fs/btrfs/extent_io.c

   1 // SPDX-License-Identifier: GPL-2.0
   2
   3 #include <linux/bitops.h>
   4 #include <linux/slab.h>
   5 #include <linux/bio.h>
   6 #include <linux/mm.h>
   7 #include <linux/pagemap.h>
   8 #include <linux/page-flags.h>
   9 #include <linux/spinlock.h>
  10 #include <linux/blkdev.h>
  11 #include <linux/swap.h>
  12 #include <linux/writeback.h>
  13 #include <linux/pagevec.h>
  14 #include <linux/prefetch.h>
  15 #include <linux/cleancache.h>
  16 #include "extent_io.h"
  17 #include "extent_map.h"
  18 #include "ctree.h"
  19 #include "btrfs_inode.h"
  20 #include "volumes.h"
  21 #include "check-integrity.h"
  22 #include "locking.h"
  23 #include "rcu-string.h"
  24 #include "backref.h"
  25 #include "disk-io.h"
  26
  27 static struct kmem_cache *extent_state_cache;
  28 static struct kmem_cache *extent_buffer_cache;
  29 static struct bio_set *btrfs_bioset;
  30
  31 static inline bool extent_state_in_tree(const struct extent_state *state)
  32 {
  33         return !RB_EMPTY_NODE(&state->rb_node);
  34 }
  35
  36 #ifdef CONFIG_BTRFS_DEBUG
  37 static LIST_HEAD(buffers);
  38 static LIST_HEAD(states);
  39
  40 static DEFINE_SPINLOCK(leak_lock);
  41
  42 static inline
  43 void btrfs_leak_debug_add(struct list_head *new, struct list_head *head)
  44 {
  45         unsigned long flags;
  46
  47         spin_lock_irqsave(&leak_lock, flags);
  48         list_add(new, head);
  49         spin_unlock_irqrestore(&leak_lock, flags);
  50 }
  51
  52 static inline
  53 void btrfs_leak_debug_del(struct list_head *entry)
  54 {
  55         unsigned long flags;
  56
  57         spin_lock_irqsave(&leak_lock, flags);
  58         list_del(entry);
  59         spin_unlock_irqrestore(&leak_lock, flags);
  60 }
  61
  62 static inline
  63 void btrfs_leak_debug_check(void)
  64 {
  65         struct extent_state *state;
  66         struct extent_buffer *eb;
  67
  68         while (!list_empty(&states)) {
  69                 state = list_entry(states.next, struct extent_state, leak_list);
  70                 pr_err("BTRFS: state leak: start %llu end %llu state %u in tree %d refs %d\n",
  71                        state->start, state->end, state->state,
  72                        extent_state_in_tree(state),
  73                        refcount_read(&state->refs));
  74                 list_del(&state->leak_list);
  75                 kmem_cache_free(extent_state_cache, state);
  76         }
  77
  78         while (!list_empty(&buffers)) {
  79                 eb = list_entry(buffers.next, struct extent_buffer, leak_list);
  80                 pr_err("BTRFS: buffer leak start %llu len %lu refs %d bflags %lu\n",
  81                        eb->start, eb->len, atomic_read(&eb->refs), eb->bflags);
  82                 list_del(&eb->leak_list);
  83                 kmem_cache_free(extent_buffer_cache, eb);
  84         }
  85 }
  86
  87 #define btrfs_debug_check_extent_io_range(tree, start, end)             \
  88         __btrfs_debug_check_extent_io_range(__func__, (tree), (start), (end))
  89 static inline void __btrfs_debug_check_extent_io_range(const char *caller,
  90                 struct extent_io_tree *tree, u64 start, u64 end)
  91 {
  92         if (tree->ops && tree->ops->check_extent_io_range)
  93                 tree->ops->check_extent_io_range(tree->private_data, caller,
  94                                                  start, end);
  95 }
  96 #else
  97 #define btrfs_leak_debug_add(new, head) do {} while (0)
  98 #define btrfs_leak_debug_del(entry)     do {} while (0)
  99 #define btrfs_leak_debug_check()        do {} while (0)
 100 #define btrfs_debug_check_extent_io_range(c, s, e)      do {} while (0)
 101 #endif
 102
 103 #define BUFFER_LRU_MAX 64
 104
 105 struct tree_entry {
 106         u64 start;
 107         u64 end;
 108         struct rb_node rb_node;
 109 };
 110
 111 struct extent_page_data {
 112         struct bio *bio;
 113         struct extent_io_tree *tree;
 114         /* tells writepage not to lock the state bits for this range
 115          * it still does the unlocking
 116          */
 117         unsigned int extent_locked:1;
 118
 119         /* tells the submit_bio code to use REQ_SYNC */
 120         unsigned int sync_io:1;
 121 };
 122
 123 static int add_extent_changeset(struct extent_state *state, unsigned bits,
 124                                  struct extent_changeset *changeset,
 125                                  int set)
 126 {
 127         int ret;
 128
 129         if (!changeset)
 130                 return 0;
 131         if (set && (state->state & bits) == bits)
 132                 return 0;
 133         if (!set && (state->state & bits) == 0)
 134                 return 0;
 135         changeset->bytes_changed += state->end - state->start + 1;
 136         ret = ulist_add(&changeset->range_changed, state->start, state->end,
 137                         GFP_ATOMIC);
 138         return ret;
 139 }
 140
 141 static void flush_write_bio(struct extent_page_data *epd);
 142
 143 static inline struct btrfs_fs_info *
 144 tree_fs_info(struct extent_io_tree *tree)
 145 {
 146         if (tree->ops)
 147                 return tree->ops->tree_fs_info(tree->private_data);
 148         return NULL;
 149 }
 150
 151 int __init extent_io_init(void)
 152 {
 153         extent_state_cache = kmem_cache_create("btrfs_extent_state",
 154                         sizeof(struct extent_state), 0,
 155                         SLAB_MEM_SPREAD, NULL);
 156         if (!extent_state_cache)
 157                 return -ENOMEM;
 158
 159         extent_buffer_cache = kmem_cache_create("btrfs_extent_buffer",
 160                         sizeof(struct extent_buffer), 0,
 161                         SLAB_MEM_SPREAD, NULL);
 162         if (!extent_buffer_cache)
 163                 goto free_state_cache;
 164
 165         btrfs_bioset = bioset_create(BIO_POOL_SIZE,
 166                                      offsetof(struct btrfs_io_bio, bio),
 167                                      BIOSET_NEED_BVECS);
 168         if (!btrfs_bioset)
 169                 goto free_buffer_cache;
 170
 171         if (bioset_integrity_create(btrfs_bioset, BIO_POOL_SIZE))
 172                 goto free_bioset;
 173
 174         return 0;
 175
 176 free_bioset:
 177         bioset_free(btrfs_bioset);
 178         btrfs_bioset = NULL;
 179
 180 free_buffer_cache:
 181         kmem_cache_destroy(extent_buffer_cache);
 182         extent_buffer_cache = NULL;
 183
 184 free_state_cache:
 185         kmem_cache_destroy(extent_state_cache);
 186         extent_state_cache = NULL;
 187         return -ENOMEM;
 188 }
 189
 190 void __cold extent_io_exit(void)
 191 {
 192         btrfs_leak_debug_check();
 193
 194         /*
 195          * Make sure all delayed rcu free are flushed before we
 196          * destroy caches.
 197          */
 198         rcu_barrier();
 199         kmem_cache_destroy(extent_state_cache);
 200         kmem_cache_destroy(extent_buffer_cache);
 201         if (btrfs_bioset)
 202                 bioset_free(btrfs_bioset);
 203 }
 204
 205 void extent_io_tree_init(struct extent_io_tree *tree,
 206                          void *private_data)
 207 {
 208         tree->state = RB_ROOT;
 209         tree->ops = NULL;
 210         tree->dirty_bytes = 0;
 211         spin_lock_init(&tree->lock);
 212         tree->private_data = private_data;
 213 }
 214
 215 static struct extent_state *alloc_extent_state(gfp_t mask)
 216 {
 217         struct extent_state *state;
 218
 219         /*
 220          * The given mask might be not appropriate for the slab allocator,
 221          * drop the unsupported bits
 222          */
 223         mask &= ~(__GFP_DMA32|__GFP_HIGHMEM);
 224         state = kmem_cache_alloc(extent_state_cache, mask);
 225         if (!state)
 226                 return state;
 227         state->state = 0;
 228         state->failrec = NULL;
 229         RB_CLEAR_NODE(&state->rb_node);
 230         btrfs_leak_debug_add(&state->leak_list, &states);
 231         refcount_set(&state->refs, 1);
 232         init_waitqueue_head(&state->wq);
 233         trace_alloc_extent_state(state, mask, _RET_IP_);
 234         return state;
 235 }
 236
 237 void free_extent_state(struct extent_state *state)
 238 {
 239         if (!state)
 240                 return;
 241         if (refcount_dec_and_test(&state->refs)) {
 242                 WARN_ON(extent_state_in_tree(state));
 243                 btrfs_leak_debug_del(&state->leak_list);
 244                 trace_free_extent_state(state, _RET_IP_);
 245                 kmem_cache_free(extent_state_cache, state);
 246         }
 247 }
 248
 249 static struct rb_node *tree_insert(struct rb_root *root,
 250                                    struct rb_node *search_start,
 251                                    u64 offset,
 252                                    struct rb_node *node,
 253                                    struct rb_node ***p_in,
 254                                    struct rb_node **parent_in)
 255 {
 256         struct rb_node **p;
 257         struct rb_node *parent = NULL;
 258         struct tree_entry *entry;
 259
 260         if (p_in && parent_in) {
 261                 p = *p_in;
 262                 parent = *parent_in;
 263                 goto do_insert;
 264         }
 265
 266         p = search_start ? &search_start : &root->rb_node;
 267         while (*p) {
 268                 parent = *p;
 269                 entry = rb_entry(parent, struct tree_entry, rb_node);
 270
 271                 if (offset < entry->start)
 272                         p = &(*p)->rb_left;
 273                 else if (offset > entry->end)
 274                         p = &(*p)->rb_right;
 275                 else
 276                         return parent;
 277         }
 278
 279 do_insert:
 280         rb_link_node(node, parent, p);
 281         rb_insert_color(node, root);
 282         return NULL;
 283 }
 284
 285 static struct rb_node *__etree_search(struct extent_io_tree *tree, u64 offset,
 286                                       struct rb_node **prev_ret,
 287                                       struct rb_node **next_ret,
 288                                       struct rb_node ***p_ret,
 289                                       struct rb_node **parent_ret)
 290 {
 291         struct rb_root *root = &tree->state;
 292         struct rb_node **n = &root->rb_node;
 293         struct rb_node *prev = NULL;
 294         struct rb_node *orig_prev = NULL;
 295         struct tree_entry *entry;
 296         struct tree_entry *prev_entry = NULL;
 297
 298         while (*n) {
 299                 prev = *n;
 300                 entry = rb_entry(prev, struct tree_entry, rb_node);
 301                 prev_entry = entry;
 302
 303                 if (offset < entry->start)
 304                         n = &(*n)->rb_left;
 305                 else if (offset > entry->end)
 306                         n = &(*n)->rb_right;
 307                 else
 308                         return *n;
 309         }
 310
 311         if (p_ret)
 312                 *p_ret = n;
 313         if (parent_ret)
 314                 *parent_ret = prev;
 315
 316         if (prev_ret) {
 317                 orig_prev = prev;
 318                 while (prev && offset > prev_entry->end) {
 319                         prev = rb_next(prev);
 320                         prev_entry = rb_entry(prev, struct tree_entry, rb_node);
 321                 }
 322                 *prev_ret = prev;
 323                 prev = orig_prev;
 324         }
 325
 326         if (next_ret) {
 327                 prev_entry = rb_entry(prev, struct tree_entry, rb_node);
 328                 while (prev && offset < prev_entry->start) {
 329                         prev = rb_prev(prev);
 330                         prev_entry = rb_entry(prev, struct tree_entry, rb_node);
 331                 }
 332                 *next_ret = prev;
 333         }
 334         return NULL;
 335 }
 336
 337 static inline struct rb_node *
 338 tree_search_for_insert(struct extent_io_tree *tree,
 339                        u64 offset,
 340                        struct rb_node ***p_ret,
 341                        struct rb_node **parent_ret)
 342 {
 343         struct rb_node *prev = NULL;
 344         struct rb_node *ret;
 345
 346         ret = __etree_search(tree, offset, &prev, NULL, p_ret, parent_ret);
 347         if (!ret)
 348                 return prev;
 349         return ret;
 350 }
 351
 352 static inline struct rb_node *tree_search(struct extent_io_tree *tree,
 353                                           u64 offset)
 354 {
 355         return tree_search_for_insert(tree, offset, NULL, NULL);
 356 }
 357
 358 static void merge_cb(struct extent_io_tree *tree, struct extent_state *new,
 359                      struct extent_state *other)
 360 {
 361         if (tree->ops && tree->ops->merge_extent_hook)
 362                 tree->ops->merge_extent_hook(tree->private_data, new, other);
 363 }
 364
 365 /*
 366  * utility function to look for merge candidates inside a given range.
 367  * Any extents with matching state are merged together into a single
 368  * extent in the tree.  Extents with EXTENT_IO in their state field
 369  * are not merged because the end_io handlers need to be able to do
 370  * operations on them without sleeping (or doing allocations/splits).
 371  *
 372  * This should be called with the tree lock held.
 373  */
 374 static void merge_state(struct extent_io_tree *tree,
 375                         struct extent_state *state)
 376 {
 377         struct extent_state *other;
 378         struct rb_node *other_node;
 379
 380         if (state->state & (EXTENT_IOBITS | EXTENT_BOUNDARY))
 381                 return;
 382
 383         other_node = rb_prev(&state->rb_node);
 384         if (other_node) {
 385                 other = rb_entry(other_node, struct extent_state, rb_node);
 386                 if (other->end == state->start - 1 &&
 387                     other->state == state->state) {
 388                         merge_cb(tree, state, other);
 389                         state->start = other->start;
 390                         rb_erase(&other->rb_node, &tree->state);
 391                         RB_CLEAR_NODE(&other->rb_node);
 392                         free_extent_state(other);
 393                 }
 394         }
 395         other_node = rb_next(&state->rb_node);
 396         if (other_node) {
 397                 other = rb_entry(other_node, struct extent_state, rb_node);
 398                 if (other->start == state->end + 1 &&
 399                     other->state == state->state) {
 400                         merge_cb(tree, state, other);
 401                         state->end = other->end;
 402                         rb_erase(&other->rb_node, &tree->state);
 403                         RB_CLEAR_NODE(&other->rb_node);
 404                         free_extent_state(other);
 405                 }
 406         }
 407 }
 408
 409 static void set_state_cb(struct extent_io_tree *tree,
 410                          struct extent_state *state, unsigned *bits)
 411 {
 412         if (tree->ops && tree->ops->set_bit_hook)
 413                 tree->ops->set_bit_hook(tree->private_data, state, bits);
 414 }
 415
 416 static void clear_state_cb(struct extent_io_tree *tree,
 417                            struct extent_state *state, unsigned *bits)
 418 {
 419         if (tree->ops && tree->ops->clear_bit_hook)
 420                 tree->ops->clear_bit_hook(tree->private_data, state, bits);
 421 }
 422
 423 static void set_state_bits(struct extent_io_tree *tree,
 424                            struct extent_state *state, unsigned *bits,
 425                            struct extent_changeset *changeset);
 426
 427 /*
 428  * insert an extent_state struct into the tree.  'bits' are set on the
 429  * struct before it is inserted.
 430  *
 431  * This may return -EEXIST if the extent is already there, in which case the
 432  * state struct is freed.
 433  *
 434  * The tree lock is not taken internally.  This is a utility function and
 435  * probably isn't what you want to call (see set/clear_extent_bit).
 436  */
 437 static int insert_state(struct extent_io_tree *tree,
 438                         struct extent_state *state, u64 start, u64 end,
 439                         struct rb_node ***p,
 440                         struct rb_node **parent,
 441                         unsigned *bits, struct extent_changeset *changeset)
 442 {
 443         struct rb_node *node;
 444
 445         if (end < start)
 446                 WARN(1, KERN_ERR "BTRFS: end < start %llu %llu\n",
 447                        end, start);
 448         state->start = start;
 449         state->end = end;
 450
 451         set_state_bits(tree, state, bits, changeset);
 452
 453         node = tree_insert(&tree->state, NULL, end, &state->rb_node, p, parent);
 454         if (node) {
 455                 struct extent_state *found;
 456                 found = rb_entry(node, struct extent_state, rb_node);
 457                 pr_err("BTRFS: found node %llu %llu on insert of %llu %llu\n",
 458                        found->start, found->end, start, end);
 459                 return -EEXIST;
 460         }
 461         merge_state(tree, state);
 462         return 0;
 463 }
 464
 465 static void split_cb(struct extent_io_tree *tree, struct extent_state *orig,
 466                      u64 split)
 467 {
 468         if (tree->ops && tree->ops->split_extent_hook)
 469                 tree->ops->split_extent_hook(tree->private_data, orig, split);
 470 }
 471
 472 /*
 473  * split a given extent state struct in two, inserting the preallocated
 474  * struct 'prealloc' as the newly created second half.  'split' indicates an
 475  * offset inside 'orig' where it should be split.
 476  *
 477  * Before calling,
 478  * the tree has 'orig' at [orig->start, orig->end].  After calling, there
 479  * are two extent state structs in the tree:
 480  * prealloc: [orig->start, split - 1]
 481  * orig: [ split, orig->end ]
 482  *
 483  * The tree locks are not taken by this function. They need to be held
 484  * by the caller.
 485  */
 486 static int split_state(struct extent_io_tree *tree, struct extent_state *orig,
 487                        struct extent_state *prealloc, u64 split)
 488 {
 489         struct rb_node *node;
 490
 491         split_cb(tree, orig, split);
 492
 493         prealloc->start = orig->start;
 494         prealloc->end = split - 1;
 495         prealloc->state = orig->state;
 496         orig->start = split;
 497
 498         node = tree_insert(&tree->state, &orig->rb_node, prealloc->end,
 499                            &prealloc->rb_node, NULL, NULL);
 500         if (node) {
 501                 free_extent_state(prealloc);
 502                 return -EEXIST;
 503         }
 504         return 0;
 505 }
 506
 507 static struct extent_state *next_state(struct extent_state *state)
 508 {
 509         struct rb_node *next = rb_next(&state->rb_node);
 510         if (next)
 511                 return rb_entry(next, struct extent_state, rb_node);
 512         else
 513                 return NULL;
 514 }
 515
 516 /*
 517  * utility function to clear some bits in an extent state struct.
 518  * it will optionally wake up any one waiting on this state (wake == 1).
 519  *
 520  * If no bits are set on the state struct after clearing things, the
 521  * struct is freed and removed from the tree
 522  */
 523 static struct extent_state *clear_state_bit(struct extent_io_tree *tree,
 524                                             struct extent_state *state,
 525                                             unsigned *bits, int wake,
 526                                             struct extent_changeset *changeset)
 527 {
 528         struct extent_state *next;
 529         unsigned bits_to_clear = *bits & ~EXTENT_CTLBITS;
 530         int ret;
 531
 532         if ((bits_to_clear & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) {
 533                 u64 range = state->end - state->start + 1;
 534                 WARN_ON(range > tree->dirty_bytes);
 535                 tree->dirty_bytes -= range;
 536         }
 537         clear_state_cb(tree, state, bits);
 538         ret = add_extent_changeset(state, bits_to_clear, changeset, 0);
 539         BUG_ON(ret < 0);
 540         state->state &= ~bits_to_clear;
 541         if (wake)
 542                 wake_up(&state->wq);
 543         if (state->state == 0) {
 544                 next = next_state(state);
 545                 if (extent_state_in_tree(state)) {
 546                         rb_erase(&state->rb_node, &tree->state);
 547                         RB_CLEAR_NODE(&state->rb_node);
 548                         free_extent_state(state);
 549                 } else {
 550                         WARN_ON(1);
 551                 }
 552         } else {
 553                 merge_state(tree, state);
 554                 next = next_state(state);
 555         }
 556         return next;
 557 }
 558
 559 static struct extent_state *
 560 alloc_extent_state_atomic(struct extent_state *prealloc)
 561 {
 562         if (!prealloc)
 563                 prealloc = alloc_extent_state(GFP_ATOMIC);
 564
 565         return prealloc;
 566 }
 567
 568 static void extent_io_tree_panic(struct extent_io_tree *tree, int err)
 569 {
 570         btrfs_panic(tree_fs_info(tree), err,
 571                     "Locking error: Extent tree was modified by another thread while locked.");
 572 }
 573
 574 /*
 575  * clear some bits on a range in the tree.  This may require splitting
 576  * or inserting elements in the tree, so the gfp mask is used to
 577  * indicate which allocations or sleeping are allowed.
 578  *
 579  * pass 'wake' == 1 to kick any sleepers, and 'delete' == 1 to remove
 580  * the given range from the tree regardless of state (ie for truncate).
 581  *
 582  * the range [start, end] is inclusive.
 583  *
 584  * This takes the tree lock, and returns 0 on success and < 0 on error.
 585  */
 586 int __clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 587                               unsigned bits, int wake, int delete,
 588                               struct extent_state **cached_state,
 589                               gfp_t mask, struct extent_changeset *changeset)
 590 {
 591         struct extent_state *state;
 592         struct extent_state *cached;
 593         struct extent_state *prealloc = NULL;
 594         struct rb_node *node;
 595         u64 last_end;
 596         int err;
 597         int clear = 0;
 598
 599         btrfs_debug_check_extent_io_range(tree, start, end);
 600
 601         if (bits & EXTENT_DELALLOC)
 602                 bits |= EXTENT_NORESERVE;
 603
 604         if (delete)
 605                 bits |= ~EXTENT_CTLBITS;
 606         bits |= EXTENT_FIRST_DELALLOC;
 607
 608         if (bits & (EXTENT_IOBITS | EXTENT_BOUNDARY))
 609                 clear = 1;
 610 again:
 611         if (!prealloc && gfpflags_allow_blocking(mask)) {
 612                 /*
 613                  * Don't care for allocation failure here because we might end
 614                  * up not needing the pre-allocated extent state at all, which
 615                  * is the case if we only have in the tree extent states that
 616                  * cover our input range and don't cover too any other range.
 617                  * If we end up needing a new extent state we allocate it later.
 618                  */
 619                 prealloc = alloc_extent_state(mask);
 620         }
 621
 622         spin_lock(&tree->lock);
 623         if (cached_state) {
 624                 cached = *cached_state;
 625
 626                 if (clear) {
 627                         *cached_state = NULL;
 628                         cached_state = NULL;
 629                 }
 630
 631                 if (cached && extent_state_in_tree(cached) &&
 632                     cached->start <= start && cached->end > start) {
 633                         if (clear)
 634                                 refcount_dec(&cached->refs);
 635                         state = cached;
 636                         goto hit_next;
 637                 }
 638                 if (clear)
 639                         free_extent_state(cached);
 640         }
 641         /*
 642          * this search will find the extents that end after
 643          * our range starts
 644          */
 645         node = tree_search(tree, start);
 646         if (!node)
 647                 goto out;
 648         state = rb_entry(node, struct extent_state, rb_node);
 649 hit_next:
 650         if (state->start > end)
 651                 goto out;
 652         WARN_ON(state->end < start);
 653         last_end = state->end;
 654
 655         /* the state doesn't have the wanted bits, go ahead */
 656         if (!(state->state & bits)) {
 657                 state = next_state(state);
 658                 goto next;
 659         }
 660
 661         /*
 662          *     | ---- desired range ---- |
 663          *  | state | or
 664          *  | ------------- state -------------- |
 665          *
 666          * We need to split the extent we found, and may flip
 667          * bits on second half.
 668          *
 669          * If the extent we found extends past our range, we
 670          * just split and search again.  It'll get split again
 671          * the next time though.
 672          *
 673          * If the extent we found is inside our range, we clear
 674          * the desired bit on it.
 675          */
 676
 677         if (state->start < start) {
 678                 prealloc = alloc_extent_state_atomic(prealloc);
 679                 BUG_ON(!prealloc);
 680                 err = split_state(tree, state, prealloc, start);
 681                 if (err)
 682                         extent_io_tree_panic(tree, err);
 683
 684                 prealloc = NULL;
 685                 if (err)
 686                         goto out;
 687                 if (state->end <= end) {
 688                         state = clear_state_bit(tree, state, &bits, wake,
 689                                                 changeset);
 690                         goto next;
 691                 }
 692                 goto search_again;
 693         }
 694         /*
 695          * | ---- desired range ---- |
 696          *                        | state |
 697          * We need to split the extent, and clear the bit
 698          * on the first half
 699          */
 700         if (state->start <= end && state->end > end) {
 701                 prealloc = alloc_extent_state_atomic(prealloc);
 702                 BUG_ON(!prealloc);
 703                 err = split_state(tree, state, prealloc, end + 1);
 704                 if (err)
 705                         extent_io_tree_panic(tree, err);
 706
 707                 if (wake)
 708                         wake_up(&state->wq);
 709
 710                 clear_state_bit(tree, prealloc, &bits, wake, changeset);
 711
 712                 prealloc = NULL;
 713                 goto out;
 714         }
 715
 716         state = clear_state_bit(tree, state, &bits, wake, changeset);
 717 next:
 718         if (last_end == (u64)-1)
 719                 goto out;
 720         start = last_end + 1;
 721         if (start <= end && state && !need_resched())
 722                 goto hit_next;
 723
 724 search_again:
 725         if (start > end)
 726                 goto out;
 727         spin_unlock(&tree->lock);
 728         if (gfpflags_allow_blocking(mask))
 729                 cond_resched();
 730         goto again;
 731
 732 out:
 733         spin_unlock(&tree->lock);
 734         if (prealloc)
 735                 free_extent_state(prealloc);
 736
 737         return 0;
 738
 739 }
 740
 741 static void wait_on_state(struct extent_io_tree *tree,
 742                           struct extent_state *state)
 743                 __releases(tree->lock)
 744                 __acquires(tree->lock)
 745 {
 746         DEFINE_WAIT(wait);
 747         prepare_to_wait(&state->wq, &wait, TASK_UNINTERRUPTIBLE);
 748         spin_unlock(&tree->lock);
 749         schedule();
 750         spin_lock(&tree->lock);
 751         finish_wait(&state->wq, &wait);
 752 }
 753
 754 /*
 755  * waits for one or more bits to clear on a range in the state tree.
 756  * The range [start, end] is inclusive.
 757  * The tree lock is taken by this function
 758  */
 759 static void wait_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 760                             unsigned long bits)
 761 {
 762         struct extent_state *state;
 763         struct rb_node *node;
 764
 765         btrfs_debug_check_extent_io_range(tree, start, end);
 766
 767         spin_lock(&tree->lock);
 768 again:
 769         while (1) {
 770                 /*
 771                  * this search will find all the extents that end after
 772                  * our range starts
 773                  */
 774                 node = tree_search(tree, start);
 775 process_node:
 776                 if (!node)
 777                         break;
 778
 779                 state = rb_entry(node, struct extent_state, rb_node);
 780
 781                 if (state->start > end)
 782                         goto out;
 783
 784                 if (state->state & bits) {
 785                         start = state->start;
 786                         refcount_inc(&state->refs);
 787                         wait_on_state(tree, state);
 788                         free_extent_state(state);
 789                         goto again;
 790                 }
 791                 start = state->end + 1;
 792
 793                 if (start > end)
 794                         break;
 795
 796                 if (!cond_resched_lock(&tree->lock)) {
 797                         node = rb_next(node);
 798                         goto process_node;
 799                 }
 800         }
 801 out:
 802         spin_unlock(&tree->lock);
 803 }
 804
 805 static void set_state_bits(struct extent_io_tree *tree,
 806                            struct extent_state *state,
 807                            unsigned *bits, struct extent_changeset *changeset)
 808 {
 809         unsigned bits_to_set = *bits & ~EXTENT_CTLBITS;
 810         int ret;
 811
 812         set_state_cb(tree, state, bits);
 813         if ((bits_to_set & EXTENT_DIRTY) && !(state->state & EXTENT_DIRTY)) {
 814                 u64 range = state->end - state->start + 1;
 815                 tree->dirty_bytes += range;
 816         }
 817         ret = add_extent_changeset(state, bits_to_set, changeset, 1);
 818         BUG_ON(ret < 0);
 819         state->state |= bits_to_set;
 820 }
 821
 822 static void cache_state_if_flags(struct extent_state *state,
 823                                  struct extent_state **cached_ptr,
 824                                  unsigned flags)
 825 {
 826         if (cached_ptr && !(*cached_ptr)) {
 827                 if (!flags || (state->state & flags)) {
 828                         *cached_ptr = state;
 829                         refcount_inc(&state->refs);
 830                 }
 831         }
 832 }
 833
 834 static void cache_state(struct extent_state *state,
 835                         struct extent_state **cached_ptr)
 836 {
 837         return cache_state_if_flags(state, cached_ptr,
 838                                     EXTENT_IOBITS | EXTENT_BOUNDARY);
 839 }
 840
 841 /*
 842  * set some bits on a range in the tree.  This may require allocations or
 843  * sleeping, so the gfp mask is used to indicate what is allowed.
 844  *
 845  * If any of the exclusive bits are set, this will fail with -EEXIST if some
 846  * part of the range already has the desired bits set.  The start of the
 847  * existing range is returned in failed_start in this case.
 848  *
 849  * [start, end] is inclusive This takes the tree lock.
 850  */
 851
 852 static int __must_check
 853 __set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 854                  unsigned bits, unsigned exclusive_bits,
 855                  u64 *failed_start, struct extent_state **cached_state,
 856                  gfp_t mask, struct extent_changeset *changeset)
 857 {
 858         struct extent_state *state;
 859         struct extent_state *prealloc = NULL;
 860         struct rb_node *node;
 861         struct rb_node **p;
 862         struct rb_node *parent;
 863         int err = 0;
 864         u64 last_start;
 865         u64 last_end;
 866
 867         btrfs_debug_check_extent_io_range(tree, start, end);
 868
 869         bits |= EXTENT_FIRST_DELALLOC;
 870 again:
 871         if (!prealloc && gfpflags_allow_blocking(mask)) {
 872                 /*
 873                  * Don't care for allocation failure here because we might end
 874                  * up not needing the pre-allocated extent state at all, which
 875                  * is the case if we only have in the tree extent states that
 876                  * cover our input range and don't cover too any other range.
 877                  * If we end up needing a new extent state we allocate it later.
 878                  */
 879                 prealloc = alloc_extent_state(mask);
 880         }
 881
 882         spin_lock(&tree->lock);
 883         if (cached_state && *cached_state) {
 884                 state = *cached_state;
 885                 if (state->start <= start && state->end > start &&
 886                     extent_state_in_tree(state)) {
 887                         node = &state->rb_node;
 888                         goto hit_next;
 889                 }
 890         }
 891         /*
 892          * this search will find all the extents that end after
 893          * our range starts.
 894          */
 895         node = tree_search_for_insert(tree, start, &p, &parent);
 896         if (!node) {
 897                 prealloc = alloc_extent_state_atomic(prealloc);
 898                 BUG_ON(!prealloc);
 899                 err = insert_state(tree, prealloc, start, end,
 900                                    &p, &parent, &bits, changeset);
 901                 if (err)
 902                         extent_io_tree_panic(tree, err);
 903
 904                 cache_state(prealloc, cached_state);
 905                 prealloc = NULL;
 906                 goto out;
 907         }
 908         state = rb_entry(node, struct extent_state, rb_node);
 909 hit_next:
 910         last_start = state->start;
 911         last_end = state->end;
 912
 913         /*
 914          * | ---- desired range ---- |
 915          * | state |
 916          *
 917          * Just lock what we found and keep going
 918          */
 919         if (state->start == start && state->end <= end) {
 920                 if (state->state & exclusive_bits) {
 921                         *failed_start = state->start;
 922                         err = -EEXIST;
 923                         goto out;
 924                 }
 925
 926                 set_state_bits(tree, state, &bits, changeset);
 927                 cache_state(state, cached_state);
 928                 merge_state(tree, state);
 929                 if (last_end == (u64)-1)
 930                         goto out;
 931                 start = last_end + 1;
 932                 state = next_state(state);
 933                 if (start < end && state && state->start == start &&
 934                     !need_resched())
 935                         goto hit_next;
 936                 goto search_again;
 937         }
 938
 939         /*
 940          *     | ---- desired range ---- |
 941          * | state |
 942          *   or
 943          * | ------------- state -------------- |
 944          *
 945          * We need to split the extent we found, and may flip bits on
 946          * second half.
 947          *
 948          * If the extent we found extends past our
 949          * range, we just split and search again.  It'll get split
 950          * again the next time though.
 951          *
 952          * If the extent we found is inside our range, we set the
 953          * desired bit on it.
 954          */
 955         if (state->start < start) {
 956                 if (state->state & exclusive_bits) {
 957                         *failed_start = start;
 958                         err = -EEXIST;
 959                         goto out;
 960                 }
 961
 962                 prealloc = alloc_extent_state_atomic(prealloc);
 963                 BUG_ON(!prealloc);
 964                 err = split_state(tree, state, prealloc, start);
 965                 if (err)
 966                         extent_io_tree_panic(tree, err);
 967
 968                 prealloc = NULL;
 969                 if (err)
 970                         goto out;
 971                 if (state->end <= end) {
 972                         set_state_bits(tree, state, &bits, changeset);
 973                         cache_state(state, cached_state);
 974                         merge_state(tree, state);
 975                         if (last_end == (u64)-1)
 976                                 goto out;
 977                         start = last_end + 1;
 978                         state = next_state(state);
 979                         if (start < end && state && state->start == start &&
 980                             !need_resched())
 981                                 goto hit_next;
 982                 }
 983                 goto search_again;
 984         }
 985         /*
 986          * | ---- desired range ---- |
 987          *     | state | or               | state |
 988          *
 989          * There's a hole, we need to insert something in it and
 990          * ignore the extent we found.
 991          */
 992         if (state->start > start) {
 993                 u64 this_end;
 994                 if (end < last_start)
 995                         this_end = end;
 996                 else
 997                         this_end = last_start - 1;
 998
 999                 prealloc = alloc_extent_state_atomic(prealloc);
1000                 BUG_ON(!prealloc);
1001
1002                 /*
1003                  * Avoid to free 'prealloc' if it can be merged with
1004                  * the later extent.
1005                  */
1006                 err = insert_state(tree, prealloc, start, this_end,
1007                                    NULL, NULL, &bits, changeset);
1008                 if (err)
1009                         extent_io_tree_panic(tree, err);
1010
1011                 cache_state(prealloc, cached_state);
1012                 prealloc = NULL;
1013                 start = this_end + 1;
1014                 goto search_again;
1015         }
1016         /*
1017          * | ---- desired range ---- |
1018          *                        | state |
1019          * We need to split the extent, and set the bit
1020          * on the first half
1021          */
1022         if (state->start <= end && state->end > end) {
1023                 if (state->state & exclusive_bits) {
1024                         *failed_start = start;
1025                         err = -EEXIST;
1026                         goto out;
1027                 }
1028
1029                 prealloc = alloc_extent_state_atomic(prealloc);
1030                 BUG_ON(!prealloc);
1031                 err = split_state(tree, state, prealloc, end + 1);
1032                 if (err)
1033                         extent_io_tree_panic(tree, err);
1034
1035                 set_state_bits(tree, prealloc, &bits, changeset);
1036                 cache_state(prealloc, cached_state);
1037                 merge_state(tree, prealloc);
1038                 prealloc = NULL;
1039                 goto out;
1040         }
1041
1042 search_again:
1043         if (start > end)
1044                 goto out;
1045         spin_unlock(&tree->lock);
1046         if (gfpflags_allow_blocking(mask))
1047                 cond_resched();
1048         goto again;
1049
1050 out:
1051         spin_unlock(&tree->lock);
1052         if (prealloc)
1053                 free_extent_state(prealloc);
1054
1055         return err;
1056
1057 }
1058
1059 int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
1060                    unsigned bits, u64 * failed_start,
1061                    struct extent_state **cached_state, gfp_t mask)
1062 {
1063         return __set_extent_bit(tree, start, end, bits, 0, failed_start,
1064                                 cached_state, mask, NULL);
1065 }
1066
1067
1068 /**
1069  * convert_extent_bit - convert all bits in a given range from one bit to
1070  *                      another
1071  * @tree:       the io tree to search
1072  * @start:      the start offset in bytes
1073  * @end:        the end offset in bytes (inclusive)
1074  * @bits:       the bits to set in this range
1075  * @clear_bits: the bits to clear in this range
1076  * @cached_state:       state that we're going to cache
1077  *
1078  * This will go through and set bits for the given range.  If any states exist
1079  * already in this range they are set with the given bit and cleared of the
1080  * clear_bits.  This is only meant to be used by things that are mergeable, ie
1081  * converting from say DELALLOC to DIRTY.  This is not meant to be used with
1082  * boundary bits like LOCK.
1083  *
1084  * All allocations are done with GFP_NOFS.
1085  */
1086 int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
1087                        unsigned bits, unsigned clear_bits,
1088                        struct extent_state **cached_state)
1089 {
1090         struct extent_state *state;
1091         struct extent_state *prealloc = NULL;
1092         struct rb_node *node;
1093         struct rb_node **p;
1094         struct rb_node *parent;
1095         int err = 0;
1096         u64 last_start;
1097         u64 last_end;
1098         bool first_iteration = true;
1099
1100         btrfs_debug_check_extent_io_range(tree, start, end);
1101
1102 again:
1103         if (!prealloc) {
1104                 /*
1105                  * Best effort, don't worry if extent state allocation fails
1106                  * here for the first iteration. We might have a cached state
1107                  * that matches exactly the target range, in which case no
1108                  * extent state allocations are needed. We'll only know this
1109                  * after locking the tree.
1110                  */
1111                 prealloc = alloc_extent_state(GFP_NOFS);
1112                 if (!prealloc && !first_iteration)
1113                         return -ENOMEM;
1114         }
1115
1116         spin_lock(&tree->lock);
1117         if (cached_state && *cached_state) {
1118                 state = *cached_state;
1119                 if (state->start <= start && state->end > start &&
1120                     extent_state_in_tree(state)) {
1121                         node = &state->rb_node;
1122                         goto hit_next;
1123                 }
1124         }
1125
1126         /*
1127          * this search will find all the extents that end after
1128          * our range starts.
1129          */
1130         node = tree_search_for_insert(tree, start, &p, &parent);
1131         if (!node) {
1132                 prealloc = alloc_extent_state_atomic(prealloc);
1133                 if (!prealloc) {
1134                         err = -ENOMEM;
1135                         goto out;
1136                 }
1137                 err = insert_state(tree, prealloc, start, end,
1138                                    &p, &parent, &bits, NULL);
1139                 if (err)
1140                         extent_io_tree_panic(tree, err);
1141                 cache_state(prealloc, cached_state);
1142                 prealloc = NULL;
1143                 goto out;
1144         }
1145         state = rb_entry(node, struct extent_state, rb_node);
1146 hit_next:
1147         last_start = state->start;
1148         last_end = state->end;
1149
1150         /*
1151          * | ---- desired range ---- |
1152          * | state |
1153          *
1154          * Just lock what we found and keep going
1155          */
1156         if (state->start == start && state->end <= end) {
1157                 set_state_bits(tree, state, &bits, NULL);
1158                 cache_state(state, cached_state);
1159                 state = clear_state_bit(tree, state, &clear_bits, 0, NULL);
1160                 if (last_end == (u64)-1)
1161                         goto out;
1162                 start = last_end + 1;
1163                 if (start < end && state && state->start == start &&
1164                     !need_resched())
1165                         goto hit_next;
1166                 goto search_again;
1167         }
1168
1169         /*
1170          *     | ---- desired range ---- |
1171          * | state |
1172          *   or
1173          * | ------------- state -------------- |
1174          *
1175          * We need to split the extent we found, and may flip bits on
1176          * second half.
1177          *
1178          * If the extent we found extends past our
1179          * range, we just split and search again.  It'll get split
1180          * again the next time though.
1181          *
1182          * If the extent we found is inside our range, we set the
1183          * desired bit on it.
1184          */
1185         if (state->start < start) {
1186                 prealloc = alloc_extent_state_atomic(prealloc);
1187                 if (!prealloc) {
1188                         err = -ENOMEM;
1189                         goto out;
1190                 }
1191                 err = split_state(tree, state, prealloc, start);
1192                 if (err)
1193                         extent_io_tree_panic(tree, err);
1194                 prealloc = NULL;
1195                 if (err)
1196                         goto out;
1197                 if (state->end <= end) {
1198                         set_state_bits(tree, state, &bits, NULL);
1199                         cache_state(state, cached_state);
1200                         state = clear_state_bit(tree, state, &clear_bits, 0,
1201                                                 NULL);
1202                         if (last_end == (u64)-1)
1203                                 goto out;
1204                         start = last_end + 1;
1205                         if (start < end && state && state->start == start &&
1206                             !need_resched())
1207                                 goto hit_next;
1208                 }
1209                 goto search_again;
1210         }
1211         /*
1212          * | ---- desired range ---- |
1213          *     | state | or               | state |
1214          *
1215          * There's a hole, we need to insert something in it and
1216          * ignore the extent we found.
1217          */
1218         if (state->start > start) {
1219                 u64 this_end;
1220                 if (end < last_start)
1221                         this_end = end;
1222                 else
1223                         this_end = last_start - 1;
1224
1225                 prealloc = alloc_extent_state_atomic(prealloc);
1226                 if (!prealloc) {
1227                         err = -ENOMEM;
1228                         goto out;
1229                 }
1230
1231                 /*
1232                  * Avoid to free 'prealloc' if it can be merged with
1233                  * the later extent.
1234                  */
1235                 err = insert_state(tree, prealloc, start, this_end,
1236                                    NULL, NULL, &bits, NULL);
1237                 if (err)
1238                         extent_io_tree_panic(tree, err);
1239                 cache_state(prealloc, cached_state);
1240                 prealloc = NULL;
1241                 start = this_end + 1;
1242                 goto search_again;
1243         }
1244         /*
1245          * | ---- desired range ---- |
1246          *                        | state |
1247          * We need to split the extent, and set the bit
1248          * on the first half
1249          */
1250         if (state->start <= end && state->end > end) {
1251                 prealloc = alloc_extent_state_atomic(prealloc);
1252                 if (!prealloc) {
1253                         err = -ENOMEM;
1254                         goto out;
1255                 }
1256
1257                 err = split_state(tree, state, prealloc, end + 1);
1258                 if (err)
1259                         extent_io_tree_panic(tree, err);
1260
1261                 set_state_bits(tree, prealloc, &bits, NULL);
1262                 cache_state(prealloc, cached_state);
1263                 clear_state_bit(tree, prealloc, &clear_bits, 0, NULL);
1264                 prealloc = NULL;
1265                 goto out;
1266         }
1267
1268 search_again:
1269         if (start > end)
1270                 goto out;
1271         spin_unlock(&tree->lock);
1272         cond_resched();
1273         first_iteration = false;
1274         goto again;
1275
1276 out:
1277         spin_unlock(&tree->lock);
1278         if (prealloc)
1279                 free_extent_state(prealloc);
1280
1281         return err;
1282 }
1283
1284 /* wrappers around set/clear extent bit */
1285 int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
1286                            unsigned bits, struct extent_changeset *changeset)
1287 {
1288         /*
1289          * We don't support EXTENT_LOCKED yet, as current changeset will
1290          * record any bits changed, so for EXTENT_LOCKED case, it will
1291          * either fail with -EEXIST or changeset will record the whole
1292          * range.
1293          */
1294         BUG_ON(bits & EXTENT_LOCKED);
1295
1296         return __set_extent_bit(tree, start, end, bits, 0, NULL, NULL, GFP_NOFS,
1297                                 changeset);
1298 }
1299
1300 int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
1301                      unsigned bits, int wake, int delete,
1302                      struct extent_state **cached)
1303 {
1304         return __clear_extent_bit(tree, start, end, bits, wake, delete,
1305                                   cached, GFP_NOFS, NULL);
1306 }
1307
1308 int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
1309                 unsigned bits, struct extent_changeset *changeset)
1310 {
1311         /*
1312          * Don't support EXTENT_LOCKED case, same reason as
1313          * set_record_extent_bits().
1314          */
1315         BUG_ON(bits & EXTENT_LOCKED);
1316
1317         return __clear_extent_bit(tree, start, end, bits, 0, 0, NULL, GFP_NOFS,
1318                                   changeset);
1319 }
1320
1321 /*
1322  * either insert or lock state struct between start and end use mask to tell
1323  * us if waiting is desired.
1324  */
1325 int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
1326                      struct extent_state **cached_state)
1327 {
1328         int err;
1329         u64 failed_start;
1330
1331         while (1) {
1332                 err = __set_extent_bit(tree, start, end, EXTENT_LOCKED,
1333                                        EXTENT_LOCKED, &failed_start,
1334                                        cached_state, GFP_NOFS, NULL);
1335                 if (err == -EEXIST) {
1336                         wait_extent_bit(tree, failed_start, end, EXTENT_LOCKED);
1337                         start = failed_start;
1338                 } else
1339                         break;
1340                 WARN_ON(start > end);
1341         }
1342         return err;
1343 }
1344
1345 int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
1346 {
1347         int err;
1348         u64 failed_start;
1349
1350         err = __set_extent_bit(tree, start, end, EXTENT_LOCKED, EXTENT_LOCKED,
1351                                &failed_start, NULL, GFP_NOFS, NULL);
1352         if (err == -EEXIST) {
1353                 if (failed_start > start)
1354                         clear_extent_bit(tree, start, failed_start - 1,
1355                                          EXTENT_LOCKED, 1, 0, NULL);
1356                 return 0;
1357         }
1358         return 1;
1359 }
1360
1361 void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end)
1362 {
1363         unsigned long index = start >> PAGE_SHIFT;
1364         unsigned long end_index = end >> PAGE_SHIFT;
1365         struct page *page;
1366
1367         while (index <= end_index) {
1368                 page = find_get_page(inode->i_mapping, index);
1369                 BUG_ON(!page); /* Pages should be in the extent_io_tree */
1370                 clear_page_dirty_for_io(page);
1371                 put_page(page);
1372                 index++;
1373         }
1374 }
1375
1376 void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end)
1377 {
1378         unsigned long index = start >> PAGE_SHIFT;
1379         unsigned long end_index = end >> PAGE_SHIFT;
1380         struct page *page;
1381
1382         while (index <= end_index) {
1383                 page = find_get_page(inode->i_mapping, index);
1384                 BUG_ON(!page); /* Pages should be in the extent_io_tree */
1385                 __set_page_dirty_nobuffers(page);
1386                 account_page_redirty(page);
1387                 put_page(page);
1388                 index++;
1389         }
1390 }
1391
1392 /*
1393  * helper function to set both pages and extents in the tree writeback
1394  */
1395 static void set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end)
1396 {
1397         tree->ops->set_range_writeback(tree->private_data, start, end);
1398 }
1399
1400 /* find the first state struct with 'bits' set after 'start', and
1401  * return it.  tree->lock must be held.  NULL will returned if
1402  * nothing was found after 'start'
1403  */
1404 static struct extent_state *
1405 find_first_extent_bit_state(struct extent_io_tree *tree,
1406                             u64 start, unsigned bits)
1407 {
1408         struct rb_node *node;
1409         struct extent_state *state;
1410
1411         /*
1412          * this search will find all the extents that end after
1413          * our range starts.
1414          */
1415         node = tree_search(tree, start);
1416         if (!node)
1417                 goto out;
1418
1419         while (1) {
1420                 state = rb_entry(node, struct extent_state, rb_node);
1421                 if (state->end >= start && (state->state & bits))
1422                         return state;
1423
1424                 node = rb_next(node);
1425                 if (!node)
1426                         break;
1427         }
1428 out:
1429         return NULL;
1430 }
1431
1432 /*
1433  * find the first offset in the io tree with 'bits' set. zero is
1434  * returned if we find something, and *start_ret and *end_ret are
1435  * set to reflect the state struct that was found.
1436  *
1437  * If nothing was found, 1 is returned. If found something, return 0.
1438  */
1439 int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
1440                           u64 *start_ret, u64 *end_ret, unsigned bits,
1441                           struct extent_state **cached_state)
1442 {
1443         struct extent_state *state;
1444         struct rb_node *n;
1445         int ret = 1;
1446
1447         spin_lock(&tree->lock);
1448         if (cached_state && *cached_state) {
1449                 state = *cached_state;
1450                 if (state->end == start - 1 && extent_state_in_tree(state)) {
1451                         n = rb_next(&state->rb_node);
1452                         while (n) {
1453                                 state = rb_entry(n, struct extent_state,
1454                                                  rb_node);
1455                                 if (state->state & bits)
1456                                         goto got_it;
1457                                 n = rb_next(n);
1458                         }
1459                         free_extent_state(*cached_state);
1460                         *cached_state = NULL;
1461                         goto out;
1462                 }
1463                 free_extent_state(*cached_state);
1464                 *cached_state = NULL;
1465         }
1466
1467         state = find_first_extent_bit_state(tree, start, bits);
1468 got_it:
1469         if (state) {
1470                 cache_state_if_flags(state, cached_state, 0);
1471                 *start_ret = state->start;
1472                 *end_ret = state->end;
1473                 ret = 0;
1474         }
1475 out:
1476         spin_unlock(&tree->lock);
1477         return ret;
1478 }
1479
1480 /*
1481  * find a contiguous range of bytes in the file marked as delalloc, not
1482  * more than 'max_bytes'.  start and end are used to return the range,
1483  *
1484  * 1 is returned if we find something, 0 if nothing was in the tree
1485  */
1486 static noinline u64 find_delalloc_range(struct extent_io_tree *tree,
1487                                         u64 *start, u64 *end, u64 max_bytes,
1488                                         struct extent_state **cached_state)
1489 {
1490         struct rb_node *node;
1491         struct extent_state *state;
1492         u64 cur_start = *start;
1493         u64 found = 0;
1494         u64 total_bytes = 0;
1495
1496         spin_lock(&tree->lock);
1497
1498         /*
1499          * this search will find all the extents that end after
1500          * our range starts.
1501          */
1502         node = tree_search(tree, cur_start);
1503         if (!node) {
1504                 if (!found)
1505                         *end = (u64)-1;
1506                 goto out;
1507         }
1508
1509         while (1) {
1510                 state = rb_entry(node, struct extent_state, rb_node);
1511                 if (found && (state->start != cur_start ||
1512                               (state->state & EXTENT_BOUNDARY))) {
1513                         goto out;
1514                 }
1515                 if (!(state->state & EXTENT_DELALLOC)) {
1516                         if (!found)
1517                                 *end = state->end;
1518                         goto out;
1519                 }
1520                 if (!found) {
1521                         *start = state->start;
1522                         *cached_state = state;
1523                         refcount_inc(&state->refs);
1524                 }
1525                 found++;
1526                 *end = state->end;
1527                 cur_start = state->end + 1;
1528                 node = rb_next(node);
1529                 total_bytes += state->end - state->start + 1;
1530                 if (total_bytes >= max_bytes)
1531                         break;
1532                 if (!node)
1533                         break;
1534         }
1535 out:
1536         spin_unlock(&tree->lock);
1537         return found;
1538 }
1539
1540 static int __process_pages_contig(struct address_space *mapping,
1541                                   struct page *locked_page,
1542                                   pgoff_t start_index, pgoff_t end_index,
1543                                   unsigned long page_ops, pgoff_t *index_ret);
1544
1545 static noinline void __unlock_for_delalloc(struct inode *inode,
1546                                            struct page *locked_page,
1547                                            u64 start, u64 end)
1548 {
1549         unsigned long index = start >> PAGE_SHIFT;
1550         unsigned long end_index = end >> PAGE_SHIFT;
1551
1552         ASSERT(locked_page);
1553         if (index == locked_page->index && end_index == index)
1554                 return;
1555
1556         __process_pages_contig(inode->i_mapping, locked_page, index, end_index,
1557                                PAGE_UNLOCK, NULL);
1558 }
1559
1560 static noinline int lock_delalloc_pages(struct inode *inode,
1561                                         struct page *locked_page,
1562                                         u64 delalloc_start,
1563                                         u64 delalloc_end)
1564 {
1565         unsigned long index = delalloc_start >> PAGE_SHIFT;
1566         unsigned long index_ret = index;
1567         unsigned long end_index = delalloc_end >> PAGE_SHIFT;
1568         int ret;
1569
1570         ASSERT(locked_page);
1571         if (index == locked_page->index && index == end_index)
1572                 return 0;
1573
1574         ret = __process_pages_contig(inode->i_mapping, locked_page, index,
1575                                      end_index, PAGE_LOCK, &index_ret);
1576         if (ret == -EAGAIN)
1577                 __unlock_for_delalloc(inode, locked_page, delalloc_start,
1578                                       (u64)index_ret << PAGE_SHIFT);
1579         return ret;
1580 }
1581
1582 /*
1583  * find a contiguous range of bytes in the file marked as delalloc, not
1584  * more than 'max_bytes'.  start and end are used to return the range,
1585  *
1586  * 1 is returned if we find something, 0 if nothing was in the tree
1587  */
1588 STATIC u64 find_lock_delalloc_range(struct inode *inode,
1589                                     struct extent_io_tree *tree,
1590                                     struct page *locked_page, u64 *start,
1591                                     u64 *end, u64 max_bytes)
1592 {
1593         u64 delalloc_start;
1594         u64 delalloc_end;
1595         u64 found;
1596         struct extent_state *cached_state = NULL;
1597         int ret;
1598         int loops = 0;
1599
1600 again:
1601         /* step one, find a bunch of delalloc bytes starting at start */
1602         delalloc_start = *start;
1603         delalloc_end = 0;
1604         found = find_delalloc_range(tree, &delalloc_start, &delalloc_end,
1605                                     max_bytes, &cached_state);
1606         if (!found || delalloc_end <= *start) {
1607                 *start = delalloc_start;
1608                 *end = delalloc_end;
1609                 free_extent_state(cached_state);
1610                 return 0;
1611         }
1612
1613         /*
1614          * start comes from the offset of locked_page.  We have to lock
1615          * pages in order, so we can't process delalloc bytes before
1616          * locked_page
1617          */
1618         if (delalloc_start < *start)
1619                 delalloc_start = *start;
1620
1621         /*
1622          * make sure to limit the number of pages we try to lock down
1623          */
1624         if (delalloc_end + 1 - delalloc_start > max_bytes)
1625                 delalloc_end = delalloc_start + max_bytes - 1;
1626
1627         /* step two, lock all the pages after the page that has start */
1628         ret = lock_delalloc_pages(inode, locked_page,
1629                                   delalloc_start, delalloc_end);
1630         if (ret == -EAGAIN) {
1631                 /* some of the pages are gone, lets avoid looping by
1632                  * shortening the size of the delalloc range we're searching
1633                  */
1634                 free_extent_state(cached_state);
1635                 cached_state = NULL;
1636                 if (!loops) {
1637                         max_bytes = PAGE_SIZE;
1638                         loops = 1;
1639                         goto again;
1640                 } else {
1641                         found = 0;
1642                         goto out_failed;
1643                 }
1644         }
1645         BUG_ON(ret); /* Only valid values are 0 and -EAGAIN */
1646
1647         /* step three, lock the state bits for the whole range */
1648         lock_extent_bits(tree, delalloc_start, delalloc_end, &cached_state);
1649
1650         /* then test to make sure it is all still delalloc */
1651         ret = test_range_bit(tree, delalloc_start, delalloc_end,
1652                              EXTENT_DELALLOC, 1, cached_state);
1653         if (!ret) {
1654                 unlock_extent_cached(tree, delalloc_start, delalloc_end,
1655                                      &cached_state);
1656                 __unlock_for_delalloc(inode, locked_page,
1657                               delalloc_start, delalloc_end);
1658                 cond_resched();
1659                 goto again;
1660         }
1661         free_extent_state(cached_state);
1662         *start = delalloc_start;
1663         *end = delalloc_end;
1664 out_failed:
1665         return found;
1666 }
1667
1668 static int __process_pages_contig(struct address_space *mapping,
1669                                   struct page *locked_page,
1670                                   pgoff_t start_index, pgoff_t end_index,
1671                                   unsigned long page_ops, pgoff_t *index_ret)
1672 {
1673         unsigned long nr_pages = end_index - start_index + 1;
1674         unsigned long pages_locked = 0;
1675         pgoff_t index = start_index;
1676         struct page *pages[16];
1677         unsigned ret;
1678         int err = 0;
1679         int i;
1680
1681         if (page_ops & PAGE_LOCK) {
1682                 ASSERT(page_ops == PAGE_LOCK);
1683                 ASSERT(index_ret && *index_ret == start_index);
1684         }
1685
1686         if ((page_ops & PAGE_SET_ERROR) && nr_pages > 0)
1687                 mapping_set_error(mapping, -EIO);
1688
1689         while (nr_pages > 0) {
1690                 ret = find_get_pages_contig(mapping, index,
1691                                      min_t(unsigned long,
1692                                      nr_pages, ARRAY_SIZE(pages)), pages);
1693                 if (ret == 0) {
1694                         /*
1695                          * Only if we're going to lock these pages,
1696                          * can we find nothing at @index.
1697                          */
1698                         ASSERT(page_ops & PAGE_LOCK);
1699                         err = -EAGAIN;
1700                         goto out;
1701                 }
1702
1703                 for (i = 0; i < ret; i++) {
1704                         if (page_ops & PAGE_SET_PRIVATE2)
1705                                 SetPagePrivate2(pages[i]);
1706
1707                         if (pages[i] == locked_page) {
1708                                 put_page(pages[i]);
1709                                 pages_locked++;
1710                                 continue;
1711                         }
1712                         if (page_ops & PAGE_CLEAR_DIRTY)
1713                                 clear_page_dirty_for_io(pages[i]);
1714                         if (page_ops & PAGE_SET_WRITEBACK)
1715                                 set_page_writeback(pages[i]);
1716                         if (page_ops & PAGE_SET_ERROR)
1717                                 SetPageError(pages[i]);
1718                         if (page_ops & PAGE_END_WRITEBACK)
1719                                 end_page_writeback(pages[i]);
1720                         if (page_ops & PAGE_UNLOCK)
1721                                 unlock_page(pages[i]);
1722                         if (page_ops & PAGE_LOCK) {
1723                                 lock_page(pages[i]);
1724                                 if (!PageDirty(pages[i]) ||
1725                                     pages[i]->mapping != mapping) {
1726                                         unlock_page(pages[i]);
1727                                         put_page(pages[i]);
1728                                         err = -EAGAIN;
1729                                         goto out;
1730                                 }
1731                         }
1732                         put_page(pages[i]);
1733                         pages_locked++;
1734                 }
1735                 nr_pages -= ret;
1736                 index += ret;
1737                 cond_resched();
1738         }
1739 out:
1740         if (err && index_ret)
1741                 *index_ret = start_index + pages_locked - 1;
1742         return err;
1743 }
1744
1745 void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
1746                                  u64 delalloc_end, struct page *locked_page,
1747                                  unsigned clear_bits,
1748                                  unsigned long page_ops)
1749 {
1750         clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, clear_bits, 1, 0,
1751                          NULL);
1752
1753         __process_pages_contig(inode->i_mapping, locked_page,
1754                                start >> PAGE_SHIFT, end >> PAGE_SHIFT,
1755                                page_ops, NULL);
1756 }
1757
1758 /*
1759  * count the number of bytes in the tree that have a given bit(s)
1760  * set.  This can be fairly slow, except for EXTENT_DIRTY which is
1761  * cached.  The total number found is returned.
1762  */
1763 u64 count_range_bits(struct extent_io_tree *tree,
1764                      u64 *start, u64 search_end, u64 max_bytes,
1765                      unsigned bits, int contig)
1766 {
1767         struct rb_node *node;
1768         struct extent_state *state;
1769         u64 cur_start = *start;
1770         u64 total_bytes = 0;
1771         u64 last = 0;
1772         int found = 0;
1773
1774         if (WARN_ON(search_end <= cur_start))
1775                 return 0;
1776
1777         spin_lock(&tree->lock);
1778         if (cur_start == 0 && bits == EXTENT_DIRTY) {
1779                 total_bytes = tree->dirty_bytes;
1780                 goto out;
1781         }
1782         /*
1783          * this search will find all the extents that end after
1784          * our range starts.
1785          */
1786         node = tree_search(tree, cur_start);
1787         if (!node)
1788                 goto out;
1789
1790         while (1) {
1791                 state = rb_entry(node, struct extent_state, rb_node);
1792                 if (state->start > search_end)
1793                         break;
1794                 if (contig && found && state->start > last + 1)
1795                         break;
1796                 if (state->end >= cur_start && (state->state & bits) == bits) {
1797                         total_bytes += min(search_end, state->end) + 1 -
1798                                        max(cur_start, state->start);
1799                         if (total_bytes >= max_bytes)
1800                                 break;
1801                         if (!found) {
1802                                 *start = max(cur_start, state->start);
1803                                 found = 1;
1804                         }
1805                         last = state->end;
1806                 } else if (contig && found) {
1807                         break;
1808                 }
1809                 node = rb_next(node);
1810                 if (!node)
1811                         break;
1812         }
1813 out:
1814         spin_unlock(&tree->lock);
1815         return total_bytes;
1816 }
1817
1818 /*
1819  * set the private field for a given byte offset in the tree.  If there isn't
1820  * an extent_state there already, this does nothing.
1821  */
1822 static noinline int set_state_failrec(struct extent_io_tree *tree, u64 start,
1823                 struct io_failure_record *failrec)
1824 {
1825         struct rb_node *node;
1826         struct extent_state *state;
1827         int ret = 0;
1828
1829         spin_lock(&tree->lock);
1830         /*
1831          * this search will find all the extents that end after
1832          * our range starts.
1833          */
1834         node = tree_search(tree, start);
1835         if (!node) {
1836                 ret = -ENOENT;
1837                 goto out;
1838         }
1839         state = rb_entry(node, struct extent_state, rb_node);
1840         if (state->start != start) {
1841                 ret = -ENOENT;
1842                 goto out;
1843         }
1844         state->failrec = failrec;
1845 out:
1846         spin_unlock(&tree->lock);
1847         return ret;
1848 }
1849
1850 static noinline int get_state_failrec(struct extent_io_tree *tree, u64 start,
1851                 struct io_failure_record **failrec)
1852 {
1853         struct rb_node *node;
1854         struct extent_state *state;
1855         int ret = 0;
1856
1857         spin_lock(&tree->lock);
1858         /*
1859          * this search will find all the extents that end after
1860          * our range starts.
1861          */
1862         node = tree_search(tree, start);
1863         if (!node) {
1864                 ret = -ENOENT;
1865                 goto out;
1866         }
1867         state = rb_entry(node, struct extent_state, rb_node);
1868         if (state->start != start) {
1869                 ret = -ENOENT;
1870                 goto out;
1871         }
1872         *failrec = state->failrec;
1873 out:
1874         spin_unlock(&tree->lock);
1875         return ret;
1876 }
1877
1878 /*
1879  * searches a range in the state tree for a given mask.
1880  * If 'filled' == 1, this returns 1 only if every extent in the tree
1881  * has the bits set.  Otherwise, 1 is returned if any bit in the
1882  * range is found set.
1883  */
1884 int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
1885                    unsigned bits, int filled, struct extent_state *cached)
1886 {
1887         struct extent_state *state = NULL;
1888         struct rb_node *node;
1889         int bitset = 0;
1890
1891         spin_lock(&tree->lock);
1892         if (cached && extent_state_in_tree(cached) && cached->start <= start &&
1893             cached->end > start)
1894                 node = &cached->rb_node;
1895         else
1896                 node = tree_search(tree, start);
1897         while (node && start <= end) {
1898                 state = rb_entry(node, struct extent_state, rb_node);
1899
1900                 if (filled && state->start > start) {
1901                         bitset = 0;
1902                         break;
1903                 }
1904
1905                 if (state->start > end)
1906                         break;
1907
1908                 if (state->state & bits) {
1909                         bitset = 1;
1910                         if (!filled)
1911                                 break;
1912                 } else if (filled) {
1913                         bitset = 0;
1914                         break;
1915                 }
1916
1917                 if (state->end == (u64)-1)
1918                         break;
1919
1920                 start = state->end + 1;
1921                 if (start > end)
1922                         break;
1923                 node = rb_next(node);
1924                 if (!node) {
1925                         if (filled)
1926                                 bitset = 0;
1927                         break;
1928                 }
1929         }
1930         spin_unlock(&tree->lock);
1931         return bitset;
1932 }
1933
1934 /*
1935  * helper function to set a given page up to date if all the
1936  * extents in the tree for that page are up to date
1937  */
1938 static void check_page_uptodate(struct extent_io_tree *tree, struct page *page)
1939 {
1940         u64 start = page_offset(page);
1941         u64 end = start + PAGE_SIZE - 1;
1942         if (test_range_bit(tree, start, end, EXTENT_UPTODATE, 1, NULL))
1943                 SetPageUptodate(page);
1944 }
1945
1946 int free_io_failure(struct extent_io_tree *failure_tree,
1947                     struct extent_io_tree *io_tree,
1948                     struct io_failure_record *rec)
1949 {
1950         int ret;
1951         int err = 0;
1952
1953         set_state_failrec(failure_tree, rec->start, NULL);
1954         ret = clear_extent_bits(failure_tree, rec->start,
1955                                 rec->start + rec->len - 1,
1956                                 EXTENT_LOCKED | EXTENT_DIRTY);
1957         if (ret)
1958                 err = ret;
1959
1960         ret = clear_extent_bits(io_tree, rec->start,
1961                                 rec->start + rec->len - 1,
1962                                 EXTENT_DAMAGED);
1963         if (ret && !err)
1964                 err = ret;
1965
1966         kfree(rec);
1967         return err;
1968 }
1969
1970 /*
1971  * this bypasses the standard btrfs submit functions deliberately, as
1972  * the standard behavior is to write all copies in a raid setup. here we only
1973  * want to write the one bad copy. so we do the mapping for ourselves and issue
1974  * submit_bio directly.
1975  * to avoid any synchronization issues, wait for the data after writing, which
1976  * actually prevents the read that triggered the error from finishing.
1977  * currently, there can be no more than two copies of every data bit. thus,
1978  * exactly one rewrite is required.
1979  */
1980 int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
1981                       u64 length, u64 logical, struct page *page,
1982                       unsigned int pg_offset, int mirror_num)
1983 {
1984         struct bio *bio;
1985         struct btrfs_device *dev;
1986         u64 map_length = 0;
1987         u64 sector;
1988         struct btrfs_bio *bbio = NULL;
1989         int ret;
1990
1991         ASSERT(!(fs_info->sb->s_flags & SB_RDONLY));
1992         BUG_ON(!mirror_num);
1993
1994         bio = btrfs_io_bio_alloc(1);
1995         bio->bi_iter.bi_size = 0;
1996         map_length = length;
1997
1998         /*
1999          * Avoid races with device replace and make sure our bbio has devices
2000          * associated to its stripes that don't go away while we are doing the
2001          * read repair operation.
2002          */
2003         btrfs_bio_counter_inc_blocked(fs_info);
2004         if (btrfs_is_parity_mirror(fs_info, logical, length)) {
2005                 /*
2006                  * Note that we don't use BTRFS_MAP_WRITE because it's supposed
2007                  * to update all raid stripes, but here we just want to correct
2008                  * bad stripe, thus BTRFS_MAP_READ is abused to only get the bad
2009                  * stripe's dev and sector.
2010                  */
2011                 ret = btrfs_map_block(fs_info, BTRFS_MAP_READ, logical,
2012                                       &map_length, &bbio, 0);
2013                 if (ret) {
2014                         btrfs_bio_counter_dec(fs_info);
2015                         bio_put(bio);
2016                         return -EIO;
2017                 }
2018                 ASSERT(bbio->mirror_num == 1);
2019         } else {
2020                 ret = btrfs_map_block(fs_info, BTRFS_MAP_WRITE, logical,
2021                                       &map_length, &bbio, mirror_num);
2022                 if (ret) {
2023                         btrfs_bio_counter_dec(fs_info);
2024                         bio_put(bio);
2025                         return -EIO;
2026                 }
2027                 BUG_ON(mirror_num != bbio->mirror_num);
2028         }
2029
2030         sector = bbio->stripes[bbio->mirror_num - 1].physical >> 9;
2031         bio->bi_iter.bi_sector = sector;
2032         dev = bbio->stripes[bbio->mirror_num - 1].dev;
2033         btrfs_put_bbio(bbio);
2034         if (!dev || !dev->bdev ||
2035             !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
2036                 btrfs_bio_counter_dec(fs_info);
2037                 bio_put(bio);
2038                 return -EIO;
2039         }
2040         bio_set_dev(bio, dev->bdev);
2041         bio->bi_opf = REQ_OP_WRITE | REQ_SYNC;
2042         bio_add_page(bio, page, length, pg_offset);
2043
2044         if (btrfsic_submit_bio_wait(bio)) {
2045                 /* try to remap that extent elsewhere? */
2046                 btrfs_bio_counter_dec(fs_info);
2047                 bio_put(bio);
2048                 btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS);
2049                 return -EIO;
2050         }
2051
2052         btrfs_info_rl_in_rcu(fs_info,
2053                 "read error corrected: ino %llu off %llu (dev %s sector %llu)",
2054                                   ino, start,
2055                                   rcu_str_deref(dev->name), sector);
2056         btrfs_bio_counter_dec(fs_info);
2057         bio_put(bio);
2058         return 0;
2059 }
2060
2061 int repair_eb_io_failure(struct btrfs_fs_info *fs_info,
2062                          struct extent_buffer *eb, int mirror_num)
2063 {
2064         u64 start = eb->start;
2065         unsigned long i, num_pages = num_extent_pages(eb->start, eb->len);
2066         int ret = 0;
2067
2068         if (sb_rdonly(fs_info->sb))
2069                 return -EROFS;
2070
2071         for (i = 0; i < num_pages; i++) {
2072                 struct page *p = eb->pages[i];
2073
2074                 ret = repair_io_failure(fs_info, 0, start, PAGE_SIZE, start, p,
2075                                         start - page_offset(p), mirror_num);
2076                 if (ret)
2077                         break;
2078                 start += PAGE_SIZE;
2079         }
2080
2081         return ret;
2082 }
2083
2084 /*
2085  * each time an IO finishes, we do a fast check in the IO failure tree
2086  * to see if we need to process or clean up an io_failure_record
2087  */
2088 int clean_io_failure(struct btrfs_fs_info *fs_info,
2089                      struct extent_io_tree *failure_tree,
2090                      struct extent_io_tree *io_tree, u64 start,
2091                      struct page *page, u64 ino, unsigned int pg_offset)
2092 {
2093         u64 private;
2094         struct io_failure_record *failrec;
2095         struct extent_state *state;
2096         int num_copies;
2097         int ret;
2098
2099         private = 0;
2100         ret = count_range_bits(failure_tree, &private, (u64)-1, 1,
2101                                EXTENT_DIRTY, 0);
2102         if (!ret)
2103                 return 0;
2104
2105         ret = get_state_failrec(failure_tree, start, &failrec);
2106         if (ret)
2107                 return 0;
2108
2109         BUG_ON(!failrec->this_mirror);
2110
2111         if (failrec->in_validation) {
2112                 /* there was no real error, just free the record */
2113                 btrfs_debug(fs_info,
2114                         "clean_io_failure: freeing dummy error at %llu",
2115                         failrec->start);
2116                 goto out;
2117         }
2118         if (sb_rdonly(fs_info->sb))
2119                 goto out;
2120
2121         spin_lock(&io_tree->lock);
2122         state = find_first_extent_bit_state(io_tree,
2123                                             failrec->start,
2124                                             EXTENT_LOCKED);
2125         spin_unlock(&io_tree->lock);
2126
2127         if (state && state->start <= failrec->start &&
2128             state->end >= failrec->start + failrec->len - 1) {
2129                 num_copies = btrfs_num_copies(fs_info, failrec->logical,
2130                                               failrec->len);
2131                 if (num_copies > 1)  {
2132                         repair_io_failure(fs_info, ino, start, failrec->len,
2133                                           failrec->logical, page, pg_offset,
2134                                           failrec->failed_mirror);
2135                 }
2136         }
2137
2138 out:
2139         free_io_failure(failure_tree, io_tree, failrec);
2140
2141         return 0;
2142 }
2143
2144 /*
2145  * Can be called when
2146  * - hold extent lock
2147  * - under ordered extent
2148  * - the inode is freeing
2149  */
2150 void btrfs_free_io_failure_record(struct btrfs_inode *inode, u64 start, u64 end)
2151 {
2152         struct extent_io_tree *failure_tree = &inode->io_failure_tree;
2153         struct io_failure_record *failrec;
2154         struct extent_state *state, *next;
2155
2156         if (RB_EMPTY_ROOT(&failure_tree->state))
2157                 return;
2158
2159         spin_lock(&failure_tree->lock);
2160         state = find_first_extent_bit_state(failure_tree, start, EXTENT_DIRTY);
2161         while (state) {
2162                 if (state->start > end)
2163                         break;
2164
2165                 ASSERT(state->end <= end);
2166
2167                 next = next_state(state);
2168
2169                 failrec = state->failrec;
2170                 free_extent_state(state);
2171                 kfree(failrec);
2172
2173                 state = next;
2174         }
2175         spin_unlock(&failure_tree->lock);
2176 }
2177
2178 int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
2179                 struct io_failure_record **failrec_ret)
2180 {
2181         struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
2182         struct io_failure_record *failrec;
2183         struct extent_map *em;
2184         struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
2185         struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
2186         struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
2187         int ret;
2188         u64 logical;
2189
2190         ret = get_state_failrec(failure_tree, start, &failrec);
2191         if (ret) {
2192                 failrec = kzalloc(sizeof(*failrec), GFP_NOFS);
2193                 if (!failrec)
2194                         return -ENOMEM;
2195
2196                 failrec->start = start;
2197                 failrec->len = end - start + 1;
2198                 failrec->this_mirror = 0;
2199                 failrec->bio_flags = 0;
2200                 failrec->in_validation = 0;
2201
2202                 read_lock(&em_tree->lock);
2203                 em = lookup_extent_mapping(em_tree, start, failrec->len);
2204                 if (!em) {
2205                         read_unlock(&em_tree->lock);
2206                         kfree(failrec);
2207                         return -EIO;
2208                 }
2209
2210                 if (em->start > start || em->start + em->len <= start) {
2211                         free_extent_map(em);
2212                         em = NULL;
2213                 }
2214                 read_unlock(&em_tree->lock);
2215                 if (!em) {
2216                         kfree(failrec);
2217                         return -EIO;
2218                 }
2219
2220                 logical = start - em->start;
2221                 logical = em->block_start + logical;
2222                 if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
2223                         logical = em->block_start;
2224                         failrec->bio_flags = EXTENT_BIO_COMPRESSED;
2225                         extent_set_compress_type(&failrec->bio_flags,
2226                                                  em->compress_type);
2227                 }
2228
2229                 btrfs_debug(fs_info,
2230                         "Get IO Failure Record: (new) logical=%llu, start=%llu, len=%llu",
2231                         logical, start, failrec->len);
2232
2233                 failrec->logical = logical;
2234                 free_extent_map(em);
2235
2236                 /* set the bits in the private failure tree */
2237                 ret = set_extent_bits(failure_tree, start, end,
2238                                         EXTENT_LOCKED | EXTENT_DIRTY);
2239                 if (ret >= 0)
2240                         ret = set_state_failrec(failure_tree, start, failrec);
2241                 /* set the bits in the inode's tree */
2242                 if (ret >= 0)
2243                         ret = set_extent_bits(tree, start, end, EXTENT_DAMAGED);
2244                 if (ret < 0) {
2245                         kfree(failrec);
2246                         return ret;
2247                 }
2248         } else {
2249                 btrfs_debug(fs_info,
2250                         "Get IO Failure Record: (found) logical=%llu, start=%llu, len=%llu, validation=%d",
2251                         failrec->logical, failrec->start, failrec->len,
2252                         failrec->in_validation);
2253                 /*
2254                  * when data can be on disk more than twice, add to failrec here
2255                  * (e.g. with a list for failed_mirror) to make
2256                  * clean_io_failure() clean all those errors at once.
2257                  */
2258         }
2259
2260         *failrec_ret = failrec;
2261
2262         return 0;
2263 }
2264
2265 bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
2266                            struct io_failure_record *failrec, int failed_mirror)
2267 {
2268         struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
2269         int num_copies;
2270
2271         num_copies = btrfs_num_copies(fs_info, failrec->logical, failrec->len);
2272         if (num_copies == 1) {
2273                 /*
2274                  * we only have a single copy of the data, so don't bother with
2275                  * all the retry and error correction code that follows. no
2276                  * matter what the error is, it is very likely to persist.
2277                  */
2278                 btrfs_debug(fs_info,
2279                         "Check Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d",
2280                         num_copies, failrec->this_mirror, failed_mirror);
2281                 return false;
2282         }
2283
2284         /*
2285          * there are two premises:
2286          *      a) deliver good data to the caller
2287          *      b) correct the bad sectors on disk
2288          */
2289         if (failed_bio_pages > 1) {
2290                 /*
2291                  * to fulfill b), we need to know the exact failing sectors, as
2292                  * we don't want to rewrite any more than the failed ones. thus,
2293                  * we need separate read requests for the failed bio
2294                  *
2295                  * if the following BUG_ON triggers, our validation request got
2296                  * merged. we need separate requests for our algorithm to work.
2297                  */
2298                 BUG_ON(failrec->in_validation);
2299                 failrec->in_validation = 1;
2300                 failrec->this_mirror = failed_mirror;
2301         } else {
2302                 /*
2303                  * we're ready to fulfill a) and b) alongside. get a good copy
2304                  * of the failed sector and if we succeed, we have setup
2305                  * everything for repair_io_failure to do the rest for us.
2306                  */
2307                 if (failrec->in_validation) {
2308                         BUG_ON(failrec->this_mirror != failed_mirror);
2309                         failrec->in_validation = 0;
2310                         failrec->this_mirror = 0;
2311                 }
2312                 failrec->failed_mirror = failed_mirror;
2313                 failrec->this_mirror++;
2314                 if (failrec->this_mirror == failed_mirror)
2315                         failrec->this_mirror++;
2316         }
2317
2318         if (failrec->this_mirror > num_copies) {
2319                 btrfs_debug(fs_info,
2320                         "Check Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d",
2321                         num_copies, failrec->this_mirror, failed_mirror);
2322                 return false;
2323         }
2324
2325         return true;
2326 }
2327
2328
2329 struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
2330                                     struct io_failure_record *failrec,
2331                                     struct page *page, int pg_offset, int icsum,
2332                                     bio_end_io_t *endio_func, void *data)
2333 {
2334         struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
2335         struct bio *bio;
2336         struct btrfs_io_bio *btrfs_failed_bio;
2337         struct btrfs_io_bio *btrfs_bio;
2338
2339         bio = btrfs_io_bio_alloc(1);
2340         bio->bi_end_io = endio_func;
2341         bio->bi_iter.bi_sector = failrec->logical >> 9;
2342         bio_set_dev(bio, fs_info->fs_devices->latest_bdev);
2343         bio->bi_iter.bi_size = 0;
2344         bio->bi_private = data;
2345
2346         btrfs_failed_bio = btrfs_io_bio(failed_bio);
2347         if (btrfs_failed_bio->csum) {
2348                 u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
2349
2350                 btrfs_bio = btrfs_io_bio(bio);
2351                 btrfs_bio->csum = btrfs_bio->csum_inline;
2352                 icsum *= csum_size;
2353                 memcpy(btrfs_bio->csum, btrfs_failed_bio->csum + icsum,
2354                        csum_size);
2355         }
2356
2357         bio_add_page(bio, page, failrec->len, pg_offset);
2358
2359         return bio;
2360 }
2361
2362 /*
2363  * this is a generic handler for readpage errors (default
2364  * readpage_io_failed_hook). if other copies exist, read those and write back
2365  * good data to the failed position. does not investigate in remapping the
2366  * failed extent elsewhere, hoping the device will be smart enough to do this as
2367  * needed
2368  */
2369
2370 static int bio_readpage_error(struct bio *failed_bio, u64 phy_offset,
2371                               struct page *page, u64 start, u64 end,
2372                               int failed_mirror)
2373 {
2374         struct io_failure_record *failrec;
2375         struct inode *inode = page->mapping->host;
2376         struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
2377         struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
2378         struct bio *bio;
2379         int read_mode = 0;
2380         blk_status_t status;
2381         int ret;
2382         unsigned failed_bio_pages = bio_pages_all(failed_bio);
2383
2384         BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
2385
2386         ret = btrfs_get_io_failure_record(inode, start, end, &failrec);
2387         if (ret)
2388                 return ret;
2389
2390         if (!btrfs_check_repairable(inode, failed_bio_pages, failrec,
2391                                     failed_mirror)) {
2392                 free_io_failure(failure_tree, tree, failrec);
2393                 return -EIO;
2394         }
2395
2396         if (failed_bio_pages > 1)
2397                 read_mode |= REQ_FAILFAST_DEV;
2398
2399         phy_offset >>= inode->i_sb->s_blocksize_bits;
2400         bio = btrfs_create_repair_bio(inode, failed_bio, failrec, page,
2401                                       start - page_offset(page),
2402                                       (int)phy_offset, failed_bio->bi_end_io,
2403                                       NULL);
2404         bio_set_op_attrs(bio, REQ_OP_READ, read_mode);
2405
2406         btrfs_debug(btrfs_sb(inode->i_sb),
2407                 "Repair Read Error: submitting new read[%#x] to this_mirror=%d, in_validation=%d",
2408                 read_mode, failrec->this_mirror, failrec->in_validation);
2409
2410         status = tree->ops->submit_bio_hook(tree->private_data, bio, failrec->this_mirror,
2411                                          failrec->bio_flags, 0);
2412         if (status) {
2413                 free_io_failure(failure_tree, tree, failrec);
2414                 bio_put(bio);
2415                 ret = blk_status_to_errno(status);
2416         }
2417
2418         return ret;
2419 }
2420
2421 /* lots and lots of room for performance fixes in the end_bio funcs */
2422
2423 void end_extent_writepage(struct page *page, int err, u64 start, u64 end)
2424 {
2425         int uptodate = (err == 0);
2426         struct extent_io_tree *tree;
2427         int ret = 0;
2428
2429         tree = &BTRFS_I(page->mapping->host)->io_tree;
2430
2431         if (tree->ops && tree->ops->writepage_end_io_hook)
2432                 tree->ops->writepage_end_io_hook(page, start, end, NULL,
2433                                 uptodate);
2434
2435         if (!uptodate) {
2436                 ClearPageUptodate(page);
2437                 SetPageError(page);
2438                 ret = err < 0 ? err : -EIO;
2439                 mapping_set_error(page->mapping, ret);
2440         }
2441 }
2442
2443 /*
2444  * after a writepage IO is done, we need to:
2445  * clear the uptodate bits on error
2446  * clear the writeback bits in the extent tree for this IO
2447  * end_page_writeback if the page has no more pending IO
2448  *
2449  * Scheduling is not allowed, so the extent state tree is expected
2450  * to have one and only one object corresponding to this IO.
2451  */
2452 static void end_bio_extent_writepage(struct bio *bio)
2453 {
2454         int error = blk_status_to_errno(bio->bi_status);
2455         struct bio_vec *bvec;
2456         u64 start;
2457         u64 end;
2458         int i;
2459
2460         ASSERT(!bio_flagged(bio, BIO_CLONED));
2461         bio_for_each_segment_all(bvec, bio, i) {
2462                 struct page *page = bvec->bv_page;
2463                 struct inode *inode = page->mapping->host;
2464                 struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
2465
2466                 /* We always issue full-page reads, but if some block
2467                  * in a page fails to read, blk_update_request() will
2468                  * advance bv_offset and adjust bv_len to compensate.
2469                  * Print a warning for nonzero offsets, and an error
2470                  * if they don't add up to a full page.  */
2471                 if (bvec->bv_offset || bvec->bv_len != PAGE_SIZE) {
2472                         if (bvec->bv_offset + bvec->bv_len != PAGE_SIZE)
2473                                 btrfs_err(fs_info,
2474                                    "partial page write in btrfs with offset %u and length %u",
2475                                         bvec->bv_offset, bvec->bv_len);
2476                         else
2477                                 btrfs_info(fs_info,
2478                                    "incomplete page write in btrfs with offset %u and length %u",
2479                                         bvec->bv_offset, bvec->bv_len);
2480                 }
2481
2482                 start = page_offset(page);
2483                 end = start + bvec->bv_offset + bvec->bv_len - 1;
2484
2485                 end_extent_writepage(page, error, start, end);
2486                 end_page_writeback(page);
2487         }
2488
2489         bio_put(bio);
2490 }
2491
2492 static void
2493 endio_readpage_release_extent(struct extent_io_tree *tree, u64 start, u64 len,
2494                               int uptodate)
2495 {
2496         struct extent_state *cached = NULL;
2497         u64 end = start + len - 1;
2498
2499         if (uptodate && tree->track_uptodate)
2500                 set_extent_uptodate(tree, start, end, &cached, GFP_ATOMIC);
2501         unlock_extent_cached_atomic(tree, start, end, &cached);
2502 }
2503
2504 /*
2505  * after a readpage IO is done, we need to:
2506  * clear the uptodate bits on error
2507  * set the uptodate bits if things worked
2508  * set the page up to date if all extents in the tree are uptodate
2509  * clear the lock bit in the extent tree
2510  * unlock the page if there are no other extents locked for it
2511  *
2512  * Scheduling is not allowed, so the extent state tree is expected
2513  * to have one and only one object corresponding to this IO.
2514  */
2515 static void end_bio_extent_readpage(struct bio *bio)
2516 {
2517         struct bio_vec *bvec;
2518         int uptodate = !bio->bi_status;
2519         struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
2520         struct extent_io_tree *tree, *failure_tree;
2521         u64 offset = 0;
2522         u64 start;
2523         u64 end;
2524         u64 len;
2525         u64 extent_start = 0;
2526         u64 extent_len = 0;
2527         int mirror;
2528         int ret;
2529         int i;
2530
2531         ASSERT(!bio_flagged(bio, BIO_CLONED));
2532         bio_for_each_segment_all(bvec, bio, i) {
2533                 struct page *page = bvec->bv_page;
2534                 struct inode *inode = page->mapping->host;
2535                 struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
2536
2537                 btrfs_debug(fs_info,
2538                         "end_bio_extent_readpage: bi_sector=%llu, err=%d, mirror=%u",
2539                         (u64)bio->bi_iter.bi_sector, bio->bi_status,
2540                         io_bio->mirror_num);
2541                 tree = &BTRFS_I(inode)->io_tree;
2542                 failure_tree = &BTRFS_I(inode)->io_failure_tree;
2543
2544                 /* We always issue full-page reads, but if some block
2545                  * in a page fails to read, blk_update_request() will
2546                  * advance bv_offset and adjust bv_len to compensate.
2547                  * Print a warning for nonzero offsets, and an error
2548                  * if they don't add up to a full page.  */
2549                 if (bvec->bv_offset || bvec->bv_len != PAGE_SIZE) {
2550                         if (bvec->bv_offset + bvec->bv_len != PAGE_SIZE)
2551                                 btrfs_err(fs_info,
2552                                         "partial page read in btrfs with offset %u and length %u",
2553                                         bvec->bv_offset, bvec->bv_len);
2554                         else
2555                                 btrfs_info(fs_info,
2556                                         "incomplete page read in btrfs with offset %u and length %u",
2557                                         bvec->bv_offset, bvec->bv_len);
2558                 }
2559
2560                 start = page_offset(page);
2561                 end = start + bvec->bv_offset + bvec->bv_len - 1;
2562                 len = bvec->bv_len;
2563
2564                 mirror = io_bio->mirror_num;
2565                 if (likely(uptodate && tree->ops)) {
2566                         ret = tree->ops->readpage_end_io_hook(io_bio, offset,
2567                                                               page, start, end,
2568                                                               mirror);
2569                         if (ret)
2570                                 uptodate = 0;
2571                         else
2572                                 clean_io_failure(BTRFS_I(inode)->root->fs_info,
2573                                                  failure_tree, tree, start,
2574                                                  page,
2575                                                  btrfs_ino(BTRFS_I(inode)), 0);
2576                 }
2577
2578                 if (likely(uptodate))
2579                         goto readpage_ok;
2580
2581                 if (tree->ops) {
2582                         ret = tree->ops->readpage_io_failed_hook(page, mirror);
2583                         if (ret == -EAGAIN) {
2584                                 /*
2585                                  * Data inode's readpage_io_failed_hook() always
2586                                  * returns -EAGAIN.
2587                                  *
2588                                  * The generic bio_readpage_error handles errors
2589                                  * the following way: If possible, new read
2590                                  * requests are created and submitted and will
2591                                  * end up in end_bio_extent_readpage as well (if
2592                                  * we're lucky, not in the !uptodate case). In
2593                                  * that case it returns 0 and we just go on with
2594                                  * the next page in our bio. If it can't handle
2595                                  * the error it will return -EIO and we remain
2596                                  * responsible for that page.
2597                                  */
2598                                 ret = bio_readpage_error(bio, offset, page,
2599                                                          start, end, mirror);
2600                                 if (ret == 0) {
2601                                         uptodate = !bio->bi_status;
2602                                         offset += len;
2603                                         continue;
2604                                 }
2605                         }
2606
2607                         /*
2608                          * metadata's readpage_io_failed_hook() always returns
2609                          * -EIO and fixes nothing.  -EIO is also returned if
2610                          * data inode error could not be fixed.
2611                          */
2612                         ASSERT(ret == -EIO);
2613                 }
2614 readpage_ok:
2615                 if (likely(uptodate)) {
2616                         loff_t i_size = i_size_read(inode);
2617                         pgoff_t end_index = i_size >> PAGE_SHIFT;
2618                         unsigned off;
2619
2620                         /* Zero out the end if this page straddles i_size */
2621                         off = i_size & (PAGE_SIZE-1);
2622                         if (page->index == end_index && off)
2623                                 zero_user_segment(page, off, PAGE_SIZE);
2624                         SetPageUptodate(page);
2625                 } else {
2626                         ClearPageUptodate(page);
2627                         SetPageError(page);
2628                 }
2629                 unlock_page(page);
2630                 offset += len;
2631
2632                 if (unlikely(!uptodate)) {
2633                         if (extent_len) {
2634                                 endio_readpage_release_extent(tree,
2635                                                               extent_start,
2636                                                               extent_len, 1);
2637                                 extent_start = 0;
2638                                 extent_len = 0;
2639                         }
2640                         endio_readpage_release_extent(tree, start,
2641                                                       end - start + 1, 0);
2642                 } else if (!extent_len) {
2643                         extent_start = start;
2644                         extent_len = end + 1 - start;
2645                 } else if (extent_start + extent_len == start) {
2646                         extent_len += end + 1 - start;
2647                 } else {
2648                         endio_readpage_release_extent(tree, extent_start,
2649                                                       extent_len, uptodate);
2650                         extent_start = start;
2651                         extent_len = end + 1 - start;
2652                 }
2653         }
2654
2655         if (extent_len)
2656                 endio_readpage_release_extent(tree, extent_start, extent_len,
2657                                               uptodate);
2658         if (io_bio->end_io)
2659                 io_bio->end_io(io_bio, blk_status_to_errno(bio->bi_status));
2660         bio_put(bio);
2661 }
2662
2663 /*
2664  * Initialize the members up to but not including 'bio'. Use after allocating a
2665  * new bio by bio_alloc_bioset as it does not initialize the bytes outside of
2666  * 'bio' because use of __GFP_ZERO is not supported.
2667  */
2668 static inline void btrfs_io_bio_init(struct btrfs_io_bio *btrfs_bio)
2669 {
2670         memset(btrfs_bio, 0, offsetof(struct btrfs_io_bio, bio));
2671 }
2672
2673 /*
2674  * The following helpers allocate a bio. As it's backed by a bioset, it'll
2675  * never fail.  We're returning a bio right now but you can call btrfs_io_bio
2676  * for the appropriate container_of magic
2677  */
2678 struct bio *btrfs_bio_alloc(struct block_device *bdev, u64 first_byte)
2679 {
2680         struct bio *bio;
2681
2682         bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, btrfs_bioset);
2683         bio_set_dev(bio, bdev);
2684         bio->bi_iter.bi_sector = first_byte >> 9;
2685         btrfs_io_bio_init(btrfs_io_bio(bio));
2686         return bio;
2687 }
2688
2689 struct bio *btrfs_bio_clone(struct bio *bio)
2690 {
2691         struct btrfs_io_bio *btrfs_bio;
2692         struct bio *new;
2693
2694         /* Bio allocation backed by a bioset does not fail */
2695         new = bio_clone_fast(bio, GFP_NOFS, btrfs_bioset);
2696         btrfs_bio = btrfs_io_bio(new);
2697         btrfs_io_bio_init(btrfs_bio);
2698         btrfs_bio->iter = bio->bi_iter;
2699         return new;
2700 }
2701
2702 struct bio *btrfs_io_bio_alloc(unsigned int nr_iovecs)
2703 {
2704         struct bio *bio;
2705
2706         /* Bio allocation backed by a bioset does not fail */
2707         bio = bio_alloc_bioset(GFP_NOFS, nr_iovecs, btrfs_bioset);
2708         btrfs_io_bio_init(btrfs_io_bio(bio));
2709         return bio;
2710 }
2711
2712 struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size)
2713 {
2714         struct bio *bio;
2715         struct btrfs_io_bio *btrfs_bio;
2716
2717         /* this will never fail when it's backed by a bioset */
2718         bio = bio_clone_fast(orig, GFP_NOFS, btrfs_bioset);
2719         ASSERT(bio);
2720
2721         btrfs_bio = btrfs_io_bio(bio);
2722         btrfs_io_bio_init(btrfs_bio);
2723
2724         bio_trim(bio, offset >> 9, size >> 9);
2725         btrfs_bio->iter = bio->bi_iter;
2726         return bio;
2727 }
2728
2729 static int __must_check submit_one_bio(struct bio *bio, int mirror_num,
2730                                        unsigned long bio_flags)
2731 {
2732         blk_status_t ret = 0;
2733         struct bio_vec *bvec = bio_last_bvec_all(bio);
2734         struct page *page = bvec->bv_page;
2735         struct extent_io_tree *tree = bio->bi_private;
2736         u64 start;
2737
2738         start = page_offset(page) + bvec->bv_offset;
2739
2740         bio->bi_private = NULL;
2741
2742         if (tree->ops)
2743                 ret = tree->ops->submit_bio_hook(tree->private_data, bio,
2744                                            mirror_num, bio_flags, start);
2745         else
2746                 btrfsic_submit_bio(bio);
2747
2748         return blk_status_to_errno(ret);
2749 }
2750
2751 /*
2752  * @opf:        bio REQ_OP_* and REQ_* flags as one value
2753  * @tree:       tree so we can call our merge_bio hook
2754  * @wbc:        optional writeback control for io accounting
2755  * @page:       page to add to the bio
2756  * @pg_offset:  offset of the new bio or to check whether we are adding
2757  *              a contiguous page to the previous one
2758  * @size:       portion of page that we want to write
2759  * @offset:     starting offset in the page
2760  * @bdev:       attach newly created bios to this bdev
2761  * @bio_ret:    must be valid pointer, newly allocated bio will be stored there
2762  * @end_io_func:     end_io callback for new bio
2763  * @mirror_num:      desired mirror to read/write
2764  * @prev_bio_flags:  flags of previous bio to see if we can merge the current one
2765  * @bio_flags:  flags of the current bio to see if we can merge them
2766  */
2767 static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree,
2768                               struct writeback_control *wbc,
2769                               struct page *page, u64 offset,
2770                               size_t size, unsigned long pg_offset,
2771                               struct block_device *bdev,
2772                               struct bio **bio_ret,
2773                               bio_end_io_t end_io_func,
2774                               int mirror_num,
2775                               unsigned long prev_bio_flags,
2776                               unsigned long bio_flags,
2777                               bool force_bio_submit)
2778 {
2779         int ret = 0;
2780         struct bio *bio;
2781         size_t page_size = min_t(size_t, size, PAGE_SIZE);
2782         sector_t sector = offset >> 9;
2783
2784         ASSERT(bio_ret);
2785
2786         if (*bio_ret) {
2787                 bool contig;
2788                 bool can_merge = true;
2789
2790                 bio = *bio_ret;
2791                 if (prev_bio_flags & EXTENT_BIO_COMPRESSED)
2792                         contig = bio->bi_iter.bi_sector == sector;
2793                 else
2794                         contig = bio_end_sector(bio) == sector;
2795
2796                 if (tree->ops && tree->ops->merge_bio_hook(page, offset,
2797                                         page_size, bio, bio_flags))
2798                         can_merge = false;
2799
2800                 if (prev_bio_flags != bio_flags || !contig || !can_merge ||
2801                     force_bio_submit ||
2802                     bio_add_page(bio, page, page_size, pg_offset) < page_size) {
2803                         ret = submit_one_bio(bio, mirror_num, prev_bio_flags);
2804                         if (ret < 0) {
2805                                 *bio_ret = NULL;
2806                                 return ret;
2807                         }
2808                         bio = NULL;
2809                 } else {
2810                         if (wbc)
2811                                 wbc_account_io(wbc, page, page_size);
2812                         return 0;
2813                 }
2814         }
2815
2816         bio = btrfs_bio_alloc(bdev, offset);
2817         bio_add_page(bio, page, page_size, pg_offset);
2818         bio->bi_end_io = end_io_func;
2819         bio->bi_private = tree;
2820         bio->bi_write_hint = page->mapping->host->i_write_hint;
2821         bio->bi_opf = opf;
2822         if (wbc) {
2823                 wbc_init_bio(wbc, bio);
2824                 wbc_account_io(wbc, page, page_size);
2825         }
2826
2827         *bio_ret = bio;
2828
2829         return ret;
2830 }
2831
2832 static void attach_extent_buffer_page(struct extent_buffer *eb,
2833                                       struct page *page)
2834 {
2835         if (!PagePrivate(page)) {
2836                 SetPagePrivate(page);
2837                 get_page(page);
2838                 set_page_private(page, (unsigned long)eb);
2839         } else {
2840                 WARN_ON(page->private != (unsigned long)eb);
2841         }
2842 }
2843
2844 void set_page_extent_mapped(struct page *page)
2845 {
2846         if (!PagePrivate(page)) {
2847                 SetPagePrivate(page);
2848                 get_page(page);
2849                 set_page_private(page, EXTENT_PAGE_PRIVATE);
2850         }
2851 }
2852
2853 static struct extent_map *
2854 __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
2855                  u64 start, u64 len, get_extent_t *get_extent,
2856                  struct extent_map **em_cached)
2857 {
2858         struct extent_map *em;
2859
2860         if (em_cached && *em_cached) {
2861                 em = *em_cached;
2862                 if (extent_map_in_tree(em) && start >= em->start &&
2863                     start < extent_map_end(em)) {
2864                         refcount_inc(&em->refs);
2865                         return em;
2866                 }
2867
2868                 free_extent_map(em);
2869                 *em_cached = NULL;
2870         }
2871
2872         em = get_extent(BTRFS_I(inode), page, pg_offset, start, len, 0);
2873         if (em_cached && !IS_ERR_OR_NULL(em)) {
2874                 BUG_ON(*em_cached);
2875                 refcount_inc(&em->refs);
2876                 *em_cached = em;
2877         }
2878         return em;
2879 }
2880 /*
2881  * basic readpage implementation.  Locked extent state structs are inserted
2882  * into the tree that are removed when the IO is done (by the end_io
2883  * handlers)
2884  * XXX JDM: This needs looking at to ensure proper page locking
2885  * return 0 on success, otherwise return error
2886  */
2887 static int __do_readpage(struct extent_io_tree *tree,
2888                          struct page *page,
2889                          get_extent_t *get_extent,
2890                          struct extent_map **em_cached,
2891                          struct bio **bio, int mirror_num,
2892                          unsigned long *bio_flags, unsigned int read_flags,
2893                          u64 *prev_em_start)
2894 {
2895         struct inode *inode = page->mapping->host;
2896         u64 start = page_offset(page);
2897         const u64 end = start + PAGE_SIZE - 1;
2898         u64 cur = start;
2899         u64 extent_offset;
2900         u64 last_byte = i_size_read(inode);
2901         u64 block_start;
2902         u64 cur_end;
2903         struct extent_map *em;
2904         struct block_device *bdev;
2905         int ret = 0;
2906         int nr = 0;
2907         size_t pg_offset = 0;
2908         size_t iosize;
2909         size_t disk_io_size;
2910         size_t blocksize = inode->i_sb->s_blocksize;
2911         unsigned long this_bio_flag = 0;
2912
2913         set_page_extent_mapped(page);
2914
2915         if (!PageUptodate(page)) {
2916                 if (cleancache_get_page(page) == 0) {
2917                         BUG_ON(blocksize != PAGE_SIZE);
2918                         unlock_extent(tree, start, end);
2919                         goto out;
2920                 }
2921         }
2922
2923         if (page->index == last_byte >> PAGE_SHIFT) {
2924                 char *userpage;
2925                 size_t zero_offset = last_byte & (PAGE_SIZE - 1);
2926
2927                 if (zero_offset) {
2928                         iosize = PAGE_SIZE - zero_offset;
2929                         userpage = kmap_atomic(page);
2930                         memset(userpage + zero_offset, 0, iosize);
2931                         flush_dcache_page(page);
2932                         kunmap_atomic(userpage);
2933                 }
2934         }
2935         while (cur <= end) {
2936                 bool force_bio_submit = false;
2937                 u64 offset;
2938
2939                 if (cur >= last_byte) {
2940                         char *userpage;
2941                         struct extent_state *cached = NULL;
2942
2943                         iosize = PAGE_SIZE - pg_offset;
2944                         userpage = kmap_atomic(page);
2945                         memset(userpage + pg_offset, 0, iosize);
2946                         flush_dcache_page(page);
2947                         kunmap_atomic(userpage);
2948                         set_extent_uptodate(tree, cur, cur + iosize - 1,
2949                                             &cached, GFP_NOFS);
2950                         unlock_extent_cached(tree, cur,
2951                                              cur + iosize - 1, &cached);
2952                         break;
2953                 }
2954                 em = __get_extent_map(inode, page, pg_offset, cur,
2955                                       end - cur + 1, get_extent, em_cached);
2956                 if (IS_ERR_OR_NULL(em)) {
2957                         SetPageError(page);
2958                         unlock_extent(tree, cur, end);
2959                         break;
2960                 }
2961                 extent_offset = cur - em->start;
2962                 BUG_ON(extent_map_end(em) <= cur);
2963                 BUG_ON(end < cur);
2964
2965                 if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
2966                         this_bio_flag |= EXTENT_BIO_COMPRESSED;
2967                         extent_set_compress_type(&this_bio_flag,
2968                                                  em->compress_type);
2969                 }
2970
2971                 iosize = min(extent_map_end(em) - cur, end - cur + 1);
2972                 cur_end = min(extent_map_end(em) - 1, end);
2973                 iosize = ALIGN(iosize, blocksize);
2974                 if (this_bio_flag & EXTENT_BIO_COMPRESSED) {
2975                         disk_io_size = em->block_len;
2976                         offset = em->block_start;
2977                 } else {
2978                         offset = em->block_start + extent_offset;
2979                         disk_io_size = iosize;
2980                 }
2981                 bdev = em->bdev;
2982                 block_start = em->block_start;
2983                 if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
2984                         block_start = EXTENT_MAP_HOLE;
2985
2986                 /*
2987                  * If we have a file range that points to a compressed extent
2988                  * and it's followed by a consecutive file range that points to
2989                  * to the same compressed extent (possibly with a different
2990                  * offset and/or length, so it either points to the whole extent
2991                  * or only part of it), we must make sure we do not submit a
2992                  * single bio to populate the pages for the 2 ranges because
2993                  * this makes the compressed extent read zero out the pages
2994                  * belonging to the 2nd range. Imagine the following scenario:
2995                  *
2996                  *  File layout
2997                  *  [0 - 8K]                     [8K - 24K]
2998                  *    |                               |
2999                  *    |                               |
3000                  * points to extent X,         points to extent X,
3001                  * offset 4K, length of 8K     offset 0, length 16K
3002                  *
3003                  * [extent X, compressed length = 4K uncompressed length = 16K]
3004                  *
3005                  * If the bio to read the compressed extent covers both ranges,
3006                  * it will decompress extent X into the pages belonging to the
3007                  * first range and then it will stop, zeroing out the remaining
3008                  * pages that belong to the other range that points to extent X.
3009                  * So here we make sure we submit 2 bios, one for the first
3010                  * range and another one for the third range. Both will target
3011                  * the same physical extent from disk, but we can't currently
3012                  * make the compressed bio endio callback populate the pages
3013                  * for both ranges because each compressed bio is tightly
3014                  * coupled with a single extent map, and each range can have
3015                  * an extent map with a different offset value relative to the
3016                  * uncompressed data of our extent and different lengths. This
3017                  * is a corner case so we prioritize correctness over
3018                  * non-optimal behavior (submitting 2 bios for the same extent).
3019                  */
3020                 if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags) &&
3021                     prev_em_start && *prev_em_start != (u64)-1 &&
3022                     *prev_em_start != em->orig_start)
3023                         force_bio_submit = true;
3024
3025                 if (prev_em_start)
3026                         *prev_em_start = em->orig_start;
3027
3028                 free_extent_map(em);
3029                 em = NULL;
3030
3031                 /* we've found a hole, just zero and go on */
3032                 if (block_start == EXTENT_MAP_HOLE) {
3033                         char *userpage;
3034                         struct extent_state *cached = NULL;
3035
3036                         userpage = kmap_atomic(page);
3037                         memset(userpage + pg_offset, 0, iosize);
3038                         flush_dcache_page(page);
3039                         kunmap_atomic(userpage);
3040
3041                         set_extent_uptodate(tree, cur, cur + iosize - 1,
3042                                             &cached, GFP_NOFS);
3043                         unlock_extent_cached(tree, cur,
3044                                              cur + iosize - 1, &cached);
3045                         cur = cur + iosize;
3046                         pg_offset += iosize;
3047                         continue;
3048                 }
3049                 /* the get_extent function already copied into the page */
3050                 if (test_range_bit(tree, cur, cur_end,
3051                                    EXTENT_UPTODATE, 1, NULL)) {
3052                         check_page_uptodate(tree, page);
3053                         unlock_extent(tree, cur, cur + iosize - 1);
3054                         cur = cur + iosize;
3055                         pg_offset += iosize;
3056                         continue;
3057                 }
3058                 /* we have an inline extent but it didn't get marked up
3059                  * to date.  Error out
3060                  */
3061                 if (block_start == EXTENT_MAP_INLINE) {
3062                         SetPageError(page);
3063                         unlock_extent(tree, cur, cur + iosize - 1);
3064                         cur = cur + iosize;
3065                         pg_offset += iosize;
3066                         continue;
3067                 }
3068
3069                 ret = submit_extent_page(REQ_OP_READ | read_flags, tree, NULL,
3070                                          page, offset, disk_io_size,
3071                                          pg_offset, bdev, bio,
3072                                          end_bio_extent_readpage, mirror_num,
3073                                          *bio_flags,
3074                                          this_bio_flag,
3075                                          force_bio_submit);
3076                 if (!ret) {
3077                         nr++;
3078                         *bio_flags = this_bio_flag;
3079                 } else {
3080                         SetPageError(page);
3081                         unlock_extent(tree, cur, cur + iosize - 1);
3082                         goto out;
3083                 }
3084                 cur = cur + iosize;
3085                 pg_offset += iosize;
3086         }
3087 out:
3088         if (!nr) {
3089                 if (!PageError(page))
3090                         SetPageUptodate(page);
3091                 unlock_page(page);
3092         }
3093         return ret;
3094 }
3095
3096 static inline void __do_contiguous_readpages(struct extent_io_tree *tree,
3097                                              struct page *pages[], int nr_pages,
3098                                              u64 start, u64 end,
3099                                              struct extent_map **em_cached,
3100                                              struct bio **bio,
3101                                              unsigned long *bio_flags,
3102                                              u64 *prev_em_start)
3103 {
3104         struct inode *inode;
3105         struct btrfs_ordered_extent *ordered;
3106         int index;
3107
3108         inode = pages[0]->mapping->host;
3109         while (1) {
3110                 lock_extent(tree, start, end);
3111                 ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
3112                                                      end - start + 1);
3113                 if (!ordered)
3114                         break;
3115                 unlock_extent(tree, start, end);
3116                 btrfs_start_ordered_extent(inode, ordered, 1);
3117                 btrfs_put_ordered_extent(ordered);
3118         }
3119
3120         for (index = 0; index < nr_pages; index++) {
3121                 __do_readpage(tree, pages[index], btrfs_get_extent, em_cached,
3122                                 bio, 0, bio_flags, 0, prev_em_start);
3123                 put_page(pages[index]);
3124         }
3125 }
3126
3127 static void __extent_readpages(struct extent_io_tree *tree,
3128                                struct page *pages[],
3129                                int nr_pages,
3130                                struct extent_map **em_cached,
3131                                struct bio **bio, unsigned long *bio_flags,
3132                                u64 *prev_em_start)
3133 {
3134         u64 start = 0;
3135         u64 end = 0;
3136         u64 page_start;
3137         int index;
3138         int first_index = 0;
3139
3140         for (index = 0; index < nr_pages; index++) {
3141                 page_start = page_offset(pages[index]);
3142                 if (!end) {
3143                         start = page_start;
3144                         end = start + PAGE_SIZE - 1;
3145                         first_index = index;
3146                 } else if (end + 1 == page_start) {
3147                         end += PAGE_SIZE;
3148                 } else {
3149                         __do_contiguous_readpages(tree, &pages[first_index],
3150                                                   index - first_index, start,
3151                                                   end, em_cached,
3152                                                   bio, bio_flags,
3153                                                   prev_em_start);
3154                         start = page_start;
3155                         end = start + PAGE_SIZE - 1;
3156                         first_index = index;
3157                 }
3158         }
3159
3160         if (end)
3161                 __do_contiguous_readpages(tree, &pages[first_index],
3162                                           index - first_index, start,
3163                                           end, em_cached, bio,
3164                                           bio_flags, prev_em_start);
3165 }
3166
3167 static int __extent_read_full_page(struct extent_io_tree *tree,
3168                                    struct page *page,
3169                                    get_extent_t *get_extent,
3170                                    struct bio **bio, int mirror_num,
3171                                    unsigned long *bio_flags,
3172                                    unsigned int read_flags)
3173 {
3174         struct inode *inode = page->mapping->host;
3175         struct btrfs_ordered_extent *ordered;
3176         u64 start = page_offset(page);
3177         u64 end = start + PAGE_SIZE - 1;
3178         int ret;
3179
3180         while (1) {
3181                 lock_extent(tree, start, end);
3182                 ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
3183                                                 PAGE_SIZE);
3184                 if (!ordered)
3185                         break;
3186                 unlock_extent(tree, start, end);
3187                 btrfs_start_ordered_extent(inode, ordered, 1);
3188                 btrfs_put_ordered_extent(ordered);
3189         }
3190
3191         ret = __do_readpage(tree, page, get_extent, NULL, bio, mirror_num,
3192                             bio_flags, read_flags, NULL);
3193         return ret;
3194 }
3195
3196 int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
3197                             get_extent_t *get_extent, int mirror_num)
3198 {
3199         struct bio *bio = NULL;
3200         unsigned long bio_flags = 0;
3201         int ret;
3202
3203         ret = __extent_read_full_page(tree, page, get_extent, &bio, mirror_num,
3204                                       &bio_flags, 0);
3205         if (bio)
3206                 ret = submit_one_bio(bio, mirror_num, bio_flags);
3207         return ret;
3208 }
3209
3210 static void update_nr_written(struct writeback_control *wbc,
3211                               unsigned long nr_written)
3212 {
3213         wbc->nr_to_write -= nr_written;
3214 }
3215
3216 /*
3217  * helper for __extent_writepage, doing all of the delayed allocation setup.
3218  *
3219  * This returns 1 if our fill_delalloc function did all the work required
3220  * to write the page (copy into inline extent).  In this case the IO has
3221  * been started and the page is already unlocked.
3222  *
3223  * This returns 0 if all went well (page still locked)
3224  * This returns < 0 if there were errors (page still locked)
3225  */
3226 static noinline_for_stack int writepage_delalloc(struct inode *inode,
3227                               struct page *page, struct writeback_control *wbc,
3228                               struct extent_page_data *epd,
3229                               u64 delalloc_start,
3230                               unsigned long *nr_written)
3231 {
3232         struct extent_io_tree *tree = epd->tree;
3233         u64 page_end = delalloc_start + PAGE_SIZE - 1;
3234         u64 nr_delalloc;
3235         u64 delalloc_to_write = 0;
3236         u64 delalloc_end = 0;
3237         int ret;
3238         int page_started = 0;
3239
3240         if (epd->extent_locked || !tree->ops || !tree->ops->fill_delalloc)
3241                 return 0;
3242
3243         while (delalloc_end < page_end) {
3244                 nr_delalloc = find_lock_delalloc_range(inode, tree,
3245                                                page,
3246                                                &delalloc_start,
3247                                                &delalloc_end,
3248                                                BTRFS_MAX_EXTENT_SIZE);
3249                 if (nr_delalloc == 0) {
3250                         delalloc_start = delalloc_end + 1;
3251                         continue;
3252                 }
3253                 ret = tree->ops->fill_delalloc(inode, page,
3254                                                delalloc_start,
3255                                                delalloc_end,
3256                                                &page_started,
3257                                                nr_written, wbc);
3258                 /* File system has been set read-only */
3259                 if (ret) {
3260                         SetPageError(page);
3261                         /* fill_delalloc should be return < 0 for error
3262                          * but just in case, we use > 0 here meaning the
3263                          * IO is started, so we don't want to return > 0
3264                          * unless things are going well.
3265                          */
3266                         ret = ret < 0 ? ret : -EIO;
3267                         goto done;
3268                 }