slub.c 96.8 KB
Newer Older
C
Christoph Lameter 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
/*
 * SLUB: A slab allocator that limits cache line use instead of queuing
 * objects in per cpu and per node lists.
 *
 * The allocator synchronizes using per slab locks and only
 * uses a centralized lock to manage a pool of partial slabs.
 *
 * (C) 2007 SGI, Christoph Lameter <clameter@sgi.com>
 */

#include <linux/mm.h>
#include <linux/module.h>
#include <linux/bit_spinlock.h>
#include <linux/interrupt.h>
#include <linux/bitops.h>
#include <linux/slab.h>
#include <linux/seq_file.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
#include <linux/mempolicy.h>
#include <linux/ctype.h>
#include <linux/kallsyms.h>
23
#include <linux/memory.h>
C
Christoph Lameter 已提交
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

/*
 * 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.
 *
C
Christoph Lameter 已提交
70 71
 * 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
C
Christoph Lameter 已提交
72
 * freed then the slab will show up again on the partial lists.
C
Christoph Lameter 已提交
73 74
 * We track full slabs for debugging purposes though because otherwise we
 * cannot scan all objects.
C
Christoph Lameter 已提交
75 76 77 78 79 80 81
 *
 * 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.
 *
82 83 84 85 86 87 88 89 90 91 92 93
 * 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
94
 * 			freelist that allows lockless access to
95 96
 * 			free objects in addition to the regular freelist
 * 			that requires the slab lock.
C
Christoph Lameter 已提交
97 98 99
 *
 * PageError		Slab requires special handling due to debug
 * 			options set. This moves	slab handling out of
100
 * 			the fast path and disables lockless freelists.
C
Christoph Lameter 已提交
101 102
 */

103 104 105 106 107 108 109 110
#define FROZEN (1 << PG_active)

#ifdef CONFIG_SLUB_DEBUG
#define SLABDEBUG (1 << PG_error)
#else
#define SLABDEBUG 0
#endif

111 112
static inline int SlabFrozen(struct page *page)
{
113
	return page->flags & FROZEN;
114 115 116 117
}

static inline void SetSlabFrozen(struct page *page)
{
118
	page->flags |= FROZEN;
119 120 121 122
}

static inline void ClearSlabFrozen(struct page *page)
{
123
	page->flags &= ~FROZEN;
124 125
}

126 127
static inline int SlabDebug(struct page *page)
{
128
	return page->flags & SLABDEBUG;
129 130 131 132
}

static inline void SetSlabDebug(struct page *page)
{
133
	page->flags |= SLABDEBUG;
134 135 136 137
}

static inline void ClearSlabDebug(struct page *page)
{
138
	page->flags &= ~SLABDEBUG;
139 140
}

C
Christoph Lameter 已提交
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
/*
 * Issues still to be resolved:
 *
 * - Support PAGE_ALLOC_DEBUG. Should be easy to do.
 *
 * - Variable sizing of the per node arrays
 */

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

#if PAGE_SHIFT <= 12

/*
 * Small page size. Make sure that we do not fragment memory
 */
#define DEFAULT_MAX_ORDER 1
#define DEFAULT_MIN_OBJECTS 4

#else

/*
 * Large page machines are customarily able to handle larger
 * page orders.
 */
#define DEFAULT_MAX_ORDER 2
#define DEFAULT_MIN_OBJECTS 8

#endif

171 172 173 174
/*
 * Mininum number of partial slabs. These will be left on the partial
 * lists even if they are empty. kmem_cache_shrink may reclaim them.
 */
C
Christoph Lameter 已提交
175
#define MIN_PARTIAL 5
C
Christoph Lameter 已提交
176

177 178 179 180 181 182 183
/*
 * 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

C
Christoph Lameter 已提交
184 185
#define DEBUG_DEFAULT_FLAGS (SLAB_DEBUG_FREE | SLAB_RED_ZONE | \
				SLAB_POISON | SLAB_STORE_USER)
C
Christoph Lameter 已提交
186

C
Christoph Lameter 已提交
187 188 189 190 191 192 193 194 195 196
/*
 * 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
197
#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long)
C
Christoph Lameter 已提交
198 199 200
#endif

#ifndef ARCH_SLAB_MINALIGN
201
#define ARCH_SLAB_MINALIGN __alignof__(unsigned long long)
C
Christoph Lameter 已提交
202 203 204
#endif

/* Internal SLUB flags */
205 206
#define __OBJECT_POISON		0x80000000 /* Poison object */
#define __SYSFS_ADD_DEFERRED	0x40000000 /* Not yet visible via sysfs */
C
Christoph Lameter 已提交
207

208 209 210 211 212
/* Not all arches define cache_line_size */
#ifndef cache_line_size
#define cache_line_size()	L1_CACHE_BYTES
#endif

C
Christoph Lameter 已提交
213 214 215 216 217 218 219 220 221
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 */
C
Christoph Lameter 已提交
222
	UP,		/* Everything works but does not show up in sysfs */
C
Christoph Lameter 已提交
223 224 225 226 227
	SYSFS		/* Sysfs up */
} slab_state = DOWN;

/* A list of all slab caches on the system */
static DECLARE_RWSEM(slub_lock);
A
Adrian Bunk 已提交
228
static LIST_HEAD(slab_caches);
C
Christoph Lameter 已提交
229

230 231 232 233 234 235 236 237 238 239 240 241
/*
 * 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 };

C
Christoph Lameter 已提交
242
#if defined(CONFIG_SYSFS) && defined(CONFIG_SLUB_DEBUG)
C
Christoph Lameter 已提交
243 244 245 246
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 *);
#else
247 248 249 250
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; }
static inline void sysfs_slab_remove(struct kmem_cache *s) {}
C
Christoph Lameter 已提交
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270
#endif

/********************************************************************
 * 			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
}

271 272
static inline struct kmem_cache_cpu *get_cpu_slab(struct kmem_cache *s, int cpu)
{
273 274 275 276 277
#ifdef CONFIG_SMP
	return s->cpu_slab[cpu];
#else
	return &s->cpu_slab;
#endif
278 279
}

280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
static inline int check_valid_pointer(struct kmem_cache *s,
				struct page *page, const void *object)
{
	void *base;

	if (!object)
		return 1;

	base = page_address(page);
	if (object < base || object >= base + s->objects * s->size ||
		(object - base) % s->size) {
		return 0;
	}

	return 1;
}

297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
/*
 * Slow version of get and set free pointer.
 *
 * This version requires touching the cache lines of kmem_cache which
 * we avoid to do in the fast alloc free paths. There we obtain the offset
 * from the page struct.
 */
static inline void *get_freepointer(struct kmem_cache *s, void *object)
{
	return *(void **)(object + s->offset);
}

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

/* Loop over all objects in a slab */
#define for_each_object(__p, __s, __addr) \
	for (__p = (__addr); __p < (__addr) + (__s)->objects * (__s)->size;\
			__p += (__s)->size)

/* Scan freelist */
#define for_each_free_object(__p, __s, __free) \
	for (__p = (__free); __p; __p = get_freepointer((__s), __p))

/* 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;
}

C
Christoph Lameter 已提交
329 330 331 332
#ifdef CONFIG_SLUB_DEBUG
/*
 * Debug settings:
 */
333 334 335
#ifdef CONFIG_SLUB_DEBUG_ON
static int slub_debug = DEBUG_DEFAULT_FLAGS;
#else
C
Christoph Lameter 已提交
336
static int slub_debug;
337
#endif
C
Christoph Lameter 已提交
338 339 340

static char *slub_debug_slabs;

C
Christoph Lameter 已提交
341 342 343 344 345 346 347 348 349 350 351 352 353
/*
 * 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) {
354
			printk(KERN_ERR "%8s 0x%p: ", text, addr + i);
C
Christoph Lameter 已提交
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
			newline = 0;
		}
		printk(" %02x", addr[i]);
		offset = i % 16;
		ascii[offset] = isgraph(addr[i]) ? addr[i] : '.';
		if (offset == 15) {
			printk(" %s\n",ascii);
			newline = 1;
		}
	}
	if (!newline) {
		i %= 16;
		while (i < 16) {
			printk("   ");
			ascii[i] = ' ';
			i++;
		}
		printk(" %s\n", ascii);
	}
}

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

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

	return p + alloc;
}

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

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

	p += alloc;
	if (addr) {
		p->addr = addr;
		p->cpu = smp_processor_id();
		p->pid = current ? current->pid : -1;
		p->when = jiffies;
	} else
		memset(p, 0, sizeof(struct track));
}

static void init_tracking(struct kmem_cache *s, void *object)
{
411 412 413 414 415
	if (!(s->flags & SLAB_STORE_USER))
		return;

	set_track(s, object, TRACK_FREE, NULL);
	set_track(s, object, TRACK_ALLOC, NULL);
C
Christoph Lameter 已提交
416 417 418 419 420 421 422
}

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

423
	printk(KERN_ERR "INFO: %s in ", s);
C
Christoph Lameter 已提交
424
	__print_symbol("%s", (unsigned long)t->addr);
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456
	printk(" age=%lu cpu=%u pid=%d\n", jiffies - t->when, t->cpu, t->pid);
}

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

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

static void print_page_info(struct page *page)
{
	printk(KERN_ERR "INFO: Slab 0x%p used=%u fp=0x%p flags=0x%04lx\n",
		page, page->inuse, page->freelist, page->flags);

}

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

	va_start(args, fmt);
	vsnprintf(buf, sizeof(buf), fmt, args);
	va_end(args);
	printk(KERN_ERR "========================================"
			"=====================================\n");
	printk(KERN_ERR "BUG %s: %s\n", s->name, buf);
	printk(KERN_ERR "----------------------------------------"
			"-------------------------------------\n\n");
C
Christoph Lameter 已提交
457 458
}

459 460 461 462 463 464 465 466 467 468 469 470
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)
C
Christoph Lameter 已提交
471 472
{
	unsigned int off;	/* Offset of last byte */
473 474 475 476 477 478 479 480 481 482 483 484 485
	u8 *addr = page_address(page);

	print_tracking(s, p);

	print_page_info(page);

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

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

	print_section("Object", p, min(s->objsize, 128));
C
Christoph Lameter 已提交
486 487 488 489 490 491 492 493 494 495

	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;

496
	if (s->flags & SLAB_STORE_USER)
C
Christoph Lameter 已提交
497 498 499 500
		off += 2 * sizeof(struct track);

	if (off != s->size)
		/* Beginning of the filler is the free pointer */
501 502 503
		print_section("Padding", p + off, s->size - off);

	dump_stack();
C
Christoph Lameter 已提交
504 505 506 507 508
}

static void object_err(struct kmem_cache *s, struct page *page,
			u8 *object, char *reason)
{
509 510
	slab_bug(s, reason);
	print_trailer(s, page, object);
C
Christoph Lameter 已提交
511 512
}

513
static void slab_err(struct kmem_cache *s, struct page *page, char *fmt, ...)
C
Christoph Lameter 已提交
514 515 516 517
{
	va_list args;
	char buf[100];

518 519
	va_start(args, fmt);
	vsnprintf(buf, sizeof(buf), fmt, args);
C
Christoph Lameter 已提交
520
	va_end(args);
521 522
	slab_bug(s, fmt);
	print_page_info(page);
C
Christoph Lameter 已提交
523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540
	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);
		p[s->objsize -1] = POISON_END;
	}

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

541
static u8 *check_bytes(u8 *start, unsigned int value, unsigned int bytes)
C
Christoph Lameter 已提交
542 543 544
{
	while (bytes) {
		if (*start != (u8)value)
545
			return start;
C
Christoph Lameter 已提交
546 547 548
		start++;
		bytes--;
	}
549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580
	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,
			u8* start, unsigned int value, unsigned int bytes)
{
	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;
C
Christoph Lameter 已提交
581 582 583 584 585 586 587 588 589
}

/*
 * 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.
C
Christoph Lameter 已提交
590
 *
C
Christoph Lameter 已提交
591 592 593 594 595
 * 	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.
C
Christoph Lameter 已提交
596 597 598
 * 	Padding is extended by another word if Redzoning is enabled and
 * 	objsize == inuse.
 *
C
Christoph Lameter 已提交
599 600 601 602
 * 	We fill with 0xbb (RED_INACTIVE) for inactive objects and with
 * 	0xcc (RED_ACTIVE) for objects in use.
 *
 * object + s->inuse
C
Christoph Lameter 已提交
603 604
 * 	Meta data starts here.
 *
C
Christoph Lameter 已提交
605 606
 * 	A. Free pointer (if we cannot overwrite object on free)
 * 	B. Tracking data for SLAB_STORE_USER
C
Christoph Lameter 已提交
607 608 609 610 611
 * 	C. Padding to reach required alignment boundary or at mininum
 * 		one word if debuggin is on to be able to detect writes
 * 		before the word boundary.
 *
 *	Padding is done using 0x5a (POISON_INUSE)
C
Christoph Lameter 已提交
612 613
 *
 * object + s->size
C
Christoph Lameter 已提交
614
 * 	Nothing is used beyond s->size.
C
Christoph Lameter 已提交
615
 *
C
Christoph Lameter 已提交
616 617
 * If slabcaches are merged then the objsize and inuse boundaries are mostly
 * ignored. And therefore no slab options that rely on these boundaries
C
Christoph Lameter 已提交
618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635
 * 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;

636 637
	return check_bytes_and_report(s, page, p, "Object padding",
				p + off, POISON_INUSE, s->size - off);
C
Christoph Lameter 已提交
638 639 640 641
}

static int slab_pad_check(struct kmem_cache *s, struct page *page)
{
642 643 644 645 646
	u8 *start;
	u8 *fault;
	u8 *end;
	int length;
	int remainder;
C
Christoph Lameter 已提交
647 648 649 650

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

651 652
	start = page_address(page);
	end = start + (PAGE_SIZE << s->order);
C
Christoph Lameter 已提交
653
	length = s->objects * s->size;
654
	remainder = end - (start + length);
C
Christoph Lameter 已提交
655 656 657
	if (!remainder)
		return 1;

658 659 660 661 662 663 664 665 666 667 668
	fault = check_bytes(start + length, POISON_INUSE, remainder);
	if (!fault)
		return 1;
	while (end > fault && end[-1] == POISON_INUSE)
		end--;

	slab_err(s, page, "Padding overwritten. 0x%p-0x%p", fault, end - 1);
	print_section("Padding", start, length);

	restore_bytes(s, "slab padding", POISON_INUSE, start, end);
	return 0;
C
Christoph Lameter 已提交
669 670 671 672 673 674 675 676 677 678 679 680
}

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;

681 682
		if (!check_bytes_and_report(s, page, object, "Redzone",
			endobject, red, s->inuse - s->objsize))
C
Christoph Lameter 已提交
683 684
			return 0;
	} else {
685 686 687
		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);
C
Christoph Lameter 已提交
688 689 690 691
	}

	if (s->flags & SLAB_POISON) {
		if (!active && (s->flags & __OBJECT_POISON) &&
692 693 694 695
			(!check_bytes_and_report(s, page, p, "Poison", p,
					POISON_FREE, s->objsize - 1) ||
			 !check_bytes_and_report(s, page, p, "Poison",
			 	p + s->objsize -1, POISON_END, 1)))
C
Christoph Lameter 已提交
696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715
			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
C
Christoph Lameter 已提交
716
		 * another error because the object count is now wrong.
C
Christoph Lameter 已提交
717 718 719 720 721 722 723 724 725 726 727 728
		 */
		set_freepointer(s, p, NULL);
		return 0;
	}
	return 1;
}

static int check_slab(struct kmem_cache *s, struct page *page)
{
	VM_BUG_ON(!irqs_disabled());

	if (!PageSlab(page)) {
729
		slab_err(s, page, "Not a valid slab page");
C
Christoph Lameter 已提交
730 731 732
		return 0;
	}
	if (page->inuse > s->objects) {
733 734
		slab_err(s, page, "inuse %u > max %u",
			s->name, page->inuse, s->objects);
C
Christoph Lameter 已提交
735 736 737 738 739 740 741 742
		return 0;
	}
	/* Slab_pad_check fixes things up after itself */
	slab_pad_check(s, page);
	return 1;
}

/*
C
Christoph Lameter 已提交
743 744
 * 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.
C
Christoph Lameter 已提交
745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761
 */
static int on_freelist(struct kmem_cache *s, struct page *page, void *search)
{
	int nr = 0;
	void *fp = page->freelist;
	void *object = NULL;

	while (fp && nr <= s->objects) {
		if (fp == search)
			return 1;
		if (!check_valid_pointer(s, page, fp)) {
			if (object) {
				object_err(s, page, object,
					"Freechain corrupt");
				set_freepointer(s, object, NULL);
				break;
			} else {
762
				slab_err(s, page, "Freepointer corrupt");
C
Christoph Lameter 已提交
763 764
				page->freelist = NULL;
				page->inuse = s->objects;
765
				slab_fix(s, "Freelist cleared");
C
Christoph Lameter 已提交
766 767 768 769 770 771 772 773 774 775
				return 0;
			}
			break;
		}
		object = fp;
		fp = get_freepointer(s, object);
		nr++;
	}

	if (page->inuse != s->objects - nr) {
776
		slab_err(s, page, "Wrong object count. Counter is %d but "
777
			"counted were %d", page->inuse, s->objects - nr);
C
Christoph Lameter 已提交
778
		page->inuse = s->objects - nr;
779
		slab_fix(s, "Object count adjusted.");
C
Christoph Lameter 已提交
780 781 782 783
	}
	return search == NULL;
}

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

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

		dump_stack();
	}
}

800
/*
C
Christoph Lameter 已提交
801
 * Tracking of fully allocated slabs for debugging purposes.
802
 */
C
Christoph Lameter 已提交
803
static void add_full(struct kmem_cache_node *n, struct page *page)
804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823
{
	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);
}

C
Christoph Lameter 已提交
824 825 826 827 828 829 830 831 832 833 834 835
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)
C
Christoph Lameter 已提交
836 837 838 839 840
{
	if (!check_slab(s, page))
		goto bad;

	if (object && !on_freelist(s, page, object)) {
841
		object_err(s, page, object, "Object already allocated");
842
		goto bad;
C
Christoph Lameter 已提交
843 844 845 846
	}

	if (!check_valid_pointer(s, page, object)) {
		object_err(s, page, object, "Freelist Pointer check fails");
847
		goto bad;
C
Christoph Lameter 已提交
848 849
	}

C
Christoph Lameter 已提交
850
	if (object && !check_object(s, page, object, 0))
C
Christoph Lameter 已提交
851 852
		goto bad;

C
Christoph Lameter 已提交
853 854 855 856 857
	/* 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);
C
Christoph Lameter 已提交
858
	return 1;
C
Christoph Lameter 已提交
859

C
Christoph Lameter 已提交
860 861 862 863 864
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
C
Christoph Lameter 已提交
865
		 * as used avoids touching the remaining objects.
C
Christoph Lameter 已提交
866
		 */
867
		slab_fix(s, "Marking all objects used");
C
Christoph Lameter 已提交
868 869 870 871 872 873
		page->inuse = s->objects;
		page->freelist = NULL;
	}
	return 0;
}

C
Christoph Lameter 已提交
874 875
static int free_debug_processing(struct kmem_cache *s, struct page *page,
						void *object, void *addr)
C
Christoph Lameter 已提交
876 877 878 879 880
{
	if (!check_slab(s, page))
		goto fail;

	if (!check_valid_pointer(s, page, object)) {
881
		slab_err(s, page, "Invalid object pointer 0x%p", object);
C
Christoph Lameter 已提交
882 883 884 885
		goto fail;
	}

	if (on_freelist(s, page, object)) {
886
		object_err(s, page, object, "Object already free");
C
Christoph Lameter 已提交
887 888 889 890 891 892 893 894
		goto fail;
	}

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

	if (unlikely(s != page->slab)) {
		if (!PageSlab(page))
895 896
			slab_err(s, page, "Attempt to free object(0x%p) "
				"outside of slab", object);
C
Christoph Lameter 已提交
897
		else
898
		if (!page->slab) {
C
Christoph Lameter 已提交
899
			printk(KERN_ERR
900
				"SLUB <none>: no slab for object 0x%p.\n",
C
Christoph Lameter 已提交
901
						object);
902 903
			dump_stack();
		}
C
Christoph Lameter 已提交
904
		else
905 906
			object_err(s, page, object,
					"page slab pointer corrupt.");
C
Christoph Lameter 已提交
907 908
		goto fail;
	}
C
Christoph Lameter 已提交
909 910 911 912 913 914 915 916

	/* Special debug activities for freeing objects */
	if (!SlabFrozen(page) && !page->freelist)
		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);
C
Christoph Lameter 已提交
917
	return 1;
C
Christoph Lameter 已提交
918

C
Christoph Lameter 已提交
919
fail:
920
	slab_fix(s, "Object at 0x%p not freed", object);
C
Christoph Lameter 已提交
921 922 923
	return 0;
}

C
Christoph Lameter 已提交
924 925
static int __init setup_slub_debug(char *str)
{
926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970
	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
	 */
	for ( ;*str && *str != ','; str++) {
		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' "
				"unknown. skipped\n",*str);
		}
C
Christoph Lameter 已提交
971 972
	}

973
check_slabs:
C
Christoph Lameter 已提交
974 975
	if (*str == ',')
		slub_debug_slabs = str + 1;
976
out:
C
Christoph Lameter 已提交
977 978 979 980 981
	return 1;
}

__setup("slub_debug", setup_slub_debug);

982 983
static unsigned long kmem_cache_flags(unsigned long objsize,
	unsigned long flags, const char *name,
984
	void (*ctor)(struct kmem_cache *, void *))
C
Christoph Lameter 已提交
985 986 987 988 989 990 991 992 993 994
{
	/*
	 * The page->offset field is only 16 bit wide. This is an offset
	 * in units of words from the beginning of an object. If the slab
	 * size is bigger then we cannot move the free pointer behind the
	 * object anymore.
	 *
	 * On 32 bit platforms the limit is 256k. On 64bit platforms
	 * the limit is 512k.
	 *
995
	 * Debugging or ctor may create a need to move the free
C
Christoph Lameter 已提交
996 997
	 * pointer. Fail if this happens.
	 */
998 999
	if (objsize >= 65535 * sizeof(void *)) {
		BUG_ON(flags & (SLAB_RED_ZONE | SLAB_POISON |
C
Christoph Lameter 已提交
1000
				SLAB_STORE_USER | SLAB_DESTROY_BY_RCU));
1001 1002
		BUG_ON(ctor);
	} else {
C
Christoph Lameter 已提交
1003 1004 1005 1006
		/*
		 * Enable debugging if selected on the kernel commandline.
		 */
		if (slub_debug && (!slub_debug_slabs ||
1007
		    strncmp(slub_debug_slabs, name,
C
Christoph Lameter 已提交
1008
		    	strlen(slub_debug_slabs)) == 0))
1009 1010 1011 1012
				flags |= slub_debug;
	}

	return flags;
C
Christoph Lameter 已提交
1013 1014
}
#else
C
Christoph Lameter 已提交
1015 1016
static inline void setup_object_debug(struct kmem_cache *s,
			struct page *page, void *object) {}
C
Christoph Lameter 已提交
1017

C
Christoph Lameter 已提交
1018 1019
static inline int alloc_debug_processing(struct kmem_cache *s,
	struct page *page, void *object, void *addr) { return 0; }
C
Christoph Lameter 已提交
1020

C
Christoph Lameter 已提交
1021 1022
static inline int free_debug_processing(struct kmem_cache *s,
	struct page *page, void *object, void *addr) { return 0; }
C
Christoph Lameter 已提交
1023 1024 1025 1026 1027

static inline int slab_pad_check(struct kmem_cache *s, struct page *page)
			{ return 1; }
static inline int check_object(struct kmem_cache *s, struct page *page,
			void *object, int active) { return 1; }
C
Christoph Lameter 已提交
1028
static inline void add_full(struct kmem_cache_node *n, struct page *page) {}
1029 1030
static inline unsigned long kmem_cache_flags(unsigned long objsize,
	unsigned long flags, const char *name,
1031
	void (*ctor)(struct kmem_cache *, void *))
1032 1033 1034
{
	return flags;
}
C
Christoph Lameter 已提交
1035 1036
#define slub_debug 0
#endif
C
Christoph Lameter 已提交
1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050
/*
 * Slab allocation and freeing
 */
static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
{
	struct page * page;
	int pages = 1 << s->order;

	if (s->order)
		flags |= __GFP_COMP;

	if (s->flags & SLAB_CACHE_DMA)
		flags |= SLUB_DMA;

1051 1052 1053
	if (s->flags & SLAB_RECLAIM_ACCOUNT)
		flags |= __GFP_RECLAIMABLE;

C
Christoph Lameter 已提交
1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072
	if (node == -1)
		page = alloc_pages(flags, s->order);
	else
		page = alloc_pages_node(node, flags, s->order);

	if (!page)
		return NULL;

	mod_zone_page_state(page_zone(page),
		(s->flags & SLAB_RECLAIM_ACCOUNT) ?
		NR_SLAB_RECLAIMABLE : NR_SLAB_UNRECLAIMABLE,
		pages);

	return page;
}

static void setup_object(struct kmem_cache *s, struct page *page,
				void *object)
{
C
Christoph Lameter 已提交
1073
	setup_object_debug(s, page, object);
1074
	if (unlikely(s->ctor))
1075
		s->ctor(s, object);
C
Christoph Lameter 已提交
1076 1077 1078 1079 1080 1081 1082 1083 1084 1085
}

static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
{
	struct page *page;
	struct kmem_cache_node *n;
	void *start;
	void *last;
	void *p;

C
Christoph Lameter 已提交
1086
	BUG_ON(flags & GFP_SLAB_BUG_MASK);
C
Christoph Lameter 已提交
1087

C
Christoph Lameter 已提交
1088 1089
	page = allocate_slab(s,
		flags & (GFP_RECLAIM_MASK | GFP_CONSTRAINT_MASK), node);
C
Christoph Lameter 已提交
1090 1091 1092 1093 1094 1095 1096 1097 1098 1099
	if (!page)
		goto out;

	n = get_node(s, page_to_nid(page));
	if (n)
		atomic_long_inc(&n->nr_slabs);
	page->slab = s;
	page->flags |= 1 << PG_slab;
	if (s->flags & (SLAB_DEBUG_FREE | SLAB_RED_ZONE | SLAB_POISON |
			SLAB_STORE_USER | SLAB_TRACE))
1100
		SetSlabDebug(page);
C
Christoph Lameter 已提交
1101 1102 1103 1104 1105 1106 1107

	start = page_address(page);

	if (unlikely(s->flags & SLAB_POISON))
		memset(start, POISON_INUSE, PAGE_SIZE << s->order);

	last = start;
1108
	for_each_object(p, s, start) {
C
Christoph Lameter 已提交
1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125
		setup_object(s, page, last);
		set_freepointer(s, last, p);
		last = p;
	}
	setup_object(s, page, last);
	set_freepointer(s, last, NULL);

	page->freelist = start;
	page->inuse = 0;
out:
	return page;
}

static void __free_slab(struct kmem_cache *s, struct page *page)
{
	int pages = 1 << s->order;

1126
	if (unlikely(SlabDebug(page))) {
C
Christoph Lameter 已提交
1127 1128 1129
		void *p;

		slab_pad_check(s, page);
1130
		for_each_object(p, s, page_address(page))
C
Christoph Lameter 已提交
1131
			check_object(s, page, p, 0);
1132
		ClearSlabDebug(page);
C
Christoph Lameter 已提交
1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169
	}

	mod_zone_page_state(page_zone(page),
		(s->flags & SLAB_RECLAIM_ACCOUNT) ?
		NR_SLAB_RECLAIMABLE : NR_SLAB_UNRECLAIMABLE,
		- pages);

	__free_pages(page, s->order);
}

static void rcu_free_slab(struct rcu_head *h)
{
	struct page *page;

	page = container_of((struct list_head *)h, struct page, lru);
	__free_slab(page->slab, page);
}

static void free_slab(struct kmem_cache *s, struct page *page)
{
	if (unlikely(s->flags & SLAB_DESTROY_BY_RCU)) {
		/*
		 * RCU free overloads the RCU head over the LRU
		 */
		struct rcu_head *head = (void *)&page->lru;

		call_rcu(head, rcu_free_slab);
	} else
		__free_slab(s, page);
}

static void discard_slab(struct kmem_cache *s, struct page *page)
{
	struct kmem_cache_node *n = get_node(s, page_to_nid(page));

	atomic_long_dec(&n->nr_slabs);
	reset_page_mapcount(page);
1170
	__ClearPageSlab(page);
C
Christoph Lameter 已提交
1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197
	free_slab(s, page);
}

/*
 * Per slab locking using the pagelock
 */
static __always_inline void slab_lock(struct page *page)
{
	bit_spin_lock(PG_locked, &page->flags);
}

static __always_inline void slab_unlock(struct page *page)
{
	bit_spin_unlock(PG_locked, &page->flags);
}

static __always_inline int slab_trylock(struct page *page)
{
	int rc = 1;

	rc = bit_spin_trylock(PG_locked, &page->flags);
	return rc;
}

/*
 * Management of partially allocated slabs
 */
C
Christoph Lameter 已提交
1198
static void add_partial_tail(struct kmem_cache_node *n, struct page *page)
C
Christoph Lameter 已提交
1199
{
C
Christoph Lameter 已提交
1200 1201 1202 1203 1204
	spin_lock(&n->list_lock);
	n->nr_partial++;
	list_add_tail(&page->lru, &n->partial);
	spin_unlock(&n->list_lock);
}
C
Christoph Lameter 已提交
1205

C
Christoph Lameter 已提交
1206 1207
static void add_partial(struct kmem_cache_node *n, struct page *page)
{
C
Christoph Lameter 已提交
1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
	spin_lock(&n->list_lock);
	n->nr_partial++;
	list_add(&page->lru, &n->partial);
	spin_unlock(&n->list_lock);
}

static void remove_partial(struct kmem_cache *s,
						struct page *page)
{
	struct kmem_cache_node *n = get_node(s, page_to_nid(page));

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

/*
C
Christoph Lameter 已提交
1226
 * Lock slab and remove from the partial list.
C
Christoph Lameter 已提交
1227
 *
C
Christoph Lameter 已提交
1228
 * Must hold list_lock.
C
Christoph Lameter 已提交
1229
 */
1230
static inline int lock_and_freeze_slab(struct kmem_cache_node *n, struct page *page)
C
Christoph Lameter 已提交
1231 1232 1233 1234
{
	if (slab_trylock(page)) {
		list_del(&page->lru);
		n->nr_partial--;
1235
		SetSlabFrozen(page);
C
Christoph Lameter 已提交
1236 1237 1238 1239 1240 1241
		return 1;
	}
	return 0;
}

/*
C
Christoph Lameter 已提交
1242
 * Try to allocate a partial slab from a specific node.
C
Christoph Lameter 已提交
1243 1244 1245 1246 1247 1248 1249 1250
 */
static struct page *get_partial_node(struct kmem_cache_node *n)
{
	struct page *page;

	/*
	 * Racy check. If we mistakenly see no partial slabs then we
	 * just allocate an empty slab. If we mistakenly try to get a
C
Christoph Lameter 已提交
1251 1252
	 * partial slab and there is none available then get_partials()
	 * will return NULL.
C
Christoph Lameter 已提交
1253 1254 1255 1256 1257 1258
	 */
	if (!n || !n->nr_partial)
		return NULL;

	spin_lock(&n->list_lock);
	list_for_each_entry(page, &n->partial, lru)
1259
		if (lock_and_freeze_slab(n, page))
C
Christoph Lameter 已提交
1260 1261 1262 1263 1264 1265 1266 1267
			goto out;
	page = NULL;
out:
	spin_unlock(&n->list_lock);
	return page;
}

/*
C
Christoph Lameter 已提交
1268
 * Get a page from somewhere. Search in increasing NUMA distances.
C
Christoph Lameter 已提交
1269 1270 1271 1272 1273 1274 1275 1276 1277
 */
static struct page *get_any_partial(struct kmem_cache *s, gfp_t flags)
{
#ifdef CONFIG_NUMA
	struct zonelist *zonelist;
	struct zone **z;
	struct page *page;

	/*
C
Christoph Lameter 已提交
1278 1279 1280 1281
	 * The defrag ratio allows a configuration of the tradeoffs between
	 * inter node defragmentation and node local allocations. A lower
	 * defrag_ratio increases the tendency to do local allocations
	 * instead of attempting to obtain partial slabs from other nodes.
C
Christoph Lameter 已提交
1282
	 *
C
Christoph Lameter 已提交
1283 1284 1285 1286
	 * If the defrag_ratio is set to 0 then kmalloc() always
	 * returns node local objects. If the ratio is higher then kmalloc()
	 * may return off node objects because partial slabs are obtained
	 * from other nodes and filled up.
C
Christoph Lameter 已提交
1287 1288
	 *
	 * If /sys/slab/xx/defrag_ratio is set to 100 (which makes
C
Christoph Lameter 已提交
1289 1290 1291 1292 1293
	 * defrag_ratio = 1000) then every (well almost) allocation will
	 * first attempt to defrag slab caches on other nodes. This means
	 * scanning over all nodes to look for partial slabs which may be
	 * expensive if we do it every time we are trying to find a slab
	 * with available objects.
C
Christoph Lameter 已提交
1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
	 */
	if (!s->defrag_ratio || get_cycles() % 1024 > s->defrag_ratio)
		return NULL;

	zonelist = &NODE_DATA(slab_node(current->mempolicy))
					->node_zonelists[gfp_zone(flags)];
	for (z = zonelist->zones; *z; z++) {
		struct kmem_cache_node *n;

		n = get_node(s, zone_to_nid(*z));

		if (n && cpuset_zone_allowed_hardwall(*z, flags) &&
C
Christoph Lameter 已提交
1306
				n->nr_partial > MIN_PARTIAL) {
C
Christoph Lameter 已提交
1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337
			page = get_partial_node(n);
			if (page)
				return page;
		}
	}
#endif
	return NULL;
}

/*
 * Get a partial page, lock it and return it.
 */
static struct page *get_partial(struct kmem_cache *s, gfp_t flags, int node)
{
	struct page *page;
	int searchnode = (node == -1) ? numa_node_id() : node;

	page = get_partial_node(get_node(s, searchnode));
	if (page || (flags & __GFP_THISNODE))
		return page;

	return get_any_partial(s, flags);
}

/*
 * Move a page back to the lists.
 *
 * Must be called with the slab lock held.
 *
 * On exit the slab lock will have been dropped.
 */
1338
static void unfreeze_slab(struct kmem_cache *s, struct page *page)
C
Christoph Lameter 已提交
1339
{
C
Christoph Lameter 已提交
1340 1341
	struct kmem_cache_node *n = get_node(s, page_to_nid(page));

1342
	ClearSlabFrozen(page);
C
Christoph Lameter 已提交
1343
	if (page->inuse) {
C
Christoph Lameter 已提交
1344

C
Christoph Lameter 已提交
1345
		if (page->freelist)
C
Christoph Lameter 已提交
1346
			add_partial(n, page);
1347
		else if (SlabDebug(page) && (s->flags & SLAB_STORE_USER))
C
Christoph Lameter 已提交
1348
			add_full(n, page);
C
Christoph Lameter 已提交
1349
		slab_unlock(page);
C
Christoph Lameter 已提交
1350

C
Christoph Lameter 已提交
1351
	} else {
C
Christoph Lameter 已提交
1352 1353
		if (n->nr_partial < MIN_PARTIAL) {
			/*
C
Christoph Lameter 已提交
1354 1355 1356 1357 1358 1359
			 * Adding an empty slab to the partial slabs in order
			 * to avoid page allocator overhead. This slab needs
			 * to come after the other slabs with objects in
			 * order to fill them up. That way the size of the
			 * partial list stays small. kmem_cache_shrink can
			 * reclaim empty slabs from the partial list.
C
Christoph Lameter 已提交
1360 1361 1362 1363 1364 1365 1366
			 */
			add_partial_tail(n, page);
			slab_unlock(page);
		} else {
			slab_unlock(page);
			discard_slab(s, page);
		}
C
Christoph Lameter 已提交
1367 1368 1369 1370 1371 1372
	}
}

/*
 * Remove the cpu slab
 */
1373
static void deactivate_slab(struct kmem_cache *s, struct kmem_cache_cpu *c)
C
Christoph Lameter 已提交
1374
{
1375
	struct page *page = c->page;
1376 1377 1378 1379 1380
	/*
	 * Merge cpu freelist into freelist. Typically we get here
	 * because both freelists are empty. So this is unlikely
	 * to occur.
	 */
1381
	while (unlikely(c->freelist)) {
1382 1383 1384
		void **object;

		/* Retrieve object from cpu_freelist */
1385
		object = c->freelist;
1386
		c->freelist = c->freelist[c->offset];
1387 1388

		/* And put onto the regular freelist */
1389
		object[c->offset] = page->freelist;
1390 1391 1392
		page->freelist = object;
		page->inuse--;
	}
1393
	c->page = NULL;
1394
	unfreeze_slab(s, page);
C
Christoph Lameter 已提交
1395 1396
}

1397
static inline void flush_slab(struct kmem_cache *s, struct kmem_cache_cpu *c)
C
Christoph Lameter 已提交
1398
{
1399 1400
	slab_lock(c->page);
	deactivate_slab(s, c);
C
Christoph Lameter 已提交
1401 1402 1403 1404 1405 1406
}

/*
 * Flush cpu slab.
 * Called from IPI handler with interrupts disabled.
 */
1407
static inline void __flush_cpu_slab(struct kmem_cache *s, int cpu)
C
Christoph Lameter 已提交
1408
{
1409
	struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
C
Christoph Lameter 已提交
1410

1411 1412
	if (likely(c && c->page))
		flush_slab(s, c);
C
Christoph Lameter 已提交
1413 1414 1415 1416 1417 1418
}

static void flush_cpu_slab(void *d)
{
	struct kmem_cache *s = d;

1419
	__flush_cpu_slab(s, smp_processor_id());
C
Christoph Lameter 已提交
1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
}

static void flush_all(struct kmem_cache *s)
{
#ifdef CONFIG_SMP
	on_each_cpu(flush_cpu_slab, s, 1, 1);
#else
	unsigned long flags;

	local_irq_save(flags);
	flush_cpu_slab(s);
	local_irq_restore(flags);
#endif
}

1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447
/*
 * Check if the objects in a per cpu structure fit numa
 * locality expectations.
 */
static inline int node_match(struct kmem_cache_cpu *c, int node)
{
#ifdef CONFIG_NUMA
	if (node != -1 && c->node != node)
		return 0;
#endif
	return 1;
}

C
Christoph Lameter 已提交
1448
/*
1449 1450 1451 1452
 * Slow path. The lockless freelist is empty or we need to perform
 * debugging duties.
 *
 * Interrupts are disabled.
C
Christoph Lameter 已提交
1453
 *
1454 1455 1456
 * Processing is still very fast if new objects have been freed to the
 * regular freelist. In that case we simply take over the regular freelist
 * as the lockless freelist and zap the regular freelist.
C
Christoph Lameter 已提交
1457
 *
1458 1459 1460
 * If that is not working then we fall back to the partial lists. We take the
 * first element of the freelist as the object to allocate now and move the
 * rest of the freelist to the lockless freelist.
C
Christoph Lameter 已提交
1461
 *
1462 1463
 * And if we were unable to get a new slab from the partial slab lists then
 * we need to allocate a new slab. This is slowest path since we may sleep.
C
Christoph Lameter 已提交
1464
 */
1465
static void *__slab_alloc(struct kmem_cache *s,
1466
		gfp_t gfpflags, int node, void *addr, struct kmem_cache_cpu *c)
C
Christoph Lameter 已提交
1467 1468
{
	void **object;
1469
	struct page *new;
C
Christoph Lameter 已提交
1470

1471
	if (!c->page)
C
Christoph Lameter 已提交
1472 1473
		goto new_slab;

1474 1475
	slab_lock(c->page);
	if (unlikely(!node_match(c, node)))
C
Christoph Lameter 已提交
1476
		goto another_slab;
1477
load_freelist:
1478
	object = c->page->freelist;
C
Christoph Lameter 已提交
1479 1480
	if (unlikely(!object))
		goto another_slab;
1481
	if (unlikely(SlabDebug(c->page)))
C
Christoph Lameter 已提交
1482 1483
		goto debug;

1484
	object = c->page->freelist;
1485
	c->freelist = object[c->offset];
1486 1487 1488 1489
	c->page->inuse = s->objects;
	c->page->freelist = NULL;
	c->node = page_to_nid(c->page);
	slab_unlock(c->page);
C
Christoph Lameter 已提交
1490 1491 1492
	return object;

another_slab:
1493
	deactivate_slab(s, c);
C
Christoph Lameter 已提交
1494 1495

new_slab:
1496 1497 1498
	new = get_partial(s, gfpflags, node);
	if (new) {
		c->page = new;
1499
		goto load_freelist;
C
Christoph Lameter 已提交
1500 1501
	}

1502 1503 1504
	if (gfpflags & __GFP_WAIT)
		local_irq_enable();

1505
	new = new_slab(s, gfpflags, node);
1506 1507 1508 1509

	if (gfpflags & __GFP_WAIT)
		local_irq_disable();

1510 1511
	if (new) {
		c = get_cpu_slab(s, smp_processor_id());
1512
		if (c->page)
1513 1514 1515 1516
			flush_slab(s, c);
		slab_lock(new);
		SetSlabFrozen(new);
		c->page = new;
1517
		goto load_freelist;
C
Christoph Lameter 已提交
1518 1519 1520
	}
	return NULL;
debug:
1521 1522
	object = c->page->freelist;
	if (!alloc_debug_processing(s, c->page, object, addr))
C
Christoph Lameter 已提交
1523
		goto another_slab;
1524

1525
	c->page->inuse++;
1526
	c->page->freelist = object[c->offset];
1527
	c->node = -1;
1528
	slab_unlock(c->page);
1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542
	return object;
}

/*
 * Inlined fastpath so that allocation functions (kmalloc, kmem_cache_alloc)
 * have the fastpath folded into their functions. So no function call
 * overhead for requests that can be satisfied on the fastpath.
 *
 * The fastpath works by first checking if the lockless freelist can be used.
 * If not then __slab_alloc is called for slow processing.
 *
 * Otherwise we can simply pick the next object from the lockless free list.
 */
static void __always_inline *slab_alloc(struct kmem_cache *s,
1543
		gfp_t gfpflags, int node, void *addr)
1544 1545 1546
{
	void **object;
	unsigned long flags;
1547
	struct kmem_cache_cpu *c;
1548 1549

	local_irq_save(flags);
1550
	c = get_cpu_slab(s, smp_processor_id());
1551
	if (unlikely(!c->freelist || !node_match(c, node)))
1552

1553
		object = __slab_alloc(s, gfpflags, node, addr, c);
1554 1555

	else {
1556
		object = c->freelist;
1557
		c->freelist = object[c->offset];
1558 1559
	}
	local_irq_restore(flags);
1560 1561

	if (unlikely((gfpflags & __GFP_ZERO) && object))
1562
		memset(object, 0, c->objsize);
1563

1564
	return object;
C
Christoph Lameter 已提交
1565 1566 1567 1568
}

void *kmem_cache_alloc(struct kmem_cache *s, gfp_t gfpflags)
{
1569
	return slab_alloc(s, gfpflags, -1, __builtin_return_address(0));
C
Christoph Lameter 已提交
1570 1571 1572 1573 1574 1575
}
EXPORT_SYMBOL(kmem_cache_alloc);

#ifdef CONFIG_NUMA
void *kmem_cache_alloc_node(struct kmem_cache *s, gfp_t gfpflags, int node)
{
1576
	return slab_alloc(s, gfpflags, node, __builtin_return_address(0));
C
Christoph Lameter 已提交
1577 1578 1579 1580 1581
}
EXPORT_SYMBOL(kmem_cache_alloc_node);
#endif

/*
1582 1583
 * Slow patch handling. This may still be called frequently since objects
 * have a longer lifetime than the cpu slabs in most processing loads.
C
Christoph Lameter 已提交
1584
 *
1585 1586 1587
 * So we still attempt to reduce cache line usage. Just take the slab
 * lock and free the item. If there is no additional partial page
 * handling required then we can return immediately.
C
Christoph Lameter 已提交
1588
 */
1589
static void __slab_free(struct kmem_cache *s, struct page *page,
1590
				void *x, void *addr, unsigned int offset)
C
Christoph Lameter 已提交
1591 1592 1593 1594 1595 1596
{
	void *prior;
	void **object = (void *)x;

	slab_lock(page);

1597
	if (unlikely(SlabDebug(page)))
C
Christoph Lameter 已提交
1598 1599
		goto debug;
checks_ok:
1600
	prior = object[offset] = page->freelist;
C
Christoph Lameter 已提交
1601 1602 1603
	page->freelist = object;
	page->inuse--;

1604
	if (unlikely(SlabFrozen(page)))
C
Christoph Lameter 已提交
1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615
		goto out_unlock;

	if (unlikely(!page->inuse))
		goto slab_empty;

	/*
	 * Objects left in the slab. If it
	 * was not on the partial list before
	 * then add it.
	 */
	if (unlikely(!prior))
C
Christoph Lameter 已提交
1616
		add_partial_tail(get_node(s, page_to_nid(page)), page);
C
Christoph Lameter 已提交
1617 1618 1619 1620 1621 1622 1623 1624

out_unlock:
	slab_unlock(page);
	return;

slab_empty:
	if (prior)
		/*
C
Christoph Lameter 已提交
1625
		 * Slab still on the partial list.
C
Christoph Lameter 已提交
1626 1627 1628 1629 1630 1631 1632 1633
		 */
		remove_partial(s, page);

	slab_unlock(page);
	discard_slab(s, page);
	return;

debug:
C
Christoph Lameter 已提交
1634
	if (!free_debug_processing(s, page, x, addr))
C
Christoph Lameter 已提交
1635 1636
		goto out_unlock;
	goto checks_ok;
C
Christoph Lameter 已提交
1637 1638
}

1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
/*
 * Fastpath with forced inlining to produce a kfree and kmem_cache_free that
 * can perform fastpath freeing without additional function calls.
 *
 * The fastpath is only possible if we are freeing to the current cpu slab
 * of this processor. This typically the case if we have just allocated
 * the item before.
 *
 * If fastpath is not possible then fall back to __slab_free where we deal
 * with all sorts of special processing.
 */
static void __always_inline slab_free(struct kmem_cache *s,
			struct page *page, void *x, void *addr)
{
	void **object = (void *)x;
	unsigned long flags;
1655
	struct kmem_cache_cpu *c;
1656 1657

	local_irq_save(flags);
P
Peter Zijlstra 已提交
1658
	debug_check_no_locks_freed(object, s->objsize);
1659
	c = get_cpu_slab(s, smp_processor_id());
1660
	if (likely(page == c->page && c->node >= 0)) {
1661
		object[c->offset] = c->freelist;
1662
		c->freelist = object;
1663
	} else
1664
		__slab_free(s, page, x, addr, c->offset);
1665 1666 1667 1668

	local_irq_restore(flags);
}

C
Christoph Lameter 已提交
1669 1670
void kmem_cache_free(struct kmem_cache *s, void *x)
{
C
Christoph Lameter 已提交
1671
	struct page *page;
C
Christoph Lameter 已提交
1672

1673
	page = virt_to_head_page(x);
C
Christoph Lameter 已提交
1674

C
Christoph Lameter 已提交
1675
	slab_free(s, page, x, __builtin_return_address(0));
C
Christoph Lameter 已提交
1676 1677 1678 1679 1680 1681
}
EXPORT_SYMBOL(kmem_cache_free);

/* Figure out on which slab object the object resides */
static struct page *get_object_page(const void *x)
{
1682
	struct page *page = virt_to_head_page(x);
C
Christoph Lameter 已提交
1683 1684 1685 1686 1687 1688 1689 1690

	if (!PageSlab(page))
		return NULL;

	return page;
}

/*
C
Christoph Lameter 已提交
1691 1692 1693 1694
 * Object placement in a slab is made very easy because we always start at
 * offset 0. If we tune the size of the object to the alignment then we can
 * get the required alignment by putting one properly sized object after
 * another.
C
Christoph Lameter 已提交
1695 1696 1697 1698
 *
 * Notice that the allocation order determines the sizes of the per cpu
 * caches. Each processor has always one slab available for allocations.
 * Increasing the allocation order reduces the number of times that slabs
C
Christoph Lameter 已提交
1699
 * must be moved on and off the partial lists and is therefore a factor in
C
Christoph Lameter 已提交
1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714
 * locking overhead.
 */

/*
 * Mininum / Maximum order of slab pages. This influences locking overhead
 * and slab fragmentation. A higher order reduces the number of partial slabs
 * and increases the number of allocations possible without having to
 * take the list_lock.
 */
static int slub_min_order;
static int slub_max_order = DEFAULT_MAX_ORDER;
static int slub_min_objects = DEFAULT_MIN_OBJECTS;

/*
 * Merge control. If this is set then no merging of slab caches will occur.
C
Christoph Lameter 已提交
1715
 * (Could be removed. This was introduced to pacify the merge skeptics.)
C
Christoph Lameter 已提交
1716 1717 1718 1719 1720 1721
 */
static int slub_nomerge;

/*
 * Calculate the order of allocation given an slab object size.
 *
C
Christoph Lameter 已提交
1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732
 * The order of allocation has significant impact on performance and other
 * system components. Generally order 0 allocations should be preferred since
 * order 0 does not cause fragmentation in the page allocator. Larger objects
 * be problematic to put into order 0 slabs because there may be too much
 * unused space left. We go to a higher order if more than 1/8th of the slab
 * would be wasted.
 *
 * In order to reach satisfactory performance we must ensure that a minimum
 * number of objects is in one slab. Otherwise we may generate too much
 * activity on the partial lists which requires taking the list_lock. This is
 * less a concern for large slabs though which are rarely used.
C
Christoph Lameter 已提交
1733
 *
C
Christoph Lameter 已提交
1734 1735 1736 1737
 * slub_max_order specifies the order where we begin to stop considering the
 * number of objects in a slab as critical. If we reach slub_max_order then
 * we try to keep the page order as low as possible. So we accept more waste
 * of space in favor of a small page order.
C
Christoph Lameter 已提交
1738
 *
C
Christoph Lameter 已提交
1739 1740 1741 1742
 * Higher order allocations also allow the placement of more objects in a
 * slab and thereby reduce object handling overhead. If the user has
 * requested a higher mininum order then we start with that one instead of
 * the smallest order which will fit the object.
C
Christoph Lameter 已提交
1743
 */
1744 1745
static inline int slab_order(int size, int min_objects,
				int max_order, int fract_leftover)
C
Christoph Lameter 已提交
1746 1747 1748
{
	int order;
	int rem;
1749
	int min_order = slub_min_order;
C
Christoph Lameter 已提交
1750

1751
	for (order = max(min_order,
1752 1753
				fls(min_objects * size - 1) - PAGE_SHIFT);
			order <= max_order; order++) {
C
Christoph Lameter 已提交
1754

1755
		unsigned long slab_size = PAGE_SIZE << order;
C
Christoph Lameter 已提交
1756

1757
		if (slab_size < min_objects * size)
C
Christoph Lameter 已提交
1758 1759 1760 1761
			continue;

		rem = slab_size % size;

1762
		if (rem <= slab_size / fract_leftover)
C
Christoph Lameter 已提交
1763 1764 1765
			break;

	}
C
Christoph Lameter 已提交
1766

C
Christoph Lameter 已提交
1767 1768 1769
	return order;
}

1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813
static inline int calculate_order(int size)
{
	int order;
	int min_objects;
	int fraction;

	/*
	 * Attempt to find best configuration for a slab. This
	 * works by first attempting to generate a layout with
	 * the best configuration and backing off gradually.
	 *
	 * First we reduce the acceptable waste in a slab. Then
	 * we reduce the minimum objects required in a slab.
	 */
	min_objects = slub_min_objects;
	while (min_objects > 1) {
		fraction = 8;
		while (fraction >= 4) {
			order = slab_order(size, min_objects,
						slub_max_order, fraction);
			if (order <= slub_max_order)
				return order;
			fraction /= 2;
		}
		min_objects /= 2;
	}

	/*
	 * We were unable to place multiple objects in a slab. Now
	 * lets see if we can place a single object there.
	 */
	order = slab_order(size, 1, slub_max_order, 1);
	if (order <= slub_max_order)
		return order;

	/*
	 * Doh this slab cannot be placed using slub_max_order.
	 */
	order = slab_order(size, 1, MAX_ORDER, 1);
	if (order <= MAX_ORDER)
		return order;
	return -ENOSYS;
}

C
Christoph Lameter 已提交
1814
/*
C
Christoph Lameter 已提交
1815
 * Figure out what the alignment of the objects will be.
C
Christoph Lameter 已提交
1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828
 */
static unsigned long calculate_alignment(unsigned long flags,
		unsigned long align, unsigned long size)
{
	/*
	 * If the user wants hardware cache aligned objects then
	 * follow that suggestion if the object is sufficiently
	 * large.
	 *
	 * The hardware cache alignment cannot override the
	 * specified alignment though. If that is greater
	 * then use it.
	 */
1829
	if ((flags & SLAB_HWCACHE_ALIGN) &&
1830 1831
			size > cache_line_size() / 2)
		return max_t(unsigned long, align, cache_line_size());
C
Christoph Lameter 已提交
1832 1833 1834 1835 1836 1837 1838

	if (align < ARCH_SLAB_MINALIGN)
		return ARCH_SLAB_MINALIGN;

	return ALIGN(align, sizeof(void *));
}

1839 1840 1841 1842 1843 1844
static void init_kmem_cache_cpu(struct kmem_cache *s,
			struct kmem_cache_cpu *c)
{
	c->page = NULL;
	c->freelist = NULL;
	c->node = 0;
1845 1846
	c->offset = s->offset / sizeof(void *);
	c->objsize = s->objsize;
1847 1848
}

C
Christoph Lameter 已提交
1849 1850 1851 1852 1853 1854
static void init_kmem_cache_node(struct kmem_cache_node *n)
{
	n->nr_partial = 0;
	atomic_long_set(&n->nr_slabs, 0);
	spin_lock_init(&n->list_lock);
	INIT_LIST_HEAD(&n->partial);
1855
#ifdef CONFIG_SLUB_DEBUG
1856
	INIT_LIST_HEAD(&n->full);
1857
#endif
C
Christoph Lameter 已提交
1858 1859
}

1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984
#ifdef CONFIG_SMP
/*
 * Per cpu array for per cpu structures.
 *
 * The per cpu array places all kmem_cache_cpu structures from one processor
 * close together meaning that it becomes possible that multiple per cpu
 * structures are contained in one cacheline. This may be particularly
 * beneficial for the kmalloc caches.
 *
 * A desktop system typically has around 60-80 slabs. With 100 here we are
 * likely able to get per cpu structures for all caches from the array defined
 * here. We must be able to cover all kmalloc caches during bootstrap.
 *
 * If the per cpu array is exhausted then fall back to kmalloc
 * of individual cachelines. No sharing is possible then.
 */
#define NR_KMEM_CACHE_CPU 100

static DEFINE_PER_CPU(struct kmem_cache_cpu,
				kmem_cache_cpu)[NR_KMEM_CACHE_CPU];

static DEFINE_PER_CPU(struct kmem_cache_cpu *, kmem_cache_cpu_free);
static cpumask_t kmem_cach_cpu_free_init_once = CPU_MASK_NONE;

static struct kmem_cache_cpu *alloc_kmem_cache_cpu(struct kmem_cache *s,
							int cpu, gfp_t flags)
{
	struct kmem_cache_cpu *c = per_cpu(kmem_cache_cpu_free, cpu);

	if (c)
		per_cpu(kmem_cache_cpu_free, cpu) =
				(void *)c->freelist;
	else {
		/* Table overflow: So allocate ourselves */
		c = kmalloc_node(
			ALIGN(sizeof(struct kmem_cache_cpu), cache_line_size()),
			flags, cpu_to_node(cpu));
		if (!c)
			return NULL;
	}

	init_kmem_cache_cpu(s, c);
	return c;
}

static void free_kmem_cache_cpu(struct kmem_cache_cpu *c, int cpu)
{
	if (c < per_cpu(kmem_cache_cpu, cpu) ||
			c > per_cpu(kmem_cache_cpu, cpu) + NR_KMEM_CACHE_CPU) {
		kfree(c);
		return;
	}
	c->freelist = (void *)per_cpu(kmem_cache_cpu_free, cpu);
	per_cpu(kmem_cache_cpu_free, cpu) = c;
}

static void free_kmem_cache_cpus(struct kmem_cache *s)
{
	int cpu;

	for_each_online_cpu(cpu) {
		struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);

		if (c) {
			s->cpu_slab[cpu] = NULL;
			free_kmem_cache_cpu(c, cpu);
		}
	}
}

static int alloc_kmem_cache_cpus(struct kmem_cache *s, gfp_t flags)
{
	int cpu;

	for_each_online_cpu(cpu) {
		struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);

		if (c)
			continue;

		c = alloc_kmem_cache_cpu(s, cpu, flags);
		if (!c) {
			free_kmem_cache_cpus(s);
			return 0;
		}
		s->cpu_slab[cpu] = c;
	}
	return 1;
}

/*
 * Initialize the per cpu array.
 */
static void init_alloc_cpu_cpu(int cpu)
{
	int i;

	if (cpu_isset(cpu, kmem_cach_cpu_free_init_once))
		return;

	for (i = NR_KMEM_CACHE_CPU - 1; i >= 0; i--)
		free_kmem_cache_cpu(&per_cpu(kmem_cache_cpu, cpu)[i], cpu);

	cpu_set(cpu, kmem_cach_cpu_free_init_once);
}

static void __init init_alloc_cpu(void)
{
	int cpu;

	for_each_online_cpu(cpu)
		init_alloc_cpu_cpu(cpu);
  }

#else
static inline void free_kmem_cache_cpus(struct kmem_cache *s) {}
static inline void init_alloc_cpu(void) {}

static inline int alloc_kmem_cache_cpus(struct kmem_cache *s, gfp_t flags)
{
	init_kmem_cache_cpu(s, &s->cpu_slab);
	return 1;
}
#endif

C
Christoph Lameter 已提交
1985 1986 1987 1988 1989 1990 1991
#ifdef CONFIG_NUMA
/*
 * No kmalloc_node yet so do it by hand. We know that this is the first
 * slab on the node for this slabcache. There are no concurrent accesses
 * possible.
 *
 * Note that this function only works on the kmalloc_node_cache
1992 1993
 * when allocating for the kmalloc_node_cache. This is used for bootstrapping
 * memory on a fresh node that has no slab structures yet.
C
Christoph Lameter 已提交
1994
 */
1995 1996
static struct kmem_cache_node *early_kmem_cache_node_alloc(gfp_t gfpflags,
							   int node)
C
Christoph Lameter 已提交
1997 1998 1999 2000 2001 2002
{
	struct page *page;
	struct kmem_cache_node *n;

	BUG_ON(kmalloc_caches->size < sizeof(struct kmem_cache_node));

2003
	page = new_slab(kmalloc_caches, gfpflags, node);
C
Christoph Lameter 已提交
2004 2005

	BUG_ON(!page);
2006 2007 2008 2009 2010 2011 2012
	if (page_to_nid(page) != node) {
		printk(KERN_ERR "SLUB: Unable to allocate memory from "
				"node %d\n", node);
		printk(KERN_ERR "SLUB: Allocating a useless per node structure "
				"in order to be able to continue\n");
	}

C
Christoph Lameter 已提交
2013 2014 2015 2016 2017
	n = page->freelist;
	BUG_ON(!n);
	page->freelist = get_freepointer(kmalloc_caches, n);
	page->inuse++;
	kmalloc_caches->node[node] = n;
2018
#ifdef CONFIG_SLUB_DEBUG
2019 2020
	init_object(kmalloc_caches, n, 1);
	init_tracking(kmalloc_caches, n);
2021
#endif
C
Christoph Lameter 已提交
2022 2023
	init_kmem_cache_node(n);
	atomic_long_inc(&n->nr_slabs);
C
Christoph Lameter 已提交
2024
	add_partial(n, page);
C
Christoph Lameter 已提交
2025 2026 2027 2028 2029 2030 2031
	return n;
}

static void free_kmem_cache_nodes(struct kmem_cache *s)
{
	int node;

C
Christoph Lameter 已提交
2032
	for_each_node_state(node, N_NORMAL_MEMORY) {
C
Christoph Lameter 已提交
2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049
		struct kmem_cache_node *n = s->node[node];
		if (n && n != &s->local_node)
			kmem_cache_free(kmalloc_caches, n);
		s->node[node] = NULL;
	}
}

static int init_kmem_cache_nodes(struct kmem_cache *s, gfp_t gfpflags)
{
	int node;
	int local_node;

	if (slab_state >= UP)
		local_node = page_to_nid(virt_to_page(s));
	else
		local_node = 0;

C
Christoph Lameter 已提交
2050
	for_each_node_state(node, N_NORMAL_MEMORY) {
C
Christoph Lameter 已提交
2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102
		struct kmem_cache_node *n;

		if (local_node == node)
			n = &s->local_node;
		else {
			if (slab_state == DOWN) {
				n = early_kmem_cache_node_alloc(gfpflags,
								node);
				continue;
			}
			n = kmem_cache_alloc_node(kmalloc_caches,
							gfpflags, node);

			if (!n) {
				free_kmem_cache_nodes(s);
				return 0;
			}

		}
		s->node[node] = n;
		init_kmem_cache_node(n);
	}
	return 1;
}
#else
static void free_kmem_cache_nodes(struct kmem_cache *s)
{
}

static int init_kmem_cache_nodes(struct kmem_cache *s, gfp_t gfpflags)
{
	init_kmem_cache_node(&s->local_node);
	return 1;
}
#endif

/*
 * calculate_sizes() determines the order and the distribution of data within
 * a slab object.
 */
static int calculate_sizes(struct kmem_cache *s)
{
	unsigned long flags = s->flags;
	unsigned long size = s->objsize;
	unsigned long align = s->align;

	/*
	 * Determine if we can poison the object itself. If the user of
	 * the slab may touch the object after free or before allocation
	 * then we should never poison the object itself.
	 */
	if ((flags & SLAB_POISON) && !(flags & SLAB_DESTROY_BY_RCU) &&
2103
			!s->ctor)
C
Christoph Lameter 已提交
2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114
		s->flags |= __OBJECT_POISON;
	else
		s->flags &= ~__OBJECT_POISON;

	/*
	 * Round up object size to the next word boundary. We can only
	 * place the free pointer at word boundaries and this determines
	 * the possible location of the free pointer.
	 */
	size = ALIGN(size, sizeof(void *));

C
Christoph Lameter 已提交
2115
#ifdef CONFIG_SLUB_DEBUG
C
Christoph Lameter 已提交
2116
	/*
C
Christoph Lameter 已提交
2117
	 * If we are Redzoning then check if there is some space between the
C
Christoph Lameter 已提交
2118
	 * end of the object and the free pointer. If not then add an
C
Christoph Lameter 已提交
2119
	 * additional word to have some bytes to store Redzone information.
C
Christoph Lameter 已提交
2120 2121 2122
	 */
	if ((flags & SLAB_RED_ZONE) && size == s->objsize)
		size += sizeof(void *);
C
Christoph Lameter 已提交
2123
#endif
C
Christoph Lameter 已提交
2124 2125

	/*
C
Christoph Lameter 已提交
2126 2127
	 * With that we have determined the number of bytes in actual use
	 * by the object. This is the potential offset to the free pointer.
C
Christoph Lameter 已提交
2128 2129 2130 2131
	 */
	s->inuse = size;

	if (((flags & (SLAB_DESTROY_BY_RCU | SLAB_POISON)) ||
2132
		s->ctor)) {
C
Christoph Lameter 已提交
2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144
		/*
		 * Relocate free pointer after the object if it is not
		 * permitted to overwrite the first word of the object on
		 * kmem_cache_free.
		 *
		 * This is the case if we do RCU, have a constructor or
		 * destructor or are poisoning the objects.
		 */
		s->offset = size;
		size += sizeof(void *);
	}

2145
#ifdef CONFIG_SLUB_DEBUG
C
Christoph Lameter 已提交
2146 2147 2148 2149 2150 2151 2152
	if (flags & SLAB_STORE_USER)
		/*
		 * Need to store information about allocs and frees after
		 * the object.
		 */
		size += 2 * sizeof(struct track);

2153
	if (flags & SLAB_RED_ZONE)
C
Christoph Lameter 已提交
2154 2155 2156 2157 2158 2159 2160 2161
		/*
		 * Add some empty padding so that we can catch
		 * overwrites from earlier objects rather than let
		 * tracking information or the free pointer be
		 * corrupted if an user writes before the start
		 * of the object.
		 */
		size += sizeof(void *);
C
Christoph Lameter 已提交
2162
#endif
C
Christoph Lameter 已提交
2163

C
Christoph Lameter 已提交
2164 2165
	/*
	 * Determine the alignment based on various parameters that the
2166 2167
	 * user specified and the dynamic determination of cache line size
	 * on bootup.
C
Christoph Lameter 已提交
2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187
	 */
	align = calculate_alignment(flags, align, s->objsize);

	/*
	 * SLUB stores one object immediately after another beginning from
	 * offset 0. In order to align the objects we have to simply size
	 * each object to conform to the alignment.
	 */
	size = ALIGN(size, align);
	s->size = size;

	s->order = calculate_order(size);
	if (s->order < 0)
		return 0;

	/*
	 * Determine the number of objects per slab
	 */
	s->objects = (PAGE_SIZE << s->order) / size;

2188
	return !!s->objects;
C
Christoph Lameter 已提交
2189 2190 2191 2192 2193 2194

}

static int kmem_cache_open(struct kmem_cache *s, gfp_t gfpflags,
		const char *name, size_t size,
		size_t align, unsigned long flags,
2195
		void (*ctor)(struct kmem_cache *, void *))
C
Christoph Lameter 已提交
2196 2197 2198 2199 2200 2201
{
	memset(s, 0, kmem_size);
	s->name = name;
	s->ctor = ctor;
	s->objsize = size;
	s->align = align;
2202
	s->flags = kmem_cache_flags(size, flags, name, ctor);
C
Christoph Lameter 已提交
2203 2204 2205 2206 2207 2208 2209 2210

	if (!calculate_sizes(s))
		goto error;

	s->refcount = 1;
#ifdef CONFIG_NUMA
	s->defrag_ratio = 100;
#endif
2211 2212
	if (!init_kmem_cache_nodes(s, gfpflags & ~SLUB_DMA))
		goto error;
C
Christoph Lameter 已提交
2213

2214
	if (alloc_kmem_cache_cpus(s, gfpflags & ~SLUB_DMA))
C
Christoph Lameter 已提交
2215
		return 1;
2216
	free_kmem_cache_nodes(s);
C
Christoph Lameter 已提交
2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238
error:
	if (flags & SLAB_PANIC)
		panic("Cannot create slab %s size=%lu realsize=%u "
			"order=%u offset=%u flags=%lx\n",
			s->name, (unsigned long)size, s->size, s->order,
			s->offset, flags);
	return 0;
}

/*
 * Check if a given pointer is valid
 */
int kmem_ptr_validate(struct kmem_cache *s, const void *object)
{
	struct page * page;

	page = get_object_page(object);

	if (!page || s != page->slab)
		/* No slab or wrong slab */
		return 0;

2239
	if (!check_valid_pointer(s, page, object))
C
Christoph Lameter 已提交
2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267
		return 0;

	/*
	 * We could also check if the object is on the slabs freelist.
	 * But this would be too expensive and it seems that the main
	 * purpose of kmem_ptr_valid is to check if the object belongs
	 * to a certain slab.
	 */
	return 1;
}
EXPORT_SYMBOL(kmem_ptr_validate);

/*
 * Determine the size of a slab object
 */
unsigned int kmem_cache_size(struct kmem_cache *s)
{
	return s->objsize;
}
EXPORT_SYMBOL(kmem_cache_size);

const char *kmem_cache_name(struct kmem_cache *s)
{
	return s->name;
}
EXPORT_SYMBOL(kmem_cache_name);

/*
C
Christoph Lameter 已提交
2268 2269
 * Attempt to free all slabs on a node. Return the number of slabs we
 * were unable to free.
C
Christoph Lameter 已提交
2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289
 */
static int free_list(struct kmem_cache *s, struct kmem_cache_node *n,
			struct list_head *list)
{
	int slabs_inuse = 0;
	unsigned long flags;
	struct page *page, *h;

	spin_lock_irqsave(&n->list_lock, flags);
	list_for_each_entry_safe(page, h, list, lru)
		if (!page->inuse) {
			list_del(&page->lru);
			discard_slab(s, page);
		} else
			slabs_inuse++;
	spin_unlock_irqrestore(&n->list_lock, flags);
	return slabs_inuse;
}

/*
C
Christoph Lameter 已提交
2290
 * Release all resources used by a slab cache.
C
Christoph Lameter 已提交
2291
 */
2292
static inline int kmem_cache_close(struct kmem_cache *s)
C
Christoph Lameter 已提交
2293 2294 2295 2296 2297 2298
{
	int node;

	flush_all(s);

	/* Attempt to free all objects */
2299
	free_kmem_cache_cpus(s);
C
Christoph Lameter 已提交
2300
	for_each_node_state(node, N_NORMAL_MEMORY) {
C
Christoph Lameter 已提交
2301 2302
		struct kmem_cache_node *n = get_node(s, node);

2303
		n->nr_partial -= free_list(s, n, &n->partial);
C
Christoph Lameter 已提交
2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320
		if (atomic_long_read(&n->nr_slabs))
			return 1;
	}
	free_kmem_cache_nodes(s);
	return 0;
}

/*
 * Close a cache and release the kmem_cache structure
 * (must be used for caches created using kmem_cache_create)
 */
void kmem_cache_destroy(struct kmem_cache *s)
{
	down_write(&slub_lock);
	s->refcount--;
	if (!s->refcount) {
		list_del(&s->list);
2321
		up_write(&slub_lock);
C
Christoph Lameter 已提交
2322 2323 2324 2325
		if (kmem_cache_close(s))
			WARN_ON(1);
		sysfs_slab_remove(s);
		kfree(s);
2326 2327
	} else
		up_write(&slub_lock);
C
Christoph Lameter 已提交
2328 2329 2330 2331 2332 2333 2334
}
EXPORT_SYMBOL(kmem_cache_destroy);

/********************************************************************
 *		Kmalloc subsystem
 *******************************************************************/

2335
struct kmem_cache kmalloc_caches[PAGE_SHIFT] __cacheline_aligned;
C
Christoph Lameter 已提交
2336 2337 2338
EXPORT_SYMBOL(kmalloc_caches);

#ifdef CONFIG_ZONE_DMA
2339
static struct kmem_cache *kmalloc_caches_dma[PAGE_SHIFT];
C
Christoph Lameter 已提交
2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386
#endif

static int __init setup_slub_min_order(char *str)
{
	get_option (&str, &slub_min_order);

	return 1;
}

__setup("slub_min_order=", setup_slub_min_order);

static int __init setup_slub_max_order(char *str)
{
	get_option (&str, &slub_max_order);

	return 1;
}

__setup("slub_max_order=", setup_slub_max_order);

static int __init setup_slub_min_objects(char *str)
{
	get_option (&str, &slub_min_objects);

	return 1;
}

__setup("slub_min_objects=", setup_slub_min_objects);

static int __init setup_slub_nomerge(char *str)
{
	slub_nomerge = 1;
	return 1;
}

__setup("slub_nomerge", setup_slub_nomerge);

static struct kmem_cache *create_kmalloc_cache(struct kmem_cache *s,
		const char *name, int size, gfp_t gfp_flags)
{
	unsigned int flags = 0;

	if (gfp_flags & SLUB_DMA)
		flags = SLAB_CACHE_DMA;

	down_write(&slub_lock);
	if (!kmem_cache_open(s, gfp_flags, name, size, ARCH_KMALLOC_MINALIGN,
2387
			flags, NULL))
C
Christoph Lameter 已提交
2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399
		goto panic;

	list_add(&s->list, &slab_caches);
	up_write(&slub_lock);
	if (sysfs_slab_add(s))
		goto panic;
	return s;

panic:
	panic("Creation of kmalloc slab %s size=%d failed.\n", name, size);
}

2400
#ifdef CONFIG_ZONE_DMA
2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417

static void sysfs_add_func(struct work_struct *w)
{
	struct kmem_cache *s;

	down_write(&slub_lock);
	list_for_each_entry(s, &slab_caches, list) {
		if (s->flags & __SYSFS_ADD_DEFERRED) {
			s->flags &= ~__SYSFS_ADD_DEFERRED;
			sysfs_slab_add(s);
		}
	}
	up_write(&slub_lock);
}

static DECLARE_WORK(sysfs_add_work, sysfs_add_func);

2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428
static noinline struct kmem_cache *dma_kmalloc_cache(int index, gfp_t flags)
{
	struct kmem_cache *s;
	char *text;
	size_t realsize;

	s = kmalloc_caches_dma[index];
	if (s)
		return s;

	/* Dynamically create dma cache */
2429 2430 2431 2432 2433 2434 2435 2436 2437
	if (flags & __GFP_WAIT)
		down_write(&slub_lock);
	else {
		if (!down_write_trylock(&slub_lock))
			goto out;
	}

	if (kmalloc_caches_dma[index])
		goto unlock_out;
2438

2439
	realsize = kmalloc_caches[index].objsize;
2440 2441 2442 2443 2444 2445 2446 2447 2448
	text = kasprintf(flags & ~SLUB_DMA, "kmalloc_dma-%d", (unsigned int)realsize),
	s = kmalloc(kmem_size, flags & ~SLUB_DMA);

	if (!s || !text || !kmem_cache_open(s, flags, text,
			realsize, ARCH_KMALLOC_MINALIGN,
			SLAB_CACHE_DMA|__SYSFS_ADD_DEFERRED, NULL)) {
		kfree(s);
		kfree(text);
		goto unlock_out;
2449
	}
2450 2451 2452 2453 2454 2455 2456

	list_add(&s->list, &slab_caches);
	kmalloc_caches_dma[index] = s;

	schedule_work(&sysfs_add_work);

unlock_out:
2457
	up_write(&slub_lock);
2458
out:
2459
	return kmalloc_caches_dma[index];
2460 2461 2462
}
#endif

2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495
/*
 * Conversion table for small slabs sizes / 8 to the index in the
 * kmalloc array. This is necessary for slabs < 192 since we have non power
 * of two cache sizes there. The size of larger slabs can be determined using
 * fls.
 */
static s8 size_index[24] = {
	3,	/* 8 */
	4,	/* 16 */
	5,	/* 24 */
	5,	/* 32 */
	6,	/* 40 */
	6,	/* 48 */
	6,	/* 56 */
	6,	/* 64 */
	1,	/* 72 */
	1,	/* 80 */
	1,	/* 88 */
	1,	/* 96 */
	7,	/* 104 */
	7,	/* 112 */
	7,	/* 120 */
	7,	/* 128 */
	2,	/* 136 */
	2,	/* 144 */
	2,	/* 152 */
	2,	/* 160 */
	2,	/* 168 */
	2,	/* 176 */
	2,	/* 184 */
	2	/* 192 */
};

C
Christoph Lameter 已提交
2496 2497
static struct kmem_cache *get_slab(size_t size, gfp_t flags)
{
2498
	int index;
C
Christoph Lameter 已提交
2499

2500 2501 2502
	if (size <= 192) {
		if (!size)
			return ZERO_SIZE_PTR;
C
Christoph Lameter 已提交
2503

2504
		index = size_index[(size - 1) / 8];
2505
	} else
2506
		index = fls(size - 1);
C
Christoph Lameter 已提交
2507 2508

#ifdef CONFIG_ZONE_DMA
2509
	if (unlikely((flags & SLUB_DMA)))
2510
		return dma_kmalloc_cache(index, flags);
2511

C
Christoph Lameter 已提交
2512 2513 2514 2515 2516 2517
#endif
	return &kmalloc_caches[index];
}

void *__kmalloc(size_t size, gfp_t flags)
{
2518
	struct kmem_cache *s;
C
Christoph Lameter 已提交
2519

2520 2521 2522 2523 2524 2525 2526
	if (unlikely(size > PAGE_SIZE / 2))
		return (void *)__get_free_pages(flags | __GFP_COMP,
							get_order(size));

	s = get_slab(size, flags);

	if (unlikely(ZERO_OR_NULL_PTR(s)))
2527 2528
		return s;

2529
	return slab_alloc(s, flags, -1, __builtin_return_address(0));
C
Christoph Lameter 已提交
2530 2531 2532 2533 2534 2535
}
EXPORT_SYMBOL(__kmalloc);

#ifdef CONFIG_NUMA
void *__kmalloc_node(size_t size, gfp_t flags, int node)
{
2536
	struct kmem_cache *s;
C
Christoph Lameter 已提交
2537

2538 2539 2540 2541 2542 2543 2544
	if (unlikely(size > PAGE_SIZE / 2))
		return (void *)__get_free_pages(flags | __GFP_COMP,
							get_order(size));

	s = get_slab(size, flags);

	if (unlikely(ZERO_OR_NULL_PTR(s)))
2545 2546
		return s;

2547
	return slab_alloc(s, flags, node, __builtin_return_address(0));
C
Christoph Lameter 已提交
2548 2549 2550 2551 2552 2553
}
EXPORT_SYMBOL(__kmalloc_node);
#endif

size_t ksize(const void *object)
{
2554
	struct page *page;
C
Christoph Lameter 已提交
2555 2556
	struct kmem_cache *s;

2557 2558
	BUG_ON(!object);
	if (unlikely(object == ZERO_SIZE_PTR))
2559 2560
		return 0;

2561
	page = virt_to_head_page(object);
C
Christoph Lameter 已提交
2562
	BUG_ON(!page);
2563 2564 2565 2566

	if (unlikely(!PageSlab(page)))
		return PAGE_SIZE << compound_order(page);

C
Christoph Lameter 已提交
2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595
	s = page->slab;
	BUG_ON(!s);

	/*
	 * Debugging requires use of the padding between object
	 * and whatever may come after it.
	 */
	if (s->flags & (SLAB_RED_ZONE | SLAB_POISON))
		return s->objsize;

	/*
	 * If we have the need to store the freelist pointer
	 * back there or track user information then we can
	 * only use the space before that information.
	 */
	if (s->flags & (SLAB_DESTROY_BY_RCU | SLAB_STORE_USER))
		return s->inuse;

	/*
	 * Else we can use all the padding etc for the allocation
	 */
	return s->size;
}
EXPORT_SYMBOL(ksize);

void kfree(const void *x)
{
	struct page *page;

2596
	if (unlikely(ZERO_OR_NULL_PTR(x)))
C
Christoph Lameter 已提交
2597 2598
		return;

2599
	page = virt_to_head_page(x);
2600 2601 2602 2603 2604
	if (unlikely(!PageSlab(page))) {
		put_page(page);
		return;
	}
	slab_free(page->slab, page, (void *)x, __builtin_return_address(0));
C
Christoph Lameter 已提交
2605 2606 2607
}
EXPORT_SYMBOL(kfree);

2608
/*
C
Christoph Lameter 已提交
2609 2610 2611 2612 2613 2614 2615 2616
 * kmem_cache_shrink removes empty slabs from the partial lists and sorts
 * the remaining slabs by the number of items in use. The slabs with the
 * most items in use come first. New allocations will then fill those up
 * and thus they can be removed from the partial lists.
 *
 * The slabs with the least items are placed last. This results in them
 * being allocated from last increasing the chance that the last objects
 * are freed in them.
2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632
 */
int kmem_cache_shrink(struct kmem_cache *s)
{
	int node;
	int i;
	struct kmem_cache_node *n;
	struct page *page;
	struct page *t;
	struct list_head *slabs_by_inuse =
		kmalloc(sizeof(struct list_head) * s->objects, GFP_KERNEL);
	unsigned long flags;

	if (!slabs_by_inuse)
		return -ENOMEM;

	flush_all(s);
C
Christoph Lameter 已提交
2633
	for_each_node_state(node, N_NORMAL_MEMORY) {
2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644
		n = get_node(s, node);

		if (!n->nr_partial)
			continue;

		for (i = 0; i < s->objects; i++)
			INIT_LIST_HEAD(slabs_by_inuse + i);

		spin_lock_irqsave(&n->list_lock, flags);

		/*
C
Christoph Lameter 已提交
2645
		 * Build lists indexed by the items in use in each slab.
2646
		 *
C
Christoph Lameter 已提交
2647 2648
		 * Note that concurrent frees may occur while we hold the
		 * list_lock. page->inuse here is the upper limit.
2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661
		 */
		list_for_each_entry_safe(page, t, &n->partial, lru) {
			if (!page->inuse && slab_trylock(page)) {
				/*
				 * Must hold slab lock here because slab_free
				 * may have freed the last object and be
				 * waiting to release the slab.
				 */
				list_del(&page->lru);
				n->nr_partial--;
				slab_unlock(page);
				discard_slab(s, page);
			} else {
2662 2663
				list_move(&page->lru,
				slabs_by_inuse + page->inuse);
2664 2665 2666 2667
			}
		}

		/*
C
Christoph Lameter 已提交
2668 2669
		 * Rebuild the partial list with the slabs filled up most
		 * first and the least used slabs at the end.
2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681
		 */
		for (i = s->objects - 1; i >= 0; i--)
			list_splice(slabs_by_inuse + i, n->partial.prev);

		spin_unlock_irqrestore(&n->list_lock, flags);
	}

	kfree(slabs_by_inuse);
	return 0;
}
EXPORT_SYMBOL(kmem_cache_shrink);

2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720
#if defined(CONFIG_NUMA) && defined(CONFIG_MEMORY_HOTPLUG)
static int slab_mem_going_offline_callback(void *arg)
{
	struct kmem_cache *s;

	down_read(&slub_lock);
	list_for_each_entry(s, &slab_caches, list)
		kmem_cache_shrink(s);
	up_read(&slub_lock);

	return 0;
}

static void slab_mem_offline_callback(void *arg)
{
	struct kmem_cache_node *n;
	struct kmem_cache *s;
	struct memory_notify *marg = arg;
	int offline_node;

	offline_node = marg->status_change_nid;

	/*
	 * If the node still has available memory. we need kmem_cache_node
	 * for it yet.
	 */
	if (offline_node < 0)
		return;

	down_read(&slub_lock);
	list_for_each_entry(s, &slab_caches, list) {
		n = get_node(s, offline_node);
		if (n) {
			/*
			 * if n->nr_slabs > 0, slabs still exist on the node
			 * that is going down. We were unable to free them,
			 * and offline_pages() function shoudn't call this
			 * callback. So, we must fail.
			 */
A
Al Viro 已提交
2721
			BUG_ON(atomic_long_read(&n->nr_slabs));
2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796

			s->node[offline_node] = NULL;
			kmem_cache_free(kmalloc_caches, n);
		}
	}
	up_read(&slub_lock);
}

static int slab_mem_going_online_callback(void *arg)
{
	struct kmem_cache_node *n;
	struct kmem_cache *s;
	struct memory_notify *marg = arg;
	int nid = marg->status_change_nid;
	int ret = 0;

	/*
	 * If the node's memory is already available, then kmem_cache_node is
	 * already created. Nothing to do.
	 */
	if (nid < 0)
		return 0;

	/*
	 * We are bringing a node online. No memory is availabe yet. We must
	 * allocate a kmem_cache_node structure in order to bring the node
	 * online.
	 */
	down_read(&slub_lock);
	list_for_each_entry(s, &slab_caches, list) {
		/*
		 * XXX: kmem_cache_alloc_node will fallback to other nodes
		 *      since memory is not yet available from the node that
		 *      is brought up.
		 */
		n = kmem_cache_alloc(kmalloc_caches, GFP_KERNEL);
		if (!n) {
			ret = -ENOMEM;
			goto out;
		}
		init_kmem_cache_node(n);
		s->node[nid] = n;
	}
out:
	up_read(&slub_lock);
	return ret;
}

static int slab_memory_callback(struct notifier_block *self,
				unsigned long action, void *arg)
{
	int ret = 0;

	switch (action) {
	case MEM_GOING_ONLINE:
		ret = slab_mem_going_online_callback(arg);
		break;
	case MEM_GOING_OFFLINE:
		ret = slab_mem_going_offline_callback(arg);
		break;
	case MEM_OFFLINE:
	case MEM_CANCEL_ONLINE:
		slab_mem_offline_callback(arg);
		break;
	case MEM_ONLINE:
	case MEM_CANCEL_OFFLINE:
		break;
	}

	ret = notifier_from_errno(ret);
	return ret;
}

#endif /* CONFIG_MEMORY_HOTPLUG */

C
Christoph Lameter 已提交
2797 2798 2799 2800 2801 2802 2803
/********************************************************************
 *			Basic setup of slabs
 *******************************************************************/

void __init kmem_cache_init(void)
{
	int i;
2804
	int caches = 0;
C
Christoph Lameter 已提交
2805

2806 2807
	init_alloc_cpu();

C
Christoph Lameter 已提交
2808 2809 2810
#ifdef CONFIG_NUMA
	/*
	 * Must first have the slab cache available for the allocations of the
C
Christoph Lameter 已提交
2811
	 * struct kmem_cache_node's. There is special bootstrap code in
C
Christoph Lameter 已提交
2812 2813 2814 2815
	 * kmem_cache_open for slab_state == DOWN.
	 */
	create_kmalloc_cache(&kmalloc_caches[0], "kmem_cache_node",
		sizeof(struct kmem_cache_node), GFP_KERNEL);
2816
	kmalloc_caches[0].refcount = -1;
2817
	caches++;
2818 2819

	hotplug_memory_notifier(slab_memory_callback, 1);
C
Christoph Lameter 已提交
2820 2821 2822 2823 2824 2825
#endif

	/* Able to allocate the per node structures */
	slab_state = PARTIAL;

	/* Caches that are not of the two-to-the-power-of size */
2826 2827
	if (KMALLOC_MIN_SIZE <= 64) {
		create_kmalloc_cache(&kmalloc_caches[1],
C
Christoph Lameter 已提交
2828
				"kmalloc-96", 96, GFP_KERNEL);
2829 2830 2831 2832
		caches++;
	}
	if (KMALLOC_MIN_SIZE <= 128) {
		create_kmalloc_cache(&kmalloc_caches[2],
C
Christoph Lameter 已提交
2833
				"kmalloc-192", 192, GFP_KERNEL);
2834 2835
		caches++;
	}
C
Christoph Lameter 已提交
2836

2837
	for (i = KMALLOC_SHIFT_LOW; i < PAGE_SHIFT; i++) {
C
Christoph Lameter 已提交
2838 2839
		create_kmalloc_cache(&kmalloc_caches[i],
			"kmalloc", 1 << i, GFP_KERNEL);
2840 2841
		caches++;
	}
C
Christoph Lameter 已提交
2842

2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857

	/*
	 * Patch up the size_index table if we have strange large alignment
	 * requirements for the kmalloc array. This is only the case for
	 * mips it seems. The standard arches will not generate any code here.
	 *
	 * Largest permitted alignment is 256 bytes due to the way we
	 * handle the index determination for the smaller caches.
	 *
	 * Make sure that nothing crazy happens if someone starts tinkering
	 * around with ARCH_KMALLOC_MINALIGN
	 */
	BUILD_BUG_ON(KMALLOC_MIN_SIZE > 256 ||
		(KMALLOC_MIN_SIZE & (KMALLOC_MIN_SIZE - 1)));

2858
	for (i = 8; i < KMALLOC_MIN_SIZE; i += 8)
2859 2860
		size_index[(i - 1) / 8] = KMALLOC_SHIFT_LOW;

C
Christoph Lameter 已提交
2861 2862 2863
	slab_state = UP;

	/* Provide the correct kmalloc names now that the caches are up */
2864
	for (i = KMALLOC_SHIFT_LOW; i < PAGE_SHIFT; i++)
C
Christoph Lameter 已提交
2865 2866 2867 2868 2869
		kmalloc_caches[i]. name =
			kasprintf(GFP_KERNEL, "kmalloc-%d", 1 << i);

#ifdef CONFIG_SMP
	register_cpu_notifier(&slab_notifier);
2870 2871 2872 2873
	kmem_size = offsetof(struct kmem_cache, cpu_slab) +
				nr_cpu_ids * sizeof(struct kmem_cache_cpu *);
#else
	kmem_size = sizeof(struct kmem_cache);
C
Christoph Lameter 已提交
2874 2875 2876 2877
#endif


	printk(KERN_INFO "SLUB: Genslabs=%d, HWalign=%d, Order=%d-%d, MinObjects=%d,"
2878 2879
		" CPUs=%d, Nodes=%d\n",
		caches, cache_line_size(),
C
Christoph Lameter 已提交
2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891
		slub_min_order, slub_max_order, slub_min_objects,
		nr_cpu_ids, nr_node_ids);
}

/*
 * Find a mergeable slab cache
 */
static int slab_unmergeable(struct kmem_cache *s)
{
	if (slub_nomerge || (s->flags & SLUB_NEVER_MERGE))
		return 1;

2892
	if (s->ctor)
C
Christoph Lameter 已提交
2893 2894
		return 1;

2895 2896 2897 2898 2899 2900
	/*
	 * We may have set a slab to be unmergeable during bootstrap.
	 */
	if (s->refcount < 0)
		return 1;

C
Christoph Lameter 已提交
2901 2902 2903 2904
	return 0;
}

static struct kmem_cache *find_mergeable(size_t size,
2905
		size_t align, unsigned long flags, const char *name,
2906
		void (*ctor)(struct kmem_cache *, void *))
C
Christoph Lameter 已提交
2907
{
2908
	struct kmem_cache *s;
C
Christoph Lameter 已提交
2909 2910 2911 2912

	if (slub_nomerge || (flags & SLUB_NEVER_MERGE))
		return NULL;

2913
	if (ctor)
C
Christoph Lameter 已提交
2914 2915 2916 2917 2918
		return NULL;

	size = ALIGN(size, sizeof(void *));
	align = calculate_alignment(flags, align, size);
	size = ALIGN(size, align);
2919
	flags = kmem_cache_flags(size, flags, name, NULL);
C
Christoph Lameter 已提交
2920

2921
	list_for_each_entry(s, &slab_caches, list) {
C
Christoph Lameter 已提交
2922 2923 2924 2925 2926 2927
		if (slab_unmergeable(s))
			continue;

		if (size > s->size)
			continue;

2928
		if ((flags & SLUB_MERGE_SAME) != (s->flags & SLUB_MERGE_SAME))
C
Christoph Lameter 已提交
2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946
				continue;
		/*
		 * Check if alignment is compatible.
		 * Courtesy of Adrian Drzewiecki
		 */
		if ((s->size & ~(align -1)) != s->size)
			continue;

		if (s->size - size >= sizeof(void *))
			continue;

		return s;
	}
	return NULL;
}

struct kmem_cache *kmem_cache_create(const char *name, size_t size,
		size_t align, unsigned long flags,
2947
		void (*ctor)(struct kmem_cache *, void *))
C
Christoph Lameter 已提交
2948 2949 2950 2951
{
	struct kmem_cache *s;

	down_write(&slub_lock);
2952
	s = find_mergeable(size, align, flags, name, ctor);
C
Christoph Lameter 已提交
2953
	if (s) {
2954 2955
		int cpu;

C
Christoph Lameter 已提交
2956 2957 2958 2959 2960 2961
		s->refcount++;
		/*
		 * Adjust the object sizes so that we clear
		 * the complete object on kzalloc.
		 */
		s->objsize = max(s->objsize, (int)size);
2962 2963 2964 2965 2966 2967 2968

		/*
		 * And then we need to update the object size in the
		 * per cpu structures
		 */
		for_each_online_cpu(cpu)
			get_cpu_slab(s, cpu)->objsize = s->objsize;
C
Christoph Lameter 已提交
2969
		s->inuse = max_t(int, s->inuse, ALIGN(size, sizeof(void *)));
2970
		up_write(&slub_lock);
C
Christoph Lameter 已提交
2971 2972
		if (sysfs_slab_alias(s, name))
			goto err;
2973 2974 2975 2976 2977
		return s;
	}
	s = kmalloc(kmem_size, GFP_KERNEL);
	if (s) {
		if (kmem_cache_open(s, GFP_KERNEL, name,
2978
				size, align, flags, ctor)) {
C
Christoph Lameter 已提交
2979
			list_add(&s->list, &slab_caches);
2980 2981 2982 2983 2984 2985
			up_write(&slub_lock);
			if (sysfs_slab_add(s))
				goto err;
			return s;
		}
		kfree(s);
C
Christoph Lameter 已提交
2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999
	}
	up_write(&slub_lock);

err:
	if (flags & SLAB_PANIC)
		panic("Cannot create slabcache %s\n", name);
	else
		s = NULL;
	return s;
}
EXPORT_SYMBOL(kmem_cache_create);

#ifdef CONFIG_SMP
/*
C
Christoph Lameter 已提交
3000 3001
 * Use the cpu notifier to insure that the cpu slabs are flushed when
 * necessary.
C
Christoph Lameter 已提交
3002 3003 3004 3005 3006
 */
static int __cpuinit slab_cpuup_callback(struct notifier_block *nfb,
		unsigned long action, void *hcpu)
{
	long cpu = (long)hcpu;
3007 3008
	struct kmem_cache *s;
	unsigned long flags;
C
Christoph Lameter 已提交
3009 3010

	switch (action) {
3011 3012 3013 3014 3015 3016 3017 3018 3019 3020
	case CPU_UP_PREPARE:
	case CPU_UP_PREPARE_FROZEN:
		init_alloc_cpu_cpu(cpu);
		down_read(&slub_lock);
		list_for_each_entry(s, &slab_caches, list)
			s->cpu_slab[cpu] = alloc_kmem_cache_cpu(s, cpu,
							GFP_KERNEL);
		up_read(&slub_lock);
		break;

C
Christoph Lameter 已提交
3021
	case CPU_UP_CANCELED:
3022
	case CPU_UP_CANCELED_FROZEN:
C
Christoph Lameter 已提交
3023
	case CPU_DEAD:
3024
	case CPU_DEAD_FROZEN:
3025 3026
		down_read(&slub_lock);
		list_for_each_entry(s, &slab_caches, list) {
3027 3028
			struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);

3029 3030 3031
			local_irq_save(flags);
			__flush_cpu_slab(s, cpu);
			local_irq_restore(flags);
3032 3033
			free_kmem_cache_cpu(c, cpu);
			s->cpu_slab[cpu] = NULL;
3034 3035
		}
		up_read(&slub_lock);
C
Christoph Lameter 已提交
3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049
		break;
	default:
		break;
	}
	return NOTIFY_OK;
}

static struct notifier_block __cpuinitdata slab_notifier =
	{ &slab_cpuup_callback, NULL, 0 };

#endif

void *__kmalloc_track_caller(size_t size, gfp_t gfpflags, void *caller)
{
3050 3051 3052 3053 3054 3055
	struct kmem_cache *s;

	if (unlikely(size > PAGE_SIZE / 2))
		return (void *)__get_free_pages(gfpflags | __GFP_COMP,
							get_order(size));
	s = get_slab(size, gfpflags);
C
Christoph Lameter 已提交
3056

3057
	if (unlikely(ZERO_OR_NULL_PTR(s)))
3058
		return s;
C
Christoph Lameter 已提交
3059

3060
	return slab_alloc(s, gfpflags, -1, caller);
C
Christoph Lameter 已提交
3061 3062 3063 3064 3065
}

void *__kmalloc_node_track_caller(size_t size, gfp_t gfpflags,
					int node, void *caller)
{
3066 3067 3068 3069 3070 3071
	struct kmem_cache *s;

	if (unlikely(size > PAGE_SIZE / 2))
		return (void *)__get_free_pages(gfpflags | __GFP_COMP,
							get_order(size));
	s = get_slab(size, gfpflags);
C
Christoph Lameter 已提交
3072

3073
	if (unlikely(ZERO_OR_NULL_PTR(s)))
3074
		return s;
C
Christoph Lameter 已提交
3075

3076
	return slab_alloc(s, gfpflags, node, caller);
C
Christoph Lameter 已提交
3077 3078
}

P
Pekka J Enberg 已提交
3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091
static unsigned long count_partial(struct kmem_cache_node *n)
{
	unsigned long flags;
	unsigned long x = 0;
	struct page *page;

	spin_lock_irqsave(&n->list_lock, flags);
	list_for_each_entry(page, &n->partial, lru)
		x += page->inuse;
	spin_unlock_irqrestore(&n->list_lock, flags);
	return x;
}

C
Christoph Lameter 已提交
3092
#if defined(CONFIG_SYSFS) && defined(CONFIG_SLUB_DEBUG)
3093 3094
static int validate_slab(struct kmem_cache *s, struct page *page,
						unsigned long *map)
3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105
{
	void *p;
	void *addr = page_address(page);

	if (!check_slab(s, page) ||
			!on_freelist(s, page, NULL))
		return 0;

	/* Now we know that a valid freelist exists */
	bitmap_zero(map, s->objects);

3106 3107
	for_each_free_object(p, s, page->freelist) {
		set_bit(slab_index(p, s, addr), map);
3108 3109 3110 3111
		if (!check_object(s, page, p, 0))
			return 0;
	}

3112 3113
	for_each_object(p, s, addr)
		if (!test_bit(slab_index(p, s, addr), map))
3114 3115 3116 3117 3118
			if (!check_object(s, page, p, 1))
				return 0;
	return 1;
}

3119 3120
static void validate_slab_slab(struct kmem_cache *s, struct page *page,
						unsigned long *map)
3121 3122
{
	if (slab_trylock(page)) {
3123
		validate_slab(s, page, map);
3124 3125 3126 3127 3128 3129
		slab_unlock(page);
	} else
		printk(KERN_INFO "SLUB %s: Skipped busy slab 0x%p\n",
			s->name, page);

	if (s->flags & DEBUG_DEFAULT_FLAGS) {
3130 3131
		if (!SlabDebug(page))
			printk(KERN_ERR "SLUB %s: SlabDebug not set "
3132 3133
				"on slab 0x%p\n", s->name, page);
	} else {
3134 3135
		if (SlabDebug(page))
			printk(KERN_ERR "SLUB %s: SlabDebug set on "
3136 3137 3138 3139
				"slab 0x%p\n", s->name, page);
	}
}

3140 3141
static int validate_slab_node(struct kmem_cache *s,
		struct kmem_cache_node *n, unsigned long *map)
3142 3143 3144 3145 3146 3147 3148 3149
{
	unsigned long count = 0;
	struct page *page;
	unsigned long flags;

	spin_lock_irqsave(&n->list_lock, flags);

	list_for_each_entry(page, &n->partial, lru) {
3150
		validate_slab_slab(s, page, map);
3151 3152 3153 3154 3155 3156 3157 3158 3159 3160
		count++;
	}
	if (count != n->nr_partial)
		printk(KERN_ERR "SLUB %s: %ld partial slabs counted but "
			"counter=%ld\n", s->name, count, n->nr_partial);

	if (!(s->flags & SLAB_STORE_USER))
		goto out;

	list_for_each_entry(page, &n->full, lru) {
3161
		validate_slab_slab(s, page, map);
3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173
		count++;
	}
	if (count != atomic_long_read(&n->nr_slabs))
		printk(KERN_ERR "SLUB: %s %ld slabs counted but "
			"counter=%ld\n", s->name, count,
			atomic_long_read(&n->nr_slabs));

out:
	spin_unlock_irqrestore(&n->list_lock, flags);
	return count;
}

3174
static long validate_slab_cache(struct kmem_cache *s)
3175 3176 3177
{
	int node;
	unsigned long count = 0;
3178 3179 3180 3181 3182
	unsigned long *map = kmalloc(BITS_TO_LONGS(s->objects) *
				sizeof(unsigned long), GFP_KERNEL);

	if (!map)
		return -ENOMEM;
3183 3184

	flush_all(s);
C
Christoph Lameter 已提交
3185
	for_each_node_state(node, N_NORMAL_MEMORY) {
3186 3187
		struct kmem_cache_node *n = get_node(s, node);

3188
		count += validate_slab_node(s, n, map);
3189
	}
3190
	kfree(map);
3191 3192 3193
	return count;
}

3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248
#ifdef SLUB_RESILIENCY_TEST
static void resiliency_test(void)
{
	u8 *p;

	printk(KERN_ERR "SLUB resiliency testing\n");
	printk(KERN_ERR "-----------------------\n");
	printk(KERN_ERR "A. Corruption after allocation\n");

	p = kzalloc(16, GFP_KERNEL);
	p[16] = 0x12;
	printk(KERN_ERR "\n1. kmalloc-16: Clobber Redzone/next pointer"
			" 0x12->0x%p\n\n", p + 16);

	validate_slab_cache(kmalloc_caches + 4);

	/* Hmmm... The next two are dangerous */
	p = kzalloc(32, GFP_KERNEL);
	p[32 + sizeof(void *)] = 0x34;
	printk(KERN_ERR "\n2. kmalloc-32: Clobber next pointer/next slab"
		 	" 0x34 -> -0x%p\n", p);
	printk(KERN_ERR "If allocated object is overwritten then not detectable\n\n");

	validate_slab_cache(kmalloc_caches + 5);
	p = kzalloc(64, GFP_KERNEL);
	p += 64 + (get_cycles() & 0xff) * sizeof(void *);
	*p = 0x56;
	printk(KERN_ERR "\n3. kmalloc-64: corrupting random byte 0x56->0x%p\n",
									p);
	printk(KERN_ERR "If allocated object is overwritten then not detectable\n\n");
	validate_slab_cache(kmalloc_caches + 6);

	printk(KERN_ERR "\nB. Corruption after free\n");
	p = kzalloc(128, GFP_KERNEL);
	kfree(p);
	*p = 0x78;
	printk(KERN_ERR "1. kmalloc-128: Clobber first word 0x78->0x%p\n\n", p);
	validate_slab_cache(kmalloc_caches + 7);

	p = kzalloc(256, GFP_KERNEL);
	kfree(p);
	p[50] = 0x9a;
	printk(KERN_ERR "\n2. kmalloc-256: Clobber 50th byte 0x9a->0x%p\n\n", p);
	validate_slab_cache(kmalloc_caches + 8);

	p = kzalloc(512, GFP_KERNEL);
	kfree(p);
	p[512] = 0xab;
	printk(KERN_ERR "\n3. kmalloc-512: Clobber redzone 0xab->0x%p\n\n", p);
	validate_slab_cache(kmalloc_caches + 9);
}
#else
static void resiliency_test(void) {};
#endif

3249
/*
C
Christoph Lameter 已提交
3250
 * Generate lists of code addresses where slabcache objects are allocated
3251 3252 3253 3254 3255 3256
 * and freed.
 */

struct location {
	unsigned long count;
	void *addr;
3257 3258 3259 3260 3261 3262 3263
	long long sum_time;
	long min_time;
	long max_time;
	long min_pid;
	long max_pid;
	cpumask_t cpus;
	nodemask_t nodes;
3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278
};

struct loc_track {
	unsigned long max;
	unsigned long count;
	struct location *loc;
};

static void free_loc_track(struct loc_track *t)
{
	if (t->max)
		free_pages((unsigned long)t->loc,
			get_order(sizeof(struct location) * t->max));
}

3279
static int alloc_loc_track(struct loc_track *t, unsigned long max, gfp_t flags)
3280 3281 3282 3283 3284 3285
{
	struct location *l;
	int order;

	order = get_order(sizeof(struct location) * max);

3286
	l = (void *)__get_free_pages(flags, order);
3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299
	if (!l)
		return 0;

	if (t->count) {
		memcpy(l, t->loc, sizeof(struct location) * t->count);
		free_loc_track(t);
	}
	t->max = max;
	t->loc = l;
	return 1;
}

static int add_location(struct loc_track *t, struct kmem_cache *s,
3300
				const struct track *track)
3301 3302 3303 3304
{
	long start, end, pos;
	struct location *l;
	void *caddr;
3305
	unsigned long age = jiffies - track->when;
3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320

	start = -1;
	end = t->count;

	for ( ; ; ) {
		pos = start + (end - start + 1) / 2;

		/*
		 * There is nothing at "end". If we end up there
		 * we need to add something to before end.
		 */
		if (pos == end)
			break;

		caddr = t->loc[pos].addr;
3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339
		if (track->addr == caddr) {

			l = &t->loc[pos];
			l->count++;
			if (track->when) {
				l->sum_time += age;
				if (age < l->min_time)
					l->min_time = age;
				if (age > l->max_time)
					l->max_time = age;

				if (track->pid < l->min_pid)
					l->min_pid = track->pid;
				if (track->pid > l->max_pid)
					l->max_pid = track->pid;

				cpu_set(track->cpu, l->cpus);
			}
			node_set(page_to_nid(virt_to_page(track)), l->nodes);
3340 3341 3342
			return 1;
		}

3343
		if (track->addr < caddr)
3344 3345 3346 3347 3348 3349
			end = pos;
		else
			start = pos;
	}

	/*
C
Christoph Lameter 已提交
3350
	 * Not found. Insert new tracking element.
3351
	 */
3352
	if (t->count >= t->max && !alloc_loc_track(t, 2 * t->max, GFP_ATOMIC))
3353 3354 3355 3356 3357 3358 3359 3360
		return 0;

	l = t->loc + pos;
	if (pos < t->count)
		memmove(l + 1, l,
			(t->count - pos) * sizeof(struct location));
	t->count++;
	l->count = 1;
3361 3362 3363 3364 3365 3366 3367 3368 3369 3370
	l->addr = track->addr;
	l->sum_time = age;
	l->min_time = age;
	l->max_time = age;
	l->min_pid = track->pid;
	l->max_pid = track->pid;
	cpus_clear(l->cpus);
	cpu_set(track->cpu, l->cpus);
	nodes_clear(l->nodes);
	node_set(page_to_nid(virt_to_page(track)), l->nodes);
3371 3372 3373 3374 3375 3376 3377
	return 1;
}

static void process_slab(struct loc_track *t, struct kmem_cache *s,
		struct page *page, enum track_item alloc)
{
	void *addr = page_address(page);
3378
	DECLARE_BITMAP(map, s->objects);
3379 3380 3381
	void *p;

	bitmap_zero(map, s->objects);
3382 3383
	for_each_free_object(p, s, page->freelist)
		set_bit(slab_index(p, s, addr), map);
3384

3385
	for_each_object(p, s, addr)
3386 3387
		if (!test_bit(slab_index(p, s, addr), map))
			add_location(t, s, get_track(s, p, alloc));
3388 3389 3390 3391 3392 3393 3394
}

static int list_locations(struct kmem_cache *s, char *buf,
					enum track_item alloc)
{
	int n = 0;
	unsigned long i;
3395
	struct loc_track t = { 0, 0, NULL };
3396 3397
	int node;

3398
	if (!alloc_loc_track(&t, PAGE_SIZE / sizeof(struct location),
3399
			GFP_TEMPORARY))
3400
		return sprintf(buf, "Out of memory\n");
3401 3402 3403 3404

	/* Push back cpu slabs */
	flush_all(s);

C
Christoph Lameter 已提交
3405
	for_each_node_state(node, N_NORMAL_MEMORY) {
3406 3407 3408 3409
		struct kmem_cache_node *n = get_node(s, node);
		unsigned long flags;
		struct page *page;

3410
		if (!atomic_long_read(&n->nr_slabs))
3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421
			continue;

		spin_lock_irqsave(&n->list_lock, flags);
		list_for_each_entry(page, &n->partial, lru)
			process_slab(&t, s, page, alloc);
		list_for_each_entry(page, &n->full, lru)
			process_slab(&t, s, page, alloc);
		spin_unlock_irqrestore(&n->list_lock, flags);
	}

	for (i = 0; i < t.count; i++) {
3422
		struct location *l = &t.loc[i];
3423 3424 3425

		if (n > PAGE_SIZE - 100)
			break;
3426 3427 3428 3429
		n += sprintf(buf + n, "%7ld ", l->count);

		if (l->addr)
			n += sprint_symbol(buf + n, (unsigned long)l->addr);
3430 3431
		else
			n += sprintf(buf + n, "<not-available>");
3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450

		if (l->sum_time != l->min_time) {
			unsigned long remainder;

			n += sprintf(buf + n, " age=%ld/%ld/%ld",
			l->min_time,
			div_long_long_rem(l->sum_time, l->count, &remainder),
			l->max_time);
		} else
			n += sprintf(buf + n, " age=%ld",
				l->min_time);

		if (l->min_pid != l->max_pid)
			n += sprintf(buf + n, " pid=%ld-%ld",
				l->min_pid, l->max_pid);
		else
			n += sprintf(buf + n, " pid=%ld",
				l->min_pid);

3451 3452
		if (num_online_cpus() > 1 && !cpus_empty(l->cpus) &&
				n < PAGE_SIZE - 60) {
3453 3454 3455 3456 3457
			n += sprintf(buf + n, " cpus=");
			n += cpulist_scnprintf(buf + n, PAGE_SIZE - n - 50,
					l->cpus);
		}

3458 3459
		if (num_online_nodes() > 1 && !nodes_empty(l->nodes) &&
				n < PAGE_SIZE - 60) {
3460 3461 3462 3463 3464
			n += sprintf(buf + n, " nodes=");
			n += nodelist_scnprintf(buf + n, PAGE_SIZE - n - 50,
					l->nodes);
		}

3465 3466 3467 3468 3469 3470 3471 3472 3473
		n += sprintf(buf + n, "\n");
	}

	free_loc_track(&t);
	if (!t.count)
		n += sprintf(buf, "No data\n");
	return n;
}

C
Christoph Lameter 已提交
3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499
enum slab_stat_type {
	SL_FULL,
	SL_PARTIAL,
	SL_CPU,
	SL_OBJECTS
};

#define SO_FULL		(1 << SL_FULL)
#define SO_PARTIAL	(1 << SL_PARTIAL)
#define SO_CPU		(1 << SL_CPU)
#define SO_OBJECTS	(1 << SL_OBJECTS)

static unsigned long slab_objects(struct kmem_cache *s,
			char *buf, unsigned long flags)
{
	unsigned long total = 0;
	int cpu;
	int node;
	int x;
	unsigned long *nodes;
	unsigned long *per_cpu;

	nodes = kzalloc(2 * sizeof(unsigned long) * nr_node_ids, GFP_KERNEL);
	per_cpu = nodes + nr_node_ids;

	for_each_possible_cpu(cpu) {
3500
		struct page *page;
3501
		int node;
3502
		struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);
C
Christoph Lameter 已提交
3503

3504 3505 3506 3507
		if (!c)
			continue;

		page = c->page;
3508 3509 3510
		node = c->node;
		if (node < 0)
			continue;
C
Christoph Lameter 已提交
3511 3512 3513 3514 3515 3516 3517 3518 3519
		if (page) {
			if (flags & SO_CPU) {
				int x = 0;

				if (flags & SO_OBJECTS)
					x = page->inuse;
				else
					x = 1;
				total += x;
3520
				nodes[node] += x;
C
Christoph Lameter 已提交
3521
			}
3522
			per_cpu[node]++;
C
Christoph Lameter 已提交
3523 3524 3525
		}
	}

C
Christoph Lameter 已提交
3526
	for_each_node_state(node, N_NORMAL_MEMORY) {
C
Christoph Lameter 已提交
3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538
		struct kmem_cache_node *n = get_node(s, node);

		if (flags & SO_PARTIAL) {
			if (flags & SO_OBJECTS)
				x = count_partial(n);
			else
				x = n->nr_partial;
			total += x;
			nodes[node] += x;
		}

		if (flags & SO_FULL) {
3539
			int full_slabs = atomic_long_read(&n->nr_slabs)
C
Christoph Lameter 已提交
3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553
					- per_cpu[node]
					- n->nr_partial;

			if (flags & SO_OBJECTS)
				x = full_slabs * s->objects;
			else
				x = full_slabs;
			total += x;
			nodes[node] += x;
		}
	}

	x = sprintf(buf, "%lu", total);
#ifdef CONFIG_NUMA
C
Christoph Lameter 已提交
3554
	for_each_node_state(node, N_NORMAL_MEMORY)
C
Christoph Lameter 已提交
3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567
		if (nodes[node])
			x += sprintf(buf + x, " N%d=%lu",
					node, nodes[node]);
#endif
	kfree(nodes);
	return x + sprintf(buf + x, "\n");
}

static int any_slab_objects(struct kmem_cache *s)
{
	int node;
	int cpu;

3568 3569 3570 3571
	for_each_possible_cpu(cpu) {
		struct kmem_cache_cpu *c = get_cpu_slab(s, cpu);

		if (c && c->page)
C
Christoph Lameter 已提交
3572
			return 1;
3573
	}
C
Christoph Lameter 已提交
3574

3575
	for_each_online_node(node) {
C
Christoph Lameter 已提交
3576 3577
		struct kmem_cache_node *n = get_node(s, node);

3578 3579 3580
		if (!n)
			continue;

3581
		if (n->nr_partial || atomic_long_read(&n->nr_slabs))
C
Christoph Lameter 已提交
3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720
			return 1;
	}
	return 0;
}

#define to_slab_attr(n) container_of(n, struct slab_attribute, attr)
#define to_slab(n) container_of(n, struct kmem_cache, kobj);

struct slab_attribute {
	struct attribute attr;
	ssize_t (*show)(struct kmem_cache *s, char *buf);
	ssize_t (*store)(struct kmem_cache *s, const char *x, size_t count);
};

#define SLAB_ATTR_RO(_name) \
	static struct slab_attribute _name##_attr = __ATTR_RO(_name)

#define SLAB_ATTR(_name) \
	static struct slab_attribute _name##_attr =  \
	__ATTR(_name, 0644, _name##_show, _name##_store)

static ssize_t slab_size_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", s->size);
}
SLAB_ATTR_RO(slab_size);

static ssize_t align_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", s->align);
}
SLAB_ATTR_RO(align);

static ssize_t object_size_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", s->objsize);
}
SLAB_ATTR_RO(object_size);

static ssize_t objs_per_slab_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", s->objects);
}
SLAB_ATTR_RO(objs_per_slab);

static ssize_t order_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", s->order);
}
SLAB_ATTR_RO(order);

static ssize_t ctor_show(struct kmem_cache *s, char *buf)
{
	if (s->ctor) {
		int n = sprint_symbol(buf, (unsigned long)s->ctor);

		return n + sprintf(buf + n, "\n");
	}
	return 0;
}
SLAB_ATTR_RO(ctor);

static ssize_t aliases_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", s->refcount - 1);
}
SLAB_ATTR_RO(aliases);

static ssize_t slabs_show(struct kmem_cache *s, char *buf)
{
	return slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU);
}
SLAB_ATTR_RO(slabs);

static ssize_t partial_show(struct kmem_cache *s, char *buf)
{
	return slab_objects(s, buf, SO_PARTIAL);
}
SLAB_ATTR_RO(partial);

static ssize_t cpu_slabs_show(struct kmem_cache *s, char *buf)
{
	return slab_objects(s, buf, SO_CPU);
}
SLAB_ATTR_RO(cpu_slabs);

static ssize_t objects_show(struct kmem_cache *s, char *buf)
{
	return slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU|SO_OBJECTS);
}
SLAB_ATTR_RO(objects);

static ssize_t sanity_checks_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_DEBUG_FREE));
}

static ssize_t sanity_checks_store(struct kmem_cache *s,
				const char *buf, size_t length)
{
	s->flags &= ~SLAB_DEBUG_FREE;
	if (buf[0] == '1')
		s->flags |= SLAB_DEBUG_FREE;
	return length;
}
SLAB_ATTR(sanity_checks);

static ssize_t trace_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_TRACE));
}

static ssize_t trace_store(struct kmem_cache *s, const char *buf,
							size_t length)
{
	s->flags &= ~SLAB_TRACE;
	if (buf[0] == '1')
		s->flags |= SLAB_TRACE;
	return length;
}
SLAB_ATTR(trace);

static ssize_t reclaim_account_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_RECLAIM_ACCOUNT));
}

static ssize_t reclaim_account_store(struct kmem_cache *s,
				const char *buf, size_t length)
{
	s->flags &= ~SLAB_RECLAIM_ACCOUNT;
	if (buf[0] == '1')
		s->flags |= SLAB_RECLAIM_ACCOUNT;
	return length;
}
SLAB_ATTR(reclaim_account);

static ssize_t hwcache_align_show(struct kmem_cache *s, char *buf)
{
3721
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_HWCACHE_ALIGN));
C
Christoph Lameter 已提交
3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795
}
SLAB_ATTR_RO(hwcache_align);

#ifdef CONFIG_ZONE_DMA
static ssize_t cache_dma_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_CACHE_DMA));
}
SLAB_ATTR_RO(cache_dma);
#endif

static ssize_t destroy_by_rcu_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_DESTROY_BY_RCU));
}
SLAB_ATTR_RO(destroy_by_rcu);

static ssize_t red_zone_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_RED_ZONE));
}

static ssize_t red_zone_store(struct kmem_cache *s,
				const char *buf, size_t length)
{
	if (any_slab_objects(s))
		return -EBUSY;

	s->flags &= ~SLAB_RED_ZONE;
	if (buf[0] == '1')
		s->flags |= SLAB_RED_ZONE;
	calculate_sizes(s);
	return length;
}
SLAB_ATTR(red_zone);

static ssize_t poison_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_POISON));
}

static ssize_t poison_store(struct kmem_cache *s,
				const char *buf, size_t length)
{
	if (any_slab_objects(s))
		return -EBUSY;

	s->flags &= ~SLAB_POISON;
	if (buf[0] == '1')
		s->flags |= SLAB_POISON;
	calculate_sizes(s);
	return length;
}
SLAB_ATTR(poison);

static ssize_t store_user_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", !!(s->flags & SLAB_STORE_USER));
}

static ssize_t store_user_store(struct kmem_cache *s,
				const char *buf, size_t length)
{
	if (any_slab_objects(s))
		return -EBUSY;

	s->flags &= ~SLAB_STORE_USER;
	if (buf[0] == '1')
		s->flags |= SLAB_STORE_USER;
	calculate_sizes(s);
	return length;
}
SLAB_ATTR(store_user);

3796 3797 3798 3799 3800 3801 3802 3803
static ssize_t validate_show(struct kmem_cache *s, char *buf)
{
	return 0;
}

static ssize_t validate_store(struct kmem_cache *s,
			const char *buf, size_t length)
{
3804 3805 3806 3807 3808 3809 3810 3811
	int ret = -EINVAL;

	if (buf[0] == '1') {
		ret = validate_slab_cache(s);
		if (ret >= 0)
			ret = length;
	}
	return ret;
3812 3813 3814
}
SLAB_ATTR(validate);

3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833
static ssize_t shrink_show(struct kmem_cache *s, char *buf)
{
	return 0;
}

static ssize_t shrink_store(struct kmem_cache *s,
			const char *buf, size_t length)
{
	if (buf[0] == '1') {
		int rc = kmem_cache_shrink(s);

		if (rc)
			return rc;
	} else
		return -EINVAL;
	return length;
}
SLAB_ATTR(shrink);

3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849
static ssize_t alloc_calls_show(struct kmem_cache *s, char *buf)
{
	if (!(s->flags & SLAB_STORE_USER))
		return -ENOSYS;
	return list_locations(s, buf, TRACK_ALLOC);
}
SLAB_ATTR_RO(alloc_calls);

static ssize_t free_calls_show(struct kmem_cache *s, char *buf)
{
	if (!(s->flags & SLAB_STORE_USER))
		return -ENOSYS;
	return list_locations(s, buf, TRACK_FREE);
}
SLAB_ATTR_RO(free_calls);

C
Christoph Lameter 已提交
3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887
#ifdef CONFIG_NUMA
static ssize_t defrag_ratio_show(struct kmem_cache *s, char *buf)
{
	return sprintf(buf, "%d\n", s->defrag_ratio / 10);
}

static ssize_t defrag_ratio_store(struct kmem_cache *s,
				const char *buf, size_t length)
{
	int n = simple_strtoul(buf, NULL, 10);

	if (n < 100)
		s->defrag_ratio = n * 10;
	return length;
}
SLAB_ATTR(defrag_ratio);
#endif

static struct attribute * slab_attrs[] = {
	&slab_size_attr.attr,
	&object_size_attr.attr,
	&objs_per_slab_attr.attr,
	&order_attr.attr,
	&objects_attr.attr,
	&slabs_attr.attr,
	&partial_attr.attr,
	&cpu_slabs_attr.attr,
	&ctor_attr.attr,
	&aliases_attr.attr,
	&align_attr.attr,
	&sanity_checks_attr.attr,
	&trace_attr.attr,
	&hwcache_align_attr.attr,
	&reclaim_account_attr.attr,
	&destroy_by_rcu_attr.attr,
	&red_zone_attr.attr,
	&poison_attr.attr,
	&store_user_attr.attr,
3888
	&validate_attr.attr,
3889
	&shrink_attr.attr,
3890 3891
	&alloc_calls_attr.attr,
	&free_calls_attr.attr,
C
Christoph Lameter 已提交
3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964
#ifdef CONFIG_ZONE_DMA
	&cache_dma_attr.attr,
#endif
#ifdef CONFIG_NUMA
	&defrag_ratio_attr.attr,
#endif
	NULL
};

static struct attribute_group slab_attr_group = {
	.attrs = slab_attrs,
};

static ssize_t slab_attr_show(struct kobject *kobj,
				struct attribute *attr,
				char *buf)
{
	struct slab_attribute *attribute;
	struct kmem_cache *s;
	int err;

	attribute = to_slab_attr(attr);
	s = to_slab(kobj);

	if (!attribute->show)
		return -EIO;

	err = attribute->show(s, buf);

	return err;
}

static ssize_t slab_attr_store(struct kobject *kobj,
				struct attribute *attr,
				const char *buf, size_t len)
{
	struct slab_attribute *attribute;
	struct kmem_cache *s;
	int err;

	attribute = to_slab_attr(attr);
	s = to_slab(kobj);

	if (!attribute->store)
		return -EIO;

	err = attribute->store(s, buf, len);

	return err;
}

static struct sysfs_ops slab_sysfs_ops = {
	.show = slab_attr_show,
	.store = slab_attr_store,
};

static struct kobj_type slab_ktype = {
	.sysfs_ops = &slab_sysfs_ops,
};

static int uevent_filter(struct kset *kset, struct kobject *kobj)
{
	struct kobj_type *ktype = get_ktype(kobj);

	if (ktype == &slab_ktype)
		return 1;
	return 0;
}

static struct kset_uevent_ops slab_uevent_ops = {
	.filter = uevent_filter,
};

3965
static struct kset *slab_kset;
C
Christoph Lameter 已提交
3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017

#define ID_STR_LENGTH 64

/* Create a unique string id for a slab cache:
 * format
 * :[flags-]size:[memory address of kmemcache]
 */
static char *create_unique_id(struct kmem_cache *s)
{
	char *name = kmalloc(ID_STR_LENGTH, GFP_KERNEL);
	char *p = name;

	BUG_ON(!name);

	*p++ = ':';
	/*
	 * First flags affecting slabcache operations. We will only
	 * get here for aliasable slabs so we do not need to support
	 * too many flags. The flags here must cover all flags that
	 * are matched during merging to guarantee that the id is
	 * unique.
	 */
	if (s->flags & SLAB_CACHE_DMA)
		*p++ = 'd';
	if (s->flags & SLAB_RECLAIM_ACCOUNT)
		*p++ = 'a';
	if (s->flags & SLAB_DEBUG_FREE)
		*p++ = 'F';
	if (p != name + 1)
		*p++ = '-';
	p += sprintf(p, "%07d", s->size);
	BUG_ON(p > name + ID_STR_LENGTH - 1);
	return name;
}

static int sysfs_slab_add(struct kmem_cache *s)
{
	int err;
	const char *name;
	int unmergeable;

	if (slab_state < SYSFS)
		/* Defer until later */
		return 0;

	unmergeable = slab_unmergeable(s);
	if (unmergeable) {
		/*
		 * Slabcache can never be merged so we can use the name proper.
		 * This is typically the case for debug situations. In that
		 * case we can catch duplicate names easily.
		 */
4018
		sysfs_remove_link(&slab_kset->kobj, s->name);
C
Christoph Lameter 已提交
4019 4020 4021 4022 4023 4024 4025 4026 4027 4028
		name = s->name;
	} else {
		/*
		 * Create a unique name for the slab as a target
		 * for the symlinks.
		 */
		name = create_unique_id(s);
	}

	kobject_set_name(&s->kobj, name);
4029
	s->kobj.kset = slab_kset;
4030
	s->kobj.ktype = &slab_ktype;
C
Christoph Lameter 已提交
4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063
	kobject_init(&s->kobj);
	err = kobject_add(&s->kobj);
	if (err)
		return err;

	err = sysfs_create_group(&s->kobj, &slab_attr_group);
	if (err)
		return err;
	kobject_uevent(&s->kobj, KOBJ_ADD);
	if (!unmergeable) {
		/* Setup first alias */
		sysfs_slab_alias(s, s->name);
		kfree(name);
	}
	return 0;
}

static void sysfs_slab_remove(struct kmem_cache *s)
{
	kobject_uevent(&s->kobj, KOBJ_REMOVE);
	kobject_del(&s->kobj);
}

/*
 * Need to buffer aliases during bootup until sysfs becomes
 * available lest we loose that information.
 */
struct saved_alias {
	struct kmem_cache *s;
	const char *name;
	struct saved_alias *next;
};

A
Adrian Bunk 已提交
4064
static struct saved_alias *alias_list;
C
Christoph Lameter 已提交
4065 4066 4067 4068 4069 4070 4071 4072 4073

static int sysfs_slab_alias(struct kmem_cache *s, const char *name)
{
	struct saved_alias *al;

	if (slab_state == SYSFS) {
		/*
		 * If we have a leftover link then remove it.
		 */
4074 4075
		sysfs_remove_link(&slab_kset->kobj, name);
		return sysfs_create_link(&slab_kset->kobj, &s->kobj, name);
C
Christoph Lameter 已提交
4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090
	}

	al = kmalloc(sizeof(struct saved_alias), GFP_KERNEL);
	if (!al)
		return -ENOMEM;

	al->s = s;
	al->name = name;
	al->next = alias_list;
	alias_list = al;
	return 0;
}

static int __init slab_sysfs_init(void)
{
4091
	struct kmem_cache *s;
C
Christoph Lameter 已提交
4092 4093
	int err;

4094 4095
	slab_kset = kset_create_and_add("slab", &slab_uevent_ops,
					&kernel_kset->kobj);
4096
	if (!slab_kset) {
C
Christoph Lameter 已提交
4097 4098 4099 4100
		printk(KERN_ERR "Cannot register slab subsystem.\n");
		return -ENOSYS;
	}

4101 4102
	slab_state = SYSFS;

4103
	list_for_each_entry(s, &slab_caches, list) {
4104
		err = sysfs_slab_add(s);
4105 4106 4107
		if (err)
			printk(KERN_ERR "SLUB: Unable to add boot slab %s"
						" to sysfs\n", s->name);
4108
	}
C
Christoph Lameter 已提交
4109 4110 4111 4112 4113 4114

	while (alias_list) {
		struct saved_alias *al = alias_list;

		alias_list = alias_list->next;
		err = sysfs_slab_alias(al->s, al->name);
4115 4116 4117
		if (err)
			printk(KERN_ERR "SLUB: Unable to add boot slab alias"
					" %s to sysfs\n", s->name);
C
Christoph Lameter 已提交
4118 4119 4120 4121 4122 4123 4124 4125 4126
		kfree(al);
	}

	resiliency_test();
	return 0;
}

__initcall(slab_sysfs_init);
#endif
P
Pekka J Enberg 已提交
4127 4128 4129 4130

/*
 * The /proc/slabinfo ABI
 */
4131 4132 4133 4134 4135 4136 4137 4138
#ifdef CONFIG_SLABINFO

ssize_t slabinfo_write(struct file *file, const char __user * buffer,
                       size_t count, loff_t *ppos)
{
	return -EINVAL;
}

P
Pekka J Enberg 已提交
4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211

static void print_slabinfo_header(struct seq_file *m)
{
	seq_puts(m, "slabinfo - version: 2.1\n");
	seq_puts(m, "# name            <active_objs> <num_objs> <objsize> "
		 "<objperslab> <pagesperslab>");
	seq_puts(m, " : tunables <limit> <batchcount> <sharedfactor>");
	seq_puts(m, " : slabdata <active_slabs> <num_slabs> <sharedavail>");
	seq_putc(m, '\n');
}

static void *s_start(struct seq_file *m, loff_t *pos)
{
	loff_t n = *pos;

	down_read(&slub_lock);
	if (!n)
		print_slabinfo_header(m);

	return seq_list_start(&slab_caches, *pos);
}

static void *s_next(struct seq_file *m, void *p, loff_t *pos)
{
	return seq_list_next(p, &slab_caches, pos);
}

static void s_stop(struct seq_file *m, void *p)
{
	up_read(&slub_lock);
}

static int s_show(struct seq_file *m, void *p)
{
	unsigned long nr_partials = 0;
	unsigned long nr_slabs = 0;
	unsigned long nr_inuse = 0;
	unsigned long nr_objs;
	struct kmem_cache *s;
	int node;

	s = list_entry(p, struct kmem_cache, list);

	for_each_online_node(node) {
		struct kmem_cache_node *n = get_node(s, node);

		if (!n)
			continue;

		nr_partials += n->nr_partial;
		nr_slabs += atomic_long_read(&n->nr_slabs);
		nr_inuse += count_partial(n);
	}

	nr_objs = nr_slabs * s->objects;
	nr_inuse += (nr_slabs - nr_partials) * s->objects;

	seq_printf(m, "%-17s %6lu %6lu %6u %4u %4d", s->name, nr_inuse,
		   nr_objs, s->size, s->objects, (1 << s->order));
	seq_printf(m, " : tunables %4u %4u %4u", 0, 0, 0);
	seq_printf(m, " : slabdata %6lu %6lu %6lu", nr_slabs, nr_slabs,
		   0UL);
	seq_putc(m, '\n');
	return 0;
}

const struct seq_operations slabinfo_op = {
	.start = s_start,
	.next = s_next,
	.stop = s_stop,
	.show = s_show,
};

4212
#endif /* CONFIG_SLABINFO */