (count)						\
-		: "memory", "ecx", "edx");				\
+								\
+	asm volatile(LOCK_PREFIX "   incl (%%eax)\n"		\
+		     "   jg	1f\n"				\
+		     "   call " #fail_fn "\n"			\
+		     "1:\n"					\
+		     : "=a" (dummy)				\
+		     : "a" (count)				\
+		     : "memory", "ecx", "edx");			\
 } while (0)
 
 #define __mutex_slowpath_needs_to_unlock()	1
@@ -104,8 +100,8 @@ do {									\
  * Additionally, if the value was < 0 originally, this function must not leave
  * it to 0 on failure.
  */
-static inline int
-__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *))
+static inline int __mutex_fastpath_trylock(atomic_t *count,
+					   int (*fail_fn)(atomic_t *))
 {
 	/*
 	 * We have two variants here. The cmpxchg based one is the best one
-- 
cgit v1.2.3


From 2c4e8830414de84cc969a1f9685f4c48b91ca6e7 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:54 -0700
Subject: include/asm-x86/mutex_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mutex_64.h | 73 +++++++++++++++++++++-------------------------
 1 file changed, 34 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/mutex_64.h b/include/asm-x86/mutex_64.h
index 6c2949a3c677..f3fae9becb38 100644
--- a/include/asm-x86/mutex_64.h
+++ b/include/asm-x86/mutex_64.h
@@ -16,23 +16,21 @@
  *
  * Atomically decrements @v and calls <fail_fn> if the result is negative.
  */
-#define __mutex_fastpath_lock(v, fail_fn)				\
-do {									\
-	unsigned long dummy;						\
-									\
-	typecheck(atomic_t *, v);					\
-	typecheck_fn(void (*)(atomic_t *), fail_fn);			\
-									\
-	__asm__ __volatile__(						\
-		LOCK_PREFIX "   decl (%%rdi)	\n"			\
-			"   jns 1f		\n"			\
-			"   call "#fail_fn"	\n"			\
-			"1:"						\
-									\
-		:"=D" (dummy)						\
-		: "D" (v)						\
-		: "rax", "rsi", "rdx", "rcx",				\
-		  "r8", "r9", "r10", "r11", "memory");			\
+#define __mutex_fastpath_lock(v, fail_fn)			\
+do {								\
+	unsigned long dummy;					\
+								\
+	typecheck(atomic_t *, v);				\
+	typecheck_fn(void (*)(atomic_t *), fail_fn);		\
+								\
+	asm volatile(LOCK_PREFIX "   decl (%%rdi)\n"		\
+		     "   jns 1f		\n"			\
+		     "   call " #fail_fn "\n"			\
+		     "1:"					\
+		     : "=D" (dummy)				\
+		     : "D" (v)					\
+		     : "rax", "rsi", "rdx", "rcx",		\
+		       "r8", "r9", "r10", "r11", "memory");	\
 } while (0)
 
 /**
@@ -45,9 +43,8 @@ do {									\
  * it wasn't 1 originally. This function returns 0 if the fastpath succeeds,
  * or anything the slow path function returns
  */
-static inline int
-__mutex_fastpath_lock_retval(atomic_t *count,
-			     int (*fail_fn)(atomic_t *))
+static inline int __mutex_fastpath_lock_retval(atomic_t *count,
+					       int (*fail_fn)(atomic_t *))
 {
 	if (unlikely(atomic_dec_return(count) < 0))
 		return fail_fn(count);
@@ -62,23 +59,21 @@ __mutex_fastpath_lock_retval(atomic_t *count,
  *
  * Atomically increments @v and calls <fail_fn> if the result is nonpositive.
  */
-#define __mutex_fastpath_unlock(v, fail_fn)				\
-do {									\
-	unsigned long dummy;						\
-									\
-	typecheck(atomic_t *, v);					\
-	typecheck_fn(void (*)(atomic_t *), fail_fn);			\
-									\
-	__asm__ __volatile__(						\
-		LOCK_PREFIX "   incl (%%rdi)	\n"			\
-			"   jg 1f		\n"			\
-			"   call "#fail_fn"	\n"			\
-			"1:			  "			\
-									\
-		:"=D" (dummy)						\
-		: "D" (v)						\
-		: "rax", "rsi", "rdx", "rcx",				\
-		  "r8", "r9", "r10", "r11", "memory");			\
+#define __mutex_fastpath_unlock(v, fail_fn)			\
+do {								\
+	unsigned long dummy;					\
+								\
+	typecheck(atomic_t *, v);				\
+	typecheck_fn(void (*)(atomic_t *), fail_fn);		\
+								\
+	asm volatile(LOCK_PREFIX "   incl (%%rdi)\n"		\
+		     "   jg 1f\n"				\
+		     "   call " #fail_fn "\n"			\
+		     "1:"					\
+		     : "=D" (dummy)				\
+		     : "D" (v)					\
+		     : "rax", "rsi", "rdx", "rcx",		\
+		       "r8", "r9", "r10", "r11", "memory");	\
 } while (0)
 
 #define __mutex_slowpath_needs_to_unlock()	1
@@ -93,8 +88,8 @@ do {									\
  * if it wasn't 1 originally. [the fallback function is never used on
  * x86_64, because all x86_64 CPUs have a CMPXCHG instruction.]
  */
-static inline int
-__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *))
+static inline int __mutex_fastpath_trylock(atomic_t *count,
+					   int (*fail_fn)(atomic_t *))
 {
 	if (likely(atomic_cmpxchg(count, 1, 0) == 1))
 		return 1;
-- 
cgit v1.2.3


From cb046eed76b7f74e619479f1aba17e74ce6f5159 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:55 -0700
Subject: include/asm-x86/numa_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/numa_64.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h
index 15fe07cde586..32c22ae0709f 100644
--- a/include/asm-x86/numa_64.h
+++ b/include/asm-x86/numa_64.h
@@ -1,11 +1,12 @@
-#ifndef _ASM_X8664_NUMA_H 
+#ifndef _ASM_X8664_NUMA_H
 #define _ASM_X8664_NUMA_H 1
 
 #include <linux/nodemask.h>
 #include <asm/apicdef.h>
 
 struct bootnode {
-	u64 start,end; 
+	u64 start;
+	u64 end;
 };
 
 extern int compute_hash_shift(struct bootnode *nodes, int numnodes);
-- 
cgit v1.2.3


From 5f4e4b7209deb3cd7cf16ebb7bf84917e4b6682a Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:56 -0700
Subject: include/asm-x86/numaq.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/numaq.h | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/numaq.h b/include/asm-x86/numaq.h
index 38f710dc37f2..94b86c31239a 100644
--- a/include/asm-x86/numaq.h
+++ b/include/asm-x86/numaq.h
@@ -3,7 +3,7 @@
  *
  * Copyright (C) 2002, IBM Corp.
  *
- * All rights reserved.          
+ * All rights reserved.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -33,7 +33,8 @@ extern int get_memcfg_numaq(void);
 /*
  * SYS_CFG_DATA_PRIV_ADDR, struct eachquadmem, and struct sys_cfg_data are the
  */
-#define SYS_CFG_DATA_PRIV_ADDR		0x0009d000 /* place for scd in private quad space */
+#define SYS_CFG_DATA_PRIV_ADDR		0x0009d000 /* place for scd in private
+						      quad space */
 
 /*
  * Communication area for each processor on lynxer-processor tests.
@@ -139,7 +140,7 @@ struct sys_cfg_data {
 	unsigned int	low_shrd_mem_base; /* 0 or 512MB or 1GB */
 	unsigned int	low_shrd_mem_quad_offset; /* 0,128M,256M,512M,1G */
 					/* may not be totally populated */
-	unsigned int	split_mem_enbl; /* 0 for no low shared memory */ 
+	unsigned int	split_mem_enbl; /* 0 for no low shared memory */
 	unsigned int	mmio_sz; /* Size of total system memory mapped I/O */
 				 /* (in MB). */
 	unsigned int	quad_spin_lock; /* Spare location used for quad */
@@ -152,7 +153,7 @@ struct sys_cfg_data {
 	/*
 	 *	memory configuration area for each quad
 	 */
-        struct	eachquadmem eq[MAX_NUMNODES];	/* indexed by quad id */
+	struct		eachquadmem eq[MAX_NUMNODES];	/* indexed by quad id */
 };
 
 static inline unsigned long *get_zholes_size(int nid)
-- 
cgit v1.2.3


From 095d1c4e61d77de4b33ea3b202aa6342f69d9891 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:57 -0700
Subject: include/asm-x86/page_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/page_32.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index 5f7257fd589b..424e82f8ae27 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -47,7 +47,10 @@ typedef unsigned long	pgdval_t;
 typedef unsigned long	pgprotval_t;
 typedef unsigned long	phys_addr_t;
 
-typedef union { pteval_t pte, pte_low; } pte_t;
+typedef union {
+	pteval_t pte;
+	pteval_t pte_low;
+} pte_t;
 
 #endif	/* __ASSEMBLY__ */
 #endif	/* CONFIG_X86_PAE */
@@ -61,7 +64,7 @@ typedef struct page *pgtable_t;
 #endif
 
 #ifndef __ASSEMBLY__
-#define __phys_addr(x)		((x)-PAGE_OFFSET)
+#define __phys_addr(x)		((x) - PAGE_OFFSET)
 #define __phys_reloc_hide(x)	RELOC_HIDE((x), 0)
 
 #ifdef CONFIG_FLATMEM
@@ -78,7 +81,7 @@ extern unsigned int __VMALLOC_RESERVE;
 extern int sysctl_legacy_va_layout;
 
 #define VMALLOC_RESERVE		((unsigned long)__VMALLOC_RESERVE)
-#define MAXMEM			(-__PAGE_OFFSET-__VMALLOC_RESERVE)
+#define MAXMEM			(-__PAGE_OFFSET - __VMALLOC_RESERVE)
 
 #ifdef CONFIG_X86_USE_3DNOW
 #include <asm/mmx.h>
-- 
cgit v1.2.3


From b20a4615944a0c106fce2aecb7ea1dbc8eefc71b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:58 -0700
Subject: include/asm-x86/page_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/page_64.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index aee05c616e05..f156778f707c 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -5,7 +5,7 @@
 
 #define THREAD_ORDER	1
 #define THREAD_SIZE  (PAGE_SIZE << THREAD_ORDER)
-#define CURRENT_MASK (~(THREAD_SIZE-1))
+#define CURRENT_MASK (~(THREAD_SIZE - 1))
 
 #define EXCEPTION_STACK_ORDER 0
 #define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER)
@@ -51,7 +51,7 @@
  * Kernel image size is limited to 512 MB (see level2_kernel_pgt in
  * arch/x86/kernel/head_64.S), and it is mapped here:
  */
-#define KERNEL_IMAGE_SIZE	(512*1024*1024)
+#define KERNEL_IMAGE_SIZE	(512 * 1024 * 1024)
 #define KERNEL_IMAGE_START	_AC(0xffffffff80000000, UL)
 
 #ifndef __ASSEMBLY__
-- 
cgit v1.2.3


From fad599854e3997a3e93559e19759a26b18c906c6 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:02:59 -0700
Subject: include/asm-x86/param.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/param.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/param.h b/include/asm-x86/param.h
index c996ec4da0c8..6f0d0422f4ca 100644
--- a/include/asm-x86/param.h
+++ b/include/asm-x86/param.h
@@ -3,8 +3,8 @@
 
 #ifdef __KERNEL__
 # define HZ		CONFIG_HZ	/* Internal kernel timer frequency */
-# define USER_HZ	100		/* .. some user interfaces are in "ticks" */
-# define CLOCKS_PER_SEC	(USER_HZ)       /* like times() */
+# define USER_HZ	100		/* some user interfaces are */
+# define CLOCKS_PER_SEC	(USER_HZ)       /* in "ticks" like times() */
 #endif
 
 #ifndef HZ
-- 
cgit v1.2.3


From 49cd740bb0b5796f34699a0f945b977f6ff34c64 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:00 -0700
Subject: include/asm-x86/paravirt.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/paravirt.h | 47 +++++++++++++++++++++++++++-------------------
 1 file changed, 28 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index d6236eb46466..0c23f7940bc4 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -231,7 +231,8 @@ struct pv_mmu_ops {
 	void (*set_pte_at)(struct mm_struct *mm, unsigned long addr,
 			   pte_t *ptep, pte_t pteval);
 	void (*set_pmd)(pmd_t *pmdp, pmd_t pmdval);
-	void (*pte_update)(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
+	void (*pte_update)(struct mm_struct *mm, unsigned long addr,
+			   pte_t *ptep);
 	void (*pte_update_defer)(struct mm_struct *mm,
 				 unsigned long addr, pte_t *ptep);
 
@@ -246,7 +247,8 @@ struct pv_mmu_ops {
 	void (*set_pte_atomic)(pte_t *ptep, pte_t pteval);
 	void (*set_pte_present)(struct mm_struct *mm, unsigned long addr,
 				pte_t *ptep, pte_t pte);
-	void (*pte_clear)(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
+	void (*pte_clear)(struct mm_struct *mm, unsigned long addr,
+			  pte_t *ptep);
 	void (*pmd_clear)(pmd_t *pmdp);
 
 #endif	/* CONFIG_X86_PAE */
@@ -274,8 +276,7 @@ struct pv_mmu_ops {
 /* This contains all the paravirt structures: we get a convenient
  * number for each function using the offset which we use to indicate
  * what to patch. */
-struct paravirt_patch_template
-{
+struct paravirt_patch_template {
 	struct pv_init_ops pv_init_ops;
 	struct pv_time_ops pv_time_ops;
 	struct pv_cpu_ops pv_cpu_ops;
@@ -660,32 +661,37 @@ static inline int paravirt_write_msr(unsigned msr, unsigned low, unsigned high)
 }
 
 /* These should all do BUG_ON(_err), but our headers are too tangled. */
-#define rdmsr(msr,val1,val2) do {		\
+#define rdmsr(msr, val1, val2)			\
+do {						\
 	int _err;				\
 	u64 _l = paravirt_read_msr(msr, &_err);	\
 	val1 = (u32)_l;				\
 	val2 = _l >> 32;			\
-} while(0)
+} while (0)
 
-#define wrmsr(msr,val1,val2) do {		\
+#define wrmsr(msr, val1, val2)			\
+do {						\
 	paravirt_write_msr(msr, val1, val2);	\
-} while(0)
+} while (0)
 
-#define rdmsrl(msr,val) do {			\
+#define rdmsrl(msr, val)			\
+do {						\
 	int _err;				\
 	val = paravirt_read_msr(msr, &_err);	\
-} while(0)
+} while (0)
 
-#define wrmsrl(msr,val)		wrmsr(msr, (u32)((u64)(val)), ((u64)(val))>>32)
-#define wrmsr_safe(msr,a,b)	paravirt_write_msr(msr, a, b)
+#define wrmsrl(msr, val)	wrmsr(msr, (u32)((u64)(val)), ((u64)(val))>>32)
+#define wrmsr_safe(msr, a, b)	paravirt_write_msr(msr, a, b)
 
 /* rdmsr with exception handling */
-#define rdmsr_safe(msr,a,b) ({			\
+#define rdmsr_safe(msr, a, b)			\
+({						\
 	int _err;				\
 	u64 _l = paravirt_read_msr(msr, &_err);	\
 	(*a) = (u32)_l;				\
 	(*b) = _l >> 32;			\
-	_err; })
+	_err;					\
+})
 
 
 static inline u64 paravirt_read_tsc(void)
@@ -693,10 +699,11 @@ static inline u64 paravirt_read_tsc(void)
 	return PVOP_CALL0(u64, pv_cpu_ops.read_tsc);
 }
 
-#define rdtscl(low) do {			\
+#define rdtscl(low)				\
+do {						\
 	u64 _l = paravirt_read_tsc();		\
 	low = (int)_l;				\
-} while(0)
+} while (0)
 
 #define rdtscll(val) (val = paravirt_read_tsc())
 
@@ -711,11 +718,12 @@ static inline unsigned long long paravirt_read_pmc(int counter)
 	return PVOP_CALL1(u64, pv_cpu_ops.read_pmc, counter);
 }
 
-#define rdpmc(counter,low,high) do {		\
+#define rdpmc(counter, low, high)		\
+do {						\
 	u64 _l = paravirt_read_pmc(counter);	\
 	low = (u32)_l;				\
 	high = _l >> 32;			\
-} while(0)
+} while (0)
 
 static inline unsigned long long paravirt_rdtscp(unsigned int *aux)
 {
@@ -794,7 +802,8 @@ static inline void set_iopl_mask(unsigned mask)
 }
 
 /* The paravirtualized I/O functions */
-static inline void slow_down_io(void) {
+static inline void slow_down_io(void)
+{
 	pv_cpu_ops.io_delay();
 #ifdef REALLY_SLOW_IO
 	pv_cpu_ops.io_delay();
-- 
cgit v1.2.3


From 5269354231c6e960941f6e1fd0737acc003d2353 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:01 -0700
Subject: include/asm-x86/parport.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/parport.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/parport.h b/include/asm-x86/parport.h
index 019cbca24a38..3c4ffeb467e9 100644
--- a/include/asm-x86/parport.h
+++ b/include/asm-x86/parport.h
@@ -1,10 +1,10 @@
 #ifndef _ASM_X86_PARPORT_H
 #define _ASM_X86_PARPORT_H
 
-static int __devinit parport_pc_find_isa_ports (int autoirq, int autodma);
-static int __devinit parport_pc_find_nonpci_ports (int autoirq, int autodma)
+static int __devinit parport_pc_find_isa_ports(int autoirq, int autodma);
+static int __devinit parport_pc_find_nonpci_ports(int autoirq, int autodma)
 {
-	return parport_pc_find_isa_ports (autoirq, autodma);
+	return parport_pc_find_isa_ports(autoirq, autodma);
 }
 
 #endif /* _ASM_X86_PARPORT_H */
-- 
cgit v1.2.3


From 3cb47d79e99ad7b3a02badd7f64fbb57a1b125b3 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:02 -0700
Subject: include/asm-x86/pci_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pci_64.h | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pci_64.h b/include/asm-x86/pci_64.h
index 374690314539..df867e5d80b1 100644
--- a/include/asm-x86/pci_64.h
+++ b/include/asm-x86/pci_64.h
@@ -1,12 +1,10 @@
 #ifndef __x8664_PCI_H
 #define __x8664_PCI_H
 
-
 #ifdef __KERNEL__
 
-
 #ifdef CONFIG_CALGARY_IOMMU
-static inline void* pci_iommu(struct pci_bus *bus)
+static inline void *pci_iommu(struct pci_bus *bus)
 {
 	struct pci_sysdata *sd = bus->sysdata;
 	return sd->iommu;
@@ -19,11 +17,10 @@ static inline void set_pci_iommu(struct pci_bus *bus, void *val)
 }
 #endif /* CONFIG_CALGARY_IOMMU */
 
-
-extern int (*pci_config_read)(int seg, int bus, int dev, int fn, int reg, int len, u32 *value);
-extern int (*pci_config_write)(int seg, int bus, int dev, int fn, int reg, int len, u32 value);
-
-
+extern int (*pci_config_read)(int seg, int bus, int dev, int fn,
+			      int reg, int len, u32 *value);
+extern int (*pci_config_write)(int seg, int bus, int dev, int fn,
+			       int reg, int len, u32 value);
 
 extern void pci_iommu_alloc(void);
 
@@ -65,5 +62,4 @@ extern void pci_iommu_alloc(void);
 
 #endif /* __KERNEL__ */
 
-
 #endif /* __x8664_PCI_H */
-- 
cgit v1.2.3


From 565c6402db09b1d1b7623ebd629f3adf6b86feaa Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:03 -0700
Subject: include/asm-x86/pci-direct.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pci-direct.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/pci-direct.h b/include/asm-x86/pci-direct.h
index 6823fa4f1afa..5b21485be573 100644
--- a/include/asm-x86/pci-direct.h
+++ b/include/asm-x86/pci-direct.h
@@ -4,7 +4,7 @@
 #include <linux/types.h>
 
 /* Direct PCI access. This is used for PCI accesses in early boot before
-   the PCI subsystem works. */ 
+   the PCI subsystem works. */
 
 extern u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset);
 extern u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset);
-- 
cgit v1.2.3


From 69bdb7bcc6b9494b1738c6f6dd3fe553c9c6978e Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:04 -0700
Subject: include/asm-x86/pci.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pci.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pci.h b/include/asm-x86/pci.h
index c61190cb9e12..ddd8e248fc0a 100644
--- a/include/asm-x86/pci.h
+++ b/include/asm-x86/pci.h
@@ -8,14 +8,13 @@
 #include <asm/scatterlist.h>
 #include <asm/io.h>
 
-
 #ifdef __KERNEL__
 
 struct pci_sysdata {
 	int		domain;		/* PCI domain */
 	int		node;		/* NUMA node */
 #ifdef CONFIG_X86_64
-	void*		iommu;		/* IOMMU private data */
+	void		*iommu;		/* IOMMU private data */
 #endif
 };
 
@@ -52,7 +51,7 @@ extern unsigned long pci_mem_start;
 #define PCIBIOS_MIN_CARDBUS_IO	0x4000
 
 void pcibios_config_init(void);
-struct pci_bus * pcibios_scan_root(int bus);
+struct pci_bus *pcibios_scan_root(int bus);
 
 void pcibios_set_master(struct pci_dev *dev);
 void pcibios_penalize_isa_irq(int irq, int active);
@@ -62,7 +61,8 @@ int pcibios_set_irq_routing(struct pci_dev *dev, int pin, int irq);
 
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
-			       enum pci_mmap_state mmap_state, int write_combine);
+			       enum pci_mmap_state mmap_state,
+			       int write_combine);
 
 
 #ifdef CONFIG_PCI
-- 
cgit v1.2.3


From 46e1abc63e736644265e7ec2897f963a4ace5993 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:05 -0700
Subject: include/asm-x86/pda.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pda.h | 80 +++++++++++++++++++++++++++------------------------
 1 file changed, 42 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pda.h b/include/asm-x86/pda.h
index d9dc209c24ad..101fb9e11954 100644
--- a/include/asm-x86/pda.h
+++ b/include/asm-x86/pda.h
@@ -57,34 +57,36 @@ extern struct x8664_pda _proxy_pda;
 
 #define pda_offset(field) offsetof(struct x8664_pda, field)
 
-#define pda_to_op(op, field, val) do {		\
-	typedef typeof(_proxy_pda.field) T__;	\
-	if (0) { T__ tmp__; tmp__ = (val); }	/* type checking */ \
-	switch (sizeof(_proxy_pda.field)) {	\
-	case 2:					\
-		asm(op "w %1,%%gs:%c2" :	\
-		    "+m" (_proxy_pda.field) :	\
-		    "ri" ((T__)val),		\
-		    "i"(pda_offset(field)));	\
-		break;				\
-	case 4:					\
-		asm(op "l %1,%%gs:%c2" :	\
-		    "+m" (_proxy_pda.field) :	\
-		    "ri" ((T__)val),		\
-		    "i" (pda_offset(field)));	\
-		break;				\
-	case 8:					\
-		asm(op "q %1,%%gs:%c2":		\
-		    "+m" (_proxy_pda.field) :	\
-		    "ri" ((T__)val),		\
-		    "i"(pda_offset(field)));	\
-		break;				\
-	default:				\
-		__bad_pda_field();		\
-	}					\
-	} while (0)
+#define pda_to_op(op, field, val)					\
+do {									\
+	typedef typeof(_proxy_pda.field) T__;				\
+	if (0) { T__ tmp__; tmp__ = (val); }	/* type checking */	\
+	switch (sizeof(_proxy_pda.field)) {				\
+	case 2:								\
+		asm(op "w %1,%%gs:%c2" :				\
+		    "+m" (_proxy_pda.field) :				\
+		    "ri" ((T__)val),					\
+		    "i"(pda_offset(field)));				\
+		break;							\
+	case 4:								\
+		asm(op "l %1,%%gs:%c2" :				\
+		    "+m" (_proxy_pda.field) :				\
+		    "ri" ((T__)val),					\
+		    "i" (pda_offset(field)));				\
+		break;							\
+	case 8:								\
+		asm(op "q %1,%%gs:%c2":					\
+		    "+m" (_proxy_pda.field) :				\
+		    "ri" ((T__)val),					\
+		    "i"(pda_offset(field)));				\
+		break;							\
+	default:							\
+		__bad_pda_field();					\
+	}								\
+} while (0)
 
-#define pda_from_op(op,field) ({		\
+#define pda_from_op(op, field)			\
+({						\
 	typeof(_proxy_pda.field) ret__;		\
 	switch (sizeof(_proxy_pda.field)) {	\
 	case 2:					\
@@ -92,23 +94,24 @@ extern struct x8664_pda _proxy_pda;
 		    "=r" (ret__) :		\
 		    "i" (pda_offset(field)),	\
 		    "m" (_proxy_pda.field));	\
-		 break;				\
+		break;				\
 	case 4:					\
 		asm(op "l %%gs:%c1,%0":		\
 		    "=r" (ret__):		\
 		    "i" (pda_offset(field)),	\
 		    "m" (_proxy_pda.field));	\
-		 break;				\
+		break;				\
 	case 8:					\
 		asm(op "q %%gs:%c1,%0":		\
 		    "=r" (ret__) :		\
 		    "i" (pda_offset(field)),	\
 		    "m" (_proxy_pda.field));	\
-		 break;				\
+		break;				\
 	default:				\
 		__bad_pda_field();		\
-       }					\
-       ret__; })
+	}					\
+	ret__;					\
+})
 
 #define read_pda(field)		pda_from_op("mov", field)
 #define write_pda(field, val)	pda_to_op("mov", field, val)
@@ -117,12 +120,13 @@ extern struct x8664_pda _proxy_pda;
 #define or_pda(field, val)	pda_to_op("or", field, val)
 
 /* This is not atomic against other CPUs -- CPU preemption needs to be off */
-#define test_and_clear_bit_pda(bit, field) ({		\
-	int old__;						\
-	asm volatile("btr %2,%%gs:%c3\n\tsbbl %0,%0"		\
-	    : "=r" (old__), "+m" (_proxy_pda.field)		\
-	    : "dIr" (bit), "i" (pda_offset(field)) : "memory");	\
-	old__;							\
+#define test_and_clear_bit_pda(bit, field)				\
+({									\
+	int old__;							\
+	asm volatile("btr %2,%%gs:%c3\n\tsbbl %0,%0"			\
+		     : "=r" (old__), "+m" (_proxy_pda.field)		\
+		     : "dIr" (bit), "i" (pda_offset(field)) : "memory");\
+	old__;								\
 })
 
 #endif
-- 
cgit v1.2.3


From bc9e3be20bab447519e864c4f9e09aae82ed0376 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:06 -0700
Subject: include/asm-x86/percpu.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/percpu.h | 104 ++++++++++++++++++++++++-----------------------
 1 file changed, 54 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/percpu.h b/include/asm-x86/percpu.h
index 0dec00f27eb4..736fc3bb8e1e 100644
--- a/include/asm-x86/percpu.h
+++ b/include/asm-x86/percpu.h
@@ -85,58 +85,62 @@ DECLARE_PER_CPU(unsigned long, this_cpu_off);
  * don't give an lvalue though). */
 extern void __bad_percpu_size(void);
 
-#define percpu_to_op(op,var,val)				\
-	do {							\
-		typedef typeof(var) T__;			\
-		if (0) { T__ tmp__; tmp__ = (val); }		\
-		switch (sizeof(var)) {				\
-		case 1:						\
-			asm(op "b %1,"__percpu_seg"%0"		\
-			    : "+m" (var)			\
-			    :"ri" ((T__)val));			\
-			break;					\
-		case 2:						\
-			asm(op "w %1,"__percpu_seg"%0"		\
-			    : "+m" (var)			\
-			    :"ri" ((T__)val));			\
-			break;					\
-		case 4:						\
-			asm(op "l %1,"__percpu_seg"%0"		\
-			    : "+m" (var)			\
-			    :"ri" ((T__)val));			\
-			break;					\
-		default: __bad_percpu_size();			\
-		}						\
-	} while (0)
-
-#define percpu_from_op(op,var)					\
-	({							\
-		typeof(var) ret__;				\
-		switch (sizeof(var)) {				\
-		case 1:						\
-			asm(op "b "__percpu_seg"%1,%0"		\
-			    : "=r" (ret__)			\
-			    : "m" (var));			\
-			break;					\
-		case 2:						\
-			asm(op "w "__percpu_seg"%1,%0"		\
-			    : "=r" (ret__)			\
-			    : "m" (var));			\
-			break;					\
-		case 4:						\
-			asm(op "l "__percpu_seg"%1,%0"		\
-			    : "=r" (ret__)			\
-			    : "m" (var));			\
-			break;					\
-		default: __bad_percpu_size();			\
-		}						\
-		ret__; })
+#define percpu_to_op(op, var, val)			\
+do {							\
+	typedef typeof(var) T__;			\
+	if (0) {					\
+		T__ tmp__;				\
+		tmp__ = (val);				\
+	}						\
+	switch (sizeof(var)) {				\
+	case 1:						\
+		asm(op "b %1,"__percpu_seg"%0"		\
+		    : "+m" (var)			\
+		    : "ri" ((T__)val));			\
+		break;					\
+	case 2:						\
+		asm(op "w %1,"__percpu_seg"%0"		\
+		    : "+m" (var)			\
+		    : "ri" ((T__)val));			\
+		break;					\
+	case 4:						\
+		asm(op "l %1,"__percpu_seg"%0"		\
+		    : "+m" (var)			\
+		    : "ri" ((T__)val));			\
+		break;					\
+	default: __bad_percpu_size();			\
+	}						\
+} while (0)
+
+#define percpu_from_op(op, var)				\
+({							\
+	typeof(var) ret__;				\
+	switch (sizeof(var)) {				\
+	case 1:						\
+		asm(op "b "__percpu_seg"%1,%0"		\
+		    : "=r" (ret__)			\
+		    : "m" (var));			\
+		break;					\
+	case 2:						\
+		asm(op "w "__percpu_seg"%1,%0"		\
+		    : "=r" (ret__)			\
+		    : "m" (var));			\
+		break;					\
+	case 4:						\
+		asm(op "l "__percpu_seg"%1,%0"		\
+		    : "=r" (ret__)			\
+		    : "m" (var));			\
+		break;					\
+	default: __bad_percpu_size();			\
+	}						\
+	ret__;						\
+})
 
 #define x86_read_percpu(var) percpu_from_op("mov", per_cpu__##var)
-#define x86_write_percpu(var,val) percpu_to_op("mov", per_cpu__##var, val)
-#define x86_add_percpu(var,val) percpu_to_op("add", per_cpu__##var, val)
-#define x86_sub_percpu(var,val) percpu_to_op("sub", per_cpu__##var, val)
-#define x86_or_percpu(var,val) percpu_to_op("or", per_cpu__##var, val)
+#define x86_write_percpu(var, val) percpu_to_op("mov", per_cpu__##var, val)
+#define x86_add_percpu(var, val) percpu_to_op("add", per_cpu__##var, val)
+#define x86_sub_percpu(var, val) percpu_to_op("sub", per_cpu__##var, val)
+#define x86_or_percpu(var, val) percpu_to_op("or", per_cpu__##var, val)
 #endif /* !__ASSEMBLY__ */
 #endif /* !CONFIG_X86_64 */
 #endif /* _ASM_X86_PERCPU_H_ */
-- 
cgit v1.2.3


From 65e05d15edfdd6ecb4426894cf6e6b5ae97602e4 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:08 -0700
Subject: include/asm-x86/pgtable-2level.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pgtable-2level.h | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable-2level.h b/include/asm-x86/pgtable-2level.h
index 701404fab308..46bc52c0eae1 100644
--- a/include/asm-x86/pgtable-2level.h
+++ b/include/asm-x86/pgtable-2level.h
@@ -26,7 +26,8 @@ static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
 	native_set_pte(ptep, pte);
 }
 
-static inline void native_set_pte_present(struct mm_struct *mm, unsigned long addr,
+static inline void native_set_pte_present(struct mm_struct *mm,
+					  unsigned long addr,
 					  pte_t *ptep, pte_t pte)
 {
 	native_set_pte(ptep, pte);
@@ -37,7 +38,8 @@ static inline void native_pmd_clear(pmd_t *pmdp)
 	native_set_pmd(pmdp, __pmd(0));
 }
 
-static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *xp)
+static inline void native_pte_clear(struct mm_struct *mm,
+				    unsigned long addr, pte_t *xp)
 {
 	*xp = native_make_pte(0);
 }
@@ -61,16 +63,18 @@ static inline pte_t native_ptep_get_and_clear(pte_t *xp)
  */
 #define PTE_FILE_MAX_BITS	29
 
-#define pte_to_pgoff(pte) \
-	((((pte).pte_low >> 1) & 0x1f ) + (((pte).pte_low >> 8) << 5 ))
+#define pte_to_pgoff(pte)						\
+	((((pte).pte_low >> 1) & 0x1f) + (((pte).pte_low >> 8) << 5))
 
-#define pgoff_to_pte(off) \
-	((pte_t) { .pte_low = (((off) & 0x1f) << 1) + (((off) >> 5) << 8) + _PAGE_FILE })
+#define pgoff_to_pte(off)						\
+	((pte_t) { .pte_low = (((off) & 0x1f) << 1) +			\
+			(((off) >> 5) << 8) + _PAGE_FILE })
 
 /* Encode and de-code a swap entry */
 #define __swp_type(x)			(((x).val >> 1) & 0x1f)
 #define __swp_offset(x)			((x).val >> 8)
-#define __swp_entry(type, offset)	((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
+#define __swp_entry(type, offset)				\
+	((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
 #define __pte_to_swp_entry(pte)		((swp_entry_t) { (pte).pte_low })
 #define __swp_entry_to_pte(x)		((pte_t) { .pte = (x).val })
 
-- 
cgit v1.2.3


From cf840147d48626d5d86d617cbc5b7cddc1bcae14 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:09 -0700
Subject: include/asm-x86/pgtable_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pgtable_32.h | 102 ++++++++++++++++++++++---------------------
 1 file changed, 53 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h
index 1e2c0d839528..c4a643674458 100644
--- a/include/asm-x86/pgtable_32.h
+++ b/include/asm-x86/pgtable_32.h
@@ -40,13 +40,13 @@ void paging_init(void);
 #ifdef CONFIG_X86_PAE
 # include <asm/pgtable-3level-defs.h>
 # define PMD_SIZE	(1UL << PMD_SHIFT)
-# define PMD_MASK	(~(PMD_SIZE-1))
+# define PMD_MASK	(~(PMD_SIZE - 1))
 #else
 # include <asm/pgtable-2level-defs.h>
 #endif
 
 #define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
-#define PGDIR_MASK	(~(PGDIR_SIZE-1))
+#define PGDIR_MASK	(~(PGDIR_SIZE - 1))
 
 #define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
 #define KERNEL_PGD_PTRS (PTRS_PER_PGD-USER_PGD_PTRS)
@@ -58,21 +58,22 @@ void paging_init(void);
  * The vmalloc() routines leaves a hole of 4kB between each vmalloced
  * area for the same reason. ;)
  */
-#define VMALLOC_OFFSET	(8*1024*1024)
-#define VMALLOC_START	(((unsigned long) high_memory + \
-			2*VMALLOC_OFFSET-1) & ~(VMALLOC_OFFSET-1))
+#define VMALLOC_OFFSET	(8 * 1024 * 1024)
+#define VMALLOC_START	(((unsigned long)high_memory + 2 * VMALLOC_OFFSET - 1) \
+			 & ~(VMALLOC_OFFSET - 1))
 #ifdef CONFIG_X86_PAE
 #define LAST_PKMAP 512
 #else
 #define LAST_PKMAP 1024
 #endif
 
-#define PKMAP_BASE ((FIXADDR_BOOT_START - PAGE_SIZE*(LAST_PKMAP + 1)) & PMD_MASK)
+#define PKMAP_BASE ((FIXADDR_BOOT_START - PAGE_SIZE * (LAST_PKMAP + 1))	\
+		    & PMD_MASK)
 
 #ifdef CONFIG_HIGHMEM
-# define VMALLOC_END	(PKMAP_BASE-2*PAGE_SIZE)
+# define VMALLOC_END	(PKMAP_BASE - 2 * PAGE_SIZE)
 #else
-# define VMALLOC_END	(FIXADDR_START-2*PAGE_SIZE)
+# define VMALLOC_END	(FIXADDR_START - 2 * PAGE_SIZE)
 #endif
 
 /*
@@ -88,16 +89,16 @@ extern unsigned long pg0[];
 #define pte_present(x)	((x).pte_low & (_PAGE_PRESENT | _PAGE_PROTNONE))
 
 /* To avoid harmful races, pmd_none(x) should check only the lower when PAE */
-#define pmd_none(x)	(!(unsigned long)pmd_val(x))
-#define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
+#define pmd_none(x)	(!(unsigned long)pmd_val((x)))
+#define pmd_present(x)	(pmd_val((x)) & _PAGE_PRESENT)
 
 extern int pmd_bad(pmd_t pmd);
 
-#define pmd_bad_v1(x)	((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
-#define	pmd_bad_v2(x)	((pmd_val(x) \
-			  & ~(PAGE_MASK | _PAGE_USER | _PAGE_PSE | _PAGE_NX)) \
-			 != _KERNPG_TABLE)
-
+#define pmd_bad_v1(x)							\
+	(_KERNPG_TABLE != (pmd_val((x)) & ~(PAGE_MASK | _PAGE_USER)))
+#define	pmd_bad_v2(x)							\
+	(_KERNPG_TABLE != (pmd_val((x)) & ~(PAGE_MASK | _PAGE_USER |	\
+					    _PAGE_PSE | _PAGE_NX)))
 
 #define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
 
@@ -123,17 +124,18 @@ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
 }
 
 /*
- * Macro to mark a page protection value as "uncacheable".  On processors which do not support
- * it, this is a no-op.
+ * Macro to mark a page protection value as "uncacheable".
+ * On processors which do not support it, this is a no-op.
  */
-#define pgprot_noncached(prot)	((boot_cpu_data.x86 > 3)					  \
-				 ? (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT)) : (prot))
+#define pgprot_noncached(prot)					\
+	((boot_cpu_data.x86 > 3)				\
+	 ? (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))	\
+	 : (prot))
 
 /*
  * Conversion functions: convert a page and protection to a page entry,
  * and a page entry and page directory to the page they refer to.
  */
-
 #define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
 
 /*
@@ -142,20 +144,20 @@ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
  * this macro returns the index of the entry in the pgd page which would
  * control the given virtual address
  */
-#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
-#define pgd_index_k(addr) pgd_index(addr)
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))
+#define pgd_index_k(addr) pgd_index((addr))
 
 /*
  * pgd_offset() returns a (pgd_t *)
  * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
  */
-#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
+#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address)))
 
 /*
  * a shortcut which implies the use of the kernel's pgd, instead
  * of a process's
  */
-#define pgd_offset_k(address) pgd_offset(&init_mm, address)
+#define pgd_offset_k(address) pgd_offset(&init_mm, (address))
 
 static inline int pud_large(pud_t pud) { return 0; }
 
@@ -165,8 +167,8 @@ static inline int pud_large(pud_t pud) { return 0; }
  * this macro returns the index of the entry in the pmd page which would
  * control the given virtual address
  */
-#define pmd_index(address) \
-		(((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
+#define pmd_index(address)				\
+	(((address) >> PMD_SHIFT) & (PTRS_PER_PMD - 1))
 
 /*
  * the pte page can be thought of an array like this: pte_t[PTRS_PER_PTE]
@@ -174,43 +176,45 @@ static inline int pud_large(pud_t pud) { return 0; }
  * this macro returns the index of the entry in the pte page which would
  * control the given virtual address
  */
-#define pte_index(address) \
-		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
+#define pte_index(address)					\
+	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset_kernel(dir, address)				\
+	((pte_t *)pmd_page_vaddr(*(dir)) +  pte_index((address)))
 
-#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
+#define pmd_page(pmd) (pfn_to_page(pmd_val((pmd)) >> PAGE_SHIFT))
 
-#define pmd_page_vaddr(pmd) \
-		((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page_vaddr(pmd)					\
+	((unsigned long)__va(pmd_val((pmd)) & PAGE_MASK))
 
 #if defined(CONFIG_HIGHPTE)
-#define pte_offset_map(dir, address) \
-	((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE0) + pte_index(address))
-#define pte_offset_map_nested(dir, address) \
-	((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE1) + pte_index(address))
-#define pte_unmap(pte) kunmap_atomic(pte, KM_PTE0)
-#define pte_unmap_nested(pte) kunmap_atomic(pte, KM_PTE1)
+#define pte_offset_map(dir, address)					\
+	((pte_t *)kmap_atomic_pte(pmd_page(*(dir)), KM_PTE0) +		\
+	 pte_index((address)))
+#define pte_offset_map_nested(dir, address)				\
+	((pte_t *)kmap_atomic_pte(pmd_page(*(dir)), KM_PTE1) +		\
+	 pte_index((address)))
+#define pte_unmap(pte) kunmap_atomic((pte), KM_PTE0)
+#define pte_unmap_nested(pte) kunmap_atomic((pte), KM_PTE1)
 #else
-#define pte_offset_map(dir, address) \
-	((pte_t *)page_address(pmd_page(*(dir))) + pte_index(address))
-#define pte_offset_map_nested(dir, address) pte_offset_map(dir, address)
+#define pte_offset_map(dir, address)					\
+	((pte_t *)page_address(pmd_page(*(dir))) + pte_index((address)))
+#define pte_offset_map_nested(dir, address) pte_offset_map((dir), (address))
 #define pte_unmap(pte) do { } while (0)
 #define pte_unmap_nested(pte) do { } while (0)
 #endif
 
 /* Clear a kernel PTE and flush it from the TLB */
-#define kpte_clear_flush(ptep, vaddr)					\
-do {									\
-	pte_clear(&init_mm, vaddr, ptep);				\
-	__flush_tlb_one(vaddr);						\
+#define kpte_clear_flush(ptep, vaddr)		\
+do {						\
+	pte_clear(&init_mm, (vaddr), (ptep));	\
+	__flush_tlb_one((vaddr));		\
 } while (0)
 
 /*
  * The i386 doesn't have any external MMU info: the kernel page
  * tables contain all the necessary information.
  */
-#define update_mmu_cache(vma,address,pte) do { } while (0)
+#define update_mmu_cache(vma, address, pte) do { } while (0)
 
 void native_pagetable_setup_start(pgd_t *base);
 void native_pagetable_setup_done(pgd_t *base);
@@ -239,7 +243,7 @@ static inline void paravirt_pagetable_setup_done(pgd_t *base)
 #define kern_addr_valid(kaddr)	(0)
 #endif
 
-#define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
-		remap_pfn_range(vma, vaddr, pfn, size, prot)
+#define io_remap_pfn_range(vma, vaddr, pfn, size, prot)	\
+	remap_pfn_range(vma, vaddr, pfn, size, prot)
 
 #endif /* _I386_PGTABLE_H */
-- 
cgit v1.2.3


From 4b01fef89a10cedbae9857e76283616af3f177cd Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:10 -0700
Subject: include/asm-x86/pgtable-3level.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pgtable-3level.h | 48 +++++++++++++++++++++++-----------------
 1 file changed, 28 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable-3level.h b/include/asm-x86/pgtable-3level.h
index 1d763eec740f..8b4a9d44b7f4 100644
--- a/include/asm-x86/pgtable-3level.h
+++ b/include/asm-x86/pgtable-3level.h
@@ -8,22 +8,26 @@
  * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
  */
 
-#define pte_ERROR(e) \
-	printk("%s:%d: bad pte %p(%08lx%08lx).\n", __FILE__, __LINE__, &(e), (e).pte_high, (e).pte_low)
-#define pmd_ERROR(e) \
-	printk("%s:%d: bad pmd %p(%016Lx).\n", __FILE__, __LINE__, &(e), pmd_val(e))
-#define pgd_ERROR(e) \
-	printk("%s:%d: bad pgd %p(%016Lx).\n", __FILE__, __LINE__, &(e), pgd_val(e))
-
+#define pte_ERROR(e)							\
+	printk("%s:%d: bad pte %p(%08lx%08lx).\n",			\
+	       __FILE__, __LINE__, &(e), (e).pte_high, (e).pte_low)
+#define pmd_ERROR(e)							\
+	printk("%s:%d: bad pmd %p(%016Lx).\n",				\
+	       __FILE__, __LINE__, &(e), pmd_val(e))
+#define pgd_ERROR(e)							\
+	printk("%s:%d: bad pgd %p(%016Lx).\n",				\
+	       __FILE__, __LINE__, &(e), pgd_val(e))
 
 static inline int pud_none(pud_t pud)
 {
 	return pud_val(pud) == 0;
 }
+
 static inline int pud_bad(pud_t pud)
 {
 	return (pud_val(pud) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER)) != 0;
 }
+
 static inline int pud_present(pud_t pud)
 {
 	return pud_val(pud) & _PAGE_PRESENT;
@@ -48,7 +52,8 @@ static inline void native_set_pte(pte_t *ptep, pte_t pte)
  * we are justified in merely clearing the PTE present bit, followed
  * by a set.  The ordering here is important.
  */
-static inline void native_set_pte_present(struct mm_struct *mm, unsigned long addr,
+static inline void native_set_pte_present(struct mm_struct *mm,
+					  unsigned long addr,
 					  pte_t *ptep, pte_t pte)
 {
 	ptep->pte_low = 0;
@@ -60,15 +65,17 @@ static inline void native_set_pte_present(struct mm_struct *mm, unsigned long ad
 
 static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
 {
-	set_64bit((unsigned long long *)(ptep),native_pte_val(pte));
+	set_64bit((unsigned long long *)(ptep), native_pte_val(pte));
 }
+
 static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
 {
-	set_64bit((unsigned long long *)(pmdp),native_pmd_val(pmd));
+	set_64bit((unsigned long long *)(pmdp), native_pmd_val(pmd));
 }
+
 static inline void native_set_pud(pud_t *pudp, pud_t pud)
 {
-	set_64bit((unsigned long long *)(pudp),native_pud_val(pud));
+	set_64bit((unsigned long long *)(pudp), native_pud_val(pud));
 }
 
 /*
@@ -76,7 +83,8 @@ static inline void native_set_pud(pud_t *pudp, pud_t pud)
  * entry, so clear the bottom half first and enforce ordering with a compiler
  * barrier.
  */
-static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr,
+				    pte_t *ptep)
 {
 	ptep->pte_low = 0;
 	smp_wmb();
@@ -107,20 +115,19 @@ static inline void pud_clear(pud_t *pudp)
 	 * current pgd to avoid unnecessary TLB flushes.
 	 */
 	pgd = read_cr3();
-	if (__pa(pudp) >= pgd && __pa(pudp) < (pgd + sizeof(pgd_t)*PTRS_PER_PGD))
+	if (__pa(pudp) >= pgd && __pa(pudp) <
+	    (pgd + sizeof(pgd_t)*PTRS_PER_PGD))
 		write_cr3(pgd);
 }
 
-#define pud_page(pud) \
-((struct page *) __va(pud_val(pud) & PAGE_MASK))
+#define pud_page(pud) ((struct page *) __va(pud_val(pud) & PAGE_MASK))
 
-#define pud_page_vaddr(pud) \
-((unsigned long) __va(pud_val(pud) & PAGE_MASK))
+#define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PAGE_MASK))
 
 
 /* Find an entry in the second-level page table.. */
-#define pmd_offset(pud, address) ((pmd_t *) pud_page(*(pud)) + \
-			pmd_index(address))
+#define pmd_offset(pud, address) ((pmd_t *)pud_page(*(pud)) +	\
+				  pmd_index(address))
 
 #ifdef CONFIG_SMP
 static inline pte_t native_ptep_get_and_clear(pte_t *ptep)
@@ -161,7 +168,8 @@ static inline unsigned long pte_pfn(pte_t pte)
  * put the 32 bits of offset into the high part.
  */
 #define pte_to_pgoff(pte) ((pte).pte_high)
-#define pgoff_to_pte(off) ((pte_t) { { .pte_low = _PAGE_FILE, .pte_high = (off) } })
+#define pgoff_to_pte(off)						\
+	((pte_t) { { .pte_low = _PAGE_FILE, .pte_high = (off) } })
 #define PTE_FILE_MAX_BITS       32
 
 /* Encode and de-code a swap entry */
-- 
cgit v1.2.3


From 7f94401e439dc1137319c48dfec0285f681eb3ad Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:11 -0700
Subject: include/asm-x86/pgtable_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pgtable_64.h | 141 +++++++++++++++++++++++--------------------
 1 file changed, 76 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 0a5081c98ae1..9fd87d0b6477 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -52,14 +52,18 @@ extern void paging_init(void);
 
 #ifndef __ASSEMBLY__
 
-#define pte_ERROR(e) \
-	printk("%s:%d: bad pte %p(%016lx).\n", __FILE__, __LINE__, &(e), pte_val(e))
-#define pmd_ERROR(e) \
-	printk("%s:%d: bad pmd %p(%016lx).\n", __FILE__, __LINE__, &(e), pmd_val(e))
-#define pud_ERROR(e) \
-	printk("%s:%d: bad pud %p(%016lx).\n", __FILE__, __LINE__, &(e), pud_val(e))
-#define pgd_ERROR(e) \
-	printk("%s:%d: bad pgd %p(%016lx).\n", __FILE__, __LINE__, &(e), pgd_val(e))
+#define pte_ERROR(e)					\
+	printk("%s:%d: bad pte %p(%016lx).\n",		\
+	       __FILE__, __LINE__, &(e), pte_val(e))
+#define pmd_ERROR(e)					\
+	printk("%s:%d: bad pmd %p(%016lx).\n",		\
+	       __FILE__, __LINE__, &(e), pmd_val(e))
+#define pud_ERROR(e)					\
+	printk("%s:%d: bad pud %p(%016lx).\n",		\
+	       __FILE__, __LINE__, &(e), pud_val(e))
+#define pgd_ERROR(e)					\
+	printk("%s:%d: bad pgd %p(%016lx).\n",		\
+	       __FILE__, __LINE__, &(e), pgd_val(e))
 
 #define pgd_none(x)	(!pgd_val(x))
 #define pud_none(x)	(!pud_val(x))
@@ -87,7 +91,8 @@ static inline pte_t native_ptep_get_and_clear(pte_t *xp)
 #ifdef CONFIG_SMP
 	return native_make_pte(xchg(&xp->pte, 0));
 #else
-	/* native_local_ptep_get_and_clear, but duplicated because of cyclic dependency */
+	/* native_local_ptep_get_and_clear,
+	   but duplicated because of cyclic dependency */
 	pte_t ret = *xp;
 	native_pte_clear(NULL, 0, xp);
 	return ret;
@@ -119,7 +124,7 @@ static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd)
 	*pgdp = pgd;
 }
 
-static inline void native_pgd_clear(pgd_t * pgd)
+static inline void native_pgd_clear(pgd_t *pgd)
 {
 	native_set_pgd(pgd, native_make_pgd(0));
 }
@@ -128,15 +133,15 @@ static inline void native_pgd_clear(pgd_t * pgd)
 
 #endif /* !__ASSEMBLY__ */
 
-#define PMD_SIZE	(_AC(1,UL) << PMD_SHIFT)
-#define PMD_MASK	(~(PMD_SIZE-1))
-#define PUD_SIZE	(_AC(1,UL) << PUD_SHIFT)
-#define PUD_MASK	(~(PUD_SIZE-1))
-#define PGDIR_SIZE	(_AC(1,UL) << PGDIR_SHIFT)
-#define PGDIR_MASK	(~(PGDIR_SIZE-1))
+#define PMD_SIZE	(_AC(1, UL) << PMD_SHIFT)
+#define PMD_MASK	(~(PMD_SIZE - 1))
+#define PUD_SIZE	(_AC(1, UL) << PUD_SHIFT)
+#define PUD_MASK	(~(PUD_SIZE - 1))
+#define PGDIR_SIZE	(_AC(1, UL) << PGDIR_SHIFT)
+#define PGDIR_MASK	(~(PGDIR_SIZE - 1))
 
 
-#define MAXMEM		 _AC(0x3fffffffffff, UL)
+#define MAXMEM		 _AC(0x00003fffffffffff, UL)
 #define VMALLOC_START    _AC(0xffffc20000000000, UL)
 #define VMALLOC_END      _AC(0xffffe1ffffffffff, UL)
 #define VMEMMAP_START	 _AC(0xffffe20000000000, UL)
@@ -163,18 +168,18 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 		~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER | _PAGE_PSE | _PAGE_NX);
 }
 
-#define pte_none(x)	(!pte_val(x))
-#define pte_present(x)	(pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
+#define pte_none(x)	(!pte_val((x)))
+#define pte_present(x)	(pte_val((x)) & (_PAGE_PRESENT | _PAGE_PROTNONE))
 
-#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))	/* FIXME: is this right? */
-#define pte_page(x)	pfn_to_page(pte_pfn(x))
-#define pte_pfn(x)  ((pte_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
+#define pages_to_mb(x)	((x) >> (20 - PAGE_SHIFT))   /* FIXME: is this right? */
+#define pte_page(x)	pfn_to_page(pte_pfn((x)))
+#define pte_pfn(x)	((pte_val((x)) & __PHYSICAL_MASK) >> PAGE_SHIFT)
 
 /*
  * Macro to mark a page protection value as "uncacheable".
  */
-#define pgprot_noncached(prot)	(__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
-
+#define pgprot_noncached(prot)					\
+	(__pgprot(pgprot_val((prot)) | _PAGE_PCD | _PAGE_PWT))
 
 /*
  * Conversion functions: convert a page and protection to a page entry,
@@ -184,77 +189,81 @@ static inline unsigned long pmd_bad(pmd_t pmd)
 /*
  * Level 4 access.
  */
-#define pgd_page_vaddr(pgd) ((unsigned long) __va((unsigned long)pgd_val(pgd) & PTE_MASK))
-#define pgd_page(pgd)		(pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT))
-#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
-#define pgd_offset(mm, addr) ((mm)->pgd + pgd_index(addr))
-#define pgd_offset_k(address) (init_level4_pgt + pgd_index(address))
+#define pgd_page_vaddr(pgd)						\
+	((unsigned long)__va((unsigned long)pgd_val((pgd)) & PTE_MASK))
+#define pgd_page(pgd)		(pfn_to_page(pgd_val((pgd)) >> PAGE_SHIFT))
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))
+#define pgd_offset(mm, address)	((mm)->pgd + pgd_index((address)))
+#define pgd_offset_k(address) (init_level4_pgt + pgd_index((address)))
 #define pgd_present(pgd) (pgd_val(pgd) & _PAGE_PRESENT)
 static inline int pgd_large(pgd_t pgd) { return 0; }
 #define mk_kernel_pgd(address) ((pgd_t){ (address) | _KERNPG_TABLE })
 
 /* PUD - Level3 access */
 /* to find an entry in a page-table-directory. */
-#define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
-#define pud_page(pud)		(pfn_to_page(pud_val(pud) >> PAGE_SHIFT))
-#define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1))
-#define pud_offset(pgd, address) ((pud_t *) pgd_page_vaddr(*(pgd)) + pud_index(address))
-#define pud_present(pud) (pud_val(pud) & _PAGE_PRESENT)
+#define pud_page_vaddr(pud)						\
+	((unsigned long)__va(pud_val((pud)) & PHYSICAL_PAGE_MASK))
+#define pud_page(pud)	(pfn_to_page(pud_val((pud)) >> PAGE_SHIFT))
+#define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD - 1))
+#define pud_offset(pgd, address)					\
+	((pud_t *)pgd_page_vaddr(*(pgd)) + pud_index((address)))
+#define pud_present(pud) (pud_val((pud)) & _PAGE_PRESENT)
 
 static inline int pud_large(pud_t pte)
 {
-	return (pud_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
-		(_PAGE_PSE|_PAGE_PRESENT);
+	return (pud_val(pte) & (_PAGE_PSE | _PAGE_PRESENT)) ==
+		(_PAGE_PSE | _PAGE_PRESENT);
 }
 
 /* PMD  - Level 2 access */
-#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
-#define pmd_page(pmd)		(pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
-
-#define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
-#define pmd_offset(dir, address) ((pmd_t *) pud_page_vaddr(*(dir)) + \
-			pmd_index(address))
-#define pmd_none(x)	(!pmd_val(x))
-#define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
-#define pfn_pmd(nr,prot) (__pmd(((nr) << PAGE_SHIFT) | pgprot_val(prot)))
-#define pmd_pfn(x)  ((pmd_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
-
-#define pte_to_pgoff(pte) ((pte_val(pte) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
-#define pgoff_to_pte(off) ((pte_t) { .pte = ((off) << PAGE_SHIFT) | _PAGE_FILE })
+#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val((pmd)) & PTE_MASK))
+#define pmd_page(pmd)		(pfn_to_page(pmd_val((pmd)) >> PAGE_SHIFT))
+
+#define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD - 1))
+#define pmd_offset(dir, address) ((pmd_t *)pud_page_vaddr(*(dir)) + \
+				  pmd_index(address))
+#define pmd_none(x)	(!pmd_val((x)))
+#define pmd_present(x)	(pmd_val((x)) & _PAGE_PRESENT)
+#define pfn_pmd(nr, prot) (__pmd(((nr) << PAGE_SHIFT) | pgprot_val((prot))))
+#define pmd_pfn(x)  ((pmd_val((x)) & __PHYSICAL_MASK) >> PAGE_SHIFT)
+
+#define pte_to_pgoff(pte) ((pte_val((pte)) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
+#define pgoff_to_pte(off) ((pte_t) { .pte = ((off) << PAGE_SHIFT) |	\
+					    _PAGE_FILE })
 #define PTE_FILE_MAX_BITS __PHYSICAL_MASK_SHIFT
 
 /* PTE - Level 1 access. */
 
 /* page, protection -> pte */
-#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
- 
-#define pte_index(address) \
-		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn((page)), (pgprot))
+
+#define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
-			pte_index(address))
+					 pte_index((address)))
 
 /* x86-64 always has all page tables mapped. */
-#define pte_offset_map(dir,address) pte_offset_kernel(dir,address)
-#define pte_offset_map_nested(dir,address) pte_offset_kernel(dir,address)
+#define pte_offset_map(dir, address) pte_offset_kernel((dir), (address))
+#define pte_offset_map_nested(dir, address) pte_offset_kernel((dir), (address))
 #define pte_unmap(pte) /* NOP */
-#define pte_unmap_nested(pte) /* NOP */ 
+#define pte_unmap_nested(pte) /* NOP */
 
-#define update_mmu_cache(vma,address,pte) do { } while (0)
+#define update_mmu_cache(vma, address, pte) do { } while (0)
 
 extern int direct_gbpages;
 
 /* Encode and de-code a swap entry */
 #define __swp_type(x)			(((x).val >> 1) & 0x3f)
 #define __swp_offset(x)			((x).val >> 8)
-#define __swp_entry(type, offset)	((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
-#define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val(pte) })
+#define __swp_entry(type, offset)	((swp_entry_t) { ((type) << 1) | \
+							 ((offset) << 8) })
+#define __pte_to_swp_entry(pte)		((swp_entry_t) { pte_val((pte)) })
 #define __swp_entry_to_pte(x)		((pte_t) { .pte = (x).val })
 
-extern int kern_addr_valid(unsigned long addr); 
+extern int kern_addr_valid(unsigned long addr);
 extern void cleanup_highmap(void);
 
-#define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
-		remap_pfn_range(vma, vaddr, pfn, size, prot)
+#define io_remap_pfn_range(vma, vaddr, pfn, size, prot)	\
+	remap_pfn_range(vma, vaddr, pfn, size, prot)
 
 #define HAVE_ARCH_UNMAPPED_AREA
 #define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN
@@ -267,8 +276,10 @@ extern void cleanup_highmap(void);
 
 /* fs/proc/kcore.c */
 #define	kc_vaddr_to_offset(v) ((v) & __VIRTUAL_MASK)
-#define	kc_offset_to_vaddr(o) \
-   (((o) & (1UL << (__VIRTUAL_MASK_SHIFT-1))) ? ((o) | (~__VIRTUAL_MASK)) : (o))
+#define	kc_offset_to_vaddr(o)				\
+	(((o) & (1UL << (__VIRTUAL_MASK_SHIFT - 1)))	\
+	 ? ((o) | ~__VIRTUAL_MASK)			\
+	 : (o))
 
 #define __HAVE_ARCH_PTE_SAME
 #endif /* !__ASSEMBLY__ */
-- 
cgit v1.2.3


From 3cbaeafeb10e38bce6c8d4764a254260d5a564bd Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:12 -0700
Subject: include/asm-x86/pgtable.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pgtable.h | 156 +++++++++++++++++++++++++++++++++++-----------
 1 file changed, 120 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index e814cfe96af2..2ce765070464 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -48,12 +48,15 @@
 #endif
 
 /* If _PAGE_PRESENT is clear, we use these: */
-#define _PAGE_FILE	_PAGE_DIRTY	/* nonlinear file mapping, saved PTE; unset:swap */
+#define _PAGE_FILE	_PAGE_DIRTY	/* nonlinear file mapping,
+					 * saved PTE; unset:swap */
 #define _PAGE_PROTNONE	_PAGE_PSE	/* if the user mapped it with PROT_NONE;
 					   pte_present gives true */
 
-#define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER |	\
+			 _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED |	\
+			 _PAGE_DIRTY)
 
 #define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
 
@@ -64,14 +67,20 @@
 #define _PAGE_CACHE_UC		(_PAGE_PCD | _PAGE_PWT)
 
 #define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
-#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-
-#define PAGE_SHARED_EXEC	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
-#define PAGE_COPY_NOEXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-#define PAGE_COPY_EXEC		__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | \
+				 _PAGE_ACCESSED | _PAGE_NX)
+
+#define PAGE_SHARED_EXEC	__pgprot(_PAGE_PRESENT | _PAGE_RW |	\
+					 _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_COPY_NOEXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER |	\
+					 _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_COPY_EXEC		__pgprot(_PAGE_PRESENT | _PAGE_USER |	\
+					 _PAGE_ACCESSED)
 #define PAGE_COPY		PAGE_COPY_NOEXEC
-#define PAGE_READONLY		__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
-#define PAGE_READONLY_EXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_READONLY		__pgprot(_PAGE_PRESENT | _PAGE_USER |	\
+					 _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_READONLY_EXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER |	\
+					 _PAGE_ACCESSED)
 
 #ifdef CONFIG_X86_32
 #define _PAGE_KERNEL_EXEC \
@@ -142,7 +151,7 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
  * ZERO_PAGE is a global shared page that is always zero: used
  * for zero-mapped memory areas etc..
  */
-extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
+extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
 
 extern spinlock_t pgd_lock;
@@ -152,30 +161,101 @@ extern struct list_head pgd_list;
  * The following only work if pte_present() is true.
  * Undefined behaviour if not..
  */
-static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
-static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
-static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
-static inline int pte_file(pte_t pte)		{ return pte_val(pte) & _PAGE_FILE; }
-static inline int pte_huge(pte_t pte)		{ return pte_val(pte) & _PAGE_PSE; }
-static inline int pte_global(pte_t pte) 	{ return pte_val(pte) & _PAGE_GLOBAL; }
-static inline int pte_exec(pte_t pte)		{ return !(pte_val(pte) & _PAGE_NX); }
-
-static inline int pmd_large(pmd_t pte) {
-	return (pmd_val(pte) & (_PAGE_PSE|_PAGE_PRESENT)) ==
-		(_PAGE_PSE|_PAGE_PRESENT);
+static inline int pte_dirty(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_DIRTY;
+}
+
+static inline int pte_young(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_ACCESSED;
+}
+
+static inline int pte_write(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_RW;
+}
+
+static inline int pte_file(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_FILE;
+}
+
+static inline int pte_huge(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_PSE;
 }
 
-static inline pte_t pte_mkclean(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_DIRTY); }
-static inline pte_t pte_mkold(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_ACCESSED); }
-static inline pte_t pte_wrprotect(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_RW); }
-static inline pte_t pte_mkexec(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_NX); }
-static inline pte_t pte_mkdirty(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_DIRTY); }
-static inline pte_t pte_mkyoung(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_ACCESSED); }
-static inline pte_t pte_mkwrite(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_RW); }
-static inline pte_t pte_mkhuge(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_PSE); }
-static inline pte_t pte_clrhuge(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_PSE); }
-static inline pte_t pte_mkglobal(pte_t pte)	{ return __pte(pte_val(pte) | _PAGE_GLOBAL); }
-static inline pte_t pte_clrglobal(pte_t pte)	{ return __pte(pte_val(pte) & ~(pteval_t)_PAGE_GLOBAL); }
+static inline int pte_global(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_GLOBAL;
+}
+
+static inline int pte_exec(pte_t pte)
+{
+	return !(pte_val(pte) & _PAGE_NX);
+}
+
+static inline int pmd_large(pmd_t pte)
+{
+	return (pmd_val(pte) & (_PAGE_PSE | _PAGE_PRESENT)) ==
+		(_PAGE_PSE | _PAGE_PRESENT);
+}
+
+static inline pte_t pte_mkclean(pte_t pte)
+{
+	return __pte(pte_val(pte) & ~(pteval_t)_PAGE_DIRTY);
+}
+
+static inline pte_t pte_mkold(pte_t pte)
+{
+	return __pte(pte_val(pte) & ~(pteval_t)_PAGE_ACCESSED);
+}
+
+static inline pte_t pte_wrprotect(pte_t pte)
+{
+	return __pte(pte_val(pte) & ~(pteval_t)_PAGE_RW);
+}
+
+static inline pte_t pte_mkexec(pte_t pte)
+{
+	return __pte(pte_val(pte) & ~(pteval_t)_PAGE_NX);
+}
+
+static inline pte_t pte_mkdirty(pte_t pte)
+{
+	return __pte(pte_val(pte) | _PAGE_DIRTY);
+}
+
+static inline pte_t pte_mkyoung(pte_t pte)
+{
+	return __pte(pte_val(pte) | _PAGE_ACCESSED);
+}
+
+static inline pte_t pte_mkwrite(pte_t pte)
+{
+	return __pte(pte_val(pte) | _PAGE_RW);
+}
+
+static inline pte_t pte_mkhuge(pte_t pte)
+{
+	return __pte(pte_val(pte) | _PAGE_PSE);
+}
+
+static inline pte_t pte_clrhuge(pte_t pte)
+{
+	return __pte(pte_val(pte) & ~(pteval_t)_PAGE_PSE);
+}
+
+static inline pte_t pte_mkglobal(pte_t pte)
+{
+	return __pte(pte_val(pte) | _PAGE_GLOBAL);
+}
+
+static inline pte_t pte_clrglobal(pte_t pte)
+{
+	return __pte(pte_val(pte) & ~(pteval_t)_PAGE_GLOBAL);
+}
 
 extern pteval_t __supported_pte_mask;
 
@@ -342,7 +422,8 @@ static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
 })
 
 #define __HAVE_ARCH_PTEP_GET_AND_CLEAR
-static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
+				       pte_t *ptep)
 {
 	pte_t pte = native_ptep_get_and_clear(ptep);
 	pte_update(mm, addr, ptep);
@@ -350,7 +431,9 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
 }
 
 #define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
-static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
+static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
+					    unsigned long addr, pte_t *ptep,
+					    int full)
 {
 	pte_t pte;
 	if (full) {
@@ -366,7 +449,8 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 }
 
 #define __HAVE_ARCH_PTEP_SET_WRPROTECT
-static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+static inline void ptep_set_wrprotect(struct mm_struct *mm,
+				      unsigned long addr, pte_t *ptep)
 {
 	clear_bit(_PAGE_BIT_RW, (unsigned long *)&ptep->pte);
 	pte_update(mm, addr, ptep);
-- 
cgit v1.2.3


From 2c5d516ca70641e26463e8d24344b515a2973c11 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:13 -0700
Subject: include/asm-x86/posix_types_32.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/posix_types_32.h | 47 +++++++++++++++++++++++-----------------
 1 file changed, 27 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/posix_types_32.h b/include/asm-x86/posix_types_32.h
index 015e539cdef5..b031efda37ec 100644
--- a/include/asm-x86/posix_types_32.h
+++ b/include/asm-x86/posix_types_32.h
@@ -45,32 +45,39 @@ typedef struct {
 #if defined(__KERNEL__)
 
 #undef	__FD_SET
-#define __FD_SET(fd,fdsetp) \
-		__asm__ __volatile__("btsl %1,%0": \
-			"+m" (*(__kernel_fd_set *) (fdsetp)):"r" ((int) (fd)))
+#define __FD_SET(fd,fdsetp)					\
+	asm volatile("btsl %1,%0":				\
+		     "+m" (*(__kernel_fd_set *)(fdsetp))	\
+		     : "r" ((int)(fd)))
 
 #undef	__FD_CLR
-#define __FD_CLR(fd,fdsetp) \
-		__asm__ __volatile__("btrl %1,%0": \
-			"+m" (*(__kernel_fd_set *) (fdsetp)):"r" ((int) (fd)))
+#define __FD_CLR(fd,fdsetp)					\
+	asm volatile("btrl %1,%0":				\
+		     "+m" (*(__kernel_fd_set *)(fdsetp))	\
+		     : "r" ((int) (fd)))
 
 #undef	__FD_ISSET
-#define __FD_ISSET(fd,fdsetp) (__extension__ ({ \
-		unsigned char __result; \
-		__asm__ __volatile__("btl %1,%2 ; setb %0" \
-			:"=q" (__result) :"r" ((int) (fd)), \
-			"m" (*(__kernel_fd_set *) (fdsetp))); \
-		__result; }))
+#define __FD_ISSET(fd,fdsetp)					\
+	(__extension__						\
+	 ({							\
+	 unsigned char __result;				\
+	 asm volatile("btl %1,%2 ; setb %0"			\
+		      : "=q" (__result)				\
+		      : "r" ((int)(fd)),			\
+			"m" (*(__kernel_fd_set *)(fdsetp)));	\
+	 __result;						\
+}))
 
 #undef	__FD_ZERO
-#define __FD_ZERO(fdsetp) \
-do { \
-	int __d0, __d1; \
-	__asm__ __volatile__("cld ; rep ; stosl" \
-			:"=m" (*(__kernel_fd_set *) (fdsetp)), \
-			  "=&c" (__d0), "=&D" (__d1) \
-			:"a" (0), "1" (__FDSET_LONGS), \
-			"2" ((__kernel_fd_set *) (fdsetp)) : "memory"); \
+#define __FD_ZERO(fdsetp)					\
+do {								\
+	int __d0, __d1;						\
+	asm volatile("cld ; rep ; stosl"			\
+		     : "=m" (*(__kernel_fd_set *)(fdsetp)),	\
+		       "=&c" (__d0), "=&D" (__d1)		\
+		     : "a" (0), "1" (__FDSET_LONGS),		\
+		       "2" ((__kernel_fd_set *)(fdsetp))	\
+		     : "memory");				\
 } while (0)
 
 #endif /* defined(__KERNEL__) */
-- 
cgit v1.2.3


From 4943aa4ec25ccc7161f4f4fcdd0018a4c1f6d4e8 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:14 -0700
Subject: include/asm-x86/posix_types_64.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/posix_types_64.h | 54 ++++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/posix_types_64.h b/include/asm-x86/posix_types_64.h
index 9926aa43775b..d6624c95854a 100644
--- a/include/asm-x86/posix_types_64.h
+++ b/include/asm-x86/posix_types_64.h
@@ -46,7 +46,7 @@ typedef unsigned long	__kernel_old_dev_t;
 #ifdef __KERNEL__
 
 #undef __FD_SET
-static __inline__ void __FD_SET(unsigned long fd, __kernel_fd_set *fdsetp)
+static inline void __FD_SET(unsigned long fd, __kernel_fd_set *fdsetp)
 {
 	unsigned long _tmp = fd / __NFDBITS;
 	unsigned long _rem = fd % __NFDBITS;
@@ -54,7 +54,7 @@ static __inline__ void __FD_SET(unsigned long fd, __kernel_fd_set *fdsetp)
 }
 
 #undef __FD_CLR
-static __inline__ void __FD_CLR(unsigned long fd, __kernel_fd_set *fdsetp)
+static inline void __FD_CLR(unsigned long fd, __kernel_fd_set *fdsetp)
 {
 	unsigned long _tmp = fd / __NFDBITS;
 	unsigned long _rem = fd % __NFDBITS;
@@ -62,7 +62,7 @@ static __inline__ void __FD_CLR(unsigned long fd, __kernel_fd_set *fdsetp)
 }
 
 #undef __FD_ISSET
-static __inline__ int __FD_ISSET(unsigned long fd, __const__ __kernel_fd_set *p)
+static inline int __FD_ISSET(unsigned long fd, __const__ __kernel_fd_set *p)
 {
 	unsigned long _tmp = fd / __NFDBITS;
 	unsigned long _rem = fd % __NFDBITS;
@@ -74,36 +74,36 @@ static __inline__ int __FD_ISSET(unsigned long fd, __const__ __kernel_fd_set *p)
  * for 256 and 1024-bit fd_sets respectively)
  */
 #undef __FD_ZERO
-static __inline__ void __FD_ZERO(__kernel_fd_set *p)
+static inline void __FD_ZERO(__kernel_fd_set *p)
 {
 	unsigned long *tmp = p->fds_bits;
 	int i;
 
 	if (__builtin_constant_p(__FDSET_LONGS)) {
 		switch (__FDSET_LONGS) {
-			case 32:
-			  tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
-			  tmp[ 4] = 0; tmp[ 5] = 0; tmp[ 6] = 0; tmp[ 7] = 0;
-			  tmp[ 8] = 0; tmp[ 9] = 0; tmp[10] = 0; tmp[11] = 0;
-			  tmp[12] = 0; tmp[13] = 0; tmp[14] = 0; tmp[15] = 0;
-			  tmp[16] = 0; tmp[17] = 0; tmp[18] = 0; tmp[19] = 0;
-			  tmp[20] = 0; tmp[21] = 0; tmp[22] = 0; tmp[23] = 0;
-			  tmp[24] = 0; tmp[25] = 0; tmp[26] = 0; tmp[27] = 0;
-			  tmp[28] = 0; tmp[29] = 0; tmp[30] = 0; tmp[31] = 0;
-			  return;
-			case 16:
-			  tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
-			  tmp[ 4] = 0; tmp[ 5] = 0; tmp[ 6] = 0; tmp[ 7] = 0;
-			  tmp[ 8] = 0; tmp[ 9] = 0; tmp[10] = 0; tmp[11] = 0;
-			  tmp[12] = 0; tmp[13] = 0; tmp[14] = 0; tmp[15] = 0;
-			  return;
-			case 8:
-			  tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
-			  tmp[ 4] = 0; tmp[ 5] = 0; tmp[ 6] = 0; tmp[ 7] = 0;
-			  return;
-			case 4:
-			  tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
-			  return;
+		case 32:
+			tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
+			tmp[ 4] = 0; tmp[ 5] = 0; tmp[ 6] = 0; tmp[ 7] = 0;
+			tmp[ 8] = 0; tmp[ 9] = 0; tmp[10] = 0; tmp[11] = 0;
+			tmp[12] = 0; tmp[13] = 0; tmp[14] = 0; tmp[15] = 0;
+			tmp[16] = 0; tmp[17] = 0; tmp[18] = 0; tmp[19] = 0;
+			tmp[20] = 0; tmp[21] = 0; tmp[22] = 0; tmp[23] = 0;
+			tmp[24] = 0; tmp[25] = 0; tmp[26] = 0; tmp[27] = 0;
+			tmp[28] = 0; tmp[29] = 0; tmp[30] = 0; tmp[31] = 0;
+			return;
+		case 16:
+			tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
+			tmp[ 4] = 0; tmp[ 5] = 0; tmp[ 6] = 0; tmp[ 7] = 0;
+			tmp[ 8] = 0; tmp[ 9] = 0; tmp[10] = 0; tmp[11] = 0;
+			tmp[12] = 0; tmp[13] = 0; tmp[14] = 0; tmp[15] = 0;
+			return;
+		case 8:
+			tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
+			tmp[ 4] = 0; tmp[ 5] = 0; tmp[ 6] = 0; tmp[ 7] = 0;
+			return;
+		case 4:
+			tmp[ 0] = 0; tmp[ 1] = 0; tmp[ 2] = 0; tmp[ 3] = 0;
+			return;
 		}
 	}
 	i = __FDSET_LONGS;
-- 
cgit v1.2.3


From cca2e6f87e3856953503aae2c0b8a1d5628796ef Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:15 -0700
Subject: include/asm-x86/processor.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor.h | 73 +++++++++++++++++++++++----------------------
 1 file changed, 38 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index b0dece41dbc4..6e26c7c717a2 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -175,12 +175,12 @@ static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 				unsigned int *ecx, unsigned int *edx)
 {
 	/* ecx is often an input as well as an output. */
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (*eax), "2" (*ecx));
+	asm("cpuid"
+	    : "=a" (*eax),
+	      "=b" (*ebx),
+	      "=c" (*ecx),
+	      "=d" (*edx)
+	    : "0" (*eax), "2" (*ecx));
 }
 
 static inline void load_cr3(pgd_t *pgdir)
@@ -427,17 +427,23 @@ static inline unsigned long native_get_debugreg(int regno)
 
 	switch (regno) {
 	case 0:
-		asm("mov %%db0, %0" :"=r" (val)); break;
+		asm("mov %%db0, %0" :"=r" (val));
+		break;
 	case 1:
-		asm("mov %%db1, %0" :"=r" (val)); break;
+		asm("mov %%db1, %0" :"=r" (val));
+		break;
 	case 2:
-		asm("mov %%db2, %0" :"=r" (val)); break;
+		asm("mov %%db2, %0" :"=r" (val));
+		break;
 	case 3:
-		asm("mov %%db3, %0" :"=r" (val)); break;
+		asm("mov %%db3, %0" :"=r" (val));
+		break;
 	case 6:
-		asm("mov %%db6, %0" :"=r" (val)); break;
+		asm("mov %%db6, %0" :"=r" (val));
+		break;
 	case 7:
-		asm("mov %%db7, %0" :"=r" (val)); break;
+		asm("mov %%db7, %0" :"=r" (val));
+		break;
 	default:
 		BUG();
 	}
@@ -478,14 +484,14 @@ static inline void native_set_iopl_mask(unsigned mask)
 #ifdef CONFIG_X86_32
 	unsigned int reg;
 
-	__asm__ __volatile__ ("pushfl;"
-			      "popl %0;"
-			      "andl %1, %0;"
-			      "orl %2, %0;"
-			      "pushl %0;"
-			      "popfl"
-				: "=&r" (reg)
-				: "i" (~X86_EFLAGS_IOPL), "r" (mask));
+	asm volatile ("pushfl;"
+		      "popl %0;"
+		      "andl %1, %0;"
+		      "orl %2, %0;"
+		      "pushl %0;"
+		      "popfl"
+		      : "=&r" (reg)
+		      : "i" (~X86_EFLAGS_IOPL), "r" (mask));
 #endif
 }
 
@@ -523,8 +529,8 @@ static inline void native_swapgs(void)
 #define set_debugreg(value, register)				\
 	native_set_debugreg(register, value)
 
-static inline void
-load_sp0(struct tss_struct *tss, struct thread_struct *thread)
+static inline void load_sp0(struct tss_struct *tss,
+			    struct thread_struct *thread)
 {
 	native_load_sp0(tss, thread);
 }
@@ -680,7 +686,7 @@ static inline unsigned int cpuid_edx(unsigned int op)
 /* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
 static inline void rep_nop(void)
 {
-	__asm__ __volatile__("rep; nop" ::: "memory");
+	asm volatile("rep; nop" ::: "memory");
 }
 
 static inline void cpu_relax(void)
@@ -694,32 +700,29 @@ static inline void sync_core(void)
 	int tmp;
 
 	asm volatile("cpuid" : "=a" (tmp) : "0" (1)
-					  : "ebx", "ecx", "edx", "memory");
+		     : "ebx", "ecx", "edx", "memory");
 }
 
-static inline void
-__monitor(const void *eax, unsigned long ecx, unsigned long edx)
+static inline void __monitor(const void *eax, unsigned long ecx,
+			     unsigned long edx)
 {
 	/* "monitor %eax, %ecx, %edx;" */
-	asm volatile(
-		".byte 0x0f, 0x01, 0xc8;"
-		:: "a" (eax), "c" (ecx), "d"(edx));
+	asm volatile(".byte 0x0f, 0x01, 0xc8;"
+		     :: "a" (eax), "c" (ecx), "d"(edx));
 }
 
 static inline void __mwait(unsigned long eax, unsigned long ecx)
 {
 	/* "mwait %eax, %ecx;" */
-	asm volatile(
-		".byte 0x0f, 0x01, 0xc9;"
-		:: "a" (eax), "c" (ecx));
+	asm volatile(".byte 0x0f, 0x01, 0xc9;"
+		     :: "a" (eax), "c" (ecx));
 }
 
 static inline void __sti_mwait(unsigned long eax, unsigned long ecx)
 {
 	/* "mwait %eax, %ecx;" */
-	asm volatile(
-		"sti; .byte 0x0f, 0x01, 0xc9;"
-		:: "a" (eax), "c" (ecx));
+	asm volatile("sti; .byte 0x0f, 0x01, 0xc9;"
+		     :: "a" (eax), "c" (ecx));
 }
 
 extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx);
-- 
cgit v1.2.3


From 708c5662975518eeea04de10cb11075d48636180 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:16 -0700
Subject: include/asm-x86/proto.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/proto.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 68563c0709ac..9da46af3b0e9 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -26,7 +26,7 @@ extern int reboot_force;
 
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);
 
-#define round_up(x,y) (((x) + (y) - 1) & ~((y)-1))
-#define round_down(x,y) ((x) & ~((y)-1))
+#define round_up(x, y) (((x) + (y) - 1) & ~((y) - 1))
+#define round_down(x, y) ((x) & ~((y) - 1))
 
 #endif
-- 
cgit v1.2.3


From 72f74fa25a468f40781e452f1d2528395090d55f Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:17 -0700
Subject: include/asm-x86/ptrace.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/ptrace.h | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index bc442461ac64..e779f2b26b32 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -140,7 +140,8 @@ extern unsigned long
 convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
 
 #ifdef CONFIG_X86_32
-extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
+extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs,
+			 int error_code);
 #else
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
 #endif
@@ -169,8 +170,8 @@ static inline int user_mode(struct pt_regs *regs)
 static inline int user_mode_vm(struct pt_regs *regs)
 {
 #ifdef CONFIG_X86_32
-	return ((regs->cs & SEGMENT_RPL_MASK) |
-		(regs->flags & VM_MASK)) >= USER_RPL;
+	return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & VM_MASK)) >=
+		USER_RPL;
 #else
 	return user_mode(regs);
 #endif
-- 
cgit v1.2.3


From 78db4c6be439cd446c94181c73f5e06a89a5aaf3 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:18 -0700
Subject: include/asm-x86/reboot.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/reboot.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/reboot.h b/include/asm-x86/reboot.h
index e9e3ffc22c07..6b5233b4f84b 100644
--- a/include/asm-x86/reboot.h
+++ b/include/asm-x86/reboot.h
@@ -3,8 +3,7 @@
 
 struct pt_regs;
 
-struct machine_ops
-{
+struct machine_ops {
 	void (*restart)(char *cmd);
 	void (*halt)(void);
 	void (*power_off)(void);
-- 
cgit v1.2.3


From c6fd5d49ec578e1078331b81ca09008fb361a8ba Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:19 -0700
Subject: include/asm-x86/resume-trace.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/resume-trace.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/resume-trace.h b/include/asm-x86/resume-trace.h
index 46f725b0bc82..2557514d7ef6 100644
--- a/include/asm-x86/resume-trace.h
+++ b/include/asm-x86/resume-trace.h
@@ -3,16 +3,17 @@
 
 #include <asm/asm.h>
 
-#define TRACE_RESUME(user) do {					\
+#define TRACE_RESUME(user)					\
+do {								\
 	if (pm_trace_enabled) {					\
 		void *tracedata;				\
 		asm volatile(_ASM_MOV_UL " $1f,%0\n"		\
-			".section .tracedata,\"a\"\n"		\
-			"1:\t.word %c1\n\t"			\
-			_ASM_PTR " %c2\n"			\
-			".previous"				\
-			:"=r" (tracedata)			\
-			: "i" (__LINE__), "i" (__FILE__));	\
+			     ".section .tracedata,\"a\"\n"	\
+			     "1:\t.word %c1\n\t"		\
+			     _ASM_PTR " %c2\n"			\
+			     ".previous"			\
+			     :"=r" (tracedata)			\
+			     : "i" (__LINE__), "i" (__FILE__));	\
 		generate_resume_trace(tracedata, user);		\
 	}							\
 } while (0)
-- 
cgit v1.2.3


From 0f4fc8c1dca86b519fed50be0962c8def8d3d446 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:20 -0700
Subject: include/asm-x86/rio.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/rio.h | 76 +++++++++++++++++++++++++--------------------------
 1 file changed, 38 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/rio.h b/include/asm-x86/rio.h
index 97cdcc9887ba..3451c576e6af 100644
--- a/include/asm-x86/rio.h
+++ b/include/asm-x86/rio.h
@@ -11,53 +11,53 @@
 #define RIO_TABLE_VERSION	3
 
 struct rio_table_hdr {
-	u8 version;      /* Version number of this data structure  */
-	u8 num_scal_dev; /* # of Scalability devices               */
-	u8 num_rio_dev;  /* # of RIO I/O devices                   */
+	u8 version;		/* Version number of this data structure  */
+	u8 num_scal_dev;	/* # of Scalability devices               */
+	u8 num_rio_dev;		/* # of RIO I/O devices                   */
 } __attribute__((packed));
 
 struct scal_detail {
-	u8 node_id;      /* Scalability Node ID                    */
-	u32 CBAR;        /* Address of 1MB register space          */
-	u8 port0node;    /* Node ID port connected to: 0xFF=None   */
-	u8 port0port;    /* Port num port connected to: 0,1,2, or  */
-	                 /* 0xFF=None                              */
-	u8 port1node;    /* Node ID port connected to: 0xFF = None */
-	u8 port1port;    /* Port num port connected to: 0,1,2, or  */
-	                 /* 0xFF=None                              */
-	u8 port2node;    /* Node ID port connected to: 0xFF = None */
-	u8 port2port;    /* Port num port connected to: 0,1,2, or  */
-	                 /* 0xFF=None                              */
-	u8 chassis_num;  /* 1 based Chassis number (1 = boot node) */
+	u8 node_id;		/* Scalability Node ID                    */
+	u32 CBAR;		/* Address of 1MB register space          */
+	u8 port0node;		/* Node ID port connected to: 0xFF=None   */
+	u8 port0port;		/* Port num port connected to: 0,1,2, or  */
+				/* 0xFF=None                              */
+	u8 port1node;		/* Node ID port connected to: 0xFF = None */
+	u8 port1port;		/* Port num port connected to: 0,1,2, or  */
+				/* 0xFF=None                              */
+	u8 port2node;		/* Node ID port connected to: 0xFF = None */
+	u8 port2port;		/* Port num port connected to: 0,1,2, or  */
+				/* 0xFF=None                              */
+	u8 chassis_num;		/* 1 based Chassis number (1 = boot node) */
 } __attribute__((packed));
 
 struct rio_detail {
-	u8 node_id;      /* RIO Node ID                            */
-	u32 BBAR;        /* Address of 1MB register space          */
-	u8 type;         /* Type of device                         */
-	u8 owner_id;     /* Node ID of Hurricane that owns this    */
-	                 /* node                                   */
-	u8 port0node;    /* Node ID port connected to: 0xFF=None   */
-	u8 port0port;    /* Port num port connected to: 0,1,2, or  */
-	                 /* 0xFF=None                              */
-	u8 port1node;    /* Node ID port connected to: 0xFF=None   */
-	u8 port1port;    /* Port num port connected to: 0,1,2, or  */
-	                 /* 0xFF=None                              */
-	u8 first_slot;   /* Lowest slot number below this Calgary  */
-	u8 status;       /* Bit 0 = 1 : the XAPIC is used          */
-	                 /*       = 0 : the XAPIC is not used, ie: */
-	                 /*            ints fwded to another XAPIC */
-	                 /*           Bits1:7 Reserved             */
-	u8 WP_index;     /* instance index - lower ones have       */
-	                 /*     lower slot numbers/PCI bus numbers */
-	u8 chassis_num;  /* 1 based Chassis number                 */
+	u8 node_id;		/* RIO Node ID                            */
+	u32 BBAR;		/* Address of 1MB register space          */
+	u8 type;		/* Type of device                         */
+	u8 owner_id;		/* Node ID of Hurricane that owns this    */
+				/* node                                   */
+	u8 port0node;		/* Node ID port connected to: 0xFF=None   */
+	u8 port0port;		/* Port num port connected to: 0,1,2, or  */
+				/* 0xFF=None                              */
+	u8 port1node;		/* Node ID port connected to: 0xFF=None   */
+	u8 port1port;		/* Port num port connected to: 0,1,2, or  */
+				/* 0xFF=None                              */
+	u8 first_slot;		/* Lowest slot number below this Calgary  */
+	u8 status;		/* Bit 0 = 1 : the XAPIC is used          */
+				/*       = 0 : the XAPIC is not used, ie: */
+				/*            ints fwded to another XAPIC */
+				/*           Bits1:7 Reserved             */
+	u8 WP_index;		/* instance index - lower ones have       */
+				/*     lower slot numbers/PCI bus numbers */
+	u8 chassis_num;		/* 1 based Chassis number                 */
 } __attribute__((packed));
 
 enum {
-	HURR_SCALABILTY	= 0,  /* Hurricane Scalability info */
-	HURR_RIOIB	= 2,  /* Hurricane RIOIB info       */
-	COMPAT_CALGARY	= 4,  /* Compatibility Calgary      */
-	ALT_CALGARY	= 5,  /* Second Planar Calgary      */
+	HURR_SCALABILTY	= 0,	/* Hurricane Scalability info */
+	HURR_RIOIB	= 2,	/* Hurricane RIOIB info       */
+	COMPAT_CALGARY	= 4,	/* Compatibility Calgary      */
+	ALT_CALGARY	= 5,	/* Second Planar Calgary      */
 };
 
 /*
-- 
cgit v1.2.3


From 6e5609a97acef44440f233ad435dd0ab563608f9 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:21 -0700
Subject: include/asm-x86/rwsem.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/rwsem.h | 169 ++++++++++++++++++++++++------------------------
 1 file changed, 86 insertions(+), 83 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/rwsem.h b/include/asm-x86/rwsem.h
index 520a379f4b80..750f2a3542b3 100644
--- a/include/asm-x86/rwsem.h
+++ b/include/asm-x86/rwsem.h
@@ -56,14 +56,16 @@ extern asmregparm struct rw_semaphore *
 /*
  * the semaphore definition
  */
-struct rw_semaphore {
-	signed long		count;
+
 #define RWSEM_UNLOCKED_VALUE		0x00000000
 #define RWSEM_ACTIVE_BIAS		0x00000001
 #define RWSEM_ACTIVE_MASK		0x0000ffff
 #define RWSEM_WAITING_BIAS		(-0x00010000)
 #define RWSEM_ACTIVE_READ_BIAS		RWSEM_ACTIVE_BIAS
 #define RWSEM_ACTIVE_WRITE_BIAS		(RWSEM_WAITING_BIAS + RWSEM_ACTIVE_BIAS)
+
+struct rw_semaphore {
+	signed long		count;
 	spinlock_t		wait_lock;
 	struct list_head	wait_list;
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
@@ -78,11 +80,13 @@ struct rw_semaphore {
 #endif
 
 
-#define __RWSEM_INITIALIZER(name) \
-{ RWSEM_UNLOCKED_VALUE, __SPIN_LOCK_UNLOCKED((name).wait_lock), \
-  LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) }
+#define __RWSEM_INITIALIZER(name)				\
+{								\
+	RWSEM_UNLOCKED_VALUE, __SPIN_LOCK_UNLOCKED((name).wait_lock), \
+	LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) \
+}
 
-#define DECLARE_RWSEM(name) \
+#define DECLARE_RWSEM(name)					\
 	struct rw_semaphore name = __RWSEM_INITIALIZER(name)
 
 extern void __init_rwsem(struct rw_semaphore *sem, const char *name,
@@ -100,16 +104,16 @@ do {								\
  */
 static inline void __down_read(struct rw_semaphore *sem)
 {
-	__asm__ __volatile__(
-		"# beginning down_read\n\t"
-LOCK_PREFIX	"  incl      (%%eax)\n\t" /* adds 0x00000001, returns the old value */
-		"  jns        1f\n"
-		"  call call_rwsem_down_read_failed\n"
-		"1:\n\t"
-		"# ending down_read\n\t"
-		: "+m" (sem->count)
-		: "a" (sem)
-		: "memory", "cc");
+	asm volatile("# beginning down_read\n\t"
+		     LOCK_PREFIX "  incl      (%%eax)\n\t"
+		     /* adds 0x00000001, returns the old value */
+		     "  jns        1f\n"
+		     "  call call_rwsem_down_read_failed\n"
+		     "1:\n\t"
+		     "# ending down_read\n\t"
+		     : "+m" (sem->count)
+		     : "a" (sem)
+		     : "memory", "cc");
 }
 
 /*
@@ -118,21 +122,20 @@ LOCK_PREFIX	"  incl      (%%eax)\n\t" /* adds 0x00000001, returns the old value
 static inline int __down_read_trylock(struct rw_semaphore *sem)
 {
 	__s32 result, tmp;
-	__asm__ __volatile__(
-		"# beginning __down_read_trylock\n\t"
-		"  movl      %0,%1\n\t"
-		"1:\n\t"
-		"  movl	     %1,%2\n\t"
-		"  addl      %3,%2\n\t"
-		"  jle	     2f\n\t"
-LOCK_PREFIX	"  cmpxchgl  %2,%0\n\t"
-		"  jnz	     1b\n\t"
-		"2:\n\t"
-		"# ending __down_read_trylock\n\t"
-		: "+m" (sem->count), "=&a" (result), "=&r" (tmp)
-		: "i" (RWSEM_ACTIVE_READ_BIAS)
-		: "memory", "cc");
-	return result>=0 ? 1 : 0;
+	asm volatile("# beginning __down_read_trylock\n\t"
+		     "  movl      %0,%1\n\t"
+		     "1:\n\t"
+		     "  movl	     %1,%2\n\t"
+		     "  addl      %3,%2\n\t"
+		     "  jle	     2f\n\t"
+		     LOCK_PREFIX "  cmpxchgl  %2,%0\n\t"
+		     "  jnz	     1b\n\t"
+		     "2:\n\t"
+		     "# ending __down_read_trylock\n\t"
+		     : "+m" (sem->count), "=&a" (result), "=&r" (tmp)
+		     : "i" (RWSEM_ACTIVE_READ_BIAS)
+		     : "memory", "cc");
+	return result >= 0 ? 1 : 0;
 }
 
 /*
@@ -143,17 +146,18 @@ static inline void __down_write_nested(struct rw_semaphore *sem, int subclass)
 	int tmp;
 
 	tmp = RWSEM_ACTIVE_WRITE_BIAS;
-	__asm__ __volatile__(
-		"# beginning down_write\n\t"
-LOCK_PREFIX	"  xadd      %%edx,(%%eax)\n\t" /* subtract 0x0000ffff, returns the old value */
-		"  testl     %%edx,%%edx\n\t" /* was the count 0 before? */
-		"  jz        1f\n"
-		"  call call_rwsem_down_write_failed\n"
-		"1:\n"
-		"# ending down_write"
-		: "+m" (sem->count), "=d" (tmp)
-		: "a" (sem), "1" (tmp)
-		: "memory", "cc");
+	asm volatile("# beginning down_write\n\t"
+		     LOCK_PREFIX "  xadd      %%edx,(%%eax)\n\t"
+		     /* subtract 0x0000ffff, returns the old value */
+		     "  testl     %%edx,%%edx\n\t"
+		     /* was the count 0 before? */
+		     "  jz        1f\n"
+		     "  call call_rwsem_down_write_failed\n"
+		     "1:\n"
+		     "# ending down_write"
+		     : "+m" (sem->count), "=d" (tmp)
+		     : "a" (sem), "1" (tmp)
+		     : "memory", "cc");
 }
 
 static inline void __down_write(struct rw_semaphore *sem)
@@ -167,7 +171,7 @@ static inline void __down_write(struct rw_semaphore *sem)
 static inline int __down_write_trylock(struct rw_semaphore *sem)
 {
 	signed long ret = cmpxchg(&sem->count,
-				  RWSEM_UNLOCKED_VALUE, 
+				  RWSEM_UNLOCKED_VALUE,
 				  RWSEM_ACTIVE_WRITE_BIAS);
 	if (ret == RWSEM_UNLOCKED_VALUE)
 		return 1;
@@ -180,16 +184,16 @@ static inline int __down_write_trylock(struct rw_semaphore *sem)
 static inline void __up_read(struct rw_semaphore *sem)
 {
 	__s32 tmp = -RWSEM_ACTIVE_READ_BIAS;
-	__asm__ __volatile__(
-		"# beginning __up_read\n\t"
-LOCK_PREFIX	"  xadd      %%edx,(%%eax)\n\t" /* subtracts 1, returns the old value */
-		"  jns        1f\n\t"
-		"  call call_rwsem_wake\n"
-		"1:\n"
-		"# ending __up_read\n"
-		: "+m" (sem->count), "=d" (tmp)
-		: "a" (sem), "1" (tmp)
-		: "memory", "cc");
+	asm volatile("# beginning __up_read\n\t"
+		     LOCK_PREFIX "  xadd      %%edx,(%%eax)\n\t"
+		     /* subtracts 1, returns the old value */
+		     "  jns        1f\n\t"
+		     "  call call_rwsem_wake\n"
+		     "1:\n"
+		     "# ending __up_read\n"
+		     : "+m" (sem->count), "=d" (tmp)
+		     : "a" (sem), "1" (tmp)
+		     : "memory", "cc");
 }
 
 /*
@@ -197,17 +201,18 @@ LOCK_PREFIX	"  xadd      %%edx,(%%eax)\n\t" /* subtracts 1, returns the old valu
  */
 static inline void __up_write(struct rw_semaphore *sem)
 {
-	__asm__ __volatile__(
-		"# beginning __up_write\n\t"
-		"  movl      %2,%%edx\n\t"
-LOCK_PREFIX	"  xaddl     %%edx,(%%eax)\n\t" /* tries to transition 0xffff0001 -> 0x00000000 */
-		"  jz       1f\n"
-		"  call call_rwsem_wake\n"
-		"1:\n\t"
-		"# ending __up_write\n"
-		: "+m" (sem->count)
-		: "a" (sem), "i" (-RWSEM_ACTIVE_WRITE_BIAS)
-		: "memory", "cc", "edx");
+	asm volatile("# beginning __up_write\n\t"
+		     "  movl      %2,%%edx\n\t"
+		     LOCK_PREFIX "  xaddl     %%edx,(%%eax)\n\t"
+		     /* tries to transition
+			0xffff0001 -> 0x00000000 */
+		     "  jz       1f\n"
+		     "  call call_rwsem_wake\n"
+		     "1:\n\t"
+		     "# ending __up_write\n"
+		     : "+m" (sem->count)
+		     : "a" (sem), "i" (-RWSEM_ACTIVE_WRITE_BIAS)
+		     : "memory", "cc", "edx");
 }
 
 /*
@@ -215,16 +220,16 @@ LOCK_PREFIX	"  xaddl     %%edx,(%%eax)\n\t" /* tries to transition 0xffff0001 ->
  */
 static inline void __downgrade_write(struct rw_semaphore *sem)
 {
-	__asm__ __volatile__(
-		"# beginning __downgrade_write\n\t"
-LOCK_PREFIX	"  addl      %2,(%%eax)\n\t" /* transitions 0xZZZZ0001 -> 0xYYYY0001 */
-		"  jns       1f\n\t"
-		"  call call_rwsem_downgrade_wake\n"
-		"1:\n\t"
-		"# ending __downgrade_write\n"
-		: "+m" (sem->count)
-		: "a" (sem), "i" (-RWSEM_WAITING_BIAS)
-		: "memory", "cc");
+	asm volatile("# beginning __downgrade_write\n\t"
+		     LOCK_PREFIX "  addl      %2,(%%eax)\n\t"
+		     /* transitions 0xZZZZ0001 -> 0xYYYY0001 */
+		     "  jns       1f\n\t"
+		     "  call call_rwsem_downgrade_wake\n"
+		     "1:\n\t"
+		     "# ending __downgrade_write\n"
+		     : "+m" (sem->count)
+		     : "a" (sem), "i" (-RWSEM_WAITING_BIAS)
+		     : "memory", "cc");
 }
 
 /*
@@ -232,10 +237,9 @@ LOCK_PREFIX	"  addl      %2,(%%eax)\n\t" /* transitions 0xZZZZ0001 -> 0xYYYY0001
  */
 static inline void rwsem_atomic_add(int delta, struct rw_semaphore *sem)
 {
-	__asm__ __volatile__(
-LOCK_PREFIX	"addl %1,%0"
-		: "+m" (sem->count)
-		: "ir" (delta));
+	asm volatile(LOCK_PREFIX "addl %1,%0"
+		     : "+m" (sem->count)
+		     : "ir" (delta));
 }
 
 /*
@@ -245,12 +249,11 @@ static inline int rwsem_atomic_update(int delta, struct rw_semaphore *sem)
 {
 	int tmp = delta;
 
-	__asm__ __volatile__(
-LOCK_PREFIX	"xadd %0,%1"
-		: "+r" (tmp), "+m" (sem->count)
-		: : "memory");
+	asm volatile(LOCK_PREFIX "xadd %0,%1"
+		     : "+r" (tmp), "+m" (sem->count)
+		     : : "memory");
 
-	return tmp+delta;
+	return tmp + delta;
 }
 
 static inline int rwsem_is_locked(struct rw_semaphore *sem)
-- 
cgit v1.2.3


From 915cd5aa0adbe0f62b7a56d6eaf6908b47f80395 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:25 -0700
Subject: include/asm-x86/setup.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/setup.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h
index f745de211191..fa6763af8d26 100644
--- a/include/asm-x86/setup.h
+++ b/include/asm-x86/setup.h
@@ -55,8 +55,8 @@ struct e820entry;
 char * __init machine_specific_memory_setup(void);
 char *memory_setup(void);
 
-int __init copy_e820_map(struct e820entry * biosmap, int nr_map);
-int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map);
+int __init copy_e820_map(struct e820entry *biosmap, int nr_map);
+int __init sanitize_e820_map(struct e820entry *biosmap, char *pnr_map);
 void __init add_memory_region(unsigned long long start,
 			      unsigned long long size, int type);
 
-- 
cgit v1.2.3


From 895b7643d6d2d04761344e18a2473255f2551e9e Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:26 -0700
Subject: include/asm-x86/sigcontext32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/sigcontext32.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/sigcontext32.h b/include/asm-x86/sigcontext32.h
index 6ffab4fd593a..57a9686fb491 100644
--- a/include/asm-x86/sigcontext32.h
+++ b/include/asm-x86/sigcontext32.h
@@ -26,7 +26,7 @@ struct _fpstate_ia32 {
 	__u32 	cw;
 	__u32	sw;
 	__u32	tag;	/* not compatible to 64bit twd */
-	__u32	ipoff;			
+	__u32	ipoff;
 	__u32	cssel;
 	__u32	dataoff;
 	__u32	datasel;
@@ -39,7 +39,7 @@ struct _fpstate_ia32 {
 	__u32	mxcsr;
 	__u32	reserved;
 	struct _fpxreg	_fxsr_st[8];
-	struct _xmmreg	_xmm[8];	/* It's actually 16 */ 
+	struct _xmmreg	_xmm[8];	/* It's actually 16 */
 	__u32	padding[56];
 };
 
-- 
cgit v1.2.3


From af1fec15de17086864fc3917e21a31e303ec0e91 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:27 -0700
Subject: include/asm-x86/sigcontext.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/sigcontext.h | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/sigcontext.h b/include/asm-x86/sigcontext.h
index d743947f4c77..2f9c884d2c0f 100644
--- a/include/asm-x86/sigcontext.h
+++ b/include/asm-x86/sigcontext.h
@@ -79,7 +79,7 @@ struct sigcontext {
 	unsigned long flags;
 	unsigned long sp_at_signal;
 	unsigned short ss, __ssh;
-	struct _fpstate __user * fpstate;
+	struct _fpstate __user *fpstate;
 	unsigned long oldmask;
 	unsigned long cr2;
 };
@@ -107,7 +107,7 @@ struct sigcontext {
 	unsigned long eflags;
 	unsigned long esp_at_signal;
 	unsigned short ss, __ssh;
-	struct _fpstate __user * fpstate;
+	struct _fpstate __user *fpstate;
 	unsigned long oldmask;
 	unsigned long cr2;
 };
@@ -121,7 +121,8 @@ struct sigcontext {
 struct _fpstate {
 	__u16	cwd;
 	__u16	swd;
-	__u16	twd;	/* Note this is not the same as the 32bit/x87/FSAVE twd */
+	__u16	twd;		/* Note this is not the same as the
+				   32bit/x87/FSAVE twd */
 	__u16	fop;
 	__u64	rip;
 	__u64	rdp;
-- 
cgit v1.2.3


From 9551b12a51bab7058ad486ba96fd0c27fdafe922 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:28 -0700
Subject: include/asm-x86/signal.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/signal.h | 48 ++++++++++++++++++++++++------------------------
 1 file changed, 24 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/signal.h b/include/asm-x86/signal.h
index aee7eca585ab..f15186d39c69 100644
--- a/include/asm-x86/signal.h
+++ b/include/asm-x86/signal.h
@@ -185,61 +185,61 @@ typedef struct sigaltstack {
 
 #define __HAVE_ARCH_SIG_BITOPS
 
-#define sigaddset(set,sig)		   \
-	(__builtin_constantp(sig) ?	   \
-	 __const_sigaddset((set),(sig)) :  \
-	 __gen_sigaddset((set),(sig)))
+#define sigaddset(set,sig)		    \
+	(__builtin_constantp(sig)	    \
+	 ? __const_sigaddset((set), (sig))  \
+	 : __gen_sigaddset((set), (sig)))
 
-static __inline__ void __gen_sigaddset(sigset_t *set, int _sig)
+static inline void __gen_sigaddset(sigset_t *set, int _sig)
 {
-	__asm__("btsl %1,%0" : "+m"(*set) : "Ir"(_sig - 1) : "cc");
+	asm("btsl %1,%0" : "+m"(*set) : "Ir"(_sig - 1) : "cc");
 }
 
-static __inline__ void __const_sigaddset(sigset_t *set, int _sig)
+static inline void __const_sigaddset(sigset_t *set, int _sig)
 {
 	unsigned long sig = _sig - 1;
 	set->sig[sig / _NSIG_BPW] |= 1 << (sig % _NSIG_BPW);
 }
 
-#define sigdelset(set,sig)		   \
-	(__builtin_constant_p(sig) ?       \
-	 __const_sigdelset((set),(sig)) :  \
-	 __gen_sigdelset((set),(sig)))
+#define sigdelset(set, sig)		    \
+	(__builtin_constant_p(sig)	    \
+	 ? __const_sigdelset((set), (sig))  \
+	 : __gen_sigdelset((set), (sig)))
 
 
-static __inline__ void __gen_sigdelset(sigset_t *set, int _sig)
+static inline void __gen_sigdelset(sigset_t *set, int _sig)
 {
-	__asm__("btrl %1,%0" : "+m"(*set) : "Ir"(_sig - 1) : "cc");
+	asm("btrl %1,%0" : "+m"(*set) : "Ir"(_sig - 1) : "cc");
 }
 
-static __inline__ void __const_sigdelset(sigset_t *set, int _sig)
+static inline void __const_sigdelset(sigset_t *set, int _sig)
 {
 	unsigned long sig = _sig - 1;
 	set->sig[sig / _NSIG_BPW] &= ~(1 << (sig % _NSIG_BPW));
 }
 
-static __inline__ int __const_sigismember(sigset_t *set, int _sig)
+static inline int __const_sigismember(sigset_t *set, int _sig)
 {
 	unsigned long sig = _sig - 1;
 	return 1 & (set->sig[sig / _NSIG_BPW] >> (sig % _NSIG_BPW));
 }
 
-static __inline__ int __gen_sigismember(sigset_t *set, int _sig)
+static inline int __gen_sigismember(sigset_t *set, int _sig)
 {
 	int ret;
-	__asm__("btl %2,%1\n\tsbbl %0,%0"
-		: "=r"(ret) : "m"(*set), "Ir"(_sig-1) : "cc");
+	asm("btl %2,%1\n\tsbbl %0,%0"
+	    : "=r"(ret) : "m"(*set), "Ir"(_sig-1) : "cc");
 	return ret;
 }
 
-#define sigismember(set,sig)			\
-	(__builtin_constant_p(sig) ?		\
-	 __const_sigismember((set),(sig)) :	\
-	 __gen_sigismember((set),(sig)))
+#define sigismember(set, sig)			\
+	(__builtin_constant_p(sig)		\
+	 ? __const_sigismember((set), (sig))	\
+	 : __gen_sigismember((set), (sig)))
 
-static __inline__ int sigfindinword(unsigned long word)
+static inline int sigfindinword(unsigned long word)
 {
-	__asm__("bsfl %1,%0" : "=r"(word) : "rm"(word) : "cc");
+	asm("bsfl %1,%0" : "=r"(word) : "rm"(word) : "cc");
 	return word;
 }
 
-- 
cgit v1.2.3


From 2fec394adf9445ae6a21c5e67d9ad3f6f5b7c8d1 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:29 -0700
Subject: include/asm-x86/smp_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp_32.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index f861d0415171..cb3ada2fedbf 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -18,8 +18,8 @@
 
 extern cpumask_t cpu_callin_map;
 
-extern void (*mtrr_hook) (void);
-extern void zap_low_mappings (void);
+extern void (*mtrr_hook)(void);
+extern void zap_low_mappings(void);
 
 #ifdef CONFIG_SMP
 /*
@@ -44,7 +44,7 @@ static inline int num_booting_cpus(void)
 
 #ifdef CONFIG_X86_LOCAL_APIC
 
-static __inline int logical_smp_processor_id(void)
+static inline int logical_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
-- 
cgit v1.2.3


From ceb7ce1052a9087bd4752424f253b883ec5e1cec Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:30 -0700
Subject: include/asm-x86/smp_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp_64.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index fd709cbba4d1..c53a011bb91c 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -24,9 +24,9 @@ extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 #define raw_smp_processor_id()	read_pda(cpunumber)
 
 #define stack_smp_processor_id()					\
-	({								\
+({									\
 	struct thread_info *ti;						\
-	__asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
+	asm("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
 	ti->cpu;							\
 })
 
@@ -46,7 +46,7 @@ static inline int num_booting_cpus(void)
 
 #define safe_smp_processor_id()		smp_processor_id()
 
-static __inline int logical_smp_processor_id(void)
+static inline int logical_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
-- 
cgit v1.2.3


From d3bf60a6e48c9a451cac345c0ad57552bb299992 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:31 -0700
Subject: include/asm-x86/spinlock.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/spinlock.h | 105 +++++++++++++++++++++------------------------
 1 file changed, 50 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index 23804c1890ff..47dfe2607bb1 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -82,7 +82,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
 	short inc = 0x0100;
 
-	__asm__ __volatile__ (
+	asm volatile (
 		LOCK_PREFIX "xaddw %w0, %1\n"
 		"1:\t"
 		"cmpb %h0, %b0\n\t"
@@ -92,9 +92,9 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 		/* don't need lfence here, because loads are in-order */
 		"jmp 1b\n"
 		"2:"
-		:"+Q" (inc), "+m" (lock->slock)
+		: "+Q" (inc), "+m" (lock->slock)
 		:
-		:"memory", "cc");
+		: "memory", "cc");
 }
 
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
@@ -104,30 +104,28 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 	int tmp;
 	short new;
 
-	asm volatile(
-		"movw %2,%w0\n\t"
-		"cmpb %h0,%b0\n\t"
-		"jne 1f\n\t"
-		"movw %w0,%w1\n\t"
-		"incb %h1\n\t"
-		"lock ; cmpxchgw %w1,%2\n\t"
-		"1:"
-		"sete %b1\n\t"
-		"movzbl %b1,%0\n\t"
-		:"=&a" (tmp), "=Q" (new), "+m" (lock->slock)
-		:
-		: "memory", "cc");
+	asm volatile("movw %2,%w0\n\t"
+		     "cmpb %h0,%b0\n\t"
+		     "jne 1f\n\t"
+		     "movw %w0,%w1\n\t"
+		     "incb %h1\n\t"
+		     "lock ; cmpxchgw %w1,%2\n\t"
+		     "1:"
+		     "sete %b1\n\t"
+		     "movzbl %b1,%0\n\t"
+		     : "=&a" (tmp), "=Q" (new), "+m" (lock->slock)
+		     :
+		     : "memory", "cc");
 
 	return tmp;
 }
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__(
-		UNLOCK_LOCK_PREFIX "incb %0"
-		:"+m" (lock->slock)
-		:
-		:"memory", "cc");
+	asm volatile(UNLOCK_LOCK_PREFIX "incb %0"
+		     : "+m" (lock->slock)
+		     :
+		     : "memory", "cc");
 }
 #else
 static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
@@ -149,21 +147,20 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 	int inc = 0x00010000;
 	int tmp;
 
-	__asm__ __volatile__ (
-		"lock ; xaddl %0, %1\n"
-		"movzwl %w0, %2\n\t"
-		"shrl $16, %0\n\t"
-		"1:\t"
-		"cmpl %0, %2\n\t"
-		"je 2f\n\t"
-		"rep ; nop\n\t"
-		"movzwl %1, %2\n\t"
-		/* don't need lfence here, because loads are in-order */
-		"jmp 1b\n"
-		"2:"
-		:"+Q" (inc), "+m" (lock->slock), "=r" (tmp)
-		:
-		:"memory", "cc");
+	asm volatile("lock ; xaddl %0, %1\n"
+		     "movzwl %w0, %2\n\t"
+		     "shrl $16, %0\n\t"
+		     "1:\t"
+		     "cmpl %0, %2\n\t"
+		     "je 2f\n\t"
+		     "rep ; nop\n\t"
+		     "movzwl %1, %2\n\t"
+		     /* don't need lfence here, because loads are in-order */
+		     "jmp 1b\n"
+		     "2:"
+		     : "+Q" (inc), "+m" (lock->slock), "=r" (tmp)
+		     :
+		     : "memory", "cc");
 }
 
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
@@ -173,31 +170,29 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 	int tmp;
 	int new;
 
-	asm volatile(
-		"movl %2,%0\n\t"
-		"movl %0,%1\n\t"
-		"roll $16, %0\n\t"
-		"cmpl %0,%1\n\t"
-		"jne 1f\n\t"
-		"addl $0x00010000, %1\n\t"
-		"lock ; cmpxchgl %1,%2\n\t"
-		"1:"
-		"sete %b1\n\t"
-		"movzbl %b1,%0\n\t"
-		:"=&a" (tmp), "=r" (new), "+m" (lock->slock)
-		:
-		: "memory", "cc");
+	asm volatile("movl %2,%0\n\t"
+		     "movl %0,%1\n\t"
+		     "roll $16, %0\n\t"
+		     "cmpl %0,%1\n\t"
+		     "jne 1f\n\t"
+		     "addl $0x00010000, %1\n\t"
+		     "lock ; cmpxchgl %1,%2\n\t"
+		     "1:"
+		     "sete %b1\n\t"
+		     "movzbl %b1,%0\n\t"
+		     : "=&a" (tmp), "=r" (new), "+m" (lock->slock)
+		     :
+		     : "memory", "cc");
 
 	return tmp;
 }
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__(
-		UNLOCK_LOCK_PREFIX "incw %0"
-		:"+m" (lock->slock)
-		:
-		:"memory", "cc");
+	asm volatile(UNLOCK_LOCK_PREFIX "incw %0"
+		     : "+m" (lock->slock)
+		     :
+		     : "memory", "cc");
 }
 #endif
 
-- 
cgit v1.2.3


From 7f3a9508b5778034091e00e696dc17ca98b8a84c Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:32 -0700
Subject: include/asm-x86/srat.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/srat.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/srat.h b/include/asm-x86/srat.h
index 165ab4bdc02b..f4bba131d068 100644
--- a/include/asm-x86/srat.h
+++ b/include/asm-x86/srat.h
@@ -1,5 +1,5 @@
 /*
- * Some of the code in this file has been gleaned from the 64 bit 
+ * Some of the code in this file has been gleaned from the 64 bit
  * discontigmem support code base.
  *
  * Copyright (C) 2002, IBM Corp.
-- 
cgit v1.2.3


From 06b0f574eafdd26d325ce5ab3c38522d0cdd7b7c Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:33 -0700
Subject: include/asm-x86/string_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/string_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/string_32.h b/include/asm-x86/string_32.h
index c5d13a86dea7..b49369ad9a61 100644
--- a/include/asm-x86/string_32.h
+++ b/include/asm-x86/string_32.h
@@ -3,7 +3,7 @@
 
 #ifdef __KERNEL__
 
-/* Let gcc decide wether to inline or use the out of line functions */
+/* Let gcc decide whether to inline or use the out of line functions */
 
 #define __HAVE_ARCH_STRCPY
 extern char *strcpy(char *dest, const char *src);
-- 
cgit v1.2.3


From 953b2f1ed6f2642e58e7c5b8c1385d132f1b5685 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:34 -0700
Subject: include/asm-x86/string_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/string_64.h | 66 ++++++++++++++++++++++-----------------------
 1 file changed, 33 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/string_64.h b/include/asm-x86/string_64.h
index e583da7918fb..52b5ab383395 100644
--- a/include/asm-x86/string_64.h
+++ b/include/asm-x86/string_64.h
@@ -3,26 +3,24 @@
 
 #ifdef __KERNEL__
 
-/* Written 2002 by Andi Kleen */ 
+/* Written 2002 by Andi Kleen */
 
-/* Only used for special circumstances. Stolen from i386/string.h */ 
-static __always_inline void *
-__inline_memcpy(void * to, const void * from, size_t n)
+/* Only used for special circumstances. Stolen from i386/string.h */
+static __always_inline void *__inline_memcpy(void *to, const void *from, size_t n)
 {
-unsigned long d0, d1, d2;
-__asm__ __volatile__(
-	"rep ; movsl\n\t"
-	"testb $2,%b4\n\t"
-	"je 1f\n\t"
-	"movsw\n"
-	"1:\ttestb $1,%b4\n\t"
-	"je 2f\n\t"
-	"movsb\n"
-	"2:"
-	: "=&c" (d0), "=&D" (d1), "=&S" (d2)
-	:"0" (n/4), "q" (n),"1" ((long) to),"2" ((long) from)
-	: "memory");
-return (to);
+	unsigned long d0, d1, d2;
+	asm volatile("rep ; movsl\n\t"
+		     "testb $2,%b4\n\t"
+		     "je 1f\n\t"
+		     "movsw\n"
+		     "1:\ttestb $1,%b4\n\t"
+		     "je 2f\n\t"
+		     "movsb\n"
+		     "2:"
+		     : "=&c" (d0), "=&D" (d1), "=&S" (d2)
+		     : "0" (n / 4), "q" (n), "1" ((long)to), "2" ((long)from)
+		     : "memory");
+	return to;
 }
 
 /* Even with __builtin_ the compiler may decide to use the out of line
@@ -32,28 +30,30 @@ return (to);
 #if (__GNUC__ == 4 && __GNUC_MINOR__ >= 3) || __GNUC__ > 4
 extern void *memcpy(void *to, const void *from, size_t len);
 #else
-extern void *__memcpy(void *to, const void *from, size_t len); 
-#define memcpy(dst,src,len) \
-	({ size_t __len = (len);				\
-	   void *__ret;						\
-	   if (__builtin_constant_p(len) && __len >= 64)	\
-		 __ret = __memcpy((dst),(src),__len);		\
-	   else							\
-		 __ret = __builtin_memcpy((dst),(src),__len);	\
-	   __ret; }) 
+extern void *__memcpy(void *to, const void *from, size_t len);
+#define memcpy(dst, src, len)					\
+({								\
+	size_t __len = (len);					\
+	void *__ret;						\
+	if (__builtin_constant_p(len) && __len >= 64)		\
+		__ret = __memcpy((dst), (src), __len);		\
+	else							\
+		__ret = __builtin_memcpy((dst), (src), __len);	\
+	__ret;							\
+})
 #endif
 
 #define __HAVE_ARCH_MEMSET
 void *memset(void *s, int c, size_t n);
 
 #define __HAVE_ARCH_MEMMOVE
-void * memmove(void * dest,const void *src,size_t count);
+void *memmove(void *dest, const void *src, size_t count);
 
-int memcmp(const void * cs,const void * ct,size_t count);
-size_t strlen(const char * s);
-char *strcpy(char * dest,const char *src);
-char *strcat(char * dest, const char * src);
-int strcmp(const char * cs,const char * ct);
+int memcmp(const void *cs, const void *ct, size_t count);
+size_t strlen(const char *s);
+char *strcpy(char *dest, const char *src);
+char *strcat(char *dest, const char *src);
+int strcmp(const char *cs, const char *ct);
 
 #endif /* __KERNEL__ */
 
-- 
cgit v1.2.3


From cf030ebd40e37cb11d1efa6677890f40f21e16f4 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:35 -0700
Subject: include/asm-x86/suspend_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/suspend_32.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/suspend_32.h b/include/asm-x86/suspend_32.h
index 1bbda3ad7796..24e1c080aa8a 100644
--- a/include/asm-x86/suspend_32.h
+++ b/include/asm-x86/suspend_32.h
@@ -10,7 +10,7 @@ static inline int arch_prepare_suspend(void) { return 0; }
 
 /* image of the saved processor state */
 struct saved_context {
-  	u16 es, fs, gs, ss;
+	u16 es, fs, gs, ss;
 	unsigned long cr0, cr2, cr3, cr4;
 	struct desc_ptr gdt;
 	struct desc_ptr idt;
@@ -32,11 +32,11 @@ extern unsigned long saved_edi;
 static inline void acpi_save_register_state(unsigned long return_point)
 {
 	saved_eip = return_point;
-	asm volatile ("movl %%esp,%0" : "=m" (saved_esp));
-	asm volatile ("movl %%ebp,%0" : "=m" (saved_ebp));
-	asm volatile ("movl %%ebx,%0" : "=m" (saved_ebx));
-	asm volatile ("movl %%edi,%0" : "=m" (saved_edi));
-	asm volatile ("movl %%esi,%0" : "=m" (saved_esi));
+	asm volatile("movl %%esp,%0" : "=m" (saved_esp));
+	asm volatile("movl %%ebp,%0" : "=m" (saved_ebp));
+	asm volatile("movl %%ebx,%0" : "=m" (saved_ebx));
+	asm volatile("movl %%edi,%0" : "=m" (saved_edi));
+	asm volatile("movl %%esi,%0" : "=m" (saved_esi));
 }
 
 #define acpi_restore_register_state()  do {} while (0)
-- 
cgit v1.2.3


From 1b17fce6078ab0672f6702097680b65124de5f05 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:36 -0700
Subject: include/asm-x86/suspend_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/suspend_64.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/suspend_64.h b/include/asm-x86/suspend_64.h
index 2eb92cb81a0d..dc3262b43072 100644
--- a/include/asm-x86/suspend_64.h
+++ b/include/asm-x86/suspend_64.h
@@ -9,8 +9,7 @@
 #include <asm/desc.h>
 #include <asm/i387.h>
 
-static inline int
-arch_prepare_suspend(void)
+static inline int arch_prepare_suspend(void)
 {
 	return 0;
 }
@@ -25,7 +24,7 @@ arch_prepare_suspend(void)
  */
 struct saved_context {
 	struct pt_regs regs;
-  	u16 ds, es, fs, gs, ss;
+	u16 ds, es, fs, gs, ss;
 	unsigned long gs_base, gs_kernel_base, fs_base;
 	unsigned long cr0, cr2, cr3, cr4, cr8;
 	unsigned long efer;
-- 
cgit v1.2.3


From a4c2d7d9285500a9b229bb7ddc7abe0212a0dab0 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:37 -0700
Subject: include/asm-x86/swiotlb.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/swiotlb.h | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/swiotlb.h b/include/asm-x86/swiotlb.h
index f9c589539a82..f5d9e74b1e4a 100644
--- a/include/asm-x86/swiotlb.h
+++ b/include/asm-x86/swiotlb.h
@@ -8,15 +8,15 @@
 extern dma_addr_t swiotlb_map_single(struct device *hwdev, void *ptr,
 				     size_t size, int dir);
 extern void *swiotlb_alloc_coherent(struct device *hwdev, size_t size,
-                       dma_addr_t *dma_handle, gfp_t flags);
+				    dma_addr_t *dma_handle, gfp_t flags);
 extern void swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
-				  size_t size, int dir);
+				 size_t size, int dir);
 extern void swiotlb_sync_single_for_cpu(struct device *hwdev,
-					 dma_addr_t dev_addr,
-					 size_t size, int dir);
+					dma_addr_t dev_addr,
+					size_t size, int dir);
 extern void swiotlb_sync_single_for_device(struct device *hwdev,
-					    dma_addr_t dev_addr,
-					    size_t size, int dir);
+					   dma_addr_t dev_addr,
+					   size_t size, int dir);
 extern void swiotlb_sync_single_range_for_cpu(struct device *hwdev,
 					      dma_addr_t dev_addr,
 					      unsigned long offset,
@@ -26,18 +26,18 @@ extern void swiotlb_sync_single_range_for_device(struct device *hwdev,
 						 unsigned long offset,
 						 size_t size, int dir);
 extern void swiotlb_sync_sg_for_cpu(struct device *hwdev,
-				     struct scatterlist *sg, int nelems,
-				     int dir);
+				    struct scatterlist *sg, int nelems,
+				    int dir);
 extern void swiotlb_sync_sg_for_device(struct device *hwdev,
-					struct scatterlist *sg, int nelems,
-					int dir);
+				       struct scatterlist *sg, int nelems,
+				       int dir);
 extern int swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg,
-		      int nents, int direction);
+			  int nents, int direction);
 extern void swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg,
-			 int nents, int direction);
+			     int nents, int direction);
 extern int swiotlb_dma_mapping_error(dma_addr_t dma_addr);
-extern void swiotlb_free_coherent (struct device *hwdev, size_t size,
-				   void *vaddr, dma_addr_t dma_handle);
+extern void swiotlb_free_coherent(struct device *hwdev, size_t size,
+				  void *vaddr, dma_addr_t dma_handle);
 extern int swiotlb_dma_supported(struct device *hwdev, u64 mask);
 extern void swiotlb_init(void);
 
-- 
cgit v1.2.3


From 26b7fcc4bde28237a906597a809b149fb06713b0 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:38 -0700
Subject: include/asm-x86/sync_bitops.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/sync_bitops.h | 56 +++++++++++++++++++++----------------------
 1 file changed, 28 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/sync_bitops.h b/include/asm-x86/sync_bitops.h
index bc249f40e0ee..f1078a5e4ed7 100644
--- a/include/asm-x86/sync_bitops.h
+++ b/include/asm-x86/sync_bitops.h
@@ -13,7 +13,7 @@
  * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
  */
 
-#define ADDR (*(volatile long *) addr)
+#define ADDR (*(volatile long *)addr)
 
 /**
  * sync_set_bit - Atomically set a bit in memory
@@ -26,12 +26,12 @@
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
-static inline void sync_set_bit(int nr, volatile unsigned long * addr)
+static inline void sync_set_bit(int nr, volatile unsigned long *addr)
 {
-	__asm__ __volatile__("lock; btsl %1,%0"
-			     :"+m" (ADDR)
-			     :"Ir" (nr)
-			     : "memory");
+	asm volatile("lock; btsl %1,%0"
+		     : "+m" (ADDR)
+		     : "Ir" (nr)
+		     : "memory");
 }
 
 /**
@@ -44,12 +44,12 @@ static inline void sync_set_bit(int nr, volatile unsigned long * addr)
  * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
  * in order to ensure changes are visible on other processors.
  */
-static inline void sync_clear_bit(int nr, volatile unsigned long * addr)
+static inline void sync_clear_bit(int nr, volatile unsigned long *addr)
 {
-	__asm__ __volatile__("lock; btrl %1,%0"
-			     :"+m" (ADDR)
-			     :"Ir" (nr)
-			     : "memory");
+	asm volatile("lock; btrl %1,%0"
+		     : "+m" (ADDR)
+		     : "Ir" (nr)
+		     : "memory");
 }
 
 /**
@@ -61,12 +61,12 @@ static inline void sync_clear_bit(int nr, volatile unsigned long * addr)
  * Note that @nr may be almost arbitrarily large; this function is not
  * restricted to acting on a single-word quantity.
  */
-static inline void sync_change_bit(int nr, volatile unsigned long * addr)
+static inline void sync_change_bit(int nr, volatile unsigned long *addr)
 {
-	__asm__ __volatile__("lock; btcl %1,%0"
-			     :"+m" (ADDR)
-			     :"Ir" (nr)
-			     : "memory");
+	asm volatile("lock; btcl %1,%0"
+		     : "+m" (ADDR)
+		     : "Ir" (nr)
+		     : "memory");
 }
 
 /**
@@ -77,13 +77,13 @@ static inline void sync_change_bit(int nr, volatile unsigned long * addr)
  * This operation is atomic and cannot be reordered.
  * It also implies a memory barrier.
  */
-static inline int sync_test_and_set_bit(int nr, volatile unsigned long * addr)
+static inline int sync_test_and_set_bit(int nr, volatile unsigned long *addr)
 {
 	int oldbit;
 
-	__asm__ __volatile__("lock; btsl %2,%1\n\tsbbl %0,%0"
-			     :"=r" (oldbit),"+m" (ADDR)
-			     :"Ir" (nr) : "memory");
+	asm volatile("lock; btsl %2,%1\n\tsbbl %0,%0"
+		     : "=r" (oldbit), "+m" (ADDR)
+		     : "Ir" (nr) : "memory");
 	return oldbit;
 }
 
@@ -95,13 +95,13 @@ static inline int sync_test_and_set_bit(int nr, volatile unsigned long * addr)
  * This operation is atomic and cannot be reordered.
  * It also implies a memory barrier.
  */
-static inline int sync_test_and_clear_bit(int nr, volatile unsigned long * addr)
+static inline int sync_test_and_clear_bit(int nr, volatile unsigned long *addr)
 {
 	int oldbit;
 
-	__asm__ __volatile__("lock; btrl %2,%1\n\tsbbl %0,%0"
-			     :"=r" (oldbit),"+m" (ADDR)
-			     :"Ir" (nr) : "memory");
+	asm volatile("lock; btrl %2,%1\n\tsbbl %0,%0"
+		     : "=r" (oldbit), "+m" (ADDR)
+		     : "Ir" (nr) : "memory");
 	return oldbit;
 }
 
@@ -113,13 +113,13 @@ static inline int sync_test_and_clear_bit(int nr, volatile unsigned long * addr)
  * This operation is atomic and cannot be reordered.
  * It also implies a memory barrier.
  */
-static inline int sync_test_and_change_bit(int nr, volatile unsigned long* addr)
+static inline int sync_test_and_change_bit(int nr, volatile unsigned long *addr)
 {
 	int oldbit;
 
-	__asm__ __volatile__("lock; btcl %2,%1\n\tsbbl %0,%0"
-			     :"=r" (oldbit),"+m" (ADDR)
-			     :"Ir" (nr) : "memory");
+	asm volatile("lock; btcl %2,%1\n\tsbbl %0,%0"
+		     : "=r" (oldbit), "+m" (ADDR)
+		     : "Ir" (nr) : "memory");
 	return oldbit;
 }
 
-- 
cgit v1.2.3


From c5386c200f55940eeeb827df172edf2e0305f23b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:39 -0700
Subject: include/asm-x86/system.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/system.h | 104 +++++++++++++++++++++++------------------------
 1 file changed, 51 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 33b0017156a7..a2f04cd79b29 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -38,35 +38,33 @@ do {									\
 	 */								\
 	unsigned long ebx, ecx, edx, esi, edi;				\
 									\
-	asm volatile(							\
-		"pushfl			\n\t"	/* save    flags */	\
-		"pushl %%ebp		\n\t"	/* save    EBP   */	\
-		"movl %%esp,%[prev_sp]	\n\t"	/* save    ESP   */	\
-		"movl %[next_sp],%%esp	\n\t"	/* restore ESP   */	\
-		"movl $1f,%[prev_ip]	\n\t"	/* save    EIP   */	\
-		"pushl %[next_ip]	\n\t"	/* restore EIP   */	\
-		"jmp __switch_to	\n"	/* regparm call  */	\
-		"1:			\t"				\
-		"popl %%ebp		\n\t"	/* restore EBP   */	\
-		"popfl			\n"	/* restore flags */	\
+	asm volatile("pushfl\n\t"		/* save    flags */	\
+		     "pushl %%ebp\n\t"		/* save    EBP   */	\
+		     "movl %%esp,%[prev_sp]\n\t"	/* save    ESP   */ \
+		     "movl %[next_sp],%%esp\n\t"	/* restore ESP   */ \
+		     "movl $1f,%[prev_ip]\n\t"	/* save    EIP   */	\
+		     "pushl %[next_ip]\n\t"	/* restore EIP   */	\
+		     "jmp __switch_to\n"	/* regparm call  */	\
+		     "1:\t"						\
+		     "popl %%ebp\n\t"		/* restore EBP   */	\
+		     "popfl\n"			/* restore flags */	\
 									\
-		  /* output parameters */				\
-		: [prev_sp] "=m" (prev->thread.sp),			\
-		  [prev_ip] "=m" (prev->thread.ip),			\
-			    "=a" (last),				\
+		     /* output parameters */				\
+		     : [prev_sp] "=m" (prev->thread.sp),		\
+		       [prev_ip] "=m" (prev->thread.ip),		\
+		       "=a" (last),					\
 									\
-		  /* clobbered output registers: */			\
-		  "=b" (ebx), "=c" (ecx), "=d" (edx),			\
-		  "=S" (esi), "=D" (edi)				\
-									\
-		  /* input parameters: */				\
-		: [next_sp]  "m" (next->thread.sp),			\
-		  [next_ip]  "m" (next->thread.ip),			\
-									\
-		  /* regparm parameters for __switch_to(): */		\
-		  [prev]     "a" (prev),				\
-		  [next]     "d" (next)					\
-	);								\
+		       /* clobbered output registers: */		\
+		       "=b" (ebx), "=c" (ecx), "=d" (edx),		\
+		       "=S" (esi), "=D" (edi)				\
+		       							\
+		       /* input parameters: */				\
+		     : [next_sp]  "m" (next->thread.sp),		\
+		       [next_ip]  "m" (next->thread.ip),		\
+		       							\
+		       /* regparm parameters for __switch_to(): */	\
+		       [prev]     "a" (prev),				\
+		       [next]     "d" (next));				\
 } while (0)
 
 /*
@@ -146,35 +144,34 @@ extern void load_gs_index(unsigned);
  */
 #define loadsegment(seg, value)			\
 	asm volatile("\n"			\
-		"1:\t"				\
-		"movl %k0,%%" #seg "\n"		\
-		"2:\n"				\
-		".section .fixup,\"ax\"\n"	\
-		"3:\t"				\
-		"movl %k1, %%" #seg "\n\t"	\
-		"jmp 2b\n"			\
-		".previous\n"			\
-		_ASM_EXTABLE(1b,3b)		\
-		: :"r" (value), "r" (0))
+		     "1:\t"			\
+		     "movl %k0,%%" #seg "\n"	\
+		     "2:\n"			\
+		     ".section .fixup,\"ax\"\n"	\
+		     "3:\t"			\
+		     "movl %k1, %%" #seg "\n\t"	\
+		     "jmp 2b\n"			\
+		     ".previous\n"		\
+		     _ASM_EXTABLE(1b,3b)	\
+		     : :"r" (value), "r" (0))
 
 
 /*
  * Save a segment register away
  */
-#define savesegment(seg, value) \
+#define savesegment(seg, value)				\
 	asm volatile("mov %%" #seg ",%0":"=rm" (value))
 
 static inline unsigned long get_limit(unsigned long segment)
 {
 	unsigned long __limit;
-	__asm__("lsll %1,%0"
-		:"=r" (__limit):"r" (segment));
-	return __limit+1;
+	asm("lsll %1,%0" : "=r" (__limit) : "r" (segment));
+	return __limit + 1;
 }
 
 static inline void native_clts(void)
 {
-	asm volatile ("clts");
+	asm volatile("clts");
 }
 
 /*
@@ -189,43 +186,43 @@ static unsigned long __force_order;
 static inline unsigned long native_read_cr0(void)
 {
 	unsigned long val;
-	asm volatile("mov %%cr0,%0\n\t" :"=r" (val), "=m" (__force_order));
+	asm volatile("mov %%cr0,%0\n\t" : "=r" (val), "=m" (__force_order));
 	return val;
 }
 
 static inline void native_write_cr0(unsigned long val)
 {
-	asm volatile("mov %0,%%cr0": :"r" (val), "m" (__force_order));
+	asm volatile("mov %0,%%cr0": : "r" (val), "m" (__force_order));
 }
 
 static inline unsigned long native_read_cr2(void)
 {
 	unsigned long val;
-	asm volatile("mov %%cr2,%0\n\t" :"=r" (val), "=m" (__force_order));
+	asm volatile("mov %%cr2,%0\n\t" : "=r" (val), "=m" (__force_order));
 	return val;
 }
 
 static inline void native_write_cr2(unsigned long val)
 {
-	asm volatile("mov %0,%%cr2": :"r" (val), "m" (__force_order));
+	asm volatile("mov %0,%%cr2": : "r" (val), "m" (__force_order));
 }
 
 static inline unsigned long native_read_cr3(void)
 {
 	unsigned long val;
-	asm volatile("mov %%cr3,%0\n\t" :"=r" (val), "=m" (__force_order));
+	asm volatile("mov %%cr3,%0\n\t" : "=r" (val), "=m" (__force_order));
 	return val;
 }
 
 static inline void native_write_cr3(unsigned long val)
 {
-	asm volatile("mov %0,%%cr3": :"r" (val), "m" (__force_order));
+	asm volatile("mov %0,%%cr3": : "r" (val), "m" (__force_order));
 }
 
 static inline unsigned long native_read_cr4(void)
 {
 	unsigned long val;
-	asm volatile("mov %%cr4,%0\n\t" :"=r" (val), "=m" (__force_order));
+	asm volatile("mov %%cr4,%0\n\t" : "=r" (val), "=m" (__force_order));
 	return val;
 }
 
@@ -237,7 +234,7 @@ static inline unsigned long native_read_cr4_safe(void)
 #ifdef CONFIG_X86_32
 	asm volatile("1: mov %%cr4, %0\n"
 		     "2:\n"
-		     _ASM_EXTABLE(1b,2b)
+		     _ASM_EXTABLE(1b, 2b)
 		     : "=r" (val), "=m" (__force_order) : "0" (0));
 #else
 	val = native_read_cr4();
@@ -247,7 +244,7 @@ static inline unsigned long native_read_cr4_safe(void)
 
 static inline void native_write_cr4(unsigned long val)
 {
-	asm volatile("mov %0,%%cr4": :"r" (val), "m" (__force_order));
+	asm volatile("mov %0,%%cr4": : "r" (val), "m" (__force_order));
 }
 
 #ifdef CONFIG_X86_64
@@ -268,6 +265,7 @@ static inline void native_wbinvd(void)
 {
 	asm volatile("wbinvd": : :"memory");
 }
+
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
@@ -300,7 +298,7 @@ static inline void clflush(volatile void *__p)
 	asm volatile("clflush %0" : "+m" (*(volatile char __force *)__p));
 }
 
-#define nop() __asm__ __volatile__ ("nop")
+#define nop() asm volatile ("nop")
 
 void disable_hlt(void);
 void enable_hlt(void);
@@ -399,7 +397,7 @@ void default_idle(void);
 # define smp_wmb()	barrier()
 #endif
 #define smp_read_barrier_depends()	read_barrier_depends()
-#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
+#define set_mb(var, value) do { (void)xchg(&var, value); } while (0)
 #else
 #define smp_mb()	barrier()
 #define smp_rmb()	barrier()
-- 
cgit v1.2.3


From 7c4d4784db93496791b798e0abab1d056b192ad0 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:40 -0700
Subject: include/asm-x86/tce.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/tce.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/tce.h b/include/asm-x86/tce.h
index cd955d3d112f..b1a4ea00df78 100644
--- a/include/asm-x86/tce.h
+++ b/include/asm-x86/tce.h
@@ -39,7 +39,7 @@ struct iommu_table;
 #define TCE_RPN_MASK     0x0000fffffffff000ULL
 
 extern void tce_build(struct iommu_table *tbl, unsigned long index,
-        unsigned int npages, unsigned long uaddr, int direction);
+		      unsigned int npages, unsigned long uaddr, int direction);
 extern void tce_free(struct iommu_table *tbl, long index, unsigned int npages);
 extern void * __init alloc_tce_table(void);
 extern void __init free_tce_table(void *tbl);
-- 
cgit v1.2.3


From 89917f28f3377fa7e38a51e9208e83b7b92542ee Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:42 -0700
Subject: include/asm-x86/thread_info_32.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/thread_info_32.h | 88 +++++++++++++++++++++-------------------
 1 file changed, 46 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/thread_info_32.h b/include/asm-x86/thread_info_32.h
index 5bd508260ffb..4e053fa561a9 100644
--- a/include/asm-x86/thread_info_32.h
+++ b/include/asm-x86/thread_info_32.h
@@ -20,7 +20,8 @@
  * low level task data that entry.S needs immediate access to
  * - this struct should fit entirely inside of one cache line
  * - this struct shares the supervisor stack pages
- * - if the contents of this structure are changed, the assembly constants must also be changed
+ * - if the contents of this structure are changed,
+ *   the assembly constants must also be changed
  */
 #ifndef __ASSEMBLY__
 
@@ -30,18 +31,16 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	unsigned long		status;		/* thread-synchronous flags */
 	__u32			cpu;		/* current CPU */
-	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
-
-
+	int			preempt_count;	/* 0 => preemptable,
+						   <0 => BUG */
 	mm_segment_t		addr_limit;	/* thread address space:
-					 	   0-0xBFFFFFFF for user-thead
-						   0-0xFFFFFFFF for kernel-thread
+						   0-0xBFFFFFFF user-thread
+						   0-0xFFFFFFFF kernel-thread
 						*/
 	void			*sysenter_return;
 	struct restart_block    restart_block;
-
-	unsigned long           previous_esp;   /* ESP of the previous stack in case
-						   of nested (IRQ) stacks
+	unsigned long           previous_esp;   /* ESP of the previous stack in
+						   case of nested (IRQ) stacks
 						*/
 	__u8			supervisor_stack[0];
 };
@@ -90,15 +89,16 @@ register unsigned long current_stack_pointer asm("esp") __used;
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
 {
-	return (struct thread_info *)(current_stack_pointer & ~(THREAD_SIZE - 1));
+	return (struct thread_info *)
+		(current_stack_pointer & ~(THREAD_SIZE - 1));
 }
 
 /* thread information allocation */
 #ifdef CONFIG_DEBUG_STACK_USAGE
-#define alloc_thread_info(tsk) ((struct thread_info *) \
-	__get_free_pages(GFP_KERNEL| __GFP_ZERO, get_order(THREAD_SIZE)))
+#define alloc_thread_info(tsk) ((struct thread_info *)			\
+	__get_free_pages(GFP_KERNEL | __GFP_ZERO, get_order(THREAD_SIZE)))
 #else
-#define alloc_thread_info(tsk) ((struct thread_info *) \
+#define alloc_thread_info(tsk) ((struct thread_info *)			\
 	__get_free_pages(GFP_KERNEL, get_order(THREAD_SIZE)))
 #endif
 
@@ -107,7 +107,7 @@ static inline struct thread_info *current_thread_info(void)
 #else /* !__ASSEMBLY__ */
 
 /* how to get the thread information struct from ASM */
-#define GET_THREAD_INFO(reg) \
+#define GET_THREAD_INFO(reg)	 \
 	movl $-THREAD_SIZE, reg; \
 	andl %esp, reg
 
@@ -119,14 +119,16 @@ static inline struct thread_info *current_thread_info(void)
 
 /*
  * thread information flags
- * - these are process state flags that various assembly files may need to access
+ * - these are process state flags that various
+ *   assembly files may need to access
  * - pending work-to-be-done flags are in LSW
  * - other flags in MSW
  */
 #define TIF_SYSCALL_TRACE	0	/* syscall trace active */
 #define TIF_SIGPENDING		1	/* signal pending */
 #define TIF_NEED_RESCHED	2	/* rescheduling necessary */
-#define TIF_SINGLESTEP		3	/* restore singlestep on return to user mode */
+#define TIF_SINGLESTEP		3	/* restore singlestep on return to
+					   user mode */
 #define TIF_IRET		4	/* return with iret */
 #define TIF_SYSCALL_EMU		5	/* syscall emulation active */
 #define TIF_SYSCALL_AUDIT	6	/* syscall auditing active */
@@ -143,36 +145,36 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_DS_AREA_MSR 	23      /* uses thread_struct.ds_area_msr */
 #define TIF_BTS_TRACE_TS        24      /* record scheduling event timestamps */
 
-#define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
-#define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
-#define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
-#define _TIF_SINGLESTEP		(1<<TIF_SINGLESTEP)
-#define _TIF_IRET		(1<<TIF_IRET)
-#define _TIF_SYSCALL_EMU	(1<<TIF_SYSCALL_EMU)
-#define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
-#define _TIF_SECCOMP		(1<<TIF_SECCOMP)
-#define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
-#define _TIF_HRTICK_RESCHED	(1<<TIF_HRTICK_RESCHED)
-#define _TIF_DEBUG		(1<<TIF_DEBUG)
-#define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
-#define _TIF_FREEZE		(1<<TIF_FREEZE)
-#define _TIF_NOTSC		(1<<TIF_NOTSC)
-#define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
-#define _TIF_DEBUGCTLMSR	(1<<TIF_DEBUGCTLMSR)
-#define _TIF_DS_AREA_MSR	(1<<TIF_DS_AREA_MSR)
-#define _TIF_BTS_TRACE_TS	(1<<TIF_BTS_TRACE_TS)
+#define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
+#define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
+#define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
+#define _TIF_SINGLESTEP		(1 << TIF_SINGLESTEP)
+#define _TIF_IRET		(1 << TIF_IRET)
+#define _TIF_SYSCALL_EMU	(1 << TIF_SYSCALL_EMU)
+#define _TIF_SYSCALL_AUDIT	(1 << TIF_SYSCALL_AUDIT)
+#define _TIF_SECCOMP		(1 << TIF_SECCOMP)
+#define _TIF_RESTORE_SIGMASK	(1 << TIF_RESTORE_SIGMASK)
+#define _TIF_HRTICK_RESCHED	(1 << TIF_HRTICK_RESCHED)
+#define _TIF_DEBUG		(1 << TIF_DEBUG)
+#define _TIF_IO_BITMAP		(1 << TIF_IO_BITMAP)
+#define _TIF_FREEZE		(1 << TIF_FREEZE)
+#define _TIF_NOTSC		(1 << TIF_NOTSC)
+#define _TIF_FORCED_TF		(1 << TIF_FORCED_TF)
+#define _TIF_DEBUGCTLMSR	(1 << TIF_DEBUGCTLMSR)
+#define _TIF_DS_AREA_MSR	(1 << TIF_DS_AREA_MSR)
+#define _TIF_BTS_TRACE_TS	(1 << TIF_BTS_TRACE_TS)
 
 /* work to do on interrupt/exception return */
-#define _TIF_WORK_MASK \
-  (0x0000FFFF & ~(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | \
-		  _TIF_SECCOMP | _TIF_SYSCALL_EMU))
+#define _TIF_WORK_MASK							\
+	(0x0000FFFF & ~(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT |	\
+			_TIF_SECCOMP | _TIF_SYSCALL_EMU))
 /* work to do on any return to u-space */
 #define _TIF_ALLWORK_MASK	(0x0000FFFF & ~_TIF_SECCOMP)
 
 /* flags to check in __switch_to() */
-#define _TIF_WORK_CTXSW \
-    (_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUGCTLMSR | \
-     _TIF_DS_AREA_MSR | _TIF_BTS_TRACE_TS)
+#define _TIF_WORK_CTXSW						\
+	(_TIF_IO_BITMAP | _TIF_NOTSC | _TIF_DEBUGCTLMSR |	\
+	 _TIF_DS_AREA_MSR | _TIF_BTS_TRACE_TS)
 #define _TIF_WORK_CTXSW_PREV _TIF_WORK_CTXSW
 #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW | _TIF_DEBUG)
 
@@ -184,8 +186,10 @@ static inline struct thread_info *current_thread_info(void)
  * ever touches our thread-synchronous status, so we don't
  * have to worry about atomic accesses.
  */
-#define TS_USEDFPU		0x0001	/* FPU was used by this task this quantum (SMP) */
-#define TS_POLLING		0x0002	/* True if in idle loop and not sleeping */
+#define TS_USEDFPU		0x0001	/* FPU was used by this task
+					   this quantum (SMP) */
+#define TS_POLLING		0x0002	/* True if in idle loop
+					   and not sleeping */
 
 #define tsk_is_polling(t) (task_thread_info(t)->status & TS_POLLING)
 
-- 
cgit v1.2.3


From b98fff30223799c5df444fef1ebcfcddf310f740 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:43 -0700
Subject: include/asm-x86/thread_info_64.h: checkpatch cleanups - formatting
 only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/thread_info_64.h | 78 +++++++++++++++++++++-------------------
 1 file changed, 41 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/thread_info_64.h b/include/asm-x86/thread_info_64.h
index 6c9b214b8fc3..1e5c6f6152cd 100644
--- a/include/asm-x86/thread_info_64.h
+++ b/include/asm-x86/thread_info_64.h
@@ -29,9 +29,9 @@ struct thread_info {
 	__u32			flags;		/* low level flags */
 	__u32			status;		/* thread synchronous flags */
 	__u32			cpu;		/* current CPU */
-	int 			preempt_count;	/* 0 => preemptable, <0 => BUG */
-
-	mm_segment_t		addr_limit;	
+	int 			preempt_count;	/* 0 => preemptable,
+						   <0 => BUG */
+	mm_segment_t		addr_limit;
 	struct restart_block    restart_block;
 #ifdef CONFIG_IA32_EMULATION
 	void __user		*sysenter_return;
@@ -61,17 +61,17 @@ struct thread_info {
 #define init_stack		(init_thread_union.stack)
 
 static inline struct thread_info *current_thread_info(void)
-{ 
+{
 	struct thread_info *ti;
 	ti = (void *)(read_pda(kernelstack) + PDA_STACKOFFSET - THREAD_SIZE);
-	return ti; 
+	return ti;
 }
 
 /* do not use in interrupt context */
 static inline struct thread_info *stack_thread_info(void)
 {
 	struct thread_info *ti;
-	__asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (~(THREAD_SIZE - 1)));
+	asm("andq %%rsp,%0; " : "=r" (ti) : "0" (~(THREAD_SIZE - 1)));
 	return ti;
 }
 
@@ -82,8 +82,8 @@ static inline struct thread_info *stack_thread_info(void)
 #define THREAD_FLAGS GFP_KERNEL
 #endif
 
-#define alloc_thread_info(tsk) \
-	((struct thread_info *) __get_free_pages(THREAD_FLAGS, THREAD_ORDER))
+#define alloc_thread_info(tsk)						\
+	((struct thread_info *)__get_free_pages(THREAD_FLAGS, THREAD_ORDER))
 
 #define free_thread_info(ti) free_pages((unsigned long) (ti), THREAD_ORDER)
 
@@ -98,7 +98,8 @@ static inline struct thread_info *stack_thread_info(void)
 
 /*
  * thread information flags
- * - these are process state flags that various assembly files may need to access
+ * - these are process state flags that various assembly files
+ *   may need to access
  * - pending work-to-be-done flags are in LSW
  * - other flags in MSW
  * Warning: layout of LSW is hardcoded in entry.S
@@ -114,7 +115,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_MCE_NOTIFY		10	/* notify userspace of an MCE */
 #define TIF_HRTICK_RESCHED	11	/* reprogram hrtick timer */
 /* 16 free */
-#define TIF_IA32		17	/* 32bit process */ 
+#define TIF_IA32		17	/* 32bit process */
 #define TIF_FORK		18	/* ret_from_fork */
 #define TIF_ABI_PENDING		19
 #define TIF_MEMDIE		20
@@ -126,39 +127,40 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_DS_AREA_MSR		26      /* uses thread_struct.ds_area_msr */
 #define TIF_BTS_TRACE_TS	27      /* record scheduling event timestamps */
 
-#define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
-#define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
-#define _TIF_SINGLESTEP		(1<<TIF_SINGLESTEP)
-#define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
-#define _TIF_IRET		(1<<TIF_IRET)
-#define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
-#define _TIF_SECCOMP		(1<<TIF_SECCOMP)
-#define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
-#define _TIF_MCE_NOTIFY		(1<<TIF_MCE_NOTIFY)
-#define _TIF_HRTICK_RESCHED	(1<<TIF_HRTICK_RESCHED)
-#define _TIF_IA32		(1<<TIF_IA32)
-#define _TIF_FORK		(1<<TIF_FORK)
-#define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
-#define _TIF_DEBUG		(1<<TIF_DEBUG)
-#define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
-#define _TIF_FREEZE		(1<<TIF_FREEZE)
-#define _TIF_FORCED_TF		(1<<TIF_FORCED_TF)
-#define _TIF_DEBUGCTLMSR	(1<<TIF_DEBUGCTLMSR)
-#define _TIF_DS_AREA_MSR	(1<<TIF_DS_AREA_MSR)
-#define _TIF_BTS_TRACE_TS	(1<<TIF_BTS_TRACE_TS)
+#define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
+#define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
+#define _TIF_SINGLESTEP		(1 << TIF_SINGLESTEP)
+#define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
+#define _TIF_IRET		(1 << TIF_IRET)
+#define _TIF_SYSCALL_AUDIT	(1 << TIF_SYSCALL_AUDIT)
+#define _TIF_SECCOMP		(1 << TIF_SECCOMP)
+#define _TIF_RESTORE_SIGMASK	(1 << TIF_RESTORE_SIGMASK)
+#define _TIF_MCE_NOTIFY		(1 << TIF_MCE_NOTIFY)
+#define _TIF_HRTICK_RESCHED	(1 << TIF_HRTICK_RESCHED)
+#define _TIF_IA32		(1 << TIF_IA32)
+#define _TIF_FORK		(1 << TIF_FORK)
+#define _TIF_ABI_PENDING	(1 << TIF_ABI_PENDING)
+#define _TIF_DEBUG		(1 << TIF_DEBUG)
+#define _TIF_IO_BITMAP		(1 << TIF_IO_BITMAP)
+#define _TIF_FREEZE		(1 << TIF_FREEZE)
+#define _TIF_FORCED_TF		(1 << TIF_FORCED_TF)
+#define _TIF_DEBUGCTLMSR	(1 << TIF_DEBUGCTLMSR)
+#define _TIF_DS_AREA_MSR	(1 << TIF_DS_AREA_MSR)
+#define _TIF_BTS_TRACE_TS	(1 << TIF_BTS_TRACE_TS)
 
 /* work to do on interrupt/exception return */
-#define _TIF_WORK_MASK \
-  (0x0000FFFF & ~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP|_TIF_SECCOMP))
+#define _TIF_WORK_MASK							\
+	(0x0000FFFF &							\
+	 ~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP|_TIF_SECCOMP))
 /* work to do on any return to user space */
 #define _TIF_ALLWORK_MASK (0x0000FFFF & ~_TIF_SECCOMP)
 
-#define _TIF_DO_NOTIFY_MASK \
+#define _TIF_DO_NOTIFY_MASK						\
 	(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY|_TIF_HRTICK_RESCHED)
 
 /* flags to check in __switch_to() */
-#define _TIF_WORK_CTXSW \
-    (_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR|_TIF_DS_AREA_MSR|_TIF_BTS_TRACE_TS)
+#define _TIF_WORK_CTXSW							\
+	(_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR|_TIF_DS_AREA_MSR|_TIF_BTS_TRACE_TS)
 #define _TIF_WORK_CTXSW_PREV _TIF_WORK_CTXSW
 #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW|_TIF_DEBUG)
 
@@ -171,9 +173,11 @@ static inline struct thread_info *stack_thread_info(void)
  * ever touches our thread-synchronous status, so we don't
  * have to worry about atomic accesses.
  */
-#define TS_USEDFPU		0x0001	/* FPU was used by this task this quantum (SMP) */
+#define TS_USEDFPU		0x0001	/* FPU was used by this task
+					   this quantum (SMP) */
 #define TS_COMPAT		0x0002	/* 32bit syscall active */
-#define TS_POLLING		0x0004	/* true if in idle loop and not sleeping */
+#define TS_POLLING		0x0004	/* true if in idle loop
+					   and not sleeping */
 
 #define tsk_is_polling(t) (task_thread_info(t)->status & TS_POLLING)
 
-- 
cgit v1.2.3


From 94cf8de0a0aff39c7b7785af4fc938ecacb79b7c Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:45 -0700
Subject: include/asm-x86/tlbflush.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/tlbflush.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/tlbflush.h b/include/asm-x86/tlbflush.h
index 3998709ed637..0c0674d94255 100644
--- a/include/asm-x86/tlbflush.h
+++ b/include/asm-x86/tlbflush.h
@@ -32,7 +32,7 @@ static inline void __native_flush_tlb_global(void)
 
 static inline void __native_flush_tlb_single(unsigned long addr)
 {
-	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory");
+	asm volatile("invlpg (%0)" ::"r" (addr) : "memory");
 }
 
 static inline void __flush_tlb_all(void)
@@ -134,8 +134,7 @@ void native_flush_tlb_others(const cpumask_t *cpumask, struct mm_struct *mm,
 #define TLBSTATE_LAZY	2
 
 #ifdef CONFIG_X86_32
-struct tlb_state
-{
+struct tlb_state {
 	struct mm_struct *active_mm;
 	int state;
 	char __cacheline_padding[L1_CACHE_BYTES-8];
-- 
cgit v1.2.3


From 5d7d03b81af05f3c291b5c6be621a2b53d187e09 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:46 -0700
Subject: include/asm-x86/topology.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/topology.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index dada89e5b152..8d1a1f3d21b4 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -71,7 +71,7 @@ static inline int cpu_to_node(int cpu)
 #ifdef CONFIG_DEBUG_PER_CPU_MAPS
 	if (x86_cpu_to_node_map_early_ptr) {
 		printk("KERN_NOTICE cpu_to_node(%d): usage too early!\n",
-			(int)cpu);
+		       (int)cpu);
 		dump_stack();
 		return ((int *)x86_cpu_to_node_map_early_ptr)[cpu];
 	}
-- 
cgit v1.2.3


From 2d86e637d15984e363e8c3f14c8f0470b4a10a3d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:47 -0700
Subject: include/asm-x86/tsc.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/tsc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/tsc.h b/include/asm-x86/tsc.h
index 7d3e27f7d484..d2d8eb5b55f5 100644
--- a/include/asm-x86/tsc.h
+++ b/include/asm-x86/tsc.h
@@ -42,7 +42,7 @@ static inline cycles_t vget_cycles(void)
 	if (!cpu_has_tsc)
 		return 0;
 #endif
-	return (cycles_t) __native_read_tsc();
+	return (cycles_t)__native_read_tsc();
 }
 
 extern void tsc_init(void);
-- 
cgit v1.2.3


From b1fcec7f2296c4b9126e1b85b52494ac8910d528 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:48 -0700
Subject: include/asm-x86/uaccess_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/uaccess_32.h | 316 +++++++++++++++++++++++++------------------
 1 file changed, 187 insertions(+), 129 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/uaccess_32.h b/include/asm-x86/uaccess_32.h
index fcc570ec4fee..8e7595c1f34e 100644
--- a/include/asm-x86/uaccess_32.h
+++ b/include/asm-x86/uaccess_32.h
@@ -32,7 +32,7 @@
 #define get_fs()	(current_thread_info()->addr_limit)
 #define set_fs(x)	(current_thread_info()->addr_limit = (x))
 
-#define segment_eq(a,b)	((a).seg == (b).seg)
+#define segment_eq(a, b)	((a).seg == (b).seg)
 
 /*
  * movsl can be slow when source and dest are not both 8-byte aligned
@@ -43,7 +43,9 @@ extern struct movsl_mask {
 } ____cacheline_aligned_in_smp movsl_mask;
 #endif
 
-#define __addr_ok(addr) ((unsigned long __force)(addr) < (current_thread_info()->addr_limit.seg))
+#define __addr_ok(addr)					\
+	((unsigned long __force)(addr) <		\
+	 (current_thread_info()->addr_limit.seg))
 
 /*
  * Test whether a block of memory is a valid user space address.
@@ -54,13 +56,16 @@ extern struct movsl_mask {
  *
  * This needs 33-bit arithmetic. We have a carry...
  */
-#define __range_ok(addr,size) ({ \
-	unsigned long flag,roksum; \
-	__chk_user_ptr(addr); \
-	asm("addl %3,%1 ; sbbl %0,%0; cmpl %1,%4; sbbl $0,%0" \
-		:"=&r" (flag), "=r" (roksum) \
-		:"1" (addr),"g" ((int)(size)),"rm" (current_thread_info()->addr_limit.seg)); \
-	flag; })
+#define __range_ok(addr, size)						\
+({									\
+	unsigned long flag, roksum;					\
+	__chk_user_ptr(addr);						\
+	asm("addl %3,%1 ; sbbl %0,%0; cmpl %1,%4; sbbl $0,%0"		\
+	    :"=&r" (flag), "=r" (roksum)				\
+	    :"1" (addr), "g" ((int)(size)),				\
+	    "rm" (current_thread_info()->addr_limit.seg));		\
+	flag;								\
+})
 
 /**
  * access_ok: - Checks if a user space pointer is valid
@@ -81,7 +86,7 @@ extern struct movsl_mask {
  * checks that the pointer is in the user space range - after calling
  * this function, memory access functions may still return -EFAULT.
  */
-#define access_ok(type,addr,size) (likely(__range_ok(addr,size) == 0))
+#define access_ok(type, addr, size) (likely(__range_ok(addr, size) == 0))
 
 /*
  * The exception table consists of pairs of addresses: the first is the
@@ -96,8 +101,7 @@ extern struct movsl_mask {
  * on our cache or tlb entries.
  */
 
-struct exception_table_entry
-{
+struct exception_table_entry {
 	unsigned long insn, fixup;
 };
 
@@ -122,13 +126,15 @@ extern void __get_user_1(void);
 extern void __get_user_2(void);
 extern void __get_user_4(void);
 
-#define __get_user_x(size,ret,x,ptr) \
-	__asm__ __volatile__("call __get_user_" #size \
-		:"=a" (ret),"=d" (x) \
-		:"0" (ptr))
+#define __get_user_x(size, ret, x, ptr)	      \
+	asm volatile("call __get_user_" #size \
+		     :"=a" (ret),"=d" (x)     \
+		     :"0" (ptr))
+
 
+/* Careful: we have to cast the result to the type of the pointer
+ * for sign reasons */
 
-/* Careful: we have to cast the result to the type of the pointer for sign reasons */
 /**
  * get_user: - Get a simple variable from user space.
  * @x:   Variable to store result.
@@ -146,15 +152,24 @@ extern void __get_user_4(void);
  * Returns zero on success, or -EFAULT on error.
  * On error, the variable @x is set to zero.
  */
-#define get_user(x,ptr)							\
-({	int __ret_gu;							\
+#define get_user(x, ptr)						\
+({									\
+	int __ret_gu;							\
 	unsigned long __val_gu;						\
 	__chk_user_ptr(ptr);						\
-	switch(sizeof (*(ptr))) {					\
-	case 1:  __get_user_x(1,__ret_gu,__val_gu,ptr); break;		\
-	case 2:  __get_user_x(2,__ret_gu,__val_gu,ptr); break;		\
-	case 4:  __get_user_x(4,__ret_gu,__val_gu,ptr); break;		\
-	default: __get_user_x(X,__ret_gu,__val_gu,ptr); break;		\
+	switch (sizeof(*(ptr))) {					\
+	case 1:								\
+		__get_user_x(1, __ret_gu, __val_gu, ptr);		\
+		break;							\
+	case 2:								\
+		__get_user_x(2, __ret_gu, __val_gu, ptr);		\
+		break;							\
+	case 4:								\
+		__get_user_x(4, __ret_gu, __val_gu, ptr);		\
+		break;							\
+	default:							\
+		__get_user_x(X, __ret_gu, __val_gu, ptr);		\
+		break;							\
 	}								\
 	(x) = (__typeof__(*(ptr)))__val_gu;				\
 	__ret_gu;							\
@@ -171,11 +186,25 @@ extern void __put_user_2(void);
 extern void __put_user_4(void);
 extern void __put_user_8(void);
 
-#define __put_user_1(x, ptr) __asm__ __volatile__("call __put_user_1":"=a" (__ret_pu):"0" ((typeof(*(ptr)))(x)), "c" (ptr))
-#define __put_user_2(x, ptr) __asm__ __volatile__("call __put_user_2":"=a" (__ret_pu):"0" ((typeof(*(ptr)))(x)), "c" (ptr))
-#define __put_user_4(x, ptr) __asm__ __volatile__("call __put_user_4":"=a" (__ret_pu):"0" ((typeof(*(ptr)))(x)), "c" (ptr))
-#define __put_user_8(x, ptr) __asm__ __volatile__("call __put_user_8":"=a" (__ret_pu):"A" ((typeof(*(ptr)))(x)), "c" (ptr))
-#define __put_user_X(x, ptr) __asm__ __volatile__("call __put_user_X":"=a" (__ret_pu):"c" (ptr))
+#define __put_user_1(x, ptr)					\
+	asm volatile("call __put_user_1" : "=a" (__ret_pu)	\
+		     : "0" ((typeof(*(ptr)))(x)), "c" (ptr))
+
+#define __put_user_2(x, ptr)					\
+	asm volatile("call __put_user_2" : "=a" (__ret_pu)	\
+		     : "0" ((typeof(*(ptr)))(x)), "c" (ptr))
+
+#define __put_user_4(x, ptr)					\
+	asm volatile("call __put_user_4" : "=a" (__ret_pu)	\
+		     : "0" ((typeof(*(ptr)))(x)), "c" (ptr))
+
+#define __put_user_8(x, ptr)					\
+	asm volatile("call __put_user_8" : "=a" (__ret_pu)	\
+		     : "A" ((typeof(*(ptr)))(x)), "c" (ptr))
+
+#define __put_user_X(x, ptr)					\
+	asm volatile("call __put_user_X" : "=a" (__ret_pu)	\
+		     : "c" (ptr))
 
 /**
  * put_user: - Write a simple value into user space.
@@ -195,32 +224,43 @@ extern void __put_user_8(void);
  */
 #ifdef CONFIG_X86_WP_WORKS_OK
 
-#define put_user(x,ptr)						\
-({	int __ret_pu;						\
+#define put_user(x, ptr)					\
+({								\
+	int __ret_pu;						\
 	__typeof__(*(ptr)) __pu_val;				\
 	__chk_user_ptr(ptr);					\
 	__pu_val = x;						\
-	switch(sizeof(*(ptr))) {				\
-	case 1: __put_user_1(__pu_val, ptr); break;		\
-	case 2: __put_user_2(__pu_val, ptr); break;		\
-	case 4: __put_user_4(__pu_val, ptr); break;		\
-	case 8: __put_user_8(__pu_val, ptr); break;		\
-	default:__put_user_X(__pu_val, ptr); break;		\
+	switch (sizeof(*(ptr))) {				\
+	case 1:							\
+		__put_user_1(__pu_val, ptr);			\
+		break;						\
+	case 2:							\
+		__put_user_2(__pu_val, ptr);			\
+		break;						\
+	case 4:							\
+		__put_user_4(__pu_val, ptr);			\
+		break;						\
+	case 8:							\
+		__put_user_8(__pu_val, ptr);			\
+		break;						\
+	default:						\
+		__put_user_X(__pu_val, ptr);			\
+		break;						\
 	}							\
 	__ret_pu;						\
 })
 
 #else
-#define put_user(x,ptr)						\
+#define put_user(x, ptr)					\
 ({								\
- 	int __ret_pu;						\
-	__typeof__(*(ptr)) __pus_tmp = x;			\
-	__ret_pu=0;						\
-	if(unlikely(__copy_to_user_ll(ptr, &__pus_tmp,		\
-				sizeof(*(ptr))) != 0))		\
- 		__ret_pu=-EFAULT;				\
- 	__ret_pu;						\
- })
+	int __ret_pu;						\
+	__typeof__(*(ptr))__pus_tmp = x;			\
+	__ret_pu = 0;						\
+	if (unlikely(__copy_to_user_ll(ptr, &__pus_tmp,		\
+				       sizeof(*(ptr))) != 0))	\
+		__ret_pu = -EFAULT;				\
+	__ret_pu;						\
+})
 
 
 #endif
@@ -245,8 +285,8 @@ extern void __put_user_8(void);
  * Returns zero on success, or -EFAULT on error.
  * On error, the variable @x is set to zero.
  */
-#define __get_user(x,ptr) \
-  __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+#define __get_user(x, ptr)				\
+	__get_user_nocheck((x), (ptr), sizeof(*(ptr)))
 
 
 /**
@@ -268,54 +308,62 @@ extern void __put_user_8(void);
  *
  * Returns zero on success, or -EFAULT on error.
  */
-#define __put_user(x,ptr) \
-  __put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
+#define __put_user(x, ptr)						\
+	__put_user_nocheck((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr)))
 
-#define __put_user_nocheck(x,ptr,size)				\
+#define __put_user_nocheck(x, ptr, size)			\
 ({								\
 	long __pu_err;						\
-	__put_user_size((x),(ptr),(size),__pu_err,-EFAULT);	\
+	__put_user_size((x), (ptr), (size), __pu_err, -EFAULT);	\
 	__pu_err;						\
 })
 
 
-#define __put_user_u64(x, addr, err)				\
-	__asm__ __volatile__(					\
-		"1:	movl %%eax,0(%2)\n"			\
-		"2:	movl %%edx,4(%2)\n"			\
-		"3:\n"						\
-		".section .fixup,\"ax\"\n"			\
-		"4:	movl %3,%0\n"				\
-		"	jmp 3b\n"				\
-		".previous\n"					\
-		_ASM_EXTABLE(1b,4b)				\
-		_ASM_EXTABLE(2b,4b)				\
-		: "=r"(err)					\
-		: "A" (x), "r" (addr), "i"(-EFAULT), "0"(err))
+#define __put_user_u64(x, addr, err)					\
+	asm volatile("1:	movl %%eax,0(%2)\n"			\
+		     "2:	movl %%edx,4(%2)\n"			\
+		     "3:\n"						\
+		     ".section .fixup,\"ax\"\n"				\
+		     "4:	movl %3,%0\n"				\
+		     "	jmp 3b\n"					\
+		     ".previous\n"					\
+		     _ASM_EXTABLE(1b, 4b)				\
+		     _ASM_EXTABLE(2b, 4b)				\
+		     : "=r" (err)					\
+		     : "A" (x), "r" (addr), "i" (-EFAULT), "0" (err))
 
 #ifdef CONFIG_X86_WP_WORKS_OK
 
-#define __put_user_size(x,ptr,size,retval,errret)			\
+#define __put_user_size(x, ptr, size, retval, errret)			\
 do {									\
 	retval = 0;							\
 	__chk_user_ptr(ptr);						\
 	switch (size) {							\
-	case 1: __put_user_asm(x,ptr,retval,"b","b","iq",errret);break;	\
-	case 2: __put_user_asm(x,ptr,retval,"w","w","ir",errret);break; \
-	case 4: __put_user_asm(x,ptr,retval,"l","","ir",errret); break;	\
-	case 8: __put_user_u64((__typeof__(*ptr))(x),ptr,retval); break;\
-	  default: __put_user_bad();					\
+	case 1:								\
+		__put_user_asm(x, ptr, retval, "b", "b", "iq", errret);	\
+		break;							\
+	case 2:								\
+		__put_user_asm(x, ptr, retval, "w", "w", "ir", errret);	\
+		break;							\
+	case 4:								\
+		__put_user_asm(x, ptr, retval, "l", "",  "ir", errret);	\
+		break;							\
+	case 8:								\
+		__put_user_u64((__typeof__(*ptr))(x), ptr, retval);	\
+		break;							\
+	default:							\
+		__put_user_bad();					\
 	}								\
 } while (0)
 
 #else
 
-#define __put_user_size(x,ptr,size,retval,errret)			\
+#define __put_user_size(x, ptr, size, retval, errret)			\
 do {									\
-	__typeof__(*(ptr)) __pus_tmp = x;				\
+	__typeof__(*(ptr))__pus_tmp = x;				\
 	retval = 0;							\
 									\
-	if(unlikely(__copy_to_user_ll(ptr, &__pus_tmp, size) != 0))	\
+	if (unlikely(__copy_to_user_ll(ptr, &__pus_tmp, size) != 0))	\
 		retval = errret;					\
 } while (0)
 
@@ -329,65 +377,70 @@ struct __large_struct { unsigned long buf[100]; };
  * aliasing issues.
  */
 #define __put_user_asm(x, addr, err, itype, rtype, ltype, errret)	\
-	__asm__ __volatile__(						\
-		"1:	mov"itype" %"rtype"1,%2\n"			\
-		"2:\n"							\
-		".section .fixup,\"ax\"\n"				\
-		"3:	movl %3,%0\n"					\
-		"	jmp 2b\n"					\
-		".previous\n"						\
-		_ASM_EXTABLE(1b,3b)					\
-		: "=r"(err)						\
-		: ltype (x), "m"(__m(addr)), "i"(errret), "0"(err))
-
-
-#define __get_user_nocheck(x,ptr,size)				\
-({								\
-	long __gu_err;						\
-	unsigned long __gu_val;					\
-	__get_user_size(__gu_val,(ptr),(size),__gu_err,-EFAULT);\
-	(x) = (__typeof__(*(ptr)))__gu_val;			\
-	__gu_err;						\
+	asm volatile("1:	mov"itype" %"rtype"1,%2\n"		\
+		     "2:\n"						\
+		     ".section .fixup,\"ax\"\n"				\
+		     "3:	movl %3,%0\n"				\
+		     "	jmp 2b\n"					\
+		     ".previous\n"					\
+		     _ASM_EXTABLE(1b, 3b)				\
+		     : "=r"(err)					\
+		     : ltype (x), "m" (__m(addr)), "i" (errret), "0" (err))
+
+
+#define __get_user_nocheck(x, ptr, size)				\
+({									\
+	long __gu_err;							\
+	unsigned long __gu_val;						\
+	__get_user_size(__gu_val, (ptr), (size), __gu_err, -EFAULT);	\
+	(x) = (__typeof__(*(ptr)))__gu_val;				\
+	__gu_err;							\
 })
 
 extern long __get_user_bad(void);
 
-#define __get_user_size(x,ptr,size,retval,errret)			\
+#define __get_user_size(x, ptr, size, retval, errret)			\
 do {									\
 	retval = 0;							\
 	__chk_user_ptr(ptr);						\
 	switch (size) {							\
-	case 1: __get_user_asm(x,ptr,retval,"b","b","=q",errret);break;	\
-	case 2: __get_user_asm(x,ptr,retval,"w","w","=r",errret);break;	\
-	case 4: __get_user_asm(x,ptr,retval,"l","","=r",errret);break;	\
-	default: (x) = __get_user_bad();				\
+	case 1:								\
+		__get_user_asm(x, ptr, retval, "b", "b", "=q", errret);	\
+		break;							\
+	case 2:								\
+		__get_user_asm(x, ptr, retval, "w", "w", "=r", errret);	\
+		break;							\
+	case 4:								\
+		__get_user_asm(x, ptr, retval, "l", "", "=r", errret);	\
+		break;							\
+	default:							\
+		(x) = __get_user_bad();					\
 	}								\
 } while (0)
 
 #define __get_user_asm(x, addr, err, itype, rtype, ltype, errret)	\
-	__asm__ __volatile__(						\
-		"1:	mov"itype" %2,%"rtype"1\n"			\
-		"2:\n"							\
-		".section .fixup,\"ax\"\n"				\
-		"3:	movl %3,%0\n"					\
-		"	xor"itype" %"rtype"1,%"rtype"1\n"		\
-		"	jmp 2b\n"					\
-		".previous\n"						\
-		_ASM_EXTABLE(1b,3b)					\
-		: "=r"(err), ltype (x)					\
-		: "m"(__m(addr)), "i"(errret), "0"(err))
-
-
-unsigned long __must_check __copy_to_user_ll(void __user *to,
-				const void *from, unsigned long n);
-unsigned long __must_check __copy_from_user_ll(void *to,
-				const void __user *from, unsigned long n);
-unsigned long __must_check __copy_from_user_ll_nozero(void *to,
-				const void __user *from, unsigned long n);
-unsigned long __must_check __copy_from_user_ll_nocache(void *to,
-				const void __user *from, unsigned long n);
-unsigned long __must_check __copy_from_user_ll_nocache_nozero(void *to,
-				const void __user *from, unsigned long n);
+	asm volatile("1:	mov"itype" %2,%"rtype"1\n"		\
+		     "2:\n"						\
+		     ".section .fixup,\"ax\"\n"				\
+		     "3:	movl %3,%0\n"				\
+		     "	xor"itype" %"rtype"1,%"rtype"1\n"		\
+		     "	jmp 2b\n"					\
+		     ".previous\n"					\
+		     _ASM_EXTABLE(1b, 3b)				\
+		     : "=r" (err), ltype (x)				\
+		     : "m" (__m(addr)), "i" (errret), "0" (err))
+
+
+unsigned long __must_check __copy_to_user_ll
+		(void __user *to, const void *from, unsigned long n);
+unsigned long __must_check __copy_from_user_ll
+		(void *to, const void __user *from, unsigned long n);
+unsigned long __must_check __copy_from_user_ll_nozero
+		(void *to, const void __user *from, unsigned long n);
+unsigned long __must_check __copy_from_user_ll_nocache
+		(void *to, const void __user *from, unsigned long n);
+unsigned long __must_check __copy_from_user_ll_nocache_nozero
+		(void *to, const void __user *from, unsigned long n);
 
 /**
  * __copy_to_user_inatomic: - Copy a block of data into user space, with less checking.
@@ -416,13 +469,16 @@ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
 
 		switch (n) {
 		case 1:
-			__put_user_size(*(u8 *)from, (u8 __user *)to, 1, ret, 1);
+			__put_user_size(*(u8 *)from, (u8 __user *)to,
+					1, ret, 1);
 			return ret;
 		case 2:
-			__put_user_size(*(u16 *)from, (u16 __user *)to, 2, ret, 2);
+			__put_user_size(*(u16 *)from, (u16 __user *)to,
+					2, ret, 2);
 			return ret;
 		case 4:
-			__put_user_size(*(u32 *)from, (u32 __user *)to, 4, ret, 4);
+			__put_user_size(*(u32 *)from, (u32 __user *)to,
+					4, ret, 4);
 			return ret;
 		}
 	}
@@ -545,19 +601,21 @@ static __always_inline unsigned long __copy_from_user_nocache(void *to,
 }
 
 static __always_inline unsigned long
-__copy_from_user_inatomic_nocache(void *to, const void __user *from, unsigned long n)
+__copy_from_user_inatomic_nocache(void *to, const void __user *from,
+				  unsigned long n)
 {
        return __copy_from_user_ll_nocache_nozero(to, from, n);
 }
 
 unsigned long __must_check copy_to_user(void __user *to,
-				const void *from, unsigned long n);
+					const void *from, unsigned long n);
 unsigned long __must_check copy_from_user(void *to,
-				const void __user *from, unsigned long n);
+					  const void __user *from,
+					  unsigned long n);
 long __must_check strncpy_from_user(char *dst, const char __user *src,
-				long count);
+				    long count);
 long __must_check __strncpy_from_user(char *dst,
-				const char __user *src, long count);
+				      const char __user *src, long count);
 
 /**
  * strlen_user: - Get the size of a string in user space.
-- 
cgit v1.2.3


From b896313e53344e79cc8bbc69f0a7d5c2b1735895 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:49 -0700
Subject: include/asm-x86/uaccess_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/uaccess_64.h | 376 ++++++++++++++++++++++++++-----------------
 1 file changed, 227 insertions(+), 149 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/uaccess_64.h b/include/asm-x86/uaccess_64.h
index b87eb4ba8f9d..b8a2f4339903 100644
--- a/include/asm-x86/uaccess_64.h
+++ b/include/asm-x86/uaccess_64.h
@@ -29,23 +29,27 @@
 #define get_fs()	(current_thread_info()->addr_limit)
 #define set_fs(x)	(current_thread_info()->addr_limit = (x))
 
-#define segment_eq(a,b)	((a).seg == (b).seg)
+#define segment_eq(a, b)	((a).seg == (b).seg)
 
-#define __addr_ok(addr) (!((unsigned long)(addr) & (current_thread_info()->addr_limit.seg)))
+#define __addr_ok(addr) (!((unsigned long)(addr) &			\
+			   (current_thread_info()->addr_limit.seg)))
 
 /*
  * Uhhuh, this needs 65-bit arithmetic. We have a carry..
  */
-#define __range_not_ok(addr,size) ({ \
-	unsigned long flag,roksum; \
-	__chk_user_ptr(addr); \
-	asm("# range_ok\n\r" \
-		"addq %3,%1 ; sbbq %0,%0 ; cmpq %1,%4 ; sbbq $0,%0"  \
-		:"=&r" (flag), "=r" (roksum) \
-		:"1" (addr),"g" ((long)(size)),"g" (current_thread_info()->addr_limit.seg)); \
-	flag; })
+#define __range_not_ok(addr, size)					\
+({									\
+	unsigned long flag, roksum;					\
+	__chk_user_ptr(addr);						\
+	asm("# range_ok\n\r"						\
+	    "addq %3,%1 ; sbbq %0,%0 ; cmpq %1,%4 ; sbbq $0,%0"		\
+	    : "=&r" (flag), "=r" (roksum)				\
+	    : "1" (addr), "g" ((long)(size)),				\
+	      "g" (current_thread_info()->addr_limit.seg));		\
+	flag;								\
+})
 
-#define access_ok(type, addr, size) (__range_not_ok(addr,size) == 0)
+#define access_ok(type, addr, size) (__range_not_ok(addr, size) == 0)
 
 /*
  * The exception table consists of pairs of addresses: the first is the
@@ -60,8 +64,7 @@
  * on our cache or tlb entries.
  */
 
-struct exception_table_entry
-{
+struct exception_table_entry {
 	unsigned long insn, fixup;
 };
 
@@ -84,23 +87,36 @@ extern int fixup_exception(struct pt_regs *regs);
  * accesses to the same area of user memory).
  */
 
-#define __get_user_x(size,ret,x,ptr) \
-	asm volatile("call __get_user_" #size \
-		:"=a" (ret),"=d" (x) \
-		:"c" (ptr) \
-		:"r8")
+#define __get_user_x(size, ret, x, ptr)		      \
+	asm volatile("call __get_user_" #size	      \
+		     : "=a" (ret),"=d" (x)	      \
+		     : "c" (ptr)		      \
+		     : "r8")
+
+/* Careful: we have to cast the result to the type of the pointer
+ * for sign reasons */
 
-/* Careful: we have to cast the result to the type of the pointer for sign reasons */
-#define get_user(x,ptr)							\
-({	unsigned long __val_gu;						\
-	int __ret_gu; 							\
+#define get_user(x, ptr)						\
+({									\
+	unsigned long __val_gu;						\
+	int __ret_gu;							\
 	__chk_user_ptr(ptr);						\
-	switch(sizeof (*(ptr))) {					\
-	case 1:  __get_user_x(1,__ret_gu,__val_gu,ptr); break;		\
-	case 2:  __get_user_x(2,__ret_gu,__val_gu,ptr); break;		\
-	case 4:  __get_user_x(4,__ret_gu,__val_gu,ptr); break;		\
-	case 8:  __get_user_x(8,__ret_gu,__val_gu,ptr); break;		\
-	default: __get_user_bad(); break;				\
+	switch (sizeof(*(ptr))) {					\
+	case 1:								\
+		__get_user_x(1, __ret_gu, __val_gu, ptr);		\
+		break;							\
+	case 2:								\
+		__get_user_x(2, __ret_gu, __val_gu, ptr);		\
+		break;							\
+	case 4:								\
+		__get_user_x(4, __ret_gu, __val_gu, ptr);		\
+		break;							\
+	case 8:								\
+		__get_user_x(8, __ret_gu, __val_gu, ptr);		\
+		break;							\
+	default:							\
+		__get_user_bad();					\
+		break;							\
 	}								\
 	(x) = (__force typeof(*(ptr)))__val_gu;				\
 	__ret_gu;							\
@@ -112,55 +128,73 @@ extern void __put_user_4(void);
 extern void __put_user_8(void);
 extern void __put_user_bad(void);
 
-#define __put_user_x(size,ret,x,ptr)					\
-	asm volatile("call __put_user_" #size			\
-		:"=a" (ret)						\
-		:"c" (ptr),"d" (x)					\
-		:"r8")
+#define __put_user_x(size, ret, x, ptr)					\
+	asm volatile("call __put_user_" #size				\
+		     :"=a" (ret)					\
+		     :"c" (ptr),"d" (x)					\
+		     :"r8")
 
-#define put_user(x,ptr)							\
-  __put_user_check((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
+#define put_user(x, ptr)						\
+	__put_user_check((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr)))
 
-#define __get_user(x,ptr) \
-  __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
-#define __put_user(x,ptr) \
-  __put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
+#define __get_user(x, ptr)						\
+	__get_user_nocheck((x), (ptr), sizeof(*(ptr)))
+#define __put_user(x, ptr)						\
+	__put_user_nocheck((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr)))
 
 #define __get_user_unaligned __get_user
 #define __put_user_unaligned __put_user
 
-#define __put_user_nocheck(x,ptr,size)			\
+#define __put_user_nocheck(x, ptr, size)		\
 ({							\
 	int __pu_err;					\
-	__put_user_size((x),(ptr),(size),__pu_err);	\
+	__put_user_size((x), (ptr), (size), __pu_err);	\
 	__pu_err;					\
 })
 
 
-#define __put_user_check(x,ptr,size)			\
-({							\
-	int __pu_err;					\
-	typeof(*(ptr)) __user *__pu_addr = (ptr);	\
-	switch (size) { 				\
-	case 1: __put_user_x(1,__pu_err,x,__pu_addr); break;	\
-	case 2: __put_user_x(2,__pu_err,x,__pu_addr); break;	\
-	case 4: __put_user_x(4,__pu_err,x,__pu_addr); break;	\
-	case 8: __put_user_x(8,__pu_err,x,__pu_addr); break;	\
-	default: __put_user_bad();			\
-	}						\
-	__pu_err;					\
+#define __put_user_check(x, ptr, size)				\
+({								\
+	int __pu_err;						\
+	typeof(*(ptr)) __user *__pu_addr = (ptr);		\
+	switch (size) {						\
+	case 1:							\
+		__put_user_x(1, __pu_err, x, __pu_addr);	\
+		break;						\
+	case 2:							\
+		__put_user_x(2, __pu_err, x, __pu_addr);	\
+		break;						\
+	case 4:							\
+		__put_user_x(4, __pu_err, x, __pu_addr);	\
+		break;						\
+	case 8:							\
+		__put_user_x(8, __pu_err, x, __pu_addr);	\
+		break;						\
+	default:						\
+		__put_user_bad();				\
+	}							\
+	__pu_err;						\
 })
 
-#define __put_user_size(x,ptr,size,retval)				\
+#define __put_user_size(x, ptr, size, retval)				\
 do {									\
 	retval = 0;							\
 	__chk_user_ptr(ptr);						\
 	switch (size) {							\
-	  case 1: __put_user_asm(x,ptr,retval,"b","b","iq",-EFAULT); break;\
-	  case 2: __put_user_asm(x,ptr,retval,"w","w","ir",-EFAULT); break;\
-	  case 4: __put_user_asm(x,ptr,retval,"l","k","ir",-EFAULT); break;\
-	  case 8: __put_user_asm(x,ptr,retval,"q","","Zr",-EFAULT); break;\
-	  default: __put_user_bad();					\
+	case 1:								\
+		__put_user_asm(x, ptr, retval, "b", "b", "iq", -EFAULT);\
+		break;							\
+	case 2:								\
+		__put_user_asm(x, ptr, retval, "w", "w", "ir", -EFAULT);\
+		break;							\
+	case 4:								\
+		__put_user_asm(x, ptr, retval, "l", "k", "ir", -EFAULT);\
+		break;							\
+	case 8:								\
+		__put_user_asm(x, ptr, retval, "q", "", "Zr", -EFAULT);	\
+		break;							\
+	default:							\
+		__put_user_bad();					\
 	}								\
 } while (0)
 
@@ -174,23 +208,22 @@ struct __large_struct { unsigned long buf[100]; };
  * aliasing issues.
  */
 #define __put_user_asm(x, addr, err, itype, rtype, ltype, errno)	\
-	asm volatile(					\
-		"1:	mov"itype" %"rtype"1,%2\n"		\
-		"2:\n"						\
-		".section .fixup,\"ax\"\n"			\
-		"3:	mov %3,%0\n"				\
-		"	jmp 2b\n"				\
-		".previous\n"					\
-		_ASM_EXTABLE(1b,3b)				\
-		: "=r"(err)					\
-		: ltype (x), "m"(__m(addr)), "i"(errno), "0"(err))
-
-
-#define __get_user_nocheck(x,ptr,size)				\
+	asm volatile("1:	mov"itype" %"rtype"1,%2\n"		\
+		     "2:\n"						\
+		     ".section .fixup, \"ax\"\n"			\
+		     "3:	mov %3,%0\n"				\
+		     "	jmp 2b\n"					\
+		     ".previous\n"					\
+		     _ASM_EXTABLE(1b, 3b)				\
+		     : "=r"(err)					\
+		     : ltype (x), "m" (__m(addr)), "i" (errno), "0" (err))
+
+
+#define __get_user_nocheck(x, ptr, size)			\
 ({								\
 	int __gu_err;						\
 	unsigned long __gu_val;					\
-	__get_user_size(__gu_val,(ptr),(size),__gu_err);	\
+	__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
 	(x) = (__force typeof(*(ptr)))__gu_val;			\
 	__gu_err;						\
 })
@@ -201,31 +234,39 @@ extern int __get_user_4(void);
 extern int __get_user_8(void);
 extern int __get_user_bad(void);
 
-#define __get_user_size(x,ptr,size,retval)				\
+#define __get_user_size(x, ptr, size, retval)				\
 do {									\
 	retval = 0;							\
 	__chk_user_ptr(ptr);						\
 	switch (size) {							\
-	  case 1: __get_user_asm(x,ptr,retval,"b","b","=q",-EFAULT); break;\
-	  case 2: __get_user_asm(x,ptr,retval,"w","w","=r",-EFAULT); break;\
-	  case 4: __get_user_asm(x,ptr,retval,"l","k","=r",-EFAULT); break;\
-	  case 8: __get_user_asm(x,ptr,retval,"q","","=r",-EFAULT); break;\
-	  default: (x) = __get_user_bad();				\
+	case 1:								\
+		__get_user_asm(x, ptr, retval, "b", "b", "=q", -EFAULT);\
+		break;							\
+	case 2:								\
+		__get_user_asm(x, ptr, retval, "w", "w", "=r", -EFAULT);\
+		break;							\
+	case 4:								\
+		__get_user_asm(x, ptr, retval, "l", "k", "=r", -EFAULT);\
+		break;							\
+	case 8:								\
+		__get_user_asm(x, ptr, retval, "q", "", "=r", -EFAULT);	\
+		break;							\
+	default:							\
+		(x) = __get_user_bad();					\
 	}								\
 } while (0)
 
 #define __get_user_asm(x, addr, err, itype, rtype, ltype, errno)	\
-	asm volatile(					\
-		"1:	mov"itype" %2,%"rtype"1\n"		\
-		"2:\n"						\
-		".section .fixup,\"ax\"\n"			\
-		"3:	mov %3,%0\n"				\
-		"	xor"itype" %"rtype"1,%"rtype"1\n"	\
-		"	jmp 2b\n"				\
-		".previous\n"					\
-		_ASM_EXTABLE(1b,3b)				\
-		: "=r"(err), ltype (x)				\
-		: "m"(__m(addr)), "i"(errno), "0"(err))
+	asm volatile("1:	mov"itype" %2,%"rtype"1\n"		\
+		     "2:\n"						\
+		     ".section .fixup, \"ax\"\n"			\
+		     "3:	mov %3,%0\n"				\
+		     "	xor"itype" %"rtype"1,%"rtype"1\n"		\
+		     "	jmp 2b\n"					\
+		     ".previous\n"					\
+		     _ASM_EXTABLE(1b, 3b)				\
+		     : "=r" (err), ltype (x)				\
+		     : "m" (__m(addr)), "i"(errno), "0"(err))
 
 /*
  * Copy To/From Userspace
@@ -244,110 +285,142 @@ copy_in_user(void __user *to, const void __user *from, unsigned len);
 
 static __always_inline __must_check
 int __copy_from_user(void *dst, const void __user *src, unsigned size)
-{ 
+{
 	int ret = 0;
 	if (!__builtin_constant_p(size))
-		return copy_user_generic(dst,(__force void *)src,size);
-	switch (size) { 
-	case 1:__get_user_asm(*(u8*)dst,(u8 __user *)src,ret,"b","b","=q",1); 
+		return copy_user_generic(dst, (__force void *)src, size);
+	switch (size) {
+	case 1:__get_user_asm(*(u8 *)dst, (u8 __user *)src,
+			      ret, "b", "b", "=q", 1);
 		return ret;
-	case 2:__get_user_asm(*(u16*)dst,(u16 __user *)src,ret,"w","w","=r",2);
+	case 2:__get_user_asm(*(u16 *)dst, (u16 __user *)src,
+			      ret, "w", "w", "=r", 2);
 		return ret;
-	case 4:__get_user_asm(*(u32*)dst,(u32 __user *)src,ret,"l","k","=r",4);
+	case 4:__get_user_asm(*(u32 *)dst, (u32 __user *)src,
+			      ret, "l", "k", "=r", 4);
+		return ret;
+	case 8:__get_user_asm(*(u64 *)dst, (u64 __user *)src,
+			      ret, "q", "", "=r", 8);
 		return ret;
-	case 8:__get_user_asm(*(u64*)dst,(u64 __user *)src,ret,"q","","=r",8);
-		return ret; 
 	case 10:
-	       	__get_user_asm(*(u64*)dst,(u64 __user *)src,ret,"q","","=r",16);
-		if (unlikely(ret)) return ret;
-		__get_user_asm(*(u16*)(8+(char*)dst),(u16 __user *)(8+(char __user *)src),ret,"w","w","=r",2);
-		return ret; 
+		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
+			       ret, "q", "", "=r", 16);
+		if (unlikely(ret))
+			return ret;
+		__get_user_asm(*(u16 *)(8 + (char *)dst),
+			       (u16 __user *)(8 + (char __user *)src),
+			       ret, "w", "w", "=r", 2);
+		return ret;
 	case 16:
-		__get_user_asm(*(u64*)dst,(u64 __user *)src,ret,"q","","=r",16);
-		if (unlikely(ret)) return ret;
-		__get_user_asm(*(u64*)(8+(char*)dst),(u64 __user *)(8+(char __user *)src),ret,"q","","=r",8);
-		return ret; 
+		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
+			       ret, "q", "", "=r", 16);
+		if (unlikely(ret))
+			return ret;
+		__get_user_asm(*(u64 *)(8 + (char *)dst),
+			       (u64 __user *)(8 + (char __user *)src),
+			       ret, "q", "", "=r", 8);
+		return ret;
 	default:
-		return copy_user_generic(dst,(__force void *)src,size); 
+		return copy_user_generic(dst, (__force void *)src, size);
 	}
-}	
+}
 
 static __always_inline __must_check
 int __copy_to_user(void __user *dst, const void *src, unsigned size)
-{ 
+{
 	int ret = 0;
 	if (!__builtin_constant_p(size))
-		return copy_user_generic((__force void *)dst,src,size);
-	switch (size) { 
-	case 1:__put_user_asm(*(u8*)src,(u8 __user *)dst,ret,"b","b","iq",1); 
+		return copy_user_generic((__force void *)dst, src, size);
+	switch (size) {
+	case 1:__put_user_asm(*(u8 *)src, (u8 __user *)dst,
+			      ret, "b", "b", "iq", 1);
 		return ret;
-	case 2:__put_user_asm(*(u16*)src,(u16 __user *)dst,ret,"w","w","ir",2);
+	case 2:__put_user_asm(*(u16 *)src, (u16 __user *)dst,
+			      ret, "w", "w", "ir", 2);
 		return ret;
-	case 4:__put_user_asm(*(u32*)src,(u32 __user *)dst,ret,"l","k","ir",4);
+	case 4:__put_user_asm(*(u32 *)src, (u32 __user *)dst,
+			      ret, "l", "k", "ir", 4);
+		return ret;
+	case 8:__put_user_asm(*(u64 *)src, (u64 __user *)dst,
+			      ret, "q", "", "ir", 8);
 		return ret;
-	case 8:__put_user_asm(*(u64*)src,(u64 __user *)dst,ret,"q","","ir",8);
-		return ret; 
 	case 10:
-		__put_user_asm(*(u64*)src,(u64 __user *)dst,ret,"q","","ir",10);
-		if (unlikely(ret)) return ret;
+		__put_user_asm(*(u64 *)src, (u64 __user *)dst,
+			       ret, "q", "", "ir", 10);
+		if (unlikely(ret))
+			return ret;
 		asm("":::"memory");
-		__put_user_asm(4[(u16*)src],4+(u16 __user *)dst,ret,"w","w","ir",2);
-		return ret; 
+		__put_user_asm(4[(u16 *)src], 4 + (u16 __user *)dst,
+			       ret, "w", "w", "ir", 2);
+		return ret;
 	case 16:
-		__put_user_asm(*(u64*)src,(u64 __user *)dst,ret,"q","","ir",16);
-		if (unlikely(ret)) return ret;
+		__put_user_asm(*(u64 *)src, (u64 __user *)dst,
+			       ret, "q", "", "ir", 16);
+		if (unlikely(ret))
+			return ret;
 		asm("":::"memory");
-		__put_user_asm(1[(u64*)src],1+(u64 __user *)dst,ret,"q","","ir",8);
-		return ret; 
+		__put_user_asm(1[(u64 *)src], 1 + (u64 __user *)dst,
+			       ret, "q", "", "ir", 8);
+		return ret;
 	default:
-		return copy_user_generic((__force void *)dst,src,size); 
+		return copy_user_generic((__force void *)dst, src, size);
 	}
-}	
+}
 
 static __always_inline __must_check
 int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
-{ 
+{
 	int ret = 0;
 	if (!__builtin_constant_p(size))
-		return copy_user_generic((__force void *)dst,(__force void *)src,size);
-	switch (size) { 
-	case 1: { 
+		return copy_user_generic((__force void *)dst,
+					 (__force void *)src, size);
+	switch (size) {
+	case 1: {
 		u8 tmp;
-		__get_user_asm(tmp,(u8 __user *)src,ret,"b","b","=q",1); 
+		__get_user_asm(tmp, (u8 __user *)src,
+			       ret, "b", "b", "=q", 1);
 		if (likely(!ret))
-			__put_user_asm(tmp,(u8 __user *)dst,ret,"b","b","iq",1); 
+			__put_user_asm(tmp, (u8 __user *)dst,
+				       ret, "b", "b", "iq", 1);
 		return ret;
 	}
-	case 2: { 
+	case 2: {
 		u16 tmp;
-		__get_user_asm(tmp,(u16 __user *)src,ret,"w","w","=r",2); 
+		__get_user_asm(tmp, (u16 __user *)src,
+			       ret, "w", "w", "=r", 2);
 		if (likely(!ret))
-			__put_user_asm(tmp,(u16 __user *)dst,ret,"w","w","ir",2); 
+			__put_user_asm(tmp, (u16 __user *)dst,
+				       ret, "w", "w", "ir", 2);
 		return ret;
 	}
 
-	case 4: { 
+	case 4: {
 		u32 tmp;
-		__get_user_asm(tmp,(u32 __user *)src,ret,"l","k","=r",4); 
+		__get_user_asm(tmp, (u32 __user *)src,
+			       ret, "l", "k", "=r", 4);
 		if (likely(!ret))
-			__put_user_asm(tmp,(u32 __user *)dst,ret,"l","k","ir",4); 
+			__put_user_asm(tmp, (u32 __user *)dst,
+				       ret, "l", "k", "ir", 4);
 		return ret;
 	}
-	case 8: { 
+	case 8: {
 		u64 tmp;
-		__get_user_asm(tmp,(u64 __user *)src,ret,"q","","=r",8); 
+		__get_user_asm(tmp, (u64 __user *)src,
+			       ret, "q", "", "=r", 8);
 		if (likely(!ret))
-			__put_user_asm(tmp,(u64 __user *)dst,ret,"q","","ir",8); 
+			__put_user_asm(tmp, (u64 __user *)dst,
+				       ret, "q", "", "ir", 8);
 		return ret;
 	}
 	default:
-		return copy_user_generic((__force void *)dst,(__force void *)src,size); 
+		return copy_user_generic((__force void *)dst,
+					 (__force void *)src, size);
 	}
-}	
+}
 
-__must_check long 
+__must_check long
 strncpy_from_user(char *dst, const char __user *src, long count);
-__must_check long 
+__must_check long
 __strncpy_from_user(char *dst, const char __user *src, long count);
 __must_check long strnlen_user(const char __user *str, long n);
 __must_check long __strnlen_user(const char __user *str, long n);
@@ -355,7 +428,8 @@ __must_check long strlen_user(const char __user *str);
 __must_check unsigned long clear_user(void __user *mem, unsigned long len);
 __must_check unsigned long __clear_user(void __user *mem, unsigned long len);
 
-__must_check long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
+__must_check long __copy_from_user_inatomic(void *dst, const void __user *src,
+					    unsigned size);
 
 static __must_check __always_inline int
 __copy_to_user_inatomic(void __user *dst, const void *src, unsigned size)
@@ -364,15 +438,19 @@ __copy_to_user_inatomic(void __user *dst, const void *src, unsigned size)
 }
 
 #define ARCH_HAS_NOCACHE_UACCESS 1
-extern long __copy_user_nocache(void *dst, const void __user *src, unsigned size, int zerorest);
+extern long __copy_user_nocache(void *dst, const void __user *src,
+				unsigned size, int zerorest);
 
-static inline int __copy_from_user_nocache(void *dst, const void __user *src, unsigned size)
+static inline int __copy_from_user_nocache(void *dst, const void __user *src,
+					   unsigned size)
 {
 	might_sleep();
 	return __copy_user_nocache(dst, src, size, 1);
 }
 
-static inline int __copy_from_user_inatomic_nocache(void *dst, const void __user *src, unsigned size)
+static inline int __copy_from_user_inatomic_nocache(void *dst,
+						    const void __user *src,
+						    unsigned size)
 {
 	return __copy_user_nocache(dst, src, size, 0);
 }
-- 
cgit v1.2.3


From 6e714b37978bbd2c7a7eb89b4474b6c2133b7796 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:50 -0700
Subject: include/asm-x86/unaligned.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/unaligned.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/unaligned.h b/include/asm-x86/unaligned.h
index 913598d4f761..d270ffe72759 100644
--- a/include/asm-x86/unaligned.h
+++ b/include/asm-x86/unaligned.h
@@ -32,6 +32,6 @@
  *
  * Note that unaligned accesses can be very expensive on some architectures.
  */
-#define put_unaligned(val, ptr) ((void)( *(ptr) = (val) ))
+#define put_unaligned(val, ptr) ((void)(*(ptr) = (val)))
 
 #endif /* _ASM_X86_UNALIGNED_H */
-- 
cgit v1.2.3


From 687fc16b65e96d72a680291670584090207cadf8 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:51 -0700
Subject: include/asm-x86/unistd_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/unistd_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index 984123a68f7c..8317d94771d3 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -81,7 +81,7 @@
 #define __NR_sigpending		 73
 #define __NR_sethostname	 74
 #define __NR_setrlimit		 75
-#define __NR_getrlimit		 76	/* Back compatible 2Gig limited rlimit */
+#define __NR_getrlimit		 76   /* Back compatible 2Gig limited rlimit */
 #define __NR_getrusage		 77
 #define __NR_gettimeofday	 78
 #define __NR_settimeofday	 79
-- 
cgit v1.2.3


From c489f4451965f4d355340ea1e60a5863c9ed2890 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:52 -0700
Subject: include/asm-x86/unistd_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/unistd_64.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index 3883ceb54ef5..fe26e36d0f51 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -2,7 +2,7 @@
 #define _ASM_X86_64_UNISTD_H_
 
 #ifndef __SYSCALL
-#define __SYSCALL(a,b)
+#define __SYSCALL(a, b)
 #endif
 
 /*
-- 
cgit v1.2.3


From 826700dc9b483d0d0de8ff4901043d54ed8b64f0 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:53 -0700
Subject: include/asm-x86/user_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/user_32.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/user_32.h b/include/asm-x86/user_32.h
index 6157da6f882c..d6e51edc259d 100644
--- a/include/asm-x86/user_32.h
+++ b/include/asm-x86/user_32.h
@@ -100,10 +100,10 @@ struct user_regs_struct {
 struct user{
 /* We start with the registers, to mimic the way that "memory" is returned
    from the ptrace(3,...) function.  */
-  struct user_regs_struct regs;		/* Where the registers are actually stored */
+  struct user_regs_struct regs;	/* Where the registers are actually stored */
 /* ptrace does not yet supply these.  Someday.... */
   int u_fpvalid;		/* True if math co-processor being used. */
-                                /* for this mess. Not yet used. */
+				/* for this mess. Not yet used. */
   struct user_i387_struct i387;	/* Math Co-processor registers. */
 /* The rest of this junk is to help gdb figure out what goes where */
   unsigned long int u_tsize;	/* Text segment size (pages). */
@@ -118,7 +118,7 @@ struct user{
   int reserved;			/* No longer used */
   unsigned long u_ar0;		/* Used by gdb to help find the values for */
 				/* the registers. */
-  struct user_i387_struct* u_fpstate;	/* Math Co-processor pointer. */
+  struct user_i387_struct *u_fpstate;	/* Math Co-processor pointer. */
   unsigned long magic;		/* To uniquely identify a core file */
   char u_comm[32];		/* User command that was responsible */
   int u_debugreg[8];
-- 
cgit v1.2.3


From a31216194c84cfa72515db4f9365ce2c68ed5791 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:54 -0700
Subject: include/asm-x86/user32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/user32.h | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/user32.h b/include/asm-x86/user32.h
index f769872debea..a3d910047879 100644
--- a/include/asm-x86/user32.h
+++ b/include/asm-x86/user32.h
@@ -1,7 +1,8 @@
 #ifndef USER32_H
 #define USER32_H 1
 
-/* IA32 compatible user structures for ptrace. These should be used for 32bit coredumps too. */
+/* IA32 compatible user structures for ptrace.
+ * These should be used for 32bit coredumps too. */
 
 struct user_i387_ia32_struct {
 	u32	cwd;
@@ -42,9 +43,9 @@ struct user_regs_struct32 {
 };
 
 struct user32 {
-  struct user_regs_struct32 regs;		/* Where the registers are actually stored */
+  struct user_regs_struct32 regs; /* Where the registers are actually stored */
   int u_fpvalid;		/* True if math co-processor being used. */
-                                /* for this mess. Not yet used. */
+				/* for this mess. Not yet used. */
   struct user_i387_ia32_struct i387;	/* Math Co-processor registers. */
 /* The rest of this junk is to help gdb figure out what goes where */
   __u32 u_tsize;	/* Text segment size (pages). */
-- 
cgit v1.2.3


From a206ea11b665cfb5360d05367eea1e9cfd3f3c8b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:55 -0700
Subject: include/asm-x86/user_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/user_64.h | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/user_64.h b/include/asm-x86/user_64.h
index 963616455609..6037b634c77f 100644
--- a/include/asm-x86/user_64.h
+++ b/include/asm-x86/user_64.h
@@ -45,12 +45,13 @@
  */
 
 /* This matches the 64bit FXSAVE format as defined by AMD. It is the same
-   as the 32bit format defined by Intel, except that the selector:offset pairs for
-   data and eip are replaced with flat 64bit pointers. */
+   as the 32bit format defined by Intel, except that the selector:offset pairs
+   for data and eip are replaced with flat 64bit pointers. */
 struct user_i387_struct {
 	unsigned short	cwd;
 	unsigned short	swd;
-	unsigned short	twd; /* Note this is not the same as the 32bit/x87/FSAVE twd */
+	unsigned short	twd;	/* Note this is not the same as
+				   the 32bit/x87/FSAVE twd */
 	unsigned short	fop;
 	__u64	rip;
 	__u64	rdp;
@@ -97,13 +98,14 @@ struct user_regs_struct {
 /* When the kernel dumps core, it starts by dumping the user struct -
    this will be used by gdb to figure out where the data and stack segments
    are within the file, and what virtual addresses to use. */
-struct user{
+
+struct user {
 /* We start with the registers, to mimic the way that "memory" is returned
    from the ptrace(3,...) function.  */
-  struct user_regs_struct regs;		/* Where the registers are actually stored */
+  struct user_regs_struct regs;	/* Where the registers are actually stored */
 /* ptrace does not yet supply these.  Someday.... */
   int u_fpvalid;		/* True if math co-processor being used. */
-                                /* for this mess. Not yet used. */
+				/* for this mess. Not yet used. */
   int pad0;
   struct user_i387_struct i387;	/* Math Co-processor registers. */
 /* The rest of this junk is to help gdb figure out what goes where */
@@ -120,7 +122,7 @@ struct user{
   int pad1;
   unsigned long u_ar0;		/* Used by gdb to help find the values for */
 				/* the registers. */
-  struct user_i387_struct* u_fpstate;	/* Math Co-processor pointer. */
+  struct user_i387_struct *u_fpstate;	/* Math Co-processor pointer. */
   unsigned long magic;		/* To uniquely identify a core file */
   char u_comm[32];		/* User command that was responsible */
   unsigned long u_debugreg[8];
-- 
cgit v1.2.3


From ac1a7b0eaa1db04143ab6132c6ce4489afbb8a18 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:56 -0700
Subject: include/asm-x86/vdso.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/vdso.h | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/vdso.h b/include/asm-x86/vdso.h
index 9bb86899abf0..86e085e003d2 100644
--- a/include/asm-x86/vdso.h
+++ b/include/asm-x86/vdso.h
@@ -8,9 +8,11 @@ extern const char VDSO64_PRELINK[];
  * Given a pointer to the vDSO image, find the pointer to VDSO64_name
  * as that symbol is defined in the vDSO sources or linker script.
  */
-#define VDSO64_SYMBOL(base, name) ({		\
-	extern const char VDSO64_##name[];	\
-	(void *) (VDSO64_##name - VDSO64_PRELINK + (unsigned long) (base)); })
+#define VDSO64_SYMBOL(base, name)					\
+({									\
+	extern const char VDSO64_##name[];				\
+	(void *)(VDSO64_##name - VDSO64_PRELINK + (unsigned long)(base)); \
+})
 #endif
 
 #if defined CONFIG_X86_32 || defined CONFIG_COMPAT
@@ -20,9 +22,11 @@ extern const char VDSO32_PRELINK[];
  * Given a pointer to the vDSO image, find the pointer to VDSO32_name
  * as that symbol is defined in the vDSO sources or linker script.
  */
-#define VDSO32_SYMBOL(base, name) ({		\
-	extern const char VDSO32_##name[];	\
-	(void *) (VDSO32_##name - VDSO32_PRELINK + (unsigned long) (base)); })
+#define VDSO32_SYMBOL(base, name)					\
+({									\
+	extern const char VDSO32_##name[];				\
+	(void *)(VDSO32_##name - VDSO32_PRELINK + (unsigned long)(base)); \
+})
 #endif
 
 /*
-- 
cgit v1.2.3


From 364fe5ef4725176324ec17f8dc3fd488d615b0de Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:57 -0700
Subject: include/asm-x86/vga.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/vga.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/vga.h b/include/asm-x86/vga.h
index 0ecf68ac03aa..0ccf804377e6 100644
--- a/include/asm-x86/vga.h
+++ b/include/asm-x86/vga.h
@@ -12,9 +12,9 @@
  *	access the videoram directly without any black magic.
  */
 
-#define VGA_MAP_MEM(x,s) (unsigned long)phys_to_virt(x)
+#define VGA_MAP_MEM(x, s) (unsigned long)phys_to_virt(x)
 
 #define vga_readb(x) (*(x))
-#define vga_writeb(x,y) (*(y) = (x))
+#define vga_writeb(x, y) (*(y) = (x))
 
 #endif
-- 
cgit v1.2.3


From 9e8a935bcff6c8154eace12277e6a9d853ef790b Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:58 -0700
Subject: include/asm-x86/vm86.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/vm86.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/vm86.h b/include/asm-x86/vm86.h
index c92fe4af52e8..a2be241ed036 100644
--- a/include/asm-x86/vm86.h
+++ b/include/asm-x86/vm86.h
@@ -42,9 +42,11 @@
 #define VM86_ARG(retval)	((retval) >> 8)
 
 #define VM86_SIGNAL	0	/* return due to signal */
-#define VM86_UNKNOWN	1	/* unhandled GP fault - IO-instruction or similar */
+#define VM86_UNKNOWN	1	/* unhandled GP fault
+				   - IO-instruction or similar */
 #define VM86_INTx	2	/* int3/int x instruction (ARG = x) */
-#define VM86_STI	3	/* sti/popf/iret instruction enabled virtual interrupts */
+#define VM86_STI	3	/* sti/popf/iret instruction enabled
+				   virtual interrupts */
 
 /*
  * Additional return values when invoking new vm86()
@@ -205,7 +207,8 @@ void release_vm86_irqs(struct task_struct *);
 #define handle_vm86_fault(a, b)
 #define release_vm86_irqs(a)
 
-static inline int handle_vm86_trap(struct kernel_vm86_regs *a, long b, int c) {
+static inline int handle_vm86_trap(struct kernel_vm86_regs *a, long b, int c)
+{
 	return 0;
 }
 
-- 
cgit v1.2.3


From 8948584eb282c4dc5c54f6f6ebbaf447a665c653 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:03:59 -0700
Subject: include/asm-x86/vmi.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/vmi.h | 88 +++++++++++++++++++++++++--------------------------
 1 file changed, 44 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/vmi.h b/include/asm-x86/vmi.h
index eb8bd892c01e..b7c0dea119fe 100644
--- a/include/asm-x86/vmi.h
+++ b/include/asm-x86/vmi.h
@@ -155,9 +155,9 @@
 
 #ifndef __ASSEMBLY__
 struct vmi_relocation_info {
-        unsigned char           *eip;
-        unsigned char           type;
-        unsigned char           reserved[3];
+	unsigned char           *eip;
+	unsigned char           type;
+	unsigned char           reserved[3];
 };
 #endif
 
@@ -173,53 +173,53 @@ struct vmi_relocation_info {
 #ifndef __ASSEMBLY__
 
 struct vrom_header {
-	u16     rom_signature;  // option ROM signature
-	u8      rom_length;     // ROM length in 512 byte chunks
-	u8      rom_entry[4];   // 16-bit code entry point
-	u8      rom_pad0;       // 4-byte align pad
-	u32     vrom_signature; // VROM identification signature
-	u8      api_version_min;// Minor version of API
-	u8      api_version_maj;// Major version of API
-	u8      jump_slots;     // Number of jump slots
-	u8      reserved1;      // Reserved for expansion
-	u32     virtual_top;    // Hypervisor virtual address start
-	u16     reserved2;      // Reserved for expansion
-	u16	license_offs;	// Offset to License string
-	u16     pci_header_offs;// Offset to PCI OPROM header
-	u16     pnp_header_offs;// Offset to PnP OPROM header
-	u32     rom_pad3;       // PnP reserverd / VMI reserved
-	u8      reserved[96];   // Reserved for headers
-	char    vmi_init[8];    // VMI_Init jump point
-	char    get_reloc[8];   // VMI_GetRelocationInfo jump point
+	u16     rom_signature;  /* option ROM signature */
+	u8      rom_length;     /* ROM length in 512 byte chunks */
+	u8      rom_entry[4];   /* 16-bit code entry point */
+	u8      rom_pad0;       /* 4-byte align pad */
+	u32     vrom_signature; /* VROM identification signature */
+	u8      api_version_min;/* Minor version of API */
+	u8      api_version_maj;/* Major version of API */
+	u8      jump_slots;     /* Number of jump slots */
+	u8      reserved1;      /* Reserved for expansion */
+	u32     virtual_top;    /* Hypervisor virtual address start */
+	u16     reserved2;      /* Reserved for expansion */
+	u16	license_offs;	/* Offset to License string */
+	u16     pci_header_offs;/* Offset to PCI OPROM header */
+	u16     pnp_header_offs;/* Offset to PnP OPROM header */
+	u32     rom_pad3;       /* PnP reserverd / VMI reserved */
+	u8      reserved[96];   /* Reserved for headers */
+	char    vmi_init[8];    /* VMI_Init jump point */
+	char    get_reloc[8];   /* VMI_GetRelocationInfo jump point */
 } __attribute__((packed));
 
 struct pnp_header {
-        char sig[4];
-        char rev;
-        char size;
-        short next;
-        short res;
-        long devID;
-        unsigned short manufacturer_offset;
-        unsigned short product_offset;
+	char sig[4];
+	char rev;
+	char size;
+	short next;
+	short res;
+	long devID;
+	unsigned short manufacturer_offset;
+	unsigned short product_offset;
 } __attribute__((packed));
 
 struct pci_header {
-        char sig[4];
-        short vendorID;
-        short deviceID;
-        short vpdData;
-        short size;
-        char rev;
-        char class;
-        char subclass;
-        char interface;
-        short chunks;
-        char rom_version_min;
-        char rom_version_maj;
-        char codetype;
-        char lastRom;
-        short reserved;
+	char sig[4];
+	short vendorID;
+	short deviceID;
+	short vpdData;
+	short size;
+	char rev;
+	char class;
+	char subclass;
+	char interface;
+	short chunks;
+	char rom_version_min;
+	char rom_version_maj;
+	char codetype;
+	char lastRom;
+	short reserved;
 } __attribute__((packed));
 
 /* Function prototypes for bootstrapping */
-- 
cgit v1.2.3


From d6ae390a0be73e6b777c6171e6b6f616462f555d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:04:00 -0700
Subject: include/asm-x86/voyager.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/voyager.h | 51 ++++++++++++++++++++++++++++-------------------
 1 file changed, 31 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/voyager.h b/include/asm-x86/voyager.h
index 91a9932937ab..9c811d2e6f91 100644
--- a/include/asm-x86/voyager.h
+++ b/include/asm-x86/voyager.h
@@ -91,8 +91,7 @@
 #define VOYAGER_WRITE_CONFIG			0x2
 #define VOYAGER_BYPASS				0xff
 
-typedef struct voyager_asic 
-{
+typedef struct voyager_asic {
 	__u8	asic_addr;	/* ASIC address; Level 4 */
 	__u8	asic_type;      /* ASIC type */
 	__u8	asic_id;	/* ASIC id */
@@ -113,7 +112,7 @@ typedef struct voyager_module {
 	__u16   largest_reg;		/* Largest register in the scan path */
 	__u16   smallest_reg;		/* Smallest register in the scan path */
 	voyager_asic_t   *asic;		/* First ASIC in scan path (CAT_I) */
-	struct   voyager_module *submodule;	/* Submodule pointer */ 
+	struct   voyager_module *submodule;	/* Submodule pointer */
 	struct   voyager_module *next;		/* Next module in linked list */
 } voyager_module_t;
 
@@ -135,7 +134,7 @@ typedef struct voyager_eeprom_hdr {
 	 __u16 cct_offset;
 	 __u16 log_length;	/* length of err log */
 	 __u16 xsum_end;	/* offset to end of
-							   checksum */
+				   checksum */
 	 __u8  reserved[4];
 	 __u8  sflag;		/* starting sentinal */
 	 __u8  part_number[13];	/* prom part number */
@@ -148,7 +147,8 @@ typedef struct voyager_eeprom_hdr {
 
 
 
-#define VOYAGER_EPROM_SIZE_OFFSET   ((__u16)(&(((voyager_eprom_hdr_t *)0)->ee_size)))
+#define VOYAGER_EPROM_SIZE_OFFSET				\
+	((__u16)(&(((voyager_eprom_hdr_t *)0)->ee_size)))
 #define VOYAGER_XSUM_END_OFFSET		0x2a
 
 /* the following three definitions are for internal table layouts
@@ -199,7 +199,7 @@ typedef struct voyager_asic_data_table {
 #define VOYAGER_WCBIC_TOM_L	0x4
 #define VOYAGER_WCBIC_TOM_H	0x5
 
-/* register defines for Voyager Memory Contol (VMC) 
+/* register defines for Voyager Memory Contol (VMC)
  * these are present on L4 machines only */
 #define	VOYAGER_VMC1		0x81
 #define VOYAGER_VMC2		0x91
@@ -334,7 +334,7 @@ typedef struct {
 
 struct QuadDescription {
 	__u8  Type;	/* for type 0 (DYADIC or MONADIC) all fields
-                         * will be zero except for slot */
+			 * will be zero except for slot */
 	__u8 StructureVersion;
 	__u32 CPI_BaseAddress;
 	__u32  LARC_BankSize;
@@ -342,7 +342,7 @@ struct QuadDescription {
 	__u8  Slot; /* Processor slots 1 - 4 */
 } __attribute__((packed));
 
-struct ProcBoardInfo { 
+struct ProcBoardInfo {
 	__u8 Type;
 	__u8 StructureVersion;
 	__u8 NumberOfBoards;
@@ -382,19 +382,30 @@ struct CPU_Info {
  * packed in it by our friend the compiler.
  */
 typedef struct {
-	__u8	Mailbox_SUS;		/* Written to by SUS to give commands/response to the OS */
-	__u8	Mailbox_OS;		/* Written to by the OS to give commands/response to SUS */
-	__u8	SUS_MailboxVersion;	/* Tells the OS which iteration of the interface SUS supports */
-	__u8	OS_MailboxVersion;	/* Tells SUS which iteration of the interface the OS supports */
-	__u32	OS_Flags;		/* Flags set by the OS as info for SUS */
-	__u32	SUS_Flags;		/* Flags set by SUS as info for the OS */
-	__u32	WatchDogPeriod;		/* Watchdog period (in seconds) which the DP uses to see if the OS is dead */
+	__u8	Mailbox_SUS;		/* Written to by SUS to give
+					   commands/response to the OS */
+	__u8	Mailbox_OS;		/* Written to by the OS to give
+					   commands/response to SUS */
+	__u8	SUS_MailboxVersion;	/* Tells the OS which iteration of the
+					   interface SUS supports */
+	__u8	OS_MailboxVersion;	/* Tells SUS which iteration of the
+					   interface the OS supports */
+	__u32	OS_Flags;		/* Flags set by the OS as info for
+					   SUS */
+	__u32	SUS_Flags;		/* Flags set by SUS as info
+					   for the OS */
+	__u32	WatchDogPeriod;		/* Watchdog period (in seconds) which
+					   the DP uses to see if the OS
+					   is dead */
 	__u32	WatchDogCount;		/* Updated by the OS on every tic. */
-	__u32	MemoryFor_SUS_ErrorLog;	/* Flat 32 bit address which tells SUS where to stuff the SUS error log on a dump */
-	MC_SlotInformation_t  MC_SlotInfo[NUMBER_OF_MC_BUSSES*SLOTS_PER_MC_BUS];	/* Storage for MCA POS data */
+	__u32	MemoryFor_SUS_ErrorLog;	/* Flat 32 bit address which tells SUS
+					   where to stuff the SUS error log
+					   on a dump */
+	MC_SlotInformation_t MC_SlotInfo[NUMBER_OF_MC_BUSSES*SLOTS_PER_MC_BUS];
+					/* Storage for MCA POS data */
 	/* All new SECOND_PASS_INTERFACE fields added from this point */
-        struct ProcBoardInfo    *BoardData;
-        struct CPU_Info         *CPU_Data;
+	struct ProcBoardInfo    *BoardData;
+	struct CPU_Info         *CPU_Data;
 	/* All new fields must be added from this point */
 } Voyager_KernelSUS_Mbox_t;
 
@@ -478,7 +489,7 @@ struct voyager_SUS {
 	__u32	SUS_errorlog;
 	/* lots of system configuration stuff under here */
 };
-	
+
 /* Variables exported by voyager_smp */
 extern __u32 voyager_extended_vic_processors;
 extern __u32 voyager_allowed_boot_processors;
-- 
cgit v1.2.3


From 8fdf765527920e30d8fd57da3a83d6bf56799114 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:04:02 -0700
Subject: include/asm-x86/xor_32.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/xor_32.h | 494 ++++++++++++++++++++++++-----------------------
 1 file changed, 248 insertions(+), 246 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/xor_32.h b/include/asm-x86/xor_32.h
index a41ef1bdd424..067b5c1835a3 100644
--- a/include/asm-x86/xor_32.h
+++ b/include/asm-x86/xor_32.h
@@ -16,12 +16,12 @@
  * Copyright (C) 1998 Ingo Molnar.
  */
 
-#define LD(x,y)		"       movq   8*("#x")(%1), %%mm"#y"   ;\n"
-#define ST(x,y)		"       movq %%mm"#y",   8*("#x")(%1)   ;\n"
-#define XO1(x,y)	"       pxor   8*("#x")(%2), %%mm"#y"   ;\n"
-#define XO2(x,y)	"       pxor   8*("#x")(%3), %%mm"#y"   ;\n"
-#define XO3(x,y)	"       pxor   8*("#x")(%4), %%mm"#y"   ;\n"
-#define XO4(x,y)	"       pxor   8*("#x")(%5), %%mm"#y"   ;\n"
+#define LD(x, y)	"       movq   8*("#x")(%1), %%mm"#y"   ;\n"
+#define ST(x, y)	"       movq %%mm"#y",   8*("#x")(%1)   ;\n"
+#define XO1(x, y)	"       pxor   8*("#x")(%2), %%mm"#y"   ;\n"
+#define XO2(x, y)	"       pxor   8*("#x")(%3), %%mm"#y"   ;\n"
+#define XO3(x, y)	"       pxor   8*("#x")(%4), %%mm"#y"   ;\n"
+#define XO4(x, y)	"       pxor   8*("#x")(%5), %%mm"#y"   ;\n"
 
 #include <asm/i387.h>
 
@@ -32,24 +32,24 @@ xor_pII_mmx_2(unsigned long bytes, unsigned long *p1, unsigned long *p2)
 
 	kernel_fpu_begin();
 
-	__asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
-#define BLOCK(i) \
-	LD(i,0)					\
-		LD(i+1,1)			\
-			LD(i+2,2)		\
-				LD(i+3,3)	\
-	XO1(i,0)				\
-	ST(i,0)					\
-		XO1(i+1,1)			\
-		ST(i+1,1)			\
-			XO1(i+2,2)		\
-			ST(i+2,2)		\
-				XO1(i+3,3)	\
-				ST(i+3,3)
+#define BLOCK(i)				\
+	LD(i, 0)				\
+		LD(i + 1, 1)			\
+			LD(i + 2, 2)		\
+				LD(i + 3, 3)	\
+	XO1(i, 0)				\
+	ST(i, 0)				\
+		XO1(i+1, 1)			\
+		ST(i+1, 1)			\
+			XO1(i + 2, 2)		\
+			ST(i + 2, 2)		\
+				XO1(i + 3, 3)	\
+				ST(i + 3, 3)
 
 	" .align 32			;\n"
-  	" 1:                            ;\n"
+	" 1:                            ;\n"
 
 	BLOCK(0)
 	BLOCK(4)
@@ -76,25 +76,25 @@ xor_pII_mmx_3(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 
 	kernel_fpu_begin();
 
-	__asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
-#define BLOCK(i) \
-	LD(i,0)					\
-		LD(i+1,1)			\
-			LD(i+2,2)		\
-				LD(i+3,3)	\
-	XO1(i,0)				\
-		XO1(i+1,1)			\
-			XO1(i+2,2)		\
-				XO1(i+3,3)	\
-	XO2(i,0)				\
-	ST(i,0)					\
-		XO2(i+1,1)			\
-		ST(i+1,1)			\
-			XO2(i+2,2)		\
-			ST(i+2,2)		\
-				XO2(i+3,3)	\
-				ST(i+3,3)
+#define BLOCK(i)				\
+	LD(i, 0)				\
+		LD(i + 1, 1)			\
+			LD(i + 2, 2)		\
+				LD(i + 3, 3)	\
+	XO1(i, 0)				\
+		XO1(i + 1, 1)			\
+			XO1(i + 2, 2)		\
+				XO1(i + 3, 3)	\
+	XO2(i, 0)				\
+	ST(i, 0)				\
+		XO2(i + 1, 1)			\
+		ST(i + 1, 1)			\
+			XO2(i + 2, 2)		\
+			ST(i + 2, 2)		\
+				XO2(i + 3, 3)	\
+				ST(i + 3, 3)
 
 	" .align 32			;\n"
 	" 1:                            ;\n"
@@ -125,29 +125,29 @@ xor_pII_mmx_4(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 
 	kernel_fpu_begin();
 
-	__asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
-#define BLOCK(i) \
-	LD(i,0)					\
-		LD(i+1,1)			\
-			LD(i+2,2)		\
-				LD(i+3,3)	\
-	XO1(i,0)				\
-		XO1(i+1,1)			\
-			XO1(i+2,2)		\
-				XO1(i+3,3)	\
-	XO2(i,0)				\
-		XO2(i+1,1)			\
-			XO2(i+2,2)		\
-				XO2(i+3,3)	\
-	XO3(i,0)				\
-	ST(i,0)					\
-		XO3(i+1,1)			\
-		ST(i+1,1)			\
-			XO3(i+2,2)		\
-			ST(i+2,2)		\
-				XO3(i+3,3)	\
-				ST(i+3,3)
+#define BLOCK(i)				\
+	LD(i, 0)				\
+		LD(i + 1, 1)			\
+			LD(i + 2, 2)		\
+				LD(i + 3, 3)	\
+	XO1(i, 0)				\
+		XO1(i + 1, 1)			\
+			XO1(i + 2, 2)		\
+				XO1(i + 3, 3)	\
+	XO2(i, 0)				\
+		XO2(i + 1, 1)			\
+			XO2(i + 2, 2)		\
+				XO2(i + 3, 3)	\
+	XO3(i, 0)				\
+	ST(i, 0)				\
+		XO3(i + 1, 1)			\
+		ST(i + 1, 1)			\
+			XO3(i + 2, 2)		\
+			ST(i + 2, 2)		\
+				XO3(i + 3, 3)	\
+				ST(i + 3, 3)
 
 	" .align 32			;\n"
 	" 1:                            ;\n"
@@ -186,35 +186,35 @@ xor_pII_mmx_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	   because we modify p4 and p5 there, but we can't mark them
 	   as read/write, otherwise we'd overflow the 10-asm-operands
 	   limit of GCC < 3.1.  */
-	__asm__ ("" : "+r" (p4), "+r" (p5));
+	asm("" : "+r" (p4), "+r" (p5));
 
-	__asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
-#define BLOCK(i) \
-	LD(i,0)					\
-		LD(i+1,1)			\
-			LD(i+2,2)		\
-				LD(i+3,3)	\
-	XO1(i,0)				\
-		XO1(i+1,1)			\
-			XO1(i+2,2)		\
-				XO1(i+3,3)	\
-	XO2(i,0)				\
-		XO2(i+1,1)			\
-			XO2(i+2,2)		\
-				XO2(i+3,3)	\
-	XO3(i,0)				\
-		XO3(i+1,1)			\
-			XO3(i+2,2)		\
-				XO3(i+3,3)	\
-	XO4(i,0)				\
-	ST(i,0)					\
-		XO4(i+1,1)			\
-		ST(i+1,1)			\
-			XO4(i+2,2)		\
-			ST(i+2,2)		\
-				XO4(i+3,3)	\
-				ST(i+3,3)
+#define BLOCK(i)				\
+	LD(i, 0)				\
+		LD(i + 1, 1)			\
+			LD(i + 2, 2)		\
+				LD(i + 3, 3)	\
+	XO1(i, 0)				\
+		XO1(i + 1, 1)			\
+			XO1(i + 2, 2)		\
+				XO1(i + 3, 3)	\
+	XO2(i, 0)				\
+		XO2(i + 1, 1)			\
+			XO2(i + 2, 2)		\
+				XO2(i + 3, 3)	\
+	XO3(i, 0)				\
+		XO3(i + 1, 1)			\
+			XO3(i + 2, 2)		\
+				XO3(i + 3, 3)	\
+	XO4(i, 0)				\
+	ST(i, 0)				\
+		XO4(i + 1, 1)			\
+		ST(i + 1, 1)			\
+			XO4(i + 2, 2)		\
+			ST(i + 2, 2)		\
+				XO4(i + 3, 3)	\
+				ST(i + 3, 3)
 
 	" .align 32			;\n"
 	" 1:                            ;\n"
@@ -233,13 +233,13 @@ xor_pII_mmx_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	"       jnz 1b                ;\n"
 	: "+r" (lines),
 	  "+r" (p1), "+r" (p2), "+r" (p3)
-	: "r" (p4), "r" (p5) 
+	: "r" (p4), "r" (p5)
 	: "memory");
 
 	/* p4 and p5 were modified, and now the variables are dead.
 	   Clobber them just to be sure nobody does something stupid
 	   like assuming they have some legal value.  */
-	__asm__ ("" : "=r" (p4), "=r" (p5));
+	asm("" : "=r" (p4), "=r" (p5));
 
 	kernel_fpu_end();
 }
@@ -259,7 +259,7 @@ xor_p5_mmx_2(unsigned long bytes, unsigned long *p1, unsigned long *p2)
 
 	kernel_fpu_begin();
 
-	__asm__ __volatile__ (
+	asm volatile(
 	" .align 32	             ;\n"
 	" 1:                         ;\n"
 	"       movq   (%1), %%mm0   ;\n"
@@ -286,7 +286,7 @@ xor_p5_mmx_2(unsigned long bytes, unsigned long *p1, unsigned long *p2)
 	"       pxor 56(%2), %%mm7   ;\n"
 	"       movq %%mm6, 48(%1)   ;\n"
 	"       movq %%mm7, 56(%1)   ;\n"
-	
+
 	"       addl $64, %1         ;\n"
 	"       addl $64, %2         ;\n"
 	"       decl %0              ;\n"
@@ -307,7 +307,7 @@ xor_p5_mmx_3(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 
 	kernel_fpu_begin();
 
-	__asm__ __volatile__ (
+	asm volatile(
 	" .align 32,0x90             ;\n"
 	" 1:                         ;\n"
 	"       movq   (%1), %%mm0   ;\n"
@@ -342,7 +342,7 @@ xor_p5_mmx_3(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	"       pxor 56(%3), %%mm7   ;\n"
 	"       movq %%mm6, 48(%1)   ;\n"
 	"       movq %%mm7, 56(%1)   ;\n"
-      
+
 	"       addl $64, %1         ;\n"
 	"       addl $64, %2         ;\n"
 	"       addl $64, %3         ;\n"
@@ -364,7 +364,7 @@ xor_p5_mmx_4(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 
 	kernel_fpu_begin();
 
-	__asm__ __volatile__ (
+	asm volatile(
 	" .align 32,0x90             ;\n"
 	" 1:                         ;\n"
 	"       movq   (%1), %%mm0   ;\n"
@@ -407,7 +407,7 @@ xor_p5_mmx_4(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	"       pxor 56(%4), %%mm7   ;\n"
 	"       movq %%mm6, 48(%1)   ;\n"
 	"       movq %%mm7, 56(%1)   ;\n"
-      
+
 	"       addl $64, %1         ;\n"
 	"       addl $64, %2         ;\n"
 	"       addl $64, %3         ;\n"
@@ -436,9 +436,9 @@ xor_p5_mmx_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	   because we modify p4 and p5 there, but we can't mark them
 	   as read/write, otherwise we'd overflow the 10-asm-operands
 	   limit of GCC < 3.1.  */
-	__asm__ ("" : "+r" (p4), "+r" (p5));
+	asm("" : "+r" (p4), "+r" (p5));
 
-	__asm__ __volatile__ (
+	asm volatile(
 	" .align 32,0x90             ;\n"
 	" 1:                         ;\n"
 	"       movq   (%1), %%mm0   ;\n"
@@ -489,7 +489,7 @@ xor_p5_mmx_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	"       pxor 56(%5), %%mm7   ;\n"
 	"       movq %%mm6, 48(%1)   ;\n"
 	"       movq %%mm7, 56(%1)   ;\n"
-      
+
 	"       addl $64, %1         ;\n"
 	"       addl $64, %2         ;\n"
 	"       addl $64, %3         ;\n"
@@ -505,7 +505,7 @@ xor_p5_mmx_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	/* p4 and p5 were modified, and now the variables are dead.
 	   Clobber them just to be sure nobody does something stupid
 	   like assuming they have some legal value.  */
-	__asm__ ("" : "=r" (p4), "=r" (p5));
+	asm("" : "=r" (p4), "=r" (p5));
 
 	kernel_fpu_end();
 }
@@ -531,11 +531,12 @@ static struct xor_block_template xor_block_p5_mmx = {
  * Copyright (C) 1999 Zach Brown (with obvious credit due Ingo)
  */
 
-#define XMMS_SAVE do {				\
+#define XMMS_SAVE				\
+do {						\
 	preempt_disable();			\
 	cr0 = read_cr0();			\
 	clts();					\
-	__asm__ __volatile__ ( 			\
+	asm volatile(				\
 		"movups %%xmm0,(%0)	;\n\t"	\
 		"movups %%xmm1,0x10(%0)	;\n\t"	\
 		"movups %%xmm2,0x20(%0)	;\n\t"	\
@@ -543,10 +544,11 @@ static struct xor_block_template xor_block_p5_mmx = {
 		:				\
 		: "r" (xmm_save) 		\
 		: "memory");			\
-} while(0)
+} while (0)
 
-#define XMMS_RESTORE do {			\
-	__asm__ __volatile__ ( 			\
+#define XMMS_RESTORE				\
+do {						\
+	asm volatile(				\
 		"sfence			;\n\t"	\
 		"movups (%0),%%xmm0	;\n\t"	\
 		"movups 0x10(%0),%%xmm1	;\n\t"	\
@@ -557,76 +559,76 @@ static struct xor_block_template xor_block_p5_mmx = {
 		: "memory");			\
 	write_cr0(cr0);				\
 	preempt_enable();			\
-} while(0)
+} while (0)
 
 #define ALIGN16 __attribute__((aligned(16)))
 
 #define OFFS(x)		"16*("#x")"
 #define PF_OFFS(x)	"256+16*("#x")"
 #define	PF0(x)		"	prefetchnta "PF_OFFS(x)"(%1)		;\n"
-#define LD(x,y)		"       movaps   "OFFS(x)"(%1), %%xmm"#y"	;\n"
-#define ST(x,y)		"       movaps %%xmm"#y",   "OFFS(x)"(%1)	;\n"
+#define LD(x, y)	"       movaps   "OFFS(x)"(%1), %%xmm"#y"	;\n"
+#define ST(x, y)	"       movaps %%xmm"#y",   "OFFS(x)"(%1)	;\n"
 #define PF1(x)		"	prefetchnta "PF_OFFS(x)"(%2)		;\n"
 #define PF2(x)		"	prefetchnta "PF_OFFS(x)"(%3)		;\n"
 #define PF3(x)		"	prefetchnta "PF_OFFS(x)"(%4)		;\n"
 #define PF4(x)		"	prefetchnta "PF_OFFS(x)"(%5)		;\n"
 #define PF5(x)		"	prefetchnta "PF_OFFS(x)"(%6)		;\n"
-#define XO1(x,y)	"       xorps   "OFFS(x)"(%2), %%xmm"#y"	;\n"
-#define XO2(x,y)	"       xorps   "OFFS(x)"(%3), %%xmm"#y"	;\n"
-#define XO3(x,y)	"       xorps   "OFFS(x)"(%4), %%xmm"#y"	;\n"
-#define XO4(x,y)	"       xorps   "OFFS(x)"(%5), %%xmm"#y"	;\n"
-#define XO5(x,y)	"       xorps   "OFFS(x)"(%6), %%xmm"#y"	;\n"
+#define XO1(x, y)	"       xorps   "OFFS(x)"(%2), %%xmm"#y"	;\n"
+#define XO2(x, y)	"       xorps   "OFFS(x)"(%3), %%xmm"#y"	;\n"
+#define XO3(x, y)	"       xorps   "OFFS(x)"(%4), %%xmm"#y"	;\n"
+#define XO4(x, y)	"       xorps   "OFFS(x)"(%5), %%xmm"#y"	;\n"
+#define XO5(x, y)	"       xorps   "OFFS(x)"(%6), %%xmm"#y"	;\n"
 
 
 static void
 xor_sse_2(unsigned long bytes, unsigned long *p1, unsigned long *p2)
 {
-        unsigned long lines = bytes >> 8;
+	unsigned long lines = bytes >> 8;
 	char xmm_save[16*4] ALIGN16;
 	int cr0;
 
 	XMMS_SAVE;
 
-        __asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
-#define BLOCK(i) \
-		LD(i,0)					\
-			LD(i+1,1)			\
+#define BLOCK(i)					\
+		LD(i, 0)				\
+			LD(i + 1, 1)			\
 		PF1(i)					\
-				PF1(i+2)		\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
-		PF0(i+4)				\
-				PF0(i+6)		\
-		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
-		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+				PF1(i + 2)		\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
+		XO1(i, 0)				\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
+		ST(i, 0)				\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addl $256, %1           ;\n"
-        "       addl $256, %2           ;\n"
-        "       decl %0                 ;\n"
-        "       jnz 1b                  ;\n"
+	"       addl $256, %1           ;\n"
+	"       addl $256, %2           ;\n"
+	"       decl %0                 ;\n"
+	"       jnz 1b                  ;\n"
 	: "+r" (lines),
 	  "+r" (p1), "+r" (p2)
 	:
-        : "memory");
+	: "memory");
 
 	XMMS_RESTORE;
 }
@@ -635,59 +637,59 @@ static void
 xor_sse_3(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	  unsigned long *p3)
 {
-        unsigned long lines = bytes >> 8;
+	unsigned long lines = bytes >> 8;
 	char xmm_save[16*4] ALIGN16;
 	int cr0;
 
 	XMMS_SAVE;
 
-        __asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
 #define BLOCK(i) \
 		PF1(i)					\
-				PF1(i+2)		\
+				PF1(i + 2)		\
 		LD(i,0)					\
-			LD(i+1,1)			\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
+			LD(i + 1, 1)			\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
 		PF2(i)					\
-				PF2(i+2)		\
-		PF0(i+4)				\
-				PF0(i+6)		\
+				PF2(i + 2)		\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
 		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
 		XO2(i,0)				\
-			XO2(i+1,1)			\
-				XO2(i+2,2)		\
-					XO2(i+3,3)	\
+			XO2(i + 1, 1)			\
+				XO2(i + 2, 2)		\
+					XO2(i + 3, 3)	\
 		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addl $256, %1           ;\n"
-        "       addl $256, %2           ;\n"
-        "       addl $256, %3           ;\n"
-        "       decl %0                 ;\n"
-        "       jnz 1b                  ;\n"
+	"       addl $256, %1           ;\n"
+	"       addl $256, %2           ;\n"
+	"       addl $256, %3           ;\n"
+	"       decl %0                 ;\n"
+	"       jnz 1b                  ;\n"
 	: "+r" (lines),
 	  "+r" (p1), "+r"(p2), "+r"(p3)
 	:
-        : "memory" );
+	: "memory" );
 
 	XMMS_RESTORE;
 }
@@ -696,66 +698,66 @@ static void
 xor_sse_4(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	  unsigned long *p3, unsigned long *p4)
 {
-        unsigned long lines = bytes >> 8;
+	unsigned long lines = bytes >> 8;
 	char xmm_save[16*4] ALIGN16;
 	int cr0;
 
 	XMMS_SAVE;
 
-        __asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
 #define BLOCK(i) \
 		PF1(i)					\
-				PF1(i+2)		\
+				PF1(i + 2)		\
 		LD(i,0)					\
-			LD(i+1,1)			\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
+			LD(i + 1, 1)			\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
 		PF2(i)					\
-				PF2(i+2)		\
+				PF2(i + 2)		\
 		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
 		PF3(i)					\
-				PF3(i+2)		\
-		PF0(i+4)				\
-				PF0(i+6)		\
+				PF3(i + 2)		\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
 		XO2(i,0)				\
-			XO2(i+1,1)			\
-				XO2(i+2,2)		\
-					XO2(i+3,3)	\
+			XO2(i + 1, 1)			\
+				XO2(i + 2, 2)		\
+					XO2(i + 3, 3)	\
 		XO3(i,0)				\
-			XO3(i+1,1)			\
-				XO3(i+2,2)		\
-					XO3(i+3,3)	\
+			XO3(i + 1, 1)			\
+				XO3(i + 2, 2)		\
+					XO3(i + 3, 3)	\
 		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addl $256, %1           ;\n"
-        "       addl $256, %2           ;\n"
-        "       addl $256, %3           ;\n"
-        "       addl $256, %4           ;\n"
-        "       decl %0                 ;\n"
-        "       jnz 1b                  ;\n"
+	"       addl $256, %1           ;\n"
+	"       addl $256, %2           ;\n"
+	"       addl $256, %3           ;\n"
+	"       addl $256, %4           ;\n"
+	"       decl %0                 ;\n"
+	"       jnz 1b                  ;\n"
 	: "+r" (lines),
 	  "+r" (p1), "+r" (p2), "+r" (p3), "+r" (p4)
 	:
-        : "memory" );
+	: "memory" );
 
 	XMMS_RESTORE;
 }
@@ -764,7 +766,7 @@ static void
 xor_sse_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	  unsigned long *p3, unsigned long *p4, unsigned long *p5)
 {
-        unsigned long lines = bytes >> 8;
+	unsigned long lines = bytes >> 8;
 	char xmm_save[16*4] ALIGN16;
 	int cr0;
 
@@ -776,65 +778,65 @@ xor_sse_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	   because we modify p4 and p5 there, but we can't mark them
 	   as read/write, otherwise we'd overflow the 10-asm-operands
 	   limit of GCC < 3.1.  */
-	__asm__ ("" : "+r" (p4), "+r" (p5));
+	asm("" : "+r" (p4), "+r" (p5));
 
-        __asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
 #define BLOCK(i) \
 		PF1(i)					\
-				PF1(i+2)		\
+				PF1(i + 2)		\
 		LD(i,0)					\
-			LD(i+1,1)			\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
+			LD(i + 1, 1)			\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
 		PF2(i)					\
-				PF2(i+2)		\
+				PF2(i + 2)		\
 		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
 		PF3(i)					\
-				PF3(i+2)		\
+				PF3(i + 2)		\
 		XO2(i,0)				\
-			XO2(i+1,1)			\
-				XO2(i+2,2)		\
-					XO2(i+3,3)	\
+			XO2(i + 1, 1)			\
+				XO2(i + 2, 2)		\
+					XO2(i + 3, 3)	\
 		PF4(i)					\
-				PF4(i+2)		\
-		PF0(i+4)				\
-				PF0(i+6)		\
+				PF4(i + 2)		\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
 		XO3(i,0)				\
-			XO3(i+1,1)			\
-				XO3(i+2,2)		\
-					XO3(i+3,3)	\
+			XO3(i + 1, 1)			\
+				XO3(i + 2, 2)		\
+					XO3(i + 3, 3)	\
 		XO4(i,0)				\
-			XO4(i+1,1)			\
-				XO4(i+2,2)		\
-					XO4(i+3,3)	\
+			XO4(i + 1, 1)			\
+				XO4(i + 2, 2)		\
+					XO4(i + 3, 3)	\
 		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addl $256, %1           ;\n"
-        "       addl $256, %2           ;\n"
-        "       addl $256, %3           ;\n"
-        "       addl $256, %4           ;\n"
-        "       addl $256, %5           ;\n"
-        "       decl %0                 ;\n"
-        "       jnz 1b                  ;\n"
+	"       addl $256, %1           ;\n"
+	"       addl $256, %2           ;\n"
+	"       addl $256, %3           ;\n"
+	"       addl $256, %4           ;\n"
+	"       addl $256, %5           ;\n"
+	"       decl %0                 ;\n"
+	"       jnz 1b                  ;\n"
 	: "+r" (lines),
 	  "+r" (p1), "+r" (p2), "+r" (p3)
 	: "r" (p4), "r" (p5)
@@ -843,17 +845,17 @@ xor_sse_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	/* p4 and p5 were modified, and now the variables are dead.
 	   Clobber them just to be sure nobody does something stupid
 	   like assuming they have some legal value.  */
-	__asm__ ("" : "=r" (p4), "=r" (p5));
+	asm("" : "=r" (p4), "=r" (p5));
 
 	XMMS_RESTORE;
 }
 
 static struct xor_block_template xor_block_pIII_sse = {
-        .name = "pIII_sse",
-        .do_2 =  xor_sse_2,
-        .do_3 =  xor_sse_3,
-        .do_4 =  xor_sse_4,
-        .do_5 = xor_sse_5,
+	.name = "pIII_sse",
+	.do_2 = xor_sse_2,
+	.do_3 = xor_sse_3,
+	.do_4 = xor_sse_4,
+	.do_5 = xor_sse_5,
 };
 
 /* Also try the generic routines.  */
@@ -861,21 +863,21 @@ static struct xor_block_template xor_block_pIII_sse = {
 
 #undef XOR_TRY_TEMPLATES
 #define XOR_TRY_TEMPLATES				\
-	do {						\
-		xor_speed(&xor_block_8regs);		\
-		xor_speed(&xor_block_8regs_p);		\
-		xor_speed(&xor_block_32regs);		\
-		xor_speed(&xor_block_32regs_p);		\
-	        if (cpu_has_xmm)			\
-			xor_speed(&xor_block_pIII_sse);	\
-	        if (cpu_has_mmx) {			\
-	                xor_speed(&xor_block_pII_mmx);	\
-	                xor_speed(&xor_block_p5_mmx);	\
-	        }					\
-	} while (0)
+do {							\
+	xor_speed(&xor_block_8regs);			\
+	xor_speed(&xor_block_8regs_p);			\
+	xor_speed(&xor_block_32regs);			\
+	xor_speed(&xor_block_32regs_p);			\
+	if (cpu_has_xmm)				\
+		xor_speed(&xor_block_pIII_sse);		\
+	if (cpu_has_mmx) {				\
+		xor_speed(&xor_block_pII_mmx);		\
+		xor_speed(&xor_block_p5_mmx);		\
+	}						\
+} while (0)
 
 /* We force the use of the SSE xor block because it can write around L2.
    We may also be able to load into the L1 only depending on how the cpu
    deals with a load to a line that is being prefetched.  */
-#define XOR_SELECT_TEMPLATE(FASTEST) \
+#define XOR_SELECT_TEMPLATE(FASTEST)			\
 	(cpu_has_xmm ? &xor_block_pIII_sse : FASTEST)
-- 
cgit v1.2.3


From 687c80540975709844699d9a96070e2cf3bad39d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 23 Mar 2008 01:04:03 -0700
Subject: include/asm-x86/xor_64.h: checkpatch cleanups - formatting only

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/xor_64.h | 294 ++++++++++++++++++++++++-----------------------
 1 file changed, 149 insertions(+), 145 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/xor_64.h b/include/asm-x86/xor_64.h
index 1eee7fcb2420..24957e39ac8a 100644
--- a/include/asm-x86/xor_64.h
+++ b/include/asm-x86/xor_64.h
@@ -24,20 +24,23 @@
  */
 
 /*
- * x86-64 changes / gcc fixes from Andi Kleen. 
+ * x86-64 changes / gcc fixes from Andi Kleen.
  * Copyright 2002 Andi Kleen, SuSE Labs.
  *
  * This hasn't been optimized for the hammer yet, but there are likely
  * no advantages to be gotten from x86-64 here anyways.
  */
 
-typedef struct { unsigned long a,b; } __attribute__((aligned(16))) xmm_store_t;
+typedef struct {
+	unsigned long a, b;
+} __attribute__((aligned(16))) xmm_store_t;
 
-/* Doesn't use gcc to save the XMM registers, because there is no easy way to 
+/* Doesn't use gcc to save the XMM registers, because there is no easy way to
    tell it to do a clts before the register saving. */
-#define XMMS_SAVE do {				\
+#define XMMS_SAVE				\
+do {						\
 	preempt_disable();			\
-	asm volatile (				\
+	asm volatile(				\
 		"movq %%cr0,%0		;\n\t"	\
 		"clts			;\n\t"	\
 		"movups %%xmm0,(%1)	;\n\t"	\
@@ -47,10 +50,11 @@ typedef struct { unsigned long a,b; } __attribute__((aligned(16))) xmm_store_t;
 		: "=&r" (cr0)			\
 		: "r" (xmm_save) 		\
 		: "memory");			\
-} while(0)
+} while (0)
 
-#define XMMS_RESTORE do {			\
-	asm volatile (				\
+#define XMMS_RESTORE				\
+do {						\
+	asm volatile(				\
 		"sfence			;\n\t"	\
 		"movups (%1),%%xmm0	;\n\t"	\
 		"movups 0x10(%1),%%xmm1	;\n\t"	\
@@ -61,72 +65,72 @@ typedef struct { unsigned long a,b; } __attribute__((aligned(16))) xmm_store_t;
 		: "r" (cr0), "r" (xmm_save)	\
 		: "memory");			\
 	preempt_enable();			\
-} while(0)
+} while (0)
 
 #define OFFS(x)		"16*("#x")"
 #define PF_OFFS(x)	"256+16*("#x")"
 #define	PF0(x)		"	prefetchnta "PF_OFFS(x)"(%[p1])		;\n"
-#define LD(x,y)		"       movaps   "OFFS(x)"(%[p1]), %%xmm"#y"	;\n"
-#define ST(x,y)		"       movaps %%xmm"#y",   "OFFS(x)"(%[p1])	;\n"
+#define LD(x, y)	"       movaps   "OFFS(x)"(%[p1]), %%xmm"#y"	;\n"
+#define ST(x, y)	"       movaps %%xmm"#y",   "OFFS(x)"(%[p1])	;\n"
 #define PF1(x)		"	prefetchnta "PF_OFFS(x)"(%[p2])		;\n"
 #define PF2(x)		"	prefetchnta "PF_OFFS(x)"(%[p3])		;\n"
 #define PF3(x)		"	prefetchnta "PF_OFFS(x)"(%[p4])		;\n"
 #define PF4(x)		"	prefetchnta "PF_OFFS(x)"(%[p5])		;\n"
 #define PF5(x)		"	prefetchnta "PF_OFFS(x)"(%[p6])		;\n"
-#define XO1(x,y)	"       xorps   "OFFS(x)"(%[p2]), %%xmm"#y"	;\n"
-#define XO2(x,y)	"       xorps   "OFFS(x)"(%[p3]), %%xmm"#y"	;\n"
-#define XO3(x,y)	"       xorps   "OFFS(x)"(%[p4]), %%xmm"#y"	;\n"
-#define XO4(x,y)	"       xorps   "OFFS(x)"(%[p5]), %%xmm"#y"	;\n"
-#define XO5(x,y)	"       xorps   "OFFS(x)"(%[p6]), %%xmm"#y"	;\n"
+#define XO1(x, y)	"       xorps   "OFFS(x)"(%[p2]), %%xmm"#y"	;\n"
+#define XO2(x, y)	"       xorps   "OFFS(x)"(%[p3]), %%xmm"#y"	;\n"
+#define XO3(x, y)	"       xorps   "OFFS(x)"(%[p4]), %%xmm"#y"	;\n"
+#define XO4(x, y)	"       xorps   "OFFS(x)"(%[p5]), %%xmm"#y"	;\n"
+#define XO5(x, y)	"       xorps   "OFFS(x)"(%[p6]), %%xmm"#y"	;\n"
 
 
 static void
 xor_sse_2(unsigned long bytes, unsigned long *p1, unsigned long *p2)
 {
-        unsigned int lines = bytes >> 8;
+	unsigned int lines = bytes >> 8;
 	unsigned long cr0;
 	xmm_store_t xmm_save[4];
 
 	XMMS_SAVE;
 
-        asm volatile (
+	asm volatile(
 #undef BLOCK
 #define BLOCK(i) \
-		LD(i,0)					\
-			LD(i+1,1)			\
+		LD(i, 0)				\
+			LD(i + 1, 1)			\
 		PF1(i)					\
-				PF1(i+2)		\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
-		PF0(i+4)				\
-				PF0(i+6)		\
-		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
-		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+				PF1(i + 2)		\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
+		XO1(i, 0)				\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
+		ST(i, 0)				\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addq %[inc], %[p1]           ;\n"
-        "       addq %[inc], %[p2]           ;\n"
+	"       addq %[inc], %[p1]           ;\n"
+	"       addq %[inc], %[p2]           ;\n"
 		"		decl %[cnt] ; jnz 1b"
 	: [p1] "+r" (p1), [p2] "+r" (p2), [cnt] "+r" (lines)
-	: [inc] "r" (256UL) 
-        : "memory");
+	: [inc] "r" (256UL)
+	: "memory");
 
 	XMMS_RESTORE;
 }
@@ -141,52 +145,52 @@ xor_sse_3(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 
 	XMMS_SAVE;
 
-        __asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
 #define BLOCK(i) \
 		PF1(i)					\
-				PF1(i+2)		\
-		LD(i,0)					\
-			LD(i+1,1)			\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
+				PF1(i + 2)		\
+		LD(i, 0)					\
+			LD(i + 1, 1)			\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
 		PF2(i)					\
-				PF2(i+2)		\
-		PF0(i+4)				\
-				PF0(i+6)		\
-		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
-		XO2(i,0)				\
-			XO2(i+1,1)			\
-				XO2(i+2,2)		\
-					XO2(i+3,3)	\
-		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+				PF2(i + 2)		\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
+		XO1(i, 0)				\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
+		XO2(i, 0)				\
+			XO2(i + 1, 1)			\
+				XO2(i + 2, 2)		\
+					XO2(i + 3, 3)	\
+		ST(i, 0)				\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addq %[inc], %[p1]           ;\n"
-        "       addq %[inc], %[p2]          ;\n"
-        "       addq %[inc], %[p3]           ;\n"
+	"       addq %[inc], %[p1]           ;\n"
+	"       addq %[inc], %[p2]          ;\n"
+	"       addq %[inc], %[p3]           ;\n"
 		"		decl %[cnt] ; jnz 1b"
 	: [cnt] "+r" (lines),
 	  [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3)
 	: [inc] "r" (256UL)
-	: "memory"); 
+	: "memory");
 	XMMS_RESTORE;
 }
 
@@ -195,64 +199,64 @@ xor_sse_4(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	  unsigned long *p3, unsigned long *p4)
 {
 	unsigned int lines = bytes >> 8;
-	xmm_store_t xmm_save[4]; 
+	xmm_store_t xmm_save[4];
 	unsigned long cr0;
 
 	XMMS_SAVE;
 
-        __asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
 #define BLOCK(i) \
 		PF1(i)					\
-				PF1(i+2)		\
-		LD(i,0)					\
-			LD(i+1,1)			\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
+				PF1(i + 2)		\
+		LD(i, 0)				\
+			LD(i + 1, 1)			\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
 		PF2(i)					\
-				PF2(i+2)		\
-		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
+				PF2(i + 2)		\
+		XO1(i, 0)				\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
 		PF3(i)					\
-				PF3(i+2)		\
-		PF0(i+4)				\
-				PF0(i+6)		\
-		XO2(i,0)				\
-			XO2(i+1,1)			\
-				XO2(i+2,2)		\
-					XO2(i+3,3)	\
-		XO3(i,0)				\
-			XO3(i+1,1)			\
-				XO3(i+2,2)		\
-					XO3(i+3,3)	\
-		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+				PF3(i + 2)		\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
+		XO2(i, 0)				\
+			XO2(i + 1, 1)			\
+				XO2(i + 2, 2)		\
+					XO2(i + 3, 3)	\
+		XO3(i, 0)				\
+			XO3(i + 1, 1)			\
+				XO3(i + 2, 2)		\
+					XO3(i + 3, 3)	\
+		ST(i, 0)				\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addq %[inc], %[p1]           ;\n"
-        "       addq %[inc], %[p2]           ;\n"
-        "       addq %[inc], %[p3]           ;\n"
-        "       addq %[inc], %[p4]           ;\n"
+	"       addq %[inc], %[p1]           ;\n"
+	"       addq %[inc], %[p2]           ;\n"
+	"       addq %[inc], %[p3]           ;\n"
+	"       addq %[inc], %[p4]           ;\n"
 	"	decl %[cnt] ; jnz 1b"
 	: [cnt] "+c" (lines),
 	  [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), [p4] "+r" (p4)
 	: [inc] "r" (256UL)
-        : "memory" );
+	: "memory" );
 
 	XMMS_RESTORE;
 }
@@ -261,70 +265,70 @@ static void
 xor_sse_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 	  unsigned long *p3, unsigned long *p4, unsigned long *p5)
 {
-        unsigned int lines = bytes >> 8;
+	unsigned int lines = bytes >> 8;
 	xmm_store_t xmm_save[4];
 	unsigned long cr0;
 
 	XMMS_SAVE;
 
-        __asm__ __volatile__ (
+	asm volatile(
 #undef BLOCK
 #define BLOCK(i) \
 		PF1(i)					\
-				PF1(i+2)		\
-		LD(i,0)					\
-			LD(i+1,1)			\
-				LD(i+2,2)		\
-					LD(i+3,3)	\
+				PF1(i + 2)		\
+		LD(i, 0)				\
+			LD(i + 1, 1)			\
+				LD(i + 2, 2)		\
+					LD(i + 3, 3)	\
 		PF2(i)					\
-				PF2(i+2)		\
-		XO1(i,0)				\
-			XO1(i+1,1)			\
-				XO1(i+2,2)		\
-					XO1(i+3,3)	\
+				PF2(i + 2)		\
+		XO1(i, 0)				\
+			XO1(i + 1, 1)			\
+				XO1(i + 2, 2)		\
+					XO1(i + 3, 3)	\
 		PF3(i)					\
-				PF3(i+2)		\
-		XO2(i,0)				\
-			XO2(i+1,1)			\
-				XO2(i+2,2)		\
-					XO2(i+3,3)	\
+				PF3(i + 2)		\
+		XO2(i, 0)				\
+			XO2(i + 1, 1)			\
+				XO2(i + 2, 2)		\
+					XO2(i + 3, 3)	\
 		PF4(i)					\
-				PF4(i+2)		\
-		PF0(i+4)				\
-				PF0(i+6)		\
-		XO3(i,0)				\
-			XO3(i+1,1)			\
-				XO3(i+2,2)		\
-					XO3(i+3,3)	\
-		XO4(i,0)				\
-			XO4(i+1,1)			\
-				XO4(i+2,2)		\
-					XO4(i+3,3)	\
-		ST(i,0)					\
-			ST(i+1,1)			\
-				ST(i+2,2)		\
-					ST(i+3,3)	\
+				PF4(i + 2)		\
+		PF0(i + 4)				\
+				PF0(i + 6)		\
+		XO3(i, 0)				\
+			XO3(i + 1, 1)			\
+				XO3(i + 2, 2)		\
+					XO3(i + 3, 3)	\
+		XO4(i, 0)				\
+			XO4(i + 1, 1)			\
+				XO4(i + 2, 2)		\
+					XO4(i + 3, 3)	\
+		ST(i, 0)				\
+			ST(i + 1, 1)			\
+				ST(i + 2, 2)		\
+					ST(i + 3, 3)	\
 
 
 		PF0(0)
 				PF0(2)
 
 	" .align 32			;\n"
-        " 1:                            ;\n"
+	" 1:                            ;\n"
 
 		BLOCK(0)
 		BLOCK(4)
 		BLOCK(8)
 		BLOCK(12)
 
-        "       addq %[inc], %[p1]           ;\n"
-        "       addq %[inc], %[p2]           ;\n"
-        "       addq %[inc], %[p3]           ;\n"
-        "       addq %[inc], %[p4]           ;\n"
-        "       addq %[inc], %[p5]           ;\n"
+	"       addq %[inc], %[p1]           ;\n"
+	"       addq %[inc], %[p2]           ;\n"
+	"       addq %[inc], %[p3]           ;\n"
+	"       addq %[inc], %[p4]           ;\n"
+	"       addq %[inc], %[p5]           ;\n"
 	"	decl %[cnt] ; jnz 1b"
 	: [cnt] "+c" (lines),
-  	  [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), [p4] "+r" (p4), 
+	  [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), [p4] "+r" (p4),
 	  [p5] "+r" (p5)
 	: [inc] "r" (256UL)
 	: "memory");
@@ -333,18 +337,18 @@ xor_sse_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
 }
 
 static struct xor_block_template xor_block_sse = {
-        .name = "generic_sse",
-        .do_2 = xor_sse_2,
-        .do_3 = xor_sse_3,
-        .do_4 = xor_sse_4,
-        .do_5 = xor_sse_5,
+	.name = "generic_sse",
+	.do_2 = xor_sse_2,
+	.do_3 = xor_sse_3,
+	.do_4 = xor_sse_4,
+	.do_5 = xor_sse_5,
 };
 
 #undef XOR_TRY_TEMPLATES
-#define XOR_TRY_TEMPLATES				\
-	do {						\
-		xor_speed(&xor_block_sse);	\
-	} while (0)
+#define XOR_TRY_TEMPLATES			\
+do {						\
+	xor_speed(&xor_block_sse);		\
+} while (0)
 
 /* We force the use of the SSE xor block because it can write around L2.
    We may also be able to load into the L1 only depending on how the cpu
-- 
cgit v1.2.3


From 7fda20f146d5d217684ffbc37c6b6c5f82c2dffd Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 29 Feb 2008 10:29:38 +0100
Subject: x86: spinlock ops are always-inlined

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/spinlock.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index 47dfe2607bb1..bc6376f1bc5a 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -78,7 +78,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock)
 	return (((tmp >> 8) & 0xff) - (tmp & 0xff)) > 1;
 }
 
-static inline void __raw_spin_lock(raw_spinlock_t *lock)
+static __always_inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
 	short inc = 0x0100;
 
@@ -99,7 +99,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 
-static inline int __raw_spin_trylock(raw_spinlock_t *lock)
+static __always_inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
 	int tmp;
 	short new;
@@ -120,7 +120,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 	return tmp;
 }
 
-static inline void __raw_spin_unlock(raw_spinlock_t *lock)
+static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
 	asm volatile(UNLOCK_LOCK_PREFIX "incb %0"
 		     : "+m" (lock->slock)
@@ -142,7 +142,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock)
 	return (((tmp >> 16) & 0xffff) - (tmp & 0xffff)) > 1;
 }
 
-static inline void __raw_spin_lock(raw_spinlock_t *lock)
+static __always_inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
 	int inc = 0x00010000;
 	int tmp;
@@ -165,7 +165,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock)
 
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 
-static inline int __raw_spin_trylock(raw_spinlock_t *lock)
+static __always_inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
 	int tmp;
 	int new;
@@ -187,7 +187,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 	return tmp;
 }
 
-static inline void __raw_spin_unlock(raw_spinlock_t *lock)
+static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
 	asm volatile(UNLOCK_LOCK_PREFIX "incw %0"
 		     : "+m" (lock->slock)
-- 
cgit v1.2.3


From 43cdf5d6e0a75c1069adc8d126b97b792ff53142 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Sat, 22 Mar 2008 18:50:22 +0100
Subject: x86: pgtable, document pde bits

Some of pde bits weren't documented, add the short description to them.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/pgtable.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 2ce765070464..f1d9f4a03f6f 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -4,13 +4,13 @@
 #define USER_PTRS_PER_PGD	((TASK_SIZE-1)/PGDIR_SIZE+1)
 #define FIRST_USER_ADDRESS	0
 
-#define _PAGE_BIT_PRESENT	0
-#define _PAGE_BIT_RW		1
-#define _PAGE_BIT_USER		2
-#define _PAGE_BIT_PWT		3
-#define _PAGE_BIT_PCD		4
-#define _PAGE_BIT_ACCESSED	5
-#define _PAGE_BIT_DIRTY		6
+#define _PAGE_BIT_PRESENT	0	/* is present */
+#define _PAGE_BIT_RW		1	/* writeable */
+#define _PAGE_BIT_USER		2	/* userspace addressable */
+#define _PAGE_BIT_PWT		3	/* page write through */
+#define _PAGE_BIT_PCD		4	/* page cache disabled */
+#define _PAGE_BIT_ACCESSED	5	/* was accessed (raised by CPU) */
+#define _PAGE_BIT_DIRTY		6	/* was written to (raised by CPU) */
 #define _PAGE_BIT_FILE		6
 #define _PAGE_BIT_PSE		7	/* 4 MB (or 2MB) page */
 #define _PAGE_BIT_PAT		7	/* on 4KB pages */
-- 
cgit v1.2.3


From 5524ea320d80e3ac6aeeec44216660831c76da08 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Tue, 11 Mar 2008 02:23:20 +0100
Subject: x86: don't set up early exception handlers for external interrupts

All of early setup runs with interrupts disabled, so there is no
need to set up early exception handlers for vectors >= 32

This saves some minor text size.

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/head64.c  | 2 +-
 arch/x86/kernel/head_64.S | 6 ++----
 include/asm-x86/segment.h | 3 ++-
 3 files changed, 5 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 48be76cda93b..d6d54faa84df 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -127,7 +127,7 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	/* Cleanup the over mapped high alias */
 	cleanup_highmap();
 
-	for (i = 0; i < IDT_ENTRIES; i++) {
+	for (i = 0; i < NUM_EXCEPTION_VECTORS; i++) {
 #ifdef CONFIG_EARLY_PRINTK
 		set_intr_gate(i, &early_idt_handlers[i]);
 #else
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 017216916dff..2c0abe0e3c68 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -278,10 +278,8 @@ bad_address:
 
 	.globl early_idt_handlers
 early_idt_handlers:
-	early_idt_tramp 0, 63
-	early_idt_tramp 64, 127
-	early_idt_tramp 128, 191
-	early_idt_tramp 192, 255
+	.set maxe, NUM_EXCEPTION_VECTORS-1
+	early_idt_tramp 0, maxe
 #endif
 
 ENTRY(early_idt_handler)
diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index 23f0535fec61..ed5131dd7d92 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -191,13 +191,14 @@
 #define SEGMENT_TI_MASK		0x4
 
 #define IDT_ENTRIES 256
+#define NUM_EXCEPTION_VECTORS 32
 #define GDT_SIZE (GDT_ENTRIES * 8)
 #define GDT_ENTRY_TLS_ENTRIES 3
 #define TLS_SIZE (GDT_ENTRY_TLS_ENTRIES * 8)
 
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
-extern const char early_idt_handlers[IDT_ENTRIES][10];
+extern const char early_idt_handlers[NUM_EXCEPTION_VECTORS][10];
 #endif
 #endif
 
-- 
cgit v1.2.3


From 7d1116a92d709c22e7db910724c9fcd2001b0499 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Wed, 12 Mar 2008 03:53:27 +0100
Subject: x86: implement true end_pfn_mapped for 32bit

Even on 32bit 2MB pages can map more memory than is in the true
max_low_pfn if end_pfn is not highmem and not aligned to 2MB.
Add a end_pfn_map similar to x86-64 that accounts for this
fact. This is important for code that really needs to know about
all mapping aliases.

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/init_32.c     | 4 ++++
 include/asm-x86/page.h    | 4 +++-
 include/asm-x86/page_64.h | 1 -
 3 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 00168e65688a..73dd0601166a 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -51,6 +51,8 @@
 
 unsigned int __VMALLOC_RESERVE = 128 << 20;
 
+unsigned long end_pfn_map;
+
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 unsigned long highstart_pfn, highend_pfn;
 
@@ -194,6 +196,7 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
 				set_pmd(pmd, pfn_pmd(pfn, prot));
 
 				pfn += PTRS_PER_PTE;
+				end_pfn_map = pfn;
 				continue;
 			}
 			pte = one_page_table_init(pmd);
@@ -208,6 +211,7 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
 
 				set_pte(pte, pfn_pte(pfn, prot));
 			}
+			end_pfn_map = pfn;
 		}
 	}
 }
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index a05b2896492f..b734939916c4 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -36,7 +36,7 @@
 #define max_pfn_mapped		end_pfn_map
 #else
 #include <asm/page_32.h>
-#define max_pfn_mapped		max_low_pfn
+#define max_pfn_mapped		end_pfn_map
 #endif	/* CONFIG_X86_64 */
 
 #define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
@@ -50,6 +50,8 @@
 
 extern int page_is_ram(unsigned long pagenr);
 
+extern unsigned long end_pfn_map;
+
 struct page;
 
 static inline void clear_user_page(void *page, unsigned long vaddr,
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index f156778f707c..54d5db634858 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -59,7 +59,6 @@ void clear_page(void *page);
 void copy_page(void *to, void *from);
 
 extern unsigned long end_pfn;
-extern unsigned long end_pfn_map;
 extern unsigned long phys_base;
 
 extern unsigned long __phys_addr(unsigned long);
-- 
cgit v1.2.3


From 67794292c8615b05f46419ba8d4fd99e7c9a5db9 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 21 Mar 2008 21:27:10 +0100
Subject: x86: replace the now useless max_pfn_mapped define

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/boot.c |  2 +-
 arch/x86/kernel/e820_64.c   | 28 ++++++++++++++--------------
 arch/x86/kernel/setup_64.c  |  2 +-
 arch/x86/mm/init_32.c       |  6 +++---
 arch/x86/mm/init_64.c       |  2 +-
 include/asm-x86/page.h      |  4 +---
 6 files changed, 21 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 956b60f3ebd5..e277c370246d 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -115,7 +115,7 @@ char *__init __acpi_map_table(unsigned long phys_addr, unsigned long size)
 	if (!phys_addr || !size)
 		return NULL;
 
-	if (phys_addr+size <= (end_pfn_map << PAGE_SHIFT) + PAGE_SIZE)
+	if (phys_addr+size <= (max_pfn_mapped << PAGE_SHIFT) + PAGE_SIZE)
 		return __va(phys_addr);
 
 	return NULL;
diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index d6ada0833876..a720f3d5ed9d 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -36,11 +36,11 @@ struct e820map e820;
 unsigned long end_pfn;
 
 /*
- * end_pfn only includes RAM, while end_pfn_map includes all e820 entries.
- * The direct mapping extends to end_pfn_map, so that we can directly access
+ * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries.
+ * The direct mapping extends to max_pfn_mapped, so that we can directly access
  * apertures, ACPI and other tables without having to play with fixmaps.
  */
-unsigned long end_pfn_map;
+unsigned long max_pfn_mapped;
 
 /*
  * Last pfn which the user wants to use.
@@ -281,16 +281,16 @@ unsigned long __init e820_end_of_ram(void)
 
 	end_pfn = find_max_pfn_with_active_regions();
 
-	if (end_pfn > end_pfn_map)
-		end_pfn_map = end_pfn;
-	if (end_pfn_map > MAXMEM>>PAGE_SHIFT)
-		end_pfn_map = MAXMEM>>PAGE_SHIFT;
+	if (end_pfn > max_pfn_mapped)
+		max_pfn_mapped = end_pfn;
+	if (max_pfn_mapped > MAXMEM>>PAGE_SHIFT)
+		max_pfn_mapped = MAXMEM>>PAGE_SHIFT;
 	if (end_pfn > end_user_pfn)
 		end_pfn = end_user_pfn;
-	if (end_pfn > end_pfn_map)
-		end_pfn = end_pfn_map;
+	if (end_pfn > max_pfn_mapped)
+		end_pfn = max_pfn_mapped;
 
-	printk(KERN_INFO "end_pfn_map = %lu\n", end_pfn_map);
+	printk(KERN_INFO "max_pfn_mapped = %lu\n", max_pfn_mapped);
 	return end_pfn;
 }
 
@@ -366,9 +366,9 @@ static int __init e820_find_active_region(const struct e820entry *ei,
 	if (*ei_startpfn >= *ei_endpfn)
 		return 0;
 
-	/* Check if end_pfn_map should be updated */
-	if (ei->type != E820_RAM && *ei_endpfn > end_pfn_map)
-		end_pfn_map = *ei_endpfn;
+	/* Check if max_pfn_mapped should be updated */
+	if (ei->type != E820_RAM && *ei_endpfn > max_pfn_mapped)
+		max_pfn_mapped = *ei_endpfn;
 
 	/* Skip if map is outside the node */
 	if (ei->type != E820_RAM || *ei_endpfn <= start_pfn ||
@@ -759,7 +759,7 @@ static int __init parse_memmap_opt(char *p)
 		saved_max_pfn = e820_end_of_ram();
 		remove_all_active_ranges();
 #endif
-		end_pfn_map = 0;
+		max_pfn_mapped = 0;
 		e820.nr_map = 0;
 		userdef = 1;
 		return 0;
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index c6fe1e4bc7c2..413b8fc31545 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -347,7 +347,7 @@ void __init setup_arch(char **cmdline_p)
 
 	check_efer();
 
-	init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
+	init_memory_mapping(0, (max_pfn_mapped << PAGE_SHIFT));
 	if (efi_enabled)
 		efi_init();
 
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 73dd0601166a..fc3ace2e88f1 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -51,7 +51,7 @@
 
 unsigned int __VMALLOC_RESERVE = 128 << 20;
 
-unsigned long end_pfn_map;
+unsigned long max_pfn_mapped;
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 unsigned long highstart_pfn, highend_pfn;
@@ -196,7 +196,7 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
 				set_pmd(pmd, pfn_pmd(pfn, prot));
 
 				pfn += PTRS_PER_PTE;
-				end_pfn_map = pfn;
+				max_pfn_mapped = pfn;
 				continue;
 			}
 			pte = one_page_table_init(pmd);
@@ -211,7 +211,7 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
 
 				set_pte(pte, pfn_pte(pfn, prot));
 			}
-			end_pfn_map = pfn;
+			max_pfn_mapped = pfn;
 		}
 	}
 }
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 210243e94d84..ef9e9cfb1fc2 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -791,7 +791,7 @@ void __init reserve_bootmem_generic(unsigned long phys, unsigned len)
 		 * This can happen with kdump kernels when accessing
 		 * firmware tables:
 		 */
-		if (pfn < end_pfn_map)
+		if (pfn < max_pfn_mapped)
 			return;
 
 		printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %u\n",
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index b734939916c4..6724a4bc6b7a 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -33,10 +33,8 @@
 
 #ifdef CONFIG_X86_64
 #include <asm/page_64.h>
-#define max_pfn_mapped		end_pfn_map
 #else
 #include <asm/page_32.h>
-#define max_pfn_mapped		end_pfn_map
 #endif	/* CONFIG_X86_64 */
 
 #define PAGE_OFFSET		((unsigned long)__PAGE_OFFSET)
@@ -50,7 +48,7 @@
 
 extern int page_is_ram(unsigned long pagenr);
 
-extern unsigned long end_pfn_map;
+extern unsigned long max_pfn_mapped;
 
 struct page;
 
-- 
cgit v1.2.3


From cc6150321903ca4c3bc9d53b0cdafb05d77d64d0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Wed, 12 Mar 2008 03:53:28 +0100
Subject: x86: account overlapped mappings in max_pfn_mapped

When end_pfn is not aligned to 2MB (or 1GB) then the kernel might
map more memory than end_pfn. Account this in max_pfn_mapped.

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup_64.c |  2 +-
 arch/x86/mm/init_64.c      | 34 +++++++++++++++++++++++-----------
 include/asm-x86/page_64.h  |  3 +++
 include/asm-x86/proto.h    |  2 --
 4 files changed, 27 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 413b8fc31545..3d76dbd9f2c0 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -347,7 +347,7 @@ void __init setup_arch(char **cmdline_p)
 
 	check_efer();
 
-	init_memory_mapping(0, (max_pfn_mapped << PAGE_SHIFT));
+	max_pfn_mapped = init_memory_mapping(0, (max_pfn_mapped << PAGE_SHIFT));
 	if (efi_enabled)
 		efi_init();
 
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index ef9e9cfb1fc2..1076097dcab2 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -313,7 +313,7 @@ __meminit void early_iounmap(void *addr, unsigned long size)
 	__flush_tlb_all();
 }
 
-static void __meminit
+static unsigned long __meminit
 phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
 {
 	int i = pmd_index(address);
@@ -335,21 +335,26 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
 		set_pte((pte_t *)pmd,
 			pfn_pte(address >> PAGE_SHIFT, PAGE_KERNEL_LARGE));
 	}
+	return address;
 }
 
-static void __meminit
+static unsigned long __meminit
 phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
 {
 	pmd_t *pmd = pmd_offset(pud, 0);
+	unsigned long last_map_addr;
+
 	spin_lock(&init_mm.page_table_lock);
-	phys_pmd_init(pmd, address, end);
+	last_map_addr = phys_pmd_init(pmd, address, end);
 	spin_unlock(&init_mm.page_table_lock);
 	__flush_tlb_all();
+	return last_map_addr;
 }
 
-static void __meminit
+static unsigned long __meminit
 phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
 {
+	unsigned long last_map_addr = end;
 	int i = pud_index(addr);
 
 	for (; i < PTRS_PER_PUD; i++, addr = (addr & PUD_MASK) + PUD_SIZE) {
@@ -368,13 +373,14 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
 
 		if (pud_val(*pud)) {
 			if (!pud_large(*pud))
-				phys_pmd_update(pud, addr, end);
+				last_map_addr = phys_pmd_update(pud, addr, end);
 			continue;
 		}
 
 		if (direct_gbpages) {
 			set_pte((pte_t *)pud,
 				pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL_LARGE));
+			last_map_addr = (addr & PUD_MASK) + PUD_SIZE;
 			continue;
 		}
 
@@ -382,12 +388,14 @@ phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
 
 		spin_lock(&init_mm.page_table_lock);
 		set_pud(pud, __pud(pmd_phys | _KERNPG_TABLE));
-		phys_pmd_init(pmd, addr, end);
+		last_map_addr = phys_pmd_init(pmd, addr, end);
 		spin_unlock(&init_mm.page_table_lock);
 
 		unmap_low_page(pmd);
 	}
 	__flush_tlb_all();
+
+	return last_map_addr >> PAGE_SHIFT;
 }
 
 static void __init find_early_table_space(unsigned long end)
@@ -542,9 +550,9 @@ static void __init early_memtest(unsigned long start, unsigned long end)
  * This runs before bootmem is initialized and gets pages directly from
  * the physical memory. To access them they are temporarily mapped.
  */
-void __init_refok init_memory_mapping(unsigned long start, unsigned long end)
+unsigned long __init_refok init_memory_mapping(unsigned long start, unsigned long end)
 {
-	unsigned long next;
+	unsigned long next, last_map_addr = end;
 	unsigned long start_phys = start, end_phys = end;
 
 	printk(KERN_INFO "init_memory_mapping\n");
@@ -577,7 +585,7 @@ void __init_refok init_memory_mapping(unsigned long start, unsigned long end)
 		next = start + PGDIR_SIZE;
 		if (next > end)
 			next = end;
-		phys_pud_init(pud, __pa(start), __pa(next));
+		last_map_addr = phys_pud_init(pud, __pa(start), __pa(next));
 		if (!after_bootmem)
 			set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
 		unmap_low_page(pud);
@@ -593,6 +601,8 @@ void __init_refok init_memory_mapping(unsigned long start, unsigned long end)
 
 	if (!after_bootmem)
 		early_memtest(start_phys, end_phys);
+
+	return last_map_addr;
 }
 
 #ifndef CONFIG_NUMA
@@ -632,11 +642,13 @@ int arch_add_memory(int nid, u64 start, u64 size)
 {
 	struct pglist_data *pgdat = NODE_DATA(nid);
 	struct zone *zone = pgdat->node_zones + ZONE_NORMAL;
-	unsigned long start_pfn = start >> PAGE_SHIFT;
+	unsigned long last_mapped_pfn, start_pfn = start >> PAGE_SHIFT;
 	unsigned long nr_pages = size >> PAGE_SHIFT;
 	int ret;
 
-	init_memory_mapping(start, start + size-1);
+	last_mapped_pfn = init_memory_mapping(start, start + size-1);
+	if (last_mapped_pfn > max_pfn_mapped)
+		max_pfn_mapped = last_mapped_pfn;
 
 	ret = __add_pages(zone, start_pfn, nr_pages);
 	WARN_ON(1);
diff --git a/include/asm-x86/page_64.h b/include/asm-x86/page_64.h
index 54d5db634858..6ea72859c491 100644
--- a/include/asm-x86/page_64.h
+++ b/include/asm-x86/page_64.h
@@ -80,6 +80,9 @@ typedef struct { pteval_t pte; } pte_t;
 
 #define vmemmap ((struct page *)VMEMMAP_START)
 
+extern unsigned long init_memory_mapping(unsigned long start,
+					 unsigned long end);
+
 #endif	/* !__ASSEMBLY__ */
 
 #ifdef CONFIG_FLATMEM
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index 9da46af3b0e9..1e17bcce450e 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -7,8 +7,6 @@
 
 extern void early_idt_handler(void);
 
-extern void init_memory_mapping(unsigned long start, unsigned long end);
-
 extern void system_call(void);
 extern void syscall_init(void);
 
-- 
cgit v1.2.3


From c9caa02c529d5e113e40cbc77254558fcdfa4215 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Wed, 12 Mar 2008 03:53:29 +0100
Subject: x86: add set_memory_4k to pageattr.c

Add a new function to force split large pages into 4k pages.
This is needed for some followup optimizations.

I had to add a new field to cpa_data to pass down the information
that try_preserve_large_page should not run.

Right now no set_page_4k() because I didn't need it and all the
specialized users I have in mind would be more comfortable with
pure addresses. I also didn't export it because it's unlikely
external code needs it.

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/pageattr.c       | 20 ++++++++++++++++----
 include/asm-x86/cacheflush.h |  1 +
 2 files changed, 17 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 270cab2e6030..7d9517abc9af 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -31,6 +31,7 @@ struct cpa_data {
 	int		numpages;
 	int		flushtlb;
 	unsigned long	pfn;
+	unsigned	force_split : 1;
 };
 
 #ifdef CONFIG_X86_64
@@ -262,6 +263,9 @@ try_preserve_large_page(pte_t *kpte, unsigned long address,
 	int i, do_split = 1;
 	unsigned int level;
 
+	if (cpa->force_split)
+		return 1;
+
 	spin_lock_irqsave(&pgd_lock, flags);
 	/*
 	 * Check for races, another CPU might have split this page
@@ -696,7 +700,8 @@ static inline int cache_attr(pgprot_t attr)
 }
 
 static int change_page_attr_set_clr(unsigned long addr, int numpages,
-				    pgprot_t mask_set, pgprot_t mask_clr)
+				    pgprot_t mask_set, pgprot_t mask_clr,
+				    int force_split)
 {
 	struct cpa_data cpa;
 	int ret, cache, checkalias;
@@ -707,7 +712,7 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages,
 	 */
 	mask_set = canon_pgprot(mask_set);
 	mask_clr = canon_pgprot(mask_clr);
-	if (!pgprot_val(mask_set) && !pgprot_val(mask_clr))
+	if (!pgprot_val(mask_set) && !pgprot_val(mask_clr) && !force_split)
 		return 0;
 
 	/* Ensure we are PAGE_SIZE aligned */
@@ -724,6 +729,7 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages,
 	cpa.mask_set = mask_set;
 	cpa.mask_clr = mask_clr;
 	cpa.flushtlb = 0;
+	cpa.force_split = force_split;
 
 	/* No alias checking for _NX bit modifications */
 	checkalias = (pgprot_val(mask_set) | pgprot_val(mask_clr)) != _PAGE_NX;
@@ -762,13 +768,13 @@ out:
 static inline int change_page_attr_set(unsigned long addr, int numpages,
 				       pgprot_t mask)
 {
-	return change_page_attr_set_clr(addr, numpages, mask, __pgprot(0));
+	return change_page_attr_set_clr(addr, numpages, mask, __pgprot(0), 0);
 }
 
 static inline int change_page_attr_clear(unsigned long addr, int numpages,
 					 pgprot_t mask)
 {
-	return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask);
+	return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask, 0);
 }
 
 int _set_memory_uc(unsigned long addr, int numpages)
@@ -847,6 +853,12 @@ int set_memory_np(unsigned long addr, int numpages)
 	return change_page_attr_clear(addr, numpages, __pgprot(_PAGE_PRESENT));
 }
 
+int set_memory_4k(unsigned long addr, int numpages)
+{
+	return change_page_attr_set_clr(addr, numpages, __pgprot(0),
+					__pgprot(0), 1);
+}
+
 int set_pages_uc(struct page *page, int numpages)
 {
 	unsigned long addr = (unsigned long)page_address(page);
diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index 7ab5b520b7bd..cb1d6f8fd003 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -45,6 +45,7 @@ int set_memory_nx(unsigned long addr, int numpages);
 int set_memory_ro(unsigned long addr, int numpages);
 int set_memory_rw(unsigned long addr, int numpages);
 int set_memory_np(unsigned long addr, int numpages);
+int set_memory_4k(unsigned long addr, int numpages);
 
 void clflush_cache_range(void *addr, unsigned int size);
 
-- 
cgit v1.2.3


From 1de87bd40e119d26533b5135677901990390bfa9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Sat, 22 Mar 2008 10:59:28 +0100
Subject: x86: re-add rdmsrl_safe

RDMSR for 64bit values with exception handling.

Makes it easier to deal with 64bit valued MSRs. The old 64bit code
base had that too as checking_rdmsrl(), but it got dropped somehow.

Signed-off-by: Andi Kleen <andi@firstfloor.org>
Cc: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/msr.h      | 8 ++++++++
 include/asm-x86/paravirt.h | 7 +++++++
 2 files changed, 15 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/msr.h b/include/asm-x86/msr.h
index 2c698a2e81f9..3707650a169b 100644
--- a/include/asm-x86/msr.h
+++ b/include/asm-x86/msr.h
@@ -150,6 +150,14 @@ static inline int wrmsr_safe(unsigned msr, unsigned low, unsigned high)
 	__err;							\
 })
 
+static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
+{
+	int err;
+
+	*p = native_read_msr_safe(msr, &err);
+	return err;
+}
+
 #define rdtscl(low)						\
 	((low) = (u32)native_read_tsc())
 
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 0c23f7940bc4..3d419398499b 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -693,6 +693,13 @@ do {						\
 	_err;					\
 })
 
+static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
+{
+	int err;
+
+	*p = paravirt_read_msr(msr, &err);
+	return err;
+}
 
 static inline u64 paravirt_read_tsc(void)
 {
-- 
cgit v1.2.3


From 8346ea17aa20e9864b0f7dc03d55f3cd5620b8c1 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Wed, 12 Mar 2008 03:53:32 +0100
Subject: x86: split large page mapping for AMD TSEG

On AMD SMM protected memory is part of the address map, but handled
internally like an MTRR. That leads to large pages getting split
internally which has some performance implications. Check for the
AMD TSEG MSR and split the large page mapping on that area
explicitely if it is part of the direct mapping.

There is also SMM ASEG, but it is in the first 1MB and already covered by
the earlier split first page patch.

Idea for this came from an earlier patch by Andreas Herrmann

On a RevF dual Socket Opteron system kernbench shows a clear
improvement from this:
(together with the earlier patches in this series, especially the
split first 2MB patch)

[lower is better]
              no split stddev         split  stddev    delta
Elapsed Time   87.146 (0.727516)     84.296 (1.09098)  -3.2%
User Time     274.537 (4.05226)     273.692 (3.34344)  -0.3%
System Time    34.907 (0.42492)      34.508 (0.26832)  -1.1%
Percent CPU   322.5   (38.3007)     326.5   (44.5128)  +1.2%

=> About 3.2% improvement in elapsed time for kernbench.

With GB pages on AMD Fam1h the impact of splitting is much higher of course,
since it would split two full GB pages (together with the first
1MB split patch) instead of two 2MB pages.  I could not benchmark
a clear difference in kernbench on gbpages, so I kept it disabled
for that case

That was only limited benchmarking of course, so if someone
was interested in running more tests for the gbpages case
that could be revisited (contributions welcome)

I didn't bother implementing this for 32bit because it is very
unlikely the 32bit lowmem mapping overlaps into the TSEG near 4GB
and the 2MB low split is already handled for both.

[ mingo@elte.hu: do it on gbpages kernels too, there's no clear reason
                 why it shouldnt help there. ]

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup_64.c  | 13 +++++++++++++
 include/asm-x86/msr-index.h |  1 +
 2 files changed, 14 insertions(+)

(limited to 'include')

diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index 3d76dbd9f2c0..b5425979501c 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -729,6 +729,19 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 
 	if (amd_apic_timer_broken())
 		disable_apic_timer = 1;
+
+	if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) {
+		unsigned long long tseg;
+
+		/*
+		 * Split up direct mapping around the TSEG SMM area.
+		 * Don't do it for gbpages because there seems very little
+		 * benefit in doing so.
+		 */
+		if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg) &&
+		(tseg >> PMD_SHIFT) < (max_pfn_mapped >> (PMD_SHIFT-PAGE_SHIFT)))
+			set_memory_4k((unsigned long)__va(tseg), 1);
+	}
 }
 
 void __cpuinit detect_ht(struct cpuinfo_x86 *c)
diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h
index af4e07f661b8..09413ad39d3c 100644
--- a/include/asm-x86/msr-index.h
+++ b/include/asm-x86/msr-index.h
@@ -112,6 +112,7 @@
 #define MSR_K8_SYSCFG			0xc0010010
 #define MSR_K8_HWCR			0xc0010015
 #define MSR_K8_ENABLE_C1E		0xc0010055
+#define MSR_K8_TSEG_ADDR		0xc0010112
 #define K8_MTRRFIXRANGE_DRAM_ENABLE	0x00040000 /* MtrrFixDramEn bit    */
 #define K8_MTRRFIXRANGE_DRAM_MODIFY	0x00080000 /* MtrrFixDramModEn bit */
 #define K8_MTRR_RDMEM_WRMEM_MASK	0x18181818 /* Mask: RdMem|WrMem    */
-- 
cgit v1.2.3


From 5af5573ee06c361378e22a9dd71dae0320e841f7 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Tue, 25 Mar 2008 13:28:56 -0300
Subject: x86: move ipi definitions to mach_ipi.h

take them out of the x86_64-only asm/mach_apic.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c               |  2 ++
 arch/x86/kernel/crash.c                 |  4 ----
 arch/x86/kernel/io_apic_64.c            |  2 ++
 arch/x86/kernel/smp.c                   |  6 +-----
 arch/x86/kernel/tlb_64.c                |  3 ++-
 include/asm-x86/mach-default/mach_ipi.h | 10 ++++++++++
 include/asm-x86/mach_apic.h             |  3 ---
 7 files changed, 17 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 5362cfd30ecd..206278f1c6f4 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -41,6 +41,8 @@
 #include <asm/timex.h>
 #include <asm/apic.h>
 
+#include <mach_ipi.h>
+
 int disable_apic_timer __cpuinitdata;
 static int apic_calibrate_pmtmr __initdata;
 int disable_apic;
diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index 9a5fa0abfcc7..2251d0ae9570 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -26,11 +26,7 @@
 #include <linux/kdebug.h>
 #include <asm/smp.h>
 
-#ifdef CONFIG_X86_32
 #include <mach_ipi.h>
-#else
-#include <asm/mach_apic.h>
-#endif
 
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 1627c0d53e0b..7d5cdf320eba 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -50,6 +50,8 @@
 #include <asm/msidef.h>
 #include <asm/hypertransport.h>
 
+#include <mach_ipi.h>
+
 struct irq_cfg {
 	cpumask_t domain;
 	cpumask_t old_domain;
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 16c52aaaca35..8f75893a6467 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -26,12 +26,8 @@
 #include <asm/tlbflush.h>
 #include <asm/mmu_context.h>
 #include <asm/proto.h>
-#ifdef CONFIG_X86_32
-#include <mach_apic.h>
 #include <mach_ipi.h>
-#else
-#include <asm/mach_apic.h>
-#endif
+#include <mach_apic.h>
 /*
  *	Some notes on x86 processor bugs affecting SMP operation:
  *
diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c
index 615d84817758..1558e513757e 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -11,11 +11,12 @@
 #include <asm/mtrr.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
-#include <asm/mach_apic.h>
 #include <asm/mmu_context.h>
 #include <asm/proto.h>
 #include <asm/apicdef.h>
 #include <asm/idle.h>
+
+#include <mach_ipi.h>
 /*
  *	Smarter SMP flushing macros.
  *		c/o Linus Torvalds.
diff --git a/include/asm-x86/mach-default/mach_ipi.h b/include/asm-x86/mach-default/mach_ipi.h
index 0dba244c86db..be323364e68f 100644
--- a/include/asm-x86/mach-default/mach_ipi.h
+++ b/include/asm-x86/mach-default/mach_ipi.h
@@ -9,10 +9,15 @@ void __send_IPI_shortcut(unsigned int shortcut, int vector);
 
 extern int no_broadcast;
 
+#ifdef CONFIG_X86_64
+#include <asm/genapic.h>
+#define send_IPI_mask (genapic->send_IPI_mask)
+#else
 static inline void send_IPI_mask(cpumask_t mask, int vector)
 {
 	send_IPI_mask_bitmask(mask, vector);
 }
+#endif
 
 static inline void __local_send_IPI_allbutself(int vector)
 {
@@ -33,6 +38,10 @@ static inline void __local_send_IPI_all(int vector)
 		__send_IPI_shortcut(APIC_DEST_ALLINC, vector);
 }
 
+#ifdef CONFIG_X86_64
+#define send_IPI_allbutself (genapic->send_IPI_allbutself)
+#define send_IPI_all (genapic->send_IPI_all)
+#else
 static inline void send_IPI_allbutself(int vector)
 {
 	/*
@@ -50,5 +59,6 @@ static inline void send_IPI_all(int vector)
 {
 	__local_send_IPI_all(vector);
 }
+#endif
 
 #endif /* __ASM_MACH_IPI_H */
diff --git a/include/asm-x86/mach_apic.h b/include/asm-x86/mach_apic.h
index 7b7115a0c1c9..1bc68c0c0cdf 100644
--- a/include/asm-x86/mach_apic.h
+++ b/include/asm-x86/mach_apic.h
@@ -20,9 +20,6 @@
 #define vector_allocation_domain	(genapic->vector_allocation_domain)
 #define apic_id_registered (genapic->apic_id_registered)
 #define init_apic_ldr (genapic->init_apic_ldr)
-#define send_IPI_mask (genapic->send_IPI_mask)
-#define send_IPI_allbutself (genapic->send_IPI_allbutself)
-#define send_IPI_all (genapic->send_IPI_all)
 #define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid)
 #define phys_pkg_id	(genapic->phys_pkg_id)
 
-- 
cgit v1.2.3


From dd46e3ca73d136aa7f9f1813e4cbb6934c3611cc Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Tue, 25 Mar 2008 18:10:46 -0300
Subject: x86: move apic declarations to mach_apic.h

take them out of the x86_64-specific asm/mach_apic.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_64.c                |  2 +-
 arch/x86/kernel/cpu/amd.c                |  2 +-
 arch/x86/kernel/io_apic_64.c             |  2 +-
 arch/x86/kernel/setup_64.c               |  2 +-
 include/asm-x86/mach-default/mach_apic.h | 83 ++++++++++++++++++--------------
 include/asm-x86/mach_apic.h              | 26 ----------
 6 files changed, 50 insertions(+), 67 deletions(-)
 delete mode 100644 include/asm-x86/mach_apic.h

(limited to 'include')

diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 206278f1c6f4..7dd6250aaf6c 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -34,7 +34,6 @@
 #include <asm/mpspec.h>
 #include <asm/hpet.h>
 #include <asm/pgalloc.h>
-#include <asm/mach_apic.h>
 #include <asm/nmi.h>
 #include <asm/idle.h>
 #include <asm/proto.h>
@@ -42,6 +41,7 @@
 #include <asm/apic.h>
 
 #include <mach_ipi.h>
+#include <mach_apic.h>
 
 int disable_apic_timer __cpuinitdata;
 static int apic_calibrate_pmtmr __initdata;
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 33d38f8305ee..0173065dc3b7 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -4,8 +4,8 @@
 #include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/apic.h>
-#include <asm/mach_apic.h>
 
+#include <mach_apic.h>
 #include "cpu.h"
 
 /*
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 7d5cdf320eba..6dd33628f28a 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -43,7 +43,6 @@
 #include <asm/smp.h>
 #include <asm/desc.h>
 #include <asm/proto.h>
-#include <asm/mach_apic.h>
 #include <asm/acpi.h>
 #include <asm/dma.h>
 #include <asm/nmi.h>
@@ -51,6 +50,7 @@
 #include <asm/hypertransport.h>
 
 #include <mach_ipi.h>
+#include <mach_apic.h>
 
 struct irq_cfg {
 	cpumask_t domain;
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index b5425979501c..540686be35d0 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -58,7 +58,6 @@
 #include <asm/mmu_context.h>
 #include <asm/proto.h>
 #include <asm/setup.h>
-#include <asm/mach_apic.h>
 #include <asm/numa.h>
 #include <asm/sections.h>
 #include <asm/dmi.h>
@@ -67,6 +66,7 @@
 #include <asm/ds.h>
 #include <asm/topology.h>
 
+#include <mach_apic.h>
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h
index 13900e8cc1ab..1f56e7d5bfdd 100644
--- a/include/asm-x86/mach-default/mach_apic.h
+++ b/include/asm-x86/mach-default/mach_apic.h
@@ -1,6 +1,8 @@
 #ifndef __ASM_MACH_APIC_H
 #define __ASM_MACH_APIC_H
 
+#ifdef CONFIG_X86_LOCAL_APIC
+
 #include <mach_apicdef.h>
 #include <asm/smp.h>
 
@@ -14,24 +16,25 @@ static inline cpumask_t target_cpus(void)
 	return cpumask_of_cpu(0);
 #endif
 } 
-#define TARGET_CPUS (target_cpus())
 
 #define NO_BALANCE_IRQ (0)
 #define esr_disable (0)
 
+#ifdef CONFIG_X86_64
+#include <asm/genapic.h>
+#define INT_DELIVERY_MODE (genapic->int_delivery_mode)
+#define INT_DEST_MODE (genapic->int_dest_mode)
+#define TARGET_CPUS	  (genapic->target_cpus())
+#define apic_id_registered (genapic->apic_id_registered)
+#define init_apic_ldr (genapic->init_apic_ldr)
+#define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid)
+#define phys_pkg_id	(genapic->phys_pkg_id)
+#define vector_allocation_domain    (genapic->vector_allocation_domain)
+extern void setup_apic_routing(void);
+#else
 #define INT_DELIVERY_MODE dest_LowestPrio
 #define INT_DEST_MODE 1     /* logical delivery broadcast to all procs */
-
-static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid)
-{
-	return physid_isset(apicid, bitmap);
-}
-
-static inline unsigned long check_apicid_present(int bit)
-{
-	return physid_isset(bit, phys_cpu_present_map);
-}
-
+#define TARGET_CPUS (target_cpus())
 /*
  * Set up the logical destination ID.
  *
@@ -49,32 +52,52 @@ static inline void init_apic_ldr(void)
 	apic_write_around(APIC_LDR, val);
 }
 
-static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map)
+static inline int apic_id_registered(void)
 {
-	return phys_map;
+	return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map);
+}
+
+static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
+{
+	return cpus_addr(cpumask)[0];
+}
+
+static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
+{
+	return cpuid_apic >> index_msb;
 }
 
-#ifdef CONFIG_X86_64
-extern void setup_apic_routing(void);
-#else
 static inline void setup_apic_routing(void)
 {
 	printk("Enabling APIC mode:  %s.  Using %d I/O APICs\n",
 					"Flat", nr_ioapics);
 }
-#endif
 
-static inline int multi_timer_check(int apic, int irq)
+static inline int apicid_to_node(int logical_apicid)
 {
 	return 0;
 }
+#endif
 
-#ifdef CONFIG_X86_32
-static inline int apicid_to_node(int logical_apicid)
+static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid)
+{
+	return physid_isset(apicid, bitmap);
+}
+
+static inline unsigned long check_apicid_present(int bit)
+{
+	return physid_isset(bit, phys_cpu_present_map);
+}
+
+static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_map)
+{
+	return phys_map;
+}
+
+static inline int multi_timer_check(int apic, int irq)
 {
 	return 0;
 }
-#endif
 
 /* Mapping from cpu number to logical apicid */
 static inline int cpu_to_logical_apicid(int cpu)
@@ -109,23 +132,9 @@ static inline int check_phys_apicid_present(int boot_cpu_physical_apicid)
 	return physid_isset(boot_cpu_physical_apicid, phys_cpu_present_map);
 }
 
-static inline int apic_id_registered(void)
-{
-	return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map);
-}
-
-static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
-{
-	return cpus_addr(cpumask)[0];
-}
-
 static inline void enable_apic_mode(void)
 {
 }
 
-static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
-{
-	return cpuid_apic >> index_msb;
-}
-
+#endif /* CONFIG_X86_LOCAL_APIC */
 #endif /* __ASM_MACH_APIC_H */
diff --git a/include/asm-x86/mach_apic.h b/include/asm-x86/mach_apic.h
deleted file mode 100644
index 1bc68c0c0cdf..000000000000
--- a/include/asm-x86/mach_apic.h
+++ /dev/null
@@ -1,26 +0,0 @@
-#ifndef __ASM_MACH_APIC_H
-#define __ASM_MACH_APIC_H
-
-/*
- * Copyright 2004 James Cleverdon, IBM.
- * Subject to the GNU Public License, v.2
- *
- * Generic APIC sub-arch defines.
- *
- * Hacked for x86-64 by James Cleverdon from i386 architecture code by
- * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and
- * James Cleverdon.
- */
-
-#include <asm/genapic.h>
-
-#define INT_DELIVERY_MODE (genapic->int_delivery_mode)
-#define INT_DEST_MODE (genapic->int_dest_mode)
-#define TARGET_CPUS	  (genapic->target_cpus())
-#define vector_allocation_domain	(genapic->vector_allocation_domain)
-#define apic_id_registered (genapic->apic_id_registered)
-#define init_apic_ldr (genapic->init_apic_ldr)
-#define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid)
-#define phys_pkg_id	(genapic->phys_pkg_id)
-
-#endif /* __ASM_MACH_APIC_H */
-- 
cgit v1.2.3


From 537e33136443bcd53ee13bc32a8f0fa46b1f3fdb Mon Sep 17 00:00:00 2001
From: Jesper Juhl <jesper.juhl@gmail.com>
Date: Wed, 26 Mar 2008 02:16:15 +0100
Subject: x86 floppy: kill off the 'register' keyword from header

When compilers became generally better at optimizing code than humans, the
register keyword became mostly useless. For the floppy driver it certainly
is since it's so slow compared to the rest of the system that optimizing
access to a single variable or two isn't going to make any real difference
So let's just leave it to the compiler - it'll do a better job anyway.

This patch does away with a few register keywords in the x86 floppy driver.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/floppy.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/floppy.h b/include/asm-x86/floppy.h
index 438b3033a250..dbe82a5c5eac 100644
--- a/include/asm-x86/floppy.h
+++ b/include/asm-x86/floppy.h
@@ -53,7 +53,7 @@ static int doing_pdma;
 
 static irqreturn_t floppy_hardint(int irq, void *dev_id)
 {
-	register unsigned char st;
+	unsigned char st;
 
 #undef TRACE_FLPY_INT
 
@@ -71,8 +71,8 @@ static irqreturn_t floppy_hardint(int irq, void *dev_id)
 #endif
 
 	{
-		register int lcount;
-		register char *lptr;
+		int lcount;
+		char *lptr;
 
 		st = 1;
 		for (lcount = virtual_dma_count, lptr = virtual_dma_addr;
-- 
cgit v1.2.3


From aa040b2f0693695ae393cd9b8a93055952dbf76f Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Sat, 22 Mar 2008 13:27:38 -0700
Subject: x86: simplify sync_test_bit(), improve

Using a naked parameterless macro could lead to other tokens being
unexpectedly replaced.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/sync_bitops.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/sync_bitops.h b/include/asm-x86/sync_bitops.h
index f1078a5e4ed7..b47a1d0b8a83 100644
--- a/include/asm-x86/sync_bitops.h
+++ b/include/asm-x86/sync_bitops.h
@@ -123,7 +123,7 @@ static inline int sync_test_and_change_bit(int nr, volatile unsigned long *addr)
 	return oldbit;
 }
 
-#define sync_test_bit test_bit
+#define sync_test_bit(nr, addr) test_bit(nr, addr)
 
 #undef ADDR
 
-- 
cgit v1.2.3


From 7219bebd72726c13c1eaaa3ade0e829e998fb3b1 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Thu, 17 Apr 2008 17:41:31 +0200
Subject: x86: add comments to describe the new api's in cacheflush.h

The new cacheflush.h API's didn't have any comments describing
how they're to be used yet and the conventions around these functions.
This patch adds comments to this effect; in order for that to be
a logical series, some prototypes had to move around.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/cacheflush.h | 63 ++++++++++++++++++++++++++++++++++++++------
 1 file changed, 55 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/cacheflush.h b/include/asm-x86/cacheflush.h
index cb1d6f8fd003..f4c0ab50d2c2 100644
--- a/include/asm-x86/cacheflush.h
+++ b/include/asm-x86/cacheflush.h
@@ -24,15 +24,34 @@
 #define copy_from_user_page(vma, page, vaddr, dst, src, len)	\
 	memcpy((dst), (src), (len))
 
-int __deprecated_for_modules change_page_attr(struct page *page, int numpages,
-					      pgprot_t prot);
 
-int set_pages_uc(struct page *page, int numpages);
-int set_pages_wb(struct page *page, int numpages);
-int set_pages_x(struct page *page, int numpages);
-int set_pages_nx(struct page *page, int numpages);
-int set_pages_ro(struct page *page, int numpages);
-int set_pages_rw(struct page *page, int numpages);
+/*
+ * The set_memory_* API can be used to change various attributes of a virtual
+ * address range. The attributes include:
+ * Cachability   : UnCached, WriteCombining, WriteBack
+ * Executability : eXeutable, NoteXecutable
+ * Read/Write    : ReadOnly, ReadWrite
+ * Presence      : NotPresent
+ *
+ * Within a catagory, the attributes are mutually exclusive.
+ *
+ * The implementation of this API will take care of various aspects that
+ * are associated with changing such attributes, such as:
+ * - Flushing TLBs
+ * - Flushing CPU caches
+ * - Making sure aliases of the memory behind the mapping don't violate
+ *   coherency rules as defined by the CPU in the system.
+ *
+ * What this API does not do:
+ * - Provide exclusion between various callers - including callers that
+ *   operation on other mappings of the same physical page
+ * - Restore default attributes when a page is freed
+ * - Guarantee that mappings other than the requested one are
+ *   in any state, other than that these do not violate rules for
+ *   the CPU you have. Do not depend on any effects on other mappings,
+ *   CPUs other than the one you have may have more relaxed rules.
+ * The caller is required to take care of these.
+ */
 
 int _set_memory_uc(unsigned long addr, int numpages);
 int _set_memory_wc(unsigned long addr, int numpages);
@@ -47,6 +66,34 @@ int set_memory_rw(unsigned long addr, int numpages);
 int set_memory_np(unsigned long addr, int numpages);
 int set_memory_4k(unsigned long addr, int numpages);
 
+/*
+ * For legacy compatibility with the old APIs, a few functions
+ * are provided that work on a "struct page".
+ * These functions operate ONLY on the 1:1 kernel mapping of the
+ * memory that the struct page represents, and internally just
+ * call the set_memory_* function. See the description of the
+ * set_memory_* function for more details on conventions.
+ *
+ * These APIs should be considered *deprecated* and are likely going to
+ * be removed in the future.
+ * The reason for this is the implicit operation on the 1:1 mapping only,
+ * making this not a generally useful API.
+ *
+ * Specifically, many users of the old APIs had a virtual address,
+ * called virt_to_page() or vmalloc_to_page() on that address to
+ * get a struct page* that the old API required.
+ * To convert these cases, use set_memory_*() on the original
+ * virtual address, do not use these functions.
+ */
+
+int set_pages_uc(struct page *page, int numpages);
+int set_pages_wb(struct page *page, int numpages);
+int set_pages_x(struct page *page, int numpages);
+int set_pages_nx(struct page *page, int numpages);
+int set_pages_ro(struct page *page, int numpages);
+int set_pages_rw(struct page *page, int numpages);
+
+
 void clflush_cache_range(void *addr, unsigned int size);
 
 void cpa_init(void);
-- 
cgit v1.2.3


From 903dcb5a1bd0ef2b09d756f646e367cd12659b6f Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Thu, 27 Mar 2008 23:55:22 +0300
Subject: x86: move generic_processor_info to apic_32.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_32.c    | 82 ++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/mpparse_32.c | 81 -------------------------------------------
 include/asm-x86/mpspec.h     |  1 +
 3 files changed, 83 insertions(+), 81 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index 65036bbaf058..a99398f71234 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1469,6 +1469,88 @@ void disconnect_bsp_APIC(int virt_wire_setup)
 	}
 }
 
+unsigned int __cpuinitdata maxcpus = NR_CPUS;
+
+void __cpuinit generic_processor_info(int apicid, int version)
+{
+	int cpu;
+	cpumask_t tmp_map;
+	physid_mask_t phys_cpu;
+
+	/*
+	 * Validate version
+	 */
+	if (version == 0x0) {
+		printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
+				"fixing up to 0x10. (tell your hw vendor)\n",
+				version);
+		version = 0x10;
+	}
+	apic_version[apicid] = version;
+
+	phys_cpu = apicid_to_cpu_present(apicid);
+	physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu);
+
+	if (num_processors >= NR_CPUS) {
+		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
+			"  Processor ignored.\n", NR_CPUS);
+		return;
+	}
+
+	if (num_processors >= maxcpus) {
+		printk(KERN_WARNING "WARNING: maxcpus limit of %i reached."
+			" Processor ignored.\n", maxcpus);
+		return;
+	}
+
+	num_processors++;
+	cpus_complement(tmp_map, cpu_present_map);
+	cpu = first_cpu(tmp_map);
+
+	if (apicid == boot_cpu_physical_apicid)
+		/*
+		 * x86_bios_cpu_apicid is required to have processors listed
+		 * in same order as logical cpu numbers. Hence the first
+		 * entry is BSP, and so on.
+		 */
+		cpu = 0;
+
+	/*
+	 * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
+	 * but we need to work other dependencies like SMP_SUSPEND etc
+	 * before this can be done without some confusion.
+	 * if (CPU_HOTPLUG_ENABLED || num_processors > 8)
+	 *       - Ashok Raj <ashok.raj@intel.com>
+	 */
+	if (num_processors > 8) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_INTEL:
+			if (!APIC_XAPIC(version)) {
+				def_to_bigsmp = 0;
+				break;
+			}
+			/* If P4 and above fall through */
+		case X86_VENDOR_AMD:
+			def_to_bigsmp = 1;
+		}
+	}
+#ifdef CONFIG_SMP
+	/* are we being called early in kernel startup? */
+	if (x86_cpu_to_apicid_early_ptr) {
+		u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr;
+		u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr;
+
+		cpu_to_apicid[cpu] = apicid;
+		bios_cpu_apicid[cpu] = apicid;
+	} else {
+		per_cpu(x86_cpu_to_apicid, cpu) = apicid;
+		per_cpu(x86_bios_cpu_apicid, cpu) = apicid;
+	}
+#endif
+	cpu_set(cpu, cpu_possible_map);
+	cpu_set(cpu, cpu_present_map);
+}
+
 /*
  * Power management
  */
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index c79d6e06c3fa..cd4522b3e90e 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -35,7 +35,6 @@
 
 /* Have we found an MP table */
 int smp_found_config;
-unsigned int __cpuinitdata maxcpus = NR_CPUS;
 
 /*
  * Various Linux-internal data structures created from the
@@ -105,86 +104,6 @@ static int mpc_record;
 static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __cpuinitdata;
 #endif
 
-static void __cpuinit generic_processor_info(int apicid, int version)
-{
-	int cpu;
-	cpumask_t tmp_map;
-	physid_mask_t phys_cpu;
-
-	/*
-	 * Validate version
-	 */
-	if (version == 0x0) {
-		printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
-				"fixing up to 0x10. (tell your hw vendor)\n",
-				version);
-		version = 0x10;
-	}
-	apic_version[apicid] = version;
-
-	phys_cpu = apicid_to_cpu_present(apicid);
-	physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu);
-
-	if (num_processors >= NR_CPUS) {
-		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
-			"  Processor ignored.\n", NR_CPUS);
-		return;
-	}
-
-	if (num_processors >= maxcpus) {
-		printk(KERN_WARNING "WARNING: maxcpus limit of %i reached."
-			" Processor ignored.\n", maxcpus);
-		return;
-	}
-
-	num_processors++;
-	cpus_complement(tmp_map, cpu_present_map);
-	cpu = first_cpu(tmp_map);
-
-	if (apicid == boot_cpu_physical_apicid)
-		/*
-		 * x86_bios_cpu_apicid is required to have processors listed
-		 * in same order as logical cpu numbers. Hence the first
-		 * entry is BSP, and so on.
-		 */
-		cpu = 0;
-
-	/*
-	 * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
-	 * but we need to work other dependencies like SMP_SUSPEND etc
-	 * before this can be done without some confusion.
-	 * if (CPU_HOTPLUG_ENABLED || num_processors > 8)
-	 *       - Ashok Raj <ashok.raj@intel.com>
-	 */
-	if (num_processors > 8) {
-		switch (boot_cpu_data.x86_vendor) {
-		case X86_VENDOR_INTEL:
-			if (!APIC_XAPIC(version)) {
-				def_to_bigsmp = 0;
-				break;
-			}
-			/* If P4 and above fall through */
-		case X86_VENDOR_AMD:
-			def_to_bigsmp = 1;
-		}
-	}
-#ifdef CONFIG_SMP
-	/* are we being called early in kernel startup? */
-	if (x86_cpu_to_apicid_early_ptr) {
-		u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr;
-		u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr;
-
-		cpu_to_apicid[cpu] = apicid;
-		bios_cpu_apicid[cpu] = apicid;
-	} else {
-		per_cpu(x86_cpu_to_apicid, cpu) = apicid;
-		per_cpu(x86_bios_cpu_apicid, cpu) = apicid;
-	}
-#endif
-	cpu_set(cpu, cpu_possible_map);
-	cpu_set(cpu, cpu_present_map);
-}
-
 static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
 {
 	int apicid;
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index eccbc581ec84..7ee54d396d60 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -43,6 +43,7 @@ extern unsigned long mp_lapic_addr;
 extern void find_smp_config(void);
 extern void get_smp_config(void);
 
+void __cpuinit generic_processor_info(int apicid, int version);
 #ifdef CONFIG_ACPI
 extern void mp_register_lapic(u8 id, u8 enabled);
 extern void mp_register_lapic_address(u64 address);
-- 
cgit v1.2.3


From 86c9835b46605fb29a3c30c6cc344d9df49e54a3 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 28 Mar 2008 11:59:57 +0100
Subject: x86: mpparse, move generic processor info to apic_32.c fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mpspec.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 7ee54d396d60..31bac12a97d1 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -1,6 +1,8 @@
 #ifndef _AM_X86_MPSPEC_H
 #define _AM_X86_MPSPEC_H
 
+#include <linux/init.h>
+
 #include <asm/mpspec_def.h>
 
 #ifdef CONFIG_X86_32
-- 
cgit v1.2.3


From fe874b3edff43f9a74d9903eb3710e5e0511faf1 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:05:56 -0300
Subject: x86: surround hard_smp_processor_id in APIC_DEFINITION

APIC_DEFINITION is not defined in x86_64, so in practice, we keep
our old code here. But as a nice side effect, the code is now
equal to smp_32.h.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp_64.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index c53a011bb91c..25206333476b 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -52,12 +52,16 @@ static inline int logical_smp_processor_id(void)
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
 }
 
-#include <mach_apicdef.h>
+# ifdef APIC_DEFINITION
+extern int hard_smp_processor_id(void);
+# else
+#  include <mach_apicdef.h>
 static inline int hard_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
 	return GET_APIC_ID(*(u32 *)(APIC_BASE + APIC_ID));
 }
+# endif /* APIC_DEFINITION */
 
 #endif
 
-- 
cgit v1.2.3


From 2ba95bcbe68d692f549fb10809f15681a25ff6fb Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:05:57 -0300
Subject: x86: provide bogus hard_smp_processor_id

We provide a bogus macro for x86_64 in case CONFIG_X86_LOCAL_APIC
is not set. It will always be set for x86_64, so the effect
is just to make the code equal to i386.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp_64.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 25206333476b..c46585e09ea1 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -46,6 +46,8 @@ static inline int num_booting_cpus(void)
 
 #define safe_smp_processor_id()		smp_processor_id()
 
+#ifdef CONFIG_X86_LOCAL_APIC
+
 static inline int logical_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
@@ -63,5 +65,13 @@ static inline int hard_smp_processor_id(void)
 }
 # endif /* APIC_DEFINITION */
 
+#else /* CONFIG_X86_LOCAL_APIC */
+
+# ifndef CONFIG_SMP
+#  define hard_smp_processor_id()	0
+# endif
+
+#endif /* CONFIG_X86_LOCAL_APIC */
+
 #endif
 
-- 
cgit v1.2.3


From 1b00084386878f25c2c591ad19cb625880d4089d Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:05:58 -0300
Subject: x86: merge hard/logical_smp_processor_id

The code is now the same between i386 and x86_64. We already
know what happens when it reaches this point: They go away
from the arch-specific headers, and suddenly appears in the common
header.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 27 +++++++++++++++++++++++++++
 include/asm-x86/smp_32.h | 27 ---------------------------
 include/asm-x86/smp_64.h | 27 ---------------------------
 3 files changed, 27 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index ef26911dc22a..e5534f19c312 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -123,6 +123,33 @@ void smp_store_cpu_info(int id);
 # include "smp_64.h"
 #endif
 
+#ifdef CONFIG_X86_LOCAL_APIC
+
+static inline int logical_smp_processor_id(void)
+{
+	/* we don't want to mark this access volatile - bad code generation */
+	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
+}
+
+# ifdef APIC_DEFINITION
+extern int hard_smp_processor_id(void);
+# else
+#  include <mach_apicdef.h>
+static inline int hard_smp_processor_id(void)
+{
+	/* we don't want to mark this access volatile - bad code generation */
+	return GET_APIC_ID(*(u32 *)(APIC_BASE + APIC_ID));
+}
+# endif /* APIC_DEFINITION */
+
+#else /* CONFIG_X86_LOCAL_APIC */
+
+# ifndef CONFIG_SMP
+#  define hard_smp_processor_id()	0
+# endif
+
+#endif /* CONFIG_X86_LOCAL_APIC */
+
 #ifdef CONFIG_HOTPLUG_CPU
 extern void cpu_exit_clear(void);
 extern void cpu_uninit(void);
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index cb3ada2fedbf..53432dbd5426 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -42,32 +42,5 @@ static inline int num_booting_cpus(void)
 #define safe_smp_processor_id()		0
 #endif /* !CONFIG_SMP */
 
-#ifdef CONFIG_X86_LOCAL_APIC
-
-static inline int logical_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
-}
-
-# ifdef APIC_DEFINITION
-extern int hard_smp_processor_id(void);
-# else
-#  include <mach_apicdef.h>
-static inline int hard_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_ID(*(u32 *)(APIC_BASE + APIC_ID));
-}
-# endif /* APIC_DEFINITION */
-
-#else /* CONFIG_X86_LOCAL_APIC */
-
-# ifndef CONFIG_SMP
-#  define hard_smp_processor_id()	0
-# endif
-
-#endif /* CONFIG_X86_LOCAL_APIC */
-
 #endif /* !ASSEMBLY */
 #endif
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index c46585e09ea1..015d36e29ade 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -46,32 +46,5 @@ static inline int num_booting_cpus(void)
 
 #define safe_smp_processor_id()		smp_processor_id()
 
-#ifdef CONFIG_X86_LOCAL_APIC
-
-static inline int logical_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
-}
-
-# ifdef APIC_DEFINITION
-extern int hard_smp_processor_id(void);
-# else
-#  include <mach_apicdef.h>
-static inline int hard_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_ID(*(u32 *)(APIC_BASE + APIC_ID));
-}
-# endif /* APIC_DEFINITION */
-
-#else /* CONFIG_X86_LOCAL_APIC */
-
-# ifndef CONFIG_SMP
-#  define hard_smp_processor_id()	0
-# endif
-
-#endif /* CONFIG_X86_LOCAL_APIC */
-
 #endif
 
-- 
cgit v1.2.3


From c1fa6c977eb978e1d09867475ec59c9a5799127f Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:05:59 -0300
Subject: x86: surround apic headers in apic definitions

Although those constants are always defined in x86_64,
and will have the effect of just including the headers
in the very way we did before, I'm doing this in a separate
patch to be conservative and avoid surprises.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp_64.h | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 015d36e29ade..b83151d7388c 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -7,9 +7,13 @@
 /*
  * We need the APIC definitions automatically as part of 'smp.h'
  */
-#include <asm/apic.h>
-#include <asm/io_apic.h>
-#include <asm/mpspec.h>
+#ifdef CONFIG_X86_LOCAL_APIC
+#  include <asm/mpspec.h>
+#  include <asm/apic.h>
+#  ifdef CONFIG_X86_IO_APIC
+#   include <asm/io_apic.h>
+#  endif
+#endif
 #include <asm/pda.h>
 #include <asm/thread_info.h>
 
-- 
cgit v1.2.3


From b23dab08fa37b302a8980e4cf925f2cb94288538 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:06:00 -0300
Subject: x86: merge includes in smp.h

move all include directives from smp_{32,64}.h to smp.h.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 13 +++++++++++++
 include/asm-x86/smp_32.h | 13 -------------
 include/asm-x86/smp_64.h | 16 ----------------
 3 files changed, 13 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index e5534f19c312..21472cea3d6c 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -5,6 +5,19 @@
 #include <linux/init.h>
 #include <asm/percpu.h>
 
+/*
+ * We need the APIC definitions automatically as part of 'smp.h'
+ */
+#ifdef CONFIG_X86_LOCAL_APIC
+# include <asm/mpspec.h>
+# include <asm/apic.h>
+# ifdef CONFIG_X86_IO_APIC
+#  include <asm/io_apic.h>
+# endif
+#endif
+#include <asm/pda.h>
+#include <asm/thread_info.h>
+
 extern cpumask_t cpu_callout_map;
 
 extern int smp_num_siblings;
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 53432dbd5426..694d3245a88f 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -2,19 +2,6 @@
 #define __ASM_SMP_H
 
 #ifndef __ASSEMBLY__
-#include <linux/cpumask.h>
-#include <linux/init.h>
-
-/*
- * We need the APIC definitions automatically as part of 'smp.h'
- */
-#ifdef CONFIG_X86_LOCAL_APIC
-# include <asm/mpspec.h>
-# include <asm/apic.h>
-# ifdef CONFIG_X86_IO_APIC
-#  include <asm/io_apic.h>
-# endif
-#endif
 
 extern cpumask_t cpu_callin_map;
 
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index b83151d7388c..eead92e30b29 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -1,22 +1,6 @@
 #ifndef __ASM_SMP_H
 #define __ASM_SMP_H
 
-#include <linux/cpumask.h>
-#include <linux/init.h>
-
-/*
- * We need the APIC definitions automatically as part of 'smp.h'
- */
-#ifdef CONFIG_X86_LOCAL_APIC
-#  include <asm/mpspec.h>
-#  include <asm/apic.h>
-#  ifdef CONFIG_X86_IO_APIC
-#   include <asm/io_apic.h>
-#  endif
-#endif
-#include <asm/pda.h>
-#include <asm/thread_info.h>
-
 extern cpumask_t cpu_initialized;
 extern cpumask_t cpu_callin_map;
 
-- 
cgit v1.2.3


From 24e8ecffa84dd560e0d4d6fcaeca6950805854e7 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:06:01 -0300
Subject: x86: split safe_smp_processor_id

This implementation in x86_64 is clean and consistent, but we
sacrifice it for the sake of being equal to i386 (since the other
way around would be harder).

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp_64.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index eead92e30b29..8ea49529f324 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -27,12 +27,12 @@ static inline int num_booting_cpus(void)
 	return cpus_weight(cpu_callout_map);
 }
 
+#define safe_smp_processor_id()		smp_processor_id()
 #else /* CONFIG_SMP */
 #define stack_smp_processor_id() 0
-
+#define safe_smp_processor_id() 0
 #endif /* !CONFIG_SMP */
 
-#define safe_smp_processor_id()		smp_processor_id()
 
 #endif
 
-- 
cgit v1.2.3


From a9c057c1d1b1080a01004ecac54308365e167b83 Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:06:02 -0300
Subject: x86: merge SMP definitions of smp.h

we merge everything that is inside CONFIG_SMP
to smp.h. They differ a little bit, so we use
CONFIG_X86_32_SMP and CONFIG_X86_64_SMP as markers.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 33 ++++++++++++++++++++++++++++++++-
 include/asm-x86/smp_32.h | 21 ---------------------
 include/asm-x86/smp_64.h | 27 ---------------------------
 3 files changed, 32 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 21472cea3d6c..57b3d86dd9ed 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -126,8 +126,39 @@ extern unsigned long setup_trampoline(void);
 
 void smp_store_cpu_info(int id);
 #define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
-#else
+
+/* We don't mark CPUs online until __cpu_up(), so we need another measure */
+static inline int num_booting_cpus(void)
+{
+	return cpus_weight(cpu_callout_map);
+}
+#endif /* CONFIG_SMP */
+
+#ifdef CONFIG_X86_32_SMP
+/*
+ * This function is needed by all SMP systems. It must _always_ be valid
+ * from the initial startup. We map APIC_BASE very early in page_setup(),
+ * so this is correct in the x86 case.
+ */
+DECLARE_PER_CPU(int, cpu_number);
+#define raw_smp_processor_id() (x86_read_percpu(cpu_number))
+extern int safe_smp_processor_id(void);
+
+#elif defined(CONFIG_X86_64_SMP)
+#define raw_smp_processor_id()	read_pda(cpunumber)
+
+#define stack_smp_processor_id()					\
+({								\
+	struct thread_info *ti;						\
+	__asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
+	ti->cpu;							\
+})
+#define safe_smp_processor_id()		smp_processor_id()
+
+#else /* !CONFIG_X86_32_SMP && !CONFIG_X86_64_SMP */
 #define cpu_physical_id(cpu)		boot_cpu_physical_apicid
+#define safe_smp_processor_id()		0
+#define stack_smp_processor_id() 	0
 #endif
 
 #ifdef CONFIG_X86_32
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
index 694d3245a88f..d9ae5ac93dfc 100644
--- a/include/asm-x86/smp_32.h
+++ b/include/asm-x86/smp_32.h
@@ -8,26 +8,5 @@ extern cpumask_t cpu_callin_map;
 extern void (*mtrr_hook)(void);
 extern void zap_low_mappings(void);
 
-#ifdef CONFIG_SMP
-/*
- * This function is needed by all SMP systems. It must _always_ be valid
- * from the initial startup. We map APIC_BASE very early in page_setup(),
- * so this is correct in the x86 case.
- */
-DECLARE_PER_CPU(int, cpu_number);
-#define raw_smp_processor_id() (x86_read_percpu(cpu_number))
-
-extern int safe_smp_processor_id(void);
-
-/* We don't mark CPUs online until __cpu_up(), so we need another measure */
-static inline int num_booting_cpus(void)
-{
-	return cpus_weight(cpu_callout_map);
-}
-
-#else /* CONFIG_SMP */
-#define safe_smp_processor_id()		0
-#endif /* !CONFIG_SMP */
-
 #endif /* !ASSEMBLY */
 #endif
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
index 8ea49529f324..058f41399798 100644
--- a/include/asm-x86/smp_64.h
+++ b/include/asm-x86/smp_64.h
@@ -7,32 +7,5 @@ extern cpumask_t cpu_callin_map;
 extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 				  void *info, int wait);
 
-#ifdef CONFIG_SMP
-
-#define raw_smp_processor_id()	read_pda(cpunumber)
-
-#define stack_smp_processor_id()					\
-({									\
-	struct thread_info *ti;						\
-	asm("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
-	ti->cpu;							\
-})
-
-/*
- * On x86 all CPUs are mapped 1:1 to the APIC space. This simplifies
- * scheduling and IPI sending and compresses data structures.
- */
-static inline int num_booting_cpus(void)
-{
-	return cpus_weight(cpu_callout_map);
-}
-
-#define safe_smp_processor_id()		smp_processor_id()
-#else /* CONFIG_SMP */
-#define stack_smp_processor_id() 0
-#define safe_smp_processor_id() 0
-#endif /* !CONFIG_SMP */
-
-
 #endif
 
-- 
cgit v1.2.3


From 8be9ac850564a409c1238cd5f53776c340aea4dc Mon Sep 17 00:00:00 2001
From: Glauber Costa <gcosta@redhat.com>
Date: Thu, 27 Mar 2008 14:06:04 -0300
Subject: x86: merge smp_32.h and smp_64.h into smp.h

Merge what's left from smp_32.h and smp_64.h into smp.h
By now, they're basically extern definitions.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h    | 11 +++++------
 include/asm-x86/smp_32.h | 12 ------------
 include/asm-x86/smp_64.h | 11 -----------
 3 files changed, 5 insertions(+), 29 deletions(-)
 delete mode 100644 include/asm-x86/smp_32.h
 delete mode 100644 include/asm-x86/smp_64.h

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 57b3d86dd9ed..bcbd25cbd863 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -19,6 +19,11 @@
 #include <asm/thread_info.h>
 
 extern cpumask_t cpu_callout_map;
+extern cpumask_t cpu_initialized;
+extern cpumask_t cpu_callin_map;
+
+extern void (*mtrr_hook)(void);
+extern void zap_low_mappings(void);
 
 extern int smp_num_siblings;
 extern unsigned int num_processors;
@@ -161,12 +166,6 @@ extern int safe_smp_processor_id(void);
 #define stack_smp_processor_id() 	0
 #endif
 
-#ifdef CONFIG_X86_32
-# include "smp_32.h"
-#else
-# include "smp_64.h"
-#endif
-
 #ifdef CONFIG_X86_LOCAL_APIC
 
 static inline int logical_smp_processor_id(void)
diff --git a/include/asm-x86/smp_32.h b/include/asm-x86/smp_32.h
deleted file mode 100644
index d9ae5ac93dfc..000000000000
--- a/include/asm-x86/smp_32.h
+++ /dev/null
@@ -1,12 +0,0 @@
-#ifndef __ASM_SMP_H
-#define __ASM_SMP_H
-
-#ifndef __ASSEMBLY__
-
-extern cpumask_t cpu_callin_map;
-
-extern void (*mtrr_hook)(void);
-extern void zap_low_mappings(void);
-
-#endif /* !ASSEMBLY */
-#endif
diff --git a/include/asm-x86/smp_64.h b/include/asm-x86/smp_64.h
deleted file mode 100644
index 058f41399798..000000000000
--- a/include/asm-x86/smp_64.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef __ASM_SMP_H
-#define __ASM_SMP_H
-
-extern cpumask_t cpu_initialized;
-extern cpumask_t cpu_callin_map;
-
-extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
-				  void *info, int wait);
-
-#endif
-
-- 
cgit v1.2.3


From fb8e8375394e1156a5a1e7ba53504b141d2365a8 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Thu, 27 Mar 2008 17:28:39 -0700
Subject: x86: sparsemem: reduce i386 PAE section size

A 1G section size makes memory hotplug too coarse in a virtual
environment.  Retuce it by a factor of 2 to 512M.  I would have liked
to make it smaller, but it runs out of reserved flags in the page flags.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/sparsemem.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/sparsemem.h b/include/asm-x86/sparsemem.h
index fa684f353aa7..9bd48b0a534b 100644
--- a/include/asm-x86/sparsemem.h
+++ b/include/asm-x86/sparsemem.h
@@ -16,7 +16,7 @@
 
 #ifdef CONFIG_X86_32
 # ifdef CONFIG_X86_PAE
-#  define SECTION_SIZE_BITS	30
+#  define SECTION_SIZE_BITS	29
 #  define MAX_PHYSADDR_BITS	36
 #  define MAX_PHYSMEM_BITS	36
 # else
-- 
cgit v1.2.3


From 6093015db2bd9e70cf20cdd23be1a50733baafdd Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 30 Mar 2008 11:45:23 +0200
Subject: x86: cleanup replace most vm86 flags with flags from
 processor-flags.h, fix

- fix build error
- fix CONFIG_HEADERS_CHECK error

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/traps_32.c | 2 +-
 include/asm-x86/Kbuild     | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 57a5704e3f6c..3284502a1bf8 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -959,7 +959,7 @@ debug_vm86:
 
 clear_TF_reenable:
 	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
-	regs->flags &= ~TF_MASK;
+	regs->flags &= ~X86_EFLAGS_TF;
 	return;
 }
 
diff --git a/include/asm-x86/Kbuild b/include/asm-x86/Kbuild
index 3b8160a2b47e..1e3554596f72 100644
--- a/include/asm-x86/Kbuild
+++ b/include/asm-x86/Kbuild
@@ -10,6 +10,7 @@ header-y += prctl.h
 header-y += ptrace-abi.h
 header-y += sigcontext32.h
 header-y += ucontext.h
+header-y += processor-flags.h
 
 unifdef-y += e820.h
 unifdef-y += ist.h
-- 
cgit v1.2.3


From 6b6891f9c545ccd45d6d8ddfd33ce27c22c271a7 Mon Sep 17 00:00:00 2001
From: "gorcunov@gmail.com" <gorcunov@gmail.com>
Date: Fri, 28 Mar 2008 17:56:57 +0300
Subject: x86: cleanup - rename VM_MASK to X86_VM_MASK

This patch renames VM_MASK to X86_VM_MASK (which
in turn defined as alias to X86_EFLAGS_VM) to better
distinguish from virtual memory flags. We can't just
use X86_EFLAGS_VM instead because it is also used
for conditional compilation

Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/traps_32.c | 8 ++++----
 arch/x86/kernel/vm86_32.c  | 2 +-
 arch/x86/mm/fault.c        | 2 +-
 include/asm-x86/ptrace.h   | 4 ++--
 include/asm-x86/vm86.h     | 4 ++--
 5 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index 3284502a1bf8..bb9107c56ff5 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -498,7 +498,7 @@ do_trap(int trapnr, int signr, char *str, int vm86, struct pt_regs *regs,
 {
 	struct task_struct *tsk = current;
 
-	if (regs->flags & VM_MASK) {
+	if (regs->flags & X86_VM_MASK) {
 		if (vm86)
 			goto vm86_trap;
 		goto trap_signal;
@@ -643,7 +643,7 @@ void __kprobes do_general_protection(struct pt_regs *regs, long error_code)
 	}
 	put_cpu();
 
-	if (regs->flags & VM_MASK)
+	if (regs->flags & X86_VM_MASK)
 		goto gp_in_vm86;
 
 	if (!user_mode(regs))
@@ -922,7 +922,7 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code)
 			goto clear_dr7;
 	}
 
-	if (regs->flags & VM_MASK)
+	if (regs->flags & X86_VM_MASK)
 		goto debug_vm86;
 
 	/* Save debug status register where ptrace can see it */
@@ -1094,7 +1094,7 @@ void do_simd_coprocessor_error(struct pt_regs *regs, long error_code)
 	 * Handle strange cache flush from user space exception
 	 * in all other cases.  This is undocumented behaviour.
 	 */
-	if (regs->flags & VM_MASK) {
+	if (regs->flags & X86_VM_MASK) {
 		handle_vm86_fault((struct kernel_vm86_regs *)regs, error_code);
 		return;
 	}
diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c
index 51040698c222..c866c00f4a85 100644
--- a/arch/x86/kernel/vm86_32.c
+++ b/arch/x86/kernel/vm86_32.c
@@ -299,7 +299,7 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
 	VEFLAGS = info->regs.pt.flags;
 	info->regs.pt.flags &= SAFE_MASK;
 	info->regs.pt.flags |= info->regs32->flags & ~SAFE_MASK;
-	info->regs.pt.flags |= VM_MASK;
+	info->regs.pt.flags |= X86_VM_MASK;
 
 	switch (info->cpu_type) {
 	case CPU_286:
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 81fcbeec3892..fd7e1798c75a 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -639,7 +639,7 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 #ifdef CONFIG_X86_32
 	/* It's safe to allow irq's after cr2 has been saved and the vmalloc
 	   fault has been handled. */
-	if (regs->flags & (X86_EFLAGS_IF|VM_MASK))
+	if (regs->flags & (X86_EFLAGS_IF | X86_VM_MASK))
 		local_irq_enable();
 
 	/*
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index e779f2b26b32..24ec061566c5 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -170,7 +170,7 @@ static inline int user_mode(struct pt_regs *regs)
 static inline int user_mode_vm(struct pt_regs *regs)
 {
 #ifdef CONFIG_X86_32
-	return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & VM_MASK)) >=
+	return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & X86_VM_MASK)) >=
 		USER_RPL;
 #else
 	return user_mode(regs);
@@ -180,7 +180,7 @@ static inline int user_mode_vm(struct pt_regs *regs)
 static inline int v8086_mode(struct pt_regs *regs)
 {
 #ifdef CONFIG_X86_32
-	return (regs->flags & VM_MASK);
+	return (regs->flags & X86_VM_MASK);
 #else
 	return 0;	/* No V86 mode support in long mode */
 #endif
diff --git a/include/asm-x86/vm86.h b/include/asm-x86/vm86.h
index a2be241ed036..f5f3dc479c34 100644
--- a/include/asm-x86/vm86.h
+++ b/include/asm-x86/vm86.h
@@ -17,9 +17,9 @@
 #define IOPL_MASK	0x00003000
 #define NT_MASK		0x00004000
 #ifdef CONFIG_VM86
-#define VM_MASK		0x00020000
+#define X86_VM_MASK	X86_EFLAGS_VM
 #else
-#define VM_MASK		0 /* ignored */
+#define X86_VM_MASK	0 /* No VM86 support */
 #endif
 #define AC_MASK		0x00040000
 #define VIF_MASK	0x00080000	/* virtual interrupt flag */
-- 
cgit v1.2.3


From a5c15d419d4b68535222b51f9054dd08d5e67470 Mon Sep 17 00:00:00 2001
From: "gorcunov@gmail.com" <gorcunov@gmail.com>
Date: Fri, 28 Mar 2008 17:56:56 +0300
Subject: x86: replace most VM86 flags with flags from processor-flags.h

Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/kprobes.c   |  2 +-
 arch/x86/kernel/signal_32.c |  4 ++--
 arch/x86/kernel/vm86_32.c   | 34 +++++++++++++++++-----------------
 include/asm-x86/vm86.h      | 10 ++--------
 4 files changed, 22 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index cc8ae90103ff..b8c6743a13da 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -489,7 +489,7 @@ static int __kprobes reenter_kprobe(struct kprobe *p, struct pt_regs *regs,
 		break;
 	case KPROBE_HIT_SS:
 		if (p == kprobe_running()) {
-			regs->flags &= ~TF_MASK;
+			regs->flags &= ~X86_EFLAGS_TF;
 			regs->flags |= kcb->kprobe_saved_flags;
 			return 0;
 		} else {
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index aa1b6a0a22e4..f1b117930837 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -419,7 +419,7 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 	 * The tracer may want to single-step inside the
 	 * handler too.
 	 */
-	regs->flags &= ~(TF_MASK | X86_EFLAGS_DF);
+	regs->flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_DF);
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
@@ -507,7 +507,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	 * The tracer may want to single-step inside the
 	 * handler too.
 	 */
-	regs->flags &= ~(TF_MASK | X86_EFLAGS_DF);
+	regs->flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_DF);
 	if (test_thread_flag(TIF_SINGLESTEP))
 		ptrace_notify(SIGTRAP);
 
diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c
index c866c00f4a85..38f566fa27d2 100644
--- a/arch/x86/kernel/vm86_32.c
+++ b/arch/x86/kernel/vm86_32.c
@@ -139,7 +139,7 @@ struct pt_regs *save_v86_state(struct kernel_vm86_regs *regs)
 		printk("no vm86_info: BAD\n");
 		do_exit(SIGSEGV);
 	}
-	set_flags(regs->pt.flags, VEFLAGS, VIF_MASK | current->thread.v86mask);
+	set_flags(regs->pt.flags, VEFLAGS, X86_EFLAGS_VIF | current->thread.v86mask);
 	tmp = copy_vm86_regs_to_user(&current->thread.vm86_info->regs, regs);
 	tmp += put_user(current->thread.screen_bitmap, &current->thread.vm86_info->screen_bitmap);
 	if (tmp) {
@@ -306,13 +306,13 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk
 		tsk->thread.v86mask = 0;
 		break;
 	case CPU_386:
-		tsk->thread.v86mask = NT_MASK | IOPL_MASK;
+		tsk->thread.v86mask = X86_EFLAGS_NT | X86_EFLAGS_IOPL;
 		break;
 	case CPU_486:
-		tsk->thread.v86mask = AC_MASK | NT_MASK | IOPL_MASK;
+		tsk->thread.v86mask = X86_EFLAGS_AC | X86_EFLAGS_NT | X86_EFLAGS_IOPL;
 		break;
 	default:
-		tsk->thread.v86mask = ID_MASK | AC_MASK | NT_MASK | IOPL_MASK;
+		tsk->thread.v86mask = X86_EFLAGS_ID | X86_EFLAGS_AC | X86_EFLAGS_NT | X86_EFLAGS_IOPL;
 		break;
 	}
 
@@ -363,24 +363,24 @@ static inline void return_to_32bit(struct kernel_vm86_regs *regs16, int retval)
 
 static inline void set_IF(struct kernel_vm86_regs *regs)
 {
-	VEFLAGS |= VIF_MASK;
-	if (VEFLAGS & VIP_MASK)
+	VEFLAGS |= X86_EFLAGS_VIF;
+	if (VEFLAGS & X86_EFLAGS_VIP)
 		return_to_32bit(regs, VM86_STI);
 }
 
 static inline void clear_IF(struct kernel_vm86_regs *regs)
 {
-	VEFLAGS &= ~VIF_MASK;
+	VEFLAGS &= ~X86_EFLAGS_VIF;
 }
 
 static inline void clear_TF(struct kernel_vm86_regs *regs)
 {
-	regs->pt.flags &= ~TF_MASK;
+	regs->pt.flags &= ~X86_EFLAGS_TF;
 }
 
 static inline void clear_AC(struct kernel_vm86_regs *regs)
 {
-	regs->pt.flags &= ~AC_MASK;
+	regs->pt.flags &= ~X86_EFLAGS_AC;
 }
 
 /*
@@ -399,7 +399,7 @@ static inline void set_vflags_long(unsigned long flags, struct kernel_vm86_regs
 {
 	set_flags(VEFLAGS, flags, current->thread.v86mask);
 	set_flags(regs->pt.flags, flags, SAFE_MASK);
-	if (flags & IF_MASK)
+	if (flags & X86_EFLAGS_IF)
 		set_IF(regs);
 	else
 		clear_IF(regs);
@@ -409,7 +409,7 @@ static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_reg
 {
 	set_flags(VFLAGS, flags, current->thread.v86mask);
 	set_flags(regs->pt.flags, flags, SAFE_MASK);
-	if (flags & IF_MASK)
+	if (flags & X86_EFLAGS_IF)
 		set_IF(regs);
 	else
 		clear_IF(regs);
@@ -419,9 +419,9 @@ static inline unsigned long get_vflags(struct kernel_vm86_regs *regs)
 {
 	unsigned long flags = regs->pt.flags & RETURN_MASK;
 
-	if (VEFLAGS & VIF_MASK)
-		flags |= IF_MASK;
-	flags |= IOPL_MASK;
+	if (VEFLAGS & X86_EFLAGS_VIF)
+		flags |= X86_EFLAGS_IF;
+	flags |= X86_EFLAGS_IOPL;
 	return flags | (VEFLAGS & current->thread.v86mask);
 }
 
@@ -573,11 +573,11 @@ void handle_vm86_fault(struct kernel_vm86_regs *regs, long error_code)
 
 #define CHECK_IF_IN_TRAP \
 	if (VMPI.vm86dbg_active && VMPI.vm86dbg_TFpendig) \
-		newflags |= TF_MASK
+		newflags |= X86_EFLAGS_TF
 #define VM86_FAULT_RETURN do { \
-	if (VMPI.force_return_for_pic  && (VEFLAGS & (IF_MASK | VIF_MASK))) \
+	if (VMPI.force_return_for_pic  && (VEFLAGS & (X86_EFLAGS_IF | X86_EFLAGS_VIF))) \
 		return_to_32bit(regs, VM86_PICRETURN); \
-	if (orig_flags & TF_MASK) \
+	if (orig_flags & X86_EFLAGS_TF) \
 		handle_vm86_trap(regs, 0, 1); \
 	return; } while (0)
 
diff --git a/include/asm-x86/vm86.h b/include/asm-x86/vm86.h
index f5f3dc479c34..074b357146df 100644
--- a/include/asm-x86/vm86.h
+++ b/include/asm-x86/vm86.h
@@ -12,19 +12,13 @@
  * Linus
  */
 
-#define TF_MASK		0x00000100
-#define IF_MASK		0x00000200
-#define IOPL_MASK	0x00003000
-#define NT_MASK		0x00004000
+#include <asm/processor-flags.h>
+
 #ifdef CONFIG_VM86
 #define X86_VM_MASK	X86_EFLAGS_VM
 #else
 #define X86_VM_MASK	0 /* No VM86 support */
 #endif
-#define AC_MASK		0x00040000
-#define VIF_MASK	0x00080000	/* virtual interrupt flag */
-#define VIP_MASK	0x00100000	/* virtual interrupt pending */
-#define ID_MASK		0x00200000
 
 #define BIOSSEG		0x0f000
 
-- 
cgit v1.2.3


From 05f2d12c3563dea8c81b301f9f3cf7919af23b13 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 28 Mar 2008 14:12:02 -0500
Subject: x86: change GET_APIC_ID() from an inline function to an out-of-line
 function

Introduce a function to read the local APIC_ID.

This change is in preparation for additional changes to
the APICID functions that will come in a later patch.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_32.c                |  4 ++--
 arch/x86/kernel/apic_64.c                | 10 +++++-----
 arch/x86/kernel/genapic_flat_64.c        |  2 +-
 arch/x86/kernel/io_apic_32.c             |  6 +++---
 arch/x86/kernel/io_apic_64.c             |  5 ++---
 arch/x86/kernel/mpparse_32.c             |  2 +-
 arch/x86/kernel/mpparse_64.c             |  2 +-
 arch/x86/kernel/smpboot.c                |  6 +++---
 include/asm-x86/mach-default/mach_apic.h |  2 +-
 include/asm-x86/mach-es7000/mach_apic.h  |  2 +-
 include/asm-x86/mach-visws/mach_apic.h   |  2 +-
 include/asm-x86/smp.h                    |  7 ++++++-
 12 files changed, 27 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index a99398f71234..4905a11b30e3 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1195,7 +1195,7 @@ void __init init_apic_mappings(void)
 	 * default configuration (or the MP table is broken).
 	 */
 	if (boot_cpu_physical_apicid == -1U)
-		boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
+		boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
 
 #ifdef CONFIG_X86_IO_APIC
 	{
@@ -1265,7 +1265,7 @@ int __init APIC_init_uniprocessor(void)
 	 * might be zero if read from MP tables. Get it from LAPIC.
 	 */
 #ifdef CONFIG_CRASH_DUMP
-	boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
+	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
 #endif
 	phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid);
 
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 4ee521ff0a3e..9b4cacdfd74f 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -650,10 +650,10 @@ int __init verify_local_APIC(void)
 	/*
 	 * The ID register is read/write in a real APIC.
 	 */
-	reg0 = apic_read(APIC_ID);
+	reg0 = read_apic_id();
 	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0);
 	apic_write(APIC_ID, reg0 ^ APIC_ID_MASK);
-	reg1 = apic_read(APIC_ID);
+	reg1 = read_apic_id();
 	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1);
 	apic_write(APIC_ID, reg0);
 	if (reg1 != (reg0 ^ APIC_ID_MASK))
@@ -892,7 +892,7 @@ void __init early_init_lapic_mapping(void)
 	 * Fetch the APIC ID of the BSP in case we have a
 	 * default configuration (or the MP table is broken).
 	 */
-	boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
+	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
 }
 
 /**
@@ -919,7 +919,7 @@ void __init init_apic_mappings(void)
 	 * Fetch the APIC ID of the BSP in case we have a
 	 * default configuration (or the MP table is broken).
 	 */
-	boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
+	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
 }
 
 /*
@@ -1140,7 +1140,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 
 	maxlvt = lapic_get_maxlvt();
 
-	apic_pm_state.apic_id = apic_read(APIC_ID);
+	apic_pm_state.apic_id = read_apic_id();
 	apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI);
 	apic_pm_state.apic_ldr = apic_read(APIC_LDR);
 	apic_pm_state.apic_dfr = apic_read(APIC_DFR);
diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c
index 07352b74bda6..6a44e8dace37 100644
--- a/arch/x86/kernel/genapic_flat_64.c
+++ b/arch/x86/kernel/genapic_flat_64.c
@@ -97,7 +97,7 @@ static void flat_send_IPI_all(int vector)
 
 static int flat_apic_id_registered(void)
 {
-	return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map);
+	return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map);
 }
 
 static unsigned int flat_cpu_mask_to_apicid(cpumask_t cpumask)
diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 8ed6eb967652..bfebe7a1966d 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -1482,8 +1482,8 @@ void /*__init*/ print_local_APIC(void * dummy)
 
 	printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
 		smp_processor_id(), hard_smp_processor_id());
-	v = apic_read(APIC_ID);
-	printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v, GET_APIC_ID(v));
+	printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v,
+			GET_APIC_ID(read_apic_id()));
 	v = apic_read(APIC_LVR);
 	printk(KERN_INFO "... APIC VERSION: %08x\n", v);
 	ver = GET_APIC_VERSION(v);
@@ -1692,7 +1692,7 @@ void disable_IO_APIC(void)
 		entry.delivery_mode   = dest_ExtINT; /* ExtInt */
 		entry.vector          = 0;
 		entry.dest.physical.physical_dest =
-					GET_APIC_ID(apic_read(APIC_ID));
+					GET_APIC_ID(read_apic_id());
 
 		/*
 		 * Add it to the IO-APIC irq-routing table:
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 6dd33628f28a..0ac92d6acf57 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1068,8 +1068,7 @@ void __apicdebuginit print_local_APIC(void * dummy)
 
 	printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
 		smp_processor_id(), hard_smp_processor_id());
-	v = apic_read(APIC_ID);
-	printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v, GET_APIC_ID(v));
+	printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v, GET_APIC_ID(read_apic_id()));
 	v = apic_read(APIC_LVR);
 	printk(KERN_INFO "... APIC VERSION: %08x\n", v);
 	ver = GET_APIC_VERSION(v);
@@ -1263,7 +1262,7 @@ void disable_IO_APIC(void)
 		entry.dest_mode       = 0; /* Physical */
 		entry.delivery_mode   = dest_ExtINT; /* ExtInt */
 		entry.vector          = 0;
-		entry.dest          = GET_APIC_ID(apic_read(APIC_ID));
+		entry.dest          = GET_APIC_ID(read_apic_id());
 
 		/*
 		 * Add it to the IO-APIC irq-routing table:
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index cd4522b3e90e..4b46a37e0634 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -802,7 +802,7 @@ void __init mp_register_lapic_address(u64 address)
 	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
 
 	if (boot_cpu_physical_apicid == -1U)
-		boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
+		boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
 
 	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
 }
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 5e789bdb34fa..29d2c40e54a2 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -631,7 +631,7 @@ void __init mp_register_lapic_address(u64 address)
 	mp_lapic_addr = (unsigned long)address;
 	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
 	if (boot_cpu_physical_apicid == -1U)
-		boot_cpu_physical_apicid  = GET_APIC_ID(apic_read(APIC_ID));
+		boot_cpu_physical_apicid  = GET_APIC_ID(read_apic_id());
 }
 void __cpuinit mp_register_lapic(u8 id, u8 enabled)
 {
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index f45d740b1b6a..5da35d2cdbd8 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -237,7 +237,7 @@ void __cpuinit smp_callin(void)
 	/*
 	 * (This works even if the APIC is not enabled.)
 	 */
-	phys_id = GET_APIC_ID(apic_read(APIC_ID));
+	phys_id = GET_APIC_ID(read_apic_id());
 	cpuid = smp_processor_id();
 	if (cpu_isset(cpuid, cpu_callin_map)) {
 		panic("%s: phys CPU#%d, CPU#%d already present??\n", __func__,
@@ -1205,9 +1205,9 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
 		return;
 	}
 
-	if (GET_APIC_ID(apic_read(APIC_ID)) != boot_cpu_physical_apicid) {
+	if (GET_APIC_ID(read_apic_id()) != boot_cpu_physical_apicid) {
 		panic("Boot APIC ID in local APIC unexpected (%d vs %d)",
-		     GET_APIC_ID(apic_read(APIC_ID)), boot_cpu_physical_apicid);
+		     GET_APIC_ID(read_apic_id()), boot_cpu_physical_apicid);
 		/* Or can we switch back to PIC here? */
 	}
 
diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h
index 1f56e7d5bfdd..14217a970c5e 100644
--- a/include/asm-x86/mach-default/mach_apic.h
+++ b/include/asm-x86/mach-default/mach_apic.h
@@ -54,7 +54,7 @@ static inline void init_apic_ldr(void)
 
 static inline int apic_id_registered(void)
 {
-	return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map);
+	return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map);
 }
 
 static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/mach-es7000/mach_apic.h
index 0137b6e142cc..fbc8ad256f5a 100644
--- a/include/asm-x86/mach-es7000/mach_apic.h
+++ b/include/asm-x86/mach-es7000/mach_apic.h
@@ -141,7 +141,7 @@ static inline void setup_portio_remap(void)
 extern unsigned int boot_cpu_physical_apicid;
 static inline int check_phys_apicid_present(int cpu_physical_apicid)
 {
-	boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
+	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
 	return (1);
 }
 
diff --git a/include/asm-x86/mach-visws/mach_apic.h b/include/asm-x86/mach-visws/mach_apic.h
index efac6f0d139f..a9ef33a8a995 100644
--- a/include/asm-x86/mach-visws/mach_apic.h
+++ b/include/asm-x86/mach-visws/mach_apic.h
@@ -23,7 +23,7 @@
 
 static inline int apic_id_registered(void)
 {
-	return physid_isset(GET_APIC_ID(apic_read(APIC_ID)), phys_cpu_present_map);
+	return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map);
 }
 
 /*
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index bcbd25cbd863..c0d693ca4357 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -174,6 +174,11 @@ static inline int logical_smp_processor_id(void)
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
 }
 
+static inline unsigned int read_apic_id(void)
+{
+	return *(u32 *)(APIC_BASE + APIC_ID);
+}
+
 # ifdef APIC_DEFINITION
 extern int hard_smp_processor_id(void);
 # else
@@ -181,7 +186,7 @@ extern int hard_smp_processor_id(void);
 static inline int hard_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_ID(*(u32 *)(APIC_BASE + APIC_ID));
+	return GET_APIC_ID(read_apic_id());
 }
 # endif /* APIC_DEFINITION */
 
-- 
cgit v1.2.3


From ae261868658773538ddda829c50224e5851c2342 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 28 Mar 2008 14:12:06 -0500
Subject: x86: add functions to determine if platform is a UV platform

Add functions that can be used to determine if an x86_64
system is a SGI "UV" system. UV systems come in 3 types and
are identified by the OEM ID in the MADT.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/boot.c  |  4 +---
 arch/x86/kernel/genapic_64.c | 25 +++++++++++++++++++++++++
 include/asm-x86/genapic_32.h |  5 +++++
 include/asm-x86/genapic_64.h |  5 +++++
 4 files changed, 36 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index e277c370246d..05878ac934db 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -56,9 +56,7 @@ EXPORT_SYMBOL(acpi_disabled);
 #ifdef	CONFIG_X86_64
 
 #include <asm/proto.h>
-
-static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; }
-
+#include <asm/genapic.h>
 
 #else				/* X86 */
 
diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c
index 4ae7b6440260..c873f60c74a6 100644
--- a/arch/x86/kernel/genapic_64.c
+++ b/arch/x86/kernel/genapic_64.c
@@ -33,6 +33,8 @@ EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
 struct genapic __read_mostly *genapic = &apic_flat;
 
+static enum uv_system_type uv_system_type;
+
 /*
  * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode.
  */
@@ -64,3 +66,26 @@ void send_IPI_self(int vector)
 {
 	__send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL);
 }
+
+int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	if (!strcmp(oem_id, "SGI")) {
+		if (!strcmp(oem_table_id, "UVL"))
+			uv_system_type = UV_LEGACY_APIC;
+		else if (!strcmp(oem_table_id, "UVX"))
+			uv_system_type = UV_X2APIC;
+		else if (!strcmp(oem_table_id, "UVH"))
+			uv_system_type = UV_NON_UNIQUE_APIC;
+	}
+	return 0;
+}
+
+enum uv_system_type get_uv_system_type(void)
+{
+	return uv_system_type;
+}
+
+int is_uv_system(void)
+{
+	return uv_system_type != UV_NONE;
+}
diff --git a/include/asm-x86/genapic_32.h b/include/asm-x86/genapic_32.h
index 5fa893dce729..f1b96932746b 100644
--- a/include/asm-x86/genapic_32.h
+++ b/include/asm-x86/genapic_32.h
@@ -114,4 +114,9 @@ struct genapic {
 
 extern struct genapic *genapic;
 
+enum uv_system_type {UV_NONE, UV_LEGACY_APIC, UV_X2APIC, UV_NON_UNIQUE_APIC};
+#define get_uv_system_type()		UV_NONE
+#define is_uv_system()			0
+
+
 #endif
diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h
index d7e516ccbaa4..914815c28ae5 100644
--- a/include/asm-x86/genapic_64.h
+++ b/include/asm-x86/genapic_64.h
@@ -33,5 +33,10 @@ extern struct genapic *genapic;
 
 extern struct genapic apic_flat;
 extern struct genapic apic_physflat;
+extern int acpi_madt_oem_check(char *, char *);
+
+enum uv_system_type {UV_NONE, UV_LEGACY_APIC, UV_X2APIC, UV_NON_UNIQUE_APIC};
+extern enum uv_system_type get_uv_system_type(void);
+extern int is_uv_system(void);
 
 #endif
-- 
cgit v1.2.3


From a65d1d644c2b65bfb99e766e7160d764b8b2bfa4 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 28 Mar 2008 14:12:08 -0500
Subject: x86: increase size of APICID

Increase the number of bits in an apicid from 8 to 32.

By default, MP_processor_info() gets the APICID from the
mpc_config_processor structure. However, this structure limits
the size of APICID to 8 bits. This patch allows the caller of
MP_processor_info() to optionally pass a larger APICID that will
be used instead of the one in the mpc_config_processor struct.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c | 4 ++--
 arch/x86/kernel/mpparse_64.c | 4 ++--
 arch/x86/mm/srat_64.c        | 6 +++++-
 include/asm-x86/apicdef.h    | 9 ++++++---
 include/asm-x86/mpspec.h     | 4 ++--
 5 files changed, 17 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 4b46a37e0634..7b7e008496e0 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -807,7 +807,7 @@ void __init mp_register_lapic_address(u64 address)
 	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
 }
 
-void __cpuinit mp_register_lapic (u8 id, u8 enabled)
+void __cpuinit mp_register_lapic (int id, u8 enabled)
 {
 	if (MAX_APICS - id <= 0) {
 		printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
@@ -862,7 +862,7 @@ static u8 uniq_ioapic_id(u8 id)
 		return id;
 }
 
-void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
+void __init mp_register_ioapic(int id, u32 address, u32 gsi_base)
 {
 	int idx = 0;
 
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 29d2c40e54a2..4840a846904e 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -633,7 +633,7 @@ void __init mp_register_lapic_address(u64 address)
 	if (boot_cpu_physical_apicid == -1U)
 		boot_cpu_physical_apicid  = GET_APIC_ID(read_apic_id());
 }
-void __cpuinit mp_register_lapic(u8 id, u8 enabled)
+void __cpuinit mp_register_lapic(int id, u8 enabled)
 {
 	if (!enabled) {
 		++disabled_cpus;
@@ -683,7 +683,7 @@ static u8 uniq_ioapic_id(u8 id)
 	return find_first_zero_bit(used, 256);
 }
 
-void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
+void __init mp_register_ioapic(int id, u32 address, u32 gsi_base)
 {
 	int idx = 0;
 
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index 04e06c8226e3..1bae9c855ceb 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -20,6 +20,7 @@
 #include <asm/proto.h>
 #include <asm/numa.h>
 #include <asm/e820.h>
+#include <asm/genapic.h>
 
 int acpi_numa __initdata;
 
@@ -148,7 +149,10 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 		return;
 	}
 
-	apic_id = pa->apic_id;
+	if (is_uv_system())
+		apic_id = (pa->apic_id << 8) | pa->local_sapic_eid;
+	else
+		apic_id = pa->apic_id;
 	apicid_to_node[apic_id] = node;
 	acpi_numa = 1;
 	printk(KERN_INFO "SRAT: PXM %u -> APIC %u -> Node %u\n",
diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index 8b244683431b..6b9008c78731 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -133,7 +133,7 @@
 # define MAX_IO_APICS 64
 #else
 # define MAX_IO_APICS 128
-# define MAX_LOCAL_APIC 256
+# define MAX_LOCAL_APIC 32768
 #endif
 
 /*
@@ -406,6 +406,9 @@ struct local_apic {
 
 #undef u32
 
-#define BAD_APICID 0xFFu
-
+#ifdef CONFIG_X86_32
+ #define BAD_APICID 0xFFu
+#else
+ #define BAD_APICID 0xFFFFu
+#endif
 #endif
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 31bac12a97d1..1f6445b147f8 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -47,9 +47,9 @@ extern void get_smp_config(void);
 
 void __cpuinit generic_processor_info(int apicid, int version);
 #ifdef CONFIG_ACPI
-extern void mp_register_lapic(u8 id, u8 enabled);
+extern void mp_register_lapic(int id, u8 enabled);
 extern void mp_register_lapic_address(u64 address);
-extern void mp_register_ioapic(u8 id, u32 address, u32 gsi_base);
+extern void mp_register_ioapic(int id, u32 address, u32 gsi_base);
 extern void mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger,
 				   u32 gsi);
 extern void mp_config_acpi_legacy_irqs(void);
-- 
cgit v1.2.3


From 0d3e865b2644e4a2250ab25c5475a0cd0d514b7e Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 28 Mar 2008 14:12:11 -0500
Subject: x86: add UV specific header for MMR definitions

Definitions of UV MMRs.
Note: this file is auto-generated by hardware design tools.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/uv/uv_mmrs.h | 373 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 373 insertions(+)
 create mode 100644 include/asm-x86/uv/uv_mmrs.h

(limited to 'include')

diff --git a/include/asm-x86/uv/uv_mmrs.h b/include/asm-x86/uv/uv_mmrs.h
new file mode 100644
index 000000000000..3b69fe6b6376
--- /dev/null
+++ b/include/asm-x86/uv/uv_mmrs.h
@@ -0,0 +1,373 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * SGI UV MMR definitions
+ *
+ * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved.
+ */
+
+#ifndef __ASM_X86_UV_MMRS__
+#define __ASM_X86_UV_MMRS__
+
+/*
+ *       AUTO GENERATED - Do not edit
+ */
+
+ #define UV_MMR_ENABLE		(1UL << 63)
+
+/* ========================================================================= */
+/*                               UVH_IPI_INT                                 */
+/* ========================================================================= */
+#define UVH_IPI_INT 0x60500UL
+#define UVH_IPI_INT_32 0x0360
+
+#define UVH_IPI_INT_VECTOR_SHFT 0
+#define UVH_IPI_INT_VECTOR_MASK 0x00000000000000ffUL
+#define UVH_IPI_INT_DELIVERY_MODE_SHFT 8
+#define UVH_IPI_INT_DELIVERY_MODE_MASK 0x0000000000000700UL
+#define UVH_IPI_INT_DESTMODE_SHFT 11
+#define UVH_IPI_INT_DESTMODE_MASK 0x0000000000000800UL
+#define UVH_IPI_INT_APIC_ID_SHFT 16
+#define UVH_IPI_INT_APIC_ID_MASK 0x0000ffffffff0000UL
+#define UVH_IPI_INT_SEND_SHFT 63
+#define UVH_IPI_INT_SEND_MASK 0x8000000000000000UL
+
+union uvh_ipi_int_u {
+    unsigned long	v;
+    struct uvh_ipi_int_s {
+	unsigned long	vector_       :  8;  /* RW */
+	unsigned long	delivery_mode :  3;  /* RW */
+	unsigned long	destmode      :  1;  /* RW */
+	unsigned long	rsvd_12_15    :  4;  /*    */
+	unsigned long	apic_id       : 32;  /* RW */
+	unsigned long	rsvd_48_62    : 15;  /*    */
+	unsigned long	send          :  1;  /* WP */
+    } s;
+};
+
+/* ========================================================================= */
+/*                   UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST                     */
+/* ========================================================================= */
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST 0x320050UL
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_32 0x009f0
+
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_ADDRESS_SHFT 4
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_ADDRESS_MASK 0x000007fffffffff0UL
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_NODE_ID_SHFT 49
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_NODE_ID_MASK 0x7ffe000000000000UL
+
+union uvh_lb_bau_intd_payload_queue_first_u {
+    unsigned long	v;
+    struct uvh_lb_bau_intd_payload_queue_first_s {
+	unsigned long	rsvd_0_3:  4;  /*    */
+	unsigned long	address : 39;  /* RW */
+	unsigned long	rsvd_43_48:  6;  /*    */
+	unsigned long	node_id : 14;  /* RW */
+	unsigned long	rsvd_63 :  1;  /*    */
+    } s;
+};
+
+/* ========================================================================= */
+/*                    UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST                     */
+/* ========================================================================= */
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST 0x320060UL
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST_32 0x009f8
+
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST_ADDRESS_SHFT 4
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST_ADDRESS_MASK 0x000007fffffffff0UL
+
+union uvh_lb_bau_intd_payload_queue_last_u {
+    unsigned long	v;
+    struct uvh_lb_bau_intd_payload_queue_last_s {
+	unsigned long	rsvd_0_3:  4;  /*    */
+	unsigned long	address : 39;  /* RW */
+	unsigned long	rsvd_43_63: 21;  /*    */
+    } s;
+};
+
+/* ========================================================================= */
+/*                    UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL                     */
+/* ========================================================================= */
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL 0x320070UL
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL_32 0x00a00
+
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL_ADDRESS_SHFT 4
+#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL_ADDRESS_MASK 0x000007fffffffff0UL
+
+union uvh_lb_bau_intd_payload_queue_tail_u {
+    unsigned long	v;
+    struct uvh_lb_bau_intd_payload_queue_tail_s {
+	unsigned long	rsvd_0_3:  4;  /*    */
+	unsigned long	address : 39;  /* RW */
+	unsigned long	rsvd_43_63: 21;  /*    */
+    } s;
+};
+
+/* ========================================================================= */
+/*                   UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE                    */
+/* ========================================================================= */
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE 0x320080UL
+
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_0_SHFT 0
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_0_MASK 0x0000000000000001UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_1_SHFT 1
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_1_MASK 0x0000000000000002UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_2_SHFT 2
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_2_MASK 0x0000000000000004UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_3_SHFT 3
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_3_MASK 0x0000000000000008UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_4_SHFT 4
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_4_MASK 0x0000000000000010UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_5_SHFT 5
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_5_MASK 0x0000000000000020UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_6_SHFT 6
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_6_MASK 0x0000000000000040UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_7_SHFT 7
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_7_MASK 0x0000000000000080UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_0_SHFT 8
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_0_MASK 0x0000000000000100UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_1_SHFT 9
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_1_MASK 0x0000000000000200UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_2_SHFT 10
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_2_MASK 0x0000000000000400UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_3_SHFT 11
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_3_MASK 0x0000000000000800UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_4_SHFT 12
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_4_MASK 0x0000000000001000UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_5_SHFT 13
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_5_MASK 0x0000000000002000UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_6_SHFT 14
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_6_MASK 0x0000000000004000UL
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_7_SHFT 15
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_7_MASK 0x0000000000008000UL
+union uvh_lb_bau_intd_software_acknowledge_u {
+    unsigned long	v;
+    struct uvh_lb_bau_intd_software_acknowledge_s {
+	unsigned long	pending_0 :  1;  /* RW, W1C */
+	unsigned long	pending_1 :  1;  /* RW, W1C */
+	unsigned long	pending_2 :  1;  /* RW, W1C */
+	unsigned long	pending_3 :  1;  /* RW, W1C */
+	unsigned long	pending_4 :  1;  /* RW, W1C */
+	unsigned long	pending_5 :  1;  /* RW, W1C */
+	unsigned long	pending_6 :  1;  /* RW, W1C */
+	unsigned long	pending_7 :  1;  /* RW, W1C */
+	unsigned long	timeout_0 :  1;  /* RW, W1C */
+	unsigned long	timeout_1 :  1;  /* RW, W1C */
+	unsigned long	timeout_2 :  1;  /* RW, W1C */
+	unsigned long	timeout_3 :  1;  /* RW, W1C */
+	unsigned long	timeout_4 :  1;  /* RW, W1C */
+	unsigned long	timeout_5 :  1;  /* RW, W1C */
+	unsigned long	timeout_6 :  1;  /* RW, W1C */
+	unsigned long	timeout_7 :  1;  /* RW, W1C */
+	unsigned long	rsvd_16_63: 48;  /*    */
+    } s;
+};
+
+/* ========================================================================= */
+/*                UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS                 */
+/* ========================================================================= */
+#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS 0x0000000000320088UL
+
+/* ========================================================================= */
+/*                     UVH_LB_BAU_SB_ACTIVATION_CONTROL                      */
+/* ========================================================================= */
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL 0x320020UL
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_32 0x009d8
+
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_INDEX_SHFT 0
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_INDEX_MASK 0x000000000000003fUL
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_PUSH_SHFT 62
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_PUSH_MASK 0x4000000000000000UL
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_INIT_SHFT 63
+#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_INIT_MASK 0x8000000000000000UL
+
+union uvh_lb_bau_sb_activation_control_u {
+    unsigned long	v;
+    struct uvh_lb_bau_sb_activation_control_s {
+	unsigned long	index :  6;  /* RW */
+	unsigned long	rsvd_6_61: 56;  /*    */
+	unsigned long	push  :  1;  /* WP */
+	unsigned long	init  :  1;  /* WP */
+    } s;
+};
+
+/* ========================================================================= */
+/*                    UVH_LB_BAU_SB_ACTIVATION_STATUS_0                      */
+/* ========================================================================= */
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_0 0x320030UL
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_0_32 0x009e0
+
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_0_STATUS_SHFT 0
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_0_STATUS_MASK 0xffffffffffffffffUL
+
+union uvh_lb_bau_sb_activation_status_0_u {
+    unsigned long	v;
+    struct uvh_lb_bau_sb_activation_status_0_s {
+	unsigned long	status : 64;  /* RW */
+    } s;
+};
+
+/* ========================================================================= */
+/*                    UVH_LB_BAU_SB_ACTIVATION_STATUS_1                      */
+/* ========================================================================= */
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_1 0x320040UL
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_1_32 0x009e8
+
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_1_STATUS_SHFT 0
+#define UVH_LB_BAU_SB_ACTIVATION_STATUS_1_STATUS_MASK 0xffffffffffffffffUL
+
+union uvh_lb_bau_sb_activation_status_1_u {
+    unsigned long	v;
+    struct uvh_lb_bau_sb_activation_status_1_s {
+	unsigned long	status : 64;  /* RW */
+    } s;
+};
+
+/* ========================================================================= */
+/*                      UVH_LB_BAU_SB_DESCRIPTOR_BASE                        */
+/* ========================================================================= */
+#define UVH_LB_BAU_SB_DESCRIPTOR_BASE 0x320010UL
+#define UVH_LB_BAU_SB_DESCRIPTOR_BASE_32 0x009d0
+
+#define UVH_LB_BAU_SB_DESCRIPTOR_BASE_PAGE_ADDRESS_SHFT 12
+#define UVH_LB_BAU_SB_DESCRIPTOR_BASE_PAGE_ADDRESS_MASK 0x000007fffffff000UL
+#define UVH_LB_BAU_SB_DESCRIPTOR_BASE_NODE_ID_SHFT 49
+#define UVH_LB_BAU_SB_DESCRIPTOR_BASE_NODE_ID_MASK 0x7ffe000000000000UL
+
+union uvh_lb_bau_sb_descriptor_base_u {
+    unsigned long	v;
+    struct uvh_lb_bau_sb_descriptor_base_s {
+	unsigned long	rsvd_0_11    : 12;  /*    */
+	unsigned long	page_address : 31;  /* RW */
+	unsigned long	rsvd_43_48   :  6;  /*    */
+	unsigned long	node_id      : 14;  /* RW */
+	unsigned long	rsvd_63      :  1;  /*    */
+    } s;
+};
+
+/* ========================================================================= */
+/*                               UVH_NODE_ID                                 */
+/* ========================================================================= */
+#define UVH_NODE_ID 0x0UL
+
+#define UVH_NODE_ID_FORCE1_SHFT 0
+#define UVH_NODE_ID_FORCE1_MASK 0x0000000000000001UL
+#define UVH_NODE_ID_MANUFACTURER_SHFT 1
+#define UVH_NODE_ID_MANUFACTURER_MASK 0x0000000000000ffeUL
+#define UVH_NODE_ID_PART_NUMBER_SHFT 12
+#define UVH_NODE_ID_PART_NUMBER_MASK 0x000000000ffff000UL
+#define UVH_NODE_ID_REVISION_SHFT 28
+#define UVH_NODE_ID_REVISION_MASK 0x00000000f0000000UL
+#define UVH_NODE_ID_NODE_ID_SHFT 32
+#define UVH_NODE_ID_NODE_ID_MASK 0x00007fff00000000UL
+#define UVH_NODE_ID_NODES_PER_BIT_SHFT 48
+#define UVH_NODE_ID_NODES_PER_BIT_MASK 0x007f000000000000UL
+#define UVH_NODE_ID_NI_PORT_SHFT 56
+#define UVH_NODE_ID_NI_PORT_MASK 0x0f00000000000000UL
+
+union uvh_node_id_u {
+    unsigned long	v;
+    struct uvh_node_id_s {
+	unsigned long	force1        :  1;  /* RO */
+	unsigned long	manufacturer  : 11;  /* RO */
+	unsigned long	part_number   : 16;  /* RO */
+	unsigned long	revision      :  4;  /* RO */
+	unsigned long	node_id       : 15;  /* RW */
+	unsigned long	rsvd_47       :  1;  /*    */
+	unsigned long	nodes_per_bit :  7;  /* RW */
+	unsigned long	rsvd_55       :  1;  /*    */
+	unsigned long	ni_port       :  4;  /* RO */
+	unsigned long	rsvd_60_63    :  4;  /*    */
+    } s;
+};
+
+/* ========================================================================= */
+/*                    UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR                      */
+/* ========================================================================= */
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR 0x1600010UL
+
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_SHFT 28
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffff0000000UL
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_GR4_SHFT 46
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_GR4_MASK 0x0000400000000000UL
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_SHFT 52
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_MASK 0x00f0000000000000UL
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63
+#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL
+
+union uvh_rh_gam_gru_overlay_config_mmr_u {
+    unsigned long	v;
+    struct uvh_rh_gam_gru_overlay_config_mmr_s {
+	unsigned long	rsvd_0_27: 28;  /*    */
+	unsigned long	base   : 18;  /* RW */
+	unsigned long	gr4    :  1;  /* RW */
+	unsigned long	rsvd_47_51:  5;  /*    */
+	unsigned long	n_gru  :  4;  /* RW */
+	unsigned long	rsvd_56_62:  7;  /*    */
+	unsigned long	enable :  1;  /* RW */
+    } s;
+};
+
+/* ========================================================================= */
+/*                    UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR                      */
+/* ========================================================================= */
+#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR 0x1600028UL
+
+#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_SHFT 26
+#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffffc000000UL
+#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_DUAL_HUB_SHFT 46
+#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_DUAL_HUB_MASK 0x0000400000000000UL
+#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63
+#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL
+
+union uvh_rh_gam_mmr_overlay_config_mmr_u {
+    unsigned long	v;
+    struct uvh_rh_gam_mmr_overlay_config_mmr_s {
+	unsigned long	rsvd_0_25: 26;  /*    */
+	unsigned long	base     : 20;  /* RW */
+	unsigned long	dual_hub :  1;  /* RW */
+	unsigned long	rsvd_47_62: 16;  /*    */
+	unsigned long	enable   :  1;  /* RW */
+    } s;
+};
+
+/* ========================================================================= */
+/*                                 UVH_RTC                                   */
+/* ========================================================================= */
+#define UVH_RTC 0x28000UL
+
+#define UVH_RTC_REAL_TIME_CLOCK_SHFT 0
+#define UVH_RTC_REAL_TIME_CLOCK_MASK 0x00ffffffffffffffUL
+
+union uvh_rtc_u {
+    unsigned long	v;
+    struct uvh_rtc_s {
+	unsigned long	real_time_clock : 56;  /* RW */
+	unsigned long	rsvd_56_63      :  8;  /*    */
+    } s;
+};
+
+/* ========================================================================= */
+/*                          UVH_SI_ADDR_MAP_CONFIG                           */
+/* ========================================================================= */
+#define UVH_SI_ADDR_MAP_CONFIG 0xc80000UL
+
+#define UVH_SI_ADDR_MAP_CONFIG_M_SKT_SHFT 0
+#define UVH_SI_ADDR_MAP_CONFIG_M_SKT_MASK 0x000000000000003fUL
+#define UVH_SI_ADDR_MAP_CONFIG_N_SKT_SHFT 8
+#define UVH_SI_ADDR_MAP_CONFIG_N_SKT_MASK 0x0000000000000f00UL
+
+union uvh_si_addr_map_config_u {
+    unsigned long	v;
+    struct uvh_si_addr_map_config_s {
+	unsigned long	m_skt :  6;  /* RW */
+	unsigned long	rsvd_6_7:  2;  /*    */
+	unsigned long	n_skt :  4;  /* RW */
+	unsigned long	rsvd_12_63: 52;  /*    */
+    } s;
+};
+
+
+#endif /* __ASM_X86_UV_MMRS__ */
-- 
cgit v1.2.3


From 952cf6d7ae52cc5423baa57e978e20e732a89ba6 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 28 Mar 2008 14:12:13 -0500
Subject: x86: define the macros and tables for the basic UV infrastructure.

Define the macros and tables for the basic UV infrastructure.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/uv/uv_hub.h | 210 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 210 insertions(+)
 create mode 100644 include/asm-x86/uv/uv_hub.h

(limited to 'include')

diff --git a/include/asm-x86/uv/uv_hub.h b/include/asm-x86/uv/uv_hub.h
new file mode 100644
index 000000000000..b4fcf9cf8951
--- /dev/null
+++ b/include/asm-x86/uv/uv_hub.h
@@ -0,0 +1,210 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * SGI UV architectural definitions
+ *
+ * Copyright (C) 2007 Silicon Graphics, Inc. All rights reserved.
+ */
+
+#ifndef __ASM_X86_UV_HUB_H__
+#define __ASM_X86_UV_HUB_H__
+
+#include <linux/numa.h>
+#include <linux/percpu.h>
+#include <asm/types.h>
+#include <asm/percpu.h>
+
+
+/*
+ * Addressing Terminology
+ *
+ * 	NASID - network ID of a router, Mbrick or Cbrick. Nasid values of
+ * 		routers always have low bit of 1, C/MBricks have low bit
+ * 		equal to 0. Most addressing macros that target UV hub chips
+ * 		right shift the NASID by 1 to exclude the always-zero bit.
+ *
+ *	SNASID - NASID right shifted by 1 bit.
+ *
+ *
+ *  Memory/UV-HUB Processor Socket Address Format:
+ *  +--------+---------------+---------------------+
+ *  |00..0000|    SNASID     |      NodeOffset     |
+ *  +--------+---------------+---------------------+
+ *           <--- N bits --->|<--------M bits ----->
+ *
+ *	M number of node offset bits (35 .. 40)
+ *	N number of SNASID bits (0 .. 10)
+ *
+ *		Note: M + N cannot currently exceed 44 (x86_64) or 46 (IA64).
+ *		The actual values are configuration dependent and are set at
+ *		boot time
+ *
+ * APICID format
+ * 	NOTE!!!!!! This is the current format of the APICID. However, code
+ * 	should assume that this will change in the future. Use functions
+ * 	in this file for all APICID bit manipulations and conversion.
+ *
+ * 		1111110000000000
+ * 		5432109876543210
+ *		nnnnnnnnnnlc0cch
+ *		sssssssssss
+ *
+ *			n  = snasid bits
+ *			l =  socket number on board
+ *			c  = core
+ *			h  = hyperthread
+ *			s  = bits that are in the socket CSR
+ *
+ *	Note: Processor only supports 12 bits in the APICID register. The ACPI
+ *	      tables hold all 16 bits. Software needs to be aware of this.
+ *
+ * 	      Unless otherwise specified, all references to APICID refer to
+ * 	      the FULL value contained in ACPI tables, not the subset in the
+ * 	      processor APICID register.
+ */
+
+
+/*
+ * Maximum number of bricks in all partitions and in all coherency domains.
+ * This is the total number of bricks accessible in the numalink fabric. It
+ * includes all C & M bricks. Routers are NOT included.
+ *
+ * This value is also the value of the maximum number of non-router NASIDs
+ * in the numalink fabric.
+ *
+ * NOTE: a brick may be 1 or 2 OS nodes. Don't get these confused.
+ */
+#define UV_MAX_NUMALINK_BLADES	16384
+
+/*
+ * Maximum number of C/Mbricks within a software SSI (hardware may support
+ * more).
+ */
+#define UV_MAX_SSI_BLADES	256
+
+/*
+ * The largest possible NASID of a C or M brick (+ 2)
+ */
+#define UV_MAX_NASID_VALUE	(UV_MAX_NUMALINK_NODES * 2)
+
+/*
+ * The following defines attributes of the HUB chip. These attributes are
+ * frequently referenced and are kept in the per-cpu data areas of each cpu.
+ * They are kept together in a struct to minimize cache misses.
+ */
+struct uv_hub_info_s {
+	unsigned long	global_mmr_base;
+	unsigned short	local_nasid;
+	unsigned short	gnode_upper;
+	unsigned short	coherency_domain_number;
+	unsigned short	numa_blade_id;
+	unsigned char	blade_processor_id;
+	unsigned char	m_val;
+	unsigned char	n_val;
+};
+DECLARE_PER_CPU(struct uv_hub_info_s, __uv_hub_info);
+#define uv_hub_info 		(&__get_cpu_var(__uv_hub_info))
+#define uv_cpu_hub_info(cpu)	(&per_cpu(__uv_hub_info, cpu))
+
+/*
+ * Local & Global MMR space macros.
+ * 	Note: macros are intended to be used ONLY by inline functions
+ * 	in this file - not by other kernel code.
+ */
+#define UV_SNASID(n)			((n) >> 1)
+#define UV_NASID(n)			((n) << 1)
+
+#define UV_LOCAL_MMR_BASE		0xf4000000UL
+#define UV_GLOBAL_MMR32_BASE		0xf8000000UL
+#define UV_GLOBAL_MMR64_BASE		(uv_hub_info->global_mmr_base)
+
+#define UV_GLOBAL_MMR32_SNASID_MASK	0x3ff
+#define UV_GLOBAL_MMR32_SNASID_SHIFT	15
+#define UV_GLOBAL_MMR64_SNASID_SHIFT	26
+
+#define UV_GLOBAL_MMR32_NASID_BITS(n)					\
+		(((UV_SNASID(n) & UV_GLOBAL_MMR32_SNASID_MASK)) <<	\
+		(UV_GLOBAL_MMR32_SNASID_SHIFT))
+
+#define UV_GLOBAL_MMR64_NASID_BITS(n)					\
+	((unsigned long)UV_SNASID(n) << UV_GLOBAL_MMR64_SNASID_SHIFT)
+
+#define UV_APIC_NASID_SHIFT	6
+
+/*
+ * Extract a NASID from an APICID (full apicid, not processor subset)
+ */
+static inline int uv_apicid_to_nasid(int apicid)
+{
+	return (UV_NASID(apicid >> UV_APIC_NASID_SHIFT));
+}
+
+/*
+ * Access global MMRs using the low memory MMR32 space. This region supports
+ * faster MMR access but not all MMRs are accessible in this space.
+ */
+static inline unsigned long *uv_global_mmr32_address(int nasid,
+				unsigned long offset)
+{
+	return __va(UV_GLOBAL_MMR32_BASE |
+		       UV_GLOBAL_MMR32_NASID_BITS(nasid) | offset);
+}
+
+static inline void uv_write_global_mmr32(int nasid, unsigned long offset,
+				 unsigned long val)
+{
+	*uv_global_mmr32_address(nasid, offset) = val;
+}
+
+static inline unsigned long uv_read_global_mmr32(int nasid,
+						 unsigned long offset)
+{
+	return *uv_global_mmr32_address(nasid, offset);
+}
+
+/*
+ * Access Global MMR space using the MMR space located at the top of physical
+ * memory.
+ */
+static inline unsigned long *uv_global_mmr64_address(int nasid,
+				unsigned long offset)
+{
+	return __va(UV_GLOBAL_MMR64_BASE |
+		       UV_GLOBAL_MMR64_NASID_BITS(nasid) | offset);
+}
+
+static inline void uv_write_global_mmr64(int nasid, unsigned long offset,
+				unsigned long val)
+{
+	*uv_global_mmr64_address(nasid, offset) = val;
+}
+
+static inline unsigned long uv_read_global_mmr64(int nasid,
+						 unsigned long offset)
+{
+	return *uv_global_mmr64_address(nasid, offset);
+}
+
+/*
+ * Access node local MMRs. Faster than using global space but only local MMRs
+ * are accessible.
+ */
+static inline unsigned long *uv_local_mmr_address(unsigned long offset)
+{
+	return __va(UV_LOCAL_MMR_BASE | offset);
+}
+
+static inline unsigned long uv_read_local_mmr(unsigned long offset)
+{
+	return *uv_local_mmr_address(offset);
+}
+
+static inline void uv_write_local_mmr(unsigned long offset, unsigned long val)
+{
+	*uv_local_mmr_address(offset) = val;
+}
+
+#endif /* __ASM_X86_UV_HUB__ */
+
-- 
cgit v1.2.3


From 8400def8252f90ecd056657c0bac806afadd8511 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 28 Mar 2008 14:12:14 -0500
Subject: x86: define the macros and tables for blade functions

Add UV macros for converting between cpu numbers, blade numbers
and node numbers. Note that these are used ONLY within x86_64 UV
modules, and are not for general kernel use.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/uv/uv_hub.h | 74 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 74 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86/uv/uv_hub.h b/include/asm-x86/uv/uv_hub.h
index b4fcf9cf8951..26b9240d1e23 100644
--- a/include/asm-x86/uv/uv_hub.h
+++ b/include/asm-x86/uv/uv_hub.h
@@ -206,5 +206,79 @@ static inline void uv_write_local_mmr(unsigned long offset, unsigned long val)
 	*uv_local_mmr_address(offset) = val;
 }
 
+/*
+ * Structures and definitions for converting between cpu, node, and blade
+ * numbers.
+ */
+struct uv_blade_info {
+	unsigned short	nr_posible_cpus;
+	unsigned short	nr_online_cpus;
+	unsigned short	nasid;
+};
+struct uv_blade_info *uv_blade_info;
+extern short *uv_node_to_blade;
+extern short *uv_cpu_to_blade;
+extern short uv_possible_blades;
+
+/* Blade-local cpu number of current cpu. Numbered 0 .. <# cpus on the blade> */
+static inline int uv_blade_processor_id(void)
+{
+	return uv_hub_info->blade_processor_id;
+}
+
+/* Blade number of current cpu. Numnbered 0 .. <#blades -1> */
+static inline int uv_numa_blade_id(void)
+{
+	return uv_hub_info->numa_blade_id;
+}
+
+/* Convert a cpu number to the the UV blade number */
+static inline int uv_cpu_to_blade_id(int cpu)
+{
+	return uv_cpu_to_blade[cpu];
+}
+
+/* Convert linux node number to the UV blade number */
+static inline int uv_node_to_blade_id(int nid)
+{
+	return uv_node_to_blade[nid];
+}
+
+/* Convert a blade id to the NASID of the blade */
+static inline int uv_blade_to_nasid(int bid)
+{
+	return uv_blade_info[bid].nasid;
+}
+
+/* Determine the number of possible cpus on a blade */
+static inline int uv_blade_nr_possible_cpus(int bid)
+{
+	return uv_blade_info[bid].nr_posible_cpus;
+}
+
+/* Determine the number of online cpus on a blade */
+static inline int uv_blade_nr_online_cpus(int bid)
+{
+	return uv_blade_info[bid].nr_online_cpus;
+}
+
+/* Convert a cpu id to the NASID of the blade containing the cpu */
+static inline int uv_cpu_to_nasid(int cpu)
+{
+	return uv_blade_info[uv_cpu_to_blade_id(cpu)].nasid;
+}
+
+/* Convert a node number to the NASID of the blade */
+static inline int uv_node_to_nasid(int nid)
+{
+	return uv_blade_info[uv_node_to_blade_id(nid)].nasid;
+}
+
+/* Maximum possible number of blades */
+static inline int uv_num_possible_blades(void)
+{
+	return uv_possible_blades;
+}
+
 #endif /* __ASM_X86_UV_HUB__ */
 
-- 
cgit v1.2.3


From ac23d4ee3f84de33c16ed7e68f9adee2386e74fb Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Fri, 28 Mar 2008 14:12:16 -0500
Subject: x86: support for new UV apic

UV supports really big systems. So big, in fact, that the APICID register
does not contain enough bits to contain an APICID that is unique across all
cpus.

The UV BIOS supports 3 APICID modes:

	- legacy mode. This mode uses the old APIC mode where
	  APICID is in bits [31:24] of the APICID register.

	- x2apic mode. This mode is whitebox-compatible. APICIDs
	  are unique across all cpus. Standard x2apic APIC operations
	  (Intel-defined) can be used for IPIs. The node identifier
	  fits within the Intel-defined portion of the APICID register.

	- x2apic-uv mode. In this mode, the APICIDs on each node have
	  unique IDs, but IDs on different node are not unique. For example,
	  if each mode has 32 cpus, the APICIDs on each node might be
	  0 - 31. Every node has the same set of IDs.
	  The UV hub is used to route IPIs/interrupts to the correct node.
	  Traditional APIC operations WILL NOT WORK.

In x2apic-uv mode, the ACPI tables all contain a full unique ID (note:
exact bit layout still changing but the following is close):

	nnnnnnnnnnlc0cch
		n = unique node number
		l = socket number on board
		c = core
		h = hyperthread

Only the "lc0cch" bits are written to the APICID register. The remaining bits are
supplied by having the get_apic_id() function "OR" the extra bits into the value
read from the APICID register. (Hmmm.. why not keep the ENTIRE APICID register
in per-cpu data....)

The x2apic-uv mode is recognized by the MADT table containing:
	  oem_id = "SGI"
	  oem_table_id = "UV-X"

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/Makefile         |   2 +-
 arch/x86/kernel/apic_64.c        |   2 +
 arch/x86/kernel/genapic_64.c     |  18 +++
 arch/x86/kernel/genx2apic_uv_x.c | 245 +++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/setup64.c        |   4 +
 arch/x86/kernel/smpboot.c        |   5 +
 include/asm-x86/genapic_64.h     |   5 +
 include/asm-x86/smp.h            |   5 +
 8 files changed, 285 insertions(+), 1 deletion(-)
 create mode 100644 arch/x86/kernel/genx2apic_uv_x.c

(limited to 'include')

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 1fe841a86f7e..0bf2fb55aa74 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -89,7 +89,7 @@ scx200-y			+= scx200_32.o
 ###
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
-        obj-y				+= genapic_64.o genapic_flat_64.o
+        obj-y				+= genapic_64.o genapic_flat_64.o genx2apic_uv_x.o
         obj-$(CONFIG_X86_PM_TIMER)	+= pmtimer_64.o
         obj-$(CONFIG_AUDIT)		+= audit_64.o
 
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 9b4cacdfd74f..8b0fad47a5d2 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -738,6 +738,7 @@ void __cpuinit setup_local_APIC(void)
 	unsigned int value;
 	int i, j;
 
+	preempt_disable();
 	value = apic_read(APIC_LVR);
 
 	BUILD_BUG_ON((SPURIOUS_APIC_VECTOR & 0x0f) != 0x0f);
@@ -831,6 +832,7 @@ void __cpuinit setup_local_APIC(void)
 	else
 		value = APIC_DM_NMI | APIC_LVT_MASKED;
 	apic_write(APIC_LVT1, value);
+	preempt_enable();
 }
 
 void __cpuinit lapic_setup_esr(void)
diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c
index 4cc1c218ae4c..910a4a777a4c 100644
--- a/arch/x86/kernel/genapic_64.c
+++ b/arch/x86/kernel/genapic_64.c
@@ -15,6 +15,7 @@
 #include <linux/kernel.h>
 #include <linux/ctype.h>
 #include <linux/init.h>
+#include <linux/hardirq.h>
 
 #include <asm/smp.h>
 #include <asm/ipi.h>
@@ -32,6 +33,7 @@ void *x86_cpu_to_apicid_early_ptr;
 #endif
 DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
+DEFINE_PER_CPU(int, x2apic_extra_bits);
 
 struct genapic __read_mostly *genapic = &apic_flat;
 
@@ -42,6 +44,9 @@ static enum uv_system_type uv_system_type;
  */
 void __init setup_apic_routing(void)
 {
+	if (uv_system_type == UV_NON_UNIQUE_APIC)
+		genapic = &apic_x2apic_uv_x;
+	else
 #ifdef CONFIG_ACPI
 	/*
 	 * Quirk: some x86_64 machines can only use physical APIC mode
@@ -82,6 +87,19 @@ int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 0;
 }
 
+unsigned int read_apic_id(void)
+{
+	unsigned int id;
+
+	WARN_ON(preemptible());
+	id = apic_read(APIC_ID);
+	if (uv_system_type >= UV_X2APIC)
+		id  |= __get_cpu_var(x2apic_extra_bits);
+	else
+		id = (id >> 24) & 0xFFu;;
+	return id;
+}
+
 enum uv_system_type get_uv_system_type(void)
 {
 	return uv_system_type;
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
new file mode 100644
index 000000000000..5d77c9cd8e15
--- /dev/null
+++ b/arch/x86/kernel/genx2apic_uv_x.c
@@ -0,0 +1,245 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * SGI UV APIC functions (note: not an Intel compatible APIC)
+ *
+ * Copyright (C) 2007 Silicon Graphics, Inc. All rights reserved.
+ */
+
+#include <linux/threads.h>
+#include <linux/cpumask.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/bootmem.h>
+#include <linux/module.h>
+#include <asm/smp.h>
+#include <asm/ipi.h>
+#include <asm/genapic.h>
+#include <asm/uv/uv_mmrs.h>
+#include <asm/uv/uv_hub.h>
+
+DEFINE_PER_CPU(struct uv_hub_info_s, __uv_hub_info);
+EXPORT_PER_CPU_SYMBOL_GPL(__uv_hub_info);
+
+struct uv_blade_info *uv_blade_info;
+EXPORT_SYMBOL_GPL(uv_blade_info);
+
+short *uv_node_to_blade;
+EXPORT_SYMBOL_GPL(uv_node_to_blade);
+
+short *uv_cpu_to_blade;
+EXPORT_SYMBOL_GPL(uv_cpu_to_blade);
+
+short uv_possible_blades;
+EXPORT_SYMBOL_GPL(uv_possible_blades);
+
+/* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
+
+static cpumask_t uv_target_cpus(void)
+{
+	return cpumask_of_cpu(0);
+}
+
+static cpumask_t uv_vector_allocation_domain(int cpu)
+{
+	cpumask_t domain = CPU_MASK_NONE;
+	cpu_set(cpu, domain);
+	return domain;
+}
+
+int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
+{
+	unsigned long val;
+	int nasid;
+
+	nasid = uv_apicid_to_nasid(phys_apicid);
+	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
+	    (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
+	    (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+	    (6 << UVH_IPI_INT_DELIVERY_MODE_SHFT);
+	uv_write_global_mmr64(nasid, UVH_IPI_INT, val);
+	return 0;
+}
+
+static void uv_send_IPI_one(int cpu, int vector)
+{
+	unsigned long val, apicid;
+	int nasid;
+
+	apicid = per_cpu(x86_cpu_to_apicid, cpu); /* ZZZ - cache node-local ? */
+	nasid = uv_apicid_to_nasid(apicid);
+	val =
+	    (1UL << UVH_IPI_INT_SEND_SHFT) | (apicid <<
+					      UVH_IPI_INT_APIC_ID_SHFT) |
+	    (vector << UVH_IPI_INT_VECTOR_SHFT);
+	uv_write_global_mmr64(nasid, UVH_IPI_INT, val);
+	printk(KERN_DEBUG
+	     "UV: IPI to cpu %d, apicid 0x%lx, vec %d, nasid%d, val 0x%lx\n",
+	     cpu, apicid, vector, nasid, val);
+}
+
+static void uv_send_IPI_mask(cpumask_t mask, int vector)
+{
+	unsigned int cpu;
+
+	for (cpu = 0; cpu < NR_CPUS; ++cpu)
+		if (cpu_isset(cpu, mask))
+			uv_send_IPI_one(cpu, vector);
+}
+
+static void uv_send_IPI_allbutself(int vector)
+{
+	cpumask_t mask = cpu_online_map;
+
+	cpu_clear(smp_processor_id(), mask);
+
+	if (!cpus_empty(mask))
+		uv_send_IPI_mask(mask, vector);
+}
+
+static void uv_send_IPI_all(int vector)
+{
+	uv_send_IPI_mask(cpu_online_map, vector);
+}
+
+static int uv_apic_id_registered(void)
+{
+	return 1;
+}
+
+static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	cpu = first_cpu(cpumask);
+	if ((unsigned)cpu < NR_CPUS)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	else
+		return BAD_APICID;
+}
+
+static unsigned int phys_pkg_id(int index_msb)
+{
+	return GET_APIC_ID(read_apic_id()) >> index_msb;
+}
+
+#ifdef ZZZ		/* Needs x2apic patch */
+static void uv_send_IPI_self(int vector)
+{
+	apic_write(APIC_SELF_IPI, vector);
+}
+#endif
+
+struct genapic apic_x2apic_uv_x = {
+	.name = "UV large system",
+	.int_delivery_mode = dest_Fixed,
+	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
+	.target_cpus = uv_target_cpus,
+	.vector_allocation_domain = uv_vector_allocation_domain,/* Fixme ZZZ */
+	.apic_id_registered = uv_apic_id_registered,
+	.send_IPI_all = uv_send_IPI_all,
+	.send_IPI_allbutself = uv_send_IPI_allbutself,
+	.send_IPI_mask = uv_send_IPI_mask,
+	/* ZZZ.send_IPI_self = uv_send_IPI_self, */
+	.cpu_mask_to_apicid = uv_cpu_mask_to_apicid,
+	.phys_pkg_id = phys_pkg_id,	/* Fixme ZZZ */
+};
+
+static __cpuinit void set_x2apic_extra_bits(int nasid)
+{
+	__get_cpu_var(x2apic_extra_bits) = ((nasid >> 1) << 6);
+}
+
+/*
+ * Called on boot cpu.
+ */
+static __init void uv_system_init(void)
+{
+	union uvh_si_addr_map_config_u m_n_config;
+	int bytes, nid, cpu, lcpu, nasid, last_nasid, blade;
+	unsigned long mmr_base;
+
+	m_n_config.v = uv_read_local_mmr(UVH_SI_ADDR_MAP_CONFIG);
+	mmr_base =
+	    uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR) &
+	    ~UV_MMR_ENABLE;
+	printk(KERN_DEBUG "UV: global MMR base 0x%lx\n", mmr_base);
+
+	last_nasid = -1;
+	for_each_possible_cpu(cpu) {
+		nid = cpu_to_node(cpu);
+		nasid = uv_apicid_to_nasid(per_cpu(x86_cpu_to_apicid, cpu));
+		if (nasid != last_nasid)
+			uv_possible_blades++;
+		last_nasid = nasid;
+	}
+	printk(KERN_DEBUG "UV: Found %d blades\n", uv_num_possible_blades());
+
+	bytes = sizeof(struct uv_blade_info) * uv_num_possible_blades();
+	uv_blade_info = alloc_bootmem_pages(bytes);
+
+	bytes = sizeof(uv_node_to_blade[0]) * num_possible_nodes();
+	uv_node_to_blade = alloc_bootmem_pages(bytes);
+	memset(uv_node_to_blade, 255, bytes);
+
+	bytes = sizeof(uv_cpu_to_blade[0]) * num_possible_cpus();
+	uv_cpu_to_blade = alloc_bootmem_pages(bytes);
+	memset(uv_cpu_to_blade, 255, bytes);
+
+	last_nasid = -1;
+	blade = -1;
+	lcpu = -1;
+	for_each_possible_cpu(cpu) {
+		nid = cpu_to_node(cpu);
+		nasid = uv_apicid_to_nasid(per_cpu(x86_cpu_to_apicid, cpu));
+		if (nasid != last_nasid) {
+			blade++;
+			lcpu = -1;
+			uv_blade_info[blade].nr_posible_cpus = 0;
+			uv_blade_info[blade].nr_online_cpus = 0;
+		}
+		last_nasid = nasid;
+		lcpu++;
+
+		uv_cpu_hub_info(cpu)->m_val = m_n_config.s.m_skt;
+		uv_cpu_hub_info(cpu)->n_val = m_n_config.s.n_skt;
+		uv_cpu_hub_info(cpu)->numa_blade_id = blade;
+		uv_cpu_hub_info(cpu)->blade_processor_id = lcpu;
+		uv_cpu_hub_info(cpu)->local_nasid = nasid;
+		uv_cpu_hub_info(cpu)->gnode_upper =
+		    nasid & ~((1 << uv_hub_info->n_val) - 1);
+		uv_cpu_hub_info(cpu)->global_mmr_base = mmr_base;
+		uv_cpu_hub_info(cpu)->coherency_domain_number = 0;/* ZZZ */
+		uv_blade_info[blade].nasid = nasid;
+		uv_blade_info[blade].nr_posible_cpus++;
+		uv_node_to_blade[nid] = blade;
+		uv_cpu_to_blade[cpu] = blade;
+
+		printk(KERN_DEBUG "UV cpu %d, apicid 0x%x, nasid %d, nid %d\n",
+		       cpu, per_cpu(x86_cpu_to_apicid, cpu), nasid, nid);
+		printk(KERN_DEBUG "UV   lcpu %d, blade %d\n", lcpu, blade);
+	}
+}
+
+/*
+ * Called on each cpu to initialize the per_cpu UV data area.
+ */
+void __cpuinit uv_cpu_init(void)
+{
+	if (!uv_node_to_blade)
+		uv_system_init();
+
+	uv_blade_info[uv_numa_blade_id()].nr_online_cpus++;
+
+	if (get_uv_system_type() == UV_NON_UNIQUE_APIC)
+		set_x2apic_extra_bits(uv_hub_info->local_nasid);
+}
diff --git a/arch/x86/kernel/setup64.c b/arch/x86/kernel/setup64.c
index 6b4e3262e8cb..4be499cd6a0d 100644
--- a/arch/x86/kernel/setup64.c
+++ b/arch/x86/kernel/setup64.c
@@ -23,6 +23,7 @@
 #include <asm/proto.h>
 #include <asm/sections.h>
 #include <asm/setup.h>
+#include <asm/genapic.h>
 
 #ifndef CONFIG_DEBUG_BOOT_PARAMS
 struct boot_params __initdata boot_params;
@@ -264,4 +265,7 @@ void __cpuinit cpu_init (void)
 	fpu_init(); 
 
 	raw_local_save_flags(kernel_eflags);
+
+	if (is_uv_system())
+		uv_cpu_init();
 }
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 5da35d2cdbd8..22bf6c29454f 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -1101,6 +1101,7 @@ static __init void disable_smp(void)
  */
 static int __init smp_sanity_check(unsigned max_cpus)
 {
+	preempt_disable();
 	if (!physid_isset(hard_smp_processor_id(), phys_cpu_present_map)) {
 		printk(KERN_WARNING "weird, boot CPU (#%d) not listed"
 				    "by the BIOS.\n", hard_smp_processor_id());
@@ -1112,6 +1113,7 @@ static int __init smp_sanity_check(unsigned max_cpus)
 	 * get out of here now!
 	 */
 	if (!smp_found_config && !acpi_lapic) {
+		preempt_enable();
 		printk(KERN_NOTICE "SMP motherboard not detected.\n");
 		disable_smp();
 		if (APIC_init_uniprocessor())
@@ -1130,6 +1132,7 @@ static int __init smp_sanity_check(unsigned max_cpus)
 			boot_cpu_physical_apicid);
 		physid_set(hard_smp_processor_id(), phys_cpu_present_map);
 	}
+	preempt_enable();
 
 	/*
 	 * If we couldn't find a local APIC, then get out of here now!
@@ -1205,11 +1208,13 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
 		return;
 	}
 
+	preempt_disable();
 	if (GET_APIC_ID(read_apic_id()) != boot_cpu_physical_apicid) {
 		panic("Boot APIC ID in local APIC unexpected (%d vs %d)",
 		     GET_APIC_ID(read_apic_id()), boot_cpu_physical_apicid);
 		/* Or can we switch back to PIC here? */
 	}
+	preempt_enable();
 
 #ifdef CONFIG_X86_32
 	connect_bsp_APIC();
diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h
index 914815c28ae5..1de931b263ce 100644
--- a/include/asm-x86/genapic_64.h
+++ b/include/asm-x86/genapic_64.h
@@ -39,4 +39,9 @@ enum uv_system_type {UV_NONE, UV_LEGACY_APIC, UV_X2APIC, UV_NON_UNIQUE_APIC};
 extern enum uv_system_type get_uv_system_type(void);
 extern int is_uv_system(void);
 
+extern struct genapic apic_x2apic_uv_x;
+DECLARE_PER_CPU(int, x2apic_extra_bits);
+extern void uv_cpu_init(void);
+extern int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip);
+
 #endif
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index c0d693ca4357..b35566264879 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -174,10 +174,15 @@ static inline int logical_smp_processor_id(void)
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
 }
 
+#ifdef CONFIG_X86_32_SMP
 static inline unsigned int read_apic_id(void)
 {
 	return *(u32 *)(APIC_BASE + APIC_ID);
 }
+#else
+extern unsigned int read_apic_id(void);
+#endif
+
 
 # ifdef APIC_DEFINITION
 extern int hard_smp_processor_id(void);
-- 
cgit v1.2.3


From a24eae88ad3767d0a4a940a10e4a9cec849b7778 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 30 Mar 2008 12:17:12 +0200
Subject: x86: uv fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/smp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index b35566264879..654724c58f5b 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -174,7 +174,7 @@ static inline int logical_smp_processor_id(void)
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
 }
 
-#ifdef CONFIG_X86_32_SMP
+#ifndef CONFIG_X86_64
 static inline unsigned int read_apic_id(void)
 {
 	return *(u32 *)(APIC_BASE + APIC_ID);
-- 
cgit v1.2.3


From b447a468fcd130aa8951672b6115c673c274e888 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Tue, 25 Mar 2008 15:06:51 -0700
Subject: x86: clean up non-smp usage of cpu maps

Cleanup references to the early cpu maps for the non-SMP configuration
and remove some functions called for SMP configurations only.

Cc: Andi Kleen <ak@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup.c    | 28 +++++++++++-----------------
 arch/x86/mm/numa_64.c      |  4 +++-
 include/asm-x86/smp.h      |  5 +++++
 include/asm-x86/topology.h | 15 +++++++++++----
 4 files changed, 30 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 1179aa06cdbf..dc7940955b7a 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -10,7 +10,7 @@
 #include <asm/setup.h>
 #include <asm/topology.h>
 
-#ifdef CONFIG_HAVE_SETUP_PER_CPU_AREA
+#if defined(CONFIG_HAVE_SETUP_PER_CPU_AREA) && defined(CONFIG_SMP)
 /*
  * Copy data used in early init routines from the initial arrays to the
  * per cpu data areas.  These arrays then become expendable and the
@@ -21,21 +21,12 @@ static void __init setup_per_cpu_maps(void)
 	int cpu;
 
 	for_each_possible_cpu(cpu) {
-#ifdef CONFIG_SMP
-		if (per_cpu_offset(cpu)) {
-#endif
-			per_cpu(x86_cpu_to_apicid, cpu) =
-						x86_cpu_to_apicid_init[cpu];
-			per_cpu(x86_bios_cpu_apicid, cpu) =
+		per_cpu(x86_cpu_to_apicid, cpu) = x86_cpu_to_apicid_init[cpu];
+		per_cpu(x86_bios_cpu_apicid, cpu) =
 						x86_bios_cpu_apicid_init[cpu];
 #ifdef CONFIG_NUMA
-			per_cpu(x86_cpu_to_node_map, cpu) =
+		per_cpu(x86_cpu_to_node_map, cpu) =
 						x86_cpu_to_node_map_init[cpu];
-#endif
-#ifdef CONFIG_SMP
-		} else
-			printk(KERN_NOTICE "per_cpu_offset zero for cpu %d\n",
-									cpu);
 #endif
 	}
 
@@ -72,17 +63,20 @@ void __init setup_per_cpu_areas(void)
 
 	/* Copy section for each CPU (we discard the original) */
 	size = PERCPU_ENOUGH_ROOM;
-
 	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n",
 			  size);
-	for_each_cpu_mask(i, cpu_possible_map) {
+
+	for_each_possible_cpu(i) {
 		char *ptr;
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 		ptr = alloc_bootmem_pages(size);
 #else
 		int node = early_cpu_to_node(i);
-		if (!node_online(node) || !NODE_DATA(node))
+		if (!node_online(node) || !NODE_DATA(node)) {
 			ptr = alloc_bootmem_pages(size);
+			printk(KERN_INFO
+			       "cpu %d has no node or node-local memory\n", i);
+		}
 		else
 			ptr = alloc_bootmem_pages_node(NODE_DATA(node), size);
 #endif
@@ -96,7 +90,7 @@ void __init setup_per_cpu_areas(void)
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 	}
 
-	/* setup percpu data maps early */
+	/* Setup percpu data maps */
 	setup_per_cpu_maps();
 }
 
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 18267a02e67a..2ea56f48f29b 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -31,13 +31,15 @@ bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct memnode memnode;
 
+#ifdef CONFIG_SMP
 int x86_cpu_to_node_map_init[NR_CPUS] = {
 	[0 ... NR_CPUS-1] = NUMA_NO_NODE
 };
 void *x86_cpu_to_node_map_early_ptr;
+EXPORT_SYMBOL(x86_cpu_to_node_map_early_ptr);
+#endif
 DEFINE_PER_CPU(int, x86_cpu_to_node_map) = NUMA_NO_NODE;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_node_map);
-EXPORT_SYMBOL(x86_cpu_to_node_map_early_ptr);
 
 s16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
 	[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 654724c58f5b..d973c11688ce 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -29,10 +29,15 @@ extern int smp_num_siblings;
 extern unsigned int num_processors;
 extern cpumask_t cpu_initialized;
 
+#ifdef CONFIG_SMP
 extern u16 x86_cpu_to_apicid_init[];
 extern u16 x86_bios_cpu_apicid_init[];
 extern void *x86_cpu_to_apicid_early_ptr;
 extern void *x86_bios_cpu_apicid_early_ptr;
+#else
+#define x86_cpu_to_apicid_early_ptr NULL
+#define x86_bios_cpu_apicid_early_ptr NULL
+#endif
 
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 DECLARE_PER_CPU(cpumask_t, cpu_core_map);
diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h
index 8d1a1f3d21b4..81a29eb08ac4 100644
--- a/include/asm-x86/topology.h
+++ b/include/asm-x86/topology.h
@@ -38,8 +38,13 @@ extern int cpu_to_node_map[];
 #endif
 
 DECLARE_PER_CPU(int, x86_cpu_to_node_map);
+
+#ifdef CONFIG_SMP
 extern int x86_cpu_to_node_map_init[];
 extern void *x86_cpu_to_node_map_early_ptr;
+#else
+#define x86_cpu_to_node_map_early_ptr NULL
+#endif
 
 extern cpumask_t node_to_cpumask_map[];
 
@@ -54,6 +59,8 @@ static inline int cpu_to_node(int cpu)
 }
 
 #else /* CONFIG_X86_64 */
+
+#ifdef CONFIG_SMP
 static inline int early_cpu_to_node(int cpu)
 {
 	int *cpu_to_node_map = x86_cpu_to_node_map_early_ptr;
@@ -65,6 +72,9 @@ static inline int early_cpu_to_node(int cpu)
 	else
 		return NUMA_NO_NODE;
 }
+#else
+#define	early_cpu_to_node(cpu)	cpu_to_node(cpu)
+#endif
 
 static inline int cpu_to_node(int cpu)
 {
@@ -76,10 +86,7 @@ static inline int cpu_to_node(int cpu)
 		return ((int *)x86_cpu_to_node_map_early_ptr)[cpu];
 	}
 #endif
-	if (per_cpu_offset(cpu))
-		return per_cpu(x86_cpu_to_node_map, cpu);
-	else
-		return NUMA_NO_NODE;
+	return per_cpu(x86_cpu_to_node_map, cpu);
 }
 #endif /* CONFIG_X86_64 */
 
-- 
cgit v1.2.3


From 61048c6328819b0973ef662f6d46f2e2bc753ceb Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Fri, 4 Apr 2008 23:41:07 +0400
Subject: x86: don't set IO APIC features if IO APIC is not enabled

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse_32.c             | 24 +++++++++++++++++++++---
 include/asm-x86/mach-default/mach_apic.h |  2 ++
 include/asm-x86/mpspec.h                 |  3 ---
 3 files changed, 23 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 4f4cfad9ae57..09cb77813680 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -221,6 +221,8 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 	}
 }
 
+#ifdef CONFIG_X86_IO_APIC
+
 static int bad_ioapic(unsigned long address)
 {
 	if (nr_ioapics >= MAX_IO_APICS) {
@@ -263,6 +265,8 @@ static void __init MP_intsrc_info (struct mpc_config_intsrc *m)
 		panic("Max # of irq sources exceeded!!\n");
 }
 
+#endif
+
 static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
 {
 	Dprintk("Lint: type %d, pol %d, trig %d, bus %d,"
@@ -421,21 +425,25 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 			}
 			case MP_IOAPIC:
 			{
+#ifdef CONFIG_X86_IO_APIC
 				struct mpc_config_ioapic *m=
 					(struct mpc_config_ioapic *)mpt;
 				MP_ioapic_info(m);
 				mpt+=sizeof(*m);
 				count+=sizeof(*m);
+#endif
 				break;
 			}
 			case MP_INTSRC:
 			{
+#ifdef CONFIG_X86_IO_APIC
 				struct mpc_config_intsrc *m=
 					(struct mpc_config_intsrc *)mpt;
 
 				MP_intsrc_info(m);
 				mpt+=sizeof(*m);
 				count+=sizeof(*m);
+#endif
 				break;
 			}
 			case MP_LINTSRC:
@@ -463,6 +471,8 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 	return num_processors;
 }
 
+#ifdef CONFIG_X86_IO_APIC
+
 static int __init ELCR_trigger(unsigned int irq)
 {
 	unsigned int port;
@@ -537,11 +547,15 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type)
 	MP_intsrc_info(&intsrc);
 }
 
+#endif
+
 static inline void __init construct_default_ISA_mptable(int mpc_default_type)
 {
 	struct mpc_config_processor processor;
 	struct mpc_config_bus bus;
+#ifdef CONFIG_X86_IO_APIC
 	struct mpc_config_ioapic ioapic;
+#endif
 	struct mpc_config_lintsrc lintsrc;
 	int linttypes[2] = { mp_ExtINT, mp_NMI };
 	int i;
@@ -597,6 +611,7 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type)
 		MP_bus_info(&bus);
 	}
 
+#ifdef CONFIG_X86_IO_APIC
 	ioapic.mpc_type = MP_IOAPIC;
 	ioapic.mpc_apicid = 2;
 	ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
@@ -608,7 +623,7 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type)
 	 * We set up most of the low 16 IO-APIC pins according to MPS rules.
 	 */
 	construct_default_ioirq_mptable(mpc_default_type);
-
+#endif
 	lintsrc.mpc_type = MP_LINTSRC;
 	lintsrc.mpc_irqflag = 0;		/* conforming */
 	lintsrc.mpc_srcbusid = 0;
@@ -670,6 +685,8 @@ void __init get_smp_config (void)
 			printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");
 			return;
 		}
+
+#ifdef CONFIG_X86_IO_APIC
 		/*
 		 * If there are no explicit MP IRQ entries, then we are
 		 * broken.  We set up most of the low 16 IO-APIC pins to
@@ -687,7 +704,7 @@ void __init get_smp_config (void)
 
 			construct_default_ioirq_mptable(0);
 		}
-
+#endif
 	} else
 		BUG();
 
@@ -967,8 +984,9 @@ void __init mp_config_acpi_legacy_irqs (void)
 	intsrc.mpc_type = MP_INTSRC;
 	intsrc.mpc_irqflag = 0;					/* Conforming */
 	intsrc.mpc_srcbus = MP_ISA_BUS;
+#ifdef CONFIG_X86_IO_APIC
 	intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid;
-
+#endif
 	/* 
 	 * Use the default configuration for the IRQs 0-15.  Unless
 	 * overridden by (MADT) interrupt source override entries.
diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h
index 14217a970c5e..0a6634f62abe 100644
--- a/include/asm-x86/mach-default/mach_apic.h
+++ b/include/asm-x86/mach-default/mach_apic.h
@@ -69,8 +69,10 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
 
 static inline void setup_apic_routing(void)
 {
+#ifdef CONFIG_X86_IO_APIC
 	printk("Enabling APIC mode:  %s.  Using %d I/O APICs\n",
 					"Flat", nr_ioapics);
+#endif
 }
 
 static inline int apicid_to_node(int logical_apicid)
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 1f6445b147f8..08cc5e027958 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -36,9 +36,6 @@ extern int mp_bus_id_to_pci_bus[MAX_MP_BUSSES];
 
 extern unsigned int boot_cpu_physical_apicid;
 extern int smp_found_config;
-extern int nr_ioapics;
-extern int mp_irq_entries;
-extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
 extern int mpc_default_type;
 extern unsigned long mp_lapic_addr;
 
-- 
cgit v1.2.3


From 9e5c5f1dd29c86307e6b3cfa75e85d0efccc1f6b Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Fri, 4 Apr 2008 23:41:26 +0400
Subject: x86: move mp_ioapic_routing to boot.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/boot.c  | 2 ++
 arch/x86/kernel/mpparse_32.c | 7 +------
 arch/x86/kernel/mpparse_64.c | 7 +------
 include/asm-x86/io_apic.h    | 7 +++++++
 4 files changed, 11 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index b33ebf6ea4f1..9cf575184536 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -320,6 +320,8 @@ acpi_parse_lapic_nmi(struct acpi_subtable_header * header, const unsigned long e
 
 #ifdef CONFIG_X86_IO_APIC
 
+struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS];
+
 static int __init
 acpi_parse_ioapic(struct acpi_subtable_header * header, const unsigned long end)
 {
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 302253cbfc6a..b6f1e4e235e3 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -837,12 +837,7 @@ void __cpuinit mp_register_lapic (int id, u8 enabled)
 #define MP_ISA_BUS		0
 #define MP_MAX_IOAPIC_PIN	127
 
-static struct mp_ioapic_routing {
-	int			apic_id;
-	int			gsi_base;
-	int			gsi_end;
-	u32			pin_programmed[4];
-} mp_ioapic_routing[MAX_IO_APICS];
+extern struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS];
 
 static int mp_find_ioapic (int gsi)
 {
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index f1015bf53cf0..813057cb2ddb 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -630,12 +630,7 @@ void __cpuinit mp_register_lapic(int id, u8 enabled)
 #define MP_ISA_BUS		0
 #define MP_MAX_IOAPIC_PIN	127
 
-static struct mp_ioapic_routing {
-	int apic_id;
-	int gsi_base;
-	int gsi_end;
-	u32 pin_programmed[4];
-} mp_ioapic_routing[MAX_IO_APICS];
+extern struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS];
 
 static int mp_find_ioapic(int gsi)
 {
diff --git a/include/asm-x86/io_apic.h b/include/asm-x86/io_apic.h
index 095e8e30a345..0c9e17c73e05 100644
--- a/include/asm-x86/io_apic.h
+++ b/include/asm-x86/io_apic.h
@@ -110,6 +110,13 @@ extern int nr_ioapic_registers[MAX_IO_APICS];
  * MP-BIOS irq configuration table structures:
  */
 
+struct mp_ioapic_routing {
+	int apic_id;
+	int gsi_base;
+	int gsi_end;
+	u32 pin_programmed[4];
+};
+
 /* I/O APIC entries */
 extern struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
 
-- 
cgit v1.2.3


From 2fe60147570231cde0d1f14711d2e34ccdf54b65 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Fri, 4 Apr 2008 23:41:44 +0400
Subject: x86: move up & smp variables to setup.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_32.c    |  3 ---
 arch/x86/kernel/apic_64.c    |  4 ----
 arch/x86/kernel/mpparse_32.c | 13 -------------
 arch/x86/kernel/mpparse_64.c |  9 ---------
 arch/x86/kernel/setup.c      |  8 ++++++++
 arch/x86/kernel/smpboot.c    |  5 -----
 include/asm-x86/smp.h        |  3 ++-
 7 files changed, 10 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index 4905a11b30e3..687208190b06 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -52,9 +52,6 @@
 
 unsigned long mp_lapic_addr;
 
-/* Processor that is doing the boot up */
-unsigned int boot_cpu_physical_apicid = -1U;
-
 DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid);
 
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 274ebabf49a2..9e8e5c050c55 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -87,10 +87,6 @@ static unsigned long apic_phys;
 
 unsigned long mp_lapic_addr;
 
-/* Processor that is doing the boot up */
-unsigned int boot_cpu_physical_apicid = -1U;
-EXPORT_SYMBOL(boot_cpu_physical_apicid);
-
 DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid);
 
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index 2b16e5c71a64..ed4b3bc0e97a 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -49,15 +49,6 @@ static int mp_current_pci_id;
 
 int pic_mode;
 
-/* Make it easy to share the UP and SMP code: */
-#ifndef CONFIG_X86_SMP
-unsigned int num_processors;
-unsigned disabled_cpus __cpuinitdata;
-#ifndef CONFIG_X86_LOCAL_APIC
-unsigned int boot_cpu_physical_apicid = -1U;
-#endif
-#endif
-
 /*
  * Intel MP BIOS table parsing routines:
  */
@@ -93,9 +84,7 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
 	int apicid;
 
 	if (!(m->mpc_cpuflag & CPU_ENABLED)) {
-#ifdef CONFIG_X86_SMP
 		disabled_cpus++;
-#endif
 		return;
 	}
 
@@ -817,9 +806,7 @@ void __cpuinit mp_register_lapic (int id, u8 enabled)
 	}
 
 	if (!enabled) {
-#ifdef CONFIG_X86_SMP
 		++disabled_cpus;
-#endif
 		return;
 	}
 
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 07c98dbd468a..f860727e9151 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -45,15 +45,6 @@ int mp_bus_id_to_pci_bus[MAX_MP_BUSSES] = {[0 ... MAX_MP_BUSSES - 1] = -1 };
 
 static int mp_current_pci_id = 0;
 
-/* Make it easy to share the UP and SMP code: */
-#ifndef CONFIG_X86_SMP
-unsigned int num_processors;
-unsigned disabled_cpus __cpuinitdata;
-#ifndef CONFIG_X86_LOCAL_APIC
-unsigned int boot_cpu_physical_apicid = -1U;
-#endif
-#endif
-
 /*
  * Intel MP BIOS table parsing routines:
  */
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 011fcdd213ff..ed157c90412e 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -12,6 +12,14 @@
 #include <asm/mpspec.h>
 #include <asm/apicdef.h>
 
+unsigned int num_processors;
+unsigned disabled_cpus __cpuinitdata;
+/* Processor that is doing the boot up */
+unsigned int boot_cpu_physical_apicid = -1U;
+EXPORT_SYMBOL(boot_cpu_physical_apicid);
+
+physid_mask_t phys_cpu_present_map;
+
 DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
 EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
 
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index abf63767cd46..21ad3f396a05 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -86,14 +86,9 @@ void *x86_bios_cpu_apicid_early_ptr;
 u8 apicid_2_node[MAX_APICID];
 #endif
 
-/* Internal processor count */
-unsigned int num_processors;
-
 /* State of each CPU */
 DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
-unsigned disabled_cpus __cpuinitdata;
-
 /* Store all idle threads, this can be reused instead of creating
 * a new thread. Also avoids complicated thread destroy functionality
 * for idle threads.
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index d973c11688ce..3496e1c299b2 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -128,7 +128,6 @@ int native_cpu_up(unsigned int cpunum);
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
 
-extern unsigned disabled_cpus;
 extern void prefill_possible_map(void);
 
 #define SMP_TRAMPOLINE_BASE 0x6000
@@ -144,6 +143,8 @@ static inline int num_booting_cpus(void)
 }
 #endif /* CONFIG_SMP */
 
+extern unsigned disabled_cpus __cpuinitdata;
+
 #ifdef CONFIG_X86_32_SMP
 /*
  * This function is needed by all SMP systems. It must _always_ be valid
-- 
cgit v1.2.3


From dfac2189c2e1fbb90ee83f15b5e404425754e9f4 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Fri, 4 Apr 2008 23:41:50 +0400
Subject: x86: move mp_register_lapic to boot.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/boot.c  | 21 ++++++++++++++++-----
 arch/x86/kernel/mpparse_32.c | 16 ----------------
 arch/x86/kernel/mpparse_64.c | 10 ----------
 include/asm-x86/mpspec.h     |  1 -
 4 files changed, 16 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 9cf575184536..11bd11847b19 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -39,6 +39,7 @@
 #include <asm/apic.h>
 #include <asm/io.h>
 #include <asm/mpspec.h>
+#include <asm/smp.h>
 
 #ifdef CONFIG_X86_LOCAL_APIC
 # include <mach_apic.h>
@@ -239,6 +240,16 @@ static int __init acpi_parse_madt(struct acpi_table_header *table)
 	return 0;
 }
 
+static void __cpuinit acpi_register_lapic(int id, u8 enabled)
+{
+	if (!enabled) {
+		++disabled_cpus;
+		return;
+	}
+
+	generic_processor_info(id, 0);
+}
+
 static int __init
 acpi_parse_lapic(struct acpi_subtable_header * header, const unsigned long end)
 {
@@ -258,8 +269,8 @@ acpi_parse_lapic(struct acpi_subtable_header * header, const unsigned long end)
 	 * to not preallocating memory for all NR_CPUS
 	 * when we use CPU hotplug.
 	 */
-	mp_register_lapic(processor->id,	/* APIC ID */
-			  processor->lapic_flags & ACPI_MADT_ENABLED);	/* Enabled? */
+	acpi_register_lapic(processor->id,	/* APIC ID */
+			    processor->lapic_flags & ACPI_MADT_ENABLED);
 
 	return 0;
 }
@@ -276,8 +287,8 @@ acpi_parse_sapic(struct acpi_subtable_header *header, const unsigned long end)
 
 	acpi_table_print_madt_entry(header);
 
-	mp_register_lapic((processor->id << 8) | processor->eid,/* APIC ID */
-		processor->lapic_flags & ACPI_MADT_ENABLED);	/* Enabled? */
+	acpi_register_lapic((processor->id << 8) | processor->eid,/* APIC ID */
+			    processor->lapic_flags & ACPI_MADT_ENABLED);
 
 	return 0;
 }
@@ -554,7 +565,7 @@ static int __cpuinit _acpi_map_lsapic(acpi_handle handle, int *pcpu)
 	buffer.pointer = NULL;
 
 	tmp_map = cpu_present_map;
-	mp_register_lapic(physid, lapic->lapic_flags & ACPI_MADT_ENABLED);
+	acpi_register_lapic(physid, lapic->lapic_flags & ACPI_MADT_ENABLED);
 
 	/*
 	 * If mp_register_lapic successfully generates a new logical cpu
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index ed4b3bc0e97a..ebec70a14198 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -797,22 +797,6 @@ void __init mp_register_lapic_address(u64 address)
 	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
 }
 
-void __cpuinit mp_register_lapic (int id, u8 enabled)
-{
-	if (MAX_APICS - id <= 0) {
-		printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
-			id, MAX_APICS);
-		return;
-	}
-
-	if (!enabled) {
-		++disabled_cpus;
-		return;
-	}
-
-	generic_processor_info(id, GET_APIC_VERSION(apic_read(APIC_LVR)));
-}
-
 #ifdef	CONFIG_X86_IO_APIC
 
 #define MP_ISA_BUS		0
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index f860727e9151..03c19a2e6e9e 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -601,16 +601,6 @@ void __init mp_register_lapic_address(u64 address)
 	if (boot_cpu_physical_apicid == -1U)
 		boot_cpu_physical_apicid  = GET_APIC_ID(read_apic_id());
 }
-void __cpuinit mp_register_lapic(int id, u8 enabled)
-{
-	if (!enabled) {
-		++disabled_cpus;
-		return;
-	}
-
-	generic_processor_info(id, 0);
-}
-
 
 #define MP_ISA_BUS		0
 #define MP_MAX_IOAPIC_PIN	127
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 08cc5e027958..c614051f4fc9 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -44,7 +44,6 @@ extern void get_smp_config(void);
 
 void __cpuinit generic_processor_info(int apicid, int version);
 #ifdef CONFIG_ACPI
-extern void mp_register_lapic(int id, u8 enabled);
 extern void mp_register_lapic_address(u64 address);
 extern void mp_register_ioapic(int id, u32 address, u32 gsi_base);
 extern void mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger,
-- 
cgit v1.2.3


From 31d2092eb0c23636b73d2c24c0c11b66470cef58 Mon Sep 17 00:00:00 2001
From: Alexey Starikovskiy <astarikovskiy@suse.de>
Date: Fri, 4 Apr 2008 23:41:57 +0400
Subject: x86: move mp_register_lapic_address to boot.c

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/boot.c  | 12 +++++++++++-
 arch/x86/kernel/mpparse_32.c | 12 ------------
 arch/x86/kernel/mpparse_64.c |  8 --------
 include/asm-x86/mpspec.h     |  1 -
 4 files changed, 11 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 11bd11847b19..057ccf1d5ad4 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -765,6 +765,16 @@ static int __init acpi_parse_fadt(struct acpi_table_header *table)
  * Parse LAPIC entries in MADT
  * returns 0 on success, < 0 on error
  */
+
+static void __init acpi_register_lapic_address(unsigned long address)
+{
+	mp_lapic_addr = address;
+
+	set_fixmap_nocache(FIX_APIC_BASE, address);
+	if (boot_cpu_physical_apicid == -1U)
+		boot_cpu_physical_apicid  = GET_APIC_ID(read_apic_id());
+}
+
 static int __init acpi_parse_madt_lapic_entries(void)
 {
 	int count;
@@ -786,7 +796,7 @@ static int __init acpi_parse_madt_lapic_entries(void)
 		return count;
 	}
 
-	mp_register_lapic_address(acpi_lapic_addr);
+	acpi_register_lapic_address(acpi_lapic_addr);
 
 	count = acpi_table_parse_madt(ACPI_MADT_TYPE_LOCAL_SAPIC,
 				      acpi_parse_sapic, MAX_APICS);
diff --git a/arch/x86/kernel/mpparse_32.c b/arch/x86/kernel/mpparse_32.c
index ebec70a14198..052043ed6499 100644
--- a/arch/x86/kernel/mpparse_32.c
+++ b/arch/x86/kernel/mpparse_32.c
@@ -785,18 +785,6 @@ void __init find_smp_config (void)
 
 #ifdef CONFIG_ACPI
 
-void __init mp_register_lapic_address(u64 address)
-{
-	mp_lapic_addr = (unsigned long) address;
-
-	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
-
-	if (boot_cpu_physical_apicid == -1U)
-		boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
-
-	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
-}
-
 #ifdef	CONFIG_X86_IO_APIC
 
 #define MP_ISA_BUS		0
diff --git a/arch/x86/kernel/mpparse_64.c b/arch/x86/kernel/mpparse_64.c
index 03c19a2e6e9e..1c3bf80b3ba6 100644
--- a/arch/x86/kernel/mpparse_64.c
+++ b/arch/x86/kernel/mpparse_64.c
@@ -594,14 +594,6 @@ void __init find_smp_config(void)
 
 #ifdef CONFIG_ACPI
 
-void __init mp_register_lapic_address(u64 address)
-{
-	mp_lapic_addr = (unsigned long)address;
-	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
-	if (boot_cpu_physical_apicid == -1U)
-		boot_cpu_physical_apicid  = GET_APIC_ID(read_apic_id());
-}
-
 #define MP_ISA_BUS		0
 #define MP_MAX_IOAPIC_PIN	127
 
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index c614051f4fc9..57a991b9c053 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -44,7 +44,6 @@ extern void get_smp_config(void);
 
 void __cpuinit generic_processor_info(int apicid, int version);
 #ifdef CONFIG_ACPI
-extern void mp_register_lapic_address(u64 address);
 extern void mp_register_ioapic(int id, u32 address, u32 gsi_base);
 extern void mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger,
 				   u32 gsi);
-- 
cgit v1.2.3


From e44b7b7525ad9d43163ab5e60c784325419e0ea6 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@suse.cz>
Date: Thu, 10 Apr 2008 23:28:10 +0200
Subject: x86: move suspend wakeup code to C

Move wakeup code to .c, so that video mode setting code can be shared
between boot and wakeup. Remove nasty assembly code in 64-bit case by
re-using trampoline code. Stack setup was fixed to clear high 16bits
of %esp, maybe that fixes some machines.

.c code sharing and morse code was done H. Peter Anvin, Sam Ravnborg
reviewed kbuild related stuff, and it seems okay to him. Rafael did
some cleanups.

[rjw:
* Made the patch stop breaking compilation on x86-32
* Added arch/x86/kernel/acpi/sleep.h
* Got rid of compiler warnings in arch/x86/kernel/acpi/sleep.c
* Fixed 32-bit compilation on x86-64 systems
* Added include/asm-x86/trampoline.h and fixed the non-SMP
  compilation on 64-bit x86
* Removed arch/x86/kernel/acpi/sleep_32.c which was not used
* Fixed some breakage caused by the integration of smpboot.c done
  under us in the meantime]

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/Kconfig                           |   2 +-
 arch/x86/boot/Makefile                     |   2 +-
 arch/x86/boot/boot.h                       |   5 +
 arch/x86/boot/video-bios.c                 |   6 +
 arch/x86/boot/video-mode.c                 | 173 ++++++++++++++++
 arch/x86/boot/video-vesa.c                 |   8 +
 arch/x86/boot/video-vga.c                  |  12 +-
 arch/x86/boot/video.c                      | 157 +--------------
 arch/x86/kernel/acpi/Makefile              |   9 +-
 arch/x86/kernel/acpi/realmode/Makefile     |  57 ++++++
 arch/x86/kernel/acpi/realmode/copy.S       |   1 +
 arch/x86/kernel/acpi/realmode/video-bios.c |   1 +
 arch/x86/kernel/acpi/realmode/video-mode.c |   1 +
 arch/x86/kernel/acpi/realmode/video-vesa.c |   1 +
 arch/x86/kernel/acpi/realmode/video-vga.c  |   1 +
 arch/x86/kernel/acpi/realmode/wakemain.c   |  81 ++++++++
 arch/x86/kernel/acpi/realmode/wakeup.S     | 113 +++++++++++
 arch/x86/kernel/acpi/realmode/wakeup.h     |  36 ++++
 arch/x86/kernel/acpi/realmode/wakeup.lds.S |  61 ++++++
 arch/x86/kernel/acpi/sleep.c               |  71 +++++--
 arch/x86/kernel/acpi/sleep.h               |  16 ++
 arch/x86/kernel/acpi/sleep_32.c            |  40 ----
 arch/x86/kernel/acpi/wakeup_32.S           | 247 ++---------------------
 arch/x86/kernel/acpi/wakeup_64.S           | 313 +----------------------------
 arch/x86/kernel/acpi/wakeup_rm.S           |  10 +
 arch/x86/kernel/e820_64.c                  |   5 +-
 arch/x86/kernel/head_64.S                  |   4 -
 arch/x86/kernel/setup_32.c                 |   4 +-
 arch/x86/kernel/setup_64.c                 |   1 +
 arch/x86/kernel/smpboot.c                  |   6 +-
 arch/x86/kernel/trampoline_64.S            |   5 -
 arch/x86/mach-voyager/voyager_smp.c        |   1 +
 include/asm-x86/smp.h                      |  13 --
 include/asm-x86/trampoline.h               |  21 ++
 34 files changed, 711 insertions(+), 773 deletions(-)
 create mode 100644 arch/x86/boot/video-mode.c
 create mode 100644 arch/x86/kernel/acpi/realmode/Makefile
 create mode 100644 arch/x86/kernel/acpi/realmode/copy.S
 create mode 100644 arch/x86/kernel/acpi/realmode/video-bios.c
 create mode 100644 arch/x86/kernel/acpi/realmode/video-mode.c
 create mode 100644 arch/x86/kernel/acpi/realmode/video-vesa.c
 create mode 100644 arch/x86/kernel/acpi/realmode/video-vga.c
 create mode 100644 arch/x86/kernel/acpi/realmode/wakemain.c
 create mode 100644 arch/x86/kernel/acpi/realmode/wakeup.S
 create mode 100644 arch/x86/kernel/acpi/realmode/wakeup.h
 create mode 100644 arch/x86/kernel/acpi/realmode/wakeup.lds.S
 create mode 100644 arch/x86/kernel/acpi/sleep.h
 delete mode 100644 arch/x86/kernel/acpi/sleep_32.c
 create mode 100644 arch/x86/kernel/acpi/wakeup_rm.S
 create mode 100644 include/asm-x86/trampoline.h

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 0a7193ae45ed..4e32b6f7d31a 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -181,7 +181,7 @@ config X86_BIOS_REBOOT
 
 config X86_TRAMPOLINE
 	bool
-	depends on X86_SMP || (X86_VOYAGER && SMP)
+	depends on X86_SMP || (X86_VOYAGER && SMP) || (64BIT && ACPI_SLEEP)
 	default y
 
 config KTIME_SCALAR
diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile
index 9695affeb584..7ee102f9c4f8 100644
--- a/arch/x86/boot/Makefile
+++ b/arch/x86/boot/Makefile
@@ -30,7 +30,7 @@ subdir-		:= compressed
 
 setup-y		+= a20.o cmdline.o copy.o cpu.o cpucheck.o edd.o
 setup-y		+= header.o main.o mca.o memory.o pm.o pmjump.o
-setup-y		+= printf.o string.o tty.o video.o version.o
+setup-y		+= printf.o string.o tty.o video.o video-mode.o version.o
 setup-$(CONFIG_X86_APM_BOOT) += apm.o
 setup-$(CONFIG_X86_VOYAGER) += voyager.o
 
diff --git a/arch/x86/boot/boot.h b/arch/x86/boot/boot.h
index 7822a4983da2..09578070bfba 100644
--- a/arch/x86/boot/boot.h
+++ b/arch/x86/boot/boot.h
@@ -286,6 +286,11 @@ int getchar_timeout(void);
 /* video.c */
 void set_video(void);
 
+/* video-mode.c */
+int set_mode(u16 mode);
+int mode_defined(u16 mode);
+void probe_cards(int unsafe);
+
 /* video-vesa.c */
 void vesa_store_edid(void);
 
diff --git a/arch/x86/boot/video-bios.c b/arch/x86/boot/video-bios.c
index ff664a117096..39e247e96172 100644
--- a/arch/x86/boot/video-bios.c
+++ b/arch/x86/boot/video-bios.c
@@ -50,6 +50,7 @@ static int set_bios_mode(u8 mode)
 	if (new_mode == mode)
 		return 0;	/* Mode change OK */
 
+#ifndef _WAKEUP
 	if (new_mode != boot_params.screen_info.orig_video_mode) {
 		/* Mode setting failed, but we didn't end up where we
 		   started.  That's bad.  Try to revert to the original
@@ -59,13 +60,18 @@ static int set_bios_mode(u8 mode)
 			     : "+a" (ax)
 			     : : "ebx", "ecx", "edx", "esi", "edi");
 	}
+#endif
 	return -1;
 }
 
 static int bios_probe(void)
 {
 	u8 mode;
+#ifdef _WAKEUP
+	u8 saved_mode = 0x03;
+#else
 	u8 saved_mode = boot_params.screen_info.orig_video_mode;
+#endif
 	u16 crtc;
 	struct mode_info *mi;
 	int nmodes = 0;
diff --git a/arch/x86/boot/video-mode.c b/arch/x86/boot/video-mode.c
new file mode 100644
index 000000000000..748e8d06290a
--- /dev/null
+++ b/arch/x86/boot/video-mode.c
@@ -0,0 +1,173 @@
+/* -*- linux-c -*- ------------------------------------------------------- *
+ *
+ *   Copyright (C) 1991, 1992 Linus Torvalds
+ *   Copyright 2007-2008 rPath, Inc. - All Rights Reserved
+ *
+ *   This file is part of the Linux kernel, and is made available under
+ *   the terms of the GNU General Public License version 2.
+ *
+ * ----------------------------------------------------------------------- */
+
+/*
+ * arch/i386/boot/video-mode.c
+ *
+ * Set the video mode.  This is separated out into a different
+ * file in order to be shared with the ACPI wakeup code.
+ */
+
+#include "boot.h"
+#include "video.h"
+#include "vesa.h"
+
+/*
+ * Common variables
+ */
+int adapter;			/* 0=CGA/MDA/HGC, 1=EGA, 2=VGA+ */
+u16 video_segment;
+int force_x, force_y;	/* Don't query the BIOS for cols/rows */
+
+int do_restore;		/* Screen contents changed during mode flip */
+int graphic_mode;	/* Graphic mode with linear frame buffer */
+
+/* Probe the video drivers and have them generate their mode lists. */
+void probe_cards(int unsafe)
+{
+	struct card_info *card;
+	static u8 probed[2];
+
+	if (probed[unsafe])
+		return;
+
+	probed[unsafe] = 1;
+
+	for (card = video_cards; card < video_cards_end; card++) {
+		if (card->unsafe == unsafe) {
+			if (card->probe)
+				card->nmodes = card->probe();
+			else
+				card->nmodes = 0;
+		}
+	}
+}
+
+/* Test if a mode is defined */
+int mode_defined(u16 mode)
+{
+	struct card_info *card;
+	struct mode_info *mi;
+	int i;
+
+	for (card = video_cards; card < video_cards_end; card++) {
+		mi = card->modes;
+		for (i = 0; i < card->nmodes; i++, mi++) {
+			if (mi->mode == mode)
+				return 1;
+		}
+	}
+
+	return 0;
+}
+
+/* Set mode (without recalc) */
+static int raw_set_mode(u16 mode, u16 *real_mode)
+{
+	int nmode, i;
+	struct card_info *card;
+	struct mode_info *mi;
+
+	/* Drop the recalc bit if set */
+	mode &= ~VIDEO_RECALC;
+
+	/* Scan for mode based on fixed ID, position, or resolution */
+	nmode = 0;
+	for (card = video_cards; card < video_cards_end; card++) {
+		mi = card->modes;
+		for (i = 0; i < card->nmodes; i++, mi++) {
+			int visible = mi->x || mi->y;
+
+			if ((mode == nmode && visible) ||
+			    mode == mi->mode ||
+			    mode == (mi->y << 8)+mi->x) {
+				*real_mode = mi->mode;
+				return card->set_mode(mi);
+			}
+
+			if (visible)
+				nmode++;
+		}
+	}
+
+	/* Nothing found?  Is it an "exceptional" (unprobed) mode? */
+	for (card = video_cards; card < video_cards_end; card++) {
+		if (mode >= card->xmode_first &&
+		    mode < card->xmode_first+card->xmode_n) {
+			struct mode_info mix;
+			*real_mode = mix.mode = mode;
+			mix.x = mix.y = 0;
+			return card->set_mode(&mix);
+		}
+	}
+
+	/* Otherwise, failure... */
+	return -1;
+}
+
+/*
+ * Recalculate the vertical video cutoff (hack!)
+ */
+static void vga_recalc_vertical(void)
+{
+	unsigned int font_size, rows;
+	u16 crtc;
+	u8 pt, ov;
+
+	set_fs(0);
+	font_size = rdfs8(0x485); /* BIOS: font size (pixels) */
+	rows = force_y ? force_y : rdfs8(0x484)+1; /* Text rows */
+
+	rows *= font_size;	/* Visible scan lines */
+	rows--;			/* ... minus one */
+
+	crtc = vga_crtc();
+
+	pt = in_idx(crtc, 0x11);
+	pt &= ~0x80;		/* Unlock CR0-7 */
+	out_idx(pt, crtc, 0x11);
+
+	out_idx((u8)rows, crtc, 0x12); /* Lower height register */
+
+	ov = in_idx(crtc, 0x07); /* Overflow register */
+	ov &= 0xbd;
+	ov |= (rows >> (8-1)) & 0x02;
+	ov |= (rows >> (9-6)) & 0x40;
+	out_idx(ov, crtc, 0x07);
+}
+
+/* Set mode (with recalc if specified) */
+int set_mode(u16 mode)
+{
+	int rv;
+	u16 real_mode;
+
+	/* Very special mode numbers... */
+	if (mode == VIDEO_CURRENT_MODE)
+		return 0;	/* Nothing to do... */
+	else if (mode == NORMAL_VGA)
+		mode = VIDEO_80x25;
+	else if (mode == EXTENDED_VGA)
+		mode = VIDEO_8POINT;
+
+	rv = raw_set_mode(mode, &real_mode);
+	if (rv)
+		return rv;
+
+	if (mode & VIDEO_RECALC)
+		vga_recalc_vertical();
+
+	/* Save the canonical mode number for the kernel, not
+	   an alias, size specification or menu position */
+#ifndef _WAKEUP
+	boot_params.hdr.vid_mode = real_mode;
+#endif
+	return 0;
+}
diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c
index 419b5c273374..5d5a3f6e8b5c 100644
--- a/arch/x86/boot/video-vesa.c
+++ b/arch/x86/boot/video-vesa.c
@@ -24,7 +24,11 @@ static struct vesa_mode_info vminfo;
 
 __videocard video_vesa;
 
+#ifndef _WAKEUP
 static void vesa_store_mode_params_graphics(void);
+#else /* _WAKEUP */
+static inline void vesa_store_mode_params_graphics(void) {}
+#endif /* _WAKEUP */
 
 static int vesa_probe(void)
 {
@@ -165,6 +169,8 @@ static int vesa_set_mode(struct mode_info *mode)
 }
 
 
+#ifndef _WAKEUP
+
 /* Switch DAC to 8-bit mode */
 static void vesa_dac_set_8bits(void)
 {
@@ -288,6 +294,8 @@ void vesa_store_edid(void)
 #endif /* CONFIG_FIRMWARE_EDID */
 }
 
+#endif /* not _WAKEUP */
+
 __videocard video_vesa =
 {
 	.card_name	= "VESA",
diff --git a/arch/x86/boot/video-vga.c b/arch/x86/boot/video-vga.c
index 7259387b7d19..330d6589a2ad 100644
--- a/arch/x86/boot/video-vga.c
+++ b/arch/x86/boot/video-vga.c
@@ -210,6 +210,8 @@ static int vga_set_mode(struct mode_info *mode)
  */
 static int vga_probe(void)
 {
+	u16 ega_bx;
+
 	static const char *card_name[] = {
 		"CGA/MDA/HGC", "EGA", "VGA"
 	};
@@ -226,12 +228,16 @@ static int vga_probe(void)
 	u8 vga_flag;
 
 	asm(INT10
-	    : "=b" (boot_params.screen_info.orig_video_ega_bx)
+	    : "=b" (ega_bx)
 	    : "a" (0x1200), "b" (0x10) /* Check EGA/VGA */
 	    : "ecx", "edx", "esi", "edi");
 
+#ifndef _WAKEUP
+	boot_params.screen_info.orig_video_ega_bx = ega_bx;
+#endif
+
 	/* If we have MDA/CGA/HGC then BL will be unchanged at 0x10 */
-	if ((u8)boot_params.screen_info.orig_video_ega_bx != 0x10) {
+	if ((u8)ega_bx != 0x10) {
 		/* EGA/VGA */
 		asm(INT10
 		    : "=a" (vga_flag)
@@ -240,7 +246,9 @@ static int vga_probe(void)
 
 		if (vga_flag == 0x1a) {
 			adapter = ADAPTER_VGA;
+#ifndef _WAKEUP
 			boot_params.screen_info.orig_video_isVGA = 1;
+#endif
 		} else {
 			adapter = ADAPTER_EGA;
 		}
diff --git a/arch/x86/boot/video.c b/arch/x86/boot/video.c
index 696d08f3843c..c1c47ba069ef 100644
--- a/arch/x86/boot/video.c
+++ b/arch/x86/boot/video.c
@@ -18,21 +18,6 @@
 #include "video.h"
 #include "vesa.h"
 
-/*
- * Mode list variables
- */
-static struct card_info cards[];    /* List of cards to probe for */
-
-/*
- * Common variables
- */
-int adapter;			/* 0=CGA/MDA/HGC, 1=EGA, 2=VGA+ */
-u16 video_segment;
-int force_x, force_y;	/* Don't query the BIOS for cols/rows */
-
-int do_restore = 0;	/* Screen contents changed during mode flip */
-int graphic_mode;	/* Graphic mode with linear frame buffer */
-
 static void store_cursor_position(void)
 {
 	u16 curpos;
@@ -107,147 +92,6 @@ static void store_mode_params(void)
 	boot_params.screen_info.orig_video_lines = y;
 }
 
-/* Probe the video drivers and have them generate their mode lists. */
-static void probe_cards(int unsafe)
-{
-	struct card_info *card;
-	static u8 probed[2];
-
-	if (probed[unsafe])
-		return;
-
-	probed[unsafe] = 1;
-
-	for (card = video_cards; card < video_cards_end; card++) {
-		if (card->unsafe == unsafe) {
-			if (card->probe)
-				card->nmodes = card->probe();
-			else
-				card->nmodes = 0;
-		}
-	}
-}
-
-/* Test if a mode is defined */
-int mode_defined(u16 mode)
-{
-	struct card_info *card;
-	struct mode_info *mi;
-	int i;
-
-	for (card = video_cards; card < video_cards_end; card++) {
-		mi = card->modes;
-		for (i = 0; i < card->nmodes; i++, mi++) {
-			if (mi->mode == mode)
-				return 1;
-		}
-	}
-
-	return 0;
-}
-
-/* Set mode (without recalc) */
-static int raw_set_mode(u16 mode, u16 *real_mode)
-{
-	int nmode, i;
-	struct card_info *card;
-	struct mode_info *mi;
-
-	/* Drop the recalc bit if set */
-	mode &= ~VIDEO_RECALC;
-
-	/* Scan for mode based on fixed ID, position, or resolution */
-	nmode = 0;
-	for (card = video_cards; card < video_cards_end; card++) {
-		mi = card->modes;
-		for (i = 0; i < card->nmodes; i++, mi++) {
-			int visible = mi->x || mi->y;
-
-			if ((mode == nmode && visible) ||
-			    mode == mi->mode ||
-			    mode == (mi->y << 8)+mi->x) {
-				*real_mode = mi->mode;
-				return card->set_mode(mi);
-			}
-
-			if (visible)
-				nmode++;
-		}
-	}
-
-	/* Nothing found?  Is it an "exceptional" (unprobed) mode? */
-	for (card = video_cards; card < video_cards_end; card++) {
-		if (mode >= card->xmode_first &&
-		    mode < card->xmode_first+card->xmode_n) {
-			struct mode_info mix;
-			*real_mode = mix.mode = mode;
-			mix.x = mix.y = 0;
-			return card->set_mode(&mix);
-		}
-	}
-
-	/* Otherwise, failure... */
-	return -1;
-}
-
-/*
- * Recalculate the vertical video cutoff (hack!)
- */
-static void vga_recalc_vertical(void)
-{
-	unsigned int font_size, rows;
-	u16 crtc;
-	u8 pt, ov;
-
-	set_fs(0);
-	font_size = rdfs8(0x485); /* BIOS: font size (pixels) */
-	rows = force_y ? force_y : rdfs8(0x484)+1; /* Text rows */
-
-	rows *= font_size;	/* Visible scan lines */
-	rows--;			/* ... minus one */
-
-	crtc = vga_crtc();
-
-	pt = in_idx(crtc, 0x11);
-	pt &= ~0x80;		/* Unlock CR0-7 */
-	out_idx(pt, crtc, 0x11);
-
-	out_idx((u8)rows, crtc, 0x12); /* Lower height register */
-
-	ov = in_idx(crtc, 0x07); /* Overflow register */
-	ov &= 0xbd;
-	ov |= (rows >> (8-1)) & 0x02;
-	ov |= (rows >> (9-6)) & 0x40;
-	out_idx(ov, crtc, 0x07);
-}
-
-/* Set mode (with recalc if specified) */
-static int set_mode(u16 mode)
-{
-	int rv;
-	u16 real_mode;
-
-	/* Very special mode numbers... */
-	if (mode == VIDEO_CURRENT_MODE)
-		return 0;	/* Nothing to do... */
-	else if (mode == NORMAL_VGA)
-		mode = VIDEO_80x25;
-	else if (mode == EXTENDED_VGA)
-		mode = VIDEO_8POINT;
-
-	rv = raw_set_mode(mode, &real_mode);
-	if (rv)
-		return rv;
-
-	if (mode & VIDEO_RECALC)
-		vga_recalc_vertical();
-
-	/* Save the canonical mode number for the kernel, not
-	   an alias, size specification or menu position */
-	boot_params.hdr.vid_mode = real_mode;
-	return 0;
-}
-
 static unsigned int get_entry(void)
 {
 	char entry_buf[4];
@@ -486,6 +330,7 @@ void set_video(void)
 		printf("Undefined video mode number: %x\n", mode);
 		mode = ASK_VGA;
 	}
+	boot_params.hdr.vid_mode = mode;
 	vesa_store_edid();
 	store_mode_params();
 
diff --git a/arch/x86/kernel/acpi/Makefile b/arch/x86/kernel/acpi/Makefile
index 19d3d6e9d09b..7335959b6aff 100644
--- a/arch/x86/kernel/acpi/Makefile
+++ b/arch/x86/kernel/acpi/Makefile
@@ -1,7 +1,14 @@
+subdir-				:= realmode
+
 obj-$(CONFIG_ACPI)		+= boot.o
-obj-$(CONFIG_ACPI_SLEEP)	+= sleep.o wakeup_$(BITS).o
+obj-$(CONFIG_ACPI_SLEEP)	+= sleep.o wakeup_rm.o wakeup_$(BITS).o
 
 ifneq ($(CONFIG_ACPI_PROCESSOR),)
 obj-y				+= cstate.o processor.o
 endif
 
+$(obj)/wakeup_rm.o:    $(obj)/realmode/wakeup.bin
+
+$(obj)/realmode/wakeup.bin: FORCE
+	$(Q)$(MAKE) $(build)=$(obj)/realmode $@
+
diff --git a/arch/x86/kernel/acpi/realmode/Makefile b/arch/x86/kernel/acpi/realmode/Makefile
new file mode 100644
index 000000000000..092900854acc
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/Makefile
@@ -0,0 +1,57 @@
+#
+# arch/x86/kernel/acpi/realmode/Makefile
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+
+targets		:= wakeup.bin wakeup.elf
+
+wakeup-y	+= wakeup.o wakemain.o video-mode.o copy.o
+
+# The link order of the video-*.o modules can matter.  In particular,
+# video-vga.o *must* be listed first, followed by video-vesa.o.
+# Hardware-specific drivers should follow in the order they should be
+# probed, and video-bios.o should typically be last.
+wakeup-y	+= video-vga.o
+wakeup-y	+= video-vesa.o
+wakeup-y	+= video-bios.o
+
+targets		+= $(wakeup-y)
+
+bootsrc		:= $(src)/../../../boot
+
+# ---------------------------------------------------------------------------
+
+# How to compile the 16-bit code.  Note we always compile for -march=i386,
+# that way we can complain to the user if the CPU is insufficient.
+# Compile with _SETUP since this is similar to the boot-time setup code.
+KBUILD_CFLAGS	:= $(LINUXINCLUDE) -g -Os -D_SETUP -D_WAKEUP -D__KERNEL__ \
+		   -I$(srctree)/$(bootsrc) \
+		   $(cflags-y) \
+		   -Wall -Wstrict-prototypes \
+		   -march=i386 -mregparm=3 \
+		   -include $(srctree)/$(bootsrc)/code16gcc.h \
+		   -fno-strict-aliasing -fomit-frame-pointer \
+		   $(call cc-option, -ffreestanding) \
+		   $(call cc-option, -fno-toplevel-reorder,\
+			$(call cc-option, -fno-unit-at-a-time)) \
+		   $(call cc-option, -fno-stack-protector) \
+		   $(call cc-option, -mpreferred-stack-boundary=2)
+KBUILD_CFLAGS	+= $(call cc-option, -m32)
+KBUILD_AFLAGS	:= $(KBUILD_CFLAGS) -D__ASSEMBLY__
+
+WAKEUP_OBJS = $(addprefix $(obj)/,$(wakeup-y))
+
+LDFLAGS_wakeup.elf	:= -T
+
+CPPFLAGS_wakeup.lds += -P -C
+
+$(obj)/wakeup.elf: $(src)/wakeup.lds $(WAKEUP_OBJS) FORCE
+	$(call if_changed,ld)
+
+OBJCOPYFLAGS_wakeup.bin	:= -O binary
+
+$(obj)/wakeup.bin: $(obj)/wakeup.elf FORCE
+	$(call if_changed,objcopy)
diff --git a/arch/x86/kernel/acpi/realmode/copy.S b/arch/x86/kernel/acpi/realmode/copy.S
new file mode 100644
index 000000000000..dc59ebee69d8
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/copy.S
@@ -0,0 +1 @@
+#include "../../../boot/copy.S"
diff --git a/arch/x86/kernel/acpi/realmode/video-bios.c b/arch/x86/kernel/acpi/realmode/video-bios.c
new file mode 100644
index 000000000000..7deabc144a27
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/video-bios.c
@@ -0,0 +1 @@
+#include "../../../boot/video-bios.c"
diff --git a/arch/x86/kernel/acpi/realmode/video-mode.c b/arch/x86/kernel/acpi/realmode/video-mode.c
new file mode 100644
index 000000000000..328ad209f113
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/video-mode.c
@@ -0,0 +1 @@
+#include "../../../boot/video-mode.c"
diff --git a/arch/x86/kernel/acpi/realmode/video-vesa.c b/arch/x86/kernel/acpi/realmode/video-vesa.c
new file mode 100644
index 000000000000..9dbb9672226a
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/video-vesa.c
@@ -0,0 +1 @@
+#include "../../../boot/video-vesa.c"
diff --git a/arch/x86/kernel/acpi/realmode/video-vga.c b/arch/x86/kernel/acpi/realmode/video-vga.c
new file mode 100644
index 000000000000..bcc81255f374
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/video-vga.c
@@ -0,0 +1 @@
+#include "../../../boot/video-vga.c"
diff --git a/arch/x86/kernel/acpi/realmode/wakemain.c b/arch/x86/kernel/acpi/realmode/wakemain.c
new file mode 100644
index 000000000000..883962d9eef2
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/wakemain.c
@@ -0,0 +1,81 @@
+#include "wakeup.h"
+#include "boot.h"
+
+static void udelay(int loops)
+{
+	while (loops--)
+		io_delay();	/* Approximately 1 us */
+}
+
+static void beep(unsigned int hz)
+{
+	u8 enable;
+
+	if (!hz) {
+		enable = 0x00;		/* Turn off speaker */
+	} else {
+		u16 div = 1193181/hz;
+
+		outb(0xb6, 0x43);	/* Ctr 2, squarewave, load, binary */
+		io_delay();
+		outb(div, 0x42);	/* LSB of counter */
+		io_delay();
+		outb(div >> 8, 0x42);	/* MSB of counter */
+		io_delay();
+
+		enable = 0x03;		/* Turn on speaker */
+	}
+	inb(0x61);		/* Dummy read of System Control Port B */
+	io_delay();
+	outb(enable, 0x61);	/* Enable timer 2 output to speaker */
+	io_delay();
+}
+
+#define DOT_HZ		880
+#define DASH_HZ		587
+#define US_PER_DOT	125000
+
+/* Okay, this is totally silly, but it's kind of fun. */
+static void send_morse(const char *pattern)
+{
+	char s;
+
+	while ((s = *pattern++)) {
+		switch (s) {
+		case '.':
+			beep(DOT_HZ);
+			udelay(US_PER_DOT);
+			beep(0);
+			udelay(US_PER_DOT);
+			break;
+		case '-':
+			beep(DASH_HZ);
+			udelay(US_PER_DOT * 3);
+			beep(0);
+			udelay(US_PER_DOT);
+			break;
+		default:	/* Assume it's a space */
+			udelay(US_PER_DOT * 3);
+			break;
+		}
+	}
+}
+
+void main(void)
+{
+	/* Kill machine if structures are wrong */
+	if (wakeup_header.real_magic != 0x12345678)
+		while (1);
+
+	if (wakeup_header.realmode_flags & 4)
+		send_morse("...-");
+
+	if (wakeup_header.realmode_flags & 1)
+		asm volatile("lcallw   $0xc000,$3");
+
+	if (wakeup_header.realmode_flags & 2) {
+		/* Need to call BIOS */
+		probe_cards(0);
+		set_mode(wakeup_header.video_mode);
+	}
+}
diff --git a/arch/x86/kernel/acpi/realmode/wakeup.S b/arch/x86/kernel/acpi/realmode/wakeup.S
new file mode 100644
index 000000000000..f9b77fb37e5b
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/wakeup.S
@@ -0,0 +1,113 @@
+/*
+ * ACPI wakeup real mode startup stub
+ */
+#include <asm/segment.h>
+#include <asm/msr-index.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+	.code16
+	.section ".header", "a"
+
+/* This should match the structure in wakeup.h */
+		.globl	wakeup_header
+wakeup_header:
+video_mode:	.short	0	/* Video mode number */
+pmode_return:	.byte	0x66, 0xea	/* ljmpl */
+		.long	0	/* offset goes here */
+		.short	__KERNEL_CS
+pmode_cr0:	.long	0	/* Saved %cr0 */
+pmode_cr3:	.long	0	/* Saved %cr3 */
+pmode_cr4:	.long	0	/* Saved %cr4 */
+pmode_efer:	.quad	0	/* Saved EFER */
+pmode_gdt:	.quad	0
+realmode_flags:	.long	0
+real_magic:	.long	0
+trampoline_segment:	.word 0
+signature:	.long	0x51ee1111
+
+	.text
+	.globl	_start
+	.code16
+wakeup_code:
+_start:
+	cli
+	cld
+
+	/* Set up segments */
+	movw	%cs, %ax
+	movw	%ax, %ds
+	movw	%ax, %es
+	movw	%ax, %ss
+
+	movl	$wakeup_stack_end, %esp
+
+	/* Clear the EFLAGS */
+	pushl	$0
+	popfl
+
+	/* Check header signature... */
+	movl	signature, %eax
+	cmpl	$0x51ee1111, %eax
+	jne	bogus_real_magic
+
+	/* Check we really have everything... */
+	movl	end_signature, %eax
+	cmpl	$0x65a22c82, %eax
+	jne	bogus_real_magic
+
+	/* Call the C code */
+	calll	main
+
+	/* Do any other stuff... */
+
+#ifndef CONFIG_64BIT
+	/* This could also be done in C code... */
+	movl	pmode_cr3, %eax
+	movl	%eax, %cr3
+
+	movl	pmode_cr4, %ecx
+	jecxz	1f
+	movl	%ecx, %cr4
+1:
+	movl	pmode_efer, %eax
+	movl	pmode_efer + 4, %edx
+	movl	%eax, %ecx
+	orl	%edx, %ecx
+	jz	1f
+	movl	$0xc0000080, %ecx
+	wrmsr
+1:
+
+	lgdtl	pmode_gdt
+
+	/* This really couldn't... */
+	movl	pmode_cr0, %eax
+	movl	%eax, %cr0
+	jmp	pmode_return
+#else
+	pushw	$0
+	pushw	trampoline_segment
+	pushw	$0
+	lret
+#endif
+
+bogus_real_magic:
+1:
+	hlt
+	jmp	1b
+
+	.data
+	.balign	4
+	.globl	HEAP, heap_end
+HEAP:
+	.long	wakeup_heap
+heap_end:
+	.long	wakeup_stack
+
+	.bss
+wakeup_heap:
+	.space	2048
+wakeup_stack:
+	.space	2048
+wakeup_stack_end:
diff --git a/arch/x86/kernel/acpi/realmode/wakeup.h b/arch/x86/kernel/acpi/realmode/wakeup.h
new file mode 100644
index 000000000000..ef8166fe8020
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/wakeup.h
@@ -0,0 +1,36 @@
+/*
+ * Definitions for the wakeup data structure at the head of the
+ * wakeup code.
+ */
+
+#ifndef ARCH_X86_KERNEL_ACPI_RM_WAKEUP_H
+#define ARCH_X86_KERNEL_ACPI_RM_WAKEUP_H
+
+#ifndef __ASSEMBLY__
+#include <linux/types.h>
+
+/* This must match data at wakeup.S */
+struct wakeup_header {
+	u16 video_mode;		/* Video mode number */
+	u16 _jmp1;		/* ljmpl opcode, 32-bit only */
+	u32 pmode_entry;	/* Protected mode resume point, 32-bit only */
+	u16 _jmp2;		/* CS value, 32-bit only */
+	u32 pmode_cr0;		/* Protected mode cr0 */
+	u32 pmode_cr3;		/* Protected mode cr3 */
+	u32 pmode_cr4;		/* Protected mode cr4 */
+	u32 pmode_efer_low;	/* Protected mode EFER */
+	u32 pmode_efer_high;
+	u64 pmode_gdt;
+	u32 realmode_flags;
+	u32 real_magic;
+	u16 trampoline_segment;	/* segment with trampoline code, 64-bit only */
+	u32 signature;		/* To check we have correct structure */
+} __attribute__((__packed__));
+
+extern struct wakeup_header wakeup_header;
+#endif
+
+#define HEADER_OFFSET 0x3f00
+#define WAKEUP_SIZE   0x4000
+
+#endif /* ARCH_X86_KERNEL_ACPI_RM_WAKEUP_H */
diff --git a/arch/x86/kernel/acpi/realmode/wakeup.lds.S b/arch/x86/kernel/acpi/realmode/wakeup.lds.S
new file mode 100644
index 000000000000..22fab6c4be15
--- /dev/null
+++ b/arch/x86/kernel/acpi/realmode/wakeup.lds.S
@@ -0,0 +1,61 @@
+/*
+ * wakeup.ld
+ *
+ * Linker script for the real-mode wakeup code
+ */
+#undef i386
+#include "wakeup.h"
+
+OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
+OUTPUT_ARCH(i386)
+ENTRY(_start)
+
+SECTIONS
+{
+	. = HEADER_OFFSET;
+	.header : {
+		 *(.header)
+	}
+
+	. = 0;
+	.text : {
+		 *(.text*)
+	}
+
+	. = ALIGN(16);
+	.rodata : {
+		*(.rodata*)
+	}
+
+	.videocards : {
+		video_cards = .;
+		*(.videocards)
+		video_cards_end = .;
+	}
+
+	. = ALIGN(16);
+	.data : {
+		 *(.data*)
+	}
+
+	.signature : {
+		end_signature = .;
+		LONG(0x65a22c82)
+	}
+
+	. = ALIGN(16);
+	.bss :	{
+		__bss_start = .;
+		*(.bss)
+		__bss_end = .;
+	}
+
+	. = ALIGN(16);
+	_end = .;
+
+	/DISCARD/ : {
+		*(.note*)
+	}
+
+	. = ASSERT(_end <= WAKEUP_SIZE, "Wakeup too big!");
+}
diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index dd78326ae47c..afc25ee9964b 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -10,30 +10,72 @@
 #include <linux/dmi.h>
 #include <linux/cpumask.h>
 
-#include <asm/smp.h>
+#include "realmode/wakeup.h"
+#include "sleep.h"
 
-/* address in low memory of the wakeup routine. */
 unsigned long acpi_wakeup_address;
 unsigned long acpi_realmode_flags;
-extern char wakeup_start, wakeup_end;
 
-extern unsigned long acpi_copy_wakeup_routine(unsigned long);
+/* address in low memory of the wakeup routine. */
+static unsigned long acpi_realmode;
+
+#ifdef CONFIG_64BIT
+static char temp_stack[10240];
+#endif
 
 /**
  * acpi_save_state_mem - save kernel state
  *
  * Create an identity mapped page table and copy the wakeup routine to
  * low memory.
+ *
+ * Note that this is too late to change acpi_wakeup_address.
  */
 int acpi_save_state_mem(void)
 {
-	if (!acpi_wakeup_address) {
-		printk(KERN_ERR "Could not allocate memory during boot, S3 disabled\n");
+	struct wakeup_header *header;
+
+	if (!acpi_realmode) {
+		printk(KERN_ERR "Could not allocate memory during boot, "
+		       "S3 disabled\n");
 		return -ENOMEM;
 	}
-	memcpy((void *)acpi_wakeup_address, &wakeup_start,
-	       &wakeup_end - &wakeup_start);
-	acpi_copy_wakeup_routine(acpi_wakeup_address);
+	memcpy((void *)acpi_realmode, &wakeup_code_start, WAKEUP_SIZE);
+
+	header = (struct wakeup_header *)(acpi_realmode + HEADER_OFFSET);
+	if (header->signature != 0x51ee1111) {
+		printk(KERN_ERR "wakeup header does not match\n");
+		return -EINVAL;
+	}
+
+	header->video_mode = saved_video_mode;
+
+#ifndef CONFIG_64BIT
+	store_gdt((struct desc_ptr *)&header->pmode_gdt);
+
+	header->pmode_efer_low = nx_enabled;
+	if (header->pmode_efer_low & 1) {
+		/* This is strange, why not save efer, always? */
+		rdmsr(MSR_EFER, header->pmode_efer_low,
+			header->pmode_efer_high);
+	}
+#endif /* !CONFIG_64BIT */
+
+	header->pmode_cr0 = read_cr0();
+	header->pmode_cr4 = read_cr4();
+	header->realmode_flags = acpi_realmode_flags;
+	header->real_magic = 0x12345678;
+
+#ifndef CONFIG_64BIT
+	header->pmode_entry = (u32)&wakeup_pmode_return;
+	header->pmode_cr3 = (u32)(swsusp_pg_dir - __PAGE_OFFSET);
+	saved_magic = 0x12345678;
+#else /* CONFIG_64BIT */
+	header->trampoline_segment = setup_trampoline() >> 4;
+	init_rsp = (unsigned long)temp_stack + 4096;
+	initial_code = (unsigned long)wakeup_long64;
+	saved_magic = 0x123456789abcdef0;
+#endif /* CONFIG_64BIT */
 
 	return 0;
 }
@@ -56,15 +98,20 @@ void acpi_restore_state_mem(void)
  */
 void __init acpi_reserve_bootmem(void)
 {
-	if ((&wakeup_end - &wakeup_start) > PAGE_SIZE*2) {
+	if ((&wakeup_code_end - &wakeup_code_start) > WAKEUP_SIZE) {
 		printk(KERN_ERR
 		       "ACPI: Wakeup code way too big, S3 disabled.\n");
 		return;
 	}
 
-	acpi_wakeup_address = (unsigned long)alloc_bootmem_low(PAGE_SIZE*2);
-	if (!acpi_wakeup_address)
+	acpi_realmode = (unsigned long)alloc_bootmem_low(WAKEUP_SIZE);
+
+	if (!acpi_realmode) {
 		printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n");
+		return;
+	}
+
+	acpi_wakeup_address = acpi_realmode;
 }
 
 
diff --git a/arch/x86/kernel/acpi/sleep.h b/arch/x86/kernel/acpi/sleep.h
new file mode 100644
index 000000000000..adbcbaa6f1df
--- /dev/null
+++ b/arch/x86/kernel/acpi/sleep.h
@@ -0,0 +1,16 @@
+/*
+ *	Variables and functions used by the code in sleep.c
+ */
+
+#include <asm/trampoline.h>
+
+extern char wakeup_code_start, wakeup_code_end;
+
+extern unsigned long saved_video_mode;
+extern long saved_magic;
+
+extern int wakeup_pmode_return;
+extern char swsusp_pg_dir[PAGE_SIZE];
+
+extern unsigned long acpi_copy_wakeup_routine(unsigned long);
+extern void wakeup_long64(void);
diff --git a/arch/x86/kernel/acpi/sleep_32.c b/arch/x86/kernel/acpi/sleep_32.c
deleted file mode 100644
index 63fe5525e026..000000000000
--- a/arch/x86/kernel/acpi/sleep_32.c
+++ /dev/null
@@ -1,40 +0,0 @@
-/*
- * sleep.c - x86-specific ACPI sleep support.
- *
- *  Copyright (C) 2001-2003 Patrick Mochel
- *  Copyright (C) 2001-2003 Pavel Machek <pavel@suse.cz>
- */
-
-#include <linux/acpi.h>
-#include <linux/bootmem.h>
-#include <linux/dmi.h>
-#include <linux/cpumask.h>
-
-#include <asm/smp.h>
-
-/* Ouch, we want to delete this. We already have better version in userspace, in
-   s2ram from suspend.sf.net project */
-static __init int reset_videomode_after_s3(const struct dmi_system_id *d)
-{
-	acpi_realmode_flags |= 2;
-	return 0;
-}
-
-static __initdata struct dmi_system_id acpisleep_dmi_table[] = {
-	{			/* Reset video mode after returning from ACPI S3 sleep */
-	 .callback = reset_videomode_after_s3,
-	 .ident = "Toshiba Satellite 4030cdt",
-	 .matches = {
-		     DMI_MATCH(DMI_PRODUCT_NAME, "S4030CDT/4.3"),
-		     },
-	 },
-	{}
-};
-
-static int __init acpisleep_dmi_init(void)
-{
-	dmi_check_system(acpisleep_dmi_table);
-	return 0;
-}
-
-core_initcall(acpisleep_dmi_init);
diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S
index f53e3277f8e5..a12e6a9fb659 100644
--- a/arch/x86/kernel/acpi/wakeup_32.S
+++ b/arch/x86/kernel/acpi/wakeup_32.S
@@ -3,178 +3,12 @@
 #include <asm/segment.h>
 #include <asm/page.h>
 
-#
-# wakeup_code runs in real mode, and at unknown address (determined at run-time).
-# Therefore it must only use relative jumps/calls. 
-#
-# Do we need to deal with A20? It is okay: ACPI specs says A20 must be enabled
-#
-# If physical address of wakeup_code is 0x12345, BIOS should call us with
-# cs = 0x1234, eip = 0x05
-#
-
-#define BEEP \
-	inb	$97, %al; 	\
-	outb	%al, $0x80; 	\
-	movb	$3, %al; 	\
-	outb	%al, $97; 	\
-	outb	%al, $0x80; 	\
-	movb	$-74, %al; 	\
-	outb	%al, $67; 	\
-	outb	%al, $0x80; 	\
-	movb	$-119, %al; 	\
-	outb	%al, $66; 	\
-	outb	%al, $0x80; 	\
-	movb	$15, %al; 	\
-	outb	%al, $66;
-
-ALIGN
-	.align	4096
-ENTRY(wakeup_start)
-wakeup_code:
-	wakeup_code_start = .
-	.code16
-
-	cli
-	cld
-
-	# setup data segment
-	movw	%cs, %ax
-	movw	%ax, %ds					# Make ds:0 point to wakeup_start
-	movw	%ax, %ss
-
-	testl   $4, realmode_flags - wakeup_code
-	jz      1f
-	BEEP
-1:
-	mov	$(wakeup_stack - wakeup_code), %sp		# Private stack is needed for ASUS board
-
-	pushl	$0						# Kill any dangerous flags
-	popfl
-
-	movl	real_magic - wakeup_code, %eax
-	cmpl	$0x12345678, %eax
-	jne	bogus_real_magic
-
-	testl	$1, realmode_flags - wakeup_code
-	jz	1f
-	lcall   $0xc000,$3
-	movw	%cs, %ax
-	movw	%ax, %ds					# Bios might have played with that
-	movw	%ax, %ss
-1:
-
-	testl	$2, realmode_flags - wakeup_code
-	jz	1f
-	mov	video_mode - wakeup_code, %ax
-	call	mode_set
-1:
-
-	# set up page table
-	movl	$swsusp_pg_dir-__PAGE_OFFSET, %eax
-	movl	%eax, %cr3
-
-	testl	$1, real_efer_save_restore - wakeup_code
-	jz	4f
-	# restore efer setting
-	movl	real_save_efer_edx - wakeup_code, %edx
-	movl	real_save_efer_eax - wakeup_code, %eax
-	mov     $0xc0000080, %ecx
-	wrmsr
-4:
-	# make sure %cr4 is set correctly (features, etc)
-	movl	real_save_cr4 - wakeup_code, %eax
-	movl	%eax, %cr4
-	
-	# need a gdt -- use lgdtl to force 32-bit operands, in case
-	# the GDT is located past 16 megabytes.
-	lgdtl	real_save_gdt - wakeup_code
-
-	movl	real_save_cr0 - wakeup_code, %eax
-	movl	%eax, %cr0
-	jmp 1f
-1:
-	movl	real_magic - wakeup_code, %eax
-	cmpl	$0x12345678, %eax
-	jne	bogus_real_magic
-
-	testl   $8, realmode_flags - wakeup_code
-	jz      1f
-	BEEP
-1:
-	ljmpl	$__KERNEL_CS, $wakeup_pmode_return
-
-real_save_gdt:	.word 0
-		.long 0
-real_save_cr0:	.long 0
-real_save_cr3:	.long 0
-real_save_cr4:	.long 0
-real_magic:	.long 0
-video_mode:	.long 0
-realmode_flags:	.long 0
-real_efer_save_restore:	.long 0
-real_save_efer_edx: 	.long 0
-real_save_efer_eax: 	.long 0
-
-bogus_real_magic:
-	jmp bogus_real_magic
-
-/* This code uses an extended set of video mode numbers. These include:
- * Aliases for standard modes
- *	NORMAL_VGA (-1)
- *	EXTENDED_VGA (-2)
- *	ASK_VGA (-3)
- * Video modes numbered by menu position -- NOT RECOMMENDED because of lack
- * of compatibility when extending the table. These are between 0x00 and 0xff.
- */
-#define VIDEO_FIRST_MENU 0x0000
-
-/* Standard BIOS video modes (BIOS number + 0x0100) */
-#define VIDEO_FIRST_BIOS 0x0100
-
-/* VESA BIOS video modes (VESA number + 0x0200) */
-#define VIDEO_FIRST_VESA 0x0200
-
-/* Video7 special modes (BIOS number + 0x0900) */
-#define VIDEO_FIRST_V7 0x0900
-
-# Setting of user mode (AX=mode ID) => CF=success
-
-# For now, we only handle VESA modes (0x0200..0x03ff).  To handle other
-# modes, we should probably compile in the video code from the boot
-# directory.
-mode_set:
-	movw	%ax, %bx
-	subb	$VIDEO_FIRST_VESA>>8, %bh
-	cmpb	$2, %bh
-	jb	check_vesa
-
-setbad:
-	clc
-	ret
-
-check_vesa:
-	orw	$0x4000, %bx			# Use linear frame buffer
-	movw	$0x4f02, %ax			# VESA BIOS mode set call
-	int	$0x10
-	cmpw	$0x004f, %ax			# AL=4f if implemented
-	jnz	setbad				# AH=0 if OK
-
-	stc
-	ret
+# Copyright 2003, 2008 Pavel Machek <pavel@suse.cz>, distribute under GPLv2
 
 	.code32
 	ALIGN
 
-.org	0x800
-wakeup_stack_begin:	# Stack grows down
-
-.org	0xff0		# Just below end of page
-wakeup_stack:
-ENTRY(wakeup_end)
-	
-.org	0x1000
-
+ENTRY(wakeup_pmode_return)
 wakeup_pmode_return:
 	movw	$__KERNEL_DS, %ax
 	movw	%ax, %ss
@@ -187,7 +21,7 @@ wakeup_pmode_return:
 	lgdt	saved_gdt
 	lidt	saved_idt
 	lldt	saved_ldt
-	ljmp	$(__KERNEL_CS),$1f
+	ljmp	$(__KERNEL_CS), $1f
 1:
 	movl	%cr3, %eax
 	movl	%eax, %cr3
@@ -201,82 +35,41 @@ wakeup_pmode_return:
 	jne	bogus_magic
 
 	# jump to place where we left off
-	movl	saved_eip,%eax
+	movl	saved_eip, %eax
 	jmp	*%eax
 
 bogus_magic:
 	jmp	bogus_magic
 
 
-##
-# acpi_copy_wakeup_routine
-#
-# Copy the above routine to low memory.
-#
-# Parameters:
-# %eax:	place to copy wakeup routine to
-#
-# Returned address is location of code in low memory (past data and stack)
-#
-ENTRY(acpi_copy_wakeup_routine)
 
-	pushl	%ebx
+save_registers:
 	sgdt	saved_gdt
 	sidt	saved_idt
 	sldt	saved_ldt
 	str	saved_tss
 
-	movl	nx_enabled, %edx
-	movl	%edx, real_efer_save_restore - wakeup_start (%eax)
-	testl	$1, real_efer_save_restore - wakeup_start (%eax)
-	jz	2f
-	# save efer setting
-	pushl	%eax
-	movl	%eax, %ebx
-	mov     $0xc0000080, %ecx
-	rdmsr
-	movl	%edx, real_save_efer_edx - wakeup_start (%ebx)
-	movl	%eax, real_save_efer_eax - wakeup_start (%ebx)
-	popl	%eax
-2:
-
-	movl    %cr3, %edx
-	movl    %edx, real_save_cr3 - wakeup_start (%eax)
-	movl    %cr4, %edx
-	movl    %edx, real_save_cr4 - wakeup_start (%eax)
-	movl	%cr0, %edx
-	movl	%edx, real_save_cr0 - wakeup_start (%eax)
-	sgdt    real_save_gdt - wakeup_start (%eax)
-
-	movl	saved_videomode, %edx
-	movl	%edx, video_mode - wakeup_start (%eax)
-	movl	acpi_realmode_flags, %edx
-	movl	%edx, realmode_flags - wakeup_start (%eax)
-	movl	$0x12345678, real_magic - wakeup_start (%eax)
-	movl	$0x12345678, saved_magic
-	popl	%ebx
-	ret
-
-save_registers:
 	leal	4(%esp), %eax
 	movl	%eax, saved_context_esp
-	movl %ebx, saved_context_ebx
-	movl %ebp, saved_context_ebp
-	movl %esi, saved_context_esi
-	movl %edi, saved_context_edi
-	pushfl ; popl saved_context_eflags
-
-	movl $ret_point, saved_eip
+	movl	%ebx, saved_context_ebx
+	movl	%ebp, saved_context_ebp
+	movl	%esi, saved_context_esi
+	movl	%edi, saved_context_edi
+	pushfl
+	popl	saved_context_eflags
+
+	movl	$ret_point, saved_eip
 	ret
 
 
 restore_registers:
-	movl saved_context_ebp, %ebp
-	movl saved_context_ebx, %ebx
-	movl saved_context_esi, %esi
-	movl saved_context_edi, %edi
-	pushl saved_context_eflags ; popfl
-	ret	
+	movl	saved_context_ebp, %ebp
+	movl	saved_context_ebx, %ebx
+	movl	saved_context_esi, %esi
+	movl	saved_context_edi, %edi
+	pushl	saved_context_eflags
+	popfl
+	ret
 
 ENTRY(do_suspend_lowlevel)
 	call	save_processor_state
diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S
index 2e1b9e0d0767..bcc293423a70 100644
--- a/arch/x86/kernel/acpi/wakeup_64.S
+++ b/arch/x86/kernel/acpi/wakeup_64.S
@@ -7,191 +7,18 @@
 #include <asm/asm-offsets.h>
 
 # Copyright 2003 Pavel Machek <pavel@suse.cz>, distribute under GPLv2
-#
-# wakeup_code runs in real mode, and at unknown address (determined at run-time).
-# Therefore it must only use relative jumps/calls. 
-#
-# Do we need to deal with A20? It is okay: ACPI specs says A20 must be enabled
-#
-# If physical address of wakeup_code is 0x12345, BIOS should call us with
-# cs = 0x1234, eip = 0x05
-#
-
-#define BEEP \
-	inb	$97, %al; 	\
-	outb	%al, $0x80; 	\
-	movb	$3, %al; 	\
-	outb	%al, $97; 	\
-	outb	%al, $0x80; 	\
-	movb	$-74, %al; 	\
-	outb	%al, $67; 	\
-	outb	%al, $0x80; 	\
-	movb	$-119, %al; 	\
-	outb	%al, $66; 	\
-	outb	%al, $0x80; 	\
-	movb	$15, %al; 	\
-	outb	%al, $66;
-
-
-ALIGN
-	.align	16
-ENTRY(wakeup_start)
-wakeup_code:
-	wakeup_code_start = .
-	.code16
-
-# Running in *copy* of this code, somewhere in low 1MB.
-
-	cli
-	cld
-	# setup data segment
-	movw	%cs, %ax
-	movw	%ax, %ds		# Make ds:0 point to wakeup_start
-	movw	%ax, %ss
-
-	# Data segment must be set up before we can see whether to beep.
-	testl   $4, realmode_flags - wakeup_code
-	jz      1f
-	BEEP
-1:
-
-					# Private stack is needed for ASUS board
-	mov	$(wakeup_stack - wakeup_code), %sp
-
-	pushl	$0			# Kill any dangerous flags
-	popfl
-
-	movl	real_magic - wakeup_code, %eax
-	cmpl	$0x12345678, %eax
-	jne	bogus_real_magic
-
-	testl	$1, realmode_flags - wakeup_code
-	jz	1f
-	lcall   $0xc000,$3
-	movw	%cs, %ax
-	movw	%ax, %ds		# Bios might have played with that
-	movw	%ax, %ss
-1:
-
-	testl	$2, realmode_flags - wakeup_code
-	jz	1f
-	mov	video_mode - wakeup_code, %ax
-	call	mode_set
-1:
-
-	mov	%ds, %ax			# Find 32bit wakeup_code addr
-	movzx   %ax, %esi			# (Convert %ds:gdt to a liner ptr)
-	shll    $4, %esi
-						# Fix up the vectors
-	addl    %esi, wakeup_32_vector - wakeup_code
-	addl    %esi, wakeup_long64_vector - wakeup_code
-	addl    %esi, gdt_48a + 2 - wakeup_code # Fixup the gdt pointer
-
-	lidtl	%ds:idt_48a - wakeup_code
-	lgdtl	%ds:gdt_48a - wakeup_code	# load gdt with whatever is
-						# appropriate
-
-	movl	$1, %eax			# protected mode (PE) bit
-	lmsw	%ax				# This is it!
-	jmp	1f
-1:
-
-	ljmpl   *(wakeup_32_vector - wakeup_code)
-
-	.balign 4
-wakeup_32_vector:
-	.long   wakeup_32 - wakeup_code
-	.word   __KERNEL32_CS, 0
-
-	.code32
-wakeup_32:
-# Running in this code, but at low address; paging is not yet turned on.
-
-	movl	$__KERNEL_DS, %eax
-	movl	%eax, %ds
-
-	/*
-	 * Prepare for entering 64bits mode
-	 */
-
-	/* Enable PAE */
-	xorl	%eax, %eax
-	btsl	$5, %eax
-	movl	%eax, %cr4
-
-	/* Setup early boot stage 4 level pagetables */
-	leal    (wakeup_level4_pgt - wakeup_code)(%esi), %eax
-	movl	%eax, %cr3
-
-        /* Check if nx is implemented */
-        movl    $0x80000001, %eax
-        cpuid
-        movl    %edx,%edi
-
-	/* Enable Long Mode */
-	xorl    %eax, %eax
-	btsl	$_EFER_LME, %eax
-
-	/* No Execute supported? */
-	btl	$20,%edi
-	jnc     1f
-	btsl	$_EFER_NX, %eax
-				
-	/* Make changes effective */
-1:	movl    $MSR_EFER, %ecx
-	xorl    %edx, %edx
-	wrmsr
-
-	xorl	%eax, %eax
-	btsl	$31, %eax			/* Enable paging and in turn activate Long Mode */
-	btsl	$0, %eax			/* Enable protected mode */
-
-	/* Make changes effective */
-	movl	%eax, %cr0
-
-	/* At this point:
-		CR4.PAE must be 1
-		CS.L must be 0
-		CR3 must point to PML4
-		Next instruction must be a branch
-		This must be on identity-mapped page
-	*/
-	/*
-	 * At this point we're in long mode but in 32bit compatibility mode
-	 * with EFER.LME = 1, CS.L = 0, CS.D = 1 (and in turn
-	 * EFER.LMA = 1). Now we want to jump in 64bit mode, to do that we load
-	 * the new gdt/idt that has __KERNEL_CS with CS.L = 1.
-	 */
-
-	/* Finally jump in 64bit mode */
-        ljmp    *(wakeup_long64_vector - wakeup_code)(%esi)
-
-	.balign 4
-wakeup_long64_vector:
-	.long   wakeup_long64 - wakeup_code
-	.word   __KERNEL_CS, 0
 
 .code64
-
-	/* Hooray, we are in Long 64-bit mode (but still running in
-	 * low memory)
-	 */
-wakeup_long64:
 	/*
-	 * We must switch to a new descriptor in kernel space for the GDT
-	 * because soon the kernel won't have access anymore to the userspace
-	 * addresses where we're currently running on. We have to do that here
-	 * because in 32bit we couldn't load a 64bit linear address.
+	 * Hooray, we are in Long 64-bit mode (but still running in low memory)
 	 */
-	lgdt	cpu_gdt_descr
-
-	movq    saved_magic, %rax
-	movq    $0x123456789abcdef0, %rdx
-	cmpq    %rdx, %rax
-	jne     bogus_64_magic
+ENTRY(wakeup_long64)
+wakeup_long64:
+	movq	saved_magic, %rax
+	movq	$0x123456789abcdef0, %rdx
+	cmpq	%rdx, %rax
+	jne	bogus_64_magic
 
-	nop
-	nop
 	movw	$__KERNEL_DS, %ax
 	movw	%ax, %ss	
 	movw	%ax, %ds
@@ -208,130 +35,8 @@ wakeup_long64:
 	movq	saved_rip, %rax
 	jmp	*%rax
 
-.code32
-
-	.align	64	
-gdta:
-	/* Its good to keep gdt in sync with one in trampoline.S */
-	.word	0, 0, 0, 0			# dummy
-	/* ??? Why I need the accessed bit set in order for this to work? */
-	.quad   0x00cf9b000000ffff              # __KERNEL32_CS
-	.quad   0x00af9b000000ffff              # __KERNEL_CS
-	.quad   0x00cf93000000ffff              # __KERNEL_DS
-
-idt_48a:
-	.word	0				# idt limit = 0
-	.word	0, 0				# idt base = 0L
-
-gdt_48a:
-	.word	0x800				# gdt limit=2048,
-						#  256 GDT entries
-	.long   gdta - wakeup_code              # gdt base (relocated in later)
-	
-real_magic:	.quad 0
-video_mode:	.quad 0
-realmode_flags:	.quad 0
-
-.code16
-bogus_real_magic:
-	jmp bogus_real_magic
-
-.code64
 bogus_64_magic:
-	jmp bogus_64_magic
-
-/* This code uses an extended set of video mode numbers. These include:
- * Aliases for standard modes
- *	NORMAL_VGA (-1)
- *	EXTENDED_VGA (-2)
- *	ASK_VGA (-3)
- * Video modes numbered by menu position -- NOT RECOMMENDED because of lack
- * of compatibility when extending the table. These are between 0x00 and 0xff.
- */
-#define VIDEO_FIRST_MENU 0x0000
-
-/* Standard BIOS video modes (BIOS number + 0x0100) */
-#define VIDEO_FIRST_BIOS 0x0100
-
-/* VESA BIOS video modes (VESA number + 0x0200) */
-#define VIDEO_FIRST_VESA 0x0200
-
-/* Video7 special modes (BIOS number + 0x0900) */
-#define VIDEO_FIRST_V7 0x0900
-
-# Setting of user mode (AX=mode ID) => CF=success
-
-# For now, we only handle VESA modes (0x0200..0x03ff).  To handle other
-# modes, we should probably compile in the video code from the boot
-# directory.
-.code16
-mode_set:
-	movw	%ax, %bx
-	subb	$VIDEO_FIRST_VESA>>8, %bh
-	cmpb	$2, %bh
-	jb	check_vesa
-
-setbad:
-	clc
-	ret
-
-check_vesa:
-	orw	$0x4000, %bx			# Use linear frame buffer
-	movw	$0x4f02, %ax			# VESA BIOS mode set call
-	int	$0x10
-	cmpw	$0x004f, %ax			# AL=4f if implemented
-	jnz	setbad				# AH=0 if OK
-
-	stc
-	ret
-
-wakeup_stack_begin:	# Stack grows down
-
-.org	0xff0
-wakeup_stack:		# Just below end of page
-
-.org   0x1000
-ENTRY(wakeup_level4_pgt)
-	.quad   level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
-	.fill   510,8,0
-	/* (2^48-(2*1024*1024*1024))/(2^39) = 511 */
-	.quad   level3_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE
-
-ENTRY(wakeup_end)
-	
-##
-# acpi_copy_wakeup_routine
-#
-# Copy the above routine to low memory.
-#
-# Parameters:
-# %rdi:	place to copy wakeup routine to
-#
-# Returned address is location of code in low memory (past data and stack)
-#
-	.code64
-ENTRY(acpi_copy_wakeup_routine)
-	pushq	%rax
-	pushq	%rdx
-
-	movl	saved_video_mode, %edx
-	movl	%edx, video_mode - wakeup_start (,%rdi)
-	movl	acpi_realmode_flags, %edx
-	movl	%edx, realmode_flags - wakeup_start (,%rdi)
-	movq	$0x12345678, real_magic - wakeup_start (,%rdi)
-	movq	$0x123456789abcdef0, %rdx
-	movq	%rdx, saved_magic
-
-	movq    saved_magic, %rax
-	movq    $0x123456789abcdef0, %rdx
-	cmpq    %rdx, %rax
-	jne     bogus_64_magic
-
-	# restore the regs we used
-	popq	%rdx
-	popq	%rax
-ENTRY(do_suspend_lowlevel_s4bios)
-	ret
+	jmp	bogus_64_magic
 
 	.align 2
 	.p2align 4,,15
@@ -414,7 +119,7 @@ do_suspend_lowlevel:
 	jmp	restore_processor_state
 .LFE5:
 .Lfe5:
-	.size	do_suspend_lowlevel,.Lfe5-do_suspend_lowlevel
+	.size	do_suspend_lowlevel, .Lfe5-do_suspend_lowlevel
 	
 .data
 ALIGN
diff --git a/arch/x86/kernel/acpi/wakeup_rm.S b/arch/x86/kernel/acpi/wakeup_rm.S
new file mode 100644
index 000000000000..6ff3b5730575
--- /dev/null
+++ b/arch/x86/kernel/acpi/wakeup_rm.S
@@ -0,0 +1,10 @@
+/*
+ * Wrapper script for the realmode binary as a transport object
+ * before copying to low memory.
+ */
+	.section ".rodata","a"
+	.globl	wakeup_code_start, wakeup_code_end
+wakeup_code_start:
+	.incbin	"arch/x86/kernel/acpi/realmode/wakeup.bin"
+wakeup_code_end:
+	.size	wakeup_code_start, .-wakeup_code_start
diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c
index a720f3d5ed9d..7f6c0c85c8f6 100644
--- a/arch/x86/kernel/e820_64.c
+++ b/arch/x86/kernel/e820_64.c
@@ -27,6 +27,7 @@
 #include <asm/setup.h>
 #include <asm/sections.h>
 #include <asm/kdebug.h>
+#include <asm/trampoline.h>
 
 struct e820map e820;
 
@@ -58,8 +59,8 @@ struct early_res {
 };
 static struct early_res early_res[MAX_EARLY_RES] __initdata = {
 	{ 0, PAGE_SIZE, "BIOS data page" },			/* BIOS data page */
-#ifdef CONFIG_SMP
-	{ SMP_TRAMPOLINE_BASE, SMP_TRAMPOLINE_BASE + 2*PAGE_SIZE, "SMP_TRAMPOLINE" },
+#ifdef CONFIG_X86_TRAMPOLINE
+	{ TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" },
 #endif
 	{}
 };
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index c1d7a877d814..10a1955bb1d1 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -132,10 +132,6 @@ ident_complete:
 	addq	%rbp, trampoline_level4_pgt + 0(%rip)
 	addq	%rbp, trampoline_level4_pgt + (511*8)(%rip)
 #endif
-#ifdef CONFIG_ACPI_SLEEP
-	addq	%rbp, wakeup_level4_pgt + 0(%rip)
-	addq	%rbp, wakeup_level4_pgt + (511*8)(%rip)
-#endif
 
 	/* Due to ENTRY(), sometimes the empty space gets filled with
 	 * zeros. Better take a jmp than relying on empty space being
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 4b198d9d0de3..5b0bffb7fcc9 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -192,7 +192,7 @@ EXPORT_SYMBOL(ist_info);
 extern void early_cpu_init(void);
 extern int root_mountflags;
 
-unsigned long saved_videomode;
+unsigned long saved_video_mode;
 
 #define RAMDISK_IMAGE_START_MASK	0x07FF
 #define RAMDISK_PROMPT_FLAG		0x8000
@@ -763,7 +763,7 @@ void __init setup_arch(char **cmdline_p)
 	edid_info = boot_params.edid_info;
 	apm_info.bios = boot_params.apm_bios_info;
 	ist_info = boot_params.ist_info;
-	saved_videomode = boot_params.hdr.vid_mode;
+	saved_video_mode = boot_params.hdr.vid_mode;
 	if( boot_params.sys_desc_table.length != 0 ) {
 		set_mca_bus(boot_params.sys_desc_table.table[3] & 0x2);
 		machine_id = boot_params.sys_desc_table.table[0];
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index b80300710c08..674ef3510cdf 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -65,6 +65,7 @@
 #include <asm/mce.h>
 #include <asm/ds.h>
 #include <asm/topology.h>
+#include <asm/trampoline.h>
 
 #include <mach_apic.h>
 #ifdef CONFIG_PARAVIRT
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index ca3929b16049..424600e671bd 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -53,6 +53,7 @@
 #include <asm/nmi.h>
 #include <asm/irq.h>
 #include <asm/smp.h>
+#include <asm/trampoline.h>
 #include <asm/cpu.h>
 #include <asm/numa.h>
 #include <asm/pgtable.h>
@@ -140,7 +141,7 @@ static atomic_t init_deasserted;
 static int boot_cpu_logical_apicid;
 
 /* ready for x86_64, no harm for x86, since it will overwrite after alloc */
-unsigned char *trampoline_base = __va(SMP_TRAMPOLINE_BASE);
+unsigned char *trampoline_base = __va(TRAMPOLINE_BASE);
 
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
@@ -554,8 +555,7 @@ cpumask_t cpu_coregroup_map(int cpu)
  * bootstrap into the page concerned. The caller
  * has made sure it's suitably aligned.
  */
-
-unsigned long __cpuinit setup_trampoline(void)
+unsigned long setup_trampoline(void)
 {
 	memcpy(trampoline_base, trampoline_data,
 	       trampoline_end - trampoline_data);
diff --git a/arch/x86/kernel/trampoline_64.S b/arch/x86/kernel/trampoline_64.S
index 2a07e67d6697..894293c598db 100644
--- a/arch/x86/kernel/trampoline_64.S
+++ b/arch/x86/kernel/trampoline_64.S
@@ -30,12 +30,7 @@
 #include <asm/msr.h>
 #include <asm/segment.h>
 
-/* We can free up trampoline after bootup if cpu hotplug is not supported. */
-#ifndef CONFIG_HOTPLUG_CPU
-.section .cpuinit.data, "aw", @progbits
-#else
 .section .rodata, "a", @progbits
-#endif
 
 .code16
 
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 4397235c2e30..be7235bf105d 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -27,6 +27,7 @@
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
 #include <asm/arch_hooks.h>
+#include <asm/trampoline.h>
 
 /* TLB state -- visible externally, indexed physically */
 DEFINE_PER_CPU_SHARED_ALIGNED(struct tlb_state, cpu_tlbstate) = { &init_mm, 0 };
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 3496e1c299b2..62ebdec394b9 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -45,22 +45,12 @@ DECLARE_PER_CPU(u16, cpu_llc_id);
 DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
 DECLARE_PER_CPU(u16, x86_bios_cpu_apicid);
 
-/*
- * Trampoline 80x86 program as an array.
- */
-extern const unsigned char trampoline_data [];
-extern const unsigned char trampoline_end  [];
-extern unsigned char *trampoline_base;
-
 /* Static state in head.S used to set up a CPU */
 extern struct {
 	void *sp;
 	unsigned short ss;
 } stack_start;
 
-extern unsigned long init_rsp;
-extern unsigned long initial_code;
-
 struct smp_ops {
 	void (*smp_prepare_boot_cpu)(void);
 	void (*smp_prepare_cpus)(unsigned max_cpus);
@@ -130,9 +120,6 @@ extern void __cpu_die(unsigned int cpu);
 
 extern void prefill_possible_map(void);
 
-#define SMP_TRAMPOLINE_BASE 0x6000
-extern unsigned long setup_trampoline(void);
-
 void smp_store_cpu_info(int id);
 #define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
 
diff --git a/include/asm-x86/trampoline.h b/include/asm-x86/trampoline.h
new file mode 100644
index 000000000000..b156b08d0131
--- /dev/null
+++ b/include/asm-x86/trampoline.h
@@ -0,0 +1,21 @@
+#ifndef __TRAMPOLINE_HEADER
+#define __TRAMPOLINE_HEADER
+
+#ifndef __ASSEMBLY__
+
+/*
+ * Trampoline 80x86 program as an array.
+ */
+extern const unsigned char trampoline_data [];
+extern const unsigned char trampoline_end  [];
+extern unsigned char *trampoline_base;
+
+extern unsigned long init_rsp;
+extern unsigned long initial_code;
+
+#define TRAMPOLINE_BASE 0x6000
+extern unsigned long setup_trampoline(void);
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __TRAMPOLINE_HEADER */
-- 
cgit v1.2.3


From c33fa9f5609e918824446ef9a75319d4a802f1f4 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 17 Apr 2008 20:05:36 +0200
Subject: uaccess: add probe_kernel_write()

add probe_kernel_read() and probe_kernel_write().

Uninlined and restricted to kernel range memory only, as suggested
by Linus.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/uaccess.h | 22 ++++++++++++++++++++++
 mm/Makefile             |  2 +-
 mm/maccess.c            | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 72 insertions(+), 1 deletion(-)
 create mode 100644 mm/maccess.c

(limited to 'include')

diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 975c963e5789..fec6decfb983 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -84,4 +84,26 @@ static inline unsigned long __copy_from_user_nocache(void *to,
 		ret;					\
 	})
 
+/*
+ * probe_kernel_read(): safely attempt to read from a location
+ * @dst: pointer to the buffer that shall take the data
+ * @src: address to read from
+ * @size: size of the data chunk
+ *
+ * Safely read from address @src to the buffer at @dst.  If a kernel fault
+ * happens, handle that and return -EFAULT.
+ */
+extern long probe_kernel_read(void *dst, void *src, size_t size);
+
+/*
+ * probe_kernel_write(): safely attempt to write to a location
+ * @dst: address to write to
+ * @src: pointer to the data that shall be written
+ * @size: size of the data chunk
+ *
+ * Safely write to address @dst from the buffer at @src.  If a kernel fault
+ * happens, handle that and return -EFAULT.
+ */
+extern long probe_kernel_write(void *dst, void *src, size_t size);
+
 #endif		/* __LINUX_UACCESS_H__ */
diff --git a/mm/Makefile b/mm/Makefile
index a5b0dd93427a..18c143b3c46c 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -8,7 +8,7 @@ mmu-$(CONFIG_MMU)	:= fremap.o highmem.o madvise.o memory.o mincore.o \
 			   vmalloc.o
 
 obj-y			:= bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
-			   page_alloc.o page-writeback.o pdflush.o \
+			   maccess.o page_alloc.o page-writeback.o pdflush.o \
 			   readahead.o swap.o truncate.o vmscan.o \
 			   prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \
 			   page_isolation.o $(mmu-y)
diff --git a/mm/maccess.c b/mm/maccess.c
new file mode 100644
index 000000000000..24f81b971403
--- /dev/null
+++ b/mm/maccess.c
@@ -0,0 +1,49 @@
+/*
+ * Access kernel memory without faulting.
+ */
+#include <linux/uaccess.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+
+/**
+ * probe_kernel_read(): safely attempt to read from a location
+ * @dst: pointer to the buffer that shall take the data
+ * @src: address to read from
+ * @size: size of the data chunk
+ *
+ * Safely read from address @src to the buffer at @dst.  If a kernel fault
+ * happens, handle that and return -EFAULT.
+ */
+long probe_kernel_read(void *dst, void *src, size_t size)
+{
+	long ret;
+
+	pagefault_disable();
+	ret = __copy_from_user_inatomic(dst,
+			(__force const void __user *)src, size);
+	pagefault_enable();
+
+	return ret ? -EFAULT : 0;
+}
+EXPORT_SYMBOL_GPL(probe_kernel_read);
+
+/**
+ * probe_kernel_write(): safely attempt to write to a location
+ * @dst: address to write to
+ * @src: pointer to the data that shall be written
+ * @size: size of the data chunk
+ *
+ * Safely write to address @dst from the buffer at @src.  If a kernel fault
+ * happens, handle that and return -EFAULT.
+ */
+long probe_kernel_write(void *dst, void *src, size_t size)
+{
+	long ret;
+
+	pagefault_disable();
+	ret = __copy_to_user_inatomic((__force void __user *)dst, src, size);
+	pagefault_enable();
+
+	return ret ? -EFAULT : 0;
+}
+EXPORT_SYMBOL_GPL(probe_kernel_write);
-- 
cgit v1.2.3


From dc7d552705215ac50a0617fcf51bb9c736255b8e Mon Sep 17 00:00:00 2001
From: Jason Wessel <jason.wessel@windriver.com>
Date: Thu, 17 Apr 2008 20:05:37 +0200
Subject: kgdb: core

kgdb core code. Handles the protocol and the arch details.

[ mingo@elte.hu: heavily modified, simplified and cleaned up. ]
[ xemul@openvz.org: use find_task_by_pid_ns ]

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jan Kiszka <jan.kiszka@web.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/kgdb.h |  271 ++++++++
 kernel/Makefile      |    1 +
 kernel/kgdb.c        | 1693 ++++++++++++++++++++++++++++++++++++++++++++++++++
 lib/Kconfig.debug    |    2 +
 lib/Kconfig.kgdb     |   27 +
 5 files changed, 1994 insertions(+)
 create mode 100644 include/linux/kgdb.h
 create mode 100644 kernel/kgdb.c
 create mode 100644 lib/Kconfig.kgdb

(limited to 'include')

diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
new file mode 100644
index 000000000000..b0985b79b638
--- /dev/null
+++ b/include/linux/kgdb.h
@@ -0,0 +1,271 @@
+/*
+ * This provides the callbacks and functions that KGDB needs to share between
+ * the core, I/O and arch-specific portions.
+ *
+ * Author: Amit Kale <amitkale@linsyssoft.com> and
+ *         Tom Rini <trini@kernel.crashing.org>
+ *
+ * 2001-2004 (c) Amit S. Kale and 2003-2005 (c) MontaVista Software, Inc.
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+#ifndef _KGDB_H_
+#define _KGDB_H_
+
+#include <linux/serial_8250.h>
+#include <linux/linkage.h>
+#include <linux/init.h>
+
+#include <asm/atomic.h>
+#include <asm/kgdb.h>
+
+struct pt_regs;
+
+/*
+ *	kgdb_skipexception - Bail out of KGDB when we've been triggered.
+ *	@exception: Exception vector number
+ *	@regs: Current &struct pt_regs.
+ *
+ *	On some architectures we need to skip a breakpoint exception when
+ *	it occurs after a breakpoint has been removed.
+ */
+extern int kgdb_skipexception(int exception, struct pt_regs *regs);
+
+/*
+ *	kgdb_post_primary_code - Save error vector/code numbers.
+ *	@regs: Original pt_regs.
+ *	@e_vector: Original error vector.
+ *	@err_code: Original error code.
+ *
+ *	This is needed on architectures which support SMP and KGDB.
+ *	This function is called after all the secondary cpus have been put
+ *	to a know spin state and the primary CPU has control over KGDB.
+ */
+extern void kgdb_post_primary_code(struct pt_regs *regs, int e_vector,
+				  int err_code);
+
+/*
+ *	kgdb_disable_hw_debug - Disable hardware debugging while we in kgdb.
+ *	@regs: Current &struct pt_regs.
+ *
+ *	This function will be called if the particular architecture must
+ *	disable hardware debugging while it is processing gdb packets or
+ *	handling exception.
+ */
+extern void kgdb_disable_hw_debug(struct pt_regs *regs);
+
+struct tasklet_struct;
+struct task_struct;
+struct uart_port;
+
+/* To enter the debugger explicitly. */
+void kgdb_breakpoint(void);
+
+extern int kgdb_connected;
+
+extern atomic_t			kgdb_setting_breakpoint;
+extern atomic_t			kgdb_cpu_doing_single_step;
+
+extern struct task_struct	*kgdb_usethread;
+extern struct task_struct	*kgdb_contthread;
+
+enum kgdb_bptype {
+	BP_BREAKPOINT = 0,
+	BP_HARDWARE_BREAKPOINT,
+	BP_WRITE_WATCHPOINT,
+	BP_READ_WATCHPOINT,
+	BP_ACCESS_WATCHPOINT
+};
+
+enum kgdb_bpstate {
+	BP_UNDEFINED = 0,
+	BP_REMOVED,
+	BP_SET,
+	BP_ACTIVE
+};
+
+struct kgdb_bkpt {
+	unsigned long		bpt_addr;
+	unsigned char		saved_instr[BREAK_INSTR_SIZE];
+	enum kgdb_bptype	type;
+	enum kgdb_bpstate	state;
+};
+
+#ifndef KGDB_MAX_BREAKPOINTS
+# define KGDB_MAX_BREAKPOINTS	1000
+#endif
+
+#define KGDB_HW_BREAKPOINT	1
+
+/*
+ * Functions each KGDB-supporting architecture must provide:
+ */
+
+/*
+ *	kgdb_arch_init - Perform any architecture specific initalization.
+ *
+ *	This function will handle the initalization of any architecture
+ *	specific callbacks.
+ */
+extern int kgdb_arch_init(void);
+
+/*
+ *	kgdb_arch_exit - Perform any architecture specific uninitalization.
+ *
+ *	This function will handle the uninitalization of any architecture
+ *	specific callbacks, for dynamic registration and unregistration.
+ */
+extern void kgdb_arch_exit(void);
+
+/*
+ *	pt_regs_to_gdb_regs - Convert ptrace regs to GDB regs
+ *	@gdb_regs: A pointer to hold the registers in the order GDB wants.
+ *	@regs: The &struct pt_regs of the current process.
+ *
+ *	Convert the pt_regs in @regs into the format for registers that
+ *	GDB expects, stored in @gdb_regs.
+ */
+extern void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs);
+
+/*
+ *	sleeping_thread_to_gdb_regs - Convert ptrace regs to GDB regs
+ *	@gdb_regs: A pointer to hold the registers in the order GDB wants.
+ *	@p: The &struct task_struct of the desired process.
+ *
+ *	Convert the register values of the sleeping process in @p to
+ *	the format that GDB expects.
+ *	This function is called when kgdb does not have access to the
+ *	&struct pt_regs and therefore it should fill the gdb registers
+ *	@gdb_regs with what has	been saved in &struct thread_struct
+ *	thread field during switch_to.
+ */
+extern void
+sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p);
+
+/*
+ *	gdb_regs_to_pt_regs - Convert GDB regs to ptrace regs.
+ *	@gdb_regs: A pointer to hold the registers we've received from GDB.
+ *	@regs: A pointer to a &struct pt_regs to hold these values in.
+ *
+ *	Convert the GDB regs in @gdb_regs into the pt_regs, and store them
+ *	in @regs.
+ */
+extern void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs);
+
+/*
+ *	kgdb_arch_handle_exception - Handle architecture specific GDB packets.
+ *	@vector: The error vector of the exception that happened.
+ *	@signo: The signal number of the exception that happened.
+ *	@err_code: The error code of the exception that happened.
+ *	@remcom_in_buffer: The buffer of the packet we have read.
+ *	@remcom_out_buffer: The buffer of %BUFMAX bytes to write a packet into.
+ *	@regs: The &struct pt_regs of the current process.
+ *
+ *	This function MUST handle the 'c' and 's' command packets,
+ *	as well packets to set / remove a hardware breakpoint, if used.
+ *	If there are additional packets which the hardware needs to handle,
+ *	they are handled here.  The code should return -1 if it wants to
+ *	process more packets, and a %0 or %1 if it wants to exit from the
+ *	kgdb callback.
+ */
+extern int
+kgdb_arch_handle_exception(int vector, int signo, int err_code,
+			   char *remcom_in_buffer,
+			   char *remcom_out_buffer,
+			   struct pt_regs *regs);
+
+/*
+ *	kgdb_roundup_cpus - Get other CPUs into a holding pattern
+ *	@flags: Current IRQ state
+ *
+ *	On SMP systems, we need to get the attention of the other CPUs
+ *	and get them be in a known state.  This should do what is needed
+ *	to get the other CPUs to call kgdb_wait(). Note that on some arches,
+ *	the NMI approach is not used for rounding up all the CPUs. For example,
+ *	in case of MIPS, smp_call_function() is used to roundup CPUs. In
+ *	this case, we have to make sure that interrupts are enabled before
+ *	calling smp_call_function(). The argument to this function is
+ *	the flags that will be used when restoring the interrupts. There is
+ *	local_irq_save() call before kgdb_roundup_cpus().
+ *
+ *	On non-SMP systems, this is not called.
+ */
+extern void kgdb_roundup_cpus(unsigned long flags);
+
+/* Optional functions. */
+extern int kgdb_validate_break_address(unsigned long addr);
+extern int kgdb_arch_set_breakpoint(unsigned long addr, char *saved_instr);
+extern int kgdb_arch_remove_breakpoint(unsigned long addr, char *bundle);
+
+/*
+ * struct kgdb_arch - Describe architecture specific values.
+ * @gdb_bpt_instr: The instruction to trigger a breakpoint.
+ * @flags: Flags for the breakpoint, currently just %KGDB_HW_BREAKPOINT.
+ * @set_breakpoint: Allow an architecture to specify how to set a software
+ * breakpoint.
+ * @remove_breakpoint: Allow an architecture to specify how to remove a
+ * software breakpoint.
+ * @set_hw_breakpoint: Allow an architecture to specify how to set a hardware
+ * breakpoint.
+ * @remove_hw_breakpoint: Allow an architecture to specify how to remove a
+ * hardware breakpoint.
+ * @remove_all_hw_break: Allow an architecture to specify how to remove all
+ * hardware breakpoints.
+ * @correct_hw_break: Allow an architecture to specify how to correct the
+ * hardware debug registers.
+ */
+struct kgdb_arch {
+	unsigned char		gdb_bpt_instr[BREAK_INSTR_SIZE];
+	unsigned long		flags;
+
+	int	(*set_breakpoint)(unsigned long, char *);
+	int	(*remove_breakpoint)(unsigned long, char *);
+	int	(*set_hw_breakpoint)(unsigned long, int, enum kgdb_bptype);
+	int	(*remove_hw_breakpoint)(unsigned long, int, enum kgdb_bptype);
+	void	(*remove_all_hw_break)(void);
+	void	(*correct_hw_break)(void);
+};
+
+/*
+ * struct kgdb_io - Describe the interface for an I/O driver to talk with KGDB.
+ * @name: Name of the I/O driver.
+ * @read_char: Pointer to a function that will return one char.
+ * @write_char: Pointer to a function that will write one char.
+ * @flush: Pointer to a function that will flush any pending writes.
+ * @init: Pointer to a function that will initialize the device.
+ * @pre_exception: Pointer to a function that will do any prep work for
+ * the I/O driver.
+ * @post_exception: Pointer to a function that will do any cleanup work
+ * for the I/O driver.
+ */
+struct kgdb_io {
+	const char		*name;
+	int			(*read_char) (void);
+	void			(*write_char) (u8);
+	void			(*flush) (void);
+	int			(*init) (void);
+	void			(*pre_exception) (void);
+	void			(*post_exception) (void);
+};
+
+extern struct kgdb_arch		arch_kgdb_ops;
+
+extern int kgdb_register_io_module(struct kgdb_io *local_kgdb_io_ops);
+extern void kgdb_unregister_io_module(struct kgdb_io *local_kgdb_io_ops);
+
+extern int kgdb_hex2long(char **ptr, long *long_val);
+extern int kgdb_mem2hex(char *mem, char *buf, int count);
+extern int kgdb_hex2mem(char *buf, char *mem, int count);
+
+extern int kgdb_isremovedbreak(unsigned long addr);
+
+extern int
+kgdb_handle_exception(int ex_vector, int signo, int err_code,
+		      struct pt_regs *regs);
+extern int kgdb_nmicallback(int cpu, void *regs);
+
+extern int			kgdb_single_step;
+extern atomic_t			kgdb_active;
+
+#endif /* _KGDB_H_ */
diff --git a/kernel/Makefile b/kernel/Makefile
index 6c584c55a6e9..05c8003718ee 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -53,6 +53,7 @@ obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
 obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
 obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
 obj-$(CONFIG_KPROBES) += kprobes.o
+obj-$(CONFIG_KGDB) += kgdb.o
 obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o
 obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
 obj-$(CONFIG_SECCOMP) += seccomp.o
diff --git a/kernel/kgdb.c b/kernel/kgdb.c
new file mode 100644
index 000000000000..017ee782bc08
--- /dev/null
+++ b/kernel/kgdb.c
@@ -0,0 +1,1693 @@
+/*
+ * KGDB stub.
+ *
+ * Maintainer: Jason Wessel <jason.wessel@windriver.com>
+ *
+ * Copyright (C) 2000-2001 VERITAS Software Corporation.
+ * Copyright (C) 2002-2004 Timesys Corporation
+ * Copyright (C) 2003-2004 Amit S. Kale <amitkale@linsyssoft.com>
+ * Copyright (C) 2004 Pavel Machek <pavel@suse.cz>
+ * Copyright (C) 2004-2006 Tom Rini <trini@kernel.crashing.org>
+ * Copyright (C) 2004-2006 LinSysSoft Technologies Pvt. Ltd.
+ * Copyright (C) 2005-2008 Wind River Systems, Inc.
+ * Copyright (C) 2007 MontaVista Software, Inc.
+ * Copyright (C) 2008 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
+ *
+ * Contributors at various stages not listed above:
+ *  Jason Wessel ( jason.wessel@windriver.com )
+ *  George Anzinger <george@mvista.com>
+ *  Anurekh Saxena (anurekh.saxena@timesys.com)
+ *  Lake Stevens Instrument Division (Glenn Engel)
+ *  Jim Kingdon, Cygnus Support.
+ *
+ * Original KGDB stub: David Grothe <dave@gcom.com>,
+ * Tigran Aivazian <tigran@sco.com>
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+#include <linux/pid_namespace.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/console.h>
+#include <linux/threads.h>
+#include <linux/uaccess.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/ptrace.h>
+#include <linux/reboot.h>
+#include <linux/string.h>
+#include <linux/delay.h>
+#include <linux/sched.h>
+#include <linux/sysrq.h>
+#include <linux/init.h>
+#include <linux/kgdb.h>
+#include <linux/pid.h>
+#include <linux/smp.h>
+#include <linux/mm.h>
+
+#include <asm/cacheflush.h>
+#include <asm/byteorder.h>
+#include <asm/atomic.h>
+#include <asm/system.h>
+
+static int kgdb_break_asap;
+
+struct kgdb_state {
+	int			ex_vector;
+	int			signo;
+	int			err_code;
+	int			cpu;
+	int			pass_exception;
+	long			threadid;
+	long			kgdb_usethreadid;
+	struct pt_regs		*linux_regs;
+};
+
+static struct debuggerinfo_struct {
+	void			*debuggerinfo;
+	struct task_struct	*task;
+} kgdb_info[NR_CPUS];
+
+/**
+ * kgdb_connected - Is a host GDB connected to us?
+ */
+int				kgdb_connected;
+EXPORT_SYMBOL_GPL(kgdb_connected);
+
+/* All the KGDB handlers are installed */
+static int			kgdb_io_module_registered;
+
+/* Guard for recursive entry */
+static int			exception_level;
+
+static struct kgdb_io		*kgdb_io_ops;
+static DEFINE_SPINLOCK(kgdb_registration_lock);
+
+/* kgdb console driver is loaded */
+static int kgdb_con_registered;
+/* determine if kgdb console output should be used */
+static int kgdb_use_con;
+
+static int __init opt_kgdb_con(char *str)
+{
+	kgdb_use_con = 1;
+	return 0;
+}
+
+early_param("kgdbcon", opt_kgdb_con);
+
+module_param(kgdb_use_con, int, 0644);
+
+/*
+ * Holds information about breakpoints in a kernel. These breakpoints are
+ * added and removed by gdb.
+ */
+static struct kgdb_bkpt		kgdb_break[KGDB_MAX_BREAKPOINTS] = {
+	[0 ... KGDB_MAX_BREAKPOINTS-1] = { .state = BP_UNDEFINED }
+};
+
+/*
+ * The CPU# of the active CPU, or -1 if none:
+ */
+atomic_t			kgdb_active = ATOMIC_INIT(-1);
+
+/*
+ * We use NR_CPUs not PERCPU, in case kgdb is used to debug early
+ * bootup code (which might not have percpu set up yet):
+ */
+static atomic_t			passive_cpu_wait[NR_CPUS];
+static atomic_t			cpu_in_kgdb[NR_CPUS];
+atomic_t			kgdb_setting_breakpoint;
+
+struct task_struct		*kgdb_usethread;
+struct task_struct		*kgdb_contthread;
+
+int				kgdb_single_step;
+
+/* Our I/O buffers. */
+static char			remcom_in_buffer[BUFMAX];
+static char			remcom_out_buffer[BUFMAX];
+
+/* Storage for the registers, in GDB format. */
+static unsigned long		gdb_regs[(NUMREGBYTES +
+					sizeof(unsigned long) - 1) /
+					sizeof(unsigned long)];
+
+/* to keep track of the CPU which is doing the single stepping*/
+atomic_t			kgdb_cpu_doing_single_step = ATOMIC_INIT(-1);
+
+/*
+ * If you are debugging a problem where roundup (the collection of
+ * all other CPUs) is a problem [this should be extremely rare],
+ * then use the nokgdbroundup option to avoid roundup. In that case
+ * the other CPUs might interfere with your debugging context, so
+ * use this with care:
+ */
+int				kgdb_do_roundup = 1;
+
+static int __init opt_nokgdbroundup(char *str)
+{
+	kgdb_do_roundup = 0;
+
+	return 0;
+}
+
+early_param("nokgdbroundup", opt_nokgdbroundup);
+
+/*
+ * Finally, some KGDB code :-)
+ */
+
+/*
+ * Weak aliases for breakpoint management,
+ * can be overriden by architectures when needed:
+ */
+int __weak kgdb_validate_break_address(unsigned long addr)
+{
+	char tmp_variable[BREAK_INSTR_SIZE];
+
+	return probe_kernel_read(tmp_variable, (char *)addr, BREAK_INSTR_SIZE);
+}
+
+int __weak kgdb_arch_set_breakpoint(unsigned long addr, char *saved_instr)
+{
+	int err;
+
+	err = probe_kernel_read(saved_instr, (char *)addr, BREAK_INSTR_SIZE);
+	if (err)
+		return err;
+
+	return probe_kernel_write((char *)addr, arch_kgdb_ops.gdb_bpt_instr,
+				  BREAK_INSTR_SIZE);
+}
+
+int __weak kgdb_arch_remove_breakpoint(unsigned long addr, char *bundle)
+{
+	return probe_kernel_write((char *)addr,
+				  (char *)bundle, BREAK_INSTR_SIZE);
+}
+
+unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs)
+{
+	return instruction_pointer(regs);
+}
+
+int __weak kgdb_arch_init(void)
+{
+	return 0;
+}
+
+/**
+ *	kgdb_disable_hw_debug - Disable hardware debugging while we in kgdb.
+ *	@regs: Current &struct pt_regs.
+ *
+ *	This function will be called if the particular architecture must
+ *	disable hardware debugging while it is processing gdb packets or
+ *	handling exception.
+ */
+void __weak kgdb_disable_hw_debug(struct pt_regs *regs)
+{
+}
+
+/*
+ * GDB remote protocol parser:
+ */
+
+static const char	hexchars[] = "0123456789abcdef";
+
+static int hex(char ch)
+{
+	if ((ch >= 'a') && (ch <= 'f'))
+		return ch - 'a' + 10;
+	if ((ch >= '0') && (ch <= '9'))
+		return ch - '0';
+	if ((ch >= 'A') && (ch <= 'F'))
+		return ch - 'A' + 10;
+	return -1;
+}
+
+/* scan for the sequence $<data>#<checksum> */
+static void get_packet(char *buffer)
+{
+	unsigned char checksum;
+	unsigned char xmitcsum;
+	int count;
+	char ch;
+
+	do {
+		/*
+		 * Spin and wait around for the start character, ignore all
+		 * other characters:
+		 */
+		while ((ch = (kgdb_io_ops->read_char())) != '$')
+			/* nothing */;
+
+		kgdb_connected = 1;
+		checksum = 0;
+		xmitcsum = -1;
+
+		count = 0;
+
+		/*
+		 * now, read until a # or end of buffer is found:
+		 */
+		while (count < (BUFMAX - 1)) {
+			ch = kgdb_io_ops->read_char();
+			if (ch == '#')
+				break;
+			checksum = checksum + ch;
+			buffer[count] = ch;
+			count = count + 1;
+		}
+		buffer[count] = 0;
+
+		if (ch == '#') {
+			xmitcsum = hex(kgdb_io_ops->read_char()) << 4;
+			xmitcsum += hex(kgdb_io_ops->read_char());
+
+			if (checksum != xmitcsum)
+				/* failed checksum */
+				kgdb_io_ops->write_char('-');
+			else
+				/* successful transfer */
+				kgdb_io_ops->write_char('+');
+			if (kgdb_io_ops->flush)
+				kgdb_io_ops->flush();
+		}
+	} while (checksum != xmitcsum);
+}
+
+/*
+ * Send the packet in buffer.
+ * Check for gdb connection if asked for.
+ */
+static void put_packet(char *buffer)
+{
+	unsigned char checksum;
+	int count;
+	char ch;
+
+	/*
+	 * $<packet info>#<checksum>.
+	 */
+	while (1) {
+		kgdb_io_ops->write_char('$');
+		checksum = 0;
+		count = 0;
+
+		while ((ch = buffer[count])) {
+			kgdb_io_ops->write_char(ch);
+			checksum += ch;
+			count++;
+		}
+
+		kgdb_io_ops->write_char('#');
+		kgdb_io_ops->write_char(hexchars[checksum >> 4]);
+		kgdb_io_ops->write_char(hexchars[checksum & 0xf]);
+		if (kgdb_io_ops->flush)
+			kgdb_io_ops->flush();
+
+		/* Now see what we get in reply. */
+		ch = kgdb_io_ops->read_char();
+
+		if (ch == 3)
+			ch = kgdb_io_ops->read_char();
+
+		/* If we get an ACK, we are done. */
+		if (ch == '+')
+			return;
+
+		/*
+		 * If we get the start of another packet, this means
+		 * that GDB is attempting to reconnect.  We will NAK
+		 * the packet being sent, and stop trying to send this
+		 * packet.
+		 */
+		if (ch == '$') {
+			kgdb_io_ops->write_char('-');
+			if (kgdb_io_ops->flush)
+				kgdb_io_ops->flush();
+			return;
+		}
+	}
+}
+
+static char *pack_hex_byte(char *pkt, u8 byte)
+{
+	*pkt++ = hexchars[byte >> 4];
+	*pkt++ = hexchars[byte & 0xf];
+
+	return pkt;
+}
+
+/*
+ * Convert the memory pointed to by mem into hex, placing result in buf.
+ * Return a pointer to the last char put in buf (null). May return an error.
+ */
+int kgdb_mem2hex(char *mem, char *buf, int count)
+{
+	char *tmp;
+	int err;
+
+	/*
+	 * We use the upper half of buf as an intermediate buffer for the
+	 * raw memory copy.  Hex conversion will work against this one.
+	 */
+	tmp = buf + count;
+
+	err = probe_kernel_read(tmp, mem, count);
+	if (!err) {
+		while (count > 0) {
+			buf = pack_hex_byte(buf, *tmp);
+			tmp++;
+			count--;
+		}
+
+		*buf = 0;
+	}
+
+	return err;
+}
+
+/*
+ * Copy the binary array pointed to by buf into mem.  Fix $, #, and
+ * 0x7d escaped with 0x7d.  Return a pointer to the character after
+ * the last byte written.
+ */
+static int kgdb_ebin2mem(char *buf, char *mem, int count)
+{
+	int err = 0;
+	char c;
+
+	while (count-- > 0) {
+		c = *buf++;
+		if (c == 0x7d)
+			c = *buf++ ^ 0x20;
+
+		err = probe_kernel_write(mem, &c, 1);
+		if (err)
+			break;
+
+		mem++;
+	}
+
+	return err;
+}
+
+/*
+ * Convert the hex array pointed to by buf into binary to be placed in mem.
+ * Return a pointer to the character AFTER the last byte written.
+ * May return an error.
+ */
+int kgdb_hex2mem(char *buf, char *mem, int count)
+{
+	char *tmp_raw;
+	char *tmp_hex;
+
+	/*
+	 * We use the upper half of buf as an intermediate buffer for the
+	 * raw memory that is converted from hex.
+	 */
+	tmp_raw = buf + count * 2;
+
+	tmp_hex = tmp_raw - 1;
+	while (tmp_hex >= buf) {
+		tmp_raw--;
+		*tmp_raw = hex(*tmp_hex--);
+		*tmp_raw |= hex(*tmp_hex--) << 4;
+	}
+
+	return probe_kernel_write(mem, tmp_raw, count);
+}
+
+/*
+ * While we find nice hex chars, build a long_val.
+ * Return number of chars processed.
+ */
+int kgdb_hex2long(char **ptr, long *long_val)
+{
+	int hex_val;
+	int num = 0;
+
+	*long_val = 0;
+
+	while (**ptr) {
+		hex_val = hex(**ptr);
+		if (hex_val < 0)
+			break;
+
+		*long_val = (*long_val << 4) | hex_val;
+		num++;
+		(*ptr)++;
+	}
+
+	return num;
+}
+
+/* Write memory due to an 'M' or 'X' packet. */
+static int write_mem_msg(int binary)
+{
+	char *ptr = &remcom_in_buffer[1];
+	unsigned long addr;
+	unsigned long length;
+	int err;
+
+	if (kgdb_hex2long(&ptr, &addr) > 0 && *(ptr++) == ',' &&
+	    kgdb_hex2long(&ptr, &length) > 0 && *(ptr++) == ':') {
+		if (binary)
+			err = kgdb_ebin2mem(ptr, (char *)addr, length);
+		else
+			err = kgdb_hex2mem(ptr, (char *)addr, length);
+		if (err)
+			return err;
+		if (CACHE_FLUSH_IS_SAFE)
+			flush_icache_range(addr, addr + length + 1);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static void error_packet(char *pkt, int error)
+{
+	error = -error;
+	pkt[0] = 'E';
+	pkt[1] = hexchars[(error / 10)];
+	pkt[2] = hexchars[(error % 10)];
+	pkt[3] = '\0';
+}
+
+/*
+ * Thread ID accessors. We represent a flat TID space to GDB, where
+ * the per CPU idle threads (which under Linux all have PID 0) are
+ * remapped to negative TIDs.
+ */
+
+#define BUF_THREAD_ID_SIZE	16
+
+static char *pack_threadid(char *pkt, unsigned char *id)
+{
+	char *limit;
+
+	limit = pkt + BUF_THREAD_ID_SIZE;
+	while (pkt < limit)
+		pkt = pack_hex_byte(pkt, *id++);
+
+	return pkt;
+}
+
+static void int_to_threadref(unsigned char *id, int value)
+{
+	unsigned char *scan;
+	int i = 4;
+
+	scan = (unsigned char *)id;
+	while (i--)
+		*scan++ = 0;
+	*scan++ = (value >> 24) & 0xff;
+	*scan++ = (value >> 16) & 0xff;
+	*scan++ = (value >> 8) & 0xff;
+	*scan++ = (value & 0xff);
+}
+
+static struct task_struct *getthread(struct pt_regs *regs, int tid)
+{
+	/*
+	 * Non-positive TIDs are remapped idle tasks:
+	 */
+	if (tid <= 0)
+		return idle_task(-tid);
+
+	/*
+	 * find_task_by_pid_ns() does not take the tasklist lock anymore
+	 * but is nicely RCU locked - hence is a pretty resilient
+	 * thing to use:
+	 */
+	return find_task_by_pid_ns(tid, &init_pid_ns);
+}
+
+/*
+ * CPU debug state control:
+ */
+
+#ifdef CONFIG_SMP
+static void kgdb_wait(struct pt_regs *regs)
+{
+	unsigned long flags;
+	int cpu;
+
+	local_irq_save(flags);
+	cpu = raw_smp_processor_id();
+	kgdb_info[cpu].debuggerinfo = regs;
+	kgdb_info[cpu].task = current;
+	/*
+	 * Make sure the above info reaches the primary CPU before
+	 * our cpu_in_kgdb[] flag setting does:
+	 */
+	smp_wmb();
+	atomic_set(&cpu_in_kgdb[cpu], 1);
+
+	/*
+	 * The primary CPU must be active to enter here, but this is
+	 * guard in case the primary CPU had not been selected if
+	 * this was an entry via nmi.
+	 */
+	while (atomic_read(&kgdb_active) == -1)
+		cpu_relax();
+
+	/* Wait till primary CPU goes completely into the debugger. */
+	while (!atomic_read(&cpu_in_kgdb[atomic_read(&kgdb_active)]))
+		cpu_relax();
+
+	/* Wait till primary CPU is done with debugging */
+	while (atomic_read(&passive_cpu_wait[cpu]))
+		cpu_relax();
+
+	kgdb_info[cpu].debuggerinfo = NULL;
+	kgdb_info[cpu].task = NULL;
+
+	/* fix up hardware debug registers on local cpu */
+	if (arch_kgdb_ops.correct_hw_break)
+		arch_kgdb_ops.correct_hw_break();
+
+	/* Signal the primary CPU that we are done: */
+	atomic_set(&cpu_in_kgdb[cpu], 0);
+	local_irq_restore(flags);
+}
+#endif
+
+/*
+ * Some architectures need cache flushes when we set/clear a
+ * breakpoint:
+ */
+static void kgdb_flush_swbreak_addr(unsigned long addr)
+{
+	if (!CACHE_FLUSH_IS_SAFE)
+		return;
+
+	if (current->mm) {
+		flush_cache_range(current->mm->mmap_cache,
+				  addr, addr + BREAK_INSTR_SIZE);
+	} else {
+		flush_icache_range(addr, addr + BREAK_INSTR_SIZE);
+	}
+}
+
+/*
+ * SW breakpoint management:
+ */
+static int kgdb_activate_sw_breakpoints(void)
+{
+	unsigned long addr;
+	int error = 0;
+	int i;
+
+	for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+		if (kgdb_break[i].state != BP_SET)
+			continue;
+
+		addr = kgdb_break[i].bpt_addr;
+		error = kgdb_arch_set_breakpoint(addr,
+				kgdb_break[i].saved_instr);
+		if (error)
+			return error;
+
+		kgdb_flush_swbreak_addr(addr);
+		kgdb_break[i].state = BP_ACTIVE;
+	}
+	return 0;
+}
+
+static int kgdb_set_sw_break(unsigned long addr)
+{
+	int err = kgdb_validate_break_address(addr);
+	int breakno = -1;
+	int i;
+
+	if (err)
+		return err;
+
+	for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+		if ((kgdb_break[i].state == BP_SET) &&
+					(kgdb_break[i].bpt_addr == addr))
+			return -EEXIST;
+	}
+	for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+		if (kgdb_break[i].state == BP_REMOVED &&
+					kgdb_break[i].bpt_addr == addr) {
+			breakno = i;
+			break;
+		}
+	}
+
+	if (breakno == -1) {
+		for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+			if (kgdb_break[i].state == BP_UNDEFINED) {
+				breakno = i;
+				break;
+			}
+		}
+	}
+
+	if (breakno == -1)
+		return -E2BIG;
+
+	kgdb_break[breakno].state = BP_SET;
+	kgdb_break[breakno].type = BP_BREAKPOINT;
+	kgdb_break[breakno].bpt_addr = addr;
+
+	return 0;
+}
+
+static int kgdb_deactivate_sw_breakpoints(void)
+{
+	unsigned long addr;
+	int error = 0;
+	int i;
+
+	for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+		if (kgdb_break[i].state != BP_ACTIVE)
+			continue;
+		addr = kgdb_break[i].bpt_addr;
+		error = kgdb_arch_remove_breakpoint(addr,
+					kgdb_break[i].saved_instr);
+		if (error)
+			return error;
+
+		kgdb_flush_swbreak_addr(addr);
+		kgdb_break[i].state = BP_SET;
+	}
+	return 0;
+}
+
+static int kgdb_remove_sw_break(unsigned long addr)
+{
+	int i;
+
+	for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+		if ((kgdb_break[i].state == BP_SET) &&
+				(kgdb_break[i].bpt_addr == addr)) {
+			kgdb_break[i].state = BP_REMOVED;
+			return 0;
+		}
+	}
+	return -ENOENT;
+}
+
+int kgdb_isremovedbreak(unsigned long addr)
+{
+	int i;
+
+	for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+		if ((kgdb_break[i].state == BP_REMOVED) &&
+					(kgdb_break[i].bpt_addr == addr))
+			return 1;
+	}
+	return 0;
+}
+
+int remove_all_break(void)
+{
+	unsigned long addr;
+	int error;
+	int i;
+
+	/* Clear memory breakpoints. */
+	for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) {
+		if (kgdb_break[i].state != BP_SET)
+			continue;
+		addr = kgdb_break[i].bpt_addr;
+		error = kgdb_arch_remove_breakpoint(addr,
+				kgdb_break[i].saved_instr);
+		if (error)
+			return error;
+		kgdb_break[i].state = BP_REMOVED;
+	}
+
+	/* Clear hardware breakpoints. */
+	if (arch_kgdb_ops.remove_all_hw_break)
+		arch_kgdb_ops.remove_all_hw_break();
+
+	return 0;
+}
+
+/*
+ * Remap normal tasks to their real PID, idle tasks to -1 ... -NR_CPUs:
+ */
+static inline int shadow_pid(int realpid)
+{
+	if (realpid)
+		return realpid;
+
+	return -1-raw_smp_processor_id();
+}
+
+static char gdbmsgbuf[BUFMAX + 1];
+
+static void kgdb_msg_write(const char *s, int len)
+{
+	char *bufptr;
+	int wcount;
+	int i;
+
+	/* 'O'utput */
+	gdbmsgbuf[0] = 'O';
+
+	/* Fill and send buffers... */
+	while (len > 0) {
+		bufptr = gdbmsgbuf + 1;
+
+		/* Calculate how many this time */
+		if ((len << 1) > (BUFMAX - 2))
+			wcount = (BUFMAX - 2) >> 1;
+		else
+			wcount = len;
+
+		/* Pack in hex chars */
+		for (i = 0; i < wcount; i++)
+			bufptr = pack_hex_byte(bufptr, s[i]);
+		*bufptr = '\0';
+
+		/* Move up */
+		s += wcount;
+		len -= wcount;
+
+		/* Write packet */
+		put_packet(gdbmsgbuf);
+	}
+}
+
+/*
+ * Return true if there is a valid kgdb I/O module.  Also if no
+ * debugger is attached a message can be printed to the console about
+ * waiting for the debugger to attach.
+ *
+ * The print_wait argument is only to be true when called from inside
+ * the core kgdb_handle_exception, because it will wait for the
+ * debugger to attach.
+ */
+static int kgdb_io_ready(int print_wait)
+{
+	if (!kgdb_io_ops)
+		return 0;
+	if (kgdb_connected)
+		return 1;
+	if (atomic_read(&kgdb_setting_breakpoint))
+		return 1;
+	if (print_wait)
+		printk(KERN_CRIT "KGDB: Waiting for remote debugger\n");
+	return 1;
+}
+
+/*
+ * All the functions that start with gdb_cmd are the various
+ * operations to implement the handlers for the gdbserial protocol
+ * where KGDB is communicating with an external debugger
+ */
+
+/* Handle the '?' status packets */
+static void gdb_cmd_status(struct kgdb_state *ks)
+{
+	/*
+	 * We know that this packet is only sent
+	 * during initial connect.  So to be safe,
+	 * we clear out our breakpoints now in case
+	 * GDB is reconnecting.
+	 */
+	remove_all_break();
+
+	remcom_out_buffer[0] = 'S';
+	pack_hex_byte(&remcom_out_buffer[1], ks->signo);
+}
+
+/* Handle the 'g' get registers request */
+static void gdb_cmd_getregs(struct kgdb_state *ks)
+{
+	struct task_struct *thread;
+	void *local_debuggerinfo;
+	int i;
+
+	thread = kgdb_usethread;
+	if (!thread) {
+		thread = kgdb_info[ks->cpu].task;
+		local_debuggerinfo = kgdb_info[ks->cpu].debuggerinfo;
+	} else {
+		local_debuggerinfo = NULL;
+		for (i = 0; i < NR_CPUS; i++) {
+			/*
+			 * Try to find the task on some other
+			 * or possibly this node if we do not
+			 * find the matching task then we try
+			 * to approximate the results.
+			 */
+			if (thread == kgdb_info[i].task)
+				local_debuggerinfo = kgdb_info[i].debuggerinfo;
+		}
+	}
+
+	/*
+	 * All threads that don't have debuggerinfo should be
+	 * in __schedule() sleeping, since all other CPUs
+	 * are in kgdb_wait, and thus have debuggerinfo.
+	 */
+	if (local_debuggerinfo) {
+		pt_regs_to_gdb_regs(gdb_regs, local_debuggerinfo);
+	} else {
+		/*
+		 * Pull stuff saved during switch_to; nothing
+		 * else is accessible (or even particularly
+		 * relevant).
+		 *
+		 * This should be enough for a stack trace.
+		 */
+		sleeping_thread_to_gdb_regs(gdb_regs, thread);
+	}
+	kgdb_mem2hex((char *)gdb_regs, remcom_out_buffer, NUMREGBYTES);
+}
+
+/* Handle the 'G' set registers request */
+static void gdb_cmd_setregs(struct kgdb_state *ks)
+{
+	kgdb_hex2mem(&remcom_in_buffer[1], (char *)gdb_regs, NUMREGBYTES);
+
+	if (kgdb_usethread && kgdb_usethread != current) {
+		error_packet(remcom_out_buffer, -EINVAL);
+	} else {
+		gdb_regs_to_pt_regs(gdb_regs, ks->linux_regs);
+		strcpy(remcom_out_buffer, "OK");
+	}
+}
+
+/* Handle the 'm' memory read bytes */
+static void gdb_cmd_memread(struct kgdb_state *ks)
+{
+	char *ptr = &remcom_in_buffer[1];
+	unsigned long length;
+	unsigned long addr;
+	int err;
+
+	if (kgdb_hex2long(&ptr, &addr) > 0 && *ptr++ == ',' &&
+					kgdb_hex2long(&ptr, &length) > 0) {
+		err = kgdb_mem2hex((char *)addr, remcom_out_buffer, length);
+		if (err)
+			error_packet(remcom_out_buffer, err);
+	} else {
+		error_packet(remcom_out_buffer, -EINVAL);
+	}
+}
+
+/* Handle the 'M' memory write bytes */
+static void gdb_cmd_memwrite(struct kgdb_state *ks)
+{
+	int err = write_mem_msg(0);
+
+	if (err)
+		error_packet(remcom_out_buffer, err);
+	else
+		strcpy(remcom_out_buffer, "OK");
+}
+
+/* Handle the 'X' memory binary write bytes */
+static void gdb_cmd_binwrite(struct kgdb_state *ks)
+{
+	int err = write_mem_msg(1);
+
+	if (err)
+		error_packet(remcom_out_buffer, err);
+	else
+		strcpy(remcom_out_buffer, "OK");
+}
+
+/* Handle the 'D' or 'k', detach or kill packets */
+static void gdb_cmd_detachkill(struct kgdb_state *ks)
+{
+	int error;
+
+	/* The detach case */
+	if (remcom_in_buffer[0] == 'D') {
+		error = remove_all_break();
+		if (error < 0) {
+			error_packet(remcom_out_buffer, error);
+		} else {
+			strcpy(remcom_out_buffer, "OK");
+			kgdb_connected = 0;
+		}
+		put_packet(remcom_out_buffer);
+	} else {
+		/*
+		 * Assume the kill case, with no exit code checking,
+		 * trying to force detach the debugger:
+		 */
+		remove_all_break();
+		kgdb_connected = 0;
+	}
+}
+
+/* Handle the 'R' reboot packets */
+static int gdb_cmd_reboot(struct kgdb_state *ks)
+{
+	/* For now, only honor R0 */
+	if (strcmp(remcom_in_buffer, "R0") == 0) {
+		printk(KERN_CRIT "Executing emergency reboot\n");
+		strcpy(remcom_out_buffer, "OK");
+		put_packet(remcom_out_buffer);
+
+		/*
+		 * Execution should not return from
+		 * machine_emergency_restart()
+		 */
+		machine_emergency_restart();
+		kgdb_connected = 0;
+
+		return 1;
+	}
+	return 0;
+}
+
+/* Handle the 'q' query packets */
+static void gdb_cmd_query(struct kgdb_state *ks)
+{
+	struct task_struct *thread;
+	unsigned char thref[8];
+	char *ptr;
+	int i;
+
+	switch (remcom_in_buffer[1]) {
+	case 's':
+	case 'f':
+		if (memcmp(remcom_in_buffer + 2, "ThreadInfo", 10)) {
+			error_packet(remcom_out_buffer, -EINVAL);
+			break;
+		}
+
+		if (remcom_in_buffer[1] == 'f')
+			ks->threadid = 1;
+
+		remcom_out_buffer[0] = 'm';
+		ptr = remcom_out_buffer + 1;
+
+		for (i = 0; i < 17; ks->threadid++) {
+			thread = getthread(ks->linux_regs, ks->threadid);
+			if (thread) {
+				int_to_threadref(thref, ks->threadid);
+				pack_threadid(ptr, thref);
+				ptr += BUF_THREAD_ID_SIZE;
+				*(ptr++) = ',';
+				i++;
+			}
+		}
+		*(--ptr) = '\0';
+		break;
+
+	case 'C':
+		/* Current thread id */
+		strcpy(remcom_out_buffer, "QC");
+		ks->threadid = shadow_pid(current->pid);
+		int_to_threadref(thref, ks->threadid);
+		pack_threadid(remcom_out_buffer + 2, thref);
+		break;
+	case 'T':
+		if (memcmp(remcom_in_buffer + 1, "ThreadExtraInfo,", 16)) {
+			error_packet(remcom_out_buffer, -EINVAL);
+			break;
+		}
+		ks->threadid = 0;
+		ptr = remcom_in_buffer + 17;
+		kgdb_hex2long(&ptr, &ks->threadid);
+		if (!getthread(ks->linux_regs, ks->threadid)) {
+			error_packet(remcom_out_buffer, -EINVAL);
+			break;
+		}
+		if (ks->threadid > 0) {
+			kgdb_mem2hex(getthread(ks->linux_regs,
+					ks->threadid)->comm,
+					remcom_out_buffer, 16);
+		} else {
+			static char tmpstr[23 + BUF_THREAD_ID_SIZE];
+
+			sprintf(tmpstr, "Shadow task %d for pid 0",
+					(int)(-ks->threadid-1));
+			kgdb_mem2hex(tmpstr, remcom_out_buffer, strlen(tmpstr));
+		}
+		break;
+	}
+}
+
+/* Handle the 'H' task query packets */
+static void gdb_cmd_task(struct kgdb_state *ks)
+{
+	struct task_struct *thread;
+	char *ptr;
+
+	switch (remcom_in_buffer[1]) {
+	case 'g':
+		ptr = &remcom_in_buffer[2];
+		kgdb_hex2long(&ptr, &ks->threadid);
+		thread = getthread(ks->linux_regs, ks->threadid);
+		if (!thread && ks->threadid > 0) {
+			error_packet(remcom_out_buffer, -EINVAL);
+			break;
+		}
+		kgdb_usethread = thread;
+		ks->kgdb_usethreadid = ks->threadid;
+		strcpy(remcom_out_buffer, "OK");
+		break;
+	case 'c':
+		ptr = &remcom_in_buffer[2];
+		kgdb_hex2long(&ptr, &ks->threadid);
+		if (!ks->threadid) {
+			kgdb_contthread = NULL;
+		} else {
+			thread = getthread(ks->linux_regs, ks->threadid);
+			if (!thread && ks->threadid > 0) {
+				error_packet(remcom_out_buffer, -EINVAL);
+				break;
+			}
+			kgdb_contthread = thread;
+		}
+		strcpy(remcom_out_buffer, "OK");
+		break;
+	}
+}
+
+/* Handle the 'T' thread query packets */
+static void gdb_cmd_thread(struct kgdb_state *ks)
+{
+	char *ptr = &remcom_in_buffer[1];
+	struct task_struct *thread;
+
+	kgdb_hex2long(&ptr, &ks->threadid);
+	thread = getthread(ks->linux_regs, ks->threadid);
+	if (thread)
+		strcpy(remcom_out_buffer, "OK");
+	else
+		error_packet(remcom_out_buffer, -EINVAL);
+}
+
+/* Handle the 'z' or 'Z' breakpoint remove or set packets */
+static void gdb_cmd_break(struct kgdb_state *ks)
+{
+	/*
+	 * Since GDB-5.3, it's been drafted that '0' is a software
+	 * breakpoint, '1' is a hardware breakpoint, so let's do that.
+	 */
+	char *bpt_type = &remcom_in_buffer[1];
+	char *ptr = &remcom_in_buffer[2];
+	unsigned long addr;
+	unsigned long length;
+	int error = 0;
+
+	if (arch_kgdb_ops.set_hw_breakpoint && *bpt_type >= '1') {
+		/* Unsupported */
+		if (*bpt_type > '4')
+			return;
+	} else {
+		if (*bpt_type != '0' && *bpt_type != '1')
+			/* Unsupported. */
+			return;
+	}
+
+	/*
+	 * Test if this is a hardware breakpoint, and
+	 * if we support it:
+	 */
+	if (*bpt_type == '1' && !(arch_kgdb_ops.flags & KGDB_HW_BREAKPOINT))
+		/* Unsupported. */
+		return;
+
+	if (*(ptr++) != ',') {
+		error_packet(remcom_out_buffer, -EINVAL);
+		return;
+	}
+	if (!kgdb_hex2long(&ptr, &addr)) {
+		error_packet(remcom_out_buffer, -EINVAL);
+		return;
+	}
+	if (*(ptr++) != ',' ||
+		!kgdb_hex2long(&ptr, &length)) {
+		error_packet(remcom_out_buffer, -EINVAL);
+		return;
+	}
+
+	if (remcom_in_buffer[0] == 'Z' && *bpt_type == '0')
+		error = kgdb_set_sw_break(addr);
+	else if (remcom_in_buffer[0] == 'z' && *bpt_type == '0')
+		error = kgdb_remove_sw_break(addr);
+	else if (remcom_in_buffer[0] == 'Z')
+		error = arch_kgdb_ops.set_hw_breakpoint(addr,
+			(int)length, *bpt_type);
+	else if (remcom_in_buffer[0] == 'z')
+		error = arch_kgdb_ops.remove_hw_breakpoint(addr,
+			(int) length, *bpt_type);
+
+	if (error == 0)
+		strcpy(remcom_out_buffer, "OK");
+	else
+		error_packet(remcom_out_buffer, error);
+}
+
+/* Handle the 'C' signal / exception passing packets */
+static int gdb_cmd_exception_pass(struct kgdb_state *ks)
+{
+	/* C09 == pass exception
+	 * C15 == detach kgdb, pass exception
+	 */
+	if (remcom_in_buffer[1] == '0' && remcom_in_buffer[2] == '9') {
+
+		ks->pass_exception = 1;
+		remcom_in_buffer[0] = 'c';
+
+	} else if (remcom_in_buffer[1] == '1' && remcom_in_buffer[2] == '5') {
+
+		ks->pass_exception = 1;
+		remcom_in_buffer[0] = 'D';
+		remove_all_break();
+		kgdb_connected = 0;
+		return 1;
+
+	} else {
+		error_packet(remcom_out_buffer, -EINVAL);
+		return 0;
+	}
+
+	/* Indicate fall through */
+	return -1;
+}
+
+/*
+ * This function performs all gdbserial command procesing
+ */
+static int gdb_serial_stub(struct kgdb_state *ks)
+{
+	int error = 0;
+	int tmp;
+
+	/* Clear the out buffer. */
+	memset(remcom_out_buffer, 0, sizeof(remcom_out_buffer));
+
+	if (kgdb_connected) {
+		unsigned char thref[8];
+		char *ptr;
+
+		/* Reply to host that an exception has occurred */
+		ptr = remcom_out_buffer;
+		*ptr++ = 'T';
+		ptr = pack_hex_byte(ptr, ks->signo);
+		ptr += strlen(strcpy(ptr, "thread:"));
+		int_to_threadref(thref, shadow_pid(current->pid));
+		ptr = pack_threadid(ptr, thref);
+		*ptr++ = ';';
+		put_packet(remcom_out_buffer);
+	}
+
+	kgdb_usethread = kgdb_info[ks->cpu].task;
+	ks->kgdb_usethreadid = shadow_pid(kgdb_info[ks->cpu].task->pid);
+	ks->pass_exception = 0;
+
+	while (1) {
+		error = 0;
+
+		/* Clear the out buffer. */
+		memset(remcom_out_buffer, 0, sizeof(remcom_out_buffer));
+
+		get_packet(remcom_in_buffer);
+
+		switch (remcom_in_buffer[0]) {
+		case '?': /* gdbserial status */
+			gdb_cmd_status(ks);
+			break;
+		case 'g': /* return the value of the CPU registers */
+			gdb_cmd_getregs(ks);
+			break;
+		case 'G': /* set the value of the CPU registers - return OK */
+			gdb_cmd_setregs(ks);
+			break;
+		case 'm': /* mAA..AA,LLLL  Read LLLL bytes at address AA..AA */
+			gdb_cmd_memread(ks);
+			break;
+		case 'M': /* MAA..AA,LLLL: Write LLLL bytes at address AA..AA */
+			gdb_cmd_memwrite(ks);
+			break;
+		case 'X': /* XAA..AA,LLLL: Write LLLL bytes at address AA..AA */
+			gdb_cmd_binwrite(ks);
+			break;
+			/* kill or detach. KGDB should treat this like a
+			 * continue.
+			 */
+		case 'D': /* Debugger detach */
+		case 'k': /* Debugger detach via kill */
+			gdb_cmd_detachkill(ks);
+			goto default_handle;
+		case 'R': /* Reboot */
+			if (gdb_cmd_reboot(ks))
+				goto default_handle;
+			break;
+		case 'q': /* query command */
+			gdb_cmd_query(ks);
+			break;
+		case 'H': /* task related */
+			gdb_cmd_task(ks);
+			break;
+		case 'T': /* Query thread status */
+			gdb_cmd_thread(ks);
+			break;
+		case 'z': /* Break point remove */
+		case 'Z': /* Break point set */
+			gdb_cmd_break(ks);
+			break;
+		case 'C': /* Exception passing */
+			tmp = gdb_cmd_exception_pass(ks);
+			if (tmp > 0)
+				goto default_handle;
+			if (tmp == 0)
+				break;
+			/* Fall through on tmp < 0 */
+		case 'c': /* Continue packet */
+		case 's': /* Single step packet */
+			if (kgdb_contthread && kgdb_contthread != current) {
+				/* Can't switch threads in kgdb */
+				error_packet(remcom_out_buffer, -EINVAL);
+				break;
+			}
+			kgdb_activate_sw_breakpoints();
+			/* Fall through to default processing */
+		default:
+default_handle:
+			error = kgdb_arch_handle_exception(ks->ex_vector,
+						ks->signo,
+						ks->err_code,
+						remcom_in_buffer,
+						remcom_out_buffer,
+						ks->linux_regs);
+			/*
+			 * Leave cmd processing on error, detach,
+			 * kill, continue, or single step.
+			 */
+			if (error >= 0 || remcom_in_buffer[0] == 'D' ||
+			    remcom_in_buffer[0] == 'k') {
+				error = 0;
+				goto kgdb_exit;
+			}
+
+		}
+
+		/* reply to the request */
+		put_packet(remcom_out_buffer);
+	}
+
+kgdb_exit:
+	if (ks->pass_exception)
+		error = 1;
+	return error;
+}
+
+static int kgdb_reenter_check(struct kgdb_state *ks)
+{
+	unsigned long addr;
+
+	if (atomic_read(&kgdb_active) != raw_smp_processor_id())
+		return 0;
+
+	/* Panic on recursive debugger calls: */
+	exception_level++;
+	addr = kgdb_arch_pc(ks->ex_vector, ks->linux_regs);
+	kgdb_deactivate_sw_breakpoints();
+
+	/*
+	 * If the break point removed ok at the place exception
+	 * occurred, try to recover and print a warning to the end
+	 * user because the user planted a breakpoint in a place that
+	 * KGDB needs in order to function.
+	 */
+	if (kgdb_remove_sw_break(addr) == 0) {
+		exception_level = 0;
+		kgdb_skipexception(ks->ex_vector, ks->linux_regs);
+		kgdb_activate_sw_breakpoints();
+		printk(KERN_CRIT "KGDB: re-enter error: breakpoint removed\n");
+		WARN_ON_ONCE(1);
+
+		return 1;
+	}
+	remove_all_break();
+	kgdb_skipexception(ks->ex_vector, ks->linux_regs);
+
+	if (exception_level > 1) {
+		dump_stack();
+		panic("Recursive entry to debugger");
+	}
+
+	printk(KERN_CRIT "KGDB: re-enter exception: ALL breakpoints killed\n");
+	dump_stack();
+	panic("Recursive entry to debugger");
+
+	return 1;
+}
+
+/*
+ * kgdb_handle_exception() - main entry point from a kernel exception
+ *
+ * Locking hierarchy:
+ *	interface locks, if any (begin_session)
+ *	kgdb lock (kgdb_active)
+ */
+int
+kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs)
+{
+	struct kgdb_state kgdb_var;
+	struct kgdb_state *ks = &kgdb_var;
+	unsigned long flags;
+	int error = 0;
+	int i, cpu;
+
+	ks->cpu			= raw_smp_processor_id();
+	ks->ex_vector		= evector;
+	ks->signo		= signo;
+	ks->ex_vector		= evector;
+	ks->err_code		= ecode;
+	ks->kgdb_usethreadid	= 0;
+	ks->linux_regs		= regs;
+
+	if (kgdb_reenter_check(ks))
+		return 0; /* Ouch, double exception ! */
+
+acquirelock:
+	/*
+	 * Interrupts will be restored by the 'trap return' code, except when
+	 * single stepping.
+	 */
+	local_irq_save(flags);
+
+	cpu = raw_smp_processor_id();
+
+	/*
+	 * Acquire the kgdb_active lock:
+	 */
+	while (atomic_cmpxchg(&kgdb_active, -1, cpu) != -1)
+		cpu_relax();
+
+	/*
+	 * Do not start the debugger connection on this CPU if the last
+	 * instance of the exception handler wanted to come into the
+	 * debugger on a different CPU via a single step
+	 */
+	if (atomic_read(&kgdb_cpu_doing_single_step) != -1 &&
+	    atomic_read(&kgdb_cpu_doing_single_step) != cpu) {
+
+		atomic_set(&kgdb_active, -1);
+		local_irq_restore(flags);
+
+		goto acquirelock;
+	}
+
+	if (!kgdb_io_ready(1)) {
+		error = 1;
+		goto kgdb_restore; /* No I/O connection, so resume the system */
+	}
+
+	/*
+	 * Don't enter if we have hit a removed breakpoint.
+	 */
+	if (kgdb_skipexception(ks->ex_vector, ks->linux_regs))
+		goto kgdb_restore;
+
+	/* Call the I/O driver's pre_exception routine */
+	if (kgdb_io_ops->pre_exception)
+		kgdb_io_ops->pre_exception();
+
+	kgdb_info[ks->cpu].debuggerinfo = ks->linux_regs;
+	kgdb_info[ks->cpu].task = current;
+
+	kgdb_disable_hw_debug(ks->linux_regs);
+
+	/*
+	 * Get the passive CPU lock which will hold all the non-primary
+	 * CPU in a spin state while the debugger is active
+	 */
+	if (!kgdb_single_step || !kgdb_contthread) {
+		for (i = 0; i < NR_CPUS; i++)
+			atomic_set(&passive_cpu_wait[i], 1);
+	}
+
+#ifdef CONFIG_SMP
+	/* Signal the other CPUs to enter kgdb_wait() */
+	if ((!kgdb_single_step || !kgdb_contthread) && kgdb_do_roundup)
+		kgdb_roundup_cpus(flags);
+#endif
+
+	/*
+	 * spin_lock code is good enough as a barrier so we don't
+	 * need one here:
+	 */
+	atomic_set(&cpu_in_kgdb[ks->cpu], 1);
+
+	/*
+	 * Wait for the other CPUs to be notified and be waiting for us:
+	 */
+	for_each_online_cpu(i) {
+		while (!atomic_read(&cpu_in_kgdb[i]))
+			cpu_relax();
+	}
+
+	/*
+	 * At this point the primary processor is completely
+	 * in the debugger and all secondary CPUs are quiescent
+	 */
+	kgdb_post_primary_code(ks->linux_regs, ks->ex_vector, ks->err_code);
+	kgdb_deactivate_sw_breakpoints();
+	kgdb_single_step = 0;
+	kgdb_contthread = NULL;
+	exception_level = 0;
+
+	/* Talk to debugger with gdbserial protocol */
+	error = gdb_serial_stub(ks);
+
+	/* Call the I/O driver's post_exception routine */
+	if (kgdb_io_ops->post_exception)
+		kgdb_io_ops->post_exception();
+
+	kgdb_info[ks->cpu].debuggerinfo = NULL;
+	kgdb_info[ks->cpu].task = NULL;
+	atomic_set(&cpu_in_kgdb[ks->cpu], 0);
+
+	if (!kgdb_single_step || !kgdb_contthread) {
+		for (i = NR_CPUS-1; i >= 0; i--)
+			atomic_set(&passive_cpu_wait[i], 0);
+		/*
+		 * Wait till all the CPUs have quit
+		 * from the debugger.
+		 */
+		for_each_online_cpu(i) {
+			while (atomic_read(&cpu_in_kgdb[i]))
+				cpu_relax();
+		}
+	}
+
+kgdb_restore:
+	/* Free kgdb_active */
+	atomic_set(&kgdb_active, -1);
+	local_irq_restore(flags);
+
+	return error;
+}
+
+int kgdb_nmicallback(int cpu, void *regs)
+{
+#ifdef CONFIG_SMP
+	if (!atomic_read(&cpu_in_kgdb[cpu]) &&
+			atomic_read(&kgdb_active) != cpu) {
+		kgdb_wait((struct pt_regs *)regs);
+		return 0;
+	}
+#endif
+	return 1;
+}
+
+void kgdb_console_write(struct console *co, const char *s, unsigned count)
+{
+	unsigned long flags;
+
+	/* If we're debugging, or KGDB has not connected, don't try
+	 * and print. */
+	if (!kgdb_connected || atomic_read(&kgdb_active) != -1)
+		return;
+
+	local_irq_save(flags);
+	kgdb_msg_write(s, count);
+	local_irq_restore(flags);
+}
+
+static struct console kgdbcons = {
+	.name		= "kgdb",
+	.write		= kgdb_console_write,
+	.flags		= CON_PRINTBUFFER | CON_ENABLED,
+	.index		= -1,
+};
+
+#ifdef CONFIG_MAGIC_SYSRQ
+static void sysrq_handle_gdb(int key, struct tty_struct *tty)
+{
+	if (!kgdb_io_ops) {
+		printk(KERN_CRIT "ERROR: No KGDB I/O module available\n");
+		return;
+	}
+	if (!kgdb_connected)
+		printk(KERN_CRIT "Entering KGDB\n");
+
+	kgdb_breakpoint();
+}
+
+static struct sysrq_key_op sysrq_gdb_op = {
+	.handler	= sysrq_handle_gdb,
+	.help_msg	= "Gdb",
+	.action_msg	= "GDB",
+};
+#endif
+
+static void kgdb_register_callbacks(void)
+{
+	if (!kgdb_io_module_registered) {
+		kgdb_io_module_registered = 1;
+		kgdb_arch_init();
+#ifdef CONFIG_MAGIC_SYSRQ
+		register_sysrq_key('g', &sysrq_gdb_op);
+#endif
+		if (kgdb_use_con && !kgdb_con_registered) {
+			register_console(&kgdbcons);
+			kgdb_con_registered = 1;
+		}
+	}
+}
+
+static void kgdb_unregister_callbacks(void)
+{
+	/*
+	 * When this routine is called KGDB should unregister from the
+	 * panic handler and clean up, making sure it is not handling any
+	 * break exceptions at the time.
+	 */
+	if (kgdb_io_module_registered) {
+		kgdb_io_module_registered = 0;
+		kgdb_arch_exit();
+#ifdef CONFIG_MAGIC_SYSRQ
+		unregister_sysrq_key('g', &sysrq_gdb_op);
+#endif
+		if (kgdb_con_registered) {
+			unregister_console(&kgdbcons);
+			kgdb_con_registered = 0;
+		}
+	}
+}
+
+static void kgdb_initial_breakpoint(void)
+{
+	kgdb_break_asap = 0;
+
+	printk(KERN_CRIT "kgdb: Waiting for connection from remote gdb...\n");
+	kgdb_breakpoint();
+}
+
+/**
+ *	kkgdb_register_io_module - register KGDB IO module
+ *	@new_kgdb_io_ops: the io ops vector
+ *
+ *	Register it with the KGDB core.
+ */
+int kgdb_register_io_module(struct kgdb_io *new_kgdb_io_ops)
+{
+	int err;
+
+	spin_lock(&kgdb_registration_lock);
+
+	if (kgdb_io_ops) {
+		spin_unlock(&kgdb_registration_lock);
+
+		printk(KERN_ERR "kgdb: Another I/O driver is already "
+				"registered with KGDB.\n");
+		return -EBUSY;
+	}
+
+	if (new_kgdb_io_ops->init) {
+		err = new_kgdb_io_ops->init();
+		if (err) {
+			spin_unlock(&kgdb_registration_lock);
+			return err;
+		}
+	}
+
+	kgdb_io_ops = new_kgdb_io_ops;
+
+	spin_unlock(&kgdb_registration_lock);
+
+	printk(KERN_INFO "kgdb: Registered I/O driver %s.\n",
+	       new_kgdb_io_ops->name);
+
+	/* Arm KGDB now. */
+	kgdb_register_callbacks();
+
+	if (kgdb_break_asap)
+		kgdb_initial_breakpoint();
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kgdb_register_io_module);
+
+/**
+ *	kkgdb_unregister_io_module - unregister KGDB IO module
+ *	@old_kgdb_io_ops: the io ops vector
+ *
+ *	Unregister it with the KGDB core.
+ */
+void kgdb_unregister_io_module(struct kgdb_io *old_kgdb_io_ops)
+{
+	BUG_ON(kgdb_connected);
+
+	/*
+	 * KGDB is no longer able to communicate out, so
+	 * unregister our callbacks and reset state.
+	 */
+	kgdb_unregister_callbacks();
+
+	spin_lock(&kgdb_registration_lock);
+
+	WARN_ON_ONCE(kgdb_io_ops != old_kgdb_io_ops);
+	kgdb_io_ops = NULL;
+
+	spin_unlock(&kgdb_registration_lock);
+
+	printk(KERN_INFO
+		"kgdb: Unregistered I/O driver %s, debugger disabled.\n",
+		old_kgdb_io_ops->name);
+}
+EXPORT_SYMBOL_GPL(kgdb_unregister_io_module);
+
+/**
+ * kgdb_breakpoint - generate breakpoint exception
+ *
+ * This function will generate a breakpoint exception.  It is used at the
+ * beginning of a program to sync up with a debugger and can be used
+ * otherwise as a quick means to stop program execution and "break" into
+ * the debugger.
+ */
+void kgdb_breakpoint(void)
+{
+	atomic_set(&kgdb_setting_breakpoint, 1);
+	wmb(); /* Sync point before breakpoint */
+	arch_kgdb_breakpoint();
+	wmb(); /* Sync point after breakpoint */
+	atomic_set(&kgdb_setting_breakpoint, 0);
+}
+EXPORT_SYMBOL_GPL(kgdb_breakpoint);
+
+static int __init opt_kgdb_wait(char *str)
+{
+	kgdb_break_asap = 1;
+
+	if (kgdb_io_module_registered)
+		kgdb_initial_breakpoint();
+
+	return 0;
+}
+
+early_param("kgdbwait", opt_kgdb_wait);
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 0796c1a090c0..e601d0e7ac5d 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -622,3 +622,5 @@ config PROVIDE_OHCI1394_DMA_INIT
 	  See Documentation/debugging-via-ohci1394.txt for more information.
 
 source "samples/Kconfig"
+
+source "lib/Kconfig.kgdb"
diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
new file mode 100644
index 000000000000..9631ba3baaf3
--- /dev/null
+++ b/lib/Kconfig.kgdb
@@ -0,0 +1,27 @@
+
+menuconfig KGDB
+	bool "KGDB: kernel debugging with remote gdb"
+	select FRAME_POINTER
+	depends on HAVE_ARCH_KGDB
+	depends on DEBUG_KERNEL && EXPERIMENTAL
+	help
+	  If you say Y here, it will be possible to remotely debug the
+	  kernel using gdb.  Documentation of kernel debugger is available
+	  at http://kgdb.sourceforge.net as well as in DocBook form
+	  in Documentation/DocBook/.  If unsure, say N.
+
+config HAVE_ARCH_KGDB_SHADOW_INFO
+	bool
+
+config HAVE_ARCH_KGDB
+	bool
+
+config KGDB_SERIAL_CONSOLE
+	tristate "KGDB: use kgdb over the serial console"
+	depends on KGDB
+	select CONSOLE_POLL
+	select MAGIC_SYSRQ
+	default y
+	help
+	  Share a serial console with kgdb. Sysrq-g must be used
+	  to break in initially.
-- 
cgit v1.2.3


From f2d937f3bf00665ccf048b3b6616ef95859b0945 Mon Sep 17 00:00:00 2001
From: Jason Wessel <jason.wessel@windriver.com>
Date: Thu, 17 Apr 2008 20:05:37 +0200
Subject: consoles: polling support, kgdboc

polled console handling support, to access a console in an irq-less
way while in debug or irq context.

absolutely zero impact as long as CONFIG_CONSOLE_POLL is disabled.
(which is the default)

[ jan.kiszka@siemens.com: lots of cleanups ]
[ mingo@elte.hu: redesign, splitups, cleanups. ]

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jan Kiszka <jan.kiszka@web.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/char/tty_io.c        |  47 +++++++++++++
 drivers/serial/8250.c        |  58 +++++++++++++++
 drivers/serial/Kconfig       |   3 +
 drivers/serial/Makefile      |   1 +
 drivers/serial/kgdboc.c      | 163 +++++++++++++++++++++++++++++++++++++++++++
 drivers/serial/serial_core.c |  72 ++++++++++++++++++-
 include/linux/serial_core.h  |   4 ++
 include/linux/tty_driver.h   |  12 ++++
 8 files changed, 357 insertions(+), 3 deletions(-)
 create mode 100644 drivers/serial/kgdboc.c

(limited to 'include')

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 613ec816ce60..4d3c7018f0c3 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -1155,6 +1155,48 @@ static struct tty_driver *get_tty_driver(dev_t device, int *index)
 	return NULL;
 }
 
+#ifdef CONFIG_CONSOLE_POLL
+
+/**
+ *	tty_find_polling_driver	-	find device of a polled tty
+ *	@name: name string to match
+ *	@line: pointer to resulting tty line nr
+ *
+ *	This routine returns a tty driver structure, given a name
+ *	and the condition that the tty driver is capable of polled
+ *	operation.
+ */
+struct tty_driver *tty_find_polling_driver(char *name, int *line)
+{
+	struct tty_driver *p, *res = NULL;
+	int tty_line = 0;
+	char *str;
+
+	mutex_lock(&tty_mutex);
+	/* Search through the tty devices to look for a match */
+	list_for_each_entry(p, &tty_drivers, tty_drivers) {
+		str = name + strlen(p->name);
+		tty_line = simple_strtoul(str, &str, 10);
+		if (*str == ',')
+			str++;
+		if (*str == '\0')
+			str = 0;
+
+		if (tty_line >= 0 && tty_line <= p->num && p->poll_init &&
+				!p->poll_init(p, tty_line, str)) {
+
+			res = p;
+			*line = tty_line;
+			break;
+		}
+	}
+	mutex_unlock(&tty_mutex);
+
+	return res;
+}
+EXPORT_SYMBOL_GPL(tty_find_polling_driver);
+#endif
+
 /**
  *	tty_check_change	-	check for POSIX terminal changes
  *	@tty: tty to check
@@ -3850,6 +3892,11 @@ void tty_set_operations(struct tty_driver *driver,
 	driver->write_proc = op->write_proc;
 	driver->tiocmget = op->tiocmget;
 	driver->tiocmset = op->tiocmset;
+#ifdef CONFIG_CONSOLE_POLL
+	driver->poll_init = op->poll_init;
+	driver->poll_get_char = op->poll_get_char;
+	driver->poll_put_char = op->poll_put_char;
+#endif
 }
 
 
diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 77f7a7f0646e..96a585e1cee8 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -1740,6 +1740,60 @@ static inline void wait_for_xmitr(struct uart_8250_port *up, int bits)
 	}
 }
 
+#ifdef CONFIG_CONSOLE_POLL
+/*
+ * Console polling routines for writing and reading from the uart while
+ * in an interrupt or debug context.
+ */
+
+static int serial8250_get_poll_char(struct uart_port *port)
+{
+	struct uart_8250_port *up = (struct uart_8250_port *)port;
+	unsigned char lsr = serial_inp(up, UART_LSR);
+
+	while (!(lsr & UART_LSR_DR))
+		lsr = serial_inp(up, UART_LSR);
+
+	return serial_inp(up, UART_RX);
+}
+
+
+static void serial8250_put_poll_char(struct uart_port *port,
+			 unsigned char c)
+{
+	unsigned int ier;
+	struct uart_8250_port *up = (struct uart_8250_port *)port;
+
+	/*
+	 *	First save the IER then disable the interrupts
+	 */
+	ier = serial_in(up, UART_IER);
+	if (up->capabilities & UART_CAP_UUE)
+		serial_out(up, UART_IER, UART_IER_UUE);
+	else
+		serial_out(up, UART_IER, 0);
+
+	wait_for_xmitr(up, BOTH_EMPTY);
+	/*
+	 *	Send the character out.
+	 *	If a LF, also do CR...
+	 */
+	serial_out(up, UART_TX, c);
+	if (c == 10) {
+		wait_for_xmitr(up, BOTH_EMPTY);
+		serial_out(up, UART_TX, 13);
+	}
+
+	/*
+	 *	Finally, wait for transmitter to become empty
+	 *	and restore the IER
+	 */
+	wait_for_xmitr(up, BOTH_EMPTY);
+	serial_out(up, UART_IER, ier);
+}
+
+#endif /* CONFIG_CONSOLE_POLL */
+
 static int serial8250_startup(struct uart_port *port)
 {
 	struct uart_8250_port *up = (struct uart_8250_port *)port;
@@ -2386,6 +2440,10 @@ static struct uart_ops serial8250_pops = {
 	.request_port	= serial8250_request_port,
 	.config_port	= serial8250_config_port,
 	.verify_port	= serial8250_verify_port,
+#ifdef CONFIG_CONSOLE_POLL
+	.poll_get_char = serial8250_get_poll_char,
+	.poll_put_char = serial8250_put_poll_char,
+#endif
 };
 
 static struct uart_8250_port serial8250_ports[UART_NR];
diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index cf627cd1b4c8..f7cd9504d811 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -961,6 +961,9 @@ config SERIAL_CORE
 config SERIAL_CORE_CONSOLE
 	bool
 
+config CONSOLE_POLL
+	bool
+
 config SERIAL_68328
 	bool "68328 serial support"
 	depends on M68328 || M68EZ328 || M68VZ328
diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile
index 640cfe44a56d..3cbea5494724 100644
--- a/drivers/serial/Makefile
+++ b/drivers/serial/Makefile
@@ -66,4 +66,5 @@ obj-$(CONFIG_SERIAL_UARTLITE) += uartlite.o
 obj-$(CONFIG_SERIAL_NETX) += netx-serial.o
 obj-$(CONFIG_SERIAL_OF_PLATFORM) += of_serial.o
 obj-$(CONFIG_SERIAL_KS8695) += serial_ks8695.o
+obj-$(CONFIG_KGDB_SERIAL_CONSOLE) += kgdboc.o
 obj-$(CONFIG_SERIAL_QE) += ucc_uart.o
diff --git a/drivers/serial/kgdboc.c b/drivers/serial/kgdboc.c
new file mode 100644
index 000000000000..341830791608
--- /dev/null
+++ b/drivers/serial/kgdboc.c
@@ -0,0 +1,163 @@
+/*
+ * Based on the same principle as kgdboe using the NETPOLL api, this
+ * driver uses a console polling api to implement a gdb serial inteface
+ * which is multiplexed on a console port.
+ *
+ * Maintainer: Jason Wessel <jason.wessel@windriver.com>
+ *
+ * 2007-2008 (c) Jason Wessel - Wind River Systems, Inc.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+#include <linux/kernel.h>
+#include <linux/ctype.h>
+#include <linux/kgdb.h>
+#include <linux/tty.h>
+
+#define MAX_CONFIG_LEN		40
+
+static struct kgdb_io		kgdboc_io_ops;
+
+/* -1 = init not run yet, 0 = unconfigured, 1 = configured. */
+static int configured		= -1;
+
+static char config[MAX_CONFIG_LEN];
+static struct kparam_string kps = {
+	.string			= config,
+	.maxlen			= MAX_CONFIG_LEN,
+};
+
+static struct tty_driver	*kgdb_tty_driver;
+static int			kgdb_tty_line;
+
+static int kgdboc_option_setup(char *opt)
+{
+	if (strlen(opt) > MAX_CONFIG_LEN) {
+		printk(KERN_ERR "kgdboc: config string too long\n");
+		return -ENOSPC;
+	}
+	strcpy(config, opt);
+
+	return 0;
+}
+
+__setup("kgdboc=", kgdboc_option_setup);
+
+static int configure_kgdboc(void)
+{
+	struct tty_driver *p;
+	int tty_line = 0;
+	int err;
+
+	err = kgdboc_option_setup(config);
+	if (err || !strlen(config) || isspace(config[0]))
+		goto noconfig;
+
+	err = -ENODEV;
+
+	p = tty_find_polling_driver(config, &tty_line);
+	if (!p)
+		goto noconfig;
+
+	kgdb_tty_driver = p;
+	kgdb_tty_line = tty_line;
+
+	err = kgdb_register_io_module(&kgdboc_io_ops);
+	if (err)
+		goto noconfig;
+
+	configured = 1;
+
+	return 0;
+
+noconfig:
+	config[0] = 0;
+	configured = 0;
+
+	return err;
+}
+
+static int __init init_kgdboc(void)
+{
+	/* Already configured? */
+	if (configured == 1)
+		return 0;
+
+	return configure_kgdboc();
+}
+
+static void cleanup_kgdboc(void)
+{
+	if (configured == 1)
+		kgdb_unregister_io_module(&kgdboc_io_ops);
+}
+
+static int kgdboc_get_char(void)
+{
+	return kgdb_tty_driver->poll_get_char(kgdb_tty_driver, kgdb_tty_line);
+}
+
+static void kgdboc_put_char(u8 chr)
+{
+	kgdb_tty_driver->poll_put_char(kgdb_tty_driver, kgdb_tty_line, chr);
+}
+
+static int param_set_kgdboc_var(const char *kmessage, struct kernel_param *kp)
+{
+	if (strlen(kmessage) >= MAX_CONFIG_LEN) {
+		printk(KERN_ERR "kgdboc: config string too long\n");
+		return -ENOSPC;
+	}
+
+	/* Only copy in the string if the init function has not run yet */
+	if (configured < 0) {
+		strcpy(config, kmessage);
+		return 0;
+	}
+
+	if (kgdb_connected) {
+		printk(KERN_ERR
+		       "kgdboc: Cannot reconfigure while KGDB is connected.\n");
+
+		return -EBUSY;
+	}
+
+	strcpy(config, kmessage);
+
+	if (configured == 1)
+		cleanup_kgdboc();
+
+	/* Go and configure with the new params. */
+	return configure_kgdboc();
+}
+
+static void kgdboc_pre_exp_handler(void)
+{
+	/* Increment the module count when the debugger is active */
+	if (!kgdb_connected)
+		try_module_get(THIS_MODULE);
+}
+
+static void kgdboc_post_exp_handler(void)
+{
+	/* decrement the module count when the debugger detaches */
+	if (!kgdb_connected)
+		module_put(THIS_MODULE);
+}
+
+static struct kgdb_io kgdboc_io_ops = {
+	.name			= "kgdboc",
+	.read_char		= kgdboc_get_char,
+	.write_char		= kgdboc_put_char,
+	.pre_exception		= kgdboc_pre_exp_handler,
+	.post_exception		= kgdboc_post_exp_handler,
+};
+
+module_init(init_kgdboc);
+module_exit(cleanup_kgdboc);
+module_param_call(kgdboc, param_set_kgdboc_var, param_get_string, &kps, 0644);
+MODULE_PARM_DESC(kgdboc, "<serial_device>[,baud]");
+MODULE_DESCRIPTION("KGDB Console TTY Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 0f5a17987cca..4d7eecbead9b 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -1827,7 +1827,7 @@ uart_get_console(struct uart_port *ports, int nr, struct console *co)
  *	options.  The format of the string is <baud><parity><bits><flow>,
  *	eg: 115200n8r
  */
-void __init
+void
 uart_parse_options(char *options, int *baud, int *parity, int *bits, int *flow)
 {
 	char *s = options;
@@ -1842,6 +1842,7 @@ uart_parse_options(char *options, int *baud, int *parity, int *bits, int *flow)
 	if (*s)
 		*flow = *s;
 }
+EXPORT_SYMBOL_GPL(uart_parse_options);
 
 struct baud_rates {
 	unsigned int rate;
@@ -1872,7 +1873,7 @@ static const struct baud_rates baud_rates[] = {
  *	@bits: number of data bits
  *	@flow: flow control character - 'r' (rts)
  */
-int __init
+int
 uart_set_options(struct uart_port *port, struct console *co,
 		 int baud, int parity, int bits, int flow)
 {
@@ -1924,10 +1925,16 @@ uart_set_options(struct uart_port *port, struct console *co,
 	port->mctrl |= TIOCM_DTR;
 
 	port->ops->set_termios(port, &termios, &dummy);
-	co->cflag = termios.c_cflag;
+	/*
+	 * Allow the setting of the UART parameters with a NULL console
+	 * too:
+	 */
+	if (co)
+		co->cflag = termios.c_cflag;
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(uart_set_options);
 #endif /* CONFIG_SERIAL_CORE_CONSOLE */
 
 static void uart_change_pm(struct uart_state *state, int pm_state)
@@ -2182,6 +2189,60 @@ uart_configure_port(struct uart_driver *drv, struct uart_state *state,
 	}
 }
 
+#ifdef CONFIG_CONSOLE_POLL
+
+static int uart_poll_init(struct tty_driver *driver, int line, char *options)
+{
+	struct uart_driver *drv = driver->driver_state;
+	struct uart_state *state = drv->state + line;
+	struct uart_port *port;
+	int baud = 9600;
+	int bits = 8;
+	int parity = 'n';
+	int flow = 'n';
+
+	if (!state || !state->port)
+		return -1;
+
+	port = state->port;
+	if (!(port->ops->poll_get_char && port->ops->poll_put_char))
+		return -1;
+
+	if (options) {
+		uart_parse_options(options, &baud, &parity, &bits, &flow);
+		return uart_set_options(port, NULL, baud, parity, bits, flow);
+	}
+
+	return 0;
+}
+
+static int uart_poll_get_char(struct tty_driver *driver, int line)
+{
+	struct uart_driver *drv = driver->driver_state;
+	struct uart_state *state = drv->state + line;
+	struct uart_port *port;
+
+	if (!state || !state->port)
+		return -1;
+
+	port = state->port;
+	return port->ops->poll_get_char(port);
+}
+
+static void uart_poll_put_char(struct tty_driver *driver, int line, char ch)
+{
+	struct uart_driver *drv = driver->driver_state;
+	struct uart_state *state = drv->state + line;
+	struct uart_port *port;
+
+	if (!state || !state->port)
+		return;
+
+	port = state->port;
+	port->ops->poll_put_char(port, ch);
+}
+#endif
+
 static const struct tty_operations uart_ops = {
 	.open		= uart_open,
 	.close		= uart_close,
@@ -2206,6 +2267,11 @@ static const struct tty_operations uart_ops = {
 #endif
 	.tiocmget	= uart_tiocmget,
 	.tiocmset	= uart_tiocmset,
+#ifdef CONFIG_CONSOLE_POLL
+	.poll_init	= uart_poll_init,
+	.poll_get_char	= uart_poll_get_char,
+	.poll_put_char	= uart_poll_put_char,
+#endif
 };
 
 /**
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 289942fc6655..7cb094a82456 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -213,6 +213,10 @@ struct uart_ops {
 	void		(*config_port)(struct uart_port *, int);
 	int		(*verify_port)(struct uart_port *, struct serial_struct *);
 	int		(*ioctl)(struct uart_port *, unsigned int, unsigned long);
+#ifdef CONFIG_CONSOLE_POLL
+	void	(*poll_put_char)(struct uart_port *, unsigned char);
+	int		(*poll_get_char)(struct uart_port *);
+#endif
 };
 
 #define UART_CONFIG_TYPE	(1 << 0)
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index 85c95cd39bc3..21f69aca4505 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -125,6 +125,7 @@
 #include <linux/cdev.h>
 
 struct tty_struct;
+struct tty_driver;
 
 struct tty_operations {
 	int  (*open)(struct tty_struct * tty, struct file * filp);
@@ -157,6 +158,11 @@ struct tty_operations {
 	int (*tiocmget)(struct tty_struct *tty, struct file *file);
 	int (*tiocmset)(struct tty_struct *tty, struct file *file,
 			unsigned int set, unsigned int clear);
+#ifdef CONFIG_CONSOLE_POLL
+	int (*poll_init)(struct tty_driver *driver, int line, char *options);
+	int (*poll_get_char)(struct tty_driver *driver, int line);
+	void (*poll_put_char)(struct tty_driver *driver, int line, char ch);
+#endif
 };
 
 struct tty_driver {
@@ -220,6 +226,11 @@ struct tty_driver {
 	int (*tiocmget)(struct tty_struct *tty, struct file *file);
 	int (*tiocmset)(struct tty_struct *tty, struct file *file,
 			unsigned int set, unsigned int clear);
+#ifdef CONFIG_CONSOLE_POLL
+	int (*poll_init)(struct tty_driver *driver, int line, char *options);
+	int (*poll_get_char)(struct tty_driver *driver, int line);
+	void (*poll_put_char)(struct tty_driver *driver, int line, char ch);
+#endif
 
 	struct list_head tty_drivers;
 };
@@ -230,6 +241,7 @@ struct tty_driver *alloc_tty_driver(int lines);
 void put_tty_driver(struct tty_driver *driver);
 void tty_set_operations(struct tty_driver *driver,
 			const struct tty_operations *op);
+extern struct tty_driver *tty_find_polling_driver(char *name, int *line);
 
 /* tty driver magic number */
 #define TTY_DRIVER_MAGIC		0x5402
-- 
cgit v1.2.3


From 82da3ff89dc2a1842cff9b0d4cbc345cb90b59e1 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 17 Apr 2008 20:05:37 +0200
Subject: x86: kgdb support

simplified and streamlined kgdb support on x86, both 32-bit and 64-bit,
based on patch from:

  Subject: kgdb: core-lite
  From: Jason Wessel <jason.wessel@windriver.com>

[ and countless other authors - see the patch for details. ]

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Jan Kiszka <jan.kiszka@web.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig         |   1 +
 arch/x86/kernel/Makefile |   1 +
 arch/x86/kernel/kgdb.c   | 417 +++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-x86/kgdb.h   |  81 +++++++++
 4 files changed, 500 insertions(+)
 create mode 100644 arch/x86/kernel/kgdb.c
 create mode 100644 include/asm-x86/kgdb.h

(limited to 'include')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6c70fed0f9a0..5c4c8d7a46cc 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -23,6 +23,7 @@ config X86
 	select HAVE_KPROBES
 	select HAVE_KRETPROBES
 	select HAVE_KVM if ((X86_32 && !X86_VOYAGER && !X86_VISWS && !X86_NUMAQ) || X86_64)
+	select HAVE_ARCH_KGDB
 
 
 config GENERIC_LOCKBREAK
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 4eb5ce841106..4a4260c7f672 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -66,6 +66,7 @@ obj-$(CONFIG_MODULES)		+= module_$(BITS).o
 obj-$(CONFIG_ACPI_SRAT) 	+= srat_32.o
 obj-$(CONFIG_EFI) 		+= efi.o efi_$(BITS).o efi_stub_$(BITS).o
 obj-$(CONFIG_DOUBLEFAULT) 	+= doublefault_32.o
+obj-$(CONFIG_KGDB)		+= kgdb.o
 obj-$(CONFIG_VM86)		+= vm86_32.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 
diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
new file mode 100644
index 000000000000..37194d6374d8
--- /dev/null
+++ b/arch/x86/kernel/kgdb.c
@@ -0,0 +1,417 @@
+/*
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ */
+
+/*
+ * Copyright (C) 2004 Amit S. Kale <amitkale@linsyssoft.com>
+ * Copyright (C) 2000-2001 VERITAS Software Corporation.
+ * Copyright (C) 2002 Andi Kleen, SuSE Labs
+ * Copyright (C) 2004 LinSysSoft Technologies Pvt. Ltd.
+ * Copyright (C) 2007 MontaVista Software, Inc.
+ * Copyright (C) 2007-2008 Jason Wessel, Wind River Systems, Inc.
+ */
+/****************************************************************************
+ *  Contributor:     Lake Stevens Instrument Division$
+ *  Written by:      Glenn Engel $
+ *  Updated by:	     Amit Kale<akale@veritas.com>
+ *  Updated by:	     Tom Rini <trini@kernel.crashing.org>
+ *  Updated by:	     Jason Wessel <jason.wessel@windriver.com>
+ *  Modified for 386 by Jim Kingdon, Cygnus Support.
+ *  Origianl kgdb, compatibility with 2.1.xx kernel by
+ *  David Grothe <dave@gcom.com>
+ *  Integrated into 2.2.5 kernel by Tigran Aivazian <tigran@sco.com>
+ *  X86_64 changes from Andi Kleen's patch merged by Jim Houston
+ */
+#include <linux/spinlock.h>
+#include <linux/kdebug.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/sched.h>
+#include <linux/delay.h>
+#include <linux/kgdb.h>
+#include <linux/init.h>
+#include <linux/smp.h>
+
+#include <asm/apicdef.h>
+#include <asm/system.h>
+
+#ifdef CONFIG_X86_32
+# include <mach_ipi.h>
+#else
+# include <asm/mach_apic.h>
+#endif
+
+/*
+ * Put the error code here just in case the user cares:
+ */
+static int gdb_x86errcode;
+
+/*
+ * Likewise, the vector number here (since GDB only gets the signal
+ * number through the usual means, and that's not very specific):
+ */
+static int gdb_x86vector = -1;
+
+/**
+ *	pt_regs_to_gdb_regs - Convert ptrace regs to GDB regs
+ *	@gdb_regs: A pointer to hold the registers in the order GDB wants.
+ *	@regs: The &struct pt_regs of the current process.
+ *
+ *	Convert the pt_regs in @regs into the format for registers that
+ *	GDB expects, stored in @gdb_regs.
+ */
+void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
+{
+	gdb_regs[GDB_AX]	= regs->ax;
+	gdb_regs[GDB_BX]	= regs->bx;
+	gdb_regs[GDB_CX]	= regs->cx;
+	gdb_regs[GDB_DX]	= regs->dx;
+	gdb_regs[GDB_SI]	= regs->si;
+	gdb_regs[GDB_DI]	= regs->di;
+	gdb_regs[GDB_BP]	= regs->bp;
+	gdb_regs[GDB_PS]	= regs->flags;
+	gdb_regs[GDB_PC]	= regs->ip;
+#ifdef CONFIG_X86_32
+	gdb_regs[GDB_DS]	= regs->ds;
+	gdb_regs[GDB_ES]	= regs->es;
+	gdb_regs[GDB_CS]	= regs->cs;
+	gdb_regs[GDB_SS]	= __KERNEL_DS;
+	gdb_regs[GDB_FS]	= 0xFFFF;
+	gdb_regs[GDB_GS]	= 0xFFFF;
+#else
+	gdb_regs[GDB_R8]	= regs->r8;
+	gdb_regs[GDB_R9]	= regs->r9;
+	gdb_regs[GDB_R10]	= regs->r10;
+	gdb_regs[GDB_R11]	= regs->r11;
+	gdb_regs[GDB_R12]	= regs->r12;
+	gdb_regs[GDB_R13]	= regs->r13;
+	gdb_regs[GDB_R14]	= regs->r14;
+	gdb_regs[GDB_R15]	= regs->r15;
+#endif
+	gdb_regs[GDB_SP]	= regs->sp;
+}
+
+/**
+ *	sleeping_thread_to_gdb_regs - Convert ptrace regs to GDB regs
+ *	@gdb_regs: A pointer to hold the registers in the order GDB wants.
+ *	@p: The &struct task_struct of the desired process.
+ *
+ *	Convert the register values of the sleeping process in @p to
+ *	the format that GDB expects.
+ *	This function is called when kgdb does not have access to the
+ *	&struct pt_regs and therefore it should fill the gdb registers
+ *	@gdb_regs with what has	been saved in &struct thread_struct
+ *	thread field during switch_to.
+ */
+void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
+{
+	gdb_regs[GDB_AX]	= 0;
+	gdb_regs[GDB_BX]	= 0;
+	gdb_regs[GDB_CX]	= 0;
+	gdb_regs[GDB_DX]	= 0;
+	gdb_regs[GDB_SI]	= 0;
+	gdb_regs[GDB_DI]	= 0;
+	gdb_regs[GDB_BP]	= *(unsigned long *)p->thread.sp;
+#ifdef CONFIG_X86_32
+	gdb_regs[GDB_DS]	= __KERNEL_DS;
+	gdb_regs[GDB_ES]	= __KERNEL_DS;
+	gdb_regs[GDB_PS]	= 0;
+	gdb_regs[GDB_CS]	= __KERNEL_CS;
+	gdb_regs[GDB_PC]	= p->thread.ip;
+	gdb_regs[GDB_SS]	= __KERNEL_DS;
+	gdb_regs[GDB_FS]	= 0xFFFF;
+	gdb_regs[GDB_GS]	= 0xFFFF;
+#else
+	gdb_regs[GDB_PS]	= *(unsigned long *)(p->thread.sp + 8);
+	gdb_regs[GDB_PC]	= 0;
+	gdb_regs[GDB_R8]	= 0;
+	gdb_regs[GDB_R9]	= 0;
+	gdb_regs[GDB_R10]	= 0;
+	gdb_regs[GDB_R11]	= 0;
+	gdb_regs[GDB_R12]	= 0;
+	gdb_regs[GDB_R13]	= 0;
+	gdb_regs[GDB_R14]	= 0;
+	gdb_regs[GDB_R15]	= 0;
+#endif
+	gdb_regs[GDB_SP]	= p->thread.sp;
+}
+
+/**
+ *	gdb_regs_to_pt_regs - Convert GDB regs to ptrace regs.
+ *	@gdb_regs: A pointer to hold the registers we've received from GDB.
+ *	@regs: A pointer to a &struct pt_regs to hold these values in.
+ *
+ *	Convert the GDB regs in @gdb_regs into the pt_regs, and store them
+ *	in @regs.
+ */
+void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs)
+{
+	regs->ax		= gdb_regs[GDB_AX];
+	regs->bx		= gdb_regs[GDB_BX];
+	regs->cx		= gdb_regs[GDB_CX];
+	regs->dx		= gdb_regs[GDB_DX];
+	regs->si		= gdb_regs[GDB_SI];
+	regs->di		= gdb_regs[GDB_DI];
+	regs->bp		= gdb_regs[GDB_BP];
+	regs->flags		= gdb_regs[GDB_PS];
+	regs->ip		= gdb_regs[GDB_PC];
+#ifdef CONFIG_X86_32
+	regs->ds		= gdb_regs[GDB_DS];
+	regs->es		= gdb_regs[GDB_ES];
+	regs->cs		= gdb_regs[GDB_CS];
+#else
+	regs->r8		= gdb_regs[GDB_R8];
+	regs->r9		= gdb_regs[GDB_R9];
+	regs->r10		= gdb_regs[GDB_R10];
+	regs->r11		= gdb_regs[GDB_R11];
+	regs->r12		= gdb_regs[GDB_R12];
+	regs->r13		= gdb_regs[GDB_R13];
+	regs->r14		= gdb_regs[GDB_R14];
+	regs->r15		= gdb_regs[GDB_R15];
+#endif
+}
+
+/**
+ *	kgdb_post_primary_code - Save error vector/code numbers.
+ *	@regs: Original pt_regs.
+ *	@e_vector: Original error vector.
+ *	@err_code: Original error code.
+ *
+ *	This is needed on architectures which support SMP and KGDB.
+ *	This function is called after all the slave cpus have been put
+ *	to a know spin state and the primary CPU has control over KGDB.
+ */
+void kgdb_post_primary_code(struct pt_regs *regs, int e_vector, int err_code)
+{
+	/* primary processor is completely in the debugger */
+	gdb_x86vector = e_vector;
+	gdb_x86errcode = err_code;
+}
+
+#ifdef CONFIG_SMP
+/**
+ *	kgdb_roundup_cpus - Get other CPUs into a holding pattern
+ *	@flags: Current IRQ state
+ *
+ *	On SMP systems, we need to get the attention of the other CPUs
+ *	and get them be in a known state.  This should do what is needed
+ *	to get the other CPUs to call kgdb_wait(). Note that on some arches,
+ *	the NMI approach is not used for rounding up all the CPUs. For example,
+ *	in case of MIPS, smp_call_function() is used to roundup CPUs. In
+ *	this case, we have to make sure that interrupts are enabled before
+ *	calling smp_call_function(). The argument to this function is
+ *	the flags that will be used when restoring the interrupts. There is
+ *	local_irq_save() call before kgdb_roundup_cpus().
+ *
+ *	On non-SMP systems, this is not called.
+ */
+void kgdb_roundup_cpus(unsigned long flags)
+{
+	send_IPI_allbutself(APIC_DM_NMI);
+}
+#endif
+
+/**
+ *	kgdb_arch_handle_exception - Handle architecture specific GDB packets.
+ *	@vector: The error vector of the exception that happened.
+ *	@signo: The signal number of the exception that happened.
+ *	@err_code: The error code of the exception that happened.
+ *	@remcom_in_buffer: The buffer of the packet we have read.
+ *	@remcom_out_buffer: The buffer of %BUFMAX bytes to write a packet into.
+ *	@regs: The &struct pt_regs of the current process.
+ *
+ *	This function MUST handle the 'c' and 's' command packets,
+ *	as well packets to set / remove a hardware breakpoint, if used.
+ *	If there are additional packets which the hardware needs to handle,
+ *	they are handled here.  The code should return -1 if it wants to
+ *	process more packets, and a %0 or %1 if it wants to exit from the
+ *	kgdb callback.
+ */
+int kgdb_arch_handle_exception(int e_vector, int signo, int err_code,
+			       char *remcomInBuffer, char *remcomOutBuffer,
+			       struct pt_regs *linux_regs)
+{
+	unsigned long addr;
+	char *ptr;
+	int newPC;
+
+	switch (remcomInBuffer[0]) {
+	case 'c':
+	case 's':
+		/* try to read optional parameter, pc unchanged if no parm */
+		ptr = &remcomInBuffer[1];
+		if (kgdb_hex2long(&ptr, &addr))
+			linux_regs->ip = addr;
+		newPC = linux_regs->ip;
+
+		/* clear the trace bit */
+		linux_regs->flags &= ~TF_MASK;
+		atomic_set(&kgdb_cpu_doing_single_step, -1);
+
+		/* set the trace bit if we're stepping */
+		if (remcomInBuffer[0] == 's') {
+			linux_regs->flags |= TF_MASK;
+			kgdb_single_step = 1;
+			if (kgdb_contthread) {
+				atomic_set(&kgdb_cpu_doing_single_step,
+					   raw_smp_processor_id());
+			}
+		}
+
+		return 0;
+	}
+
+	/* this means that we do not want to exit from the handler: */
+	return -1;
+}
+
+static inline int
+single_step_cont(struct pt_regs *regs, struct die_args *args)
+{
+	/*
+	 * Single step exception from kernel space to user space so
+	 * eat the exception and continue the process:
+	 */
+	printk(KERN_ERR "KGDB: trap/step from kernel to user space, "
+			"resuming...\n");
+	kgdb_arch_handle_exception(args->trapnr, args->signr,
+				   args->err, "c", "", regs);
+
+	return NOTIFY_STOP;
+}
+
+static int __kgdb_notify(struct die_args *args, unsigned long cmd)
+{
+	struct pt_regs *regs = args->regs;
+
+	switch (cmd) {
+	case DIE_NMI:
+		if (atomic_read(&kgdb_active) != -1) {
+			/* KGDB CPU roundup */
+			kgdb_nmicallback(raw_smp_processor_id(), regs);
+			return NOTIFY_STOP;
+		}
+		return NOTIFY_DONE;
+
+	case DIE_NMI_IPI:
+		if (atomic_read(&kgdb_active) != -1) {
+			/* KGDB CPU roundup: */
+			if (kgdb_nmicallback(raw_smp_processor_id(), regs))
+				return NOTIFY_DONE;
+			return NOTIFY_STOP;
+		}
+		return NOTIFY_DONE;
+
+	case DIE_NMIWATCHDOG:
+		if (atomic_read(&kgdb_active) != -1) {
+			/* KGDB CPU roundup: */
+			kgdb_nmicallback(raw_smp_processor_id(), regs);
+			return NOTIFY_STOP;
+		}
+		/* Enter debugger: */
+		break;
+
+	case DIE_DEBUG:
+		if (atomic_read(&kgdb_cpu_doing_single_step) ==
+			raw_smp_processor_id() &&
+			user_mode(regs))
+			return single_step_cont(regs, args);
+		/* fall through */
+	default:
+		if (user_mode(regs))
+			return NOTIFY_DONE;
+	}
+
+	if (kgdb_handle_exception(args->trapnr, args->signr, args->err, regs))
+		return NOTIFY_DONE;
+
+	return NOTIFY_STOP;
+}
+
+static int
+kgdb_notify(struct notifier_block *self, unsigned long cmd, void *ptr)
+{
+	unsigned long flags;
+	int ret;
+
+	local_irq_save(flags);
+	ret = __kgdb_notify(ptr, cmd);
+	local_irq_restore(flags);
+
+	return ret;
+}
+
+static struct notifier_block kgdb_notifier = {
+	.notifier_call	= kgdb_notify,
+
+	/*
+	 * Lowest-prio notifier priority, we want to be notified last:
+	 */
+	.priority	= -INT_MAX,
+};
+
+/**
+ *	kgdb_arch_init - Perform any architecture specific initalization.
+ *
+ *	This function will handle the initalization of any architecture
+ *	specific callbacks.
+ */
+int kgdb_arch_init(void)
+{
+	return register_die_notifier(&kgdb_notifier);
+}
+
+/**
+ *	kgdb_arch_exit - Perform any architecture specific uninitalization.
+ *
+ *	This function will handle the uninitalization of any architecture
+ *	specific callbacks, for dynamic registration and unregistration.
+ */
+void kgdb_arch_exit(void)
+{
+	unregister_die_notifier(&kgdb_notifier);
+}
+
+/**
+ *
+ *	kgdb_skipexception - Bail out of KGDB when we've been triggered.
+ *	@exception: Exception vector number
+ *	@regs: Current &struct pt_regs.
+ *
+ *	On some architectures we need to skip a breakpoint exception when
+ *	it occurs after a breakpoint has been removed.
+ *
+ * Skip an int3 exception when it occurs after a breakpoint has been
+ * removed. Backtrack eip by 1 since the int3 would have caused it to
+ * increment by 1.
+ */
+int kgdb_skipexception(int exception, struct pt_regs *regs)
+{
+	if (exception == 3 && kgdb_isremovedbreak(regs->ip - 1)) {
+		regs->ip -= 1;
+		return 1;
+	}
+	return 0;
+}
+
+unsigned long kgdb_arch_pc(int exception, struct pt_regs *regs)
+{
+	if (exception == 3)
+		return instruction_pointer(regs) - 1;
+	return instruction_pointer(regs);
+}
+
+struct kgdb_arch arch_kgdb_ops = {
+	/* Breakpoint instruction: */
+	.gdb_bpt_instr		= { 0xcc },
+};
diff --git a/include/asm-x86/kgdb.h b/include/asm-x86/kgdb.h
new file mode 100644
index 000000000000..484c47554f3b
--- /dev/null
+++ b/include/asm-x86/kgdb.h
@@ -0,0 +1,81 @@
+#ifndef _ASM_KGDB_H_
+#define _ASM_KGDB_H_
+
+/*
+ * Copyright (C) 2001-2004 Amit S. Kale
+ * Copyright (C) 2008 Wind River Systems, Inc.
+ */
+
+/*
+ * BUFMAX defines the maximum number of characters in inbound/outbound
+ * buffers at least NUMREGBYTES*2 are needed for register packets
+ * Longer buffer is needed to list all threads
+ */
+#define BUFMAX			1024
+
+/*
+ *  Note that this register image is in a different order than
+ *  the register image that Linux produces at interrupt time.
+ *
+ *  Linux's register image is defined by struct pt_regs in ptrace.h.
+ *  Just why GDB uses a different order is a historical mystery.
+ */
+#ifdef CONFIG_X86_32
+enum regnames {
+	GDB_AX,			/* 0 */
+	GDB_CX,			/* 1 */
+	GDB_DX,			/* 2 */
+	GDB_BX,			/* 3 */
+	GDB_SP,			/* 4 */
+	GDB_BP,			/* 5 */
+	GDB_SI,			/* 6 */
+	GDB_DI,			/* 7 */
+	GDB_PC,			/* 8 also known as eip */
+	GDB_PS,			/* 9 also known as eflags */
+	GDB_CS,			/* 10 */
+	GDB_SS,			/* 11 */
+	GDB_DS,			/* 12 */
+	GDB_ES,			/* 13 */
+	GDB_FS,			/* 14 */
+	GDB_GS,			/* 15 */
+};
+#else /* ! CONFIG_X86_32 */
+enum regnames {
+	GDB_AX,			/* 0 */
+	GDB_DX,			/* 1 */
+	GDB_CX,			/* 2 */
+	GDB_BX,			/* 3 */
+	GDB_SI,			/* 4 */
+	GDB_DI,			/* 5 */
+	GDB_BP,			/* 6 */
+	GDB_SP,			/* 7 */
+	GDB_R8,			/* 8 */
+	GDB_R9,			/* 9 */
+	GDB_R10,		/* 10 */
+	GDB_R11,		/* 11 */
+	GDB_R12,		/* 12 */
+	GDB_R13,		/* 13 */
+	GDB_R14,		/* 14 */
+	GDB_R15,		/* 15 */
+	GDB_PC,			/* 16 */
+	GDB_PS,			/* 17 */
+};
+#endif /* CONFIG_X86_32 */
+
+/*
+ * Number of bytes of registers:
+ */
+#ifdef CONFIG_X86_32
+# define NUMREGBYTES		64
+#else
+# define NUMREGBYTES		((GDB_PS+1)*8)
+#endif
+
+static inline void arch_kgdb_breakpoint(void)
+{
+	asm("   int $3");
+}
+#define BREAK_INSTR_SIZE	1
+#define CACHE_FLUSH_IS_SAFE	1
+
+#endif				/* _ASM_KGDB_H_ */
-- 
cgit v1.2.3


From d359752407f8916c29ad53a5c30ac73e338f2797 Mon Sep 17 00:00:00 2001
From: Jason Wessel <jason.wessel@windriver.com>
Date: Fri, 15 Feb 2008 14:55:53 -0600
Subject: kgdb: fix NMI hangs

This patch fixes the hang regression with kgdb when the NMI interrupt
comes in while the master core is returning from an exception.

Adjust the NMI logic such that KGDB will not stop NMI exceptions from
occurring by in general returning NOTIFY_DONE.  It is not possible to
distinguish the debug NMI sync vs the normal NMI apic interrupt so
kgdb needs to catch the unknown NMI if it the debugger was previously
active on one of the cpus.

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/kgdb.c     | 18 +++++++++++++++---
 arch/x86/kernel/traps_32.c |  2 ++
 arch/x86/kernel/traps_64.c |  2 ++
 include/asm-x86/kdebug.h   |  1 +
 4 files changed, 20 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index 37194d6374d8..5d7a21119bf8 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -41,6 +41,7 @@
 #include <linux/kgdb.h>
 #include <linux/init.h>
 #include <linux/smp.h>
+#include <linux/nmi.h>
 
 #include <asm/apicdef.h>
 #include <asm/system.h>
@@ -290,6 +291,8 @@ single_step_cont(struct pt_regs *regs, struct die_args *args)
 	return NOTIFY_STOP;
 }
 
+static int was_in_debug_nmi[NR_CPUS];
+
 static int __kgdb_notify(struct die_args *args, unsigned long cmd)
 {
 	struct pt_regs *regs = args->regs;
@@ -299,15 +302,24 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd)
 		if (atomic_read(&kgdb_active) != -1) {
 			/* KGDB CPU roundup */
 			kgdb_nmicallback(raw_smp_processor_id(), regs);
+			was_in_debug_nmi[raw_smp_processor_id()] = 1;
+			touch_nmi_watchdog();
 			return NOTIFY_STOP;
 		}
 		return NOTIFY_DONE;
 
 	case DIE_NMI_IPI:
 		if (atomic_read(&kgdb_active) != -1) {
-			/* KGDB CPU roundup: */
-			if (kgdb_nmicallback(raw_smp_processor_id(), regs))
-				return NOTIFY_DONE;
+			/* KGDB CPU roundup */
+			kgdb_nmicallback(raw_smp_processor_id(), regs);
+			was_in_debug_nmi[raw_smp_processor_id()] = 1;
+			touch_nmi_watchdog();
+		}
+		return NOTIFY_DONE;
+
+	case DIE_NMIUNKNOWN:
+		if (was_in_debug_nmi[raw_smp_processor_id()]) {
+			was_in_debug_nmi[raw_smp_processor_id()] = 0;
 			return NOTIFY_STOP;
 		}
 		return NOTIFY_DONE;
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c
index b22c01e05a18..c5421f30d678 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps_32.c
@@ -708,6 +708,8 @@ io_check_error(unsigned char reason, struct pt_regs * regs)
 static __kprobes void
 unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 {
+	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
+		return;
 #ifdef CONFIG_MCA
 	/* Might actually be able to figure out what the guilty party
 	* is. */
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
index 045466681911..055b1650c69d 100644
--- a/arch/x86/kernel/traps_64.c
+++ b/arch/x86/kernel/traps_64.c
@@ -806,6 +806,8 @@ io_check_error(unsigned char reason, struct pt_regs * regs)
 static __kprobes void
 unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 {
+	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
+		return;
 	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
 		reason);
 	printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n");
diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index 99dcbafa1511..35231240ebdc 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -20,6 +20,7 @@ enum die_val {
 	DIE_CALL,
 	DIE_NMI_IPI,
 	DIE_PAGE_FAULT,
+	DIE_NMIUNKNOWN,
 };
 
 extern void printk_address(unsigned long address, int reliable);
-- 
cgit v1.2.3


From 7c3078b637882303b1dcf6a16229d0e35f6b60a5 Mon Sep 17 00:00:00 2001
From: Jason Wessel <jason.wessel@windriver.com>
Date: Fri, 15 Feb 2008 14:55:54 -0600
Subject: kgdb: clocksource watchdog

In order to not trip the clocksource watchdog, kgdb must touch the
clocksource watchdog on the return to normal system run state.

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/clocksource.h |  1 +
 kernel/kgdb.c               |  4 ++++
 kernel/time/clocksource.c   | 12 ++++++++++++
 3 files changed, 17 insertions(+)

(limited to 'include')

diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index 85778a4b1209..35094479ca55 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -216,6 +216,7 @@ static inline void clocksource_calculate_interval(struct clocksource *c,
 /* used to install a new clocksource */
 extern int clocksource_register(struct clocksource*);
 extern void clocksource_unregister(struct clocksource*);
+extern void clocksource_touch_watchdog(void);
 extern struct clocksource* clocksource_get_next(void);
 extern void clocksource_change_rating(struct clocksource *cs, int rating);
 extern void clocksource_resume(void);
diff --git a/kernel/kgdb.c b/kernel/kgdb.c
index 017ee782bc08..e3f603740425 100644
--- a/kernel/kgdb.c
+++ b/kernel/kgdb.c
@@ -28,6 +28,7 @@
  * kind, whether express or implied.
  */
 #include <linux/pid_namespace.h>
+#include <linux/clocksource.h>
 #include <linux/interrupt.h>
 #include <linux/spinlock.h>
 #include <linux/console.h>
@@ -574,6 +575,7 @@ static void kgdb_wait(struct pt_regs *regs)
 
 	/* Signal the primary CPU that we are done: */
 	atomic_set(&cpu_in_kgdb[cpu], 0);
+	clocksource_touch_watchdog();
 	local_irq_restore(flags);
 }
 #endif
@@ -1396,6 +1398,7 @@ acquirelock:
 	    atomic_read(&kgdb_cpu_doing_single_step) != cpu) {
 
 		atomic_set(&kgdb_active, -1);
+		clocksource_touch_watchdog();
 		local_irq_restore(flags);
 
 		goto acquirelock;
@@ -1487,6 +1490,7 @@ acquirelock:
 kgdb_restore:
 	/* Free kgdb_active */
 	atomic_set(&kgdb_active, -1);
+	clocksource_touch_watchdog();
 	local_irq_restore(flags);
 
 	return error;
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 7f60097d443a..f61402b1f2d0 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -221,6 +221,18 @@ void clocksource_resume(void)
 	spin_unlock_irqrestore(&clocksource_lock, flags);
 }
 
+/**
+ * clocksource_touch_watchdog - Update watchdog
+ *
+ * Update the watchdog after exception contexts such as kgdb so as not
+ * to incorrectly trip the watchdog.
+ *
+ */
+void clocksource_touch_watchdog(void)
+{
+	clocksource_resume_watchdog();
+}
+
 /**
  * clocksource_get_next - Returns the selected clocksource
  *
-- 
cgit v1.2.3


From e3e2aaf7dc0d82a055e084cfd48b9257c0c66b68 Mon Sep 17 00:00:00 2001
From: Jason Wessel <jason.wessel@windriver.com>
Date: Thu, 20 Mar 2008 13:43:45 -0500
Subject: kgdb: add documentation

Add in the kgdb documentation for kgdb.

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 Documentation/DocBook/Makefile  |   2 +-
 Documentation/DocBook/kgdb.tmpl | 435 ++++++++++++++++++++++++++++++++++++++++
 MAINTAINERS                     |   6 +
 include/linux/kgdb.h            |  52 +++--
 4 files changed, 473 insertions(+), 22 deletions(-)
 create mode 100644 Documentation/DocBook/kgdb.tmpl

(limited to 'include')

diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile
index 300e1707893f..e471bc466a7e 100644
--- a/Documentation/DocBook/Makefile
+++ b/Documentation/DocBook/Makefile
@@ -9,7 +9,7 @@
 DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \
 	    kernel-hacking.xml kernel-locking.xml deviceiobook.xml \
 	    procfs-guide.xml writing_usb_driver.xml networking.xml \
-	    kernel-api.xml filesystems.xml lsm.xml usb.xml \
+	    kernel-api.xml filesystems.xml lsm.xml usb.xml kgdb.xml \
 	    gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \
 	    genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml
 
diff --git a/Documentation/DocBook/kgdb.tmpl b/Documentation/DocBook/kgdb.tmpl
new file mode 100644
index 000000000000..95e5f84cbf56
--- /dev/null
+++ b/Documentation/DocBook/kgdb.tmpl
@@ -0,0 +1,435 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
+	"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
+
+<book id="kgdbOnLinux">
+ <bookinfo>
+  <title>Using kgdb and the kgdb Internals</title>
+
+  <authorgroup>
+   <author>
+    <firstname>Jason</firstname>
+    <surname>Wessel</surname>
+    <affiliation>
+     <address>
+      <email>jason.wessel@windriver.com</email>
+     </address>
+    </affiliation>
+   </author>
+  </authorgroup>
+
+  <authorgroup>
+   <author>
+    <firstname>Tom</firstname>
+    <surname>Rini</surname>
+    <affiliation>
+     <address>
+      <email>trini@kernel.crashing.org</email>
+     </address>
+    </affiliation>
+   </author>
+  </authorgroup>
+
+  <authorgroup>
+   <author>
+    <firstname>Amit S.</firstname>
+    <surname>Kale</surname>
+    <affiliation>
+     <address>
+      <email>amitkale@linsyssoft.com</email>
+     </address>
+    </affiliation>
+   </author>
+  </authorgroup>
+
+  <copyright>
+   <year>2008</year>
+   <holder>Wind River Systems, Inc.</holder>
+  </copyright>
+  <copyright>
+   <year>2004-2005</year>
+   <holder>MontaVista Software, Inc.</holder>
+  </copyright>
+  <copyright>
+   <year>2004</year>
+   <holder>Amit S. Kale</holder>
+  </copyright>
+
+  <legalnotice>
+   <para>
+   This file is licensed under the terms of the GNU General Public License
+   version 2. This program is licensed "as is" without any warranty of any
+   kind, whether express or implied.
+   </para>
+
+  </legalnotice>
+ </bookinfo>
+
+<toc></toc>
+  <chapter id="Introduction">
+    <title>Introduction</title>
+    <para>
+    kgdb is a source level debugger for linux kernel. It is used along
+    with gdb to debug a linux kernel.  The expectation is that gdb can
+    be used to "break in" to the kernel to inspect memory, variables
+    and look through a cal stack information similar to what an
+    application developer would use gdb for.  It is possible to place
+    breakpoints in kernel code and perform some limited execution
+    stepping.
+    </para>
+    <para>
+    Two machines are required for using kgdb. One of these machines is a
+    development machine and the other is a test machine.  The kernel
+    to be debugged runs on the test machine. The development machine
+    runs an instance of gdb against the vmlinux file which contains
+    the symbols (not boot image such as bzImage, zImage, uImage...).
+    In gdb the developer specifies the connection parameters and
+    connects to kgdb.  Depending on which kgdb I/O modules exist in
+    the kernel for a given architecture, it may be possible to debug
+    the test machine's kernel with the development machine using a
+    rs232 or ethernet connection.
+    </para>
+  </chapter>
+  <chapter id="CompilingAKernel">
+    <title>Compiling a kernel</title>
+    <para>
+    To enable <symbol>CONFIG_KGDB</symbol>, look under the "Kernel debugging"
+    and then select "KGDB: kernel debugging with remote gdb".
+    </para>
+    <para>
+    Next you should choose one of more I/O drivers to interconnect debugging
+    host and debugged target.  Early boot debugging requires a KGDB
+    I/O driver that supports early debugging and the driver must be
+    built into the kernel directly. Kgdb I/O driver configuration
+    takes place via kernel or module parameters, see following
+    chapter.
+    </para>
+    <para>
+    The kgdb test compile options are described in the kgdb test suite chapter.
+    </para>
+
+  </chapter>
+  <chapter id="EnableKGDB">
+   <title>Enable kgdb for debugging</title>
+   <para>
+   In order to use kgdb you must activate it by passing configuration
+   information to one of the kgdb I/O drivers.  If you do not pass any
+   configuration information kgdb will not do anything at all.  Kgdb
+   will only actively hook up to the kernel trap hooks if a kgdb I/O
+   driver is loaded and configured.  If you unconfigure a kgdb I/O
+   driver, kgdb will unregister all the kernel hook points.
+   </para>
+   <para>
+   All drivers can be reconfigured at run time, if
+   <symbol>CONFIG_SYSFS</symbol> and <symbol>CONFIG_MODULES</symbol>
+   are enabled, by echo'ing a new config string to
+   <constant>/sys/module/&lt;driver&gt;/parameter/&lt;option&gt;</constant>.
+   The driver can be unconfigured by passing an empty string.  You cannot
+   change the configuration while the debugger is attached.  Make sure
+   to detach the debugger with the <constant>detach</constant> command
+   prior to trying unconfigure a kgdb I/O driver.
+   </para>
+   <sect1 id="kgdbwait">
+   <title>Kernel parameter: kgdbwait</title>
+   <para>
+   The Kernel command line option <constant>kgdbwait</constant> makes
+   kgdb wait for a debugger connection during booting of a kernel.  You
+   can only use this option you compiled a kgdb I/O driver into the
+   kernel and you specified the I/O driver configuration as a kernel
+   command line option.  The kgdbwait parameter should always follow the
+   configuration parameter for the kgdb I/O driver in the kernel
+   command line else the I/O driver will not be configured prior to
+   asking the kernel to use it to wait.
+   </para>
+   <para>
+   The kernel will stop and wait as early as the I/O driver and
+   architecture will allow when you use this option.  If you build the
+   kgdb I/O driver as a kernel module kgdbwait will not do anything.
+   </para>
+   </sect1>
+  <sect1 id="kgdboc">
+  <title>Kernel parameter: kgdboc</title>
+  <para>
+  The kgdboc driver was originally an abbreviation meant to stand for
+  "kgdb over console".  Kgdboc is designed to work with a single
+  serial port as example, and it was meant to cover the circumstance
+  where you wanted to use a serial console as your primary console as
+  well as using it to perform kernel debugging.
+  </para>
+  <sect2 id="UsingKgdboc">
+  <title>Using kgdboc</title>
+  <para>
+  You can configure kgdboc via sysfs or a module or kernel boot line
+  parameter depending on if you build with CONFIG_KGDBOC as a module
+  or built-in.
+  <orderedlist>
+  <listitem><para>From the module load or build-in</para>
+  <para><constant>kgdboc=&lt;tty-device&gt;,[baud]</constant></para>
+  <para>
+  The example here would be if your console port was typically ttyS0, you would use something like <constant>kgdboc=ttyS0,115200</constant> or on the ARM Versatile AB you would likely use <constant>kgdboc=ttyAMA0,115200</constant>
+  </para>
+  </listitem>
+  <listitem><para>From sysfs</para>
+  <para><constant>echo ttyS0 &gt; /sys/module/kgdboc/parameters/kgdboc</constant></para>
+  </listitem>
+  </orderedlist>
+  </para>
+  <para>
+  NOTE: Kgdboc does not support interrupting the target via the
+  gdb remote protocol.  You must manually send a sysrq-g unless you
+  have a proxy that splits console output to a terminal problem and
+  has a separate port for the debugger to connect to that sends the
+  sysrq-g for you.
+  </para>
+  <para>When using kgdboc with no debugger proxy, you can end up
+  connecting the debugger for one of two entry points.  If an
+  exception occurs after you have loaded kgdboc a message should print
+  on the console stating it is waiting for the debugger.  In case you
+  disconnect your terminal program and then connect the debugger in
+  its place.  If you want to interrupt the target system and forcibly
+  enter a debug session you have to issue a Sysrq sequence and then
+  type the letter <constant>g</constant>.  Then you disconnect the
+  terminal session and connect gdb.  Your options if you don't like
+  this are to hack gdb to send the sysrq-g for you as well as on the
+  initial connect, or to use a debugger proxy that allows an
+  unmodified gdb to do the debugging.
+  </para>
+  </sect2>
+  <sect2 id="kgdbocDesign">
+  <title>kgdboc internals</title>
+  <para>
+  The kgdboc driver is actually a very thin driver that relies on the
+  underlying low level to the hardware driver having "polling hooks"
+  which the to which the tty driver is attached.  In the initial
+  implementation of kgdboc it the serial_core was changed to expose a
+  low level uart hook for doing polled mode reading and writing of a
+  single character while in an atomic context.  When kgdb makes an I/O
+  request to the debugger, kgdboc invokes a call back in the serial
+  core which in turn uses the call back in the uart driver.  It is
+  certainly possible to extend kgdboc to work with non-uart based
+  consoles in the future.
+  </para>
+  <para>
+  When using kgdboc with a uart, the uart driver must implement two callbacks in the <constant>struct uart_ops</constant>. Example from drivers/8250.c:<programlisting>
+#ifdef CONFIG_CONSOLE_POLL
+	.poll_get_char = serial8250_get_poll_char,
+	.poll_put_char = serial8250_put_poll_char,
+#endif
+  </programlisting>
+  Any implementation specifics around creating a polling driver use the
+  <constant>#ifdef CONFIG_CONSOLE_POLL</constant>, as shown above.
+  Keep in mind that polling hooks have to be implemented in such a way
+  that they can be called from an atomic context and have to restore
+  the state of the uart chip on return such that the system can return
+  to normal when the debugger detaches.  You need to be very careful
+  with any kind of lock you consider, because failing here is most
+  going to mean pressing the reset button.
+  </para>
+  </sect2>
+  </sect1>
+  <sect1 id="kgdbcon">
+  <title>Kernel parameter: kgdbcon</title>
+  <para>
+  Kgdb supports using the gdb serial protocol to send console messages
+  to the debugger when the debugger is connected and running.  There
+  are two ways to activate this feature.
+  <orderedlist>
+  <listitem><para>Activate with the kernel command line option:</para>
+  <para><constant>kgdbcon</constant></para>
+  </listitem>
+  <listitem><para>Use sysfs before configuring an io driver</para>
+  <para>
+  <constant>echo 1 &gt; /sys/module/kgdb/parameters/kgdb_use_con</constant>
+  </para>
+  <para>
+  NOTE: If you do this after you configure the kgdb I/O driver, the
+  setting will not take effect until the next point the I/O is
+  reconfigured.
+  </para>
+  </listitem>
+  </orderedlist>
+  </para>
+  <para>
+  IMPORTANT NOTE: Using this option with kgdb over the console
+  (kgdboc) or kgdb over ethernet (kgdboe) is not supported.
+  </para>
+  </sect1>
+  </chapter>
+  <chapter id="ConnectingGDB">
+  <title>Connecting gdb</title>
+    <para>
+    If you are using kgdboc, you need to have used kgdbwait as a boot
+    argument, issued a sysrq-g, or the system you are going to debug
+    has already taken an exception and is waiting for the debugger to
+    attach before you can connect gdb.
+    </para>
+    <para>
+    If you are not using different kgdb I/O driver other than kgdboc,
+    you should be able to connect and the target will automatically
+    respond.
+    </para>
+    <para>
+    Example (using a serial port):
+    </para>
+    <programlisting>
+    % gdb ./vmlinux
+    (gdb) set remotebaud 115200
+    (gdb) target remote /dev/ttyS0
+    </programlisting>
+    <para>
+    Example (kgdb to a terminal server):
+    </para>
+    <programlisting>
+    % gdb ./vmlinux
+    (gdb) target remote udp:192.168.2.2:6443
+    </programlisting>
+    <para>
+    Example (kgdb over ethernet):
+    </para>
+    <programlisting>
+    % gdb ./vmlinux
+    (gdb) target remote udp:192.168.2.2:6443
+    </programlisting>
+    <para>
+    Once connected, you can debug a kernel the way you would debug an
+    application program.
+    </para>
+    <para>
+    If you are having problems connecting or something is going
+    seriously wrong while debugging, it will most often be the case
+    that you want to enable gdb to be verbose about its target
+    communications.  You do this prior to issuing the <constant>target
+    remote</constant> command by typing in: <constant>set remote debug 1</constant>
+    </para>
+  </chapter>
+  <chapter id="KGDBTestSuite">
+    <title>kgdb Test Suite</title>
+    <para>
+    When kgdb is enabled in the kernel config you can also elect to
+    enable the config parameter KGDB_TESTS.  Turning this on will
+    enable a special kgdb I/O module which is designed to test the
+    kgdb internal functions.
+    </para>
+    <para>
+    The kgdb tests are mainly intended for developers to test the kgdb
+    internals as well as a tool for developing a new kgdb architecture
+    specific implementation.  These tests are not really for end users
+    of the Linux kernel.  The primary source of documentation would be
+    to look in the drivers/misc/kgdbts.c file.
+    </para>
+    <para>
+    The kgdb test suite can also be configured at compile time to run
+    the core set of tests by setting the kernel config parameter
+    KGDB_TESTS_ON_BOOT.  This particular option is aimed at automated
+    regression testing and does not require modifying the kernel boot
+    config arguments.  If this is turned on, the kgdb test suite can
+    be disabled by specifying "kgdbts=" as a kernel boot argument.
+    </para>
+  </chapter>
+  <chapter id="CommonBackEndReq">
+    <title>Architecture Specifics</title>
+      <para>
+      Kgdb is organized into three basic components:
+      <orderedlist>
+      <listitem><para>kgdb core</para>
+      <para>
+      The kgdb core is found in kernel/kgdb.c.  It contains:
+      <itemizedlist>
+      <listitem><para>All the logic to implement the gdb serial protocol</para></listitem>
+      <listitem><para>A generic OS exception handler which includes sync'ing the processors into a stopped state on an multi cpu system.</para></listitem>
+      <listitem><para>The API to talk to the kgdb I/O drivers</para></listitem>
+      <listitem><para>The API to make calls to the arch specific kgdb implementation</para></listitem>
+      <listitem><para>The logic to perform safe memory reads and writes to memory while using the debugger</para></listitem>
+      <listitem><para>A full implementation for software breakpoints unless overridden by the arch</para></listitem>
+      </itemizedlist>
+      </para>
+      </listitem>
+      <listitem><para>kgdb arch specific implementation</para>
+      <para>
+      This implementation is generally found in arch/*/kernel/kgdb.c.
+      As an example, arch/x86/kernel/kgdb.c contains the specifics to
+      implement HW breakpoint as well as the initialization to
+      dynamically register and unregister for the trap handlers on
+      this architecture.  The arch specific portion implements:
+      <itemizedlist>
+      <listitem><para>contains an arch specific trap catcher which
+      invokes kgdb_handle_exception() to start kgdb about doing its
+      work</para></listitem>
+      <listitem><para>translation to and from gdb specific packet format to pt_regs</para></listitem>
+      <listitem><para>Registration and unregistration of architecture specific trap hooks</para></listitem>
+      <listitem><para>Any special exception handling and cleanup</para></listitem>
+      <listitem><para>NMI exception handling and cleanup</para></listitem>
+      <listitem><para>(optional)HW breakpoints</para></listitem>
+      </itemizedlist>
+      </para>
+      </listitem>
+      <listitem><para>kgdb I/O driver</para>
+      <para>
+      Each kgdb I/O driver has to provide an configuration
+      initialization, and cleanup handler for when it
+      unloads/unconfigures.  Any given kgdb I/O driver has to operate
+      very closely with the hardware and must do it in such a way that
+      does not enable interrupts or change other parts of the system
+      context without completely restoring them.  Every kgdb I/O
+      driver must provide a read and write character interface.  The
+      kgdb core will repeatedly "poll" a kgdb I/O driver for characters
+      when it needs input.  The I/O driver is expected to return
+      immediately if there is no data available.  Doing so allows for
+      the future possibility to touch watch dog hardware in such a way
+      as to have a target system not reset when these are enabled.
+      </para>
+      </listitem>
+      </orderedlist>
+      </para>
+      <para>
+      If you are intent on adding kgdb architecture specific support
+      for a new architecture, the architecture should define
+      <constant>HAVE_ARCH_KGDB</constant> in the architecture specific
+      Kconfig file.  This will enable kgdb for the architecture, and
+      at that point you must create an architecture specific kgdb
+      implementation.
+      </para>
+      <para>
+      There are a few flags which must be set on every architecture in
+      their &lt;asm/kgdb.h&gt; file.  These are:
+      <itemizedlist>
+        <listitem>
+	  <para>
+	  NUMREGBYTES: The size in bytes of all of the registers, so
+	  that we can ensure they will all fit into a packet.
+	  </para>
+	  <para>
+	  BUFMAX: The size in bytes of the buffer GDB will read into.
+	  This must be larger than NUMREGBYTES.
+	  </para>
+	  <para>
+	  CACHE_FLUSH_IS_SAFE: Set to 1 if it is always safe to call
+	  flush_cache_range or flush_icache_range.  On some architectures,
+	  these functions may not be safe to call on SMP since we keep other
+	  CPUs in a holding pattern.
+	  </para>
+	</listitem>
+      </itemizedlist>
+      </para>
+      <para>
+      There are also the following functions for the common backend,
+      found in kernel/kgdb.c, that must be supplied by the
+      architecture-specific backend unless marked as (optional), in
+      which case a default function maybe used if the architecture
+      does not need to provide a specific implementation.
+      </para>
+!Iinclude/linux/kgdb.h
+  </chapter>
+  <chapter id="credits">
+     <title>Credits</title>
+	<para>
+		The following people have contributed to this document:
+		<orderedlist>
+			<listitem><para>Amit Kale<email>amitkale@linsyssoft.com</email></para></listitem>
+			<listitem><para>Tom Rini<email>trini@kernel.crashing.org</email></para></listitem>
+			<listitem><para>Jason Wessel<email>jason.wessel@windriver.com</email></para></listitem>
+		</orderedlist>
+	</para>
+  </chapter>
+</book>
+
diff --git a/MAINTAINERS b/MAINTAINERS
index e46775868019..3eceebb48c92 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2319,6 +2319,12 @@ L:	linux-kernel@vger.kernel.org
 L:	kexec@lists.infradead.org
 S:	Maintained
 
+KGDB
+P:	Jason Wessel
+M:	jason.wessel@windriver.com
+L:	kgdb-bugreport@lists.sourceforge.net
+S:	Maintained
+
 KPROBES
 P:	Ananth N Mavinakayanahalli
 M:	ananth@in.ibm.com
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index b0985b79b638..9757b1a6d9dc 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -22,31 +22,34 @@
 
 struct pt_regs;
 
-/*
- *	kgdb_skipexception - Bail out of KGDB when we've been triggered.
+/**
+ *	kgdb_skipexception - (optional) exit kgdb_handle_exception early
  *	@exception: Exception vector number
  *	@regs: Current &struct pt_regs.
  *
- *	On some architectures we need to skip a breakpoint exception when
- *	it occurs after a breakpoint has been removed.
+ *	On some architectures it is required to skip a breakpoint
+ *	exception when it occurs after a breakpoint has been removed.
+ *	This can be implemented in the architecture specific portion of
+ *	for kgdb.
  */
 extern int kgdb_skipexception(int exception, struct pt_regs *regs);
 
-/*
- *	kgdb_post_primary_code - Save error vector/code numbers.
+/**
+ *	kgdb_post_primary_code - (optional) Save error vector/code numbers.
  *	@regs: Original pt_regs.
  *	@e_vector: Original error vector.
  *	@err_code: Original error code.
  *
- *	This is needed on architectures which support SMP and KGDB.
- *	This function is called after all the secondary cpus have been put
- *	to a know spin state and the primary CPU has control over KGDB.
+ *	This is usually needed on architectures which support SMP and
+ *	KGDB.  This function is called after all the secondary cpus have
+ *	been put to a know spin state and the primary CPU has control over
+ *	KGDB.
  */
 extern void kgdb_post_primary_code(struct pt_regs *regs, int e_vector,
 				  int err_code);
 
-/*
- *	kgdb_disable_hw_debug - Disable hardware debugging while we in kgdb.
+/**
+ *	kgdb_disable_hw_debug - (optional) Disable hardware debugging hook
  *	@regs: Current &struct pt_regs.
  *
  *	This function will be called if the particular architecture must
@@ -59,7 +62,14 @@ struct tasklet_struct;
 struct task_struct;
 struct uart_port;
 
-/* To enter the debugger explicitly. */
+/**
+ *	kgdb_breakpoint - compiled in breakpoint
+ *
+ *	This will be impelmented a static inline per architecture.  This
+ *	function is called by the kgdb core to execute an architecture
+ *	specific trap to cause kgdb to enter the exception processing.
+ *
+ */
 void kgdb_breakpoint(void);
 
 extern int kgdb_connected;
@@ -102,7 +112,7 @@ struct kgdb_bkpt {
  * Functions each KGDB-supporting architecture must provide:
  */
 
-/*
+/**
  *	kgdb_arch_init - Perform any architecture specific initalization.
  *
  *	This function will handle the initalization of any architecture
@@ -110,7 +120,7 @@ struct kgdb_bkpt {
  */
 extern int kgdb_arch_init(void);
 
-/*
+/**
  *	kgdb_arch_exit - Perform any architecture specific uninitalization.
  *
  *	This function will handle the uninitalization of any architecture
@@ -118,7 +128,7 @@ extern int kgdb_arch_init(void);
  */
 extern void kgdb_arch_exit(void);
 
-/*
+/**
  *	pt_regs_to_gdb_regs - Convert ptrace regs to GDB regs
  *	@gdb_regs: A pointer to hold the registers in the order GDB wants.
  *	@regs: The &struct pt_regs of the current process.
@@ -128,7 +138,7 @@ extern void kgdb_arch_exit(void);
  */
 extern void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs);
 
-/*
+/**
  *	sleeping_thread_to_gdb_regs - Convert ptrace regs to GDB regs
  *	@gdb_regs: A pointer to hold the registers in the order GDB wants.
  *	@p: The &struct task_struct of the desired process.
@@ -143,7 +153,7 @@ extern void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs);
 extern void
 sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p);
 
-/*
+/**
  *	gdb_regs_to_pt_regs - Convert GDB regs to ptrace regs.
  *	@gdb_regs: A pointer to hold the registers we've received from GDB.
  *	@regs: A pointer to a &struct pt_regs to hold these values in.
@@ -153,7 +163,7 @@ sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p);
  */
 extern void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs);
 
-/*
+/**
  *	kgdb_arch_handle_exception - Handle architecture specific GDB packets.
  *	@vector: The error vector of the exception that happened.
  *	@signo: The signal number of the exception that happened.
@@ -175,7 +185,7 @@ kgdb_arch_handle_exception(int vector, int signo, int err_code,
 			   char *remcom_out_buffer,
 			   struct pt_regs *regs);
 
-/*
+/**
  *	kgdb_roundup_cpus - Get other CPUs into a holding pattern
  *	@flags: Current IRQ state
  *
@@ -198,7 +208,7 @@ extern int kgdb_validate_break_address(unsigned long addr);
 extern int kgdb_arch_set_breakpoint(unsigned long addr, char *saved_instr);
 extern int kgdb_arch_remove_breakpoint(unsigned long addr, char *bundle);
 
-/*
+/**
  * struct kgdb_arch - Describe architecture specific values.
  * @gdb_bpt_instr: The instruction to trigger a breakpoint.
  * @flags: Flags for the breakpoint, currently just %KGDB_HW_BREAKPOINT.
@@ -227,7 +237,7 @@ struct kgdb_arch {
 	void	(*correct_hw_break)(void);
 };
 
-/*
+/**
  * struct kgdb_io - Describe the interface for an I/O driver to talk with KGDB.
  * @name: Name of the I/O driver.
  * @read_char: Pointer to a function that will return one char.
-- 
cgit v1.2.3


From cac1f3c8a80f3fc0b4489d1d3ba29214677ffab2 Mon Sep 17 00:00:00 2001
From: Paul Gortmaker <paul.gortmaker@windriver.com>
Date: Tue, 15 Apr 2008 12:49:21 -0400
Subject: phylib: factor out get_phy_id from within get_phy_device

We were already doing what amounts to a get_phy_id from within
get_phy_device, and rather than duplicate this for the TBIPA
probing, we might as well just factor it out and make it available
instead.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/phy/phy_device.c | 38 +++++++++++++++++++++++++++++---------
 include/linux/phy.h          |  1 +
 2 files changed, 30 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index f4c4fd85425f..8b1121b02f98 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -86,35 +86,55 @@ struct phy_device* phy_device_create(struct mii_bus *bus, int addr, int phy_id)
 EXPORT_SYMBOL(phy_device_create);
 
 /**
- * get_phy_device - reads the specified PHY device and returns its @phy_device struct
+ * get_phy_id - reads the specified addr for its ID.
  * @bus: the target MII bus
  * @addr: PHY address on the MII bus
+ * @phy_id: where to store the ID retrieved.
  *
  * Description: Reads the ID registers of the PHY at @addr on the
- *   @bus, then allocates and returns the phy_device to represent it.
+ *   @bus, stores it in @phy_id and returns zero on success.
  */
-struct phy_device * get_phy_device(struct mii_bus *bus, int addr)
+int get_phy_id(struct mii_bus *bus, int addr, u32 *phy_id)
 {
 	int phy_reg;
-	u32 phy_id;
-	struct phy_device *dev = NULL;
 
 	/* Grab the bits from PHYIR1, and put them
 	 * in the upper half */
 	phy_reg = bus->read(bus, addr, MII_PHYSID1);
 
 	if (phy_reg < 0)
-		return ERR_PTR(phy_reg);
+		return -EIO;
 
-	phy_id = (phy_reg & 0xffff) << 16;
+	*phy_id = (phy_reg & 0xffff) << 16;
 
 	/* Grab the bits from PHYIR2, and put them in the lower half */
 	phy_reg = bus->read(bus, addr, MII_PHYSID2);
 
 	if (phy_reg < 0)
-		return ERR_PTR(phy_reg);
+		return -EIO;
+
+	*phy_id |= (phy_reg & 0xffff);
+
+	return 0;
+}
+
+/**
+ * get_phy_device - reads the specified PHY device and returns its @phy_device struct
+ * @bus: the target MII bus
+ * @addr: PHY address on the MII bus
+ *
+ * Description: Reads the ID registers of the PHY at @addr on the
+ *   @bus, then allocates and returns the phy_device to represent it.
+ */
+struct phy_device * get_phy_device(struct mii_bus *bus, int addr)
+{
+	struct phy_device *dev = NULL;
+	u32 phy_id;
+	int r;
 
-	phy_id |= (phy_reg & 0xffff);
+	r = get_phy_id(bus, addr, &phy_id);
+	if (r)
+		return ERR_PTR(r);
 
 	/* If the phy_id is all Fs, there is no device there */
 	if (0xffffffff == phy_id)
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 2d838448415c..779cbcd65f62 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -381,6 +381,7 @@ struct phy_driver {
 
 int phy_read(struct phy_device *phydev, u16 regnum);
 int phy_write(struct phy_device *phydev, u16 regnum, u16 val);
+int get_phy_id(struct mii_bus *bus, int addr, u32 *phy_id);
 struct phy_device* get_phy_device(struct mii_bus *bus, int addr);
 int phy_clear_interrupt(struct phy_device *phydev);
 int phy_config_interrupt(struct phy_device *phydev, u32 interrupts);
-- 
cgit v1.2.3


From cf48062658e7ab3bc55e10c65676c3c73c16f8bf Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 24 Jan 2008 00:05:14 +0900
Subject: libata: prefer hardreset

When both soft and hard resets are available, libata preferred
softreset till now.  The logic behind it was to be softer to devices;
however, this doesn't really help much.  Rationales for the change:

* BIOS may freeze lock certain things during boot and softreset can't
  unlock those.  This by itself is okay but during operation PHY event
  or other error conditions can trigger hardreset and the device may
  end up with different configuration.

  For example, after a hardreset, previously unlockable HPA can be
  unlocked resulting in different device size and thus revalidation
  failure.  Similar condition can occur during or after resume.

* Certain ATAPI devices require hardreset to recover after certain
  error conditions.  On PATA, this is done by issuing the DEVICE RESET
  command.  On SATA, COMRESET has equivalent effect.  The problem is
  that DEVICE RESET needs its own execution protocol.

  For SFF controllers with bare TF access, it can be easily
  implemented but more advanced controllers (e.g. ahci and sata_sil24)
  require specialized implementations.  Simply using hardreset solves
  the problem nicely.

* COMRESET initialization sequence is the norm in SATA land and many
  SATA devices don't work properly if only SRST is used.  For example,
  some PMPs behave this way and libata works around by always issuing
  hardreset if the host supports PMP.

  Like the above example, libata has developed a number of mechanisms
  aiming to promote softreset to hardreset if softreset is not going
  to work.  This approach is time consuming and error prone.

  Also, note that, dependingon how you read the specs, it could be
  argued that PMP fan-out ports require COMRESET to start operation.
  In fact, all the PMPs on the market except one don't work properly
  if COMRESET is not issued to fan-out ports after PMP reset.

* COMRESET is an integral part of SATA connection and any working
  device should be able to handle COMRESET properly.  After all, it's
  the way to signal hardreset during reboot.  This is the most used
  and recommended (at least by the ahci spec) method of resetting
  devices.

So, this patch makes libata prefer hardreset over softreset by making
the following changes.

* Rename ATA_EH_RESET_MASK to ATA_EH_RESET and use it whereever
  ATA_EH_{SOFT|HARD}RESET used to be used.  ATA_EH_{SOFT|HARD}RESET is
  now only used to tell prereset whether soft or hard reset will be
  issued.

* Strip out now unneeded promote-to-hardreset logics from
  ata_eh_reset(), ata_std_prereset(), sata_pmp_std_prereset() and
  other places.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c        | 10 ++---
 drivers/ata/libata-core.c | 19 ++-------
 drivers/ata/libata-eh.c   | 99 +++++++++++++++++------------------------------
 drivers/ata/libata-pmp.c  | 28 ++++----------
 drivers/ata/libata-scsi.c |  4 +-
 drivers/ata/sata_fsl.c    |  2 +-
 drivers/ata/sata_mv.c     | 31 ++++-----------
 drivers/ata/sata_nv.c     | 12 +++---
 drivers/ata/sata_sil24.c  | 54 +++++++++++++-------------
 drivers/ata/sata_via.c    |  2 +-
 include/linux/libata.h    |  8 ++--
 11 files changed, 101 insertions(+), 168 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index b1eb4e24c86a..f6bbd52b1547 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1663,7 +1663,7 @@ static void ahci_error_intr(struct ata_port *ap, u32 irq_stat)
 		u32 *unk = (u32 *)(pp->rx_fis + RX_FIS_UNK);
 
 		active_ehi->err_mask |= AC_ERR_HSM;
-		active_ehi->action |= ATA_EH_SOFTRESET;
+		active_ehi->action |= ATA_EH_RESET;
 		ata_ehi_push_desc(active_ehi,
 				  "unknown FIS %08x %08x %08x %08x" ,
 				  unk[0], unk[1], unk[2], unk[3]);
@@ -1671,19 +1671,19 @@ static void ahci_error_intr(struct ata_port *ap, u32 irq_stat)
 
 	if (ap->nr_pmp_links && (irq_stat & PORT_IRQ_BAD_PMP)) {
 		active_ehi->err_mask |= AC_ERR_HSM;
-		active_ehi->action |= ATA_EH_SOFTRESET;
+		active_ehi->action |= ATA_EH_RESET;
 		ata_ehi_push_desc(active_ehi, "incorrect PMP");
 	}
 
 	if (irq_stat & (PORT_IRQ_HBUS_ERR | PORT_IRQ_HBUS_DATA_ERR)) {
 		host_ehi->err_mask |= AC_ERR_HOST_BUS;
-		host_ehi->action |= ATA_EH_SOFTRESET;
+		host_ehi->action |= ATA_EH_RESET;
 		ata_ehi_push_desc(host_ehi, "host bus error");
 	}
 
 	if (irq_stat & PORT_IRQ_IF_ERR) {
 		host_ehi->err_mask |= AC_ERR_ATA_BUS;
-		host_ehi->action |= ATA_EH_SOFTRESET;
+		host_ehi->action |= ATA_EH_RESET;
 		ata_ehi_push_desc(host_ehi, "interface fatal error");
 	}
 
@@ -1771,7 +1771,7 @@ static void ahci_port_intr(struct ata_port *ap)
 	/* while resetting, invalid completions are expected */
 	if (unlikely(rc < 0 && !resetting)) {
 		ehi->err_mask |= AC_ERR_HSM;
-		ehi->action |= ATA_EH_SOFTRESET;
+		ehi->action |= ATA_EH_RESET;
 		ata_port_freeze(ap);
 	}
 }
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index be95fdb69726..02e7ba43a3b3 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3949,17 +3949,6 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
 	const unsigned long *timing = sata_ehc_deb_timing(ehc);
 	int rc;
 
-	/* handle link resume */
-	if ((ehc->i.flags & ATA_EHI_RESUME_LINK) &&
-	    (link->flags & ATA_LFLAG_HRST_TO_RESUME))
-		ehc->i.action |= ATA_EH_HARDRESET;
-
-	/* Some PMPs don't work with only SRST, force hardreset if PMP
-	 * is supported.
-	 */
-	if (ap->flags & ATA_FLAG_PMP)
-		ehc->i.action |= ATA_EH_HARDRESET;
-
 	/* if we're about to do hardreset, nothing more to do */
 	if (ehc->i.action & ATA_EH_HARDRESET)
 		return 0;
@@ -6055,9 +6044,9 @@ void ata_qc_issue(struct ata_queued_cmd *qc)
 		if (ata_sg_setup(qc))
 			goto sg_err;
 
-	/* if device is sleeping, schedule softreset and abort the link */
+	/* if device is sleeping, schedule reset and abort the link */
 	if (unlikely(qc->dev->flags & ATA_DFLAG_SLEEPING)) {
-		link->eh_info.action |= ATA_EH_SOFTRESET;
+		link->eh_info.action |= ATA_EH_RESET;
 		ata_ehi_push_desc(&link->eh_info, "waking up from sleep");
 		ata_link_abort(link);
 		return;
@@ -6634,7 +6623,7 @@ int ata_host_suspend(struct ata_host *host, pm_message_t mesg)
  */
 void ata_host_resume(struct ata_host *host)
 {
-	ata_host_request_pm(host, PMSG_ON, ATA_EH_SOFTRESET,
+	ata_host_request_pm(host, PMSG_ON, ATA_EH_RESET,
 			    ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 0);
 	host->dev->power.power_state = PMSG_ON;
 
@@ -7171,7 +7160,7 @@ int ata_host_register(struct ata_host *host, struct scsi_host_template *sht)
 
 			ehi->probe_mask =
 				(1 << ata_link_max_devices(&ap->link)) - 1;
-			ehi->action |= ATA_EH_SOFTRESET;
+			ehi->action |= ATA_EH_RESET;
 			ehi->flags |= ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET;
 
 			ap->pflags &= ~ATA_PFLAG_INITIALIZING;
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index a5830329eda4..f7cae6400155 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1079,16 +1079,9 @@ void ata_eh_about_to_do(struct ata_link *link, struct ata_device *dev,
 
 	spin_lock_irqsave(ap->lock, flags);
 
-	/* Reset is represented by combination of actions and EHI
-	 * flags.  Suck in all related bits before clearing eh_info to
-	 * avoid losing requested action.
-	 */
-	if (action & ATA_EH_RESET_MASK) {
-		ehc->i.action |= ehi->action & ATA_EH_RESET_MASK;
+	/* suck in and clear reset modifier */
+	if (action & ATA_EH_RESET) {
 		ehc->i.flags |= ehi->flags & ATA_EHI_RESET_MODIFIER_MASK;
-
-		/* make sure all reset actions are cleared & clear EHI flags */
-		action |= ATA_EH_RESET_MASK;
 		ehi->flags &= ~ATA_EHI_RESET_MODIFIER_MASK;
 	}
 
@@ -1117,11 +1110,9 @@ void ata_eh_done(struct ata_link *link, struct ata_device *dev,
 {
 	struct ata_eh_context *ehc = &link->eh_context;
 
-	/* if reset is complete, clear all reset actions & reset modifier */
-	if (action & ATA_EH_RESET_MASK) {
-		action |= ATA_EH_RESET_MASK;
+	/* if reset is complete, clear reset modifier */
+	if (action & ATA_EH_RESET)
 		ehc->i.flags &= ~ATA_EHI_RESET_MODIFIER_MASK;
-	}
 
 	ata_eh_clear_action(link, dev, &ehc->i, action);
 }
@@ -1329,20 +1320,20 @@ static void ata_eh_analyze_serror(struct ata_link *link)
 
 	if (serror & SERR_PERSISTENT) {
 		err_mask |= AC_ERR_ATA_BUS;
-		action |= ATA_EH_HARDRESET;
+		action |= ATA_EH_RESET;
 	}
 	if (serror &
 	    (SERR_DATA_RECOVERED | SERR_COMM_RECOVERED | SERR_DATA)) {
 		err_mask |= AC_ERR_ATA_BUS;
-		action |= ATA_EH_SOFTRESET;
+		action |= ATA_EH_RESET;
 	}
 	if (serror & SERR_PROTOCOL) {
 		err_mask |= AC_ERR_HSM;
-		action |= ATA_EH_SOFTRESET;
+		action |= ATA_EH_RESET;
 	}
 	if (serror & SERR_INTERNAL) {
 		err_mask |= AC_ERR_SYSTEM;
-		action |= ATA_EH_HARDRESET;
+		action |= ATA_EH_RESET;
 	}
 
 	/* Determine whether a hotplug event has occurred.  Both
@@ -1448,7 +1439,7 @@ static unsigned int ata_eh_analyze_tf(struct ata_queued_cmd *qc,
 
 	if ((stat & (ATA_BUSY | ATA_DRQ | ATA_DRDY)) != ATA_DRDY) {
 		qc->err_mask |= AC_ERR_HSM;
-		return ATA_EH_SOFTRESET;
+		return ATA_EH_RESET;
 	}
 
 	if (stat & (ATA_ERR | ATA_DF))
@@ -1484,7 +1475,7 @@ static unsigned int ata_eh_analyze_tf(struct ata_queued_cmd *qc,
 	}
 
 	if (qc->err_mask & (AC_ERR_HSM | AC_ERR_TIMEOUT | AC_ERR_ATA_BUS))
-		action |= ATA_EH_SOFTRESET;
+		action |= ATA_EH_RESET;
 
 	return action;
 }
@@ -1685,7 +1676,7 @@ static unsigned int ata_eh_speed_down(struct ata_device *dev,
 	if (verdict & ATA_EH_SPDN_SPEED_DOWN) {
 		/* speed down SATA link speed if possible */
 		if (sata_down_spd_limit(link) == 0) {
-			action |= ATA_EH_HARDRESET;
+			action |= ATA_EH_RESET;
 			goto done;
 		}
 
@@ -1705,7 +1696,7 @@ static unsigned int ata_eh_speed_down(struct ata_device *dev,
 			dev->spdn_cnt++;
 
 			if (ata_down_xfermask_limit(dev, sel) == 0) {
-				action |= ATA_EH_SOFTRESET;
+				action |= ATA_EH_RESET;
 				goto done;
 			}
 		}
@@ -1719,7 +1710,7 @@ static unsigned int ata_eh_speed_down(struct ata_device *dev,
 	    (dev->xfer_shift != ATA_SHIFT_PIO)) {
 		if (ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO) == 0) {
 			dev->spdn_cnt = 0;
-			action |= ATA_EH_SOFTRESET;
+			action |= ATA_EH_RESET;
 			goto done;
 		}
 	}
@@ -1764,9 +1755,9 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 		ehc->i.serror |= serror;
 		ata_eh_analyze_serror(link);
 	} else if (rc != -EOPNOTSUPP) {
-		/* SError read failed, force hardreset and probing */
+		/* SError read failed, force reset and probing */
 		ata_ehi_schedule_probe(&ehc->i);
-		ehc->i.action |= ATA_EH_HARDRESET;
+		ehc->i.action |= ATA_EH_RESET;
 		ehc->i.err_mask |= AC_ERR_OTHER;
 	}
 
@@ -1814,7 +1805,7 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 	/* enforce default EH actions */
 	if (ap->pflags & ATA_PFLAG_FROZEN ||
 	    all_err_mask & (AC_ERR_HSM | AC_ERR_TIMEOUT))
-		ehc->i.action |= ATA_EH_SOFTRESET;
+		ehc->i.action |= ATA_EH_RESET;
 	else if (((eflags & ATA_EFLAG_IS_IO) && all_err_mask) ||
 		 (!(eflags & ATA_EFLAG_IS_IO) && (all_err_mask & ~AC_ERR_DEV)))
 		ehc->i.action |= ATA_EH_REVALIDATE;
@@ -2118,7 +2109,6 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	int try = 0;
 	struct ata_device *dev;
 	unsigned long deadline, now;
-	unsigned int tmp_action;
 	ata_reset_fn_t reset;
 	unsigned long flags;
 	u32 sstatus;
@@ -2129,7 +2119,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	ap->pflags |= ATA_PFLAG_RESETTING;
 	spin_unlock_irqrestore(ap->lock, flags);
 
-	ata_eh_about_to_do(link, NULL, ehc->i.action & ATA_EH_RESET_MASK);
+	ata_eh_about_to_do(link, NULL, ATA_EH_RESET);
 
 	ata_link_for_each_dev(dev, link) {
 		/* If we issue an SRST then an ATA drive (not ATAPI)
@@ -2159,17 +2149,15 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		goto done;
 	}
 
-	/* Determine which reset to use and record in ehc->i.action.
-	 * prereset() may examine and modify it.
-	 */
-	if (softreset && (!hardreset || (!(lflags & ATA_LFLAG_NO_SRST) &&
-					 !sata_set_spd_needed(link) &&
-					 !(ehc->i.action & ATA_EH_HARDRESET))))
-		tmp_action = ATA_EH_SOFTRESET;
-	else
-		tmp_action = ATA_EH_HARDRESET;
-
-	ehc->i.action = (ehc->i.action & ~ATA_EH_RESET_MASK) | tmp_action;
+	/* prefer hardreset */
+	ehc->i.action &= ~ATA_EH_RESET;
+	if (hardreset) {
+		reset = hardreset;
+		ehc->i.action = ATA_EH_HARDRESET;
+	} else {
+		reset = softreset;
+		ehc->i.action = ATA_EH_SOFTRESET;
+	}
 
 	if (prereset) {
 		rc = prereset(link, jiffies + ATA_EH_PRERESET_TIMEOUT);
@@ -2177,7 +2165,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 			if (rc == -ENOENT) {
 				ata_link_printk(link, KERN_DEBUG,
 						"port disabled. ignoring.\n");
-				ehc->i.action &= ~ATA_EH_RESET_MASK;
+				ehc->i.action &= ~ATA_EH_RESET;
 
 				ata_link_for_each_dev(dev, link)
 					classes[dev->devno] = ATA_DEV_NONE;
@@ -2190,12 +2178,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		}
 	}
 
-	/* prereset() might have modified ehc->i.action */
-	if (ehc->i.action & ATA_EH_HARDRESET)
-		reset = hardreset;
-	else if (ehc->i.action & ATA_EH_SOFTRESET)
-		reset = softreset;
-	else {
+	/* prereset() might have cleared ATA_EH_RESET */
+	if (!(ehc->i.action & ATA_EH_RESET)) {
 		/* prereset told us not to reset, bang classes and return */
 		ata_link_for_each_dev(dev, link)
 			classes[dev->devno] = ATA_DEV_NONE;
@@ -2203,14 +2187,6 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		goto out;
 	}
 
-	/* did prereset() screw up?  if so, fix up to avoid oopsing */
-	if (!reset) {
-		if (softreset)
-			reset = softreset;
-		else
-			reset = hardreset;
-	}
-
  retry:
 	deadline = jiffies + ata_eh_reset_timeouts[try++];
 
@@ -2240,7 +2216,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 			goto fail;
 		}
 
-		ata_eh_about_to_do(link, NULL, ATA_EH_RESET_MASK);
+		ata_eh_about_to_do(link, NULL, ATA_EH_RESET);
 		rc = ata_do_reset(link, reset, classes, deadline);
 	}
 
@@ -2290,7 +2266,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		postreset(link, classes);
 
 	/* reset successful, schedule revalidation */
-	ata_eh_done(link, NULL, ehc->i.action & ATA_EH_RESET_MASK);
+	ata_eh_done(link, NULL, ATA_EH_RESET);
 	ehc->i.action |= ATA_EH_REVALIDATE;
 
 	rc = 0;
@@ -2548,7 +2524,7 @@ static int ata_eh_schedule_probe(struct ata_device *dev)
 	ata_eh_detach_dev(dev);
 	ata_dev_init(dev);
 	ehc->did_probe_mask |= (1 << dev->devno);
-	ehc->i.action |= ATA_EH_SOFTRESET;
+	ehc->i.action |= ATA_EH_RESET;
 	ehc->saved_xfer_mode[dev->devno] = 0;
 	ehc->saved_ncq_enabled &= ~(1 << dev->devno);
 
@@ -2592,12 +2568,7 @@ static int ata_eh_handle_dev_fail(struct ata_device *dev, int err)
 
 		return 1;
 	} else {
-		/* soft didn't work?  be haaaaard */
-		if (ehc->i.flags & ATA_EHI_DID_RESET)
-			ehc->i.action |= ATA_EH_HARDRESET;
-		else
-			ehc->i.action |= ATA_EH_SOFTRESET;
-
+		ehc->i.action |= ATA_EH_RESET;
 		return 0;
 	}
 }
@@ -2690,7 +2661,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 			ehc->i.action = 0;
 
 		/* do we need to reset? */
-		if (ehc->i.action & ATA_EH_RESET_MASK)
+		if (ehc->i.action & ATA_EH_RESET)
 			reset = 1;
 
 		ata_link_for_each_dev(dev, link)
@@ -2708,7 +2679,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 		ata_port_for_each_link(link, ap) {
 			struct ata_eh_context *ehc = &link->eh_context;
 
-			if (!(ehc->i.action & ATA_EH_RESET_MASK))
+			if (!(ehc->i.action & ATA_EH_RESET))
 				continue;
 
 			rc = ata_eh_reset(link, ata_link_nr_vacant(link),
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index d91f5090ba9d..8439fc8efdd6 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -194,15 +194,6 @@ int sata_pmp_std_prereset(struct ata_link *link, unsigned long deadline)
 	const unsigned long *timing = sata_ehc_deb_timing(ehc);
 	int rc;
 
-	/* force HRST? */
-	if (link->flags & ATA_LFLAG_NO_SRST)
-		ehc->i.action |= ATA_EH_HARDRESET;
-
-	/* handle link resume */
-	if ((ehc->i.flags & ATA_EHI_RESUME_LINK) &&
-	    (link->flags & ATA_LFLAG_HRST_TO_RESUME))
-		ehc->i.action |= ATA_EH_HARDRESET;
-
 	/* if we're about to do hardreset, nothing more to do */
 	if (ehc->i.action & ATA_EH_HARDRESET)
 		return 0;
@@ -445,7 +436,7 @@ static int sata_pmp_init_links(struct ata_port *ap, int nr_ports)
 
 		link->flags = 0;
 		ehc->i.probe_mask |= 1;
-		ehc->i.action |= ATA_EH_SOFTRESET;
+		ehc->i.action |= ATA_EH_RESET;
 		ehc->i.flags |= ATA_EHI_RESUME_LINK;
 	}
 
@@ -840,13 +831,12 @@ static int sata_pmp_eh_recover_pmp(struct ata_port *ap,
  retry:
 	ehc->classes[0] = ATA_DEV_UNKNOWN;
 
-	if (ehc->i.action & ATA_EH_RESET_MASK) {
+	if (ehc->i.action & ATA_EH_RESET) {
 		struct ata_link *tlink;
 
 		ata_eh_freeze_port(ap);
 
 		/* reset */
-		ehc->i.action = ATA_EH_HARDRESET;
 		rc = ata_eh_reset(link, 0, prereset, softreset, hardreset,
 				  postreset);
 		if (rc) {
@@ -890,11 +880,11 @@ static int sata_pmp_eh_recover_pmp(struct ata_port *ap,
 				reval_failed = 1;
 
 			ata_dev_printk(dev, KERN_WARNING,
-				       "retrying hardreset%s\n",
+				       "retrying reset%s\n",
 				       sleep ? " in 5 secs" : "");
 			if (sleep)
 				ssleep(5);
-			ehc->i.action |= ATA_EH_HARDRESET;
+			ehc->i.action |= ATA_EH_RESET;
 			goto retry;
 		} else {
 			ata_dev_printk(dev, KERN_ERR, "failed to recover PMP "
@@ -938,10 +928,8 @@ static int sata_pmp_eh_handle_disabled_links(struct ata_port *ap)
 		/* Some PMPs require hardreset sequence to get
 		 * SError.N working.
 		 */
-		if ((link->flags & ATA_LFLAG_HRST_TO_RESUME) &&
-		    (link->eh_context.i.flags & ATA_EHI_RESUME_LINK))
-			sata_link_hardreset(link, sata_deb_timing_normal,
-					    jiffies + ATA_TMOUT_INTERNAL_QUICK);
+		sata_link_hardreset(link, sata_deb_timing_normal,
+				    jiffies + ATA_TMOUT_INTERNAL_QUICK);
 
 		/* unconditionally clear SError.N */
 		rc = sata_scr_write(link, SCR_ERROR, SERR_PHYRDY_CHG);
@@ -1124,7 +1112,7 @@ static int sata_pmp_eh_recover(struct ata_port *ap,
 
  link_fail:
 	if (sata_pmp_handle_link_fail(link, link_tries)) {
-		pmp_ehc->i.action |= ATA_EH_HARDRESET;
+		pmp_ehc->i.action |= ATA_EH_RESET;
 		goto retry;
 	}
 
@@ -1142,7 +1130,7 @@ static int sata_pmp_eh_recover(struct ata_port *ap,
 	if (--pmp_tries) {
 		ata_port_printk(ap, KERN_WARNING,
 				"failed to recover PMP, retrying in 5 secs\n");
-		pmp_ehc->i.action |= ATA_EH_HARDRESET;
+		pmp_ehc->i.action |= ATA_EH_RESET;
 		ssleep(5);
 		goto retry;
 	}
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 15795394b0a8..1f036a7b14f3 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -3508,7 +3508,7 @@ static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
 		ata_port_for_each_link(link, ap) {
 			struct ata_eh_info *ehi = &link->eh_info;
 			ehi->probe_mask |= (1 << ata_link_max_devices(link)) - 1;
-			ehi->action |= ATA_EH_SOFTRESET;
+			ehi->action |= ATA_EH_RESET;
 		}
 	} else {
 		struct ata_device *dev = ata_find_dev(ap, devno);
@@ -3516,7 +3516,7 @@ static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
 		if (dev) {
 			struct ata_eh_info *ehi = &dev->link->eh_info;
 			ehi->probe_mask |= 1 << dev->devno;
-			ehi->action |= ATA_EH_SOFTRESET;
+			ehi->action |= ATA_EH_RESET;
 			ehi->flags |= ATA_EHI_RESUME_LINK;
 		} else
 			rc = -EINVAL;
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 9d1e3cad4aa9..fbd423ad23bb 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -996,7 +996,7 @@ static void sata_fsl_error_intr(struct ata_port *ap)
 	/* handle fatal errors */
 	if (hstatus & FATAL_ERROR_DECODE) {
 		err_mask |= AC_ERR_ATA_BUS;
-		action |= ATA_EH_SOFTRESET;
+		action |= ATA_EH_RESET;
 		/* how will fatal error interrupts be completed ?? */
 		freeze = 1;
 	}
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index 6ebebde8454a..a4944c8ad46d 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -1524,14 +1524,14 @@ static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc)
 			EDMA_ERR_CRQB_PAR | EDMA_ERR_CRPB_PAR |
 			EDMA_ERR_INTRL_PAR)) {
 		err_mask |= AC_ERR_ATA_BUS;
-		action |= ATA_EH_HARDRESET;
+		action |= ATA_EH_RESET;
 		ata_ehi_push_desc(ehi, "parity error");
 	}
 	if (edma_err_cause & (EDMA_ERR_DEV_DCON | EDMA_ERR_DEV_CON)) {
 		ata_ehi_hotplugged(ehi);
 		ata_ehi_push_desc(ehi, edma_err_cause & EDMA_ERR_DEV_DCON ?
 			"dev disconnect" : "dev connect");
-		action |= ATA_EH_HARDRESET;
+		action |= ATA_EH_RESET;
 	}
 
 	if (IS_GEN_I(hpriv)) {
@@ -1555,7 +1555,7 @@ static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc)
 			sata_scr_read(&ap->link, SCR_ERROR, &serr);
 			sata_scr_write_flush(&ap->link, SCR_ERROR, serr);
 			err_mask = AC_ERR_ATA_BUS;
-			action |= ATA_EH_HARDRESET;
+			action |= ATA_EH_RESET;
 		}
 	}
 
@@ -1564,7 +1564,7 @@ static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc)
 
 	if (!err_mask) {
 		err_mask = AC_ERR_OTHER;
-		action |= ATA_EH_HARDRESET;
+		action |= ATA_EH_RESET;
 	}
 
 	ehi->serror |= serr;
@@ -1780,7 +1780,7 @@ static void mv_pci_error(struct ata_host *host, void __iomem *mmio)
 				ata_ehi_push_desc(ehi,
 					"PCI err cause 0x%08x", err_cause);
 			err_mask = AC_ERR_HOST_BUS;
-			ehi->action = ATA_EH_HARDRESET;
+			ehi->action = ATA_EH_RESET;
 			qc = ata_qc_from_tag(ap, ap->link.active_tag);
 			if (qc)
 				qc->err_mask |= err_mask;
@@ -2449,28 +2449,13 @@ static int mv_prereset(struct ata_link *link, unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
 	struct mv_port_priv *pp	= ap->private_data;
-	struct ata_eh_context *ehc = &link->eh_context;
-	int rc;
 
-	rc = mv_stop_dma(ap);
-	if (rc)
-		ehc->i.action |= ATA_EH_HARDRESET;
+	mv_stop_dma(ap);
 
-	if (!(pp->pp_flags & MV_PP_FLAG_HAD_A_RESET)) {
+	if (!(pp->pp_flags & MV_PP_FLAG_HAD_A_RESET))
 		pp->pp_flags |= MV_PP_FLAG_HAD_A_RESET;
-		ehc->i.action |= ATA_EH_HARDRESET;
-	}
-
-	/* if we're about to do hardreset, nothing more to do */
-	if (ehc->i.action & ATA_EH_HARDRESET)
-		return 0;
-
-	if (ata_link_online(link))
-		rc = ata_wait_ready(ap, deadline);
-	else
-		rc = -ENODEV;
 
-	return rc;
+	return 0;
 }
 
 static int mv_hardreset(struct ata_link *link, unsigned int *class,
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index ed5473bf7a0a..ce02e15c857c 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -929,7 +929,7 @@ static int nv_adma_check_cpb(struct ata_port *ap, int cpb_num, int force_err)
 					"notifier for tag %d with no cmd?\n",
 					cpb_num);
 			ehi->err_mask |= AC_ERR_HSM;
-			ehi->action |= ATA_EH_SOFTRESET;
+			ehi->action |= ATA_EH_RESET;
 			ata_port_freeze(ap);
 			return 1;
 		}
@@ -1892,7 +1892,7 @@ static void nv_swncq_error_handler(struct ata_port *ap)
 
 	if (ap->link.sactive) {
 		nv_swncq_ncq_stop(ap);
-		ehc->i.action |= ATA_EH_HARDRESET;
+		ehc->i.action |= ATA_EH_RESET;
 	}
 
 	ata_bmdma_drive_eh(ap, ata_std_prereset, ata_std_softreset,
@@ -2173,7 +2173,7 @@ static int nv_swncq_sdbfis(struct ata_port *ap)
 		ata_ehi_clear_desc(ehi);
 		ata_ehi_push_desc(ehi, "BMDMA stat 0x%x", host_stat);
 		ehi->err_mask |= AC_ERR_HOST_BUS;
-		ehi->action |= ATA_EH_SOFTRESET;
+		ehi->action |= ATA_EH_RESET;
 		return -EINVAL;
 	}
 
@@ -2188,7 +2188,7 @@ static int nv_swncq_sdbfis(struct ata_port *ap)
 		ata_ehi_push_desc(ehi, "illegal SWNCQ:qc_active transition"
 				  "(%08x->%08x)", pp->qc_active, sactive);
 		ehi->err_mask |= AC_ERR_HSM;
-		ehi->action |= ATA_EH_HARDRESET;
+		ehi->action |= ATA_EH_RESET;
 		return -EINVAL;
 	}
 	for (i = 0; i < ATA_MAX_QUEUE; i++) {
@@ -2324,7 +2324,7 @@ static void nv_swncq_host_interrupt(struct ata_port *ap, u16 fis)
 		ata_ehi_push_desc(ehi, "Ata error. fis:0x%X", fis);
 		ehi->err_mask |= AC_ERR_DEV;
 		ehi->serror |= serror;
-		ehi->action |= ATA_EH_SOFTRESET;
+		ehi->action |= ATA_EH_RESET;
 		ata_port_freeze(ap);
 		return;
 	}
@@ -2356,7 +2356,7 @@ static void nv_swncq_host_interrupt(struct ata_port *ap, u16 fis)
 		if (pp->ncq_flags & (ncq_saw_sdb | ncq_saw_backout)) {
 			ata_ehi_push_desc(ehi, "illegal fis transaction");
 			ehi->err_mask |= AC_ERR_HSM;
-			ehi->action |= ATA_EH_HARDRESET;
+			ehi->action |= ATA_EH_RESET;
 			goto irq_error;
 		}
 
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index df7988df7908..aa8d0323c9bb 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -286,45 +286,45 @@ static struct sil24_cerr_info {
 				    "device error via D2H FIS" },
 	[PORT_CERR_SDB]		= { AC_ERR_DEV, 0,
 				    "device error via SDB FIS" },
-	[PORT_CERR_DATA]	= { AC_ERR_ATA_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_DATA]	= { AC_ERR_ATA_BUS, ATA_EH_RESET,
 				    "error in data FIS" },
-	[PORT_CERR_SEND]	= { AC_ERR_ATA_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_SEND]	= { AC_ERR_ATA_BUS, ATA_EH_RESET,
 				    "failed to transmit command FIS" },
-	[PORT_CERR_INCONSISTENT] = { AC_ERR_HSM, ATA_EH_SOFTRESET,
+	[PORT_CERR_INCONSISTENT] = { AC_ERR_HSM, ATA_EH_RESET,
 				     "protocol mismatch" },
-	[PORT_CERR_DIRECTION]	= { AC_ERR_HSM, ATA_EH_SOFTRESET,
+	[PORT_CERR_DIRECTION]	= { AC_ERR_HSM, ATA_EH_RESET,
 				    "data directon mismatch" },
-	[PORT_CERR_UNDERRUN]	= { AC_ERR_HSM, ATA_EH_SOFTRESET,
+	[PORT_CERR_UNDERRUN]	= { AC_ERR_HSM, ATA_EH_RESET,
 				    "ran out of SGEs while writing" },
-	[PORT_CERR_OVERRUN]	= { AC_ERR_HSM, ATA_EH_SOFTRESET,
+	[PORT_CERR_OVERRUN]	= { AC_ERR_HSM, ATA_EH_RESET,
 				    "ran out of SGEs while reading" },
-	[PORT_CERR_PKT_PROT]	= { AC_ERR_HSM, ATA_EH_SOFTRESET,
+	[PORT_CERR_PKT_PROT]	= { AC_ERR_HSM, ATA_EH_RESET,
 				    "invalid data directon for ATAPI CDB" },
-	[PORT_CERR_SGT_BOUNDARY] = { AC_ERR_SYSTEM, ATA_EH_SOFTRESET,
+	[PORT_CERR_SGT_BOUNDARY] = { AC_ERR_SYSTEM, ATA_EH_RESET,
 				     "SGT not on qword boundary" },
-	[PORT_CERR_SGT_TGTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_SGT_TGTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI target abort while fetching SGT" },
-	[PORT_CERR_SGT_MSTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_SGT_MSTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI master abort while fetching SGT" },
-	[PORT_CERR_SGT_PCIPERR]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_SGT_PCIPERR]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI parity error while fetching SGT" },
-	[PORT_CERR_CMD_BOUNDARY] = { AC_ERR_SYSTEM, ATA_EH_SOFTRESET,
+	[PORT_CERR_CMD_BOUNDARY] = { AC_ERR_SYSTEM, ATA_EH_RESET,
 				     "PRB not on qword boundary" },
-	[PORT_CERR_CMD_TGTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_CMD_TGTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI target abort while fetching PRB" },
-	[PORT_CERR_CMD_MSTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_CMD_MSTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI master abort while fetching PRB" },
-	[PORT_CERR_CMD_PCIPERR]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_CMD_PCIPERR]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI parity error while fetching PRB" },
-	[PORT_CERR_XFR_UNDEF]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_XFR_UNDEF]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "undefined error while transferring data" },
-	[PORT_CERR_XFR_TGTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_XFR_TGTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI target abort while transferring data" },
-	[PORT_CERR_XFR_MSTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_XFR_MSTABRT]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI master abort while transferring data" },
-	[PORT_CERR_XFR_PCIPERR]	= { AC_ERR_HOST_BUS, ATA_EH_SOFTRESET,
+	[PORT_CERR_XFR_PCIPERR]	= { AC_ERR_HOST_BUS, ATA_EH_RESET,
 				    "PCI parity error while transferring data" },
-	[PORT_CERR_SENDSERVICE]	= { AC_ERR_HSM, ATA_EH_SOFTRESET,
+	[PORT_CERR_SENDSERVICE]	= { AC_ERR_HSM, ATA_EH_RESET,
 				    "FIS received while sending service FIS" },
 };
 
@@ -616,7 +616,7 @@ static int sil24_init_port(struct ata_port *ap)
 
 	if ((tmp & (PORT_CS_INIT | PORT_CS_RDY)) != PORT_CS_RDY) {
 		pp->do_port_rst = 1;
-		ap->link.eh_context.i.action |= ATA_EH_HARDRESET;
+		ap->link.eh_context.i.action |= ATA_EH_RESET;
 		return -EIO;
 	}
 
@@ -1022,7 +1022,7 @@ static void sil24_error_intr(struct ata_port *ap)
 
 	if (irq_stat & PORT_IRQ_UNK_FIS) {
 		ehi->err_mask |= AC_ERR_HSM;
-		ehi->action |= ATA_EH_SOFTRESET;
+		ehi->action |= ATA_EH_RESET;
 		ata_ehi_push_desc(ehi, "unknown FIS");
 		freeze = 1;
 	}
@@ -1043,7 +1043,7 @@ static void sil24_error_intr(struct ata_port *ap)
 		 */
 		if (ap->nr_active_links >= 3) {
 			ehi->err_mask |= AC_ERR_OTHER;
-			ehi->action |= ATA_EH_HARDRESET;
+			ehi->action |= ATA_EH_RESET;
 			ata_ehi_push_desc(ehi, "PMP DMA CS errata");
 			pp->do_port_rst = 1;
 			freeze = 1;
@@ -1064,7 +1064,7 @@ static void sil24_error_intr(struct ata_port *ap)
 						  irq_stat);
 			} else {
 				err_mask |= AC_ERR_HSM;
-				action |= ATA_EH_HARDRESET;
+				action |= ATA_EH_RESET;
 				freeze = 1;
 			}
 		} else
@@ -1078,12 +1078,12 @@ static void sil24_error_intr(struct ata_port *ap)
 		if (ci && ci->desc) {
 			err_mask |= ci->err_mask;
 			action |= ci->action;
-			if (action & ATA_EH_RESET_MASK)
+			if (action & ATA_EH_RESET)
 				freeze = 1;
 			ata_ehi_push_desc(ehi, "%s", ci->desc);
 		} else {
 			err_mask |= AC_ERR_OTHER;
-			action |= ATA_EH_SOFTRESET;
+			action |= ATA_EH_RESET;
 			freeze = 1;
 			ata_ehi_push_desc(ehi, "unknown command error %d",
 					  cerr);
@@ -1153,7 +1153,7 @@ static inline void sil24_host_intr(struct ata_port *ap)
 	if (rc < 0) {
 		struct ata_eh_info *ehi = &ap->link.eh_info;
 		ehi->err_mask |= AC_ERR_HSM;
-		ehi->action |= ATA_EH_SOFTRESET;
+		ehi->action |= ATA_EH_RESET;
 		ata_port_freeze(ap);
 		return;
 	}
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 0d03f44824fb..c0e0f1d18d50 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -320,7 +320,7 @@ static int vt6420_prereset(struct ata_link *link, unsigned long deadline)
 
 	if (!online) {
 		/* tell EH to bail */
-		ehc->i.action &= ~ATA_EH_RESET_MASK;
+		ehc->i.action &= ~ATA_EH_RESET;
 		return 0;
 	}
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 37ee881c42ac..c63cfb3b222b 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -292,12 +292,12 @@ enum {
 
 	/* reset / recovery action types */
 	ATA_EH_REVALIDATE	= (1 << 0),
-	ATA_EH_SOFTRESET	= (1 << 1),
-	ATA_EH_HARDRESET	= (1 << 2),
+	ATA_EH_SOFTRESET	= (1 << 1), /* meaningful only in ->prereset */
+	ATA_EH_HARDRESET	= (1 << 2), /* meaningful only in ->prereset */
+	ATA_EH_RESET		= ATA_EH_SOFTRESET | ATA_EH_HARDRESET,
 	ATA_EH_ENABLE_LINK	= (1 << 3),
 	ATA_EH_LPM		= (1 << 4),  /* link power management action */
 
-	ATA_EH_RESET_MASK	= ATA_EH_SOFTRESET | ATA_EH_HARDRESET,
 	ATA_EH_PERDEV_MASK	= ATA_EH_REVALIDATE,
 
 	/* ata_eh_info->flags */
@@ -1097,7 +1097,7 @@ extern void ata_ehi_clear_desc(struct ata_eh_info *ehi);
 static inline void ata_ehi_schedule_probe(struct ata_eh_info *ehi)
 {
 	ehi->flags |= ATA_EHI_RESUME_LINK;
-	ehi->action |= ATA_EH_SOFTRESET;
+	ehi->action |= ATA_EH_RESET;
 	ehi->probe_mask |= (1 << ATA_MAX_DEVICES) - 1;
 }
 
-- 
cgit v1.2.3


From d692abd92f22a81b38d52c39601871003262841c Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 24 Jan 2008 00:05:14 +0900
Subject: libata: kill ATA_LFLAG_HRST_TO_RESUME

Now that hardreset is the preferred method of resetting, there's no
need for ATA_LFLAG_HRST_TO_RESUME flag.  Kill it.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c       |  2 +-
 drivers/ata/libata-pmp.c | 16 ----------------
 drivers/ata/sata_nv.c    |  5 -----
 drivers/ata/sata_sil.c   |  5 -----
 include/linux/libata.h   |  1 -
 5 files changed, 1 insertion(+), 28 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index f6bbd52b1547..66d6c8821087 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -416,7 +416,7 @@ static const struct ata_port_info ahci_port_info[] = {
 	{
 		AHCI_HFLAGS	(AHCI_HFLAG_NO_NCQ | AHCI_HFLAG_NO_PMP),
 		.flags		= AHCI_FLAG_COMMON,
-		.link_flags	= AHCI_LFLAG_COMMON | ATA_LFLAG_HRST_TO_RESUME,
+		.link_flags	= AHCI_LFLAG_COMMON,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_vt8251_ops,
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 8439fc8efdd6..7f8bcffa81ad 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -453,9 +453,6 @@ static void sata_pmp_quirks(struct ata_port *ap)
 	if (vendor == 0x1095 && devid == 0x3726) {
 		/* sil3726 quirks */
 		ata_port_for_each_link(link, ap) {
-			/* SError.N need a kick in the ass to get working */
-			link->flags |= ATA_LFLAG_HRST_TO_RESUME;
-
 			/* class code report is unreliable */
 			if (link->pmp < 5)
 				link->flags |= ATA_LFLAG_ASSUME_ATA;
@@ -468,9 +465,6 @@ static void sata_pmp_quirks(struct ata_port *ap)
 	} else if (vendor == 0x1095 && devid == 0x4723) {
 		/* sil4723 quirks */
 		ata_port_for_each_link(link, ap) {
-			/* SError.N need a kick in the ass to get working */
-			link->flags |= ATA_LFLAG_HRST_TO_RESUME;
-
 			/* class code report is unreliable */
 			if (link->pmp < 2)
 				link->flags |= ATA_LFLAG_ASSUME_ATA;
@@ -483,9 +477,6 @@ static void sata_pmp_quirks(struct ata_port *ap)
 	} else if (vendor == 0x1095 && devid == 0x4726) {
 		/* sil4726 quirks */
 		ata_port_for_each_link(link, ap) {
-			/* SError.N need a kick in the ass to get working */
-			link->flags |= ATA_LFLAG_HRST_TO_RESUME;
-
 			/* Class code report is unreliable and SRST
 			 * times out under certain configurations.
 			 * Config device can be at port 0 or 5 and
@@ -513,13 +504,6 @@ static void sata_pmp_quirks(struct ata_port *ap)
 		 * otherwise.  Don't try hard to recover it.
 		 */
 		ap->pmp_link[ap->nr_pmp_links - 1].flags |= ATA_LFLAG_NO_RETRY;
-	} else if (vendor == 0x11ab && devid == 0x4140) {
-		/* Marvell 88SM4140 quirks.  Fan-out ports require PHY
-		 * reset to work; other than that, it behaves very
-		 * nicely.
-		 */
-		ata_port_for_each_link(link, ap)
-			link->flags |= ATA_LFLAG_HRST_TO_RESUME;
 	}
 }
 
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index ce02e15c857c..75b76535c720 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -579,7 +579,6 @@ static const struct ata_port_info nv_port_info[] = {
 	{
 		.sht		= &nv_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
-		.link_flags	= ATA_LFLAG_HRST_TO_RESUME,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
@@ -590,7 +589,6 @@ static const struct ata_port_info nv_port_info[] = {
 	{
 		.sht		= &nv_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
-		.link_flags	= ATA_LFLAG_HRST_TO_RESUME,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
@@ -601,7 +599,6 @@ static const struct ata_port_info nv_port_info[] = {
 	{
 		.sht		= &nv_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
-		.link_flags	= ATA_LFLAG_HRST_TO_RESUME,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
@@ -613,7 +610,6 @@ static const struct ata_port_info nv_port_info[] = {
 		.sht		= &nv_adma_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_MMIO | ATA_FLAG_NCQ,
-		.link_flags	= ATA_LFLAG_HRST_TO_RESUME,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
@@ -625,7 +621,6 @@ static const struct ata_port_info nv_port_info[] = {
 		.sht		= &nv_swncq_sht,
 		.flags	        = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_NCQ,
-		.link_flags	= ATA_LFLAG_HRST_TO_RESUME,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index 0b8191b52f97..7052915a31b6 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -60,7 +60,6 @@ enum {
 
 	SIL_DFL_PORT_FLAGS	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_MMIO,
-	SIL_DFL_LINK_FLAGS	= ATA_LFLAG_HRST_TO_RESUME,
 
 	/*
 	 * Controller IDs
@@ -215,7 +214,6 @@ static const struct ata_port_info sil_port_info[] = {
 	/* sil_3112 */
 	{
 		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_MOD15WRITE,
-		.link_flags	= SIL_DFL_LINK_FLAGS,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= ATA_UDMA5,
@@ -225,7 +223,6 @@ static const struct ata_port_info sil_port_info[] = {
 	{
 		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_MOD15WRITE |
 				  SIL_FLAG_NO_SATA_IRQ,
-		.link_flags	= SIL_DFL_LINK_FLAGS,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= ATA_UDMA5,
@@ -234,7 +231,6 @@ static const struct ata_port_info sil_port_info[] = {
 	/* sil_3512 */
 	{
 		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
-		.link_flags	= SIL_DFL_LINK_FLAGS,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= ATA_UDMA5,
@@ -243,7 +239,6 @@ static const struct ata_port_info sil_port_info[] = {
 	/* sil_3114 */
 	{
 		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
-		.link_flags	= SIL_DFL_LINK_FLAGS,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= ATA_UDMA5,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index c63cfb3b222b..1524af6f018b 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -163,7 +163,6 @@ enum {
 	ATA_DEV_NONE		= 9,	/* no device */
 
 	/* struct ata_link flags */
-	ATA_LFLAG_HRST_TO_RESUME = (1 << 0), /* hardreset to resume link */
 	ATA_LFLAG_SKIP_D2H_BSY	= (1 << 1), /* can't wait for the first D2H
 					     * Register FIS clearing BSY */
 	ATA_LFLAG_NO_SRST	= (1 << 2), /* avoid softreset */
-- 
cgit v1.2.3


From 672b2d65ba83a6f3f801fd3d58851aa9c0725a54 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 24 Jan 2008 00:05:14 +0900
Subject: libata: kill ATA_EHI_RESUME_LINK

ATA_EHI_RESUME_LINK has two functions - promote reset to hardreset if
ATA_LFLAG_HRST_TO_RESUME is set and preventing EH from shortcutting
reset action when probing is requested.  The former is gone now and
the latter can easily be achieved by making EH to perform at least one
reset if reset is requested, which also makes more sense than
depending on RESUME_LINK flag.

As ATA_EHI_RESUME_LINK was the only EHI reset modifier, this also
kills reset modifier handling.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-eh.c   | 21 ++++++++-------------
 drivers/ata/libata-pmp.c  |  1 -
 drivers/ata/libata-scsi.c |  1 -
 include/linux/libata.h    |  3 ---
 4 files changed, 8 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index f7cae6400155..e6584fa7f456 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1079,12 +1079,6 @@ void ata_eh_about_to_do(struct ata_link *link, struct ata_device *dev,
 
 	spin_lock_irqsave(ap->lock, flags);
 
-	/* suck in and clear reset modifier */
-	if (action & ATA_EH_RESET) {
-		ehc->i.flags |= ehi->flags & ATA_EHI_RESET_MODIFIER_MASK;
-		ehi->flags &= ~ATA_EHI_RESET_MODIFIER_MASK;
-	}
-
 	ata_eh_clear_action(link, dev, ehi, action);
 
 	if (!(ehc->i.flags & ATA_EHI_QUIET))
@@ -1110,10 +1104,6 @@ void ata_eh_done(struct ata_link *link, struct ata_device *dev,
 {
 	struct ata_eh_context *ehc = &link->eh_context;
 
-	/* if reset is complete, clear reset modifier */
-	if (action & ATA_EH_RESET)
-		ehc->i.flags &= ~ATA_EHI_RESET_MODIFIER_MASK;
-
 	ata_eh_clear_action(link, dev, &ehc->i, action);
 }
 
@@ -2491,6 +2481,7 @@ static int ata_link_nr_vacant(struct ata_link *link)
 
 static int ata_eh_skip_recovery(struct ata_link *link)
 {
+	struct ata_port *ap = link->ap;
 	struct ata_eh_context *ehc = &link->eh_context;
 	struct ata_device *dev;
 
@@ -2498,9 +2489,13 @@ static int ata_eh_skip_recovery(struct ata_link *link)
 	if (link->flags & ATA_LFLAG_DISABLED)
 		return 1;
 
-	/* thaw frozen port, resume link and recover failed devices */
-	if ((link->ap->pflags & ATA_PFLAG_FROZEN) ||
-	    (ehc->i.flags & ATA_EHI_RESUME_LINK) || ata_link_nr_enabled(link))
+	/* thaw frozen port and recover failed devices */
+	if ((ap->pflags & ATA_PFLAG_FROZEN) || ata_link_nr_enabled(link))
+		return 0;
+
+	/* reset at least once if reset is requested */
+	if ((ehc->i.action & ATA_EH_RESET) &&
+	    !(ehc->i.flags & ATA_EHI_DID_RESET))
 		return 0;
 
 	/* skip if class codes for all vacant slots are ATA_DEV_NONE */
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 7f8bcffa81ad..df1d3252b9e6 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -437,7 +437,6 @@ static int sata_pmp_init_links(struct ata_port *ap, int nr_ports)
 		link->flags = 0;
 		ehc->i.probe_mask |= 1;
 		ehc->i.action |= ATA_EH_RESET;
-		ehc->i.flags |= ATA_EHI_RESUME_LINK;
 	}
 
 	return 0;
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 1f036a7b14f3..caffca7dd76f 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -3517,7 +3517,6 @@ static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
 			struct ata_eh_info *ehi = &dev->link->eh_info;
 			ehi->probe_mask |= 1 << dev->devno;
 			ehi->action |= ATA_EH_RESET;
-			ehi->flags |= ATA_EHI_RESUME_LINK;
 		} else
 			rc = -EINVAL;
 	}
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 1524af6f018b..4093e3b6a8b7 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -301,7 +301,6 @@ enum {
 
 	/* ata_eh_info->flags */
 	ATA_EHI_HOTPLUGGED	= (1 << 0),  /* could have been hotplugged */
-	ATA_EHI_RESUME_LINK	= (1 << 1),  /* resume link (reset modifier) */
 	ATA_EHI_NO_AUTOPSY	= (1 << 2),  /* no autopsy */
 	ATA_EHI_QUIET		= (1 << 3),  /* be quiet */
 
@@ -312,7 +311,6 @@ enum {
 	ATA_EHI_POST_SETMODE	= (1 << 20), /* revaildating after setmode */
 
 	ATA_EHI_DID_RESET	= ATA_EHI_DID_SOFTRESET | ATA_EHI_DID_HARDRESET,
-	ATA_EHI_RESET_MODIFIER_MASK = ATA_EHI_RESUME_LINK,
 
 	/* max tries if error condition is still set after ->error_handler */
 	ATA_EH_MAX_TRIES	= 5,
@@ -1095,7 +1093,6 @@ extern void ata_ehi_clear_desc(struct ata_eh_info *ehi);
 
 static inline void ata_ehi_schedule_probe(struct ata_eh_info *ehi)
 {
-	ehi->flags |= ATA_EHI_RESUME_LINK;
 	ehi->action |= ATA_EH_RESET;
 	ehi->probe_mask |= (1 << ATA_MAX_DEVICES) - 1;
 }
-- 
cgit v1.2.3


From 8cebf274dd1c955a6e03385a85fd6569ce445946 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 24 Jan 2008 00:05:14 +0900
Subject: libata: kill ATA_LFLAG_SKIP_D2H_BSY

Some controllers can't reliably record the initial D2H FIS after SATA
link is brought online for whatever reason.  Advanced controllers
which don't have traditional TF register based interface often have
this problem as they don't really have the TF registers to update
while the controller and link are being initialized.

SKIP_D2H_BSY works around the problem by skipping the wait for device
readiness before issuing SRST, so for such controllers libata issues
SRST blindly and hopes for the best.

Now that libata defaults to hardreset, this workaround is no longer
necessary.  For controllers which have support for hardreset, SRST is
never issued by itself.  It is only issued as follow-up SRST for
device classification and PMP initialization, so there's no need to
wait for it from prereset.

Kill ATA_LFLAG_SKIP_D2H_BSY.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c        | 7 -------
 drivers/ata/libata-core.c | 6 ++----
 drivers/ata/sata_fsl.c    | 2 --
 drivers/ata/sata_sil24.c  | 4 ----
 include/linux/libata.h    | 2 --
 5 files changed, 2 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 66d6c8821087..1bd258e5390f 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -198,7 +198,6 @@ enum {
 					  ATA_FLAG_MMIO | ATA_FLAG_PIO_DMA |
 					  ATA_FLAG_ACPI_SATA | ATA_FLAG_AN |
 					  ATA_FLAG_IPM,
-	AHCI_LFLAG_COMMON		= ATA_LFLAG_SKIP_D2H_BSY,
 
 	ICH_MAP				= 0x90, /* ICH MAP register */
 };
@@ -407,7 +406,6 @@ static const struct ata_port_info ahci_port_info[] = {
 	/* board_ahci */
 	{
 		.flags		= AHCI_FLAG_COMMON,
-		.link_flags	= AHCI_LFLAG_COMMON,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_ops,
@@ -416,7 +414,6 @@ static const struct ata_port_info ahci_port_info[] = {
 	{
 		AHCI_HFLAGS	(AHCI_HFLAG_NO_NCQ | AHCI_HFLAG_NO_PMP),
 		.flags		= AHCI_FLAG_COMMON,
-		.link_flags	= AHCI_LFLAG_COMMON,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_vt8251_ops,
@@ -425,7 +422,6 @@ static const struct ata_port_info ahci_port_info[] = {
 	{
 		AHCI_HFLAGS	(AHCI_HFLAG_IGN_IRQ_IF_ERR),
 		.flags		= AHCI_FLAG_COMMON,
-		.link_flags	= AHCI_LFLAG_COMMON,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_ops,
@@ -436,7 +432,6 @@ static const struct ata_port_info ahci_port_info[] = {
 				 AHCI_HFLAG_32BIT_ONLY |
 				 AHCI_HFLAG_SECT255 | AHCI_HFLAG_NO_PMP),
 		.flags		= AHCI_FLAG_COMMON,
-		.link_flags	= AHCI_LFLAG_COMMON,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_ops,
@@ -447,7 +442,6 @@ static const struct ata_port_info ahci_port_info[] = {
 				 AHCI_HFLAG_MV_PATA),
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_MMIO | ATA_FLAG_PIO_DMA,
-		.link_flags	= AHCI_LFLAG_COMMON,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_ops,
@@ -457,7 +451,6 @@ static const struct ata_port_info ahci_port_info[] = {
 		AHCI_HFLAGS	(AHCI_HFLAG_IGN_SERR_INTERNAL |
 				 AHCI_HFLAG_NO_PMP),
 		.flags		= AHCI_FLAG_COMMON,
-		.link_flags	= AHCI_LFLAG_COMMON,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_ops,
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 02e7ba43a3b3..baef749ec528 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3962,10 +3962,8 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
 					"link for reset (errno=%d)\n", rc);
 	}
 
-	/* Wait for !BSY if the controller can wait for the first D2H
-	 * Reg FIS and we don't know that no device is attached.
-	 */
-	if (!(link->flags & ATA_LFLAG_SKIP_D2H_BSY) && !ata_link_offline(link)) {
+	/* wait for !BSY if we don't know that no device is attached */
+	if (!ata_link_offline(link)) {
 		rc = ata_wait_ready(ap, deadline);
 		if (rc && rc != -ENODEV) {
 			ata_link_printk(link, KERN_WARNING, "device not ready "
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index fbd423ad23bb..d23b690ed1a1 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -35,7 +35,6 @@ enum {
 	SATA_FSL_HOST_FLAGS	= (ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				ATA_FLAG_MMIO | ATA_FLAG_PIO_DMA |
 				ATA_FLAG_NCQ),
-	SATA_FSL_HOST_LFLAGS	= ATA_LFLAG_SKIP_D2H_BSY,
 
 	SATA_FSL_MAX_CMDS	= SATA_FSL_QUEUE_DEPTH,
 	SATA_FSL_CMD_HDR_SIZE	= 16,	/* 4 DWORDS */
@@ -1241,7 +1240,6 @@ static const struct ata_port_operations sata_fsl_ops = {
 static const struct ata_port_info sata_fsl_port_info[] = {
 	{
 	 .flags = SATA_FSL_HOST_FLAGS,
-	 .link_flags = SATA_FSL_HOST_LFLAGS,
 	 .pio_mask = 0x1f,	/* pio 0-4 */
 	 .udma_mask = 0x7f,	/* udma 0-6 */
 	 .port_ops = &sata_fsl_ops,
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index aa8d0323c9bb..ba0c00e8ee7f 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -254,7 +254,6 @@ enum {
 				  ATA_FLAG_MMIO | ATA_FLAG_PIO_DMA |
 				  ATA_FLAG_NCQ | ATA_FLAG_ACPI_SATA |
 				  ATA_FLAG_AN | ATA_FLAG_PMP,
-	SIL24_COMMON_LFLAGS	= ATA_LFLAG_SKIP_D2H_BSY,
 	SIL24_FLAG_PCIX_IRQ_WOC	= (1 << 24), /* IRQ loss errata on PCI-X */
 
 	IRQ_STAT_4PORTS		= 0xf,
@@ -449,7 +448,6 @@ static const struct ata_port_info sil24_port_info[] = {
 	{
 		.flags		= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(4) |
 				  SIL24_FLAG_PCIX_IRQ_WOC,
-		.link_flags	= SIL24_COMMON_LFLAGS,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= ATA_UDMA5,		/* udma0-5 */
@@ -458,7 +456,6 @@ static const struct ata_port_info sil24_port_info[] = {
 	/* sil_3132 */
 	{
 		.flags		= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(2),
-		.link_flags	= SIL24_COMMON_LFLAGS,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= ATA_UDMA5,		/* udma0-5 */
@@ -467,7 +464,6 @@ static const struct ata_port_info sil24_port_info[] = {
 	/* sil_3131/sil_3531 */
 	{
 		.flags		= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(1),
-		.link_flags	= SIL24_COMMON_LFLAGS,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= ATA_UDMA5,		/* udma0-5 */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 4093e3b6a8b7..6eec11957e54 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -163,8 +163,6 @@ enum {
 	ATA_DEV_NONE		= 9,	/* no device */
 
 	/* struct ata_link flags */
-	ATA_LFLAG_SKIP_D2H_BSY	= (1 << 1), /* can't wait for the first D2H
-					     * Register FIS clearing BSY */
 	ATA_LFLAG_NO_SRST	= (1 << 2), /* avoid softreset */
 	ATA_LFLAG_ASSUME_ATA	= (1 << 3), /* assume ATA class */
 	ATA_LFLAG_ASSUME_SEMB	= (1 << 4), /* assume SEMB class */
-- 
cgit v1.2.3


From b558edddb1c42c70a30cfe494984d4be409f7b2b Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 24 Jan 2008 00:05:14 +0900
Subject: libata: kill ata_ehi_schedule_probe()

ata_ehi_schedule_probe() was created to hide details of link-resuming
reset magic.  Now that all the softreset workarounds are gone,
scheduling probe is very simple - set probe_mask and request RESET.
Kill ata_ehi_schedule_probe() and open code it.  This also increases
consistency as ata_ehi_schedule_probe() couldn't cover individual
device probings so they were open-coded even when the helper existed.

While at it, define ATA_ALL_DEVICES as mask of all possible devices on
a link and always use it when requesting probe on link level for
simplicity and consistency.  Setting extra bits in the probe_mask
doesn't hurt anybody.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c |  3 +--
 drivers/ata/libata-eh.c   |  2 +-
 drivers/ata/libata-pmp.c  | 12 ++++++++----
 drivers/ata/libata-scsi.c |  2 +-
 include/linux/libata.h    | 12 ++++--------
 5 files changed, 15 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index baef749ec528..4dc429fd0056 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -7156,8 +7156,7 @@ int ata_host_register(struct ata_host *host, struct scsi_host_template *sht)
 			/* kick EH for boot probing */
 			spin_lock_irqsave(ap->lock, flags);
 
-			ehi->probe_mask =
-				(1 << ata_link_max_devices(&ap->link)) - 1;
+			ehi->probe_mask |= ATA_ALL_DEVICES;
 			ehi->action |= ATA_EH_RESET;
 			ehi->flags |= ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET;
 
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index e6584fa7f456..0d0a2c0ab9e7 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1746,7 +1746,7 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 		ata_eh_analyze_serror(link);
 	} else if (rc != -EOPNOTSUPP) {
 		/* SError read failed, force reset and probing */
-		ata_ehi_schedule_probe(&ehc->i);
+		ehc->i.probe_mask |= ATA_ALL_DEVICES;
 		ehc->i.action |= ATA_EH_RESET;
 		ehc->i.err_mask |= AC_ERR_OTHER;
 	}
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index df1d3252b9e6..39e036c8a2bc 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -435,7 +435,7 @@ static int sata_pmp_init_links(struct ata_port *ap, int nr_ports)
 		struct ata_eh_context *ehc = &link->eh_context;
 
 		link->flags = 0;
-		ehc->i.probe_mask |= 1;
+		ehc->i.probe_mask |= ATA_ALL_DEVICES;
 		ehc->i.action |= ATA_EH_RESET;
 	}
 
@@ -831,8 +831,12 @@ static int sata_pmp_eh_recover_pmp(struct ata_port *ap,
 		ata_eh_thaw_port(ap);
 
 		/* PMP is reset, SErrors cannot be trusted, scan all */
-		ata_port_for_each_link(tlink, ap)
-			ata_ehi_schedule_probe(&tlink->eh_context.i);
+		ata_port_for_each_link(tlink, ap) {
+			struct ata_eh_context *ehc = &tlink->eh_context;
+
+			ehc->i.probe_mask |= ATA_ALL_DEVICES;
+			ehc->i.action |= ATA_EH_RESET;
+		}
 	}
 
 	/* If revalidation is requested, revalidate and reconfigure;
@@ -847,7 +851,7 @@ static int sata_pmp_eh_recover_pmp(struct ata_port *ap,
 		tries--;
 
 		if (rc == -ENODEV) {
-			ehc->i.probe_mask |= 1;
+			ehc->i.probe_mask |= ATA_ALL_DEVICES;
 			detach = 1;
 			/* give it just two more chances */
 			tries = min(tries, 2);
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index caffca7dd76f..798ba5e45710 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -3507,7 +3507,7 @@ static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
 
 		ata_port_for_each_link(link, ap) {
 			struct ata_eh_info *ehi = &link->eh_info;
-			ehi->probe_mask |= (1 << ata_link_max_devices(link)) - 1;
+			ehi->probe_mask |= ATA_ALL_DEVICES;
 			ehi->action |= ATA_EH_RESET;
 		}
 	} else {
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 6eec11957e54..bc60132c7d33 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -122,6 +122,8 @@ enum {
 
 	ATAPI_MAX_DRAIN		= 16 << 10,
 
+	ATA_ALL_DEVICES		= (1 << ATA_MAX_DEVICES) - 1,
+
 	ATA_SHT_EMULATED	= 1,
 	ATA_SHT_CMD_PER_LUN	= 1,
 	ATA_SHT_THIS_ID		= -1,
@@ -1089,17 +1091,11 @@ extern void ata_ehi_push_desc(struct ata_eh_info *ehi, const char *fmt, ...)
 	__attribute__ ((format (printf, 2, 3)));
 extern void ata_ehi_clear_desc(struct ata_eh_info *ehi);
 
-static inline void ata_ehi_schedule_probe(struct ata_eh_info *ehi)
-{
-	ehi->action |= ATA_EH_RESET;
-	ehi->probe_mask |= (1 << ATA_MAX_DEVICES) - 1;
-}
-
 static inline void ata_ehi_hotplugged(struct ata_eh_info *ehi)
 {
-	ata_ehi_schedule_probe(ehi);
+	ehi->probe_mask |= (1 << ATA_MAX_DEVICES) - 1;
 	ehi->flags |= ATA_EHI_HOTPLUGGED;
-	ehi->action |= ATA_EH_ENABLE_LINK;
+	ehi->action |= ATA_EH_RESET | ATA_EH_ENABLE_LINK;
 	ehi->err_mask |= AC_ERR_ATA_BUS;
 }
 
-- 
cgit v1.2.3


From c1bc899f5805771926c9198e2ab4d77122c356a1 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:47 +0900
Subject: libata: reorganize ata_port_operations

Over the time, ops in ata_port_operations has become a bit confusing.
Reorganize.  SFF/BMDMA ops are separated into separate a group as they
will be taken out of ata_port_operations later.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 include/linux/libata.h | 117 ++++++++++++++++++++++++++-----------------------
 1 file changed, 61 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/include/linux/libata.h b/include/linux/libata.h
index bc60132c7d33..9476a479691b 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -665,69 +665,74 @@ struct ata_port {
 };
 
 struct ata_port_operations {
-	void (*dev_config) (struct ata_device *);
+	/*
+	 * Command execution
+	 */
+	int  (*qc_defer)(struct ata_queued_cmd *qc);
+	int  (*check_atapi_dma)(struct ata_queued_cmd *qc);
+	void (*qc_prep)(struct ata_queued_cmd *qc);
+	unsigned int (*qc_issue)(struct ata_queued_cmd *qc);
 
-	void (*set_piomode) (struct ata_port *, struct ata_device *);
-	void (*set_dmamode) (struct ata_port *, struct ata_device *);
-	unsigned long (*mode_filter) (struct ata_device *, unsigned long);
+	/*
+	 * Configuration and exception handling
+	 */
+	int  (*cable_detect)(struct ata_port *ap);
+	unsigned long (*mode_filter)(struct ata_device *dev, unsigned long xfer_mask);
+	void (*set_piomode)(struct ata_port *ap, struct ata_device *dev);
+	void (*set_dmamode)(struct ata_port *ap, struct ata_device *dev);
+	int  (*set_mode)(struct ata_link *link, struct ata_device **r_failed_dev);
 
-	void (*tf_load) (struct ata_port *ap, const struct ata_taskfile *tf);
-	void (*tf_read) (struct ata_port *ap, struct ata_taskfile *tf);
+	void (*dev_config)(struct ata_device *dev);
 
-	void (*exec_command)(struct ata_port *ap, const struct ata_taskfile *tf);
+	void (*freeze)(struct ata_port *ap);
+	void (*thaw)(struct ata_port *ap);
+	void (*error_handler)(struct ata_port *ap);
+	void (*post_internal_cmd)(struct ata_queued_cmd *qc);
+
+	/*
+	 * Optional features
+	 */
+	int  (*scr_read)(struct ata_port *ap, unsigned int sc_reg, u32 *val);
+	int  (*scr_write)(struct ata_port *ap, unsigned int sc_reg, u32 val);
+	void (*pmp_attach)(struct ata_port *ap);
+	void (*pmp_detach)(struct ata_port *ap);
+	int  (*enable_pm)(struct ata_port *ap, enum link_pm policy);
+	void (*disable_pm)(struct ata_port *ap);
+
+	/*
+	 * Start, stop, suspend and resume
+	 */
+	int  (*port_suspend)(struct ata_port *ap, pm_message_t mesg);
+	int  (*port_resume)(struct ata_port *ap);
+	int  (*port_start)(struct ata_port *ap);
+	void (*port_stop)(struct ata_port *ap);
+	void (*host_stop)(struct ata_host *host);
+
+	/*
+	 * SFF / taskfile oriented ops
+	 */
+	void (*dev_select)(struct ata_port *ap, unsigned int device);
 	u8   (*check_status)(struct ata_port *ap);
 	u8   (*check_altstatus)(struct ata_port *ap);
-	void (*dev_select)(struct ata_port *ap, unsigned int device);
-
-	void (*phy_reset) (struct ata_port *ap); /* obsolete */
-	int  (*set_mode) (struct ata_link *link, struct ata_device **r_failed_dev);
-
-	int (*cable_detect) (struct ata_port *ap);
-
-	int  (*check_atapi_dma) (struct ata_queued_cmd *qc);
-
-	void (*bmdma_setup) (struct ata_queued_cmd *qc);
-	void (*bmdma_start) (struct ata_queued_cmd *qc);
-
-	unsigned int (*data_xfer) (struct ata_device *dev, unsigned char *buf,
-				   unsigned int buflen, int rw);
-
-	int (*qc_defer) (struct ata_queued_cmd *qc);
-	void (*qc_prep) (struct ata_queued_cmd *qc);
-	unsigned int (*qc_issue) (struct ata_queued_cmd *qc);
-
-	/* port multiplier */
-	void (*pmp_attach) (struct ata_port *ap);
-	void (*pmp_detach) (struct ata_port *ap);
-
-	/* Error handlers.  ->error_handler overrides ->eng_timeout and
-	 * indicates that new-style EH is in place.
+	void (*tf_load)(struct ata_port *ap, const struct ata_taskfile *tf);
+	void (*tf_read)(struct ata_port *ap, struct ata_taskfile *tf);
+	void (*exec_command)(struct ata_port *ap, const struct ata_taskfile *tf);
+	unsigned int (*data_xfer)(struct ata_device *dev, unsigned char *buf,
+				  unsigned int buflen, int rw);
+	u8   (*irq_on)(struct ata_port *);
+
+	void (*irq_clear)(struct ata_port *);
+	void (*bmdma_setup)(struct ata_queued_cmd *qc);
+	void (*bmdma_start)(struct ata_queued_cmd *qc);
+	void (*bmdma_stop)(struct ata_queued_cmd *qc);
+	u8   (*bmdma_status)(struct ata_port *ap);
+
+	/*
+	 * Obsolete
 	 */
-	void (*eng_timeout) (struct ata_port *ap); /* obsolete */
-
-	void (*freeze) (struct ata_port *ap);
-	void (*thaw) (struct ata_port *ap);
-	void (*error_handler) (struct ata_port *ap);
-	void (*post_internal_cmd) (struct ata_queued_cmd *qc);
-
+	void (*phy_reset)(struct ata_port *ap);
+	void (*eng_timeout)(struct ata_port *ap);
 	irq_handler_t irq_handler;
-	void (*irq_clear) (struct ata_port *);
-	u8 (*irq_on) (struct ata_port *);
-
-	int (*scr_read) (struct ata_port *ap, unsigned int sc_reg, u32 *val);
-	int (*scr_write) (struct ata_port *ap, unsigned int sc_reg, u32 val);
-
-	int (*port_suspend) (struct ata_port *ap, pm_message_t mesg);
-	int (*port_resume) (struct ata_port *ap);
-	int (*enable_pm) (struct ata_port *ap, enum link_pm policy);
-	void (*disable_pm) (struct ata_port *ap);
-	int (*port_start) (struct ata_port *ap);
-	void (*port_stop) (struct ata_port *ap);
-
-	void (*host_stop) (struct ata_host *host);
-
-	void (*bmdma_stop) (struct ata_queued_cmd *qc);
-	u8   (*bmdma_status) (struct ata_port *ap);
 };
 
 struct ata_port_info {
-- 
cgit v1.2.3


From 358f9a77a668660729e705fde9c3cf69f013aa98 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:47 +0900
Subject: libata: implement and use ata_noop_irq_clear()

->irq_clear() is used to clear IRQ bit of a SFF controller and isn't
useful for drivers which don't use libata SFF HSM implementation.
However, it's a required callback and many drivers implement their own
noop version as placeholder.  This patch implements ata_noop_irq_clear
and use it to replace those custom placeholders.

Also, SFF drivers which don't support BMDMA don't need to use
ata_bmdma_irq_clear().  It becomes noop if BMDMA address isn't
initialized.  Convert them to use ata_noop_irq_clear().

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c           | 12 +++---------
 drivers/ata/libata-core.c    |  1 +
 drivers/ata/libata-sff.c     |  8 ++++++++
 drivers/ata/pata_ali.c       |  2 +-
 drivers/ata/pata_at32.c      |  7 +------
 drivers/ata/pata_icside.c    |  7 +------
 drivers/ata/pata_isapnp.c    |  2 +-
 drivers/ata/pata_ixp4xx_cf.c |  2 +-
 drivers/ata/pata_legacy.c    | 22 +++++++++++-----------
 drivers/ata/pata_mpc52xx.c   |  2 +-
 drivers/ata/pata_mpiix.c     |  2 +-
 drivers/ata/pata_ns87410.c   |  2 +-
 drivers/ata/pata_pcmcia.c    |  4 ++--
 drivers/ata/pata_platform.c  |  2 +-
 drivers/ata/pata_qdi.c       |  4 ++--
 drivers/ata/pata_rb500_cf.c  |  6 +-----
 drivers/ata/pata_winbond.c   |  2 +-
 drivers/ata/pdc_adma.c       |  8 +-------
 drivers/ata/sata_fsl.c       |  7 +------
 drivers/ata/sata_inic162x.c  |  7 +------
 drivers/ata/sata_mv.c        | 11 +++--------
 drivers/ata/sata_qstor.c     |  8 +-------
 drivers/ata/sata_sil24.c     |  8 +-------
 include/linux/libata.h       |  1 +
 24 files changed, 47 insertions(+), 90 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 1bd258e5390f..492e521715d6 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -244,7 +244,6 @@ static int ahci_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
 static int ahci_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
 static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 static unsigned int ahci_qc_issue(struct ata_queued_cmd *qc);
-static void ahci_irq_clear(struct ata_port *ap);
 static int ahci_port_start(struct ata_port *ap);
 static void ahci_port_stop(struct ata_port *ap);
 static void ahci_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
@@ -307,7 +306,7 @@ static const struct ata_port_operations ahci_ops = {
 	.qc_prep		= ahci_qc_prep,
 	.qc_issue		= ahci_qc_issue,
 
-	.irq_clear		= ahci_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 
 	.scr_read		= ahci_scr_read,
 	.scr_write		= ahci_scr_write,
@@ -343,7 +342,7 @@ static const struct ata_port_operations ahci_vt8251_ops = {
 	.qc_prep		= ahci_qc_prep,
 	.qc_issue		= ahci_qc_issue,
 
-	.irq_clear		= ahci_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 
 	.scr_read		= ahci_scr_read,
 	.scr_write		= ahci_scr_write,
@@ -377,7 +376,7 @@ static const struct ata_port_operations ahci_p5wdh_ops = {
 	.qc_prep		= ahci_qc_prep,
 	.qc_issue		= ahci_qc_issue,
 
-	.irq_clear		= ahci_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 
 	.scr_read		= ahci_scr_read,
 	.scr_write		= ahci_scr_write,
@@ -1769,11 +1768,6 @@ static void ahci_port_intr(struct ata_port *ap)
 	}
 }
 
-static void ahci_irq_clear(struct ata_port *ap)
-{
-	/* TODO */
-}
-
 static irqreturn_t ahci_interrupt(int irq, void *dev_instance)
 {
 	struct ata_host *host = dev_instance;
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 4dc429fd0056..394edf937cf2 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -7824,6 +7824,7 @@ EXPORT_SYMBOL_GPL(ata_noop_qc_prep);
 EXPORT_SYMBOL_GPL(ata_bmdma_setup);
 EXPORT_SYMBOL_GPL(ata_bmdma_start);
 EXPORT_SYMBOL_GPL(ata_bmdma_irq_clear);
+EXPORT_SYMBOL_GPL(ata_noop_irq_clear);
 EXPORT_SYMBOL_GPL(ata_bmdma_status);
 EXPORT_SYMBOL_GPL(ata_bmdma_stop);
 EXPORT_SYMBOL_GPL(ata_bmdma_freeze);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 20dc572fb45a..1cf03d41aa33 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -302,6 +302,14 @@ void ata_bmdma_irq_clear(struct ata_port *ap)
 	iowrite8(ioread8(mmio + ATA_DMA_STATUS), mmio + ATA_DMA_STATUS);
 }
 
+/**
+ *	ata_noop_irq_clear - Noop placeholder for irq_clear
+ *	@ap: Port associated with this ATA transaction.
+ */
+void ata_noop_irq_clear(struct ata_port *ap)
+{
+}
+
 /**
  *	ata_bmdma_status - Read PCI IDE BMDMA status
  *	@ap: Port associated with this ATA transaction.
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index 3814aebefb2d..84b748aef386 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -380,7 +380,7 @@ static struct ata_port_operations ali_early_port_ops = {
 	.data_xfer	= ata_data_xfer,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pata_at32.c b/drivers/ata/pata_at32.c
index db057b183d60..27c959f35c2c 100644
--- a/drivers/ata/pata_at32.c
+++ b/drivers/ata/pata_at32.c
@@ -166,11 +166,6 @@ static void pata_at32_set_piomode(struct ata_port *ap, struct ata_device *adev)
 	}
 }
 
-static void pata_at32_irq_clear(struct ata_port *ap)
-{
-	/* No DMA controller yet */
-}
-
 static struct scsi_host_template at32_sht = {
 	.module			= THIS_MODULE,
 	.name			= DRV_NAME,
@@ -208,7 +203,7 @@ static struct ata_port_operations at32_port_ops = {
 
 	.data_xfer		= ata_data_xfer,
 
-	.irq_clear		= pata_at32_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.port_start		= ata_sff_port_start,
diff --git a/drivers/ata/pata_icside.c b/drivers/ata/pata_icside.c
index f97068be2d79..e1230cae7ff1 100644
--- a/drivers/ata/pata_icside.c
+++ b/drivers/ata/pata_icside.c
@@ -322,11 +322,6 @@ static struct scsi_host_template pata_icside_sht = {
 	.bios_param		= ata_std_bios_param,
 };
 
-/* wish this was exported from libata-core */
-static void ata_dummy_noret(struct ata_port *port)
-{
-}
-
 static void pata_icside_postreset(struct ata_link *link, unsigned int *classes)
 {
 	struct ata_port *ap = link->ap;
@@ -380,7 +375,7 @@ static struct ata_port_operations pata_icside_port_ops = {
 	.error_handler		= pata_icside_error_handler,
 	.post_internal_cmd	= pata_icside_bmdma_stop,
 
-	.irq_clear		= ata_dummy_noret,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.bmdma_stop		= pata_icside_bmdma_stop,
diff --git a/drivers/ata/pata_isapnp.c b/drivers/ata/pata_isapnp.c
index 4320e7986321..ef561de0c24d 100644
--- a/drivers/ata/pata_isapnp.c
+++ b/drivers/ata/pata_isapnp.c
@@ -55,7 +55,7 @@ static struct ata_port_operations isapnp_port_ops = {
 
 	.data_xfer	= ata_data_xfer,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c
index 030878fedeb5..83e38cc077e1 100644
--- a/drivers/ata/pata_ixp4xx_cf.c
+++ b/drivers/ata/pata_ixp4xx_cf.c
@@ -126,7 +126,7 @@ static struct ata_port_operations ixp4xx_port_ops = {
 	.cable_detect		= ata_cable_40wire,
 
 	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.port_start		= ata_port_start,
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
index 50fe08ebe23c..6ac02f7d5289 100644
--- a/drivers/ata/pata_legacy.c
+++ b/drivers/ata/pata_legacy.c
@@ -252,7 +252,7 @@ static struct ata_port_operations simple_port_ops = {
 	.data_xfer	= ata_data_xfer_noirq,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -279,7 +279,7 @@ static struct ata_port_operations legacy_port_ops = {
 	.data_xfer	= ata_data_xfer_noirq,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -393,7 +393,7 @@ static struct ata_port_operations pdc20230_port_ops = {
 	.data_xfer	= pdc_data_xfer_vlb,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -447,7 +447,7 @@ static struct ata_port_operations ht6560a_port_ops = {
 	.data_xfer	= ata_data_xfer,	/* Check vlb/noirq */
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -512,7 +512,7 @@ static struct ata_port_operations ht6560b_port_ops = {
 	.data_xfer	= ata_data_xfer,    /* FIXME: Check 32bit and noirq */
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -633,7 +633,7 @@ static struct ata_port_operations opti82c611a_port_ops = {
 	.data_xfer	= ata_data_xfer,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -765,7 +765,7 @@ static struct ata_port_operations opti82c46x_port_ops = {
 	.data_xfer	= ata_data_xfer,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -950,7 +950,7 @@ static struct ata_port_operations qdi6500_port_ops = {
 	.data_xfer	= vlb32_data_xfer,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -977,7 +977,7 @@ static struct ata_port_operations qdi6580_port_ops = {
 	.data_xfer	= vlb32_data_xfer,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -1004,7 +1004,7 @@ static struct ata_port_operations qdi6580dp_port_ops = {
 	.data_xfer	= vlb32_data_xfer,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -1095,7 +1095,7 @@ static struct ata_port_operations winbond_port_ops = {
 
 	.data_xfer	= vlb32_data_xfer,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index 5413ebfa72e5..d84e0c8ea02f 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -296,7 +296,7 @@ static struct ata_port_operations mpc52xx_ata_port_ops = {
 	.qc_prep		= ata_qc_prep,
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_data_xfer,
-	.irq_clear		= ata_bmdma_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 	.port_start		= ata_port_start,
 };
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index c0d9e0cf208c..ced6372749b3 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -187,7 +187,7 @@ static struct ata_port_operations mpiix_port_ops = {
 	.qc_issue	= mpiix_qc_issue_prot,
 	.data_xfer	= ata_data_xfer,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index 9fe66fd75017..d182bdf31ee1 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -182,7 +182,7 @@ static struct ata_port_operations ns87410_port_ops = {
 	.data_xfer	= ata_data_xfer,
 
 	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index 3e7f6a9da28b..9881a9e004a4 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -164,7 +164,7 @@ static struct ata_port_operations pcmcia_port_ops = {
 
 	.data_xfer	= ata_data_xfer_noirq,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -189,7 +189,7 @@ static struct ata_port_operations pcmcia_8bit_port_ops = {
 
 	.data_xfer	= ata_data_xfer_8bit,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c
index aad7adc6ea56..a8429f5c3006 100644
--- a/drivers/ata/pata_platform.c
+++ b/drivers/ata/pata_platform.c
@@ -86,7 +86,7 @@ static struct ata_port_operations pata_platform_port_ops = {
 
 	.data_xfer		= ata_data_xfer_noirq,
 
-	.irq_clear		= ata_bmdma_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.port_start		= ata_dummy_ret0,
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
index 9f308ed76cc8..60238d5748a7 100644
--- a/drivers/ata/pata_qdi.c
+++ b/drivers/ata/pata_qdi.c
@@ -191,7 +191,7 @@ static struct ata_port_operations qdi6500_port_ops = {
 
 	.data_xfer	= qdi_data_xfer,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
@@ -217,7 +217,7 @@ static struct ata_port_operations qdi6580_port_ops = {
 
 	.data_xfer	= qdi_data_xfer,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pata_rb500_cf.c b/drivers/ata/pata_rb500_cf.c
index 4ce9b03fe6c8..08c32af1c84d 100644
--- a/drivers/ata/pata_rb500_cf.c
+++ b/drivers/ata/pata_rb500_cf.c
@@ -117,10 +117,6 @@ static irqreturn_t rb500_pata_irq_handler(int irq, void *dev_instance)
 	return IRQ_HANDLED;
 }
 
-static void rb500_pata_irq_clear(struct ata_port *ap)
-{
-}
-
 static int rb500_pata_port_start(struct ata_port *ap)
 {
 	return 0;
@@ -144,7 +140,7 @@ static struct ata_port_operations rb500_pata_port_ops = {
 	.error_handler		= ata_bmdma_error_handler,
 
 	.irq_handler		= rb500_pata_irq_handler,
-	.irq_clear		= rb500_pata_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.port_start		= rb500_pata_port_start,
diff --git a/drivers/ata/pata_winbond.c b/drivers/ata/pata_winbond.c
index 99c92eda217b..5318248782bb 100644
--- a/drivers/ata/pata_winbond.c
+++ b/drivers/ata/pata_winbond.c
@@ -159,7 +159,7 @@ static struct ata_port_operations winbond_port_ops = {
 
 	.data_xfer	= winbond_data_xfer,
 
-	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_clear	= ata_noop_irq_clear,
 	.irq_on		= ata_irq_on,
 
 	.port_start	= ata_sff_port_start,
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index 8e1b7e9c0ae4..bc2d12a2da30 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -138,7 +138,6 @@ static unsigned int adma_qc_issue(struct ata_queued_cmd *qc);
 static int adma_check_atapi_dma(struct ata_queued_cmd *qc);
 static void adma_bmdma_stop(struct ata_queued_cmd *qc);
 static u8 adma_bmdma_status(struct ata_port *ap);
-static void adma_irq_clear(struct ata_port *ap);
 static void adma_freeze(struct ata_port *ap);
 static void adma_thaw(struct ata_port *ap);
 static void adma_error_handler(struct ata_port *ap);
@@ -174,7 +173,7 @@ static const struct ata_port_operations adma_ata_ops = {
 	.freeze			= adma_freeze,
 	.thaw			= adma_thaw,
 	.error_handler		= adma_error_handler,
-	.irq_clear		= adma_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 	.port_start		= adma_port_start,
 	.port_stop		= adma_port_stop,
@@ -223,11 +222,6 @@ static u8 adma_bmdma_status(struct ata_port *ap)
 	return 0;
 }
 
-static void adma_irq_clear(struct ata_port *ap)
-{
-	/* nothing */
-}
-
 static void adma_reset_engine(struct ata_port *ap)
 {
 	void __iomem *chan = ADMA_PORT_REGS(ap);
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index d23b690ed1a1..031a512cbaa8 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -933,11 +933,6 @@ static void sata_fsl_post_internal_cmd(struct ata_queued_cmd *qc)
 	}
 }
 
-static void sata_fsl_irq_clear(struct ata_port *ap)
-{
-	/* unused */
-}
-
 static void sata_fsl_error_intr(struct ata_port *ap)
 {
 	struct ata_link *link = &ap->link;
@@ -1223,7 +1218,7 @@ static const struct ata_port_operations sata_fsl_ops = {
 
 	.qc_prep = sata_fsl_qc_prep,
 	.qc_issue = sata_fsl_qc_issue,
-	.irq_clear = sata_fsl_irq_clear,
+	.irq_clear = ata_noop_irq_clear,
 
 	.scr_read = sata_fsl_scr_read,
 	.scr_write = sata_fsl_scr_write,
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 59e65edc5820..74f14369dc8d 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -266,11 +266,6 @@ static u8 inic_bmdma_status(struct ata_port *ap)
 	return ATA_DMA_INTR;
 }
 
-static void inic_irq_clear(struct ata_port *ap)
-{
-	/* noop */
-}
-
 static void inic_host_intr(struct ata_port *ap)
 {
 	void __iomem *port_base = inic_port_base(ap);
@@ -555,7 +550,7 @@ static struct ata_port_operations inic_port_ops = {
 	.bmdma_stop		= inic_bmdma_stop,
 	.bmdma_status		= inic_bmdma_status,
 
-	.irq_clear		= inic_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.qc_prep	 	= ata_qc_prep,
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index a4944c8ad46d..b3b3da4eaa03 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -461,7 +461,6 @@ struct mv_hw_ops {
 	void (*reset_bus)(struct ata_host *host, void __iomem *mmio);
 };
 
-static void mv_irq_clear(struct ata_port *ap);
 static int mv_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val);
 static int mv_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val);
 static int mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val);
@@ -564,7 +563,7 @@ static const struct ata_port_operations mv5_ops = {
 	.qc_issue		= mv_qc_issue,
 	.data_xfer		= ata_data_xfer,
 
-	.irq_clear		= mv_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.error_handler		= mv_error_handler,
@@ -592,7 +591,7 @@ static const struct ata_port_operations mv6_ops = {
 	.qc_issue		= mv_qc_issue,
 	.data_xfer		= ata_data_xfer,
 
-	.irq_clear		= mv_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.error_handler		= mv_error_handler,
@@ -620,7 +619,7 @@ static const struct ata_port_operations mv_iie_ops = {
 	.qc_issue		= mv_qc_issue,
 	.data_xfer		= ata_data_xfer,
 
-	.irq_clear		= mv_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 
 	.error_handler		= mv_error_handler,
@@ -801,10 +800,6 @@ static inline int mv_get_hc_count(unsigned long port_flags)
 	return ((port_flags & MV_FLAG_DUAL_HC) ? 2 : 1);
 }
 
-static void mv_irq_clear(struct ata_port *ap)
-{
-}
-
 static void mv_set_edma_ptrs(void __iomem *port_mmio,
 			     struct mv_host_priv *hpriv,
 			     struct mv_port_priv *pp)
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 91cc12c82040..3c8e97f251f9 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -121,7 +121,6 @@ static unsigned int qs_qc_issue(struct ata_queued_cmd *qc);
 static int qs_check_atapi_dma(struct ata_queued_cmd *qc);
 static void qs_bmdma_stop(struct ata_queued_cmd *qc);
 static u8 qs_bmdma_status(struct ata_port *ap);
-static void qs_irq_clear(struct ata_port *ap);
 static void qs_freeze(struct ata_port *ap);
 static void qs_thaw(struct ata_port *ap);
 static void qs_error_handler(struct ata_port *ap);
@@ -157,7 +156,7 @@ static const struct ata_port_operations qs_ata_ops = {
 	.freeze			= qs_freeze,
 	.thaw			= qs_thaw,
 	.error_handler		= qs_error_handler,
-	.irq_clear		= qs_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 	.irq_on			= ata_irq_on,
 	.scr_read		= qs_scr_read,
 	.scr_write		= qs_scr_write,
@@ -211,11 +210,6 @@ static u8 qs_bmdma_status(struct ata_port *ap)
 	return 0;
 }
 
-static void qs_irq_clear(struct ata_port *ap)
-{
-	/* nothing */
-}
-
 static inline void qs_enter_reg_mode(struct ata_port *ap)
 {
 	u8 __iomem *chan = qs_mmio_base(ap->host) + (ap->port_no * 0x4000);
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index ba0c00e8ee7f..b85464d51f68 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -348,7 +348,6 @@ static void sil24_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
 static int sil24_qc_defer(struct ata_queued_cmd *qc);
 static void sil24_qc_prep(struct ata_queued_cmd *qc);
 static unsigned int sil24_qc_issue(struct ata_queued_cmd *qc);
-static void sil24_irq_clear(struct ata_port *ap);
 static void sil24_pmp_attach(struct ata_port *ap);
 static void sil24_pmp_detach(struct ata_port *ap);
 static void sil24_freeze(struct ata_port *ap);
@@ -416,7 +415,7 @@ static const struct ata_port_operations sil24_ops = {
 	.qc_prep		= sil24_qc_prep,
 	.qc_issue		= sil24_qc_issue,
 
-	.irq_clear		= sil24_irq_clear,
+	.irq_clear		= ata_noop_irq_clear,
 
 	.scr_read		= sil24_scr_read,
 	.scr_write		= sil24_scr_write,
@@ -921,11 +920,6 @@ static unsigned int sil24_qc_issue(struct ata_queued_cmd *qc)
 	return 0;
 }
 
-static void sil24_irq_clear(struct ata_port *ap)
-{
-	/* unused */
-}
-
 static void sil24_pmp_attach(struct ata_port *ap)
 {
 	sil24_config_pmp(ap, 1);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 9476a479691b..639298af583e 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -896,6 +896,7 @@ extern void ata_bmdma_start(struct ata_queued_cmd *qc);
 extern void ata_bmdma_stop(struct ata_queued_cmd *qc);
 extern u8   ata_bmdma_status(struct ata_port *ap);
 extern void ata_bmdma_irq_clear(struct ata_port *ap);
+extern void ata_noop_irq_clear(struct ata_port *ap);
 extern void ata_bmdma_freeze(struct ata_port *ap);
 extern void ata_bmdma_thaw(struct ata_port *ap);
 extern void ata_bmdma_drive_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
-- 
cgit v1.2.3


From 68d1d07b510bb57a504588adc2bd2758adea0965 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:49 +0900
Subject: libata: implement and use SHT initializers

libata lets low level drivers build scsi_host_template and register it
to the SCSI layer.  This allows low level drivers high level of
flexibility but also burdens them with lots of boilerplate entries.

This patch implements SHT initializers which can be used to initialize
all the boilerplate entries in a sht.  Three variants of them are
implemented - BASE, BMDMA and NCQ - for different types of drivers.
Note that entries can be overriden by putting individual initializers
after the helper macro.

All sht tables are identical before and after this patch.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c              | 15 +----------
 drivers/ata/ata_generic.c       | 16 +-----------
 drivers/ata/ata_piix.c          | 16 +-----------
 drivers/ata/pata_acpi.c         | 17 +-----------
 drivers/ata/pata_ali.c          | 16 +-----------
 drivers/ata/pata_amd.c          | 16 +-----------
 drivers/ata/pata_artop.c        | 16 +-----------
 drivers/ata/pata_at32.c         | 16 +-----------
 drivers/ata/pata_atiixp.c       | 15 +----------
 drivers/ata/pata_bf54x.c        | 14 +---------
 drivers/ata/pata_cmd640.c       | 16 +-----------
 drivers/ata/pata_cmd64x.c       | 16 +-----------
 drivers/ata/pata_cs5520.c       | 15 +----------
 drivers/ata/pata_cs5530.c       | 17 ++----------
 drivers/ata/pata_cs5535.c       | 16 +-----------
 drivers/ata/pata_cs5536.c       | 16 +-----------
 drivers/ata/pata_cypress.c      | 16 +-----------
 drivers/ata/pata_efar.c         | 16 +-----------
 drivers/ata/pata_hpt366.c       | 16 +-----------
 drivers/ata/pata_hpt37x.c       | 16 +-----------
 drivers/ata/pata_hpt3x2n.c      | 16 +-----------
 drivers/ata/pata_hpt3x3.c       | 16 +-----------
 drivers/ata/pata_icside.c       | 14 +---------
 drivers/ata/pata_isapnp.c       | 16 +-----------
 drivers/ata/pata_it8213.c       | 16 +-----------
 drivers/ata/pata_it821x.c       | 16 +-----------
 drivers/ata/pata_ixp4xx_cf.c    | 16 +-----------
 drivers/ata/pata_jmicron.c      | 17 +-----------
 drivers/ata/pata_legacy.c       | 16 +-----------
 drivers/ata/pata_marvell.c      | 17 +-----------
 drivers/ata/pata_mpc52xx.c      | 16 +-----------
 drivers/ata/pata_mpiix.c        | 16 +-----------
 drivers/ata/pata_netcell.c      | 17 +-----------
 drivers/ata/pata_ninja32.c      | 16 +-----------
 drivers/ata/pata_ns87410.c      | 16 +-----------
 drivers/ata/pata_ns87415.c      | 16 +-----------
 drivers/ata/pata_oldpiix.c      | 16 +-----------
 drivers/ata/pata_opti.c         | 16 +-----------
 drivers/ata/pata_optidma.c      | 16 +-----------
 drivers/ata/pata_pcmcia.c       | 16 +-----------
 drivers/ata/pata_pdc2027x.c     | 16 +-----------
 drivers/ata/pata_pdc202xx_old.c | 16 +-----------
 drivers/ata/pata_platform.c     | 16 +-----------
 drivers/ata/pata_qdi.c          | 16 +-----------
 drivers/ata/pata_radisys.c      | 16 +-----------
 drivers/ata/pata_rb500_cf.c     | 17 +-----------
 drivers/ata/pata_rz1000.c       | 16 +-----------
 drivers/ata/pata_sc1200.c       | 17 ++----------
 drivers/ata/pata_scc.c          | 16 +-----------
 drivers/ata/pata_serverworks.c  | 16 +-----------
 drivers/ata/pata_sil680.c       | 16 +-----------
 drivers/ata/pata_sis.c          | 16 +-----------
 drivers/ata/pata_sl82c105.c     | 16 +-----------
 drivers/ata/pata_triflex.c      | 16 +-----------
 drivers/ata/pata_via.c          | 16 +-----------
 drivers/ata/pata_winbond.c      | 16 +-----------
 drivers/ata/pdc_adma.c          | 14 +---------
 drivers/ata/sata_fsl.c          | 14 +---------
 drivers/ata/sata_inic162x.c     | 16 +-----------
 drivers/ata/sata_mv.c           | 28 ++------------------
 drivers/ata/sata_nv.c           | 42 +++---------------------------
 drivers/ata/sata_promise.c      | 14 +---------
 drivers/ata/sata_qstor.c        | 14 +---------
 drivers/ata/sata_sil.c          | 16 +-----------
 drivers/ata/sata_sil24.c        | 14 +---------
 drivers/ata/sata_sis.c          | 16 +-----------
 drivers/ata/sata_svw.c          | 16 +-----------
 drivers/ata/sata_sx4.c          | 14 +---------
 drivers/ata/sata_uli.c          | 16 +-----------
 drivers/ata/sata_via.c          | 16 +-----------
 drivers/ata/sata_vsc.c          | 16 +-----------
 include/linux/libata.h          | 57 +++++++++++++++++++++++++++++++++++++++++
 72 files changed, 133 insertions(+), 1086 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index c6ea44a7f2a9..8862595cb2cf 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -62,7 +62,6 @@ enum {
 	AHCI_MAX_PORTS		= 32,
 	AHCI_MAX_SG		= 168, /* hardware max is 64K */
 	AHCI_DMA_BOUNDARY	= 0xffffffff,
-	AHCI_USE_CLUSTERING	= 1,
 	AHCI_MAX_CMDS		= 32,
 	AHCI_CMD_SZ		= 32,
 	AHCI_CMD_SLOT_SZ	= AHCI_MAX_CMDS * AHCI_CMD_SZ,
@@ -274,22 +273,10 @@ static struct class_device_attribute *ahci_shost_attrs[] = {
 };
 
 static struct scsi_host_template ahci_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.change_queue_depth	= ata_scsi_change_queue_depth,
+	ATA_NCQ_SHT(DRV_NAME),
 	.can_queue		= AHCI_MAX_CMDS - 1,
-	.this_id		= ATA_SHT_THIS_ID,
 	.sg_tablesize		= AHCI_MAX_SG,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= AHCI_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= AHCI_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 	.shost_attrs		= ahci_shost_attrs,
 };
 
diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index db4c3cb78fda..5c64ce134c6c 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -95,21 +95,7 @@ static int generic_set_mode(struct ata_link *link, struct ata_device **unused)
 }
 
 static struct scsi_host_template generic_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations generic_port_ops = {
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 067760a16889..9f887b2c92df 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -291,21 +291,7 @@ static struct pci_driver piix_pci_driver = {
 };
 
 static struct scsi_host_template piix_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations piix_pata_ops = {
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index bdc3b9d7395c..187545c0898a 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -232,22 +232,7 @@ static int pacpi_port_start(struct ata_port *ap)
 }
 
 static struct scsi_host_template pacpi_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	/* Use standard CHS mapping rules */
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations pacpi_ops = {
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index 84b748aef386..f3d6d9b345ba 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -339,21 +339,7 @@ static int ali_check_atapi_dma(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template ali_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 /*
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 5e1bc13a756b..90d786dfbec3 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -353,21 +353,7 @@ static void nv_host_stop(struct ata_host *host)
 }
 
 static struct scsi_host_template amd_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations amd33_port_ops = {
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index a238c7bd0bba..7bfb7e8bdca2 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -314,21 +314,7 @@ static void artop6260_set_dmamode (struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template artop_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations artop6210_ops = {
diff --git a/drivers/ata/pata_at32.c b/drivers/ata/pata_at32.c
index 27c959f35c2c..d7b7b7fde362 100644
--- a/drivers/ata/pata_at32.c
+++ b/drivers/ata/pata_at32.c
@@ -167,21 +167,7 @@ static void pata_at32_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template at32_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations at32_port_ops = {
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 408bdc1a9776..645c47271ff5 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -221,21 +221,8 @@ static void atiixp_bmdma_stop(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template atiixp_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BMDMA_SHT(DRV_NAME),
 	.sg_tablesize		= LIBATA_DUMB_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static struct ata_port_operations atiixp_port_ops = {
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 146c202d5834..6c75fcac3cf4 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1357,21 +1357,9 @@ static int bfin_port_start(struct ata_port *ap)
 }
 
 static struct scsi_host_template bfin_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BASE_SHT(DRV_NAME),
 	.sg_tablesize		= SG_NONE,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static const struct ata_port_operations bfin_pata_ops = {
diff --git a/drivers/ata/pata_cmd640.c b/drivers/ata/pata_cmd640.c
index 0ef1d1ded1f8..26562b814400 100644
--- a/drivers/ata/pata_cmd640.c
+++ b/drivers/ata/pata_cmd640.c
@@ -166,21 +166,7 @@ static int cmd640_port_start(struct ata_port *ap)
 }
 
 static struct scsi_host_template cmd640_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations cmd640_port_ops = {
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index e8c1262341ee..6aea05cc0940 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -266,21 +266,7 @@ static void cmd646r1_bmdma_stop(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template cmd64x_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations cmd64x_port_ops = {
diff --git a/drivers/ata/pata_cs5520.c b/drivers/ata/pata_cs5520.c
index 44ad2c9d488f..7e643099a444 100644
--- a/drivers/ata/pata_cs5520.c
+++ b/drivers/ata/pata_cs5520.c
@@ -140,21 +140,8 @@ static void cs5520_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template cs5520_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BMDMA_SHT(DRV_NAME),
 	.sg_tablesize		= LIBATA_DUMB_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static struct ata_port_operations cs5520_port_ops = {
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
index f876aeddf1a1..0bb03dabcf18 100644
--- a/drivers/ata/pata_cs5530.c
+++ b/drivers/ata/pata_cs5530.c
@@ -161,21 +161,8 @@ static unsigned int cs5530_qc_issue_prot(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template cs5530_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_DUMB_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
+	.sg_tablesize	= LIBATA_DUMB_MAX_PRD,
 };
 
 static struct ata_port_operations cs5530_port_ops = {
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
index 01324530d052..48a18349c1d8 100644
--- a/drivers/ata/pata_cs5535.c
+++ b/drivers/ata/pata_cs5535.c
@@ -158,21 +158,7 @@ static void cs5535_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template cs5535_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations cs5535_port_ops = {
diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c
index 391aa888f8fd..f02d9107ef3b 100644
--- a/drivers/ata/pata_cs5536.c
+++ b/drivers/ata/pata_cs5536.c
@@ -221,21 +221,7 @@ static void cs5536_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template cs5536_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations cs5536_port_ops = {
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
index fc5f9c4e5d87..07fa1ab36315 100644
--- a/drivers/ata/pata_cypress.c
+++ b/drivers/ata/pata_cypress.c
@@ -110,21 +110,7 @@ static void cy82c693_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template cy82c693_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations cy82c693_port_ops = {
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index dc33220fe5b2..8700d9dcd8c9 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -233,21 +233,7 @@ static void efar_set_dmamode (struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template efar_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations efar_ops = {
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
index a82089048f58..a30028de41c0 100644
--- a/drivers/ata/pata_hpt366.c
+++ b/drivers/ata/pata_hpt366.c
@@ -290,21 +290,7 @@ static void hpt366_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template hpt36x_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 /*
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
index 2ddcd07a7518..7d6fac43e2f9 100644
--- a/drivers/ata/pata_hpt37x.c
+++ b/drivers/ata/pata_hpt37x.c
@@ -619,21 +619,7 @@ static void hpt37x_bmdma_stop(struct ata_queued_cmd *qc)
 
 
 static struct scsi_host_template hpt37x_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 /*
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
index 3a517cb9bd3e..aa380c46b168 100644
--- a/drivers/ata/pata_hpt3x2n.c
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -339,21 +339,7 @@ static unsigned int hpt3x2n_qc_issue_prot(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template hpt3x2n_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 /*
diff --git a/drivers/ata/pata_hpt3x3.c b/drivers/ata/pata_hpt3x3.c
index c09f95a4a0d1..9837ab0181e8 100644
--- a/drivers/ata/pata_hpt3x3.c
+++ b/drivers/ata/pata_hpt3x3.c
@@ -102,21 +102,7 @@ static int hpt3x3_atapi_dma(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template hpt3x3_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations hpt3x3_port_ops = {
diff --git a/drivers/ata/pata_icside.c b/drivers/ata/pata_icside.c
index e1230cae7ff1..88a1c7ae0a4d 100644
--- a/drivers/ata/pata_icside.c
+++ b/drivers/ata/pata_icside.c
@@ -305,21 +305,9 @@ static int icside_dma_init(struct pata_icside_info *info)
 
 
 static struct scsi_host_template pata_icside_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BASE_SHT(DRV_NAME),
 	.sg_tablesize		= PATA_ICSIDE_MAX_SG,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= ~0, /* no dma boundaries */
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static void pata_icside_postreset(struct ata_link *link, unsigned int *classes)
diff --git a/drivers/ata/pata_isapnp.c b/drivers/ata/pata_isapnp.c
index ef561de0c24d..91ca4d50db04 100644
--- a/drivers/ata/pata_isapnp.c
+++ b/drivers/ata/pata_isapnp.c
@@ -20,21 +20,7 @@
 #define DRV_VERSION "0.2.2"
 
 static struct scsi_host_template isapnp_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations isapnp_port_ops = {
diff --git a/drivers/ata/pata_it8213.c b/drivers/ata/pata_it8213.c
index 25c49c2e1519..678a05b304d8 100644
--- a/drivers/ata/pata_it8213.c
+++ b/drivers/ata/pata_it8213.c
@@ -243,21 +243,7 @@ static void it8213_set_dmamode (struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template it8213_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations it8213_ops = {
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index 6bdbb7140dfa..7d969c911731 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -632,21 +632,7 @@ static int it821x_port_start(struct ata_port *ap)
 }
 
 static struct scsi_host_template it821x_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations it821x_smart_port_ops = {
diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c
index 6eb8cc9a3f12..b7e8e825a869 100644
--- a/drivers/ata/pata_ixp4xx_cf.c
+++ b/drivers/ata/pata_ixp4xx_cf.c
@@ -88,21 +88,7 @@ static unsigned int ixp4xx_mmio_data_xfer(struct ata_device *dev,
 }
 
 static struct scsi_host_template ixp4xx_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations ixp4xx_port_ops = {
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index 7d36fa85435a..69781af7b1bb 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -122,22 +122,7 @@ static void jmicron_error_handler(struct ata_port *ap)
 /* No PIO or DMA methods needed for this device */
 
 static struct scsi_host_template jmicron_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	/* Use standard CHS mapping rules */
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations jmicron_ops = {
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
index 6ac02f7d5289..5329b954c5f2 100644
--- a/drivers/ata/pata_legacy.c
+++ b/drivers/ata/pata_legacy.c
@@ -208,21 +208,7 @@ static int legacy_set_mode(struct ata_link *link, struct ata_device **unused)
 }
 
 static struct scsi_host_template legacy_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 /*
diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c
index c4ee9b45301f..9de6e429d0d1 100644
--- a/drivers/ata/pata_marvell.c
+++ b/drivers/ata/pata_marvell.c
@@ -92,22 +92,7 @@ static void marvell_error_handler(struct ata_port *ap)
 /* No PIO or DMA methods needed for this device */
 
 static struct scsi_host_template marvell_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	/* Use standard CHS mapping rules */
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations marvell_ops = {
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index fefe71dbed1a..4117b618a9d9 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -265,21 +265,7 @@ mpc52xx_ata_error_handler(struct ata_port *ap)
 
 
 static struct scsi_host_template mpc52xx_ata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations mpc52xx_ata_port_ops = {
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index ced6372749b3..e8e6837110b4 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -151,21 +151,7 @@ static unsigned int mpiix_qc_issue_prot(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template mpiix_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations mpiix_port_ops = {
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
index 9fd1a84c01d3..11f200a2a156 100644
--- a/drivers/ata/pata_netcell.c
+++ b/drivers/ata/pata_netcell.c
@@ -21,22 +21,7 @@
 /* No PIO or DMA methods needed for this device */
 
 static struct scsi_host_template netcell_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	/* Use standard CHS mapping rules */
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations netcell_ops = {
diff --git a/drivers/ata/pata_ninja32.c b/drivers/ata/pata_ninja32.c
index 15dd649f89ee..ce3b07cab8bc 100644
--- a/drivers/ata/pata_ninja32.c
+++ b/drivers/ata/pata_ninja32.c
@@ -79,21 +79,7 @@ static void ninja32_dev_select(struct ata_port *ap, unsigned int device)
 }
 
 static struct scsi_host_template ninja32_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations ninja32_port_ops = {
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index d182bdf31ee1..d2f85f107d15 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -144,21 +144,7 @@ static unsigned int ns87410_qc_issue_prot(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template ns87410_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations ns87410_port_ops = {
diff --git a/drivers/ata/pata_ns87415.c b/drivers/ata/pata_ns87415.c
index 93eb958cb0c9..78d634423cbf 100644
--- a/drivers/ata/pata_ns87415.c
+++ b/drivers/ata/pata_ns87415.c
@@ -366,21 +366,7 @@ static const struct ata_port_operations ns87415_pata_ops = {
 };
 
 static struct scsi_host_template ns87415_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index 44da09ace52c..45f9b3eb5b45 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -220,21 +220,7 @@ static unsigned int oldpiix_qc_issue_prot(struct ata_queued_cmd *qc)
 
 
 static struct scsi_host_template oldpiix_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations oldpiix_pata_ops = {
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index 1e865f138d1c..1deacfa0be07 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -165,21 +165,7 @@ static void opti_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template opti_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations opti_port_ops = {
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index 3f9d03599f23..7495758a86fe 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -350,21 +350,7 @@ static int optidma_set_mode(struct ata_link *link, struct ata_device **r_failed)
 }
 
 static struct scsi_host_template optidma_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations optidma_port_ops = {
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index 9881a9e004a4..c05b36c94d51 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -128,21 +128,7 @@ static unsigned int ata_data_xfer_8bit(struct ata_device *dev,
 
 
 static struct scsi_host_template pcmcia_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations pcmcia_port_ops = {
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
index 511c89b9bae8..229d9acd934a 100644
--- a/drivers/ata/pata_pdc2027x.c
+++ b/drivers/ata/pata_pdc2027x.c
@@ -129,21 +129,7 @@ static struct pci_driver pdc2027x_pci_driver = {
 };
 
 static struct scsi_host_template pdc2027x_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations pdc2027x_pata100_ops = {
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
index 3ed866723e0c..564ee0798ec1 100644
--- a/drivers/ata/pata_pdc202xx_old.c
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -262,21 +262,7 @@ static int pdc2026x_check_atapi_dma(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template pdc202xx_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations pdc2024x_port_ops = {
diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c
index 602f5562d6fb..cd2d03a4591a 100644
--- a/drivers/ata/pata_platform.c
+++ b/drivers/ata/pata_platform.c
@@ -47,21 +47,7 @@ static int pata_platform_set_mode(struct ata_link *link, struct ata_device **unu
 }
 
 static struct scsi_host_template pata_platform_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations pata_platform_port_ops = {
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
index 60238d5748a7..ccb8682300b8 100644
--- a/drivers/ata/pata_qdi.c
+++ b/drivers/ata/pata_qdi.c
@@ -154,21 +154,7 @@ static unsigned int qdi_data_xfer(struct ata_device *dev, unsigned char *buf,
 }
 
 static struct scsi_host_template qdi_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations qdi6500_port_ops = {
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
index 8109b08fc024..3981bf84d093 100644
--- a/drivers/ata/pata_radisys.c
+++ b/drivers/ata/pata_radisys.c
@@ -185,21 +185,7 @@ static unsigned int radisys_qc_issue_prot(struct ata_queued_cmd *qc)
 
 
 static struct scsi_host_template radisys_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations radisys_pata_ops = {
diff --git a/drivers/ata/pata_rb500_cf.c b/drivers/ata/pata_rb500_cf.c
index 22cb9e1a02f5..4543c980342c 100644
--- a/drivers/ata/pata_rb500_cf.c
+++ b/drivers/ata/pata_rb500_cf.c
@@ -142,22 +142,7 @@ static struct ata_port_operations rb500_pata_port_ops = {
 /* ------------------------------------------------------------------------ */
 
 static struct scsi_host_template rb500_pata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
-	.proc_name		= DRV_NAME,
-
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 /* ------------------------------------------------------------------------ */
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
index 75b252111106..80909a607d36 100644
--- a/drivers/ata/pata_rz1000.c
+++ b/drivers/ata/pata_rz1000.c
@@ -53,21 +53,7 @@ static int rz1000_set_mode(struct ata_link *link, struct ata_device **unused)
 
 
 static struct scsi_host_template rz1000_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations rz1000_port_ops = {
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
index 725a8586cd6e..38ce6e12ee3d 100644
--- a/drivers/ata/pata_sc1200.c
+++ b/drivers/ata/pata_sc1200.c
@@ -179,21 +179,8 @@ static unsigned int sc1200_qc_issue_prot(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template sc1200_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_DUMB_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
+	.sg_tablesize	= LIBATA_DUMB_MAX_PRD,
 };
 
 static struct ata_port_operations sc1200_port_ops = {
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index 6c016deeaed8..1833e9ef522e 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -968,21 +968,7 @@ static void scc_port_stop (struct ata_port *ap)
 }
 
 static struct scsi_host_template scc_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations scc_pata_ops = {
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index 6702df37cfed..318a36988900 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -298,21 +298,7 @@ static void serverworks_set_dmamode(struct ata_port *ap, struct ata_device *adev
 }
 
 static struct scsi_host_template serverworks_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations serverworks_osb4_port_ops = {
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index f4dc09718cab..7812815a34c3 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -192,21 +192,7 @@ static void sil680_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template sil680_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations sil680_port_ops = {
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index abda90f51247..dcd8457a8377 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -514,21 +514,7 @@ static void sis_133_set_dmamode (struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template sis_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations sis_133_ops = {
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index 6c37181341ea..ece366bced0c 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -235,21 +235,7 @@ static int sl82c105_qc_defer(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template sl82c105_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations sl82c105_port_ops = {
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index 403eafcffe12..510569957d10 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -180,21 +180,7 @@ static void triflex_bmdma_stop(struct ata_queued_cmd *qc)
 }
 
 static struct scsi_host_template triflex_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations triflex_port_ops = {
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index 24430f70f00e..a7bc860e1310 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -335,21 +335,7 @@ static void via_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct scsi_host_template via_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations via_port_ops = {
diff --git a/drivers/ata/pata_winbond.c b/drivers/ata/pata_winbond.c
index 5318248782bb..9bafae9d5fe0 100644
--- a/drivers/ata/pata_winbond.c
+++ b/drivers/ata/pata_winbond.c
@@ -122,21 +122,7 @@ static unsigned int winbond_data_xfer(struct ata_device *dev,
 }
 
 static struct scsi_host_template winbond_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_PIO_SHT(DRV_NAME),
 };
 
 static struct ata_port_operations winbond_port_ops = {
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index bc2d12a2da30..fdf62de57cfc 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -143,21 +143,9 @@ static void adma_thaw(struct ata_port *ap);
 static void adma_error_handler(struct ata_port *ap);
 
 static struct scsi_host_template adma_ata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
-	.proc_name		= DRV_NAME,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BASE_SHT(DRV_NAME),
 	.sg_tablesize		= LIBATA_MAX_PRD,
 	.dma_boundary		= ADMA_DMA_BOUNDARY,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.use_clustering		= ENABLE_CLUSTERING,
-	.emulated		= ATA_SHT_EMULATED,
 };
 
 static const struct ata_port_operations adma_ata_ops = {
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 8389ecac62cd..f50381b4ba06 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -1191,22 +1191,10 @@ static int sata_fsl_init_controller(struct ata_host *host)
  * scsi mid-layer and libata interface structures
  */
 static struct scsi_host_template sata_fsl_sht = {
-	.module = THIS_MODULE,
-	.name = "sata_fsl",
-	.ioctl = ata_scsi_ioctl,
-	.queuecommand = ata_scsi_queuecmd,
-	.change_queue_depth = ata_scsi_change_queue_depth,
+	ATA_NCQ_SHT("sata_fsl"),
 	.can_queue = SATA_FSL_QUEUE_DEPTH,
-	.this_id = ATA_SHT_THIS_ID,
 	.sg_tablesize = SATA_FSL_MAX_PRD_USABLE,
-	.cmd_per_lun = ATA_SHT_CMD_PER_LUN,
-	.emulated = ATA_SHT_EMULATED,
-	.use_clustering = ATA_SHT_USE_CLUSTERING,
-	.proc_name = "sata_fsl",
 	.dma_boundary = ATA_DMA_BOUNDARY,
-	.slave_configure = ata_scsi_slave_config,
-	.slave_destroy = ata_scsi_slave_destroy,
-	.bios_param = ata_std_bios_param,
 };
 
 static const struct ata_port_operations sata_fsl_ops = {
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 74f14369dc8d..bb853df865da 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -109,21 +109,7 @@ struct inic_port_priv {
 };
 
 static struct scsi_host_template inic_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const int scr_map[] = {
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index 4685bce745bb..52d41edadb72 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -514,40 +514,16 @@ static int __mv_stop_dma(struct ata_port *ap);
  * PRDs for 64K boundaries in mv_fill_sg().
  */
 static struct scsi_host_template mv5_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BASE_SHT(DRV_NAME),
 	.sg_tablesize		= MV_MAX_SG_CT / 2,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= 1,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= MV_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static struct scsi_host_template mv6_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.change_queue_depth	= ata_scsi_change_queue_depth,
+	ATA_NCQ_SHT(DRV_NAME),
 	.can_queue		= MV_MAX_Q_DEPTH - 1,
-	.this_id		= ATA_SHT_THIS_ID,
 	.sg_tablesize		= MV_MAX_SG_CT / 2,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= 1,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= MV_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static const struct ata_port_operations mv5_ops = {
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index bd5b6c35ee5d..9e2b4cef48f2 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -385,59 +385,23 @@ static struct pci_driver nv_pci_driver = {
 };
 
 static struct scsi_host_template nv_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static struct scsi_host_template nv_adma_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.change_queue_depth	= ata_scsi_change_queue_depth,
+	ATA_NCQ_SHT(DRV_NAME),
 	.can_queue		= NV_ADMA_MAX_CPBS,
-	.this_id		= ATA_SHT_THIS_ID,
 	.sg_tablesize		= NV_ADMA_SGTBL_TOTAL_LEN,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= NV_ADMA_DMA_BOUNDARY,
 	.slave_configure	= nv_adma_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static struct scsi_host_template nv_swncq_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.change_queue_depth	= ata_scsi_change_queue_depth,
+	ATA_NCQ_SHT(DRV_NAME),
 	.can_queue		= ATA_MAX_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
 	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= ATA_DMA_BOUNDARY,
 	.slave_configure	= nv_swncq_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static const struct ata_port_operations nv_generic_ops = {
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index 11c1afea2db2..37c32ab3b23b 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -155,21 +155,9 @@ static int pdc_pata_cable_detect(struct ata_port *ap);
 static int pdc_sata_cable_detect(struct ata_port *ap);
 
 static struct scsi_host_template pdc_ata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BASE_SHT(DRV_NAME),
 	.sg_tablesize		= PDC_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static const struct ata_port_operations pdc_sata_ops = {
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 3c8e97f251f9..2566d0926aab 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -126,21 +126,9 @@ static void qs_thaw(struct ata_port *ap);
 static void qs_error_handler(struct ata_port *ap);
 
 static struct scsi_host_template qs_ata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BASE_SHT(DRV_NAME),
 	.sg_tablesize		= QS_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ENABLE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= QS_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static const struct ata_port_operations qs_ata_ops = {
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index 53f0bae3be43..738c1a8ae3b6 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -167,21 +167,7 @@ static struct pci_driver sil_pci_driver = {
 };
 
 static struct scsi_host_template sil_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations sil_ops = {
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index b85464d51f68..7fa63950d81a 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -384,22 +384,10 @@ static struct pci_driver sil24_pci_driver = {
 };
 
 static struct scsi_host_template sil24_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.change_queue_depth	= ata_scsi_change_queue_depth,
+	ATA_NCQ_SHT(DRV_NAME),
 	.can_queue		= SIL24_MAX_CMDS,
-	.this_id		= ATA_SHT_THIS_ID,
 	.sg_tablesize		= SIL24_MAX_SGE,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static const struct ata_port_operations sil24_ops = {
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index a8adef9786b7..4becb7fde5e7 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -86,21 +86,7 @@ static struct pci_driver sis_pci_driver = {
 };
 
 static struct scsi_host_template sis_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations sis_ops = {
diff --git a/drivers/ata/sata_svw.c b/drivers/ata/sata_svw.c
index aa690142fa90..c8768396e006 100644
--- a/drivers/ata/sata_svw.c
+++ b/drivers/ata/sata_svw.c
@@ -327,24 +327,10 @@ static int k2_sata_proc_info(struct Scsi_Host *shost, char *page, char **start,
 
 
 static struct scsi_host_template k2_sata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
+	ATA_BMDMA_SHT(DRV_NAME),
 #ifdef CONFIG_PPC_OF
 	.proc_info		= k2_sata_proc_info,
 #endif
-	.bios_param		= ata_std_bios_param,
 };
 
 
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index e3d56bc6726d..1802f92180e4 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -236,21 +236,9 @@ static unsigned int pdc20621_qc_issue_prot(struct ata_queued_cmd *qc);
 
 
 static struct scsi_host_template pdc_sata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
+	ATA_BASE_SHT(DRV_NAME),
 	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
 	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
 };
 
 static const struct ata_port_operations pdc_20621_ops = {
diff --git a/drivers/ata/sata_uli.c b/drivers/ata/sata_uli.c
index f7fc0450478b..764d7064fa59 100644
--- a/drivers/ata/sata_uli.c
+++ b/drivers/ata/sata_uli.c
@@ -76,21 +76,7 @@ static struct pci_driver uli_pci_driver = {
 };
 
 static struct scsi_host_template uli_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations uli_ops = {
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index f66ffd7da54e..9be877cb7f57 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -100,21 +100,7 @@ static struct pci_driver svia_pci_driver = {
 };
 
 static struct scsi_host_template svia_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 static const struct ata_port_operations vt6420_sata_ops = {
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index 099a2ba4cd4f..fd6855f0bf48 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -300,21 +300,7 @@ out:
 
 
 static struct scsi_host_template vsc_sata_sht = {
-	.module			= THIS_MODULE,
-	.name			= DRV_NAME,
-	.ioctl			= ata_scsi_ioctl,
-	.queuecommand		= ata_scsi_queuecmd,
-	.can_queue		= ATA_DEF_QUEUE,
-	.this_id		= ATA_SHT_THIS_ID,
-	.sg_tablesize		= LIBATA_MAX_PRD,
-	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
-	.emulated		= ATA_SHT_EMULATED,
-	.use_clustering		= ATA_SHT_USE_CLUSTERING,
-	.proc_name		= DRV_NAME,
-	.dma_boundary		= ATA_DMA_BOUNDARY,
-	.slave_configure	= ata_scsi_slave_config,
-	.slave_destroy		= ata_scsi_slave_destroy,
-	.bios_param		= ata_std_bios_param,
+	ATA_BMDMA_SHT(DRV_NAME),
 };
 
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 639298af583e..eccc38e17568 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1070,6 +1070,63 @@ extern void ata_do_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
 		      ata_reset_fn_t softreset, ata_reset_fn_t hardreset,
 		      ata_postreset_fn_t postreset);
 
+/*
+ * Base operations to inherit from and initializers for sht
+ *
+ * Operations
+ *
+ * base  : Common to all libata drivers.
+ * sata  : SATA controllers w/ native interface.
+ * pmp   : SATA controllers w/ PMP support.
+ * sff   : SFF ATA controllers w/o BMDMA support.
+ * bmdma : SFF ATA controllers w/ BMDMA support.
+ *
+ * sht initializers
+ *
+ * BASE  : Common to all libata drivers.  The user must set
+ *	   sg_tablesize and dma_boundary.
+ * PIO   : SFF ATA controllers w/ only PIO support.
+ * BMDMA : SFF ATA controllers w/ BMDMA support.  sg_tablesize and
+ *	   dma_boundary are set to BMDMA limits.
+ * NCQ   : SATA controllers supporting NCQ.  The user must set
+ *	   sg_tablesize, dma_boundary and can_queue.
+ */
+extern const struct ata_port_operations ata_base_port_ops;
+extern const struct ata_port_operations sata_port_ops;
+extern const struct ata_port_operations sata_pmp_port_ops;
+extern const struct ata_port_operations ata_sff_port_ops;
+extern const struct ata_port_operations ata_bmdma_port_ops;
+
+#define ATA_BASE_SHT(drv_name)					\
+	.module			= THIS_MODULE,			\
+	.name			= drv_name,			\
+	.ioctl			= ata_scsi_ioctl,		\
+	.queuecommand		= ata_scsi_queuecmd,		\
+	.can_queue		= ATA_DEF_QUEUE,		\
+	.this_id		= ATA_SHT_THIS_ID,		\
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,		\
+	.emulated		= ATA_SHT_EMULATED,		\
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,	\
+	.proc_name		= drv_name,			\
+	.slave_configure	= ata_scsi_slave_config,	\
+	.slave_destroy		= ata_scsi_slave_destroy,	\
+	.bios_param		= ata_std_bios_param
+
+/* PIO only, sg_tablesize and dma_boundary limits can be removed */
+#define ATA_PIO_SHT(drv_name)					\
+	ATA_BASE_SHT(drv_name),					\
+	.sg_tablesize		= LIBATA_MAX_PRD,		\
+	.dma_boundary		= ATA_DMA_BOUNDARY
+
+#define ATA_BMDMA_SHT(drv_name)					\
+	ATA_BASE_SHT(drv_name),					\
+	.sg_tablesize		= LIBATA_MAX_PRD,		\
+	.dma_boundary		= ATA_DMA_BOUNDARY
+
+#define ATA_NCQ_SHT(drv_name)					\
+	ATA_BASE_SHT(drv_name),					\
+	.change_queue_depth	= ata_scsi_change_queue_depth
+
 /*
  * printk helpers
  */
-- 
cgit v1.2.3


From 029cfd6b74fc5c517865fad78cf4a3ea8d9b664a Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:49 +0900
Subject: libata: implement and use ops inheritance

libata lets low level drivers build ata_port_operations table and
register it with libata core layer.  This allows low level drivers
high level of flexibility but also burdens them with lots of
boilerplate entries.

This becomes worse for drivers which support related similar
controllers which differ slightly.  They share most of the operations
except for a few.  However, the driver still needs to list all
operations for each variant.  This results in large number of
duplicate entries, which is not only inefficient but also error-prone
as it becomes very difficult to tell what the actual differences are.

This duplicate boilerplates all over the low level drivers also make
updating the core layer exteremely difficult and error-prone.  When
compounded with multi-branched development model, it ends up
accumulating inconsistencies over time.  Some of those inconsistencies
cause immediate problems and fixed.  Others just remain there dormant
making maintenance increasingly difficult.

To rectify the problem, this patch implements ata_port_operations
inheritance.  To allow LLDs to easily re-use their own ops tables
overriding only specific methods, this patch implements poor man's
class inheritance.  An ops table has ->inherits field which can be set
to any ops table as long as it doesn't create a loop.  When the host
is started, the inheritance chain is followed and any operation which
isn't specified is taken from the nearest ancestor which has it
specified.  This operation is called finalization and done only once
per an ops table and the LLD doesn't have to do anything special about
it other than making the ops table non-const such that libata can
update it.

libata provides four base ops tables lower drivers can inherit from -
base, sata, pmp, sff and bmdma.  To avoid overriding these ops
accidentaly, these ops are declared const and LLDs should always
inherit these instead of using them directly.

After finalization, all the ops table are identical before and after
the patch except for setting .irq_handler to ata_interrupt in drivers
which didn't use to.  The .irq_handler doesn't have any actual effect
and the field will soon be removed by later patch.

* sata_sx4 is still using old style EH and currently doesn't take
  advantage of ops inheritance.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c              |  94 ++-------------
 drivers/ata/ata_generic.c       |  31 +----
 drivers/ata/ata_piix.c          | 149 +++--------------------
 drivers/ata/libata-core.c       | 113 ++++++++++++++++-
 drivers/ata/pata_acpi.c         |  35 +-----
 drivers/ata/pata_ali.c          | 122 ++-----------------
 drivers/ata/pata_amd.c          | 191 ++++-------------------------
 drivers/ata/pata_artop.c        |  61 +---------
 drivers/ata/pata_at32.c         |  23 +---
 drivers/ata/pata_atiixp.c       |  32 +----
 drivers/ata/pata_bf54x.c        |   4 +-
 drivers/ata/pata_cmd640.c       |  33 +----
 drivers/ata/pata_cmd64x.c       |  96 ++-------------
 drivers/ata/pata_cs5520.c       |  29 +----
 drivers/ata/pata_cs5530.c       |  31 +----
 drivers/ata/pata_cs5535.c       |  31 +----
 drivers/ata/pata_cs5536.c       |  31 +----
 drivers/ata/pata_cypress.c      |  31 +----
 drivers/ata/pata_efar.c         |  30 +----
 drivers/ata/pata_hpt366.c       |  32 +----
 drivers/ata/pata_hpt37x.c       | 122 ++-----------------
 drivers/ata/pata_hpt3x2n.c      |  32 +----
 drivers/ata/pata_hpt3x3.c       |  33 +----
 drivers/ata/pata_icside.c       |  32 ++---
 drivers/ata/pata_isapnp.c       |  21 +---
 drivers/ata/pata_it8213.c       |  31 +----
 drivers/ata/pata_it821x.c       |  57 ++-------
 drivers/ata/pata_ixp4xx_cf.c    |  23 +---
 drivers/ata/pata_jmicron.c      |  31 +----
 drivers/ata/pata_legacy.c       | 260 +++-------------------------------------
 drivers/ata/pata_marvell.c      |  33 +----
 drivers/ata/pata_mpc52xx.c      |  18 +--
 drivers/ata/pata_mpiix.c        |  23 +---
 drivers/ata/pata_netcell.c      |  32 +----
 drivers/ata/pata_ninja32.c      |  30 +----
 drivers/ata/pata_ns87410.c      |  25 +---
 drivers/ata/pata_ns87415.c      |  68 ++---------
 drivers/ata/pata_oldpiix.c      |  31 +----
 drivers/ata/pata_opti.c         |  23 +---
 drivers/ata/pata_optidma.c      |  63 +---------
 drivers/ata/pata_pcmcia.c       |  48 +-------
 drivers/ata/pata_pdc2027x.c     |  56 +--------
 drivers/ata/pata_pdc202xx_old.c |  75 +++---------
 drivers/ata/pata_platform.c     |  24 +---
 drivers/ata/pata_qdi.c          |  47 +-------
 drivers/ata/pata_radisys.c      |  32 +----
 drivers/ata/pata_rb500_cf.c     |  16 +--
 drivers/ata/pata_rz1000.c       |  25 +---
 drivers/ata/pata_sc1200.c       |  33 +----
 drivers/ata/pata_scc.c          |   7 +-
 drivers/ata/pata_serverworks.c  |  63 +---------
 drivers/ata/pata_sil680.c       |  30 +----
 drivers/ata/pata_sis.c          | 177 +++------------------------
 drivers/ata/pata_sl82c105.c     |  34 +-----
 drivers/ata/pata_triflex.c      |  32 +----
 drivers/ata/pata_via.c          |  61 +---------
 drivers/ata/pata_winbond.c      |  25 +---
 drivers/ata/pdc_adma.c          |  19 +--
 drivers/ata/sata_fsl.c          |   5 +-
 drivers/ata/sata_inic162x.c     |  19 +--
 drivers/ata/sata_mv.c           |  66 ++--------
 drivers/ata/sata_nv.c           | 113 +++--------------
 drivers/ata/sata_promise.c      |  62 +++-------
 drivers/ata/sata_qstor.c        |  20 ++--
 drivers/ata/sata_sil.c          |  21 +---
 drivers/ata/sata_sil24.c        |  20 ++--
 drivers/ata/sata_sis.c          |  23 +---
 drivers/ata/sata_svw.c          |  18 +--
 drivers/ata/sata_sx4.c          |   3 +-
 drivers/ata/sata_uli.c          |  28 +----
 drivers/ata/sata_via.c          |  88 ++------------
 drivers/ata/sata_vsc.c          |  19 +--
 include/linux/libata.h          |  23 +++-
 73 files changed, 523 insertions(+), 2996 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 8862595cb2cf..dacb3ef0c3e6 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -280,118 +280,46 @@ static struct scsi_host_template ahci_sht = {
 	.shost_attrs		= ahci_shost_attrs,
 };
 
-static const struct ata_port_operations ahci_ops = {
+static struct ata_port_operations ahci_ops = {
+	.inherits		= &sata_pmp_port_ops,
+
 	.check_status		= ahci_check_status,
 	.check_altstatus	= ahci_check_status,
-	.dev_select		= ata_noop_dev_select,
-
-	.dev_config		= ahci_dev_config,
 
 	.tf_read		= ahci_tf_read,
-
 	.qc_defer		= sata_pmp_qc_defer_cmd_switch,
 	.qc_prep		= ahci_qc_prep,
 	.qc_issue		= ahci_qc_issue,
 
-	.irq_clear		= ata_noop_irq_clear,
-
-	.scr_read		= ahci_scr_read,
-	.scr_write		= ahci_scr_write,
-
 	.freeze			= ahci_freeze,
 	.thaw			= ahci_thaw,
-
 	.error_handler		= ahci_error_handler,
 	.post_internal_cmd	= ahci_post_internal_cmd,
-
-	.pmp_attach		= ahci_pmp_attach,
-	.pmp_detach		= ahci_pmp_detach,
-
-#ifdef CONFIG_PM
-	.port_suspend		= ahci_port_suspend,
-	.port_resume		= ahci_port_resume,
-#endif
-	.enable_pm		= ahci_enable_alpm,
-	.disable_pm		= ahci_disable_alpm,
-
-	.port_start		= ahci_port_start,
-	.port_stop		= ahci_port_stop,
-};
-
-static const struct ata_port_operations ahci_vt8251_ops = {
-	.check_status		= ahci_check_status,
-	.check_altstatus	= ahci_check_status,
-	.dev_select		= ata_noop_dev_select,
-
 	.dev_config		= ahci_dev_config,
 
-	.tf_read		= ahci_tf_read,
-
-	.qc_defer		= sata_pmp_qc_defer_cmd_switch,
-	.qc_prep		= ahci_qc_prep,
-	.qc_issue		= ahci_qc_issue,
-
-	.irq_clear		= ata_noop_irq_clear,
-
 	.scr_read		= ahci_scr_read,
 	.scr_write		= ahci_scr_write,
-
-	.freeze			= ahci_freeze,
-	.thaw			= ahci_thaw,
-
-	.error_handler		= ahci_vt8251_error_handler,
-	.post_internal_cmd	= ahci_post_internal_cmd,
-
 	.pmp_attach		= ahci_pmp_attach,
 	.pmp_detach		= ahci_pmp_detach,
 
+	.enable_pm		= ahci_enable_alpm,
+	.disable_pm		= ahci_disable_alpm,
 #ifdef CONFIG_PM
 	.port_suspend		= ahci_port_suspend,
 	.port_resume		= ahci_port_resume,
 #endif
-	.enable_pm		= ahci_enable_alpm,
-	.disable_pm		= ahci_disable_alpm,
-
 	.port_start		= ahci_port_start,
 	.port_stop		= ahci_port_stop,
 };
 
-static const struct ata_port_operations ahci_p5wdh_ops = {
-	.check_status		= ahci_check_status,
-	.check_altstatus	= ahci_check_status,
-	.dev_select		= ata_noop_dev_select,
-
-	.dev_config		= ahci_dev_config,
-
-	.tf_read		= ahci_tf_read,
-
-	.qc_defer		= sata_pmp_qc_defer_cmd_switch,
-	.qc_prep		= ahci_qc_prep,
-	.qc_issue		= ahci_qc_issue,
-
-	.irq_clear		= ata_noop_irq_clear,
-
-	.scr_read		= ahci_scr_read,
-	.scr_write		= ahci_scr_write,
-
-	.freeze			= ahci_freeze,
-	.thaw			= ahci_thaw,
+static struct ata_port_operations ahci_vt8251_ops = {
+	.inherits		= &ahci_ops,
+	.error_handler		= ahci_vt8251_error_handler,
+};
 
+static struct ata_port_operations ahci_p5wdh_ops = {
+	.inherits		= &ahci_ops,
 	.error_handler		= ahci_p5wdh_error_handler,
-	.post_internal_cmd	= ahci_post_internal_cmd,
-
-	.pmp_attach		= ahci_pmp_attach,
-	.pmp_detach		= ahci_pmp_detach,
-
-#ifdef CONFIG_PM
-	.port_suspend		= ahci_port_suspend,
-	.port_resume		= ahci_port_resume,
-#endif
-	.enable_pm		= ahci_enable_alpm,
-	.disable_pm		= ahci_disable_alpm,
-
-	.port_start		= ahci_port_start,
-	.port_stop		= ahci_port_stop,
 };
 
 #define AHCI_HFLAGS(flags)	.private_data	= (void *)(flags)
diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index 5c64ce134c6c..0b5b515ae159 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -99,36 +99,9 @@ static struct scsi_host_template generic_sht = {
 };
 
 static struct ata_port_operations generic_port_ops = {
-	.set_mode	= generic_set_mode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.data_xfer	= ata_data_xfer,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.inherits	= &ata_bmdma_port_ops,
 	.cable_detect	= ata_cable_unknown,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.set_mode	= generic_set_mode,
 };
 
 static int all_generic_ide;		/* Set to claim all devices */
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 9f887b2c92df..bb46b61a7c6b 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -294,155 +294,34 @@ static struct scsi_host_template piix_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations piix_pata_ops = {
+static struct ata_port_operations piix_pata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= piix_set_piomode,
 	.set_dmamode		= piix_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= piix_pata_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_40wire,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
+};
 
-	.port_start		= ata_sff_port_start,
+static struct ata_port_operations piix_vmw_ops = {
+	.inherits		= &piix_pata_ops,
+	.bmdma_status		= piix_vmw_bmdma_status,
 };
 
-static const struct ata_port_operations ich_pata_ops = {
-	.set_piomode		= piix_set_piomode,
-	.set_dmamode		= ich_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= piix_pata_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+static struct ata_port_operations ich_pata_ops = {
+	.inherits		= &piix_pata_ops,
 	.cable_detect		= ich_pata_cable_detect,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
+	.set_dmamode		= ich_set_dmamode,
 };
 
-static const struct ata_port_operations piix_sata_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
+static struct ata_port_operations piix_sata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 };
 
-static const struct ata_port_operations piix_vmw_ops = {
-	.set_piomode		= piix_set_piomode,
-	.set_dmamode		= piix_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= piix_vmw_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= piix_pata_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_40wire,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
-};
-
-static const struct ata_port_operations piix_sidpr_sata_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
+static struct ata_port_operations piix_sidpr_sata_ops = {
+	.inherits		= &piix_sata_ops,
 	.scr_read		= piix_sidpr_scr_read,
 	.scr_write		= piix_sidpr_scr_write,
-
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= piix_sidpr_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 static const struct piix_map_db ich5_map_db = {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 394edf937cf2..32fa9ee397b6 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -74,6 +74,56 @@ const unsigned long sata_deb_timing_normal[]		= {   5,  100, 2000 };
 const unsigned long sata_deb_timing_hotplug[]		= {  25,  500, 2000 };
 const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
+const struct ata_port_operations ata_base_port_ops = {
+	.irq_clear		= ata_noop_irq_clear,
+};
+
+const struct ata_port_operations sata_port_ops = {
+	.inherits		= &ata_base_port_ops,
+
+	.qc_defer		= ata_std_qc_defer,
+	.dev_select		= ata_noop_dev_select,
+};
+
+const struct ata_port_operations sata_pmp_port_ops = {
+	.inherits		= &sata_port_ops,
+};
+
+const struct ata_port_operations ata_sff_port_ops = {
+	.inherits		= &ata_base_port_ops,
+
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= ata_bmdma_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.dev_select		= ata_std_dev_select,
+	.check_status		= ata_check_status,
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.exec_command		= ata_exec_command,
+	.data_xfer		= ata_data_xfer,
+	.irq_on			= ata_irq_on,
+
+	.port_start		= ata_sff_port_start,
+	.irq_handler		= ata_interrupt,
+};
+
+const struct ata_port_operations ata_bmdma_port_ops = {
+	.inherits		= &ata_sff_port_ops,
+
+	.mode_filter		= ata_pci_default_filter,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.irq_clear		= ata_bmdma_irq_clear,
+};
+
 static unsigned int ata_dev_init_params(struct ata_device *dev,
 					u16 heads, u16 sectors);
 static unsigned int ata_dev_set_xfermode(struct ata_device *dev);
@@ -6971,6 +7021,56 @@ static void ata_host_stop(struct device *gendev, void *res)
 		host->ops->host_stop(host);
 }
 
+/**
+ *	ata_finalize_port_ops - finalize ata_port_operations
+ *	@ops: ata_port_operations to finalize
+ *
+ *	An ata_port_operations can inherit from another ops and that
+ *	ops can again inherit from another.  This can go on as many
+ *	times as necessary as long as there is no loop in the
+ *	inheritance chain.
+ *
+ *	Ops tables are finalized when the host is started.  NULL or
+ *	unspecified entries are inherited from the closet ancestor
+ *	which has the method and the entry is populated with it.
+ *	After finalization, the ops table directly points to all the
+ *	methods and ->inherits is no longer necessary and cleared.
+ *
+ *	Using ATA_OP_NULL, inheriting ops can force a method to NULL.
+ *
+ *	LOCKING:
+ *	None.
+ */
+static void ata_finalize_port_ops(struct ata_port_operations *ops)
+{
+	static spinlock_t lock = SPIN_LOCK_UNLOCKED;
+	const struct ata_port_operations *cur;
+	void **begin = (void **)ops;
+	void **end = (void **)&ops->inherits;
+	void **pp;
+
+	if (!ops || !ops->inherits)
+		return;
+
+	spin_lock(&lock);
+
+	for (cur = ops->inherits; cur; cur = cur->inherits) {
+		void **inherit = (void **)cur;
+
+		for (pp = begin; pp < end; pp++, inherit++)
+			if (!*pp)
+				*pp = *inherit;
+	}
+
+	for (pp = begin; pp < end; pp++)
+		if (IS_ERR(*pp))
+			*pp = NULL;
+
+	ops->inherits = NULL;
+
+	spin_unlock(&lock);
+}
+
 /**
  *	ata_host_start - start and freeze ports of an ATA host
  *	@host: ATA host to start ports for
@@ -6996,9 +7096,13 @@ int ata_host_start(struct ata_host *host)
 	if (host->flags & ATA_HOST_STARTED)
 		return 0;
 
+	ata_finalize_port_ops(host->ops);
+
 	for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap = host->ports[i];
 
+		ata_finalize_port_ops(ap->ops);
+
 		if (!host->ops && !ata_port_is_dummy(ap))
 			host->ops = ap->ops;
 
@@ -7060,7 +7164,7 @@ int ata_host_start(struct ata_host *host)
  */
 /* KILLME - the only user left is ipr */
 void ata_host_init(struct ata_host *host, struct device *dev,
-		   unsigned long flags, const struct ata_port_operations *ops)
+		   unsigned long flags, struct ata_port_operations *ops)
 {
 	spin_lock_init(&host->lock);
 	host->dev = dev;
@@ -7749,7 +7853,7 @@ static unsigned int ata_dummy_qc_issue(struct ata_queued_cmd *qc)
 	return AC_ERR_SYSTEM;
 }
 
-const struct ata_port_operations ata_dummy_port_ops = {
+struct ata_port_operations ata_dummy_port_ops = {
 	.check_status		= ata_dummy_check_status,
 	.check_altstatus	= ata_dummy_check_status,
 	.dev_select		= ata_noop_dev_select,
@@ -7777,6 +7881,11 @@ const struct ata_port_info ata_dummy_port_info = {
 EXPORT_SYMBOL_GPL(sata_deb_timing_normal);
 EXPORT_SYMBOL_GPL(sata_deb_timing_hotplug);
 EXPORT_SYMBOL_GPL(sata_deb_timing_long);
+EXPORT_SYMBOL_GPL(ata_base_port_ops);
+EXPORT_SYMBOL_GPL(sata_port_ops);
+EXPORT_SYMBOL_GPL(sata_pmp_port_ops);
+EXPORT_SYMBOL_GPL(ata_sff_port_ops);
+EXPORT_SYMBOL_GPL(ata_bmdma_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_info);
 EXPORT_SYMBOL_GPL(ata_std_bios_param);
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index 187545c0898a..35ad488db6ed 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -235,39 +235,14 @@ static struct scsi_host_template pacpi_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations pacpi_ops = {
+static struct ata_port_operations pacpi_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.qc_issue		= pacpi_qc_issue_prot,
+	.cable_detect		= pacpi_cable_detect,
+	.mode_filter		= pacpi_mode_filter,
 	.set_piomode		= pacpi_set_piomode,
 	.set_dmamode		= pacpi_set_dmamode,
-	.mode_filter		= pacpi_mode_filter,
-
-	/* Task file is PCI ATA format, use helpers */
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= pacpi_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= pacpi_cable_detect,
-
-	/* BMDMA handling is PCI ATA format, use helpers */
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= pacpi_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	/* Timeout handling */
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	/* Generic PATA PCI ATA helpers */
 	.port_start		= pacpi_port_start,
 };
 
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index f3d6d9b345ba..b00a9cf72c31 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -347,29 +347,15 @@ static struct scsi_host_template ali_sht = {
  */
 
 static struct ata_port_operations ali_early_port_ops = {
-	.set_piomode	= ali_set_piomode,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.inherits	= &ata_sff_port_ops,
 	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= ali_set_piomode,
+};
 
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+static const struct ata_port_operations ali_dma_base_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.set_piomode	= ali_set_piomode,
+	.set_dmamode	= ali_set_dmamode,
 };
 
 /*
@@ -377,115 +363,31 @@ static struct ata_port_operations ali_early_port_ops = {
  *	detect
  */
 static struct ata_port_operations ali_20_port_ops = {
-	.set_piomode	= ali_set_piomode,
-	.set_dmamode	= ali_set_dmamode,
+	.inherits	= &ali_dma_base_ops,
+	.cable_detect	= ata_cable_40wire,
 	.mode_filter	= ali_20_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
 	.check_atapi_dma = ali_check_atapi_dma,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
 	.dev_config	= ali_lock_sectors,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /*
  *	Port operations for DMA capable ALi with cable detect
  */
 static struct ata_port_operations ali_c2_port_ops = {
-	.set_piomode	= ali_set_piomode,
-	.set_dmamode	= ali_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
+	.inherits	= &ali_dma_base_ops,
 	.check_atapi_dma = ali_check_atapi_dma,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-	.dev_config	= ali_lock_sectors,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
 	.cable_detect	= ali_c2_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.dev_config	= ali_lock_sectors,
 };
 
 /*
  *	Port operations for DMA capable ALi with cable detect and LBA48
  */
 static struct ata_port_operations ali_c5_port_ops = {
-	.set_piomode	= ali_set_piomode,
-	.set_dmamode	= ali_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
+	.inherits	= &ali_dma_base_ops,
 	.check_atapi_dma = ali_check_atapi_dma,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
 	.dev_config	= ali_warn_atapi_dma,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
 	.cable_detect	= ali_c2_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 90d786dfbec3..b0cb4eaf273c 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -356,204 +356,57 @@ static struct scsi_host_template amd_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
+static const struct ata_port_operations amd_base_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.error_handler	= amd_error_handler,
+};
+
 static struct ata_port_operations amd33_port_ops = {
+	.inherits	= &amd_base_port_ops,
+	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= amd33_set_piomode,
 	.set_dmamode	= amd33_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= amd_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations amd66_port_ops = {
+	.inherits	= &amd_base_port_ops,
+	.cable_detect	= ata_cable_unknown,
 	.set_piomode	= amd66_set_piomode,
 	.set_dmamode	= amd66_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= amd_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_unknown,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations amd100_port_ops = {
+	.inherits	= &amd_base_port_ops,
+	.cable_detect	= ata_cable_unknown,
 	.set_piomode	= amd100_set_piomode,
 	.set_dmamode	= amd100_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= amd_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_unknown,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations amd133_port_ops = {
+	.inherits	= &amd_base_port_ops,
+	.cable_detect	= amd_cable_detect,
 	.set_piomode	= amd133_set_piomode,
 	.set_dmamode	= amd133_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= amd_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= amd_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
+};
 
-	.port_start	= ata_sff_port_start,
+static const struct ata_port_operations nv_base_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= ata_cable_ignore,
+	.mode_filter	= nv_mode_filter,
+	.error_handler	= nv_error_handler,
+	.host_stop	= nv_host_stop,
 };
 
 static struct ata_port_operations nv100_port_ops = {
+	.inherits	= &nv_base_port_ops,
 	.set_piomode	= nv100_set_piomode,
 	.set_dmamode	= nv100_set_dmamode,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= nv_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_ignore,
-	.mode_filter	= nv_mode_filter,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
-	.host_stop	= nv_host_stop,
 };
 
 static struct ata_port_operations nv133_port_ops = {
+	.inherits	= &nv_base_port_ops,
 	.set_piomode	= nv133_set_piomode,
 	.set_dmamode	= nv133_set_dmamode,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= nv_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_ignore,
-	.mode_filter	= nv_mode_filter,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
-	.host_stop	= nv_host_stop,
 };
 
 static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index 7bfb7e8bdca2..0101e5aef3e0 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -317,69 +317,20 @@ static struct scsi_host_template artop_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations artop6210_ops = {
+static struct ata_port_operations artop6210_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= artop6210_set_piomode,
 	.set_dmamode		= artop6210_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= artop6210_error_handler,
-	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_40wire,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations artop6260_ops = {
+static struct ata_port_operations artop6260_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.cable_detect		= artop6260_cable_detect,
 	.set_piomode		= artop6260_set_piomode,
 	.set_dmamode		= artop6260_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= artop6260_error_handler,
-	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= artop6260_cable_detect,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_at32.c b/drivers/ata/pata_at32.c
index d7b7b7fde362..528315587532 100644
--- a/drivers/ata/pata_at32.c
+++ b/drivers/ata/pata_at32.c
@@ -171,28 +171,9 @@ static struct scsi_host_template at32_sht = {
 };
 
 static struct ata_port_operations at32_port_ops = {
-	.set_piomode		= pata_at32_set_piomode,
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+	.inherits		= &ata_sff_port_ops,
 	.cable_detect		= ata_cable_40wire,
-
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
-	.data_xfer		= ata_data_xfer,
-
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
+	.set_piomode		= pata_at32_set_piomode,
 };
 
 static int __init pata_at32_init_one(struct device *dev,
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 645c47271ff5..2655f6a17ad3 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -226,36 +226,16 @@ static struct scsi_host_template atiixp_sht = {
 };
 
 static struct ata_port_operations atiixp_port_ops = {
-	.set_piomode	= atiixp_set_piomode,
-	.set_dmamode	= atiixp_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= atiixp_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= atiixp_cable_detect,
+	.inherits	= &ata_bmdma_port_ops,
 
-	.bmdma_setup 	= ata_bmdma_setup,
+	.qc_prep 	= ata_dumb_qc_prep,
 	.bmdma_start 	= atiixp_bmdma_start,
 	.bmdma_stop	= atiixp_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_dumb_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
 
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= atiixp_cable_detect,
+	.set_piomode	= atiixp_set_piomode,
+	.set_dmamode	= atiixp_set_dmamode,
+	.error_handler	= atiixp_error_handler,
 };
 
 static int atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 6c75fcac3cf4..7a22ef483061 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1363,6 +1363,8 @@ static struct scsi_host_template bfin_sht = {
 };
 
 static const struct ata_port_operations bfin_pata_ops = {
+	.inherits		= &ata_sff_port_ops,
+
 	.set_piomode		= bfin_set_piomode,
 	.set_dmamode		= bfin_set_dmamode,
 
@@ -1380,14 +1382,12 @@ static const struct ata_port_operations bfin_pata_ops = {
 	.data_xfer		= bfin_data_xfer,
 
 	.qc_prep		= ata_noop_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
 
 	.freeze			= bfin_bmdma_freeze,
 	.thaw			= bfin_bmdma_thaw,
 	.error_handler		= bfin_error_handler,
 	.post_internal_cmd	= bfin_bmdma_stop,
 
-	.irq_handler		= ata_interrupt,
 	.irq_clear		= bfin_irq_clear,
 	.irq_on			= bfin_irq_on,
 
diff --git a/drivers/ata/pata_cmd640.c b/drivers/ata/pata_cmd640.c
index 26562b814400..061c891c8a66 100644
--- a/drivers/ata/pata_cmd640.c
+++ b/drivers/ata/pata_cmd640.c
@@ -170,35 +170,12 @@ static struct scsi_host_template cmd640_sht = {
 };
 
 static struct ata_port_operations cmd640_port_ops = {
-	.set_piomode	= cmd640_set_piomode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= cmd640_qc_issue_prot,
-
-	/* In theory this is not needed once we kill the prefetcher */
+	.inherits	= &ata_bmdma_port_ops,
+	/* In theory xfer_noirq is not needed once we kill the prefetcher */
 	.data_xfer	= ata_data_xfer_noirq,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
+	.qc_issue	= cmd640_qc_issue_prot,
+	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= cmd640_set_piomode,
 	.port_start	= cmd640_port_start,
 };
 
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index 6aea05cc0940..1ac8ecfb97e2 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -269,103 +269,27 @@ static struct scsi_host_template cmd64x_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static struct ata_port_operations cmd64x_port_ops = {
+static const struct ata_port_operations cmd64x_base_ops = {
+	.inherits	= &ata_bmdma_port_ops,
 	.set_piomode	= cmd64x_set_piomode,
 	.set_dmamode	= cmd64x_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
-static struct ata_port_operations cmd646r1_port_ops = {
-	.set_piomode	= cmd64x_set_piomode,
-	.set_dmamode	= cmd64x_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+static struct ata_port_operations cmd64x_port_ops = {
+	.inherits	= &cmd64x_base_ops,
 	.cable_detect	= ata_cable_40wire,
+};
 
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
+static struct ata_port_operations cmd646r1_port_ops = {
+	.inherits	= &cmd64x_base_ops,
 	.bmdma_stop	= cmd646r1_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
 };
 
 static struct ata_port_operations cmd648_port_ops = {
-	.set_piomode	= cmd64x_set_piomode,
-	.set_dmamode	= cmd64x_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= cmd648_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
+	.inherits	= &cmd64x_base_ops,
 	.bmdma_stop	= cmd648_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= cmd648_cable_detect,
 };
 
 static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_cs5520.c b/drivers/ata/pata_cs5520.c
index 7e643099a444..46d0ce32ee5a 100644
--- a/drivers/ata/pata_cs5520.c
+++ b/drivers/ata/pata_cs5520.c
@@ -145,34 +145,11 @@ static struct scsi_host_template cs5520_sht = {
 };
 
 static struct ata_port_operations cs5520_port_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.qc_prep		= ata_dumb_qc_prep,
+	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= cs5520_set_piomode,
 	.set_dmamode		= cs5520_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_40wire,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_dumb_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 static int __devinit cs5520_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
index 0bb03dabcf18..e4a16a578cac 100644
--- a/drivers/ata/pata_cs5530.c
+++ b/drivers/ata/pata_cs5530.c
@@ -166,37 +166,14 @@ static struct scsi_host_template cs5530_sht = {
 };
 
 static struct ata_port_operations cs5530_port_ops = {
-	.set_piomode	= cs5530_set_piomode,
-	.set_dmamode	= cs5530_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
+	.inherits	= &ata_bmdma_port_ops,
 
 	.qc_prep 	= ata_dumb_qc_prep,
 	.qc_issue	= cs5530_qc_issue_prot,
 
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= cs5530_set_piomode,
+	.set_dmamode	= cs5530_set_dmamode,
 };
 
 static const struct dmi_system_id palmax_dmi_table[] = {
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
index 48a18349c1d8..f910a8aa7437 100644
--- a/drivers/ata/pata_cs5535.c
+++ b/drivers/ata/pata_cs5535.c
@@ -162,37 +162,10 @@ static struct scsi_host_template cs5535_sht = {
 };
 
 static struct ata_port_operations cs5535_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= cs5535_cable_detect,
 	.set_piomode	= cs5535_set_piomode,
 	.set_dmamode	= cs5535_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= cs5535_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c
index f02d9107ef3b..075ee6a7be39 100644
--- a/drivers/ata/pata_cs5536.c
+++ b/drivers/ata/pata_cs5536.c
@@ -225,37 +225,10 @@ static struct scsi_host_template cs5536_sht = {
 };
 
 static struct ata_port_operations cs5536_port_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.cable_detect		= cs5536_cable_detect,
 	.set_piomode		= cs5536_set_piomode,
 	.set_dmamode		= cs5536_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= cs5536_cable_detect,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
index 07fa1ab36315..c459553e7d1e 100644
--- a/drivers/ata/pata_cypress.c
+++ b/drivers/ata/pata_cypress.c
@@ -114,37 +114,10 @@ static struct scsi_host_template cy82c693_sht = {
 };
 
 static struct ata_port_operations cy82c693_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= cy82c693_set_piomode,
 	.set_dmamode	= cy82c693_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static int cy82c693_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index 8700d9dcd8c9..ef62fc642c17 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -236,36 +236,12 @@ static struct scsi_host_template efar_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations efar_ops = {
+static struct ata_port_operations efar_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.cable_detect		= efar_cable_detect,
 	.set_piomode		= efar_set_piomode,
 	.set_dmamode		= efar_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= efar_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= efar_cable_detect,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
index a30028de41c0..788955f57ff8 100644
--- a/drivers/ata/pata_hpt366.c
+++ b/drivers/ata/pata_hpt366.c
@@ -298,37 +298,11 @@ static struct scsi_host_template hpt36x_sht = {
  */
 
 static struct ata_port_operations hpt366_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= hpt36x_cable_detect,
+	.mode_filter	= hpt366_filter,
 	.set_piomode	= hpt366_set_piomode,
 	.set_dmamode	= hpt366_set_dmamode,
-	.mode_filter	= hpt366_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= hpt36x_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
index 7d6fac43e2f9..c42eec70d297 100644
--- a/drivers/ata/pata_hpt37x.c
+++ b/drivers/ata/pata_hpt37x.c
@@ -627,36 +627,15 @@ static struct scsi_host_template hpt37x_sht = {
  */
 
 static struct ata_port_operations hpt370_port_ops = {
-	.set_piomode	= hpt370_set_piomode,
-	.set_dmamode	= hpt370_set_dmamode,
-	.mode_filter	= hpt370_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= hpt37x_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.inherits	= &ata_bmdma_port_ops,
 
-	.bmdma_setup 	= ata_bmdma_setup,
 	.bmdma_start 	= hpt370_bmdma_start,
 	.bmdma_stop	= hpt370_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
 
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.mode_filter	= hpt370_filter,
+	.set_piomode	= hpt370_set_piomode,
+	.set_dmamode	= hpt370_set_dmamode,
+	.error_handler	= hpt37x_error_handler,
 };
 
 /*
@@ -664,36 +643,8 @@ static struct ata_port_operations hpt370_port_ops = {
  */
 
 static struct ata_port_operations hpt370a_port_ops = {
-	.set_piomode	= hpt370_set_piomode,
-	.set_dmamode	= hpt370_set_dmamode,
+	.inherits	= &hpt370_port_ops,
 	.mode_filter	= hpt370a_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= hpt37x_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= hpt370_bmdma_start,
-	.bmdma_stop	= hpt370_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /*
@@ -702,36 +653,13 @@ static struct ata_port_operations hpt370a_port_ops = {
  */
 
 static struct ata_port_operations hpt372_port_ops = {
-	.set_piomode	= hpt372_set_piomode,
-	.set_dmamode	= hpt372_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= hpt37x_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.inherits	= &ata_bmdma_port_ops,
 
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
 	.bmdma_stop	= hpt37x_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
 
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.set_piomode	= hpt372_set_piomode,
+	.set_dmamode	= hpt372_set_dmamode,
+	.error_handler	= hpt37x_error_handler,
 };
 
 /*
@@ -740,36 +668,8 @@ static struct ata_port_operations hpt372_port_ops = {
  */
 
 static struct ata_port_operations hpt374_port_ops = {
-	.set_piomode	= hpt372_set_piomode,
-	.set_dmamode	= hpt372_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
+	.inherits	= &hpt372_port_ops,
 	.error_handler	= hpt374_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= hpt37x_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
index aa380c46b168..b77b1279d757 100644
--- a/drivers/ata/pata_hpt3x2n.c
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -347,37 +347,15 @@ static struct scsi_host_template hpt3x2n_sht = {
  */
 
 static struct ata_port_operations hpt3x2n_port_ops = {
-	.set_piomode	= hpt3x2n_set_piomode,
-	.set_dmamode	= hpt3x2n_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
+	.inherits	= &ata_bmdma_port_ops,
 
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= hpt3x2n_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= hpt3x2n_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
 	.bmdma_stop	= hpt3x2n_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= hpt3x2n_qc_issue_prot,
 
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= hpt3x2n_cable_detect,
+	.set_piomode	= hpt3x2n_set_piomode,
+	.set_dmamode	= hpt3x2n_set_dmamode,
+	.error_handler	= hpt3x2n_error_handler,
 };
 
 /**
diff --git a/drivers/ata/pata_hpt3x3.c b/drivers/ata/pata_hpt3x3.c
index 9837ab0181e8..8857d029ac2e 100644
--- a/drivers/ata/pata_hpt3x3.c
+++ b/drivers/ata/pata_hpt3x3.c
@@ -106,40 +106,13 @@ static struct scsi_host_template hpt3x3_sht = {
 };
 
 static struct ata_port_operations hpt3x3_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.check_atapi_dma= hpt3x3_atapi_dma,
+	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= hpt3x3_set_piomode,
 #if defined(CONFIG_PATA_HPT3X3_DMA)
 	.set_dmamode	= hpt3x3_set_dmamode,
 #endif
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-	.check_atapi_dma= hpt3x3_atapi_dma,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_icside.c b/drivers/ata/pata_icside.c
index 88a1c7ae0a4d..ff16b0eaa2c2 100644
--- a/drivers/ata/pata_icside.c
+++ b/drivers/ata/pata_icside.c
@@ -339,35 +339,19 @@ static void pata_icside_error_handler(struct ata_port *ap)
 }
 
 static struct ata_port_operations pata_icside_port_ops = {
-	.set_dmamode		= pata_icside_set_dmamode,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-
-	.cable_detect		= ata_cable_40wire,
-
-	.bmdma_setup		= pata_icside_bmdma_setup,
-	.bmdma_start		= pata_icside_bmdma_start,
-
-	.data_xfer		= ata_data_xfer_noirq,
-
+	.inherits		= &ata_sff_port_ops,
 	/* no need to build any PRD tables for DMA */
 	.qc_prep		= ata_noop_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_data_xfer_noirq,
+	.bmdma_setup		= pata_icside_bmdma_setup,
+	.bmdma_start		= pata_icside_bmdma_start,
+	.bmdma_stop		= pata_icside_bmdma_stop,
+	.bmdma_status		= pata_icside_bmdma_status,
 
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
+	.cable_detect		= ata_cable_40wire,
+	.set_dmamode		= pata_icside_set_dmamode,
 	.error_handler		= pata_icside_error_handler,
 	.post_internal_cmd	= pata_icside_bmdma_stop,
-
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.bmdma_stop		= pata_icside_bmdma_stop,
-	.bmdma_status		= pata_icside_bmdma_status,
 };
 
 static void __devinit
diff --git a/drivers/ata/pata_isapnp.c b/drivers/ata/pata_isapnp.c
index 91ca4d50db04..085913ec6f68 100644
--- a/drivers/ata/pata_isapnp.c
+++ b/drivers/ata/pata_isapnp.c
@@ -24,27 +24,8 @@ static struct scsi_host_template isapnp_sht = {
 };
 
 static struct ata_port_operations isapnp_port_ops = {
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.inherits	= &ata_sff_port_ops,
 	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_it8213.c b/drivers/ata/pata_it8213.c
index 678a05b304d8..9ce89522e764 100644
--- a/drivers/ata/pata_it8213.c
+++ b/drivers/ata/pata_it8213.c
@@ -246,36 +246,13 @@ static struct scsi_host_template it8213_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations it8213_ops = {
+
+static struct ata_port_operations it8213_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.cable_detect		= it8213_cable_detect,
 	.set_piomode		= it8213_set_piomode,
 	.set_dmamode		= it8213_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= it8213_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= it8213_cable_detect,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index 7d969c911731..669d224d30ca 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -636,71 +636,30 @@ static struct scsi_host_template it821x_sht = {
 };
 
 static struct ata_port_operations it821x_smart_port_ops = {
-	.set_mode	= it821x_smart_set_mode,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.mode_filter	= ata_pci_default_filter,
+	.inherits	= &ata_bmdma_port_ops,
 
-	.check_status 	= ata_check_status,
 	.check_atapi_dma= it821x_check_atapi_dma,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-	.dev_config	= it821x_dev_config,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= it821x_ident_hack,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= it821x_smart_qc_issue_prot,
 
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
+	.cable_detect	= it821x_ident_hack,
+	.set_mode	= it821x_smart_set_mode,
+	.dev_config	= it821x_dev_config,
 
 	.port_start	= it821x_port_start,
 };
 
 static struct ata_port_operations it821x_passthru_port_ops = {
-	.set_piomode	= it821x_passthru_set_piomode,
-	.set_dmamode	= it821x_passthru_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
+	.inherits	= &ata_bmdma_port_ops,
 
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
 	.check_atapi_dma= it821x_check_atapi_dma,
 	.dev_select 	= it821x_passthru_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_unknown,
-
-	.bmdma_setup 	= ata_bmdma_setup,
 	.bmdma_start 	= it821x_passthru_bmdma_start,
 	.bmdma_stop	= it821x_passthru_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= it821x_passthru_qc_issue_prot,
 
-	.data_xfer	= ata_data_xfer,
-
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_handler	= ata_interrupt,
-	.irq_on		= ata_irq_on,
+	.cable_detect	= ata_cable_unknown,
+	.set_piomode	= it821x_passthru_set_piomode,
+	.set_dmamode	= it821x_passthru_set_dmamode,
 
 	.port_start	= it821x_port_start,
 };
diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c
index b7e8e825a869..d02629aa20da 100644
--- a/drivers/ata/pata_ixp4xx_cf.c
+++ b/drivers/ata/pata_ixp4xx_cf.c
@@ -92,29 +92,10 @@ static struct scsi_host_template ixp4xx_sht = {
 };
 
 static struct ata_port_operations ixp4xx_port_ops = {
-	.set_mode		= ixp4xx_set_mode,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status 		= ata_check_status,
-	.dev_select 		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.qc_prep 		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
+	.inherits		= &ata_sff_port_ops,
 	.data_xfer		= ixp4xx_mmio_data_xfer,
 	.cable_detect		= ata_cable_40wire,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
+	.set_mode		= ixp4xx_set_mode,
 };
 
 static void ixp4xx_setup_port(struct ata_port *ap,
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index 69781af7b1bb..61ff5c6b4568 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -125,36 +125,9 @@ static struct scsi_host_template jmicron_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations jmicron_ops = {
-	/* Task file is PCI ATA format, use helpers */
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
+static struct ata_port_operations jmicron_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.error_handler		= jmicron_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	/* BMDMA handling is PCI ATA format, use helpers */
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	/* IRQ-related hooks */
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	/* Generic PATA PCI ATA helpers */
-	.port_start		= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
index 5329b954c5f2..2474068596f4 100644
--- a/drivers/ata/pata_legacy.c
+++ b/drivers/ata/pata_legacy.c
@@ -211,6 +211,11 @@ static struct scsi_host_template legacy_sht = {
 	ATA_PIO_SHT(DRV_NAME),
 };
 
+static const struct ata_port_operations legacy_base_port_ops = {
+	.inherits	= &ata_sff_port_ops,
+	.cable_detect	= ata_cable_40wire,
+};
+
 /*
  *	These ops are used if the user indicates the hardware
  *	snoops the commands to decide on the mode and handles the
@@ -220,55 +225,14 @@ static struct scsi_host_template legacy_sht = {
  */
 
 static struct ata_port_operations simple_port_ops = {
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
+	.inherits	= &legacy_base_port_ops,
 	.data_xfer	= ata_data_xfer_noirq,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations legacy_port_ops = {
-	.set_mode	= legacy_set_mode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-	.cable_detect	= ata_cable_40wire,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
+	.inherits	= &legacy_base_port_ops,
 	.data_xfer	= ata_data_xfer_noirq,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.set_mode	= legacy_set_mode,
 };
 
 /*
@@ -359,30 +323,9 @@ static unsigned int pdc_data_xfer_vlb(struct ata_device *dev,
 }
 
 static struct ata_port_operations pdc20230_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= pdc20230_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
 	.data_xfer	= pdc_data_xfer_vlb,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /*
@@ -413,30 +356,8 @@ static void ht6560a_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct ata_port_operations ht6560a_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= ht6560a_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,	/* Check vlb/noirq */
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /*
@@ -478,30 +399,8 @@ static void ht6560b_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 static struct ata_port_operations ht6560b_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= ht6560b_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,    /* FIXME: Check 32bit and noirq */
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /*
@@ -599,30 +498,8 @@ static void opti82c611a_set_piomode(struct ata_port *ap,
 
 
 static struct ata_port_operations opti82c611a_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= opti82c611a_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /*
@@ -731,30 +608,9 @@ static unsigned int opti82c46x_qc_issue_prot(struct ata_queued_cmd *qc)
 }
 
 static struct ata_port_operations opti82c46x_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= opti82c46x_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= opti82c46x_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static void qdi6500_set_piomode(struct ata_port *ap, struct ata_device *adev)
@@ -916,84 +772,22 @@ static int qdi_port(struct platform_device *dev,
 }
 
 static struct ata_port_operations qdi6500_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= qdi6500_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= qdi_qc_issue_prot,
-
 	.data_xfer	= vlb32_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations qdi6580_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= qdi6580_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
 	.data_xfer	= vlb32_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations qdi6580dp_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= qdi6580dp_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= qdi_qc_issue_prot,
-
 	.data_xfer	= vlb32_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static DEFINE_SPINLOCK(winbond_lock);
@@ -1062,29 +856,9 @@ static int winbond_port(struct platform_device *dev,
 }
 
 static struct ata_port_operations winbond_port_ops = {
+	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= winbond_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
 	.data_xfer	= vlb32_data_xfer,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct legacy_controller controllers[] = {
diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c
index 9de6e429d0d1..286310fc5910 100644
--- a/drivers/ata/pata_marvell.c
+++ b/drivers/ata/pata_marvell.c
@@ -95,37 +95,10 @@ static struct scsi_host_template marvell_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations marvell_ops = {
-	/* Task file is PCI ATA format, use helpers */
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= marvell_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+static struct ata_port_operations marvell_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.cable_detect		= marvell_cable_detect,
-
-	/* BMDMA handling is PCI ATA format, use helpers */
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	/* Timeout handling */
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	/* Generic PATA PCI ATA helpers */
-	.port_start		= ata_sff_port_start,
+	.error_handler		= marvell_error_handler,
 };
 
 
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index 4117b618a9d9..ac7c0822b1a7 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -269,22 +269,12 @@ static struct scsi_host_template mpc52xx_ata_sht = {
 };
 
 static struct ata_port_operations mpc52xx_ata_port_ops = {
-	.set_piomode		= mpc52xx_ata_set_piomode,
+	.inherits		= &ata_sff_port_ops,
 	.dev_select		= mpc52xx_ata_dev_select,
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= mpc52xx_ata_error_handler,
 	.cable_detect		= ata_cable_40wire,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-	.port_start		= ata_sff_port_start,
+	.set_piomode		= mpc52xx_ata_set_piomode,
+	.error_handler		= mpc52xx_ata_error_handler,
+	.post_internal_cmd	= ATA_OP_NULL,
 };
 
 static int __devinit
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index e8e6837110b4..dab54f8a272d 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -155,28 +155,11 @@ static struct scsi_host_template mpiix_sht = {
 };
 
 static struct ata_port_operations mpiix_port_ops = {
+	.inherits	= &ata_sff_port_ops,
+	.qc_issue	= mpiix_qc_issue_prot,
+	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= mpiix_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
 	.error_handler	= mpiix_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= mpiix_qc_issue_prot,
-	.data_xfer	= ata_data_xfer,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static int mpiix_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
index 11f200a2a156..65389d1837b3 100644
--- a/drivers/ata/pata_netcell.c
+++ b/drivers/ata/pata_netcell.c
@@ -24,37 +24,9 @@ static struct scsi_host_template netcell_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations netcell_ops = {
-	/* Task file is PCI ATA format, use helpers */
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+static struct ata_port_operations netcell_ops = {
+	.inherits	= &ata_bmdma_port_ops,
 	.cable_detect		= ata_cable_80wire,
-
-	/* BMDMA handling is PCI ATA format, use helpers */
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	/* IRQ-related hooks */
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	/* Generic PATA PCI ATA helpers */
-	.port_start		= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_ninja32.c b/drivers/ata/pata_ninja32.c
index ce3b07cab8bc..8213d081f313 100644
--- a/drivers/ata/pata_ninja32.c
+++ b/drivers/ata/pata_ninja32.c
@@ -83,36 +83,10 @@ static struct scsi_host_template ninja32_sht = {
 };
 
 static struct ata_port_operations ninja32_port_ops = {
-	.set_piomode	= ninja32_set_piomode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
+	.inherits	= &ata_bmdma_port_ops,
 	.dev_select 	= ninja32_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
 	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.set_piomode	= ninja32_set_piomode,
 };
 
 static int ninja32_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index d2f85f107d15..5b1982fa0be1 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -148,30 +148,11 @@ static struct scsi_host_template ns87410_sht = {
 };
 
 static struct ata_port_operations ns87410_port_ops = {
+	.inherits	= &ata_sff_port_ops,
+	.qc_issue	= ns87410_qc_issue_prot,
+	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= ns87410_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
 	.error_handler	= ns87410_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ns87410_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static int ns87410_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_ns87415.c b/drivers/ata/pata_ns87415.c
index 78d634423cbf..38d86a262dbb 100644
--- a/drivers/ata/pata_ns87415.c
+++ b/drivers/ata/pata_ns87415.c
@@ -297,73 +297,29 @@ static u8 ns87560_bmdma_status(struct ata_port *ap)
 {
 	return ns87560_read_buggy(ap->ioaddr.bmdma_addr + ATA_DMA_STATUS);
 }
+#endif		/* 87560 SuperIO Support */
 
-static const struct ata_port_operations ns87560_pata_ops = {
-	.set_piomode		= ns87415_set_piomode,
-	.mode_filter		= ata_pci_default_filter,
+static struct ata_port_operations ns87415_pata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 
-	.tf_load		= ata_tf_load,
-	.tf_read		= ns87560_tf_read,
-	.check_status		= ns87560_check_status,
 	.check_atapi_dma	= ns87415_check_atapi_dma,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_40wire,
-
 	.bmdma_setup		= ns87415_bmdma_setup,
 	.bmdma_start		= ns87415_bmdma_start,
 	.bmdma_stop		= ns87415_bmdma_stop,
-	.bmdma_status		= ns87560_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
 	.irq_clear		= ns87415_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
-};
-
-#endif		/* 87560 SuperIO Support */
-
-
-static const struct ata_port_operations ns87415_pata_ops = {
-	.set_piomode		= ns87415_set_piomode,
-	.mode_filter		= ata_pci_default_filter,
 
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.check_atapi_dma	= ns87415_check_atapi_dma,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
 	.cable_detect		= ata_cable_40wire,
+	.set_piomode		= ns87415_set_piomode,
+};
 
-	.bmdma_setup		= ns87415_bmdma_setup,
-	.bmdma_start		= ns87415_bmdma_start,
-	.bmdma_stop		= ns87415_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ns87415_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
+#if defined(CONFIG_SUPERIO)
+static struct ata_port_operations ns87560_pata_ops = {
+	.inherits		= &ns87415_pata_ops,
+	.tf_read		= ns87560_tf_read,
+	.check_status		= ns87560_check_status,
+	.bmdma_status		= ns87560_bmdma_status,
 };
+#endif
 
 static struct scsi_host_template ns87415_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index 45f9b3eb5b45..f6062b37310d 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -223,36 +223,13 @@ static struct scsi_host_template oldpiix_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations oldpiix_pata_ops = {
+static struct ata_port_operations oldpiix_pata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.qc_issue		= oldpiix_qc_issue_prot,
+	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= oldpiix_set_piomode,
 	.set_dmamode		= oldpiix_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= oldpiix_pata_error_handler,
-	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_40wire,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= oldpiix_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index 1deacfa0be07..c4a0795c3ff4 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -169,29 +169,10 @@ static struct scsi_host_template opti_sht = {
 };
 
 static struct ata_port_operations opti_port_ops = {
+	.inherits	= &ata_sff_port_ops,
+	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= opti_set_piomode,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
 	.error_handler	= opti_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static int opti_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index 7495758a86fe..eb4b08190e3a 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -354,73 +354,18 @@ static struct scsi_host_template optidma_sht = {
 };
 
 static struct ata_port_operations optidma_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= optidma_set_pio_mode,
 	.set_dmamode	= optidma_set_dma_mode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.error_handler	= optidma_error_handler,
 	.set_mode	= optidma_set_mode,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.error_handler	= optidma_error_handler,
 };
 
 static struct ata_port_operations optiplus_port_ops = {
+	.inherits	= &optidma_port_ops,
 	.set_piomode	= optiplus_set_pio_mode,
 	.set_dmamode	= optiplus_set_dma_mode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.error_handler	= optidma_error_handler,
-	.set_mode	= optidma_set_mode,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index c05b36c94d51..57efbf05c95f 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -132,53 +132,17 @@ static struct scsi_host_template pcmcia_sht = {
 };
 
 static struct ata_port_operations pcmcia_port_ops = {
-	.set_mode	= pcmcia_set_mode,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
+	.inherits	= &ata_sff_port_ops,
 	.data_xfer	= ata_data_xfer_noirq,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_mode	= pcmcia_set_mode,
 };
 
 static struct ata_port_operations pcmcia_8bit_port_ops = {
-	.set_mode	= pcmcia_set_mode_8bit,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
+	.inherits	= &ata_sff_port_ops,
 	.data_xfer	= ata_data_xfer_8bit,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_mode	= pcmcia_set_mode_8bit,
 };
 
 #define CS_CHECK(fn, ret) \
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
index 229d9acd934a..f619c20dd192 100644
--- a/drivers/ata/pata_pdc2027x.c
+++ b/drivers/ata/pata_pdc2027x.c
@@ -133,66 +133,18 @@ static struct scsi_host_template pdc2027x_sht = {
 };
 
 static struct ata_port_operations pdc2027x_pata100_ops = {
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
+	.inherits		= &ata_bmdma_port_ops,
 	.check_atapi_dma	= pdc2027x_check_atapi_dma,
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= pdc2027x_error_handler,
-	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
 	.cable_detect		= pdc2027x_cable_detect,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
+	.error_handler		= pdc2027x_error_handler,
 };
 
 static struct ata_port_operations pdc2027x_pata133_ops = {
+	.inherits		= &pdc2027x_pata100_ops,
+	.mode_filter		= pdc2027x_mode_filter,
 	.set_piomode		= pdc2027x_set_piomode,
 	.set_dmamode		= pdc2027x_set_dmamode,
 	.set_mode		= pdc2027x_set_mode,
-	.mode_filter		= pdc2027x_mode_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.check_atapi_dma	= pdc2027x_check_atapi_dma,
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= pdc2027x_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= pdc2027x_cable_detect,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 static struct ata_port_info pdc2027x_port_info[] = {
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
index 564ee0798ec1..4daac20df0bc 100644
--- a/drivers/ata/pata_pdc202xx_old.c
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -266,69 +266,24 @@ static struct scsi_host_template pdc202xx_sht = {
 };
 
 static struct ata_port_operations pdc2024x_port_ops = {
-	.set_piomode	= pdc202xx_set_piomode,
-	.set_dmamode	= pdc202xx_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.inherits		= &ata_bmdma_port_ops,
+
+	.cable_detect		= ata_cable_40wire,
+	.set_piomode		= pdc202xx_set_piomode,
+	.set_dmamode		= pdc202xx_set_dmamode,
 };
 
 static struct ata_port_operations pdc2026x_port_ops = {
-	.set_piomode	= pdc202xx_set_piomode,
-	.set_dmamode	= pdc202xx_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-	.dev_config	= pdc2026x_dev_config,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= pdc2026x_cable_detect,
-
-	.check_atapi_dma= pdc2026x_check_atapi_dma,
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= pdc2026x_bmdma_start,
-	.bmdma_stop	= pdc2026x_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= pdc2026x_port_start,
+	.inherits		= &pdc2024x_port_ops,
+
+	.check_atapi_dma	= pdc2026x_check_atapi_dma,
+	.bmdma_start		= pdc2026x_bmdma_start,
+	.bmdma_stop		= pdc2026x_bmdma_stop,
+
+	.cable_detect		= pdc2026x_cable_detect,
+	.dev_config		= pdc2026x_dev_config,
+
+	.port_start		= pdc2026x_port_start,
 };
 
 static int pdc202xx_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c
index cd2d03a4591a..0588c9b7e73e 100644
--- a/drivers/ata/pata_platform.c
+++ b/drivers/ata/pata_platform.c
@@ -51,27 +51,11 @@ static struct scsi_host_template pata_platform_sht = {
 };
 
 static struct ata_port_operations pata_platform_port_ops = {
-	.set_mode		= pata_platform_set_mode,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_unknown,
-
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
+	.inherits		= &ata_sff_port_ops,
 	.data_xfer		= ata_data_xfer_noirq,
-
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
+	.cable_detect		= ata_cable_unknown,
+	.set_mode		= pata_platform_set_mode,
+	.port_start		= ATA_OP_NULL,
 };
 
 static void pata_platform_setup_port(struct ata_ioports *ioaddr,
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
index ccb8682300b8..d16b343d2a62 100644
--- a/drivers/ata/pata_qdi.c
+++ b/drivers/ata/pata_qdi.c
@@ -158,55 +158,16 @@ static struct scsi_host_template qdi_sht = {
 };
 
 static struct ata_port_operations qdi6500_port_ops = {
-	.set_piomode	= qdi6500_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
+	.inherits	= &ata_sff_port_ops,
 	.qc_issue	= qdi_qc_issue_prot,
-
 	.data_xfer	= qdi_data_xfer,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= qdi6500_set_piomode,
 };
 
 static struct ata_port_operations qdi6580_port_ops = {
+	.inherits	= &qdi6500_port_ops,
 	.set_piomode	= qdi6580_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= qdi_qc_issue_prot,
-
-	.data_xfer	= qdi_data_xfer,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
index 3981bf84d093..94e60b3a1ec6 100644
--- a/drivers/ata/pata_radisys.c
+++ b/drivers/ata/pata_radisys.c
@@ -188,36 +188,12 @@ static struct scsi_host_template radisys_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations radisys_pata_ops = {
+static struct ata_port_operations radisys_pata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.qc_issue		= radisys_qc_issue_prot,
+	.cable_detect		= ata_cable_unknown,
 	.set_piomode		= radisys_set_piomode,
 	.set_dmamode		= radisys_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= ata_cable_unknown,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= radisys_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 
diff --git a/drivers/ata/pata_rb500_cf.c b/drivers/ata/pata_rb500_cf.c
index 4543c980342c..7affceec1c29 100644
--- a/drivers/ata/pata_rb500_cf.c
+++ b/drivers/ata/pata_rb500_cf.c
@@ -118,25 +118,11 @@ static irqreturn_t rb500_pata_irq_handler(int irq, void *dev_instance)
 }
 
 static struct ata_port_operations rb500_pata_port_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-
+	.inherits		= &ata_sff_port_ops,
 	.exec_command		= rb500_pata_exec_command,
-	.check_status 		= ata_check_status,
-	.dev_select 		= ata_std_dev_select,
-
 	.data_xfer		= rb500_pata_data_xfer,
-
-	.qc_prep 		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
 	.freeze			= rb500_pata_freeze,
 	.thaw			= rb500_pata_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
 };
 
 /* ------------------------------------------------------------------------ */
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
index 80909a607d36..a2aef7328bfc 100644
--- a/drivers/ata/pata_rz1000.c
+++ b/drivers/ata/pata_rz1000.c
@@ -57,30 +57,9 @@ static struct scsi_host_template rz1000_sht = {
 };
 
 static struct ata_port_operations rz1000_port_ops = {
-	.set_mode	= rz1000_set_mode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.inherits	= &ata_sff_port_ops,
 	.cable_detect	= ata_cable_40wire,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.set_mode	= rz1000_set_mode,
 };
 
 static int rz1000_fifo_disable(struct pci_dev *pdev)
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
index 38ce6e12ee3d..362b7f829d8e 100644
--- a/drivers/ata/pata_sc1200.c
+++ b/drivers/ata/pata_sc1200.c
@@ -184,37 +184,12 @@ static struct scsi_host_template sc1200_sht = {
 };
 
 static struct ata_port_operations sc1200_port_ops = {
-	.set_piomode	= sc1200_set_piomode,
-	.set_dmamode	= sc1200_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
+	.inherits	= &ata_bmdma_port_ops,
 	.qc_prep 	= ata_dumb_qc_prep,
 	.qc_issue	= sc1200_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= sc1200_set_piomode,
+	.set_dmamode	= sc1200_set_dmamode,
 };
 
 /**
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index 1833e9ef522e..033d1f3a82de 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -972,6 +972,8 @@ static struct scsi_host_template scc_sht = {
 };
 
 static const struct ata_port_operations scc_pata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+
 	.set_piomode		= scc_set_piomode,
 	.set_dmamode		= scc_set_dmamode,
 	.mode_filter		= scc_mode_filter,
@@ -989,12 +991,7 @@ static const struct ata_port_operations scc_pata_ops = {
 	.bmdma_status		= scc_bmdma_status,
 	.data_xfer		= scc_data_xfer,
 
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
 	.freeze			= scc_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-
 	.error_handler		= scc_error_handler,
 	.post_internal_cmd	= scc_bmdma_stop,
 
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index 318a36988900..627abcf85c6e 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -302,71 +302,16 @@ static struct scsi_host_template serverworks_sht = {
 };
 
 static struct ata_port_operations serverworks_osb4_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= serverworks_cable_detect,
+	.mode_filter	= serverworks_osb4_filter,
 	.set_piomode	= serverworks_set_piomode,
 	.set_dmamode	= serverworks_set_dmamode,
-	.mode_filter	= serverworks_osb4_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= serverworks_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations serverworks_csb_port_ops = {
-	.set_piomode	= serverworks_set_piomode,
-	.set_dmamode	= serverworks_set_dmamode,
+	.inherits	= &serverworks_osb4_port_ops,
 	.mode_filter	= serverworks_csb_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= serverworks_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static int serverworks_fixup_osb4(struct pci_dev *pdev)
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index 7812815a34c3..0936f534d9c7 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -196,36 +196,10 @@ static struct scsi_host_template sil680_sht = {
 };
 
 static struct ata_port_operations sil680_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= sil680_cable_detect,
 	.set_piomode	= sil680_set_piomode,
 	.set_dmamode	= sil680_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= sil680_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index dcd8457a8377..3ed628670cd7 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -517,196 +517,51 @@ static struct scsi_host_template sis_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations sis_133_ops = {
+static struct ata_port_operations sis_133_for_sata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.set_piomode		= sis_133_set_piomode,
 	.set_dmamode		= sis_133_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= sis_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
 	.cable_detect		= sis_133_cable_detect,
+};
 
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
+static struct ata_port_operations sis_base_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.error_handler		= sis_error_handler,
 };
 
-static const struct ata_port_operations sis_133_for_sata_ops = {
+static struct ata_port_operations sis_133_ops = {
+	.inherits		= &sis_base_ops,
 	.set_piomode		= sis_133_set_piomode,
 	.set_dmamode		= sis_133_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
 	.cable_detect		= sis_133_cable_detect,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations sis_133_early_ops = {
+static struct ata_port_operations sis_133_early_ops = {
+	.inherits		= &sis_base_ops,
 	.set_piomode		= sis_100_set_piomode,
 	.set_dmamode		= sis_133_early_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= sis_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
 	.cable_detect		= sis_66_cable_detect,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations sis_100_ops = {
+static struct ata_port_operations sis_100_ops = {
+	.inherits		= &sis_base_ops,
 	.set_piomode		= sis_100_set_piomode,
 	.set_dmamode		= sis_100_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= sis_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
 	.cable_detect		= sis_66_cable_detect,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations sis_66_ops = {
+static struct ata_port_operations sis_66_ops = {
+	.inherits		= &sis_base_ops,
 	.set_piomode		= sis_old_set_piomode,
 	.set_dmamode		= sis_66_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
 	.cable_detect		= sis_66_cable_detect,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= sis_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations sis_old_ops = {
+static struct ata_port_operations sis_old_ops = {
+	.inherits		= &sis_base_ops,
 	.set_piomode		= sis_old_set_piomode,
 	.set_dmamode		= sis_old_set_dmamode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= sis_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
 	.cable_detect		= ata_cable_40wire,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_handler		= ata_interrupt,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
 static const struct ata_port_info sis_info = {
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index ece366bced0c..0dd8e2f69558 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -239,37 +239,13 @@ static struct scsi_host_template sl82c105_sht = {
 };
 
 static struct ata_port_operations sl82c105_port_ops = {
-	.set_piomode	= sl82c105_set_piomode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= sl82c105_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
+	.inherits	= &ata_bmdma_port_ops,
+	.qc_defer	= sl82c105_qc_defer,
 	.bmdma_start 	= sl82c105_bmdma_start,
 	.bmdma_stop	= sl82c105_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_defer	= sl82c105_qc_defer,
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= sl82c105_set_piomode,
+	.error_handler	= sl82c105_error_handler,
 };
 
 /**
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index 510569957d10..bc4956ef0931 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -184,36 +184,12 @@ static struct scsi_host_template triflex_sht = {
 };
 
 static struct ata_port_operations triflex_port_ops = {
-	.set_piomode	= triflex_set_piomode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= triflex_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.bmdma_setup 	= ata_bmdma_setup,
+	.inherits	= &ata_bmdma_port_ops,
 	.bmdma_start 	= triflex_bmdma_start,
 	.bmdma_stop	= triflex_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= triflex_set_piomode,
+	.error_handler	= triflex_error_handler,
 };
 
 static int triflex_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index a7bc860e1310..d1edb1b27480 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -339,71 +339,16 @@ static struct scsi_host_template via_sht = {
 };
 
 static struct ata_port_operations via_port_ops = {
+	.inherits	= &ata_bmdma_port_ops,
+	.cable_detect	= via_cable_detect,
 	.set_piomode	= via_set_piomode,
 	.set_dmamode	= via_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
 	.error_handler	= via_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= via_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
-	.data_xfer	= ata_data_xfer,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 static struct ata_port_operations via_port_ops_noirq = {
-	.set_piomode	= via_set_piomode,
-	.set_dmamode	= via_set_dmamode,
-	.mode_filter	= ata_pci_default_filter,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= via_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= via_cable_detect,
-
-	.bmdma_setup 	= ata_bmdma_setup,
-	.bmdma_start 	= ata_bmdma_start,
-	.bmdma_stop	= ata_bmdma_stop,
-	.bmdma_status 	= ata_bmdma_status,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
+	.inherits	= &via_port_ops,
 	.data_xfer	= ata_data_xfer_noirq,
-
-	.irq_handler	= ata_interrupt,
-	.irq_clear	= ata_bmdma_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
 };
 
 /**
diff --git a/drivers/ata/pata_winbond.c b/drivers/ata/pata_winbond.c
index 9bafae9d5fe0..f235bb0d6139 100644
--- a/drivers/ata/pata_winbond.c
+++ b/drivers/ata/pata_winbond.c
@@ -126,29 +126,10 @@ static struct scsi_host_template winbond_sht = {
 };
 
 static struct ata_port_operations winbond_port_ops = {
-	.set_piomode	= winbond_set_piomode,
-
-	.tf_load	= ata_tf_load,
-	.tf_read	= ata_tf_read,
-	.check_status 	= ata_check_status,
-	.exec_command	= ata_exec_command,
-	.dev_select 	= ata_std_dev_select,
-
-	.freeze		= ata_bmdma_freeze,
-	.thaw		= ata_bmdma_thaw,
-	.error_handler	= ata_bmdma_error_handler,
-	.post_internal_cmd = ata_bmdma_post_internal_cmd,
-	.cable_detect	= ata_cable_40wire,
-
-	.qc_prep 	= ata_qc_prep,
-	.qc_issue	= ata_qc_issue_prot,
-
+	.inherits	= &ata_sff_port_ops,
 	.data_xfer	= winbond_data_xfer,
-
-	.irq_clear	= ata_noop_irq_clear,
-	.irq_on		= ata_irq_on,
-
-	.port_start	= ata_sff_port_start,
+	.cable_detect	= ata_cable_40wire,
+	.set_piomode	= winbond_set_piomode,
 };
 
 /**
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index fdf62de57cfc..a5706149af6b 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -148,26 +148,29 @@ static struct scsi_host_template adma_ata_sht = {
 	.dma_boundary		= ADMA_DMA_BOUNDARY,
 };
 
-static const struct ata_port_operations adma_ata_ops = {
+static struct ata_port_operations adma_ata_ops = {
+	.inherits		= &ata_base_port_ops,
+
+	.dev_select		= ata_std_dev_select,
 	.tf_load		= ata_tf_load,
 	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
 	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-	.check_atapi_dma	= adma_check_atapi_dma,
+	.exec_command		= ata_exec_command,
 	.data_xfer		= ata_data_xfer,
+	.check_atapi_dma	= adma_check_atapi_dma,
+	.bmdma_stop		= adma_bmdma_stop,
+	.bmdma_status		= adma_bmdma_status,
 	.qc_prep		= adma_qc_prep,
 	.qc_issue		= adma_qc_issue,
+	.irq_on			= ata_irq_on,
+
 	.freeze			= adma_freeze,
 	.thaw			= adma_thaw,
 	.error_handler		= adma_error_handler,
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
+
 	.port_start		= adma_port_start,
 	.port_stop		= adma_port_stop,
 	.host_stop		= adma_host_stop,
-	.bmdma_stop		= adma_bmdma_stop,
-	.bmdma_status		= adma_bmdma_status,
 };
 
 static struct ata_port_info adma_port_info[] = {
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index f50381b4ba06..865030ae8f8a 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -1198,16 +1198,15 @@ static struct scsi_host_template sata_fsl_sht = {
 };
 
 static const struct ata_port_operations sata_fsl_ops = {
+	.inherits = &sata_port_ops,
+
 	.check_status = sata_fsl_check_status,
 	.check_altstatus = sata_fsl_check_status,
-	.dev_select = ata_noop_dev_select,
 
 	.tf_read = sata_fsl_tf_read,
 
-	.qc_defer = ata_std_qc_defer,
 	.qc_prep = sata_fsl_qc_prep,
 	.qc_issue = sata_fsl_qc_issue,
-	.irq_clear = ata_noop_irq_clear,
 
 	.scr_read = sata_fsl_scr_read,
 	.scr_write = sata_fsl_scr_write,
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index bb853df865da..047f80f5825c 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -522,26 +522,13 @@ static int inic_port_start(struct ata_port *ap)
 }
 
 static struct ata_port_operations inic_port_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.scr_read		= inic_scr_read,
-	.scr_write		= inic_scr_write,
+	.inherits		= &ata_sff_port_ops,
 
 	.bmdma_setup		= inic_bmdma_setup,
 	.bmdma_start		= inic_bmdma_start,
 	.bmdma_stop		= inic_bmdma_stop,
 	.bmdma_status		= inic_bmdma_status,
-
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.qc_prep	 	= ata_qc_prep,
 	.qc_issue		= inic_qc_issue,
-	.data_xfer		= ata_data_xfer,
 
 	.freeze			= inic_freeze,
 	.thaw			= inic_thaw,
@@ -549,8 +536,10 @@ static struct ata_port_operations inic_port_ops = {
 	.post_internal_cmd	= inic_post_internal_cmd,
 	.dev_config		= inic_dev_config,
 
-	.port_resume		= inic_port_resume,
+	.scr_read		= inic_scr_read,
+	.scr_write		= inic_scr_write,
 
+	.port_resume		= inic_port_resume,
 	.port_start		= inic_port_start,
 };
 
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index 52d41edadb72..f341a82d27bf 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -526,23 +526,16 @@ static struct scsi_host_template mv6_sht = {
 	.dma_boundary		= MV_DMA_BOUNDARY,
 };
 
-static const struct ata_port_operations mv5_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
+static struct ata_port_operations mv5_ops = {
+	.inherits		= &ata_sff_port_ops,
 
 	.qc_prep		= mv_qc_prep,
 	.qc_issue		= mv_qc_issue,
-	.data_xfer		= ata_data_xfer,
 
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.error_handler		= mv_error_handler,
 	.freeze			= mv_eh_freeze,
 	.thaw			= mv_eh_thaw,
+	.error_handler		= mv_error_handler,
+	.post_internal_cmd	= ATA_OP_NULL,
 
 	.scr_read		= mv5_scr_read,
 	.scr_write		= mv5_scr_write,
@@ -551,57 +544,18 @@ static const struct ata_port_operations mv5_ops = {
 	.port_stop		= mv_port_stop,
 };
 
-static const struct ata_port_operations mv6_ops = {
-	.dev_config             = mv6_dev_config,
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.qc_prep		= mv_qc_prep,
-	.qc_issue		= mv_qc_issue,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.error_handler		= mv_error_handler,
-	.freeze			= mv_eh_freeze,
-	.thaw			= mv_eh_thaw,
+static struct ata_port_operations mv6_ops = {
+	.inherits		= &mv5_ops,
 	.qc_defer		= ata_std_qc_defer,
-
+	.dev_config             = mv6_dev_config,
 	.scr_read		= mv_scr_read,
 	.scr_write		= mv_scr_write,
-
-	.port_start		= mv_port_start,
-	.port_stop		= mv_port_stop,
 };
 
-static const struct ata_port_operations mv_iie_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
+static struct ata_port_operations mv_iie_ops = {
+	.inherits		= &mv6_ops,
+	.dev_config		= ATA_OP_NULL,
 	.qc_prep		= mv_qc_prep_iie,
-	.qc_issue		= mv_qc_issue,
-	.data_xfer		= ata_data_xfer,
-
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.error_handler		= mv_error_handler,
-	.freeze			= mv_eh_freeze,
-	.thaw			= mv_eh_thaw,
-	.qc_defer		= ata_std_qc_defer,
-
-	.scr_read		= mv_scr_read,
-	.scr_write		= mv_scr_write,
-
-	.port_start		= mv_port_start,
-	.port_stop		= mv_port_stop,
 };
 
 static const struct ata_port_info mv_port_info[] = {
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 9e2b4cef48f2..7b7ba0e26903 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -404,106 +404,41 @@ static struct scsi_host_template nv_swncq_sht = {
 	.slave_configure	= nv_swncq_slave_config,
 };
 
-static const struct ata_port_operations nv_generic_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
+static struct ata_port_operations nv_generic_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.error_handler		= nv_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
 	.scr_read		= nv_scr_read,
 	.scr_write		= nv_scr_write,
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations nv_nf2_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.mode_filter		= ata_pci_default_filter,
+static struct ata_port_operations nv_nf2_ops = {
+	.inherits		= &nv_generic_ops,
 	.freeze			= nv_nf2_freeze,
 	.thaw			= nv_nf2_thaw,
-	.error_handler		= nv_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-	.scr_read		= nv_scr_read,
-	.scr_write		= nv_scr_write,
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations nv_ck804_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.mode_filter		= ata_pci_default_filter,
+static struct ata_port_operations nv_ck804_ops = {
+	.inherits		= &nv_generic_ops,
 	.freeze			= nv_ck804_freeze,
 	.thaw			= nv_ck804_thaw,
-	.error_handler		= nv_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-	.scr_read		= nv_scr_read,
-	.scr_write		= nv_scr_write,
-	.port_start		= ata_sff_port_start,
 	.host_stop		= nv_ck804_host_stop,
 };
 
-static const struct ata_port_operations nv_adma_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= nv_adma_tf_read,
+static struct ata_port_operations nv_adma_ops = {
+	.inherits		= &nv_generic_ops,
+
 	.check_atapi_dma	= nv_adma_check_atapi_dma,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
+	.tf_read		= nv_adma_tf_read,
 	.qc_defer		= ata_std_qc_defer,
 	.qc_prep		= nv_adma_qc_prep,
 	.qc_issue		= nv_adma_qc_issue,
-	.mode_filter		= ata_pci_default_filter,
+	.irq_clear		= nv_adma_irq_clear,
+
 	.freeze			= nv_adma_freeze,
 	.thaw			= nv_adma_thaw,
 	.error_handler		= nv_adma_error_handler,
 	.post_internal_cmd	= nv_adma_post_internal_cmd,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= nv_adma_irq_clear,
-	.irq_on			= ata_irq_on,
-	.scr_read		= nv_scr_read,
-	.scr_write		= nv_scr_write,
+
 	.port_start		= nv_adma_port_start,
 	.port_stop		= nv_adma_port_stop,
 #ifdef CONFIG_PM
@@ -513,29 +448,17 @@ static const struct ata_port_operations nv_adma_ops = {
 	.host_stop		= nv_adma_host_stop,
 };
 
-static const struct ata_port_operations nv_swncq_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
+static struct ata_port_operations nv_swncq_ops = {
+	.inherits		= &nv_generic_ops,
+
 	.qc_defer		= ata_std_qc_defer,
 	.qc_prep		= nv_swncq_qc_prep,
 	.qc_issue		= nv_swncq_qc_issue,
-	.mode_filter		= ata_pci_default_filter,
+
 	.freeze			= nv_mcp55_freeze,
 	.thaw			= nv_mcp55_thaw,
 	.error_handler		= nv_swncq_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-	.scr_read		= nv_scr_read,
-	.scr_write		= nv_scr_write,
+
 #ifdef CONFIG_PM
 	.port_suspend		= nv_swncq_port_suspend,
 	.port_resume		= nv_swncq_port_resume,
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index 37c32ab3b23b..e09b975c973d 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -160,74 +160,42 @@ static struct scsi_host_template pdc_ata_sht = {
 	.dma_boundary		= ATA_DMA_BOUNDARY,
 };
 
-static const struct ata_port_operations pdc_sata_ops = {
+static const struct ata_port_operations pdc_common_ops = {
+	.inherits		= &ata_sff_port_ops,
+
 	.tf_load		= pdc_tf_load_mmio,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
 	.exec_command		= pdc_exec_command_mmio,
-	.dev_select		= ata_std_dev_select,
 	.check_atapi_dma	= pdc_check_atapi_dma,
-
 	.qc_prep		= pdc_qc_prep,
 	.qc_issue		= pdc_qc_issue_prot,
-	.freeze			= pdc_sata_freeze,
-	.thaw			= pdc_sata_thaw,
-	.error_handler		= pdc_sata_error_handler,
-	.post_internal_cmd	= pdc_post_internal_cmd,
-	.cable_detect		= pdc_sata_cable_detect,
-	.data_xfer		= ata_data_xfer,
 	.irq_clear		= pdc_irq_clear,
-	.irq_on			= ata_irq_on,
 
-	.scr_read		= pdc_sata_scr_read,
-	.scr_write		= pdc_sata_scr_write,
-	.port_start		= pdc_sata_port_start,
+	.post_internal_cmd	= pdc_post_internal_cmd,
 };
 
-/* First-generation chips need a more restrictive ->check_atapi_dma op */
-static const struct ata_port_operations pdc_old_sata_ops = {
-	.tf_load		= pdc_tf_load_mmio,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= pdc_exec_command_mmio,
-	.dev_select		= ata_std_dev_select,
-	.check_atapi_dma	= pdc_old_sata_check_atapi_dma,
-
-	.qc_prep		= pdc_qc_prep,
-	.qc_issue		= pdc_qc_issue_prot,
+static struct ata_port_operations pdc_sata_ops = {
+	.inherits		= &pdc_common_ops,
+	.cable_detect		= pdc_sata_cable_detect,
 	.freeze			= pdc_sata_freeze,
 	.thaw			= pdc_sata_thaw,
 	.error_handler		= pdc_sata_error_handler,
-	.post_internal_cmd	= pdc_post_internal_cmd,
-	.cable_detect		= pdc_sata_cable_detect,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= pdc_irq_clear,
-	.irq_on			= ata_irq_on,
-
 	.scr_read		= pdc_sata_scr_read,
 	.scr_write		= pdc_sata_scr_write,
 	.port_start		= pdc_sata_port_start,
 };
 
-static const struct ata_port_operations pdc_pata_ops = {
-	.tf_load		= pdc_tf_load_mmio,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= pdc_exec_command_mmio,
-	.dev_select		= ata_std_dev_select,
-	.check_atapi_dma	= pdc_check_atapi_dma,
+/* First-generation chips need a more restrictive ->check_atapi_dma op */
+static struct ata_port_operations pdc_old_sata_ops = {
+	.inherits		= &pdc_sata_ops,
+	.check_atapi_dma	= pdc_old_sata_check_atapi_dma,
+};
 
-	.qc_prep		= pdc_qc_prep,
-	.qc_issue		= pdc_qc_issue_prot,
+static struct ata_port_operations pdc_pata_ops = {
+	.inherits		= &pdc_common_ops,
+	.cable_detect		= pdc_pata_cable_detect,
 	.freeze			= pdc_freeze,
 	.thaw			= pdc_thaw,
 	.error_handler		= pdc_pata_error_handler,
-	.post_internal_cmd	= pdc_post_internal_cmd,
-	.cable_detect		= pdc_pata_cable_detect,
-	.data_xfer		= ata_data_xfer,
-	.irq_clear		= pdc_irq_clear,
-	.irq_on			= ata_irq_on,
-
 	.port_start		= pdc_common_port_start,
 };
 
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 2566d0926aab..107ef09814de 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -131,27 +131,25 @@ static struct scsi_host_template qs_ata_sht = {
 	.dma_boundary		= QS_DMA_BOUNDARY,
 };
 
-static const struct ata_port_operations qs_ata_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
+static struct ata_port_operations qs_ata_ops = {
+	.inherits		= &ata_sff_port_ops,
+
 	.check_atapi_dma	= qs_check_atapi_dma,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
+	.bmdma_stop		= qs_bmdma_stop,
+	.bmdma_status		= qs_bmdma_status,
 	.qc_prep		= qs_qc_prep,
 	.qc_issue		= qs_qc_issue,
-	.data_xfer		= ata_data_xfer,
+
 	.freeze			= qs_freeze,
 	.thaw			= qs_thaw,
 	.error_handler		= qs_error_handler,
-	.irq_clear		= ata_noop_irq_clear,
-	.irq_on			= ata_irq_on,
+	.post_internal_cmd	= ATA_OP_NULL,
+
 	.scr_read		= qs_scr_read,
 	.scr_write		= qs_scr_write,
+
 	.port_start		= qs_port_start,
 	.host_stop		= qs_host_stop,
-	.bmdma_stop		= qs_bmdma_stop,
-	.bmdma_status		= qs_bmdma_status,
 };
 
 static const struct ata_port_info qs_port_info[] = {
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index 738c1a8ae3b6..eac7ca73cfa0 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -170,31 +170,14 @@ static struct scsi_host_template sil_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations sil_ops = {
+static struct ata_port_operations sil_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.dev_config		= sil_dev_config,
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
 	.set_mode		= sil_set_mode,
-	.mode_filter		= ata_pci_default_filter,
-	.bmdma_setup            = ata_bmdma_setup,
-	.bmdma_start            = ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
 	.freeze			= sil_freeze,
 	.thaw			= sil_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
 	.scr_read		= sil_scr_read,
 	.scr_write		= sil_scr_write,
-	.port_start		= ata_sff_port_start,
 };
 
 static const struct ata_port_info sil_port_info[] = {
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index 7fa63950d81a..363fb90e1047 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -390,34 +390,28 @@ static struct scsi_host_template sil24_sht = {
 	.dma_boundary		= ATA_DMA_BOUNDARY,
 };
 
-static const struct ata_port_operations sil24_ops = {
-	.dev_config		= sil24_dev_config,
+static struct ata_port_operations sil24_ops = {
+	.inherits		= &sata_pmp_port_ops,
 
 	.check_status		= sil24_check_status,
 	.check_altstatus	= sil24_check_status,
-	.dev_select		= ata_noop_dev_select,
-
 	.tf_read		= sil24_tf_read,
-
 	.qc_defer		= sil24_qc_defer,
 	.qc_prep		= sil24_qc_prep,
 	.qc_issue		= sil24_qc_issue,
 
-	.irq_clear		= ata_noop_irq_clear,
+	.freeze			= sil24_freeze,
+	.thaw			= sil24_thaw,
+	.error_handler		= sil24_error_handler,
+	.post_internal_cmd	= sil24_post_internal_cmd,
+	.dev_config		= sil24_dev_config,
 
 	.scr_read		= sil24_scr_read,
 	.scr_write		= sil24_scr_write,
-
 	.pmp_attach		= sil24_pmp_attach,
 	.pmp_detach		= sil24_pmp_detach,
 
-	.freeze			= sil24_freeze,
-	.thaw			= sil24_thaw,
-	.error_handler		= sil24_error_handler,
-	.post_internal_cmd	= sil24_post_internal_cmd,
-
 	.port_start		= sil24_port_start,
-
 #ifdef CONFIG_PM
 	.port_resume		= sil24_port_resume,
 #endif
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index 4becb7fde5e7..9089c7ab5000 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -89,29 +89,10 @@ static struct scsi_host_template sis_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations sis_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-	.bmdma_setup            = ata_bmdma_setup,
-	.bmdma_start            = ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
+static struct ata_port_operations sis_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.scr_read		= sis_scr_read,
 	.scr_write		= sis_scr_write,
-	.port_start		= ata_sff_port_start,
 };
 
 static const struct ata_port_info sis_port_info = {
diff --git a/drivers/ata/sata_svw.c b/drivers/ata/sata_svw.c
index c8768396e006..8636f164256e 100644
--- a/drivers/ata/sata_svw.c
+++ b/drivers/ata/sata_svw.c
@@ -334,30 +334,16 @@ static struct scsi_host_template k2_sata_sht = {
 };
 
 
-static const struct ata_port_operations k2_sata_ops = {
+static struct ata_port_operations k2_sata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.tf_load		= k2_sata_tf_load,
 	.tf_read		= k2_sata_tf_read,
 	.check_status		= k2_stat_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
 	.check_atapi_dma	= k2_sata_check_atapi_dma,
 	.bmdma_setup		= k2_bmdma_setup_mmio,
 	.bmdma_start		= k2_bmdma_start_mmio,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
 	.scr_read		= k2_sata_scr_read,
 	.scr_write		= k2_sata_scr_write,
-	.port_start		= ata_sff_port_start,
 };
 
 static const struct ata_port_info k2_port_info[] = {
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 1802f92180e4..8138cda86a66 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -241,7 +241,8 @@ static struct scsi_host_template pdc_sata_sht = {
 	.dma_boundary		= ATA_DMA_BOUNDARY,
 };
 
-static const struct ata_port_operations pdc_20621_ops = {
+/* TODO: inherit from base port_ops after converting to new EH */
+static struct ata_port_operations pdc_20621_ops = {
 	.tf_load		= pdc_tf_load_mmio,
 	.tf_read		= ata_tf_read,
 	.check_status		= ata_check_status,
diff --git a/drivers/ata/sata_uli.c b/drivers/ata/sata_uli.c
index 764d7064fa59..6ecd13fefa1a 100644
--- a/drivers/ata/sata_uli.c
+++ b/drivers/ata/sata_uli.c
@@ -79,34 +79,10 @@ static struct scsi_host_template uli_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations uli_ops = {
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup            = ata_bmdma_setup,
-	.bmdma_start            = ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.mode_filter		= ata_pci_default_filter,
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
+static struct ata_port_operations uli_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.scr_read		= uli_scr_read,
 	.scr_write		= uli_scr_write,
-
-	.port_start		= ata_sff_port_start,
 };
 
 static const struct ata_port_info uli_port_info = {
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 9be877cb7f57..6326bcf8ea5d 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -103,97 +103,23 @@ static struct scsi_host_template svia_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
 
-static const struct ata_port_operations vt6420_sata_ops = {
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup            = ata_bmdma_setup,
-	.bmdma_start            = ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
+static struct ata_port_operations vt6420_sata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.freeze			= svia_noop_freeze,
-	.thaw			= ata_bmdma_thaw,
 	.error_handler		= vt6420_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations vt6421_pata_ops = {
+static struct ata_port_operations vt6421_pata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+	.cable_detect		= vt6421_pata_cable_detect,
 	.set_piomode		= vt6421_set_pio_mode,
 	.set_dmamode		= vt6421_set_dma_mode,
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup            = ata_bmdma_setup,
-	.bmdma_start            = ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.cable_detect		= vt6421_pata_cable_detect,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
 };
 
-static const struct ata_port_operations vt6421_sata_ops = {
-	.mode_filter		= ata_pci_default_filter,
-
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
-	.exec_command		= ata_exec_command,
-	.dev_select		= ata_std_dev_select,
-
-	.bmdma_setup            = ata_bmdma_setup,
-	.bmdma_start            = ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
-
+static struct ata_port_operations vt6421_sata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.scr_read		= svia_scr_read,
 	.scr_write		= svia_scr_write,
-
-	.port_start		= ata_sff_port_start,
 };
 
 static const struct ata_port_info vt6420_port_info = {
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index fd6855f0bf48..8045a72dc559 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -304,29 +304,14 @@ static struct scsi_host_template vsc_sata_sht = {
 };
 
 
-static const struct ata_port_operations vsc_sata_ops = {
+static struct ata_port_operations vsc_sata_ops = {
+	.inherits		= &ata_bmdma_port_ops,
 	.tf_load		= vsc_sata_tf_load,
 	.tf_read		= vsc_sata_tf_read,
-	.exec_command		= ata_exec_command,
-	.check_status		= ata_check_status,
-	.dev_select		= ata_std_dev_select,
-	.bmdma_setup            = ata_bmdma_setup,
-	.bmdma_start            = ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
-	.mode_filter		= ata_pci_default_filter,
 	.freeze			= vsc_freeze,
 	.thaw			= vsc_thaw,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-	.irq_clear		= ata_bmdma_irq_clear,
-	.irq_on			= ata_irq_on,
 	.scr_read		= vsc_sata_scr_read,
 	.scr_write		= vsc_sata_scr_write,
-	.port_start		= ata_sff_port_start,
 };
 
 static void __devinit vsc_sata_setup_port(struct ata_ioports *port,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index eccc38e17568..46aa4ab64891 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -433,7 +433,7 @@ struct ata_host {
 	void __iomem * const	*iomap;
 	unsigned int		n_ports;
 	void			*private_data;
-	const struct ata_port_operations *ops;
+	struct ata_port_operations *ops;
 	unsigned long		flags;
 #ifdef CONFIG_ATA_ACPI
 	acpi_handle		acpi_handle;
@@ -602,7 +602,7 @@ struct ata_link {
 
 struct ata_port {
 	struct Scsi_Host	*scsi_host; /* our co-allocated scsi host */
-	const struct ata_port_operations *ops;
+	struct ata_port_operations *ops;
 	spinlock_t		*lock;
 	unsigned long		flags;	/* ATA_FLAG_xxx */
 	unsigned int		pflags; /* ATA_PFLAG_xxx */
@@ -664,6 +664,13 @@ struct ata_port {
 	u8			sector_buf[ATA_SECT_SIZE]; /* owned by EH */
 };
 
+/* The following initializer overrides a method to NULL whether one of
+ * its parent has the method defined or not.  This is equivalent to
+ * ERR_PTR(-ENOENT).  Unfortunately, ERR_PTR doesn't render a constant
+ * expression and thus can't be used as an initializer.
+ */
+#define ATA_OP_NULL		(void *)(unsigned long)(-ENOENT)
+
 struct ata_port_operations {
 	/*
 	 * Command execution
@@ -733,6 +740,12 @@ struct ata_port_operations {
 	void (*phy_reset)(struct ata_port *ap);
 	void (*eng_timeout)(struct ata_port *ap);
 	irq_handler_t irq_handler;
+
+	/*
+	 * ->inherits must be the last field and all the preceding
+	 * fields must be pointers.
+	 */
+	const struct ata_port_operations	*inherits;
 };
 
 struct ata_port_info {
@@ -742,7 +755,7 @@ struct ata_port_info {
 	unsigned long		pio_mask;
 	unsigned long		mwdma_mask;
 	unsigned long		udma_mask;
-	const struct ata_port_operations *port_ops;
+	struct ata_port_operations *port_ops;
 	irq_handler_t		irq_handler;
 	void 			*private_data;
 };
@@ -765,7 +778,7 @@ extern const unsigned long sata_deb_timing_normal[];
 extern const unsigned long sata_deb_timing_hotplug[];
 extern const unsigned long sata_deb_timing_long[];
 
-extern const struct ata_port_operations ata_dummy_port_ops;
+extern struct ata_port_operations ata_dummy_port_ops;
 extern const struct ata_port_info ata_dummy_port_info;
 
 static inline const unsigned long *
@@ -812,7 +825,7 @@ extern int ata_host_activate(struct ata_host *host, int irq,
 			     struct scsi_host_template *sht);
 extern void ata_host_detach(struct ata_host *host);
 extern void ata_host_init(struct ata_host *, struct device *,
-			  unsigned long, const struct ata_port_operations *);
+			  unsigned long, struct ata_port_operations *);
 extern int ata_scsi_detect(struct scsi_host_template *sht);
 extern int ata_scsi_ioctl(struct scsi_device *dev, int cmd, void __user *arg);
 extern int ata_scsi_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *));
-- 
cgit v1.2.3


From 1bd5b715a305f6f13455e89becbd839010dd14b5 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:49 +0900
Subject: libata: make ata_pci_init_one() not use ops->irq_handler and pi->sht

ata_pci_init_one() is the only function which uses ops->irq_handler
and pi->sht.  Other initialization functions take the same information
as arguments.  This causes confusion and duplicate unused entries in
structures.

Make ata_pci_init_one() take sht as an argument and use ata_interrupt
implicitly.  All current users use ata_interrupt and if different irq
handler is necessary open coding ata_pci_init_one() using
ata_prepare_sff_host() and ata_activate_sff_host can be done under ten
lines including error handling and driver which requires custom
interrupt handler is likely to require custom initialization anyway.

As ata_pci_init_one() was the last user of ops->irq_handler, this
patch also kills the field.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ata_generic.c       |  3 +--
 drivers/ata/libata-core.c       |  1 -
 drivers/ata/libata-sff.c        |  7 ++++---
 drivers/ata/pata_acpi.c         |  3 +--
 drivers/ata/pata_ali.c          |  9 +--------
 drivers/ata/pata_amd.c          | 12 +-----------
 drivers/ata/pata_artop.c        |  6 +-----
 drivers/ata/pata_atiixp.c       |  3 +--
 drivers/ata/pata_cmd640.c       |  3 +--
 drivers/ata/pata_cmd64x.c       |  8 +-------
 drivers/ata/pata_cs5530.c       |  4 +---
 drivers/ata/pata_cs5535.c       |  3 +--
 drivers/ata/pata_cs5536.c       |  3 +--
 drivers/ata/pata_cypress.c      |  3 +--
 drivers/ata/pata_efar.c         |  3 +--
 drivers/ata/pata_hpt366.c       |  3 +--
 drivers/ata/pata_hpt37x.c       |  8 +-------
 drivers/ata/pata_hpt3x2n.c      |  3 +--
 drivers/ata/pata_it8213.c       |  3 +--
 drivers/ata/pata_it821x.c       |  4 +---
 drivers/ata/pata_jmicron.c      |  3 +--
 drivers/ata/pata_marvell.c      |  4 +---
 drivers/ata/pata_netcell.c      |  3 +--
 drivers/ata/pata_ns87410.c      |  3 +--
 drivers/ata/pata_ns87415.c      |  4 +---
 drivers/ata/pata_oldpiix.c      |  3 +--
 drivers/ata/pata_opti.c         |  3 +--
 drivers/ata/pata_optidma.c      |  4 +---
 drivers/ata/pata_pdc202xx_old.c |  5 +----
 drivers/ata/pata_radisys.c      |  3 +--
 drivers/ata/pata_rz1000.c       |  3 +--
 drivers/ata/pata_sc1200.c       |  3 +--
 drivers/ata/pata_serverworks.c  |  6 +-----
 drivers/ata/pata_sil680.c       |  4 +---
 drivers/ata/pata_sis.c          | 10 +---------
 drivers/ata/pata_sl82c105.c     |  4 +---
 drivers/ata/pata_triflex.c      |  3 +--
 drivers/ata/pata_via.c          |  8 +-------
 include/linux/libata.h          |  4 ++--
 39 files changed, 42 insertions(+), 130 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index 0b5b515ae159..a912ee01a47c 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -120,7 +120,6 @@ static int ata_generic_init_one(struct pci_dev *dev, const struct pci_device_id
 {
 	u16 command;
 	static const struct ata_port_info info = {
-		.sht = &generic_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -153,7 +152,7 @@ static int ata_generic_init_one(struct pci_dev *dev, const struct pci_device_id
 	if (dev->vendor == PCI_VENDOR_ID_AL)
 	    	ata_pci_clear_simplex(dev);
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &generic_sht);
 }
 
 static struct pci_device_id ata_generic[] = {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 32fa9ee397b6..abbe3229480c 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -109,7 +109,6 @@ const struct ata_port_operations ata_sff_port_ops = {
 	.irq_on			= ata_irq_on,
 
 	.port_start		= ata_sff_port_start,
-	.irq_handler		= ata_interrupt,
 };
 
 const struct ata_port_operations ata_bmdma_port_ops = {
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 1cf03d41aa33..a9d5898cbbc4 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -826,6 +826,7 @@ int ata_pci_activate_sff_host(struct ata_host *host,
  *	ata_pci_init_one - Initialize/register PCI IDE host controller
  *	@pdev: Controller to be initialized
  *	@ppi: array of port_info, must be enough for two ports
+ *	@sht: scsi_host_template to use when registering the host
  *
  *	This is a helper function which can be called from a driver's
  *	xxx_init_one() probe function if the hardware uses traditional
@@ -846,7 +847,8 @@ int ata_pci_activate_sff_host(struct ata_host *host,
  *	Zero on success, negative on errno-based value on error.
  */
 int ata_pci_init_one(struct pci_dev *pdev,
-		     const struct ata_port_info * const * ppi)
+		     const struct ata_port_info * const * ppi,
+		     struct scsi_host_template *sht)
 {
 	struct device *dev = &pdev->dev;
 	const struct ata_port_info *pi = NULL;
@@ -882,8 +884,7 @@ int ata_pci_init_one(struct pci_dev *pdev,
 		goto out;
 
 	pci_set_master(pdev);
-	rc = ata_pci_activate_sff_host(host, pi->port_ops->irq_handler,
-				       pi->sht);
+	rc = ata_pci_activate_sff_host(host, ata_interrupt, sht);
  out:
 	if (rc == 0)
 		devres_remove_group(&pdev->dev, NULL);
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index 35ad488db6ed..3edde51750da 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -264,7 +264,6 @@ static struct ata_port_operations pacpi_ops = {
 static int pacpi_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht		= &pacpi_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
 
 		.pio_mask	= 0x1f,
@@ -274,7 +273,7 @@ static int pacpi_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 		.port_ops	= &pacpi_ops,
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &pacpi_sht);
 }
 
 static const struct pci_device_id pacpi_pci_tbl[] = {
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index b00a9cf72c31..f2924996f6e3 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -463,14 +463,12 @@ static void ali_init_chipset(struct pci_dev *pdev)
 static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info_early = {
-		.sht = &ali_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.port_ops = &ali_early_port_ops
 	};
 	/* Revision 0x20 added DMA */
 	static const struct ata_port_info info_20 = {
-		.sht = &ali_sht,
 		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_PIO_LBA48,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -478,7 +476,6 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* Revision 0x20 with support logic added UDMA */
 	static const struct ata_port_info info_20_udma = {
-		.sht = &ali_sht,
 		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_PIO_LBA48,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -487,7 +484,6 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* Revision 0xC2 adds UDMA66 */
 	static const struct ata_port_info info_c2 = {
-		.sht = &ali_sht,
 		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_PIO_LBA48,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -496,7 +492,6 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* Revision 0xC3 is UDMA66 for now */
 	static const struct ata_port_info info_c3 = {
-		.sht = &ali_sht,
 		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_PIO_LBA48,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -505,7 +500,6 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* Revision 0xC4 is UDMA100 */
 	static const struct ata_port_info info_c4 = {
-		.sht = &ali_sht,
 		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_PIO_LBA48,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -514,7 +508,6 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* Revision 0xC5 is UDMA133 with LBA48 DMA */
 	static const struct ata_port_info info_c5 = {
-		.sht = &ali_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -559,7 +552,7 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	        	ppi[0] = &info_20_udma;
 		pci_dev_put(isa_bridge);
 	}
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &ali_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index b0cb4eaf273c..644702cac6ee 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -413,7 +413,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info[10] = {
 		{	/* 0: AMD 7401 */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,	/* No SWDMA */
@@ -421,7 +420,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd33_port_ops
 		},
 		{	/* 1: Early AMD7409 - no swdma */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -429,7 +427,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd66_port_ops
 		},
 		{	/* 2: AMD 7409, no swdma errata */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -437,7 +434,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd66_port_ops
 		},
 		{	/* 3: AMD 7411 */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -445,7 +441,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd100_port_ops
 		},
 		{	/* 4: AMD 7441 */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -453,7 +448,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd100_port_ops
 		},
 		{	/* 5: AMD 8111*/
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -461,7 +455,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd133_port_ops
 		},
 		{	/* 6: AMD 8111 UDMA 100 (Serenade) */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -469,7 +462,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd133_port_ops
 		},
 		{	/* 7: Nvidia Nforce */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -477,7 +469,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &nv100_port_ops
 		},
 		{	/* 8: Nvidia Nforce2 and later */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -485,7 +476,6 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &nv133_port_ops
 		},
 		{	/* 9: AMD CS5536 (Geode companion) */
-			.sht = &amd_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -544,7 +534,7 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	/* And fire it up */
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &amd_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index 0101e5aef3e0..698a53c96111 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -352,7 +352,6 @@ static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static int printed_version;
 	static const struct ata_port_info info_6210 = {
-		.sht		= &artop_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -360,7 +359,6 @@ static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 		.port_ops	= &artop6210_ops,
 	};
 	static const struct ata_port_info info_626x = {
-		.sht		= &artop_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -368,7 +366,6 @@ static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 		.port_ops	= &artop6260_ops,
 	};
 	static const struct ata_port_info info_628x = {
-		.sht		= &artop_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -376,7 +373,6 @@ static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 		.port_ops	= &artop6260_ops,
 	};
 	static const struct ata_port_info info_628x_fast = {
-		.sht		= &artop_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -434,7 +430,7 @@ static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 
 	BUG_ON(ppi[0] == NULL);
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &artop_sht);
 }
 
 static const struct pci_device_id artop_pci_tbl[] = {
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 2655f6a17ad3..6fe433ba62bd 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -241,7 +241,6 @@ static struct ata_port_operations atiixp_port_ops = {
 static int atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &atiixp_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x06,	/* No MWDMA0 support */
@@ -249,7 +248,7 @@ static int atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.port_ops = &atiixp_port_ops
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &atiixp_sht);
 }
 
 static const struct pci_device_id atiixp[] = {
diff --git a/drivers/ata/pata_cmd640.c b/drivers/ata/pata_cmd640.c
index 061c891c8a66..efd2bb5747b4 100644
--- a/drivers/ata/pata_cmd640.c
+++ b/drivers/ata/pata_cmd640.c
@@ -211,7 +211,6 @@ static void cmd640_hardware_init(struct pci_dev *pdev)
 static int cmd640_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &cmd640_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.port_ops = &cmd640_port_ops
@@ -225,7 +224,7 @@ static int cmd640_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	cmd640_hardware_init(pdev);
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &cmd640_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index 1ac8ecfb97e2..bfd72ef9cd31 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -298,21 +298,18 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	static const struct ata_port_info cmd_info[6] = {
 		{	/* CMD 643 - no UDMA */
-			.sht = &cmd64x_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
 			.port_ops = &cmd64x_port_ops
 		},
 		{	/* CMD 646 with broken UDMA */
-			.sht = &cmd64x_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
 			.port_ops = &cmd64x_port_ops
 		},
 		{	/* CMD 646 with working UDMA */
-			.sht = &cmd64x_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -320,14 +317,12 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &cmd64x_port_ops
 		},
 		{	/* CMD 646 rev 1  */
-			.sht = &cmd64x_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
 			.port_ops = &cmd646r1_port_ops
 		},
 		{	/* CMD 648 */
-			.sht = &cmd64x_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -335,7 +330,6 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &cmd648_port_ops
 		},
 		{	/* CMD 649 */
-			.sht = &cmd64x_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -379,7 +373,7 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	pci_write_config_byte(pdev, UDIDETCR0, 0xF0);
 #endif
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &cmd64x_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
index e4a16a578cac..c632ce499d33 100644
--- a/drivers/ata/pata_cs5530.c
+++ b/drivers/ata/pata_cs5530.c
@@ -298,7 +298,6 @@ fail_put:
 static int cs5530_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &cs5530_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -307,7 +306,6 @@ static int cs5530_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* The docking connector doesn't do UDMA, and it seems not MWDMA */
 	static const struct ata_port_info info_palmax_secondary = {
-		.sht = &cs5530_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.port_ops = &cs5530_port_ops
@@ -327,7 +325,7 @@ static int cs5530_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		ppi[1] = &info_palmax_secondary;
 
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &cs5530_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
index f910a8aa7437..d78cf95cbe45 100644
--- a/drivers/ata/pata_cs5535.c
+++ b/drivers/ata/pata_cs5535.c
@@ -181,7 +181,6 @@ static struct ata_port_operations cs5535_port_ops = {
 static int cs5535_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &cs5535_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -200,7 +199,7 @@ static int cs5535_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	rdmsr(ATAC_CH0D1_PIO, timings, dummy);
 	if (CS5535_BAD_PIO(timings))
 		wrmsr(ATAC_CH0D1_PIO, 0xF7F4F7F4UL, 0);
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &cs5535_sht);
 }
 
 static const struct pci_device_id cs5535[] = {
diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c
index 075ee6a7be39..f7c0e4e319ed 100644
--- a/drivers/ata/pata_cs5536.c
+++ b/drivers/ata/pata_cs5536.c
@@ -241,7 +241,6 @@ static struct ata_port_operations cs5536_port_ops = {
 static int cs5536_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &cs5536_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -262,7 +261,7 @@ static int cs5536_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		return -ENODEV;
 	}
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &cs5536_sht);
 }
 
 static const struct pci_device_id cs5536[] = {
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
index c459553e7d1e..cbd6670ea0de 100644
--- a/drivers/ata/pata_cypress.c
+++ b/drivers/ata/pata_cypress.c
@@ -123,7 +123,6 @@ static struct ata_port_operations cy82c693_port_ops = {
 static int cy82c693_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &cy82c693_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -137,7 +136,7 @@ static int cy82c693_init_one(struct pci_dev *pdev, const struct pci_device_id *i
 	if (PCI_FUNC(pdev->devfn) != 1)
 		return -ENODEV;
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &cy82c693_sht);
 }
 
 static const struct pci_device_id cy82c693[] = {
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index ef62fc642c17..0260edac2370 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -263,7 +263,6 @@ static int efar_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	static int printed_version;
 	static const struct ata_port_info info = {
-		.sht		= &efar_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma1-2 */
@@ -276,7 +275,7 @@ static int efar_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &efar_sht);
 }
 
 static const struct pci_device_id efar_pci_tbl[] = {
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
index 788955f57ff8..b62d398ed84b 100644
--- a/drivers/ata/pata_hpt366.c
+++ b/drivers/ata/pata_hpt366.c
@@ -350,7 +350,6 @@ static void hpt36x_init_chipset(struct pci_dev *dev)
 static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info_hpt366 = {
-		.sht = &hpt36x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -394,7 +393,7 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 			break;
 	}
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &hpt36x_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
index c42eec70d297..a43c19753669 100644
--- a/drivers/ata/pata_hpt37x.c
+++ b/drivers/ata/pata_hpt37x.c
@@ -783,7 +783,6 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	/* HPT370 - UDMA100 */
 	static const struct ata_port_info info_hpt370 = {
-		.sht = &hpt37x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -792,7 +791,6 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	};
 	/* HPT370A - UDMA100 */
 	static const struct ata_port_info info_hpt370a = {
-		.sht = &hpt37x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -801,7 +799,6 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	};
 	/* HPT370 - UDMA100 */
 	static const struct ata_port_info info_hpt370_33 = {
-		.sht = &hpt37x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -810,7 +807,6 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	};
 	/* HPT370A - UDMA100 */
 	static const struct ata_port_info info_hpt370a_33 = {
-		.sht = &hpt37x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -819,7 +815,6 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	};
 	/* HPT371, 372 and friends - UDMA133 */
 	static const struct ata_port_info info_hpt372 = {
-		.sht = &hpt37x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -828,7 +823,6 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	};
 	/* HPT374 - UDMA100 */
 	static const struct ata_port_info info_hpt374 = {
-		.sht = &hpt37x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -1051,7 +1045,7 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	port_info = *port;
 	port_info.private_data = private_data;
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &hpt37x_sht);
 }
 
 static const struct pci_device_id hpt37x[] = {
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
index b77b1279d757..2c178c30116c 100644
--- a/drivers/ata/pata_hpt3x2n.c
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -452,7 +452,6 @@ static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	/* HPT372N and friends - UDMA133 */
 	static const struct ata_port_info info = {
-		.sht = &hpt3x2n_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -568,7 +567,7 @@ static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	}
 
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &hpt3x2n_sht);
 }
 
 static const struct pci_device_id hpt3x2n[] = {
diff --git a/drivers/ata/pata_it8213.c b/drivers/ata/pata_it8213.c
index 9ce89522e764..291a0d6e2434 100644
--- a/drivers/ata/pata_it8213.c
+++ b/drivers/ata/pata_it8213.c
@@ -274,7 +274,6 @@ static int it8213_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 {
 	static int printed_version;
 	static const struct ata_port_info info = {
-		.sht		= &it8213_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -288,7 +287,7 @@ static int it8213_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &it8213_sht);
 }
 
 static const struct pci_device_id it8213_pci_tbl[] = {
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index 669d224d30ca..63c5cf0d1fee 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -687,14 +687,12 @@ static int it821x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	u8 conf;
 
 	static const struct ata_port_info info_smart = {
-		.sht = &it821x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
 		.port_ops = &it821x_smart_port_ops
 	};
 	static const struct ata_port_info info_passthru = {
-		.sht = &it821x_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -724,7 +722,7 @@ static int it821x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	else
 		ppi[0] = &info_smart;
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &it821x_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index 61ff5c6b4568..859e47a600cc 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -148,7 +148,6 @@ static struct ata_port_operations jmicron_ops = {
 static int jmicron_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht		= &jmicron_sht,
 		.flags	= ATA_FLAG_SLAVE_POSS,
 
 		.pio_mask	= 0x1f,
@@ -159,7 +158,7 @@ static int jmicron_init_one (struct pci_dev *pdev, const struct pci_device_id *i
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &jmicron_sht);
 }
 
 static const struct pci_device_id jmicron_pci_tbl[] = {
diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c
index 286310fc5910..d8da4f344c0a 100644
--- a/drivers/ata/pata_marvell.c
+++ b/drivers/ata/pata_marvell.c
@@ -119,7 +119,6 @@ static struct ata_port_operations marvell_ops = {
 static int marvell_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht		= &marvell_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 
 		.pio_mask	= 0x1f,
@@ -129,7 +128,6 @@ static int marvell_init_one (struct pci_dev *pdev, const struct pci_device_id *i
 		.port_ops	= &marvell_ops,
 	};
 	static const struct ata_port_info info_sata = {
-		.sht		= &marvell_sht,
 		/* Slave possible as its magically mapped not real */
 		.flags		= ATA_FLAG_SLAVE_POSS,
 
@@ -144,7 +142,7 @@ static int marvell_init_one (struct pci_dev *pdev, const struct pci_device_id *i
 	if (pdev->device == 0x6101)
 		ppi[1] = &ata_dummy_port_info;
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &marvell_sht);
 }
 
 static const struct pci_device_id marvell_pci_tbl[] = {
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
index 65389d1837b3..ae50a5e85cf1 100644
--- a/drivers/ata/pata_netcell.c
+++ b/drivers/ata/pata_netcell.c
@@ -48,7 +48,6 @@ static int netcell_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 {
 	static int printed_version;
 	static const struct ata_port_info info = {
-		.sht		= &netcell_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		/* Actually we don't really care about these as the
 		   firmware deals with it */
@@ -72,7 +71,7 @@ static int netcell_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	ata_pci_clear_simplex(pdev);
 
 	/* And let the library code do the work */
-	return ata_pci_init_one(pdev, port_info);
+	return ata_pci_init_one(pdev, port_info, &netcell_sht);
 }
 
 static const struct pci_device_id netcell_pci_tbl[] = {
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index 5b1982fa0be1..1bdca8f1e767 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -158,13 +158,12 @@ static struct ata_port_operations ns87410_port_ops = {
 static int ns87410_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &ns87410_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x0F,
 		.port_ops = &ns87410_port_ops
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &ns87410_sht);
 }
 
 static const struct pci_device_id ns87410[] = {
diff --git a/drivers/ata/pata_ns87415.c b/drivers/ata/pata_ns87415.c
index 38d86a262dbb..42508940e4a9 100644
--- a/drivers/ata/pata_ns87415.c
+++ b/drivers/ata/pata_ns87415.c
@@ -345,7 +345,6 @@ static int ns87415_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 {
 	static int printed_version;
 	static const struct ata_port_info info = {
-		.sht		= &ns87415_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -355,7 +354,6 @@ static int ns87415_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	int rc;
 #if defined(CONFIG_SUPERIO)
 	static const struct ata_port_info info87560 = {
-		.sht		= &ns87415_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -377,7 +375,7 @@ static int ns87415_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	pci_write_config_byte(pdev, 0x55, 0xEE);
 	/* Select PIO0 8bit clocking */
 	pci_write_config_byte(pdev, 0x54, 0xB7);
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &ns87415_sht);
 }
 
 static const struct pci_device_id ns87415_pci_tbl[] = {
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index f6062b37310d..9e3afadbd04a 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -252,7 +252,6 @@ static int oldpiix_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 {
 	static int printed_version;
 	static const struct ata_port_info info = {
-		.sht		= &oldpiix_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma1-2 */
@@ -264,7 +263,7 @@ static int oldpiix_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &oldpiix_sht);
 }
 
 static const struct pci_device_id oldpiix_pci_tbl[] = {
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index c4a0795c3ff4..8601d9c3cb39 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -178,7 +178,6 @@ static struct ata_port_operations opti_port_ops = {
 static int opti_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &opti_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.port_ops = &opti_port_ops
@@ -189,7 +188,7 @@ static int opti_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!printed_version++)
 		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &opti_sht);
 }
 
 static const struct pci_device_id opti[] = {
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index eb4b08190e3a..c376f9ef77c8 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -414,14 +414,12 @@ done_nomsg:		/* Wrong chip revision */
 static int optidma_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info_82c700 = {
-		.sht = &optidma_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
 		.port_ops = &optidma_port_ops
 	};
 	static const struct ata_port_info info_82c700_udma = {
-		.sht = &optidma_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -447,7 +445,7 @@ static int optidma_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (optiplus_with_udma(dev))
 		ppi[0] = &info_82c700_udma;
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &optidma_sht);
 }
 
 static const struct pci_device_id optidma[] = {
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
index 4daac20df0bc..5545fbab6a7e 100644
--- a/drivers/ata/pata_pdc202xx_old.c
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -290,7 +290,6 @@ static int pdc202xx_init_one(struct pci_dev *dev, const struct pci_device_id *id
 {
 	static const struct ata_port_info info[3] = {
 		{
-			.sht = &pdc202xx_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -298,7 +297,6 @@ static int pdc202xx_init_one(struct pci_dev *dev, const struct pci_device_id *id
 			.port_ops = &pdc2024x_port_ops
 		},
 		{
-			.sht = &pdc202xx_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -306,7 +304,6 @@ static int pdc202xx_init_one(struct pci_dev *dev, const struct pci_device_id *id
 			.port_ops = &pdc2026x_port_ops
 		},
 		{
-			.sht = &pdc202xx_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -327,7 +324,7 @@ static int pdc202xx_init_one(struct pci_dev *dev, const struct pci_device_id *id
 				return -ENODEV;
 		}
 	}
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &pdc202xx_sht);
 }
 
 static const struct pci_device_id pdc202xx[] = {
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
index 94e60b3a1ec6..145d5ba92795 100644
--- a/drivers/ata/pata_radisys.c
+++ b/drivers/ata/pata_radisys.c
@@ -216,7 +216,6 @@ static int radisys_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 {
 	static int printed_version;
 	static const struct ata_port_info info = {
-		.sht		= &radisys_sht,
 		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma1-2 */
@@ -229,7 +228,7 @@ static int radisys_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &radisys_sht);
 }
 
 static const struct pci_device_id radisys_pci_tbl[] = {
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
index a2aef7328bfc..04be6aee4354 100644
--- a/drivers/ata/pata_rz1000.c
+++ b/drivers/ata/pata_rz1000.c
@@ -89,7 +89,6 @@ static int rz1000_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 {
 	static int printed_version;
 	static const struct ata_port_info info = {
-		.sht = &rz1000_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.port_ops = &rz1000_port_ops
@@ -100,7 +99,7 @@ static int rz1000_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 		printk(KERN_DEBUG DRV_NAME " version " DRV_VERSION "\n");
 
 	if (rz1000_fifo_disable(pdev) == 0)
-		return ata_pci_init_one(pdev, ppi);
+		return ata_pci_init_one(pdev, ppi, &rz1000_sht);
 
 	printk(KERN_ERR DRV_NAME ": failed to disable read-ahead on chipset..\n");
 	/* Not safe to use so skip */
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
index 362b7f829d8e..38c7fb0bebe9 100644
--- a/drivers/ata/pata_sc1200.c
+++ b/drivers/ata/pata_sc1200.c
@@ -204,7 +204,6 @@ static struct ata_port_operations sc1200_port_ops = {
 static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &sc1200_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -214,7 +213,7 @@ static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	/* Can't enable port 2 yet, see top comments */
 	const struct ata_port_info *ppi[] = { &info, &ata_dummy_port_info };
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &sc1200_sht);
 }
 
 static const struct pci_device_id sc1200[] = {
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index 627abcf85c6e..515b5b70a555 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -399,28 +399,24 @@ static int serverworks_init_one(struct pci_dev *pdev, const struct pci_device_id
 {
 	static const struct ata_port_info info[4] = {
 		{ /* OSB4 */
-			.sht = &serverworks_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
 			.udma_mask = 0x07,
 			.port_ops = &serverworks_osb4_port_ops
 		}, { /* OSB4 no UDMA */
-			.sht = &serverworks_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
 			.udma_mask = 0x00,
 			.port_ops = &serverworks_osb4_port_ops
 		}, { /* CSB5 */
-			.sht = &serverworks_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
 			.udma_mask = ATA_UDMA4,
 			.port_ops = &serverworks_csb_port_ops
 		}, { /* CSB5 - later revisions*/
-			.sht = &serverworks_sht,
 			.flags = ATA_FLAG_SLAVE_POSS,
 			.pio_mask = 0x1f,
 			.mwdma_mask = 0x07,
@@ -465,7 +461,7 @@ static int serverworks_init_one(struct pci_dev *pdev, const struct pci_device_id
 	if (pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB5IDE)
 		ata_pci_clear_simplex(pdev);
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &serverworks_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index 0936f534d9c7..5313deeffa6d 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -282,7 +282,6 @@ static int __devinit sil680_init_one(struct pci_dev *pdev,
 				     const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &sil680_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -290,7 +289,6 @@ static int __devinit sil680_init_one(struct pci_dev *pdev,
 		.port_ops = &sil680_port_ops
 	};
 	static const struct ata_port_info info_slow = {
-		.sht = &sil680_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -364,7 +362,7 @@ static int __devinit sil680_init_one(struct pci_dev *pdev,
 				 &sil680_sht);
 
 use_ioports:
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &sil680_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index 3ed628670cd7..32be13ba5f06 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -565,7 +565,6 @@ static struct ata_port_operations sis_old_ops = {
 };
 
 static const struct ata_port_info sis_info = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.mwdma_mask	= 0x07,
@@ -573,7 +572,6 @@ static const struct ata_port_info sis_info = {
 	.port_ops	= &sis_old_ops,
 };
 static const struct ata_port_info sis_info33 = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.mwdma_mask	= 0x07,
@@ -581,42 +579,36 @@ static const struct ata_port_info sis_info33 = {
 	.port_ops	= &sis_old_ops,
 };
 static const struct ata_port_info sis_info66 = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.udma_mask	= ATA_UDMA4,	/* UDMA 66 */
 	.port_ops	= &sis_66_ops,
 };
 static const struct ata_port_info sis_info100 = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.udma_mask	= ATA_UDMA5,
 	.port_ops	= &sis_100_ops,
 };
 static const struct ata_port_info sis_info100_early = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS,
 	.udma_mask	= ATA_UDMA5,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.port_ops	= &sis_66_ops,
 };
 static const struct ata_port_info sis_info133 = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.udma_mask	= ATA_UDMA6,
 	.port_ops	= &sis_133_ops,
 };
 const struct ata_port_info sis_info133_for_sata = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.udma_mask	= ATA_UDMA6,
 	.port_ops	= &sis_133_for_sata_ops,
 };
 static const struct ata_port_info sis_info133_early = {
-	.sht		= &sis_sht,
 	.flags		= ATA_FLAG_SLAVE_POSS,
 	.pio_mask	= 0x1f,	/* pio0-4 */
 	.udma_mask	= ATA_UDMA6,
@@ -844,7 +836,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	sis_fixup(pdev, chipset);
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &sis_sht);
 }
 
 static const struct pci_device_id sis_pci_tbl[] = {
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index 0dd8e2f69558..2d14b2505c7d 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -289,14 +289,12 @@ static int sl82c105_bridge_revision(struct pci_dev *pdev)
 static int sl82c105_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info_dma = {
-		.sht = &sl82c105_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
 		.port_ops = &sl82c105_port_ops
 	};
 	static const struct ata_port_info info_early = {
-		.sht = &sl82c105_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.port_ops = &sl82c105_port_ops
@@ -325,7 +323,7 @@ static int sl82c105_init_one(struct pci_dev *dev, const struct pci_device_id *id
 	val |= CTRL_P0EN | CTRL_P0F16 | CTRL_P1F16;
 	pci_write_config_dword(dev, 0x40, val);
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &sl82c105_sht);
 }
 
 static const struct pci_device_id sl82c105[] = {
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index bc4956ef0931..86dc66c37389 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -195,7 +195,6 @@ static struct ata_port_operations triflex_port_ops = {
 static int triflex_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	static const struct ata_port_info info = {
-		.sht = &triflex_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -207,7 +206,7 @@ static int triflex_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!printed_version++)
 		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(dev, ppi);
+	return ata_pci_init_one(dev, ppi, &triflex_sht);
 }
 
 static const struct pci_device_id triflex[] = {
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index d1edb1b27480..e66bb85ad3d1 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -398,7 +398,6 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	/* Early VIA without UDMA support */
 	static const struct ata_port_info via_mwdma_info = {
-		.sht = &via_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -406,7 +405,6 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* Ditto with IRQ masking required */
 	static const struct ata_port_info via_mwdma_info_borked = {
-		.sht = &via_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -414,7 +412,6 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* VIA UDMA 33 devices (and borked 66) */
 	static const struct ata_port_info via_udma33_info = {
-		.sht = &via_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -423,7 +420,6 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* VIA UDMA 66 devices */
 	static const struct ata_port_info via_udma66_info = {
-		.sht = &via_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -432,7 +428,6 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* VIA UDMA 100 devices */
 	static const struct ata_port_info via_udma100_info = {
-		.sht = &via_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -441,7 +436,6 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	};
 	/* UDMA133 with bad AST (All current 133) */
 	static const struct ata_port_info via_udma133_info = {
-		.sht = &via_sht,
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
@@ -532,7 +526,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	/* We have established the device type, now fire it up */
 	type.private_data = (void *)config;
 
-	return ata_pci_init_one(pdev, ppi);
+	return ata_pci_init_one(pdev, ppi, &via_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 46aa4ab64891..5494119854de 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -739,7 +739,6 @@ struct ata_port_operations {
 	 */
 	void (*phy_reset)(struct ata_port *ap);
 	void (*eng_timeout)(struct ata_port *ap);
-	irq_handler_t irq_handler;
 
 	/*
 	 * ->inherits must be the last field and all the preceding
@@ -1020,7 +1019,8 @@ static inline int ata_acpi_cbl_80wire(struct ata_port *ap,
 struct pci_dev;
 
 extern int ata_pci_init_one(struct pci_dev *pdev,
-			     const struct ata_port_info * const * ppi);
+			    const struct ata_port_info * const * ppi,
+			    struct scsi_host_template *sht);
 extern void ata_pci_remove_one(struct pci_dev *pdev);
 #ifdef CONFIG_PM
 extern void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t mesg);
-- 
cgit v1.2.3


From 887125e3740283be25564bfc6fb5d24974b651ab Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:49 +0900
Subject: libata: stop overloading port_info->private_data

port_info->private_data is currently used for two purposes - to record
private data about the port_info or to specify host->private_data to
use when allocating ata_host.

This overloading is confusing and counter-intuitive in that
port_info->private_data becomes host->private_data instead of
port->private_data.  In addition, port_info and host don't correspond
to each other 1-to-1.  Currently, the first non-NULL
port_info->private_data is used.

This patch makes port_info->private_data just be what it is -
private_data for the port_info where LLD can jot down extra info.
libata no longer sets host->private_data to the first non-NULL
port_info->private_data, @host_priv argument is added to
ata_pci_init_one() instead.  LLDs which use ata_pci_init_one() can use
this argument to pass in pointer to host private data.  LLDs which
don't should use init-register model anyway and can initialize
host->private_data directly.

Adding @host_priv instead of using init-register model for LLDs which
use ata_pci_init_one() is suggested by Alan Cox.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
---
 drivers/ata/ata_generic.c       |  2 +-
 drivers/ata/libata-core.c       |  2 --
 drivers/ata/libata-sff.c        |  4 +++-
 drivers/ata/pata_acpi.c         | 18 ++----------------
 drivers/ata/pata_ali.c          |  2 +-
 drivers/ata/pata_amd.c          | 27 +++++++--------------------
 drivers/ata/pata_artop.c        |  2 +-
 drivers/ata/pata_atiixp.c       |  2 +-
 drivers/ata/pata_cmd640.c       |  2 +-
 drivers/ata/pata_cmd64x.c       |  2 +-
 drivers/ata/pata_cs5530.c       |  2 +-
 drivers/ata/pata_cs5535.c       |  2 +-
 drivers/ata/pata_cs5536.c       |  2 +-
 drivers/ata/pata_cypress.c      |  2 +-
 drivers/ata/pata_efar.c         |  2 +-
 drivers/ata/pata_hpt366.c       | 12 ++++++------
 drivers/ata/pata_hpt37x.c       | 33 ++++++++++++++-------------------
 drivers/ata/pata_hpt3x2n.c      |  9 ++++-----
 drivers/ata/pata_it8213.c       |  2 +-
 drivers/ata/pata_it821x.c       |  2 +-
 drivers/ata/pata_jmicron.c      | 18 ++----------------
 drivers/ata/pata_marvell.c      | 18 ++----------------
 drivers/ata/pata_netcell.c      |  2 +-
 drivers/ata/pata_ns87410.c      |  2 +-
 drivers/ata/pata_ns87415.c      |  2 +-
 drivers/ata/pata_oldpiix.c      |  2 +-
 drivers/ata/pata_opti.c         |  2 +-
 drivers/ata/pata_optidma.c      |  2 +-
 drivers/ata/pata_pdc202xx_old.c |  2 +-
 drivers/ata/pata_radisys.c      |  2 +-
 drivers/ata/pata_rz1000.c       |  2 +-
 drivers/ata/pata_sc1200.c       |  2 +-
 drivers/ata/pata_serverworks.c  |  2 +-
 drivers/ata/pata_sil680.c       |  2 +-
 drivers/ata/pata_sis.c          |  8 +++-----
 drivers/ata/pata_sl82c105.c     |  2 +-
 drivers/ata/pata_triflex.c      |  2 +-
 drivers/ata/pata_via.c          | 19 ++++++++-----------
 include/linux/libata.h          |  2 +-
 39 files changed, 79 insertions(+), 145 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index a912ee01a47c..b23e2a1099c5 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -152,7 +152,7 @@ static int ata_generic_init_one(struct pci_dev *dev, const struct pci_device_id
 	if (dev->vendor == PCI_VENDOR_ID_AL)
 	    	ata_pci_clear_simplex(dev);
 
-	return ata_pci_init_one(dev, ppi, &generic_sht);
+	return ata_pci_init_one(dev, ppi, &generic_sht, NULL);
 }
 
 static struct pci_device_id ata_generic[] = {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index abbe3229480c..aa6bcd79d60a 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -6995,8 +6995,6 @@ struct ata_host *ata_host_alloc_pinfo(struct device *dev,
 
 		if (!host->ops && (pi->port_ops != &ata_dummy_port_ops))
 			host->ops = pi->port_ops;
-		if (!host->private_data && pi->private_data)
-			host->private_data = pi->private_data;
 	}
 
 	return host;
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index a9d5898cbbc4..6223ec042c80 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -827,6 +827,7 @@ int ata_pci_activate_sff_host(struct ata_host *host,
  *	@pdev: Controller to be initialized
  *	@ppi: array of port_info, must be enough for two ports
  *	@sht: scsi_host_template to use when registering the host
+ *	@host_priv: host private_data
  *
  *	This is a helper function which can be called from a driver's
  *	xxx_init_one() probe function if the hardware uses traditional
@@ -848,7 +849,7 @@ int ata_pci_activate_sff_host(struct ata_host *host,
  */
 int ata_pci_init_one(struct pci_dev *pdev,
 		     const struct ata_port_info * const * ppi,
-		     struct scsi_host_template *sht)
+		     struct scsi_host_template *sht, void *host_priv)
 {
 	struct device *dev = &pdev->dev;
 	const struct ata_port_info *pi = NULL;
@@ -882,6 +883,7 @@ int ata_pci_init_one(struct pci_dev *pdev,
 	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
 	if (rc)
 		goto out;
+	host->private_data = host_priv;
 
 	pci_set_master(pdev);
 	rc = ata_pci_activate_sff_host(host, ata_interrupt, sht);
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index 3edde51750da..d337f3209caf 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -67,20 +67,6 @@ static int pacpi_cable_detect(struct ata_port *ap)
 		return ATA_CBL_PATA40;
 }
 
-/**
- *	pacpi_error_handler - Setup and error handler
- *	@ap: Port to handle
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void pacpi_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, pacpi_pre_reset, ata_std_softreset, NULL,
-			   ata_std_postreset);
-}
-
 /**
  *	pacpi_discover_modes	-	filter non ACPI modes
  *	@adev: ATA device
@@ -242,7 +228,7 @@ static struct ata_port_operations pacpi_ops = {
 	.mode_filter		= pacpi_mode_filter,
 	.set_piomode		= pacpi_set_piomode,
 	.set_dmamode		= pacpi_set_dmamode,
-	.error_handler		= pacpi_error_handler,
+	.prereset		= pacpi_pre_reset,
 	.port_start		= pacpi_port_start,
 };
 
@@ -273,7 +259,7 @@ static int pacpi_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 		.port_ops	= &pacpi_ops,
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(pdev, ppi, &pacpi_sht);
+	return ata_pci_init_one(pdev, ppi, &pacpi_sht, NULL);
 }
 
 static const struct pci_device_id pacpi_pci_tbl[] = {
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index f2924996f6e3..43c558f20f92 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -552,7 +552,7 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	        	ppi[0] = &info_20_udma;
 		pci_dev_put(isa_bridge);
 	}
-	return ata_pci_init_one(pdev, ppi, &ali_sht);
+	return ata_pci_init_one(pdev, ppi, &ali_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 644702cac6ee..09c8286b6890 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -144,12 +144,6 @@ static int amd_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-static void amd_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, amd_pre_reset, ata_std_softreset, NULL,
-			   ata_std_postreset);
-}
-
 static int amd_cable_detect(struct ata_port *ap)
 {
 	static const u32 bitmask[2] = {0x03, 0x0C};
@@ -300,13 +294,6 @@ static int nv_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-static void nv_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, nv_pre_reset,
-			       ata_std_softreset, NULL,
-			       ata_std_postreset);
-}
-
 /**
  *	nv100_set_piomode	-	set initial PIO mode data
  *	@ap: ATA interface
@@ -358,7 +345,7 @@ static struct scsi_host_template amd_sht = {
 
 static const struct ata_port_operations amd_base_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
-	.error_handler	= amd_error_handler,
+	.prereset	= amd_pre_reset,
 };
 
 static struct ata_port_operations amd33_port_ops = {
@@ -393,7 +380,7 @@ static const struct ata_port_operations nv_base_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 	.cable_detect	= ata_cable_ignore,
 	.mode_filter	= nv_mode_filter,
-	.error_handler	= nv_error_handler,
+	.prereset	= nv_pre_reset,
 	.host_stop	= nv_host_stop,
 };
 
@@ -483,10 +470,10 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			.port_ops = &amd100_port_ops
 		}
 	};
-	struct ata_port_info pi;
-	const struct ata_port_info *ppi[] = { &pi, NULL };
+	const struct ata_port_info *ppi[] = { NULL, NULL };
 	static int printed_version;
 	int type = id->driver_data;
+	void *hpriv = NULL;
 	u8 fifo;
 	int rc;
 
@@ -511,7 +498,7 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	/*
 	 * Okay, type is determined now.  Apply type-specific workarounds.
 	 */
-	pi = info[type];
+	ppi[0] = &info[type];
 
 	if (type < 3)
 		ata_pci_clear_simplex(pdev);
@@ -530,11 +517,11 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		u32 udma;
 
 		pci_read_config_dword(pdev, 0x60, &udma);
-		pi.private_data = (void *)(unsigned long)udma;
+		hpriv = (void *)(unsigned long)udma;
 	}
 
 	/* And fire it up */
-	return ata_pci_init_one(pdev, ppi, &amd_sht);
+	return ata_pci_init_one(pdev, ppi, &amd_sht, hpriv);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index 698a53c96111..ebd15cadf15f 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -430,7 +430,7 @@ static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 
 	BUG_ON(ppi[0] == NULL);
 
-	return ata_pci_init_one(pdev, ppi, &artop_sht);
+	return ata_pci_init_one(pdev, ppi, &artop_sht, NULL);
 }
 
 static const struct pci_device_id artop_pci_tbl[] = {
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 6fe433ba62bd..0bea7e75d2d6 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -248,7 +248,7 @@ static int atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.port_ops = &atiixp_port_ops
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(dev, ppi, &atiixp_sht);
+	return ata_pci_init_one(dev, ppi, &atiixp_sht, NULL);
 }
 
 static const struct pci_device_id atiixp[] = {
diff --git a/drivers/ata/pata_cmd640.c b/drivers/ata/pata_cmd640.c
index efd2bb5747b4..27219b00edf4 100644
--- a/drivers/ata/pata_cmd640.c
+++ b/drivers/ata/pata_cmd640.c
@@ -224,7 +224,7 @@ static int cmd640_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	cmd640_hardware_init(pdev);
 
-	return ata_pci_init_one(pdev, ppi, &cmd640_sht);
+	return ata_pci_init_one(pdev, ppi, &cmd640_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index bfd72ef9cd31..f0e566623614 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -373,7 +373,7 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	pci_write_config_byte(pdev, UDIDETCR0, 0xF0);
 #endif
 
-	return ata_pci_init_one(pdev, ppi, &cmd64x_sht);
+	return ata_pci_init_one(pdev, ppi, &cmd64x_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
index c632ce499d33..ac3ad55d7c3c 100644
--- a/drivers/ata/pata_cs5530.c
+++ b/drivers/ata/pata_cs5530.c
@@ -325,7 +325,7 @@ static int cs5530_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		ppi[1] = &info_palmax_secondary;
 
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(pdev, ppi, &cs5530_sht);
+	return ata_pci_init_one(pdev, ppi, &cs5530_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
index d78cf95cbe45..5c0762ebf58c 100644
--- a/drivers/ata/pata_cs5535.c
+++ b/drivers/ata/pata_cs5535.c
@@ -199,7 +199,7 @@ static int cs5535_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	rdmsr(ATAC_CH0D1_PIO, timings, dummy);
 	if (CS5535_BAD_PIO(timings))
 		wrmsr(ATAC_CH0D1_PIO, 0xF7F4F7F4UL, 0);
-	return ata_pci_init_one(dev, ppi, &cs5535_sht);
+	return ata_pci_init_one(dev, ppi, &cs5535_sht, NULL);
 }
 
 static const struct pci_device_id cs5535[] = {
diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c
index f7c0e4e319ed..2d34b9145dcb 100644
--- a/drivers/ata/pata_cs5536.c
+++ b/drivers/ata/pata_cs5536.c
@@ -261,7 +261,7 @@ static int cs5536_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		return -ENODEV;
 	}
 
-	return ata_pci_init_one(dev, ppi, &cs5536_sht);
+	return ata_pci_init_one(dev, ppi, &cs5536_sht, NULL);
 }
 
 static const struct pci_device_id cs5536[] = {
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
index cbd6670ea0de..ae14969e1dfe 100644
--- a/drivers/ata/pata_cypress.c
+++ b/drivers/ata/pata_cypress.c
@@ -136,7 +136,7 @@ static int cy82c693_init_one(struct pci_dev *pdev, const struct pci_device_id *i
 	if (PCI_FUNC(pdev->devfn) != 1)
 		return -ENODEV;
 
-	return ata_pci_init_one(pdev, ppi, &cy82c693_sht);
+	return ata_pci_init_one(pdev, ppi, &cy82c693_sht, NULL);
 }
 
 static const struct pci_device_id cy82c693[] = {
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index 0260edac2370..2f5b4848456a 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -275,7 +275,7 @@ static int efar_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &efar_sht);
+	return ata_pci_init_one(pdev, ppi, &efar_sht, NULL);
 }
 
 static const struct pci_device_id efar_pci_tbl[] = {
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
index b62d398ed84b..c2d4923d4db7 100644
--- a/drivers/ata/pata_hpt366.c
+++ b/drivers/ata/pata_hpt366.c
@@ -356,9 +356,9 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.udma_mask = ATA_UDMA4,
 		.port_ops = &hpt366_port_ops
 	};
-	struct ata_port_info info = info_hpt366;
-	const struct ata_port_info *ppi[] = { &info, NULL };
+	const struct ata_port_info *ppi[] = { &info_hpt366, NULL };
 
+	void *hpriv = NULL;
 	u32 class_rev;
 	u32 reg1;
 	int rc;
@@ -383,17 +383,17 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	/* info_hpt366 is safe against re-entry so we can scribble on it */
 	switch((reg1 & 0x700) >> 8) {
 		case 5:
-			info.private_data = &hpt366_40;
+			hpriv = &hpt366_40;
 			break;
 		case 9:
-			info.private_data = &hpt366_25;
+			hpriv = &hpt366_25;
 			break;
 		default:
-			info.private_data = &hpt366_33;
+			hpriv = &hpt366_33;
 			break;
 	}
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(dev, ppi, &hpt36x_sht);
+	return ata_pci_init_one(dev, ppi, &hpt36x_sht, hpriv);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
index a43c19753669..fb37e3a161fc 100644
--- a/drivers/ata/pata_hpt37x.c
+++ b/drivers/ata/pata_hpt37x.c
@@ -831,10 +831,8 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	};
 
 	static const int MHz[4] = { 33, 40, 50, 66 };
-	const struct ata_port_info *port;
 	void *private_data = NULL;
-	struct ata_port_info port_info;
-	const struct ata_port_info *ppi[] = { &port_info, NULL };
+	const struct ata_port_info *ppi[] = { NULL, NULL };
 
 	u8 irqmask;
 	u32 class_rev;
@@ -866,17 +864,17 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 
 		switch(class_rev) {
 			case 3:
-				port = &info_hpt370;
+				ppi[0] = &info_hpt370;
 				chip_table = &hpt370;
 				prefer_dpll = 0;
 				break;
 			case 4:
-				port = &info_hpt370a;
+				ppi[0] = &info_hpt370a;
 				chip_table = &hpt370a;
 				prefer_dpll = 0;
 				break;
 			case 5:
-				port = &info_hpt372;
+				ppi[0] = &info_hpt372;
 				chip_table = &hpt372;
 				break;
 			default:
@@ -889,21 +887,21 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 				/* 372N if rev >= 2*/
 				if (class_rev >= 2)
 					return -ENODEV;
-				port = &info_hpt372;
+				ppi[0] = &info_hpt372;
 				chip_table = &hpt372a;
 				break;
 			case PCI_DEVICE_ID_TTI_HPT302:
 				/* 302N if rev > 1 */
 				if (class_rev > 1)
 					return -ENODEV;
-				port = &info_hpt372;
+				ppi[0] = &info_hpt372;
 				/* Check this */
 				chip_table = &hpt302;
 				break;
 			case PCI_DEVICE_ID_TTI_HPT371:
 				if (class_rev > 1)
 					return -ENODEV;
-				port = &info_hpt372;
+				ppi[0] = &info_hpt372;
 				chip_table = &hpt371;
 				/* Single channel device, master is not present
 				   but the BIOS (or us for non x86) must mark it
@@ -914,7 +912,7 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 				break;
 			case PCI_DEVICE_ID_TTI_HPT374:
 				chip_table = &hpt374;
-				port = &info_hpt374;
+				ppi[0] = &info_hpt374;
 				break;
 			default:
 				printk(KERN_ERR "pata_hpt37x: PCI table is bogus please report (%d).\n", dev->device);
@@ -993,7 +991,7 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		int dpll, adjust;
 
 		/* Compute DPLL */
-		dpll = (port->udma_mask & 0xC0) ? 3 : 2;
+		dpll = (ppi[0]->udma_mask & 0xC0) ? 3 : 2;
 
 		f_low = (MHz[clock_slot] * 48) / MHz[dpll];
 		f_high = f_low + 2;
@@ -1033,19 +1031,16 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		 *	about lack of UDMA133 support on lower clocks
  		 */
 
-		if (clock_slot < 2 && port == &info_hpt370)
-			port = &info_hpt370_33;
-		if (clock_slot < 2 && port == &info_hpt370a)
-			port = &info_hpt370a_33;
+		if (clock_slot < 2 && ppi[0] == &info_hpt370)
+			ppi[0] = &info_hpt370_33;
+		if (clock_slot < 2 && ppi[0] == &info_hpt370a)
+			ppi[0] = &info_hpt370a_33;
 		printk(KERN_INFO "pata_hpt37x: %s using %dMHz bus clock.\n",
 		       chip_table->name, MHz[clock_slot]);
 	}
 
 	/* Now kick off ATA set up */
-	port_info = *port;
-	port_info.private_data = private_data;
-
-	return ata_pci_init_one(dev, ppi, &hpt37x_sht);
+	return ata_pci_init_one(dev, ppi, &hpt37x_sht, private_data);
 }
 
 static const struct pci_device_id hpt37x[] = {
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
index 2c178c30116c..c774be93ae04 100644
--- a/drivers/ata/pata_hpt3x2n.c
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -458,8 +458,7 @@ static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.udma_mask = ATA_UDMA6,
 		.port_ops = &hpt3x2n_port_ops
 	};
-	struct ata_port_info port = info;
-	const struct ata_port_info *ppi[] = { &port, NULL };
+	const struct ata_port_info *ppi[] = { &info, NULL };
 
 	u8 irqmask;
 	u32 class_rev;
@@ -468,6 +467,7 @@ static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	unsigned int f_low, f_high;
 	int adjust;
 	unsigned long iobase = pci_resource_start(dev, 4);
+	void *hpriv = NULL;
 	int rc;
 
 	rc = pcim_enable_device(dev);
@@ -554,9 +554,8 @@ static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	       pci_mhz);
 	/* Set our private data up. We only need a few flags so we use
 	   it directly */
-	port.private_data = NULL;
 	if (pci_mhz > 60) {
-		port.private_data = (void *)PCI66;
+		hpriv = (void *)PCI66;
 		/*
 		 * On  HPT371N, if ATA clock is 66 MHz we must set bit 2 in
 		 * the MISC. register to stretch the UltraDMA Tss timing.
@@ -567,7 +566,7 @@ static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	}
 
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(dev, ppi, &hpt3x2n_sht);
+	return ata_pci_init_one(dev, ppi, &hpt3x2n_sht, hpriv);
 }
 
 static const struct pci_device_id hpt3x2n[] = {
diff --git a/drivers/ata/pata_it8213.c b/drivers/ata/pata_it8213.c
index 291a0d6e2434..d23a46b75028 100644
--- a/drivers/ata/pata_it8213.c
+++ b/drivers/ata/pata_it8213.c
@@ -287,7 +287,7 @@ static int it8213_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &it8213_sht);
+	return ata_pci_init_one(pdev, ppi, &it8213_sht, NULL);
 }
 
 static const struct pci_device_id it8213_pci_tbl[] = {
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index 63c5cf0d1fee..6a8a4ddf5bfe 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -722,7 +722,7 @@ static int it821x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	else
 		ppi[0] = &info_smart;
 
-	return ata_pci_init_one(pdev, ppi, &it821x_sht);
+	return ata_pci_init_one(pdev, ppi, &it821x_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index 859e47a600cc..317f3474e0ba 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -105,20 +105,6 @@ static int jmicron_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	jmicron_error_handler - Setup and error handler
- *	@ap: Port to handle
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void jmicron_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, jmicron_pre_reset, ata_std_softreset, NULL,
-			   ata_std_postreset);
-}
-
 /* No PIO or DMA methods needed for this device */
 
 static struct scsi_host_template jmicron_sht = {
@@ -127,7 +113,7 @@ static struct scsi_host_template jmicron_sht = {
 
 static struct ata_port_operations jmicron_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.error_handler		= jmicron_error_handler,
+	.prereset		= jmicron_pre_reset,
 };
 
 
@@ -158,7 +144,7 @@ static int jmicron_init_one (struct pci_dev *pdev, const struct pci_device_id *i
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
 
-	return ata_pci_init_one(pdev, ppi, &jmicron_sht);
+	return ata_pci_init_one(pdev, ppi, &jmicron_sht, NULL);
 }
 
 static const struct pci_device_id jmicron_pci_tbl[] = {
diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c
index d8da4f344c0a..d38e64cd6097 100644
--- a/drivers/ata/pata_marvell.c
+++ b/drivers/ata/pata_marvell.c
@@ -75,20 +75,6 @@ static int marvell_cable_detect(struct ata_port *ap)
 	return 0;	/* Our BUG macro needs the right markup */
 }
 
-/**
- *	marvell_error_handler - Setup and error handler
- *	@ap: Port to handle
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void marvell_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, marvell_pre_reset, ata_std_softreset, NULL,
-			   ata_std_postreset);
-}
-
 /* No PIO or DMA methods needed for this device */
 
 static struct scsi_host_template marvell_sht = {
@@ -98,7 +84,7 @@ static struct scsi_host_template marvell_sht = {
 static struct ata_port_operations marvell_ops = {
 	.inherits		= &ata_bmdma_port_ops,
 	.cable_detect		= marvell_cable_detect,
-	.error_handler		= marvell_error_handler,
+	.prereset		= marvell_pre_reset,
 };
 
 
@@ -142,7 +128,7 @@ static int marvell_init_one (struct pci_dev *pdev, const struct pci_device_id *i
 	if (pdev->device == 0x6101)
 		ppi[1] = &ata_dummy_port_info;
 
-	return ata_pci_init_one(pdev, ppi, &marvell_sht);
+	return ata_pci_init_one(pdev, ppi, &marvell_sht, NULL);
 }
 
 static const struct pci_device_id marvell_pci_tbl[] = {
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
index ae50a5e85cf1..349182840d24 100644
--- a/drivers/ata/pata_netcell.c
+++ b/drivers/ata/pata_netcell.c
@@ -71,7 +71,7 @@ static int netcell_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	ata_pci_clear_simplex(pdev);
 
 	/* And let the library code do the work */
-	return ata_pci_init_one(pdev, port_info, &netcell_sht);
+	return ata_pci_init_one(pdev, port_info, &netcell_sht, NULL);
 }
 
 static const struct pci_device_id netcell_pci_tbl[] = {
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index 1bdca8f1e767..5a043e426480 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -163,7 +163,7 @@ static int ns87410_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.port_ops = &ns87410_port_ops
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(dev, ppi, &ns87410_sht);
+	return ata_pci_init_one(dev, ppi, &ns87410_sht, NULL);
 }
 
 static const struct pci_device_id ns87410[] = {
diff --git a/drivers/ata/pata_ns87415.c b/drivers/ata/pata_ns87415.c
index 42508940e4a9..cdd79d6fc0ee 100644
--- a/drivers/ata/pata_ns87415.c
+++ b/drivers/ata/pata_ns87415.c
@@ -375,7 +375,7 @@ static int ns87415_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	pci_write_config_byte(pdev, 0x55, 0xEE);
 	/* Select PIO0 8bit clocking */
 	pci_write_config_byte(pdev, 0x54, 0xB7);
-	return ata_pci_init_one(pdev, ppi, &ns87415_sht);
+	return ata_pci_init_one(pdev, ppi, &ns87415_sht, NULL);
 }
 
 static const struct pci_device_id ns87415_pci_tbl[] = {
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index 9e3afadbd04a..7001b756819e 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -263,7 +263,7 @@ static int oldpiix_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &oldpiix_sht);
+	return ata_pci_init_one(pdev, ppi, &oldpiix_sht, NULL);
 }
 
 static const struct pci_device_id oldpiix_pci_tbl[] = {
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index 8601d9c3cb39..5a5f20e03fc0 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -188,7 +188,7 @@ static int opti_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!printed_version++)
 		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(dev, ppi, &opti_sht);
+	return ata_pci_init_one(dev, ppi, &opti_sht, NULL);
 }
 
 static const struct pci_device_id opti[] = {
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index c376f9ef77c8..ba2819ff964b 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -445,7 +445,7 @@ static int optidma_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (optiplus_with_udma(dev))
 		ppi[0] = &info_82c700_udma;
 
-	return ata_pci_init_one(dev, ppi, &optidma_sht);
+	return ata_pci_init_one(dev, ppi, &optidma_sht, NULL);
 }
 
 static const struct pci_device_id optidma[] = {
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
index 5545fbab6a7e..8214100e3ac1 100644
--- a/drivers/ata/pata_pdc202xx_old.c
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -324,7 +324,7 @@ static int pdc202xx_init_one(struct pci_dev *dev, const struct pci_device_id *id
 				return -ENODEV;
 		}
 	}
-	return ata_pci_init_one(dev, ppi, &pdc202xx_sht);
+	return ata_pci_init_one(dev, ppi, &pdc202xx_sht, NULL);
 }
 
 static const struct pci_device_id pdc202xx[] = {
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
index 145d5ba92795..9ab84fc3798d 100644
--- a/drivers/ata/pata_radisys.c
+++ b/drivers/ata/pata_radisys.c
@@ -228,7 +228,7 @@ static int radisys_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &radisys_sht);
+	return ata_pci_init_one(pdev, ppi, &radisys_sht, NULL);
 }
 
 static const struct pci_device_id radisys_pci_tbl[] = {
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
index 04be6aee4354..462b72a31280 100644
--- a/drivers/ata/pata_rz1000.c
+++ b/drivers/ata/pata_rz1000.c
@@ -99,7 +99,7 @@ static int rz1000_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 		printk(KERN_DEBUG DRV_NAME " version " DRV_VERSION "\n");
 
 	if (rz1000_fifo_disable(pdev) == 0)
-		return ata_pci_init_one(pdev, ppi, &rz1000_sht);
+		return ata_pci_init_one(pdev, ppi, &rz1000_sht, NULL);
 
 	printk(KERN_ERR DRV_NAME ": failed to disable read-ahead on chipset..\n");
 	/* Not safe to use so skip */
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
index 38c7fb0bebe9..42efacf73c79 100644
--- a/drivers/ata/pata_sc1200.c
+++ b/drivers/ata/pata_sc1200.c
@@ -213,7 +213,7 @@ static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	/* Can't enable port 2 yet, see top comments */
 	const struct ata_port_info *ppi[] = { &info, &ata_dummy_port_info };
 
-	return ata_pci_init_one(dev, ppi, &sc1200_sht);
+	return ata_pci_init_one(dev, ppi, &sc1200_sht, NULL);
 }
 
 static const struct pci_device_id sc1200[] = {
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index 515b5b70a555..2f4f9b0f89de 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -461,7 +461,7 @@ static int serverworks_init_one(struct pci_dev *pdev, const struct pci_device_id
 	if (pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB5IDE)
 		ata_pci_clear_simplex(pdev);
 
-	return ata_pci_init_one(pdev, ppi, &serverworks_sht);
+	return ata_pci_init_one(pdev, ppi, &serverworks_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index 5313deeffa6d..63fafc6d6da3 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -362,7 +362,7 @@ static int __devinit sil680_init_one(struct pci_dev *pdev,
 				 &sil680_sht);
 
 use_ioports:
-	return ata_pci_init_one(pdev, ppi, &sil680_sht);
+	return ata_pci_init_one(pdev, ppi, &sil680_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index 32be13ba5f06..28abfc26e7a4 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -690,8 +690,7 @@ static void sis_fixup(struct pci_dev *pdev, struct sis_chipset *sis)
 static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	static int printed_version;
-	struct ata_port_info port;
-	const struct ata_port_info *ppi[] = { &port, NULL };
+	const struct ata_port_info *ppi[] = { NULL, NULL };
 	struct pci_dev *host = NULL;
 	struct sis_chipset *chipset = NULL;
 	struct sis_chipset *sets;
@@ -831,12 +830,11 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (chipset == NULL)
 		return -ENODEV;
 
-	port = *chipset->info;
-	port.private_data = chipset;
+	ppi[0] = chipset->info;
 
 	sis_fixup(pdev, chipset);
 
-	return ata_pci_init_one(pdev, ppi, &sis_sht);
+	return ata_pci_init_one(pdev, ppi, &sis_sht, chipset);
 }
 
 static const struct pci_device_id sis_pci_tbl[] = {
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index 2d14b2505c7d..1d97f920bd2b 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -323,7 +323,7 @@ static int sl82c105_init_one(struct pci_dev *dev, const struct pci_device_id *id
 	val |= CTRL_P0EN | CTRL_P0F16 | CTRL_P1F16;
 	pci_write_config_dword(dev, 0x40, val);
 
-	return ata_pci_init_one(dev, ppi, &sl82c105_sht);
+	return ata_pci_init_one(dev, ppi, &sl82c105_sht, NULL);
 }
 
 static const struct pci_device_id sl82c105[] = {
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index 86dc66c37389..f07b0e5df222 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -206,7 +206,7 @@ static int triflex_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!printed_version++)
 		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(dev, ppi, &triflex_sht);
+	return ata_pci_init_one(dev, ppi, &triflex_sht, NULL);
 }
 
 static const struct pci_device_id triflex[] = {
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index e66bb85ad3d1..f4092cbd566f 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -442,8 +442,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		.udma_mask = ATA_UDMA6,	/* FIXME: should check north bridge */
 		.port_ops = &via_port_ops
 	};
-	struct ata_port_info type;
-	const struct ata_port_info *ppi[] = { &type, NULL };
+	const struct ata_port_info *ppi[] = { NULL, NULL };
 	struct pci_dev *isa = NULL;
 	const struct via_isa_bridge *config;
 	static int printed_version;
@@ -491,25 +490,25 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	switch(config->flags & VIA_UDMA) {
 		case VIA_UDMA_NONE:
 			if (config->flags & VIA_NO_UNMASK)
-				type = via_mwdma_info_borked;
+				ppi[0] = &via_mwdma_info_borked;
 			else
-				type = via_mwdma_info;
+				ppi[0] = &via_mwdma_info;
 			break;
 		case VIA_UDMA_33:
-			type = via_udma33_info;
+			ppi[0] = &via_udma33_info;
 			break;
 		case VIA_UDMA_66:
-			type = via_udma66_info;
+			ppi[0] = &via_udma66_info;
 			/* The 66 MHz devices require we enable the clock */
 			pci_read_config_dword(pdev, 0x50, &timing);
 			timing |= 0x80008;
 			pci_write_config_dword(pdev, 0x50, timing);
 			break;
 		case VIA_UDMA_100:
-			type = via_udma100_info;
+			ppi[0] = &via_udma100_info;
 			break;
 		case VIA_UDMA_133:
-			type = via_udma133_info;
+			ppi[0] = &via_udma133_info;
 			break;
 		default:
 			WARN_ON(1);
@@ -524,9 +523,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	/* We have established the device type, now fire it up */
-	type.private_data = (void *)config;
-
-	return ata_pci_init_one(pdev, ppi, &via_sht);
+	return ata_pci_init_one(pdev, ppi, &via_sht, (void *)config);
 }
 
 #ifdef CONFIG_PM
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 5494119854de..e7f10a88efe1 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1020,7 +1020,7 @@ struct pci_dev;
 
 extern int ata_pci_init_one(struct pci_dev *pdev,
 			    const struct ata_port_info * const * ppi,
-			    struct scsi_host_template *sht);
+			    struct scsi_host_template *sht, void *host_priv);
 extern void ata_pci_remove_one(struct pci_dev *pdev);
 #ifdef CONFIG_PM
 extern void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t mesg);
-- 
cgit v1.2.3


From 959471936241bd83da7d0a76411cef6772140fe6 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:49 +0900
Subject: libata: kill port_info->sht and ->irq_handler

libata core layer doesn't care about sht or ->irq_handler.  Those are
only of interest to the LLD during initialization.  This is confusing
and has caused several drivers to have duplicate unused initializers
for these fields.

Currently only sata_nv uses these fields.  Make sata_nv use
->private_data, which is supposed to carry LLD-specific information,
instead and kill ->sht and ->irq_handler.  nv_pi_priv structure is
defined and struct literals are used to initialize private_data.
Notational overhead is negligible.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/sata_nv.c  | 29 +++++++++++++++++------------
 include/linux/libata.h |  2 --
 2 files changed, 17 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 7b7ba0e26903..5637b082bc85 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -466,58 +466,61 @@ static struct ata_port_operations nv_swncq_ops = {
 	.port_start		= nv_swncq_port_start,
 };
 
+struct nv_pi_priv {
+	irq_handler_t			irq_handler;
+	struct scsi_host_template	*sht;
+};
+
+#define NV_PI_PRIV(_irq_handler, _sht) \
+	&(struct nv_pi_priv){ .irq_handler = _irq_handler, .sht = _sht }
+
 static const struct ata_port_info nv_port_info[] = {
 	/* generic */
 	{
-		.sht		= &nv_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
 		.port_ops	= &nv_generic_ops,
-		.irq_handler	= nv_generic_interrupt,
+		.private_data	= NV_PI_PRIV(nv_generic_interrupt, &nv_sht),
 	},
 	/* nforce2/3 */
 	{
-		.sht		= &nv_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
 		.port_ops	= &nv_nf2_ops,
-		.irq_handler	= nv_nf2_interrupt,
+		.private_data	= NV_PI_PRIV(nv_nf2_interrupt, &nv_sht),
 	},
 	/* ck804 */
 	{
-		.sht		= &nv_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
 		.port_ops	= &nv_ck804_ops,
-		.irq_handler	= nv_ck804_interrupt,
+		.private_data	= NV_PI_PRIV(nv_ck804_interrupt, &nv_sht),
 	},
 	/* ADMA */
 	{
-		.sht		= &nv_adma_sht,
 		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_MMIO | ATA_FLAG_NCQ,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
 		.port_ops	= &nv_adma_ops,
-		.irq_handler	= nv_adma_interrupt,
+		.private_data	= NV_PI_PRIV(nv_adma_interrupt, &nv_adma_sht),
 	},
 	/* SWNCQ */
 	{
-		.sht		= &nv_swncq_sht,
 		.flags	        = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_NCQ,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
 		.port_ops	= &nv_swncq_ops,
-		.irq_handler	= nv_swncq_interrupt,
+		.private_data	= NV_PI_PRIV(nv_swncq_interrupt, &nv_swncq_sht),
 	},
 };
 
@@ -2316,6 +2319,7 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	static int printed_version;
 	const struct ata_port_info *ppi[] = { NULL, NULL };
+	struct nv_pi_priv *ipriv;
 	struct ata_host *host;
 	struct nv_host_priv *hpriv;
 	int rc;
@@ -2352,6 +2356,7 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	ppi[0] = &nv_port_info[type];
+	ipriv = ppi[0]->private_data;
 	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
 	if (rc)
 		return rc;
@@ -2390,8 +2395,8 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		nv_swncq_host_init(host);
 
 	pci_set_master(pdev);
-	return ata_host_activate(host, pdev->irq, ppi[0]->irq_handler,
-				 IRQF_SHARED, ppi[0]->sht);
+	return ata_host_activate(host, pdev->irq, ipriv->irq_handler,
+				 IRQF_SHARED, ipriv->sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/include/linux/libata.h b/include/linux/libata.h
index e7f10a88efe1..88c6fa84ed74 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -748,14 +748,12 @@ struct ata_port_operations {
 };
 
 struct ata_port_info {
-	struct scsi_host_template	*sht;
 	unsigned long		flags;
 	unsigned long		link_flags;
 	unsigned long		pio_mask;
 	unsigned long		mwdma_mask;
 	unsigned long		udma_mask;
 	struct ata_port_operations *port_ops;
-	irq_handler_t		irq_handler;
 	void 			*private_data;
 };
 
-- 
cgit v1.2.3


From a1efdaba2dbd6fb89e23a87b66d3f4dd92c9f5af Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 12:22:50 +0900
Subject: libata: make reset related methods proper port operations

Currently reset methods are not specified directly in the
ata_port_operations table.  If a LLD wants to use custom reset
methods, it should construct and use a error_handler which uses those
reset methods.  It's done this way for two reasons.

First, the ops table already contained too many methods and adding
four more of them would noticeably increase the amount of necessary
boilerplate code all over low level drivers.

Second, as ->error_handler uses those reset methods, it can get
confusing.  ie. By overriding ->error_handler, those reset ops can be
made useless making layering a bit hazy.

Now that ops table uses inheritance, the first problem doesn't exist
anymore.  The second isn't completely solved but is relieved by
providing default values - most drivers can just override what it has
implemented and don't have to concern itself about higher level
callbacks.  In fact, there currently is no driver which actually
modifies error handling behavior.  Drivers which override
->error_handler just wraps the standard error handler only to prepare
the controller for EH.  I don't think making ops layering strict has
any noticeable benefit.

This patch makes ->prereset, ->softreset, ->hardreset, ->postreset and
their PMP counterparts propoer ops.  Default ops are provided in the
base ops tables and drivers are converted to override individual reset
methods instead of creating custom error_handler.

* ata_std_error_handler() doesn't use sata_std_hardreset() if SCRs
  aren't accessible.  sata_promise doesn't need to use separate
  error_handlers for PATA and SATA anymore.

* softreset is broken for sata_inic162x and sata_sx4.  As libata now
  always prefers hardreset, this doesn't really matter but the ops are
  forced to NULL using ATA_OP_NULL for documentation purpose.

* pata_hpt374 needs to use different prereset for the first and second
  PCI functions.  This used to be done by branching from
  hpt374_error_handler().  The proper way to do this is to use
  separate ops and port_info tables for each function.  Converted.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c          | 53 +++++++++++++--------------------------
 drivers/ata/ata_piix.c      | 21 ++++------------
 drivers/ata/libata-core.c   | 14 +++++++++--
 drivers/ata/libata-eh.c     | 25 +++++++++++++++++++
 drivers/ata/libata-pmp.c    | 47 ++++++++--------------------------
 drivers/ata/libata-sff.c    | 49 ++++++++++++------------------------
 drivers/ata/pata_artop.c    | 34 ++-----------------------
 drivers/ata/pata_atiixp.c   |  7 +-----
 drivers/ata/pata_bf54x.c    | 14 ++---------
 drivers/ata/pata_efar.c     | 15 +----------
 drivers/ata/pata_hpt37x.c   | 61 +++++++++++++++------------------------------
 drivers/ata/pata_hpt3x2n.c  | 16 ++----------
 drivers/ata/pata_icside.c   |  8 +-----
 drivers/ata/pata_it8213.c   | 15 +----------
 drivers/ata/pata_mpc52xx.c  | 10 --------
 drivers/ata/pata_mpiix.c    | 16 +-----------
 drivers/ata/pata_ns87410.c  | 16 +-----------
 drivers/ata/pata_oldpiix.c  | 16 +-----------
 drivers/ata/pata_opti.c     | 17 +------------
 drivers/ata/pata_optidma.c  | 17 +------------
 drivers/ata/pata_pdc2027x.c | 19 ++------------
 drivers/ata/pata_scc.c      | 15 +++--------
 drivers/ata/pata_sis.c      | 15 +----------
 drivers/ata/pata_sl82c105.c |  8 +-----
 drivers/ata/pata_triflex.c  |  7 +-----
 drivers/ata/pata_via.c      | 14 +----------
 drivers/ata/pdc_adma.c      | 11 +++-----
 drivers/ata/sata_fsl.c      | 12 +--------
 drivers/ata/sata_inic162x.c |  5 ++--
 drivers/ata/sata_mv.c       | 16 ++++++------
 drivers/ata/sata_nv.c       | 17 ++++---------
 drivers/ata/sata_promise.c  | 22 +++-------------
 drivers/ata/sata_qstor.c    |  6 +++--
 drivers/ata/sata_sil24.c    | 18 +++++++++----
 drivers/ata/sata_via.c      | 10 ++------
 drivers/scsi/ipr.c          |  3 ++-
 include/linux/libata.h      | 19 +++++++-------
 37 files changed, 186 insertions(+), 502 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index dacb3ef0c3e6..3efa9904f7a0 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -252,9 +252,18 @@ static void ahci_freeze(struct ata_port *ap);
 static void ahci_thaw(struct ata_port *ap);
 static void ahci_pmp_attach(struct ata_port *ap);
 static void ahci_pmp_detach(struct ata_port *ap);
+static int ahci_softreset(struct ata_link *link, unsigned int *class,
+			  unsigned long deadline);
+static int ahci_hardreset(struct ata_link *link, unsigned int *class,
+			  unsigned long deadline);
+static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class,
+				 unsigned long deadline);
+static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
+				unsigned long deadline);
+static void ahci_postreset(struct ata_link *link, unsigned int *class);
+static int ahci_pmp_softreset(struct ata_link *link, unsigned int *class,
+			      unsigned long deadline);
 static void ahci_error_handler(struct ata_port *ap);
-static void ahci_vt8251_error_handler(struct ata_port *ap);
-static void ahci_p5wdh_error_handler(struct ata_port *ap);
 static void ahci_post_internal_cmd(struct ata_queued_cmd *qc);
 static int ahci_port_resume(struct ata_port *ap);
 static void ahci_dev_config(struct ata_device *dev);
@@ -293,6 +302,10 @@ static struct ata_port_operations ahci_ops = {
 
 	.freeze			= ahci_freeze,
 	.thaw			= ahci_thaw,
+	.softreset		= ahci_softreset,
+	.hardreset		= ahci_hardreset,
+	.postreset		= ahci_postreset,
+	.pmp_softreset		= ahci_pmp_softreset,
 	.error_handler		= ahci_error_handler,
 	.post_internal_cmd	= ahci_post_internal_cmd,
 	.dev_config		= ahci_dev_config,
@@ -314,12 +327,12 @@ static struct ata_port_operations ahci_ops = {
 
 static struct ata_port_operations ahci_vt8251_ops = {
 	.inherits		= &ahci_ops,
-	.error_handler		= ahci_vt8251_error_handler,
+	.hardreset		= ahci_vt8251_hardreset,
 };
 
 static struct ata_port_operations ahci_p5wdh_ops = {
 	.inherits		= &ahci_ops,
-	.error_handler		= ahci_p5wdh_error_handler,
+	.hardreset		= ahci_p5wdh_hardreset,
 };
 
 #define AHCI_HFLAGS(flags)	.private_data	= (void *)(flags)
@@ -1796,37 +1809,7 @@ static void ahci_error_handler(struct ata_port *ap)
 		ahci_start_engine(ap);
 	}
 
-	/* perform recovery */
-	sata_pmp_do_eh(ap, ata_std_prereset, ahci_softreset,
-		       ahci_hardreset, ahci_postreset,
-		       sata_pmp_std_prereset, ahci_pmp_softreset,
-		       sata_pmp_std_hardreset, sata_pmp_std_postreset);
-}
-
-static void ahci_vt8251_error_handler(struct ata_port *ap)
-{
-	if (!(ap->pflags & ATA_PFLAG_FROZEN)) {
-		/* restart engine */
-		ahci_stop_engine(ap);
-		ahci_start_engine(ap);
-	}
-
-	/* perform recovery */
-	ata_do_eh(ap, ata_std_prereset, ahci_softreset, ahci_vt8251_hardreset,
-		  ahci_postreset);
-}
-
-static void ahci_p5wdh_error_handler(struct ata_port *ap)
-{
-	if (!(ap->pflags & ATA_PFLAG_FROZEN)) {
-		/* restart engine */
-		ahci_stop_engine(ap);
-		ahci_start_engine(ap);
-	}
-
-	/* perform recovery */
-	ata_do_eh(ap, ata_std_prereset, ahci_softreset, ahci_p5wdh_hardreset,
-		  ahci_postreset);
+	sata_pmp_error_handler(ap);
 }
 
 static void ahci_post_internal_cmd(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index bb46b61a7c6b..eafb984313f6 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -162,15 +162,16 @@ struct piix_host_priv {
 
 static int piix_init_one(struct pci_dev *pdev,
 			 const struct pci_device_id *ent);
-static void piix_pata_error_handler(struct ata_port *ap);
+static int piix_pata_prereset(struct ata_link *link, unsigned long deadline);
 static void piix_set_piomode(struct ata_port *ap, struct ata_device *adev);
 static void piix_set_dmamode(struct ata_port *ap, struct ata_device *adev);
 static void ich_set_dmamode(struct ata_port *ap, struct ata_device *adev);
 static int ich_pata_cable_detect(struct ata_port *ap);
 static u8 piix_vmw_bmdma_status(struct ata_port *ap);
+static int piix_sidpr_hardreset(struct ata_link *link, unsigned int *class,
+				unsigned long deadline);
 static int piix_sidpr_scr_read(struct ata_port *ap, unsigned int reg, u32 *val);
 static int piix_sidpr_scr_write(struct ata_port *ap, unsigned int reg, u32 val);
-static void piix_sidpr_error_handler(struct ata_port *ap);
 #ifdef CONFIG_PM
 static int piix_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg);
 static int piix_pci_device_resume(struct pci_dev *pdev);
@@ -299,7 +300,7 @@ static struct ata_port_operations piix_pata_ops = {
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= piix_set_piomode,
 	.set_dmamode		= piix_set_dmamode,
-	.error_handler		= piix_pata_error_handler,
+	.prereset		= piix_pata_prereset,
 };
 
 static struct ata_port_operations piix_vmw_ops = {
@@ -319,9 +320,9 @@ static struct ata_port_operations piix_sata_ops = {
 
 static struct ata_port_operations piix_sidpr_sata_ops = {
 	.inherits		= &piix_sata_ops,
+	.hardreset		= piix_sidpr_hardreset,
 	.scr_read		= piix_sidpr_scr_read,
 	.scr_write		= piix_sidpr_scr_write,
-	.error_handler		= piix_sidpr_error_handler,
 };
 
 static const struct piix_map_db ich5_map_db = {
@@ -645,12 +646,6 @@ static int piix_pata_prereset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-static void piix_pata_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, piix_pata_prereset, ata_std_softreset, NULL,
-			   ata_std_postreset);
-}
-
 /**
  *	piix_set_piomode - Initialize host controller PATA PIO timings
  *	@ap: Port whose timings we are configuring
@@ -1057,12 +1052,6 @@ static int piix_sidpr_hardreset(struct ata_link *link, unsigned int *class,
 	return -EAGAIN;
 }
 
-static void piix_sidpr_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, ata_std_prereset, ata_std_softreset,
-			   piix_sidpr_hardreset, ata_std_postreset);
-}
-
 #ifdef CONFIG_PM
 static int piix_broken_suspend(void)
 {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index aa6bcd79d60a..345f4a6865a8 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -76,6 +76,10 @@ const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
 const struct ata_port_operations ata_base_port_ops = {
 	.irq_clear		= ata_noop_irq_clear,
+	.prereset		= ata_std_prereset,
+	.hardreset		= sata_std_hardreset,
+	.postreset		= ata_std_postreset,
+	.error_handler		= ata_std_error_handler,
 };
 
 const struct ata_port_operations sata_port_ops = {
@@ -87,6 +91,11 @@ const struct ata_port_operations sata_port_ops = {
 
 const struct ata_port_operations sata_pmp_port_ops = {
 	.inherits		= &sata_port_ops,
+
+	.pmp_prereset		= sata_pmp_std_prereset,
+	.pmp_hardreset		= sata_pmp_std_hardreset,
+	.pmp_postreset		= sata_pmp_std_postreset,
+	.error_handler		= sata_pmp_error_handler,
 };
 
 const struct ata_port_operations ata_sff_port_ops = {
@@ -97,6 +106,7 @@ const struct ata_port_operations ata_sff_port_ops = {
 
 	.freeze			= ata_bmdma_freeze,
 	.thaw			= ata_bmdma_thaw,
+	.softreset		= ata_std_softreset,
 	.error_handler		= ata_bmdma_error_handler,
 	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
 
@@ -7935,7 +7945,6 @@ EXPORT_SYMBOL_GPL(ata_bmdma_status);
 EXPORT_SYMBOL_GPL(ata_bmdma_stop);
 EXPORT_SYMBOL_GPL(ata_bmdma_freeze);
 EXPORT_SYMBOL_GPL(ata_bmdma_thaw);
-EXPORT_SYMBOL_GPL(ata_bmdma_drive_eh);
 EXPORT_SYMBOL_GPL(ata_bmdma_error_handler);
 EXPORT_SYMBOL_GPL(ata_bmdma_post_internal_cmd);
 EXPORT_SYMBOL_GPL(ata_port_probe);
@@ -8005,7 +8014,7 @@ EXPORT_SYMBOL_GPL(sata_pmp_qc_defer_cmd_switch);
 EXPORT_SYMBOL_GPL(sata_pmp_std_prereset);
 EXPORT_SYMBOL_GPL(sata_pmp_std_hardreset);
 EXPORT_SYMBOL_GPL(sata_pmp_std_postreset);
-EXPORT_SYMBOL_GPL(sata_pmp_do_eh);
+EXPORT_SYMBOL_GPL(sata_pmp_error_handler);
 
 EXPORT_SYMBOL_GPL(__ata_ehi_push_desc);
 EXPORT_SYMBOL_GPL(ata_ehi_push_desc);
@@ -8024,6 +8033,7 @@ EXPORT_SYMBOL_GPL(ata_eh_thaw_port);
 EXPORT_SYMBOL_GPL(ata_eh_qc_complete);
 EXPORT_SYMBOL_GPL(ata_eh_qc_retry);
 EXPORT_SYMBOL_GPL(ata_do_eh);
+EXPORT_SYMBOL_GPL(ata_std_error_handler);
 EXPORT_SYMBOL_GPL(ata_irq_on);
 EXPORT_SYMBOL_GPL(ata_dev_try_classify);
 
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 0d0a2c0ab9e7..ec32082356cb 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2814,6 +2814,7 @@ void ata_eh_finish(struct ata_port *ap)
 /**
  *	ata_do_eh - do standard error handling
  *	@ap: host port to handle error for
+ *
  *	@prereset: prereset method (can be NULL)
  *	@softreset: softreset method (can be NULL)
  *	@hardreset: hardreset method (can be NULL)
@@ -2844,6 +2845,30 @@ void ata_do_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
 	ata_eh_finish(ap);
 }
 
+/**
+ *	ata_std_error_handler - standard error handler
+ *	@ap: host port to handle error for
+ *
+ *	Standard error handler
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep).
+ */
+void ata_std_error_handler(struct ata_port *ap)
+{
+	struct ata_port_operations *ops = ap->ops;
+	ata_reset_fn_t hardreset = ops->hardreset;
+
+	/* sata_std_hardreset is inherited to all drivers from
+	 * ata_base_port_ops.  Ignore it if SCR access is not
+	 * available.
+	 */
+	if (hardreset == sata_std_hardreset && !sata_scr_valid(&ap->link))
+		hardreset = NULL;
+
+	ata_do_eh(ap, ops->prereset, ops->softreset, hardreset, ops->postreset);
+}
+
 #ifdef CONFIG_PM
 /**
  *	ata_eh_handle_port_suspend - perform port suspend operation
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 39e036c8a2bc..a7cb1498c9b2 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -962,14 +962,6 @@ static int sata_pmp_handle_link_fail(struct ata_link *link, int *link_tries)
 /**
  *	sata_pmp_eh_recover - recover PMP-enabled port
  *	@ap: ATA port to recover
- *	@prereset: prereset method (can be NULL)
- *	@softreset: softreset method
- *	@hardreset: hardreset method
- *	@postreset: postreset method (can be NULL)
- *	@pmp_prereset: PMP prereset method (can be NULL)
- *	@pmp_softreset: PMP softreset method (can be NULL)
- *	@pmp_hardreset: PMP hardreset method (can be NULL)
- *	@pmp_postreset: PMP postreset method (can be NULL)
  *
  *	Drive EH recovery operation for PMP enabled port @ap.  This
  *	function recovers host and PMP ports with proper retrials and
@@ -982,12 +974,9 @@ static int sata_pmp_handle_link_fail(struct ata_link *link, int *link_tries)
  *	RETURNS:
  *	0 on success, -errno on failure.
  */
-static int sata_pmp_eh_recover(struct ata_port *ap,
-		ata_prereset_fn_t prereset, ata_reset_fn_t softreset,
-		ata_reset_fn_t hardreset, ata_postreset_fn_t postreset,
-		ata_prereset_fn_t pmp_prereset, ata_reset_fn_t pmp_softreset,
-		ata_reset_fn_t pmp_hardreset, ata_postreset_fn_t pmp_postreset)
+static int sata_pmp_eh_recover(struct ata_port *ap)
 {
+	struct ata_port_operations *ops = ap->ops;
 	int pmp_tries, link_tries[SATA_PMP_MAX_PORTS];
 	struct ata_link *pmp_link = &ap->link;
 	struct ata_device *pmp_dev = pmp_link->device;
@@ -1005,8 +994,8 @@ static int sata_pmp_eh_recover(struct ata_port *ap,
  retry:
 	/* PMP attached? */
 	if (!ap->nr_pmp_links) {
-		rc = ata_eh_recover(ap, prereset, softreset, hardreset,
-				    postreset, NULL);
+		rc = ata_eh_recover(ap, ops->prereset, ops->softreset,
+				    ops->hardreset, ops->postreset, NULL);
 		if (rc) {
 			ata_link_for_each_dev(dev, &ap->link)
 				ata_dev_disable(dev);
@@ -1024,8 +1013,8 @@ static int sata_pmp_eh_recover(struct ata_port *ap,
 	}
 
 	/* recover pmp */
-	rc = sata_pmp_eh_recover_pmp(ap, prereset, softreset, hardreset,
-				     postreset);
+	rc = sata_pmp_eh_recover_pmp(ap, ops->prereset, ops->softreset,
+				     ops->hardreset, ops->postreset);
 	if (rc)
 		goto pmp_fail;
 
@@ -1035,8 +1024,8 @@ static int sata_pmp_eh_recover(struct ata_port *ap,
 		goto pmp_fail;
 
 	/* recover links */
-	rc = ata_eh_recover(ap, pmp_prereset, pmp_softreset, pmp_hardreset,
-			    pmp_postreset, &link);
+	rc = ata_eh_recover(ap, ops->pmp_prereset, ops->pmp_softreset,
+			    ops->pmp_hardreset, ops->pmp_postreset, &link);
 	if (rc)
 		goto link_fail;
 
@@ -1132,16 +1121,8 @@ static int sata_pmp_eh_recover(struct ata_port *ap,
 }
 
 /**
- *	sata_pmp_do_eh - do standard error handling for PMP-enabled host
+ *	sata_pmp_error_handler - do standard error handling for PMP-enabled host
  *	@ap: host port to handle error for
- *	@prereset: prereset method (can be NULL)
- *	@softreset: softreset method
- *	@hardreset: hardreset method
- *	@postreset: postreset method (can be NULL)
- *	@pmp_prereset: PMP prereset method (can be NULL)
- *	@pmp_softreset: PMP softreset method (can be NULL)
- *	@pmp_hardreset: PMP hardreset method (can be NULL)
- *	@pmp_postreset: PMP postreset method (can be NULL)
  *
  *	Perform standard error handling sequence for PMP-enabled host
  *	@ap.
@@ -1149,16 +1130,10 @@ static int sata_pmp_eh_recover(struct ata_port *ap,
  *	LOCKING:
  *	Kernel thread context (may sleep).
  */
-void sata_pmp_do_eh(struct ata_port *ap,
-		ata_prereset_fn_t prereset, ata_reset_fn_t softreset,
-		ata_reset_fn_t hardreset, ata_postreset_fn_t postreset,
-		ata_prereset_fn_t pmp_prereset, ata_reset_fn_t pmp_softreset,
-		ata_reset_fn_t pmp_hardreset, ata_postreset_fn_t pmp_postreset)
+void sata_pmp_error_handler(struct ata_port *ap)
 {
 	ata_eh_autopsy(ap);
 	ata_eh_report(ap);
-	sata_pmp_eh_recover(ap, prereset, softreset, hardreset, postreset,
-			    pmp_prereset, pmp_softreset, pmp_hardreset,
-			    pmp_postreset);
+	sata_pmp_eh_recover(ap);
 	ata_eh_finish(ap);
 }
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 6223ec042c80..2a229a1d3211 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -396,28 +396,21 @@ void ata_bmdma_thaw(struct ata_port *ap)
 }
 
 /**
- *	ata_bmdma_drive_eh - Perform EH with given methods for BMDMA controller
+ *	ata_bmdma_error_handler - Stock error handler for BMDMA controller
  *	@ap: port to handle error for
- *	@prereset: prereset method (can be NULL)
- *	@softreset: softreset method (can be NULL)
- *	@hardreset: hardreset method (can be NULL)
- *	@postreset: postreset method (can be NULL)
  *
- *	Handle error for ATA BMDMA controller.  It can handle both
+ *	Stock error handler for BMDMA controller.  It can handle both
  *	PATA and SATA controllers.  Many controllers should be able to
  *	use this EH as-is or with some added handling before and
  *	after.
  *
- *	This function is intended to be used for constructing
- *	->error_handler callback by low level drivers.
- *
  *	LOCKING:
  *	Kernel thread context (may sleep)
  */
-void ata_bmdma_drive_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
-			ata_reset_fn_t softreset, ata_reset_fn_t hardreset,
-			ata_postreset_fn_t postreset)
+void ata_bmdma_error_handler(struct ata_port *ap)
 {
+	ata_reset_fn_t softreset = ap->ops->softreset;
+	ata_reset_fn_t hardreset = ap->ops->hardreset;
 	struct ata_queued_cmd *qc;
 	unsigned long flags;
 	int thaw = 0;
@@ -460,29 +453,19 @@ void ata_bmdma_drive_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
 		ata_eh_thaw_port(ap);
 
 	/* PIO and DMA engines have been stopped, perform recovery */
-	ata_do_eh(ap, prereset, softreset, hardreset, postreset);
-}
-
-/**
- *	ata_bmdma_error_handler - Stock error handler for BMDMA controller
- *	@ap: port to handle error for
- *
- *	Stock error handler for BMDMA controller.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep)
- */
-void ata_bmdma_error_handler(struct ata_port *ap)
-{
-	ata_reset_fn_t softreset = NULL, hardreset = NULL;
 
-	if (ap->ioaddr.ctl_addr)
-		softreset = ata_std_softreset;
-	if (sata_scr_valid(&ap->link))
-		hardreset = sata_std_hardreset;
+	/* ata_std_softreset and sata_std_hardreset are inherited to
+	 * all SFF drivers from ata_sff_port_ops.  Ignore softreset if
+	 * ctl isn't accessible.  Ignore hardreset if SCR access isn't
+	 * available.
+	 */
+	if (softreset == ata_std_softreset && !ap->ioaddr.ctl_addr)
+		softreset = NULL;
+	if (hardreset == sata_std_hardreset && !sata_scr_valid(&ap->link))
+		hardreset = NULL;
 
-	ata_bmdma_drive_eh(ap, ata_std_prereset, softreset, hardreset,
-			   ata_std_postreset);
+	ata_do_eh(ap, ap->ops->prereset, softreset, hardreset,
+		  ap->ops->postreset);
 }
 
 /**
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index ebd15cadf15f..b6d8c4d0e6c2 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -55,21 +55,6 @@ static int artop6210_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	artop6210_error_handler - Probe specified port on PATA host controller
- *	@ap: Port to probe
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void artop6210_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, artop6210_pre_reset,
-				    ata_std_softreset, NULL,
-				    ata_std_postreset);
-}
-
 /**
  *	artop6260_pre_reset	-	check for 40/80 pin
  *	@link: link
@@ -113,21 +98,6 @@ static int artop6260_cable_detect(struct ata_port *ap)
 	return ATA_CBL_PATA80;
 }
 
-/**
- *	artop6260_error_handler - Probe specified port on PATA host controller
- *	@ap: Port to probe
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void artop6260_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, artop6260_pre_reset,
-				    ata_std_softreset, NULL,
-				    ata_std_postreset);
-}
-
 /**
  *	artop6210_load_piomode - Load a set of PATA PIO timings
  *	@ap: Port whose timings we are configuring
@@ -322,7 +292,7 @@ static struct ata_port_operations artop6210_ops = {
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= artop6210_set_piomode,
 	.set_dmamode		= artop6210_set_dmamode,
-	.error_handler		= artop6210_error_handler,
+	.prereset		= artop6210_pre_reset,
 };
 
 static struct ata_port_operations artop6260_ops = {
@@ -330,7 +300,7 @@ static struct ata_port_operations artop6260_ops = {
 	.cable_detect		= artop6260_cable_detect,
 	.set_piomode		= artop6260_set_piomode,
 	.set_dmamode		= artop6260_set_dmamode,
-	.error_handler		= artop6260_error_handler,
+	.prereset		= artop6260_pre_reset,
 };
 
 
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 0bea7e75d2d6..56a65baddd4a 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -48,11 +48,6 @@ static int atiixp_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-static void atiixp_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, atiixp_pre_reset, ata_std_softreset, NULL,   ata_std_postreset);
-}
-
 static int atiixp_cable_detect(struct ata_port *ap)
 {
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
@@ -235,7 +230,7 @@ static struct ata_port_operations atiixp_port_ops = {
 	.cable_detect	= atiixp_cable_detect,
 	.set_piomode	= atiixp_set_piomode,
 	.set_dmamode	= atiixp_set_dmamode,
-	.error_handler	= atiixp_error_handler,
+	.prereset	= atiixp_pre_reset,
 };
 
 static int atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 7a22ef483061..eea275acb2a8 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1314,17 +1314,6 @@ static void bfin_std_postreset(struct ata_link *link, unsigned int *classes)
 	write_atapi_register(base, ATA_REG_CTRL, ap->ctl);
 }
 
-/**
- *	bfin_error_handler - Stock error handler for DMA controller
- *	@ap: port to handle error for
- */
-
-static void bfin_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, ata_std_prereset, bfin_std_softreset, NULL,
-			   bfin_std_postreset);
-}
-
 static void bfin_port_stop(struct ata_port *ap)
 {
 	dev_dbg(ap->dev, "in atapi port stop\n");
@@ -1385,7 +1374,8 @@ static const struct ata_port_operations bfin_pata_ops = {
 
 	.freeze			= bfin_bmdma_freeze,
 	.thaw			= bfin_bmdma_thaw,
-	.error_handler		= bfin_error_handler,
+	.softreset		= bfin_std_softreset,
+	.postreset		= bfin_std_postreset,
 	.post_internal_cmd	= bfin_bmdma_stop,
 
 	.irq_clear		= bfin_irq_clear,
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index 2f5b4848456a..1d839a57068e 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -48,19 +48,6 @@ static int efar_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	efar_probe_reset - Probe specified port on PATA host controller
- *	@ap: Port to probe
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void efar_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, efar_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	efar_cable_detect	-	check for 40/80 pin
  *	@ap: Port
@@ -241,7 +228,7 @@ static struct ata_port_operations efar_ops = {
 	.cable_detect		= efar_cable_detect,
 	.set_piomode		= efar_set_piomode,
 	.set_dmamode		= efar_set_dmamode,
-	.error_handler		= efar_error_handler,
+	.prereset		= efar_pre_reset,
 };
 
 
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
index fb37e3a161fc..c10fcd31418d 100644
--- a/drivers/ata/pata_hpt37x.c
+++ b/drivers/ata/pata_hpt37x.c
@@ -341,19 +341,7 @@ static int hpt37x_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	hpt37x_error_handler	-	reset the hpt374
- *	@ap: ATA port to reset
- *
- *	Perform probe for HPT37x, except for HPT374 channel 2
- */
-
-static void hpt37x_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, hpt37x_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
-static int hpt374_pre_reset(struct ata_link *link, unsigned long deadline)
+static int hpt374_fn1_pre_reset(struct ata_link *link, unsigned long deadline)
 {
 	static const struct pci_bits hpt37x_enable_bits[] = {
 		{ 0x50, 1, 0x04, 0x04 },
@@ -389,25 +377,6 @@ static int hpt374_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	hpt374_error_handler	-	reset the hpt374
- *	@classes:
- *
- *	The 374 cable detect is a little different due to the extra
- *	channels. The function 0 channels work like usual but function 1
- *	is special
- */
-
-static void hpt374_error_handler(struct ata_port *ap)
-{
-	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
-
-	if (!(PCI_FUNC(pdev->devfn) & 1))
-		hpt37x_error_handler(ap);
-	else
-		ata_bmdma_drive_eh(ap, hpt374_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	hpt370_set_piomode		-	PIO setup
  *	@ap: ATA interface
@@ -635,7 +604,7 @@ static struct ata_port_operations hpt370_port_ops = {
 	.mode_filter	= hpt370_filter,
 	.set_piomode	= hpt370_set_piomode,
 	.set_dmamode	= hpt370_set_dmamode,
-	.error_handler	= hpt37x_error_handler,
+	.prereset	= hpt37x_pre_reset,
 };
 
 /*
@@ -659,17 +628,17 @@ static struct ata_port_operations hpt372_port_ops = {
 
 	.set_piomode	= hpt372_set_piomode,
 	.set_dmamode	= hpt372_set_dmamode,
-	.error_handler	= hpt37x_error_handler,
+	.prereset	= hpt37x_pre_reset,
 };
 
 /*
  *	Configuration for HPT374. Mode setting works like 372 and friends
- *	but we have a different cable detection procedure.
+ *	but we have a different cable detection procedure for function 1.
  */
 
-static struct ata_port_operations hpt374_port_ops = {
+static struct ata_port_operations hpt374_fn1_port_ops = {
 	.inherits	= &hpt372_port_ops,
-	.error_handler	= hpt374_error_handler,
+	.prereset	= hpt374_fn1_pre_reset,
 };
 
 /**
@@ -821,13 +790,20 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.udma_mask = ATA_UDMA6,
 		.port_ops = &hpt372_port_ops
 	};
-	/* HPT374 - UDMA100 */
-	static const struct ata_port_info info_hpt374 = {
+	/* HPT374 - UDMA100, function 1 uses different prereset method */
+	static const struct ata_port_info info_hpt374_fn0 = {
+		.flags = ATA_FLAG_SLAVE_POSS,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = ATA_UDMA5,
+		.port_ops = &hpt372_port_ops
+	};
+	static const struct ata_port_info info_hpt374_fn1 = {
 		.flags = ATA_FLAG_SLAVE_POSS,
 		.pio_mask = 0x1f,
 		.mwdma_mask = 0x07,
 		.udma_mask = ATA_UDMA5,
-		.port_ops = &hpt374_port_ops
+		.port_ops = &hpt374_fn1_port_ops
 	};
 
 	static const int MHz[4] = { 33, 40, 50, 66 };
@@ -912,7 +888,10 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 				break;
 			case PCI_DEVICE_ID_TTI_HPT374:
 				chip_table = &hpt374;
-				ppi[0] = &info_hpt374;
+				if (!(PCI_FUNC(dev->devfn) & 1))
+					*ppi = &info_hpt374_fn0;
+				else
+					*ppi = &info_hpt374_fn1;
 				break;
 			default:
 				printk(KERN_ERR "pata_hpt37x: PCI table is bogus please report (%d).\n", dev->device);
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
index c774be93ae04..cd44ee3d3cc1 100644
--- a/drivers/ata/pata_hpt3x2n.c
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -148,7 +148,7 @@ static int hpt3x2n_cable_detect(struct ata_port *ap)
  *	Reset the hardware and state machine,
  */
 
-static int hpt3xn_pre_reset(struct ata_link *link, unsigned long deadline)
+static int hpt3x2n_pre_reset(struct ata_link *link, unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
@@ -159,18 +159,6 @@ static int hpt3xn_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	hpt3x2n_error_handler	-	probe the hpt3x2n bus
- *	@ap: ATA port to reset
- *
- *	Perform the probe reset handling for the 3x2N
- */
-
-static void hpt3x2n_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, hpt3xn_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	hpt3x2n_set_piomode		-	PIO setup
  *	@ap: ATA interface
@@ -355,7 +343,7 @@ static struct ata_port_operations hpt3x2n_port_ops = {
 	.cable_detect	= hpt3x2n_cable_detect,
 	.set_piomode	= hpt3x2n_set_piomode,
 	.set_dmamode	= hpt3x2n_set_dmamode,
-	.error_handler	= hpt3x2n_error_handler,
+	.prereset	= hpt3x2n_pre_reset,
 };
 
 /**
diff --git a/drivers/ata/pata_icside.c b/drivers/ata/pata_icside.c
index ff16b0eaa2c2..13d43e9dd279 100644
--- a/drivers/ata/pata_icside.c
+++ b/drivers/ata/pata_icside.c
@@ -332,12 +332,6 @@ static void pata_icside_postreset(struct ata_link *link, unsigned int *classes)
 	}
 }
 
-static void pata_icside_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, ata_std_prereset, ata_std_softreset, NULL,
-			   pata_icside_postreset);
-}
-
 static struct ata_port_operations pata_icside_port_ops = {
 	.inherits		= &ata_sff_port_ops,
 	/* no need to build any PRD tables for DMA */
@@ -350,7 +344,7 @@ static struct ata_port_operations pata_icside_port_ops = {
 
 	.cable_detect		= ata_cable_40wire,
 	.set_dmamode		= pata_icside_set_dmamode,
-	.error_handler		= pata_icside_error_handler,
+	.postreset		= pata_icside_postreset,
 	.post_internal_cmd	= pata_icside_bmdma_stop,
 };
 
diff --git a/drivers/ata/pata_it8213.c b/drivers/ata/pata_it8213.c
index d23a46b75028..84ab89e8a247 100644
--- a/drivers/ata/pata_it8213.c
+++ b/drivers/ata/pata_it8213.c
@@ -43,19 +43,6 @@ static int it8213_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	it8213_error_handler - Probe specified port on PATA host controller
- *	@ap: Port to probe
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void it8213_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, it8213_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	it8213_cable_detect	-	check for 40/80 pin
  *	@ap: Port
@@ -252,7 +239,7 @@ static struct ata_port_operations it8213_ops = {
 	.cable_detect		= it8213_cable_detect,
 	.set_piomode		= it8213_set_piomode,
 	.set_dmamode		= it8213_set_dmamode,
-	.error_handler		= it8213_error_handler,
+	.prereset		= it8213_pre_reset,
 };
 
 
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index ac7c0822b1a7..fec93196710e 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -255,15 +255,6 @@ mpc52xx_ata_dev_select(struct ata_port *ap, unsigned int device)
 	ata_std_dev_select(ap,device);
 }
 
-static void
-mpc52xx_ata_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, ata_std_prereset, ata_std_softreset, NULL,
-			ata_std_postreset);
-}
-
-
-
 static struct scsi_host_template mpc52xx_ata_sht = {
 	ATA_PIO_SHT(DRV_NAME),
 };
@@ -273,7 +264,6 @@ static struct ata_port_operations mpc52xx_ata_port_ops = {
 	.dev_select		= mpc52xx_ata_dev_select,
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= mpc52xx_ata_set_piomode,
-	.error_handler		= mpc52xx_ata_error_handler,
 	.post_internal_cmd	= ATA_OP_NULL,
 };
 
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index dab54f8a272d..1b9d0d412ebf 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -58,20 +58,6 @@ static int mpiix_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	mpiix_error_handler		-	probe reset
- *	@ap: ATA port
- *
- *	Perform the ATA probe and bus reset sequence plus specific handling
- *	for this hardware. The MPIIX has the enable bits in a different place
- *	to PIIX4 and friends. As a pure PIO device it has no cable detect
- */
-
-static void mpiix_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, mpiix_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	mpiix_set_piomode	-	set initial PIO mode data
  *	@ap: ATA interface
@@ -159,7 +145,7 @@ static struct ata_port_operations mpiix_port_ops = {
 	.qc_issue	= mpiix_qc_issue_prot,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= mpiix_set_piomode,
-	.error_handler	= mpiix_error_handler,
+	.prereset	= mpiix_pre_reset,
 };
 
 static int mpiix_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index 5a043e426480..4d2eefee7387 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -53,20 +53,6 @@ static int ns87410_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	ns87410_error_handler		-	probe reset
- *	@ap: ATA port
- *
- *	Perform the ATA probe and bus reset sequence plus specific handling
- *	for this hardware. The MPIIX has the enable bits in a different place
- *	to PIIX4 and friends. As a pure PIO device it has no cable detect
- */
-
-static void ns87410_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, ns87410_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	ns87410_set_piomode	-	set initial PIO mode data
  *	@ap: ATA interface
@@ -152,7 +138,7 @@ static struct ata_port_operations ns87410_port_ops = {
 	.qc_issue	= ns87410_qc_issue_prot,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= ns87410_set_piomode,
-	.error_handler	= ns87410_error_handler,
+	.prereset	= ns87410_pre_reset,
 };
 
 static int ns87410_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index 7001b756819e..c1da79a76439 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -50,20 +50,6 @@ static int oldpiix_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	oldpiix_pata_error_handler - Probe specified port on PATA host controller
- *	@ap: Port to probe
- *	@classes:
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void oldpiix_pata_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, oldpiix_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	oldpiix_set_piomode - Initialize host controller PATA PIO timings
  *	@ap: Port whose timings we are configuring
@@ -229,7 +215,7 @@ static struct ata_port_operations oldpiix_pata_ops = {
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= oldpiix_set_piomode,
 	.set_dmamode		= oldpiix_set_dmamode,
-	.error_handler		= oldpiix_pata_error_handler,
+	.prereset		= oldpiix_pre_reset,
 };
 
 
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index 5a5f20e03fc0..4ddd03a67775 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -67,21 +67,6 @@ static int opti_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	opti_probe_reset		-	probe reset
- *	@ap: ATA port
- *
- *	Perform the ATA probe and bus reset sequence plus specific handling
- *	for this hardware. The Opti needs little handling - we have no UDMA66
- *	capability that needs cable detection. All we must do is check the port
- *	is enabled.
- */
-
-static void opti_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, opti_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	opti_write_reg		-	control register setup
  *	@ap: ATA port
@@ -172,7 +157,7 @@ static struct ata_port_operations opti_port_ops = {
 	.inherits	= &ata_sff_port_ops,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= opti_set_piomode,
-	.error_handler	= opti_error_handler,
+	.prereset	= opti_pre_reset,
 };
 
 static int opti_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index ba2819ff964b..36ac147de178 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -67,21 +67,6 @@ static int optidma_pre_reset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	optidma_probe_reset		-	probe reset
- *	@ap: ATA port
- *
- *	Perform the ATA probe and bus reset sequence plus specific handling
- *	for this hardware. The Opti needs little handling - we have no UDMA66
- *	capability that needs cable detection. All we must do is check the port
- *	is enabled.
- */
-
-static void optidma_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, optidma_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	optidma_unlock		-	unlock control registers
  *	@ap: ATA port
@@ -359,7 +344,7 @@ static struct ata_port_operations optidma_port_ops = {
 	.set_piomode	= optidma_set_pio_mode,
 	.set_dmamode	= optidma_set_dma_mode,
 	.set_mode	= optidma_set_mode,
-	.error_handler	= optidma_error_handler,
+	.prereset	= optidma_pre_reset,
 };
 
 static struct ata_port_operations optiplus_port_ops = {
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
index f619c20dd192..d235c9f92d09 100644
--- a/drivers/ata/pata_pdc2027x.c
+++ b/drivers/ata/pata_pdc2027x.c
@@ -63,7 +63,7 @@ enum {
 };
 
 static int pdc2027x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-static void pdc2027x_error_handler(struct ata_port *ap);
+static int pdc2027x_prereset(struct ata_link *link, unsigned long deadline);
 static void pdc2027x_set_piomode(struct ata_port *ap, struct ata_device *adev);
 static void pdc2027x_set_dmamode(struct ata_port *ap, struct ata_device *adev);
 static int pdc2027x_check_atapi_dma(struct ata_queued_cmd *qc);
@@ -136,7 +136,7 @@ static struct ata_port_operations pdc2027x_pata100_ops = {
 	.inherits		= &ata_bmdma_port_ops,
 	.check_atapi_dma	= pdc2027x_check_atapi_dma,
 	.cable_detect		= pdc2027x_cable_detect,
-	.error_handler		= pdc2027x_error_handler,
+	.prereset		= pdc2027x_prereset,
 };
 
 static struct ata_port_operations pdc2027x_pata133_ops = {
@@ -251,21 +251,6 @@ static int pdc2027x_prereset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-/**
- *	pdc2027x_error_handler - Perform reset on PATA port and classify
- *	@ap: Port to reset
- *
- *	Reset PATA phy and classify attached devices.
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void pdc2027x_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, pdc2027x_prereset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	pdc2720x_mode_filter	-	mode selection filter
  *	@adev: ATA device
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index 033d1f3a82de..07f2d7a6f1a8 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -904,17 +904,6 @@ static void scc_std_postreset(struct ata_link *link, unsigned int *classes)
 	DPRINTK("EXIT\n");
 }
 
-/**
- *	scc_error_handler - Stock error handler for BMDMA controller
- *	@ap: port to handle error for
- */
-
-static void scc_error_handler (struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, scc_pata_prereset, scc_std_softreset, NULL,
-			   scc_std_postreset);
-}
-
 /**
  *	scc_bmdma_irq_clear - Clear PCI IDE BMDMA interrupt.
  *	@ap: Port associated with this ATA transaction.
@@ -992,7 +981,9 @@ static const struct ata_port_operations scc_pata_ops = {
 	.data_xfer		= scc_data_xfer,
 
 	.freeze			= scc_bmdma_freeze,
-	.error_handler		= scc_error_handler,
+	.prereset		= scc_pata_prereset,
+	.softreset		= scc_std_softreset,
+	.postreset		= scc_std_postreset,
 	.post_internal_cmd	= scc_bmdma_stop,
 
 	.irq_clear		= scc_bmdma_irq_clear,
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index 28abfc26e7a4..793e6714df8c 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -160,19 +160,6 @@ static int sis_pre_reset(struct ata_link *link, unsigned long deadline)
 }
 
 
-/**
- *	sis_error_handler - Probe specified port on PATA host controller
- *	@ap: Port to probe
- *
- *	LOCKING:
- *	None (inherited from caller).
- */
-
-static void sis_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, sis_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	sis_set_fifo	-	Set RWP fifo bits for this device
  *	@ap: Port
@@ -526,7 +513,7 @@ static struct ata_port_operations sis_133_for_sata_ops = {
 
 static struct ata_port_operations sis_base_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.error_handler		= sis_error_handler,
+	.prereset		= sis_pre_reset,
 };
 
 static struct ata_port_operations sis_133_ops = {
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index 1d97f920bd2b..bee11ca8f55a 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -64,12 +64,6 @@ static int sl82c105_pre_reset(struct ata_link *link, unsigned long deadline)
 }
 
 
-static void sl82c105_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, sl82c105_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
-
 /**
  *	sl82c105_configure_piomode	-	set chip PIO timing
  *	@ap: ATA interface
@@ -245,7 +239,7 @@ static struct ata_port_operations sl82c105_port_ops = {
 	.bmdma_stop	= sl82c105_bmdma_stop,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= sl82c105_set_piomode,
-	.error_handler	= sl82c105_error_handler,
+	.prereset	= sl82c105_pre_reset,
 };
 
 /**
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index f07b0e5df222..bd546a389ce1 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -71,11 +71,6 @@ static int triflex_prereset(struct ata_link *link, unsigned long deadline)
 
 
 
-static void triflex_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, triflex_prereset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	triflex_load_timing		-	timing configuration
  *	@ap: ATA interface
@@ -189,7 +184,7 @@ static struct ata_port_operations triflex_port_ops = {
 	.bmdma_stop	= triflex_bmdma_stop,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= triflex_set_piomode,
-	.error_handler	= triflex_error_handler,
+	.prereset	= triflex_prereset,
 };
 
 static int triflex_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index f4092cbd566f..2928fa173132 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -214,18 +214,6 @@ static int via_pre_reset(struct ata_link *link, unsigned long deadline)
 }
 
 
-/**
- *	via_error_handler		-	reset for VIA chips
- *	@ap: ATA port
- *
- *	Handle the reset callback for the later chips with cable detect
- */
-
-static void via_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, via_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
-}
-
 /**
  *	via_do_set_mode	-	set initial PIO mode data
  *	@ap: ATA interface
@@ -343,7 +331,7 @@ static struct ata_port_operations via_port_ops = {
 	.cable_detect	= via_cable_detect,
 	.set_piomode	= via_set_piomode,
 	.set_dmamode	= via_set_dmamode,
-	.error_handler	= via_error_handler,
+	.prereset	= via_pre_reset,
 };
 
 static struct ata_port_operations via_port_ops_noirq = {
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index a5706149af6b..5ed065d0ab4c 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -140,7 +140,7 @@ static void adma_bmdma_stop(struct ata_queued_cmd *qc);
 static u8 adma_bmdma_status(struct ata_port *ap);
 static void adma_freeze(struct ata_port *ap);
 static void adma_thaw(struct ata_port *ap);
-static void adma_error_handler(struct ata_port *ap);
+static int adma_prereset(struct ata_link *link, unsigned long deadline);
 
 static struct scsi_host_template adma_ata_sht = {
 	ATA_BASE_SHT(DRV_NAME),
@@ -166,7 +166,8 @@ static struct ata_port_operations adma_ata_ops = {
 
 	.freeze			= adma_freeze,
 	.thaw			= adma_thaw,
-	.error_handler		= adma_error_handler,
+	.prereset		= adma_prereset,
+	.softreset		= ata_std_softreset,
 
 	.port_start		= adma_port_start,
 	.port_stop		= adma_port_stop,
@@ -292,12 +293,6 @@ static int adma_prereset(struct ata_link *link, unsigned long deadline)
 	return ata_std_prereset(link, deadline);
 }
 
-static void adma_error_handler(struct ata_port *ap)
-{
-	ata_do_eh(ap, adma_prereset, ata_std_softreset, NULL,
-		  ata_std_postreset);
-}
-
 static int adma_fill_sg(struct ata_queued_cmd *qc)
 {
 	struct scatterlist *sg;
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 865030ae8f8a..676302fdaa97 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -912,16 +912,6 @@ err:
 	return -EIO;
 }
 
-static void sata_fsl_error_handler(struct ata_port *ap)
-{
-
-	DPRINTK("in xx_error_handler\n");
-
-	/* perform recovery */
-	ata_do_eh(ap, ata_std_prereset, sata_fsl_softreset, sata_std_hardreset,
-		  ata_std_postreset);
-}
-
 static void sata_fsl_post_internal_cmd(struct ata_queued_cmd *qc)
 {
 	if (qc->flags & ATA_QCFLAG_FAILED)
@@ -1213,7 +1203,7 @@ static const struct ata_port_operations sata_fsl_ops = {
 
 	.freeze = sata_fsl_freeze,
 	.thaw = sata_fsl_thaw,
-	.error_handler = sata_fsl_error_handler,
+	.softreset = sata_fsl_softreset,
 	.post_internal_cmd = sata_fsl_post_internal_cmd,
 
 	.port_start = sata_fsl_port_start,
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 047f80f5825c..ba1c09953517 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -452,8 +452,7 @@ static void inic_error_handler(struct ata_port *ap)
 	spin_unlock_irqrestore(ap->lock, flags);
 
 	/* PIO and DMA engines have been stopped, perform recovery */
-	ata_do_eh(ap, ata_std_prereset, NULL, inic_hardreset,
-		  ata_std_postreset);
+	ata_std_error_handler(ap);
 }
 
 static void inic_post_internal_cmd(struct ata_queued_cmd *qc)
@@ -532,6 +531,8 @@ static struct ata_port_operations inic_port_ops = {
 
 	.freeze			= inic_freeze,
 	.thaw			= inic_thaw,
+	.softreset		= ATA_OP_NULL,	/* softreset is broken */
+	.hardreset		= inic_hardreset,
 	.error_handler		= inic_error_handler,
 	.post_internal_cmd	= inic_post_internal_cmd,
 	.dev_config		= inic_dev_config,
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index f341a82d27bf..9a89390531b1 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -470,7 +470,10 @@ static void mv_port_stop(struct ata_port *ap);
 static void mv_qc_prep(struct ata_queued_cmd *qc);
 static void mv_qc_prep_iie(struct ata_queued_cmd *qc);
 static unsigned int mv_qc_issue(struct ata_queued_cmd *qc);
-static void mv_error_handler(struct ata_port *ap);
+static int mv_prereset(struct ata_link *link, unsigned long deadline);
+static int mv_hardreset(struct ata_link *link, unsigned int *class,
+			unsigned long deadline);
+static void mv_postreset(struct ata_link *link, unsigned int *classes);
 static void mv_eh_freeze(struct ata_port *ap);
 static void mv_eh_thaw(struct ata_port *ap);
 static void mv6_dev_config(struct ata_device *dev);
@@ -534,7 +537,10 @@ static struct ata_port_operations mv5_ops = {
 
 	.freeze			= mv_eh_freeze,
 	.thaw			= mv_eh_thaw,
-	.error_handler		= mv_error_handler,
+	.prereset		= mv_prereset,
+	.hardreset		= mv_hardreset,
+	.postreset		= mv_postreset,
+	.error_handler		= ata_std_error_handler, /* avoid SFF EH */
 	.post_internal_cmd	= ATA_OP_NULL,
 
 	.scr_read		= mv5_scr_read,
@@ -2415,12 +2421,6 @@ static void mv_postreset(struct ata_link *link, unsigned int *classes)
 	iowrite8(ap->ctl, ap->ioaddr.ctl_addr);
 }
 
-static void mv_error_handler(struct ata_port *ap)
-{
-	ata_do_eh(ap, mv_prereset, ata_std_softreset,
-		  mv_hardreset, mv_postreset);
-}
-
 static void mv_eh_freeze(struct ata_port *ap)
 {
 	struct mv_host_priv *hpriv = ap->host->private_data;
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 5637b082bc85..b2eb5724cf85 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -309,7 +309,8 @@ static void nv_nf2_freeze(struct ata_port *ap);
 static void nv_nf2_thaw(struct ata_port *ap);
 static void nv_ck804_freeze(struct ata_port *ap);
 static void nv_ck804_thaw(struct ata_port *ap);
-static void nv_error_handler(struct ata_port *ap);
+static int nv_hardreset(struct ata_link *link, unsigned int *class,
+			unsigned long deadline);
 static int nv_adma_slave_config(struct scsi_device *sdev);
 static int nv_adma_check_atapi_dma(struct ata_queued_cmd *qc);
 static void nv_adma_qc_prep(struct ata_queued_cmd *qc);
@@ -406,7 +407,7 @@ static struct scsi_host_template nv_swncq_sht = {
 
 static struct ata_port_operations nv_generic_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.error_handler		= nv_error_handler,
+	.hardreset		= nv_hardreset,
 	.scr_read		= nv_scr_read,
 	.scr_write		= nv_scr_write,
 };
@@ -1599,12 +1600,6 @@ static int nv_hardreset(struct ata_link *link, unsigned int *class,
 	return sata_std_hardreset(link, &dummy, deadline);
 }
 
-static void nv_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, ata_std_prereset, ata_std_softreset,
-			   nv_hardreset, ata_std_postreset);
-}
-
 static void nv_adma_error_handler(struct ata_port *ap)
 {
 	struct nv_adma_port_priv *pp = ap->private_data;
@@ -1658,8 +1653,7 @@ static void nv_adma_error_handler(struct ata_port *ap)
 		readw(mmio + NV_ADMA_CTL);	/* flush posted write */
 	}
 
-	ata_bmdma_drive_eh(ap, ata_std_prereset, ata_std_softreset,
-			   nv_hardreset, ata_std_postreset);
+	ata_bmdma_error_handler(ap);
 }
 
 static void nv_swncq_qc_to_dq(struct ata_port *ap, struct ata_queued_cmd *qc)
@@ -1785,8 +1779,7 @@ static void nv_swncq_error_handler(struct ata_port *ap)
 		ehc->i.action |= ATA_EH_RESET;
 	}
 
-	ata_bmdma_drive_eh(ap, ata_std_prereset, ata_std_softreset,
-			   nv_hardreset, ata_std_postreset);
+	ata_bmdma_error_handler(ap);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index e09b975c973d..91659dc15caf 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -148,8 +148,7 @@ static void pdc_freeze(struct ata_port *ap);
 static void pdc_sata_freeze(struct ata_port *ap);
 static void pdc_thaw(struct ata_port *ap);
 static void pdc_sata_thaw(struct ata_port *ap);
-static void pdc_pata_error_handler(struct ata_port *ap);
-static void pdc_sata_error_handler(struct ata_port *ap);
+static void pdc_error_handler(struct ata_port *ap);
 static void pdc_post_internal_cmd(struct ata_queued_cmd *qc);
 static int pdc_pata_cable_detect(struct ata_port *ap);
 static int pdc_sata_cable_detect(struct ata_port *ap);
@@ -171,6 +170,7 @@ static const struct ata_port_operations pdc_common_ops = {
 	.irq_clear		= pdc_irq_clear,
 
 	.post_internal_cmd	= pdc_post_internal_cmd,
+	.error_handler		= pdc_error_handler,
 };
 
 static struct ata_port_operations pdc_sata_ops = {
@@ -178,7 +178,6 @@ static struct ata_port_operations pdc_sata_ops = {
 	.cable_detect		= pdc_sata_cable_detect,
 	.freeze			= pdc_sata_freeze,
 	.thaw			= pdc_sata_thaw,
-	.error_handler		= pdc_sata_error_handler,
 	.scr_read		= pdc_sata_scr_read,
 	.scr_write		= pdc_sata_scr_write,
 	.port_start		= pdc_sata_port_start,
@@ -195,7 +194,6 @@ static struct ata_port_operations pdc_pata_ops = {
 	.cable_detect		= pdc_pata_cable_detect,
 	.freeze			= pdc_freeze,
 	.thaw			= pdc_thaw,
-	.error_handler		= pdc_pata_error_handler,
 	.port_start		= pdc_common_port_start,
 };
 
@@ -694,24 +692,12 @@ static void pdc_sata_thaw(struct ata_port *ap)
 	readl(host_mmio + hotplug_offset); /* flush */
 }
 
-static void pdc_common_error_handler(struct ata_port *ap, ata_reset_fn_t hardreset)
+static void pdc_error_handler(struct ata_port *ap)
 {
 	if (!(ap->pflags & ATA_PFLAG_FROZEN))
 		pdc_reset_port(ap);
 
-	/* perform recovery */
-	ata_do_eh(ap, ata_std_prereset, ata_std_softreset, hardreset,
-		  ata_std_postreset);
-}
-
-static void pdc_pata_error_handler(struct ata_port *ap)
-{
-	pdc_common_error_handler(ap, NULL);
-}
-
-static void pdc_sata_error_handler(struct ata_port *ap)
-{
-	pdc_common_error_handler(ap, sata_std_hardreset);
+	ata_std_error_handler(ap);
 }
 
 static void pdc_post_internal_cmd(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 107ef09814de..2ceb0990bcd8 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -123,6 +123,7 @@ static void qs_bmdma_stop(struct ata_queued_cmd *qc);
 static u8 qs_bmdma_status(struct ata_port *ap);
 static void qs_freeze(struct ata_port *ap);
 static void qs_thaw(struct ata_port *ap);
+static int qs_prereset(struct ata_link *link, unsigned long deadline);
 static void qs_error_handler(struct ata_port *ap);
 
 static struct scsi_host_template qs_ata_sht = {
@@ -142,6 +143,8 @@ static struct ata_port_operations qs_ata_ops = {
 
 	.freeze			= qs_freeze,
 	.thaw			= qs_thaw,
+	.prereset		= qs_prereset,
+	.softreset		= ATA_OP_NULL,
 	.error_handler		= qs_error_handler,
 	.post_internal_cmd	= ATA_OP_NULL,
 
@@ -250,8 +253,7 @@ static int qs_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
 static void qs_error_handler(struct ata_port *ap)
 {
 	qs_enter_reg_mode(ap);
-	ata_do_eh(ap, qs_prereset, NULL, sata_std_hardreset,
-		  ata_std_postreset);
+	ata_std_error_handler(ap);
 }
 
 static int qs_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index 363fb90e1047..67df1d753305 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -352,6 +352,14 @@ static void sil24_pmp_attach(struct ata_port *ap);
 static void sil24_pmp_detach(struct ata_port *ap);
 static void sil24_freeze(struct ata_port *ap);
 static void sil24_thaw(struct ata_port *ap);
+static int sil24_softreset(struct ata_link *link, unsigned int *class,
+			   unsigned long deadline);
+static int sil24_hardreset(struct ata_link *link, unsigned int *class,
+			   unsigned long deadline);
+static int sil24_pmp_softreset(struct ata_link *link, unsigned int *class,
+			       unsigned long deadline);
+static int sil24_pmp_hardreset(struct ata_link *link, unsigned int *class,
+			       unsigned long deadline);
 static void sil24_error_handler(struct ata_port *ap);
 static void sil24_post_internal_cmd(struct ata_queued_cmd *qc);
 static int sil24_port_start(struct ata_port *ap);
@@ -402,6 +410,10 @@ static struct ata_port_operations sil24_ops = {
 
 	.freeze			= sil24_freeze,
 	.thaw			= sil24_thaw,
+	.softreset		= sil24_softreset,
+	.hardreset		= sil24_hardreset,
+	.pmp_softreset		= sil24_pmp_softreset,
+	.pmp_hardreset		= sil24_pmp_hardreset,
 	.error_handler		= sil24_error_handler,
 	.post_internal_cmd	= sil24_post_internal_cmd,
 	.dev_config		= sil24_dev_config,
@@ -1181,11 +1193,7 @@ static void sil24_error_handler(struct ata_port *ap)
 	if (sil24_init_port(ap))
 		ata_eh_freeze_port(ap);
 
-	/* perform recovery */
-	sata_pmp_do_eh(ap, ata_std_prereset, sil24_softreset, sil24_hardreset,
-		       ata_std_postreset, sata_pmp_std_prereset,
-		       sil24_pmp_softreset, sil24_pmp_hardreset,
-		       sata_pmp_std_postreset);
+	sata_pmp_error_handler(ap);
 
 	pp->do_port_rst = 0;
 }
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 6326bcf8ea5d..402fd7333d48 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -71,7 +71,7 @@ static int svia_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 static int svia_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
 static int svia_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
 static void svia_noop_freeze(struct ata_port *ap);
-static void vt6420_error_handler(struct ata_port *ap);
+static int vt6420_prereset(struct ata_link *link, unsigned long deadline);
 static int vt6421_pata_cable_detect(struct ata_port *ap);
 static void vt6421_set_pio_mode(struct ata_port *ap, struct ata_device *adev);
 static void vt6421_set_dma_mode(struct ata_port *ap, struct ata_device *adev);
@@ -106,7 +106,7 @@ static struct scsi_host_template svia_sht = {
 static struct ata_port_operations vt6420_sata_ops = {
 	.inherits		= &ata_bmdma_port_ops,
 	.freeze			= svia_noop_freeze,
-	.error_handler		= vt6420_error_handler,
+	.prereset		= vt6420_prereset,
 };
 
 static struct ata_port_operations vt6421_pata_ops = {
@@ -247,12 +247,6 @@ static int vt6420_prereset(struct ata_link *link, unsigned long deadline)
 	return 0;
 }
 
-static void vt6420_error_handler(struct ata_port *ap)
-{
-	ata_bmdma_drive_eh(ap, vt6420_prereset, ata_std_softreset, NULL,
-			   ata_std_postreset);
-}
-
 static int vt6421_pata_cable_detect(struct ata_port *ap)
 {
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index c72014a3e7d4..79fd2436bd70 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -3937,7 +3937,7 @@ static int __ipr_eh_dev_reset(struct scsi_cmnd * scsi_cmd)
 	if (ipr_is_gata(res) && res->sata_port) {
 		ap = res->sata_port->ap;
 		spin_unlock_irq(scsi_cmd->device->host->host_lock);
-		ata_do_eh(ap, NULL, NULL, ipr_sata_reset, NULL);
+		ata_std_error_handler(ap);
 		spin_lock_irq(scsi_cmd->device->host->host_lock);
 
 		list_for_each_entry(ipr_cmd, &ioa_cfg->pending_q, queue) {
@@ -5275,6 +5275,7 @@ static struct ata_port_operations ipr_sata_ops = {
 	.check_altstatus = ipr_ata_check_altstatus,
 	.dev_select = ata_noop_dev_select,
 	.phy_reset = ipr_ata_phy_reset,
+	.hardreset = ipr_sata_reset,
 	.post_internal_cmd = ipr_ata_post_internal,
 	.tf_read = ipr_tf_read,
 	.qc_prep = ata_noop_qc_prep,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 88c6fa84ed74..01c233303aee 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -693,6 +693,14 @@ struct ata_port_operations {
 
 	void (*freeze)(struct ata_port *ap);
 	void (*thaw)(struct ata_port *ap);
+	ata_prereset_fn_t	prereset;
+	ata_reset_fn_t		softreset;
+	ata_reset_fn_t		hardreset;
+	ata_postreset_fn_t	postreset;
+	ata_prereset_fn_t	pmp_prereset;
+	ata_reset_fn_t		pmp_softreset;
+	ata_reset_fn_t		pmp_hardreset;
+	ata_postreset_fn_t	pmp_postreset;
 	void (*error_handler)(struct ata_port *ap);
 	void (*post_internal_cmd)(struct ata_queued_cmd *qc);
 
@@ -909,10 +917,6 @@ extern void ata_bmdma_irq_clear(struct ata_port *ap);
 extern void ata_noop_irq_clear(struct ata_port *ap);
 extern void ata_bmdma_freeze(struct ata_port *ap);
 extern void ata_bmdma_thaw(struct ata_port *ap);
-extern void ata_bmdma_drive_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
-			       ata_reset_fn_t softreset,
-			       ata_reset_fn_t hardreset,
-			       ata_postreset_fn_t postreset);
 extern void ata_bmdma_error_handler(struct ata_port *ap);
 extern void ata_bmdma_post_internal_cmd(struct ata_queued_cmd *qc);
 extern int ata_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
@@ -1056,11 +1060,7 @@ extern int sata_pmp_std_prereset(struct ata_link *link, unsigned long deadline);
 extern int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
 				  unsigned long deadline);
 extern void sata_pmp_std_postreset(struct ata_link *link, unsigned int *class);
-extern void sata_pmp_do_eh(struct ata_port *ap,
-		ata_prereset_fn_t prereset, ata_reset_fn_t softreset,
-		ata_reset_fn_t hardreset, ata_postreset_fn_t postreset,
-		ata_prereset_fn_t pmp_prereset, ata_reset_fn_t pmp_softreset,
-		ata_reset_fn_t pmp_hardreset, ata_postreset_fn_t pmp_postreset);
+extern void sata_pmp_error_handler(struct ata_port *ap);
 
 /*
  * EH
@@ -1080,6 +1080,7 @@ extern void ata_eh_qc_retry(struct ata_queued_cmd *qc);
 extern void ata_do_eh(struct ata_port *ap, ata_prereset_fn_t prereset,
 		      ata_reset_fn_t softreset, ata_reset_fn_t hardreset,
 		      ata_postreset_fn_t postreset);
+extern void ata_std_error_handler(struct ata_port *ap);
 
 /*
  * Base operations to inherit from and initializers for sht
-- 
cgit v1.2.3


From 624d5c514eed18d5a93062e9d86d67065175f30a Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 22:16:41 +0900
Subject: libata: reorganize SFF related stuff

* Move SFF related functions from libata-core.c to libata-sff.c.

  ata_[bmdma_]sff_port_ops, ata_devchk(), ata_dev_try_classify(),
  ata_std_dev_select(), ata_tf_to_host(), ata_busy_sleep(),
  ata_wait_after_reset(), ata_wait_ready(), ata_bus_post_reset(),
  ata_bus_softreset(), ata_bus_reset(), ata_std_softreset(),
  sata_std_hardreset(), ata_fill_sg(), ata_fill_sg_dumb(),
  ata_qc_prep(), ata_dump_qc_prep(), ata_data_xfer(),
  ata_data_xfer_noirq(), ata_pio_sector(), ata_pio_sectors(),
  atapi_send_cdb(), __atapi_pio_bytes(), atapi_pio_bytes(),
  ata_hsm_ok_in_wq(), ata_hsm_qc_complete(), ata_hsm_move(),
  ata_pio_task(), ata_qc_issue_prot(), ata_host_intr(),
  ata_interrupt(), ata_std_ports()

* Make ata_pio_queue_task() global as it's now called from
  libata-sff.c.

* Move SFF related stuff in include/linux/libata.h and
  drivers/ata/libata.h into one place.  While at it, move timing
  constants into the global enum definition and fortify comments a
  bit.

This patch strictly moves stuff around and as such doesn't cause any
functional difference.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 2114 +++-----------------------------------------
 drivers/ata/libata-sff.c  | 2151 +++++++++++++++++++++++++++++++++++++++++----
 drivers/ata/libata.h      |    8 +-
 include/linux/libata.h    |  379 ++++----
 4 files changed, 2321 insertions(+), 2331 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 7860d9f60ae4..34c068f18350 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -46,7 +46,6 @@
 #include <linux/init.h>
 #include <linux/list.h>
 #include <linux/mm.h>
-#include <linux/highmem.h>
 #include <linux/spinlock.h>
 #include <linux/blkdev.h>
 #include <linux/delay.h>
@@ -98,41 +97,6 @@ const struct ata_port_operations sata_pmp_port_ops = {
 	.error_handler		= sata_pmp_error_handler,
 };
 
-const struct ata_port_operations ata_sff_port_ops = {
-	.inherits		= &ata_base_port_ops,
-
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.softreset		= ata_std_softreset,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.dev_select		= ata_std_dev_select,
-	.check_status		= ata_check_status,
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.data_xfer		= ata_data_xfer,
-	.irq_on			= ata_irq_on,
-
-	.port_start		= ata_sff_port_start,
-};
-
-const struct ata_port_operations ata_bmdma_port_ops = {
-	.inherits		= &ata_sff_port_ops,
-
-	.mode_filter		= ata_pci_default_filter,
-
-	.bmdma_setup		= ata_bmdma_setup,
-	.bmdma_start		= ata_bmdma_start,
-	.bmdma_stop		= ata_bmdma_stop,
-	.bmdma_status		= ata_bmdma_status,
-	.irq_clear		= ata_bmdma_irq_clear,
-};
-
 static unsigned int ata_dev_init_params(struct ata_device *dev,
 					u16 heads, u16 sectors);
 static unsigned int ata_dev_set_xfermode(struct ata_device *dev);
@@ -422,6 +386,14 @@ int atapi_cmd_type(u8 opcode)
 	}
 }
 
+/**
+ *	ata_noop_irq_clear - Noop placeholder for irq_clear
+ *	@ap: Port associated with this ATA transaction.
+ */
+void ata_noop_irq_clear(struct ata_port *ap)
+{
+}
+
 /**
  *	ata_tf_to_fis - Convert ATA taskfile to SATA FIS structure
  *	@tf: Taskfile to convert
@@ -1102,50 +1074,6 @@ static void ata_lpm_disable(struct ata_host *host)
 }
 #endif	/* CONFIG_PM */
 
-
-/**
- *	ata_devchk - PATA device presence detection
- *	@ap: ATA channel to examine
- *	@device: Device to examine (starting at zero)
- *
- *	This technique was originally described in
- *	Hale Landis's ATADRVR (www.ata-atapi.com), and
- *	later found its way into the ATA/ATAPI spec.
- *
- *	Write a pattern to the ATA shadow registers,
- *	and if a device is present, it will respond by
- *	correctly storing and echoing back the
- *	ATA shadow register contents.
- *
- *	LOCKING:
- *	caller.
- */
-
-static unsigned int ata_devchk(struct ata_port *ap, unsigned int device)
-{
-	struct ata_ioports *ioaddr = &ap->ioaddr;
-	u8 nsect, lbal;
-
-	ap->ops->dev_select(ap, device);
-
-	iowrite8(0x55, ioaddr->nsect_addr);
-	iowrite8(0xaa, ioaddr->lbal_addr);
-
-	iowrite8(0xaa, ioaddr->nsect_addr);
-	iowrite8(0x55, ioaddr->lbal_addr);
-
-	iowrite8(0x55, ioaddr->nsect_addr);
-	iowrite8(0xaa, ioaddr->lbal_addr);
-
-	nsect = ioread8(ioaddr->nsect_addr);
-	lbal = ioread8(ioaddr->lbal_addr);
-
-	if ((nsect == 0x55) && (lbal == 0xaa))
-		return 1;	/* we found a device */
-
-	return 0;		/* nothing found */
-}
-
 /**
  *	ata_dev_classify - determine device type based on ATA-spec signature
  *	@tf: ATA taskfile register set for device to be identified
@@ -1205,75 +1133,6 @@ unsigned int ata_dev_classify(const struct ata_taskfile *tf)
 	return ATA_DEV_UNKNOWN;
 }
 
-/**
- *	ata_dev_try_classify - Parse returned ATA device signature
- *	@dev: ATA device to classify (starting at zero)
- *	@present: device seems present
- *	@r_err: Value of error register on completion
- *
- *	After an event -- SRST, E.D.D., or SATA COMRESET -- occurs,
- *	an ATA/ATAPI-defined set of values is placed in the ATA
- *	shadow registers, indicating the results of device detection
- *	and diagnostics.
- *
- *	Select the ATA device, and read the values from the ATA shadow
- *	registers.  Then parse according to the Error register value,
- *	and the spec-defined values examined by ata_dev_classify().
- *
- *	LOCKING:
- *	caller.
- *
- *	RETURNS:
- *	Device type - %ATA_DEV_ATA, %ATA_DEV_ATAPI or %ATA_DEV_NONE.
- */
-unsigned int ata_dev_try_classify(struct ata_device *dev, int present,
-				  u8 *r_err)
-{
-	struct ata_port *ap = dev->link->ap;
-	struct ata_taskfile tf;
-	unsigned int class;
-	u8 err;
-
-	ap->ops->dev_select(ap, dev->devno);
-
-	memset(&tf, 0, sizeof(tf));
-
-	ap->ops->tf_read(ap, &tf);
-	err = tf.feature;
-	if (r_err)
-		*r_err = err;
-
-	/* see if device passed diags: continue and warn later */
-	if (err == 0)
-		/* diagnostic fail : do nothing _YET_ */
-		dev->horkage |= ATA_HORKAGE_DIAGNOSTIC;
-	else if (err == 1)
-		/* do nothing */ ;
-	else if ((dev->devno == 0) && (err == 0x81))
-		/* do nothing */ ;
-	else
-		return ATA_DEV_NONE;
-
-	/* determine if device is ATA or ATAPI */
-	class = ata_dev_classify(&tf);
-
-	if (class == ATA_DEV_UNKNOWN) {
-		/* If the device failed diagnostic, it's likely to
-		 * have reported incorrect device signature too.
-		 * Assume ATA device if the device seems present but
-		 * device signature is invalid with diagnostic
-		 * failure.
-		 */
-		if (present && (dev->horkage & ATA_HORKAGE_DIAGNOSTIC))
-			class = ATA_DEV_ATA;
-		else
-			class = ATA_DEV_NONE;
-	} else if ((class == ATA_DEV_ATA) && (ata_chk_status(ap) == 0))
-		class = ATA_DEV_NONE;
-
-	return class;
-}
-
 /**
  *	ata_id_string - Convert IDENTIFY DEVICE page into string
  *	@id: IDENTIFY DEVICE results we will examine
@@ -1597,73 +1456,6 @@ void ata_noop_dev_select(struct ata_port *ap, unsigned int device)
 {
 }
 
-
-/**
- *	ata_std_dev_select - Select device 0/1 on ATA bus
- *	@ap: ATA channel to manipulate
- *	@device: ATA device (numbered from zero) to select
- *
- *	Use the method defined in the ATA specification to
- *	make either device 0, or device 1, active on the
- *	ATA channel.  Works with both PIO and MMIO.
- *
- *	May be used as the dev_select() entry in ata_port_operations.
- *
- *	LOCKING:
- *	caller.
- */
-
-void ata_std_dev_select(struct ata_port *ap, unsigned int device)
-{
-	u8 tmp;
-
-	if (device == 0)
-		tmp = ATA_DEVICE_OBS;
-	else
-		tmp = ATA_DEVICE_OBS | ATA_DEV1;
-
-	iowrite8(tmp, ap->ioaddr.device_addr);
-	ata_pause(ap);		/* needed; also flushes, for mmio */
-}
-
-/**
- *	ata_dev_select - Select device 0/1 on ATA bus
- *	@ap: ATA channel to manipulate
- *	@device: ATA device (numbered from zero) to select
- *	@wait: non-zero to wait for Status register BSY bit to clear
- *	@can_sleep: non-zero if context allows sleeping
- *
- *	Use the method defined in the ATA specification to
- *	make either device 0, or device 1, active on the
- *	ATA channel.
- *
- *	This is a high-level version of ata_std_dev_select(),
- *	which additionally provides the services of inserting
- *	the proper pauses and status polling, where needed.
- *
- *	LOCKING:
- *	caller.
- */
-
-void ata_dev_select(struct ata_port *ap, unsigned int device,
-			   unsigned int wait, unsigned int can_sleep)
-{
-	if (ata_msg_probe(ap))
-		ata_port_printk(ap, KERN_INFO, "ata_dev_select: ENTER, "
-				"device %u, wait %u\n", device, wait);
-
-	if (wait)
-		ata_wait_idle(ap);
-
-	ap->ops->dev_select(ap, device);
-
-	if (wait) {
-		if (can_sleep && ap->link.device[device].class == ATA_DEV_ATAPI)
-			msleep(150);
-		ata_wait_idle(ap);
-	}
-}
-
 /**
  *	ata_dump_id - IDENTIFY DEVICE info debugging output
  *	@id: IDENTIFY DEVICE page to dump
@@ -1791,8 +1583,7 @@ unsigned long ata_id_xfermask(const u16 *id)
  *	LOCKING:
  *	Inherited from caller.
  */
-static void ata_pio_queue_task(struct ata_port *ap, void *data,
-			       unsigned long delay)
+void ata_pio_queue_task(struct ata_port *ap, void *data, unsigned long delay)
 {
 	ap->port_task_data = data;
 
@@ -3531,353 +3322,6 @@ int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 	return rc;
 }
 
-/**
- *	ata_tf_to_host - issue ATA taskfile to host controller
- *	@ap: port to which command is being issued
- *	@tf: ATA taskfile register set
- *
- *	Issues ATA taskfile register set to ATA host controller,
- *	with proper synchronization with interrupt handler and
- *	other threads.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- */
-
-static inline void ata_tf_to_host(struct ata_port *ap,
-				  const struct ata_taskfile *tf)
-{
-	ap->ops->tf_load(ap, tf);
-	ap->ops->exec_command(ap, tf);
-}
-
-/**
- *	ata_busy_sleep - sleep until BSY clears, or timeout
- *	@ap: port containing status register to be polled
- *	@tmout_pat: impatience timeout
- *	@tmout: overall timeout
- *
- *	Sleep until ATA Status register bit BSY clears,
- *	or a timeout occurs.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep).
- *
- *	RETURNS:
- *	0 on success, -errno otherwise.
- */
-int ata_busy_sleep(struct ata_port *ap,
-		   unsigned long tmout_pat, unsigned long tmout)
-{
-	unsigned long timer_start, timeout;
-	u8 status;
-
-	status = ata_busy_wait(ap, ATA_BUSY, 300);
-	timer_start = jiffies;
-	timeout = timer_start + tmout_pat;
-	while (status != 0xff && (status & ATA_BUSY) &&
-	       time_before(jiffies, timeout)) {
-		msleep(50);
-		status = ata_busy_wait(ap, ATA_BUSY, 3);
-	}
-
-	if (status != 0xff && (status & ATA_BUSY))
-		ata_port_printk(ap, KERN_WARNING,
-				"port is slow to respond, please be patient "
-				"(Status 0x%x)\n", status);
-
-	timeout = timer_start + tmout;
-	while (status != 0xff && (status & ATA_BUSY) &&
-	       time_before(jiffies, timeout)) {
-		msleep(50);
-		status = ata_chk_status(ap);
-	}
-
-	if (status == 0xff)
-		return -ENODEV;
-
-	if (status & ATA_BUSY) {
-		ata_port_printk(ap, KERN_ERR, "port failed to respond "
-				"(%lu secs, Status 0x%x)\n",
-				tmout / HZ, status);
-		return -EBUSY;
-	}
-
-	return 0;
-}
-
-/**
- *	ata_wait_after_reset - wait before checking status after reset
- *	@ap: port containing status register to be polled
- *	@deadline: deadline jiffies for the operation
- *
- *	After reset, we need to pause a while before reading status.
- *	Also, certain combination of controller and device report 0xff
- *	for some duration (e.g. until SATA PHY is up and running)
- *	which is interpreted as empty port in ATA world.  This
- *	function also waits for such devices to get out of 0xff
- *	status.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep).
- */
-void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline)
-{
-	unsigned long until = jiffies + ATA_TMOUT_FF_WAIT;
-
-	if (time_before(until, deadline))
-		deadline = until;
-
-	/* Spec mandates ">= 2ms" before checking status.  We wait
-	 * 150ms, because that was the magic delay used for ATAPI
-	 * devices in Hale Landis's ATADRVR, for the period of time
-	 * between when the ATA command register is written, and then
-	 * status is checked.  Because waiting for "a while" before
-	 * checking status is fine, post SRST, we perform this magic
-	 * delay here as well.
-	 *
-	 * Old drivers/ide uses the 2mS rule and then waits for ready.
-	 */
-	msleep(150);
-
-	/* Wait for 0xff to clear.  Some SATA devices take a long time
-	 * to clear 0xff after reset.  For example, HHD424020F7SV00
-	 * iVDR needs >= 800ms while.  Quantum GoVault needs even more
-	 * than that.
-	 *
-	 * Note that some PATA controllers (pata_ali) explode if
-	 * status register is read more than once when there's no
-	 * device attached.
-	 */
-	if (ap->flags & ATA_FLAG_SATA) {
-		while (1) {
-			u8 status = ata_chk_status(ap);
-
-			if (status != 0xff || time_after(jiffies, deadline))
-				return;
-
-			msleep(50);
-		}
-	}
-}
-
-/**
- *	ata_wait_ready - sleep until BSY clears, or timeout
- *	@ap: port containing status register to be polled
- *	@deadline: deadline jiffies for the operation
- *
- *	Sleep until ATA Status register bit BSY clears, or timeout
- *	occurs.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep).
- *
- *	RETURNS:
- *	0 on success, -errno otherwise.
- */
-int ata_wait_ready(struct ata_port *ap, unsigned long deadline)
-{
-	unsigned long start = jiffies;
-	int warned = 0;
-
-	while (1) {
-		u8 status = ata_chk_status(ap);
-		unsigned long now = jiffies;
-
-		if (!(status & ATA_BUSY))
-			return 0;
-		if (!ata_link_online(&ap->link) && status == 0xff)
-			return -ENODEV;
-		if (time_after(now, deadline))
-			return -EBUSY;
-
-		if (!warned && time_after(now, start + 5 * HZ) &&
-		    (deadline - now > 3 * HZ)) {
-			ata_port_printk(ap, KERN_WARNING,
-				"port is slow to respond, please be patient "
-				"(Status 0x%x)\n", status);
-			warned = 1;
-		}
-
-		msleep(50);
-	}
-}
-
-static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
-			      unsigned long deadline)
-{
-	struct ata_ioports *ioaddr = &ap->ioaddr;
-	unsigned int dev0 = devmask & (1 << 0);
-	unsigned int dev1 = devmask & (1 << 1);
-	int rc, ret = 0;
-
-	/* if device 0 was found in ata_devchk, wait for its
-	 * BSY bit to clear
-	 */
-	if (dev0) {
-		rc = ata_wait_ready(ap, deadline);
-		if (rc) {
-			if (rc != -ENODEV)
-				return rc;
-			ret = rc;
-		}
-	}
-
-	/* if device 1 was found in ata_devchk, wait for register
-	 * access briefly, then wait for BSY to clear.
-	 */
-	if (dev1) {
-		int i;
-
-		ap->ops->dev_select(ap, 1);
-
-		/* Wait for register access.  Some ATAPI devices fail
-		 * to set nsect/lbal after reset, so don't waste too
-		 * much time on it.  We're gonna wait for !BSY anyway.
-		 */
-		for (i = 0; i < 2; i++) {
-			u8 nsect, lbal;
-
-			nsect = ioread8(ioaddr->nsect_addr);
-			lbal = ioread8(ioaddr->lbal_addr);
-			if ((nsect == 1) && (lbal == 1))
-				break;
-			msleep(50);	/* give drive a breather */
-		}
-
-		rc = ata_wait_ready(ap, deadline);
-		if (rc) {
-			if (rc != -ENODEV)
-				return rc;
-			ret = rc;
-		}
-	}
-
-	/* is all this really necessary? */
-	ap->ops->dev_select(ap, 0);
-	if (dev1)
-		ap->ops->dev_select(ap, 1);
-	if (dev0)
-		ap->ops->dev_select(ap, 0);
-
-	return ret;
-}
-
-static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
-			     unsigned long deadline)
-{
-	struct ata_ioports *ioaddr = &ap->ioaddr;
-
-	DPRINTK("ata%u: bus reset via SRST\n", ap->print_id);
-
-	/* software reset.  causes dev0 to be selected */
-	iowrite8(ap->ctl, ioaddr->ctl_addr);
-	udelay(20);	/* FIXME: flush */
-	iowrite8(ap->ctl | ATA_SRST, ioaddr->ctl_addr);
-	udelay(20);	/* FIXME: flush */
-	iowrite8(ap->ctl, ioaddr->ctl_addr);
-
-	/* wait a while before checking status */
-	ata_wait_after_reset(ap, deadline);
-
-	/* Before we perform post reset processing we want to see if
-	 * the bus shows 0xFF because the odd clown forgets the D7
-	 * pulldown resistor.
-	 */
-	if (ata_chk_status(ap) == 0xFF)
-		return -ENODEV;
-
-	return ata_bus_post_reset(ap, devmask, deadline);
-}
-
-/**
- *	ata_bus_reset - reset host port and associated ATA channel
- *	@ap: port to reset
- *
- *	This is typically the first time we actually start issuing
- *	commands to the ATA channel.  We wait for BSY to clear, then
- *	issue EXECUTE DEVICE DIAGNOSTIC command, polling for its
- *	result.  Determine what devices, if any, are on the channel
- *	by looking at the device 0/1 error register.  Look at the signature
- *	stored in each device's taskfile registers, to determine if
- *	the device is ATA or ATAPI.
- *
- *	LOCKING:
- *	PCI/etc. bus probe sem.
- *	Obtains host lock.
- *
- *	SIDE EFFECTS:
- *	Sets ATA_FLAG_DISABLED if bus reset fails.
- */
-
-void ata_bus_reset(struct ata_port *ap)
-{
-	struct ata_device *device = ap->link.device;
-	struct ata_ioports *ioaddr = &ap->ioaddr;
-	unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS;
-	u8 err;
-	unsigned int dev0, dev1 = 0, devmask = 0;
-	int rc;
-
-	DPRINTK("ENTER, host %u, port %u\n", ap->print_id, ap->port_no);
-
-	/* determine if device 0/1 are present */
-	if (ap->flags & ATA_FLAG_SATA_RESET)
-		dev0 = 1;
-	else {
-		dev0 = ata_devchk(ap, 0);
-		if (slave_possible)
-			dev1 = ata_devchk(ap, 1);
-	}
-
-	if (dev0)
-		devmask |= (1 << 0);
-	if (dev1)
-		devmask |= (1 << 1);
-
-	/* select device 0 again */
-	ap->ops->dev_select(ap, 0);
-
-	/* issue bus reset */
-	if (ap->flags & ATA_FLAG_SRST) {
-		rc = ata_bus_softreset(ap, devmask, jiffies + 40 * HZ);
-		if (rc && rc != -ENODEV)
-			goto err_out;
-	}
-
-	/*
-	 * determine by signature whether we have ATA or ATAPI devices
-	 */
-	device[0].class = ata_dev_try_classify(&device[0], dev0, &err);
-	if ((slave_possible) && (err != 0x81))
-		device[1].class = ata_dev_try_classify(&device[1], dev1, &err);
-
-	/* is double-select really necessary? */
-	if (device[1].class != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 1);
-	if (device[0].class != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 0);
-
-	/* if no devices were detected, disable this port */
-	if ((device[0].class == ATA_DEV_NONE) &&
-	    (device[1].class == ATA_DEV_NONE))
-		goto err_out;
-
-	if (ap->flags & (ATA_FLAG_SATA_RESET | ATA_FLAG_SRST)) {
-		/* set up device control for ATA_FLAG_SATA_RESET */
-		iowrite8(ap->ctl, ioaddr->ctl_addr);
-	}
-
-	DPRINTK("EXIT\n");
-	return;
-
-err_out:
-	ata_port_printk(ap, KERN_ERR, "disabling port\n");
-	ata_port_disable(ap);
-
-	DPRINTK("EXIT\n");
-}
-
 /**
  *	sata_link_debounce - debounce SATA phy status
  *	@link: ATA link to debounce SATA phy status for
@@ -4034,12 +3478,12 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
 }
 
 /**
- *	ata_std_softreset - reset host port via ATA SRST
- *	@link: ATA link to reset
- *	@classes: resulting classes of attached devices
+ *	sata_link_hardreset - reset link via SATA phy reset
+ *	@link: link to reset
+ *	@timing: timing parameters { interval, duratinon, timeout } in msec
  *	@deadline: deadline jiffies for the operation
  *
- *	Reset host port using ATA SRST.
+ *	SATA phy-reset @link using DET bits of SControl register.
  *
  *	LOCKING:
  *	Kernel thread context (may sleep)
@@ -4047,70 +3491,10 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_std_softreset(struct ata_link *link, unsigned int *classes,
-		      unsigned long deadline)
+int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
+			unsigned long deadline)
 {
-	struct ata_port *ap = link->ap;
-	unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS;
-	unsigned int devmask = 0;
-	int rc;
-	u8 err;
-
-	DPRINTK("ENTER\n");
-
-	if (ata_link_offline(link)) {
-		classes[0] = ATA_DEV_NONE;
-		goto out;
-	}
-
-	/* determine if device 0/1 are present */
-	if (ata_devchk(ap, 0))
-		devmask |= (1 << 0);
-	if (slave_possible && ata_devchk(ap, 1))
-		devmask |= (1 << 1);
-
-	/* select device 0 again */
-	ap->ops->dev_select(ap, 0);
-
-	/* issue bus reset */
-	DPRINTK("about to softreset, devmask=%x\n", devmask);
-	rc = ata_bus_softreset(ap, devmask, deadline);
-	/* if link is occupied, -ENODEV too is an error */
-	if (rc && (rc != -ENODEV || sata_scr_valid(link))) {
-		ata_link_printk(link, KERN_ERR, "SRST failed (errno=%d)\n", rc);
-		return rc;
-	}
-
-	/* determine by signature whether we have ATA or ATAPI devices */
-	classes[0] = ata_dev_try_classify(&link->device[0],
-					  devmask & (1 << 0), &err);
-	if (slave_possible && err != 0x81)
-		classes[1] = ata_dev_try_classify(&link->device[1],
-						  devmask & (1 << 1), &err);
-
- out:
-	DPRINTK("EXIT, classes[0]=%u [1]=%u\n", classes[0], classes[1]);
-	return 0;
-}
-
-/**
- *	sata_link_hardreset - reset link via SATA phy reset
- *	@link: link to reset
- *	@timing: timing parameters { interval, duratinon, timeout } in msec
- *	@deadline: deadline jiffies for the operation
- *
- *	SATA phy-reset @link using DET bits of SControl register.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep)
- *
- *	RETURNS:
- *	0 on success, -errno otherwise.
- */
-int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
-			unsigned long deadline)
-{
-	u32 scontrol;
+	u32 scontrol;
 	int rc;
 
 	DPRINTK("ENTER\n");
@@ -4153,74 +3537,6 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 	return rc;
 }
 
-/**
- *	sata_std_hardreset - reset host port via SATA phy reset
- *	@link: link to reset
- *	@class: resulting class of attached device
- *	@deadline: deadline jiffies for the operation
- *
- *	SATA phy-reset host port using DET bits of SControl register,
- *	wait for !BSY and classify the attached device.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep)
- *
- *	RETURNS:
- *	0 on success, -errno otherwise.
- */
-int sata_std_hardreset(struct ata_link *link, unsigned int *class,
-		       unsigned long deadline)
-{
-	struct ata_port *ap = link->ap;
-	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
-	int rc;
-
-	DPRINTK("ENTER\n");
-
-	/* do hardreset */
-	rc = sata_link_hardreset(link, timing, deadline);
-	if (rc) {
-		ata_link_printk(link, KERN_ERR,
-				"COMRESET failed (errno=%d)\n", rc);
-		return rc;
-	}
-
-	/* TODO: phy layer with polling, timeouts, etc. */
-	if (ata_link_offline(link)) {
-		*class = ATA_DEV_NONE;
-		DPRINTK("EXIT, link offline\n");
-		return 0;
-	}
-
-	/* wait a while before checking status */
-	ata_wait_after_reset(ap, deadline);
-
-	/* If PMP is supported, we have to do follow-up SRST.  Note
-	 * that some PMPs don't send D2H Reg FIS after hardreset at
-	 * all if the first port is empty.  Wait for it just for a
-	 * second and request follow-up SRST.
-	 */
-	if (ap->flags & ATA_FLAG_PMP) {
-		ata_wait_ready(ap, jiffies + HZ);
-		return -EAGAIN;
-	}
-
-	rc = ata_wait_ready(ap, deadline);
-	/* link occupied, -ENODEV too is an error */
-	if (rc) {
-		ata_link_printk(link, KERN_ERR,
-				"COMRESET failed (errno=%d)\n", rc);
-		return rc;
-	}
-
-	ap->ops->dev_select(ap, 0);	/* probably unnecessary */
-
-	*class = ata_dev_try_classify(link->device, 1, NULL);
-
-	DPRINTK("EXIT, class=%u\n", *class);
-	return 0;
-}
-
 /**
  *	ata_std_postreset - standard postreset callback
  *	@link: the target ata_link
@@ -4803,112 +4119,6 @@ void ata_sg_clean(struct ata_queued_cmd *qc)
 	qc->sg = NULL;
 }
 
-/**
- *	ata_fill_sg - Fill PCI IDE PRD table
- *	@qc: Metadata associated with taskfile to be transferred
- *
- *	Fill PCI IDE PRD (scatter-gather) table with segments
- *	associated with the current disk command.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- */
-static void ata_fill_sg(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-	struct scatterlist *sg;
-	unsigned int si, pi;
-
-	pi = 0;
-	for_each_sg(qc->sg, sg, qc->n_elem, si) {
-		u32 addr, offset;
-		u32 sg_len, len;
-
-		/* determine if physical DMA addr spans 64K boundary.
-		 * Note h/w doesn't support 64-bit, so we unconditionally
-		 * truncate dma_addr_t to u32.
-		 */
-		addr = (u32) sg_dma_address(sg);
-		sg_len = sg_dma_len(sg);
-
-		while (sg_len) {
-			offset = addr & 0xffff;
-			len = sg_len;
-			if ((offset + sg_len) > 0x10000)
-				len = 0x10000 - offset;
-
-			ap->prd[pi].addr = cpu_to_le32(addr);
-			ap->prd[pi].flags_len = cpu_to_le32(len & 0xffff);
-			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", pi, addr, len);
-
-			pi++;
-			sg_len -= len;
-			addr += len;
-		}
-	}
-
-	ap->prd[pi - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
-}
-
-/**
- *	ata_fill_sg_dumb - Fill PCI IDE PRD table
- *	@qc: Metadata associated with taskfile to be transferred
- *
- *	Fill PCI IDE PRD (scatter-gather) table with segments
- *	associated with the current disk command. Perform the fill
- *	so that we avoid writing any length 64K records for
- *	controllers that don't follow the spec.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- */
-static void ata_fill_sg_dumb(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-	struct scatterlist *sg;
-	unsigned int si, pi;
-
-	pi = 0;
-	for_each_sg(qc->sg, sg, qc->n_elem, si) {
-		u32 addr, offset;
-		u32 sg_len, len, blen;
-
-		/* determine if physical DMA addr spans 64K boundary.
-		 * Note h/w doesn't support 64-bit, so we unconditionally
-		 * truncate dma_addr_t to u32.
-		 */
-		addr = (u32) sg_dma_address(sg);
-		sg_len = sg_dma_len(sg);
-
-		while (sg_len) {
-			offset = addr & 0xffff;
-			len = sg_len;
-			if ((offset + sg_len) > 0x10000)
-				len = 0x10000 - offset;
-
-			blen = len & 0xffff;
-			ap->prd[pi].addr = cpu_to_le32(addr);
-			if (blen == 0) {
-			   /* Some PATA chipsets like the CS5530 can't
-			      cope with 0x0000 meaning 64K as the spec says */
-				ap->prd[pi].flags_len = cpu_to_le32(0x8000);
-				blen = 0x8000;
-				ap->prd[++pi].addr = cpu_to_le32(addr + 0x8000);
-			}
-			ap->prd[pi].flags_len = cpu_to_le32(blen);
-			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", pi, addr, len);
-
-			pi++;
-			sg_len -= len;
-			addr += len;
-		}
-	}
-
-	ap->prd[pi - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
-}
-
 /**
  *	ata_check_atapi_dma - Check whether ATAPI DMA can be supported
  *	@qc: Metadata associated with taskfile to check
@@ -4918,858 +4128,132 @@ static void ata_fill_sg_dumb(struct ata_queued_cmd *qc)
  *	supplied PACKET command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- *	RETURNS: 0 when ATAPI DMA can be used
- *               nonzero otherwise
- */
-int ata_check_atapi_dma(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-
-	/* Don't allow DMA if it isn't multiple of 16 bytes.  Quite a
-	 * few ATAPI devices choke on such DMA requests.
-	 */
-	if (unlikely(qc->nbytes & 15))
-		return 1;
-
-	if (ap->ops->check_atapi_dma)
-		return ap->ops->check_atapi_dma(qc);
-
-	return 0;
-}
-
-/**
- *	ata_std_qc_defer - Check whether a qc needs to be deferred
- *	@qc: ATA command in question
- *
- *	Non-NCQ commands cannot run with any other command, NCQ or
- *	not.  As upper layer only knows the queue depth, we are
- *	responsible for maintaining exclusion.  This function checks
- *	whether a new command @qc can be issued.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- *	RETURNS:
- *	ATA_DEFER_* if deferring is needed, 0 otherwise.
- */
-int ata_std_qc_defer(struct ata_queued_cmd *qc)
-{
-	struct ata_link *link = qc->dev->link;
-
-	if (qc->tf.protocol == ATA_PROT_NCQ) {
-		if (!ata_tag_valid(link->active_tag))
-			return 0;
-	} else {
-		if (!ata_tag_valid(link->active_tag) && !link->sactive)
-			return 0;
-	}
-
-	return ATA_DEFER_LINK;
-}
-
-/**
- *	ata_qc_prep - Prepare taskfile for submission
- *	@qc: Metadata associated with taskfile to be prepared
- *
- *	Prepare ATA taskfile for submission.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- */
-void ata_qc_prep(struct ata_queued_cmd *qc)
-{
-	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
-		return;
-
-	ata_fill_sg(qc);
-}
-
-/**
- *	ata_dumb_qc_prep - Prepare taskfile for submission
- *	@qc: Metadata associated with taskfile to be prepared
- *
- *	Prepare ATA taskfile for submission.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- */
-void ata_dumb_qc_prep(struct ata_queued_cmd *qc)
-{
-	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
-		return;
-
-	ata_fill_sg_dumb(qc);
-}
-
-void ata_noop_qc_prep(struct ata_queued_cmd *qc) { }
-
-/**
- *	ata_sg_init - Associate command with scatter-gather table.
- *	@qc: Command to be associated
- *	@sg: Scatter-gather table.
- *	@n_elem: Number of elements in s/g table.
- *
- *	Initialize the data-related elements of queued_cmd @qc
- *	to point to a scatter-gather table @sg, containing @n_elem
- *	elements.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- */
-void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
-		 unsigned int n_elem)
-{
-	qc->sg = sg;
-	qc->n_elem = n_elem;
-	qc->cursg = qc->sg;
-}
-
-/**
- *	ata_sg_setup - DMA-map the scatter-gather table associated with a command.
- *	@qc: Command with scatter-gather table to be mapped.
- *
- *	DMA-map the scatter-gather table associated with queued_cmd @qc.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- *	RETURNS:
- *	Zero on success, negative on error.
- *
- */
-static int ata_sg_setup(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-	unsigned int n_elem;
-
-	VPRINTK("ENTER, ata%u\n", ap->print_id);
-
-	n_elem = dma_map_sg(ap->dev, qc->sg, qc->n_elem, qc->dma_dir);
-	if (n_elem < 1)
-		return -1;
-
-	DPRINTK("%d sg elements mapped\n", n_elem);
-
-	qc->n_elem = n_elem;
-	qc->flags |= ATA_QCFLAG_DMAMAP;
-
-	return 0;
-}
-
-/**
- *	swap_buf_le16 - swap halves of 16-bit words in place
- *	@buf:  Buffer to swap
- *	@buf_words:  Number of 16-bit words in buffer.
- *
- *	Swap halves of 16-bit words if needed to convert from
- *	little-endian byte order to native cpu byte order, or
- *	vice-versa.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-void swap_buf_le16(u16 *buf, unsigned int buf_words)
-{
-#ifdef __BIG_ENDIAN
-	unsigned int i;
-
-	for (i = 0; i < buf_words; i++)
-		buf[i] = le16_to_cpu(buf[i]);
-#endif /* __BIG_ENDIAN */
-}
-
-/**
- *	ata_data_xfer - Transfer data by PIO
- *	@dev: device to target
- *	@buf: data buffer
- *	@buflen: buffer length
- *	@rw: read/write
- *
- *	Transfer data from/to the device data register by PIO.
- *
- *	LOCKING:
- *	Inherited from caller.
- *
- *	RETURNS:
- *	Bytes consumed.
- */
-unsigned int ata_data_xfer(struct ata_device *dev, unsigned char *buf,
-			   unsigned int buflen, int rw)
-{
-	struct ata_port *ap = dev->link->ap;
-	void __iomem *data_addr = ap->ioaddr.data_addr;
-	unsigned int words = buflen >> 1;
-
-	/* Transfer multiple of 2 bytes */
-	if (rw == READ)
-		ioread16_rep(data_addr, buf, words);
-	else
-		iowrite16_rep(data_addr, buf, words);
-
-	/* Transfer trailing 1 byte, if any. */
-	if (unlikely(buflen & 0x01)) {
-		__le16 align_buf[1] = { 0 };
-		unsigned char *trailing_buf = buf + buflen - 1;
-
-		if (rw == READ) {
-			align_buf[0] = cpu_to_le16(ioread16(data_addr));
-			memcpy(trailing_buf, align_buf, 1);
-		} else {
-			memcpy(align_buf, trailing_buf, 1);
-			iowrite16(le16_to_cpu(align_buf[0]), data_addr);
-		}
-		words++;
-	}
-
-	return words << 1;
-}
-
-/**
- *	ata_data_xfer_noirq - Transfer data by PIO
- *	@dev: device to target
- *	@buf: data buffer
- *	@buflen: buffer length
- *	@rw: read/write
- *
- *	Transfer data from/to the device data register by PIO. Do the
- *	transfer with interrupts disabled.
- *
- *	LOCKING:
- *	Inherited from caller.
- *
- *	RETURNS:
- *	Bytes consumed.
- */
-unsigned int ata_data_xfer_noirq(struct ata_device *dev, unsigned char *buf,
-				 unsigned int buflen, int rw)
-{
-	unsigned long flags;
-	unsigned int consumed;
-
-	local_irq_save(flags);
-	consumed = ata_data_xfer(dev, buf, buflen, rw);
-	local_irq_restore(flags);
-
-	return consumed;
-}
-
-
-/**
- *	ata_pio_sector - Transfer a sector of data.
- *	@qc: Command on going
- *
- *	Transfer qc->sect_size bytes of data from/to the ATA device.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-
-static void ata_pio_sector(struct ata_queued_cmd *qc)
-{
-	int do_write = (qc->tf.flags & ATA_TFLAG_WRITE);
-	struct ata_port *ap = qc->ap;
-	struct page *page;
-	unsigned int offset;
-	unsigned char *buf;
-
-	if (qc->curbytes == qc->nbytes - qc->sect_size)
-		ap->hsm_task_state = HSM_ST_LAST;
-
-	page = sg_page(qc->cursg);
-	offset = qc->cursg->offset + qc->cursg_ofs;
-
-	/* get the current page and offset */
-	page = nth_page(page, (offset >> PAGE_SHIFT));
-	offset %= PAGE_SIZE;
-
-	DPRINTK("data %s\n", qc->tf.flags & ATA_TFLAG_WRITE ? "write" : "read");
-
-	if (PageHighMem(page)) {
-		unsigned long flags;
-
-		/* FIXME: use a bounce buffer */
-		local_irq_save(flags);
-		buf = kmap_atomic(page, KM_IRQ0);
-
-		/* do the actual data transfer */
-		ap->ops->data_xfer(qc->dev, buf + offset, qc->sect_size, do_write);
-
-		kunmap_atomic(buf, KM_IRQ0);
-		local_irq_restore(flags);
-	} else {
-		buf = page_address(page);
-		ap->ops->data_xfer(qc->dev, buf + offset, qc->sect_size, do_write);
-	}
-
-	qc->curbytes += qc->sect_size;
-	qc->cursg_ofs += qc->sect_size;
-
-	if (qc->cursg_ofs == qc->cursg->length) {
-		qc->cursg = sg_next(qc->cursg);
-		qc->cursg_ofs = 0;
-	}
-}
-
-/**
- *	ata_pio_sectors - Transfer one or many sectors.
- *	@qc: Command on going
- *
- *	Transfer one or many sectors of data from/to the
- *	ATA device for the DRQ request.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-
-static void ata_pio_sectors(struct ata_queued_cmd *qc)
-{
-	if (is_multi_taskfile(&qc->tf)) {
-		/* READ/WRITE MULTIPLE */
-		unsigned int nsect;
-
-		WARN_ON(qc->dev->multi_count == 0);
-
-		nsect = min((qc->nbytes - qc->curbytes) / qc->sect_size,
-			    qc->dev->multi_count);
-		while (nsect--)
-			ata_pio_sector(qc);
-	} else
-		ata_pio_sector(qc);
-
-	ata_altstatus(qc->ap); /* flush */
-}
-
-/**
- *	atapi_send_cdb - Write CDB bytes to hardware
- *	@ap: Port to which ATAPI device is attached.
- *	@qc: Taskfile currently active
- *
- *	When device has indicated its readiness to accept
- *	a CDB, this function is called.  Send the CDB.
- *
- *	LOCKING:
- *	caller.
- */
-
-static void atapi_send_cdb(struct ata_port *ap, struct ata_queued_cmd *qc)
-{
-	/* send SCSI cdb */
-	DPRINTK("send cdb\n");
-	WARN_ON(qc->dev->cdb_len < 12);
-
-	ap->ops->data_xfer(qc->dev, qc->cdb, qc->dev->cdb_len, 1);
-	ata_altstatus(ap); /* flush */
-
-	switch (qc->tf.protocol) {
-	case ATAPI_PROT_PIO:
-		ap->hsm_task_state = HSM_ST;
-		break;
-	case ATAPI_PROT_NODATA:
-		ap->hsm_task_state = HSM_ST_LAST;
-		break;
-	case ATAPI_PROT_DMA:
-		ap->hsm_task_state = HSM_ST_LAST;
-		/* initiate bmdma */
-		ap->ops->bmdma_start(qc);
-		break;
-	}
-}
-
-/**
- *	__atapi_pio_bytes - Transfer data from/to the ATAPI device.
- *	@qc: Command on going
- *	@bytes: number of bytes
- *
- *	Transfer Transfer data from/to the ATAPI device.
- *
- *	LOCKING:
- *	Inherited from caller.
- *
- */
-static int __atapi_pio_bytes(struct ata_queued_cmd *qc, unsigned int bytes)
-{
-	int rw = (qc->tf.flags & ATA_TFLAG_WRITE) ? WRITE : READ;
-	struct ata_port *ap = qc->ap;
-	struct ata_device *dev = qc->dev;
-	struct ata_eh_info *ehi = &dev->link->eh_info;
-	struct scatterlist *sg;
-	struct page *page;
-	unsigned char *buf;
-	unsigned int offset, count, consumed;
-
-next_sg:
-	sg = qc->cursg;
-	if (unlikely(!sg)) {
-		ata_ehi_push_desc(ehi, "unexpected or too much trailing data "
-				  "buf=%u cur=%u bytes=%u",
-				  qc->nbytes, qc->curbytes, bytes);
-		return -1;
-	}
-
-	page = sg_page(sg);
-	offset = sg->offset + qc->cursg_ofs;
-
-	/* get the current page and offset */
-	page = nth_page(page, (offset >> PAGE_SHIFT));
-	offset %= PAGE_SIZE;
-
-	/* don't overrun current sg */
-	count = min(sg->length - qc->cursg_ofs, bytes);
-
-	/* don't cross page boundaries */
-	count = min(count, (unsigned int)PAGE_SIZE - offset);
-
-	DPRINTK("data %s\n", qc->tf.flags & ATA_TFLAG_WRITE ? "write" : "read");
-
-	if (PageHighMem(page)) {
-		unsigned long flags;
-
-		/* FIXME: use bounce buffer */
-		local_irq_save(flags);
-		buf = kmap_atomic(page, KM_IRQ0);
-
-		/* do the actual data transfer */
-		consumed = ap->ops->data_xfer(dev,  buf + offset, count, rw);
-
-		kunmap_atomic(buf, KM_IRQ0);
-		local_irq_restore(flags);
-	} else {
-		buf = page_address(page);
-		consumed = ap->ops->data_xfer(dev,  buf + offset, count, rw);
-	}
-
-	bytes -= min(bytes, consumed);
-	qc->curbytes += count;
-	qc->cursg_ofs += count;
-
-	if (qc->cursg_ofs == sg->length) {
-		qc->cursg = sg_next(qc->cursg);
-		qc->cursg_ofs = 0;
-	}
-
-	/* consumed can be larger than count only for the last transfer */
-	WARN_ON(qc->cursg && count != consumed);
-
-	if (bytes)
-		goto next_sg;
-	return 0;
-}
-
-/**
- *	atapi_pio_bytes - Transfer data from/to the ATAPI device.
- *	@qc: Command on going
- *
- *	Transfer Transfer data from/to the ATAPI device.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-
-static void atapi_pio_bytes(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-	struct ata_device *dev = qc->dev;
-	struct ata_eh_info *ehi = &dev->link->eh_info;
-	unsigned int ireason, bc_lo, bc_hi, bytes;
-	int i_write, do_write = (qc->tf.flags & ATA_TFLAG_WRITE) ? 1 : 0;
-
-	/* Abuse qc->result_tf for temp storage of intermediate TF
-	 * here to save some kernel stack usage.
-	 * For normal completion, qc->result_tf is not relevant. For
-	 * error, qc->result_tf is later overwritten by ata_qc_complete().
-	 * So, the correctness of qc->result_tf is not affected.
-	 */
-	ap->ops->tf_read(ap, &qc->result_tf);
-	ireason = qc->result_tf.nsect;
-	bc_lo = qc->result_tf.lbam;
-	bc_hi = qc->result_tf.lbah;
-	bytes = (bc_hi << 8) | bc_lo;
-
-	/* shall be cleared to zero, indicating xfer of data */
-	if (unlikely(ireason & (1 << 0)))
-		goto atapi_check;
-
-	/* make sure transfer direction matches expected */
-	i_write = ((ireason & (1 << 1)) == 0) ? 1 : 0;
-	if (unlikely(do_write != i_write))
-		goto atapi_check;
-
-	if (unlikely(!bytes))
-		goto atapi_check;
-
-	VPRINTK("ata%u: xfering %d bytes\n", ap->print_id, bytes);
-
-	if (unlikely(__atapi_pio_bytes(qc, bytes)))
-		goto err_out;
-	ata_altstatus(ap); /* flush */
-
-	return;
-
- atapi_check:
-	ata_ehi_push_desc(ehi, "ATAPI check failed (ireason=0x%x bytes=%u)",
-			  ireason, bytes);
- err_out:
-	qc->err_mask |= AC_ERR_HSM;
-	ap->hsm_task_state = HSM_ST_ERR;
-}
-
-/**
- *	ata_hsm_ok_in_wq - Check if the qc can be handled in the workqueue.
- *	@ap: the target ata_port
- *	@qc: qc on going
- *
- *	RETURNS:
- *	1 if ok in workqueue, 0 otherwise.
- */
-
-static inline int ata_hsm_ok_in_wq(struct ata_port *ap, struct ata_queued_cmd *qc)
-{
-	if (qc->tf.flags & ATA_TFLAG_POLLING)
-		return 1;
-
-	if (ap->hsm_task_state == HSM_ST_FIRST) {
-		if (qc->tf.protocol == ATA_PROT_PIO &&
-		    (qc->tf.flags & ATA_TFLAG_WRITE))
-		    return 1;
-
-		if (ata_is_atapi(qc->tf.protocol) &&
-		    !(qc->dev->flags & ATA_DFLAG_CDB_INTR))
-			return 1;
-	}
-
-	return 0;
-}
-
-/**
- *	ata_hsm_qc_complete - finish a qc running on standard HSM
- *	@qc: Command to complete
- *	@in_wq: 1 if called from workqueue, 0 otherwise
- *
- *	Finish @qc which is running on standard HSM.
- *
- *	LOCKING:
- *	If @in_wq is zero, spin_lock_irqsave(host lock).
- *	Otherwise, none on entry and grabs host lock.
- */
-static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
-{
-	struct ata_port *ap = qc->ap;
-	unsigned long flags;
-
-	if (ap->ops->error_handler) {
-		if (in_wq) {
-			spin_lock_irqsave(ap->lock, flags);
-
-			/* EH might have kicked in while host lock is
-			 * released.
-			 */
-			qc = ata_qc_from_tag(ap, qc->tag);
-			if (qc) {
-				if (likely(!(qc->err_mask & AC_ERR_HSM))) {
-					ap->ops->irq_on(ap);
-					ata_qc_complete(qc);
-				} else
-					ata_port_freeze(ap);
-			}
-
-			spin_unlock_irqrestore(ap->lock, flags);
-		} else {
-			if (likely(!(qc->err_mask & AC_ERR_HSM)))
-				ata_qc_complete(qc);
-			else
-				ata_port_freeze(ap);
-		}
-	} else {
-		if (in_wq) {
-			spin_lock_irqsave(ap->lock, flags);
-			ap->ops->irq_on(ap);
-			ata_qc_complete(qc);
-			spin_unlock_irqrestore(ap->lock, flags);
-		} else
-			ata_qc_complete(qc);
-	}
-}
-
-/**
- *	ata_hsm_move - move the HSM to the next state.
- *	@ap: the target ata_port
- *	@qc: qc on going
- *	@status: current device status
- *	@in_wq: 1 if called from workqueue, 0 otherwise
- *
- *	RETURNS:
- *	1 when poll next status needed, 0 otherwise.
- */
-int ata_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
-		 u8 status, int in_wq)
-{
-	unsigned long flags = 0;
-	int poll_next;
-
-	WARN_ON((qc->flags & ATA_QCFLAG_ACTIVE) == 0);
-
-	/* Make sure ata_qc_issue_prot() does not throw things
-	 * like DMA polling into the workqueue. Notice that
-	 * in_wq is not equivalent to (qc->tf.flags & ATA_TFLAG_POLLING).
-	 */
-	WARN_ON(in_wq != ata_hsm_ok_in_wq(ap, qc));
-
-fsm_start:
-	DPRINTK("ata%u: protocol %d task_state %d (dev_stat 0x%X)\n",
-		ap->print_id, qc->tf.protocol, ap->hsm_task_state, status);
-
-	switch (ap->hsm_task_state) {
-	case HSM_ST_FIRST:
-		/* Send first data block or PACKET CDB */
-
-		/* If polling, we will stay in the work queue after
-		 * sending the data. Otherwise, interrupt handler
-		 * takes over after sending the data.
-		 */
-		poll_next = (qc->tf.flags & ATA_TFLAG_POLLING);
-
-		/* check device status */
-		if (unlikely((status & ATA_DRQ) == 0)) {
-			/* handle BSY=0, DRQ=0 as error */
-			if (likely(status & (ATA_ERR | ATA_DF)))
-				/* device stops HSM for abort/error */
-				qc->err_mask |= AC_ERR_DEV;
-			else
-				/* HSM violation. Let EH handle this */
-				qc->err_mask |= AC_ERR_HSM;
-
-			ap->hsm_task_state = HSM_ST_ERR;
-			goto fsm_start;
-		}
-
-		/* Device should not ask for data transfer (DRQ=1)
-		 * when it finds something wrong.
-		 * We ignore DRQ here and stop the HSM by
-		 * changing hsm_task_state to HSM_ST_ERR and
-		 * let the EH abort the command or reset the device.
-		 */
-		if (unlikely(status & (ATA_ERR | ATA_DF))) {
-			/* Some ATAPI tape drives forget to clear the ERR bit
-			 * when doing the next command (mostly request sense).
-			 * We ignore ERR here to workaround and proceed sending
-			 * the CDB.
-			 */
-			if (!(qc->dev->horkage & ATA_HORKAGE_STUCK_ERR)) {
-				ata_port_printk(ap, KERN_WARNING,
-						"DRQ=1 with device error, "
-						"dev_stat 0x%X\n", status);
-				qc->err_mask |= AC_ERR_HSM;
-				ap->hsm_task_state = HSM_ST_ERR;
-				goto fsm_start;
-			}
-		}
-
-		/* Send the CDB (atapi) or the first data block (ata pio out).
-		 * During the state transition, interrupt handler shouldn't
-		 * be invoked before the data transfer is complete and
-		 * hsm_task_state is changed. Hence, the following locking.
-		 */
-		if (in_wq)
-			spin_lock_irqsave(ap->lock, flags);
-
-		if (qc->tf.protocol == ATA_PROT_PIO) {
-			/* PIO data out protocol.
-			 * send first data block.
-			 */
-
-			/* ata_pio_sectors() might change the state
-			 * to HSM_ST_LAST. so, the state is changed here
-			 * before ata_pio_sectors().
-			 */
-			ap->hsm_task_state = HSM_ST;
-			ata_pio_sectors(qc);
-		} else
-			/* send CDB */
-			atapi_send_cdb(ap, qc);
-
-		if (in_wq)
-			spin_unlock_irqrestore(ap->lock, flags);
-
-		/* if polling, ata_pio_task() handles the rest.
-		 * otherwise, interrupt handler takes over from here.
-		 */
-		break;
-
-	case HSM_ST:
-		/* complete command or read/write the data register */
-		if (qc->tf.protocol == ATAPI_PROT_PIO) {
-			/* ATAPI PIO protocol */
-			if ((status & ATA_DRQ) == 0) {
-				/* No more data to transfer or device error.
-				 * Device error will be tagged in HSM_ST_LAST.
-				 */
-				ap->hsm_task_state = HSM_ST_LAST;
-				goto fsm_start;
-			}
-
-			/* Device should not ask for data transfer (DRQ=1)
-			 * when it finds something wrong.
-			 * We ignore DRQ here and stop the HSM by
-			 * changing hsm_task_state to HSM_ST_ERR and
-			 * let the EH abort the command or reset the device.
-			 */
-			if (unlikely(status & (ATA_ERR | ATA_DF))) {
-				ata_port_printk(ap, KERN_WARNING, "DRQ=1 with "
-						"device error, dev_stat 0x%X\n",
-						status);
-				qc->err_mask |= AC_ERR_HSM;
-				ap->hsm_task_state = HSM_ST_ERR;
-				goto fsm_start;
-			}
-
-			atapi_pio_bytes(qc);
-
-			if (unlikely(ap->hsm_task_state == HSM_ST_ERR))
-				/* bad ireason reported by device */
-				goto fsm_start;
-
-		} else {
-			/* ATA PIO protocol */
-			if (unlikely((status & ATA_DRQ) == 0)) {
-				/* handle BSY=0, DRQ=0 as error */
-				if (likely(status & (ATA_ERR | ATA_DF)))
-					/* device stops HSM for abort/error */
-					qc->err_mask |= AC_ERR_DEV;
-				else
-					/* HSM violation. Let EH handle this.
-					 * Phantom devices also trigger this
-					 * condition.  Mark hint.
-					 */
-					qc->err_mask |= AC_ERR_HSM |
-							AC_ERR_NODEV_HINT;
-
-				ap->hsm_task_state = HSM_ST_ERR;
-				goto fsm_start;
-			}
-
-			/* For PIO reads, some devices may ask for
-			 * data transfer (DRQ=1) alone with ERR=1.
-			 * We respect DRQ here and transfer one
-			 * block of junk data before changing the
-			 * hsm_task_state to HSM_ST_ERR.
-			 *
-			 * For PIO writes, ERR=1 DRQ=1 doesn't make
-			 * sense since the data block has been
-			 * transferred to the device.
-			 */
-			if (unlikely(status & (ATA_ERR | ATA_DF))) {
-				/* data might be corrputed */
-				qc->err_mask |= AC_ERR_DEV;
-
-				if (!(qc->tf.flags & ATA_TFLAG_WRITE)) {
-					ata_pio_sectors(qc);
-					status = ata_wait_idle(ap);
-				}
-
-				if (status & (ATA_BUSY | ATA_DRQ))
-					qc->err_mask |= AC_ERR_HSM;
-
-				/* ata_pio_sectors() might change the
-				 * state to HSM_ST_LAST. so, the state
-				 * is changed after ata_pio_sectors().
-				 */
-				ap->hsm_task_state = HSM_ST_ERR;
-				goto fsm_start;
-			}
+ *	spin_lock_irqsave(host lock)
+ *
+ *	RETURNS: 0 when ATAPI DMA can be used
+ *               nonzero otherwise
+ */
+int ata_check_atapi_dma(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
 
-			ata_pio_sectors(qc);
+	/* Don't allow DMA if it isn't multiple of 16 bytes.  Quite a
+	 * few ATAPI devices choke on such DMA requests.
+	 */
+	if (unlikely(qc->nbytes & 15))
+		return 1;
 
-			if (ap->hsm_task_state == HSM_ST_LAST &&
-			    (!(qc->tf.flags & ATA_TFLAG_WRITE))) {
-				/* all data read */
-				status = ata_wait_idle(ap);
-				goto fsm_start;
-			}
-		}
+	if (ap->ops->check_atapi_dma)
+		return ap->ops->check_atapi_dma(qc);
 
-		poll_next = 1;
-		break;
+	return 0;
+}
 
-	case HSM_ST_LAST:
-		if (unlikely(!ata_ok(status))) {
-			qc->err_mask |= __ac_err_mask(status);
-			ap->hsm_task_state = HSM_ST_ERR;
-			goto fsm_start;
-		}
+/**
+ *	ata_std_qc_defer - Check whether a qc needs to be deferred
+ *	@qc: ATA command in question
+ *
+ *	Non-NCQ commands cannot run with any other command, NCQ or
+ *	not.  As upper layer only knows the queue depth, we are
+ *	responsible for maintaining exclusion.  This function checks
+ *	whether a new command @qc can be issued.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ *	RETURNS:
+ *	ATA_DEFER_* if deferring is needed, 0 otherwise.
+ */
+int ata_std_qc_defer(struct ata_queued_cmd *qc)
+{
+	struct ata_link *link = qc->dev->link;
 
-		/* no more data to transfer */
-		DPRINTK("ata%u: dev %u command complete, drv_stat 0x%x\n",
-			ap->print_id, qc->dev->devno, status);
+	if (qc->tf.protocol == ATA_PROT_NCQ) {
+		if (!ata_tag_valid(link->active_tag))
+			return 0;
+	} else {
+		if (!ata_tag_valid(link->active_tag) && !link->sactive)
+			return 0;
+	}
 
-		WARN_ON(qc->err_mask);
+	return ATA_DEFER_LINK;
+}
 
-		ap->hsm_task_state = HSM_ST_IDLE;
+void ata_noop_qc_prep(struct ata_queued_cmd *qc) { }
 
-		/* complete taskfile transaction */
-		ata_hsm_qc_complete(qc, in_wq);
+/**
+ *	ata_sg_init - Associate command with scatter-gather table.
+ *	@qc: Command to be associated
+ *	@sg: Scatter-gather table.
+ *	@n_elem: Number of elements in s/g table.
+ *
+ *	Initialize the data-related elements of queued_cmd @qc
+ *	to point to a scatter-gather table @sg, containing @n_elem
+ *	elements.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ */
+void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
+		 unsigned int n_elem)
+{
+	qc->sg = sg;
+	qc->n_elem = n_elem;
+	qc->cursg = qc->sg;
+}
 
-		poll_next = 0;
-		break;
+/**
+ *	ata_sg_setup - DMA-map the scatter-gather table associated with a command.
+ *	@qc: Command with scatter-gather table to be mapped.
+ *
+ *	DMA-map the scatter-gather table associated with queued_cmd @qc.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ *	RETURNS:
+ *	Zero on success, negative on error.
+ *
+ */
+static int ata_sg_setup(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	unsigned int n_elem;
 
-	case HSM_ST_ERR:
-		/* make sure qc->err_mask is available to
-		 * know what's wrong and recover
-		 */
-		WARN_ON(qc->err_mask == 0);
+	VPRINTK("ENTER, ata%u\n", ap->print_id);
 
-		ap->hsm_task_state = HSM_ST_IDLE;
+	n_elem = dma_map_sg(ap->dev, qc->sg, qc->n_elem, qc->dma_dir);
+	if (n_elem < 1)
+		return -1;
 
-		/* complete taskfile transaction */
-		ata_hsm_qc_complete(qc, in_wq);
+	DPRINTK("%d sg elements mapped\n", n_elem);
 
-		poll_next = 0;
-		break;
-	default:
-		poll_next = 0;
-		BUG();
-	}
+	qc->n_elem = n_elem;
+	qc->flags |= ATA_QCFLAG_DMAMAP;
 
-	return poll_next;
+	return 0;
 }
 
-static void ata_pio_task(struct work_struct *work)
+/**
+ *	swap_buf_le16 - swap halves of 16-bit words in place
+ *	@buf:  Buffer to swap
+ *	@buf_words:  Number of 16-bit words in buffer.
+ *
+ *	Swap halves of 16-bit words if needed to convert from
+ *	little-endian byte order to native cpu byte order, or
+ *	vice-versa.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+void swap_buf_le16(u16 *buf, unsigned int buf_words)
 {
-	struct ata_port *ap =
-		container_of(work, struct ata_port, port_task.work);
-	struct ata_queued_cmd *qc = ap->port_task_data;
-	u8 status;
-	int poll_next;
-
-fsm_start:
-	WARN_ON(ap->hsm_task_state == HSM_ST_IDLE);
-
-	/*
-	 * This is purely heuristic.  This is a fast path.
-	 * Sometimes when we enter, BSY will be cleared in
-	 * a chk-status or two.  If not, the drive is probably seeking
-	 * or something.  Snooze for a couple msecs, then
-	 * chk-status again.  If still busy, queue delayed work.
-	 */
-	status = ata_busy_wait(ap, ATA_BUSY, 5);
-	if (status & ATA_BUSY) {
-		msleep(2);
-		status = ata_busy_wait(ap, ATA_BUSY, 10);
-		if (status & ATA_BUSY) {
-			ata_pio_queue_task(ap, qc, ATA_SHORT_PAUSE);
-			return;
-		}
-	}
-
-	/* move the HSM */
-	poll_next = ata_hsm_move(ap, qc, status, 1);
+#ifdef __BIG_ENDIAN
+	unsigned int i;
 
-	/* another command or interrupt handler
-	 * may be running at this point.
-	 */
-	if (poll_next)
-		goto fsm_start;
+	for (i = 0; i < buf_words; i++)
+		buf[i] = le16_to_cpu(buf[i]);
+#endif /* __BIG_ENDIAN */
 }
 
 /**
@@ -6121,285 +4605,6 @@ err:
 	ata_qc_complete(qc);
 }
 
-/**
- *	ata_qc_issue_prot - issue taskfile to device in proto-dependent manner
- *	@qc: command to issue to device
- *
- *	Using various libata functions and hooks, this function
- *	starts an ATA command.  ATA commands are grouped into
- *	classes called "protocols", and issuing each type of protocol
- *	is slightly different.
- *
- *	May be used as the qc_issue() entry in ata_port_operations.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- *	RETURNS:
- *	Zero on success, AC_ERR_* mask on failure
- */
-
-unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-
-	/* Use polling pio if the LLD doesn't handle
-	 * interrupt driven pio and atapi CDB interrupt.
-	 */
-	if (ap->flags & ATA_FLAG_PIO_POLLING) {
-		switch (qc->tf.protocol) {
-		case ATA_PROT_PIO:
-		case ATA_PROT_NODATA:
-		case ATAPI_PROT_PIO:
-		case ATAPI_PROT_NODATA:
-			qc->tf.flags |= ATA_TFLAG_POLLING;
-			break;
-		case ATAPI_PROT_DMA:
-			if (qc->dev->flags & ATA_DFLAG_CDB_INTR)
-				/* see ata_dma_blacklisted() */
-				BUG();
-			break;
-		default:
-			break;
-		}
-	}
-
-	/* select the device */
-	ata_dev_select(ap, qc->dev->devno, 1, 0);
-
-	/* start the command */
-	switch (qc->tf.protocol) {
-	case ATA_PROT_NODATA:
-		if (qc->tf.flags & ATA_TFLAG_POLLING)
-			ata_qc_set_polling(qc);
-
-		ata_tf_to_host(ap, &qc->tf);
-		ap->hsm_task_state = HSM_ST_LAST;
-
-		if (qc->tf.flags & ATA_TFLAG_POLLING)
-			ata_pio_queue_task(ap, qc, 0);
-
-		break;
-
-	case ATA_PROT_DMA:
-		WARN_ON(qc->tf.flags & ATA_TFLAG_POLLING);
-
-		ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
-		ap->ops->bmdma_setup(qc);	    /* set up bmdma */
-		ap->ops->bmdma_start(qc);	    /* initiate bmdma */
-		ap->hsm_task_state = HSM_ST_LAST;
-		break;
-
-	case ATA_PROT_PIO:
-		if (qc->tf.flags & ATA_TFLAG_POLLING)
-			ata_qc_set_polling(qc);
-
-		ata_tf_to_host(ap, &qc->tf);
-
-		if (qc->tf.flags & ATA_TFLAG_WRITE) {
-			/* PIO data out protocol */
-			ap->hsm_task_state = HSM_ST_FIRST;
-			ata_pio_queue_task(ap, qc, 0);
-
-			/* always send first data block using
-			 * the ata_pio_task() codepath.
-			 */
-		} else {
-			/* PIO data in protocol */
-			ap->hsm_task_state = HSM_ST;
-
-			if (qc->tf.flags & ATA_TFLAG_POLLING)
-				ata_pio_queue_task(ap, qc, 0);
-
-			/* if polling, ata_pio_task() handles the rest.
-			 * otherwise, interrupt handler takes over from here.
-			 */
-		}
-
-		break;
-
-	case ATAPI_PROT_PIO:
-	case ATAPI_PROT_NODATA:
-		if (qc->tf.flags & ATA_TFLAG_POLLING)
-			ata_qc_set_polling(qc);
-
-		ata_tf_to_host(ap, &qc->tf);
-
-		ap->hsm_task_state = HSM_ST_FIRST;
-
-		/* send cdb by polling if no cdb interrupt */
-		if ((!(qc->dev->flags & ATA_DFLAG_CDB_INTR)) ||
-		    (qc->tf.flags & ATA_TFLAG_POLLING))
-			ata_pio_queue_task(ap, qc, 0);
-		break;
-
-	case ATAPI_PROT_DMA:
-		WARN_ON(qc->tf.flags & ATA_TFLAG_POLLING);
-
-		ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
-		ap->ops->bmdma_setup(qc);	    /* set up bmdma */
-		ap->hsm_task_state = HSM_ST_FIRST;
-
-		/* send cdb by polling if no cdb interrupt */
-		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
-			ata_pio_queue_task(ap, qc, 0);
-		break;
-
-	default:
-		WARN_ON(1);
-		return AC_ERR_SYSTEM;
-	}
-
-	return 0;
-}
-
-/**
- *	ata_host_intr - Handle host interrupt for given (port, task)
- *	@ap: Port on which interrupt arrived (possibly...)
- *	@qc: Taskfile currently active in engine
- *
- *	Handle host interrupt for given queued command.  Currently,
- *	only DMA interrupts are handled.  All other commands are
- *	handled via polling with interrupts disabled (nIEN bit).
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- *
- *	RETURNS:
- *	One if interrupt was handled, zero if not (shared irq).
- */
-
-inline unsigned int ata_host_intr(struct ata_port *ap,
-				  struct ata_queued_cmd *qc)
-{
-	struct ata_eh_info *ehi = &ap->link.eh_info;
-	u8 status, host_stat = 0;
-
-	VPRINTK("ata%u: protocol %d task_state %d\n",
-		ap->print_id, qc->tf.protocol, ap->hsm_task_state);
-
-	/* Check whether we are expecting interrupt in this state */
-	switch (ap->hsm_task_state) {
-	case HSM_ST_FIRST:
-		/* Some pre-ATAPI-4 devices assert INTRQ
-		 * at this state when ready to receive CDB.
-		 */
-
-		/* Check the ATA_DFLAG_CDB_INTR flag is enough here.
-		 * The flag was turned on only for atapi devices.  No
-		 * need to check ata_is_atapi(qc->tf.protocol) again.
-		 */
-		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
-			goto idle_irq;
-		break;
-	case HSM_ST_LAST:
-		if (qc->tf.protocol == ATA_PROT_DMA ||
-		    qc->tf.protocol == ATAPI_PROT_DMA) {
-			/* check status of DMA engine */
-			host_stat = ap->ops->bmdma_status(ap);
-			VPRINTK("ata%u: host_stat 0x%X\n",
-				ap->print_id, host_stat);
-
-			/* if it's not our irq... */
-			if (!(host_stat & ATA_DMA_INTR))
-				goto idle_irq;
-
-			/* before we do anything else, clear DMA-Start bit */
-			ap->ops->bmdma_stop(qc);
-
-			if (unlikely(host_stat & ATA_DMA_ERR)) {
-				/* error when transfering data to/from memory */
-				qc->err_mask |= AC_ERR_HOST_BUS;
-				ap->hsm_task_state = HSM_ST_ERR;
-			}
-		}
-		break;
-	case HSM_ST:
-		break;
-	default:
-		goto idle_irq;
-	}
-
-	/* check altstatus */
-	status = ata_altstatus(ap);
-	if (status & ATA_BUSY)
-		goto idle_irq;
-
-	/* check main status, clearing INTRQ */
-	status = ata_chk_status(ap);
-	if (unlikely(status & ATA_BUSY))
-		goto idle_irq;
-
-	/* ack bmdma irq events */
-	ap->ops->irq_clear(ap);
-
-	ata_hsm_move(ap, qc, status, 0);
-
-	if (unlikely(qc->err_mask) && (qc->tf.protocol == ATA_PROT_DMA ||
-				       qc->tf.protocol == ATAPI_PROT_DMA))
-		ata_ehi_push_desc(ehi, "BMDMA stat 0x%x", host_stat);
-
-	return 1;	/* irq handled */
-
-idle_irq:
-	ap->stats.idle_irq++;
-
-#ifdef ATA_IRQ_TRAP
-	if ((ap->stats.idle_irq % 1000) == 0) {
-		ata_chk_status(ap);
-		ap->ops->irq_clear(ap);
-		ata_port_printk(ap, KERN_WARNING, "irq trap\n");
-		return 1;
-	}
-#endif
-	return 0;	/* irq not handled */
-}
-
-/**
- *	ata_interrupt - Default ATA host interrupt handler
- *	@irq: irq line (unused)
- *	@dev_instance: pointer to our ata_host information structure
- *
- *	Default interrupt handler for PCI IDE devices.  Calls
- *	ata_host_intr() for each port that is not disabled.
- *
- *	LOCKING:
- *	Obtains host lock during operation.
- *
- *	RETURNS:
- *	IRQ_NONE or IRQ_HANDLED.
- */
-
-irqreturn_t ata_interrupt(int irq, void *dev_instance)
-{
-	struct ata_host *host = dev_instance;
-	unsigned int i;
-	unsigned int handled = 0;
-	unsigned long flags;
-
-	/* TODO: make _irqsave conditional on x86 PCI IDE legacy mode */
-	spin_lock_irqsave(&host->lock, flags);
-
-	for (i = 0; i < host->n_ports; i++) {
-		struct ata_port *ap;
-
-		ap = host->ports[i];
-		if (ap &&
-		    !(ap->flags & ATA_FLAG_DISABLED)) {
-			struct ata_queued_cmd *qc;
-
-			qc = ata_qc_from_tag(ap, ap->link.active_tag);
-			if (qc && (!(qc->tf.flags & ATA_TFLAG_POLLING)) &&
-			    (qc->flags & ATA_QCFLAG_ACTIVE))
-				handled |= ata_host_intr(ap, qc);
-		}
-	}
-
-	spin_unlock_irqrestore(&host->lock, flags);
-
-	return IRQ_RETVAL(handled);
-}
-
 /**
  *	sata_scr_valid - test whether SCRs are accessible
  *	@link: ATA link to test SCR accessibility for
@@ -7432,33 +5637,6 @@ void ata_host_detach(struct ata_host *host)
 	ata_acpi_dissociate(host);
 }
 
-/**
- *	ata_std_ports - initialize ioaddr with standard port offsets.
- *	@ioaddr: IO address structure to be initialized
- *
- *	Utility function which initializes data_addr, error_addr,
- *	feature_addr, nsect_addr, lbal_addr, lbam_addr, lbah_addr,
- *	device_addr, status_addr, and command_addr to standard offsets
- *	relative to cmd_addr.
- *
- *	Does not set ctl_addr, altstatus_addr, bmdma_addr, or scr_addr.
- */
-
-void ata_std_ports(struct ata_ioports *ioaddr)
-{
-	ioaddr->data_addr = ioaddr->cmd_addr + ATA_REG_DATA;
-	ioaddr->error_addr = ioaddr->cmd_addr + ATA_REG_ERR;
-	ioaddr->feature_addr = ioaddr->cmd_addr + ATA_REG_FEATURE;
-	ioaddr->nsect_addr = ioaddr->cmd_addr + ATA_REG_NSECT;
-	ioaddr->lbal_addr = ioaddr->cmd_addr + ATA_REG_LBAL;
-	ioaddr->lbam_addr = ioaddr->cmd_addr + ATA_REG_LBAM;
-	ioaddr->lbah_addr = ioaddr->cmd_addr + ATA_REG_LBAH;
-	ioaddr->device_addr = ioaddr->cmd_addr + ATA_REG_DEVICE;
-	ioaddr->status_addr = ioaddr->cmd_addr + ATA_REG_STATUS;
-	ioaddr->command_addr = ioaddr->cmd_addr + ATA_REG_CMD;
-}
-
-
 #ifdef CONFIG_PCI
 
 /**
@@ -7890,12 +6068,9 @@ EXPORT_SYMBOL_GPL(sata_deb_timing_long);
 EXPORT_SYMBOL_GPL(ata_base_port_ops);
 EXPORT_SYMBOL_GPL(sata_port_ops);
 EXPORT_SYMBOL_GPL(sata_pmp_port_ops);
-EXPORT_SYMBOL_GPL(ata_sff_port_ops);
-EXPORT_SYMBOL_GPL(ata_bmdma_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_info);
 EXPORT_SYMBOL_GPL(ata_std_bios_param);
-EXPORT_SYMBOL_GPL(ata_std_ports);
 EXPORT_SYMBOL_GPL(ata_host_init);
 EXPORT_SYMBOL_GPL(ata_host_alloc);
 EXPORT_SYMBOL_GPL(ata_host_alloc_pinfo);
@@ -7904,14 +6079,9 @@ EXPORT_SYMBOL_GPL(ata_host_register);
 EXPORT_SYMBOL_GPL(ata_host_activate);
 EXPORT_SYMBOL_GPL(ata_host_detach);
 EXPORT_SYMBOL_GPL(ata_sg_init);
-EXPORT_SYMBOL_GPL(ata_hsm_move);
 EXPORT_SYMBOL_GPL(ata_qc_complete);
 EXPORT_SYMBOL_GPL(ata_qc_complete_multiple);
-EXPORT_SYMBOL_GPL(ata_qc_issue_prot);
-EXPORT_SYMBOL_GPL(ata_tf_load);
-EXPORT_SYMBOL_GPL(ata_tf_read);
 EXPORT_SYMBOL_GPL(ata_noop_dev_select);
-EXPORT_SYMBOL_GPL(ata_std_dev_select);
 EXPORT_SYMBOL_GPL(sata_print_link_status);
 EXPORT_SYMBOL_GPL(atapi_cmd_type);
 EXPORT_SYMBOL_GPL(ata_tf_to_fis);
@@ -7923,54 +6093,27 @@ EXPORT_SYMBOL_GPL(ata_xfer_mode2mask);
 EXPORT_SYMBOL_GPL(ata_xfer_mode2shift);
 EXPORT_SYMBOL_GPL(ata_mode_string);
 EXPORT_SYMBOL_GPL(ata_id_xfermask);
-EXPORT_SYMBOL_GPL(ata_check_status);
-EXPORT_SYMBOL_GPL(ata_altstatus);
-EXPORT_SYMBOL_GPL(ata_exec_command);
 EXPORT_SYMBOL_GPL(ata_port_start);
-EXPORT_SYMBOL_GPL(ata_sff_port_start);
-EXPORT_SYMBOL_GPL(ata_interrupt);
 EXPORT_SYMBOL_GPL(ata_do_set_mode);
-EXPORT_SYMBOL_GPL(ata_data_xfer);
-EXPORT_SYMBOL_GPL(ata_data_xfer_noirq);
 EXPORT_SYMBOL_GPL(ata_std_qc_defer);
-EXPORT_SYMBOL_GPL(ata_qc_prep);
-EXPORT_SYMBOL_GPL(ata_dumb_qc_prep);
 EXPORT_SYMBOL_GPL(ata_noop_qc_prep);
-EXPORT_SYMBOL_GPL(ata_bmdma_setup);
-EXPORT_SYMBOL_GPL(ata_bmdma_start);
-EXPORT_SYMBOL_GPL(ata_bmdma_irq_clear);
 EXPORT_SYMBOL_GPL(ata_noop_irq_clear);
-EXPORT_SYMBOL_GPL(ata_bmdma_status);
-EXPORT_SYMBOL_GPL(ata_bmdma_stop);
-EXPORT_SYMBOL_GPL(ata_bmdma_freeze);
-EXPORT_SYMBOL_GPL(ata_bmdma_thaw);
-EXPORT_SYMBOL_GPL(ata_bmdma_error_handler);
-EXPORT_SYMBOL_GPL(ata_bmdma_post_internal_cmd);
 EXPORT_SYMBOL_GPL(ata_port_probe);
 EXPORT_SYMBOL_GPL(ata_dev_disable);
 EXPORT_SYMBOL_GPL(sata_set_spd);
 EXPORT_SYMBOL_GPL(sata_link_debounce);
 EXPORT_SYMBOL_GPL(sata_link_resume);
-EXPORT_SYMBOL_GPL(ata_bus_reset);
-EXPORT_SYMBOL_GPL(ata_std_prereset);
-EXPORT_SYMBOL_GPL(ata_std_softreset);
 EXPORT_SYMBOL_GPL(sata_link_hardreset);
-EXPORT_SYMBOL_GPL(sata_std_hardreset);
-EXPORT_SYMBOL_GPL(ata_std_postreset);
 EXPORT_SYMBOL_GPL(ata_dev_classify);
 EXPORT_SYMBOL_GPL(ata_dev_pair);
 EXPORT_SYMBOL_GPL(ata_port_disable);
 EXPORT_SYMBOL_GPL(ata_ratelimit);
 EXPORT_SYMBOL_GPL(ata_wait_register);
-EXPORT_SYMBOL_GPL(ata_busy_sleep);
-EXPORT_SYMBOL_GPL(ata_wait_after_reset);
-EXPORT_SYMBOL_GPL(ata_wait_ready);
 EXPORT_SYMBOL_GPL(ata_scsi_ioctl);
 EXPORT_SYMBOL_GPL(ata_scsi_queuecmd);
 EXPORT_SYMBOL_GPL(ata_scsi_slave_config);
 EXPORT_SYMBOL_GPL(ata_scsi_slave_destroy);
 EXPORT_SYMBOL_GPL(ata_scsi_change_queue_depth);
-EXPORT_SYMBOL_GPL(ata_host_intr);
 EXPORT_SYMBOL_GPL(sata_scr_valid);
 EXPORT_SYMBOL_GPL(sata_scr_read);
 EXPORT_SYMBOL_GPL(sata_scr_write);
@@ -7993,11 +6136,6 @@ EXPORT_SYMBOL_GPL(ata_timing_cycle2mode);
 
 #ifdef CONFIG_PCI
 EXPORT_SYMBOL_GPL(pci_test_config_bits);
-EXPORT_SYMBOL_GPL(ata_pci_init_sff_host);
-EXPORT_SYMBOL_GPL(ata_pci_init_bmdma);
-EXPORT_SYMBOL_GPL(ata_pci_prepare_sff_host);
-EXPORT_SYMBOL_GPL(ata_pci_activate_sff_host);
-EXPORT_SYMBOL_GPL(ata_pci_init_one);
 EXPORT_SYMBOL_GPL(ata_pci_remove_one);
 #ifdef CONFIG_PM
 EXPORT_SYMBOL_GPL(ata_pci_device_do_suspend);
@@ -8005,8 +6143,6 @@ EXPORT_SYMBOL_GPL(ata_pci_device_do_resume);
 EXPORT_SYMBOL_GPL(ata_pci_device_suspend);
 EXPORT_SYMBOL_GPL(ata_pci_device_resume);
 #endif /* CONFIG_PM */
-EXPORT_SYMBOL_GPL(ata_pci_default_filter);
-EXPORT_SYMBOL_GPL(ata_pci_clear_simplex);
 #endif /* CONFIG_PCI */
 
 EXPORT_SYMBOL_GPL(sata_pmp_qc_defer_cmd_switch);
@@ -8033,8 +6169,6 @@ EXPORT_SYMBOL_GPL(ata_eh_qc_complete);
 EXPORT_SYMBOL_GPL(ata_eh_qc_retry);
 EXPORT_SYMBOL_GPL(ata_do_eh);
 EXPORT_SYMBOL_GPL(ata_std_error_handler);
-EXPORT_SYMBOL_GPL(ata_irq_on);
-EXPORT_SYMBOL_GPL(ata_dev_try_classify);
 
 EXPORT_SYMBOL_GPL(ata_cable_40wire);
 EXPORT_SYMBOL_GPL(ata_cable_80wire);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 40645ed125b1..840ae6da59bc 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -35,9 +35,185 @@
 #include <linux/kernel.h>
 #include <linux/pci.h>
 #include <linux/libata.h>
+#include <linux/highmem.h>
 
 #include "libata.h"
 
+const struct ata_port_operations ata_sff_port_ops = {
+	.inherits		= &ata_base_port_ops,
+
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.softreset		= ata_std_softreset,
+	.error_handler		= ata_bmdma_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.dev_select		= ata_std_dev_select,
+	.check_status		= ata_check_status,
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.exec_command		= ata_exec_command,
+	.data_xfer		= ata_data_xfer,
+	.irq_on			= ata_irq_on,
+
+	.port_start		= ata_sff_port_start,
+};
+
+const struct ata_port_operations ata_bmdma_port_ops = {
+	.inherits		= &ata_sff_port_ops,
+
+	.mode_filter		= ata_pci_default_filter,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.irq_clear		= ata_bmdma_irq_clear,
+};
+
+/**
+ *	ata_fill_sg - Fill PCI IDE PRD table
+ *	@qc: Metadata associated with taskfile to be transferred
+ *
+ *	Fill PCI IDE PRD (scatter-gather) table with segments
+ *	associated with the current disk command.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ */
+static void ata_fill_sg(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct scatterlist *sg;
+	unsigned int si, pi;
+
+	pi = 0;
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
+		u32 addr, offset;
+		u32 sg_len, len;
+
+		/* determine if physical DMA addr spans 64K boundary.
+		 * Note h/w doesn't support 64-bit, so we unconditionally
+		 * truncate dma_addr_t to u32.
+		 */
+		addr = (u32) sg_dma_address(sg);
+		sg_len = sg_dma_len(sg);
+
+		while (sg_len) {
+			offset = addr & 0xffff;
+			len = sg_len;
+			if ((offset + sg_len) > 0x10000)
+				len = 0x10000 - offset;
+
+			ap->prd[pi].addr = cpu_to_le32(addr);
+			ap->prd[pi].flags_len = cpu_to_le32(len & 0xffff);
+			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", pi, addr, len);
+
+			pi++;
+			sg_len -= len;
+			addr += len;
+		}
+	}
+
+	ap->prd[pi - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
+}
+
+/**
+ *	ata_fill_sg_dumb - Fill PCI IDE PRD table
+ *	@qc: Metadata associated with taskfile to be transferred
+ *
+ *	Fill PCI IDE PRD (scatter-gather) table with segments
+ *	associated with the current disk command. Perform the fill
+ *	so that we avoid writing any length 64K records for
+ *	controllers that don't follow the spec.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ */
+static void ata_fill_sg_dumb(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct scatterlist *sg;
+	unsigned int si, pi;
+
+	pi = 0;
+	for_each_sg(qc->sg, sg, qc->n_elem, si) {
+		u32 addr, offset;
+		u32 sg_len, len, blen;
+
+		/* determine if physical DMA addr spans 64K boundary.
+		 * Note h/w doesn't support 64-bit, so we unconditionally
+		 * truncate dma_addr_t to u32.
+		 */
+		addr = (u32) sg_dma_address(sg);
+		sg_len = sg_dma_len(sg);
+
+		while (sg_len) {
+			offset = addr & 0xffff;
+			len = sg_len;
+			if ((offset + sg_len) > 0x10000)
+				len = 0x10000 - offset;
+
+			blen = len & 0xffff;
+			ap->prd[pi].addr = cpu_to_le32(addr);
+			if (blen == 0) {
+			   /* Some PATA chipsets like the CS5530 can't
+			      cope with 0x0000 meaning 64K as the spec says */
+				ap->prd[pi].flags_len = cpu_to_le32(0x8000);
+				blen = 0x8000;
+				ap->prd[++pi].addr = cpu_to_le32(addr + 0x8000);
+			}
+			ap->prd[pi].flags_len = cpu_to_le32(blen);
+			VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", pi, addr, len);
+
+			pi++;
+			sg_len -= len;
+			addr += len;
+		}
+	}
+
+	ap->prd[pi - 1].flags_len |= cpu_to_le32(ATA_PRD_EOT);
+}
+
+/**
+ *	ata_qc_prep - Prepare taskfile for submission
+ *	@qc: Metadata associated with taskfile to be prepared
+ *
+ *	Prepare ATA taskfile for submission.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ */
+void ata_qc_prep(struct ata_queued_cmd *qc)
+{
+	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
+		return;
+
+	ata_fill_sg(qc);
+}
+
+/**
+ *	ata_dumb_qc_prep - Prepare taskfile for submission
+ *	@qc: Metadata associated with taskfile to be prepared
+ *
+ *	Prepare ATA taskfile for submission.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ */
+void ata_dumb_qc_prep(struct ata_queued_cmd *qc)
+{
+	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
+		return;
+
+	ata_fill_sg_dumb(qc);
+}
+
 /**
  *	ata_check_status - Read device status reg & clear interrupt
  *	@ap: port where the device is
@@ -67,223 +243,1746 @@ u8 ata_check_status(struct ata_port *ap)
  *	LOCKING:
  *	Inherited from caller.
  */
-u8 ata_altstatus(struct ata_port *ap)
+u8 ata_altstatus(struct ata_port *ap)
+{
+	if (ap->ops->check_altstatus)
+		return ap->ops->check_altstatus(ap);
+
+	return ioread8(ap->ioaddr.altstatus_addr);
+}
+
+/**
+ *	ata_busy_sleep - sleep until BSY clears, or timeout
+ *	@ap: port containing status register to be polled
+ *	@tmout_pat: impatience timeout
+ *	@tmout: overall timeout
+ *
+ *	Sleep until ATA Status register bit BSY clears,
+ *	or a timeout occurs.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep).
+ *
+ *	RETURNS:
+ *	0 on success, -errno otherwise.
+ */
+int ata_busy_sleep(struct ata_port *ap,
+		   unsigned long tmout_pat, unsigned long tmout)
+{
+	unsigned long timer_start, timeout;
+	u8 status;
+
+	status = ata_busy_wait(ap, ATA_BUSY, 300);
+	timer_start = jiffies;
+	timeout = timer_start + tmout_pat;
+	while (status != 0xff && (status & ATA_BUSY) &&
+	       time_before(jiffies, timeout)) {
+		msleep(50);
+		status = ata_busy_wait(ap, ATA_BUSY, 3);
+	}
+
+	if (status != 0xff && (status & ATA_BUSY))
+		ata_port_printk(ap, KERN_WARNING,
+				"port is slow to respond, please be patient "
+				"(Status 0x%x)\n", status);
+
+	timeout = timer_start + tmout;
+	while (status != 0xff && (status & ATA_BUSY) &&
+	       time_before(jiffies, timeout)) {
+		msleep(50);
+		status = ata_chk_status(ap);
+	}
+
+	if (status == 0xff)
+		return -ENODEV;
+
+	if (status & ATA_BUSY) {
+		ata_port_printk(ap, KERN_ERR, "port failed to respond "
+				"(%lu secs, Status 0x%x)\n",
+				tmout / HZ, status);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+/**
+ *	ata_wait_ready - sleep until BSY clears, or timeout
+ *	@ap: port containing status register to be polled
+ *	@deadline: deadline jiffies for the operation
+ *
+ *	Sleep until ATA Status register bit BSY clears, or timeout
+ *	occurs.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep).
+ *
+ *	RETURNS:
+ *	0 on success, -errno otherwise.
+ */
+int ata_wait_ready(struct ata_port *ap, unsigned long deadline)
+{
+	unsigned long start = jiffies;
+	int warned = 0;
+
+	while (1) {
+		u8 status = ata_chk_status(ap);
+		unsigned long now = jiffies;
+
+		if (!(status & ATA_BUSY))
+			return 0;
+		if (!ata_link_online(&ap->link) && status == 0xff)
+			return -ENODEV;
+		if (time_after(now, deadline))
+			return -EBUSY;
+
+		if (!warned && time_after(now, start + 5 * HZ) &&
+		    (deadline - now > 3 * HZ)) {
+			ata_port_printk(ap, KERN_WARNING,
+				"port is slow to respond, please be patient "
+				"(Status 0x%x)\n", status);
+			warned = 1;
+		}
+
+		msleep(50);
+	}
+}
+
+/**
+ *	ata_std_dev_select - Select device 0/1 on ATA bus
+ *	@ap: ATA channel to manipulate
+ *	@device: ATA device (numbered from zero) to select
+ *
+ *	Use the method defined in the ATA specification to
+ *	make either device 0, or device 1, active on the
+ *	ATA channel.  Works with both PIO and MMIO.
+ *
+ *	May be used as the dev_select() entry in ata_port_operations.
+ *
+ *	LOCKING:
+ *	caller.
+ */
+void ata_std_dev_select(struct ata_port *ap, unsigned int device)
+{
+	u8 tmp;
+
+	if (device == 0)
+		tmp = ATA_DEVICE_OBS;
+	else
+		tmp = ATA_DEVICE_OBS | ATA_DEV1;
+
+	iowrite8(tmp, ap->ioaddr.device_addr);
+	ata_pause(ap);		/* needed; also flushes, for mmio */
+}
+
+/**
+ *	ata_dev_select - Select device 0/1 on ATA bus
+ *	@ap: ATA channel to manipulate
+ *	@device: ATA device (numbered from zero) to select
+ *	@wait: non-zero to wait for Status register BSY bit to clear
+ *	@can_sleep: non-zero if context allows sleeping
+ *
+ *	Use the method defined in the ATA specification to
+ *	make either device 0, or device 1, active on the
+ *	ATA channel.
+ *
+ *	This is a high-level version of ata_std_dev_select(),
+ *	which additionally provides the services of inserting
+ *	the proper pauses and status polling, where needed.
+ *
+ *	LOCKING:
+ *	caller.
+ */
+void ata_dev_select(struct ata_port *ap, unsigned int device,
+			   unsigned int wait, unsigned int can_sleep)
+{
+	if (ata_msg_probe(ap))
+		ata_port_printk(ap, KERN_INFO, "ata_dev_select: ENTER, "
+				"device %u, wait %u\n", device, wait);
+
+	if (wait)
+		ata_wait_idle(ap);
+
+	ap->ops->dev_select(ap, device);
+
+	if (wait) {
+		if (can_sleep && ap->link.device[device].class == ATA_DEV_ATAPI)
+			msleep(150);
+		ata_wait_idle(ap);
+	}
+}
+
+/**
+ *	ata_irq_on - Enable interrupts on a port.
+ *	@ap: Port on which interrupts are enabled.
+ *
+ *	Enable interrupts on a legacy IDE device using MMIO or PIO,
+ *	wait for idle, clear any pending interrupts.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+u8 ata_irq_on(struct ata_port *ap)
+{
+	struct ata_ioports *ioaddr = &ap->ioaddr;
+	u8 tmp;
+
+	ap->ctl &= ~ATA_NIEN;
+	ap->last_ctl = ap->ctl;
+
+	if (ioaddr->ctl_addr)
+		iowrite8(ap->ctl, ioaddr->ctl_addr);
+	tmp = ata_wait_idle(ap);
+
+	ap->ops->irq_clear(ap);
+
+	return tmp;
+}
+
+/**
+ *	ata_bmdma_irq_clear - Clear PCI IDE BMDMA interrupt.
+ *	@ap: Port associated with this ATA transaction.
+ *
+ *	Clear interrupt and error flags in DMA status register.
+ *
+ *	May be used as the irq_clear() entry in ata_port_operations.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ */
+void ata_bmdma_irq_clear(struct ata_port *ap)
+{
+	void __iomem *mmio = ap->ioaddr.bmdma_addr;
+
+	if (!mmio)
+		return;
+
+	iowrite8(ioread8(mmio + ATA_DMA_STATUS), mmio + ATA_DMA_STATUS);
+}
+
+/**
+ *	ata_tf_load - send taskfile registers to host controller
+ *	@ap: Port to which output is sent
+ *	@tf: ATA taskfile register set
+ *
+ *	Outputs ATA taskfile to standard ATA host controller.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
+{
+	struct ata_ioports *ioaddr = &ap->ioaddr;
+	unsigned int is_addr = tf->flags & ATA_TFLAG_ISADDR;
+
+	if (tf->ctl != ap->last_ctl) {
+		if (ioaddr->ctl_addr)
+			iowrite8(tf->ctl, ioaddr->ctl_addr);
+		ap->last_ctl = tf->ctl;
+		ata_wait_idle(ap);
+	}
+
+	if (is_addr && (tf->flags & ATA_TFLAG_LBA48)) {
+		WARN_ON(!ioaddr->ctl_addr);
+		iowrite8(tf->hob_feature, ioaddr->feature_addr);
+		iowrite8(tf->hob_nsect, ioaddr->nsect_addr);
+		iowrite8(tf->hob_lbal, ioaddr->lbal_addr);
+		iowrite8(tf->hob_lbam, ioaddr->lbam_addr);
+		iowrite8(tf->hob_lbah, ioaddr->lbah_addr);
+		VPRINTK("hob: feat 0x%X nsect 0x%X, lba 0x%X 0x%X 0x%X\n",
+			tf->hob_feature,
+			tf->hob_nsect,
+			tf->hob_lbal,
+			tf->hob_lbam,
+			tf->hob_lbah);
+	}
+
+	if (is_addr) {
+		iowrite8(tf->feature, ioaddr->feature_addr);
+		iowrite8(tf->nsect, ioaddr->nsect_addr);
+		iowrite8(tf->lbal, ioaddr->lbal_addr);
+		iowrite8(tf->lbam, ioaddr->lbam_addr);
+		iowrite8(tf->lbah, ioaddr->lbah_addr);
+		VPRINTK("feat 0x%X nsect 0x%X lba 0x%X 0x%X 0x%X\n",
+			tf->feature,
+			tf->nsect,
+			tf->lbal,
+			tf->lbam,
+			tf->lbah);
+	}
+
+	if (tf->flags & ATA_TFLAG_DEVICE) {
+		iowrite8(tf->device, ioaddr->device_addr);
+		VPRINTK("device 0x%X\n", tf->device);
+	}
+
+	ata_wait_idle(ap);
+}
+
+/**
+ *	ata_tf_read - input device's ATA taskfile shadow registers
+ *	@ap: Port from which input is read
+ *	@tf: ATA taskfile register set for storing input
+ *
+ *	Reads ATA taskfile registers for currently-selected device
+ *	into @tf. Assumes the device has a fully SFF compliant task file
+ *	layout and behaviour. If you device does not (eg has a different
+ *	status method) then you will need to provide a replacement tf_read
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
+{
+	struct ata_ioports *ioaddr = &ap->ioaddr;
+
+	tf->command = ata_check_status(ap);
+	tf->feature = ioread8(ioaddr->error_addr);
+	tf->nsect = ioread8(ioaddr->nsect_addr);
+	tf->lbal = ioread8(ioaddr->lbal_addr);
+	tf->lbam = ioread8(ioaddr->lbam_addr);
+	tf->lbah = ioread8(ioaddr->lbah_addr);
+	tf->device = ioread8(ioaddr->device_addr);
+
+	if (tf->flags & ATA_TFLAG_LBA48) {
+		if (likely(ioaddr->ctl_addr)) {
+			iowrite8(tf->ctl | ATA_HOB, ioaddr->ctl_addr);
+			tf->hob_feature = ioread8(ioaddr->error_addr);
+			tf->hob_nsect = ioread8(ioaddr->nsect_addr);
+			tf->hob_lbal = ioread8(ioaddr->lbal_addr);
+			tf->hob_lbam = ioread8(ioaddr->lbam_addr);
+			tf->hob_lbah = ioread8(ioaddr->lbah_addr);
+			iowrite8(tf->ctl, ioaddr->ctl_addr);
+			ap->last_ctl = tf->ctl;
+		} else
+			WARN_ON(1);
+	}
+}
+
+/**
+ *	ata_exec_command - issue ATA command to host controller
+ *	@ap: port to which command is being issued
+ *	@tf: ATA taskfile register set
+ *
+ *	Issues ATA command, with proper synchronization with interrupt
+ *	handler / other threads.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ */
+void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf)
+{
+	DPRINTK("ata%u: cmd 0x%X\n", ap->print_id, tf->command);
+
+	iowrite8(tf->command, ap->ioaddr.command_addr);
+	ata_pause(ap);
+}
+
+/**
+ *	ata_tf_to_host - issue ATA taskfile to host controller
+ *	@ap: port to which command is being issued
+ *	@tf: ATA taskfile register set
+ *
+ *	Issues ATA taskfile register set to ATA host controller,
+ *	with proper synchronization with interrupt handler and
+ *	other threads.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ */
+static inline void ata_tf_to_host(struct ata_port *ap,
+				  const struct ata_taskfile *tf)
+{
+	ap->ops->tf_load(ap, tf);
+	ap->ops->exec_command(ap, tf);
+}
+
+/**
+ *	ata_data_xfer - Transfer data by PIO
+ *	@dev: device to target
+ *	@buf: data buffer
+ *	@buflen: buffer length
+ *	@rw: read/write
+ *
+ *	Transfer data from/to the device data register by PIO.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ *
+ *	RETURNS:
+ *	Bytes consumed.
+ */
+unsigned int ata_data_xfer(struct ata_device *dev, unsigned char *buf,
+			   unsigned int buflen, int rw)
+{
+	struct ata_port *ap = dev->link->ap;
+	void __iomem *data_addr = ap->ioaddr.data_addr;
+	unsigned int words = buflen >> 1;
+
+	/* Transfer multiple of 2 bytes */
+	if (rw == READ)
+		ioread16_rep(data_addr, buf, words);
+	else
+		iowrite16_rep(data_addr, buf, words);
+
+	/* Transfer trailing 1 byte, if any. */
+	if (unlikely(buflen & 0x01)) {
+		__le16 align_buf[1] = { 0 };
+		unsigned char *trailing_buf = buf + buflen - 1;
+
+		if (rw == READ) {
+			align_buf[0] = cpu_to_le16(ioread16(data_addr));
+			memcpy(trailing_buf, align_buf, 1);
+		} else {
+			memcpy(align_buf, trailing_buf, 1);
+			iowrite16(le16_to_cpu(align_buf[0]), data_addr);
+		}
+		words++;
+	}
+
+	return words << 1;
+}
+
+/**
+ *	ata_data_xfer_noirq - Transfer data by PIO
+ *	@dev: device to target
+ *	@buf: data buffer
+ *	@buflen: buffer length
+ *	@rw: read/write
+ *
+ *	Transfer data from/to the device data register by PIO. Do the
+ *	transfer with interrupts disabled.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ *
+ *	RETURNS:
+ *	Bytes consumed.
+ */
+unsigned int ata_data_xfer_noirq(struct ata_device *dev, unsigned char *buf,
+				 unsigned int buflen, int rw)
+{
+	unsigned long flags;
+	unsigned int consumed;
+
+	local_irq_save(flags);
+	consumed = ata_data_xfer(dev, buf, buflen, rw);
+	local_irq_restore(flags);
+
+	return consumed;
+}
+
+/**
+ *	ata_pio_sector - Transfer a sector of data.
+ *	@qc: Command on going
+ *
+ *	Transfer qc->sect_size bytes of data from/to the ATA device.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+static void ata_pio_sector(struct ata_queued_cmd *qc)
+{
+	int do_write = (qc->tf.flags & ATA_TFLAG_WRITE);
+	struct ata_port *ap = qc->ap;
+	struct page *page;
+	unsigned int offset;
+	unsigned char *buf;
+
+	if (qc->curbytes == qc->nbytes - qc->sect_size)
+		ap->hsm_task_state = HSM_ST_LAST;
+
+	page = sg_page(qc->cursg);
+	offset = qc->cursg->offset + qc->cursg_ofs;
+
+	/* get the current page and offset */
+	page = nth_page(page, (offset >> PAGE_SHIFT));
+	offset %= PAGE_SIZE;
+
+	DPRINTK("data %s\n", qc->tf.flags & ATA_TFLAG_WRITE ? "write" : "read");
+
+	if (PageHighMem(page)) {
+		unsigned long flags;
+
+		/* FIXME: use a bounce buffer */
+		local_irq_save(flags);
+		buf = kmap_atomic(page, KM_IRQ0);
+
+		/* do the actual data transfer */
+		ap->ops->data_xfer(qc->dev, buf + offset, qc->sect_size, do_write);
+
+		kunmap_atomic(buf, KM_IRQ0);
+		local_irq_restore(flags);
+	} else {
+		buf = page_address(page);
+		ap->ops->data_xfer(qc->dev, buf + offset, qc->sect_size, do_write);
+	}
+
+	qc->curbytes += qc->sect_size;
+	qc->cursg_ofs += qc->sect_size;
+
+	if (qc->cursg_ofs == qc->cursg->length) {
+		qc->cursg = sg_next(qc->cursg);
+		qc->cursg_ofs = 0;
+	}
+}
+
+/**
+ *	ata_pio_sectors - Transfer one or many sectors.
+ *	@qc: Command on going
+ *
+ *	Transfer one or many sectors of data from/to the
+ *	ATA device for the DRQ request.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+static void ata_pio_sectors(struct ata_queued_cmd *qc)
+{
+	if (is_multi_taskfile(&qc->tf)) {
+		/* READ/WRITE MULTIPLE */
+		unsigned int nsect;
+
+		WARN_ON(qc->dev->multi_count == 0);
+
+		nsect = min((qc->nbytes - qc->curbytes) / qc->sect_size,
+			    qc->dev->multi_count);
+		while (nsect--)
+			ata_pio_sector(qc);
+	} else
+		ata_pio_sector(qc);
+
+	ata_altstatus(qc->ap); /* flush */
+}
+
+/**
+ *	atapi_send_cdb - Write CDB bytes to hardware
+ *	@ap: Port to which ATAPI device is attached.
+ *	@qc: Taskfile currently active
+ *
+ *	When device has indicated its readiness to accept
+ *	a CDB, this function is called.  Send the CDB.
+ *
+ *	LOCKING:
+ *	caller.
+ */
+static void atapi_send_cdb(struct ata_port *ap, struct ata_queued_cmd *qc)
+{
+	/* send SCSI cdb */
+	DPRINTK("send cdb\n");
+	WARN_ON(qc->dev->cdb_len < 12);
+
+	ap->ops->data_xfer(qc->dev, qc->cdb, qc->dev->cdb_len, 1);
+	ata_altstatus(ap); /* flush */
+
+	switch (qc->tf.protocol) {
+	case ATAPI_PROT_PIO:
+		ap->hsm_task_state = HSM_ST;
+		break;
+	case ATAPI_PROT_NODATA:
+		ap->hsm_task_state = HSM_ST_LAST;
+		break;
+	case ATAPI_PROT_DMA:
+		ap->hsm_task_state = HSM_ST_LAST;
+		/* initiate bmdma */
+		ap->ops->bmdma_start(qc);
+		break;
+	}
+}
+
+/**
+ *	__atapi_pio_bytes - Transfer data from/to the ATAPI device.
+ *	@qc: Command on going
+ *	@bytes: number of bytes
+ *
+ *	Transfer Transfer data from/to the ATAPI device.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ *
+ */
+static int __atapi_pio_bytes(struct ata_queued_cmd *qc, unsigned int bytes)
+{
+	int rw = (qc->tf.flags & ATA_TFLAG_WRITE) ? WRITE : READ;
+	struct ata_port *ap = qc->ap;
+	struct ata_device *dev = qc->dev;
+	struct ata_eh_info *ehi = &dev->link->eh_info;
+	struct scatterlist *sg;
+	struct page *page;
+	unsigned char *buf;
+	unsigned int offset, count, consumed;
+
+next_sg:
+	sg = qc->cursg;
+	if (unlikely(!sg)) {
+		ata_ehi_push_desc(ehi, "unexpected or too much trailing data "
+				  "buf=%u cur=%u bytes=%u",
+				  qc->nbytes, qc->curbytes, bytes);
+		return -1;
+	}
+
+	page = sg_page(sg);
+	offset = sg->offset + qc->cursg_ofs;
+
+	/* get the current page and offset */
+	page = nth_page(page, (offset >> PAGE_SHIFT));
+	offset %= PAGE_SIZE;
+
+	/* don't overrun current sg */
+	count = min(sg->length - qc->cursg_ofs, bytes);
+
+	/* don't cross page boundaries */
+	count = min(count, (unsigned int)PAGE_SIZE - offset);
+
+	DPRINTK("data %s\n", qc->tf.flags & ATA_TFLAG_WRITE ? "write" : "read");
+
+	if (PageHighMem(page)) {
+		unsigned long flags;
+
+		/* FIXME: use bounce buffer */
+		local_irq_save(flags);
+		buf = kmap_atomic(page, KM_IRQ0);
+
+		/* do the actual data transfer */
+		consumed = ap->ops->data_xfer(dev,  buf + offset, count, rw);
+
+		kunmap_atomic(buf, KM_IRQ0);
+		local_irq_restore(flags);
+	} else {
+		buf = page_address(page);
+		consumed = ap->ops->data_xfer(dev,  buf + offset, count, rw);
+	}
+
+	bytes -= min(bytes, consumed);
+	qc->curbytes += count;
+	qc->cursg_ofs += count;
+
+	if (qc->cursg_ofs == sg->length) {
+		qc->cursg = sg_next(qc->cursg);
+		qc->cursg_ofs = 0;
+	}
+
+	/* consumed can be larger than count only for the last transfer */
+	WARN_ON(qc->cursg && count != consumed);
+
+	if (bytes)
+		goto next_sg;
+	return 0;
+}
+
+/**
+ *	atapi_pio_bytes - Transfer data from/to the ATAPI device.
+ *	@qc: Command on going
+ *
+ *	Transfer Transfer data from/to the ATAPI device.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+static void atapi_pio_bytes(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *dev = qc->dev;
+	struct ata_eh_info *ehi = &dev->link->eh_info;
+	unsigned int ireason, bc_lo, bc_hi, bytes;
+	int i_write, do_write = (qc->tf.flags & ATA_TFLAG_WRITE) ? 1 : 0;
+
+	/* Abuse qc->result_tf for temp storage of intermediate TF
+	 * here to save some kernel stack usage.
+	 * For normal completion, qc->result_tf is not relevant. For
+	 * error, qc->result_tf is later overwritten by ata_qc_complete().
+	 * So, the correctness of qc->result_tf is not affected.
+	 */
+	ap->ops->tf_read(ap, &qc->result_tf);
+	ireason = qc->result_tf.nsect;
+	bc_lo = qc->result_tf.lbam;
+	bc_hi = qc->result_tf.lbah;
+	bytes = (bc_hi << 8) | bc_lo;
+
+	/* shall be cleared to zero, indicating xfer of data */
+	if (unlikely(ireason & (1 << 0)))
+		goto atapi_check;
+
+	/* make sure transfer direction matches expected */
+	i_write = ((ireason & (1 << 1)) == 0) ? 1 : 0;
+	if (unlikely(do_write != i_write))
+		goto atapi_check;
+
+	if (unlikely(!bytes))
+		goto atapi_check;
+
+	VPRINTK("ata%u: xfering %d bytes\n", ap->print_id, bytes);
+
+	if (unlikely(__atapi_pio_bytes(qc, bytes)))
+		goto err_out;
+	ata_altstatus(ap); /* flush */
+
+	return;
+
+ atapi_check:
+	ata_ehi_push_desc(ehi, "ATAPI check failed (ireason=0x%x bytes=%u)",
+			  ireason, bytes);
+ err_out:
+	qc->err_mask |= AC_ERR_HSM;
+	ap->hsm_task_state = HSM_ST_ERR;
+}
+
+/**
+ *	ata_hsm_ok_in_wq - Check if the qc can be handled in the workqueue.
+ *	@ap: the target ata_port
+ *	@qc: qc on going
+ *
+ *	RETURNS:
+ *	1 if ok in workqueue, 0 otherwise.
+ */
+static inline int ata_hsm_ok_in_wq(struct ata_port *ap, struct ata_queued_cmd *qc)
+{
+	if (qc->tf.flags & ATA_TFLAG_POLLING)
+		return 1;
+
+	if (ap->hsm_task_state == HSM_ST_FIRST) {
+		if (qc->tf.protocol == ATA_PROT_PIO &&
+		    (qc->tf.flags & ATA_TFLAG_WRITE))
+		    return 1;
+
+		if (ata_is_atapi(qc->tf.protocol) &&
+		    !(qc->dev->flags & ATA_DFLAG_CDB_INTR))
+			return 1;
+	}
+
+	return 0;
+}
+
+/**
+ *	ata_hsm_qc_complete - finish a qc running on standard HSM
+ *	@qc: Command to complete
+ *	@in_wq: 1 if called from workqueue, 0 otherwise
+ *
+ *	Finish @qc which is running on standard HSM.
+ *
+ *	LOCKING:
+ *	If @in_wq is zero, spin_lock_irqsave(host lock).
+ *	Otherwise, none on entry and grabs host lock.
+ */
+static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
+{
+	struct ata_port *ap = qc->ap;
+	unsigned long flags;
+
+	if (ap->ops->error_handler) {
+		if (in_wq) {
+			spin_lock_irqsave(ap->lock, flags);
+
+			/* EH might have kicked in while host lock is
+			 * released.
+			 */
+			qc = ata_qc_from_tag(ap, qc->tag);
+			if (qc) {
+				if (likely(!(qc->err_mask & AC_ERR_HSM))) {
+					ap->ops->irq_on(ap);
+					ata_qc_complete(qc);
+				} else
+					ata_port_freeze(ap);
+			}
+
+			spin_unlock_irqrestore(ap->lock, flags);
+		} else {
+			if (likely(!(qc->err_mask & AC_ERR_HSM)))
+				ata_qc_complete(qc);
+			else
+				ata_port_freeze(ap);
+		}
+	} else {
+		if (in_wq) {
+			spin_lock_irqsave(ap->lock, flags);
+			ap->ops->irq_on(ap);
+			ata_qc_complete(qc);
+			spin_unlock_irqrestore(ap->lock, flags);
+		} else
+			ata_qc_complete(qc);
+	}
+}
+
+/**
+ *	ata_hsm_move - move the HSM to the next state.
+ *	@ap: the target ata_port
+ *	@qc: qc on going
+ *	@status: current device status
+ *	@in_wq: 1 if called from workqueue, 0 otherwise
+ *
+ *	RETURNS:
+ *	1 when poll next status needed, 0 otherwise.
+ */
+int ata_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
+		 u8 status, int in_wq)
+{
+	unsigned long flags = 0;
+	int poll_next;
+
+	WARN_ON((qc->flags & ATA_QCFLAG_ACTIVE) == 0);
+
+	/* Make sure ata_qc_issue_prot() does not throw things
+	 * like DMA polling into the workqueue. Notice that
+	 * in_wq is not equivalent to (qc->tf.flags & ATA_TFLAG_POLLING).
+	 */
+	WARN_ON(in_wq != ata_hsm_ok_in_wq(ap, qc));
+
+fsm_start:
+	DPRINTK("ata%u: protocol %d task_state %d (dev_stat 0x%X)\n",
+		ap->print_id, qc->tf.protocol, ap->hsm_task_state, status);
+
+	switch (ap->hsm_task_state) {
+	case HSM_ST_FIRST:
+		/* Send first data block or PACKET CDB */
+
+		/* If polling, we will stay in the work queue after
+		 * sending the data. Otherwise, interrupt handler
+		 * takes over after sending the data.
+		 */
+		poll_next = (qc->tf.flags & ATA_TFLAG_POLLING);
+
+		/* check device status */
+		if (unlikely((status & ATA_DRQ) == 0)) {
+			/* handle BSY=0, DRQ=0 as error */
+			if (likely(status & (ATA_ERR | ATA_DF)))
+				/* device stops HSM for abort/error */
+				qc->err_mask |= AC_ERR_DEV;
+			else
+				/* HSM violation. Let EH handle this */
+				qc->err_mask |= AC_ERR_HSM;
+
+			ap->hsm_task_state = HSM_ST_ERR;
+			goto fsm_start;
+		}
+
+		/* Device should not ask for data transfer (DRQ=1)
+		 * when it finds something wrong.
+		 * We ignore DRQ here and stop the HSM by
+		 * changing hsm_task_state to HSM_ST_ERR and
+		 * let the EH abort the command or reset the device.
+		 */
+		if (unlikely(status & (ATA_ERR | ATA_DF))) {
+			/* Some ATAPI tape drives forget to clear the ERR bit
+			 * when doing the next command (mostly request sense).
+			 * We ignore ERR here to workaround and proceed sending
+			 * the CDB.
+			 */
+			if (!(qc->dev->horkage & ATA_HORKAGE_STUCK_ERR)) {
+				ata_port_printk(ap, KERN_WARNING,
+						"DRQ=1 with device error, "
+						"dev_stat 0x%X\n", status);
+				qc->err_mask |= AC_ERR_HSM;
+				ap->hsm_task_state = HSM_ST_ERR;
+				goto fsm_start;
+			}
+		}
+
+		/* Send the CDB (atapi) or the first data block (ata pio out).
+		 * During the state transition, interrupt handler shouldn't
+		 * be invoked before the data transfer is complete and
+		 * hsm_task_state is changed. Hence, the following locking.
+		 */
+		if (in_wq)
+			spin_lock_irqsave(ap->lock, flags);
+
+		if (qc->tf.protocol == ATA_PROT_PIO) {
+			/* PIO data out protocol.
+			 * send first data block.
+			 */
+
+			/* ata_pio_sectors() might change the state
+			 * to HSM_ST_LAST. so, the state is changed here
+			 * before ata_pio_sectors().
+			 */
+			ap->hsm_task_state = HSM_ST;
+			ata_pio_sectors(qc);
+		} else
+			/* send CDB */
+			atapi_send_cdb(ap, qc);
+
+		if (in_wq)
+			spin_unlock_irqrestore(ap->lock, flags);
+
+		/* if polling, ata_pio_task() handles the rest.
+		 * otherwise, interrupt handler takes over from here.
+		 */
+		break;
+
+	case HSM_ST:
+		/* complete command or read/write the data register */
+		if (qc->tf.protocol == ATAPI_PROT_PIO) {
+			/* ATAPI PIO protocol */
+			if ((status & ATA_DRQ) == 0) {
+				/* No more data to transfer or device error.
+				 * Device error will be tagged in HSM_ST_LAST.
+				 */
+				ap->hsm_task_state = HSM_ST_LAST;
+				goto fsm_start;
+			}
+
+			/* Device should not ask for data transfer (DRQ=1)
+			 * when it finds something wrong.
+			 * We ignore DRQ here and stop the HSM by
+			 * changing hsm_task_state to HSM_ST_ERR and
+			 * let the EH abort the command or reset the device.
+			 */
+			if (unlikely(status & (ATA_ERR | ATA_DF))) {
+				ata_port_printk(ap, KERN_WARNING, "DRQ=1 with "
+						"device error, dev_stat 0x%X\n",
+						status);
+				qc->err_mask |= AC_ERR_HSM;
+				ap->hsm_task_state = HSM_ST_ERR;
+				goto fsm_start;
+			}
+
+			atapi_pio_bytes(qc);
+
+			if (unlikely(ap->hsm_task_state == HSM_ST_ERR))
+				/* bad ireason reported by device */
+				goto fsm_start;
+
+		} else {
+			/* ATA PIO protocol */
+			if (unlikely((status & ATA_DRQ) == 0)) {
+				/* handle BSY=0, DRQ=0 as error */
+				if (likely(status & (ATA_ERR | ATA_DF)))
+					/* device stops HSM for abort/error */
+					qc->err_mask |= AC_ERR_DEV;
+				else
+					/* HSM violation. Let EH handle this.
+					 * Phantom devices also trigger this
+					 * condition.  Mark hint.
+					 */
+					qc->err_mask |= AC_ERR_HSM |
+							AC_ERR_NODEV_HINT;
+
+				ap->hsm_task_state = HSM_ST_ERR;
+				goto fsm_start;
+			}
+
+			/* For PIO reads, some devices may ask for
+			 * data transfer (DRQ=1) alone with ERR=1.
+			 * We respect DRQ here and transfer one
+			 * block of junk data before changing the
+			 * hsm_task_state to HSM_ST_ERR.
+			 *
+			 * For PIO writes, ERR=1 DRQ=1 doesn't make
+			 * sense since the data block has been
+			 * transferred to the device.
+			 */
+			if (unlikely(status & (ATA_ERR | ATA_DF))) {
+				/* data might be corrputed */
+				qc->err_mask |= AC_ERR_DEV;
+
+				if (!(qc->tf.flags & ATA_TFLAG_WRITE)) {
+					ata_pio_sectors(qc);
+					status = ata_wait_idle(ap);
+				}
+
+				if (status & (ATA_BUSY | ATA_DRQ))
+					qc->err_mask |= AC_ERR_HSM;
+
+				/* ata_pio_sectors() might change the
+				 * state to HSM_ST_LAST. so, the state
+				 * is changed after ata_pio_sectors().
+				 */
+				ap->hsm_task_state = HSM_ST_ERR;
+				goto fsm_start;
+			}
+
+			ata_pio_sectors(qc);
+
+			if (ap->hsm_task_state == HSM_ST_LAST &&
+			    (!(qc->tf.flags & ATA_TFLAG_WRITE))) {
+				/* all data read */
+				status = ata_wait_idle(ap);
+				goto fsm_start;
+			}
+		}
+
+		poll_next = 1;
+		break;
+
+	case HSM_ST_LAST:
+		if (unlikely(!ata_ok(status))) {
+			qc->err_mask |= __ac_err_mask(status);
+			ap->hsm_task_state = HSM_ST_ERR;
+			goto fsm_start;
+		}
+
+		/* no more data to transfer */
+		DPRINTK("ata%u: dev %u command complete, drv_stat 0x%x\n",
+			ap->print_id, qc->dev->devno, status);
+
+		WARN_ON(qc->err_mask);
+
+		ap->hsm_task_state = HSM_ST_IDLE;
+
+		/* complete taskfile transaction */
+		ata_hsm_qc_complete(qc, in_wq);
+
+		poll_next = 0;
+		break;
+
+	case HSM_ST_ERR:
+		/* make sure qc->err_mask is available to
+		 * know what's wrong and recover
+		 */
+		WARN_ON(qc->err_mask == 0);
+
+		ap->hsm_task_state = HSM_ST_IDLE;
+
+		/* complete taskfile transaction */
+		ata_hsm_qc_complete(qc, in_wq);
+
+		poll_next = 0;
+		break;
+	default:
+		poll_next = 0;
+		BUG();
+	}
+
+	return poll_next;
+}
+
+void ata_pio_task(struct work_struct *work)
+{
+	struct ata_port *ap =
+		container_of(work, struct ata_port, port_task.work);
+	struct ata_queued_cmd *qc = ap->port_task_data;
+	u8 status;
+	int poll_next;
+
+fsm_start:
+	WARN_ON(ap->hsm_task_state == HSM_ST_IDLE);
+
+	/*
+	 * This is purely heuristic.  This is a fast path.
+	 * Sometimes when we enter, BSY will be cleared in
+	 * a chk-status or two.  If not, the drive is probably seeking
+	 * or something.  Snooze for a couple msecs, then
+	 * chk-status again.  If still busy, queue delayed work.
+	 */
+	status = ata_busy_wait(ap, ATA_BUSY, 5);
+	if (status & ATA_BUSY) {
+		msleep(2);
+		status = ata_busy_wait(ap, ATA_BUSY, 10);
+		if (status & ATA_BUSY) {
+			ata_pio_queue_task(ap, qc, ATA_SHORT_PAUSE);
+			return;
+		}
+	}
+
+	/* move the HSM */
+	poll_next = ata_hsm_move(ap, qc, status, 1);
+
+	/* another command or interrupt handler
+	 * may be running at this point.
+	 */
+	if (poll_next)
+		goto fsm_start;
+}
+
+/**
+ *	ata_qc_issue_prot - issue taskfile to device in proto-dependent manner
+ *	@qc: command to issue to device
+ *
+ *	Using various libata functions and hooks, this function
+ *	starts an ATA command.  ATA commands are grouped into
+ *	classes called "protocols", and issuing each type of protocol
+ *	is slightly different.
+ *
+ *	May be used as the qc_issue() entry in ata_port_operations.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ *	RETURNS:
+ *	Zero on success, AC_ERR_* mask on failure
+ */
+unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+
+	/* Use polling pio if the LLD doesn't handle
+	 * interrupt driven pio and atapi CDB interrupt.
+	 */
+	if (ap->flags & ATA_FLAG_PIO_POLLING) {
+		switch (qc->tf.protocol) {
+		case ATA_PROT_PIO:
+		case ATA_PROT_NODATA:
+		case ATAPI_PROT_PIO:
+		case ATAPI_PROT_NODATA:
+			qc->tf.flags |= ATA_TFLAG_POLLING;
+			break;
+		case ATAPI_PROT_DMA:
+			if (qc->dev->flags & ATA_DFLAG_CDB_INTR)
+				/* see ata_dma_blacklisted() */
+				BUG();
+			break;
+		default:
+			break;
+		}
+	}
+
+	/* select the device */
+	ata_dev_select(ap, qc->dev->devno, 1, 0);
+
+	/* start the command */
+	switch (qc->tf.protocol) {
+	case ATA_PROT_NODATA:
+		if (qc->tf.flags & ATA_TFLAG_POLLING)
+			ata_qc_set_polling(qc);
+
+		ata_tf_to_host(ap, &qc->tf);
+		ap->hsm_task_state = HSM_ST_LAST;
+
+		if (qc->tf.flags & ATA_TFLAG_POLLING)
+			ata_pio_queue_task(ap, qc, 0);
+
+		break;
+
+	case ATA_PROT_DMA:
+		WARN_ON(qc->tf.flags & ATA_TFLAG_POLLING);
+
+		ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
+		ap->ops->bmdma_setup(qc);	    /* set up bmdma */
+		ap->ops->bmdma_start(qc);	    /* initiate bmdma */
+		ap->hsm_task_state = HSM_ST_LAST;
+		break;
+
+	case ATA_PROT_PIO:
+		if (qc->tf.flags & ATA_TFLAG_POLLING)
+			ata_qc_set_polling(qc);
+
+		ata_tf_to_host(ap, &qc->tf);
+
+		if (qc->tf.flags & ATA_TFLAG_WRITE) {
+			/* PIO data out protocol */
+			ap->hsm_task_state = HSM_ST_FIRST;
+			ata_pio_queue_task(ap, qc, 0);
+
+			/* always send first data block using
+			 * the ata_pio_task() codepath.
+			 */
+		} else {
+			/* PIO data in protocol */
+			ap->hsm_task_state = HSM_ST;
+
+			if (qc->tf.flags & ATA_TFLAG_POLLING)
+				ata_pio_queue_task(ap, qc, 0);
+
+			/* if polling, ata_pio_task() handles the rest.
+			 * otherwise, interrupt handler takes over from here.
+			 */
+		}
+
+		break;
+
+	case ATAPI_PROT_PIO:
+	case ATAPI_PROT_NODATA:
+		if (qc->tf.flags & ATA_TFLAG_POLLING)
+			ata_qc_set_polling(qc);
+
+		ata_tf_to_host(ap, &qc->tf);
+
+		ap->hsm_task_state = HSM_ST_FIRST;
+
+		/* send cdb by polling if no cdb interrupt */
+		if ((!(qc->dev->flags & ATA_DFLAG_CDB_INTR)) ||
+		    (qc->tf.flags & ATA_TFLAG_POLLING))
+			ata_pio_queue_task(ap, qc, 0);
+		break;
+
+	case ATAPI_PROT_DMA:
+		WARN_ON(qc->tf.flags & ATA_TFLAG_POLLING);
+
+		ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
+		ap->ops->bmdma_setup(qc);	    /* set up bmdma */
+		ap->hsm_task_state = HSM_ST_FIRST;
+
+		/* send cdb by polling if no cdb interrupt */
+		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
+			ata_pio_queue_task(ap, qc, 0);
+		break;
+
+	default:
+		WARN_ON(1);
+		return AC_ERR_SYSTEM;
+	}
+
+	return 0;
+}
+
+/**
+ *	ata_host_intr - Handle host interrupt for given (port, task)
+ *	@ap: Port on which interrupt arrived (possibly...)
+ *	@qc: Taskfile currently active in engine
+ *
+ *	Handle host interrupt for given queued command.  Currently,
+ *	only DMA interrupts are handled.  All other commands are
+ *	handled via polling with interrupts disabled (nIEN bit).
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ *	RETURNS:
+ *	One if interrupt was handled, zero if not (shared irq).
+ */
+inline unsigned int ata_host_intr(struct ata_port *ap,
+				  struct ata_queued_cmd *qc)
+{
+	struct ata_eh_info *ehi = &ap->link.eh_info;
+	u8 status, host_stat = 0;
+
+	VPRINTK("ata%u: protocol %d task_state %d\n",
+		ap->print_id, qc->tf.protocol, ap->hsm_task_state);
+
+	/* Check whether we are expecting interrupt in this state */
+	switch (ap->hsm_task_state) {
+	case HSM_ST_FIRST:
+		/* Some pre-ATAPI-4 devices assert INTRQ
+		 * at this state when ready to receive CDB.
+		 */
+
+		/* Check the ATA_DFLAG_CDB_INTR flag is enough here.
+		 * The flag was turned on only for atapi devices.  No
+		 * need to check ata_is_atapi(qc->tf.protocol) again.
+		 */
+		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
+			goto idle_irq;
+		break;
+	case HSM_ST_LAST:
+		if (qc->tf.protocol == ATA_PROT_DMA ||
+		    qc->tf.protocol == ATAPI_PROT_DMA) {
+			/* check status of DMA engine */
+			host_stat = ap->ops->bmdma_status(ap);
+			VPRINTK("ata%u: host_stat 0x%X\n",
+				ap->print_id, host_stat);
+
+			/* if it's not our irq... */
+			if (!(host_stat & ATA_DMA_INTR))
+				goto idle_irq;
+
+			/* before we do anything else, clear DMA-Start bit */
+			ap->ops->bmdma_stop(qc);
+
+			if (unlikely(host_stat & ATA_DMA_ERR)) {
+				/* error when transfering data to/from memory */
+				qc->err_mask |= AC_ERR_HOST_BUS;
+				ap->hsm_task_state = HSM_ST_ERR;
+			}
+		}
+		break;
+	case HSM_ST:
+		break;
+	default:
+		goto idle_irq;
+	}
+
+	/* check altstatus */
+	status = ata_altstatus(ap);
+	if (status & ATA_BUSY)
+		goto idle_irq;
+
+	/* check main status, clearing INTRQ */
+	status = ata_chk_status(ap);
+	if (unlikely(status & ATA_BUSY))
+		goto idle_irq;
+
+	/* ack bmdma irq events */
+	ap->ops->irq_clear(ap);
+
+	ata_hsm_move(ap, qc, status, 0);
+
+	if (unlikely(qc->err_mask) && (qc->tf.protocol == ATA_PROT_DMA ||
+				       qc->tf.protocol == ATAPI_PROT_DMA))
+		ata_ehi_push_desc(ehi, "BMDMA stat 0x%x", host_stat);
+
+	return 1;	/* irq handled */
+
+idle_irq:
+	ap->stats.idle_irq++;
+
+#ifdef ATA_IRQ_TRAP
+	if ((ap->stats.idle_irq % 1000) == 0) {
+		ata_chk_status(ap);
+		ap->ops->irq_clear(ap);
+		ata_port_printk(ap, KERN_WARNING, "irq trap\n");
+		return 1;
+	}
+#endif
+	return 0;	/* irq not handled */
+}
+
+/**
+ *	ata_interrupt - Default ATA host interrupt handler
+ *	@irq: irq line (unused)
+ *	@dev_instance: pointer to our ata_host information structure
+ *
+ *	Default interrupt handler for PCI IDE devices.  Calls
+ *	ata_host_intr() for each port that is not disabled.
+ *
+ *	LOCKING:
+ *	Obtains host lock during operation.
+ *
+ *	RETURNS:
+ *	IRQ_NONE or IRQ_HANDLED.
+ */
+irqreturn_t ata_interrupt(int irq, void *dev_instance)
+{
+	struct ata_host *host = dev_instance;
+	unsigned int i;
+	unsigned int handled = 0;
+	unsigned long flags;
+
+	/* TODO: make _irqsave conditional on x86 PCI IDE legacy mode */
+	spin_lock_irqsave(&host->lock, flags);
+
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap;
+
+		ap = host->ports[i];
+		if (ap &&
+		    !(ap->flags & ATA_FLAG_DISABLED)) {
+			struct ata_queued_cmd *qc;
+
+			qc = ata_qc_from_tag(ap, ap->link.active_tag);
+			if (qc && (!(qc->tf.flags & ATA_TFLAG_POLLING)) &&
+			    (qc->flags & ATA_QCFLAG_ACTIVE))
+				handled |= ata_host_intr(ap, qc);
+		}
+	}
+
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	return IRQ_RETVAL(handled);
+}
+
+/**
+ *	ata_bmdma_freeze - Freeze BMDMA controller port
+ *	@ap: port to freeze
+ *
+ *	Freeze BMDMA controller port.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+void ata_bmdma_freeze(struct ata_port *ap)
+{
+	struct ata_ioports *ioaddr = &ap->ioaddr;
+
+	ap->ctl |= ATA_NIEN;
+	ap->last_ctl = ap->ctl;
+
+	if (ioaddr->ctl_addr)
+		iowrite8(ap->ctl, ioaddr->ctl_addr);
+
+	/* Under certain circumstances, some controllers raise IRQ on
+	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
+	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
+	 */
+	ata_chk_status(ap);
+
+	ap->ops->irq_clear(ap);
+}
+
+/**
+ *	ata_bmdma_thaw - Thaw BMDMA controller port
+ *	@ap: port to thaw
+ *
+ *	Thaw BMDMA controller port.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+void ata_bmdma_thaw(struct ata_port *ap)
 {
-	if (ap->ops->check_altstatus)
-		return ap->ops->check_altstatus(ap);
-
-	return ioread8(ap->ioaddr.altstatus_addr);
+	/* clear & re-enable interrupts */
+	ata_chk_status(ap);
+	ap->ops->irq_clear(ap);
+	ap->ops->irq_on(ap);
 }
 
 /**
- *	ata_irq_on - Enable interrupts on a port.
- *	@ap: Port on which interrupts are enabled.
+ *	ata_devchk - PATA device presence detection
+ *	@ap: ATA channel to examine
+ *	@device: Device to examine (starting at zero)
  *
- *	Enable interrupts on a legacy IDE device using MMIO or PIO,
- *	wait for idle, clear any pending interrupts.
+ *	This technique was originally described in
+ *	Hale Landis's ATADRVR (www.ata-atapi.com), and
+ *	later found its way into the ATA/ATAPI spec.
+ *
+ *	Write a pattern to the ATA shadow registers,
+ *	and if a device is present, it will respond by
+ *	correctly storing and echoing back the
+ *	ATA shadow register contents.
  *
  *	LOCKING:
- *	Inherited from caller.
+ *	caller.
  */
-u8 ata_irq_on(struct ata_port *ap)
+static unsigned int ata_devchk(struct ata_port *ap, unsigned int device)
 {
 	struct ata_ioports *ioaddr = &ap->ioaddr;
-	u8 tmp;
+	u8 nsect, lbal;
 
-	ap->ctl &= ~ATA_NIEN;
-	ap->last_ctl = ap->ctl;
+	ap->ops->dev_select(ap, device);
 
-	if (ioaddr->ctl_addr)
-		iowrite8(ap->ctl, ioaddr->ctl_addr);
-	tmp = ata_wait_idle(ap);
+	iowrite8(0x55, ioaddr->nsect_addr);
+	iowrite8(0xaa, ioaddr->lbal_addr);
 
-	ap->ops->irq_clear(ap);
+	iowrite8(0xaa, ioaddr->nsect_addr);
+	iowrite8(0x55, ioaddr->lbal_addr);
 
-	return tmp;
+	iowrite8(0x55, ioaddr->nsect_addr);
+	iowrite8(0xaa, ioaddr->lbal_addr);
+
+	nsect = ioread8(ioaddr->nsect_addr);
+	lbal = ioread8(ioaddr->lbal_addr);
+
+	if ((nsect == 0x55) && (lbal == 0xaa))
+		return 1;	/* we found a device */
+
+	return 0;		/* nothing found */
 }
 
 /**
- *	ata_bmdma_irq_clear - Clear PCI IDE BMDMA interrupt.
- *	@ap: Port associated with this ATA transaction.
+ *	ata_dev_try_classify - Parse returned ATA device signature
+ *	@dev: ATA device to classify (starting at zero)
+ *	@present: device seems present
+ *	@r_err: Value of error register on completion
  *
- *	Clear interrupt and error flags in DMA status register.
+ *	After an event -- SRST, E.D.D., or SATA COMRESET -- occurs,
+ *	an ATA/ATAPI-defined set of values is placed in the ATA
+ *	shadow registers, indicating the results of device detection
+ *	and diagnostics.
  *
- *	May be used as the irq_clear() entry in ata_port_operations.
+ *	Select the ATA device, and read the values from the ATA shadow
+ *	registers.  Then parse according to the Error register value,
+ *	and the spec-defined values examined by ata_dev_classify().
  *
  *	LOCKING:
- *	spin_lock_irqsave(host lock)
+ *	caller.
+ *
+ *	RETURNS:
+ *	Device type - %ATA_DEV_ATA, %ATA_DEV_ATAPI or %ATA_DEV_NONE.
  */
-void ata_bmdma_irq_clear(struct ata_port *ap)
+unsigned int ata_dev_try_classify(struct ata_device *dev, int present,
+				  u8 *r_err)
 {
-	void __iomem *mmio = ap->ioaddr.bmdma_addr;
+	struct ata_port *ap = dev->link->ap;
+	struct ata_taskfile tf;
+	unsigned int class;
+	u8 err;
+
+	ap->ops->dev_select(ap, dev->devno);
+
+	memset(&tf, 0, sizeof(tf));
+
+	ap->ops->tf_read(ap, &tf);
+	err = tf.feature;
+	if (r_err)
+		*r_err = err;
+
+	/* see if device passed diags: continue and warn later */
+	if (err == 0)
+		/* diagnostic fail : do nothing _YET_ */
+		dev->horkage |= ATA_HORKAGE_DIAGNOSTIC;
+	else if (err == 1)
+		/* do nothing */ ;
+	else if ((dev->devno == 0) && (err == 0x81))
+		/* do nothing */ ;
+	else
+		return ATA_DEV_NONE;
 
-	if (!mmio)
-		return;
+	/* determine if device is ATA or ATAPI */
+	class = ata_dev_classify(&tf);
 
-	iowrite8(ioread8(mmio + ATA_DMA_STATUS), mmio + ATA_DMA_STATUS);
+	if (class == ATA_DEV_UNKNOWN) {
+		/* If the device failed diagnostic, it's likely to
+		 * have reported incorrect device signature too.
+		 * Assume ATA device if the device seems present but
+		 * device signature is invalid with diagnostic
+		 * failure.
+		 */
+		if (present && (dev->horkage & ATA_HORKAGE_DIAGNOSTIC))
+			class = ATA_DEV_ATA;
+		else
+			class = ATA_DEV_NONE;
+	} else if ((class == ATA_DEV_ATA) && (ata_chk_status(ap) == 0))
+		class = ATA_DEV_NONE;
+
+	return class;
 }
 
-/**
- *	ata_tf_load - send taskfile registers to host controller
- *	@ap: Port to which output is sent
- *	@tf: ATA taskfile register set
- *
- *	Outputs ATA taskfile to standard ATA host controller.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
+static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
+			      unsigned long deadline)
 {
 	struct ata_ioports *ioaddr = &ap->ioaddr;
-	unsigned int is_addr = tf->flags & ATA_TFLAG_ISADDR;
+	unsigned int dev0 = devmask & (1 << 0);
+	unsigned int dev1 = devmask & (1 << 1);
+	int rc, ret = 0;
 
-	if (tf->ctl != ap->last_ctl) {
-		if (ioaddr->ctl_addr)
-			iowrite8(tf->ctl, ioaddr->ctl_addr);
-		ap->last_ctl = tf->ctl;
-		ata_wait_idle(ap);
+	/* if device 0 was found in ata_devchk, wait for its
+	 * BSY bit to clear
+	 */
+	if (dev0) {
+		rc = ata_wait_ready(ap, deadline);
+		if (rc) {
+			if (rc != -ENODEV)
+				return rc;
+			ret = rc;
+		}
 	}
 
-	if (is_addr && (tf->flags & ATA_TFLAG_LBA48)) {
-		WARN_ON(!ioaddr->ctl_addr);
-		iowrite8(tf->hob_feature, ioaddr->feature_addr);
-		iowrite8(tf->hob_nsect, ioaddr->nsect_addr);
-		iowrite8(tf->hob_lbal, ioaddr->lbal_addr);
-		iowrite8(tf->hob_lbam, ioaddr->lbam_addr);
-		iowrite8(tf->hob_lbah, ioaddr->lbah_addr);
-		VPRINTK("hob: feat 0x%X nsect 0x%X, lba 0x%X 0x%X 0x%X\n",
-			tf->hob_feature,
-			tf->hob_nsect,
-			tf->hob_lbal,
-			tf->hob_lbam,
-			tf->hob_lbah);
-	}
+	/* if device 1 was found in ata_devchk, wait for register
+	 * access briefly, then wait for BSY to clear.
+	 */
+	if (dev1) {
+		int i;
 
-	if (is_addr) {
-		iowrite8(tf->feature, ioaddr->feature_addr);
-		iowrite8(tf->nsect, ioaddr->nsect_addr);
-		iowrite8(tf->lbal, ioaddr->lbal_addr);
-		iowrite8(tf->lbam, ioaddr->lbam_addr);
-		iowrite8(tf->lbah, ioaddr->lbah_addr);
-		VPRINTK("feat 0x%X nsect 0x%X lba 0x%X 0x%X 0x%X\n",
-			tf->feature,
-			tf->nsect,
-			tf->lbal,
-			tf->lbam,
-			tf->lbah);
-	}
+		ap->ops->dev_select(ap, 1);
 
-	if (tf->flags & ATA_TFLAG_DEVICE) {
-		iowrite8(tf->device, ioaddr->device_addr);
-		VPRINTK("device 0x%X\n", tf->device);
+		/* Wait for register access.  Some ATAPI devices fail
+		 * to set nsect/lbal after reset, so don't waste too
+		 * much time on it.  We're gonna wait for !BSY anyway.
+		 */
+		for (i = 0; i < 2; i++) {
+			u8 nsect, lbal;
+
+			nsect = ioread8(ioaddr->nsect_addr);
+			lbal = ioread8(ioaddr->lbal_addr);
+			if ((nsect == 1) && (lbal == 1))
+				break;
+			msleep(50);	/* give drive a breather */
+		}
+
+		rc = ata_wait_ready(ap, deadline);
+		if (rc) {
+			if (rc != -ENODEV)
+				return rc;
+			ret = rc;
+		}
 	}
 
-	ata_wait_idle(ap);
+	/* is all this really necessary? */
+	ap->ops->dev_select(ap, 0);
+	if (dev1)
+		ap->ops->dev_select(ap, 1);
+	if (dev0)
+		ap->ops->dev_select(ap, 0);
+
+	return ret;
 }
 
 /**
- *	ata_tf_read - input device's ATA taskfile shadow registers
- *	@ap: Port from which input is read
- *	@tf: ATA taskfile register set for storing input
+ *	ata_wait_after_reset - wait before checking status after reset
+ *	@ap: port containing status register to be polled
+ *	@deadline: deadline jiffies for the operation
  *
- *	Reads ATA taskfile registers for currently-selected device
- *	into @tf. Assumes the device has a fully SFF compliant task file
- *	layout and behaviour. If you device does not (eg has a different
- *	status method) then you will need to provide a replacement tf_read
+ *	After reset, we need to pause a while before reading status.
+ *	Also, certain combination of controller and device report 0xff
+ *	for some duration (e.g. until SATA PHY is up and running)
+ *	which is interpreted as empty port in ATA world.  This
+ *	function also waits for such devices to get out of 0xff
+ *	status.
  *
  *	LOCKING:
- *	Inherited from caller.
+ *	Kernel thread context (may sleep).
  */
-void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
+void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline)
 {
-	struct ata_ioports *ioaddr = &ap->ioaddr;
+	unsigned long until = jiffies + ATA_TMOUT_FF_WAIT;
+
+	if (time_before(until, deadline))
+		deadline = until;
+
+	/* Spec mandates ">= 2ms" before checking status.  We wait
+	 * 150ms, because that was the magic delay used for ATAPI
+	 * devices in Hale Landis's ATADRVR, for the period of time
+	 * between when the ATA command register is written, and then
+	 * status is checked.  Because waiting for "a while" before
+	 * checking status is fine, post SRST, we perform this magic
+	 * delay here as well.
+	 *
+	 * Old drivers/ide uses the 2mS rule and then waits for ready.
+	 */
+	msleep(150);
 
-	tf->command = ata_check_status(ap);
-	tf->feature = ioread8(ioaddr->error_addr);
-	tf->nsect = ioread8(ioaddr->nsect_addr);
-	tf->lbal = ioread8(ioaddr->lbal_addr);
-	tf->lbam = ioread8(ioaddr->lbam_addr);
-	tf->lbah = ioread8(ioaddr->lbah_addr);
-	tf->device = ioread8(ioaddr->device_addr);
+	/* Wait for 0xff to clear.  Some SATA devices take a long time
+	 * to clear 0xff after reset.  For example, HHD424020F7SV00
+	 * iVDR needs >= 800ms while.  Quantum GoVault needs even more
+	 * than that.
+	 *
+	 * Note that some PATA controllers (pata_ali) explode if
+	 * status register is read more than once when there's no
+	 * device attached.
+	 */
+	if (ap->flags & ATA_FLAG_SATA) {
+		while (1) {
+			u8 status = ata_chk_status(ap);
 
-	if (tf->flags & ATA_TFLAG_LBA48) {
-		if (likely(ioaddr->ctl_addr)) {
-			iowrite8(tf->ctl | ATA_HOB, ioaddr->ctl_addr);
-			tf->hob_feature = ioread8(ioaddr->error_addr);
-			tf->hob_nsect = ioread8(ioaddr->nsect_addr);
-			tf->hob_lbal = ioread8(ioaddr->lbal_addr);
-			tf->hob_lbam = ioread8(ioaddr->lbam_addr);
-			tf->hob_lbah = ioread8(ioaddr->lbah_addr);
-			iowrite8(tf->ctl, ioaddr->ctl_addr);
-			ap->last_ctl = tf->ctl;
-		} else
-			WARN_ON(1);
+			if (status != 0xff || time_after(jiffies, deadline))
+				return;
+
+			msleep(50);
+		}
 	}
 }
 
-/**
- *	ata_exec_command - issue ATA command to host controller
- *	@ap: port to which command is being issued
- *	@tf: ATA taskfile register set
- *
- *	Issues ATA command, with proper synchronization with interrupt
- *	handler / other threads.
- *
- *	LOCKING:
- *	spin_lock_irqsave(host lock)
- */
-void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf)
+static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
+			     unsigned long deadline)
 {
-	DPRINTK("ata%u: cmd 0x%X\n", ap->print_id, tf->command);
+	struct ata_ioports *ioaddr = &ap->ioaddr;
 
-	iowrite8(tf->command, ap->ioaddr.command_addr);
-	ata_pause(ap);
+	DPRINTK("ata%u: bus reset via SRST\n", ap->print_id);
+
+	/* software reset.  causes dev0 to be selected */
+	iowrite8(ap->ctl, ioaddr->ctl_addr);
+	udelay(20);	/* FIXME: flush */
+	iowrite8(ap->ctl | ATA_SRST, ioaddr->ctl_addr);
+	udelay(20);	/* FIXME: flush */
+	iowrite8(ap->ctl, ioaddr->ctl_addr);
+
+	/* wait a while before checking status */
+	ata_wait_after_reset(ap, deadline);
+
+	/* Before we perform post reset processing we want to see if
+	 * the bus shows 0xFF because the odd clown forgets the D7
+	 * pulldown resistor.
+	 */
+	if (ata_chk_status(ap) == 0xFF)
+		return -ENODEV;
+
+	return ata_bus_post_reset(ap, devmask, deadline);
 }
 
 /**
- *	ata_bmdma_freeze - Freeze BMDMA controller port
- *	@ap: port to freeze
+ *	ata_std_softreset - reset host port via ATA SRST
+ *	@link: ATA link to reset
+ *	@classes: resulting classes of attached devices
+ *	@deadline: deadline jiffies for the operation
  *
- *	Freeze BMDMA controller port.
+ *	Reset host port using ATA SRST.
  *
  *	LOCKING:
- *	Inherited from caller.
+ *	Kernel thread context (may sleep)
+ *
+ *	RETURNS:
+ *	0 on success, -errno otherwise.
  */
-void ata_bmdma_freeze(struct ata_port *ap)
+int ata_std_softreset(struct ata_link *link, unsigned int *classes,
+		      unsigned long deadline)
 {
-	struct ata_ioports *ioaddr = &ap->ioaddr;
+	struct ata_port *ap = link->ap;
+	unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS;
+	unsigned int devmask = 0;
+	int rc;
+	u8 err;
 
-	ap->ctl |= ATA_NIEN;
-	ap->last_ctl = ap->ctl;
+	DPRINTK("ENTER\n");
 
-	if (ioaddr->ctl_addr)
-		iowrite8(ap->ctl, ioaddr->ctl_addr);
+	if (ata_link_offline(link)) {
+		classes[0] = ATA_DEV_NONE;
+		goto out;
+	}
 
-	/* Under certain circumstances, some controllers raise IRQ on
-	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
-	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
-	 */
-	ata_chk_status(ap);
+	/* determine if device 0/1 are present */
+	if (ata_devchk(ap, 0))
+		devmask |= (1 << 0);
+	if (slave_possible && ata_devchk(ap, 1))
+		devmask |= (1 << 1);
+
+	/* select device 0 again */
+	ap->ops->dev_select(ap, 0);
+
+	/* issue bus reset */
+	DPRINTK("about to softreset, devmask=%x\n", devmask);
+	rc = ata_bus_softreset(ap, devmask, deadline);
+	/* if link is occupied, -ENODEV too is an error */
+	if (rc && (rc != -ENODEV || sata_scr_valid(link))) {
+		ata_link_printk(link, KERN_ERR, "SRST failed (errno=%d)\n", rc);
+		return rc;
+	}
 
-	ap->ops->irq_clear(ap);
+	/* determine by signature whether we have ATA or ATAPI devices */
+	classes[0] = ata_dev_try_classify(&link->device[0],
+					  devmask & (1 << 0), &err);
+	if (slave_possible && err != 0x81)
+		classes[1] = ata_dev_try_classify(&link->device[1],
+						  devmask & (1 << 1), &err);
+
+ out:
+	DPRINTK("EXIT, classes[0]=%u [1]=%u\n", classes[0], classes[1]);
+	return 0;
 }
 
 /**
- *	ata_bmdma_thaw - Thaw BMDMA controller port
- *	@ap: port to thaw
+ *	sata_std_hardreset - reset host port via SATA phy reset
+ *	@link: link to reset
+ *	@class: resulting class of attached device
+ *	@deadline: deadline jiffies for the operation
  *
- *	Thaw BMDMA controller port.
+ *	SATA phy-reset host port using DET bits of SControl register,
+ *	wait for !BSY and classify the attached device.
  *
  *	LOCKING:
- *	Inherited from caller.
+ *	Kernel thread context (may sleep)
+ *
+ *	RETURNS:
+ *	0 on success, -errno otherwise.
  */
-void ata_bmdma_thaw(struct ata_port *ap)
+int sata_std_hardreset(struct ata_link *link, unsigned int *class,
+		       unsigned long deadline)
 {
-	/* clear & re-enable interrupts */
-	ata_chk_status(ap);
-	ap->ops->irq_clear(ap);
-	ap->ops->irq_on(ap);
+	struct ata_port *ap = link->ap;
+	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
+	int rc;
+
+	DPRINTK("ENTER\n");
+
+	/* do hardreset */
+	rc = sata_link_hardreset(link, timing, deadline);
+	if (rc) {
+		ata_link_printk(link, KERN_ERR,
+				"COMRESET failed (errno=%d)\n", rc);
+		return rc;
+	}
+
+	/* TODO: phy layer with polling, timeouts, etc. */
+	if (ata_link_offline(link)) {
+		*class = ATA_DEV_NONE;
+		DPRINTK("EXIT, link offline\n");
+		return 0;
+	}
+
+	/* wait a while before checking status */
+	ata_wait_after_reset(ap, deadline);
+
+	/* If PMP is supported, we have to do follow-up SRST.  Note
+	 * that some PMPs don't send D2H Reg FIS after hardreset at
+	 * all if the first port is empty.  Wait for it just for a
+	 * second and request follow-up SRST.
+	 */
+	if (ap->flags & ATA_FLAG_PMP) {
+		ata_wait_ready(ap, jiffies + HZ);
+		return -EAGAIN;
+	}
+
+	rc = ata_wait_ready(ap, deadline);
+	/* link occupied, -ENODEV too is an error */
+	if (rc) {
+		ata_link_printk(link, KERN_ERR,
+				"COMRESET failed (errno=%d)\n", rc);
+		return rc;
+	}
+
+	ap->ops->dev_select(ap, 0);	/* probably unnecessary */
+
+	*class = ata_dev_try_classify(link->device, 1, NULL);
+
+	DPRINTK("EXIT, class=%u\n", *class);
+	return 0;
 }
 
 /**
@@ -393,6 +2092,31 @@ int ata_sff_port_start(struct ata_port *ap)
 	return 0;
 }
 
+/**
+ *	ata_std_ports - initialize ioaddr with standard port offsets.
+ *	@ioaddr: IO address structure to be initialized
+ *
+ *	Utility function which initializes data_addr, error_addr,
+ *	feature_addr, nsect_addr, lbal_addr, lbam_addr, lbah_addr,
+ *	device_addr, status_addr, and command_addr to standard offsets
+ *	relative to cmd_addr.
+ *
+ *	Does not set ctl_addr, altstatus_addr, bmdma_addr, or scr_addr.
+ */
+void ata_std_ports(struct ata_ioports *ioaddr)
+{
+	ioaddr->data_addr = ioaddr->cmd_addr + ATA_REG_DATA;
+	ioaddr->error_addr = ioaddr->cmd_addr + ATA_REG_ERR;
+	ioaddr->feature_addr = ioaddr->cmd_addr + ATA_REG_FEATURE;
+	ioaddr->nsect_addr = ioaddr->cmd_addr + ATA_REG_NSECT;
+	ioaddr->lbal_addr = ioaddr->cmd_addr + ATA_REG_LBAL;
+	ioaddr->lbam_addr = ioaddr->cmd_addr + ATA_REG_LBAM;
+	ioaddr->lbah_addr = ioaddr->cmd_addr + ATA_REG_LBAH;
+	ioaddr->device_addr = ioaddr->cmd_addr + ATA_REG_DEVICE;
+	ioaddr->status_addr = ioaddr->cmd_addr + ATA_REG_STATUS;
+	ioaddr->command_addr = ioaddr->cmd_addr + ATA_REG_CMD;
+}
+
 /**
  *	ata_bmdma_setup - Set up PCI IDE BMDMA transaction
  *	@qc: Info associated with this ATA transaction.
@@ -494,11 +2218,94 @@ u8 ata_bmdma_status(struct ata_port *ap)
 }
 
 /**
- *	ata_noop_irq_clear - Noop placeholder for irq_clear
- *	@ap: Port associated with this ATA transaction.
+ *	ata_bus_reset - reset host port and associated ATA channel
+ *	@ap: port to reset
+ *
+ *	This is typically the first time we actually start issuing
+ *	commands to the ATA channel.  We wait for BSY to clear, then
+ *	issue EXECUTE DEVICE DIAGNOSTIC command, polling for its
+ *	result.  Determine what devices, if any, are on the channel
+ *	by looking at the device 0/1 error register.  Look at the signature
+ *	stored in each device's taskfile registers, to determine if
+ *	the device is ATA or ATAPI.
+ *
+ *	LOCKING:
+ *	PCI/etc. bus probe sem.
+ *	Obtains host lock.
+ *
+ *	SIDE EFFECTS:
+ *	Sets ATA_FLAG_DISABLED if bus reset fails.
+ *
+ *	DEPRECATED:
+ *	This function is only for drivers which still use old EH and
+ *	will be removed soon.
  */
-void ata_noop_irq_clear(struct ata_port *ap)
+void ata_bus_reset(struct ata_port *ap)
 {
+	struct ata_device *device = ap->link.device;
+	struct ata_ioports *ioaddr = &ap->ioaddr;
+	unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS;
+	u8 err;
+	unsigned int dev0, dev1 = 0, devmask = 0;
+	int rc;
+
+	DPRINTK("ENTER, host %u, port %u\n", ap->print_id, ap->port_no);
+
+	/* determine if device 0/1 are present */
+	if (ap->flags & ATA_FLAG_SATA_RESET)
+		dev0 = 1;
+	else {
+		dev0 = ata_devchk(ap, 0);
+		if (slave_possible)
+			dev1 = ata_devchk(ap, 1);
+	}
+
+	if (dev0)
+		devmask |= (1 << 0);
+	if (dev1)
+		devmask |= (1 << 1);
+
+	/* select device 0 again */
+	ap->ops->dev_select(ap, 0);
+
+	/* issue bus reset */
+	if (ap->flags & ATA_FLAG_SRST) {
+		rc = ata_bus_softreset(ap, devmask, jiffies + 40 * HZ);
+		if (rc && rc != -ENODEV)
+			goto err_out;
+	}
+
+	/*
+	 * determine by signature whether we have ATA or ATAPI devices
+	 */
+	device[0].class = ata_dev_try_classify(&device[0], dev0, &err);
+	if ((slave_possible) && (err != 0x81))
+		device[1].class = ata_dev_try_classify(&device[1], dev1, &err);
+
+	/* is double-select really necessary? */
+	if (device[1].class != ATA_DEV_NONE)
+		ap->ops->dev_select(ap, 1);
+	if (device[0].class != ATA_DEV_NONE)
+		ap->ops->dev_select(ap, 0);
+
+	/* if no devices were detected, disable this port */
+	if ((device[0].class == ATA_DEV_NONE) &&
+	    (device[1].class == ATA_DEV_NONE))
+		goto err_out;
+
+	if (ap->flags & (ATA_FLAG_SATA_RESET | ATA_FLAG_SRST)) {
+		/* set up device control for ATA_FLAG_SATA_RESET */
+		iowrite8(ap->ctl, ioaddr->ctl_addr);
+	}
+
+	DPRINTK("EXIT\n");
+	return;
+
+err_out:
+	ata_port_printk(ap, KERN_ERR, "disabling port\n");
+	ata_port_disable(ap);
+
+	DPRINTK("EXIT\n");
 }
 
 #ifdef CONFIG_PCI
@@ -914,3 +2721,49 @@ int ata_pci_init_one(struct pci_dev *pdev,
 
 #endif /* CONFIG_PCI */
 
+EXPORT_SYMBOL_GPL(ata_sff_port_ops);
+EXPORT_SYMBOL_GPL(ata_bmdma_port_ops);
+EXPORT_SYMBOL_GPL(ata_qc_prep);
+EXPORT_SYMBOL_GPL(ata_dumb_qc_prep);
+EXPORT_SYMBOL_GPL(ata_std_dev_select);
+EXPORT_SYMBOL_GPL(ata_check_status);
+EXPORT_SYMBOL_GPL(ata_altstatus);
+EXPORT_SYMBOL_GPL(ata_busy_sleep);
+EXPORT_SYMBOL_GPL(ata_wait_ready);
+EXPORT_SYMBOL_GPL(ata_tf_load);
+EXPORT_SYMBOL_GPL(ata_tf_read);
+EXPORT_SYMBOL_GPL(ata_exec_command);
+EXPORT_SYMBOL_GPL(ata_data_xfer);
+EXPORT_SYMBOL_GPL(ata_data_xfer_noirq);
+EXPORT_SYMBOL_GPL(ata_irq_on);
+EXPORT_SYMBOL_GPL(ata_bmdma_irq_clear);
+EXPORT_SYMBOL_GPL(ata_hsm_move);
+EXPORT_SYMBOL_GPL(ata_qc_issue_prot);
+EXPORT_SYMBOL_GPL(ata_host_intr);
+EXPORT_SYMBOL_GPL(ata_interrupt);
+EXPORT_SYMBOL_GPL(ata_bmdma_freeze);
+EXPORT_SYMBOL_GPL(ata_bmdma_thaw);
+EXPORT_SYMBOL_GPL(ata_std_prereset);
+EXPORT_SYMBOL_GPL(ata_dev_try_classify);
+EXPORT_SYMBOL_GPL(ata_wait_after_reset);
+EXPORT_SYMBOL_GPL(ata_std_softreset);
+EXPORT_SYMBOL_GPL(sata_std_hardreset);
+EXPORT_SYMBOL_GPL(ata_std_postreset);
+EXPORT_SYMBOL_GPL(ata_bmdma_error_handler);
+EXPORT_SYMBOL_GPL(ata_bmdma_post_internal_cmd);
+EXPORT_SYMBOL_GPL(ata_sff_port_start);
+EXPORT_SYMBOL_GPL(ata_std_ports);
+EXPORT_SYMBOL_GPL(ata_bmdma_setup);
+EXPORT_SYMBOL_GPL(ata_bmdma_start);
+EXPORT_SYMBOL_GPL(ata_bmdma_stop);
+EXPORT_SYMBOL_GPL(ata_bmdma_status);
+EXPORT_SYMBOL_GPL(ata_bus_reset);
+#ifdef CONFIG_PCI
+EXPORT_SYMBOL_GPL(ata_pci_clear_simplex);
+EXPORT_SYMBOL_GPL(ata_pci_default_filter);
+EXPORT_SYMBOL_GPL(ata_pci_init_bmdma);
+EXPORT_SYMBOL_GPL(ata_pci_init_sff_host);
+EXPORT_SYMBOL_GPL(ata_pci_prepare_sff_host);
+EXPORT_SYMBOL_GPL(ata_pci_activate_sff_host);
+EXPORT_SYMBOL_GPL(ata_pci_init_one);
+#endif /* CONFIG_PCI */
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index aa884f71a12a..a69f663c7402 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -67,6 +67,8 @@ extern int ata_build_rw_tf(struct ata_taskfile *tf, struct ata_device *dev,
 			   unsigned int tag);
 extern u64 ata_tf_read_block(struct ata_taskfile *tf, struct ata_device *dev);
 extern void ata_dev_disable(struct ata_device *dev);
+extern void ata_pio_queue_task(struct ata_port *ap, void *data,
+			       unsigned long delay);
 extern void ata_port_flush_task(struct ata_port *ap);
 extern unsigned ata_exec_internal(struct ata_device *dev,
 				  struct ata_taskfile *tf, const u8 *cdb,
@@ -91,8 +93,6 @@ extern void ata_qc_free(struct ata_queued_cmd *qc);
 extern void ata_qc_issue(struct ata_queued_cmd *qc);
 extern void __ata_qc_complete(struct ata_queued_cmd *qc);
 extern int ata_check_atapi_dma(struct ata_queued_cmd *qc);
-extern void ata_dev_select(struct ata_port *ap, unsigned int device,
-                           unsigned int wait, unsigned int can_sleep);
 extern void swap_buf_le16(u16 *buf, unsigned int buf_words);
 extern int ata_flush_cache(struct ata_device *dev);
 extern void ata_dev_init(struct ata_device *dev);
@@ -194,7 +194,9 @@ extern int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 extern void ata_eh_finish(struct ata_port *ap);
 
 /* libata-sff.c */
+extern void ata_dev_select(struct ata_port *ap, unsigned int device,
+                           unsigned int wait, unsigned int can_sleep);
 extern u8 ata_irq_on(struct ata_port *ap);
-
+extern void ata_pio_task(struct work_struct *work);
 
 #endif /* __LIBATA_H__ */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 01c233303aee..673f34b256ba 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -349,6 +349,22 @@ enum {
 	ATAPI_READ_CD		= 2,		/* READ CD [MSF] */
 	ATAPI_PASS_THRU		= 3,		/* SAT pass-thru */
 	ATAPI_MISC		= 4,		/* the rest */
+
+	/* Timing constants */
+	ATA_TIMING_SETUP	= (1 << 0),
+	ATA_TIMING_ACT8B	= (1 << 1),
+	ATA_TIMING_REC8B	= (1 << 2),
+	ATA_TIMING_CYC8B	= (1 << 3),
+	ATA_TIMING_8BIT		= ATA_TIMING_ACT8B | ATA_TIMING_REC8B |
+				  ATA_TIMING_CYC8B,
+	ATA_TIMING_ACTIVE	= (1 << 4),
+	ATA_TIMING_RECOVER	= (1 << 5),
+	ATA_TIMING_CYCLE	= (1 << 6),
+	ATA_TIMING_UDMA		= (1 << 7),
+	ATA_TIMING_ALL		= ATA_TIMING_SETUP | ATA_TIMING_ACT8B |
+				  ATA_TIMING_REC8B | ATA_TIMING_CYC8B |
+				  ATA_TIMING_ACTIVE | ATA_TIMING_RECOVER |
+				  ATA_TIMING_CYCLE | ATA_TIMING_UDMA,
 };
 
 enum ata_xfer_mask {
@@ -779,6 +795,9 @@ struct ata_timing {
 
 #define FIT(v, vmin, vmax)	max_t(short, min_t(short, v, vmax), vmin)
 
+/*
+ * Core layer - drivers/ata/libata-core.c
+ */
 extern const unsigned long sata_deb_timing_normal[];
 extern const unsigned long sata_deb_timing_hotplug[];
 extern const unsigned long sata_deb_timing_long[];
@@ -802,22 +821,14 @@ static inline int ata_port_is_dummy(struct ata_port *ap)
 
 extern void sata_print_link_status(struct ata_link *link);
 extern void ata_port_probe(struct ata_port *);
-extern void ata_bus_reset(struct ata_port *ap);
 extern int sata_set_spd(struct ata_link *link);
 extern int sata_link_debounce(struct ata_link *link,
 			const unsigned long *params, unsigned long deadline);
 extern int sata_link_resume(struct ata_link *link, const unsigned long *params,
 			    unsigned long deadline);
-extern int ata_std_prereset(struct ata_link *link, unsigned long deadline);
-extern int ata_std_softreset(struct ata_link *link, unsigned int *classes,
-			     unsigned long deadline);
 extern int sata_link_hardreset(struct ata_link *link,
 			const unsigned long *timing, unsigned long deadline);
-extern int sata_std_hardreset(struct ata_link *link, unsigned int *class,
-			      unsigned long deadline);
-extern void ata_std_postreset(struct ata_link *link, unsigned int *classes);
 extern void ata_port_disable(struct ata_port *);
-extern void ata_std_ports(struct ata_ioports *ioaddr);
 
 extern struct ata_host *ata_host_alloc(struct device *dev, int max_ports);
 extern struct ata_host *ata_host_alloc_pinfo(struct device *dev,
@@ -843,7 +854,6 @@ extern void ata_sas_port_stop(struct ata_port *ap);
 extern int ata_sas_slave_configure(struct scsi_device *, struct ata_port *);
 extern int ata_sas_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *),
 			    struct ata_port *ap);
-extern unsigned int ata_host_intr(struct ata_port *ap, struct ata_queued_cmd *qc);
 extern int sata_scr_valid(struct ata_link *link);
 extern int sata_scr_read(struct ata_link *link, int reg, u32 *val);
 extern int sata_scr_write(struct ata_link *link, int reg, u32 val);
@@ -855,21 +865,9 @@ extern int ata_host_suspend(struct ata_host *host, pm_message_t mesg);
 extern void ata_host_resume(struct ata_host *host);
 #endif
 extern int ata_ratelimit(void);
-extern int ata_busy_sleep(struct ata_port *ap,
-			  unsigned long timeout_pat, unsigned long timeout);
-extern void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline);
-extern int ata_wait_ready(struct ata_port *ap, unsigned long deadline);
 extern u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
 			     unsigned long interval_msec,
 			     unsigned long timeout_msec);
-extern unsigned int ata_dev_try_classify(struct ata_device *dev, int present,
-					 u8 *r_err);
-
-/*
- * Default driver ops implementations
- */
-extern void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf);
-extern void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
 extern int atapi_cmd_type(u8 opcode);
 extern void ata_tf_to_fis(const struct ata_taskfile *tf,
 			  u8 pmp, int is_cmd, u8 *fis);
@@ -885,22 +883,9 @@ extern int ata_xfer_mode2shift(unsigned long xfer_mode);
 extern const char *ata_mode_string(unsigned long xfer_mask);
 extern unsigned long ata_id_xfermask(const u16 *id);
 extern void ata_noop_dev_select(struct ata_port *ap, unsigned int device);
-extern void ata_std_dev_select(struct ata_port *ap, unsigned int device);
-extern u8 ata_check_status(struct ata_port *ap);
-extern u8 ata_altstatus(struct ata_port *ap);
-extern void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf);
 extern int ata_port_start(struct ata_port *ap);
-extern int ata_sff_port_start(struct ata_port *ap);
-extern irqreturn_t ata_interrupt(int irq, void *dev_instance);
-extern unsigned int ata_data_xfer(struct ata_device *dev,
-			unsigned char *buf, unsigned int buflen, int rw);
-extern unsigned int ata_data_xfer_noirq(struct ata_device *dev,
-			unsigned char *buf, unsigned int buflen, int rw);
 extern int ata_std_qc_defer(struct ata_queued_cmd *qc);
-extern void ata_dumb_qc_prep(struct ata_queued_cmd *qc);
-extern void ata_qc_prep(struct ata_queued_cmd *qc);
 extern void ata_noop_qc_prep(struct ata_queued_cmd *qc);
-extern unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc);
 extern void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
 		 unsigned int n_elem);
 extern unsigned int ata_dev_classify(const struct ata_taskfile *tf);
@@ -909,18 +894,7 @@ extern void ata_id_string(const u16 *id, unsigned char *s,
 			  unsigned int ofs, unsigned int len);
 extern void ata_id_c_string(const u16 *id, unsigned char *s,
 			    unsigned int ofs, unsigned int len);
-extern void ata_bmdma_setup(struct ata_queued_cmd *qc);
-extern void ata_bmdma_start(struct ata_queued_cmd *qc);
-extern void ata_bmdma_stop(struct ata_queued_cmd *qc);
-extern u8   ata_bmdma_status(struct ata_port *ap);
-extern void ata_bmdma_irq_clear(struct ata_port *ap);
 extern void ata_noop_irq_clear(struct ata_port *ap);
-extern void ata_bmdma_freeze(struct ata_port *ap);
-extern void ata_bmdma_thaw(struct ata_port *ap);
-extern void ata_bmdma_error_handler(struct ata_port *ap);
-extern void ata_bmdma_post_internal_cmd(struct ata_queued_cmd *qc);
-extern int ata_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
-			u8 status, int in_wq);
 extern void ata_qc_complete(struct ata_queued_cmd *qc);
 extern int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active,
 				    void (*finish_qc)(struct ata_queued_cmd *));
@@ -935,7 +909,6 @@ extern int ata_scsi_change_queue_depth(struct scsi_device *sdev,
 				       int queue_depth);
 extern struct ata_device *ata_dev_pair(struct ata_device *adev);
 extern int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev);
-extern u8 ata_irq_on(struct ata_port *ap);
 
 extern int ata_cable_40wire(struct ata_port *ap);
 extern int ata_cable_80wire(struct ata_port *ap);
@@ -943,10 +916,7 @@ extern int ata_cable_sata(struct ata_port *ap);
 extern int ata_cable_ignore(struct ata_port *ap);
 extern int ata_cable_unknown(struct ata_port *ap);
 
-/*
- * Timing helpers
- */
-
+/* Timing helpers */
 extern unsigned int ata_pio_need_iordy(const struct ata_device *);
 extern const struct ata_timing *ata_timing_find_mode(u8 xfer_mode);
 extern int ata_timing_compute(struct ata_device *, unsigned short,
@@ -956,24 +926,31 @@ extern void ata_timing_merge(const struct ata_timing *,
 			     unsigned int);
 extern u8 ata_timing_cycle2mode(unsigned int xfer_shift, int cycle);
 
-enum {
-	ATA_TIMING_SETUP	= (1 << 0),
-	ATA_TIMING_ACT8B	= (1 << 1),
-	ATA_TIMING_REC8B	= (1 << 2),
-	ATA_TIMING_CYC8B	= (1 << 3),
-	ATA_TIMING_8BIT		= ATA_TIMING_ACT8B | ATA_TIMING_REC8B |
-				  ATA_TIMING_CYC8B,
-	ATA_TIMING_ACTIVE	= (1 << 4),
-	ATA_TIMING_RECOVER	= (1 << 5),
-	ATA_TIMING_CYCLE	= (1 << 6),
-	ATA_TIMING_UDMA		= (1 << 7),
-	ATA_TIMING_ALL		= ATA_TIMING_SETUP | ATA_TIMING_ACT8B |
-				  ATA_TIMING_REC8B | ATA_TIMING_CYC8B |
-				  ATA_TIMING_ACTIVE | ATA_TIMING_RECOVER |
-				  ATA_TIMING_CYCLE | ATA_TIMING_UDMA,
+/* PCI */
+#ifdef CONFIG_PCI
+struct pci_dev;
+
+struct pci_bits {
+	unsigned int		reg;	/* PCI config register to read */
+	unsigned int		width;	/* 1 (8 bit), 2 (16 bit), 4 (32 bit) */
+	unsigned long		mask;
+	unsigned long		val;
 };
 
-/* libata-acpi.c */
+extern int pci_test_config_bits(struct pci_dev *pdev, const struct pci_bits *bits);
+extern void ata_pci_remove_one(struct pci_dev *pdev);
+
+#ifdef CONFIG_PM
+extern void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t mesg);
+extern int __must_check ata_pci_device_do_resume(struct pci_dev *pdev);
+extern int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg);
+extern int ata_pci_device_resume(struct pci_dev *pdev);
+#endif /* CONFIG_PM */
+#endif /* CONFIG_PCI */
+
+/*
+ * ACPI - drivers/ata/libata-acpi.c
+ */
 #ifdef CONFIG_ATA_ACPI
 static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap)
 {
@@ -1017,43 +994,8 @@ static inline int ata_acpi_cbl_80wire(struct ata_port *ap,
 }
 #endif
 
-#ifdef CONFIG_PCI
-struct pci_dev;
-
-extern int ata_pci_init_one(struct pci_dev *pdev,
-			    const struct ata_port_info * const * ppi,
-			    struct scsi_host_template *sht, void *host_priv);
-extern void ata_pci_remove_one(struct pci_dev *pdev);
-#ifdef CONFIG_PM
-extern void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t mesg);
-extern int __must_check ata_pci_device_do_resume(struct pci_dev *pdev);
-extern int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg);
-extern int ata_pci_device_resume(struct pci_dev *pdev);
-#endif
-extern int ata_pci_clear_simplex(struct pci_dev *pdev);
-
-struct pci_bits {
-	unsigned int		reg;	/* PCI config register to read */
-	unsigned int		width;	/* 1 (8 bit), 2 (16 bit), 4 (32 bit) */
-	unsigned long		mask;
-	unsigned long		val;
-};
-
-extern int ata_pci_init_sff_host(struct ata_host *host);
-extern int ata_pci_init_bmdma(struct ata_host *host);
-extern int ata_pci_prepare_sff_host(struct pci_dev *pdev,
-				    const struct ata_port_info * const * ppi,
-				    struct ata_host **r_host);
-extern int ata_pci_activate_sff_host(struct ata_host *host,
-				     irq_handler_t irq_handler,
-				     struct scsi_host_template *sht);
-extern int pci_test_config_bits(struct pci_dev *pdev, const struct pci_bits *bits);
-extern unsigned long ata_pci_default_filter(struct ata_device *dev,
-					    unsigned long xfer_mask);
-#endif /* CONFIG_PCI */
-
 /*
- * PMP
+ * PMP - drivers/ata/libata-pmp.c
  */
 extern int sata_pmp_qc_defer_cmd_switch(struct ata_queued_cmd *qc);
 extern int sata_pmp_std_prereset(struct ata_link *link, unsigned long deadline);
@@ -1063,7 +1005,7 @@ extern void sata_pmp_std_postreset(struct ata_link *link, unsigned int *class);
 extern void sata_pmp_error_handler(struct ata_port *ap);
 
 /*
- * EH
+ * EH - drivers/ata/libata-eh.c
  */
 extern void ata_port_schedule_eh(struct ata_port *ap);
 extern int ata_link_abort(struct ata_link *link);
@@ -1106,8 +1048,6 @@ extern void ata_std_error_handler(struct ata_port *ap);
 extern const struct ata_port_operations ata_base_port_ops;
 extern const struct ata_port_operations sata_port_ops;
 extern const struct ata_port_operations sata_pmp_port_ops;
-extern const struct ata_port_operations ata_sff_port_ops;
-extern const struct ata_port_operations ata_bmdma_port_ops;
 
 #define ATA_BASE_SHT(drv_name)					\
 	.module			= THIS_MODULE,			\
@@ -1124,17 +1064,6 @@ extern const struct ata_port_operations ata_bmdma_port_ops;
 	.slave_destroy		= ata_scsi_slave_destroy,	\
 	.bios_param		= ata_std_bios_param
 
-/* PIO only, sg_tablesize and dma_boundary limits can be removed */
-#define ATA_PIO_SHT(drv_name)					\
-	ATA_BASE_SHT(drv_name),					\
-	.sg_tablesize		= LIBATA_MAX_PRD,		\
-	.dma_boundary		= ATA_DMA_BOUNDARY
-
-#define ATA_BMDMA_SHT(drv_name)					\
-	ATA_BASE_SHT(drv_name),					\
-	.sg_tablesize		= LIBATA_MAX_PRD,		\
-	.dma_boundary		= ATA_DMA_BOUNDARY
-
 #define ATA_NCQ_SHT(drv_name)					\
 	ATA_BASE_SHT(drv_name),					\
 	.change_queue_depth	= ata_scsi_change_queue_depth
@@ -1287,11 +1216,6 @@ static inline struct ata_link *ata_port_next_link(struct ata_link *link)
 	for ((dev) = (link)->device + ata_link_max_devices(link) - 1; \
 	     (dev) >= (link)->device || ((dev) = NULL); (dev)--)
 
-static inline u8 ata_chk_status(struct ata_port *ap)
-{
-	return ap->ops->check_status(ap);
-}
-
 /**
  *	ata_ncq_enabled - Test whether NCQ is enabled
  *	@dev: ATA device to test for
@@ -1308,74 +1232,6 @@ static inline int ata_ncq_enabled(struct ata_device *dev)
 			      ATA_DFLAG_NCQ)) == ATA_DFLAG_NCQ;
 }
 
-/**
- *	ata_pause - Flush writes and pause 400 nanoseconds.
- *	@ap: Port to wait for.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-
-static inline void ata_pause(struct ata_port *ap)
-{
-	ata_altstatus(ap);
-	ndelay(400);
-}
-
-
-/**
- *	ata_busy_wait - Wait for a port status register
- *	@ap: Port to wait for.
- *	@bits: bits that must be clear
- *	@max: number of 10uS waits to perform
- *
- *	Waits up to max*10 microseconds for the selected bits in the port's
- *	status register to be cleared.
- *	Returns final value of status register.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-
-static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits,
-			       unsigned int max)
-{
-	u8 status;
-
-	do {
-		udelay(10);
-		status = ata_chk_status(ap);
-		max--;
-	} while (status != 0xff && (status & bits) && (max > 0));
-
-	return status;
-}
-
-
-/**
- *	ata_wait_idle - Wait for a port to be idle.
- *	@ap: Port to wait for.
- *
- *	Waits up to 10ms for port's BUSY and DRQ signals to clear.
- *	Returns final value of status register.
- *
- *	LOCKING:
- *	Inherited from caller.
- */
-
-static inline u8 ata_wait_idle(struct ata_port *ap)
-{
-	u8 status = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
-
-#ifdef ATA_DEBUG
-	if (status != 0xff && (status & (ATA_BUSY | ATA_DRQ)))
-		ata_port_printk(ap, KERN_DEBUG, "abnormal Status 0x%X\n",
-				status);
-#endif
-
-	return status;
-}
-
 static inline void ata_qc_set_polling(struct ata_queued_cmd *qc)
 {
 	qc->tf.ctl |= ATA_NIEN;
@@ -1468,4 +1324,149 @@ static inline struct ata_port *ata_shost_to_port(struct Scsi_Host *host)
 	return *(struct ata_port **)&host->hostdata[0];
 }
 
+/**************************************************************************
+ * SFF - drivers/ata/libata-sff.c
+ */
+extern const struct ata_port_operations ata_sff_port_ops;
+extern const struct ata_port_operations ata_bmdma_port_ops;
+
+/* PIO only, sg_tablesize and dma_boundary limits can be removed */
+#define ATA_PIO_SHT(drv_name)					\
+	ATA_BASE_SHT(drv_name),					\
+	.sg_tablesize		= LIBATA_MAX_PRD,		\
+	.dma_boundary		= ATA_DMA_BOUNDARY
+
+#define ATA_BMDMA_SHT(drv_name)					\
+	ATA_BASE_SHT(drv_name),					\
+	.sg_tablesize		= LIBATA_MAX_PRD,		\
+	.dma_boundary		= ATA_DMA_BOUNDARY
+
+extern void ata_qc_prep(struct ata_queued_cmd *qc);
+extern void ata_dumb_qc_prep(struct ata_queued_cmd *qc);
+extern void ata_std_dev_select(struct ata_port *ap, unsigned int device);
+extern u8 ata_check_status(struct ata_port *ap);
+extern u8 ata_altstatus(struct ata_port *ap);
+extern int ata_busy_sleep(struct ata_port *ap,
+			  unsigned long timeout_pat, unsigned long timeout);
+extern int ata_wait_ready(struct ata_port *ap, unsigned long deadline);
+extern void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf);
+extern void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
+extern void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf);
+extern unsigned int ata_data_xfer(struct ata_device *dev,
+			unsigned char *buf, unsigned int buflen, int rw);
+extern unsigned int ata_data_xfer_noirq(struct ata_device *dev,
+			unsigned char *buf, unsigned int buflen, int rw);
+extern u8 ata_irq_on(struct ata_port *ap);
+extern void ata_bmdma_irq_clear(struct ata_port *ap);
+extern int ata_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
+			u8 status, int in_wq);
+extern unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc);
+extern unsigned int ata_host_intr(struct ata_port *ap, struct ata_queued_cmd *qc);
+extern irqreturn_t ata_interrupt(int irq, void *dev_instance);
+extern void ata_bmdma_freeze(struct ata_port *ap);
+extern void ata_bmdma_thaw(struct ata_port *ap);
+extern int ata_std_prereset(struct ata_link *link, unsigned long deadline);
+extern unsigned int ata_dev_try_classify(struct ata_device *dev, int present,
+					 u8 *r_err);
+extern void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline);
+extern int ata_std_softreset(struct ata_link *link, unsigned int *classes,
+			     unsigned long deadline);
+extern int sata_std_hardreset(struct ata_link *link, unsigned int *class,
+			      unsigned long deadline);
+extern void ata_std_postreset(struct ata_link *link, unsigned int *classes);
+extern void ata_bmdma_error_handler(struct ata_port *ap);
+extern void ata_bmdma_post_internal_cmd(struct ata_queued_cmd *qc);
+extern int ata_sff_port_start(struct ata_port *ap);
+extern void ata_std_ports(struct ata_ioports *ioaddr);
+extern void ata_bmdma_setup(struct ata_queued_cmd *qc);
+extern void ata_bmdma_start(struct ata_queued_cmd *qc);
+extern void ata_bmdma_stop(struct ata_queued_cmd *qc);
+extern u8 ata_bmdma_status(struct ata_port *ap);
+extern void ata_bus_reset(struct ata_port *ap);
+
+#ifdef CONFIG_PCI
+extern int ata_pci_clear_simplex(struct pci_dev *pdev);
+extern unsigned long ata_pci_default_filter(struct ata_device *dev,
+					    unsigned long xfer_mask);
+extern int ata_pci_init_bmdma(struct ata_host *host);
+extern int ata_pci_init_sff_host(struct ata_host *host);
+extern int ata_pci_prepare_sff_host(struct pci_dev *pdev,
+				    const struct ata_port_info * const * ppi,
+				    struct ata_host **r_host);
+extern int ata_pci_activate_sff_host(struct ata_host *host,
+				     irq_handler_t irq_handler,
+				     struct scsi_host_template *sht);
+extern int ata_pci_init_one(struct pci_dev *pdev,
+			    const struct ata_port_info * const * ppi,
+			    struct scsi_host_template *sht, void *host_priv);
+#endif /* CONFIG_PCI */
+
+static inline u8 ata_chk_status(struct ata_port *ap)
+{
+	return ap->ops->check_status(ap);
+}
+
+/**
+ *	ata_pause - Flush writes and pause 400 nanoseconds.
+ *	@ap: Port to wait for.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+static inline void ata_pause(struct ata_port *ap)
+{
+	ata_altstatus(ap);
+	ndelay(400);
+}
+
+/**
+ *	ata_busy_wait - Wait for a port status register
+ *	@ap: Port to wait for.
+ *	@bits: bits that must be clear
+ *	@max: number of 10uS waits to perform
+ *
+ *	Waits up to max*10 microseconds for the selected bits in the port's
+ *	status register to be cleared.
+ *	Returns final value of status register.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits,
+			       unsigned int max)
+{
+	u8 status;
+
+	do {
+		udelay(10);
+		status = ata_chk_status(ap);
+		max--;
+	} while (status != 0xff && (status & bits) && (max > 0));
+
+	return status;
+}
+
+/**
+ *	ata_wait_idle - Wait for a port to be idle.
+ *	@ap: Port to wait for.
+ *
+ *	Waits up to 10ms for port's BUSY and DRQ signals to clear.
+ *	Returns final value of status register.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+static inline u8 ata_wait_idle(struct ata_port *ap)
+{
+	u8 status = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
+
+#ifdef ATA_DEBUG
+	if (status != 0xff && (status & (ATA_BUSY | ATA_DRQ)))
+		ata_port_printk(ap, KERN_DEBUG, "abnormal Status 0x%X\n",
+				status);
+#endif
+
+	return status;
+}
+
 #endif /* __LINUX_LIBATA_H__ */
-- 
cgit v1.2.3


From 071ce34d57924edb76b76f7de460eb4991463959 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 22:16:42 +0900
Subject: libata: move ata_pci_default_filter() out of CONFIG_PCI

ata_pci_default_filter() doesn't really have anything to do with PCI.
It's generally applicable to BMDMA controllers.  Move it out of
CONFIG_PCI.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-sff.c | 22 +++++++++++-----------
 include/linux/libata.h   |  4 ++--
 2 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 840ae6da59bc..ebdd46bc13c4 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -2117,6 +2117,16 @@ void ata_std_ports(struct ata_ioports *ioaddr)
 	ioaddr->command_addr = ioaddr->cmd_addr + ATA_REG_CMD;
 }
 
+unsigned long ata_pci_default_filter(struct ata_device *adev, unsigned long xfer_mask)
+{
+	/* Filter out DMA modes if the device has been configured by
+	   the BIOS as PIO only */
+
+	if (adev->link->ap->ioaddr.bmdma_addr == NULL)
+		xfer_mask &= ~(ATA_MASK_MWDMA | ATA_MASK_UDMA);
+	return xfer_mask;
+}
+
 /**
  *	ata_bmdma_setup - Set up PCI IDE BMDMA transaction
  *	@qc: Info associated with this ATA transaction.
@@ -2335,16 +2345,6 @@ int ata_pci_clear_simplex(struct pci_dev *pdev)
 	return 0;
 }
 
-unsigned long ata_pci_default_filter(struct ata_device *adev, unsigned long xfer_mask)
-{
-	/* Filter out DMA modes if the device has been configured by
-	   the BIOS as PIO only */
-
-	if (adev->link->ap->ioaddr.bmdma_addr == NULL)
-		xfer_mask &= ~(ATA_MASK_MWDMA | ATA_MASK_UDMA);
-	return xfer_mask;
-}
-
 /**
  *	ata_pci_init_bmdma - acquire PCI BMDMA resources and init ATA host
  *	@host: target ATA host
@@ -2725,6 +2725,7 @@ EXPORT_SYMBOL_GPL(ata_sff_port_ops);
 EXPORT_SYMBOL_GPL(ata_bmdma_port_ops);
 EXPORT_SYMBOL_GPL(ata_qc_prep);
 EXPORT_SYMBOL_GPL(ata_dumb_qc_prep);
+EXPORT_SYMBOL_GPL(ata_pci_default_filter);
 EXPORT_SYMBOL_GPL(ata_std_dev_select);
 EXPORT_SYMBOL_GPL(ata_check_status);
 EXPORT_SYMBOL_GPL(ata_altstatus);
@@ -2760,7 +2761,6 @@ EXPORT_SYMBOL_GPL(ata_bmdma_status);
 EXPORT_SYMBOL_GPL(ata_bus_reset);
 #ifdef CONFIG_PCI
 EXPORT_SYMBOL_GPL(ata_pci_clear_simplex);
-EXPORT_SYMBOL_GPL(ata_pci_default_filter);
 EXPORT_SYMBOL_GPL(ata_pci_init_bmdma);
 EXPORT_SYMBOL_GPL(ata_pci_init_sff_host);
 EXPORT_SYMBOL_GPL(ata_pci_prepare_sff_host);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 673f34b256ba..53b8db05a1fb 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1378,6 +1378,8 @@ extern void ata_bmdma_error_handler(struct ata_port *ap);
 extern void ata_bmdma_post_internal_cmd(struct ata_queued_cmd *qc);
 extern int ata_sff_port_start(struct ata_port *ap);
 extern void ata_std_ports(struct ata_ioports *ioaddr);
+extern unsigned long ata_pci_default_filter(struct ata_device *dev,
+					    unsigned long xfer_mask);
 extern void ata_bmdma_setup(struct ata_queued_cmd *qc);
 extern void ata_bmdma_start(struct ata_queued_cmd *qc);
 extern void ata_bmdma_stop(struct ata_queued_cmd *qc);
@@ -1386,8 +1388,6 @@ extern void ata_bus_reset(struct ata_port *ap);
 
 #ifdef CONFIG_PCI
 extern int ata_pci_clear_simplex(struct pci_dev *pdev);
-extern unsigned long ata_pci_default_filter(struct ata_device *dev,
-					    unsigned long xfer_mask);
 extern int ata_pci_init_bmdma(struct ata_host *host);
 extern int ata_pci_init_sff_host(struct ata_host *host);
 extern int ata_pci_prepare_sff_host(struct pci_dev *pdev,
-- 
cgit v1.2.3


From 6fd36390117f7844ad147377878ddb52088f583a Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 25 Mar 2008 22:16:44 +0900
Subject: libata: kill ata_chk_status()

ata_chk_status() just calls ops->check_status and it only adds
confusion with other status functions.  Kill it.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-sff.c    | 22 +++++++++++-----------
 drivers/ata/pata_bf54x.c    |  2 +-
 drivers/ata/pata_scc.c      |  2 +-
 drivers/ata/sata_inic162x.c | 10 +++++-----
 drivers/ata/sata_sil.c      |  6 +++---
 drivers/ata/sata_via.c      |  2 +-
 drivers/ata/sata_vsc.c      |  2 +-
 include/linux/libata.h      |  7 +------
 8 files changed, 24 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index ebdd46bc13c4..c601dcef2925 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -290,7 +290,7 @@ int ata_busy_sleep(struct ata_port *ap,
 	while (status != 0xff && (status & ATA_BUSY) &&
 	       time_before(jiffies, timeout)) {
 		msleep(50);
-		status = ata_chk_status(ap);
+		status = ap->ops->check_status(ap);
 	}
 
 	if (status == 0xff)
@@ -326,7 +326,7 @@ int ata_wait_ready(struct ata_port *ap, unsigned long deadline)
 	int warned = 0;
 
 	while (1) {
-		u8 status = ata_chk_status(ap);
+		u8 status = ap->ops->check_status(ap);
 		unsigned long now = jiffies;
 
 		if (!(status & ATA_BUSY))
@@ -1486,7 +1486,7 @@ inline unsigned int ata_host_intr(struct ata_port *ap,
 		goto idle_irq;
 
 	/* check main status, clearing INTRQ */
-	status = ata_chk_status(ap);
+	status = ap->ops->check_status(ap);
 	if (unlikely(status & ATA_BUSY))
 		goto idle_irq;
 
@@ -1506,7 +1506,7 @@ idle_irq:
 
 #ifdef ATA_IRQ_TRAP
 	if ((ap->stats.idle_irq % 1000) == 0) {
-		ata_chk_status(ap);
+		ap->ops->check_status(ap);
 		ap->ops->irq_clear(ap);
 		ata_port_printk(ap, KERN_WARNING, "irq trap\n");
 		return 1;
@@ -1582,7 +1582,7 @@ void ata_bmdma_freeze(struct ata_port *ap)
 	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
 	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
 	 */
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 
 	ap->ops->irq_clear(ap);
 }
@@ -1599,7 +1599,7 @@ void ata_bmdma_freeze(struct ata_port *ap)
 void ata_bmdma_thaw(struct ata_port *ap)
 {
 	/* clear & re-enable interrupts */
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 	ap->ops->irq_clear(ap);
 	ap->ops->irq_on(ap);
 }
@@ -1709,7 +1709,7 @@ unsigned int ata_dev_try_classify(struct ata_device *dev, int present,
 			class = ATA_DEV_ATA;
 		else
 			class = ATA_DEV_NONE;
-	} else if ((class == ATA_DEV_ATA) && (ata_chk_status(ap) == 0))
+	} else if ((class == ATA_DEV_ATA) && (ap->ops->check_status(ap) == 0))
 		class = ATA_DEV_NONE;
 
 	return class;
@@ -1820,7 +1820,7 @@ void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline)
 	 */
 	if (ap->flags & ATA_FLAG_SATA) {
 		while (1) {
-			u8 status = ata_chk_status(ap);
+			u8 status = ap->ops->check_status(ap);
 
 			if (status != 0xff || time_after(jiffies, deadline))
 				return;
@@ -1851,7 +1851,7 @@ static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
 	 * the bus shows 0xFF because the odd clown forgets the D7
 	 * pulldown resistor.
 	 */
-	if (ata_chk_status(ap) == 0xFF)
+	if (ap->ops->check_status(ap) == 0xFF)
 		return -ENODEV;
 
 	return ata_bus_post_reset(ap, devmask, deadline);
@@ -2034,7 +2034,7 @@ void ata_bmdma_error_handler(struct ata_port *ap)
 	}
 
 	ata_altstatus(ap);
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 	ap->ops->irq_clear(ap);
 
 	spin_unlock_irqrestore(ap->lock, flags);
@@ -2725,7 +2725,6 @@ EXPORT_SYMBOL_GPL(ata_sff_port_ops);
 EXPORT_SYMBOL_GPL(ata_bmdma_port_ops);
 EXPORT_SYMBOL_GPL(ata_qc_prep);
 EXPORT_SYMBOL_GPL(ata_dumb_qc_prep);
-EXPORT_SYMBOL_GPL(ata_pci_default_filter);
 EXPORT_SYMBOL_GPL(ata_std_dev_select);
 EXPORT_SYMBOL_GPL(ata_check_status);
 EXPORT_SYMBOL_GPL(ata_altstatus);
@@ -2754,6 +2753,7 @@ EXPORT_SYMBOL_GPL(ata_bmdma_error_handler);
 EXPORT_SYMBOL_GPL(ata_bmdma_post_internal_cmd);
 EXPORT_SYMBOL_GPL(ata_sff_port_start);
 EXPORT_SYMBOL_GPL(ata_std_ports);
+EXPORT_SYMBOL_GPL(ata_pci_default_filter);
 EXPORT_SYMBOL_GPL(ata_bmdma_setup);
 EXPORT_SYMBOL_GPL(ata_bmdma_start);
 EXPORT_SYMBOL_GPL(ata_bmdma_stop);
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index eea275acb2a8..457ac800cd5e 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1264,7 +1264,7 @@ static void bfin_bmdma_freeze(struct ata_port *ap)
 	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
 	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
 	 */
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 
 	bfin_irq_clear(ap);
 }
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index 07f2d7a6f1a8..85d33637dff9 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -854,7 +854,7 @@ static void scc_bmdma_freeze (struct ata_port *ap)
 	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
 	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
 	 */
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 
 	ap->ops->irq_clear(ap);
 }
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index ba1c09953517..12fbf3868fe2 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -267,14 +267,14 @@ static void inic_host_intr(struct ata_port *ap)
 			ata_qc_from_tag(ap, ap->link.active_tag);
 
 		if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) {
-			ata_chk_status(ap);	/* clear ATA interrupt */
+			ap->ops->check_status(ap); /* clear ATA interrupt */
 			return;
 		}
 
 		if (likely(ata_host_intr(ap, qc)))
 			return;
 
-		ata_chk_status(ap);	/* clear ATA interrupt */
+		ap->ops->check_status(ap); /* clear ATA interrupt */
 		ata_port_printk(ap, KERN_WARNING, "unhandled "
 				"interrupt, irq_stat=%x\n", irq_stat);
 		return;
@@ -351,7 +351,7 @@ static unsigned int inic_qc_issue(struct ata_queued_cmd *qc)
 	 */
 	if (unlikely(qc->tf.command == ATA_CMD_ID_ATA ||
 		     qc->tf.command == ATA_CMD_ID_ATAPI)) {
-		u8 stat = ata_chk_status(ap);
+		u8 stat = ap->ops->check_status(ap);
 		if (stat == 0x7f || stat == 0xff)
 			return AC_ERR_HSM;
 	}
@@ -365,7 +365,7 @@ static void inic_freeze(struct ata_port *ap)
 
 	__inic_set_pirq_mask(ap, PIRQ_MASK_FREEZE);
 
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 	writeb(0xff, port_base + PORT_IRQ_STAT);
 
 	readb(port_base + PORT_IRQ_STAT); /* flush */
@@ -375,7 +375,7 @@ static void inic_thaw(struct ata_port *ap)
 {
 	void __iomem *port_base = inic_port_base(ap);
 
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 	writeb(0xff, port_base + PORT_IRQ_STAT);
 
 	__inic_set_pirq_mask(ap, PIRQ_MASK_OTHER);
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index eac7ca73cfa0..659dfcbdc1b2 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -369,7 +369,7 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 
 	if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) {
 		/* this sometimes happens, just clear IRQ */
-		ata_chk_status(ap);
+		ap->ops->check_status(ap);
 		return;
 	}
 
@@ -405,7 +405,7 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 	}
 
 	/* check main status, clearing INTRQ */
-	status = ata_chk_status(ap);
+	status = ap->ops->check_status(ap);
 	if (unlikely(status & ATA_BUSY))
 		goto err_hsm;
 
@@ -480,7 +480,7 @@ static void sil_thaw(struct ata_port *ap)
 	u32 tmp;
 
 	/* clear IRQ */
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 	ata_bmdma_irq_clear(ap);
 
 	/* turn on SATA IRQ if supported */
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 402fd7333d48..4bc6e849af2c 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -173,7 +173,7 @@ static void svia_noop_freeze(struct ata_port *ap)
 	/* Some VIA controllers choke if ATA_NIEN is manipulated in
 	 * certain way.  Leave it alone and just clear pending IRQ.
 	 */
-	ata_chk_status(ap);
+	ap->ops->check_status(ap);
 	ata_bmdma_irq_clear(ap);
 }
 
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index 8045a72dc559..fb3a88722664 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -251,7 +251,7 @@ static void vsc_port_intr(u8 port_status, struct ata_port *ap)
 	 * simply clear the interrupt
 	 */
 	if (unlikely(!handled))
-		ata_chk_status(ap);
+		ap->ops->check_status(ap);
 }
 
 /*
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 53b8db05a1fb..61a7f8d06971 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1401,11 +1401,6 @@ extern int ata_pci_init_one(struct pci_dev *pdev,
 			    struct scsi_host_template *sht, void *host_priv);
 #endif /* CONFIG_PCI */
 
-static inline u8 ata_chk_status(struct ata_port *ap)
-{
-	return ap->ops->check_status(ap);
-}
-
 /**
  *	ata_pause - Flush writes and pause 400 nanoseconds.
  *	@ap: Port to wait for.
@@ -1439,7 +1434,7 @@ static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits,
 
 	do {
 		udelay(10);
-		status = ata_chk_status(ap);
+		status = ap->ops->check_status(ap);
 		max--;
 	} while (status != 0xff && (status & bits) && (max > 0));
 
-- 
cgit v1.2.3


From 03faab7827e4e45823fd27c47b84c133e20a0cd0 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 27 Mar 2008 19:14:24 +0900
Subject: libata: implement ATA_QCFLAG_RETRY

Currently whether a command should be retried after failure is
determined inside ata_eh_finish().  Add ATA_QCFLAG_RETRY and move the
logic into ata_eh_autopsy().  This makes things clearer and helps
extending retry determination logic.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-eh.c | 18 ++++++++----------
 include/linux/libata.h  |  1 +
 2 files changed, 9 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index ec32082356cb..cc8548e1572a 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1785,6 +1785,11 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 		if (qc->flags & ATA_QCFLAG_SENSE_VALID)
 			qc->err_mask &= ~(AC_ERR_DEV | AC_ERR_OTHER);
 
+		/* determine whether the command is worth retrying */
+		if (!(qc->err_mask & AC_ERR_INVALID) &&
+		    ((qc->flags & ATA_QCFLAG_IO) || qc->err_mask != AC_ERR_DEV))
+			qc->flags |= ATA_QCFLAG_RETRY;
+
 		/* accumulate error info */
 		ehc->i.dev = qc->dev;
 		all_err_mask |= qc->err_mask;
@@ -2783,18 +2788,11 @@ void ata_eh_finish(struct ata_port *ap)
 			/* FIXME: Once EH migration is complete,
 			 * generate sense data in this function,
 			 * considering both err_mask and tf.
-			 *
-			 * There's no point in retrying invalid
-			 * (detected by libata) and non-IO device
-			 * errors (rejected by device).  Finish them
-			 * immediately.
 			 */
-			if ((qc->err_mask & AC_ERR_INVALID) ||
-			    (!(qc->flags & ATA_QCFLAG_IO) &&
-			     qc->err_mask == AC_ERR_DEV))
-				ata_eh_qc_complete(qc);
-			else
+			if (qc->flags & ATA_QCFLAG_RETRY)
 				ata_eh_qc_retry(qc);
+			else
+				ata_eh_qc_complete(qc);
 		} else {
 			if (qc->flags & ATA_QCFLAG_SENSE_VALID) {
 				ata_eh_qc_complete(qc);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 61a7f8d06971..b25ea6ab1be9 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -224,6 +224,7 @@ enum {
 	ATA_QCFLAG_RESULT_TF	= (1 << 4), /* result TF requested */
 	ATA_QCFLAG_CLEAR_EXCL	= (1 << 5), /* clear excl_link on completion */
 	ATA_QCFLAG_QUIET	= (1 << 6), /* don't report device error */
+	ATA_QCFLAG_RETRY	= (1 << 7), /* retry after failure */
 
 	ATA_QCFLAG_FAILED	= (1 << 16), /* cmd failed and is owned by EH */
 	ATA_QCFLAG_SENSE_VALID	= (1 << 17), /* sense data valid */
-- 
cgit v1.2.3


From 83c063dd730cb56bf3fc89b70250ff9a398fec1e Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Thu, 28 Feb 2008 21:43:13 +0900
Subject: use ATA_TAG_INTERNAL in ata_tag_internal()

It should be ATA_TAG_INTERNAL.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 include/linux/libata.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/libata.h b/include/linux/libata.h
index b25ea6ab1be9..a05de2ba7a72 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1121,7 +1121,7 @@ static inline unsigned int ata_tag_valid(unsigned int tag)
 
 static inline unsigned int ata_tag_internal(unsigned int tag)
 {
-	return tag == ATA_MAX_QUEUE - 1;
+	return tag == ATA_TAG_INTERNAL;
 }
 
 /*
-- 
cgit v1.2.3


From 9363c3825ea9ad76561eb48a395349dd29211ed6 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:16 +0900
Subject: libata: rename SFF functions

SFF functions have confusing names.  Some have sff prefix, some have
bmdma, some std, some pci and some none.  Unify the naming by...

* SFF functions which are common to both BMDMA and non-BMDMA are
  prefixed with ata_sff_.

* SFF functions which are specific to BMDMA are prefixed with
  ata_bmdma_.

* SFF functions which are specific to PCI but apply to both BMDMA and
  non-BMDMA are prefixed with ata_pci_sff_.

* SFF functions which are specific to PCI and BMDMA are prefixed with
  ata_pci_bmdma_.

* Drop generic prefixes from LLD specific routines.  For example,
  bfin_std_dev_select -> bfin_dev_select.

The following renames are noteworthy.

  ata_qc_issue_prot() -> ata_sff_qc_issue()
  ata_pci_default_filter() -> ata_bmdma_mode_filter()
  ata_dev_try_classify() -> ata_sff_dev_classify()

This rename is in preparation of separating SFF support out of libata
core layer.  This patch strictly renames functions and doesn't
introduce any behavior difference.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c              |  10 +-
 drivers/ata/ata_generic.c       |   4 +-
 drivers/ata/ata_piix.c          |   6 +-
 drivers/ata/libata-core.c       |  16 +-
 drivers/ata/libata-eh.c         |   2 +-
 drivers/ata/libata-sff.c        | 344 ++++++++++++++++++++--------------------
 drivers/ata/pata_acpi.c         |  16 +-
 drivers/ata/pata_ali.c          |   6 +-
 drivers/ata/pata_amd.c          |  10 +-
 drivers/ata/pata_artop.c        |   6 +-
 drivers/ata/pata_at32.c         |   2 +-
 drivers/ata/pata_atiixp.c       |   6 +-
 drivers/ata/pata_bf54x.c        |  84 +++++-----
 drivers/ata/pata_cmd640.c       |  16 +-
 drivers/ata/pata_cmd64x.c       |   4 +-
 drivers/ata/pata_cs5520.c       |   8 +-
 drivers/ata/pata_cs5530.c       |  12 +-
 drivers/ata/pata_cs5535.c       |   2 +-
 drivers/ata/pata_cs5536.c       |   2 +-
 drivers/ata/pata_cypress.c      |   2 +-
 drivers/ata/pata_efar.c         |   4 +-
 drivers/ata/pata_hpt366.c       |   4 +-
 drivers/ata/pata_hpt37x.c       |  10 +-
 drivers/ata/pata_hpt3x2n.c      |  10 +-
 drivers/ata/pata_hpt3x3.c       |   6 +-
 drivers/ata/pata_icside.c       |   8 +-
 drivers/ata/pata_isapnp.c       |   4 +-
 drivers/ata/pata_it8213.c       |   4 +-
 drivers/ata/pata_it821x.c       |  20 +--
 drivers/ata/pata_ixp4xx_cf.c    |   4 +-
 drivers/ata/pata_jmicron.c      |   4 +-
 drivers/ata/pata_legacy.c       |  32 ++--
 drivers/ata/pata_marvell.c      |   4 +-
 drivers/ata/pata_mpc52xx.c      |   4 +-
 drivers/ata/pata_mpiix.c        |  18 +--
 drivers/ata/pata_netcell.c      |   4 +-
 drivers/ata/pata_ninja32.c      |   6 +-
 drivers/ata/pata_ns87410.c      |  12 +-
 drivers/ata/pata_ns87415.c      |   8 +-
 drivers/ata/pata_oldpiix.c      |  12 +-
 drivers/ata/pata_opti.c         |   4 +-
 drivers/ata/pata_optidma.c      |   4 +-
 drivers/ata/pata_pcmcia.c       |   6 +-
 drivers/ata/pata_pdc2027x.c     |  10 +-
 drivers/ata/pata_pdc202xx_old.c |   2 +-
 drivers/ata/pata_platform.c     |   4 +-
 drivers/ata/pata_qdi.c          |  14 +-
 drivers/ata/pata_radisys.c      |  10 +-
 drivers/ata/pata_rb500_cf.c     |   6 +-
 drivers/ata/pata_rz1000.c       |   2 +-
 drivers/ata/pata_sc1200.c       |  12 +-
 drivers/ata/pata_scc.c          |  80 +++++-----
 drivers/ata/pata_serverworks.c  |  12 +-
 drivers/ata/pata_sil680.c       |  10 +-
 drivers/ata/pata_sis.c          |   4 +-
 drivers/ata/pata_sl82c105.c     |   4 +-
 drivers/ata/pata_triflex.c      |   4 +-
 drivers/ata/pata_via.c          |   6 +-
 drivers/ata/pata_winbond.c      |   6 +-
 drivers/ata/pdc_adma.c          |  12 +-
 drivers/ata/sata_inic162x.c     |  12 +-
 drivers/ata/sata_mv.c           |   6 +-
 drivers/ata/sata_nv.c           |  32 ++--
 drivers/ata/sata_promise.c      |  12 +-
 drivers/ata/sata_qstor.c        |  10 +-
 drivers/ata/sata_sil.c          |   8 +-
 drivers/ata/sata_sis.c          |   6 +-
 drivers/ata/sata_svw.c          |   4 +-
 drivers/ata/sata_sx4.c          |  26 +--
 drivers/ata/sata_uli.c          |  12 +-
 drivers/ata/sata_via.c          |  12 +-
 drivers/ata/sata_vsc.c          |   4 +-
 include/linux/libata.h          | 103 ++++++------
 73 files changed, 599 insertions(+), 596 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 3efa9904f7a0..771509c9a3fb 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1303,9 +1303,9 @@ static int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 	ahci_exec_polled_cmd(ap, pmp, &tf, 0, 0, 0);
 
 	/* wait a while before checking status */
-	ata_wait_after_reset(ap, deadline);
+	ata_sff_wait_after_reset(ap, deadline);
 
-	rc = ata_wait_ready(ap, deadline);
+	rc = ata_sff_wait_ready(ap, deadline);
 	/* link occupied, -ENODEV too is an error */
 	if (rc) {
 		reason = "device not ready";
@@ -1350,7 +1350,7 @@ static int ahci_hardreset(struct ata_link *link, unsigned int *class,
 	tf.command = 0x80;
 	ata_tf_to_fis(&tf, 0, 0, d2h_fis);
 
-	rc = sata_std_hardreset(link, class, deadline);
+	rc = sata_sff_hardreset(link, class, deadline);
 
 	ahci_start_engine(ap);
 
@@ -1431,7 +1431,7 @@ static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
 	 * have to be reset again.  For most cases, this should
 	 * suffice while making probing snappish enough.
 	 */
-	rc = ata_wait_ready(ap, jiffies + 2 * HZ);
+	rc = ata_sff_wait_ready(ap, jiffies + 2 * HZ);
 	if (rc)
 		ahci_kick_engine(ap, 0);
 
@@ -1444,7 +1444,7 @@ static void ahci_postreset(struct ata_link *link, unsigned int *class)
 	void __iomem *port_mmio = ahci_port_base(ap);
 	u32 new_tmp, tmp;
 
-	ata_std_postreset(link, class);
+	ata_sff_postreset(link, class);
 
 	/* Make sure port's ATAPI bit is set appropriately */
 	new_tmp = tmp = readl(port_mmio + PORT_CMD);
diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index b23e2a1099c5..47aeccd52fa9 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -150,9 +150,9 @@ static int ata_generic_init_one(struct pci_dev *dev, const struct pci_device_id
 		return -ENODEV;
 
 	if (dev->vendor == PCI_VENDOR_ID_AL)
-	    	ata_pci_clear_simplex(dev);
+		ata_pci_bmdma_clear_simplex(dev);
 
-	return ata_pci_init_one(dev, ppi, &generic_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &generic_sht, NULL);
 }
 
 static struct pci_device_id ata_generic[] = {
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index e113f2f80275..7ab76a413cdf 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -629,7 +629,7 @@ static int piix_pata_prereset(struct ata_link *link, unsigned long deadline)
 
 	if (!pci_test_config_bits(pdev, &piix_enable_bits[ap->port_no]))
 		return -ENOENT;
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -1493,7 +1493,7 @@ static int __devinit piix_init_one(struct pci_dev *pdev,
 		hpriv->map = piix_init_sata_map(pdev, port_info,
 					piix_map_db_table[ent->driver_data]);
 
-	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
+	rc = ata_pci_sff_prepare_host(pdev, ppi, &host);
 	if (rc)
 		return rc;
 	host->private_data = hpriv;
@@ -1527,7 +1527,7 @@ static int __devinit piix_init_one(struct pci_dev *pdev,
 	}
 
 	pci_set_master(pdev);
-	return ata_pci_activate_sff_host(host, ata_interrupt, &piix_sht);
+	return ata_pci_sff_activate_host(host, ata_sff_interrupt, &piix_sht);
 }
 
 static int __init piix_init(void)
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index fa205d7c5bea..cbdbfb5eaeaa 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -75,9 +75,9 @@ const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
 const struct ata_port_operations ata_base_port_ops = {
 	.irq_clear		= ata_noop_irq_clear,
-	.prereset		= ata_std_prereset,
-	.hardreset		= sata_std_hardreset,
-	.postreset		= ata_std_postreset,
+	.prereset		= ata_sff_prereset,
+	.hardreset		= sata_sff_hardreset,
+	.postreset		= ata_sff_postreset,
 	.error_handler		= ata_std_error_handler,
 };
 
@@ -3425,7 +3425,7 @@ int sata_link_resume(struct ata_link *link, const unsigned long *params,
 }
 
 /**
- *	ata_std_prereset - prepare for reset
+ *	ata_sff_prereset - prepare for reset
  *	@link: ATA link to be reset
  *	@deadline: deadline jiffies for the operation
  *
@@ -3441,7 +3441,7 @@ int sata_link_resume(struct ata_link *link, const unsigned long *params,
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_std_prereset(struct ata_link *link, unsigned long deadline)
+int ata_sff_prereset(struct ata_link *link, unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
 	struct ata_eh_context *ehc = &link->eh_context;
@@ -3463,7 +3463,7 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
 
 	/* wait for !BSY if we don't know that no device is attached */
 	if (!ata_link_offline(link)) {
-		rc = ata_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(ap, deadline);
 		if (rc && rc != -ENODEV) {
 			ata_link_printk(link, KERN_WARNING, "device not ready "
 					"(errno=%d), forcing hardreset\n", rc);
@@ -3535,7 +3535,7 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 }
 
 /**
- *	ata_std_postreset - standard postreset callback
+ *	ata_sff_postreset - standard postreset callback
  *	@link: the target ata_link
  *	@classes: classes of attached devices
  *
@@ -3546,7 +3546,7 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
  *	LOCKING:
  *	Kernel thread context (may sleep)
  */
-void ata_std_postreset(struct ata_link *link, unsigned int *classes)
+void ata_sff_postreset(struct ata_link *link, unsigned int *classes)
 {
 	struct ata_port *ap = link->ap;
 	u32 serror;
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index cc8548e1572a..f4f9c2783821 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2861,7 +2861,7 @@ void ata_std_error_handler(struct ata_port *ap)
 	 * ata_base_port_ops.  Ignore it if SCR access is not
 	 * available.
 	 */
-	if (hardreset == sata_std_hardreset && !sata_scr_valid(&ap->link))
+	if (hardreset == sata_sff_hardreset && !sata_scr_valid(&ap->link))
 		hardreset = NULL;
 
 	ata_do_eh(ap, ops->prereset, ops->softreset, hardreset, ops->postreset);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 5208ca21f634..8544321293d4 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -42,22 +42,22 @@
 const struct ata_port_operations ata_sff_port_ops = {
 	.inherits		= &ata_base_port_ops,
 
-	.qc_prep		= ata_qc_prep,
-	.qc_issue		= ata_qc_issue_prot,
-
-	.freeze			= ata_bmdma_freeze,
-	.thaw			= ata_bmdma_thaw,
-	.softreset		= ata_std_softreset,
-	.error_handler		= ata_bmdma_error_handler,
-	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
-
-	.dev_select		= ata_std_dev_select,
-	.check_status		= ata_check_status,
-	.tf_load		= ata_tf_load,
-	.tf_read		= ata_tf_read,
-	.exec_command		= ata_exec_command,
-	.data_xfer		= ata_data_xfer,
-	.irq_on			= ata_irq_on,
+	.qc_prep		= ata_sff_qc_prep,
+	.qc_issue		= ata_sff_qc_issue,
+
+	.freeze			= ata_sff_freeze,
+	.thaw			= ata_sff_thaw,
+	.softreset		= ata_sff_softreset,
+	.error_handler		= ata_sff_error_handler,
+	.post_internal_cmd	= ata_sff_post_internal_cmd,
+
+	.dev_select		= ata_sff_dev_select,
+	.check_status		= ata_sff_check_status,
+	.tf_load		= ata_sff_tf_load,
+	.tf_read		= ata_sff_tf_read,
+	.exec_command		= ata_sff_exec_command,
+	.data_xfer		= ata_sff_data_xfer,
+	.irq_on			= ata_sff_irq_on,
 
 	.port_start		= ata_sff_port_start,
 };
@@ -65,13 +65,13 @@ const struct ata_port_operations ata_sff_port_ops = {
 const struct ata_port_operations ata_bmdma_port_ops = {
 	.inherits		= &ata_sff_port_ops,
 
-	.mode_filter		= ata_pci_default_filter,
+	.mode_filter		= ata_bmdma_mode_filter,
 
 	.bmdma_setup		= ata_bmdma_setup,
 	.bmdma_start		= ata_bmdma_start,
 	.bmdma_stop		= ata_bmdma_stop,
 	.bmdma_status		= ata_bmdma_status,
-	.irq_clear		= ata_bmdma_irq_clear,
+	.irq_clear		= ata_sff_irq_clear,
 };
 
 /**
@@ -181,7 +181,7 @@ static void ata_fill_sg_dumb(struct ata_queued_cmd *qc)
 }
 
 /**
- *	ata_qc_prep - Prepare taskfile for submission
+ *	ata_sff_qc_prep - Prepare taskfile for submission
  *	@qc: Metadata associated with taskfile to be prepared
  *
  *	Prepare ATA taskfile for submission.
@@ -189,7 +189,7 @@ static void ata_fill_sg_dumb(struct ata_queued_cmd *qc)
  *	LOCKING:
  *	spin_lock_irqsave(host lock)
  */
-void ata_qc_prep(struct ata_queued_cmd *qc)
+void ata_sff_qc_prep(struct ata_queued_cmd *qc)
 {
 	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
 		return;
@@ -198,7 +198,7 @@ void ata_qc_prep(struct ata_queued_cmd *qc)
 }
 
 /**
- *	ata_dumb_qc_prep - Prepare taskfile for submission
+ *	ata_sff_dumb_qc_prep - Prepare taskfile for submission
  *	@qc: Metadata associated with taskfile to be prepared
  *
  *	Prepare ATA taskfile for submission.
@@ -206,7 +206,7 @@ void ata_qc_prep(struct ata_queued_cmd *qc)
  *	LOCKING:
  *	spin_lock_irqsave(host lock)
  */
-void ata_dumb_qc_prep(struct ata_queued_cmd *qc)
+void ata_sff_dumb_qc_prep(struct ata_queued_cmd *qc)
 {
 	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
 		return;
@@ -215,7 +215,7 @@ void ata_dumb_qc_prep(struct ata_queued_cmd *qc)
 }
 
 /**
- *	ata_check_status - Read device status reg & clear interrupt
+ *	ata_sff_check_status - Read device status reg & clear interrupt
  *	@ap: port where the device is
  *
  *	Reads ATA taskfile status register for currently-selected device
@@ -225,13 +225,13 @@ void ata_dumb_qc_prep(struct ata_queued_cmd *qc)
  *	LOCKING:
  *	Inherited from caller.
  */
-u8 ata_check_status(struct ata_port *ap)
+u8 ata_sff_check_status(struct ata_port *ap)
 {
 	return ioread8(ap->ioaddr.status_addr);
 }
 
 /**
- *	ata_altstatus - Read device alternate status reg
+ *	ata_sff_altstatus - Read device alternate status reg
  *	@ap: port where the device is
  *
  *	Reads ATA taskfile alternate status register for
@@ -243,7 +243,7 @@ u8 ata_check_status(struct ata_port *ap)
  *	LOCKING:
  *	Inherited from caller.
  */
-u8 ata_altstatus(struct ata_port *ap)
+u8 ata_sff_altstatus(struct ata_port *ap)
 {
 	if (ap->ops->check_altstatus)
 		return ap->ops->check_altstatus(ap);
@@ -252,7 +252,7 @@ u8 ata_altstatus(struct ata_port *ap)
 }
 
 /**
- *	ata_busy_sleep - sleep until BSY clears, or timeout
+ *	ata_sff_busy_sleep - sleep until BSY clears, or timeout
  *	@ap: port containing status register to be polled
  *	@tmout_pat: impatience timeout
  *	@tmout: overall timeout
@@ -266,19 +266,19 @@ u8 ata_altstatus(struct ata_port *ap)
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_busy_sleep(struct ata_port *ap,
-		   unsigned long tmout_pat, unsigned long tmout)
+int ata_sff_busy_sleep(struct ata_port *ap,
+		       unsigned long tmout_pat, unsigned long tmout)
 {
 	unsigned long timer_start, timeout;
 	u8 status;
 
-	status = ata_busy_wait(ap, ATA_BUSY, 300);
+	status = ata_sff_busy_wait(ap, ATA_BUSY, 300);
 	timer_start = jiffies;
 	timeout = timer_start + tmout_pat;
 	while (status != 0xff && (status & ATA_BUSY) &&
 	       time_before(jiffies, timeout)) {
 		msleep(50);
-		status = ata_busy_wait(ap, ATA_BUSY, 3);
+		status = ata_sff_busy_wait(ap, ATA_BUSY, 3);
 	}
 
 	if (status != 0xff && (status & ATA_BUSY))
@@ -307,7 +307,7 @@ int ata_busy_sleep(struct ata_port *ap,
 }
 
 /**
- *	ata_wait_ready - sleep until BSY clears, or timeout
+ *	ata_sff_wait_ready - sleep until BSY clears, or timeout
  *	@ap: port containing status register to be polled
  *	@deadline: deadline jiffies for the operation
  *
@@ -320,7 +320,7 @@ int ata_busy_sleep(struct ata_port *ap,
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_wait_ready(struct ata_port *ap, unsigned long deadline)
+int ata_sff_wait_ready(struct ata_port *ap, unsigned long deadline)
 {
 	unsigned long start = jiffies;
 	int warned = 0;
@@ -349,7 +349,7 @@ int ata_wait_ready(struct ata_port *ap, unsigned long deadline)
 }
 
 /**
- *	ata_std_dev_select - Select device 0/1 on ATA bus
+ *	ata_sff_dev_select - Select device 0/1 on ATA bus
  *	@ap: ATA channel to manipulate
  *	@device: ATA device (numbered from zero) to select
  *
@@ -362,7 +362,7 @@ int ata_wait_ready(struct ata_port *ap, unsigned long deadline)
  *	LOCKING:
  *	caller.
  */
-void ata_std_dev_select(struct ata_port *ap, unsigned int device)
+void ata_sff_dev_select(struct ata_port *ap, unsigned int device)
 {
 	u8 tmp;
 
@@ -372,7 +372,7 @@ void ata_std_dev_select(struct ata_port *ap, unsigned int device)
 		tmp = ATA_DEVICE_OBS | ATA_DEV1;
 
 	iowrite8(tmp, ap->ioaddr.device_addr);
-	ata_pause(ap);		/* needed; also flushes, for mmio */
+	ata_sff_pause(ap);	/* needed; also flushes, for mmio */
 }
 
 /**
@@ -386,9 +386,9 @@ void ata_std_dev_select(struct ata_port *ap, unsigned int device)
  *	make either device 0, or device 1, active on the
  *	ATA channel.
  *
- *	This is a high-level version of ata_std_dev_select(),
- *	which additionally provides the services of inserting
- *	the proper pauses and status polling, where needed.
+ *	This is a high-level version of ata_sff_dev_select(), which
+ *	additionally provides the services of inserting the proper
+ *	pauses and status polling, where needed.
  *
  *	LOCKING:
  *	caller.
@@ -413,7 +413,7 @@ void ata_dev_select(struct ata_port *ap, unsigned int device,
 }
 
 /**
- *	ata_irq_on - Enable interrupts on a port.
+ *	ata_sff_irq_on - Enable interrupts on a port.
  *	@ap: Port on which interrupts are enabled.
  *
  *	Enable interrupts on a legacy IDE device using MMIO or PIO,
@@ -422,7 +422,7 @@ void ata_dev_select(struct ata_port *ap, unsigned int device,
  *	LOCKING:
  *	Inherited from caller.
  */
-u8 ata_irq_on(struct ata_port *ap)
+u8 ata_sff_irq_on(struct ata_port *ap)
 {
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 	u8 tmp;
@@ -440,7 +440,7 @@ u8 ata_irq_on(struct ata_port *ap)
 }
 
 /**
- *	ata_bmdma_irq_clear - Clear PCI IDE BMDMA interrupt.
+ *	ata_sff_irq_clear - Clear PCI IDE BMDMA interrupt.
  *	@ap: Port associated with this ATA transaction.
  *
  *	Clear interrupt and error flags in DMA status register.
@@ -450,7 +450,7 @@ u8 ata_irq_on(struct ata_port *ap)
  *	LOCKING:
  *	spin_lock_irqsave(host lock)
  */
-void ata_bmdma_irq_clear(struct ata_port *ap)
+void ata_sff_irq_clear(struct ata_port *ap)
 {
 	void __iomem *mmio = ap->ioaddr.bmdma_addr;
 
@@ -461,7 +461,7 @@ void ata_bmdma_irq_clear(struct ata_port *ap)
 }
 
 /**
- *	ata_tf_load - send taskfile registers to host controller
+ *	ata_sff_tf_load - send taskfile registers to host controller
  *	@ap: Port to which output is sent
  *	@tf: ATA taskfile register set
  *
@@ -470,7 +470,7 @@ void ata_bmdma_irq_clear(struct ata_port *ap)
  *	LOCKING:
  *	Inherited from caller.
  */
-void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
+void ata_sff_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
 {
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 	unsigned int is_addr = tf->flags & ATA_TFLAG_ISADDR;
@@ -520,7 +520,7 @@ void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
 }
 
 /**
- *	ata_tf_read - input device's ATA taskfile shadow registers
+ *	ata_sff_tf_read - input device's ATA taskfile shadow registers
  *	@ap: Port from which input is read
  *	@tf: ATA taskfile register set for storing input
  *
@@ -532,11 +532,11 @@ void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
  *	LOCKING:
  *	Inherited from caller.
  */
-void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
+void ata_sff_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
 {
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 
-	tf->command = ata_check_status(ap);
+	tf->command = ata_sff_check_status(ap);
 	tf->feature = ioread8(ioaddr->error_addr);
 	tf->nsect = ioread8(ioaddr->nsect_addr);
 	tf->lbal = ioread8(ioaddr->lbal_addr);
@@ -560,7 +560,7 @@ void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
 }
 
 /**
- *	ata_exec_command - issue ATA command to host controller
+ *	ata_sff_exec_command - issue ATA command to host controller
  *	@ap: port to which command is being issued
  *	@tf: ATA taskfile register set
  *
@@ -570,12 +570,12 @@ void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
  *	LOCKING:
  *	spin_lock_irqsave(host lock)
  */
-void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf)
+void ata_sff_exec_command(struct ata_port *ap, const struct ata_taskfile *tf)
 {
 	DPRINTK("ata%u: cmd 0x%X\n", ap->print_id, tf->command);
 
 	iowrite8(tf->command, ap->ioaddr.command_addr);
-	ata_pause(ap);
+	ata_sff_pause(ap);
 }
 
 /**
@@ -598,7 +598,7 @@ static inline void ata_tf_to_host(struct ata_port *ap,
 }
 
 /**
- *	ata_data_xfer - Transfer data by PIO
+ *	ata_sff_data_xfer - Transfer data by PIO
  *	@dev: device to target
  *	@buf: data buffer
  *	@buflen: buffer length
@@ -612,8 +612,8 @@ static inline void ata_tf_to_host(struct ata_port *ap,
  *	RETURNS:
  *	Bytes consumed.
  */
-unsigned int ata_data_xfer(struct ata_device *dev, unsigned char *buf,
-			   unsigned int buflen, int rw)
+unsigned int ata_sff_data_xfer(struct ata_device *dev, unsigned char *buf,
+			       unsigned int buflen, int rw)
 {
 	struct ata_port *ap = dev->link->ap;
 	void __iomem *data_addr = ap->ioaddr.data_addr;
@@ -644,7 +644,7 @@ unsigned int ata_data_xfer(struct ata_device *dev, unsigned char *buf,
 }
 
 /**
- *	ata_data_xfer_noirq - Transfer data by PIO
+ *	ata_sff_data_xfer_noirq - Transfer data by PIO
  *	@dev: device to target
  *	@buf: data buffer
  *	@buflen: buffer length
@@ -659,14 +659,14 @@ unsigned int ata_data_xfer(struct ata_device *dev, unsigned char *buf,
  *	RETURNS:
  *	Bytes consumed.
  */
-unsigned int ata_data_xfer_noirq(struct ata_device *dev, unsigned char *buf,
-				 unsigned int buflen, int rw)
+unsigned int ata_sff_data_xfer_noirq(struct ata_device *dev, unsigned char *buf,
+				     unsigned int buflen, int rw)
 {
 	unsigned long flags;
 	unsigned int consumed;
 
 	local_irq_save(flags);
-	consumed = ata_data_xfer(dev, buf, buflen, rw);
+	consumed = ata_sff_data_xfer(dev, buf, buflen, rw);
 	local_irq_restore(flags);
 
 	return consumed;
@@ -752,7 +752,7 @@ static void ata_pio_sectors(struct ata_queued_cmd *qc)
 	} else
 		ata_pio_sector(qc);
 
-	ata_altstatus(qc->ap); /* flush */
+	ata_sff_altstatus(qc->ap); /* flush */
 }
 
 /**
@@ -773,7 +773,7 @@ static void atapi_send_cdb(struct ata_port *ap, struct ata_queued_cmd *qc)
 	WARN_ON(qc->dev->cdb_len < 12);
 
 	ap->ops->data_xfer(qc->dev, qc->cdb, qc->dev->cdb_len, 1);
-	ata_altstatus(ap); /* flush */
+	ata_sff_altstatus(ap); /* flush */
 
 	switch (qc->tf.protocol) {
 	case ATAPI_PROT_PIO:
@@ -915,7 +915,7 @@ static void atapi_pio_bytes(struct ata_queued_cmd *qc)
 
 	if (unlikely(__atapi_pio_bytes(qc, bytes)))
 		goto err_out;
-	ata_altstatus(ap); /* flush */
+	ata_sff_altstatus(ap); /* flush */
 
 	return;
 
@@ -1004,7 +1004,7 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
 }
 
 /**
- *	ata_hsm_move - move the HSM to the next state.
+ *	ata_sff_hsm_move - move the HSM to the next state.
  *	@ap: the target ata_port
  *	@qc: qc on going
  *	@status: current device status
@@ -1013,15 +1013,15 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
  *	RETURNS:
  *	1 when poll next status needed, 0 otherwise.
  */
-int ata_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
-		 u8 status, int in_wq)
+int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
+		     u8 status, int in_wq)
 {
 	unsigned long flags = 0;
 	int poll_next;
 
 	WARN_ON((qc->flags & ATA_QCFLAG_ACTIVE) == 0);
 
-	/* Make sure ata_qc_issue_prot() does not throw things
+	/* Make sure ata_sff_qc_issue() does not throw things
 	 * like DMA polling into the workqueue. Notice that
 	 * in_wq is not equivalent to (qc->tf.flags & ATA_TFLAG_POLLING).
 	 */
@@ -1263,10 +1263,10 @@ fsm_start:
 	 * or something.  Snooze for a couple msecs, then
 	 * chk-status again.  If still busy, queue delayed work.
 	 */
-	status = ata_busy_wait(ap, ATA_BUSY, 5);
+	status = ata_sff_busy_wait(ap, ATA_BUSY, 5);
 	if (status & ATA_BUSY) {
 		msleep(2);
-		status = ata_busy_wait(ap, ATA_BUSY, 10);
+		status = ata_sff_busy_wait(ap, ATA_BUSY, 10);
 		if (status & ATA_BUSY) {
 			ata_pio_queue_task(ap, qc, ATA_SHORT_PAUSE);
 			return;
@@ -1274,7 +1274,7 @@ fsm_start:
 	}
 
 	/* move the HSM */
-	poll_next = ata_hsm_move(ap, qc, status, 1);
+	poll_next = ata_sff_hsm_move(ap, qc, status, 1);
 
 	/* another command or interrupt handler
 	 * may be running at this point.
@@ -1284,7 +1284,7 @@ fsm_start:
 }
 
 /**
- *	ata_qc_issue_prot - issue taskfile to device in proto-dependent manner
+ *	ata_sff_qc_issue - issue taskfile to device in proto-dependent manner
  *	@qc: command to issue to device
  *
  *	Using various libata functions and hooks, this function
@@ -1300,7 +1300,7 @@ fsm_start:
  *	RETURNS:
  *	Zero on success, AC_ERR_* mask on failure
  */
-unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
+unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 
@@ -1415,7 +1415,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
 }
 
 /**
- *	ata_host_intr - Handle host interrupt for given (port, task)
+ *	ata_sff_host_intr - Handle host interrupt for given (port, task)
  *	@ap: Port on which interrupt arrived (possibly...)
  *	@qc: Taskfile currently active in engine
  *
@@ -1429,8 +1429,8 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
  *	RETURNS:
  *	One if interrupt was handled, zero if not (shared irq).
  */
-inline unsigned int ata_host_intr(struct ata_port *ap,
-				  struct ata_queued_cmd *qc)
+inline unsigned int ata_sff_host_intr(struct ata_port *ap,
+				      struct ata_queued_cmd *qc)
 {
 	struct ata_eh_info *ehi = &ap->link.eh_info;
 	u8 status, host_stat = 0;
@@ -1481,7 +1481,7 @@ inline unsigned int ata_host_intr(struct ata_port *ap,
 	}
 
 	/* check altstatus */
-	status = ata_altstatus(ap);
+	status = ata_sff_altstatus(ap);
 	if (status & ATA_BUSY)
 		goto idle_irq;
 
@@ -1493,7 +1493,7 @@ inline unsigned int ata_host_intr(struct ata_port *ap,
 	/* ack bmdma irq events */
 	ap->ops->irq_clear(ap);
 
-	ata_hsm_move(ap, qc, status, 0);
+	ata_sff_hsm_move(ap, qc, status, 0);
 
 	if (unlikely(qc->err_mask) && (qc->tf.protocol == ATA_PROT_DMA ||
 				       qc->tf.protocol == ATAPI_PROT_DMA))
@@ -1516,12 +1516,12 @@ idle_irq:
 }
 
 /**
- *	ata_interrupt - Default ATA host interrupt handler
+ *	ata_sff_interrupt - Default ATA host interrupt handler
  *	@irq: irq line (unused)
  *	@dev_instance: pointer to our ata_host information structure
  *
  *	Default interrupt handler for PCI IDE devices.  Calls
- *	ata_host_intr() for each port that is not disabled.
+ *	ata_sff_host_intr() for each port that is not disabled.
  *
  *	LOCKING:
  *	Obtains host lock during operation.
@@ -1529,7 +1529,7 @@ idle_irq:
  *	RETURNS:
  *	IRQ_NONE or IRQ_HANDLED.
  */
-irqreturn_t ata_interrupt(int irq, void *dev_instance)
+irqreturn_t ata_sff_interrupt(int irq, void *dev_instance)
 {
 	struct ata_host *host = dev_instance;
 	unsigned int i;
@@ -1550,7 +1550,7 @@ irqreturn_t ata_interrupt(int irq, void *dev_instance)
 			qc = ata_qc_from_tag(ap, ap->link.active_tag);
 			if (qc && (!(qc->tf.flags & ATA_TFLAG_POLLING)) &&
 			    (qc->flags & ATA_QCFLAG_ACTIVE))
-				handled |= ata_host_intr(ap, qc);
+				handled |= ata_sff_host_intr(ap, qc);
 		}
 	}
 
@@ -1560,7 +1560,7 @@ irqreturn_t ata_interrupt(int irq, void *dev_instance)
 }
 
 /**
- *	ata_bmdma_freeze - Freeze BMDMA controller port
+ *	ata_sff_freeze - Freeze SFF controller port
  *	@ap: port to freeze
  *
  *	Freeze BMDMA controller port.
@@ -1568,7 +1568,7 @@ irqreturn_t ata_interrupt(int irq, void *dev_instance)
  *	LOCKING:
  *	Inherited from caller.
  */
-void ata_bmdma_freeze(struct ata_port *ap)
+void ata_sff_freeze(struct ata_port *ap)
 {
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 
@@ -1588,15 +1588,15 @@ void ata_bmdma_freeze(struct ata_port *ap)
 }
 
 /**
- *	ata_bmdma_thaw - Thaw BMDMA controller port
+ *	ata_sff_thaw - Thaw SFF controller port
  *	@ap: port to thaw
  *
- *	Thaw BMDMA controller port.
+ *	Thaw SFF controller port.
  *
  *	LOCKING:
  *	Inherited from caller.
  */
-void ata_bmdma_thaw(struct ata_port *ap)
+void ata_sff_thaw(struct ata_port *ap)
 {
 	/* clear & re-enable interrupts */
 	ap->ops->check_status(ap);
@@ -1647,7 +1647,7 @@ static unsigned int ata_devchk(struct ata_port *ap, unsigned int device)
 }
 
 /**
- *	ata_dev_try_classify - Parse returned ATA device signature
+ *	ata_sff_dev_classify - Parse returned ATA device signature
  *	@dev: ATA device to classify (starting at zero)
  *	@present: device seems present
  *	@r_err: Value of error register on completion
@@ -1667,7 +1667,7 @@ static unsigned int ata_devchk(struct ata_port *ap, unsigned int device)
  *	RETURNS:
  *	Device type - %ATA_DEV_ATA, %ATA_DEV_ATAPI or %ATA_DEV_NONE.
  */
-unsigned int ata_dev_try_classify(struct ata_device *dev, int present,
+unsigned int ata_sff_dev_classify(struct ata_device *dev, int present,
 				  u8 *r_err)
 {
 	struct ata_port *ap = dev->link->ap;
@@ -1727,7 +1727,7 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	 * BSY bit to clear
 	 */
 	if (dev0) {
-		rc = ata_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(ap, deadline);
 		if (rc) {
 			if (rc != -ENODEV)
 				return rc;
@@ -1757,7 +1757,7 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 			msleep(50);	/* give drive a breather */
 		}
 
-		rc = ata_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(ap, deadline);
 		if (rc) {
 			if (rc != -ENODEV)
 				return rc;
@@ -1776,7 +1776,7 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 }
 
 /**
- *	ata_wait_after_reset - wait before checking status after reset
+ *	ata_sff_wait_after_reset - wait before checking status after reset
  *	@ap: port containing status register to be polled
  *	@deadline: deadline jiffies for the operation
  *
@@ -1790,7 +1790,7 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
  *	LOCKING:
  *	Kernel thread context (may sleep).
  */
-void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline)
+void ata_sff_wait_after_reset(struct ata_port *ap, unsigned long deadline)
 {
 	unsigned long until = jiffies + ATA_TMOUT_FF_WAIT;
 
@@ -1845,7 +1845,7 @@ static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
 	iowrite8(ap->ctl, ioaddr->ctl_addr);
 
 	/* wait a while before checking status */
-	ata_wait_after_reset(ap, deadline);
+	ata_sff_wait_after_reset(ap, deadline);
 
 	/* Before we perform post reset processing we want to see if
 	 * the bus shows 0xFF because the odd clown forgets the D7
@@ -1858,7 +1858,7 @@ static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
 }
 
 /**
- *	ata_std_softreset - reset host port via ATA SRST
+ *	ata_sff_softreset - reset host port via ATA SRST
  *	@link: ATA link to reset
  *	@classes: resulting classes of attached devices
  *	@deadline: deadline jiffies for the operation
@@ -1871,7 +1871,7 @@ static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_std_softreset(struct ata_link *link, unsigned int *classes,
+int ata_sff_softreset(struct ata_link *link, unsigned int *classes,
 		      unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
@@ -1906,10 +1906,10 @@ int ata_std_softreset(struct ata_link *link, unsigned int *classes,
 	}
 
 	/* determine by signature whether we have ATA or ATAPI devices */
-	classes[0] = ata_dev_try_classify(&link->device[0],
+	classes[0] = ata_sff_dev_classify(&link->device[0],
 					  devmask & (1 << 0), &err);
 	if (slave_possible && err != 0x81)
-		classes[1] = ata_dev_try_classify(&link->device[1],
+		classes[1] = ata_sff_dev_classify(&link->device[1],
 						  devmask & (1 << 1), &err);
 
  out:
@@ -1918,7 +1918,7 @@ int ata_std_softreset(struct ata_link *link, unsigned int *classes,
 }
 
 /**
- *	sata_std_hardreset - reset host port via SATA phy reset
+ *	sata_sff_hardreset - reset host port via SATA phy reset
  *	@link: link to reset
  *	@class: resulting class of attached device
  *	@deadline: deadline jiffies for the operation
@@ -1932,7 +1932,7 @@ int ata_std_softreset(struct ata_link *link, unsigned int *classes,
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int sata_std_hardreset(struct ata_link *link, unsigned int *class,
+int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
 		       unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
@@ -1957,7 +1957,7 @@ int sata_std_hardreset(struct ata_link *link, unsigned int *class,
 	}
 
 	/* wait a while before checking status */
-	ata_wait_after_reset(ap, deadline);
+	ata_sff_wait_after_reset(ap, deadline);
 
 	/* If PMP is supported, we have to do follow-up SRST.  Note
 	 * that some PMPs don't send D2H Reg FIS after hardreset at
@@ -1965,11 +1965,11 @@ int sata_std_hardreset(struct ata_link *link, unsigned int *class,
 	 * second and request follow-up SRST.
 	 */
 	if (ap->flags & ATA_FLAG_PMP) {
-		ata_wait_ready(ap, jiffies + HZ);
+		ata_sff_wait_ready(ap, jiffies + HZ);
 		return -EAGAIN;
 	}
 
-	rc = ata_wait_ready(ap, deadline);
+	rc = ata_sff_wait_ready(ap, deadline);
 	/* link occupied, -ENODEV too is an error */
 	if (rc) {
 		ata_link_printk(link, KERN_ERR,
@@ -1979,17 +1979,17 @@ int sata_std_hardreset(struct ata_link *link, unsigned int *class,
 
 	ap->ops->dev_select(ap, 0);	/* probably unnecessary */
 
-	*class = ata_dev_try_classify(link->device, 1, NULL);
+	*class = ata_sff_dev_classify(link->device, 1, NULL);
 
 	DPRINTK("EXIT, class=%u\n", *class);
 	return 0;
 }
 
 /**
- *	ata_bmdma_error_handler - Stock error handler for BMDMA controller
+ *	ata_sff_error_handler - Stock error handler for BMDMA controller
  *	@ap: port to handle error for
  *
- *	Stock error handler for BMDMA controller.  It can handle both
+ *	Stock error handler for SFF controller.  It can handle both
  *	PATA and SATA controllers.  Many controllers should be able to
  *	use this EH as-is or with some added handling before and
  *	after.
@@ -1997,7 +1997,7 @@ int sata_std_hardreset(struct ata_link *link, unsigned int *class,
  *	LOCKING:
  *	Kernel thread context (may sleep)
  */
-void ata_bmdma_error_handler(struct ata_port *ap)
+void ata_sff_error_handler(struct ata_port *ap)
 {
 	ata_reset_fn_t softreset = ap->ops->softreset;
 	ata_reset_fn_t hardreset = ap->ops->hardreset;
@@ -2034,7 +2034,7 @@ void ata_bmdma_error_handler(struct ata_port *ap)
 		ap->ops->bmdma_stop(qc);
 	}
 
-	ata_altstatus(ap);
+	ata_sff_altstatus(ap);
 	ap->ops->check_status(ap);
 	ap->ops->irq_clear(ap);
 
@@ -2045,14 +2045,14 @@ void ata_bmdma_error_handler(struct ata_port *ap)
 
 	/* PIO and DMA engines have been stopped, perform recovery */
 
-	/* ata_std_softreset and sata_std_hardreset are inherited to
+	/* ata_sff_softreset and sata_sff_hardreset are inherited to
 	 * all SFF drivers from ata_sff_port_ops.  Ignore softreset if
 	 * ctl isn't accessible.  Ignore hardreset if SCR access isn't
 	 * available.
 	 */
-	if (softreset == ata_std_softreset && !ap->ioaddr.ctl_addr)
+	if (softreset == ata_sff_softreset && !ap->ioaddr.ctl_addr)
 		softreset = NULL;
-	if (hardreset == sata_std_hardreset && !sata_scr_valid(&ap->link))
+	if (hardreset == sata_sff_hardreset && !sata_scr_valid(&ap->link))
 		hardreset = NULL;
 
 	ata_do_eh(ap, ap->ops->prereset, softreset, hardreset,
@@ -2060,14 +2060,13 @@ void ata_bmdma_error_handler(struct ata_port *ap)
 }
 
 /**
- *	ata_bmdma_post_internal_cmd - Stock post_internal_cmd for
- *				      BMDMA controller
+ *	ata_sff_post_internal_cmd - Stock post_internal_cmd for SFF controller
  *	@qc: internal command to clean up
  *
  *	LOCKING:
  *	Kernel thread context (may sleep)
  */
-void ata_bmdma_post_internal_cmd(struct ata_queued_cmd *qc)
+void ata_sff_post_internal_cmd(struct ata_queued_cmd *qc)
 {
 	if (qc->ap->ioaddr.bmdma_addr)
 		ata_bmdma_stop(qc);
@@ -2094,7 +2093,7 @@ int ata_sff_port_start(struct ata_port *ap)
 }
 
 /**
- *	ata_std_ports - initialize ioaddr with standard port offsets.
+ *	ata_sff_std_ports - initialize ioaddr with standard port offsets.
  *	@ioaddr: IO address structure to be initialized
  *
  *	Utility function which initializes data_addr, error_addr,
@@ -2104,7 +2103,7 @@ int ata_sff_port_start(struct ata_port *ap)
  *
  *	Does not set ctl_addr, altstatus_addr, bmdma_addr, or scr_addr.
  */
-void ata_std_ports(struct ata_ioports *ioaddr)
+void ata_sff_std_ports(struct ata_ioports *ioaddr)
 {
 	ioaddr->data_addr = ioaddr->cmd_addr + ATA_REG_DATA;
 	ioaddr->error_addr = ioaddr->cmd_addr + ATA_REG_ERR;
@@ -2118,7 +2117,8 @@ void ata_std_ports(struct ata_ioports *ioaddr)
 	ioaddr->command_addr = ioaddr->cmd_addr + ATA_REG_CMD;
 }
 
-unsigned long ata_pci_default_filter(struct ata_device *adev, unsigned long xfer_mask)
+unsigned long ata_bmdma_mode_filter(struct ata_device *adev,
+				    unsigned long xfer_mask)
 {
 	/* Filter out DMA modes if the device has been configured by
 	   the BIOS as PIO only */
@@ -2209,7 +2209,7 @@ void ata_bmdma_stop(struct ata_queued_cmd *qc)
 		 mmio + ATA_DMA_CMD);
 
 	/* one-PIO-cycle guaranteed wait, per spec, for HDMA1:0 transition */
-	ata_altstatus(ap);        /* dummy read */
+	ata_sff_altstatus(ap);        /* dummy read */
 }
 
 /**
@@ -2289,9 +2289,9 @@ void ata_bus_reset(struct ata_port *ap)
 	/*
 	 * determine by signature whether we have ATA or ATAPI devices
 	 */
-	device[0].class = ata_dev_try_classify(&device[0], dev0, &err);
+	device[0].class = ata_sff_dev_classify(&device[0], dev0, &err);
 	if ((slave_possible) && (err != 0x81))
-		device[1].class = ata_dev_try_classify(&device[1], dev1, &err);
+		device[1].class = ata_sff_dev_classify(&device[1], dev1, &err);
 
 	/* is double-select really necessary? */
 	if (device[1].class != ATA_DEV_NONE)
@@ -2322,7 +2322,7 @@ err_out:
 #ifdef CONFIG_PCI
 
 /**
- *	ata_pci_clear_simplex	-	attempt to kick device out of simplex
+ *	ata_pci_bmdma_clear_simplex -	attempt to kick device out of simplex
  *	@pdev: PCI device
  *
  *	Some PCI ATA devices report simplex mode but in fact can be told to
@@ -2330,7 +2330,7 @@ err_out:
  *	perform the task on such devices. Calling it on other devices will
  *	have -undefined- behaviour.
  */
-int ata_pci_clear_simplex(struct pci_dev *pdev)
+int ata_pci_bmdma_clear_simplex(struct pci_dev *pdev)
 {
 	unsigned long bmdma = pci_resource_start(pdev, 4);
 	u8 simplex;
@@ -2347,7 +2347,7 @@ int ata_pci_clear_simplex(struct pci_dev *pdev)
 }
 
 /**
- *	ata_pci_init_bmdma - acquire PCI BMDMA resources and init ATA host
+ *	ata_pci_bmdma_init - acquire PCI BMDMA resources and init ATA host
  *	@host: target ATA host
  *
  *	Acquire PCI BMDMA resources and initialize @host accordingly.
@@ -2358,7 +2358,7 @@ int ata_pci_clear_simplex(struct pci_dev *pdev)
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_pci_init_bmdma(struct ata_host *host)
+int ata_pci_bmdma_init(struct ata_host *host)
 {
 	struct device *gdev = host->dev;
 	struct pci_dev *pdev = to_pci_dev(gdev);
@@ -2418,7 +2418,7 @@ static int ata_resources_present(struct pci_dev *pdev, int port)
 }
 
 /**
- *	ata_pci_init_sff_host - acquire native PCI ATA resources and init host
+ *	ata_pci_sff_init_host - acquire native PCI ATA resources and init host
  *	@host: target ATA host
  *
  *	Acquire native PCI ATA resources for @host and initialize the
@@ -2436,7 +2436,7 @@ static int ata_resources_present(struct pci_dev *pdev, int port)
  *	0 if at least one port is initialized, -ENODEV if no port is
  *	available.
  */
-int ata_pci_init_sff_host(struct ata_host *host)
+int ata_pci_sff_init_host(struct ata_host *host)
 {
 	struct device *gdev = host->dev;
 	struct pci_dev *pdev = to_pci_dev(gdev);
@@ -2478,7 +2478,7 @@ int ata_pci_init_sff_host(struct ata_host *host)
 		ap->ioaddr.altstatus_addr =
 		ap->ioaddr.ctl_addr = (void __iomem *)
 			((unsigned long)iomap[base + 1] | ATA_PCI_CTL_OFS);
-		ata_std_ports(&ap->ioaddr);
+		ata_sff_std_ports(&ap->ioaddr);
 
 		ata_port_desc(ap, "cmd 0x%llx ctl 0x%llx",
 			(unsigned long long)pci_resource_start(pdev, base),
@@ -2496,7 +2496,7 @@ int ata_pci_init_sff_host(struct ata_host *host)
 }
 
 /**
- *	ata_pci_prepare_sff_host - helper to prepare native PCI ATA host
+ *	ata_pci_sff_prepare_host - helper to prepare native PCI ATA host
  *	@pdev: target PCI device
  *	@ppi: array of port_info, must be enough for two ports
  *	@r_host: out argument for the initialized ATA host
@@ -2510,7 +2510,7 @@ int ata_pci_init_sff_host(struct ata_host *host)
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_pci_prepare_sff_host(struct pci_dev *pdev,
+int ata_pci_sff_prepare_host(struct pci_dev *pdev,
 			     const struct ata_port_info * const * ppi,
 			     struct ata_host **r_host)
 {
@@ -2528,12 +2528,12 @@ int ata_pci_prepare_sff_host(struct pci_dev *pdev,
 		goto err_out;
 	}
 
-	rc = ata_pci_init_sff_host(host);
+	rc = ata_pci_sff_init_host(host);
 	if (rc)
 		goto err_out;
 
 	/* init DMA related stuff */
-	rc = ata_pci_init_bmdma(host);
+	rc = ata_pci_bmdma_init(host);
 	if (rc)
 		goto err_bmdma;
 
@@ -2554,7 +2554,7 @@ int ata_pci_prepare_sff_host(struct pci_dev *pdev,
 }
 
 /**
- *	ata_pci_activate_sff_host - start SFF host, request IRQ and register it
+ *	ata_pci_sff_activate_host - start SFF host, request IRQ and register it
  *	@host: target SFF ATA host
  *	@irq_handler: irq_handler used when requesting IRQ(s)
  *	@sht: scsi_host_template to use when registering the host
@@ -2569,7 +2569,7 @@ int ata_pci_prepare_sff_host(struct pci_dev *pdev,
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_pci_activate_sff_host(struct ata_host *host,
+int ata_pci_sff_activate_host(struct ata_host *host,
 			      irq_handler_t irq_handler,
 			      struct scsi_host_template *sht)
 {
@@ -2647,7 +2647,7 @@ int ata_pci_activate_sff_host(struct ata_host *host,
 }
 
 /**
- *	ata_pci_init_one - Initialize/register PCI IDE host controller
+ *	ata_pci_sff_init_one - Initialize/register PCI IDE host controller
  *	@pdev: Controller to be initialized
  *	@ppi: array of port_info, must be enough for two ports
  *	@sht: scsi_host_template to use when registering the host
@@ -2671,9 +2671,9 @@ int ata_pci_activate_sff_host(struct ata_host *host,
  *	RETURNS:
  *	Zero on success, negative on errno-based value on error.
  */
-int ata_pci_init_one(struct pci_dev *pdev,
-		     const struct ata_port_info * const * ppi,
-		     struct scsi_host_template *sht, void *host_priv)
+int ata_pci_sff_init_one(struct pci_dev *pdev,
+			 const struct ata_port_info * const * ppi,
+			 struct scsi_host_template *sht, void *host_priv)
 {
 	struct device *dev = &pdev->dev;
 	const struct ata_port_info *pi = NULL;
@@ -2704,13 +2704,13 @@ int ata_pci_init_one(struct pci_dev *pdev,
 		goto out;
 
 	/* prepare and activate SFF host */
-	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
+	rc = ata_pci_sff_prepare_host(pdev, ppi, &host);
 	if (rc)
 		goto out;
 	host->private_data = host_priv;
 
 	pci_set_master(pdev);
-	rc = ata_pci_activate_sff_host(host, ata_interrupt, sht);
+	rc = ata_pci_sff_activate_host(host, ata_sff_interrupt, sht);
  out:
 	if (rc == 0)
 		devres_remove_group(&pdev->dev, NULL);
@@ -2724,47 +2724,47 @@ int ata_pci_init_one(struct pci_dev *pdev,
 
 EXPORT_SYMBOL_GPL(ata_sff_port_ops);
 EXPORT_SYMBOL_GPL(ata_bmdma_port_ops);
-EXPORT_SYMBOL_GPL(ata_qc_prep);
-EXPORT_SYMBOL_GPL(ata_dumb_qc_prep);
-EXPORT_SYMBOL_GPL(ata_std_dev_select);
-EXPORT_SYMBOL_GPL(ata_check_status);
-EXPORT_SYMBOL_GPL(ata_altstatus);
-EXPORT_SYMBOL_GPL(ata_busy_sleep);
-EXPORT_SYMBOL_GPL(ata_wait_ready);
-EXPORT_SYMBOL_GPL(ata_tf_load);
-EXPORT_SYMBOL_GPL(ata_tf_read);
-EXPORT_SYMBOL_GPL(ata_exec_command);
-EXPORT_SYMBOL_GPL(ata_data_xfer);
-EXPORT_SYMBOL_GPL(ata_data_xfer_noirq);
-EXPORT_SYMBOL_GPL(ata_irq_on);
-EXPORT_SYMBOL_GPL(ata_bmdma_irq_clear);
-EXPORT_SYMBOL_GPL(ata_hsm_move);
-EXPORT_SYMBOL_GPL(ata_qc_issue_prot);
-EXPORT_SYMBOL_GPL(ata_host_intr);
-EXPORT_SYMBOL_GPL(ata_interrupt);
-EXPORT_SYMBOL_GPL(ata_bmdma_freeze);
-EXPORT_SYMBOL_GPL(ata_bmdma_thaw);
-EXPORT_SYMBOL_GPL(ata_std_prereset);
-EXPORT_SYMBOL_GPL(ata_dev_try_classify);
-EXPORT_SYMBOL_GPL(ata_wait_after_reset);
-EXPORT_SYMBOL_GPL(ata_std_softreset);
-EXPORT_SYMBOL_GPL(sata_std_hardreset);
-EXPORT_SYMBOL_GPL(ata_std_postreset);
-EXPORT_SYMBOL_GPL(ata_bmdma_error_handler);
-EXPORT_SYMBOL_GPL(ata_bmdma_post_internal_cmd);
+EXPORT_SYMBOL_GPL(ata_sff_qc_prep);
+EXPORT_SYMBOL_GPL(ata_sff_dumb_qc_prep);
+EXPORT_SYMBOL_GPL(ata_sff_dev_select);
+EXPORT_SYMBOL_GPL(ata_sff_check_status);
+EXPORT_SYMBOL_GPL(ata_sff_altstatus);
+EXPORT_SYMBOL_GPL(ata_sff_busy_sleep);
+EXPORT_SYMBOL_GPL(ata_sff_wait_ready);
+EXPORT_SYMBOL_GPL(ata_sff_tf_load);
+EXPORT_SYMBOL_GPL(ata_sff_tf_read);
+EXPORT_SYMBOL_GPL(ata_sff_exec_command);
+EXPORT_SYMBOL_GPL(ata_sff_data_xfer);
+EXPORT_SYMBOL_GPL(ata_sff_data_xfer_noirq);
+EXPORT_SYMBOL_GPL(ata_sff_irq_on);
+EXPORT_SYMBOL_GPL(ata_sff_irq_clear);
+EXPORT_SYMBOL_GPL(ata_sff_hsm_move);
+EXPORT_SYMBOL_GPL(ata_sff_qc_issue);
+EXPORT_SYMBOL_GPL(ata_sff_host_intr);
+EXPORT_SYMBOL_GPL(ata_sff_interrupt);
+EXPORT_SYMBOL_GPL(ata_sff_freeze);
+EXPORT_SYMBOL_GPL(ata_sff_thaw);
+EXPORT_SYMBOL_GPL(ata_sff_prereset);
+EXPORT_SYMBOL_GPL(ata_sff_dev_classify);
+EXPORT_SYMBOL_GPL(ata_sff_wait_after_reset);
+EXPORT_SYMBOL_GPL(ata_sff_softreset);
+EXPORT_SYMBOL_GPL(sata_sff_hardreset);
+EXPORT_SYMBOL_GPL(ata_sff_postreset);
+EXPORT_SYMBOL_GPL(ata_sff_error_handler);
+EXPORT_SYMBOL_GPL(ata_sff_post_internal_cmd);
 EXPORT_SYMBOL_GPL(ata_sff_port_start);
-EXPORT_SYMBOL_GPL(ata_std_ports);
-EXPORT_SYMBOL_GPL(ata_pci_default_filter);
+EXPORT_SYMBOL_GPL(ata_sff_std_ports);
+EXPORT_SYMBOL_GPL(ata_bmdma_mode_filter);
 EXPORT_SYMBOL_GPL(ata_bmdma_setup);
 EXPORT_SYMBOL_GPL(ata_bmdma_start);
 EXPORT_SYMBOL_GPL(ata_bmdma_stop);
 EXPORT_SYMBOL_GPL(ata_bmdma_status);
 EXPORT_SYMBOL_GPL(ata_bus_reset);
 #ifdef CONFIG_PCI
-EXPORT_SYMBOL_GPL(ata_pci_clear_simplex);
-EXPORT_SYMBOL_GPL(ata_pci_init_bmdma);
-EXPORT_SYMBOL_GPL(ata_pci_init_sff_host);
-EXPORT_SYMBOL_GPL(ata_pci_prepare_sff_host);
-EXPORT_SYMBOL_GPL(ata_pci_activate_sff_host);
-EXPORT_SYMBOL_GPL(ata_pci_init_one);
+EXPORT_SYMBOL_GPL(ata_pci_bmdma_clear_simplex);
+EXPORT_SYMBOL_GPL(ata_pci_bmdma_init);
+EXPORT_SYMBOL_GPL(ata_pci_sff_init_host);
+EXPORT_SYMBOL_GPL(ata_pci_sff_prepare_host);
+EXPORT_SYMBOL_GPL(ata_pci_sff_activate_host);
+EXPORT_SYMBOL_GPL(ata_pci_sff_init_one);
 #endif /* CONFIG_PCI */
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index d337f3209caf..c5f91e629945 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -47,7 +47,7 @@ static int pacpi_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (ap->acpi_handle == NULL || ata_acpi_gtm(ap, &acpi->gtm) < 0)
 		return -ENODEV;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -106,7 +106,7 @@ static unsigned long pacpi_discover_modes(struct ata_port *ap, struct ata_device
 static unsigned long pacpi_mode_filter(struct ata_device *adev, unsigned long mask)
 {
 	struct pata_acpi *acpi = adev->link->ap->private_data;
-	return ata_pci_default_filter(adev, mask & acpi->mask[adev->devno]);
+	return ata_bmdma_mode_filter(adev, mask & acpi->mask[adev->devno]);
 }
 
 /**
@@ -162,7 +162,7 @@ static void pacpi_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	pacpi_qc_issue_prot	-	command issue
+ *	pacpi_qc_issue	-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -170,14 +170,14 @@ static void pacpi_set_dmamode(struct ata_port *ap, struct ata_device *adev)
  *	neccessary.
  */
 
-static unsigned int pacpi_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int pacpi_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
 	struct pata_acpi *acpi = ap->private_data;
 
 	if (acpi->gtm.flags & 0x10)
-		return ata_qc_issue_prot(qc);
+		return ata_sff_qc_issue(qc);
 
 	if (adev != acpi->last) {
 		pacpi_set_piomode(ap, adev);
@@ -185,7 +185,7 @@ static unsigned int pacpi_qc_issue_prot(struct ata_queued_cmd *qc)
 			pacpi_set_dmamode(ap, adev);
 		acpi->last = adev;
 	}
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 /**
@@ -223,7 +223,7 @@ static struct scsi_host_template pacpi_sht = {
 
 static struct ata_port_operations pacpi_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.qc_issue		= pacpi_qc_issue_prot,
+	.qc_issue		= pacpi_qc_issue,
 	.cable_detect		= pacpi_cable_detect,
 	.mode_filter		= pacpi_mode_filter,
 	.set_piomode		= pacpi_set_piomode,
@@ -259,7 +259,7 @@ static int pacpi_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 		.port_ops	= &pacpi_ops,
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(pdev, ppi, &pacpi_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &pacpi_sht, NULL);
 }
 
 static const struct pci_device_id pacpi_pci_tbl[] = {
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index 43c558f20f92..fcabe46f262b 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -121,7 +121,7 @@ static unsigned long ali_20_filter(struct ata_device *adev, unsigned long mask)
 	ata_id_c_string(adev->id, model_num, ATA_ID_PROD, sizeof(model_num));
 	if (strstr(model_num, "WDC"))
 		return mask &= ~ATA_MASK_UDMA;
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 /**
@@ -449,7 +449,7 @@ static void ali_init_chipset(struct pci_dev *pdev)
 	}
 	pci_dev_put(isa_bridge);
 	pci_dev_put(north);
-	ata_pci_clear_simplex(pdev);
+	ata_pci_bmdma_clear_simplex(pdev);
 }
 /**
  *	ali_init_one		-	discovery callback
@@ -552,7 +552,7 @@ static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	        	ppi[0] = &info_20_udma;
 		pci_dev_put(isa_bridge);
 	}
-	return ata_pci_init_one(pdev, ppi, &ali_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &ali_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 33074c34105c..26665c396485 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -143,7 +143,7 @@ static int amd_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &amd_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 static int amd_cable_detect(struct ata_port *ap)
@@ -293,7 +293,7 @@ static int nv_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &nv_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -503,7 +503,7 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	ppi[0] = &info[type];
 
 	if (type < 3)
-		ata_pci_clear_simplex(pdev);
+		ata_pci_bmdma_clear_simplex(pdev);
 
 	/* Check for AMD7411 */
 	if (type == 3)
@@ -523,7 +523,7 @@ static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	/* And fire it up */
-	return ata_pci_init_one(pdev, ppi, &amd_sht, hpriv);
+	return ata_pci_sff_init_one(pdev, ppi, &amd_sht, hpriv);
 }
 
 #ifdef CONFIG_PM
@@ -546,7 +546,7 @@ static int amd_reinit_one(struct pci_dev *pdev)
 			pci_write_config_byte(pdev, 0x41, fifo | 0xF0);
 		if (pdev->device == PCI_DEVICE_ID_AMD_VIPER_7409 ||
 		    pdev->device == PCI_DEVICE_ID_AMD_COBRA_7401)
-		    	ata_pci_clear_simplex(pdev);
+			ata_pci_bmdma_clear_simplex(pdev);
 	}
 
 	ata_host_resume(host);
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index b6d8c4d0e6c2..0f513bc11193 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -52,7 +52,7 @@ static int artop6210_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -78,7 +78,7 @@ static int artop6260_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (pdev->device % 1 && !pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -400,7 +400,7 @@ static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
 
 	BUG_ON(ppi[0] == NULL);
 
-	return ata_pci_init_one(pdev, ppi, &artop_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &artop_sht, NULL);
 }
 
 static const struct pci_device_id artop_pci_tbl[] = {
diff --git a/drivers/ata/pata_at32.c b/drivers/ata/pata_at32.c
index 528315587532..3e8651d78952 100644
--- a/drivers/ata/pata_at32.c
+++ b/drivers/ata/pata_at32.c
@@ -223,7 +223,7 @@ static int __init pata_at32_init_one(struct device *dev,
 	host->private_data = info;
 
 	/* Register ATA device and return */
-	return ata_host_activate(host, info->irq, ata_interrupt,
+	return ata_host_activate(host, info->irq, ata_sff_interrupt,
 				 IRQF_SHARED | IRQF_TRIGGER_RISING,
 				 &at32_sht);
 }
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 56a65baddd4a..78738fb4223b 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -45,7 +45,7 @@ static int atiixp_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &atiixp_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 static int atiixp_cable_detect(struct ata_port *ap)
@@ -223,7 +223,7 @@ static struct scsi_host_template atiixp_sht = {
 static struct ata_port_operations atiixp_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 
-	.qc_prep 	= ata_dumb_qc_prep,
+	.qc_prep 	= ata_sff_dumb_qc_prep,
 	.bmdma_start 	= atiixp_bmdma_start,
 	.bmdma_stop	= atiixp_bmdma_stop,
 
@@ -243,7 +243,7 @@ static int atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.port_ops = &atiixp_port_ops
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(dev, ppi, &atiixp_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &atiixp_sht, NULL);
 }
 
 static const struct pci_device_id atiixp[] = {
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 457ac800cd5e..c854e882d4a9 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -674,7 +674,7 @@ static void read_atapi_data(void __iomem *base,
  *	@ap: Port to which output is sent
  *	@tf: ATA taskfile register set
  *
- *	Note: Original code is ata_tf_load().
+ *	Note: Original code is ata_sff_tf_load().
  */
 
 static void bfin_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
@@ -745,7 +745,7 @@ static u8 bfin_check_status(struct ata_port *ap)
  *	@ap: Port from which input is read
  *	@tf: ATA taskfile register set for storing input
  *
- *	Note: Original code is ata_tf_read().
+ *	Note: Original code is ata_sff_tf_read().
  */
 
 static void bfin_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
@@ -775,7 +775,7 @@ static void bfin_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
  *	@ap: port to which command is being issued
  *	@tf: ATA taskfile register set
  *
- *	Note: Original code is ata_exec_command().
+ *	Note: Original code is ata_sff_exec_command().
  */
 
 static void bfin_exec_command(struct ata_port *ap,
@@ -785,7 +785,7 @@ static void bfin_exec_command(struct ata_port *ap,
 	dev_dbg(ap->dev, "ata%u: cmd 0x%X\n", ap->print_id, tf->command);
 
 	write_atapi_register(base, ATA_REG_CMD, tf->command);
-	ata_pause(ap);
+	ata_sff_pause(ap);
 }
 
 /**
@@ -800,14 +800,14 @@ static u8 bfin_check_altstatus(struct ata_port *ap)
 }
 
 /**
- *	bfin_std_dev_select - Select device 0/1 on ATA bus
+ *	bfin_dev_select - Select device 0/1 on ATA bus
  *	@ap: ATA channel to manipulate
  *	@device: ATA device (numbered from zero) to select
  *
- *	Note: Original code is ata_std_dev_select().
+ *	Note: Original code is ata_sff_dev_select().
  */
 
-static void bfin_std_dev_select(struct ata_port *ap, unsigned int device)
+static void bfin_dev_select(struct ata_port *ap, unsigned int device)
 {
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
 	u8 tmp;
@@ -818,7 +818,7 @@ static void bfin_std_dev_select(struct ata_port *ap, unsigned int device)
 		tmp = ATA_DEVICE_OBS | ATA_DEV1;
 
 	write_atapi_register(base, ATA_REG_DEVICE, tmp);
-	ata_pause(ap);
+	ata_sff_pause(ap);
 }
 
 /**
@@ -977,7 +977,7 @@ static unsigned int bfin_devchk(struct ata_port *ap,
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
 	u8 nsect, lbal;
 
-	bfin_std_dev_select(ap, device);
+	bfin_dev_select(ap, device);
 
 	write_atapi_register(base, ATA_REG_NSECT, 0x55);
 	write_atapi_register(base, ATA_REG_LBAL, 0xaa);
@@ -1014,7 +1014,7 @@ static void bfin_bus_post_reset(struct ata_port *ap, unsigned int devmask)
 	 * BSY bit to clear
 	 */
 	if (dev0)
-		ata_busy_sleep(ap, ATA_TMOUT_BOOT_QUICK, ATA_TMOUT_BOOT);
+		ata_sff_busy_sleep(ap, ATA_TMOUT_BOOT_QUICK, ATA_TMOUT_BOOT);
 
 	/* if device 1 was found in ata_devchk, wait for
 	 * register access, then wait for BSY to clear
@@ -1023,7 +1023,7 @@ static void bfin_bus_post_reset(struct ata_port *ap, unsigned int devmask)
 	while (dev1) {
 		u8 nsect, lbal;
 
-		bfin_std_dev_select(ap, 1);
+		bfin_dev_select(ap, 1);
 		nsect = read_atapi_register(base, ATA_REG_NSECT);
 		lbal = read_atapi_register(base, ATA_REG_LBAL);
 		if ((nsect == 1) && (lbal == 1))
@@ -1035,14 +1035,14 @@ static void bfin_bus_post_reset(struct ata_port *ap, unsigned int devmask)
 		msleep(50);	/* give drive a breather */
 	}
 	if (dev1)
-		ata_busy_sleep(ap, ATA_TMOUT_BOOT_QUICK, ATA_TMOUT_BOOT);
+		ata_sff_busy_sleep(ap, ATA_TMOUT_BOOT_QUICK, ATA_TMOUT_BOOT);
 
 	/* is all this really necessary? */
-	bfin_std_dev_select(ap, 0);
+	bfin_dev_select(ap, 0);
 	if (dev1)
-		bfin_std_dev_select(ap, 1);
+		bfin_dev_select(ap, 1);
 	if (dev0)
-		bfin_std_dev_select(ap, 0);
+		bfin_dev_select(ap, 0);
 }
 
 /**
@@ -1088,15 +1088,15 @@ static unsigned int bfin_bus_softreset(struct ata_port *ap,
 }
 
 /**
- *	bfin_std_softreset - reset host port via ATA SRST
+ *	bfin_softreset - reset host port via ATA SRST
  *	@ap: port to reset
  *	@classes: resulting classes of attached devices
  *
- *	Note: Original code is ata_std_softreset().
+ *	Note: Original code is ata_sff_softreset().
  */
 
-static int bfin_std_softreset(struct ata_link *link, unsigned int *classes,
-		unsigned long deadline)
+static int bfin_softreset(struct ata_link *link, unsigned int *classes,
+			  unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
 	unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS;
@@ -1115,7 +1115,7 @@ static int bfin_std_softreset(struct ata_link *link, unsigned int *classes,
 		devmask |= (1 << 1);
 
 	/* select device 0 again */
-	bfin_std_dev_select(ap, 0);
+	bfin_dev_select(ap, 0);
 
 	/* issue bus reset */
 	err_mask = bfin_bus_softreset(ap, devmask);
@@ -1126,10 +1126,10 @@ static int bfin_std_softreset(struct ata_link *link, unsigned int *classes,
 	}
 
 	/* determine by signature whether we have ATA or ATAPI devices */
-	classes[0] = ata_dev_try_classify(&ap->link.device[0],
+	classes[0] = ata_sff_dev_classify(&ap->link.device[0],
 				devmask & (1 << 0), &err);
 	if (slave_possible && err != 0x81)
-		classes[1] = ata_dev_try_classify(&ap->link.device[1],
+		classes[1] = ata_sff_dev_classify(&ap->link.device[1],
 					devmask & (1 << 1), &err);
 
  out:
@@ -1167,7 +1167,7 @@ static unsigned char bfin_bmdma_status(struct ata_port *ap)
  *	@buflen: buffer length
  *	@write_data: read/write
  *
- *	Note: Original code is ata_data_xfer().
+ *	Note: Original code is ata_sff_data_xfer().
  */
 
 static unsigned int bfin_data_xfer(struct ata_device *dev, unsigned char *buf,
@@ -1206,7 +1206,7 @@ static unsigned int bfin_data_xfer(struct ata_device *dev, unsigned char *buf,
  *	bfin_irq_clear - Clear ATAPI interrupt.
  *	@ap: Port associated with this ATA transaction.
  *
- *	Note: Original code is ata_bmdma_irq_clear().
+ *	Note: Original code is ata_sff_irq_clear().
  */
 
 static void bfin_irq_clear(struct ata_port *ap)
@@ -1223,7 +1223,7 @@ static void bfin_irq_clear(struct ata_port *ap)
  *	bfin_irq_on - Enable interrupts on a port.
  *	@ap: Port on which interrupts are enabled.
  *
- *	Note: Original code is ata_irq_on().
+ *	Note: Original code is ata_sff_irq_on().
  */
 
 static unsigned char bfin_irq_on(struct ata_port *ap)
@@ -1244,13 +1244,13 @@ static unsigned char bfin_irq_on(struct ata_port *ap)
 }
 
 /**
- *	bfin_bmdma_freeze - Freeze DMA controller port
+ *	bfin_freeze - Freeze DMA controller port
  *	@ap: port to freeze
  *
- *	Note: Original code is ata_bmdma_freeze().
+ *	Note: Original code is ata_sff_freeze().
  */
 
-static void bfin_bmdma_freeze(struct ata_port *ap)
+static void bfin_freeze(struct ata_port *ap)
 {
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
 
@@ -1270,13 +1270,13 @@ static void bfin_bmdma_freeze(struct ata_port *ap)
 }
 
 /**
- *	bfin_bmdma_thaw - Thaw DMA controller port
+ *	bfin_thaw - Thaw DMA controller port
  *	@ap: port to thaw
  *
- *	Note: Original code is ata_bmdma_thaw().
+ *	Note: Original code is ata_sff_thaw().
  */
 
-void bfin_bmdma_thaw(struct ata_port *ap)
+void bfin_thaw(struct ata_port *ap)
 {
 	bfin_check_status(ap);
 	bfin_irq_clear(ap);
@@ -1284,14 +1284,14 @@ void bfin_bmdma_thaw(struct ata_port *ap)
 }
 
 /**
- *	bfin_std_postreset - standard postreset callback
+ *	bfin_postreset - standard postreset callback
  *	@ap: the target ata_port
  *	@classes: classes of attached devices
  *
- *	Note: Original code is ata_std_postreset().
+ *	Note: Original code is ata_sff_postreset().
  */
 
-static void bfin_std_postreset(struct ata_link *link, unsigned int *classes)
+static void bfin_postreset(struct ata_link *link, unsigned int *classes)
 {
 	struct ata_port *ap = link->ap;
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
@@ -1301,9 +1301,9 @@ static void bfin_std_postreset(struct ata_link *link, unsigned int *classes)
 
 	/* is double-select really necessary? */
 	if (classes[0] != ATA_DEV_NONE)
-		bfin_std_dev_select(ap, 1);
+		bfin_dev_select(ap, 1);
 	if (classes[1] != ATA_DEV_NONE)
-		bfin_std_dev_select(ap, 0);
+		bfin_dev_select(ap, 0);
 
 	/* bail out if no device is present */
 	if (classes[0] == ATA_DEV_NONE && classes[1] == ATA_DEV_NONE) {
@@ -1362,7 +1362,7 @@ static const struct ata_port_operations bfin_pata_ops = {
 	.exec_command		= bfin_exec_command,
 	.check_status		= bfin_check_status,
 	.check_altstatus	= bfin_check_altstatus,
-	.dev_select		= bfin_std_dev_select,
+	.dev_select		= bfin_dev_select,
 
 	.bmdma_setup		= bfin_bmdma_setup,
 	.bmdma_start		= bfin_bmdma_start,
@@ -1372,10 +1372,10 @@ static const struct ata_port_operations bfin_pata_ops = {
 
 	.qc_prep		= ata_noop_qc_prep,
 
-	.freeze			= bfin_bmdma_freeze,
-	.thaw			= bfin_bmdma_thaw,
-	.softreset		= bfin_std_softreset,
-	.postreset		= bfin_std_postreset,
+	.freeze			= bfin_freeze,
+	.thaw			= bfin_thaw,
+	.softreset		= bfin_softreset,
+	.postreset		= bfin_postreset,
 	.post_internal_cmd	= bfin_bmdma_stop,
 
 	.irq_clear		= bfin_irq_clear,
@@ -1513,7 +1513,7 @@ static int __devinit bfin_atapi_probe(struct platform_device *pdev)
 	}
 
 	if (ata_host_activate(host, platform_get_irq(pdev, 0),
-		ata_interrupt, IRQF_SHARED, &bfin_sht) != 0) {
+		ata_sff_interrupt, IRQF_SHARED, &bfin_sht) != 0) {
 		peripheral_free_list(atapi_io_port);
 		dev_err(&pdev->dev, "Fail to attach ATAPI device\n");
 		return -ENODEV;
diff --git a/drivers/ata/pata_cmd640.c b/drivers/ata/pata_cmd640.c
index 27219b00edf4..a907cf478891 100644
--- a/drivers/ata/pata_cmd640.c
+++ b/drivers/ata/pata_cmd640.c
@@ -107,8 +107,8 @@ static void cmd640_set_piomode(struct ata_port *ap, struct ata_device *adev)
 		pci_write_config_byte(pdev, arttim + 1, (t.active << 4) | t.recover);
 	} else {
 		/* Save the shared timings for channel, they will be loaded
-		   by qc_issue_prot. Reloading the setup time is expensive
-		   so we keep a merged one loaded */
+		   by qc_issue. Reloading the setup time is expensive so we
+		   keep a merged one loaded */
 		pci_read_config_byte(pdev, ARTIM23, &reg);
 		reg &= 0x3F;
 		reg |= t.setup;
@@ -119,14 +119,14 @@ static void cmd640_set_piomode(struct ata_port *ap, struct ata_device *adev)
 
 
 /**
- *	cmd640_qc_issue_prot	-	command preparation hook
+ *	cmd640_qc_issue	-	command preparation hook
  *	@qc: Command to be issued
  *
  *	Channel 1 has shared timings. We must reprogram the
  *	clock each drive 2/3 switch we do.
  */
 
-static unsigned int cmd640_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int cmd640_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -137,7 +137,7 @@ static unsigned int cmd640_qc_issue_prot(struct ata_queued_cmd *qc)
 		pci_write_config_byte(pdev, DRWTIM23, timing->reg58[adev->devno]);
 		timing->last = adev->devno;
 	}
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 /**
@@ -172,8 +172,8 @@ static struct scsi_host_template cmd640_sht = {
 static struct ata_port_operations cmd640_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 	/* In theory xfer_noirq is not needed once we kill the prefetcher */
-	.data_xfer	= ata_data_xfer_noirq,
-	.qc_issue	= cmd640_qc_issue_prot,
+	.data_xfer	= ata_sff_data_xfer_noirq,
+	.qc_issue	= cmd640_qc_issue,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= cmd640_set_piomode,
 	.port_start	= cmd640_port_start,
@@ -224,7 +224,7 @@ static int cmd640_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	cmd640_hardware_init(pdev);
 
-	return ata_pci_init_one(pdev, ppi, &cmd640_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &cmd640_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index f0e566623614..ddd09b7d98c9 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -349,7 +349,7 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	class_rev &= 0xFF;
 
 	if (id->driver_data == 0)	/* 643 */
-		ata_pci_clear_simplex(pdev);
+		ata_pci_bmdma_clear_simplex(pdev);
 
 	if (pdev->device == PCI_DEVICE_ID_CMD_646) {
 		/* Does UDMA work ? */
@@ -373,7 +373,7 @@ static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	pci_write_config_byte(pdev, UDIDETCR0, 0xF0);
 #endif
 
-	return ata_pci_init_one(pdev, ppi, &cmd64x_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &cmd64x_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cs5520.c b/drivers/ata/pata_cs5520.c
index 46d0ce32ee5a..1186bcd2781c 100644
--- a/drivers/ata/pata_cs5520.c
+++ b/drivers/ata/pata_cs5520.c
@@ -146,7 +146,7 @@ static struct scsi_host_template cs5520_sht = {
 
 static struct ata_port_operations cs5520_port_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.qc_prep		= ata_dumb_qc_prep,
+	.qc_prep		= ata_sff_dumb_qc_prep,
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= cs5520_set_piomode,
 	.set_dmamode		= cs5520_set_dmamode,
@@ -227,7 +227,7 @@ static int __devinit cs5520_init_one(struct pci_dev *pdev, const struct pci_devi
 	ioaddr->ctl_addr = iomap[1];
 	ioaddr->altstatus_addr = iomap[1];
 	ioaddr->bmdma_addr = iomap[4];
-	ata_std_ports(ioaddr);
+	ata_sff_std_ports(ioaddr);
 
 	ata_port_desc(host->ports[0],
 		      "cmd 0x%x ctl 0x%x", cmd_port[0], ctl_port[0]);
@@ -238,7 +238,7 @@ static int __devinit cs5520_init_one(struct pci_dev *pdev, const struct pci_devi
 	ioaddr->ctl_addr = iomap[3];
 	ioaddr->altstatus_addr = iomap[3];
 	ioaddr->bmdma_addr = iomap[4] + 8;
-	ata_std_ports(ioaddr);
+	ata_sff_std_ports(ioaddr);
 
 	ata_port_desc(host->ports[1],
 		      "cmd 0x%x ctl 0x%x", cmd_port[1], ctl_port[1]);
@@ -258,7 +258,7 @@ static int __devinit cs5520_init_one(struct pci_dev *pdev, const struct pci_devi
 			continue;
 
 		rc = devm_request_irq(&pdev->dev, irq[ap->port_no],
-				      ata_interrupt, 0, DRV_NAME, host);
+				      ata_sff_interrupt, 0, DRV_NAME, host);
 		if (rc)
 			return rc;
 
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
index ac3ad55d7c3c..744beebaaf49 100644
--- a/drivers/ata/pata_cs5530.c
+++ b/drivers/ata/pata_cs5530.c
@@ -133,7 +133,7 @@ static void cs5530_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	cs5530_qc_issue_prot	-	command issue
+ *	cs5530_qc_issue		-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -142,7 +142,7 @@ static void cs5530_set_dmamode(struct ata_port *ap, struct ata_device *adev)
  *	one MWDMA/UDMA bit.
  */
 
-static unsigned int cs5530_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int cs5530_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -157,7 +157,7 @@ static unsigned int cs5530_qc_issue_prot(struct ata_queued_cmd *qc)
 		    	cs5530_set_dmamode(ap, adev);
 	}
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static struct scsi_host_template cs5530_sht = {
@@ -168,8 +168,8 @@ static struct scsi_host_template cs5530_sht = {
 static struct ata_port_operations cs5530_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 
-	.qc_prep 	= ata_dumb_qc_prep,
-	.qc_issue	= cs5530_qc_issue_prot,
+	.qc_prep 	= ata_sff_dumb_qc_prep,
+	.qc_issue	= cs5530_qc_issue,
 
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= cs5530_set_piomode,
@@ -325,7 +325,7 @@ static int cs5530_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		ppi[1] = &info_palmax_secondary;
 
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(pdev, ppi, &cs5530_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &cs5530_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
index 5c0762ebf58c..f1b6556f0483 100644
--- a/drivers/ata/pata_cs5535.c
+++ b/drivers/ata/pata_cs5535.c
@@ -199,7 +199,7 @@ static int cs5535_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	rdmsr(ATAC_CH0D1_PIO, timings, dummy);
 	if (CS5535_BAD_PIO(timings))
 		wrmsr(ATAC_CH0D1_PIO, 0xF7F4F7F4UL, 0);
-	return ata_pci_init_one(dev, ppi, &cs5535_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &cs5535_sht, NULL);
 }
 
 static const struct pci_device_id cs5535[] = {
diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c
index 2d34b9145dcb..73f8332cb679 100644
--- a/drivers/ata/pata_cs5536.c
+++ b/drivers/ata/pata_cs5536.c
@@ -261,7 +261,7 @@ static int cs5536_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		return -ENODEV;
 	}
 
-	return ata_pci_init_one(dev, ppi, &cs5536_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &cs5536_sht, NULL);
 }
 
 static const struct pci_device_id cs5536[] = {
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
index ae14969e1dfe..a9c3218e22fd 100644
--- a/drivers/ata/pata_cypress.c
+++ b/drivers/ata/pata_cypress.c
@@ -136,7 +136,7 @@ static int cy82c693_init_one(struct pci_dev *pdev, const struct pci_device_id *i
 	if (PCI_FUNC(pdev->devfn) != 1)
 		return -ENODEV;
 
-	return ata_pci_init_one(pdev, ppi, &cy82c693_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &cy82c693_sht, NULL);
 }
 
 static const struct pci_device_id cy82c693[] = {
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index 1d839a57068e..9fba82976ba6 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -45,7 +45,7 @@ static int efar_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &efar_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -262,7 +262,7 @@ static int efar_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &efar_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &efar_sht, NULL);
 }
 
 static const struct pci_device_id efar_pci_tbl[] = {
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
index c2d4923d4db7..f2b83eabc7c7 100644
--- a/drivers/ata/pata_hpt366.c
+++ b/drivers/ata/pata_hpt366.c
@@ -184,7 +184,7 @@ static unsigned long hpt366_filter(struct ata_device *adev, unsigned long mask)
 		if (hpt_dma_blacklisted(adev, "UDMA4", bad_ata66_4))
 			mask &= ~(0xF0 << ATA_SHIFT_UDMA);
 	}
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 /**
@@ -393,7 +393,7 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 			break;
 	}
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(dev, ppi, &hpt36x_sht, hpriv);
+	return ata_pci_sff_init_one(dev, ppi, &hpt36x_sht, hpriv);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
index c10fcd31418d..42163998de9a 100644
--- a/drivers/ata/pata_hpt37x.c
+++ b/drivers/ata/pata_hpt37x.c
@@ -283,7 +283,7 @@ static unsigned long hpt370_filter(struct ata_device *adev, unsigned long mask)
 		if (hpt_dma_blacklisted(adev, "UDMA100", bad_ata100_5))
 			mask &= ~(0xE0 << ATA_SHIFT_UDMA);
 	}
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 /**
@@ -299,7 +299,7 @@ static unsigned long hpt370a_filter(struct ata_device *adev, unsigned long mask)
 		if (hpt_dma_blacklisted(adev, "UDMA100", bad_ata100_5))
 			mask &= ~(0xE0 << ATA_SHIFT_UDMA);
 	}
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 /**
@@ -338,7 +338,7 @@ static int hpt37x_pre_reset(struct ata_link *link, unsigned long deadline)
 	pci_write_config_byte(pdev, 0x50 + 4 * ap->port_no, 0x37);
 	udelay(100);
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 static int hpt374_fn1_pre_reset(struct ata_link *link, unsigned long deadline)
@@ -374,7 +374,7 @@ static int hpt374_fn1_pre_reset(struct ata_link *link, unsigned long deadline)
 	pci_write_config_byte(pdev, 0x50 + 4 * ap->port_no, 0x37);
 	udelay(100);
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -1019,7 +1019,7 @@ static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	}
 
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(dev, ppi, &hpt37x_sht, private_data);
+	return ata_pci_sff_init_one(dev, ppi, &hpt37x_sht, private_data);
 }
 
 static const struct pci_device_id hpt37x[] = {
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
index cd44ee3d3cc1..d5c9fd7b82bb 100644
--- a/drivers/ata/pata_hpt3x2n.c
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -156,7 +156,7 @@ static int hpt3x2n_pre_reset(struct ata_link *link, unsigned long deadline)
 	pci_write_config_byte(pdev, 0x50 + 4 * ap->port_no, 0x37);
 	udelay(100);
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -308,7 +308,7 @@ static int hpt3x2n_use_dpll(struct ata_port *ap, int writing)
 	return 0;
 }
 
-static unsigned int hpt3x2n_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int hpt3x2n_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_taskfile *tf = &qc->tf;
 	struct ata_port *ap = qc->ap;
@@ -323,7 +323,7 @@ static unsigned int hpt3x2n_qc_issue_prot(struct ata_queued_cmd *qc)
 				hpt3x2n_set_clock(ap, 0x23);
 		}
 	}
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static struct scsi_host_template hpt3x2n_sht = {
@@ -338,7 +338,7 @@ static struct ata_port_operations hpt3x2n_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 
 	.bmdma_stop	= hpt3x2n_bmdma_stop,
-	.qc_issue	= hpt3x2n_qc_issue_prot,
+	.qc_issue	= hpt3x2n_qc_issue,
 
 	.cable_detect	= hpt3x2n_cable_detect,
 	.set_piomode	= hpt3x2n_set_piomode,
@@ -554,7 +554,7 @@ static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	}
 
 	/* Now kick off ATA set up */
-	return ata_pci_init_one(dev, ppi, &hpt3x2n_sht, hpriv);
+	return ata_pci_sff_init_one(dev, ppi, &hpt3x2n_sht, hpriv);
 }
 
 static const struct pci_device_id hpt3x2n[] = {
diff --git a/drivers/ata/pata_hpt3x3.c b/drivers/ata/pata_hpt3x3.c
index 8857d029ac2e..f11a320337c0 100644
--- a/drivers/ata/pata_hpt3x3.c
+++ b/drivers/ata/pata_hpt3x3.c
@@ -202,15 +202,15 @@ static int hpt3x3_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 		ioaddr->altstatus_addr =
 		ioaddr->ctl_addr = base + offset_ctl[i];
 		ioaddr->scr_addr = NULL;
-		ata_std_ports(ioaddr);
+		ata_sff_std_ports(ioaddr);
 		ioaddr->bmdma_addr = base + 8 * i;
 
 		ata_port_pbar_desc(ap, 4, -1, "ioport");
 		ata_port_pbar_desc(ap, 4, offset_cmd[i], "cmd");
 	}
 	pci_set_master(pdev);
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &hpt3x3_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &hpt3x3_sht);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_icside.c b/drivers/ata/pata_icside.c
index 13d43e9dd279..52de9f908fb0 100644
--- a/drivers/ata/pata_icside.c
+++ b/drivers/ata/pata_icside.c
@@ -270,7 +270,7 @@ static void pata_icside_bmdma_stop(struct ata_queued_cmd *qc)
 	disable_dma(state->dma);
 
 	/* see ata_bmdma_stop */
-	ata_altstatus(ap);
+	ata_sff_altstatus(ap);
 }
 
 static u8 pata_icside_bmdma_status(struct ata_port *ap)
@@ -316,7 +316,7 @@ static void pata_icside_postreset(struct ata_link *link, unsigned int *classes)
 	struct pata_icside_state *state = ap->host->private_data;
 
 	if (classes[0] != ATA_DEV_NONE || classes[1] != ATA_DEV_NONE)
-		return ata_std_postreset(link, classes);
+		return ata_sff_postreset(link, classes);
 
 	state->port[ap->port_no].disabled = 1;
 
@@ -336,7 +336,7 @@ static struct ata_port_operations pata_icside_port_ops = {
 	.inherits		= &ata_sff_port_ops,
 	/* no need to build any PRD tables for DMA */
 	.qc_prep		= ata_noop_qc_prep,
-	.data_xfer		= ata_data_xfer_noirq,
+	.data_xfer		= ata_sff_data_xfer_noirq,
 	.bmdma_setup		= pata_icside_bmdma_setup,
 	.bmdma_start		= pata_icside_bmdma_start,
 	.bmdma_stop		= pata_icside_bmdma_stop,
@@ -481,7 +481,7 @@ static int __devinit pata_icside_add_ports(struct pata_icside_info *info)
 		pata_icside_setup_ioaddr(ap, info->base, info, info->port[i]);
 	}
 
-	return ata_host_activate(host, ec->irq, ata_interrupt, 0,
+	return ata_host_activate(host, ec->irq, ata_sff_interrupt, 0,
 				 &pata_icside_sht);
 }
 
diff --git a/drivers/ata/pata_isapnp.c b/drivers/ata/pata_isapnp.c
index 085913ec6f68..6a111baab523 100644
--- a/drivers/ata/pata_isapnp.c
+++ b/drivers/ata/pata_isapnp.c
@@ -50,7 +50,7 @@ static int isapnp_init_one(struct pnp_dev *idev, const struct pnp_device_id *dev
 
 	if (pnp_irq_valid(idev, 0)) {
 		irq = pnp_irq(idev, 0);
-		handler = ata_interrupt;
+		handler = ata_sff_interrupt;
 	}
 
 	/* allocate host */
@@ -78,7 +78,7 @@ static int isapnp_init_one(struct pnp_dev *idev, const struct pnp_device_id *dev
 		ap->ioaddr.ctl_addr = ctl_addr;
 	}
 
-	ata_std_ports(&ap->ioaddr);
+	ata_sff_std_ports(&ap->ioaddr);
 
 	ata_port_desc(ap, "cmd 0x%llx ctl 0x%llx",
 		      (unsigned long long)pnp_port_start(idev, 0),
diff --git a/drivers/ata/pata_it8213.c b/drivers/ata/pata_it8213.c
index 84ab89e8a247..c113d7c079c8 100644
--- a/drivers/ata/pata_it8213.c
+++ b/drivers/ata/pata_it8213.c
@@ -40,7 +40,7 @@ static int it8213_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &it8213_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -274,7 +274,7 @@ static int it8213_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &it8213_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &it8213_sht, NULL);
 }
 
 static const struct pci_device_id it8213_pci_tbl[] = {
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index 6a8a4ddf5bfe..88e37cfcfc4f 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -395,11 +395,11 @@ static void it821x_passthru_dev_select(struct ata_port *ap,
 		it821x_program(ap, adev, itdev->pio[adev->devno]);
 		itdev->last_device = device;
 	}
-	ata_std_dev_select(ap, device);
+	ata_sff_dev_select(ap, device);
 }
 
 /**
- *	it821x_smart_qc_issue_prot	-	wrap qc issue prot
+ *	it821x_smart_qc_issue		-	wrap qc issue prot
  *	@qc: command
  *
  *	Wrap the command issue sequence for the IT821x. We need to
@@ -407,7 +407,7 @@ static void it821x_passthru_dev_select(struct ata_port *ap,
  *	usual happenings kick off
  */
 
-static unsigned int it821x_smart_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int it821x_smart_qc_issue(struct ata_queued_cmd *qc)
 {
 	switch(qc->tf.command)
 	{
@@ -427,14 +427,14 @@ static unsigned int it821x_smart_qc_issue_prot(struct ata_queued_cmd *qc)
 		case ATA_CMD_ID_ATA:
 		/* Arguably should just no-op this one */
 		case ATA_CMD_SET_FEATURES:
-			return ata_qc_issue_prot(qc);
+			return ata_sff_qc_issue(qc);
 	}
 	printk(KERN_DEBUG "it821x: can't process command 0x%02X\n", qc->tf.command);
 	return AC_ERR_DEV;
 }
 
 /**
- *	it821x_passthru_qc_issue_prot	-	wrap qc issue prot
+ *	it821x_passthru_qc_issue	-	wrap qc issue prot
  *	@qc: command
  *
  *	Wrap the command issue sequence for the IT821x. We need to
@@ -442,10 +442,10 @@ static unsigned int it821x_smart_qc_issue_prot(struct ata_queued_cmd *qc)
  *	usual happenings kick off
  */
 
-static unsigned int it821x_passthru_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int it821x_passthru_qc_issue(struct ata_queued_cmd *qc)
 {
 	it821x_passthru_dev_select(qc->ap, qc->dev->devno);
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 /**
@@ -639,7 +639,7 @@ static struct ata_port_operations it821x_smart_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 
 	.check_atapi_dma= it821x_check_atapi_dma,
-	.qc_issue	= it821x_smart_qc_issue_prot,
+	.qc_issue	= it821x_smart_qc_issue,
 
 	.cable_detect	= it821x_ident_hack,
 	.set_mode	= it821x_smart_set_mode,
@@ -655,7 +655,7 @@ static struct ata_port_operations it821x_passthru_port_ops = {
 	.dev_select 	= it821x_passthru_dev_select,
 	.bmdma_start 	= it821x_passthru_bmdma_start,
 	.bmdma_stop	= it821x_passthru_bmdma_stop,
-	.qc_issue	= it821x_passthru_qc_issue_prot,
+	.qc_issue	= it821x_passthru_qc_issue,
 
 	.cable_detect	= ata_cable_unknown,
 	.set_piomode	= it821x_passthru_set_piomode,
@@ -722,7 +722,7 @@ static int it821x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	else
 		ppi[0] = &info_smart;
 
-	return ata_pci_init_one(pdev, ppi, &it821x_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &it821x_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c
index d02629aa20da..283a8fba79b3 100644
--- a/drivers/ata/pata_ixp4xx_cf.c
+++ b/drivers/ata/pata_ixp4xx_cf.c
@@ -110,7 +110,7 @@ static void ixp4xx_setup_port(struct ata_port *ap,
 	ioaddr->altstatus_addr	= data->cs1 + 0x06;
 	ioaddr->ctl_addr	= data->cs1 + 0x06;
 
-	ata_std_ports(ioaddr);
+	ata_sff_std_ports(ioaddr);
 
 #ifndef __ARMEB__
 
@@ -186,7 +186,7 @@ static __devinit int ixp4xx_pata_probe(struct platform_device *pdev)
 	dev_printk(KERN_INFO, &pdev->dev, "version " DRV_VERSION "\n");
 
 	/* activate host */
-	return ata_host_activate(host, irq, ata_interrupt, 0, &ixp4xx_sht);
+	return ata_host_activate(host, irq, ata_sff_interrupt, 0, &ixp4xx_sht);
 }
 
 static __devexit int ixp4xx_pata_remove(struct platform_device *dev)
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index 317f3474e0ba..73b7596816b4 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -102,7 +102,7 @@ static int jmicron_pre_reset(struct ata_link *link, unsigned long deadline)
 		ap->cbl = ATA_CBL_SATA;
 		break;
 	}
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /* No PIO or DMA methods needed for this device */
@@ -144,7 +144,7 @@ static int jmicron_init_one (struct pci_dev *pdev, const struct pci_device_id *i
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
 
-	return ata_pci_init_one(pdev, ppi, &jmicron_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &jmicron_sht, NULL);
 }
 
 static const struct pci_device_id jmicron_pci_tbl[] = {
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
index 2474068596f4..f13f10a55ef3 100644
--- a/drivers/ata/pata_legacy.c
+++ b/drivers/ata/pata_legacy.c
@@ -226,12 +226,12 @@ static const struct ata_port_operations legacy_base_port_ops = {
 
 static struct ata_port_operations simple_port_ops = {
 	.inherits	= &legacy_base_port_ops,
-	.data_xfer	= ata_data_xfer_noirq,
+	.data_xfer	= ata_sff_data_xfer_noirq,
 };
 
 static struct ata_port_operations legacy_port_ops = {
 	.inherits	= &legacy_base_port_ops,
-	.data_xfer	= ata_data_xfer_noirq,
+	.data_xfer	= ata_sff_data_xfer_noirq,
 	.set_mode	= legacy_set_mode,
 };
 
@@ -317,7 +317,7 @@ static unsigned int pdc_data_xfer_vlb(struct ata_device *dev,
 		}
 		local_irq_restore(flags);
 	} else
-		buflen = ata_data_xfer_noirq(dev, buf, buflen, rw);
+		buflen = ata_sff_data_xfer_noirq(dev, buf, buflen, rw);
 
 	return buflen;
 }
@@ -579,7 +579,7 @@ static void opti82c46x_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	opt82c465mv_qc_issue_prot	-	command issue
+ *	opt82c465mv_qc_issue		-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -593,7 +593,7 @@ static void opti82c46x_set_piomode(struct ata_port *ap, struct ata_device *adev)
  *	FIXME: dual channel needs ->serialize support
  */
 
-static unsigned int opti82c46x_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int opti82c46x_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -604,13 +604,13 @@ static unsigned int opti82c46x_qc_issue_prot(struct ata_queued_cmd *qc)
 	    && ap->host->private_data != NULL)
 		opti82c46x_set_piomode(ap, adev);
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static struct ata_port_operations opti82c46x_port_ops = {
 	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= opti82c46x_set_piomode,
-	.qc_issue	= opti82c46x_qc_issue_prot,
+	.qc_issue	= opti82c46x_qc_issue,
 };
 
 static void qdi6500_set_piomode(struct ata_port *ap, struct ata_device *adev)
@@ -644,7 +644,7 @@ static void qdi6500_set_piomode(struct ata_port *ap, struct ata_device *adev)
  *	@irq: interrupt line
  *
  *	In dual channel mode the 6580 has one clock per channel and we have
- *	to software clockswitch in qc_issue_prot.
+ *	to software clockswitch in qc_issue.
  */
 
 static void qdi6580dp_set_piomode(struct ata_port *ap, struct ata_device *adev)
@@ -710,14 +710,14 @@ static void qdi6580_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	qdi_qc_issue_prot	-	command issue
+ *	qdi_qc_issue		-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
  *	this interface so that we can load the correct ATA timings.
  */
 
-static unsigned int qdi_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int qdi_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -730,7 +730,7 @@ static unsigned int qdi_qc_issue_prot(struct ata_queued_cmd *qc)
 							2 * ap->port_no);
 		}
 	}
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static unsigned int vlb32_data_xfer(struct ata_device *adev, unsigned char *buf,
@@ -759,7 +759,7 @@ static unsigned int vlb32_data_xfer(struct ata_device *adev, unsigned char *buf,
 		}
 		return (buflen + 3) & ~3;
 	} else
-		return ata_data_xfer(adev, buf, buflen, rw);
+		return ata_sff_data_xfer(adev, buf, buflen, rw);
 }
 
 static int qdi_port(struct platform_device *dev,
@@ -774,7 +774,7 @@ static int qdi_port(struct platform_device *dev,
 static struct ata_port_operations qdi6500_port_ops = {
 	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= qdi6500_set_piomode,
-	.qc_issue	= qdi_qc_issue_prot,
+	.qc_issue	= qdi_qc_issue,
 	.data_xfer	= vlb32_data_xfer,
 };
 
@@ -1016,13 +1016,13 @@ static __init int legacy_init_one(struct legacy_probe *probe)
 	ap->ioaddr.cmd_addr = io_addr;
 	ap->ioaddr.altstatus_addr = ctrl_addr;
 	ap->ioaddr.ctl_addr = ctrl_addr;
-	ata_std_ports(&ap->ioaddr);
+	ata_sff_std_ports(&ap->ioaddr);
 	ap->host->private_data = ld;
 
 	ata_port_desc(ap, "cmd 0x%lx ctl 0x%lx", io, io + 0x0206);
 
-	ret = ata_host_activate(host, probe->irq, ata_interrupt, 0,
-								&legacy_sht);
+	ret = ata_host_activate(host, probe->irq, ata_sff_interrupt, 0,
+				&legacy_sht);
 	if (ret)
 		goto fail;
 	ld->platform_dev = pdev;
diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c
index d38e64cd6097..24a011b25024 100644
--- a/drivers/ata/pata_marvell.c
+++ b/drivers/ata/pata_marvell.c
@@ -55,7 +55,7 @@ static int marvell_pre_reset(struct ata_link *link, unsigned long deadline)
 	    (!(devices & 0x10)))	/* PATA enable ? */
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 static int marvell_cable_detect(struct ata_port *ap)
@@ -128,7 +128,7 @@ static int marvell_init_one (struct pci_dev *pdev, const struct pci_device_id *i
 	if (pdev->device == 0x6101)
 		ppi[1] = &ata_dummy_port_info;
 
-	return ata_pci_init_one(pdev, ppi, &marvell_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &marvell_sht, NULL);
 }
 
 static const struct pci_device_id marvell_pci_tbl[] = {
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index fec93196710e..5d1d32a39c4d 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -252,7 +252,7 @@ mpc52xx_ata_dev_select(struct ata_port *ap, unsigned int device)
 	if (device != priv->csel)
 		mpc52xx_ata_apply_timings(priv, device);
 
-	ata_std_dev_select(ap,device);
+	ata_sff_dev_select(ap,device);
 }
 
 static struct scsi_host_template mpc52xx_ata_sht = {
@@ -305,7 +305,7 @@ mpc52xx_ata_init_one(struct device *dev, struct mpc52xx_ata_priv *priv,
 	ata_port_desc(ap, "ata_regs 0x%lx", raw_ata_regs);
 
 	/* activate host */
-	return ata_host_activate(host, priv->ata_irq, ata_interrupt, 0,
+	return ata_host_activate(host, priv->ata_irq, ata_sff_interrupt, 0,
 				 &mpc52xx_ata_sht);
 }
 
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index 1b9d0d412ebf..7d7e3fdab71f 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -55,7 +55,7 @@ static int mpiix_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &mpiix_enable_bits))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -69,8 +69,8 @@ static int mpiix_pre_reset(struct ata_link *link, unsigned long deadline)
  *
  *	This would get very ugly because we can only program timing for one
  *	device at a time, the other gets PIO0. Fortunately libata calls
- *	our qc_issue_prot command before a command is issued so we can
- *	flip the timings back and forth to reduce the pain.
+ *	our qc_issue command before a command is issued so we can flip the
+ *	timings back and forth to reduce the pain.
  */
 
 static void mpiix_set_piomode(struct ata_port *ap, struct ata_device *adev)
@@ -110,7 +110,7 @@ static void mpiix_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	mpiix_qc_issue_prot	-	command issue
+ *	mpiix_qc_issue		-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -120,7 +120,7 @@ static void mpiix_set_piomode(struct ata_port *ap, struct ata_device *adev)
  *	be made PIO0.
  */
 
-static unsigned int mpiix_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int mpiix_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -133,7 +133,7 @@ static unsigned int mpiix_qc_issue_prot(struct ata_queued_cmd *qc)
 	if (adev->pio_mode && adev != ap->private_data)
 		mpiix_set_piomode(ap, adev);
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static struct scsi_host_template mpiix_sht = {
@@ -142,7 +142,7 @@ static struct scsi_host_template mpiix_sht = {
 
 static struct ata_port_operations mpiix_port_ops = {
 	.inherits	= &ata_sff_port_ops,
-	.qc_issue	= mpiix_qc_issue_prot,
+	.qc_issue	= mpiix_qc_issue,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= mpiix_set_piomode,
 	.prereset	= mpiix_pre_reset,
@@ -207,10 +207,10 @@ static int mpiix_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	ap->ioaddr.altstatus_addr = ctl_addr;
 
 	/* Let libata fill in the port details */
-	ata_std_ports(&ap->ioaddr);
+	ata_sff_std_ports(&ap->ioaddr);
 
 	/* activate host */
-	return ata_host_activate(host, irq, ata_interrupt, IRQF_SHARED,
+	return ata_host_activate(host, irq, ata_sff_interrupt, IRQF_SHARED,
 				 &mpiix_sht);
 }
 
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
index 349182840d24..d9719c8b9dbe 100644
--- a/drivers/ata/pata_netcell.c
+++ b/drivers/ata/pata_netcell.c
@@ -68,10 +68,10 @@ static int netcell_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 		return rc;
 
 	/* Any chip specific setup/optimisation/messages here */
-	ata_pci_clear_simplex(pdev);
+	ata_pci_bmdma_clear_simplex(pdev);
 
 	/* And let the library code do the work */
-	return ata_pci_init_one(pdev, port_info, &netcell_sht, NULL);
+	return ata_pci_sff_init_one(pdev, port_info, &netcell_sht, NULL);
 }
 
 static const struct pci_device_id netcell_pci_tbl[] = {
diff --git a/drivers/ata/pata_ninja32.c b/drivers/ata/pata_ninja32.c
index 8213d081f313..c9c0ea67c71a 100644
--- a/drivers/ata/pata_ninja32.c
+++ b/drivers/ata/pata_ninja32.c
@@ -73,7 +73,7 @@ static void ninja32_dev_select(struct ata_port *ap, unsigned int device)
 	struct ata_device *adev = &ap->link.device[device];
 	if (ap->private_data != adev) {
 		iowrite8(0xd6, ap->ioaddr.bmdma_addr + 0x1f);
-		ata_std_dev_select(ap, device);
+		ata_sff_dev_select(ap, device);
 		ninja32_set_piomode(ap, adev);
 	}
 }
@@ -132,7 +132,7 @@ static int ninja32_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	ap->ioaddr.ctl_addr = base + 0x1E;
 	ap->ioaddr.altstatus_addr = base + 0x1E;
 	ap->ioaddr.bmdma_addr = base;
-	ata_std_ports(&ap->ioaddr);
+	ata_sff_std_ports(&ap->ioaddr);
 
 	iowrite8(0x05, base + 0x01);	/* Enable interrupt lines */
 	iowrite8(0xBE, base + 0x02);	/* Burst, ?? setup */
@@ -142,7 +142,7 @@ static int ninja32_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	iowrite8(0xa4, base + 0x1c);	/* Unknown */
 	iowrite8(0x83, base + 0x1d);	/* BMDMA control: WAIT0 */
 	/* FIXME: Should we disable them at remove ? */
-	return ata_host_activate(host, dev->irq, ata_interrupt,
+	return ata_host_activate(host, dev->irq, ata_sff_interrupt,
 				 IRQF_SHARED, &ninja32_sht);
 }
 
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index 4d2eefee7387..76d2455bc453 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -50,7 +50,7 @@ static int ns87410_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &ns87410_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -105,7 +105,7 @@ static void ns87410_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	ns87410_qc_issue_prot	-	command issue
+ *	ns87410_qc_issue	-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -113,7 +113,7 @@ static void ns87410_set_piomode(struct ata_port *ap, struct ata_device *adev)
  *	necessary.
  */
 
-static unsigned int ns87410_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int ns87410_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -126,7 +126,7 @@ static unsigned int ns87410_qc_issue_prot(struct ata_queued_cmd *qc)
 	if (adev->pio_mode && adev != ap->private_data)
 		ns87410_set_piomode(ap, adev);
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static struct scsi_host_template ns87410_sht = {
@@ -135,7 +135,7 @@ static struct scsi_host_template ns87410_sht = {
 
 static struct ata_port_operations ns87410_port_ops = {
 	.inherits	= &ata_sff_port_ops,
-	.qc_issue	= ns87410_qc_issue_prot,
+	.qc_issue	= ns87410_qc_issue,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= ns87410_set_piomode,
 	.prereset	= ns87410_pre_reset,
@@ -149,7 +149,7 @@ static int ns87410_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.port_ops = &ns87410_port_ops
 	};
 	const struct ata_port_info *ppi[] = { &info, NULL };
-	return ata_pci_init_one(dev, ppi, &ns87410_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &ns87410_sht, NULL);
 }
 
 static const struct pci_device_id ns87410[] = {
diff --git a/drivers/ata/pata_ns87415.c b/drivers/ata/pata_ns87415.c
index cdd79d6fc0ee..03a52cefc010 100644
--- a/drivers/ata/pata_ns87415.c
+++ b/drivers/ata/pata_ns87415.c
@@ -172,14 +172,14 @@ static void ns87415_bmdma_stop(struct ata_queued_cmd *qc)
 }
 
 /**
- *	ns87415_bmdma_irq_clear		-	Clear interrupt
+ *	ns87415_irq_clear		-	Clear interrupt
  *	@ap: Channel to clear
  *
  *	Erratum: Due to a chip bug regisers 02 and 0A bit 1 and 2 (the
  *	error bits) are reset by writing to register 00 or 08.
  */
 
-static void ns87415_bmdma_irq_clear(struct ata_port *ap)
+static void ns87415_irq_clear(struct ata_port *ap)
 {
 	void __iomem *mmio = ap->ioaddr.bmdma_addr;
 
@@ -306,7 +306,7 @@ static struct ata_port_operations ns87415_pata_ops = {
 	.bmdma_setup		= ns87415_bmdma_setup,
 	.bmdma_start		= ns87415_bmdma_start,
 	.bmdma_stop		= ns87415_bmdma_stop,
-	.irq_clear		= ns87415_bmdma_irq_clear,
+	.irq_clear		= ns87415_irq_clear,
 
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= ns87415_set_piomode,
@@ -375,7 +375,7 @@ static int ns87415_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	pci_write_config_byte(pdev, 0x55, 0xEE);
 	/* Select PIO0 8bit clocking */
 	pci_write_config_byte(pdev, 0x54, 0xB7);
-	return ata_pci_init_one(pdev, ppi, &ns87415_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &ns87415_sht, NULL);
 }
 
 static const struct pci_device_id ns87415_pci_tbl[] = {
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index c1da79a76439..e678af383d13 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -47,7 +47,7 @@ static int oldpiix_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &oldpiix_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -181,7 +181,7 @@ static void oldpiix_set_dmamode (struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	oldpiix_qc_issue_prot	-	command issue
+ *	oldpiix_qc_issue	-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -191,7 +191,7 @@ static void oldpiix_set_dmamode (struct ata_port *ap, struct ata_device *adev)
  *	be made PIO0.
  */
 
-static unsigned int oldpiix_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int oldpiix_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -201,7 +201,7 @@ static unsigned int oldpiix_qc_issue_prot(struct ata_queued_cmd *qc)
 		if (adev->dma_mode)
 			oldpiix_set_dmamode(ap, adev);
 	}
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 
@@ -211,7 +211,7 @@ static struct scsi_host_template oldpiix_sht = {
 
 static struct ata_port_operations oldpiix_pata_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.qc_issue		= oldpiix_qc_issue_prot,
+	.qc_issue		= oldpiix_qc_issue,
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= oldpiix_set_piomode,
 	.set_dmamode		= oldpiix_set_dmamode,
@@ -249,7 +249,7 @@ static int oldpiix_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &oldpiix_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &oldpiix_sht, NULL);
 }
 
 static const struct pci_device_id oldpiix_pci_tbl[] = {
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index 4ddd03a67775..fb2cf661b0e8 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -64,7 +64,7 @@ static int opti_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &opti_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -173,7 +173,7 @@ static int opti_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!printed_version++)
 		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(dev, ppi, &opti_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &opti_sht, NULL);
 }
 
 static const struct pci_device_id opti[] = {
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index 36ac147de178..4cd744456313 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -64,7 +64,7 @@ static int optidma_pre_reset(struct ata_link *link, unsigned long deadline)
 	if (ap->port_no && !pci_test_config_bits(pdev, &optidma_enable_bits))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -430,7 +430,7 @@ static int optidma_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (optiplus_with_udma(dev))
 		ppi[0] = &info_82c700_udma;
 
-	return ata_pci_init_one(dev, ppi, &optidma_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &optidma_sht, NULL);
 }
 
 static const struct pci_device_id optidma[] = {
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index 57efbf05c95f..2e206c5f869a 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -133,7 +133,7 @@ static struct scsi_host_template pcmcia_sht = {
 
 static struct ata_port_operations pcmcia_port_ops = {
 	.inherits	= &ata_sff_port_ops,
-	.data_xfer	= ata_data_xfer_noirq,
+	.data_xfer	= ata_sff_data_xfer_noirq,
 	.cable_detect	= ata_cable_40wire,
 	.set_mode	= pcmcia_set_mode,
 };
@@ -323,13 +323,13 @@ next_entry:
 		ap->ioaddr.cmd_addr = io_addr + 0x10 * p;
 		ap->ioaddr.altstatus_addr = ctl_addr + 0x10 * p;
 		ap->ioaddr.ctl_addr = ctl_addr + 0x10 * p;
-		ata_std_ports(&ap->ioaddr);
+		ata_sff_std_ports(&ap->ioaddr);
 
 		ata_port_desc(ap, "cmd 0x%lx ctl 0x%lx", io_base, ctl_base);
 	}
 
 	/* activate */
-	ret = ata_host_activate(host, pdev->irq.AssignedIRQ, ata_interrupt,
+	ret = ata_host_activate(host, pdev->irq.AssignedIRQ, ata_sff_interrupt,
 				IRQF_SHARED, &pcmcia_sht);
 	if (ret)
 		goto failed;
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
index d235c9f92d09..0e1c2c1134d3 100644
--- a/drivers/ata/pata_pdc2027x.c
+++ b/drivers/ata/pata_pdc2027x.c
@@ -248,7 +248,7 @@ static int pdc2027x_prereset(struct ata_link *link, unsigned long deadline)
 	/* Check whether port enabled */
 	if (!pdc2027x_port_enabled(link->ap))
 		return -ENOENT;
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
@@ -265,7 +265,7 @@ static unsigned long pdc2027x_mode_filter(struct ata_device *adev, unsigned long
 	struct ata_device *pair = ata_dev_pair(adev);
 
 	if (adev->class != ATA_DEV_ATA || adev->devno == 0 || pair == NULL)
-		return ata_pci_default_filter(adev, mask);
+		return ata_bmdma_mode_filter(adev, mask);
 
 	/* Check for slave of a Maxtor at UDMA6 */
 	ata_id_c_string(pair->id, model_num, ATA_ID_PROD,
@@ -274,7 +274,7 @@ static unsigned long pdc2027x_mode_filter(struct ata_device *adev, unsigned long
 	if (strstr(model_num, "Maxtor") == NULL && pair->dma_mode == XFER_UDMA_6)
 		mask &= ~ (1 << (6 + ATA_SHIFT_UDMA));
 
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 /**
@@ -759,8 +759,8 @@ static int __devinit pdc2027x_init_one(struct pci_dev *pdev, const struct pci_de
 		return -EIO;
 
 	pci_set_master(pdev);
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &pdc2027x_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &pdc2027x_sht);
 }
 
 /**
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
index 8214100e3ac1..d2673060bc8d 100644
--- a/drivers/ata/pata_pdc202xx_old.c
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -324,7 +324,7 @@ static int pdc202xx_init_one(struct pci_dev *dev, const struct pci_device_id *id
 				return -ENODEV;
 		}
 	}
-	return ata_pci_init_one(dev, ppi, &pdc202xx_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &pdc202xx_sht, NULL);
 }
 
 static const struct pci_device_id pdc202xx[] = {
diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c
index 0588c9b7e73e..1edfc13d05d2 100644
--- a/drivers/ata/pata_platform.c
+++ b/drivers/ata/pata_platform.c
@@ -52,7 +52,7 @@ static struct scsi_host_template pata_platform_sht = {
 
 static struct ata_port_operations pata_platform_port_ops = {
 	.inherits		= &ata_sff_port_ops,
-	.data_xfer		= ata_data_xfer_noirq,
+	.data_xfer		= ata_sff_data_xfer_noirq,
 	.cable_detect		= ata_cable_unknown,
 	.set_mode		= pata_platform_set_mode,
 	.port_start		= ATA_OP_NULL,
@@ -176,7 +176,7 @@ int __devinit __pata_platform_probe(struct device *dev,
 		      (unsigned long long)ctl_res->start);
 
 	/* activate */
-	return ata_host_activate(host, irq, irq ? ata_interrupt : NULL,
+	return ata_host_activate(host, irq, irq ? ata_sff_interrupt : NULL,
 				 irq_flags, &pata_platform_sht);
 }
 EXPORT_SYMBOL_GPL(__pata_platform_probe);
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
index d16b343d2a62..0d81dd5fdc38 100644
--- a/drivers/ata/pata_qdi.c
+++ b/drivers/ata/pata_qdi.c
@@ -102,14 +102,14 @@ static void qdi6580_set_piomode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	qdi_qc_issue_prot	-	command issue
+ *	qdi_qc_issue		-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
  *	this interface so that we can load the correct ATA timings.
  */
 
-static unsigned int qdi_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int qdi_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -121,7 +121,7 @@ static unsigned int qdi_qc_issue_prot(struct ata_queued_cmd *qc)
 			outb(qdi->clock[adev->devno], qdi->timing);
 		}
 	}
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static unsigned int qdi_data_xfer(struct ata_device *dev, unsigned char *buf,
@@ -148,7 +148,7 @@ static unsigned int qdi_data_xfer(struct ata_device *dev, unsigned char *buf,
 			buflen += 4 - slop;
 		}
 	} else
-		buflen = ata_data_xfer(dev, buf, buflen, rw);
+		buflen = ata_sff_data_xfer(dev, buf, buflen, rw);
 
 	return buflen;
 }
@@ -159,7 +159,7 @@ static struct scsi_host_template qdi_sht = {
 
 static struct ata_port_operations qdi6500_port_ops = {
 	.inherits	= &ata_sff_port_ops,
-	.qc_issue	= qdi_qc_issue_prot,
+	.qc_issue	= qdi_qc_issue,
 	.data_xfer	= qdi_data_xfer,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= qdi6500_set_piomode,
@@ -223,7 +223,7 @@ static __init int qdi_init_one(unsigned long port, int type, unsigned long io, i
 	ap->ioaddr.cmd_addr = io_addr;
 	ap->ioaddr.altstatus_addr = ctl_addr;
 	ap->ioaddr.ctl_addr = ctl_addr;
-	ata_std_ports(&ap->ioaddr);
+	ata_sff_std_ports(&ap->ioaddr);
 
 	ata_port_desc(ap, "cmd %lx ctl %lx", io, ctl);
 
@@ -239,7 +239,7 @@ static __init int qdi_init_one(unsigned long port, int type, unsigned long io, i
 	printk(KERN_INFO DRV_NAME": qd%d at 0x%lx.\n", type, io);
 
 	/* activate */
-	ret = ata_host_activate(host, irq, ata_interrupt, 0, &qdi_sht);
+	ret = ata_host_activate(host, irq, ata_sff_interrupt, 0, &qdi_sht);
 	if (ret)
 		goto fail;
 
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
index 9ab84fc3798d..1c0d9fa7ee54 100644
--- a/drivers/ata/pata_radisys.c
+++ b/drivers/ata/pata_radisys.c
@@ -156,7 +156,7 @@ static void radisys_set_dmamode (struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	radisys_qc_issue_prot	-	command issue
+ *	radisys_qc_issue	-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -166,7 +166,7 @@ static void radisys_set_dmamode (struct ata_port *ap, struct ata_device *adev)
  *	be made PIO0.
  */
 
-static unsigned int radisys_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int radisys_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -180,7 +180,7 @@ static unsigned int radisys_qc_issue_prot(struct ata_queued_cmd *qc)
 				radisys_set_piomode(ap, adev);
 		}
 	}
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 
@@ -190,7 +190,7 @@ static struct scsi_host_template radisys_sht = {
 
 static struct ata_port_operations radisys_pata_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.qc_issue		= radisys_qc_issue_prot,
+	.qc_issue		= radisys_qc_issue,
 	.cable_detect		= ata_cable_unknown,
 	.set_piomode		= radisys_set_piomode,
 	.set_dmamode		= radisys_set_dmamode,
@@ -228,7 +228,7 @@ static int radisys_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 		dev_printk(KERN_DEBUG, &pdev->dev,
 			   "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(pdev, ppi, &radisys_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &radisys_sht, NULL);
 }
 
 static const struct pci_device_id radisys_pci_tbl[] = {
diff --git a/drivers/ata/pata_rb500_cf.c b/drivers/ata/pata_rb500_cf.c
index 7affceec1c29..5b23d79bc92b 100644
--- a/drivers/ata/pata_rb500_cf.c
+++ b/drivers/ata/pata_rb500_cf.c
@@ -57,7 +57,7 @@ static inline void rb500_pata_finish_io(struct ata_port *ap)
 	struct ata_host *ah = ap->host;
 	struct rb500_cf_info *info = ah->private_data;
 
-	ata_altstatus(ap);
+	ata_sff_altstatus(ap);
 	ndelay(RB500_CF_IO_DELAY);
 
 	set_irq_type(info->irq, IRQ_TYPE_LEVEL_HIGH);
@@ -109,7 +109,7 @@ static irqreturn_t rb500_pata_irq_handler(int irq, void *dev_instance)
 	if (gpio_get_value(info->gpio_line)) {
 		set_irq_type(info->irq, IRQ_TYPE_LEVEL_LOW);
 		if (!info->frozen)
-			ata_interrupt(info->irq, dev_instance);
+			ata_sff_interrupt(info->irq, dev_instance);
 	} else {
 		set_irq_type(info->irq, IRQ_TYPE_LEVEL_HIGH);
 	}
@@ -148,7 +148,7 @@ static void rb500_pata_setup_ports(struct ata_host *ah)
 	ap->ioaddr.ctl_addr	= info->iobase + RB500_CF_REG_CTRL;
 	ap->ioaddr.altstatus_addr = info->iobase + RB500_CF_REG_CTRL;
 
-	ata_std_ports(&ap->ioaddr);
+	ata_sff_std_ports(&ap->ioaddr);
 
 	ap->ioaddr.data_addr	= info->iobase + RB500_CF_REG_DATA;
 }
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
index 462b72a31280..7dfd1f3f6f3a 100644
--- a/drivers/ata/pata_rz1000.c
+++ b/drivers/ata/pata_rz1000.c
@@ -99,7 +99,7 @@ static int rz1000_init_one (struct pci_dev *pdev, const struct pci_device_id *en
 		printk(KERN_DEBUG DRV_NAME " version " DRV_VERSION "\n");
 
 	if (rz1000_fifo_disable(pdev) == 0)
-		return ata_pci_init_one(pdev, ppi, &rz1000_sht, NULL);
+		return ata_pci_sff_init_one(pdev, ppi, &rz1000_sht, NULL);
 
 	printk(KERN_ERR DRV_NAME ": failed to disable read-ahead on chipset..\n");
 	/* Not safe to use so skip */
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
index 42efacf73c79..cbab397e3db7 100644
--- a/drivers/ata/pata_sc1200.c
+++ b/drivers/ata/pata_sc1200.c
@@ -151,7 +151,7 @@ static void sc1200_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 }
 
 /**
- *	sc1200_qc_issue_prot	-	command issue
+ *	sc1200_qc_issue		-	command issue
  *	@qc: command pending
  *
  *	Called when the libata layer is about to issue a command. We wrap
@@ -160,7 +160,7 @@ static void sc1200_set_dmamode(struct ata_port *ap, struct ata_device *adev)
  *	one MWDMA/UDMA bit.
  */
 
-static unsigned int sc1200_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int sc1200_qc_issue(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct ata_device *adev = qc->dev;
@@ -175,7 +175,7 @@ static unsigned int sc1200_qc_issue_prot(struct ata_queued_cmd *qc)
 		    	sc1200_set_dmamode(ap, adev);
 	}
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static struct scsi_host_template sc1200_sht = {
@@ -185,8 +185,8 @@ static struct scsi_host_template sc1200_sht = {
 
 static struct ata_port_operations sc1200_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
-	.qc_prep 	= ata_dumb_qc_prep,
-	.qc_issue	= sc1200_qc_issue_prot,
+	.qc_prep 	= ata_sff_dumb_qc_prep,
+	.qc_issue	= sc1200_qc_issue,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= sc1200_set_piomode,
 	.set_dmamode	= sc1200_set_dmamode,
@@ -213,7 +213,7 @@ static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	/* Can't enable port 2 yet, see top comments */
 	const struct ata_port_info *ppi[] = { &info, &ata_dummy_port_info };
 
-	return ata_pci_init_one(dev, ppi, &sc1200_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &sc1200_sht, NULL);
 }
 
 static const struct pci_device_id sc1200[] = {
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index fba5bed0a641..701d0addae73 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -266,7 +266,7 @@ unsigned long scc_mode_filter(struct ata_device *adev, unsigned long mask)
 		printk(KERN_INFO "%s: limit ATAPI UDMA to UDMA4\n", DRV_NAME);
 		mask &= ~(0xE0 << ATA_SHIFT_UDMA);
 	}
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 /**
@@ -274,7 +274,7 @@ unsigned long scc_mode_filter(struct ata_device *adev, unsigned long mask)
  *	@ap: Port to which output is sent
  *	@tf: ATA taskfile register set
  *
- *	Note: Original code is ata_tf_load().
+ *	Note: Original code is ata_sff_tf_load().
  */
 
 static void scc_tf_load (struct ata_port *ap, const struct ata_taskfile *tf)
@@ -341,7 +341,7 @@ static u8 scc_check_status (struct ata_port *ap)
  *	@ap: Port from which input is read
  *	@tf: ATA taskfile register set for storing input
  *
- *	Note: Original code is ata_tf_read().
+ *	Note: Original code is ata_sff_tf_read().
  */
 
 static void scc_tf_read (struct ata_port *ap, struct ata_taskfile *tf)
@@ -373,7 +373,7 @@ static void scc_tf_read (struct ata_port *ap, struct ata_taskfile *tf)
  *	@ap: port to which command is being issued
  *	@tf: ATA taskfile register set
  *
- *	Note: Original code is ata_exec_command().
+ *	Note: Original code is ata_sff_exec_command().
  */
 
 static void scc_exec_command (struct ata_port *ap,
@@ -382,7 +382,7 @@ static void scc_exec_command (struct ata_port *ap,
 	DPRINTK("ata%u: cmd 0x%X\n", ap->print_id, tf->command);
 
 	out_be32(ap->ioaddr.command_addr, tf->command);
-	ata_pause(ap);
+	ata_sff_pause(ap);
 }
 
 /**
@@ -396,14 +396,14 @@ static u8 scc_check_altstatus (struct ata_port *ap)
 }
 
 /**
- *	scc_std_dev_select - Select device 0/1 on ATA bus
+ *	scc_dev_select - Select device 0/1 on ATA bus
  *	@ap: ATA channel to manipulate
  *	@device: ATA device (numbered from zero) to select
  *
- *	Note: Original code is ata_std_dev_select().
+ *	Note: Original code is ata_sff_dev_select().
  */
 
-static void scc_std_dev_select (struct ata_port *ap, unsigned int device)
+static void scc_dev_select (struct ata_port *ap, unsigned int device)
 {
 	u8 tmp;
 
@@ -413,7 +413,7 @@ static void scc_std_dev_select (struct ata_port *ap, unsigned int device)
 		tmp = ATA_DEVICE_OBS | ATA_DEV1;
 
 	out_be32(ap->ioaddr.device_addr, tmp);
-	ata_pause(ap);
+	ata_sff_pause(ap);
 }
 
 /**
@@ -514,7 +514,7 @@ static int scc_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	 * BSY bit to clear
 	 */
 	if (dev0) {
-		rc = ata_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(ap, deadline);
 		if (rc && rc != -ENODEV)
 			return rc;
 	}
@@ -535,7 +535,7 @@ static int scc_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 		msleep(50);	/* give drive a breather */
 	}
 	if (dev1) {
-		rc = ata_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(ap, deadline);
 		if (rc && rc != -ENODEV)
 			return rc;
 	}
@@ -571,7 +571,7 @@ static unsigned int scc_bus_softreset(struct ata_port *ap, unsigned int devmask,
 	out_be32(ioaddr->ctl_addr, ap->ctl);
 
 	/* wait a while before checking status */
-	ata_wait_after_reset(ap, deadline);
+	ata_sff_wait_after_reset(ap, deadline);
 
 	/* Before we perform post reset processing we want to see if
 	 * the bus shows 0xFF because the odd clown forgets the D7
@@ -586,16 +586,16 @@ static unsigned int scc_bus_softreset(struct ata_port *ap, unsigned int devmask,
 }
 
 /**
- *	scc_std_softreset - reset host port via ATA SRST
+ *	scc_softreset - reset host port via ATA SRST
  *	@ap: port to reset
  *	@classes: resulting classes of attached devices
  *	@deadline: deadline jiffies for the operation
  *
- *	Note: Original code is ata_std_softreset().
+ *	Note: Original code is ata_sff_softreset().
  */
 
-static int scc_std_softreset(struct ata_link *link, unsigned int *classes,
-                             unsigned long deadline)
+static int scc_softreset(struct ata_link *link, unsigned int *classes,
+			 unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
 	unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS;
@@ -628,10 +628,10 @@ static int scc_std_softreset(struct ata_link *link, unsigned int *classes,
 	}
 
 	/* determine by signature whether we have ATA or ATAPI devices */
-	classes[0] = ata_dev_try_classify(&ap->link.device[0],
+	classes[0] = ata_sff_dev_classify(&ap->link.device[0],
 					  devmask & (1 << 0), &err);
 	if (slave_possible && err != 0x81)
-		classes[1] = ata_dev_try_classify(&ap->link.device[1],
+		classes[1] = ata_sff_dev_classify(&ap->link.device[1],
 						  devmask & (1 << 1), &err);
 
  out:
@@ -695,7 +695,7 @@ static void scc_bmdma_stop (struct ata_queued_cmd *qc)
 			printk(KERN_WARNING "%s: Internal Bus Error\n", DRV_NAME);
 			out_be32(bmid_base + SCC_DMA_INTST, INTSTS_BMSINT);
 			/* TBD: SW reset */
-			scc_std_softreset(&ap->link, &classes, deadline);
+			scc_softreset(&ap->link, &classes, deadline);
 			continue;
 		}
 
@@ -721,7 +721,7 @@ static void scc_bmdma_stop (struct ata_queued_cmd *qc)
 		 in_be32(bmid_base + SCC_DMA_CMD) & ~ATA_DMA_START);
 
 	/* one-PIO-cycle guaranteed wait, per spec, for HDMA1:0 transition */
-	ata_altstatus(ap);	/* dummy read */
+	ata_sff_altstatus(ap);	/* dummy read */
 }
 
 /**
@@ -742,7 +742,7 @@ static u8 scc_bmdma_status (struct ata_port *ap)
 		return host_stat;
 
 	/* errata A252,A308 workaround: Step4 */
-	if ((ata_altstatus(ap) & ATA_ERR) && (int_status & INTSTS_INTRQ))
+	if ((ata_sff_altstatus(ap) & ATA_ERR) && (int_status & INTSTS_INTRQ))
 		return (host_stat | ATA_DMA_INTR);
 
 	/* errata A308 workaround Step5 */
@@ -773,7 +773,7 @@ static u8 scc_bmdma_status (struct ata_port *ap)
  *	@buflen: buffer length
  *	@rw: read/write
  *
- *	Note: Original code is ata_data_xfer().
+ *	Note: Original code is ata_sff_data_xfer().
  */
 
 static unsigned int scc_data_xfer (struct ata_device *dev, unsigned char *buf,
@@ -815,7 +815,7 @@ static unsigned int scc_data_xfer (struct ata_device *dev, unsigned char *buf,
  *	scc_irq_on - Enable interrupts on a port.
  *	@ap: Port on which interrupts are enabled.
  *
- *	Note: Original code is ata_irq_on().
+ *	Note: Original code is ata_sff_irq_on().
  */
 
 static u8 scc_irq_on (struct ata_port *ap)
@@ -835,13 +835,13 @@ static u8 scc_irq_on (struct ata_port *ap)
 }
 
 /**
- *	scc_bmdma_freeze - Freeze BMDMA controller port
+ *	scc_freeze - Freeze BMDMA controller port
  *	@ap: port to freeze
  *
- *	Note: Original code is ata_bmdma_freeze().
+ *	Note: Original code is ata_sff_freeze().
  */
 
-static void scc_bmdma_freeze (struct ata_port *ap)
+static void scc_freeze (struct ata_port *ap)
 {
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 
@@ -868,18 +868,18 @@ static void scc_bmdma_freeze (struct ata_port *ap)
 static int scc_pata_prereset(struct ata_link *link, unsigned long deadline)
 {
 	link->ap->cbl = ATA_CBL_PATA80;
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 /**
- *	scc_std_postreset - standard postreset callback
+ *	scc_postreset - standard postreset callback
  *	@ap: the target ata_port
  *	@classes: classes of attached devices
  *
- *	Note: Original code is ata_std_postreset().
+ *	Note: Original code is ata_sff_postreset().
  */
 
-static void scc_std_postreset(struct ata_link *link, unsigned int *classes)
+static void scc_postreset(struct ata_link *link, unsigned int *classes)
 {
 	struct ata_port *ap = link->ap;
 
@@ -905,13 +905,13 @@ static void scc_std_postreset(struct ata_link *link, unsigned int *classes)
 }
 
 /**
- *	scc_bmdma_irq_clear - Clear PCI IDE BMDMA interrupt.
+ *	scc_irq_clear - Clear PCI IDE BMDMA interrupt.
  *	@ap: Port associated with this ATA transaction.
  *
- *	Note: Original code is ata_bmdma_irq_clear().
+ *	Note: Original code is ata_sff_irq_clear().
  */
 
-static void scc_bmdma_irq_clear (struct ata_port *ap)
+static void scc_irq_clear (struct ata_port *ap)
 {
 	void __iomem *mmio = ap->ioaddr.bmdma_addr;
 
@@ -972,7 +972,7 @@ static struct ata_port_operations scc_pata_ops = {
 	.exec_command		= scc_exec_command,
 	.check_status		= scc_check_status,
 	.check_altstatus	= scc_check_altstatus,
-	.dev_select		= scc_std_dev_select,
+	.dev_select		= scc_dev_select,
 
 	.bmdma_setup		= scc_bmdma_setup,
 	.bmdma_start		= scc_bmdma_start,
@@ -980,13 +980,13 @@ static struct ata_port_operations scc_pata_ops = {
 	.bmdma_status		= scc_bmdma_status,
 	.data_xfer		= scc_data_xfer,
 
-	.freeze			= scc_bmdma_freeze,
+	.freeze			= scc_freeze,
 	.prereset		= scc_pata_prereset,
-	.softreset		= scc_std_softreset,
-	.postreset		= scc_std_postreset,
+	.softreset		= scc_softreset,
+	.postreset		= scc_postreset,
 	.post_internal_cmd	= scc_bmdma_stop,
 
-	.irq_clear		= scc_bmdma_irq_clear,
+	.irq_clear		= scc_irq_clear,
 	.irq_on			= scc_irq_on,
 
 	.port_start		= scc_port_start,
@@ -1140,8 +1140,8 @@ static int scc_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rc)
 		return rc;
 
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &scc_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &scc_sht);
 }
 
 static struct pci_driver scc_pci_driver = {
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index 2f4f9b0f89de..ffd26d0dc50d 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -199,7 +199,7 @@ static unsigned long serverworks_osb4_filter(struct ata_device *adev, unsigned l
 {
 	if (adev->class == ATA_DEV_ATA)
 		mask &= ~ATA_MASK_UDMA;
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 
@@ -219,7 +219,7 @@ static unsigned long serverworks_csb_filter(struct ata_device *adev, unsigned lo
 
 	/* Disk, UDMA */
 	if (adev->class != ATA_DEV_ATA)
-		return ata_pci_default_filter(adev, mask);
+		return ata_bmdma_mode_filter(adev, mask);
 
 	/* Actually do need to check */
 	ata_id_c_string(adev->id, model_num, ATA_ID_PROD, sizeof(model_num));
@@ -228,7 +228,7 @@ static unsigned long serverworks_csb_filter(struct ata_device *adev, unsigned lo
 		if (!strcmp(p, model_num))
 			mask &= ~(0xE0 << ATA_SHIFT_UDMA);
 	}
-	return ata_pci_default_filter(adev, mask);
+	return ata_bmdma_mode_filter(adev, mask);
 }
 
 /**
@@ -459,9 +459,9 @@ static int serverworks_init_one(struct pci_dev *pdev, const struct pci_device_id
 		serverworks_fixup_ht1000(pdev);
 
 	if (pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB5IDE)
-		ata_pci_clear_simplex(pdev);
+		ata_pci_bmdma_clear_simplex(pdev);
 
-	return ata_pci_init_one(pdev, ppi, &serverworks_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &serverworks_sht, NULL);
 }
 
 #ifdef CONFIG_PM
@@ -482,7 +482,7 @@ static int serverworks_reinit_one(struct pci_dev *pdev)
 			serverworks_fixup_osb4(pdev);
 			break;
 		case PCI_DEVICE_ID_SERVERWORKS_CSB5IDE:
-			ata_pci_clear_simplex(pdev);
+			ata_pci_bmdma_clear_simplex(pdev);
 			/* fall through */
 		case PCI_DEVICE_ID_SERVERWORKS_CSB6IDE:
 		case PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2:
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index 63fafc6d6da3..720b8645f58a 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -350,19 +350,19 @@ static int __devinit sil680_init_one(struct pci_dev *pdev,
 	host->ports[0]->ioaddr.cmd_addr = mmio_base + 0x80;
 	host->ports[0]->ioaddr.ctl_addr = mmio_base + 0x8a;
 	host->ports[0]->ioaddr.altstatus_addr = mmio_base + 0x8a;
-	ata_std_ports(&host->ports[0]->ioaddr);
+	ata_sff_std_ports(&host->ports[0]->ioaddr);
 	host->ports[1]->ioaddr.bmdma_addr = mmio_base + 0x08;
 	host->ports[1]->ioaddr.cmd_addr = mmio_base + 0xc0;
 	host->ports[1]->ioaddr.ctl_addr = mmio_base + 0xca;
 	host->ports[1]->ioaddr.altstatus_addr = mmio_base + 0xca;
-	ata_std_ports(&host->ports[1]->ioaddr);
+	ata_sff_std_ports(&host->ports[1]->ioaddr);
 
 	/* Register & activate */
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &sil680_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &sil680_sht);
 
 use_ioports:
-	return ata_pci_init_one(pdev, ppi, &sil680_sht, NULL);
+	return ata_pci_sff_init_one(pdev, ppi, &sil680_sht, NULL);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index 793e6714df8c..e82c66e8d31b 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -156,7 +156,7 @@ static int sis_pre_reset(struct ata_link *link, unsigned long deadline)
 	/* Clear the FIFO settings. We can't enable the FIFO until
 	   we know we are poking at a disk */
 	pci_write_config_byte(pdev, 0x4B, 0);
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 
@@ -821,7 +821,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	sis_fixup(pdev, chipset);
 
-	return ata_pci_init_one(pdev, ppi, &sis_sht, chipset);
+	return ata_pci_sff_init_one(pdev, ppi, &sis_sht, chipset);
 }
 
 static const struct pci_device_id sis_pci_tbl[] = {
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index bee11ca8f55a..70d94fb28a5f 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -60,7 +60,7 @@ static int sl82c105_pre_reset(struct ata_link *link, unsigned long deadline)
 
 	if (ap->port_no && !pci_test_config_bits(pdev, &sl82c105_enable_bits[ap->port_no]))
 		return -ENOENT;
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 
@@ -317,7 +317,7 @@ static int sl82c105_init_one(struct pci_dev *dev, const struct pci_device_id *id
 	val |= CTRL_P0EN | CTRL_P0F16 | CTRL_P1F16;
 	pci_write_config_dword(dev, 0x40, val);
 
-	return ata_pci_init_one(dev, ppi, &sl82c105_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &sl82c105_sht, NULL);
 }
 
 static const struct pci_device_id sl82c105[] = {
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index bd546a389ce1..b181261f2743 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -66,7 +66,7 @@ static int triflex_prereset(struct ata_link *link, unsigned long deadline)
 	if (!pci_test_config_bits(pdev, &triflex_enable_bits[ap->port_no]))
 		return -ENOENT;
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 
@@ -201,7 +201,7 @@ static int triflex_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 	if (!printed_version++)
 		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
 
-	return ata_pci_init_one(dev, ppi, &triflex_sht, NULL);
+	return ata_pci_sff_init_one(dev, ppi, &triflex_sht, NULL);
 }
 
 static const struct pci_device_id triflex[] = {
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index 2928fa173132..4b85f84fbe76 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -210,7 +210,7 @@ static int via_pre_reset(struct ata_link *link, unsigned long deadline)
 			return -ENOENT;
 	}
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 
@@ -336,7 +336,7 @@ static struct ata_port_operations via_port_ops = {
 
 static struct ata_port_operations via_port_ops_noirq = {
 	.inherits	= &via_port_ops,
-	.data_xfer	= ata_data_xfer_noirq,
+	.data_xfer	= ata_sff_data_xfer_noirq,
 };
 
 /**
@@ -511,7 +511,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	/* We have established the device type, now fire it up */
-	return ata_pci_init_one(pdev, ppi, &via_sht, (void *)config);
+	return ata_pci_sff_init_one(pdev, ppi, &via_sht, (void *)config);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/ata/pata_winbond.c b/drivers/ata/pata_winbond.c
index f235bb0d6139..cc18231e9334 100644
--- a/drivers/ata/pata_winbond.c
+++ b/drivers/ata/pata_winbond.c
@@ -116,7 +116,7 @@ static unsigned int winbond_data_xfer(struct ata_device *dev,
 			buflen += 4 - slop;
 		}
 	} else
-		buflen = ata_data_xfer(dev, buf, buflen, rw);
+		buflen = ata_sff_data_xfer(dev, buf, buflen, rw);
 
 	return buflen;
 }
@@ -198,7 +198,7 @@ static __init int winbond_init_one(unsigned long port)
 		ap->ioaddr.cmd_addr = cmd_addr;
 		ap->ioaddr.altstatus_addr = ctl_addr;
 		ap->ioaddr.ctl_addr = ctl_addr;
-		ata_std_ports(&ap->ioaddr);
+		ata_sff_std_ports(&ap->ioaddr);
 
 		/* hook in a private data structure per channel */
 		host->private_data = &winbond_data[nr_winbond_host];
@@ -206,7 +206,7 @@ static __init int winbond_init_one(unsigned long port)
 		winbond_data[nr_winbond_host].platform_dev = pdev;
 
 		/* activate */
-		rc = ata_host_activate(host, 14 + i, ata_interrupt, 0,
+		rc = ata_host_activate(host, 14 + i, ata_sff_interrupt, 0,
 				       &winbond_sht);
 		if (rc)
 			goto err_unregister;
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index c431bf36f9ba..be53545c9f64 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -208,7 +208,7 @@ static void adma_reinit_engine(struct ata_port *ap)
 
 	/* mask/clear ATA interrupts */
 	writeb(ATA_NIEN, ap->ioaddr.ctl_addr);
-	ata_check_status(ap);
+	ata_sff_check_status(ap);
 
 	/* reset the ADMA engine */
 	adma_reset_engine(ap);
@@ -243,7 +243,7 @@ static void adma_freeze(struct ata_port *ap)
 
 	/* mask/clear ATA interrupts */
 	writeb(ATA_NIEN, ap->ioaddr.ctl_addr);
-	ata_check_status(ap);
+	ata_sff_check_status(ap);
 
 	/* reset ADMA to idle state */
 	writew(aPIOMD4 | aNIEN | aRSTADM, chan + ADMA_CONTROL);
@@ -266,7 +266,7 @@ static int adma_prereset(struct ata_link *link, unsigned long deadline)
 		pp->state = adma_state_mmio;
 	adma_reinit_engine(ap);
 
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 static int adma_fill_sg(struct ata_queued_cmd *qc)
@@ -322,7 +322,7 @@ static void adma_qc_prep(struct ata_queued_cmd *qc)
 
 	adma_enter_reg_mode(qc->ap);
 	if (qc->tf.protocol != ATA_PROT_DMA) {
-		ata_qc_prep(qc);
+		ata_sff_qc_prep(qc);
 		return;
 	}
 
@@ -421,7 +421,7 @@ static unsigned int adma_qc_issue(struct ata_queued_cmd *qc)
 	}
 
 	pp->state = adma_state_mmio;
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static inline unsigned int adma_intr_pkt(struct ata_host *host)
@@ -492,7 +492,7 @@ static inline unsigned int adma_intr_mmio(struct ata_host *host)
 			if (qc && (!(qc->tf.flags & ATA_TFLAG_POLLING))) {
 
 				/* check main status, clearing INTRQ */
-				u8 status = ata_check_status(ap);
+				u8 status = ata_sff_check_status(ap);
 				if ((status & ATA_BUSY))
 					continue;
 				DPRINTK("ata%u: protocol %d (dev_stat 0x%X)\n",
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 12fbf3868fe2..6e6fca4c20b6 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -271,7 +271,7 @@ static void inic_host_intr(struct ata_port *ap)
 			return;
 		}
 
-		if (likely(ata_host_intr(ap, qc)))
+		if (likely(ata_sff_host_intr(ap, qc)))
 			return;
 
 		ap->ops->check_status(ap); /* clear ATA interrupt */
@@ -356,7 +356,7 @@ static unsigned int inic_qc_issue(struct ata_queued_cmd *qc)
 			return AC_ERR_HSM;
 	}
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static void inic_freeze(struct ata_port *ap)
@@ -418,9 +418,9 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class,
 		struct ata_taskfile tf;
 
 		/* wait a while before checking status */
-		ata_wait_after_reset(ap, deadline);
+		ata_sff_wait_after_reset(ap, deadline);
 
-		rc = ata_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(ap, deadline);
 		/* link occupied, -ENODEV too is an error */
 		if (rc) {
 			ata_link_printk(link, KERN_WARNING, "device not ready "
@@ -428,7 +428,7 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class,
 			return rc;
 		}
 
-		ata_tf_read(ap, &tf);
+		ata_sff_tf_read(ap, &tf);
 		*class = ata_dev_classify(&tf);
 		if (*class == ATA_DEV_UNKNOWN)
 			*class = ATA_DEV_NONE;
@@ -663,7 +663,7 @@ static int inic_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			((unsigned long)iomap[2 * i + 1] | ATA_PCI_CTL_OFS);
 		port->scr_addr = iomap[MMIO_BAR] + offset + PORT_SCR;
 
-		ata_std_ports(port);
+		ata_sff_std_ports(port);
 
 		ata_port_pbar_desc(ap, MMIO_BAR, -1, "mmio");
 		ata_port_pbar_desc(ap, MMIO_BAR, offset, "port");
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index 16c15ed3536e..fa75df634c75 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -1386,7 +1386,7 @@ static unsigned int mv_qc_issue(struct ata_queued_cmd *qc)
 		 * shadow block, etc registers.
 		 */
 		mv_stop_edma(ap);
-		return ata_qc_issue_prot(qc);
+		return ata_sff_qc_issue(qc);
 	}
 
 	mv_start_dma(ap, port_mmio, pp, qc->tf.protocol);
@@ -2362,7 +2362,7 @@ comreset_retry:
 	 */
 	retry = 20;
 	while (1) {
-		u8 drv_stat = ata_check_status(ap);
+		u8 drv_stat = ata_sff_check_status(ap);
 		if ((drv_stat != 0x80) && (drv_stat != 0x7f))
 			break;
 		msleep(500);
@@ -2377,7 +2377,7 @@ comreset_retry:
 	 */
 
 	/* finally, read device signature from TF registers */
-	*class = ata_dev_try_classify(ap->link.device, 1, NULL);
+	*class = ata_sff_dev_classify(ap->link.device, 1, NULL);
 
 	writelfl(0, port_mmio + EDMA_ERR_IRQ_CAUSE_OFS);
 
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 95fd0cd28b49..63cc43765f04 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -730,7 +730,7 @@ static void nv_adma_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
 	   ADMA mode could abort outstanding commands. */
 	nv_adma_register_mode(ap);
 
-	ata_tf_read(ap, tf);
+	ata_sff_tf_read(ap, tf);
 }
 
 static unsigned int nv_adma_tf_to_cpb(struct ata_taskfile *tf, __le16 *cpb)
@@ -844,12 +844,12 @@ static int nv_host_intr(struct ata_port *ap, u8 irq_stat)
 
 	/* DEV interrupt w/ no active qc? */
 	if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) {
-		ata_check_status(ap);
+		ata_sff_check_status(ap);
 		return 1;
 	}
 
 	/* handle interrupt */
-	return ata_host_intr(ap, qc);
+	return ata_sff_host_intr(ap, qc);
 }
 
 static irqreturn_t nv_adma_interrupt(int irq, void *dev_instance)
@@ -1028,7 +1028,7 @@ static void nv_adma_irq_clear(struct ata_port *ap)
 	u32 notifier_clears[2];
 
 	if (pp->flags & NV_ADMA_ATAPI_SETUP_COMPLETE) {
-		ata_bmdma_irq_clear(ap);
+		ata_sff_irq_clear(ap);
 		return;
 	}
 
@@ -1059,7 +1059,7 @@ static void nv_adma_post_internal_cmd(struct ata_queued_cmd *qc)
 	struct nv_adma_port_priv *pp = qc->ap->private_data;
 
 	if (pp->flags & NV_ADMA_PORT_REGISTER_MODE)
-		ata_bmdma_post_internal_cmd(qc);
+		ata_sff_post_internal_cmd(qc);
 }
 
 static int nv_adma_port_start(struct ata_port *ap)
@@ -1336,7 +1336,7 @@ static void nv_adma_qc_prep(struct ata_queued_cmd *qc)
 		BUG_ON(!(pp->flags & NV_ADMA_ATAPI_SETUP_COMPLETE) &&
 			(qc->flags & ATA_QCFLAG_DMAMAP));
 		nv_adma_register_mode(qc->ap);
-		ata_qc_prep(qc);
+		ata_sff_qc_prep(qc);
 		return;
 	}
 
@@ -1395,7 +1395,7 @@ static unsigned int nv_adma_qc_issue(struct ata_queued_cmd *qc)
 		BUG_ON(!(pp->flags & NV_ADMA_ATAPI_SETUP_COMPLETE) &&
 			(qc->flags & ATA_QCFLAG_DMAMAP));
 		nv_adma_register_mode(qc->ap);
-		return ata_qc_issue_prot(qc);
+		return ata_sff_qc_issue(qc);
 	} else
 		nv_adma_mode(qc->ap);
 
@@ -1436,7 +1436,7 @@ static irqreturn_t nv_generic_interrupt(int irq, void *dev_instance)
 
 			qc = ata_qc_from_tag(ap, ap->link.active_tag);
 			if (qc && (!(qc->tf.flags & ATA_TFLAG_POLLING)))
-				handled += ata_host_intr(ap, qc);
+				handled += ata_sff_host_intr(ap, qc);
 			else
 				// No request pending?  Clear interrupt status
 				// anyway, in case there's one pending.
@@ -1571,7 +1571,7 @@ static void nv_mcp55_freeze(struct ata_port *ap)
 	mask = readl(mmio_base + NV_INT_ENABLE_MCP55);
 	mask &= ~(NV_INT_ALL_MCP55 << shift);
 	writel(mask, mmio_base + NV_INT_ENABLE_MCP55);
-	ata_bmdma_freeze(ap);
+	ata_sff_freeze(ap);
 }
 
 static void nv_mcp55_thaw(struct ata_port *ap)
@@ -1585,7 +1585,7 @@ static void nv_mcp55_thaw(struct ata_port *ap)
 	mask = readl(mmio_base + NV_INT_ENABLE_MCP55);
 	mask |= (NV_INT_MASK_MCP55 << shift);
 	writel(mask, mmio_base + NV_INT_ENABLE_MCP55);
-	ata_bmdma_thaw(ap);
+	ata_sff_thaw(ap);
 }
 
 static int nv_hardreset(struct ata_link *link, unsigned int *class,
@@ -1597,7 +1597,7 @@ static int nv_hardreset(struct ata_link *link, unsigned int *class,
 	 * some controllers.  Don't classify on hardreset.  For more
 	 * info, see http://bugzilla.kernel.org/show_bug.cgi?id=3352
 	 */
-	return sata_std_hardreset(link, &dummy, deadline);
+	return sata_sff_hardreset(link, &dummy, deadline);
 }
 
 static void nv_adma_error_handler(struct ata_port *ap)
@@ -1653,7 +1653,7 @@ static void nv_adma_error_handler(struct ata_port *ap)
 		readw(mmio + NV_ADMA_CTL);	/* flush posted write */
 	}
 
-	ata_bmdma_error_handler(ap);
+	ata_sff_error_handler(ap);
 }
 
 static void nv_swncq_qc_to_dq(struct ata_port *ap, struct ata_queued_cmd *qc)
@@ -1779,7 +1779,7 @@ static void nv_swncq_error_handler(struct ata_port *ap)
 		ehc->i.action |= ATA_EH_RESET;
 	}
 
-	ata_bmdma_error_handler(ap);
+	ata_sff_error_handler(ap);
 }
 
 #ifdef CONFIG_PM
@@ -1925,7 +1925,7 @@ static int nv_swncq_port_start(struct ata_port *ap)
 static void nv_swncq_qc_prep(struct ata_queued_cmd *qc)
 {
 	if (qc->tf.protocol != ATA_PROT_NCQ) {
-		ata_qc_prep(qc);
+		ata_sff_qc_prep(qc);
 		return;
 	}
 
@@ -2001,7 +2001,7 @@ static unsigned int nv_swncq_qc_issue(struct ata_queued_cmd *qc)
 	struct nv_swncq_port_priv *pp = ap->private_data;
 
 	if (qc->tf.protocol != ATA_PROT_NCQ)
-		return ata_qc_issue_prot(qc);
+		return ata_sff_qc_issue(qc);
 
 	DPRINTK("Enter\n");
 
@@ -2350,7 +2350,7 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	ppi[0] = &nv_port_info[type];
 	ipriv = ppi[0]->private_data;
-	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
+	rc = ata_pci_sff_prepare_host(pdev, ppi, &host);
 	if (rc)
 		return rc;
 
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index 5a0a03a08abc..9923e860eae3 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -143,7 +143,7 @@ static void pdc_exec_command_mmio(struct ata_port *ap, const struct ata_taskfile
 static int pdc_check_atapi_dma(struct ata_queued_cmd *qc);
 static int pdc_old_sata_check_atapi_dma(struct ata_queued_cmd *qc);
 static void pdc_irq_clear(struct ata_port *ap);
-static unsigned int pdc_qc_issue_prot(struct ata_queued_cmd *qc);
+static unsigned int pdc_qc_issue(struct ata_queued_cmd *qc);
 static void pdc_freeze(struct ata_port *ap);
 static void pdc_sata_freeze(struct ata_port *ap);
 static void pdc_thaw(struct ata_port *ap);
@@ -166,7 +166,7 @@ static const struct ata_port_operations pdc_common_ops = {
 	.exec_command		= pdc_exec_command_mmio,
 	.check_atapi_dma	= pdc_check_atapi_dma,
 	.qc_prep		= pdc_qc_prep,
-	.qc_issue		= pdc_qc_issue_prot,
+	.qc_issue		= pdc_qc_issue,
 	.irq_clear		= pdc_irq_clear,
 
 	.post_internal_cmd	= pdc_post_internal_cmd,
@@ -894,7 +894,7 @@ static inline void pdc_packet_start(struct ata_queued_cmd *qc)
 	readl(ap->ioaddr.cmd_addr + PDC_PKT_SUBMIT); /* flush */
 }
 
-static unsigned int pdc_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int pdc_qc_issue(struct ata_queued_cmd *qc)
 {
 	switch (qc->tf.protocol) {
 	case ATAPI_PROT_NODATA:
@@ -914,20 +914,20 @@ static unsigned int pdc_qc_issue_prot(struct ata_queued_cmd *qc)
 		break;
 	}
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static void pdc_tf_load_mmio(struct ata_port *ap, const struct ata_taskfile *tf)
 {
 	WARN_ON(tf->protocol == ATA_PROT_DMA || tf->protocol == ATAPI_PROT_DMA);
-	ata_tf_load(ap, tf);
+	ata_sff_tf_load(ap, tf);
 }
 
 static void pdc_exec_command_mmio(struct ata_port *ap,
 				  const struct ata_taskfile *tf)
 {
 	WARN_ON(tf->protocol == ATA_PROT_DMA || tf->protocol == ATAPI_PROT_DMA);
-	ata_exec_command(ap, tf);
+	ata_sff_exec_command(ap, tf);
 }
 
 static int pdc_check_atapi_dma(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 2ceb0990bcd8..1600107047cf 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -239,7 +239,7 @@ static int qs_prereset(struct ata_link *link, unsigned long deadline)
 	struct ata_port *ap = link->ap;
 
 	qs_reset_channel_logic(ap);
-	return ata_std_prereset(link, deadline);
+	return ata_sff_prereset(link, deadline);
 }
 
 static int qs_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
@@ -303,7 +303,7 @@ static void qs_qc_prep(struct ata_queued_cmd *qc)
 
 	qs_enter_reg_mode(qc->ap);
 	if (qc->tf.protocol != ATA_PROT_DMA) {
-		ata_qc_prep(qc);
+		ata_sff_qc_prep(qc);
 		return;
 	}
 
@@ -362,7 +362,7 @@ static unsigned int qs_qc_issue(struct ata_queued_cmd *qc)
 	}
 
 	pp->state = qs_state_mmio;
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static void qs_do_or_die(struct ata_queued_cmd *qc, u8 status)
@@ -451,7 +451,7 @@ static inline unsigned int qs_intr_mmio(struct ata_host *host)
 				 * and pretend we knew it was ours.. (ugh).
 				 * This does not affect packet mode.
 				 */
-				ata_check_status(ap);
+				ata_sff_check_status(ap);
 				handled = 1;
 				continue;
 			}
@@ -459,7 +459,7 @@ static inline unsigned int qs_intr_mmio(struct ata_host *host)
 			if (!pp || pp->state != qs_state_mmio)
 				continue;
 			if (!(qc->tf.flags & ATA_TFLAG_POLLING))
-				handled |= ata_host_intr(ap, qc);
+				handled |= ata_sff_host_intr(ap, qc);
 		}
 	}
 	return handled;
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index 659dfcbdc1b2..987313b68f3b 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -410,10 +410,10 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 		goto err_hsm;
 
 	/* ack bmdma irq events */
-	ata_bmdma_irq_clear(ap);
+	ata_sff_irq_clear(ap);
 
 	/* kick HSM in the ass */
-	ata_hsm_move(ap, qc, status, 0);
+	ata_sff_hsm_move(ap, qc, status, 0);
 
 	if (unlikely(qc->err_mask) && ata_is_dma(qc->tf.protocol))
 		ata_ehi_push_desc(ehi, "BMDMA2 stat 0x%x", bmdma2);
@@ -481,7 +481,7 @@ static void sil_thaw(struct ata_port *ap)
 
 	/* clear IRQ */
 	ap->ops->check_status(ap);
-	ata_bmdma_irq_clear(ap);
+	ata_sff_irq_clear(ap);
 
 	/* turn on SATA IRQ if supported */
 	if (!(ap->flags & SIL_FLAG_NO_SATA_IRQ))
@@ -655,7 +655,7 @@ static int sil_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		ioaddr->ctl_addr = mmio_base + sil_port[i].ctl;
 		ioaddr->bmdma_addr = mmio_base + sil_port[i].bmdma;
 		ioaddr->scr_addr = mmio_base + sil_port[i].scr;
-		ata_std_ports(ioaddr);
+		ata_sff_std_ports(ioaddr);
 
 		ata_port_pbar_desc(ap, SIL_MMIO_BAR, -1, "mmio");
 		ata_port_pbar_desc(ap, SIL_MMIO_BAR, sil_port[i].tf, "tf");
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index 9089c7ab5000..6b8e45ba32e8 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -309,7 +309,7 @@ static int sis_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		break;
 	}
 
-	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
+	rc = ata_pci_sff_prepare_host(pdev, ppi, &host);
 	if (rc)
 		return rc;
 
@@ -327,8 +327,8 @@ static int sis_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_master(pdev);
 	pci_intx(pdev, 1);
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &sis_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &sis_sht);
 }
 
 static int __init sis_init(void)
diff --git a/drivers/ata/sata_svw.c b/drivers/ata/sata_svw.c
index 8636f164256e..7b941106f7de 100644
--- a/drivers/ata/sata_svw.c
+++ b/drivers/ata/sata_svw.c
@@ -492,8 +492,8 @@ static int k2_sata_init_one(struct pci_dev *pdev, const struct pci_device_id *en
 	writel(0x0, mmio_base + K2_SATA_SIM_OFFSET);
 
 	pci_set_master(pdev);
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &k2_sata_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &k2_sata_sht);
 }
 
 /* 0x240 is device ID for Apple K2 device
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 8d7f39532d4a..40be2ff60902 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -232,7 +232,7 @@ static void pdc20621_get_from_dimm(struct ata_host *host,
 static void pdc20621_put_to_dimm(struct ata_host *host,
 				 void *psource, u32 offset, u32 size);
 static void pdc20621_irq_clear(struct ata_port *ap);
-static unsigned int pdc20621_qc_issue_prot(struct ata_queued_cmd *qc);
+static unsigned int pdc20621_qc_issue(struct ata_queued_cmd *qc);
 
 
 static struct scsi_host_template pdc_sata_sht = {
@@ -244,17 +244,17 @@ static struct scsi_host_template pdc_sata_sht = {
 /* TODO: inherit from base port_ops after converting to new EH */
 static struct ata_port_operations pdc_20621_ops = {
 	.tf_load		= pdc_tf_load_mmio,
-	.tf_read		= ata_tf_read,
-	.check_status		= ata_check_status,
+	.tf_read		= ata_sff_tf_read,
+	.check_status		= ata_sff_check_status,
 	.exec_command		= pdc_exec_command_mmio,
-	.dev_select		= ata_std_dev_select,
+	.dev_select		= ata_sff_dev_select,
 	.phy_reset		= pdc_20621_phy_reset,
 	.qc_prep		= pdc20621_qc_prep,
-	.qc_issue		= pdc20621_qc_issue_prot,
-	.data_xfer		= ata_data_xfer,
+	.qc_issue		= pdc20621_qc_issue,
+	.data_xfer		= ata_sff_data_xfer,
 	.eng_timeout		= pdc_eng_timeout,
 	.irq_clear		= pdc20621_irq_clear,
-	.irq_on			= ata_irq_on,
+	.irq_on			= ata_sff_irq_on,
 	.port_start		= pdc_port_start,
 };
 
@@ -682,7 +682,7 @@ static void pdc20621_packet_start(struct ata_queued_cmd *qc)
 	}
 }
 
-static unsigned int pdc20621_qc_issue_prot(struct ata_queued_cmd *qc)
+static unsigned int pdc20621_qc_issue(struct ata_queued_cmd *qc)
 {
 	switch (qc->tf.protocol) {
 	case ATA_PROT_DMA:
@@ -698,7 +698,7 @@ static unsigned int pdc20621_qc_issue_prot(struct ata_queued_cmd *qc)
 		break;
 	}
 
-	return ata_qc_issue_prot(qc);
+	return ata_sff_qc_issue(qc);
 }
 
 static inline unsigned int pdc20621_host_intr(struct ata_port *ap,
@@ -770,7 +770,7 @@ static inline unsigned int pdc20621_host_intr(struct ata_port *ap,
 	/* command completion, but no data xfer */
 	} else if (qc->tf.protocol == ATA_PROT_NODATA) {
 
-		status = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
+		status = ata_sff_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
 		DPRINTK("BUS_NODATA (drv_stat 0x%X)\n", status);
 		qc->err_mask |= ac_err_mask(status);
 		ata_qc_complete(qc);
@@ -879,7 +879,7 @@ static void pdc_eng_timeout(struct ata_port *ap)
 		break;
 
 	default:
-		drv_stat = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
+		drv_stat = ata_sff_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
 
 		ata_port_printk(ap, KERN_ERR,
 				"unknown timeout, cmd 0x%x stat 0x%x\n",
@@ -898,7 +898,7 @@ static void pdc_tf_load_mmio(struct ata_port *ap, const struct ata_taskfile *tf)
 {
 	WARN_ON(tf->protocol == ATA_PROT_DMA ||
 		tf->protocol == ATA_PROT_NODATA);
-	ata_tf_load(ap, tf);
+	ata_sff_tf_load(ap, tf);
 }
 
 
@@ -906,7 +906,7 @@ static void pdc_exec_command_mmio(struct ata_port *ap, const struct ata_taskfile
 {
 	WARN_ON(tf->protocol == ATA_PROT_DMA ||
 		tf->protocol == ATA_PROT_NODATA);
-	ata_exec_command(ap, tf);
+	ata_sff_exec_command(ap, tf);
 }
 
 
diff --git a/drivers/ata/sata_uli.c b/drivers/ata/sata_uli.c
index 6ecd13fefa1a..f277cea904ce 100644
--- a/drivers/ata/sata_uli.c
+++ b/drivers/ata/sata_uli.c
@@ -175,11 +175,11 @@ static int uli_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	host->private_data = hpriv;
 
 	/* the first two ports are standard SFF */
-	rc = ata_pci_init_sff_host(host);
+	rc = ata_pci_sff_init_host(host);
 	if (rc)
 		return rc;
 
-	rc = ata_pci_init_bmdma(host);
+	rc = ata_pci_bmdma_init(host);
 	if (rc)
 		return rc;
 
@@ -200,7 +200,7 @@ static int uli_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			((unsigned long)iomap[1] | ATA_PCI_CTL_OFS) + 4;
 		ioaddr->bmdma_addr = iomap[4] + 16;
 		hpriv->scr_cfg_addr[2] = ULI5287_BASE + ULI5287_OFFS*4;
-		ata_std_ports(ioaddr);
+		ata_sff_std_ports(ioaddr);
 
 		ata_port_desc(host->ports[2],
 			"cmd 0x%llx ctl 0x%llx bmdma 0x%llx",
@@ -215,7 +215,7 @@ static int uli_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			((unsigned long)iomap[3] | ATA_PCI_CTL_OFS) + 4;
 		ioaddr->bmdma_addr = iomap[4] + 24;
 		hpriv->scr_cfg_addr[3] = ULI5287_BASE + ULI5287_OFFS*5;
-		ata_std_ports(ioaddr);
+		ata_sff_std_ports(ioaddr);
 
 		ata_port_desc(host->ports[2],
 			"cmd 0x%llx ctl 0x%llx bmdma 0x%llx",
@@ -242,8 +242,8 @@ static int uli_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_master(pdev);
 	pci_intx(pdev, 1);
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &uli_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &uli_sht);
 }
 
 static int __init uli_init(void)
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 4bc6e849af2c..9323bc2a2785 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -174,7 +174,7 @@ static void svia_noop_freeze(struct ata_port *ap)
 	 * certain way.  Leave it alone and just clear pending IRQ.
 	 */
 	ap->ops->check_status(ap);
-	ata_bmdma_irq_clear(ap);
+	ata_sff_irq_clear(ap);
 }
 
 /**
@@ -242,7 +242,7 @@ static int vt6420_prereset(struct ata_link *link, unsigned long deadline)
 
  skip_scr:
 	/* wait for !BSY */
-	ata_wait_ready(ap, deadline);
+	ata_sff_wait_ready(ap, deadline);
 
 	return 0;
 }
@@ -304,7 +304,7 @@ static void vt6421_init_addrs(struct ata_port *ap)
 	ioaddr->bmdma_addr = bmdma_addr;
 	ioaddr->scr_addr = vt6421_scr_addr(iomap[5], ap->port_no);
 
-	ata_std_ports(ioaddr);
+	ata_sff_std_ports(ioaddr);
 
 	ata_port_pbar_desc(ap, ap->port_no, -1, "port");
 	ata_port_pbar_desc(ap, 4, ap->port_no * 8, "bmdma");
@@ -316,7 +316,7 @@ static int vt6420_prepare_host(struct pci_dev *pdev, struct ata_host **r_host)
 	struct ata_host *host;
 	int rc;
 
-	rc = ata_pci_prepare_sff_host(pdev, ppi, &host);
+	rc = ata_pci_sff_prepare_host(pdev, ppi, &host);
 	if (rc)
 		return rc;
 	*r_host = host;
@@ -448,8 +448,8 @@ static int svia_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	svia_configure(pdev);
 
 	pci_set_master(pdev);
-	return ata_host_activate(host, pdev->irq, ata_interrupt, IRQF_SHARED,
-				 &svia_sht);
+	return ata_host_activate(host, pdev->irq, ata_sff_interrupt,
+				 IRQF_SHARED, &svia_sht);
 }
 
 static int __init svia_init(void)
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index fb3a88722664..2c3c7693c750 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -200,7 +200,7 @@ static void vsc_sata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 	u16 nsect, lbal, lbam, lbah, feature;
 
-	tf->command = ata_check_status(ap);
+	tf->command = ata_sff_check_status(ap);
 	tf->device = readw(ioaddr->device_addr);
 	feature = readw(ioaddr->error_addr);
 	nsect = readw(ioaddr->nsect_addr);
@@ -243,7 +243,7 @@ static void vsc_port_intr(u8 port_status, struct ata_port *ap)
 
 	qc = ata_qc_from_tag(ap, ap->link.active_tag);
 	if (qc && likely(!(qc->tf.flags & ATA_TFLAG_POLLING)))
-		handled = ata_host_intr(ap, qc);
+		handled = ata_sff_host_intr(ap, qc);
 
 	/* We received an interrupt during a polled command,
 	 * or some other spurious condition.  Interrupt reporting
diff --git a/include/linux/libata.h b/include/linux/libata.h
index a05de2ba7a72..66663bfe2c71 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1342,45 +1342,48 @@ extern const struct ata_port_operations ata_bmdma_port_ops;
 	.sg_tablesize		= LIBATA_MAX_PRD,		\
 	.dma_boundary		= ATA_DMA_BOUNDARY
 
-extern void ata_qc_prep(struct ata_queued_cmd *qc);
-extern void ata_dumb_qc_prep(struct ata_queued_cmd *qc);
-extern void ata_std_dev_select(struct ata_port *ap, unsigned int device);
-extern u8 ata_check_status(struct ata_port *ap);
-extern u8 ata_altstatus(struct ata_port *ap);
-extern int ata_busy_sleep(struct ata_port *ap,
-			  unsigned long timeout_pat, unsigned long timeout);
-extern int ata_wait_ready(struct ata_port *ap, unsigned long deadline);
-extern void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf);
-extern void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
-extern void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf);
-extern unsigned int ata_data_xfer(struct ata_device *dev,
+extern void ata_sff_qc_prep(struct ata_queued_cmd *qc);
+extern void ata_sff_dumb_qc_prep(struct ata_queued_cmd *qc);
+extern void ata_sff_dev_select(struct ata_port *ap, unsigned int device);
+extern u8 ata_sff_check_status(struct ata_port *ap);
+extern u8 ata_sff_altstatus(struct ata_port *ap);
+extern int ata_sff_busy_sleep(struct ata_port *ap,
+			      unsigned long timeout_pat, unsigned long timeout);
+extern int ata_sff_wait_ready(struct ata_port *ap, unsigned long deadline);
+extern void ata_sff_tf_load(struct ata_port *ap, const struct ata_taskfile *tf);
+extern void ata_sff_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
+extern void ata_sff_exec_command(struct ata_port *ap,
+				 const struct ata_taskfile *tf);
+extern unsigned int ata_sff_data_xfer(struct ata_device *dev,
 			unsigned char *buf, unsigned int buflen, int rw);
-extern unsigned int ata_data_xfer_noirq(struct ata_device *dev,
+extern unsigned int ata_sff_data_xfer_noirq(struct ata_device *dev,
 			unsigned char *buf, unsigned int buflen, int rw);
-extern u8 ata_irq_on(struct ata_port *ap);
-extern void ata_bmdma_irq_clear(struct ata_port *ap);
-extern int ata_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
-			u8 status, int in_wq);
-extern unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc);
-extern unsigned int ata_host_intr(struct ata_port *ap, struct ata_queued_cmd *qc);
-extern irqreturn_t ata_interrupt(int irq, void *dev_instance);
-extern void ata_bmdma_freeze(struct ata_port *ap);
-extern void ata_bmdma_thaw(struct ata_port *ap);
-extern int ata_std_prereset(struct ata_link *link, unsigned long deadline);
-extern unsigned int ata_dev_try_classify(struct ata_device *dev, int present,
-					 u8 *r_err);
-extern void ata_wait_after_reset(struct ata_port *ap, unsigned long deadline);
-extern int ata_std_softreset(struct ata_link *link, unsigned int *classes,
+extern u8 ata_sff_irq_on(struct ata_port *ap);
+extern void ata_sff_irq_clear(struct ata_port *ap);
+extern int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
+			    u8 status, int in_wq);
+extern unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc);
+extern unsigned int ata_sff_host_intr(struct ata_port *ap,
+				      struct ata_queued_cmd *qc);
+extern irqreturn_t ata_sff_interrupt(int irq, void *dev_instance);
+extern void ata_sff_freeze(struct ata_port *ap);
+extern void ata_sff_thaw(struct ata_port *ap);
+extern int ata_sff_prereset(struct ata_link *link, unsigned long deadline);
+extern unsigned int ata_sff_dev_classify(struct ata_device *dev, int present,
+					  u8 *r_err);
+extern void ata_sff_wait_after_reset(struct ata_port *ap,
+				     unsigned long deadline);
+extern int ata_sff_softreset(struct ata_link *link, unsigned int *classes,
 			     unsigned long deadline);
-extern int sata_std_hardreset(struct ata_link *link, unsigned int *class,
-			      unsigned long deadline);
-extern void ata_std_postreset(struct ata_link *link, unsigned int *classes);
-extern void ata_bmdma_error_handler(struct ata_port *ap);
-extern void ata_bmdma_post_internal_cmd(struct ata_queued_cmd *qc);
+extern int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
+			       unsigned long deadline);
+extern void ata_sff_postreset(struct ata_link *link, unsigned int *classes);
+extern void ata_sff_error_handler(struct ata_port *ap);
+extern void ata_sff_post_internal_cmd(struct ata_queued_cmd *qc);
 extern int ata_sff_port_start(struct ata_port *ap);
-extern void ata_std_ports(struct ata_ioports *ioaddr);
-extern unsigned long ata_pci_default_filter(struct ata_device *dev,
-					    unsigned long xfer_mask);
+extern void ata_sff_std_ports(struct ata_ioports *ioaddr);
+extern unsigned long ata_bmdma_mode_filter(struct ata_device *dev,
+					   unsigned long xfer_mask);
 extern void ata_bmdma_setup(struct ata_queued_cmd *qc);
 extern void ata_bmdma_start(struct ata_queued_cmd *qc);
 extern void ata_bmdma_stop(struct ata_queued_cmd *qc);
@@ -1388,35 +1391,35 @@ extern u8 ata_bmdma_status(struct ata_port *ap);
 extern void ata_bus_reset(struct ata_port *ap);
 
 #ifdef CONFIG_PCI
-extern int ata_pci_clear_simplex(struct pci_dev *pdev);
-extern int ata_pci_init_bmdma(struct ata_host *host);
-extern int ata_pci_init_sff_host(struct ata_host *host);
-extern int ata_pci_prepare_sff_host(struct pci_dev *pdev,
+extern int ata_pci_bmdma_clear_simplex(struct pci_dev *pdev);
+extern int ata_pci_bmdma_init(struct ata_host *host);
+extern int ata_pci_sff_init_host(struct ata_host *host);
+extern int ata_pci_sff_prepare_host(struct pci_dev *pdev,
 				    const struct ata_port_info * const * ppi,
 				    struct ata_host **r_host);
-extern int ata_pci_activate_sff_host(struct ata_host *host,
+extern int ata_pci_sff_activate_host(struct ata_host *host,
 				     irq_handler_t irq_handler,
 				     struct scsi_host_template *sht);
-extern int ata_pci_init_one(struct pci_dev *pdev,
-			    const struct ata_port_info * const * ppi,
-			    struct scsi_host_template *sht, void *host_priv);
+extern int ata_pci_sff_init_one(struct pci_dev *pdev,
+				const struct ata_port_info * const * ppi,
+				struct scsi_host_template *sht, void *host_priv);
 #endif /* CONFIG_PCI */
 
 /**
- *	ata_pause - Flush writes and pause 400 nanoseconds.
+ *	ata_sff_pause - Flush writes and pause 400 nanoseconds.
  *	@ap: Port to wait for.
  *
  *	LOCKING:
  *	Inherited from caller.
  */
-static inline void ata_pause(struct ata_port *ap)
+static inline void ata_sff_pause(struct ata_port *ap)
 {
-	ata_altstatus(ap);
+	ata_sff_altstatus(ap);
 	ndelay(400);
 }
 
 /**
- *	ata_busy_wait - Wait for a port status register
+ *	ata_sff_busy_wait - Wait for a port status register
  *	@ap: Port to wait for.
  *	@bits: bits that must be clear
  *	@max: number of 10uS waits to perform
@@ -1428,8 +1431,8 @@ static inline void ata_pause(struct ata_port *ap)
  *	LOCKING:
  *	Inherited from caller.
  */
-static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits,
-			       unsigned int max)
+static inline u8 ata_sff_busy_wait(struct ata_port *ap, unsigned int bits,
+				   unsigned int max)
 {
 	u8 status;
 
@@ -1454,7 +1457,7 @@ static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits,
  */
 static inline u8 ata_wait_idle(struct ata_port *ap)
 {
-	u8 status = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
+	u8 status = ata_sff_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000);
 
 #ifdef ATA_DEBUG
 	if (status != 0xff && (status & (ATA_BUSY | ATA_DRQ)))
-- 
cgit v1.2.3


From 5682ed33aae05d10a25c95633ef9d9c062825888 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:16 +0900
Subject: libata: rename SFF port ops

Add sff_ prefix to SFF specific port ops.

This rename is in preparation of separating SFF support out of libata
core layer.  This patch strictly renames ops and doesn't introduce any
behavior difference.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c            |   6 +--
 drivers/ata/libata-core.c     |  18 +++----
 drivers/ata/libata-scsi.c     |   2 +-
 drivers/ata/libata-sff.c      | 109 ++++++++++++++++++++++--------------------
 drivers/ata/pata_bf54x.c      |  20 ++++----
 drivers/ata/pata_cmd640.c     |   2 +-
 drivers/ata/pata_icside.c     |   4 +-
 drivers/ata/pata_it821x.c     |   2 +-
 drivers/ata/pata_ixp4xx_cf.c  |   2 +-
 drivers/ata/pata_legacy.c     |  14 +++---
 drivers/ata/pata_mpc52xx.c    |   2 +-
 drivers/ata/pata_ninja32.c    |   2 +-
 drivers/ata/pata_ns87415.c    |   8 ++--
 drivers/ata/pata_pcmcia.c     |   4 +-
 drivers/ata/pata_platform.c   |   2 +-
 drivers/ata/pata_qdi.c        |   2 +-
 drivers/ata/pata_rb500_cf.c   |   4 +-
 drivers/ata/pata_scc.c        |  42 ++++++++--------
 drivers/ata/pata_via.c        |   2 +-
 drivers/ata/pata_winbond.c    |   2 +-
 drivers/ata/sata_fsl.c        |   6 +--
 drivers/ata/sata_inic162x.c   |  12 ++---
 drivers/ata/sata_nv.c         |  20 ++++----
 drivers/ata/sata_promise.c    |   6 +--
 drivers/ata/sata_sil.c        |   6 +--
 drivers/ata/sata_sil24.c      |   6 +--
 drivers/ata/sata_svw.c        |  10 ++--
 drivers/ata/sata_sx4.c        |  16 +++----
 drivers/ata/sata_via.c        |   2 +-
 drivers/ata/sata_vsc.c        |   6 +--
 drivers/scsi/ipr.c            |   8 ++--
 drivers/scsi/libsas/sas_ata.c |   8 ++--
 include/linux/libata.h        |  25 +++++-----
 33 files changed, 192 insertions(+), 188 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 771509c9a3fb..c5e4501daa74 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -292,10 +292,10 @@ static struct scsi_host_template ahci_sht = {
 static struct ata_port_operations ahci_ops = {
 	.inherits		= &sata_pmp_port_ops,
 
-	.check_status		= ahci_check_status,
-	.check_altstatus	= ahci_check_status,
+	.sff_check_status	= ahci_check_status,
+	.sff_check_altstatus	= ahci_check_status,
 
-	.tf_read		= ahci_tf_read,
+	.sff_tf_read		= ahci_tf_read,
 	.qc_defer		= sata_pmp_qc_defer_cmd_switch,
 	.qc_prep		= ahci_qc_prep,
 	.qc_issue		= ahci_qc_issue,
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index cbdbfb5eaeaa..a01e02c5ce7a 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -74,7 +74,7 @@ const unsigned long sata_deb_timing_hotplug[]		= {  25,  500, 2000 };
 const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
 const struct ata_port_operations ata_base_port_ops = {
-	.irq_clear		= ata_noop_irq_clear,
+	.sff_irq_clear		= ata_noop_irq_clear,
 	.prereset		= ata_sff_prereset,
 	.hardreset		= sata_sff_hardreset,
 	.postreset		= ata_sff_postreset,
@@ -85,7 +85,7 @@ const struct ata_port_operations sata_port_ops = {
 	.inherits		= &ata_base_port_ops,
 
 	.qc_defer		= ata_std_qc_defer,
-	.dev_select		= ata_noop_dev_select,
+	.sff_dev_select		= ata_noop_dev_select,
 };
 
 const struct ata_port_operations sata_pmp_port_ops = {
@@ -3563,9 +3563,9 @@ void ata_sff_postreset(struct ata_link *link, unsigned int *classes)
 
 	/* is double-select really necessary? */
 	if (classes[0] != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 1);
+		ap->ops->sff_dev_select(ap, 1);
 	if (classes[1] != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 0);
+		ap->ops->sff_dev_select(ap, 0);
 
 	/* bail out if no device is present */
 	if (classes[0] == ATA_DEV_NONE && classes[1] == ATA_DEV_NONE) {
@@ -4416,7 +4416,7 @@ static void fill_result_tf(struct ata_queued_cmd *qc)
 	struct ata_port *ap = qc->ap;
 
 	qc->result_tf.flags = qc->tf.flags;
-	ap->ops->tf_read(ap, &qc->result_tf);
+	ap->ops->sff_tf_read(ap, &qc->result_tf);
 }
 
 static void ata_verify_xfer(struct ata_queued_cmd *qc)
@@ -6049,16 +6049,16 @@ static unsigned int ata_dummy_qc_issue(struct ata_queued_cmd *qc)
 }
 
 struct ata_port_operations ata_dummy_port_ops = {
-	.check_status		= ata_dummy_check_status,
-	.check_altstatus	= ata_dummy_check_status,
-	.dev_select		= ata_noop_dev_select,
+	.sff_check_status	= ata_dummy_check_status,
+	.sff_check_altstatus	= ata_dummy_check_status,
+	.sff_dev_select		= ata_noop_dev_select,
 	.qc_prep		= ata_noop_qc_prep,
 	.qc_issue		= ata_dummy_qc_issue,
 	.freeze			= ata_dummy_noret,
 	.thaw			= ata_dummy_noret,
 	.error_handler		= ata_dummy_noret,
 	.post_internal_cmd	= ata_dummy_qc_noret,
-	.irq_clear		= ata_dummy_noret,
+	.sff_irq_clear		= ata_dummy_noret,
 	.port_start		= ata_dummy_ret0,
 	.port_stop		= ata_dummy_noret,
 };
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 798ba5e45710..f8be92836a6e 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -2393,7 +2393,7 @@ static void atapi_request_sense(struct ata_queued_cmd *qc)
 	/* FIXME: is this needed? */
 	memset(cmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE);
 
-	ap->ops->tf_read(ap, &qc->tf);
+	ap->ops->sff_tf_read(ap, &qc->tf);
 
 	/* fill these in, for the case where they are -not- overwritten */
 	cmd->sense_buffer[0] = 0x70;
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 8544321293d4..04024a556660 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -51,13 +51,13 @@ const struct ata_port_operations ata_sff_port_ops = {
 	.error_handler		= ata_sff_error_handler,
 	.post_internal_cmd	= ata_sff_post_internal_cmd,
 
-	.dev_select		= ata_sff_dev_select,
-	.check_status		= ata_sff_check_status,
-	.tf_load		= ata_sff_tf_load,
-	.tf_read		= ata_sff_tf_read,
-	.exec_command		= ata_sff_exec_command,
-	.data_xfer		= ata_sff_data_xfer,
-	.irq_on			= ata_sff_irq_on,
+	.sff_dev_select		= ata_sff_dev_select,
+	.sff_check_status	= ata_sff_check_status,
+	.sff_tf_load		= ata_sff_tf_load,
+	.sff_tf_read		= ata_sff_tf_read,
+	.sff_exec_command	= ata_sff_exec_command,
+	.sff_data_xfer		= ata_sff_data_xfer,
+	.sff_irq_on		= ata_sff_irq_on,
 
 	.port_start		= ata_sff_port_start,
 };
@@ -71,7 +71,7 @@ const struct ata_port_operations ata_bmdma_port_ops = {
 	.bmdma_start		= ata_bmdma_start,
 	.bmdma_stop		= ata_bmdma_stop,
 	.bmdma_status		= ata_bmdma_status,
-	.irq_clear		= ata_sff_irq_clear,
+	.sff_irq_clear		= ata_sff_irq_clear,
 };
 
 /**
@@ -245,8 +245,8 @@ u8 ata_sff_check_status(struct ata_port *ap)
  */
 u8 ata_sff_altstatus(struct ata_port *ap)
 {
-	if (ap->ops->check_altstatus)
-		return ap->ops->check_altstatus(ap);
+	if (ap->ops->sff_check_altstatus)
+		return ap->ops->sff_check_altstatus(ap);
 
 	return ioread8(ap->ioaddr.altstatus_addr);
 }
@@ -290,7 +290,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 	while (status != 0xff && (status & ATA_BUSY) &&
 	       time_before(jiffies, timeout)) {
 		msleep(50);
-		status = ap->ops->check_status(ap);
+		status = ap->ops->sff_check_status(ap);
 	}
 
 	if (status == 0xff)
@@ -326,7 +326,7 @@ int ata_sff_wait_ready(struct ata_port *ap, unsigned long deadline)
 	int warned = 0;
 
 	while (1) {
-		u8 status = ap->ops->check_status(ap);
+		u8 status = ap->ops->sff_check_status(ap);
 		unsigned long now = jiffies;
 
 		if (!(status & ATA_BUSY))
@@ -403,7 +403,7 @@ void ata_dev_select(struct ata_port *ap, unsigned int device,
 	if (wait)
 		ata_wait_idle(ap);
 
-	ap->ops->dev_select(ap, device);
+	ap->ops->sff_dev_select(ap, device);
 
 	if (wait) {
 		if (can_sleep && ap->link.device[device].class == ATA_DEV_ATAPI)
@@ -434,7 +434,7 @@ u8 ata_sff_irq_on(struct ata_port *ap)
 		iowrite8(ap->ctl, ioaddr->ctl_addr);
 	tmp = ata_wait_idle(ap);
 
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_irq_clear(ap);
 
 	return tmp;
 }
@@ -593,8 +593,8 @@ void ata_sff_exec_command(struct ata_port *ap, const struct ata_taskfile *tf)
 static inline void ata_tf_to_host(struct ata_port *ap,
 				  const struct ata_taskfile *tf)
 {
-	ap->ops->tf_load(ap, tf);
-	ap->ops->exec_command(ap, tf);
+	ap->ops->sff_tf_load(ap, tf);
+	ap->ops->sff_exec_command(ap, tf);
 }
 
 /**
@@ -709,13 +709,15 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
 		buf = kmap_atomic(page, KM_IRQ0);
 
 		/* do the actual data transfer */
-		ap->ops->data_xfer(qc->dev, buf + offset, qc->sect_size, do_write);
+		ap->ops->sff_data_xfer(qc->dev, buf + offset, qc->sect_size,
+				       do_write);
 
 		kunmap_atomic(buf, KM_IRQ0);
 		local_irq_restore(flags);
 	} else {
 		buf = page_address(page);
-		ap->ops->data_xfer(qc->dev, buf + offset, qc->sect_size, do_write);
+		ap->ops->sff_data_xfer(qc->dev, buf + offset, qc->sect_size,
+				       do_write);
 	}
 
 	qc->curbytes += qc->sect_size;
@@ -772,7 +774,7 @@ static void atapi_send_cdb(struct ata_port *ap, struct ata_queued_cmd *qc)
 	DPRINTK("send cdb\n");
 	WARN_ON(qc->dev->cdb_len < 12);
 
-	ap->ops->data_xfer(qc->dev, qc->cdb, qc->dev->cdb_len, 1);
+	ap->ops->sff_data_xfer(qc->dev, qc->cdb, qc->dev->cdb_len, 1);
 	ata_sff_altstatus(ap); /* flush */
 
 	switch (qc->tf.protocol) {
@@ -844,13 +846,13 @@ next_sg:
 		buf = kmap_atomic(page, KM_IRQ0);
 
 		/* do the actual data transfer */
-		consumed = ap->ops->data_xfer(dev,  buf + offset, count, rw);
+		consumed = ap->ops->sff_data_xfer(dev,  buf + offset, count, rw);
 
 		kunmap_atomic(buf, KM_IRQ0);
 		local_irq_restore(flags);
 	} else {
 		buf = page_address(page);
-		consumed = ap->ops->data_xfer(dev,  buf + offset, count, rw);
+		consumed = ap->ops->sff_data_xfer(dev,  buf + offset, count, rw);
 	}
 
 	bytes -= min(bytes, consumed);
@@ -893,7 +895,7 @@ static void atapi_pio_bytes(struct ata_queued_cmd *qc)
 	 * error, qc->result_tf is later overwritten by ata_qc_complete().
 	 * So, the correctness of qc->result_tf is not affected.
 	 */
-	ap->ops->tf_read(ap, &qc->result_tf);
+	ap->ops->sff_tf_read(ap, &qc->result_tf);
 	ireason = qc->result_tf.nsect;
 	bc_lo = qc->result_tf.lbam;
 	bc_hi = qc->result_tf.lbah;
@@ -979,7 +981,7 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
 			qc = ata_qc_from_tag(ap, qc->tag);
 			if (qc) {
 				if (likely(!(qc->err_mask & AC_ERR_HSM))) {
-					ap->ops->irq_on(ap);
+					ap->ops->sff_irq_on(ap);
 					ata_qc_complete(qc);
 				} else
 					ata_port_freeze(ap);
@@ -995,7 +997,7 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
 	} else {
 		if (in_wq) {
 			spin_lock_irqsave(ap->lock, flags);
-			ap->ops->irq_on(ap);
+			ap->ops->sff_irq_on(ap);
 			ata_qc_complete(qc);
 			spin_unlock_irqrestore(ap->lock, flags);
 		} else
@@ -1345,7 +1347,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc)
 	case ATA_PROT_DMA:
 		WARN_ON(qc->tf.flags & ATA_TFLAG_POLLING);
 
-		ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
+		ap->ops->sff_tf_load(ap, &qc->tf);  /* load tf registers */
 		ap->ops->bmdma_setup(qc);	    /* set up bmdma */
 		ap->ops->bmdma_start(qc);	    /* initiate bmdma */
 		ap->hsm_task_state = HSM_ST_LAST;
@@ -1397,7 +1399,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc)
 	case ATAPI_PROT_DMA:
 		WARN_ON(qc->tf.flags & ATA_TFLAG_POLLING);
 
-		ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
+		ap->ops->sff_tf_load(ap, &qc->tf);  /* load tf registers */
 		ap->ops->bmdma_setup(qc);	    /* set up bmdma */
 		ap->hsm_task_state = HSM_ST_FIRST;
 
@@ -1486,12 +1488,12 @@ inline unsigned int ata_sff_host_intr(struct ata_port *ap,
 		goto idle_irq;
 
 	/* check main status, clearing INTRQ */
-	status = ap->ops->check_status(ap);
+	status = ap->ops->sff_check_status(ap);
 	if (unlikely(status & ATA_BUSY))
 		goto idle_irq;
 
 	/* ack bmdma irq events */
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_irq_clear(ap);
 
 	ata_sff_hsm_move(ap, qc, status, 0);
 
@@ -1506,8 +1508,8 @@ idle_irq:
 
 #ifdef ATA_IRQ_TRAP
 	if ((ap->stats.idle_irq % 1000) == 0) {
-		ap->ops->check_status(ap);
-		ap->ops->irq_clear(ap);
+		ap->ops->sff_check_status(ap);
+		ap->ops->sff_irq_clear(ap);
 		ata_port_printk(ap, KERN_WARNING, "irq trap\n");
 		return 1;
 	}
@@ -1582,9 +1584,9 @@ void ata_sff_freeze(struct ata_port *ap)
 	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
 	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
 	 */
-	ap->ops->check_status(ap);
+	ap->ops->sff_check_status(ap);
 
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_irq_clear(ap);
 }
 
 /**
@@ -1599,9 +1601,9 @@ void ata_sff_freeze(struct ata_port *ap)
 void ata_sff_thaw(struct ata_port *ap)
 {
 	/* clear & re-enable interrupts */
-	ap->ops->check_status(ap);
-	ap->ops->irq_clear(ap);
-	ap->ops->irq_on(ap);
+	ap->ops->sff_check_status(ap);
+	ap->ops->sff_irq_clear(ap);
+	ap->ops->sff_irq_on(ap);
 }
 
 /**
@@ -1626,7 +1628,7 @@ static unsigned int ata_devchk(struct ata_port *ap, unsigned int device)
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 	u8 nsect, lbal;
 
-	ap->ops->dev_select(ap, device);
+	ap->ops->sff_dev_select(ap, device);
 
 	iowrite8(0x55, ioaddr->nsect_addr);
 	iowrite8(0xaa, ioaddr->lbal_addr);
@@ -1675,11 +1677,11 @@ unsigned int ata_sff_dev_classify(struct ata_device *dev, int present,
 	unsigned int class;
 	u8 err;
 
-	ap->ops->dev_select(ap, dev->devno);
+	ap->ops->sff_dev_select(ap, dev->devno);
 
 	memset(&tf, 0, sizeof(tf));
 
-	ap->ops->tf_read(ap, &tf);
+	ap->ops->sff_tf_read(ap, &tf);
 	err = tf.feature;
 	if (r_err)
 		*r_err = err;
@@ -1709,7 +1711,8 @@ unsigned int ata_sff_dev_classify(struct ata_device *dev, int present,
 			class = ATA_DEV_ATA;
 		else
 			class = ATA_DEV_NONE;
-	} else if ((class == ATA_DEV_ATA) && (ap->ops->check_status(ap) == 0))
+	} else if ((class == ATA_DEV_ATA) &&
+		   (ap->ops->sff_check_status(ap) == 0))
 		class = ATA_DEV_NONE;
 
 	return class;
@@ -1741,7 +1744,7 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	if (dev1) {
 		int i;
 
-		ap->ops->dev_select(ap, 1);
+		ap->ops->sff_dev_select(ap, 1);
 
 		/* Wait for register access.  Some ATAPI devices fail
 		 * to set nsect/lbal after reset, so don't waste too
@@ -1766,11 +1769,11 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	}
 
 	/* is all this really necessary? */
-	ap->ops->dev_select(ap, 0);
+	ap->ops->sff_dev_select(ap, 0);
 	if (dev1)
-		ap->ops->dev_select(ap, 1);
+		ap->ops->sff_dev_select(ap, 1);
 	if (dev0)
-		ap->ops->dev_select(ap, 0);
+		ap->ops->sff_dev_select(ap, 0);
 
 	return ret;
 }
@@ -1820,7 +1823,7 @@ void ata_sff_wait_after_reset(struct ata_port *ap, unsigned long deadline)
 	 */
 	if (ap->flags & ATA_FLAG_SATA) {
 		while (1) {
-			u8 status = ap->ops->check_status(ap);
+			u8 status = ap->ops->sff_check_status(ap);
 
 			if (status != 0xff || time_after(jiffies, deadline))
 				return;
@@ -1851,7 +1854,7 @@ static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
 	 * the bus shows 0xFF because the odd clown forgets the D7
 	 * pulldown resistor.
 	 */
-	if (ap->ops->check_status(ap) == 0xFF)
+	if (ap->ops->sff_check_status(ap) == 0xFF)
 		return -ENODEV;
 
 	return ata_bus_post_reset(ap, devmask, deadline);
@@ -1894,7 +1897,7 @@ int ata_sff_softreset(struct ata_link *link, unsigned int *classes,
 		devmask |= (1 << 1);
 
 	/* select device 0 again */
-	ap->ops->dev_select(ap, 0);
+	ap->ops->sff_dev_select(ap, 0);
 
 	/* issue bus reset */
 	DPRINTK("about to softreset, devmask=%x\n", devmask);
@@ -1977,7 +1980,7 @@ int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
 		return rc;
 	}
 
-	ap->ops->dev_select(ap, 0);	/* probably unnecessary */
+	ap->ops->sff_dev_select(ap, 0);	/* probably unnecessary */
 
 	*class = ata_sff_dev_classify(link->device, 1, NULL);
 
@@ -2035,8 +2038,8 @@ void ata_sff_error_handler(struct ata_port *ap)
 	}
 
 	ata_sff_altstatus(ap);
-	ap->ops->check_status(ap);
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_check_status(ap);
+	ap->ops->sff_irq_clear(ap);
 
 	spin_unlock_irqrestore(ap->lock, flags);
 
@@ -2153,7 +2156,7 @@ void ata_bmdma_setup(struct ata_queued_cmd *qc)
 	iowrite8(dmactl, ap->ioaddr.bmdma_addr + ATA_DMA_CMD);
 
 	/* issue r/w command */
-	ap->ops->exec_command(ap, &qc->tf);
+	ap->ops->sff_exec_command(ap, &qc->tf);
 }
 
 /**
@@ -2277,7 +2280,7 @@ void ata_bus_reset(struct ata_port *ap)
 		devmask |= (1 << 1);
 
 	/* select device 0 again */
-	ap->ops->dev_select(ap, 0);
+	ap->ops->sff_dev_select(ap, 0);
 
 	/* issue bus reset */
 	if (ap->flags & ATA_FLAG_SRST) {
@@ -2295,9 +2298,9 @@ void ata_bus_reset(struct ata_port *ap)
 
 	/* is double-select really necessary? */
 	if (device[1].class != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 1);
+		ap->ops->sff_dev_select(ap, 1);
 	if (device[0].class != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 0);
+		ap->ops->sff_dev_select(ap, 0);
 
 	/* if no devices were detected, disable this port */
 	if ((device[0].class == ATA_DEV_NONE) &&
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index c854e882d4a9..d98bd7455e6b 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1264,7 +1264,7 @@ static void bfin_freeze(struct ata_port *ap)
 	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
 	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
 	 */
-	ap->ops->check_status(ap);
+	ap->ops->sff_check_status(ap);
 
 	bfin_irq_clear(ap);
 }
@@ -1357,18 +1357,18 @@ static const struct ata_port_operations bfin_pata_ops = {
 	.set_piomode		= bfin_set_piomode,
 	.set_dmamode		= bfin_set_dmamode,
 
-	.tf_load		= bfin_tf_load,
-	.tf_read		= bfin_tf_read,
-	.exec_command		= bfin_exec_command,
-	.check_status		= bfin_check_status,
-	.check_altstatus	= bfin_check_altstatus,
-	.dev_select		= bfin_dev_select,
+	.sff_tf_load		= bfin_tf_load,
+	.sff_tf_read		= bfin_tf_read,
+	.sff_exec_command	= bfin_exec_command,
+	.sff_check_status	= bfin_check_status,
+	.sff_check_altstatus	= bfin_check_altstatus,
+	.sff_dev_select		= bfin_dev_select,
 
 	.bmdma_setup		= bfin_bmdma_setup,
 	.bmdma_start		= bfin_bmdma_start,
 	.bmdma_stop		= bfin_bmdma_stop,
 	.bmdma_status		= bfin_bmdma_status,
-	.data_xfer		= bfin_data_xfer,
+	.sff_data_xfer		= bfin_data_xfer,
 
 	.qc_prep		= ata_noop_qc_prep,
 
@@ -1378,8 +1378,8 @@ static const struct ata_port_operations bfin_pata_ops = {
 	.postreset		= bfin_postreset,
 	.post_internal_cmd	= bfin_bmdma_stop,
 
-	.irq_clear		= bfin_irq_clear,
-	.irq_on			= bfin_irq_on,
+	.sff_irq_clear		= bfin_irq_clear,
+	.sff_irq_on		= bfin_irq_on,
 
 	.port_start		= bfin_port_start,
 	.port_stop		= bfin_port_stop,
diff --git a/drivers/ata/pata_cmd640.c b/drivers/ata/pata_cmd640.c
index a907cf478891..2de30b990278 100644
--- a/drivers/ata/pata_cmd640.c
+++ b/drivers/ata/pata_cmd640.c
@@ -172,7 +172,7 @@ static struct scsi_host_template cmd640_sht = {
 static struct ata_port_operations cmd640_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 	/* In theory xfer_noirq is not needed once we kill the prefetcher */
-	.data_xfer	= ata_sff_data_xfer_noirq,
+	.sff_data_xfer	= ata_sff_data_xfer_noirq,
 	.qc_issue	= cmd640_qc_issue,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= cmd640_set_piomode,
diff --git a/drivers/ata/pata_icside.c b/drivers/ata/pata_icside.c
index 52de9f908fb0..17138436423d 100644
--- a/drivers/ata/pata_icside.c
+++ b/drivers/ata/pata_icside.c
@@ -250,7 +250,7 @@ static void pata_icside_bmdma_setup(struct ata_queued_cmd *qc)
 	set_dma_mode(state->dma, write ? DMA_MODE_WRITE : DMA_MODE_READ);
 
 	/* issue r/w command */
-	ap->ops->exec_command(ap, &qc->tf);
+	ap->ops->sff_exec_command(ap, &qc->tf);
 }
 
 static void pata_icside_bmdma_start(struct ata_queued_cmd *qc)
@@ -336,7 +336,7 @@ static struct ata_port_operations pata_icside_port_ops = {
 	.inherits		= &ata_sff_port_ops,
 	/* no need to build any PRD tables for DMA */
 	.qc_prep		= ata_noop_qc_prep,
-	.data_xfer		= ata_sff_data_xfer_noirq,
+	.sff_data_xfer		= ata_sff_data_xfer_noirq,
 	.bmdma_setup		= pata_icside_bmdma_setup,
 	.bmdma_start		= pata_icside_bmdma_start,
 	.bmdma_stop		= pata_icside_bmdma_stop,
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index 88e37cfcfc4f..e10816931b2f 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -652,7 +652,7 @@ static struct ata_port_operations it821x_passthru_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 
 	.check_atapi_dma= it821x_check_atapi_dma,
-	.dev_select 	= it821x_passthru_dev_select,
+	.sff_dev_select	= it821x_passthru_dev_select,
 	.bmdma_start 	= it821x_passthru_bmdma_start,
 	.bmdma_stop	= it821x_passthru_bmdma_stop,
 	.qc_issue	= it821x_passthru_qc_issue,
diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c
index 283a8fba79b3..8a175f23b907 100644
--- a/drivers/ata/pata_ixp4xx_cf.c
+++ b/drivers/ata/pata_ixp4xx_cf.c
@@ -93,7 +93,7 @@ static struct scsi_host_template ixp4xx_sht = {
 
 static struct ata_port_operations ixp4xx_port_ops = {
 	.inherits		= &ata_sff_port_ops,
-	.data_xfer		= ixp4xx_mmio_data_xfer,
+	.sff_data_xfer		= ixp4xx_mmio_data_xfer,
 	.cable_detect		= ata_cable_40wire,
 	.set_mode		= ixp4xx_set_mode,
 };
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
index f13f10a55ef3..7af4b29cc422 100644
--- a/drivers/ata/pata_legacy.c
+++ b/drivers/ata/pata_legacy.c
@@ -226,12 +226,12 @@ static const struct ata_port_operations legacy_base_port_ops = {
 
 static struct ata_port_operations simple_port_ops = {
 	.inherits	= &legacy_base_port_ops,
-	.data_xfer	= ata_sff_data_xfer_noirq,
+	.sff_data_xfer	= ata_sff_data_xfer_noirq,
 };
 
 static struct ata_port_operations legacy_port_ops = {
 	.inherits	= &legacy_base_port_ops,
-	.data_xfer	= ata_sff_data_xfer_noirq,
+	.sff_data_xfer	= ata_sff_data_xfer_noirq,
 	.set_mode	= legacy_set_mode,
 };
 
@@ -325,7 +325,7 @@ static unsigned int pdc_data_xfer_vlb(struct ata_device *dev,
 static struct ata_port_operations pdc20230_port_ops = {
 	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= pdc20230_set_piomode,
-	.data_xfer	= pdc_data_xfer_vlb,
+	.sff_data_xfer	= pdc_data_xfer_vlb,
 };
 
 /*
@@ -775,19 +775,19 @@ static struct ata_port_operations qdi6500_port_ops = {
 	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= qdi6500_set_piomode,
 	.qc_issue	= qdi_qc_issue,
-	.data_xfer	= vlb32_data_xfer,
+	.sff_data_xfer	= vlb32_data_xfer,
 };
 
 static struct ata_port_operations qdi6580_port_ops = {
 	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= qdi6580_set_piomode,
-	.data_xfer	= vlb32_data_xfer,
+	.sff_data_xfer	= vlb32_data_xfer,
 };
 
 static struct ata_port_operations qdi6580dp_port_ops = {
 	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= qdi6580dp_set_piomode,
-	.data_xfer	= vlb32_data_xfer,
+	.sff_data_xfer	= vlb32_data_xfer,
 };
 
 static DEFINE_SPINLOCK(winbond_lock);
@@ -858,7 +858,7 @@ static int winbond_port(struct platform_device *dev,
 static struct ata_port_operations winbond_port_ops = {
 	.inherits	= &legacy_base_port_ops,
 	.set_piomode	= winbond_set_piomode,
-	.data_xfer	= vlb32_data_xfer,
+	.sff_data_xfer	= vlb32_data_xfer,
 };
 
 static struct legacy_controller controllers[] = {
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index 5d1d32a39c4d..bc79df6e7cb0 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -261,7 +261,7 @@ static struct scsi_host_template mpc52xx_ata_sht = {
 
 static struct ata_port_operations mpc52xx_ata_port_ops = {
 	.inherits		= &ata_sff_port_ops,
-	.dev_select		= mpc52xx_ata_dev_select,
+	.sff_dev_select		= mpc52xx_ata_dev_select,
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= mpc52xx_ata_set_piomode,
 	.post_internal_cmd	= ATA_OP_NULL,
diff --git a/drivers/ata/pata_ninja32.c b/drivers/ata/pata_ninja32.c
index c9c0ea67c71a..565e67cd13fa 100644
--- a/drivers/ata/pata_ninja32.c
+++ b/drivers/ata/pata_ninja32.c
@@ -84,7 +84,7 @@ static struct scsi_host_template ninja32_sht = {
 
 static struct ata_port_operations ninja32_port_ops = {
 	.inherits	= &ata_bmdma_port_ops,
-	.dev_select 	= ninja32_dev_select,
+	.sff_dev_select = ninja32_dev_select,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= ninja32_set_piomode,
 };
diff --git a/drivers/ata/pata_ns87415.c b/drivers/ata/pata_ns87415.c
index 03a52cefc010..ae92b0049bd5 100644
--- a/drivers/ata/pata_ns87415.c
+++ b/drivers/ata/pata_ns87415.c
@@ -138,7 +138,7 @@ static void ns87415_bmdma_setup(struct ata_queued_cmd *qc)
 		dmactl |= ATA_DMA_WR;
 	iowrite8(dmactl, ap->ioaddr.bmdma_addr + ATA_DMA_CMD);
 	/* issue r/w command */
-	ap->ops->exec_command(ap, &qc->tf);
+	ap->ops->sff_exec_command(ap, &qc->tf);
 }
 
 /**
@@ -306,7 +306,7 @@ static struct ata_port_operations ns87415_pata_ops = {
 	.bmdma_setup		= ns87415_bmdma_setup,
 	.bmdma_start		= ns87415_bmdma_start,
 	.bmdma_stop		= ns87415_bmdma_stop,
-	.irq_clear		= ns87415_irq_clear,
+	.sff_irq_clear		= ns87415_irq_clear,
 
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= ns87415_set_piomode,
@@ -315,8 +315,8 @@ static struct ata_port_operations ns87415_pata_ops = {
 #if defined(CONFIG_SUPERIO)
 static struct ata_port_operations ns87560_pata_ops = {
 	.inherits		= &ns87415_pata_ops,
-	.tf_read		= ns87560_tf_read,
-	.check_status		= ns87560_check_status,
+	.sff_tf_read		= ns87560_tf_read,
+	.sff_check_status	= ns87560_check_status,
 	.bmdma_status		= ns87560_bmdma_status,
 };
 #endif
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index 2e206c5f869a..3d39f9dfec5a 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -133,14 +133,14 @@ static struct scsi_host_template pcmcia_sht = {
 
 static struct ata_port_operations pcmcia_port_ops = {
 	.inherits	= &ata_sff_port_ops,
-	.data_xfer	= ata_sff_data_xfer_noirq,
+	.sff_data_xfer	= ata_sff_data_xfer_noirq,
 	.cable_detect	= ata_cable_40wire,
 	.set_mode	= pcmcia_set_mode,
 };
 
 static struct ata_port_operations pcmcia_8bit_port_ops = {
 	.inherits	= &ata_sff_port_ops,
-	.data_xfer	= ata_data_xfer_8bit,
+	.sff_data_xfer	= ata_data_xfer_8bit,
 	.cable_detect	= ata_cable_40wire,
 	.set_mode	= pcmcia_set_mode_8bit,
 };
diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c
index 1edfc13d05d2..6527c56c34a3 100644
--- a/drivers/ata/pata_platform.c
+++ b/drivers/ata/pata_platform.c
@@ -52,7 +52,7 @@ static struct scsi_host_template pata_platform_sht = {
 
 static struct ata_port_operations pata_platform_port_ops = {
 	.inherits		= &ata_sff_port_ops,
-	.data_xfer		= ata_sff_data_xfer_noirq,
+	.sff_data_xfer		= ata_sff_data_xfer_noirq,
 	.cable_detect		= ata_cable_unknown,
 	.set_mode		= pata_platform_set_mode,
 	.port_start		= ATA_OP_NULL,
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
index 0d81dd5fdc38..bf45cf017753 100644
--- a/drivers/ata/pata_qdi.c
+++ b/drivers/ata/pata_qdi.c
@@ -160,7 +160,7 @@ static struct scsi_host_template qdi_sht = {
 static struct ata_port_operations qdi6500_port_ops = {
 	.inherits	= &ata_sff_port_ops,
 	.qc_issue	= qdi_qc_issue,
-	.data_xfer	= qdi_data_xfer,
+	.sff_data_xfer	= qdi_data_xfer,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= qdi6500_set_piomode,
 };
diff --git a/drivers/ata/pata_rb500_cf.c b/drivers/ata/pata_rb500_cf.c
index 5b23d79bc92b..800ae4601f44 100644
--- a/drivers/ata/pata_rb500_cf.c
+++ b/drivers/ata/pata_rb500_cf.c
@@ -119,8 +119,8 @@ static irqreturn_t rb500_pata_irq_handler(int irq, void *dev_instance)
 
 static struct ata_port_operations rb500_pata_port_ops = {
 	.inherits		= &ata_sff_port_ops,
-	.exec_command		= rb500_pata_exec_command,
-	.data_xfer		= rb500_pata_data_xfer,
+	.sff_exec_command	= rb500_pata_exec_command,
+	.sff_data_xfer		= rb500_pata_data_xfer,
 	.freeze			= rb500_pata_freeze,
 	.thaw			= rb500_pata_thaw,
 };
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index 701d0addae73..2b9da715c704 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -441,7 +441,7 @@ static void scc_bmdma_setup (struct ata_queued_cmd *qc)
 	out_be32(mmio + SCC_DMA_CMD, dmactl);
 
 	/* issue r/w command */
-	ap->ops->exec_command(ap, &qc->tf);
+	ap->ops->sff_exec_command(ap, &qc->tf);
 }
 
 /**
@@ -476,7 +476,7 @@ static unsigned int scc_devchk (struct ata_port *ap,
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 	u8 nsect, lbal;
 
-	ap->ops->dev_select(ap, device);
+	ap->ops->sff_dev_select(ap, device);
 
 	out_be32(ioaddr->nsect_addr, 0x55);
 	out_be32(ioaddr->lbal_addr, 0xaa);
@@ -525,7 +525,7 @@ static int scc_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	while (dev1) {
 		u8 nsect, lbal;
 
-		ap->ops->dev_select(ap, 1);
+		ap->ops->sff_dev_select(ap, 1);
 		nsect = in_be32(ioaddr->nsect_addr);
 		lbal = in_be32(ioaddr->lbal_addr);
 		if ((nsect == 1) && (lbal == 1))
@@ -541,11 +541,11 @@ static int scc_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	}
 
 	/* is all this really necessary? */
-	ap->ops->dev_select(ap, 0);
+	ap->ops->sff_dev_select(ap, 0);
 	if (dev1)
-		ap->ops->dev_select(ap, 1);
+		ap->ops->sff_dev_select(ap, 1);
 	if (dev0)
-		ap->ops->dev_select(ap, 0);
+		ap->ops->sff_dev_select(ap, 0);
 
 	return 0;
 }
@@ -616,7 +616,7 @@ static int scc_softreset(struct ata_link *link, unsigned int *classes,
 		devmask |= (1 << 1);
 
 	/* select device 0 again */
-	ap->ops->dev_select(ap, 0);
+	ap->ops->sff_dev_select(ap, 0);
 
 	/* issue bus reset */
 	DPRINTK("about to softreset, devmask=%x\n", devmask);
@@ -829,7 +829,7 @@ static u8 scc_irq_on (struct ata_port *ap)
 	out_be32(ioaddr->ctl_addr, ap->ctl);
 	tmp = ata_wait_idle(ap);
 
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_irq_clear(ap);
 
 	return tmp;
 }
@@ -854,9 +854,9 @@ static void scc_freeze (struct ata_port *ap)
 	 * ATA_NIEN manipulation.  Also, many controllers fail to mask
 	 * previously pending IRQ on ATA_NIEN assertion.  Clear it.
 	 */
-	ap->ops->check_status(ap);
+	ap->ops->sff_check_status(ap);
 
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_irq_clear(ap);
 }
 
 /**
@@ -887,9 +887,9 @@ static void scc_postreset(struct ata_link *link, unsigned int *classes)
 
 	/* is double-select really necessary? */
 	if (classes[0] != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 1);
+		ap->ops->sff_dev_select(ap, 1);
 	if (classes[1] != ATA_DEV_NONE)
-		ap->ops->dev_select(ap, 0);
+		ap->ops->sff_dev_select(ap, 0);
 
 	/* bail out if no device is present */
 	if (classes[0] == ATA_DEV_NONE && classes[1] == ATA_DEV_NONE) {
@@ -967,18 +967,18 @@ static struct ata_port_operations scc_pata_ops = {
 	.set_dmamode		= scc_set_dmamode,
 	.mode_filter		= scc_mode_filter,
 
-	.tf_load		= scc_tf_load,
-	.tf_read		= scc_tf_read,
-	.exec_command		= scc_exec_command,
-	.check_status		= scc_check_status,
-	.check_altstatus	= scc_check_altstatus,
-	.dev_select		= scc_dev_select,
+	.sff_tf_load		= scc_tf_load,
+	.sff_tf_read		= scc_tf_read,
+	.sff_exec_command	= scc_exec_command,
+	.sff_check_status	= scc_check_status,
+	.sff_check_altstatus	= scc_check_altstatus,
+	.sff_dev_select		= scc_dev_select,
 
 	.bmdma_setup		= scc_bmdma_setup,
 	.bmdma_start		= scc_bmdma_start,
 	.bmdma_stop		= scc_bmdma_stop,
 	.bmdma_status		= scc_bmdma_status,
-	.data_xfer		= scc_data_xfer,
+	.sff_data_xfer		= scc_data_xfer,
 
 	.freeze			= scc_freeze,
 	.prereset		= scc_pata_prereset,
@@ -986,8 +986,8 @@ static struct ata_port_operations scc_pata_ops = {
 	.postreset		= scc_postreset,
 	.post_internal_cmd	= scc_bmdma_stop,
 
-	.irq_clear		= scc_irq_clear,
-	.irq_on			= scc_irq_on,
+	.sff_irq_clear		= scc_irq_clear,
+	.sff_irq_on		= scc_irq_on,
 
 	.port_start		= scc_port_start,
 	.port_stop		= scc_port_stop,
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index 4b85f84fbe76..d4840748fb5c 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -336,7 +336,7 @@ static struct ata_port_operations via_port_ops = {
 
 static struct ata_port_operations via_port_ops_noirq = {
 	.inherits	= &via_port_ops,
-	.data_xfer	= ata_sff_data_xfer_noirq,
+	.sff_data_xfer	= ata_sff_data_xfer_noirq,
 };
 
 /**
diff --git a/drivers/ata/pata_winbond.c b/drivers/ata/pata_winbond.c
index cc18231e9334..6e52a3573fbf 100644
--- a/drivers/ata/pata_winbond.c
+++ b/drivers/ata/pata_winbond.c
@@ -127,7 +127,7 @@ static struct scsi_host_template winbond_sht = {
 
 static struct ata_port_operations winbond_port_ops = {
 	.inherits	= &ata_sff_port_ops,
-	.data_xfer	= winbond_data_xfer,
+	.sff_data_xfer	= winbond_data_xfer,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= winbond_set_piomode,
 };
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 676302fdaa97..37482d4e883f 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -1190,10 +1190,10 @@ static struct scsi_host_template sata_fsl_sht = {
 static const struct ata_port_operations sata_fsl_ops = {
 	.inherits = &sata_port_ops,
 
-	.check_status = sata_fsl_check_status,
-	.check_altstatus = sata_fsl_check_status,
+	.sff_check_status = sata_fsl_check_status,
+	.sff_check_altstatus = sata_fsl_check_status,
 
-	.tf_read = sata_fsl_tf_read,
+	.sff_tf_read = sata_fsl_tf_read,
 
 	.qc_prep = sata_fsl_qc_prep,
 	.qc_issue = sata_fsl_qc_issue,
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 6e6fca4c20b6..0b5a736a45e3 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -222,7 +222,7 @@ static void inic_bmdma_setup(struct ata_queued_cmd *qc)
 	writeb(pp->cached_prdctl, port_base + PORT_PRD_CTL);
 
 	/* issue r/w command */
-	ap->ops->exec_command(ap, &qc->tf);
+	ap->ops->sff_exec_command(ap, &qc->tf);
 }
 
 static void inic_bmdma_start(struct ata_queued_cmd *qc)
@@ -267,14 +267,14 @@ static void inic_host_intr(struct ata_port *ap)
 			ata_qc_from_tag(ap, ap->link.active_tag);
 
 		if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) {
-			ap->ops->check_status(ap); /* clear ATA interrupt */
+			ap->ops->sff_check_status(ap); /* clear ATA interrupt */
 			return;
 		}
 
 		if (likely(ata_sff_host_intr(ap, qc)))
 			return;
 
-		ap->ops->check_status(ap); /* clear ATA interrupt */
+		ap->ops->sff_check_status(ap); /* clear ATA interrupt */
 		ata_port_printk(ap, KERN_WARNING, "unhandled "
 				"interrupt, irq_stat=%x\n", irq_stat);
 		return;
@@ -351,7 +351,7 @@ static unsigned int inic_qc_issue(struct ata_queued_cmd *qc)
 	 */
 	if (unlikely(qc->tf.command == ATA_CMD_ID_ATA ||
 		     qc->tf.command == ATA_CMD_ID_ATAPI)) {
-		u8 stat = ap->ops->check_status(ap);
+		u8 stat = ap->ops->sff_check_status(ap);
 		if (stat == 0x7f || stat == 0xff)
 			return AC_ERR_HSM;
 	}
@@ -365,7 +365,7 @@ static void inic_freeze(struct ata_port *ap)
 
 	__inic_set_pirq_mask(ap, PIRQ_MASK_FREEZE);
 
-	ap->ops->check_status(ap);
+	ap->ops->sff_check_status(ap);
 	writeb(0xff, port_base + PORT_IRQ_STAT);
 
 	readb(port_base + PORT_IRQ_STAT); /* flush */
@@ -375,7 +375,7 @@ static void inic_thaw(struct ata_port *ap)
 {
 	void __iomem *port_base = inic_port_base(ap);
 
-	ap->ops->check_status(ap);
+	ap->ops->sff_check_status(ap);
 	writeb(0xff, port_base + PORT_IRQ_STAT);
 
 	__inic_set_pirq_mask(ap, PIRQ_MASK_OTHER);
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 63cc43765f04..109b07495721 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -429,11 +429,11 @@ static struct ata_port_operations nv_adma_ops = {
 	.inherits		= &nv_generic_ops,
 
 	.check_atapi_dma	= nv_adma_check_atapi_dma,
-	.tf_read		= nv_adma_tf_read,
+	.sff_tf_read		= nv_adma_tf_read,
 	.qc_defer		= ata_std_qc_defer,
 	.qc_prep		= nv_adma_qc_prep,
 	.qc_issue		= nv_adma_qc_issue,
-	.irq_clear		= nv_adma_irq_clear,
+	.sff_irq_clear		= nv_adma_irq_clear,
 
 	.freeze			= nv_adma_freeze,
 	.thaw			= nv_adma_thaw,
@@ -1440,7 +1440,7 @@ static irqreturn_t nv_generic_interrupt(int irq, void *dev_instance)
 			else
 				// No request pending?  Clear interrupt status
 				// anyway, in case there's one pending.
-				ap->ops->check_status(ap);
+				ap->ops->sff_check_status(ap);
 		}
 
 	}
@@ -1739,7 +1739,7 @@ static void nv_swncq_ncq_stop(struct ata_port *ap)
 		pp->dhfis_bits, pp->dmafis_bits, pp->sdbfis_bits);
 
 	ata_port_printk(ap, KERN_ERR, "ATA_REG 0x%X ERR_REG 0x%X\n",
-			ap->ops->check_status(ap),
+			ap->ops->sff_check_status(ap),
 			ioread8(ap->ioaddr.error_addr));
 
 	sactive = readl(pp->sactive_block);
@@ -1765,7 +1765,7 @@ static void nv_swncq_ncq_stop(struct ata_port *ap)
 	}
 
 	nv_swncq_pp_reinit(ap);
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_irq_clear(ap);
 	__ata_bmdma_stop(ap);
 	nv_swncq_irq_clear(ap, 0xffff);
 }
@@ -1987,8 +1987,8 @@ static unsigned int nv_swncq_issue_atacmd(struct ata_port *ap,
 	pp->dmafis_bits &= ~(1 << qc->tag);
 	pp->qc_active |= (0x1 << qc->tag);
 
-	ap->ops->tf_load(ap, &qc->tf);	 /* load tf registers */
-	ap->ops->exec_command(ap, &qc->tf);
+	ap->ops->sff_tf_load(ap, &qc->tf);	 /* load tf registers */
+	ap->ops->sff_exec_command(ap, &qc->tf);
 
 	DPRINTK("Issued tag %u\n", qc->tag);
 
@@ -2060,7 +2060,7 @@ static int nv_swncq_sdbfis(struct ata_port *ap)
 		return -EINVAL;
 	}
 
-	ap->ops->irq_clear(ap);
+	ap->ops->sff_irq_clear(ap);
 	__ata_bmdma_stop(ap);
 
 	sactive = readl(pp->sactive_block);
@@ -2182,7 +2182,7 @@ static void nv_swncq_host_interrupt(struct ata_port *ap, u16 fis)
 	u8 ata_stat;
 	int rc = 0;
 
-	ata_stat = ap->ops->check_status(ap);
+	ata_stat = ap->ops->sff_check_status(ap);
 	nv_swncq_irq_clear(ap, fis);
 	if (!fis)
 		return;
@@ -2245,7 +2245,7 @@ static void nv_swncq_host_interrupt(struct ata_port *ap, u16 fis)
 
 		if (!(fis & NV_SWNCQ_IRQ_DMASETUP) &&
 		    !(pp->ncq_flags & ncq_saw_dmas)) {
-			ata_stat = ap->ops->check_status(ap);
+			ata_stat = ap->ops->sff_check_status(ap);
 			if (ata_stat & ATA_BUSY)
 				goto irq_exit;
 
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index 9923e860eae3..5a10dc5048ad 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -162,12 +162,12 @@ static struct scsi_host_template pdc_ata_sht = {
 static const struct ata_port_operations pdc_common_ops = {
 	.inherits		= &ata_sff_port_ops,
 
-	.tf_load		= pdc_tf_load_mmio,
-	.exec_command		= pdc_exec_command_mmio,
+	.sff_tf_load		= pdc_tf_load_mmio,
+	.sff_exec_command	= pdc_exec_command_mmio,
 	.check_atapi_dma	= pdc_check_atapi_dma,
 	.qc_prep		= pdc_qc_prep,
 	.qc_issue		= pdc_qc_issue,
-	.irq_clear		= pdc_irq_clear,
+	.sff_irq_clear		= pdc_irq_clear,
 
 	.post_internal_cmd	= pdc_post_internal_cmd,
 	.error_handler		= pdc_error_handler,
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index 987313b68f3b..88bf4212590f 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -369,7 +369,7 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 
 	if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) {
 		/* this sometimes happens, just clear IRQ */
-		ap->ops->check_status(ap);
+		ap->ops->sff_check_status(ap);
 		return;
 	}
 
@@ -405,7 +405,7 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 	}
 
 	/* check main status, clearing INTRQ */
-	status = ap->ops->check_status(ap);
+	status = ap->ops->sff_check_status(ap);
 	if (unlikely(status & ATA_BUSY))
 		goto err_hsm;
 
@@ -480,7 +480,7 @@ static void sil_thaw(struct ata_port *ap)
 	u32 tmp;
 
 	/* clear IRQ */
-	ap->ops->check_status(ap);
+	ap->ops->sff_check_status(ap);
 	ata_sff_irq_clear(ap);
 
 	/* turn on SATA IRQ if supported */
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index 67df1d753305..b83851f6e068 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -401,9 +401,9 @@ static struct scsi_host_template sil24_sht = {
 static struct ata_port_operations sil24_ops = {
 	.inherits		= &sata_pmp_port_ops,
 
-	.check_status		= sil24_check_status,
-	.check_altstatus	= sil24_check_status,
-	.tf_read		= sil24_tf_read,
+	.sff_check_status	= sil24_check_status,
+	.sff_check_altstatus	= sil24_check_status,
+	.sff_tf_read		= sil24_tf_read,
 	.qc_defer		= sil24_qc_defer,
 	.qc_prep		= sil24_qc_prep,
 	.qc_issue		= sil24_qc_issue,
diff --git a/drivers/ata/sata_svw.c b/drivers/ata/sata_svw.c
index 7b941106f7de..16aa6839aa5a 100644
--- a/drivers/ata/sata_svw.c
+++ b/drivers/ata/sata_svw.c
@@ -233,7 +233,7 @@ static void k2_bmdma_setup_mmio(struct ata_queued_cmd *qc)
 
 	/* issue r/w command if this is not a ATA DMA command*/
 	if (qc->tf.protocol != ATA_PROT_DMA)
-		ap->ops->exec_command(ap, &qc->tf);
+		ap->ops->sff_exec_command(ap, &qc->tf);
 }
 
 /**
@@ -269,7 +269,7 @@ static void k2_bmdma_start_mmio(struct ata_queued_cmd *qc)
 	   and the start command. */
 	/* issue r/w command if the access is to ATA*/
 	if (qc->tf.protocol == ATA_PROT_DMA)
-		ap->ops->exec_command(ap, &qc->tf);
+		ap->ops->sff_exec_command(ap, &qc->tf);
 }
 
 
@@ -336,9 +336,9 @@ static struct scsi_host_template k2_sata_sht = {
 
 static struct ata_port_operations k2_sata_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.tf_load		= k2_sata_tf_load,
-	.tf_read		= k2_sata_tf_read,
-	.check_status		= k2_stat_check_status,
+	.sff_tf_load		= k2_sata_tf_load,
+	.sff_tf_read		= k2_sata_tf_read,
+	.sff_check_status	= k2_stat_check_status,
 	.check_atapi_dma	= k2_sata_check_atapi_dma,
 	.bmdma_setup		= k2_bmdma_setup_mmio,
 	.bmdma_start		= k2_bmdma_start_mmio,
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 40be2ff60902..6107eff731f3 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -243,18 +243,18 @@ static struct scsi_host_template pdc_sata_sht = {
 
 /* TODO: inherit from base port_ops after converting to new EH */
 static struct ata_port_operations pdc_20621_ops = {
-	.tf_load		= pdc_tf_load_mmio,
-	.tf_read		= ata_sff_tf_read,
-	.check_status		= ata_sff_check_status,
-	.exec_command		= pdc_exec_command_mmio,
-	.dev_select		= ata_sff_dev_select,
+	.sff_tf_load		= pdc_tf_load_mmio,
+	.sff_tf_read		= ata_sff_tf_read,
+	.sff_check_status	= ata_sff_check_status,
+	.sff_exec_command	= pdc_exec_command_mmio,
+	.sff_dev_select		= ata_sff_dev_select,
 	.phy_reset		= pdc_20621_phy_reset,
 	.qc_prep		= pdc20621_qc_prep,
 	.qc_issue		= pdc20621_qc_issue,
-	.data_xfer		= ata_sff_data_xfer,
+	.sff_data_xfer		= ata_sff_data_xfer,
 	.eng_timeout		= pdc_eng_timeout,
-	.irq_clear		= pdc20621_irq_clear,
-	.irq_on			= ata_sff_irq_on,
+	.sff_irq_clear		= pdc20621_irq_clear,
+	.sff_irq_on		= ata_sff_irq_on,
 	.port_start		= pdc_port_start,
 };
 
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 9323bc2a2785..e5df37689740 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -173,7 +173,7 @@ static void svia_noop_freeze(struct ata_port *ap)
 	/* Some VIA controllers choke if ATA_NIEN is manipulated in
 	 * certain way.  Leave it alone and just clear pending IRQ.
 	 */
-	ap->ops->check_status(ap);
+	ap->ops->sff_check_status(ap);
 	ata_sff_irq_clear(ap);
 }
 
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index 2c3c7693c750..f3d635c0a2e9 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -251,7 +251,7 @@ static void vsc_port_intr(u8 port_status, struct ata_port *ap)
 	 * simply clear the interrupt
 	 */
 	if (unlikely(!handled))
-		ap->ops->check_status(ap);
+		ap->ops->sff_check_status(ap);
 }
 
 /*
@@ -306,8 +306,8 @@ static struct scsi_host_template vsc_sata_sht = {
 
 static struct ata_port_operations vsc_sata_ops = {
 	.inherits		= &ata_bmdma_port_ops,
-	.tf_load		= vsc_sata_tf_load,
-	.tf_read		= vsc_sata_tf_read,
+	.sff_tf_load		= vsc_sata_tf_load,
+	.sff_tf_read		= vsc_sata_tf_read,
 	.freeze			= vsc_freeze,
 	.thaw			= vsc_thaw,
 	.scr_read		= vsc_sata_scr_read,
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 79fd2436bd70..5bddae19d1fb 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -5271,13 +5271,13 @@ static u8 ipr_ata_check_altstatus(struct ata_port *ap)
 }
 
 static struct ata_port_operations ipr_sata_ops = {
-	.check_status = ipr_ata_check_status,
-	.check_altstatus = ipr_ata_check_altstatus,
-	.dev_select = ata_noop_dev_select,
+	.sff_check_status = ipr_ata_check_status,
+	.sff_check_altstatus = ipr_ata_check_altstatus,
+	.sff_dev_select = ata_noop_dev_select,
 	.phy_reset = ipr_ata_phy_reset,
 	.hardreset = ipr_sata_reset,
 	.post_internal_cmd = ipr_ata_post_internal,
-	.tf_read = ipr_tf_read,
+	.sff_tf_read = ipr_tf_read,
 	.qc_prep = ata_noop_qc_prep,
 	.qc_issue = ipr_qc_issue,
 	.port_start = ata_sas_port_start,
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index b0e5ac372a32..a1664b87927a 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -348,12 +348,12 @@ static int sas_ata_scr_read(struct ata_port *ap, unsigned int sc_reg_in,
 }
 
 static struct ata_port_operations sas_sata_ops = {
-	.check_status		= sas_ata_check_status,
-	.check_altstatus	= sas_ata_check_status,
-	.dev_select		= ata_noop_dev_select,
+	.sff_check_status	= sas_ata_check_status,
+	.sff_check_altstatus	= sas_ata_check_status,
+	.sff_dev_select		= ata_noop_dev_select,
 	.phy_reset		= sas_ata_phy_reset,
 	.post_internal_cmd	= sas_ata_post_internal,
-	.tf_read		= sas_ata_tf_read,
+	.sff_tf_read		= sas_ata_tf_read,
 	.qc_prep		= ata_noop_qc_prep,
 	.qc_issue		= sas_ata_qc_issue,
 	.port_start		= ata_sas_port_start,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 66663bfe2c71..584eca19b8f1 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -743,17 +743,18 @@ struct ata_port_operations {
 	/*
 	 * SFF / taskfile oriented ops
 	 */
-	void (*dev_select)(struct ata_port *ap, unsigned int device);
-	u8   (*check_status)(struct ata_port *ap);
-	u8   (*check_altstatus)(struct ata_port *ap);
-	void (*tf_load)(struct ata_port *ap, const struct ata_taskfile *tf);
-	void (*tf_read)(struct ata_port *ap, struct ata_taskfile *tf);
-	void (*exec_command)(struct ata_port *ap, const struct ata_taskfile *tf);
-	unsigned int (*data_xfer)(struct ata_device *dev, unsigned char *buf,
-				  unsigned int buflen, int rw);
-	u8   (*irq_on)(struct ata_port *);
-
-	void (*irq_clear)(struct ata_port *);
+	void (*sff_dev_select)(struct ata_port *ap, unsigned int device);
+	u8   (*sff_check_status)(struct ata_port *ap);
+	u8   (*sff_check_altstatus)(struct ata_port *ap);
+	void (*sff_tf_load)(struct ata_port *ap, const struct ata_taskfile *tf);
+	void (*sff_tf_read)(struct ata_port *ap, struct ata_taskfile *tf);
+	void (*sff_exec_command)(struct ata_port *ap,
+				 const struct ata_taskfile *tf);
+	unsigned int (*sff_data_xfer)(struct ata_device *dev,
+			unsigned char *buf, unsigned int buflen, int rw);
+	u8   (*sff_irq_on)(struct ata_port *);
+	void (*sff_irq_clear)(struct ata_port *);
+
 	void (*bmdma_setup)(struct ata_queued_cmd *qc);
 	void (*bmdma_start)(struct ata_queued_cmd *qc);
 	void (*bmdma_stop)(struct ata_queued_cmd *qc);
@@ -1438,7 +1439,7 @@ static inline u8 ata_sff_busy_wait(struct ata_port *ap, unsigned int bits,
 
 	do {
 		udelay(10);
-		status = ap->ops->check_status(ap);
+		status = ap->ops->sff_check_status(ap);
 		max--;
 	} while (status != 0xff && (status & bits) && (max > 0));
 
-- 
cgit v1.2.3


From 288623a06c652239d2f57d271af12bb024cf7218 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:17 +0900
Subject: libata: clean up port_ops->sff_irq_clear()

->sff_irq_clear() is called only from SFF interrupt handler, so there
is no reason to initialize it for non-SFF controllers.  Also,
ata_sff_irq_clear() can handle both BMDMA and non-BMDMA SFF
controllers.

This patch kills ata_noop_irq_clear() and removes it from base
port_ops and sets ->sff_irq_clear to ata_sff_irq_clear() in sff
port_ops instead of bmdma port_ops.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c | 10 ----------
 drivers/ata/libata-sff.c  |  2 +-
 include/linux/libata.h    |  1 -
 3 files changed, 1 insertion(+), 12 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index a01e02c5ce7a..fd912ccb90f8 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -74,7 +74,6 @@ const unsigned long sata_deb_timing_hotplug[]		= {  25,  500, 2000 };
 const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
 const struct ata_port_operations ata_base_port_ops = {
-	.sff_irq_clear		= ata_noop_irq_clear,
 	.prereset		= ata_sff_prereset,
 	.hardreset		= sata_sff_hardreset,
 	.postreset		= ata_sff_postreset,
@@ -386,14 +385,6 @@ int atapi_cmd_type(u8 opcode)
 	}
 }
 
-/**
- *	ata_noop_irq_clear - Noop placeholder for irq_clear
- *	@ap: Port associated with this ATA transaction.
- */
-void ata_noop_irq_clear(struct ata_port *ap)
-{
-}
-
 /**
  *	ata_tf_to_fis - Convert ATA taskfile to SATA FIS structure
  *	@tf: Taskfile to convert
@@ -6108,7 +6099,6 @@ EXPORT_SYMBOL_GPL(ata_port_start);
 EXPORT_SYMBOL_GPL(ata_do_set_mode);
 EXPORT_SYMBOL_GPL(ata_std_qc_defer);
 EXPORT_SYMBOL_GPL(ata_noop_qc_prep);
-EXPORT_SYMBOL_GPL(ata_noop_irq_clear);
 EXPORT_SYMBOL_GPL(ata_port_probe);
 EXPORT_SYMBOL_GPL(ata_dev_disable);
 EXPORT_SYMBOL_GPL(sata_set_spd);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 04024a556660..c11601617134 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -58,6 +58,7 @@ const struct ata_port_operations ata_sff_port_ops = {
 	.sff_exec_command	= ata_sff_exec_command,
 	.sff_data_xfer		= ata_sff_data_xfer,
 	.sff_irq_on		= ata_sff_irq_on,
+	.sff_irq_clear		= ata_sff_irq_clear,
 
 	.port_start		= ata_sff_port_start,
 };
@@ -71,7 +72,6 @@ const struct ata_port_operations ata_bmdma_port_ops = {
 	.bmdma_start		= ata_bmdma_start,
 	.bmdma_stop		= ata_bmdma_stop,
 	.bmdma_status		= ata_bmdma_status,
-	.sff_irq_clear		= ata_sff_irq_clear,
 };
 
 /**
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 584eca19b8f1..603712b59cf3 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -896,7 +896,6 @@ extern void ata_id_string(const u16 *id, unsigned char *s,
 			  unsigned int ofs, unsigned int len);
 extern void ata_id_c_string(const u16 *id, unsigned char *s,
 			    unsigned int ofs, unsigned int len);
-extern void ata_noop_irq_clear(struct ata_port *ap);
 extern void ata_qc_complete(struct ata_queued_cmd *qc);
 extern int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active,
 				    void (*finish_qc)(struct ata_queued_cmd *));
-- 
cgit v1.2.3


From 0aa1113d544226bc2c4a20d6ac1d71170512a361 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:18 +0900
Subject: libata: separate out ata_std_prereset() from ata_sff_prereset()

Separate out generic ATA portion from ata_sff_prereset() into
ata_std_prereset() and implement ata_sff_prereset() using the std
version.  Waiting for device readiness is the only SFF specific part.

ata_base_port_ops now has ata_std_prereset() for its prereset and
ata_sff_port_ops overrides it to ata_sff_prereset().  This change can
affect pdc_adma, ahci, sata_fsl and sata_sil24.  pdc_adma implements
its own prereset using ata_sff_prereset() and the rest has hardreset
and thus are unaffected by this change.

This change reflects real world situation.  There is no generic way to
wait for device readiness for non-SFF controllers and some of them
don't have any mechanism for that.  Non-sff drivers which don't have
hardreset should wrap ata_std_prereset() and wait for device readiness
itself but there's no such driver now and isn't likely to be popular
in the future either.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c | 17 ++++-------------
 drivers/ata/libata-sff.c  | 43 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/libata.h    |  1 +
 3 files changed, 48 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index fd912ccb90f8..5f771bb44332 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -74,7 +74,7 @@ const unsigned long sata_deb_timing_hotplug[]		= {  25,  500, 2000 };
 const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
 const struct ata_port_operations ata_base_port_ops = {
-	.prereset		= ata_sff_prereset,
+	.prereset		= ata_std_prereset,
 	.hardreset		= sata_sff_hardreset,
 	.postreset		= ata_sff_postreset,
 	.error_handler		= ata_std_error_handler,
@@ -3416,7 +3416,7 @@ int sata_link_resume(struct ata_link *link, const unsigned long *params,
 }
 
 /**
- *	ata_sff_prereset - prepare for reset
+ *	ata_std_prereset - prepare for reset
  *	@link: ATA link to be reset
  *	@deadline: deadline jiffies for the operation
  *
@@ -3432,7 +3432,7 @@ int sata_link_resume(struct ata_link *link, const unsigned long *params,
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_sff_prereset(struct ata_link *link, unsigned long deadline)
+int ata_std_prereset(struct ata_link *link, unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
 	struct ata_eh_context *ehc = &link->eh_context;
@@ -3452,16 +3452,6 @@ int ata_sff_prereset(struct ata_link *link, unsigned long deadline)
 					"link for reset (errno=%d)\n", rc);
 	}
 
-	/* wait for !BSY if we don't know that no device is attached */
-	if (!ata_link_offline(link)) {
-		rc = ata_sff_wait_ready(ap, deadline);
-		if (rc && rc != -ENODEV) {
-			ata_link_printk(link, KERN_WARNING, "device not ready "
-					"(errno=%d), forcing hardreset\n", rc);
-			ehc->i.action |= ATA_EH_HARDRESET;
-		}
-	}
-
 	return 0;
 }
 
@@ -6104,6 +6094,7 @@ EXPORT_SYMBOL_GPL(ata_dev_disable);
 EXPORT_SYMBOL_GPL(sata_set_spd);
 EXPORT_SYMBOL_GPL(sata_link_debounce);
 EXPORT_SYMBOL_GPL(sata_link_resume);
+EXPORT_SYMBOL_GPL(ata_std_prereset);
 EXPORT_SYMBOL_GPL(sata_link_hardreset);
 EXPORT_SYMBOL_GPL(ata_dev_classify);
 EXPORT_SYMBOL_GPL(ata_dev_pair);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index c11601617134..9234bc047956 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -47,6 +47,7 @@ const struct ata_port_operations ata_sff_port_ops = {
 
 	.freeze			= ata_sff_freeze,
 	.thaw			= ata_sff_thaw,
+	.prereset		= ata_sff_prereset,
 	.softreset		= ata_sff_softreset,
 	.error_handler		= ata_sff_error_handler,
 	.post_internal_cmd	= ata_sff_post_internal_cmd,
@@ -1606,6 +1607,48 @@ void ata_sff_thaw(struct ata_port *ap)
 	ap->ops->sff_irq_on(ap);
 }
 
+/**
+ *	ata_sff_prereset - prepare SFF link for reset
+ *	@link: SFF link to be reset
+ *	@deadline: deadline jiffies for the operation
+ *
+ *	SFF link @link is about to be reset.  Initialize it.  It first
+ *	calls ata_std_prereset() and wait for !BSY if the port is
+ *	being softreset.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep)
+ *
+ *	RETURNS:
+ *	0 on success, -errno otherwise.
+ */
+int ata_sff_prereset(struct ata_link *link, unsigned long deadline)
+{
+	struct ata_port *ap = link->ap;
+	struct ata_eh_context *ehc = &link->eh_context;
+	int rc;
+
+	rc = ata_std_prereset(link, deadline);
+	if (rc)
+		return rc;
+
+	/* if we're about to do hardreset, nothing more to do */
+	if (ehc->i.action & ATA_EH_HARDRESET)
+		return 0;
+
+	/* wait for !BSY if we don't know that no device is attached */
+	if (!ata_link_offline(link)) {
+		rc = ata_sff_wait_ready(ap, deadline);
+		if (rc && rc != -ENODEV) {
+			ata_link_printk(link, KERN_WARNING, "device not ready "
+					"(errno=%d), forcing hardreset\n", rc);
+			ehc->i.action |= ATA_EH_HARDRESET;
+		}
+	}
+
+	return 0;
+}
+
 /**
  *	ata_devchk - PATA device presence detection
  *	@ap: ATA channel to examine
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 603712b59cf3..595ede55fe4c 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -824,6 +824,7 @@ static inline int ata_port_is_dummy(struct ata_port *ap)
 extern void sata_print_link_status(struct ata_link *link);
 extern void ata_port_probe(struct ata_port *);
 extern int sata_set_spd(struct ata_link *link);
+extern int ata_std_prereset(struct ata_link *link, unsigned long deadline);
 extern int sata_link_debounce(struct ata_link *link,
 			const unsigned long *params, unsigned long deadline);
 extern int sata_link_resume(struct ata_link *link, const unsigned long *params,
-- 
cgit v1.2.3


From 203c75b8245c5386044721d9c5eda5c6b71b3d14 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:18 +0900
Subject: libata: separate out ata_std_postreset() from ata_sff_postreset()

Separate out generic ATA portion from ata_sff_postreset() into
ata_std_postreset() and implement ata_sff_postreset() using the std
version.

ata_base_port_ops now has ata_std_postreset() for its postreset and
ata_sff_port_ops overrides it to ata_sff_postreset().

This change affects pdc_adma, ahci, sata_fsl and sata_sil24.  pdc_adma
now specifies postreset to ata_sff_postreset() explicitly.  sata_fsl
and sata_sil24 now use ata_std_postreset() which makes no difference
to them.  ahci now calls ata_std_postreset() from its own postreset
method, which causes no behavior difference.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c        |  2 +-
 drivers/ata/libata-core.c | 24 ++++--------------------
 drivers/ata/libata-sff.c  | 36 ++++++++++++++++++++++++++++++++++++
 include/linux/libata.h    |  1 +
 4 files changed, 42 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index c5e4501daa74..939dc1d4e50d 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1444,7 +1444,7 @@ static void ahci_postreset(struct ata_link *link, unsigned int *class)
 	void __iomem *port_mmio = ahci_port_base(ap);
 	u32 new_tmp, tmp;
 
-	ata_sff_postreset(link, class);
+	ata_std_postreset(link, class);
 
 	/* Make sure port's ATAPI bit is set appropriately */
 	new_tmp = tmp = readl(port_mmio + PORT_CMD);
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 5f771bb44332..eaead76c9443 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -76,7 +76,7 @@ const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 const struct ata_port_operations ata_base_port_ops = {
 	.prereset		= ata_std_prereset,
 	.hardreset		= sata_sff_hardreset,
-	.postreset		= ata_sff_postreset,
+	.postreset		= ata_std_postreset,
 	.error_handler		= ata_std_error_handler,
 };
 
@@ -3516,7 +3516,7 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 }
 
 /**
- *	ata_sff_postreset - standard postreset callback
+ *	ata_std_postreset - standard postreset callback
  *	@link: the target ata_link
  *	@classes: classes of attached devices
  *
@@ -3527,9 +3527,8 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
  *	LOCKING:
  *	Kernel thread context (may sleep)
  */
-void ata_sff_postreset(struct ata_link *link, unsigned int *classes)
+void ata_std_postreset(struct ata_link *link, unsigned int *classes)
 {
-	struct ata_port *ap = link->ap;
 	u32 serror;
 
 	DPRINTK("ENTER\n");
@@ -3542,22 +3541,6 @@ void ata_sff_postreset(struct ata_link *link, unsigned int *classes)
 		sata_scr_write(link, SCR_ERROR, serror);
 	link->eh_info.serror = 0;
 
-	/* is double-select really necessary? */
-	if (classes[0] != ATA_DEV_NONE)
-		ap->ops->sff_dev_select(ap, 1);
-	if (classes[1] != ATA_DEV_NONE)
-		ap->ops->sff_dev_select(ap, 0);
-
-	/* bail out if no device is present */
-	if (classes[0] == ATA_DEV_NONE && classes[1] == ATA_DEV_NONE) {
-		DPRINTK("EXIT, no device\n");
-		return;
-	}
-
-	/* set up device control */
-	if (ap->ioaddr.ctl_addr)
-		iowrite8(ap->ctl, ap->ioaddr.ctl_addr);
-
 	DPRINTK("EXIT\n");
 }
 
@@ -6096,6 +6079,7 @@ EXPORT_SYMBOL_GPL(sata_link_debounce);
 EXPORT_SYMBOL_GPL(sata_link_resume);
 EXPORT_SYMBOL_GPL(ata_std_prereset);
 EXPORT_SYMBOL_GPL(sata_link_hardreset);
+EXPORT_SYMBOL_GPL(ata_std_postreset);
 EXPORT_SYMBOL_GPL(ata_dev_classify);
 EXPORT_SYMBOL_GPL(ata_dev_pair);
 EXPORT_SYMBOL_GPL(ata_port_disable);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 9234bc047956..e530baccc9cb 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -49,6 +49,7 @@ const struct ata_port_operations ata_sff_port_ops = {
 	.thaw			= ata_sff_thaw,
 	.prereset		= ata_sff_prereset,
 	.softreset		= ata_sff_softreset,
+	.postreset		= ata_sff_postreset,
 	.error_handler		= ata_sff_error_handler,
 	.post_internal_cmd	= ata_sff_post_internal_cmd,
 
@@ -2031,6 +2032,41 @@ int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
 	return 0;
 }
 
+/**
+ *	ata_sff_postreset - SFF postreset callback
+ *	@link: the target SFF ata_link
+ *	@classes: classes of attached devices
+ *
+ *	This function is invoked after a successful reset.  It first
+ *	calls ata_std_postreset() and performs SFF specific postreset
+ *	processing.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep)
+ */
+void ata_sff_postreset(struct ata_link *link, unsigned int *classes)
+{
+	struct ata_port *ap = link->ap;
+
+	ata_std_postreset(link, classes);
+
+	/* is double-select really necessary? */
+	if (classes[0] != ATA_DEV_NONE)
+		ap->ops->sff_dev_select(ap, 1);
+	if (classes[1] != ATA_DEV_NONE)
+		ap->ops->sff_dev_select(ap, 0);
+
+	/* bail out if no device is present */
+	if (classes[0] == ATA_DEV_NONE && classes[1] == ATA_DEV_NONE) {
+		DPRINTK("EXIT, no device\n");
+		return;
+	}
+
+	/* set up device control */
+	if (ap->ioaddr.ctl_addr)
+		iowrite8(ap->ctl, ap->ioaddr.ctl_addr);
+}
+
 /**
  *	ata_sff_error_handler - Stock error handler for BMDMA controller
  *	@ap: port to handle error for
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 595ede55fe4c..6e14c27319d5 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -831,6 +831,7 @@ extern int sata_link_resume(struct ata_link *link, const unsigned long *params,
 			    unsigned long deadline);
 extern int sata_link_hardreset(struct ata_link *link,
 			const unsigned long *timing, unsigned long deadline);
+extern void ata_std_postreset(struct ata_link *link, unsigned int *classes);
 extern void ata_port_disable(struct ata_port *);
 
 extern struct ata_host *ata_host_alloc(struct device *dev, int max_ports);
-- 
cgit v1.2.3


From 705e76beb90b97421e1f61e857c4246799781bb5 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:19 +0900
Subject: libata: restructure SFF post-reset readiness waits

Previously, post-softreset readiness is waited as follows.

1. ata_sff_wait_after_reset() waits for 150ms and then for
   ATA_TMOUT_FF_WAIT if status is 0xff and other conditions meet.

2. ata_bus_softreset() finishes with -ENODEV if status is still 0xff.
   If not, continue to #3.

3. ata_bus_post_reset() waits readiness of dev0 and/or dev1 depending
   on devmask using ata_sff_wait_ready().

And for post-hardreset readiness,

1. ata_sff_wait_after_reset() waits for 150ms and then for
   ATA_TMOUT_FF_WAIT if status is 0xff and other conditions meet.

2. sata_sff_hardreset waits for device readiness using
   ata_sff_wait_ready().

This patch merges and unifies post-reset readiness waits into
ata_sff_wait_ready() and ata_sff_wait_after_reset().

ATA_TMOUT_FF_WAIT handling is merged into ata_sff_wait_ready().  If TF
status is 0xff, link status is unknown and the port is SATA, it will
continue polling till ATA_TMOUT_FF_WAIT.

ata_sff_wait_after_reset() is updated to perform the following steps.

1. waits for 150ms.

2. waits for dev0 readiness using ata_sff_wait_ready().  Note that
   this is done regardless of devmask, as ata_sff_wait_ready() handles
   0xff status correctly, this preserves the original behavior except
   that it may wait longer after softreset if link is online but
   status is 0xff.  This behavior change is very unlikely to cause any
   actual difference and is intended.  It brings softreset behavior to
   that of hardreset.

3. waits for dev1 readiness just the same way ata_bus_post_reset() did.

Now both soft and hard resets call ata_sff_wait_after_reset() after
reset to wait for readiness after resets.  As
ata_sff_wait_after_reset() contains calls to ->sff_dev_select(),
explicit call near the end of sata_sff_hardreset() is removed.

This change makes reset implementation simpler and more consistent.

While at it, make the magical 150ms wait post-reset wait duration a
constant and ata_sff_wait_ready() and ata_sff_wait_after_reset() take
@link instead of @ap.  This is to make them consistent with other
reset helpers and ease core changes.

pata_scc is updated accordingly.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c          |  11 +--
 drivers/ata/libata-sff.c    | 161 ++++++++++++++++++--------------------------
 drivers/ata/pata_scc.c      |  91 ++++++++++++++-----------
 drivers/ata/sata_inic162x.c |   6 +-
 drivers/ata/sata_via.c      |   2 +-
 include/linux/libata.h      |  18 ++++-
 6 files changed, 139 insertions(+), 150 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 939dc1d4e50d..45a67a9ad8ab 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1302,10 +1302,8 @@ static int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 	tf.ctl &= ~ATA_SRST;
 	ahci_exec_polled_cmd(ap, pmp, &tf, 0, 0, 0);
 
-	/* wait a while before checking status */
-	ata_sff_wait_after_reset(ap, deadline);
-
-	rc = ata_sff_wait_ready(ap, deadline);
+	/* wait for link to become ready */
+	rc = ata_sff_wait_after_reset(link, 1, deadline);
 	/* link occupied, -ENODEV too is an error */
 	if (rc) {
 		reason = "device not ready";
@@ -1415,9 +1413,6 @@ static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
 	if (rc || ata_link_offline(link))
 		return rc;
 
-	/* spec mandates ">= 2ms" before checking status */
-	msleep(150);
-
 	/* The pseudo configuration device on SIMG4726 attached to
 	 * ASUS P5W-DH Deluxe doesn't send signature FIS after
 	 * hardreset if no device is attached to the first downstream
@@ -1431,7 +1426,7 @@ static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
 	 * have to be reset again.  For most cases, this should
 	 * suffice while making probing snappish enough.
 	 */
-	rc = ata_sff_wait_ready(ap, jiffies + 2 * HZ);
+	rc = ata_sff_wait_after_reset(link, 1, jiffies + 2 * HZ);
 	if (rc)
 		ahci_kick_engine(ap, 0);
 
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index e530baccc9cb..6e8de3c1595e 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -310,7 +310,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 
 /**
  *	ata_sff_wait_ready - sleep until BSY clears, or timeout
- *	@ap: port containing status register to be polled
+ *	@link: SFF link to wait ready status for
  *	@deadline: deadline jiffies for the operation
  *
  *	Sleep until ATA Status register bit BSY clears, or timeout
@@ -322,26 +322,52 @@ int ata_sff_busy_sleep(struct ata_port *ap,
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_sff_wait_ready(struct ata_port *ap, unsigned long deadline)
+int ata_sff_wait_ready(struct ata_link *link, unsigned long deadline)
 {
+	struct ata_port *ap = link->ap;
 	unsigned long start = jiffies;
+	unsigned long nodev_deadline = start + ATA_TMOUT_FF_WAIT;
 	int warned = 0;
 
+	if (time_after(nodev_deadline, deadline))
+		nodev_deadline = deadline;
+
 	while (1) {
 		u8 status = ap->ops->sff_check_status(ap);
 		unsigned long now = jiffies;
 
 		if (!(status & ATA_BUSY))
 			return 0;
-		if (!ata_link_online(&ap->link) && status == 0xff)
-			return -ENODEV;
+
+		/* No device status could be transient.  Ignore it if
+		 * link is online.  Also, some SATA devices take a
+		 * long time to clear 0xff after reset.  For example,
+		 * HHD424020F7SV00 iVDR needs >= 800ms while Quantum
+		 * GoVault needs even more than that.  Wait for
+		 * ATA_TMOUT_FF_WAIT on -ENODEV if link isn't offline.
+		 *
+		 * Note that some PATA controllers (pata_ali) explode
+		 * if status register is read more than once when
+		 * there's no device attached.
+		 */
+		if (status == 0xff) {
+			if (ata_link_online(link))
+				status = ATA_BUSY;
+			else if ((link->ap->flags & ATA_FLAG_SATA) &&
+				 !ata_link_offline(link) &&
+				 time_before(now, nodev_deadline))
+				status = ATA_BUSY;
+			if (status == 0xff)
+				return -ENODEV;
+		}
+
 		if (time_after(now, deadline))
 			return -EBUSY;
 
 		if (!warned && time_after(now, start + 5 * HZ) &&
 		    (deadline - now > 3 * HZ)) {
-			ata_port_printk(ap, KERN_WARNING,
-				"port is slow to respond, please be patient "
+			ata_link_printk(link, KERN_WARNING,
+				"link is slow to respond, please be patient "
 				"(Status 0x%x)\n", status);
 			warned = 1;
 		}
@@ -1625,7 +1651,6 @@ void ata_sff_thaw(struct ata_port *ap)
  */
 int ata_sff_prereset(struct ata_link *link, unsigned long deadline)
 {
-	struct ata_port *ap = link->ap;
 	struct ata_eh_context *ehc = &link->eh_context;
 	int rc;
 
@@ -1639,7 +1664,7 @@ int ata_sff_prereset(struct ata_link *link, unsigned long deadline)
 
 	/* wait for !BSY if we don't know that no device is attached */
 	if (!ata_link_offline(link)) {
-		rc = ata_sff_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(link, deadline);
 		if (rc && rc != -ENODEV) {
 			ata_link_printk(link, KERN_WARNING, "device not ready "
 					"(errno=%d), forcing hardreset\n", rc);
@@ -1762,25 +1787,41 @@ unsigned int ata_sff_dev_classify(struct ata_device *dev, int present,
 	return class;
 }
 
-static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
-			      unsigned long deadline)
+/**
+ *	ata_sff_wait_after_reset - wait for devices to become ready after reset
+ *	@link: SFF link which is just reset
+ *	@devmask: mask of present devices
+ *	@deadline: deadline jiffies for the operation
+ *
+ *	Wait devices attached to SFF @link to become ready after
+ *	reset.  It contains preceding 150ms wait to avoid accessing TF
+ *	status register too early.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep).
+ *
+ *	RETURNS:
+ *	0 on success, -ENODEV if some or all of devices in @devmask
+ *	don't seem to exist.  -errno on other errors.
+ */
+int ata_sff_wait_after_reset(struct ata_link *link, unsigned int devmask,
+			     unsigned long deadline)
 {
+	struct ata_port *ap = link->ap;
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 	unsigned int dev0 = devmask & (1 << 0);
 	unsigned int dev1 = devmask & (1 << 1);
 	int rc, ret = 0;
 
-	/* if device 0 was found in ata_devchk, wait for its
-	 * BSY bit to clear
+	msleep(ATA_WAIT_AFTER_RESET_MSECS);
+
+	/* always check readiness of the master device */
+	rc = ata_sff_wait_ready(link, deadline);
+	/* -ENODEV means the odd clown forgot the D7 pulldown resistor
+	 * and TF status is 0xff, bail out on it too.
 	 */
-	if (dev0) {
-		rc = ata_sff_wait_ready(ap, deadline);
-		if (rc) {
-			if (rc != -ENODEV)
-				return rc;
-			ret = rc;
-		}
-	}
+	if (rc)
+		return rc;
 
 	/* if device 1 was found in ata_devchk, wait for register
 	 * access briefly, then wait for BSY to clear.
@@ -1804,7 +1845,7 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 			msleep(50);	/* give drive a breather */
 		}
 
-		rc = ata_sff_wait_ready(ap, deadline);
+		rc = ata_sff_wait_ready(link, deadline);
 		if (rc) {
 			if (rc != -ENODEV)
 				return rc;
@@ -1822,61 +1863,6 @@ static int ata_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	return ret;
 }
 
-/**
- *	ata_sff_wait_after_reset - wait before checking status after reset
- *	@ap: port containing status register to be polled
- *	@deadline: deadline jiffies for the operation
- *
- *	After reset, we need to pause a while before reading status.
- *	Also, certain combination of controller and device report 0xff
- *	for some duration (e.g. until SATA PHY is up and running)
- *	which is interpreted as empty port in ATA world.  This
- *	function also waits for such devices to get out of 0xff
- *	status.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep).
- */
-void ata_sff_wait_after_reset(struct ata_port *ap, unsigned long deadline)
-{
-	unsigned long until = jiffies + ATA_TMOUT_FF_WAIT;
-
-	if (time_before(until, deadline))
-		deadline = until;
-
-	/* Spec mandates ">= 2ms" before checking status.  We wait
-	 * 150ms, because that was the magic delay used for ATAPI
-	 * devices in Hale Landis's ATADRVR, for the period of time
-	 * between when the ATA command register is written, and then
-	 * status is checked.  Because waiting for "a while" before
-	 * checking status is fine, post SRST, we perform this magic
-	 * delay here as well.
-	 *
-	 * Old drivers/ide uses the 2mS rule and then waits for ready.
-	 */
-	msleep(150);
-
-	/* Wait for 0xff to clear.  Some SATA devices take a long time
-	 * to clear 0xff after reset.  For example, HHD424020F7SV00
-	 * iVDR needs >= 800ms while.  Quantum GoVault needs even more
-	 * than that.
-	 *
-	 * Note that some PATA controllers (pata_ali) explode if
-	 * status register is read more than once when there's no
-	 * device attached.
-	 */
-	if (ap->flags & ATA_FLAG_SATA) {
-		while (1) {
-			u8 status = ap->ops->sff_check_status(ap);
-
-			if (status != 0xff || time_after(jiffies, deadline))
-				return;
-
-			msleep(50);
-		}
-	}
-}
-
 static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
 			     unsigned long deadline)
 {
@@ -1891,17 +1877,8 @@ static int ata_bus_softreset(struct ata_port *ap, unsigned int devmask,
 	udelay(20);	/* FIXME: flush */
 	iowrite8(ap->ctl, ioaddr->ctl_addr);
 
-	/* wait a while before checking status */
-	ata_sff_wait_after_reset(ap, deadline);
-
-	/* Before we perform post reset processing we want to see if
-	 * the bus shows 0xFF because the odd clown forgets the D7
-	 * pulldown resistor.
-	 */
-	if (ap->ops->sff_check_status(ap) == 0xFF)
-		return -ENODEV;
-
-	return ata_bus_post_reset(ap, devmask, deadline);
+	/* wait the port to become ready */
+	return ata_sff_wait_after_reset(&ap->link, devmask, deadline);
 }
 
 /**
@@ -2003,20 +1980,18 @@ int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
 		return 0;
 	}
 
-	/* wait a while before checking status */
-	ata_sff_wait_after_reset(ap, deadline);
-
 	/* If PMP is supported, we have to do follow-up SRST.  Note
 	 * that some PMPs don't send D2H Reg FIS after hardreset at
 	 * all if the first port is empty.  Wait for it just for a
 	 * second and request follow-up SRST.
 	 */
 	if (ap->flags & ATA_FLAG_PMP) {
-		ata_sff_wait_ready(ap, jiffies + HZ);
+		ata_sff_wait_after_reset(link, 1, jiffies + HZ);
 		return -EAGAIN;
 	}
 
-	rc = ata_sff_wait_ready(ap, deadline);
+	/* wait for the link to become online */
+	rc = ata_sff_wait_after_reset(link, 1, deadline);
 	/* link occupied, -ENODEV too is an error */
 	if (rc) {
 		ata_link_printk(link, KERN_ERR,
@@ -2024,8 +1999,6 @@ int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
 		return rc;
 	}
 
-	ap->ops->sff_dev_select(ap, 0);	/* probably unnecessary */
-
 	*class = ata_sff_dev_classify(link->device, 1, NULL);
 
 	DPRINTK("EXIT, class=%u\n", *class);
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index 2b9da715c704..accc275e74cc 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -497,47 +497,68 @@ static unsigned int scc_devchk (struct ata_port *ap,
 }
 
 /**
- *	scc_bus_post_reset - PATA device post reset
+ *	scc_wait_after_reset - wait for devices to become ready after reset
  *
- *	Note: Original code is ata_bus_post_reset().
+ *	Note: Original code is ata_sff_wait_after_reset
  */
 
-static int scc_bus_post_reset(struct ata_port *ap, unsigned int devmask,
-                              unsigned long deadline)
+int scc_wait_after_reset(struct ata_link *link, unsigned int devmask,
+			 unsigned long deadline)
 {
+	struct ata_port *ap = link->ap;
 	struct ata_ioports *ioaddr = &ap->ioaddr;
 	unsigned int dev0 = devmask & (1 << 0);
 	unsigned int dev1 = devmask & (1 << 1);
-	int rc;
+	int rc, ret = 0;
+
+	/* Spec mandates ">= 2ms" before checking status.  We wait
+	 * 150ms, because that was the magic delay used for ATAPI
+	 * devices in Hale Landis's ATADRVR, for the period of time
+	 * between when the ATA command register is written, and then
+	 * status is checked.  Because waiting for "a while" before
+	 * checking status is fine, post SRST, we perform this magic
+	 * delay here as well.
+	 *
+	 * Old drivers/ide uses the 2mS rule and then waits for ready.
+	 */
+	msleep(150);
 
-	/* if device 0 was found in ata_devchk, wait for its
-	 * BSY bit to clear
+	/* always check readiness of the master device */
+	rc = ata_sff_wait_ready(link, deadline);
+	/* -ENODEV means the odd clown forgot the D7 pulldown resistor
+	 * and TF status is 0xff, bail out on it too.
 	 */
-	if (dev0) {
-		rc = ata_sff_wait_ready(ap, deadline);
-		if (rc && rc != -ENODEV)
-			return rc;
-	}
+	if (rc)
+		return rc;
 
-	/* if device 1 was found in ata_devchk, wait for
-	 * register access, then wait for BSY to clear
+	/* if device 1 was found in ata_devchk, wait for register
+	 * access briefly, then wait for BSY to clear.
 	 */
-	while (dev1) {
-		u8 nsect, lbal;
+	if (dev1) {
+		int i;
 
 		ap->ops->sff_dev_select(ap, 1);
-		nsect = in_be32(ioaddr->nsect_addr);
-		lbal = in_be32(ioaddr->lbal_addr);
-		if ((nsect == 1) && (lbal == 1))
-			break;
-		if (time_after(jiffies, deadline))
-			return -EBUSY;
-		msleep(50);	/* give drive a breather */
-	}
-	if (dev1) {
-		rc = ata_sff_wait_ready(ap, deadline);
-		if (rc && rc != -ENODEV)
-			return rc;
+
+		/* Wait for register access.  Some ATAPI devices fail
+		 * to set nsect/lbal after reset, so don't waste too
+		 * much time on it.  We're gonna wait for !BSY anyway.
+		 */
+		for (i = 0; i < 2; i++) {
+			u8 nsect, lbal;
+
+			nsect = in_be32(ioaddr->nsect_addr);
+			lbal = in_be32(ioaddr->lbal_addr);
+			if ((nsect == 1) && (lbal == 1))
+				break;
+			msleep(50);	/* give drive a breather */
+		}
+
+		rc = ata_sff_wait_ready(link, deadline);
+		if (rc) {
+			if (rc != -ENODEV)
+				return rc;
+			ret = rc;
+		}
 	}
 
 	/* is all this really necessary? */
@@ -547,7 +568,7 @@ static int scc_bus_post_reset(struct ata_port *ap, unsigned int devmask,
 	if (dev0)
 		ap->ops->sff_dev_select(ap, 0);
 
-	return 0;
+	return ret;
 }
 
 /**
@@ -570,17 +591,7 @@ static unsigned int scc_bus_softreset(struct ata_port *ap, unsigned int devmask,
 	udelay(20);
 	out_be32(ioaddr->ctl_addr, ap->ctl);
 
-	/* wait a while before checking status */
-	ata_sff_wait_after_reset(ap, deadline);
-
-	/* Before we perform post reset processing we want to see if
-	 * the bus shows 0xFF because the odd clown forgets the D7
-	 * pulldown resistor.
-	 */
-	if (scc_check_status(ap) == 0xFF)
-		return 0;
-
-	scc_bus_post_reset(ap, devmask, deadline);
+	scc_wait_after_reset(&ap->link, devmask, deadlien);
 
 	return 0;
 }
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 0b5a736a45e3..9f47d0022453 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -417,10 +417,8 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class,
 	if (ata_link_online(link)) {
 		struct ata_taskfile tf;
 
-		/* wait a while before checking status */
-		ata_sff_wait_after_reset(ap, deadline);
-
-		rc = ata_sff_wait_ready(ap, deadline);
+		/* wait for link to become ready */
+		rc = ata_sff_wait_after_reset(link, 1, deadline);
 		/* link occupied, -ENODEV too is an error */
 		if (rc) {
 			ata_link_printk(link, KERN_WARNING, "device not ready "
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index e5df37689740..96deeb354e16 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -242,7 +242,7 @@ static int vt6420_prereset(struct ata_link *link, unsigned long deadline)
 
  skip_scr:
 	/* wait for !BSY */
-	ata_sff_wait_ready(ap, deadline);
+	ata_sff_wait_ready(link, deadline);
 
 	return 0;
 }
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 6e14c27319d5..da5560244787 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -249,6 +249,18 @@ enum {
 	 */
 	ATA_TMOUT_FF_WAIT	= 4 * HZ / 5,
 
+	/* Spec mandates to wait for ">= 2ms" before checking status
+	 * after reset.  We wait 150ms, because that was the magic
+	 * delay used for ATAPI devices in Hale Landis's ATADRVR, for
+	 * the period of time between when the ATA command register is
+	 * written, and then status is checked.  Because waiting for
+	 * "a while" before checking status is fine, post SRST, we
+	 * perform this magic delay here as well.
+	 *
+	 * Old drivers/ide uses the 2mS rule and then waits for ready.
+	 */
+	ATA_WAIT_AFTER_RESET_MSECS = 150,
+
 	/* ATA bus states */
 	BUS_UNKNOWN		= 0,
 	BUS_DMA			= 1,
@@ -1351,7 +1363,7 @@ extern u8 ata_sff_check_status(struct ata_port *ap);
 extern u8 ata_sff_altstatus(struct ata_port *ap);
 extern int ata_sff_busy_sleep(struct ata_port *ap,
 			      unsigned long timeout_pat, unsigned long timeout);
-extern int ata_sff_wait_ready(struct ata_port *ap, unsigned long deadline);
+extern int ata_sff_wait_ready(struct ata_link *link, unsigned long deadline);
 extern void ata_sff_tf_load(struct ata_port *ap, const struct ata_taskfile *tf);
 extern void ata_sff_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
 extern void ata_sff_exec_command(struct ata_port *ap,
@@ -1373,8 +1385,8 @@ extern void ata_sff_thaw(struct ata_port *ap);
 extern int ata_sff_prereset(struct ata_link *link, unsigned long deadline);
 extern unsigned int ata_sff_dev_classify(struct ata_device *dev, int present,
 					  u8 *r_err);
-extern void ata_sff_wait_after_reset(struct ata_port *ap,
-				     unsigned long deadline);
+extern int ata_sff_wait_after_reset(struct ata_link *link, unsigned int devmask,
+				    unsigned long deadline);
 extern int ata_sff_softreset(struct ata_link *link, unsigned int *classes,
 			     unsigned long deadline);
 extern int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
-- 
cgit v1.2.3


From aa2731ad9ad80ac3fca48bd1c4cf0eceede4810e Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:19 +0900
Subject: libata: separate out ata_wait_ready() and implement
 ata_wait_after_reset()

Factor out waiting logic (which is common to all ATA controllers) from
ata_sff_wait_ready() into ata_wait_ready().  ata_wait_ready() takes
@check_ready function pointer and uses it to poll for readiness.  This
allows non-SFF controllers to use ata_wait_ready() to wait for link
readiness.

This patch also implements ata_wait_after_reset() - generic version of
ata_sff_wait_after_reset() - using ata_wait_ready().

ata_sff_wait_ready() is reimplemented using ata_wait_ready() and
ata_sff_check_ready().  Functionality remains the same.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c | 98 +++++++++++++++++++++++++++++++++++++++++++++++
 drivers/ata/libata-sff.c  | 62 ++++++------------------------
 drivers/ata/libata.h      |  2 +
 include/linux/libata.h    |  2 +
 4 files changed, 114 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index eaead76c9443..3bad6f189190 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3310,6 +3310,103 @@ int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 	return rc;
 }
 
+/**
+ *	ata_wait_ready - wait for link to become ready
+ *	@link: link to be waited on
+ *	@deadline: deadline jiffies for the operation
+ *	@check_ready: callback to check link readiness
+ *
+ *	Wait for @link to become ready.  @check_ready should return
+ *	positive number if @link is ready, 0 if it isn't, -ENODEV if
+ *	link doesn't seem to be occupied, other errno for other error
+ *	conditions.
+ *
+ *	Transient -ENODEV conditions are allowed for
+ *	ATA_TMOUT_FF_WAIT.
+ *
+ *	LOCKING:
+ *	EH context.
+ *
+ *	RETURNS:
+ *	0 if @linke is ready before @deadline; otherwise, -errno.
+ */
+int ata_wait_ready(struct ata_link *link, unsigned long deadline,
+		   int (*check_ready)(struct ata_link *link))
+{
+	unsigned long start = jiffies;
+	unsigned long nodev_deadline = start + ATA_TMOUT_FF_WAIT;
+	int warned = 0;
+
+	if (time_after(nodev_deadline, deadline))
+		nodev_deadline = deadline;
+
+	while (1) {
+		unsigned long now = jiffies;
+		int ready, tmp;
+
+		ready = tmp = check_ready(link);
+		if (ready > 0)
+			return 0;
+
+		/* -ENODEV could be transient.  Ignore -ENODEV if link
+		 * is online.  Also, some SATA devices take a long
+		 * time to clear 0xff after reset.  For example,
+		 * HHD424020F7SV00 iVDR needs >= 800ms while Quantum
+		 * GoVault needs even more than that.  Wait for
+		 * ATA_TMOUT_FF_WAIT on -ENODEV if link isn't offline.
+		 *
+		 * Note that some PATA controllers (pata_ali) explode
+		 * if status register is read more than once when
+		 * there's no device attached.
+		 */
+		if (ready == -ENODEV) {
+			if (ata_link_online(link))
+				ready = 0;
+			else if ((link->ap->flags & ATA_FLAG_SATA) &&
+				 !ata_link_offline(link) &&
+				 time_before(now, nodev_deadline))
+				ready = 0;
+		}
+
+		if (ready)
+			return ready;
+		if (time_after(now, deadline))
+			return -EBUSY;
+
+		if (!warned && time_after(now, start + 5 * HZ) &&
+		    (deadline - now > 3 * HZ)) {
+			ata_link_printk(link, KERN_WARNING,
+				"link is slow to respond, please be patient "
+				"(ready=%d)\n", tmp);
+			warned = 1;
+		}
+
+		msleep(50);
+	}
+}
+
+/**
+ *	ata_wait_after_reset - wait for link to become ready after reset
+ *	@link: link to be waited on
+ *	@deadline: deadline jiffies for the operation
+ *	@check_ready: callback to check link readiness
+ *
+ *	Wait for @link to become ready after reset.
+ *
+ *	LOCKING:
+ *	EH context.
+ *
+ *	RETURNS:
+ *	0 if @linke is ready before @deadline; otherwise, -errno.
+ */
+extern int ata_wait_after_reset(struct ata_link *link, unsigned long deadline,
+				int (*check_ready)(struct ata_link *link))
+{
+	msleep(ATA_WAIT_AFTER_RESET_MSECS);
+
+	return ata_wait_ready(link, deadline, check_ready);
+}
+
 /**
  *	sata_link_debounce - debounce SATA phy status
  *	@link: ATA link to debounce SATA phy status for
@@ -6075,6 +6172,7 @@ EXPORT_SYMBOL_GPL(ata_noop_qc_prep);
 EXPORT_SYMBOL_GPL(ata_port_probe);
 EXPORT_SYMBOL_GPL(ata_dev_disable);
 EXPORT_SYMBOL_GPL(sata_set_spd);
+EXPORT_SYMBOL_GPL(ata_wait_after_reset);
 EXPORT_SYMBOL_GPL(sata_link_debounce);
 EXPORT_SYMBOL_GPL(sata_link_resume);
 EXPORT_SYMBOL_GPL(ata_std_prereset);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 6e8de3c1595e..78912c5011ad 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -308,6 +308,17 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 	return 0;
 }
 
+static int ata_sff_check_ready(struct ata_link *link)
+{
+	u8 status = link->ap->ops->sff_check_status(link->ap);
+
+	if (!(status & ATA_BUSY))
+		return 1;
+	if (status == 0xff)
+		return -ENODEV;
+	return 0;
+}
+
 /**
  *	ata_sff_wait_ready - sleep until BSY clears, or timeout
  *	@link: SFF link to wait ready status for
@@ -324,56 +335,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
  */
 int ata_sff_wait_ready(struct ata_link *link, unsigned long deadline)
 {
-	struct ata_port *ap = link->ap;
-	unsigned long start = jiffies;
-	unsigned long nodev_deadline = start + ATA_TMOUT_FF_WAIT;
-	int warned = 0;
-
-	if (time_after(nodev_deadline, deadline))
-		nodev_deadline = deadline;
-
-	while (1) {
-		u8 status = ap->ops->sff_check_status(ap);
-		unsigned long now = jiffies;
-
-		if (!(status & ATA_BUSY))
-			return 0;
-
-		/* No device status could be transient.  Ignore it if
-		 * link is online.  Also, some SATA devices take a
-		 * long time to clear 0xff after reset.  For example,
-		 * HHD424020F7SV00 iVDR needs >= 800ms while Quantum
-		 * GoVault needs even more than that.  Wait for
-		 * ATA_TMOUT_FF_WAIT on -ENODEV if link isn't offline.
-		 *
-		 * Note that some PATA controllers (pata_ali) explode
-		 * if status register is read more than once when
-		 * there's no device attached.
-		 */
-		if (status == 0xff) {
-			if (ata_link_online(link))
-				status = ATA_BUSY;
-			else if ((link->ap->flags & ATA_FLAG_SATA) &&
-				 !ata_link_offline(link) &&
-				 time_before(now, nodev_deadline))
-				status = ATA_BUSY;
-			if (status == 0xff)
-				return -ENODEV;
-		}
-
-		if (time_after(now, deadline))
-			return -EBUSY;
-
-		if (!warned && time_after(now, start + 5 * HZ) &&
-		    (deadline - now > 3 * HZ)) {
-			ata_link_printk(link, KERN_WARNING,
-				"link is slow to respond, please be patient "
-				"(Status 0x%x)\n", status);
-			warned = 1;
-		}
-
-		msleep(50);
-	}
+	return ata_wait_ready(link, deadline, ata_sff_check_ready);
 }
 
 /**
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 11b5f67a19d4..08af43e2c081 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -81,6 +81,8 @@ extern unsigned ata_exec_internal_sg(struct ata_device *dev,
 				     int dma_dir, struct scatterlist *sg,
 				     unsigned int n_elem, unsigned long timeout);
 extern unsigned int ata_do_simple_cmd(struct ata_device *dev, u8 cmd);
+extern int ata_wait_ready(struct ata_link *link, unsigned long deadline,
+			  int (*check_ready)(struct ata_link *link));
 extern int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class,
 			   unsigned int flags, u16 *id);
 extern int ata_dev_reread_id(struct ata_device *dev, unsigned int readid_flags);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index da5560244787..4bbf2524e473 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -837,6 +837,8 @@ extern void sata_print_link_status(struct ata_link *link);
 extern void ata_port_probe(struct ata_port *);
 extern int sata_set_spd(struct ata_link *link);
 extern int ata_std_prereset(struct ata_link *link, unsigned long deadline);
+extern int ata_wait_after_reset(struct ata_link *link, unsigned long deadline,
+				int (*check_ready)(struct ata_link *link));
 extern int sata_link_debounce(struct ata_link *link,
 			const unsigned long *params, unsigned long deadline);
 extern int sata_link_resume(struct ata_link *link, const unsigned long *params,
-- 
cgit v1.2.3


From 9dadd45b24145d6aee2fabb28d7aef972301892b Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:19 +0900
Subject: libata: move generic hardreset code from sata_sff_hardreset() to
 sata_link_hardreset()

sata_sff_hardreset() contains link readiness wait logic which isn't
SFF specific.  Move that part into sata_link_hardreset(), which now
takes two more parameters - @online and @check_ready.  Both are
optional.  The former is out parameter for link onlineness after
reset.  The latter is used to wait for link readiness after hardreset.

Users of sata_link_hardreset() is updated to use new funtionality and
ahci_hardreset() is updated to use sata_link_hardreset() instead of
sata_sff_hardreset().  This doesn't really cause any behavior change.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c        | 35 +++++++++++++++++++--------------
 drivers/ata/ata_piix.c    |  2 +-
 drivers/ata/libata-core.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++-
 drivers/ata/libata-pmp.c  |  7 ++++---
 drivers/ata/libata-sff.c  | 49 +++++++++-------------------------------------
 include/linux/libata.h    | 10 +++++++++-
 6 files changed, 92 insertions(+), 61 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 7e251a2cbda5..0f553aaa6f79 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1343,10 +1343,12 @@ static int ahci_softreset(struct ata_link *link, unsigned int *class,
 static int ahci_hardreset(struct ata_link *link, unsigned int *class,
 			  unsigned long deadline)
 {
+	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
 	struct ata_port *ap = link->ap;
 	struct ahci_port_priv *pp = ap->private_data;
 	u8 *d2h_fis = pp->rx_fis + RX_FIS_D2H_REG;
 	struct ata_taskfile tf;
+	bool online;
 	int rc;
 
 	DPRINTK("ENTER\n");
@@ -1358,14 +1360,14 @@ static int ahci_hardreset(struct ata_link *link, unsigned int *class,
 	tf.command = 0x80;
 	ata_tf_to_fis(&tf, 0, 0, d2h_fis);
 
-	rc = sata_sff_hardreset(link, class, deadline);
+	rc = sata_link_hardreset(link, timing, deadline, &online,
+				 ahci_check_ready);
 
 	ahci_start_engine(ap);
 
-	if (rc == 0 && ata_link_online(link))
+	*class = ATA_DEV_NONE;
+	if (online)
 		*class = ahci_dev_classify(ap);
-	if (rc != -EAGAIN && *class == ATA_DEV_UNKNOWN)
-		*class = ATA_DEV_NONE;
 
 	DPRINTK("EXIT, rc=%d, class=%u\n", rc, *class);
 	return rc;
@@ -1376,6 +1378,7 @@ static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class,
 {
 	struct ata_port *ap = link->ap;
 	u32 serror;
+	bool online;
 	int rc;
 
 	DPRINTK("ENTER\n");
@@ -1383,7 +1386,7 @@ static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class,
 	ahci_stop_engine(ap);
 
 	rc = sata_link_hardreset(link, sata_ehc_deb_timing(&link->eh_context),
-				 deadline);
+				 deadline, &online, NULL);
 
 	/* vt8251 needs SError cleared for the port to operate */
 	ahci_scr_read(ap, SCR_ERROR, &serror);
@@ -1396,7 +1399,8 @@ static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class,
 	/* vt8251 doesn't clear BSY on signature FIS reception,
 	 * request follow-up softreset.
 	 */
-	return rc ?: -EAGAIN;
+	*class = ATA_DEV_NONE;
+	return online ? -EAGAIN : rc;
 }
 
 static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
@@ -1406,6 +1410,7 @@ static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
 	struct ahci_port_priv *pp = ap->private_data;
 	u8 *d2h_fis = pp->rx_fis + RX_FIS_D2H_REG;
 	struct ata_taskfile tf;
+	bool online;
 	int rc;
 
 	ahci_stop_engine(ap);
@@ -1416,13 +1421,10 @@ static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
 	ata_tf_to_fis(&tf, 0, 0, d2h_fis);
 
 	rc = sata_link_hardreset(link, sata_ehc_deb_timing(&link->eh_context),
-				 deadline);
+				 deadline, &online, NULL);
 
 	ahci_start_engine(ap);
 
-	if (rc || ata_link_offline(link))
-		return rc;
-
 	/* The pseudo configuration device on SIMG4726 attached to
 	 * ASUS P5W-DH Deluxe doesn't send signature FIS after
 	 * hardreset if no device is attached to the first downstream
@@ -1436,11 +1438,14 @@ static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
 	 * have to be reset again.  For most cases, this should
 	 * suffice while making probing snappish enough.
 	 */
-	rc = ata_wait_after_reset(link, jiffies + 2 * HZ, ahci_check_ready);
-	if (rc)
-		ahci_kick_engine(ap, 0);
-
-	return 0;
+	if (online) {
+		rc = ata_wait_after_reset(link, jiffies + 2 * HZ,
+					  ahci_check_ready);
+		if (rc)
+			ahci_kick_engine(ap, 0);
+	}
+	*class = ATA_DEV_NONE;
+	return rc;
 }
 
 static void ahci_postreset(struct ata_link *link, unsigned int *class)
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 7ab76a413cdf..f59a55bfade4 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -1022,7 +1022,7 @@ static int piix_sidpr_hardreset(struct ata_link *link, unsigned int *class,
 	int rc;
 
 	/* do hardreset */
-	rc = sata_link_hardreset(link, timing, deadline);
+	rc = sata_link_hardreset(link, timing, deadline, NULL, NULL);
 	if (rc) {
 		ata_link_printk(link, KERN_ERR,
 				"COMRESET failed (errno=%d)\n", rc);
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 3bad6f189190..b607292b6480 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3557,8 +3557,18 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
  *	@link: link to reset
  *	@timing: timing parameters { interval, duratinon, timeout } in msec
  *	@deadline: deadline jiffies for the operation
+ *	@online: optional out parameter indicating link onlineness
+ *	@check_ready: optional callback to check link readiness
  *
  *	SATA phy-reset @link using DET bits of SControl register.
+ *	After hardreset, link readiness is waited upon using
+ *	ata_wait_ready() if @check_ready is specified.  LLDs are
+ *	allowed to not specify @check_ready and wait itself after this
+ *	function returns.  Device classification is LLD's
+ *	responsibility.
+ *
+ *	*@online is set to one iff reset succeeded and @link is online
+ *	after reset.
  *
  *	LOCKING:
  *	Kernel thread context (may sleep)
@@ -3567,13 +3577,17 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
  *	0 on success, -errno otherwise.
  */
 int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
-			unsigned long deadline)
+			unsigned long deadline,
+			bool *online, int (*check_ready)(struct ata_link *))
 {
 	u32 scontrol;
 	int rc;
 
 	DPRINTK("ENTER\n");
 
+	if (online)
+		*online = false;
+
 	if (sata_set_spd_needed(link)) {
 		/* SATA spec says nothing about how to reconfigure
 		 * spd.  To be on the safe side, turn off phy during
@@ -3607,7 +3621,41 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 
 	/* bring link back */
 	rc = sata_link_resume(link, timing, deadline);
+	if (rc)
+		goto out;
+	/* if link is offline nothing more to do */
+	if (ata_link_offline(link))
+		goto out;
+
+	/* Link is online.  From this point, -ENODEV too is an error. */
+	if (online)
+		*online = true;
+
+	if ((link->ap->flags & ATA_FLAG_PMP) && ata_is_host_link(link)) {
+		/* If PMP is supported, we have to do follow-up SRST.
+		 * Some PMPs don't send D2H Reg FIS after hardreset if
+		 * the first port is empty.  Wait only for
+		 * ATA_TMOUT_PMP_SRST_WAIT.
+		 */
+		if (check_ready) {
+			unsigned long pmp_deadline;
+
+			pmp_deadline = jiffies + ATA_TMOUT_PMP_SRST_WAIT;
+			if (time_after(pmp_deadline, deadline))
+				pmp_deadline = deadline;
+			ata_wait_ready(link, pmp_deadline, check_ready);
+		}
+		rc = -EAGAIN;
+		goto out;
+	}
+
+	rc = 0;
+	if (check_ready)
+		rc = ata_wait_ready(link, deadline, check_ready);
  out:
+	if (rc && rc != -EAGAIN)
+		ata_link_printk(link, KERN_ERR,
+				"COMRESET failed (errno=%d)\n", rc);
 	DPRINTK("EXIT, rc=%d\n", rc);
 	return rc;
 }
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index a7cb1498c9b2..7f1a87f01ab2 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -239,13 +239,14 @@ int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
 			   unsigned long deadline)
 {
 	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
+	bool online;
 	u32 tmp;
 	int rc;
 
 	DPRINTK("ENTER\n");
 
 	/* do hardreset */
-	rc = sata_link_hardreset(link, timing, deadline);
+	rc = sata_link_hardreset(link, timing, deadline, &online, NULL);
 	if (rc) {
 		ata_link_printk(link, KERN_ERR,
 				"COMRESET failed (errno=%d)\n", rc);
@@ -261,7 +262,7 @@ int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
 	}
 
 	/* if device is present, follow up with srst to wait for !BSY */
-	if (ata_link_online(link))
+	if (online)
 		rc = -EAGAIN;
  out:
 	/* if SCR isn't accessible, we need to reset the PMP */
@@ -916,7 +917,7 @@ static int sata_pmp_eh_handle_disabled_links(struct ata_port *ap)
 		 * SError.N working.
 		 */
 		sata_link_hardreset(link, sata_deb_timing_normal,
-				    jiffies + ATA_TMOUT_INTERNAL_QUICK);
+				jiffies + ATA_TMOUT_INTERNAL_QUICK, NULL, NULL);
 
 		/* unconditionally clear SError.N */
 		rc = sata_scr_write(link, SCR_ERROR, SERR_PHYRDY_CHG);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 78912c5011ad..0b97e84d3af6 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -1921,50 +1921,19 @@ int ata_sff_softreset(struct ata_link *link, unsigned int *classes,
 int sata_sff_hardreset(struct ata_link *link, unsigned int *class,
 		       unsigned long deadline)
 {
-	struct ata_port *ap = link->ap;
-	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
+	struct ata_eh_context *ehc = &link->eh_context;
+	const unsigned long *timing = sata_ehc_deb_timing(ehc);
+	bool online;
 	int rc;
 
-	DPRINTK("ENTER\n");
-
-	/* do hardreset */
-	rc = sata_link_hardreset(link, timing, deadline);
-	if (rc) {
-		ata_link_printk(link, KERN_ERR,
-				"COMRESET failed (errno=%d)\n", rc);
-		return rc;
-	}
-
-	/* TODO: phy layer with polling, timeouts, etc. */
-	if (ata_link_offline(link)) {
-		*class = ATA_DEV_NONE;
-		DPRINTK("EXIT, link offline\n");
-		return 0;
-	}
-
-	/* If PMP is supported, we have to do follow-up SRST.  Note
-	 * that some PMPs don't send D2H Reg FIS after hardreset at
-	 * all if the first port is empty.  Wait for it just for a
-	 * second and request follow-up SRST.
-	 */
-	if (ap->flags & ATA_FLAG_PMP) {
-		ata_sff_wait_after_reset(link, 1, jiffies + HZ);
-		return -EAGAIN;
-	}
-
-	/* wait for the link to become online */
-	rc = ata_sff_wait_after_reset(link, 1, deadline);
-	/* link occupied, -ENODEV too is an error */
-	if (rc) {
-		ata_link_printk(link, KERN_ERR,
-				"COMRESET failed (errno=%d)\n", rc);
-		return rc;
-	}
-
-	*class = ata_sff_dev_classify(link->device, 1, NULL);
+	rc = sata_link_hardreset(link, timing, deadline, &online,
+				 ata_sff_check_ready);
+	*class = ATA_DEV_NONE;
+	if (online)
+		*class = ata_sff_dev_classify(link->device, 1, NULL);
 
 	DPRINTK("EXIT, class=%u\n", *class);
-	return 0;
+	return rc;
 }
 
 /**
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 4bbf2524e473..d9ebce2bf5e7 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -261,6 +261,13 @@ enum {
 	 */
 	ATA_WAIT_AFTER_RESET_MSECS = 150,
 
+	/* If PMP is supported, we have to do follow-up SRST.  As some
+	 * PMPs don't send D2H Reg FIS after hardreset, LLDs are
+	 * advised to wait only for the following duration before
+	 * doing SRST.
+	 */
+	ATA_TMOUT_PMP_SRST_WAIT	= 1 * HZ,
+
 	/* ATA bus states */
 	BUS_UNKNOWN		= 0,
 	BUS_DMA			= 1,
@@ -844,7 +851,8 @@ extern int sata_link_debounce(struct ata_link *link,
 extern int sata_link_resume(struct ata_link *link, const unsigned long *params,
 			    unsigned long deadline);
 extern int sata_link_hardreset(struct ata_link *link,
-			const unsigned long *timing, unsigned long deadline);
+			const unsigned long *timing, unsigned long deadline,
+			bool *online, int (*check_ready)(struct ata_link *));
 extern void ata_std_postreset(struct ata_link *link, unsigned int *classes);
 extern void ata_port_disable(struct ata_port *);
 
-- 
cgit v1.2.3


From 57c9efdfb3cee5d4564fcb5f70555e2edb1bc52a Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:19 +0900
Subject: libata: implement and use sata_std_hardreset()

Implement sata_std_hardreset(), which simply wraps around
sata_link_hardreset().  sata_std_hardreset() becomes new standard
hardreset method for sata_port_ops and sata_sff_hardreset() moves from
ata_base_port_ops to ata_sff_port_ops, which is where it really
belongs.

ata_is_builtin_hardreset() is added so that both
ata_std_error_handler() and ata_sff_error_handler() skip both builtin
hardresets if SCR isn't accessible.

piix_sidpr_hardreset() in ata_piix.c is identical to
sata_std_hardreset() in functionality and got replaced with the
standard function.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ata_piix.c    | 27 +--------------------------
 drivers/ata/libata-core.c | 30 +++++++++++++++++++++++++++++-
 drivers/ata/libata-eh.c   |  7 ++-----
 drivers/ata/libata-sff.c  |  9 ++++-----
 drivers/ata/libata.h      |  9 +++++++++
 include/linux/libata.h    |  2 ++
 6 files changed, 47 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index f59a55bfade4..b7c38eeb498f 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -165,8 +165,6 @@ static void piix_set_dmamode(struct ata_port *ap, struct ata_device *adev);
 static void ich_set_dmamode(struct ata_port *ap, struct ata_device *adev);
 static int ich_pata_cable_detect(struct ata_port *ap);
 static u8 piix_vmw_bmdma_status(struct ata_port *ap);
-static int piix_sidpr_hardreset(struct ata_link *link, unsigned int *class,
-				unsigned long deadline);
 static int piix_sidpr_scr_read(struct ata_port *ap, unsigned int reg, u32 *val);
 static int piix_sidpr_scr_write(struct ata_port *ap, unsigned int reg, u32 val);
 #ifdef CONFIG_PM
@@ -319,7 +317,7 @@ static struct ata_port_operations piix_sata_ops = {
 
 static struct ata_port_operations piix_sidpr_sata_ops = {
 	.inherits		= &piix_sata_ops,
-	.hardreset		= piix_sidpr_hardreset,
+	.hardreset		= sata_std_hardreset,
 	.scr_read		= piix_sidpr_scr_read,
 	.scr_write		= piix_sidpr_scr_write,
 };
@@ -1015,29 +1013,6 @@ static int piix_sidpr_scr_write(struct ata_port *ap, unsigned int reg, u32 val)
 	return 0;
 }
 
-static int piix_sidpr_hardreset(struct ata_link *link, unsigned int *class,
-				unsigned long deadline)
-{
-	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
-	int rc;
-
-	/* do hardreset */
-	rc = sata_link_hardreset(link, timing, deadline, NULL, NULL);
-	if (rc) {
-		ata_link_printk(link, KERN_ERR,
-				"COMRESET failed (errno=%d)\n", rc);
-		return rc;
-	}
-
-	/* TODO: phy layer with polling, timeouts, etc. */
-	if (ata_link_offline(link)) {
-		*class = ATA_DEV_NONE;
-		return 0;
-	}
-
-	return -EAGAIN;
-}
-
 #ifdef CONFIG_PM
 static int piix_broken_suspend(void)
 {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index b607292b6480..c4fd4afbf349 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -75,7 +75,6 @@ const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
 const struct ata_port_operations ata_base_port_ops = {
 	.prereset		= ata_std_prereset,
-	.hardreset		= sata_sff_hardreset,
 	.postreset		= ata_std_postreset,
 	.error_handler		= ata_std_error_handler,
 };
@@ -84,6 +83,7 @@ const struct ata_port_operations sata_port_ops = {
 	.inherits		= &ata_base_port_ops,
 
 	.qc_defer		= ata_std_qc_defer,
+	.hardreset		= sata_std_hardreset,
 	.sff_dev_select		= ata_noop_dev_select,
 };
 
@@ -3660,6 +3660,33 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 	return rc;
 }
 
+/**
+ *	sata_std_hardreset - COMRESET w/o waiting or classification
+ *	@link: link to reset
+ *	@class: resulting class of attached device
+ *	@deadline: deadline jiffies for the operation
+ *
+ *	Standard SATA COMRESET w/o waiting or classification.
+ *
+ *	LOCKING:
+ *	Kernel thread context (may sleep)
+ *
+ *	RETURNS:
+ *	0 if link offline, -EAGAIN if link online, -errno on errors.
+ */
+int sata_std_hardreset(struct ata_link *link, unsigned int *class,
+		       unsigned long deadline)
+{
+	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
+	bool online;
+	int rc;
+
+	/* do hardreset */
+	rc = sata_link_hardreset(link, timing, deadline, &online, NULL);
+	*class = ATA_DEV_NONE;
+	return online ? -EAGAIN : rc;
+}
+
 /**
  *	ata_std_postreset - standard postreset callback
  *	@link: the target ata_link
@@ -6225,6 +6252,7 @@ EXPORT_SYMBOL_GPL(sata_link_debounce);
 EXPORT_SYMBOL_GPL(sata_link_resume);
 EXPORT_SYMBOL_GPL(ata_std_prereset);
 EXPORT_SYMBOL_GPL(sata_link_hardreset);
+EXPORT_SYMBOL_GPL(sata_std_hardreset);
 EXPORT_SYMBOL_GPL(ata_std_postreset);
 EXPORT_SYMBOL_GPL(ata_dev_classify);
 EXPORT_SYMBOL_GPL(ata_dev_pair);
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index f4f9c2783821..21687bbd9a70 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2857,11 +2857,8 @@ void ata_std_error_handler(struct ata_port *ap)
 	struct ata_port_operations *ops = ap->ops;
 	ata_reset_fn_t hardreset = ops->hardreset;
 
-	/* sata_std_hardreset is inherited to all drivers from
-	 * ata_base_port_ops.  Ignore it if SCR access is not
-	 * available.
-	 */
-	if (hardreset == sata_sff_hardreset && !sata_scr_valid(&ap->link))
+	/* ignore built-in hardreset if SCR access is not available */
+	if (ata_is_builtin_hardreset(hardreset) && !sata_scr_valid(&ap->link))
 		hardreset = NULL;
 
 	ata_do_eh(ap, ops->prereset, ops->softreset, hardreset, ops->postreset);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 0b97e84d3af6..f464ca1fa261 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -49,6 +49,7 @@ const struct ata_port_operations ata_sff_port_ops = {
 	.thaw			= ata_sff_thaw,
 	.prereset		= ata_sff_prereset,
 	.softreset		= ata_sff_softreset,
+	.hardreset		= sata_sff_hardreset,
 	.postreset		= ata_sff_postreset,
 	.error_handler		= ata_sff_error_handler,
 	.post_internal_cmd	= ata_sff_post_internal_cmd,
@@ -2031,14 +2032,12 @@ void ata_sff_error_handler(struct ata_port *ap)
 
 	/* PIO and DMA engines have been stopped, perform recovery */
 
-	/* ata_sff_softreset and sata_sff_hardreset are inherited to
-	 * all SFF drivers from ata_sff_port_ops.  Ignore softreset if
-	 * ctl isn't accessible.  Ignore hardreset if SCR access isn't
-	 * available.
+	/* Ignore ata_sff_softreset if ctl isn't accessible and
+	 * built-in hardresets if SCR access isn't available.
 	 */
 	if (softreset == ata_sff_softreset && !ap->ioaddr.ctl_addr)
 		softreset = NULL;
-	if (hardreset == sata_sff_hardreset && !sata_scr_valid(&ap->link))
+	if (ata_is_builtin_hardreset(hardreset) && !sata_scr_valid(&ap->link))
 		hardreset = NULL;
 
 	ata_do_eh(ap, ap->ops->prereset, softreset, hardreset,
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 08af43e2c081..87f54a1db3b0 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -38,6 +38,15 @@ struct ata_scsi_args {
 	void			(*done)(struct scsi_cmnd *);
 };
 
+static inline int ata_is_builtin_hardreset(ata_reset_fn_t reset)
+{
+	if (reset == sata_std_hardreset)
+		return 1;
+	if (reset == sata_sff_hardreset)
+		return 1;
+	return 0;
+}
+
 /* libata-core.c */
 enum {
 	/* flags for ata_dev_read_id() */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index d9ebce2bf5e7..c060cd3cba66 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -853,6 +853,8 @@ extern int sata_link_resume(struct ata_link *link, const unsigned long *params,
 extern int sata_link_hardreset(struct ata_link *link,
 			const unsigned long *timing, unsigned long deadline,
 			bool *online, int (*check_ready)(struct ata_link *));
+extern int sata_std_hardreset(struct ata_link *link, unsigned int *class,
+			      unsigned long deadline);
 extern void ata_std_postreset(struct ata_link *link, unsigned int *classes);
 extern void ata_port_disable(struct ata_port *);
 
-- 
cgit v1.2.3


From ac371987a81c61c2efbd6931245cdcaf43baad89 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:19 +0900
Subject: libata: clear SError after link resume

SError used to be cleared in ->postreset.  This has small hotplug race
condition.  If a device is plugged in after reset is complete but
postreset hasn't run yet, its hotplug event gets lost when SError is
cleared.  This patch makes sata_link_resume() clear SError.  This
kills the race condition and makes a lot of sense as some PMP and host
PHYs don't work properly without SError cleared.

This change makes sata_pmp_std_{pre|post}_reset()'s unnecessary as
they become identical to ata_std counterparts.  It also simplifies
sata_pmp_hardreset() and ahci_vt8251_hardreset().

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c        |  5 ---
 drivers/ata/libata-core.c | 35 +++++++++++-------
 drivers/ata/libata-pmp.c  | 93 +----------------------------------------------
 include/linux/libata.h    |  2 -
 4 files changed, 23 insertions(+), 112 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 0f553aaa6f79..a69bcca4eb1b 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1377,7 +1377,6 @@ static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class,
 				 unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
-	u32 serror;
 	bool online;
 	int rc;
 
@@ -1388,10 +1387,6 @@ static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class,
 	rc = sata_link_hardreset(link, sata_ehc_deb_timing(&link->eh_context),
 				 deadline, &online, NULL);
 
-	/* vt8251 needs SError cleared for the port to operate */
-	ahci_scr_read(ap, SCR_ERROR, &serror);
-	ahci_scr_write(ap, SCR_ERROR, serror);
-
 	ahci_start_engine(ap);
 
 	DPRINTK("EXIT, rc=%d, class=%u\n", rc, *class);
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index c4fd4afbf349..e00b620f161a 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -90,9 +90,9 @@ const struct ata_port_operations sata_port_ops = {
 const struct ata_port_operations sata_pmp_port_ops = {
 	.inherits		= &sata_port_ops,
 
-	.pmp_prereset		= sata_pmp_std_prereset,
+	.pmp_prereset		= ata_std_prereset,
 	.pmp_hardreset		= sata_pmp_std_hardreset,
-	.pmp_postreset		= sata_pmp_std_postreset,
+	.pmp_postreset		= ata_std_postreset,
 	.error_handler		= sata_pmp_error_handler,
 };
 
@@ -3493,7 +3493,7 @@ int sata_link_debounce(struct ata_link *link, const unsigned long *params,
 int sata_link_resume(struct ata_link *link, const unsigned long *params,
 		     unsigned long deadline)
 {
-	u32 scontrol;
+	u32 scontrol, serror;
 	int rc;
 
 	if ((rc = sata_scr_read(link, SCR_CONTROL, &scontrol)))
@@ -3509,7 +3509,25 @@ int sata_link_resume(struct ata_link *link, const unsigned long *params,
 	 */
 	msleep(200);
 
-	return sata_link_debounce(link, params, deadline);
+	if ((rc = sata_link_debounce(link, params, deadline)))
+		return rc;
+
+	/* Clear SError.  PMP and some host PHYs require this to
+	 * operate and clearing should be done before checking PHY
+	 * online status to avoid race condition (hotplugging between
+	 * link resume and status check).
+	 */
+	if (!(rc = sata_scr_read(link, SCR_ERROR, &serror)))
+		rc = sata_scr_write(link, SCR_ERROR, serror);
+	if (rc == 0 || rc == -EINVAL) {
+		unsigned long flags;
+
+		spin_lock_irqsave(link->ap->lock, flags);
+		link->eh_info.serror = 0;
+		spin_unlock_irqrestore(link->ap->lock, flags);
+		rc = 0;
+	}
+	return rc;
 }
 
 /**
@@ -3701,18 +3719,11 @@ int sata_std_hardreset(struct ata_link *link, unsigned int *class,
  */
 void ata_std_postreset(struct ata_link *link, unsigned int *classes)
 {
-	u32 serror;
-
 	DPRINTK("ENTER\n");
 
 	/* print link status */
 	sata_print_link_status(link);
 
-	/* clear SError */
-	if (sata_scr_read(link, SCR_ERROR, &serror) == 0)
-		sata_scr_write(link, SCR_ERROR, serror);
-	link->eh_info.serror = 0;
-
 	DPRINTK("EXIT\n");
 }
 
@@ -6296,9 +6307,7 @@ EXPORT_SYMBOL_GPL(ata_pci_device_resume);
 #endif /* CONFIG_PCI */
 
 EXPORT_SYMBOL_GPL(sata_pmp_qc_defer_cmd_switch);
-EXPORT_SYMBOL_GPL(sata_pmp_std_prereset);
 EXPORT_SYMBOL_GPL(sata_pmp_std_hardreset);
-EXPORT_SYMBOL_GPL(sata_pmp_std_postreset);
 EXPORT_SYMBOL_GPL(sata_pmp_error_handler);
 
 EXPORT_SYMBOL_GPL(__ata_ehi_push_desc);
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 7f1a87f01ab2..2f8a9577c26d 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -175,49 +175,6 @@ int sata_pmp_scr_write(struct ata_link *link, int reg, u32 val)
 	return 0;
 }
 
-/**
- *	sata_pmp_std_prereset - prepare PMP link for reset
- *	@link: link to be reset
- *	@deadline: deadline jiffies for the operation
- *
- *	@link is about to be reset.  Initialize it.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep)
- *
- *	RETURNS:
- *	0 on success, -errno otherwise.
- */
-int sata_pmp_std_prereset(struct ata_link *link, unsigned long deadline)
-{
-	struct ata_eh_context *ehc = &link->eh_context;
-	const unsigned long *timing = sata_ehc_deb_timing(ehc);
-	int rc;
-
-	/* if we're about to do hardreset, nothing more to do */
-	if (ehc->i.action & ATA_EH_HARDRESET)
-		return 0;
-
-	/* resume link */
-	rc = sata_link_resume(link, timing, deadline);
-	if (rc) {
-		/* phy resume failed */
-		ata_link_printk(link, KERN_WARNING, "failed to resume link "
-				"for reset (errno=%d)\n", rc);
-		return rc;
-	}
-
-	/* clear SError bits including .X which blocks the port when set */
-	rc = sata_scr_write(link, SCR_ERROR, 0xffffffff);
-	if (rc) {
-		ata_link_printk(link, KERN_ERR,
-				"failed to clear SError (errno=%d)\n", rc);
-		return rc;
-	}
-
-	return 0;
-}
-
 /**
  *	sata_pmp_std_hardreset - standard hardreset method for PMP link
  *	@link: link to be reset
@@ -238,33 +195,13 @@ int sata_pmp_std_prereset(struct ata_link *link, unsigned long deadline)
 int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
 			   unsigned long deadline)
 {
-	const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context);
-	bool online;
 	u32 tmp;
 	int rc;
 
 	DPRINTK("ENTER\n");
 
-	/* do hardreset */
-	rc = sata_link_hardreset(link, timing, deadline, &online, NULL);
-	if (rc) {
-		ata_link_printk(link, KERN_ERR,
-				"COMRESET failed (errno=%d)\n", rc);
-		goto out;
-	}
-
-	/* clear SError bits including .X which blocks the port when set */
-	rc = sata_scr_write(link, SCR_ERROR, 0xffffffff);
-	if (rc) {
-		ata_link_printk(link, KERN_ERR, "failed to clear SError "
-				"during hardreset (errno=%d)\n", rc);
-		goto out;
-	}
+	rc = sata_std_hardreset(link, class, deadline);
 
-	/* if device is present, follow up with srst to wait for !BSY */
-	if (online)
-		rc = -EAGAIN;
- out:
 	/* if SCR isn't accessible, we need to reset the PMP */
 	if (rc && rc != -EAGAIN && sata_scr_read(link, SCR_STATUS, &tmp))
 		rc = -ERESTART;
@@ -273,34 +210,6 @@ int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
 	return rc;
 }
 
-/**
- *	ata_std_postreset - standard postreset method for PMP link
- *	@link: the target ata_link
- *	@classes: classes of attached devices
- *
- *	This function is invoked after a successful reset.  Note that
- *	the device might have been reset more than once using
- *	different reset methods before postreset is invoked.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep)
- */
-void sata_pmp_std_postreset(struct ata_link *link, unsigned int *class)
-{
-	u32 serror;
-
-	DPRINTK("ENTER\n");
-
-	/* clear SError */
-	if (sata_scr_read(link, SCR_ERROR, &serror) == 0)
-		sata_scr_write(link, SCR_ERROR, serror);
-
-	/* print link status */
-	sata_print_link_status(link);
-
-	DPRINTK("EXIT\n");
-}
-
 /**
  *	sata_pmp_read_gscr - read GSCR block of SATA PMP
  *	@dev: PMP device
diff --git a/include/linux/libata.h b/include/linux/libata.h
index c060cd3cba66..b9188371b12a 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1025,10 +1025,8 @@ static inline int ata_acpi_cbl_80wire(struct ata_port *ap,
  * PMP - drivers/ata/libata-pmp.c
  */
 extern int sata_pmp_qc_defer_cmd_switch(struct ata_queued_cmd *qc);
-extern int sata_pmp_std_prereset(struct ata_link *link, unsigned long deadline);
 extern int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
 				  unsigned long deadline);
-extern void sata_pmp_std_postreset(struct ata_link *link, unsigned int *class);
 extern void sata_pmp_error_handler(struct ata_port *ap);
 
 /*
-- 
cgit v1.2.3


From 5958e3025fd9d97429163e074d9cfa3848f51f28 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:20 +0900
Subject: libata: move PMP SCR access failure during reset to ata_eh_reset()

If PMP fan-out reset fails and SCR isn't accessible, PMP should be
reset.  This used to be tested by sata_pmp_std_hardreset() and
communicated to EH by -ERESTART.  However, this logic is generic and
doesn't really have much to do with specific hardreset implementation.

This patch moves SCR access failure detection logic to ata_eh_reset()
where it belongs.  As this makes sata_pmp_std_hardreset() identical to
sata_std_hardreset(), the function is killed and replaced with the
standard method.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c |  3 +--
 drivers/ata/libata-eh.c   |  5 +++++
 drivers/ata/libata-pmp.c  | 35 -----------------------------------
 drivers/ata/sata_sil24.c  |  2 +-
 include/linux/libata.h    |  2 --
 5 files changed, 7 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index e00b620f161a..2da579b46bdd 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -91,7 +91,7 @@ const struct ata_port_operations sata_pmp_port_ops = {
 	.inherits		= &sata_port_ops,
 
 	.pmp_prereset		= ata_std_prereset,
-	.pmp_hardreset		= sata_pmp_std_hardreset,
+	.pmp_hardreset		= sata_std_hardreset,
 	.pmp_postreset		= ata_std_postreset,
 	.error_handler		= sata_pmp_error_handler,
 };
@@ -6307,7 +6307,6 @@ EXPORT_SYMBOL_GPL(ata_pci_device_resume);
 #endif /* CONFIG_PCI */
 
 EXPORT_SYMBOL_GPL(sata_pmp_qc_defer_cmd_switch);
-EXPORT_SYMBOL_GPL(sata_pmp_std_hardreset);
 EXPORT_SYMBOL_GPL(sata_pmp_error_handler);
 
 EXPORT_SYMBOL_GPL(__ata_ehi_push_desc);
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 21687bbd9a70..d8c4a45dcf26 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2276,6 +2276,11 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	return rc;
 
  fail:
+	/* if SCR isn't accessible on a fan-out port, PMP needs to be reset */
+	if (!ata_is_host_link(link) &&
+	    sata_scr_read(link, SCR_STATUS, &sstatus))
+		rc = -ERESTART;
+
 	if (rc == -ERESTART || try >= max_tries)
 		goto out;
 
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 2f8a9577c26d..9c998611b644 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -175,41 +175,6 @@ int sata_pmp_scr_write(struct ata_link *link, int reg, u32 val)
 	return 0;
 }
 
-/**
- *	sata_pmp_std_hardreset - standard hardreset method for PMP link
- *	@link: link to be reset
- *	@class: resulting class of attached device
- *	@deadline: deadline jiffies for the operation
- *
- *	Hardreset PMP port @link.  Note that this function doesn't
- *	wait for BSY clearance.  There simply isn't a generic way to
- *	wait the event.  Instead, this function return -EAGAIN thus
- *	telling libata-EH to followup with softreset.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep)
- *
- *	RETURNS:
- *	0 on success, -errno otherwise.
- */
-int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
-			   unsigned long deadline)
-{
-	u32 tmp;
-	int rc;
-
-	DPRINTK("ENTER\n");
-
-	rc = sata_std_hardreset(link, class, deadline);
-
-	/* if SCR isn't accessible, we need to reset the PMP */
-	if (rc && rc != -EAGAIN && sata_scr_read(link, SCR_STATUS, &tmp))
-		rc = -ERESTART;
-
-	DPRINTK("EXIT, rc=%d\n", rc);
-	return rc;
-}
-
 /**
  *	sata_pmp_read_gscr - read GSCR block of SATA PMP
  *	@dev: PMP device
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index b83851f6e068..fc9d48cd8122 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -944,7 +944,7 @@ static int sil24_pmp_hardreset(struct ata_link *link, unsigned int *class,
 		return rc;
 	}
 
-	return sata_pmp_std_hardreset(link, class, deadline);
+	return sata_std_hardreset(link, class, deadline);
 }
 
 static void sil24_freeze(struct ata_port *ap)
diff --git a/include/linux/libata.h b/include/linux/libata.h
index b9188371b12a..2b5a0b77e179 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1025,8 +1025,6 @@ static inline int ata_acpi_cbl_80wire(struct ata_port *ap,
  * PMP - drivers/ata/libata-pmp.c
  */
 extern int sata_pmp_qc_defer_cmd_switch(struct ata_queued_cmd *qc);
-extern int sata_pmp_std_hardreset(struct ata_link *link, unsigned int *class,
-				  unsigned long deadline);
 extern void sata_pmp_error_handler(struct ata_port *ap);
 
 /*
-- 
cgit v1.2.3


From 22183bf569c8600ff414ac25f23134044e0ef453 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:20 +0900
Subject: libata: add qc_fill_rtf port operation

On command completion, ata_qc_complete() directly called ops->tf_read
to fill qc->result_tf.  This patch adds ops->qc_fill_rtf to replace
hardcoded ops->tf_read usage.

ata_sff_qc_fill_rtf() which uses ops->tf_read to fill result_tf is
implemented and set in ata_base_port_ops and other ops tables which
don't inherit from ata_base_port_ops, so this patch doesn't introduce
any behavior change.

ops->qc_fill_rtf() is similar to ops->sff_tf_read() but can only be
called when a command finishes.  As some non-SFF controllers don't
have TF registers defined unless they're associated with in-flight
commands, this limited operation makes life easier for those drivers
and help lifting SFF assumptions from libata core layer.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c     |  3 ++-
 drivers/ata/libata-sff.c      | 20 ++++++++++++++++++++
 drivers/ata/sata_sx4.c        |  1 +
 drivers/scsi/ipr.c            |  1 +
 drivers/scsi/libsas/sas_ata.c |  1 +
 include/linux/libata.h        |  2 ++
 6 files changed, 27 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 51876b93c1b7..3b822124e97e 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -74,6 +74,7 @@ const unsigned long sata_deb_timing_hotplug[]		= {  25,  500, 2000 };
 const unsigned long sata_deb_timing_long[]		= { 100, 2000, 5000 };
 
 const struct ata_port_operations ata_base_port_ops = {
+	.qc_fill_rtf		= ata_sff_qc_fill_rtf,
 	.prereset		= ata_std_prereset,
 	.postreset		= ata_std_postreset,
 	.error_handler		= ata_std_error_handler,
@@ -4562,7 +4563,7 @@ static void fill_result_tf(struct ata_queued_cmd *qc)
 	struct ata_port *ap = qc->ap;
 
 	qc->result_tf.flags = qc->tf.flags;
-	ap->ops->sff_tf_read(ap, &qc->result_tf);
+	ap->ops->qc_fill_rtf(qc);
 }
 
 static void ata_verify_xfer(struct ata_queued_cmd *qc)
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 5be8a6058dac..5ae813f54420 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -1407,6 +1407,25 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc)
 	return 0;
 }
 
+/**
+ *	ata_sff_qc_fill_rtf - fill result TF using ->sff_tf_read
+ *	@qc: qc to fill result TF for
+ *
+ *	@qc is finished and result TF needs to be filled.  Fill it
+ *	using ->sff_tf_read.
+ *
+ *	LOCKING:
+ *	spin_lock_irqsave(host lock)
+ *
+ *	RETURNS:
+ *	true indicating that result TF is successfully filled.
+ */
+bool ata_sff_qc_fill_rtf(struct ata_queued_cmd *qc)
+{
+	qc->ap->ops->sff_tf_read(qc->ap, &qc->result_tf);
+	return true;
+}
+
 /**
  *	ata_sff_host_intr - Handle host interrupt for given (port, task)
  *	@ap: Port on which interrupt arrived (possibly...)
@@ -2724,6 +2743,7 @@ EXPORT_SYMBOL_GPL(ata_sff_irq_on);
 EXPORT_SYMBOL_GPL(ata_sff_irq_clear);
 EXPORT_SYMBOL_GPL(ata_sff_hsm_move);
 EXPORT_SYMBOL_GPL(ata_sff_qc_issue);
+EXPORT_SYMBOL_GPL(ata_sff_qc_fill_rtf);
 EXPORT_SYMBOL_GPL(ata_sff_host_intr);
 EXPORT_SYMBOL_GPL(ata_sff_interrupt);
 EXPORT_SYMBOL_GPL(ata_sff_freeze);
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 6107eff731f3..ec04b8d3c791 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -251,6 +251,7 @@ static struct ata_port_operations pdc_20621_ops = {
 	.phy_reset		= pdc_20621_phy_reset,
 	.qc_prep		= pdc20621_qc_prep,
 	.qc_issue		= pdc20621_qc_issue,
+	.qc_fill_rtf		= ata_sff_qc_fill_rtf,
 	.sff_data_xfer		= ata_sff_data_xfer,
 	.eng_timeout		= pdc_eng_timeout,
 	.sff_irq_clear		= pdc20621_irq_clear,
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 5bddae19d1fb..2ecd32991522 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -5280,6 +5280,7 @@ static struct ata_port_operations ipr_sata_ops = {
 	.sff_tf_read = ipr_tf_read,
 	.qc_prep = ata_noop_qc_prep,
 	.qc_issue = ipr_qc_issue,
+	.qc_fill_rtf = ata_sff_qc_fill_rtf,
 	.port_start = ata_sas_port_start,
 	.port_stop = ata_sas_port_stop
 };
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index a1664b87927a..2ec255839dcd 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -356,6 +356,7 @@ static struct ata_port_operations sas_sata_ops = {
 	.sff_tf_read		= sas_ata_tf_read,
 	.qc_prep		= ata_noop_qc_prep,
 	.qc_issue		= sas_ata_qc_issue,
+	.qc_fill_rtf		= ata_sff_qc_fill_rtf,
 	.port_start		= ata_sas_port_start,
 	.port_stop		= ata_sas_port_stop,
 	.scr_read		= sas_ata_scr_read,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 2b5a0b77e179..bb4200d42f0d 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -715,6 +715,7 @@ struct ata_port_operations {
 	int  (*check_atapi_dma)(struct ata_queued_cmd *qc);
 	void (*qc_prep)(struct ata_queued_cmd *qc);
 	unsigned int (*qc_issue)(struct ata_queued_cmd *qc);
+	bool (*qc_fill_rtf)(struct ata_queued_cmd *qc);
 
 	/*
 	 * Configuration and exception handling
@@ -1385,6 +1386,7 @@ extern void ata_sff_irq_clear(struct ata_port *ap);
 extern int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
 			    u8 status, int in_wq);
 extern unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc);
+extern bool ata_sff_qc_fill_rtf(struct ata_queued_cmd *qc);
 extern unsigned int ata_sff_host_intr(struct ata_port *ap,
 				      struct ata_queued_cmd *qc);
 extern irqreturn_t ata_sff_interrupt(int irq, void *dev_instance);
-- 
cgit v1.2.3


From 79f97dadfe9b4b561634d202225ba2fa910dc225 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:20 +0900
Subject: libata: drop @finish_qc from ata_qc_complete_multiple()

ata_qc_complete_multiple() took @finish_qc and called it on every qc
before completing it.  This was to give opportunity to update TF cache
before ata_qc_complete() tries to fill result_tf.  Now that result TF
is a separate operation, this is no longer necessary.

Update sata_sil24, which was the only user of this mechanism, such
that it implements its own ops->qc_fill_rtf() and drop @finish_qc from
ata_qc_complete_multiple().

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c        |  2 +-
 drivers/ata/libata-core.c |  6 +-----
 drivers/ata/sata_sil24.c  | 19 +++++++++----------
 include/linux/libata.h    |  3 +--
 4 files changed, 12 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 3071a2341be3..1389c64e0027 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1696,7 +1696,7 @@ static void ahci_port_intr(struct ata_port *ap)
 	else
 		qc_active = readl(port_mmio + PORT_CMD_ISSUE);
 
-	rc = ata_qc_complete_multiple(ap, qc_active, NULL);
+	rc = ata_qc_complete_multiple(ap, qc_active);
 
 	/* while resetting, invalid completions are expected */
 	if (unlikely(rc < 0 && !resetting)) {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 3b822124e97e..dcc5a28e26c1 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4673,7 +4673,6 @@ void ata_qc_complete(struct ata_queued_cmd *qc)
  *	ata_qc_complete_multiple - Complete multiple qcs successfully
  *	@ap: port in question
  *	@qc_active: new qc_active mask
- *	@finish_qc: LLDD callback invoked before completing a qc
  *
  *	Complete in-flight commands.  This functions is meant to be
  *	called from low-level driver's interrupt routine to complete
@@ -4686,8 +4685,7 @@ void ata_qc_complete(struct ata_queued_cmd *qc)
  *	RETURNS:
  *	Number of completed commands on success, -errno otherwise.
  */
-int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active,
-			     void (*finish_qc)(struct ata_queued_cmd *))
+int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active)
 {
 	int nr_done = 0;
 	u32 done_mask;
@@ -4708,8 +4706,6 @@ int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active,
 			continue;
 
 		if ((qc = ata_qc_from_tag(ap, i))) {
-			if (finish_qc)
-				finish_qc(qc);
 			ata_qc_complete(qc);
 			nr_done++;
 		}
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index fc9d48cd8122..79952f825b40 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -348,6 +348,7 @@ static void sil24_tf_read(struct ata_port *ap, struct ata_taskfile *tf);
 static int sil24_qc_defer(struct ata_queued_cmd *qc);
 static void sil24_qc_prep(struct ata_queued_cmd *qc);
 static unsigned int sil24_qc_issue(struct ata_queued_cmd *qc);
+static bool sil24_qc_fill_rtf(struct ata_queued_cmd *qc);
 static void sil24_pmp_attach(struct ata_port *ap);
 static void sil24_pmp_detach(struct ata_port *ap);
 static void sil24_freeze(struct ata_port *ap);
@@ -407,6 +408,7 @@ static struct ata_port_operations sil24_ops = {
 	.qc_defer		= sil24_qc_defer,
 	.qc_prep		= sil24_qc_prep,
 	.qc_issue		= sil24_qc_issue,
+	.qc_fill_rtf		= sil24_qc_fill_rtf,
 
 	.freeze			= sil24_freeze,
 	.thaw			= sil24_thaw,
@@ -914,6 +916,12 @@ static unsigned int sil24_qc_issue(struct ata_queued_cmd *qc)
 	return 0;
 }
 
+static bool sil24_qc_fill_rtf(struct ata_queued_cmd *qc)
+{
+	sil24_read_tf(qc->ap, qc->tag, &qc->result_tf);
+	return true;
+}
+
 static void sil24_pmp_attach(struct ata_port *ap)
 {
 	sil24_config_pmp(ap, 1);
@@ -1098,15 +1106,6 @@ static void sil24_error_intr(struct ata_port *ap)
 	}
 }
 
-static void sil24_finish_qc(struct ata_queued_cmd *qc)
-{
-	struct ata_port *ap = qc->ap;
-	struct sil24_port_priv *pp = ap->private_data;
-
-	if (qc->flags & ATA_QCFLAG_RESULT_TF)
-		sil24_read_tf(ap, qc->tag, &pp->tf);
-}
-
 static inline void sil24_host_intr(struct ata_port *ap)
 {
 	void __iomem *port = ap->ioaddr.cmd_addr;
@@ -1131,7 +1130,7 @@ static inline void sil24_host_intr(struct ata_port *ap)
 	}
 
 	qc_active = slot_stat & ~HOST_SSTAT_ATTN;
-	rc = ata_qc_complete_multiple(ap, qc_active, sil24_finish_qc);
+	rc = ata_qc_complete_multiple(ap, qc_active);
 	if (rc > 0)
 		return;
 	if (rc < 0) {
diff --git a/include/linux/libata.h b/include/linux/libata.h
index bb4200d42f0d..1d8b6b7de0a2 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -924,8 +924,7 @@ extern void ata_id_string(const u16 *id, unsigned char *s,
 extern void ata_id_c_string(const u16 *id, unsigned char *s,
 			    unsigned int ofs, unsigned int len);
 extern void ata_qc_complete(struct ata_queued_cmd *qc);
-extern int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active,
-				    void (*finish_qc)(struct ata_queued_cmd *));
+extern int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active);
 extern void ata_scsi_simulate(struct ata_device *dev, struct scsi_cmnd *cmd,
 			      void (*done)(struct scsi_cmnd *));
 extern int ata_std_bios_param(struct scsi_device *sdev,
-- 
cgit v1.2.3


From c9f75b04ed5ed65a058d18a8a8dda50632a96de8 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:21 +0900
Subject: libata: kill ata_noop_dev_select()

Now that SFF assumptions are separated out from non-SFF reset
sequence, port_ops->sff_dev_select() is no longer necessary for
non-SFF controllers.  Kill ata_noop_dev_select() and ->sff_dev_select
initialization from base and other non-SFF port_ops.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c     | 19 -------------------
 drivers/scsi/ipr.c            |  1 -
 drivers/scsi/libsas/sas_ata.c |  1 -
 include/linux/libata.h        |  1 -
 4 files changed, 22 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index e8598eeeec39..f23f50af551c 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -84,7 +84,6 @@ const struct ata_port_operations sata_port_ops = {
 
 	.qc_defer		= ata_std_qc_defer,
 	.hardreset		= sata_std_hardreset,
-	.sff_dev_select		= ata_noop_dev_select,
 };
 
 const struct ata_port_operations sata_pmp_port_ops = {
@@ -1431,22 +1430,6 @@ static int ata_hpa_resize(struct ata_device *dev)
 	return 0;
 }
 
-/**
- *	ata_noop_dev_select - Select device 0/1 on ATA bus
- *	@ap: ATA channel to manipulate
- *	@device: ATA device (numbered from zero) to select
- *
- *	This function performs no actual function.
- *
- *	May be used as the dev_select() entry in ata_port_operations.
- *
- *	LOCKING:
- *	caller.
- */
-void ata_noop_dev_select(struct ata_port *ap, unsigned int device)
-{
-}
-
 /**
  *	ata_dump_id - IDENTIFY DEVICE info debugging output
  *	@id: IDENTIFY DEVICE page to dump
@@ -6193,7 +6176,6 @@ static unsigned int ata_dummy_qc_issue(struct ata_queued_cmd *qc)
 struct ata_port_operations ata_dummy_port_ops = {
 	.sff_check_status	= ata_dummy_check_status,
 	.sff_check_altstatus	= ata_dummy_check_status,
-	.sff_dev_select		= ata_noop_dev_select,
 	.qc_prep		= ata_noop_qc_prep,
 	.qc_issue		= ata_dummy_qc_issue,
 	.freeze			= ata_dummy_noret,
@@ -6234,7 +6216,6 @@ EXPORT_SYMBOL_GPL(ata_host_detach);
 EXPORT_SYMBOL_GPL(ata_sg_init);
 EXPORT_SYMBOL_GPL(ata_qc_complete);
 EXPORT_SYMBOL_GPL(ata_qc_complete_multiple);
-EXPORT_SYMBOL_GPL(ata_noop_dev_select);
 EXPORT_SYMBOL_GPL(sata_print_link_status);
 EXPORT_SYMBOL_GPL(atapi_cmd_type);
 EXPORT_SYMBOL_GPL(ata_tf_to_fis);
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 5a95ea7c4dba..65dc18dea845 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -5247,7 +5247,6 @@ static bool ipr_qc_fill_rtf(struct ata_queued_cmd *qc)
 }
 
 static struct ata_port_operations ipr_sata_ops = {
-	.sff_dev_select = ata_noop_dev_select,
 	.phy_reset = ipr_ata_phy_reset,
 	.hardreset = ipr_sata_reset,
 	.post_internal_cmd = ipr_ata_post_internal,
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index 5c114bb7e907..a4811e4106df 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -344,7 +344,6 @@ static int sas_ata_scr_read(struct ata_port *ap, unsigned int sc_reg_in,
 }
 
 static struct ata_port_operations sas_sata_ops = {
-	.sff_dev_select		= ata_noop_dev_select,
 	.phy_reset		= sas_ata_phy_reset,
 	.post_internal_cmd	= sas_ata_post_internal,
 	.qc_prep		= ata_noop_qc_prep,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 1d8b6b7de0a2..037db1883bae 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -911,7 +911,6 @@ extern unsigned long ata_xfer_mode2mask(u8 xfer_mode);
 extern int ata_xfer_mode2shift(unsigned long xfer_mode);
 extern const char *ata_mode_string(unsigned long xfer_mask);
 extern unsigned long ata_id_xfermask(const u16 *id);
-extern void ata_noop_dev_select(struct ata_port *ap, unsigned int device);
 extern int ata_port_start(struct ata_port *ap);
 extern int ata_std_qc_defer(struct ata_queued_cmd *qc);
 extern void ata_noop_qc_prep(struct ata_queued_cmd *qc);
-- 
cgit v1.2.3


From 127102aea2ea9ec4e9ca233e2b1a75c8d3b058c4 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:21 +0900
Subject: libata: make SFF support optional

Now that SFF support is completely separated out from the core layer,
it can be made optional.  Add CONFIG_ATA_SFF and let SFF drivers
depend on it.  If CONFIG_ATA_SFF isn't set, all codes in libata-sff.c
and data structures for SFF support are disabled.  This saves good
number of bytes for small systems.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/Kconfig       | 44 +++++++++++++++++++++++++++-----------------
 drivers/ata/Makefile      |  4 ++--
 drivers/ata/libata-core.c |  2 ++
 drivers/ata/libata-scsi.c |  2 ++
 drivers/ata/libata.h      |  4 ++++
 include/linux/libata.h    |  9 +++++++++
 6 files changed, 46 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index ea665c249035..48c8fc55391e 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -49,6 +49,32 @@ config SATA_AHCI
 
 	  If unsure, say N.
 
+config SATA_SIL24
+	tristate "Silicon Image 3124/3132 SATA support"
+	depends on PCI
+	help
+	  This option enables support for Silicon Image 3124/3132 Serial ATA.
+
+	  If unsure, say N.
+
+config SATA_FSL
+	tristate "Freescale 3.0Gbps SATA support"
+	depends on FSL_SOC
+	help
+	  This option enables support for Freescale 3.0Gbps SATA controller.
+	  It can be found on MPC837x and MPC8315.
+
+	  If unsure, say N.
+
+config ATA_SFF
+	bool "ATA SFF support"
+	default y
+	help
+	  This option adds support for ATA controllers with SFF
+	  compliant or similar programming interface.
+
+if ATA_SFF
+
 config SATA_SVW
 	tristate "ServerWorks Frodo / Apple K2 SATA support"
 	depends on PCI
@@ -125,14 +151,6 @@ config SATA_SIL
 
 	  If unsure, say N.
 
-config SATA_SIL24
-	tristate "Silicon Image 3124/3132 SATA support"
-	depends on PCI
-	help
-	  This option enables support for Silicon Image 3124/3132 Serial ATA.
-
-	  If unsure, say N.
-
 config SATA_SIS
 	tristate "SiS 964/965/966/180 SATA support"
 	depends on PCI
@@ -183,15 +201,6 @@ config PATA_ACPI
 	  firmware in the BIOS. This driver can sometimes handle
 	  otherwise unsupported hardware.
 
-config SATA_FSL
-	tristate "Freescale 3.0Gbps SATA support"
-	depends on FSL_SOC
-	help
-	  This option enables support for Freescale 3.0Gbps SATA controller.
-	  It can be found on MPC837x and MPC8315.
-
-	  If unsure, say N.
-
 config PATA_ALI
 	tristate "ALi PATA support (Experimental)"
 	depends on PCI && EXPERIMENTAL
@@ -679,4 +688,5 @@ config PATA_BF54X
 
 	  If unsure, say N.
 
+endif # ATA_SFF
 endif # ATA
diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile
index 0511e6f0bb58..e6e41b2c731c 100644
--- a/drivers/ata/Makefile
+++ b/drivers/ata/Makefile
@@ -78,6 +78,6 @@ obj-$(CONFIG_ATA_GENERIC)	+= ata_generic.o
 # Should be last libata driver
 obj-$(CONFIG_PATA_LEGACY)	+= pata_legacy.o
 
-libata-objs	:= libata-core.o libata-scsi.o libata-sff.o libata-eh.o \
-		   libata-pmp.o
+libata-objs	:= libata-core.o libata-scsi.o libata-eh.o libata-pmp.o
+libata-$(CONFIG_ATA_SFF)	+= libata-sff.o
 libata-$(CONFIG_ATA_ACPI)	+= libata-acpi.o
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 3a94c69c7fe7..ca60af0cb051 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5195,7 +5195,9 @@ struct ata_port *ata_port_alloc(struct ata_host *host)
 	ap->msg_enable = ATA_MSG_DRV | ATA_MSG_ERR | ATA_MSG_WARN;
 #endif
 
+#ifdef CONFIG_ATA_SFF
 	INIT_DELAYED_WORK(&ap->port_task, ata_pio_task);
+#endif
 	INIT_DELAYED_WORK(&ap->hotplug_task, ata_scsi_hotplug);
 	INIT_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan);
 	INIT_LIST_HEAD(&ap->eh_done_q);
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index f8be92836a6e..a70881c408e5 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -2393,7 +2393,9 @@ static void atapi_request_sense(struct ata_queued_cmd *qc)
 	/* FIXME: is this needed? */
 	memset(cmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE);
 
+#ifdef CONFIG_ATA_SFF
 	ap->ops->sff_tf_read(ap, &qc->tf);
+#endif
 
 	/* fill these in, for the case where they are -not- overwritten */
 	cmd->sense_buffer[0] = 0x70;
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 87f54a1db3b0..6b70a624828c 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -42,8 +42,10 @@ static inline int ata_is_builtin_hardreset(ata_reset_fn_t reset)
 {
 	if (reset == sata_std_hardreset)
 		return 1;
+#ifdef CONFIG_ATA_SFF
 	if (reset == sata_sff_hardreset)
 		return 1;
+#endif
 	return 0;
 }
 
@@ -206,9 +208,11 @@ extern int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 extern void ata_eh_finish(struct ata_port *ap);
 
 /* libata-sff.c */
+#ifdef CONFIG_ATA_SFF
 extern void ata_dev_select(struct ata_port *ap, unsigned int device,
                            unsigned int wait, unsigned int can_sleep);
 extern u8 ata_irq_on(struct ata_port *ap);
 extern void ata_pio_task(struct work_struct *work);
+#endif /* CONFIG_ATA_SFF */
 
 #endif /* __LIBATA_H__ */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 037db1883bae..db77b90003fd 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -445,6 +445,7 @@ enum link_pm {
 };
 extern struct class_device_attribute class_device_attr_link_power_management_policy;
 
+#ifdef CONFIG_ATA_SFF
 struct ata_ioports {
 	void __iomem		*cmd_addr;
 	void __iomem		*data_addr;
@@ -462,6 +463,7 @@ struct ata_ioports {
 	void __iomem		*bmdma_addr;
 	void __iomem		*scr_addr;
 };
+#endif /* CONFIG_ATA_SFF */
 
 struct ata_host {
 	spinlock_t		lock;
@@ -648,7 +650,9 @@ struct ata_port {
 	struct ata_prd		*prd;	 /* our SG list */
 	dma_addr_t		prd_dma; /* and its DMA mapping */
 
+#ifdef CONFIG_ATA_SFF
 	struct ata_ioports	ioaddr;	/* ATA cmd/ctl/dma register blocks */
+#endif /* CONFIG_ATA_SFF */
 
 	u8			ctl;	/* cache of ATA control register */
 	u8			last_ctl;	/* Cache last written value */
@@ -760,6 +764,7 @@ struct ata_port_operations {
 	void (*port_stop)(struct ata_port *ap);
 	void (*host_stop)(struct ata_host *host);
 
+#ifdef CONFIG_ATA_SFF
 	/*
 	 * SFF / taskfile oriented ops
 	 */
@@ -779,6 +784,7 @@ struct ata_port_operations {
 	void (*bmdma_start)(struct ata_queued_cmd *qc);
 	void (*bmdma_stop)(struct ata_queued_cmd *qc);
 	u8   (*bmdma_status)(struct ata_port *ap);
+#endif /* CONFIG_ATA_SFF */
 
 	/*
 	 * Obsolete
@@ -1349,6 +1355,8 @@ static inline struct ata_port *ata_shost_to_port(struct Scsi_Host *host)
 /**************************************************************************
  * SFF - drivers/ata/libata-sff.c
  */
+#ifdef CONFIG_ATA_SFF
+
 extern const struct ata_port_operations ata_sff_port_ops;
 extern const struct ata_port_operations ata_bmdma_port_ops;
 
@@ -1489,5 +1497,6 @@ static inline u8 ata_wait_idle(struct ata_port *ap)
 
 	return status;
 }
+#endif /* CONFIG_ATA_SFF */
 
 #endif /* __LINUX_LIBATA_H__ */
-- 
cgit v1.2.3


From 48515f6c006c2a9d7b624ee8ad068018c2d3fe0e Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:21 +0900
Subject: libata: separate PMP support code from core code

Most of PMP support code is already in libata-pmp.c.  All that are in
libata-core.c are sata_pmp_port_ops and EXPORTs.  Move them to
libata-pmp.c.  Also, collect PMP related prototypes and declarations
in header files and move them right above of SFF stuff.

This change is to make PMP support optional.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/libata-core.c | 13 -------------
 drivers/ata/libata-pmp.c  | 12 ++++++++++++
 drivers/ata/libata.h      | 10 +++++-----
 include/linux/libata.h    | 17 ++++++++++-------
 4 files changed, 27 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index ca60af0cb051..b2d5d63fb6c9 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -86,15 +86,6 @@ const struct ata_port_operations sata_port_ops = {
 	.hardreset		= sata_std_hardreset,
 };
 
-const struct ata_port_operations sata_pmp_port_ops = {
-	.inherits		= &sata_port_ops,
-
-	.pmp_prereset		= ata_std_prereset,
-	.pmp_hardreset		= sata_std_hardreset,
-	.pmp_postreset		= ata_std_postreset,
-	.error_handler		= sata_pmp_error_handler,
-};
-
 static unsigned int ata_dev_init_params(struct ata_device *dev,
 					u16 heads, u16 sectors);
 static unsigned int ata_dev_set_xfermode(struct ata_device *dev);
@@ -6192,7 +6183,6 @@ EXPORT_SYMBOL_GPL(sata_deb_timing_hotplug);
 EXPORT_SYMBOL_GPL(sata_deb_timing_long);
 EXPORT_SYMBOL_GPL(ata_base_port_ops);
 EXPORT_SYMBOL_GPL(sata_port_ops);
-EXPORT_SYMBOL_GPL(sata_pmp_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_info);
 EXPORT_SYMBOL_GPL(ata_std_bios_param);
@@ -6272,9 +6262,6 @@ EXPORT_SYMBOL_GPL(ata_pci_device_resume);
 #endif /* CONFIG_PM */
 #endif /* CONFIG_PCI */
 
-EXPORT_SYMBOL_GPL(sata_pmp_qc_defer_cmd_switch);
-EXPORT_SYMBOL_GPL(sata_pmp_error_handler);
-
 EXPORT_SYMBOL_GPL(__ata_ehi_push_desc);
 EXPORT_SYMBOL_GPL(ata_ehi_push_desc);
 EXPORT_SYMBOL_GPL(ata_ehi_clear_desc);
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 9c998611b644..bb10c0630791 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -11,6 +11,14 @@
 #include <linux/libata.h>
 #include "libata.h"
 
+const struct ata_port_operations sata_pmp_port_ops = {
+	.inherits		= &sata_port_ops,
+	.pmp_prereset		= ata_std_prereset,
+	.pmp_hardreset		= sata_std_hardreset,
+	.pmp_postreset		= ata_std_postreset,
+	.error_handler		= sata_pmp_error_handler,
+};
+
 /**
  *	sata_pmp_read - read PMP register
  *	@link: link to read PMP register for
@@ -1012,3 +1020,7 @@ void sata_pmp_error_handler(struct ata_port *ap)
 	sata_pmp_eh_recover(ap);
 	ata_eh_finish(ap);
 }
+
+EXPORT_SYMBOL_GPL(sata_pmp_port_ops);
+EXPORT_SYMBOL_GPL(sata_pmp_qc_defer_cmd_switch);
+EXPORT_SYMBOL_GPL(sata_pmp_error_handler);
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 6b70a624828c..42b30e38495d 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -179,11 +179,6 @@ extern void ata_schedule_scsi_eh(struct Scsi_Host *shost);
 extern void ata_scsi_dev_rescan(struct work_struct *work);
 extern int ata_bus_probe(struct ata_port *ap);
 
-/* libata-pmp.c */
-extern int sata_pmp_scr_read(struct ata_link *link, int reg, u32 *val);
-extern int sata_pmp_scr_write(struct ata_link *link, int reg, u32 val);
-extern int sata_pmp_attach(struct ata_device *dev);
-
 /* libata-eh.c */
 extern enum scsi_eh_timer_return ata_scsi_timed_out(struct scsi_cmnd *cmd);
 extern void ata_scsi_error(struct Scsi_Host *host);
@@ -207,6 +202,11 @@ extern int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 			  struct ata_link **r_failed_disk);
 extern void ata_eh_finish(struct ata_port *ap);
 
+/* libata-pmp.c */
+extern int sata_pmp_scr_read(struct ata_link *link, int reg, u32 *val);
+extern int sata_pmp_scr_write(struct ata_link *link, int reg, u32 val);
+extern int sata_pmp_attach(struct ata_device *dev);
+
 /* libata-sff.c */
 #ifdef CONFIG_ATA_SFF
 extern void ata_dev_select(struct ata_port *ap, unsigned int device,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index db77b90003fd..eb86d6f39635 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1026,12 +1026,6 @@ static inline int ata_acpi_cbl_80wire(struct ata_port *ap,
 }
 #endif
 
-/*
- * PMP - drivers/ata/libata-pmp.c
- */
-extern int sata_pmp_qc_defer_cmd_switch(struct ata_queued_cmd *qc);
-extern void sata_pmp_error_handler(struct ata_port *ap);
-
 /*
  * EH - drivers/ata/libata-eh.c
  */
@@ -1075,7 +1069,6 @@ extern void ata_std_error_handler(struct ata_port *ap);
  */
 extern const struct ata_port_operations ata_base_port_ops;
 extern const struct ata_port_operations sata_port_ops;
-extern const struct ata_port_operations sata_pmp_port_ops;
 
 #define ATA_BASE_SHT(drv_name)					\
 	.module			= THIS_MODULE,			\
@@ -1352,6 +1345,16 @@ static inline struct ata_port *ata_shost_to_port(struct Scsi_Host *host)
 	return *(struct ata_port **)&host->hostdata[0];
 }
 
+
+/**************************************************************************
+ * PMP - drivers/ata/libata-pmp.c
+ */
+extern const struct ata_port_operations sata_pmp_port_ops;
+
+extern int sata_pmp_qc_defer_cmd_switch(struct ata_queued_cmd *qc);
+extern void sata_pmp_error_handler(struct ata_port *ap);
+
+
 /**************************************************************************
  * SFF - drivers/ata/libata-sff.c
  */
-- 
cgit v1.2.3


From 071f44b1d2c051641b62a3571223314737ccbe59 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:22 +0900
Subject: libata: implement PMP helpers

Implement helpers to test whether PMP is supported, attached and
determine pmp number to use when issuing SRST to a link.  While at it,
move ata_is_host_link() so that it's together with the two new PMP
helpers.

This change simplifies LLDs and helps making PMP support optional.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/ahci.c        | 30 ++++++------------------------
 drivers/ata/libata-acpi.c |  2 +-
 drivers/ata/libata-core.c |  4 ++--
 drivers/ata/libata-eh.c   | 14 +++++++-------
 drivers/ata/libata-pmp.c  |  6 +++---
 drivers/ata/libata-scsi.c |  6 +++---
 drivers/ata/sata_sil24.c  | 27 +++++++--------------------
 include/linux/libata.h    | 38 +++++++++++++++++++++++++++++---------
 8 files changed, 58 insertions(+), 69 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 6281f7f9eae6..0de6432ee026 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -260,8 +260,6 @@ static int ahci_vt8251_hardreset(struct ata_link *link, unsigned int *class,
 static int ahci_p5wdh_hardreset(struct ata_link *link, unsigned int *class,
 				unsigned long deadline);
 static void ahci_postreset(struct ata_link *link, unsigned int *class);
-static int ahci_pmp_softreset(struct ata_link *link, unsigned int *class,
-			      unsigned long deadline);
 static void ahci_error_handler(struct ata_port *ap);
 static void ahci_post_internal_cmd(struct ata_queued_cmd *qc);
 static int ahci_port_resume(struct ata_port *ap);
@@ -301,7 +299,7 @@ static struct ata_port_operations ahci_ops = {
 	.softreset		= ahci_softreset,
 	.hardreset		= ahci_hardreset,
 	.postreset		= ahci_postreset,
-	.pmp_softreset		= ahci_pmp_softreset,
+	.pmp_softreset		= ahci_softreset,
 	.error_handler		= ahci_error_handler,
 	.post_internal_cmd	= ahci_post_internal_cmd,
 	.dev_config		= ahci_dev_config,
@@ -1263,10 +1261,11 @@ static int ahci_check_ready(struct ata_link *link)
 	return 0;
 }
 
-static int ahci_do_softreset(struct ata_link *link, unsigned int *class,
-			     int pmp, unsigned long deadline)
+static int ahci_softreset(struct ata_link *link, unsigned int *class,
+			  unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
+	int pmp = sata_srst_pmp(link);
 	const char *reason = NULL;
 	unsigned long now, msecs;
 	struct ata_taskfile tf;
@@ -1326,17 +1325,6 @@ static int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 	return rc;
 }
 
-static int ahci_softreset(struct ata_link *link, unsigned int *class,
-			  unsigned long deadline)
-{
-	int pmp = 0;
-
-	if (link->ap->flags & ATA_FLAG_PMP)
-		pmp = SATA_PMP_CTRL_PORT;
-
-	return ahci_do_softreset(link, class, pmp, deadline);
-}
-
 static int ahci_hardreset(struct ata_link *link, unsigned int *class,
 			  unsigned long deadline)
 {
@@ -1457,12 +1445,6 @@ static void ahci_postreset(struct ata_link *link, unsigned int *class)
 	}
 }
 
-static int ahci_pmp_softreset(struct ata_link *link, unsigned int *class,
-			      unsigned long deadline)
-{
-	return ahci_do_softreset(link, class, link->pmp, deadline);
-}
-
 static unsigned int ahci_fill_sg(struct ata_queued_cmd *qc, void *cmd_tbl)
 {
 	struct scatterlist *sg;
@@ -1581,7 +1563,7 @@ static void ahci_error_intr(struct ata_port *ap, u32 irq_stat)
 				  unk[0], unk[1], unk[2], unk[3]);
 	}
 
-	if (ap->nr_pmp_links && (irq_stat & PORT_IRQ_BAD_PMP)) {
+	if (sata_pmp_attached(ap) && (irq_stat & PORT_IRQ_BAD_PMP)) {
 		active_ehi->err_mask |= AC_ERR_HSM;
 		active_ehi->action |= ATA_EH_RESET;
 		ata_ehi_push_desc(active_ehi, "incorrect PMP");
@@ -1847,7 +1829,7 @@ static int ahci_port_resume(struct ata_port *ap)
 	ahci_power_up(ap);
 	ahci_start_port(ap);
 
-	if (ap->nr_pmp_links)
+	if (sata_pmp_attached(ap))
 		ahci_pmp_attach(ap);
 	else
 		ahci_pmp_detach(ap);
diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c
index bf98a566adac..f88a4f940e11 100644
--- a/drivers/ata/libata-acpi.c
+++ b/drivers/ata/libata-acpi.c
@@ -77,7 +77,7 @@ void ata_acpi_associate_sata_port(struct ata_port *ap)
 {
 	WARN_ON(!(ap->flags & ATA_FLAG_ACPI_SATA));
 
-	if (!ap->nr_pmp_links) {
+	if (!sata_pmp_attached(ap)) {
 		acpi_integer adr = SATA_ADR(ap->port_no, NO_PORT_MULT);
 
 		ap->link.device->acpi_handle =
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index b2d5d63fb6c9..3401248180c9 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -2278,7 +2278,7 @@ int ata_dev_configure(struct ata_device *dev)
 		 * changed notifications and ATAPI ANs.
 		 */
 		if ((ap->flags & ATA_FLAG_AN) && ata_id_has_atapi_AN(id) &&
-		    (!ap->nr_pmp_links ||
+		    (!sata_pmp_attached(ap) ||
 		     sata_scr_read(&ap->link, SCR_NOTIFICATION, &sntf) == 0)) {
 			unsigned int err_mask;
 
@@ -3623,7 +3623,7 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 	if (online)
 		*online = true;
 
-	if ((link->ap->flags & ATA_FLAG_PMP) && ata_is_host_link(link)) {
+	if (sata_pmp_supported(link->ap) && ata_is_host_link(link)) {
 		/* If PMP is supported, we have to do follow-up SRST.
 		 * Some PMPs don't send D2H Reg FIS after hardreset if
 		 * the first port is empty.  Wait only for
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 4ec1397434c0..99f83bdc572b 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -873,9 +873,9 @@ int sata_async_notification(struct ata_port *ap)
 	if (rc == 0)
 		sata_scr_write(&ap->link, SCR_NOTIFICATION, sntf);
 
-	if (!ap->nr_pmp_links || rc) {
+	if (!sata_pmp_attached(ap) || rc) {
 		/* PMP is not attached or SNTF is not available */
-		if (!ap->nr_pmp_links) {
+		if (!sata_pmp_attached(ap)) {
 			/* PMP is not attached.  Check whether ATAPI
 			 * AN is configured.  If so, notify media
 			 * change.
@@ -1853,7 +1853,7 @@ void ata_eh_autopsy(struct ata_port *ap)
 	/* Autopsy of fanout ports can affect host link autopsy.
 	 * Perform host link autopsy last.
 	 */
-	if (ap->nr_pmp_links)
+	if (sata_pmp_attached(ap))
 		ata_eh_link_autopsy(&ap->link);
 }
 
@@ -2076,7 +2076,7 @@ static int ata_eh_followup_srst_needed(struct ata_link *link,
 	}
 	if (rc != 0)
 		return 0;
-	if ((link->ap->flags & ATA_FLAG_PMP) && ata_is_host_link(link))
+	if (sata_pmp_supported(link->ap) && ata_is_host_link(link))
 		return 1;
 	return 0;
 }
@@ -2668,7 +2668,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 		/* if PMP is attached, this function only deals with
 		 * downstream links, port should stay thawed.
 		 */
-		if (!ap->nr_pmp_links)
+		if (!sata_pmp_attached(ap))
 			ata_eh_freeze_port(ap);
 
 		ata_port_for_each_link(link, ap) {
@@ -2687,7 +2687,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 			}
 		}
 
-		if (!ap->nr_pmp_links)
+		if (!sata_pmp_attached(ap))
 			ata_eh_thaw_port(ap);
 	}
 
@@ -2731,7 +2731,7 @@ dev_fail:
 			/* PMP reset requires working host port.
 			 * Can't retry if it's frozen.
 			 */
-			if (ap->nr_pmp_links)
+			if (sata_pmp_attached(ap))
 				goto out;
 			break;
 		}
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index bb10c0630791..ff1822a7da38 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -411,7 +411,7 @@ int sata_pmp_attach(struct ata_device *dev)
 	int rc;
 
 	/* is it hanging off the right place? */
-	if (!(ap->flags & ATA_FLAG_PMP)) {
+	if (!sata_pmp_supported(ap)) {
 		ata_dev_printk(dev, KERN_ERR,
 			       "host does not support Port Multiplier\n");
 		return -EINVAL;
@@ -876,7 +876,7 @@ static int sata_pmp_eh_recover(struct ata_port *ap)
 
  retry:
 	/* PMP attached? */
-	if (!ap->nr_pmp_links) {
+	if (!sata_pmp_attached(ap)) {
 		rc = ata_eh_recover(ap, ops->prereset, ops->softreset,
 				    ops->hardreset, ops->postreset, NULL);
 		if (rc) {
@@ -983,7 +983,7 @@ static int sata_pmp_eh_recover(struct ata_port *ap)
 	if (ap->pflags & ATA_PFLAG_UNLOADING)
 		return rc;
 
-	if (!ap->nr_pmp_links)
+	if (!sata_pmp_attached(ap))
 		goto retry;
 
 	if (--pmp_tries) {
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index a70881c408e5..fedf62de9460 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -2617,7 +2617,7 @@ static unsigned int atapi_xlat(struct ata_queued_cmd *qc)
 
 static struct ata_device *ata_find_dev(struct ata_port *ap, int devno)
 {
-	if (ap->nr_pmp_links == 0) {
+	if (!sata_pmp_attached(ap)) {
 		if (likely(devno < ata_link_max_devices(&ap->link)))
 			return &ap->link.device[devno];
 	} else {
@@ -2634,7 +2634,7 @@ static struct ata_device *__ata_scsi_find_dev(struct ata_port *ap,
 	int devno;
 
 	/* skip commands not addressed to targets we simulate */
-	if (ap->nr_pmp_links == 0) {
+	if (!sata_pmp_attached(ap)) {
 		if (unlikely(scsidev->channel || scsidev->lun))
 			return NULL;
 		devno = scsidev->id;
@@ -3492,7 +3492,7 @@ static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
 	if (lun != SCAN_WILD_CARD && lun)
 		return -EINVAL;
 
-	if (ap->nr_pmp_links == 0) {
+	if (!sata_pmp_attached(ap)) {
 		if (channel != SCAN_WILD_CARD && channel)
 			return -EINVAL;
 		devno = id;
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index 6039614e956c..068789361895 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -354,8 +354,6 @@ static int sil24_softreset(struct ata_link *link, unsigned int *class,
 			   unsigned long deadline);
 static int sil24_hardreset(struct ata_link *link, unsigned int *class,
 			   unsigned long deadline);
-static int sil24_pmp_softreset(struct ata_link *link, unsigned int *class,
-			       unsigned long deadline);
 static int sil24_pmp_hardreset(struct ata_link *link, unsigned int *class,
 			       unsigned long deadline);
 static void sil24_error_handler(struct ata_port *ap);
@@ -408,7 +406,7 @@ static struct ata_port_operations sil24_ops = {
 	.thaw			= sil24_thaw,
 	.softreset		= sil24_softreset,
 	.hardreset		= sil24_hardreset,
-	.pmp_softreset		= sil24_pmp_softreset,
+	.pmp_softreset		= sil24_softreset,
 	.pmp_hardreset		= sil24_pmp_hardreset,
 	.error_handler		= sil24_error_handler,
 	.post_internal_cmd	= sil24_post_internal_cmd,
@@ -588,7 +586,7 @@ static int sil24_init_port(struct ata_port *ap)
 	u32 tmp;
 
 	/* clear PMP error status */
-	if (ap->nr_pmp_links)
+	if (sata_pmp_attached(ap))
 		sil24_clear_pmp(ap);
 
 	writel(PORT_CS_INIT, port + PORT_CTRL_STAT);
@@ -653,10 +651,11 @@ static int sil24_exec_polled_cmd(struct ata_port *ap, int pmp,
 	return rc;
 }
 
-static int sil24_do_softreset(struct ata_link *link, unsigned int *class,
-			      int pmp, unsigned long deadline)
+static int sil24_softreset(struct ata_link *link, unsigned int *class,
+			   unsigned long deadline)
 {
 	struct ata_port *ap = link->ap;
+	int pmp = sata_srst_pmp(link);
 	unsigned long timeout_msec = 0;
 	struct ata_taskfile tf;
 	const char *reason;
@@ -706,12 +705,6 @@ static int sil24_do_softreset(struct ata_link *link, unsigned int *class,
 	return -EIO;
 }
 
-static int sil24_softreset(struct ata_link *link, unsigned int *class,
-			   unsigned long deadline)
-{
-	return sil24_do_softreset(link, class, SATA_PMP_CTRL_PORT, deadline);
-}
-
 static int sil24_hardreset(struct ata_link *link, unsigned int *class,
 			   unsigned long deadline)
 {
@@ -926,12 +919,6 @@ static void sil24_pmp_detach(struct ata_port *ap)
 	sil24_config_pmp(ap, 0);
 }
 
-static int sil24_pmp_softreset(struct ata_link *link, unsigned int *class,
-			       unsigned long deadline)
-{
-	return sil24_do_softreset(link, class, link->pmp, deadline);
-}
-
 static int sil24_pmp_hardreset(struct ata_link *link, unsigned int *class,
 			       unsigned long deadline)
 {
@@ -1034,7 +1021,7 @@ static void sil24_error_intr(struct ata_port *ap)
 		}
 
 		/* find out the offending link and qc */
-		if (ap->nr_pmp_links) {
+		if (sata_pmp_attached(ap)) {
 			context = readl(port + PORT_CONTEXT);
 			pmp = (context >> 5) & 0xf;
 
@@ -1082,7 +1069,7 @@ static void sil24_error_intr(struct ata_port *ap)
 		ehi->action |= action;
 
 		/* if PMP, resume */
-		if (ap->nr_pmp_links)
+		if (sata_pmp_attached(ap))
 			writel(PORT_CS_PMP_RESUME, port + PORT_CTRL_STAT);
 	}
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index eb86d6f39635..1908bf484743 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1089,6 +1089,31 @@ extern const struct ata_port_operations sata_port_ops;
 	ATA_BASE_SHT(drv_name),					\
 	.change_queue_depth	= ata_scsi_change_queue_depth
 
+/*
+ * PMP helpers
+ */
+static inline bool sata_pmp_supported(struct ata_port *ap)
+{
+	return ap->flags & ATA_FLAG_PMP;
+}
+
+static inline bool sata_pmp_attached(struct ata_port *ap)
+{
+	return ap->nr_pmp_links != 0;
+}
+
+static inline int ata_is_host_link(const struct ata_link *link)
+{
+	return link == &link->ap->link;
+}
+
+static inline int sata_srst_pmp(struct ata_link *link)
+{
+	if (sata_pmp_supported(link->ap) && ata_is_host_link(link))
+		return SATA_PMP_CTRL_PORT;
+	return link->pmp;
+}
+
 /*
  * printk helpers
  */
@@ -1096,7 +1121,7 @@ extern const struct ata_port_operations sata_port_ops;
 	printk("%sata%u: "fmt, lv, (ap)->print_id , ##args)
 
 #define ata_link_printk(link, lv, fmt, args...) do { \
-	if ((link)->ap->nr_pmp_links) \
+	if (sata_pmp_attached((link)->ap)) \
 		printk("%sata%u.%02u: "fmt, lv, (link)->ap->print_id,	\
 		       (link)->pmp , ##args); \
 	else \
@@ -1182,11 +1207,6 @@ static inline unsigned int ata_dev_absent(const struct ata_device *dev)
 /*
  * link helpers
  */
-static inline int ata_is_host_link(const struct ata_link *link)
-{
-	return link == &link->ap->link;
-}
-
 static inline int ata_link_max_devices(const struct ata_link *link)
 {
 	if (ata_is_host_link(link) && link->ap->flags & ATA_FLAG_SLAVE_POSS)
@@ -1201,7 +1221,7 @@ static inline int ata_link_active(struct ata_link *link)
 
 static inline struct ata_link *ata_port_first_link(struct ata_port *ap)
 {
-	if (ap->nr_pmp_links)
+	if (sata_pmp_attached(ap))
 		return ap->pmp_link;
 	return &ap->link;
 }
@@ -1210,8 +1230,8 @@ static inline struct ata_link *ata_port_next_link(struct ata_link *link)
 {
 	struct ata_port *ap = link->ap;
 
-	if (link == &ap->link) {
-		if (!ap->nr_pmp_links)
+	if (ata_is_host_link(link)) {
+		if (!sata_pmp_attached(ap))
 			return NULL;
 		return ap->pmp_link;
 	}
-- 
cgit v1.2.3


From 88fcd5627563722483427a55113c0a83f56e8080 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 7 Apr 2008 22:47:22 +0900
Subject: libata: make PMP support optional

Make PMP support optional by adding CONFIG_SATA_PMP and leaving out
libata-pmp.c if it isn't set.  PMP helpers return constant values if
PMP support is not enabled and PMP declarations alias non-PMP
counterparts.  This makes the compiler to leave out PMP related part
out and LLDs to use non-PMP counterparts automatically.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/ata/Kconfig    |  6 ++++++
 drivers/ata/Makefile   |  3 ++-
 drivers/ata/libata.h   | 17 +++++++++++++++++
 include/linux/libata.h | 27 +++++++++++++++++++++++++++
 4 files changed, 52 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index 48c8fc55391e..3eb6035b61b8 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -41,6 +41,12 @@ config ATA_ACPI
 	  You can disable this at kernel boot time by using the
 	  option libata.noacpi=1
 
+config SATA_PMP
+	bool "SATA Port Multiplier support"
+	default y
+	help
+	  This option adds support for SATA Port Multipliers.
+
 config SATA_AHCI
 	tristate "AHCI SATA support"
 	depends on PCI
diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile
index e6e41b2c731c..1fbc2aa648b7 100644
--- a/drivers/ata/Makefile
+++ b/drivers/ata/Makefile
@@ -78,6 +78,7 @@ obj-$(CONFIG_ATA_GENERIC)	+= ata_generic.o
 # Should be last libata driver
 obj-$(CONFIG_PATA_LEGACY)	+= pata_legacy.o
 
-libata-objs	:= libata-core.o libata-scsi.o libata-eh.o libata-pmp.o
+libata-objs	:= libata-core.o libata-scsi.o libata-eh.o
 libata-$(CONFIG_ATA_SFF)	+= libata-sff.o
+libata-$(CONFIG_SATA_PMP)	+= libata-pmp.o
 libata-$(CONFIG_ATA_ACPI)	+= libata-acpi.o
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 42b30e38495d..4aeeabb10a47 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -203,9 +203,26 @@ extern int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 extern void ata_eh_finish(struct ata_port *ap);
 
 /* libata-pmp.c */
+#ifdef CONFIG_SATA_PMP
 extern int sata_pmp_scr_read(struct ata_link *link, int reg, u32 *val);
 extern int sata_pmp_scr_write(struct ata_link *link, int reg, u32 val);
 extern int sata_pmp_attach(struct ata_device *dev);
+#else /* CONFIG_SATA_PMP */
+static inline int sata_pmp_scr_read(struct ata_link *link, int reg, u32 *val)
+{
+	return -EINVAL;
+}
+
+static inline int sata_pmp_scr_write(struct ata_link *link, int reg, u32 val)
+{
+	return -EINVAL;
+}
+
+static inline int sata_pmp_attach(struct ata_device *dev)
+{
+	return -EINVAL;
+}
+#endif /* CONFIG_SATA_PMP */
 
 /* libata-sff.c */
 #ifdef CONFIG_ATA_SFF
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 1908bf484743..165734a2dd47 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1092,6 +1092,7 @@ extern const struct ata_port_operations sata_port_ops;
 /*
  * PMP helpers
  */
+#ifdef CONFIG_SATA_PMP
 static inline bool sata_pmp_supported(struct ata_port *ap)
 {
 	return ap->flags & ATA_FLAG_PMP;
@@ -1106,6 +1107,22 @@ static inline int ata_is_host_link(const struct ata_link *link)
 {
 	return link == &link->ap->link;
 }
+#else /* CONFIG_SATA_PMP */
+static inline bool sata_pmp_supported(struct ata_port *ap)
+{
+	return false;
+}
+
+static inline bool sata_pmp_attached(struct ata_port *ap)
+{
+	return false;
+}
+
+static inline int ata_is_host_link(const struct ata_link *link)
+{
+	return 1;
+}
+#endif /* CONFIG_SATA_PMP */
 
 static inline int sata_srst_pmp(struct ata_link *link)
 {
@@ -1369,11 +1386,21 @@ static inline struct ata_port *ata_shost_to_port(struct Scsi_Host *host)
 /**************************************************************************
  * PMP - drivers/ata/libata-pmp.c
  */
+#ifdef CONFIG_SATA_PMP
+
 extern const struct ata_port_operations sata_pmp_port_ops;
 
 extern int sata_pmp_qc_defer_cmd_switch(struct ata_queued_cmd *qc);
 extern void sata_pmp_error_handler(struct ata_port *ap);
 
+#else /* CONFIG_SATA_PMP */
+
+#define sata_pmp_port_ops		sata_port_ops
+#define sata_pmp_qc_defer_cmd_switch	ata_std_qc_defer
+#define sata_pmp_error_handler		ata_std_error_handler
+
+#endif /* CONFIG_SATA_PMP */
+
 
 /**************************************************************************
  * SFF - drivers/ata/libata-sff.c
-- 
cgit v1.2.3


From 19242d7233df7d658405d4b7ee1758d21414cfaa Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Thu, 17 Apr 2008 20:17:25 -0700
Subject: async_tx: fix multiple dependency submission

Shrink struct dma_async_tx_descriptor and introduce
async_tx_channel_switch to properly inject a channel switch interrupt in
the descriptor stream.  This simplifies the locking model as drivers no
longer need to handle dma_async_tx_descriptor.lock.

Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_tx.c | 197 +++++++++++++++++++++++++++++++++++++--------
 drivers/dma/dmaengine.c    |   2 -
 drivers/dma/iop-adma.c     |   9 ++-
 include/linux/dmaengine.h  |   9 +--
 4 files changed, 170 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 2be3bae89930..69756164b61d 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -89,13 +89,19 @@ dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
 		iter = tx;
 
 		/* find the root of the unsubmitted dependency chain */
-		while (iter->cookie == -EBUSY) {
+		do {
 			parent = iter->parent;
-			if (parent && parent->cookie == -EBUSY)
-				iter = iter->parent;
-			else
+			if (!parent)
 				break;
-		}
+			else
+				iter = parent;
+		} while (parent);
+
+		/* there is a small window for ->parent == NULL and
+		 * ->cookie == -EBUSY
+		 */
+		while (iter->cookie == -EBUSY)
+			cpu_relax();
 
 		status = dma_sync_wait(iter->chan, iter->cookie);
 	} while (status == DMA_IN_PROGRESS || (iter != tx));
@@ -111,24 +117,33 @@ EXPORT_SYMBOL_GPL(dma_wait_for_async_tx);
 void
 async_tx_run_dependencies(struct dma_async_tx_descriptor *tx)
 {
-	struct dma_async_tx_descriptor *dep_tx, *_dep_tx;
-	struct dma_device *dev;
+	struct dma_async_tx_descriptor *next = tx->next;
 	struct dma_chan *chan;
 
-	list_for_each_entry_safe(dep_tx, _dep_tx, &tx->depend_list,
-		depend_node) {
-		chan = dep_tx->chan;
-		dev = chan->device;
-		/* we can't depend on ourselves */
-		BUG_ON(chan == tx->chan);
-		list_del(&dep_tx->depend_node);
-		tx->tx_submit(dep_tx);
-
-		/* we need to poke the engine as client code does not
-		 * know about dependency submission events
-		 */
-		dev->device_issue_pending(chan);
+	if (!next)
+		return;
+
+	tx->next = NULL;
+	chan = next->chan;
+
+	/* keep submitting up until a channel switch is detected
+	 * in that case we will be called again as a result of
+	 * processing the interrupt from async_tx_channel_switch
+	 */
+	while (next && next->chan == chan) {
+		struct dma_async_tx_descriptor *_next;
+
+		spin_lock_bh(&next->lock);
+		next->parent = NULL;
+		_next = next->next;
+		next->next = NULL;
+		spin_unlock_bh(&next->lock);
+
+		next->tx_submit(next);
+		next = _next;
 	}
+
+	chan->device->device_issue_pending(chan);
 }
 EXPORT_SYMBOL_GPL(async_tx_run_dependencies);
 
@@ -397,6 +412,92 @@ static void __exit async_tx_exit(void)
 }
 #endif
 
+
+/**
+ * async_tx_channel_switch - queue an interrupt descriptor with a dependency
+ * 	pre-attached.
+ * @depend_tx: the operation that must finish before the new operation runs
+ * @tx: the new operation
+ */
+static void
+async_tx_channel_switch(struct dma_async_tx_descriptor *depend_tx,
+			struct dma_async_tx_descriptor *tx)
+{
+	struct dma_chan *chan;
+	struct dma_device *device;
+	struct dma_async_tx_descriptor *intr_tx = (void *) ~0;
+
+	/* first check to see if we can still append to depend_tx */
+	spin_lock_bh(&depend_tx->lock);
+	if (depend_tx->parent && depend_tx->chan == tx->chan) {
+		tx->parent = depend_tx;
+		depend_tx->next = tx;
+		intr_tx = NULL;
+	}
+	spin_unlock_bh(&depend_tx->lock);
+
+	if (!intr_tx)
+		return;
+
+	chan = depend_tx->chan;
+	device = chan->device;
+
+	/* see if we can schedule an interrupt
+	 * otherwise poll for completion
+	 */
+	if (dma_has_cap(DMA_INTERRUPT, device->cap_mask))
+		intr_tx = device->device_prep_dma_interrupt(chan);
+	else
+		intr_tx = NULL;
+
+	if (intr_tx) {
+		intr_tx->callback = NULL;
+		intr_tx->callback_param = NULL;
+		tx->parent = intr_tx;
+		/* safe to set ->next outside the lock since we know we are
+		 * not submitted yet
+		 */
+		intr_tx->next = tx;
+
+		/* check if we need to append */
+		spin_lock_bh(&depend_tx->lock);
+		if (depend_tx->parent) {
+			intr_tx->parent = depend_tx;
+			depend_tx->next = intr_tx;
+			async_tx_ack(intr_tx);
+			intr_tx = NULL;
+		}
+		spin_unlock_bh(&depend_tx->lock);
+
+		if (intr_tx) {
+			intr_tx->parent = NULL;
+			intr_tx->tx_submit(intr_tx);
+			async_tx_ack(intr_tx);
+		}
+	} else {
+		if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
+			panic("%s: DMA_ERROR waiting for depend_tx\n",
+			      __func__);
+		tx->tx_submit(tx);
+	}
+}
+
+
+/**
+ * submit_disposition - while holding depend_tx->lock we must avoid submitting
+ * 	new operations to prevent a circular locking dependency with
+ * 	drivers that already hold a channel lock when calling
+ * 	async_tx_run_dependencies.
+ * @ASYNC_TX_SUBMITTED: we were able to append the new operation under the lock
+ * @ASYNC_TX_CHANNEL_SWITCH: when the lock is dropped schedule a channel switch
+ * @ASYNC_TX_DIRECT_SUBMIT: when the lock is dropped submit directly
+ */
+enum submit_disposition {
+	ASYNC_TX_SUBMITTED,
+	ASYNC_TX_CHANNEL_SWITCH,
+	ASYNC_TX_DIRECT_SUBMIT,
+};
+
 void
 async_tx_submit(struct dma_chan *chan, struct dma_async_tx_descriptor *tx,
 	enum async_tx_flags flags, struct dma_async_tx_descriptor *depend_tx,
@@ -405,28 +506,54 @@ async_tx_submit(struct dma_chan *chan, struct dma_async_tx_descriptor *tx,
 	tx->callback = cb_fn;
 	tx->callback_param = cb_param;
 
-	/* set this new tx to run after depend_tx if:
-	 * 1/ a dependency exists (depend_tx is !NULL)
-	 * 2/ the tx can not be submitted to the current channel
-	 */
-	if (depend_tx && depend_tx->chan != chan) {
-		/* if ack is already set then we cannot be sure
+	if (depend_tx) {
+		enum submit_disposition s;
+
+		/* sanity check the dependency chain:
+		 * 1/ if ack is already set then we cannot be sure
 		 * we are referring to the correct operation
+		 * 2/ dependencies are 1:1 i.e. two transactions can
+		 * not depend on the same parent
 		 */
-		BUG_ON(depend_tx->ack);
+		BUG_ON(depend_tx->ack || depend_tx->next || tx->parent);
 
-		tx->parent = depend_tx;
+		/* the lock prevents async_tx_run_dependencies from missing
+		 * the setting of ->next when ->parent != NULL
+		 */
 		spin_lock_bh(&depend_tx->lock);
-		list_add_tail(&tx->depend_node, &depend_tx->depend_list);
-		if (depend_tx->cookie == 0) {
-			struct dma_chan *dep_chan = depend_tx->chan;
-			struct dma_device *dep_dev = dep_chan->device;
-			dep_dev->device_dependency_added(dep_chan);
+		if (depend_tx->parent) {
+			/* we have a parent so we can not submit directly
+			 * if we are staying on the same channel: append
+			 * else: channel switch
+			 */
+			if (depend_tx->chan == chan) {
+				tx->parent = depend_tx;
+				depend_tx->next = tx;
+				s = ASYNC_TX_SUBMITTED;
+			} else
+				s = ASYNC_TX_CHANNEL_SWITCH;
+		} else {
+			/* we do not have a parent so we may be able to submit
+			 * directly if we are staying on the same channel
+			 */
+			if (depend_tx->chan == chan)
+				s = ASYNC_TX_DIRECT_SUBMIT;
+			else
+				s = ASYNC_TX_CHANNEL_SWITCH;
 		}
 		spin_unlock_bh(&depend_tx->lock);
 
-		/* schedule an interrupt to trigger the channel switch */
-		async_trigger_callback(ASYNC_TX_ACK, depend_tx, NULL, NULL);
+		switch (s) {
+		case ASYNC_TX_SUBMITTED:
+			break;
+		case ASYNC_TX_CHANNEL_SWITCH:
+			async_tx_channel_switch(depend_tx, tx);
+			break;
+		case ASYNC_TX_DIRECT_SUBMIT:
+			tx->parent = NULL;
+			tx->tx_submit(tx);
+			break;
+		}
 	} else {
 		tx->parent = NULL;
 		tx->tx_submit(tx);
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 8db0e7f9d3f4..9cb898a76bb3 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -600,8 +600,6 @@ void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx,
 {
 	tx->chan = chan;
 	spin_lock_init(&tx->lock);
-	INIT_LIST_HEAD(&tx->depend_node);
-	INIT_LIST_HEAD(&tx->depend_list);
 }
 EXPORT_SYMBOL(dma_async_tx_descriptor_init);
 
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index f82b0906d466..21854cd7190f 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -63,7 +63,6 @@ iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
 	struct iop_adma_chan *iop_chan, dma_cookie_t cookie)
 {
 	BUG_ON(desc->async_tx.cookie < 0);
-	spin_lock_bh(&desc->async_tx.lock);
 	if (desc->async_tx.cookie > 0) {
 		cookie = desc->async_tx.cookie;
 		desc->async_tx.cookie = 0;
@@ -101,7 +100,6 @@ iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
 
 	/* run dependent operations */
 	async_tx_run_dependencies(&desc->async_tx);
-	spin_unlock_bh(&desc->async_tx.lock);
 
 	return cookie;
 }
@@ -275,8 +273,11 @@ iop_adma_slot_cleanup(struct iop_adma_chan *iop_chan)
 
 static void iop_adma_tasklet(unsigned long data)
 {
-	struct iop_adma_chan *chan = (struct iop_adma_chan *) data;
-	__iop_adma_slot_cleanup(chan);
+	struct iop_adma_chan *iop_chan = (struct iop_adma_chan *) data;
+
+	spin_lock(&iop_chan->lock);
+	__iop_adma_slot_cleanup(iop_chan);
+	spin_unlock(&iop_chan->lock);
 }
 
 static struct iop_adma_desc_slot *
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 34d440698293..91252a7e4d03 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -221,11 +221,9 @@ typedef void (*dma_async_tx_callback)(void *dma_async_param);
  * @callback: routine to call after this operation is complete
  * @callback_param: general parameter to pass to the callback routine
  * ---async_tx api specific fields---
- * @depend_list: at completion this list of transactions are submitted
- * @depend_node: allow this transaction to be executed after another
- *	transaction has completed, possibly on another channel
+ * @next: at completion submit this descriptor
  * @parent: pointer to the next level up in the dependency chain
- * @lock: protect the dependency list
+ * @lock: protect the parent and next pointers
  */
 struct dma_async_tx_descriptor {
 	dma_cookie_t cookie;
@@ -236,8 +234,7 @@ struct dma_async_tx_descriptor {
 	dma_cookie_t (*tx_submit)(struct dma_async_tx_descriptor *tx);
 	dma_async_tx_callback callback;
 	void *callback_param;
-	struct list_head depend_list;
-	struct list_head depend_node;
+	struct dma_async_tx_descriptor *next;
 	struct dma_async_tx_descriptor *parent;
 	spinlock_t lock;
 };
-- 
cgit v1.2.3


From ce4d65a5db77e1568c82d5151a746f627c4f6ed5 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Thu, 17 Apr 2008 20:17:26 -0700
Subject: async_tx: kill ->device_dependency_added

DMA drivers no longer need to be notified of dependency submission
events as async_tx_run_dependencies and async_tx_channel_switch will
handle the scheduling and execution of dependent operations.

[sfr@canb.auug.org.au: extend this for fsldma]
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   |  1 -
 drivers/dma/fsldma.c      |  8 --------
 drivers/dma/ioat_dma.c    | 12 ------------
 drivers/dma/iop-adma.c    |  7 -------
 include/linux/dmaengine.h |  2 --
 5 files changed, 30 deletions(-)

(limited to 'include')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 9cb898a76bb3..af6911a75dae 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -362,7 +362,6 @@ int dma_async_device_register(struct dma_device *device)
 
 	BUG_ON(!device->device_alloc_chan_resources);
 	BUG_ON(!device->device_free_chan_resources);
-	BUG_ON(!device->device_dependency_added);
 	BUG_ON(!device->device_is_tx_complete);
 	BUG_ON(!device->device_issue_pending);
 	BUG_ON(!device->dev);
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index d8ae18dbf1a7..95b36b7934a5 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -658,13 +658,6 @@ static void fsl_dma_memcpy_issue_pending(struct dma_chan *chan)
 	fsl_chan_xfer_ld_queue(fsl_chan);
 }
 
-static void fsl_dma_dependency_added(struct dma_chan *chan)
-{
-	struct fsl_dma_chan *fsl_chan = to_fsl_chan(chan);
-
-	fsl_chan_ld_cleanup(fsl_chan);
-}
-
 /**
  * fsl_dma_is_complete - Determine the DMA status
  * @fsl_chan : Freescale DMA channel
@@ -1089,7 +1082,6 @@ static int __devinit of_fsl_dma_probe(struct of_device *dev,
 	fdev->common.device_prep_dma_memcpy = fsl_dma_prep_memcpy;
 	fdev->common.device_is_tx_complete = fsl_dma_is_complete;
 	fdev->common.device_issue_pending = fsl_dma_memcpy_issue_pending;
-	fdev->common.device_dependency_added = fsl_dma_dependency_added;
 	fdev->common.dev = &dev->dev;
 
 	irq = irq_of_parse_and_map(dev->node, 0);
diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
index 4017d9e7acd2..1517fe4e2d14 100644
--- a/drivers/dma/ioat_dma.c
+++ b/drivers/dma/ioat_dma.c
@@ -924,17 +924,6 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 	spin_unlock_bh(&ioat_chan->cleanup_lock);
 }
 
-static void ioat_dma_dependency_added(struct dma_chan *chan)
-{
-	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
-	spin_lock_bh(&ioat_chan->desc_lock);
-	if (ioat_chan->pending == 0) {
-		spin_unlock_bh(&ioat_chan->desc_lock);
-		ioat_dma_memcpy_cleanup(ioat_chan);
-	} else
-		spin_unlock_bh(&ioat_chan->desc_lock);
-}
-
 /**
  * ioat_dma_is_complete - poll the status of a IOAT DMA transaction
  * @chan: IOAT DMA channel handle
@@ -1316,7 +1305,6 @@ struct ioatdma_device *ioat_dma_probe(struct pci_dev *pdev,
 
 	dma_cap_set(DMA_MEMCPY, device->common.cap_mask);
 	device->common.device_is_tx_complete = ioat_dma_is_complete;
-	device->common.device_dependency_added = ioat_dma_dependency_added;
 	switch (device->version) {
 	case IOAT_VER_1_2:
 		device->common.device_prep_dma_memcpy = ioat1_dma_prep_memcpy;
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index 21854cd7190f..2aa3df50c842 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -672,12 +672,6 @@ iop_adma_prep_dma_zero_sum(struct dma_chan *chan, dma_addr_t *dma_src,
 	return sw_desc ? &sw_desc->async_tx : NULL;
 }
 
-static void iop_adma_dependency_added(struct dma_chan *chan)
-{
-	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
-	tasklet_schedule(&iop_chan->irq_tasklet);
-}
-
 static void iop_adma_free_chan_resources(struct dma_chan *chan)
 {
 	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
@@ -1178,7 +1172,6 @@ static int __devinit iop_adma_probe(struct platform_device *pdev)
 	dma_dev->device_free_chan_resources = iop_adma_free_chan_resources;
 	dma_dev->device_is_tx_complete = iop_adma_is_complete;
 	dma_dev->device_issue_pending = iop_adma_issue_pending;
-	dma_dev->device_dependency_added = iop_adma_dependency_added;
 	dma_dev->dev = &pdev->dev;
 
 	/* set prep routines based on capability */
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 91252a7e4d03..cd34df78c6aa 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -258,7 +258,6 @@ struct dma_async_tx_descriptor {
  * @device_prep_dma_zero_sum: prepares a zero_sum operation
  * @device_prep_dma_memset: prepares a memset operation
  * @device_prep_dma_interrupt: prepares an end of chain interrupt operation
- * @device_dependency_added: async_tx notifies the channel about new deps
  * @device_issue_pending: push pending transactions to hardware
  */
 struct dma_device {
@@ -293,7 +292,6 @@ struct dma_device {
 	struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
 		struct dma_chan *chan);
 
-	void (*device_dependency_added)(struct dma_chan *chan);
 	enum dma_status (*device_is_tx_complete)(struct dma_chan *chan,
 			dma_cookie_t cookie, dma_cookie_t *last,
 			dma_cookie_t *used);
-- 
cgit v1.2.3


From c4fe15541d0ef5cc8cc1ce43057663851f8fc387 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Thu, 17 Apr 2008 20:17:26 -0700
Subject: iop-adma: remove the workaround for missed interrupts on iop3xx

This workaround was covering the dependency submission bug in async_tx.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/iop-adma.c                 | 5 -----
 include/asm-arm/arch-iop13xx/adma.h    | 5 -----
 include/asm-arm/hardware/iop3xx-adma.h | 8 --------
 include/asm-arm/hardware/iop_adma.h    | 2 --
 4 files changed, 20 deletions(-)

(limited to 'include')

diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index 2aa3df50c842..93252294f32b 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -255,8 +255,6 @@ static void __iop_adma_slot_cleanup(struct iop_adma_chan *iop_chan)
 
 	BUG_ON(!seen_current);
 
-	iop_chan_idle(busy, iop_chan);
-
 	if (cookie > 0) {
 		iop_chan->completed_cookie = cookie;
 		pr_debug("\tcompleted cookie %d\n", cookie);
@@ -1226,9 +1224,6 @@ static int __devinit iop_adma_probe(struct platform_device *pdev)
 	}
 
 	spin_lock_init(&iop_chan->lock);
-	init_timer(&iop_chan->cleanup_watchdog);
-	iop_chan->cleanup_watchdog.data = (unsigned long) iop_chan;
-	iop_chan->cleanup_watchdog.function = iop_adma_tasklet;
 	INIT_LIST_HEAD(&iop_chan->chain);
 	INIT_LIST_HEAD(&iop_chan->all_slots);
 	INIT_RCU_HEAD(&iop_chan->common.rcu);
diff --git a/include/asm-arm/arch-iop13xx/adma.h b/include/asm-arm/arch-iop13xx/adma.h
index efd9a5eb1008..90d14ee564f5 100644
--- a/include/asm-arm/arch-iop13xx/adma.h
+++ b/include/asm-arm/arch-iop13xx/adma.h
@@ -454,11 +454,6 @@ static inline void iop_chan_append(struct iop_adma_chan *chan)
 	__raw_writel(adma_accr, ADMA_ACCR(chan));
 }
 
-static inline void iop_chan_idle(int busy, struct iop_adma_chan *chan)
-{
-	do { } while (0);
-}
-
 static inline u32 iop_chan_get_status(struct iop_adma_chan *chan)
 {
 	return __raw_readl(ADMA_ACSR(chan));
diff --git a/include/asm-arm/hardware/iop3xx-adma.h b/include/asm-arm/hardware/iop3xx-adma.h
index 5c529e6a5e3b..84d635b0a71a 100644
--- a/include/asm-arm/hardware/iop3xx-adma.h
+++ b/include/asm-arm/hardware/iop3xx-adma.h
@@ -767,20 +767,12 @@ static inline int iop_desc_get_zero_result(struct iop_adma_desc_slot *desc)
 static inline void iop_chan_append(struct iop_adma_chan *chan)
 {
 	u32 dma_chan_ctrl;
-	/* workaround dropped interrupts on 3xx */
-	mod_timer(&chan->cleanup_watchdog, jiffies + msecs_to_jiffies(3));
 
 	dma_chan_ctrl = __raw_readl(DMA_CCR(chan));
 	dma_chan_ctrl |= 0x2;
 	__raw_writel(dma_chan_ctrl, DMA_CCR(chan));
 }
 
-static inline void iop_chan_idle(int busy, struct iop_adma_chan *chan)
-{
-	if (!busy)
-		del_timer(&chan->cleanup_watchdog);
-}
-
 static inline u32 iop_chan_get_status(struct iop_adma_chan *chan)
 {
 	return __raw_readl(DMA_CSR(chan));
diff --git a/include/asm-arm/hardware/iop_adma.h b/include/asm-arm/hardware/iop_adma.h
index ca8e71f44346..cb7e3611bcba 100644
--- a/include/asm-arm/hardware/iop_adma.h
+++ b/include/asm-arm/hardware/iop_adma.h
@@ -51,7 +51,6 @@ struct iop_adma_device {
  * @common: common dmaengine channel object members
  * @last_used: place holder for allocation to continue from where it left off
  * @all_slots: complete domain of slots usable by the channel
- * @cleanup_watchdog: workaround missed interrupts on iop3xx
  * @slots_allocated: records the actual size of the descriptor slot pool
  * @irq_tasklet: bottom half where iop_adma_slot_cleanup runs
  */
@@ -65,7 +64,6 @@ struct iop_adma_chan {
 	struct dma_chan common;
 	struct iop_adma_desc_slot *last_used;
 	struct list_head all_slots;
-	struct timer_list cleanup_watchdog;
 	int slots_allocated;
 	struct tasklet_struct irq_tasklet;
 };
-- 
cgit v1.2.3


From 636bdeaa1243327501edfd2a597ed7443eb4239a Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Thu, 17 Apr 2008 20:17:26 -0700
Subject: dmaengine: ack to flags: make use of the unused bits in the 'ack'
 field

'ack' is currently a simple integer that flags whether or not a client is done
touching fields in the given descriptor.  It is effectively just a single bit
of information.  Converting this to a flags parameter allows the other bits to
be put to use to control completion actions, like dma-unmap, and capture
results, like xor-zero-sum == 0.

Changes are one of:
1/ convert all open-coded ->ack manipulations to use async_tx_ack
   and async_tx_test_ack.
2/ set the ack bit at prep time where possible
3/ make drivers store the flags at prep time
4/ add flags to the device_prep_dma_interrupt prototype

Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_memcpy.c |  2 +-
 crypto/async_tx/async_tx.c     |  9 +++++----
 crypto/async_tx/async_xor.c    |  2 +-
 drivers/dma/dmaengine.c        | 12 ++++++------
 drivers/dma/fsldma.c           | 10 +++++-----
 drivers/dma/ioat_dma.c         | 24 ++++++++++++------------
 drivers/dma/iop-adma.c         | 39 +++++++++++++++++++++------------------
 include/linux/dmaengine.h      | 25 ++++++++++++++++++-------
 8 files changed, 69 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/crypto/async_tx/async_memcpy.c b/crypto/async_tx/async_memcpy.c
index 84caa4efc0d4..a5eda80e8427 100644
--- a/crypto/async_tx/async_memcpy.c
+++ b/crypto/async_tx/async_memcpy.c
@@ -77,7 +77,7 @@ async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
 			/* if ack is already set then we cannot be sure
 			 * we are referring to the correct operation
 			 */
-			BUG_ON(depend_tx->ack);
+			BUG_ON(async_tx_test_ack(depend_tx));
 			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
 				panic("%s: DMA_ERROR waiting for depend_tx\n",
 					__func__);
diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 69756164b61d..c6e772fc5ccd 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -446,7 +446,7 @@ async_tx_channel_switch(struct dma_async_tx_descriptor *depend_tx,
 	 * otherwise poll for completion
 	 */
 	if (dma_has_cap(DMA_INTERRUPT, device->cap_mask))
-		intr_tx = device->device_prep_dma_interrupt(chan);
+		intr_tx = device->device_prep_dma_interrupt(chan, 0);
 	else
 		intr_tx = NULL;
 
@@ -515,7 +515,8 @@ async_tx_submit(struct dma_chan *chan, struct dma_async_tx_descriptor *tx,
 		 * 2/ dependencies are 1:1 i.e. two transactions can
 		 * not depend on the same parent
 		 */
-		BUG_ON(depend_tx->ack || depend_tx->next || tx->parent);
+		BUG_ON(async_tx_test_ack(depend_tx) || depend_tx->next ||
+		       tx->parent);
 
 		/* the lock prevents async_tx_run_dependencies from missing
 		 * the setting of ->next when ->parent != NULL
@@ -594,7 +595,7 @@ async_trigger_callback(enum async_tx_flags flags,
 		if (device && !dma_has_cap(DMA_INTERRUPT, device->cap_mask))
 			device = NULL;
 
-		tx = device ? device->device_prep_dma_interrupt(chan) : NULL;
+		tx = device ? device->device_prep_dma_interrupt(chan, 0) : NULL;
 	} else
 		tx = NULL;
 
@@ -610,7 +611,7 @@ async_trigger_callback(enum async_tx_flags flags,
 			/* if ack is already set then we cannot be sure
 			 * we are referring to the correct operation
 			 */
-			BUG_ON(depend_tx->ack);
+			BUG_ON(async_tx_test_ack(depend_tx));
 			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
 				panic("%s: DMA_ERROR waiting for depend_tx\n",
 					__func__);
diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index 1c445c7bdab7..3a0dddca5a10 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -191,7 +191,7 @@ async_xor(struct page *dest, struct page **src_list, unsigned int offset,
 				/* if ack is already set then we cannot be sure
 				 * we are referring to the correct operation
 				 */
-				BUG_ON(depend_tx->ack);
+				BUG_ON(async_tx_test_ack(depend_tx));
 				if (dma_wait_for_async_tx(depend_tx) ==
 					DMA_ERROR)
 					panic("%s: DMA_ERROR waiting for "
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index af6911a75dae..d6dc70fd7527 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -478,7 +478,8 @@ dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest,
 
 	dma_src = dma_map_single(dev->dev, src, len, DMA_TO_DEVICE);
 	dma_dest = dma_map_single(dev->dev, dest, len, DMA_FROM_DEVICE);
-	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, 0);
+	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len,
+					 DMA_CTRL_ACK);
 
 	if (!tx) {
 		dma_unmap_single(dev->dev, dma_src, len, DMA_TO_DEVICE);
@@ -486,7 +487,6 @@ dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest,
 		return -ENOMEM;
 	}
 
-	tx->ack = 1;
 	tx->callback = NULL;
 	cookie = tx->tx_submit(tx);
 
@@ -524,7 +524,8 @@ dma_async_memcpy_buf_to_pg(struct dma_chan *chan, struct page *page,
 
 	dma_src = dma_map_single(dev->dev, kdata, len, DMA_TO_DEVICE);
 	dma_dest = dma_map_page(dev->dev, page, offset, len, DMA_FROM_DEVICE);
-	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, 0);
+	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len,
+					 DMA_CTRL_ACK);
 
 	if (!tx) {
 		dma_unmap_single(dev->dev, dma_src, len, DMA_TO_DEVICE);
@@ -532,7 +533,6 @@ dma_async_memcpy_buf_to_pg(struct dma_chan *chan, struct page *page,
 		return -ENOMEM;
 	}
 
-	tx->ack = 1;
 	tx->callback = NULL;
 	cookie = tx->tx_submit(tx);
 
@@ -573,7 +573,8 @@ dma_async_memcpy_pg_to_pg(struct dma_chan *chan, struct page *dest_pg,
 	dma_src = dma_map_page(dev->dev, src_pg, src_off, len, DMA_TO_DEVICE);
 	dma_dest = dma_map_page(dev->dev, dest_pg, dest_off, len,
 				DMA_FROM_DEVICE);
-	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, 0);
+	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len,
+					 DMA_CTRL_ACK);
 
 	if (!tx) {
 		dma_unmap_page(dev->dev, dma_src, len, DMA_TO_DEVICE);
@@ -581,7 +582,6 @@ dma_async_memcpy_pg_to_pg(struct dma_chan *chan, struct page *dest_pg,
 		return -ENOMEM;
 	}
 
-	tx->ack = 1;
 	tx->callback = NULL;
 	cookie = tx->tx_submit(tx);
 
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 95b36b7934a5..054eabffc185 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -412,7 +412,7 @@ static void fsl_dma_free_chan_resources(struct dma_chan *chan)
 }
 
 static struct dma_async_tx_descriptor *
-fsl_dma_prep_interrupt(struct dma_chan *chan)
+fsl_dma_prep_interrupt(struct dma_chan *chan, unsigned long flags)
 {
 	struct fsl_dma_chan *fsl_chan;
 	struct fsl_desc_sw *new;
@@ -429,7 +429,7 @@ fsl_dma_prep_interrupt(struct dma_chan *chan)
 	}
 
 	new->async_tx.cookie = -EBUSY;
-	new->async_tx.ack = 0;
+	new->async_tx.flags = flags;
 
 	/* Insert the link descriptor to the LD ring */
 	list_add_tail(&new->node, &new->async_tx.tx_list);
@@ -482,7 +482,7 @@ static struct dma_async_tx_descriptor *fsl_dma_prep_memcpy(
 			set_desc_next(fsl_chan, &prev->hw, new->async_tx.phys);
 
 		new->async_tx.cookie = 0;
-		new->async_tx.ack = 1;
+		async_tx_ack(&new->async_tx);
 
 		prev = new;
 		len -= copy;
@@ -493,7 +493,7 @@ static struct dma_async_tx_descriptor *fsl_dma_prep_memcpy(
 		list_add_tail(&new->node, &first->async_tx.tx_list);
 	} while (len);
 
-	new->async_tx.ack = 0; /* client is in control of this ack */
+	new->async_tx.flags = flags; /* client is in control of this ack */
 	new->async_tx.cookie = -EBUSY;
 
 	/* Set End-of-link to the last link descriptor of new list*/
@@ -874,7 +874,7 @@ static int fsl_dma_self_test(struct fsl_dma_chan *fsl_chan)
 	async_tx_ack(tx3);
 
 	/* Interrupt tx test */
-	tx1 = fsl_dma_prep_interrupt(chan);
+	tx1 = fsl_dma_prep_interrupt(chan, 0);
 	async_tx_ack(tx1);
 	cookie = fsl_dma_tx_submit(tx1);
 
diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
index 1517fe4e2d14..318e8a22d814 100644
--- a/drivers/dma/ioat_dma.c
+++ b/drivers/dma/ioat_dma.c
@@ -212,14 +212,14 @@ static dma_cookie_t ioat1_tx_submit(struct dma_async_tx_descriptor *tx)
 	u32 copy;
 	size_t len;
 	dma_addr_t src, dst;
-	int orig_ack;
+	unsigned long orig_flags;
 	unsigned int desc_count = 0;
 
 	/* src and dest and len are stored in the initial descriptor */
 	len = first->len;
 	src = first->src;
 	dst = first->dst;
-	orig_ack = first->async_tx.ack;
+	orig_flags = first->async_tx.flags;
 	new = first;
 
 	spin_lock_bh(&ioat_chan->desc_lock);
@@ -228,7 +228,7 @@ static dma_cookie_t ioat1_tx_submit(struct dma_async_tx_descriptor *tx)
 	do {
 		copy = min_t(size_t, len, ioat_chan->xfercap);
 
-		new->async_tx.ack = 1;
+		async_tx_ack(&new->async_tx);
 
 		hw = new->hw;
 		hw->size = copy;
@@ -264,7 +264,7 @@ static dma_cookie_t ioat1_tx_submit(struct dma_async_tx_descriptor *tx)
 	}
 
 	new->tx_cnt = desc_count;
-	new->async_tx.ack = orig_ack; /* client is in control of this ack */
+	new->async_tx.flags = orig_flags; /* client is in control of this ack */
 
 	/* store the original values for use in later cleanup */
 	if (new != first) {
@@ -304,14 +304,14 @@ static dma_cookie_t ioat2_tx_submit(struct dma_async_tx_descriptor *tx)
 	u32 copy;
 	size_t len;
 	dma_addr_t src, dst;
-	int orig_ack;
+	unsigned long orig_flags;
 	unsigned int desc_count = 0;
 
 	/* src and dest and len are stored in the initial descriptor */
 	len = first->len;
 	src = first->src;
 	dst = first->dst;
-	orig_ack = first->async_tx.ack;
+	orig_flags = first->async_tx.flags;
 	new = first;
 
 	/*
@@ -321,7 +321,7 @@ static dma_cookie_t ioat2_tx_submit(struct dma_async_tx_descriptor *tx)
 	do {
 		copy = min_t(size_t, len, ioat_chan->xfercap);
 
-		new->async_tx.ack = 1;
+		async_tx_ack(&new->async_tx);
 
 		hw = new->hw;
 		hw->size = copy;
@@ -349,7 +349,7 @@ static dma_cookie_t ioat2_tx_submit(struct dma_async_tx_descriptor *tx)
 	}
 
 	new->tx_cnt = desc_count;
-	new->async_tx.ack = orig_ack; /* client is in control of this ack */
+	new->async_tx.flags = orig_flags; /* client is in control of this ack */
 
 	/* store the original values for use in later cleanup */
 	if (new != first) {
@@ -714,7 +714,7 @@ static struct dma_async_tx_descriptor *ioat1_dma_prep_memcpy(
 		new->len = len;
 		new->dst = dma_dest;
 		new->src = dma_src;
-		new->async_tx.ack = 0;
+		new->async_tx.flags = flags;
 		return &new->async_tx;
 	} else
 		return NULL;
@@ -742,7 +742,7 @@ static struct dma_async_tx_descriptor *ioat2_dma_prep_memcpy(
 		new->len = len;
 		new->dst = dma_dest;
 		new->src = dma_src;
-		new->async_tx.ack = 0;
+		new->async_tx.flags = flags;
 		return &new->async_tx;
 	} else
 		return NULL;
@@ -842,7 +842,7 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 				 * a completed entry, but not the last, so clean
 				 * up if the client is done with the descriptor
 				 */
-				if (desc->async_tx.ack) {
+				if (async_tx_test_ack(&desc->async_tx)) {
 					list_del(&desc->node);
 					list_add_tail(&desc->node,
 						      &ioat_chan->free_desc);
@@ -979,7 +979,7 @@ static void ioat_dma_start_null_desc(struct ioat_dma_chan *ioat_chan)
 	desc->hw->size = 0;
 	desc->hw->src_addr = 0;
 	desc->hw->dst_addr = 0;
-	desc->async_tx.ack = 1;
+	async_tx_ack(&desc->async_tx);
 	switch (ioat_chan->device->version) {
 	case IOAT_VER_1_2:
 		desc->hw->next = 0;
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index 93252294f32b..762b729672e0 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -111,7 +111,7 @@ iop_adma_clean_slot(struct iop_adma_desc_slot *desc,
 	/* the client is allowed to attach dependent operations
 	 * until 'ack' is set
 	 */
-	if (!desc->async_tx.ack)
+	if (!async_tx_test_ack(&desc->async_tx))
 		return 0;
 
 	/* leave the last descriptor in the chain
@@ -148,7 +148,7 @@ static void __iop_adma_slot_cleanup(struct iop_adma_chan *iop_chan)
 			"this_desc: %#x next_desc: %#x ack: %d\n",
 			iter->async_tx.cookie, iter->idx, busy,
 			iter->async_tx.phys, iop_desc_get_next_desc(iter),
-			iter->async_tx.ack);
+			async_tx_test_ack(&iter->async_tx));
 		prefetch(_iter);
 		prefetch(&_iter->async_tx);
 
@@ -338,9 +338,7 @@ retry:
 
 				/* pre-ack all but the last descriptor */
 				if (num_slots != slots_per_op)
-					iter->async_tx.ack = 1;
-				else
-					iter->async_tx.ack = 0;
+					async_tx_ack(&iter->async_tx);
 
 				list_add_tail(&iter->chain_node, &chain);
 				alloc_tail = iter;
@@ -513,7 +511,7 @@ static int iop_adma_alloc_chan_resources(struct dma_chan *chan)
 }
 
 static struct dma_async_tx_descriptor *
-iop_adma_prep_dma_interrupt(struct dma_chan *chan)
+iop_adma_prep_dma_interrupt(struct dma_chan *chan, unsigned long flags)
 {
 	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
 	struct iop_adma_desc_slot *sw_desc, *grp_start;
@@ -528,6 +526,7 @@ iop_adma_prep_dma_interrupt(struct dma_chan *chan)
 		grp_start = sw_desc->group_head;
 		iop_desc_init_interrupt(grp_start, iop_chan);
 		grp_start->unmap_len = 0;
+		sw_desc->async_tx.flags = flags;
 	}
 	spin_unlock_bh(&iop_chan->lock);
 
@@ -560,6 +559,7 @@ iop_adma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dma_dest,
 		iop_desc_set_memcpy_src_addr(grp_start, dma_src);
 		sw_desc->unmap_src_cnt = 1;
 		sw_desc->unmap_len = len;
+		sw_desc->async_tx.flags = flags;
 	}
 	spin_unlock_bh(&iop_chan->lock);
 
@@ -592,6 +592,7 @@ iop_adma_prep_dma_memset(struct dma_chan *chan, dma_addr_t dma_dest,
 		iop_desc_set_dest_addr(grp_start, iop_chan, dma_dest);
 		sw_desc->unmap_src_cnt = 1;
 		sw_desc->unmap_len = len;
+		sw_desc->async_tx.flags = flags;
 	}
 	spin_unlock_bh(&iop_chan->lock);
 
@@ -625,6 +626,7 @@ iop_adma_prep_dma_xor(struct dma_chan *chan, dma_addr_t dma_dest,
 		iop_desc_set_dest_addr(grp_start, iop_chan, dma_dest);
 		sw_desc->unmap_src_cnt = src_cnt;
 		sw_desc->unmap_len = len;
+		sw_desc->async_tx.flags = flags;
 		while (src_cnt--)
 			iop_desc_set_xor_src_addr(grp_start, src_cnt,
 						  dma_src[src_cnt]);
@@ -661,6 +663,7 @@ iop_adma_prep_dma_zero_sum(struct dma_chan *chan, dma_addr_t *dma_src,
 			__func__, grp_start->xor_check_result);
 		sw_desc->unmap_src_cnt = src_cnt;
 		sw_desc->unmap_len = len;
+		sw_desc->async_tx.flags = flags;
 		while (src_cnt--)
 			iop_desc_set_zero_sum_src_addr(grp_start, src_cnt,
 						       dma_src[src_cnt]);
@@ -847,11 +850,11 @@ static int __devinit iop_adma_memcpy_self_test(struct iop_adma_device *device)
 	src_dma = dma_map_single(dma_chan->device->dev, src,
 				IOP_ADMA_TEST_SIZE, DMA_TO_DEVICE);
 	tx = iop_adma_prep_dma_memcpy(dma_chan, dest_dma, src_dma,
-				      IOP_ADMA_TEST_SIZE, 1);
+				      IOP_ADMA_TEST_SIZE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 
 	cookie = iop_adma_tx_submit(tx);
 	iop_adma_issue_pending(dma_chan);
-	async_tx_ack(tx);
 	msleep(1);
 
 	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) !=
@@ -947,11 +950,11 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
 		dma_srcs[i] = dma_map_page(dma_chan->device->dev, xor_srcs[i],
 					   0, PAGE_SIZE, DMA_TO_DEVICE);
 	tx = iop_adma_prep_dma_xor(dma_chan, dest_dma, dma_srcs,
-				   IOP_ADMA_NUM_SRC_TEST, PAGE_SIZE, 1);
+				   IOP_ADMA_NUM_SRC_TEST, PAGE_SIZE,
+				   DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 
 	cookie = iop_adma_tx_submit(tx);
 	iop_adma_issue_pending(dma_chan);
-	async_tx_ack(tx);
 	msleep(8);
 
 	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) !=
@@ -994,11 +997,11 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
 					   DMA_TO_DEVICE);
 	tx = iop_adma_prep_dma_zero_sum(dma_chan, dma_srcs,
 					IOP_ADMA_NUM_SRC_TEST + 1, PAGE_SIZE,
-					&zero_sum_result, 1);
+					&zero_sum_result,
+					DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 
 	cookie = iop_adma_tx_submit(tx);
 	iop_adma_issue_pending(dma_chan);
-	async_tx_ack(tx);
 	msleep(8);
 
 	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) != DMA_SUCCESS) {
@@ -1018,11 +1021,11 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
 	/* test memset */
 	dma_addr = dma_map_page(dma_chan->device->dev, dest, 0,
 			PAGE_SIZE, DMA_FROM_DEVICE);
-	tx = iop_adma_prep_dma_memset(dma_chan, dma_addr, 0, PAGE_SIZE, 1);
+	tx = iop_adma_prep_dma_memset(dma_chan, dma_addr, 0, PAGE_SIZE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 
 	cookie = iop_adma_tx_submit(tx);
 	iop_adma_issue_pending(dma_chan);
-	async_tx_ack(tx);
 	msleep(8);
 
 	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) != DMA_SUCCESS) {
@@ -1050,11 +1053,11 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
 					   DMA_TO_DEVICE);
 	tx = iop_adma_prep_dma_zero_sum(dma_chan, dma_srcs,
 					IOP_ADMA_NUM_SRC_TEST + 1, PAGE_SIZE,
-					&zero_sum_result, 1);
+					&zero_sum_result,
+					DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 
 	cookie = iop_adma_tx_submit(tx);
 	iop_adma_issue_pending(dma_chan);
-	async_tx_ack(tx);
 	msleep(8);
 
 	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) != DMA_SUCCESS) {
@@ -1287,7 +1290,7 @@ static void iop_chan_start_null_memcpy(struct iop_adma_chan *iop_chan)
 		grp_start = sw_desc->group_head;
 
 		list_splice_init(&sw_desc->async_tx.tx_list, &iop_chan->chain);
-		sw_desc->async_tx.ack = 1;
+		async_tx_ack(&sw_desc->async_tx);
 		iop_desc_init_memcpy(grp_start, 0);
 		iop_desc_set_byte_count(grp_start, iop_chan, 0);
 		iop_desc_set_dest_addr(grp_start, iop_chan, 0);
@@ -1343,7 +1346,7 @@ static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan)
 	if (sw_desc) {
 		grp_start = sw_desc->group_head;
 		list_splice_init(&sw_desc->async_tx.tx_list, &iop_chan->chain);
-		sw_desc->async_tx.ack = 1;
+		async_tx_ack(&sw_desc->async_tx);
 		iop_desc_init_null_xor(grp_start, 2, 0);
 		iop_desc_set_byte_count(grp_start, iop_chan, 0);
 		iop_desc_set_dest_addr(grp_start, iop_chan, 0);
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index cd34df78c6aa..b4d84ed6187d 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -95,12 +95,17 @@ enum dma_transaction_type {
 #define DMA_TX_TYPE_END (DMA_INTERRUPT + 1)
 
 /**
- * enum dma_prep_flags - DMA flags to augment operation preparation
+ * enum dma_ctrl_flags - DMA flags to augment operation preparation,
+ * 	control completion, and communicate status.
  * @DMA_PREP_INTERRUPT - trigger an interrupt (callback) upon completion of
  * 	this transaction
+ * @DMA_CTRL_ACK - the descriptor cannot be reused until the client
+ * 	acknowledges receipt, i.e. has has a chance to establish any
+ * 	dependency chains
  */
-enum dma_prep_flags {
+enum dma_ctrl_flags {
 	DMA_PREP_INTERRUPT = (1 << 0),
+	DMA_CTRL_ACK = (1 << 1),
 };
 
 /**
@@ -211,8 +216,8 @@ typedef void (*dma_async_tx_callback)(void *dma_async_param);
  * ---dma generic offload fields---
  * @cookie: tracking cookie for this transaction, set to -EBUSY if
  *	this tx is sitting on a dependency list
- * @ack: the descriptor can not be reused until the client acknowledges
- *	receipt, i.e. has has a chance to establish any dependency chains
+ * @flags: flags to augment operation preparation, control completion, and
+ * 	communicate status
  * @phys: physical address of the descriptor
  * @tx_list: driver common field for operations that require multiple
  *	descriptors
@@ -227,7 +232,7 @@ typedef void (*dma_async_tx_callback)(void *dma_async_param);
  */
 struct dma_async_tx_descriptor {
 	dma_cookie_t cookie;
-	int ack;
+	enum dma_ctrl_flags flags; /* not a 'long' to pack with cookie */
 	dma_addr_t phys;
 	struct list_head tx_list;
 	struct dma_chan *chan;
@@ -290,7 +295,7 @@ struct dma_device {
 		struct dma_chan *chan, dma_addr_t dest, int value, size_t len,
 		unsigned long flags);
 	struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
-		struct dma_chan *chan);
+		struct dma_chan *chan, unsigned long flags);
 
 	enum dma_status (*device_is_tx_complete)(struct dma_chan *chan,
 			dma_cookie_t cookie, dma_cookie_t *last,
@@ -316,7 +321,13 @@ void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx,
 static inline void
 async_tx_ack(struct dma_async_tx_descriptor *tx)
 {
-	tx->ack = 1;
+	tx->flags |= DMA_CTRL_ACK;
+}
+
+static inline int
+async_tx_test_ack(struct dma_async_tx_descriptor *tx)
+{
+	return tx->flags & DMA_CTRL_ACK;
 }
 
 #define first_dma_cap(mask) __first_dma_cap(&(mask))
-- 
cgit v1.2.3


From a594eeb1a1d320981fccc29584b6f21fcebd765f Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 18 Apr 2008 00:46:20 +0200
Subject: IDE: remove ide=reverse IDE core

This option is obsolete and can be removed safely.

It allows us to remove the pci_get_device_reverse() function from the
PCI core.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 Documentation/ide/ide.txt           |  2 --
 Documentation/kernel-parameters.txt |  2 +-
 drivers/ide/Kconfig                 |  9 +--------
 drivers/ide/ide-scan-pci.c          |  9 ++-------
 drivers/ide/ide.c                   | 12 ------------
 include/linux/ide.h                 |  1 -
 6 files changed, 4 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/Documentation/ide/ide.txt b/Documentation/ide/ide.txt
index 818676aad45a..ce7c01af30f6 100644
--- a/Documentation/ide/ide.txt
+++ b/Documentation/ide/ide.txt
@@ -269,8 +269,6 @@ Summary of ide driver parameters for kernel command line
 			  ability to bit test for detection is currently
 			  unknown.
 
- "ide=reverse"		: formerly called to pci sub-system, but now local.
-
  "ide=doubler"		: probe/support IDE doublers on Amiga
 
 There may be more options than shown -- use the source, Luke!
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index dafd001bf833..228d85a3f319 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -763,7 +763,7 @@ and is between 256 and 4096 characters. It is defined in the file
 			Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
 
 	ide=		[HW] (E)IDE subsystem
-			Format: ide=nodma or ide=doubler or ide=reverse
+			Format: ide=nodma or ide=doubler
 			See Documentation/ide/ide.txt.
 
 	ide?=		[HW] (E)IDE subsystem
diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index eed6d8e1b5c7..ac5875783430 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -416,12 +416,6 @@ config BLK_DEV_OFFBOARD
 	  This can improve the usability of some boot managers such as lilo
 	  when booting from a drive on an off-board controller.
 
-	  If you say Y here, and you actually want to reverse the device scan
-	  order as explained above, you also need to issue the kernel command
-	  line option "ide=reverse". (Try "man bootparam" or see the
-	  documentation of your boot loader (lilo or loadlin) about how to
-	  pass options to the kernel at boot time.)
-
 	  Note that, if you do this, the order of the hd* devices will be
 	  rearranged which may require modification of fstab and other files.
 
@@ -615,8 +609,7 @@ config BLK_DEV_HPT366
 	  reference to device 0x80. The other solution is to say Y to "Boot
 	  off-board chipsets first support" (CONFIG_BLK_DEV_OFFBOARD) unless
 	  your mother board has the chipset natively mounted. Regardless one
-	  should use the fore mentioned option and call at LILO or include
-	  "ide=reverse" in LILO's append-line.
+	  should use the fore mentioned option and call at LILO.
 
 	  This driver requires dynamic tuning of the chipset during the
 	  ide-probe at boot. It is reported to support DVD II drives, by the
diff --git a/drivers/ide/ide-scan-pci.c b/drivers/ide/ide-scan-pci.c
index 93d2e41be853..98888da1b600 100644
--- a/drivers/ide/ide-scan-pci.c
+++ b/drivers/ide/ide-scan-pci.c
@@ -88,13 +88,8 @@ static int __init ide_scan_pcibus(void)
 	struct list_head *l, *n;
 
 	pre_init = 0;
-	if (!ide_scan_direction)
-		while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)))
-			ide_scan_pcidev(dev);
-	else
-		while ((dev = pci_get_device_reverse(PCI_ANY_ID, PCI_ANY_ID,
-						     dev)))
-			ide_scan_pcidev(dev);
+	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)))
+		ide_scan_pcidev(dev);
 
 	/*
 	 *	Hand the drivers over to the PCI layer now we
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index fc69fe2e3ec0..5f545153391a 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -90,10 +90,6 @@ static int system_bus_speed;	/* holds what we think is VESA/PCI bus speed */
 DEFINE_MUTEX(ide_cfg_mtx);
  __cacheline_aligned_in_smp DEFINE_SPINLOCK(ide_lock);
 
-#ifdef CONFIG_IDEPCI_PCIBUS_ORDER
-int ide_scan_direction; /* THIS was formerly 2.2.x pci=reverse */
-#endif
-
 int noautodma = 0;
 
 #ifdef CONFIG_BLK_DEV_IDEACPI
@@ -1217,14 +1213,6 @@ static int __init ide_setup(char *s)
 		goto obsolete_option;
 	}
 
-#ifdef CONFIG_IDEPCI_PCIBUS_ORDER
-	if (!strcmp(s, "ide=reverse")) {
-		ide_scan_direction = 1;
-		printk(" : Enabled support for IDE inverse scan order.\n");
-		goto obsolete_option;
-	}
-#endif
-
 #ifdef CONFIG_BLK_DEV_IDEACPI
 	if (!strcmp(s, "ide=noacpi")) {
 		//printk(" : Disable IDE ACPI support.\n");
diff --git a/include/linux/ide.h b/include/linux/ide.h
index bc26b2f27359..43d2968a4e13 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -990,7 +990,6 @@ extern void do_ide_request(struct request_queue *);
 void ide_init_disk(struct gendisk *, ide_drive_t *);
 
 #ifdef CONFIG_IDEPCI_PCIBUS_ORDER
-extern int ide_scan_direction;
 extern int __ide_pci_register_driver(struct pci_driver *driver, struct module *owner, const char *mod_name);
 #define ide_pci_register_driver(d) __ide_pci_register_driver(d, THIS_MODULE, KBUILD_MODNAME)
 #else
-- 
cgit v1.2.3


From 4f0eee4d877e3b617b6a22d209d52b3dfca2b2a7 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:21 +0200
Subject: ide: use ide_find_port() instead of ide_deprecated_find_port()

* Use ide_find_port() instead of ide_deprecated_find_port() in bast-ide/
  palm_bk3710/ide-cs/delkin_cb host drivers and in ide_register_hw().

* Remove no longer needed ide_deprecated_find_port().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/bast-ide.c    |  2 +-
 drivers/ide/arm/palm_bk3710.c |  2 +-
 drivers/ide/ide.c             | 27 +--------------------------
 drivers/ide/legacy/ide-cs.c   |  2 +-
 drivers/ide/pci/delkin_cb.c   |  2 +-
 include/linux/ide.h           |  1 -
 6 files changed, 5 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/arm/bast-ide.c b/drivers/ide/arm/bast-ide.c
index 161d30c8481e..a22da7ab2b57 100644
--- a/drivers/ide/arm/bast-ide.c
+++ b/drivers/ide/arm/bast-ide.c
@@ -41,7 +41,7 @@ static int __init bastide_register(unsigned int base, unsigned int aux, int irq)
 	hw.io_ports[IDE_CONTROL_OFFSET] = aux + (6 * 0x20);
 	hw.irq = irq;
 
-	hwif = ide_deprecated_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
 	if (hwif == NULL)
 		goto out;
 
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 8e1f6bd33887..0a722503c102 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -378,7 +378,7 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	hw.irq = irq->start;
 	hw.chipset = ide_palm3710;
 
-	hwif = ide_deprecated_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
 	if (hwif == NULL)
 		goto out;
 
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 67ce697a9d51..bd5ff7d987d0 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -618,31 +618,6 @@ void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
 }
 EXPORT_SYMBOL_GPL(ide_init_port_hw);
 
-ide_hwif_t *ide_deprecated_find_port(unsigned long base)
-{
-	ide_hwif_t *hwif;
-	int i;
-
-	for (i = 0; i < MAX_HWIFS; i++) {
-		hwif = &ide_hwifs[i];
-		if (hwif->io_ports[IDE_DATA_OFFSET] == base)
-			goto found;
-	}
-
-	for (i = 0; i < MAX_HWIFS; i++) {
-		hwif = &ide_hwifs[i];
-		if (hwif->hold)
-			continue;
-		if (!hwif->present && hwif->mate == NULL)
-			goto found;
-	}
-
-	hwif = NULL;
-found:
-	return hwif;
-}
-EXPORT_SYMBOL_GPL(ide_deprecated_find_port);
-
 /**
  *	ide_register_hw		-	register IDE interface
  *	@hw: hardware registers
@@ -662,7 +637,7 @@ int ide_register_hw(hw_regs_t *hw, void (*quirkproc)(ide_drive_t *),
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	do {
-		hwif = ide_deprecated_find_port(hw->io_ports[IDE_DATA_OFFSET]);
+		hwif = ide_find_port(hw->io_ports[IDE_DATA_OFFSET]);
 		if (hwif)
 			goto found;
 		for (index = 0; index < MAX_HWIFS; index++)
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index 15ccf6944ae2..de2e5944809e 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -156,7 +156,7 @@ static int idecs_register(unsigned long io, unsigned long ctl, unsigned long irq
     hw.chipset = ide_pci;
     hw.dev = &handle->dev;
 
-    hwif = ide_deprecated_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+    hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
     if (hwif == NULL)
 	return -1;
 
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index 3f9cd64c26a6..e08e13a0bb6e 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -78,7 +78,7 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 	hw.irq = dev->irq;
 	hw.chipset = ide_pci;		/* this enables IRQ sharing */
 
-	hwif = ide_deprecated_find_port(hw.io_ports[IDE_DATA_OFFSET]);
+	hwif = ide_find_port(hw.io_ports[IDE_DATA_OFFSET]);
 	if (hwif == NULL)
 		goto out_disable;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 43d2968a4e13..1b423958a894 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -186,7 +186,6 @@ typedef struct hw_regs_s {
 } hw_regs_t;
 
 struct hwif_s * ide_find_port(unsigned long);
-struct hwif_s *ide_deprecated_find_port(unsigned long);
 void ide_init_port_data(struct hwif_s *, unsigned int);
 void ide_init_port_hw(struct hwif_s *, hw_regs_t *);
 
-- 
cgit v1.2.3


From 5b0c4b30a625927340a3e7f565aa4de8b60489cc Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:22 +0200
Subject: ide: remove IDE devices from /proc/ide/ before unregistering them

IDE devices need to be removed from /proc/ide/ _before_ being unregistered:

* Drop 'ide_hwif_t *hwif' argument from destroy_proc_ide_device()
  and use drive->hwif instead.

* Rename destroy_proc_ide_device() to ide_proc_unregister_device().

* Call ide_proc_unregister_device() in drive_release_dev().

* Remove no longer needed destroy_proc_ide_drives().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c |  2 ++
 drivers/ide/ide-proc.c  | 16 ++--------------
 include/linux/ide.h     |  2 ++
 3 files changed, 6 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 9db8978ad394..dea314ce33d3 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1199,6 +1199,8 @@ static void drive_release_dev (struct device *dev)
 {
 	ide_drive_t *drive = container_of(dev, ide_drive_t, gendev);
 
+	ide_proc_unregister_device(drive);
+
 	spin_lock_irq(&ide_lock);
 	ide_remove_drive_from_hwgroup(drive);
 	kfree(drive->id);
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index bab88ca7f7ec..77025d1057b5 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -764,27 +764,16 @@ void ide_proc_port_register_devices(ide_hwif_t *hwif)
 	}
 }
 
-static void destroy_proc_ide_device(ide_hwif_t *hwif, ide_drive_t *drive)
+void ide_proc_unregister_device(ide_drive_t *drive)
 {
 	if (drive->proc) {
 		ide_remove_proc_entries(drive->proc, generic_drive_entries);
 		remove_proc_entry(drive->name, proc_ide_root);
-		remove_proc_entry(drive->name, hwif->proc);
+		remove_proc_entry(drive->name, drive->hwif->proc);
 		drive->proc = NULL;
 	}
 }
 
-static void destroy_proc_ide_drives(ide_hwif_t *hwif)
-{
-	int	d;
-
-	for (d = 0; d < MAX_DRIVES; d++) {
-		ide_drive_t *drive = &hwif->drives[d];
-		if (drive->proc)
-			destroy_proc_ide_device(hwif, drive);
-	}
-}
-
 static ide_proc_entry_t hwif_entries[] = {
 	{ "channel",	S_IFREG|S_IRUGO,	proc_ide_read_channel,	NULL },
 	{ "mate",	S_IFREG|S_IRUGO,	proc_ide_read_mate,	NULL },
@@ -816,7 +805,6 @@ EXPORT_SYMBOL_GPL(ide_pci_create_host_proc);
 void ide_proc_unregister_port(ide_hwif_t *hwif)
 {
 	if (hwif->proc) {
-		destroy_proc_ide_drives(hwif);
 		ide_remove_proc_entries(hwif->proc, hwif_entries);
 		remove_proc_entry(hwif->name, proc_ide_root);
 		hwif->proc = NULL;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 1b423958a894..f9449ecd79d9 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -690,6 +690,7 @@ void proc_ide_create(void);
 void proc_ide_destroy(void);
 void ide_proc_register_port(ide_hwif_t *);
 void ide_proc_port_register_devices(ide_hwif_t *);
+void ide_proc_unregister_device(ide_drive_t *);
 void ide_proc_unregister_port(ide_hwif_t *);
 void ide_proc_register_driver(ide_drive_t *, ide_driver_t *);
 void ide_proc_unregister_driver(ide_drive_t *, ide_driver_t *);
@@ -723,6 +724,7 @@ static inline void proc_ide_create(void) { ; }
 static inline void proc_ide_destroy(void) { ; }
 static inline void ide_proc_register_port(ide_hwif_t *hwif) { ; }
 static inline void ide_proc_port_register_devices(ide_hwif_t *hwif) { ; }
+static inline void ide_proc_unregister_device(ide_drive_t *drive) { ; }
 static inline void ide_proc_unregister_port(ide_hwif_t *hwif) { ; }
 static inline void ide_proc_register_driver(ide_drive_t *drive, ide_driver_t *driver) { ; }
 static inline void ide_proc_unregister_driver(ide_drive_t *drive, ide_driver_t *driver) { ; }
-- 
cgit v1.2.3


From 2dde7861afa23cd59db83515cb0b810b92b220aa Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:23 +0200
Subject: ide: rework PowerMac media-bay support (take 2)

Rework PowerMac media-bay support in such way that instead of
un/registering the IDE interface we un/register IDE devices:

* Add ide_port_scan() helper for probing+registerering devices on a port.

* Rename ide_port_unregister_devices() to __ide_port_unregister_devices().

* Add ide_port_unregister_devices() helper for unregistering devices on a port.

* Add 'ide_hwif_t *cd_port' to 'struct media_bay_info', pass 'hwif' instead
  of hwif->index to media_bay_set_ide_infos() and use it to setup 'cd_port'.

* Use ide_port_unregister_devices() instead of ide_unregister()
  and ide_port_scan() instead of ide_register_hw() in media_bay_step().

* Unexport ide_register_hw() and make it static.

v2:
* Fix build by adding <linux/ide.h> include to <asm-powerpc/mediabay.h>.
  (Reported by Michael/Kamalesh/Andrew).

Cc: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>
Cc: Michael Ellerman <michael@ellerman.id.au>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c        | 18 ++++++++++++++++++
 drivers/ide/ide.c              | 22 ++++++++++++++++------
 drivers/ide/ppc/pmac.c         |  3 ++-
 drivers/macintosh/mediabay.c   | 17 +++++++----------
 include/asm-powerpc/mediabay.h |  6 +++++-
 include/linux/ide.h            |  6 ++----
 6 files changed, 50 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index f3ee098b69c7..468c4ac4181d 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1494,3 +1494,21 @@ int ide_device_add(u8 idx[4], const struct ide_port_info *d)
 	return ide_device_add_all(idx_all, d);
 }
 EXPORT_SYMBOL_GPL(ide_device_add);
+
+void ide_port_scan(ide_hwif_t *hwif)
+{
+	ide_port_cable_detect(hwif);
+	ide_port_init_devices(hwif);
+
+	if (ide_probe_port(hwif) < 0)
+		return;
+
+	hwif->present = 1;
+
+	ide_port_tune_devices(hwif);
+	ide_acpi_port_init_devices(hwif);
+	ide_port_setup_devices(hwif);
+	hwif_register_devices(hwif);
+	ide_proc_port_register_devices(hwif);
+}
+EXPORT_SYMBOL_GPL(ide_port_scan);
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 47c44d15ad4a..a8b5b08dd6e2 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -502,7 +502,7 @@ void ide_remove_port_from_hwgroup(ide_hwif_t *hwif)
 }
 
 /* Called with ide_lock held. */
-static void ide_port_unregister_devices(ide_hwif_t *hwif)
+static void __ide_port_unregister_devices(ide_hwif_t *hwif)
 {
 	int i;
 
@@ -518,6 +518,18 @@ static void ide_port_unregister_devices(ide_hwif_t *hwif)
 	}
 }
 
+void ide_port_unregister_devices(ide_hwif_t *hwif)
+{
+	mutex_lock(&ide_cfg_mtx);
+	spin_lock_irq(&ide_lock);
+	__ide_port_unregister_devices(hwif);
+	hwif->present = 0;
+	ide_port_init_devices_data(hwif);
+	spin_unlock_irq(&ide_lock);
+	mutex_unlock(&ide_cfg_mtx);
+}
+EXPORT_SYMBOL_GPL(ide_port_unregister_devices);
+
 /**
  *	ide_unregister		-	free an IDE interface
  *	@index: index of interface (will change soon to a pointer)
@@ -558,7 +570,7 @@ void ide_unregister(unsigned int index, int init_default, int restore)
 	hwif = &ide_hwifs[index];
 	if (!hwif->present)
 		goto abort;
-	ide_port_unregister_devices(hwif);
+	__ide_port_unregister_devices(hwif);
 	hwif->present = 0;
 
 	spin_unlock_irq(&ide_lock);
@@ -648,8 +660,8 @@ EXPORT_SYMBOL_GPL(ide_init_port_hw);
  *	Returns -1 on error.
  */
 
-int ide_register_hw(hw_regs_t *hw, void (*quirkproc)(ide_drive_t *),
-		    ide_hwif_t **hwifp)
+static int ide_register_hw(hw_regs_t *hw, void (*quirkproc)(ide_drive_t *),
+			   ide_hwif_t **hwifp)
 {
 	int index, retry = 1;
 	ide_hwif_t *hwif;
@@ -683,8 +695,6 @@ found:
 	return hwif->present ? index : -1;
 }
 
-EXPORT_SYMBOL(ide_register_hw);
-
 /*
  *	Locks for IDE setting functionality
  */
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index d9ca52e6cdab..7889f5f5c49a 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1088,7 +1088,8 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif, hw_regs_t *hw)
 	if (np->parent && np->parent->name
 	    && strcasecmp(np->parent->name, "media-bay") == 0) {
 #ifdef CONFIG_PMAC_MEDIABAY
-		media_bay_set_ide_infos(np->parent, pmif->regbase, pmif->irq, hwif->index);
+		media_bay_set_ide_infos(np->parent, pmif->regbase, pmif->irq,
+					hwif);
 #endif /* CONFIG_PMAC_MEDIABAY */
 		pmif->mediabay = 1;
 		if (!bidp)
diff --git a/drivers/macintosh/mediabay.c b/drivers/macintosh/mediabay.c
index bd8a1d14b45d..82add26cc665 100644
--- a/drivers/macintosh/mediabay.c
+++ b/drivers/macintosh/mediabay.c
@@ -79,6 +79,7 @@ struct media_bay_info {
 	int				sleeping;
 	struct semaphore		lock;
 #ifdef CONFIG_BLK_DEV_IDE_PMAC
+	ide_hwif_t			*cd_port;
 	void __iomem			*cd_base;
 	int				cd_irq;
 	int				cd_retry;
@@ -448,7 +449,7 @@ int check_media_bay_by_base(unsigned long base, int what)
 }
 
 int media_bay_set_ide_infos(struct device_node* which_bay, unsigned long base,
-			    int irq, int index)
+			    int irq, ide_hwif_t *hwif)
 {
 	int	i;
 
@@ -456,10 +457,11 @@ int media_bay_set_ide_infos(struct device_node* which_bay, unsigned long base,
 		struct media_bay_info* bay = &media_bays[i];
 
 		if (bay->mdev && which_bay == bay->mdev->ofdev.node) {
-			int timeout = 5000;
+			int timeout = 5000, index = hwif->index;
 			
 			down(&bay->lock);
 
+			bay->cd_port	= hwif;
  			bay->cd_base	= (void __iomem *) base;
 			bay->cd_irq	= irq;
 
@@ -551,15 +553,10 @@ static void media_bay_step(int i)
 			bay->timer = 0;
 			bay->state = mb_up;
 			if (bay->cd_index < 0) {
-				hw_regs_t hw;
-
 				printk("mediabay %d, registering IDE...\n", i);
 				pmu_suspend();
-				ide_init_hwif_ports(&hw, (unsigned long) bay->cd_base, (unsigned long) 0, NULL);
-				hw.irq = bay->cd_irq;
-				hw.chipset = ide_pmac;
-				bay->cd_index =
-					ide_register_hw(&hw, NULL, NULL);
+				ide_port_scan(bay->cd_port);
+				bay->cd_index = bay->cd_port->index;
 				pmu_resume();
 			}
 			if (bay->cd_index == -1) {
@@ -589,7 +586,7 @@ static void media_bay_step(int i)
     	        if (bay->cd_index >= 0) {
 			printk(KERN_DEBUG "Unregistering mb %d ide, index:%d\n", i,
 			       bay->cd_index);
-			ide_unregister(bay->cd_index, 1, 1);
+			ide_port_unregister_devices(bay->cd_port);
 			bay->cd_index = -1;
 		}
 	    	if (bay->cd_retry) {
diff --git a/include/asm-powerpc/mediabay.h b/include/asm-powerpc/mediabay.h
index de83fe196309..df111c362a7f 100644
--- a/include/asm-powerpc/mediabay.h
+++ b/include/asm-powerpc/mediabay.h
@@ -22,10 +22,14 @@ int check_media_bay(struct device_node *which_bay, int what);
 /* Number of bays in the machine or 0 */
 extern int media_bay_count;
 
+#ifdef CONFIG_BLK_DEV_IDE_PMAC
+#include <linux/ide.h>
+
 int check_media_bay_by_base(unsigned long base, int what);
 /* called by IDE PMAC host driver to register IDE controller for media bay */
 int media_bay_set_ide_infos(struct device_node *which_bay, unsigned long base,
-			    int irq, int index);
+			    int irq, ide_hwif_t *hwif);
+#endif
 
 #endif /* __KERNEL__ */
 #endif /* _PPC_MEDIABAY_H */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f9449ecd79d9..9aaad7e70593 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -189,10 +189,6 @@ struct hwif_s * ide_find_port(unsigned long);
 void ide_init_port_data(struct hwif_s *, unsigned int);
 void ide_init_port_hw(struct hwif_s *, hw_regs_t *);
 
-struct ide_drive_s;
-int ide_register_hw(hw_regs_t *, void (*)(struct ide_drive_s *),
-		    struct hwif_s **);
-
 static inline void ide_std_init_ports(hw_regs_t *hw,
 				      unsigned long io_addr,
 				      unsigned long ctl_addr)
@@ -1204,6 +1200,8 @@ void ide_undecoded_slave(ide_drive_t *);
 
 int ide_device_add_all(u8 *idx, const struct ide_port_info *);
 int ide_device_add(u8 idx[4], const struct ide_port_info *);
+void ide_port_unregister_devices(ide_hwif_t *);
+void ide_port_scan(ide_hwif_t *);
 
 static inline void *ide_get_hwifdata (ide_hwif_t * hwif)
 {
-- 
cgit v1.2.3


From 50672e5d7486c9ab312432cbe180ac071f1de8e0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:23 +0200
Subject: ide: remove dead/obsolete ->busproc method

->busproc method is used by HDIO_SET_BUSSTATE ioctl but it has no chance
of working as intended (in 2.4.x days) because to issue an ioctl there
is a device node needed and:

- for BUSSTATE_TRISTATE+OFF it is too late (devices are already gone)

- for BUSSTATE_TRISTATE+ON it is too early (devices are not registered yet)

Just remove ->busproc method for now (it was only implemented by hpt366,
siimage and tc86c001 host drivers).

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide.c          |  3 ---
 drivers/ide/pci/hpt366.c   | 59 ----------------------------------------------
 drivers/ide/pci/sgiioc4.c  |  1 -
 drivers/ide/pci/siimage.c  | 43 ---------------------------------
 drivers/ide/pci/tc86c001.c | 36 ----------------------------
 include/linux/ide.h        |  2 --
 6 files changed, 144 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index a8b5b08dd6e2..cb18ba8de22d 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -414,7 +414,6 @@ static void ide_hwif_restore(ide_hwif_t *hwif, ide_hwif_t *tmp_hwif)
 	hwif->resetproc			= tmp_hwif->resetproc;
 	hwif->maskproc			= tmp_hwif->maskproc;
 	hwif->quirkproc			= tmp_hwif->quirkproc;
-	hwif->busproc			= tmp_hwif->busproc;
 
 	hwif->ata_input_data		= tmp_hwif->ata_input_data;
 	hwif->ata_output_data		= tmp_hwif->ata_output_data;
@@ -1071,8 +1070,6 @@ int generic_ide_ioctl(ide_drive_t *drive, struct file *file, struct block_device
 		case HDIO_SET_BUSSTATE:
 			if (!capable(CAP_SYS_ADMIN))
 				return -EACCES;
-			if (HWIF(drive)->busproc)
-				return HWIF(drive)->busproc(drive, (int)arg);
 			return -EOPNOTSUPP;
 		default:
 			return -EINVAL;
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 6357bb6269ab..d03a231d965e 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -929,64 +929,6 @@ static void hpt3xxn_rw_disk(ide_drive_t *drive, struct request *rq)
 	hpt3xxn_set_clock(HWIF(drive), rq_data_dir(rq) ? 0x23 : 0x21);
 }
 
-/* 
- * Set/get power state for a drive.
- * NOTE: affects both drives on each channel.
- *
- * When we turn the power back on, we need to re-initialize things.
- */
-#define TRISTATE_BIT  0x8000
-
-static int hpt3xx_busproc(ide_drive_t *drive, int state)
-{
-	ide_hwif_t *hwif	= HWIF(drive);
-	struct pci_dev *dev	= to_pci_dev(hwif->dev);
-	u8  mcr_addr		= hwif->select_data + 2;
-	u8  resetmask		= hwif->channel ? 0x80 : 0x40;
-	u8  bsr2		= 0;
-	u16 mcr			= 0;
-
-	hwif->bus_state = state;
-
-	/* Grab the status. */
-	pci_read_config_word(dev, mcr_addr, &mcr);
-	pci_read_config_byte(dev, 0x59, &bsr2);
-
-	/*
-	 * Set the state. We don't set it if we don't need to do so.
-	 * Make sure that the drive knows that it has failed if it's off.
-	 */
-	switch (state) {
-	case BUSSTATE_ON:
-		if (!(bsr2 & resetmask))
-			return 0;
-		hwif->drives[0].failures = hwif->drives[1].failures = 0;
-
-		pci_write_config_byte(dev, 0x59, bsr2 & ~resetmask);
-		pci_write_config_word(dev, mcr_addr, mcr & ~TRISTATE_BIT);
-		return 0;
-	case BUSSTATE_OFF:
-		if ((bsr2 & resetmask) && !(mcr & TRISTATE_BIT))
-			return 0;
-		mcr &= ~TRISTATE_BIT;
-		break;
-	case BUSSTATE_TRISTATE:
-		if ((bsr2 & resetmask) &&  (mcr & TRISTATE_BIT))
-			return 0;
-		mcr |= TRISTATE_BIT;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	hwif->drives[0].failures = hwif->drives[0].max_failures + 1;
-	hwif->drives[1].failures = hwif->drives[1].max_failures + 1;
-
-	pci_write_config_word(dev, mcr_addr, mcr);
-	pci_write_config_byte(dev, 0x59, bsr2 | resetmask);
-	return 0;
-}
-
 /**
  *	hpt37x_calibrate_dpll	-	calibrate the DPLL
  *	@dev: PCI device
@@ -1334,7 +1276,6 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
 
 	hwif->quirkproc		= &hpt3xx_quirkproc;
 	hwif->maskproc		= &hpt3xx_maskproc;
-	hwif->busproc		= &hpt3xx_busproc;
 
 	hwif->udma_filter	= &hpt3xx_udma_filter;
 	hwif->mdma_filter	= &hpt3xx_mdma_filter;
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 054626497be4..9046a69117ff 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -562,7 +562,6 @@ ide_init_sgiioc4(ide_hwif_t * hwif)
 						clear interrupts */
 	hwif->maskproc = &sgiioc4_maskproc;	/* Mask on/off NIEN register */
 	hwif->quirkproc = NULL;
-	hwif->busproc = NULL;
 
 	hwif->INB = &sgiioc4_INB;
 
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 8d624afe8529..b6be1b45f329 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -369,48 +369,6 @@ static int siimage_mmio_ide_dma_test_irq (ide_drive_t *drive)
 	return 0;
 }
 
-/**
- *	sil_sata_busproc	-	bus isolation IOCTL
- *	@drive: drive to isolate/restore
- *	@state: bus state to set
- *
- *	Used by the SII3112 to handle bus isolation. As this is a 
- *	SATA controller the work required is quite limited, we 
- *	just have to clean up the statistics
- */
-
-static int sil_sata_busproc(ide_drive_t * drive, int state)
-{
-	ide_hwif_t *hwif	= HWIF(drive);
-	struct pci_dev *dev	= to_pci_dev(hwif->dev);
-	u32 stat_config		= 0;
-	unsigned long addr	= siimage_selreg(hwif, 0);
-
-	if (hwif->mmio)
-		stat_config = readl((void __iomem *)addr);
-	else
-		pci_read_config_dword(dev, addr, &stat_config);
-
-	switch (state) {
-		case BUSSTATE_ON:
-			hwif->drives[0].failures = 0;
-			hwif->drives[1].failures = 0;
-			break;
-		case BUSSTATE_OFF:
-			hwif->drives[0].failures = hwif->drives[0].max_failures + 1;
-			hwif->drives[1].failures = hwif->drives[1].max_failures + 1;
-			break;
-		case BUSSTATE_TRISTATE:
-			hwif->drives[0].failures = hwif->drives[0].max_failures + 1;
-			hwif->drives[1].failures = hwif->drives[1].max_failures + 1;
-			break;
-		default:
-			return -EINVAL;
-	}
-	hwif->bus_state = state;
-	return 0;
-}
-
 /**
  *	sil_sata_reset_poll	-	wait for SATA reset
  *	@drive: drive we are resetting
@@ -818,7 +776,6 @@ static void __devinit init_hwif_siimage(ide_hwif_t *hwif)
 	if (sata) {
 		static int first = 1;
 
-		hwif->busproc = &sil_sata_busproc;
 		hwif->reset_poll = &sil_sata_reset_poll;
 		hwif->pre_reset = &sil_sata_pre_reset;
 		hwif->udma_filter = &sil_sata_udma_filter;
diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index 2ef2ed2f2b32..1e4a6262bcef 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -126,40 +126,6 @@ static void tc86c001_dma_start(ide_drive_t *drive)
 	ide_dma_start(drive);
 }
 
-static int tc86c001_busproc(ide_drive_t *drive, int state)
-{
-	ide_hwif_t *hwif	= HWIF(drive);
-	unsigned long sc_base	= hwif->config_data;
-	u16 scr1;
-
-	/* System Control 1 Register bit 11 (ATA Hard Reset) read */
-	scr1 = inw(sc_base + 0x00);
-
-	switch (state) {
-		case BUSSTATE_ON:
-			if (!(scr1 & 0x0800))
-				return 0;
-			scr1 &= ~0x0800;
-
-			hwif->drives[0].failures = hwif->drives[1].failures = 0;
-			break;
-		case BUSSTATE_OFF:
-			if (scr1 & 0x0800)
-				return 0;
-			scr1 |= 0x0800;
-
-			hwif->drives[0].failures = hwif->drives[0].max_failures + 1;
-			hwif->drives[1].failures = hwif->drives[1].max_failures + 1;
-			break;
-		default:
-			return -EINVAL;
-	}
-
-	/* System Control 1 Register bit 11 (ATA Hard Reset) write */
-	outw(scr1, sc_base + 0x00);
-	return 0;
-}
-
 static u8 __devinit tc86c001_cable_detect(ide_hwif_t *hwif)
 {
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
@@ -194,8 +160,6 @@ static void __devinit init_hwif_tc86c001(ide_hwif_t *hwif)
 	hwif->set_pio_mode = &tc86c001_set_pio_mode;
 	hwif->set_dma_mode = &tc86c001_set_mode;
 
-	hwif->busproc	= &tc86c001_busproc;
-
 	hwif->cable_detect = tc86c001_cable_detect;
 
 	if (!hwif->dma_base)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 9aaad7e70593..e43570a19200 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -502,8 +502,6 @@ typedef struct hwif_s {
 	void	(*maskproc)(ide_drive_t *, int);
 	/* check host's drive quirk list */
 	void	(*quirkproc)(ide_drive_t *);
-	/* driver soft-power interface */
-	int	(*busproc)(ide_drive_t *, int);
 #endif
 	u8 (*mdma_filter)(ide_drive_t *);
 	u8 (*udma_filter)(ide_drive_t *);
-- 
cgit v1.2.3


From f74c91413ec6140ee0553180c5f56fdd27c22a2e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:23 +0200
Subject: ide: add warm-plug support for IDE devices (take 2)

* Add 'struct class ide_port_class' ('ide_port' class) and a 'struct
  device *portdev' ('ide_port' class device) in ide_hwif_t.

* Register 'ide_port' class in ide_init() and unregister it in
  cleanup_module().

* Create ->portdev in ide_register_port () and unregister it in
  ide_unregister().

* Add "delete_devices" class device attribute for unregistering IDE devices
  on a port and "scan" one for probing+registering IDE devices on a port.

* Add ide_sysfs_register_port() helper for registering "delete_devices"
  and "scan" attributes with ->portdev.  Call it in ide_device_add_all().

* Document IDE warm-plug support in Documentation/ide/warm-plug-howto.txt.

v2:
* Convert patch from using 'struct class_device' to use 'struct device'.
  (thanks to Kay Sievers for doing it)

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 Documentation/ide/warm-plug-howto.txt | 13 +++++++
 drivers/ide/ide-probe.c               | 71 ++++++++++++++++++++++++++++++++++-
 drivers/ide/ide.c                     | 24 ++++++++++++
 include/linux/ide.h                   |  5 ++-
 4 files changed, 110 insertions(+), 3 deletions(-)
 create mode 100644 Documentation/ide/warm-plug-howto.txt

(limited to 'include')

diff --git a/Documentation/ide/warm-plug-howto.txt b/Documentation/ide/warm-plug-howto.txt
new file mode 100644
index 000000000000..d5885468b072
--- /dev/null
+++ b/Documentation/ide/warm-plug-howto.txt
@@ -0,0 +1,13 @@
+
+IDE warm-plug HOWTO
+===================
+
+To warm-plug devices on a port 'idex':
+
+# echo -n "1" > /sys/class/ide_port/idex/delete_devices
+
+unplug old device(s) and plug new device(s)
+
+# echo -n "1" > /sys/class/ide_port/idex/scan
+
+done
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 468c4ac4181d..510254ab3c9b 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -623,7 +623,7 @@ static void hwif_release_dev (struct device *dev)
 	complete(&hwif->gendev_rel_comp);
 }
 
-static void ide_register_port(ide_hwif_t *hwif)
+static int ide_register_port(ide_hwif_t *hwif)
 {
 	int ret;
 
@@ -639,9 +639,23 @@ static void ide_register_port(ide_hwif_t *hwif)
 	}
 	hwif->gendev.release = hwif_release_dev;
 	ret = device_register(&hwif->gendev);
-	if (ret < 0)
+	if (ret < 0) {
 		printk(KERN_WARNING "IDE: %s: device_register error: %d\n",
 			__FUNCTION__, ret);
+		goto out;
+	}
+
+	get_device(&hwif->gendev);
+
+	hwif->portdev = device_create(ide_port_class, &hwif->gendev,
+				      MKDEV(0, 0), hwif->name);
+	if (IS_ERR(hwif->portdev)) {
+		ret = PTR_ERR(hwif->portdev);
+		device_unregister(&hwif->gendev);
+	}
+	dev_set_drvdata(hwif->portdev, hwif);
+out:
+	return ret;
 }
 
 /**
@@ -1378,6 +1392,58 @@ static void ide_port_cable_detect(ide_hwif_t *hwif)
 	}
 }
 
+static ssize_t store_delete_devices(struct device *portdev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t n)
+{
+	ide_hwif_t *hwif = dev_get_drvdata(portdev);
+
+	if (strncmp(buf, "1", n))
+		return -EINVAL;
+
+	ide_port_unregister_devices(hwif);
+
+	return n;
+};
+
+static DEVICE_ATTR(delete_devices, S_IWUSR, NULL, store_delete_devices);
+
+static ssize_t store_scan(struct device *portdev,
+			  struct device_attribute *attr,
+			  const char *buf, size_t n)
+{
+	ide_hwif_t *hwif = dev_get_drvdata(portdev);
+
+	if (strncmp(buf, "1", n))
+		return -EINVAL;
+
+	ide_port_unregister_devices(hwif);
+	ide_port_scan(hwif);
+
+	return n;
+};
+
+static DEVICE_ATTR(scan, S_IWUSR, NULL, store_scan);
+
+static struct device_attribute *ide_port_attrs[] = {
+	&dev_attr_delete_devices,
+	&dev_attr_scan,
+	NULL
+};
+
+static int ide_sysfs_register_port(ide_hwif_t *hwif)
+{
+	int i, rc;
+
+	for (i = 0; ide_port_attrs[i]; i++) {
+		rc = device_create_file(hwif->portdev, ide_port_attrs[i]);
+		if (rc)
+			break;
+	}
+
+	return rc;
+}
+
 int ide_device_add_all(u8 *idx, const struct ide_port_info *d)
 {
 	ide_hwif_t *hwif, *mate = NULL;
@@ -1474,6 +1540,7 @@ int ide_device_add_all(u8 *idx, const struct ide_port_info *d)
 		hwif = &ide_hwifs[idx[i]];
 
 		if (hwif->present) {
+			ide_sysfs_register_port(hwif);
 			ide_proc_register_port(hwif);
 			ide_proc_port_register_devices(hwif);
 		}
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index cb18ba8de22d..d791b1ffb586 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -78,6 +78,8 @@
 /* default maximum number of failures */
 #define IDE_DEFAULT_MAX_FAILURES 	1
 
+struct class *ide_port_class;
+
 static const u8 ide_hwif_to_major[] = { IDE0_MAJOR, IDE1_MAJOR,
 					IDE2_MAJOR, IDE3_MAJOR,
 					IDE4_MAJOR, IDE5_MAJOR,
@@ -591,6 +593,7 @@ void ide_unregister(unsigned int index, int init_default, int restore)
 
 	ide_remove_port_from_hwgroup(hwif);
 
+	device_unregister(hwif->portdev);
 	device_unregister(&hwif->gendev);
 	wait_for_completion(&hwif->gendev_rel_comp);
 
@@ -1590,6 +1593,13 @@ struct bus_type ide_bus_type = {
 
 EXPORT_SYMBOL_GPL(ide_bus_type);
 
+static void ide_port_class_release(struct device *portdev)
+{
+	ide_hwif_t *hwif = dev_get_drvdata(portdev);
+
+	put_device(&hwif->gendev);
+}
+
 /*
  * This is gets invoked once during initialization, to set *everything* up
  */
@@ -1610,11 +1620,23 @@ static int __init ide_init(void)
 		return ret;
 	}
 
+	ide_port_class = class_create(THIS_MODULE, "ide_port");
+	if (IS_ERR(ide_port_class)) {
+		ret = PTR_ERR(ide_port_class);
+		goto out_port_class;
+	}
+	ide_port_class->dev_release = ide_port_class_release;
+
 	init_ide_data();
 
 	proc_ide_create();
 
 	return 0;
+
+out_port_class:
+	bus_unregister(&ide_bus_type);
+
+	return ret;
 }
 
 #ifdef MODULE
@@ -1651,6 +1673,8 @@ void __exit cleanup_module (void)
 
 	proc_ide_destroy();
 
+	class_destroy(ide_port_class);
+
 	bus_unregister(&ide_bus_type);
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e43570a19200..9cebf3054080 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -579,7 +579,9 @@ typedef struct hwif_s {
 	unsigned	mmio       : 1; /* host uses MMIO */
 	unsigned	straight8  : 1;	/* Alan's straight 8 check */
 
-	struct device	gendev;
+	struct device		gendev;
+	struct device		*portdev;
+
 	struct completion gendev_rel_comp; /* To deal with device release() */
 
 	void		*hwif_data;	/* extra hwif data */
@@ -1275,6 +1277,7 @@ extern struct mutex ide_cfg_mtx;
 #define local_irq_set(flags)	do { local_save_flags((flags)); local_irq_enable_in_hardirq(); } while (0)
 
 extern struct bus_type ide_bus_type;
+extern struct class *ide_port_class;
 
 /* check if CACHE FLUSH (EXT) command is supported (bits defined in ATA-6) */
 #define ide_id_has_flush_cache(id)	((id)->cfs_enable_2 & 0x3000)
-- 
cgit v1.2.3


From 9a0e77f28b50128df0c9e26ae489e44e29a7270a Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:24 +0200
Subject: ide: remove obsoleted "idex=base[,ctl[,irq]]" kernel parameters (take
 2)

* Remove obsoleted "idex=base[,ctl[,irq]]" kernel parameters
  and update Documentation/ide/ide.txt.

* Remove no longer needed ide_forced chipset type.

v2:
* is_chipset_set[] -> is_chipset_set in ide.c.

* Documentation/ide/ide.txt fix.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 Documentation/ide/ide.txt | 33 +--------------------------------
 drivers/ide/ide-generic.c |  3 +--
 drivers/ide/ide-probe.c   |  3 +--
 drivers/ide/ide-proc.c    |  3 ---
 drivers/ide/ide.c         | 41 ++++++++++++-----------------------------
 drivers/ide/pci/cmd640.c  |  2 +-
 drivers/ide/setup-pci.c   | 11 -----------
 include/linux/ide.h       |  2 +-
 8 files changed, 17 insertions(+), 81 deletions(-)

(limited to 'include')

diff --git a/Documentation/ide/ide.txt b/Documentation/ide/ide.txt
index ae4f4f43c25c..18c02df2f78f 100644
--- a/Documentation/ide/ide.txt
+++ b/Documentation/ide/ide.txt
@@ -71,29 +71,6 @@ This driver automatically probes for most IDE interfaces (including all PCI
 ones), for the drives/geometries attached to those interfaces, and for the IRQ
 lines being used by the interfaces (normally 14, 15 for ide0/ide1).
 
-For special cases, interfaces may be specified using kernel "command line"
-options.  For example,
-
-	ide3=0x168,0x36e,10	/* ioports 0x168-0x16f,0x36e, irq 10 */
-
-Normally the irq number need not be specified, as ide.c will probe for it:
-
-	ide3=0x168,0x36e	/* ioports 0x168-0x16f,0x36e */
-
-The standard port, and irq values are these:
-
-	ide0=0x1f0,0x3f6,14
-	ide1=0x170,0x376,15
-	ide2=0x1e8,0x3ee,11
-	ide3=0x168,0x36e,10
-
-Note that the first parameter reserves 8 contiguous ioports, whereas the
-second value denotes a single ioport. If in doubt, do a 'cat /proc/ioports'.
-
-In all probability the device uses these ports and IRQs if it is attached
-to the appropriate ide channel.  Pass the parameter for the correct ide
-channel to the kernel, as explained above.
-
 Any number of interfaces may share a single IRQ if necessary, at a slight
 performance penalty, whether on separate cards or a single VLB card.
 The IDE driver automatically detects and handles this.  However, this may
@@ -199,7 +176,7 @@ When ide.c is used as a module, you can pass command line parameters to the
 driver using the "options=" keyword to insmod, while replacing any ',' with
 ';'.  For example:
 
-	insmod ide.o options="ide0=serialize ide1=serialize ide2=0x1e8;0x3ee;11"
+	insmod ide.o options="hda=nodma hdb=nodma"
 
 
 ================================================================================
@@ -240,14 +217,6 @@ Summary of ide driver parameters for kernel command line
 			  As for VLB, it is safest to not specify it.
 			  Bigger values are safer than smaller ones.
 
- "idex=base"		: probe for an interface at the addr specified,
-			  where "base" is usually 0x1f0 or 0x170
-			  and "ctl" is assumed to be "base"+0x206
-
- "idex=base,ctl"	: specify both base and ctl
-
- "idex=base,ctl,irq"	: specify base, ctl, and irq number
-
  "idex=serialize"	: do not overlap operations on idex. Please note
 			  that you will have to specify this option for
 			  both the respective primary and secondary channel
diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index 387574fe4b2b..bae41459192d 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -93,8 +93,7 @@ static int __init ide_generic_init(void)
 		ide_hwif_t *hwif = &ide_hwifs[i];
 
 		if (hwif->io_ports[IDE_DATA_OFFSET] &&
-		    (hwif->chipset == ide_unknown ||
-		     hwif->chipset == ide_forced))
+		    hwif->chipset == ide_unknown)
 			idx[i] = i;
 		else
 			idx[i] = 0xff;
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 8ef5194f6d47..33cb5e5a249b 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1519,8 +1519,7 @@ int ide_device_add_all(u8 *idx, const struct ide_port_info *d)
 		hwif = &ide_hwifs[idx[i]];
 
 		if (hwif->present) {
-			if (hwif->chipset == ide_unknown ||
-			    hwif->chipset == ide_forced)
+			if (hwif->chipset == ide_unknown)
 				hwif->chipset = ide_generic;
 			hwif_register_devices(hwif);
 		}
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index 77025d1057b5..edd7f186dc4d 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -46,9 +46,6 @@ static int proc_ide_read_imodel
 	int		len;
 	const char	*name;
 
-	/*
-	 * Neither ide_unknown nor ide_forced should be set at this point.
-	 */
 	switch (hwif->chipset) {
 		case ide_generic:	name = "generic";	break;
 		case ide_pci:		name = "pci";		break;
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 96126b3b12e9..7e789c97a8b8 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -1170,7 +1170,7 @@ extern int probe_ht6560b;
 extern int probe_qd65xx;
 extern int cmd640_vlb;
 
-static int __initdata is_chipset_set[MAX_HWIFS];
+static int __initdata is_chipset_set;
 
 /*
  * ide_setup() gets called VERY EARLY during initialization,
@@ -1328,8 +1328,6 @@ static int __init ide_setup(char *s)
 			"minus10", "four", "qd65xx", "ht6560b", "cmd640_vlb",
 			"dtc2278", "umc8672", "ali14xx", NULL };
 
-		hw_regs_t hwregs;
-
 		hw = s[3] - '0';
 		hwif = &ide_hwifs[hw];
 		i = match_parm(&s[4], ide_words, vals, 3);
@@ -1338,19 +1336,14 @@ static int __init ide_setup(char *s)
 		 * Cryptic check to ensure chipset not already set for hwif.
 		 * Note: we can't depend on hwif->chipset here.
 		 */
-		if ((i >= -18 && i <= -11) || (i > 0 && i <= 3)) {
+		if (i >= -18 && i <= -11) {
 			/* chipset already specified */
-			if (is_chipset_set[hw])
+			if (is_chipset_set)
 				goto bad_option;
-			if (i > -18 && i <= -11) {
-				/* these drivers are for "ide0=" only */
-				if (hw != 0)
-					goto bad_hwif;
-				/* chipset already specified for 2nd port */
-				if (is_chipset_set[hw+1])
-					goto bad_option;
-			}
-			is_chipset_set[hw] = 1;
+			/* these drivers are for "ide0=" only */
+			if (hw != 0)
+				goto bad_hwif;
+			is_chipset_set = 1;
 			printk("\n");
 		}
 
@@ -1430,21 +1423,11 @@ static int __init ide_setup(char *s)
 			case -1: /* "noprobe" */
 				hwif->noprobe = 1;
 				goto obsolete_option;
-
-			case 1:	/* base */
-				vals[1] = vals[0] + 0x206; /* default ctl */
-			case 2: /* base,ctl */
-				vals[2] = 0;	/* default irq = probe for it */
-			case 3: /* base,ctl,irq */
-				memset(&hwregs, 0, sizeof(hwregs));
-				ide_init_hwif_ports(&hwregs, vals[0], vals[1], &hwif->irq);
-				memcpy(hwif->io_ports, hwregs.io_ports, sizeof(hwif->io_ports));
-				hwif->irq      = vals[2];
-				hwif->noprobe  = 0;
-				hwif->chipset  = ide_forced;
-				goto obsolete_option;
-
-			case 0: goto bad_option;
+			case 0:
+			case 1:
+			case 2:
+			case 3:
+				goto bad_option;
 			default:
 				printk(" -- SUPPORT NOT CONFIGURED IN THIS KERNEL\n");
 				return 1;
diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c
index 29fbc5ead03b..58a95f62e383 100644
--- a/drivers/ide/pci/cmd640.c
+++ b/drivers/ide/pci/cmd640.c
@@ -415,7 +415,7 @@ static void __init setup_device_ptrs (void)
 	cmd_hwif1 = &ide_hwifs[1]; /* default, if not found below */
 	for (i = 0; i < MAX_HWIFS; i++) {
 		ide_hwif_t *hwif = &ide_hwifs[i];
-		if (hwif->chipset == ide_unknown || hwif->chipset == ide_forced) {
+		if (hwif->chipset == ide_unknown) {
 			if (hwif->io_ports[IDE_DATA_OFFSET] == 0x1f0)
 				cmd_hwif0 = hwif;
 			else if (hwif->io_ports[IDE_DATA_OFFSET] == 0x170)
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 634e3f6a9608..ea66c996e4ec 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -40,17 +40,6 @@ static ide_hwif_t *ide_match_hwif(unsigned long io_base, u8 bootable, const char
 	int h;
 	ide_hwif_t *hwif;
 
-	/*
-	 * Look for a hwif with matching io_base specified using
-	 * parameters to ide_setup().
-	 */
-	for (h = 0; h < MAX_HWIFS; ++h) {
-		hwif = &ide_hwifs[h];
-		if (hwif->io_ports[IDE_DATA_OFFSET] == io_base) {
-			if (hwif->chipset == ide_forced)
-				return hwif; /* a perfect match */
-		}
-	}
 	/*
 	 * Look for a hwif with matching io_base default value.
 	 * If chipset is "ide_unknown", then claim that hwif slot.
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 9cebf3054080..67f83c60845f 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -169,7 +169,7 @@ enum {		ide_unknown,	ide_generic,	ide_pci,
 		ide_rz1000,	ide_trm290,
 		ide_cmd646,	ide_cy82c693,	ide_4drives,
 		ide_pmac,	ide_etrax100,	ide_acorn,
-		ide_au1xxx,	ide_palm3710,	ide_forced
+		ide_au1xxx,	ide_palm3710
 };
 
 typedef u8 hwif_chipset_t;
-- 
cgit v1.2.3


From 93de00fd1c70e1a23a73a865e0f9abfe74a7a719 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:24 +0200
Subject: ide: remove broken/dangerous HDIO_[UNREGISTER,SCAN]_HWIF ioctls (take
 3)

hdparm explicitely marks HDIO_[UNREGISTER,SCAN]_HWIF ioctls as DANGEROUS
and given the number of bugs we can assume that there are no real users:

* DMA has no chance of working because DMA resources are released by
  ide_unregister() and they are never allocated again.

* Since ide_init_hwif_ports() is used for ->io_ports[] setup the ioctls
  don't work for almost all hosts with "non-standard" (== non ISA-like)
  layout of IDE taskfile registers (there is a lot of such host drivers).

* ide_port_init_devices() is not called when probing IDE devices so:
  - drive->autotune is never set and IDE host/devices are not programmed
    for the correct PIO/DMA transfer modes (=> possible data corruption)
  - host specific I/O 32-bit and IRQ unmasking settings are not applied
    (=> possible data corruption)
  - host specific ->port_init_devs method is not called (=> no luck with
    ht6560b, qd65xx and opti621 host drivers)

* ->rw_disk method is not preserved (=> no HPT3xxN chipsets support).

* ->serialized flag is not preserved (=> possible data corruption when
   using icside, aec62xx (ATP850UF chipset), cmd640, cs5530, hpt366
   (HPT3xxN chipsets), rz1000, sc1200, dtc2278 and ht6560b host drivers).

* ->ack_intr method is not preserved (=> needed by ide-cris, buddha,
  gayle and macide host drivers).

* ->sata_scr[] and sata_misc[] is cleared by ide_unregister() and it
  isn't initialized again (SiI3112 support needs them).

* To issue an ioctl() there need to be at least one IDE device present
  in the system.

* ->cable_detect method is not preserved + it is not called when probing
  IDE devices so cable detection is broken (however since DMA support is
  also broken it doesn't really matter ;-).

* Some objects which may have already been freed in ide_unregister()
  are restored by ide_hwif_restore() (i.e. ->hwgroup).

* ide_register_hw() may unregister unrelated IDE ports if free ide_hwifs[]
  slot cannot be found.

* When IDE host drivers are modular unregistered port may be re-used by
  different host driver that owned it first causing subtle bugs.

Since we now have a proper warm-plug support remove these ioctls,
then remove no longer needed:
- ide_register_hw() and ide_hwif_restore() functions
- 'init_default' and 'restore' arguments of ide_unregister()
- zeroeing of hwif->{dma,extra}_* fields in ide_unregister()

As an added bonus IDE core code size shrinks by ~3kB (x86-32).

v2:
* fix ide_unregister() arguments in cleanup_module() (Andrew Morton).

v3:
* fix ide_unregister() arguments in palm_bk3710.c.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 block/compat_ioctl.c              |   1 -
 drivers/ide/arm/bast-ide.c        |   2 +-
 drivers/ide/arm/palm_bk3710.c     |   2 +-
 drivers/ide/arm/rapide.c          |   2 +-
 drivers/ide/ide-pnp.c             |   2 +-
 drivers/ide/ide.c                 | 199 +-------------------------------------
 drivers/ide/legacy/ide-cs.c       |   4 +-
 drivers/ide/legacy/ide_platform.c |   2 +-
 drivers/ide/mips/au1xxx-ide.c     |   2 +-
 drivers/ide/pci/delkin_cb.c       |   4 +-
 drivers/ide/pci/scc_pata.c        |   2 +-
 include/linux/hdreg.h             |   4 +-
 include/linux/ide.h               |   2 +-
 13 files changed, 19 insertions(+), 209 deletions(-)

(limited to 'include')

diff --git a/block/compat_ioctl.c b/block/compat_ioctl.c
index b73373216b0e..c70d0b6f666f 100644
--- a/block/compat_ioctl.c
+++ b/block/compat_ioctl.c
@@ -624,7 +624,6 @@ static int compat_blkdev_driver_ioctl(struct inode *inode, struct file *file,
 	case HDIO_GET_IDENTITY:
 	case HDIO_DRIVE_TASK:
 	case HDIO_DRIVE_CMD:
-	case HDIO_SCAN_HWIF:
 	/* 0x330 is reserved -- it used to be HDIO_GETGEO_BIG */
 	case 0x330:
 	/* 0x02 -- Floppy ioctls */
diff --git a/drivers/ide/arm/bast-ide.c b/drivers/ide/arm/bast-ide.c
index a22da7ab2b57..d2196436788b 100644
--- a/drivers/ide/arm/bast-ide.c
+++ b/drivers/ide/arm/bast-ide.c
@@ -48,7 +48,7 @@ static int __init bastide_register(unsigned int base, unsigned int aux, int irq)
 	i = hwif->index;
 
 	if (hwif->present)
-		ide_unregister(i, 0, 0);
+		ide_unregister(i);
 	else if (!hwif->hold)
 		ide_init_port_data(hwif, i);
 
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 0a722503c102..73899ef4ab0d 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -385,7 +385,7 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	i = hwif->index;
 
 	if (hwif->present)
-		ide_unregister(i, 0, 0);
+		ide_unregister(i);
 	else if (!hwif->hold)
 		ide_init_port_data(hwif, i);
 
diff --git a/drivers/ide/arm/rapide.c b/drivers/ide/arm/rapide.c
index efba00d2fc37..b30adcf321c3 100644
--- a/drivers/ide/arm/rapide.c
+++ b/drivers/ide/arm/rapide.c
@@ -76,7 +76,7 @@ static void __devexit rapide_remove(struct expansion_card *ec)
 
 	ecard_set_drvdata(ec, NULL);
 
-	ide_unregister(hwif->index, 0, 0);
+	ide_unregister(hwif->index);
 
 	ecard_release_resources(ec);
 }
diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c
index c14bb5380c25..34c2ad36ce54 100644
--- a/drivers/ide/ide-pnp.c
+++ b/drivers/ide/ide-pnp.c
@@ -62,7 +62,7 @@ static void idepnp_remove(struct pnp_dev * dev)
 	ide_hwif_t *hwif = pnp_get_drvdata(dev);
 
 	if (hwif)
-		ide_unregister(hwif->index, 0, 0);
+		ide_unregister(hwif->index);
 	else
 		printk(KERN_ERR "idepnp: Unable to remove device, please report.\n");
 }
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 7e789c97a8b8..1121d9cb2a9b 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -362,107 +362,6 @@ void ide_hwif_release_regions(ide_hwif_t *hwif)
 			release_region(hwif->io_ports[i], 1);
 }
 
-/**
- *	ide_hwif_restore	-	restore hwif to template
- *	@hwif: hwif to update
- *	@tmp_hwif: template
- *
- *	Restore hwif to a previous state by copying most settings
- *	from the template.
- */
-
-static void ide_hwif_restore(ide_hwif_t *hwif, ide_hwif_t *tmp_hwif)
-{
-	hwif->hwgroup			= tmp_hwif->hwgroup;
-
-	hwif->gendev.parent		= tmp_hwif->gendev.parent;
-
-	hwif->proc			= tmp_hwif->proc;
-
-	hwif->major			= tmp_hwif->major;
-	hwif->straight8			= tmp_hwif->straight8;
-	hwif->bus_state			= tmp_hwif->bus_state;
-
-	hwif->host_flags		= tmp_hwif->host_flags;
-
-	hwif->pio_mask			= tmp_hwif->pio_mask;
-
-	hwif->ultra_mask		= tmp_hwif->ultra_mask;
-	hwif->mwdma_mask		= tmp_hwif->mwdma_mask;
-	hwif->swdma_mask		= tmp_hwif->swdma_mask;
-
-	hwif->cbl			= tmp_hwif->cbl;
-
-	hwif->chipset			= tmp_hwif->chipset;
-	hwif->hold			= tmp_hwif->hold;
-
-	hwif->dev			= tmp_hwif->dev;
-
-#ifdef CONFIG_BLK_DEV_IDEPCI
-	hwif->cds			= tmp_hwif->cds;
-#endif
-
-	hwif->set_pio_mode		= tmp_hwif->set_pio_mode;
-	hwif->set_dma_mode		= tmp_hwif->set_dma_mode;
-	hwif->mdma_filter		= tmp_hwif->mdma_filter;
-	hwif->udma_filter		= tmp_hwif->udma_filter;
-	hwif->selectproc		= tmp_hwif->selectproc;
-	hwif->reset_poll		= tmp_hwif->reset_poll;
-	hwif->pre_reset			= tmp_hwif->pre_reset;
-	hwif->resetproc			= tmp_hwif->resetproc;
-	hwif->maskproc			= tmp_hwif->maskproc;
-	hwif->quirkproc			= tmp_hwif->quirkproc;
-
-	hwif->ata_input_data		= tmp_hwif->ata_input_data;
-	hwif->ata_output_data		= tmp_hwif->ata_output_data;
-	hwif->atapi_input_bytes		= tmp_hwif->atapi_input_bytes;
-	hwif->atapi_output_bytes	= tmp_hwif->atapi_output_bytes;
-
-	hwif->dma_host_set		= tmp_hwif->dma_host_set;
-	hwif->dma_setup			= tmp_hwif->dma_setup;
-	hwif->dma_exec_cmd		= tmp_hwif->dma_exec_cmd;
-	hwif->dma_start			= tmp_hwif->dma_start;
-	hwif->ide_dma_end		= tmp_hwif->ide_dma_end;
-	hwif->ide_dma_test_irq		= tmp_hwif->ide_dma_test_irq;
-	hwif->ide_dma_clear_irq		= tmp_hwif->ide_dma_clear_irq;
-	hwif->dma_lost_irq		= tmp_hwif->dma_lost_irq;
-	hwif->dma_timeout		= tmp_hwif->dma_timeout;
-
-	hwif->OUTB			= tmp_hwif->OUTB;
-	hwif->OUTBSYNC			= tmp_hwif->OUTBSYNC;
-	hwif->OUTW			= tmp_hwif->OUTW;
-	hwif->OUTSW			= tmp_hwif->OUTSW;
-	hwif->OUTSL			= tmp_hwif->OUTSL;
-
-	hwif->INB			= tmp_hwif->INB;
-	hwif->INW			= tmp_hwif->INW;
-	hwif->INSW			= tmp_hwif->INSW;
-	hwif->INSL			= tmp_hwif->INSL;
-
-	hwif->sg_max_nents		= tmp_hwif->sg_max_nents;
-
-	hwif->mmio			= tmp_hwif->mmio;
-	hwif->rqsize			= tmp_hwif->rqsize;
-
-#ifndef CONFIG_BLK_DEV_IDECS
-	hwif->irq			= tmp_hwif->irq;
-#endif
-
-	hwif->dma_base			= tmp_hwif->dma_base;
-	hwif->dma_command		= tmp_hwif->dma_command;
-	hwif->dma_vendor1		= tmp_hwif->dma_vendor1;
-	hwif->dma_status		= tmp_hwif->dma_status;
-	hwif->dma_vendor3		= tmp_hwif->dma_vendor3;
-	hwif->dma_prdtable		= tmp_hwif->dma_prdtable;
-
-	hwif->config_data		= tmp_hwif->config_data;
-	hwif->select_data		= tmp_hwif->select_data;
-	hwif->extra_base		= tmp_hwif->extra_base;
-	hwif->extra_ports		= tmp_hwif->extra_ports;
-
-	hwif->hwif_data			= tmp_hwif->hwif_data;
-}
-
 void ide_remove_port_from_hwgroup(ide_hwif_t *hwif)
 {
 	ide_hwgroup_t *hwgroup = hwif->hwgroup;
@@ -530,8 +429,6 @@ EXPORT_SYMBOL_GPL(ide_port_unregister_devices);
 /**
  *	ide_unregister		-	free an IDE interface
  *	@index: index of interface (will change soon to a pointer)
- *	@init_default: init default hwif flag
- *	@restore: restore hwif flag
  *
  *	Perform the final unregister of an IDE interface. At the moment
  *	we don't refcount interfaces so this will also get split up.
@@ -551,10 +448,9 @@ EXPORT_SYMBOL_GPL(ide_port_unregister_devices);
  *	This is raving bonkers.
  */
 
-void ide_unregister(unsigned int index, int init_default, int restore)
+void ide_unregister(unsigned int index)
 {
 	ide_hwif_t *hwif, *g;
-	static ide_hwif_t tmp_hwif; /* protected by ide_cfg_mtx */
 	ide_hwgroup_t *hwgroup;
 	int irq_count = 0;
 
@@ -601,34 +497,14 @@ void ide_unregister(unsigned int index, int init_default, int restore)
 	unregister_blkdev(hwif->major, hwif->name);
 	spin_lock_irq(&ide_lock);
 
-	if (hwif->dma_base) {
-		(void) ide_release_dma(hwif);
-
-		hwif->dma_base = 0;
-		hwif->dma_command = 0;
-		hwif->dma_vendor1 = 0;
-		hwif->dma_status = 0;
-		hwif->dma_vendor3 = 0;
-		hwif->dma_prdtable = 0;
-
-		hwif->extra_base  = 0;
-		hwif->extra_ports = 0;
-	}
+	if (hwif->dma_base)
+		(void)ide_release_dma(hwif);
 
 	ide_hwif_release_regions(hwif);
 
-	/* copy original settings */
-	tmp_hwif = *hwif;
-
 	/* restore hwif data to pristine status */
 	ide_init_port_data(hwif, index);
 
-	if (init_default)
-		init_hwif_default(hwif, index);
-
-	if (restore)
-		ide_hwif_restore(hwif, &tmp_hwif);
-
 abort:
 	spin_unlock_irq(&ide_lock);
 	mutex_unlock(&ide_cfg_mtx);
@@ -647,52 +523,6 @@ void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
 }
 EXPORT_SYMBOL_GPL(ide_init_port_hw);
 
-/**
- *	ide_register_hw		-	register IDE interface
- *	@hw: hardware registers
- *	@quirkproc: quirkproc function
- *	@hwifp: pointer to returned hwif
- *
- *	Register an IDE interface, specifying exactly the registers etc.
- *
- *	Returns -1 on error.
- */
-
-static int ide_register_hw(hw_regs_t *hw, void (*quirkproc)(ide_drive_t *),
-			   ide_hwif_t **hwifp)
-{
-	int index, retry = 1;
-	ide_hwif_t *hwif;
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-
-	do {
-		hwif = ide_find_port(hw->io_ports[IDE_DATA_OFFSET]);
-		if (hwif)
-			goto found;
-		for (index = 0; index < MAX_HWIFS; index++)
-			ide_unregister(index, 1, 1);
-	} while (retry--);
-	return -1;
-found:
-	index = hwif->index;
-	if (hwif->present)
-		ide_unregister(index, 0, 1);
-	else if (!hwif->hold)
-		ide_init_port_data(hwif, index);
-
-	ide_init_port_hw(hwif, hw);
-	hwif->quirkproc = quirkproc;
-
-	idx[0] = index;
-
-	ide_device_add(idx, NULL);
-
-	if (hwifp)
-		*hwifp = hwif;
-
-	return hwif->present ? index : -1;
-}
-
 /*
  *	Locks for IDE setting functionality
  */
@@ -995,27 +825,6 @@ int generic_ide_ioctl(ide_drive_t *drive, struct file *file, struct block_device
 			if (!capable(CAP_SYS_RAWIO))
 				return -EACCES;
 			return ide_task_ioctl(drive, cmd, arg);
-
-		case HDIO_SCAN_HWIF:
-		{
-			hw_regs_t hw;
-			int args[3];
-			if (!capable(CAP_SYS_RAWIO)) return -EACCES;
-			if (copy_from_user(args, p, 3 * sizeof(int)))
-				return -EFAULT;
-			memset(&hw, 0, sizeof(hw));
-			ide_init_hwif_ports(&hw, (unsigned long) args[0],
-					    (unsigned long) args[1], NULL);
-			hw.irq = args[2];
-			if (ide_register_hw(&hw, NULL, NULL) == -1)
-				return -EIO;
-			return 0;
-		}
-	        case HDIO_UNREGISTER_HWIF:
-			if (!capable(CAP_SYS_RAWIO)) return -EACCES;
-			/* (arg > MAX_HWIFS) checked in function */
-			ide_unregister(arg, 1, 1);
-			return 0;
 		case HDIO_SET_NICE:
 			if (!capable(CAP_SYS_ADMIN)) return -EACCES;
 			if (arg != (arg & ((1 << IDE_NICE_DSC_OVERLAP) | (1 << IDE_NICE_1))))
@@ -1648,7 +1457,7 @@ void __exit cleanup_module (void)
 	int index;
 
 	for (index = 0; index < MAX_HWIFS; ++index)
-		ide_unregister(index, 0, 0);
+		ide_unregister(index);
 
 	proc_ide_destroy();
 
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index de2e5944809e..2b0b4958881a 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -163,7 +163,7 @@ static int idecs_register(unsigned long io, unsigned long ctl, unsigned long irq
     i = hwif->index;
 
     if (hwif->present)
-	ide_unregister(i, 0, 0);
+	ide_unregister(i);
     else if (!hwif->hold)
 	ide_init_port_data(hwif, i);
 
@@ -360,7 +360,7 @@ void ide_release(struct pcmcia_device *link)
     if (info->ndev) {
 	/* FIXME: if this fails we need to queue the cleanup somehow
 	   -- need to investigate the required PCMCIA magic */
-	ide_unregister(info->hd, 0, 0);
+	ide_unregister(info->hd);
     }
     info->ndev = 0;
 
diff --git a/drivers/ide/legacy/ide_platform.c b/drivers/ide/legacy/ide_platform.c
index 688fcae17488..249651e2da42 100644
--- a/drivers/ide/legacy/ide_platform.c
+++ b/drivers/ide/legacy/ide_platform.c
@@ -122,7 +122,7 @@ static int __devexit plat_ide_remove(struct platform_device *pdev)
 {
 	ide_hwif_t *hwif = pdev->dev.driver_data;
 
-	ide_unregister(hwif->index, 0, 0);
+	ide_unregister(hwif->index);
 
 	return 0;
 }
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 85c016bdfd38..ee76023f3737 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -673,7 +673,7 @@ static int au_ide_remove(struct device *dev)
 	ide_hwif_t *hwif = dev_get_drvdata(dev);
 	_auide_hwif *ahwif = &auide_hwif;
 
-	ide_unregister(hwif->index, 0, 0);
+	ide_unregister(hwif->index);
 
 	iounmap((void *)ahwif->regbase);
 
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index e08e13a0bb6e..89570df52f0a 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -85,7 +85,7 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 	i = hwif->index;
 
 	if (hwif->present)
-		ide_unregister(i, 0, 0);
+		ide_unregister(i);
 	else if (!hwif->hold)
 		ide_init_port_data(hwif, i);
 
@@ -120,7 +120,7 @@ delkin_cb_remove (struct pci_dev *dev)
 	ide_hwif_t *hwif = pci_get_drvdata(dev);
 
 	if (hwif)
-		ide_unregister(hwif->index, 0, 0);
+		ide_unregister(hwif->index);
 
 	pci_disable_device(dev);
 }
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 238e3e181e87..085c1b58a99c 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -736,7 +736,7 @@ static void __devexit scc_remove(struct pci_dev *dev)
 		hwif->dmatable_cpu = NULL;
 	}
 
-	ide_unregister(hwif->index, 0, 0);
+	ide_unregister(hwif->index);
 
 	hwif->chipset = ide_unknown;
 	iounmap((void*)ports->dma);
diff --git a/include/linux/hdreg.h b/include/linux/hdreg.h
index e38e75967e74..c37e9241fae7 100644
--- a/include/linux/hdreg.h
+++ b/include/linux/hdreg.h
@@ -422,9 +422,11 @@ struct hd_geometry {
 #define HDIO_SET_NOWERR		0x0325	/* change ignore-write-error flag */
 #define HDIO_SET_DMA		0x0326	/* change use-dma flag */
 #define HDIO_SET_PIO_MODE	0x0327	/* reconfig interface to new speed */
+#ifndef __KERNEL__
 #define HDIO_SCAN_HWIF		0x0328	/* register and (re)scan interface */
-#define HDIO_SET_NICE		0x0329	/* set nice flags */
 #define HDIO_UNREGISTER_HWIF	0x032a  /* unregister interface */
+#endif
+#define HDIO_SET_NICE		0x0329	/* set nice flags */
 #define HDIO_SET_WCACHE		0x032b	/* change write cache enable-disable */
 #define HDIO_SET_ACOUSTIC	0x032c	/* change acoustic behavior */
 #define HDIO_SET_BUSSTATE	0x032d	/* set the bus state of the hwif */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 67f83c60845f..478ddf7e21d5 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1191,7 +1191,7 @@ static inline void ide_acpi_set_state(ide_hwif_t *hwif, int on) {}
 void ide_remove_port_from_hwgroup(ide_hwif_t *);
 extern int ide_hwif_request_regions(ide_hwif_t *hwif);
 extern void ide_hwif_release_regions(ide_hwif_t* hwif);
-void ide_unregister(unsigned int, int, int);
+void ide_unregister(unsigned int);
 
 void ide_register_region(struct gendisk *);
 void ide_unregister_region(struct gendisk *);
-- 
cgit v1.2.3


From 2304dc6481f9d4fb4f0cb5b72497dfe1694cef9c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:24 +0200
Subject: ide: remove ->hold field from ide_hwif_t (take 2)

->hold is write-only now, remove it.

v2:
* v1 missed bast-ide, palm_bk3710, ide-cs and delkin_cb host drivers.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/bast-ide.c    | 2 +-
 drivers/ide/arm/palm_bk3710.c | 2 +-
 drivers/ide/legacy/ide-cs.c   | 2 +-
 drivers/ide/mips/au1xxx-ide.c | 3 ---
 drivers/ide/pci/delkin_cb.c   | 2 +-
 drivers/ide/ppc/pmac.c        | 1 -
 include/linux/ide.h           | 1 -
 7 files changed, 4 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/arm/bast-ide.c b/drivers/ide/arm/bast-ide.c
index d2196436788b..ec46c44b061c 100644
--- a/drivers/ide/arm/bast-ide.c
+++ b/drivers/ide/arm/bast-ide.c
@@ -49,7 +49,7 @@ static int __init bastide_register(unsigned int base, unsigned int aux, int irq)
 
 	if (hwif->present)
 		ide_unregister(i);
-	else if (!hwif->hold)
+	else
 		ide_init_port_data(hwif, i);
 
 	ide_init_port_hw(hwif, &hw);
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 73899ef4ab0d..474162cdf665 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -386,7 +386,7 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 
 	if (hwif->present)
 		ide_unregister(i);
-	else if (!hwif->hold)
+	else
 		ide_init_port_data(hwif, i);
 
 	ide_init_port_hw(hwif, &hw);
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index 2b0b4958881a..9a23b94f2939 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -164,7 +164,7 @@ static int idecs_register(unsigned long io, unsigned long ctl, unsigned long irq
 
     if (hwif->present)
 	ide_unregister(i);
-    else if (!hwif->hold)
+    else
 	ide_init_port_data(hwif, i);
 
     ide_init_port_hw(hwif, &hw);
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index ee76023f3737..9b628248f2f4 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -613,9 +613,6 @@ static int au_ide_probe(struct device *dev)
 
 	hwif->dev = dev;
 
-	/* hold should be on in all cases */
-	hwif->hold                      = 1;
-
 	hwif->mmio  = 1;
 
 	/* If the user has selected DDMA assisted copies,
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index 89570df52f0a..961698d655eb 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -86,7 +86,7 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 
 	if (hwif->present)
 		ide_unregister(i);
-	else if (!hwif->hold)
+	else
 		ide_init_port_data(hwif, i);
 
 	ide_init_port_hw(hwif, &hw);
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 7889f5f5c49a..b37dcfbdaad6 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1120,7 +1120,6 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif, hw_regs_t *hw)
 	hwif->hwif_data = pmif;
 	ide_init_port_hw(hwif, hw);
 	hwif->noprobe = pmif->mediabay;
-	hwif->hold = pmif->mediabay;
 	hwif->cbl = pmif->cable_80 ? ATA_CBL_PATA80 : ATA_CBL_PATA40;
 	hwif->set_pio_mode = pmac_ide_set_pio_mode;
 	if (pmif->kind == controller_un_ata6
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 478ddf7e21d5..7744ac9d1ff9 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -571,7 +571,6 @@ typedef struct hwif_s {
 
 	unsigned	noprobe    : 1;	/* don't probe for this interface */
 	unsigned	present    : 1;	/* this interface exists */
-	unsigned	hold       : 1; /* this interface is always present */
 	unsigned	serialized : 1;	/* serialized all channel operation */
 	unsigned	sharing_irq: 1;	/* 1 = sharing irq with another hwif */
 	unsigned	reset      : 1;	/* reset after probe */
-- 
cgit v1.2.3


From e6bfa38a4803646e212ef542b957344e790c3733 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:25 +0200
Subject: ide: remove ide_init_hwif_ports()

ide_init_hwif_ports() is only used by init_ide_data() now, inline it there.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide.c   | 11 +++++++++--
 include/linux/ide.h | 32 --------------------------------
 2 files changed, 9 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index a60b956db519..9e4d503bc716 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -197,13 +197,20 @@ static void __init init_ide_data (void)
 	/* Initialise all interface structures */
 	for (index = 0; index < MAX_HWIFS; ++index) {
 		ide_hwif_t *hwif = &ide_hwifs[index];
+		unsigned long io_addr = ide_default_io_base(index);
+		unsigned long ctl_addr = ide_default_io_ctl(io_addr);
 
 		ide_init_port_data(hwif, index);
 
+#ifdef CONFIG_IDE_ARCH_OBSOLETE_INIT
 		memset(&hw, 0, sizeof(hw));
-		ide_init_hwif_ports(&hw, ide_default_io_base(index), 0,
-				    &hwif->irq);
+		ide_std_init_ports(&hw, io_addr, ctl_addr);
+# ifdef CONFIG_PPC32
+		if (ppc_ide_md.ide_init_hwif)
+			ppc_ide_md.ide_init_hwif(&hw, io_addr, 0, &hwif->irq);
+# endif
 		memcpy(hwif->io_ports, hw.io_ports, sizeof(hw.io_ports));
+#endif
 		hwif->noprobe = !hwif->io_ports[IDE_DATA_OFFSET];
 #if !defined(CONFIG_PPC32) || !defined(CONFIG_PCI)
 		hwif->irq =
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 7744ac9d1ff9..f4c7db572a16 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -215,38 +215,6 @@ static inline void ide_std_init_ports(hw_regs_t *hw,
 # define ide_init_default_irq(base)	(0)
 #endif
 
-#ifdef CONFIG_IDE_ARCH_OBSOLETE_INIT
-static inline void ide_init_hwif_ports(hw_regs_t *hw,
-				       unsigned long io_addr,
-				       unsigned long ctl_addr,
-				       int *irq)
-{
-	if (!ctl_addr)
-		ide_std_init_ports(hw, io_addr, ide_default_io_ctl(io_addr));
-	else
-		ide_std_init_ports(hw, io_addr, ctl_addr);
-
-	if (irq)
-		*irq = 0;
-
-	hw->io_ports[IDE_IRQ_OFFSET] = 0;
-
-#ifdef CONFIG_PPC32
-	if (ppc_ide_md.ide_init_hwif)
-		ppc_ide_md.ide_init_hwif(hw, io_addr, ctl_addr, irq);
-#endif
-}
-#else
-static inline void ide_init_hwif_ports(hw_regs_t *hw,
-				       unsigned long io_addr,
-				       unsigned long ctl_addr,
-				       int *irq)
-{
-	if (io_addr || ctl_addr)
-		printk(KERN_WARNING "%s: must not be called\n", __FUNCTION__);
-}
-#endif /* CONFIG_IDE_ARCH_OBSOLETE_INIT */
-
 /* Currently only m68k, apus and m8xx need it */
 #ifndef IDE_ARCH_ACK_INTR
 # define ide_ack_intr(hwif) (1)
-- 
cgit v1.2.3


From 7616c0ad2087c7d244b8985390c63059a6223c45 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:26 +0200
Subject: ide: add ide_atapi_{discard_data,write_zeros} inline helpers

Add ide_atapi_{discard_data,write_zeros} inline helpers to <linux/ide.h>
and use them instead of home-brewn helpers in ide-{floppy,tape,scsi}.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 24 +++---------------------
 drivers/ide/ide-tape.c   | 14 ++------------
 drivers/scsi/ide-scsi.c  | 18 +++---------------
 include/linux/ide.h      | 22 ++++++++++++++++++++++
 4 files changed, 30 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 973f5f6c815e..170c60d93f55 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -229,23 +229,6 @@ static void ide_floppy_put(struct ide_floppy_obj *floppy)
 	mutex_unlock(&idefloppy_ref_mutex);
 }
 
-/*
- * Too bad. The drive wants to send us data which we are not ready to accept.
- * Just throw it away.
- */
-static void idefloppy_discard_data(ide_drive_t *drive, unsigned int bcount)
-{
-	while (bcount--)
-		(void) HWIF(drive)->INB(IDE_DATA_REG);
-}
-
-static void idefloppy_write_zeros(ide_drive_t *drive, unsigned int bcount)
-{
-	while (bcount--)
-		HWIF(drive)->OUTB(0, IDE_DATA_REG);
-}
-
-
 /*
  * Used to finish servicing a request. For read/write requests, we will call
  * ide_end_request to pass to the next buffer.
@@ -313,10 +296,9 @@ static void ide_floppy_io_buffers(ide_drive_t *drive, idefloppy_pc_t *pc,
 		printk(KERN_ERR "%s: leftover data in %s, bcount == %d\n",
 				drive->name, __func__, bcount);
 		if (direction)
-			idefloppy_write_zeros(drive, bcount);
+			ide_atapi_write_zeros(drive, bcount);
 		else
-			idefloppy_discard_data(drive, bcount);
-
+			ide_atapi_discard_data(drive, bcount);
 	}
 }
 
@@ -541,7 +523,7 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 				printk(KERN_ERR "ide-floppy: The floppy wants "
 					"to send us more data than expected "
 					"- discarding data\n");
-				idefloppy_discard_data(drive, bcount);
+				ide_atapi_discard_data(drive, bcount);
 
 				ide_set_handler(drive,
 						&idefloppy_pc_intr,
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 0598ecfd5f37..bfdc4f449797 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -518,16 +518,6 @@ static struct ide_tape_obj *ide_tape_chrdev_get(unsigned int i)
 	return tape;
 }
 
-/*
- * Too bad. The drive wants to send us data which we are not ready to accept.
- * Just throw it away.
- */
-static void idetape_discard_data(ide_drive_t *drive, unsigned int bcount)
-{
-	while (bcount--)
-		(void) HWIF(drive)->INB(IDE_DATA_REG);
-}
-
 static void idetape_input_buffers(ide_drive_t *drive, idetape_pc_t *pc,
 				  unsigned int bcount)
 {
@@ -538,7 +528,7 @@ static void idetape_input_buffers(ide_drive_t *drive, idetape_pc_t *pc,
 		if (bh == NULL) {
 			printk(KERN_ERR "ide-tape: bh == NULL in "
 				"idetape_input_buffers\n");
-			idetape_discard_data(drive, bcount);
+			ide_atapi_discard_data(drive, bcount);
 			return;
 		}
 		count = min(
@@ -1152,7 +1142,7 @@ static ide_startstop_t idetape_pc_intr(ide_drive_t *drive)
 				printk(KERN_ERR "ide-tape: The tape wants to "
 					"send us more data than expected "
 					"- discarding data\n");
-				idetape_discard_data(drive, bcount);
+				ide_atapi_discard_data(drive, bcount);
 				ide_set_handler(drive, &idetape_pc_intr,
 						IDETAPE_WAIT_CMD, NULL);
 				return ide_started;
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 68e5c632c5d5..3c3b3502c4d4 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -152,18 +152,6 @@ static inline idescsi_scsi_t *drive_to_idescsi(ide_drive_t *ide_drive)
  */
 #define IDESCSI_PC_RQ			90
 
-static void idescsi_discard_data (ide_drive_t *drive, unsigned int bcount)
-{
-	while (bcount--)
-		(void) HWIF(drive)->INB(IDE_DATA_REG);
-}
-
-static void idescsi_output_zeros (ide_drive_t *drive, unsigned int bcount)
-{
-	while (bcount--)
-		HWIF(drive)->OUTB(0, IDE_DATA_REG);
-}
-
 /*
  *	PIO data transfer routines using the scatter gather table.
  */
@@ -200,7 +188,7 @@ static void idescsi_input_buffers (ide_drive_t *drive, idescsi_pc_t *pc, unsigne
 
 	if (bcount) {
 		printk (KERN_ERR "ide-scsi: scatter gather table too small, discarding data\n");
-		idescsi_discard_data (drive, bcount);
+		ide_atapi_discard_data(drive, bcount);
 	}
 }
 
@@ -237,7 +225,7 @@ static void idescsi_output_buffers (ide_drive_t *drive, idescsi_pc_t *pc, unsign
 
 	if (bcount) {
 		printk (KERN_ERR "ide-scsi: scatter gather table too small, padding with zeros\n");
-		idescsi_output_zeros (drive, bcount);
+		ide_atapi_write_zeros(drive, bcount);
 	}
 }
 
@@ -463,7 +451,7 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 				}
 				pc->actually_transferred += temp;
 				pc->current_position += temp;
-				idescsi_discard_data(drive, bcount - temp);
+				ide_atapi_discard_data(drive, bcount - temp);
 				ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
 				return ide_started;
 			}
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f4c7db572a16..2eb99cab4a3d 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1297,4 +1297,26 @@ static inline u8 ide_read_error(ide_drive_t *drive)
 	return hwif->INB(hwif->io_ports[IDE_ERROR_OFFSET]);
 }
 
+/*
+ * Too bad. The drive wants to send us data which we are not ready to accept.
+ * Just throw it away.
+ */
+static inline void ide_atapi_discard_data(ide_drive_t *drive, unsigned bcount)
+{
+	ide_hwif_t *hwif = drive->hwif;
+
+	/* FIXME: use ->atapi_input_bytes */
+	while (bcount--)
+		(void)hwif->INB(hwif->io_ports[IDE_DATA_OFFSET]);
+}
+
+static inline void ide_atapi_write_zeros(ide_drive_t *drive, unsigned bcount)
+{
+	ide_hwif_t *hwif = drive->hwif;
+
+	/* FIXME: use ->atapi_output_bytes */
+	while (bcount--)
+		hwif->OUTB(0, hwif->io_ports[IDE_DATA_OFFSET]);
+}
+
 #endif /* _IDE_H */
-- 
cgit v1.2.3


From 23579a2a170265aacf78069f4817a41c1d6e9323 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:26 +0200
Subject: ide: remove IDE_*_REG macros

* Add IDE_{ALTSTATUS,IREASON,BCOUNTL,BCOUNTH}_OFFSET defines.

* Remove IDE_*_REG macros - this results in more readable
  and slightly smaller code.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/cris/ide-cris.c  | 18 +++++++---
 drivers/ide/ide-cd.c         | 10 +++---
 drivers/ide/ide-floppy.c     | 14 ++++----
 drivers/ide/ide-io.c         | 38 ++++++++++++--------
 drivers/ide/ide-iops.c       | 86 +++++++++++++++++++++++++-------------------
 drivers/ide/ide-probe.c      | 19 +++++-----
 drivers/ide/ide-tape.c       | 12 +++----
 drivers/ide/ide-taskfile.c   | 29 ++++++++-------
 drivers/ide/legacy/ht6560b.c | 13 +++----
 drivers/ide/pci/hpt366.c     |  2 +-
 drivers/ide/pci/scc_pata.c   |  6 ++--
 drivers/ide/pci/sgiioc4.c    | 19 +++++-----
 drivers/ide/ppc/pmac.c       |  3 +-
 drivers/scsi/ide-scsi.c      | 15 ++++----
 include/linux/ide.h          | 29 +++++----------
 15 files changed, 175 insertions(+), 138 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/cris/ide-cris.c b/drivers/ide/cris/ide-cris.c
index c8ffbaf29a88..31266d278095 100644
--- a/drivers/ide/cris/ide-cris.c
+++ b/drivers/ide/cris/ide-cris.c
@@ -228,7 +228,10 @@ cris_ide_fill_descriptor(cris_dma_descr_type *d, void* buf, unsigned int len, in
 static void
 cris_ide_start_dma(ide_drive_t *drive, cris_dma_descr_type *d, int dir,int type,int len)
 {
-	reg_ata_rw_ctrl2 ctrl2 = REG_TYPE_CONV(reg_ata_rw_ctrl2, int, IDE_DATA_REG);
+	ide_hwif_t *hwif = drive->hwif;
+
+	reg_ata_rw_ctrl2 ctrl2 = REG_TYPE_CONV(reg_ata_rw_ctrl2, int,
+					       hwif->io_ports[IDE_DATA_OFFSET]);
 	reg_ata_rw_trf_cnt trf_cnt = {0};
 
 	mycontext.saved_data = (dma_descr_data*)virt_to_phys(d);
@@ -264,8 +267,12 @@ cris_ide_wait_dma(int dir)
 
 static int cris_dma_test_irq(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	int intr = REG_RD_INT(ata, regi_ata, r_intr);
-	reg_ata_rw_ctrl2 ctrl2 = REG_TYPE_CONV(reg_ata_rw_ctrl2, int, IDE_DATA_REG);
+
+	reg_ata_rw_ctrl2 ctrl2 = REG_TYPE_CONV(reg_ata_rw_ctrl2, int,
+					       hwif->io_ports[IDE_DATA_OFFSET]);
+
 	return intr & (1 << ctrl2.sel) ? 1 : 0;
 }
 
@@ -523,7 +530,8 @@ static void cris_ide_start_dma(ide_drive_t *drive, cris_dma_descr_type *d, int d
 	                          IO_STATE(R_ATA_CTRL_DATA, handsh, dma);
 	*R_ATA_CTRL_DATA =
 		cmd |
-		IO_FIELD(R_ATA_CTRL_DATA, data, IDE_DATA_REG) |
+		IO_FIELD(R_ATA_CTRL_DATA, data,
+			 drive->hwif->io_ports[IDE_DATA_OFFSET]) |
 		IO_STATE(R_ATA_CTRL_DATA, src_dst,  dma)  |
 		IO_STATE(R_ATA_CTRL_DATA, multi,    on)   |
 		IO_STATE(R_ATA_CTRL_DATA, dma_size, word);
@@ -541,7 +549,9 @@ cris_ide_wait_dma(int dir)
 static int cris_dma_test_irq(ide_drive_t *drive)
 {
 	int intr = *R_IRQ_MASK0_RD;
-	int bus = IO_EXTRACT(R_ATA_CTRL_DATA, sel, IDE_DATA_REG);
+	int bus = IO_EXTRACT(R_ATA_CTRL_DATA, sel,
+			     drive->hwif->io_ports[IDE_DATA_OFFSET]);
+
 	return intr & (1 << (bus + IO_BITNR(R_IRQ_MASK0_RD, ata_irq0))) ? 1 : 0;
 }
 
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index c8d0e8715997..396000208f81 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -542,7 +542,8 @@ static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive,
 
 		/* packet command */
 		spin_lock_irqsave(&ide_lock, flags);
-		hwif->OUTBSYNC(drive, WIN_PACKETCMD, IDE_COMMAND_REG);
+		hwif->OUTBSYNC(drive, WIN_PACKETCMD,
+			       hwif->io_ports[IDE_COMMAND_OFFSET]);
 		ndelay(400);
 		spin_unlock_irqrestore(&ide_lock, flags);
 
@@ -992,6 +993,7 @@ static int cdrom_newpc_intr_dummy_cb(struct request *rq)
 
 static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	struct cdrom_info *info = drive->driver_data;
 	struct request *rq = HWGROUP(drive)->rq;
 	xfer_func_t *xferfunc;
@@ -1032,9 +1034,9 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 	/*
 	 * ok we fall to pio :/
 	 */
-	ireason = HWIF(drive)->INB(IDE_IREASON_REG) & 0x3;
-	lowcyl  = HWIF(drive)->INB(IDE_BCOUNTL_REG);
-	highcyl = HWIF(drive)->INB(IDE_BCOUNTH_REG);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]) & 0x3;
+	lowcyl  = hwif->INB(hwif->io_ports[IDE_BCOUNTL_OFFSET]);
+	highcyl = hwif->INB(hwif->io_ports[IDE_BCOUNTH_OFFSET]);
 
 	len = lowcyl + (256 * highcyl);
 
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 170c60d93f55..4ce67bdb5d5e 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -498,10 +498,10 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
 	}
 
 	/* Get the number of bytes to transfer */
-	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
-		  hwif->INB(IDE_BCOUNTL_REG);
+	bcount = (hwif->INB(hwif->io_ports[IDE_BCOUNTH_OFFSET]) << 8) |
+		  hwif->INB(hwif->io_ports[IDE_BCOUNTL_OFFSET]);
 	/* on this interrupt */
-	ireason = hwif->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 
 	if (ireason & CD) {
 		printk(KERN_ERR "ide-floppy: CoD != 0 in %s\n", __func__);
@@ -562,6 +562,7 @@ static ide_startstop_t idefloppy_pc_intr (ide_drive_t *drive)
  */
 static ide_startstop_t idefloppy_transfer_pc(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	ide_startstop_t startstop;
 	idefloppy_floppy_t *floppy = drive->driver_data;
 	u8 ireason;
@@ -571,7 +572,7 @@ static ide_startstop_t idefloppy_transfer_pc(ide_drive_t *drive)
 				"initiated yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason = drive->hwif->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 	if ((ireason & CD) == 0 || (ireason & IO)) {
 		printk(KERN_ERR "ide-floppy: (IO,CoD) != (0,1) while "
 				"issuing a packet command\n");
@@ -608,6 +609,7 @@ static int idefloppy_transfer_pc2(ide_drive_t *drive)
 
 static ide_startstop_t idefloppy_transfer_pc1(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	idefloppy_floppy_t *floppy = drive->driver_data;
 	ide_startstop_t startstop;
 	u8 ireason;
@@ -617,7 +619,7 @@ static ide_startstop_t idefloppy_transfer_pc1(ide_drive_t *drive)
 				"initiated yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason = drive->hwif->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 	if ((ireason & CD) == 0 || (ireason & IO)) {
 		printk(KERN_ERR "ide-floppy: (IO,CoD) != (0,1) "
 				"while issuing a packet command\n");
@@ -723,7 +725,7 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 		return ide_started;
 	} else {
 		/* Issue the packet command */
-		HWIF(drive)->OUTB(WIN_PACKETCMD, IDE_COMMAND_REG);
+		hwif->OUTB(WIN_PACKETCMD, hwif->io_ports[IDE_COMMAND_OFFSET]);
 		return (*pkt_xfer_routine) (drive);
 	}
 }
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 715379605a7b..31e5afadb7e9 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -301,39 +301,45 @@ void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
 	struct ide_taskfile *tf = &task->tf;
 
 	if (task->tf_flags & IDE_TFLAG_IN_DATA) {
-		u16 data = hwif->INW(IDE_DATA_REG);
+		u16 data = hwif->INW(hwif->io_ports[IDE_DATA_OFFSET]);
 
 		tf->data = data & 0xff;
 		tf->hob_data = (data >> 8) & 0xff;
 	}
 
 	/* be sure we're looking at the low order bits */
-	hwif->OUTB(drive->ctl & ~0x80, IDE_CONTROL_REG);
+	hwif->OUTB(drive->ctl & ~0x80, hwif->io_ports[IDE_CONTROL_OFFSET]);
 
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
-		tf->nsect  = hwif->INB(IDE_NSECTOR_REG);
+		tf->nsect  = hwif->INB(hwif->io_ports[IDE_NSECTOR_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_IN_LBAL)
-		tf->lbal   = hwif->INB(IDE_SECTOR_REG);
+		tf->lbal   = hwif->INB(hwif->io_ports[IDE_SECTOR_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_IN_LBAM)
-		tf->lbam   = hwif->INB(IDE_LCYL_REG);
+		tf->lbam   = hwif->INB(hwif->io_ports[IDE_LCYL_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_IN_LBAH)
-		tf->lbah   = hwif->INB(IDE_HCYL_REG);
+		tf->lbah   = hwif->INB(hwif->io_ports[IDE_HCYL_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_IN_DEVICE)
-		tf->device = hwif->INB(IDE_SELECT_REG);
+		tf->device = hwif->INB(hwif->io_ports[IDE_SELECT_OFFSET]);
 
 	if (task->tf_flags & IDE_TFLAG_LBA48) {
-		hwif->OUTB(drive->ctl | 0x80, IDE_CONTROL_REG);
+		hwif->OUTB(drive->ctl | 0x80,
+			   hwif->io_ports[IDE_CONTROL_OFFSET]);
 
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_FEATURE)
-			tf->hob_feature = hwif->INB(IDE_FEATURE_REG);
+			tf->hob_feature =
+				hwif->INB(hwif->io_ports[IDE_FEATURE_OFFSET]);
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_NSECT)
-			tf->hob_nsect   = hwif->INB(IDE_NSECTOR_REG);
+			tf->hob_nsect   =
+				hwif->INB(hwif->io_ports[IDE_NSECTOR_OFFSET]);
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_LBAL)
-			tf->hob_lbal    = hwif->INB(IDE_SECTOR_REG);
+			tf->hob_lbal    =
+				hwif->INB(hwif->io_ports[IDE_SECTOR_OFFSET]);
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_LBAM)
-			tf->hob_lbam    = hwif->INB(IDE_LCYL_REG);
+			tf->hob_lbam    =
+				hwif->INB(hwif->io_ports[IDE_LCYL_OFFSET]);
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_LBAH)
-			tf->hob_lbah    = hwif->INB(IDE_HCYL_REG);
+			tf->hob_lbah    =
+				hwif->INB(hwif->io_ports[IDE_HCYL_OFFSET]);
 	}
 }
 
@@ -448,7 +454,8 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8
 		if (err == ABRT_ERR) {
 			if (drive->select.b.lba &&
 			    /* some newer drives don't support WIN_SPECIFY */
-			    hwif->INB(IDE_COMMAND_REG) == WIN_SPECIFY)
+			    hwif->INB(hwif->io_ports[IDE_COMMAND_OFFSET]) ==
+				WIN_SPECIFY)
 				return ide_stopped;
 		} else if ((err & BAD_CRC) == BAD_CRC) {
 			/* UDMA crc error, just retry the operation */
@@ -500,7 +507,8 @@ static ide_startstop_t ide_atapi_error(ide_drive_t *drive, struct request *rq, u
 
 	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		hwif->OUTB(WIN_IDLEIMMEDIATE, IDE_COMMAND_REG);
+		hwif->OUTB(WIN_IDLEIMMEDIATE,
+			   hwif->io_ports[IDE_COMMAND_OFFSET]);
 
 	if (rq->errors >= ERROR_MAX) {
 		ide_kill_rq(drive, rq);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index e77cee0e5d65..45944219eea0 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -158,9 +158,12 @@ EXPORT_SYMBOL(default_hwif_mmiops);
 
 void SELECT_DRIVE (ide_drive_t *drive)
 {
-	if (HWIF(drive)->selectproc)
-		HWIF(drive)->selectproc(drive);
-	HWIF(drive)->OUTB(drive->select.all, IDE_SELECT_REG);
+	ide_hwif_t *hwif = drive->hwif;
+
+	if (hwif->selectproc)
+		hwif->selectproc(drive);
+
+	hwif->OUTB(drive->select.all, hwif->io_ports[IDE_SELECT_OFFSET]);
 }
 
 void SELECT_MASK (ide_drive_t *drive, int mask)
@@ -194,15 +197,18 @@ static void ata_input_data(ide_drive_t *drive, void *buffer, u32 wcount)
 	if (io_32bit) {
 		if (io_32bit & 2) {
 			unsigned long flags;
+
 			local_irq_save(flags);
-			ata_vlb_sync(drive, IDE_NSECTOR_REG);
-			hwif->INSL(IDE_DATA_REG, buffer, wcount);
+			ata_vlb_sync(drive, hwif->io_ports[IDE_NSECTOR_OFFSET]);
+			hwif->INSL(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+				   wcount);
 			local_irq_restore(flags);
 		} else
-			hwif->INSL(IDE_DATA_REG, buffer, wcount);
-	} else {
-		hwif->INSW(IDE_DATA_REG, buffer, wcount<<1);
-	}
+			hwif->INSL(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+				   wcount);
+	} else
+		hwif->INSW(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+			   wcount << 1);
 }
 
 /*
@@ -216,15 +222,18 @@ static void ata_output_data(ide_drive_t *drive, void *buffer, u32 wcount)
 	if (io_32bit) {
 		if (io_32bit & 2) {
 			unsigned long flags;
+
 			local_irq_save(flags);
-			ata_vlb_sync(drive, IDE_NSECTOR_REG);
-			hwif->OUTSL(IDE_DATA_REG, buffer, wcount);
+			ata_vlb_sync(drive, hwif->io_ports[IDE_NSECTOR_OFFSET]);
+			hwif->OUTSL(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+				    wcount);
 			local_irq_restore(flags);
 		} else
-			hwif->OUTSL(IDE_DATA_REG, buffer, wcount);
-	} else {
-		hwif->OUTSW(IDE_DATA_REG, buffer, wcount<<1);
-	}
+			hwif->OUTSL(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+				    wcount);
+	} else
+		hwif->OUTSW(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+			    wcount << 1);
 }
 
 /*
@@ -243,13 +252,15 @@ static void atapi_input_bytes(ide_drive_t *drive, void *buffer, u32 bytecount)
 #if defined(CONFIG_ATARI) || defined(CONFIG_Q40)
 	if (MACH_IS_ATARI || MACH_IS_Q40) {
 		/* Atari has a byte-swapped IDE interface */
-		insw_swapw(IDE_DATA_REG, buffer, bytecount / 2);
+		insw_swapw(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+			   bytecount / 2);
 		return;
 	}
 #endif /* CONFIG_ATARI || CONFIG_Q40 */
 	hwif->ata_input_data(drive, buffer, bytecount / 4);
 	if ((bytecount & 0x03) >= 2)
-		hwif->INSW(IDE_DATA_REG, ((u8 *)buffer)+(bytecount & ~0x03), 1);
+		hwif->INSW(hwif->io_ports[IDE_DATA_OFFSET],
+			   (u8 *)buffer + (bytecount & ~0x03), 1);
 }
 
 static void atapi_output_bytes(ide_drive_t *drive, void *buffer, u32 bytecount)
@@ -260,13 +271,15 @@ static void atapi_output_bytes(ide_drive_t *drive, void *buffer, u32 bytecount)
 #if defined(CONFIG_ATARI) || defined(CONFIG_Q40)
 	if (MACH_IS_ATARI || MACH_IS_Q40) {
 		/* Atari has a byte-swapped IDE interface */
-		outsw_swapw(IDE_DATA_REG, buffer, bytecount / 2);
+		outsw_swapw(hwif->io_ports[IDE_DATA_OFFSET], buffer,
+			    bytecount / 2);
 		return;
 	}
 #endif /* CONFIG_ATARI || CONFIG_Q40 */
 	hwif->ata_output_data(drive, buffer, bytecount / 4);
 	if ((bytecount & 0x03) >= 2)
-		hwif->OUTSW(IDE_DATA_REG, ((u8*)buffer)+(bytecount & ~0x03), 1);
+		hwif->OUTSW(hwif->io_ports[IDE_DATA_OFFSET],
+			    (u8 *)buffer + (bytecount & ~0x03), 1);
 }
 
 void default_hwif_transport(ide_hwif_t *hwif)
@@ -429,7 +442,7 @@ int drive_is_ready (ide_drive_t *drive)
 	 * an interrupt with another pci card/device.  We make no assumptions
 	 * about possible isa-pnp and pci-pnp issues yet.
 	 */
-	if (IDE_CONTROL_REG)
+	if (hwif->io_ports[IDE_CONTROL_OFFSET])
 		stat = ide_read_altstatus(drive);
 	else
 		/* Note: this may clear a pending IRQ!! */
@@ -631,7 +644,7 @@ int ide_driveid_update(ide_drive_t *drive)
 	SELECT_MASK(drive, 1);
 	ide_set_irq(drive, 1);
 	msleep(50);
-	hwif->OUTB(WIN_IDENTIFY, IDE_COMMAND_REG);
+	hwif->OUTB(WIN_IDENTIFY, hwif->io_ports[IDE_COMMAND_OFFSET]);
 	timeout = jiffies + WAIT_WORSTCASE;
 	do {
 		if (time_after(jiffies, timeout)) {
@@ -718,9 +731,10 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 	SELECT_MASK(drive, 0);
 	udelay(1);
 	ide_set_irq(drive, 0);
-	hwif->OUTB(speed, IDE_NSECTOR_REG);
-	hwif->OUTB(SETFEATURES_XFER, IDE_FEATURE_REG);
-	hwif->OUTBSYNC(drive, WIN_SETFEATURES, IDE_COMMAND_REG);
+	hwif->OUTB(speed, hwif->io_ports[IDE_NSECTOR_OFFSET]);
+	hwif->OUTB(SETFEATURES_XFER, hwif->io_ports[IDE_FEATURE_OFFSET]);
+	hwif->OUTBSYNC(drive, WIN_SETFEATURES,
+		       hwif->io_ports[IDE_COMMAND_OFFSET]);
 	if (drive->quirk_list == 2)
 		ide_set_irq(drive, 1);
 
@@ -828,7 +842,7 @@ void ide_execute_command(ide_drive_t *drive, u8 cmd, ide_handler_t *handler,
 
 	spin_lock_irqsave(&ide_lock, flags);
 	__ide_set_handler(drive, handler, timeout, expiry);
-	hwif->OUTBSYNC(drive, cmd, IDE_COMMAND_REG);
+	hwif->OUTBSYNC(drive, cmd, hwif->io_ports[IDE_COMMAND_OFFSET]);
 	/*
 	 * Drive takes 400nS to respond, we must avoid the IRQ being
 	 * serviced before that.
@@ -1009,7 +1023,8 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	unsigned long flags;
 	ide_hwif_t *hwif;
 	ide_hwgroup_t *hwgroup;
-	
+	u8 ctl;
+
 	spin_lock_irqsave(&ide_lock, flags);
 	hwif = HWIF(drive);
 	hwgroup = HWGROUP(drive);
@@ -1023,7 +1038,8 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		pre_reset(drive);
 		SELECT_DRIVE(drive);
 		udelay (20);
-		hwif->OUTBSYNC(drive, WIN_SRST, IDE_COMMAND_REG);
+		hwif->OUTBSYNC(drive, WIN_SRST,
+			       hwif->io_ports[IDE_COMMAND_OFFSET]);
 		ndelay(400);
 		hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
 		hwgroup->polling = 1;
@@ -1039,7 +1055,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	for (unit = 0; unit < MAX_DRIVES; ++unit)
 		pre_reset(&hwif->drives[unit]);
 
-	if (!IDE_CONTROL_REG) {
+	if (hwif->io_ports[IDE_CONTROL_OFFSET] == 0) {
 		spin_unlock_irqrestore(&ide_lock, flags);
 		return ide_stopped;
 	}
@@ -1054,16 +1070,14 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * recover from reset very quickly, saving us the first 50ms wait time.
 	 */
 	/* set SRST and nIEN */
-	hwif->OUTBSYNC(drive, drive->ctl|6,IDE_CONTROL_REG);
+	hwif->OUTBSYNC(drive, drive->ctl|6, hwif->io_ports[IDE_CONTROL_OFFSET]);
 	/* more than enough time */
 	udelay(10);
-	if (drive->quirk_list == 2) {
-		/* clear SRST and nIEN */
-		hwif->OUTBSYNC(drive, drive->ctl, IDE_CONTROL_REG);
-	} else {
-		/* clear SRST, leave nIEN */
-		hwif->OUTBSYNC(drive, drive->ctl|2, IDE_CONTROL_REG);
-	}
+	if (drive->quirk_list == 2)
+		ctl = drive->ctl;	/* clear SRST and nIEN */
+	else
+		ctl = drive->ctl | 2;	/* clear SRST, leave nIEN */
+	hwif->OUTBSYNC(drive, ctl, hwif->io_ports[IDE_CONTROL_OFFSET]);
 	/* more than enough time */
 	udelay(10);
 	hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 33cb5e5a249b..10ccf278d5be 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -271,7 +271,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 	/* take a deep breath */
 	msleep(50);
 
-	if (IDE_CONTROL_REG) {
+	if (hwif->io_ports[IDE_CONTROL_OFFSET]) {
 		a = ide_read_altstatus(drive);
 		s = ide_read_status(drive);
 		if ((a ^ s) & ~INDEX_STAT)
@@ -289,10 +289,10 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 	 */
 	if ((cmd == WIN_PIDENTIFY))
 		/* disable dma & overlap */
-		hwif->OUTB(0, IDE_FEATURE_REG);
+		hwif->OUTB(0, hwif->io_ports[IDE_FEATURE_OFFSET]);
 
 	/* ask drive for ID */
-	hwif->OUTB(cmd, IDE_COMMAND_REG);
+	hwif->OUTB(cmd, hwif->io_ports[IDE_COMMAND_OFFSET]);
 
 	timeout = ((cmd == WIN_IDENTIFY) ? WAIT_WORSTCASE : WAIT_PIDENTIFY) / 2;
 	timeout += jiffies;
@@ -353,7 +353,7 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 	 * interrupts during the identify-phase that
 	 * the irq handler isn't expecting.
 	 */
-	if (IDE_CONTROL_REG) {
+	if (hwif->io_ports[IDE_CONTROL_OFFSET]) {
 		if (!hwif->irq) {
 			autoprobe = 1;
 			cookie = probe_irq_on();
@@ -445,7 +445,8 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 	msleep(50);
 	SELECT_DRIVE(drive);
 	msleep(50);
-	if (hwif->INB(IDE_SELECT_REG) != drive->select.all && !drive->present) {
+	if (hwif->INB(hwif->io_ports[IDE_SELECT_OFFSET]) != drive->select.all &&
+	    !drive->present) {
 		if (drive->select.b.unit != 0) {
 			/* exit with drive0 selected */
 			SELECT_DRIVE(&hwif->drives[0]);
@@ -477,9 +478,11 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 			printk(KERN_ERR "%s: no response (status = 0x%02x), "
 					"resetting drive\n", drive->name, stat);
 			msleep(50);
-			hwif->OUTB(drive->select.all, IDE_SELECT_REG);
+			hwif->OUTB(drive->select.all,
+				   hwif->io_ports[IDE_SELECT_OFFSET]);
 			msleep(50);
-			hwif->OUTB(WIN_SRST, IDE_COMMAND_REG);
+			hwif->OUTB(WIN_SRST,
+				   hwif->io_ports[IDE_COMMAND_OFFSET]);
 			(void)ide_busy_sleep(hwif);
 			rc = try_to_identify(drive, cmd);
 		}
@@ -515,7 +518,7 @@ static void enable_nest (ide_drive_t *drive)
 	printk("%s: enabling %s -- ", hwif->name, drive->id->model);
 	SELECT_DRIVE(drive);
 	msleep(50);
-	hwif->OUTB(EXABYTE_ENABLE_NEST, IDE_COMMAND_REG);
+	hwif->OUTB(EXABYTE_ENABLE_NEST, hwif->io_ports[IDE_COMMAND_OFFSET]);
 
 	if (ide_busy_sleep(hwif)) {
 		printk(KERN_CONT "failed (timeout)\n");
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index bfdc4f449797..8c39146b6088 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1117,10 +1117,10 @@ static ide_startstop_t idetape_pc_intr(ide_drive_t *drive)
 		return ide_do_reset(drive);
 	}
 	/* Get the number of bytes to transfer on this interrupt. */
-	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
-		  hwif->INB(IDE_BCOUNTL_REG);
+	bcount = (hwif->INB(hwif->io_ports[IDE_BCOUNTH_OFFSET]) << 8) |
+		  hwif->INB(hwif->io_ports[IDE_BCOUNTL_OFFSET]);
 
-	ireason = hwif->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 
 	if (ireason & CD) {
 		printk(KERN_ERR "ide-tape: CoD != 0 in %s\n", __func__);
@@ -1224,12 +1224,12 @@ static ide_startstop_t idetape_transfer_pc(ide_drive_t *drive)
 				"yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason = hwif->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 	while (retries-- && ((ireason & CD) == 0 || (ireason & IO))) {
 		printk(KERN_ERR "ide-tape: (IO,CoD != (0,1) while issuing "
 				"a packet command, retrying\n");
 		udelay(100);
-		ireason = hwif->INB(IDE_IREASON_REG);
+		ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 		if (retries == 0) {
 			printk(KERN_ERR "ide-tape: (IO,CoD != (0,1) while "
 					"issuing a packet command, ignoring\n");
@@ -1323,7 +1323,7 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive, idetape_pc_t *pc)
 				    IDETAPE_WAIT_CMD, NULL);
 		return ide_started;
 	} else {
-		hwif->OUTB(WIN_PACKETCMD, IDE_COMMAND_REG);
+		hwif->OUTB(WIN_PACKETCMD, hwif->io_ports[IDE_COMMAND_OFFSET]);
 		return idetape_transfer_pc(drive);
 	}
 }
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 4c86a8d84b4c..155cc904f4eb 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -59,32 +59,34 @@ void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 		SELECT_MASK(drive, 0);
 
 	if (task->tf_flags & IDE_TFLAG_OUT_DATA)
-		hwif->OUTW((tf->hob_data << 8) | tf->data, IDE_DATA_REG);
+		hwif->OUTW((tf->hob_data << 8) | tf->data,
+			   hwif->io_ports[IDE_DATA_OFFSET]);
 
 	if (task->tf_flags & IDE_TFLAG_OUT_HOB_FEATURE)
-		hwif->OUTB(tf->hob_feature, IDE_FEATURE_REG);
+		hwif->OUTB(tf->hob_feature, hwif->io_ports[IDE_FEATURE_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_HOB_NSECT)
-		hwif->OUTB(tf->hob_nsect, IDE_NSECTOR_REG);
+		hwif->OUTB(tf->hob_nsect, hwif->io_ports[IDE_NSECTOR_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_HOB_LBAL)
-		hwif->OUTB(tf->hob_lbal, IDE_SECTOR_REG);
+		hwif->OUTB(tf->hob_lbal, hwif->io_ports[IDE_SECTOR_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_HOB_LBAM)
-		hwif->OUTB(tf->hob_lbam, IDE_LCYL_REG);
+		hwif->OUTB(tf->hob_lbam, hwif->io_ports[IDE_LCYL_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_HOB_LBAH)
-		hwif->OUTB(tf->hob_lbah, IDE_HCYL_REG);
+		hwif->OUTB(tf->hob_lbah, hwif->io_ports[IDE_HCYL_OFFSET]);
 
 	if (task->tf_flags & IDE_TFLAG_OUT_FEATURE)
-		hwif->OUTB(tf->feature, IDE_FEATURE_REG);
+		hwif->OUTB(tf->feature, hwif->io_ports[IDE_FEATURE_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_NSECT)
-		hwif->OUTB(tf->nsect, IDE_NSECTOR_REG);
+		hwif->OUTB(tf->nsect, hwif->io_ports[IDE_NSECTOR_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_LBAL)
-		hwif->OUTB(tf->lbal, IDE_SECTOR_REG);
+		hwif->OUTB(tf->lbal, hwif->io_ports[IDE_SECTOR_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_LBAM)
-		hwif->OUTB(tf->lbam, IDE_LCYL_REG);
+		hwif->OUTB(tf->lbam, hwif->io_ports[IDE_LCYL_OFFSET]);
 	if (task->tf_flags & IDE_TFLAG_OUT_LBAH)
-		hwif->OUTB(tf->lbah, IDE_HCYL_REG);
+		hwif->OUTB(tf->lbah, hwif->io_ports[IDE_HCYL_OFFSET]);
 
 	if (task->tf_flags & IDE_TFLAG_OUT_DEVICE)
-		hwif->OUTB((tf->device & HIHI) | drive->select.all, IDE_SELECT_REG);
+		hwif->OUTB((tf->device & HIHI) | drive->select.all,
+			   hwif->io_ports[IDE_SELECT_OFFSET]);
 }
 
 int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
@@ -152,7 +154,8 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	switch (task->data_phase) {
 	case TASKFILE_MULTI_OUT:
 	case TASKFILE_OUT:
-		hwif->OUTBSYNC(drive, tf->command, IDE_COMMAND_REG);
+		hwif->OUTBSYNC(drive, tf->command,
+			       hwif->io_ports[IDE_COMMAND_OFFSET]);
 		ndelay(400);	/* FIXME */
 		return pre_task_out_intr(drive, task->rq);
 	case TASKFILE_MULTI_IN:
diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c
index 78ca68e60f97..314e6c6aeb6c 100644
--- a/drivers/ide/legacy/ht6560b.c
+++ b/drivers/ide/legacy/ht6560b.c
@@ -82,7 +82,7 @@
  * out how they setup those cycle time interfacing values, as they at Holtek
  * call them. IDESETUP.COM that is supplied with the drivers figures out
  * optimal values and fetches those values to drivers. I found out that
- * they use IDE_SELECT_REG to fetch timings to the ide board right after
+ * they use Select register to fetch timings to the ide board right after
  * interface switching. After that it was quite easy to add code to
  * ht6560b.c.
  *
@@ -127,6 +127,7 @@
  */
 static void ht6560b_selectproc (ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long flags;
 	static u8 current_select = 0;
 	static u8 current_timing = 0;
@@ -155,8 +156,8 @@ static void ht6560b_selectproc (ide_drive_t *drive)
 		/*
 		 * Set timing for this drive:
 		 */
-		outb(timing, IDE_SELECT_REG);
-		(void)inb(IDE_STATUS_REG);
+		outb(timing, hwif->io_ports[IDE_SELECT_OFFSET]);
+		(void)inb(hwif->io_ports[IDE_STATUS_OFFSET]);
 #ifdef DEBUG
 		printk("ht6560b: %s: select=%#x timing=%#x\n",
 			drive->name, select, timing);
@@ -193,9 +194,9 @@ static int __init try_to_init_ht6560b(void)
 	 * Ht6560b autodetected
 	 */
 	outb(HT_CONFIG_DEFAULT, HT_CONFIG_PORT);
-	outb(HT_TIMING_DEFAULT, 0x1f6);  /* IDE_SELECT_REG */
-	(void) inb(0x1f7);               /* IDE_STATUS_REG */
-	
+	outb(HT_TIMING_DEFAULT, 0x1f6);	/* Select register */
+	(void)inb(0x1f7);		/* Status register */
+
 	printk("ht6560b " HT6560B_VERSION
 	       ": chipset detected and initialized"
 #ifdef DEBUG
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index d03a231d965e..82d0e318a1fe 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -760,7 +760,7 @@ static void hpt3xx_maskproc(ide_drive_t *drive, int mask)
 		}
 	} else
 		outb(mask ? (drive->ctl | 2) : (drive->ctl & ~2),
-		     IDE_CONTROL_REG);
+		     hwif->io_ports[IDE_CONTROL_OFFSET]);
 }
 
 /*
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 1a560dc1eac1..ef07c7a8b97a 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -334,7 +334,8 @@ static int scc_ide_dma_end(ide_drive_t * drive)
 
 	/* errata A308 workaround: Step5 (check data loss) */
 	/* We don't check non ide_disk because it is limited to UDMA4 */
-	if (!(in_be32((void __iomem *)IDE_ALTSTATUS_REG) & ERR_STAT) &&
+	if (!(in_be32((void __iomem *)hwif->io_ports[IDE_ALTSTATUS_OFFSET])
+	      & ERR_STAT) &&
 	    drive->media == ide_disk && drive->current_speed > XFER_UDMA_4) {
 		reg = in_be32((void __iomem *)intsts_port);
 		if (!(reg & INTSTS_ACTEINT)) {
@@ -437,7 +438,8 @@ static int scc_dma_test_irq(ide_drive_t *drive)
 	u32 int_stat = in_be32((void __iomem *)hwif->dma_base + 0x014);
 
 	/* SCC errata A252,A308 workaround: Step4 */
-	if ((in_be32((void __iomem *)IDE_ALTSTATUS_REG) & ERR_STAT) &&
+	if ((in_be32((void __iomem *)hwif->io_ports[IDE_ALTSTATUS_OFFSET])
+	     & ERR_STAT) &&
 	    (int_stat & INTSTS_INTRQ))
 		return 1;
 
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 9046a69117ff..9d1a3038af9b 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -112,10 +112,9 @@ static void
 sgiioc4_maskproc(ide_drive_t * drive, int mask)
 {
 	writeb(mask ? (drive->ctl | 2) : (drive->ctl & ~2),
-	       (void __iomem *)IDE_CONTROL_REG);
+	       (void __iomem *)drive->hwif->io_ports[IDE_CONTROL_OFFSET]);
 }
 
-
 static int
 sgiioc4_checkirq(ide_hwif_t * hwif)
 {
@@ -142,18 +141,18 @@ sgiioc4_clearirq(ide_drive_t * drive)
 	intr_reg = readl((void __iomem *)other_ir);
 	if (intr_reg & 0x03) { /* Valid IOC4-IDE interrupt */
 		/*
-		 * Using sgiioc4_INB to read the IDE_STATUS_REG has a side effect
-		 * of clearing the interrupt.  The first read should clear it
-		 * if it is set.  The second read should return a "clear" status
-		 * if it got cleared.  If not, then spin for a bit trying to
-		 * clear it.
+		 * Using sgiioc4_INB to read the Status register has a side
+		 * effect of clearing the interrupt.  The first read should
+		 * clear it if it is set.  The second read should return
+		 * a "clear" status if it got cleared.  If not, then spin
+		 * for a bit trying to clear it.
 		 */
-		u8 stat = sgiioc4_INB(IDE_STATUS_REG);
+		u8 stat = sgiioc4_INB(hwif->io_ports[IDE_STATUS_OFFSET]);
 		int count = 0;
-		stat = sgiioc4_INB(IDE_STATUS_REG);
+		stat = sgiioc4_INB(hwif->io_ports[IDE_STATUS_OFFSET]);
 		while ((stat & 0x80) && (count++ < 100)) {
 			udelay(1);
-			stat = sgiioc4_INB(IDE_STATUS_REG);
+			stat = sgiioc4_INB(hwif->io_ports[IDE_STATUS_OFFSET]);
 		}
 
 		if (intr_reg & 0x02) {
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index b37dcfbdaad6..03a77713caf2 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -450,7 +450,8 @@ pmac_ide_init_hwif_ports(hw_regs_t *hw,
 	hw->dev = &pmac_ide[ix].mdev->ofdev.dev;
 }
 
-#define PMAC_IDE_REG(x) ((void __iomem *)(IDE_DATA_REG+(x)))
+#define PMAC_IDE_REG(x) \
+	((void __iomem *)((drive)->hwif->io_ports[IDE_DATA_OFFSET] + (x)))
 
 /*
  * Apply the timings of the proper unit (master/slave) to the shared
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 3c3b3502c4d4..7fea769cf291 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -275,9 +275,12 @@ static int idescsi_end_request(ide_drive_t *, int, int);
 static ide_startstop_t
 idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 {
+	ide_hwif_t *hwif = drive->hwif;
+
 	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		HWIF(drive)->OUTB(WIN_IDLEIMMEDIATE,IDE_COMMAND_REG);
+		hwif->OUTB(WIN_IDLEIMMEDIATE,
+			   hwif->io_ports[IDE_COMMAND_OFFSET]);
 
 	rq->errors++;
 
@@ -423,9 +426,9 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 		idescsi_end_request (drive, 1, 0);
 		return ide_stopped;
 	}
-	bcount = (hwif->INB(IDE_BCOUNTH_REG) << 8) |
-		  hwif->INB(IDE_BCOUNTL_REG);
-	ireason = hwif->INB(IDE_IREASON_REG);
+	bcount = (hwif->INB(hwif->io_ports[IDE_BCOUNTH_OFFSET]) << 8) |
+		  hwif->INB(hwif->io_ports[IDE_BCOUNTL_OFFSET]);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 
 	if (ireason & CD) {
 		printk(KERN_ERR "ide-scsi: CoD != 0 in idescsi_pc_intr\n");
@@ -497,7 +500,7 @@ static ide_startstop_t idescsi_transfer_pc(ide_drive_t *drive)
 			"initiated yet DRQ isn't asserted\n");
 		return startstop;
 	}
-	ireason = hwif->INB(IDE_IREASON_REG);
+	ireason = hwif->INB(hwif->io_ports[IDE_IREASON_OFFSET]);
 	if ((ireason & CD) == 0 || (ireason & IO)) {
 		printk(KERN_ERR "ide-scsi: (IO,CoD) != (0,1) while "
 				"issuing a packet command\n");
@@ -587,7 +590,7 @@ static ide_startstop_t idescsi_issue_pc (ide_drive_t *drive, idescsi_pc_t *pc)
 		return ide_started;
 	} else {
 		/* Issue the packet command */
-		HWIF(drive)->OUTB(WIN_PACKETCMD, IDE_COMMAND_REG);
+		hwif->OUTB(WIN_PACKETCMD, hwif->io_ports[IDE_COMMAND_OFFSET]);
 		return idescsi_transfer_pc(drive);
 	}
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 2eb99cab4a3d..3b691cce00e1 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -82,24 +82,10 @@ typedef unsigned char	byte;	/* used everywhere */
 
 #define IDE_FEATURE_OFFSET	IDE_ERROR_OFFSET
 #define IDE_COMMAND_OFFSET	IDE_STATUS_OFFSET
-
-#define IDE_DATA_REG		(HWIF(drive)->io_ports[IDE_DATA_OFFSET])
-#define IDE_ERROR_REG		(HWIF(drive)->io_ports[IDE_ERROR_OFFSET])
-#define IDE_NSECTOR_REG		(HWIF(drive)->io_ports[IDE_NSECTOR_OFFSET])
-#define IDE_SECTOR_REG		(HWIF(drive)->io_ports[IDE_SECTOR_OFFSET])
-#define IDE_LCYL_REG		(HWIF(drive)->io_ports[IDE_LCYL_OFFSET])
-#define IDE_HCYL_REG		(HWIF(drive)->io_ports[IDE_HCYL_OFFSET])
-#define IDE_SELECT_REG		(HWIF(drive)->io_ports[IDE_SELECT_OFFSET])
-#define IDE_STATUS_REG		(HWIF(drive)->io_ports[IDE_STATUS_OFFSET])
-#define IDE_CONTROL_REG		(HWIF(drive)->io_ports[IDE_CONTROL_OFFSET])
-#define IDE_IRQ_REG		(HWIF(drive)->io_ports[IDE_IRQ_OFFSET])
-
-#define IDE_FEATURE_REG		IDE_ERROR_REG
-#define IDE_COMMAND_REG		IDE_STATUS_REG
-#define IDE_ALTSTATUS_REG	IDE_CONTROL_REG
-#define IDE_IREASON_REG		IDE_NSECTOR_REG
-#define IDE_BCOUNTL_REG		IDE_LCYL_REG
-#define IDE_BCOUNTH_REG		IDE_HCYL_REG
+#define IDE_ALTSTATUS_OFFSET	IDE_CONTROL_OFFSET
+#define IDE_IREASON_OFFSET	IDE_NSECTOR_OFFSET
+#define IDE_BCOUNTL_OFFSET	IDE_LCYL_OFFSET
+#define IDE_BCOUNTH_OFFSET	IDE_HCYL_OFFSET
 
 #define OK_STAT(stat,good,bad)	(((stat)&((good)|(bad)))==(good))
 #define BAD_R_STAT		(BUSY_STAT   | ERR_STAT)
@@ -369,7 +355,7 @@ typedef struct ide_drive_s {
         u8	wcache;		/* status of write cache */
 	u8	acoustic;	/* acoustic management */
 	u8	media;		/* disk, cdrom, tape, floppy, ... */
-	u8	ctl;		/* "normal" value for IDE_CONTROL_REG */
+	u8	ctl;		/* "normal" value for Control register */
 	u8	ready_stat;	/* min status value for drive ready */
 	u8	mult_count;	/* current multiple sector setting */
 	u8	mult_req;	/* requested multiple sector setting */
@@ -1273,7 +1259,10 @@ static inline ide_drive_t *ide_get_paired_drive(ide_drive_t *drive)
 
 static inline void ide_set_irq(ide_drive_t *drive, int on)
 {
-	drive->hwif->OUTB(drive->ctl | (on ? 0 : 2), IDE_CONTROL_REG);
+	ide_hwif_t *hwif = drive->hwif;
+
+	hwif->OUTB(drive->ctl | (on ? 0 : 2),
+		   hwif->io_ports[IDE_CONTROL_OFFSET]);
 }
 
 static inline u8 ide_read_status(ide_drive_t *drive)
-- 
cgit v1.2.3


From 8303b46e18b58b2d0257e6842e60b50ac880a6d1 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Fri, 18 Apr 2008 00:46:26 +0200
Subject: ide: add generic packet command representation ide_atapi_pc

This new struct unifies ide{-floppy,-tape,-scsi}'s view of a packet command. For now,
it represents the common denominator between the three drivers while adding driver-
specific members at the end of the struct which will be merged/simplified into the
generic ATAPI handling code in later steps, or removed completely.

Bart:
- move struct ide_atapi_pc outside of #ifdef/#endif CONFIG_IDE_PROC_FS

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 47 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)

(limited to 'include')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index 3b691cce00e1..e9b1def1e66a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -595,6 +595,53 @@ int set_io_32bit(ide_drive_t *, int);
 int set_pio_mode(ide_drive_t *, int);
 int set_using_dma(ide_drive_t *, int);
 
+struct ide_atapi_pc {
+	/* actual packet bytes */
+	u8 c[12];
+	/* incremented on each retry */
+	int retries;
+	int error;
+
+	/* bytes to transfer */
+	int req_xfer;
+	/* bytes actually transferred */
+	int xferred;
+
+	/* data buffer */
+	u8 *buf;
+	/* current buffer position */
+	u8 *cur_pos;
+	int buf_size;
+	/* missing/available data on the current buffer */
+	int b_count;
+
+	/* the corresponding request */
+	struct request *rq;
+
+	unsigned long flags;
+
+	/*
+	 * those are more or less driver-specific and some of them are subject
+	 * to change/removal later.
+	 */
+	u8 pc_buf[256];
+	void (*idefloppy_callback) (ide_drive_t *);
+	ide_startstop_t (*idetape_callback) (ide_drive_t *);
+
+	/* idetape only */
+	struct idetape_bh *bh;
+	char *b_data;
+
+	/* idescsi only for now */
+	struct scatterlist *sg;
+	unsigned int sg_cnt;
+
+	struct scsi_cmnd *scsi_cmd;
+	void (*done) (struct scsi_cmnd *);
+
+	unsigned long timeout;
+};
+
 #ifdef CONFIG_IDE_PROC_FS
 /*
  * configurable drive settings
-- 
cgit v1.2.3


From eaec3e7ded9dbc88bad393c076b65f4b7b11d30d Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Fri, 18 Apr 2008 00:46:27 +0200
Subject: ide: use generic ATAPI packet command flags in ide-{floppy,tape}

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 14 --------------
 drivers/ide/ide-tape.c   | 16 ----------------
 include/linux/ide.h      | 15 +++++++++++++++
 3 files changed, 15 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index bf1ef60a5a07..5f133dfb541c 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -78,20 +78,6 @@
  */
 #define IDEFLOPPY_PC_STACK		(10 + IDEFLOPPY_MAX_PC_RETRIES)
 
-/* Packet command flag bits. */
-enum {
-	/* 1 when we prefer to use DMA if possible */
-	PC_FLAG_DMA_RECOMMENDED	= (1 << 0),
-	/* 1 while DMA in progress */
-	PC_FLAG_DMA_IN_PROGRESS	= (1 << 1),
-	/* 1 when encountered problem during DMA */
-	PC_FLAG_DMA_ERROR	= (1 << 2),
-	/* Data direction */
-	PC_FLAG_WRITING		= (1 << 3),
-	/* Suppress error reporting */
-	PC_FLAG_SUPPRESS_ERROR	= (1 << 4),
-};
-
 /* format capacities descriptor codes */
 #define CAPACITY_INVALID	0x00
 #define CAPACITY_UNFORMATTED	0x01
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 3f9dcca6f092..f43fd070f1b6 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -181,22 +181,6 @@ struct idetape_bh {
 	char *b_data;
 };
 
-/* Packet command flag bits. */
-enum {
-	/* Set when an error is considered normal - We won't retry */
-	PC_FLAG_ABORT		= (1 << 0),
-	/* 1 When polling for DSC on a media access command */
-	PC_FLAG_WAIT_FOR_DSC	= (1 << 1),
-	/* 1 when we prefer to use DMA if possible */
-	PC_FLAG_DMA_RECOMMENDED	= (1 << 2),
-	/* 1 while DMA in progress */
-	PC_FLAG_DMA_IN_PROGRESS	= (1 << 3),
-	/* 1 when encountered problem during DMA */
-	PC_FLAG_DMA_ERROR	= (1 << 4),
-	/* Data direction */
-	PC_FLAG_WRITING		= (1 <<	5),
-};
-
 /* Tape door status */
 #define DOOR_UNLOCKED			0
 #define DOOR_LOCKED			1
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e9b1def1e66a..65445b7efc63 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -595,6 +595,21 @@ int set_io_32bit(ide_drive_t *, int);
 int set_pio_mode(ide_drive_t *, int);
 int set_using_dma(ide_drive_t *, int);
 
+/* ATAPI packet command flags */
+enum {
+	/* set when an error is considered normal - no retry (ide-tape) */
+	PC_FLAG_ABORT			= (1 << 0),
+	PC_FLAG_SUPPRESS_ERROR		= (1 << 1),
+	PC_FLAG_WAIT_FOR_DSC		= (1 << 2),
+	PC_FLAG_DMA_OK			= (1 << 3),
+	PC_FLAG_DMA_RECOMMENDED		= (1 << 4),
+	PC_FLAG_DMA_IN_PROGRESS		= (1 << 5),
+	PC_FLAG_DMA_ERROR		= (1 << 6),
+	PC_FLAG_WRITING			= (1 << 7),
+	/* command timed out */
+	PC_FLAG_TIMEDOUT		= (1 << 8),
+};
+
 struct ide_atapi_pc {
 	/* actual packet bytes */
 	u8 c[12];
-- 
cgit v1.2.3


From 1d850bd0b291efd1740b8d87ce025363436eb6dc Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:32 +0200
Subject: ppc: remove ppc_ide_md

* Add special cases for pplus and prep to ide_default_{irq,io_base}()
  (+ FIXMEs about the need to use IDE platform host driver instead).

* Remove no longer needed ppc_ide_md and struct ide_machdep_calls.

* Then remove <linux/ide.h> include from:
  - arch/powerpc/kernel/setup_32.c
  - arch/ppc/kernel/ppc_ksyms.c
  - arch/ppc/kernel/setup.c
  - arch/ppc/platforms/pplus.c
  - arch/ppc/platforms/prep_setup.c

There should be no functional changes caused by this patch.

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 arch/powerpc/kernel/setup_32.c  |  8 -------
 arch/ppc/kernel/ppc_ksyms.c     |  5 -----
 arch/ppc/kernel/setup.c         |  2 --
 arch/ppc/platforms/pplus.c      | 35 ------------------------------
 arch/ppc/platforms/prep_setup.c | 38 ---------------------------------
 drivers/ide/ide.c               |  4 ----
 include/asm-powerpc/ide.h       | 47 ++++++++++++++++++++++++++++-------------
 7 files changed, 32 insertions(+), 107 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index cd870a823d18..06d918d94dd1 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -10,9 +10,6 @@
 #include <linux/reboot.h>
 #include <linux/delay.h>
 #include <linux/initrd.h>
-#if defined(CONFIG_IDE) || defined(CONFIG_IDE_MODULE)
-#include <linux/ide.h>
-#endif
 #include <linux/tty.h>
 #include <linux/bootmem.h>
 #include <linux/seq_file.h>
@@ -51,11 +48,6 @@
 
 extern void bootx_init(unsigned long r4, unsigned long phys);
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
-struct ide_machdep_calls ppc_ide_md;
-EXPORT_SYMBOL(ppc_ide_md);
-#endif
-
 int boot_cpuid;
 EXPORT_SYMBOL_GPL(boot_cpuid);
 int boot_cpuid_phys;
diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c
index c35350250cfc..2ba659f401be 100644
--- a/arch/ppc/kernel/ppc_ksyms.c
+++ b/arch/ppc/kernel/ppc_ksyms.c
@@ -12,7 +12,6 @@
 #include <linux/irq.h>
 #include <linux/pci.h>
 #include <linux/delay.h>
-#include <linux/ide.h>
 #include <linux/pm.h>
 #include <linux/bitops.h>
 
@@ -124,10 +123,6 @@ EXPORT_SYMBOL(__ioremap);
 EXPORT_SYMBOL(iounmap);
 EXPORT_SYMBOL(ioremap_bot);	/* aka VMALLOC_END */
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
-EXPORT_SYMBOL(ppc_ide_md);
-#endif
-
 #ifdef CONFIG_PCI
 EXPORT_SYMBOL(isa_io_base);
 EXPORT_SYMBOL(isa_mem_base);
diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c
index 294055902f0c..bfddfdee0b65 100644
--- a/arch/ppc/kernel/setup.c
+++ b/arch/ppc/kernel/setup.c
@@ -10,7 +10,6 @@
 #include <linux/reboot.h>
 #include <linux/delay.h>
 #include <linux/initrd.h>
-#include <linux/ide.h>
 #include <linux/screen_info.h>
 #include <linux/bootmem.h>
 #include <linux/seq_file.h>
@@ -57,7 +56,6 @@ extern void ppc6xx_idle(void);
 extern void power4_idle(void);
 
 extern boot_infos_t *boot_infos;
-struct ide_machdep_calls ppc_ide_md;
 
 /* Used with the BI_MEMSIZE bootinfo parameter to store the memory
    size value reported by the boot loader. */
diff --git a/arch/ppc/platforms/pplus.c b/arch/ppc/platforms/pplus.c
index 384cdfea70fa..cbcac85c7a78 100644
--- a/arch/ppc/platforms/pplus.c
+++ b/arch/ppc/platforms/pplus.c
@@ -19,7 +19,6 @@
 #include <linux/ioport.h>
 #include <linux/console.h>
 #include <linux/pci.h>
-#include <linux/ide.h>
 #include <linux/seq_file.h>
 #include <linux/root_dev.h>
 
@@ -668,35 +667,6 @@ static void __init pplus_init_IRQ(void)
 		ppc_md.progress("init_irq: exit", 0);
 }
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
-/*
- * IDE stuff.
- */
-static int pplus_ide_default_irq(unsigned long base)
-{
-	switch (base) {
-	case 0x1f0:
-		return 14;
-	case 0x170:
-		return 15;
-	default:
-		return 0;
-	}
-}
-
-static unsigned long pplus_ide_default_io_base(int index)
-{
-	switch (index) {
-	case 0:
-		return 0x1f0;
-	case 1:
-		return 0x170;
-	default:
-		return 0;
-	}
-}
-#endif
-
 #ifdef CONFIG_SMP
 /* PowerPlus (MTX) support */
 static int __init smp_pplus_probe(void)
@@ -862,11 +832,6 @@ platform_init(unsigned long r3, unsigned long r4, unsigned long r5,
 	ppc_md.find_end_of_memory = pplus_find_end_of_memory;
 	ppc_md.setup_io_mappings = pplus_map_io;
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
-	ppc_ide_md.default_irq = pplus_ide_default_irq;
-	ppc_ide_md.default_io_base = pplus_ide_default_io_base;
-#endif
-
 #ifdef CONFIG_SERIAL_TEXT_DEBUG
 	ppc_md.progress = gen550_progress;
 #endif				/* CONFIG_SERIAL_TEXT_DEBUG */
diff --git a/arch/ppc/platforms/prep_setup.c b/arch/ppc/platforms/prep_setup.c
index 38449855d5ff..465b658c927d 100644
--- a/arch/ppc/platforms/prep_setup.c
+++ b/arch/ppc/platforms/prep_setup.c
@@ -33,7 +33,6 @@
 #include <linux/console.h>
 #include <linux/timex.h>
 #include <linux/pci.h>
-#include <linux/ide.h>
 #include <linux/seq_file.h>
 #include <linux/root_dev.h>
 
@@ -894,38 +893,6 @@ prep_init_IRQ(void)
 		i8259_init(MPC10X_MAPA_PCI_INTACK_ADDR, 0);
 }
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
-/*
- * IDE stuff.
- */
-static int
-prep_ide_default_irq(unsigned long base)
-{
-	switch (base) {
-		case 0x1f0: return 13;
-		case 0x170: return 13;
-		case 0x1e8: return 11;
-		case 0x168: return 10;
-		case 0xfff0: return 14;		/* MCP(N)750 ide0 */
-		case 0xffe0: return 15;		/* MCP(N)750 ide1 */
-		default: return 0;
-	}
-}
-
-static unsigned long
-prep_ide_default_io_base(int index)
-{
-	switch (index) {
-		case 0: return 0x1f0;
-		case 1: return 0x170;
-		case 2: return 0x1e8;
-		case 3: return 0x168;
-		default:
-			return 0;
-	}
-}
-#endif
-
 #ifdef CONFIG_SMP
 /* PReP (MTX) support */
 static int __init
@@ -1070,11 +1037,6 @@ prep_init(unsigned long r3, unsigned long r4, unsigned long r5,
 
 	ppc_md.setup_io_mappings = prep_map_io;
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
-	ppc_ide_md.default_irq = prep_ide_default_irq;
-	ppc_ide_md.default_io_base = prep_ide_default_io_base;
-#endif
-
 #ifdef CONFIG_SMP
 	smp_ops			 = &prep_smp_ops;
 #endif /* CONFIG_SMP */
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 7e6418fe3a84..c2fb5c964a51 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -205,10 +205,6 @@ static void __init init_ide_data (void)
 #ifdef CONFIG_IDE_ARCH_OBSOLETE_INIT
 		memset(&hw, 0, sizeof(hw));
 		ide_std_init_ports(&hw, io_addr, ctl_addr);
-# ifdef CONFIG_PPC32
-		if (ppc_ide_md.ide_init_hwif)
-			ppc_ide_md.ide_init_hwif(&hw, io_addr, 0, &hwif->irq);
-# endif
 		memcpy(hwif->io_ports, hw.io_ports, sizeof(hw.io_ports));
 #endif
 		hwif->noprobe = !hwif->io_ports[IDE_DATA_OFFSET];
diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index 6d50310ecaea..06549456c953 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -31,30 +31,47 @@
 #include <linux/hdreg.h>
 #include <linux/ioport.h>
 
-struct ide_machdep_calls {
-        int         (*default_irq)(unsigned long base);
-        unsigned long (*default_io_base)(int index);
-        void        (*ide_init_hwif)(hw_regs_t *hw,
-                                     unsigned long data_port,
-                                     unsigned long ctrl_port,
-                                     int *irq);
-};
-
-extern struct ide_machdep_calls ppc_ide_md;
-
 #define IDE_ARCH_OBSOLETE_DEFAULTS
 
+/* FIXME: use ide_platform host driver */
 static __inline__ int ide_default_irq(unsigned long base)
 {
-	if (ppc_ide_md.default_irq)
-		return ppc_ide_md.default_irq(base);
+#ifdef CONFIG_PPLUS
+	switch (base) {
+	case 0x1f0:	return 14;
+	case 0x170:	return 15;
+	}
+#endif
+#ifdef CONFIG_PPC_PREP
+	switch (base) {
+	case 0x1f0:	return 13;
+	case 0x170:	return 13;
+	case 0x1e8:	return 11;
+	case 0x168:	return 10;
+	case 0xfff0:	return 14;	/* MCP(N)750 ide0 */
+	case 0xffe0:	return 15;	/* MCP(N)750 ide1 */
+	}
+#endif
 	return 0;
 }
 
+/* FIXME: use ide_platform host driver */
 static __inline__ unsigned long ide_default_io_base(int index)
 {
-	if (ppc_ide_md.default_io_base)
-		return ppc_ide_md.default_io_base(index);
+#ifdef CONFIG_PPLUS
+	switch (index) {
+	case 0:		return 0x1f0;
+	case 1:		return 0x170;
+	}
+#endif
+#ifdef CONFIG_PPC_PREP
+	switch (index) {
+	case 0:		return 0x1f0;
+	case 1:		return 0x170;
+	case 2:		return 0x1e8;
+	case 3:		return 0x168;
+	}
+#endif
 	return 0;
 }
 
-- 
cgit v1.2.3


From 0e33555fffdc8490630d98070e76e5fe031bcac2 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:33 +0200
Subject: ide: add CONFIG_IDE_ARCH_OBSOLETE_DEFAULTS (take 2)

* Add CONFIG_IDE_ARCH_OBSOLETE_DEFAULTS to drivers/ide/Kconfig and use
  it instead of defining IDE_ARCH_OBSOLETE_DEFAULTS in <arch/ide.h>.

v2:
* Define ide_default_irq() in ide-probe.c/ns87415.c if not already defined
  and drop defining ide_default_irq() for CONFIG_IDE_ARCH_OBSOLETE_DEFAULTS=n.

  [ Thanks to Stephen Rothwell and David Miller for noticing the problem. ]

Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Kconfig                 | 3 +++
 drivers/ide/ide-probe.c             | 4 ++++
 drivers/ide/ide.c                   | 4 ++++
 drivers/ide/pci/ns87415.c           | 4 ++++
 include/asm-alpha/ide.h             | 3 ---
 include/asm-ia64/ide.h              | 2 --
 include/asm-m32r/ide.h              | 2 --
 include/asm-mips/mach-generic/ide.h | 2 --
 include/asm-powerpc/ide.h           | 2 --
 include/asm-x86/ide.h               | 2 --
 include/linux/ide.h                 | 7 -------
 11 files changed, 15 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 4dc2761e9704..a57893c03b7a 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -1092,6 +1092,9 @@ config BLK_DEV_IDEDMA
 config IDE_ARCH_OBSOLETE_INIT
 	def_bool ALPHA || (ARM && !ARCH_L7200) || BLACKFIN || X86 || IA64 || M32R || MIPS || PARISC || PPC || (SUPERH64 && BLK_DEV_IDEPCI) || SPARC
 
+config IDE_ARCH_OBSOLETE_DEFAULTS
+	def_bool ALPHA || X86 || IA64 || M32R || MIPS || PPC32
+
 endif
 
 config BLK_DEV_HD_ONLY
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 10ccf278d5be..6a196c27b0aa 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1233,6 +1233,10 @@ static void drive_release_dev (struct device *dev)
 	complete(&drive->gendev_rel_comp);
 }
 
+#ifndef ide_default_irq
+#define ide_default_irq(irq) 0
+#endif
+
 static int hwif_init(ide_hwif_t *hwif)
 {
 	int old_irq;
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index c2fb5c964a51..a1a02c74d77f 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -165,6 +165,10 @@ static void ide_port_init_devices_data(ide_hwif_t *hwif)
 	}
 }
 
+#ifndef CONFIG_IDE_ARCH_OBSOLETE_DEFAULTS
+# define ide_default_io_base(index)	(0)
+# define ide_init_default_irq(base)	(0)
+#endif
 
 /*
  * init_ide_data() sets reasonable default values into all fields
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index bf0d3b2931f1..75513320aad9 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -181,6 +181,10 @@ static int ns87415_ide_dma_setup(ide_drive_t *drive)
 	return 1;
 }
 
+#ifndef ide_default_irq
+#define ide_default_irq(irq) 0
+#endif
+
 static void __devinit init_hwif_ns87415 (ide_hwif_t *hwif)
 {
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
diff --git a/include/asm-alpha/ide.h b/include/asm-alpha/ide.h
index b7bf68d0407b..a2feed30bb68 100644
--- a/include/asm-alpha/ide.h
+++ b/include/asm-alpha/ide.h
@@ -13,9 +13,6 @@
 
 #ifdef __KERNEL__
 
-
-#define IDE_ARCH_OBSOLETE_DEFAULTS
-
 static inline int ide_default_irq(unsigned long base)
 {
 	switch (base) {
diff --git a/include/asm-ia64/ide.h b/include/asm-ia64/ide.h
index 1ccf23809329..09c2a05e1c8a 100644
--- a/include/asm-ia64/ide.h
+++ b/include/asm-ia64/ide.h
@@ -16,8 +16,6 @@
 
 #include <linux/irq.h>
 
-#define IDE_ARCH_OBSOLETE_DEFAULTS
-
 static inline int ide_default_irq(unsigned long base)
 {
 	switch (base) {
diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
index 5d2044e529ab..feb7f0d7aca9 100644
--- a/include/asm-m32r/ide.h
+++ b/include/asm-m32r/ide.h
@@ -23,8 +23,6 @@
 # endif
 #endif
 
-#define IDE_ARCH_OBSOLETE_DEFAULTS
-
 static __inline__ int ide_default_irq(unsigned long base)
 {
 	switch (base) {
diff --git a/include/asm-mips/mach-generic/ide.h b/include/asm-mips/mach-generic/ide.h
index 4ec2b930dfbb..45e24474cf43 100644
--- a/include/asm-mips/mach-generic/ide.h
+++ b/include/asm-mips/mach-generic/ide.h
@@ -27,8 +27,6 @@
 # endif
 #endif
 
-#define IDE_ARCH_OBSOLETE_DEFAULTS
-
 static __inline__ int ide_probe_legacy(void)
 {
 #ifdef CONFIG_PCI
diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index 06549456c953..fef2ef1dbe86 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -31,8 +31,6 @@
 #include <linux/hdreg.h>
 #include <linux/ioport.h>
 
-#define IDE_ARCH_OBSOLETE_DEFAULTS
-
 /* FIXME: use ide_platform host driver */
 static __inline__ int ide_default_irq(unsigned long base)
 {
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
index c2552d8bebf7..58080a7111de 100644
--- a/include/asm-x86/ide.h
+++ b/include/asm-x86/ide.h
@@ -20,8 +20,6 @@
 # endif
 #endif
 
-#define IDE_ARCH_OBSOLETE_DEFAULTS
-
 static __inline__ int ide_default_irq(unsigned long base)
 {
 	switch (base) {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 65445b7efc63..6c39482fd1a1 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -194,13 +194,6 @@ static inline void ide_std_init_ports(hw_regs_t *hw,
 #define MAX_HWIFS	CONFIG_IDE_MAX_HWIFS
 #endif
 
-/* needed on alpha, x86/x86_64, ia64, mips, ppc32 and sh */
-#ifndef IDE_ARCH_OBSOLETE_DEFAULTS
-# define ide_default_io_base(index)	(0)
-# define ide_default_irq(base)		(0)
-# define ide_init_default_irq(base)	(0)
-#endif
-
 /* Currently only m68k, apus and m8xx need it */
 #ifndef IDE_ARCH_ACK_INTR
 # define ide_ack_intr(hwif) (1)
-- 
cgit v1.2.3


From 9dfcd15a6dc1ef81307295e08b797fa9212be71a Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:34 +0200
Subject: ide: remove ide_default_io_ctl() macro

It is always == '((base) + 0x206)' if CONFIG_IDE_ARCH_OBSOLETE_DEFAULTS=y
and it is not needed otherwise (arm, blackfin, parisc, ppc64, sh, sparc[64]).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide.c                   | 2 +-
 include/asm-alpha/ide.h             | 2 --
 include/asm-arm/ide.h               | 8 --------
 include/asm-blackfin/ide.h          | 4 ----
 include/asm-ia64/ide.h              | 2 --
 include/asm-m32r/ide.h              | 2 --
 include/asm-mips/mach-generic/ide.h | 2 --
 include/asm-parisc/ide.h            | 2 --
 include/asm-powerpc/ide.h           | 2 --
 include/asm-sh/ide.h                | 3 ---
 include/asm-sparc/ide.h             | 2 --
 include/asm-sparc64/ide.h           | 2 --
 include/asm-x86/ide.h               | 2 --
 13 files changed, 1 insertion(+), 34 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 9b9b686fbacb..a93f127b8bf2 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -202,7 +202,7 @@ static void __init init_ide_data (void)
 	for (index = 0; index < MAX_HWIFS; ++index) {
 		ide_hwif_t *hwif = &ide_hwifs[index];
 		unsigned long io_addr = ide_default_io_base(index);
-		unsigned long ctl_addr = ide_default_io_ctl(io_addr);
+		unsigned long ctl_addr = io_addr + 0x206;
 
 		ide_init_port_data(hwif, index);
 
diff --git a/include/asm-alpha/ide.h b/include/asm-alpha/ide.h
index a2feed30bb68..e67c38c96a6b 100644
--- a/include/asm-alpha/ide.h
+++ b/include/asm-alpha/ide.h
@@ -37,8 +37,6 @@ static inline unsigned long ide_default_io_base(int index)
 	}
 }
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #ifdef CONFIG_PCI
 #define ide_init_default_irq(base)	(0)
 #else
diff --git a/include/asm-arm/ide.h b/include/asm-arm/ide.h
index f348fcf3150b..88f4d231ce4f 100644
--- a/include/asm-arm/ide.h
+++ b/include/asm-arm/ide.h
@@ -17,14 +17,6 @@
 #define MAX_HWIFS	4
 #endif
 
-#if !defined(CONFIG_ARCH_L7200)
-# ifdef CONFIG_ARCH_CLPS7500
-#  define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-# else
-#  define ide_default_io_ctl(base)	(0)
-# endif
-#endif /* !ARCH_L7200 */
-
 #define __ide_mm_insw(port,addr,len)	readsw(port,addr,len)
 #define __ide_mm_insl(port,addr,len)	readsl(port,addr,len)
 #define __ide_mm_outsw(port,addr,len)	writesw(port,addr,len)
diff --git a/include/asm-blackfin/ide.h b/include/asm-blackfin/ide.h
index 121e272581d6..5b88de115bf4 100644
--- a/include/asm-blackfin/ide.h
+++ b/include/asm-blackfin/ide.h
@@ -19,10 +19,6 @@
 
 #define MAX_HWIFS	1
 
-/* Legacy ... BLK_DEV_IDECS */
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
-
 #include <asm-generic/ide_iops.h>
 
 /****************************************************************************/
diff --git a/include/asm-ia64/ide.h b/include/asm-ia64/ide.h
index 09c2a05e1c8a..989edb6f1285 100644
--- a/include/asm-ia64/ide.h
+++ b/include/asm-ia64/ide.h
@@ -44,8 +44,6 @@ static inline unsigned long ide_default_io_base(int index)
 	}
 }
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #ifdef CONFIG_PCI
 #define ide_init_default_irq(base)	(0)
 #else
diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
index feb7f0d7aca9..41bd33861ebe 100644
--- a/include/asm-m32r/ide.h
+++ b/include/asm-m32r/ide.h
@@ -63,8 +63,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	}
 }
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #ifdef CONFIG_BLK_DEV_IDEPCI
 #define ide_init_default_irq(base)     (0)
 #else
diff --git a/include/asm-mips/mach-generic/ide.h b/include/asm-mips/mach-generic/ide.h
index 45e24474cf43..7a9093187fe9 100644
--- a/include/asm-mips/mach-generic/ide.h
+++ b/include/asm-mips/mach-generic/ide.h
@@ -96,8 +96,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	}
 }
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #ifdef CONFIG_BLK_DEV_IDEPCI
 #define ide_init_default_irq(base)	(0)
 #else
diff --git a/include/asm-parisc/ide.h b/include/asm-parisc/ide.h
index be8760fbc8ee..db0c94410095 100644
--- a/include/asm-parisc/ide.h
+++ b/include/asm-parisc/ide.h
@@ -17,8 +17,6 @@
 #define MAX_HWIFS	2
 #endif
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #define ide_request_irq(irq,hand,flg,dev,id)	request_irq((irq),(hand),(flg),(dev),(id))
 #define ide_free_irq(irq,dev_id)		free_irq((irq), (dev_id))
 #define ide_request_region(from,extent,name)	request_region((from), (extent), (name))
diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index fef2ef1dbe86..17c51efff67c 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -86,8 +86,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 
 #endif /* __powerpc64__ */
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #endif /* __KERNEL__ */
 
 #endif /* _ASM_POWERPC_IDE_H */
diff --git a/include/asm-sh/ide.h b/include/asm-sh/ide.h
index 9f8e9142dc33..58e0bdd52be4 100644
--- a/include/asm-sh/ide.h
+++ b/include/asm-sh/ide.h
@@ -14,9 +14,6 @@
 
 #ifdef __KERNEL__
 
-
-#define ide_default_io_ctl(base)	(0)
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-sparc/ide.h b/include/asm-sparc/ide.h
index 4076cb5d1581..afd1736ed480 100644
--- a/include/asm-sparc/ide.h
+++ b/include/asm-sparc/ide.h
@@ -17,8 +17,6 @@
 #undef  MAX_HWIFS
 #define MAX_HWIFS	2
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #define __ide_insl(data_reg, buffer, wcount) \
 	__ide_insw(data_reg, buffer, (wcount)<<1)
 #define __ide_outsl(data_reg, buffer, wcount) \
diff --git a/include/asm-sparc64/ide.h b/include/asm-sparc64/ide.h
index ac7eb210b941..c5fdabe0b42d 100644
--- a/include/asm-sparc64/ide.h
+++ b/include/asm-sparc64/ide.h
@@ -24,8 +24,6 @@
 # endif
 #endif
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #define __ide_insl(data_reg, buffer, wcount) \
 	__ide_insw(data_reg, buffer, (wcount)<<1)
 #define __ide_outsl(data_reg, buffer, wcount) \
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
index 58080a7111de..800edccd33f1 100644
--- a/include/asm-x86/ide.h
+++ b/include/asm-x86/ide.h
@@ -58,8 +58,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	}
 }
 
-#define ide_default_io_ctl(base)	((base) + 0x206) /* obsolete */
-
 #ifdef CONFIG_BLK_DEV_IDEPCI
 #define ide_init_default_irq(base)	(0)
 #else
-- 
cgit v1.2.3


From 273b8385e5817a4765f82257004c5ec661a6a5b2 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 18 Apr 2008 00:46:35 +0200
Subject: ide: remove ide_init_default_irq() macro

* Use ide_default_irq() instead of ide_init_default_irq() in
  ide_generic host driver (so the correct IRQ is always set
  regardless of CONFIG_PCI / CONFIG_BLK_DEV_IDEPCI).

* Remove no longer needed ide_init_default_irq() macro.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-generic.c           | 2 +-
 include/asm-alpha/ide.h             | 6 ------
 include/asm-ia64/ide.h              | 6 ------
 include/asm-m32r/ide.h              | 6 ------
 include/asm-mips/mach-generic/ide.h | 6 ------
 include/asm-powerpc/ide.h           | 6 ------
 include/asm-x86/ide.h               | 6 ------
 7 files changed, 1 insertion(+), 37 deletions(-)

(limited to 'include')

diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index 4f996d01587a..25fda0a3263f 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -97,7 +97,7 @@ static int __init ide_generic_init(void)
 		if (hwif->chipset == ide_unknown && io_addr) {
 			memset(&hw, 0, sizeof(hw));
 			ide_std_init_ports(&hw, io_addr, io_addr + 0x206);
-			hw.irq = ide_init_default_irq(io_addr);
+			hw.irq = ide_default_irq(io_addr);
 			ide_init_port_hw(hwif, &hw);
 
 			idx[i] = i;
diff --git a/include/asm-alpha/ide.h b/include/asm-alpha/ide.h
index e67c38c96a6b..f44129abc02c 100644
--- a/include/asm-alpha/ide.h
+++ b/include/asm-alpha/ide.h
@@ -37,12 +37,6 @@ static inline unsigned long ide_default_io_base(int index)
 	}
 }
 
-#ifdef CONFIG_PCI
-#define ide_init_default_irq(base)	(0)
-#else
-#define ide_init_default_irq(base)	ide_default_irq(base)
-#endif
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-ia64/ide.h b/include/asm-ia64/ide.h
index 989edb6f1285..8fa3f8cd067a 100644
--- a/include/asm-ia64/ide.h
+++ b/include/asm-ia64/ide.h
@@ -44,12 +44,6 @@ static inline unsigned long ide_default_io_base(int index)
 	}
 }
 
-#ifdef CONFIG_PCI
-#define ide_init_default_irq(base)	(0)
-#else
-#define ide_init_default_irq(base)	ide_default_irq(base)
-#endif
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
index 41bd33861ebe..1e7f6474d130 100644
--- a/include/asm-m32r/ide.h
+++ b/include/asm-m32r/ide.h
@@ -63,12 +63,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	}
 }
 
-#ifdef CONFIG_BLK_DEV_IDEPCI
-#define ide_init_default_irq(base)     (0)
-#else
-#define ide_init_default_irq(base)     ide_default_irq(base)
-#endif
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-mips/mach-generic/ide.h b/include/asm-mips/mach-generic/ide.h
index 7a9093187fe9..0f6c251f5fec 100644
--- a/include/asm-mips/mach-generic/ide.h
+++ b/include/asm-mips/mach-generic/ide.h
@@ -96,12 +96,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	}
 }
 
-#ifdef CONFIG_BLK_DEV_IDEPCI
-#define ide_init_default_irq(base)	(0)
-#else
-#define ide_init_default_irq(base)	ide_default_irq(base)
-#endif
-
 /* MIPS port and memory-mapped I/O string operations.  */
 static inline void __ide_flush_prologue(void)
 {
diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index 17c51efff67c..3d90bf7d3d73 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -73,12 +73,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	return 0;
 }
 
-#ifdef CONFIG_PCI
-#define ide_init_default_irq(base)	(0)
-#else
-#define ide_init_default_irq(base)	ide_default_irq(base)
-#endif
-
 #ifdef CONFIG_BLK_DEV_MPC8xx_IDE
 #define IDE_ARCH_ACK_INTR  1
 #define ide_ack_intr(hwif) ((hwif)->ack_intr ? (hwif)->ack_intr(hwif) : 1)
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
index 800edccd33f1..cf9c98e5bdb5 100644
--- a/include/asm-x86/ide.h
+++ b/include/asm-x86/ide.h
@@ -58,12 +58,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	}
 }
 
-#ifdef CONFIG_BLK_DEV_IDEPCI
-#define ide_init_default_irq(base)	(0)
-#else
-#define ide_init_default_irq(base)	ide_default_irq(base)
-#endif
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From e56d8b8a2ee5fb7f63ceba58e1c0fb3c844888a4 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 17 Apr 2008 23:17:34 -0700
Subject: [INET]: Drop the inet_inherit_port() call.

As I can see from the code, two places (tcp_v6_syn_recv_sock and
dccp_v6_request_recv_sock) that call this one already run with
BHs disabled, so it's safe to call __inet_inherit_port there.

Besides (in case I missed smth with code review) the calltrace
tcp_v6_syn_recv_sock
 `- tcp_v4_syn_recv_sock
     `- __inet_inherit_port
and the similar for DCCP are valid, but assumes BHs to be disabled.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 7 -------
 net/dccp/ipv6.c               | 2 +-
 net/ipv6/tcp_ipv6.c           | 2 +-
 3 files changed, 2 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 5525227c5e92..5ec91d88b517 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -235,13 +235,6 @@ static inline void __inet_inherit_port(struct sock *sk, struct sock *child)
 	spin_unlock(&head->lock);
 }
 
-static inline void inet_inherit_port(struct sock *sk, struct sock *child)
-{
-	local_bh_disable();
-	__inet_inherit_port(sk, child);
-	local_bh_enable();
-}
-
 extern void inet_put_port(struct sock *sk);
 
 extern void inet_listen_wlock(struct inet_hashinfo *hashinfo);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index f8371c7e5e80..9b1129bb7ece 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -625,7 +625,7 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk,
 	newinet->daddr = newinet->saddr = newinet->rcv_saddr = LOOPBACK4_IPV6;
 
 	__inet6_hash(newsk);
-	inet_inherit_port(sk, newsk);
+	__inet_inherit_port(sk, newsk);
 
 	return newsk;
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 231c4dddfb8c..715965f0fac0 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1534,7 +1534,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 #endif
 
 	__inet6_hash(newsk);
-	inet_inherit_port(sk, newsk);
+	__inet_inherit_port(sk, newsk);
 
 	return newsk;
 
-- 
cgit v1.2.3


From 53083773dcbd3c80477e2ace143e361e1e806745 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Thu, 17 Apr 2008 23:18:15 -0700
Subject: [INET]: Uninline the __inet_inherit_port call.

This deblats ~200 bytes when ipv6 and dccp are 'y'.

Besides, this will ease compilation issues for patches
I'm working on to make inet hash tables more scalable
wrt net namespaces.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 14 +-------------
 net/ipv4/inet_hashtables.c    | 16 ++++++++++++++++
 2 files changed, 17 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 5ec91d88b517..735b926a3497 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -221,19 +221,7 @@ static inline int inet_sk_listen_hashfn(const struct sock *sk)
 }
 
 /* Caller must disable local BH processing. */
-static inline void __inet_inherit_port(struct sock *sk, struct sock *child)
-{
-	struct inet_hashinfo *table = sk->sk_prot->h.hashinfo;
-	const int bhash = inet_bhashfn(inet_sk(child)->num, table->bhash_size);
-	struct inet_bind_hashbucket *head = &table->bhash[bhash];
-	struct inet_bind_bucket *tb;
-
-	spin_lock(&head->lock);
-	tb = inet_csk(sk)->icsk_bind_hash;
-	sk_add_bind_node(child, &tb->owners);
-	inet_csk(child)->icsk_bind_hash = tb;
-	spin_unlock(&head->lock);
-}
+extern void __inet_inherit_port(struct sock *sk, struct sock *child);
 
 extern void inet_put_port(struct sock *sk);
 
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 1612184ecc6f..2023d37b2708 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -92,6 +92,22 @@ void inet_put_port(struct sock *sk)
 
 EXPORT_SYMBOL(inet_put_port);
 
+void __inet_inherit_port(struct sock *sk, struct sock *child)
+{
+	struct inet_hashinfo *table = sk->sk_prot->h.hashinfo;
+	const int bhash = inet_bhashfn(inet_sk(child)->num, table->bhash_size);
+	struct inet_bind_hashbucket *head = &table->bhash[bhash];
+	struct inet_bind_bucket *tb;
+
+	spin_lock(&head->lock);
+	tb = inet_csk(sk)->icsk_bind_hash;
+	sk_add_bind_node(child, &tb->owners);
+	inet_csk(child)->icsk_bind_hash = tb;
+	spin_unlock(&head->lock);
+}
+
+EXPORT_SYMBOL_GPL(__inet_inherit_port);
+
 /*
  * This lock without WQ_FLAG_EXCLUSIVE is good on UP and it can be very bad on SMP.
  * Look, when several writers sleep and reader wakes them up, all but one
-- 
cgit v1.2.3


From 38d1c069db8c87eb6cb10ca1ede9d9b673531ddd Mon Sep 17 00:00:00 2001
From: Boaz Harrosh <bharrosh@panasas.com>
Date: Fri, 18 Apr 2008 10:11:51 -0500
Subject: [SCSI] iscsi: extended cdb support

Support for extended CDBs in iscsi.
All we need is to check if command spills over 16 bytes then allocate
an iscsi-extended-header for the leftovers.

Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Reviewed-by: Pete Wyckoff <pw@osc.edu>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libiscsi.c    | 55 +++++++++++++++++++++++++++++++++++++++++-----
 include/scsi/iscsi_proto.h |  6 +++--
 2 files changed, 54 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index bdd7de7da39a..2f6b0955ff01 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -137,6 +137,45 @@ static int iscsi_add_hdr(struct iscsi_cmd_task *ctask, unsigned len)
 	return 0;
 }
 
+/*
+ * make an extended cdb AHS
+ */
+static int iscsi_prep_ecdb_ahs(struct iscsi_cmd_task *ctask)
+{
+	struct scsi_cmnd *cmd = ctask->sc;
+	unsigned rlen, pad_len;
+	unsigned short ahslength;
+	struct iscsi_ecdb_ahdr *ecdb_ahdr;
+	int rc;
+
+	ecdb_ahdr = iscsi_next_hdr(ctask);
+	rlen = cmd->cmd_len - ISCSI_CDB_SIZE;
+
+	BUG_ON(rlen > sizeof(ecdb_ahdr->ecdb));
+	ahslength = rlen + sizeof(ecdb_ahdr->reserved);
+
+	pad_len = iscsi_padding(rlen);
+
+	rc = iscsi_add_hdr(ctask, sizeof(ecdb_ahdr->ahslength) +
+	                   sizeof(ecdb_ahdr->ahstype) + ahslength + pad_len);
+	if (rc)
+		return rc;
+
+	if (pad_len)
+		memset(&ecdb_ahdr->ecdb[rlen], 0, pad_len);
+
+	ecdb_ahdr->ahslength = cpu_to_be16(ahslength);
+	ecdb_ahdr->ahstype = ISCSI_AHSTYPE_CDB;
+	ecdb_ahdr->reserved = 0;
+	memcpy(ecdb_ahdr->ecdb, cmd->cmnd + ISCSI_CDB_SIZE, rlen);
+
+	debug_scsi("iscsi_prep_ecdb_ahs: varlen_cdb_len %d "
+		   "rlen %d pad_len %d ahs_length %d iscsi_headers_size %u\n",
+		   cmd->cmd_len, rlen, pad_len, ahslength, ctask->hdr_len);
+
+	return 0;
+}
+
 /**
  * iscsi_prep_scsi_cmd_pdu - prep iscsi scsi cmd pdu
  * @ctask: iscsi cmd task
@@ -150,7 +189,7 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 	struct iscsi_session *session = conn->session;
 	struct iscsi_cmd *hdr = ctask->hdr;
 	struct scsi_cmnd *sc = ctask->sc;
-	unsigned hdrlength;
+	unsigned hdrlength, cmd_len;
 	int rc;
 
 	ctask->hdr_len = 0;
@@ -165,10 +204,16 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 	hdr->cmdsn = cpu_to_be32(session->cmdsn);
 	session->cmdsn++;
 	hdr->exp_statsn = cpu_to_be32(conn->exp_statsn);
-	memcpy(hdr->cdb, sc->cmnd, sc->cmd_len);
-	if (sc->cmd_len < MAX_COMMAND_SIZE)
-		memset(&hdr->cdb[sc->cmd_len], 0,
-			MAX_COMMAND_SIZE - sc->cmd_len);
+	cmd_len = sc->cmd_len;
+	if (cmd_len < ISCSI_CDB_SIZE)
+		memset(&hdr->cdb[cmd_len], 0, ISCSI_CDB_SIZE - cmd_len);
+	else if (cmd_len > ISCSI_CDB_SIZE) {
+		rc = iscsi_prep_ecdb_ahs(ctask);
+		if (rc)
+			return rc;
+		cmd_len = ISCSI_CDB_SIZE;
+	}
+	memcpy(hdr->cdb, sc->cmnd, cmd_len);
 
 	ctask->imm_count = 0;
 	if (sc->sc_data_direction == DMA_TO_DEVICE) {
diff --git a/include/scsi/iscsi_proto.h b/include/scsi/iscsi_proto.h
index 5ffec8ad6964..e0593bfae622 100644
--- a/include/scsi/iscsi_proto.h
+++ b/include/scsi/iscsi_proto.h
@@ -112,6 +112,7 @@ struct iscsi_ahs_hdr {
 
 #define ISCSI_AHSTYPE_CDB		1
 #define ISCSI_AHSTYPE_RLENGTH		2
+#define ISCSI_CDB_SIZE			16
 
 /* iSCSI PDU Header */
 struct iscsi_cmd {
@@ -125,7 +126,7 @@ struct iscsi_cmd {
 	__be32 data_length;
 	__be32 cmdsn;
 	__be32 exp_statsn;
-	uint8_t cdb[16];	/* SCSI Command Block */
+	uint8_t cdb[ISCSI_CDB_SIZE];	/* SCSI Command Block */
 	/* Additional Data (Command Dependent) */
 };
 
@@ -154,7 +155,8 @@ struct iscsi_ecdb_ahdr {
 	__be16 ahslength;	/* CDB length - 15, including reserved byte */
 	uint8_t ahstype;
 	uint8_t reserved;
-	uint8_t ecdb[260 - 16];	/* 4-byte aligned extended CDB spillover */
+	/* 4-byte aligned extended CDB spillover */
+	uint8_t ecdb[260 - ISCSI_CDB_SIZE];
 };
 
 /* SCSI Response Header */
-- 
cgit v1.2.3


From 8a076191f373abaeb4aa5f6755d22e49db98940f Mon Sep 17 00:00:00 2001
From: "Ahmed S. Darwish" <darwish.07@gmail.com>
Date: Sat, 1 Mar 2008 21:51:09 +0200
Subject: LSM: Introduce inode_getsecid and ipc_getsecid hooks

Introduce inode_getsecid(inode, secid) and ipc_getsecid(ipcp, secid)
LSM hooks. These hooks will be used instead of similar exported
SELinux interfaces.

Let {inode,ipc,task}_getsecid hooks set the secid to 0 by default
if CONFIG_SECURITY is not defined or if the hook is set to
NULL (dummy). This is done to notify the caller that no valid
secid exists.

Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Acked-by: James Morris <jmorris@namei.org>
Reviewed-by: Paul Moore <paul.moore@hp.com>
---
 include/linux/security.h | 30 +++++++++++++++++++++++++++++-
 security/dummy.c         | 16 +++++++++++++++-
 security/security.c      | 10 ++++++++++
 3 files changed, 54 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index c673dfd4dffc..45717d9d9656 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -468,6 +468,11 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@dentry is the dentry being changed.
  *	Return 0 on success.  If error is returned, then the operation
  *	causing setuid bit removal is failed.
+ * @inode_getsecid:
+ *	Get the secid associated with the node.
+ *	@inode contains a pointer to the inode.
+ *	@secid contains a pointer to the location where result will be saved.
+ *	In case of failure, @secid will be set to zero.
  *
  * Security hooks for file operations
  *
@@ -636,6 +641,8 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  * @task_getsecid:
  *	Retrieve the security identifier of the process @p.
  *	@p contains the task_struct for the process and place is into @secid.
+ *	In case of failure, @secid will be set to zero.
+ *
  * @task_setgroups:
  *	Check permission before setting the supplementary group set of the
  *	current process.
@@ -997,6 +1004,11 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@ipcp contains the kernel IPC permission structure
  *	@flag contains the desired (requested) permission set
  *	Return 0 if permission is granted.
+ * @ipc_getsecid:
+ *	Get the secid associated with the ipc object.
+ *	@ipcp contains the kernel IPC permission structure.
+ *	@secid contains a pointer to the location where result will be saved.
+ *	In case of failure, @secid will be set to zero.
  *
  * Security hooks for individual messages held in System V IPC message queues
  * @msg_msg_alloc_security:
@@ -1317,6 +1329,7 @@ struct security_operations {
 	int (*inode_getsecurity)(const struct inode *inode, const char *name, void **buffer, bool alloc);
   	int (*inode_setsecurity)(struct inode *inode, const char *name, const void *value, size_t size, int flags);
   	int (*inode_listsecurity)(struct inode *inode, char *buffer, size_t buffer_size);
+	void (*inode_getsecid)(const struct inode *inode, u32 *secid);
 
 	int (*file_permission) (struct file * file, int mask);
 	int (*file_alloc_security) (struct file * file);
@@ -1369,6 +1382,7 @@ struct security_operations {
 	void (*task_to_inode)(struct task_struct *p, struct inode *inode);
 
 	int (*ipc_permission) (struct kern_ipc_perm * ipcp, short flag);
+	void (*ipc_getsecid) (struct kern_ipc_perm *ipcp, u32 *secid);
 
 	int (*msg_msg_alloc_security) (struct msg_msg * msg);
 	void (*msg_msg_free_security) (struct msg_msg * msg);
@@ -1578,6 +1592,7 @@ int security_inode_killpriv(struct dentry *dentry);
 int security_inode_getsecurity(const struct inode *inode, const char *name, void **buffer, bool alloc);
 int security_inode_setsecurity(struct inode *inode, const char *name, const void *value, size_t size, int flags);
 int security_inode_listsecurity(struct inode *inode, char *buffer, size_t buffer_size);
+void security_inode_getsecid(const struct inode *inode, u32 *secid);
 int security_file_permission(struct file *file, int mask);
 int security_file_alloc(struct file *file);
 void security_file_free(struct file *file);
@@ -1622,6 +1637,7 @@ int security_task_prctl(int option, unsigned long arg2, unsigned long arg3,
 void security_task_reparent_to_init(struct task_struct *p);
 void security_task_to_inode(struct task_struct *p, struct inode *inode);
 int security_ipc_permission(struct kern_ipc_perm *ipcp, short flag);
+void security_ipc_getsecid(struct kern_ipc_perm *ipcp, u32 *secid);
 int security_msg_msg_alloc(struct msg_msg *msg);
 void security_msg_msg_free(struct msg_msg *msg);
 int security_msg_queue_alloc(struct msg_queue *msq);
@@ -2022,6 +2038,11 @@ static inline int security_inode_listsecurity(struct inode *inode, char *buffer,
 	return 0;
 }
 
+static inline void security_inode_getsecid(const struct inode *inode, u32 *secid)
+{
+	*secid = 0;
+}
+
 static inline int security_file_permission (struct file *file, int mask)
 {
 	return 0;
@@ -2137,7 +2158,9 @@ static inline int security_task_getsid (struct task_struct *p)
 }
 
 static inline void security_task_getsecid (struct task_struct *p, u32 *secid)
-{ }
+{
+	*secid = 0;
+}
 
 static inline int security_task_setgroups (struct group_info *group_info)
 {
@@ -2216,6 +2239,11 @@ static inline int security_ipc_permission (struct kern_ipc_perm *ipcp,
 	return 0;
 }
 
+static inline void security_ipc_getsecid(struct kern_ipc_perm *ipcp, u32 *secid)
+{
+	*secid = 0;
+}
+
 static inline int security_msg_msg_alloc (struct msg_msg * msg)
 {
 	return 0;
diff --git a/security/dummy.c b/security/dummy.c
index 78d8f92310a4..fb2e942efbb6 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -424,6 +424,11 @@ static int dummy_inode_listsecurity(struct inode *inode, char *buffer, size_t bu
 	return 0;
 }
 
+static void dummy_inode_getsecid(const struct inode *inode, u32 *secid)
+{
+	*secid = 0;
+}
+
 static int dummy_file_permission (struct file *file, int mask)
 {
 	return 0;
@@ -542,7 +547,9 @@ static int dummy_task_getsid (struct task_struct *p)
 }
 
 static void dummy_task_getsecid (struct task_struct *p, u32 *secid)
-{ }
+{
+	*secid = 0;
+}
 
 static int dummy_task_setgroups (struct group_info *group_info)
 {
@@ -616,6 +623,11 @@ static int dummy_ipc_permission (struct kern_ipc_perm *ipcp, short flag)
 	return 0;
 }
 
+static void dummy_ipc_getsecid(struct kern_ipc_perm *ipcp, u32 *secid)
+{
+	*secid = 0;
+}
+
 static int dummy_msg_msg_alloc_security (struct msg_msg *msg)
 {
 	return 0;
@@ -1058,6 +1070,7 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, inode_getsecurity);
 	set_to_dummy_if_null(ops, inode_setsecurity);
 	set_to_dummy_if_null(ops, inode_listsecurity);
+	set_to_dummy_if_null(ops, inode_getsecid);
 	set_to_dummy_if_null(ops, file_permission);
 	set_to_dummy_if_null(ops, file_alloc_security);
 	set_to_dummy_if_null(ops, file_free_security);
@@ -1094,6 +1107,7 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, task_reparent_to_init);
  	set_to_dummy_if_null(ops, task_to_inode);
 	set_to_dummy_if_null(ops, ipc_permission);
+	set_to_dummy_if_null(ops, ipc_getsecid);
 	set_to_dummy_if_null(ops, msg_msg_alloc_security);
 	set_to_dummy_if_null(ops, msg_msg_free_security);
 	set_to_dummy_if_null(ops, msg_queue_alloc_security);
diff --git a/security/security.c b/security/security.c
index 9beecac933b4..290482bdbbb0 100644
--- a/security/security.c
+++ b/security/security.c
@@ -523,6 +523,11 @@ int security_inode_listsecurity(struct inode *inode, char *buffer, size_t buffer
 	return security_ops->inode_listsecurity(inode, buffer, buffer_size);
 }
 
+void security_inode_getsecid(const struct inode *inode, u32 *secid)
+{
+	security_ops->inode_getsecid(inode, secid);
+}
+
 int security_file_permission(struct file *file, int mask)
 {
 	return security_ops->file_permission(file, mask);
@@ -712,6 +717,11 @@ int security_ipc_permission(struct kern_ipc_perm *ipcp, short flag)
 	return security_ops->ipc_permission(ipcp, flag);
 }
 
+void security_ipc_getsecid(struct kern_ipc_perm *ipcp, u32 *secid)
+{
+	security_ops->ipc_getsecid(ipcp, secid);
+}
+
 int security_msg_msg_alloc(struct msg_msg *msg)
 {
 	return security_ops->msg_msg_alloc_security(msg);
-- 
cgit v1.2.3


From 6b89a74be0fbbc6cc639d5cf7dcf8e6ee0f120a7 Mon Sep 17 00:00:00 2001
From: "Ahmed S. Darwish" <darwish.07@gmail.com>
Date: Sat, 1 Mar 2008 21:58:32 +0200
Subject: SELinux: remove redundant exports

Remove the following exported SELinux interfaces:
selinux_get_inode_sid(inode, sid)
selinux_get_ipc_sid(ipcp, sid)
selinux_get_task_sid(tsk, sid)
selinux_sid_to_string(sid, ctx, len)

They can be substitued with the following generic equivalents
respectively:
new LSM hook, inode_getsecid(inode, secid)
new LSM hook, ipc_getsecid*(ipcp, secid)
LSM hook, task_getsecid(tsk, secid)
LSM hook, sid_to_secctx(sid, ctx, len)

Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Acked-by: James Morris <jmorris@namei.org>
Reviewed-by: Paul Moore <paul.moore@hp.com>
---
 include/linux/selinux.h    | 62 ----------------------------------------------
 security/selinux/exports.c | 42 -------------------------------
 2 files changed, 104 deletions(-)

(limited to 'include')

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index 8c2cc4c02526..24b0af1c4cac 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -16,7 +16,6 @@
 
 struct selinux_audit_rule;
 struct audit_context;
-struct inode;
 struct kern_ipc_perm;
 
 #ifdef CONFIG_SECURITY_SELINUX
@@ -69,45 +68,6 @@ int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
  */
 void selinux_audit_set_callback(int (*callback)(void));
 
-/**
- *     selinux_sid_to_string - map a security context ID to a string
- *     @sid: security context ID to be converted.
- *     @ctx: address of context string to be returned
- *     @ctxlen: length of returned context string.
- *
- *     Returns 0 if successful, -errno if not.  On success, the context
- *     string will be allocated internally, and the caller must call
- *     kfree() on it after use.
- */
-int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen);
-
-/**
- *     selinux_get_inode_sid - get the inode's security context ID
- *     @inode: inode structure to get the sid from.
- *     @sid: pointer to security context ID to be filled in.
- *
- *     Returns nothing
- */
-void selinux_get_inode_sid(const struct inode *inode, u32 *sid);
-
-/**
- *     selinux_get_ipc_sid - get the ipc security context ID
- *     @ipcp: ipc structure to get the sid from.
- *     @sid: pointer to security context ID to be filled in.
- *
- *     Returns nothing
- */
-void selinux_get_ipc_sid(const struct kern_ipc_perm *ipcp, u32 *sid);
-
-/**
- *     selinux_get_task_sid - return the SID of task
- *     @tsk: the task whose SID will be returned
- *     @sid: pointer to security context ID to be filled in.
- *
- *     Returns nothing
- */
-void selinux_get_task_sid(struct task_struct *tsk, u32 *sid);
-
 /**
  *     selinux_string_to_sid - map a security context string to a security ID
  *     @str: the security context string to be mapped
@@ -175,28 +135,6 @@ static inline void selinux_audit_set_callback(int (*callback)(void))
 	return;
 }
 
-static inline int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen)
-{
-       *ctx = NULL;
-       *ctxlen = 0;
-       return 0;
-}
-
-static inline void selinux_get_inode_sid(const struct inode *inode, u32 *sid)
-{
-	*sid = 0;
-}
-
-static inline void selinux_get_ipc_sid(const struct kern_ipc_perm *ipcp, u32 *sid)
-{
-	*sid = 0;
-}
-
-static inline void selinux_get_task_sid(struct task_struct *tsk, u32 *sid)
-{
-	*sid = 0;
-}
-
 static inline int selinux_string_to_sid(const char *str, u32 *sid)
 {
        *sid = 0;
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index 87d2bb3ea355..64af2d3409ef 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -25,48 +25,6 @@
 /* SECMARK reference count */
 extern atomic_t selinux_secmark_refcount;
 
-int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen)
-{
-	if (selinux_enabled)
-		return security_sid_to_context(sid, ctx, ctxlen);
-	else {
-		*ctx = NULL;
-		*ctxlen = 0;
-	}
-
-	return 0;
-}
-
-void selinux_get_inode_sid(const struct inode *inode, u32 *sid)
-{
-	if (selinux_enabled) {
-		struct inode_security_struct *isec = inode->i_security;
-		*sid = isec->sid;
-		return;
-	}
-	*sid = 0;
-}
-
-void selinux_get_ipc_sid(const struct kern_ipc_perm *ipcp, u32 *sid)
-{
-	if (selinux_enabled) {
-		struct ipc_security_struct *isec = ipcp->security;
-		*sid = isec->sid;
-		return;
-	}
-	*sid = 0;
-}
-
-void selinux_get_task_sid(struct task_struct *tsk, u32 *sid)
-{
-	if (selinux_enabled) {
-		struct task_security_struct *tsec = tsk->security;
-		*sid = tsec->sid;
-		return;
-	}
-	*sid = 0;
-}
-
 int selinux_string_to_sid(char *str, u32 *sid)
 {
 	if (selinux_enabled)
-- 
cgit v1.2.3


From 03d37d25e0f91b28c4b6d002be6221f1af4b19d8 Mon Sep 17 00:00:00 2001
From: "Ahmed S. Darwish" <darwish.07@gmail.com>
Date: Sat, 1 Mar 2008 22:00:05 +0200
Subject: LSM/Audit: Introduce generic Audit LSM hooks

Introduce a generic Audit interface for security modules
by adding the following new LSM hooks:

audit_rule_init(field, op, rulestr, lsmrule)
audit_rule_known(krule)
audit_rule_match(secid, field, op, rule, actx)
audit_rule_free(rule)

Those hooks are only available if CONFIG_AUDIT is enabled.

Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Acked-by: James Morris <jmorris@namei.org>
Reviewed-by: Paul Moore <paul.moore@hp.com>
---
 include/linux/security.h | 72 ++++++++++++++++++++++++++++++++++++++++++++++++
 security/dummy.c         | 31 ++++++++++++++++++++-
 security/security.c      | 25 +++++++++++++++++
 3 files changed, 127 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index 45717d9d9656..697f228daf19 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -37,6 +37,7 @@
 extern unsigned securebits;
 
 struct ctl_table;
+struct audit_krule;
 
 /*
  * These functions are in security/capability.c and are used
@@ -1235,6 +1236,37 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@secdata contains the security context.
  *	@seclen contains the length of the security context.
  *
+ * Security hooks for Audit
+ *
+ * @audit_rule_init:
+ *	Allocate and initialize an LSM audit rule structure.
+ *	@field contains the required Audit action. Fields flags are defined in include/linux/audit.h
+ *	@op contains the operator the rule uses.
+ *	@rulestr contains the context where the rule will be applied to.
+ *	@lsmrule contains a pointer to receive the result.
+ *	Return 0 if @lsmrule has been successfully set,
+ *	-EINVAL in case of an invalid rule.
+ *
+ * @audit_rule_known:
+ *	Specifies whether given @rule contains any fields related to current LSM.
+ *	@rule contains the audit rule of interest.
+ *	Return 1 in case of relation found, 0 otherwise.
+ *
+ * @audit_rule_match:
+ *	Determine if given @secid matches a rule previously approved
+ *	by @audit_rule_known.
+ *	@secid contains the security id in question.
+ *	@field contains the field which relates to current LSM.
+ *	@op contains the operator that will be used for matching.
+ *	@rule points to the audit rule that will be checked against.
+ *	@actx points to the audit context associated with the check.
+ *	Return 1 if secid matches the rule, 0 if it does not, -ERRNO on failure.
+ *
+ * @audit_rule_free:
+ *	Deallocate the LSM audit rule structure previously allocated by
+ *	audit_rule_init.
+ *	@rule contains the allocated rule
+ *
  * This is the main security structure.
  */
 struct security_operations {
@@ -1494,6 +1526,13 @@ struct security_operations {
 
 #endif	/* CONFIG_KEYS */
 
+#ifdef CONFIG_AUDIT
+	int (*audit_rule_init)(u32 field, u32 op, char *rulestr, void **lsmrule);
+	int (*audit_rule_known)(struct audit_krule *krule);
+	int (*audit_rule_match)(u32 secid, u32 field, u32 op, void *lsmrule,
+				struct audit_context *actx);
+	void (*audit_rule_free)(void *lsmrule);
+#endif /* CONFIG_AUDIT */
 };
 
 /* prototypes */
@@ -2700,5 +2739,38 @@ static inline int security_key_permission(key_ref_t key_ref,
 #endif
 #endif /* CONFIG_KEYS */
 
+#ifdef CONFIG_AUDIT
+#ifdef CONFIG_SECURITY
+int security_audit_rule_init(u32 field, u32 op, char *rulestr, void **lsmrule);
+int security_audit_rule_known(struct audit_krule *krule);
+int security_audit_rule_match(u32 secid, u32 field, u32 op, void *lsmrule,
+			      struct audit_context *actx);
+void security_audit_rule_free(void *lsmrule);
+
+#else
+
+static inline int security_audit_rule_init(u32 field, u32 op, char *rulestr,
+					   void **lsmrule)
+{
+	return 0;
+}
+
+static inline int security_audit_rule_known(struct audit_krule *krule)
+{
+	return 0;
+}
+
+static inline int security_audit_rule_match(u32 secid, u32 field, u32 op,
+				   void *lsmrule, struct audit_context *actx)
+{
+	return 0;
+}
+
+static inline void security_audit_rule_free(void *lsmrule)
+{ }
+
+#endif /* CONFIG_SECURITY */
+#endif /* CONFIG_AUDIT */
+
 #endif /* ! __LINUX_SECURITY_H */
 
diff --git a/security/dummy.c b/security/dummy.c
index fb2e942efbb6..1ac9f8e66aa2 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -993,6 +993,30 @@ static inline int dummy_key_permission(key_ref_t key_ref,
 }
 #endif /* CONFIG_KEYS */
 
+#ifdef CONFIG_AUDIT
+static inline int dummy_audit_rule_init(u32 field, u32 op, char *rulestr,
+					void **lsmrule)
+{
+	return 0;
+}
+
+static inline int dummy_audit_rule_known(struct audit_krule *krule)
+{
+	return 0;
+}
+
+static inline int dummy_audit_rule_match(u32 secid, u32 field, u32 op,
+					 void *lsmrule,
+					 struct audit_context *actx)
+{
+	return 0;
+}
+
+static inline void dummy_audit_rule_free(void *lsmrule)
+{ }
+
+#endif /* CONFIG_AUDIT */
+
 struct security_operations dummy_security_ops;
 
 #define set_to_dummy_if_null(ops, function)				\
@@ -1182,6 +1206,11 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, key_free);
 	set_to_dummy_if_null(ops, key_permission);
 #endif	/* CONFIG_KEYS */
-
+#ifdef CONFIG_AUDIT
+	set_to_dummy_if_null(ops, audit_rule_init);
+	set_to_dummy_if_null(ops, audit_rule_known);
+	set_to_dummy_if_null(ops, audit_rule_match);
+	set_to_dummy_if_null(ops, audit_rule_free);
+#endif
 }
 
diff --git a/security/security.c b/security/security.c
index 290482bdbbb0..2ef593ec70f3 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1120,3 +1120,28 @@ int security_key_permission(key_ref_t key_ref,
 }
 
 #endif	/* CONFIG_KEYS */
+
+#ifdef CONFIG_AUDIT
+
+int security_audit_rule_init(u32 field, u32 op, char *rulestr, void **lsmrule)
+{
+	return security_ops->audit_rule_init(field, op, rulestr, lsmrule);
+}
+
+int security_audit_rule_known(struct audit_krule *krule)
+{
+	return security_ops->audit_rule_known(krule);
+}
+
+void security_audit_rule_free(void *lsmrule)
+{
+	security_ops->audit_rule_free(lsmrule);
+}
+
+int security_audit_rule_match(u32 secid, u32 field, u32 op, void *lsmrule,
+			      struct audit_context *actx)
+{
+	return security_ops->audit_rule_match(secid, field, op, lsmrule, actx);
+}
+
+#endif /* CONFIG_AUDIT */
-- 
cgit v1.2.3


From 9d57a7f9e23dc30783d245280fc9907cf2c87837 Mon Sep 17 00:00:00 2001
From: "Ahmed S. Darwish" <darwish.07@gmail.com>
Date: Sat, 1 Mar 2008 22:03:14 +0200
Subject: SELinux: use new audit hooks, remove redundant exports

Setup the new Audit LSM hooks for SELinux.
Remove the now redundant exported SELinux Audit interface.

Audit: Export 'audit_krule' and 'audit_field' to the public
since their internals are needed by the implementation of the
new LSM hook 'audit_rule_known'.

Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Acked-by: James Morris <jmorris@namei.org>
---
 include/linux/audit.h          | 29 +++++++++++++++++
 include/linux/selinux.h        | 72 ------------------------------------------
 kernel/audit.h                 | 25 ---------------
 security/selinux/hooks.c       |  8 +++++
 security/selinux/ss/services.c | 45 +++++++++++++++++++-------
 5 files changed, 71 insertions(+), 108 deletions(-)

(limited to 'include')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 2af9ec025015..04869c96016b 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -353,6 +353,33 @@ struct netlink_skb_parms;
 struct linux_binprm;
 struct mq_attr;
 struct mqstat;
+struct audit_watch;
+struct audit_tree;
+
+struct audit_krule {
+	int			vers_ops;
+	u32			flags;
+	u32			listnr;
+	u32			action;
+	u32			mask[AUDIT_BITMASK_SIZE];
+	u32			buflen; /* for data alloc on list rules */
+	u32			field_count;
+	char			*filterkey; /* ties events to rules */
+	struct audit_field	*fields;
+	struct audit_field	*arch_f; /* quick access to arch field */
+	struct audit_field	*inode_f; /* quick access to an inode field */
+	struct audit_watch	*watch;	/* associated watch */
+	struct audit_tree	*tree;	/* associated watched tree */
+	struct list_head	rlist;	/* entry in audit_{watch,tree}.rules list */
+};
+
+struct audit_field {
+	u32				type;
+	u32				val;
+	u32				op;
+	char				*se_str;
+	void				*se_rule;
+};
 
 #define AUDITSC_INVALID 0
 #define AUDITSC_SUCCESS 1
@@ -536,6 +563,8 @@ extern void		    audit_log_d_path(struct audit_buffer *ab,
 					     const char *prefix,
 					     struct path *path);
 extern void		    audit_log_lost(const char *message);
+extern int		    audit_update_lsm_rules(void);
+
 				/* Private API (for audit.c only) */
 extern int audit_filter_user(struct netlink_skb_parms *cb, int type);
 extern int audit_filter_type(int type);
diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index 24b0af1c4cac..20f965d4b041 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -20,54 +20,6 @@ struct kern_ipc_perm;
 
 #ifdef CONFIG_SECURITY_SELINUX
 
-/**
- *	selinux_audit_rule_init - alloc/init an selinux audit rule structure.
- *	@field: the field this rule refers to
- *	@op: the operater the rule uses
- *	@rulestr: the text "target" of the rule
- *	@rule: pointer to the new rule structure returned via this
- *
- *	Returns 0 if successful, -errno if not.  On success, the rule structure
- *	will be allocated internally.  The caller must free this structure with
- *	selinux_audit_rule_free() after use.
- */
-int selinux_audit_rule_init(u32 field, u32 op, char *rulestr,
-                            struct selinux_audit_rule **rule);
-
-/**
- *	selinux_audit_rule_free - free an selinux audit rule structure.
- *	@rule: pointer to the audit rule to be freed
- *
- *	This will free all memory associated with the given rule.
- *	If @rule is NULL, no operation is performed.
- */
-void selinux_audit_rule_free(struct selinux_audit_rule *rule);
-
-/**
- *	selinux_audit_rule_match - determine if a context ID matches a rule.
- *	@sid: the context ID to check
- *	@field: the field this rule refers to
- *	@op: the operater the rule uses
- *	@rule: pointer to the audit rule to check against
- *	@actx: the audit context (can be NULL) associated with the check
- *
- *	Returns 1 if the context id matches the rule, 0 if it does not, and
- *	-errno on failure.
- */
-int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
-                             struct selinux_audit_rule *rule,
-                             struct audit_context *actx);
-
-/**
- *	selinux_audit_set_callback - set the callback for policy reloads.
- *	@callback: the function to call when the policy is reloaded
- *
- *	This sets the function callback function that will update the rules
- *	upon policy reloads.  This callback should rebuild all existing rules
- *	using selinux_audit_rule_init().
- */
-void selinux_audit_set_callback(int (*callback)(void));
-
 /**
  *     selinux_string_to_sid - map a security context string to a security ID
  *     @str: the security context string to be mapped
@@ -111,30 +63,6 @@ void selinux_secmark_refcount_inc(void);
 void selinux_secmark_refcount_dec(void);
 #else
 
-static inline int selinux_audit_rule_init(u32 field, u32 op,
-                                          char *rulestr,
-                                          struct selinux_audit_rule **rule)
-{
-	return -EOPNOTSUPP;
-}
-
-static inline void selinux_audit_rule_free(struct selinux_audit_rule *rule)
-{
-	return;
-}
-
-static inline int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
-                                           struct selinux_audit_rule *rule,
-                                           struct audit_context *actx)
-{
-	return 0;
-}
-
-static inline void selinux_audit_set_callback(int (*callback)(void))
-{
-	return;
-}
-
 static inline int selinux_string_to_sid(const char *str, u32 *sid)
 {
        *sid = 0;
diff --git a/kernel/audit.h b/kernel/audit.h
index 2554bd524fd1..3cfc54ee3e1f 100644
--- a/kernel/audit.h
+++ b/kernel/audit.h
@@ -65,34 +65,9 @@ struct audit_watch {
 	struct list_head	rules;	/* associated rules */
 };
 
-struct audit_field {
-	u32				type;
-	u32				val;
-	u32				op;
-	char				*se_str;
-	struct selinux_audit_rule	*se_rule;
-};
-
 struct audit_tree;
 struct audit_chunk;
 
-struct audit_krule {
-	int			vers_ops;
-	u32			flags;
-	u32			listnr;
-	u32			action;
-	u32			mask[AUDIT_BITMASK_SIZE];
-	u32			buflen; /* for data alloc on list rules */
-	u32			field_count;
-	char			*filterkey; /* ties events to rules */
-	struct audit_field	*fields;
-	struct audit_field	*arch_f; /* quick access to arch field */
-	struct audit_field	*inode_f; /* quick access to an inode field */
-	struct audit_watch	*watch;	/* associated watch */
-	struct audit_tree	*tree;	/* associated watched tree */
-	struct list_head	rlist;	/* entry in audit_{watch,tree}.rules list */
-};
-
 struct audit_entry {
 	struct list_head	list;
 	struct rcu_head		rcu;
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index bfffaa52e0cb..a2f7e9cf78c5 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -83,6 +83,7 @@
 #include "netport.h"
 #include "xfrm.h"
 #include "netlabel.h"
+#include "audit.h"
 
 #define XATTR_SELINUX_SUFFIX "selinux"
 #define XATTR_NAME_SELINUX XATTR_SECURITY_PREFIX XATTR_SELINUX_SUFFIX
@@ -5478,6 +5479,13 @@ static struct security_operations selinux_ops = {
 	.key_free =                     selinux_key_free,
 	.key_permission =               selinux_key_permission,
 #endif
+
+#ifdef CONFIG_AUDIT
+	.audit_rule_init =		selinux_audit_rule_init,
+	.audit_rule_known =		selinux_audit_rule_known,
+	.audit_rule_match =		selinux_audit_rule_match,
+	.audit_rule_free =		selinux_audit_rule_free,
+#endif
 };
 
 static __init int selinux_init(void)
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index d75050819b06..1e0df5ec1bcd 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -57,6 +57,7 @@
 #include "netlabel.h"
 #include "xfrm.h"
 #include "ebitmap.h"
+#include "audit.h"
 
 extern void selnl_notify_policyload(u32 seqno);
 unsigned int policydb_loaded_version;
@@ -2296,21 +2297,23 @@ struct selinux_audit_rule {
 	struct context au_ctxt;
 };
 
-void selinux_audit_rule_free(struct selinux_audit_rule *rule)
+void selinux_audit_rule_free(void *vrule)
 {
+	struct selinux_audit_rule *rule = vrule;
+
 	if (rule) {
 		context_destroy(&rule->au_ctxt);
 		kfree(rule);
 	}
 }
 
-int selinux_audit_rule_init(u32 field, u32 op, char *rulestr,
-                            struct selinux_audit_rule **rule)
+int selinux_audit_rule_init(u32 field, u32 op, char *rulestr, void **vrule)
 {
 	struct selinux_audit_rule *tmprule;
 	struct role_datum *roledatum;
 	struct type_datum *typedatum;
 	struct user_datum *userdatum;
+	struct selinux_audit_rule **rule = (struct selinux_audit_rule **)vrule;
 	int rc = 0;
 
 	*rule = NULL;
@@ -2397,12 +2400,37 @@ int selinux_audit_rule_init(u32 field, u32 op, char *rulestr,
 	return rc;
 }
 
-int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
-                             struct selinux_audit_rule *rule,
+/* Check to see if the rule contains any selinux fields */
+int selinux_audit_rule_known(struct audit_krule *rule)
+{
+	int i;
+
+	for (i = 0; i < rule->field_count; i++) {
+		struct audit_field *f = &rule->fields[i];
+		switch (f->type) {
+		case AUDIT_SUBJ_USER:
+		case AUDIT_SUBJ_ROLE:
+		case AUDIT_SUBJ_TYPE:
+		case AUDIT_SUBJ_SEN:
+		case AUDIT_SUBJ_CLR:
+		case AUDIT_OBJ_USER:
+		case AUDIT_OBJ_ROLE:
+		case AUDIT_OBJ_TYPE:
+		case AUDIT_OBJ_LEV_LOW:
+		case AUDIT_OBJ_LEV_HIGH:
+			return 1;
+		}
+	}
+
+	return 0;
+}
+
+int selinux_audit_rule_match(u32 sid, u32 field, u32 op, void *vrule,
                              struct audit_context *actx)
 {
 	struct context *ctxt;
 	struct mls_level *level;
+	struct selinux_audit_rule *rule = vrule;
 	int match = 0;
 
 	if (!rule) {
@@ -2509,7 +2537,7 @@ out:
 	return match;
 }
 
-static int (*aurule_callback)(void) = NULL;
+static int (*aurule_callback)(void) = audit_update_lsm_rules;
 
 static int aurule_avc_callback(u32 event, u32 ssid, u32 tsid,
                                u16 class, u32 perms, u32 *retained)
@@ -2534,11 +2562,6 @@ static int __init aurule_init(void)
 }
 __initcall(aurule_init);
 
-void selinux_audit_set_callback(int (*callback)(void))
-{
-	aurule_callback = callback;
-}
-
 #ifdef CONFIG_NETLABEL
 /**
  * security_netlbl_cache_add - Add an entry to the NetLabel cache
-- 
cgit v1.2.3


From 04305e4aff8b0533dc05f9f6f1a34d0796bd985f Mon Sep 17 00:00:00 2001
From: "Ahmed S. Darwish" <darwish.07@gmail.com>
Date: Sat, 19 Apr 2008 09:59:43 +1000
Subject: Audit: Final renamings and cleanup

Rename the se_str and se_rule audit fields elements to
lsm_str and lsm_rule to avoid confusion.

Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Acked-by: James Morris <jmorris@namei.org>
---
 include/linux/audit.h            |  4 +--
 kernel/auditfilter.c             | 40 ++++++++++++-------------
 kernel/auditsc.c                 | 12 ++++----
 security/selinux/include/audit.h | 65 ++++++++++++++++++++++++++++++++++++++++
 4 files changed, 93 insertions(+), 28 deletions(-)
 create mode 100644 security/selinux/include/audit.h

(limited to 'include')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 04869c96016b..4ccb048cae1d 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -377,8 +377,8 @@ struct audit_field {
 	u32				type;
 	u32				val;
 	u32				op;
-	char				*se_str;
-	void				*se_rule;
+	char				*lsm_str;
+	void				*lsm_rule;
 };
 
 #define AUDITSC_INVALID 0
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 7c69cb5e44fb..28fef6bf8534 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -139,8 +139,8 @@ static inline void audit_free_rule(struct audit_entry *e)
 	if (e->rule.fields)
 		for (i = 0; i < e->rule.field_count; i++) {
 			struct audit_field *f = &e->rule.fields[i];
-			kfree(f->se_str);
-			security_audit_rule_free(f->se_rule);
+			kfree(f->lsm_str);
+			security_audit_rule_free(f->lsm_rule);
 		}
 	kfree(e->rule.fields);
 	kfree(e->rule.filterkey);
@@ -554,8 +554,8 @@ static struct audit_entry *audit_data_to_entry(struct audit_rule_data *data,
 		f->op = data->fieldflags[i] & AUDIT_OPERATORS;
 		f->type = data->fields[i];
 		f->val = data->values[i];
-		f->se_str = NULL;
-		f->se_rule = NULL;
+		f->lsm_str = NULL;
+		f->lsm_rule = NULL;
 		switch(f->type) {
 		case AUDIT_PID:
 		case AUDIT_UID:
@@ -598,7 +598,7 @@ static struct audit_entry *audit_data_to_entry(struct audit_rule_data *data,
 			entry->rule.buflen += f->val;
 
 			err = security_audit_rule_init(f->type, f->op, str,
-						       (void **)&f->se_rule);
+						       (void **)&f->lsm_rule);
 			/* Keep currently invalid fields around in case they
 			 * become valid after a policy reload. */
 			if (err == -EINVAL) {
@@ -610,7 +610,7 @@ static struct audit_entry *audit_data_to_entry(struct audit_rule_data *data,
 				kfree(str);
 				goto exit_free;
 			} else
-				f->se_str = str;
+				f->lsm_str = str;
 			break;
 		case AUDIT_WATCH:
 			str = audit_unpack_string(&bufp, &remain, f->val);
@@ -754,7 +754,7 @@ static struct audit_rule_data *audit_krule_to_data(struct audit_krule *krule)
 		case AUDIT_OBJ_LEV_LOW:
 		case AUDIT_OBJ_LEV_HIGH:
 			data->buflen += data->values[i] =
-				audit_pack_string(&bufp, f->se_str);
+				audit_pack_string(&bufp, f->lsm_str);
 			break;
 		case AUDIT_WATCH:
 			data->buflen += data->values[i] =
@@ -806,7 +806,7 @@ static int audit_compare_rule(struct audit_krule *a, struct audit_krule *b)
 		case AUDIT_OBJ_TYPE:
 		case AUDIT_OBJ_LEV_LOW:
 		case AUDIT_OBJ_LEV_HIGH:
-			if (strcmp(a->fields[i].se_str, b->fields[i].se_str))
+			if (strcmp(a->fields[i].lsm_str, b->fields[i].lsm_str))
 				return 1;
 			break;
 		case AUDIT_WATCH:
@@ -862,28 +862,28 @@ out:
 	return new;
 }
 
-/* Duplicate LSM field information.  The se_rule is opaque, so must be
+/* Duplicate LSM field information.  The lsm_rule is opaque, so must be
  * re-initialized. */
 static inline int audit_dupe_lsm_field(struct audit_field *df,
 					   struct audit_field *sf)
 {
 	int ret = 0;
-	char *se_str;
+	char *lsm_str;
 
-	/* our own copy of se_str */
-	se_str = kstrdup(sf->se_str, GFP_KERNEL);
-	if (unlikely(!se_str))
+	/* our own copy of lsm_str */
+	lsm_str = kstrdup(sf->lsm_str, GFP_KERNEL);
+	if (unlikely(!lsm_str))
 		return -ENOMEM;
-	df->se_str = se_str;
+	df->lsm_str = lsm_str;
 
-	/* our own (refreshed) copy of se_rule */
-	ret = security_audit_rule_init(df->type, df->op, df->se_str,
-				       (void **)&df->se_rule);
+	/* our own (refreshed) copy of lsm_rule */
+	ret = security_audit_rule_init(df->type, df->op, df->lsm_str,
+				       (void **)&df->lsm_rule);
 	/* Keep currently invalid fields around in case they
 	 * become valid after a policy reload. */
 	if (ret == -EINVAL) {
 		printk(KERN_WARNING "audit rule for LSM \'%s\' is "
-		       "invalid\n", df->se_str);
+		       "invalid\n", df->lsm_str);
 		ret = 0;
 	}
 
@@ -930,7 +930,7 @@ static struct audit_entry *audit_dupe_rule(struct audit_krule *old,
 	new->tree = old->tree;
 	memcpy(new->fields, old->fields, sizeof(struct audit_field) * fcount);
 
-	/* deep copy this information, updating the se_rule fields, because
+	/* deep copy this information, updating the lsm_rule fields, because
 	 * the originals will all be freed when the old rule is freed. */
 	for (i = 0; i < fcount; i++) {
 		switch (new->fields[i].type) {
@@ -1762,7 +1762,7 @@ unlock_and_return:
 	return result;
 }
 
-/* This function will re-initialize the se_rule field of all applicable rules.
+/* This function will re-initialize the lsm_rule field of all applicable rules.
  * It will traverse the filter lists serarching for rules that contain LSM
  * specific filter fields.  When such a rule is found, it is copied, the
  * LSM field is re-initialized, and the old rule is replaced with the
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index c0700535e5c5..56e56ed594a8 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -527,14 +527,14 @@ static int audit_filter_rules(struct task_struct *tsk,
 			   match for now to avoid losing information that
 			   may be wanted.   An error message will also be
 			   logged upon error */
-			if (f->se_rule) {
+			if (f->lsm_rule) {
 				if (need_sid) {
 					security_task_getsecid(tsk, &sid);
 					need_sid = 0;
 				}
 				result = security_audit_rule_match(sid, f->type,
 				                                  f->op,
-				                                  f->se_rule,
+				                                  f->lsm_rule,
 				                                  ctx);
 			}
 			break;
@@ -545,18 +545,18 @@ static int audit_filter_rules(struct task_struct *tsk,
 		case AUDIT_OBJ_LEV_HIGH:
 			/* The above note for AUDIT_SUBJ_USER...AUDIT_SUBJ_CLR
 			   also applies here */
-			if (f->se_rule) {
+			if (f->lsm_rule) {
 				/* Find files that match */
 				if (name) {
 					result = security_audit_rule_match(
 					           name->osid, f->type, f->op,
-					           f->se_rule, ctx);
+					           f->lsm_rule, ctx);
 				} else if (ctx) {
 					for (j = 0; j < ctx->name_count; j++) {
 						if (security_audit_rule_match(
 						      ctx->names[j].osid,
 						      f->type, f->op,
-						      f->se_rule, ctx)) {
+						      f->lsm_rule, ctx)) {
 							++result;
 							break;
 						}
@@ -569,7 +569,7 @@ static int audit_filter_rules(struct task_struct *tsk,
 					     aux = aux->next) {
 						if (aux->type == AUDIT_IPC) {
 							struct audit_aux_data_ipcctl *axi = (void *)aux;
-							if (security_audit_rule_match(axi->osid, f->type, f->op, f->se_rule, ctx)) {
+							if (security_audit_rule_match(axi->osid, f->type, f->op, f->lsm_rule, ctx)) {
 								++result;
 								break;
 							}
diff --git a/security/selinux/include/audit.h b/security/selinux/include/audit.h
new file mode 100644
index 000000000000..6c8b9ef15579
--- /dev/null
+++ b/security/selinux/include/audit.h
@@ -0,0 +1,65 @@
+/*
+ * SELinux support for the Audit LSM hooks
+ *
+ * Most of below header was moved from include/linux/selinux.h which 
+ * is released under below copyrights:
+ *
+ * Author: James Morris <jmorris@redhat.com>
+ *
+ * Copyright (C) 2005 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ * Copyright (C) 2006 Trusted Computer Solutions, Inc. <dgoeddel@trustedcs.com>
+ * Copyright (C) 2006 IBM Corporation, Timothy R. Chavez <tinytim@us.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2,
+ * as published by the Free Software Foundation.
+ */
+
+#ifndef _SELINUX_AUDIT_H
+#define _SELINUX_AUDIT_H
+
+/**
+ *	selinux_audit_rule_init - alloc/init an selinux audit rule structure.
+ *	@field: the field this rule refers to
+ *	@op: the operater the rule uses
+ *	@rulestr: the text "target" of the rule
+ *	@rule: pointer to the new rule structure returned via this
+ *
+ *	Returns 0 if successful, -errno if not.  On success, the rule structure
+ *	will be allocated internally.  The caller must free this structure with
+ *	selinux_audit_rule_free() after use.
+ */
+int selinux_audit_rule_init(u32 field, u32 op, char *rulestr, void **rule);
+
+/**
+ *	selinux_audit_rule_free - free an selinux audit rule structure.
+ *	@rule: pointer to the audit rule to be freed
+ *
+ *	This will free all memory associated with the given rule.
+ *	If @rule is NULL, no operation is performed.
+ */
+void selinux_audit_rule_free(void *rule);
+
+/**
+ *	selinux_audit_rule_match - determine if a context ID matches a rule.
+ *	@sid: the context ID to check
+ *	@field: the field this rule refers to
+ *	@op: the operater the rule uses
+ *	@rule: pointer to the audit rule to check against
+ *	@actx: the audit context (can be NULL) associated with the check
+ *
+ *	Returns 1 if the context id matches the rule, 0 if it does not, and
+ *	-errno on failure.
+ */
+int selinux_audit_rule_match(u32 sid, u32 field, u32 op, void *rule,
+                             struct audit_context *actx);
+
+/**
+ *	selinux_audit_rule_known - check to see if rule contains selinux fields.
+ *	@rule: rule to be checked
+ *	Returns 1 if there are selinux fields specified in the rule, 0 otherwise.
+ */
+int selinux_audit_rule_known(struct audit_krule *krule);
+
+#endif /* _SELINUX_AUDIT_H */
+
-- 
cgit v1.2.3


From 076c54c5bcaed2081c0cba94a6f77c4d470236ad Mon Sep 17 00:00:00 2001
From: "Ahmed S. Darwish" <darwish.07@gmail.com>
Date: Thu, 6 Mar 2008 18:09:10 +0200
Subject: Security: Introduce security= boot parameter

Add the security= boot parameter. This is done to avoid LSM
registration clashes in case of more than one bult-in module.

User can choose a security module to enable at boot. If no
security= boot parameter is specified, only the first LSM
asking for registration will be loaded. An invalid security
module name will be treated as if no module has been chosen.

LSM modules must check now if they are allowed to register
by calling security_module_enable(ops) first. Modify SELinux
and SMACK to do so.

Do not let SMACK register smackfs if it was not chosen on
boot. Smackfs assumes that smack hooks are registered and
the initial task security setup (swapper->security) is done.

Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Acked-by: James Morris <jmorris@namei.org>
---
 Documentation/kernel-parameters.txt |  6 ++++++
 include/linux/security.h            | 12 ++++++++++++
 security/dummy.c                    |  4 +++-
 security/security.c                 | 38 ++++++++++++++++++++++++++++++++++++-
 security/selinux/hooks.c            |  7 +++++++
 security/smack/smack.h              |  2 ++
 security/smack/smack_lsm.c          |  7 ++++++-
 security/smack/smackfs.c            | 11 ++++++++++-
 8 files changed, 83 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 256a2162503c..4b0f1ae31a4c 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -366,6 +366,12 @@ and is between 256 and 4096 characters. It is defined in the file
 			possible to determine what the correct size should be.
 			This option provides an override for these situations.
 
+	security=	[SECURITY] Choose a security module to enable at boot.
+			If this boot parameter is not specified, only the first
+			security module asking for security registration will be
+			loaded. An invalid security module name will be treated
+			as if no module has been chosen.
+
 	capability.disable=
 			[SECURITY] Disable capabilities.  This would normally
 			be used only if an alternative security model is to be
diff --git a/include/linux/security.h b/include/linux/security.h
index 697f228daf19..f4116d6ed64b 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -36,6 +36,9 @@
 
 extern unsigned securebits;
 
+/* Maximum number of letters for an LSM name string */
+#define SECURITY_NAME_MAX	10
+
 struct ctl_table;
 struct audit_krule;
 
@@ -137,6 +140,12 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
 /**
  * struct security_operations - main security structure
  *
+ * Security module identifier.
+ *
+ * @name:
+ *	A string that acts as a unique identifeir for the LSM with max number
+ *	of characters = SECURITY_NAME_MAX.
+ *
  * Security hooks for program execution operations.
  *
  * @bprm_alloc_security:
@@ -1270,6 +1279,8 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  * This is the main security structure.
  */
 struct security_operations {
+	char name[SECURITY_NAME_MAX + 1];
+
 	int (*ptrace) (struct task_struct * parent, struct task_struct * child);
 	int (*capget) (struct task_struct * target,
 		       kernel_cap_t * effective,
@@ -1537,6 +1548,7 @@ struct security_operations {
 
 /* prototypes */
 extern int security_init	(void);
+extern int security_module_enable(struct security_operations *ops);
 extern int register_security	(struct security_operations *ops);
 extern int mod_reg_security	(const char *name, struct security_operations *ops);
 extern struct dentry *securityfs_create_file(const char *name, mode_t mode,
diff --git a/security/dummy.c b/security/dummy.c
index 1ac9f8e66aa2..d797a4196b89 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -1017,7 +1017,9 @@ static inline void dummy_audit_rule_free(void *lsmrule)
 
 #endif /* CONFIG_AUDIT */
 
-struct security_operations dummy_security_ops;
+struct security_operations dummy_security_ops = {
+	.name = "dummy",
+};
 
 #define set_to_dummy_if_null(ops, function)				\
 	do {								\
diff --git a/security/security.c b/security/security.c
index 2ef593ec70f3..dd0c6baed494 100644
--- a/security/security.c
+++ b/security/security.c
@@ -17,6 +17,8 @@
 #include <linux/kernel.h>
 #include <linux/security.h>
 
+/* Boot-time LSM user choice */
+static __initdata char chosen_lsm[SECURITY_NAME_MAX + 1];
 
 /* things that live in dummy.c */
 extern struct security_operations dummy_security_ops;
@@ -67,13 +69,47 @@ int __init security_init(void)
 	return 0;
 }
 
+/* Save user chosen LSM */
+static int __init choose_lsm(char *str)
+{
+	strncpy(chosen_lsm, str, SECURITY_NAME_MAX);
+	return 1;
+}
+__setup("security=", choose_lsm);
+
+/**
+ * security_module_enable - Load given security module on boot ?
+ * @ops: a pointer to the struct security_operations that is to be checked.
+ *
+ * Each LSM must pass this method before registering its own operations
+ * to avoid security registration races. This method may also be used
+ * to check if your LSM is currently loaded.
+ *
+ * Return true if:
+ *	-The passed LSM is the one chosen by user at boot time,
+ *	-or user didsn't specify a specific LSM and we're the first to ask
+ *	 for registeration permissoin,
+ *	-or the passed LSM is currently loaded.
+ * Otherwise, return false.
+ */
+int __init security_module_enable(struct security_operations *ops)
+{
+	if (!*chosen_lsm)
+		strncpy(chosen_lsm, ops->name, SECURITY_NAME_MAX);
+	else if (strncmp(ops->name, chosen_lsm, SECURITY_NAME_MAX))
+		return 0;
+
+	return 1;
+}
+
 /**
  * register_security - registers a security framework with the kernel
  * @ops: a pointer to the struct security_options that is to be registered
  *
  * This function is to allow a security module to register itself with the
  * kernel security subsystem.  Some rudimentary checking is done on the @ops
- * value passed to this function.
+ * value passed to this function. You'll need to check first if your LSM
+ * is allowed to register its @ops by calling security_module_enable(@ops).
  *
  * If there is already a security module registered with the kernel,
  * an error will be returned.  Otherwise 0 is returned on success.
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index a2f7e9cf78c5..f9927f02bc3d 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -5295,6 +5295,8 @@ static int selinux_key_permission(key_ref_t key_ref,
 #endif
 
 static struct security_operations selinux_ops = {
+	.name =				"selinux",
+
 	.ptrace =			selinux_ptrace,
 	.capget =			selinux_capget,
 	.capset_check =			selinux_capset_check,
@@ -5492,6 +5494,11 @@ static __init int selinux_init(void)
 {
 	struct task_security_struct *tsec;
 
+	if (!security_module_enable(&selinux_ops)) {
+		selinux_enabled = 0;
+		return 0;
+	}
+
 	if (!selinux_enabled) {
 		printk(KERN_INFO "SELinux:  Disabled at boot.\n");
 		return 0;
diff --git a/security/smack/smack.h b/security/smack/smack.h
index 62c1e982849d..4a4477f5afdc 100644
--- a/security/smack/smack.h
+++ b/security/smack/smack.h
@@ -15,6 +15,7 @@
 
 #include <linux/capability.h>
 #include <linux/spinlock.h>
+#include <linux/security.h>
 #include <net/netlabel.h>
 
 /*
@@ -187,6 +188,7 @@ extern struct smack_known smack_known_star;
 extern struct smack_known smack_known_unset;
 
 extern struct smk_list_entry *smack_list;
+extern struct security_operations smack_ops;
 
 /*
  * Stricly for CIPSO level manipulation.
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 732ba27923c4..904bdc01a12b 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -2424,7 +2424,9 @@ static void smack_release_secctx(char *secdata, u32 seclen)
 {
 }
 
-static struct security_operations smack_ops = {
+struct security_operations smack_ops = {
+	.name =				"smack",
+
 	.ptrace = 			smack_ptrace,
 	.capget = 			cap_capget,
 	.capset_check = 		cap_capset_check,
@@ -2557,6 +2559,9 @@ static struct security_operations smack_ops = {
  */
 static __init int smack_init(void)
 {
+	if (!security_module_enable(&smack_ops))
+		return 0;
+
 	printk(KERN_INFO "Smack:  Initializing.\n");
 
 	/*
diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c
index cfae8afcc262..6ba283783b70 100644
--- a/security/smack/smackfs.c
+++ b/security/smack/smackfs.c
@@ -965,12 +965,21 @@ static struct vfsmount *smackfs_mount;
  *
  * register the smackfs
  *
- * Returns 0 unless the registration fails.
+ * Do not register smackfs if Smack wasn't enabled
+ * on boot. We can not put this method normally under the
+ * smack_init() code path since the security subsystem get
+ * initialized before the vfs caches.
+ *
+ * Returns true if we were not chosen on boot or if
+ * we were chosen and filesystem registration succeeded.
  */
 static int __init init_smk_fs(void)
 {
 	int err;
 
+	if (!security_module_enable(&smack_ops))
+		return 0;
+
 	err = register_filesystem(&smk_fs_type);
 	if (!err) {
 		smackfs_mount = kern_mount(&smk_fs_type);
-- 
cgit v1.2.3