fs/bcachefs/btree_gc.c

   1 // SPDX-License-Identifier: GPL-2.0
   2 /*
   3  * Copyright (C) 2010 Kent Overstreet <[email protected]>
   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(trans, &iter, b);
 393         iter.prefetch = true;
 394
 395         while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 396                 BUG_ON(bpos_lt(k.k->p, b->data->min_key));
 397                 BUG_ON(bpos_gt(k.k->p, b->data->max_key));
 398
 399                 bch2_btree_and_journal_iter_advance(&iter);
 400                 bch2_bkey_buf_reassemble(&cur_k, c, k);
 401
 402                 cur = bch2_btree_node_get_noiter(trans, cur_k.k,
 403                                         b->c.btree_id, b->c.level - 1,
 404                                         false);
 405                 ret = PTR_ERR_OR_ZERO(cur);
 406
 407                 printbuf_reset(&buf);
 408                 bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(cur_k.k));
 409
 410                 if (mustfix_fsck_err_on(bch2_err_matches(ret, EIO), c,
 411                                 btree_node_unreadable,
 412                                 "Topology repair: unreadable btree node at btree %s level %u:\n"
 413                                 "  %s",
 414                                 bch2_btree_id_str(b->c.btree_id),
 415                                 b->c.level - 1,
 416                                 buf.buf)) {
 417                         bch2_btree_node_evict(trans, cur_k.k);
 418                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 419                                                       b->c.level, cur_k.k->k.p);
 420                         cur = NULL;
 421                         if (ret)
 422                                 break;
 423                         continue;
 424                 }
 425
 426                 bch_err_msg(c, ret, "getting btree node");
 427                 if (ret)
 428                         break;
 429
 430                 ret = btree_repair_node_boundaries(c, b, prev, cur);
 431
 432                 if (ret == DROP_THIS_NODE) {
 433                         six_unlock_read(&cur->c.lock);
 434                         bch2_btree_node_evict(trans, cur_k.k);
 435                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 436                                                       b->c.level, cur_k.k->k.p);
 437                         cur = NULL;
 438                         if (ret)
 439                                 break;
 440                         continue;
 441                 }
 442
 443                 if (prev)
 444                         six_unlock_read(&prev->c.lock);
 445                 prev = NULL;
 446
 447                 if (ret == DROP_PREV_NODE) {
 448                         bch2_btree_node_evict(trans, prev_k.k);
 449                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 450                                                       b->c.level, prev_k.k->k.p);
 451                         if (ret)
 452                                 break;
 453
 454                         bch2_btree_and_journal_iter_exit(&iter);
 455                         bch2_bkey_buf_exit(&prev_k, c);
 456                         bch2_bkey_buf_exit(&cur_k, c);
 457                         goto again;
 458                 } else if (ret)
 459                         break;
 460
 461                 prev = cur;
 462                 cur = NULL;
 463                 bch2_bkey_buf_copy(&prev_k, c, cur_k.k);
 464         }
 465
 466         if (!ret && !IS_ERR_OR_NULL(prev)) {
 467                 BUG_ON(cur);
 468                 ret = btree_repair_node_end(c, b, prev);
 469         }
 470
 471         if (!IS_ERR_OR_NULL(prev))
 472                 six_unlock_read(&prev->c.lock);
 473         prev = NULL;
 474         if (!IS_ERR_OR_NULL(cur))
 475                 six_unlock_read(&cur->c.lock);
 476         cur = NULL;
 477
 478         if (ret)
 479                 goto err;
 480
 481         bch2_btree_and_journal_iter_exit(&iter);
 482         bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b);
 483         iter.prefetch = true;
 484
 485         while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 486                 bch2_bkey_buf_reassemble(&cur_k, c, k);
 487                 bch2_btree_and_journal_iter_advance(&iter);
 488
 489                 cur = bch2_btree_node_get_noiter(trans, cur_k.k,
 490                                         b->c.btree_id, b->c.level - 1,
 491                                         false);
 492                 ret = PTR_ERR_OR_ZERO(cur);
 493
 494                 bch_err_msg(c, ret, "getting btree node");
 495                 if (ret)
 496                         goto err;
 497
 498                 ret = bch2_btree_repair_topology_recurse(trans, cur);
 499                 six_unlock_read(&cur->c.lock);
 500                 cur = NULL;
 501
 502                 if (ret == DROP_THIS_NODE) {
 503                         bch2_btree_node_evict(trans, cur_k.k);
 504                         ret = bch2_journal_key_delete(c, b->c.btree_id,
 505                                                       b->c.level, cur_k.k->k.p);
 506                         dropped_children = true;
 507                 }
 508
 509                 if (ret)
 510                         goto err;
 511
 512                 have_child = true;
 513         }
 514
 515         printbuf_reset(&buf);
 516         bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key));
 517
 518         if (mustfix_fsck_err_on(!have_child, c,
 519                         btree_node_topology_interior_node_empty,
 520                         "empty interior btree node at btree %s level %u\n"
 521                         "  %s",
 522                         bch2_btree_id_str(b->c.btree_id),
 523                         b->c.level, buf.buf))
 524                 ret = DROP_THIS_NODE;
 525 err:
 526 fsck_err:
 527         if (!IS_ERR_OR_NULL(prev))
 528                 six_unlock_read(&prev->c.lock);
 529         if (!IS_ERR_OR_NULL(cur))
 530                 six_unlock_read(&cur->c.lock);
 531
 532         bch2_btree_and_journal_iter_exit(&iter);
 533         bch2_bkey_buf_exit(&prev_k, c);
 534         bch2_bkey_buf_exit(&cur_k, c);
 535
 536         if (!ret && dropped_children)
 537                 goto again;
 538
 539         printbuf_exit(&buf);
 540         return ret;
 541 }
 542
 543 int bch2_check_topology(struct bch_fs *c)
 544 {
 545         struct btree_trans *trans = bch2_trans_get(c);
 546         struct btree *b;
 547         unsigned i;
 548         int ret = 0;
 549
 550         for (i = 0; i < btree_id_nr_alive(c) && !ret; i++) {
 551                 struct btree_root *r = bch2_btree_id_root(c, i);
 552
 553                 if (!r->alive)
 554                         continue;
 555
 556                 b = r->b;
 557                 if (btree_node_fake(b))
 558                         continue;
 559
 560                 btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_read);
 561                 ret = bch2_btree_repair_topology_recurse(trans, b);
 562                 six_unlock_read(&b->c.lock);
 563
 564                 if (ret == DROP_THIS_NODE) {
 565                         bch_err(c, "empty btree root - repair unimplemented");
 566                         ret = -BCH_ERR_fsck_repair_unimplemented;
 567                 }
 568         }
 569
 570         bch2_trans_put(trans);
 571
 572         return ret;
 573 }
 574
 575 static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id,
 576                                unsigned level, bool is_root,
 577                                struct bkey_s_c *k)
 578 {
 579         struct bch_fs *c = trans->c;
 580         struct bkey_ptrs_c ptrs_c = bch2_bkey_ptrs_c(*k);
 581         const union bch_extent_entry *entry_c;
 582         struct extent_ptr_decoded p = { 0 };
 583         bool do_update = false;
 584         struct printbuf buf = PRINTBUF;
 585         int ret = 0;
 586
 587         /*
 588          * XXX
 589          * use check_bucket_ref here
 590          */
 591         bkey_for_each_ptr_decode(k->k, ptrs_c, p, entry_c) {
 592                 struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev);
 593                 struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr);
 594                 enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, &entry_c->ptr);
 595
 596                 if (fsck_err_on(!g->gen_valid,
 597                                 c, ptr_to_missing_alloc_key,
 598                                 "bucket %u:%zu data type %s ptr gen %u missing in alloc btree\n"
 599                                 "while marking %s",
 600                                 p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
 601                                 bch2_data_type_str(ptr_data_type(k->k, &p.ptr)),
 602                                 p.ptr.gen,
 603                                 (printbuf_reset(&buf),
 604                                  bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) {
 605                         if (!p.ptr.cached) {
 606                                 g->gen_valid            = true;
 607                                 g->gen                  = p.ptr.gen;
 608                         } else {
 609                                 do_update = true;
 610                         }
 611                 }
 612
 613                 if (fsck_err_on(gen_cmp(p.ptr.gen, g->gen) > 0,
 614                                 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_type_str(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 (fsck_err_on(gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX,
 635                                 c, ptr_gen_newer_than_bucket_gen,
 636                                 "bucket %u:%zu gen %u data type %s: ptr gen %u too stale\n"
 637                                 "while marking %s",
 638                                 p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen,
 639                                 bch2_data_type_str(ptr_data_type(k->k, &p.ptr)),
 640                                 p.ptr.gen,
 641                                 (printbuf_reset(&buf),
 642                                  bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))
 643                         do_update = true;
 644
 645                 if (fsck_err_on(!p.ptr.cached && gen_cmp(p.ptr.gen, g->gen) < 0,
 646                                 c, stale_dirty_ptr,
 647                                 "bucket %u:%zu data type %s stale dirty ptr: %u < %u\n"
 648                                 "while marking %s",
 649                                 p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
 650                                 bch2_data_type_str(ptr_data_type(k->k, &p.ptr)),
 651                                 p.ptr.gen, g->gen,
 652                                 (printbuf_reset(&buf),
 653                                  bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))
 654                         do_update = true;
 655
 656                 if (data_type != BCH_DATA_btree && p.ptr.gen != g->gen)
 657                         continue;
 658
 659                 if (fsck_err_on(bucket_data_type(g->data_type) &&
 660                                 bucket_data_type(g->data_type) != data_type, c,
 661                                 ptr_bucket_data_type_mismatch,
 662                                 "bucket %u:%zu different types of data in same bucket: %s, %s\n"
 663                                 "while marking %s",
 664                                 p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr),
 665                                 bch2_data_type_str(g->data_type),
 666                                 bch2_data_type_str(data_type),
 667                                 (printbuf_reset(&buf),
 668                                  bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) {
 669                         if (data_type == BCH_DATA_btree) {
 670                                 g->data_type    = data_type;
 671                                 set_bit(BCH_FS_need_another_gc, &c->flags);
 672                         } else {
 673                                 do_update = true;
 674                         }
 675                 }
 676
 677                 if (p.has_ec) {
 678                         struct gc_stripe *m = genradix_ptr(&c->gc_stripes, p.ec.idx);
 679
 680                         if (fsck_err_on(!m || !m->alive, c,
 681                                         ptr_to_missing_stripe,
 682                                         "pointer to nonexistent stripe %llu\n"
 683                                         "while marking %s",
 684                                         (u64) p.ec.idx,
 685                                         (printbuf_reset(&buf),
 686                                          bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))
 687                                 do_update = true;
 688
 689                         if (fsck_err_on(m && m->alive && !bch2_ptr_matches_stripe_m(m, p), c,
 690                                         ptr_to_incorrect_stripe,
 691                                         "pointer does not match stripe %llu\n"
 692                                         "while marking %s",
 693                                         (u64) p.ec.idx,
 694                                         (printbuf_reset(&buf),
 695                                          bch2_bkey_val_to_text(&buf, c, *k), buf.buf)))
 696                                 do_update = true;
 697                 }
 698         }
 699
 700         if (do_update) {
 701                 struct bkey_ptrs ptrs;
 702                 union bch_extent_entry *entry;
 703                 struct bch_extent_ptr *ptr;
 704                 struct bkey_i *new;
 705
 706                 if (is_root) {
 707                         bch_err(c, "cannot update btree roots yet");
 708                         ret = -EINVAL;
 709                         goto err;
 710                 }
 711
 712                 new = kmalloc(bkey_bytes(k->k), GFP_KERNEL);
 713                 if (!new) {
 714                         ret = -BCH_ERR_ENOMEM_gc_repair_key;
 715                         bch_err_msg(c, ret, "allocating new key");
 716                         goto err;
 717                 }
 718
 719                 bkey_reassemble(new, *k);
 720
 721                 if (level) {
 722                         /*
 723                          * We don't want to drop btree node pointers - if the
 724                          * btree node isn't there anymore, the read path will
 725                          * sort it out:
 726                          */
 727                         ptrs = bch2_bkey_ptrs(bkey_i_to_s(new));
 728                         bkey_for_each_ptr(ptrs, ptr) {
 729                                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
 730                                 struct bucket *g = PTR_GC_BUCKET(ca, ptr);
 731
 732                                 ptr->gen = g->gen;
 733                         }
 734                 } else {
 735                         bch2_bkey_drop_ptrs(bkey_i_to_s(new), ptr, ({
 736                                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
 737                                 struct bucket *g = PTR_GC_BUCKET(ca, ptr);
 738                                 enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, ptr);
 739
 740                                 (ptr->cached &&
 741                                  (!g->gen_valid || gen_cmp(ptr->gen, g->gen) > 0)) ||
 742                                 (!ptr->cached &&
 743                                  gen_cmp(ptr->gen, g->gen) < 0) ||
 744                                 gen_cmp(g->gen, ptr->gen) > BUCKET_GC_GEN_MAX ||
 745                                 (g->data_type &&
 746                                  g->data_type != data_type);
 747                         }));
 748 again:
 749                         ptrs = bch2_bkey_ptrs(bkey_i_to_s(new));
 750                         bkey_extent_entry_for_each(ptrs, entry) {
 751                                 if (extent_entry_type(entry) == BCH_EXTENT_ENTRY_stripe_ptr) {
 752                                         struct gc_stripe *m = genradix_ptr(&c->gc_stripes,
 753                                                                         entry->stripe_ptr.idx);
 754                                         union bch_extent_entry *next_ptr;
 755
 756                                         bkey_extent_entry_for_each_from(ptrs, next_ptr, entry)
 757                                                 if (extent_entry_type(next_ptr) == BCH_EXTENT_ENTRY_ptr)
 758                                                         goto found;
 759                                         next_ptr = NULL;
 760 found:
 761                                         if (!next_ptr) {
 762                                                 bch_err(c, "aieee, found stripe ptr with no data ptr");
 763                                                 continue;
 764                                         }
 765
 766                                         if (!m || !m->alive ||
 767                                             !__bch2_ptr_matches_stripe(&m->ptrs[entry->stripe_ptr.block],
 768                                                                        &next_ptr->ptr,
 769                                                                        m->sectors)) {
 770                                                 bch2_bkey_extent_entry_drop(new, entry);
 771                                                 goto again;
 772                                         }
 773                                 }
 774                         }
 775                 }
 776
 777                 ret = bch2_journal_key_insert_take(c, btree_id, level, new);
 778                 if (ret) {
 779                         kfree(new);
 780                         goto err;
 781                 }
 782
 783                 if (level)
 784                         bch2_btree_node_update_key_early(trans, btree_id, level - 1, *k, new);
 785
 786                 if (0) {
 787                         printbuf_reset(&buf);
 788                         bch2_bkey_val_to_text(&buf, c, *k);
 789                         bch_info(c, "updated %s", buf.buf);
 790
 791                         printbuf_reset(&buf);
 792                         bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(new));
 793                         bch_info(c, "new key %s", buf.buf);
 794                 }
 795
 796                 *k = bkey_i_to_s_c(new);
 797         }
 798 err:
 799 fsck_err:
 800         printbuf_exit(&buf);
 801         return ret;
 802 }
 803
 804 /* marking of btree keys/nodes: */
 805
 806 static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id,
 807                             unsigned level, bool is_root,
 808                             struct bkey_s_c *k,
 809                             bool initial)
 810 {
 811         struct bch_fs *c = trans->c;
 812         struct bkey deleted = KEY(0, 0, 0);
 813         struct bkey_s_c old = (struct bkey_s_c) { &deleted, NULL };
 814         int ret = 0;
 815
 816         deleted.p = k->k->p;
 817
 818         if (initial) {
 819                 BUG_ON(bch2_journal_seq_verify &&
 820                        k->k->version.lo > atomic64_read(&c->journal.seq));
 821
 822                 ret = bch2_check_fix_ptrs(trans, btree_id, level, is_root, k);
 823                 if (ret)
 824                         goto err;
 825
 826                 if (fsck_err_on(k->k->version.lo > atomic64_read(&c->key_version), c,
 827                                 bkey_version_in_future,
 828                                 "key version number higher than recorded: %llu > %llu",
 829                                 k->k->version.lo,
 830                                 atomic64_read(&c->key_version)))
 831                         atomic64_set(&c->key_version, k->k->version.lo);
 832         }
 833
 834         ret = commit_do(trans, NULL, NULL, 0,
 835                         bch2_key_trigger(trans, btree_id, level, old, unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_GC));
 836 fsck_err:
 837 err:
 838         bch_err_fn(c, ret);
 839         return ret;
 840 }
 841
 842 static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool initial)
 843 {
 844         struct bch_fs *c = trans->c;
 845         struct btree_node_iter iter;
 846         struct bkey unpacked;
 847         struct bkey_s_c k;
 848         struct bkey_buf prev, cur;
 849         int ret = 0;
 850
 851         if (!btree_node_type_needs_gc(btree_node_type(b)))
 852                 return 0;
 853
 854         bch2_btree_node_iter_init_from_start(&iter, b);
 855         bch2_bkey_buf_init(&prev);
 856         bch2_bkey_buf_init(&cur);
 857         bkey_init(&prev.k->k);
 858
 859         while ((k = bch2_btree_node_iter_peek_unpack(&iter, b, &unpacked)).k) {
 860                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false,
 861                                        &k, initial);
 862                 if (ret)
 863                         break;
 864
 865                 bch2_btree_node_iter_advance(&iter, b);
 866
 867                 if (b->c.level) {
 868                         bch2_bkey_buf_reassemble(&cur, c, k);
 869
 870                         ret = bch2_gc_check_topology(c, b, &prev, cur,
 871                                         bch2_btree_node_iter_end(&iter));
 872                         if (ret)
 873                                 break;
 874                 }
 875         }
 876
 877         bch2_bkey_buf_exit(&cur, c);
 878         bch2_bkey_buf_exit(&prev, c);
 879         return ret;
 880 }
 881
 882 static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id,
 883                          bool initial, bool metadata_only)
 884 {
 885         struct bch_fs *c = trans->c;
 886         struct btree_iter iter;
 887         struct btree *b;
 888         unsigned depth = metadata_only ? 1 : 0;
 889         int ret = 0;
 890
 891         gc_pos_set(c, gc_pos_btree(btree_id, POS_MIN, 0));
 892
 893         __for_each_btree_node(trans, iter, btree_id, POS_MIN,
 894                               0, depth, BTREE_ITER_PREFETCH, b, ret) {
 895                 bch2_verify_btree_nr_keys(b);
 896
 897                 gc_pos_set(c, gc_pos_btree_node(b));
 898
 899                 ret = btree_gc_mark_node(trans, b, initial);
 900                 if (ret)
 901                         break;
 902         }
 903         bch2_trans_iter_exit(trans, &iter);
 904
 905         if (ret)
 906                 return ret;
 907
 908         mutex_lock(&c->btree_root_lock);
 909         b = bch2_btree_id_root(c, btree_id)->b;
 910         if (!btree_node_fake(b)) {
 911                 struct bkey_s_c k = bkey_i_to_s_c(&b->key);
 912
 913                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1,
 914                                        true, &k, initial);
 915         }
 916         gc_pos_set(c, gc_pos_btree_root(b->c.btree_id));
 917         mutex_unlock(&c->btree_root_lock);
 918
 919         return ret;
 920 }
 921
 922 static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b,
 923                                       unsigned target_depth)
 924 {
 925         struct bch_fs *c = trans->c;
 926         struct btree_and_journal_iter iter;
 927         struct bkey_s_c k;
 928         struct bkey_buf cur, prev;
 929         struct printbuf buf = PRINTBUF;
 930         int ret = 0;
 931
 932         bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b);
 933         bch2_bkey_buf_init(&prev);
 934         bch2_bkey_buf_init(&cur);
 935         bkey_init(&prev.k->k);
 936
 937         while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 938                 BUG_ON(bpos_lt(k.k->p, b->data->min_key));
 939                 BUG_ON(bpos_gt(k.k->p, b->data->max_key));
 940
 941                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level,
 942                                        false, &k, true);
 943                 if (ret)
 944                         goto fsck_err;
 945
 946                 if (b->c.level) {
 947                         bch2_bkey_buf_reassemble(&cur, c, k);
 948                         k = bkey_i_to_s_c(cur.k);
 949
 950                         bch2_btree_and_journal_iter_advance(&iter);
 951
 952                         ret = bch2_gc_check_topology(c, b,
 953                                         &prev, cur,
 954                                         !bch2_btree_and_journal_iter_peek(&iter).k);
 955                         if (ret)
 956                                 goto fsck_err;
 957                 } else {
 958                         bch2_btree_and_journal_iter_advance(&iter);
 959                 }
 960         }
 961
 962         if (b->c.level > target_depth) {
 963                 bch2_btree_and_journal_iter_exit(&iter);
 964                 bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b);
 965                 iter.prefetch = true;
 966
 967                 while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) {
 968                         struct btree *child;
 969
 970                         bch2_bkey_buf_reassemble(&cur, c, k);
 971                         bch2_btree_and_journal_iter_advance(&iter);
 972
 973                         child = bch2_btree_node_get_noiter(trans, cur.k,
 974                                                 b->c.btree_id, b->c.level - 1,
 975                                                 false);
 976                         ret = PTR_ERR_OR_ZERO(child);
 977
 978                         if (bch2_err_matches(ret, EIO)) {
 979                                 bch2_topology_error(c);
 980
 981                                 if (__fsck_err(c,
 982                                           FSCK_CAN_FIX|
 983                                           FSCK_CAN_IGNORE|
 984                                           FSCK_NO_RATELIMIT,
 985                                           btree_node_read_error,
 986                                           "Unreadable btree node at btree %s level %u:\n"
 987                                           "  %s",
 988                                           bch2_btree_id_str(b->c.btree_id),
 989                                           b->c.level - 1,
 990                                           (printbuf_reset(&buf),
 991                                            bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(cur.k)), buf.buf)) &&
 992                                     should_restart_for_topology_repair(c)) {
 993                                         bch_info(c, "Halting mark and sweep to start topology repair pass");
 994                                         ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology);
 995                                         goto fsck_err;
 996                                 } else {
 997                                         /* Continue marking when opted to not
 998                                          * fix the error: */
 999                                         ret = 0;
1000                                         set_bit(BCH_FS_initial_gc_unfixed, &c->flags);
1001                                         continue;
1002                                 }
1003                         } else if (ret) {
1004                                 bch_err_msg(c, ret, "getting btree node");
1005                                 break;
1006                         }
1007
1008                         ret = bch2_gc_btree_init_recurse(trans, child,
1009                                                          target_depth);
1010                         six_unlock_read(&child->c.lock);
1011
1012                         if (ret)
1013                                 break;
1014                 }
1015         }
1016 fsck_err:
1017         bch2_bkey_buf_exit(&cur, c);
1018         bch2_bkey_buf_exit(&prev, c);
1019         bch2_btree_and_journal_iter_exit(&iter);
1020         printbuf_exit(&buf);
1021         return ret;
1022 }
1023
1024 static int bch2_gc_btree_init(struct btree_trans *trans,
1025                               enum btree_id btree_id,
1026                               bool metadata_only)
1027 {
1028         struct bch_fs *c = trans->c;
1029         struct btree *b;
1030         unsigned target_depth = metadata_only ? 1 : 0;
1031         struct printbuf buf = PRINTBUF;
1032         int ret = 0;
1033
1034         b = bch2_btree_id_root(c, btree_id)->b;
1035
1036         if (btree_node_fake(b))
1037                 return 0;
1038
1039         six_lock_read(&b->c.lock, NULL, NULL);
1040         printbuf_reset(&buf);
1041         bch2_bpos_to_text(&buf, b->data->min_key);
1042         if (mustfix_fsck_err_on(!bpos_eq(b->data->min_key, POS_MIN), c,
1043                                 btree_root_bad_min_key,
1044                         "btree root with incorrect min_key: %s", buf.buf)) {
1045                 bch_err(c, "repair unimplemented");
1046                 ret = -BCH_ERR_fsck_repair_unimplemented;
1047                 goto fsck_err;
1048         }
1049
1050         printbuf_reset(&buf);
1051         bch2_bpos_to_text(&buf, b->data->max_key);
1052         if (mustfix_fsck_err_on(!bpos_eq(b->data->max_key, SPOS_MAX), c,
1053                                 btree_root_bad_max_key,
1054                         "btree root with incorrect max_key: %s", buf.buf)) {
1055                 bch_err(c, "repair unimplemented");
1056                 ret = -BCH_ERR_fsck_repair_unimplemented;
1057                 goto fsck_err;
1058         }
1059
1060         if (b->c.level >= target_depth)
1061                 ret = bch2_gc_btree_init_recurse(trans, b, target_depth);
1062
1063         if (!ret) {
1064                 struct bkey_s_c k = bkey_i_to_s_c(&b->key);
1065
1066                 ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, true,
1067                                        &k, true);
1068         }
1069 fsck_err:
1070         six_unlock_read(&b->c.lock);
1071
1072         bch_err_fn(c, ret);
1073         printbuf_exit(&buf);
1074         return ret;
1075 }
1076
1077 static inline int btree_id_gc_phase_cmp(enum btree_id l, enum btree_id r)
1078 {
1079         return  (int) btree_id_to_gc_phase(l) -
1080                 (int) btree_id_to_gc_phase(r);
1081 }
1082
1083 static int bch2_gc_btrees(struct bch_fs *c, bool initial, bool metadata_only)
1084 {
1085         struct btree_trans *trans = bch2_trans_get(c);
1086         enum btree_id ids[BTREE_ID_NR];
1087         unsigned i;
1088         int ret = 0;
1089
1090         for (i = 0; i < BTREE_ID_NR; i++)
1091                 ids[i] = i;
1092         bubble_sort(ids, BTREE_ID_NR, btree_id_gc_phase_cmp);
1093
1094         for (i = 0; i < BTREE_ID_NR && !ret; i++)
1095                 ret = initial
1096                         ? bch2_gc_btree_init(trans, ids[i], metadata_only)
1097                         : bch2_gc_btree(trans, ids[i], initial, metadata_only);
1098
1099         for (i = BTREE_ID_NR; i < btree_id_nr_alive(c) && !ret; i++) {
1100                 if (!bch2_btree_id_root(c, i)->alive)
1101                         continue;
1102
1103                 ret = initial
1104                         ? bch2_gc_btree_init(trans, i, metadata_only)
1105                         : bch2_gc_btree(trans, i, initial, metadata_only);
1106         }
1107
1108         bch2_trans_put(trans);
1109         bch_err_fn(c, ret);
1110         return ret;
1111 }
1112
1113 static void mark_metadata_sectors(struct bch_fs *c, struct bch_dev *ca,
1114                                   u64 start, u64 end,
1115                                   enum bch_data_type type,
1116                                   unsigned flags)
1117 {
1118         u64 b = sector_to_bucket(ca, start);
1119
1120         do {
1121                 unsigned sectors =
1122                         min_t(u64, bucket_to_sector(ca, b + 1), end) - start;
1123
1124                 bch2_mark_metadata_bucket(c, ca, b, type, sectors,
1125                                           gc_phase(GC_PHASE_SB), flags);
1126                 b++;
1127                 start += sectors;
1128         } while (start < end);
1129 }
1130
1131 static void bch2_mark_dev_superblock(struct bch_fs *c, struct bch_dev *ca,
1132                                      unsigned flags)
1133 {
1134         struct bch_sb_layout *layout = &ca->disk_sb.sb->layout;
1135         unsigned i;
1136         u64 b;
1137
1138         for (i = 0; i < layout->nr_superblocks; i++) {
1139                 u64 offset = le64_to_cpu(layout->sb_offset[i]);
1140
1141                 if (offset == BCH_SB_SECTOR)
1142                         mark_metadata_sectors(c, ca, 0, BCH_SB_SECTOR,
1143                                               BCH_DATA_sb, flags);
1144
1145                 mark_metadata_sectors(c, ca, offset,
1146                                       offset + (1 << layout->sb_max_size_bits),
1147                                       BCH_DATA_sb, flags);
1148         }
1149
1150         for (i = 0; i < ca->journal.nr; i++) {
1151                 b = ca->journal.buckets[i];
1152                 bch2_mark_metadata_bucket(c, ca, b, BCH_DATA_journal,
1153                                           ca->mi.bucket_size,
1154                                           gc_phase(GC_PHASE_SB), flags);
1155         }
1156 }
1157
1158 static void bch2_mark_superblocks(struct bch_fs *c)
1159 {
1160         mutex_lock(&c->sb_lock);
1161         gc_pos_set(c, gc_phase(GC_PHASE_SB));
1162
1163         for_each_online_member(c, ca)
1164                 bch2_mark_dev_superblock(c, ca, BTREE_TRIGGER_GC);
1165         mutex_unlock(&c->sb_lock);
1166 }
1167
1168 #if 0
1169 /* Also see bch2_pending_btree_node_free_insert_done() */
1170 static void bch2_mark_pending_btree_node_frees(struct bch_fs *c)
1171 {
1172         struct btree_update *as;
1173         struct pending_btree_node_free *d;
1174
1175         mutex_lock(&c->btree_interior_update_lock);
1176         gc_pos_set(c, gc_phase(GC_PHASE_PENDING_DELETE));
1177
1178         for_each_pending_btree_node_free(c, as, d)
1179                 if (d->index_update_done)
1180                         bch2_mark_key(c, bkey_i_to_s_c(&d->key), BTREE_TRIGGER_GC);
1181
1182         mutex_unlock(&c->btree_interior_update_lock);
1183 }
1184 #endif
1185
1186 static void bch2_gc_free(struct bch_fs *c)
1187 {
1188         genradix_free(&c->reflink_gc_table);
1189         genradix_free(&c->gc_stripes);
1190
1191         for_each_member_device(c, ca) {
1192                 kvfree(rcu_dereference_protected(ca->buckets_gc, 1));
1193                 ca->buckets_gc = NULL;
1194
1195                 free_percpu(ca->usage_gc);
1196                 ca->usage_gc = NULL;
1197         }
1198
1199         free_percpu(c->usage_gc);
1200         c->usage_gc = NULL;
1201 }
1202
1203 static int bch2_gc_done(struct bch_fs *c,
1204                         bool initial, bool metadata_only)
1205 {
1206         struct bch_dev *ca = NULL;
1207         struct printbuf buf = PRINTBUF;
1208         bool verify = !metadata_only &&
1209                 !c->opts.reconstruct_alloc &&
1210                 (!initial || (c->sb.compat & (1ULL << BCH_COMPAT_alloc_info)));
1211         unsigned i;
1212         int ret = 0;
1213
1214         percpu_down_write(&c->mark_lock);
1215
1216 #define copy_field(_err, _f, _msg, ...)                                 \
1217         if (dst->_f != src->_f &&                                       \
1218             (!verify ||                                                 \
1219              fsck_err(c, _err, _msg ": got %llu, should be %llu"        \
1220                       , ##__VA_ARGS__, dst->_f, src->_f)))              \
1221                 dst->_f = src->_f
1222 #define copy_dev_field(_err, _f, _msg, ...)                             \
1223         copy_field(_err, _f, "dev %u has wrong " _msg, ca->dev_idx, ##__VA_ARGS__)
1224 #define copy_fs_field(_err, _f, _msg, ...)                              \
1225         copy_field(_err, _f, "fs has wrong " _msg, ##__VA_ARGS__)
1226
1227         for (i = 0; i < ARRAY_SIZE(c->usage); i++)
1228                 bch2_fs_usage_acc_to_base(c, i);
1229
1230         __for_each_member_device(c, ca) {
1231                 struct bch_dev_usage *dst = ca->usage_base;
1232                 struct bch_dev_usage *src = (void *)
1233                         bch2_acc_percpu_u64s((u64 __percpu *) ca->usage_gc,
1234                                              dev_usage_u64s());
1235
1236                 for (i = 0; i < BCH_DATA_NR; i++) {
1237                         copy_dev_field(dev_usage_buckets_wrong,
1238                                        d[i].buckets,    "%s buckets", bch2_data_type_str(i));
1239                         copy_dev_field(dev_usage_sectors_wrong,
1240                                        d[i].sectors,    "%s sectors", bch2_data_type_str(i));
1241                         copy_dev_field(dev_usage_fragmented_wrong,
1242                                        d[i].fragmented, "%s fragmented", bch2_data_type_str(i));
1243                 }
1244         }
1245
1246         {
1247                 unsigned nr = fs_usage_u64s(c);
1248                 struct bch_fs_usage *dst = c->usage_base;
1249                 struct bch_fs_usage *src = (void *)
1250                         bch2_acc_percpu_u64s((u64 __percpu *) c->usage_gc, nr);
1251
1252                 copy_fs_field(fs_usage_hidden_wrong,
1253                               b.hidden,         "hidden");
1254                 copy_fs_field(fs_usage_btree_wrong,
1255                               b.btree,          "btree");
1256
1257                 if (!metadata_only) {
1258                         copy_fs_field(fs_usage_data_wrong,
1259                                       b.data,   "data");
1260                         copy_fs_field(fs_usage_cached_wrong,
1261                                       b.cached, "cached");
1262                         copy_fs_field(fs_usage_reserved_wrong,
1263                                       b.reserved,       "reserved");
1264                         copy_fs_field(fs_usage_nr_inodes_wrong,
1265                                       b.nr_inodes,"nr_inodes");
1266
1267                         for (i = 0; i < BCH_REPLICAS_MAX; i++)
1268                                 copy_fs_field(fs_usage_persistent_reserved_wrong,
1269                                               persistent_reserved[i],
1270                                               "persistent_reserved[%i]", i);
1271                 }
1272
1273                 for (i = 0; i < c->replicas.nr; i++) {
1274                         struct bch_replicas_entry_v1 *e =
1275                                 cpu_replicas_entry(&c->replicas, i);
1276
1277                         if (metadata_only &&
1278                             (e->data_type == BCH_DATA_user ||
1279                              e->data_type == BCH_DATA_cached))
1280                                 continue;
1281
1282                         printbuf_reset(&buf);
1283                         bch2_replicas_entry_to_text(&buf, e);
1284
1285                         copy_fs_field(fs_usage_replicas_wrong,
1286                                       replicas[i], "%s", buf.buf);
1287                 }
1288         }
1289
1290 #undef copy_fs_field
1291 #undef copy_dev_field
1292 #undef copy_stripe_field
1293 #undef copy_field
1294 fsck_err:
1295         if (ca)
1296                 percpu_ref_put(&ca->ref);
1297         bch_err_fn(c, ret);
1298
1299         percpu_up_write(&c->mark_lock);
1300         printbuf_exit(&buf);
1301         return ret;
1302 }
1303
1304 static int bch2_gc_start(struct bch_fs *c)
1305 {
1306         BUG_ON(c->usage_gc);
1307
1308         c->usage_gc = __alloc_percpu_gfp(fs_usage_u64s(c) * sizeof(u64),
1309                                          sizeof(u64), GFP_KERNEL);
1310         if (!c->usage_gc) {
1311                 bch_err(c, "error allocating c->usage_gc");
1312                 return -BCH_ERR_ENOMEM_gc_start;
1313         }
1314
1315         for_each_member_device(c, ca) {
1316                 BUG_ON(ca->usage_gc);
1317
1318                 ca->usage_gc = alloc_percpu(struct bch_dev_usage);
1319                 if (!ca->usage_gc) {
1320                         bch_err(c, "error allocating ca->usage_gc");
1321                         percpu_ref_put(&ca->ref);
1322                         return -BCH_ERR_ENOMEM_gc_start;
1323                 }
1324
1325                 this_cpu_write(ca->usage_gc->d[BCH_DATA_free].buckets,
1326                                ca->mi.nbuckets - ca->mi.first_bucket);
1327         }
1328
1329         return 0;
1330 }
1331
1332 static int bch2_gc_reset(struct bch_fs *c)
1333 {
1334         for_each_member_device(c, ca) {
1335                 free_percpu(ca->usage_gc);
1336                 ca->usage_gc = NULL;
1337         }
1338
1339         free_percpu(c->usage_gc);
1340         c->usage_gc = NULL;
1341
1342         return bch2_gc_start(c);
1343 }
1344
1345 /* returns true if not equal */
1346 static inline bool bch2_alloc_v4_cmp(struct bch_alloc_v4 l,
1347                                      struct bch_alloc_v4 r)
1348 {
1349         return  l.gen != r.gen                          ||
1350                 l.oldest_gen != r.oldest_gen            ||
1351                 l.data_type != r.data_type              ||
1352                 l.dirty_sectors != r.dirty_sectors      ||
1353                 l.cached_sectors != r.cached_sectors     ||
1354                 l.stripe_redundancy != r.stripe_redundancy ||
1355                 l.stripe != r.stripe;
1356 }
1357
1358 static int bch2_alloc_write_key(struct btree_trans *trans,
1359                                 struct btree_iter *iter,
1360                                 struct bkey_s_c k,
1361                                 bool metadata_only)
1362 {
1363         struct bch_fs *c = trans->c;
1364         struct bch_dev *ca = bch_dev_bkey_exists(c, iter->pos.inode);
1365         struct bucket old_gc, gc, *b;
1366         struct bkey_i_alloc_v4 *a;
1367         struct bch_alloc_v4 old_convert, new;
1368         const struct bch_alloc_v4 *old;
1369         int ret;
1370
1371         old = bch2_alloc_to_v4(k, &old_convert);
1372         new = *old;
1373
1374         percpu_down_read(&c->mark_lock);
1375         b = gc_bucket(ca, iter->pos.offset);
1376         old_gc = *b;
1377
1378         if ((old->data_type == BCH_DATA_sb ||
1379              old->data_type == BCH_DATA_journal) &&
1380             !bch2_dev_is_online(ca)) {
1381                 b->data_type = old->data_type;
1382                 b->dirty_sectors = old->dirty_sectors;
1383         }
1384
1385         /*
1386          * b->data_type doesn't yet include need_discard & need_gc_gen states -
1387          * fix that here:
1388          */
1389         b->data_type = __alloc_data_type(b->dirty_sectors,
1390                                          b->cached_sectors,
1391                                          b->stripe,
1392                                          *old,
1393                                          b->data_type);
1394         gc = *b;
1395         percpu_up_read(&c->mark_lock);
1396
1397         if (gc.data_type != old_gc.data_type ||
1398             gc.dirty_sectors != old_gc.dirty_sectors)
1399                 bch2_dev_usage_update_m(c, ca, &old_gc, &gc);
1400
1401         if (metadata_only &&
1402             gc.data_type != BCH_DATA_sb &&
1403             gc.data_type != BCH_DATA_journal &&
1404             gc.data_type != BCH_DATA_btree)
1405                 return 0;
1406
1407         if (gen_after(old->gen, gc.gen))
1408                 return 0;
1409
1410         if (fsck_err_on(new.data_type != gc.data_type, c,
1411                         alloc_key_data_type_wrong,
1412                         "bucket %llu:%llu gen %u has wrong data_type"
1413                         ": got %s, should be %s",
1414                         iter->pos.inode, iter->pos.offset,
1415                         gc.gen,
1416                         bch2_data_type_str(new.data_type),
1417                         bch2_data_type_str(gc.data_type)))
1418                 new.data_type = gc.data_type;
1419
1420 #define copy_bucket_field(_errtype, _f)                                 \
1421         if (fsck_err_on(new._f != gc._f, c, _errtype,                   \
1422                         "bucket %llu:%llu gen %u data type %s has wrong " #_f   \
1423                         ": got %u, should be %u",                       \
1424                         iter->pos.inode, iter->pos.offset,              \
1425                         gc.gen,                                         \
1426                         bch2_data_type_str(gc.data_type),               \
1427                         new._f, gc._f))                                 \
1428                 new._f = gc._f;                                         \
1429
1430         copy_bucket_field(alloc_key_gen_wrong,
1431                           gen);
1432         copy_bucket_field(alloc_key_dirty_sectors_wrong,
1433                           dirty_sectors);
1434         copy_bucket_field(alloc_key_cached_sectors_wrong,
1435                           cached_sectors);
1436         copy_bucket_field(alloc_key_stripe_wrong,
1437                           stripe);
1438         copy_bucket_field(alloc_key_stripe_redundancy_wrong,
1439                           stripe_redundancy);
1440 #undef copy_bucket_field
1441
1442         if (!bch2_alloc_v4_cmp(*old, new))
1443                 return 0;
1444
1445         a = bch2_alloc_to_v4_mut(trans, k);
1446         ret = PTR_ERR_OR_ZERO(a);
1447         if (ret)
1448                 return ret;
1449
1450         a->v = new;
1451
1452         /*
1453          * The trigger normally makes sure this is set, but we're not running
1454          * triggers:
1455          */
1456         if (a->v.data_type == BCH_DATA_cached && !a->v.io_time[READ])
1457                 a->v.io_time[READ] = max_t(u64, 1, atomic64_read(&c->io_clock[READ].now));
1458
1459         ret = bch2_trans_update(trans, iter, &a->k_i, BTREE_TRIGGER_NORUN);
1460 fsck_err:
1461         return ret;
1462 }
1463
1464 static int bch2_gc_alloc_done(struct bch_fs *c, bool metadata_only)
1465 {
1466         int ret = 0;
1467
1468         for_each_member_device(c, ca) {
1469                 ret = bch2_trans_run(c,
1470                         for_each_btree_key_upto_commit(trans, iter, BTREE_ID_alloc,
1471                                         POS(ca->dev_idx, ca->mi.first_bucket),
1472                                         POS(ca->dev_idx, ca->mi.nbuckets - 1),
1473                                         BTREE_ITER_SLOTS|BTREE_ITER_PREFETCH, k,
1474                                         NULL, NULL, BCH_TRANS_COMMIT_lazy_rw,
1475                                 bch2_alloc_write_key(trans, &iter, k, metadata_only)));
1476                 if (ret) {
1477                         percpu_ref_put(&ca->ref);
1478                         break;
1479                 }
1480         }
1481
1482         bch_err_fn(c, ret);
1483         return ret;
1484 }
1485
1486 static int bch2_gc_alloc_start(struct bch_fs *c, bool metadata_only)
1487 {
1488         for_each_member_device(c, ca) {
1489                 struct bucket_array *buckets = kvmalloc(sizeof(struct bucket_array) +
1490                                 ca->mi.nbuckets * sizeof(struct bucket),
1491                                 GFP_KERNEL|__GFP_ZERO);
1492                 if (!buckets) {
1493                         percpu_ref_put(&ca->ref);
1494                         bch_err(c, "error allocating ca->buckets[gc]");
1495                         return -BCH_ERR_ENOMEM_gc_alloc_start;
1496                 }
1497
1498                 buckets->first_bucket   = ca->mi.first_bucket;
1499                 buckets->nbuckets       = ca->mi.nbuckets;
1500                 rcu_assign_pointer(ca->buckets_gc, buckets);
1501         }
1502
1503         int ret = bch2_trans_run(c,
1504                 for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN,
1505                                          BTREE_ITER_PREFETCH, k, ({
1506                         struct bch_dev *ca = bch_dev_bkey_exists(c, k.k->p.inode);
1507                         struct bucket *g = gc_bucket(ca, k.k->p.offset);
1508
1509                         struct bch_alloc_v4 a_convert;
1510                         const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert);
1511
1512                         g->gen_valid    = 1;
1513                         g->gen          = a->gen;
1514
1515                         if (metadata_only &&
1516                             (a->data_type == BCH_DATA_user ||
1517                              a->data_type == BCH_DATA_cached ||
1518                              a->data_type == BCH_DATA_parity)) {
1519                                 g->data_type            = a->data_type;
1520                                 g->dirty_sectors        = a->dirty_sectors;
1521                                 g->cached_sectors       = a->cached_sectors;
1522                                 g->stripe               = a->stripe;
1523                                 g->stripe_redundancy    = a->stripe_redundancy;
1524                         }
1525
1526                         0;
1527                 })));
1528         bch_err_fn(c, ret);
1529         return ret;
1530 }
1531
1532 static void bch2_gc_alloc_reset(struct bch_fs *c, bool metadata_only)
1533 {
1534         for_each_member_device(c, ca) {
1535                 struct bucket_array *buckets = gc_bucket_array(ca);
1536                 struct bucket *g;
1537
1538                 for_each_bucket(g, buckets) {
1539                         if (metadata_only &&
1540                             (g->data_type == BCH_DATA_user ||
1541                              g->data_type == BCH_DATA_cached ||
1542                              g->data_type == BCH_DATA_parity))
1543                                 continue;
1544                         g->data_type = 0;
1545                         g->dirty_sectors = 0;
1546                         g->cached_sectors = 0;
1547                 }
1548         }
1549 }
1550
1551 static int bch2_gc_write_reflink_key(struct btree_trans *trans,
1552                                      struct btree_iter *iter,
1553                                      struct bkey_s_c k,
1554                                      size_t *idx)
1555 {
1556         struct bch_fs *c = trans->c;
1557         const __le64 *refcount = bkey_refcount_c(k);
1558         struct printbuf buf = PRINTBUF;
1559         struct reflink_gc *r;
1560         int ret = 0;
1561
1562         if (!refcount)
1563                 return 0;
1564
1565         while ((r = genradix_ptr(&c->reflink_gc_table, *idx)) &&
1566                r->offset < k.k->p.offset)
1567                 ++*idx;
1568
1569         if (!r ||
1570             r->offset != k.k->p.offset ||
1571             r->size != k.k->size) {
1572                 bch_err(c, "unexpected inconsistency walking reflink table at gc finish");
1573                 return -EINVAL;
1574         }
1575
1576         if (fsck_err_on(r->refcount != le64_to_cpu(*refcount), c,
1577                         reflink_v_refcount_wrong,
1578                         "reflink key has wrong refcount:\n"
1579                         "  %s\n"
1580                         "  should be %u",
1581                         (bch2_bkey_val_to_text(&buf, c, k), buf.buf),
1582                         r->refcount)) {
1583                 struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, k);
1584                 ret = PTR_ERR_OR_ZERO(new);
1585                 if (ret)
1586                         return ret;
1587
1588                 if (!r->refcount)
1589                         new->k.type = KEY_TYPE_deleted;
1590                 else
1591                         *bkey_refcount(bkey_i_to_s(new)) = cpu_to_le64(r->refcount);
1592                 ret = bch2_trans_update(trans, iter, new, 0);
1593         }
1594 fsck_err:
1595         printbuf_exit(&buf);
1596         return ret;
1597 }
1598
1599 static int bch2_gc_reflink_done(struct bch_fs *c, bool metadata_only)
1600 {
1601         size_t idx = 0;
1602
1603         if (metadata_only)
1604                 return 0;
1605
1606         int ret = bch2_trans_run(c,
1607                 for_each_btree_key_commit(trans, iter,
1608                                 BTREE_ID_reflink, POS_MIN,
1609                                 BTREE_ITER_PREFETCH, k,
1610                                 NULL, NULL, BCH_TRANS_COMMIT_no_enospc,
1611                         bch2_gc_write_reflink_key(trans, &iter, k, &idx)));
1612         c->reflink_gc_nr = 0;
1613         return ret;
1614 }
1615
1616 static int bch2_gc_reflink_start(struct bch_fs *c,
1617                                  bool metadata_only)
1618 {
1619
1620         if (metadata_only)
1621                 return 0;
1622
1623         c->reflink_gc_nr = 0;
1624
1625         int ret = bch2_trans_run(c,
1626                 for_each_btree_key(trans, iter, BTREE_ID_reflink, POS_MIN,
1627                                    BTREE_ITER_PREFETCH, k, ({
1628                         const __le64 *refcount = bkey_refcount_c(k);
1629
1630                         if (!refcount)
1631                                 continue;
1632
1633                         struct reflink_gc *r = genradix_ptr_alloc(&c->reflink_gc_table,
1634                                                         c->reflink_gc_nr++, GFP_KERNEL);
1635                         if (!r) {
1636                                 ret = -BCH_ERR_ENOMEM_gc_reflink_start;
1637                                 break;
1638                         }
1639
1640                         r->offset       = k.k->p.offset;
1641                         r->size         = k.k->size;
1642                         r->refcount     = 0;
1643                         0;
1644                 })));
1645
1646         bch_err_fn(c, ret);
1647         return ret;
1648 }
1649
1650 static void bch2_gc_reflink_reset(struct bch_fs *c, bool metadata_only)
1651 {
1652         struct genradix_iter iter;
1653         struct reflink_gc *r;
1654
1655         genradix_for_each(&c->reflink_gc_table, iter, r)
1656                 r->refcount = 0;
1657 }
1658
1659 static int bch2_gc_write_stripes_key(struct btree_trans *trans,
1660                                      struct btree_iter *iter,
1661                                      struct bkey_s_c k)
1662 {
1663         struct bch_fs *c = trans->c;
1664         struct printbuf buf = PRINTBUF;
1665         const struct bch_stripe *s;
1666         struct gc_stripe *m;
1667         bool bad = false;
1668         unsigned i;
1669         int ret = 0;
1670
1671         if (k.k->type != KEY_TYPE_stripe)
1672                 return 0;
1673
1674         s = bkey_s_c_to_stripe(k).v;
1675         m = genradix_ptr(&c->gc_stripes, k.k->p.offset);
1676
1677         for (i = 0; i < s->nr_blocks; i++) {
1678                 u32 old = stripe_blockcount_get(s, i);
1679                 u32 new = (m ? m->block_sectors[i] : 0);
1680
1681                 if (old != new) {
1682                         prt_printf(&buf, "stripe block %u has wrong sector count: got %u, should be %u\n",
1683                                    i, old, new);
1684                         bad = true;
1685                 }
1686         }
1687
1688         if (bad)
1689                 bch2_bkey_val_to_text(&buf, c, k);
1690
1691         if (fsck_err_on(bad, c, stripe_sector_count_wrong,
1692                         "%s", buf.buf)) {
1693                 struct bkey_i_stripe *new;
1694
1695                 new = bch2_trans_kmalloc(trans, bkey_bytes(k.k));
1696                 ret = PTR_ERR_OR_ZERO(new);
1697                 if (ret)
1698                         return ret;
1699
1700                 bkey_reassemble(&new->k_i, k);
1701
1702                 for (i = 0; i < new->v.nr_blocks; i++)
1703                         stripe_blockcount_set(&new->v, i, m ? m->block_sectors[i] : 0);
1704
1705                 ret = bch2_trans_update(trans, iter, &new->k_i, 0);
1706         }
1707 fsck_err:
1708         printbuf_exit(&buf);
1709         return ret;
1710 }
1711
1712 static int bch2_gc_stripes_done(struct bch_fs *c, bool metadata_only)
1713 {
1714         if (metadata_only)
1715                 return 0;
1716
1717         return bch2_trans_run(c,
1718                 for_each_btree_key_commit(trans, iter,
1719                                 BTREE_ID_stripes, POS_MIN,
1720                                 BTREE_ITER_PREFETCH, k,
1721                                 NULL, NULL, BCH_TRANS_COMMIT_no_enospc,
1722                         bch2_gc_write_stripes_key(trans, &iter, k)));
1723 }
1724
1725 static void bch2_gc_stripes_reset(struct bch_fs *c, bool metadata_only)
1726 {
1727         genradix_free(&c->gc_stripes);
1728 }
1729
1730 /**
1731  * bch2_gc - walk _all_ references to buckets, and recompute them:
1732  *
1733  * @c:                  filesystem object
1734  * @initial:            are we in recovery?
1735  * @metadata_only:      are we just checking metadata references, or everything?
1736  *
1737  * Returns: 0 on success, or standard errcode on failure
1738  *
1739  * Order matters here:
1740  *  - Concurrent GC relies on the fact that we have a total ordering for
1741  *    everything that GC walks - see  gc_will_visit_node(),
1742  *    gc_will_visit_root()
1743  *
1744  *  - also, references move around in the course of index updates and
1745  *    various other crap: everything needs to agree on the ordering
1746  *    references are allowed to move around in - e.g., we're allowed to
1747  *    start with a reference owned by an open_bucket (the allocator) and
1748  *    move it to the btree, but not the reverse.
1749  *
1750  *    This is necessary to ensure that gc doesn't miss references that
1751  *    move around - if references move backwards in the ordering GC
1752  *    uses, GC could skip past them
1753  */
1754 int bch2_gc(struct bch_fs *c, bool initial, bool metadata_only)
1755 {
1756         unsigned iter = 0;
1757         int ret;
1758
1759         lockdep_assert_held(&c->state_lock);
1760
1761         down_write(&c->gc_lock);
1762
1763         bch2_btree_interior_updates_flush(c);
1764
1765         ret   = bch2_gc_start(c) ?:
1766                 bch2_gc_alloc_start(c, metadata_only) ?:
1767                 bch2_gc_reflink_start(c, metadata_only);
1768         if (ret)
1769                 goto out;
1770 again:
1771         gc_pos_set(c, gc_phase(GC_PHASE_START));
1772
1773         bch2_mark_superblocks(c);
1774
1775         ret = bch2_gc_btrees(c, initial, metadata_only);
1776
1777         if (ret)
1778                 goto out;
1779
1780 #if 0
1781         bch2_mark_pending_btree_node_frees(c);
1782 #endif
1783         c->gc_count++;
1784
1785         if (test_bit(BCH_FS_need_another_gc, &c->flags) ||
1786             (!iter && bch2_test_restart_gc)) {
1787                 if (iter++ > 2) {
1788                         bch_info(c, "Unable to fix bucket gens, looping");
1789                         ret = -EINVAL;
1790                         goto out;
1791                 }
1792
1793                 /*
1794                  * XXX: make sure gens we fixed got saved
1795                  */
1796                 bch_info(c, "Second GC pass needed, restarting:");
1797                 clear_bit(BCH_FS_need_another_gc, &c->flags);
1798                 __gc_pos_set(c, gc_phase(GC_PHASE_NOT_RUNNING));
1799
1800                 bch2_gc_stripes_reset(c, metadata_only);
1801                 bch2_gc_alloc_reset(c, metadata_only);
1802                 bch2_gc_reflink_reset(c, metadata_only);
1803                 ret = bch2_gc_reset(c);
1804                 if (ret)
1805                         goto out;
1806
1807                 /* flush fsck errors, reset counters */
1808                 bch2_flush_fsck_errs(c);
1809                 goto again;
1810         }
1811 out:
1812         if (!ret) {
1813                 bch2_journal_block(&c->journal);
1814
1815                 ret   = bch2_gc_alloc_done(c, metadata_only) ?:
1816                         bch2_gc_done(c, initial, metadata_only) ?:
1817                         bch2_gc_stripes_done(c, metadata_only) ?:
1818                         bch2_gc_reflink_done(c, metadata_only);
1819
1820                 bch2_journal_unblock(&c->journal);
1821         }
1822
1823         percpu_down_write(&c->mark_lock);
1824         /* Indicates that gc is no longer in progress: */
1825         __gc_pos_set(c, gc_phase(GC_PHASE_NOT_RUNNING));
1826
1827         bch2_gc_free(c);
1828         percpu_up_write(&c->mark_lock);
1829
1830         up_write(&c->gc_lock);
1831
1832         /*
1833          * At startup, allocations can happen directly instead of via the
1834          * allocator thread - issue wakeup in case they blocked on gc_lock:
1835          */
1836         closure_wake_up(&c->freelist_wait);
1837         bch_err_fn(c, ret);
1838         return ret;
1839 }
1840
1841 static int gc_btree_gens_key(struct btree_trans *trans,
1842                              struct btree_iter *iter,
1843                              struct bkey_s_c k)
1844 {
1845         struct bch_fs *c = trans->c;
1846         struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
1847         struct bkey_i *u;
1848         int ret;
1849
1850         percpu_down_read(&c->mark_lock);
1851         bkey_for_each_ptr(ptrs, ptr) {
1852                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
1853
1854                 if (ptr_stale(ca, ptr) > 16) {
1855                         percpu_up_read(&c->mark_lock);
1856                         goto update;
1857                 }
1858         }
1859
1860         bkey_for_each_ptr(ptrs, ptr) {
1861                 struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
1862                 u8 *gen = &ca->oldest_gen[PTR_BUCKET_NR(ca, ptr)];
1863
1864                 if (gen_after(*gen, ptr->gen))
1865                         *gen = ptr->gen;
1866         }
1867         percpu_up_read(&c->mark_lock);
1868         return 0;
1869 update:
1870         u = bch2_bkey_make_mut(trans, iter, &k, 0);
1871         ret = PTR_ERR_OR_ZERO(u);
1872         if (ret)
1873                 return ret;
1874
1875         bch2_extent_normalize(c, bkey_i_to_s(u));
1876         return 0;
1877 }
1878
1879 static int bch2_alloc_write_oldest_gen(struct btree_trans *trans, struct btree_iter *iter,
1880                                        struct bkey_s_c k)
1881 {
1882         struct bch_dev *ca = bch_dev_bkey_exists(trans->c, iter->pos.inode);
1883         struct bch_alloc_v4 a_convert;
1884         const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert);
1885         struct bkey_i_alloc_v4 *a_mut;
1886         int ret;
1887
1888         if (a->oldest_gen == ca->oldest_gen[iter->pos.offset])
1889                 return 0;
1890
1891         a_mut = bch2_alloc_to_v4_mut(trans, k);
1892         ret = PTR_ERR_OR_ZERO(a_mut);
1893         if (ret)
1894                 return ret;
1895
1896         a_mut->v.oldest_gen = ca->oldest_gen[iter->pos.offset];
1897         a_mut->v.data_type = alloc_data_type(a_mut->v, a_mut->v.data_type);
1898
1899         return bch2_trans_update(trans, iter, &a_mut->k_i, 0);
1900 }
1901
1902 int bch2_gc_gens(struct bch_fs *c)
1903 {
1904         u64 b, start_time = local_clock();
1905         int ret;
1906
1907         /*
1908          * Ideally we would be using state_lock and not gc_lock here, but that
1909          * introduces a deadlock in the RO path - we currently take the state
1910          * lock at the start of going RO, thus the gc thread may get stuck:
1911          */
1912         if (!mutex_trylock(&c->gc_gens_lock))
1913                 return 0;
1914
1915         trace_and_count(c, gc_gens_start, c);
1916         down_read(&c->gc_lock);
1917
1918         for_each_member_device(c, ca) {
1919                 struct bucket_gens *gens = bucket_gens(ca);
1920
1921                 BUG_ON(ca->oldest_gen);
1922
1923                 ca->oldest_gen = kvmalloc(gens->nbuckets, GFP_KERNEL);
1924                 if (!ca->oldest_gen) {
1925                         percpu_ref_put(&ca->ref);
1926                         ret = -BCH_ERR_ENOMEM_gc_gens;
1927                         goto err;
1928                 }
1929
1930                 for (b = gens->first_bucket;
1931                      b < gens->nbuckets; b++)
1932                         ca->oldest_gen[b] = gens->b[b];
1933         }
1934
1935         for (unsigned i = 0; i < BTREE_ID_NR; i++)
1936                 if (btree_type_has_ptrs(i)) {
1937                         c->gc_gens_btree = i;
1938                         c->gc_gens_pos = POS_MIN;
1939
1940                         ret = bch2_trans_run(c,
1941                                 for_each_btree_key_commit(trans, iter, i,
1942                                                 POS_MIN,
1943                                                 BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS,
1944                                                 k,
1945                                                 NULL, NULL,
1946                                                 BCH_TRANS_COMMIT_no_enospc,
1947                                         gc_btree_gens_key(trans, &iter, k)));
1948                         if (ret)
1949                                 goto err;
1950                 }
1951
1952         ret = bch2_trans_run(c,
1953                 for_each_btree_key_commit(trans, iter, BTREE_ID_alloc,
1954                                 POS_MIN,
1955                                 BTREE_ITER_PREFETCH,
1956                                 k,
1957                                 NULL, NULL,
1958                                 BCH_TRANS_COMMIT_no_enospc,
1959                         bch2_alloc_write_oldest_gen(trans, &iter, k)));
1960         if (ret)
1961                 goto err;
1962
1963         c->gc_gens_btree        = 0;
1964         c->gc_gens_pos          = POS_MIN;
1965
1966         c->gc_count++;
1967
1968         bch2_time_stats_update(&c->times[BCH_TIME_btree_gc], start_time);
1969         trace_and_count(c, gc_gens_end, c);
1970 err:
1971         for_each_member_device(c, ca) {
1972                 kvfree(ca->oldest_gen);
1973                 ca->oldest_gen = NULL;
1974         }
1975
1976         up_read(&c->gc_lock);
1977         mutex_unlock(&c->gc_gens_lock);
1978         if (!bch2_err_matches(ret, EROFS))
1979                 bch_err_fn(c, ret);
1980         return ret;
1981 }
1982
1983 static int bch2_gc_thread(void *arg)
1984 {
1985         struct bch_fs *c = arg;
1986         struct io_clock *clock = &c->io_clock[WRITE];
1987         unsigned long last = atomic64_read(&clock->now);
1988         unsigned last_kick = atomic_read(&c->kick_gc);
1989
1990         set_freezable();
1991
1992         while (1) {
1993                 while (1) {
1994                         set_current_state(TASK_INTERRUPTIBLE);
1995
1996                         if (kthread_should_stop()) {
1997                                 __set_current_state(TASK_RUNNING);
1998                                 return 0;
1999                         }
2000
2001                         if (atomic_read(&c->kick_gc) != last_kick)
2002                                 break;
2003
2004                         if (c->btree_gc_periodic) {
2005                                 unsigned long next = last + c->capacity / 16;
2006
2007                                 if (atomic64_read(&clock->now) >= next)
2008                                         break;
2009
2010                                 bch2_io_clock_schedule_timeout(clock, next);
2011                         } else {
2012                                 schedule();
2013                         }
2014
2015                         try_to_freeze();
2016                 }
2017                 __set_current_state(TASK_RUNNING);
2018
2019                 last = atomic64_read(&clock->now);
2020                 last_kick = atomic_read(&c->kick_gc);
2021
2022                 /*
2023                  * Full gc is currently incompatible with btree key cache:
2024                  */
2025 #if 0
2026                 ret = bch2_gc(c, false, false);
2027 #else
2028                 bch2_gc_gens(c);
2029 #endif
2030                 debug_check_no_locks_held();
2031         }
2032
2033         return 0;
2034 }
2035
2036 void bch2_gc_thread_stop(struct bch_fs *c)
2037 {
2038         struct task_struct *p;
2039
2040         p = c->gc_thread;
2041         c->gc_thread = NULL;
2042
2043         if (p) {
2044                 kthread_stop(p);
2045                 put_task_struct(p);
2046         }
2047 }
2048
2049 int bch2_gc_thread_start(struct bch_fs *c)
2050 {
2051         struct task_struct *p;
2052
2053         if (c->gc_thread)
2054                 return 0;
2055
2056         p = kthread_create(bch2_gc_thread, c, "bch-gc/%s", c->name);
2057         if (IS_ERR(p)) {
2058                 bch_err_fn(c, PTR_ERR(p));
2059                 return PTR_ERR(p);
2060         }
2061
2062         get_task_struct(p);
2063         c->gc_thread = p;
2064         wake_up_process(p);
2065         return 0;
2066 }