slab.h 20.9 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
#include <linux/workqueue.h>
19
#include <linux/percpu-refcount.h>
G
Glauber Costa 已提交
20

L
Linus Torvalds 已提交
21

22 23
/*
 * Flags to pass to kmem_cache_create().
24
 * The ones marked DEBUG are only valid if CONFIG_DEBUG_SLAB is set.
L
Linus Torvalds 已提交
25
 */
26
/* DEBUG: Perform (expensive) checks on alloc/free */
27
#define SLAB_CONSISTENCY_CHECKS	((slab_flags_t __force)0x00000100U)
28
/* DEBUG: Red zone objs in a cache */
29
#define SLAB_RED_ZONE		((slab_flags_t __force)0x00000400U)
30
/* DEBUG: Poison objects */
31
#define SLAB_POISON		((slab_flags_t __force)0x00000800U)
32
/* Align objs on cache lines */
33
#define SLAB_HWCACHE_ALIGN	((slab_flags_t __force)0x00002000U)
34
/* Use GFP_DMA memory */
35
#define SLAB_CACHE_DMA		((slab_flags_t __force)0x00004000U)
36 37
/* Use GFP_DMA32 memory */
#define SLAB_CACHE_DMA32	((slab_flags_t __force)0x00008000U)
38
/* DEBUG: Store the last owner for bug hunting */
39
#define SLAB_STORE_USER		((slab_flags_t __force)0x00010000U)
40
/* Panic if kmem_cache_create() fails */
41
#define SLAB_PANIC		((slab_flags_t __force)0x00040000U)
42
/*
43
 * SLAB_TYPESAFE_BY_RCU - **WARNING** READ THIS!
44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
 *
 * 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();
 *
68 69 70 71 72 73 74 75
 * 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.
76 77
 *
 * Note that SLAB_TYPESAFE_BY_RCU was originally named SLAB_DESTROY_BY_RCU.
78
 */
79
/* Defer freeing slabs to RCU */
80
#define SLAB_TYPESAFE_BY_RCU	((slab_flags_t __force)0x00080000U)
81
/* Spread some memory over cpuset */
82
#define SLAB_MEM_SPREAD		((slab_flags_t __force)0x00100000U)
83
/* Trace allocations and frees */
84
#define SLAB_TRACE		((slab_flags_t __force)0x00200000U)
L
Linus Torvalds 已提交
85

86 87
/* Flag to prevent checks on free */
#ifdef CONFIG_DEBUG_OBJECTS
88
# define SLAB_DEBUG_OBJECTS	((slab_flags_t __force)0x00400000U)
89
#else
90
# define SLAB_DEBUG_OBJECTS	0
91 92
#endif

93
/* Avoid kmemleak tracing */
94
#define SLAB_NOLEAKTRACE	((slab_flags_t __force)0x00800000U)
95

96
/* Fault injection mark */
97
#ifdef CONFIG_FAILSLAB
98
# define SLAB_FAILSLAB		((slab_flags_t __force)0x02000000U)
99
#else
100
# define SLAB_FAILSLAB		0
101
#endif
102
/* Account to memcg */
103
#ifdef CONFIG_MEMCG_KMEM
104
# define SLAB_ACCOUNT		((slab_flags_t __force)0x04000000U)
V
Vladimir Davydov 已提交
105
#else
106
# define SLAB_ACCOUNT		0
V
Vladimir Davydov 已提交
107
#endif
V
Vegard Nossum 已提交
108

A
Alexander Potapenko 已提交
109
#ifdef CONFIG_KASAN
110
#define SLAB_KASAN		((slab_flags_t __force)0x08000000U)
A
Alexander Potapenko 已提交
111
#else
112
#define SLAB_KASAN		0
A
Alexander Potapenko 已提交
113 114
#endif

115
/* The following flags affect the page allocator grouping pages by mobility */
116
/* Objects are reclaimable */
117
#define SLAB_RECLAIM_ACCOUNT	((slab_flags_t __force)0x00020000U)
118
#define SLAB_TEMPORARY		SLAB_RECLAIM_ACCOUNT	/* Objects are short-lived */
119 120 121 122

/* Slab deactivation flag */
#define SLAB_DEACTIVATED	((slab_flags_t __force)0x10000000U)

123 124 125 126 127 128 129 130 131 132
/*
 * 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)

133
#define ZERO_OR_NULL_PTR(x) ((unsigned long)(x) <= \
134 135
				(unsigned long)ZERO_SIZE_PTR)

136
#include <linux/kasan.h>
137

138
struct mem_cgroup;
139 140 141 142
/*
 * struct kmem_cache related prototypes
 */
void __init kmem_cache_init(void);
143
bool slab_is_available(void);
L
Linus Torvalds 已提交
144

145 146
extern bool usercopy_fallback;

147 148
struct kmem_cache *kmem_cache_create(const char *name, unsigned int size,
			unsigned int align, slab_flags_t flags,
149 150
			void (*ctor)(void *));
struct kmem_cache *kmem_cache_create_usercopy(const char *name,
151 152
			unsigned int size, unsigned int align,
			slab_flags_t flags,
153
			unsigned int useroffset, unsigned int usersize,
154
			void (*ctor)(void *));
155 156
void kmem_cache_destroy(struct kmem_cache *);
int kmem_cache_shrink(struct kmem_cache *);
157 158

void memcg_create_kmem_cache(struct mem_cgroup *, struct kmem_cache *);
159
void memcg_deactivate_kmem_caches(struct mem_cgroup *, struct mem_cgroup *);
160

161 162 163 164 165 166 167 168
/*
 * 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.
 */
169 170 171 172 173 174 175 176 177 178 179 180 181 182
#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)
183

184 185 186 187 188 189 190
/*
 * 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 *);
191
size_t __ksize(const void *);
192 193
size_t ksize(const void *);

K
Kees Cook 已提交
194
#ifdef CONFIG_HAVE_HARDENED_USERCOPY_ALLOCATOR
195 196
void __check_heap_object(const void *ptr, unsigned long n, struct page *page,
			bool to_user);
K
Kees Cook 已提交
197
#else
198 199
static inline void __check_heap_object(const void *ptr, unsigned long n,
				       struct page *page, bool to_user) { }
K
Kees Cook 已提交
200 201
#endif

202 203 204 205 206 207 208 209 210 211 212 213 214
/*
 * 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

215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
/*
 * 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)

233
/*
234 235 236 237 238 239
 * Kmalloc array related definitions
 */

#ifdef CONFIG_SLAB
/*
 * The largest kmalloc size supported by the SLAB allocators is
240 241 242 243 244 245 246
 * 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.
 */
247 248
#define KMALLOC_SHIFT_HIGH	((MAX_ORDER + PAGE_SHIFT - 1) <= 25 ? \
				(MAX_ORDER + PAGE_SHIFT - 1) : 25)
249
#define KMALLOC_SHIFT_MAX	KMALLOC_SHIFT_HIGH
250
#ifndef KMALLOC_SHIFT_LOW
251
#define KMALLOC_SHIFT_LOW	5
252
#endif
253 254 255
#endif

#ifdef CONFIG_SLUB
256
/*
257 258
 * SLUB directly allocates requests fitting in to an order-1 page
 * (PAGE_SIZE*2).  Larger requests are passed to the page allocator.
259 260
 */
#define KMALLOC_SHIFT_HIGH	(PAGE_SHIFT + 1)
261
#define KMALLOC_SHIFT_MAX	(MAX_ORDER + PAGE_SHIFT - 1)
262
#ifndef KMALLOC_SHIFT_LOW
263 264
#define KMALLOC_SHIFT_LOW	3
#endif
265
#endif
266

267 268
#ifdef CONFIG_SLOB
/*
269
 * SLOB passes all requests larger than one page to the page allocator.
270 271 272 273
 * No kmalloc array is necessary since objects of different sizes can
 * be allocated from the same page.
 */
#define KMALLOC_SHIFT_HIGH	PAGE_SHIFT
274
#define KMALLOC_SHIFT_MAX	(MAX_ORDER + PAGE_SHIFT - 1)
275 276 277 278 279
#ifndef KMALLOC_SHIFT_LOW
#define KMALLOC_SHIFT_LOW	3
#endif
#endif

280 281 282 283 284 285
/* 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)
286

287 288 289
/*
 * Kmalloc subsystem.
 */
290
#ifndef KMALLOC_MIN_SIZE
291
#define KMALLOC_MIN_SIZE (1 << KMALLOC_SHIFT_LOW)
292 293
#endif

J
Joonsoo Kim 已提交
294 295 296 297 298 299 300 301 302 303 304
/*
 * 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)

305 306 307 308
/*
 * Whenever changing this, take care of that kmalloc_type() and
 * create_kmalloc_caches() still work as intended.
 */
309 310
enum kmalloc_cache_type {
	KMALLOC_NORMAL = 0,
311
	KMALLOC_RECLAIM,
312 313 314 315 316 317
#ifdef CONFIG_ZONE_DMA
	KMALLOC_DMA,
#endif
	NR_KMALLOC_TYPES
};

318
#ifndef CONFIG_SLOB
319 320 321 322 323
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)
{
324
#ifdef CONFIG_ZONE_DMA
325 326 327 328 329 330
	/*
	 * The most common case is KMALLOC_NORMAL, so test for it
	 * with a single branch for both flags.
	 */
	if (likely((flags & (__GFP_DMA | __GFP_RECLAIMABLE)) == 0))
		return KMALLOC_NORMAL;
331 332

	/*
333 334
	 * At least one of the flags has to be set. If both are, __GFP_DMA
	 * is more important.
335
	 */
336 337 338 339
	return flags & __GFP_DMA ? KMALLOC_DMA : KMALLOC_RECLAIM;
#else
	return flags & __GFP_RECLAIMABLE ? KMALLOC_RECLAIM : KMALLOC_NORMAL;
#endif
340 341
}

342 343 344 345 346
/*
 * Figure out which kmalloc slab an allocation of a certain size
 * belongs to.
 * 0 = zero alloc
 * 1 =  65 .. 96 bytes
347 348
 * 2 = 129 .. 192 bytes
 * n = 2^(n-1)+1 .. 2^n
349
 */
350
static __always_inline unsigned int kmalloc_index(size_t size)
351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390
{
	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;
}
391
#endif /* !CONFIG_SLOB */
392

393 394
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;
395
void kmem_cache_free(struct kmem_cache *, void *);
396

397
/*
J
Jesper Dangaard Brouer 已提交
398
 * Bulk allocation and freeing operations. These are accelerated in an
399 400 401 402 403 404
 * 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 **);
405
int kmem_cache_alloc_bulk(struct kmem_cache *, gfp_t, size_t, void **);
406

407 408 409 410 411 412 413 414 415
/*
 * 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);
}

416
#ifdef CONFIG_NUMA
417 418
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;
419 420 421 422 423 424 425 426 427 428 429 430 431
#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
432
extern void *kmem_cache_alloc_trace(struct kmem_cache *, gfp_t, size_t) __assume_slab_alignment __malloc;
433 434 435 436

#ifdef CONFIG_NUMA
extern void *kmem_cache_alloc_node_trace(struct kmem_cache *s,
					   gfp_t gfpflags,
437
					   int node, size_t size) __assume_slab_alignment __malloc;
438 439 440 441 442 443 444 445 446 447 448 449 450 451
#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)
{
452 453
	void *ret = kmem_cache_alloc(s, flags);

454
	ret = kasan_kmalloc(s, ret, size, flags);
455
	return ret;
456 457 458 459 460 461 462
}

static __always_inline void *
kmem_cache_alloc_node_trace(struct kmem_cache *s,
			      gfp_t gfpflags,
			      int node, size_t size)
{
463 464
	void *ret = kmem_cache_alloc_node(s, gfpflags, node);

465
	ret = kasan_kmalloc(s, ret, size, gfpflags);
466
	return ret;
467 468 469
}
#endif /* CONFIG_TRACING */

470
extern void *kmalloc_order(size_t size, gfp_t flags, unsigned int order) __assume_page_alignment __malloc;
471 472

#ifdef CONFIG_TRACING
473
extern void *kmalloc_order_trace(size_t size, gfp_t flags, unsigned int order) __assume_page_alignment __malloc;
474 475 476 477 478 479
#else
static __always_inline void *
kmalloc_order_trace(size_t size, gfp_t flags, unsigned int order)
{
	return kmalloc_order(size, flags, order);
}
480 481
#endif

482 483 484 485 486 487 488 489 490
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.
491
 * @flags: the type of memory to allocate.
492 493 494
 *
 * kmalloc is the normal method of allocating memory
 * for objects smaller than page size in the kernel.
495
 *
496 497 498
 * The @flags argument may be one of the GFP flags defined at
 * include/linux/gfp.h and described at
 * :ref:`Documentation/core-api/mm-api.rst <mm-api-gfp-flags>`
499
 *
500
 * The recommended usage of the @flags is described at
501
 * :ref:`Documentation/core-api/memory-allocation.rst <memory-allocation>`
502
 *
503
 * Below is a brief outline of the most useful GFP flags
504
 *
505 506
 * %GFP_KERNEL
 *	Allocate normal kernel ram. May sleep.
507
 *
508 509
 * %GFP_NOWAIT
 *	Allocation will not sleep.
510
 *
511 512
 * %GFP_ATOMIC
 *	Allocation will not sleep.  May use emergency pools.
513
 *
514 515
 * %GFP_HIGHUSER
 *	Allocate memory from high memory on behalf of user.
516 517 518 519
 *
 * Also it is possible to set different flags by OR'ing
 * in one or more of the following additional @flags:
 *
520 521
 * %__GFP_HIGH
 *	This allocation has high priority and may use emergency pools.
522
 *
523 524 525
 * %__GFP_NOFAIL
 *	Indicate that this allocation is in no way allowed to fail
 *	(think twice before using).
526
 *
527 528 529
 * %__GFP_NORETRY
 *	If memory is not immediately available,
 *	then give up at once.
530
 *
531 532
 * %__GFP_NOWARN
 *	If allocation fails, don't issue any warnings.
533
 *
534 535 536
 * %__GFP_RETRY_MAYFAIL
 *	Try really hard to succeed the allocation but fail
 *	eventually.
537 538 539 540
 */
static __always_inline void *kmalloc(size_t size, gfp_t flags)
{
	if (__builtin_constant_p(size)) {
541 542 543
#ifndef CONFIG_SLOB
		unsigned int index;
#endif
544 545 546
		if (size > KMALLOC_MAX_CACHE_SIZE)
			return kmalloc_large(size, flags);
#ifndef CONFIG_SLOB
547
		index = kmalloc_index(size);
548

549 550
		if (!index)
			return ZERO_SIZE_PTR;
551

552 553 554
		return kmem_cache_alloc_trace(
				kmalloc_caches[kmalloc_type(flags)][index],
				flags, size);
555 556 557 558 559
#endif
	}
	return __kmalloc(size, flags);
}

560 561 562 563 564
/*
 * Determine size used for the nth kmalloc cache.
 * return size or 0 if a kmalloc cache for that
 * size does not exist
 */
565
static __always_inline unsigned int kmalloc_size(unsigned int n)
566
{
567
#ifndef CONFIG_SLOB
568
	if (n > 2)
569
		return 1U << n;
570 571 572 573 574 575

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

	if (n == 2 && KMALLOC_MIN_SIZE <= 64)
		return 192;
576
#endif
577 578 579
	return 0;
}

580 581 582 583
static __always_inline void *kmalloc_node(size_t size, gfp_t flags, int node)
{
#ifndef CONFIG_SLOB
	if (__builtin_constant_p(size) &&
584
		size <= KMALLOC_MAX_CACHE_SIZE) {
585
		unsigned int i = kmalloc_index(size);
586 587 588 589

		if (!i)
			return ZERO_SIZE_PTR;

590 591
		return kmem_cache_alloc_node_trace(
				kmalloc_caches[kmalloc_type(flags)][i],
592 593 594 595 596 597
						flags, node, size);
	}
#endif
	return __kmalloc_node(size, flags, node);
}

598 599
int memcg_update_all_caches(int num_memcgs);

600 601 602 603 604
/**
 * kmalloc_array - allocate memory for an array.
 * @n: number of elements.
 * @size: element size.
 * @flags: the type of memory to allocate (see kmalloc).
605
 */
X
Xi Wang 已提交
606
static inline void *kmalloc_array(size_t n, size_t size, gfp_t flags)
L
Linus Torvalds 已提交
607
{
608 609 610
	size_t bytes;

	if (unlikely(check_mul_overflow(n, size, &bytes)))
P
Paul Mundt 已提交
611
		return NULL;
612
	if (__builtin_constant_p(n) && __builtin_constant_p(size))
613 614
		return kmalloc(bytes, flags);
	return __kmalloc(bytes, flags);
X
Xi Wang 已提交
615 616 617 618 619 620 621 622 623 624 625
}

/**
 * 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
626 627
}

628 629 630 631 632 633 634 635
/*
 * 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.
 */
636
extern void *__kmalloc_track_caller(size_t, gfp_t, unsigned long);
637
#define kmalloc_track_caller(size, flags) \
638
	__kmalloc_track_caller(size, flags, _RET_IP_)
L
Linus Torvalds 已提交
639

640 641 642
static inline void *kmalloc_array_node(size_t n, size_t size, gfp_t flags,
				       int node)
{
643 644 645
	size_t bytes;

	if (unlikely(check_mul_overflow(n, size, &bytes)))
646 647
		return NULL;
	if (__builtin_constant_p(n) && __builtin_constant_p(size))
648 649
		return kmalloc_node(bytes, flags, node);
	return __kmalloc_node(bytes, flags, node);
650 651 652 653 654 655 656 657
}

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);
}


658
#ifdef CONFIG_NUMA
659
extern void *__kmalloc_node_track_caller(size_t, gfp_t, int, unsigned long);
660 661
#define kmalloc_node_track_caller(size, flags, node) \
	__kmalloc_node_track_caller(size, flags, node, \
662
			_RET_IP_)
663

664 665 666 667
#else /* CONFIG_NUMA */

#define kmalloc_node_track_caller(size, flags, node) \
	kmalloc_track_caller(size, flags)
668

P
Pascal Terjan 已提交
669
#endif /* CONFIG_NUMA */
670

671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688
/*
 * 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
689 690 691 692 693 694 695 696 697 698 699
/**
 * 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);
}

700
unsigned int kmem_cache_size(struct kmem_cache *s);
701 702
void __init kmem_cache_init_late(void);

703 704 705 706 707 708 709 710
#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
711
#endif	/* _LINUX_SLAB_H */