Merge branch 'android-4.4-p' of https://android.googlesource.com/kernel/common into lineage-18.1-caf-msm8998

This brings LA.UM.9.2.r1-02700-SDMxx0.0 up to date with
https://android.googlesource.com/kernel/common/ android-4.4-p at commit:
f5978a07daf67 Merge 4.4.267 into android-4.4-p

Conflicts:
	arch/alpha/include/asm/Kbuild
	drivers/mmc/core/mmc.c
	drivers/usb/gadget/configfs.c

Change-Id: I978d923e97c18f284edbd32c0c19ac70002f7d83

43 files changed, 274 insertions, 436 deletions

diff --git a/arch/alpha/include/asm/Kbuild b/arch/alpha/include/asm/Kbuild
index 74252220cb46..f5cb9b28e5fb 100644
--- a/arch/alpha/include/asm/Kbuild
+++ b/arch/alpha/include/asm/Kbuild
@@ -3,6 +3,7 @@
 generic-y += clkdev.h
 generic-y += cputime.h
 generic-y += exec.h
+generic-y += export.h
 generic-y += hash.h
 generic-y += irq_work.h
 generic-y += mcs_spinlock.h
diff --git a/arch/alpha/include/asm/uaccess.h b/arch/alpha/include/asm/uaccess.h
index c0ddbbf73400..89413a29cb07 100644
--- a/arch/alpha/include/asm/uaccess.h
+++ b/arch/alpha/include/asm/uaccess.h
@@ -341,45 +341,17 @@ __asm__ __volatile__("1: stb %r2,%1\n"				\
  * Complex access routines
  */
 
-/* This little bit of silliness is to get the GP loaded for a function
-   that ordinarily wouldn't.  Otherwise we could have it done by the macro
-   directly, which can be optimized the linker.  */
-#ifdef MODULE
-#define __module_address(sym)		"r"(sym),
-#define __module_call(ra, arg, sym)	"jsr $" #ra ",(%" #arg ")," #sym
-#else
-#define __module_address(sym)
-#define __module_call(ra, arg, sym)	"bsr $" #ra "," #sym " !samegp"
-#endif
-
-extern void __copy_user(void);
-
-extern inline long
-__copy_tofrom_user_nocheck(void *to, const void *from, long len)
-{
-	register void * __cu_to __asm__("$6") = to;
-	register const void * __cu_from __asm__("$7") = from;
-	register long __cu_len __asm__("$0") = len;
-
-	__asm__ __volatile__(
-		__module_call(28, 3, __copy_user)
-		: "=r" (__cu_len), "=r" (__cu_from), "=r" (__cu_to)
-		: __module_address(__copy_user)
-		  "0" (__cu_len), "1" (__cu_from), "2" (__cu_to)
-		: "$1", "$2", "$3", "$4", "$5", "$28", "memory");
-
-	return __cu_len;
-}
+extern long __copy_user(void *to, const void *from, long len);
 
-#define __copy_to_user(to, from, n)					\
-({									\
-	__chk_user_ptr(to);						\
-	__copy_tofrom_user_nocheck((__force void *)(to), (from), (n));	\
+#define __copy_to_user(to, from, n)			\
+({							\
+	__chk_user_ptr(to);				\
+	__copy_user((__force void *)(to), (from), (n));	\
 })
-#define __copy_from_user(to, from, n)					\
-({									\
-	__chk_user_ptr(from);						\
-	__copy_tofrom_user_nocheck((to), (__force void *)(from), (n));	\
+#define __copy_from_user(to, from, n)			\
+({							\
+	__chk_user_ptr(from);				\
+	__copy_user((to), (__force void *)(from), (n));	\
 })
 
 #define __copy_to_user_inatomic __copy_to_user
@@ -389,35 +361,22 @@ extern inline long
 copy_to_user(void __user *to, const void *from, long n)
 {
 	if (likely(__access_ok((unsigned long)to, n, get_fs())))
-		n = __copy_tofrom_user_nocheck((__force void *)to, from, n);
+		n = __copy_user((__force void *)to, from, n);
 	return n;
 }
 
 extern inline long
 copy_from_user(void *to, const void __user *from, long n)
 {
+	long res = n;
 	if (likely(__access_ok((unsigned long)from, n, get_fs())))
-		n = __copy_tofrom_user_nocheck(to, (__force void *)from, n);
-	else
-		memset(to, 0, n);
-	return n;
+		res = __copy_from_user_inatomic(to, from, n);
+	if (unlikely(res))
+		memset(to + (n - res), 0, res);
+	return res;
 }
 
-extern void __do_clear_user(void);
-
-extern inline long
-__clear_user(void __user *to, long len)
-{
-	register void __user * __cl_to __asm__("$6") = to;
-	register long __cl_len __asm__("$0") = len;
-	__asm__ __volatile__(
-		__module_call(28, 2, __do_clear_user)
-		: "=r"(__cl_len), "=r"(__cl_to)
-		: __module_address(__do_clear_user)
-		  "0"(__cl_len), "1"(__cl_to)
-		: "$1", "$2", "$3", "$4", "$5", "$28", "memory");
-	return __cl_len;
-}
+extern long __clear_user(void __user *to, long len);
 
 extern inline long
 clear_user(void __user *to, long len)
@@ -427,9 +386,6 @@ clear_user(void __user *to, long len)
 	return len;
 }
 
-#undef __module_address
-#undef __module_call
-
 #define user_addr_max() \
         (segment_eq(get_fs(), USER_DS) ? TASK_SIZE : ~0UL)
 
diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index 3ecac0106c8a..8ce13d7a2ad3 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -8,7 +8,7 @@ ccflags-y	:= -Wno-sign-compare
 
 obj-y    := entry.o traps.o process.o osf_sys.o irq.o \
 	    irq_alpha.o signal.o setup.o ptrace.o time.o \
-	    alpha_ksyms.o systbls.o err_common.o io.o
+	    systbls.o err_common.o io.o
 
 obj-$(CONFIG_VGA_HOSE)	+= console.o
 obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/alpha/kernel/alpha_ksyms.c b/arch/alpha/kernel/alpha_ksyms.c
deleted file mode 100644
index f4c7ab6f43b0..000000000000
--- a/arch/alpha/kernel/alpha_ksyms.c
+++ /dev/null
@@ -1,102 +0,0 @@
-/*
- * linux/arch/alpha/kernel/alpha_ksyms.c
- *
- * Export the alpha-specific functions that are needed for loadable
- * modules.
- */
-
-#include <linux/module.h>
-#include <asm/console.h>
-#include <asm/uaccess.h>
-#include <asm/checksum.h>
-#include <asm/fpu.h>
-#include <asm/machvec.h>
-
-#include <linux/syscalls.h>
-
-/* these are C runtime functions with special calling conventions: */
-extern void __divl (void);
-extern void __reml (void);
-extern void __divq (void);
-extern void __remq (void);
-extern void __divlu (void);
-extern void __remlu (void);
-extern void __divqu (void);
-extern void __remqu (void);
-
-EXPORT_SYMBOL(alpha_mv);
-EXPORT_SYMBOL(callback_getenv);
-EXPORT_SYMBOL(callback_setenv);
-EXPORT_SYMBOL(callback_save_env);
-
-/* platform dependent support */
-EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strcpy);
-EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strncpy);
-EXPORT_SYMBOL(strncat);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strrchr);
-EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(__memcpy);
-EXPORT_SYMBOL(__memset);
-EXPORT_SYMBOL(___memset);
-EXPORT_SYMBOL(__memsetw);
-EXPORT_SYMBOL(__constant_c_memset);
-EXPORT_SYMBOL(copy_page);
-EXPORT_SYMBOL(clear_page);
-
-EXPORT_SYMBOL(alpha_read_fp_reg);
-EXPORT_SYMBOL(alpha_read_fp_reg_s);
-EXPORT_SYMBOL(alpha_write_fp_reg);
-EXPORT_SYMBOL(alpha_write_fp_reg_s);
-
-/* Networking helper routines. */
-EXPORT_SYMBOL(csum_tcpudp_magic);
-EXPORT_SYMBOL(ip_compute_csum);
-EXPORT_SYMBOL(ip_fast_csum);
-EXPORT_SYMBOL(csum_partial_copy_nocheck);
-EXPORT_SYMBOL(csum_partial_copy_from_user);
-EXPORT_SYMBOL(csum_ipv6_magic);
-
-#ifdef CONFIG_MATHEMU_MODULE
-extern long (*alpha_fp_emul_imprecise)(struct pt_regs *, unsigned long);
-extern long (*alpha_fp_emul) (unsigned long pc);
-EXPORT_SYMBOL(alpha_fp_emul_imprecise);
-EXPORT_SYMBOL(alpha_fp_emul);
-#endif
-
-/*
- * The following are specially called from the uaccess assembly stubs.
- */
-EXPORT_SYMBOL(__copy_user);
-EXPORT_SYMBOL(__do_clear_user);
-
-/* 
- * SMP-specific symbols.
- */
-
-#ifdef CONFIG_SMP
-EXPORT_SYMBOL(_atomic_dec_and_lock);
-#endif /* CONFIG_SMP */
-
-/*
- * The following are special because they're not called
- * explicitly (the C compiler or assembler generates them in
- * response to division operations).  Fortunately, their
- * interface isn't gonna change any time soon now, so it's OK
- * to leave it out of version control.
- */
-# undef memcpy
-# undef memset
-EXPORT_SYMBOL(__divl);
-EXPORT_SYMBOL(__divlu);
-EXPORT_SYMBOL(__divq);
-EXPORT_SYMBOL(__divqu);
-EXPORT_SYMBOL(__reml);
-EXPORT_SYMBOL(__remlu);
-EXPORT_SYMBOL(__remq);
-EXPORT_SYMBOL(__remqu);
-EXPORT_SYMBOL(memcpy);
-EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(memchr);
diff --git a/arch/alpha/kernel/machvec_impl.h b/arch/alpha/kernel/machvec_impl.h
index f54bdf658cd0..8c6516025efb 100644
--- a/arch/alpha/kernel/machvec_impl.h
+++ b/arch/alpha/kernel/machvec_impl.h
@@ -144,9 +144,11 @@
    else beforehand.  Fine.  We'll do it ourselves.  */
 #if 0
 #define ALIAS_MV(system) \
-  struct alpha_machine_vector alpha_mv __attribute__((alias(#system "_mv")));
+  struct alpha_machine_vector alpha_mv __attribute__((alias(#system "_mv"))); \
+  EXPORT_SYMBOL(alpha_mv);
 #else
 #define ALIAS_MV(system) \
-  asm(".global alpha_mv\nalpha_mv = " #system "_mv");
+  asm(".global alpha_mv\nalpha_mv = " #system "_mv"); \
+  EXPORT_SYMBOL(alpha_mv);
 #endif
 #endif /* GENERIC */
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index b20af76f12c1..4811e54069fc 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -115,6 +115,7 @@ unsigned long alpha_agpgart_size = DEFAULT_AGP_APER_SIZE;
 
 #ifdef CONFIG_ALPHA_GENERIC
 struct alpha_machine_vector alpha_mv;
+EXPORT_SYMBOL(alpha_mv);
 #endif
 
 #ifndef alpha_using_srm
diff --git a/arch/alpha/lib/Makefile b/arch/alpha/lib/Makefile
index 59660743237c..a80815960364 100644
--- a/arch/alpha/lib/Makefile
+++ b/arch/alpha/lib/Makefile
@@ -20,12 +20,8 @@ lib-y =	__divqu.o __remqu.o __divlu.o __remlu.o \
 	checksum.o \
 	csum_partial_copy.o \
 	$(ev67-y)strlen.o \
-	$(ev67-y)strcat.o \
-	strcpy.o \
-	$(ev67-y)strncat.o \
-	strncpy.o \
-	$(ev6-y)stxcpy.o \
-	$(ev6-y)stxncpy.o \
+	stycpy.o \
+	styncpy.o \
 	$(ev67-y)strchr.o \
 	$(ev67-y)strrchr.o \
 	$(ev6-y)memchr.o \
@@ -46,11 +42,20 @@ AFLAGS___remqu.o =       -DREM
 AFLAGS___divlu.o = -DDIV       -DINTSIZE
 AFLAGS___remlu.o =       -DREM -DINTSIZE
 
-$(obj)/__divqu.o: $(obj)/$(ev6-y)divide.S
-	$(cmd_as_o_S)
-$(obj)/__remqu.o: $(obj)/$(ev6-y)divide.S
-	$(cmd_as_o_S)
-$(obj)/__divlu.o: $(obj)/$(ev6-y)divide.S
-	$(cmd_as_o_S)
-$(obj)/__remlu.o: $(obj)/$(ev6-y)divide.S
-	$(cmd_as_o_S)
+$(addprefix $(obj)/,__divqu.o __remqu.o __divlu.o __remlu.o): \
+						$(src)/$(ev6-y)divide.S FORCE
+	$(call if_changed_rule,as_o_S)
+
+# There are direct branches between {str*cpy,str*cat} and stx*cpy.
+# Ensure the branches are within range by merging these objects.
+
+LDFLAGS_stycpy.o := -r
+LDFLAGS_styncpy.o := -r
+
+$(obj)/stycpy.o: $(obj)/strcpy.o $(obj)/$(ev67-y)strcat.o \
+		 $(obj)/$(ev6-y)stxcpy.o FORCE
+	$(call if_changed,ld)
+
+$(obj)/styncpy.o: $(obj)/strncpy.o $(obj)/$(ev67-y)strncat.o \
+		 $(obj)/$(ev6-y)stxncpy.o FORCE
+	$(call if_changed,ld)
diff --git a/arch/alpha/lib/callback_srm.S b/arch/alpha/lib/callback_srm.S
index 8804bec2c644..6093addc931a 100644
--- a/arch/alpha/lib/callback_srm.S
+++ b/arch/alpha/lib/callback_srm.S
@@ -3,6 +3,7 @@
  */
 
 #include <asm/console.h>
+#include <asm/export.h>
 
 .text
 #define HWRPB_CRB_OFFSET 0xc0
@@ -92,6 +93,10 @@ CALLBACK(reset_env, CCB_RESET_ENV, 4)
 CALLBACK(save_env, CCB_SAVE_ENV, 1)
 CALLBACK(pswitch, CCB_PSWITCH, 3)
 CALLBACK(bios_emul, CCB_BIOS_EMUL, 5)
+
+EXPORT_SYMBOL(callback_getenv)
+EXPORT_SYMBOL(callback_setenv)
+EXPORT_SYMBOL(callback_save_env)
 	
 .data
 __alpha_using_srm:		# For use by bootpheader
diff --git a/arch/alpha/lib/checksum.c b/arch/alpha/lib/checksum.c
index 199f6efa83fa..65197c3c0845 100644
--- a/arch/alpha/lib/checksum.c
+++ b/arch/alpha/lib/checksum.c
@@ -50,6 +50,7 @@ __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
 		(__force u64)saddr + (__force u64)daddr +
 		(__force u64)sum + ((len + proto) << 8));
 }
+EXPORT_SYMBOL(csum_tcpudp_magic);
 
 __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
 				   unsigned short len,
@@ -148,6 +149,7 @@ __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 {
 	return (__force __sum16)~do_csum(iph,ihl*4);
 }
+EXPORT_SYMBOL(ip_fast_csum);
 
 /*
  * computes the checksum of a memory block at buff, length len,
@@ -182,3 +184,4 @@ __sum16 ip_compute_csum(const void *buff, int len)
 {
 	return (__force __sum16)~from64to16(do_csum(buff,len));
 }
+EXPORT_SYMBOL(ip_compute_csum);
diff --git a/arch/alpha/lib/clear_page.S b/arch/alpha/lib/clear_page.S
index a221ae266e29..263d7393c0e7 100644
--- a/arch/alpha/lib/clear_page.S
+++ b/arch/alpha/lib/clear_page.S
@@ -3,7 +3,7 @@
  *
  * Zero an entire page.
  */
-
+#include <asm/export.h>
 	.text
 	.align 4
 	.global clear_page
@@ -37,3 +37,4 @@ clear_page:
 	nop
 
 	.end clear_page
+	EXPORT_SYMBOL(clear_page)
diff --git a/arch/alpha/lib/clear_user.S b/arch/alpha/lib/clear_user.S
index 8860316c1957..006f469fef73 100644
--- a/arch/alpha/lib/clear_user.S
+++ b/arch/alpha/lib/clear_user.S
@@ -8,22 +8,8 @@
  * right "bytes left to zero" value (and that it is updated only _after_
  * a successful copy).  There is also some rather minor exception setup
  * stuff.
- *
- * NOTE! This is not directly C-callable, because the calling semantics
- * are different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	exception pointer in $7
- *	return address in $28 (exceptions expect it there)
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6
  */
+#include <asm/export.h>
 
 /* Allow an exception for an insn; exit if we get one.  */
 #define EX(x,y...)			\
@@ -37,62 +23,63 @@
 	.set noreorder
 	.align 4
 
-	.globl __do_clear_user
-	.ent __do_clear_user
-	.frame	$30, 0, $28
+	.globl __clear_user
+	.ent __clear_user
+	.frame	$30, 0, $26
 	.prologue 0
 
 $loop:
 	and	$1, 3, $4	# e0    :
 	beq	$4, 1f		# .. e1 :
 
-0:	EX( stq_u $31, 0($6) )	# e0    : zero one word
+0:	EX( stq_u $31, 0($16) )	# e0    : zero one word
 	subq	$0, 8, $0	# .. e1 :
 	subq	$4, 1, $4	# e0    :
-	addq	$6, 8, $6	# .. e1 :
+	addq	$16, 8, $16	# .. e1 :
 	bne	$4, 0b		# e1    :
 	unop			#       :
 
 1:	bic	$1, 3, $1	# e0    :
 	beq	$1, $tail	# .. e1 :
 
-2:	EX( stq_u $31, 0($6) )	# e0    : zero four words
+2:	EX( stq_u $31, 0($16) )	# e0    : zero four words
 	subq	$0, 8, $0	# .. e1 :
-	EX( stq_u $31, 8($6) )	# e0    :
+	EX( stq_u $31, 8($16) )	# e0    :
 	subq	$0, 8, $0	# .. e1 :
-	EX( stq_u $31, 16($6) )	# e0    :
+	EX( stq_u $31, 16($16) )	# e0    :
 	subq	$0, 8, $0	# .. e1 :
-	EX( stq_u $31, 24($6) )	# e0    :
+	EX( stq_u $31, 24($16) )	# e0    :
 	subq	$0, 8, $0	# .. e1 :
 	subq	$1, 4, $1	# e0    :
-	addq	$6, 32, $6	# .. e1 :
+	addq	$16, 32, $16	# .. e1 :
 	bne	$1, 2b		# e1    :
 
 $tail:
 	bne	$2, 1f		# e1    : is there a tail to do?
-	ret	$31, ($28), 1	# .. e1 :
+	ret	$31, ($26), 1	# .. e1 :
 
-1:	EX( ldq_u $5, 0($6) )	# e0    :
+1:	EX( ldq_u $5, 0($16) )	# e0    :
 	clr	$0		# .. e1 :
 	nop			# e1    :
 	mskqh	$5, $0, $5	# e0    :
-	EX( stq_u $5, 0($6) )	# e0    :
-	ret	$31, ($28), 1	# .. e1 :
+	EX( stq_u $5, 0($16) )	# e0    :
+	ret	$31, ($26), 1	# .. e1 :
 
-__do_clear_user:
-	and	$6, 7, $4	# e0    : find dest misalignment
+__clear_user:
+	and	$17, $17, $0
+	and	$16, 7, $4	# e0    : find dest misalignment
 	beq	$0, $zerolength # .. e1 :
 	addq	$0, $4, $1	# e0    : bias counter
 	and	$1, 7, $2	# e1    : number of bytes in tail
 	srl	$1, 3, $1	# e0    :
 	beq	$4, $loop	# .. e1 :
 
-	EX( ldq_u $5, 0($6) )	# e0    : load dst word to mask back in
+	EX( ldq_u $5, 0($16) )	# e0    : load dst word to mask back in
 	beq	$1, $oneword	# .. e1 : sub-word store?
 
-	mskql	$5, $6, $5	# e0    : take care of misaligned head
-	addq	$6, 8, $6	# .. e1 :
-	EX( stq_u $5, -8($6) )	# e0    :
+	mskql	$5, $16, $5	# e0    : take care of misaligned head
+	addq	$16, 8, $16	# .. e1 :
+	EX( stq_u $5, -8($16) )	# e0    :
 	addq	$0, $4, $0	# .. e1 : bytes left -= 8 - misalignment
 	subq	$1, 1, $1	# e0    :
 	subq	$0, 8, $0	# .. e1 :
@@ -100,14 +87,15 @@ __do_clear_user:
 	unop			#       :
 
 $oneword:
-	mskql	$5, $6, $4	# e0    :
+	mskql	$5, $16, $4	# e0    :
 	mskqh	$5, $2, $5	# e0    :
 	or	$5, $4, $5	# e1    :
-	EX( stq_u $5, 0($6) )	# e0    :
+	EX( stq_u $5, 0($16) )	# e0    :
 	clr	$0		# .. e1 :
 
 $zerolength:
 $exception:
-	ret	$31, ($28), 1	# .. e1 :
+	ret	$31, ($26), 1	# .. e1 :
 
-	.end __do_clear_user
+	.end __clear_user
+	EXPORT_SYMBOL(__clear_user)
diff --git a/arch/alpha/lib/copy_page.S b/arch/alpha/lib/copy_page.S
index 9f3b97459cc6..2ee0bd0508c5 100644
--- a/arch/alpha/lib/copy_page.S
+++ b/arch/alpha/lib/copy_page.S
@@ -3,7 +3,7 @@
  *
  * Copy an entire page.
  */
-
+#include <asm/export.h>
 	.text
 	.align 4
 	.global copy_page
@@ -47,3 +47,4 @@ copy_page:
 	nop
 
 	.end copy_page
+	EXPORT_SYMBOL(copy_page)
diff --git a/arch/alpha/lib/copy_user.S b/arch/alpha/lib/copy_user.S
index 6f3fab9eb434..159f1b7e6e49 100644
--- a/arch/alpha/lib/copy_user.S
+++ b/arch/alpha/lib/copy_user.S
@@ -9,23 +9,10 @@
  * contains the right "bytes left to copy" value (and that it is updated
  * only _after_ a successful copy). There is also some rather minor
  * exception setup stuff..
- *
- * NOTE! This is not directly C-callable, because the calling semantics are
- * different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	source address in $7
- *	return address in $28
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6,$7
  */
 
+#include <asm/export.h>
+
 /* Allow an exception for an insn; exit if we get one.  */
 #define EXI(x,y...)			\
 	99: x,##y;			\
@@ -47,58 +34,59 @@
 	.ent __copy_user
 __copy_user:
 	.prologue 0
-	and $6,7,$3
+	and $18,$18,$0
+	and $16,7,$3
 	beq $0,$35
 	beq $3,$36
 	subq $3,8,$3
 	.align 4
 $37:
-	EXI( ldq_u $1,0($7) )
-	EXO( ldq_u $2,0($6) )
-	extbl $1,$7,$1
-	mskbl $2,$6,$2
-	insbl $1,$6,$1
+	EXI( ldq_u $1,0($17) )
+	EXO( ldq_u $2,0($16) )
+	extbl $1,$17,$1
+	mskbl $2,$16,$2
+	insbl $1,$16,$1
 	addq $3,1,$3
 	bis $1,$2,$1
-	EXO( stq_u $1,0($6) )
+	EXO( stq_u $1,0($16) )
 	subq $0,1,$0
-	addq $6,1,$6
-	addq $7,1,$7
+	addq $16,1,$16
+	addq $17,1,$17
 	beq $0,$41
 	bne $3,$37
 $36:
-	and $7,7,$1
+	and $17,7,$1
 	bic $0,7,$4
 	beq $1,$43
 	beq $4,$48
-	EXI( ldq_u $3,0($7) )
+	EXI( ldq_u $3,0($17) )
 	.align 4
 $50:
-	EXI( ldq_u $2,8($7) )
+	EXI( ldq_u $2,8($17) )
 	subq $4,8,$4
-	extql $3,$7,$3
-	extqh $2,$7,$1
+	extql $3,$17,$3
+	extqh $2,$17,$1
 	bis $3,$1,$1
-	EXO( stq $1,0($6) )
-	addq $7,8,$7
+	EXO( stq $1,0($16) )
+	addq $17,8,$17
 	subq $0,8,$0
-	addq $6,8,$6
+	addq $16,8,$16
 	bis $2,$2,$3
 	bne $4,$50
 $48:
 	beq $0,$41
 	.align 4
 $57:
-	EXI( ldq_u $1,0($7) )
-	EXO( ldq_u $2,0($6) )
-	extbl $1,$7,$1
-	mskbl $2,$6,$2
-	insbl $1,$6,$1
+	EXI( ldq_u $1,0($17) )
+	EXO( ldq_u $2,0($16) )
+	extbl $1,$17,$1
+	mskbl $2,$16,$2
+	insbl $1,$16,$1
 	bis $1,$2,$1
-	EXO( stq_u $1,0($6) )
+	EXO( stq_u $1,0($16) )
 	subq $0,1,$0
-	addq $6,1,$6
-	addq $7,1,$7
+	addq $16,1,$16
+	addq $17,1,$17
 	bne $0,$57
 	br $31,$41
 	.align 4
@@ -106,40 +94,27 @@ $43:
 	beq $4,$65
 	.align 4
 $66:
-	EXI( ldq $1,0($7) )
+	EXI( ldq $1,0($17) )
 	subq $4,8,$4
-	EXO( stq $1,0($6) )
-	addq $7,8,$7
+	EXO( stq $1,0($16) )
+	addq $17,8,$17
 	subq $0,8,$0
-	addq $6,8,$6
+	addq $16,8,$16
 	bne $4,$66
 $65:
 	beq $0,$41
-	EXI( ldq $2,0($7) )
-	EXO( ldq $1,0($6) )
+	EXI( ldq $2,0($17) )
+	EXO( ldq $1,0($16) )
 	mskql $2,$0,$2
 	mskqh $1,$0,$1
 	bis $2,$1,$2
-	EXO( stq $2,0($6) )
+	EXO( stq $2,0($16) )
 	bis $31,$31,$0
 $41:
 $35:
-$exitout:
-	ret $31,($28),1
-
 $exitin:
-	/* A stupid byte-by-byte zeroing of the rest of the output
-	   buffer.  This cures security holes by never leaving 
-	   random kernel data around to be copied elsewhere.  */
-
-	mov $0,$1
-$101:
-	EXO ( ldq_u $2,0($6) )
-	subq $1,1,$1
-	mskbl $2,$6,$2
-	EXO ( stq_u $2,0($6) )
-	addq $6,1,$6
-	bgt $1,$101
-	ret $31,($28),1
+$exitout:
+	ret $31,($26),1
 
 	.end __copy_user
+EXPORT_SYMBOL(__copy_user)
diff --git a/arch/alpha/lib/csum_ipv6_magic.S b/arch/alpha/lib/csum_ipv6_magic.S
index 2c2acb96deb6..e74b4544b0cc 100644
--- a/arch/alpha/lib/csum_ipv6_magic.S
+++ b/arch/alpha/lib/csum_ipv6_magic.S
@@ -12,6 +12,7 @@
  * added by Ivan Kokshaysky <ink@jurassic.park.msu.ru>
  */
 
+#include <asm/export.h>
 	.globl csum_ipv6_magic
 	.align 4
 	.ent csum_ipv6_magic
@@ -113,3 +114,4 @@ csum_ipv6_magic:
 	ret			# .. e1 :
 
 	.end csum_ipv6_magic
+	EXPORT_SYMBOL(csum_ipv6_magic)
diff --git a/arch/alpha/lib/csum_partial_copy.c b/arch/alpha/lib/csum_partial_copy.c
index 5675dca8dbb1..b4ff3b683bcd 100644
--- a/arch/alpha/lib/csum_partial_copy.c
+++ b/arch/alpha/lib/csum_partial_copy.c
@@ -374,6 +374,7 @@ csum_partial_copy_from_user(const void __user *src, void *dst, int len,
 	}
 	return (__force __wsum)checksum;
 }
+EXPORT_SYMBOL(csum_partial_copy_from_user);
 
 __wsum
 csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum)
@@ -386,3 +387,4 @@ csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum)
 	set_fs(oldfs);
 	return checksum;
 }
+EXPORT_SYMBOL(csum_partial_copy_nocheck);
diff --git a/arch/alpha/lib/dec_and_lock.c b/arch/alpha/lib/dec_and_lock.c
index f9f5fe830e9f..4221b40167ee 100644
--- a/arch/alpha/lib/dec_and_lock.c
+++ b/arch/alpha/lib/dec_and_lock.c
@@ -7,6 +7,7 @@
 
 #include <linux/spinlock.h>
 #include <linux/atomic.h>
+#include <linux/export.h>
 
   asm (".text					\n\
 	.global _atomic_dec_and_lock		\n\
@@ -39,3 +40,4 @@ static int __used atomic_dec_and_lock_1(atomic_t *atomic, spinlock_t *lock)
 	spin_unlock(lock);
 	return 0;
 }
+EXPORT_SYMBOL(_atomic_dec_and_lock);
diff --git a/arch/alpha/lib/divide.S b/arch/alpha/lib/divide.S
index 2d1a0484a99e..1e33bd127621 100644
--- a/arch/alpha/lib/divide.S
+++ b/arch/alpha/lib/divide.S
@@ -45,6 +45,7 @@
  *	$28 - compare status
  */
 
+#include <asm/export.h>
 #define halt .long 0
 
 /*
@@ -151,6 +152,7 @@ ufunction:
 	addq	$30,STACK,$30
 	ret	$31,($23),1
 	.end	ufunction
+EXPORT_SYMBOL(ufunction)
 
 /*
  * Uhh.. Ugly signed division. I'd rather not have it at all, but
@@ -193,3 +195,4 @@ sfunction:
 	addq	$30,STACK,$30
 	ret	$31,($23),1
 	.end	sfunction
+EXPORT_SYMBOL(sfunction)
diff --git a/arch/alpha/lib/ev6-clear_page.S b/arch/alpha/lib/ev6-clear_page.S
index adf4f7be0e2b..abe99e69a194 100644
--- a/arch/alpha/lib/ev6-clear_page.S
+++ b/arch/alpha/lib/ev6-clear_page.S
@@ -3,7 +3,7 @@
  *
  * Zero an entire page.
  */
-
+#include <asm/export.h>
         .text
         .align 4
         .global clear_page
@@ -52,3 +52,4 @@ clear_page:
 	nop
 
 	.end clear_page
+	EXPORT_SYMBOL(clear_page)
diff --git a/arch/alpha/lib/ev6-clear_user.S b/arch/alpha/lib/ev6-clear_user.S
index 4f42a16b7f53..e179e4757ef8 100644
--- a/arch/alpha/lib/ev6-clear_user.S
+++ b/arch/alpha/lib/ev6-clear_user.S
@@ -9,21 +9,6 @@
  * a successful copy).  There is also some rather minor exception setup
  * stuff.
  *
- * NOTE! This is not directly C-callable, because the calling semantics
- * are different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	exception pointer in $7
- *	return address in $28 (exceptions expect it there)
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6
- *
  * Much of the information about 21264 scheduling/coding comes from:
  *	Compiler Writer's Guide for the Alpha 21264
  *	abbreviated as 'CWG' in other comments here
@@ -43,6 +28,7 @@
  *	want to leave a hole (and we also want to avoid repeating lots of work)
  */
 
+#include <asm/export.h>
 /* Allow an exception for an insn; exit if we get one.  */
 #define EX(x,y...)			\
 	99: x,##y;			\
@@ -55,14 +41,15 @@
 	.set noreorder
 	.align 4
 
-	.globl __do_clear_user
-	.ent __do_clear_user
-	.frame	$30, 0, $28
+	.globl __clear_user
+	.ent __clear_user
+	.frame	$30, 0, $26
 	.prologue 0
 
 				# Pipeline info : Slotting & Comments
-__do_clear_user:
-	and	$6, 7, $4	# .. E  .. ..	: find dest head misalignment
+__clear_user:
+	and	$17, $17, $0
+	and	$16, 7, $4	# .. E  .. ..	: find dest head misalignment
 	beq	$0, $zerolength # U  .. .. ..	:  U L U L
 
 	addq	$0, $4, $1	# .. .. .. E	: bias counter
@@ -74,14 +61,14 @@ __do_clear_user:
 
 /*
  * Head is not aligned.  Write (8 - $4) bytes to head of destination
- * This means $6 is known to be misaligned
+ * This means $16 is known to be misaligned
  */
-	EX( ldq_u $5, 0($6) )	# .. .. .. L	: load dst word to mask back in
+	EX( ldq_u $5, 0($16) )	# .. .. .. L	: load dst word to mask back in
 	beq	$1, $onebyte	# .. .. U  ..	: sub-word store?
-	mskql	$5, $6, $5	# .. U  .. ..	: take care of misaligned head
-	addq	$6, 8, $6	# E  .. .. .. 	: L U U L
+	mskql	$5, $16, $5	# .. U  .. ..	: take care of misaligned head
+	addq	$16, 8, $16	# E  .. .. .. 	: L U U L
 
-	EX( stq_u $5, -8($6) )	# .. .. .. L	:
+	EX( stq_u $5, -8($16) )	# .. .. .. L	:
 	subq	$1, 1, $1	# .. .. E  ..	:
 	addq	$0, $4, $0	# .. E  .. ..	: bytes left -= 8 - misalignment
 	subq	$0, 8, $0	# E  .. .. ..	: U L U L
@@ -92,11 +79,11 @@ __do_clear_user:
  * values upon initial entry to the loop
  * $1 is number of quadwords to clear (zero is a valid value)
  * $2 is number of trailing bytes (0..7) ($2 never used...)
- * $6 is known to be aligned 0mod8
+ * $16 is known to be aligned 0mod8
  */
 $headalign:
 	subq	$1, 16, $4	# .. .. .. E	: If < 16, we can not use the huge loop
-	and	$6, 0x3f, $2	# .. .. E  ..	: Forward work for huge loop
+	and	$16, 0x3f, $2	# .. .. E  ..	: Forward work for huge loop
 	subq	$2, 0x40, $3	# .. E  .. ..	: bias counter (huge loop)
 	blt	$4, $trailquad	# U  .. .. ..	: U L U L
 
@@ -113,21 +100,21 @@ $headalign:
 	beq	$3, $bigalign	# U  .. .. ..	: U L U L : Aligned 0mod64
 
 $alignmod64:
-	EX( stq_u $31, 0($6) )	# .. .. .. L
+	EX( stq_u $31, 0($16) )	# .. .. .. L
 	addq	$3, 8, $3	# .. .. E  ..
 	subq	$0, 8, $0	# .. E  .. ..
 	nop			# E  .. .. ..	: U L U L
 
 	nop			# .. .. .. E
 	subq	$1, 1, $1	# .. .. E  ..
-	addq	$6, 8, $6	# .. E  .. ..
+	addq	$16, 8, $16	# .. E  .. ..
 	blt	$3, $alignmod64	# U  .. .. ..	: U L U L
 
 $bigalign:
 /*
  * $0 is the number of bytes left
  * $1 is the number of quads left
- * $6 is aligned 0mod64
+ * $16 is aligned 0mod64
  * we know that we'll be taking a minimum of one trip through
  * CWG Section 3.7.6: do not expect a sustained store rate of > 1/cycle
  * We are _not_ going to update $0 after every single store.  That
@@ -144,39 +131,39 @@ $bigalign:
 	nop			# E :
 	nop			# E :
 	nop			# E :
-	bis	$6,$6,$3	# E : U L U L : Initial wh64 address is dest
+	bis	$16,$16,$3	# E : U L U L : Initial wh64 address is dest
 	/* This might actually help for the current trip... */
 
 $do_wh64:
 	wh64	($3)		# .. .. .. L1	: memory subsystem hint
 	subq	$1, 16, $4	# .. .. E  ..	: Forward calculation - repeat the loop?
-	EX( stq_u $31, 0($6) )	# .. L  .. ..
+	EX( stq_u $31, 0($16) )	# .. L  .. ..
 	subq	$0, 8, $0	# E  .. .. ..	: U L U L
 
-	addq	$6, 128, $3	# E : Target address of wh64
-	EX( stq_u $31, 8($6) )	# L :
-	EX( stq_u $31, 16($6) )	# L :
+	addq	$16, 128, $3	# E : Target address of wh64
+	EX( stq_u $31, 8($16) )	# L :
+	EX( stq_u $31, 16($16) )	# L :
 	subq	$0, 16, $0	# E : U L L U
 
 	nop			# E :
-	EX( stq_u $31, 24($6) )	# L :
-	EX( stq_u $31, 32($6) )	# L :
+	EX( stq_u $31, 24($16) )	# L :
+	EX( stq_u $31, 32($16) )	# L :
 	subq	$0, 168, $5	# E : U L L U : two trips through the loop left?
 	/* 168 = 192 - 24, since we've already completed some stores */
 
 	subq	$0, 16, $0	# E :
-	EX( stq_u $31, 40($6) )	# L :
-	EX( stq_u $31, 48($6) )	# L :
-	cmovlt	$5, $6, $3	# E : U L L U : Latency 2, extra mapping cycle
+	EX( stq_u $31, 40($16) )	# L :
+	EX( stq_u $31, 48($16) )	# L :
+	cmovlt	$5, $16, $3	# E : U L L U : Latency 2, extra mapping cycle
 
 	subq	$1, 8, $1	# E :
 	subq	$0, 16, $0	# E :
-	EX( stq_u $31, 56($6) )	# L :
+	EX( stq_u $31, 56($16) )	# L :
 	nop			# E : U L U L
 
 	nop			# E :
 	subq	$0, 8, $0	# E :
-	addq	$6, 64, $6	# E :
+	addq	$16, 64, $16	# E :
 	bge	$4, $do_wh64	# U : U L U L
 
 $trailquad:
@@ -189,14 +176,14 @@ $trailquad:
 	beq	$1, $trailbytes	# U  .. .. ..	: U L U L : Only 0..7 bytes to go
 
 $onequad:
-	EX( stq_u $31, 0($6) )	# .. .. .. L
+	EX( stq_u $31, 0($16) )	# .. .. .. L
 	subq	$1, 1, $1	# .. .. E  ..
 	subq	$0, 8, $0	# .. E  .. ..
 	nop			# E  .. .. ..	: U L U L
 
 	nop			# .. .. .. E
 	nop			# .. .. E  ..
-	addq	$6, 8, $6	# .. E  .. ..
+	addq	$16, 8, $16	# .. E  .. ..
 	bgt	$1, $onequad	# U  .. .. ..	: U L U L
 
 	# We have an unknown number of bytes left to go.
@@ -210,9 +197,9 @@ $trailbytes:
 	# so we will use $0 as the loop counter
 	# We know for a fact that $0 > 0 zero due to previous context
 $onebyte:
-	EX( stb $31, 0($6) )	# .. .. .. L
+	EX( stb $31, 0($16) )	# .. .. .. L
 	subq	$0, 1, $0	# .. .. E  ..	:
-	addq	$6, 1, $6	# .. E  .. ..	:
+	addq	$16, 1, $16	# .. E  .. ..	:
 	bgt	$0, $onebyte	# U  .. .. ..	: U L U L
 
 $zerolength:
@@ -220,6 +207,6 @@ $exception:			# Destination for exception recovery(?)
 	nop			# .. .. .. E	:
 	nop			# .. .. E  ..	:
 	nop			# .. E  .. ..	:
-	ret	$31, ($28), 1	# L0 .. .. ..	: L U L U
-	.end __do_clear_user
-
+	ret	$31, ($26), 1	# L0 .. .. ..	: L U L U
+	.end __clear_user
+	EXPORT_SYMBOL(__clear_user)
diff --git a/arch/alpha/lib/ev6-copy_page.S b/arch/alpha/lib/ev6-copy_page.S
index b789db192754..77935061bddb 100644
--- a/arch/alpha/lib/ev6-copy_page.S
+++ b/arch/alpha/lib/ev6-copy_page.S
@@ -56,7 +56,7 @@
    destination pages are in the dcache, but it is my guess that this is
    less important than the dcache miss case.  */
 
-
+#include <asm/export.h>
 	.text
 	.align 4
 	.global copy_page
@@ -201,3 +201,4 @@ copy_page:
 	nop
 
 	.end copy_page
+	EXPORT_SYMBOL(copy_page)
diff --git a/arch/alpha/lib/ev6-copy_user.S b/arch/alpha/lib/ev6-copy_user.S
index db42ffe9c350..35e6710d0700 100644
--- a/arch/alpha/lib/ev6-copy_user.S
+++ b/arch/alpha/lib/ev6-copy_user.S
@@ -12,21 +12,6 @@
  * only _after_ a successful copy). There is also some rather minor
  * exception setup stuff..
  *
- * NOTE! This is not directly C-callable, because the calling semantics are
- * different:
- *
- * Inputs:
- *	length in $0
- *	destination address in $6
- *	source address in $7
- *	return address in $28
- *
- * Outputs:
- *	bytes left to copy in $0
- *
- * Clobbers:
- *	$1,$2,$3,$4,$5,$6,$7
- *
  * Much of the information about 21264 scheduling/coding comes from:
  *	Compiler Writer's Guide for the Alpha 21264
  *	abbreviated as 'CWG' in other comments here
@@ -37,6 +22,7 @@
  *	L	- lower subcluster; L0 - subcluster L0; L1 - subcluster L1
  */
 
+#include <asm/export.h>
 /* Allow an exception for an insn; exit if we get one.  */
 #define EXI(x,y...)			\
 	99: x,##y;			\
@@ -59,10 +45,11 @@
 				# Pipeline info: Slotting & Comments
 __copy_user:
 	.prologue 0
-	subq $0, 32, $1		# .. E  .. ..	: Is this going to be a small copy?
+	andq $18, $18, $0
+	subq $18, 32, $1	# .. E  .. ..	: Is this going to be a small copy?
 	beq $0, $zerolength	# U  .. .. ..	: U L U L
 
-	and $6,7,$3		# .. .. .. E	: is leading dest misalignment
+	and $16,7,$3		# .. .. .. E	: is leading dest misalignment
 	ble $1, $onebyteloop	# .. .. U  ..	: 1st branch : small amount of data
 	beq $3, $destaligned	# .. U  .. ..	: 2nd (one cycle fetcher stall)
 	subq $3, 8, $3		# E  .. .. ..	: L U U L : trip counter
@@ -72,17 +59,17 @@ __copy_user:
  * We know we have at least one trip through this loop
  */
 $aligndest:
-	EXI( ldbu $1,0($7) )	# .. .. .. L	: Keep loads separate from stores
-	addq $6,1,$6		# .. .. E  ..	: Section 3.8 in the CWG
+	EXI( ldbu $1,0($17) )	# .. .. .. L	: Keep loads separate from stores
+	addq $16,1,$16		# .. .. E  ..	: Section 3.8 in the CWG
 	addq $3,1,$3		# .. E  .. ..	:
 	nop			# E  .. .. ..	: U L U L
 
 /*
- * the -1 is to compensate for the inc($6) done in a previous quadpack
+ * the -1 is to compensate for the inc($16) done in a previous quadpack
  * which allows us zero dependencies within either quadpack in the loop
  */
-	EXO( stb $1,-1($6) )	# .. .. .. L	:
-	addq $7,1,$7		# .. .. E  ..	: Section 3.8 in the CWG
+	EXO( stb $1,-1($16) )	# .. .. .. L	:
+	addq $17,1,$17		# .. .. E  ..	: Section 3.8 in the CWG
 	subq $0,1,$0		# .. E  .. ..	:
 	bne $3, $aligndest	# U  .. .. ..	: U L U L
 
@@ -91,29 +78,29 @@ $aligndest:
  * If we arrived via branch, we have a minimum of 32 bytes
  */
 $destaligned:
-	and $7,7,$1		# .. .. .. E	: Check _current_ source alignment
+	and $17,7,$1		# .. .. .. E	: Check _current_ source alignment
 	bic $0,7,$4		# .. .. E  ..	: number bytes as a quadword loop
-	EXI( ldq_u $3,0($7) )	# .. L  .. ..	: Forward fetch for fallthrough code
+	EXI( ldq_u $3,0($17) )	# .. L  .. ..	: Forward fetch for fallthrough code
 	beq $1,$quadaligned	# U  .. .. ..	: U L U L
 
 /*
- * In the worst case, we've just executed an ldq_u here from 0($7)
+ * In the worst case, we've just executed an ldq_u here from 0($17)
  * and we'll repeat it once if we take the branch
  */
 
 /* Misaligned quadword loop - not unrolled.  Leave it that way. */
 $misquad:
-	EXI( ldq_u $2,8($7) )	# .. .. .. L	:
+	EXI( ldq_u $2,8($17) )	# .. .. .. L	:
 	subq $4,8,$4		# .. .. E  ..	:
-	extql $3,$7,$3		# .. U  .. ..	:
-	extqh $2,$7,$1		# U  .. .. ..	: U U L L
+	extql $3,$17,$3		# .. U  .. ..	:
+	extqh $2,$17,$1		# U  .. .. ..	: U U L L
 
 	bis $3,$1,$1		# .. .. .. E	:
-	EXO( stq $1,0($6) )	# .. .. L  ..	:
-	addq $7,8,$7		# .. E  .. ..	:
+	EXO( stq $1,0($16) )	# .. .. L  ..	:
+	addq $17,8,$17		# .. E  .. ..	:
 	subq $0,8,$0		# E  .. .. ..	: U L L U
 
-	addq $6,8,$6		# .. .. .. E	:
+	addq $16,8,$16		# .. .. .. E	:
 	bis $2,$2,$3		# .. .. E  ..	:
 	nop			# .. E  .. ..	:
 	bne $4,$misquad		# U  .. .. ..	: U L U L
@@ -124,8 +111,8 @@ $misquad:
 	beq $0,$zerolength	# U  .. .. ..	: U L U L
 
 /* We know we have at least one trip through the byte loop */
-	EXI ( ldbu $2,0($7) )	# .. .. .. L	: No loads in the same quad
-	addq $6,1,$6		# .. .. E  ..	: as the store (Section 3.8 in CWG)
+	EXI ( ldbu $2,0($17) )	# .. .. .. L	: No loads in the same quad
+	addq $16,1,$16		# .. .. E  ..	: as the store (Section 3.8 in CWG)
 	nop			# .. E  .. ..	:
 	br $31, $dirtyentry	# L0 .. .. ..	: L U U L
 /* Do the trailing byte loop load, then hop into the store part of the loop */
@@ -135,8 +122,8 @@ $misquad:
  * Based upon the usage context, it's worth the effort to unroll this loop
  * $0 - number of bytes to be moved
  * $4 - number of bytes to move as quadwords
- * $6 is current destination address
- * $7 is current source address
+ * $16 is current destination address
+ * $17 is current source address
  */
 $quadaligned:
 	subq	$4, 32, $2	# .. .. .. E	: do not unroll for small stuff
@@ -154,29 +141,29 @@ $quadaligned:
  * instruction memory hint instruction).
  */
 $unroll4:
-	EXI( ldq $1,0($7) )	# .. .. .. L
-	EXI( ldq $2,8($7) )	# .. .. L  ..
+	EXI( ldq $1,0($17) )	# .. .. .. L
+	EXI( ldq $2,8($17) )	# .. .. L  ..
 	subq	$4,32,$4	# .. E  .. ..
 	nop			# E  .. .. ..	: U U L L
 
-	addq	$7,16,$7	# .. .. .. E
-	EXO( stq $1,0($6) )	# .. .. L  ..
-	EXO( stq $2,8($6) )	# .. L  .. ..
+	addq	$17,16,$17	# .. .. .. E
+	EXO( stq $1,0($16) )	# .. .. L  ..
+	EXO( stq $2,8($16) )	# .. L  .. ..
 	subq	$0,16,$0	# E  .. .. ..	: U L L U
 
-	addq	$6,16,$6	# .. .. .. E
-	EXI( ldq $1,0($7) )	# .. .. L  ..
-	EXI( ldq $2,8($7) )	# .. L  .. ..
+	addq	$16,16,$16	# .. .. .. E
+	EXI( ldq $1,0($17) )	# .. .. L  ..
+	EXI( ldq $2,8($17) )	# .. L  .. ..
 	subq	$4, 32, $3	# E  .. .. ..	: U U L L : is there enough for another trip?
 
-	EXO( stq $1,0($6) )	# .. .. .. L
-	EXO( stq $2,8($6) )	# .. .. L  ..
+	EXO( stq $1,0($16) )	# .. .. .. L
+	EXO( stq $2,8($16) )	# .. .. L  ..
 	subq	$0,16,$0	# .. E  .. ..
-	addq	$7,16,$7	# E  .. .. ..	: U L L U
+	addq	$17,16,$17	# E  .. .. ..	: U L L U
 
 	nop			# .. .. .. E
 	nop			# .. .. E  ..
-	addq	$6,16,$6	# .. E  .. ..
+	addq	$16,16,$16	# .. E  .. ..
 	bgt	$3,$unroll4	# U  .. .. ..	: U L U L
 
 	nop
@@ -185,14 +172,14 @@ $unroll4:
 	beq	$4, $noquads
 
 $onequad:
-	EXI( ldq $1,0($7) )
+	EXI( ldq $1,0($17) )
 	subq	$4,8,$4
-	addq	$7,8,$7
+	addq	$17,8,$17
 	nop
 
-	EXO( stq $1,0($6) )
+	EXO( stq $1,0($16) )
 	subq	$0,8,$0
-	addq	$6,8,$6
+	addq	$16,8,$16
 	bne	$4,$onequad
 
 $noquads:
@@ -206,54 +193,33 @@ $noquads:
  * There's no point in doing a lot of complex alignment calculations to try to
  * to quadword stuff for a small amount of data.
  *	$0 - remaining number of bytes left to copy
- *	$6 - current dest addr
- *	$7 - current source addr
+ *	$16 - current dest addr
+ *	$17 - current source addr
  */
 
 $onebyteloop:
-	EXI ( ldbu $2,0($7) )	# .. .. .. L	: No loads in the same quad
-	addq $6,1,$6		# .. .. E  ..	: as the store (Section 3.8 in CWG)
+	EXI ( ldbu $2,0($17) )	# .. .. .. L	: No loads in the same quad
+	addq $16,1,$16		# .. .. E  ..	: as the store (Section 3.8 in CWG)
 	nop			# .. E  .. ..	:
 	nop			# E  .. .. ..	: U L U L
 
 $dirtyentry:
 /*
- * the -1 is to compensate for the inc($6) done in a previous quadpack
+ * the -1 is to compensate for the inc($16) done in a previous quadpack
  * which allows us zero dependencies within either quadpack in the loop
  */
-	EXO ( stb $2,-1($6) )	# .. .. .. L	:
-	addq $7,1,$7		# .. .. E  ..	: quadpack as the load
+	EXO ( stb $2,-1($16) )	# .. .. .. L	:
+	addq $17,1,$17		# .. .. E  ..	: quadpack as the load
 	subq $0,1,$0		# .. E  .. ..	: change count _after_ copy
 	bgt $0,$onebyteloop	# U  .. .. ..	: U L U L
 
 $zerolength:
+$exitin:
 $exitout:			# Destination for exception recovery(?)
 	nop			# .. .. .. E
 	nop			# .. .. E  ..
 	nop			# .. E  .. ..
-	ret $31,($28),1		# L0 .. .. ..	: L U L U
-
-$exitin:
-
-	/* A stupid byte-by-byte zeroing of the rest of the output
-	   buffer.  This cures security holes by never leaving 
-	   random kernel data around to be copied elsewhere.  */
-
-	nop
-	nop
-	nop
-	mov	$0,$1
-
-$101:
-	EXO ( stb $31,0($6) )	# L
-	subq $1,1,$1		# E
-	addq $6,1,$6		# E
-	bgt $1,$101		# U
-
-	nop
-	nop
-	nop
-	ret $31,($28),1		# L0
+	ret $31,($26),1		# L0 .. .. ..	: L U L U
 
 	.end __copy_user
-
+	EXPORT_SYMBOL(__copy_user)
diff --git a/arch/alpha/lib/ev6-csum_ipv6_magic.S b/arch/alpha/lib/ev6-csum_ipv6_magic.S
index fc0bc399f872..de62627ac4fe 100644
--- a/arch/alpha/lib/ev6-csum_ipv6_magic.S
+++ b/arch/alpha/lib/ev6-csum_ipv6_magic.S
@@ -52,6 +52,7 @@
  * may cause additional delay in rare cases (load-load replay traps).
  */
 
+#include <asm/export.h>
 	.globl csum_ipv6_magic
 	.align 4
 	.ent csum_ipv6_magic
@@ -148,3 +149,4 @@ csum_ipv6_magic:
 	ret			# L0 : L U L U
 
 	.end csum_ipv6_magic
+	EXPORT_SYMBOL(csum_ipv6_magic)
diff --git a/arch/alpha/lib/ev6-divide.S b/arch/alpha/lib/ev6-divide.S
index 2a82b9be93fa..d18dc0e96e3d 100644
--- a/arch/alpha/lib/ev6-divide.S
+++ b/arch/alpha/lib/ev6-divide.S
@@ -55,6 +55,7 @@
  * Try not to change the actual algorithm if possible for consistency.
  */
 
+#include <asm/export.h>
 #define halt .long 0
 
 /*
@@ -205,6 +206,7 @@ ufunction:
 	addq	$30,STACK,$30		# E :
 	ret	$31,($23),1		# L0 : L U U L
 	.end	ufunction
+EXPORT_SYMBOL(ufunction)
 
 /*
  * Uhh.. Ugly signed division. I'd rather not have it at all, but
@@ -257,3 +259,4 @@ sfunction:
 	addq	$30,STACK,$30		# E :
 	ret	$31,($23),1		# L0 : L U U L
 	.end	sfunction
+EXPORT_SYMBOL(sfunction)
diff --git a/arch/alpha/lib/ev6-memchr.S b/arch/alpha/lib/ev6-memchr.S
index 1a5f71b9d8b1..419adc53ccb4 100644
--- a/arch/alpha/lib/ev6-memchr.S
+++ b/arch/alpha/lib/ev6-memchr.S
@@ -27,7 +27,7 @@
  *	L	- lower subcluster; L0 - subcluster L0; L1 - subcluster L1
  * Try not to change the actual algorithm if possible for consistency.
  */
-
+#include <asm/export.h>
         .set noreorder
         .set noat
 
@@ -189,3 +189,4 @@ $not_found:
 	ret			# L0 :
 
         .end memchr
+	EXPORT_SYMBOL(memchr)
diff --git a/arch/alpha/lib/ev6-memcpy.S b/arch/alpha/lib/ev6-memcpy.S
index 52b37b0f2af5..b19798b2efc0 100644
--- a/arch/alpha/lib/ev6-memcpy.S
+++ b/arch/alpha/lib/ev6-memcpy.S
@@ -19,7 +19,7 @@
  * Temp usage notes:
  *	$1,$2,		- scratch
  */
-
+#include <asm/export.h>
 	.set noreorder
 	.set noat
 
@@ -242,6 +242,7 @@ $nomoredata:
 	nop				# E :
 
 	.end memcpy
+	EXPORT_SYMBOL(memcpy)
 
 /* For backwards module compatibility.  */
 __memcpy = memcpy
diff --git a/arch/alpha/lib/ev6-memset.S b/arch/alpha/lib/ev6-memset.S
index 356bb2fdd705..fed21c6893e8 100644
--- a/arch/alpha/lib/ev6-memset.S
+++ b/arch/alpha/lib/ev6-memset.S
@@ -26,7 +26,7 @@
  * as fixes will need to be made in multiple places.  The performance gain
  * is worth it.
  */
-
+#include <asm/export.h>
 	.set noat
 	.set noreorder
 .text
@@ -229,6 +229,7 @@ end_b:
 	nop
 	ret $31,($26),1		# L0 :
 	.end ___memset
+	EXPORT_SYMBOL(___memset)
 
 	/*
 	 * This is the original body of code, prior to replication and
@@ -406,6 +407,7 @@ end:
 	nop
 	ret $31,($26),1		# L0 :
 	.end __constant_c_memset
+	EXPORT_SYMBOL(__constant_c_memset)
 
 	/*
 	 * This is a replicant of the __constant_c_memset code, rescheduled
@@ -594,6 +596,9 @@ end_w:
 	ret $31,($26),1		# L0 :
 
 	.end __memsetw
+	EXPORT_SYMBOL(__memsetw)
 
 memset = ___memset
 __memset = ___memset
+	EXPORT_SYMBOL(memset)
+	EXPORT_SYMBOL(__memset)
diff --git a/arch/alpha/lib/ev67-strcat.S b/arch/alpha/lib/ev67-strcat.S
index c426fe3ed72f..b69f60419be1 100644
--- a/arch/alpha/lib/ev67-strcat.S
+++ b/arch/alpha/lib/ev67-strcat.S
@@ -19,7 +19,7 @@
  * string once.
  */
 
-
+#include <asm/export.h>
 	.text
 
 	.align 4
@@ -52,3 +52,4 @@ $found:	cttz	$2, $3		# U0 :
 	br	__stxcpy	# L0 :
 
 	.end strcat
+	EXPORT_SYMBOL(strcat)
diff --git a/arch/alpha/lib/ev67-strchr.S b/arch/alpha/lib/ev67-strchr.S
index fbb7b4ffade9..ea8f2f35db9c 100644
--- a/arch/alpha/lib/ev67-strchr.S
+++ b/arch/alpha/lib/ev67-strchr.S
@@ -15,7 +15,7 @@
  *	L	- lower subcluster; L0 - subcluster L0; L1 - subcluster L1
  * Try not to change the actual algorithm if possible for consistency.
  */
-
+#include <asm/export.h>
 #include <asm/regdef.h>
 
 	.set noreorder
@@ -86,3 +86,4 @@ $found:	negq    t0, t1		# E : clear all but least set bit
 	ret			# L0 :
 
 	.end strchr
+	EXPORT_SYMBOL(strchr)
diff --git a/arch/alpha/lib/ev67-strlen.S b/arch/alpha/lib/ev67-strlen.S
index 503928072523..736fd41884a8 100644
--- a/arch/alpha/lib/ev67-strlen.S
+++ b/arch/alpha/lib/ev67-strlen.S
@@ -17,7 +17,7 @@
  *	U	- upper subcluster; U0 - subcluster U0; U1 - subcluster U1
  *	L	- lower subcluster; L0 - subcluster L0; L1 - subcluster L1
  */
-
+#include <asm/export.h>
 	.set noreorder
 	.set noat
 
@@ -47,3 +47,4 @@ $found:
 	ret	$31, ($26)	# L0 :
 
 	.end	strlen
+	EXPORT_SYMBOL(strlen)
diff --git a/arch/alpha/lib/ev67-strncat.S b/arch/alpha/lib/ev67-strncat.S
index 4ae716cd2bfb..cd35cbade73a 100644
--- a/arch/alpha/lib/ev67-strncat.S
+++ b/arch/alpha/lib/ev67-strncat.S
@@ -20,7 +20,7 @@
  * Try not to change the actual algorithm if possible for consistency.
  */
 
-
+#include <asm/export.h>
 	.text
 
 	.align 4
@@ -92,3 +92,4 @@ $zerocount:
 	ret			# L0 :
 
 	.end strncat
+	EXPORT_SYMBOL(strncat)
diff --git a/arch/alpha/lib/ev67-strrchr.S b/arch/alpha/lib/ev67-strrchr.S
index dd0d8c6b9f59..747455f0328c 100644
--- a/arch/alpha/lib/ev67-strrchr.S
+++ b/arch/alpha/lib/ev67-strrchr.S
@@ -18,7 +18,7 @@
  *	L	- lower subcluster; L0 - subcluster L0; L1 - subcluster L1
  */
 
-
+#include <asm/export.h>
 #include <asm/regdef.h>
 
 	.set noreorder
@@ -107,3 +107,4 @@ $eos:
 	nop
 
 	.end strrchr
+	EXPORT_SYMBOL(strrchr)
diff --git a/arch/alpha/lib/fpreg.c b/arch/alpha/lib/fpreg.c
index 05017ba34c3c..4aa6dbfa14ee 100644
--- a/arch/alpha/lib/fpreg.c
+++ b/arch/alpha/lib/fpreg.c
@@ -4,6 +4,9 @@
  * (C) Copyright 1998 Linus Torvalds
  */
 
+#include <linux/compiler.h>
+#include <linux/export.h>
+
 #if defined(CONFIG_ALPHA_EV6) || defined(CONFIG_ALPHA_EV67)
 #define STT(reg,val)  asm volatile ("ftoit $f"#reg",%0" : "=r"(val));
 #else
@@ -52,6 +55,7 @@ alpha_read_fp_reg (unsigned long reg)
 	}
 	return val;
 }
+EXPORT_SYMBOL(alpha_read_fp_reg);
 
 #if defined(CONFIG_ALPHA_EV6) || defined(CONFIG_ALPHA_EV67)
 #define LDT(reg,val)  asm volatile ("itoft %0,$f"#reg : : "r"(val));
@@ -97,6 +101,7 @@ alpha_write_fp_reg (unsigned long reg, unsigned long val)
 	      case 31: LDT(31, val); break;
 	}
 }
+EXPORT_SYMBOL(alpha_write_fp_reg);
 
 #if defined(CONFIG_ALPHA_EV6) || defined(CONFIG_ALPHA_EV67)
 #define STS(reg,val)  asm volatile ("ftois $f"#reg",%0" : "=r"(val));
@@ -146,6 +151,7 @@ alpha_read_fp_reg_s (unsigned long reg)
 	}
 	return val;
 }
+EXPORT_SYMBOL(alpha_read_fp_reg_s);
 
 #if defined(CONFIG_ALPHA_EV6) || defined(CONFIG_ALPHA_EV67)
 #define LDS(reg,val)  asm volatile ("itofs %0,$f"#reg : : "r"(val));
@@ -191,3 +197,4 @@ alpha_write_fp_reg_s (unsigned long reg, unsigned long val)
 	      case 31: LDS(31, val); break;
 	}
 }
+EXPORT_SYMBOL(alpha_write_fp_reg_s);
diff --git a/arch/alpha/lib/memchr.S b/arch/alpha/lib/memchr.S
index 14427eeb555e..c13d3eca2e05 100644
--- a/arch/alpha/lib/memchr.S
+++ b/arch/alpha/lib/memchr.S
@@ -31,7 +31,7 @@ For correctness consider that:
       - only minimum number of quadwords may be accessed
       - the third argument is an unsigned long
 */
-
+#include <asm/export.h>
         .set noreorder
         .set noat
 
@@ -162,3 +162,4 @@ $not_found:
 	ret			# .. e1 :
 
         .end memchr
+	EXPORT_SYMBOL(memchr)
diff --git a/arch/alpha/lib/memcpy.c b/arch/alpha/lib/memcpy.c
index 64083fc73238..57d9291ad172 100644
--- a/arch/alpha/lib/memcpy.c
+++ b/arch/alpha/lib/memcpy.c
@@ -16,6 +16,7 @@
  */
 
 #include <linux/types.h>
+#include <linux/export.h>
 
 /*
  * This should be done in one go with ldq_u*2/mask/stq_u. Do it
@@ -158,6 +159,4 @@ void * memcpy(void * dest, const void *src, size_t n)
 	__memcpy_unaligned_up ((unsigned long) dest, (unsigned long) src, n);
 	return dest;
 }
-
-/* For backward modules compatibility, define __memcpy.  */
-asm("__memcpy = memcpy; .globl __memcpy");
+EXPORT_SYMBOL(memcpy);
diff --git a/arch/alpha/lib/memmove.S b/arch/alpha/lib/memmove.S
index eb3b6e02242f..ff6a39d38385 100644
--- a/arch/alpha/lib/memmove.S
+++ b/arch/alpha/lib/memmove.S
@@ -6,7 +6,7 @@
  * This is hand-massaged output from the original memcpy.c.  We defer to
  * memcpy whenever possible; the backwards copy loops are not unrolled.
  */
-        
+#include <asm/export.h>
 	.set noat
 	.set noreorder
 	.text
@@ -179,3 +179,4 @@ $egress:
 	nop
 
 	.end memmove
+	EXPORT_SYMBOL(memmove)
diff --git a/arch/alpha/lib/memset.S b/arch/alpha/lib/memset.S
index 76ccc6d1f364..89a26f5e89de 100644
--- a/arch/alpha/lib/memset.S
+++ b/arch/alpha/lib/memset.S
@@ -13,7 +13,7 @@
  * The scheduling comments are according to the EV5 documentation (and done by 
  * hand, so they might well be incorrect, please do tell me about it..)
  */
-
+#include <asm/export.h>
 	.set noat
 	.set noreorder
 .text
@@ -106,6 +106,8 @@ within_one_quad:
 end:
 	ret $31,($26),1		/* E1 */
 	.end ___memset
+EXPORT_SYMBOL(___memset)
+EXPORT_SYMBOL(__constant_c_memset)
 
 	.align 5
 	.ent __memsetw
@@ -122,6 +124,9 @@ __memsetw:
 	br __constant_c_memset	/* .. E1 */
 
 	.end __memsetw
+EXPORT_SYMBOL(__memsetw)
 
 memset = ___memset
 __memset = ___memset
+	EXPORT_SYMBOL(memset)
+	EXPORT_SYMBOL(__memset)
diff --git a/arch/alpha/lib/strcat.S b/arch/alpha/lib/strcat.S
index 393f50384878..249837b03d4b 100644
--- a/arch/alpha/lib/strcat.S
+++ b/arch/alpha/lib/strcat.S
@@ -4,6 +4,7 @@
  *
  * Append a null-terminated string from SRC to DST.
  */
+#include <asm/export.h>
 
 	.text
 
@@ -50,3 +51,4 @@ $found:	negq    $2, $3		# clear all but least set bit
 	br	__stxcpy
 
 	.end strcat
+EXPORT_SYMBOL(strcat);
diff --git a/arch/alpha/lib/strchr.S b/arch/alpha/lib/strchr.S
index 011a175e8329..7412a173ea39 100644
--- a/arch/alpha/lib/strchr.S
+++ b/arch/alpha/lib/strchr.S
@@ -5,7 +5,7 @@
  * Return the address of a given character within a null-terminated
  * string, or null if it is not found.
  */
-
+#include <asm/export.h>
 #include <asm/regdef.h>
 
 	.set noreorder
@@ -68,3 +68,4 @@ $retnull:
 	ret			# .. e1 :
 
 	.end strchr
+	EXPORT_SYMBOL(strchr)
diff --git a/arch/alpha/lib/strcpy.S b/arch/alpha/lib/strcpy.S
index e0728e4ad21f..98deae1e4d08 100644
--- a/arch/alpha/lib/strcpy.S
+++ b/arch/alpha/lib/strcpy.S
@@ -5,7 +5,7 @@
  * Copy a null-terminated string from SRC to DST.  Return a pointer
  * to the null-terminator in the source.
  */
-
+#include <asm/export.h>
 	.text
 
 	.align 3
@@ -21,3 +21,4 @@ strcpy:
 	br	__stxcpy	# do the copy
 
 	.end strcpy
+	EXPORT_SYMBOL(strcpy)
diff --git a/arch/alpha/lib/strlen.S b/arch/alpha/lib/strlen.S
index fe63353de152..79c416f71bac 100644
--- a/arch/alpha/lib/strlen.S
+++ b/arch/alpha/lib/strlen.S
@@ -11,7 +11,7 @@
  *	  do this instead of the 9 instructions that
  *	  binary search needs).
  */
-
+#include <asm/export.h>
 	.set noreorder
 	.set noat
 
@@ -55,3 +55,4 @@ done:	subq	$0, $16, $0
 	ret	$31, ($26)
 
 	.end	strlen
+	EXPORT_SYMBOL(strlen)
diff --git a/arch/alpha/lib/strncat.S b/arch/alpha/lib/strncat.S
index a8278163c972..6c29ea60869a 100644
--- a/arch/alpha/lib/strncat.S
+++ b/arch/alpha/lib/strncat.S
@@ -9,7 +9,7 @@
  * past count, whereas libc may write to count+1.  This follows the generic
  * implementation in lib/string.c and is, IMHO, more sensible.
  */
-
+#include <asm/export.h>
 	.text
 
 	.align 3
@@ -82,3 +82,4 @@ $zerocount:
 	ret
 
 	.end strncat
+	EXPORT_SYMBOL(strncat)
diff --git a/arch/alpha/lib/strncpy.S b/arch/alpha/lib/strncpy.S
index a46f7f3ad8c7..e102cf1567dd 100644
--- a/arch/alpha/lib/strncpy.S
+++ b/arch/alpha/lib/strncpy.S
@@ -10,7 +10,7 @@
  * version has cropped that bit o' nastiness as well as assuming that
  * __stxncpy is in range of a branch.
  */
-
+#include <asm/export.h>
 	.set noat
 	.set noreorder
 
@@ -79,3 +79,4 @@ $zerolen:
 	ret
 
 	.end	strncpy
+	EXPORT_SYMBOL(strncpy)
diff --git a/arch/alpha/lib/strrchr.S b/arch/alpha/lib/strrchr.S
index 1970dc07cfd1..4bc6cb4b9812 100644
--- a/arch/alpha/lib/strrchr.S
+++ b/arch/alpha/lib/strrchr.S
@@ -5,7 +5,7 @@
  * Return the address of the last occurrence of a given character
  * within a null-terminated string, or null if it is not found.
  */
-
+#include <asm/export.h>
 #include <asm/regdef.h>
 
 	.set noreorder
@@ -85,3 +85,4 @@ $retnull:
 	ret			# .. e1 :
 
 	.end strrchr
+	EXPORT_SYMBOL(strrchr)