Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6:
slub: avoid leaking caches or refcounts on sysfs error
slab: Fix comment on #endif
slab: remove GFP_THISNODE clearing from alloc_slabmgmt()
slub: Add might_sleep_if() to slab_alloc()
SLUB: failslab support
slub: Fix incorrect use of loose
slab: Update the kmem_cache_create documentation regarding the name parameter
slub: make early_kmem_cache_node_alloc void
slab: unsigned slabp->inuse cannot be less than 0
slub - fix get_object_page comment
SLUB: Replace __builtin_return_address(0) with _RET_IP_.
SLUB: cleanup - define macros instead of hardcoded numbers
diff --git a/include/linux/fault-inject.h b/include/linux/fault-inject.h
index 32368c4..06ca9b2 100644
--- a/include/linux/fault-inject.h
+++ b/include/linux/fault-inject.h
@@ -81,4 +81,13 @@
#endif /* CONFIG_FAULT_INJECTION */
+#ifdef CONFIG_FAILSLAB
+extern bool should_failslab(size_t size, gfp_t gfpflags);
+#else
+static inline bool should_failslab(size_t size, gfp_t gfpflags)
+{
+ return false;
+}
+#endif /* CONFIG_FAILSLAB */
+
#endif /* _LINUX_FAULT_INJECT_H */
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 000da12..f96d13c 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -253,9 +253,9 @@
* request comes from.
*/
#if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_SLUB)
-extern void *__kmalloc_track_caller(size_t, gfp_t, void*);
+extern void *__kmalloc_track_caller(size_t, gfp_t, unsigned long);
#define kmalloc_track_caller(size, flags) \
- __kmalloc_track_caller(size, flags, __builtin_return_address(0))
+ __kmalloc_track_caller(size, flags, _RET_IP_)
#else
#define kmalloc_track_caller(size, flags) \
__kmalloc(size, flags)
@@ -271,10 +271,10 @@
* allocation request comes from.
*/
#if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_SLUB)
-extern void *__kmalloc_node_track_caller(size_t, gfp_t, int, void *);
+extern void *__kmalloc_node_track_caller(size_t, gfp_t, int, unsigned long);
#define kmalloc_node_track_caller(size, flags, node) \
__kmalloc_node_track_caller(size, flags, node, \
- __builtin_return_address(0))
+ _RET_IP_)
#else
#define kmalloc_node_track_caller(size, flags, node) \
__kmalloc_node(size, flags, node)
@@ -285,7 +285,7 @@
#define kmalloc_node_track_caller(size, flags, node) \
kmalloc_track_caller(size, flags)
-#endif /* DEBUG_SLAB */
+#endif /* CONFIG_NUMA */
/*
* Shortcuts
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index eae594c..2e75478 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -730,6 +730,7 @@
config FAILSLAB
bool "Fault-injection capability for kmalloc"
depends on FAULT_INJECTION
+ depends on SLAB || SLUB
help
Provide fault-injection capability for kmalloc.
diff --git a/mm/Makefile b/mm/Makefile
index c06b45a..51c2770 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -28,6 +28,7 @@
obj-$(CONFIG_MMU_NOTIFIER) += mmu_notifier.o
obj-$(CONFIG_SLAB) += slab.o
obj-$(CONFIG_SLUB) += slub.o
+obj-$(CONFIG_FAILSLAB) += failslab.o
obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
obj-$(CONFIG_FS_XIP) += filemap_xip.o
obj-$(CONFIG_MIGRATION) += migrate.o
diff --git a/mm/failslab.c b/mm/failslab.c
new file mode 100644
index 0000000..7c6ea64
--- /dev/null
+++ b/mm/failslab.c
@@ -0,0 +1,59 @@
+#include <linux/fault-inject.h>
+
+static struct {
+ struct fault_attr attr;
+ u32 ignore_gfp_wait;
+#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS
+ struct dentry *ignore_gfp_wait_file;
+#endif
+} failslab = {
+ .attr = FAULT_ATTR_INITIALIZER,
+ .ignore_gfp_wait = 1,
+};
+
+bool should_failslab(size_t size, gfp_t gfpflags)
+{
+ if (gfpflags & __GFP_NOFAIL)
+ return false;
+
+ if (failslab.ignore_gfp_wait && (gfpflags & __GFP_WAIT))
+ return false;
+
+ return should_fail(&failslab.attr, size);
+}
+
+static int __init setup_failslab(char *str)
+{
+ return setup_fault_attr(&failslab.attr, str);
+}
+__setup("failslab=", setup_failslab);
+
+#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS
+
+static int __init failslab_debugfs_init(void)
+{
+ mode_t mode = S_IFREG | S_IRUSR | S_IWUSR;
+ struct dentry *dir;
+ int err;
+
+ err = init_fault_attr_dentries(&failslab.attr, "failslab");
+ if (err)
+ return err;
+ dir = failslab.attr.dentries.dir;
+
+ failslab.ignore_gfp_wait_file =
+ debugfs_create_bool("ignore-gfp-wait", mode, dir,
+ &failslab.ignore_gfp_wait);
+
+ if (!failslab.ignore_gfp_wait_file) {
+ err = -ENOMEM;
+ debugfs_remove(failslab.ignore_gfp_wait_file);
+ cleanup_fault_attr_dentries(&failslab.attr);
+ }
+
+ return err;
+}
+
+late_initcall(failslab_debugfs_init);
+
+#endif /* CONFIG_FAULT_INJECTION_DEBUG_FS */
diff --git a/mm/slab.c b/mm/slab.c
index 0918751..f97e564 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2123,6 +2123,8 @@
*
* @name must be valid until the cache is destroyed. This implies that
* the module calling this has to destroy the cache before getting unloaded.
+ * Note that kmem_cache_name() is not guaranteed to return the same pointer,
+ * therefore applications must manage it themselves.
*
* The flags are
*
@@ -2609,7 +2611,7 @@
if (OFF_SLAB(cachep)) {
/* Slab management obj is off-slab. */
slabp = kmem_cache_alloc_node(cachep->slabp_cache,
- local_flags & ~GFP_THISNODE, nodeid);
+ local_flags, nodeid);
if (!slabp)
return NULL;
} else {
@@ -2997,7 +2999,7 @@
* there must be at least one object available for
* allocation.
*/
- BUG_ON(slabp->inuse < 0 || slabp->inuse >= cachep->num);
+ BUG_ON(slabp->inuse >= cachep->num);
while (slabp->inuse < cachep->num && batchcount--) {
STATS_INC_ALLOCED(cachep);
@@ -3106,79 +3108,14 @@
#define cache_alloc_debugcheck_after(a,b,objp,d) (objp)
#endif
-#ifdef CONFIG_FAILSLAB
-
-static struct failslab_attr {
-
- struct fault_attr attr;
-
- u32 ignore_gfp_wait;
-#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS
- struct dentry *ignore_gfp_wait_file;
-#endif
-
-} failslab = {
- .attr = FAULT_ATTR_INITIALIZER,
- .ignore_gfp_wait = 1,
-};
-
-static int __init setup_failslab(char *str)
-{
- return setup_fault_attr(&failslab.attr, str);
-}
-__setup("failslab=", setup_failslab);
-
-static int should_failslab(struct kmem_cache *cachep, gfp_t flags)
+static bool slab_should_failslab(struct kmem_cache *cachep, gfp_t flags)
{
if (cachep == &cache_cache)
- return 0;
- if (flags & __GFP_NOFAIL)
- return 0;
- if (failslab.ignore_gfp_wait && (flags & __GFP_WAIT))
- return 0;
+ return false;
- return should_fail(&failslab.attr, obj_size(cachep));
+ return should_failslab(obj_size(cachep), flags);
}
-#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS
-
-static int __init failslab_debugfs(void)
-{
- mode_t mode = S_IFREG | S_IRUSR | S_IWUSR;
- struct dentry *dir;
- int err;
-
- err = init_fault_attr_dentries(&failslab.attr, "failslab");
- if (err)
- return err;
- dir = failslab.attr.dentries.dir;
-
- failslab.ignore_gfp_wait_file =
- debugfs_create_bool("ignore-gfp-wait", mode, dir,
- &failslab.ignore_gfp_wait);
-
- if (!failslab.ignore_gfp_wait_file) {
- err = -ENOMEM;
- debugfs_remove(failslab.ignore_gfp_wait_file);
- cleanup_fault_attr_dentries(&failslab.attr);
- }
-
- return err;
-}
-
-late_initcall(failslab_debugfs);
-
-#endif /* CONFIG_FAULT_INJECTION_DEBUG_FS */
-
-#else /* CONFIG_FAILSLAB */
-
-static inline int should_failslab(struct kmem_cache *cachep, gfp_t flags)
-{
- return 0;
-}
-
-#endif /* CONFIG_FAILSLAB */
-
static inline void *____cache_alloc(struct kmem_cache *cachep, gfp_t flags)
{
void *objp;
@@ -3381,7 +3318,7 @@
unsigned long save_flags;
void *ptr;
- if (should_failslab(cachep, flags))
+ if (slab_should_failslab(cachep, flags))
return NULL;
cache_alloc_debugcheck_before(cachep, flags);
@@ -3457,7 +3394,7 @@
unsigned long save_flags;
void *objp;
- if (should_failslab(cachep, flags))
+ if (slab_should_failslab(cachep, flags))
return NULL;
cache_alloc_debugcheck_before(cachep, flags);
@@ -3686,9 +3623,9 @@
EXPORT_SYMBOL(__kmalloc_node);
void *__kmalloc_node_track_caller(size_t size, gfp_t flags,
- int node, void *caller)
+ int node, unsigned long caller)
{
- return __do_kmalloc_node(size, flags, node, caller);
+ return __do_kmalloc_node(size, flags, node, (void *)caller);
}
EXPORT_SYMBOL(__kmalloc_node_track_caller);
#else
@@ -3730,9 +3667,9 @@
}
EXPORT_SYMBOL(__kmalloc);
-void *__kmalloc_track_caller(size_t size, gfp_t flags, void *caller)
+void *__kmalloc_track_caller(size_t size, gfp_t flags, unsigned long caller)
{
- return __do_kmalloc(size, flags, caller);
+ return __do_kmalloc(size, flags, (void *)caller);
}
EXPORT_SYMBOL(__kmalloc_track_caller);
diff --git a/mm/slub.c b/mm/slub.c
index a2cd47d..6cb7ad1 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -24,6 +24,7 @@
#include <linux/kallsyms.h>
#include <linux/memory.h>
#include <linux/math64.h>
+#include <linux/fault-inject.h>
/*
* Lock order:
@@ -153,6 +154,10 @@
#define ARCH_SLAB_MINALIGN __alignof__(unsigned long long)
#endif
+#define OO_SHIFT 16
+#define OO_MASK ((1 << OO_SHIFT) - 1)
+#define MAX_OBJS_PER_PAGE 65535 /* since page.objects is u16 */
+
/* Internal SLUB flags */
#define __OBJECT_POISON 0x80000000 /* Poison object */
#define __SYSFS_ADD_DEFERRED 0x40000000 /* Not yet visible via sysfs */
@@ -178,7 +183,7 @@
* Tracking user of a slab.
*/
struct track {
- void *addr; /* Called from address */
+ unsigned long addr; /* Called from address */
int cpu; /* Was running on cpu */
int pid; /* Pid context */
unsigned long when; /* When did the operation occur */
@@ -290,7 +295,7 @@
unsigned long size)
{
struct kmem_cache_order_objects x = {
- (order << 16) + (PAGE_SIZE << order) / size
+ (order << OO_SHIFT) + (PAGE_SIZE << order) / size
};
return x;
@@ -298,12 +303,12 @@
static inline int oo_order(struct kmem_cache_order_objects x)
{
- return x.x >> 16;
+ return x.x >> OO_SHIFT;
}
static inline int oo_objects(struct kmem_cache_order_objects x)
{
- return x.x & ((1 << 16) - 1);
+ return x.x & OO_MASK;
}
#ifdef CONFIG_SLUB_DEBUG
@@ -367,7 +372,7 @@
}
static void set_track(struct kmem_cache *s, void *object,
- enum track_item alloc, void *addr)
+ enum track_item alloc, unsigned long addr)
{
struct track *p;
@@ -391,8 +396,8 @@
if (!(s->flags & SLAB_STORE_USER))
return;
- set_track(s, object, TRACK_FREE, NULL);
- set_track(s, object, TRACK_ALLOC, NULL);
+ set_track(s, object, TRACK_FREE, 0UL);
+ set_track(s, object, TRACK_ALLOC, 0UL);
}
static void print_track(const char *s, struct track *t)
@@ -401,7 +406,7 @@
return;
printk(KERN_ERR "INFO: %s in %pS age=%lu cpu=%u pid=%d\n",
- s, t->addr, jiffies - t->when, t->cpu, t->pid);
+ s, (void *)t->addr, jiffies - t->when, t->cpu, t->pid);
}
static void print_tracking(struct kmem_cache *s, void *object)
@@ -692,7 +697,7 @@
if (!check_valid_pointer(s, page, get_freepointer(s, p))) {
object_err(s, page, p, "Freepointer corrupt");
/*
- * No choice but to zap it and thus loose the remainder
+ * No choice but to zap it and thus lose the remainder
* of the free objects in this slab. May cause
* another error because the object count is now wrong.
*/
@@ -764,8 +769,8 @@
}
max_objects = (PAGE_SIZE << compound_order(page)) / s->size;
- if (max_objects > 65535)
- max_objects = 65535;
+ if (max_objects > MAX_OBJS_PER_PAGE)
+ max_objects = MAX_OBJS_PER_PAGE;
if (page->objects != max_objects) {
slab_err(s, page, "Wrong number of objects. Found %d but "
@@ -866,7 +871,7 @@
}
static int alloc_debug_processing(struct kmem_cache *s, struct page *page,
- void *object, void *addr)
+ void *object, unsigned long addr)
{
if (!check_slab(s, page))
goto bad;
@@ -906,7 +911,7 @@
}
static int free_debug_processing(struct kmem_cache *s, struct page *page,
- void *object, void *addr)
+ void *object, unsigned long addr)
{
if (!check_slab(s, page))
goto fail;
@@ -1029,10 +1034,10 @@
struct page *page, void *object) {}
static inline int alloc_debug_processing(struct kmem_cache *s,
- struct page *page, void *object, void *addr) { return 0; }
+ struct page *page, void *object, unsigned long addr) { return 0; }
static inline int free_debug_processing(struct kmem_cache *s,
- struct page *page, void *object, void *addr) { return 0; }
+ struct page *page, void *object, unsigned long addr) { return 0; }
static inline int slab_pad_check(struct kmem_cache *s, struct page *page)
{ return 1; }
@@ -1499,8 +1504,8 @@
* we need to allocate a new slab. This is the slowest path since it involves
* a call to the page allocator and the setup of a new slab.
*/
-static void *__slab_alloc(struct kmem_cache *s,
- gfp_t gfpflags, int node, void *addr, struct kmem_cache_cpu *c)
+static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
+ unsigned long addr, struct kmem_cache_cpu *c)
{
void **object;
struct page *new;
@@ -1584,13 +1589,18 @@
* Otherwise we can simply pick the next object from the lockless free list.
*/
static __always_inline void *slab_alloc(struct kmem_cache *s,
- gfp_t gfpflags, int node, void *addr)
+ gfp_t gfpflags, int node, unsigned long addr)
{
void **object;
struct kmem_cache_cpu *c;
unsigned long flags;
unsigned int objsize;
+ might_sleep_if(gfpflags & __GFP_WAIT);
+
+ if (should_failslab(s->objsize, gfpflags))
+ return NULL;
+
local_irq_save(flags);
c = get_cpu_slab(s, smp_processor_id());
objsize = c->objsize;
@@ -1613,14 +1623,14 @@
void *kmem_cache_alloc(struct kmem_cache *s, gfp_t gfpflags)
{
- return slab_alloc(s, gfpflags, -1, __builtin_return_address(0));
+ return slab_alloc(s, gfpflags, -1, _RET_IP_);
}
EXPORT_SYMBOL(kmem_cache_alloc);
#ifdef CONFIG_NUMA
void *kmem_cache_alloc_node(struct kmem_cache *s, gfp_t gfpflags, int node)
{
- return slab_alloc(s, gfpflags, node, __builtin_return_address(0));
+ return slab_alloc(s, gfpflags, node, _RET_IP_);
}
EXPORT_SYMBOL(kmem_cache_alloc_node);
#endif
@@ -1634,7 +1644,7 @@
* handling required then we can return immediately.
*/
static void __slab_free(struct kmem_cache *s, struct page *page,
- void *x, void *addr, unsigned int offset)
+ void *x, unsigned long addr, unsigned int offset)
{
void *prior;
void **object = (void *)x;
@@ -1704,7 +1714,7 @@
* with all sorts of special processing.
*/
static __always_inline void slab_free(struct kmem_cache *s,
- struct page *page, void *x, void *addr)
+ struct page *page, void *x, unsigned long addr)
{
void **object = (void *)x;
struct kmem_cache_cpu *c;
@@ -1731,11 +1741,11 @@
page = virt_to_head_page(x);
- slab_free(s, page, x, __builtin_return_address(0));
+ slab_free(s, page, x, _RET_IP_);
}
EXPORT_SYMBOL(kmem_cache_free);
-/* Figure out on which slab object the object resides */
+/* Figure out on which slab page the object resides */
static struct page *get_object_page(const void *x)
{
struct page *page = virt_to_head_page(x);
@@ -1807,8 +1817,8 @@
int rem;
int min_order = slub_min_order;
- if ((PAGE_SIZE << min_order) / size > 65535)
- return get_order(size * 65535) - 1;
+ if ((PAGE_SIZE << min_order) / size > MAX_OBJS_PER_PAGE)
+ return get_order(size * MAX_OBJS_PER_PAGE) - 1;
for (order = max(min_order,
fls(min_objects * size - 1) - PAGE_SHIFT);
@@ -2073,8 +2083,7 @@
* when allocating for the kmalloc_node_cache. This is used for bootstrapping
* memory on a fresh node that has no slab structures yet.
*/
-static struct kmem_cache_node *early_kmem_cache_node_alloc(gfp_t gfpflags,
- int node)
+static void early_kmem_cache_node_alloc(gfp_t gfpflags, int node)
{
struct page *page;
struct kmem_cache_node *n;
@@ -2112,7 +2121,6 @@
local_irq_save(flags);
add_partial(n, page, 0);
local_irq_restore(flags);
- return n;
}
static void free_kmem_cache_nodes(struct kmem_cache *s)
@@ -2144,8 +2152,7 @@
n = &s->local_node;
else {
if (slab_state == DOWN) {
- n = early_kmem_cache_node_alloc(gfpflags,
- node);
+ early_kmem_cache_node_alloc(gfpflags, node);
continue;
}
n = kmem_cache_alloc_node(kmalloc_caches,
@@ -2659,7 +2666,7 @@
if (unlikely(ZERO_OR_NULL_PTR(s)))
return s;
- return slab_alloc(s, flags, -1, __builtin_return_address(0));
+ return slab_alloc(s, flags, -1, _RET_IP_);
}
EXPORT_SYMBOL(__kmalloc);
@@ -2687,7 +2694,7 @@
if (unlikely(ZERO_OR_NULL_PTR(s)))
return s;
- return slab_alloc(s, flags, node, __builtin_return_address(0));
+ return slab_alloc(s, flags, node, _RET_IP_);
}
EXPORT_SYMBOL(__kmalloc_node);
#endif
@@ -2744,7 +2751,7 @@
put_page(page);
return;
}
- slab_free(page->slab, page, object, __builtin_return_address(0));
+ slab_free(page->slab, page, object, _RET_IP_);
}
EXPORT_SYMBOL(kfree);
@@ -3123,8 +3130,12 @@
s->inuse = max_t(int, s->inuse, ALIGN(size, sizeof(void *)));
up_write(&slub_lock);
- if (sysfs_slab_alias(s, name))
+ if (sysfs_slab_alias(s, name)) {
+ down_write(&slub_lock);
+ s->refcount--;
+ up_write(&slub_lock);
goto err;
+ }
return s;
}
@@ -3134,8 +3145,13 @@
size, align, flags, ctor)) {
list_add(&s->list, &slab_caches);
up_write(&slub_lock);
- if (sysfs_slab_add(s))
+ if (sysfs_slab_add(s)) {
+ down_write(&slub_lock);
+ list_del(&s->list);
+ up_write(&slub_lock);
+ kfree(s);
goto err;
+ }
return s;
}
kfree(s);
@@ -3202,7 +3218,7 @@
#endif
-void *__kmalloc_track_caller(size_t size, gfp_t gfpflags, void *caller)
+void *__kmalloc_track_caller(size_t size, gfp_t gfpflags, unsigned long caller)
{
struct kmem_cache *s;
@@ -3218,7 +3234,7 @@
}
void *__kmalloc_node_track_caller(size_t size, gfp_t gfpflags,
- int node, void *caller)
+ int node, unsigned long caller)
{
struct kmem_cache *s;
@@ -3429,7 +3445,7 @@
struct location {
unsigned long count;
- void *addr;
+ unsigned long addr;
long long sum_time;
long min_time;
long max_time;
@@ -3477,7 +3493,7 @@
{
long start, end, pos;
struct location *l;
- void *caddr;
+ unsigned long caddr;
unsigned long age = jiffies - track->when;
start = -1;
@@ -4345,7 +4361,7 @@
/*
* Need to buffer aliases during bootup until sysfs becomes
- * available lest we loose that information.
+ * available lest we lose that information.
*/
struct saved_alias {
struct kmem_cache *s;