git.sesse.net Git - bcachefs-tools-debian/blob - libbcachefs/btree_write_buffer.c

   1 // SPDX-License-Identifier: GPL-2.0
   2
   3 #include "bcachefs.h"
   4 #include "btree_locking.h"
   5 #include "btree_update.h"
   6 #include "btree_update_interior.h"
   7 #include "btree_write_buffer.h"
   8 #include "error.h"
   9 #include "journal.h"
  10 #include "journal_reclaim.h"
  11
  12 #include <linux/sort.h>
  13
  14 static int bch2_btree_write_buffer_journal_flush(struct journal *,
  15                                 struct journal_entry_pin *, u64);
  16
  17 static int btree_write_buffered_key_cmp(const void *_l, const void *_r)
  18 {
  19         const struct btree_write_buffered_key *l = _l;
  20         const struct btree_write_buffered_key *r = _r;
  21
  22         return  cmp_int(l->btree, r->btree) ?:
  23                 bpos_cmp(l->k.k.p, r->k.k.p) ?:
  24                 cmp_int(l->journal_seq, r->journal_seq) ?:
  25                 cmp_int(l->journal_offset, r->journal_offset);
  26 }
  27
  28 static int btree_write_buffered_journal_cmp(const void *_l, const void *_r)
  29 {
  30         const struct btree_write_buffered_key *l = _l;
  31         const struct btree_write_buffered_key *r = _r;
  32
  33         return  cmp_int(l->journal_seq, r->journal_seq);
  34 }
  35
  36 static int bch2_btree_write_buffer_flush_one(struct btree_trans *trans,
  37                                              struct btree_iter *iter,
  38                                              struct btree_write_buffered_key *wb,
  39                                              unsigned commit_flags,
  40                                              bool *write_locked,
  41                                              size_t *fast)
  42 {
  43         struct bch_fs *c = trans->c;
  44         struct btree_path *path;
  45         int ret;
  46
  47         ret = bch2_btree_iter_traverse(iter);
  48         if (ret)
  49                 return ret;
  50
  51         /*
  52          * We can't clone a path that has write locks: unshare it now, before
  53          * set_pos and traverse():
  54          */
  55         if (iter->path->ref > 1)
  56                 iter->path = __bch2_btree_path_make_mut(trans, iter->path, true, _THIS_IP_);
  57
  58         path = iter->path;
  59
  60         if (!*write_locked) {
  61                 ret = bch2_btree_node_lock_write(trans, path, &path->l[0].b->c);
  62                 if (ret)
  63                         return ret;
  64
  65                 bch2_btree_node_prep_for_write(trans, path, path->l[0].b);
  66                 *write_locked = true;
  67         }
  68
  69         if (!bch2_btree_node_insert_fits(c, path->l[0].b, wb->k.k.u64s)) {
  70                 bch2_btree_node_unlock_write(trans, path, path->l[0].b);
  71                 *write_locked = false;
  72                 goto trans_commit;
  73         }
  74
  75         bch2_btree_insert_key_leaf(trans, path, &wb->k, wb->journal_seq);
  76         (*fast)++;
  77         return 0;
  78 trans_commit:
  79         trans->journal_res.seq = wb->journal_seq;
  80
  81         return  bch2_trans_update(trans, iter, &wb->k,
  82                                   BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?:
  83                 bch2_trans_commit(trans, NULL, NULL,
  84                                   commit_flags|
  85                                   BCH_TRANS_COMMIT_no_check_rw|
  86                                   BCH_TRANS_COMMIT_no_enospc|
  87                                   BCH_TRANS_COMMIT_no_journal_res|
  88                                   BCH_TRANS_COMMIT_journal_reclaim);
  89 }
  90
  91 static union btree_write_buffer_state btree_write_buffer_switch(struct btree_write_buffer *wb)
  92 {
  93         union btree_write_buffer_state old, new;
  94         u64 v = READ_ONCE(wb->state.v);
  95
  96         do {
  97                 old.v = new.v = v;
  98
  99                 new.nr = 0;
 100                 new.idx++;
 101         } while ((v = atomic64_cmpxchg_acquire(&wb->state.counter, old.v, new.v)) != old.v);
 102
 103         while (old.idx == 0 ? wb->state.ref0 : wb->state.ref1)
 104                 cpu_relax();
 105
 106         smp_mb();
 107
 108         return old;
 109 }
 110
 111 /*
 112  * Update a btree with a write buffered key using the journal seq of the
 113  * original write buffer insert.
 114  *
 115  * It is not safe to rejournal the key once it has been inserted into the write
 116  * buffer because that may break recovery ordering. For example, the key may
 117  * have already been modified in the active write buffer in a seq that comes
 118  * before the current transaction. If we were to journal this key again and
 119  * crash, recovery would process updates in the wrong order.
 120  */
 121 static int
 122 btree_write_buffered_insert(struct btree_trans *trans,
 123                           struct btree_write_buffered_key *wb)
 124 {
 125         struct btree_iter iter;
 126         int ret;
 127
 128         bch2_trans_iter_init(trans, &iter, wb->btree, bkey_start_pos(&wb->k.k),
 129                              BTREE_ITER_CACHED|BTREE_ITER_INTENT);
 130
 131         trans->journal_res.seq = wb->journal_seq;
 132
 133         ret   = bch2_btree_iter_traverse(&iter) ?:
 134                 bch2_trans_update(trans, &iter, &wb->k,
 135                                   BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE);
 136         bch2_trans_iter_exit(trans, &iter);
 137         return ret;
 138 }
 139
 140 int bch2_btree_write_buffer_flush_locked(struct btree_trans *trans)
 141 {
 142         struct bch_fs *c = trans->c;
 143         struct journal *j = &c->journal;
 144         struct btree_write_buffer *wb = &c->btree_write_buffer;
 145         struct journal_entry_pin pin;
 146         struct btree_write_buffered_key *i, *keys;
 147         struct btree_iter iter = { NULL };
 148         size_t nr = 0, skipped = 0, fast = 0, slowpath = 0;
 149         bool write_locked = false;
 150         union btree_write_buffer_state s;
 151         int ret = 0;
 152
 153         memset(&pin, 0, sizeof(pin));
 154
 155         bch2_journal_pin_copy(j, &pin, &wb->journal_pin,
 156                               bch2_btree_write_buffer_journal_flush);
 157         bch2_journal_pin_drop(j, &wb->journal_pin);
 158
 159         s = btree_write_buffer_switch(wb);
 160         keys = wb->keys[s.idx];
 161         nr = s.nr;
 162
 163         if (race_fault())
 164                 goto slowpath;
 165
 166         /*
 167          * We first sort so that we can detect and skip redundant updates, and
 168          * then we attempt to flush in sorted btree order, as this is most
 169          * efficient.
 170          *
 171          * However, since we're not flushing in the order they appear in the
 172          * journal we won't be able to drop our journal pin until everything is
 173          * flushed - which means this could deadlock the journal if we weren't
 174          * passing BCH_TRANS_COMMIT_journal_reclaim. This causes the update to fail
 175          * if it would block taking a journal reservation.
 176          *
 177          * If that happens, simply skip the key so we can optimistically insert
 178          * as many keys as possible in the fast path.
 179          */
 180         sort(keys, nr, sizeof(keys[0]),
 181              btree_write_buffered_key_cmp, NULL);
 182
 183         for (i = keys; i < keys + nr; i++) {
 184                 if (i + 1 < keys + nr &&
 185                     i[0].btree == i[1].btree &&
 186                     bpos_eq(i[0].k.k.p, i[1].k.k.p)) {
 187                         skipped++;
 188                         i->journal_seq = 0;
 189                         continue;
 190                 }
 191
 192                 if (write_locked &&
 193                     (iter.path->btree_id != i->btree ||
 194                      bpos_gt(i->k.k.p, iter.path->l[0].b->key.k.p))) {
 195                         bch2_btree_node_unlock_write(trans, iter.path, iter.path->l[0].b);
 196                         write_locked = false;
 197                 }
 198
 199                 if (!iter.path || iter.path->btree_id != i->btree) {
 200                         bch2_trans_iter_exit(trans, &iter);
 201                         bch2_trans_iter_init(trans, &iter, i->btree, i->k.k.p,
 202                                              BTREE_ITER_INTENT|BTREE_ITER_ALL_SNAPSHOTS);
 203                 }
 204
 205                 bch2_btree_iter_set_pos(&iter, i->k.k.p);
 206                 iter.path->preserve = false;
 207
 208                 do {
 209                         ret = bch2_btree_write_buffer_flush_one(trans, &iter, i, 0,
 210                                                                 &write_locked, &fast);
 211                         if (!write_locked)
 212                                 bch2_trans_begin(trans);
 213                 } while (bch2_err_matches(ret, BCH_ERR_transaction_restart));
 214
 215                 if (ret == -BCH_ERR_journal_reclaim_would_deadlock) {
 216                         slowpath++;
 217                         continue;
 218                 }
 219                 if (ret)
 220                         break;
 221
 222                 i->journal_seq = 0;
 223         }
 224
 225         if (write_locked)
 226                 bch2_btree_node_unlock_write(trans, iter.path, iter.path->l[0].b);
 227         bch2_trans_iter_exit(trans, &iter);
 228
 229         trace_write_buffer_flush(trans, nr, skipped, fast, wb->size);
 230
 231         if (slowpath)
 232                 goto slowpath;
 233
 234         bch2_fs_fatal_err_on(ret, c, "%s: insert error %s", __func__, bch2_err_str(ret));
 235 out:
 236         bch2_journal_pin_drop(j, &pin);
 237         return ret;
 238 slowpath:
 239         trace_and_count(c, write_buffer_flush_slowpath, trans, slowpath, nr);
 240
 241         /*
 242          * Now sort the rest by journal seq and bump the journal pin as we go.
 243          * The slowpath zapped the seq of keys that were successfully flushed so
 244          * we can skip those here.
 245          */
 246         sort(keys, nr, sizeof(keys[0]),
 247              btree_write_buffered_journal_cmp,
 248              NULL);
 249
 250         for (i = keys; i < keys + nr; i++) {
 251                 if (!i->journal_seq)
 252                         continue;
 253
 254                 bch2_journal_pin_update(j, i->journal_seq, &pin,
 255                               bch2_btree_write_buffer_journal_flush);
 256
 257                 ret = commit_do(trans, NULL, NULL,
 258                                 BCH_WATERMARK_reclaim|
 259                                 BCH_TRANS_COMMIT_no_check_rw|
 260                                 BCH_TRANS_COMMIT_no_enospc|
 261                                 BCH_TRANS_COMMIT_no_journal_res|
 262                                 BCH_TRANS_COMMIT_journal_reclaim,
 263                                 btree_write_buffered_insert(trans, i));
 264                 if (bch2_fs_fatal_err_on(ret, c, "%s: insert error %s", __func__, bch2_err_str(ret)))
 265                         break;
 266         }
 267
 268         goto out;
 269 }
 270
 271 int bch2_btree_write_buffer_flush_sync(struct btree_trans *trans)
 272 {
 273         struct bch_fs *c = trans->c;
 274
 275         if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_btree_write_buffer))
 276                 return -BCH_ERR_erofs_no_writes;
 277
 278         trace_and_count(c, write_buffer_flush_sync, trans, _RET_IP_);
 279
 280         bch2_trans_unlock(trans);
 281         mutex_lock(&c->btree_write_buffer.flush_lock);
 282         int ret = bch2_btree_write_buffer_flush_locked(trans);
 283         mutex_unlock(&c->btree_write_buffer.flush_lock);
 284         bch2_write_ref_put(c, BCH_WRITE_REF_btree_write_buffer);
 285         return ret;
 286 }
 287
 288 int bch2_btree_write_buffer_flush_nocheck_rw(struct btree_trans *trans)
 289 {
 290         struct bch_fs *c = trans->c;
 291         struct btree_write_buffer *wb = &c->btree_write_buffer;
 292         int ret = 0;
 293
 294         if (mutex_trylock(&wb->flush_lock)) {
 295                 ret = bch2_btree_write_buffer_flush_locked(trans);
 296                 mutex_unlock(&wb->flush_lock);
 297         }
 298
 299         return ret;
 300 }
 301
 302 int bch2_btree_write_buffer_tryflush(struct btree_trans *trans)
 303 {
 304         struct bch_fs *c = trans->c;
 305
 306         if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_btree_write_buffer))
 307                 return -BCH_ERR_erofs_no_writes;
 308
 309         int ret = bch2_btree_write_buffer_flush_nocheck_rw(trans);
 310         bch2_write_ref_put(c, BCH_WRITE_REF_btree_write_buffer);
 311         return ret;
 312 }
 313
 314 static int bch2_btree_write_buffer_journal_flush(struct journal *j,
 315                                 struct journal_entry_pin *_pin, u64 seq)
 316 {
 317         struct bch_fs *c = container_of(j, struct bch_fs, journal);
 318         struct btree_write_buffer *wb = &c->btree_write_buffer;
 319
 320         mutex_lock(&wb->flush_lock);
 321         int ret = bch2_trans_run(c, bch2_btree_write_buffer_flush_locked(trans));
 322         mutex_unlock(&wb->flush_lock);
 323
 324         return ret;
 325 }
 326
 327 static inline u64 btree_write_buffer_ref(int idx)
 328 {
 329         return ((union btree_write_buffer_state) {
 330                 .ref0 = idx == 0,
 331                 .ref1 = idx == 1,
 332         }).v;
 333 }
 334
 335 int bch2_btree_insert_keys_write_buffer(struct btree_trans *trans)
 336 {
 337         struct bch_fs *c = trans->c;
 338         struct btree_write_buffer *wb = &c->btree_write_buffer;
 339         struct btree_write_buffered_key *i;
 340         union btree_write_buffer_state old, new;
 341         int ret = 0;
 342         u64 v;
 343
 344         trans_for_each_wb_update(trans, i) {
 345                 EBUG_ON(i->k.k.u64s > BTREE_WRITE_BUFERED_U64s_MAX);
 346
 347                 i->journal_seq          = trans->journal_res.seq;
 348                 i->journal_offset       = trans->journal_res.offset;
 349         }
 350
 351         preempt_disable();
 352         v = READ_ONCE(wb->state.v);
 353         do {
 354                 old.v = new.v = v;
 355
 356                 new.v += btree_write_buffer_ref(new.idx);
 357                 new.nr += trans->nr_wb_updates;
 358                 if (new.nr > wb->size) {
 359                         ret = -BCH_ERR_btree_insert_need_flush_buffer;
 360                         goto out;
 361                 }
 362         } while ((v = atomic64_cmpxchg_acquire(&wb->state.counter, old.v, new.v)) != old.v);
 363
 364         memcpy(wb->keys[new.idx] + old.nr,
 365                trans->wb_updates,
 366                sizeof(trans->wb_updates[0]) * trans->nr_wb_updates);
 367
 368         bch2_journal_pin_add(&c->journal, trans->journal_res.seq, &wb->journal_pin,
 369                              bch2_btree_write_buffer_journal_flush);
 370
 371         atomic64_sub_return_release(btree_write_buffer_ref(new.idx), &wb->state.counter);
 372 out:
 373         preempt_enable();
 374         return ret;
 375 }
 376
 377 void bch2_fs_btree_write_buffer_exit(struct bch_fs *c)
 378 {
 379         struct btree_write_buffer *wb = &c->btree_write_buffer;
 380
 381         BUG_ON(wb->state.nr && !bch2_journal_error(&c->journal));
 382
 383         kvfree(wb->keys[1]);
 384         kvfree(wb->keys[0]);
 385 }
 386
 387 int bch2_fs_btree_write_buffer_init(struct bch_fs *c)
 388 {
 389         struct btree_write_buffer *wb = &c->btree_write_buffer;
 390
 391         mutex_init(&wb->flush_lock);
 392         wb->size = c->opts.btree_write_buffer_size;
 393
 394         wb->keys[0] = kvmalloc_array(wb->size, sizeof(*wb->keys[0]), GFP_KERNEL);
 395         wb->keys[1] = kvmalloc_array(wb->size, sizeof(*wb->keys[1]), GFP_KERNEL);
 396         if (!wb->keys[0] || !wb->keys[1])
 397                 return -BCH_ERR_ENOMEM_fs_btree_write_buffer_init;
 398
 399         return 0;
 400 }