slab.c 97.2 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
/*
 * linux/mm/slab.c
 * Written by Mark Hemment, 1996/97.
 * (markhe@nextd.demon.co.uk)
 *
 * kmem_cache_destroy() + some cleanup - 1999 Andrea Arcangeli
 *
 * Major cleanup, different bufctl logic, per-cpu arrays
 *	(c) 2000 Manfred Spraul
 *
 * Cleanup, make the head arrays unconditional, preparation for NUMA
 * 	(c) 2002 Manfred Spraul
 *
 * An implementation of the Slab Allocator as described in outline in;
 *	UNIX Internals: The New Frontiers by Uresh Vahalia
 *	Pub: Prentice Hall	ISBN 0-13-101908-2
 * or with a little more detail in;
 *	The Slab Allocator: An Object-Caching Kernel Memory Allocator
 *	Jeff Bonwick (Sun Microsystems).
 *	Presented at: USENIX Summer 1994 Technical Conference
 *
 * The memory is organized in caches, one cache for each object type.
 * (e.g. inode_cache, dentry_cache, buffer_head, vm_area_struct)
 * Each cache consists out of many slabs (they are small (usually one
 * page long) and always contiguous), and each slab contains multiple
 * initialized objects.
 *
 * This means, that your constructor is used only for newly allocated
 * slabs and you must pass objects with the same intializations to
 * kmem_cache_free.
 *
 * Each cache can only support one memory type (GFP_DMA, GFP_HIGHMEM,
 * normal). If you need a special memory type, then must create a new
 * cache for that memory type.
 *
 * In order to reduce fragmentation, the slabs are sorted in 3 groups:
 *   full slabs with 0 free objects
 *   partial slabs
 *   empty slabs with no allocated objects
 *
 * If partial slabs exist, then new allocations come from these slabs,
 * otherwise from empty slabs or new slabs are allocated.
 *
 * kmem_cache_destroy() CAN CRASH if you try to allocate from the cache
 * during kmem_cache_destroy(). The caller must prevent concurrent allocs.
 *
 * Each cache has a short per-cpu head array, most allocs
 * and frees go into that array, and if that array overflows, then 1/2
 * of the entries in the array are given back into the global cache.
 * The head array is strictly LIFO and should improve the cache hit rates.
 * On SMP, it additionally reduces the spinlock operations.
 *
 * The c_cpuarray may not be read with enabled local interrupts - 
 * it's changed with a smp_call_function().
 *
 * SMP synchronization:
 *  constructors and destructors are called without any locking.
 *  Several members in kmem_cache_t and struct slab never change, they
 *	are accessed without any locking.
 *  The per-cpu arrays are never accessed from the wrong cpu, no locking,
 *  	and local interrupts are disabled so slab code is preempt-safe.
 *  The non-constant members are protected with a per-cache irq spinlock.
 *
 * Many thanks to Mark Hemment, who wrote another per-cpu slab patch
 * in 2000 - many ideas in the current implementation are derived from
 * his patch.
 *
 * Further notes from the original documentation:
 *
 * 11 April '97.  Started multi-threading - markhe
Ingo Molnar's avatar
Ingo Molnar committed
71
 *	The global cache-chain is protected by the mutex 'cache_chain_mutex'.
Linus Torvalds's avatar
Linus Torvalds committed
72
73
74
75
76
77
 *	The sem is only needed when accessing/extending the cache-chain, which
 *	can never happen inside an interrupt (kmem_cache_create(),
 *	kmem_cache_shrink() and kmem_cache_reap()).
 *
 *	At present, each engine can be growing a cache.  This should be blocked.
 *
78
79
80
81
82
83
84
85
86
 * 15 March 2005. NUMA slab allocator.
 *	Shai Fultheim <shai@scalex86.org>.
 *	Shobhit Dayal <shobhit@calsoftinc.com>
 *	Alok N Kataria <alokk@calsoftinc.com>
 *	Christoph Lameter <christoph@lameter.com>
 *
 *	Modified the slab allocator to be node aware on NUMA systems.
 *	Each node has its own list of partial, free and full slabs.
 *	All object allocations for a node occur from node specific slab lists.
Linus Torvalds's avatar
Linus Torvalds committed
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
 */

#include	<linux/config.h>
#include	<linux/slab.h>
#include	<linux/mm.h>
#include	<linux/swap.h>
#include	<linux/cache.h>
#include	<linux/interrupt.h>
#include	<linux/init.h>
#include	<linux/compiler.h>
#include	<linux/seq_file.h>
#include	<linux/notifier.h>
#include	<linux/kallsyms.h>
#include	<linux/cpu.h>
#include	<linux/sysctl.h>
#include	<linux/module.h>
#include	<linux/rcupdate.h>
104
#include	<linux/string.h>
105
#include	<linux/nodemask.h>
106
#include	<linux/mempolicy.h>
Ingo Molnar's avatar
Ingo Molnar committed
107
#include	<linux/mutex.h>
Linus Torvalds's avatar
Linus Torvalds committed
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202

#include	<asm/uaccess.h>
#include	<asm/cacheflush.h>
#include	<asm/tlbflush.h>
#include	<asm/page.h>

/*
 * DEBUG	- 1 for kmem_cache_create() to honour; SLAB_DEBUG_INITIAL,
 *		  SLAB_RED_ZONE & SLAB_POISON.
 *		  0 for faster, smaller code (especially in the critical paths).
 *
 * STATS	- 1 to collect stats for /proc/slabinfo.
 *		  0 for faster, smaller code (especially in the critical paths).
 *
 * FORCED_DEBUG	- 1 enables SLAB_RED_ZONE and SLAB_POISON (if possible)
 */

#ifdef CONFIG_DEBUG_SLAB
#define	DEBUG		1
#define	STATS		1
#define	FORCED_DEBUG	1
#else
#define	DEBUG		0
#define	STATS		0
#define	FORCED_DEBUG	0
#endif

/* Shouldn't this be in a header file somewhere? */
#define	BYTES_PER_WORD		sizeof(void *)

#ifndef cache_line_size
#define cache_line_size()	L1_CACHE_BYTES
#endif

#ifndef ARCH_KMALLOC_MINALIGN
/*
 * Enforce a minimum alignment for the kmalloc caches.
 * Usually, the kmalloc caches are cache_line_size() aligned, except when
 * DEBUG and FORCED_DEBUG are enabled, then they are BYTES_PER_WORD aligned.
 * Some archs want to perform DMA into kmalloc caches and need a guaranteed
 * alignment larger than BYTES_PER_WORD. ARCH_KMALLOC_MINALIGN allows that.
 * Note that this flag disables some debug features.
 */
#define ARCH_KMALLOC_MINALIGN 0
#endif

#ifndef ARCH_SLAB_MINALIGN
/*
 * Enforce a minimum alignment for all caches.
 * Intended for archs that get misalignment faults even for BYTES_PER_WORD
 * aligned buffers. Includes ARCH_KMALLOC_MINALIGN.
 * If possible: Do not enable this flag for CONFIG_DEBUG_SLAB, it disables
 * some debug features.
 */
#define ARCH_SLAB_MINALIGN 0
#endif

#ifndef ARCH_KMALLOC_FLAGS
#define ARCH_KMALLOC_FLAGS SLAB_HWCACHE_ALIGN
#endif

/* Legal flag mask for kmem_cache_create(). */
#if DEBUG
# define CREATE_MASK	(SLAB_DEBUG_INITIAL | SLAB_RED_ZONE | \
			 SLAB_POISON | SLAB_HWCACHE_ALIGN | \
			 SLAB_NO_REAP | SLAB_CACHE_DMA | \
			 SLAB_MUST_HWCACHE_ALIGN | SLAB_STORE_USER | \
			 SLAB_RECLAIM_ACCOUNT | SLAB_PANIC | \
			 SLAB_DESTROY_BY_RCU)
#else
# define CREATE_MASK	(SLAB_HWCACHE_ALIGN | SLAB_NO_REAP | \
			 SLAB_CACHE_DMA | SLAB_MUST_HWCACHE_ALIGN | \
			 SLAB_RECLAIM_ACCOUNT | SLAB_PANIC | \
			 SLAB_DESTROY_BY_RCU)
#endif

/*
 * kmem_bufctl_t:
 *
 * Bufctl's are used for linking objs within a slab
 * linked offsets.
 *
 * This implementation relies on "struct page" for locating the cache &
 * slab an object belongs to.
 * This allows the bufctl structure to be small (one int), but limits
 * the number of objects a slab (not a cache) can contain when off-slab
 * bufctls are used. The limit is the size of the largest general cache
 * that does not use off-slab slabs.
 * For 32bit archs with 4 kB pages, is this 56.
 * This is not serious, as it is only for large objects, when it is unwise
 * to have too many per slab.
 * Note: This limit can be raised by introducing a general cache whose size
 * is less than 512 (PAGE_SIZE<<3), but greater than 256.
 */

203
typedef unsigned int kmem_bufctl_t;
Linus Torvalds's avatar
Linus Torvalds committed
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
#define BUFCTL_END	(((kmem_bufctl_t)(~0U))-0)
#define BUFCTL_FREE	(((kmem_bufctl_t)(~0U))-1)
#define	SLAB_LIMIT	(((kmem_bufctl_t)(~0U))-2)

/* Max number of objs-per-slab for caches which use off-slab slabs.
 * Needed to avoid a possible looping condition in cache_grow().
 */
static unsigned long offslab_limit;

/*
 * struct slab
 *
 * Manages the objs in a slab. Placed either at the beginning of mem allocated
 * for a slab, or allocated from an general cache.
 * Slabs are chained into three list: fully used, partial, fully free slabs.
 */
struct slab {
221
222
223
224
225
226
	struct list_head list;
	unsigned long colouroff;
	void *s_mem;		/* including colour offset */
	unsigned int inuse;	/* num of objs active in slab */
	kmem_bufctl_t free;
	unsigned short nodeid;
Linus Torvalds's avatar
Linus Torvalds committed
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
};

/*
 * struct slab_rcu
 *
 * slab_destroy on a SLAB_DESTROY_BY_RCU cache uses this structure to
 * arrange for kmem_freepages to be called via RCU.  This is useful if
 * we need to approach a kernel structure obliquely, from its address
 * obtained without the usual locking.  We can lock the structure to
 * stabilize it and check it's still at the given address, only if we
 * can be sure that the memory has not been meanwhile reused for some
 * other kind of object (which our subsystem's lock might corrupt).
 *
 * rcu_read_lock before reading the address, then rcu_read_unlock after
 * taking the spinlock within the structure expected at that address.
 *
 * We assume struct slab_rcu can overlay struct slab when destroying.
 */
struct slab_rcu {
246
247
248
	struct rcu_head head;
	kmem_cache_t *cachep;
	void *addr;
Linus Torvalds's avatar
Linus Torvalds committed
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
};

/*
 * struct array_cache
 *
 * Purpose:
 * - LIFO ordering, to hand out cache-warm objects from _alloc
 * - reduce the number of linked list operations
 * - reduce spinlock operations
 *
 * The limit is stored in the per-cpu structure to reduce the data cache
 * footprint.
 *
 */
struct array_cache {
	unsigned int avail;
	unsigned int limit;
	unsigned int batchcount;
	unsigned int touched;
268
269
270
271
272
273
274
	spinlock_t lock;
	void *entry[0];		/*
				 * Must have this definition in here for the proper
				 * alignment of array_cache. Also simplifies accessing
				 * the entries.
				 * [0] is for gcc 2.95. It should really be [].
				 */
Linus Torvalds's avatar
Linus Torvalds committed
275
276
277
278
279
280
281
282
};

/* bootstrap: The caches do not work without cpuarrays anymore,
 * but the cpuarrays are allocated from the generic caches...
 */
#define BOOT_CPUCACHE_ENTRIES	1
struct arraycache_init {
	struct array_cache cache;
283
	void *entries[BOOT_CPUCACHE_ENTRIES];
Linus Torvalds's avatar
Linus Torvalds committed
284
285
286
};

/*
287
 * The slab lists for all objects.
Linus Torvalds's avatar
Linus Torvalds committed
288
289
 */
struct kmem_list3 {
290
291
292
293
294
295
296
297
298
299
	struct list_head slabs_partial;	/* partial list first, better asm code */
	struct list_head slabs_full;
	struct list_head slabs_free;
	unsigned long free_objects;
	unsigned long next_reap;
	int free_touched;
	unsigned int free_limit;
	spinlock_t list_lock;
	struct array_cache *shared;	/* shared per node */
	struct array_cache **alien;	/* on other nodes */
Linus Torvalds's avatar
Linus Torvalds committed
300
301
};

302
303
304
305
306
307
308
309
310
311
/*
 * Need this for bootstrapping a per node allocator.
 */
#define NUM_INIT_LISTS (2 * MAX_NUMNODES + 1)
struct kmem_list3 __initdata initkmem_list3[NUM_INIT_LISTS];
#define	CACHE_CACHE 0
#define	SIZE_AC 1
#define	SIZE_L3 (1 + MAX_NUMNODES)

/*
312
 * This function must be completely optimized away if
313
314
315
316
 * a constant is passed to it. Mostly the same as
 * what is in linux/slab.h except it returns an
 * index.
 */
317
static __always_inline int index_of(const size_t size)
318
{
319
320
	extern void __bad_size(void);

321
322
323
324
325
326
327
328
329
330
	if (__builtin_constant_p(size)) {
		int i = 0;

#define CACHE(x) \
	if (size <=x) \
		return i; \
	else \
		i++;
#include "linux/kmalloc_sizes.h"
#undef CACHE
331
		__bad_size();
332
	} else
333
		__bad_size();
334
335
336
337
338
	return 0;
}

#define INDEX_AC index_of(sizeof(struct arraycache_init))
#define INDEX_L3 index_of(sizeof(struct kmem_list3))
Linus Torvalds's avatar
Linus Torvalds committed
339

340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
static inline void kmem_list3_init(struct kmem_list3 *parent)
{
	INIT_LIST_HEAD(&parent->slabs_full);
	INIT_LIST_HEAD(&parent->slabs_partial);
	INIT_LIST_HEAD(&parent->slabs_free);
	parent->shared = NULL;
	parent->alien = NULL;
	spin_lock_init(&parent->list_lock);
	parent->free_objects = 0;
	parent->free_touched = 0;
}

#define MAKE_LIST(cachep, listp, slab, nodeid)	\
	do {	\
		INIT_LIST_HEAD(listp);		\
		list_splice(&(cachep->nodelists[nodeid]->slab), listp); \
	} while (0)

#define	MAKE_ALL_LISTS(cachep, ptr, nodeid)			\
	do {					\
	MAKE_LIST((cachep), (&(ptr)->slabs_full), slabs_full, nodeid);	\
	MAKE_LIST((cachep), (&(ptr)->slabs_partial), slabs_partial, nodeid); \
	MAKE_LIST((cachep), (&(ptr)->slabs_free), slabs_free, nodeid);	\
	} while (0)
Linus Torvalds's avatar
Linus Torvalds committed
364
365
366
367
368
369

/*
 * kmem_cache_t
 *
 * manages a cache.
 */
370

371
struct kmem_cache {
Linus Torvalds's avatar
Linus Torvalds committed
372
/* 1) per-cpu data, touched during every alloc/free */
373
374
375
376
	struct array_cache *array[NR_CPUS];
	unsigned int batchcount;
	unsigned int limit;
	unsigned int shared;
377
	unsigned int buffer_size;
378
/* 2) touched by every alloc & free from the backend */
379
380
381
382
	struct kmem_list3 *nodelists[MAX_NUMNODES];
	unsigned int flags;	/* constant flags */
	unsigned int num;	/* # of objs per slab */
	spinlock_t spinlock;
Linus Torvalds's avatar
Linus Torvalds committed
383
384
385

/* 3) cache_grow/shrink */
	/* order of pgs per slab (2^n) */
386
	unsigned int gfporder;
Linus Torvalds's avatar
Linus Torvalds committed
387
388

	/* force GFP flags, e.g. GFP_DMA */
389
	gfp_t gfpflags;
Linus Torvalds's avatar
Linus Torvalds committed
390

391
392
393
394
395
396
	size_t colour;		/* cache colouring range */
	unsigned int colour_off;	/* colour offset */
	unsigned int colour_next;	/* cache colouring */
	kmem_cache_t *slabp_cache;
	unsigned int slab_size;
	unsigned int dflags;	/* dynamic flags */
Linus Torvalds's avatar
Linus Torvalds committed
397
398

	/* constructor func */
399
	void (*ctor) (void *, kmem_cache_t *, unsigned long);
Linus Torvalds's avatar
Linus Torvalds committed
400
401

	/* de-constructor func */
402
	void (*dtor) (void *, kmem_cache_t *, unsigned long);
Linus Torvalds's avatar
Linus Torvalds committed
403
404

/* 4) cache creation/removal */
405
406
	const char *name;
	struct list_head next;
Linus Torvalds's avatar
Linus Torvalds committed
407
408
409

/* 5) statistics */
#if STATS
410
411
412
413
414
415
416
417
418
419
420
421
422
	unsigned long num_active;
	unsigned long num_allocations;
	unsigned long high_mark;
	unsigned long grown;
	unsigned long reaped;
	unsigned long errors;
	unsigned long max_freeable;
	unsigned long node_allocs;
	unsigned long node_frees;
	atomic_t allochit;
	atomic_t allocmiss;
	atomic_t freehit;
	atomic_t freemiss;
Linus Torvalds's avatar
Linus Torvalds committed
423
424
#endif
#if DEBUG
425
426
427
428
429
430
431
432
	/*
	 * If debugging is enabled, then the allocator can add additional
	 * fields and/or padding to every object. buffer_size contains the total
	 * object size including these internal fields, the following two
	 * variables contain the offset to the user object and its size.
	 */
	int obj_offset;
	int obj_size;
Linus Torvalds's avatar
Linus Torvalds committed
433
434
435
436
437
438
439
440
441
442
#endif
};

#define CFLGS_OFF_SLAB		(0x80000000UL)
#define	OFF_SLAB(x)	((x)->flags & CFLGS_OFF_SLAB)

#define BATCHREFILL_LIMIT	16
/* Optimization question: fewer reaps means less 
 * probability for unnessary cpucache drain/refill cycles.
 *
Adrian Bunk's avatar
Adrian Bunk committed
443
 * OTOH the cpuarrays can contain lots of objects,
Linus Torvalds's avatar
Linus Torvalds committed
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
 * which could lock up otherwise freeable slabs.
 */
#define REAPTIMEOUT_CPUC	(2*HZ)
#define REAPTIMEOUT_LIST3	(4*HZ)

#if STATS
#define	STATS_INC_ACTIVE(x)	((x)->num_active++)
#define	STATS_DEC_ACTIVE(x)	((x)->num_active--)
#define	STATS_INC_ALLOCED(x)	((x)->num_allocations++)
#define	STATS_INC_GROWN(x)	((x)->grown++)
#define	STATS_INC_REAPED(x)	((x)->reaped++)
#define	STATS_SET_HIGH(x)	do { if ((x)->num_active > (x)->high_mark) \
					(x)->high_mark = (x)->num_active; \
				} while (0)
#define	STATS_INC_ERR(x)	((x)->errors++)
#define	STATS_INC_NODEALLOCS(x)	((x)->node_allocs++)
460
#define	STATS_INC_NODEFREES(x)	((x)->node_frees++)
Linus Torvalds's avatar
Linus Torvalds committed
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
#define	STATS_SET_FREEABLE(x, i) \
				do { if ((x)->max_freeable < i) \
					(x)->max_freeable = i; \
				} while (0)

#define STATS_INC_ALLOCHIT(x)	atomic_inc(&(x)->allochit)
#define STATS_INC_ALLOCMISS(x)	atomic_inc(&(x)->allocmiss)
#define STATS_INC_FREEHIT(x)	atomic_inc(&(x)->freehit)
#define STATS_INC_FREEMISS(x)	atomic_inc(&(x)->freemiss)
#else
#define	STATS_INC_ACTIVE(x)	do { } while (0)
#define	STATS_DEC_ACTIVE(x)	do { } while (0)
#define	STATS_INC_ALLOCED(x)	do { } while (0)
#define	STATS_INC_GROWN(x)	do { } while (0)
#define	STATS_INC_REAPED(x)	do { } while (0)
#define	STATS_SET_HIGH(x)	do { } while (0)
#define	STATS_INC_ERR(x)	do { } while (0)
#define	STATS_INC_NODEALLOCS(x)	do { } while (0)
479
#define	STATS_INC_NODEFREES(x)	do { } while (0)
Linus Torvalds's avatar
Linus Torvalds committed
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
#define	STATS_SET_FREEABLE(x, i) \
				do { } while (0)

#define STATS_INC_ALLOCHIT(x)	do { } while (0)
#define STATS_INC_ALLOCMISS(x)	do { } while (0)
#define STATS_INC_FREEHIT(x)	do { } while (0)
#define STATS_INC_FREEMISS(x)	do { } while (0)
#endif

#if DEBUG
/* Magic nums for obj red zoning.
 * Placed in the first word before and the first word after an obj.
 */
#define	RED_INACTIVE	0x5A2CF071UL	/* when obj is inactive */
#define	RED_ACTIVE	0x170FC2A5UL	/* when obj is active */

/* ...and for poisoning */
#define	POISON_INUSE	0x5a	/* for use-uninitialised poisoning */
#define POISON_FREE	0x6b	/* for use-after-free poisoning */
#define	POISON_END	0xa5	/* end-byte of poisoning */

/* memory layout of objects:
 * 0		: objp
503
 * 0 .. cachep->obj_offset - BYTES_PER_WORD - 1: padding. This ensures that
Linus Torvalds's avatar
Linus Torvalds committed
504
505
 * 		the end of an object is aligned with the end of the real
 * 		allocation. Catches writes behind the end of the allocation.
506
 * cachep->obj_offset - BYTES_PER_WORD .. cachep->obj_offset - 1:
Linus Torvalds's avatar
Linus Torvalds committed
507
 * 		redzone word.
508
509
510
 * cachep->obj_offset: The real object.
 * cachep->buffer_size - 2* BYTES_PER_WORD: redzone word [BYTES_PER_WORD long]
 * cachep->buffer_size - 1* BYTES_PER_WORD: last caller address [BYTES_PER_WORD long]
Linus Torvalds's avatar
Linus Torvalds committed
511
 */
512
static int obj_offset(kmem_cache_t *cachep)
Linus Torvalds's avatar
Linus Torvalds committed
513
{
514
	return cachep->obj_offset;
Linus Torvalds's avatar
Linus Torvalds committed
515
516
}

517
static int obj_size(kmem_cache_t *cachep)
Linus Torvalds's avatar
Linus Torvalds committed
518
{
519
	return cachep->obj_size;
Linus Torvalds's avatar
Linus Torvalds committed
520
521
522
523
524
}

static unsigned long *dbg_redzone1(kmem_cache_t *cachep, void *objp)
{
	BUG_ON(!(cachep->flags & SLAB_RED_ZONE));
525
	return (unsigned long*) (objp+obj_offset(cachep)-BYTES_PER_WORD);
Linus Torvalds's avatar
Linus Torvalds committed
526
527
528
529
530
531
}

static unsigned long *dbg_redzone2(kmem_cache_t *cachep, void *objp)
{
	BUG_ON(!(cachep->flags & SLAB_RED_ZONE));
	if (cachep->flags & SLAB_STORE_USER)
532
		return (unsigned long *)(objp + cachep->buffer_size -
533
					 2 * BYTES_PER_WORD);
534
	return (unsigned long *)(objp + cachep->buffer_size - BYTES_PER_WORD);
Linus Torvalds's avatar
Linus Torvalds committed
535
536
537
538
539
}

static void **dbg_userword(kmem_cache_t *cachep, void *objp)
{
	BUG_ON(!(cachep->flags & SLAB_STORE_USER));
540
	return (void **)(objp + cachep->buffer_size - BYTES_PER_WORD);
Linus Torvalds's avatar
Linus Torvalds committed
541
542
543
544
}

#else

545
546
#define obj_offset(x)			0
#define obj_size(cachep)		(cachep->buffer_size)
Linus Torvalds's avatar
Linus Torvalds committed
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
#define dbg_redzone1(cachep, objp)	({BUG(); (unsigned long *)NULL;})
#define dbg_redzone2(cachep, objp)	({BUG(); (unsigned long *)NULL;})
#define dbg_userword(cachep, objp)	({BUG(); (void **)NULL;})

#endif

/*
 * Maximum size of an obj (in 2^order pages)
 * and absolute limit for the gfp order.
 */
#if defined(CONFIG_LARGE_ALLOCS)
#define	MAX_OBJ_ORDER	13	/* up to 32Mb */
#define	MAX_GFP_ORDER	13	/* up to 32Mb */
#elif defined(CONFIG_MMU)
#define	MAX_OBJ_ORDER	5	/* 32 pages */
#define	MAX_GFP_ORDER	5	/* 32 pages */
#else
#define	MAX_OBJ_ORDER	8	/* up to 1Mb */
#define	MAX_GFP_ORDER	8	/* up to 1Mb */
#endif

/*
 * Do not go above this order unless 0 objects fit into the slab.
 */
#define	BREAK_GFP_ORDER_HI	1
#define	BREAK_GFP_ORDER_LO	0
static int slab_break_gfp_order = BREAK_GFP_ORDER_LO;

575
/* Functions for storing/retrieving the cachep and or slab from the
Linus Torvalds's avatar
Linus Torvalds committed
576
577
578
 * global 'mem_map'. These are used to find the slab an obj belongs to.
 * With kfree(), these are used to find the cache which an obj belongs to.
 */
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
static inline void page_set_cache(struct page *page, struct kmem_cache *cache)
{
	page->lru.next = (struct list_head *)cache;
}

static inline struct kmem_cache *page_get_cache(struct page *page)
{
	return (struct kmem_cache *)page->lru.next;
}

static inline void page_set_slab(struct page *page, struct slab *slab)
{
	page->lru.prev = (struct list_head *)slab;
}

static inline struct slab *page_get_slab(struct page *page)
{
	return (struct slab *)page->lru.prev;
}
Linus Torvalds's avatar
Linus Torvalds committed
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616

/* These are the default caches for kmalloc. Custom caches can have other sizes. */
struct cache_sizes malloc_sizes[] = {
#define CACHE(x) { .cs_size = (x) },
#include <linux/kmalloc_sizes.h>
	CACHE(ULONG_MAX)
#undef CACHE
};
EXPORT_SYMBOL(malloc_sizes);

/* Must match cache_sizes above. Out of line to keep cache footprint low. */
struct cache_names {
	char *name;
	char *name_dma;
};

static struct cache_names __initdata cache_names[] = {
#define CACHE(x) { .name = "size-" #x, .name_dma = "size-" #x "(DMA)" },
#include <linux/kmalloc_sizes.h>
617
	{NULL,}
Linus Torvalds's avatar
Linus Torvalds committed
618
619
620
621
#undef CACHE
};

static struct arraycache_init initarray_cache __initdata =
622
    { {0, BOOT_CPUCACHE_ENTRIES, 1, 0} };
Linus Torvalds's avatar
Linus Torvalds committed
623
static struct arraycache_init initarray_generic =
624
    { {0, BOOT_CPUCACHE_ENTRIES, 1, 0} };
Linus Torvalds's avatar
Linus Torvalds committed
625
626
627

/* internal cache of cache description objs */
static kmem_cache_t cache_cache = {
628
629
630
	.batchcount = 1,
	.limit = BOOT_CPUCACHE_ENTRIES,
	.shared = 1,
631
	.buffer_size = sizeof(kmem_cache_t),
632
633
634
	.flags = SLAB_NO_REAP,
	.spinlock = SPIN_LOCK_UNLOCKED,
	.name = "kmem_cache",
Linus Torvalds's avatar
Linus Torvalds committed
635
#if DEBUG
636
	.obj_size = sizeof(kmem_cache_t),
Linus Torvalds's avatar
Linus Torvalds committed
637
638
639
640
#endif
};

/* Guard access to the cache-chain. */
Ingo Molnar's avatar
Ingo Molnar committed
641
static DEFINE_MUTEX(cache_chain_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
static struct list_head cache_chain;

/*
 * vm_enough_memory() looks at this to determine how many
 * slab-allocated pages are possibly freeable under pressure
 *
 * SLAB_RECLAIM_ACCOUNT turns this on per-slab
 */
atomic_t slab_reclaim_pages;

/*
 * chicken and egg problem: delay the per-cpu array allocation
 * until the general caches are up.
 */
static enum {
	NONE,
658
659
	PARTIAL_AC,
	PARTIAL_L3,
Linus Torvalds's avatar
Linus Torvalds committed
660
661
662
663
664
	FULL
} g_cpucache_up;

static DEFINE_PER_CPU(struct work_struct, reap_work);

665
666
667
static void free_block(kmem_cache_t *cachep, void **objpp, int len, int node);
static void enable_cpucache(kmem_cache_t *cachep);
static void cache_reap(void *unused);
668
static int __node_shrink(kmem_cache_t *cachep, int node);
Linus Torvalds's avatar
Linus Torvalds committed
669
670
671
672
673
674

static inline struct array_cache *ac_data(kmem_cache_t *cachep)
{
	return cachep->array[smp_processor_id()];
}

675
static inline kmem_cache_t *__find_general_cachep(size_t size, gfp_t gfpflags)
Linus Torvalds's avatar
Linus Torvalds committed
676
677
678
679
680
{
	struct cache_sizes *csizep = malloc_sizes;

#if DEBUG
	/* This happens if someone tries to call
681
682
683
	 * kmem_cache_create(), or __kmalloc(), before
	 * the generic caches are initialized.
	 */
684
	BUG_ON(malloc_sizes[INDEX_AC].cs_cachep == NULL);
Linus Torvalds's avatar
Linus Torvalds committed
685
686
687
688
689
#endif
	while (size > csizep->cs_size)
		csizep++;

	/*
690
	 * Really subtle: The last entry with cs->cs_size==ULONG_MAX
Linus Torvalds's avatar
Linus Torvalds committed
691
692
693
694
695
696
697
698
	 * has cs_{dma,}cachep==NULL. Thus no special case
	 * for large kmalloc calls required.
	 */
	if (unlikely(gfpflags & GFP_DMA))
		return csizep->cs_dmacachep;
	return csizep->cs_cachep;
}

699
kmem_cache_t *kmem_find_general_cachep(size_t size, gfp_t gfpflags)
700
701
702
703
704
{
	return __find_general_cachep(size, gfpflags);
}
EXPORT_SYMBOL(kmem_find_general_cachep);

705
static size_t slab_mgmt_size(size_t nr_objs, size_t align)
Linus Torvalds's avatar
Linus Torvalds committed
706
{
707
708
	return ALIGN(sizeof(struct slab)+nr_objs*sizeof(kmem_bufctl_t), align);
}
Linus Torvalds's avatar
Linus Torvalds committed
709

710
711
712
713
714
715
716
717
718
/* Calculate the number of objects and left-over bytes for a given
   buffer size. */
static void cache_estimate(unsigned long gfporder, size_t buffer_size,
			   size_t align, int flags, size_t *left_over,
			   unsigned int *num)
{
	int nr_objs;
	size_t mgmt_size;
	size_t slab_size = PAGE_SIZE << gfporder;
Linus Torvalds's avatar
Linus Torvalds committed
719

720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
	/*
	 * The slab management structure can be either off the slab or
	 * on it. For the latter case, the memory allocated for a
	 * slab is used for:
	 *
	 * - The struct slab
	 * - One kmem_bufctl_t for each object
	 * - Padding to respect alignment of @align
	 * - @buffer_size bytes for each object
	 *
	 * If the slab management structure is off the slab, then the
	 * alignment will already be calculated into the size. Because
	 * the slabs are all pages aligned, the objects will be at the
	 * correct alignment when allocated.
	 */
	if (flags & CFLGS_OFF_SLAB) {
		mgmt_size = 0;
		nr_objs = slab_size / buffer_size;

		if (nr_objs > SLAB_LIMIT)
			nr_objs = SLAB_LIMIT;
	} else {
		/*
		 * Ignore padding for the initial guess. The padding
		 * is at most @align-1 bytes, and @buffer_size is at
		 * least @align. In the worst case, this result will
		 * be one greater than the number of objects that fit
		 * into the memory allocation when taking the padding
		 * into account.
		 */
		nr_objs = (slab_size - sizeof(struct slab)) /
			  (buffer_size + sizeof(kmem_bufctl_t));

		/*
		 * This calculated number will be either the right
		 * amount, or one greater than what we want.
		 */
		if (slab_mgmt_size(nr_objs, align) + nr_objs*buffer_size
		       > slab_size)
			nr_objs--;

		if (nr_objs > SLAB_LIMIT)
			nr_objs = SLAB_LIMIT;

		mgmt_size = slab_mgmt_size(nr_objs, align);
	}
	*num = nr_objs;
	*left_over = slab_size - nr_objs*buffer_size - mgmt_size;
Linus Torvalds's avatar
Linus Torvalds committed
768
769
770
771
772
773
774
}

#define slab_error(cachep, msg) __slab_error(__FUNCTION__, cachep, msg)

static void __slab_error(const char *function, kmem_cache_t *cachep, char *msg)
{
	printk(KERN_ERR "slab error in %s(): cache `%s': %s\n",
775
	       function, cachep->name, msg);
Linus Torvalds's avatar
Linus Torvalds committed
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
	dump_stack();
}

/*
 * Initiate the reap timer running on the target CPU.  We run at around 1 to 2Hz
 * via the workqueue/eventd.
 * Add the CPU number into the expiration time to minimize the possibility of
 * the CPUs getting into lockstep and contending for the global cache chain
 * lock.
 */
static void __devinit start_cpu_timer(int cpu)
{
	struct work_struct *reap_work = &per_cpu(reap_work, cpu);

	/*
	 * When this gets called from do_initcalls via cpucache_init(),
	 * init_workqueues() has already run, so keventd will be setup
	 * at that time.
	 */
	if (keventd_up() && reap_work->func == NULL) {
		INIT_WORK(reap_work, cache_reap, NULL);
		schedule_delayed_work_on(cpu, reap_work, HZ + 3 * cpu);
	}
}

801
static struct array_cache *alloc_arraycache(int node, int entries,
802
					    int batchcount)
Linus Torvalds's avatar
Linus Torvalds committed
803
{
804
	int memsize = sizeof(void *) * entries + sizeof(struct array_cache);
Linus Torvalds's avatar
Linus Torvalds committed
805
806
	struct array_cache *nc = NULL;

807
	nc = kmalloc_node(memsize, GFP_KERNEL, node);
Linus Torvalds's avatar
Linus Torvalds committed
808
809
810
811
812
	if (nc) {
		nc->avail = 0;
		nc->limit = entries;
		nc->batchcount = batchcount;
		nc->touched = 0;
813
		spin_lock_init(&nc->lock);
Linus Torvalds's avatar
Linus Torvalds committed
814
815
816
817
	}
	return nc;
}

818
#ifdef CONFIG_NUMA
819
820
static void *__cache_alloc_node(kmem_cache_t *, gfp_t, int);

821
822
823
static inline struct array_cache **alloc_alien_cache(int node, int limit)
{
	struct array_cache **ac_ptr;
824
	int memsize = sizeof(void *) * MAX_NUMNODES;
825
826
827
828
829
830
831
832
833
834
835
836
837
	int i;

	if (limit > 1)
		limit = 12;
	ac_ptr = kmalloc_node(memsize, GFP_KERNEL, node);
	if (ac_ptr) {
		for_each_node(i) {
			if (i == node || !node_online(i)) {
				ac_ptr[i] = NULL;
				continue;
			}
			ac_ptr[i] = alloc_arraycache(node, limit, 0xbaadf00d);
			if (!ac_ptr[i]) {
838
				for (i--; i <= 0; i--)
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
					kfree(ac_ptr[i]);
				kfree(ac_ptr);
				return NULL;
			}
		}
	}
	return ac_ptr;
}

static inline void free_alien_cache(struct array_cache **ac_ptr)
{
	int i;

	if (!ac_ptr)
		return;

	for_each_node(i)
856
	    kfree(ac_ptr[i]);
857
858
859
860

	kfree(ac_ptr);
}

861
862
static inline void __drain_alien_cache(kmem_cache_t *cachep,
				       struct array_cache *ac, int node)
863
864
865
866
867
{
	struct kmem_list3 *rl3 = cachep->nodelists[node];

	if (ac->avail) {
		spin_lock(&rl3->list_lock);
868
		free_block(cachep, ac->entry, ac->avail, node);
869
870
871
872
873
874
875
		ac->avail = 0;
		spin_unlock(&rl3->list_lock);
	}
}

static void drain_alien_cache(kmem_cache_t *cachep, struct kmem_list3 *l3)
{
876
	int i = 0;
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
	struct array_cache *ac;
	unsigned long flags;

	for_each_online_node(i) {
		ac = l3->alien[i];
		if (ac) {
			spin_lock_irqsave(&ac->lock, flags);
			__drain_alien_cache(cachep, ac, i);
			spin_unlock_irqrestore(&ac->lock, flags);
		}
	}
}
#else
#define alloc_alien_cache(node, limit) do { } while (0)
#define free_alien_cache(ac_ptr) do { } while (0)
#define drain_alien_cache(cachep, l3) do { } while (0)
#endif

Linus Torvalds's avatar
Linus Torvalds committed
895
static int __devinit cpuup_callback(struct notifier_block *nfb,
896
				    unsigned long action, void *hcpu)
Linus Torvalds's avatar
Linus Torvalds committed
897
898
{
	long cpu = (long)hcpu;
899
	kmem_cache_t *cachep;
900
901
902
	struct kmem_list3 *l3 = NULL;
	int node = cpu_to_node(cpu);
	int memsize = sizeof(struct kmem_list3);
Linus Torvalds's avatar
Linus Torvalds committed
903
904
905

	switch (action) {
	case CPU_UP_PREPARE:
Ingo Molnar's avatar
Ingo Molnar committed
906
		mutex_lock(&cache_chain_mutex);
907
908
909
910
911
912
		/* we need to do this right in the beginning since
		 * alloc_arraycache's are going to use this list.
		 * kmalloc_node allows us to add the slab to the right
		 * kmem_list3 and not this cpu's kmem_list3
		 */

Linus Torvalds's avatar
Linus Torvalds committed
913
		list_for_each_entry(cachep, &cache_chain, next) {
914
915
916
917
918
919
			/* setup the size64 kmemlist for cpu before we can
			 * begin anything. Make sure some other cpu on this
			 * node has not already allocated this
			 */
			if (!cachep->nodelists[node]) {
				if (!(l3 = kmalloc_node(memsize,
920
							GFP_KERNEL, node)))
921
922
923
					goto bad;
				kmem_list3_init(l3);
				l3->next_reap = jiffies + REAPTIMEOUT_LIST3 +
924
				    ((unsigned long)cachep) % REAPTIMEOUT_LIST3;
925
926
927

				cachep->nodelists[node] = l3;
			}
Linus Torvalds's avatar
Linus Torvalds committed
928

929
930
			spin_lock_irq(&cachep->nodelists[node]->list_lock);
			cachep->nodelists[node]->free_limit =
931
932
			    (1 + nr_cpus_node(node)) *
			    cachep->batchcount + cachep->num;
933
934
935
936
			spin_unlock_irq(&cachep->nodelists[node]->list_lock);
		}

		/* Now we can go ahead with allocating the shared array's
937
		   & array cache's */
938
		list_for_each_entry(cachep, &cache_chain, next) {
939
940
			struct array_cache *nc;

941
			nc = alloc_arraycache(node, cachep->limit,
942
					      cachep->batchcount);
Linus Torvalds's avatar
Linus Torvalds committed
943
944
945
946
			if (!nc)
				goto bad;
			cachep->array[cpu] = nc;

947
948
949
950
			l3 = cachep->nodelists[node];
			BUG_ON(!l3);
			if (!l3->shared) {
				if (!(nc = alloc_arraycache(node,
951
952
953
954
							    cachep->shared *
							    cachep->batchcount,
							    0xbaadf00d)))
					goto bad;
955
956

				/* we are serialised from CPU_DEAD or
957
				   CPU_UP_CANCELLED by the cpucontrol lock */
958
959
				l3->shared = nc;
			}
Linus Torvalds's avatar
Linus Torvalds committed
960
		}
Ingo Molnar's avatar
Ingo Molnar committed
961
		mutex_unlock(&cache_chain_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
962
963
964
965
966
967
968
969
		break;
	case CPU_ONLINE:
		start_cpu_timer(cpu);
		break;
#ifdef CONFIG_HOTPLUG_CPU
	case CPU_DEAD:
		/* fall thru */
	case CPU_UP_CANCELED:
Ingo Molnar's avatar
Ingo Molnar committed
970
		mutex_lock(&cache_chain_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
971
972
973

		list_for_each_entry(cachep, &cache_chain, next) {
			struct array_cache *nc;
974
			cpumask_t mask;
Linus Torvalds's avatar
Linus Torvalds committed
975

976
			mask = node_to_cpumask(node);
Linus Torvalds's avatar
Linus Torvalds committed
977
978
979
980
			spin_lock_irq(&cachep->spinlock);
			/* cpu is dead; no one can alloc from it. */
			nc = cachep->array[cpu];
			cachep->array[cpu] = NULL;
981
982
983
984
985
986
987
988
989
990
			l3 = cachep->nodelists[node];

			if (!l3)
				goto unlock_cache;

			spin_lock(&l3->list_lock);

			/* Free limit for this kmem_list3 */
			l3->free_limit -= cachep->batchcount;
			if (nc)
991
				free_block(cachep, nc->entry, nc->avail, node);
992
993

			if (!cpus_empty(mask)) {
994
995
996
				spin_unlock(&l3->list_lock);
				goto unlock_cache;
			}
997
998
999

			if (l3->shared) {
				free_block(cachep, l3->shared->entry,
1000
					   l3->shared->avail, node);