2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* SLUB: A slab allocator that limits cache line use instead of queuing
|
|
|
|
* objects in per cpu and per node lists.
|
|
|
|
*
|
|
|
|
* The allocator synchronizes using per slab locks and only
|
|
|
|
* uses a centralized lock to manage a pool of partial slabs.
|
|
|
|
*
|
|
|
|
* (C) 2007 SGI, Christoph Lameter <clameter@sgi.com>
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/bit_spinlock.h>
|
|
|
|
#include <linux/interrupt.h>
|
|
|
|
#include <linux/bitops.h>
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/seq_file.h>
|
|
|
|
#include <linux/cpu.h>
|
|
|
|
#include <linux/cpuset.h>
|
|
|
|
#include <linux/mempolicy.h>
|
|
|
|
#include <linux/ctype.h>
|
|
|
|
#include <linux/kallsyms.h>
|
2007-10-22 01:41:37 +02:00
|
|
|
#include <linux/memory.h>
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Lock order:
|
|
|
|
* 1. slab_lock(page)
|
|
|
|
* 2. slab->list_lock
|
|
|
|
*
|
|
|
|
* The slab_lock protects operations on the object of a particular
|
|
|
|
* slab and its metadata in the page struct. If the slab lock
|
|
|
|
* has been taken then no allocations nor frees can be performed
|
|
|
|
* on the objects in the slab nor can the slab be added or removed
|
|
|
|
* from the partial or full lists since this would mean modifying
|
|
|
|
* the page_struct of the slab.
|
|
|
|
*
|
|
|
|
* The list_lock protects the partial and full list on each node and
|
|
|
|
* the partial slab counter. If taken then no new slabs may be added or
|
|
|
|
* removed from the lists nor make the number of partial slabs be modified.
|
|
|
|
* (Note that the total number of slabs is an atomic value that may be
|
|
|
|
* modified without taking the list lock).
|
|
|
|
*
|
|
|
|
* The list_lock is a centralized lock and thus we avoid taking it as
|
|
|
|
* much as possible. As long as SLUB does not have to handle partial
|
|
|
|
* slabs, operations can continue without any centralized lock. F.e.
|
|
|
|
* allocating a long series of objects that fill up slabs does not require
|
|
|
|
* the list lock.
|
|
|
|
*
|
|
|
|
* The lock order is sometimes inverted when we are trying to get a slab
|
|
|
|
* off a list. We take the list_lock and then look for a page on the list
|
|
|
|
* to use. While we do that objects in the slabs may be freed. We can
|
|
|
|
* only operate on the slab if we have also taken the slab_lock. So we use
|
|
|
|
* a slab_trylock() on the slab. If trylock was successful then no frees
|
|
|
|
* can occur anymore and we can use the slab for allocations etc. If the
|
|
|
|
* slab_trylock() does not succeed then frees are in progress in the slab and
|
|
|
|
* 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
|
|
|
|
* no danger of cacheline contention.
|
|
|
|
*
|
|
|
|
* Interrupts are disabled during allocation and deallocation in order to
|
|
|
|
* make the slab allocator safe to use in the context of an irq. In addition
|
|
|
|
* interrupts are disabled to ensure that the processor does not change
|
|
|
|
* while handling per_cpu slabs, due to kernel preemption.
|
|
|
|
*
|
|
|
|
* SLUB assigns one slab for allocation to each processor.
|
|
|
|
* Allocations only occur from these slabs called cpu slabs.
|
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Slabs with free elements are kept on a partial list and during regular
|
|
|
|
* operations no list for full slabs is used. If an object in a full slab is
|
2007-05-06 23:49:36 +02:00
|
|
|
* freed then the slab will show up again on the partial lists.
|
2007-05-09 11:32:39 +02:00
|
|
|
* We track full slabs for debugging purposes though because otherwise we
|
|
|
|
* cannot scan all objects.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
|
|
|
* Slabs are freed when they become empty. Teardown and setup is
|
|
|
|
* minimal so we rely on the page allocators per cpu caches for
|
|
|
|
* fast frees and allocs.
|
|
|
|
*
|
|
|
|
* Overloading of page flags that are otherwise used for LRU management.
|
|
|
|
*
|
2007-05-17 07:10:53 +02:00
|
|
|
* PageActive The slab is frozen and exempt from list processing.
|
|
|
|
* This means that the slab is dedicated to a purpose
|
|
|
|
* such as satisfying allocations for a specific
|
|
|
|
* processor. Objects may be freed in the slab while
|
|
|
|
* it is frozen but slab_free will then skip the usual
|
|
|
|
* list operations. It is up to the processor holding
|
|
|
|
* the slab to integrate the slab into the slab lists
|
|
|
|
* when the slab is no longer needed.
|
|
|
|
*
|
|
|
|
* One use of this flag is to mark slabs that are
|
|
|
|
* used for allocations. Then such a slab becomes a cpu
|
|
|
|
* slab. The cpu slab may be equipped with an additional
|
2007-10-16 10:26:05 +02:00
|
|
|
* freelist that allows lockless access to
|
2007-05-10 12:15:16 +02:00
|
|
|
* free objects in addition to the regular freelist
|
|
|
|
* that requires the slab lock.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
|
|
|
* PageError Slab requires special handling due to debug
|
|
|
|
* options set. This moves slab handling out of
|
2007-05-10 12:15:16 +02:00
|
|
|
* the fast path and disables lockless freelists.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
|
2007-05-17 07:10:56 +02:00
|
|
|
#define FROZEN (1 << PG_active)
|
|
|
|
|
|
|
|
#ifdef CONFIG_SLUB_DEBUG
|
|
|
|
#define SLABDEBUG (1 << PG_error)
|
|
|
|
#else
|
|
|
|
#define SLABDEBUG 0
|
|
|
|
#endif
|
|
|
|
|
2007-05-17 07:10:53 +02:00
|
|
|
static inline int SlabFrozen(struct page *page)
|
|
|
|
{
|
2007-05-17 07:10:56 +02:00
|
|
|
return page->flags & FROZEN;
|
2007-05-17 07:10:53 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void SetSlabFrozen(struct page *page)
|
|
|
|
{
|
2007-05-17 07:10:56 +02:00
|
|
|
page->flags |= FROZEN;
|
2007-05-17 07:10:53 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ClearSlabFrozen(struct page *page)
|
|
|
|
{
|
2007-05-17 07:10:56 +02:00
|
|
|
page->flags &= ~FROZEN;
|
2007-05-17 07:10:53 +02:00
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:42 +02:00
|
|
|
static inline int SlabDebug(struct page *page)
|
|
|
|
{
|
2007-05-17 07:10:56 +02:00
|
|
|
return page->flags & SLABDEBUG;
|
2007-05-09 11:32:42 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void SetSlabDebug(struct page *page)
|
|
|
|
{
|
2007-05-17 07:10:56 +02:00
|
|
|
page->flags |= SLABDEBUG;
|
2007-05-09 11:32:42 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ClearSlabDebug(struct page *page)
|
|
|
|
{
|
2007-05-17 07:10:56 +02:00
|
|
|
page->flags &= ~SLABDEBUG;
|
2007-05-09 11:32:42 +02:00
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Issues still to be resolved:
|
|
|
|
*
|
|
|
|
* - Support PAGE_ALLOC_DEBUG. Should be easy to do.
|
|
|
|
*
|
|
|
|
* - Variable sizing of the per node arrays
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* Enable to test recovery from slab corruption on boot */
|
|
|
|
#undef SLUB_RESILIENCY_TEST
|
|
|
|
|
|
|
|
#if PAGE_SHIFT <= 12
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Small page size. Make sure that we do not fragment memory
|
|
|
|
*/
|
|
|
|
#define DEFAULT_MAX_ORDER 1
|
|
|
|
#define DEFAULT_MIN_OBJECTS 4
|
|
|
|
|
|
|
|
#else
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Large page machines are customarily able to handle larger
|
|
|
|
* page orders.
|
|
|
|
*/
|
|
|
|
#define DEFAULT_MAX_ORDER 2
|
|
|
|
#define DEFAULT_MIN_OBJECTS 8
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
2007-05-06 23:49:46 +02:00
|
|
|
/*
|
|
|
|
* Mininum number of partial slabs. These will be left on the partial
|
|
|
|
* lists even if they are empty. kmem_cache_shrink may reclaim them.
|
|
|
|
*/
|
2007-12-21 23:37:37 +01:00
|
|
|
#define MIN_PARTIAL 5
|
2007-05-06 23:49:44 +02:00
|
|
|
|
2007-05-06 23:49:46 +02:00
|
|
|
/*
|
|
|
|
* Maximum number of desirable partial slabs.
|
|
|
|
* The existence of more partial slabs makes kmem_cache_shrink
|
|
|
|
* sort the partial list by the number of objects in the.
|
|
|
|
*/
|
|
|
|
#define MAX_PARTIAL 10
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#define DEBUG_DEFAULT_FLAGS (SLAB_DEBUG_FREE | SLAB_RED_ZONE | \
|
|
|
|
SLAB_POISON | SLAB_STORE_USER)
|
2007-05-09 11:32:39 +02:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Set of flags that will prevent slab merging
|
|
|
|
*/
|
|
|
|
#define SLUB_NEVER_MERGE (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER | \
|
|
|
|
SLAB_TRACE | SLAB_DESTROY_BY_RCU)
|
|
|
|
|
|
|
|
#define SLUB_MERGE_SAME (SLAB_DEBUG_FREE | SLAB_RECLAIM_ACCOUNT | \
|
|
|
|
SLAB_CACHE_DMA)
|
|
|
|
|
|
|
|
#ifndef ARCH_KMALLOC_MINALIGN
|
2007-05-06 23:49:37 +02:00
|
|
|
#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long)
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef ARCH_SLAB_MINALIGN
|
2007-05-06 23:49:37 +02:00
|
|
|
#define ARCH_SLAB_MINALIGN __alignof__(unsigned long long)
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Internal SLUB flags */
|
2007-08-08 00:11:48 +02:00
|
|
|
#define __OBJECT_POISON 0x80000000 /* Poison object */
|
|
|
|
#define __SYSFS_ADD_DEFERRED 0x40000000 /* Not yet visible via sysfs */
|
2008-02-14 23:28:01 +01:00
|
|
|
#define __KMALLOC_CACHE 0x20000000 /* objects freed using kfree */
|
|
|
|
#define __PAGE_ALLOC_FALLBACK 0x10000000 /* Allow fallback to page alloc */
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-05-09 11:32:35 +02:00
|
|
|
/* Not all arches define cache_line_size */
|
|
|
|
#ifndef cache_line_size
|
|
|
|
#define cache_line_size() L1_CACHE_BYTES
|
|
|
|
#endif
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
static int kmem_size = sizeof(struct kmem_cache);
|
|
|
|
|
|
|
|
#ifdef CONFIG_SMP
|
|
|
|
static struct notifier_block slab_notifier;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
static enum {
|
|
|
|
DOWN, /* No slab functionality available */
|
|
|
|
PARTIAL, /* kmem_cache_open() works but kmalloc does not */
|
2007-05-09 11:32:39 +02:00
|
|
|
UP, /* Everything works but does not show up in sysfs */
|
2007-05-06 23:49:36 +02:00
|
|
|
SYSFS /* Sysfs up */
|
|
|
|
} slab_state = DOWN;
|
|
|
|
|
|
|
|
/* A list of all slab caches on the system */
|
|
|
|
static DECLARE_RWSEM(slub_lock);
|
2007-07-17 13:03:27 +02:00
|
|
|
static LIST_HEAD(slab_caches);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-05-09 11:32:43 +02:00
|
|
|
/*
|
|
|
|
* Tracking user of a slab.
|
|
|
|
*/
|
|
|
|
struct track {
|
|
|
|
void *addr; /* Called from address */
|
|
|
|
int cpu; /* Was running on cpu */
|
|
|
|
int pid; /* Pid context */
|
|
|
|
unsigned long when; /* When did the operation occur */
|
|
|
|
};
|
|
|
|
|
|
|
|
enum track_item { TRACK_ALLOC, TRACK_FREE };
|
|
|
|
|
2007-05-09 11:32:44 +02:00
|
|
|
#if defined(CONFIG_SYSFS) && defined(CONFIG_SLUB_DEBUG)
|
2007-05-06 23:49:36 +02:00
|
|
|
static int sysfs_slab_add(struct kmem_cache *);
|
|
|
|
static int sysfs_slab_alias(struct kmem_cache *, const char *);
|
|
|
|
static void sysfs_slab_remove(struct kmem_cache *);
|
2008-02-08 02:47:41 +01:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#else
|
2007-07-17 13:03:24 +02:00
|
|
|
static inline int sysfs_slab_add(struct kmem_cache *s) { return 0; }
|
|
|
|
static inline int sysfs_slab_alias(struct kmem_cache *s, const char *p)
|
|
|
|
{ return 0; }
|
2008-01-08 07:29:05 +01:00
|
|
|
static inline void sysfs_slab_remove(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
kfree(s);
|
|
|
|
}
|
2008-02-08 02:47:41 +01:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
|
2008-02-08 02:47:41 +01:00
|
|
|
static inline void stat(struct kmem_cache_cpu *c, enum stat_item si)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_SLUB_STATS
|
|
|
|
c->stat[si]++;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/********************************************************************
|
|
|
|
* Core slab cache functions
|
|
|
|
*******************************************************************/
|
|
|
|
|
|
|
|
int slab_is_available(void)
|
|
|
|
{
|
|
|
|
return slab_state >= UP;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline struct kmem_cache_node *get_node(struct kmem_cache *s, int node)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
return s->node[node];
|
|
|
|
#else
|
|
|
|
return &s->local_node;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
static inline struct kmem_cache_cpu *get_cpu_slab(struct kmem_cache *s, int cpu)
|
|
|
|
{
|
2007-10-16 10:26:08 +02:00
|
|
|
#ifdef CONFIG_SMP
|
|
|
|
return s->cpu_slab[cpu];
|
|
|
|
#else
|
|
|
|
return &s->cpu_slab;
|
|
|
|
#endif
|
2007-10-16 10:26:05 +02:00
|
|
|
}
|
|
|
|
|
2008-02-16 08:45:26 +01:00
|
|
|
/* Verify that a pointer has an address that is valid within a slab page */
|
2007-05-09 11:32:43 +02:00
|
|
|
static inline int check_valid_pointer(struct kmem_cache *s,
|
|
|
|
struct page *page, const void *object)
|
|
|
|
{
|
|
|
|
void *base;
|
|
|
|
|
2008-03-01 22:40:44 +01:00
|
|
|
if (!object)
|
2007-05-09 11:32:43 +02:00
|
|
|
return 1;
|
|
|
|
|
2008-03-01 22:40:44 +01:00
|
|
|
base = page_address(page);
|
2007-05-09 11:32:43 +02:00
|
|
|
if (object < base || object >= base + s->objects * s->size ||
|
|
|
|
(object - base) % s->size) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:40 +02:00
|
|
|
/*
|
|
|
|
* Slow version of get and set free pointer.
|
|
|
|
*
|
|
|
|
* This version requires touching the cache lines of kmem_cache which
|
|
|
|
* we avoid to do in the fast alloc free paths. There we obtain the offset
|
|
|
|
* from the page struct.
|
|
|
|
*/
|
|
|
|
static inline void *get_freepointer(struct kmem_cache *s, void *object)
|
|
|
|
{
|
|
|
|
return *(void **)(object + s->offset);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_freepointer(struct kmem_cache *s, void *object, void *fp)
|
|
|
|
{
|
|
|
|
*(void **)(object + s->offset) = fp;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Loop over all objects in a slab */
|
|
|
|
#define for_each_object(__p, __s, __addr) \
|
|
|
|
for (__p = (__addr); __p < (__addr) + (__s)->objects * (__s)->size;\
|
|
|
|
__p += (__s)->size)
|
|
|
|
|
|
|
|
/* Scan freelist */
|
|
|
|
#define for_each_free_object(__p, __s, __free) \
|
2008-03-01 22:40:44 +01:00
|
|
|
for (__p = (__free); __p; __p = get_freepointer((__s), __p))
|
2007-05-09 11:32:40 +02:00
|
|
|
|
|
|
|
/* Determine object index from a given position */
|
|
|
|
static inline int slab_index(void *p, struct kmem_cache *s, void *addr)
|
|
|
|
{
|
|
|
|
return (p - addr) / s->size;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:44 +02:00
|
|
|
#ifdef CONFIG_SLUB_DEBUG
|
|
|
|
/*
|
|
|
|
* Debug settings:
|
|
|
|
*/
|
2007-07-16 08:38:14 +02:00
|
|
|
#ifdef CONFIG_SLUB_DEBUG_ON
|
|
|
|
static int slub_debug = DEBUG_DEFAULT_FLAGS;
|
|
|
|
#else
|
2007-05-09 11:32:44 +02:00
|
|
|
static int slub_debug;
|
2007-07-16 08:38:14 +02:00
|
|
|
#endif
|
2007-05-09 11:32:44 +02:00
|
|
|
|
|
|
|
static char *slub_debug_slabs;
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Object debugging
|
|
|
|
*/
|
|
|
|
static void print_section(char *text, u8 *addr, unsigned int length)
|
|
|
|
{
|
|
|
|
int i, offset;
|
|
|
|
int newline = 1;
|
|
|
|
char ascii[17];
|
|
|
|
|
|
|
|
ascii[16] = 0;
|
|
|
|
|
|
|
|
for (i = 0; i < length; i++) {
|
|
|
|
if (newline) {
|
2007-07-17 13:03:18 +02:00
|
|
|
printk(KERN_ERR "%8s 0x%p: ", text, addr + i);
|
2007-05-06 23:49:36 +02:00
|
|
|
newline = 0;
|
|
|
|
}
|
2008-01-08 08:20:27 +01:00
|
|
|
printk(KERN_CONT " %02x", addr[i]);
|
2007-05-06 23:49:36 +02:00
|
|
|
offset = i % 16;
|
|
|
|
ascii[offset] = isgraph(addr[i]) ? addr[i] : '.';
|
|
|
|
if (offset == 15) {
|
2008-01-08 08:20:27 +01:00
|
|
|
printk(KERN_CONT " %s\n", ascii);
|
2007-05-06 23:49:36 +02:00
|
|
|
newline = 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (!newline) {
|
|
|
|
i %= 16;
|
|
|
|
while (i < 16) {
|
2008-01-08 08:20:27 +01:00
|
|
|
printk(KERN_CONT " ");
|
2007-05-06 23:49:36 +02:00
|
|
|
ascii[i] = ' ';
|
|
|
|
i++;
|
|
|
|
}
|
2008-01-08 08:20:27 +01:00
|
|
|
printk(KERN_CONT " %s\n", ascii);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct track *get_track(struct kmem_cache *s, void *object,
|
|
|
|
enum track_item alloc)
|
|
|
|
{
|
|
|
|
struct track *p;
|
|
|
|
|
|
|
|
if (s->offset)
|
|
|
|
p = object + s->offset + sizeof(void *);
|
|
|
|
else
|
|
|
|
p = object + s->inuse;
|
|
|
|
|
|
|
|
return p + alloc;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void set_track(struct kmem_cache *s, void *object,
|
|
|
|
enum track_item alloc, void *addr)
|
|
|
|
{
|
|
|
|
struct track *p;
|
|
|
|
|
|
|
|
if (s->offset)
|
|
|
|
p = object + s->offset + sizeof(void *);
|
|
|
|
else
|
|
|
|
p = object + s->inuse;
|
|
|
|
|
|
|
|
p += alloc;
|
|
|
|
if (addr) {
|
|
|
|
p->addr = addr;
|
|
|
|
p->cpu = smp_processor_id();
|
|
|
|
p->pid = current ? current->pid : -1;
|
|
|
|
p->when = jiffies;
|
|
|
|
} else
|
|
|
|
memset(p, 0, sizeof(struct track));
|
|
|
|
}
|
|
|
|
|
|
|
|
static void init_tracking(struct kmem_cache *s, void *object)
|
|
|
|
{
|
2007-07-17 13:03:18 +02:00
|
|
|
if (!(s->flags & SLAB_STORE_USER))
|
|
|
|
return;
|
|
|
|
|
|
|
|
set_track(s, object, TRACK_FREE, NULL);
|
|
|
|
set_track(s, object, TRACK_ALLOC, NULL);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static void print_track(const char *s, struct track *t)
|
|
|
|
{
|
|
|
|
if (!t->addr)
|
|
|
|
return;
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
printk(KERN_ERR "INFO: %s in ", s);
|
2007-05-06 23:49:36 +02:00
|
|
|
__print_symbol("%s", (unsigned long)t->addr);
|
2007-07-17 13:03:18 +02:00
|
|
|
printk(" age=%lu cpu=%u pid=%d\n", jiffies - t->when, t->cpu, t->pid);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void print_tracking(struct kmem_cache *s, void *object)
|
|
|
|
{
|
|
|
|
if (!(s->flags & SLAB_STORE_USER))
|
|
|
|
return;
|
|
|
|
|
|
|
|
print_track("Allocated", get_track(s, object, TRACK_ALLOC));
|
|
|
|
print_track("Freed", get_track(s, object, TRACK_FREE));
|
|
|
|
}
|
|
|
|
|
|
|
|
static void print_page_info(struct page *page)
|
|
|
|
{
|
|
|
|
printk(KERN_ERR "INFO: Slab 0x%p used=%u fp=0x%p flags=0x%04lx\n",
|
|
|
|
page, page->inuse, page->freelist, page->flags);
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
static void slab_bug(struct kmem_cache *s, char *fmt, ...)
|
|
|
|
{
|
|
|
|
va_list args;
|
|
|
|
char buf[100];
|
|
|
|
|
|
|
|
va_start(args, fmt);
|
|
|
|
vsnprintf(buf, sizeof(buf), fmt, args);
|
|
|
|
va_end(args);
|
|
|
|
printk(KERN_ERR "========================================"
|
|
|
|
"=====================================\n");
|
|
|
|
printk(KERN_ERR "BUG %s: %s\n", s->name, buf);
|
|
|
|
printk(KERN_ERR "----------------------------------------"
|
|
|
|
"-------------------------------------\n\n");
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
static void slab_fix(struct kmem_cache *s, char *fmt, ...)
|
|
|
|
{
|
|
|
|
va_list args;
|
|
|
|
char buf[100];
|
|
|
|
|
|
|
|
va_start(args, fmt);
|
|
|
|
vsnprintf(buf, sizeof(buf), fmt, args);
|
|
|
|
va_end(args);
|
|
|
|
printk(KERN_ERR "FIX %s: %s\n", s->name, buf);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void print_trailer(struct kmem_cache *s, struct page *page, u8 *p)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
unsigned int off; /* Offset of last byte */
|
2008-03-01 22:40:44 +01:00
|
|
|
u8 *addr = page_address(page);
|
2007-07-17 13:03:18 +02:00
|
|
|
|
|
|
|
print_tracking(s, p);
|
|
|
|
|
|
|
|
print_page_info(page);
|
|
|
|
|
|
|
|
printk(KERN_ERR "INFO: Object 0x%p @offset=%tu fp=0x%p\n\n",
|
|
|
|
p, p - addr, get_freepointer(s, p));
|
|
|
|
|
|
|
|
if (p > addr + 16)
|
|
|
|
print_section("Bytes b4", p - 16, 16);
|
|
|
|
|
|
|
|
print_section("Object", p, min(s->objsize, 128));
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
if (s->flags & SLAB_RED_ZONE)
|
|
|
|
print_section("Redzone", p + s->objsize,
|
|
|
|
s->inuse - s->objsize);
|
|
|
|
|
|
|
|
if (s->offset)
|
|
|
|
off = s->offset + sizeof(void *);
|
|
|
|
else
|
|
|
|
off = s->inuse;
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
if (s->flags & SLAB_STORE_USER)
|
2007-05-06 23:49:36 +02:00
|
|
|
off += 2 * sizeof(struct track);
|
|
|
|
|
|
|
|
if (off != s->size)
|
|
|
|
/* Beginning of the filler is the free pointer */
|
2007-07-17 13:03:18 +02:00
|
|
|
print_section("Padding", p + off, s->size - off);
|
|
|
|
|
|
|
|
dump_stack();
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static void object_err(struct kmem_cache *s, struct page *page,
|
|
|
|
u8 *object, char *reason)
|
|
|
|
{
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_bug(s, reason);
|
|
|
|
print_trailer(s, page, object);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
static void slab_err(struct kmem_cache *s, struct page *page, char *fmt, ...)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
va_list args;
|
|
|
|
char buf[100];
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
va_start(args, fmt);
|
|
|
|
vsnprintf(buf, sizeof(buf), fmt, args);
|
2007-05-06 23:49:36 +02:00
|
|
|
va_end(args);
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_bug(s, fmt);
|
|
|
|
print_page_info(page);
|
2007-05-06 23:49:36 +02:00
|
|
|
dump_stack();
|
|
|
|
}
|
|
|
|
|
|
|
|
static void init_object(struct kmem_cache *s, void *object, int active)
|
|
|
|
{
|
|
|
|
u8 *p = object;
|
|
|
|
|
|
|
|
if (s->flags & __OBJECT_POISON) {
|
|
|
|
memset(p, POISON_FREE, s->objsize - 1);
|
2008-01-08 08:20:27 +01:00
|
|
|
p[s->objsize - 1] = POISON_END;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (s->flags & SLAB_RED_ZONE)
|
|
|
|
memset(p + s->objsize,
|
|
|
|
active ? SLUB_RED_ACTIVE : SLUB_RED_INACTIVE,
|
|
|
|
s->inuse - s->objsize);
|
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
static u8 *check_bytes(u8 *start, unsigned int value, unsigned int bytes)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
while (bytes) {
|
|
|
|
if (*start != (u8)value)
|
2007-07-17 13:03:18 +02:00
|
|
|
return start;
|
2007-05-06 23:49:36 +02:00
|
|
|
start++;
|
|
|
|
bytes--;
|
|
|
|
}
|
2007-07-17 13:03:18 +02:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void restore_bytes(struct kmem_cache *s, char *message, u8 data,
|
|
|
|
void *from, void *to)
|
|
|
|
{
|
|
|
|
slab_fix(s, "Restoring 0x%p-0x%p=0x%x\n", from, to - 1, data);
|
|
|
|
memset(from, data, to - from);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int check_bytes_and_report(struct kmem_cache *s, struct page *page,
|
|
|
|
u8 *object, char *what,
|
2008-01-08 08:20:27 +01:00
|
|
|
u8 *start, unsigned int value, unsigned int bytes)
|
2007-07-17 13:03:18 +02:00
|
|
|
{
|
|
|
|
u8 *fault;
|
|
|
|
u8 *end;
|
|
|
|
|
|
|
|
fault = check_bytes(start, value, bytes);
|
|
|
|
if (!fault)
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
end = start + bytes;
|
|
|
|
while (end > fault && end[-1] == value)
|
|
|
|
end--;
|
|
|
|
|
|
|
|
slab_bug(s, "%s overwritten", what);
|
|
|
|
printk(KERN_ERR "INFO: 0x%p-0x%p. First byte 0x%x instead of 0x%x\n",
|
|
|
|
fault, end - 1, fault[0], value);
|
|
|
|
print_trailer(s, page, object);
|
|
|
|
|
|
|
|
restore_bytes(s, what, value, fault, end);
|
|
|
|
return 0;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Object layout:
|
|
|
|
*
|
|
|
|
* object address
|
|
|
|
* Bytes of the object to be managed.
|
|
|
|
* If the freepointer may overlay the object then the free
|
|
|
|
* pointer is the first word of the object.
|
2007-05-09 11:32:39 +02:00
|
|
|
*
|
2007-05-06 23:49:36 +02:00
|
|
|
* Poisoning uses 0x6b (POISON_FREE) and the last byte is
|
|
|
|
* 0xa5 (POISON_END)
|
|
|
|
*
|
|
|
|
* object + s->objsize
|
|
|
|
* Padding to reach word boundary. This is also used for Redzoning.
|
2007-05-09 11:32:39 +02:00
|
|
|
* Padding is extended by another word if Redzoning is enabled and
|
|
|
|
* objsize == inuse.
|
|
|
|
*
|
2007-05-06 23:49:36 +02:00
|
|
|
* We fill with 0xbb (RED_INACTIVE) for inactive objects and with
|
|
|
|
* 0xcc (RED_ACTIVE) for objects in use.
|
|
|
|
*
|
|
|
|
* object + s->inuse
|
2007-05-09 11:32:39 +02:00
|
|
|
* Meta data starts here.
|
|
|
|
*
|
2007-05-06 23:49:36 +02:00
|
|
|
* A. Free pointer (if we cannot overwrite object on free)
|
|
|
|
* B. Tracking data for SLAB_STORE_USER
|
2007-05-09 11:32:39 +02:00
|
|
|
* C. Padding to reach required alignment boundary or at mininum
|
2008-02-16 08:45:26 +01:00
|
|
|
* one word if debugging is on to be able to detect writes
|
2007-05-09 11:32:39 +02:00
|
|
|
* before the word boundary.
|
|
|
|
*
|
|
|
|
* Padding is done using 0x5a (POISON_INUSE)
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
|
|
|
* object + s->size
|
2007-05-09 11:32:39 +02:00
|
|
|
* Nothing is used beyond s->size.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* If slabcaches are merged then the objsize and inuse boundaries are mostly
|
|
|
|
* ignored. And therefore no slab options that rely on these boundaries
|
2007-05-06 23:49:36 +02:00
|
|
|
* may be used with merged slabcaches.
|
|
|
|
*/
|
|
|
|
|
|
|
|
static int check_pad_bytes(struct kmem_cache *s, struct page *page, u8 *p)
|
|
|
|
{
|
|
|
|
unsigned long off = s->inuse; /* The end of info */
|
|
|
|
|
|
|
|
if (s->offset)
|
|
|
|
/* Freepointer is placed after the object. */
|
|
|
|
off += sizeof(void *);
|
|
|
|
|
|
|
|
if (s->flags & SLAB_STORE_USER)
|
|
|
|
/* We also have user information there */
|
|
|
|
off += 2 * sizeof(struct track);
|
|
|
|
|
|
|
|
if (s->size == off)
|
|
|
|
return 1;
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
return check_bytes_and_report(s, page, p, "Object padding",
|
|
|
|
p + off, POISON_INUSE, s->size - off);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int slab_pad_check(struct kmem_cache *s, struct page *page)
|
|
|
|
{
|
2007-07-17 13:03:18 +02:00
|
|
|
u8 *start;
|
|
|
|
u8 *fault;
|
|
|
|
u8 *end;
|
|
|
|
int length;
|
|
|
|
int remainder;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
if (!(s->flags & SLAB_POISON))
|
|
|
|
return 1;
|
|
|
|
|
2008-03-01 22:40:44 +01:00
|
|
|
start = page_address(page);
|
2007-07-17 13:03:18 +02:00
|
|
|
end = start + (PAGE_SIZE << s->order);
|
2007-05-06 23:49:36 +02:00
|
|
|
length = s->objects * s->size;
|
2007-07-17 13:03:18 +02:00
|
|
|
remainder = end - (start + length);
|
2007-05-06 23:49:36 +02:00
|
|
|
if (!remainder)
|
|
|
|
return 1;
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
fault = check_bytes(start + length, POISON_INUSE, remainder);
|
|
|
|
if (!fault)
|
|
|
|
return 1;
|
|
|
|
while (end > fault && end[-1] == POISON_INUSE)
|
|
|
|
end--;
|
|
|
|
|
|
|
|
slab_err(s, page, "Padding overwritten. 0x%p-0x%p", fault, end - 1);
|
|
|
|
print_section("Padding", start, length);
|
|
|
|
|
|
|
|
restore_bytes(s, "slab padding", POISON_INUSE, start, end);
|
|
|
|
return 0;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int check_object(struct kmem_cache *s, struct page *page,
|
|
|
|
void *object, int active)
|
|
|
|
{
|
|
|
|
u8 *p = object;
|
|
|
|
u8 *endobject = object + s->objsize;
|
|
|
|
|
|
|
|
if (s->flags & SLAB_RED_ZONE) {
|
|
|
|
unsigned int red =
|
|
|
|
active ? SLUB_RED_ACTIVE : SLUB_RED_INACTIVE;
|
|
|
|
|
2007-07-17 13:03:18 +02:00
|
|
|
if (!check_bytes_and_report(s, page, object, "Redzone",
|
|
|
|
endobject, red, s->inuse - s->objsize))
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
} else {
|
2008-02-06 02:57:39 +01:00
|
|
|
if ((s->flags & SLAB_POISON) && s->objsize < s->inuse) {
|
|
|
|
check_bytes_and_report(s, page, p, "Alignment padding",
|
|
|
|
endobject, POISON_INUSE, s->inuse - s->objsize);
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (s->flags & SLAB_POISON) {
|
|
|
|
if (!active && (s->flags & __OBJECT_POISON) &&
|
2007-07-17 13:03:18 +02:00
|
|
|
(!check_bytes_and_report(s, page, p, "Poison", p,
|
|
|
|
POISON_FREE, s->objsize - 1) ||
|
|
|
|
!check_bytes_and_report(s, page, p, "Poison",
|
2008-01-08 08:20:27 +01:00
|
|
|
p + s->objsize - 1, POISON_END, 1)))
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
/*
|
|
|
|
* check_pad_bytes cleans up on its own.
|
|
|
|
*/
|
|
|
|
check_pad_bytes(s, page, p);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!s->offset && active)
|
|
|
|
/*
|
|
|
|
* Object and freepointer overlap. Cannot check
|
|
|
|
* freepointer while object is allocated.
|
|
|
|
*/
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
/* Check free pointer validity */
|
|
|
|
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
|
|
|
|
* of the free objects in this slab. May cause
|
2007-05-09 11:32:39 +02:00
|
|
|
* another error because the object count is now wrong.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2008-03-01 22:40:44 +01:00
|
|
|
set_freepointer(s, p, NULL);
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int check_slab(struct kmem_cache *s, struct page *page)
|
|
|
|
{
|
|
|
|
VM_BUG_ON(!irqs_disabled());
|
|
|
|
|
|
|
|
if (!PageSlab(page)) {
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_err(s, page, "Not a valid slab page");
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (page->inuse > s->objects) {
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_err(s, page, "inuse %u > max %u",
|
|
|
|
s->name, page->inuse, s->objects);
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
/* Slab_pad_check fixes things up after itself */
|
|
|
|
slab_pad_check(s, page);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Determine if a certain object on a page is on the freelist. Must hold the
|
|
|
|
* slab lock to guarantee that the chains are in a consistent state.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static int on_freelist(struct kmem_cache *s, struct page *page, void *search)
|
|
|
|
{
|
|
|
|
int nr = 0;
|
|
|
|
void *fp = page->freelist;
|
|
|
|
void *object = NULL;
|
|
|
|
|
2008-03-01 22:40:44 +01:00
|
|
|
while (fp && nr <= s->objects) {
|
2007-05-06 23:49:36 +02:00
|
|
|
if (fp == search)
|
|
|
|
return 1;
|
|
|
|
if (!check_valid_pointer(s, page, fp)) {
|
|
|
|
if (object) {
|
|
|
|
object_err(s, page, object,
|
|
|
|
"Freechain corrupt");
|
2008-03-01 22:40:44 +01:00
|
|
|
set_freepointer(s, object, NULL);
|
2007-05-06 23:49:36 +02:00
|
|
|
break;
|
|
|
|
} else {
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_err(s, page, "Freepointer corrupt");
|
2008-03-01 22:40:44 +01:00
|
|
|
page->freelist = NULL;
|
2007-05-06 23:49:36 +02:00
|
|
|
page->inuse = s->objects;
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_fix(s, "Freelist cleared");
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
object = fp;
|
|
|
|
fp = get_freepointer(s, object);
|
|
|
|
nr++;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (page->inuse != s->objects - nr) {
|
2007-05-06 23:49:47 +02:00
|
|
|
slab_err(s, page, "Wrong object count. Counter is %d but "
|
2007-07-17 13:03:18 +02:00
|
|
|
"counted were %d", page->inuse, s->objects - nr);
|
2007-05-06 23:49:36 +02:00
|
|
|
page->inuse = s->objects - nr;
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_fix(s, "Object count adjusted.");
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
return search == NULL;
|
|
|
|
}
|
|
|
|
|
2007-05-17 07:11:00 +02:00
|
|
|
static void trace(struct kmem_cache *s, struct page *page, void *object, int alloc)
|
|
|
|
{
|
|
|
|
if (s->flags & SLAB_TRACE) {
|
|
|
|
printk(KERN_INFO "TRACE %s %s 0x%p inuse=%d fp=0x%p\n",
|
|
|
|
s->name,
|
|
|
|
alloc ? "alloc" : "free",
|
|
|
|
object, page->inuse,
|
|
|
|
page->freelist);
|
|
|
|
|
|
|
|
if (!alloc)
|
|
|
|
print_section("Object", (void *)object, s->objsize);
|
|
|
|
|
|
|
|
dump_stack();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:42 +02:00
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Tracking of fully allocated slabs for debugging purposes.
|
2007-05-06 23:49:42 +02:00
|
|
|
*/
|
2007-05-06 23:49:44 +02:00
|
|
|
static void add_full(struct kmem_cache_node *n, struct page *page)
|
2007-05-06 23:49:42 +02:00
|
|
|
{
|
|
|
|
spin_lock(&n->list_lock);
|
|
|
|
list_add(&page->lru, &n->full);
|
|
|
|
spin_unlock(&n->list_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void remove_full(struct kmem_cache *s, struct page *page)
|
|
|
|
{
|
|
|
|
struct kmem_cache_node *n;
|
|
|
|
|
|
|
|
if (!(s->flags & SLAB_STORE_USER))
|
|
|
|
return;
|
|
|
|
|
|
|
|
n = get_node(s, page_to_nid(page));
|
|
|
|
|
|
|
|
spin_lock(&n->list_lock);
|
|
|
|
list_del(&page->lru);
|
|
|
|
spin_unlock(&n->list_lock);
|
|
|
|
}
|
|
|
|
|
2007-05-17 07:11:00 +02:00
|
|
|
static void setup_object_debug(struct kmem_cache *s, struct page *page,
|
|
|
|
void *object)
|
|
|
|
{
|
|
|
|
if (!(s->flags & (SLAB_STORE_USER|SLAB_RED_ZONE|__OBJECT_POISON)))
|
|
|
|
return;
|
|
|
|
|
|
|
|
init_object(s, object, 0);
|
|
|
|
init_tracking(s, object);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int alloc_debug_processing(struct kmem_cache *s, struct page *page,
|
|
|
|
void *object, void *addr)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
if (!check_slab(s, page))
|
|
|
|
goto bad;
|
|
|
|
|
2008-02-16 08:45:24 +01:00
|
|
|
if (!on_freelist(s, page, object)) {
|
2007-07-17 13:03:18 +02:00
|
|
|
object_err(s, page, object, "Object already allocated");
|
2007-05-06 23:49:47 +02:00
|
|
|
goto bad;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (!check_valid_pointer(s, page, object)) {
|
|
|
|
object_err(s, page, object, "Freelist Pointer check fails");
|
2007-05-06 23:49:47 +02:00
|
|
|
goto bad;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2008-02-16 08:45:24 +01:00
|
|
|
if (!check_object(s, page, object, 0))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto bad;
|
|
|
|
|
2007-05-17 07:11:00 +02:00
|
|
|
/* Success perform special debug activities for allocs */
|
|
|
|
if (s->flags & SLAB_STORE_USER)
|
|
|
|
set_track(s, object, TRACK_ALLOC, addr);
|
|
|
|
trace(s, page, object, 1);
|
|
|
|
init_object(s, object, 1);
|
2007-05-06 23:49:36 +02:00
|
|
|
return 1;
|
2007-05-17 07:11:00 +02:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
bad:
|
|
|
|
if (PageSlab(page)) {
|
|
|
|
/*
|
|
|
|
* If this is a slab page then lets do the best we can
|
|
|
|
* to avoid issues in the future. Marking all objects
|
2007-05-09 11:32:39 +02:00
|
|
|
* as used avoids touching the remaining objects.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_fix(s, "Marking all objects used");
|
2007-05-06 23:49:36 +02:00
|
|
|
page->inuse = s->objects;
|
2008-03-01 22:40:44 +01:00
|
|
|
page->freelist = NULL;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-05-17 07:11:00 +02:00
|
|
|
static int free_debug_processing(struct kmem_cache *s, struct page *page,
|
|
|
|
void *object, void *addr)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
if (!check_slab(s, page))
|
|
|
|
goto fail;
|
|
|
|
|
|
|
|
if (!check_valid_pointer(s, page, object)) {
|
2007-05-06 23:49:47 +02:00
|
|
|
slab_err(s, page, "Invalid object pointer 0x%p", object);
|
2007-05-06 23:49:36 +02:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (on_freelist(s, page, object)) {
|
2007-07-17 13:03:18 +02:00
|
|
|
object_err(s, page, object, "Object already free");
|
2007-05-06 23:49:36 +02:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!check_object(s, page, object, 1))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (unlikely(s != page->slab)) {
|
2008-02-06 02:57:39 +01:00
|
|
|
if (!PageSlab(page)) {
|
2007-05-06 23:49:47 +02:00
|
|
|
slab_err(s, page, "Attempt to free object(0x%p) "
|
|
|
|
"outside of slab", object);
|
2008-02-06 02:57:39 +01:00
|
|
|
} else if (!page->slab) {
|
2007-05-06 23:49:36 +02:00
|
|
|
printk(KERN_ERR
|
2007-05-06 23:49:47 +02:00
|
|
|
"SLUB <none>: no slab for object 0x%p.\n",
|
2007-05-06 23:49:36 +02:00
|
|
|
object);
|
2007-05-06 23:49:47 +02:00
|
|
|
dump_stack();
|
2008-01-08 08:20:27 +01:00
|
|
|
} else
|
2007-07-17 13:03:18 +02:00
|
|
|
object_err(s, page, object,
|
|
|
|
"page slab pointer corrupt.");
|
2007-05-06 23:49:36 +02:00
|
|
|
goto fail;
|
|
|
|
}
|
2007-05-17 07:11:00 +02:00
|
|
|
|
|
|
|
/* Special debug activities for freeing objects */
|
2008-03-01 22:40:44 +01:00
|
|
|
if (!SlabFrozen(page) && !page->freelist)
|
2007-05-17 07:11:00 +02:00
|
|
|
remove_full(s, page);
|
|
|
|
if (s->flags & SLAB_STORE_USER)
|
|
|
|
set_track(s, object, TRACK_FREE, addr);
|
|
|
|
trace(s, page, object, 0);
|
|
|
|
init_object(s, object, 0);
|
2007-05-06 23:49:36 +02:00
|
|
|
return 1;
|
2007-05-17 07:11:00 +02:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
fail:
|
2007-07-17 13:03:18 +02:00
|
|
|
slab_fix(s, "Object at 0x%p not freed", object);
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:44 +02:00
|
|
|
static int __init setup_slub_debug(char *str)
|
|
|
|
{
|
2007-07-16 08:38:14 +02:00
|
|
|
slub_debug = DEBUG_DEFAULT_FLAGS;
|
|
|
|
if (*str++ != '=' || !*str)
|
|
|
|
/*
|
|
|
|
* No options specified. Switch on full debugging.
|
|
|
|
*/
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
if (*str == ',')
|
|
|
|
/*
|
|
|
|
* No options but restriction on slabs. This means full
|
|
|
|
* debugging for slabs matching a pattern.
|
|
|
|
*/
|
|
|
|
goto check_slabs;
|
|
|
|
|
|
|
|
slub_debug = 0;
|
|
|
|
if (*str == '-')
|
|
|
|
/*
|
|
|
|
* Switch off all debugging measures.
|
|
|
|
*/
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Determine which debug features should be switched on
|
|
|
|
*/
|
2008-01-08 08:20:27 +01:00
|
|
|
for (; *str && *str != ','; str++) {
|
2007-07-16 08:38:14 +02:00
|
|
|
switch (tolower(*str)) {
|
|
|
|
case 'f':
|
|
|
|
slub_debug |= SLAB_DEBUG_FREE;
|
|
|
|
break;
|
|
|
|
case 'z':
|
|
|
|
slub_debug |= SLAB_RED_ZONE;
|
|
|
|
break;
|
|
|
|
case 'p':
|
|
|
|
slub_debug |= SLAB_POISON;
|
|
|
|
break;
|
|
|
|
case 'u':
|
|
|
|
slub_debug |= SLAB_STORE_USER;
|
|
|
|
break;
|
|
|
|
case 't':
|
|
|
|
slub_debug |= SLAB_TRACE;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
printk(KERN_ERR "slub_debug option '%c' "
|
2008-01-08 08:20:27 +01:00
|
|
|
"unknown. skipped\n", *str);
|
2007-07-16 08:38:14 +02:00
|
|
|
}
|
2007-05-09 11:32:44 +02:00
|
|
|
}
|
|
|
|
|
2007-07-16 08:38:14 +02:00
|
|
|
check_slabs:
|
2007-05-09 11:32:44 +02:00
|
|
|
if (*str == ',')
|
|
|
|
slub_debug_slabs = str + 1;
|
2007-07-16 08:38:14 +02:00
|
|
|
out:
|
2007-05-09 11:32:44 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
__setup("slub_debug", setup_slub_debug);
|
|
|
|
|
2007-09-12 00:24:11 +02:00
|
|
|
static unsigned long kmem_cache_flags(unsigned long objsize,
|
|
|
|
unsigned long flags, const char *name,
|
2007-10-17 08:25:51 +02:00
|
|
|
void (*ctor)(struct kmem_cache *, void *))
|
2007-05-09 11:32:44 +02:00
|
|
|
{
|
|
|
|
/*
|
2008-02-16 08:45:24 +01:00
|
|
|
* Enable debugging if selected on the kernel commandline.
|
2007-05-09 11:32:44 +02:00
|
|
|
*/
|
2008-02-16 08:45:24 +01:00
|
|
|
if (slub_debug && (!slub_debug_slabs ||
|
|
|
|
strncmp(slub_debug_slabs, name, strlen(slub_debug_slabs)) == 0))
|
|
|
|
flags |= slub_debug;
|
2007-09-12 00:24:11 +02:00
|
|
|
|
|
|
|
return flags;
|
2007-05-09 11:32:44 +02:00
|
|
|
}
|
|
|
|
#else
|
2007-05-17 07:11:00 +02:00
|
|
|
static inline void setup_object_debug(struct kmem_cache *s,
|
|
|
|
struct page *page, void *object) {}
|
2007-05-09 11:32:44 +02:00
|
|
|
|
2007-05-17 07:11:00 +02:00
|
|
|
static inline int alloc_debug_processing(struct kmem_cache *s,
|
|
|
|
struct page *page, void *object, void *addr) { return 0; }
|
2007-05-09 11:32:44 +02:00
|
|
|
|
2007-05-17 07:11:00 +02:00
|
|
|
static inline int free_debug_processing(struct kmem_cache *s,
|
|
|
|
struct page *page, void *object, void *addr) { return 0; }
|
2007-05-09 11:32:44 +02:00
|
|
|
|
|
|
|
static inline int slab_pad_check(struct kmem_cache *s, struct page *page)
|
|
|
|
{ return 1; }
|
|
|
|
static inline int check_object(struct kmem_cache *s, struct page *page,
|
|
|
|
void *object, int active) { return 1; }
|
2007-05-17 07:11:00 +02:00
|
|
|
static inline void add_full(struct kmem_cache_node *n, struct page *page) {}
|
2007-09-12 00:24:11 +02:00
|
|
|
static inline unsigned long kmem_cache_flags(unsigned long objsize,
|
|
|
|
unsigned long flags, const char *name,
|
2007-10-17 08:25:51 +02:00
|
|
|
void (*ctor)(struct kmem_cache *, void *))
|
2007-09-12 00:24:11 +02:00
|
|
|
{
|
|
|
|
return flags;
|
|
|
|
}
|
2007-05-09 11:32:44 +02:00
|
|
|
#define slub_debug 0
|
|
|
|
#endif
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Slab allocation and freeing
|
|
|
|
*/
|
|
|
|
static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
|
|
|
|
{
|
2008-01-08 08:20:27 +01:00
|
|
|
struct page *page;
|
2007-05-06 23:49:36 +02:00
|
|
|
int pages = 1 << s->order;
|
|
|
|
|
2008-02-14 23:21:32 +01:00
|
|
|
flags |= s->allocflags;
|
2007-10-16 10:25:52 +02:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
if (node == -1)
|
|
|
|
page = alloc_pages(flags, s->order);
|
|
|
|
else
|
|
|
|
page = alloc_pages_node(node, flags, s->order);
|
|
|
|
|
|
|
|
if (!page)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
mod_zone_page_state(page_zone(page),
|
|
|
|
(s->flags & SLAB_RECLAIM_ACCOUNT) ?
|
|
|
|
NR_SLAB_RECLAIMABLE : NR_SLAB_UNRECLAIMABLE,
|
|
|
|
pages);
|
|
|
|
|
|
|
|
return page;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void setup_object(struct kmem_cache *s, struct page *page,
|
|
|
|
void *object)
|
|
|
|
{
|
2007-05-17 07:11:00 +02:00
|
|
|
setup_object_debug(s, page, object);
|
2007-05-06 23:50:17 +02:00
|
|
|
if (unlikely(s->ctor))
|
2007-10-17 08:25:51 +02:00
|
|
|
s->ctor(s, object);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
|
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
struct kmem_cache_node *n;
|
|
|
|
void *start;
|
|
|
|
void *last;
|
|
|
|
void *p;
|
|
|
|
|
2007-10-16 10:25:41 +02:00
|
|
|
BUG_ON(flags & GFP_SLAB_BUG_MASK);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:25:41 +02:00
|
|
|
page = allocate_slab(s,
|
|
|
|
flags & (GFP_RECLAIM_MASK | GFP_CONSTRAINT_MASK), node);
|
2007-05-06 23:49:36 +02:00
|
|
|
if (!page)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
n = get_node(s, page_to_nid(page));
|
|
|
|
if (n)
|
|
|
|
atomic_long_inc(&n->nr_slabs);
|
|
|
|
page->slab = s;
|
|
|
|
page->flags |= 1 << PG_slab;
|
|
|
|
if (s->flags & (SLAB_DEBUG_FREE | SLAB_RED_ZONE | SLAB_POISON |
|
|
|
|
SLAB_STORE_USER | SLAB_TRACE))
|
2007-05-09 11:32:42 +02:00
|
|
|
SetSlabDebug(page);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
start = page_address(page);
|
|
|
|
|
|
|
|
if (unlikely(s->flags & SLAB_POISON))
|
|
|
|
memset(start, POISON_INUSE, PAGE_SIZE << s->order);
|
|
|
|
|
|
|
|
last = start;
|
2007-05-09 11:32:40 +02:00
|
|
|
for_each_object(p, s, start) {
|
2007-05-06 23:49:36 +02:00
|
|
|
setup_object(s, page, last);
|
|
|
|
set_freepointer(s, last, p);
|
|
|
|
last = p;
|
|
|
|
}
|
|
|
|
setup_object(s, page, last);
|
2008-03-01 22:40:44 +01:00
|
|
|
set_freepointer(s, last, NULL);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
page->freelist = start;
|
|
|
|
page->inuse = 0;
|
|
|
|
out:
|
|
|
|
return page;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __free_slab(struct kmem_cache *s, struct page *page)
|
|
|
|
{
|
|
|
|
int pages = 1 << s->order;
|
|
|
|
|
2007-05-17 07:10:50 +02:00
|
|
|
if (unlikely(SlabDebug(page))) {
|
2007-05-06 23:49:36 +02:00
|
|
|
void *p;
|
|
|
|
|
|
|
|
slab_pad_check(s, page);
|
2008-03-01 22:40:44 +01:00
|
|
|
for_each_object(p, s, page_address(page))
|
2007-05-06 23:49:36 +02:00
|
|
|
check_object(s, page, p, 0);
|
2007-07-26 20:54:34 +02:00
|
|
|
ClearSlabDebug(page);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
mod_zone_page_state(page_zone(page),
|
|
|
|
(s->flags & SLAB_RECLAIM_ACCOUNT) ?
|
|
|
|
NR_SLAB_RECLAIMABLE : NR_SLAB_UNRECLAIMABLE,
|
2008-01-08 08:20:27 +01:00
|
|
|
-pages);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
__free_pages(page, s->order);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void rcu_free_slab(struct rcu_head *h)
|
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
|
|
|
|
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;
|
|
|
|
|
|
|
|
call_rcu(head, rcu_free_slab);
|
|
|
|
} else
|
|
|
|
__free_slab(s, page);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void discard_slab(struct kmem_cache *s, struct page *page)
|
|
|
|
{
|
|
|
|
struct kmem_cache_node *n = get_node(s, page_to_nid(page));
|
|
|
|
|
|
|
|
atomic_long_dec(&n->nr_slabs);
|
|
|
|
reset_page_mapcount(page);
|
2007-05-09 11:32:42 +02:00
|
|
|
__ClearPageSlab(page);
|
2007-05-06 23:49:36 +02:00
|
|
|
free_slab(s, page);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Per slab locking using the pagelock
|
|
|
|
*/
|
|
|
|
static __always_inline void slab_lock(struct page *page)
|
|
|
|
{
|
|
|
|
bit_spin_lock(PG_locked, &page->flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline void slab_unlock(struct page *page)
|
|
|
|
{
|
2008-01-08 08:20:27 +01:00
|
|
|
__bit_spin_unlock(PG_locked, &page->flags);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline int slab_trylock(struct page *page)
|
|
|
|
{
|
|
|
|
int rc = 1;
|
|
|
|
|
|
|
|
rc = bit_spin_trylock(PG_locked, &page->flags);
|
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Management of partially allocated slabs
|
|
|
|
*/
|
2008-01-08 08:20:27 +01:00
|
|
|
static void add_partial(struct kmem_cache_node *n,
|
|
|
|
struct page *page, int tail)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
2007-05-06 23:49:44 +02:00
|
|
|
spin_lock(&n->list_lock);
|
|
|
|
n->nr_partial++;
|
2008-01-08 08:20:27 +01:00
|
|
|
if (tail)
|
|
|
|
list_add_tail(&page->lru, &n->partial);
|
|
|
|
else
|
|
|
|
list_add(&page->lru, &n->partial);
|
2007-05-06 23:49:36 +02:00
|
|
|
spin_unlock(&n->list_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void remove_partial(struct kmem_cache *s,
|
|
|
|
struct page *page)
|
|
|
|
{
|
|
|
|
struct kmem_cache_node *n = get_node(s, page_to_nid(page));
|
|
|
|
|
|
|
|
spin_lock(&n->list_lock);
|
|
|
|
list_del(&page->lru);
|
|
|
|
n->nr_partial--;
|
|
|
|
spin_unlock(&n->list_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Lock slab and remove from the partial list.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Must hold list_lock.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2007-05-17 07:10:53 +02:00
|
|
|
static inline int lock_and_freeze_slab(struct kmem_cache_node *n, struct page *page)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
if (slab_trylock(page)) {
|
|
|
|
list_del(&page->lru);
|
|
|
|
n->nr_partial--;
|
2007-05-17 07:10:53 +02:00
|
|
|
SetSlabFrozen(page);
|
2007-05-06 23:49:36 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Try to allocate a partial slab from a specific node.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static struct page *get_partial_node(struct kmem_cache_node *n)
|
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Racy check. If we mistakenly see no partial slabs then we
|
|
|
|
* just allocate an empty slab. If we mistakenly try to get a
|
2007-05-09 11:32:39 +02:00
|
|
|
* partial slab and there is none available then get_partials()
|
|
|
|
* will return NULL.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
if (!n || !n->nr_partial)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
spin_lock(&n->list_lock);
|
|
|
|
list_for_each_entry(page, &n->partial, lru)
|
2007-05-17 07:10:53 +02:00
|
|
|
if (lock_and_freeze_slab(n, page))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto out;
|
|
|
|
page = NULL;
|
|
|
|
out:
|
|
|
|
spin_unlock(&n->list_lock);
|
|
|
|
return page;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Get a page from somewhere. Search in increasing NUMA distances.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static struct page *get_any_partial(struct kmem_cache *s, gfp_t flags)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
struct zonelist *zonelist;
|
|
|
|
struct zone **z;
|
|
|
|
struct page *page;
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* The defrag ratio allows a configuration of the tradeoffs between
|
|
|
|
* inter node defragmentation and node local allocations. A lower
|
|
|
|
* defrag_ratio increases the tendency to do local allocations
|
|
|
|
* instead of attempting to obtain partial slabs from other nodes.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* If the defrag_ratio is set to 0 then kmalloc() always
|
|
|
|
* returns node local objects. If the ratio is higher then kmalloc()
|
|
|
|
* may return off node objects because partial slabs are obtained
|
|
|
|
* from other nodes and filled up.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2008-02-16 08:45:26 +01:00
|
|
|
* If /sys/kernel/slab/xx/defrag_ratio is set to 100 (which makes
|
2007-05-09 11:32:39 +02:00
|
|
|
* defrag_ratio = 1000) then every (well almost) allocation will
|
|
|
|
* first attempt to defrag slab caches on other nodes. This means
|
|
|
|
* scanning over all nodes to look for partial slabs which may be
|
|
|
|
* expensive if we do it every time we are trying to find a slab
|
|
|
|
* with available objects.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2008-01-08 08:20:26 +01:00
|
|
|
if (!s->remote_node_defrag_ratio ||
|
|
|
|
get_cycles() % 1024 > s->remote_node_defrag_ratio)
|
2007-05-06 23:49:36 +02:00
|
|
|
return NULL;
|
|
|
|
|
2008-02-06 02:57:39 +01:00
|
|
|
zonelist = &NODE_DATA(
|
|
|
|
slab_node(current->mempolicy))->node_zonelists[gfp_zone(flags)];
|
2007-05-06 23:49:36 +02:00
|
|
|
for (z = zonelist->zones; *z; z++) {
|
|
|
|
struct kmem_cache_node *n;
|
|
|
|
|
|
|
|
n = get_node(s, zone_to_nid(*z));
|
|
|
|
|
|
|
|
if (n && cpuset_zone_allowed_hardwall(*z, flags) &&
|
2007-05-06 23:49:44 +02:00
|
|
|
n->nr_partial > MIN_PARTIAL) {
|
2007-05-06 23:49:36 +02:00
|
|
|
page = get_partial_node(n);
|
|
|
|
if (page)
|
|
|
|
return page;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Get a partial page, lock it and return it.
|
|
|
|
*/
|
|
|
|
static struct page *get_partial(struct kmem_cache *s, gfp_t flags, int node)
|
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
int searchnode = (node == -1) ? numa_node_id() : node;
|
|
|
|
|
|
|
|
page = get_partial_node(get_node(s, searchnode));
|
|
|
|
if (page || (flags & __GFP_THISNODE))
|
|
|
|
return page;
|
|
|
|
|
|
|
|
return get_any_partial(s, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Move a page back to the lists.
|
|
|
|
*
|
|
|
|
* Must be called with the slab lock held.
|
|
|
|
*
|
|
|
|
* On exit the slab lock will have been dropped.
|
|
|
|
*/
|
2008-01-08 08:20:27 +01:00
|
|
|
static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
2007-05-06 23:49:44 +02:00
|
|
|
struct kmem_cache_node *n = get_node(s, page_to_nid(page));
|
2008-02-08 02:47:41 +01:00
|
|
|
struct kmem_cache_cpu *c = get_cpu_slab(s, smp_processor_id());
|
2007-05-06 23:49:44 +02:00
|
|
|
|
2007-05-17 07:10:53 +02:00
|
|
|
ClearSlabFrozen(page);
|
2007-05-06 23:49:36 +02:00
|
|
|
if (page->inuse) {
|
2007-05-06 23:49:44 +02:00
|
|
|
|
2008-03-01 22:40:44 +01:00
|
|
|
if (page->freelist) {
|
2008-01-08 08:20:27 +01:00
|
|
|
add_partial(n, page, tail);
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, tail ? DEACTIVATE_TO_TAIL : DEACTIVATE_TO_HEAD);
|
|
|
|
} else {
|
|
|
|
stat(c, DEACTIVATE_FULL);
|
|
|
|
if (SlabDebug(page) && (s->flags & SLAB_STORE_USER))
|
|
|
|
add_full(n, page);
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
slab_unlock(page);
|
|
|
|
} else {
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, DEACTIVATE_EMPTY);
|
2007-05-06 23:49:44 +02:00
|
|
|
if (n->nr_partial < MIN_PARTIAL) {
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Adding an empty slab to the partial slabs in order
|
|
|
|
* to avoid page allocator overhead. This slab needs
|
|
|
|
* to come after the other slabs with objects in
|
2008-02-16 08:45:26 +01:00
|
|
|
* so that the others get filled first. That way the
|
|
|
|
* size of the partial list stays small.
|
|
|
|
*
|
|
|
|
* kmem_cache_shrink can reclaim any empty slabs from the
|
|
|
|
* partial list.
|
2007-05-06 23:49:44 +02:00
|
|
|
*/
|
2008-01-08 08:20:27 +01:00
|
|
|
add_partial(n, page, 1);
|
2007-05-06 23:49:44 +02:00
|
|
|
slab_unlock(page);
|
|
|
|
} else {
|
|
|
|
slab_unlock(page);
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(get_cpu_slab(s, raw_smp_processor_id()), FREE_SLAB);
|
2007-05-06 23:49:44 +02:00
|
|
|
discard_slab(s, page);
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Remove the cpu slab
|
|
|
|
*/
|
2007-10-16 10:26:05 +02:00
|
|
|
static void deactivate_slab(struct kmem_cache *s, struct kmem_cache_cpu *c)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
2007-10-16 10:26:05 +02:00
|
|
|
struct page *page = c->page;
|
2008-01-08 08:20:27 +01:00
|
|
|
int tail = 1;
|
2008-02-08 02:47:41 +01:00
|
|
|
|
2008-03-04 20:10:17 +01:00
|
|
|
if (page->freelist)
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, DEACTIVATE_REMOTE_FREES);
|
2007-05-10 12:15:16 +02:00
|
|
|
/*
|
2008-02-16 08:45:26 +01:00
|
|
|
* Merge cpu freelist into slab freelist. Typically we get here
|
2007-05-10 12:15:16 +02:00
|
|
|
* because both freelists are empty. So this is unlikely
|
|
|
|
* to occur.
|
|
|
|
*/
|
2008-03-01 22:40:44 +01:00
|
|
|
while (unlikely(c->freelist)) {
|
2007-05-10 12:15:16 +02:00
|
|
|
void **object;
|
|
|
|
|
2008-01-08 08:20:27 +01:00
|
|
|
tail = 0; /* Hot objects. Put the slab first */
|
|
|
|
|
2007-05-10 12:15:16 +02:00
|
|
|
/* Retrieve object from cpu_freelist */
|
2007-10-16 10:26:05 +02:00
|
|
|
object = c->freelist;
|
2007-10-16 10:26:06 +02:00
|
|
|
c->freelist = c->freelist[c->offset];
|
2007-05-10 12:15:16 +02:00
|
|
|
|
|
|
|
/* And put onto the regular freelist */
|
2007-10-16 10:26:06 +02:00
|
|
|
object[c->offset] = page->freelist;
|
2007-05-10 12:15:16 +02:00
|
|
|
page->freelist = object;
|
|
|
|
page->inuse--;
|
|
|
|
}
|
2007-10-16 10:26:05 +02:00
|
|
|
c->page = NULL;
|
2008-01-08 08:20:27 +01:00
|
|
|
unfreeze_slab(s, page, tail);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
static inline void flush_slab(struct kmem_cache *s, struct kmem_cache_cpu *c)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, CPUSLAB_FLUSH);
|
2007-10-16 10:26:05 +02:00
|
|
|
slab_lock(c->page);
|
|
|
|
deactivate_slab(s, c);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Flush cpu slab.
|
2008-02-16 08:45:26 +01:00
|
|
|
*
|
2007-05-06 23:49:36 +02:00
|
|
|
* Called from IPI handler with interrupts disabled.
|
|
|
|
*/
|
2007-07-17 13:03:24 +02:00
|
|
|
static inline void __flush_cpu_slab(struct kmem_cache *s, int cpu)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
2007-10-16 10:26:05 +02:00
|
|
|
struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
if (likely(c && c->page))
|
|
|
|
flush_slab(s, c);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static void flush_cpu_slab(void *d)
|
|
|
|
{
|
|
|
|
struct kmem_cache *s = d;
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
__flush_cpu_slab(s, smp_processor_id());
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static void flush_all(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_SMP
|
|
|
|
on_each_cpu(flush_cpu_slab, s, 1, 1);
|
|
|
|
#else
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
local_irq_save(flags);
|
|
|
|
flush_cpu_slab(s);
|
|
|
|
local_irq_restore(flags);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
/*
|
|
|
|
* Check if the objects in a per cpu structure fit numa
|
|
|
|
* locality expectations.
|
|
|
|
*/
|
|
|
|
static inline int node_match(struct kmem_cache_cpu *c, int node)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
if (node != -1 && c->node != node)
|
|
|
|
return 0;
|
|
|
|
#endif
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
2007-05-10 12:15:16 +02:00
|
|
|
* Slow path. The lockless freelist is empty or we need to perform
|
|
|
|
* debugging duties.
|
|
|
|
*
|
|
|
|
* Interrupts are disabled.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-10 12:15:16 +02:00
|
|
|
* Processing is still very fast if new objects have been freed to the
|
|
|
|
* regular freelist. In that case we simply take over the regular freelist
|
|
|
|
* as the lockless freelist and zap the regular freelist.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-10 12:15:16 +02:00
|
|
|
* If that is not working then we fall back to the partial lists. We take the
|
|
|
|
* first element of the freelist as the object to allocate now and move the
|
|
|
|
* rest of the freelist to the lockless freelist.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-10 12:15:16 +02:00
|
|
|
* And if we were unable to get a new slab from the partial slab lists then
|
2008-02-16 08:45:26 +01:00
|
|
|
* 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.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2007-05-10 12:15:16 +02:00
|
|
|
static void *__slab_alloc(struct kmem_cache *s,
|
2007-10-16 10:26:05 +02:00
|
|
|
gfp_t gfpflags, int node, void *addr, struct kmem_cache_cpu *c)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
void **object;
|
2007-10-16 10:26:05 +02:00
|
|
|
struct page *new;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2008-03-28 04:56:33 +01:00
|
|
|
/* We handle __GFP_ZERO in the caller */
|
|
|
|
gfpflags &= ~__GFP_ZERO;
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
if (!c->page)
|
2007-05-06 23:49:36 +02:00
|
|
|
goto new_slab;
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
slab_lock(c->page);
|
|
|
|
if (unlikely(!node_match(c, node)))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto another_slab;
|
2008-02-16 08:45:26 +01:00
|
|
|
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, ALLOC_REFILL);
|
2008-02-16 08:45:26 +01:00
|
|
|
|
2007-05-10 12:15:16 +02:00
|
|
|
load_freelist:
|
2007-10-16 10:26:05 +02:00
|
|
|
object = c->page->freelist;
|
2008-03-01 22:40:44 +01:00
|
|
|
if (unlikely(!object))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto another_slab;
|
2007-10-16 10:26:05 +02:00
|
|
|
if (unlikely(SlabDebug(c->page)))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto debug;
|
|
|
|
|
2007-10-16 10:26:06 +02:00
|
|
|
c->freelist = object[c->offset];
|
2007-10-16 10:26:05 +02:00
|
|
|
c->page->inuse = s->objects;
|
2008-03-01 22:40:44 +01:00
|
|
|
c->page->freelist = NULL;
|
2007-10-16 10:26:05 +02:00
|
|
|
c->node = page_to_nid(c->page);
|
2008-01-08 08:20:30 +01:00
|
|
|
unlock_out:
|
2007-10-16 10:26:05 +02:00
|
|
|
slab_unlock(c->page);
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, ALLOC_SLOWPATH);
|
2007-05-06 23:49:36 +02:00
|
|
|
return object;
|
|
|
|
|
|
|
|
another_slab:
|
2007-10-16 10:26:05 +02:00
|
|
|
deactivate_slab(s, c);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
new_slab:
|
2007-10-16 10:26:05 +02:00
|
|
|
new = get_partial(s, gfpflags, node);
|
|
|
|
if (new) {
|
|
|
|
c->page = new;
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, ALLOC_FROM_PARTIAL);
|
2007-05-10 12:15:16 +02:00
|
|
|
goto load_freelist;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2007-10-17 08:25:51 +02:00
|
|
|
if (gfpflags & __GFP_WAIT)
|
|
|
|
local_irq_enable();
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
new = new_slab(s, gfpflags, node);
|
2007-10-17 08:25:51 +02:00
|
|
|
|
|
|
|
if (gfpflags & __GFP_WAIT)
|
|
|
|
local_irq_disable();
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
if (new) {
|
|
|
|
c = get_cpu_slab(s, smp_processor_id());
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, ALLOC_SLAB);
|
2007-11-05 20:31:58 +01:00
|
|
|
if (c->page)
|
2007-10-16 10:26:05 +02:00
|
|
|
flush_slab(s, c);
|
|
|
|
slab_lock(new);
|
|
|
|
SetSlabFrozen(new);
|
|
|
|
c->page = new;
|
2007-05-17 07:10:53 +02:00
|
|
|
goto load_freelist;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
Revert "SLUB: Alternate fast paths using cmpxchg_local"
This reverts commit 1f84260c8ce3b1ce26d4c1d6dedc2f33a3a29c0c, which is
suspected to be the reason for some very occasional and hard-to-trigger
crashes that usually look related to memory allocation (mostly reported
in networking, but since that's generally the most common source of
shortlived allocations - and allocations in interrupt contexts - that in
itself is not a big clue).
See for example
http://bugzilla.kernel.org/show_bug.cgi?id=9973
http://lkml.org/lkml/2008/2/19/278
etc.
One promising suspicion for what the root cause of bug is (which also
explains why it's so hard to trigger in practice) came from Eric
Dumazet:
"I wonder how SLUB_FASTPATH is supposed to work, since it is affected
by a classical ABA problem of lockless algo.
cmpxchg_local(&c->freelist, object, object[c->offset]) can succeed,
while an interrupt came (on this cpu), and several allocations were
done, and one free was performed at the end of this interruption, so
'object' was recycled.
c->freelist can then contain the previous value (object), but
object[c->offset] was changed by IRQ.
We then put back in freelist an already allocated object."
but another reason for the revert is simply that everybody agrees that
this code was the main suspect just by virtue of the pattern of oopses.
Cc: Torsten Kaiser <just.for.lkml@googlemail.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-19 18:08:49 +01:00
|
|
|
|
2008-02-14 23:28:01 +01:00
|
|
|
/*
|
|
|
|
* No memory available.
|
|
|
|
*
|
|
|
|
* If the slab uses higher order allocs but the object is
|
|
|
|
* smaller than a page size then we can fallback in emergencies
|
|
|
|
* to the page allocator via kmalloc_large. The page allocator may
|
|
|
|
* have failed to obtain a higher order page and we can try to
|
|
|
|
* allocate a single page if the object fits into a single page.
|
|
|
|
* That is only possible if certain conditions are met that are being
|
|
|
|
* checked when a slab is created.
|
|
|
|
*/
|
2008-03-13 07:57:49 +01:00
|
|
|
if (!(gfpflags & __GFP_NORETRY) &&
|
|
|
|
(s->flags & __PAGE_ALLOC_FALLBACK)) {
|
|
|
|
if (gfpflags & __GFP_WAIT)
|
|
|
|
local_irq_enable();
|
|
|
|
object = kmalloc_large(s->objsize, gfpflags);
|
|
|
|
if (gfpflags & __GFP_WAIT)
|
|
|
|
local_irq_disable();
|
|
|
|
return object;
|
|
|
|
}
|
2008-02-14 23:28:01 +01:00
|
|
|
return NULL;
|
2007-05-06 23:49:36 +02:00
|
|
|
debug:
|
2007-10-16 10:26:05 +02:00
|
|
|
if (!alloc_debug_processing(s, c->page, object, addr))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto another_slab;
|
2007-05-10 12:15:16 +02:00
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
c->page->inuse++;
|
2007-10-16 10:26:06 +02:00
|
|
|
c->page->freelist = object[c->offset];
|
2007-10-16 10:26:07 +02:00
|
|
|
c->node = -1;
|
2008-01-08 08:20:30 +01:00
|
|
|
goto unlock_out;
|
2007-05-10 12:15:16 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Inlined fastpath so that allocation functions (kmalloc, kmem_cache_alloc)
|
|
|
|
* have the fastpath folded into their functions. So no function call
|
|
|
|
* overhead for requests that can be satisfied on the fastpath.
|
|
|
|
*
|
|
|
|
* The fastpath works by first checking if the lockless freelist can be used.
|
|
|
|
* If not then __slab_alloc is called for slow processing.
|
|
|
|
*
|
|
|
|
* Otherwise we can simply pick the next object from the lockless free list.
|
|
|
|
*/
|
2008-01-08 08:20:27 +01:00
|
|
|
static __always_inline void *slab_alloc(struct kmem_cache *s,
|
2007-07-17 13:03:28 +02:00
|
|
|
gfp_t gfpflags, int node, void *addr)
|
2007-05-10 12:15:16 +02:00
|
|
|
{
|
|
|
|
void **object;
|
2007-10-16 10:26:05 +02:00
|
|
|
struct kmem_cache_cpu *c;
|
2008-01-08 08:20:30 +01:00
|
|
|
unsigned long flags;
|
|
|
|
|
2007-05-10 12:15:16 +02:00
|
|
|
local_irq_save(flags);
|
2007-10-16 10:26:05 +02:00
|
|
|
c = get_cpu_slab(s, smp_processor_id());
|
2008-03-01 22:40:44 +01:00
|
|
|
if (unlikely(!c->freelist || !node_match(c, node)))
|
2007-05-10 12:15:16 +02:00
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
object = __slab_alloc(s, gfpflags, node, addr, c);
|
2007-05-10 12:15:16 +02:00
|
|
|
|
|
|
|
else {
|
2007-10-16 10:26:05 +02:00
|
|
|
object = c->freelist;
|
2007-10-16 10:26:06 +02:00
|
|
|
c->freelist = object[c->offset];
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, ALLOC_FASTPATH);
|
2007-05-10 12:15:16 +02:00
|
|
|
}
|
|
|
|
local_irq_restore(flags);
|
2007-07-17 13:03:23 +02:00
|
|
|
|
|
|
|
if (unlikely((gfpflags & __GFP_ZERO) && object))
|
2007-10-16 10:26:09 +02:00
|
|
|
memset(object, 0, c->objsize);
|
2007-07-17 13:03:23 +02:00
|
|
|
|
2007-05-10 12:15:16 +02:00
|
|
|
return object;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
void *kmem_cache_alloc(struct kmem_cache *s, gfp_t gfpflags)
|
|
|
|
{
|
2007-07-17 13:03:28 +02:00
|
|
|
return slab_alloc(s, gfpflags, -1, __builtin_return_address(0));
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_alloc);
|
|
|
|
|
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
void *kmem_cache_alloc_node(struct kmem_cache *s, gfp_t gfpflags, int node)
|
|
|
|
{
|
2007-07-17 13:03:28 +02:00
|
|
|
return slab_alloc(s, gfpflags, node, __builtin_return_address(0));
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_alloc_node);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/*
|
2007-05-10 12:15:16 +02:00
|
|
|
* Slow patch handling. This may still be called frequently since objects
|
|
|
|
* have a longer lifetime than the cpu slabs in most processing loads.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-10 12:15:16 +02:00
|
|
|
* So we still attempt to reduce cache line usage. Just take the slab
|
|
|
|
* lock and free the item. If there is no additional partial page
|
|
|
|
* handling required then we can return immediately.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2007-05-10 12:15:16 +02:00
|
|
|
static void __slab_free(struct kmem_cache *s, struct page *page,
|
2007-10-16 10:26:06 +02:00
|
|
|
void *x, void *addr, unsigned int offset)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
void *prior;
|
|
|
|
void **object = (void *)x;
|
2008-02-08 02:47:41 +01:00
|
|
|
struct kmem_cache_cpu *c;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2008-02-08 02:47:41 +01:00
|
|
|
c = get_cpu_slab(s, raw_smp_processor_id());
|
|
|
|
stat(c, FREE_SLOWPATH);
|
2007-05-06 23:49:36 +02:00
|
|
|
slab_lock(page);
|
|
|
|
|
2007-05-09 11:32:42 +02:00
|
|
|
if (unlikely(SlabDebug(page)))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto debug;
|
2008-02-16 08:45:26 +01:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
checks_ok:
|
2007-10-16 10:26:06 +02:00
|
|
|
prior = object[offset] = page->freelist;
|
2007-05-06 23:49:36 +02:00
|
|
|
page->freelist = object;
|
|
|
|
page->inuse--;
|
|
|
|
|
2008-02-08 02:47:41 +01:00
|
|
|
if (unlikely(SlabFrozen(page))) {
|
|
|
|
stat(c, FREE_FROZEN);
|
2007-05-06 23:49:36 +02:00
|
|
|
goto out_unlock;
|
2008-02-08 02:47:41 +01:00
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
if (unlikely(!page->inuse))
|
|
|
|
goto slab_empty;
|
|
|
|
|
|
|
|
/*
|
2008-02-16 08:45:26 +01:00
|
|
|
* Objects left in the slab. If it was not on the partial list before
|
2007-05-06 23:49:36 +02:00
|
|
|
* then add it.
|
|
|
|
*/
|
2008-03-01 22:40:44 +01:00
|
|
|
if (unlikely(!prior)) {
|
2008-01-08 08:20:27 +01:00
|
|
|
add_partial(get_node(s, page_to_nid(page)), page, 1);
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, FREE_ADD_PARTIAL);
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
out_unlock:
|
|
|
|
slab_unlock(page);
|
|
|
|
return;
|
|
|
|
|
|
|
|
slab_empty:
|
2008-03-01 22:40:44 +01:00
|
|
|
if (prior) {
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Slab still on the partial list.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
remove_partial(s, page);
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, FREE_REMOVE_PARTIAL);
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
slab_unlock(page);
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, FREE_SLAB);
|
2007-05-06 23:49:36 +02:00
|
|
|
discard_slab(s, page);
|
|
|
|
return;
|
|
|
|
|
|
|
|
debug:
|
2007-05-17 07:11:00 +02:00
|
|
|
if (!free_debug_processing(s, page, x, addr))
|
2007-05-06 23:49:42 +02:00
|
|
|
goto out_unlock;
|
|
|
|
goto checks_ok;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2007-05-10 12:15:16 +02:00
|
|
|
/*
|
|
|
|
* Fastpath with forced inlining to produce a kfree and kmem_cache_free that
|
|
|
|
* can perform fastpath freeing without additional function calls.
|
|
|
|
*
|
|
|
|
* The fastpath is only possible if we are freeing to the current cpu slab
|
|
|
|
* of this processor. This typically the case if we have just allocated
|
|
|
|
* the item before.
|
|
|
|
*
|
|
|
|
* If fastpath is not possible then fall back to __slab_free where we deal
|
|
|
|
* with all sorts of special processing.
|
|
|
|
*/
|
2008-01-08 08:20:27 +01:00
|
|
|
static __always_inline void slab_free(struct kmem_cache *s,
|
2007-05-10 12:15:16 +02:00
|
|
|
struct page *page, void *x, void *addr)
|
|
|
|
{
|
|
|
|
void **object = (void *)x;
|
2007-10-16 10:26:05 +02:00
|
|
|
struct kmem_cache_cpu *c;
|
2008-01-08 08:20:30 +01:00
|
|
|
unsigned long flags;
|
|
|
|
|
2007-05-10 12:15:16 +02:00
|
|
|
local_irq_save(flags);
|
2007-10-16 10:26:05 +02:00
|
|
|
c = get_cpu_slab(s, smp_processor_id());
|
2008-02-16 08:45:25 +01:00
|
|
|
debug_check_no_locks_freed(object, c->objsize);
|
2007-10-16 10:26:07 +02:00
|
|
|
if (likely(page == c->page && c->node >= 0)) {
|
2007-10-16 10:26:06 +02:00
|
|
|
object[c->offset] = c->freelist;
|
2007-10-16 10:26:05 +02:00
|
|
|
c->freelist = object;
|
2008-02-08 02:47:41 +01:00
|
|
|
stat(c, FREE_FASTPATH);
|
2007-05-10 12:15:16 +02:00
|
|
|
} else
|
2007-10-16 10:26:06 +02:00
|
|
|
__slab_free(s, page, x, addr, c->offset);
|
2007-05-10 12:15:16 +02:00
|
|
|
|
|
|
|
local_irq_restore(flags);
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
void kmem_cache_free(struct kmem_cache *s, void *x)
|
|
|
|
{
|
2007-05-06 23:49:42 +02:00
|
|
|
struct page *page;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-05-06 23:49:41 +02:00
|
|
|
page = virt_to_head_page(x);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-05-06 23:49:42 +02:00
|
|
|
slab_free(s, page, x, __builtin_return_address(0));
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_free);
|
|
|
|
|
|
|
|
/* Figure out on which slab object the object resides */
|
|
|
|
static struct page *get_object_page(const void *x)
|
|
|
|
{
|
2007-05-06 23:49:41 +02:00
|
|
|
struct page *page = virt_to_head_page(x);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
if (!PageSlab(page))
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
return page;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Object placement in a slab is made very easy because we always start at
|
|
|
|
* offset 0. If we tune the size of the object to the alignment then we can
|
|
|
|
* get the required alignment by putting one properly sized object after
|
|
|
|
* another.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
|
|
|
* Notice that the allocation order determines the sizes of the per cpu
|
|
|
|
* caches. Each processor has always one slab available for allocations.
|
|
|
|
* Increasing the allocation order reduces the number of times that slabs
|
2007-05-09 11:32:39 +02:00
|
|
|
* must be moved on and off the partial lists and is therefore a factor in
|
2007-05-06 23:49:36 +02:00
|
|
|
* locking overhead.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Mininum / Maximum order of slab pages. This influences locking overhead
|
|
|
|
* and slab fragmentation. A higher order reduces the number of partial slabs
|
|
|
|
* and increases the number of allocations possible without having to
|
|
|
|
* take the list_lock.
|
|
|
|
*/
|
|
|
|
static int slub_min_order;
|
|
|
|
static int slub_max_order = DEFAULT_MAX_ORDER;
|
|
|
|
static int slub_min_objects = DEFAULT_MIN_OBJECTS;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Merge control. If this is set then no merging of slab caches will occur.
|
2007-05-09 11:32:39 +02:00
|
|
|
* (Could be removed. This was introduced to pacify the merge skeptics.)
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static int slub_nomerge;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Calculate the order of allocation given an slab object size.
|
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* The order of allocation has significant impact on performance and other
|
|
|
|
* system components. Generally order 0 allocations should be preferred since
|
|
|
|
* order 0 does not cause fragmentation in the page allocator. Larger objects
|
|
|
|
* be problematic to put into order 0 slabs because there may be too much
|
|
|
|
* unused space left. We go to a higher order if more than 1/8th of the slab
|
|
|
|
* would be wasted.
|
|
|
|
*
|
|
|
|
* In order to reach satisfactory performance we must ensure that a minimum
|
|
|
|
* number of objects is in one slab. Otherwise we may generate too much
|
|
|
|
* activity on the partial lists which requires taking the list_lock. This is
|
|
|
|
* less a concern for large slabs though which are rarely used.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* slub_max_order specifies the order where we begin to stop considering the
|
|
|
|
* number of objects in a slab as critical. If we reach slub_max_order then
|
|
|
|
* we try to keep the page order as low as possible. So we accept more waste
|
|
|
|
* of space in favor of a small page order.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Higher order allocations also allow the placement of more objects in a
|
|
|
|
* slab and thereby reduce object handling overhead. If the user has
|
|
|
|
* requested a higher mininum order then we start with that one instead of
|
|
|
|
* the smallest order which will fit the object.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2007-05-09 11:32:46 +02:00
|
|
|
static inline int slab_order(int size, int min_objects,
|
|
|
|
int max_order, int fract_leftover)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
int order;
|
|
|
|
int rem;
|
2007-07-17 13:03:20 +02:00
|
|
|
int min_order = slub_min_order;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-07-17 13:03:20 +02:00
|
|
|
for (order = max(min_order,
|
2007-05-09 11:32:46 +02:00
|
|
|
fls(min_objects * size - 1) - PAGE_SHIFT);
|
|
|
|
order <= max_order; order++) {
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-05-09 11:32:46 +02:00
|
|
|
unsigned long slab_size = PAGE_SIZE << order;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-05-09 11:32:46 +02:00
|
|
|
if (slab_size < min_objects * size)
|
2007-05-06 23:49:36 +02:00
|
|
|
continue;
|
|
|
|
|
|
|
|
rem = slab_size % size;
|
|
|
|
|
2007-05-09 11:32:46 +02:00
|
|
|
if (rem <= slab_size / fract_leftover)
|
2007-05-06 23:49:36 +02:00
|
|
|
break;
|
|
|
|
|
|
|
|
}
|
2007-05-09 11:32:39 +02:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
return order;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:46 +02:00
|
|
|
static inline int calculate_order(int size)
|
|
|
|
{
|
|
|
|
int order;
|
|
|
|
int min_objects;
|
|
|
|
int fraction;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Attempt to find best configuration for a slab. This
|
|
|
|
* works by first attempting to generate a layout with
|
|
|
|
* the best configuration and backing off gradually.
|
|
|
|
*
|
|
|
|
* First we reduce the acceptable waste in a slab. Then
|
|
|
|
* we reduce the minimum objects required in a slab.
|
|
|
|
*/
|
|
|
|
min_objects = slub_min_objects;
|
|
|
|
while (min_objects > 1) {
|
|
|
|
fraction = 8;
|
|
|
|
while (fraction >= 4) {
|
|
|
|
order = slab_order(size, min_objects,
|
|
|
|
slub_max_order, fraction);
|
|
|
|
if (order <= slub_max_order)
|
|
|
|
return order;
|
|
|
|
fraction /= 2;
|
|
|
|
}
|
|
|
|
min_objects /= 2;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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);
|
|
|
|
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);
|
|
|
|
if (order <= MAX_ORDER)
|
|
|
|
return order;
|
|
|
|
return -ENOSYS;
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Figure out what the alignment of the objects will be.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static unsigned long calculate_alignment(unsigned long flags,
|
|
|
|
unsigned long align, unsigned long size)
|
|
|
|
{
|
|
|
|
/*
|
2008-02-16 08:45:26 +01:00
|
|
|
* If the user wants hardware cache aligned objects then follow that
|
|
|
|
* suggestion if the object is sufficiently large.
|
2007-05-06 23:49:36 +02:00
|
|
|
*
|
2008-02-16 08:45:26 +01:00
|
|
|
* The hardware cache alignment cannot override the specified
|
|
|
|
* alignment though. If that is greater then use it.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2008-03-05 23:05:56 +01:00
|
|
|
if (flags & SLAB_HWCACHE_ALIGN) {
|
|
|
|
unsigned long ralign = cache_line_size();
|
|
|
|
while (size <= ralign / 2)
|
|
|
|
ralign /= 2;
|
|
|
|
align = max(align, ralign);
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
if (align < ARCH_SLAB_MINALIGN)
|
2008-03-05 23:05:56 +01:00
|
|
|
align = ARCH_SLAB_MINALIGN;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
return ALIGN(align, sizeof(void *));
|
|
|
|
}
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
static void init_kmem_cache_cpu(struct kmem_cache *s,
|
|
|
|
struct kmem_cache_cpu *c)
|
|
|
|
{
|
|
|
|
c->page = NULL;
|
2008-03-01 22:40:44 +01:00
|
|
|
c->freelist = NULL;
|
2007-10-16 10:26:05 +02:00
|
|
|
c->node = 0;
|
2007-10-16 10:26:09 +02:00
|
|
|
c->offset = s->offset / sizeof(void *);
|
|
|
|
c->objsize = s->objsize;
|
2007-10-16 10:26:05 +02:00
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
static void init_kmem_cache_node(struct kmem_cache_node *n)
|
|
|
|
{
|
|
|
|
n->nr_partial = 0;
|
|
|
|
atomic_long_set(&n->nr_slabs, 0);
|
|
|
|
spin_lock_init(&n->list_lock);
|
|
|
|
INIT_LIST_HEAD(&n->partial);
|
2007-07-17 13:03:32 +02:00
|
|
|
#ifdef CONFIG_SLUB_DEBUG
|
2007-05-06 23:49:42 +02:00
|
|
|
INIT_LIST_HEAD(&n->full);
|
2007-07-17 13:03:32 +02:00
|
|
|
#endif
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2007-10-16 10:26:08 +02:00
|
|
|
#ifdef CONFIG_SMP
|
|
|
|
/*
|
|
|
|
* Per cpu array for per cpu structures.
|
|
|
|
*
|
|
|
|
* The per cpu array places all kmem_cache_cpu structures from one processor
|
|
|
|
* close together meaning that it becomes possible that multiple per cpu
|
|
|
|
* structures are contained in one cacheline. This may be particularly
|
|
|
|
* beneficial for the kmalloc caches.
|
|
|
|
*
|
|
|
|
* A desktop system typically has around 60-80 slabs. With 100 here we are
|
|
|
|
* likely able to get per cpu structures for all caches from the array defined
|
|
|
|
* here. We must be able to cover all kmalloc caches during bootstrap.
|
|
|
|
*
|
|
|
|
* If the per cpu array is exhausted then fall back to kmalloc
|
|
|
|
* of individual cachelines. No sharing is possible then.
|
|
|
|
*/
|
|
|
|
#define NR_KMEM_CACHE_CPU 100
|
|
|
|
|
|
|
|
static DEFINE_PER_CPU(struct kmem_cache_cpu,
|
|
|
|
kmem_cache_cpu)[NR_KMEM_CACHE_CPU];
|
|
|
|
|
|
|
|
static DEFINE_PER_CPU(struct kmem_cache_cpu *, kmem_cache_cpu_free);
|
|
|
|
static cpumask_t kmem_cach_cpu_free_init_once = CPU_MASK_NONE;
|
|
|
|
|
|
|
|
static struct kmem_cache_cpu *alloc_kmem_cache_cpu(struct kmem_cache *s,
|
|
|
|
int cpu, gfp_t flags)
|
|
|
|
{
|
|
|
|
struct kmem_cache_cpu *c = per_cpu(kmem_cache_cpu_free, cpu);
|
|
|
|
|
|
|
|
if (c)
|
|
|
|
per_cpu(kmem_cache_cpu_free, cpu) =
|
|
|
|
(void *)c->freelist;
|
|
|
|
else {
|
|
|
|
/* Table overflow: So allocate ourselves */
|
|
|
|
c = kmalloc_node(
|
|
|
|
ALIGN(sizeof(struct kmem_cache_cpu), cache_line_size()),
|
|
|
|
flags, cpu_to_node(cpu));
|
|
|
|
if (!c)
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
init_kmem_cache_cpu(s, c);
|
|
|
|
return c;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void free_kmem_cache_cpu(struct kmem_cache_cpu *c, int cpu)
|
|
|
|
{
|
|
|
|
if (c < per_cpu(kmem_cache_cpu, cpu) ||
|
|
|
|
c > per_cpu(kmem_cache_cpu, cpu) + NR_KMEM_CACHE_CPU) {
|
|
|
|
kfree(c);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
c->freelist = (void *)per_cpu(kmem_cache_cpu_free, cpu);
|
|
|
|
per_cpu(kmem_cache_cpu_free, cpu) = c;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void free_kmem_cache_cpus(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
int cpu;
|
|
|
|
|
|
|
|
for_each_online_cpu(cpu) {
|
|
|
|
struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
|
|
|
|
|
|
|
|
if (c) {
|
|
|
|
s->cpu_slab[cpu] = NULL;
|
|
|
|
free_kmem_cache_cpu(c, cpu);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static int alloc_kmem_cache_cpus(struct kmem_cache *s, gfp_t flags)
|
|
|
|
{
|
|
|
|
int cpu;
|
|
|
|
|
|
|
|
for_each_online_cpu(cpu) {
|
|
|
|
struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
|
|
|
|
|
|
|
|
if (c)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
c = alloc_kmem_cache_cpu(s, cpu, flags);
|
|
|
|
if (!c) {
|
|
|
|
free_kmem_cache_cpus(s);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
s->cpu_slab[cpu] = c;
|
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Initialize the per cpu array.
|
|
|
|
*/
|
|
|
|
static void init_alloc_cpu_cpu(int cpu)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
if (cpu_isset(cpu, kmem_cach_cpu_free_init_once))
|
|
|
|
return;
|
|
|
|
|
|
|
|
for (i = NR_KMEM_CACHE_CPU - 1; i >= 0; i--)
|
|
|
|
free_kmem_cache_cpu(&per_cpu(kmem_cache_cpu, cpu)[i], cpu);
|
|
|
|
|
|
|
|
cpu_set(cpu, kmem_cach_cpu_free_init_once);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __init init_alloc_cpu(void)
|
|
|
|
{
|
|
|
|
int cpu;
|
|
|
|
|
|
|
|
for_each_online_cpu(cpu)
|
|
|
|
init_alloc_cpu_cpu(cpu);
|
|
|
|
}
|
|
|
|
|
|
|
|
#else
|
|
|
|
static inline void free_kmem_cache_cpus(struct kmem_cache *s) {}
|
|
|
|
static inline void init_alloc_cpu(void) {}
|
|
|
|
|
|
|
|
static inline int alloc_kmem_cache_cpus(struct kmem_cache *s, gfp_t flags)
|
|
|
|
{
|
|
|
|
init_kmem_cache_cpu(s, &s->cpu_slab);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
/*
|
|
|
|
* No kmalloc_node yet so do it by hand. We know that this is the first
|
|
|
|
* slab on the node for this slabcache. There are no concurrent accesses
|
|
|
|
* possible.
|
|
|
|
*
|
|
|
|
* Note that this function only works on the kmalloc_node_cache
|
2007-10-16 10:26:08 +02:00
|
|
|
* when allocating for the kmalloc_node_cache. This is used for bootstrapping
|
|
|
|
* memory on a fresh node that has no slab structures yet.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2007-10-16 10:24:18 +02:00
|
|
|
static struct kmem_cache_node *early_kmem_cache_node_alloc(gfp_t gfpflags,
|
|
|
|
int node)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
struct kmem_cache_node *n;
|
2008-01-08 08:20:28 +01:00
|
|
|
unsigned long flags;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
BUG_ON(kmalloc_caches->size < sizeof(struct kmem_cache_node));
|
|
|
|
|
2007-08-22 23:01:57 +02:00
|
|
|
page = new_slab(kmalloc_caches, gfpflags, node);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
BUG_ON(!page);
|
2007-08-22 23:01:57 +02:00
|
|
|
if (page_to_nid(page) != node) {
|
|
|
|
printk(KERN_ERR "SLUB: Unable to allocate memory from "
|
|
|
|
"node %d\n", node);
|
|
|
|
printk(KERN_ERR "SLUB: Allocating a useless per node structure "
|
|
|
|
"in order to be able to continue\n");
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
n = page->freelist;
|
|
|
|
BUG_ON(!n);
|
|
|
|
page->freelist = get_freepointer(kmalloc_caches, n);
|
|
|
|
page->inuse++;
|
|
|
|
kmalloc_caches->node[node] = n;
|
2007-07-17 13:03:32 +02:00
|
|
|
#ifdef CONFIG_SLUB_DEBUG
|
2007-07-17 13:03:21 +02:00
|
|
|
init_object(kmalloc_caches, n, 1);
|
|
|
|
init_tracking(kmalloc_caches, n);
|
2007-07-17 13:03:32 +02:00
|
|
|
#endif
|
2007-05-06 23:49:36 +02:00
|
|
|
init_kmem_cache_node(n);
|
|
|
|
atomic_long_inc(&n->nr_slabs);
|
2008-02-16 08:45:26 +01:00
|
|
|
|
2008-01-08 08:20:28 +01:00
|
|
|
/*
|
|
|
|
* lockdep requires consistent irq usage for each lock
|
|
|
|
* so even though there cannot be a race this early in
|
|
|
|
* the boot sequence, we still disable irqs.
|
|
|
|
*/
|
|
|
|
local_irq_save(flags);
|
2008-01-08 08:20:27 +01:00
|
|
|
add_partial(n, page, 0);
|
2008-01-08 08:20:28 +01:00
|
|
|
local_irq_restore(flags);
|
2007-05-06 23:49:36 +02:00
|
|
|
return n;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void free_kmem_cache_nodes(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
int node;
|
|
|
|
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY) {
|
2007-05-06 23:49:36 +02:00
|
|
|
struct kmem_cache_node *n = s->node[node];
|
|
|
|
if (n && n != &s->local_node)
|
|
|
|
kmem_cache_free(kmalloc_caches, n);
|
|
|
|
s->node[node] = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static int init_kmem_cache_nodes(struct kmem_cache *s, gfp_t gfpflags)
|
|
|
|
{
|
|
|
|
int node;
|
|
|
|
int local_node;
|
|
|
|
|
|
|
|
if (slab_state >= UP)
|
|
|
|
local_node = page_to_nid(virt_to_page(s));
|
|
|
|
else
|
|
|
|
local_node = 0;
|
|
|
|
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY) {
|
2007-05-06 23:49:36 +02:00
|
|
|
struct kmem_cache_node *n;
|
|
|
|
|
|
|
|
if (local_node == node)
|
|
|
|
n = &s->local_node;
|
|
|
|
else {
|
|
|
|
if (slab_state == DOWN) {
|
|
|
|
n = early_kmem_cache_node_alloc(gfpflags,
|
|
|
|
node);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
n = kmem_cache_alloc_node(kmalloc_caches,
|
|
|
|
gfpflags, node);
|
|
|
|
|
|
|
|
if (!n) {
|
|
|
|
free_kmem_cache_nodes(s);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
s->node[node] = n;
|
|
|
|
init_kmem_cache_node(n);
|
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
static void free_kmem_cache_nodes(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
static int init_kmem_cache_nodes(struct kmem_cache *s, gfp_t gfpflags)
|
|
|
|
{
|
|
|
|
init_kmem_cache_node(&s->local_node);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/*
|
|
|
|
* calculate_sizes() determines the order and the distribution of data within
|
|
|
|
* a slab object.
|
|
|
|
*/
|
|
|
|
static int calculate_sizes(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
unsigned long flags = s->flags;
|
|
|
|
unsigned long size = s->objsize;
|
|
|
|
unsigned long align = s->align;
|
|
|
|
|
2008-02-16 08:45:25 +01:00
|
|
|
/*
|
|
|
|
* Round up object size to the next word boundary. We can only
|
|
|
|
* place the free pointer at word boundaries and this determines
|
|
|
|
* the possible location of the free pointer.
|
|
|
|
*/
|
|
|
|
size = ALIGN(size, sizeof(void *));
|
|
|
|
|
|
|
|
#ifdef CONFIG_SLUB_DEBUG
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Determine if we can poison the object itself. If the user of
|
|
|
|
* the slab may touch the object after free or before allocation
|
|
|
|
* then we should never poison the object itself.
|
|
|
|
*/
|
|
|
|
if ((flags & SLAB_POISON) && !(flags & SLAB_DESTROY_BY_RCU) &&
|
2007-05-17 07:10:50 +02:00
|
|
|
!s->ctor)
|
2007-05-06 23:49:36 +02:00
|
|
|
s->flags |= __OBJECT_POISON;
|
|
|
|
else
|
|
|
|
s->flags &= ~__OBJECT_POISON;
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* If we are Redzoning then check if there is some space between the
|
2007-05-06 23:49:36 +02:00
|
|
|
* end of the object and the free pointer. If not then add an
|
2007-05-09 11:32:39 +02:00
|
|
|
* additional word to have some bytes to store Redzone information.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
if ((flags & SLAB_RED_ZONE) && size == s->objsize)
|
|
|
|
size += sizeof(void *);
|
2007-05-09 11:32:44 +02:00
|
|
|
#endif
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* With that we have determined the number of bytes in actual use
|
|
|
|
* by the object. This is the potential offset to the free pointer.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
s->inuse = size;
|
|
|
|
|
|
|
|
if (((flags & (SLAB_DESTROY_BY_RCU | SLAB_POISON)) ||
|
2007-05-17 07:10:50 +02:00
|
|
|
s->ctor)) {
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Relocate free pointer after the object if it is not
|
|
|
|
* permitted to overwrite the first word of the object on
|
|
|
|
* kmem_cache_free.
|
|
|
|
*
|
|
|
|
* This is the case if we do RCU, have a constructor or
|
|
|
|
* destructor or are poisoning the objects.
|
|
|
|
*/
|
|
|
|
s->offset = size;
|
|
|
|
size += sizeof(void *);
|
|
|
|
}
|
|
|
|
|
2007-05-23 22:57:31 +02:00
|
|
|
#ifdef CONFIG_SLUB_DEBUG
|
2007-05-06 23:49:36 +02:00
|
|
|
if (flags & SLAB_STORE_USER)
|
|
|
|
/*
|
|
|
|
* Need to store information about allocs and frees after
|
|
|
|
* the object.
|
|
|
|
*/
|
|
|
|
size += 2 * sizeof(struct track);
|
|
|
|
|
2007-05-09 11:32:36 +02:00
|
|
|
if (flags & SLAB_RED_ZONE)
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Add some empty padding so that we can catch
|
|
|
|
* overwrites from earlier objects rather than let
|
|
|
|
* tracking information or the free pointer be
|
|
|
|
* corrupted if an user writes before the start
|
|
|
|
* of the object.
|
|
|
|
*/
|
|
|
|
size += sizeof(void *);
|
2007-05-09 11:32:44 +02:00
|
|
|
#endif
|
2007-05-09 11:32:39 +02:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Determine the alignment based on various parameters that the
|
2007-05-09 11:32:35 +02:00
|
|
|
* user specified and the dynamic determination of cache line size
|
|
|
|
* on bootup.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
align = calculate_alignment(flags, align, s->objsize);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* SLUB stores one object immediately after another beginning from
|
|
|
|
* offset 0. In order to align the objects we have to simply size
|
|
|
|
* each object to conform to the alignment.
|
|
|
|
*/
|
|
|
|
size = ALIGN(size, align);
|
|
|
|
s->size = size;
|
|
|
|
|
2008-02-14 23:28:01 +01:00
|
|
|
if ((flags & __KMALLOC_CACHE) &&
|
|
|
|
PAGE_SIZE / size < slub_min_objects) {
|
|
|
|
/*
|
|
|
|
* Kmalloc cache that would not have enough objects in
|
|
|
|
* an order 0 page. Kmalloc slabs can fallback to
|
|
|
|
* page allocator order 0 allocs so take a reasonably large
|
|
|
|
* order that will allows us a good number of objects.
|
|
|
|
*/
|
|
|
|
s->order = max(slub_max_order, PAGE_ALLOC_COSTLY_ORDER);
|
|
|
|
s->flags |= __PAGE_ALLOC_FALLBACK;
|
|
|
|
s->allocflags |= __GFP_NOWARN;
|
|
|
|
} else
|
|
|
|
s->order = calculate_order(size);
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
if (s->order < 0)
|
|
|
|
return 0;
|
|
|
|
|
2008-02-14 23:21:32 +01:00
|
|
|
s->allocflags = 0;
|
|
|
|
if (s->order)
|
|
|
|
s->allocflags |= __GFP_COMP;
|
|
|
|
|
|
|
|
if (s->flags & SLAB_CACHE_DMA)
|
|
|
|
s->allocflags |= SLUB_DMA;
|
|
|
|
|
|
|
|
if (s->flags & SLAB_RECLAIM_ACCOUNT)
|
|
|
|
s->allocflags |= __GFP_RECLAIMABLE;
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* Determine the number of objects per slab
|
|
|
|
*/
|
|
|
|
s->objects = (PAGE_SIZE << s->order) / size;
|
|
|
|
|
2007-10-16 10:26:06 +02:00
|
|
|
return !!s->objects;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
static int kmem_cache_open(struct kmem_cache *s, gfp_t gfpflags,
|
|
|
|
const char *name, size_t size,
|
|
|
|
size_t align, unsigned long flags,
|
2007-10-17 08:25:51 +02:00
|
|
|
void (*ctor)(struct kmem_cache *, void *))
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
memset(s, 0, kmem_size);
|
|
|
|
s->name = name;
|
|
|
|
s->ctor = ctor;
|
|
|
|
s->objsize = size;
|
|
|
|
s->align = align;
|
2007-09-12 00:24:11 +02:00
|
|
|
s->flags = kmem_cache_flags(size, flags, name, ctor);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
if (!calculate_sizes(s))
|
|
|
|
goto error;
|
|
|
|
|
|
|
|
s->refcount = 1;
|
|
|
|
#ifdef CONFIG_NUMA
|
2008-01-08 08:20:26 +01:00
|
|
|
s->remote_node_defrag_ratio = 100;
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
2007-10-16 10:26:05 +02:00
|
|
|
if (!init_kmem_cache_nodes(s, gfpflags & ~SLUB_DMA))
|
|
|
|
goto error;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
if (alloc_kmem_cache_cpus(s, gfpflags & ~SLUB_DMA))
|
2007-05-06 23:49:36 +02:00
|
|
|
return 1;
|
2007-10-16 10:26:08 +02:00
|
|
|
free_kmem_cache_nodes(s);
|
2007-05-06 23:49:36 +02:00
|
|
|
error:
|
|
|
|
if (flags & SLAB_PANIC)
|
|
|
|
panic("Cannot create slab %s size=%lu realsize=%u "
|
|
|
|
"order=%u offset=%u flags=%lx\n",
|
|
|
|
s->name, (unsigned long)size, s->size, s->order,
|
|
|
|
s->offset, flags);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check if a given pointer is valid
|
|
|
|
*/
|
|
|
|
int kmem_ptr_validate(struct kmem_cache *s, const void *object)
|
|
|
|
{
|
2008-01-08 08:20:27 +01:00
|
|
|
struct page *page;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
page = get_object_page(object);
|
|
|
|
|
|
|
|
if (!page || s != page->slab)
|
|
|
|
/* No slab or wrong slab */
|
|
|
|
return 0;
|
|
|
|
|
2007-05-09 11:32:37 +02:00
|
|
|
if (!check_valid_pointer(s, page, object))
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We could also check if the object is on the slabs freelist.
|
|
|
|
* But this would be too expensive and it seems that the main
|
2008-02-16 08:45:26 +01:00
|
|
|
* purpose of kmem_ptr_valid() is to check if the object belongs
|
2007-05-06 23:49:36 +02:00
|
|
|
* to a certain slab.
|
|
|
|
*/
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_ptr_validate);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Determine the size of a slab object
|
|
|
|
*/
|
|
|
|
unsigned int kmem_cache_size(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
return s->objsize;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_size);
|
|
|
|
|
|
|
|
const char *kmem_cache_name(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
return s->name;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_name);
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Attempt to free all slabs on a node. Return the number of slabs we
|
|
|
|
* were unable to free.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static int free_list(struct kmem_cache *s, struct kmem_cache_node *n,
|
|
|
|
struct list_head *list)
|
|
|
|
{
|
|
|
|
int slabs_inuse = 0;
|
|
|
|
unsigned long flags;
|
|
|
|
struct page *page, *h;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&n->list_lock, flags);
|
|
|
|
list_for_each_entry_safe(page, h, list, lru)
|
|
|
|
if (!page->inuse) {
|
|
|
|
list_del(&page->lru);
|
|
|
|
discard_slab(s, page);
|
|
|
|
} else
|
|
|
|
slabs_inuse++;
|
|
|
|
spin_unlock_irqrestore(&n->list_lock, flags);
|
|
|
|
return slabs_inuse;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Release all resources used by a slab cache.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
2007-07-17 13:03:24 +02:00
|
|
|
static inline int kmem_cache_close(struct kmem_cache *s)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
int node;
|
|
|
|
|
|
|
|
flush_all(s);
|
|
|
|
|
|
|
|
/* Attempt to free all objects */
|
2007-10-16 10:26:08 +02:00
|
|
|
free_kmem_cache_cpus(s);
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY) {
|
2007-05-06 23:49:36 +02:00
|
|
|
struct kmem_cache_node *n = get_node(s, node);
|
|
|
|
|
2007-05-06 23:49:46 +02:00
|
|
|
n->nr_partial -= free_list(s, n, &n->partial);
|
2007-05-06 23:49:36 +02:00
|
|
|
if (atomic_long_read(&n->nr_slabs))
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
free_kmem_cache_nodes(s);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Close a cache and release the kmem_cache structure
|
|
|
|
* (must be used for caches created using kmem_cache_create)
|
|
|
|
*/
|
|
|
|
void kmem_cache_destroy(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
down_write(&slub_lock);
|
|
|
|
s->refcount--;
|
|
|
|
if (!s->refcount) {
|
|
|
|
list_del(&s->list);
|
2007-07-17 13:03:31 +02:00
|
|
|
up_write(&slub_lock);
|
2007-05-06 23:49:36 +02:00
|
|
|
if (kmem_cache_close(s))
|
|
|
|
WARN_ON(1);
|
|
|
|
sysfs_slab_remove(s);
|
2007-07-17 13:03:31 +02:00
|
|
|
} else
|
|
|
|
up_write(&slub_lock);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_destroy);
|
|
|
|
|
|
|
|
/********************************************************************
|
|
|
|
* Kmalloc subsystem
|
|
|
|
*******************************************************************/
|
|
|
|
|
2008-02-14 23:28:09 +01:00
|
|
|
struct kmem_cache kmalloc_caches[PAGE_SHIFT + 1] __cacheline_aligned;
|
2007-05-06 23:49:36 +02:00
|
|
|
EXPORT_SYMBOL(kmalloc_caches);
|
|
|
|
|
|
|
|
#ifdef CONFIG_ZONE_DMA
|
2008-02-14 23:28:09 +01:00
|
|
|
static struct kmem_cache *kmalloc_caches_dma[PAGE_SHIFT + 1];
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
|
|
|
|
static int __init setup_slub_min_order(char *str)
|
|
|
|
{
|
2008-01-08 08:20:27 +01:00
|
|
|
get_option(&str, &slub_min_order);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
__setup("slub_min_order=", setup_slub_min_order);
|
|
|
|
|
|
|
|
static int __init setup_slub_max_order(char *str)
|
|
|
|
{
|
2008-01-08 08:20:27 +01:00
|
|
|
get_option(&str, &slub_max_order);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
__setup("slub_max_order=", setup_slub_max_order);
|
|
|
|
|
|
|
|
static int __init setup_slub_min_objects(char *str)
|
|
|
|
{
|
2008-01-08 08:20:27 +01:00
|
|
|
get_option(&str, &slub_min_objects);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
__setup("slub_min_objects=", setup_slub_min_objects);
|
|
|
|
|
|
|
|
static int __init setup_slub_nomerge(char *str)
|
|
|
|
{
|
|
|
|
slub_nomerge = 1;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
__setup("slub_nomerge", setup_slub_nomerge);
|
|
|
|
|
|
|
|
static struct kmem_cache *create_kmalloc_cache(struct kmem_cache *s,
|
|
|
|
const char *name, int size, gfp_t gfp_flags)
|
|
|
|
{
|
|
|
|
unsigned int flags = 0;
|
|
|
|
|
|
|
|
if (gfp_flags & SLUB_DMA)
|
|
|
|
flags = SLAB_CACHE_DMA;
|
|
|
|
|
|
|
|
down_write(&slub_lock);
|
|
|
|
if (!kmem_cache_open(s, gfp_flags, name, size, ARCH_KMALLOC_MINALIGN,
|
2008-02-14 23:28:01 +01:00
|
|
|
flags | __KMALLOC_CACHE, NULL))
|
2007-05-06 23:49:36 +02:00
|
|
|
goto panic;
|
|
|
|
|
|
|
|
list_add(&s->list, &slab_caches);
|
|
|
|
up_write(&slub_lock);
|
|
|
|
if (sysfs_slab_add(s))
|
|
|
|
goto panic;
|
|
|
|
return s;
|
|
|
|
|
|
|
|
panic:
|
|
|
|
panic("Creation of kmalloc slab %s size=%d failed.\n", name, size);
|
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:24 +02:00
|
|
|
#ifdef CONFIG_ZONE_DMA
|
2007-08-08 00:11:48 +02:00
|
|
|
|
|
|
|
static void sysfs_add_func(struct work_struct *w)
|
|
|
|
{
|
|
|
|
struct kmem_cache *s;
|
|
|
|
|
|
|
|
down_write(&slub_lock);
|
|
|
|
list_for_each_entry(s, &slab_caches, list) {
|
|
|
|
if (s->flags & __SYSFS_ADD_DEFERRED) {
|
|
|
|
s->flags &= ~__SYSFS_ADD_DEFERRED;
|
|
|
|
sysfs_slab_add(s);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
up_write(&slub_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
static DECLARE_WORK(sysfs_add_work, sysfs_add_func);
|
|
|
|
|
2007-07-17 13:03:24 +02:00
|
|
|
static noinline struct kmem_cache *dma_kmalloc_cache(int index, gfp_t flags)
|
|
|
|
{
|
|
|
|
struct kmem_cache *s;
|
|
|
|
char *text;
|
|
|
|
size_t realsize;
|
|
|
|
|
|
|
|
s = kmalloc_caches_dma[index];
|
|
|
|
if (s)
|
|
|
|
return s;
|
|
|
|
|
|
|
|
/* Dynamically create dma cache */
|
2007-08-08 00:11:48 +02:00
|
|
|
if (flags & __GFP_WAIT)
|
|
|
|
down_write(&slub_lock);
|
|
|
|
else {
|
|
|
|
if (!down_write_trylock(&slub_lock))
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (kmalloc_caches_dma[index])
|
|
|
|
goto unlock_out;
|
2007-07-17 13:03:24 +02:00
|
|
|
|
2007-07-17 13:03:27 +02:00
|
|
|
realsize = kmalloc_caches[index].objsize;
|
2008-02-06 02:57:39 +01:00
|
|
|
text = kasprintf(flags & ~SLUB_DMA, "kmalloc_dma-%d",
|
|
|
|
(unsigned int)realsize);
|
2007-08-08 00:11:48 +02:00
|
|
|
s = kmalloc(kmem_size, flags & ~SLUB_DMA);
|
|
|
|
|
|
|
|
if (!s || !text || !kmem_cache_open(s, flags, text,
|
|
|
|
realsize, ARCH_KMALLOC_MINALIGN,
|
|
|
|
SLAB_CACHE_DMA|__SYSFS_ADD_DEFERRED, NULL)) {
|
|
|
|
kfree(s);
|
|
|
|
kfree(text);
|
|
|
|
goto unlock_out;
|
2007-07-17 13:03:25 +02:00
|
|
|
}
|
2007-08-08 00:11:48 +02:00
|
|
|
|
|
|
|
list_add(&s->list, &slab_caches);
|
|
|
|
kmalloc_caches_dma[index] = s;
|
|
|
|
|
|
|
|
schedule_work(&sysfs_add_work);
|
|
|
|
|
|
|
|
unlock_out:
|
2007-07-17 13:03:25 +02:00
|
|
|
up_write(&slub_lock);
|
2007-08-08 00:11:48 +02:00
|
|
|
out:
|
2007-07-17 13:03:25 +02:00
|
|
|
return kmalloc_caches_dma[index];
|
2007-07-17 13:03:24 +02:00
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2007-07-17 13:03:26 +02:00
|
|
|
/*
|
|
|
|
* Conversion table for small slabs sizes / 8 to the index in the
|
|
|
|
* kmalloc array. This is necessary for slabs < 192 since we have non power
|
|
|
|
* of two cache sizes there. The size of larger slabs can be determined using
|
|
|
|
* fls.
|
|
|
|
*/
|
|
|
|
static s8 size_index[24] = {
|
|
|
|
3, /* 8 */
|
|
|
|
4, /* 16 */
|
|
|
|
5, /* 24 */
|
|
|
|
5, /* 32 */
|
|
|
|
6, /* 40 */
|
|
|
|
6, /* 48 */
|
|
|
|
6, /* 56 */
|
|
|
|
6, /* 64 */
|
|
|
|
1, /* 72 */
|
|
|
|
1, /* 80 */
|
|
|
|
1, /* 88 */
|
|
|
|
1, /* 96 */
|
|
|
|
7, /* 104 */
|
|
|
|
7, /* 112 */
|
|
|
|
7, /* 120 */
|
|
|
|
7, /* 128 */
|
|
|
|
2, /* 136 */
|
|
|
|
2, /* 144 */
|
|
|
|
2, /* 152 */
|
|
|
|
2, /* 160 */
|
|
|
|
2, /* 168 */
|
|
|
|
2, /* 176 */
|
|
|
|
2, /* 184 */
|
|
|
|
2 /* 192 */
|
|
|
|
};
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
static struct kmem_cache *get_slab(size_t size, gfp_t flags)
|
|
|
|
{
|
2007-07-17 13:03:26 +02:00
|
|
|
int index;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-07-17 13:03:26 +02:00
|
|
|
if (size <= 192) {
|
|
|
|
if (!size)
|
|
|
|
return ZERO_SIZE_PTR;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-07-17 13:03:26 +02:00
|
|
|
index = size_index[(size - 1) / 8];
|
2007-10-16 10:24:38 +02:00
|
|
|
} else
|
2007-07-17 13:03:26 +02:00
|
|
|
index = fls(size - 1);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
#ifdef CONFIG_ZONE_DMA
|
2007-07-17 13:03:26 +02:00
|
|
|
if (unlikely((flags & SLUB_DMA)))
|
2007-07-17 13:03:24 +02:00
|
|
|
return dma_kmalloc_cache(index, flags);
|
2007-07-17 13:03:26 +02:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
return &kmalloc_caches[index];
|
|
|
|
}
|
|
|
|
|
|
|
|
void *__kmalloc(size_t size, gfp_t flags)
|
|
|
|
{
|
2007-10-16 10:24:38 +02:00
|
|
|
struct kmem_cache *s;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2008-02-14 23:28:09 +01:00
|
|
|
if (unlikely(size > PAGE_SIZE))
|
2008-02-11 21:47:46 +01:00
|
|
|
return kmalloc_large(size, flags);
|
2007-10-16 10:24:38 +02:00
|
|
|
|
|
|
|
s = get_slab(size, flags);
|
|
|
|
|
|
|
|
if (unlikely(ZERO_OR_NULL_PTR(s)))
|
2007-07-17 13:03:22 +02:00
|
|
|
return s;
|
|
|
|
|
2007-07-17 13:03:28 +02:00
|
|
|
return slab_alloc(s, flags, -1, __builtin_return_address(0));
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(__kmalloc);
|
|
|
|
|
2008-03-01 22:56:40 +01:00
|
|
|
static void *kmalloc_large_node(size_t size, gfp_t flags, int node)
|
|
|
|
{
|
|
|
|
struct page *page = alloc_pages_node(node, flags | __GFP_COMP,
|
|
|
|
get_order(size));
|
|
|
|
|
|
|
|
if (page)
|
|
|
|
return page_address(page);
|
|
|
|
else
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
void *__kmalloc_node(size_t size, gfp_t flags, int node)
|
|
|
|
{
|
2007-10-16 10:24:38 +02:00
|
|
|
struct kmem_cache *s;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2008-02-14 23:28:09 +01:00
|
|
|
if (unlikely(size > PAGE_SIZE))
|
2008-03-01 22:56:40 +01:00
|
|
|
return kmalloc_large_node(size, flags, node);
|
2007-10-16 10:24:38 +02:00
|
|
|
|
|
|
|
s = get_slab(size, flags);
|
|
|
|
|
|
|
|
if (unlikely(ZERO_OR_NULL_PTR(s)))
|
2007-07-17 13:03:22 +02:00
|
|
|
return s;
|
|
|
|
|
2007-07-17 13:03:28 +02:00
|
|
|
return slab_alloc(s, flags, node, __builtin_return_address(0));
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(__kmalloc_node);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
size_t ksize(const void *object)
|
|
|
|
{
|
2007-06-08 22:46:49 +02:00
|
|
|
struct page *page;
|
2007-05-06 23:49:36 +02:00
|
|
|
struct kmem_cache *s;
|
|
|
|
|
2007-10-16 10:24:46 +02:00
|
|
|
if (unlikely(object == ZERO_SIZE_PTR))
|
2007-06-08 22:46:49 +02:00
|
|
|
return 0;
|
|
|
|
|
2007-12-05 08:45:30 +01:00
|
|
|
page = virt_to_head_page(object);
|
|
|
|
|
|
|
|
if (unlikely(!PageSlab(page)))
|
|
|
|
return PAGE_SIZE << compound_order(page);
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
s = page->slab;
|
|
|
|
|
2008-02-16 08:45:25 +01:00
|
|
|
#ifdef CONFIG_SLUB_DEBUG
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* 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;
|
|
|
|
|
2008-02-16 08:45:25 +01:00
|
|
|
#endif
|
2007-05-06 23:49:36 +02:00
|
|
|
/*
|
|
|
|
* 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;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(ksize);
|
|
|
|
|
|
|
|
void kfree(const void *x)
|
|
|
|
{
|
|
|
|
struct page *page;
|
2008-02-08 02:47:41 +01:00
|
|
|
void *object = (void *)x;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:24:44 +02:00
|
|
|
if (unlikely(ZERO_OR_NULL_PTR(x)))
|
2007-05-06 23:49:36 +02:00
|
|
|
return;
|
|
|
|
|
2007-05-06 23:49:41 +02:00
|
|
|
page = virt_to_head_page(x);
|
2007-10-16 10:24:38 +02:00
|
|
|
if (unlikely(!PageSlab(page))) {
|
|
|
|
put_page(page);
|
|
|
|
return;
|
|
|
|
}
|
2008-02-08 02:47:41 +01:00
|
|
|
slab_free(page->slab, page, object, __builtin_return_address(0));
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kfree);
|
|
|
|
|
2008-03-19 21:42:07 +01:00
|
|
|
#if defined(SLUB_DEBUG) || defined(CONFIG_SLABINFO)
|
2008-01-08 08:20:26 +01:00
|
|
|
static unsigned long count_partial(struct kmem_cache_node *n)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
unsigned long x = 0;
|
|
|
|
struct page *page;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&n->list_lock, flags);
|
|
|
|
list_for_each_entry(page, &n->partial, lru)
|
|
|
|
x += page->inuse;
|
|
|
|
spin_unlock_irqrestore(&n->list_lock, flags);
|
|
|
|
return x;
|
|
|
|
}
|
2008-03-19 21:42:07 +01:00
|
|
|
#endif
|
2008-01-08 08:20:26 +01:00
|
|
|
|
2007-05-06 23:49:46 +02:00
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* kmem_cache_shrink removes empty slabs from the partial lists and sorts
|
|
|
|
* the remaining slabs by the number of items in use. The slabs with the
|
|
|
|
* most items in use come first. New allocations will then fill those up
|
|
|
|
* and thus they can be removed from the partial lists.
|
|
|
|
*
|
|
|
|
* The slabs with the least items are placed last. This results in them
|
|
|
|
* being allocated from last increasing the chance that the last objects
|
|
|
|
* are freed in them.
|
2007-05-06 23:49:46 +02:00
|
|
|
*/
|
|
|
|
int kmem_cache_shrink(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
int node;
|
|
|
|
int i;
|
|
|
|
struct kmem_cache_node *n;
|
|
|
|
struct page *page;
|
|
|
|
struct page *t;
|
|
|
|
struct list_head *slabs_by_inuse =
|
|
|
|
kmalloc(sizeof(struct list_head) * s->objects, GFP_KERNEL);
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
if (!slabs_by_inuse)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
flush_all(s);
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY) {
|
2007-05-06 23:49:46 +02:00
|
|
|
n = get_node(s, node);
|
|
|
|
|
|
|
|
if (!n->nr_partial)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
for (i = 0; i < s->objects; i++)
|
|
|
|
INIT_LIST_HEAD(slabs_by_inuse + i);
|
|
|
|
|
|
|
|
spin_lock_irqsave(&n->list_lock, flags);
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Build lists indexed by the items in use in each slab.
|
2007-05-06 23:49:46 +02:00
|
|
|
*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Note that concurrent frees may occur while we hold the
|
|
|
|
* list_lock. page->inuse here is the upper limit.
|
2007-05-06 23:49:46 +02:00
|
|
|
*/
|
|
|
|
list_for_each_entry_safe(page, t, &n->partial, lru) {
|
|
|
|
if (!page->inuse && slab_trylock(page)) {
|
|
|
|
/*
|
|
|
|
* Must hold slab lock here because slab_free
|
|
|
|
* may have freed the last object and be
|
|
|
|
* waiting to release the slab.
|
|
|
|
*/
|
|
|
|
list_del(&page->lru);
|
|
|
|
n->nr_partial--;
|
|
|
|
slab_unlock(page);
|
|
|
|
discard_slab(s, page);
|
|
|
|
} else {
|
2007-07-30 22:06:46 +02:00
|
|
|
list_move(&page->lru,
|
|
|
|
slabs_by_inuse + page->inuse);
|
2007-05-06 23:49:46 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Rebuild the partial list with the slabs filled up most
|
|
|
|
* first and the least used slabs at the end.
|
2007-05-06 23:49:46 +02:00
|
|
|
*/
|
|
|
|
for (i = s->objects - 1; i >= 0; i--)
|
|
|
|
list_splice(slabs_by_inuse + i, n->partial.prev);
|
|
|
|
|
|
|
|
spin_unlock_irqrestore(&n->list_lock, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
kfree(slabs_by_inuse);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_shrink);
|
|
|
|
|
2007-10-22 01:41:37 +02:00
|
|
|
#if defined(CONFIG_NUMA) && defined(CONFIG_MEMORY_HOTPLUG)
|
|
|
|
static int slab_mem_going_offline_callback(void *arg)
|
|
|
|
{
|
|
|
|
struct kmem_cache *s;
|
|
|
|
|
|
|
|
down_read(&slub_lock);
|
|
|
|
list_for_each_entry(s, &slab_caches, list)
|
|
|
|
kmem_cache_shrink(s);
|
|
|
|
up_read(&slub_lock);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void slab_mem_offline_callback(void *arg)
|
|
|
|
{
|
|
|
|
struct kmem_cache_node *n;
|
|
|
|
struct kmem_cache *s;
|
|
|
|
struct memory_notify *marg = arg;
|
|
|
|
int offline_node;
|
|
|
|
|
|
|
|
offline_node = marg->status_change_nid;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the node still has available memory. we need kmem_cache_node
|
|
|
|
* for it yet.
|
|
|
|
*/
|
|
|
|
if (offline_node < 0)
|
|
|
|
return;
|
|
|
|
|
|
|
|
down_read(&slub_lock);
|
|
|
|
list_for_each_entry(s, &slab_caches, list) {
|
|
|
|
n = get_node(s, offline_node);
|
|
|
|
if (n) {
|
|
|
|
/*
|
|
|
|
* if n->nr_slabs > 0, slabs still exist on the node
|
|
|
|
* that is going down. We were unable to free them,
|
|
|
|
* and offline_pages() function shoudn't call this
|
|
|
|
* callback. So, we must fail.
|
|
|
|
*/
|
2007-10-29 05:42:55 +01:00
|
|
|
BUG_ON(atomic_long_read(&n->nr_slabs));
|
2007-10-22 01:41:37 +02:00
|
|
|
|
|
|
|
s->node[offline_node] = NULL;
|
|
|
|
kmem_cache_free(kmalloc_caches, n);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
up_read(&slub_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int slab_mem_going_online_callback(void *arg)
|
|
|
|
{
|
|
|
|
struct kmem_cache_node *n;
|
|
|
|
struct kmem_cache *s;
|
|
|
|
struct memory_notify *marg = arg;
|
|
|
|
int nid = marg->status_change_nid;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the node's memory is already available, then kmem_cache_node is
|
|
|
|
* already created. Nothing to do.
|
|
|
|
*/
|
|
|
|
if (nid < 0)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We are bringing a node online. No memory is availabe yet. We must
|
|
|
|
* allocate a kmem_cache_node structure in order to bring the node
|
|
|
|
* online.
|
|
|
|
*/
|
|
|
|
down_read(&slub_lock);
|
|
|
|
list_for_each_entry(s, &slab_caches, list) {
|
|
|
|
/*
|
|
|
|
* XXX: kmem_cache_alloc_node will fallback to other nodes
|
|
|
|
* since memory is not yet available from the node that
|
|
|
|
* is brought up.
|
|
|
|
*/
|
|
|
|
n = kmem_cache_alloc(kmalloc_caches, GFP_KERNEL);
|
|
|
|
if (!n) {
|
|
|
|
ret = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
init_kmem_cache_node(n);
|
|
|
|
s->node[nid] = n;
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
up_read(&slub_lock);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int slab_memory_callback(struct notifier_block *self,
|
|
|
|
unsigned long action, void *arg)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
switch (action) {
|
|
|
|
case MEM_GOING_ONLINE:
|
|
|
|
ret = slab_mem_going_online_callback(arg);
|
|
|
|
break;
|
|
|
|
case MEM_GOING_OFFLINE:
|
|
|
|
ret = slab_mem_going_offline_callback(arg);
|
|
|
|
break;
|
|
|
|
case MEM_OFFLINE:
|
|
|
|
case MEM_CANCEL_ONLINE:
|
|
|
|
slab_mem_offline_callback(arg);
|
|
|
|
break;
|
|
|
|
case MEM_ONLINE:
|
|
|
|
case MEM_CANCEL_OFFLINE:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
ret = notifier_from_errno(ret);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* CONFIG_MEMORY_HOTPLUG */
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
/********************************************************************
|
|
|
|
* Basic setup of slabs
|
|
|
|
*******************************************************************/
|
|
|
|
|
|
|
|
void __init kmem_cache_init(void)
|
|
|
|
{
|
|
|
|
int i;
|
2007-06-16 19:16:13 +02:00
|
|
|
int caches = 0;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:26:08 +02:00
|
|
|
init_alloc_cpu();
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
/*
|
|
|
|
* Must first have the slab cache available for the allocations of the
|
2007-05-09 11:32:39 +02:00
|
|
|
* struct kmem_cache_node's. There is special bootstrap code in
|
2007-05-06 23:49:36 +02:00
|
|
|
* kmem_cache_open for slab_state == DOWN.
|
|
|
|
*/
|
|
|
|
create_kmalloc_cache(&kmalloc_caches[0], "kmem_cache_node",
|
|
|
|
sizeof(struct kmem_cache_node), GFP_KERNEL);
|
2007-05-31 09:40:51 +02:00
|
|
|
kmalloc_caches[0].refcount = -1;
|
2007-06-16 19:16:13 +02:00
|
|
|
caches++;
|
2007-10-22 01:41:37 +02:00
|
|
|
|
|
|
|
hotplug_memory_notifier(slab_memory_callback, 1);
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Able to allocate the per node structures */
|
|
|
|
slab_state = PARTIAL;
|
|
|
|
|
|
|
|
/* Caches that are not of the two-to-the-power-of size */
|
2007-06-16 19:16:13 +02:00
|
|
|
if (KMALLOC_MIN_SIZE <= 64) {
|
|
|
|
create_kmalloc_cache(&kmalloc_caches[1],
|
2007-05-06 23:49:36 +02:00
|
|
|
"kmalloc-96", 96, GFP_KERNEL);
|
2007-06-16 19:16:13 +02:00
|
|
|
caches++;
|
|
|
|
}
|
|
|
|
if (KMALLOC_MIN_SIZE <= 128) {
|
|
|
|
create_kmalloc_cache(&kmalloc_caches[2],
|
2007-05-06 23:49:36 +02:00
|
|
|
"kmalloc-192", 192, GFP_KERNEL);
|
2007-06-16 19:16:13 +02:00
|
|
|
caches++;
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2008-02-14 23:28:09 +01:00
|
|
|
for (i = KMALLOC_SHIFT_LOW; i <= PAGE_SHIFT; i++) {
|
2007-05-06 23:49:36 +02:00
|
|
|
create_kmalloc_cache(&kmalloc_caches[i],
|
|
|
|
"kmalloc", 1 << i, GFP_KERNEL);
|
2007-06-16 19:16:13 +02:00
|
|
|
caches++;
|
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-07-17 13:03:26 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Patch up the size_index table if we have strange large alignment
|
|
|
|
* requirements for the kmalloc array. This is only the case for
|
2008-02-16 08:45:26 +01:00
|
|
|
* MIPS it seems. The standard arches will not generate any code here.
|
2007-07-17 13:03:26 +02:00
|
|
|
*
|
|
|
|
* Largest permitted alignment is 256 bytes due to the way we
|
|
|
|
* handle the index determination for the smaller caches.
|
|
|
|
*
|
|
|
|
* Make sure that nothing crazy happens if someone starts tinkering
|
|
|
|
* around with ARCH_KMALLOC_MINALIGN
|
|
|
|
*/
|
|
|
|
BUILD_BUG_ON(KMALLOC_MIN_SIZE > 256 ||
|
|
|
|
(KMALLOC_MIN_SIZE & (KMALLOC_MIN_SIZE - 1)));
|
|
|
|
|
2007-07-17 13:03:28 +02:00
|
|
|
for (i = 8; i < KMALLOC_MIN_SIZE; i += 8)
|
2007-07-17 13:03:26 +02:00
|
|
|
size_index[(i - 1) / 8] = KMALLOC_SHIFT_LOW;
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
slab_state = UP;
|
|
|
|
|
|
|
|
/* Provide the correct kmalloc names now that the caches are up */
|
2008-02-14 23:28:09 +01:00
|
|
|
for (i = KMALLOC_SHIFT_LOW; i <= PAGE_SHIFT; i++)
|
2007-05-06 23:49:36 +02:00
|
|
|
kmalloc_caches[i]. name =
|
|
|
|
kasprintf(GFP_KERNEL, "kmalloc-%d", 1 << i);
|
|
|
|
|
|
|
|
#ifdef CONFIG_SMP
|
|
|
|
register_cpu_notifier(&slab_notifier);
|
2007-10-16 10:26:08 +02:00
|
|
|
kmem_size = offsetof(struct kmem_cache, cpu_slab) +
|
|
|
|
nr_cpu_ids * sizeof(struct kmem_cache_cpu *);
|
|
|
|
#else
|
|
|
|
kmem_size = sizeof(struct kmem_cache);
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
|
2008-02-06 02:57:39 +01:00
|
|
|
printk(KERN_INFO
|
|
|
|
"SLUB: Genslabs=%d, HWalign=%d, Order=%d-%d, MinObjects=%d,"
|
2007-06-16 19:16:13 +02:00
|
|
|
" CPUs=%d, Nodes=%d\n",
|
|
|
|
caches, cache_line_size(),
|
2007-05-06 23:49:36 +02:00
|
|
|
slub_min_order, slub_max_order, slub_min_objects,
|
|
|
|
nr_cpu_ids, nr_node_ids);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Find a mergeable slab cache
|
|
|
|
*/
|
|
|
|
static int slab_unmergeable(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
if (slub_nomerge || (s->flags & SLUB_NEVER_MERGE))
|
|
|
|
return 1;
|
|
|
|
|
2008-02-14 23:28:09 +01:00
|
|
|
if ((s->flags & __PAGE_ALLOC_FALLBACK))
|
2008-02-14 23:28:01 +01:00
|
|
|
return 1;
|
|
|
|
|
2007-05-17 07:10:50 +02:00
|
|
|
if (s->ctor)
|
2007-05-06 23:49:36 +02:00
|
|
|
return 1;
|
|
|
|
|
2007-05-31 09:40:51 +02:00
|
|
|
/*
|
|
|
|
* We may have set a slab to be unmergeable during bootstrap.
|
|
|
|
*/
|
|
|
|
if (s->refcount < 0)
|
|
|
|
return 1;
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct kmem_cache *find_mergeable(size_t size,
|
2007-09-12 00:24:11 +02:00
|
|
|
size_t align, unsigned long flags, const char *name,
|
2007-10-17 08:25:51 +02:00
|
|
|
void (*ctor)(struct kmem_cache *, void *))
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
2007-07-17 13:03:19 +02:00
|
|
|
struct kmem_cache *s;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
if (slub_nomerge || (flags & SLUB_NEVER_MERGE))
|
|
|
|
return NULL;
|
|
|
|
|
2007-05-17 07:10:50 +02:00
|
|
|
if (ctor)
|
2007-05-06 23:49:36 +02:00
|
|
|
return NULL;
|
|
|
|
|
|
|
|
size = ALIGN(size, sizeof(void *));
|
|
|
|
align = calculate_alignment(flags, align, size);
|
|
|
|
size = ALIGN(size, align);
|
2007-09-12 00:24:11 +02:00
|
|
|
flags = kmem_cache_flags(size, flags, name, NULL);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-07-17 13:03:19 +02:00
|
|
|
list_for_each_entry(s, &slab_caches, list) {
|
2007-05-06 23:49:36 +02:00
|
|
|
if (slab_unmergeable(s))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
if (size > s->size)
|
|
|
|
continue;
|
|
|
|
|
2007-09-12 00:24:11 +02:00
|
|
|
if ((flags & SLUB_MERGE_SAME) != (s->flags & SLUB_MERGE_SAME))
|
2007-05-06 23:49:36 +02:00
|
|
|
continue;
|
|
|
|
/*
|
|
|
|
* Check if alignment is compatible.
|
|
|
|
* Courtesy of Adrian Drzewiecki
|
|
|
|
*/
|
2008-01-08 08:20:27 +01:00
|
|
|
if ((s->size & ~(align - 1)) != s->size)
|
2007-05-06 23:49:36 +02:00
|
|
|
continue;
|
|
|
|
|
|
|
|
if (s->size - size >= sizeof(void *))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
struct kmem_cache *kmem_cache_create(const char *name, size_t size,
|
|
|
|
size_t align, unsigned long flags,
|
2007-10-17 08:25:51 +02:00
|
|
|
void (*ctor)(struct kmem_cache *, void *))
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
struct kmem_cache *s;
|
|
|
|
|
|
|
|
down_write(&slub_lock);
|
2007-09-12 00:24:11 +02:00
|
|
|
s = find_mergeable(size, align, flags, name, ctor);
|
2007-05-06 23:49:36 +02:00
|
|
|
if (s) {
|
2007-10-16 10:26:09 +02:00
|
|
|
int cpu;
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
s->refcount++;
|
|
|
|
/*
|
|
|
|
* Adjust the object sizes so that we clear
|
|
|
|
* the complete object on kzalloc.
|
|
|
|
*/
|
|
|
|
s->objsize = max(s->objsize, (int)size);
|
2007-10-16 10:26:09 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* And then we need to update the object size in the
|
|
|
|
* per cpu structures
|
|
|
|
*/
|
|
|
|
for_each_online_cpu(cpu)
|
|
|
|
get_cpu_slab(s, cpu)->objsize = s->objsize;
|
2008-02-16 08:45:26 +01:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
s->inuse = max_t(int, s->inuse, ALIGN(size, sizeof(void *)));
|
2007-07-17 13:03:31 +02:00
|
|
|
up_write(&slub_lock);
|
2008-02-16 08:45:26 +01:00
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
if (sysfs_slab_alias(s, name))
|
|
|
|
goto err;
|
2007-07-17 13:03:31 +02:00
|
|
|
return s;
|
|
|
|
}
|
2008-02-16 08:45:26 +01:00
|
|
|
|
2007-07-17 13:03:31 +02:00
|
|
|
s = kmalloc(kmem_size, GFP_KERNEL);
|
|
|
|
if (s) {
|
|
|
|
if (kmem_cache_open(s, GFP_KERNEL, name,
|
2007-05-17 07:10:50 +02:00
|
|
|
size, align, flags, ctor)) {
|
2007-05-06 23:49:36 +02:00
|
|
|
list_add(&s->list, &slab_caches);
|
2007-07-17 13:03:31 +02:00
|
|
|
up_write(&slub_lock);
|
|
|
|
if (sysfs_slab_add(s))
|
|
|
|
goto err;
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
kfree(s);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
up_write(&slub_lock);
|
|
|
|
|
|
|
|
err:
|
|
|
|
if (flags & SLAB_PANIC)
|
|
|
|
panic("Cannot create slabcache %s\n", name);
|
|
|
|
else
|
|
|
|
s = NULL;
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kmem_cache_create);
|
|
|
|
|
|
|
|
#ifdef CONFIG_SMP
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Use the cpu notifier to insure that the cpu slabs are flushed when
|
|
|
|
* necessary.
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static int __cpuinit slab_cpuup_callback(struct notifier_block *nfb,
|
|
|
|
unsigned long action, void *hcpu)
|
|
|
|
{
|
|
|
|
long cpu = (long)hcpu;
|
2007-07-17 13:03:19 +02:00
|
|
|
struct kmem_cache *s;
|
|
|
|
unsigned long flags;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
switch (action) {
|
2007-10-16 10:26:08 +02:00
|
|
|
case CPU_UP_PREPARE:
|
|
|
|
case CPU_UP_PREPARE_FROZEN:
|
|
|
|
init_alloc_cpu_cpu(cpu);
|
|
|
|
down_read(&slub_lock);
|
|
|
|
list_for_each_entry(s, &slab_caches, list)
|
|
|
|
s->cpu_slab[cpu] = alloc_kmem_cache_cpu(s, cpu,
|
|
|
|
GFP_KERNEL);
|
|
|
|
up_read(&slub_lock);
|
|
|
|
break;
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
case CPU_UP_CANCELED:
|
2007-05-09 11:35:10 +02:00
|
|
|
case CPU_UP_CANCELED_FROZEN:
|
2007-05-06 23:49:36 +02:00
|
|
|
case CPU_DEAD:
|
2007-05-09 11:35:10 +02:00
|
|
|
case CPU_DEAD_FROZEN:
|
2007-07-17 13:03:19 +02:00
|
|
|
down_read(&slub_lock);
|
|
|
|
list_for_each_entry(s, &slab_caches, list) {
|
2007-10-16 10:26:08 +02:00
|
|
|
struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
|
|
|
|
|
2007-07-17 13:03:19 +02:00
|
|
|
local_irq_save(flags);
|
|
|
|
__flush_cpu_slab(s, cpu);
|
|
|
|
local_irq_restore(flags);
|
2007-10-16 10:26:08 +02:00
|
|
|
free_kmem_cache_cpu(c, cpu);
|
|
|
|
s->cpu_slab[cpu] = NULL;
|
2007-07-17 13:03:19 +02:00
|
|
|
}
|
|
|
|
up_read(&slub_lock);
|
2007-05-06 23:49:36 +02:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return NOTIFY_OK;
|
|
|
|
}
|
|
|
|
|
2008-01-08 08:20:27 +01:00
|
|
|
static struct notifier_block __cpuinitdata slab_notifier = {
|
2008-02-06 02:57:39 +01:00
|
|
|
.notifier_call = slab_cpuup_callback
|
2008-01-08 08:20:27 +01:00
|
|
|
};
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
void *__kmalloc_track_caller(size_t size, gfp_t gfpflags, void *caller)
|
|
|
|
{
|
2007-10-16 10:24:38 +02:00
|
|
|
struct kmem_cache *s;
|
|
|
|
|
2008-02-14 23:28:09 +01:00
|
|
|
if (unlikely(size > PAGE_SIZE))
|
2008-02-11 21:47:46 +01:00
|
|
|
return kmalloc_large(size, gfpflags);
|
|
|
|
|
2007-10-16 10:24:38 +02:00
|
|
|
s = get_slab(size, gfpflags);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:24:44 +02:00
|
|
|
if (unlikely(ZERO_OR_NULL_PTR(s)))
|
2007-07-17 13:03:22 +02:00
|
|
|
return s;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-07-17 13:03:28 +02:00
|
|
|
return slab_alloc(s, gfpflags, -1, caller);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
void *__kmalloc_node_track_caller(size_t size, gfp_t gfpflags,
|
|
|
|
int node, void *caller)
|
|
|
|
{
|
2007-10-16 10:24:38 +02:00
|
|
|
struct kmem_cache *s;
|
|
|
|
|
2008-02-14 23:28:09 +01:00
|
|
|
if (unlikely(size > PAGE_SIZE))
|
2008-03-01 22:56:40 +01:00
|
|
|
return kmalloc_large_node(size, gfpflags, node);
|
2008-02-11 21:47:46 +01:00
|
|
|
|
2007-10-16 10:24:38 +02:00
|
|
|
s = get_slab(size, gfpflags);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:24:44 +02:00
|
|
|
if (unlikely(ZERO_OR_NULL_PTR(s)))
|
2007-07-17 13:03:22 +02:00
|
|
|
return s;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-07-17 13:03:28 +02:00
|
|
|
return slab_alloc(s, gfpflags, node, caller);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:44 +02:00
|
|
|
#if defined(CONFIG_SYSFS) && defined(CONFIG_SLUB_DEBUG)
|
2007-07-17 13:03:30 +02:00
|
|
|
static int validate_slab(struct kmem_cache *s, struct page *page,
|
|
|
|
unsigned long *map)
|
2007-05-06 23:49:43 +02:00
|
|
|
{
|
|
|
|
void *p;
|
2008-03-01 22:40:44 +01:00
|
|
|
void *addr = page_address(page);
|
2007-05-06 23:49:43 +02:00
|
|
|
|
|
|
|
if (!check_slab(s, page) ||
|
|
|
|
!on_freelist(s, page, NULL))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/* Now we know that a valid freelist exists */
|
|
|
|
bitmap_zero(map, s->objects);
|
|
|
|
|
2007-05-09 11:32:40 +02:00
|
|
|
for_each_free_object(p, s, page->freelist) {
|
|
|
|
set_bit(slab_index(p, s, addr), map);
|
2007-05-06 23:49:43 +02:00
|
|
|
if (!check_object(s, page, p, 0))
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:40 +02:00
|
|
|
for_each_object(p, s, addr)
|
|
|
|
if (!test_bit(slab_index(p, s, addr), map))
|
2007-05-06 23:49:43 +02:00
|
|
|
if (!check_object(s, page, p, 1))
|
|
|
|
return 0;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:30 +02:00
|
|
|
static void validate_slab_slab(struct kmem_cache *s, struct page *page,
|
|
|
|
unsigned long *map)
|
2007-05-06 23:49:43 +02:00
|
|
|
{
|
|
|
|
if (slab_trylock(page)) {
|
2007-07-17 13:03:30 +02:00
|
|
|
validate_slab(s, page, map);
|
2007-05-06 23:49:43 +02:00
|
|
|
slab_unlock(page);
|
|
|
|
} else
|
|
|
|
printk(KERN_INFO "SLUB %s: Skipped busy slab 0x%p\n",
|
|
|
|
s->name, page);
|
|
|
|
|
|
|
|
if (s->flags & DEBUG_DEFAULT_FLAGS) {
|
2007-05-09 11:32:42 +02:00
|
|
|
if (!SlabDebug(page))
|
|
|
|
printk(KERN_ERR "SLUB %s: SlabDebug not set "
|
2007-05-06 23:49:43 +02:00
|
|
|
"on slab 0x%p\n", s->name, page);
|
|
|
|
} else {
|
2007-05-09 11:32:42 +02:00
|
|
|
if (SlabDebug(page))
|
|
|
|
printk(KERN_ERR "SLUB %s: SlabDebug set on "
|
2007-05-06 23:49:43 +02:00
|
|
|
"slab 0x%p\n", s->name, page);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:30 +02:00
|
|
|
static int validate_slab_node(struct kmem_cache *s,
|
|
|
|
struct kmem_cache_node *n, unsigned long *map)
|
2007-05-06 23:49:43 +02:00
|
|
|
{
|
|
|
|
unsigned long count = 0;
|
|
|
|
struct page *page;
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&n->list_lock, flags);
|
|
|
|
|
|
|
|
list_for_each_entry(page, &n->partial, lru) {
|
2007-07-17 13:03:30 +02:00
|
|
|
validate_slab_slab(s, page, map);
|
2007-05-06 23:49:43 +02:00
|
|
|
count++;
|
|
|
|
}
|
|
|
|
if (count != n->nr_partial)
|
|
|
|
printk(KERN_ERR "SLUB %s: %ld partial slabs counted but "
|
|
|
|
"counter=%ld\n", s->name, count, n->nr_partial);
|
|
|
|
|
|
|
|
if (!(s->flags & SLAB_STORE_USER))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
list_for_each_entry(page, &n->full, lru) {
|
2007-07-17 13:03:30 +02:00
|
|
|
validate_slab_slab(s, page, map);
|
2007-05-06 23:49:43 +02:00
|
|
|
count++;
|
|
|
|
}
|
|
|
|
if (count != atomic_long_read(&n->nr_slabs))
|
|
|
|
printk(KERN_ERR "SLUB: %s %ld slabs counted but "
|
|
|
|
"counter=%ld\n", s->name, count,
|
|
|
|
atomic_long_read(&n->nr_slabs));
|
|
|
|
|
|
|
|
out:
|
|
|
|
spin_unlock_irqrestore(&n->list_lock, flags);
|
|
|
|
return count;
|
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:30 +02:00
|
|
|
static long validate_slab_cache(struct kmem_cache *s)
|
2007-05-06 23:49:43 +02:00
|
|
|
{
|
|
|
|
int node;
|
|
|
|
unsigned long count = 0;
|
2007-07-17 13:03:30 +02:00
|
|
|
unsigned long *map = kmalloc(BITS_TO_LONGS(s->objects) *
|
|
|
|
sizeof(unsigned long), GFP_KERNEL);
|
|
|
|
|
|
|
|
if (!map)
|
|
|
|
return -ENOMEM;
|
2007-05-06 23:49:43 +02:00
|
|
|
|
|
|
|
flush_all(s);
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY) {
|
2007-05-06 23:49:43 +02:00
|
|
|
struct kmem_cache_node *n = get_node(s, node);
|
|
|
|
|
2007-07-17 13:03:30 +02:00
|
|
|
count += validate_slab_node(s, n, map);
|
2007-05-06 23:49:43 +02:00
|
|
|
}
|
2007-07-17 13:03:30 +02:00
|
|
|
kfree(map);
|
2007-05-06 23:49:43 +02:00
|
|
|
return count;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:41 +02:00
|
|
|
#ifdef SLUB_RESILIENCY_TEST
|
|
|
|
static void resiliency_test(void)
|
|
|
|
{
|
|
|
|
u8 *p;
|
|
|
|
|
|
|
|
printk(KERN_ERR "SLUB resiliency testing\n");
|
|
|
|
printk(KERN_ERR "-----------------------\n");
|
|
|
|
printk(KERN_ERR "A. Corruption after allocation\n");
|
|
|
|
|
|
|
|
p = kzalloc(16, GFP_KERNEL);
|
|
|
|
p[16] = 0x12;
|
|
|
|
printk(KERN_ERR "\n1. kmalloc-16: Clobber Redzone/next pointer"
|
|
|
|
" 0x12->0x%p\n\n", p + 16);
|
|
|
|
|
|
|
|
validate_slab_cache(kmalloc_caches + 4);
|
|
|
|
|
|
|
|
/* Hmmm... The next two are dangerous */
|
|
|
|
p = kzalloc(32, GFP_KERNEL);
|
|
|
|
p[32 + sizeof(void *)] = 0x34;
|
|
|
|
printk(KERN_ERR "\n2. kmalloc-32: Clobber next pointer/next slab"
|
2008-02-06 02:57:39 +01:00
|
|
|
" 0x34 -> -0x%p\n", p);
|
|
|
|
printk(KERN_ERR
|
|
|
|
"If allocated object is overwritten then not detectable\n\n");
|
2007-05-09 11:32:41 +02:00
|
|
|
|
|
|
|
validate_slab_cache(kmalloc_caches + 5);
|
|
|
|
p = kzalloc(64, GFP_KERNEL);
|
|
|
|
p += 64 + (get_cycles() & 0xff) * sizeof(void *);
|
|
|
|
*p = 0x56;
|
|
|
|
printk(KERN_ERR "\n3. kmalloc-64: corrupting random byte 0x56->0x%p\n",
|
|
|
|
p);
|
2008-02-06 02:57:39 +01:00
|
|
|
printk(KERN_ERR
|
|
|
|
"If allocated object is overwritten then not detectable\n\n");
|
2007-05-09 11:32:41 +02:00
|
|
|
validate_slab_cache(kmalloc_caches + 6);
|
|
|
|
|
|
|
|
printk(KERN_ERR "\nB. Corruption after free\n");
|
|
|
|
p = kzalloc(128, GFP_KERNEL);
|
|
|
|
kfree(p);
|
|
|
|
*p = 0x78;
|
|
|
|
printk(KERN_ERR "1. kmalloc-128: Clobber first word 0x78->0x%p\n\n", p);
|
|
|
|
validate_slab_cache(kmalloc_caches + 7);
|
|
|
|
|
|
|
|
p = kzalloc(256, GFP_KERNEL);
|
|
|
|
kfree(p);
|
|
|
|
p[50] = 0x9a;
|
2008-02-06 02:57:39 +01:00
|
|
|
printk(KERN_ERR "\n2. kmalloc-256: Clobber 50th byte 0x9a->0x%p\n\n",
|
|
|
|
p);
|
2007-05-09 11:32:41 +02:00
|
|
|
validate_slab_cache(kmalloc_caches + 8);
|
|
|
|
|
|
|
|
p = kzalloc(512, GFP_KERNEL);
|
|
|
|
kfree(p);
|
|
|
|
p[512] = 0xab;
|
|
|
|
printk(KERN_ERR "\n3. kmalloc-512: Clobber redzone 0xab->0x%p\n\n", p);
|
|
|
|
validate_slab_cache(kmalloc_caches + 9);
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
static void resiliency_test(void) {};
|
|
|
|
#endif
|
|
|
|
|
2007-05-06 23:49:45 +02:00
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Generate lists of code addresses where slabcache objects are allocated
|
2007-05-06 23:49:45 +02:00
|
|
|
* and freed.
|
|
|
|
*/
|
|
|
|
|
|
|
|
struct location {
|
|
|
|
unsigned long count;
|
|
|
|
void *addr;
|
2007-05-09 11:32:45 +02:00
|
|
|
long long sum_time;
|
|
|
|
long min_time;
|
|
|
|
long max_time;
|
|
|
|
long min_pid;
|
|
|
|
long max_pid;
|
|
|
|
cpumask_t cpus;
|
|
|
|
nodemask_t nodes;
|
2007-05-06 23:49:45 +02:00
|
|
|
};
|
|
|
|
|
|
|
|
struct loc_track {
|
|
|
|
unsigned long max;
|
|
|
|
unsigned long count;
|
|
|
|
struct location *loc;
|
|
|
|
};
|
|
|
|
|
|
|
|
static void free_loc_track(struct loc_track *t)
|
|
|
|
{
|
|
|
|
if (t->max)
|
|
|
|
free_pages((unsigned long)t->loc,
|
|
|
|
get_order(sizeof(struct location) * t->max));
|
|
|
|
}
|
|
|
|
|
2007-07-17 13:03:20 +02:00
|
|
|
static int alloc_loc_track(struct loc_track *t, unsigned long max, gfp_t flags)
|
2007-05-06 23:49:45 +02:00
|
|
|
{
|
|
|
|
struct location *l;
|
|
|
|
int order;
|
|
|
|
|
|
|
|
order = get_order(sizeof(struct location) * max);
|
|
|
|
|
2007-07-17 13:03:20 +02:00
|
|
|
l = (void *)__get_free_pages(flags, order);
|
2007-05-06 23:49:45 +02:00
|
|
|
if (!l)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (t->count) {
|
|
|
|
memcpy(l, t->loc, sizeof(struct location) * t->count);
|
|
|
|
free_loc_track(t);
|
|
|
|
}
|
|
|
|
t->max = max;
|
|
|
|
t->loc = l;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int add_location(struct loc_track *t, struct kmem_cache *s,
|
2007-05-09 11:32:45 +02:00
|
|
|
const struct track *track)
|
2007-05-06 23:49:45 +02:00
|
|
|
{
|
|
|
|
long start, end, pos;
|
|
|
|
struct location *l;
|
|
|
|
void *caddr;
|
2007-05-09 11:32:45 +02:00
|
|
|
unsigned long age = jiffies - track->when;
|
2007-05-06 23:49:45 +02:00
|
|
|
|
|
|
|
start = -1;
|
|
|
|
end = t->count;
|
|
|
|
|
|
|
|
for ( ; ; ) {
|
|
|
|
pos = start + (end - start + 1) / 2;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* There is nothing at "end". If we end up there
|
|
|
|
* we need to add something to before end.
|
|
|
|
*/
|
|
|
|
if (pos == end)
|
|
|
|
break;
|
|
|
|
|
|
|
|
caddr = t->loc[pos].addr;
|
2007-05-09 11:32:45 +02:00
|
|
|
if (track->addr == caddr) {
|
|
|
|
|
|
|
|
l = &t->loc[pos];
|
|
|
|
l->count++;
|
|
|
|
if (track->when) {
|
|
|
|
l->sum_time += age;
|
|
|
|
if (age < l->min_time)
|
|
|
|
l->min_time = age;
|
|
|
|
if (age > l->max_time)
|
|
|
|
l->max_time = age;
|
|
|
|
|
|
|
|
if (track->pid < l->min_pid)
|
|
|
|
l->min_pid = track->pid;
|
|
|
|
if (track->pid > l->max_pid)
|
|
|
|
l->max_pid = track->pid;
|
|
|
|
|
|
|
|
cpu_set(track->cpu, l->cpus);
|
|
|
|
}
|
|
|
|
node_set(page_to_nid(virt_to_page(track)), l->nodes);
|
2007-05-06 23:49:45 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:45 +02:00
|
|
|
if (track->addr < caddr)
|
2007-05-06 23:49:45 +02:00
|
|
|
end = pos;
|
|
|
|
else
|
|
|
|
start = pos;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-05-09 11:32:39 +02:00
|
|
|
* Not found. Insert new tracking element.
|
2007-05-06 23:49:45 +02:00
|
|
|
*/
|
2007-07-17 13:03:20 +02:00
|
|
|
if (t->count >= t->max && !alloc_loc_track(t, 2 * t->max, GFP_ATOMIC))
|
2007-05-06 23:49:45 +02:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
l = t->loc + pos;
|
|
|
|
if (pos < t->count)
|
|
|
|
memmove(l + 1, l,
|
|
|
|
(t->count - pos) * sizeof(struct location));
|
|
|
|
t->count++;
|
|
|
|
l->count = 1;
|
2007-05-09 11:32:45 +02:00
|
|
|
l->addr = track->addr;
|
|
|
|
l->sum_time = age;
|
|
|
|
l->min_time = age;
|
|
|
|
l->max_time = age;
|
|
|
|
l->min_pid = track->pid;
|
|
|
|
l->max_pid = track->pid;
|
|
|
|
cpus_clear(l->cpus);
|
|
|
|
cpu_set(track->cpu, l->cpus);
|
|
|
|
nodes_clear(l->nodes);
|
|
|
|
node_set(page_to_nid(virt_to_page(track)), l->nodes);
|
2007-05-06 23:49:45 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void process_slab(struct loc_track *t, struct kmem_cache *s,
|
|
|
|
struct page *page, enum track_item alloc)
|
|
|
|
{
|
2008-03-01 22:40:44 +01:00
|
|
|
void *addr = page_address(page);
|
2007-05-09 11:32:40 +02:00
|
|
|
DECLARE_BITMAP(map, s->objects);
|
2007-05-06 23:49:45 +02:00
|
|
|
void *p;
|
|
|
|
|
|
|
|
bitmap_zero(map, s->objects);
|
2007-05-09 11:32:40 +02:00
|
|
|
for_each_free_object(p, s, page->freelist)
|
|
|
|
set_bit(slab_index(p, s, addr), map);
|
2007-05-06 23:49:45 +02:00
|
|
|
|
2007-05-09 11:32:40 +02:00
|
|
|
for_each_object(p, s, addr)
|
2007-05-09 11:32:45 +02:00
|
|
|
if (!test_bit(slab_index(p, s, addr), map))
|
|
|
|
add_location(t, s, get_track(s, p, alloc));
|
2007-05-06 23:49:45 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int list_locations(struct kmem_cache *s, char *buf,
|
|
|
|
enum track_item alloc)
|
|
|
|
{
|
2008-02-01 00:20:50 +01:00
|
|
|
int len = 0;
|
2007-05-06 23:49:45 +02:00
|
|
|
unsigned long i;
|
2007-07-17 13:03:20 +02:00
|
|
|
struct loc_track t = { 0, 0, NULL };
|
2007-05-06 23:49:45 +02:00
|
|
|
int node;
|
|
|
|
|
2007-07-17 13:03:20 +02:00
|
|
|
if (!alloc_loc_track(&t, PAGE_SIZE / sizeof(struct location),
|
2007-10-16 10:26:09 +02:00
|
|
|
GFP_TEMPORARY))
|
2007-07-17 13:03:20 +02:00
|
|
|
return sprintf(buf, "Out of memory\n");
|
2007-05-06 23:49:45 +02:00
|
|
|
|
|
|
|
/* Push back cpu slabs */
|
|
|
|
flush_all(s);
|
|
|
|
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY) {
|
2007-05-06 23:49:45 +02:00
|
|
|
struct kmem_cache_node *n = get_node(s, node);
|
|
|
|
unsigned long flags;
|
|
|
|
struct page *page;
|
|
|
|
|
2007-08-22 23:01:56 +02:00
|
|
|
if (!atomic_long_read(&n->nr_slabs))
|
2007-05-06 23:49:45 +02:00
|
|
|
continue;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&n->list_lock, flags);
|
|
|
|
list_for_each_entry(page, &n->partial, lru)
|
|
|
|
process_slab(&t, s, page, alloc);
|
|
|
|
list_for_each_entry(page, &n->full, lru)
|
|
|
|
process_slab(&t, s, page, alloc);
|
|
|
|
spin_unlock_irqrestore(&n->list_lock, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < t.count; i++) {
|
2007-05-09 11:32:45 +02:00
|
|
|
struct location *l = &t.loc[i];
|
2007-05-06 23:49:45 +02:00
|
|
|
|
2008-02-01 00:20:50 +01:00
|
|
|
if (len > PAGE_SIZE - 100)
|
2007-05-06 23:49:45 +02:00
|
|
|
break;
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf + len, "%7ld ", l->count);
|
2007-05-09 11:32:45 +02:00
|
|
|
|
|
|
|
if (l->addr)
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprint_symbol(buf + len, (unsigned long)l->addr);
|
2007-05-06 23:49:45 +02:00
|
|
|
else
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf + len, "<not-available>");
|
2007-05-09 11:32:45 +02:00
|
|
|
|
|
|
|
if (l->sum_time != l->min_time) {
|
|
|
|
unsigned long remainder;
|
|
|
|
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf + len, " age=%ld/%ld/%ld",
|
2007-05-09 11:32:45 +02:00
|
|
|
l->min_time,
|
|
|
|
div_long_long_rem(l->sum_time, l->count, &remainder),
|
|
|
|
l->max_time);
|
|
|
|
} else
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf + len, " age=%ld",
|
2007-05-09 11:32:45 +02:00
|
|
|
l->min_time);
|
|
|
|
|
|
|
|
if (l->min_pid != l->max_pid)
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf + len, " pid=%ld-%ld",
|
2007-05-09 11:32:45 +02:00
|
|
|
l->min_pid, l->max_pid);
|
|
|
|
else
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf + len, " pid=%ld",
|
2007-05-09 11:32:45 +02:00
|
|
|
l->min_pid);
|
|
|
|
|
2007-06-24 02:16:32 +02:00
|
|
|
if (num_online_cpus() > 1 && !cpus_empty(l->cpus) &&
|
2008-02-01 00:20:50 +01:00
|
|
|
len < PAGE_SIZE - 60) {
|
|
|
|
len += sprintf(buf + len, " cpus=");
|
|
|
|
len += cpulist_scnprintf(buf + len, PAGE_SIZE - len - 50,
|
2007-05-09 11:32:45 +02:00
|
|
|
l->cpus);
|
|
|
|
}
|
|
|
|
|
2007-06-24 02:16:32 +02:00
|
|
|
if (num_online_nodes() > 1 && !nodes_empty(l->nodes) &&
|
2008-02-01 00:20:50 +01:00
|
|
|
len < PAGE_SIZE - 60) {
|
|
|
|
len += sprintf(buf + len, " nodes=");
|
|
|
|
len += nodelist_scnprintf(buf + len, PAGE_SIZE - len - 50,
|
2007-05-09 11:32:45 +02:00
|
|
|
l->nodes);
|
|
|
|
}
|
|
|
|
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf + len, "\n");
|
2007-05-06 23:49:45 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
free_loc_track(&t);
|
|
|
|
if (!t.count)
|
2008-02-01 00:20:50 +01:00
|
|
|
len += sprintf(buf, "No data\n");
|
|
|
|
return len;
|
2007-05-06 23:49:45 +02:00
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
enum slab_stat_type {
|
|
|
|
SL_FULL,
|
|
|
|
SL_PARTIAL,
|
|
|
|
SL_CPU,
|
|
|
|
SL_OBJECTS
|
|
|
|
};
|
|
|
|
|
|
|
|
#define SO_FULL (1 << SL_FULL)
|
|
|
|
#define SO_PARTIAL (1 << SL_PARTIAL)
|
|
|
|
#define SO_CPU (1 << SL_CPU)
|
|
|
|
#define SO_OBJECTS (1 << SL_OBJECTS)
|
|
|
|
|
2008-03-02 21:28:24 +01:00
|
|
|
static ssize_t show_slab_objects(struct kmem_cache *s,
|
|
|
|
char *buf, unsigned long flags)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
|
|
|
unsigned long total = 0;
|
|
|
|
int cpu;
|
|
|
|
int node;
|
|
|
|
int x;
|
|
|
|
unsigned long *nodes;
|
|
|
|
unsigned long *per_cpu;
|
|
|
|
|
|
|
|
nodes = kzalloc(2 * sizeof(unsigned long) * nr_node_ids, GFP_KERNEL);
|
2008-03-02 21:28:24 +01:00
|
|
|
if (!nodes)
|
|
|
|
return -ENOMEM;
|
2007-05-06 23:49:36 +02:00
|
|
|
per_cpu = nodes + nr_node_ids;
|
|
|
|
|
|
|
|
for_each_possible_cpu(cpu) {
|
2007-10-16 10:26:05 +02:00
|
|
|
struct page *page;
|
|
|
|
struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
if (!c)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
page = c->page;
|
2007-10-16 10:26:07 +02:00
|
|
|
node = c->node;
|
|
|
|
if (node < 0)
|
|
|
|
continue;
|
2007-05-06 23:49:36 +02:00
|
|
|
if (page) {
|
|
|
|
if (flags & SO_CPU) {
|
|
|
|
if (flags & SO_OBJECTS)
|
|
|
|
x = page->inuse;
|
|
|
|
else
|
|
|
|
x = 1;
|
|
|
|
total += x;
|
2007-10-16 10:26:07 +02:00
|
|
|
nodes[node] += x;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
2007-10-16 10:26:07 +02:00
|
|
|
per_cpu[node]++;
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY) {
|
2007-05-06 23:49:36 +02:00
|
|
|
struct kmem_cache_node *n = get_node(s, node);
|
|
|
|
|
|
|
|
if (flags & SO_PARTIAL) {
|
|
|
|
if (flags & SO_OBJECTS)
|
|
|
|
x = count_partial(n);
|
|
|
|
else
|
|
|
|
x = n->nr_partial;
|
|
|
|
total += x;
|
|
|
|
nodes[node] += x;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (flags & SO_FULL) {
|
2007-08-22 23:01:56 +02:00
|
|
|
int full_slabs = atomic_long_read(&n->nr_slabs)
|
2007-05-06 23:49:36 +02:00
|
|
|
- per_cpu[node]
|
|
|
|
- n->nr_partial;
|
|
|
|
|
|
|
|
if (flags & SO_OBJECTS)
|
|
|
|
x = full_slabs * s->objects;
|
|
|
|
else
|
|
|
|
x = full_slabs;
|
|
|
|
total += x;
|
|
|
|
nodes[node] += x;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
x = sprintf(buf, "%lu", total);
|
|
|
|
#ifdef CONFIG_NUMA
|
2007-10-16 10:25:33 +02:00
|
|
|
for_each_node_state(node, N_NORMAL_MEMORY)
|
2007-05-06 23:49:36 +02:00
|
|
|
if (nodes[node])
|
|
|
|
x += sprintf(buf + x, " N%d=%lu",
|
|
|
|
node, nodes[node]);
|
|
|
|
#endif
|
|
|
|
kfree(nodes);
|
|
|
|
return x + sprintf(buf + x, "\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
static int any_slab_objects(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
int node;
|
|
|
|
int cpu;
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
for_each_possible_cpu(cpu) {
|
|
|
|
struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
|
|
|
|
|
|
|
|
if (c && c->page)
|
2007-05-06 23:49:36 +02:00
|
|
|
return 1;
|
2007-10-16 10:26:05 +02:00
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
for_each_online_node(node) {
|
2007-05-06 23:49:36 +02:00
|
|
|
struct kmem_cache_node *n = get_node(s, node);
|
|
|
|
|
2007-10-16 10:26:05 +02:00
|
|
|
if (!n)
|
|
|
|
continue;
|
|
|
|
|
2007-08-22 23:01:56 +02:00
|
|
|
if (n->nr_partial || atomic_long_read(&n->nr_slabs))
|
2007-05-06 23:49:36 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define to_slab_attr(n) container_of(n, struct slab_attribute, attr)
|
|
|
|
#define to_slab(n) container_of(n, struct kmem_cache, kobj);
|
|
|
|
|
|
|
|
struct slab_attribute {
|
|
|
|
struct attribute attr;
|
|
|
|
ssize_t (*show)(struct kmem_cache *s, char *buf);
|
|
|
|
ssize_t (*store)(struct kmem_cache *s, const char *x, size_t count);
|
|
|
|
};
|
|
|
|
|
|
|
|
#define SLAB_ATTR_RO(_name) \
|
|
|
|
static struct slab_attribute _name##_attr = __ATTR_RO(_name)
|
|
|
|
|
|
|
|
#define SLAB_ATTR(_name) \
|
|
|
|
static struct slab_attribute _name##_attr = \
|
|
|
|
__ATTR(_name, 0644, _name##_show, _name##_store)
|
|
|
|
|
|
|
|
static ssize_t slab_size_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", s->size);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(slab_size);
|
|
|
|
|
|
|
|
static ssize_t align_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", s->align);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(align);
|
|
|
|
|
|
|
|
static ssize_t object_size_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", s->objsize);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(object_size);
|
|
|
|
|
|
|
|
static ssize_t objs_per_slab_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", s->objects);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(objs_per_slab);
|
|
|
|
|
|
|
|
static ssize_t order_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", s->order);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(order);
|
|
|
|
|
|
|
|
static ssize_t ctor_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
if (s->ctor) {
|
|
|
|
int n = sprint_symbol(buf, (unsigned long)s->ctor);
|
|
|
|
|
|
|
|
return n + sprintf(buf + n, "\n");
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(ctor);
|
|
|
|
|
|
|
|
static ssize_t aliases_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", s->refcount - 1);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(aliases);
|
|
|
|
|
|
|
|
static ssize_t slabs_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
2008-02-16 00:22:21 +01:00
|
|
|
return show_slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(slabs);
|
|
|
|
|
|
|
|
static ssize_t partial_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
2008-02-16 00:22:21 +01:00
|
|
|
return show_slab_objects(s, buf, SO_PARTIAL);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(partial);
|
|
|
|
|
|
|
|
static ssize_t cpu_slabs_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
2008-02-16 00:22:21 +01:00
|
|
|
return show_slab_objects(s, buf, SO_CPU);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(cpu_slabs);
|
|
|
|
|
|
|
|
static ssize_t objects_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
2008-02-16 00:22:21 +01:00
|
|
|
return show_slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU|SO_OBJECTS);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(objects);
|
|
|
|
|
|
|
|
static ssize_t sanity_checks_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_DEBUG_FREE));
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t sanity_checks_store(struct kmem_cache *s,
|
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
|
|
|
s->flags &= ~SLAB_DEBUG_FREE;
|
|
|
|
if (buf[0] == '1')
|
|
|
|
s->flags |= SLAB_DEBUG_FREE;
|
|
|
|
return length;
|
|
|
|
}
|
|
|
|
SLAB_ATTR(sanity_checks);
|
|
|
|
|
|
|
|
static ssize_t trace_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_TRACE));
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t trace_store(struct kmem_cache *s, const char *buf,
|
|
|
|
size_t length)
|
|
|
|
{
|
|
|
|
s->flags &= ~SLAB_TRACE;
|
|
|
|
if (buf[0] == '1')
|
|
|
|
s->flags |= SLAB_TRACE;
|
|
|
|
return length;
|
|
|
|
}
|
|
|
|
SLAB_ATTR(trace);
|
|
|
|
|
|
|
|
static ssize_t reclaim_account_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_RECLAIM_ACCOUNT));
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t reclaim_account_store(struct kmem_cache *s,
|
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
|
|
|
s->flags &= ~SLAB_RECLAIM_ACCOUNT;
|
|
|
|
if (buf[0] == '1')
|
|
|
|
s->flags |= SLAB_RECLAIM_ACCOUNT;
|
|
|
|
return length;
|
|
|
|
}
|
|
|
|
SLAB_ATTR(reclaim_account);
|
|
|
|
|
|
|
|
static ssize_t hwcache_align_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
2007-05-06 23:49:56 +02:00
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_HWCACHE_ALIGN));
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(hwcache_align);
|
|
|
|
|
|
|
|
#ifdef CONFIG_ZONE_DMA
|
|
|
|
static ssize_t cache_dma_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_CACHE_DMA));
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(cache_dma);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
static ssize_t destroy_by_rcu_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_DESTROY_BY_RCU));
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(destroy_by_rcu);
|
|
|
|
|
|
|
|
static ssize_t red_zone_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_RED_ZONE));
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t red_zone_store(struct kmem_cache *s,
|
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
|
|
|
if (any_slab_objects(s))
|
|
|
|
return -EBUSY;
|
|
|
|
|
|
|
|
s->flags &= ~SLAB_RED_ZONE;
|
|
|
|
if (buf[0] == '1')
|
|
|
|
s->flags |= SLAB_RED_ZONE;
|
|
|
|
calculate_sizes(s);
|
|
|
|
return length;
|
|
|
|
}
|
|
|
|
SLAB_ATTR(red_zone);
|
|
|
|
|
|
|
|
static ssize_t poison_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_POISON));
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t poison_store(struct kmem_cache *s,
|
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
|
|
|
if (any_slab_objects(s))
|
|
|
|
return -EBUSY;
|
|
|
|
|
|
|
|
s->flags &= ~SLAB_POISON;
|
|
|
|
if (buf[0] == '1')
|
|
|
|
s->flags |= SLAB_POISON;
|
|
|
|
calculate_sizes(s);
|
|
|
|
return length;
|
|
|
|
}
|
|
|
|
SLAB_ATTR(poison);
|
|
|
|
|
|
|
|
static ssize_t store_user_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return sprintf(buf, "%d\n", !!(s->flags & SLAB_STORE_USER));
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t store_user_store(struct kmem_cache *s,
|
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
|
|
|
if (any_slab_objects(s))
|
|
|
|
return -EBUSY;
|
|
|
|
|
|
|
|
s->flags &= ~SLAB_STORE_USER;
|
|
|
|
if (buf[0] == '1')
|
|
|
|
s->flags |= SLAB_STORE_USER;
|
|
|
|
calculate_sizes(s);
|
|
|
|
return length;
|
|
|
|
}
|
|
|
|
SLAB_ATTR(store_user);
|
|
|
|
|
2007-05-06 23:49:43 +02:00
|
|
|
static ssize_t validate_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t validate_store(struct kmem_cache *s,
|
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
2007-07-17 13:03:30 +02:00
|
|
|
int ret = -EINVAL;
|
|
|
|
|
|
|
|
if (buf[0] == '1') {
|
|
|
|
ret = validate_slab_cache(s);
|
|
|
|
if (ret >= 0)
|
|
|
|
ret = length;
|
|
|
|
}
|
|
|
|
return ret;
|
2007-05-06 23:49:43 +02:00
|
|
|
}
|
|
|
|
SLAB_ATTR(validate);
|
|
|
|
|
2007-05-06 23:49:46 +02:00
|
|
|
static ssize_t shrink_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t shrink_store(struct kmem_cache *s,
|
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
|
|
|
if (buf[0] == '1') {
|
|
|
|
int rc = kmem_cache_shrink(s);
|
|
|
|
|
|
|
|
if (rc)
|
|
|
|
return rc;
|
|
|
|
} else
|
|
|
|
return -EINVAL;
|
|
|
|
return length;
|
|
|
|
}
|
|
|
|
SLAB_ATTR(shrink);
|
|
|
|
|
2007-05-06 23:49:45 +02:00
|
|
|
static ssize_t alloc_calls_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
if (!(s->flags & SLAB_STORE_USER))
|
|
|
|
return -ENOSYS;
|
|
|
|
return list_locations(s, buf, TRACK_ALLOC);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(alloc_calls);
|
|
|
|
|
|
|
|
static ssize_t free_calls_show(struct kmem_cache *s, char *buf)
|
|
|
|
{
|
|
|
|
if (!(s->flags & SLAB_STORE_USER))
|
|
|
|
return -ENOSYS;
|
|
|
|
return list_locations(s, buf, TRACK_FREE);
|
|
|
|
}
|
|
|
|
SLAB_ATTR_RO(free_calls);
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
#ifdef CONFIG_NUMA
|
2008-01-08 08:20:26 +01:00
|
|
|
static ssize_t remote_node_defrag_ratio_show(struct kmem_cache *s, char *buf)
|
2007-05-06 23:49:36 +02:00
|
|
|
{
|
2008-01-08 08:20:26 +01:00
|
|
|
return sprintf(buf, "%d\n", s->remote_node_defrag_ratio / 10);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
2008-01-08 08:20:26 +01:00
|
|
|
static ssize_t remote_node_defrag_ratio_store(struct kmem_cache *s,
|
2007-05-06 23:49:36 +02:00
|
|
|
const char *buf, size_t length)
|
|
|
|
{
|
|
|
|
int n = simple_strtoul(buf, NULL, 10);
|
|
|
|
|
|
|
|
if (n < 100)
|
2008-01-08 08:20:26 +01:00
|
|
|
s->remote_node_defrag_ratio = n * 10;
|
2007-05-06 23:49:36 +02:00
|
|
|
return length;
|
|
|
|
}
|
2008-01-08 08:20:26 +01:00
|
|
|
SLAB_ATTR(remote_node_defrag_ratio);
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
|
2008-02-08 02:47:41 +01:00
|
|
|
#ifdef CONFIG_SLUB_STATS
|
|
|
|
static int show_stat(struct kmem_cache *s, char *buf, enum stat_item si)
|
|
|
|
{
|
|
|
|
unsigned long sum = 0;
|
|
|
|
int cpu;
|
|
|
|
int len;
|
|
|
|
int *data = kmalloc(nr_cpu_ids * sizeof(int), GFP_KERNEL);
|
|
|
|
|
|
|
|
if (!data)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
for_each_online_cpu(cpu) {
|
|
|
|
unsigned x = get_cpu_slab(s, cpu)->stat[si];
|
|
|
|
|
|
|
|
data[cpu] = x;
|
|
|
|
sum += x;
|
|
|
|
}
|
|
|
|
|
|
|
|
len = sprintf(buf, "%lu", sum);
|
|
|
|
|
|
|
|
for_each_online_cpu(cpu) {
|
|
|
|
if (data[cpu] && len < PAGE_SIZE - 20)
|
|
|
|
len += sprintf(buf + len, " c%d=%u", cpu, data[cpu]);
|
|
|
|
}
|
|
|
|
kfree(data);
|
|
|
|
return len + sprintf(buf + len, "\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
#define STAT_ATTR(si, text) \
|
|
|
|
static ssize_t text##_show(struct kmem_cache *s, char *buf) \
|
|
|
|
{ \
|
|
|
|
return show_stat(s, buf, si); \
|
|
|
|
} \
|
|
|
|
SLAB_ATTR_RO(text); \
|
|
|
|
|
|
|
|
STAT_ATTR(ALLOC_FASTPATH, alloc_fastpath);
|
|
|
|
STAT_ATTR(ALLOC_SLOWPATH, alloc_slowpath);
|
|
|
|
STAT_ATTR(FREE_FASTPATH, free_fastpath);
|
|
|
|
STAT_ATTR(FREE_SLOWPATH, free_slowpath);
|
|
|
|
STAT_ATTR(FREE_FROZEN, free_frozen);
|
|
|
|
STAT_ATTR(FREE_ADD_PARTIAL, free_add_partial);
|
|
|
|
STAT_ATTR(FREE_REMOVE_PARTIAL, free_remove_partial);
|
|
|
|
STAT_ATTR(ALLOC_FROM_PARTIAL, alloc_from_partial);
|
|
|
|
STAT_ATTR(ALLOC_SLAB, alloc_slab);
|
|
|
|
STAT_ATTR(ALLOC_REFILL, alloc_refill);
|
|
|
|
STAT_ATTR(FREE_SLAB, free_slab);
|
|
|
|
STAT_ATTR(CPUSLAB_FLUSH, cpuslab_flush);
|
|
|
|
STAT_ATTR(DEACTIVATE_FULL, deactivate_full);
|
|
|
|
STAT_ATTR(DEACTIVATE_EMPTY, deactivate_empty);
|
|
|
|
STAT_ATTR(DEACTIVATE_TO_HEAD, deactivate_to_head);
|
|
|
|
STAT_ATTR(DEACTIVATE_TO_TAIL, deactivate_to_tail);
|
|
|
|
STAT_ATTR(DEACTIVATE_REMOTE_FREES, deactivate_remote_frees);
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
2008-01-08 08:20:27 +01:00
|
|
|
static struct attribute *slab_attrs[] = {
|
2007-05-06 23:49:36 +02:00
|
|
|
&slab_size_attr.attr,
|
|
|
|
&object_size_attr.attr,
|
|
|
|
&objs_per_slab_attr.attr,
|
|
|
|
&order_attr.attr,
|
|
|
|
&objects_attr.attr,
|
|
|
|
&slabs_attr.attr,
|
|
|
|
&partial_attr.attr,
|
|
|
|
&cpu_slabs_attr.attr,
|
|
|
|
&ctor_attr.attr,
|
|
|
|
&aliases_attr.attr,
|
|
|
|
&align_attr.attr,
|
|
|
|
&sanity_checks_attr.attr,
|
|
|
|
&trace_attr.attr,
|
|
|
|
&hwcache_align_attr.attr,
|
|
|
|
&reclaim_account_attr.attr,
|
|
|
|
&destroy_by_rcu_attr.attr,
|
|
|
|
&red_zone_attr.attr,
|
|
|
|
&poison_attr.attr,
|
|
|
|
&store_user_attr.attr,
|
2007-05-06 23:49:43 +02:00
|
|
|
&validate_attr.attr,
|
2007-05-06 23:49:46 +02:00
|
|
|
&shrink_attr.attr,
|
2007-05-06 23:49:45 +02:00
|
|
|
&alloc_calls_attr.attr,
|
|
|
|
&free_calls_attr.attr,
|
2007-05-06 23:49:36 +02:00
|
|
|
#ifdef CONFIG_ZONE_DMA
|
|
|
|
&cache_dma_attr.attr,
|
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_NUMA
|
2008-01-08 08:20:26 +01:00
|
|
|
&remote_node_defrag_ratio_attr.attr,
|
2008-02-08 02:47:41 +01:00
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_SLUB_STATS
|
|
|
|
&alloc_fastpath_attr.attr,
|
|
|
|
&alloc_slowpath_attr.attr,
|
|
|
|
&free_fastpath_attr.attr,
|
|
|
|
&free_slowpath_attr.attr,
|
|
|
|
&free_frozen_attr.attr,
|
|
|
|
&free_add_partial_attr.attr,
|
|
|
|
&free_remove_partial_attr.attr,
|
|
|
|
&alloc_from_partial_attr.attr,
|
|
|
|
&alloc_slab_attr.attr,
|
|
|
|
&alloc_refill_attr.attr,
|
|
|
|
&free_slab_attr.attr,
|
|
|
|
&cpuslab_flush_attr.attr,
|
|
|
|
&deactivate_full_attr.attr,
|
|
|
|
&deactivate_empty_attr.attr,
|
|
|
|
&deactivate_to_head_attr.attr,
|
|
|
|
&deactivate_to_tail_attr.attr,
|
|
|
|
&deactivate_remote_frees_attr.attr,
|
2007-05-06 23:49:36 +02:00
|
|
|
#endif
|
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct attribute_group slab_attr_group = {
|
|
|
|
.attrs = slab_attrs,
|
|
|
|
};
|
|
|
|
|
|
|
|
static ssize_t slab_attr_show(struct kobject *kobj,
|
|
|
|
struct attribute *attr,
|
|
|
|
char *buf)
|
|
|
|
{
|
|
|
|
struct slab_attribute *attribute;
|
|
|
|
struct kmem_cache *s;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
attribute = to_slab_attr(attr);
|
|
|
|
s = to_slab(kobj);
|
|
|
|
|
|
|
|
if (!attribute->show)
|
|
|
|
return -EIO;
|
|
|
|
|
|
|
|
err = attribute->show(s, buf);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t slab_attr_store(struct kobject *kobj,
|
|
|
|
struct attribute *attr,
|
|
|
|
const char *buf, size_t len)
|
|
|
|
{
|
|
|
|
struct slab_attribute *attribute;
|
|
|
|
struct kmem_cache *s;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
attribute = to_slab_attr(attr);
|
|
|
|
s = to_slab(kobj);
|
|
|
|
|
|
|
|
if (!attribute->store)
|
|
|
|
return -EIO;
|
|
|
|
|
|
|
|
err = attribute->store(s, buf, len);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2008-01-08 07:29:05 +01:00
|
|
|
static void kmem_cache_release(struct kobject *kobj)
|
|
|
|
{
|
|
|
|
struct kmem_cache *s = to_slab(kobj);
|
|
|
|
|
|
|
|
kfree(s);
|
|
|
|
}
|
|
|
|
|
2007-05-06 23:49:36 +02:00
|
|
|
static struct sysfs_ops slab_sysfs_ops = {
|
|
|
|
.show = slab_attr_show,
|
|
|
|
.store = slab_attr_store,
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct kobj_type slab_ktype = {
|
|
|
|
.sysfs_ops = &slab_sysfs_ops,
|
2008-01-08 07:29:05 +01:00
|
|
|
.release = kmem_cache_release
|
2007-05-06 23:49:36 +02:00
|
|
|
};
|
|
|
|
|
|
|
|
static int uevent_filter(struct kset *kset, struct kobject *kobj)
|
|
|
|
{
|
|
|
|
struct kobj_type *ktype = get_ktype(kobj);
|
|
|
|
|
|
|
|
if (ktype == &slab_ktype)
|
|
|
|
return 1;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct kset_uevent_ops slab_uevent_ops = {
|
|
|
|
.filter = uevent_filter,
|
|
|
|
};
|
|
|
|
|
2007-11-01 16:29:06 +01:00
|
|
|
static struct kset *slab_kset;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
#define ID_STR_LENGTH 64
|
|
|
|
|
|
|
|
/* Create a unique string id for a slab cache:
|
2008-02-16 08:45:26 +01:00
|
|
|
*
|
|
|
|
* Format :[flags-]size
|
2007-05-06 23:49:36 +02:00
|
|
|
*/
|
|
|
|
static char *create_unique_id(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
char *name = kmalloc(ID_STR_LENGTH, GFP_KERNEL);
|
|
|
|
char *p = name;
|
|
|
|
|
|
|
|
BUG_ON(!name);
|
|
|
|
|
|
|
|
*p++ = ':';
|
|
|
|
/*
|
|
|
|
* First flags affecting slabcache operations. We will only
|
|
|
|
* get here for aliasable slabs so we do not need to support
|
|
|
|
* too many flags. The flags here must cover all flags that
|
|
|
|
* are matched during merging to guarantee that the id is
|
|
|
|
* unique.
|
|
|
|
*/
|
|
|
|
if (s->flags & SLAB_CACHE_DMA)
|
|
|
|
*p++ = 'd';
|
|
|
|
if (s->flags & SLAB_RECLAIM_ACCOUNT)
|
|
|
|
*p++ = 'a';
|
|
|
|
if (s->flags & SLAB_DEBUG_FREE)
|
|
|
|
*p++ = 'F';
|
|
|
|
if (p != name + 1)
|
|
|
|
*p++ = '-';
|
|
|
|
p += sprintf(p, "%07d", s->size);
|
|
|
|
BUG_ON(p > name + ID_STR_LENGTH - 1);
|
|
|
|
return name;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int sysfs_slab_add(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
const char *name;
|
|
|
|
int unmergeable;
|
|
|
|
|
|
|
|
if (slab_state < SYSFS)
|
|
|
|
/* Defer until later */
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
unmergeable = slab_unmergeable(s);
|
|
|
|
if (unmergeable) {
|
|
|
|
/*
|
|
|
|
* Slabcache can never be merged so we can use the name proper.
|
|
|
|
* This is typically the case for debug situations. In that
|
|
|
|
* case we can catch duplicate names easily.
|
|
|
|
*/
|
2007-11-01 16:29:06 +01:00
|
|
|
sysfs_remove_link(&slab_kset->kobj, s->name);
|
2007-05-06 23:49:36 +02:00
|
|
|
name = s->name;
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* Create a unique name for the slab as a target
|
|
|
|
* for the symlinks.
|
|
|
|
*/
|
|
|
|
name = create_unique_id(s);
|
|
|
|
}
|
|
|
|
|
2007-11-01 16:29:06 +01:00
|
|
|
s->kobj.kset = slab_kset;
|
2007-12-18 07:05:35 +01:00
|
|
|
err = kobject_init_and_add(&s->kobj, &slab_ktype, NULL, name);
|
|
|
|
if (err) {
|
|
|
|
kobject_put(&s->kobj);
|
2007-05-06 23:49:36 +02:00
|
|
|
return err;
|
2007-12-18 07:05:35 +01:00
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
err = sysfs_create_group(&s->kobj, &slab_attr_group);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
kobject_uevent(&s->kobj, KOBJ_ADD);
|
|
|
|
if (!unmergeable) {
|
|
|
|
/* Setup first alias */
|
|
|
|
sysfs_slab_alias(s, s->name);
|
|
|
|
kfree(name);
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void sysfs_slab_remove(struct kmem_cache *s)
|
|
|
|
{
|
|
|
|
kobject_uevent(&s->kobj, KOBJ_REMOVE);
|
|
|
|
kobject_del(&s->kobj);
|
2008-01-08 07:29:05 +01:00
|
|
|
kobject_put(&s->kobj);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Need to buffer aliases during bootup until sysfs becomes
|
|
|
|
* available lest we loose that information.
|
|
|
|
*/
|
|
|
|
struct saved_alias {
|
|
|
|
struct kmem_cache *s;
|
|
|
|
const char *name;
|
|
|
|
struct saved_alias *next;
|
|
|
|
};
|
|
|
|
|
2007-07-17 13:03:27 +02:00
|
|
|
static struct saved_alias *alias_list;
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
static int sysfs_slab_alias(struct kmem_cache *s, const char *name)
|
|
|
|
{
|
|
|
|
struct saved_alias *al;
|
|
|
|
|
|
|
|
if (slab_state == SYSFS) {
|
|
|
|
/*
|
|
|
|
* If we have a leftover link then remove it.
|
|
|
|
*/
|
2007-11-01 16:29:06 +01:00
|
|
|
sysfs_remove_link(&slab_kset->kobj, name);
|
|
|
|
return sysfs_create_link(&slab_kset->kobj, &s->kobj, name);
|
2007-05-06 23:49:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
al = kmalloc(sizeof(struct saved_alias), GFP_KERNEL);
|
|
|
|
if (!al)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
al->s = s;
|
|
|
|
al->name = name;
|
|
|
|
al->next = alias_list;
|
|
|
|
alias_list = al;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __init slab_sysfs_init(void)
|
|
|
|
{
|
2007-07-17 13:03:19 +02:00
|
|
|
struct kmem_cache *s;
|
2007-05-06 23:49:36 +02:00
|
|
|
int err;
|
|
|
|
|
2007-11-06 19:36:58 +01:00
|
|
|
slab_kset = kset_create_and_add("slab", &slab_uevent_ops, kernel_kobj);
|
2007-11-01 16:29:06 +01:00
|
|
|
if (!slab_kset) {
|
2007-05-06 23:49:36 +02:00
|
|
|
printk(KERN_ERR "Cannot register slab subsystem.\n");
|
|
|
|
return -ENOSYS;
|
|
|
|
}
|
|
|
|
|
2007-05-09 11:32:39 +02:00
|
|
|
slab_state = SYSFS;
|
|
|
|
|
2007-07-17 13:03:19 +02:00
|
|
|
list_for_each_entry(s, &slab_caches, list) {
|
2007-05-09 11:32:39 +02:00
|
|
|
err = sysfs_slab_add(s);
|
2007-08-31 08:56:26 +02:00
|
|
|
if (err)
|
|
|
|
printk(KERN_ERR "SLUB: Unable to add boot slab %s"
|
|
|
|
" to sysfs\n", s->name);
|
2007-05-09 11:32:39 +02:00
|
|
|
}
|
2007-05-06 23:49:36 +02:00
|
|
|
|
|
|
|
while (alias_list) {
|
|
|
|
struct saved_alias *al = alias_list;
|
|
|
|
|
|
|
|
alias_list = alias_list->next;
|
|
|
|
err = sysfs_slab_alias(al->s, al->name);
|
2007-08-31 08:56:26 +02:00
|
|
|
if (err)
|
|
|
|
printk(KERN_ERR "SLUB: Unable to add boot slab alias"
|
|
|
|
" %s to sysfs\n", s->name);
|
2007-05-06 23:49:36 +02:00
|
|
|
kfree(al);
|
|
|
|
}
|
|
|
|
|
|
|
|
resiliency_test();
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
__initcall(slab_sysfs_init);
|
|
|
|
#endif
|
2008-01-01 17:23:28 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The /proc/slabinfo ABI
|
|
|
|
*/
|
2008-01-02 22:04:48 +01:00
|
|
|
#ifdef CONFIG_SLABINFO
|
|
|
|
|
|
|
|
ssize_t slabinfo_write(struct file *file, const char __user * buffer,
|
|
|
|
size_t count, loff_t *ppos)
|
|
|
|
{
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
2008-01-01 17:23:28 +01:00
|
|
|
|
|
|
|
static void print_slabinfo_header(struct seq_file *m)
|
|
|
|
{
|
|
|
|
seq_puts(m, "slabinfo - version: 2.1\n");
|
|
|
|
seq_puts(m, "# name <active_objs> <num_objs> <objsize> "
|
|
|
|
"<objperslab> <pagesperslab>");
|
|
|
|
seq_puts(m, " : tunables <limit> <batchcount> <sharedfactor>");
|
|
|
|
seq_puts(m, " : slabdata <active_slabs> <num_slabs> <sharedavail>");
|
|
|
|
seq_putc(m, '\n');
|
|
|
|
}
|
|
|
|
|
|
|
|
static void *s_start(struct seq_file *m, loff_t *pos)
|
|
|
|
{
|
|
|
|
loff_t n = *pos;
|
|
|
|
|
|
|
|
down_read(&slub_lock);
|
|
|
|
if (!n)
|
|
|
|
print_slabinfo_header(m);
|
|
|
|
|
|
|
|
return seq_list_start(&slab_caches, *pos);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void *s_next(struct seq_file *m, void *p, loff_t *pos)
|
|
|
|
{
|
|
|
|
return seq_list_next(p, &slab_caches, pos);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void s_stop(struct seq_file *m, void *p)
|
|
|
|
{
|
|
|
|
up_read(&slub_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int s_show(struct seq_file *m, void *p)
|
|
|
|
{
|
|
|
|
unsigned long nr_partials = 0;
|
|
|
|
unsigned long nr_slabs = 0;
|
|
|
|
unsigned long nr_inuse = 0;
|
|
|
|
unsigned long nr_objs;
|
|
|
|
struct kmem_cache *s;
|
|
|
|
int node;
|
|
|
|
|
|
|
|
s = list_entry(p, struct kmem_cache, list);
|
|
|
|
|
|
|
|
for_each_online_node(node) {
|
|
|
|
struct kmem_cache_node *n = get_node(s, node);
|
|
|
|
|
|
|
|
if (!n)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
nr_partials += n->nr_partial;
|
|
|
|
nr_slabs += atomic_long_read(&n->nr_slabs);
|
|
|
|
nr_inuse += count_partial(n);
|
|
|
|
}
|
|
|
|
|
|
|
|
nr_objs = nr_slabs * s->objects;
|
|
|
|
nr_inuse += (nr_slabs - nr_partials) * s->objects;
|
|
|
|
|
|
|
|
seq_printf(m, "%-17s %6lu %6lu %6u %4u %4d", s->name, nr_inuse,
|
|
|
|
nr_objs, s->size, s->objects, (1 << s->order));
|
|
|
|
seq_printf(m, " : tunables %4u %4u %4u", 0, 0, 0);
|
|
|
|
seq_printf(m, " : slabdata %6lu %6lu %6lu", nr_slabs, nr_slabs,
|
|
|
|
0UL);
|
|
|
|
seq_putc(m, '\n');
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
const struct seq_operations slabinfo_op = {
|
|
|
|
.start = s_start,
|
|
|
|
.next = s_next,
|
|
|
|
.stop = s_stop,
|
|
|
|
.show = s_show,
|
|
|
|
};
|
|
|
|
|
2008-01-02 22:04:48 +01:00
|
|
|
#endif /* CONFIG_SLABINFO */
|