1 files changed, 213 insertions, 119 deletions

diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 9f909622a25e..5d6030235d7a 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -31,8 +31,6 @@
 #include <asm/tlbflush.h>
 #include <asm/shmparam.h>
 
-bool vmap_lazy_unmap __read_mostly = true;
-
 /*** Page table manipulation functions ***/
 
 static void vunmap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end)
@@ -263,8 +261,15 @@ struct vmap_area {
 };
 
 static DEFINE_SPINLOCK(vmap_area_lock);
-static struct rb_root vmap_area_root = RB_ROOT;
 static LIST_HEAD(vmap_area_list);
+static struct rb_root vmap_area_root = RB_ROOT;
+
+/* The vmap cache globals are protected by vmap_area_lock */
+static struct rb_node *free_vmap_cache;
+static unsigned long cached_hole_size;
+static unsigned long cached_vstart;
+static unsigned long cached_align;
+
 static unsigned long vmap_area_pcpu_hole;
 
 static struct vmap_area *__find_vmap_area(unsigned long addr)
@@ -293,13 +298,13 @@ static void __insert_vmap_area(struct vmap_area *va)
 	struct rb_node *tmp;
 
 	while (*p) {
-		struct vmap_area *tmp;
+		struct vmap_area *tmp_va;
 
 		parent = *p;
-		tmp = rb_entry(parent, struct vmap_area, rb_node);
-		if (va->va_start < tmp->va_end)
+		tmp_va = rb_entry(parent, struct vmap_area, rb_node);
+		if (va->va_start < tmp_va->va_end)
 			p = &(*p)->rb_left;
-		else if (va->va_end > tmp->va_start)
+		else if (va->va_end > tmp_va->va_start)
 			p = &(*p)->rb_right;
 		else
 			BUG();
@@ -333,9 +338,11 @@ static struct vmap_area *alloc_vmap_area(unsigned long size,
 	struct rb_node *n;
 	unsigned long addr;
 	int purged = 0;
+	struct vmap_area *first;
 
 	BUG_ON(!size);
 	BUG_ON(size & ~PAGE_MASK);
+	BUG_ON(!is_power_of_2(align));
 
 	va = kmalloc_node(sizeof(struct vmap_area),
 			gfp_mask & GFP_RECLAIM_MASK, node);
@@ -343,79 +350,106 @@ static struct vmap_area *alloc_vmap_area(unsigned long size,
 		return ERR_PTR(-ENOMEM);
 
 retry:
-	addr = ALIGN(vstart, align);
-
 	spin_lock(&vmap_area_lock);
-	if (addr + size - 1 < addr)
-		goto overflow;
+	/*
+	 * Invalidate cache if we have more permissive parameters.
+	 * cached_hole_size notes the largest hole noticed _below_
+	 * the vmap_area cached in free_vmap_cache: if size fits
+	 * into that hole, we want to scan from vstart to reuse
+	 * the hole instead of allocating above free_vmap_cache.
+	 * Note that __free_vmap_area may update free_vmap_cache
+	 * without updating cached_hole_size or cached_align.
+	 */
+	if (!free_vmap_cache ||
+			size < cached_hole_size ||
+			vstart < cached_vstart ||
+			align < cached_align) {
+nocache:
+		cached_hole_size = 0;
+		free_vmap_cache = NULL;
+	}
+	/* record if we encounter less permissive parameters */
+	cached_vstart = vstart;
+	cached_align = align;
+
+	/* find starting point for our search */
+	if (free_vmap_cache) {
+		first = rb_entry(free_vmap_cache, struct vmap_area, rb_node);
+		addr = ALIGN(first->va_end + PAGE_SIZE, align);
+		if (addr < vstart)
+			goto nocache;
+		if (addr + size - 1 < addr)
+			goto overflow;
+
+	} else {
+		addr = ALIGN(vstart, align);
+		if (addr + size - 1 < addr)
+			goto overflow;
 
-	/* XXX: could have a last_hole cache */
-	n = vmap_area_root.rb_node;
-	if (n) {
-		struct vmap_area *first = NULL;
+		n = vmap_area_root.rb_node;
+		first = NULL;
 
-		do {
+		while (n) {
 			struct vmap_area *tmp;
 			tmp = rb_entry(n, struct vmap_area, rb_node);
 			if (tmp->va_end >= addr) {
-				if (!first && tmp->va_start < addr + size)
-					first = tmp;
-				n = n->rb_left;
-			} else {
 				first = tmp;
+				if (tmp->va_start <= addr)
+					break;
+				n = n->rb_left;
+			} else
 				n = n->rb_right;
-			}
-		} while (n);
+		}
 
 		if (!first)
 			goto found;
-
-		if (first->va_end < addr) {
-			n = rb_next(&first->rb_node);
-			if (n)
-				first = rb_entry(n, struct vmap_area, rb_node);
-			else
-				goto found;
-		}
-
-		while (addr + size > first->va_start && addr + size <= vend) {
-			addr = ALIGN(first->va_end + PAGE_SIZE, align);
-			if (addr + size - 1 < addr)
-				goto overflow;
-
-			n = rb_next(&first->rb_node);
-			if (n)
-				first = rb_entry(n, struct vmap_area, rb_node);
-			else
-				goto found;
-		}
 	}
-found:
-	if (addr + size > vend) {
-overflow:
-		spin_unlock(&vmap_area_lock);
-		if (!purged) {
-			purge_vmap_area_lazy();
-			purged = 1;
-			goto retry;
-		}
-		if (printk_ratelimit())
-			printk(KERN_WARNING
-				"vmap allocation for size %lu failed: "
-				"use vmalloc=<size> to increase size.\n", size);
-		kfree(va);
-		return ERR_PTR(-EBUSY);
+
+	/* from the starting point, walk areas until a suitable hole is found */
+	while (addr + size >= first->va_start && addr + size <= vend) {
+		if (addr + cached_hole_size < first->va_start)
+			cached_hole_size = first->va_start - addr;
+		addr = ALIGN(first->va_end + PAGE_SIZE, align);
+		if (addr + size - 1 < addr)
+			goto overflow;
+
+		n = rb_next(&first->rb_node);
+		if (n)
+			first = rb_entry(n, struct vmap_area, rb_node);
+		else
+			goto found;
 	}
 
-	BUG_ON(addr & (align-1));
+found:
+	if (addr + size > vend)
+		goto overflow;
 
 	va->va_start = addr;
 	va->va_end = addr + size;
 	va->flags = 0;
 	__insert_vmap_area(va);
+	free_vmap_cache = &va->rb_node;
 	spin_unlock(&vmap_area_lock);
 
+	BUG_ON(va->va_start & (align-1));
+	BUG_ON(va->va_start < vstart);
+	BUG_ON(va->va_end > vend);
+
 	return va;
+
+overflow:
+	spin_unlock(&vmap_area_lock);
+	if (!purged) {
+		purge_vmap_area_lazy();
+		purged = 1;
+		goto retry;
+	}
+	if (printk_ratelimit())
+		printk(KERN_WARNING
+			"vmap allocation for size %lu failed: "
+			"use vmalloc=<size> to increase size.\n", size);
+	kfree(va);
+	return ERR_PTR(-EBUSY);
 }
 
 static void rcu_free_va(struct rcu_head *head)
@@ -428,6 +462,22 @@ static void rcu_free_va(struct rcu_head *head)
 static void __free_vmap_area(struct vmap_area *va)
 {
 	BUG_ON(RB_EMPTY_NODE(&va->rb_node));
+
+	if (free_vmap_cache) {
+		if (va->va_end < cached_vstart) {
+			free_vmap_cache = NULL;
+		} else {
+			struct vmap_area *cache;
+			cache = rb_entry(free_vmap_cache, struct vmap_area, rb_node);
+			if (va->va_start <= cache->va_start) {
+				free_vmap_cache = rb_prev(&va->rb_node);
+				/*
+				 * We don't try to update cached_hole_size or
+				 * cached_align, but it won't go very wrong.
+				 */
+			}
+		}
+	}
 	rb_erase(&va->rb_node, &vmap_area_root);
 	RB_CLEAR_NODE(&va->rb_node);
 	list_del_rcu(&va->list);
@@ -503,9 +553,6 @@ static unsigned long lazy_max_pages(void)
 {
 	unsigned int log;
 
-	if (!vmap_lazy_unmap)
-		return 0;
-
 	log = fls(num_online_cpus());
 
 	return log * (32UL * 1024 * 1024 / PAGE_SIZE);
@@ -566,7 +613,6 @@ static void __purge_vmap_area_lazy(unsigned long *start, unsigned long *end,
 			if (va->va_end > *end)
 				*end = va->va_end;
 			nr += (va->va_end - va->va_start) >> PAGE_SHIFT;
-			unmap_vmap_area(va);
 			list_add_tail(&va->purge_list, &valist);
 			va->flags |= VM_LAZY_FREEING;
 			va->flags &= ~VM_LAZY_FREE;
@@ -611,10 +657,11 @@ static void purge_vmap_area_lazy(void)
 }
 
 /*
- * Free and unmap a vmap area, caller ensuring flush_cache_vunmap had been
- * called for the correct range previously.
+ * Free a vmap area, caller ensuring that the area has been unmapped
+ * and flush_cache_vunmap had been called for the correct range
+ * previously.
  */
-static void free_unmap_vmap_area_noflush(struct vmap_area *va)
+static void free_vmap_area_noflush(struct vmap_area *va)
 {
 	va->flags |= VM_LAZY_FREE;
 	atomic_add((va->va_end - va->va_start) >> PAGE_SHIFT, &vmap_lazy_nr);
@@ -623,6 +670,16 @@ static void free_unmap_vmap_area_noflush(struct vmap_area *va)
 }
 
 /*
+ * Free and unmap a vmap area, caller ensuring flush_cache_vunmap had been
+ * called for the correct range previously.
+ */
+static void free_unmap_vmap_area_noflush(struct vmap_area *va)
+{
+	unmap_vmap_area(va);
+	free_vmap_area_noflush(va);
+}
+
+/*
  * Free and unmap a vmap area
  */
 static void free_unmap_vmap_area(struct vmap_area *va)
@@ -743,7 +800,7 @@ static struct vmap_block *new_vmap_block(gfp_t gfp_mask)
 	va = alloc_vmap_area(VMAP_BLOCK_SIZE, VMAP_BLOCK_SIZE,
 					VMALLOC_START, VMALLOC_END,
 					node, gfp_mask);
-	if (unlikely(IS_ERR(va))) {
+	if (IS_ERR(va)) {
 		kfree(vb);
 		return ERR_CAST(va);
 	}
@@ -798,7 +855,7 @@ static void free_vmap_block(struct vmap_block *vb)
 	spin_unlock(&vmap_block_tree_lock);
 	BUG_ON(tmp != vb);
 
-	free_unmap_vmap_area_noflush(vb->va);
+	free_vmap_area_noflush(vb->va);
 	call_rcu(&vb->rcu_head, rcu_free_vb);
 }
 
@@ -936,6 +993,8 @@ static void vb_free(const void *addr, unsigned long size)
 	rcu_read_unlock();
 	BUG_ON(!vb);
 
+	vunmap_page_range((unsigned long)addr, (unsigned long)addr + size);
+
 	spin_lock(&vb->lock);
 	BUG_ON(bitmap_allocate_region(vb->dirty_map, offset >> PAGE_SHIFT, order));
 
@@ -988,7 +1047,6 @@ void vm_unmap_aliases(void)
 
 				s = vb->va->va_start + (i << PAGE_SHIFT);
 				e = vb->va->va_start + (j << PAGE_SHIFT);
-				vunmap_page_range(s, e);
 				flush = 1;
 
 				if (s < start)
@@ -1169,6 +1227,7 @@ void unmap_kernel_range_noflush(unsigned long addr, unsigned long size)
 {
 	vunmap_page_range(addr, addr + size);
 }
+EXPORT_SYMBOL_GPL(unmap_kernel_range_noflush);
 
 /**
  * unmap_kernel_range - unmap kernel VM area and flush cache and TLB
@@ -1309,13 +1368,6 @@ struct vm_struct *get_vm_area_caller(unsigned long size, unsigned long flags,
 						-1, GFP_KERNEL, caller);
 }
 
-struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags,
-				   int node, gfp_t gfp_mask)
-{
-	return __get_vm_area_node(size, 1, flags, VMALLOC_START, VMALLOC_END,
-				  node, gfp_mask, __builtin_return_address(0));
-}
-
 static struct vm_struct *find_vm_area(const void *addr)
 {
 	struct vmap_area *va;
@@ -1531,25 +1583,12 @@ fail:
 	return NULL;
 }
 
-void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot)
-{
-	void *addr = __vmalloc_area_node(area, gfp_mask, prot, -1,
-					 __builtin_return_address(0));
-
-	/*
-	 * 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.
-	 */
-	kmemleak_alloc(addr, area->size - PAGE_SIZE, 3, gfp_mask);
-
-	return addr;
-}
-
 /**
- *	__vmalloc_node  -  allocate virtually contiguous memory
+ *	__vmalloc_node_range  -  allocate virtually contiguous memory
  *	@size:		allocation size
  *	@align:		desired alignment
+ *	@start:		vm area range start
+ *	@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
@@ -1559,9 +1598,9 @@ void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot)
  *	allocator with @gfp_mask flags.  Map them into contiguous
  *	kernel virtual space, using a pagetable protection of @prot.
  */
-static void *__vmalloc_node(unsigned long size, unsigned long align,
-			    gfp_t gfp_mask, pgprot_t prot,
-			    int node, void *caller)
+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)
 {
 	struct vm_struct *area;
 	void *addr;
@@ -1571,8 +1610,8 @@ static void *__vmalloc_node(unsigned long size, unsigned long align,
 	if (!size || (size >> PAGE_SHIFT) > totalram_pages)
 		return NULL;
 
-	area = __get_vm_area_node(size, align, VM_ALLOC, VMALLOC_START,
-				  VMALLOC_END, node, gfp_mask, caller);
+	area = __get_vm_area_node(size, align, VM_ALLOC, start, end, node,
+				  gfp_mask, caller);
 
 	if (!area)
 		return NULL;
@@ -1589,6 +1628,27 @@ static void *__vmalloc_node(unsigned long size, unsigned long align,
 	return addr;
 }
 
+/**
+ *	__vmalloc_node  -  allocate virtually contiguous memory
+ *	@size:		allocation size
+ *	@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
+ *	@caller:	caller's return address
+ *
+ *	Allocate enough pages to cover @size from the page level
+ *	allocator with @gfp_mask flags.  Map them into contiguous
+ *	kernel virtual space, using a pagetable protection of @prot.
+ */
+static void *__vmalloc_node(unsigned long size, unsigned long align,
+			    gfp_t gfp_mask, pgprot_t prot,
+			    int node, void *caller)
+{
+	return __vmalloc_node_range(size, align, VMALLOC_START, VMALLOC_END,
+				gfp_mask, prot, node, caller);
+}
+
 void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
 {
 	return __vmalloc_node(size, 1, gfp_mask, prot, -1,
@@ -1596,6 +1656,13 @@ void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
 }
 EXPORT_SYMBOL(__vmalloc);
 
+static inline void *__vmalloc_node_flags(unsigned long size,
+					int node, gfp_t flags)
+{
+	return __vmalloc_node(size, 1, flags, PAGE_KERNEL,
+					node, __builtin_return_address(0));
+}
+
 /**
  *	vmalloc  -  allocate virtually contiguous memory
  *	@size:		allocation size
@@ -1607,12 +1674,28 @@ EXPORT_SYMBOL(__vmalloc);
  */
 void *vmalloc(unsigned long size)
 {
-	return __vmalloc_node(size, 1, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL,
-					-1, __builtin_return_address(0));
+	return __vmalloc_node_flags(size, -1, GFP_KERNEL | __GFP_HIGHMEM);
 }
 EXPORT_SYMBOL(vmalloc);
 
 /**
+ *	vzalloc - allocate virtually contiguous memory with zero fill
+ *	@size:	allocation size
+ *	Allocate enough pages to cover @size from the page level
+ *	allocator and map them into contiguous kernel virtual space.
+ *	The memory allocated is set to zero.
+ *
+ *	For tight control over page level allocator and protection flags
+ *	use __vmalloc() instead.
+ */
+void *vzalloc(unsigned long size)
+{
+	return __vmalloc_node_flags(size, -1,
+				GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO);
+}
+EXPORT_SYMBOL(vzalloc);
+
+/**
  * vmalloc_user - allocate zeroed virtually contiguous memory for userspace
  * @size: allocation size
  *
@@ -1653,6 +1736,25 @@ void *vmalloc_node(unsigned long size, int node)
 }
 EXPORT_SYMBOL(vmalloc_node);
 
+/**
+ * vzalloc_node - allocate memory on a specific node with zero fill
+ * @size:	allocation size
+ * @node:	numa node
+ *
+ * Allocate enough pages to cover @size from the page level
+ * allocator and map them into contiguous kernel virtual space.
+ * The memory allocated is set to zero.
+ *
+ * For tight control over page level allocator and protection flags
+ * use __vmalloc_node() instead.
+ */
+void *vzalloc_node(unsigned long size, int node)
+{
+	return __vmalloc_node_flags(size, node,
+			 GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO);
+}
+EXPORT_SYMBOL(vzalloc_node);
+
 #ifndef PAGE_KERNEL_EXEC
 # define PAGE_KERNEL_EXEC PAGE_KERNEL
 #endif
@@ -1901,8 +2003,6 @@ finished:
  *	should know vmalloc() area is valid and can use memcpy().
  *	This is for routines which have to access vmalloc area without
  *	any informaion, as /dev/kmem.
- *
- *	The caller should guarantee KM_USER1 is not used.
  */
 
 long vwrite(char *buf, char *addr, unsigned long count)
@@ -2155,17 +2255,16 @@ static unsigned long pvm_determine_end(struct vmap_area **pnext,
  * @sizes: array containing size of each area
  * @nr_vms: the number of areas to allocate
  * @align: alignment, all entries in @offsets and @sizes must be aligned to this
- * @gfp_mask: allocation mask
  *
  * Returns: kmalloc'd vm_struct pointer array pointing to allocated
  *	    vm_structs on success, %NULL on failure
  *
  * Percpu allocator wants to use congruent vm areas so that it can
  * maintain the offsets among percpu areas.  This function allocates
- * congruent vmalloc areas for it.  These areas tend to be scattered
- * pretty far, distance between two areas easily going up to
- * gigabytes.  To avoid interacting with regular vmallocs, these areas
- * are allocated from top.
+ * congruent vmalloc areas for it with GFP_KERNEL.  These areas tend to
+ * be scattered pretty far, distance between two areas easily going up
+ * to gigabytes.  To avoid interacting with regular vmallocs, these
+ * areas are allocated from top.
  *
  * Despite its complicated look, this allocator is rather simple.  It
  * does everything top-down and scans areas from the end looking for
@@ -2176,7 +2275,7 @@ static unsigned long pvm_determine_end(struct vmap_area **pnext,
  */
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				     const size_t *sizes, int nr_vms,
-				     size_t align, gfp_t gfp_mask)
+				     size_t align)
 {
 	const unsigned long vmalloc_start = ALIGN(VMALLOC_START, align);
 	const unsigned long vmalloc_end = VMALLOC_END & ~(align - 1);
@@ -2186,8 +2285,6 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 	unsigned long base, start, end, last_end;
 	bool purged = false;
 
-	gfp_mask &= GFP_RECLAIM_MASK;
-
 	/* verify parameters and allocate data structures */
 	BUG_ON(align & ~PAGE_MASK || !is_power_of_2(align));
 	for (last_area = 0, area = 0; area < nr_vms; area++) {
@@ -2220,14 +2317,14 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 		return NULL;
 	}
 
-	vms = kzalloc(sizeof(vms[0]) * nr_vms, gfp_mask);
-	vas = kzalloc(sizeof(vas[0]) * nr_vms, gfp_mask);
+	vms = kzalloc(sizeof(vms[0]) * nr_vms, GFP_KERNEL);
+	vas = kzalloc(sizeof(vas[0]) * nr_vms, GFP_KERNEL);
 	if (!vas || !vms)
 		goto err_free;
 
 	for (area = 0; area < nr_vms; area++) {
-		vas[area] = kzalloc(sizeof(struct vmap_area), gfp_mask);
-		vms[area] = kzalloc(sizeof(struct vm_struct), gfp_mask);
+		vas[area] = kzalloc(sizeof(struct vmap_area), GFP_KERNEL);
+		vms[area] = kzalloc(sizeof(struct vm_struct), GFP_KERNEL);
 		if (!vas[area] || !vms[area])
 			goto err_free;
 	}
@@ -2350,6 +2447,7 @@ 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)
 {
 	loff_t n = *pos;
 	struct vm_struct *v;
@@ -2376,6 +2474,7 @@ static void *s_next(struct seq_file *m, void *p, loff_t *pos)
 }
 
 static void s_stop(struct seq_file *m, void *p)
+	__releases(&vmlist_lock)
 {
 	read_unlock(&vmlist_lock);
 }
@@ -2406,13 +2505,8 @@ static int s_show(struct seq_file *m, void *p)
 	seq_printf(m, "0x%p-0x%p %7ld",
 		v->addr, v->addr + v->size, v->size);
 
-	if (v->caller) {
-		char buff[KSYM_SYMBOL_LEN];
-
-		seq_putc(m, ' ');
-		sprint_symbol(buff, (unsigned long)v->caller);
-		seq_puts(m, buff);
-	}
+	if (v->caller)
+		seq_printf(m, " %pS", v->caller);
 
 	if (v->nr_pages)
 		seq_printf(m, " pages=%d", v->nr_pages);