Blame view
fs/mbcache.c
24.1 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 |
/* * linux/fs/mbcache.c * (C) 2001-2002 Andreas Gruenbacher, <a.gruenbacher@computer.org> */ /* * Filesystem Meta Information Block Cache (mbcache) * * The mbcache caches blocks of block devices that need to be located * by their device/block number, as well as by other criteria (such * as the block's contents). * * There can only be one cache entry in a cache per device and block number. * Additional indexes need not be unique in this sense. The number of * additional indexes (=other criteria) can be hardwired at compile time * or specified at cache create time. * * Each cache entry is of fixed size. An entry may be `valid' or `invalid' * in the cache. A valid entry is in the main hash tables of the cache, * and may also be in the lru list. An invalid entry is not in any hashes * or lists. * * A valid cache entry is only in the lru list if no handles refer to it. * Invalid cache entries will be freed when the last handle to the cache * entry is released. Entries that cannot be freed immediately are put * back on the lru list. */ |
1f3e55fe0
|
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 |
/* * Lock descriptions and usage: * * Each hash chain of both the block and index hash tables now contains * a built-in lock used to serialize accesses to the hash chain. * * Accesses to global data structures mb_cache_list and mb_cache_lru_list * are serialized via the global spinlock mb_cache_spinlock. * * Each mb_cache_entry contains a spinlock, e_entry_lock, to serialize * accesses to its local data, such as e_used and e_queued. * * Lock ordering: * * Each block hash chain's lock has the highest lock order, followed by an * index hash chain's lock, mb_cache_bg_lock (used to implement mb_cache_entry's * lock), and mb_cach_spinlock, with the lowest order. While holding * either a block or index hash chain lock, a thread can acquire an * mc_cache_bg_lock, which in turn can also acquire mb_cache_spinlock. * * Synchronization: * * Since both mb_cache_entry_get and mb_cache_entry_find scan the block and * index hash chian, it needs to lock the corresponding hash chain. For each * mb_cache_entry within the chain, it needs to lock the mb_cache_entry to * prevent either any simultaneous release or free on the entry and also * to serialize accesses to either the e_used or e_queued member of the entry. * * To avoid having a dangling reference to an already freed * mb_cache_entry, an mb_cache_entry is only freed when it is not on a * block hash chain and also no longer being referenced, both e_used, * and e_queued are 0's. When an mb_cache_entry is explicitly freed it is * first removed from a block hash chain. */ |
1da177e4c
|
62 63 64 65 66 67 68 69 |
#include <linux/kernel.h> #include <linux/module.h> #include <linux/hash.h> #include <linux/fs.h> #include <linux/mm.h> #include <linux/slab.h> #include <linux/sched.h> |
3e037e521
|
70 |
#include <linux/list_bl.h> |
1da177e4c
|
71 |
#include <linux/mbcache.h> |
3e037e521
|
72 |
#include <linux/init.h> |
1f3e55fe0
|
73 |
#include <linux/blockgroup_lock.h> |
ec7756ae1
|
74 |
#include <linux/log2.h> |
1da177e4c
|
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 |
#ifdef MB_CACHE_DEBUG # define mb_debug(f...) do { \ printk(KERN_DEBUG f); \ printk(" "); \ } while (0) #define mb_assert(c) do { if (!(c)) \ printk(KERN_ERR "assertion " #c " failed "); \ } while(0) #else # define mb_debug(f...) do { } while(0) # define mb_assert(c) do { } while(0) #endif #define mb_error(f...) do { \ printk(KERN_ERR f); \ printk(" "); \ } while(0) #define MB_CACHE_WRITER ((unsigned short)~0U >> 1) |
ec7756ae1
|
97 |
#define MB_CACHE_ENTRY_LOCK_BITS ilog2(NR_BG_LOCKS) |
1f3e55fe0
|
98 99 |
#define MB_CACHE_ENTRY_LOCK_INDEX(ce) \ (hash_long((unsigned long)ce, MB_CACHE_ENTRY_LOCK_BITS)) |
75c96f858
|
100 |
static DECLARE_WAIT_QUEUE_HEAD(mb_cache_queue); |
1f3e55fe0
|
101 |
static struct blockgroup_lock *mb_cache_bg_lock; |
9c191f701
|
102 |
static struct kmem_cache *mb_cache_kmem_cache; |
1f3e55fe0
|
103 |
|
1da177e4c
|
104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 |
MODULE_AUTHOR("Andreas Gruenbacher <a.gruenbacher@computer.org>"); MODULE_DESCRIPTION("Meta block cache (for extended attributes)"); MODULE_LICENSE("GPL"); EXPORT_SYMBOL(mb_cache_create); EXPORT_SYMBOL(mb_cache_shrink); EXPORT_SYMBOL(mb_cache_destroy); EXPORT_SYMBOL(mb_cache_entry_alloc); EXPORT_SYMBOL(mb_cache_entry_insert); EXPORT_SYMBOL(mb_cache_entry_release); EXPORT_SYMBOL(mb_cache_entry_free); EXPORT_SYMBOL(mb_cache_entry_get); #if !defined(MB_CACHE_INDEXES_COUNT) || (MB_CACHE_INDEXES_COUNT > 0) EXPORT_SYMBOL(mb_cache_entry_find_first); EXPORT_SYMBOL(mb_cache_entry_find_next); #endif |
1da177e4c
|
120 121 122 123 124 125 126 127 128 |
/* * Global data: list of all mbcache's, lru list, and a spinlock for * accessing cache data structures on SMP machines. The lru list is * global across all mbcaches. */ static LIST_HEAD(mb_cache_list); static LIST_HEAD(mb_cache_lru_list); static DEFINE_SPINLOCK(mb_cache_spinlock); |
1da177e4c
|
129 |
|
1f3e55fe0
|
130 131 132 133 134 135 136 137 138 139 140 141 142 |
static inline void __spin_lock_mb_cache_entry(struct mb_cache_entry *ce) { spin_lock(bgl_lock_ptr(mb_cache_bg_lock, MB_CACHE_ENTRY_LOCK_INDEX(ce))); } static inline void __spin_unlock_mb_cache_entry(struct mb_cache_entry *ce) { spin_unlock(bgl_lock_ptr(mb_cache_bg_lock, MB_CACHE_ENTRY_LOCK_INDEX(ce))); } |
1da177e4c
|
143 |
static inline int |
3e037e521
|
144 |
__mb_cache_entry_is_block_hashed(struct mb_cache_entry *ce) |
1da177e4c
|
145 |
{ |
3e037e521
|
146 |
return !hlist_bl_unhashed(&ce->e_block_list); |
1da177e4c
|
147 |
} |
3e037e521
|
148 149 |
static inline void __mb_cache_entry_unhash_block(struct mb_cache_entry *ce) |
1da177e4c
|
150 |
{ |
3e037e521
|
151 152 153 154 155 156 157 158 |
if (__mb_cache_entry_is_block_hashed(ce)) hlist_bl_del_init(&ce->e_block_list); } static inline int __mb_cache_entry_is_index_hashed(struct mb_cache_entry *ce) { return !hlist_bl_unhashed(&ce->e_index.o_list); |
1da177e4c
|
159 |
} |
3e037e521
|
160 161 162 163 164 165 |
static inline void __mb_cache_entry_unhash_index(struct mb_cache_entry *ce) { if (__mb_cache_entry_is_index_hashed(ce)) hlist_bl_del_init(&ce->e_index.o_list); } |
1f3e55fe0
|
166 167 168 169 170 171 172 173 |
/* * __mb_cache_entry_unhash_unlock() * * This function is called to unhash both the block and index hash * chain. * It assumes both the block and index hash chain is locked upon entry. * It also unlock both hash chains both exit */ |
3e037e521
|
174 |
static inline void |
1f3e55fe0
|
175 |
__mb_cache_entry_unhash_unlock(struct mb_cache_entry *ce) |
3e037e521
|
176 177 |
{ __mb_cache_entry_unhash_index(ce); |
1f3e55fe0
|
178 |
hlist_bl_unlock(ce->e_index_hash_p); |
3e037e521
|
179 |
__mb_cache_entry_unhash_block(ce); |
1f3e55fe0
|
180 |
hlist_bl_unlock(ce->e_block_hash_p); |
3e037e521
|
181 |
} |
1da177e4c
|
182 |
|
858119e15
|
183 |
static void |
27496a8c6
|
184 |
__mb_cache_entry_forget(struct mb_cache_entry *ce, gfp_t gfp_mask) |
1da177e4c
|
185 186 |
{ struct mb_cache *cache = ce->e_cache; |
1f3e55fe0
|
187 |
mb_assert(!(ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt))); |
2aec7c523
|
188 189 |
kmem_cache_free(cache->c_entry_cache, ce); atomic_dec(&cache->c_entry_count); |
1da177e4c
|
190 |
} |
858119e15
|
191 |
static void |
1f3e55fe0
|
192 |
__mb_cache_entry_release(struct mb_cache_entry *ce) |
1da177e4c
|
193 |
{ |
1f3e55fe0
|
194 195 |
/* First lock the entry to serialize access to its local data. */ __spin_lock_mb_cache_entry(ce); |
1da177e4c
|
196 197 198 199 200 |
/* Wake up all processes queuing for this cache entry. */ if (ce->e_queued) wake_up_all(&mb_cache_queue); if (ce->e_used >= MB_CACHE_WRITER) ce->e_used -= MB_CACHE_WRITER; |
1f3e55fe0
|
201 202 203 204 |
/* * Make sure that all cache entries on lru_list have * both e_used and e_qued of 0s. */ |
1da177e4c
|
205 |
ce->e_used--; |
1f3e55fe0
|
206 207 208 |
if (!(ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt))) { if (!__mb_cache_entry_is_block_hashed(ce)) { __spin_unlock_mb_cache_entry(ce); |
1da177e4c
|
209 |
goto forget; |
1f3e55fe0
|
210 211 212 213 214 215 216 217 218 |
} /* * Need access to lru list, first drop entry lock, * then reacquire the lock in the proper order. */ spin_lock(&mb_cache_spinlock); if (list_empty(&ce->e_lru_list)) list_add_tail(&ce->e_lru_list, &mb_cache_lru_list); spin_unlock(&mb_cache_spinlock); |
1da177e4c
|
219 |
} |
1f3e55fe0
|
220 |
__spin_unlock_mb_cache_entry(ce); |
1da177e4c
|
221 222 |
return; forget: |
1f3e55fe0
|
223 |
mb_assert(list_empty(&ce->e_lru_list)); |
1da177e4c
|
224 225 |
__mb_cache_entry_forget(ce, GFP_KERNEL); } |
1da177e4c
|
226 |
/* |
1ab6c4997
|
227 |
* mb_cache_shrink_scan() memory pressure callback |
1da177e4c
|
228 229 230 231 |
* * This function is called by the kernel memory management when memory * gets low. * |
7f8275d0d
|
232 |
* @shrink: (ignored) |
1495f230f
|
233 |
* @sc: shrink_control passed from reclaim |
1da177e4c
|
234 |
* |
1ab6c4997
|
235 |
* Returns the number of objects freed. |
1da177e4c
|
236 |
*/ |
1ab6c4997
|
237 238 |
static unsigned long mb_cache_shrink_scan(struct shrinker *shrink, struct shrink_control *sc) |
1da177e4c
|
239 240 |
{ LIST_HEAD(free_list); |
e566d48c9
|
241 |
struct mb_cache_entry *entry, *tmp; |
1495f230f
|
242 243 |
int nr_to_scan = sc->nr_to_scan; gfp_t gfp_mask = sc->gfp_mask; |
1ab6c4997
|
244 |
unsigned long freed = 0; |
1da177e4c
|
245 |
|
1da177e4c
|
246 |
mb_debug("trying to free %d entries", nr_to_scan); |
e566d48c9
|
247 |
spin_lock(&mb_cache_spinlock); |
1f3e55fe0
|
248 |
while ((nr_to_scan-- > 0) && !list_empty(&mb_cache_lru_list)) { |
1da177e4c
|
249 250 |
struct mb_cache_entry *ce = list_entry(mb_cache_lru_list.next, |
1f3e55fe0
|
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 |
struct mb_cache_entry, e_lru_list); list_del_init(&ce->e_lru_list); if (ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt)) continue; spin_unlock(&mb_cache_spinlock); /* Prevent any find or get operation on the entry */ hlist_bl_lock(ce->e_block_hash_p); hlist_bl_lock(ce->e_index_hash_p); /* Ignore if it is touched by a find/get */ if (ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt) || !list_empty(&ce->e_lru_list)) { hlist_bl_unlock(ce->e_index_hash_p); hlist_bl_unlock(ce->e_block_hash_p); spin_lock(&mb_cache_spinlock); continue; } __mb_cache_entry_unhash_unlock(ce); list_add_tail(&ce->e_lru_list, &free_list); spin_lock(&mb_cache_spinlock); |
1ab6c4997
|
270 271 |
} spin_unlock(&mb_cache_spinlock); |
1f3e55fe0
|
272 |
|
1ab6c4997
|
273 274 |
list_for_each_entry_safe(entry, tmp, &free_list, e_lru_list) { __mb_cache_entry_forget(entry, gfp_mask); |
1f3e55fe0
|
275 |
freed++; |
1da177e4c
|
276 |
} |
1ab6c4997
|
277 278 279 280 281 282 283 284 285 286 |
return freed; } static unsigned long mb_cache_shrink_count(struct shrinker *shrink, struct shrink_control *sc) { struct mb_cache *cache; unsigned long count = 0; spin_lock(&mb_cache_spinlock); |
e566d48c9
|
287 288 289 290 291 |
list_for_each_entry(cache, &mb_cache_list, c_cache_list) { mb_debug("cache %s (%d)", cache->c_name, atomic_read(&cache->c_entry_count)); count += atomic_read(&cache->c_entry_count); } |
1da177e4c
|
292 |
spin_unlock(&mb_cache_spinlock); |
1ab6c4997
|
293 |
|
55f841ce9
|
294 |
return vfs_pressure_ratio(count); |
1da177e4c
|
295 |
} |
1ab6c4997
|
296 297 298 299 300 |
static struct shrinker mb_cache_shrinker = { .count_objects = mb_cache_shrink_count, .scan_objects = mb_cache_shrink_scan, .seeks = DEFAULT_SEEKS, }; |
1da177e4c
|
301 302 303 304 305 306 307 308 309 310 |
/* * mb_cache_create() create a new cache * * All entries in one cache are equal size. Cache entries may be from * multiple devices. If this is the first mbcache created, registers * the cache with kernel memory management. Returns NULL if no more * memory was available. * * @name: name of the cache (informal) |
1da177e4c
|
311 312 313 |
* @bucket_bits: log2(number of hash buckets) */ struct mb_cache * |
2aec7c523
|
314 |
mb_cache_create(const char *name, int bucket_bits) |
1da177e4c
|
315 |
{ |
2aec7c523
|
316 |
int n, bucket_count = 1 << bucket_bits; |
1da177e4c
|
317 |
struct mb_cache *cache = NULL; |
1f3e55fe0
|
318 319 320 321 322 323 324 |
if (!mb_cache_bg_lock) { mb_cache_bg_lock = kmalloc(sizeof(struct blockgroup_lock), GFP_KERNEL); if (!mb_cache_bg_lock) return NULL; bgl_lock_init(mb_cache_bg_lock); } |
2aec7c523
|
325 |
cache = kmalloc(sizeof(struct mb_cache), GFP_KERNEL); |
1da177e4c
|
326 |
if (!cache) |
2aec7c523
|
327 |
return NULL; |
1da177e4c
|
328 |
cache->c_name = name; |
1da177e4c
|
329 330 |
atomic_set(&cache->c_entry_count, 0); cache->c_bucket_bits = bucket_bits; |
3e037e521
|
331 332 |
cache->c_block_hash = kmalloc(bucket_count * sizeof(struct hlist_bl_head), GFP_KERNEL); |
1da177e4c
|
333 334 335 |
if (!cache->c_block_hash) goto fail; for (n=0; n<bucket_count; n++) |
3e037e521
|
336 337 338 |
INIT_HLIST_BL_HEAD(&cache->c_block_hash[n]); cache->c_index_hash = kmalloc(bucket_count * sizeof(struct hlist_bl_head), GFP_KERNEL); |
2aec7c523
|
339 340 341 |
if (!cache->c_index_hash) goto fail; for (n=0; n<bucket_count; n++) |
3e037e521
|
342 |
INIT_HLIST_BL_HEAD(&cache->c_index_hash[n]); |
9c191f701
|
343 344 345 346 347 348 349 350 |
if (!mb_cache_kmem_cache) { mb_cache_kmem_cache = kmem_cache_create(name, sizeof(struct mb_cache_entry), 0, SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD, NULL); if (!mb_cache_kmem_cache) goto fail2; } cache->c_entry_cache = mb_cache_kmem_cache; |
1da177e4c
|
351 |
|
3a48ee8a4
|
352 353 354 355 356 |
/* * Set an upper limit on the number of cache entries so that the hash * chains won't grow too long. */ cache->c_max_entries = bucket_count << 4; |
1da177e4c
|
357 358 359 360 |
spin_lock(&mb_cache_spinlock); list_add(&cache->c_cache_list, &mb_cache_list); spin_unlock(&mb_cache_spinlock); return cache; |
2aec7c523
|
361 362 |
fail2: kfree(cache->c_index_hash); |
1da177e4c
|
363 |
fail: |
2aec7c523
|
364 365 |
kfree(cache->c_block_hash); kfree(cache); |
1da177e4c
|
366 367 368 369 370 371 372 |
return NULL; } /* * mb_cache_shrink() * |
7f927fcc2
|
373 |
* Removes all cache entries of a device from the cache. All cache entries |
1da177e4c
|
374 375 376 |
* currently in use cannot be freed, and thus remain in the cache. All others * are freed. * |
1da177e4c
|
377 378 379 |
* @bdev: which device's cache entries to shrink */ void |
8c52ab42c
|
380 |
mb_cache_shrink(struct block_device *bdev) |
1da177e4c
|
381 382 |
{ LIST_HEAD(free_list); |
1f3e55fe0
|
383 384 |
struct list_head *l; struct mb_cache_entry *ce, *tmp; |
1da177e4c
|
385 |
|
1f3e55fe0
|
386 |
l = &mb_cache_lru_list; |
1da177e4c
|
387 |
spin_lock(&mb_cache_spinlock); |
1f3e55fe0
|
388 389 390 |
while (!list_is_last(l, &mb_cache_lru_list)) { l = l->next; ce = list_entry(l, struct mb_cache_entry, e_lru_list); |
1da177e4c
|
391 |
if (ce->e_bdev == bdev) { |
1f3e55fe0
|
392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 |
list_del_init(&ce->e_lru_list); if (ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt)) continue; spin_unlock(&mb_cache_spinlock); /* * Prevent any find or get operation on the entry. */ hlist_bl_lock(ce->e_block_hash_p); hlist_bl_lock(ce->e_index_hash_p); /* Ignore if it is touched by a find/get */ if (ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt) || !list_empty(&ce->e_lru_list)) { hlist_bl_unlock(ce->e_index_hash_p); hlist_bl_unlock(ce->e_block_hash_p); l = &mb_cache_lru_list; spin_lock(&mb_cache_spinlock); continue; } __mb_cache_entry_unhash_unlock(ce); mb_assert(!(ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt))); list_add_tail(&ce->e_lru_list, &free_list); l = &mb_cache_lru_list; spin_lock(&mb_cache_spinlock); |
1da177e4c
|
418 419 420 |
} } spin_unlock(&mb_cache_spinlock); |
1f3e55fe0
|
421 422 423 |
list_for_each_entry_safe(ce, tmp, &free_list, e_lru_list) { __mb_cache_entry_forget(ce, GFP_KERNEL); |
1da177e4c
|
424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 |
} } /* * mb_cache_destroy() * * Shrinks the cache to its minimum possible size (hopefully 0 entries), * and then destroys it. If this was the last mbcache, un-registers the * mbcache from kernel memory management. */ void mb_cache_destroy(struct mb_cache *cache) { LIST_HEAD(free_list); |
1f3e55fe0
|
439 |
struct mb_cache_entry *ce, *tmp; |
1da177e4c
|
440 441 |
spin_lock(&mb_cache_spinlock); |
1f3e55fe0
|
442 443 |
list_for_each_entry_safe(ce, tmp, &mb_cache_lru_list, e_lru_list) { if (ce->e_cache == cache) |
1da177e4c
|
444 |
list_move_tail(&ce->e_lru_list, &free_list); |
1da177e4c
|
445 446 447 |
} list_del(&cache->c_cache_list); spin_unlock(&mb_cache_spinlock); |
1f3e55fe0
|
448 449 450 451 452 453 454 455 456 457 458 |
list_for_each_entry_safe(ce, tmp, &free_list, e_lru_list) { list_del_init(&ce->e_lru_list); /* * Prevent any find or get operation on the entry. */ hlist_bl_lock(ce->e_block_hash_p); hlist_bl_lock(ce->e_index_hash_p); mb_assert(!(ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt))); __mb_cache_entry_unhash_unlock(ce); __mb_cache_entry_forget(ce, GFP_KERNEL); |
1da177e4c
|
459 460 461 462 463 464 465 |
} if (atomic_read(&cache->c_entry_count) > 0) { mb_error("cache %s: %d orphaned entries", cache->c_name, atomic_read(&cache->c_entry_count)); } |
9c191f701
|
466 467 468 469 |
if (list_empty(&mb_cache_list)) { kmem_cache_destroy(mb_cache_kmem_cache); mb_cache_kmem_cache = NULL; } |
2aec7c523
|
470 |
kfree(cache->c_index_hash); |
1da177e4c
|
471 472 473 |
kfree(cache->c_block_hash); kfree(cache); } |
1da177e4c
|
474 475 476 477 478 479 480 481 482 |
/* * mb_cache_entry_alloc() * * Allocates a new cache entry. The new entry will not be valid initially, * and thus cannot be looked up yet. It should be filled with data, and * then inserted into the cache using mb_cache_entry_insert(). Returns NULL * if no more memory was available. */ struct mb_cache_entry * |
335e92e8a
|
483 |
mb_cache_entry_alloc(struct mb_cache *cache, gfp_t gfp_flags) |
1da177e4c
|
484 |
{ |
1f3e55fe0
|
485 |
struct mb_cache_entry *ce; |
3a48ee8a4
|
486 487 |
if (atomic_read(&cache->c_entry_count) >= cache->c_max_entries) { |
1f3e55fe0
|
488 489 490 |
struct list_head *l; l = &mb_cache_lru_list; |
3a48ee8a4
|
491 |
spin_lock(&mb_cache_spinlock); |
1f3e55fe0
|
492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 |
while (!list_is_last(l, &mb_cache_lru_list)) { l = l->next; ce = list_entry(l, struct mb_cache_entry, e_lru_list); if (ce->e_cache == cache) { list_del_init(&ce->e_lru_list); if (ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt)) continue; spin_unlock(&mb_cache_spinlock); /* * Prevent any find or get operation on the * entry. */ hlist_bl_lock(ce->e_block_hash_p); hlist_bl_lock(ce->e_index_hash_p); /* Ignore if it is touched by a find/get */ if (ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt) || !list_empty(&ce->e_lru_list)) { hlist_bl_unlock(ce->e_index_hash_p); hlist_bl_unlock(ce->e_block_hash_p); l = &mb_cache_lru_list; spin_lock(&mb_cache_spinlock); continue; } mb_assert(list_empty(&ce->e_lru_list)); mb_assert(!(ce->e_used || ce->e_queued || atomic_read(&ce->e_refcnt))); __mb_cache_entry_unhash_unlock(ce); goto found; } |
3a48ee8a4
|
523 524 525 |
} spin_unlock(&mb_cache_spinlock); } |
1f3e55fe0
|
526 527 528 529 530 531 532 533 534 535 536 537 |
ce = kmem_cache_alloc(cache->c_entry_cache, gfp_flags); if (!ce) return NULL; atomic_inc(&cache->c_entry_count); INIT_LIST_HEAD(&ce->e_lru_list); INIT_HLIST_BL_NODE(&ce->e_block_list); INIT_HLIST_BL_NODE(&ce->e_index.o_list); ce->e_cache = cache; ce->e_queued = 0; atomic_set(&ce->e_refcnt, 0); found: |
3e037e521
|
538 539 |
ce->e_block_hash_p = &cache->c_block_hash[0]; ce->e_index_hash_p = &cache->c_index_hash[0]; |
3a48ee8a4
|
540 |
ce->e_used = 1 + MB_CACHE_WRITER; |
1da177e4c
|
541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 |
return ce; } /* * mb_cache_entry_insert() * * Inserts an entry that was allocated using mb_cache_entry_alloc() into * the cache. After this, the cache entry can be looked up, but is not yet * in the lru list as the caller still holds a handle to it. Returns 0 on * success, or -EBUSY if a cache entry for that device + inode exists * already (this may happen after a failed lookup, but when another process * has inserted the same cache entry in the meantime). * * @bdev: device the cache entry belongs to * @block: block number |
2aec7c523
|
557 |
* @key: lookup key |
1da177e4c
|
558 559 560 |
*/ int mb_cache_entry_insert(struct mb_cache_entry *ce, struct block_device *bdev, |
2aec7c523
|
561 |
sector_t block, unsigned int key) |
1da177e4c
|
562 563 564 |
{ struct mb_cache *cache = ce->e_cache; unsigned int bucket; |
3e037e521
|
565 |
struct hlist_bl_node *l; |
3e037e521
|
566 567 568 |
struct hlist_bl_head *block_hash_p; struct hlist_bl_head *index_hash_p; struct mb_cache_entry *lce; |
1da177e4c
|
569 |
|
3e037e521
|
570 |
mb_assert(ce); |
1da177e4c
|
571 572 |
bucket = hash_long((unsigned long)bdev + (block & 0xffffffff), cache->c_bucket_bits); |
3e037e521
|
573 |
block_hash_p = &cache->c_block_hash[bucket]; |
1f3e55fe0
|
574 |
hlist_bl_lock(block_hash_p); |
3e037e521
|
575 |
hlist_bl_for_each_entry(lce, l, block_hash_p, e_block_list) { |
1f3e55fe0
|
576 577 578 579 |
if (lce->e_bdev == bdev && lce->e_block == block) { hlist_bl_unlock(block_hash_p); return -EBUSY; } |
1da177e4c
|
580 |
} |
3e037e521
|
581 |
mb_assert(!__mb_cache_entry_is_block_hashed(ce)); |
1f3e55fe0
|
582 583 |
__mb_cache_entry_unhash_block(ce); __mb_cache_entry_unhash_index(ce); |
1da177e4c
|
584 585 |
ce->e_bdev = bdev; ce->e_block = block; |
3e037e521
|
586 |
ce->e_block_hash_p = block_hash_p; |
2aec7c523
|
587 |
ce->e_index.o_key = key; |
1f3e55fe0
|
588 589 |
hlist_bl_add_head(&ce->e_block_list, block_hash_p); hlist_bl_unlock(block_hash_p); |
2aec7c523
|
590 |
bucket = hash_long(key, cache->c_bucket_bits); |
3e037e521
|
591 |
index_hash_p = &cache->c_index_hash[bucket]; |
1f3e55fe0
|
592 |
hlist_bl_lock(index_hash_p); |
3e037e521
|
593 594 |
ce->e_index_hash_p = index_hash_p; hlist_bl_add_head(&ce->e_index.o_list, index_hash_p); |
1f3e55fe0
|
595 596 |
hlist_bl_unlock(index_hash_p); return 0; |
1da177e4c
|
597 598 599 600 601 602 603 604 605 606 607 608 609 |
} /* * mb_cache_entry_release() * * Release a handle to a cache entry. When the last handle to a cache entry * is released it is either freed (if it is invalid) or otherwise inserted * in to the lru list. */ void mb_cache_entry_release(struct mb_cache_entry *ce) { |
1f3e55fe0
|
610 |
__mb_cache_entry_release(ce); |
1da177e4c
|
611 612 613 614 615 616 |
} /* * mb_cache_entry_free() * |
1da177e4c
|
617 618 619 620 |
*/ void mb_cache_entry_free(struct mb_cache_entry *ce) { |
1f3e55fe0
|
621 |
mb_assert(ce); |
1da177e4c
|
622 |
mb_assert(list_empty(&ce->e_lru_list)); |
1f3e55fe0
|
623 624 625 626 627 628 629 |
hlist_bl_lock(ce->e_index_hash_p); __mb_cache_entry_unhash_index(ce); hlist_bl_unlock(ce->e_index_hash_p); hlist_bl_lock(ce->e_block_hash_p); __mb_cache_entry_unhash_block(ce); hlist_bl_unlock(ce->e_block_hash_p); __mb_cache_entry_release(ce); |
1da177e4c
|
630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 |
} /* * mb_cache_entry_get() * * Get a cache entry by device / block number. (There can only be one entry * in the cache per device and block.) Returns NULL if no such cache entry * exists. The returned cache entry is locked for exclusive access ("single * writer"). */ struct mb_cache_entry * mb_cache_entry_get(struct mb_cache *cache, struct block_device *bdev, sector_t block) { unsigned int bucket; |
3e037e521
|
646 |
struct hlist_bl_node *l; |
1da177e4c
|
647 |
struct mb_cache_entry *ce; |
3e037e521
|
648 |
struct hlist_bl_head *block_hash_p; |
1da177e4c
|
649 650 651 |
bucket = hash_long((unsigned long)bdev + (block & 0xffffffff), cache->c_bucket_bits); |
3e037e521
|
652 |
block_hash_p = &cache->c_block_hash[bucket]; |
1f3e55fe0
|
653 654 |
/* First serialize access to the block corresponding hash chain. */ hlist_bl_lock(block_hash_p); |
3e037e521
|
655 656 |
hlist_bl_for_each_entry(ce, l, block_hash_p, e_block_list) { mb_assert(ce->e_block_hash_p == block_hash_p); |
1da177e4c
|
657 |
if (ce->e_bdev == bdev && ce->e_block == block) { |
1f3e55fe0
|
658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 |
/* * Prevent a free from removing the entry. */ atomic_inc(&ce->e_refcnt); hlist_bl_unlock(block_hash_p); __spin_lock_mb_cache_entry(ce); atomic_dec(&ce->e_refcnt); if (ce->e_used > 0) { DEFINE_WAIT(wait); while (ce->e_used > 0) { ce->e_queued++; prepare_to_wait(&mb_cache_queue, &wait, TASK_UNINTERRUPTIBLE); __spin_unlock_mb_cache_entry(ce); schedule(); __spin_lock_mb_cache_entry(ce); ce->e_queued--; } finish_wait(&mb_cache_queue, &wait); } ce->e_used += 1 + MB_CACHE_WRITER; __spin_unlock_mb_cache_entry(ce); |
1da177e4c
|
680 |
|
1f3e55fe0
|
681 682 |
if (!list_empty(&ce->e_lru_list)) { spin_lock(&mb_cache_spinlock); |
1da177e4c
|
683 |
list_del_init(&ce->e_lru_list); |
1da177e4c
|
684 |
spin_unlock(&mb_cache_spinlock); |
1da177e4c
|
685 |
} |
3e037e521
|
686 |
if (!__mb_cache_entry_is_block_hashed(ce)) { |
1f3e55fe0
|
687 |
__mb_cache_entry_release(ce); |
1da177e4c
|
688 689 |
return NULL; } |
1f3e55fe0
|
690 |
return ce; |
1da177e4c
|
691 692 |
} } |
1f3e55fe0
|
693 694 |
hlist_bl_unlock(block_hash_p); return NULL; |
1da177e4c
|
695 696 697 698 699 |
} #if !defined(MB_CACHE_INDEXES_COUNT) || (MB_CACHE_INDEXES_COUNT > 0) static struct mb_cache_entry * |
3e037e521
|
700 |
__mb_cache_entry_find(struct hlist_bl_node *l, struct hlist_bl_head *head, |
2aec7c523
|
701 |
struct block_device *bdev, unsigned int key) |
1da177e4c
|
702 |
{ |
1f3e55fe0
|
703 704 |
/* The index hash chain is alredy acquire by caller. */ |
3e037e521
|
705 |
while (l != NULL) { |
1da177e4c
|
706 |
struct mb_cache_entry *ce = |
3e037e521
|
707 708 709 |
hlist_bl_entry(l, struct mb_cache_entry, e_index.o_list); mb_assert(ce->e_index_hash_p == head); |
2aec7c523
|
710 |
if (ce->e_bdev == bdev && ce->e_index.o_key == key) { |
1f3e55fe0
|
711 712 713 714 715 716 717 718 |
/* * Prevent a free from removing the entry. */ atomic_inc(&ce->e_refcnt); hlist_bl_unlock(head); __spin_lock_mb_cache_entry(ce); atomic_dec(&ce->e_refcnt); ce->e_used++; |
1da177e4c
|
719 720 |
/* Incrementing before holding the lock gives readers priority over writers. */ |
1f3e55fe0
|
721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 |
if (ce->e_used >= MB_CACHE_WRITER) { DEFINE_WAIT(wait); while (ce->e_used >= MB_CACHE_WRITER) { ce->e_queued++; prepare_to_wait(&mb_cache_queue, &wait, TASK_UNINTERRUPTIBLE); __spin_unlock_mb_cache_entry(ce); schedule(); __spin_lock_mb_cache_entry(ce); ce->e_queued--; } finish_wait(&mb_cache_queue, &wait); } __spin_unlock_mb_cache_entry(ce); if (!list_empty(&ce->e_lru_list)) { |
1da177e4c
|
737 |
spin_lock(&mb_cache_spinlock); |
1f3e55fe0
|
738 739 |
list_del_init(&ce->e_lru_list); spin_unlock(&mb_cache_spinlock); |
1da177e4c
|
740 |
} |
3e037e521
|
741 |
if (!__mb_cache_entry_is_block_hashed(ce)) { |
1f3e55fe0
|
742 |
__mb_cache_entry_release(ce); |
1da177e4c
|
743 744 745 746 747 748 |
return ERR_PTR(-EAGAIN); } return ce; } l = l->next; } |
1f3e55fe0
|
749 |
hlist_bl_unlock(head); |
1da177e4c
|
750 751 752 753 754 755 756 757 |
return NULL; } /* * mb_cache_entry_find_first() * * Find the first cache entry on a given device with a certain key in |
25985edce
|
758 |
* an additional index. Additional matches can be found with |
1da177e4c
|
759 760 761 762 |
* mb_cache_entry_find_next(). Returns NULL if no match was found. The * returned cache entry is locked for shared access ("multiple readers"). * * @cache: the cache to search |
1da177e4c
|
763 764 765 766 |
* @bdev: the device the cache entry should belong to * @key: the key in the index */ struct mb_cache_entry * |
2aec7c523
|
767 768 |
mb_cache_entry_find_first(struct mb_cache *cache, struct block_device *bdev, unsigned int key) |
1da177e4c
|
769 770 |
{ unsigned int bucket = hash_long(key, cache->c_bucket_bits); |
3e037e521
|
771 772 773 |
struct hlist_bl_node *l; struct mb_cache_entry *ce = NULL; struct hlist_bl_head *index_hash_p; |
1da177e4c
|
774 |
|
3e037e521
|
775 |
index_hash_p = &cache->c_index_hash[bucket]; |
1f3e55fe0
|
776 |
hlist_bl_lock(index_hash_p); |
3e037e521
|
777 778 779 |
if (!hlist_bl_empty(index_hash_p)) { l = hlist_bl_first(index_hash_p); ce = __mb_cache_entry_find(l, index_hash_p, bdev, key); |
1f3e55fe0
|
780 781 |
} else hlist_bl_unlock(index_hash_p); |
1da177e4c
|
782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 |
return ce; } /* * mb_cache_entry_find_next() * * Find the next cache entry on a given device with a certain key in an * additional index. Returns NULL if no match could be found. The previous * entry is atomatically released, so that mb_cache_entry_find_next() can * be called like this: * * entry = mb_cache_entry_find_first(); * while (entry) { * ... * entry = mb_cache_entry_find_next(entry, ...); * } * * @prev: The previous match |
1da177e4c
|
801 802 803 804 |
* @bdev: the device the cache entry should belong to * @key: the key in the index */ struct mb_cache_entry * |
2aec7c523
|
805 |
mb_cache_entry_find_next(struct mb_cache_entry *prev, |
1da177e4c
|
806 807 808 809 |
struct block_device *bdev, unsigned int key) { struct mb_cache *cache = prev->e_cache; unsigned int bucket = hash_long(key, cache->c_bucket_bits); |
3e037e521
|
810 |
struct hlist_bl_node *l; |
1da177e4c
|
811 |
struct mb_cache_entry *ce; |
3e037e521
|
812 |
struct hlist_bl_head *index_hash_p; |
1da177e4c
|
813 |
|
3e037e521
|
814 815 |
index_hash_p = &cache->c_index_hash[bucket]; mb_assert(prev->e_index_hash_p == index_hash_p); |
1f3e55fe0
|
816 |
hlist_bl_lock(index_hash_p); |
3e037e521
|
817 |
mb_assert(!hlist_bl_empty(index_hash_p)); |
2aec7c523
|
818 |
l = prev->e_index.o_list.next; |
3e037e521
|
819 |
ce = __mb_cache_entry_find(l, index_hash_p, bdev, key); |
1f3e55fe0
|
820 |
__mb_cache_entry_release(prev); |
1da177e4c
|
821 822 823 824 825 826 827 |
return ce; } #endif /* !defined(MB_CACHE_INDEXES_COUNT) || (MB_CACHE_INDEXES_COUNT > 0) */ static int __init init_mbcache(void) { |
8e1f936b7
|
828 |
register_shrinker(&mb_cache_shrinker); |
1da177e4c
|
829 830 831 832 833 |
return 0; } static void __exit exit_mbcache(void) { |
8e1f936b7
|
834 |
unregister_shrinker(&mb_cache_shrinker); |
1da177e4c
|
835 836 837 838 |
} module_init(init_mbcache) module_exit(exit_mbcache) |