slab.h 22.3 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0 */
L
Linus Torvalds 已提交
2
/*
3 4
 * Written by Mark Hemment, 1996 (markhe@nextd.demon.co.uk).
 *
C
Christoph Lameter 已提交
5
 * (C) SGI 2006, Christoph Lameter
6 7
 * 	Cleaned up and restructured to ease the addition of alternative
 * 	implementations of SLAB allocators.
8 9
 * (C) Linux Foundation 2008-2013
 *      Unified interface for all slab allocators
L
Linus Torvalds 已提交
10 11 12 13 14
 */

#ifndef _LINUX_SLAB_H
#define	_LINUX_SLAB_H

15
#include <linux/gfp.h>
16
#include <linux/overflow.h>
17
#include <linux/types.h>
G
Glauber Costa 已提交
18 19
#include <linux/workqueue.h>

L
Linus Torvalds 已提交
20

21 22
/*
 * Flags to pass to kmem_cache_create().
23
 * The ones marked DEBUG are only valid if CONFIG_DEBUG_SLAB is set.
L
Linus Torvalds 已提交
24
 */
25
/* DEBUG: Perform (expensive) checks on alloc/free */
26
#define SLAB_CONSISTENCY_CHECKS	((slab_flags_t __force)0x00000100U)
27
/* DEBUG: Red zone objs in a cache */
28
#define SLAB_RED_ZONE		((slab_flags_t __force)0x00000400U)
29
/* DEBUG: Poison objects */
30
#define SLAB_POISON		((slab_flags_t __force)0x00000800U)
31
/* Align objs on cache lines */
32
#define SLAB_HWCACHE_ALIGN	((slab_flags_t __force)0x00002000U)
33
/* Use GFP_DMA memory */
34
#define SLAB_CACHE_DMA		((slab_flags_t __force)0x00004000U)
35
/* DEBUG: Store the last owner for bug hunting */
36
#define SLAB_STORE_USER		((slab_flags_t __force)0x00010000U)
37
/* Panic if kmem_cache_create() fails */
38
#define SLAB_PANIC		((slab_flags_t __force)0x00040000U)
39
/*
40
 * SLAB_TYPESAFE_BY_RCU - **WARNING** READ THIS!
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
 *
 * This delays freeing the SLAB page by a grace period, it does _NOT_
 * delay object freeing. This means that if you do kmem_cache_free()
 * that memory location is free to be reused at any time. Thus it may
 * be possible to see another object there in the same RCU grace period.
 *
 * This feature only ensures the memory location backing the object
 * stays valid, the trick to using this is relying on an independent
 * object validation pass. Something like:
 *
 *  rcu_read_lock()
 * again:
 *  obj = lockless_lookup(key);
 *  if (obj) {
 *    if (!try_get_ref(obj)) // might fail for free objects
 *      goto again;
 *
 *    if (obj->key != key) { // not the object we expected
 *      put_ref(obj);
 *      goto again;
 *    }
 *  }
 *  rcu_read_unlock();
 *
65 66 67 68 69 70 71 72
 * This is useful if we need to approach a kernel structure obliquely,
 * from its address obtained without the usual locking. We can lock
 * the structure to stabilize it and check it's still at the given address,
 * only if we can be sure that the memory has not been meanwhile reused
 * for some other kind of object (which our subsystem's lock might corrupt).
 *
 * rcu_read_lock before reading the address, then rcu_read_unlock after
 * taking the spinlock within the structure expected at that address.
73 74
 *
 * Note that SLAB_TYPESAFE_BY_RCU was originally named SLAB_DESTROY_BY_RCU.
75
 */
76
/* Defer freeing slabs to RCU */
77
#define SLAB_TYPESAFE_BY_RCU	((slab_flags_t __force)0x00080000U)
78
/* Spread some memory over cpuset */
79
#define SLAB_MEM_SPREAD		((slab_flags_t __force)0x00100000U)
80
/* Trace allocations and frees */
81
#define SLAB_TRACE		((slab_flags_t __force)0x00200000U)
L
Linus Torvalds 已提交
82

83 84
/* Flag to prevent checks on free */
#ifdef CONFIG_DEBUG_OBJECTS
85
# define SLAB_DEBUG_OBJECTS	((slab_flags_t __force)0x00400000U)
86
#else
87
# define SLAB_DEBUG_OBJECTS	0
88 89
#endif

90
/* Avoid kmemleak tracing */
91
#define SLAB_NOLEAKTRACE	((slab_flags_t __force)0x00800000U)
92

93
/* Fault injection mark */
94
#ifdef CONFIG_FAILSLAB
95
# define SLAB_FAILSLAB		((slab_flags_t __force)0x02000000U)
96
#else
97
# define SLAB_FAILSLAB		0
98
#endif
99
/* Account to memcg */
100
#ifdef CONFIG_MEMCG_KMEM
101
# define SLAB_ACCOUNT		((slab_flags_t __force)0x04000000U)
V
Vladimir Davydov 已提交
102
#else
103
# define SLAB_ACCOUNT		0
V
Vladimir Davydov 已提交
104
#endif
V
Vegard Nossum 已提交
105

A
Alexander Potapenko 已提交
106
#ifdef CONFIG_KASAN
107
#define SLAB_KASAN		((slab_flags_t __force)0x08000000U)
A
Alexander Potapenko 已提交
108
#else
109
#define SLAB_KASAN		0
A
Alexander Potapenko 已提交
110 111
#endif

112
/* The following flags affect the page allocator grouping pages by mobility */
113
/* Objects are reclaimable */
114
#define SLAB_RECLAIM_ACCOUNT	((slab_flags_t __force)0x00020000U)
115
#define SLAB_TEMPORARY		SLAB_RECLAIM_ACCOUNT	/* Objects are short-lived */
116 117 118 119 120 121 122 123 124 125
/*
 * ZERO_SIZE_PTR will be returned for zero sized kmalloc requests.
 *
 * Dereferencing ZERO_SIZE_PTR will lead to a distinct access fault.
 *
 * ZERO_SIZE_PTR can be passed to kfree though in the same way that NULL can.
 * Both make kfree a no-op.
 */
#define ZERO_SIZE_PTR ((void *)16)

126
#define ZERO_OR_NULL_PTR(x) ((unsigned long)(x) <= \
127 128
				(unsigned long)ZERO_SIZE_PTR)

129
#include <linux/kasan.h>
130

131
struct mem_cgroup;
132 133 134 135
/*
 * struct kmem_cache related prototypes
 */
void __init kmem_cache_init(void);
136
bool slab_is_available(void);
L
Linus Torvalds 已提交
137

138 139
extern bool usercopy_fallback;

140 141
struct kmem_cache *kmem_cache_create(const char *name, unsigned int size,
			unsigned int align, slab_flags_t flags,
142 143
			void (*ctor)(void *));
struct kmem_cache *kmem_cache_create_usercopy(const char *name,
144 145
			unsigned int size, unsigned int align,
			slab_flags_t flags,
146
			unsigned int useroffset, unsigned int usersize,
147
			void (*ctor)(void *));
148 149
void kmem_cache_destroy(struct kmem_cache *);
int kmem_cache_shrink(struct kmem_cache *);
150 151 152 153

void memcg_create_kmem_cache(struct mem_cgroup *, struct kmem_cache *);
void memcg_deactivate_kmem_caches(struct mem_cgroup *);
void memcg_destroy_kmem_caches(struct mem_cgroup *);
154

155 156 157 158 159 160 161 162
/*
 * Please use this macro to create slab caches. Simply specify the
 * name of the structure and maybe some flags that are listed above.
 *
 * The alignment of the struct determines object alignment. If you
 * f.e. add ____cacheline_aligned_in_smp to the struct declaration
 * then the objects will be properly aligned in SMP configurations.
 */
163 164 165 166 167 168 169 170 171 172 173 174 175 176
#define KMEM_CACHE(__struct, __flags)					\
		kmem_cache_create(#__struct, sizeof(struct __struct),	\
			__alignof__(struct __struct), (__flags), NULL)

/*
 * To whitelist a single field for copying to/from usercopy, use this
 * macro instead for KMEM_CACHE() above.
 */
#define KMEM_CACHE_USERCOPY(__struct, __flags, __field)			\
		kmem_cache_create_usercopy(#__struct,			\
			sizeof(struct __struct),			\
			__alignof__(struct __struct), (__flags),	\
			offsetof(struct __struct, __field),		\
			sizeof_field(struct __struct, __field), NULL)
177

178 179 180 181 182 183 184 185 186
/*
 * Common kmalloc functions provided by all allocators
 */
void * __must_check __krealloc(const void *, size_t, gfp_t);
void * __must_check krealloc(const void *, size_t, gfp_t);
void kfree(const void *);
void kzfree(const void *);
size_t ksize(const void *);

K
Kees Cook 已提交
187
#ifdef CONFIG_HAVE_HARDENED_USERCOPY_ALLOCATOR
188 189
void __check_heap_object(const void *ptr, unsigned long n, struct page *page,
			bool to_user);
K
Kees Cook 已提交
190
#else
191 192
static inline void __check_heap_object(const void *ptr, unsigned long n,
				       struct page *page, bool to_user) { }
K
Kees Cook 已提交
193 194
#endif

195 196 197 198 199 200 201 202 203 204 205 206 207
/*
 * Some archs want to perform DMA into kmalloc caches and need a guaranteed
 * alignment larger than the alignment of a 64-bit integer.
 * Setting ARCH_KMALLOC_MINALIGN in arch headers allows that.
 */
#if defined(ARCH_DMA_MINALIGN) && ARCH_DMA_MINALIGN > 8
#define ARCH_KMALLOC_MINALIGN ARCH_DMA_MINALIGN
#define KMALLOC_MIN_SIZE ARCH_DMA_MINALIGN
#define KMALLOC_SHIFT_LOW ilog2(ARCH_DMA_MINALIGN)
#else
#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long)
#endif

208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
/*
 * Setting ARCH_SLAB_MINALIGN in arch headers allows a different alignment.
 * Intended for arches that get misalignment faults even for 64 bit integer
 * aligned buffers.
 */
#ifndef ARCH_SLAB_MINALIGN
#define ARCH_SLAB_MINALIGN __alignof__(unsigned long long)
#endif

/*
 * kmalloc and friends return ARCH_KMALLOC_MINALIGN aligned
 * pointers. kmem_cache_alloc and friends return ARCH_SLAB_MINALIGN
 * aligned pointers.
 */
#define __assume_kmalloc_alignment __assume_aligned(ARCH_KMALLOC_MINALIGN)
#define __assume_slab_alignment __assume_aligned(ARCH_SLAB_MINALIGN)
#define __assume_page_alignment __assume_aligned(PAGE_SIZE)

226
/*
227 228 229 230 231 232
 * Kmalloc array related definitions
 */

#ifdef CONFIG_SLAB
/*
 * The largest kmalloc size supported by the SLAB allocators is
233 234 235 236 237 238 239
 * 32 megabyte (2^25) or the maximum allocatable page order if that is
 * less than 32 MB.
 *
 * WARNING: Its not easy to increase this value since the allocators have
 * to do various tricks to work around compiler limitations in order to
 * ensure proper constant folding.
 */
240 241
#define KMALLOC_SHIFT_HIGH	((MAX_ORDER + PAGE_SHIFT - 1) <= 25 ? \
				(MAX_ORDER + PAGE_SHIFT - 1) : 25)
242
#define KMALLOC_SHIFT_MAX	KMALLOC_SHIFT_HIGH
243
#ifndef KMALLOC_SHIFT_LOW
244
#define KMALLOC_SHIFT_LOW	5
245
#endif
246 247 248
#endif

#ifdef CONFIG_SLUB
249
/*
250 251
 * SLUB directly allocates requests fitting in to an order-1 page
 * (PAGE_SIZE*2).  Larger requests are passed to the page allocator.
252 253
 */
#define KMALLOC_SHIFT_HIGH	(PAGE_SHIFT + 1)
254
#define KMALLOC_SHIFT_MAX	(MAX_ORDER + PAGE_SHIFT - 1)
255
#ifndef KMALLOC_SHIFT_LOW
256 257
#define KMALLOC_SHIFT_LOW	3
#endif
258
#endif
259

260 261
#ifdef CONFIG_SLOB
/*
262
 * SLOB passes all requests larger than one page to the page allocator.
263 264 265 266
 * No kmalloc array is necessary since objects of different sizes can
 * be allocated from the same page.
 */
#define KMALLOC_SHIFT_HIGH	PAGE_SHIFT
267
#define KMALLOC_SHIFT_MAX	(MAX_ORDER + PAGE_SHIFT - 1)
268 269 270 271 272
#ifndef KMALLOC_SHIFT_LOW
#define KMALLOC_SHIFT_LOW	3
#endif
#endif

273 274 275 276 277 278
/* Maximum allocatable size */
#define KMALLOC_MAX_SIZE	(1UL << KMALLOC_SHIFT_MAX)
/* Maximum size for which we actually use a slab cache */
#define KMALLOC_MAX_CACHE_SIZE	(1UL << KMALLOC_SHIFT_HIGH)
/* Maximum order allocatable via the slab allocagtor */
#define KMALLOC_MAX_ORDER	(KMALLOC_SHIFT_MAX - PAGE_SHIFT)
279

280 281 282
/*
 * Kmalloc subsystem.
 */
283
#ifndef KMALLOC_MIN_SIZE
284
#define KMALLOC_MIN_SIZE (1 << KMALLOC_SHIFT_LOW)
285 286
#endif

J
Joonsoo Kim 已提交
287 288 289 290 291 292 293 294 295 296 297
/*
 * This restriction comes from byte sized index implementation.
 * Page size is normally 2^12 bytes and, in this case, if we want to use
 * byte sized index which can represent 2^8 entries, the size of the object
 * should be equal or greater to 2^12 / 2^8 = 2^4 = 16.
 * If minimum size of kmalloc is less than 16, we use it as minimum object
 * size and give up to use byte sized index.
 */
#define SLAB_OBJ_MIN_SIZE      (KMALLOC_MIN_SIZE < 16 ? \
                               (KMALLOC_MIN_SIZE) : 16)

298 299 300 301 302 303 304 305
enum kmalloc_cache_type {
	KMALLOC_NORMAL = 0,
#ifdef CONFIG_ZONE_DMA
	KMALLOC_DMA,
#endif
	NR_KMALLOC_TYPES
};

306
#ifndef CONFIG_SLOB
307 308 309 310 311 312 313
extern struct kmem_cache *
kmalloc_caches[NR_KMALLOC_TYPES][KMALLOC_SHIFT_HIGH + 1];

static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags)
{
	int is_dma = 0;

314
#ifdef CONFIG_ZONE_DMA
315
	is_dma = !!(flags & __GFP_DMA);
316 317
#endif

318 319 320
	return is_dma;
}

321 322 323 324 325
/*
 * Figure out which kmalloc slab an allocation of a certain size
 * belongs to.
 * 0 = zero alloc
 * 1 =  65 .. 96 bytes
326 327
 * 2 = 129 .. 192 bytes
 * n = 2^(n-1)+1 .. 2^n
328
 */
329
static __always_inline unsigned int kmalloc_index(size_t size)
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
{
	if (!size)
		return 0;

	if (size <= KMALLOC_MIN_SIZE)
		return KMALLOC_SHIFT_LOW;

	if (KMALLOC_MIN_SIZE <= 32 && size > 64 && size <= 96)
		return 1;
	if (KMALLOC_MIN_SIZE <= 64 && size > 128 && size <= 192)
		return 2;
	if (size <=          8) return 3;
	if (size <=         16) return 4;
	if (size <=         32) return 5;
	if (size <=         64) return 6;
	if (size <=        128) return 7;
	if (size <=        256) return 8;
	if (size <=        512) return 9;
	if (size <=       1024) return 10;
	if (size <=   2 * 1024) return 11;
	if (size <=   4 * 1024) return 12;
	if (size <=   8 * 1024) return 13;
	if (size <=  16 * 1024) return 14;
	if (size <=  32 * 1024) return 15;
	if (size <=  64 * 1024) return 16;
	if (size <= 128 * 1024) return 17;
	if (size <= 256 * 1024) return 18;
	if (size <= 512 * 1024) return 19;
	if (size <= 1024 * 1024) return 20;
	if (size <=  2 * 1024 * 1024) return 21;
	if (size <=  4 * 1024 * 1024) return 22;
	if (size <=  8 * 1024 * 1024) return 23;
	if (size <=  16 * 1024 * 1024) return 24;
	if (size <=  32 * 1024 * 1024) return 25;
	if (size <=  64 * 1024 * 1024) return 26;
	BUG();

	/* Will never be reached. Needed because the compiler may complain */
	return -1;
}
370
#endif /* !CONFIG_SLOB */
371

372 373
void *__kmalloc(size_t size, gfp_t flags) __assume_kmalloc_alignment __malloc;
void *kmem_cache_alloc(struct kmem_cache *, gfp_t flags) __assume_slab_alignment __malloc;
374
void kmem_cache_free(struct kmem_cache *, void *);
375

376
/*
J
Jesper Dangaard Brouer 已提交
377
 * Bulk allocation and freeing operations. These are accelerated in an
378 379 380 381 382 383
 * allocator specific way to avoid taking locks repeatedly or building
 * metadata structures unnecessarily.
 *
 * Note that interrupts must be enabled when calling these functions.
 */
void kmem_cache_free_bulk(struct kmem_cache *, size_t, void **);
384
int kmem_cache_alloc_bulk(struct kmem_cache *, gfp_t, size_t, void **);
385

386 387 388 389 390 391 392 393 394
/*
 * Caller must not use kfree_bulk() on memory not originally allocated
 * by kmalloc(), because the SLOB allocator cannot handle this.
 */
static __always_inline void kfree_bulk(size_t size, void **p)
{
	kmem_cache_free_bulk(NULL, size, p);
}

395
#ifdef CONFIG_NUMA
396 397
void *__kmalloc_node(size_t size, gfp_t flags, int node) __assume_kmalloc_alignment __malloc;
void *kmem_cache_alloc_node(struct kmem_cache *, gfp_t flags, int node) __assume_slab_alignment __malloc;
398 399 400 401 402 403 404 405 406 407 408 409 410
#else
static __always_inline void *__kmalloc_node(size_t size, gfp_t flags, int node)
{
	return __kmalloc(size, flags);
}

static __always_inline void *kmem_cache_alloc_node(struct kmem_cache *s, gfp_t flags, int node)
{
	return kmem_cache_alloc(s, flags);
}
#endif

#ifdef CONFIG_TRACING
411
extern void *kmem_cache_alloc_trace(struct kmem_cache *, gfp_t, size_t) __assume_slab_alignment __malloc;
412 413 414 415

#ifdef CONFIG_NUMA
extern void *kmem_cache_alloc_node_trace(struct kmem_cache *s,
					   gfp_t gfpflags,
416
					   int node, size_t size) __assume_slab_alignment __malloc;
417 418 419 420 421 422 423 424 425 426 427 428 429 430
#else
static __always_inline void *
kmem_cache_alloc_node_trace(struct kmem_cache *s,
			      gfp_t gfpflags,
			      int node, size_t size)
{
	return kmem_cache_alloc_trace(s, gfpflags, size);
}
#endif /* CONFIG_NUMA */

#else /* CONFIG_TRACING */
static __always_inline void *kmem_cache_alloc_trace(struct kmem_cache *s,
		gfp_t flags, size_t size)
{
431 432
	void *ret = kmem_cache_alloc(s, flags);

433
	kasan_kmalloc(s, ret, size, flags);
434
	return ret;
435 436 437 438 439 440 441
}

static __always_inline void *
kmem_cache_alloc_node_trace(struct kmem_cache *s,
			      gfp_t gfpflags,
			      int node, size_t size)
{
442 443
	void *ret = kmem_cache_alloc_node(s, gfpflags, node);

444
	kasan_kmalloc(s, ret, size, gfpflags);
445
	return ret;
446 447 448
}
#endif /* CONFIG_TRACING */

449
extern void *kmalloc_order(size_t size, gfp_t flags, unsigned int order) __assume_page_alignment __malloc;
450 451

#ifdef CONFIG_TRACING
452
extern void *kmalloc_order_trace(size_t size, gfp_t flags, unsigned int order) __assume_page_alignment __malloc;
453 454 455 456 457 458
#else
static __always_inline void *
kmalloc_order_trace(size_t size, gfp_t flags, unsigned int order)
{
	return kmalloc_order(size, flags, order);
}
459 460
#endif

461 462 463 464 465 466 467 468 469
static __always_inline void *kmalloc_large(size_t size, gfp_t flags)
{
	unsigned int order = get_order(size);
	return kmalloc_order_trace(size, flags, order);
}

/**
 * kmalloc - allocate memory
 * @size: how many bytes of memory are required.
470
 * @flags: the type of memory to allocate.
471 472 473
 *
 * kmalloc is the normal method of allocating memory
 * for objects smaller than page size in the kernel.
474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491
 *
 * The @flags argument may be one of:
 *
 * %GFP_USER - Allocate memory on behalf of user.  May sleep.
 *
 * %GFP_KERNEL - Allocate normal kernel ram.  May sleep.
 *
 * %GFP_ATOMIC - Allocation will not sleep.  May use emergency pools.
 *   For example, use this inside interrupt handlers.
 *
 * %GFP_HIGHUSER - Allocate pages from high memory.
 *
 * %GFP_NOIO - Do not do any I/O at all while trying to get memory.
 *
 * %GFP_NOFS - Do not make any fs calls while trying to get memory.
 *
 * %GFP_NOWAIT - Allocation will not sleep.
 *
492
 * %__GFP_THISNODE - Allocate node-local memory only.
493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510
 *
 * %GFP_DMA - Allocation suitable for DMA.
 *   Should only be used for kmalloc() caches. Otherwise, use a
 *   slab created with SLAB_DMA.
 *
 * Also it is possible to set different flags by OR'ing
 * in one or more of the following additional @flags:
 *
 * %__GFP_HIGH - This allocation has high priority and may use emergency pools.
 *
 * %__GFP_NOFAIL - Indicate that this allocation is in no way allowed to fail
 *   (think twice before using).
 *
 * %__GFP_NORETRY - If memory is not immediately available,
 *   then give up at once.
 *
 * %__GFP_NOWARN - If allocation fails, don't issue any warnings.
 *
511 512
 * %__GFP_RETRY_MAYFAIL - Try really hard to succeed the allocation but fail
 *   eventually.
513 514 515 516
 *
 * There are other flags available as well, but these are not intended
 * for general use, and so are not documented here. For a full list of
 * potential flags, always refer to linux/gfp.h.
517 518 519 520
 */
static __always_inline void *kmalloc(size_t size, gfp_t flags)
{
	if (__builtin_constant_p(size)) {
521 522 523
#ifndef CONFIG_SLOB
		unsigned int index;
#endif
524 525 526
		if (size > KMALLOC_MAX_CACHE_SIZE)
			return kmalloc_large(size, flags);
#ifndef CONFIG_SLOB
527
		index = kmalloc_index(size);
528

529 530
		if (!index)
			return ZERO_SIZE_PTR;
531

532 533 534
		return kmem_cache_alloc_trace(
				kmalloc_caches[kmalloc_type(flags)][index],
				flags, size);
535 536 537 538 539
#endif
	}
	return __kmalloc(size, flags);
}

540 541 542 543 544
/*
 * Determine size used for the nth kmalloc cache.
 * return size or 0 if a kmalloc cache for that
 * size does not exist
 */
545
static __always_inline unsigned int kmalloc_size(unsigned int n)
546
{
547
#ifndef CONFIG_SLOB
548
	if (n > 2)
549
		return 1U << n;
550 551 552 553 554 555

	if (n == 1 && KMALLOC_MIN_SIZE <= 32)
		return 96;

	if (n == 2 && KMALLOC_MIN_SIZE <= 64)
		return 192;
556
#endif
557 558 559
	return 0;
}

560 561 562 563
static __always_inline void *kmalloc_node(size_t size, gfp_t flags, int node)
{
#ifndef CONFIG_SLOB
	if (__builtin_constant_p(size) &&
564
		size <= KMALLOC_MAX_CACHE_SIZE) {
565
		unsigned int i = kmalloc_index(size);
566 567 568 569

		if (!i)
			return ZERO_SIZE_PTR;

570 571
		return kmem_cache_alloc_node_trace(
				kmalloc_caches[kmalloc_type(flags)][i],
572 573 574 575 576 577
						flags, node, size);
	}
#endif
	return __kmalloc_node(size, flags, node);
}

578 579 580 581 582
struct memcg_cache_array {
	struct rcu_head rcu;
	struct kmem_cache *entries[0];
};

G
Glauber Costa 已提交
583 584 585 586
/*
 * This is the main placeholder for memcg-related information in kmem caches.
 * Both the root cache and the child caches will have it. For the root cache,
 * this will hold a dynamically allocated array large enough to hold
587 588 589
 * information about the currently limited memcgs in the system. To allow the
 * array to be accessed without taking any locks, on relocation we free the old
 * version only after a grace period.
G
Glauber Costa 已提交
590
 *
T
Tejun Heo 已提交
591
 * Root and child caches hold different metadata.
G
Glauber Costa 已提交
592
 *
T
Tejun Heo 已提交
593 594
 * @root_cache:	Common to root and child caches.  NULL for root, pointer to
 *		the root cache for children.
595
 *
T
Tejun Heo 已提交
596 597 598 599 600 601
 * The following fields are specific to root caches.
 *
 * @memcg_caches: kmemcg ID indexed table of child caches.  This table is
 *		used to index child cachces during allocation and cleared
 *		early during shutdown.
 *
602 603
 * @root_caches_node: List node for slab_root_caches list.
 *
T
Tejun Heo 已提交
604 605 606 607 608 609 610 611 612
 * @children:	List of all child caches.  While the child caches are also
 *		reachable through @memcg_caches, a child cache remains on
 *		this list until it is actually destroyed.
 *
 * The following fields are specific to child caches.
 *
 * @memcg:	Pointer to the memcg this cache belongs to.
 *
 * @children_node: List node for @root_cache->children list.
613 614
 *
 * @kmem_caches_node: List node for @memcg->kmem_caches list.
G
Glauber Costa 已提交
615 616
 */
struct memcg_cache_params {
T
Tejun Heo 已提交
617
	struct kmem_cache *root_cache;
G
Glauber Costa 已提交
618
	union {
T
Tejun Heo 已提交
619 620
		struct {
			struct memcg_cache_array __rcu *memcg_caches;
621
			struct list_head __root_caches_node;
T
Tejun Heo 已提交
622
			struct list_head children;
623
			bool dying;
T
Tejun Heo 已提交
624
		};
625 626
		struct {
			struct mem_cgroup *memcg;
T
Tejun Heo 已提交
627
			struct list_head children_node;
628
			struct list_head kmem_caches_node;
629 630 631 632 633 634

			void (*deact_fn)(struct kmem_cache *);
			union {
				struct rcu_head deact_rcu_head;
				struct work_struct deact_work;
			};
635
		};
G
Glauber Costa 已提交
636 637 638
	};
};

639 640
int memcg_update_all_caches(int num_memcgs);

641 642 643 644 645
/**
 * kmalloc_array - allocate memory for an array.
 * @n: number of elements.
 * @size: element size.
 * @flags: the type of memory to allocate (see kmalloc).
646
 */
X
Xi Wang 已提交
647
static inline void *kmalloc_array(size_t n, size_t size, gfp_t flags)
L
Linus Torvalds 已提交
648
{
649 650 651
	size_t bytes;

	if (unlikely(check_mul_overflow(n, size, &bytes)))
P
Paul Mundt 已提交
652
		return NULL;
653
	if (__builtin_constant_p(n) && __builtin_constant_p(size))
654 655
		return kmalloc(bytes, flags);
	return __kmalloc(bytes, flags);
X
Xi Wang 已提交
656 657 658 659 660 661 662 663 664 665 666
}

/**
 * kcalloc - allocate memory for an array. The memory is set to zero.
 * @n: number of elements.
 * @size: element size.
 * @flags: the type of memory to allocate (see kmalloc).
 */
static inline void *kcalloc(size_t n, size_t size, gfp_t flags)
{
	return kmalloc_array(n, size, flags | __GFP_ZERO);
L
Linus Torvalds 已提交
667 668
}

669 670 671 672 673 674 675 676
/*
 * kmalloc_track_caller is a special version of kmalloc that records the
 * calling function of the routine calling it for slab leak tracking instead
 * of just the calling function (confusing, eh?).
 * It's useful when the call to kmalloc comes from a widely-used standard
 * allocator where we care about the real place the memory allocation
 * request comes from.
 */
677
extern void *__kmalloc_track_caller(size_t, gfp_t, unsigned long);
678
#define kmalloc_track_caller(size, flags) \
679
	__kmalloc_track_caller(size, flags, _RET_IP_)
L
Linus Torvalds 已提交
680

681 682 683
static inline void *kmalloc_array_node(size_t n, size_t size, gfp_t flags,
				       int node)
{
684 685 686
	size_t bytes;

	if (unlikely(check_mul_overflow(n, size, &bytes)))
687 688
		return NULL;
	if (__builtin_constant_p(n) && __builtin_constant_p(size))
689 690
		return kmalloc_node(bytes, flags, node);
	return __kmalloc_node(bytes, flags, node);
691 692 693 694 695 696 697 698
}

static inline void *kcalloc_node(size_t n, size_t size, gfp_t flags, int node)
{
	return kmalloc_array_node(n, size, flags | __GFP_ZERO, node);
}


699
#ifdef CONFIG_NUMA
700
extern void *__kmalloc_node_track_caller(size_t, gfp_t, int, unsigned long);
701 702
#define kmalloc_node_track_caller(size, flags, node) \
	__kmalloc_node_track_caller(size, flags, node, \
703
			_RET_IP_)
704

705 706 707 708
#else /* CONFIG_NUMA */

#define kmalloc_node_track_caller(size, flags, node) \
	kmalloc_track_caller(size, flags)
709

P
Pascal Terjan 已提交
710
#endif /* CONFIG_NUMA */
711

712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729
/*
 * Shortcuts
 */
static inline void *kmem_cache_zalloc(struct kmem_cache *k, gfp_t flags)
{
	return kmem_cache_alloc(k, flags | __GFP_ZERO);
}

/**
 * kzalloc - allocate memory. The memory is set to zero.
 * @size: how many bytes of memory are required.
 * @flags: the type of memory to allocate (see kmalloc).
 */
static inline void *kzalloc(size_t size, gfp_t flags)
{
	return kmalloc(size, flags | __GFP_ZERO);
}

J
Jeff Layton 已提交
730 731 732 733 734 735 736 737 738 739 740
/**
 * kzalloc_node - allocate zeroed memory from a particular memory node.
 * @size: how many bytes of memory are required.
 * @flags: the type of memory to allocate (see kmalloc).
 * @node: memory node from which to allocate
 */
static inline void *kzalloc_node(size_t size, gfp_t flags, int node)
{
	return kmalloc_node(size, flags | __GFP_ZERO, node);
}

741
unsigned int kmem_cache_size(struct kmem_cache *s);
742 743
void __init kmem_cache_init_late(void);

744 745 746 747 748 749 750 751
#if defined(CONFIG_SMP) && defined(CONFIG_SLAB)
int slab_prepare_cpu(unsigned int cpu);
int slab_dead_cpu(unsigned int cpu);
#else
#define slab_prepare_cpu	NULL
#define slab_dead_cpu		NULL
#endif

L
Linus Torvalds 已提交
752
#endif	/* _LINUX_SLAB_H */