| rjw | 1f88458 | 2022-01-06 17:20:42 +0800 | [diff] [blame] | 1 | /* | 
|  | 2 | * Copyright (C) 2002 Sistina Software (UK) Limited. | 
|  | 3 | * Copyright (C) 2006 Red Hat GmbH | 
|  | 4 | * | 
|  | 5 | * This file is released under the GPL. | 
|  | 6 | * | 
|  | 7 | * Kcopyd provides a simple interface for copying an area of one | 
|  | 8 | * block-device to one or more other block-devices, with an asynchronous | 
|  | 9 | * completion notification. | 
|  | 10 | */ | 
|  | 11 |  | 
|  | 12 | #include <linux/types.h> | 
|  | 13 | #include <linux/atomic.h> | 
|  | 14 | #include <linux/blkdev.h> | 
|  | 15 | #include <linux/fs.h> | 
|  | 16 | #include <linux/init.h> | 
|  | 17 | #include <linux/list.h> | 
|  | 18 | #include <linux/mempool.h> | 
|  | 19 | #include <linux/module.h> | 
|  | 20 | #include <linux/pagemap.h> | 
|  | 21 | #include <linux/slab.h> | 
|  | 22 | #include <linux/vmalloc.h> | 
|  | 23 | #include <linux/workqueue.h> | 
|  | 24 | #include <linux/mutex.h> | 
|  | 25 | #include <linux/delay.h> | 
|  | 26 | #include <linux/device-mapper.h> | 
|  | 27 | #include <linux/dm-kcopyd.h> | 
|  | 28 |  | 
|  | 29 | #include "dm-core.h" | 
|  | 30 |  | 
|  | 31 | #define SUB_JOB_SIZE	128 | 
|  | 32 | #define SPLIT_COUNT	8 | 
|  | 33 | #define MIN_JOBS	8 | 
|  | 34 | #define RESERVE_PAGES	(DIV_ROUND_UP(SUB_JOB_SIZE << SECTOR_SHIFT, PAGE_SIZE)) | 
|  | 35 |  | 
|  | 36 | /*----------------------------------------------------------------- | 
|  | 37 | * Each kcopyd client has its own little pool of preallocated | 
|  | 38 | * pages for kcopyd io. | 
|  | 39 | *---------------------------------------------------------------*/ | 
|  | 40 | struct dm_kcopyd_client { | 
|  | 41 | struct page_list *pages; | 
|  | 42 | unsigned nr_reserved_pages; | 
|  | 43 | unsigned nr_free_pages; | 
|  | 44 |  | 
|  | 45 | struct dm_io_client *io_client; | 
|  | 46 |  | 
|  | 47 | wait_queue_head_t destroyq; | 
|  | 48 | atomic_t nr_jobs; | 
|  | 49 |  | 
|  | 50 | mempool_t *job_pool; | 
|  | 51 |  | 
|  | 52 | struct workqueue_struct *kcopyd_wq; | 
|  | 53 | struct work_struct kcopyd_work; | 
|  | 54 |  | 
|  | 55 | struct dm_kcopyd_throttle *throttle; | 
|  | 56 |  | 
|  | 57 | /* | 
|  | 58 | * We maintain four lists of jobs: | 
|  | 59 | * | 
|  | 60 | * i)   jobs waiting for pages | 
|  | 61 | * ii)  jobs that have pages, and are waiting for the io to be issued. | 
|  | 62 | * iii) jobs that don't need to do any IO and just run a callback | 
|  | 63 | * iv) jobs that have completed. | 
|  | 64 | * | 
|  | 65 | * All four of these are protected by job_lock. | 
|  | 66 | */ | 
|  | 67 | spinlock_t job_lock; | 
|  | 68 | struct list_head callback_jobs; | 
|  | 69 | struct list_head complete_jobs; | 
|  | 70 | struct list_head io_jobs; | 
|  | 71 | struct list_head pages_jobs; | 
|  | 72 | }; | 
|  | 73 |  | 
|  | 74 | static struct page_list zero_page_list; | 
|  | 75 |  | 
|  | 76 | static DEFINE_SPINLOCK(throttle_spinlock); | 
|  | 77 |  | 
|  | 78 | /* | 
|  | 79 | * IO/IDLE accounting slowly decays after (1 << ACCOUNT_INTERVAL_SHIFT) period. | 
|  | 80 | * When total_period >= (1 << ACCOUNT_INTERVAL_SHIFT) the counters are divided | 
|  | 81 | * by 2. | 
|  | 82 | */ | 
|  | 83 | #define ACCOUNT_INTERVAL_SHIFT		SHIFT_HZ | 
|  | 84 |  | 
|  | 85 | /* | 
|  | 86 | * Sleep this number of milliseconds. | 
|  | 87 | * | 
|  | 88 | * The value was decided experimentally. | 
|  | 89 | * Smaller values seem to cause an increased copy rate above the limit. | 
|  | 90 | * The reason for this is unknown but possibly due to jiffies rounding errors | 
|  | 91 | * or read/write cache inside the disk. | 
|  | 92 | */ | 
|  | 93 | #define SLEEP_MSEC			100 | 
|  | 94 |  | 
|  | 95 | /* | 
|  | 96 | * Maximum number of sleep events. There is a theoretical livelock if more | 
|  | 97 | * kcopyd clients do work simultaneously which this limit avoids. | 
|  | 98 | */ | 
|  | 99 | #define MAX_SLEEPS			10 | 
|  | 100 |  | 
|  | 101 | static void io_job_start(struct dm_kcopyd_throttle *t) | 
|  | 102 | { | 
|  | 103 | unsigned throttle, now, difference; | 
|  | 104 | int slept = 0, skew; | 
|  | 105 |  | 
|  | 106 | if (unlikely(!t)) | 
|  | 107 | return; | 
|  | 108 |  | 
|  | 109 | try_again: | 
|  | 110 | spin_lock_irq(&throttle_spinlock); | 
|  | 111 |  | 
|  | 112 | throttle = ACCESS_ONCE(t->throttle); | 
|  | 113 |  | 
|  | 114 | if (likely(throttle >= 100)) | 
|  | 115 | goto skip_limit; | 
|  | 116 |  | 
|  | 117 | now = jiffies; | 
|  | 118 | difference = now - t->last_jiffies; | 
|  | 119 | t->last_jiffies = now; | 
|  | 120 | if (t->num_io_jobs) | 
|  | 121 | t->io_period += difference; | 
|  | 122 | t->total_period += difference; | 
|  | 123 |  | 
|  | 124 | /* | 
|  | 125 | * Maintain sane values if we got a temporary overflow. | 
|  | 126 | */ | 
|  | 127 | if (unlikely(t->io_period > t->total_period)) | 
|  | 128 | t->io_period = t->total_period; | 
|  | 129 |  | 
|  | 130 | if (unlikely(t->total_period >= (1 << ACCOUNT_INTERVAL_SHIFT))) { | 
|  | 131 | int shift = fls(t->total_period >> ACCOUNT_INTERVAL_SHIFT); | 
|  | 132 | t->total_period >>= shift; | 
|  | 133 | t->io_period >>= shift; | 
|  | 134 | } | 
|  | 135 |  | 
|  | 136 | skew = t->io_period - throttle * t->total_period / 100; | 
|  | 137 |  | 
|  | 138 | if (unlikely(skew > 0) && slept < MAX_SLEEPS) { | 
|  | 139 | slept++; | 
|  | 140 | spin_unlock_irq(&throttle_spinlock); | 
|  | 141 | msleep(SLEEP_MSEC); | 
|  | 142 | goto try_again; | 
|  | 143 | } | 
|  | 144 |  | 
|  | 145 | skip_limit: | 
|  | 146 | t->num_io_jobs++; | 
|  | 147 |  | 
|  | 148 | spin_unlock_irq(&throttle_spinlock); | 
|  | 149 | } | 
|  | 150 |  | 
|  | 151 | static void io_job_finish(struct dm_kcopyd_throttle *t) | 
|  | 152 | { | 
|  | 153 | unsigned long flags; | 
|  | 154 |  | 
|  | 155 | if (unlikely(!t)) | 
|  | 156 | return; | 
|  | 157 |  | 
|  | 158 | spin_lock_irqsave(&throttle_spinlock, flags); | 
|  | 159 |  | 
|  | 160 | t->num_io_jobs--; | 
|  | 161 |  | 
|  | 162 | if (likely(ACCESS_ONCE(t->throttle) >= 100)) | 
|  | 163 | goto skip_limit; | 
|  | 164 |  | 
|  | 165 | if (!t->num_io_jobs) { | 
|  | 166 | unsigned now, difference; | 
|  | 167 |  | 
|  | 168 | now = jiffies; | 
|  | 169 | difference = now - t->last_jiffies; | 
|  | 170 | t->last_jiffies = now; | 
|  | 171 |  | 
|  | 172 | t->io_period += difference; | 
|  | 173 | t->total_period += difference; | 
|  | 174 |  | 
|  | 175 | /* | 
|  | 176 | * Maintain sane values if we got a temporary overflow. | 
|  | 177 | */ | 
|  | 178 | if (unlikely(t->io_period > t->total_period)) | 
|  | 179 | t->io_period = t->total_period; | 
|  | 180 | } | 
|  | 181 |  | 
|  | 182 | skip_limit: | 
|  | 183 | spin_unlock_irqrestore(&throttle_spinlock, flags); | 
|  | 184 | } | 
|  | 185 |  | 
|  | 186 |  | 
|  | 187 | static void wake(struct dm_kcopyd_client *kc) | 
|  | 188 | { | 
|  | 189 | queue_work(kc->kcopyd_wq, &kc->kcopyd_work); | 
|  | 190 | } | 
|  | 191 |  | 
|  | 192 | /* | 
|  | 193 | * Obtain one page for the use of kcopyd. | 
|  | 194 | */ | 
|  | 195 | static struct page_list *alloc_pl(gfp_t gfp) | 
|  | 196 | { | 
|  | 197 | struct page_list *pl; | 
|  | 198 |  | 
|  | 199 | pl = kmalloc(sizeof(*pl), gfp); | 
|  | 200 | if (!pl) | 
|  | 201 | return NULL; | 
|  | 202 |  | 
|  | 203 | pl->page = alloc_page(gfp); | 
|  | 204 | if (!pl->page) { | 
|  | 205 | kfree(pl); | 
|  | 206 | return NULL; | 
|  | 207 | } | 
|  | 208 |  | 
|  | 209 | return pl; | 
|  | 210 | } | 
|  | 211 |  | 
|  | 212 | static void free_pl(struct page_list *pl) | 
|  | 213 | { | 
|  | 214 | __free_page(pl->page); | 
|  | 215 | kfree(pl); | 
|  | 216 | } | 
|  | 217 |  | 
|  | 218 | /* | 
|  | 219 | * Add the provided pages to a client's free page list, releasing | 
|  | 220 | * back to the system any beyond the reserved_pages limit. | 
|  | 221 | */ | 
|  | 222 | static void kcopyd_put_pages(struct dm_kcopyd_client *kc, struct page_list *pl) | 
|  | 223 | { | 
|  | 224 | struct page_list *next; | 
|  | 225 |  | 
|  | 226 | do { | 
|  | 227 | next = pl->next; | 
|  | 228 |  | 
|  | 229 | if (kc->nr_free_pages >= kc->nr_reserved_pages) | 
|  | 230 | free_pl(pl); | 
|  | 231 | else { | 
|  | 232 | pl->next = kc->pages; | 
|  | 233 | kc->pages = pl; | 
|  | 234 | kc->nr_free_pages++; | 
|  | 235 | } | 
|  | 236 |  | 
|  | 237 | pl = next; | 
|  | 238 | } while (pl); | 
|  | 239 | } | 
|  | 240 |  | 
|  | 241 | static int kcopyd_get_pages(struct dm_kcopyd_client *kc, | 
|  | 242 | unsigned int nr, struct page_list **pages) | 
|  | 243 | { | 
|  | 244 | struct page_list *pl; | 
|  | 245 |  | 
|  | 246 | *pages = NULL; | 
|  | 247 |  | 
|  | 248 | do { | 
|  | 249 | pl = alloc_pl(__GFP_NOWARN | __GFP_NORETRY | __GFP_KSWAPD_RECLAIM); | 
|  | 250 | if (unlikely(!pl)) { | 
|  | 251 | /* Use reserved pages */ | 
|  | 252 | pl = kc->pages; | 
|  | 253 | if (unlikely(!pl)) | 
|  | 254 | goto out_of_memory; | 
|  | 255 | kc->pages = pl->next; | 
|  | 256 | kc->nr_free_pages--; | 
|  | 257 | } | 
|  | 258 | pl->next = *pages; | 
|  | 259 | *pages = pl; | 
|  | 260 | } while (--nr); | 
|  | 261 |  | 
|  | 262 | return 0; | 
|  | 263 |  | 
|  | 264 | out_of_memory: | 
|  | 265 | if (*pages) | 
|  | 266 | kcopyd_put_pages(kc, *pages); | 
|  | 267 | return -ENOMEM; | 
|  | 268 | } | 
|  | 269 |  | 
|  | 270 | /* | 
|  | 271 | * These three functions resize the page pool. | 
|  | 272 | */ | 
|  | 273 | static void drop_pages(struct page_list *pl) | 
|  | 274 | { | 
|  | 275 | struct page_list *next; | 
|  | 276 |  | 
|  | 277 | while (pl) { | 
|  | 278 | next = pl->next; | 
|  | 279 | free_pl(pl); | 
|  | 280 | pl = next; | 
|  | 281 | } | 
|  | 282 | } | 
|  | 283 |  | 
|  | 284 | /* | 
|  | 285 | * Allocate and reserve nr_pages for the use of a specific client. | 
|  | 286 | */ | 
|  | 287 | static int client_reserve_pages(struct dm_kcopyd_client *kc, unsigned nr_pages) | 
|  | 288 | { | 
|  | 289 | unsigned i; | 
|  | 290 | struct page_list *pl = NULL, *next; | 
|  | 291 |  | 
|  | 292 | for (i = 0; i < nr_pages; i++) { | 
|  | 293 | next = alloc_pl(GFP_KERNEL); | 
|  | 294 | if (!next) { | 
|  | 295 | if (pl) | 
|  | 296 | drop_pages(pl); | 
|  | 297 | return -ENOMEM; | 
|  | 298 | } | 
|  | 299 | next->next = pl; | 
|  | 300 | pl = next; | 
|  | 301 | } | 
|  | 302 |  | 
|  | 303 | kc->nr_reserved_pages += nr_pages; | 
|  | 304 | kcopyd_put_pages(kc, pl); | 
|  | 305 |  | 
|  | 306 | return 0; | 
|  | 307 | } | 
|  | 308 |  | 
|  | 309 | static void client_free_pages(struct dm_kcopyd_client *kc) | 
|  | 310 | { | 
|  | 311 | BUG_ON(kc->nr_free_pages != kc->nr_reserved_pages); | 
|  | 312 | drop_pages(kc->pages); | 
|  | 313 | kc->pages = NULL; | 
|  | 314 | kc->nr_free_pages = kc->nr_reserved_pages = 0; | 
|  | 315 | } | 
|  | 316 |  | 
|  | 317 | /*----------------------------------------------------------------- | 
|  | 318 | * kcopyd_jobs need to be allocated by the *clients* of kcopyd, | 
|  | 319 | * for this reason we use a mempool to prevent the client from | 
|  | 320 | * ever having to do io (which could cause a deadlock). | 
|  | 321 | *---------------------------------------------------------------*/ | 
|  | 322 | struct kcopyd_job { | 
|  | 323 | struct dm_kcopyd_client *kc; | 
|  | 324 | struct list_head list; | 
|  | 325 | unsigned long flags; | 
|  | 326 |  | 
|  | 327 | /* | 
|  | 328 | * Error state of the job. | 
|  | 329 | */ | 
|  | 330 | int read_err; | 
|  | 331 | unsigned long write_err; | 
|  | 332 |  | 
|  | 333 | /* | 
|  | 334 | * Either READ or WRITE | 
|  | 335 | */ | 
|  | 336 | int rw; | 
|  | 337 | struct dm_io_region source; | 
|  | 338 |  | 
|  | 339 | /* | 
|  | 340 | * The destinations for the transfer. | 
|  | 341 | */ | 
|  | 342 | unsigned int num_dests; | 
|  | 343 | struct dm_io_region dests[DM_KCOPYD_MAX_REGIONS]; | 
|  | 344 |  | 
|  | 345 | struct page_list *pages; | 
|  | 346 |  | 
|  | 347 | /* | 
|  | 348 | * Set this to ensure you are notified when the job has | 
|  | 349 | * completed.  'context' is for callback to use. | 
|  | 350 | */ | 
|  | 351 | dm_kcopyd_notify_fn fn; | 
|  | 352 | void *context; | 
|  | 353 |  | 
|  | 354 | /* | 
|  | 355 | * These fields are only used if the job has been split | 
|  | 356 | * into more manageable parts. | 
|  | 357 | */ | 
|  | 358 | struct mutex lock; | 
|  | 359 | atomic_t sub_jobs; | 
|  | 360 | sector_t progress; | 
|  | 361 | sector_t write_offset; | 
|  | 362 |  | 
|  | 363 | struct kcopyd_job *master_job; | 
|  | 364 | }; | 
|  | 365 |  | 
|  | 366 | static struct kmem_cache *_job_cache; | 
|  | 367 |  | 
|  | 368 | int __init dm_kcopyd_init(void) | 
|  | 369 | { | 
|  | 370 | _job_cache = kmem_cache_create("kcopyd_job", | 
|  | 371 | sizeof(struct kcopyd_job) * (SPLIT_COUNT + 1), | 
|  | 372 | __alignof__(struct kcopyd_job), 0, NULL); | 
|  | 373 | if (!_job_cache) | 
|  | 374 | return -ENOMEM; | 
|  | 375 |  | 
|  | 376 | zero_page_list.next = &zero_page_list; | 
|  | 377 | zero_page_list.page = ZERO_PAGE(0); | 
|  | 378 |  | 
|  | 379 | return 0; | 
|  | 380 | } | 
|  | 381 |  | 
|  | 382 | void dm_kcopyd_exit(void) | 
|  | 383 | { | 
|  | 384 | kmem_cache_destroy(_job_cache); | 
|  | 385 | _job_cache = NULL; | 
|  | 386 | } | 
|  | 387 |  | 
|  | 388 | /* | 
|  | 389 | * Functions to push and pop a job onto the head of a given job | 
|  | 390 | * list. | 
|  | 391 | */ | 
|  | 392 | static struct kcopyd_job *pop_io_job(struct list_head *jobs, | 
|  | 393 | struct dm_kcopyd_client *kc) | 
|  | 394 | { | 
|  | 395 | struct kcopyd_job *job; | 
|  | 396 |  | 
|  | 397 | /* | 
|  | 398 | * For I/O jobs, pop any read, any write without sequential write | 
|  | 399 | * constraint and sequential writes that are at the right position. | 
|  | 400 | */ | 
|  | 401 | list_for_each_entry(job, jobs, list) { | 
|  | 402 | if (job->rw == READ || !test_bit(DM_KCOPYD_WRITE_SEQ, &job->flags)) { | 
|  | 403 | list_del(&job->list); | 
|  | 404 | return job; | 
|  | 405 | } | 
|  | 406 |  | 
|  | 407 | if (job->write_offset == job->master_job->write_offset) { | 
|  | 408 | job->master_job->write_offset += job->source.count; | 
|  | 409 | list_del(&job->list); | 
|  | 410 | return job; | 
|  | 411 | } | 
|  | 412 | } | 
|  | 413 |  | 
|  | 414 | return NULL; | 
|  | 415 | } | 
|  | 416 |  | 
|  | 417 | static struct kcopyd_job *pop(struct list_head *jobs, | 
|  | 418 | struct dm_kcopyd_client *kc) | 
|  | 419 | { | 
|  | 420 | struct kcopyd_job *job = NULL; | 
|  | 421 | unsigned long flags; | 
|  | 422 |  | 
|  | 423 | spin_lock_irqsave(&kc->job_lock, flags); | 
|  | 424 |  | 
|  | 425 | if (!list_empty(jobs)) { | 
|  | 426 | if (jobs == &kc->io_jobs) | 
|  | 427 | job = pop_io_job(jobs, kc); | 
|  | 428 | else { | 
|  | 429 | job = list_entry(jobs->next, struct kcopyd_job, list); | 
|  | 430 | list_del(&job->list); | 
|  | 431 | } | 
|  | 432 | } | 
|  | 433 | spin_unlock_irqrestore(&kc->job_lock, flags); | 
|  | 434 |  | 
|  | 435 | return job; | 
|  | 436 | } | 
|  | 437 |  | 
|  | 438 | static void push(struct list_head *jobs, struct kcopyd_job *job) | 
|  | 439 | { | 
|  | 440 | unsigned long flags; | 
|  | 441 | struct dm_kcopyd_client *kc = job->kc; | 
|  | 442 |  | 
|  | 443 | spin_lock_irqsave(&kc->job_lock, flags); | 
|  | 444 | list_add_tail(&job->list, jobs); | 
|  | 445 | spin_unlock_irqrestore(&kc->job_lock, flags); | 
|  | 446 | } | 
|  | 447 |  | 
|  | 448 |  | 
|  | 449 | static void push_head(struct list_head *jobs, struct kcopyd_job *job) | 
|  | 450 | { | 
|  | 451 | unsigned long flags; | 
|  | 452 | struct dm_kcopyd_client *kc = job->kc; | 
|  | 453 |  | 
|  | 454 | spin_lock_irqsave(&kc->job_lock, flags); | 
|  | 455 | list_add(&job->list, jobs); | 
|  | 456 | spin_unlock_irqrestore(&kc->job_lock, flags); | 
|  | 457 | } | 
|  | 458 |  | 
|  | 459 | /* | 
|  | 460 | * These three functions process 1 item from the corresponding | 
|  | 461 | * job list. | 
|  | 462 | * | 
|  | 463 | * They return: | 
|  | 464 | * < 0: error | 
|  | 465 | *   0: success | 
|  | 466 | * > 0: can't process yet. | 
|  | 467 | */ | 
|  | 468 | static int run_complete_job(struct kcopyd_job *job) | 
|  | 469 | { | 
|  | 470 | void *context = job->context; | 
|  | 471 | int read_err = job->read_err; | 
|  | 472 | unsigned long write_err = job->write_err; | 
|  | 473 | dm_kcopyd_notify_fn fn = job->fn; | 
|  | 474 | struct dm_kcopyd_client *kc = job->kc; | 
|  | 475 |  | 
|  | 476 | if (job->pages && job->pages != &zero_page_list) | 
|  | 477 | kcopyd_put_pages(kc, job->pages); | 
|  | 478 | /* | 
|  | 479 | * If this is the master job, the sub jobs have already | 
|  | 480 | * completed so we can free everything. | 
|  | 481 | */ | 
|  | 482 | if (job->master_job == job) | 
|  | 483 | mempool_free(job, kc->job_pool); | 
|  | 484 | fn(read_err, write_err, context); | 
|  | 485 |  | 
|  | 486 | if (atomic_dec_and_test(&kc->nr_jobs)) | 
|  | 487 | wake_up(&kc->destroyq); | 
|  | 488 |  | 
|  | 489 | cond_resched(); | 
|  | 490 |  | 
|  | 491 | return 0; | 
|  | 492 | } | 
|  | 493 |  | 
|  | 494 | static void complete_io(unsigned long error, void *context) | 
|  | 495 | { | 
|  | 496 | struct kcopyd_job *job = (struct kcopyd_job *) context; | 
|  | 497 | struct dm_kcopyd_client *kc = job->kc; | 
|  | 498 |  | 
|  | 499 | io_job_finish(kc->throttle); | 
|  | 500 |  | 
|  | 501 | if (error) { | 
|  | 502 | if (op_is_write(job->rw)) | 
|  | 503 | job->write_err |= error; | 
|  | 504 | else | 
|  | 505 | job->read_err = 1; | 
|  | 506 |  | 
|  | 507 | if (!test_bit(DM_KCOPYD_IGNORE_ERROR, &job->flags)) { | 
|  | 508 | push(&kc->complete_jobs, job); | 
|  | 509 | wake(kc); | 
|  | 510 | return; | 
|  | 511 | } | 
|  | 512 | } | 
|  | 513 |  | 
|  | 514 | if (op_is_write(job->rw)) | 
|  | 515 | push(&kc->complete_jobs, job); | 
|  | 516 |  | 
|  | 517 | else { | 
|  | 518 | job->rw = WRITE; | 
|  | 519 | push(&kc->io_jobs, job); | 
|  | 520 | } | 
|  | 521 |  | 
|  | 522 | wake(kc); | 
|  | 523 | } | 
|  | 524 |  | 
|  | 525 | /* | 
|  | 526 | * Request io on as many buffer heads as we can currently get for | 
|  | 527 | * a particular job. | 
|  | 528 | */ | 
|  | 529 | static int run_io_job(struct kcopyd_job *job) | 
|  | 530 | { | 
|  | 531 | int r; | 
|  | 532 | struct dm_io_request io_req = { | 
|  | 533 | .bi_op = job->rw, | 
|  | 534 | .bi_op_flags = 0, | 
|  | 535 | .mem.type = DM_IO_PAGE_LIST, | 
|  | 536 | .mem.ptr.pl = job->pages, | 
|  | 537 | .mem.offset = 0, | 
|  | 538 | .notify.fn = complete_io, | 
|  | 539 | .notify.context = job, | 
|  | 540 | .client = job->kc->io_client, | 
|  | 541 | }; | 
|  | 542 |  | 
|  | 543 | /* | 
|  | 544 | * If we need to write sequentially and some reads or writes failed, | 
|  | 545 | * no point in continuing. | 
|  | 546 | */ | 
|  | 547 | if (test_bit(DM_KCOPYD_WRITE_SEQ, &job->flags) && | 
|  | 548 | job->master_job->write_err) { | 
|  | 549 | job->write_err = job->master_job->write_err; | 
|  | 550 | return -EIO; | 
|  | 551 | } | 
|  | 552 |  | 
|  | 553 | io_job_start(job->kc->throttle); | 
|  | 554 |  | 
|  | 555 | if (job->rw == READ) | 
|  | 556 | r = dm_io(&io_req, 1, &job->source, NULL); | 
|  | 557 | else | 
|  | 558 | r = dm_io(&io_req, job->num_dests, job->dests, NULL); | 
|  | 559 |  | 
|  | 560 | return r; | 
|  | 561 | } | 
|  | 562 |  | 
|  | 563 | static int run_pages_job(struct kcopyd_job *job) | 
|  | 564 | { | 
|  | 565 | int r; | 
|  | 566 | unsigned nr_pages = dm_div_up(job->dests[0].count, PAGE_SIZE >> 9); | 
|  | 567 |  | 
|  | 568 | r = kcopyd_get_pages(job->kc, nr_pages, &job->pages); | 
|  | 569 | if (!r) { | 
|  | 570 | /* this job is ready for io */ | 
|  | 571 | push(&job->kc->io_jobs, job); | 
|  | 572 | return 0; | 
|  | 573 | } | 
|  | 574 |  | 
|  | 575 | if (r == -ENOMEM) | 
|  | 576 | /* can't complete now */ | 
|  | 577 | return 1; | 
|  | 578 |  | 
|  | 579 | return r; | 
|  | 580 | } | 
|  | 581 |  | 
|  | 582 | /* | 
|  | 583 | * Run through a list for as long as possible.  Returns the count | 
|  | 584 | * of successful jobs. | 
|  | 585 | */ | 
|  | 586 | static int process_jobs(struct list_head *jobs, struct dm_kcopyd_client *kc, | 
|  | 587 | int (*fn) (struct kcopyd_job *)) | 
|  | 588 | { | 
|  | 589 | struct kcopyd_job *job; | 
|  | 590 | int r, count = 0; | 
|  | 591 |  | 
|  | 592 | while ((job = pop(jobs, kc))) { | 
|  | 593 |  | 
|  | 594 | r = fn(job); | 
|  | 595 |  | 
|  | 596 | if (r < 0) { | 
|  | 597 | /* error this rogue job */ | 
|  | 598 | if (op_is_write(job->rw)) | 
|  | 599 | job->write_err = (unsigned long) -1L; | 
|  | 600 | else | 
|  | 601 | job->read_err = 1; | 
|  | 602 | push(&kc->complete_jobs, job); | 
|  | 603 | wake(kc); | 
|  | 604 | break; | 
|  | 605 | } | 
|  | 606 |  | 
|  | 607 | if (r > 0) { | 
|  | 608 | /* | 
|  | 609 | * We couldn't service this job ATM, so | 
|  | 610 | * push this job back onto the list. | 
|  | 611 | */ | 
|  | 612 | push_head(jobs, job); | 
|  | 613 | break; | 
|  | 614 | } | 
|  | 615 |  | 
|  | 616 | count++; | 
|  | 617 | } | 
|  | 618 |  | 
|  | 619 | return count; | 
|  | 620 | } | 
|  | 621 |  | 
|  | 622 | /* | 
|  | 623 | * kcopyd does this every time it's woken up. | 
|  | 624 | */ | 
|  | 625 | static void do_work(struct work_struct *work) | 
|  | 626 | { | 
|  | 627 | struct dm_kcopyd_client *kc = container_of(work, | 
|  | 628 | struct dm_kcopyd_client, kcopyd_work); | 
|  | 629 | struct blk_plug plug; | 
|  | 630 | unsigned long flags; | 
|  | 631 |  | 
|  | 632 | /* | 
|  | 633 | * The order that these are called is *very* important. | 
|  | 634 | * complete jobs can free some pages for pages jobs. | 
|  | 635 | * Pages jobs when successful will jump onto the io jobs | 
|  | 636 | * list.  io jobs call wake when they complete and it all | 
|  | 637 | * starts again. | 
|  | 638 | */ | 
|  | 639 | spin_lock_irqsave(&kc->job_lock, flags); | 
|  | 640 | list_splice_tail_init(&kc->callback_jobs, &kc->complete_jobs); | 
|  | 641 | spin_unlock_irqrestore(&kc->job_lock, flags); | 
|  | 642 |  | 
|  | 643 | blk_start_plug(&plug); | 
|  | 644 | process_jobs(&kc->complete_jobs, kc, run_complete_job); | 
|  | 645 | process_jobs(&kc->pages_jobs, kc, run_pages_job); | 
|  | 646 | process_jobs(&kc->io_jobs, kc, run_io_job); | 
|  | 647 | blk_finish_plug(&plug); | 
|  | 648 | } | 
|  | 649 |  | 
|  | 650 | /* | 
|  | 651 | * If we are copying a small region we just dispatch a single job | 
|  | 652 | * to do the copy, otherwise the io has to be split up into many | 
|  | 653 | * jobs. | 
|  | 654 | */ | 
|  | 655 | static void dispatch_job(struct kcopyd_job *job) | 
|  | 656 | { | 
|  | 657 | struct dm_kcopyd_client *kc = job->kc; | 
|  | 658 | atomic_inc(&kc->nr_jobs); | 
|  | 659 | if (unlikely(!job->source.count)) | 
|  | 660 | push(&kc->callback_jobs, job); | 
|  | 661 | else if (job->pages == &zero_page_list) | 
|  | 662 | push(&kc->io_jobs, job); | 
|  | 663 | else | 
|  | 664 | push(&kc->pages_jobs, job); | 
|  | 665 | wake(kc); | 
|  | 666 | } | 
|  | 667 |  | 
|  | 668 | static void segment_complete(int read_err, unsigned long write_err, | 
|  | 669 | void *context) | 
|  | 670 | { | 
|  | 671 | /* FIXME: tidy this function */ | 
|  | 672 | sector_t progress = 0; | 
|  | 673 | sector_t count = 0; | 
|  | 674 | struct kcopyd_job *sub_job = (struct kcopyd_job *) context; | 
|  | 675 | struct kcopyd_job *job = sub_job->master_job; | 
|  | 676 | struct dm_kcopyd_client *kc = job->kc; | 
|  | 677 |  | 
|  | 678 | mutex_lock(&job->lock); | 
|  | 679 |  | 
|  | 680 | /* update the error */ | 
|  | 681 | if (read_err) | 
|  | 682 | job->read_err = 1; | 
|  | 683 |  | 
|  | 684 | if (write_err) | 
|  | 685 | job->write_err |= write_err; | 
|  | 686 |  | 
|  | 687 | /* | 
|  | 688 | * Only dispatch more work if there hasn't been an error. | 
|  | 689 | */ | 
|  | 690 | if ((!job->read_err && !job->write_err) || | 
|  | 691 | test_bit(DM_KCOPYD_IGNORE_ERROR, &job->flags)) { | 
|  | 692 | /* get the next chunk of work */ | 
|  | 693 | progress = job->progress; | 
|  | 694 | count = job->source.count - progress; | 
|  | 695 | if (count) { | 
|  | 696 | if (count > SUB_JOB_SIZE) | 
|  | 697 | count = SUB_JOB_SIZE; | 
|  | 698 |  | 
|  | 699 | job->progress += count; | 
|  | 700 | } | 
|  | 701 | } | 
|  | 702 | mutex_unlock(&job->lock); | 
|  | 703 |  | 
|  | 704 | if (count) { | 
|  | 705 | int i; | 
|  | 706 |  | 
|  | 707 | *sub_job = *job; | 
|  | 708 | sub_job->write_offset = progress; | 
|  | 709 | sub_job->source.sector += progress; | 
|  | 710 | sub_job->source.count = count; | 
|  | 711 |  | 
|  | 712 | for (i = 0; i < job->num_dests; i++) { | 
|  | 713 | sub_job->dests[i].sector += progress; | 
|  | 714 | sub_job->dests[i].count = count; | 
|  | 715 | } | 
|  | 716 |  | 
|  | 717 | sub_job->fn = segment_complete; | 
|  | 718 | sub_job->context = sub_job; | 
|  | 719 | dispatch_job(sub_job); | 
|  | 720 |  | 
|  | 721 | } else if (atomic_dec_and_test(&job->sub_jobs)) { | 
|  | 722 |  | 
|  | 723 | /* | 
|  | 724 | * Queue the completion callback to the kcopyd thread. | 
|  | 725 | * | 
|  | 726 | * Some callers assume that all the completions are called | 
|  | 727 | * from a single thread and don't race with each other. | 
|  | 728 | * | 
|  | 729 | * We must not call the callback directly here because this | 
|  | 730 | * code may not be executing in the thread. | 
|  | 731 | */ | 
|  | 732 | push(&kc->complete_jobs, job); | 
|  | 733 | wake(kc); | 
|  | 734 | } | 
|  | 735 | } | 
|  | 736 |  | 
|  | 737 | /* | 
|  | 738 | * Create some sub jobs to share the work between them. | 
|  | 739 | */ | 
|  | 740 | static void split_job(struct kcopyd_job *master_job) | 
|  | 741 | { | 
|  | 742 | int i; | 
|  | 743 |  | 
|  | 744 | atomic_inc(&master_job->kc->nr_jobs); | 
|  | 745 |  | 
|  | 746 | atomic_set(&master_job->sub_jobs, SPLIT_COUNT); | 
|  | 747 | for (i = 0; i < SPLIT_COUNT; i++) { | 
|  | 748 | master_job[i + 1].master_job = master_job; | 
|  | 749 | segment_complete(0, 0u, &master_job[i + 1]); | 
|  | 750 | } | 
|  | 751 | } | 
|  | 752 |  | 
|  | 753 | int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from, | 
|  | 754 | unsigned int num_dests, struct dm_io_region *dests, | 
|  | 755 | unsigned int flags, dm_kcopyd_notify_fn fn, void *context) | 
|  | 756 | { | 
|  | 757 | struct kcopyd_job *job; | 
|  | 758 | int i; | 
|  | 759 |  | 
|  | 760 | /* | 
|  | 761 | * Allocate an array of jobs consisting of one master job | 
|  | 762 | * followed by SPLIT_COUNT sub jobs. | 
|  | 763 | */ | 
|  | 764 | job = mempool_alloc(kc->job_pool, GFP_NOIO); | 
|  | 765 |  | 
|  | 766 | /* | 
|  | 767 | * set up for the read. | 
|  | 768 | */ | 
|  | 769 | job->kc = kc; | 
|  | 770 | job->flags = flags; | 
|  | 771 | job->read_err = 0; | 
|  | 772 | job->write_err = 0; | 
|  | 773 |  | 
|  | 774 | job->num_dests = num_dests; | 
|  | 775 | memcpy(&job->dests, dests, sizeof(*dests) * num_dests); | 
|  | 776 |  | 
|  | 777 | /* | 
|  | 778 | * If one of the destination is a host-managed zoned block device, | 
|  | 779 | * we need to write sequentially. If one of the destination is a | 
|  | 780 | * host-aware device, then leave it to the caller to choose what to do. | 
|  | 781 | */ | 
|  | 782 | if (!test_bit(DM_KCOPYD_WRITE_SEQ, &job->flags)) { | 
|  | 783 | for (i = 0; i < job->num_dests; i++) { | 
|  | 784 | if (bdev_zoned_model(dests[i].bdev) == BLK_ZONED_HM) { | 
|  | 785 | set_bit(DM_KCOPYD_WRITE_SEQ, &job->flags); | 
|  | 786 | break; | 
|  | 787 | } | 
|  | 788 | } | 
|  | 789 | } | 
|  | 790 |  | 
|  | 791 | /* | 
|  | 792 | * If we need to write sequentially, errors cannot be ignored. | 
|  | 793 | */ | 
|  | 794 | if (test_bit(DM_KCOPYD_WRITE_SEQ, &job->flags) && | 
|  | 795 | test_bit(DM_KCOPYD_IGNORE_ERROR, &job->flags)) | 
|  | 796 | clear_bit(DM_KCOPYD_IGNORE_ERROR, &job->flags); | 
|  | 797 |  | 
|  | 798 | if (from) { | 
|  | 799 | job->source = *from; | 
|  | 800 | job->pages = NULL; | 
|  | 801 | job->rw = READ; | 
|  | 802 | } else { | 
|  | 803 | memset(&job->source, 0, sizeof job->source); | 
|  | 804 | job->source.count = job->dests[0].count; | 
|  | 805 | job->pages = &zero_page_list; | 
|  | 806 |  | 
|  | 807 | /* | 
|  | 808 | * Use WRITE ZEROES to optimize zeroing if all dests support it. | 
|  | 809 | */ | 
|  | 810 | job->rw = REQ_OP_WRITE_ZEROES; | 
|  | 811 | for (i = 0; i < job->num_dests; i++) | 
|  | 812 | if (!bdev_write_zeroes_sectors(job->dests[i].bdev)) { | 
|  | 813 | job->rw = WRITE; | 
|  | 814 | break; | 
|  | 815 | } | 
|  | 816 | } | 
|  | 817 |  | 
|  | 818 | job->fn = fn; | 
|  | 819 | job->context = context; | 
|  | 820 | job->master_job = job; | 
|  | 821 | job->write_offset = 0; | 
|  | 822 |  | 
|  | 823 | if (job->source.count <= SUB_JOB_SIZE) | 
|  | 824 | dispatch_job(job); | 
|  | 825 | else { | 
|  | 826 | mutex_init(&job->lock); | 
|  | 827 | job->progress = 0; | 
|  | 828 | split_job(job); | 
|  | 829 | } | 
|  | 830 |  | 
|  | 831 | return 0; | 
|  | 832 | } | 
|  | 833 | EXPORT_SYMBOL(dm_kcopyd_copy); | 
|  | 834 |  | 
|  | 835 | int dm_kcopyd_zero(struct dm_kcopyd_client *kc, | 
|  | 836 | unsigned num_dests, struct dm_io_region *dests, | 
|  | 837 | unsigned flags, dm_kcopyd_notify_fn fn, void *context) | 
|  | 838 | { | 
|  | 839 | return dm_kcopyd_copy(kc, NULL, num_dests, dests, flags, fn, context); | 
|  | 840 | } | 
|  | 841 | EXPORT_SYMBOL(dm_kcopyd_zero); | 
|  | 842 |  | 
|  | 843 | void *dm_kcopyd_prepare_callback(struct dm_kcopyd_client *kc, | 
|  | 844 | dm_kcopyd_notify_fn fn, void *context) | 
|  | 845 | { | 
|  | 846 | struct kcopyd_job *job; | 
|  | 847 |  | 
|  | 848 | job = mempool_alloc(kc->job_pool, GFP_NOIO); | 
|  | 849 |  | 
|  | 850 | memset(job, 0, sizeof(struct kcopyd_job)); | 
|  | 851 | job->kc = kc; | 
|  | 852 | job->fn = fn; | 
|  | 853 | job->context = context; | 
|  | 854 | job->master_job = job; | 
|  | 855 |  | 
|  | 856 | atomic_inc(&kc->nr_jobs); | 
|  | 857 |  | 
|  | 858 | return job; | 
|  | 859 | } | 
|  | 860 | EXPORT_SYMBOL(dm_kcopyd_prepare_callback); | 
|  | 861 |  | 
|  | 862 | void dm_kcopyd_do_callback(void *j, int read_err, unsigned long write_err) | 
|  | 863 | { | 
|  | 864 | struct kcopyd_job *job = j; | 
|  | 865 | struct dm_kcopyd_client *kc = job->kc; | 
|  | 866 |  | 
|  | 867 | job->read_err = read_err; | 
|  | 868 | job->write_err = write_err; | 
|  | 869 |  | 
|  | 870 | push(&kc->callback_jobs, job); | 
|  | 871 | wake(kc); | 
|  | 872 | } | 
|  | 873 | EXPORT_SYMBOL(dm_kcopyd_do_callback); | 
|  | 874 |  | 
|  | 875 | /* | 
|  | 876 | * Cancels a kcopyd job, eg. someone might be deactivating a | 
|  | 877 | * mirror. | 
|  | 878 | */ | 
|  | 879 | #if 0 | 
|  | 880 | int kcopyd_cancel(struct kcopyd_job *job, int block) | 
|  | 881 | { | 
|  | 882 | /* FIXME: finish */ | 
|  | 883 | return -1; | 
|  | 884 | } | 
|  | 885 | #endif  /*  0  */ | 
|  | 886 |  | 
|  | 887 | /*----------------------------------------------------------------- | 
|  | 888 | * Client setup | 
|  | 889 | *---------------------------------------------------------------*/ | 
|  | 890 | struct dm_kcopyd_client *dm_kcopyd_client_create(struct dm_kcopyd_throttle *throttle) | 
|  | 891 | { | 
|  | 892 | int r = -ENOMEM; | 
|  | 893 | struct dm_kcopyd_client *kc; | 
|  | 894 |  | 
|  | 895 | kc = kzalloc(sizeof(*kc), GFP_KERNEL); | 
|  | 896 | if (!kc) | 
|  | 897 | return ERR_PTR(-ENOMEM); | 
|  | 898 |  | 
|  | 899 | spin_lock_init(&kc->job_lock); | 
|  | 900 | INIT_LIST_HEAD(&kc->callback_jobs); | 
|  | 901 | INIT_LIST_HEAD(&kc->complete_jobs); | 
|  | 902 | INIT_LIST_HEAD(&kc->io_jobs); | 
|  | 903 | INIT_LIST_HEAD(&kc->pages_jobs); | 
|  | 904 | kc->throttle = throttle; | 
|  | 905 |  | 
|  | 906 | kc->job_pool = mempool_create_slab_pool(MIN_JOBS, _job_cache); | 
|  | 907 | if (!kc->job_pool) | 
|  | 908 | goto bad_slab; | 
|  | 909 |  | 
|  | 910 | INIT_WORK(&kc->kcopyd_work, do_work); | 
|  | 911 | kc->kcopyd_wq = alloc_workqueue("kcopyd", WQ_MEM_RECLAIM, 0); | 
|  | 912 | if (!kc->kcopyd_wq) | 
|  | 913 | goto bad_workqueue; | 
|  | 914 |  | 
|  | 915 | kc->pages = NULL; | 
|  | 916 | kc->nr_reserved_pages = kc->nr_free_pages = 0; | 
|  | 917 | r = client_reserve_pages(kc, RESERVE_PAGES); | 
|  | 918 | if (r) | 
|  | 919 | goto bad_client_pages; | 
|  | 920 |  | 
|  | 921 | kc->io_client = dm_io_client_create(); | 
|  | 922 | if (IS_ERR(kc->io_client)) { | 
|  | 923 | r = PTR_ERR(kc->io_client); | 
|  | 924 | goto bad_io_client; | 
|  | 925 | } | 
|  | 926 |  | 
|  | 927 | init_waitqueue_head(&kc->destroyq); | 
|  | 928 | atomic_set(&kc->nr_jobs, 0); | 
|  | 929 |  | 
|  | 930 | return kc; | 
|  | 931 |  | 
|  | 932 | bad_io_client: | 
|  | 933 | client_free_pages(kc); | 
|  | 934 | bad_client_pages: | 
|  | 935 | destroy_workqueue(kc->kcopyd_wq); | 
|  | 936 | bad_workqueue: | 
|  | 937 | mempool_destroy(kc->job_pool); | 
|  | 938 | bad_slab: | 
|  | 939 | kfree(kc); | 
|  | 940 |  | 
|  | 941 | return ERR_PTR(r); | 
|  | 942 | } | 
|  | 943 | EXPORT_SYMBOL(dm_kcopyd_client_create); | 
|  | 944 |  | 
|  | 945 | void dm_kcopyd_client_destroy(struct dm_kcopyd_client *kc) | 
|  | 946 | { | 
|  | 947 | /* Wait for completion of all jobs submitted by this client. */ | 
|  | 948 | wait_event(kc->destroyq, !atomic_read(&kc->nr_jobs)); | 
|  | 949 |  | 
|  | 950 | BUG_ON(!list_empty(&kc->callback_jobs)); | 
|  | 951 | BUG_ON(!list_empty(&kc->complete_jobs)); | 
|  | 952 | BUG_ON(!list_empty(&kc->io_jobs)); | 
|  | 953 | BUG_ON(!list_empty(&kc->pages_jobs)); | 
|  | 954 | destroy_workqueue(kc->kcopyd_wq); | 
|  | 955 | dm_io_client_destroy(kc->io_client); | 
|  | 956 | client_free_pages(kc); | 
|  | 957 | mempool_destroy(kc->job_pool); | 
|  | 958 | kfree(kc); | 
|  | 959 | } | 
|  | 960 | EXPORT_SYMBOL(dm_kcopyd_client_destroy); |