1 files changed, 328 insertions, 232 deletions

diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 86ce9a526c1..f64632b6719 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -27,10 +27,32 @@
 #include <linux/pfn.h>
 #include <linux/kmemleak.h>
 #include <linux/atomic.h>
+#include <linux/compiler.h>
+#include <linux/llist.h>
+
 #include <asm/uaccess.h>
 #include <asm/tlbflush.h>
 #include <asm/shmparam.h>
 
+struct vfree_deferred {
+	struct llist_head list;
+	struct work_struct wq;
+};
+static DEFINE_PER_CPU(struct vfree_deferred, vfree_deferred);
+
+static void __vunmap(const void *, int);
+
+static void free_work(struct work_struct *w)
+{
+	struct vfree_deferred *p = container_of(w, struct vfree_deferred, wq);
+	struct llist_node *llnode = llist_del_all(&p->list);
+	while (llnode) {
+		void *p = llnode;
+		llnode = llist_next(llnode);
+		__vunmap(p, 1);
+	}
+}
+
 /*** Page table manipulation functions ***/
 
 static void vunmap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end)
@@ -249,19 +271,9 @@ 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 */
@@ -282,7 +294,7 @@ static struct vmap_area *__find_vmap_area(unsigned long addr)
 		va = rb_entry(n, struct vmap_area, rb_node);
 		if (addr < va->va_start)
 			n = n->rb_left;
-		else if (addr > va->va_start)
+		else if (addr >= va->va_end)
 			n = n->rb_right;
 		else
 			return va;
@@ -313,7 +325,7 @@ 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;
@@ -349,6 +361,12 @@ static struct vmap_area *alloc_vmap_area(unsigned long size,
 	if (unlikely(!va))
 		return ERR_PTR(-ENOMEM);
 
+	/*
+	 * Only scan the relevant parts containing pointers to other objects
+	 * to avoid false negatives.
+	 */
+	kmemleak_scan_area(&va->rb_node, SIZE_MAX, gfp_mask & GFP_RECLAIM_MASK);
+
 retry:
 	spin_lock(&vmap_area_lock);
 	/*
@@ -378,12 +396,12 @@ nocache:
 		addr = ALIGN(first->va_end, align);
 		if (addr < vstart)
 			goto nocache;
-		if (addr + size - 1 < addr)
+		if (addr + size < addr)
 			goto overflow;
 
 	} else {
 		addr = ALIGN(vstart, align);
-		if (addr + size - 1 < addr)
+		if (addr + size < addr)
 			goto overflow;
 
 		n = vmap_area_root.rb_node;
@@ -410,14 +428,14 @@ nocache:
 		if (addr + cached_hole_size < first->va_start)
 			cached_hole_size = first->va_start - addr;
 		addr = ALIGN(first->va_end, align);
-		if (addr + size - 1 < addr)
+		if (addr + size < addr)
 			goto overflow;
 
-		n = rb_next(&first->rb_node);
-		if (n)
-			first = rb_entry(n, struct vmap_area, rb_node);
-		else
+		if (list_is_last(&first->list, &vmap_area_list))
 			goto found;
+
+		first = list_entry(first->list.next,
+				struct vmap_area, list);
 	}
 
 found:
@@ -742,9 +760,7 @@ struct vmap_block_queue {
 struct vmap_block {
 	spinlock_t lock;
 	struct vmap_area *va;
-	struct vmap_block_queue *vbq;
 	unsigned long free, dirty;
-	DECLARE_BITMAP(alloc_map, VMAP_BBMAP_BITS);
 	DECLARE_BITMAP(dirty_map, VMAP_BBMAP_BITS);
 	struct list_head free_list;
 	struct rcu_head rcu_head;
@@ -810,7 +826,6 @@ static struct vmap_block *new_vmap_block(gfp_t gfp_mask)
 	vb->va = va;
 	vb->free = VMAP_BBMAP_BITS;
 	vb->dirty = 0;
-	bitmap_zero(vb->alloc_map, VMAP_BBMAP_BITS);
 	bitmap_zero(vb->dirty_map, VMAP_BBMAP_BITS);
 	INIT_LIST_HEAD(&vb->free_list);
 
@@ -822,7 +837,6 @@ static struct vmap_block *new_vmap_block(gfp_t gfp_mask)
 	radix_tree_preload_end();
 
 	vbq = &get_cpu_var(vmap_block_queue);
-	vb->vbq = vbq;
 	spin_lock(&vbq->lock);
 	list_add_rcu(&vb->free_list, &vbq->free);
 	spin_unlock(&vbq->lock);
@@ -863,7 +877,6 @@ static void purge_fragmented_blocks(int cpu)
 		if (vb->free + vb->dirty == VMAP_BBMAP_BITS && vb->dirty != VMAP_BBMAP_BITS) {
 			vb->free = 0; /* prevent further allocs after releasing lock */
 			vb->dirty = VMAP_BBMAP_BITS; /* prevent purging it again */
-			bitmap_fill(vb->alloc_map, VMAP_BBMAP_BITS);
 			bitmap_fill(vb->dirty_map, VMAP_BBMAP_BITS);
 			spin_lock(&vbq->lock);
 			list_del_rcu(&vb->free_list);
@@ -881,11 +894,6 @@ static void purge_fragmented_blocks(int cpu)
 	}
 }
 
-static void purge_fragmented_blocks_thiscpu(void)
-{
-	purge_fragmented_blocks(smp_processor_id());
-}
-
 static void purge_fragmented_blocks_allcpus(void)
 {
 	int cpu;
@@ -900,10 +908,17 @@ static void *vb_alloc(unsigned long size, gfp_t gfp_mask)
 	struct vmap_block *vb;
 	unsigned long addr = 0;
 	unsigned int order;
-	int purge = 0;
 
 	BUG_ON(size & ~PAGE_MASK);
 	BUG_ON(size > PAGE_SIZE*VMAP_MAX_ALLOC);
+	if (WARN_ON(size == 0)) {
+		/*
+		 * Allocating 0 bytes isn't what caller wants since
+		 * get_order(0) returns funny result. Just warn and terminate
+		 * early.
+		 */
+		return NULL;
+	}
 	order = get_order(size);
 
 again:
@@ -916,17 +931,7 @@ again:
 		if (vb->free < 1UL << order)
 			goto next;
 
-		i = bitmap_find_free_region(vb->alloc_map,
-						VMAP_BBMAP_BITS, order);
-
-		if (i < 0) {
-			if (vb->free + vb->dirty == VMAP_BBMAP_BITS) {
-				/* fragmented and no outstanding allocations */
-				BUG_ON(vb->dirty != VMAP_BBMAP_BITS);
-				purge = 1;
-			}
-			goto next;
-		}
+		i = VMAP_BBMAP_BITS - vb->free;
 		addr = vb->va->va_start + (i << PAGE_SHIFT);
 		BUG_ON(addr_to_vb_idx(addr) !=
 				addr_to_vb_idx(vb->va->va_start));
@@ -942,9 +947,6 @@ next:
 		spin_unlock(&vb->lock);
 	}
 
-	if (purge)
-		purge_fragmented_blocks_thiscpu();
-
 	put_cpu_var(vmap_block_queue);
 	rcu_read_unlock();
 
@@ -1022,15 +1024,16 @@ void vm_unmap_aliases(void)
 
 		rcu_read_lock();
 		list_for_each_entry_rcu(vb, &vbq->free, free_list) {
-			int i;
+			int i, j;
 
 			spin_lock(&vb->lock);
 			i = find_first_bit(vb->dirty_map, VMAP_BBMAP_BITS);
-			while (i < VMAP_BBMAP_BITS) {
+			if (i < VMAP_BBMAP_BITS) {
 				unsigned long s, e;
-				int j;
-				j = find_next_zero_bit(vb->dirty_map,
-					VMAP_BBMAP_BITS, i);
+
+				j = find_last_bit(vb->dirty_map,
+							VMAP_BBMAP_BITS);
+				j = j + 1; /* need exclusive index */
 
 				s = vb->va->va_start + (i << PAGE_SHIFT);
 				e = vb->va->va_start + (j << PAGE_SHIFT);
@@ -1040,10 +1043,6 @@ void vm_unmap_aliases(void)
 					start = s;
 				if (e > end)
 					end = e;
-
-				i = j;
-				i = find_next_bit(vb->dirty_map,
-							VMAP_BBMAP_BITS, i);
 			}
 			spin_unlock(&vb->lock);
 		}
@@ -1086,6 +1085,12 @@ EXPORT_SYMBOL(vm_unmap_ram);
  * @node: prefer to allocate data structures on this node
  * @prot: memory protection to use. PAGE_KERNEL for regular RAM
  *
+ * If you use this function for less than VMAP_MAX_ALLOC pages, it could be
+ * faster than vmap so it's good.  But if you mix long-life and short-life
+ * objects with vm_map_ram(), it could consume lots of address space through
+ * fragmentation (especially on a 32bit machine).  You could see failures in
+ * the end.  Please use this function for short-lived objects.
+ *
  * Returns: a pointer to the address that has been mapped, or %NULL on failure
  */
 void *vm_map_ram(struct page **pages, unsigned int count, int node, pgprot_t prot)
@@ -1117,6 +1122,7 @@ 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
@@ -1176,18 +1182,23 @@ void __init vmalloc_init(void)
 
 	for_each_possible_cpu(i) {
 		struct vmap_block_queue *vbq;
+		struct vfree_deferred *p;
 
 		vbq = &per_cpu(vmap_block_queue, i);
 		spin_lock_init(&vbq->lock);
 		INIT_LIST_HEAD(&vbq->free);
+		p = &per_cpu(vfree_deferred, i);
+		init_llist_head(&p->list);
+		INIT_WORK(&p->wq, free_work);
 	}
 
 	/* Import existing vmlist entries. */
 	for (tmp = vmlist; tmp; tmp = tmp->next) {
 		va = kzalloc(sizeof(struct vmap_area), GFP_NOWAIT);
-		va->flags = tmp->flags | VM_VM_AREA;
+		va->flags = VM_VM_AREA;
 		va->va_start = (unsigned long)tmp->addr;
 		va->va_end = va->va_start + tmp->size;
+		va->vm = tmp;
 		__insert_vmap_area(va);
 	}
 
@@ -1257,11 +1268,12 @@ void unmap_kernel_range(unsigned long addr, unsigned long size)
 	vunmap_page_range(addr, end);
 	flush_tlb_kernel_range(addr, end);
 }
+EXPORT_SYMBOL_GPL(unmap_kernel_range);
 
 int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages)
 {
 	unsigned long addr = (unsigned long)area->addr;
-	unsigned long end = addr + area->size - PAGE_SIZE;
+	unsigned long end = addr + get_vm_area_size(area);
 	int err;
 
 	err = vmap_page_range(addr, end, prot, *pages);
@@ -1274,61 +1286,40 @@ 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, void *caller)
+			      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)
-{
-	struct vm_struct *tmp, **p;
-
-	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, void *caller)
+static void clear_vm_uninitialized_flag(struct vm_struct *vm)
 {
-	setup_vmalloc_vm(vm, va, flags, caller);
-	insert_vmalloc_vmlist(vm);
+	/*
+	 * Before removing VM_UNINITIALIZED,
+	 * we should make sure that vm has proper values.
+	 * Pair with smp_rmb() in show_numa_info().
+	 */
+	smp_wmb();
+	vm->flags &= ~VM_UNINITIALIZED;
 }
 
 static struct vm_struct *__get_vm_area_node(unsigned long size,
 		unsigned long align, unsigned long flags, unsigned long start,
-		unsigned long end, int node, gfp_t gfp_mask, void *caller)
+		unsigned long end, int node, gfp_t gfp_mask, const void *caller)
 {
 	struct vmap_area *va;
 	struct vm_struct *area;
 
 	BUG_ON(in_interrupt());
-	if (flags & VM_IOREMAP) {
-		int bit = fls(size);
-
-		if (bit > IOREMAP_MAX_ORDER)
-			bit = IOREMAP_MAX_ORDER;
-		else if (bit < PAGE_SHIFT)
-			bit = PAGE_SHIFT;
-
-		align = 1ul << bit;
-	}
+	if (flags & VM_IOREMAP)
+		align = 1ul << clamp(fls(size), PAGE_SHIFT, IOREMAP_MAX_ORDER);
 
 	size = PAGE_ALIGN(size);
 	if (unlikely(!size))
@@ -1349,17 +1340,7 @@ static struct vm_struct *__get_vm_area_node(unsigned long size,
 		return NULL;
 	}
 
-	/*
-	 * 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.
-	 */
-	if (flags & VM_UNLIST)
-		setup_vmalloc_vm(area, va, flags, caller);
-	else
-		insert_vmalloc_vm(area, va, flags, caller);
+	setup_vmalloc_vm(area, va, flags, caller);
 
 	return area;
 }
@@ -1367,17 +1348,17 @@ static struct vm_struct *__get_vm_area_node(unsigned long size,
 struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags,
 				unsigned long start, unsigned long end)
 {
-	return __get_vm_area_node(size, 1, flags, start, end, -1, GFP_KERNEL,
-						__builtin_return_address(0));
+	return __get_vm_area_node(size, 1, flags, start, end, NUMA_NO_NODE,
+				  GFP_KERNEL, __builtin_return_address(0));
 }
 EXPORT_SYMBOL_GPL(__get_vm_area);
 
 struct vm_struct *__get_vm_area_caller(unsigned long size, unsigned long flags,
 				       unsigned long start, unsigned long end,
-				       void *caller)
+				       const void *caller)
 {
-	return __get_vm_area_node(size, 1, flags, start, end, -1, GFP_KERNEL,
-				  caller);
+	return __get_vm_area_node(size, 1, flags, start, end, NUMA_NO_NODE,
+				  GFP_KERNEL, caller);
 }
 
 /**
@@ -1392,17 +1373,26 @@ struct vm_struct *__get_vm_area_caller(unsigned long size, unsigned long flags,
 struct vm_struct *get_vm_area(unsigned long size, unsigned long flags)
 {
 	return __get_vm_area_node(size, 1, flags, VMALLOC_START, VMALLOC_END,
-				-1, GFP_KERNEL, __builtin_return_address(0));
+				  NUMA_NO_NODE, GFP_KERNEL,
+				  __builtin_return_address(0));
 }
 
 struct vm_struct *get_vm_area_caller(unsigned long size, unsigned long flags,
-				void *caller)
+				const void *caller)
 {
 	return __get_vm_area_node(size, 1, flags, VMALLOC_START, VMALLOC_END,
-						-1, GFP_KERNEL, caller);
+				  NUMA_NO_NODE, GFP_KERNEL, caller);
 }
 
-static struct vm_struct *find_vm_area(const void *addr)
+/**
+ *	find_vm_area  -  find a continuous kernel virtual area
+ *	@addr:		base address
+ *
+ *	Search for the kernel VM area starting at @addr, and return it.
+ *	It is up to the caller to do all required locking to keep the returned
+ *	pointer valid.
+ */
+struct vm_struct *find_vm_area(const void *addr)
 {
 	struct vmap_area *va;
 
@@ -1429,19 +1419,10 @@ 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);
@@ -1459,10 +1440,9 @@ static void __vunmap(const void *addr, int deallocate_pages)
 	if (!addr)
 		return;
 
-	if ((PAGE_SIZE-1) & (unsigned long)addr) {
-		WARN(1, KERN_ERR "Trying to vfree() bad address (%p)\n", addr);
+	if (WARN(!PAGE_ALIGNED(addr), "Trying to vfree() bad address (%p)\n",
+			addr))
 		return;
-	}
 
 	area = remove_vm_area(addr);
 	if (unlikely(!area)) {
@@ -1493,7 +1473,7 @@ static void __vunmap(const void *addr, int deallocate_pages)
 	kfree(area);
 	return;
 }
-
+ 
 /**
  *	vfree  -  release memory allocated by vmalloc()
  *	@addr:		memory base address
@@ -1502,15 +1482,26 @@ static void __vunmap(const void *addr, int deallocate_pages)
  *	obtained from vmalloc(), vmalloc_32() or __vmalloc(). If @addr is
  *	NULL, no operation is performed.
  *
- *	Must not be called in interrupt context.
+ *	Must not be called in NMI context (strictly speaking, only if we don't
+ *	have CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG, but making the calling
+ *	conventions for vfree() arch-depenedent would be a really bad idea)
+ *
+ *	NOTE: assumes that the object at *addr has a size >= sizeof(llist_node)
  */
 void vfree(const void *addr)
 {
-	BUG_ON(in_interrupt());
+	BUG_ON(in_nmi());
 
 	kmemleak_free(addr);
 
-	__vunmap(addr, 1);
+	if (!addr)
+		return;
+	if (unlikely(in_interrupt())) {
+		struct vfree_deferred *p = this_cpu_ptr(&vfree_deferred);
+		if (llist_add((struct llist_node *)addr, &p->list))
+			schedule_work(&p->wq);
+	} else
+		__vunmap(addr, 1);
 }
 EXPORT_SYMBOL(vfree);
 
@@ -1527,7 +1518,8 @@ void vunmap(const void *addr)
 {
 	BUG_ON(in_interrupt());
 	might_sleep();
-	__vunmap(addr, 0);
+	if (addr)
+		__vunmap(addr, 0);
 }
 EXPORT_SYMBOL(vunmap);
 
@@ -1567,29 +1559,28 @@ EXPORT_SYMBOL(vmap);
 
 static void *__vmalloc_node(unsigned long size, unsigned long align,
 			    gfp_t gfp_mask, pgprot_t prot,
-			    int node, void *caller);
+			    int node, const void *caller);
 static void *__vmalloc_area_node(struct vm_struct *area, gfp_t gfp_mask,
-				 pgprot_t prot, int node, void *caller)
+				 pgprot_t prot, int node)
 {
 	const int order = 0;
 	struct page **pages;
 	unsigned int nr_pages, array_size, i;
 	gfp_t nested_gfp = (gfp_mask & GFP_RECLAIM_MASK) | __GFP_ZERO;
 
-	nr_pages = (area->size - PAGE_SIZE) >> PAGE_SHIFT;
+	nr_pages = get_vm_area_size(area) >> PAGE_SHIFT;
 	array_size = (nr_pages * sizeof(struct page *));
 
 	area->nr_pages = nr_pages;
 	/* Please note that the recursion is strictly bounded. */
 	if (array_size > PAGE_SIZE) {
 		pages = __vmalloc_node(array_size, 1, nested_gfp|__GFP_HIGHMEM,
-				PAGE_KERNEL, node, caller);
+				PAGE_KERNEL, node, area->caller);
 		area->flags |= VM_VPAGES;
 	} else {
 		pages = kmalloc_node(array_size, nested_gfp, node);
 	}
 	area->pages = pages;
-	area->caller = caller;
 	if (!area->pages) {
 		remove_vm_area(area->addr);
 		kfree(area);
@@ -1600,7 +1591,7 @@ static void *__vmalloc_area_node(struct vm_struct *area, gfp_t gfp_mask,
 		struct page *page;
 		gfp_t tmp_mask = gfp_mask | __GFP_NOWARN;
 
-		if (node < 0)
+		if (node == NUMA_NO_NODE)
 			page = alloc_page(tmp_mask);
 		else
 			page = alloc_pages_node(node, tmp_mask, order);
@@ -1633,7 +1624,7 @@ fail:
  *	@end:		vm area range end
  *	@gfp_mask:	flags for the page level allocator
  *	@prot:		protection mask for the allocated pages
- *	@node:		node to use for allocation or -1
+ *	@node:		node to use for allocation or NUMA_NO_NODE
  *	@caller:	caller's return address
  *
  *	Allocate enough pages to cover @size from the page level
@@ -1642,7 +1633,7 @@ fail:
  */
 void *__vmalloc_node_range(unsigned long size, unsigned long align,
 			unsigned long start, unsigned long end, gfp_t gfp_mask,
-			pgprot_t prot, int node, void *caller)
+			pgprot_t prot, int node, const void *caller)
 {
 	struct vm_struct *area;
 	void *addr;
@@ -1652,27 +1643,28 @@ void *__vmalloc_node_range(unsigned long size, unsigned long align,
 	if (!size || (size >> PAGE_SHIFT) > totalram_pages)
 		goto fail;
 
-	area = __get_vm_area_node(size, align, VM_ALLOC | VM_UNLIST,
+	area = __get_vm_area_node(size, align, VM_ALLOC | VM_UNINITIALIZED,
 				  start, end, node, gfp_mask, caller);
 	if (!area)
 		goto fail;
 
-	addr = __vmalloc_area_node(area, gfp_mask, prot, node, caller);
+	addr = __vmalloc_area_node(area, gfp_mask, prot, node);
 	if (!addr)
 		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_UNINITIALIZED
+	 * 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_uninitialized_flag(area);
 
 	/*
-	 * A ref_count = 3 is needed because the vm_struct and vmap_area
-	 * structures allocated in the __get_vm_area_node() function contain
-	 * references to the virtual address of the vmalloc'ed block.
+	 * A ref_count = 2 is needed because vm_struct allocated in
+	 * __get_vm_area_node() contains a reference to the virtual address of
+	 * the vmalloc'ed block.
 	 */
-	kmemleak_alloc(addr, real_size, 3, gfp_mask);
+	kmemleak_alloc(addr, real_size, 2, gfp_mask);
 
 	return addr;
 
@@ -1689,7 +1681,7 @@ fail:
  *	@align:		desired alignment
  *	@gfp_mask:	flags for the page level allocator
  *	@prot:		protection mask for the allocated pages
- *	@node:		node to use for allocation or -1
+ *	@node:		node to use for allocation or NUMA_NO_NODE
  *	@caller:	caller's return address
  *
  *	Allocate enough pages to cover @size from the page level
@@ -1698,7 +1690,7 @@ fail:
  */
 static void *__vmalloc_node(unsigned long size, unsigned long align,
 			    gfp_t gfp_mask, pgprot_t prot,
-			    int node, void *caller)
+			    int node, const void *caller)
 {
 	return __vmalloc_node_range(size, align, VMALLOC_START, VMALLOC_END,
 				gfp_mask, prot, node, caller);
@@ -1706,7 +1698,7 @@ static void *__vmalloc_node(unsigned long size, unsigned long align,
 
 void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
 {
-	return __vmalloc_node(size, 1, gfp_mask, prot, -1,
+	return __vmalloc_node(size, 1, gfp_mask, prot, NUMA_NO_NODE,
 				__builtin_return_address(0));
 }
 EXPORT_SYMBOL(__vmalloc);
@@ -1729,7 +1721,8 @@ static inline void *__vmalloc_node_flags(unsigned long size,
  */
 void *vmalloc(unsigned long size)
 {
-	return __vmalloc_node_flags(size, -1, GFP_KERNEL | __GFP_HIGHMEM);
+	return __vmalloc_node_flags(size, NUMA_NO_NODE,
+				    GFP_KERNEL | __GFP_HIGHMEM);
 }
 EXPORT_SYMBOL(vmalloc);
 
@@ -1745,7 +1738,7 @@ EXPORT_SYMBOL(vmalloc);
  */
 void *vzalloc(unsigned long size)
 {
-	return __vmalloc_node_flags(size, -1,
+	return __vmalloc_node_flags(size, NUMA_NO_NODE,
 				GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO);
 }
 EXPORT_SYMBOL(vzalloc);
@@ -1764,7 +1757,8 @@ void *vmalloc_user(unsigned long size)
 
 	ret = __vmalloc_node(size, SHMLBA,
 			     GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO,
-			     PAGE_KERNEL, -1, __builtin_return_address(0));
+			     PAGE_KERNEL, NUMA_NO_NODE,
+			     __builtin_return_address(0));
 	if (ret) {
 		area = find_vm_area(ret);
 		area->flags |= VM_USERMAP;
@@ -1829,7 +1823,7 @@ EXPORT_SYMBOL(vzalloc_node);
 void *vmalloc_exec(unsigned long size)
 {
 	return __vmalloc_node(size, 1, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC,
-			      -1, __builtin_return_address(0));
+			      NUMA_NO_NODE, __builtin_return_address(0));
 }
 
 #if defined(CONFIG_64BIT) && defined(CONFIG_ZONE_DMA32)
@@ -1850,7 +1844,7 @@ void *vmalloc_exec(unsigned long size)
 void *vmalloc_32(unsigned long size)
 {
 	return __vmalloc_node(size, 1, GFP_VMALLOC32, PAGE_KERNEL,
-			      -1, __builtin_return_address(0));
+			      NUMA_NO_NODE, __builtin_return_address(0));
 }
 EXPORT_SYMBOL(vmalloc_32);
 
@@ -1867,7 +1861,7 @@ void *vmalloc_32_user(unsigned long size)
 	void *ret;
 
 	ret = __vmalloc_node(size, 1, GFP_VMALLOC32 | __GFP_ZERO, PAGE_KERNEL,
-			     -1, __builtin_return_address(0));
+			     NUMA_NO_NODE, __builtin_return_address(0));
 	if (ret) {
 		area = find_vm_area(ret);
 		area->flags |= VM_USERMAP;
@@ -1906,9 +1900,9 @@ static int aligned_vread(char *buf, char *addr, unsigned long count)
 			 * we can expect USER0 is not used (see vread/vwrite's
 			 * function description)
 			 */
-			void *map = kmap_atomic(p, KM_USER0);
+			void *map = kmap_atomic(p);
 			memcpy(buf, map + offset, length);
-			kunmap_atomic(map, KM_USER0);
+			kunmap_atomic(map);
 		} else
 			memset(buf, 0, length);
 
@@ -1945,9 +1939,9 @@ static int aligned_vwrite(char *buf, char *addr, unsigned long count)
 			 * we can expect USER0 is not used (see vread/vwrite's
 			 * function description)
 			 */
-			void *map = kmap_atomic(p, KM_USER0);
+			void *map = kmap_atomic(p);
 			memcpy(map + offset, buf, length);
-			kunmap_atomic(map, KM_USER0);
+			kunmap_atomic(map);
 		}
 		addr += length;
 		buf += length;
@@ -1974,9 +1968,7 @@ static int aligned_vwrite(char *buf, char *addr, unsigned long count)
  *	IOREMAP area is treated as memory hole and no copy is done.
  *
  *	If [addr...addr+count) doesn't includes any intersects with alive
- *	vm_struct area, returns 0.
- *	@buf should be kernel's buffer. Because	this function uses KM_USER0,
- *	the caller should guarantee KM_USER0 is not used.
+ *	vm_struct area, returns 0. @buf should be kernel's buffer.
  *
  *	Note: In usual ops, vread() is never necessary because the caller
  *	should know vmalloc() area is valid and can use memcpy().
@@ -1987,7 +1979,8 @@ 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;
@@ -1996,10 +1989,17 @@ 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 + get_vm_area_size(vm))
 			continue;
 		while (addr < vaddr) {
 			if (count == 0)
@@ -2009,10 +2009,10 @@ long vread(char *buf, char *addr, unsigned long count)
 			addr++;
 			count--;
 		}
-		n = vaddr + tmp->size - PAGE_SIZE - addr;
+		n = vaddr + get_vm_area_size(vm) - 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);
@@ -2021,7 +2021,7 @@ 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;
@@ -2050,9 +2050,7 @@ finished:
  *	IOREMAP area is treated as memory hole and no copy is done.
  *
  *	If [addr...addr+count) doesn't includes any intersects with alive
- *	vm_struct area, returns 0.
- *	@buf should be kernel's buffer. Because	this function uses KM_USER0,
- *	the caller should guarantee KM_USER0 is not used.
+ *	vm_struct area, returns 0. @buf should be kernel's buffer.
  *
  *	Note: In usual ops, vwrite() is never necessary because the caller
  *	should know vmalloc() area is valid and can use memcpy().
@@ -2062,7 +2060,8 @@ 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;
@@ -2072,10 +2071,17 @@ 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 + get_vm_area_size(vm))
 			continue;
 		while (addr < vaddr) {
 			if (count == 0)
@@ -2084,10 +2090,10 @@ long vwrite(char *buf, char *addr, unsigned long count)
 			addr++;
 			count--;
 		}
-		n = vaddr + tmp->size - PAGE_SIZE - addr;
+		n = vaddr + get_vm_area_size(vm) - addr;
 		if (n > count)
 			n = count;
-		if (!(tmp->flags & VM_IOREMAP)) {
+		if (!(vm->flags & VM_IOREMAP)) {
 			aligned_vwrite(buf, addr, n);
 			copied++;
 		}
@@ -2096,49 +2102,50 @@ 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;
 }
 
 /**
- *	remap_vmalloc_range  -  map vmalloc pages to userspace
- *	@vma:		vma to cover (map full range of vma)
- *	@addr:		vmalloc memory
- *	@pgoff:		number of pages into addr before first page to map
+ *	remap_vmalloc_range_partial  -  map vmalloc pages to userspace
+ *	@vma:		vma to cover
+ *	@uaddr:		target user address to start at
+ *	@kaddr:		virtual address of vmalloc kernel memory
+ *	@size:		size of map area
  *
  *	Returns:	0 for success, -Exxx on failure
  *
- *	This function checks that addr is a valid vmalloc'ed area, and
- *	that it is big enough to cover the vma. Will return failure if
- *	that criteria isn't met.
+ *	This function checks that @kaddr is a valid vmalloc'ed area,
+ *	and that it is big enough to cover the range starting at
+ *	@uaddr in @vma. Will return failure if that criteria isn't
+ *	met.
  *
  *	Similar to remap_pfn_range() (see mm/memory.c)
  */
-int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
-						unsigned long pgoff)
+int remap_vmalloc_range_partial(struct vm_area_struct *vma, unsigned long uaddr,
+				void *kaddr, unsigned long size)
 {
 	struct vm_struct *area;
-	unsigned long uaddr = vma->vm_start;
-	unsigned long usize = vma->vm_end - vma->vm_start;
 
-	if ((PAGE_SIZE-1) & (unsigned long)addr)
+	size = PAGE_ALIGN(size);
+
+	if (!PAGE_ALIGNED(uaddr) || !PAGE_ALIGNED(kaddr))
 		return -EINVAL;
 
-	area = find_vm_area(addr);
+	area = find_vm_area(kaddr);
 	if (!area)
 		return -EINVAL;
 
 	if (!(area->flags & VM_USERMAP))
 		return -EINVAL;
 
-	if (usize + (pgoff << PAGE_SHIFT) > area->size - PAGE_SIZE)
+	if (kaddr + size > area->addr + area->size)
 		return -EINVAL;
 
-	addr += pgoff << PAGE_SHIFT;
 	do {
-		struct page *page = vmalloc_to_page(addr);
+		struct page *page = vmalloc_to_page(kaddr);
 		int ret;
 
 		ret = vm_insert_page(vma, uaddr, page);
@@ -2146,22 +2153,44 @@ int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
 			return ret;
 
 		uaddr += PAGE_SIZE;
-		addr += PAGE_SIZE;
-		usize -= PAGE_SIZE;
-	} while (usize > 0);
+		kaddr += PAGE_SIZE;
+		size -= PAGE_SIZE;
+	} while (size > 0);
 
-	/* Prevent "things" like memory migration? VM_flags need a cleanup... */
-	vma->vm_flags |= VM_RESERVED;
+	vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP;
 
 	return 0;
 }
+EXPORT_SYMBOL(remap_vmalloc_range_partial);
+
+/**
+ *	remap_vmalloc_range  -  map vmalloc pages to userspace
+ *	@vma:		vma to cover (map full range of vma)
+ *	@addr:		vmalloc memory
+ *	@pgoff:		number of pages into addr before first page to map
+ *
+ *	Returns:	0 for success, -Exxx on failure
+ *
+ *	This function checks that addr is a valid vmalloc'ed area, and
+ *	that it is big enough to cover the vma. Will return failure if
+ *	that criteria isn't met.
+ *
+ *	Similar to remap_pfn_range() (see mm/memory.c)
+ */
+int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
+						unsigned long pgoff)
+{
+	return remap_vmalloc_range_partial(vma, vma->vm_start,
+					   addr + (pgoff << PAGE_SHIFT),
+					   vma->vm_end - vma->vm_start);
+}
 EXPORT_SYMBOL(remap_vmalloc_range);
 
 /*
  * Implement a stub for vmalloc_sync_all() if the architecture chose not to
  * have one.
  */
-void  __attribute__((weak)) vmalloc_sync_all(void)
+void __weak vmalloc_sync_all(void)
 {
 }
 
@@ -2375,8 +2404,8 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 		return NULL;
 	}
 
-	vms = kzalloc(sizeof(vms[0]) * nr_vms, GFP_KERNEL);
-	vas = kzalloc(sizeof(vas[0]) * nr_vms, GFP_KERNEL);
+	vms = kcalloc(nr_vms, sizeof(vms[0]), GFP_KERNEL);
+	vas = kcalloc(nr_vms, sizeof(vas[0]), GFP_KERNEL);
 	if (!vas || !vms)
 		goto err_free2;
 
@@ -2468,8 +2497,8 @@ found:
 
 	/* insert all vm's */
 	for (area = 0; area < nr_vms; area++)
-		insert_vmalloc_vm(vms[area], vas[area], VM_ALLOC,
-				  pcpu_get_vm_areas);
+		setup_vmalloc_vm(vms[area], vas[area], VM_ALLOC,
+				 pcpu_get_vm_areas);
 
 	kfree(vas);
 	return vms;
@@ -2504,19 +2533,19 @@ 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;
 
@@ -2524,26 +2553,35 @@ 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)
 {
-	if (NUMA_BUILD) {
+	if (IS_ENABLED(CONFIG_NUMA)) {
 		unsigned int nr, *counters = m->private;
 
 		if (!counters)
 			return;
 
+		/* Pair with smp_wmb() in clear_vm_uninitialized_flag() */
+		smp_rmb();
+		if (v->flags & VM_UNINITIALIZED)
+			return;
+
 		memset(counters, 0, nr_node_ids * sizeof(unsigned int));
 
 		for (nr = 0; nr < v->nr_pages; nr++)
@@ -2557,9 +2595,19 @@ 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;
+
+	/*
+	 * s_show can encounter race with remove_vm_area, !VM_VM_AREA on
+	 * behalf of vmap area is being tear down or vm_map_ram allocation.
+	 */
+	if (!(va->flags & VM_VM_AREA))
+		return 0;
 
-	seq_printf(m, "0x%p-0x%p %7ld",
+	v = va->vm;
+
+	seq_printf(m, "0x%pK-0x%pK %7ld",
 		v->addr, v->addr + v->size, v->size);
 
 	if (v->caller)
@@ -2572,19 +2620,19 @@ static int s_show(struct seq_file *m, void *p)
 		seq_printf(m, " phys=%llx", (unsigned long long)v->phys_addr);
 
 	if (v->flags & VM_IOREMAP)
-		seq_printf(m, " ioremap");
+		seq_puts(m, " ioremap");
 
 	if (v->flags & VM_ALLOC)
-		seq_printf(m, " vmalloc");
+		seq_puts(m, " vmalloc");
 
 	if (v->flags & VM_MAP)
-		seq_printf(m, " vmap");
+		seq_puts(m, " vmap");
 
 	if (v->flags & VM_USERMAP)
-		seq_printf(m, " user");
+		seq_puts(m, " user");
 
 	if (v->flags & VM_VPAGES)
-		seq_printf(m, " vpages");
+		seq_puts(m, " vpages");
 
 	show_numa_info(m, v);
 	seq_putc(m, '\n');
@@ -2603,7 +2651,7 @@ static int vmalloc_open(struct inode *inode, struct file *file)
 	unsigned int *ptr = NULL;
 	int ret;
 
-	if (NUMA_BUILD) {
+	if (IS_ENABLED(CONFIG_NUMA)) {
 		ptr = kmalloc(nr_node_ids * sizeof(unsigned int), GFP_KERNEL);
 		if (ptr == NULL)
 			return -ENOMEM;
@@ -2630,5 +2678,53 @@ 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