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

   1 // SPDX-License-Identifier: GPL-2.0
   2 /*
   3  * Copyright (C) 2010 Kent Overstreet <kent.overstreet@gmail.com>
   4  * Copyright (C) 2014 Datera Inc.
   5  */
   6
   7 #include "bcachefs.h"
   8 #include "alloc_background.h"
   9 #include "alloc_foreground.h"
  10 #include "bkey_methods.h"
  11 #include "bkey_buf.h"
  12 #include "btree_journal_iter.h"
  13 #include "btree_key_cache.h"
  14 #include "btree_locking.h"
  15 #include "btree_update_interior.h"
  16 #include "btree_io.h"
  17 #include "btree_gc.h"
  18 #include "buckets.h"
  19 #include "clock.h"
  20 #include "debug.h"
  21 #include "ec.h"
  22 #include "error.h"
  23 #include "extents.h"
  24 #include "journal.h"
  25 #include "keylist.h"
  26 #include "move.h"
  27 #include "recovery.h"
  28 #include "reflink.h"
  29 #include "replicas.h"
  30 #include "super-io.h"
  31 #include "trace.h"
  32
  33 #include <linux/slab.h>
  34 #include <linux/bitops.h>
  35 #include <linux/freezer.h>
  36 #include <linux/kthread.h>
  37 #include <linux/preempt.h>
  38 #include <linux/rcupdate.h>
  39 #include <linux/sched/task.h>
  40
  41 #define DROP_THIS_NODE          10
  42 #define DROP_PREV_NODE          11
  43
  44 static struct bkey_s unsafe_bkey_s_c_to_s(struct bkey_s_c k)
  45 {
  46         return (struct bkey_s) {{{
  47                 (struct bkey *) k.k,
  48                 (struct bch_val *) k.v
  49         }}};
  50 }
  51
  52 static bool should_restart_for_topology_repair(struct bch_fs *c)
  53 {
  54         return c->opts.fix_errors != FSCK_FIX_no &&
  55                 !(c->recovery_passes_complete & BIT_ULL(BCH_RECOVERY_PASS_check_topology));
  56 }
  57
  58 static inline void __gc_pos_set(struct bch_fs *c, struct gc_pos new_pos)
  59 {
  60         preempt_disable();
  61         write_seqcount_begin(&c->gc_pos_lock);
  62         c->gc_pos = new_pos;
  63         write_seqcount_end(&c->gc_pos_lock);
  64         preempt_enable();
  65 }
  66
  67 static inline void gc_pos_set(struct bch_fs *c, struct gc_pos new_pos)
  68 {
  69         BUG_ON(gc_pos_cmp(new_pos, c->gc_pos) <= 0);
  70         __gc_pos_set(c, new_pos);
  71 }
  72
  73 /*
  74  * Missing: if an interior btree node is empty, we need to do something -
  75  * perhaps just kill it
  76  */
  77 static int bch2_gc_check_topology(struct bch_fs *c,
  78                                   struct btree *b,
  79                                   struct bkey_buf *prev,
  80                                   struct bkey_buf cur,
  81                                   bool is_last)
  82 {
  83         struct bpos node_start  = b->data->min_key;
  84         struct bpos node_end    = b->data->max_key;
  85         struct bpos expected_start = bkey_deleted(&prev->k->k)
  86                 ? node_start
  87                 : bpos_successor(prev->k->k.p);
  88         struct printbuf buf1 = PRINTBUF, buf2 = PRINTBUF;
  89         int ret = 0;
  90
  91         if (cur.k->k.type == KEY_TYPE_btree_ptr_v2) {
  92                 struct bkey_i_btree_ptr_v2 *bp = bkey_i_to_btree_ptr_v2(cur.k);
  93
  94                 if (!bpos_eq(expected_start, bp->v.min_key)) {
  95                         bch2_topology_error(c);
  96
  97                         if (bkey_deleted(&prev->k->k)) {
  98                                 prt_printf(&buf1, "start of node: ");
  99                                 bch2_bpos_to_text(&buf1, node_start);
 100                         } else {
 101                                 bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(prev->k));
 102                         }
 103                         bch2_bkey_val_to_text(&buf2, c, bkey_i_to_s_c(cur.k));
 104
 105                         if (__fsck_err(c,
 106                                        FSCK_CAN_FIX|
 107                                        FSCK_CAN_IGNORE|
 108                                        FSCK_NO_RATELIMIT,
 109                                        btree_node_topology_bad_min_key,
 110                                        "btree node with incorrect min_key at btree %s level %u:\n"
 111                                        "  prev %s\n"
 112                                        "  cur %s",
 113                                        bch2_btree_id_str(b->c.btree_id), b->c.level,
 114                                        buf1.buf, buf2.buf) && should_restart_for_topology_repair(c)) {
 115                                 bch_info(c, "Halting mark and sweep to start topology repair pass");
 116                                 ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology);
 117                                 goto err;
 118                         } else {
 119                                 set_bit(BCH_FS_initial_gc_unfixed, &c->flags);
 120                         }
 121                 }
 122         }
 123
 124         if (is_last && !bpos_eq(cur.k->k.p, node_end)) {
 125                 bch2_topology_error(c);
 126
 127                 printbuf_reset(&buf1);
 128                 printbuf_reset(&buf2);
 129
 130                 bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(cur.k));
 131                 bch2_bpos_to_text(&buf2, node_end);
 132
 133                 if (__fsck_err(c, FSCK_CAN_FIX|FSCK_CAN_IGNORE|FSCK_NO_RATELIMIT,
 134                           btree_node_topology_bad_max_key,
 135                           "btree node with incorrect max_key at btree %s level %u:\n"
 136                           "  %s\n"
 137                           "  expected %s",
 138                           bch2_btree_id_str(b->c.btree_id), b->c.level,
 139                           buf1.buf, buf2.buf) &&
 140                     should_restart_for_topology_repair(c)) {
 141                         bch_info(c, "Halting mark and sweep to start topology repair pass");
 142                         ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology);
 143                         goto err;
 144                 } else {
 145                         set_bit(BCH_FS_initial_gc_unfixed, &c->flags);
 146                 }
 147         }
 148
 149         bch2_bkey_buf_copy(prev, c, cur.k);
 150 err:
 151 fsck_err:
 152         printbuf_exit(&buf2);
 153         printbuf_exit(&buf1);
 154         return ret;
 155 }
 156
 157 static void btree_ptr_to_v2(struct btree *b, struct bkey_i_btree_ptr_v2 *dst)
 158 {
 159         switch (b->key.k.type) {
 160         case KEY_TYPE_btree_ptr: {
 161                 struct bkey_i_btree_ptr *src = bkey_i_to_btree_ptr(&b->key);
 162
 163                 dst->k.p                = src->k.p;
 164                 dst->v.mem_ptr          = 0;
 165                 dst->v.seq              = b->data->keys.seq;
 166                 dst->v.sectors_written  = 0;
 167                 dst->v.flags            = 0;
 168                 dst->v.min_key          = b->data->min_key;
 169                 set_bkey_val_bytes(&dst->k, sizeof(dst->v) + bkey_val_bytes(&src->k));
 170                 memcpy(dst->v.start, src->v.start, bkey_val_bytes(&src->k));
 171                 break;
 172         }
 173         case KEY_TYPE_btree_ptr_v2:
 174                 bkey_copy(&dst->k_i, &b->key);
 175                 break;
 176         default:
 177                 BUG();
 178         }
 179 }
 180
 181 static void bch2_btree_node_update_key_early(struct btree_trans *trans,
 182                                              enum btree_id btree, unsigned level,
 183                                              struct bkey_s_c old, struct bkey_i *new)
 184 {
 185         struct bch_fs *c = trans->c;
 186         struct btree *b;
 187         struct bkey_buf tmp;
 188         int ret;
 189
 190         bch2_bkey_buf_init(&tmp);
 191         bch2_bkey_buf_reassemble(&tmp, c, old);
 192
 193         b = bch2_btree_node_get_noiter(trans, tmp.k, btree, level, true);
 194         if (!IS_ERR_OR_NULL(b)) {
 195                 mutex_lock(&c->btree_cache.lock);
 196
 197                 bch2_btree_node_hash_remove(&c->btree_cache, b);
 198
 199                 bkey_copy(&b->key, new);
 200                 ret = __bch2_btree_node_hash_insert(&c->btree_cache, b);
 201                 BUG_ON(ret);
 202
 203                 mutex_unlock(&c->btree_cache.lock);
 204                 six_unlock_read(&b->c.lock);
 205         }
 206
 207         bch2_bkey_buf_exit(&tmp, c);
 208 }
 209
 210 static int set_node_min(struct bch_fs *c, struct btree *b, struct bpos new_min)
 211 {
 212         struct bkey_i_btree_ptr_v2 *new;
 213         int ret;
 214
 215         new = kmalloc_array(BKEY_BTREE_PTR_U64s_MAX, sizeof(u64), GFP_KERNEL);
 216         if (!new)
 217                 return -BCH_ERR_ENOMEM_gc_repair_key;
 218
 219         btree_ptr_to_v2(b, new);
 220         b->data->min_key        = new_min;
 221         new->v.min_key          = new_min;
 222         SET_BTREE_PTR_RANGE_UPDATED(&new->v, true);
 223
 224         ret = bch2_journal_key_insert_take(c, b->c.btree_id, b->c.level + 1, &new->k_i);
 225         if (ret) {
 226                 kfree(new);
 227                 return ret;
 228         }
 229
 230         bch2_btree_node_drop_keys_outside_node(b);
 231         bkey_copy(&b->key, &new->k_i);
 232         return 0;
 233 }
 234
 235 static int set_node_max(struct bch_fs *c, struct btree *b, struct bpos new_max)
 236 {
 237         struct bkey_i_btree_ptr_v2 *new;
 238         int ret;
 239
 240         ret = bch2_journal_key_delete(c, b->c.btree_id, b->c.level + 1, b->key.k.p);
 241         if (ret)
 242                 return ret;
 243
 244         new = kmalloc_array(BKEY_BTREE_PTR_U64s_MAX, sizeof(u64), GFP_KERNEL);
 245         if (!new)
 246                 return -BCH_ERR_ENOMEM_gc_repair_key;
 247
 248         btree_ptr_to_v2(b, new);
 249         b->data->max_key        = new_max;
 250         new->k.p                = new_max;
 251         SET_BTREE_PTR_RANGE_UPDATED(&new->v, true);
 252
 253         ret = bch2_journal_key_insert_take(c, b->c.btree_id, b->c.level + 1, &new->k_i);
 254         if (ret) {
 255                 kfree(new);
 256                 return ret;
 257         }
 258
 259         bch2_btree_node_drop_keys_outside_node(b);
 260
 261         mutex_lock(&c->btree_cache.lock);
 262         bch2_btree_node_hash_remove(&c->btree_cache, b);
 263
 264         bkey_copy(&b->key, &new->k_i);
 265         ret = __bch2_btree_node_hash_insert(&c->btree_cache, b);
 266         BUG_ON(ret);
 267         mutex_unlock(&c->btree_cache.lock);
 268         return 0;
 269 }
 270
 271 static int btree_repair_node_boundaries(struct bch_fs *c, struct btree *b,
 272                                         struct btree *prev, struct btree *cur)
 273 {
 274         struct bpos expected_start = !prev
 275                 ? b->data->min_key
 276                 : bpos_successor(prev->key.k.p);
 277         struct printbuf buf1 = PRINTBUF, buf2 = PRINTBUF;
 278         int ret = 0;
 279
 280         if (!prev) {
 281                 prt_printf(&buf1, "start of node: ");
 282                 bch2_bpos_to_text(&buf1, b->data->min_key);
 283         } else {
 284                 bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(&prev->key));
 285         }
 286
 287         bch2_bkey_val_to_text(&buf2, c, bkey_i_to_s_c(&cur->key));
 288
 289         if (prev &&
 290             bpos_gt(expected_start, cur->data->min_key) &&
 291             BTREE_NODE_SEQ(cur->data) > BTREE_NODE_SEQ(prev->data)) {
 292                 /* cur overwrites prev: */
 293
 294                 if (mustfix_fsck_err_on(bpos_ge(prev->data->min_key,
 295                                                 cur->data->min_key), c,
 296                                 btree_node_topology_overwritten_by_next_node,
 297                                 "btree node overwritten by next node at btree %s level %u:\n"
 298                                 "  node %s\n"
 299                                 "  next %s",
 300                                 bch2_btree_id_str(b->c.btree_id), b->c.level,
 301                                 buf1.buf, buf2.buf)) {
 302                         ret = DROP_PREV_NODE;
 303                         goto out;
 304                 }
 305
 306                 if (mustfix_fsck_err_on(!bpos_eq(prev->key.k.p,
 307                                                  bpos_predecessor(cur->data->min_key)), c,
 308                                 btree_node_topology_bad_max_key,
 309                                 "btree node with incorrect max_key at btree %s level %u:\n"
 310                                 "  node %s\n"
 311                                 "  next %s",
 312                                 bch2_btree_id_str(b->c.btree_id), b->c.level,
 313                                 buf1.buf, buf2.buf))
 314                         ret = set_node_max(c, prev,
 315                                            bpos_predecessor(cur->data->min_key));
 316         } else {
 317                 /* prev overwrites cur: */
 318
 319                 if (mustfix_fsck_err_on(bpos_ge(expected_start,
 320                                                 cur->data->max_key), c,
 321                                 btree_node_topology_overwritten_by_prev_node,
 322                                 "btree node overwritten by prev node at btree %s level %u:\n"
 323                                 "  prev %s\n"
 324                                 "  node %s",
 325                                 bch2_btree_id_str(b->c.btree_id), b->c.level,
 326                                 buf1.buf, buf2.buf)) {
 327                         ret = DROP_THIS_NODE;
 328                         goto out;
 329                 }
 330
 331                 if (mustfix_fsck_err_on(!bpos_eq(expected_start, cur->data->min_key), c,
 332                                 btree_node_topology_bad_min_key,
 333                                 "btree node with incorrect min_key at btree %s level %u:\n"
 334                                 "  prev %s\n"
 335                                 "  node %s",
 336                                 bch2_btree_id_str(b->c.btree_id), b->c.level,
 337                                 buf1.buf, buf2.buf))
 338                         ret = set_node_min(c, cur, expected_start);
 339         }
 340 out:
 341 fsck_err:
 342         printbuf_exit(&buf2);
 343         printbuf_exit(&buf1);
 344         return ret;
 345 }
 346
 347 static int btree_repair_node_end(struct bch_fs *c, struct btree *b,
 348                                  struct btree *child)
 349 {
 350         struct printbuf buf1 = PRINTBUF, buf2 = PRINTBUF;
 351         int ret = 0;
 352
 353         bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(&child->key));
 354         bch2_bpos_to_text(&buf2, b->key.k.p);
 355
 356         if (mustfix_fsck_err_on(!bpos_eq(child->key.k.p, b->key.k.p), c,
 357                                 btree_node_topology_bad_max_key,
 358                         "btree node with incorrect max_key at btree %s level %u:\n"
 359                         "  %s\n"
 360                         "  expected %s",
 361                         bch2_btree_id_str(b->c.btree_id), b->c.level,
 362                         buf1.buf, buf2.buf)) {
 363                 ret = set_node_max(c, child, b->key.k.p);
 364                 if (ret)
 365                         goto err;
 366         }
 367 err:
 368 fsck_err:
 369         printbuf_exit(&buf2);
 370         printbuf_exit(&buf1);
 371         return ret;
 372 }
 373
 374 static int bch2_btree_repair_topology_recurse(struct btree_trans *trans, struct btree *b)
 375 {
 376         struct bch_fs *c = trans->c;
 377         struct btree_and_journal_iter iter;
 378         struct bkey_s_c k;
 379         struct bkey_buf prev_k, cur_k;
 380         struct btree *prev = NULL, *cur = NULL;
 381         bool have_child, dropped_children = false;
 382         struct printbuf buf = PRINTBUF;
 383         int ret = 0;
 384
 385         if (!b->c.level)
 386                 return 0;
 387 again:
 388         prev = NULL;
 389         have_child = dropped_children = false;
 390         bch2_bkey_buf_init(&prev_k);
 391         bch2_bkey_buf_init(&cur_k);
 392         bch2_btree_and_journal_iter_init_node_iter(&iter, c, b);
 393
 394         while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 395                 BUG_ON(bpos_lt(k.k->p, b->data->min_key));
 396                 BUG_ON(bpos_gt(k.k->p, b->data->max_key));
 397
 398                 bch2_btree_and_journal_iter_advance(&iter);
 399                 bch2_bkey_buf_reassemble(&cur_k, c, k);
 400
 401                 cur = bch2_btree_node_get_noiter(trans, cur_k.k,
 402                                         b->c.btree_id, b->c.level - 1,
 403                                         false);
 404                 ret = PTR_ERR_OR_ZERO(cur);
 405
 406                 printbuf_reset(&buf);
 407                 bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(cur_k.k));
 408
 409                 if (mustfix_fsck_err_on(ret == -EIO, c,
 410                                 btree_node_unreadable,
 411                                 "Topology repair: unreadable btree node at btree %s level %u:\n"
 412                                 "  %s",
 413                                 bch2_btree_id_str(b->c.btree_id),
 414                                 b->c.level - 1,
 415                                 buf.buf)) {
 416                         bch2_btree_node_evict(trans, cur_k.k);
 417                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 418                                                       b->c.level, cur_k.k->k.p);
 419                         cur = NULL;
 420                         if (ret)
 421                                 break;
 422                         continue;
 423                 }
 424
 425                 bch_err_msg(c, ret, "getting btree node");
 426                 if (ret)
 427                         break;
 428
 429                 ret = btree_repair_node_boundaries(c, b, prev, cur);
 430
 431                 if (ret == DROP_THIS_NODE) {
 432                         six_unlock_read(&cur->c.lock);
 433                         bch2_btree_node_evict(trans, cur_k.k);
 434                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 435                                                       b->c.level, cur_k.k->k.p);
 436                         cur = NULL;
 437                         if (ret)
 438                                 break;
 439                         continue;
 440                 }
 441
 442                 if (prev)
 443                         six_unlock_read(&prev->c.lock);
 444                 prev = NULL;
 445
 446                 if (ret == DROP_PREV_NODE) {
 447                         bch2_btree_node_evict(trans, prev_k.k);
 448                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 449                                                       b->c.level, prev_k.k->k.p);
 450                         if (ret)
 451                                 break;
 452
 453                         bch2_btree_and_journal_iter_exit(&iter);
 454                         bch2_bkey_buf_exit(&prev_k, c);
 455                         bch2_bkey_buf_exit(&cur_k, c);
 456                         goto again;
 457                 } else if (ret)
 458                         break;
 459
 460                 prev = cur;
 461                 cur = NULL;
 462                 bch2_bkey_buf_copy(&prev_k, c, cur_k.k);
 463         }
 464
 465         if (!ret && !IS_ERR_OR_NULL(prev)) {
 466                 BUG_ON(cur);
 467                 ret = btree_repair_node_end(c, b, prev);
 468         }
 469
 470         if (!IS_ERR_OR_NULL(prev))
 471                 six_unlock_read(&prev->c.lock);
 472         prev = NULL;
 473         if (!IS_ERR_OR_NULL(cur))
 474                 six_unlock_read(&cur->c.lock);
 475         cur = NULL;
 476
 477         if (ret)
 478                 goto err;
 479
 480         bch2_btree_and_journal_iter_exit(&iter);
 481         bch2_btree_and_journal_iter_init_node_iter(&iter, c, b);
 482
 483         while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 484                 bch2_bkey_buf_reassemble(&cur_k, c, k);
 485                 bch2_btree_and_journal_iter_advance(&iter);
 486
 487                 cur = bch2_btree_node_get_noiter(trans, cur_k.k,
 488                                         b->c.btree_id, b->c.level - 1,
 489                                         false);
 490                 ret = PTR_ERR_OR_ZERO(cur);
 491
 492                 bch_err_msg(c, ret, "getting btree node");
 493                 if (ret)
 494                         goto err;
 495
 496                 ret = bch2_btree_repair_topology_recurse(trans, cur);
 497                 six_unlock_read(&cur->c.lock);
 498                 cur = NULL;
 499
 500                 if (ret == DROP_THIS_NODE) {
 501                         bch2_btree_node_evict(trans, cur_k.k);
 502                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 503                                                       b->c.level, cur_k.k->k.p);
 504                         dropped_children = true;
 505                 }
 506
 507                 if (ret)
 508                         goto err;
 509
 510                 have_child = true;
 511         }
 512
 513         printbuf_reset(&buf);
 514         bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key));
 515
 516         if (mustfix_fsck_err_on(!have_child, c,
 517                         btree_node_topology_interior_node_empty,
 518                         "empty interior btree node at btree %s level %u\n"
 519                         "  %s",
 520                         bch2_btree_id_str(b->c.btree_id),
 521                         b->c.level, buf.buf))
 522                 ret = DROP_THIS_NODE;
 523 err:
 524 fsck_err:
 525         if (!IS_ERR_OR_NULL(prev))
 526                 six_unlock_read(&prev->c.lock);
 527         if (!IS_ERR_OR_NULL(cur))
 528                 six_unlock_read(&cur->c.lock);
 529
 530         bch2_btree_and_journal_iter_exit(&iter);
 531         bch2_bkey_buf_exit(&prev_k, c);
 532         bch2_bkey_buf_exit(&cur_k, c);
 533
 534         if (!ret && dropped_children)
 535                 goto again;
 536
 537         printbuf_exit(&buf);
 538         return ret;
 539 }
 540
 541 int bch2_check_topology(struct bch_fs *c)
 542 {
 543         struct btree_trans *trans = bch2_trans_get(c);
 544         struct btree *b;
 545         unsigned i;
 546         int ret = 0;
 547
 548         for (i = 0; i < btree_id_nr_alive(c) && !ret; i++) {
 549                 struct btree_root *r = bch2_btree_id_root(c, i);
 550
 551                 if (!r->alive)
 552                         continue;
 553
 554                 b = r->b;
 555                 if (btree_node_fake(b))
 556                         continue;
 557
 558                 btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_read);
 559                 ret = bch2_btree_repair_topology_recurse(trans, b);
 560                 six_unlock_read(&b->c.lock);
 561
 562                 if (ret == DROP_THIS_NODE) {
 563                         bch_err(c, "empty btree root - repair unimplemented");
 564                         ret = -BCH_ERR_fsck_repair_unimplemented;
 565                 }
 566         }
 567
 568         bch2_trans_put(trans);
 569
 570         return ret;
 571 }
 572
 573 static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id,
 574                                unsigned level, bool is_root,
 575                                struct bkey_s_c *k)
 576 {
 577         struct bch_fs *c = trans->c;
 578         struct bkey_ptrs_c ptrs_c = bch2_bkey_ptrs_c(*k);
 579         const union bch_extent_entry *entry_c;
 580         struct extent_ptr_decoded p = { 0 };
 581         bool do_update = false;
 582         struct printbuf buf = PRINTBUF;
 583         int ret = 0;
 584
 585         /*
 586          * XXX
 587          * use check_bucket_ref here
 588          */
 589         bkey_for_each_ptr_decode(k->k, ptrs_c, p, entry_c) {
 590                 struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev);
 591                 struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr);
 592                 enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, &entry_c->ptr);
 593
 594                 if (!g->gen_valid &&
 595                     (c->opts.reconstruct_alloc ||
 596                      fsck_err(c, ptr_to_missing_alloc_key,
 597                               "bucket %u:%zu data type %s ptr gen %u missing in alloc btree\n"
 598                               "while marking %s",
 599                               p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
 600                               bch2_data_types[ptr_data_type(k->k, &p.ptr)],
 601                               p.ptr.gen,
 602                               (printbuf_reset(&buf),
 603                                bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))) {
 604                         if (!p.ptr.cached) {
 605                                 g->gen_valid            = true;
 606                                 g->gen                  = p.ptr.gen;
 607                         } else {
 608                                 do_update = true;
 609                         }
 610                 }
 611
 612                 if (gen_cmp(p.ptr.gen, g->gen) > 0 &&
 613                     (c->opts.reconstruct_alloc ||
 614                      fsck_err(c, ptr_gen_newer_than_bucket_gen,
 615                               "bucket %u:%zu data type %s ptr gen in the future: %u > %u\n"
 616                               "while marking %s",
 617                               p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
 618                               bch2_data_types[ptr_data_type(k->k, &p.ptr)],
 619                               p.ptr.gen, g->gen,
 620                               (printbuf_reset(&buf),
 621                                bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))) {
 622                         if (!p.ptr.cached) {
 623                                 g->gen_valid            = true;
 624                                 g->gen                  = p.ptr.gen;
 625                                 g->data_type            = 0;
 626                                 g->dirty_sectors        = 0;
 627                                 g->cached_sectors       = 0;
 628                                 set_bit(BCH_FS_need_another_gc, &c->flags);
 629                         } else {
 630                                 do_update = true;
 631                         }
 632                 }
 633
 634                 if (gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX &&
 635                     (c->opts.reconstruct_alloc ||
 636                      fsck_err(c, ptr_gen_newer_than_bucket_gen,
 637                               "bucket %u:%zu gen %u data type %s: ptr gen %u too stale\n"
 638                               "while marking %s",
 639                               p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen,
 640                               bch2_data_types[ptr_data_type(k->k, &p.ptr)],
 641                               p.ptr.gen,
 642                               (printbuf_reset(&buf),
 643                                bch2_bkey_val_to_text(&buf, c, *k), buf.buf))))
 644                         do_update = true;
 645
 646                 if (!p.ptr.cached && gen_cmp(p.ptr.gen, g->gen) < 0 &&
 647                     (c->opts.reconstruct_alloc ||
 648                      fsck_err(c, stale_dirty_ptr,
 649                               "bucket %u:%zu data type %s stale dirty ptr: %u < %u\n"
 650                               "while marking %s",
 651                               p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
 652                               bch2_data_types[ptr_data_type(k->k, &p.ptr)],
 653                               p.ptr.gen, g->gen,
 654                               (printbuf_reset(&buf),
 655                                bch2_bkey_val_to_text(&buf, c, *k), buf.buf))))
 656                         do_update = true;
 657
 658                 if (data_type != BCH_DATA_btree && p.ptr.gen != g->gen)
 659                         continue;
 660
 661                 if (fsck_err_on(bucket_data_type(g->data_type) &&
 662                                 bucket_data_type(g->data_type) != data_type, c,
 663                                 ptr_bucket_data_type_mismatch,
 664                                 "bucket %u:%zu different types of data in same bucket: %s, %s\n"
 665                                 "while marking %s",
 666                                 p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
 667                                 bch2_data_types[g->data_type],
 668                                 bch2_data_types[data_type],
 669                                 (printbuf_reset(&buf),
 670                                  bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) {
 671                         if (data_type == BCH_DATA_btree) {
 672                                 g->data_type    = data_type;
 673                                 set_bit(BCH_FS_need_another_gc, &c->flags);
 674                         } else {
 675                                 do_update = true;
 676                         }
 677                 }
 678
 679                 if (p.has_ec) {
 680                         struct gc_stripe *m = genradix_ptr(&c->gc_stripes, p.ec.idx);
 681
 682                         if (fsck_err_on(!m || !m->alive, c,
 683                                         ptr_to_missing_stripe,
 684                                         "pointer to nonexistent stripe %llu\n"
 685                                         "while marking %s",
 686                                         (u64) p.ec.idx,
 687                                         (printbuf_reset(&buf),
 688                                          bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))
 689                                 do_update = true;
 690
 691                         if (fsck_err_on(m && m->alive && !bch2_ptr_matches_stripe_m(m, p), c,
 692                                         ptr_to_incorrect_stripe,
 693                                         "pointer does not match stripe %llu\n"
 694                                         "while marking %s",
 695                                         (u64) p.ec.idx,
 696                                         (printbuf_reset(&buf),
 697                                          bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))
 698                                 do_update = true;
 699                 }
 700         }
 701
 702         if (do_update) {
 703                 struct bkey_ptrs ptrs;
 704                 union bch_extent_entry *entry;
 705                 struct bch_extent_ptr *ptr;
 706                 struct bkey_i *new;
 707
 708                 if (is_root) {
 709                         bch_err(c, "cannot update btree roots yet");
 710                         ret = -EINVAL;
 711                         goto err;
 712                 }
 713
 714                 new = kmalloc(bkey_bytes(k->k), GFP_KERNEL);
 715                 if (!new) {
 716                         ret = -BCH_ERR_ENOMEM_gc_repair_key;
 717                         bch_err_msg(c, ret, "allocating new key");
 718                         goto err;
 719                 }
 720
 721                 bkey_reassemble(new, *k);
 722
 723                 if (level) {
 724                         /*
 725                          * We don't want to drop btree node pointers - if the
 726                          * btree node isn't there anymore, the read path will
 727                          * sort it out:
 728                          */
 729                         ptrs = bch2_bkey_ptrs(bkey_i_to_s(new));
 730                         bkey_for_each_ptr(ptrs, ptr) {
 731                                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
 732                                 struct bucket *g = PTR_GC_BUCKET(ca, ptr);
 733
 734                                 ptr->gen = g->gen;
 735                         }
 736                 } else {
 737                         bch2_bkey_drop_ptrs(bkey_i_to_s(new), ptr, ({
 738                                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
 739                                 struct bucket *g = PTR_GC_BUCKET(ca, ptr);
 740                                 enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, ptr);
 741
 742                                 (ptr->cached &&
 743                                  (!g->gen_valid || gen_cmp(ptr->gen, g->gen) > 0)) ||
 744                                 (!ptr->cached &&
 745                                  gen_cmp(ptr->gen, g->gen) < 0) ||
 746                                 gen_cmp(g->gen, ptr->gen) > BUCKET_GC_GEN_MAX ||
 747                                 (g->data_type &&
 748                                  g->data_type != data_type);
 749                         }));
 750 again:
 751                         ptrs = bch2_bkey_ptrs(bkey_i_to_s(new));
 752                         bkey_extent_entry_for_each(ptrs, entry) {
 753                                 if (extent_entry_type(entry) == BCH_EXTENT_ENTRY_stripe_ptr) {
 754                                         struct gc_stripe *m = genradix_ptr(&c->gc_stripes,
 755                                                                         entry->stripe_ptr.idx);
 756                                         union bch_extent_entry *next_ptr;
 757
 758                                         bkey_extent_entry_for_each_from(ptrs, next_ptr, entry)
 759                                                 if (extent_entry_type(next_ptr) == BCH_EXTENT_ENTRY_ptr)
 760                                                         goto found;
 761                                         next_ptr = NULL;
 762 found:
 763                                         if (!next_ptr) {
 764                                                 bch_err(c, "aieee, found stripe ptr with no data ptr");
 765                                                 continue;
 766                                         }
 767
 768                                         if (!m || !m->alive ||
 769                                             !__bch2_ptr_matches_stripe(&m->ptrs[entry->stripe_ptr.block],
 770                                                                        &next_ptr->ptr,
 771                                                                        m->sectors)) {
 772                                                 bch2_bkey_extent_entry_drop(new, entry);
 773                                                 goto again;
 774                                         }
 775                                 }
 776                         }
 777                 }
 778
 779                 ret = bch2_journal_key_insert_take(c, btree_id, level, new);
 780                 if (ret) {
 781                         kfree(new);
 782                         goto err;
 783                 }
 784
 785                 if (level)
 786                         bch2_btree_node_update_key_early(trans, btree_id, level - 1, *k, new);
 787
 788                 if (0) {
 789                         printbuf_reset(&buf);
 790                         bch2_bkey_val_to_text(&buf, c, *k);
 791                         bch_info(c, "updated %s", buf.buf);
 792
 793                         printbuf_reset(&buf);
 794                         bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(new));
 795                         bch_info(c, "new key %s", buf.buf);
 796                 }
 797
 798                 *k = bkey_i_to_s_c(new);
 799         }
 800 err:
 801 fsck_err:
 802         printbuf_exit(&buf);
 803         return ret;
 804 }
 805
 806 /* marking of btree keys/nodes: */
 807
 808 static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id,
 809                             unsigned level, bool is_root,
 810                             struct bkey_s_c *k,
 811                             bool initial)
 812 {
 813         struct bch_fs *c = trans->c;
 814         struct bkey deleted = KEY(0, 0, 0);
 815         struct bkey_s_c old = (struct bkey_s_c) { &deleted, NULL };
 816         int ret = 0;
 817
 818         deleted.p = k->k->p;
 819
 820         if (initial) {
 821                 BUG_ON(bch2_journal_seq_verify &&
 822                        k->k->version.lo > atomic64_read(&c->journal.seq));
 823
 824                 ret = bch2_check_fix_ptrs(trans, btree_id, level, is_root, k);
 825                 if (ret)
 826                         goto err;
 827
 828                 if (fsck_err_on(k->k->version.lo > atomic64_read(&c->key_version), c,
 829                                 bkey_version_in_future,
 830                                 "key version number higher than recorded: %llu > %llu",
 831                                 k->k->version.lo,
 832                                 atomic64_read(&c->key_version)))
 833                         atomic64_set(&c->key_version, k->k->version.lo);
 834         }
 835
 836         ret = commit_do(trans, NULL, NULL, 0,
 837                         bch2_key_trigger(trans, btree_id, level, old, unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_GC));
 838 fsck_err:
 839 err:
 840         bch_err_fn(c, ret);
 841         return ret;
 842 }
 843
 844 static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool initial)
 845 {
 846         struct bch_fs *c = trans->c;
 847         struct btree_node_iter iter;
 848         struct bkey unpacked;
 849         struct bkey_s_c k;
 850         struct bkey_buf prev, cur;
 851         int ret = 0;
 852
 853         if (!btree_node_type_needs_gc(btree_node_type(b)))
 854                 return 0;
 855
 856         bch2_btree_node_iter_init_from_start(&iter, b);
 857         bch2_bkey_buf_init(&prev);
 858         bch2_bkey_buf_init(&cur);
 859         bkey_init(&prev.k->k);
 860
 861         while ((k = bch2_btree_node_iter_peek_unpack(&iter, b, &unpacked)).k) {
 862                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false,
 863                                        &k, initial);
 864                 if (ret)
 865                         break;
 866
 867                 bch2_btree_node_iter_advance(&iter, b);
 868
 869                 if (b->c.level) {
 870                         bch2_bkey_buf_reassemble(&cur, c, k);
 871
 872                         ret = bch2_gc_check_topology(c, b, &prev, cur,
 873                                         bch2_btree_node_iter_end(&iter));
 874                         if (ret)
 875                                 break;
 876                 }
 877         }
 878
 879         bch2_bkey_buf_exit(&cur, c);
 880         bch2_bkey_buf_exit(&prev, c);
 881         return ret;
 882 }
 883
 884 static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id,
 885                          bool initial, bool metadata_only)
 886 {
 887         struct bch_fs *c = trans->c;
 888         struct btree_iter iter;
 889         struct btree *b;
 890         unsigned depth = metadata_only ? 1 : 0;
 891         int ret = 0;
 892
 893         gc_pos_set(c, gc_pos_btree(btree_id, POS_MIN, 0));
 894
 895         __for_each_btree_node(trans, iter, btree_id, POS_MIN,
 896                               0, depth, BTREE_ITER_PREFETCH, b, ret) {
 897                 bch2_verify_btree_nr_keys(b);
 898
 899                 gc_pos_set(c, gc_pos_btree_node(b));
 900
 901                 ret = btree_gc_mark_node(trans, b, initial);
 902                 if (ret)
 903                         break;
 904         }
 905         bch2_trans_iter_exit(trans, &iter);
 906
 907         if (ret)
 908                 return ret;
 909
 910         mutex_lock(&c->btree_root_lock);
 911         b = bch2_btree_id_root(c, btree_id)->b;
 912         if (!btree_node_fake(b)) {
 913                 struct bkey_s_c k = bkey_i_to_s_c(&b->key);
 914
 915                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1,
 916                                        true, &k, initial);
 917         }
 918         gc_pos_set(c, gc_pos_btree_root(b->c.btree_id));
 919         mutex_unlock(&c->btree_root_lock);
 920
 921         return ret;
 922 }
 923
 924 static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b,
 925                                       unsigned target_depth)
 926 {
 927         struct bch_fs *c = trans->c;
 928         struct btree_and_journal_iter iter;
 929         struct bkey_s_c k;
 930         struct bkey_buf cur, prev;
 931         struct printbuf buf = PRINTBUF;
 932         int ret = 0;
 933
 934         bch2_btree_and_journal_iter_init_node_iter(&iter, c, b);
 935         bch2_bkey_buf_init(&prev);
 936         bch2_bkey_buf_init(&cur);
 937         bkey_init(&prev.k->k);
 938
 939         while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 940                 BUG_ON(bpos_lt(k.k->p, b->data->min_key));
 941                 BUG_ON(bpos_gt(k.k->p, b->data->max_key));
 942
 943                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level,
 944                                        false, &k, true);
 945                 if (ret)
 946                         goto fsck_err;
 947
 948                 if (b->c.level) {
 949                         bch2_bkey_buf_reassemble(&cur, c, k);
 950                         k = bkey_i_to_s_c(cur.k);
 951
 952                         bch2_btree_and_journal_iter_advance(&iter);
 953
 954                         ret = bch2_gc_check_topology(c, b,
 955                                         &prev, cur,
 956                                         !bch2_btree_and_journal_iter_peek(&iter).k);
 957                         if (ret)
 958                                 goto fsck_err;
 959                 } else {
 960                         bch2_btree_and_journal_iter_advance(&iter);
 961                 }
 962         }
 963
 964         if (b->c.level > target_depth) {
 965                 bch2_btree_and_journal_iter_exit(&iter);
 966                 bch2_btree_and_journal_iter_init_node_iter(&iter, c, b);
 967
 968                 while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 969                         struct btree *child;
 970
 971                         bch2_bkey_buf_reassemble(&cur, c, k);
 972                         bch2_btree_and_journal_iter_advance(&iter);
 973
 974                         child = bch2_btree_node_get_noiter(trans, cur.k,
 975                                                 b->c.btree_id, b->c.level - 1,
 976                                                 false);
 977                         ret = PTR_ERR_OR_ZERO(child);
 978
 979                         if (ret == -EIO) {
 980                                 bch2_topology_error(c);
 981
 982                                 if (__fsck_err(c,
 983                                           FSCK_CAN_FIX|
 984                                           FSCK_CAN_IGNORE|
 985                                           FSCK_NO_RATELIMIT,
 986                                           btree_node_read_error,
 987                                           "Unreadable btree node at btree %s level %u:\n"
 988                                           "  %s",
 989                                           bch2_btree_id_str(b->c.btree_id),
 990                                           b->c.level - 1,
 991                                           (printbuf_reset(&buf),
 992                                            bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(cur.k)), buf.buf)) &&
 993                                     should_restart_for_topology_repair(c)) {
 994                                         bch_info(c, "Halting mark and sweep to start topology repair pass");
 995                                         ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology);
 996                                         goto fsck_err;
 997                                 } else {
 998                                         /* Continue marking when opted to not
 999                                          * fix the error: */
1000                                         ret = 0;
1001                                         set_bit(BCH_FS_initial_gc_unfixed, &c->flags);
1002                                         continue;
1003                                 }
1004                         } else if (ret) {
1005                                 bch_err_msg(c, ret, "getting btree node");
1006                                 break;
1007                         }
1008
1009                         ret = bch2_gc_btree_init_recurse(trans, child,
1010                                                          target_depth);
1011                         six_unlock_read(&child->c.lock);
1012
1013                         if (ret)
1014                                 break;
1015                 }
1016         }
1017 fsck_err:
1018         bch2_bkey_buf_exit(&cur, c);
1019         bch2_bkey_buf_exit(&prev, c);
1020         bch2_btree_and_journal_iter_exit(&iter);
1021         printbuf_exit(&buf);
1022         return ret;
1023 }
1024
1025 static int bch2_gc_btree_init(struct btree_trans *trans,
1026                               enum btree_id btree_id,
1027                               bool metadata_only)
1028 {
1029         struct bch_fs *c = trans->c;
1030         struct btree *b;
1031         unsigned target_depth = metadata_only ? 1 : 0;
1032         struct printbuf buf = PRINTBUF;
1033         int ret = 0;
1034
1035         b = bch2_btree_id_root(c, btree_id)->b;
1036
1037         if (btree_node_fake(b))
1038                 return 0;
1039
1040         six_lock_read(&b->c.lock, NULL, NULL);
1041         printbuf_reset(&buf);
1042         bch2_bpos_to_text(&buf, b->data->min_key);
1043         if (mustfix_fsck_err_on(!bpos_eq(b->data->min_key, POS_MIN), c,
1044                                 btree_root_bad_min_key,
1045                         "btree root with incorrect min_key: %s", buf.buf)) {
1046                 bch_err(c, "repair unimplemented");
1047                 ret = -BCH_ERR_fsck_repair_unimplemented;
1048                 goto fsck_err;
1049         }
1050
1051         printbuf_reset(&buf);
1052         bch2_bpos_to_text(&buf, b->data->max_key);
1053         if (mustfix_fsck_err_on(!bpos_eq(b->data->max_key, SPOS_MAX), c,
1054                                 btree_root_bad_max_key,
1055                         "btree root with incorrect max_key: %s", buf.buf)) {
1056                 bch_err(c, "repair unimplemented");
1057                 ret = -BCH_ERR_fsck_repair_unimplemented;
1058                 goto fsck_err;
1059         }
1060
1061         if (b->c.level >= target_depth)
1062                 ret = bch2_gc_btree_init_recurse(trans, b, target_depth);
1063
1064         if (!ret) {
1065                 struct bkey_s_c k = bkey_i_to_s_c(&b->key);
1066
1067                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, true,
1068                                        &k, true);
1069         }
1070 fsck_err:
1071         six_unlock_read(&b->c.lock);
1072
1073         bch_err_fn(c, ret);
1074         printbuf_exit(&buf);
1075         return ret;
1076 }
1077
1078 static inline int btree_id_gc_phase_cmp(enum btree_id l, enum btree_id r)
1079 {
1080         return  (int) btree_id_to_gc_phase(l) -
1081                 (int) btree_id_to_gc_phase(r);
1082 }
1083
1084 static int bch2_gc_btrees(struct bch_fs *c, bool initial, bool metadata_only)
1085 {
1086         struct btree_trans *trans = bch2_trans_get(c);
1087         enum btree_id ids[BTREE_ID_NR];
1088         unsigned i;
1089         int ret = 0;
1090
1091         for (i = 0; i < BTREE_ID_NR; i++)
1092                 ids[i] = i;
1093         bubble_sort(ids, BTREE_ID_NR, btree_id_gc_phase_cmp);
1094
1095         for (i = 0; i < BTREE_ID_NR && !ret; i++)
1096                 ret = initial
1097                         ? bch2_gc_btree_init(trans, ids[i], metadata_only)
1098                         : bch2_gc_btree(trans, ids[i], initial, metadata_only);
1099
1100         for (i = BTREE_ID_NR; i < btree_id_nr_alive(c) && !ret; i++) {
1101                 if (!bch2_btree_id_root(c, i)->alive)
1102                         continue;
1103
1104                 ret = initial
1105                         ? bch2_gc_btree_init(trans, i, metadata_only)
1106                         : bch2_gc_btree(trans, i, initial, metadata_only);
1107         }
1108
1109         bch2_trans_put(trans);
1110         bch_err_fn(c, ret);
1111         return ret;
1112 }
1113
1114 static void mark_metadata_sectors(struct bch_fs *c, struct bch_dev *ca,
1115                                   u64 start, u64 end,
1116                                   enum bch_data_type type,
1117                                   unsigned flags)
1118 {
1119         u64 b = sector_to_bucket(ca, start);
1120
1121         do {
1122                 unsigned sectors =
1123                         min_t(u64, bucket_to_sector(ca, b + 1), end) - start;
1124
1125                 bch2_mark_metadata_bucket(c, ca, b, type, sectors,
1126                                           gc_phase(GC_PHASE_SB), flags);
1127                 b++;
1128                 start += sectors;
1129         } while (start < end);
1130 }
1131
1132 static void bch2_mark_dev_superblock(struct bch_fs *c, struct bch_dev *ca,
1133                                      unsigned flags)
1134 {
1135         struct bch_sb_layout *layout = &ca->disk_sb.sb->layout;
1136         unsigned i;
1137         u64 b;
1138
1139         for (i = 0; i < layout->nr_superblocks; i++) {
1140                 u64 offset = le64_to_cpu(layout->sb_offset[i]);
1141
1142                 if (offset == BCH_SB_SECTOR)
1143                         mark_metadata_sectors(c, ca, 0, BCH_SB_SECTOR,
1144                                               BCH_DATA_sb, flags);
1145
1146                 mark_metadata_sectors(c, ca, offset,
1147                                       offset + (1 << layout->sb_max_size_bits),
1148                                       BCH_DATA_sb, flags);
1149         }
1150
1151         for (i = 0; i < ca->journal.nr; i++) {
1152                 b = ca->journal.buckets[i];
1153                 bch2_mark_metadata_bucket(c, ca, b, BCH_DATA_journal,
1154                                           ca->mi.bucket_size,
1155                                           gc_phase(GC_PHASE_SB), flags);
1156         }
1157 }
1158
1159 static void bch2_mark_superblocks(struct bch_fs *c)
1160 {
1161         mutex_lock(&c->sb_lock);
1162         gc_pos_set(c, gc_phase(GC_PHASE_SB));
1163
1164         for_each_online_member(c, ca)
1165                 bch2_mark_dev_superblock(c, ca, BTREE_TRIGGER_GC);
1166         mutex_unlock(&c->sb_lock);
1167 }
1168
1169 #if 0
1170 /* Also see bch2_pending_btree_node_free_insert_done() */
1171 static void bch2_mark_pending_btree_node_frees(struct bch_fs *c)
1172 {
1173         struct btree_update *as;
1174         struct pending_btree_node_free *d;
1175
1176         mutex_lock(&c->btree_interior_update_lock);
1177         gc_pos_set(c, gc_phase(GC_PHASE_PENDING_DELETE));
1178
1179         for_each_pending_btree_node_free(c, as, d)
1180                 if (d->index_update_done)
1181                         bch2_mark_key(c, bkey_i_to_s_c(&d->key), BTREE_TRIGGER_GC);
1182
1183         mutex_unlock(&c->btree_interior_update_lock);
1184 }
1185 #endif
1186
1187 static void bch2_gc_free(struct bch_fs *c)
1188 {
1189         genradix_free(&c->reflink_gc_table);
1190         genradix_free(&c->gc_stripes);
1191
1192         for_each_member_device(c, ca) {
1193                 kvpfree(rcu_dereference_protected(ca->buckets_gc, 1),
1194                         sizeof(struct bucket_array) +
1195                         ca->mi.nbuckets * sizeof(struct bucket));
1196                 ca->buckets_gc = NULL;
1197
1198                 free_percpu(ca->usage_gc);
1199                 ca->usage_gc = NULL;
1200         }
1201
1202         free_percpu(c->usage_gc);
1203         c->usage_gc = NULL;
1204 }
1205
1206 static int bch2_gc_done(struct bch_fs *c,
1207                         bool initial, bool metadata_only)
1208 {
1209         struct bch_dev *ca = NULL;
1210         struct printbuf buf = PRINTBUF;
1211         bool verify = !metadata_only &&
1212                 !c->opts.reconstruct_alloc &&
1213                 (!initial || (c->sb.compat & (1ULL << BCH_COMPAT_alloc_info)));
1214         unsigned i;
1215         int ret = 0;
1216
1217         percpu_down_write(&c->mark_lock);
1218
1219 #define copy_field(_err, _f, _msg, ...)                                 \
1220         if (dst->_f != src->_f &&                                       \
1221             (!verify ||                                                 \
1222              fsck_err(c, _err, _msg ": got %llu, should be %llu"        \
1223                       , ##__VA_ARGS__, dst->_f, src->_f)))              \
1224                 dst->_f = src->_f
1225 #define copy_dev_field(_err, _f, _msg, ...)                             \
1226         copy_field(_err, _f, "dev %u has wrong " _msg, ca->dev_idx, ##__VA_ARGS__)
1227 #define copy_fs_field(_err, _f, _msg, ...)                              \
1228         copy_field(_err, _f, "fs has wrong " _msg, ##__VA_ARGS__)
1229
1230         for (i = 0; i < ARRAY_SIZE(c->usage); i++)
1231                 bch2_fs_usage_acc_to_base(c, i);
1232
1233         __for_each_member_device(c, ca) {
1234                 struct bch_dev_usage *dst = ca->usage_base;
1235                 struct bch_dev_usage *src = (void *)
1236                         bch2_acc_percpu_u64s((u64 __percpu *) ca->usage_gc,
1237                                              dev_usage_u64s());
1238
1239                 for (i = 0; i < BCH_DATA_NR; i++) {
1240                         copy_dev_field(dev_usage_buckets_wrong,
1241                                        d[i].buckets,    "%s buckets", bch2_data_types[i]);
1242                         copy_dev_field(dev_usage_sectors_wrong,
1243                                        d[i].sectors,    "%s sectors", bch2_data_types[i]);
1244                         copy_dev_field(dev_usage_fragmented_wrong,
1245                                        d[i].fragmented, "%s fragmented", bch2_data_types[i]);
1246                 }
1247         }
1248
1249         {
1250                 unsigned nr = fs_usage_u64s(c);
1251                 struct bch_fs_usage *dst = c->usage_base;
1252                 struct bch_fs_usage *src = (void *)
1253                         bch2_acc_percpu_u64s((u64 __percpu *) c->usage_gc, nr);
1254
1255                 copy_fs_field(fs_usage_hidden_wrong,
1256                               hidden,           "hidden");
1257                 copy_fs_field(fs_usage_btree_wrong,
1258                               btree,            "btree");
1259
1260                 if (!metadata_only) {
1261                         copy_fs_field(fs_usage_data_wrong,
1262                                       data,     "data");
1263                         copy_fs_field(fs_usage_cached_wrong,
1264                                       cached,   "cached");
1265                         copy_fs_field(fs_usage_reserved_wrong,
1266                                       reserved, "reserved");
1267                         copy_fs_field(fs_usage_nr_inodes_wrong,
1268                                       nr_inodes,"nr_inodes");
1269
1270                         for (i = 0; i < BCH_REPLICAS_MAX; i++)
1271                                 copy_fs_field(fs_usage_persistent_reserved_wrong,
1272                                               persistent_reserved[i],
1273                                               "persistent_reserved[%i]", i);
1274                 }
1275
1276                 for (i = 0; i < c->replicas.nr; i++) {
1277                         struct bch_replicas_entry_v1 *e =
1278                                 cpu_replicas_entry(&c->replicas, i);
1279
1280                         if (metadata_only &&
1281                             (e->data_type == BCH_DATA_user ||
1282                              e->data_type == BCH_DATA_cached))
1283                                 continue;
1284
1285                         printbuf_reset(&buf);
1286                         bch2_replicas_entry_to_text(&buf, e);
1287
1288                         copy_fs_field(fs_usage_replicas_wrong,
1289                                       replicas[i], "%s", buf.buf);
1290                 }
1291         }
1292
1293 #undef copy_fs_field
1294 #undef copy_dev_field
1295 #undef copy_stripe_field
1296 #undef copy_field
1297 fsck_err:
1298         if (ca)
1299                 percpu_ref_put(&ca->ref);
1300         bch_err_fn(c, ret);
1301
1302         percpu_up_write(&c->mark_lock);
1303         printbuf_exit(&buf);
1304         return ret;
1305 }
1306
1307 static int bch2_gc_start(struct bch_fs *c)
1308 {
1309         BUG_ON(c->usage_gc);
1310
1311         c->usage_gc = __alloc_percpu_gfp(fs_usage_u64s(c) * sizeof(u64),
1312                                          sizeof(u64), GFP_KERNEL);
1313         if (!c->usage_gc) {
1314                 bch_err(c, "error allocating c->usage_gc");
1315                 return -BCH_ERR_ENOMEM_gc_start;
1316         }
1317
1318         for_each_member_device(c, ca) {
1319                 BUG_ON(ca->usage_gc);
1320
1321                 ca->usage_gc = alloc_percpu(struct bch_dev_usage);
1322                 if (!ca->usage_gc) {
1323                         bch_err(c, "error allocating ca->usage_gc");
1324                         percpu_ref_put(&ca->ref);
1325                         return -BCH_ERR_ENOMEM_gc_start;
1326                 }
1327
1328                 this_cpu_write(ca->usage_gc->d[BCH_DATA_free].buckets,
1329                                ca->mi.nbuckets - ca->mi.first_bucket);
1330         }
1331
1332         return 0;
1333 }
1334
1335 static int bch2_gc_reset(struct bch_fs *c)
1336 {
1337         for_each_member_device(c, ca) {
1338                 free_percpu(ca->usage_gc);
1339                 ca->usage_gc = NULL;
1340         }
1341
1342         free_percpu(c->usage_gc);
1343         c->usage_gc = NULL;
1344
1345         return bch2_gc_start(c);
1346 }
1347
1348 /* returns true if not equal */
1349 static inline bool bch2_alloc_v4_cmp(struct bch_alloc_v4 l,
1350                                      struct bch_alloc_v4 r)
1351 {
1352         return  l.gen != r.gen                          ||
1353                 l.oldest_gen != r.oldest_gen            ||
1354                 l.data_type != r.data_type              ||
1355                 l.dirty_sectors != r.dirty_sectors      ||
1356                 l.cached_sectors != r.cached_sectors     ||
1357                 l.stripe_redundancy != r.stripe_redundancy ||
1358                 l.stripe != r.stripe;
1359 }
1360
1361 static int bch2_alloc_write_key(struct btree_trans *trans,
1362                                 struct btree_iter *iter,
1363                                 struct bkey_s_c k,
1364                                 bool metadata_only)
1365 {
1366         struct bch_fs *c = trans->c;
1367         struct bch_dev *ca = bch_dev_bkey_exists(c, iter->pos.inode);
1368         struct bucket gc, *b;
1369         struct bkey_i_alloc_v4 *a;
1370         struct bch_alloc_v4 old_convert, new;
1371         const struct bch_alloc_v4 *old;
1372         enum bch_data_type type;
1373         int ret;
1374
1375         old = bch2_alloc_to_v4(k, &old_convert);
1376         new = *old;
1377
1378         percpu_down_read(&c->mark_lock);
1379         b = gc_bucket(ca, iter->pos.offset);
1380
1381         /*
1382          * b->data_type doesn't yet include need_discard & need_gc_gen states -
1383          * fix that here:
1384          */
1385         type = __alloc_data_type(b->dirty_sectors,
1386                                  b->cached_sectors,
1387                                  b->stripe,
1388                                  *old,
1389                                  b->data_type);
1390         if (b->data_type != type) {
1391                 struct bch_dev_usage *u;
1392
1393                 preempt_disable();
1394                 u = this_cpu_ptr(ca->usage_gc);
1395                 u->d[b->data_type].buckets--;
1396                 b->data_type = type;
1397                 u->d[b->data_type].buckets++;
1398                 preempt_enable();
1399         }
1400
1401         gc = *b;
1402         percpu_up_read(&c->mark_lock);
1403
1404         if (metadata_only &&
1405             gc.data_type != BCH_DATA_sb &&
1406             gc.data_type != BCH_DATA_journal &&
1407             gc.data_type != BCH_DATA_btree)
1408                 return 0;
1409
1410         if (gen_after(old->gen, gc.gen))
1411                 return 0;
1412
1413         if (c->opts.reconstruct_alloc ||
1414             fsck_err_on(new.data_type != gc.data_type, c,
1415                         alloc_key_data_type_wrong,
1416                         "bucket %llu:%llu gen %u has wrong data_type"
1417                         ": got %s, should be %s",
1418                         iter->pos.inode, iter->pos.offset,
1419                         gc.gen,
1420                         bch2_data_types[new.data_type],
1421                         bch2_data_types[gc.data_type]))
1422                 new.data_type = gc.data_type;
1423
1424 #define copy_bucket_field(_errtype, _f)                                 \
1425         if (c->opts.reconstruct_alloc ||                                \
1426             fsck_err_on(new._f != gc._f, c, _errtype,                   \
1427                         "bucket %llu:%llu gen %u data type %s has wrong " #_f   \
1428                         ": got %u, should be %u",                       \
1429                         iter->pos.inode, iter->pos.offset,              \
1430                         gc.gen,                                         \
1431                         bch2_data_types[gc.data_type],                  \
1432                         new._f, gc._f))                                 \
1433                 new._f = gc._f;                                         \
1434
1435         copy_bucket_field(alloc_key_gen_wrong,
1436                           gen);
1437         copy_bucket_field(alloc_key_dirty_sectors_wrong,
1438                           dirty_sectors);
1439         copy_bucket_field(alloc_key_cached_sectors_wrong,
1440                           cached_sectors);
1441         copy_bucket_field(alloc_key_stripe_wrong,
1442                           stripe);
1443         copy_bucket_field(alloc_key_stripe_redundancy_wrong,
1444                           stripe_redundancy);
1445 #undef copy_bucket_field
1446
1447         if (!bch2_alloc_v4_cmp(*old, new))
1448                 return 0;
1449
1450         a = bch2_alloc_to_v4_mut(trans, k);
1451         ret = PTR_ERR_OR_ZERO(a);
1452         if (ret)
1453                 return ret;
1454
1455         a->v = new;
1456
1457         /*
1458          * The trigger normally makes sure this is set, but we're not running
1459          * triggers:
1460          */
1461         if (a->v.data_type == BCH_DATA_cached && !a->v.io_time[READ])
1462                 a->v.io_time[READ] = max_t(u64, 1, atomic64_read(&c->io_clock[READ].now));
1463
1464         ret = bch2_trans_update(trans, iter, &a->k_i, BTREE_TRIGGER_NORUN);
1465 fsck_err:
1466         return ret;
1467 }
1468
1469 static int bch2_gc_alloc_done(struct bch_fs *c, bool metadata_only)
1470 {
1471         int ret = 0;
1472
1473         for_each_member_device(c, ca) {
1474                 ret = bch2_trans_run(c,
1475                         for_each_btree_key_upto_commit(trans, iter, BTREE_ID_alloc,
1476                                         POS(ca->dev_idx, ca->mi.first_bucket),
1477                                         POS(ca->dev_idx, ca->mi.nbuckets - 1),
1478                                         BTREE_ITER_SLOTS|BTREE_ITER_PREFETCH, k,
1479                                         NULL, NULL, BCH_TRANS_COMMIT_lazy_rw,
1480                                 bch2_alloc_write_key(trans, &iter, k, metadata_only)));
1481                 if (ret) {
1482                         percpu_ref_put(&ca->ref);
1483                         break;
1484                 }
1485         }
1486
1487         bch_err_fn(c, ret);
1488         return ret;
1489 }
1490
1491 static int bch2_gc_alloc_start(struct bch_fs *c, bool metadata_only)
1492 {
1493         for_each_member_device(c, ca) {
1494                 struct bucket_array *buckets = kvpmalloc(sizeof(struct bucket_array) +
1495                                 ca->mi.nbuckets * sizeof(struct bucket),
1496                                 GFP_KERNEL|__GFP_ZERO);
1497                 if (!buckets) {
1498                         percpu_ref_put(&ca->ref);
1499                         bch_err(c, "error allocating ca->buckets[gc]");
1500                         return -BCH_ERR_ENOMEM_gc_alloc_start;
1501                 }
1502
1503                 buckets->first_bucket   = ca->mi.first_bucket;
1504                 buckets->nbuckets       = ca->mi.nbuckets;
1505                 rcu_assign_pointer(ca->buckets_gc, buckets);
1506         }
1507
1508         int ret = bch2_trans_run(c,
1509                 for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN,
1510                                          BTREE_ITER_PREFETCH, k, ({
1511                         struct bch_dev *ca = bch_dev_bkey_exists(c, k.k->p.inode);
1512                         struct bucket *g = gc_bucket(ca, k.k->p.offset);
1513
1514                         struct bch_alloc_v4 a_convert;
1515                         const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert);
1516
1517                         g->gen_valid    = 1;
1518                         g->gen          = a->gen;
1519
1520                         if (metadata_only &&
1521                             (a->data_type == BCH_DATA_user ||
1522                              a->data_type == BCH_DATA_cached ||
1523                              a->data_type == BCH_DATA_parity)) {
1524                                 g->data_type            = a->data_type;
1525                                 g->dirty_sectors        = a->dirty_sectors;
1526                                 g->cached_sectors       = a->cached_sectors;
1527                                 g->stripe               = a->stripe;
1528                                 g->stripe_redundancy    = a->stripe_redundancy;
1529                         }
1530
1531                         0;
1532                 })));
1533         bch_err_fn(c, ret);
1534         return ret;
1535 }
1536
1537 static void bch2_gc_alloc_reset(struct bch_fs *c, bool metadata_only)
1538 {
1539         for_each_member_device(c, ca) {
1540                 struct bucket_array *buckets = gc_bucket_array(ca);
1541                 struct bucket *g;
1542
1543                 for_each_bucket(g, buckets) {
1544                         if (metadata_only &&
1545                             (g->data_type == BCH_DATA_user ||
1546                              g->data_type == BCH_DATA_cached ||
1547                              g->data_type == BCH_DATA_parity))
1548                                 continue;
1549                         g->data_type = 0;
1550                         g->dirty_sectors = 0;
1551                         g->cached_sectors = 0;
1552                 }
1553         }
1554 }
1555
1556 static int bch2_gc_write_reflink_key(struct btree_trans *trans,
1557                                      struct btree_iter *iter,
1558                                      struct bkey_s_c k,
1559                                      size_t *idx)
1560 {
1561         struct bch_fs *c = trans->c;
1562         const __le64 *refcount = bkey_refcount_c(k);
1563         struct printbuf buf = PRINTBUF;
1564         struct reflink_gc *r;
1565         int ret = 0;
1566
1567         if (!refcount)
1568                 return 0;
1569
1570         while ((r = genradix_ptr(&c->reflink_gc_table, *idx)) &&
1571                r->offset < k.k->p.offset)
1572                 ++*idx;
1573
1574         if (!r ||
1575             r->offset != k.k->p.offset ||
1576             r->size != k.k->size) {
1577                 bch_err(c, "unexpected inconsistency walking reflink table at gc finish");
1578                 return -EINVAL;
1579         }
1580
1581         if (fsck_err_on(r->refcount != le64_to_cpu(*refcount), c,
1582                         reflink_v_refcount_wrong,
1583                         "reflink key has wrong refcount:\n"
1584                         "  %s\n"
1585                         "  should be %u",
1586                         (bch2_bkey_val_to_text(&buf, c, k), buf.buf),
1587                         r->refcount)) {
1588                 struct bkey_i *new = bch2_bkey_make_mut(trans, iter, &k, 0);
1589
1590                 ret = PTR_ERR_OR_ZERO(new);
1591                 if (ret)
1592                         return ret;
1593
1594                 if (!r->refcount)
1595                         new->k.type = KEY_TYPE_deleted;
1596                 else
1597                         *bkey_refcount(bkey_i_to_s(new)) = cpu_to_le64(r->refcount);
1598         }
1599 fsck_err:
1600         printbuf_exit(&buf);
1601         return ret;
1602 }
1603
1604 static int bch2_gc_reflink_done(struct bch_fs *c, bool metadata_only)
1605 {
1606         size_t idx = 0;
1607
1608         if (metadata_only)
1609                 return 0;
1610
1611         int ret = bch2_trans_run(c,
1612                 for_each_btree_key_commit(trans, iter,
1613                                 BTREE_ID_reflink, POS_MIN,
1614                                 BTREE_ITER_PREFETCH, k,
1615                                 NULL, NULL, BCH_TRANS_COMMIT_no_enospc,
1616                         bch2_gc_write_reflink_key(trans, &iter, k, &idx)));
1617         c->reflink_gc_nr = 0;
1618         return ret;
1619 }
1620
1621 static int bch2_gc_reflink_start(struct bch_fs *c,
1622                                  bool metadata_only)
1623 {
1624
1625         if (metadata_only)
1626                 return 0;
1627
1628         c->reflink_gc_nr = 0;
1629
1630         int ret = bch2_trans_run(c,
1631                 for_each_btree_key(trans, iter, BTREE_ID_reflink, POS_MIN,
1632                                    BTREE_ITER_PREFETCH, k, ({
1633                         const __le64 *refcount = bkey_refcount_c(k);
1634
1635                         if (!refcount)
1636                                 continue;
1637
1638                         struct reflink_gc *r = genradix_ptr_alloc(&c->reflink_gc_table,
1639                                                         c->reflink_gc_nr++, GFP_KERNEL);
1640                         if (!r) {
1641                                 ret = -BCH_ERR_ENOMEM_gc_reflink_start;
1642                                 break;
1643                         }
1644
1645                         r->offset       = k.k->p.offset;
1646                         r->size         = k.k->size;
1647                         r->refcount     = 0;
1648                         0;
1649                 })));
1650
1651         bch_err_fn(c, ret);
1652         return ret;
1653 }
1654
1655 static void bch2_gc_reflink_reset(struct bch_fs *c, bool metadata_only)
1656 {
1657         struct genradix_iter iter;
1658         struct reflink_gc *r;
1659
1660         genradix_for_each(&c->reflink_gc_table, iter, r)
1661                 r->refcount = 0;
1662 }
1663
1664 static int bch2_gc_write_stripes_key(struct btree_trans *trans,
1665                                      struct btree_iter *iter,
1666                                      struct bkey_s_c k)
1667 {
1668         struct bch_fs *c = trans->c;
1669         struct printbuf buf = PRINTBUF;
1670         const struct bch_stripe *s;
1671         struct gc_stripe *m;
1672         bool bad = false;
1673         unsigned i;
1674         int ret = 0;
1675
1676         if (k.k->type != KEY_TYPE_stripe)
1677                 return 0;
1678
1679         s = bkey_s_c_to_stripe(k).v;
1680         m = genradix_ptr(&c->gc_stripes, k.k->p.offset);
1681
1682         for (i = 0; i < s->nr_blocks; i++) {
1683                 u32 old = stripe_blockcount_get(s, i);
1684                 u32 new = (m ? m->block_sectors[i] : 0);
1685
1686                 if (old != new) {
1687                         prt_printf(&buf, "stripe block %u has wrong sector count: got %u, should be %u\n",
1688                                    i, old, new);
1689                         bad = true;
1690                 }
1691         }
1692
1693         if (bad)
1694                 bch2_bkey_val_to_text(&buf, c, k);
1695
1696         if (fsck_err_on(bad, c, stripe_sector_count_wrong,
1697                         "%s", buf.buf)) {
1698                 struct bkey_i_stripe *new;
1699
1700                 new = bch2_trans_kmalloc(trans, bkey_bytes(k.k));
1701                 ret = PTR_ERR_OR_ZERO(new);
1702                 if (ret)
1703                         return ret;
1704
1705                 bkey_reassemble(&new->k_i, k);
1706
1707                 for (i = 0; i < new->v.nr_blocks; i++)
1708                         stripe_blockcount_set(&new->v, i, m ? m->block_sectors[i] : 0);
1709
1710                 ret = bch2_trans_update(trans, iter, &new->k_i, 0);
1711         }
1712 fsck_err:
1713         printbuf_exit(&buf);
1714         return ret;
1715 }
1716
1717 static int bch2_gc_stripes_done(struct bch_fs *c, bool metadata_only)
1718 {
1719         if (metadata_only)
1720                 return 0;
1721
1722         return bch2_trans_run(c,
1723                 for_each_btree_key_commit(trans, iter,
1724                                 BTREE_ID_stripes, POS_MIN,
1725                                 BTREE_ITER_PREFETCH, k,
1726                                 NULL, NULL, BCH_TRANS_COMMIT_no_enospc,
1727                         bch2_gc_write_stripes_key(trans, &iter, k)));
1728 }
1729
1730 static void bch2_gc_stripes_reset(struct bch_fs *c, bool metadata_only)
1731 {
1732         genradix_free(&c->gc_stripes);
1733 }
1734
1735 /**
1736  * bch2_gc - walk _all_ references to buckets, and recompute them:
1737  *
1738  * @c:                  filesystem object
1739  * @initial:            are we in recovery?
1740  * @metadata_only:      are we just checking metadata references, or everything?
1741  *
1742  * Returns: 0 on success, or standard errcode on failure
1743  *
1744  * Order matters here:
1745  *  - Concurrent GC relies on the fact that we have a total ordering for
1746  *    everything that GC walks - see  gc_will_visit_node(),
1747  *    gc_will_visit_root()
1748  *
1749  *  - also, references move around in the course of index updates and
1750  *    various other crap: everything needs to agree on the ordering
1751  *    references are allowed to move around in - e.g., we're allowed to
1752  *    start with a reference owned by an open_bucket (the allocator) and
1753  *    move it to the btree, but not the reverse.
1754  *
1755  *    This is necessary to ensure that gc doesn't miss references that
1756  *    move around - if references move backwards in the ordering GC
1757  *    uses, GC could skip past them
1758  */
1759 int bch2_gc(struct bch_fs *c, bool initial, bool metadata_only)
1760 {
1761         unsigned iter = 0;
1762         int ret;
1763
1764         lockdep_assert_held(&c->state_lock);
1765
1766         down_write(&c->gc_lock);
1767
1768         bch2_btree_interior_updates_flush(c);
1769
1770         ret   = bch2_gc_start(c) ?:
1771                 bch2_gc_alloc_start(c, metadata_only) ?:
1772                 bch2_gc_reflink_start(c, metadata_only);
1773         if (ret)
1774                 goto out;
1775 again:
1776         gc_pos_set(c, gc_phase(GC_PHASE_START));
1777
1778         bch2_mark_superblocks(c);
1779
1780         ret = bch2_gc_btrees(c, initial, metadata_only);
1781
1782         if (ret)
1783                 goto out;
1784
1785 #if 0
1786         bch2_mark_pending_btree_node_frees(c);
1787 #endif
1788         c->gc_count++;
1789
1790         if (test_bit(BCH_FS_need_another_gc, &c->flags) ||
1791             (!iter && bch2_test_restart_gc)) {
1792                 if (iter++ > 2) {
1793                         bch_info(c, "Unable to fix bucket gens, looping");
1794                         ret = -EINVAL;
1795                         goto out;
1796                 }
1797
1798                 /*
1799                  * XXX: make sure gens we fixed got saved
1800                  */
1801                 bch_info(c, "Second GC pass needed, restarting:");
1802                 clear_bit(BCH_FS_need_another_gc, &c->flags);
1803                 __gc_pos_set(c, gc_phase(GC_PHASE_NOT_RUNNING));
1804
1805                 bch2_gc_stripes_reset(c, metadata_only);
1806                 bch2_gc_alloc_reset(c, metadata_only);
1807                 bch2_gc_reflink_reset(c, metadata_only);
1808                 ret = bch2_gc_reset(c);
1809                 if (ret)
1810                         goto out;
1811
1812                 /* flush fsck errors, reset counters */
1813                 bch2_flush_fsck_errs(c);
1814                 goto again;
1815         }
1816 out:
1817         if (!ret) {
1818                 bch2_journal_block(&c->journal);
1819
1820                 ret   = bch2_gc_stripes_done(c, metadata_only) ?:
1821                         bch2_gc_reflink_done(c, metadata_only) ?:
1822                         bch2_gc_alloc_done(c, metadata_only) ?:
1823                         bch2_gc_done(c, initial, metadata_only);
1824
1825                 bch2_journal_unblock(&c->journal);
1826         }
1827
1828         percpu_down_write(&c->mark_lock);
1829         /* Indicates that gc is no longer in progress: */
1830         __gc_pos_set(c, gc_phase(GC_PHASE_NOT_RUNNING));
1831
1832         bch2_gc_free(c);
1833         percpu_up_write(&c->mark_lock);
1834
1835         up_write(&c->gc_lock);
1836
1837         /*
1838          * At startup, allocations can happen directly instead of via the
1839          * allocator thread - issue wakeup in case they blocked on gc_lock:
1840          */
1841         closure_wake_up(&c->freelist_wait);
1842         bch_err_fn(c, ret);
1843         return ret;
1844 }
1845
1846 static int gc_btree_gens_key(struct btree_trans *trans,
1847                              struct btree_iter *iter,
1848                              struct bkey_s_c k)
1849 {
1850         struct bch_fs *c = trans->c;
1851         struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
1852         struct bkey_i *u;
1853         int ret;
1854
1855         percpu_down_read(&c->mark_lock);
1856         bkey_for_each_ptr(ptrs, ptr) {
1857                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
1858
1859                 if (ptr_stale(ca, ptr) > 16) {
1860                         percpu_up_read(&c->mark_lock);
1861                         goto update;
1862                 }
1863         }
1864
1865         bkey_for_each_ptr(ptrs, ptr) {
1866                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
1867                 u8 *gen = &ca->oldest_gen[PTR_BUCKET_NR(ca, ptr)];
1868
1869                 if (gen_after(*gen, ptr->gen))
1870                         *gen = ptr->gen;
1871         }
1872         percpu_up_read(&c->mark_lock);
1873         return 0;
1874 update:
1875         u = bch2_bkey_make_mut(trans, iter, &k, 0);
1876         ret = PTR_ERR_OR_ZERO(u);
1877         if (ret)
1878                 return ret;
1879
1880         bch2_extent_normalize(c, bkey_i_to_s(u));
1881         return 0;
1882 }
1883
1884 static int bch2_alloc_write_oldest_gen(struct btree_trans *trans, struct btree_iter *iter,
1885                                        struct bkey_s_c k)
1886 {
1887         struct bch_dev *ca = bch_dev_bkey_exists(trans->c, iter->pos.inode);
1888         struct bch_alloc_v4 a_convert;
1889         const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert);
1890         struct bkey_i_alloc_v4 *a_mut;
1891         int ret;
1892
1893         if (a->oldest_gen == ca->oldest_gen[iter->pos.offset])
1894                 return 0;
1895
1896         a_mut = bch2_alloc_to_v4_mut(trans, k);
1897         ret = PTR_ERR_OR_ZERO(a_mut);
1898         if (ret)
1899                 return ret;
1900
1901         a_mut->v.oldest_gen = ca->oldest_gen[iter->pos.offset];
1902         a_mut->v.data_type = alloc_data_type(a_mut->v, a_mut->v.data_type);
1903
1904         return bch2_trans_update(trans, iter, &a_mut->k_i, 0);
1905 }
1906
1907 int bch2_gc_gens(struct bch_fs *c)
1908 {
1909         u64 b, start_time = local_clock();
1910         int ret;
1911
1912         /*
1913          * Ideally we would be using state_lock and not gc_lock here, but that
1914          * introduces a deadlock in the RO path - we currently take the state
1915          * lock at the start of going RO, thus the gc thread may get stuck:
1916          */
1917         if (!mutex_trylock(&c->gc_gens_lock))
1918                 return 0;
1919
1920         trace_and_count(c, gc_gens_start, c);
1921         down_read(&c->gc_lock);
1922
1923         for_each_member_device(c, ca) {
1924                 struct bucket_gens *gens = bucket_gens(ca);
1925
1926                 BUG_ON(ca->oldest_gen);
1927
1928                 ca->oldest_gen = kvmalloc(gens->nbuckets, GFP_KERNEL);
1929                 if (!ca->oldest_gen) {
1930                         percpu_ref_put(&ca->ref);
1931                         ret = -BCH_ERR_ENOMEM_gc_gens;
1932                         goto err;
1933                 }
1934
1935                 for (b = gens->first_bucket;
1936                      b < gens->nbuckets; b++)
1937                         ca->oldest_gen[b] = gens->b[b];
1938         }
1939
1940         for (unsigned i = 0; i < BTREE_ID_NR; i++)
1941                 if (btree_type_has_ptrs(i)) {
1942                         c->gc_gens_btree = i;
1943                         c->gc_gens_pos = POS_MIN;
1944
1945                         ret = bch2_trans_run(c,
1946                                 for_each_btree_key_commit(trans, iter, i,
1947                                                 POS_MIN,
1948                                                 BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS,
1949                                                 k,
1950                                                 NULL, NULL,
1951                                                 BCH_TRANS_COMMIT_no_enospc,
1952                                         gc_btree_gens_key(trans, &iter, k)));
1953                         if (ret)
1954                                 goto err;
1955                 }
1956
1957         ret = bch2_trans_run(c,
1958                 for_each_btree_key_commit(trans, iter, BTREE_ID_alloc,
1959                                 POS_MIN,
1960                                 BTREE_ITER_PREFETCH,
1961                                 k,
1962                                 NULL, NULL,
1963                                 BCH_TRANS_COMMIT_no_enospc,
1964                         bch2_alloc_write_oldest_gen(trans, &iter, k)));
1965         if (ret)
1966                 goto err;
1967
1968         c->gc_gens_btree        = 0;
1969         c->gc_gens_pos          = POS_MIN;
1970
1971         c->gc_count++;
1972
1973         bch2_time_stats_update(&c->times[BCH_TIME_btree_gc], start_time);
1974         trace_and_count(c, gc_gens_end, c);
1975 err:
1976         for_each_member_device(c, ca) {
1977                 kvfree(ca->oldest_gen);
1978                 ca->oldest_gen = NULL;
1979         }
1980
1981         up_read(&c->gc_lock);
1982         mutex_unlock(&c->gc_gens_lock);
1983         if (!bch2_err_matches(ret, EROFS))
1984                 bch_err_fn(c, ret);
1985         return ret;
1986 }
1987
1988 static int bch2_gc_thread(void *arg)
1989 {
1990         struct bch_fs *c = arg;
1991         struct io_clock *clock = &c->io_clock[WRITE];
1992         unsigned long last = atomic64_read(&clock->now);
1993         unsigned last_kick = atomic_read(&c->kick_gc);
1994
1995         set_freezable();
1996
1997         while (1) {
1998                 while (1) {
1999                         set_current_state(TASK_INTERRUPTIBLE);
2000
2001                         if (kthread_should_stop()) {
2002                                 __set_current_state(TASK_RUNNING);
2003                                 return 0;
2004                         }
2005
2006                         if (atomic_read(&c->kick_gc) != last_kick)
2007                                 break;
2008
2009                         if (c->btree_gc_periodic) {
2010                                 unsigned long next = last + c->capacity / 16;
2011
2012                                 if (atomic64_read(&clock->now) >= next)
2013                                         break;
2014
2015                                 bch2_io_clock_schedule_timeout(clock, next);
2016                         } else {
2017                                 schedule();
2018                         }
2019
2020                         try_to_freeze();
2021                 }
2022                 __set_current_state(TASK_RUNNING);
2023
2024                 last = atomic64_read(&clock->now);
2025                 last_kick = atomic_read(&c->kick_gc);
2026
2027                 /*
2028                  * Full gc is currently incompatible with btree key cache:
2029                  */
2030 #if 0
2031                 ret = bch2_gc(c, false, false);
2032 #else
2033                 bch2_gc_gens(c);
2034 #endif
2035                 debug_check_no_locks_held();
2036         }
2037
2038         return 0;
2039 }
2040
2041 void bch2_gc_thread_stop(struct bch_fs *c)
2042 {
2043         struct task_struct *p;
2044
2045         p = c->gc_thread;
2046         c->gc_thread = NULL;
2047
2048         if (p) {
2049                 kthread_stop(p);
2050                 put_task_struct(p);
2051         }
2052 }
2053
2054 int bch2_gc_thread_start(struct bch_fs *c)
2055 {
2056         struct task_struct *p;
2057
2058         if (c->gc_thread)
2059                 return 0;
2060
2061         p = kthread_create(bch2_gc_thread, c, "bch-gc/%s", c->name);
2062         if (IS_ERR(p)) {
2063                 bch_err_fn(c, PTR_ERR(p));
2064                 return PTR_ERR(p);
2065         }
2066
2067         get_task_struct(p);
2068         c->gc_thread = p;
2069         wake_up_process(p);
2070         return 0;
2071 }