slab.c 94.9 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
71
72
73
74
75
76
77
/*
 * 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
 *	The global cache-chain is protected by the semaphore 'cache_chain_sem'.
 *	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>
Linus Torvalds's avatar
Linus Torvalds committed
106
107
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

#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.
 */

201
typedef unsigned int kmem_bufctl_t;
Linus Torvalds's avatar
Linus Torvalds committed
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
#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 {
219
220
221
222
223
224
	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
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
};

/*
 * 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 {
244
245
246
	struct rcu_head head;
	kmem_cache_t *cachep;
	void *addr;
Linus Torvalds's avatar
Linus Torvalds committed
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
};

/*
 * 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;
266
267
268
269
270
271
272
	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
273
274
275
276
277
278
279
280
};

/* 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;
281
	void *entries[BOOT_CPUCACHE_ENTRIES];
Linus Torvalds's avatar
Linus Torvalds committed
282
283
284
};

/*
285
 * The slab lists for all objects.
Linus Torvalds's avatar
Linus Torvalds committed
286
287
 */
struct kmem_list3 {
288
289
290
291
292
293
294
295
296
297
	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
298
299
};

300
301
302
303
304
305
306
307
308
309
/*
 * 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)

/*
310
 * This function must be completely optimized away if
311
312
313
314
 * a constant is passed to it. Mostly the same as
 * what is in linux/slab.h except it returns an
 * index.
 */
315
static __always_inline int index_of(const size_t size)
316
317
318
319
320
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
		{
			extern void __bad_size(void);
			__bad_size();
		}
331
332
	} else
		BUG();
333
334
335
336
337
	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
338

339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
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
363
364
365
366
367
368

/*
 * kmem_cache_t
 *
 * manages a cache.
 */
369

370
struct kmem_cache {
Linus Torvalds's avatar
Linus Torvalds committed
371
/* 1) per-cpu data, touched during every alloc/free */
372
373
374
375
376
	struct array_cache *array[NR_CPUS];
	unsigned int batchcount;
	unsigned int limit;
	unsigned int shared;
	unsigned int objsize;
377
/* 2) touched by every alloc & free from the backend */
378
379
380
381
	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
382
383
384

/* 3) cache_grow/shrink */
	/* order of pgs per slab (2^n) */
385
	unsigned int gfporder;
Linus Torvalds's avatar
Linus Torvalds committed
386
387

	/* force GFP flags, e.g. GFP_DMA */
388
	gfp_t gfpflags;
Linus Torvalds's avatar
Linus Torvalds committed
389

390
391
392
393
394
395
	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
396
397

	/* constructor func */
398
	void (*ctor) (void *, kmem_cache_t *, unsigned long);
Linus Torvalds's avatar
Linus Torvalds committed
399
400

	/* de-constructor func */
401
	void (*dtor) (void *, kmem_cache_t *, unsigned long);
Linus Torvalds's avatar
Linus Torvalds committed
402
403

/* 4) cache creation/removal */
404
405
	const char *name;
	struct list_head next;
Linus Torvalds's avatar
Linus Torvalds committed
406
407
408

/* 5) statistics */
#if STATS
409
410
411
412
413
414
415
416
417
418
419
420
421
	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
422
423
#endif
#if DEBUG
424
425
	int dbghead;
	int reallen;
Linus Torvalds's avatar
Linus Torvalds committed
426
427
428
429
430
431
432
433
434
435
#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
436
 * OTOH the cpuarrays can contain lots of objects,
Linus Torvalds's avatar
Linus Torvalds committed
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
 * 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++)
453
#define	STATS_INC_NODEFREES(x)	((x)->node_frees++)
Linus Torvalds's avatar
Linus Torvalds committed
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
#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)
472
#define	STATS_INC_NODEFREES(x)	do { } while (0)
Linus Torvalds's avatar
Linus Torvalds committed
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
#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
 * 0 .. cachep->dbghead - BYTES_PER_WORD - 1: padding. This ensures that
 * 		the end of an object is aligned with the end of the real
 * 		allocation. Catches writes behind the end of the allocation.
 * cachep->dbghead - BYTES_PER_WORD .. cachep->dbghead - 1:
 * 		redzone word.
 * cachep->dbghead: The real object.
 * cachep->objsize - 2* BYTES_PER_WORD: redzone word [BYTES_PER_WORD long]
 * cachep->objsize - 1* BYTES_PER_WORD: last caller address [BYTES_PER_WORD long]
 */
static int obj_dbghead(kmem_cache_t *cachep)
{
	return cachep->dbghead;
}

static int obj_reallen(kmem_cache_t *cachep)
{
	return cachep->reallen;
}

static unsigned long *dbg_redzone1(kmem_cache_t *cachep, void *objp)
{
	BUG_ON(!(cachep->flags & SLAB_RED_ZONE));
	return (unsigned long*) (objp+obj_dbghead(cachep)-BYTES_PER_WORD);
}

static unsigned long *dbg_redzone2(kmem_cache_t *cachep, void *objp)
{
	BUG_ON(!(cachep->flags & SLAB_RED_ZONE));
	if (cachep->flags & SLAB_STORE_USER)
525
526
527
		return (unsigned long *)(objp + cachep->objsize -
					 2 * BYTES_PER_WORD);
	return (unsigned long *)(objp + cachep->objsize - BYTES_PER_WORD);
Linus Torvalds's avatar
Linus Torvalds committed
528
529
530
531
532
}

static void **dbg_userword(kmem_cache_t *cachep, void *objp)
{
	BUG_ON(!(cachep->flags & SLAB_STORE_USER));
533
	return (void **)(objp + cachep->objsize - BYTES_PER_WORD);
Linus Torvalds's avatar
Linus Torvalds committed
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
}

#else

#define obj_dbghead(x)			0
#define obj_reallen(cachep)		(cachep->objsize)
#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;

568
/* Functions for storing/retrieving the cachep and or slab from the
Linus Torvalds's avatar
Linus Torvalds committed
569
570
571
 * 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.
 */
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
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
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609

/* 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>
610
	{NULL,}
Linus Torvalds's avatar
Linus Torvalds committed
611
612
613
614
#undef CACHE
};

static struct arraycache_init initarray_cache __initdata =
615
    { {0, BOOT_CPUCACHE_ENTRIES, 1, 0} };
Linus Torvalds's avatar
Linus Torvalds committed
616
static struct arraycache_init initarray_generic =
617
    { {0, BOOT_CPUCACHE_ENTRIES, 1, 0} };
Linus Torvalds's avatar
Linus Torvalds committed
618
619
620

/* internal cache of cache description objs */
static kmem_cache_t cache_cache = {
621
622
623
624
625
626
627
	.batchcount = 1,
	.limit = BOOT_CPUCACHE_ENTRIES,
	.shared = 1,
	.objsize = sizeof(kmem_cache_t),
	.flags = SLAB_NO_REAP,
	.spinlock = SPIN_LOCK_UNLOCKED,
	.name = "kmem_cache",
Linus Torvalds's avatar
Linus Torvalds committed
628
#if DEBUG
629
	.reallen = sizeof(kmem_cache_t),
Linus Torvalds's avatar
Linus Torvalds committed
630
631
632
633
#endif
};

/* Guard access to the cache-chain. */
634
static struct semaphore cache_chain_sem;
Linus Torvalds's avatar
Linus Torvalds committed
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
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,
651
652
	PARTIAL_AC,
	PARTIAL_L3,
Linus Torvalds's avatar
Linus Torvalds committed
653
654
655
656
657
	FULL
} g_cpucache_up;

static DEFINE_PER_CPU(struct work_struct, reap_work);

658
659
660
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);
661
static int __node_shrink(kmem_cache_t *cachep, int node);
Linus Torvalds's avatar
Linus Torvalds committed
662
663
664
665
666
667

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

668
static inline kmem_cache_t *__find_general_cachep(size_t size, gfp_t gfpflags)
Linus Torvalds's avatar
Linus Torvalds committed
669
670
671
672
673
{
	struct cache_sizes *csizep = malloc_sizes;

#if DEBUG
	/* This happens if someone tries to call
674
675
676
	 * kmem_cache_create(), or __kmalloc(), before
	 * the generic caches are initialized.
	 */
677
	BUG_ON(malloc_sizes[INDEX_AC].cs_cachep == NULL);
Linus Torvalds's avatar
Linus Torvalds committed
678
679
680
681
682
#endif
	while (size > csizep->cs_size)
		csizep++;

	/*
683
	 * Really subtle: The last entry with cs->cs_size==ULONG_MAX
Linus Torvalds's avatar
Linus Torvalds committed
684
685
686
687
688
689
690
691
	 * 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;
}

692
kmem_cache_t *kmem_find_general_cachep(size_t size, gfp_t gfpflags)
693
694
695
696
697
{
	return __find_general_cachep(size, gfpflags);
}
EXPORT_SYMBOL(kmem_find_general_cachep);

Linus Torvalds's avatar
Linus Torvalds committed
698
699
/* Cal the num objs, wastage, and bytes left over for a given slab size. */
static void cache_estimate(unsigned long gfporder, size_t size, size_t align,
700
			   int flags, size_t *left_over, unsigned int *num)
Linus Torvalds's avatar
Linus Torvalds committed
701
702
{
	int i;
703
	size_t wastage = PAGE_SIZE << gfporder;
Linus Torvalds's avatar
Linus Torvalds committed
704
705
706
707
708
709
710
711
	size_t extra = 0;
	size_t base = 0;

	if (!(flags & CFLGS_OFF_SLAB)) {
		base = sizeof(struct slab);
		extra = sizeof(kmem_bufctl_t);
	}
	i = 0;
712
	while (i * size + ALIGN(base + i * extra, align) <= wastage)
Linus Torvalds's avatar
Linus Torvalds committed
713
714
715
716
717
718
719
720
		i++;
	if (i > 0)
		i--;

	if (i > SLAB_LIMIT)
		i = SLAB_LIMIT;

	*num = i;
721
722
	wastage -= i * size;
	wastage -= ALIGN(base + i * extra, align);
Linus Torvalds's avatar
Linus Torvalds committed
723
724
725
726
727
728
729
730
	*left_over = wastage;
}

#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",
731
	       function, cachep->name, msg);
Linus Torvalds's avatar
Linus Torvalds committed
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
	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);
	}
}

757
static struct array_cache *alloc_arraycache(int node, int entries,
758
					    int batchcount)
Linus Torvalds's avatar
Linus Torvalds committed
759
{
760
	int memsize = sizeof(void *) * entries + sizeof(struct array_cache);
Linus Torvalds's avatar
Linus Torvalds committed
761
762
	struct array_cache *nc = NULL;

763
	nc = kmalloc_node(memsize, GFP_KERNEL, node);
Linus Torvalds's avatar
Linus Torvalds committed
764
765
766
767
768
	if (nc) {
		nc->avail = 0;
		nc->limit = entries;
		nc->batchcount = batchcount;
		nc->touched = 0;
769
		spin_lock_init(&nc->lock);
Linus Torvalds's avatar
Linus Torvalds committed
770
771
772
773
	}
	return nc;
}

774
775
776
777
#ifdef CONFIG_NUMA
static inline struct array_cache **alloc_alien_cache(int node, int limit)
{
	struct array_cache **ac_ptr;
778
	int memsize = sizeof(void *) * MAX_NUMNODES;
779
780
781
782
783
784
785
786
787
788
789
790
791
	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]) {
792
				for (i--; i <= 0; i--)
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
					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)
810
	    kfree(ac_ptr[i]);
811
812
813
814

	kfree(ac_ptr);
}

815
816
static inline void __drain_alien_cache(kmem_cache_t *cachep,
				       struct array_cache *ac, int node)
817
818
819
820
821
{
	struct kmem_list3 *rl3 = cachep->nodelists[node];

	if (ac->avail) {
		spin_lock(&rl3->list_lock);
822
		free_block(cachep, ac->entry, ac->avail, node);
823
824
825
826
827
828
829
		ac->avail = 0;
		spin_unlock(&rl3->list_lock);
	}
}

static void drain_alien_cache(kmem_cache_t *cachep, struct kmem_list3 *l3)
{
830
	int i = 0;
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
	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
849
static int __devinit cpuup_callback(struct notifier_block *nfb,
850
				    unsigned long action, void *hcpu)
Linus Torvalds's avatar
Linus Torvalds committed
851
852
{
	long cpu = (long)hcpu;
853
	kmem_cache_t *cachep;
854
855
856
	struct kmem_list3 *l3 = NULL;
	int node = cpu_to_node(cpu);
	int memsize = sizeof(struct kmem_list3);
Linus Torvalds's avatar
Linus Torvalds committed
857
858
859
860

	switch (action) {
	case CPU_UP_PREPARE:
		down(&cache_chain_sem);
861
862
863
864
865
866
		/* 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
867
		list_for_each_entry(cachep, &cache_chain, next) {
868
869
870
871
872
873
			/* 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,
874
							GFP_KERNEL, node)))
875
876
877
					goto bad;
				kmem_list3_init(l3);
				l3->next_reap = jiffies + REAPTIMEOUT_LIST3 +
878
				    ((unsigned long)cachep) % REAPTIMEOUT_LIST3;
879
880
881

				cachep->nodelists[node] = l3;
			}
Linus Torvalds's avatar
Linus Torvalds committed
882

883
884
			spin_lock_irq(&cachep->nodelists[node]->list_lock);
			cachep->nodelists[node]->free_limit =
885
886
			    (1 + nr_cpus_node(node)) *
			    cachep->batchcount + cachep->num;
887
888
889
890
			spin_unlock_irq(&cachep->nodelists[node]->list_lock);
		}

		/* Now we can go ahead with allocating the shared array's
891
		   & array cache's */
892
		list_for_each_entry(cachep, &cache_chain, next) {
893
894
			struct array_cache *nc;

895
			nc = alloc_arraycache(node, cachep->limit,
896
					      cachep->batchcount);
Linus Torvalds's avatar
Linus Torvalds committed
897
898
899
900
			if (!nc)
				goto bad;
			cachep->array[cpu] = nc;

901
902
903
904
			l3 = cachep->nodelists[node];
			BUG_ON(!l3);
			if (!l3->shared) {
				if (!(nc = alloc_arraycache(node,
905
906
907
908
							    cachep->shared *
							    cachep->batchcount,
							    0xbaadf00d)))
					goto bad;
909
910

				/* we are serialised from CPU_DEAD or
911
				   CPU_UP_CANCELLED by the cpucontrol lock */
912
913
				l3->shared = nc;
			}
Linus Torvalds's avatar
Linus Torvalds committed
914
915
916
917
918
919
920
921
922
923
924
925
926
927
		}
		up(&cache_chain_sem);
		break;
	case CPU_ONLINE:
		start_cpu_timer(cpu);
		break;
#ifdef CONFIG_HOTPLUG_CPU
	case CPU_DEAD:
		/* fall thru */
	case CPU_UP_CANCELED:
		down(&cache_chain_sem);

		list_for_each_entry(cachep, &cache_chain, next) {
			struct array_cache *nc;
928
			cpumask_t mask;
Linus Torvalds's avatar
Linus Torvalds committed
929

930
			mask = node_to_cpumask(node);
Linus Torvalds's avatar
Linus Torvalds committed
931
932
933
934
			spin_lock_irq(&cachep->spinlock);
			/* cpu is dead; no one can alloc from it. */
			nc = cachep->array[cpu];
			cachep->array[cpu] = NULL;
935
936
937
938
939
940
941
942
943
944
			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)
945
				free_block(cachep, nc->entry, nc->avail, node);
946
947

			if (!cpus_empty(mask)) {
948
949
950
				spin_unlock(&l3->list_lock);
				goto unlock_cache;
			}
951
952
953

			if (l3->shared) {
				free_block(cachep, l3->shared->entry,
954
					   l3->shared->avail, node);
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
				kfree(l3->shared);
				l3->shared = NULL;
			}
			if (l3->alien) {
				drain_alien_cache(cachep, l3);
				free_alien_cache(l3->alien);
				l3->alien = NULL;
			}

			/* free slabs belonging to this node */
			if (__node_shrink(cachep, node)) {
				cachep->nodelists[node] = NULL;
				spin_unlock(&l3->list_lock);
				kfree(l3);
			} else {
				spin_unlock(&l3->list_lock);
			}
972
		      unlock_cache:
Linus Torvalds's avatar
Linus Torvalds committed
973
974
975
976
977
978
979
980
			spin_unlock_irq(&cachep->spinlock);
			kfree(nc);
		}
		up(&cache_chain_sem);
		break;
#endif
	}
	return NOTIFY_OK;
981
      bad:
Linus Torvalds's avatar
Linus Torvalds committed
982
983
984
985
986
987
	up(&cache_chain_sem);
	return NOTIFY_BAD;
}

static struct notifier_block cpucache_notifier = { &cpuup_callback, NULL, 0 };

988
989
990
/*
 * swap the static kmem_list3 with kmalloced memory
 */
991
static void init_list(kmem_cache_t *cachep, struct kmem_list3 *list, int nodeid)
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
{
	struct kmem_list3 *ptr;

	BUG_ON(cachep->nodelists[nodeid] != list);
	ptr = kmalloc_node(sizeof(struct kmem_list3), GFP_KERNEL, nodeid);
	BUG_ON(!ptr);

	local_irq_disable();
	memcpy(ptr, list, sizeof(struct kmem_list3));
	MAKE_ALL_LISTS(cachep, ptr, nodeid);
	cachep->nodelists[nodeid] = ptr;
	local_irq_enable();
}

Linus Torvalds's avatar
Linus Torvalds committed
1006
1007
1008
1009
1010
1011
1012
1013
/* Initialisation.
 * Called after the gfp() functions have been enabled, and before smp_init().
 */
void __init kmem_cache_init(void)
{
	size_t left_over;
	struct cache_sizes *sizes;
	struct cache_names *names;
1014
1015
1016
1017
1018
1019
1020
	int i;

	for (i = 0; i < NUM_INIT_LISTS; i++) {
		kmem_list3_init(&initkmem_list3[i]);
		if (i < MAX_NUMNODES)
			cache_cache.nodelists[i] = NULL;
	}
Linus Torvalds's avatar
Linus Torvalds committed
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033

	/*
	 * Fragmentation resistance on low memory - only use bigger
	 * page orders on machines with more than 32MB of memory.
	 */
	if (num_physpages > (32 << 20) >> PAGE_SHIFT)
		slab_break_gfp_order = BREAK_GFP_ORDER_HI;

	/* Bootstrap is tricky, because several objects are allocated
	 * from caches that do not exist yet:
	 * 1) initialize the cache_cache cache: it contains the kmem_cache_t
	 *    structures of all caches, except cache_cache itself: cache_cache
	 *    is statically allocated.
1034
1035
1036
	 *    Initially an __init data area is used for the head array and the
	 *    kmem_list3 structures, it's replaced with a kmalloc allocated
	 *    array at the end of the bootstrap.
Linus Torvalds's avatar
Linus Torvalds committed
1037
	 * 2) Create the first kmalloc cache.
1038
1039
1040
1041
	 *    The kmem_cache_t for the new cache is allocated normally.
	 *    An __init data area is used for the head array.
	 * 3) Create the remaining kmalloc caches, with minimally sized
	 *    head arrays.
Linus Torvalds's avatar
Linus Torvalds committed
1042
1043
	 * 4) Replace the __init data head arrays for cache_cache and the first
	 *    kmalloc cache with kmalloc allocated arrays.
1044
1045
1046
	 * 5) Replace the __init data for kmem_list3 for cache_cache and
	 *    the other cache's with kmalloc allocated memory.
	 * 6) Resize the head arrays of the kmalloc caches to their final sizes.
Linus Torvalds's avatar
Linus Torvalds committed
1047
1048
1049
1050
1051
1052
1053
1054
	 */

	/* 1) create the cache_cache */
	init_MUTEX(&cache_chain_sem);
	INIT_LIST_HEAD(&cache_chain);
	list_add(&cache_cache.next, &cache_chain);
	cache_cache.colour_off = cache_line_size();
	cache_cache.array[smp_processor_id()] = &initarray_cache.cache;
1055
	cache_cache.nodelists[numa_node_id()] = &initkmem_list3[CACHE_CACHE];
Linus Torvalds's avatar
Linus Torvalds committed
1056
1057
1058
1059

	cache_cache.objsize = ALIGN(cache_cache.objsize, cache_line_size());

	cache_estimate(0, cache_cache.objsize, cache_line_size(), 0,
1060
		       &left_over, &cache_cache.num);
Linus Torvalds's avatar
Linus Torvalds committed
1061
1062
1063
	if (!cache_cache.num)
		BUG();

1064
	cache_cache.colour = left_over / cache_cache.colour_off;
Linus Torvalds's avatar
Linus Torvalds committed
1065
	cache_cache.colour_next = 0;
1066
1067
	cache_cache.slab_size = ALIGN(cache_cache.num * sizeof(kmem_bufctl_t) +
				      sizeof(struct slab), cache_line_size());
Linus Torvalds's avatar
Linus Torvalds committed
1068
1069
1070
1071
1072

	/* 2+3) create the kmalloc caches */
	sizes = malloc_sizes;
	names = cache_names;

1073
1074
1075
1076
1077
1078
	/* Initialize the caches that provide memory for the array cache
	 * and the kmem_list3 structures first.
	 * Without this, further allocations will bug
	 */

	sizes[INDEX_AC].cs_cachep = kmem_cache_create(names[INDEX_AC].name,
1079
1080
1081
1082
						      sizes[INDEX_AC].cs_size,
						      ARCH_KMALLOC_MINALIGN,
						      (ARCH_KMALLOC_FLAGS |
						       SLAB_PANIC), NULL, NULL);
1083
1084
1085

	if (INDEX_AC != INDEX_L3)
		sizes[INDEX_L3].cs_cachep =
1086
1087
1088
1089
1090
		    kmem_cache_create(names[INDEX_L3].name,
				      sizes[INDEX_L3].cs_size,
				      ARCH_KMALLOC_MINALIGN,
				      (ARCH_KMALLOC_FLAGS | SLAB_PANIC), NULL,
				      NULL);
1091

Linus Torvalds's avatar
Linus Torvalds committed
1092
	while (sizes->cs_size != ULONG_MAX) {
1093
1094
		/*
		 * For performance, all the general caches are L1 aligned.
Linus Torvalds's avatar
Linus Torvalds committed
1095
1096
1097
		 * This should be particularly beneficial on SMP boxes, as it
		 * eliminates "false sharing".
		 * Note for systems short on memory removing the alignment will
1098
1099
		 * allow tighter packing of the smaller caches.
		 */
1100
		if (!sizes->cs_cachep)
1101
			sizes->cs_cachep = kmem_cache_create(names->name,
1102
1103
1104
1105
1106
							     sizes->cs_size,
							     ARCH_KMALLOC_MINALIGN,
							     (ARCH_KMALLOC_FLAGS
							      | SLAB_PANIC),
							     NULL, NULL);
Linus Torvalds's avatar
Linus Torvalds committed
1107
1108
1109

		/* Inc off-slab bufctl limit until the ceiling is hit. */
		if (!(OFF_SLAB(sizes->cs_cachep))) {
1110
			offslab_limit = sizes->cs_size - sizeof(struct slab);
Linus Torvalds's avatar
Linus Torvalds committed
1111
1112
1113
1114
			offslab_limit /= sizeof(kmem_bufctl_t);
		}

		sizes->cs_dmacachep = kmem_cache_create(names->name_dma,
1115
1116
1117
1118
1119
1120
							sizes->cs_size,
							ARCH_KMALLOC_MINALIGN,
							(ARCH_KMALLOC_FLAGS |
							 SLAB_CACHE_DMA |
							 SLAB_PANIC), NULL,
							NULL);
Linus Torvalds's avatar
Linus Torvalds committed
1121
1122
1123
1124
1125
1126

		sizes++;
		names++;
	}
	/* 4) Replace the bootstrap head arrays */
	{
1127
		void *ptr;
1128

Linus Torvalds's avatar
Linus Torvalds committed
1129
		ptr = kmalloc(sizeof(struct arraycache_init), GFP_KERNEL);
1130

Linus Torvalds's avatar
Linus Torvalds committed
1131
1132
		local_irq_disable();
		BUG_ON(ac_data(&cache_cache) != &initarray_cache.cache);
1133
		memcpy(ptr, ac_data(&cache_cache),
1134
		       sizeof(struct arraycache_init));
Linus Torvalds's avatar
Linus Torvalds committed
1135
1136
		cache_cache.array[smp_processor_id()] = ptr;
		local_irq_enable();
1137

Linus Torvalds's avatar
Linus Torvalds committed
1138
		ptr = kmalloc(sizeof(struct arraycache_init), GFP_KERNEL);
1139

Linus Torvalds's avatar
Linus Torvalds committed
1140
		local_irq_disable();
1141
		BUG_ON(ac_data(malloc_sizes[INDEX_AC].cs_cachep)
1142
		       != &initarray_generic.cache);
1143
		memcpy(ptr, ac_data(malloc_sizes[INDEX_AC].cs_cachep),
1144
		       sizeof(struct arraycache_init));
1145
		malloc_sizes[INDEX_AC].cs_cachep->array[smp_processor_id()] =
1146
		    ptr;
Linus Torvalds's avatar
Linus Torvalds committed
1147
1148
		local_irq_enable();
	}
1149
1150
1151
1152
1153
	/* 5) Replace the bootstrap kmem_list3's */
	{
		int node;
		/* Replace the static kmem_list3 structures for the boot cpu */
		init_list(&cache_cache, &initkmem_list3[CACHE_CACHE],
1154
			  numa_node_id());
1155
1156
1157

		for_each_online_node(node) {
			init_list(malloc_sizes[INDEX_AC].cs_cachep,
1158
				  &initkmem_list3[SIZE_AC + node], node);
1159
1160
1161

			if (INDEX_AC != INDEX_L3) {
				init_list(malloc_sizes[INDEX_L3].cs_cachep,
1162
1163
					  &initkmem_list3[SIZE_L3 + node],
					  node);
1164
1165
1166
			}
		}
	}
Linus Torvalds's avatar
Linus Torvalds committed
1167

1168
	/* 6) resize the head arrays to their final sizes */
Linus Torvalds's avatar
Linus Torvalds committed
1169
1170
1171
1172
	{
		kmem_cache_t *cachep;
		down(&cache_chain_sem);
		list_for_each_entry(cachep, &cache_chain, next)
1173
		    enable_cpucache(cachep);
Linus Torvalds's avatar
Linus Torvalds committed
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
		up(&cache_chain_sem);
	}

	/* Done! */
	g_cpucache_up = FULL;

	/* Register a cpu startup notifier callback
	 * that initializes ac_data for all new cpus
	 */
	register_cpu_notifier(&cpucache_notifier);

	/* The reap timers are started later, with a module init call:
	 * That part of the kernel is not yet operational.
	 */
}

static int __init cpucache_init(void)
{
	int cpu;

	/* 
	 * Register the timers that return unneeded
	 * pages to gfp.
	 */
1198
	for_each_online_cpu(cpu)
1199
	    start_cpu_timer(cpu);
Linus Torvalds's avatar
Linus Torvalds committed
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212

	return 0;
}

__initcall(cpucache_init);

/*
 * Interface to system's page allocator. No need to hold the cache-lock.
 *
 * If we requested dmaable memory, we will get it. Even if we
 * did not request dmaable memory, we might get it, but that
 * would be relatively rare and ignorable.
 */
1213
static void *kmem_getpages(kmem_cache_t *cachep, gfp_t flags, int nodeid)
Linus Torvalds's avatar
Linus Torvalds committed
1214
1215
1216
1217
1218
1219
{
	struct page *page;
	void *addr;
	int i;

	flags |= cachep->gfpflags;
1220
	page = alloc_pages_node(nodeid, flags, cachep->gfporder);
Linus Torvalds's avatar
Linus Torvalds committed
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
	if (!page)
		return NULL;
	addr = page_address(page);

	i = (1 << cachep->gfporder);
	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
		atomic_add(i, &slab_reclaim_pages);
	add_page_state(nr_slab, i);
	while (i--) {
		SetPageSlab(page);
		page++;
	}
	return addr;
}

/*
 * Interface to system's page release.
 */
static void kmem_freepages(kmem_cache_t *cachep, void *addr)
{
1241
	unsigned long i = (1 << cachep->gfporder);
Linus Torvalds's avatar
Linus Torvalds committed
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
	struct page *page = virt_to_page(addr);
	const unsigned long nr_freed = i;

	while (i--) {
		if (!TestClearPageSlab(page))
			BUG();
		page++;
	}
	sub_page_state(nr_slab, nr_freed);
	if (current->reclaim_state)
		current->reclaim_state->reclaimed_slab += nr_freed;
	free_pages((unsigned long)addr, cachep->gfporder);
1254
1255
	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
		atomic_sub(1 << cachep->gfporder, &slab_reclaim_pages);
Linus Torvalds's avatar
Linus Torvalds committed
1256
1257
1258
1259
}

static void kmem_rcu_free(struct rcu_head *head)
{
1260
	struct slab_rcu *slab_rcu = (struct slab_rcu *)head;
Linus Torvalds's avatar
Linus Torvalds committed
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
	kmem_cache_t *cachep = slab_rcu->cachep;

	kmem_freepages(cachep, slab_rcu->addr);
	if (OFF_SLAB(cachep))
		kmem_cache_free(cachep->slabp_cache, slab_rcu);
}

#if DEBUG

#ifdef CONFIG_DEBUG_PAGEALLOC
static void store_stackinfo(kmem_cache_t *cachep, unsigned long *addr,
1272
			    unsigned long caller)
Linus Torvalds's avatar
Linus Torvalds committed
1273
1274
1275
{
	int size = obj_reallen(cachep);

1276
	addr = (unsigned long *)&((char *)addr)[obj_dbghead(cachep)];
Linus Torvalds's avatar
Linus Torvalds committed
1277

1278
	if (size < 5 * sizeof(unsigned long))
Linus Torvalds's avatar
Linus Torvalds committed
1279
1280
		return;

1281
1282
1283
1284
	*addr++ = 0x12345678;
	*addr++ = caller;
	*addr++ = smp_processor_id();
	size -= 3 * sizeof(unsigned long);
Linus Torvalds's avatar
Linus Torvalds committed
1285
1286
1287
1288
1289
1290
1291
	{
		unsigned long *sptr = &caller;
		unsigned long svalue;

		while (!kstack_end(sptr)) {
			svalue = *sptr++;
			if (kernel_text_address(svalue)) {
1292
				*addr++ = svalue;
Linus Torvalds's avatar
Linus Torvalds committed
1293
1294
1295
1296
1297
1298
1299
				size -= sizeof(unsigned long);
				if (size <= sizeof(unsigned long))
					break;
			}
		}

	}
1300
	*addr++ = 0x87654321;
Linus Torvalds's avatar
Linus Torvalds committed
1301
1302
1303
1304
1305
1306
}
#endif

static void poison_obj(kmem_cache_t *cachep, void *addr, unsigned char val)
{
	int size = obj_reallen(cachep);
1307
	addr = &((char *)addr)[obj_dbghead(cachep)];
Linus Torvalds's avatar
Linus Torvalds committed
1308
1309

	memset(addr, val, size);
1310
	*(unsigned char *)(addr + size - 1) = POISON_END;
Linus Torvalds's avatar
Linus Torvalds committed
1311
1312
1313
1314
1315
1316
}

static void dump_line(char *data, int offset, int limit)
{
	int i;
	printk(KERN_ERR "%03x:", offset);
1317
1318
	for (i = 0; i < limit; i++) {
		printk(" %02x", (unsigned char)data[offset + i]);
Linus Torvalds's avatar
Linus Torvalds committed
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
	}
	printk("\n");
}
#endif

#if DEBUG

static void print_objinfo(kmem_cache_t *cachep, void *objp, int lines)
{
	int i, size;
	char *realobj;

	if (cachep->flags & SLAB_RED_ZONE) {
		printk(KERN_ERR "Redzone: 0x%lx/0x%lx.\n",
1333
1334
		       *dbg_redzone1(cachep, objp),
		       *dbg_redzone2(cachep, objp));
Linus Torvalds's avatar
Linus Torvalds committed
1335
1336
1337
1338
	}

	if (cachep->flags & SLAB_STORE_USER) {
		printk(KERN_ERR "Last user: [<%p>]",
1339
		       *dbg_userword(cachep, objp));
Linus Torvalds's avatar
Linus Torvalds committed
1340
		print_symbol("(%s)",
1341
			     (unsigned long)*dbg_userword(cachep, objp));
Linus Torvalds's avatar
Linus Torvalds committed
1342
1343
		printk("\n");
	}
1344
	realobj = (char *)objp + obj_dbghead(cachep);
Linus Torvalds's avatar
Linus Torvalds committed
1345
	size = obj_reallen(cachep);
1346
	for (i = 0; i < size && lines; i += 16, lines--) {
Linus Torvalds's avatar
Linus Torvalds committed
1347
1348
		int limit;
		limit = 16;
1349
1350
		if (i + limit > size)
			limit = size - i;
Linus Torvalds's avatar
Linus Torvalds committed
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
		dump_line(realobj, i, limit);
	}
}

static void check_poison_obj(kmem_cache_t *cachep, void *objp)
{
	char *realobj;
	int size, i;
	int lines = 0;

1361
	realobj = (char *)objp + obj_dbghead(cachep);
Linus Torvalds's avatar
Linus Torvalds committed
1362
1363
	size = obj_reallen(cachep);

1364
	for (i = 0; i < size; i++) {
Linus Torvalds's avatar
Linus Torvalds committed
1365
		char exp = POISON_FREE;
1366
		if (i == size - 1)
Linus Torvalds's avatar
Linus Torvalds committed
1367
1368
1369
1370
1371
1372
			exp = POISON_END;
		if (realobj[i] != exp) {
			int limit;
			/* Mismatch ! */
			/* Print header */
			if (lines == 0) {
1373
1374
1375
				printk(KERN_ERR
				       "Slab corruption: start=%p, len=%d\n",
				       realobj, size);
Linus Torvalds's avatar
Linus Torvalds committed
1376
1377
1378
				print_objinfo(cachep, objp, 0);
			}
			/* Hexdump the affected line */
1379
			i = (i / 16) * 16;
Linus Torvalds's avatar
Linus Torvalds committed
1380
			limit = 16;
1381
1382
			if (i + limit > size)
				limit = size - i;
Linus Torvalds's avatar
Linus Torvalds committed
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
			dump_line(realobj, i, limit);
			i += 16;
			lines++;
			/* Limit to 5 lines */
			if (lines > 5)
				break;
		}
	}
	if (lines != 0) {
		/* Print some data about the neighboring objects, if they
		 * exist:
		 */
1395
		struct slab *slabp = page_get_slab(virt_to_page(objp));
Linus Torvalds's avatar
Linus Torvalds committed
1396
1397
		int objnr;

1398
		objnr = (objp - slabp->s_mem) / cachep->objsize;
Linus Torvalds's avatar
Linus Torvalds committed
1399
		if (objnr) {
1400
1401
			objp = slabp->s_mem + (objnr - 1) * cachep->objsize;
			realobj = (char *)objp + obj_dbghead(cachep);
Linus Torvalds's avatar
Linus Torvalds committed
1402
			printk(KERN_ERR "Prev obj: start=%p, len=%d\n",
1403
			       realobj, size);
Linus Torvalds's avatar
Linus Torvalds committed
1404
1405
			print_objinfo(cachep, objp, 2);
		}
1406
1407
1408
		if (objnr + 1 < cachep->num) {
			objp = slabp->s_mem + (objnr + 1) * cachep->objsize;
			realobj = (char *)objp + obj_dbghead(cachep);
Linus Torvalds's avatar
Linus Torvalds committed
1409
			printk(KERN_ERR "Next obj: start=%p, len=%d\n",
1410
			       realobj, size);
Linus Torvalds's avatar
Linus Torvalds committed
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
			print_objinfo(cachep, objp, 2);
		}
	}
}
#endif

/* Destroy all the objs in a slab, and release the mem back to the system.
 * Before calling the slab must have been unlinked from the cache.
 * The cache-lock is not held/needed.
 */
1421
static void slab_destroy(kmem_cache_t *cachep, struct slab *slabp)
Linus Torvalds's avatar
Linus Torvalds committed
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
{
	void *addr = slabp->s_mem - slabp->colouroff;

#if DEBUG
	int i;
	for (i = 0; i < cachep->num; i++) {
		void *objp = slabp->s_mem + cachep->objsize * i;

		if (cachep->flags & SLAB_POISON) {
#ifdef CONFIG_DEBUG_PAGEALLOC
1432
1433
1434
1435
1436
			if ((cachep->objsize % PAGE_SIZE) == 0
			    && OFF_SLAB(cachep))
				kernel_map_pages(virt_to_page(objp),
						 cachep->objsize / PAGE_SIZE,
						 1);
Linus Torvalds's avatar
Linus Torvalds committed
1437
1438
1439
1440
1441
1442
1443
1444
1445
			else
				check_poison_obj(cachep, objp);
#else
			check_poison_obj(cachep, objp);
#endif
		}
		if (cachep->flags & SLAB_RED_ZONE) {
			if (*dbg_redzone1(cachep, objp) != RED_INACTIVE)
				slab_error(cachep, "start of a freed object "
1446
					   "was overwritten");
Linus Torvalds's avatar
Linus Torvalds committed
1447
1448
			if (*dbg_redzone2(cachep, objp) != RED_INACTIVE)
				slab_error(cachep, "end of a freed object "
1449
					   "was overwritten");
Linus Torvalds's avatar
Linus Torvalds committed
1450
1451
		}
		if (cachep->dtor && !(cachep->flags & SLAB_POISON))
1452
			(cachep->dtor) (objp + obj_dbghead(cachep), cachep, 0);
Linus Torvalds's avatar
Linus Torvalds committed
1453
1454
1455
1456
1457
	}
#else
	if (cachep->dtor) {
		int i;
		for (i = 0; i < cachep->num; i++) {
1458
1459
			void *objp = slabp->s_mem + cachep->objsize * i;
			(cachep->dtor) (objp, cachep, 0);
Linus Torvalds's avatar
Linus Torvalds committed
1460
1461
1462
1463
1464
1465
1466
		}
	}
#endif

	if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU)) {
		struct slab_rcu *slab_rcu;

1467
		slab_rcu = (struct slab_rcu *)slabp;
Linus Torvalds's avatar
Linus Torvalds committed
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
		slab_rcu->cachep = cachep;
		slab_rcu->addr = addr;
		call_rcu(&slab_rcu->head, kmem_rcu_free);
	} else {
		kmem_freepages(cachep, addr);
		if (OFF_SLAB(cachep))
			kmem_cache_free(cachep->slabp_cache, slabp);
	}
}

1478
1479
1480
1481
1482
1483
1484
/* For setting up all the kmem_list3s for cache whose objsize is same
   as size of kmem_list3. */
static inline void set_up_list3s(kmem_cache_t *cachep, int index)
{
	int node;

	for_each_online_node(node) {
1485
		cachep->nodelists[node] = &initkmem_list3[index + node];
1486
		cachep->nodelists[node]->next_reap = jiffies +
1487
1488
		    REAPTIMEOUT_LIST3 +
		    ((unsigned long)cachep) % REAPTIMEOUT_LIST3;
1489
1490
1491
	}
}

1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
/**
 * calculate_slab_order - calculate size (page order) of slabs and the number
 *                        of objects per slab.
 *
 * This could be made much more intelligent.  For now, try to avoid using
 * high order pages for slabs.  When the gfp() functions are more friendly
 * towards high-order requests, this should be changed.
 */
static inline size_t calculate_slab_order(kmem_cache_t *cachep, size_t size,
					  size_t align, gfp_t flags)
{
	size_t left_over = 0;

1505
	for (;; cachep->gfporder++) {
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
		unsigned int num;
		size_t remainder;

		if (cachep->gfporder > MAX_GFP_ORDER) {
			cachep->num = 0;
			break;
		}

		cache_estimate(cachep->gfporder, size, align, flags,
			       &remainder, &num);
		if (!num)
			continue;
		/* More than offslab_limit objects will cause problems */
		if (flags & CFLGS_OFF_SLAB && cachep->num > offslab_limit)
			break;

		cachep->num = num;
		left_over = remainder;

		/*
		 * Large number of objects is good, but very large slabs are
		 * currently bad for the gfp()s.
		 */
		if (cachep->gfporder >= slab_break_gfp_order)
			break;

		if ((left_over * 8) <= (PAGE_SIZE << cachep->gfporder))
			/* Acceptable internal fragmentation */
			break;
	}
	return left_over;
}

Linus Torvalds's avatar
Linus Torvalds committed
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
/**
 * kmem_cache_create - Create a cache.
 * @name: A string which is used in /proc/slabinfo to identify this cache.
 * @size: The size of objects to be created in this cache.
 * @align: The required alignment for the objects.
 * @flags: SLAB flags
 * @ctor: A constructor for the objects.
 * @dtor: A destructor for the objects.
 *
 * Returns a ptr to the cache on success, NULL on failure.
 * Cannot be called within a int, but can be interrupted.
 * The @ctor is run when new pages are allocated by the cache
 * and the @dtor is run before the pages are handed back.
 *
 * @name must be valid until the cache is destroyed. This implies that
 * the module calling this has to destroy the cache before getting 
 * unloaded.
 * 
 * The flags are
 *
 * %SLAB_POISON - Poison the slab with a known test pattern (a5a5a5a5)
 * to catch references to uninitialised memory.
 *
 * %SLAB_RED_ZONE - Insert `Red' zones around the allocated memory to check
 * for buffer overruns.
 *
 * %SLAB_NO_REAP - Don't automatically reap this cache when we're under
 * memory pressure.
 *
 * %SLAB_HWCACHE_ALIGN - Align the objects in this cache to a hardware
 * cacheline.  This can be beneficial if you're counting cycles as closely
 * as davem.
 */
kmem_cache_t *
kmem_cache_create (const char *name, size_t size, size_t align,
	unsigned long flags, void (*ctor)(void*, kmem_cache_t *, unsigned long),
	void (*dtor)(void*, kmem_cache_t *, unsigned long))
{
	size_t left_over, slab_size, ralign;
	kmem_cache_t *cachep = NULL;
1579
	struct list_head *p;
Linus Torvalds's avatar
Linus Torvalds committed
1580
1581
1582
1583
1584

	/*
	 * Sanity checks... these are all serious usage bugs.
	 */
	if ((!name) ||
1585
1586
1587
1588
1589
1590
1591
	    in_interrupt() ||
	    (size < BYTES_PER_WORD) ||
	    (size > (1 << MAX_OBJ_ORDER) * PAGE_SIZE) || (dtor && !ctor)) {
		printk(KERN_ERR "%s: Early error in slab %s\n",
		       __FUNCTION__, name);
		BUG();
	}
Linus Torvalds's avatar
Linus Torvalds committed
1592

1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
	down(&cache_chain_sem);

	list_for_each(p, &cache_chain) {
		kmem_cache_t *pc = list_entry(p, kmem_cache_t, next);
		mm_segment_t old_fs = get_fs();
		char tmp;
		int res;

		/*
		 * This happens when the module gets unloaded and doesn't
		 * destroy its slab cache and no-one else reuses the vmalloc
		 * area of the module.  Print a warning.
		 */
		set_fs(KERNEL_DS);
		res = __get_user(tmp, pc->name);
		set_fs(old_fs);
		if (res) {
			printk("SLAB: cache with size %d has lost its name\n",
1611
			       pc->objsize);
1612
1613
1614
			continue;
		}

1615
		if (!strcmp(pc->name, name)) {
1616
1617
1618
1619
1620
1621
			printk("kmem_cache_create: duplicate cache %s\n", name);
			dump_stack();
			goto oops;
		}
	}

Linus Torvalds's avatar
Linus Torvalds committed
1622
1623
1624
1625
1626
#if DEBUG
	WARN_ON(strchr(name, ' '));	/* It confuses parsers */
	if ((flags & SLAB_DEBUG_INITIAL) && !ctor) {
		/* No constructor, but inital state check requested */
		printk(KERN_ERR "%s: No con, but init state check "
1627
		       "requested - %s\n", __FUNCTION__, name);
Linus Torvalds's avatar
Linus Torvalds committed
1628
1629
1630
1631
1632
1633
1634
1635
1636
		flags &= ~SLAB_DEBUG_INITIAL;
	}
#if FORCED_DEBUG
	/*
	 * Enable redzoning and last user accounting, except for caches with
	 * large objects, if the increased size would increase the object size
	 * above the next power of two: caches with object sizes just above a
	 * power of two have a significant amount of internal fragmentation.
	 */
1637
1638
1639
	if ((size < 4096
	     || fls(size - 1) == fls(size - 1 + 3 * BYTES_PER_WORD)))
		flags |= SLAB_RED_ZONE | SLAB_STORE_USER;
Linus Torvalds's avatar