Blame view
mm/slab_common.c
44.6 KB
b24413180
|
1 |
// SPDX-License-Identifier: GPL-2.0 |
039363f38
|
2 3 4 5 6 7 8 9 10 11 12 |
/* * Slab allocator functions that are independent of the allocator strategy * * (C) 2012 Christoph Lameter <cl@linux.com> */ #include <linux/slab.h> #include <linux/mm.h> #include <linux/poison.h> #include <linux/interrupt.h> #include <linux/memory.h> |
1c99ba291
|
13 |
#include <linux/cache.h> |
039363f38
|
14 15 |
#include <linux/compiler.h> #include <linux/module.h> |
20cea9683
|
16 17 |
#include <linux/cpu.h> #include <linux/uaccess.h> |
b7454ad3c
|
18 19 |
#include <linux/seq_file.h> #include <linux/proc_fs.h> |
fcf8a1e48
|
20 |
#include <linux/debugfs.h> |
039363f38
|
21 22 23 |
#include <asm/cacheflush.h> #include <asm/tlbflush.h> #include <asm/page.h> |
2633d7a02
|
24 |
#include <linux/memcontrol.h> |
928cec9cd
|
25 26 |
#define CREATE_TRACE_POINTS |
f1b6eb6e6
|
27 |
#include <trace/events/kmem.h> |
039363f38
|
28 |
|
97d066091
|
29 30 31 |
#include "slab.h" enum slab_state slab_state; |
18004c5d4
|
32 33 |
LIST_HEAD(slab_caches); DEFINE_MUTEX(slab_mutex); |
9b030cb86
|
34 |
struct kmem_cache *kmem_cache; |
97d066091
|
35 |
|
2d891fbc3
|
36 37 38 39 40 41 42 |
#ifdef CONFIG_HARDENED_USERCOPY bool usercopy_fallback __ro_after_init = IS_ENABLED(CONFIG_HARDENED_USERCOPY_FALLBACK); module_param(usercopy_fallback, bool, 0400); MODULE_PARM_DESC(usercopy_fallback, "WARN instead of reject usercopy whitelist violations"); #endif |
657dc2f97
|
43 44 45 46 |
static LIST_HEAD(slab_caches_to_rcu_destroy); static void slab_caches_to_rcu_destroy_workfn(struct work_struct *work); static DECLARE_WORK(slab_caches_to_rcu_destroy_work, slab_caches_to_rcu_destroy_workfn); |
07f361b2b
|
47 |
/* |
423c929cb
|
48 49 50 |
* Set of flags that will prevent slab merging */ #define SLAB_NEVER_MERGE (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER | \ |
5f0d5a3ae
|
51 |
SLAB_TRACE | SLAB_TYPESAFE_BY_RCU | SLAB_NOLEAKTRACE | \ |
7ed2f9e66
|
52 |
SLAB_FAILSLAB | SLAB_KASAN) |
423c929cb
|
53 |
|
230e9fc28
|
54 |
#define SLAB_MERGE_SAME (SLAB_RECLAIM_ACCOUNT | SLAB_CACHE_DMA | \ |
6d6ea1e96
|
55 |
SLAB_CACHE_DMA32 | SLAB_ACCOUNT) |
423c929cb
|
56 57 58 |
/* * Merge control. If this is set then no merging of slab caches will occur. |
423c929cb
|
59 |
*/ |
7660a6fdd
|
60 |
static bool slab_nomerge = !IS_ENABLED(CONFIG_SLAB_MERGE_DEFAULT); |
423c929cb
|
61 62 63 |
static int __init setup_slab_nomerge(char *str) { |
7660a6fdd
|
64 |
slab_nomerge = true; |
423c929cb
|
65 66 67 68 69 70 71 72 73 74 |
return 1; } #ifdef CONFIG_SLUB __setup_param("slub_nomerge", slub_nomerge, setup_slab_nomerge, 0); #endif __setup("slab_nomerge", setup_slab_nomerge); /* |
07f361b2b
|
75 76 77 78 79 80 81 |
* Determine the size of a slab object */ unsigned int kmem_cache_size(struct kmem_cache *s) { return s->object_size; } EXPORT_SYMBOL(kmem_cache_size); |
77be4b136
|
82 |
#ifdef CONFIG_DEBUG_VM |
f4957d5bd
|
83 |
static int kmem_cache_sanity_check(const char *name, unsigned int size) |
039363f38
|
84 |
{ |
039363f38
|
85 86 |
if (!name || in_interrupt() || size < sizeof(void *) || size > KMALLOC_MAX_SIZE) { |
77be4b136
|
87 88 89 |
pr_err("kmem_cache_create(%s) integrity check failed ", name); return -EINVAL; |
039363f38
|
90 |
} |
b920536aa
|
91 |
|
20cea9683
|
92 |
WARN_ON(strchr(name, ' ')); /* It confuses parsers */ |
77be4b136
|
93 94 95 |
return 0; } #else |
f4957d5bd
|
96 |
static inline int kmem_cache_sanity_check(const char *name, unsigned int size) |
77be4b136
|
97 98 99 |
{ return 0; } |
20cea9683
|
100 |
#endif |
484748f0b
|
101 102 103 |
void __kmem_cache_free_bulk(struct kmem_cache *s, size_t nr, void **p) { size_t i; |
ca2571955
|
104 105 106 107 108 109 |
for (i = 0; i < nr; i++) { if (s) kmem_cache_free(s, p[i]); else kfree(p[i]); } |
484748f0b
|
110 |
} |
865762a81
|
111 |
int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t nr, |
484748f0b
|
112 113 114 115 116 117 118 119 |
void **p) { size_t i; for (i = 0; i < nr; i++) { void *x = p[i] = kmem_cache_alloc(s, flags); if (!x) { __kmem_cache_free_bulk(s, i, p); |
865762a81
|
120 |
return 0; |
484748f0b
|
121 122 |
} } |
865762a81
|
123 |
return i; |
484748f0b
|
124 |
} |
84c07d11a
|
125 |
#ifdef CONFIG_MEMCG_KMEM |
510ded33e
|
126 127 |
LIST_HEAD(slab_root_caches); |
63b02ef7d
|
128 |
static DEFINE_SPINLOCK(memcg_kmem_wq_lock); |
510ded33e
|
129 |
|
f0a3a24b5
|
130 |
static void kmemcg_cache_shutdown(struct percpu_ref *percpu_ref); |
f7ce3190c
|
131 |
void slab_init_memcg_params(struct kmem_cache *s) |
33a690c45
|
132 |
{ |
9eeadc8b6
|
133 |
s->memcg_params.root_cache = NULL; |
f7ce3190c
|
134 |
RCU_INIT_POINTER(s->memcg_params.memcg_caches, NULL); |
9eeadc8b6
|
135 |
INIT_LIST_HEAD(&s->memcg_params.children); |
92ee383f6
|
136 |
s->memcg_params.dying = false; |
f7ce3190c
|
137 138 139 |
} static int init_memcg_params(struct kmem_cache *s, |
c03914b7a
|
140 |
struct kmem_cache *root_cache) |
f7ce3190c
|
141 142 |
{ struct memcg_cache_array *arr; |
33a690c45
|
143 |
|
9eeadc8b6
|
144 |
if (root_cache) { |
f0a3a24b5
|
145 146 147 148 149 |
int ret = percpu_ref_init(&s->memcg_params.refcnt, kmemcg_cache_shutdown, 0, GFP_KERNEL); if (ret) return ret; |
f7ce3190c
|
150 |
s->memcg_params.root_cache = root_cache; |
9eeadc8b6
|
151 |
INIT_LIST_HEAD(&s->memcg_params.children_node); |
bc2791f85
|
152 |
INIT_LIST_HEAD(&s->memcg_params.kmem_caches_node); |
33a690c45
|
153 |
return 0; |
f7ce3190c
|
154 |
} |
33a690c45
|
155 |
|
f7ce3190c
|
156 |
slab_init_memcg_params(s); |
33a690c45
|
157 |
|
f7ce3190c
|
158 159 |
if (!memcg_nr_cache_ids) return 0; |
33a690c45
|
160 |
|
f80c7dab9
|
161 162 163 |
arr = kvzalloc(sizeof(struct memcg_cache_array) + memcg_nr_cache_ids * sizeof(void *), GFP_KERNEL); |
f7ce3190c
|
164 165 |
if (!arr) return -ENOMEM; |
33a690c45
|
166 |
|
f7ce3190c
|
167 |
RCU_INIT_POINTER(s->memcg_params.memcg_caches, arr); |
33a690c45
|
168 169 |
return 0; } |
f7ce3190c
|
170 |
static void destroy_memcg_params(struct kmem_cache *s) |
33a690c45
|
171 |
{ |
b749ecfaf
|
172 |
if (is_root_cache(s)) { |
f80c7dab9
|
173 |
kvfree(rcu_access_pointer(s->memcg_params.memcg_caches)); |
b749ecfaf
|
174 175 176 |
} else { mem_cgroup_put(s->memcg_params.memcg); WRITE_ONCE(s->memcg_params.memcg, NULL); |
f0a3a24b5
|
177 |
percpu_ref_exit(&s->memcg_params.refcnt); |
b749ecfaf
|
178 |
} |
f80c7dab9
|
179 180 181 182 183 184 185 186 |
} static void free_memcg_params(struct rcu_head *rcu) { struct memcg_cache_array *old; old = container_of(rcu, struct memcg_cache_array, rcu); kvfree(old); |
33a690c45
|
187 |
} |
f7ce3190c
|
188 |
static int update_memcg_params(struct kmem_cache *s, int new_array_size) |
6f817f4cd
|
189 |
{ |
f7ce3190c
|
190 |
struct memcg_cache_array *old, *new; |
6f817f4cd
|
191 |
|
f80c7dab9
|
192 193 |
new = kvzalloc(sizeof(struct memcg_cache_array) + new_array_size * sizeof(void *), GFP_KERNEL); |
f7ce3190c
|
194 |
if (!new) |
6f817f4cd
|
195 |
return -ENOMEM; |
f7ce3190c
|
196 197 198 199 200 |
old = rcu_dereference_protected(s->memcg_params.memcg_caches, lockdep_is_held(&slab_mutex)); if (old) memcpy(new->entries, old->entries, memcg_nr_cache_ids * sizeof(void *)); |
6f817f4cd
|
201 |
|
f7ce3190c
|
202 203 |
rcu_assign_pointer(s->memcg_params.memcg_caches, new); if (old) |
f80c7dab9
|
204 |
call_rcu(&old->rcu, free_memcg_params); |
6f817f4cd
|
205 206 |
return 0; } |
55007d849
|
207 208 209 210 |
int memcg_update_all_caches(int num_memcgs) { struct kmem_cache *s; int ret = 0; |
55007d849
|
211 |
|
05257a1a3
|
212 |
mutex_lock(&slab_mutex); |
510ded33e
|
213 |
list_for_each_entry(s, &slab_root_caches, root_caches_node) { |
f7ce3190c
|
214 |
ret = update_memcg_params(s, num_memcgs); |
55007d849
|
215 |
/* |
55007d849
|
216 217 218 219 |
* Instead of freeing the memory, we'll just leave the caches * up to this point in an updated state. */ if (ret) |
05257a1a3
|
220 |
break; |
55007d849
|
221 |
} |
55007d849
|
222 223 224 |
mutex_unlock(&slab_mutex); return ret; } |
657dc2f97
|
225 |
|
c03914b7a
|
226 |
void memcg_link_cache(struct kmem_cache *s, struct mem_cgroup *memcg) |
657dc2f97
|
227 |
{ |
510ded33e
|
228 229 230 |
if (is_root_cache(s)) { list_add(&s->root_caches_node, &slab_root_caches); } else { |
f0a3a24b5
|
231 |
css_get(&memcg->css); |
c03914b7a
|
232 |
s->memcg_params.memcg = memcg; |
510ded33e
|
233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
list_add(&s->memcg_params.children_node, &s->memcg_params.root_cache->memcg_params.children); list_add(&s->memcg_params.kmem_caches_node, &s->memcg_params.memcg->kmem_caches); } } static void memcg_unlink_cache(struct kmem_cache *s) { if (is_root_cache(s)) { list_del(&s->root_caches_node); } else { list_del(&s->memcg_params.children_node); list_del(&s->memcg_params.kmem_caches_node); } |
657dc2f97
|
248 |
} |
33a690c45
|
249 |
#else |
f7ce3190c
|
250 |
static inline int init_memcg_params(struct kmem_cache *s, |
c03914b7a
|
251 |
struct kmem_cache *root_cache) |
33a690c45
|
252 253 254 |
{ return 0; } |
f7ce3190c
|
255 |
static inline void destroy_memcg_params(struct kmem_cache *s) |
33a690c45
|
256 257 |
{ } |
657dc2f97
|
258 |
|
510ded33e
|
259 |
static inline void memcg_unlink_cache(struct kmem_cache *s) |
657dc2f97
|
260 261 |
{ } |
84c07d11a
|
262 |
#endif /* CONFIG_MEMCG_KMEM */ |
55007d849
|
263 |
|
77be4b136
|
264 |
/* |
692ae74aa
|
265 266 267 |
* Figure out what the alignment of the objects will be given a set of * flags, a user specified alignment and the size of the objects. */ |
f4957d5bd
|
268 269 |
static unsigned int calculate_alignment(slab_flags_t flags, unsigned int align, unsigned int size) |
692ae74aa
|
270 271 272 273 274 275 276 277 278 |
{ /* * 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. */ if (flags & SLAB_HWCACHE_ALIGN) { |
f4957d5bd
|
279 |
unsigned int ralign; |
692ae74aa
|
280 281 282 283 284 285 286 287 288 289 290 291 292 293 |
ralign = cache_line_size(); while (size <= ralign / 2) ralign /= 2; align = max(align, ralign); } if (align < ARCH_SLAB_MINALIGN) align = ARCH_SLAB_MINALIGN; return ALIGN(align, sizeof(void *)); } /* |
423c929cb
|
294 295 296 297 298 299 300 301 302 303 304 305 |
* Find a mergeable slab cache */ int slab_unmergeable(struct kmem_cache *s) { if (slab_nomerge || (s->flags & SLAB_NEVER_MERGE)) return 1; if (!is_root_cache(s)) return 1; if (s->ctor) return 1; |
8eb8284b4
|
306 307 |
if (s->usersize) return 1; |
423c929cb
|
308 309 310 311 312 313 314 315 |
/* * We may have set a slab to be unmergeable during bootstrap. */ if (s->refcount < 0) return 1; return 0; } |
f4957d5bd
|
316 |
struct kmem_cache *find_mergeable(unsigned int size, unsigned int align, |
d50112edd
|
317 |
slab_flags_t flags, const char *name, void (*ctor)(void *)) |
423c929cb
|
318 319 |
{ struct kmem_cache *s; |
c6e28895a
|
320 |
if (slab_nomerge) |
423c929cb
|
321 322 323 324 325 326 327 328 329 |
return NULL; if (ctor) return NULL; size = ALIGN(size, sizeof(void *)); align = calculate_alignment(flags, align, size); size = ALIGN(size, align); flags = kmem_cache_flags(size, flags, name, NULL); |
c6e28895a
|
330 331 |
if (flags & SLAB_NEVER_MERGE) return NULL; |
510ded33e
|
332 |
list_for_each_entry_reverse(s, &slab_root_caches, root_caches_node) { |
423c929cb
|
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 |
if (slab_unmergeable(s)) continue; if (size > s->size) continue; if ((flags & SLAB_MERGE_SAME) != (s->flags & SLAB_MERGE_SAME)) 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; |
95069ac8d
|
350 351 352 |
if (IS_ENABLED(CONFIG_SLAB) && align && (align > s->align || s->align % align)) continue; |
423c929cb
|
353 354 355 356 |
return s; } return NULL; } |
c9a77a792
|
357 |
static struct kmem_cache *create_cache(const char *name, |
613a5eb56
|
358 |
unsigned int object_size, unsigned int align, |
7bbdb81ee
|
359 360 |
slab_flags_t flags, unsigned int useroffset, unsigned int usersize, void (*ctor)(void *), |
c9a77a792
|
361 |
struct mem_cgroup *memcg, struct kmem_cache *root_cache) |
794b1248b
|
362 363 364 |
{ struct kmem_cache *s; int err; |
8eb8284b4
|
365 366 |
if (WARN_ON(useroffset + usersize > object_size)) useroffset = usersize = 0; |
794b1248b
|
367 368 369 370 371 372 |
err = -ENOMEM; s = kmem_cache_zalloc(kmem_cache, GFP_KERNEL); if (!s) goto out; s->name = name; |
613a5eb56
|
373 |
s->size = s->object_size = object_size; |
794b1248b
|
374 375 |
s->align = align; s->ctor = ctor; |
8eb8284b4
|
376 377 |
s->useroffset = useroffset; s->usersize = usersize; |
794b1248b
|
378 |
|
c03914b7a
|
379 |
err = init_memcg_params(s, root_cache); |
794b1248b
|
380 381 382 383 384 385 386 387 388 |
if (err) goto out_free_cache; err = __kmem_cache_create(s, flags); if (err) goto out_free_cache; s->refcount = 1; list_add(&s->list, &slab_caches); |
c03914b7a
|
389 |
memcg_link_cache(s, memcg); |
794b1248b
|
390 391 392 393 394 395 |
out: if (err) return ERR_PTR(err); return s; out_free_cache: |
f7ce3190c
|
396 |
destroy_memcg_params(s); |
7c4da061f
|
397 |
kmem_cache_free(kmem_cache, s); |
794b1248b
|
398 399 |
goto out; } |
459068554
|
400 |
|
f496990f1
|
401 402 403 |
/** * kmem_cache_create_usercopy - Create a cache with a region suitable * for copying to userspace |
77be4b136
|
404 405 406 407 |
* @name: A string which is used in /proc/slabinfo to identify this cache. * @size: The size of objects to be created in this cache. * @align: The required alignment for the objects. * @flags: SLAB flags |
8eb8284b4
|
408 409 |
* @useroffset: Usercopy region offset * @usersize: Usercopy region size |
77be4b136
|
410 411 |
* @ctor: A constructor for the objects. * |
77be4b136
|
412 413 414 415 416 417 418 419 |
* Cannot be called within a interrupt, but can be interrupted. * The @ctor is run when new pages are allocated by the cache. * * The flags are * * %SLAB_POISON - Poison the slab with a known test pattern (a5a5a5a5) * to catch references to uninitialised memory. * |
f496990f1
|
420 |
* %SLAB_RED_ZONE - Insert `Red` zones around the allocated memory to check |
77be4b136
|
421 422 423 424 425 |
* for buffer overruns. * * %SLAB_HWCACHE_ALIGN - Align the objects in this cache to a hardware * cacheline. This can be beneficial if you're counting cycles as closely * as davem. |
f496990f1
|
426 427 |
* * Return: a pointer to the cache on success, NULL on failure. |
77be4b136
|
428 |
*/ |
2633d7a02
|
429 |
struct kmem_cache * |
f4957d5bd
|
430 431 |
kmem_cache_create_usercopy(const char *name, unsigned int size, unsigned int align, |
7bbdb81ee
|
432 433 |
slab_flags_t flags, unsigned int useroffset, unsigned int usersize, |
8eb8284b4
|
434 |
void (*ctor)(void *)) |
77be4b136
|
435 |
{ |
40911a798
|
436 |
struct kmem_cache *s = NULL; |
3dec16ea3
|
437 |
const char *cache_name; |
3965fc365
|
438 |
int err; |
039363f38
|
439 |
|
77be4b136
|
440 |
get_online_cpus(); |
03afc0e25
|
441 |
get_online_mems(); |
05257a1a3
|
442 |
memcg_get_cache_ids(); |
03afc0e25
|
443 |
|
77be4b136
|
444 |
mutex_lock(&slab_mutex); |
686d550d2
|
445 |
|
794b1248b
|
446 |
err = kmem_cache_sanity_check(name, size); |
3aa24f519
|
447 |
if (err) { |
3965fc365
|
448 |
goto out_unlock; |
3aa24f519
|
449 |
} |
686d550d2
|
450 |
|
e70954fd6
|
451 452 453 454 455 |
/* Refuse requests with allocator specific flags */ if (flags & ~SLAB_FLAGS_PERMITTED) { err = -EINVAL; goto out_unlock; } |
d8843922f
|
456 457 458 459 460 461 462 |
/* * Some allocators will constraint the set of valid flags to a subset * of all flags. We expect them to define CACHE_CREATE_MASK in this * case, and we'll just provide them with a sanitized version of the * passed flags. */ flags &= CACHE_CREATE_MASK; |
686d550d2
|
463 |
|
8eb8284b4
|
464 465 466 467 468 469 470 |
/* Fail closed on bad usersize of useroffset values. */ if (WARN_ON(!usersize && useroffset) || WARN_ON(size < usersize || size - usersize < useroffset)) usersize = useroffset = 0; if (!usersize) s = __kmem_cache_alias(name, size, align, flags, ctor); |
794b1248b
|
471 |
if (s) |
3965fc365
|
472 |
goto out_unlock; |
2633d7a02
|
473 |
|
3dec16ea3
|
474 |
cache_name = kstrdup_const(name, GFP_KERNEL); |
794b1248b
|
475 476 477 478 |
if (!cache_name) { err = -ENOMEM; goto out_unlock; } |
7c9adf5a5
|
479 |
|
613a5eb56
|
480 |
s = create_cache(cache_name, size, |
c9a77a792
|
481 |
calculate_alignment(flags, align, size), |
8eb8284b4
|
482 |
flags, useroffset, usersize, ctor, NULL, NULL); |
794b1248b
|
483 484 |
if (IS_ERR(s)) { err = PTR_ERR(s); |
3dec16ea3
|
485 |
kfree_const(cache_name); |
794b1248b
|
486 |
} |
3965fc365
|
487 488 |
out_unlock: |
20cea9683
|
489 |
mutex_unlock(&slab_mutex); |
03afc0e25
|
490 |
|
05257a1a3
|
491 |
memcg_put_cache_ids(); |
03afc0e25
|
492 |
put_online_mems(); |
20cea9683
|
493 |
put_online_cpus(); |
ba3253c78
|
494 |
if (err) { |
686d550d2
|
495 496 497 498 499 |
if (flags & SLAB_PANIC) panic("kmem_cache_create: Failed to create slab '%s'. Error %d ", name, err); else { |
1170532bb
|
500 501 |
pr_warn("kmem_cache_create(%s) failed with error %d ", |
686d550d2
|
502 503 504 |
name, err); dump_stack(); } |
686d550d2
|
505 506 |
return NULL; } |
039363f38
|
507 508 |
return s; } |
8eb8284b4
|
509 |
EXPORT_SYMBOL(kmem_cache_create_usercopy); |
f496990f1
|
510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 |
/** * kmem_cache_create - Create a cache. * @name: A string which is used in /proc/slabinfo to identify this cache. * @size: The size of objects to be created in this cache. * @align: The required alignment for the objects. * @flags: SLAB flags * @ctor: A constructor for the objects. * * Cannot be called within a interrupt, but can be interrupted. * The @ctor is run when new pages are allocated by the cache. * * The flags are * * %SLAB_POISON - Poison the slab with a known test pattern (a5a5a5a5) * to catch references to uninitialised memory. * * %SLAB_RED_ZONE - Insert `Red` zones around the allocated memory to check * for buffer overruns. * * %SLAB_HWCACHE_ALIGN - Align the objects in this cache to a hardware * cacheline. This can be beneficial if you're counting cycles as closely * as davem. * * Return: a pointer to the cache on success, NULL on failure. */ |
8eb8284b4
|
535 |
struct kmem_cache * |
f4957d5bd
|
536 |
kmem_cache_create(const char *name, unsigned int size, unsigned int align, |
8eb8284b4
|
537 538 |
slab_flags_t flags, void (*ctor)(void *)) { |
6d07d1cd3
|
539 |
return kmem_cache_create_usercopy(name, size, align, flags, 0, 0, |
8eb8284b4
|
540 541 |
ctor); } |
794b1248b
|
542 |
EXPORT_SYMBOL(kmem_cache_create); |
2633d7a02
|
543 |
|
657dc2f97
|
544 |
static void slab_caches_to_rcu_destroy_workfn(struct work_struct *work) |
d5b3cf713
|
545 |
{ |
657dc2f97
|
546 547 |
LIST_HEAD(to_destroy); struct kmem_cache *s, *s2; |
d5b3cf713
|
548 |
|
657dc2f97
|
549 |
/* |
5f0d5a3ae
|
550 |
* On destruction, SLAB_TYPESAFE_BY_RCU kmem_caches are put on the |
657dc2f97
|
551 552 553 554 555 556 557 558 559 560 |
* @slab_caches_to_rcu_destroy list. The slab pages are freed * through RCU and and the associated kmem_cache are dereferenced * while freeing the pages, so the kmem_caches should be freed only * after the pending RCU operations are finished. As rcu_barrier() * is a pretty slow operation, we batch all pending destructions * asynchronously. */ mutex_lock(&slab_mutex); list_splice_init(&slab_caches_to_rcu_destroy, &to_destroy); mutex_unlock(&slab_mutex); |
d5b3cf713
|
561 |
|
657dc2f97
|
562 563 564 565 566 567 568 569 570 571 572 573 |
if (list_empty(&to_destroy)) return; rcu_barrier(); list_for_each_entry_safe(s, s2, &to_destroy, list) { #ifdef SLAB_SUPPORTS_SYSFS sysfs_slab_release(s); #else slab_kmem_cache_release(s); #endif } |
d5b3cf713
|
574 |
} |
657dc2f97
|
575 |
static int shutdown_cache(struct kmem_cache *s) |
d5b3cf713
|
576 |
{ |
f9fa1d919
|
577 578 |
/* free asan quarantined objects */ kasan_cache_shutdown(s); |
657dc2f97
|
579 580 |
if (__kmem_cache_shutdown(s) != 0) return -EBUSY; |
d5b3cf713
|
581 |
|
510ded33e
|
582 |
memcg_unlink_cache(s); |
657dc2f97
|
583 |
list_del(&s->list); |
d5b3cf713
|
584 |
|
5f0d5a3ae
|
585 |
if (s->flags & SLAB_TYPESAFE_BY_RCU) { |
d50d82faa
|
586 587 588 |
#ifdef SLAB_SUPPORTS_SYSFS sysfs_slab_unlink(s); #endif |
657dc2f97
|
589 590 591 |
list_add_tail(&s->list, &slab_caches_to_rcu_destroy); schedule_work(&slab_caches_to_rcu_destroy_work); } else { |
d5b3cf713
|
592 |
#ifdef SLAB_SUPPORTS_SYSFS |
d50d82faa
|
593 |
sysfs_slab_unlink(s); |
bf5eb3de3
|
594 |
sysfs_slab_release(s); |
d5b3cf713
|
595 596 597 598 |
#else slab_kmem_cache_release(s); #endif } |
657dc2f97
|
599 600 |
return 0; |
d5b3cf713
|
601 |
} |
84c07d11a
|
602 |
#ifdef CONFIG_MEMCG_KMEM |
794b1248b
|
603 |
/* |
776ed0f03
|
604 |
* memcg_create_kmem_cache - Create a cache for a memory cgroup. |
794b1248b
|
605 606 607 608 609 610 611 |
* @memcg: The memory cgroup the new cache is for. * @root_cache: The parent of the new cache. * * This function attempts to create a kmem cache that will serve allocation * requests going from @memcg to @root_cache. The new cache inherits properties * from its parent. */ |
d5b3cf713
|
612 613 |
void memcg_create_kmem_cache(struct mem_cgroup *memcg, struct kmem_cache *root_cache) |
2633d7a02
|
614 |
{ |
3e0350a36
|
615 |
static char memcg_name_buf[NAME_MAX + 1]; /* protected by slab_mutex */ |
33398cf2f
|
616 |
struct cgroup_subsys_state *css = &memcg->css; |
f7ce3190c
|
617 |
struct memcg_cache_array *arr; |
bd6731458
|
618 |
struct kmem_cache *s = NULL; |
794b1248b
|
619 |
char *cache_name; |
f7ce3190c
|
620 |
int idx; |
794b1248b
|
621 622 |
get_online_cpus(); |
03afc0e25
|
623 |
get_online_mems(); |
794b1248b
|
624 |
mutex_lock(&slab_mutex); |
2a4db7eb9
|
625 |
/* |
567e9ab2e
|
626 |
* The memory cgroup could have been offlined while the cache |
2a4db7eb9
|
627 628 |
* creation work was pending. */ |
570332978
|
629 |
if (memcg->kmem_state != KMEM_ONLINE) |
2a4db7eb9
|
630 |
goto out_unlock; |
f7ce3190c
|
631 632 633 |
idx = memcg_cache_id(memcg); arr = rcu_dereference_protected(root_cache->memcg_params.memcg_caches, lockdep_is_held(&slab_mutex)); |
d5b3cf713
|
634 635 636 637 638 |
/* * Since per-memcg caches are created asynchronously on first * allocation (see memcg_kmem_get_cache()), several threads can try to * create the same cache, but only one of them may succeed. */ |
f7ce3190c
|
639 |
if (arr->entries[idx]) |
d5b3cf713
|
640 |
goto out_unlock; |
f1008365b
|
641 |
cgroup_name(css->cgroup, memcg_name_buf, sizeof(memcg_name_buf)); |
73f576c04
|
642 643 |
cache_name = kasprintf(GFP_KERNEL, "%s(%llu:%s)", root_cache->name, css->serial_nr, memcg_name_buf); |
794b1248b
|
644 645 |
if (!cache_name) goto out_unlock; |
c9a77a792
|
646 |
s = create_cache(cache_name, root_cache->object_size, |
613a5eb56
|
647 |
root_cache->align, |
f773e36de
|
648 |
root_cache->flags & CACHE_CREATE_MASK, |
8eb8284b4
|
649 |
root_cache->useroffset, root_cache->usersize, |
f773e36de
|
650 |
root_cache->ctor, memcg, root_cache); |
d5b3cf713
|
651 652 653 654 655 |
/* * If we could not create a memcg cache, do not complain, because * that's not critical at all as we can always proceed with the root * cache. */ |
bd6731458
|
656 |
if (IS_ERR(s)) { |
794b1248b
|
657 |
kfree(cache_name); |
d5b3cf713
|
658 |
goto out_unlock; |
bd6731458
|
659 |
} |
794b1248b
|
660 |
|
d5b3cf713
|
661 |
/* |
f0a3a24b5
|
662 |
* Since readers won't lock (see memcg_kmem_get_cache()), we need a |
d5b3cf713
|
663 664 665 666 |
* barrier here to ensure nobody will see the kmem_cache partially * initialized. */ smp_wmb(); |
f7ce3190c
|
667 |
arr->entries[idx] = s; |
d5b3cf713
|
668 |
|
794b1248b
|
669 670 |
out_unlock: mutex_unlock(&slab_mutex); |
03afc0e25
|
671 672 |
put_online_mems(); |
794b1248b
|
673 |
put_online_cpus(); |
2633d7a02
|
674 |
} |
b8529907b
|
675 |
|
0b14e8aa6
|
676 |
static void kmemcg_workfn(struct work_struct *work) |
01fb58bcb
|
677 678 |
{ struct kmem_cache *s = container_of(work, struct kmem_cache, |
0b14e8aa6
|
679 |
memcg_params.work); |
01fb58bcb
|
680 681 682 683 684 |
get_online_cpus(); get_online_mems(); mutex_lock(&slab_mutex); |
0b14e8aa6
|
685 |
s->memcg_params.work_fn(s); |
01fb58bcb
|
686 687 688 689 |
mutex_unlock(&slab_mutex); put_online_mems(); put_online_cpus(); |
01fb58bcb
|
690 |
} |
0b14e8aa6
|
691 |
static void kmemcg_rcufn(struct rcu_head *head) |
01fb58bcb
|
692 693 |
{ struct kmem_cache *s = container_of(head, struct kmem_cache, |
0b14e8aa6
|
694 |
memcg_params.rcu_head); |
01fb58bcb
|
695 696 |
/* |
0b14e8aa6
|
697 |
* We need to grab blocking locks. Bounce to ->work. The |
01fb58bcb
|
698 699 700 |
* work item shares the space with the RCU head and can't be * initialized eariler. */ |
0b14e8aa6
|
701 702 |
INIT_WORK(&s->memcg_params.work, kmemcg_workfn); queue_work(memcg_kmem_cache_wq, &s->memcg_params.work); |
01fb58bcb
|
703 |
} |
f0a3a24b5
|
704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 |
static void kmemcg_cache_shutdown_fn(struct kmem_cache *s) { WARN_ON(shutdown_cache(s)); } static void kmemcg_cache_shutdown(struct percpu_ref *percpu_ref) { struct kmem_cache *s = container_of(percpu_ref, struct kmem_cache, memcg_params.refcnt); unsigned long flags; spin_lock_irqsave(&memcg_kmem_wq_lock, flags); if (s->memcg_params.root_cache->memcg_params.dying) goto unlock; s->memcg_params.work_fn = kmemcg_cache_shutdown_fn; INIT_WORK(&s->memcg_params.work, kmemcg_workfn); queue_work(memcg_kmem_cache_wq, &s->memcg_params.work); unlock: spin_unlock_irqrestore(&memcg_kmem_wq_lock, flags); } static void kmemcg_cache_deactivate_after_rcu(struct kmem_cache *s) { __kmemcg_cache_deactivate_after_rcu(s); percpu_ref_kill(&s->memcg_params.refcnt); } |
434866947
|
732 |
static void kmemcg_cache_deactivate(struct kmem_cache *s) |
01fb58bcb
|
733 |
{ |
f0a3a24b5
|
734 |
if (WARN_ON_ONCE(is_root_cache(s))) |
01fb58bcb
|
735 |
return; |
434866947
|
736 |
__kmemcg_cache_deactivate(s); |
fcf8a1e48
|
737 |
s->flags |= SLAB_DEACTIVATED; |
434866947
|
738 |
|
63b02ef7d
|
739 740 741 742 743 744 |
/* * memcg_kmem_wq_lock is used to synchronize memcg_params.dying * flag and make sure that no new kmem_cache deactivation tasks * are queued (see flush_memcg_workqueue() ). */ spin_lock_irq(&memcg_kmem_wq_lock); |
92ee383f6
|
745 |
if (s->memcg_params.root_cache->memcg_params.dying) |
63b02ef7d
|
746 |
goto unlock; |
92ee383f6
|
747 |
|
f0a3a24b5
|
748 |
s->memcg_params.work_fn = kmemcg_cache_deactivate_after_rcu; |
0b14e8aa6
|
749 |
call_rcu(&s->memcg_params.rcu_head, kmemcg_rcufn); |
63b02ef7d
|
750 751 |
unlock: spin_unlock_irq(&memcg_kmem_wq_lock); |
01fb58bcb
|
752 |
} |
fb2f2b0ad
|
753 754 |
void memcg_deactivate_kmem_caches(struct mem_cgroup *memcg, struct mem_cgroup *parent) |
2a4db7eb9
|
755 756 757 |
{ int idx; struct memcg_cache_array *arr; |
d6e0b7fa1
|
758 |
struct kmem_cache *s, *c; |
fb2f2b0ad
|
759 |
unsigned int nr_reparented; |
2a4db7eb9
|
760 761 |
idx = memcg_cache_id(memcg); |
d6e0b7fa1
|
762 763 |
get_online_cpus(); get_online_mems(); |
2a4db7eb9
|
764 |
mutex_lock(&slab_mutex); |
510ded33e
|
765 |
list_for_each_entry(s, &slab_root_caches, root_caches_node) { |
2a4db7eb9
|
766 767 |
arr = rcu_dereference_protected(s->memcg_params.memcg_caches, lockdep_is_held(&slab_mutex)); |
d6e0b7fa1
|
768 769 770 |
c = arr->entries[idx]; if (!c) continue; |
434866947
|
771 |
kmemcg_cache_deactivate(c); |
2a4db7eb9
|
772 773 |
arr->entries[idx] = NULL; } |
fb2f2b0ad
|
774 775 776 777 778 779 780 781 782 783 784 785 |
nr_reparented = 0; list_for_each_entry(s, &memcg->kmem_caches, memcg_params.kmem_caches_node) { WRITE_ONCE(s->memcg_params.memcg, parent); css_put(&memcg->css); nr_reparented++; } if (nr_reparented) { list_splice_init(&memcg->kmem_caches, &parent->kmem_caches); css_get_many(&parent->css, nr_reparented); } |
2a4db7eb9
|
786 |
mutex_unlock(&slab_mutex); |
d6e0b7fa1
|
787 788 789 |
put_online_mems(); put_online_cpus(); |
2a4db7eb9
|
790 |
} |
657dc2f97
|
791 |
static int shutdown_memcg_caches(struct kmem_cache *s) |
d60fdcc9e
|
792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 |
{ struct memcg_cache_array *arr; struct kmem_cache *c, *c2; LIST_HEAD(busy); int i; BUG_ON(!is_root_cache(s)); /* * First, shutdown active caches, i.e. caches that belong to online * memory cgroups. */ arr = rcu_dereference_protected(s->memcg_params.memcg_caches, lockdep_is_held(&slab_mutex)); for_each_memcg_cache_index(i) { c = arr->entries[i]; if (!c) continue; |
657dc2f97
|
810 |
if (shutdown_cache(c)) |
d60fdcc9e
|
811 812 813 814 815 |
/* * The cache still has objects. Move it to a temporary * list so as not to try to destroy it for a second * time while iterating over inactive caches below. */ |
9eeadc8b6
|
816 |
list_move(&c->memcg_params.children_node, &busy); |
d60fdcc9e
|
817 818 819 820 821 822 823 824 825 826 827 828 829 830 |
else /* * The cache is empty and will be destroyed soon. Clear * the pointer to it in the memcg_caches array so that * it will never be accessed even if the root cache * stays alive. */ arr->entries[i] = NULL; } /* * Second, shutdown all caches left from memory cgroups that are now * offline. */ |
9eeadc8b6
|
831 832 |
list_for_each_entry_safe(c, c2, &s->memcg_params.children, memcg_params.children_node) |
657dc2f97
|
833 |
shutdown_cache(c); |
d60fdcc9e
|
834 |
|
9eeadc8b6
|
835 |
list_splice(&busy, &s->memcg_params.children); |
d60fdcc9e
|
836 837 838 839 840 |
/* * A cache being destroyed must be empty. In particular, this means * that all per memcg caches attached to it must be empty too. */ |
9eeadc8b6
|
841 |
if (!list_empty(&s->memcg_params.children)) |
d60fdcc9e
|
842 843 844 |
return -EBUSY; return 0; } |
92ee383f6
|
845 846 847 |
static void flush_memcg_workqueue(struct kmem_cache *s) { |
63b02ef7d
|
848 |
spin_lock_irq(&memcg_kmem_wq_lock); |
92ee383f6
|
849 |
s->memcg_params.dying = true; |
63b02ef7d
|
850 |
spin_unlock_irq(&memcg_kmem_wq_lock); |
92ee383f6
|
851 852 |
/* |
434866947
|
853 |
* SLAB and SLUB deactivate the kmem_caches through call_rcu. Make |
92ee383f6
|
854 855 |
* sure all registered rcu callbacks have been invoked. */ |
434866947
|
856 |
rcu_barrier(); |
92ee383f6
|
857 858 859 860 861 862 |
/* * SLAB and SLUB create memcg kmem_caches through workqueue and SLUB * deactivates the memcg kmem_caches through workqueue. Make sure all * previous workitems on workqueue are processed. */ |
904c1db4b
|
863 864 |
if (likely(memcg_kmem_cache_wq)) flush_workqueue(memcg_kmem_cache_wq); |
e4d09b31a
|
865 866 867 868 869 870 871 872 873 874 875 876 |
/* * If we're racing with children kmem_cache deactivation, it might * take another rcu grace period to complete their destruction. * At this moment the corresponding percpu_ref_kill() call should be * done, but it might take another rcu grace period to complete * switching to the atomic mode. * Please, note that we check without grabbing the slab_mutex. It's safe * because at this moment the children list can't grow. */ if (!list_empty(&s->memcg_params.children)) rcu_barrier(); |
92ee383f6
|
877 |
} |
d60fdcc9e
|
878 |
#else |
657dc2f97
|
879 |
static inline int shutdown_memcg_caches(struct kmem_cache *s) |
d60fdcc9e
|
880 881 882 |
{ return 0; } |
92ee383f6
|
883 884 885 886 |
static inline void flush_memcg_workqueue(struct kmem_cache *s) { } |
84c07d11a
|
887 |
#endif /* CONFIG_MEMCG_KMEM */ |
97d066091
|
888 |
|
41a212859
|
889 890 |
void slab_kmem_cache_release(struct kmem_cache *s) { |
52b4b950b
|
891 |
__kmem_cache_release(s); |
f7ce3190c
|
892 |
destroy_memcg_params(s); |
3dec16ea3
|
893 |
kfree_const(s->name); |
41a212859
|
894 895 |
kmem_cache_free(kmem_cache, s); } |
945cf2b61
|
896 897 |
void kmem_cache_destroy(struct kmem_cache *s) { |
d60fdcc9e
|
898 |
int err; |
d5b3cf713
|
899 |
|
3942d2991
|
900 901 |
if (unlikely(!s)) return; |
92ee383f6
|
902 |
flush_memcg_workqueue(s); |
945cf2b61
|
903 |
get_online_cpus(); |
03afc0e25
|
904 |
get_online_mems(); |
945cf2b61
|
905 |
mutex_lock(&slab_mutex); |
b8529907b
|
906 |
|
945cf2b61
|
907 |
s->refcount--; |
b8529907b
|
908 909 |
if (s->refcount) goto out_unlock; |
657dc2f97
|
910 |
err = shutdown_memcg_caches(s); |
d60fdcc9e
|
911 |
if (!err) |
657dc2f97
|
912 |
err = shutdown_cache(s); |
b8529907b
|
913 |
|
cd918c557
|
914 |
if (err) { |
756a025f0
|
915 916 917 |
pr_err("kmem_cache_destroy %s: Slab cache still has objects ", s->name); |
cd918c557
|
918 919 |
dump_stack(); } |
b8529907b
|
920 921 |
out_unlock: mutex_unlock(&slab_mutex); |
d5b3cf713
|
922 |
|
03afc0e25
|
923 |
put_online_mems(); |
945cf2b61
|
924 925 926 |
put_online_cpus(); } EXPORT_SYMBOL(kmem_cache_destroy); |
03afc0e25
|
927 928 929 930 931 932 |
/** * kmem_cache_shrink - Shrink a cache. * @cachep: The cache to shrink. * * Releases as many slabs as possible for a cache. * To help debugging, a zero exit status indicates all slabs were released. |
a862f68a8
|
933 934 |
* * Return: %0 if all slabs were released, non-zero otherwise |
03afc0e25
|
935 936 937 938 939 940 941 |
*/ int kmem_cache_shrink(struct kmem_cache *cachep) { int ret; get_online_cpus(); get_online_mems(); |
55834c590
|
942 |
kasan_cache_shrink(cachep); |
c9fc58640
|
943 |
ret = __kmem_cache_shrink(cachep); |
03afc0e25
|
944 945 946 947 948 |
put_online_mems(); put_online_cpus(); return ret; } EXPORT_SYMBOL(kmem_cache_shrink); |
04f768a39
|
949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 |
/** * kmem_cache_shrink_all - shrink a cache and all memcg caches for root cache * @s: The cache pointer */ void kmem_cache_shrink_all(struct kmem_cache *s) { struct kmem_cache *c; if (!IS_ENABLED(CONFIG_MEMCG_KMEM) || !is_root_cache(s)) { kmem_cache_shrink(s); return; } get_online_cpus(); get_online_mems(); kasan_cache_shrink(s); __kmem_cache_shrink(s); /* * We have to take the slab_mutex to protect from the memcg list * modification. */ mutex_lock(&slab_mutex); for_each_memcg_cache(c, s) { /* * Don't need to shrink deactivated memcg caches. */ if (s->flags & SLAB_DEACTIVATED) continue; kasan_cache_shrink(c); __kmem_cache_shrink(c); } mutex_unlock(&slab_mutex); put_online_mems(); put_online_cpus(); } |
fda901241
|
985 |
bool slab_is_available(void) |
97d066091
|
986 987 988 |
{ return slab_state >= UP; } |
b7454ad3c
|
989 |
|
45530c447
|
990 991 |
#ifndef CONFIG_SLOB /* Create a cache during boot when no slab services are available yet */ |
361d575e5
|
992 993 994 |
void __init create_boot_cache(struct kmem_cache *s, const char *name, unsigned int size, slab_flags_t flags, unsigned int useroffset, unsigned int usersize) |
45530c447
|
995 996 |
{ int err; |
59bb47985
|
997 |
unsigned int align = ARCH_KMALLOC_MINALIGN; |
45530c447
|
998 999 1000 |
s->name = name; s->size = s->object_size = size; |
59bb47985
|
1001 1002 1003 1004 1005 1006 1007 1008 |
/* * For power of two sizes, guarantee natural alignment for kmalloc * caches, regardless of SL*B debugging options. */ if (is_power_of_2(size)) align = max(align, size); s->align = calculate_alignment(flags, align, size); |
8eb8284b4
|
1009 1010 |
s->useroffset = useroffset; s->usersize = usersize; |
f7ce3190c
|
1011 1012 |
slab_init_memcg_params(s); |
45530c447
|
1013 1014 1015 |
err = __kmem_cache_create(s, flags); if (err) |
361d575e5
|
1016 1017 |
panic("Creation of kmalloc slab %s size=%u failed. Reason %d ", |
45530c447
|
1018 1019 1020 1021 |
name, size, err); s->refcount = -1; /* Exempt from merging for now */ } |
55de8b9c6
|
1022 1023 1024 |
struct kmem_cache *__init create_kmalloc_cache(const char *name, unsigned int size, slab_flags_t flags, unsigned int useroffset, unsigned int usersize) |
45530c447
|
1025 1026 1027 1028 1029 1030 |
{ struct kmem_cache *s = kmem_cache_zalloc(kmem_cache, GFP_NOWAIT); if (!s) panic("Out of memory when creating slab %s ", name); |
6c0c21adc
|
1031 |
create_boot_cache(s, name, size, flags, useroffset, usersize); |
45530c447
|
1032 |
list_add(&s->list, &slab_caches); |
c03914b7a
|
1033 |
memcg_link_cache(s, NULL); |
45530c447
|
1034 1035 1036 |
s->refcount = 1; return s; } |
cc252eae8
|
1037 |
struct kmem_cache * |
a07057dce
|
1038 1039 |
kmalloc_caches[NR_KMALLOC_TYPES][KMALLOC_SHIFT_HIGH + 1] __ro_after_init = { /* initialization for https://bugs.llvm.org/show_bug.cgi?id=42570 */ }; |
9425c58e5
|
1040 |
EXPORT_SYMBOL(kmalloc_caches); |
f97d5f634
|
1041 |
/* |
2c59dd654
|
1042 1043 1044 1045 1046 |
* 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. */ |
d5f866550
|
1047 |
static u8 size_index[24] __ro_after_init = { |
2c59dd654
|
1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 |
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 */ }; |
ac914d08b
|
1073 |
static inline unsigned int size_index_elem(unsigned int bytes) |
2c59dd654
|
1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 |
{ return (bytes - 1) / 8; } /* * Find the kmem_cache structure that serves a given size of * allocation */ struct kmem_cache *kmalloc_slab(size_t size, gfp_t flags) { |
d5f866550
|
1084 |
unsigned int index; |
2c59dd654
|
1085 1086 1087 1088 1089 1090 |
if (size <= 192) { if (!size) return ZERO_SIZE_PTR; index = size_index[size_index_elem(size)]; |
61448479a
|
1091 |
} else { |
221d7da66
|
1092 |
if (WARN_ON_ONCE(size > KMALLOC_MAX_CACHE_SIZE)) |
61448479a
|
1093 |
return NULL; |
2c59dd654
|
1094 |
index = fls(size - 1); |
61448479a
|
1095 |
} |
2c59dd654
|
1096 |
|
cc252eae8
|
1097 |
return kmalloc_caches[kmalloc_type(flags)][index]; |
2c59dd654
|
1098 1099 1100 |
} /* |
4066c33d0
|
1101 1102 1103 1104 |
* kmalloc_info[] is to make slub_debug=,kmalloc-xx option work at boot time. * kmalloc_index() supports up to 2^26=64MB, so the final entry of the table is * kmalloc-67108864. */ |
af3b5f876
|
1105 |
const struct kmalloc_info_struct kmalloc_info[] __initconst = { |
4066c33d0
|
1106 1107 1108 1109 1110 |
{NULL, 0}, {"kmalloc-96", 96}, {"kmalloc-192", 192}, {"kmalloc-8", 8}, {"kmalloc-16", 16}, {"kmalloc-32", 32}, {"kmalloc-64", 64}, {"kmalloc-128", 128}, {"kmalloc-256", 256}, {"kmalloc-512", 512}, |
f0d778741
|
1111 1112 1113 1114 1115 1116 1117 1118 1119 |
{"kmalloc-1k", 1024}, {"kmalloc-2k", 2048}, {"kmalloc-4k", 4096}, {"kmalloc-8k", 8192}, {"kmalloc-16k", 16384}, {"kmalloc-32k", 32768}, {"kmalloc-64k", 65536}, {"kmalloc-128k", 131072}, {"kmalloc-256k", 262144}, {"kmalloc-512k", 524288}, {"kmalloc-1M", 1048576}, {"kmalloc-2M", 2097152}, {"kmalloc-4M", 4194304}, {"kmalloc-8M", 8388608}, {"kmalloc-16M", 16777216}, {"kmalloc-32M", 33554432}, {"kmalloc-64M", 67108864} |
4066c33d0
|
1120 1121 1122 |
}; /* |
34cc6990d
|
1123 1124 1125 1126 1127 1128 1129 1130 1131 |
* 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 |
f97d5f634
|
1132 |
*/ |
34cc6990d
|
1133 |
void __init setup_kmalloc_cache_index_table(void) |
f97d5f634
|
1134 |
{ |
ac914d08b
|
1135 |
unsigned int i; |
f97d5f634
|
1136 |
|
2c59dd654
|
1137 1138 1139 1140 |
BUILD_BUG_ON(KMALLOC_MIN_SIZE > 256 || (KMALLOC_MIN_SIZE & (KMALLOC_MIN_SIZE - 1))); for (i = 8; i < KMALLOC_MIN_SIZE; i += 8) { |
ac914d08b
|
1141 |
unsigned int elem = size_index_elem(i); |
2c59dd654
|
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 |
if (elem >= ARRAY_SIZE(size_index)) break; size_index[elem] = KMALLOC_SHIFT_LOW; } if (KMALLOC_MIN_SIZE >= 64) { /* * The 96 byte size cache is not used if the alignment * is 64 byte. */ for (i = 64 + 8; i <= 96; i += 8) size_index[size_index_elem(i)] = 7; } if (KMALLOC_MIN_SIZE >= 128) { /* * The 192 byte sized cache is not used if the alignment * is 128 byte. Redirect kmalloc to use the 256 byte cache * instead. */ for (i = 128 + 8; i <= 192; i += 8) size_index[size_index_elem(i)] = 8; } |
34cc6990d
|
1167 |
} |
f0d778741
|
1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 |
static const char * kmalloc_cache_name(const char *prefix, unsigned int size) { static const char units[3] = "\0kM"; int idx = 0; while (size >= 1024 && (size % 1024 == 0)) { size /= 1024; idx++; } return kasprintf(GFP_NOWAIT, "%s-%u%c", prefix, size, units[idx]); } |
1291523f2
|
1182 1183 |
static void __init new_kmalloc_cache(int idx, int type, slab_flags_t flags) |
a9730fca9
|
1184 |
{ |
1291523f2
|
1185 1186 1187 1188 |
const char *name; if (type == KMALLOC_RECLAIM) { flags |= SLAB_RECLAIM_ACCOUNT; |
f0d778741
|
1189 |
name = kmalloc_cache_name("kmalloc-rcl", |
1291523f2
|
1190 1191 1192 1193 1194 1195 1196 |
kmalloc_info[idx].size); BUG_ON(!name); } else { name = kmalloc_info[idx].name; } kmalloc_caches[type][idx] = create_kmalloc_cache(name, |
6c0c21adc
|
1197 1198 |
kmalloc_info[idx].size, flags, 0, kmalloc_info[idx].size); |
a9730fca9
|
1199 |
} |
34cc6990d
|
1200 1201 1202 1203 1204 |
/* * Create the kmalloc array. Some of the regular kmalloc arrays * may already have been created because they were needed to * enable allocations for slab creation. */ |
d50112edd
|
1205 |
void __init create_kmalloc_caches(slab_flags_t flags) |
34cc6990d
|
1206 |
{ |
1291523f2
|
1207 |
int i, type; |
34cc6990d
|
1208 |
|
1291523f2
|
1209 1210 1211 1212 |
for (type = KMALLOC_NORMAL; type <= KMALLOC_RECLAIM; type++) { for (i = KMALLOC_SHIFT_LOW; i <= KMALLOC_SHIFT_HIGH; i++) { if (!kmalloc_caches[type][i]) new_kmalloc_cache(i, type, flags); |
f97d5f634
|
1213 |
|
1291523f2
|
1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 |
/* * Caches that are not of the two-to-the-power-of size. * These have to be created immediately after the * earlier power of two caches */ if (KMALLOC_MIN_SIZE <= 32 && i == 6 && !kmalloc_caches[type][1]) new_kmalloc_cache(1, type, flags); if (KMALLOC_MIN_SIZE <= 64 && i == 7 && !kmalloc_caches[type][2]) new_kmalloc_cache(2, type, flags); } |
8a965b3ba
|
1226 |
} |
f97d5f634
|
1227 1228 |
/* Kmalloc array is now usable */ slab_state = UP; |
f97d5f634
|
1229 1230 |
#ifdef CONFIG_ZONE_DMA for (i = 0; i <= KMALLOC_SHIFT_HIGH; i++) { |
cc252eae8
|
1231 |
struct kmem_cache *s = kmalloc_caches[KMALLOC_NORMAL][i]; |
f97d5f634
|
1232 1233 |
if (s) { |
0be70327e
|
1234 |
unsigned int size = kmalloc_size(i); |
f0d778741
|
1235 |
const char *n = kmalloc_cache_name("dma-kmalloc", size); |
f97d5f634
|
1236 1237 |
BUG_ON(!n); |
cc252eae8
|
1238 1239 |
kmalloc_caches[KMALLOC_DMA][i] = create_kmalloc_cache( n, size, SLAB_CACHE_DMA | flags, 0, 0); |
f97d5f634
|
1240 1241 1242 1243 |
} } #endif } |
45530c447
|
1244 |
#endif /* !CONFIG_SLOB */ |
cea371f4f
|
1245 1246 1247 1248 1249 |
/* * To avoid unnecessary overhead, we pass through large allocation requests * directly to the page allocator. We use __GFP_COMP, because we will need to * know the allocation order to free the pages properly in kfree. */ |
52383431b
|
1250 1251 |
void *kmalloc_order(size_t size, gfp_t flags, unsigned int order) { |
6a486c0ad
|
1252 |
void *ret = NULL; |
52383431b
|
1253 1254 1255 |
struct page *page; flags |= __GFP_COMP; |
4949148ad
|
1256 |
page = alloc_pages(flags, order); |
6a486c0ad
|
1257 1258 1259 1260 1261 |
if (likely(page)) { ret = page_address(page); mod_node_page_state(page_pgdat(page), NR_SLAB_UNRECLAIMABLE, 1 << order); } |
0116523cf
|
1262 |
ret = kasan_kmalloc_large(ret, size, flags); |
a2f775751
|
1263 |
/* As ret might get tagged, call kmemleak hook after KASAN. */ |
53128245b
|
1264 |
kmemleak_alloc(ret, size, 1, flags); |
52383431b
|
1265 1266 1267 |
return ret; } EXPORT_SYMBOL(kmalloc_order); |
f1b6eb6e6
|
1268 1269 1270 1271 1272 1273 1274 1275 1276 |
#ifdef CONFIG_TRACING void *kmalloc_order_trace(size_t size, gfp_t flags, unsigned int order) { void *ret = kmalloc_order(size, flags, order); trace_kmalloc(_RET_IP_, ret, size, PAGE_SIZE << order, flags); return ret; } EXPORT_SYMBOL(kmalloc_order_trace); #endif |
45530c447
|
1277 |
|
7c00fce98
|
1278 1279 1280 |
#ifdef CONFIG_SLAB_FREELIST_RANDOM /* Randomize a generic freelist */ static void freelist_randomize(struct rnd_state *state, unsigned int *list, |
302d55d51
|
1281 |
unsigned int count) |
7c00fce98
|
1282 |
{ |
7c00fce98
|
1283 |
unsigned int rand; |
302d55d51
|
1284 |
unsigned int i; |
7c00fce98
|
1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 |
for (i = 0; i < count; i++) list[i] = i; /* Fisher-Yates shuffle */ for (i = count - 1; i > 0; i--) { rand = prandom_u32_state(state); rand %= (i + 1); swap(list[i], list[rand]); } } /* Create a random sequence per cache */ int cache_random_seq_create(struct kmem_cache *cachep, unsigned int count, gfp_t gfp) { struct rnd_state state; if (count < 2 || cachep->random_seq) return 0; cachep->random_seq = kcalloc(count, sizeof(unsigned int), gfp); if (!cachep->random_seq) return -ENOMEM; /* Get best entropy at this stage of boot */ prandom_seed_state(&state, get_random_long()); freelist_randomize(&state, cachep->random_seq, count); return 0; } /* Destroy the per-cache random freelist sequence */ void cache_random_seq_destroy(struct kmem_cache *cachep) { kfree(cachep->random_seq); cachep->random_seq = NULL; } #endif /* CONFIG_SLAB_FREELIST_RANDOM */ |
5b3657710
|
1324 |
#if defined(CONFIG_SLAB) || defined(CONFIG_SLUB_DEBUG) |
e9b4db2b8
|
1325 |
#ifdef CONFIG_SLAB |
0825a6f98
|
1326 |
#define SLABINFO_RIGHTS (0600) |
e9b4db2b8
|
1327 |
#else |
0825a6f98
|
1328 |
#define SLABINFO_RIGHTS (0400) |
e9b4db2b8
|
1329 |
#endif |
b047501cd
|
1330 |
static void print_slabinfo_header(struct seq_file *m) |
bcee6e2a1
|
1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 |
{ /* * Output format version, so at least we can change it * without _too_ many complaints. */ #ifdef CONFIG_DEBUG_SLAB seq_puts(m, "slabinfo - version: 2.1 (statistics) "); #else seq_puts(m, "slabinfo - version: 2.1 "); #endif |
756a025f0
|
1343 |
seq_puts(m, "# name <active_objs> <num_objs> <objsize> <objperslab> <pagesperslab>"); |
bcee6e2a1
|
1344 1345 1346 |
seq_puts(m, " : tunables <limit> <batchcount> <sharedfactor>"); seq_puts(m, " : slabdata <active_slabs> <num_slabs> <sharedavail>"); #ifdef CONFIG_DEBUG_SLAB |
756a025f0
|
1347 |
seq_puts(m, " : globalstat <listallocs> <maxobjs> <grown> <reaped> <error> <maxfreeable> <nodeallocs> <remotefrees> <alienoverflow>"); |
bcee6e2a1
|
1348 1349 1350 1351 1352 |
seq_puts(m, " : cpustat <allochit> <allocmiss> <freehit> <freemiss>"); #endif seq_putc(m, ' '); } |
1df3b26f2
|
1353 |
void *slab_start(struct seq_file *m, loff_t *pos) |
b7454ad3c
|
1354 |
{ |
b7454ad3c
|
1355 |
mutex_lock(&slab_mutex); |
510ded33e
|
1356 |
return seq_list_start(&slab_root_caches, *pos); |
b7454ad3c
|
1357 |
} |
276a2439c
|
1358 |
void *slab_next(struct seq_file *m, void *p, loff_t *pos) |
b7454ad3c
|
1359 |
{ |
510ded33e
|
1360 |
return seq_list_next(p, &slab_root_caches, pos); |
b7454ad3c
|
1361 |
} |
276a2439c
|
1362 |
void slab_stop(struct seq_file *m, void *p) |
b7454ad3c
|
1363 1364 1365 |
{ mutex_unlock(&slab_mutex); } |
749c54151
|
1366 1367 1368 1369 1370 |
static void memcg_accumulate_slabinfo(struct kmem_cache *s, struct slabinfo *info) { struct kmem_cache *c; struct slabinfo sinfo; |
749c54151
|
1371 1372 1373 |
if (!is_root_cache(s)) return; |
426589f57
|
1374 |
for_each_memcg_cache(c, s) { |
749c54151
|
1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 |
memset(&sinfo, 0, sizeof(sinfo)); get_slabinfo(c, &sinfo); info->active_slabs += sinfo.active_slabs; info->num_slabs += sinfo.num_slabs; info->shared_avail += sinfo.shared_avail; info->active_objs += sinfo.active_objs; info->num_objs += sinfo.num_objs; } } |
b047501cd
|
1385 |
static void cache_show(struct kmem_cache *s, struct seq_file *m) |
b7454ad3c
|
1386 |
{ |
0d7561c61
|
1387 1388 1389 1390 |
struct slabinfo sinfo; memset(&sinfo, 0, sizeof(sinfo)); get_slabinfo(s, &sinfo); |
749c54151
|
1391 |
memcg_accumulate_slabinfo(s, &sinfo); |
0d7561c61
|
1392 |
seq_printf(m, "%-17s %6lu %6lu %6u %4u %4d", |
749c54151
|
1393 |
cache_name(s), sinfo.active_objs, sinfo.num_objs, s->size, |
0d7561c61
|
1394 1395 1396 1397 1398 1399 1400 1401 1402 |
sinfo.objects_per_slab, (1 << sinfo.cache_order)); seq_printf(m, " : tunables %4u %4u %4u", sinfo.limit, sinfo.batchcount, sinfo.shared); seq_printf(m, " : slabdata %6lu %6lu %6lu", sinfo.active_slabs, sinfo.num_slabs, sinfo.shared_avail); slabinfo_show_stats(m, s); seq_putc(m, ' '); |
b7454ad3c
|
1403 |
} |
1df3b26f2
|
1404 |
static int slab_show(struct seq_file *m, void *p) |
749c54151
|
1405 |
{ |
510ded33e
|
1406 |
struct kmem_cache *s = list_entry(p, struct kmem_cache, root_caches_node); |
749c54151
|
1407 |
|
510ded33e
|
1408 |
if (p == slab_root_caches.next) |
1df3b26f2
|
1409 |
print_slabinfo_header(m); |
510ded33e
|
1410 |
cache_show(s, m); |
b047501cd
|
1411 1412 |
return 0; } |
852d8be0a
|
1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 |
void dump_unreclaimable_slab(void) { struct kmem_cache *s, *s2; struct slabinfo sinfo; /* * Here acquiring slab_mutex is risky since we don't prefer to get * sleep in oom path. But, without mutex hold, it may introduce a * risk of crash. * Use mutex_trylock to protect the list traverse, dump nothing * without acquiring the mutex. */ if (!mutex_trylock(&slab_mutex)) { pr_warn("excessive unreclaimable slab but cannot dump stats "); return; } pr_info("Unreclaimable slab info: "); pr_info("Name Used Total "); list_for_each_entry_safe(s, s2, &slab_caches, list) { if (!is_root_cache(s) || (s->flags & SLAB_RECLAIM_ACCOUNT)) continue; get_slabinfo(s, &sinfo); if (sinfo.num_objs > 0) pr_info("%-17s %10luKB %10luKB ", cache_name(s), (sinfo.active_objs * s->size) / 1024, (sinfo.num_objs * s->size) / 1024); } mutex_unlock(&slab_mutex); } |
5b3657710
|
1450 |
#if defined(CONFIG_MEMCG) |
bc2791f85
|
1451 1452 |
void *memcg_slab_start(struct seq_file *m, loff_t *pos) { |
aa9694bb7
|
1453 |
struct mem_cgroup *memcg = mem_cgroup_from_seq(m); |
bc2791f85
|
1454 1455 1456 1457 1458 1459 1460 |
mutex_lock(&slab_mutex); return seq_list_start(&memcg->kmem_caches, *pos); } void *memcg_slab_next(struct seq_file *m, void *p, loff_t *pos) { |
aa9694bb7
|
1461 |
struct mem_cgroup *memcg = mem_cgroup_from_seq(m); |
bc2791f85
|
1462 1463 1464 1465 1466 1467 1468 1469 |
return seq_list_next(p, &memcg->kmem_caches, pos); } void memcg_slab_stop(struct seq_file *m, void *p) { mutex_unlock(&slab_mutex); } |
b047501cd
|
1470 1471 |
int memcg_slab_show(struct seq_file *m, void *p) { |
bc2791f85
|
1472 1473 |
struct kmem_cache *s = list_entry(p, struct kmem_cache, memcg_params.kmem_caches_node); |
aa9694bb7
|
1474 |
struct mem_cgroup *memcg = mem_cgroup_from_seq(m); |
b047501cd
|
1475 |
|
bc2791f85
|
1476 |
if (p == memcg->kmem_caches.next) |
b047501cd
|
1477 |
print_slabinfo_header(m); |
bc2791f85
|
1478 |
cache_show(s, m); |
b047501cd
|
1479 |
return 0; |
749c54151
|
1480 |
} |
b047501cd
|
1481 |
#endif |
749c54151
|
1482 |
|
b7454ad3c
|
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 |
/* * slabinfo_op - iterator that generates /proc/slabinfo * * Output layout: * cache-name * num-active-objs * total-objs * object size * num-active-slabs * total-slabs * num-pages-per-slab * + further values on SMP and with statistics enabled */ static const struct seq_operations slabinfo_op = { |
1df3b26f2
|
1497 |
.start = slab_start, |
276a2439c
|
1498 1499 |
.next = slab_next, .stop = slab_stop, |
1df3b26f2
|
1500 |
.show = slab_show, |
b7454ad3c
|
1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 |
}; static int slabinfo_open(struct inode *inode, struct file *file) { return seq_open(file, &slabinfo_op); } static const struct file_operations proc_slabinfo_operations = { .open = slabinfo_open, .read = seq_read, .write = slabinfo_write, .llseek = seq_lseek, .release = seq_release, }; static int __init slab_proc_init(void) { |
e9b4db2b8
|
1518 1519 |
proc_create("slabinfo", SLABINFO_RIGHTS, NULL, &proc_slabinfo_operations); |
b7454ad3c
|
1520 1521 1522 |
return 0; } module_init(slab_proc_init); |
fcf8a1e48
|
1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 |
#if defined(CONFIG_DEBUG_FS) && defined(CONFIG_MEMCG_KMEM) /* * Display information about kmem caches that have child memcg caches. */ static int memcg_slabinfo_show(struct seq_file *m, void *unused) { struct kmem_cache *s, *c; struct slabinfo sinfo; mutex_lock(&slab_mutex); seq_puts(m, "# <name> <css_id[:dead|deact]> <active_objs> <num_objs>"); seq_puts(m, " <active_slabs> <num_slabs> "); list_for_each_entry(s, &slab_root_caches, root_caches_node) { /* * Skip kmem caches that don't have any memcg children. */ if (list_empty(&s->memcg_params.children)) continue; memset(&sinfo, 0, sizeof(sinfo)); get_slabinfo(s, &sinfo); seq_printf(m, "%-17s root %6lu %6lu %6lu %6lu ", cache_name(s), sinfo.active_objs, sinfo.num_objs, sinfo.active_slabs, sinfo.num_slabs); for_each_memcg_cache(c, s) { struct cgroup_subsys_state *css; char *status = ""; css = &c->memcg_params.memcg->css; if (!(css->flags & CSS_ONLINE)) status = ":dead"; else if (c->flags & SLAB_DEACTIVATED) status = ":deact"; memset(&sinfo, 0, sizeof(sinfo)); get_slabinfo(c, &sinfo); seq_printf(m, "%-17s %4d%-6s %6lu %6lu %6lu %6lu ", cache_name(c), css->id, status, sinfo.active_objs, sinfo.num_objs, sinfo.active_slabs, sinfo.num_slabs); } } mutex_unlock(&slab_mutex); return 0; } DEFINE_SHOW_ATTRIBUTE(memcg_slabinfo); static int __init memcg_slabinfo_init(void) { debugfs_create_file("memcg_slabinfo", S_IFREG | S_IRUGO, NULL, NULL, &memcg_slabinfo_fops); return 0; } late_initcall(memcg_slabinfo_init); #endif /* CONFIG_DEBUG_FS && CONFIG_MEMCG_KMEM */ |
5b3657710
|
1584 |
#endif /* CONFIG_SLAB || CONFIG_SLUB_DEBUG */ |
928cec9cd
|
1585 1586 1587 1588 1589 1590 1591 1592 1593 |
static __always_inline void *__do_krealloc(const void *p, size_t new_size, gfp_t flags) { void *ret; size_t ks = 0; if (p) ks = ksize(p); |
0316bec22
|
1594 |
if (ks >= new_size) { |
0116523cf
|
1595 |
p = kasan_krealloc((void *)p, new_size, flags); |
928cec9cd
|
1596 |
return (void *)p; |
0316bec22
|
1597 |
} |
928cec9cd
|
1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 |
ret = kmalloc_track_caller(new_size, flags); if (ret && p) memcpy(ret, p, ks); return ret; } /** * __krealloc - like krealloc() but don't free @p. * @p: object to reallocate memory for. * @new_size: how many bytes of memory are required. * @flags: the type of memory to allocate. * * This function is like krealloc() except it never frees the originally * allocated buffer. Use this if you don't want to free the buffer immediately * like, for example, with RCU. |
a862f68a8
|
1615 1616 |
* * Return: pointer to the allocated memory or %NULL in case of error |
928cec9cd
|
1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 |
*/ void *__krealloc(const void *p, size_t new_size, gfp_t flags) { if (unlikely(!new_size)) return ZERO_SIZE_PTR; return __do_krealloc(p, new_size, flags); } EXPORT_SYMBOL(__krealloc); /** * krealloc - reallocate memory. The contents will remain unchanged. * @p: object to reallocate memory for. * @new_size: how many bytes of memory are required. * @flags: the type of memory to allocate. * * The contents of the object pointed to are preserved up to the * lesser of the new and old sizes. If @p is %NULL, krealloc() * behaves exactly like kmalloc(). If @new_size is 0 and @p is not a * %NULL pointer, the object pointed to is freed. |
a862f68a8
|
1638 1639 |
* * Return: pointer to the allocated memory or %NULL in case of error |
928cec9cd
|
1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 |
*/ void *krealloc(const void *p, size_t new_size, gfp_t flags) { void *ret; if (unlikely(!new_size)) { kfree(p); return ZERO_SIZE_PTR; } ret = __do_krealloc(p, new_size, flags); |
772a2fa50
|
1651 |
if (ret && kasan_reset_tag(p) != kasan_reset_tag(ret)) |
928cec9cd
|
1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 |
kfree(p); return ret; } EXPORT_SYMBOL(krealloc); /** * kzfree - like kfree but zero memory * @p: object to free memory of * * The memory of the object @p points to is zeroed before freed. * If @p is %NULL, kzfree() does nothing. * * Note: this function zeroes the whole allocated buffer which can be a good * deal bigger than the requested buffer size passed to kmalloc(). So be * careful when using this function in performance sensitive code. */ void kzfree(const void *p) { size_t ks; void *mem = (void *)p; if (unlikely(ZERO_OR_NULL_PTR(mem))) return; ks = ksize(mem); memset(mem, 0, ks); kfree(mem); } EXPORT_SYMBOL(kzfree); |
10d1f8cb3
|
1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 |
/** * ksize - get the actual amount of memory allocated for a given object * @objp: Pointer to the object * * kmalloc may internally round up allocations and return more memory * than requested. ksize() can be used to determine the actual amount of * memory allocated. The caller may use this additional memory, even though * a smaller amount of memory was initially specified with the kmalloc call. * The caller must guarantee that objp points to a valid object previously * allocated with either kmalloc() or kmem_cache_alloc(). The object * must not be freed during the duration of the call. * * Return: size of the actual memory used by @objp in bytes */ size_t ksize(const void *objp) { |
0d4ca4c9b
|
1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 |
size_t size; if (WARN_ON_ONCE(!objp)) return 0; /* * We need to check that the pointed to object is valid, and only then * unpoison the shadow memory below. We use __kasan_check_read(), to * generate a more useful report at the time ksize() is called (rather * than later where behaviour is undefined due to potential * use-after-free or double-free). * * If the pointed to memory is invalid we return 0, to avoid users of * ksize() writing to and potentially corrupting the memory region. * * We want to perform the check before __ksize(), to avoid potentially * crashing in __ksize() due to accessing invalid metadata. */ if (unlikely(objp == ZERO_SIZE_PTR) || !__kasan_check_read(objp, 1)) return 0; size = __ksize(objp); |
10d1f8cb3
|
1718 1719 1720 1721 1722 1723 1724 1725 |
/* * We assume that ksize callers could use whole allocated area, * so we need to unpoison this area. */ kasan_unpoison_shadow(objp, size); return size; } EXPORT_SYMBOL(ksize); |
928cec9cd
|
1726 1727 1728 1729 1730 1731 1732 |
/* Tracepoints definitions. */ EXPORT_TRACEPOINT_SYMBOL(kmalloc); EXPORT_TRACEPOINT_SYMBOL(kmem_cache_alloc); EXPORT_TRACEPOINT_SYMBOL(kmalloc_node); EXPORT_TRACEPOINT_SYMBOL(kmem_cache_alloc_node); EXPORT_TRACEPOINT_SYMBOL(kfree); EXPORT_TRACEPOINT_SYMBOL(kmem_cache_free); |
4f6923fbb
|
1733 1734 1735 1736 1737 1738 1739 1740 |
int should_failslab(struct kmem_cache *s, gfp_t gfpflags) { if (__should_failslab(s, gfpflags)) return -ENOMEM; return 0; } ALLOW_ERROR_INJECTION(should_failslab, ERRNO); |