Blame view
mm/zswap.c
35.1 KB
c942fddf8
|
1 |
// SPDX-License-Identifier: GPL-2.0-or-later |
2b2811178
|
2 3 4 5 6 7 8 9 10 11 |
/* * zswap.c - zswap driver file * * zswap is a backend for frontswap that takes pages that are in the process * of being swapped out and attempts to compress and store them in a * RAM-based memory pool. This can result in a significant I/O reduction on * the swap device and, in the case where decompressing from RAM is faster * than reading from the swap device, can also improve workload performance. * * Copyright (C) 2012 Seth Jennings <sjenning@linux.vnet.ibm.com> |
2b2811178
|
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 |
*/ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/module.h> #include <linux/cpu.h> #include <linux/highmem.h> #include <linux/slab.h> #include <linux/spinlock.h> #include <linux/types.h> #include <linux/atomic.h> #include <linux/frontswap.h> #include <linux/rbtree.h> #include <linux/swap.h> #include <linux/crypto.h> #include <linux/mempool.h> |
12d79d64b
|
28 |
#include <linux/zpool.h> |
2b2811178
|
29 30 31 32 33 34 |
#include <linux/mm_types.h> #include <linux/page-flags.h> #include <linux/swapops.h> #include <linux/writeback.h> #include <linux/pagemap.h> |
45190f01d
|
35 |
#include <linux/workqueue.h> |
2b2811178
|
36 37 38 39 |
/********************************* * statistics **********************************/ |
12d79d64b
|
40 41 |
/* Total bytes used by the compressed storage */ static u64 zswap_pool_total_size; |
2b2811178
|
42 43 |
/* The number of compressed pages currently stored in zswap */ static atomic_t zswap_stored_pages = ATOMIC_INIT(0); |
a85f878b4
|
44 45 |
/* The number of same-value filled pages currently stored in zswap */ static atomic_t zswap_same_filled_pages = ATOMIC_INIT(0); |
2b2811178
|
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
/* * The statistics below are not protected from concurrent access for * performance reasons so they may not be a 100% accurate. However, * they do provide useful information on roughly how many times a * certain event is occurring. */ /* Pool limit was hit (see zswap_max_pool_percent) */ static u64 zswap_pool_limit_hit; /* Pages written back when pool limit was reached */ static u64 zswap_written_back_pages; /* Store failed due to a reclaim failure after pool limit was reached */ static u64 zswap_reject_reclaim_fail; /* Compressed page was too big for the allocator to (optimally) store */ static u64 zswap_reject_compress_poor; /* Store failed because underlying allocator could not get memory */ static u64 zswap_reject_alloc_fail; /* Store failed because the entry metadata could not be allocated (rare) */ static u64 zswap_reject_kmemcache_fail; /* Duplicate store was encountered (rare) */ static u64 zswap_duplicate_entry; |
45190f01d
|
68 69 70 71 |
/* Shrinker work queue */ static struct workqueue_struct *shrink_wq; /* Pool limit was hit, we need to calm down */ static bool zswap_pool_reached_full; |
2b2811178
|
72 73 74 |
/********************************* * tunables **********************************/ |
c00ed16a9
|
75 |
|
bae21db88
|
76 |
#define ZSWAP_PARAM_UNSET "" |
bb8b93b5b
|
77 78 |
/* Enable/disable zswap */ static bool zswap_enabled = IS_ENABLED(CONFIG_ZSWAP_DEFAULT_ON); |
d7b028f56
|
79 80 81 82 83 84 85 |
static int zswap_enabled_param_set(const char *, const struct kernel_param *); static struct kernel_param_ops zswap_enabled_param_ops = { .set = zswap_enabled_param_set, .get = param_get_bool, }; module_param_cb(enabled, &zswap_enabled_param_ops, &zswap_enabled, 0644); |
2b2811178
|
86 |
|
90b0fc26d
|
87 |
/* Crypto compressor to use */ |
bb8b93b5b
|
88 |
static char *zswap_compressor = CONFIG_ZSWAP_COMPRESSOR_DEFAULT; |
90b0fc26d
|
89 90 91 92 |
static int zswap_compressor_param_set(const char *, const struct kernel_param *); static struct kernel_param_ops zswap_compressor_param_ops = { .set = zswap_compressor_param_set, |
c99b42c35
|
93 94 |
.get = param_get_charp, .free = param_free_charp, |
90b0fc26d
|
95 96 |
}; module_param_cb(compressor, &zswap_compressor_param_ops, |
c99b42c35
|
97 |
&zswap_compressor, 0644); |
2b2811178
|
98 |
|
90b0fc26d
|
99 |
/* Compressed storage zpool to use */ |
bb8b93b5b
|
100 |
static char *zswap_zpool_type = CONFIG_ZSWAP_ZPOOL_DEFAULT; |
90b0fc26d
|
101 102 |
static int zswap_zpool_param_set(const char *, const struct kernel_param *); static struct kernel_param_ops zswap_zpool_param_ops = { |
c99b42c35
|
103 104 105 |
.set = zswap_zpool_param_set, .get = param_get_charp, .free = param_free_charp, |
90b0fc26d
|
106 |
}; |
c99b42c35
|
107 |
module_param_cb(zpool, &zswap_zpool_param_ops, &zswap_zpool_type, 0644); |
12d79d64b
|
108 |
|
90b0fc26d
|
109 110 111 |
/* The maximum percentage of memory that the compressed pool can occupy */ static unsigned int zswap_max_pool_percent = 20; module_param_named(max_pool_percent, zswap_max_pool_percent, uint, 0644); |
60105e124
|
112 |
|
45190f01d
|
113 114 115 116 |
/* The threshold for accepting new pages after the max_pool_percent was hit */ static unsigned int zswap_accept_thr_percent = 90; /* of max pool size */ module_param_named(accept_threshold_percent, zswap_accept_thr_percent, uint, 0644); |
a85f878b4
|
117 118 119 120 |
/* Enable/disable handling same-value filled pages (enabled by default) */ static bool zswap_same_filled_pages_enabled = true; module_param_named(same_filled_pages_enabled, zswap_same_filled_pages_enabled, bool, 0644); |
2b2811178
|
121 |
/********************************* |
f1c54846e
|
122 |
* data structures |
2b2811178
|
123 |
**********************************/ |
2b2811178
|
124 |
|
f1c54846e
|
125 126 127 128 129 |
struct zswap_pool { struct zpool *zpool; struct crypto_comp * __percpu *tfm; struct kref kref; struct list_head list; |
45190f01d
|
130 131 |
struct work_struct release_work; struct work_struct shrink_work; |
cab7a7e5b
|
132 |
struct hlist_node node; |
f1c54846e
|
133 |
char tfm_name[CRYPTO_MAX_ALG_NAME]; |
2b2811178
|
134 |
}; |
2b2811178
|
135 136 137 138 139 140 141 |
/* * struct zswap_entry * * This structure contains the metadata for tracking a single compressed * page within zswap. * * rbnode - links the entry into red-black tree for the appropriate swap type |
f1c54846e
|
142 |
* offset - the swap offset for the entry. Index into the red-black tree. |
2b2811178
|
143 144 |
* refcount - the number of outstanding reference to the entry. This is needed * to protect against premature freeing of the entry by code |
6b4525164
|
145 |
* concurrent calls to load, invalidate, and writeback. The lock |
2b2811178
|
146 147 148 |
* for the zswap_tree structure that contains the entry must * be held while changing the refcount. Since the lock must * be held, there is no reason to also make refcount atomic. |
2b2811178
|
149 |
* length - the length in bytes of the compressed page data. Needed during |
a85f878b4
|
150 |
* decompression. For a same value filled page length is 0. |
f1c54846e
|
151 152 |
* pool - the zswap_pool the entry's data is in * handle - zpool allocation handle that stores the compressed page data |
a85f878b4
|
153 |
* value - value of the same-value filled pages which have same content |
2b2811178
|
154 155 156 157 158 159 |
*/ struct zswap_entry { struct rb_node rbnode; pgoff_t offset; int refcount; unsigned int length; |
f1c54846e
|
160 |
struct zswap_pool *pool; |
a85f878b4
|
161 162 163 164 |
union { unsigned long handle; unsigned long value; }; |
2b2811178
|
165 166 167 168 169 170 171 172 173 174 175 176 177 178 |
}; struct zswap_header { swp_entry_t swpentry; }; /* * The tree lock in the zswap_tree struct protects a few things: * - the rbtree * - the refcount field of each entry in the tree */ struct zswap_tree { struct rb_root rbroot; spinlock_t lock; |
2b2811178
|
179 180 181 |
}; static struct zswap_tree *zswap_trees[MAX_SWAPFILES]; |
f1c54846e
|
182 183 184 185 |
/* RCU-protected iteration */ static LIST_HEAD(zswap_pools); /* protects zswap_pools list modification */ static DEFINE_SPINLOCK(zswap_pools_lock); |
32a4e1690
|
186 187 |
/* pool counter to provide unique names to zpool */ static atomic_t zswap_pools_count = ATOMIC_INIT(0); |
f1c54846e
|
188 |
|
90b0fc26d
|
189 190 |
/* used by param callback function */ static bool zswap_init_started; |
d7b028f56
|
191 192 |
/* fatal error during init */ static bool zswap_init_failed; |
ae3d89a7e
|
193 194 |
/* init completed, but couldn't create the initial pool */ static bool zswap_has_pool; |
f1c54846e
|
195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 |
/********************************* * helpers and fwd declarations **********************************/ #define zswap_pool_debug(msg, p) \ pr_debug("%s pool %s/%s ", msg, (p)->tfm_name, \ zpool_get_type((p)->zpool)) static int zswap_writeback_entry(struct zpool *pool, unsigned long handle); static int zswap_pool_get(struct zswap_pool *pool); static void zswap_pool_put(struct zswap_pool *pool); static const struct zpool_ops zswap_zpool_ops = { .evict = zswap_writeback_entry }; static bool zswap_is_full(void) { |
ca79b0c21
|
214 215 |
return totalram_pages() * zswap_max_pool_percent / 100 < DIV_ROUND_UP(zswap_pool_total_size, PAGE_SIZE); |
f1c54846e
|
216 |
} |
45190f01d
|
217 218 219 220 221 222 |
static bool zswap_can_accept(void) { return totalram_pages() * zswap_accept_thr_percent / 100 * zswap_max_pool_percent / 100 > DIV_ROUND_UP(zswap_pool_total_size, PAGE_SIZE); } |
f1c54846e
|
223 224 225 226 227 228 229 230 231 232 233 234 235 236 |
static void zswap_update_total_size(void) { struct zswap_pool *pool; u64 total = 0; rcu_read_lock(); list_for_each_entry_rcu(pool, &zswap_pools, list) total += zpool_get_total_size(pool->zpool); rcu_read_unlock(); zswap_pool_total_size = total; } |
2b2811178
|
237 238 239 240 |
/********************************* * zswap entry functions **********************************/ static struct kmem_cache *zswap_entry_cache; |
dd01d7d89
|
241 |
static int __init zswap_entry_cache_create(void) |
2b2811178
|
242 243 |
{ zswap_entry_cache = KMEM_CACHE(zswap_entry, 0); |
5d2d42de1
|
244 |
return zswap_entry_cache == NULL; |
2b2811178
|
245 |
} |
c119239b1
|
246 |
static void __init zswap_entry_cache_destroy(void) |
2b2811178
|
247 248 249 250 251 252 253 254 255 256 257 |
{ kmem_cache_destroy(zswap_entry_cache); } static struct zswap_entry *zswap_entry_cache_alloc(gfp_t gfp) { struct zswap_entry *entry; entry = kmem_cache_alloc(zswap_entry_cache, gfp); if (!entry) return NULL; entry->refcount = 1; |
0ab0abcf5
|
258 |
RB_CLEAR_NODE(&entry->rbnode); |
2b2811178
|
259 260 261 262 263 264 265 |
return entry; } static void zswap_entry_cache_free(struct zswap_entry *entry) { kmem_cache_free(zswap_entry_cache, entry); } |
2b2811178
|
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 |
/********************************* * rbtree functions **********************************/ static struct zswap_entry *zswap_rb_search(struct rb_root *root, pgoff_t offset) { struct rb_node *node = root->rb_node; struct zswap_entry *entry; while (node) { entry = rb_entry(node, struct zswap_entry, rbnode); if (entry->offset > offset) node = node->rb_left; else if (entry->offset < offset) node = node->rb_right; else return entry; } return NULL; } /* * In the case that a entry with the same offset is found, a pointer to * the existing entry is stored in dupentry and the function returns -EEXIST */ static int zswap_rb_insert(struct rb_root *root, struct zswap_entry *entry, struct zswap_entry **dupentry) { struct rb_node **link = &root->rb_node, *parent = NULL; struct zswap_entry *myentry; while (*link) { parent = *link; myentry = rb_entry(parent, struct zswap_entry, rbnode); if (myentry->offset > entry->offset) link = &(*link)->rb_left; else if (myentry->offset < entry->offset) link = &(*link)->rb_right; else { *dupentry = myentry; return -EEXIST; } } rb_link_node(&entry->rbnode, parent, link); rb_insert_color(&entry->rbnode, root); return 0; } |
0ab0abcf5
|
312 313 314 315 316 317 318 319 320 |
static void zswap_rb_erase(struct rb_root *root, struct zswap_entry *entry) { if (!RB_EMPTY_NODE(&entry->rbnode)) { rb_erase(&entry->rbnode, root); RB_CLEAR_NODE(&entry->rbnode); } } /* |
12d79d64b
|
321 |
* Carries out the common pattern of freeing and entry's zpool allocation, |
0ab0abcf5
|
322 323 |
* freeing the entry itself, and decrementing the number of stored pages. */ |
60105e124
|
324 |
static void zswap_free_entry(struct zswap_entry *entry) |
0ab0abcf5
|
325 |
{ |
a85f878b4
|
326 327 328 329 330 331 |
if (!entry->length) atomic_dec(&zswap_same_filled_pages); else { zpool_free(entry->pool->zpool, entry->handle); zswap_pool_put(entry->pool); } |
0ab0abcf5
|
332 333 |
zswap_entry_cache_free(entry); atomic_dec(&zswap_stored_pages); |
f1c54846e
|
334 |
zswap_update_total_size(); |
0ab0abcf5
|
335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 |
} /* caller must hold the tree lock */ static void zswap_entry_get(struct zswap_entry *entry) { entry->refcount++; } /* caller must hold the tree lock * remove from the tree and free it, if nobody reference the entry */ static void zswap_entry_put(struct zswap_tree *tree, struct zswap_entry *entry) { int refcount = --entry->refcount; BUG_ON(refcount < 0); if (refcount == 0) { zswap_rb_erase(&tree->rbroot, entry); |
60105e124
|
354 |
zswap_free_entry(entry); |
0ab0abcf5
|
355 356 357 358 359 360 361 |
} } /* caller must hold the tree lock */ static struct zswap_entry *zswap_entry_find_get(struct rb_root *root, pgoff_t offset) { |
b0c9865fd
|
362 |
struct zswap_entry *entry; |
0ab0abcf5
|
363 364 365 366 367 368 369 |
entry = zswap_rb_search(root, offset); if (entry) zswap_entry_get(entry); return entry; } |
2b2811178
|
370 371 372 373 |
/********************************* * per-cpu code **********************************/ static DEFINE_PER_CPU(u8 *, zswap_dstmem); |
ad7ed7708
|
374 |
static int zswap_dstmem_prepare(unsigned int cpu) |
2b2811178
|
375 |
{ |
2b2811178
|
376 |
u8 *dst; |
ad7ed7708
|
377 |
dst = kmalloc_node(PAGE_SIZE * 2, GFP_KERNEL, cpu_to_node(cpu)); |
2b2695f5f
|
378 |
if (!dst) |
ad7ed7708
|
379 |
return -ENOMEM; |
2b2695f5f
|
380 |
|
ad7ed7708
|
381 382 |
per_cpu(zswap_dstmem, cpu) = dst; return 0; |
2b2811178
|
383 |
} |
ad7ed7708
|
384 |
static int zswap_dstmem_dead(unsigned int cpu) |
2b2811178
|
385 |
{ |
ad7ed7708
|
386 |
u8 *dst; |
2b2811178
|
387 |
|
ad7ed7708
|
388 389 390 |
dst = per_cpu(zswap_dstmem, cpu); kfree(dst); per_cpu(zswap_dstmem, cpu) = NULL; |
f1c54846e
|
391 |
|
f1c54846e
|
392 |
return 0; |
f1c54846e
|
393 |
} |
cab7a7e5b
|
394 |
static int zswap_cpu_comp_prepare(unsigned int cpu, struct hlist_node *node) |
f1c54846e
|
395 |
{ |
cab7a7e5b
|
396 |
struct zswap_pool *pool = hlist_entry(node, struct zswap_pool, node); |
f1c54846e
|
397 |
struct crypto_comp *tfm; |
cab7a7e5b
|
398 399 |
if (WARN_ON(*per_cpu_ptr(pool->tfm, cpu))) return 0; |
f1c54846e
|
400 |
|
cab7a7e5b
|
401 402 403 404 405 406 407 408 |
tfm = crypto_alloc_comp(pool->tfm_name, 0, 0); if (IS_ERR_OR_NULL(tfm)) { pr_err("could not alloc crypto comp %s : %ld ", pool->tfm_name, PTR_ERR(tfm)); return -ENOMEM; } *per_cpu_ptr(pool->tfm, cpu) = tfm; |
2b2811178
|
409 |
return 0; |
2b2811178
|
410 |
} |
cab7a7e5b
|
411 |
static int zswap_cpu_comp_dead(unsigned int cpu, struct hlist_node *node) |
f1c54846e
|
412 |
{ |
cab7a7e5b
|
413 414 |
struct zswap_pool *pool = hlist_entry(node, struct zswap_pool, node); struct crypto_comp *tfm; |
f1c54846e
|
415 |
|
cab7a7e5b
|
416 417 418 419 420 |
tfm = *per_cpu_ptr(pool->tfm, cpu); if (!IS_ERR_OR_NULL(tfm)) crypto_free_comp(tfm); *per_cpu_ptr(pool->tfm, cpu) = NULL; return 0; |
f1c54846e
|
421 |
} |
2b2811178
|
422 |
/********************************* |
f1c54846e
|
423 |
* pool functions |
2b2811178
|
424 |
**********************************/ |
f1c54846e
|
425 426 |
static struct zswap_pool *__zswap_pool_current(void) |
2b2811178
|
427 |
{ |
f1c54846e
|
428 429 430 |
struct zswap_pool *pool; pool = list_first_or_null_rcu(&zswap_pools, typeof(*pool), list); |
ae3d89a7e
|
431 432 433 |
WARN_ONCE(!pool && zswap_has_pool, "%s: no page storage pool! ", __func__); |
f1c54846e
|
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 |
return pool; } static struct zswap_pool *zswap_pool_current(void) { assert_spin_locked(&zswap_pools_lock); return __zswap_pool_current(); } static struct zswap_pool *zswap_pool_current_get(void) { struct zswap_pool *pool; rcu_read_lock(); pool = __zswap_pool_current(); |
ae3d89a7e
|
452 |
if (!zswap_pool_get(pool)) |
f1c54846e
|
453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 |
pool = NULL; rcu_read_unlock(); return pool; } static struct zswap_pool *zswap_pool_last_get(void) { struct zswap_pool *pool, *last = NULL; rcu_read_lock(); list_for_each_entry_rcu(pool, &zswap_pools, list) last = pool; |
ae3d89a7e
|
468 469 470 471 |
WARN_ONCE(!last && zswap_has_pool, "%s: no page storage pool! ", __func__); if (!zswap_pool_get(last)) |
f1c54846e
|
472 473 474 475 476 477 |
last = NULL; rcu_read_unlock(); return last; } |
8bc8b228d
|
478 |
/* type and compressor must be null-terminated */ |
f1c54846e
|
479 480 481 482 483 484 485 |
static struct zswap_pool *zswap_pool_find_get(char *type, char *compressor) { struct zswap_pool *pool; assert_spin_locked(&zswap_pools_lock); list_for_each_entry_rcu(pool, &zswap_pools, list) { |
8bc8b228d
|
486 |
if (strcmp(pool->tfm_name, compressor)) |
f1c54846e
|
487 |
continue; |
8bc8b228d
|
488 |
if (strcmp(zpool_get_type(pool->zpool), type)) |
f1c54846e
|
489 490 491 492 493 494 495 496 497 |
continue; /* if we can't get it, it's about to be destroyed */ if (!zswap_pool_get(pool)) continue; return pool; } return NULL; } |
45190f01d
|
498 499 500 501 502 503 504 505 506 |
static void shrink_worker(struct work_struct *w) { struct zswap_pool *pool = container_of(w, typeof(*pool), shrink_work); if (zpool_shrink(pool->zpool, 1, NULL)) zswap_reject_reclaim_fail++; zswap_pool_put(pool); } |
f1c54846e
|
507 508 509 |
static struct zswap_pool *zswap_pool_create(char *type, char *compressor) { struct zswap_pool *pool; |
32a4e1690
|
510 |
char name[38]; /* 'zswap' + 32 char (max) num + \0 */ |
d0164adc8
|
511 |
gfp_t gfp = __GFP_NORETRY | __GFP_NOWARN | __GFP_KSWAPD_RECLAIM; |
cab7a7e5b
|
512 |
int ret; |
f1c54846e
|
513 |
|
bae21db88
|
514 515 516 517 518 519 520 521 522 523 |
if (!zswap_has_pool) { /* if either are unset, pool initialization failed, and we * need both params to be set correctly before trying to * create a pool. */ if (!strcmp(type, ZSWAP_PARAM_UNSET)) return NULL; if (!strcmp(compressor, ZSWAP_PARAM_UNSET)) return NULL; } |
f1c54846e
|
524 |
pool = kzalloc(sizeof(*pool), GFP_KERNEL); |
f4ae0ce0f
|
525 |
if (!pool) |
f1c54846e
|
526 |
return NULL; |
f1c54846e
|
527 |
|
32a4e1690
|
528 529 530 531 |
/* unique name for each pool specifically required by zsmalloc */ snprintf(name, 38, "zswap%x", atomic_inc_return(&zswap_pools_count)); pool->zpool = zpool_create_pool(type, name, gfp, &zswap_zpool_ops); |
f1c54846e
|
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 |
if (!pool->zpool) { pr_err("%s zpool not available ", type); goto error; } pr_debug("using %s zpool ", zpool_get_type(pool->zpool)); strlcpy(pool->tfm_name, compressor, sizeof(pool->tfm_name)); pool->tfm = alloc_percpu(struct crypto_comp *); if (!pool->tfm) { pr_err("percpu alloc failed "); goto error; } |
cab7a7e5b
|
547 548 549 |
ret = cpuhp_state_add_instance(CPUHP_MM_ZSWP_POOL_PREPARE, &pool->node); if (ret) |
f1c54846e
|
550 551 552 553 554 555 556 557 558 |
goto error; pr_debug("using %s compressor ", pool->tfm_name); /* being the current pool takes 1 ref; this func expects the * caller to always add the new pool as the current pool */ kref_init(&pool->kref); INIT_LIST_HEAD(&pool->list); |
45190f01d
|
559 |
INIT_WORK(&pool->shrink_work, shrink_worker); |
f1c54846e
|
560 561 562 563 564 565 566 567 568 569 570 571 |
zswap_pool_debug("created", pool); return pool; error: free_percpu(pool->tfm); if (pool->zpool) zpool_destroy_pool(pool->zpool); kfree(pool); return NULL; } |
c99b42c35
|
572 |
static __init struct zswap_pool *__zswap_pool_create_fallback(void) |
f1c54846e
|
573 |
{ |
bae21db88
|
574 575 576 |
bool has_comp, has_zpool; has_comp = crypto_has_comp(zswap_compressor, 0, 0); |
bb8b93b5b
|
577 578 |
if (!has_comp && strcmp(zswap_compressor, CONFIG_ZSWAP_COMPRESSOR_DEFAULT)) { |
f1c54846e
|
579 580 |
pr_err("compressor %s not available, using default %s ", |
bb8b93b5b
|
581 |
zswap_compressor, CONFIG_ZSWAP_COMPRESSOR_DEFAULT); |
c99b42c35
|
582 |
param_free_charp(&zswap_compressor); |
bb8b93b5b
|
583 |
zswap_compressor = CONFIG_ZSWAP_COMPRESSOR_DEFAULT; |
bae21db88
|
584 |
has_comp = crypto_has_comp(zswap_compressor, 0, 0); |
f1c54846e
|
585 |
} |
bae21db88
|
586 587 588 589 590 591 592 593 594 |
if (!has_comp) { pr_err("default compressor %s not available ", zswap_compressor); param_free_charp(&zswap_compressor); zswap_compressor = ZSWAP_PARAM_UNSET; } has_zpool = zpool_has_pool(zswap_zpool_type); |
bb8b93b5b
|
595 596 |
if (!has_zpool && strcmp(zswap_zpool_type, CONFIG_ZSWAP_ZPOOL_DEFAULT)) { |
f1c54846e
|
597 598 |
pr_err("zpool %s not available, using default %s ", |
bb8b93b5b
|
599 |
zswap_zpool_type, CONFIG_ZSWAP_ZPOOL_DEFAULT); |
c99b42c35
|
600 |
param_free_charp(&zswap_zpool_type); |
bb8b93b5b
|
601 |
zswap_zpool_type = CONFIG_ZSWAP_ZPOOL_DEFAULT; |
bae21db88
|
602 |
has_zpool = zpool_has_pool(zswap_zpool_type); |
f1c54846e
|
603 |
} |
bae21db88
|
604 605 606 607 608 609 610 611 612 613 |
if (!has_zpool) { pr_err("default zpool %s not available ", zswap_zpool_type); param_free_charp(&zswap_zpool_type); zswap_zpool_type = ZSWAP_PARAM_UNSET; } if (!has_comp || !has_zpool) return NULL; |
f1c54846e
|
614 615 616 617 618 619 620 |
return zswap_pool_create(zswap_zpool_type, zswap_compressor); } static void zswap_pool_destroy(struct zswap_pool *pool) { zswap_pool_debug("destroying", pool); |
cab7a7e5b
|
621 |
cpuhp_state_remove_instance(CPUHP_MM_ZSWP_POOL_PREPARE, &pool->node); |
f1c54846e
|
622 623 624 625 626 627 628 |
free_percpu(pool->tfm); zpool_destroy_pool(pool->zpool); kfree(pool); } static int __must_check zswap_pool_get(struct zswap_pool *pool) { |
ae3d89a7e
|
629 630 |
if (!pool) return 0; |
f1c54846e
|
631 632 |
return kref_get_unless_zero(&pool->kref); } |
200867af4
|
633 |
static void __zswap_pool_release(struct work_struct *work) |
f1c54846e
|
634 |
{ |
45190f01d
|
635 636 |
struct zswap_pool *pool = container_of(work, typeof(*pool), release_work); |
200867af4
|
637 638 |
synchronize_rcu(); |
f1c54846e
|
639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 |
/* nobody should have been able to get a kref... */ WARN_ON(kref_get_unless_zero(&pool->kref)); /* pool is now off zswap_pools list and has no references. */ zswap_pool_destroy(pool); } static void __zswap_pool_empty(struct kref *kref) { struct zswap_pool *pool; pool = container_of(kref, typeof(*pool), kref); spin_lock(&zswap_pools_lock); WARN_ON(pool == zswap_pool_current()); list_del_rcu(&pool->list); |
200867af4
|
658 |
|
45190f01d
|
659 660 |
INIT_WORK(&pool->release_work, __zswap_pool_release); schedule_work(&pool->release_work); |
f1c54846e
|
661 662 663 664 665 666 667 |
spin_unlock(&zswap_pools_lock); } static void zswap_pool_put(struct zswap_pool *pool) { kref_put(&pool->kref, __zswap_pool_empty); |
2b2811178
|
668 |
} |
2b2811178
|
669 |
/********************************* |
90b0fc26d
|
670 671 |
* param callbacks **********************************/ |
c99b42c35
|
672 |
/* val must be a null-terminated string */ |
90b0fc26d
|
673 674 675 676 |
static int __zswap_param_set(const char *val, const struct kernel_param *kp, char *type, char *compressor) { struct zswap_pool *pool, *put_pool = NULL; |
c99b42c35
|
677 |
char *s = strstrip((char *)val); |
90b0fc26d
|
678 |
int ret; |
d7b028f56
|
679 680 681 682 683 |
if (zswap_init_failed) { pr_err("can't set param, initialization failed "); return -ENODEV; } |
c99b42c35
|
684 |
/* no change required */ |
ae3d89a7e
|
685 |
if (!strcmp(s, *(char **)kp->arg) && zswap_has_pool) |
c99b42c35
|
686 |
return 0; |
90b0fc26d
|
687 688 689 690 691 |
/* if this is load-time (pre-init) param setting, * don't create a pool; that's done during init. */ if (!zswap_init_started) |
c99b42c35
|
692 |
return param_set_charp(s, kp); |
90b0fc26d
|
693 694 |
if (!type) { |
c99b42c35
|
695 696 697 |
if (!zpool_has_pool(s)) { pr_err("zpool %s not available ", s); |
90b0fc26d
|
698 699 |
return -ENOENT; } |
c99b42c35
|
700 |
type = s; |
90b0fc26d
|
701 |
} else if (!compressor) { |
c99b42c35
|
702 703 704 |
if (!crypto_has_comp(s, 0, 0)) { pr_err("compressor %s not available ", s); |
90b0fc26d
|
705 706 |
return -ENOENT; } |
c99b42c35
|
707 708 709 710 |
compressor = s; } else { WARN_ON(1); return -EINVAL; |
90b0fc26d
|
711 712 713 714 715 716 717 |
} spin_lock(&zswap_pools_lock); pool = zswap_pool_find_get(type, compressor); if (pool) { zswap_pool_debug("using existing", pool); |
fd5bb66cd
|
718 |
WARN_ON(pool == zswap_pool_current()); |
90b0fc26d
|
719 |
list_del_rcu(&pool->list); |
90b0fc26d
|
720 |
} |
fd5bb66cd
|
721 722 723 724 |
spin_unlock(&zswap_pools_lock); if (!pool) pool = zswap_pool_create(type, compressor); |
90b0fc26d
|
725 |
if (pool) |
c99b42c35
|
726 |
ret = param_set_charp(s, kp); |
90b0fc26d
|
727 728 |
else ret = -EINVAL; |
fd5bb66cd
|
729 |
spin_lock(&zswap_pools_lock); |
90b0fc26d
|
730 731 732 |
if (!ret) { put_pool = zswap_pool_current(); list_add_rcu(&pool->list, &zswap_pools); |
ae3d89a7e
|
733 |
zswap_has_pool = true; |
90b0fc26d
|
734 735 736 737 738 739 740 |
} else if (pool) { /* add the possibly pre-existing pool to the end of the pools * list; if it's new (and empty) then it'll be removed and * destroyed by the put after we drop the lock */ list_add_tail_rcu(&pool->list, &zswap_pools); put_pool = pool; |
fd5bb66cd
|
741 742 743 744 745 |
} spin_unlock(&zswap_pools_lock); if (!zswap_has_pool && !pool) { |
ae3d89a7e
|
746 747 748 749 750 751 752 753 |
/* if initial pool creation failed, and this pool creation also * failed, maybe both compressor and zpool params were bad. * Allow changing this param, so pool creation will succeed * when the other param is changed. We already verified this * param is ok in the zpool_has_pool() or crypto_has_comp() * checks above. */ ret = param_set_charp(s, kp); |
90b0fc26d
|
754 |
} |
90b0fc26d
|
755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 |
/* drop the ref from either the old current pool, * or the new pool we failed to add */ if (put_pool) zswap_pool_put(put_pool); return ret; } static int zswap_compressor_param_set(const char *val, const struct kernel_param *kp) { return __zswap_param_set(val, kp, zswap_zpool_type, NULL); } static int zswap_zpool_param_set(const char *val, const struct kernel_param *kp) { return __zswap_param_set(val, kp, NULL, zswap_compressor); } |
d7b028f56
|
775 776 777 778 779 780 781 782 |
static int zswap_enabled_param_set(const char *val, const struct kernel_param *kp) { if (zswap_init_failed) { pr_err("can't enable, initialization failed "); return -ENODEV; } |
ae3d89a7e
|
783 784 785 786 787 |
if (!zswap_has_pool && zswap_init_started) { pr_err("can't enable, no pool configured "); return -ENODEV; } |
d7b028f56
|
788 789 790 |
return param_set_bool(val, kp); } |
90b0fc26d
|
791 |
/********************************* |
2b2811178
|
792 793 794 795 796 797 |
* writeback code **********************************/ /* return enum for zswap_get_swap_cache_page */ enum zswap_get_swap_ret { ZSWAP_SWAPCACHE_NEW, ZSWAP_SWAPCACHE_EXIST, |
67d13fe84
|
798 |
ZSWAP_SWAPCACHE_FAIL, |
2b2811178
|
799 800 801 802 803 804 805 806 807 808 809 810 811 |
}; /* * zswap_get_swap_cache_page * * This is an adaption of read_swap_cache_async() * * This function tries to find a page with the given swap entry * in the swapper_space address space (the swap cache). If the page * is found, it is returned in retpage. Otherwise, a page is allocated, * added to the swap cache, and returned in retpage. * * If success, the swap cache page is returned in retpage |
67d13fe84
|
812 813 814 815 |
* Returns ZSWAP_SWAPCACHE_EXIST if page was already in the swap cache * Returns ZSWAP_SWAPCACHE_NEW if the new page needs to be populated, * the new page is added to swapcache and locked * Returns ZSWAP_SWAPCACHE_FAIL on error |
2b2811178
|
816 817 818 819 |
*/ static int zswap_get_swap_cache_page(swp_entry_t entry, struct page **retpage) { |
5b999aadb
|
820 |
bool page_was_allocated; |
2b2811178
|
821 |
|
5b999aadb
|
822 823 824 825 826 |
*retpage = __read_swap_cache_async(entry, GFP_KERNEL, NULL, 0, &page_was_allocated); if (page_was_allocated) return ZSWAP_SWAPCACHE_NEW; if (!*retpage) |
67d13fe84
|
827 |
return ZSWAP_SWAPCACHE_FAIL; |
2b2811178
|
828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 |
return ZSWAP_SWAPCACHE_EXIST; } /* * Attempts to free an entry by adding a page to the swap cache, * decompressing the entry data into the page, and issuing a * bio write to write the page back to the swap device. * * This can be thought of as a "resumed writeback" of the page * to the swap device. We are basically resuming the same swap * writeback path that was intercepted with the frontswap_store() * in the first place. After the page has been decompressed into * the swap cache, the compressed version stored by zswap can be * freed. */ |
12d79d64b
|
843 |
static int zswap_writeback_entry(struct zpool *pool, unsigned long handle) |
2b2811178
|
844 845 846 847 848 849 850 |
{ struct zswap_header *zhdr; swp_entry_t swpentry; struct zswap_tree *tree; pgoff_t offset; struct zswap_entry *entry; struct page *page; |
f1c54846e
|
851 |
struct crypto_comp *tfm; |
2b2811178
|
852 853 |
u8 *src, *dst; unsigned int dlen; |
0ab0abcf5
|
854 |
int ret; |
2b2811178
|
855 856 857 858 859 |
struct writeback_control wbc = { .sync_mode = WB_SYNC_NONE, }; /* extract swpentry from data */ |
12d79d64b
|
860 |
zhdr = zpool_map_handle(pool, handle, ZPOOL_MM_RO); |
2b2811178
|
861 |
swpentry = zhdr->swpentry; /* here */ |
2b2811178
|
862 863 |
tree = zswap_trees[swp_type(swpentry)]; offset = swp_offset(swpentry); |
2b2811178
|
864 865 866 |
/* find and ref zswap entry */ spin_lock(&tree->lock); |
0ab0abcf5
|
867 |
entry = zswap_entry_find_get(&tree->rbroot, offset); |
2b2811178
|
868 869 870 |
if (!entry) { /* entry was invalidated */ spin_unlock(&tree->lock); |
068619e32
|
871 |
zpool_unmap_handle(pool, handle); |
2b2811178
|
872 873 |
return 0; } |
2b2811178
|
874 875 876 877 878 |
spin_unlock(&tree->lock); BUG_ON(offset != entry->offset); /* try to allocate swap cache page */ switch (zswap_get_swap_cache_page(swpentry, &page)) { |
67d13fe84
|
879 |
case ZSWAP_SWAPCACHE_FAIL: /* no memory or invalidate happened */ |
2b2811178
|
880 881 |
ret = -ENOMEM; goto fail; |
67d13fe84
|
882 |
case ZSWAP_SWAPCACHE_EXIST: |
2b2811178
|
883 |
/* page is already in the swap cache, ignore for now */ |
09cbfeaf1
|
884 |
put_page(page); |
2b2811178
|
885 886 887 888 889 890 |
ret = -EEXIST; goto fail; case ZSWAP_SWAPCACHE_NEW: /* page is locked */ /* decompress */ dlen = PAGE_SIZE; |
068619e32
|
891 |
src = (u8 *)zhdr + sizeof(struct zswap_header); |
2b2811178
|
892 |
dst = kmap_atomic(page); |
f1c54846e
|
893 894 895 896 |
tfm = *get_cpu_ptr(entry->pool->tfm); ret = crypto_comp_decompress(tfm, src, entry->length, dst, &dlen); put_cpu_ptr(entry->pool->tfm); |
2b2811178
|
897 |
kunmap_atomic(dst); |
2b2811178
|
898 899 900 901 902 903 |
BUG_ON(ret); BUG_ON(dlen != PAGE_SIZE); /* page is up to date */ SetPageUptodate(page); } |
b349acc76
|
904 905 |
/* move it to the tail of the inactive list after end_writeback */ SetPageReclaim(page); |
2b2811178
|
906 907 |
/* start writeback */ __swap_writepage(page, &wbc, end_swap_bio_write); |
09cbfeaf1
|
908 |
put_page(page); |
2b2811178
|
909 910 911 |
zswap_written_back_pages++; spin_lock(&tree->lock); |
2b2811178
|
912 |
/* drop local reference */ |
0ab0abcf5
|
913 |
zswap_entry_put(tree, entry); |
2b2811178
|
914 915 |
/* |
0ab0abcf5
|
916 917 918 919 920 921 922 923 |
* There are two possible situations for entry here: * (1) refcount is 1(normal case), entry is valid and on the tree * (2) refcount is 0, entry is freed and not on the tree * because invalidate happened during writeback * search the tree and free the entry if find entry */ if (entry == zswap_rb_search(&tree->rbroot, offset)) zswap_entry_put(tree, entry); |
2b2811178
|
924 |
spin_unlock(&tree->lock); |
2b2811178
|
925 |
|
0ab0abcf5
|
926 927 928 929 930 931 932 933 934 |
goto end; /* * if we get here due to ZSWAP_SWAPCACHE_EXIST * a load may happening concurrently * it is safe and okay to not free the entry * if we free the entry in the following put * it it either okay to return !0 */ |
2b2811178
|
935 936 |
fail: spin_lock(&tree->lock); |
0ab0abcf5
|
937 |
zswap_entry_put(tree, entry); |
2b2811178
|
938 |
spin_unlock(&tree->lock); |
0ab0abcf5
|
939 940 |
end: |
068619e32
|
941 |
zpool_unmap_handle(pool, handle); |
2b2811178
|
942 943 |
return ret; } |
a85f878b4
|
944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 |
static int zswap_is_page_same_filled(void *ptr, unsigned long *value) { unsigned int pos; unsigned long *page; page = (unsigned long *)ptr; for (pos = 1; pos < PAGE_SIZE / sizeof(*page); pos++) { if (page[pos] != page[0]) return 0; } *value = page[0]; return 1; } static void zswap_fill_page(void *ptr, unsigned long value) { unsigned long *page; page = (unsigned long *)ptr; memset_l(page, value, PAGE_SIZE / sizeof(unsigned long)); } |
2b2811178
|
965 966 967 968 969 970 971 972 973 |
/********************************* * frontswap hooks **********************************/ /* attempts to compress and store an single page */ static int zswap_frontswap_store(unsigned type, pgoff_t offset, struct page *page) { struct zswap_tree *tree = zswap_trees[type]; struct zswap_entry *entry, *dupentry; |
f1c54846e
|
974 |
struct crypto_comp *tfm; |
2b2811178
|
975 |
int ret; |
9c3760eb8
|
976 |
unsigned int hlen, dlen = PAGE_SIZE; |
a85f878b4
|
977 |
unsigned long handle, value; |
2b2811178
|
978 979 |
char *buf; u8 *src, *dst; |
9c3760eb8
|
980 |
struct zswap_header zhdr = { .swpentry = swp_entry(type, offset) }; |
d2fcd82bb
|
981 |
gfp_t gfp; |
2b2811178
|
982 |
|
7ba716698
|
983 984 985 986 987 |
/* THP isn't supported */ if (PageTransHuge(page)) { ret = -EINVAL; goto reject; } |
c00ed16a9
|
988 |
if (!zswap_enabled || !tree) { |
2b2811178
|
989 990 991 992 993 994 |
ret = -ENODEV; goto reject; } /* reclaim space if needed */ if (zswap_is_full()) { |
45190f01d
|
995 |
struct zswap_pool *pool; |
2b2811178
|
996 |
zswap_pool_limit_hit++; |
45190f01d
|
997 998 999 1000 1001 1002 1003 |
zswap_pool_reached_full = true; pool = zswap_pool_last_get(); if (pool) queue_work(shrink_wq, &pool->shrink_work); ret = -ENOMEM; goto reject; } |
16e536ef4
|
1004 |
|
45190f01d
|
1005 1006 |
if (zswap_pool_reached_full) { if (!zswap_can_accept()) { |
16e536ef4
|
1007 1008 |
ret = -ENOMEM; goto reject; |
45190f01d
|
1009 1010 |
} else zswap_pool_reached_full = false; |
2b2811178
|
1011 1012 1013 1014 1015 1016 1017 1018 1019 |
} /* allocate entry */ entry = zswap_entry_cache_alloc(GFP_KERNEL); if (!entry) { zswap_reject_kmemcache_fail++; ret = -ENOMEM; goto reject; } |
a85f878b4
|
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 |
if (zswap_same_filled_pages_enabled) { src = kmap_atomic(page); if (zswap_is_page_same_filled(src, &value)) { kunmap_atomic(src); entry->offset = offset; entry->length = 0; entry->value = value; atomic_inc(&zswap_same_filled_pages); goto insert_entry; } kunmap_atomic(src); } |
f1c54846e
|
1032 1033 1034 1035 1036 1037 |
/* if entry is successfully added, it keeps the reference */ entry->pool = zswap_pool_current_get(); if (!entry->pool) { ret = -EINVAL; goto freepage; } |
2b2811178
|
1038 1039 |
/* compress */ dst = get_cpu_var(zswap_dstmem); |
f1c54846e
|
1040 |
tfm = *get_cpu_ptr(entry->pool->tfm); |
2b2811178
|
1041 |
src = kmap_atomic(page); |
f1c54846e
|
1042 |
ret = crypto_comp_compress(tfm, src, PAGE_SIZE, dst, &dlen); |
2b2811178
|
1043 |
kunmap_atomic(src); |
f1c54846e
|
1044 |
put_cpu_ptr(entry->pool->tfm); |
2b2811178
|
1045 1046 |
if (ret) { ret = -EINVAL; |
f1c54846e
|
1047 |
goto put_dstmem; |
2b2811178
|
1048 1049 1050 |
} /* store */ |
9c3760eb8
|
1051 |
hlen = zpool_evictable(entry->pool->zpool) ? sizeof(zhdr) : 0; |
d2fcd82bb
|
1052 1053 1054 1055 |
gfp = __GFP_NORETRY | __GFP_NOWARN | __GFP_KSWAPD_RECLAIM; if (zpool_malloc_support_movable(entry->pool->zpool)) gfp |= __GFP_HIGHMEM | __GFP_MOVABLE; ret = zpool_malloc(entry->pool->zpool, hlen + dlen, gfp, &handle); |
2b2811178
|
1056 1057 |
if (ret == -ENOSPC) { zswap_reject_compress_poor++; |
f1c54846e
|
1058 |
goto put_dstmem; |
2b2811178
|
1059 1060 1061 |
} if (ret) { zswap_reject_alloc_fail++; |
f1c54846e
|
1062 |
goto put_dstmem; |
2b2811178
|
1063 |
} |
9c3760eb8
|
1064 1065 1066 |
buf = zpool_map_handle(entry->pool->zpool, handle, ZPOOL_MM_RW); memcpy(buf, &zhdr, hlen); memcpy(buf + hlen, dst, dlen); |
f1c54846e
|
1067 |
zpool_unmap_handle(entry->pool->zpool, handle); |
2b2811178
|
1068 1069 1070 1071 1072 1073 |
put_cpu_var(zswap_dstmem); /* populate entry */ entry->offset = offset; entry->handle = handle; entry->length = dlen; |
a85f878b4
|
1074 |
insert_entry: |
2b2811178
|
1075 1076 1077 1078 1079 1080 1081 |
/* map */ spin_lock(&tree->lock); do { ret = zswap_rb_insert(&tree->rbroot, entry, &dupentry); if (ret == -EEXIST) { zswap_duplicate_entry++; /* remove from rbtree */ |
0ab0abcf5
|
1082 1083 |
zswap_rb_erase(&tree->rbroot, dupentry); zswap_entry_put(tree, dupentry); |
2b2811178
|
1084 1085 1086 1087 1088 1089 |
} } while (ret == -EEXIST); spin_unlock(&tree->lock); /* update stats */ atomic_inc(&zswap_stored_pages); |
f1c54846e
|
1090 |
zswap_update_total_size(); |
2b2811178
|
1091 1092 |
return 0; |
f1c54846e
|
1093 |
put_dstmem: |
2b2811178
|
1094 |
put_cpu_var(zswap_dstmem); |
f1c54846e
|
1095 1096 |
zswap_pool_put(entry->pool); freepage: |
2b2811178
|
1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 |
zswap_entry_cache_free(entry); reject: return ret; } /* * returns 0 if the page was successfully decompressed * return -1 on entry not found or error */ static int zswap_frontswap_load(unsigned type, pgoff_t offset, struct page *page) { struct zswap_tree *tree = zswap_trees[type]; struct zswap_entry *entry; |
f1c54846e
|
1111 |
struct crypto_comp *tfm; |
2b2811178
|
1112 1113 |
u8 *src, *dst; unsigned int dlen; |
0ab0abcf5
|
1114 |
int ret; |
2b2811178
|
1115 1116 1117 |
/* find */ spin_lock(&tree->lock); |
0ab0abcf5
|
1118 |
entry = zswap_entry_find_get(&tree->rbroot, offset); |
2b2811178
|
1119 1120 1121 1122 1123 |
if (!entry) { /* entry was written back */ spin_unlock(&tree->lock); return -1; } |
2b2811178
|
1124 |
spin_unlock(&tree->lock); |
a85f878b4
|
1125 1126 1127 1128 1129 1130 |
if (!entry->length) { dst = kmap_atomic(page); zswap_fill_page(dst, entry->value); kunmap_atomic(dst); goto freeentry; } |
2b2811178
|
1131 1132 |
/* decompress */ dlen = PAGE_SIZE; |
9c3760eb8
|
1133 1134 1135 |
src = zpool_map_handle(entry->pool->zpool, entry->handle, ZPOOL_MM_RO); if (zpool_evictable(entry->pool->zpool)) src += sizeof(struct zswap_header); |
2b2811178
|
1136 |
dst = kmap_atomic(page); |
f1c54846e
|
1137 1138 1139 |
tfm = *get_cpu_ptr(entry->pool->tfm); ret = crypto_comp_decompress(tfm, src, entry->length, dst, &dlen); put_cpu_ptr(entry->pool->tfm); |
2b2811178
|
1140 |
kunmap_atomic(dst); |
f1c54846e
|
1141 |
zpool_unmap_handle(entry->pool->zpool, entry->handle); |
2b2811178
|
1142 |
BUG_ON(ret); |
a85f878b4
|
1143 |
freeentry: |
2b2811178
|
1144 |
spin_lock(&tree->lock); |
0ab0abcf5
|
1145 |
zswap_entry_put(tree, entry); |
2b2811178
|
1146 |
spin_unlock(&tree->lock); |
2b2811178
|
1147 1148 1149 1150 1151 1152 1153 1154 |
return 0; } /* frees an entry in zswap */ static void zswap_frontswap_invalidate_page(unsigned type, pgoff_t offset) { struct zswap_tree *tree = zswap_trees[type]; struct zswap_entry *entry; |
2b2811178
|
1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 |
/* find */ spin_lock(&tree->lock); entry = zswap_rb_search(&tree->rbroot, offset); if (!entry) { /* entry was written back */ spin_unlock(&tree->lock); return; } /* remove from rbtree */ |
0ab0abcf5
|
1166 |
zswap_rb_erase(&tree->rbroot, entry); |
2b2811178
|
1167 1168 |
/* drop the initial reference from entry creation */ |
0ab0abcf5
|
1169 |
zswap_entry_put(tree, entry); |
2b2811178
|
1170 1171 |
spin_unlock(&tree->lock); |
2b2811178
|
1172 1173 1174 1175 1176 1177 |
} /* frees all zswap entries for the given swap type */ static void zswap_frontswap_invalidate_area(unsigned type) { struct zswap_tree *tree = zswap_trees[type]; |
0bd42136f
|
1178 |
struct zswap_entry *entry, *n; |
2b2811178
|
1179 1180 1181 1182 1183 1184 |
if (!tree) return; /* walk the tree and free everything */ spin_lock(&tree->lock); |
0ab0abcf5
|
1185 |
rbtree_postorder_for_each_entry_safe(entry, n, &tree->rbroot, rbnode) |
60105e124
|
1186 |
zswap_free_entry(entry); |
2b2811178
|
1187 1188 |
tree->rbroot = RB_ROOT; spin_unlock(&tree->lock); |
aa9bca05a
|
1189 1190 |
kfree(tree); zswap_trees[type] = NULL; |
2b2811178
|
1191 |
} |
2b2811178
|
1192 1193 1194 |
static void zswap_frontswap_init(unsigned type) { struct zswap_tree *tree; |
9cd1f701c
|
1195 |
tree = kzalloc(sizeof(*tree), GFP_KERNEL); |
60105e124
|
1196 1197 1198 1199 1200 |
if (!tree) { pr_err("alloc failed, zswap disabled for swap type %d ", type); return; } |
2b2811178
|
1201 1202 1203 |
tree->rbroot = RB_ROOT; spin_lock_init(&tree->lock); zswap_trees[type] = tree; |
2b2811178
|
1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 |
} static struct frontswap_ops zswap_frontswap_ops = { .store = zswap_frontswap_store, .load = zswap_frontswap_load, .invalidate_page = zswap_frontswap_invalidate_page, .invalidate_area = zswap_frontswap_invalidate_area, .init = zswap_frontswap_init }; /********************************* * debugfs functions **********************************/ #ifdef CONFIG_DEBUG_FS #include <linux/debugfs.h> static struct dentry *zswap_debugfs_root; static int __init zswap_debugfs_init(void) { if (!debugfs_initialized()) return -ENODEV; zswap_debugfs_root = debugfs_create_dir("zswap", NULL); |
2b2811178
|
1228 |
|
0825a6f98
|
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 |
debugfs_create_u64("pool_limit_hit", 0444, zswap_debugfs_root, &zswap_pool_limit_hit); debugfs_create_u64("reject_reclaim_fail", 0444, zswap_debugfs_root, &zswap_reject_reclaim_fail); debugfs_create_u64("reject_alloc_fail", 0444, zswap_debugfs_root, &zswap_reject_alloc_fail); debugfs_create_u64("reject_kmemcache_fail", 0444, zswap_debugfs_root, &zswap_reject_kmemcache_fail); debugfs_create_u64("reject_compress_poor", 0444, zswap_debugfs_root, &zswap_reject_compress_poor); debugfs_create_u64("written_back_pages", 0444, zswap_debugfs_root, &zswap_written_back_pages); debugfs_create_u64("duplicate_entry", 0444, zswap_debugfs_root, &zswap_duplicate_entry); debugfs_create_u64("pool_total_size", 0444, zswap_debugfs_root, &zswap_pool_total_size); debugfs_create_atomic_t("stored_pages", 0444, zswap_debugfs_root, &zswap_stored_pages); |
a85f878b4
|
1247 |
debugfs_create_atomic_t("same_filled_pages", 0444, |
0825a6f98
|
1248 |
zswap_debugfs_root, &zswap_same_filled_pages); |
2b2811178
|
1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 |
return 0; } static void __exit zswap_debugfs_exit(void) { debugfs_remove_recursive(zswap_debugfs_root); } #else static int __init zswap_debugfs_init(void) { return 0; } static void __exit zswap_debugfs_exit(void) { } #endif /********************************* * module init and exit **********************************/ static int __init init_zswap(void) { |
f1c54846e
|
1271 |
struct zswap_pool *pool; |
ad7ed7708
|
1272 |
int ret; |
60105e124
|
1273 |
|
90b0fc26d
|
1274 |
zswap_init_started = true; |
2b2811178
|
1275 1276 1277 |
if (zswap_entry_cache_create()) { pr_err("entry cache creation failed "); |
f1c54846e
|
1278 |
goto cache_fail; |
2b2811178
|
1279 |
} |
f1c54846e
|
1280 |
|
ad7ed7708
|
1281 1282 1283 |
ret = cpuhp_setup_state(CPUHP_MM_ZSWP_MEM_PREPARE, "mm/zswap:prepare", zswap_dstmem_prepare, zswap_dstmem_dead); if (ret) { |
f1c54846e
|
1284 1285 1286 |
pr_err("dstmem alloc failed "); goto dstmem_fail; |
2b2811178
|
1287 |
} |
f1c54846e
|
1288 |
|
cab7a7e5b
|
1289 1290 1291 1292 1293 1294 |
ret = cpuhp_setup_state_multi(CPUHP_MM_ZSWP_POOL_PREPARE, "mm/zswap_pool:prepare", zswap_cpu_comp_prepare, zswap_cpu_comp_dead); if (ret) goto hp_fail; |
f1c54846e
|
1295 |
pool = __zswap_pool_create_fallback(); |
ae3d89a7e
|
1296 1297 1298 1299 1300 1301 1302 |
if (pool) { pr_info("loaded using pool %s/%s ", pool->tfm_name, zpool_get_type(pool->zpool)); list_add(&pool->list, &zswap_pools); zswap_has_pool = true; } else { |
f1c54846e
|
1303 1304 |
pr_err("pool creation failed "); |
ae3d89a7e
|
1305 |
zswap_enabled = false; |
2b2811178
|
1306 |
} |
60105e124
|
1307 |
|
45190f01d
|
1308 1309 1310 |
shrink_wq = create_workqueue("zswap-shrink"); if (!shrink_wq) goto fallback_fail; |
2b2811178
|
1311 1312 1313 1314 1315 |
frontswap_register_ops(&zswap_frontswap_ops); if (zswap_debugfs_init()) pr_warn("debugfs initialization failed "); return 0; |
f1c54846e
|
1316 |
|
45190f01d
|
1317 |
fallback_fail: |
38aeb071b
|
1318 1319 |
if (pool) zswap_pool_destroy(pool); |
cab7a7e5b
|
1320 |
hp_fail: |
ad7ed7708
|
1321 |
cpuhp_remove_state(CPUHP_MM_ZSWP_MEM_PREPARE); |
f1c54846e
|
1322 |
dstmem_fail: |
c119239b1
|
1323 |
zswap_entry_cache_destroy(); |
f1c54846e
|
1324 |
cache_fail: |
d7b028f56
|
1325 1326 1327 |
/* if built-in, we aren't unloaded on failure; don't allow use */ zswap_init_failed = true; zswap_enabled = false; |
2b2811178
|
1328 1329 1330 1331 1332 1333 |
return -ENOMEM; } /* must be late so crypto has time to come up */ late_initcall(init_zswap); MODULE_LICENSE("GPL"); |
68386da82
|
1334 |
MODULE_AUTHOR("Seth Jennings <sjennings@variantweb.net>"); |
2b2811178
|
1335 |
MODULE_DESCRIPTION("Compressed cache for swap pages"); |