Merge branch 'move_extents' of git://oss.oracle.com/git/tye/linux-2.6 into ocfs2-merge-window

Conflicts:
	fs/ocfs2/ioctl.c

7 files changed, 119 insertions, 171 deletions

diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c
index 3cc95dd0a3a6..075ddada4911 100644
--- a/arch/s390/mm/extmem.c
+++ b/arch/s390/mm/extmem.c
@@ -412,6 +412,7 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long
 	struct dcss_segment *seg;
 	int rc, diag_cc;
 
+	start_addr = end_addr = 0;
 	seg = kmalloc(sizeof(*seg), GFP_KERNEL | GFP_DMA);
 	if (seg == NULL) {
 		rc = -ENOMEM;
@@ -573,6 +574,7 @@ segment_modify_shared (char *name, int do_nonshared)
 	unsigned long start_addr, end_addr, dummy;
 	int rc, diag_cc;
 
+	start_addr = end_addr = 0;
 	mutex_lock(&dcss_lock);
 	seg = segment_by_name (name);
 	if (seg == NULL) {
@@ -681,8 +683,6 @@ void
 segment_save(char *name)
 {
 	struct dcss_segment *seg;
-	int startpfn = 0;
-	int endpfn = 0;
 	char cmd1[160];
 	char cmd2[80];
 	int i, response;
@@ -698,8 +698,6 @@ segment_save(char *name)
 		goto out;
 	}
 
-	startpfn = seg->start_addr >> PAGE_SHIFT;
-	endpfn = (seg->end) >> PAGE_SHIFT;
 	sprintf(cmd1, "DEFSEG %s", name);
 	for (i=0; i<seg->segcnt; i++) {
 		sprintf(cmd1+strlen(cmd1), " %lX-%lX %s",
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index ab988135e5c6..a0f9e730f26a 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -225,33 +225,6 @@ static noinline void do_sigbus(struct pt_regs *regs, long int_code,
 	force_sig_info(SIGBUS, &si, tsk);
 }
 
-#ifdef CONFIG_S390_EXEC_PROTECT
-static noinline int signal_return(struct pt_regs *regs, long int_code,
-				  unsigned long trans_exc_code)
-{
-	u16 instruction;
-	int rc;
-
-	rc = __get_user(instruction, (u16 __user *) regs->psw.addr);
-
-	if (!rc && instruction == 0x0a77) {
-		clear_tsk_thread_flag(current, TIF_PER_TRAP);
-		if (is_compat_task())
-			sys32_sigreturn();
-		else
-			sys_sigreturn();
-	} else if (!rc && instruction == 0x0aad) {
-		clear_tsk_thread_flag(current, TIF_PER_TRAP);
-		if (is_compat_task())
-			sys32_rt_sigreturn();
-		else
-			sys_rt_sigreturn();
-	} else
-		do_sigsegv(regs, int_code, SEGV_MAPERR, trans_exc_code);
-	return 0;
-}
-#endif /* CONFIG_S390_EXEC_PROTECT */
-
 static noinline void do_fault_error(struct pt_regs *regs, long int_code,
 				    unsigned long trans_exc_code, int fault)
 {
@@ -259,13 +232,6 @@ static noinline void do_fault_error(struct pt_regs *regs, long int_code,
 
 	switch (fault) {
 	case VM_FAULT_BADACCESS:
-#ifdef CONFIG_S390_EXEC_PROTECT
-		if ((regs->psw.mask & PSW_MASK_ASC) == PSW_ASC_SECONDARY &&
-		    (trans_exc_code & 3) == 0) {
-			signal_return(regs, int_code, trans_exc_code);
-			break;
-		}
-#endif /* CONFIG_S390_EXEC_PROTECT */
 	case VM_FAULT_BADMAP:
 		/* Bad memory access. Check if it is kernel or user space. */
 		if (regs->psw.mask & PSW_MASK_PSTATE) {
@@ -414,11 +380,6 @@ void __kprobes do_dat_exception(struct pt_regs *regs, long pgm_int_code,
 	int access, fault;
 
 	access = VM_READ | VM_EXEC | VM_WRITE;
-#ifdef CONFIG_S390_EXEC_PROTECT
-	if ((regs->psw.mask & PSW_MASK_ASC) == PSW_ASC_SECONDARY &&
-	    (trans_exc_code & 3) == 0)
-		access = VM_EXEC;
-#endif
 	fault = do_exception(regs, access, trans_exc_code);
 	if (unlikely(fault))
 		do_fault_error(regs, pgm_int_code & 255, trans_exc_code, fault);
@@ -491,22 +452,28 @@ static int __init nopfault(char *str)
 
 __setup("nopfault", nopfault);
 
-typedef struct {
-	__u16 refdiagc;
-	__u16 reffcode;
-	__u16 refdwlen;
-	__u16 refversn;
-	__u64 refgaddr;
-	__u64 refselmk;
-	__u64 refcmpmk;
-	__u64 reserved;
-} __attribute__ ((packed, aligned(8))) pfault_refbk_t;
+struct pfault_refbk {
+	u16 refdiagc;
+	u16 reffcode;
+	u16 refdwlen;
+	u16 refversn;
+	u64 refgaddr;
+	u64 refselmk;
+	u64 refcmpmk;
+	u64 reserved;
+} __attribute__ ((packed, aligned(8)));
 
 int pfault_init(void)
 {
-	pfault_refbk_t refbk =
-		{ 0x258, 0, 5, 2, __LC_CURRENT, 1ULL << 48, 1ULL << 48,
-		  __PF_RES_FIELD };
+	struct pfault_refbk refbk = {
+		.refdiagc = 0x258,
+		.reffcode = 0,
+		.refdwlen = 5,
+		.refversn = 2,
+		.refgaddr = __LC_CURRENT_PID,
+		.refselmk = 1ULL << 48,
+		.refcmpmk = 1ULL << 48,
+		.reserved = __PF_RES_FIELD };
         int rc;
 
 	if (!MACHINE_IS_VM || pfault_disable)
@@ -524,8 +491,12 @@ int pfault_init(void)
 
 void pfault_fini(void)
 {
-	pfault_refbk_t refbk =
-	{ 0x258, 1, 5, 2, 0ULL, 0ULL, 0ULL, 0ULL };
+	struct pfault_refbk refbk = {
+		.refdiagc = 0x258,
+		.reffcode = 1,
+		.refdwlen = 5,
+		.refversn = 2,
+	};
 
 	if (!MACHINE_IS_VM || pfault_disable)
 		return;
@@ -537,11 +508,15 @@ void pfault_fini(void)
 		: : "a" (&refbk), "m" (refbk) : "cc");
 }
 
+static DEFINE_SPINLOCK(pfault_lock);
+static LIST_HEAD(pfault_list);
+
 static void pfault_interrupt(unsigned int ext_int_code,
 			     unsigned int param32, unsigned long param64)
 {
 	struct task_struct *tsk;
 	__u16 subcode;
+	pid_t pid;
 
 	/*
 	 * Get the external interruption subcode & pfault
@@ -553,44 +528,79 @@ static void pfault_interrupt(unsigned int ext_int_code,
 	if ((subcode & 0xff00) != __SUBCODE_MASK)
 		return;
 	kstat_cpu(smp_processor_id()).irqs[EXTINT_PFL]++;
-
-	/*
-	 * Get the token (= address of the task structure of the affected task).
-	 */
-#ifdef CONFIG_64BIT
-	tsk = (struct task_struct *) param64;
-#else
-	tsk = (struct task_struct *) param32;
-#endif
-
+	if (subcode & 0x0080) {
+		/* Get the token (= pid of the affected task). */
+		pid = sizeof(void *) == 4 ? param32 : param64;
+		rcu_read_lock();
+		tsk = find_task_by_pid_ns(pid, &init_pid_ns);
+		if (tsk)
+			get_task_struct(tsk);
+		rcu_read_unlock();
+		if (!tsk)
+			return;
+	} else {
+		tsk = current;
+	}
+	spin_lock(&pfault_lock);
 	if (subcode & 0x0080) {
 		/* signal bit is set -> a page has been swapped in by VM */
-		if (xchg(&tsk->thread.pfault_wait, -1) != 0) {
+		if (tsk->thread.pfault_wait == 1) {
 			/* Initial interrupt was faster than the completion
 			 * interrupt. pfault_wait is valid. Set pfault_wait
 			 * back to zero and wake up the process. This can
 			 * safely be done because the task is still sleeping
 			 * and can't produce new pfaults. */
 			tsk->thread.pfault_wait = 0;
+			list_del(&tsk->thread.list);
 			wake_up_process(tsk);
-			put_task_struct(tsk);
+		} else {
+			/* Completion interrupt was faster than initial
+			 * interrupt. Set pfault_wait to -1 so the initial
+			 * interrupt doesn't put the task to sleep. */
+			tsk->thread.pfault_wait = -1;
 		}
+		put_task_struct(tsk);
 	} else {
 		/* signal bit not set -> a real page is missing. */
-		get_task_struct(tsk);
-		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
-		if (xchg(&tsk->thread.pfault_wait, 1) != 0) {
+		if (tsk->thread.pfault_wait == -1) {
 			/* Completion interrupt was faster than the initial
-			 * interrupt (swapped in a -1 for pfault_wait). Set
-			 * pfault_wait back to zero and exit. This can be
-			 * done safely because tsk is running in kernel 
-			 * mode and can't produce new pfaults. */
+			 * interrupt (pfault_wait == -1). Set pfault_wait
+			 * back to zero and exit. */
 			tsk->thread.pfault_wait = 0;
-			set_task_state(tsk, TASK_RUNNING);
-			put_task_struct(tsk);
-		} else
+		} else {
+			/* Initial interrupt arrived before completion
+			 * interrupt. Let the task sleep. */
+			tsk->thread.pfault_wait = 1;
+			list_add(&tsk->thread.list, &pfault_list);
+			set_task_state(tsk, TASK_UNINTERRUPTIBLE);
 			set_tsk_need_resched(tsk);
+		}
+	}
+	spin_unlock(&pfault_lock);
+}
+
+static int __cpuinit pfault_cpu_notify(struct notifier_block *self,
+				       unsigned long action, void *hcpu)
+{
+	struct thread_struct *thread, *next;
+	struct task_struct *tsk;
+
+	switch (action) {
+	case CPU_DEAD:
+	case CPU_DEAD_FROZEN:
+		spin_lock_irq(&pfault_lock);
+		list_for_each_entry_safe(thread, next, &pfault_list, list) {
+			thread->pfault_wait = 0;
+			list_del(&thread->list);
+			tsk = container_of(thread, struct task_struct, thread);
+			wake_up_process(tsk);
+		}
+		spin_unlock_irq(&pfault_lock);
+		break;
+	default:
+		break;
 	}
+	return NOTIFY_OK;
 }
 
 static int __init pfault_irq_init(void)
@@ -599,22 +609,21 @@ static int __init pfault_irq_init(void)
 
 	if (!MACHINE_IS_VM)
 		return 0;
-	/*
-	 * Try to get pfault pseudo page faults going.
-	 */
 	rc = register_external_interrupt(0x2603, pfault_interrupt);
-	if (rc) {
-		pfault_disable = 1;
-		return rc;
-	}
-	if (pfault_init() == 0)
-		return 0;
+	if (rc)
+		goto out_extint;
+	rc = pfault_init() == 0 ? 0 : -EOPNOTSUPP;
+	if (rc)
+		goto out_pfault;
+	hotcpu_notifier(pfault_cpu_notify, 0);
+	return 0;
 
-	/* Tough luck, no pfault. */
-	pfault_disable = 1;
+out_pfault:
 	unregister_external_interrupt(0x2603, pfault_interrupt);
-	return 0;
+out_extint:
+	pfault_disable = 1;
+	return rc;
 }
 early_initcall(pfault_irq_init);
 
-#endif
+#endif /* CONFIG_PFAULT */
diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c
index 639cd21f2218..a4d856db9154 100644
--- a/arch/s390/mm/hugetlbpage.c
+++ b/arch/s390/mm/hugetlbpage.c
@@ -13,7 +13,6 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 				   pte_t *pteptr, pte_t pteval)
 {
 	pmd_t *pmdp = (pmd_t *) pteptr;
-	pte_t shadow_pteval = pteval;
 	unsigned long mask;
 
 	if (!MACHINE_HAS_HPAGE) {
@@ -21,18 +20,9 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 		mask = pte_val(pteval) &
 				(_SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO);
 		pte_val(pteval) = (_SEGMENT_ENTRY + __pa(pteptr)) | mask;
-		if (mm->context.noexec) {
-			pteptr += PTRS_PER_PTE;
-			pte_val(shadow_pteval) =
-					(_SEGMENT_ENTRY + __pa(pteptr)) | mask;
-		}
 	}
 
 	pmd_val(*pmdp) = pte_val(pteval);
-	if (mm->context.noexec) {
-		pmdp = get_shadow_table(pmdp);
-		pmd_val(*pmdp) = pte_val(shadow_pteval);
-	}
 }
 
 int arch_prepare_hugepage(struct page *page)
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index bb409332a484..dfefc2171691 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -175,7 +175,8 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 		pmd = pmd_offset(pud, address);
 		pte = pte_offset_kernel(pmd, address);
 		if (!enable) {
-			ptep_invalidate(&init_mm, address, pte);
+			__ptep_ipte(address, pte);
+			pte_val(*pte) = _PAGE_TYPE_EMPTY;
 			continue;
 		}
 		*pte = mk_pte_phys(address, __pgprot(_PAGE_TYPE_RW));
diff --git a/arch/s390/mm/pageattr.c b/arch/s390/mm/pageattr.c
index f05edcc3beff..d013ed39743b 100644
--- a/arch/s390/mm/pageattr.c
+++ b/arch/s390/mm/pageattr.c
@@ -28,7 +28,7 @@ static void change_page_attr(unsigned long addr, int numpages,
 
 		pte = *ptep;
 		pte = set(pte);
-		ptep_invalidate(&init_mm, addr, ptep);
+		__ptep_ipte(addr, ptep);
 		*ptep = pte;
 		addr += PAGE_SIZE;
 	}
diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c
index e1850c28cd68..8d4330642512 100644
--- a/arch/s390/mm/pgtable.c
+++ b/arch/s390/mm/pgtable.c
@@ -40,7 +40,6 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 static DEFINE_PER_CPU(struct rcu_table_freelist *, rcu_table_freelist);
 
 static void __page_table_free(struct mm_struct *mm, unsigned long *table);
-static void __crst_table_free(struct mm_struct *mm, unsigned long *table);
 
 static struct rcu_table_freelist *rcu_table_freelist_get(struct mm_struct *mm)
 {
@@ -67,7 +66,7 @@ static void rcu_table_freelist_callback(struct rcu_head *head)
 	while (batch->pgt_index > 0)
 		__page_table_free(batch->mm, batch->table[--batch->pgt_index]);
 	while (batch->crst_index < RCU_FREELIST_SIZE)
-		__crst_table_free(batch->mm, batch->table[batch->crst_index++]);
+		crst_table_free(batch->mm, batch->table[batch->crst_index++]);
 	free_page((unsigned long) batch);
 }
 
@@ -125,63 +124,33 @@ static int __init parse_vmalloc(char *arg)
 }
 early_param("vmalloc", parse_vmalloc);
 
-unsigned long *crst_table_alloc(struct mm_struct *mm, int noexec)
+unsigned long *crst_table_alloc(struct mm_struct *mm)
 {
 	struct page *page = alloc_pages(GFP_KERNEL, ALLOC_ORDER);
 
 	if (!page)
 		return NULL;
-	page->index = 0;
-	if (noexec) {
-		struct page *shadow = alloc_pages(GFP_KERNEL, ALLOC_ORDER);
-		if (!shadow) {
-			__free_pages(page, ALLOC_ORDER);
-			return NULL;
-		}
-		page->index = page_to_phys(shadow);
-	}
-	spin_lock_bh(&mm->context.list_lock);
-	list_add(&page->lru, &mm->context.crst_list);
-	spin_unlock_bh(&mm->context.list_lock);
 	return (unsigned long *) page_to_phys(page);
 }
 
-static void __crst_table_free(struct mm_struct *mm, unsigned long *table)
-{
-	unsigned long *shadow = get_shadow_table(table);
-
-	if (shadow)
-		free_pages((unsigned long) shadow, ALLOC_ORDER);
-	free_pages((unsigned long) table, ALLOC_ORDER);
-}
-
 void crst_table_free(struct mm_struct *mm, unsigned long *table)
 {
-	struct page *page = virt_to_page(table);
-
-	spin_lock_bh(&mm->context.list_lock);
-	list_del(&page->lru);
-	spin_unlock_bh(&mm->context.list_lock);
-	__crst_table_free(mm, table);
+	free_pages((unsigned long) table, ALLOC_ORDER);
 }
 
 void crst_table_free_rcu(struct mm_struct *mm, unsigned long *table)
 {
 	struct rcu_table_freelist *batch;
-	struct page *page = virt_to_page(table);
 
-	spin_lock_bh(&mm->context.list_lock);
-	list_del(&page->lru);
-	spin_unlock_bh(&mm->context.list_lock);
 	if (atomic_read(&mm->mm_users) < 2 &&
 	    cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) {
-		__crst_table_free(mm, table);
+		crst_table_free(mm, table);
 		return;
 	}
 	batch = rcu_table_freelist_get(mm);
 	if (!batch) {
 		smp_call_function(smp_sync, NULL, 1);
-		__crst_table_free(mm, table);
+		crst_table_free(mm, table);
 		return;
 	}
 	batch->table[--batch->crst_index] = table;
@@ -197,7 +166,7 @@ int crst_table_upgrade(struct mm_struct *mm, unsigned long limit)
 
 	BUG_ON(limit > (1UL << 53));
 repeat:
-	table = crst_table_alloc(mm, mm->context.noexec);
+	table = crst_table_alloc(mm);
 	if (!table)
 		return -ENOMEM;
 	spin_lock_bh(&mm->page_table_lock);
@@ -273,7 +242,7 @@ unsigned long *page_table_alloc(struct mm_struct *mm)
 	unsigned long *table;
 	unsigned long bits;
 
-	bits = (mm->context.noexec || mm->context.has_pgste) ? 3UL : 1UL;
+	bits = (mm->context.has_pgste) ? 3UL : 1UL;
 	spin_lock_bh(&mm->context.list_lock);
 	page = NULL;
 	if (!list_empty(&mm->context.pgtable_list)) {
@@ -329,7 +298,7 @@ void page_table_free(struct mm_struct *mm, unsigned long *table)
 	struct page *page;
 	unsigned long bits;
 
-	bits = (mm->context.noexec || mm->context.has_pgste) ? 3UL : 1UL;
+	bits = (mm->context.has_pgste) ? 3UL : 1UL;
 	bits <<= (__pa(table) & (PAGE_SIZE - 1)) / 256 / sizeof(unsigned long);
 	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
 	spin_lock_bh(&mm->context.list_lock);
@@ -366,7 +335,7 @@ void page_table_free_rcu(struct mm_struct *mm, unsigned long *table)
 		page_table_free(mm, table);
 		return;
 	}
-	bits = (mm->context.noexec || mm->context.has_pgste) ? 3UL : 1UL;
+	bits = (mm->context.has_pgste) ? 3UL : 1UL;
 	bits <<= (__pa(table) & (PAGE_SIZE - 1)) / 256 / sizeof(unsigned long);
 	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
 	spin_lock_bh(&mm->context.list_lock);
@@ -379,25 +348,6 @@ void page_table_free_rcu(struct mm_struct *mm, unsigned long *table)
 		rcu_table_freelist_finish();
 }
 
-void disable_noexec(struct mm_struct *mm, struct task_struct *tsk)
-{
-	struct page *page;
-
-	spin_lock_bh(&mm->context.list_lock);
-	/* Free shadow region and segment tables. */
-	list_for_each_entry(page, &mm->context.crst_list, lru)
-		if (page->index) {
-			free_pages((unsigned long) page->index, ALLOC_ORDER);
-			page->index = 0;
-		}
-	/* "Free" second halves of page tables. */
-	list_for_each_entry(page, &mm->context.pgtable_list, lru)
-		page->flags &= ~SECOND_HALVES;
-	spin_unlock_bh(&mm->context.list_lock);
-	mm->context.noexec = 0;
-	update_mm(mm, tsk);
-}
-
 /*
  * switch on pgstes for its userspace process (for kvm)
  */
diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index 34c43f23b28c..8c1970d1dd91 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -95,7 +95,7 @@ static int vmem_add_mem(unsigned long start, unsigned long size, int ro)
 			pu_dir = vmem_pud_alloc();
 			if (!pu_dir)
 				goto out;
-			pgd_populate_kernel(&init_mm, pg_dir, pu_dir);
+			pgd_populate(&init_mm, pg_dir, pu_dir);
 		}
 
 		pu_dir = pud_offset(pg_dir, address);
@@ -103,7 +103,7 @@ static int vmem_add_mem(unsigned long start, unsigned long size, int ro)
 			pm_dir = vmem_pmd_alloc();
 			if (!pm_dir)
 				goto out;
-			pud_populate_kernel(&init_mm, pu_dir, pm_dir);
+			pud_populate(&init_mm, pu_dir, pm_dir);
 		}
 
 		pte = mk_pte_phys(address, __pgprot(ro ? _PAGE_RO : 0));
@@ -123,7 +123,7 @@ static int vmem_add_mem(unsigned long start, unsigned long size, int ro)
 			pt_dir = vmem_pte_alloc();
 			if (!pt_dir)
 				goto out;
-			pmd_populate_kernel(&init_mm, pm_dir, pt_dir);
+			pmd_populate(&init_mm, pm_dir, pt_dir);
 		}
 
 		pt_dir = pte_offset_kernel(pm_dir, address);
@@ -159,7 +159,7 @@ static void vmem_remove_range(unsigned long start, unsigned long size)
 			continue;
 
 		if (pmd_huge(*pm_dir)) {
-			pmd_clear_kernel(pm_dir);
+			pmd_clear(pm_dir);
 			address += HPAGE_SIZE - PAGE_SIZE;
 			continue;
 		}
@@ -192,7 +192,7 @@ int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node)
 			pu_dir = vmem_pud_alloc();
 			if (!pu_dir)
 				goto out;
-			pgd_populate_kernel(&init_mm, pg_dir, pu_dir);
+			pgd_populate(&init_mm, pg_dir, pu_dir);
 		}
 
 		pu_dir = pud_offset(pg_dir, address);
@@ -200,7 +200,7 @@ int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node)
 			pm_dir = vmem_pmd_alloc();
 			if (!pm_dir)
 				goto out;
-			pud_populate_kernel(&init_mm, pu_dir, pm_dir);
+			pud_populate(&init_mm, pu_dir, pm_dir);
 		}
 
 		pm_dir = pmd_offset(pu_dir, address);
@@ -208,7 +208,7 @@ int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node)
 			pt_dir = vmem_pte_alloc();
 			if (!pt_dir)
 				goto out;
-			pmd_populate_kernel(&init_mm, pm_dir, pt_dir);
+			pmd_populate(&init_mm, pm_dir, pt_dir);
 		}
 
 		pt_dir = pte_offset_kernel(pm_dir, address);