drivers/md/dm-kcopyd.c

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