drivers/md/bcache/journal.c

   1 /*
   2  * bcache journalling code, for btree insertions
   3  *
   4  * Copyright 2012 Google, Inc.
   5  */
   6
   7 #include "bcache.h"
   8 #include "btree.h"
   9 #include "debug.h"
  10 #include "extents.h"
  11
  12 #include <trace/events/bcache.h>
  13
  14 /*
  15  * Journal replay/recovery:
  16  *
  17  * This code is all driven from run_cache_set(); we first read the journal
  18  * entries, do some other stuff, then we mark all the keys in the journal
  19  * entries (same as garbage collection would), then we replay them - reinserting
  20  * them into the cache in precisely the same order as they appear in the
  21  * journal.
  22  *
  23  * We only journal keys that go in leaf nodes, which simplifies things quite a
  24  * bit.
  25  */
  26
  27 static void journal_read_endio(struct bio *bio)
  28 {
  29         struct closure *cl = bio->bi_private;
  30         closure_put(cl);
  31 }
  32
  33 static int journal_read_bucket(struct cache *ca, struct list_head *list,
  34                                unsigned bucket_index)
  35 {
  36         struct journal_device *ja = &ca->journal;
  37         struct bio *bio = &ja->bio;
  38
  39         struct journal_replay *i;
  40         struct jset *j, *data = ca->set->journal.w[0].data;
  41         struct closure cl;
  42         unsigned len, left, offset = 0;
  43         int ret = 0;
  44         sector_t bucket = bucket_to_sector(ca->set, ca->sb.d[bucket_index]);
  45
  46         closure_init_stack(&cl);
  47
  48         pr_debug("reading %u", bucket_index);
  49
  50         while (offset < ca->sb.bucket_size) {
  51 reread:         left = ca->sb.bucket_size - offset;
  52                 len = min_t(unsigned, left, PAGE_SECTORS << JSET_BITS);
  53
  54                 bio_reset(bio);
  55                 bio->bi_iter.bi_sector  = bucket + offset;
  56                 bio->bi_bdev    = ca->bdev;
  57                 bio->bi_iter.bi_size    = len << 9;
  58
  59                 bio->bi_end_io  = journal_read_endio;
  60                 bio->bi_private = &cl;
  61                 bio_set_op_attrs(bio, REQ_OP_READ, 0);
  62                 bch_bio_map(bio, data);
  63
  64                 closure_bio_submit(bio, &cl);
  65                 closure_sync(&cl);
  66
  67                 /* This function could be simpler now since we no longer write
  68                  * journal entries that overlap bucket boundaries; this means
  69                  * the start of a bucket will always have a valid journal entry
  70                  * if it has any journal entries at all.
  71                  */
  72
  73                 j = data;
  74                 while (len) {
  75                         struct list_head *where;
  76                         size_t blocks, bytes = set_bytes(j);
  77
  78                         if (j->magic != jset_magic(&ca->sb)) {
  79                                 pr_debug("%u: bad magic", bucket_index);
  80                                 return ret;
  81                         }
  82
  83                         if (bytes > left << 9 ||
  84                             bytes > PAGE_SIZE << JSET_BITS) {
  85                                 pr_info("%u: too big, %zu bytes, offset %u",
  86                                         bucket_index, bytes, offset);
  87                                 return ret;
  88                         }
  89
  90                         if (bytes > len << 9)
  91                                 goto reread;
  92
  93                         if (j->csum != csum_set(j)) {
  94                                 pr_info("%u: bad csum, %zu bytes, offset %u",
  95                                         bucket_index, bytes, offset);
  96                                 return ret;
  97                         }
  98
  99                         blocks = set_blocks(j, block_bytes(ca->set));
 100
 101                         while (!list_empty(list)) {
 102                                 i = list_first_entry(list,
 103                                         struct journal_replay, list);
 104                                 if (i->j.seq >= j->last_seq)
 105                                         break;
 106                                 list_del(&i->list);
 107                                 kfree(i);
 108                         }
 109
 110                         list_for_each_entry_reverse(i, list, list) {
 111                                 if (j->seq == i->j.seq)
 112                                         goto next_set;
 113
 114                                 if (j->seq < i->j.last_seq)
 115                                         goto next_set;
 116
 117                                 if (j->seq > i->j.seq) {
 118                                         where = &i->list;
 119                                         goto add;
 120                                 }
 121                         }
 122
 123                         where = list;
 124 add:
 125                         i = kmalloc(offsetof(struct journal_replay, j) +
 126                                     bytes, GFP_KERNEL);
 127                         if (!i)
 128                                 return -ENOMEM;
 129                         memcpy(&i->j, j, bytes);
 130                         list_add(&i->list, where);
 131                         ret = 1;
 132
 133                         ja->seq[bucket_index] = j->seq;
 134 next_set:
 135                         offset  += blocks * ca->sb.block_size;
 136                         len     -= blocks * ca->sb.block_size;
 137                         j = ((void *) j) + blocks * block_bytes(ca);
 138                 }
 139         }
 140
 141         return ret;
 142 }
 143
 144 int bch_journal_read(struct cache_set *c, struct list_head *list)
 145 {
 146 #define read_bucket(b)                                                  \
 147         ({                                                              \
 148                 int ret = journal_read_bucket(ca, list, b);             \
 149                 __set_bit(b, bitmap);                                   \
 150                 if (ret < 0)                                            \
 151                         return ret;                                     \
 152                 ret;                                                    \
 153         })
 154
 155         struct cache *ca;
 156         unsigned iter;
 157
 158         for_each_cache(ca, c, iter) {
 159                 struct journal_device *ja = &ca->journal;
 160                 DECLARE_BITMAP(bitmap, SB_JOURNAL_BUCKETS);
 161                 unsigned i, l, r, m;
 162                 uint64_t seq;
 163
 164                 bitmap_zero(bitmap, SB_JOURNAL_BUCKETS);
 165                 pr_debug("%u journal buckets", ca->sb.njournal_buckets);
 166
 167                 /*
 168                  * Read journal buckets ordered by golden ratio hash to quickly
 169                  * find a sequence of buckets with valid journal entries
 170                  */
 171                 for (i = 0; i < ca->sb.njournal_buckets; i++) {
 172                         l = (i * 2654435769U) % ca->sb.njournal_buckets;
 173
 174                         if (test_bit(l, bitmap))
 175                                 break;
 176
 177                         if (read_bucket(l))
 178                                 goto bsearch;
 179                 }
 180
 181                 /*
 182                  * If that fails, check all the buckets we haven't checked
 183                  * already
 184                  */
 185                 pr_debug("falling back to linear search");
 186
 187                 for (l = find_first_zero_bit(bitmap, ca->sb.njournal_buckets);
 188                      l < ca->sb.njournal_buckets;
 189                      l = find_next_zero_bit(bitmap, ca->sb.njournal_buckets, l + 1))
 190                         if (read_bucket(l))
 191                                 goto bsearch;
 192
 193                 /* no journal entries on this device? */
 194                 if (l == ca->sb.njournal_buckets)
 195                         continue;
 196 bsearch:
 197                 BUG_ON(list_empty(list));
 198
 199                 /* Binary search */
 200                 m = l;
 201                 r = find_next_bit(bitmap, ca->sb.njournal_buckets, l + 1);
 202                 pr_debug("starting binary search, l %u r %u", l, r);
 203
 204                 while (l + 1 < r) {
 205                         seq = list_entry(list->prev, struct journal_replay,
 206                                          list)->j.seq;
 207
 208                         m = (l + r) >> 1;
 209                         read_bucket(m);
 210
 211                         if (seq != list_entry(list->prev, struct journal_replay,
 212                                               list)->j.seq)
 213                                 l = m;
 214                         else
 215                                 r = m;
 216                 }
 217
 218                 /*
 219                  * Read buckets in reverse order until we stop finding more
 220                  * journal entries
 221                  */
 222                 pr_debug("finishing up: m %u njournal_buckets %u",
 223                          m, ca->sb.njournal_buckets);
 224                 l = m;
 225
 226                 while (1) {
 227                         if (!l--)
 228                                 l = ca->sb.njournal_buckets - 1;
 229
 230                         if (l == m)
 231                                 break;
 232
 233                         if (test_bit(l, bitmap))
 234                                 continue;
 235
 236                         if (!read_bucket(l))
 237                                 break;
 238                 }
 239
 240                 seq = 0;
 241
 242                 for (i = 0; i < ca->sb.njournal_buckets; i++)
 243                         if (ja->seq[i] > seq) {
 244                                 seq = ja->seq[i];
 245                                 /*
 246                                  * When journal_reclaim() goes to allocate for
 247                                  * the first time, it'll use the bucket after
 248                                  * ja->cur_idx
 249                                  */
 250                                 ja->cur_idx = i;
 251                                 ja->last_idx = ja->discard_idx = (i + 1) %
 252                                         ca->sb.njournal_buckets;
 253
 254                         }
 255         }
 256
 257         if (!list_empty(list))
 258                 c->journal.seq = list_entry(list->prev,
 259                                             struct journal_replay,
 260                                             list)->j.seq;
 261
 262         return 0;
 263 #undef read_bucket
 264 }
 265
 266 void bch_journal_mark(struct cache_set *c, struct list_head *list)
 267 {
 268         atomic_t p = { 0 };
 269         struct bkey *k;
 270         struct journal_replay *i;
 271         struct journal *j = &c->journal;
 272         uint64_t last = j->seq;
 273
 274         /*
 275          * journal.pin should never fill up - we never write a journal
 276          * entry when it would fill up. But if for some reason it does, we
 277          * iterate over the list in reverse order so that we can just skip that
 278          * refcount instead of bugging.
 279          */
 280
 281         list_for_each_entry_reverse(i, list, list) {
 282                 BUG_ON(last < i->j.seq);
 283                 i->pin = NULL;
 284
 285                 while (last-- != i->j.seq)
 286                         if (fifo_free(&j->pin) > 1) {
 287                                 fifo_push_front(&j->pin, p);
 288                                 atomic_set(&fifo_front(&j->pin), 0);
 289                         }
 290
 291                 if (fifo_free(&j->pin) > 1) {
 292                         fifo_push_front(&j->pin, p);
 293                         i->pin = &fifo_front(&j->pin);
 294                         atomic_set(i->pin, 1);
 295                 }
 296
 297                 for (k = i->j.start;
 298                      k < bset_bkey_last(&i->j);
 299                      k = bkey_next(k))
 300                         if (!__bch_extent_invalid(c, k)) {
 301                                 unsigned j;
 302
 303                                 for (j = 0; j < KEY_PTRS(k); j++)
 304                                         if (ptr_available(c, k, j))
 305                                                 atomic_inc(&PTR_BUCKET(c, k, j)->pin);
 306
 307                                 bch_initial_mark_key(c, 0, k);
 308                         }
 309         }
 310 }
 311
 312 int bch_journal_replay(struct cache_set *s, struct list_head *list)
 313 {
 314         int ret = 0, keys = 0, entries = 0;
 315         struct bkey *k;
 316         struct journal_replay *i =
 317                 list_entry(list->prev, struct journal_replay, list);
 318
 319         uint64_t start = i->j.last_seq, end = i->j.seq, n = start;
 320         struct keylist keylist;
 321
 322         list_for_each_entry(i, list, list) {
 323                 BUG_ON(i->pin && atomic_read(i->pin) != 1);
 324
 325                 cache_set_err_on(n != i->j.seq, s,
 326 "bcache: journal entries %llu-%llu missing! (replaying %llu-%llu)",
 327                                  n, i->j.seq - 1, start, end);
 328
 329                 for (k = i->j.start;
 330                      k < bset_bkey_last(&i->j);
 331                      k = bkey_next(k)) {
 332                         trace_bcache_journal_replay_key(k);
 333
 334                         bch_keylist_init_single(&keylist, k);
 335
 336                         ret = bch_btree_insert(s, &keylist, i->pin, NULL);
 337                         if (ret)
 338                                 goto err;
 339
 340                         BUG_ON(!bch_keylist_empty(&keylist));
 341                         keys++;
 342
 343                         cond_resched();
 344                 }
 345
 346                 if (i->pin)
 347                         atomic_dec(i->pin);
 348                 n = i->j.seq + 1;
 349                 entries++;
 350         }
 351
 352         pr_info("journal replay done, %i keys in %i entries, seq %llu",
 353                 keys, entries, end);
 354 err:
 355         while (!list_empty(list)) {
 356                 i = list_first_entry(list, struct journal_replay, list);
 357                 list_del(&i->list);
 358                 kfree(i);
 359         }
 360
 361         return ret;
 362 }
 363
 364 /* Journalling */
 365
 366 static void btree_flush_write(struct cache_set *c)
 367 {
 368         /*
 369          * Try to find the btree node with that references the oldest journal
 370          * entry, best is our current candidate and is locked if non NULL:
 371          */
 372         struct btree *b, *best;
 373         unsigned i;
 374 retry:
 375         best = NULL;
 376
 377         for_each_cached_btree(b, c, i)
 378                 if (btree_current_write(b)->journal) {
 379                         if (!best)
 380                                 best = b;
 381                         else if (journal_pin_cmp(c,
 382                                         btree_current_write(best)->journal,
 383                                         btree_current_write(b)->journal)) {
 384                                 best = b;
 385                         }
 386                 }
 387
 388         b = best;
 389         if (b) {
 390                 mutex_lock(&b->write_lock);
 391                 if (!btree_current_write(b)->journal) {
 392                         mutex_unlock(&b->write_lock);
 393                         /* We raced */
 394                         goto retry;
 395                 }
 396
 397                 __bch_btree_node_write(b, NULL);
 398                 mutex_unlock(&b->write_lock);
 399         }
 400 }
 401
 402 #define last_seq(j)     ((j)->seq - fifo_used(&(j)->pin) + 1)
 403
 404 static void journal_discard_endio(struct bio *bio)
 405 {
 406         struct journal_device *ja =
 407                 container_of(bio, struct journal_device, discard_bio);
 408         struct cache *ca = container_of(ja, struct cache, journal);
 409
 410         atomic_set(&ja->discard_in_flight, DISCARD_DONE);
 411
 412         closure_wake_up(&ca->set->journal.wait);
 413         closure_put(&ca->set->cl);
 414 }
 415
 416 static void journal_discard_work(struct work_struct *work)
 417 {
 418         struct journal_device *ja =
 419                 container_of(work, struct journal_device, discard_work);
 420
 421         submit_bio(&ja->discard_bio);
 422 }
 423
 424 static void do_journal_discard(struct cache *ca)
 425 {
 426         struct journal_device *ja = &ca->journal;
 427         struct bio *bio = &ja->discard_bio;
 428
 429         if (!ca->discard) {
 430                 ja->discard_idx = ja->last_idx;
 431                 return;
 432         }
 433
 434         switch (atomic_read(&ja->discard_in_flight)) {
 435         case DISCARD_IN_FLIGHT:
 436                 return;
 437
 438         case DISCARD_DONE:
 439                 ja->discard_idx = (ja->discard_idx + 1) %
 440                         ca->sb.njournal_buckets;
 441
 442                 atomic_set(&ja->discard_in_flight, DISCARD_READY);
 443                 /* fallthrough */
 444
 445         case DISCARD_READY:
 446                 if (ja->discard_idx == ja->last_idx)
 447                         return;
 448
 449                 atomic_set(&ja->discard_in_flight, DISCARD_IN_FLIGHT);
 450
 451                 bio_init(bio);
 452                 bio_set_op_attrs(bio, REQ_OP_DISCARD, 0);
 453                 bio->bi_iter.bi_sector  = bucket_to_sector(ca->set,
 454                                                 ca->sb.d[ja->discard_idx]);
 455                 bio->bi_bdev            = ca->bdev;
 456                 bio->bi_max_vecs        = 1;
 457                 bio->bi_io_vec          = bio->bi_inline_vecs;
 458                 bio->bi_iter.bi_size    = bucket_bytes(ca);
 459                 bio->bi_end_io          = journal_discard_endio;
 460
 461                 closure_get(&ca->set->cl);
 462                 INIT_WORK(&ja->discard_work, journal_discard_work);
 463                 schedule_work(&ja->discard_work);
 464         }
 465 }
 466
 467 static void journal_reclaim(struct cache_set *c)
 468 {
 469         struct bkey *k = &c->journal.key;
 470         struct cache *ca;
 471         uint64_t last_seq;
 472         unsigned iter, n = 0;
 473         atomic_t p;
 474
 475         while (!atomic_read(&fifo_front(&c->journal.pin)))
 476                 fifo_pop(&c->journal.pin, p);
 477
 478         last_seq = last_seq(&c->journal);
 479
 480         /* Update last_idx */
 481
 482         for_each_cache(ca, c, iter) {
 483                 struct journal_device *ja = &ca->journal;
 484
 485                 while (ja->last_idx != ja->cur_idx &&
 486                        ja->seq[ja->last_idx] < last_seq)
 487                         ja->last_idx = (ja->last_idx + 1) %
 488                                 ca->sb.njournal_buckets;
 489         }
 490
 491         for_each_cache(ca, c, iter)
 492                 do_journal_discard(ca);
 493
 494         if (c->journal.blocks_free)
 495                 goto out;
 496
 497         /*
 498          * Allocate:
 499          * XXX: Sort by free journal space
 500          */
 501
 502         for_each_cache(ca, c, iter) {
 503                 struct journal_device *ja = &ca->journal;
 504                 unsigned next = (ja->cur_idx + 1) % ca->sb.njournal_buckets;
 505
 506                 /* No space available on this device */
 507                 if (next == ja->discard_idx)
 508                         continue;
 509
 510                 ja->cur_idx = next;
 511                 k->ptr[n++] = PTR(0,
 512                                   bucket_to_sector(c, ca->sb.d[ja->cur_idx]),
 513                                   ca->sb.nr_this_dev);
 514         }
 515
 516         bkey_init(k);
 517         SET_KEY_PTRS(k, n);
 518
 519         if (n)
 520                 c->journal.blocks_free = c->sb.bucket_size >> c->block_bits;
 521 out:
 522         if (!journal_full(&c->journal))
 523                 __closure_wake_up(&c->journal.wait);
 524 }
 525
 526 void bch_journal_next(struct journal *j)
 527 {
 528         atomic_t p = { 1 };
 529
 530         j->cur = (j->cur == j->w)
 531                 ? &j->w[1]
 532                 : &j->w[0];
 533
 534         /*
 535          * The fifo_push() needs to happen at the same time as j->seq is
 536          * incremented for last_seq() to be calculated correctly
 537          */
 538         BUG_ON(!fifo_push(&j->pin, p));
 539         atomic_set(&fifo_back(&j->pin), 1);
 540
 541         j->cur->data->seq       = ++j->seq;
 542         j->cur->dirty           = false;
 543         j->cur->need_write      = false;
 544         j->cur->data->keys      = 0;
 545
 546         if (fifo_full(&j->pin))
 547                 pr_debug("journal_pin full (%zu)", fifo_used(&j->pin));
 548 }
 549
 550 static void journal_write_endio(struct bio *bio)
 551 {
 552         struct journal_write *w = bio->bi_private;
 553
 554         cache_set_err_on(bio->bi_error, w->c, "journal io error");
 555         closure_put(&w->c->journal.io);
 556 }
 557
 558 static void journal_write(struct closure *);
 559
 560 static void journal_write_done(struct closure *cl)
 561 {
 562         struct journal *j = container_of(cl, struct journal, io);
 563         struct journal_write *w = (j->cur == j->w)
 564                 ? &j->w[1]
 565                 : &j->w[0];
 566
 567         __closure_wake_up(&w->wait);
 568         continue_at_nobarrier(cl, journal_write, system_wq);
 569 }
 570
 571 static void journal_write_unlock(struct closure *cl)
 572 {
 573         struct cache_set *c = container_of(cl, struct cache_set, journal.io);
 574
 575         c->journal.io_in_flight = 0;
 576         spin_unlock(&c->journal.lock);
 577 }
 578
 579 static void journal_write_unlocked(struct closure *cl)
 580         __releases(c->journal.lock)
 581 {
 582         struct cache_set *c = container_of(cl, struct cache_set, journal.io);
 583         struct cache *ca;
 584         struct journal_write *w = c->journal.cur;
 585         struct bkey *k = &c->journal.key;
 586         unsigned i, sectors = set_blocks(w->data, block_bytes(c)) *
 587                 c->sb.block_size;
 588
 589         struct bio *bio;
 590         struct bio_list list;
 591         bio_list_init(&list);
 592
 593         if (!w->need_write) {
 594                 closure_return_with_destructor(cl, journal_write_unlock);
 595                 return;
 596         } else if (journal_full(&c->journal)) {
 597                 journal_reclaim(c);
 598                 spin_unlock(&c->journal.lock);
 599
 600                 btree_flush_write(c);
 601                 continue_at(cl, journal_write, system_wq);
 602                 return;
 603         }
 604
 605         c->journal.blocks_free -= set_blocks(w->data, block_bytes(c));
 606
 607         w->data->btree_level = c->root->level;
 608
 609         bkey_copy(&w->data->btree_root, &c->root->key);
 610         bkey_copy(&w->data->uuid_bucket, &c->uuid_bucket);
 611
 612         for_each_cache(ca, c, i)
 613                 w->data->prio_bucket[ca->sb.nr_this_dev] = ca->prio_buckets[0];
 614
 615         w->data->magic          = jset_magic(&c->sb);
 616         w->data->version        = BCACHE_JSET_VERSION;
 617         w->data->last_seq       = last_seq(&c->journal);
 618         w->data->csum           = csum_set(w->data);
 619
 620         for (i = 0; i < KEY_PTRS(k); i++) {
 621                 ca = PTR_CACHE(c, k, i);
 622                 bio = &ca->journal.bio;
 623
 624                 atomic_long_add(sectors, &ca->meta_sectors_written);
 625
 626                 bio_reset(bio);
 627                 bio->bi_iter.bi_sector  = PTR_OFFSET(k, i);
 628                 bio->bi_bdev    = ca->bdev;
 629                 bio->bi_iter.bi_size = sectors << 9;
 630
 631                 bio->bi_end_io  = journal_write_endio;
 632                 bio->bi_private = w;
 633                 bio_set_op_attrs(bio, REQ_OP_WRITE,
 634                                  REQ_SYNC|REQ_META|REQ_PREFLUSH|REQ_FUA);
 635                 bch_bio_map(bio, w->data);
 636
 637                 trace_bcache_journal_write(bio);
 638                 bio_list_add(&list, bio);
 639
 640                 SET_PTR_OFFSET(k, i, PTR_OFFSET(k, i) + sectors);
 641
 642                 ca->journal.seq[ca->journal.cur_idx] = w->data->seq;
 643         }
 644
 645         atomic_dec_bug(&fifo_back(&c->journal.pin));
 646         bch_journal_next(&c->journal);
 647         journal_reclaim(c);
 648
 649         spin_unlock(&c->journal.lock);
 650
 651         while ((bio = bio_list_pop(&list)))
 652                 closure_bio_submit(bio, cl);
 653
 654         continue_at(cl, journal_write_done, NULL);
 655 }
 656
 657 static void journal_write(struct closure *cl)
 658 {
 659         struct cache_set *c = container_of(cl, struct cache_set, journal.io);
 660
 661         spin_lock(&c->journal.lock);
 662         journal_write_unlocked(cl);
 663 }
 664
 665 static void journal_try_write(struct cache_set *c)
 666         __releases(c->journal.lock)
 667 {
 668         struct closure *cl = &c->journal.io;
 669         struct journal_write *w = c->journal.cur;
 670
 671         w->need_write = true;
 672
 673         if (!c->journal.io_in_flight) {
 674                 c->journal.io_in_flight = 1;
 675                 closure_call(cl, journal_write_unlocked, NULL, &c->cl);
 676         } else {
 677                 spin_unlock(&c->journal.lock);
 678         }
 679 }
 680
 681 static struct journal_write *journal_wait_for_write(struct cache_set *c,
 682                                                     unsigned nkeys)
 683 {
 684         size_t sectors;
 685         struct closure cl;
 686         bool wait = false;
 687
 688         closure_init_stack(&cl);
 689
 690         spin_lock(&c->journal.lock);
 691
 692         while (1) {
 693                 struct journal_write *w = c->journal.cur;
 694
 695                 sectors = __set_blocks(w->data, w->data->keys + nkeys,
 696                                        block_bytes(c)) * c->sb.block_size;
 697
 698                 if (sectors <= min_t(size_t,
 699                                      c->journal.blocks_free * c->sb.block_size,
 700                                      PAGE_SECTORS << JSET_BITS))
 701                         return w;
 702
 703                 if (wait)
 704                         closure_wait(&c->journal.wait, &cl);
 705
 706                 if (!journal_full(&c->journal)) {
 707                         if (wait)
 708                                 trace_bcache_journal_entry_full(c);
 709
 710                         /*
 711                          * XXX: If we were inserting so many keys that they
 712                          * won't fit in an _empty_ journal write, we'll
 713                          * deadlock. For now, handle this in
 714                          * bch_keylist_realloc() - but something to think about.
 715                          */
 716                         BUG_ON(!w->data->keys);
 717
 718                         journal_try_write(c); /* unlocks */
 719                 } else {
 720                         if (wait)
 721                                 trace_bcache_journal_full(c);
 722
 723                         journal_reclaim(c);
 724                         spin_unlock(&c->journal.lock);
 725
 726                         btree_flush_write(c);
 727                 }
 728
 729                 closure_sync(&cl);
 730                 spin_lock(&c->journal.lock);
 731                 wait = true;
 732         }
 733 }
 734
 735 static void journal_write_work(struct work_struct *work)
 736 {
 737         struct cache_set *c = container_of(to_delayed_work(work),
 738                                            struct cache_set,
 739                                            journal.work);
 740         spin_lock(&c->journal.lock);
 741         if (c->journal.cur->dirty)
 742                 journal_try_write(c);
 743         else
 744                 spin_unlock(&c->journal.lock);
 745 }
 746
 747 /*
 748  * Entry point to the journalling code - bio_insert() and btree_invalidate()
 749  * pass bch_journal() a list of keys to be journalled, and then
 750  * bch_journal() hands those same keys off to btree_insert_async()
 751  */
 752
 753 atomic_t *bch_journal(struct cache_set *c,
 754                       struct keylist *keys,
 755                       struct closure *parent)
 756 {
 757         struct journal_write *w;
 758         atomic_t *ret;
 759
 760         if (!CACHE_SYNC(&c->sb))
 761                 return NULL;
 762
 763         w = journal_wait_for_write(c, bch_keylist_nkeys(keys));
 764
 765         memcpy(bset_bkey_last(w->data), keys->keys, bch_keylist_bytes(keys));
 766         w->data->keys += bch_keylist_nkeys(keys);
 767
 768         ret = &fifo_back(&c->journal.pin);
 769         atomic_inc(ret);
 770
 771         if (parent) {
 772                 closure_wait(&w->wait, parent);
 773                 journal_try_write(c);
 774         } else if (!w->dirty) {
 775                 w->dirty = true;
 776                 schedule_delayed_work(&c->journal.work,
 777                                       msecs_to_jiffies(c->journal_delay_ms));
 778                 spin_unlock(&c->journal.lock);
 779         } else {
 780                 spin_unlock(&c->journal.lock);
 781         }
 782
 783
 784         return ret;
 785 }
 786
 787 void bch_journal_meta(struct cache_set *c, struct closure *cl)
 788 {
 789         struct keylist keys;
 790         atomic_t *ref;
 791
 792         bch_keylist_init(&keys);
 793
 794         ref = bch_journal(c, &keys, cl);
 795         if (ref)
 796                 atomic_dec_bug(ref);
 797 }
 798
 799 void bch_journal_free(struct cache_set *c)
 800 {
 801         free_pages((unsigned long) c->journal.w[1].data, JSET_BITS);
 802         free_pages((unsigned long) c->journal.w[0].data, JSET_BITS);
 803         free_fifo(&c->journal.pin);
 804 }
 805
 806 int bch_journal_alloc(struct cache_set *c)
 807 {
 808         struct journal *j = &c->journal;
 809
 810         spin_lock_init(&j->lock);
 811         INIT_DELAYED_WORK(&j->work, journal_write_work);
 812
 813         c->journal_delay_ms = 100;
 814
 815         j->w[0].c = c;
 816         j->w[1].c = c;
 817
 818         if (!(init_fifo(&j->pin, JOURNAL_PIN, GFP_KERNEL)) ||
 819             !(j->w[0].data = (void *) __get_free_pages(GFP_KERNEL, JSET_BITS)) ||
 820             !(j->w[1].data = (void *) __get_free_pages(GFP_KERNEL, JSET_BITS)))
 821                 return -ENOMEM;
 822
 823         return 0;
 824 }