Merge branch 'akpm' (incoming from Andrew)

Merge first batch of fixes from Andrew Morton:

 - A couple of kthread changes

 - A few minor audit patches

 - A number of fbdev patches.  Florian remains AWOL so I'm picking up
   some of these.

 - A few kbuild things

 - ocfs2 updates

 - Almost all of the MM queue

(And in the meantime, I already have the second big batch from Andrew
pending in my mailbox ;^)

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (149 commits)
  memcg: take reference before releasing rcu_read_lock
  mem hotunplug: fix kfree() of bootmem memory
  mmKconfig: add an option to disable bounce
  mm, nobootmem: do memset() after memblock_reserve()
  mm, nobootmem: clean-up of free_low_memory_core_early()
  fs/buffer.c: remove unnecessary init operation after allocating buffer_head.
  numa, cpu hotplug: change links of CPU and node when changing node number by onlining CPU
  mm: fix memory_hotplug.c printk format warning
  mm: swap: mark swap pages writeback before queueing for direct IO
  swap: redirty page if page write fails on swap file
  mm, memcg: give exiting processes access to memory reserves
  thp: fix huge zero page logic for page with pfn == 0
  memcg: avoid accessing memcg after releasing reference
  fs: fix fsync() error reporting
  memblock: fix missing comment of memblock_insert_region()
  mm: Remove unused parameter of pages_correctly_reserved()
  firmware, memmap: fix firmware_map_entry leak
  mm/vmstat: add note on safety of drain_zonestat
  mm: thp: add split tail pages to shrink page list in page reclaim
  mm: allow for outstanding swap writeback accounting
  ...

1 files changed, 140 insertions, 78 deletions

diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 0f751f2068c3..72043d6c88c0 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -249,19 +249,9 @@ EXPORT_SYMBOL(vmalloc_to_pfn);
 #define VM_LAZY_FREEING	0x02
 #define VM_VM_AREA	0x04
 
-struct vmap_area {
-	unsigned long va_start;
-	unsigned long va_end;
-	unsigned long flags;
-	struct rb_node rb_node;		/* address sorted rbtree */
-	struct list_head list;		/* address sorted list */
-	struct list_head purge_list;	/* "lazy purge" list */
-	struct vm_struct *vm;
-	struct rcu_head rcu_head;
-};
-
 static DEFINE_SPINLOCK(vmap_area_lock);
-static LIST_HEAD(vmap_area_list);
+/* Export for kexec only */
+LIST_HEAD(vmap_area_list);
 static struct rb_root vmap_area_root = RB_ROOT;
 
 /* The vmap cache globals are protected by vmap_area_lock */
@@ -313,7 +303,7 @@ static void __insert_vmap_area(struct vmap_area *va)
 	rb_link_node(&va->rb_node, parent, p);
 	rb_insert_color(&va->rb_node, &vmap_area_root);
 
-	/* address-sort this list so it is usable like the vmlist */
+	/* address-sort this list */
 	tmp = rb_prev(&va->rb_node);
 	if (tmp) {
 		struct vmap_area *prev;
@@ -1125,6 +1115,7 @@ void *vm_map_ram(struct page **pages, unsigned int count, int node, pgprot_t pro
 }
 EXPORT_SYMBOL(vm_map_ram);
 
+static struct vm_struct *vmlist __initdata;
 /**
  * vm_area_add_early - add vmap area early during boot
  * @vm: vm_struct to add
@@ -1283,41 +1274,35 @@ int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages)
 }
 EXPORT_SYMBOL_GPL(map_vm_area);
 
-/*** Old vmalloc interfaces ***/
-DEFINE_RWLOCK(vmlist_lock);
-struct vm_struct *vmlist;
-
 static void setup_vmalloc_vm(struct vm_struct *vm, struct vmap_area *va,
 			      unsigned long flags, const void *caller)
 {
+	spin_lock(&vmap_area_lock);
 	vm->flags = flags;
 	vm->addr = (void *)va->va_start;
 	vm->size = va->va_end - va->va_start;
 	vm->caller = caller;
 	va->vm = vm;
 	va->flags |= VM_VM_AREA;
+	spin_unlock(&vmap_area_lock);
 }
 
-static void insert_vmalloc_vmlist(struct vm_struct *vm)
+static void clear_vm_unlist(struct vm_struct *vm)
 {
-	struct vm_struct *tmp, **p;
-
+	/*
+	 * Before removing VM_UNLIST,
+	 * we should make sure that vm has proper values.
+	 * Pair with smp_rmb() in show_numa_info().
+	 */
+	smp_wmb();
 	vm->flags &= ~VM_UNLIST;
-	write_lock(&vmlist_lock);
-	for (p = &vmlist; (tmp = *p) != NULL; p = &tmp->next) {
-		if (tmp->addr >= vm->addr)
-			break;
-	}
-	vm->next = *p;
-	*p = vm;
-	write_unlock(&vmlist_lock);
 }
 
 static void insert_vmalloc_vm(struct vm_struct *vm, struct vmap_area *va,
 			      unsigned long flags, const void *caller)
 {
 	setup_vmalloc_vm(vm, va, flags, caller);
-	insert_vmalloc_vmlist(vm);
+	clear_vm_unlist(vm);
 }
 
 static struct vm_struct *__get_vm_area_node(unsigned long size,
@@ -1360,10 +1345,9 @@ static struct vm_struct *__get_vm_area_node(unsigned long size,
 
 	/*
 	 * When this function is called from __vmalloc_node_range,
-	 * we do not add vm_struct to vmlist here to avoid
-	 * accessing uninitialized members of vm_struct such as
-	 * pages and nr_pages fields. They will be set later.
-	 * To distinguish it from others, we use a VM_UNLIST flag.
+	 * we add VM_UNLIST flag to avoid accessing uninitialized
+	 * members of vm_struct such as pages and nr_pages fields.
+	 * They will be set later.
 	 */
 	if (flags & VM_UNLIST)
 		setup_vmalloc_vm(area, va, flags, caller);
@@ -1447,19 +1431,10 @@ struct vm_struct *remove_vm_area(const void *addr)
 	if (va && va->flags & VM_VM_AREA) {
 		struct vm_struct *vm = va->vm;
 
-		if (!(vm->flags & VM_UNLIST)) {
-			struct vm_struct *tmp, **p;
-			/*
-			 * remove from list and disallow access to
-			 * this vm_struct before unmap. (address range
-			 * confliction is maintained by vmap.)
-			 */
-			write_lock(&vmlist_lock);
-			for (p = &vmlist; (tmp = *p) != vm; p = &tmp->next)
-				;
-			*p = tmp->next;
-			write_unlock(&vmlist_lock);
-		}
+		spin_lock(&vmap_area_lock);
+		va->vm = NULL;
+		va->flags &= ~VM_VM_AREA;
+		spin_unlock(&vmap_area_lock);
 
 		vmap_debug_free_range(va->va_start, va->va_end);
 		free_unmap_vmap_area(va);
@@ -1680,10 +1655,11 @@ void *__vmalloc_node_range(unsigned long size, unsigned long align,
 		return NULL;
 
 	/*
-	 * In this function, newly allocated vm_struct is not added
-	 * to vmlist at __get_vm_area_node(). so, it is added here.
+	 * In this function, newly allocated vm_struct has VM_UNLIST flag.
+	 * It means that vm_struct is not fully initialized.
+	 * Now, it is fully initialized, so remove this flag here.
 	 */
-	insert_vmalloc_vmlist(area);
+	clear_vm_unlist(area);
 
 	/*
 	 * A ref_count = 3 is needed because the vm_struct and vmap_area
@@ -2005,7 +1981,8 @@ static int aligned_vwrite(char *buf, char *addr, unsigned long count)
 
 long vread(char *buf, char *addr, unsigned long count)
 {
-	struct vm_struct *tmp;
+	struct vmap_area *va;
+	struct vm_struct *vm;
 	char *vaddr, *buf_start = buf;
 	unsigned long buflen = count;
 	unsigned long n;
@@ -2014,10 +1991,17 @@ long vread(char *buf, char *addr, unsigned long count)
 	if ((unsigned long) addr + count < count)
 		count = -(unsigned long) addr;
 
-	read_lock(&vmlist_lock);
-	for (tmp = vmlist; count && tmp; tmp = tmp->next) {
-		vaddr = (char *) tmp->addr;
-		if (addr >= vaddr + tmp->size - PAGE_SIZE)
+	spin_lock(&vmap_area_lock);
+	list_for_each_entry(va, &vmap_area_list, list) {
+		if (!count)
+			break;
+
+		if (!(va->flags & VM_VM_AREA))
+			continue;
+
+		vm = va->vm;
+		vaddr = (char *) vm->addr;
+		if (addr >= vaddr + vm->size - PAGE_SIZE)
 			continue;
 		while (addr < vaddr) {
 			if (count == 0)
@@ -2027,10 +2011,10 @@ long vread(char *buf, char *addr, unsigned long count)
 			addr++;
 			count--;
 		}
-		n = vaddr + tmp->size - PAGE_SIZE - addr;
+		n = vaddr + vm->size - PAGE_SIZE - addr;
 		if (n > count)
 			n = count;
-		if (!(tmp->flags & VM_IOREMAP))
+		if (!(vm->flags & VM_IOREMAP))
 			aligned_vread(buf, addr, n);
 		else /* IOREMAP area is treated as memory hole */
 			memset(buf, 0, n);
@@ -2039,7 +2023,7 @@ long vread(char *buf, char *addr, unsigned long count)
 		count -= n;
 	}
 finished:
-	read_unlock(&vmlist_lock);
+	spin_unlock(&vmap_area_lock);
 
 	if (buf == buf_start)
 		return 0;
@@ -2078,7 +2062,8 @@ finished:
 
 long vwrite(char *buf, char *addr, unsigned long count)
 {
-	struct vm_struct *tmp;
+	struct vmap_area *va;
+	struct vm_struct *vm;
 	char *vaddr;
 	unsigned long n, buflen;
 	int copied = 0;
@@ -2088,10 +2073,17 @@ long vwrite(char *buf, char *addr, unsigned long count)
 		count = -(unsigned long) addr;
 	buflen = count;
 
-	read_lock(&vmlist_lock);
-	for (tmp = vmlist; count && tmp; tmp = tmp->next) {
-		vaddr = (char *) tmp->addr;
-		if (addr >= vaddr + tmp->size - PAGE_SIZE)
+	spin_lock(&vmap_area_lock);
+	list_for_each_entry(va, &vmap_area_list, list) {
+		if (!count)
+			break;
+
+		if (!(va->flags & VM_VM_AREA))
+			continue;
+
+		vm = va->vm;
+		vaddr = (char *) vm->addr;
+		if (addr >= vaddr + vm->size - PAGE_SIZE)
 			continue;
 		while (addr < vaddr) {
 			if (count == 0)
@@ -2100,10 +2092,10 @@ long vwrite(char *buf, char *addr, unsigned long count)
 			addr++;
 			count--;
 		}
-		n = vaddr + tmp->size - PAGE_SIZE - addr;
+		n = vaddr + vm->size - PAGE_SIZE - addr;
 		if (n > count)
 			n = count;
-		if (!(tmp->flags & VM_IOREMAP)) {
+		if (!(vm->flags & VM_IOREMAP)) {
 			aligned_vwrite(buf, addr, n);
 			copied++;
 		}
@@ -2112,7 +2104,7 @@ long vwrite(char *buf, char *addr, unsigned long count)
 		count -= n;
 	}
 finished:
-	read_unlock(&vmlist_lock);
+	spin_unlock(&vmap_area_lock);
 	if (!copied)
 		return 0;
 	return buflen;
@@ -2519,19 +2511,19 @@ void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms)
 
 #ifdef CONFIG_PROC_FS
 static void *s_start(struct seq_file *m, loff_t *pos)
-	__acquires(&vmlist_lock)
+	__acquires(&vmap_area_lock)
 {
 	loff_t n = *pos;
-	struct vm_struct *v;
+	struct vmap_area *va;
 
-	read_lock(&vmlist_lock);
-	v = vmlist;
-	while (n > 0 && v) {
+	spin_lock(&vmap_area_lock);
+	va = list_entry((&vmap_area_list)->next, typeof(*va), list);
+	while (n > 0 && &va->list != &vmap_area_list) {
 		n--;
-		v = v->next;
+		va = list_entry(va->list.next, typeof(*va), list);
 	}
-	if (!n)
-		return v;
+	if (!n && &va->list != &vmap_area_list)
+		return va;
 
 	return NULL;
 
@@ -2539,16 +2531,20 @@ static void *s_start(struct seq_file *m, loff_t *pos)
 
 static void *s_next(struct seq_file *m, void *p, loff_t *pos)
 {
-	struct vm_struct *v = p;
+	struct vmap_area *va = p, *next;
 
 	++*pos;
-	return v->next;
+	next = list_entry(va->list.next, typeof(*va), list);
+	if (&next->list != &vmap_area_list)
+		return next;
+
+	return NULL;
 }
 
 static void s_stop(struct seq_file *m, void *p)
-	__releases(&vmlist_lock)
+	__releases(&vmap_area_lock)
 {
-	read_unlock(&vmlist_lock);
+	spin_unlock(&vmap_area_lock);
 }
 
 static void show_numa_info(struct seq_file *m, struct vm_struct *v)
@@ -2559,6 +2555,11 @@ static void show_numa_info(struct seq_file *m, struct vm_struct *v)
 		if (!counters)
 			return;
 
+		/* Pair with smp_wmb() in clear_vm_unlist() */
+		smp_rmb();
+		if (v->flags & VM_UNLIST)
+			return;
+
 		memset(counters, 0, nr_node_ids * sizeof(unsigned int));
 
 		for (nr = 0; nr < v->nr_pages; nr++)
@@ -2572,7 +2573,20 @@ static void show_numa_info(struct seq_file *m, struct vm_struct *v)
 
 static int s_show(struct seq_file *m, void *p)
 {
-	struct vm_struct *v = p;
+	struct vmap_area *va = p;
+	struct vm_struct *v;
+
+	if (va->flags & (VM_LAZY_FREE | VM_LAZY_FREEING))
+		return 0;
+
+	if (!(va->flags & VM_VM_AREA)) {
+		seq_printf(m, "0x%pK-0x%pK %7ld vm_map_ram\n",
+			(void *)va->va_start, (void *)va->va_end,
+					va->va_end - va->va_start);
+		return 0;
+	}
+
+	v = va->vm;
 
 	seq_printf(m, "0x%pK-0x%pK %7ld",
 		v->addr, v->addr + v->size, v->size);
@@ -2645,5 +2659,53 @@ static int __init proc_vmalloc_init(void)
 	return 0;
 }
 module_init(proc_vmalloc_init);
+
+void get_vmalloc_info(struct vmalloc_info *vmi)
+{
+	struct vmap_area *va;
+	unsigned long free_area_size;
+	unsigned long prev_end;
+
+	vmi->used = 0;
+	vmi->largest_chunk = 0;
+
+	prev_end = VMALLOC_START;
+
+	spin_lock(&vmap_area_lock);
+
+	if (list_empty(&vmap_area_list)) {
+		vmi->largest_chunk = VMALLOC_TOTAL;
+		goto out;
+	}
+
+	list_for_each_entry(va, &vmap_area_list, list) {
+		unsigned long addr = va->va_start;
+
+		/*
+		 * Some archs keep another range for modules in vmalloc space
+		 */
+		if (addr < VMALLOC_START)
+			continue;
+		if (addr >= VMALLOC_END)
+			break;
+
+		if (va->flags & (VM_LAZY_FREE | VM_LAZY_FREEING))
+			continue;
+
+		vmi->used += (va->va_end - va->va_start);
+
+		free_area_size = addr - prev_end;
+		if (vmi->largest_chunk < free_area_size)
+			vmi->largest_chunk = free_area_size;
+
+		prev_end = va->va_end;
+	}
+
+	if (VMALLOC_END - prev_end > vmi->largest_chunk)
+		vmi->largest_chunk = VMALLOC_END - prev_end;
+
+out:
+	spin_unlock(&vmap_area_lock);
+}
 #endif