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

   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                                         BCH_TRANS_COMMIT_no_enospc|
 565                                         BCH_TRANS_COMMIT_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                                 BCH_TRANS_COMMIT_no_enospc|
 585                                 BCH_TRANS_COMMIT_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                         BCH_TRANS_COMMIT_no_enospc|BCH_TRANS_COMMIT_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                                         BCH_TRANS_COMMIT_no_enospc|
1426                                         BCH_TRANS_COMMIT_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, BCH_TRANS_COMMIT_no_enospc|BCH_TRANS_COMMIT_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, BCH_TRANS_COMMIT_no_enospc|BCH_TRANS_COMMIT_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                                   BCH_TRANS_COMMIT_no_enospc);
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                                   BCH_TRANS_COMMIT_no_enospc);
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                                                   BCH_TRANS_COMMIT_lazy_rw|
1888                                                   BCH_TRANS_COMMIT_no_enospc);
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                                                   BCH_TRANS_COMMIT_lazy_rw|
1909                                                   BCH_TRANS_COMMIT_no_enospc);
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 }