Mike Snitzer | 4f81a41 | 2012-10-12 21:02:13 +0100 | [diff] [blame] | 1 | /* |
| 2 | * Copyright (C) 2012 Red Hat, Inc. |
| 3 | * |
| 4 | * This file is released under the GPL. |
| 5 | */ |
| 6 | |
| 7 | #include "dm.h" |
| 8 | #include "dm-bio-prison.h" |
| 9 | |
| 10 | #include <linux/spinlock.h> |
| 11 | #include <linux/mempool.h> |
| 12 | #include <linux/module.h> |
| 13 | #include <linux/slab.h> |
| 14 | |
| 15 | /*----------------------------------------------------------------*/ |
| 16 | |
| 17 | struct dm_bio_prison_cell { |
| 18 | struct hlist_node list; |
| 19 | struct dm_bio_prison *prison; |
| 20 | struct dm_cell_key key; |
| 21 | struct bio *holder; |
| 22 | struct bio_list bios; |
| 23 | }; |
| 24 | |
| 25 | struct dm_bio_prison { |
| 26 | spinlock_t lock; |
| 27 | mempool_t *cell_pool; |
| 28 | |
| 29 | unsigned nr_buckets; |
| 30 | unsigned hash_mask; |
| 31 | struct hlist_head *cells; |
| 32 | }; |
| 33 | |
| 34 | /*----------------------------------------------------------------*/ |
| 35 | |
| 36 | static uint32_t calc_nr_buckets(unsigned nr_cells) |
| 37 | { |
| 38 | uint32_t n = 128; |
| 39 | |
| 40 | nr_cells /= 4; |
| 41 | nr_cells = min(nr_cells, 8192u); |
| 42 | |
| 43 | while (n < nr_cells) |
| 44 | n <<= 1; |
| 45 | |
| 46 | return n; |
| 47 | } |
| 48 | |
| 49 | static struct kmem_cache *_cell_cache; |
| 50 | |
| 51 | /* |
| 52 | * @nr_cells should be the number of cells you want in use _concurrently_. |
| 53 | * Don't confuse it with the number of distinct keys. |
| 54 | */ |
| 55 | struct dm_bio_prison *dm_bio_prison_create(unsigned nr_cells) |
| 56 | { |
| 57 | unsigned i; |
| 58 | uint32_t nr_buckets = calc_nr_buckets(nr_cells); |
| 59 | size_t len = sizeof(struct dm_bio_prison) + |
| 60 | (sizeof(struct hlist_head) * nr_buckets); |
| 61 | struct dm_bio_prison *prison = kmalloc(len, GFP_KERNEL); |
| 62 | |
| 63 | if (!prison) |
| 64 | return NULL; |
| 65 | |
| 66 | spin_lock_init(&prison->lock); |
| 67 | prison->cell_pool = mempool_create_slab_pool(nr_cells, _cell_cache); |
| 68 | if (!prison->cell_pool) { |
| 69 | kfree(prison); |
| 70 | return NULL; |
| 71 | } |
| 72 | |
| 73 | prison->nr_buckets = nr_buckets; |
| 74 | prison->hash_mask = nr_buckets - 1; |
| 75 | prison->cells = (struct hlist_head *) (prison + 1); |
| 76 | for (i = 0; i < nr_buckets; i++) |
| 77 | INIT_HLIST_HEAD(prison->cells + i); |
| 78 | |
| 79 | return prison; |
| 80 | } |
| 81 | EXPORT_SYMBOL_GPL(dm_bio_prison_create); |
| 82 | |
| 83 | void dm_bio_prison_destroy(struct dm_bio_prison *prison) |
| 84 | { |
| 85 | mempool_destroy(prison->cell_pool); |
| 86 | kfree(prison); |
| 87 | } |
| 88 | EXPORT_SYMBOL_GPL(dm_bio_prison_destroy); |
| 89 | |
| 90 | static uint32_t hash_key(struct dm_bio_prison *prison, struct dm_cell_key *key) |
| 91 | { |
| 92 | const unsigned long BIG_PRIME = 4294967291UL; |
| 93 | uint64_t hash = key->block * BIG_PRIME; |
| 94 | |
| 95 | return (uint32_t) (hash & prison->hash_mask); |
| 96 | } |
| 97 | |
| 98 | static int keys_equal(struct dm_cell_key *lhs, struct dm_cell_key *rhs) |
| 99 | { |
| 100 | return (lhs->virtual == rhs->virtual) && |
| 101 | (lhs->dev == rhs->dev) && |
| 102 | (lhs->block == rhs->block); |
| 103 | } |
| 104 | |
| 105 | static struct dm_bio_prison_cell *__search_bucket(struct hlist_head *bucket, |
| 106 | struct dm_cell_key *key) |
| 107 | { |
| 108 | struct dm_bio_prison_cell *cell; |
| 109 | struct hlist_node *tmp; |
| 110 | |
| 111 | hlist_for_each_entry(cell, tmp, bucket, list) |
| 112 | if (keys_equal(&cell->key, key)) |
| 113 | return cell; |
| 114 | |
| 115 | return NULL; |
| 116 | } |
| 117 | |
| 118 | /* |
| 119 | * This may block if a new cell needs allocating. You must ensure that |
| 120 | * cells will be unlocked even if the calling thread is blocked. |
| 121 | * |
| 122 | * Returns 1 if the cell was already held, 0 if @inmate is the new holder. |
| 123 | */ |
| 124 | int dm_bio_detain(struct dm_bio_prison *prison, struct dm_cell_key *key, |
| 125 | struct bio *inmate, struct dm_bio_prison_cell **ref) |
| 126 | { |
| 127 | int r = 1; |
| 128 | unsigned long flags; |
| 129 | uint32_t hash = hash_key(prison, key); |
| 130 | struct dm_bio_prison_cell *cell, *cell2; |
| 131 | |
| 132 | BUG_ON(hash > prison->nr_buckets); |
| 133 | |
| 134 | spin_lock_irqsave(&prison->lock, flags); |
| 135 | |
| 136 | cell = __search_bucket(prison->cells + hash, key); |
| 137 | if (cell) { |
| 138 | bio_list_add(&cell->bios, inmate); |
| 139 | goto out; |
| 140 | } |
| 141 | |
| 142 | /* |
| 143 | * Allocate a new cell |
| 144 | */ |
| 145 | spin_unlock_irqrestore(&prison->lock, flags); |
| 146 | cell2 = mempool_alloc(prison->cell_pool, GFP_NOIO); |
| 147 | spin_lock_irqsave(&prison->lock, flags); |
| 148 | |
| 149 | /* |
| 150 | * We've been unlocked, so we have to double check that |
| 151 | * nobody else has inserted this cell in the meantime. |
| 152 | */ |
| 153 | cell = __search_bucket(prison->cells + hash, key); |
| 154 | if (cell) { |
| 155 | mempool_free(cell2, prison->cell_pool); |
| 156 | bio_list_add(&cell->bios, inmate); |
| 157 | goto out; |
| 158 | } |
| 159 | |
| 160 | /* |
| 161 | * Use new cell. |
| 162 | */ |
| 163 | cell = cell2; |
| 164 | |
| 165 | cell->prison = prison; |
| 166 | memcpy(&cell->key, key, sizeof(cell->key)); |
| 167 | cell->holder = inmate; |
| 168 | bio_list_init(&cell->bios); |
| 169 | hlist_add_head(&cell->list, prison->cells + hash); |
| 170 | |
| 171 | r = 0; |
| 172 | |
| 173 | out: |
| 174 | spin_unlock_irqrestore(&prison->lock, flags); |
| 175 | |
| 176 | *ref = cell; |
| 177 | |
| 178 | return r; |
| 179 | } |
| 180 | EXPORT_SYMBOL_GPL(dm_bio_detain); |
| 181 | |
| 182 | /* |
| 183 | * @inmates must have been initialised prior to this call |
| 184 | */ |
| 185 | static void __cell_release(struct dm_bio_prison_cell *cell, struct bio_list *inmates) |
| 186 | { |
| 187 | struct dm_bio_prison *prison = cell->prison; |
| 188 | |
| 189 | hlist_del(&cell->list); |
| 190 | |
| 191 | if (inmates) { |
| 192 | bio_list_add(inmates, cell->holder); |
| 193 | bio_list_merge(inmates, &cell->bios); |
| 194 | } |
| 195 | |
| 196 | mempool_free(cell, prison->cell_pool); |
| 197 | } |
| 198 | |
| 199 | void dm_cell_release(struct dm_bio_prison_cell *cell, struct bio_list *bios) |
| 200 | { |
| 201 | unsigned long flags; |
| 202 | struct dm_bio_prison *prison = cell->prison; |
| 203 | |
| 204 | spin_lock_irqsave(&prison->lock, flags); |
| 205 | __cell_release(cell, bios); |
| 206 | spin_unlock_irqrestore(&prison->lock, flags); |
| 207 | } |
| 208 | EXPORT_SYMBOL_GPL(dm_cell_release); |
| 209 | |
| 210 | /* |
Mike Snitzer | 4f81a41 | 2012-10-12 21:02:13 +0100 | [diff] [blame] | 211 | * Sometimes we don't want the holder, just the additional bios. |
| 212 | */ |
| 213 | static void __cell_release_no_holder(struct dm_bio_prison_cell *cell, struct bio_list *inmates) |
| 214 | { |
| 215 | struct dm_bio_prison *prison = cell->prison; |
| 216 | |
| 217 | hlist_del(&cell->list); |
| 218 | bio_list_merge(inmates, &cell->bios); |
| 219 | |
| 220 | mempool_free(cell, prison->cell_pool); |
| 221 | } |
| 222 | |
| 223 | void dm_cell_release_no_holder(struct dm_bio_prison_cell *cell, struct bio_list *inmates) |
| 224 | { |
| 225 | unsigned long flags; |
| 226 | struct dm_bio_prison *prison = cell->prison; |
| 227 | |
| 228 | spin_lock_irqsave(&prison->lock, flags); |
| 229 | __cell_release_no_holder(cell, inmates); |
| 230 | spin_unlock_irqrestore(&prison->lock, flags); |
| 231 | } |
| 232 | EXPORT_SYMBOL_GPL(dm_cell_release_no_holder); |
| 233 | |
| 234 | void dm_cell_error(struct dm_bio_prison_cell *cell) |
| 235 | { |
| 236 | struct dm_bio_prison *prison = cell->prison; |
| 237 | struct bio_list bios; |
| 238 | struct bio *bio; |
| 239 | unsigned long flags; |
| 240 | |
| 241 | bio_list_init(&bios); |
| 242 | |
| 243 | spin_lock_irqsave(&prison->lock, flags); |
| 244 | __cell_release(cell, &bios); |
| 245 | spin_unlock_irqrestore(&prison->lock, flags); |
| 246 | |
| 247 | while ((bio = bio_list_pop(&bios))) |
| 248 | bio_io_error(bio); |
| 249 | } |
| 250 | EXPORT_SYMBOL_GPL(dm_cell_error); |
| 251 | |
| 252 | /*----------------------------------------------------------------*/ |
| 253 | |
| 254 | #define DEFERRED_SET_SIZE 64 |
| 255 | |
| 256 | struct dm_deferred_entry { |
| 257 | struct dm_deferred_set *ds; |
| 258 | unsigned count; |
| 259 | struct list_head work_items; |
| 260 | }; |
| 261 | |
| 262 | struct dm_deferred_set { |
| 263 | spinlock_t lock; |
| 264 | unsigned current_entry; |
| 265 | unsigned sweeper; |
| 266 | struct dm_deferred_entry entries[DEFERRED_SET_SIZE]; |
| 267 | }; |
| 268 | |
| 269 | struct dm_deferred_set *dm_deferred_set_create(void) |
| 270 | { |
| 271 | int i; |
| 272 | struct dm_deferred_set *ds; |
| 273 | |
| 274 | ds = kmalloc(sizeof(*ds), GFP_KERNEL); |
| 275 | if (!ds) |
| 276 | return NULL; |
| 277 | |
| 278 | spin_lock_init(&ds->lock); |
| 279 | ds->current_entry = 0; |
| 280 | ds->sweeper = 0; |
| 281 | for (i = 0; i < DEFERRED_SET_SIZE; i++) { |
| 282 | ds->entries[i].ds = ds; |
| 283 | ds->entries[i].count = 0; |
| 284 | INIT_LIST_HEAD(&ds->entries[i].work_items); |
| 285 | } |
| 286 | |
| 287 | return ds; |
| 288 | } |
| 289 | EXPORT_SYMBOL_GPL(dm_deferred_set_create); |
| 290 | |
| 291 | void dm_deferred_set_destroy(struct dm_deferred_set *ds) |
| 292 | { |
| 293 | kfree(ds); |
| 294 | } |
| 295 | EXPORT_SYMBOL_GPL(dm_deferred_set_destroy); |
| 296 | |
| 297 | struct dm_deferred_entry *dm_deferred_entry_inc(struct dm_deferred_set *ds) |
| 298 | { |
| 299 | unsigned long flags; |
| 300 | struct dm_deferred_entry *entry; |
| 301 | |
| 302 | spin_lock_irqsave(&ds->lock, flags); |
| 303 | entry = ds->entries + ds->current_entry; |
| 304 | entry->count++; |
| 305 | spin_unlock_irqrestore(&ds->lock, flags); |
| 306 | |
| 307 | return entry; |
| 308 | } |
| 309 | EXPORT_SYMBOL_GPL(dm_deferred_entry_inc); |
| 310 | |
| 311 | static unsigned ds_next(unsigned index) |
| 312 | { |
| 313 | return (index + 1) % DEFERRED_SET_SIZE; |
| 314 | } |
| 315 | |
| 316 | static void __sweep(struct dm_deferred_set *ds, struct list_head *head) |
| 317 | { |
| 318 | while ((ds->sweeper != ds->current_entry) && |
| 319 | !ds->entries[ds->sweeper].count) { |
| 320 | list_splice_init(&ds->entries[ds->sweeper].work_items, head); |
| 321 | ds->sweeper = ds_next(ds->sweeper); |
| 322 | } |
| 323 | |
| 324 | if ((ds->sweeper == ds->current_entry) && !ds->entries[ds->sweeper].count) |
| 325 | list_splice_init(&ds->entries[ds->sweeper].work_items, head); |
| 326 | } |
| 327 | |
| 328 | void dm_deferred_entry_dec(struct dm_deferred_entry *entry, struct list_head *head) |
| 329 | { |
| 330 | unsigned long flags; |
| 331 | |
| 332 | spin_lock_irqsave(&entry->ds->lock, flags); |
| 333 | BUG_ON(!entry->count); |
| 334 | --entry->count; |
| 335 | __sweep(entry->ds, head); |
| 336 | spin_unlock_irqrestore(&entry->ds->lock, flags); |
| 337 | } |
| 338 | EXPORT_SYMBOL_GPL(dm_deferred_entry_dec); |
| 339 | |
| 340 | /* |
| 341 | * Returns 1 if deferred or 0 if no pending items to delay job. |
| 342 | */ |
| 343 | int dm_deferred_set_add_work(struct dm_deferred_set *ds, struct list_head *work) |
| 344 | { |
| 345 | int r = 1; |
| 346 | unsigned long flags; |
| 347 | unsigned next_entry; |
| 348 | |
| 349 | spin_lock_irqsave(&ds->lock, flags); |
| 350 | if ((ds->sweeper == ds->current_entry) && |
| 351 | !ds->entries[ds->current_entry].count) |
| 352 | r = 0; |
| 353 | else { |
| 354 | list_add(work, &ds->entries[ds->current_entry].work_items); |
| 355 | next_entry = ds_next(ds->current_entry); |
| 356 | if (!ds->entries[next_entry].count) |
| 357 | ds->current_entry = next_entry; |
| 358 | } |
| 359 | spin_unlock_irqrestore(&ds->lock, flags); |
| 360 | |
| 361 | return r; |
| 362 | } |
| 363 | EXPORT_SYMBOL_GPL(dm_deferred_set_add_work); |
| 364 | |
| 365 | /*----------------------------------------------------------------*/ |
| 366 | |
| 367 | static int __init dm_bio_prison_init(void) |
| 368 | { |
| 369 | _cell_cache = KMEM_CACHE(dm_bio_prison_cell, 0); |
| 370 | if (!_cell_cache) |
| 371 | return -ENOMEM; |
| 372 | |
| 373 | return 0; |
| 374 | } |
| 375 | |
| 376 | static void __exit dm_bio_prison_exit(void) |
| 377 | { |
| 378 | kmem_cache_destroy(_cell_cache); |
| 379 | _cell_cache = NULL; |
| 380 | } |
| 381 | |
| 382 | /* |
| 383 | * module hooks |
| 384 | */ |
| 385 | module_init(dm_bio_prison_init); |
| 386 | module_exit(dm_bio_prison_exit); |
| 387 | |
| 388 | MODULE_DESCRIPTION(DM_NAME " bio prison"); |
| 389 | MODULE_AUTHOR("Joe Thornber <dm-devel@redhat.com>"); |
| 390 | MODULE_LICENSE("GPL"); |