1 // SPDX-License-Identifier: GPL-2.0 2 #include "bcachefs.h" 3 #include "alloc_background.h" 4 #include "alloc_foreground.h" 5 #include "backpointers.h" 6 #include "btree_cache.h" 7 #include "btree_io.h" 8 #include "btree_key_cache.h" 9 #include "btree_update.h" 10 #include "btree_update_interior.h" 11 #include "btree_gc.h" 12 #include "btree_write_buffer.h" 13 #include "buckets.h" 14 #include "buckets_waiting_for_journal.h" 15 #include "clock.h" 16 #include "debug.h" 17 #include "ec.h" 18 #include "error.h" 19 #include "lru.h" 20 #include "recovery.h" 21 #include "trace.h" 22 #include "varint.h" 23 24 #include <linux/kthread.h> 25 #include <linux/math64.h> 26 #include <linux/random.h> 27 #include <linux/rculist.h> 28 #include <linux/rcupdate.h> 29 #include <linux/sched/task.h> 30 #include <linux/sort.h> 31 32 /* Persistent alloc info: */ 33 34 static const unsigned BCH_ALLOC_V1_FIELD_BYTES[] = { 35 #define x(name, bits) [BCH_ALLOC_FIELD_V1_##name] = bits / 8, 36 BCH_ALLOC_FIELDS_V1() 37 #undef x 38 }; 39 40 struct bkey_alloc_unpacked { 41 u64 journal_seq; 42 u8 gen; 43 u8 oldest_gen; 44 u8 data_type; 45 bool need_discard:1; 46 bool need_inc_gen:1; 47 #define x(_name, _bits) u##_bits _name; 48 BCH_ALLOC_FIELDS_V2() 49 #undef x 50 }; 51 52 static inline u64 alloc_field_v1_get(const struct bch_alloc *a, 53 const void **p, unsigned field) 54 { 55 unsigned bytes = BCH_ALLOC_V1_FIELD_BYTES[field]; 56 u64 v; 57 58 if (!(a->fields & (1 << field))) 59 return 0; 60 61 switch (bytes) { 62 case 1: 63 v = *((const u8 *) *p); 64 break; 65 case 2: 66 v = le16_to_cpup(*p); 67 break; 68 case 4: 69 v = le32_to_cpup(*p); 70 break; 71 case 8: 72 v = le64_to_cpup(*p); 73 break; 74 default: 75 BUG(); 76 } 77 78 *p += bytes; 79 return v; 80 } 81 82 static void bch2_alloc_unpack_v1(struct bkey_alloc_unpacked *out, 83 struct bkey_s_c k) 84 { 85 const struct bch_alloc *in = bkey_s_c_to_alloc(k).v; 86 const void *d = in->data; 87 unsigned idx = 0; 88 89 out->gen = in->gen; 90 91 #define x(_name, _bits) out->_name = alloc_field_v1_get(in, &d, idx++); 92 BCH_ALLOC_FIELDS_V1() 93 #undef x 94 } 95 96 static int bch2_alloc_unpack_v2(struct bkey_alloc_unpacked *out, 97 struct bkey_s_c k) 98 { 99 struct bkey_s_c_alloc_v2 a = bkey_s_c_to_alloc_v2(k); 100 const u8 *in = a.v->data; 101 const u8 *end = bkey_val_end(a); 102 unsigned fieldnr = 0; 103 int ret; 104 u64 v; 105 106 out->gen = a.v->gen; 107 out->oldest_gen = a.v->oldest_gen; 108 out->data_type = a.v->data_type; 109 110 #define x(_name, _bits) \ 111 if (fieldnr < a.v->nr_fields) { \ 112 ret = bch2_varint_decode_fast(in, end, &v); \ 113 if (ret < 0) \ 114 return ret; \ 115 in += ret; \ 116 } else { \ 117 v = 0; \ 118 } \ 119 out->_name = v; \ 120 if (v != out->_name) \ 121 return -1; \ 122 fieldnr++; 123 124 BCH_ALLOC_FIELDS_V2() 125 #undef x 126 return 0; 127 } 128 129 static int bch2_alloc_unpack_v3(struct bkey_alloc_unpacked *out, 130 struct bkey_s_c k) 131 { 132 struct bkey_s_c_alloc_v3 a = bkey_s_c_to_alloc_v3(k); 133 const u8 *in = a.v->data; 134 const u8 *end = bkey_val_end(a); 135 unsigned fieldnr = 0; 136 int ret; 137 u64 v; 138 139 out->gen = a.v->gen; 140 out->oldest_gen = a.v->oldest_gen; 141 out->data_type = a.v->data_type; 142 out->need_discard = BCH_ALLOC_V3_NEED_DISCARD(a.v); 143 out->need_inc_gen = BCH_ALLOC_V3_NEED_INC_GEN(a.v); 144 out->journal_seq = le64_to_cpu(a.v->journal_seq); 145 146 #define x(_name, _bits) \ 147 if (fieldnr < a.v->nr_fields) { \ 148 ret = bch2_varint_decode_fast(in, end, &v); \ 149 if (ret < 0) \ 150 return ret; \ 151 in += ret; \ 152 } else { \ 153 v = 0; \ 154 } \ 155 out->_name = v; \ 156 if (v != out->_name) \ 157 return -1; \ 158 fieldnr++; 159 160 BCH_ALLOC_FIELDS_V2() 161 #undef x 162 return 0; 163 } 164 165 static struct bkey_alloc_unpacked bch2_alloc_unpack(struct bkey_s_c k) 166 { 167 struct bkey_alloc_unpacked ret = { .gen = 0 }; 168 169 switch (k.k->type) { 170 case KEY_TYPE_alloc: 171 bch2_alloc_unpack_v1(&ret, k); 172 break; 173 case KEY_TYPE_alloc_v2: 174 bch2_alloc_unpack_v2(&ret, k); 175 break; 176 case KEY_TYPE_alloc_v3: 177 bch2_alloc_unpack_v3(&ret, k); 178 break; 179 } 180 181 return ret; 182 } 183 184 static unsigned bch_alloc_v1_val_u64s(const struct bch_alloc *a) 185 { 186 unsigned i, bytes = offsetof(struct bch_alloc, data); 187 188 for (i = 0; i < ARRAY_SIZE(BCH_ALLOC_V1_FIELD_BYTES); i++) 189 if (a->fields & (1 << i)) 190 bytes += BCH_ALLOC_V1_FIELD_BYTES[i]; 191 192 return DIV_ROUND_UP(bytes, sizeof(u64)); 193 } 194 195 int bch2_alloc_v1_invalid(struct bch_fs *c, struct bkey_s_c k, 196 enum bkey_invalid_flags flags, 197 struct printbuf *err) 198 { 199 struct bkey_s_c_alloc a = bkey_s_c_to_alloc(k); 200 int ret = 0; 201 202 /* allow for unknown fields */ 203 bkey_fsck_err_on(bkey_val_u64s(a.k) < bch_alloc_v1_val_u64s(a.v), c, err, 204 alloc_v1_val_size_bad, 205 "incorrect value size (%zu < %u)", 206 bkey_val_u64s(a.k), bch_alloc_v1_val_u64s(a.v)); 207 fsck_err: 208 return ret; 209 } 210 211 int bch2_alloc_v2_invalid(struct bch_fs *c, struct bkey_s_c k, 212 enum bkey_invalid_flags flags, 213 struct printbuf *err) 214 { 215 struct bkey_alloc_unpacked u; 216 int ret = 0; 217 218 bkey_fsck_err_on(bch2_alloc_unpack_v2(&u, k), c, err, 219 alloc_v2_unpack_error, 220 "unpack error"); 221 fsck_err: 222 return ret; 223 } 224 225 int bch2_alloc_v3_invalid(struct bch_fs *c, struct bkey_s_c k, 226 enum bkey_invalid_flags flags, 227 struct printbuf *err) 228 { 229 struct bkey_alloc_unpacked u; 230 int ret = 0; 231 232 bkey_fsck_err_on(bch2_alloc_unpack_v3(&u, k), c, err, 233 alloc_v2_unpack_error, 234 "unpack error"); 235 fsck_err: 236 return ret; 237 } 238 239 int bch2_alloc_v4_invalid(struct bch_fs *c, struct bkey_s_c k, 240 enum bkey_invalid_flags flags, struct printbuf *err) 241 { 242 struct bkey_s_c_alloc_v4 a = bkey_s_c_to_alloc_v4(k); 243 int ret = 0; 244 245 bkey_fsck_err_on(alloc_v4_u64s(a.v) > bkey_val_u64s(k.k), c, err, 246 alloc_v4_val_size_bad, 247 "bad val size (%u > %zu)", 248 alloc_v4_u64s(a.v), bkey_val_u64s(k.k)); 249 250 bkey_fsck_err_on(!BCH_ALLOC_V4_BACKPOINTERS_START(a.v) && 251 BCH_ALLOC_V4_NR_BACKPOINTERS(a.v), c, err, 252 alloc_v4_backpointers_start_bad, 253 "invalid backpointers_start"); 254 255 bkey_fsck_err_on(alloc_data_type(*a.v, a.v->data_type) != a.v->data_type, c, err, 256 alloc_key_data_type_bad, 257 "invalid data type (got %u should be %u)", 258 a.v->data_type, alloc_data_type(*a.v, a.v->data_type)); 259 260 switch (a.v->data_type) { 261 case BCH_DATA_free: 262 case BCH_DATA_need_gc_gens: 263 case BCH_DATA_need_discard: 264 bkey_fsck_err_on(a.v->dirty_sectors || 265 a.v->cached_sectors || 266 a.v->stripe, c, err, 267 alloc_key_empty_but_have_data, 268 "empty data type free but have data"); 269 break; 270 case BCH_DATA_sb: 271 case BCH_DATA_journal: 272 case BCH_DATA_btree: 273 case BCH_DATA_user: 274 case BCH_DATA_parity: 275 bkey_fsck_err_on(!a.v->dirty_sectors, c, err, 276 alloc_key_dirty_sectors_0, 277 "data_type %s but dirty_sectors==0", 278 bch2_data_types[a.v->data_type]); 279 break; 280 case BCH_DATA_cached: 281 bkey_fsck_err_on(!a.v->cached_sectors || 282 a.v->dirty_sectors || 283 a.v->stripe, c, err, 284 alloc_key_cached_inconsistency, 285 "data type inconsistency"); 286 287 bkey_fsck_err_on(!a.v->io_time[READ] && 288 c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_to_lru_refs, 289 c, err, 290 alloc_key_cached_but_read_time_zero, 291 "cached bucket with read_time == 0"); 292 break; 293 case BCH_DATA_stripe: 294 break; 295 } 296 fsck_err: 297 return ret; 298 } 299 300 void bch2_alloc_v4_swab(struct bkey_s k) 301 { 302 struct bch_alloc_v4 *a = bkey_s_to_alloc_v4(k).v; 303 struct bch_backpointer *bp, *bps; 304 305 a->journal_seq = swab64(a->journal_seq); 306 a->flags = swab32(a->flags); 307 a->dirty_sectors = swab32(a->dirty_sectors); 308 a->cached_sectors = swab32(a->cached_sectors); 309 a->io_time[0] = swab64(a->io_time[0]); 310 a->io_time[1] = swab64(a->io_time[1]); 311 a->stripe = swab32(a->stripe); 312 a->nr_external_backpointers = swab32(a->nr_external_backpointers); 313 a->fragmentation_lru = swab64(a->fragmentation_lru); 314 315 bps = alloc_v4_backpointers(a); 316 for (bp = bps; bp < bps + BCH_ALLOC_V4_NR_BACKPOINTERS(a); bp++) { 317 bp->bucket_offset = swab40(bp->bucket_offset); 318 bp->bucket_len = swab32(bp->bucket_len); 319 bch2_bpos_swab(&bp->pos); 320 } 321 } 322 323 void bch2_alloc_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) 324 { 325 struct bch_alloc_v4 _a; 326 const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &_a); 327 unsigned i; 328 329 prt_newline(out); 330 printbuf_indent_add(out, 2); 331 332 prt_printf(out, "gen %u oldest_gen %u data_type %s", 333 a->gen, a->oldest_gen, 334 a->data_type < BCH_DATA_NR 335 ? bch2_data_types[a->data_type] 336 : "(invalid data type)"); 337 prt_newline(out); 338 prt_printf(out, "journal_seq %llu", a->journal_seq); 339 prt_newline(out); 340 prt_printf(out, "need_discard %llu", BCH_ALLOC_V4_NEED_DISCARD(a)); 341 prt_newline(out); 342 prt_printf(out, "need_inc_gen %llu", BCH_ALLOC_V4_NEED_INC_GEN(a)); 343 prt_newline(out); 344 prt_printf(out, "dirty_sectors %u", a->dirty_sectors); 345 prt_newline(out); 346 prt_printf(out, "cached_sectors %u", a->cached_sectors); 347 prt_newline(out); 348 prt_printf(out, "stripe %u", a->stripe); 349 prt_newline(out); 350 prt_printf(out, "stripe_redundancy %u", a->stripe_redundancy); 351 prt_newline(out); 352 prt_printf(out, "io_time[READ] %llu", a->io_time[READ]); 353 prt_newline(out); 354 prt_printf(out, "io_time[WRITE] %llu", a->io_time[WRITE]); 355 prt_newline(out); 356 prt_printf(out, "fragmentation %llu", a->fragmentation_lru); 357 prt_newline(out); 358 prt_printf(out, "bp_start %llu", BCH_ALLOC_V4_BACKPOINTERS_START(a)); 359 prt_newline(out); 360 361 if (BCH_ALLOC_V4_NR_BACKPOINTERS(a)) { 362 struct bkey_s_c_alloc_v4 a_raw = bkey_s_c_to_alloc_v4(k); 363 const struct bch_backpointer *bps = alloc_v4_backpointers_c(a_raw.v); 364 365 prt_printf(out, "backpointers: %llu", BCH_ALLOC_V4_NR_BACKPOINTERS(a_raw.v)); 366 printbuf_indent_add(out, 2); 367 368 for (i = 0; i < BCH_ALLOC_V4_NR_BACKPOINTERS(a_raw.v); i++) { 369 prt_newline(out); 370 bch2_backpointer_to_text(out, &bps[i]); 371 } 372 373 printbuf_indent_sub(out, 2); 374 } 375 376 printbuf_indent_sub(out, 2); 377 } 378 379 void __bch2_alloc_to_v4(struct bkey_s_c k, struct bch_alloc_v4 *out) 380 { 381 if (k.k->type == KEY_TYPE_alloc_v4) { 382 void *src, *dst; 383 384 *out = *bkey_s_c_to_alloc_v4(k).v; 385 386 src = alloc_v4_backpointers(out); 387 SET_BCH_ALLOC_V4_BACKPOINTERS_START(out, BCH_ALLOC_V4_U64s); 388 dst = alloc_v4_backpointers(out); 389 390 if (src < dst) 391 memset(src, 0, dst - src); 392 393 SET_BCH_ALLOC_V4_NR_BACKPOINTERS(out, 0); 394 } else { 395 struct bkey_alloc_unpacked u = bch2_alloc_unpack(k); 396 397 *out = (struct bch_alloc_v4) { 398 .journal_seq = u.journal_seq, 399 .flags = u.need_discard, 400 .gen = u.gen, 401 .oldest_gen = u.oldest_gen, 402 .data_type = u.data_type, 403 .stripe_redundancy = u.stripe_redundancy, 404 .dirty_sectors = u.dirty_sectors, 405 .cached_sectors = u.cached_sectors, 406 .io_time[READ] = u.read_time, 407 .io_time[WRITE] = u.write_time, 408 .stripe = u.stripe, 409 }; 410 411 SET_BCH_ALLOC_V4_BACKPOINTERS_START(out, BCH_ALLOC_V4_U64s); 412 } 413 } 414 415 static noinline struct bkey_i_alloc_v4 * 416 __bch2_alloc_to_v4_mut(struct btree_trans *trans, struct bkey_s_c k) 417 { 418 struct bkey_i_alloc_v4 *ret; 419 420 ret = bch2_trans_kmalloc(trans, max(bkey_bytes(k.k), sizeof(struct bkey_i_alloc_v4))); 421 if (IS_ERR(ret)) 422 return ret; 423 424 if (k.k->type == KEY_TYPE_alloc_v4) { 425 void *src, *dst; 426 427 bkey_reassemble(&ret->k_i, k); 428 429 src = alloc_v4_backpointers(&ret->v); 430 SET_BCH_ALLOC_V4_BACKPOINTERS_START(&ret->v, BCH_ALLOC_V4_U64s); 431 dst = alloc_v4_backpointers(&ret->v); 432 433 if (src < dst) 434 memset(src, 0, dst - src); 435 436 SET_BCH_ALLOC_V4_NR_BACKPOINTERS(&ret->v, 0); 437 set_alloc_v4_u64s(ret); 438 } else { 439 bkey_alloc_v4_init(&ret->k_i); 440 ret->k.p = k.k->p; 441 bch2_alloc_to_v4(k, &ret->v); 442 } 443 return ret; 444 } 445 446 static inline struct bkey_i_alloc_v4 *bch2_alloc_to_v4_mut_inlined(struct btree_trans *trans, struct bkey_s_c k) 447 { 448 struct bkey_s_c_alloc_v4 a; 449 450 if (likely(k.k->type == KEY_TYPE_alloc_v4) && 451 ((a = bkey_s_c_to_alloc_v4(k), true) && 452 BCH_ALLOC_V4_NR_BACKPOINTERS(a.v) == 0)) 453 return bch2_bkey_make_mut_noupdate_typed(trans, k, alloc_v4); 454 455 return __bch2_alloc_to_v4_mut(trans, k); 456 } 457 458 struct bkey_i_alloc_v4 *bch2_alloc_to_v4_mut(struct btree_trans *trans, struct bkey_s_c k) 459 { 460 return bch2_alloc_to_v4_mut_inlined(trans, k); 461 } 462 463 struct bkey_i_alloc_v4 * 464 bch2_trans_start_alloc_update(struct btree_trans *trans, struct btree_iter *iter, 465 struct bpos pos) 466 { 467 struct bkey_s_c k; 468 struct bkey_i_alloc_v4 *a; 469 int ret; 470 471 k = bch2_bkey_get_iter(trans, iter, BTREE_ID_alloc, pos, 472 BTREE_ITER_WITH_UPDATES| 473 BTREE_ITER_CACHED| 474 BTREE_ITER_INTENT); 475 ret = bkey_err(k); 476 if (unlikely(ret)) 477 return ERR_PTR(ret); 478 479 a = bch2_alloc_to_v4_mut_inlined(trans, k); 480 ret = PTR_ERR_OR_ZERO(a); 481 if (unlikely(ret)) 482 goto err; 483 return a; 484 err: 485 bch2_trans_iter_exit(trans, iter); 486 return ERR_PTR(ret); 487 } 488 489 static struct bpos alloc_gens_pos(struct bpos pos, unsigned *offset) 490 { 491 *offset = pos.offset & KEY_TYPE_BUCKET_GENS_MASK; 492 493 pos.offset >>= KEY_TYPE_BUCKET_GENS_BITS; 494 return pos; 495 } 496 497 static struct bpos bucket_gens_pos_to_alloc(struct bpos pos, unsigned offset) 498 { 499 pos.offset <<= KEY_TYPE_BUCKET_GENS_BITS; 500 pos.offset += offset; 501 return pos; 502 } 503 504 static unsigned alloc_gen(struct bkey_s_c k, unsigned offset) 505 { 506 return k.k->type == KEY_TYPE_bucket_gens 507 ? bkey_s_c_to_bucket_gens(k).v->gens[offset] 508 : 0; 509 } 510 511 int bch2_bucket_gens_invalid(struct bch_fs *c, struct bkey_s_c k, 512 enum bkey_invalid_flags flags, 513 struct printbuf *err) 514 { 515 int ret = 0; 516 517 bkey_fsck_err_on(bkey_val_bytes(k.k) != sizeof(struct bch_bucket_gens), c, err, 518 bucket_gens_val_size_bad, 519 "bad val size (%zu != %zu)", 520 bkey_val_bytes(k.k), sizeof(struct bch_bucket_gens)); 521 fsck_err: 522 return ret; 523 } 524 525 void bch2_bucket_gens_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) 526 { 527 struct bkey_s_c_bucket_gens g = bkey_s_c_to_bucket_gens(k); 528 unsigned i; 529 530 for (i = 0; i < ARRAY_SIZE(g.v->gens); i++) { 531 if (i) 532 prt_char(out, ' '); 533 prt_printf(out, "%u", g.v->gens[i]); 534 } 535 } 536 537 int bch2_bucket_gens_init(struct bch_fs *c) 538 { 539 struct btree_trans *trans = bch2_trans_get(c); 540 struct btree_iter iter; 541 struct bkey_s_c k; 542 struct bch_alloc_v4 a; 543 struct bkey_i_bucket_gens g; 544 bool have_bucket_gens_key = false; 545 unsigned offset; 546 struct bpos pos; 547 u8 gen; 548 int ret; 549 550 for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, 551 BTREE_ITER_PREFETCH, k, ret) { 552 /* 553 * Not a fsck error because this is checked/repaired by 554 * bch2_check_alloc_key() which runs later: 555 */ 556 if (!bch2_dev_bucket_exists(c, k.k->p)) 557 continue; 558 559 gen = bch2_alloc_to_v4(k, &a)->gen; 560 pos = alloc_gens_pos(iter.pos, &offset); 561 562 if (have_bucket_gens_key && bkey_cmp(iter.pos, pos)) { 563 ret = commit_do(trans, NULL, NULL, 564 BTREE_INSERT_NOFAIL| 565 BTREE_INSERT_LAZY_RW, 566 bch2_btree_insert_trans(trans, BTREE_ID_bucket_gens, &g.k_i, 0)); 567 if (ret) 568 break; 569 have_bucket_gens_key = false; 570 } 571 572 if (!have_bucket_gens_key) { 573 bkey_bucket_gens_init(&g.k_i); 574 g.k.p = pos; 575 have_bucket_gens_key = true; 576 } 577 578 g.v.gens[offset] = gen; 579 } 580 bch2_trans_iter_exit(trans, &iter); 581 582 if (have_bucket_gens_key && !ret) 583 ret = commit_do(trans, NULL, NULL, 584 BTREE_INSERT_NOFAIL| 585 BTREE_INSERT_LAZY_RW, 586 bch2_btree_insert_trans(trans, BTREE_ID_bucket_gens, &g.k_i, 0)); 587 588 bch2_trans_put(trans); 589 590 if (ret) 591 bch_err_fn(c, ret); 592 return ret; 593 } 594 595 int bch2_alloc_read(struct bch_fs *c) 596 { 597 struct btree_trans *trans = bch2_trans_get(c); 598 struct btree_iter iter; 599 struct bkey_s_c k; 600 struct bch_dev *ca; 601 int ret; 602 603 down_read(&c->gc_lock); 604 605 if (c->sb.version_upgrade_complete >= bcachefs_metadata_version_bucket_gens) { 606 const struct bch_bucket_gens *g; 607 u64 b; 608 609 for_each_btree_key(trans, iter, BTREE_ID_bucket_gens, POS_MIN, 610 BTREE_ITER_PREFETCH, k, ret) { 611 u64 start = bucket_gens_pos_to_alloc(k.k->p, 0).offset; 612 u64 end = bucket_gens_pos_to_alloc(bpos_nosnap_successor(k.k->p), 0).offset; 613 614 if (k.k->type != KEY_TYPE_bucket_gens) 615 continue; 616 617 g = bkey_s_c_to_bucket_gens(k).v; 618 619 /* 620 * Not a fsck error because this is checked/repaired by 621 * bch2_check_alloc_key() which runs later: 622 */ 623 if (!bch2_dev_exists2(c, k.k->p.inode)) 624 continue; 625 626 ca = bch_dev_bkey_exists(c, k.k->p.inode); 627 628 for (b = max_t(u64, ca->mi.first_bucket, start); 629 b < min_t(u64, ca->mi.nbuckets, end); 630 b++) 631 *bucket_gen(ca, b) = g->gens[b & KEY_TYPE_BUCKET_GENS_MASK]; 632 } 633 bch2_trans_iter_exit(trans, &iter); 634 } else { 635 struct bch_alloc_v4 a; 636 637 for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, 638 BTREE_ITER_PREFETCH, k, ret) { 639 /* 640 * Not a fsck error because this is checked/repaired by 641 * bch2_check_alloc_key() which runs later: 642 */ 643 if (!bch2_dev_bucket_exists(c, k.k->p)) 644 continue; 645 646 ca = bch_dev_bkey_exists(c, k.k->p.inode); 647 648 *bucket_gen(ca, k.k->p.offset) = bch2_alloc_to_v4(k, &a)->gen; 649 } 650 bch2_trans_iter_exit(trans, &iter); 651 } 652 653 bch2_trans_put(trans); 654 up_read(&c->gc_lock); 655 656 if (ret) 657 bch_err_fn(c, ret); 658 659 return ret; 660 } 661 662 /* Free space/discard btree: */ 663 664 static int bch2_bucket_do_index(struct btree_trans *trans, 665 struct bkey_s_c alloc_k, 666 const struct bch_alloc_v4 *a, 667 bool set) 668 { 669 struct bch_fs *c = trans->c; 670 struct bch_dev *ca = bch_dev_bkey_exists(c, alloc_k.k->p.inode); 671 struct btree_iter iter; 672 struct bkey_s_c old; 673 struct bkey_i *k; 674 enum btree_id btree; 675 enum bch_bkey_type old_type = !set ? KEY_TYPE_set : KEY_TYPE_deleted; 676 enum bch_bkey_type new_type = set ? KEY_TYPE_set : KEY_TYPE_deleted; 677 struct printbuf buf = PRINTBUF; 678 int ret; 679 680 if (a->data_type != BCH_DATA_free && 681 a->data_type != BCH_DATA_need_discard) 682 return 0; 683 684 k = bch2_trans_kmalloc_nomemzero(trans, sizeof(*k)); 685 if (IS_ERR(k)) 686 return PTR_ERR(k); 687 688 bkey_init(&k->k); 689 k->k.type = new_type; 690 691 switch (a->data_type) { 692 case BCH_DATA_free: 693 btree = BTREE_ID_freespace; 694 k->k.p = alloc_freespace_pos(alloc_k.k->p, *a); 695 bch2_key_resize(&k->k, 1); 696 break; 697 case BCH_DATA_need_discard: 698 btree = BTREE_ID_need_discard; 699 k->k.p = alloc_k.k->p; 700 break; 701 default: 702 return 0; 703 } 704 705 old = bch2_bkey_get_iter(trans, &iter, btree, 706 bkey_start_pos(&k->k), 707 BTREE_ITER_INTENT); 708 ret = bkey_err(old); 709 if (ret) 710 return ret; 711 712 if (ca->mi.freespace_initialized && 713 c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info && 714 bch2_trans_inconsistent_on(old.k->type != old_type, trans, 715 "incorrect key when %s %s:%llu:%llu:0 (got %s should be %s)\n" 716 " for %s", 717 set ? "setting" : "clearing", 718 bch2_btree_id_str(btree), 719 iter.pos.inode, 720 iter.pos.offset, 721 bch2_bkey_types[old.k->type], 722 bch2_bkey_types[old_type], 723 (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf))) { 724 ret = -EIO; 725 goto err; 726 } 727 728 ret = bch2_trans_update(trans, &iter, k, 0); 729 err: 730 bch2_trans_iter_exit(trans, &iter); 731 printbuf_exit(&buf); 732 return ret; 733 } 734 735 static noinline int bch2_bucket_gen_update(struct btree_trans *trans, 736 struct bpos bucket, u8 gen) 737 { 738 struct btree_iter iter; 739 unsigned offset; 740 struct bpos pos = alloc_gens_pos(bucket, &offset); 741 struct bkey_i_bucket_gens *g; 742 struct bkey_s_c k; 743 int ret; 744 745 g = bch2_trans_kmalloc(trans, sizeof(*g)); 746 ret = PTR_ERR_OR_ZERO(g); 747 if (ret) 748 return ret; 749 750 k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_bucket_gens, pos, 751 BTREE_ITER_INTENT| 752 BTREE_ITER_WITH_UPDATES); 753 ret = bkey_err(k); 754 if (ret) 755 return ret; 756 757 if (k.k->type != KEY_TYPE_bucket_gens) { 758 bkey_bucket_gens_init(&g->k_i); 759 g->k.p = iter.pos; 760 } else { 761 bkey_reassemble(&g->k_i, k); 762 } 763 764 g->v.gens[offset] = gen; 765 766 ret = bch2_trans_update(trans, &iter, &g->k_i, 0); 767 bch2_trans_iter_exit(trans, &iter); 768 return ret; 769 } 770 771 int bch2_trans_mark_alloc(struct btree_trans *trans, 772 enum btree_id btree_id, unsigned level, 773 struct bkey_s_c old, struct bkey_i *new, 774 unsigned flags) 775 { 776 struct bch_fs *c = trans->c; 777 struct bch_alloc_v4 old_a_convert, *new_a; 778 const struct bch_alloc_v4 *old_a; 779 u64 old_lru, new_lru; 780 int ret = 0; 781 782 /* 783 * Deletion only happens in the device removal path, with 784 * BTREE_TRIGGER_NORUN: 785 */ 786 BUG_ON(new->k.type != KEY_TYPE_alloc_v4); 787 788 old_a = bch2_alloc_to_v4(old, &old_a_convert); 789 new_a = &bkey_i_to_alloc_v4(new)->v; 790 791 new_a->data_type = alloc_data_type(*new_a, new_a->data_type); 792 793 if (new_a->dirty_sectors > old_a->dirty_sectors || 794 new_a->cached_sectors > old_a->cached_sectors) { 795 new_a->io_time[READ] = max_t(u64, 1, atomic64_read(&c->io_clock[READ].now)); 796 new_a->io_time[WRITE]= max_t(u64, 1, atomic64_read(&c->io_clock[WRITE].now)); 797 SET_BCH_ALLOC_V4_NEED_INC_GEN(new_a, true); 798 SET_BCH_ALLOC_V4_NEED_DISCARD(new_a, true); 799 } 800 801 if (data_type_is_empty(new_a->data_type) && 802 BCH_ALLOC_V4_NEED_INC_GEN(new_a) && 803 !bch2_bucket_is_open_safe(c, new->k.p.inode, new->k.p.offset)) { 804 new_a->gen++; 805 SET_BCH_ALLOC_V4_NEED_INC_GEN(new_a, false); 806 } 807 808 if (old_a->data_type != new_a->data_type || 809 (new_a->data_type == BCH_DATA_free && 810 alloc_freespace_genbits(*old_a) != alloc_freespace_genbits(*new_a))) { 811 ret = bch2_bucket_do_index(trans, old, old_a, false) ?: 812 bch2_bucket_do_index(trans, bkey_i_to_s_c(new), new_a, true); 813 if (ret) 814 return ret; 815 } 816 817 if (new_a->data_type == BCH_DATA_cached && 818 !new_a->io_time[READ]) 819 new_a->io_time[READ] = max_t(u64, 1, atomic64_read(&c->io_clock[READ].now)); 820 821 old_lru = alloc_lru_idx_read(*old_a); 822 new_lru = alloc_lru_idx_read(*new_a); 823 824 if (old_lru != new_lru) { 825 ret = bch2_lru_change(trans, new->k.p.inode, 826 bucket_to_u64(new->k.p), 827 old_lru, new_lru); 828 if (ret) 829 return ret; 830 } 831 832 new_a->fragmentation_lru = alloc_lru_idx_fragmentation(*new_a, 833 bch_dev_bkey_exists(c, new->k.p.inode)); 834 835 if (old_a->fragmentation_lru != new_a->fragmentation_lru) { 836 ret = bch2_lru_change(trans, 837 BCH_LRU_FRAGMENTATION_START, 838 bucket_to_u64(new->k.p), 839 old_a->fragmentation_lru, new_a->fragmentation_lru); 840 if (ret) 841 return ret; 842 } 843 844 if (old_a->gen != new_a->gen) { 845 ret = bch2_bucket_gen_update(trans, new->k.p, new_a->gen); 846 if (ret) 847 return ret; 848 } 849 850 return 0; 851 } 852 853 /* 854 * This synthesizes deleted extents for holes, similar to BTREE_ITER_SLOTS for 855 * extents style btrees, but works on non-extents btrees: 856 */ 857 static struct bkey_s_c bch2_get_key_or_hole(struct btree_iter *iter, struct bpos end, struct bkey *hole) 858 { 859 struct bkey_s_c k = bch2_btree_iter_peek_slot(iter); 860 861 if (bkey_err(k)) 862 return k; 863 864 if (k.k->type) { 865 return k; 866 } else { 867 struct btree_iter iter2; 868 struct bpos next; 869 870 bch2_trans_copy_iter(&iter2, iter); 871 872 if (!bpos_eq(iter->path->l[0].b->key.k.p, SPOS_MAX)) 873 end = bkey_min(end, bpos_nosnap_successor(iter->path->l[0].b->key.k.p)); 874 875 end = bkey_min(end, POS(iter->pos.inode, iter->pos.offset + U32_MAX - 1)); 876 877 /* 878 * btree node min/max is a closed interval, upto takes a half 879 * open interval: 880 */ 881 k = bch2_btree_iter_peek_upto(&iter2, end); 882 next = iter2.pos; 883 bch2_trans_iter_exit(iter->trans, &iter2); 884 885 BUG_ON(next.offset >= iter->pos.offset + U32_MAX); 886 887 if (bkey_err(k)) 888 return k; 889 890 bkey_init(hole); 891 hole->p = iter->pos; 892 893 bch2_key_resize(hole, next.offset - iter->pos.offset); 894 return (struct bkey_s_c) { hole, NULL }; 895 } 896 } 897 898 static bool next_bucket(struct bch_fs *c, struct bpos *bucket) 899 { 900 struct bch_dev *ca; 901 unsigned iter; 902 903 if (bch2_dev_bucket_exists(c, *bucket)) 904 return true; 905 906 if (bch2_dev_exists2(c, bucket->inode)) { 907 ca = bch_dev_bkey_exists(c, bucket->inode); 908 909 if (bucket->offset < ca->mi.first_bucket) { 910 bucket->offset = ca->mi.first_bucket; 911 return true; 912 } 913 914 bucket->inode++; 915 bucket->offset = 0; 916 } 917 918 rcu_read_lock(); 919 iter = bucket->inode; 920 ca = __bch2_next_dev(c, &iter, NULL); 921 if (ca) 922 *bucket = POS(ca->dev_idx, ca->mi.first_bucket); 923 rcu_read_unlock(); 924 925 return ca != NULL; 926 } 927 928 static struct bkey_s_c bch2_get_key_or_real_bucket_hole(struct btree_iter *iter, struct bkey *hole) 929 { 930 struct bch_fs *c = iter->trans->c; 931 struct bkey_s_c k; 932 again: 933 k = bch2_get_key_or_hole(iter, POS_MAX, hole); 934 if (bkey_err(k)) 935 return k; 936 937 if (!k.k->type) { 938 struct bpos bucket = bkey_start_pos(k.k); 939 940 if (!bch2_dev_bucket_exists(c, bucket)) { 941 if (!next_bucket(c, &bucket)) 942 return bkey_s_c_null; 943 944 bch2_btree_iter_set_pos(iter, bucket); 945 goto again; 946 } 947 948 if (!bch2_dev_bucket_exists(c, k.k->p)) { 949 struct bch_dev *ca = bch_dev_bkey_exists(c, bucket.inode); 950 951 bch2_key_resize(hole, ca->mi.nbuckets - bucket.offset); 952 } 953 } 954 955 return k; 956 } 957 958 static noinline_for_stack 959 int bch2_check_alloc_key(struct btree_trans *trans, 960 struct bkey_s_c alloc_k, 961 struct btree_iter *alloc_iter, 962 struct btree_iter *discard_iter, 963 struct btree_iter *freespace_iter, 964 struct btree_iter *bucket_gens_iter) 965 { 966 struct bch_fs *c = trans->c; 967 struct bch_dev *ca; 968 struct bch_alloc_v4 a_convert; 969 const struct bch_alloc_v4 *a; 970 unsigned discard_key_type, freespace_key_type; 971 unsigned gens_offset; 972 struct bkey_s_c k; 973 struct printbuf buf = PRINTBUF; 974 int ret; 975 976 if (fsck_err_on(!bch2_dev_bucket_exists(c, alloc_k.k->p), c, 977 alloc_key_to_missing_dev_bucket, 978 "alloc key for invalid device:bucket %llu:%llu", 979 alloc_k.k->p.inode, alloc_k.k->p.offset)) 980 return bch2_btree_delete_at(trans, alloc_iter, 0); 981 982 ca = bch_dev_bkey_exists(c, alloc_k.k->p.inode); 983 if (!ca->mi.freespace_initialized) 984 return 0; 985 986 a = bch2_alloc_to_v4(alloc_k, &a_convert); 987 988 discard_key_type = a->data_type == BCH_DATA_need_discard ? KEY_TYPE_set : 0; 989 bch2_btree_iter_set_pos(discard_iter, alloc_k.k->p); 990 k = bch2_btree_iter_peek_slot(discard_iter); 991 ret = bkey_err(k); 992 if (ret) 993 goto err; 994 995 if (k.k->type != discard_key_type && 996 (c->opts.reconstruct_alloc || 997 fsck_err(c, need_discard_key_wrong, 998 "incorrect key in need_discard btree (got %s should be %s)\n" 999 " %s", 1000 bch2_bkey_types[k.k->type], 1001 bch2_bkey_types[discard_key_type], 1002 (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf)))) { 1003 struct bkey_i *update = 1004 bch2_trans_kmalloc(trans, sizeof(*update)); 1005 1006 ret = PTR_ERR_OR_ZERO(update); 1007 if (ret) 1008 goto err; 1009 1010 bkey_init(&update->k); 1011 update->k.type = discard_key_type; 1012 update->k.p = discard_iter->pos; 1013 1014 ret = bch2_trans_update(trans, discard_iter, update, 0); 1015 if (ret) 1016 goto err; 1017 } 1018 1019 freespace_key_type = a->data_type == BCH_DATA_free ? KEY_TYPE_set : 0; 1020 bch2_btree_iter_set_pos(freespace_iter, alloc_freespace_pos(alloc_k.k->p, *a)); 1021 k = bch2_btree_iter_peek_slot(freespace_iter); 1022 ret = bkey_err(k); 1023 if (ret) 1024 goto err; 1025 1026 if (k.k->type != freespace_key_type && 1027 (c->opts.reconstruct_alloc || 1028 fsck_err(c, freespace_key_wrong, 1029 "incorrect key in freespace btree (got %s should be %s)\n" 1030 " %s", 1031 bch2_bkey_types[k.k->type], 1032 bch2_bkey_types[freespace_key_type], 1033 (printbuf_reset(&buf), 1034 bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf)))) { 1035 struct bkey_i *update = 1036 bch2_trans_kmalloc(trans, sizeof(*update)); 1037 1038 ret = PTR_ERR_OR_ZERO(update); 1039 if (ret) 1040 goto err; 1041 1042 bkey_init(&update->k); 1043 update->k.type = freespace_key_type; 1044 update->k.p = freespace_iter->pos; 1045 bch2_key_resize(&update->k, 1); 1046 1047 ret = bch2_trans_update(trans, freespace_iter, update, 0); 1048 if (ret) 1049 goto err; 1050 } 1051 1052 bch2_btree_iter_set_pos(bucket_gens_iter, alloc_gens_pos(alloc_k.k->p, &gens_offset)); 1053 k = bch2_btree_iter_peek_slot(bucket_gens_iter); 1054 ret = bkey_err(k); 1055 if (ret) 1056 goto err; 1057 1058 if (a->gen != alloc_gen(k, gens_offset) && 1059 (c->opts.reconstruct_alloc || 1060 fsck_err(c, bucket_gens_key_wrong, 1061 "incorrect gen in bucket_gens btree (got %u should be %u)\n" 1062 " %s", 1063 alloc_gen(k, gens_offset), a->gen, 1064 (printbuf_reset(&buf), 1065 bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf)))) { 1066 struct bkey_i_bucket_gens *g = 1067 bch2_trans_kmalloc(trans, sizeof(*g)); 1068 1069 ret = PTR_ERR_OR_ZERO(g); 1070 if (ret) 1071 goto err; 1072 1073 if (k.k->type == KEY_TYPE_bucket_gens) { 1074 bkey_reassemble(&g->k_i, k); 1075 } else { 1076 bkey_bucket_gens_init(&g->k_i); 1077 g->k.p = alloc_gens_pos(alloc_k.k->p, &gens_offset); 1078 } 1079 1080 g->v.gens[gens_offset] = a->gen; 1081 1082 ret = bch2_trans_update(trans, bucket_gens_iter, &g->k_i, 0); 1083 if (ret) 1084 goto err; 1085 } 1086 err: 1087 fsck_err: 1088 printbuf_exit(&buf); 1089 return ret; 1090 } 1091 1092 static noinline_for_stack 1093 int bch2_check_alloc_hole_freespace(struct btree_trans *trans, 1094 struct bpos start, 1095 struct bpos *end, 1096 struct btree_iter *freespace_iter) 1097 { 1098 struct bch_fs *c = trans->c; 1099 struct bch_dev *ca; 1100 struct bkey_s_c k; 1101 struct printbuf buf = PRINTBUF; 1102 int ret; 1103 1104 ca = bch_dev_bkey_exists(c, start.inode); 1105 if (!ca->mi.freespace_initialized) 1106 return 0; 1107 1108 bch2_btree_iter_set_pos(freespace_iter, start); 1109 1110 k = bch2_btree_iter_peek_slot(freespace_iter); 1111 ret = bkey_err(k); 1112 if (ret) 1113 goto err; 1114 1115 *end = bkey_min(k.k->p, *end); 1116 1117 if (k.k->type != KEY_TYPE_set && 1118 (c->opts.reconstruct_alloc || 1119 fsck_err(c, freespace_hole_missing, 1120 "hole in alloc btree missing in freespace btree\n" 1121 " device %llu buckets %llu-%llu", 1122 freespace_iter->pos.inode, 1123 freespace_iter->pos.offset, 1124 end->offset))) { 1125 struct bkey_i *update = 1126 bch2_trans_kmalloc(trans, sizeof(*update)); 1127 1128 ret = PTR_ERR_OR_ZERO(update); 1129 if (ret) 1130 goto err; 1131 1132 bkey_init(&update->k); 1133 update->k.type = KEY_TYPE_set; 1134 update->k.p = freespace_iter->pos; 1135 bch2_key_resize(&update->k, 1136 min_t(u64, U32_MAX, end->offset - 1137 freespace_iter->pos.offset)); 1138 1139 ret = bch2_trans_update(trans, freespace_iter, update, 0); 1140 if (ret) 1141 goto err; 1142 } 1143 err: 1144 fsck_err: 1145 printbuf_exit(&buf); 1146 return ret; 1147 } 1148 1149 static noinline_for_stack 1150 int bch2_check_alloc_hole_bucket_gens(struct btree_trans *trans, 1151 struct bpos start, 1152 struct bpos *end, 1153 struct btree_iter *bucket_gens_iter) 1154 { 1155 struct bch_fs *c = trans->c; 1156 struct bkey_s_c k; 1157 struct printbuf buf = PRINTBUF; 1158 unsigned i, gens_offset, gens_end_offset; 1159 int ret; 1160 1161 if (c->sb.version < bcachefs_metadata_version_bucket_gens) 1162 return 0; 1163 1164 bch2_btree_iter_set_pos(bucket_gens_iter, alloc_gens_pos(start, &gens_offset)); 1165 1166 k = bch2_btree_iter_peek_slot(bucket_gens_iter); 1167 ret = bkey_err(k); 1168 if (ret) 1169 goto err; 1170 1171 if (bkey_cmp(alloc_gens_pos(start, &gens_offset), 1172 alloc_gens_pos(*end, &gens_end_offset))) 1173 gens_end_offset = KEY_TYPE_BUCKET_GENS_NR; 1174 1175 if (k.k->type == KEY_TYPE_bucket_gens) { 1176 struct bkey_i_bucket_gens g; 1177 bool need_update = false; 1178 1179 bkey_reassemble(&g.k_i, k); 1180 1181 for (i = gens_offset; i < gens_end_offset; i++) { 1182 if (fsck_err_on(g.v.gens[i], c, 1183 bucket_gens_hole_wrong, 1184 "hole in alloc btree at %llu:%llu with nonzero gen in bucket_gens btree (%u)", 1185 bucket_gens_pos_to_alloc(k.k->p, i).inode, 1186 bucket_gens_pos_to_alloc(k.k->p, i).offset, 1187 g.v.gens[i])) { 1188 g.v.gens[i] = 0; 1189 need_update = true; 1190 } 1191 } 1192 1193 if (need_update) { 1194 struct bkey_i *u = bch2_trans_kmalloc(trans, sizeof(g)); 1195 1196 ret = PTR_ERR_OR_ZERO(u); 1197 if (ret) 1198 goto err; 1199 1200 memcpy(u, &g, sizeof(g)); 1201 1202 ret = bch2_trans_update(trans, bucket_gens_iter, u, 0); 1203 if (ret) 1204 goto err; 1205 } 1206 } 1207 1208 *end = bkey_min(*end, bucket_gens_pos_to_alloc(bpos_nosnap_successor(k.k->p), 0)); 1209 err: 1210 fsck_err: 1211 printbuf_exit(&buf); 1212 return ret; 1213 } 1214 1215 static noinline_for_stack int __bch2_check_discard_freespace_key(struct btree_trans *trans, 1216 struct btree_iter *iter) 1217 { 1218 struct bch_fs *c = trans->c; 1219 struct btree_iter alloc_iter; 1220 struct bkey_s_c alloc_k; 1221 struct bch_alloc_v4 a_convert; 1222 const struct bch_alloc_v4 *a; 1223 u64 genbits; 1224 struct bpos pos; 1225 enum bch_data_type state = iter->btree_id == BTREE_ID_need_discard 1226 ? BCH_DATA_need_discard 1227 : BCH_DATA_free; 1228 struct printbuf buf = PRINTBUF; 1229 int ret; 1230 1231 pos = iter->pos; 1232 pos.offset &= ~(~0ULL << 56); 1233 genbits = iter->pos.offset & (~0ULL << 56); 1234 1235 alloc_k = bch2_bkey_get_iter(trans, &alloc_iter, BTREE_ID_alloc, pos, 0); 1236 ret = bkey_err(alloc_k); 1237 if (ret) 1238 return ret; 1239 1240 if (fsck_err_on(!bch2_dev_bucket_exists(c, pos), c, 1241 need_discard_freespace_key_to_invalid_dev_bucket, 1242 "entry in %s btree for nonexistant dev:bucket %llu:%llu", 1243 bch2_btree_id_str(iter->btree_id), pos.inode, pos.offset)) 1244 goto delete; 1245 1246 a = bch2_alloc_to_v4(alloc_k, &a_convert); 1247 1248 if (fsck_err_on(a->data_type != state || 1249 (state == BCH_DATA_free && 1250 genbits != alloc_freespace_genbits(*a)), c, 1251 need_discard_freespace_key_bad, 1252 "%s\n incorrectly set at %s:%llu:%llu:0 (free %u, genbits %llu should be %llu)", 1253 (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf), 1254 bch2_btree_id_str(iter->btree_id), 1255 iter->pos.inode, 1256 iter->pos.offset, 1257 a->data_type == state, 1258 genbits >> 56, alloc_freespace_genbits(*a) >> 56)) 1259 goto delete; 1260 out: 1261 fsck_err: 1262 set_btree_iter_dontneed(&alloc_iter); 1263 bch2_trans_iter_exit(trans, &alloc_iter); 1264 printbuf_exit(&buf); 1265 return ret; 1266 delete: 1267 ret = bch2_btree_delete_extent_at(trans, iter, 1268 iter->btree_id == BTREE_ID_freespace ? 1 : 0, 0) ?: 1269 bch2_trans_commit(trans, NULL, NULL, 1270 BTREE_INSERT_NOFAIL|BTREE_INSERT_LAZY_RW); 1271 goto out; 1272 } 1273 1274 static int bch2_check_discard_freespace_key(struct btree_trans *trans, 1275 struct btree_iter *iter, 1276 struct bpos end) 1277 { 1278 if (!btree_id_is_extents(iter->btree_id)) { 1279 return __bch2_check_discard_freespace_key(trans, iter); 1280 } else { 1281 int ret = 0; 1282 1283 while (!bkey_eq(iter->pos, end) && 1284 !(ret = btree_trans_too_many_iters(trans) ?: 1285 __bch2_check_discard_freespace_key(trans, iter))) 1286 bch2_btree_iter_set_pos(iter, bpos_nosnap_successor(iter->pos)); 1287 1288 return ret; 1289 } 1290 } 1291 1292 /* 1293 * We've already checked that generation numbers in the bucket_gens btree are 1294 * valid for buckets that exist; this just checks for keys for nonexistent 1295 * buckets. 1296 */ 1297 static noinline_for_stack 1298 int bch2_check_bucket_gens_key(struct btree_trans *trans, 1299 struct btree_iter *iter, 1300 struct bkey_s_c k) 1301 { 1302 struct bch_fs *c = trans->c; 1303 struct bkey_i_bucket_gens g; 1304 struct bch_dev *ca; 1305 u64 start = bucket_gens_pos_to_alloc(k.k->p, 0).offset; 1306 u64 end = bucket_gens_pos_to_alloc(bpos_nosnap_successor(k.k->p), 0).offset; 1307 u64 b; 1308 bool need_update = false, dev_exists; 1309 struct printbuf buf = PRINTBUF; 1310 int ret = 0; 1311 1312 BUG_ON(k.k->type != KEY_TYPE_bucket_gens); 1313 bkey_reassemble(&g.k_i, k); 1314 1315 /* if no bch_dev, skip out whether we repair or not */ 1316 dev_exists = bch2_dev_exists2(c, k.k->p.inode); 1317 if (!dev_exists) { 1318 if (fsck_err_on(!dev_exists, c, 1319 bucket_gens_to_invalid_dev, 1320 "bucket_gens key for invalid device:\n %s", 1321 (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { 1322 ret = bch2_btree_delete_at(trans, iter, 0); 1323 } 1324 goto out; 1325 } 1326 1327 ca = bch_dev_bkey_exists(c, k.k->p.inode); 1328 if (fsck_err_on(end <= ca->mi.first_bucket || 1329 start >= ca->mi.nbuckets, c, 1330 bucket_gens_to_invalid_buckets, 1331 "bucket_gens key for invalid buckets:\n %s", 1332 (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { 1333 ret = bch2_btree_delete_at(trans, iter, 0); 1334 goto out; 1335 } 1336 1337 for (b = start; b < ca->mi.first_bucket; b++) 1338 if (fsck_err_on(g.v.gens[b & KEY_TYPE_BUCKET_GENS_MASK], c, 1339 bucket_gens_nonzero_for_invalid_buckets, 1340 "bucket_gens key has nonzero gen for invalid bucket")) { 1341 g.v.gens[b & KEY_TYPE_BUCKET_GENS_MASK] = 0; 1342 need_update = true; 1343 } 1344 1345 for (b = ca->mi.nbuckets; b < end; b++) 1346 if (fsck_err_on(g.v.gens[b & KEY_TYPE_BUCKET_GENS_MASK], c, 1347 bucket_gens_nonzero_for_invalid_buckets, 1348 "bucket_gens key has nonzero gen for invalid bucket")) { 1349 g.v.gens[b & KEY_TYPE_BUCKET_GENS_MASK] = 0; 1350 need_update = true; 1351 } 1352 1353 if (need_update) { 1354 struct bkey_i *u = bch2_trans_kmalloc(trans, sizeof(g)); 1355 1356 ret = PTR_ERR_OR_ZERO(u); 1357 if (ret) 1358 goto out; 1359 1360 memcpy(u, &g, sizeof(g)); 1361 ret = bch2_trans_update(trans, iter, u, 0); 1362 } 1363 out: 1364 fsck_err: 1365 printbuf_exit(&buf); 1366 return ret; 1367 } 1368 1369 int bch2_check_alloc_info(struct bch_fs *c) 1370 { 1371 struct btree_trans *trans = bch2_trans_get(c); 1372 struct btree_iter iter, discard_iter, freespace_iter, bucket_gens_iter; 1373 struct bkey hole; 1374 struct bkey_s_c k; 1375 int ret = 0; 1376 1377 bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, POS_MIN, 1378 BTREE_ITER_PREFETCH); 1379 bch2_trans_iter_init(trans, &discard_iter, BTREE_ID_need_discard, POS_MIN, 1380 BTREE_ITER_PREFETCH); 1381 bch2_trans_iter_init(trans, &freespace_iter, BTREE_ID_freespace, POS_MIN, 1382 BTREE_ITER_PREFETCH); 1383 bch2_trans_iter_init(trans, &bucket_gens_iter, BTREE_ID_bucket_gens, POS_MIN, 1384 BTREE_ITER_PREFETCH); 1385 1386 while (1) { 1387 struct bpos next; 1388 1389 bch2_trans_begin(trans); 1390 1391 k = bch2_get_key_or_real_bucket_hole(&iter, &hole); 1392 ret = bkey_err(k); 1393 if (ret) 1394 goto bkey_err; 1395 1396 if (!k.k) 1397 break; 1398 1399 if (k.k->type) { 1400 next = bpos_nosnap_successor(k.k->p); 1401 1402 ret = bch2_check_alloc_key(trans, 1403 k, &iter, 1404 &discard_iter, 1405 &freespace_iter, 1406 &bucket_gens_iter); 1407 if (ret) 1408 goto bkey_err; 1409 } else { 1410 next = k.k->p; 1411 1412 ret = bch2_check_alloc_hole_freespace(trans, 1413 bkey_start_pos(k.k), 1414 &next, 1415 &freespace_iter) ?: 1416 bch2_check_alloc_hole_bucket_gens(trans, 1417 bkey_start_pos(k.k), 1418 &next, 1419 &bucket_gens_iter); 1420 if (ret) 1421 goto bkey_err; 1422 } 1423 1424 ret = bch2_trans_commit(trans, NULL, NULL, 1425 BTREE_INSERT_NOFAIL| 1426 BTREE_INSERT_LAZY_RW); 1427 if (ret) 1428 goto bkey_err; 1429 1430 bch2_btree_iter_set_pos(&iter, next); 1431 bkey_err: 1432 if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) 1433 continue; 1434 if (ret) 1435 break; 1436 } 1437 bch2_trans_iter_exit(trans, &bucket_gens_iter); 1438 bch2_trans_iter_exit(trans, &freespace_iter); 1439 bch2_trans_iter_exit(trans, &discard_iter); 1440 bch2_trans_iter_exit(trans, &iter); 1441 1442 if (ret < 0) 1443 goto err; 1444 1445 ret = for_each_btree_key2(trans, iter, 1446 BTREE_ID_need_discard, POS_MIN, 1447 BTREE_ITER_PREFETCH, k, 1448 bch2_check_discard_freespace_key(trans, &iter, k.k->p)) ?: 1449 for_each_btree_key2(trans, iter, 1450 BTREE_ID_freespace, POS_MIN, 1451 BTREE_ITER_PREFETCH, k, 1452 bch2_check_discard_freespace_key(trans, &iter, k.k->p)) ?: 1453 for_each_btree_key_commit(trans, iter, 1454 BTREE_ID_bucket_gens, POS_MIN, 1455 BTREE_ITER_PREFETCH, k, 1456 NULL, NULL, BTREE_INSERT_NOFAIL|BTREE_INSERT_LAZY_RW, 1457 bch2_check_bucket_gens_key(trans, &iter, k)); 1458 err: 1459 bch2_trans_put(trans); 1460 if (ret) 1461 bch_err_fn(c, ret); 1462 return ret; 1463 } 1464 1465 static int bch2_check_alloc_to_lru_ref(struct btree_trans *trans, 1466 struct btree_iter *alloc_iter) 1467 { 1468 struct bch_fs *c = trans->c; 1469 struct btree_iter lru_iter; 1470 struct bch_alloc_v4 a_convert; 1471 const struct bch_alloc_v4 *a; 1472 struct bkey_s_c alloc_k, lru_k; 1473 struct printbuf buf = PRINTBUF; 1474 int ret; 1475 1476 alloc_k = bch2_btree_iter_peek(alloc_iter); 1477 if (!alloc_k.k) 1478 return 0; 1479 1480 ret = bkey_err(alloc_k); 1481 if (ret) 1482 return ret; 1483 1484 a = bch2_alloc_to_v4(alloc_k, &a_convert); 1485 1486 if (a->data_type != BCH_DATA_cached) 1487 return 0; 1488 1489 lru_k = bch2_bkey_get_iter(trans, &lru_iter, BTREE_ID_lru, 1490 lru_pos(alloc_k.k->p.inode, 1491 bucket_to_u64(alloc_k.k->p), 1492 a->io_time[READ]), 0); 1493 ret = bkey_err(lru_k); 1494 if (ret) 1495 return ret; 1496 1497 if (fsck_err_on(!a->io_time[READ], c, 1498 alloc_key_cached_but_read_time_zero, 1499 "cached bucket with read_time 0\n" 1500 " %s", 1501 (printbuf_reset(&buf), 1502 bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf)) || 1503 fsck_err_on(lru_k.k->type != KEY_TYPE_set, c, 1504 alloc_key_to_missing_lru_entry, 1505 "missing lru entry\n" 1506 " %s", 1507 (printbuf_reset(&buf), 1508 bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf))) { 1509 u64 read_time = a->io_time[READ] ?: 1510 atomic64_read(&c->io_clock[READ].now); 1511 1512 ret = bch2_lru_set(trans, 1513 alloc_k.k->p.inode, 1514 bucket_to_u64(alloc_k.k->p), 1515 read_time); 1516 if (ret) 1517 goto err; 1518 1519 if (a->io_time[READ] != read_time) { 1520 struct bkey_i_alloc_v4 *a_mut = 1521 bch2_alloc_to_v4_mut(trans, alloc_k); 1522 ret = PTR_ERR_OR_ZERO(a_mut); 1523 if (ret) 1524 goto err; 1525 1526 a_mut->v.io_time[READ] = read_time; 1527 ret = bch2_trans_update(trans, alloc_iter, 1528 &a_mut->k_i, BTREE_TRIGGER_NORUN); 1529 if (ret) 1530 goto err; 1531 } 1532 } 1533 err: 1534 fsck_err: 1535 bch2_trans_iter_exit(trans, &lru_iter); 1536 printbuf_exit(&buf); 1537 return ret; 1538 } 1539 1540 int bch2_check_alloc_to_lru_refs(struct bch_fs *c) 1541 { 1542 struct btree_iter iter; 1543 struct bkey_s_c k; 1544 int ret = 0; 1545 1546 ret = bch2_trans_run(c, 1547 for_each_btree_key_commit(trans, iter, BTREE_ID_alloc, 1548 POS_MIN, BTREE_ITER_PREFETCH, k, 1549 NULL, NULL, BTREE_INSERT_NOFAIL|BTREE_INSERT_LAZY_RW, 1550 bch2_check_alloc_to_lru_ref(trans, &iter))); 1551 if (ret) 1552 bch_err_fn(c, ret); 1553 return ret; 1554 } 1555 1556 static int bch2_discard_one_bucket(struct btree_trans *trans, 1557 struct btree_iter *need_discard_iter, 1558 struct bpos *discard_pos_done, 1559 u64 *seen, 1560 u64 *open, 1561 u64 *need_journal_commit, 1562 u64 *discarded) 1563 { 1564 struct bch_fs *c = trans->c; 1565 struct bpos pos = need_discard_iter->pos; 1566 struct btree_iter iter = { NULL }; 1567 struct bkey_s_c k; 1568 struct bch_dev *ca; 1569 struct bkey_i_alloc_v4 *a; 1570 struct printbuf buf = PRINTBUF; 1571 int ret = 0; 1572 1573 ca = bch_dev_bkey_exists(c, pos.inode); 1574 if (!percpu_ref_tryget(&ca->io_ref)) { 1575 bch2_btree_iter_set_pos(need_discard_iter, POS(pos.inode + 1, 0)); 1576 return 0; 1577 } 1578 1579 if (bch2_bucket_is_open_safe(c, pos.inode, pos.offset)) { 1580 (*open)++; 1581 goto out; 1582 } 1583 1584 if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, 1585 c->journal.flushed_seq_ondisk, 1586 pos.inode, pos.offset)) { 1587 (*need_journal_commit)++; 1588 goto out; 1589 } 1590 1591 k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_alloc, 1592 need_discard_iter->pos, 1593 BTREE_ITER_CACHED); 1594 ret = bkey_err(k); 1595 if (ret) 1596 goto out; 1597 1598 a = bch2_alloc_to_v4_mut(trans, k); 1599 ret = PTR_ERR_OR_ZERO(a); 1600 if (ret) 1601 goto out; 1602 1603 if (BCH_ALLOC_V4_NEED_INC_GEN(&a->v)) { 1604 a->v.gen++; 1605 SET_BCH_ALLOC_V4_NEED_INC_GEN(&a->v, false); 1606 goto write; 1607 } 1608 1609 if (a->v.journal_seq > c->journal.flushed_seq_ondisk) { 1610 if (c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info) { 1611 bch2_trans_inconsistent(trans, 1612 "clearing need_discard but journal_seq %llu > flushed_seq %llu\n" 1613 "%s", 1614 a->v.journal_seq, 1615 c->journal.flushed_seq_ondisk, 1616 (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); 1617 ret = -EIO; 1618 } 1619 goto out; 1620 } 1621 1622 if (a->v.data_type != BCH_DATA_need_discard) { 1623 if (c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info) { 1624 bch2_trans_inconsistent(trans, 1625 "bucket incorrectly set in need_discard btree\n" 1626 "%s", 1627 (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); 1628 ret = -EIO; 1629 } 1630 1631 goto out; 1632 } 1633 1634 if (!bkey_eq(*discard_pos_done, iter.pos) && 1635 ca->mi.discard && !c->opts.nochanges) { 1636 /* 1637 * This works without any other locks because this is the only 1638 * thread that removes items from the need_discard tree 1639 */ 1640 bch2_trans_unlock(trans); 1641 blkdev_issue_discard(ca->disk_sb.bdev, 1642 k.k->p.offset * ca->mi.bucket_size, 1643 ca->mi.bucket_size, 1644 GFP_KERNEL); 1645 *discard_pos_done = iter.pos; 1646 1647 ret = bch2_trans_relock_notrace(trans); 1648 if (ret) 1649 goto out; 1650 } 1651 1652 SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); 1653 a->v.data_type = alloc_data_type(a->v, a->v.data_type); 1654 write: 1655 ret = bch2_trans_update(trans, &iter, &a->k_i, 0) ?: 1656 bch2_trans_commit(trans, NULL, NULL, 1657 BCH_WATERMARK_btree| 1658 BTREE_INSERT_NOFAIL); 1659 if (ret) 1660 goto out; 1661 1662 this_cpu_inc(c->counters[BCH_COUNTER_bucket_discard]); 1663 (*discarded)++; 1664 out: 1665 (*seen)++; 1666 bch2_trans_iter_exit(trans, &iter); 1667 percpu_ref_put(&ca->io_ref); 1668 printbuf_exit(&buf); 1669 return ret; 1670 } 1671 1672 static void bch2_do_discards_work(struct work_struct *work) 1673 { 1674 struct bch_fs *c = container_of(work, struct bch_fs, discard_work); 1675 struct btree_iter iter; 1676 struct bkey_s_c k; 1677 u64 seen = 0, open = 0, need_journal_commit = 0, discarded = 0; 1678 struct bpos discard_pos_done = POS_MAX; 1679 int ret; 1680 1681 /* 1682 * We're doing the commit in bch2_discard_one_bucket instead of using 1683 * for_each_btree_key_commit() so that we can increment counters after 1684 * successful commit: 1685 */ 1686 ret = bch2_trans_run(c, 1687 for_each_btree_key2(trans, iter, 1688 BTREE_ID_need_discard, POS_MIN, 0, k, 1689 bch2_discard_one_bucket(trans, &iter, &discard_pos_done, 1690 &seen, 1691 &open, 1692 &need_journal_commit, 1693 &discarded))); 1694 1695 if (need_journal_commit * 2 > seen) 1696 bch2_journal_flush_async(&c->journal, NULL); 1697 1698 bch2_write_ref_put(c, BCH_WRITE_REF_discard); 1699 1700 trace_discard_buckets(c, seen, open, need_journal_commit, discarded, 1701 bch2_err_str(ret)); 1702 } 1703 1704 void bch2_do_discards(struct bch_fs *c) 1705 { 1706 if (bch2_write_ref_tryget(c, BCH_WRITE_REF_discard) && 1707 !queue_work(c->write_ref_wq, &c->discard_work)) 1708 bch2_write_ref_put(c, BCH_WRITE_REF_discard); 1709 } 1710 1711 static int invalidate_one_bucket(struct btree_trans *trans, 1712 struct btree_iter *lru_iter, 1713 struct bkey_s_c lru_k, 1714 s64 *nr_to_invalidate) 1715 { 1716 struct bch_fs *c = trans->c; 1717 struct btree_iter alloc_iter = { NULL }; 1718 struct bkey_i_alloc_v4 *a = NULL; 1719 struct printbuf buf = PRINTBUF; 1720 struct bpos bucket = u64_to_bucket(lru_k.k->p.offset); 1721 unsigned cached_sectors; 1722 int ret = 0; 1723 1724 if (*nr_to_invalidate <= 0) 1725 return 1; 1726 1727 if (!bch2_dev_bucket_exists(c, bucket)) { 1728 prt_str(&buf, "lru entry points to invalid bucket"); 1729 goto err; 1730 } 1731 1732 if (bch2_bucket_is_open_safe(c, bucket.inode, bucket.offset)) 1733 return 0; 1734 1735 a = bch2_trans_start_alloc_update(trans, &alloc_iter, bucket); 1736 ret = PTR_ERR_OR_ZERO(a); 1737 if (ret) 1738 goto out; 1739 1740 /* We expect harmless races here due to the btree write buffer: */ 1741 if (lru_pos_time(lru_iter->pos) != alloc_lru_idx_read(a->v)) 1742 goto out; 1743 1744 BUG_ON(a->v.data_type != BCH_DATA_cached); 1745 1746 if (!a->v.cached_sectors) 1747 bch_err(c, "invalidating empty bucket, confused"); 1748 1749 cached_sectors = a->v.cached_sectors; 1750 1751 SET_BCH_ALLOC_V4_NEED_INC_GEN(&a->v, false); 1752 a->v.gen++; 1753 a->v.data_type = 0; 1754 a->v.dirty_sectors = 0; 1755 a->v.cached_sectors = 0; 1756 a->v.io_time[READ] = atomic64_read(&c->io_clock[READ].now); 1757 a->v.io_time[WRITE] = atomic64_read(&c->io_clock[WRITE].now); 1758 1759 ret = bch2_trans_update(trans, &alloc_iter, &a->k_i, 1760 BTREE_TRIGGER_BUCKET_INVALIDATE) ?: 1761 bch2_trans_commit(trans, NULL, NULL, 1762 BCH_WATERMARK_btree| 1763 BTREE_INSERT_NOFAIL); 1764 if (ret) 1765 goto out; 1766 1767 trace_and_count(c, bucket_invalidate, c, bucket.inode, bucket.offset, cached_sectors); 1768 --*nr_to_invalidate; 1769 out: 1770 bch2_trans_iter_exit(trans, &alloc_iter); 1771 printbuf_exit(&buf); 1772 return ret; 1773 err: 1774 prt_str(&buf, "\n lru key: "); 1775 bch2_bkey_val_to_text(&buf, c, lru_k); 1776 1777 prt_str(&buf, "\n lru entry: "); 1778 bch2_lru_pos_to_text(&buf, lru_iter->pos); 1779 1780 prt_str(&buf, "\n alloc key: "); 1781 if (!a) 1782 bch2_bpos_to_text(&buf, bucket); 1783 else 1784 bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&a->k_i)); 1785 1786 bch_err(c, "%s", buf.buf); 1787 if (c->curr_recovery_pass > BCH_RECOVERY_PASS_check_lrus) { 1788 bch2_inconsistent_error(c); 1789 ret = -EINVAL; 1790 } 1791 1792 goto out; 1793 } 1794 1795 static void bch2_do_invalidates_work(struct work_struct *work) 1796 { 1797 struct bch_fs *c = container_of(work, struct bch_fs, invalidate_work); 1798 struct bch_dev *ca; 1799 struct btree_trans *trans = bch2_trans_get(c); 1800 struct btree_iter iter; 1801 struct bkey_s_c k; 1802 unsigned i; 1803 int ret = 0; 1804 1805 ret = bch2_btree_write_buffer_flush(trans); 1806 if (ret) 1807 goto err; 1808 1809 for_each_member_device(ca, c, i) { 1810 s64 nr_to_invalidate = 1811 should_invalidate_buckets(ca, bch2_dev_usage_read(ca)); 1812 1813 ret = for_each_btree_key2_upto(trans, iter, BTREE_ID_lru, 1814 lru_pos(ca->dev_idx, 0, 0), 1815 lru_pos(ca->dev_idx, U64_MAX, LRU_TIME_MAX), 1816 BTREE_ITER_INTENT, k, 1817 invalidate_one_bucket(trans, &iter, k, &nr_to_invalidate)); 1818 1819 if (ret < 0) { 1820 percpu_ref_put(&ca->ref); 1821 break; 1822 } 1823 } 1824 err: 1825 bch2_trans_put(trans); 1826 bch2_write_ref_put(c, BCH_WRITE_REF_invalidate); 1827 } 1828 1829 void bch2_do_invalidates(struct bch_fs *c) 1830 { 1831 if (bch2_write_ref_tryget(c, BCH_WRITE_REF_invalidate) && 1832 !queue_work(c->write_ref_wq, &c->invalidate_work)) 1833 bch2_write_ref_put(c, BCH_WRITE_REF_invalidate); 1834 } 1835 1836 int bch2_dev_freespace_init(struct bch_fs *c, struct bch_dev *ca, 1837 u64 bucket_start, u64 bucket_end) 1838 { 1839 struct btree_trans *trans = bch2_trans_get(c); 1840 struct btree_iter iter; 1841 struct bkey_s_c k; 1842 struct bkey hole; 1843 struct bpos end = POS(ca->dev_idx, bucket_end); 1844 struct bch_member *m; 1845 unsigned long last_updated = jiffies; 1846 int ret; 1847 1848 BUG_ON(bucket_start > bucket_end); 1849 BUG_ON(bucket_end > ca->mi.nbuckets); 1850 1851 bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, 1852 POS(ca->dev_idx, max_t(u64, ca->mi.first_bucket, bucket_start)), 1853 BTREE_ITER_PREFETCH); 1854 /* 1855 * Scan the alloc btree for every bucket on @ca, and add buckets to the 1856 * freespace/need_discard/need_gc_gens btrees as needed: 1857 */ 1858 while (1) { 1859 if (last_updated + HZ * 10 < jiffies) { 1860 bch_info(ca, "%s: currently at %llu/%llu", 1861 __func__, iter.pos.offset, ca->mi.nbuckets); 1862 last_updated = jiffies; 1863 } 1864 1865 bch2_trans_begin(trans); 1866 1867 if (bkey_ge(iter.pos, end)) { 1868 ret = 0; 1869 break; 1870 } 1871 1872 k = bch2_get_key_or_hole(&iter, end, &hole); 1873 ret = bkey_err(k); 1874 if (ret) 1875 goto bkey_err; 1876 1877 if (k.k->type) { 1878 /* 1879 * We process live keys in the alloc btree one at a 1880 * time: 1881 */ 1882 struct bch_alloc_v4 a_convert; 1883 const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert); 1884 1885 ret = bch2_bucket_do_index(trans, k, a, true) ?: 1886 bch2_trans_commit(trans, NULL, NULL, 1887 BTREE_INSERT_LAZY_RW| 1888 BTREE_INSERT_NOFAIL); 1889 if (ret) 1890 goto bkey_err; 1891 1892 bch2_btree_iter_advance(&iter); 1893 } else { 1894 struct bkey_i *freespace; 1895 1896 freespace = bch2_trans_kmalloc(trans, sizeof(*freespace)); 1897 ret = PTR_ERR_OR_ZERO(freespace); 1898 if (ret) 1899 goto bkey_err; 1900 1901 bkey_init(&freespace->k); 1902 freespace->k.type = KEY_TYPE_set; 1903 freespace->k.p = k.k->p; 1904 freespace->k.size = k.k->size; 1905 1906 ret = bch2_btree_insert_trans(trans, BTREE_ID_freespace, freespace, 0) ?: 1907 bch2_trans_commit(trans, NULL, NULL, 1908 BTREE_INSERT_LAZY_RW| 1909 BTREE_INSERT_NOFAIL); 1910 if (ret) 1911 goto bkey_err; 1912 1913 bch2_btree_iter_set_pos(&iter, k.k->p); 1914 } 1915 bkey_err: 1916 if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) 1917 continue; 1918 if (ret) 1919 break; 1920 } 1921 1922 bch2_trans_iter_exit(trans, &iter); 1923 bch2_trans_put(trans); 1924 1925 if (ret < 0) { 1926 bch_err_msg(ca, ret, "initializing free space"); 1927 return ret; 1928 } 1929 1930 mutex_lock(&c->sb_lock); 1931 m = bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx); 1932 SET_BCH_MEMBER_FREESPACE_INITIALIZED(m, true); 1933 mutex_unlock(&c->sb_lock); 1934 1935 return 0; 1936 } 1937 1938 int bch2_fs_freespace_init(struct bch_fs *c) 1939 { 1940 struct bch_dev *ca; 1941 unsigned i; 1942 int ret = 0; 1943 bool doing_init = false; 1944 1945 /* 1946 * We can crash during the device add path, so we need to check this on 1947 * every mount: 1948 */ 1949 1950 for_each_member_device(ca, c, i) { 1951 if (ca->mi.freespace_initialized) 1952 continue; 1953 1954 if (!doing_init) { 1955 bch_info(c, "initializing freespace"); 1956 doing_init = true; 1957 } 1958 1959 ret = bch2_dev_freespace_init(c, ca, 0, ca->mi.nbuckets); 1960 if (ret) { 1961 percpu_ref_put(&ca->ref); 1962 bch_err_fn(c, ret); 1963 return ret; 1964 } 1965 } 1966 1967 if (doing_init) { 1968 mutex_lock(&c->sb_lock); 1969 bch2_write_super(c); 1970 mutex_unlock(&c->sb_lock); 1971 bch_verbose(c, "done initializing freespace"); 1972 } 1973 1974 return 0; 1975 } 1976 1977 /* Bucket IO clocks: */ 1978 1979 int bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev, 1980 size_t bucket_nr, int rw) 1981 { 1982 struct bch_fs *c = trans->c; 1983 struct btree_iter iter; 1984 struct bkey_i_alloc_v4 *a; 1985 u64 now; 1986 int ret = 0; 1987 1988 a = bch2_trans_start_alloc_update(trans, &iter, POS(dev, bucket_nr)); 1989 ret = PTR_ERR_OR_ZERO(a); 1990 if (ret) 1991 return ret; 1992 1993 now = atomic64_read(&c->io_clock[rw].now); 1994 if (a->v.io_time[rw] == now) 1995 goto out; 1996 1997 a->v.io_time[rw] = now; 1998 1999 ret = bch2_trans_update(trans, &iter, &a->k_i, 0) ?: 2000 bch2_trans_commit(trans, NULL, NULL, 0); 2001 out: 2002 bch2_trans_iter_exit(trans, &iter); 2003 return ret; 2004 } 2005 2006 /* Startup/shutdown (ro/rw): */ 2007 2008 void bch2_recalc_capacity(struct bch_fs *c) 2009 { 2010 struct bch_dev *ca; 2011 u64 capacity = 0, reserved_sectors = 0, gc_reserve; 2012 unsigned bucket_size_max = 0; 2013 unsigned long ra_pages = 0; 2014 unsigned i; 2015 2016 lockdep_assert_held(&c->state_lock); 2017 2018 for_each_online_member(ca, c, i) { 2019 struct backing_dev_info *bdi = ca->disk_sb.bdev->bd_disk->bdi; 2020 2021 ra_pages += bdi->ra_pages; 2022 } 2023 2024 bch2_set_ra_pages(c, ra_pages); 2025 2026 for_each_rw_member(ca, c, i) { 2027 u64 dev_reserve = 0; 2028 2029 /* 2030 * We need to reserve buckets (from the number 2031 * of currently available buckets) against 2032 * foreground writes so that mainly copygc can 2033 * make forward progress. 2034 * 2035 * We need enough to refill the various reserves 2036 * from scratch - copygc will use its entire 2037 * reserve all at once, then run against when 2038 * its reserve is refilled (from the formerly 2039 * available buckets). 2040 * 2041 * This reserve is just used when considering if 2042 * allocations for foreground writes must wait - 2043 * not -ENOSPC calculations. 2044 */ 2045 2046 dev_reserve += ca->nr_btree_reserve * 2; 2047 dev_reserve += ca->mi.nbuckets >> 6; /* copygc reserve */ 2048 2049 dev_reserve += 1; /* btree write point */ 2050 dev_reserve += 1; /* copygc write point */ 2051 dev_reserve += 1; /* rebalance write point */ 2052 2053 dev_reserve *= ca->mi.bucket_size; 2054 2055 capacity += bucket_to_sector(ca, ca->mi.nbuckets - 2056 ca->mi.first_bucket); 2057 2058 reserved_sectors += dev_reserve * 2; 2059 2060 bucket_size_max = max_t(unsigned, bucket_size_max, 2061 ca->mi.bucket_size); 2062 } 2063 2064 gc_reserve = c->opts.gc_reserve_bytes 2065 ? c->opts.gc_reserve_bytes >> 9 2066 : div64_u64(capacity * c->opts.gc_reserve_percent, 100); 2067 2068 reserved_sectors = max(gc_reserve, reserved_sectors); 2069 2070 reserved_sectors = min(reserved_sectors, capacity); 2071 2072 c->capacity = capacity - reserved_sectors; 2073 2074 c->bucket_size_max = bucket_size_max; 2075 2076 /* Wake up case someone was waiting for buckets */ 2077 closure_wake_up(&c->freelist_wait); 2078 } 2079 2080 u64 bch2_min_rw_member_capacity(struct bch_fs *c) 2081 { 2082 struct bch_dev *ca; 2083 unsigned i; 2084 u64 ret = U64_MAX; 2085 2086 for_each_rw_member(ca, c, i) 2087 ret = min(ret, ca->mi.nbuckets * ca->mi.bucket_size); 2088 return ret; 2089 } 2090 2091 static bool bch2_dev_has_open_write_point(struct bch_fs *c, struct bch_dev *ca) 2092 { 2093 struct open_bucket *ob; 2094 bool ret = false; 2095 2096 for (ob = c->open_buckets; 2097 ob < c->open_buckets + ARRAY_SIZE(c->open_buckets); 2098 ob++) { 2099 spin_lock(&ob->lock); 2100 if (ob->valid && !ob->on_partial_list && 2101 ob->dev == ca->dev_idx) 2102 ret = true; 2103 spin_unlock(&ob->lock); 2104 } 2105 2106 return ret; 2107 } 2108 2109 /* device goes ro: */ 2110 void bch2_dev_allocator_remove(struct bch_fs *c, struct bch_dev *ca) 2111 { 2112 unsigned i; 2113 2114 /* First, remove device from allocation groups: */ 2115 2116 for (i = 0; i < ARRAY_SIZE(c->rw_devs); i++) 2117 clear_bit(ca->dev_idx, c->rw_devs[i].d); 2118 2119 /* 2120 * Capacity is calculated based off of devices in allocation groups: 2121 */ 2122 bch2_recalc_capacity(c); 2123 2124 bch2_open_buckets_stop(c, ca, false); 2125 2126 /* 2127 * Wake up threads that were blocked on allocation, so they can notice 2128 * the device can no longer be removed and the capacity has changed: 2129 */ 2130 closure_wake_up(&c->freelist_wait); 2131 2132 /* 2133 * journal_res_get() can block waiting for free space in the journal - 2134 * it needs to notice there may not be devices to allocate from anymore: 2135 */ 2136 wake_up(&c->journal.wait); 2137 2138 /* Now wait for any in flight writes: */ 2139 2140 closure_wait_event(&c->open_buckets_wait, 2141 !bch2_dev_has_open_write_point(c, ca)); 2142 } 2143 2144 /* device goes rw: */ 2145 void bch2_dev_allocator_add(struct bch_fs *c, struct bch_dev *ca) 2146 { 2147 unsigned i; 2148 2149 for (i = 0; i < ARRAY_SIZE(c->rw_devs); i++) 2150 if (ca->mi.data_allowed & (1 << i)) 2151 set_bit(ca->dev_idx, c->rw_devs[i].d); 2152 } 2153 2154 void bch2_fs_allocator_background_init(struct bch_fs *c) 2155 { 2156 spin_lock_init(&c->freelist_lock); 2157 INIT_WORK(&c->discard_work, bch2_do_discards_work); 2158 INIT_WORK(&c->invalidate_work, bch2_do_invalidates_work); 2159 } 2160