aboutsummaryrefslogtreecommitdiff
path: root/mm/slub.c
diff options
context:
space:
mode:
Diffstat (limited to 'mm/slub.c')
-rw-r--r--mm/slub.c376
1 files changed, 309 insertions, 67 deletions
diff --git a/mm/slub.c b/mm/slub.c
index e15aa7f193c..94d2a33a866 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -64,7 +64,7 @@
* we must stay away from it for a while since we may cause a bouncing
* cacheline if we try to acquire the lock. So go onto the next slab.
* If all pages are busy then we may allocate a new slab instead of reusing
- * a partial slab. A new slab has noone operating on it and thus there is
+ * a partial slab. A new slab has no one operating on it and thus there is
* no danger of cacheline contention.
*
* Interrupts are disabled during allocation and deallocation in order to
@@ -217,7 +217,7 @@ static inline void sysfs_slab_remove(struct kmem_cache *s)
#endif
-static inline void stat(struct kmem_cache *s, enum stat_item si)
+static inline void stat(const struct kmem_cache *s, enum stat_item si)
{
#ifdef CONFIG_SLUB_STATS
__this_cpu_inc(s->cpu_slab->stat[si]);
@@ -281,11 +281,40 @@ static inline int slab_index(void *p, struct kmem_cache *s, void *addr)
return (p - addr) / s->size;
}
+static inline size_t slab_ksize(const struct kmem_cache *s)
+{
+#ifdef CONFIG_SLUB_DEBUG
+ /*
+ * Debugging requires use of the padding between object
+ * and whatever may come after it.
+ */
+ if (s->flags & (SLAB_RED_ZONE | SLAB_POISON))
+ return s->objsize;
+
+#endif
+ /*
+ * If we have the need to store the freelist pointer
+ * back there or track user information then we can
+ * only use the space before that information.
+ */
+ if (s->flags & (SLAB_DESTROY_BY_RCU | SLAB_STORE_USER))
+ return s->inuse;
+ /*
+ * Else we can use all the padding etc for the allocation
+ */
+ return s->size;
+}
+
+static inline int order_objects(int order, unsigned long size, int reserved)
+{
+ return ((PAGE_SIZE << order) - reserved) / size;
+}
+
static inline struct kmem_cache_order_objects oo_make(int order,
- unsigned long size)
+ unsigned long size, int reserved)
{
struct kmem_cache_order_objects x = {
- (order << OO_SHIFT) + (PAGE_SIZE << order) / size
+ (order << OO_SHIFT) + order_objects(order, size, reserved)
};
return x;
@@ -617,7 +646,7 @@ static int slab_pad_check(struct kmem_cache *s, struct page *page)
return 1;
start = page_address(page);
- length = (PAGE_SIZE << compound_order(page));
+ length = (PAGE_SIZE << compound_order(page)) - s->reserved;
end = start + length;
remainder = length % s->size;
if (!remainder)
@@ -698,7 +727,7 @@ static int check_slab(struct kmem_cache *s, struct page *page)
return 0;
}
- maxobj = (PAGE_SIZE << compound_order(page)) / s->size;
+ maxobj = order_objects(compound_order(page), s->size, s->reserved);
if (page->objects > maxobj) {
slab_err(s, page, "objects %u > max %u",
s->name, page->objects, maxobj);
@@ -748,7 +777,7 @@ static int on_freelist(struct kmem_cache *s, struct page *page, void *search)
nr++;
}
- max_objects = (PAGE_SIZE << compound_order(page)) / s->size;
+ max_objects = order_objects(compound_order(page), s->size, s->reserved);
if (max_objects > MAX_OBJS_PER_PAGE)
max_objects = MAX_OBJS_PER_PAGE;
@@ -800,21 +829,31 @@ static inline int slab_pre_alloc_hook(struct kmem_cache *s, gfp_t flags)
static inline void slab_post_alloc_hook(struct kmem_cache *s, gfp_t flags, void *object)
{
flags &= gfp_allowed_mask;
- kmemcheck_slab_alloc(s, flags, object, s->objsize);
+ kmemcheck_slab_alloc(s, flags, object, slab_ksize(s));
kmemleak_alloc_recursive(object, s->objsize, 1, s->flags, flags);
}
static inline void slab_free_hook(struct kmem_cache *s, void *x)
{
kmemleak_free_recursive(x, s->flags);
-}
-static inline void slab_free_hook_irq(struct kmem_cache *s, void *object)
-{
- kmemcheck_slab_free(s, object, s->objsize);
- debug_check_no_locks_freed(object, s->objsize);
+ /*
+ * Trouble is that we may no longer disable interupts in the fast path
+ * So in order to make the debug calls that expect irqs to be
+ * disabled we need to disable interrupts temporarily.
+ */
+#if defined(CONFIG_KMEMCHECK) || defined(CONFIG_LOCKDEP)
+ {
+ unsigned long flags;
+
+ local_irq_save(flags);
+ kmemcheck_slab_free(s, x, s->objsize);
+ debug_check_no_locks_freed(x, s->objsize);
+ local_irq_restore(flags);
+ }
+#endif
if (!(s->flags & SLAB_DEBUG_OBJECTS))
- debug_check_no_obj_freed(object, s->objsize);
+ debug_check_no_obj_freed(x, s->objsize);
}
/*
@@ -1101,9 +1140,6 @@ static inline void slab_post_alloc_hook(struct kmem_cache *s, gfp_t flags,
static inline void slab_free_hook(struct kmem_cache *s, void *x) {}
-static inline void slab_free_hook_irq(struct kmem_cache *s,
- void *object) {}
-
#endif /* CONFIG_SLUB_DEBUG */
/*
@@ -1249,21 +1285,38 @@ static void __free_slab(struct kmem_cache *s, struct page *page)
__free_pages(page, order);
}
+#define need_reserve_slab_rcu \
+ (sizeof(((struct page *)NULL)->lru) < sizeof(struct rcu_head))
+
static void rcu_free_slab(struct rcu_head *h)
{
struct page *page;
- page = container_of((struct list_head *)h, struct page, lru);
+ if (need_reserve_slab_rcu)
+ page = virt_to_head_page(h);
+ else
+ page = container_of((struct list_head *)h, struct page, lru);
+
__free_slab(page->slab, page);
}
static void free_slab(struct kmem_cache *s, struct page *page)
{
if (unlikely(s->flags & SLAB_DESTROY_BY_RCU)) {
- /*
- * RCU free overloads the RCU head over the LRU
- */
- struct rcu_head *head = (void *)&page->lru;
+ struct rcu_head *head;
+
+ if (need_reserve_slab_rcu) {
+ int order = compound_order(page);
+ int offset = (PAGE_SIZE << order) - s->reserved;
+
+ VM_BUG_ON(s->reserved != sizeof(*head));
+ head = page_address(page) + offset;
+ } else {
+ /*
+ * RCU free overloads the RCU head over the LRU
+ */
+ head = (void *)&page->lru;
+ }
call_rcu(head, rcu_free_slab);
} else
@@ -1487,6 +1540,78 @@ static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail)
}
}
+#ifdef CONFIG_CMPXCHG_LOCAL
+#ifdef CONFIG_PREEMPT
+/*
+ * Calculate the next globally unique transaction for disambiguiation
+ * during cmpxchg. The transactions start with the cpu number and are then
+ * incremented by CONFIG_NR_CPUS.
+ */
+#define TID_STEP roundup_pow_of_two(CONFIG_NR_CPUS)
+#else
+/*
+ * No preemption supported therefore also no need to check for
+ * different cpus.
+ */
+#define TID_STEP 1
+#endif
+
+static inline unsigned long next_tid(unsigned long tid)
+{
+ return tid + TID_STEP;
+}
+
+static inline unsigned int tid_to_cpu(unsigned long tid)
+{
+ return tid % TID_STEP;
+}
+
+static inline unsigned long tid_to_event(unsigned long tid)
+{
+ return tid / TID_STEP;
+}
+
+static inline unsigned int init_tid(int cpu)
+{
+ return cpu;
+}
+
+static inline void note_cmpxchg_failure(const char *n,
+ const struct kmem_cache *s, unsigned long tid)
+{
+#ifdef SLUB_DEBUG_CMPXCHG
+ unsigned long actual_tid = __this_cpu_read(s->cpu_slab->tid);
+
+ printk(KERN_INFO "%s %s: cmpxchg redo ", n, s->name);
+
+#ifdef CONFIG_PREEMPT
+ if (tid_to_cpu(tid) != tid_to_cpu(actual_tid))
+ printk("due to cpu change %d -> %d\n",
+ tid_to_cpu(tid), tid_to_cpu(actual_tid));
+ else
+#endif
+ if (tid_to_event(tid) != tid_to_event(actual_tid))
+ printk("due to cpu running other code. Event %ld->%ld\n",
+ tid_to_event(tid), tid_to_event(actual_tid));
+ else
+ printk("for unknown reason: actual=%lx was=%lx target=%lx\n",
+ actual_tid, tid, next_tid(tid));
+#endif
+ stat(s, CMPXCHG_DOUBLE_CPU_FAIL);
+}
+
+#endif
+
+void init_kmem_cache_cpus(struct kmem_cache *s)
+{
+#ifdef CONFIG_CMPXCHG_LOCAL
+ int cpu;
+
+ for_each_possible_cpu(cpu)
+ per_cpu_ptr(s->cpu_slab, cpu)->tid = init_tid(cpu);
+#endif
+
+}
/*
* Remove the cpu slab
*/
@@ -1518,6 +1643,9 @@ static void deactivate_slab(struct kmem_cache *s, struct kmem_cache_cpu *c)
page->inuse--;
}
c->page = NULL;
+#ifdef CONFIG_CMPXCHG_LOCAL
+ c->tid = next_tid(c->tid);
+#endif
unfreeze_slab(s, page, tail);
}
@@ -1652,6 +1780,19 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
{
void **object;
struct page *new;
+#ifdef CONFIG_CMPXCHG_LOCAL
+ unsigned long flags;
+
+ local_irq_save(flags);
+#ifdef CONFIG_PREEMPT
+ /*
+ * We may have been preempted and rescheduled on a different
+ * cpu before disabling interrupts. Need to reload cpu area
+ * pointer.
+ */
+ c = this_cpu_ptr(s->cpu_slab);
+#endif
+#endif
/* We handle __GFP_ZERO in the caller */
gfpflags &= ~__GFP_ZERO;
@@ -1678,6 +1819,10 @@ load_freelist:
c->node = page_to_nid(c->page);
unlock_out:
slab_unlock(c->page);
+#ifdef CONFIG_CMPXCHG_LOCAL
+ c->tid = next_tid(c->tid);
+ local_irq_restore(flags);
+#endif
stat(s, ALLOC_SLOWPATH);
return object;
@@ -1713,6 +1858,9 @@ new_slab:
}
if (!(gfpflags & __GFP_NOWARN) && printk_ratelimit())
slab_out_of_memory(s, gfpflags, node);
+#ifdef CONFIG_CMPXCHG_LOCAL
+ local_irq_restore(flags);
+#endif
return NULL;
debug:
if (!alloc_debug_processing(s, c->page, object, addr))
@@ -1739,23 +1887,76 @@ static __always_inline void *slab_alloc(struct kmem_cache *s,
{
void **object;
struct kmem_cache_cpu *c;
+#ifdef CONFIG_CMPXCHG_LOCAL
+ unsigned long tid;
+#else
unsigned long flags;
+#endif
if (slab_pre_alloc_hook(s, gfpflags))
return NULL;
+#ifndef CONFIG_CMPXCHG_LOCAL
local_irq_save(flags);
+#else
+redo:
+#endif
+
+ /*
+ * Must read kmem_cache cpu data via this cpu ptr. Preemption is
+ * enabled. We may switch back and forth between cpus while
+ * reading from one cpu area. That does not matter as long
+ * as we end up on the original cpu again when doing the cmpxchg.
+ */
c = __this_cpu_ptr(s->cpu_slab);
+
+#ifdef CONFIG_CMPXCHG_LOCAL
+ /*
+ * The transaction ids are globally unique per cpu and per operation on
+ * a per cpu queue. Thus they can be guarantee that the cmpxchg_double
+ * occurs on the right processor and that there was no operation on the
+ * linked list in between.
+ */
+ tid = c->tid;
+ barrier();
+#endif
+
object = c->freelist;
if (unlikely(!object || !node_match(c, node)))
object = __slab_alloc(s, gfpflags, node, addr, c);
else {
+#ifdef CONFIG_CMPXCHG_LOCAL
+ /*
+ * The cmpxchg will only match if there was no additional
+ * operation and if we are on the right processor.
+ *
+ * The cmpxchg does the following atomically (without lock semantics!)
+ * 1. Relocate first pointer to the current per cpu area.
+ * 2. Verify that tid and freelist have not been changed
+ * 3. If they were not changed replace tid and freelist
+ *
+ * Since this is without lock semantics the protection is only against
+ * code executing on this cpu *not* from access by other cpus.
+ */
+ if (unlikely(!this_cpu_cmpxchg_double(
+ s->cpu_slab->freelist, s->cpu_slab->tid,
+ object, tid,
+ get_freepointer(s, object), next_tid(tid)))) {
+
+ note_cmpxchg_failure("slab_alloc", s, tid);
+ goto redo;
+ }
+#else
c->freelist = get_freepointer(s, object);
+#endif
stat(s, ALLOC_FASTPATH);
}
+
+#ifndef CONFIG_CMPXCHG_LOCAL
local_irq_restore(flags);
+#endif
if (unlikely(gfpflags & __GFP_ZERO) && object)
memset(object, 0, s->objsize);
@@ -1833,9 +2034,13 @@ static void __slab_free(struct kmem_cache *s, struct page *page,
{
void *prior;
void **object = (void *)x;
+#ifdef CONFIG_CMPXCHG_LOCAL
+ unsigned long flags;
- stat(s, FREE_SLOWPATH);
+ local_irq_save(flags);
+#endif
slab_lock(page);
+ stat(s, FREE_SLOWPATH);
if (kmem_cache_debug(s))
goto debug;
@@ -1865,6 +2070,9 @@ checks_ok:
out_unlock:
slab_unlock(page);
+#ifdef CONFIG_CMPXCHG_LOCAL
+ local_irq_restore(flags);
+#endif
return;
slab_empty:
@@ -1876,6 +2084,9 @@ slab_empty:
stat(s, FREE_REMOVE_PARTIAL);
}
slab_unlock(page);
+#ifdef CONFIG_CMPXCHG_LOCAL
+ local_irq_restore(flags);
+#endif
stat(s, FREE_SLAB);
discard_slab(s, page);
return;
@@ -1902,23 +2113,56 @@ static __always_inline void slab_free(struct kmem_cache *s,
{
void **object = (void *)x;
struct kmem_cache_cpu *c;
+#ifdef CONFIG_CMPXCHG_LOCAL
+ unsigned long tid;
+#else
unsigned long flags;
+#endif
slab_free_hook(s, x);
+#ifndef CONFIG_CMPXCHG_LOCAL
local_irq_save(flags);
+
+#else
+redo:
+#endif
+
+ /*
+ * Determine the currently cpus per cpu slab.
+ * The cpu may change afterward. However that does not matter since
+ * data is retrieved via this pointer. If we are on the same cpu
+ * during the cmpxchg then the free will succedd.
+ */
c = __this_cpu_ptr(s->cpu_slab);
- slab_free_hook_irq(s, x);
+#ifdef CONFIG_CMPXCHG_LOCAL
+ tid = c->tid;
+ barrier();
+#endif
if (likely(page == c->page && c->node != NUMA_NO_NODE)) {
set_freepointer(s, object, c->freelist);
+
+#ifdef CONFIG_CMPXCHG_LOCAL
+ if (unlikely(!this_cpu_cmpxchg_double(
+ s->cpu_slab->freelist, s->cpu_slab->tid,
+ c->freelist, tid,
+ object, next_tid(tid)))) {
+
+ note_cmpxchg_failure("slab_free", s, tid);
+ goto redo;
+ }
+#else
c->freelist = object;
+#endif
stat(s, FREE_FASTPATH);
} else
__slab_free(s, page, x, addr);
+#ifndef CONFIG_CMPXCHG_LOCAL
local_irq_restore(flags);
+#endif
}
void kmem_cache_free(struct kmem_cache *s, void *x)
@@ -1988,13 +2232,13 @@ static int slub_nomerge;
* the smallest order which will fit the object.
*/
static inline int slab_order(int size, int min_objects,
- int max_order, int fract_leftover)
+ int max_order, int fract_leftover, int reserved)
{
int order;
int rem;
int min_order = slub_min_order;
- if ((PAGE_SIZE << min_order) / size > MAX_OBJS_PER_PAGE)
+ if (order_objects(min_order, size, reserved) > MAX_OBJS_PER_PAGE)
return get_order(size * MAX_OBJS_PER_PAGE) - 1;
for (order = max(min_order,
@@ -2003,10 +2247,10 @@ static inline int slab_order(int size, int min_objects,
unsigned long slab_size = PAGE_SIZE << order;
- if (slab_size < min_objects * size)
+ if (slab_size < min_objects * size + reserved)
continue;
- rem = slab_size % size;
+ rem = (slab_size - reserved) % size;
if (rem <= slab_size / fract_leftover)
break;
@@ -2016,7 +2260,7 @@ static inline int slab_order(int size, int min_objects,
return order;
}
-static inline int calculate_order(int size)
+static inline int calculate_order(int size, int reserved)
{
int order;
int min_objects;
@@ -2034,14 +2278,14 @@ static inline int calculate_order(int size)
min_objects = slub_min_objects;
if (!min_objects)
min_objects = 4 * (fls(nr_cpu_ids) + 1);
- max_objects = (PAGE_SIZE << slub_max_order)/size;
+ max_objects = order_objects(slub_max_order, size, reserved);
min_objects = min(min_objects, max_objects);
while (min_objects > 1) {
fraction = 16;
while (fraction >= 4) {
order = slab_order(size, min_objects,
- slub_max_order, fraction);
+ slub_max_order, fraction, reserved);
if (order <= slub_max_order)
return order;
fraction /= 2;
@@ -2053,14 +2297,14 @@ static inline int calculate_order(int size)
* We were unable to place multiple objects in a slab. Now
* lets see if we can place a single object there.
*/
- order = slab_order(size, 1, slub_max_order, 1);
+ order = slab_order(size, 1, slub_max_order, 1, reserved);
if (order <= slub_max_order)
return order;
/*
* Doh this slab cannot be placed using slub_max_order.
*/
- order = slab_order(size, 1, MAX_ORDER, 1);
+ order = slab_order(size, 1, MAX_ORDER, 1, reserved);
if (order < MAX_ORDER)
return order;
return -ENOSYS;
@@ -2110,9 +2354,23 @@ static inline int alloc_kmem_cache_cpus(struct kmem_cache *s)
BUILD_BUG_ON(PERCPU_DYNAMIC_EARLY_SIZE <
SLUB_PAGE_SHIFT * sizeof(struct kmem_cache_cpu));
+#ifdef CONFIG_CMPXCHG_LOCAL
+ /*
+ * Must align to double word boundary for the double cmpxchg instructions
+ * to work.
+ */
+ s->cpu_slab = __alloc_percpu(sizeof(struct kmem_cache_cpu), 2 * sizeof(void *));
+#else
+ /* Regular alignment is sufficient */
s->cpu_slab = alloc_percpu(struct kmem_cache_cpu);
+#endif
+
+ if (!s->cpu_slab)
+ return 0;
- return s->cpu_slab != NULL;
+ init_kmem_cache_cpus(s);
+
+ return 1;
}
static struct kmem_cache *kmem_cache_node;
@@ -2311,7 +2569,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
if (forced_order >= 0)
order = forced_order;
else
- order = calculate_order(size);
+ order = calculate_order(size, s->reserved);
if (order < 0)
return 0;
@@ -2329,8 +2587,8 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
/*
* Determine the number of objects per slab
*/
- s->oo = oo_make(order, size);
- s->min = oo_make(get_order(size), size);
+ s->oo = oo_make(order, size, s->reserved);
+ s->min = oo_make(get_order(size), size, s->reserved);
if (oo_objects(s->oo) > oo_objects(s->max))
s->max = s->oo;
@@ -2349,6 +2607,10 @@ static int kmem_cache_open(struct kmem_cache *s,
s->objsize = size;
s->align = align;
s->flags = kmem_cache_flags(size, flags, name, ctor);
+ s->reserved = 0;
+
+ if (need_reserve_slab_rcu && (s->flags & SLAB_DESTROY_BY_RCU))
+ s->reserved = sizeof(struct rcu_head);
if (!calculate_sizes(s, -1))
goto error;
@@ -2399,12 +2661,6 @@ unsigned int kmem_cache_size(struct kmem_cache *s)
}
EXPORT_SYMBOL(kmem_cache_size);
-const char *kmem_cache_name(struct kmem_cache *s)
-{
- return s->name;
-}
-EXPORT_SYMBOL(kmem_cache_name);
-
static void list_slab_objects(struct kmem_cache *s, struct page *page,
const char *text)
{
@@ -2696,7 +2952,6 @@ EXPORT_SYMBOL(__kmalloc_node);
size_t ksize(const void *object)
{
struct page *page;
- struct kmem_cache *s;
if (unlikely(object == ZERO_SIZE_PTR))
return 0;
@@ -2707,28 +2962,8 @@ size_t ksize(const void *object)
WARN_ON(!PageCompound(page));
return PAGE_SIZE << compound_order(page);
}
- s = page->slab;
-
-#ifdef CONFIG_SLUB_DEBUG
- /*
- * Debugging requires use of the padding between object
- * and whatever may come after it.
- */
- if (s->flags & (SLAB_RED_ZONE | SLAB_POISON))
- return s->objsize;
-#endif
- /*
- * If we have the need to store the freelist pointer
- * back there or track user information then we can
- * only use the space before that information.
- */
- if (s->flags & (SLAB_DESTROY_BY_RCU | SLAB_STORE_USER))
- return s->inuse;
- /*
- * Else we can use all the padding etc for the allocation
- */
- return s->size;
+ return slab_ksize(page->slab);
}
EXPORT_SYMBOL(ksize);
@@ -3312,7 +3547,7 @@ void *__kmalloc_track_caller(size_t size, gfp_t gfpflags, unsigned long caller)
ret = slab_alloc(s, gfpflags, NUMA_NO_NODE, caller);
- /* Honor the call site pointer we recieved. */
+ /* Honor the call site pointer we received. */
trace_kmalloc(caller, ret, size, s->size, gfpflags);
return ret;
@@ -3342,7 +3577,7 @@ void *__kmalloc_node_track_caller(size_t size, gfp_t gfpflags,
ret = slab_alloc(s, gfpflags, node, caller);
- /* Honor the call site pointer we recieved. */
+ /* Honor the call site pointer we received. */
trace_kmalloc_node(caller, ret, size, s->size, gfpflags, node);
return ret;
@@ -4017,6 +4252,12 @@ static ssize_t destroy_by_rcu_show(struct kmem_cache *s, char *buf)
}
SLAB_ATTR_RO(destroy_by_rcu);
+static ssize_t reserved_show(struct kmem_cache *s, char *buf)
+{
+ return sprintf(buf, "%d\n", s->reserved);
+}
+SLAB_ATTR_RO(reserved);
+
#ifdef CONFIG_SLUB_DEBUG
static ssize_t slabs_show(struct kmem_cache *s, char *buf)
{
@@ -4303,6 +4544,7 @@ static struct attribute *slab_attrs[] = {
&reclaim_account_attr.attr,
&destroy_by_rcu_attr.attr,
&shrink_attr.attr,
+ &reserved_attr.attr,
#ifdef CONFIG_SLUB_DEBUG
&total_objects_attr.attr,
&slabs_attr.attr,
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2024-12-27 18:05:54 +0000