slub.c 105 KB
Newer Older
Christoph Lameter's avatar
Christoph Lameter committed
1
2
3
4
5
6
7
/*
 * SLUB: A slab allocator that limits cache line use instead of queuing
 * objects in per cpu and per node lists.
 *
 * The allocator synchronizes using per slab locks and only
 * uses a centralized lock to manage a pool of partial slabs.
 *
Christoph Lameter's avatar
Christoph Lameter committed
8
 * (C) 2007 SGI, Christoph Lameter
Christoph Lameter's avatar
Christoph Lameter committed
9
10
11
12
13
14
15
16
17
18
19
20
21
 */

#include <linux/mm.h>
#include <linux/module.h>
#include <linux/bit_spinlock.h>
#include <linux/interrupt.h>
#include <linux/bitops.h>
#include <linux/slab.h>
#include <linux/seq_file.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
#include <linux/mempolicy.h>
#include <linux/ctype.h>
22
#include <linux/debugobjects.h>
Christoph Lameter's avatar
Christoph Lameter committed
23
#include <linux/kallsyms.h>
24
#include <linux/memory.h>
Roman Zippel's avatar
Roman Zippel committed
25
#include <linux/math64.h>
Christoph Lameter's avatar
Christoph Lameter committed
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

/*
 * Lock order:
 *   1. slab_lock(page)
 *   2. slab->list_lock
 *
 *   The slab_lock protects operations on the object of a particular
 *   slab and its metadata in the page struct. If the slab lock
 *   has been taken then no allocations nor frees can be performed
 *   on the objects in the slab nor can the slab be added or removed
 *   from the partial or full lists since this would mean modifying
 *   the page_struct of the slab.
 *
 *   The list_lock protects the partial and full list on each node and
 *   the partial slab counter. If taken then no new slabs may be added or
 *   removed from the lists nor make the number of partial slabs be modified.
 *   (Note that the total number of slabs is an atomic value that may be
 *   modified without taking the list lock).
 *
 *   The list_lock is a centralized lock and thus we avoid taking it as
 *   much as possible. As long as SLUB does not have to handle partial
 *   slabs, operations can continue without any centralized lock. F.e.
 *   allocating a long series of objects that fill up slabs does not require
 *   the list lock.
 *
 *   The lock order is sometimes inverted when we are trying to get a slab
 *   off a list. We take the list_lock and then look for a page on the list
 *   to use. While we do that objects in the slabs may be freed. We can
 *   only operate on the slab if we have also taken the slab_lock. So we use
 *   a slab_trylock() on the slab. If trylock was successful then no frees
 *   can occur anymore and we can use the slab for allocations etc. If the
 *   slab_trylock() does not succeed then frees are in progress in the slab and
 *   we must stay away from it for a while since we may cause a bouncing
 *   cacheline if we try to acquire the lock. So go onto the next slab.
 *   If all pages are busy then we may allocate a new slab instead of reusing
 *   a partial slab. A new slab has noone operating on it and thus there is
 *   no danger of cacheline contention.
 *
 *   Interrupts are disabled during allocation and deallocation in order to
 *   make the slab allocator safe to use in the context of an irq. In addition
 *   interrupts are disabled to ensure that the processor does not change
 *   while handling per_cpu slabs, due to kernel preemption.
 *
 * SLUB assigns one slab for allocation to each processor.
 * Allocations only occur from these slabs called cpu slabs.
 *
Christoph Lameter's avatar
Christoph Lameter committed
72
73
 * Slabs with free elements are kept on a partial list and during regular
 * operations no list for full slabs is used. If an object in a full slab is
Christoph Lameter's avatar
Christoph Lameter committed
74
 * freed then the slab will show up again on the partial lists.
Christoph Lameter's avatar
Christoph Lameter committed
75
76
 * We track full slabs for debugging purposes though because otherwise we
 * cannot scan all objects.
Christoph Lameter's avatar
Christoph Lameter committed
77
78
79
80
81
82
83
 *
 * Slabs are freed when they become empty. Teardown and setup is
 * minimal so we rely on the page allocators per cpu caches for
 * fast frees and allocs.
 *
 * Overloading of page flags that are otherwise used for LRU management.
 *
84
85
86
87
88
89
90
91
92
93
94
95
 * PageActive 		The slab is frozen and exempt from list processing.
 * 			This means that the slab is dedicated to a purpose
 * 			such as satisfying allocations for a specific
 * 			processor. Objects may be freed in the slab while
 * 			it is frozen but slab_free will then skip the usual
 * 			list operations. It is up to the processor holding
 * 			the slab to integrate the slab into the slab lists
 * 			when the slab is no longer needed.
 *
 * 			One use of this flag is to mark slabs that are
 * 			used for allocations. Then such a slab becomes a cpu
 * 			slab. The cpu slab may be equipped with an additional
96
 * 			freelist that allows lockless access to
97
98
 * 			free objects in addition to the regular freelist
 * 			that requires the slab lock.
Christoph Lameter's avatar
Christoph Lameter committed
99
100
101
 *
 * PageError		Slab requires special handling due to debug
 * 			options set. This moves	slab handling out of
102
 * 			the fast path and disables lockless freelists.
Christoph Lameter's avatar
Christoph Lameter committed
103
104
 */

105
#ifdef CONFIG_SLUB_DEBUG
106
#define SLABDEBUG 1
107
108
109
110
#else
#define SLABDEBUG 0
#endif

Christoph Lameter's avatar
Christoph Lameter committed
111
112
113
114
115
116
117
118
119
120
121
/*
 * Issues still to be resolved:
 *
 * - Support PAGE_ALLOC_DEBUG. Should be easy to do.
 *
 * - Variable sizing of the per node arrays
 */

/* Enable to test recovery from slab corruption on boot */
#undef SLUB_RESILIENCY_TEST

122
123
124
125
/*
 * Mininum number of partial slabs. These will be left on the partial
 * lists even if they are empty. kmem_cache_shrink may reclaim them.
 */
126
#define MIN_PARTIAL 5
Christoph Lameter's avatar
Christoph Lameter committed
127

128
129
130
131
132
133
134
/*
 * Maximum number of desirable partial slabs.
 * The existence of more partial slabs makes kmem_cache_shrink
 * sort the partial list by the number of objects in the.
 */
#define MAX_PARTIAL 10

Christoph Lameter's avatar
Christoph Lameter committed
135
136
#define DEBUG_DEFAULT_FLAGS (SLAB_DEBUG_FREE | SLAB_RED_ZONE | \
				SLAB_POISON | SLAB_STORE_USER)
Christoph Lameter's avatar
Christoph Lameter committed
137

Christoph Lameter's avatar
Christoph Lameter committed
138
139
140
141
142
143
144
145
146
147
/*
 * Set of flags that will prevent slab merging
 */
#define SLUB_NEVER_MERGE (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER | \
		SLAB_TRACE | SLAB_DESTROY_BY_RCU)

#define SLUB_MERGE_SAME (SLAB_DEBUG_FREE | SLAB_RECLAIM_ACCOUNT | \
		SLAB_CACHE_DMA)

#ifndef ARCH_KMALLOC_MINALIGN
148
#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long)
Christoph Lameter's avatar
Christoph Lameter committed
149
150
151
#endif

#ifndef ARCH_SLAB_MINALIGN
152
#define ARCH_SLAB_MINALIGN __alignof__(unsigned long long)
Christoph Lameter's avatar
Christoph Lameter committed
153
154
155
#endif

/* Internal SLUB flags */
156
157
#define __OBJECT_POISON		0x80000000 /* Poison object */
#define __SYSFS_ADD_DEFERRED	0x40000000 /* Not yet visible via sysfs */
Christoph Lameter's avatar
Christoph Lameter committed
158
159
160
161
162
163
164
165
166
167

static int kmem_size = sizeof(struct kmem_cache);

#ifdef CONFIG_SMP
static struct notifier_block slab_notifier;
#endif

static enum {
	DOWN,		/* No slab functionality available */
	PARTIAL,	/* kmem_cache_open() works but kmalloc does not */
Christoph Lameter's avatar
Christoph Lameter committed
168
	UP,		/* Everything works but does not show up in sysfs */
Christoph Lameter's avatar
Christoph Lameter committed
169
170
171
172
173
	SYSFS		/* Sysfs up */
} slab_state = DOWN;

/* A list of all slab caches on the system */
static DECLARE_RWSEM(slub_lock);
Adrian Bunk's avatar
Adrian Bunk committed
174
static LIST_HEAD(slab_caches);
Christoph Lameter's avatar
Christoph Lameter committed
175

176
177
178
179
180
181
182
183
184
185
186
187
/*
 * Tracking user of a slab.
 */
struct track {
	void *addr;		/* Called from address */
	int cpu;		/* Was running on cpu */
	int pid;		/* Pid context */
	unsigned long when;	/* When did the operation occur */
};

enum track_item { TRACK_ALLOC, TRACK_FREE };

188
#ifdef CONFIG_SLUB_DEBUG
Christoph Lameter's avatar
Christoph Lameter committed
189
190
191
static int sysfs_slab_add(struct kmem_cache *);
static int sysfs_slab_alias(struct kmem_cache *, const char *);
static void sysfs_slab_remove(struct kmem_cache *);
192

Christoph Lameter's avatar
Christoph Lameter committed
193
#else
194
195
196
static inline int sysfs_slab_add(struct kmem_cache *s) { return 0; }
static inline int sysfs_slab_alias(struct kmem_cache *s, const char *p)
							{ return 0; }
197
198
199
200
static inline void sysfs_slab_remove(struct kmem_cache *s)
{
	kfree(s);
}
201

Christoph Lameter's avatar
Christoph Lameter committed
202
203
#endif

204
205
206
207
208
209
210
static inline void stat(struct kmem_cache_cpu *c, enum stat_item si)
{
#ifdef CONFIG_SLUB_STATS
	c->stat[si]++;
#endif
}

Christoph Lameter's avatar
Christoph Lameter committed
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
/********************************************************************
 * 			Core slab cache functions
 *******************************************************************/

int slab_is_available(void)
{
	return slab_state >= UP;
}

static inline struct kmem_cache_node *get_node(struct kmem_cache *s, int node)
{
#ifdef CONFIG_NUMA
	return s->node[node];
#else
	return &s->local_node;
#endif
}

229
230
static inline struct kmem_cache_cpu *get_cpu_slab(struct kmem_cache *s, int cpu)
{
231
232
233
234
235
#ifdef CONFIG_SMP
	return s->cpu_slab[cpu];
#else
	return &s->cpu_slab;
#endif
236
237
}

Christoph Lameter's avatar
Christoph Lameter committed
238
/* Verify that a pointer has an address that is valid within a slab page */
239
240
241
242
243
static inline int check_valid_pointer(struct kmem_cache *s,
				struct page *page, const void *object)
{
	void *base;

244
	if (!object)
245
246
		return 1;

247
	base = page_address(page);
248
	if (object < base || object >= base + page->objects * s->size ||
249
250
251
252
253
254
255
		(object - base) % s->size) {
		return 0;
	}

	return 1;
}

256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
/*
 * Slow version of get and set free pointer.
 *
 * This version requires touching the cache lines of kmem_cache which
 * we avoid to do in the fast alloc free paths. There we obtain the offset
 * from the page struct.
 */
static inline void *get_freepointer(struct kmem_cache *s, void *object)
{
	return *(void **)(object + s->offset);
}

static inline void set_freepointer(struct kmem_cache *s, void *object, void *fp)
{
	*(void **)(object + s->offset) = fp;
}

/* Loop over all objects in a slab */
274
275
#define for_each_object(__p, __s, __addr, __objects) \
	for (__p = (__addr); __p < (__addr) + (__objects) * (__s)->size;\
276
277
278
279
			__p += (__s)->size)

/* Scan freelist */
#define for_each_free_object(__p, __s, __free) \
280
	for (__p = (__free); __p; __p = get_freepointer((__s), __p))
281
282
283
284
285
286
287

/* Determine object index from a given position */
static inline int slab_index(void *p, struct kmem_cache *s, void *addr)
{
	return (p - addr) / s->size;
}

288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
static inline struct kmem_cache_order_objects oo_make(int order,
						unsigned long size)
{
	struct kmem_cache_order_objects x = {
		(order << 16) + (PAGE_SIZE << order) / size
	};

	return x;
}

static inline int oo_order(struct kmem_cache_order_objects x)
{
	return x.x >> 16;
}

static inline int oo_objects(struct kmem_cache_order_objects x)
{
	return x.x & ((1 << 16) - 1);
}

308
309
310
311
#ifdef CONFIG_SLUB_DEBUG
/*
 * Debug settings:
 */
312
313
314
#ifdef CONFIG_SLUB_DEBUG_ON
static int slub_debug = DEBUG_DEFAULT_FLAGS;
#else
315
static int slub_debug;
316
#endif
317
318
319

static char *slub_debug_slabs;

Christoph Lameter's avatar
Christoph Lameter committed
320
321
322
323
324
325
326
327
328
329
330
331
332
/*
 * Object debugging
 */
static void print_section(char *text, u8 *addr, unsigned int length)
{
	int i, offset;
	int newline = 1;
	char ascii[17];

	ascii[16] = 0;

	for (i = 0; i < length; i++) {
		if (newline) {
333
			printk(KERN_ERR "%8s 0x%p: ", text, addr + i);
Christoph Lameter's avatar
Christoph Lameter committed
334
335
			newline = 0;
		}
336
		printk(KERN_CONT " %02x", addr[i]);
Christoph Lameter's avatar
Christoph Lameter committed
337
338
339
		offset = i % 16;
		ascii[offset] = isgraph(addr[i]) ? addr[i] : '.';
		if (offset == 15) {
340
			printk(KERN_CONT " %s\n", ascii);
Christoph Lameter's avatar
Christoph Lameter committed
341
342
343
344
345
346
			newline = 1;
		}
	}
	if (!newline) {
		i %= 16;
		while (i < 16) {
347
			printk(KERN_CONT "   ");
Christoph Lameter's avatar
Christoph Lameter committed
348
349
350
			ascii[i] = ' ';
			i++;
		}
351
		printk(KERN_CONT " %s\n", ascii);
Christoph Lameter's avatar
Christoph Lameter committed
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
	}
}

static struct track *get_track(struct kmem_cache *s, void *object,
	enum track_item alloc)
{
	struct track *p;

	if (s->offset)
		p = object + s->offset + sizeof(void *);
	else
		p = object + s->inuse;

	return p + alloc;
}

static void set_track(struct kmem_cache *s, void *object,
				enum track_item alloc, void *addr)
{
	struct track *p;

	if (s->offset)
		p = object + s->offset + sizeof(void *);
	else
		p = object + s->inuse;

	p += alloc;
	if (addr) {
		p->addr = addr;
		p->cpu = smp_processor_id();
382
		p->pid = current->pid;
Christoph Lameter's avatar
Christoph Lameter committed
383
384
385
386
387
388
389
		p->when = jiffies;
	} else
		memset(p, 0, sizeof(struct track));
}

static void init_tracking(struct kmem_cache *s, void *object)
{
390
391
392
393
394
	if (!(s->flags & SLAB_STORE_USER))
		return;

	set_track(s, object, TRACK_FREE, NULL);
	set_track(s, object, TRACK_ALLOC, NULL);
Christoph Lameter's avatar
Christoph Lameter committed
395
396
397
398
399
400
401
}

static void print_track(const char *s, struct track *t)
{
	if (!t->addr)
		return;

402
403
	printk(KERN_ERR "INFO: %s in %pS age=%lu cpu=%u pid=%d\n",
		s, t->addr, jiffies - t->when, t->cpu, t->pid);
404
405
406
407
408
409
410
411
412
413
414
415
416
}

static void print_tracking(struct kmem_cache *s, void *object)
{
	if (!(s->flags & SLAB_STORE_USER))
		return;

	print_track("Allocated", get_track(s, object, TRACK_ALLOC));
	print_track("Freed", get_track(s, object, TRACK_FREE));
}

static void print_page_info(struct page *page)
{
417
418
	printk(KERN_ERR "INFO: Slab 0x%p objects=%u used=%u fp=0x%p flags=0x%04lx\n",
		page, page->objects, page->inuse, page->freelist, page->flags);
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434

}

static void slab_bug(struct kmem_cache *s, char *fmt, ...)
{
	va_list args;
	char buf[100];

	va_start(args, fmt);
	vsnprintf(buf, sizeof(buf), fmt, args);
	va_end(args);
	printk(KERN_ERR "========================================"
			"=====================================\n");
	printk(KERN_ERR "BUG %s: %s\n", s->name, buf);
	printk(KERN_ERR "----------------------------------------"
			"-------------------------------------\n\n");
Christoph Lameter's avatar
Christoph Lameter committed
435
436
}

437
438
439
440
441
442
443
444
445
446
447
448
static void slab_fix(struct kmem_cache *s, char *fmt, ...)
{
	va_list args;
	char buf[100];

	va_start(args, fmt);
	vsnprintf(buf, sizeof(buf), fmt, args);
	va_end(args);
	printk(KERN_ERR "FIX %s: %s\n", s->name, buf);
}

static void print_trailer(struct kmem_cache *s, struct page *page, u8 *p)
Christoph Lameter's avatar
Christoph Lameter committed
449
450
{
	unsigned int off;	/* Offset of last byte */
451
	u8 *addr = page_address(page);
452
453
454
455
456
457
458
459
460
461
462

	print_tracking(s, p);

	print_page_info(page);

	printk(KERN_ERR "INFO: Object 0x%p @offset=%tu fp=0x%p\n\n",
			p, p - addr, get_freepointer(s, p));

	if (p > addr + 16)
		print_section("Bytes b4", p - 16, 16);

463
	print_section("Object", p, min_t(unsigned long, s->objsize, PAGE_SIZE));
Christoph Lameter's avatar
Christoph Lameter committed
464
465
466
467
468
469
470
471
472
473

	if (s->flags & SLAB_RED_ZONE)
		print_section("Redzone", p + s->objsize,
			s->inuse - s->objsize);

	if (s->offset)
		off = s->offset + sizeof(void *);
	else
		off = s->inuse;

474
	if (s->flags & SLAB_STORE_USER)
Christoph Lameter's avatar
Christoph Lameter committed
475
476
477
478
		off += 2 * sizeof(struct track);

	if (off != s->size)
		/* Beginning of the filler is the free pointer */
479
480
481
		print_section("Padding", p + off, s->size - off);

	dump_stack();
Christoph Lameter's avatar
Christoph Lameter committed
482
483
484
485
486
}

static void object_err(struct kmem_cache *s, struct page *page,
			u8 *object, char *reason)
{
487
	slab_bug(s, "%s", reason);
488
	print_trailer(s, page, object);
Christoph Lameter's avatar
Christoph Lameter committed
489
490
}

491
static void slab_err(struct kmem_cache *s, struct page *page, char *fmt, ...)
Christoph Lameter's avatar
Christoph Lameter committed
492
493
494
495
{
	va_list args;
	char buf[100];

496
497
	va_start(args, fmt);
	vsnprintf(buf, sizeof(buf), fmt, args);
Christoph Lameter's avatar
Christoph Lameter committed
498
	va_end(args);
499
	slab_bug(s, "%s", buf);
500
	print_page_info(page);
Christoph Lameter's avatar
Christoph Lameter committed
501
502
503
504
505
506
507
508
509
	dump_stack();
}

static void init_object(struct kmem_cache *s, void *object, int active)
{
	u8 *p = object;

	if (s->flags & __OBJECT_POISON) {
		memset(p, POISON_FREE, s->objsize - 1);
510
		p[s->objsize - 1] = POISON_END;
Christoph Lameter's avatar
Christoph Lameter committed
511
512
513
514
515
516
517
518
	}

	if (s->flags & SLAB_RED_ZONE)
		memset(p + s->objsize,
			active ? SLUB_RED_ACTIVE : SLUB_RED_INACTIVE,
			s->inuse - s->objsize);
}

519
static u8 *check_bytes(u8 *start, unsigned int value, unsigned int bytes)
Christoph Lameter's avatar
Christoph Lameter committed
520
521
522
{
	while (bytes) {
		if (*start != (u8)value)
523
			return start;
Christoph Lameter's avatar
Christoph Lameter committed
524
525
526
		start++;
		bytes--;
	}
527
528
529
530
531
532
533
534
535
536
537
538
	return NULL;
}

static void restore_bytes(struct kmem_cache *s, char *message, u8 data,
						void *from, void *to)
{
	slab_fix(s, "Restoring 0x%p-0x%p=0x%x\n", from, to - 1, data);
	memset(from, data, to - from);
}

static int check_bytes_and_report(struct kmem_cache *s, struct page *page,
			u8 *object, char *what,
539
			u8 *start, unsigned int value, unsigned int bytes)
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
{
	u8 *fault;
	u8 *end;

	fault = check_bytes(start, value, bytes);
	if (!fault)
		return 1;

	end = start + bytes;
	while (end > fault && end[-1] == value)
		end--;

	slab_bug(s, "%s overwritten", what);
	printk(KERN_ERR "INFO: 0x%p-0x%p. First byte 0x%x instead of 0x%x\n",
					fault, end - 1, fault[0], value);
	print_trailer(s, page, object);

	restore_bytes(s, what, value, fault, end);
	return 0;
Christoph Lameter's avatar
Christoph Lameter committed
559
560
561
562
563
564
565
566
567
}

/*
 * Object layout:
 *
 * object address
 * 	Bytes of the object to be managed.
 * 	If the freepointer may overlay the object then the free
 * 	pointer is the first word of the object.
Christoph Lameter's avatar
Christoph Lameter committed
568
 *
Christoph Lameter's avatar
Christoph Lameter committed
569
570
571
572
573
 * 	Poisoning uses 0x6b (POISON_FREE) and the last byte is
 * 	0xa5 (POISON_END)
 *
 * object + s->objsize
 * 	Padding to reach word boundary. This is also used for Redzoning.
Christoph Lameter's avatar
Christoph Lameter committed
574
575
576
 * 	Padding is extended by another word if Redzoning is enabled and
 * 	objsize == inuse.
 *
Christoph Lameter's avatar
Christoph Lameter committed
577
578
579
580
 * 	We fill with 0xbb (RED_INACTIVE) for inactive objects and with
 * 	0xcc (RED_ACTIVE) for objects in use.
 *
 * object + s->inuse
Christoph Lameter's avatar
Christoph Lameter committed
581
582
 * 	Meta data starts here.
 *
Christoph Lameter's avatar
Christoph Lameter committed
583
584
 * 	A. Free pointer (if we cannot overwrite object on free)
 * 	B. Tracking data for SLAB_STORE_USER
Christoph Lameter's avatar
Christoph Lameter committed
585
 * 	C. Padding to reach required alignment boundary or at mininum
Christoph Lameter's avatar
Christoph Lameter committed
586
 * 		one word if debugging is on to be able to detect writes
Christoph Lameter's avatar
Christoph Lameter committed
587
588
589
 * 		before the word boundary.
 *
 *	Padding is done using 0x5a (POISON_INUSE)
Christoph Lameter's avatar
Christoph Lameter committed
590
591
 *
 * object + s->size
Christoph Lameter's avatar
Christoph Lameter committed
592
 * 	Nothing is used beyond s->size.
Christoph Lameter's avatar
Christoph Lameter committed
593
 *
Christoph Lameter's avatar
Christoph Lameter committed
594
595
 * If slabcaches are merged then the objsize and inuse boundaries are mostly
 * ignored. And therefore no slab options that rely on these boundaries
Christoph Lameter's avatar
Christoph Lameter committed
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
 * may be used with merged slabcaches.
 */

static int check_pad_bytes(struct kmem_cache *s, struct page *page, u8 *p)
{
	unsigned long off = s->inuse;	/* The end of info */

	if (s->offset)
		/* Freepointer is placed after the object. */
		off += sizeof(void *);

	if (s->flags & SLAB_STORE_USER)
		/* We also have user information there */
		off += 2 * sizeof(struct track);

	if (s->size == off)
		return 1;

614
615
	return check_bytes_and_report(s, page, p, "Object padding",
				p + off, POISON_INUSE, s->size - off);
Christoph Lameter's avatar
Christoph Lameter committed
616
617
}

618
/* Check the pad bytes at the end of a slab page */
Christoph Lameter's avatar
Christoph Lameter committed
619
620
static int slab_pad_check(struct kmem_cache *s, struct page *page)
{
621
622
623
624
625
	u8 *start;
	u8 *fault;
	u8 *end;
	int length;
	int remainder;
Christoph Lameter's avatar
Christoph Lameter committed
626
627
628
629

	if (!(s->flags & SLAB_POISON))
		return 1;

630
	start = page_address(page);
631
	length = (PAGE_SIZE << compound_order(page));
632
633
	end = start + length;
	remainder = length % s->size;
Christoph Lameter's avatar
Christoph Lameter committed
634
635
636
	if (!remainder)
		return 1;

637
	fault = check_bytes(end - remainder, POISON_INUSE, remainder);
638
639
640
641
642
643
	if (!fault)
		return 1;
	while (end > fault && end[-1] == POISON_INUSE)
		end--;

	slab_err(s, page, "Padding overwritten. 0x%p-0x%p", fault, end - 1);
644
	print_section("Padding", end - remainder, remainder);
645
646
647

	restore_bytes(s, "slab padding", POISON_INUSE, start, end);
	return 0;
Christoph Lameter's avatar
Christoph Lameter committed
648
649
650
651
652
653
654
655
656
657
658
659
}

static int check_object(struct kmem_cache *s, struct page *page,
					void *object, int active)
{
	u8 *p = object;
	u8 *endobject = object + s->objsize;

	if (s->flags & SLAB_RED_ZONE) {
		unsigned int red =
			active ? SLUB_RED_ACTIVE : SLUB_RED_INACTIVE;

660
661
		if (!check_bytes_and_report(s, page, object, "Redzone",
			endobject, red, s->inuse - s->objsize))
Christoph Lameter's avatar
Christoph Lameter committed
662
663
			return 0;
	} else {
Ingo Molnar's avatar
Ingo Molnar committed
664
665
666
667
		if ((s->flags & SLAB_POISON) && s->objsize < s->inuse) {
			check_bytes_and_report(s, page, p, "Alignment padding",
				endobject, POISON_INUSE, s->inuse - s->objsize);
		}
Christoph Lameter's avatar
Christoph Lameter committed
668
669
670
671
	}

	if (s->flags & SLAB_POISON) {
		if (!active && (s->flags & __OBJECT_POISON) &&
672
673
674
			(!check_bytes_and_report(s, page, p, "Poison", p,
					POISON_FREE, s->objsize - 1) ||
			 !check_bytes_and_report(s, page, p, "Poison",
675
				p + s->objsize - 1, POISON_END, 1)))
Christoph Lameter's avatar
Christoph Lameter committed
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
			return 0;
		/*
		 * check_pad_bytes cleans up on its own.
		 */
		check_pad_bytes(s, page, p);
	}

	if (!s->offset && active)
		/*
		 * Object and freepointer overlap. Cannot check
		 * freepointer while object is allocated.
		 */
		return 1;

	/* Check free pointer validity */
	if (!check_valid_pointer(s, page, get_freepointer(s, p))) {
		object_err(s, page, p, "Freepointer corrupt");
		/*
		 * No choice but to zap it and thus loose the remainder
		 * of the free objects in this slab. May cause
Christoph Lameter's avatar
Christoph Lameter committed
696
		 * another error because the object count is now wrong.
Christoph Lameter's avatar
Christoph Lameter committed
697
		 */
698
		set_freepointer(s, p, NULL);
Christoph Lameter's avatar
Christoph Lameter committed
699
700
701
702
703
704
705
		return 0;
	}
	return 1;
}

static int check_slab(struct kmem_cache *s, struct page *page)
{
706
707
	int maxobj;

Christoph Lameter's avatar
Christoph Lameter committed
708
709
710
	VM_BUG_ON(!irqs_disabled());

	if (!PageSlab(page)) {
711
		slab_err(s, page, "Not a valid slab page");
Christoph Lameter's avatar
Christoph Lameter committed
712
713
		return 0;
	}
714
715
716
717
718
719
720
721

	maxobj = (PAGE_SIZE << compound_order(page)) / s->size;
	if (page->objects > maxobj) {
		slab_err(s, page, "objects %u > max %u",
			s->name, page->objects, maxobj);
		return 0;
	}
	if (page->inuse > page->objects) {
722
		slab_err(s, page, "inuse %u > max %u",
723
			s->name, page->inuse, page->objects);
Christoph Lameter's avatar
Christoph Lameter committed
724
725
726
727
728
729
730
731
		return 0;
	}
	/* Slab_pad_check fixes things up after itself */
	slab_pad_check(s, page);
	return 1;
}

/*
Christoph Lameter's avatar
Christoph Lameter committed
732
733
 * Determine if a certain object on a page is on the freelist. Must hold the
 * slab lock to guarantee that the chains are in a consistent state.
Christoph Lameter's avatar
Christoph Lameter committed
734
735
736
737
738
739
 */
static int on_freelist(struct kmem_cache *s, struct page *page, void *search)
{
	int nr = 0;
	void *fp = page->freelist;
	void *object = NULL;
740
	unsigned long max_objects;
Christoph Lameter's avatar
Christoph Lameter committed
741

742
	while (fp && nr <= page->objects) {
Christoph Lameter's avatar
Christoph Lameter committed
743
744
745
746
747
748
		if (fp == search)
			return 1;
		if (!check_valid_pointer(s, page, fp)) {
			if (object) {
				object_err(s, page, object,
					"Freechain corrupt");
749
				set_freepointer(s, object, NULL);
Christoph Lameter's avatar
Christoph Lameter committed
750
751
				break;
			} else {
752
				slab_err(s, page, "Freepointer corrupt");
753
				page->freelist = NULL;
754
				page->inuse = page->objects;
755
				slab_fix(s, "Freelist cleared");
Christoph Lameter's avatar
Christoph Lameter committed
756
757
758
759
760
761
762
763
764
				return 0;
			}
			break;
		}
		object = fp;
		fp = get_freepointer(s, object);
		nr++;
	}

765
766
767
768
769
770
771
772
773
774
	max_objects = (PAGE_SIZE << compound_order(page)) / s->size;
	if (max_objects > 65535)
		max_objects = 65535;

	if (page->objects != max_objects) {
		slab_err(s, page, "Wrong number of objects. Found %d but "
			"should be %d", page->objects, max_objects);
		page->objects = max_objects;
		slab_fix(s, "Number of objects adjusted.");
	}
775
	if (page->inuse != page->objects - nr) {
776
		slab_err(s, page, "Wrong object count. Counter is %d but "
777
778
			"counted were %d", page->inuse, page->objects - nr);
		page->inuse = page->objects - nr;
779
		slab_fix(s, "Object count adjusted.");
Christoph Lameter's avatar
Christoph Lameter committed
780
781
782
783
	}
	return search == NULL;
}

784
785
static void trace(struct kmem_cache *s, struct page *page, void *object,
								int alloc)
Christoph Lameter's avatar
Christoph Lameter committed
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
{
	if (s->flags & SLAB_TRACE) {
		printk(KERN_INFO "TRACE %s %s 0x%p inuse=%d fp=0x%p\n",
			s->name,
			alloc ? "alloc" : "free",
			object, page->inuse,
			page->freelist);

		if (!alloc)
			print_section("Object", (void *)object, s->objsize);

		dump_stack();
	}
}

801
/*
Christoph Lameter's avatar
Christoph Lameter committed
802
 * Tracking of fully allocated slabs for debugging purposes.
803
 */
Christoph Lameter's avatar
Christoph Lameter committed
804
static void add_full(struct kmem_cache_node *n, struct page *page)
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
{
	spin_lock(&n->list_lock);
	list_add(&page->lru, &n->full);
	spin_unlock(&n->list_lock);
}

static void remove_full(struct kmem_cache *s, struct page *page)
{
	struct kmem_cache_node *n;

	if (!(s->flags & SLAB_STORE_USER))
		return;

	n = get_node(s, page_to_nid(page));

	spin_lock(&n->list_lock);
	list_del(&page->lru);
	spin_unlock(&n->list_lock);
}

825
826
827
828
829
830
831
832
/* Tracking of the number of slabs for debugging purposes */
static inline unsigned long slabs_node(struct kmem_cache *s, int node)
{
	struct kmem_cache_node *n = get_node(s, node);

	return atomic_long_read(&n->nr_slabs);
}

833
static inline void inc_slabs_node(struct kmem_cache *s, int node, int objects)
834
835
836
837
838
839
840
841
842
{
	struct kmem_cache_node *n = get_node(s, node);

	/*
	 * May be called early in order to allocate a slab for the
	 * kmem_cache_node structure. Solve the chicken-egg
	 * dilemma by deferring the increment of the count during
	 * bootstrap (see early_kmem_cache_node_alloc).
	 */
843
	if (!NUMA_BUILD || n) {
844
		atomic_long_inc(&n->nr_slabs);
845
846
		atomic_long_add(objects, &n->total_objects);
	}
847
}
848
static inline void dec_slabs_node(struct kmem_cache *s, int node, int objects)
849
850
851
852
{
	struct kmem_cache_node *n = get_node(s, node);

	atomic_long_dec(&n->nr_slabs);
853
	atomic_long_sub(objects, &n->total_objects);
854
855
856
}

/* Object debug checks for alloc/free paths */
Christoph Lameter's avatar
Christoph Lameter committed
857
858
859
860
861
862
863
864
865
866
867
868
static void setup_object_debug(struct kmem_cache *s, struct page *page,
								void *object)
{
	if (!(s->flags & (SLAB_STORE_USER|SLAB_RED_ZONE|__OBJECT_POISON)))
		return;

	init_object(s, object, 0);
	init_tracking(s, object);
}

static int alloc_debug_processing(struct kmem_cache *s, struct page *page,
						void *object, void *addr)
Christoph Lameter's avatar
Christoph Lameter committed
869
870
871
872
{
	if (!check_slab(s, page))
		goto bad;

873
	if (!on_freelist(s, page, object)) {
874
		object_err(s, page, object, "Object already allocated");
875
		goto bad;
Christoph Lameter's avatar
Christoph Lameter committed
876
877
878
879
	}

	if (!check_valid_pointer(s, page, object)) {
		object_err(s, page, object, "Freelist Pointer check fails");
880
		goto bad;
Christoph Lameter's avatar
Christoph Lameter committed
881
882
	}

883
	if (!check_object(s, page, object, 0))
Christoph Lameter's avatar
Christoph Lameter committed
884
885
		goto bad;

Christoph Lameter's avatar
Christoph Lameter committed
886
887
888
889
890
	/* Success perform special debug activities for allocs */
	if (s->flags & SLAB_STORE_USER)
		set_track(s, object, TRACK_ALLOC, addr);
	trace(s, page, object, 1);
	init_object(s, object, 1);
Christoph Lameter's avatar
Christoph Lameter committed
891
	return 1;
Christoph Lameter's avatar
Christoph Lameter committed
892

Christoph Lameter's avatar
Christoph Lameter committed
893
894
895
896
897
bad:
	if (PageSlab(page)) {
		/*
		 * If this is a slab page then lets do the best we can
		 * to avoid issues in the future. Marking all objects
Christoph Lameter's avatar
Christoph Lameter committed
898
		 * as used avoids touching the remaining objects.
Christoph Lameter's avatar
Christoph Lameter committed
899
		 */
900
		slab_fix(s, "Marking all objects used");
901
		page->inuse = page->objects;
902
		page->freelist = NULL;
Christoph Lameter's avatar
Christoph Lameter committed
903
904
905
906
	}
	return 0;
}

Christoph Lameter's avatar
Christoph Lameter committed
907
908
static int free_debug_processing(struct kmem_cache *s, struct page *page,
						void *object, void *addr)
Christoph Lameter's avatar
Christoph Lameter committed
909
910
911
912
913
{
	if (!check_slab(s, page))
		goto fail;

	if (!check_valid_pointer(s, page, object)) {
914
		slab_err(s, page, "Invalid object pointer 0x%p", object);
Christoph Lameter's avatar
Christoph Lameter committed
915
916
917
918
		goto fail;
	}

	if (on_freelist(s, page, object)) {
919
		object_err(s, page, object, "Object already free");
Christoph Lameter's avatar
Christoph Lameter committed
920
921
922
923
924
925
926
		goto fail;
	}

	if (!check_object(s, page, object, 1))
		return 0;

	if (unlikely(s != page->slab)) {
Ingo Molnar's avatar
Ingo Molnar committed
927
		if (!PageSlab(page)) {
928
929
			slab_err(s, page, "Attempt to free object(0x%p) "
				"outside of slab", object);
Ingo Molnar's avatar
Ingo Molnar committed
930
		} else if (!page->slab) {
Christoph Lameter's avatar
Christoph Lameter committed
931
			printk(KERN_ERR
932
				"SLUB <none>: no slab for object 0x%p.\n",
Christoph Lameter's avatar
Christoph Lameter committed
933
						object);
934
			dump_stack();
935
		} else
936
937
			object_err(s, page, object,
					"page slab pointer corrupt.");
Christoph Lameter's avatar
Christoph Lameter committed
938
939
		goto fail;
	}
Christoph Lameter's avatar
Christoph Lameter committed
940
941

	/* Special debug activities for freeing objects */
942
	if (!PageSlubFrozen(page) && !page->freelist)
Christoph Lameter's avatar
Christoph Lameter committed
943
944
945
946
947
		remove_full(s, page);
	if (s->flags & SLAB_STORE_USER)
		set_track(s, object, TRACK_FREE, addr);
	trace(s, page, object, 0);
	init_object(s, object, 0);
Christoph Lameter's avatar
Christoph Lameter committed
948
	return 1;
Christoph Lameter's avatar
Christoph Lameter committed
949

Christoph Lameter's avatar
Christoph Lameter committed
950
fail:
951
	slab_fix(s, "Object at 0x%p not freed", object);
Christoph Lameter's avatar
Christoph Lameter committed
952
953
954
	return 0;
}

955
956
static int __init setup_slub_debug(char *str)
{
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
	slub_debug = DEBUG_DEFAULT_FLAGS;
	if (*str++ != '=' || !*str)
		/*
		 * No options specified. Switch on full debugging.
		 */
		goto out;

	if (*str == ',')
		/*
		 * No options but restriction on slabs. This means full
		 * debugging for slabs matching a pattern.
		 */
		goto check_slabs;

	slub_debug = 0;
	if (*str == '-')
		/*
		 * Switch off all debugging measures.
		 */
		goto out;

	/*
	 * Determine which debug features should be switched on
	 */
981
	for (; *str && *str != ','; str++) {
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
		switch (tolower(*str)) {
		case 'f':
			slub_debug |= SLAB_DEBUG_FREE;
			break;
		case 'z':
			slub_debug |= SLAB_RED_ZONE;
			break;
		case 'p':
			slub_debug |= SLAB_POISON;
			break;
		case 'u':
			slub_debug |= SLAB_STORE_USER;
			break;
		case 't':
			slub_debug |= SLAB_TRACE;
			break;
		default:
			printk(KERN_ERR "slub_debug option '%c' "
1000
				"unknown. skipped\n", *str);