fs/btrfs/extent-tree.c

   1 /*
   2  * Copyright (C) 2007 Oracle.  All rights reserved.
   3  *
   4  * This program is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU General Public
   6  * License v2 as published by the Free Software Foundation.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  11  * General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public
  14  * License along with this program; if not, write to the
  15  * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  16  * Boston, MA 021110-1307, USA.
  17  */
  18 #include <linux/sched.h>
  19 #include <linux/pagemap.h>
  20 #include <linux/writeback.h>
  21 #include <linux/blkdev.h>
  22 #include <linux/sort.h>
  23 #include <linux/rcupdate.h>
  24 #include <linux/kthread.h>
  25 #include "compat.h"
  26 #include "hash.h"
  27 #include "ctree.h"
  28 #include "disk-io.h"
  29 #include "print-tree.h"
  30 #include "transaction.h"
  31 #include "volumes.h"
  32 #include "locking.h"
  33 #include "free-space-cache.h"
  34
  35 static int update_block_group(struct btrfs_trans_handle *trans,
  36                               struct btrfs_root *root,
  37                               u64 bytenr, u64 num_bytes, int alloc,
  38                               int mark_free);
  39 static int update_reserved_extents(struct btrfs_block_group_cache *cache,
  40                                    u64 num_bytes, int reserve);
  41 static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
  42                                 struct btrfs_root *root,
  43                                 u64 bytenr, u64 num_bytes, u64 parent,
  44                                 u64 root_objectid, u64 owner_objectid,
  45                                 u64 owner_offset, int refs_to_drop,
  46                                 struct btrfs_delayed_extent_op *extra_op);
  47 static void __run_delayed_extent_op(struct btrfs_delayed_extent_op *extent_op,
  48                                     struct extent_buffer *leaf,
  49                                     struct btrfs_extent_item *ei);
  50 static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
  51                                       struct btrfs_root *root,
  52                                       u64 parent, u64 root_objectid,
  53                                       u64 flags, u64 owner, u64 offset,
  54                                       struct btrfs_key *ins, int ref_mod);
  55 static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
  56                                      struct btrfs_root *root,
  57                                      u64 parent, u64 root_objectid,
  58                                      u64 flags, struct btrfs_disk_key *key,
  59                                      int level, struct btrfs_key *ins);
  60 static int do_chunk_alloc(struct btrfs_trans_handle *trans,
  61                           struct btrfs_root *extent_root, u64 alloc_bytes,
  62                           u64 flags, int force);
  63 static int pin_down_bytes(struct btrfs_trans_handle *trans,
  64                           struct btrfs_root *root,
  65                           struct btrfs_path *path,
  66                           u64 bytenr, u64 num_bytes,
  67                           int is_data, int reserved,
  68                           struct extent_buffer **must_clean);
  69 static int find_next_key(struct btrfs_path *path, int level,
  70                          struct btrfs_key *key);
  71 static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
  72                             int dump_block_groups);
  73
  74 static noinline int
  75 block_group_cache_done(struct btrfs_block_group_cache *cache)
  76 {
  77         smp_mb();
  78         return cache->cached == BTRFS_CACHE_FINISHED;
  79 }
  80
  81 static int block_group_bits(struct btrfs_block_group_cache *cache, u64 bits)
  82 {
  83         return (cache->flags & bits) == bits;
  84 }
  85
  86 /*
  87  * this adds the block group to the fs_info rb tree for the block group
  88  * cache
  89  */
  90 static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
  91                                 struct btrfs_block_group_cache *block_group)
  92 {
  93         struct rb_node **p;
  94         struct rb_node *parent = NULL;
  95         struct btrfs_block_group_cache *cache;
  96
  97         spin_lock(&info->block_group_cache_lock);
  98         p = &info->block_group_cache_tree.rb_node;
  99
 100         while (*p) {
 101                 parent = *p;
 102                 cache = rb_entry(parent, struct btrfs_block_group_cache,
 103                                  cache_node);
 104                 if (block_group->key.objectid < cache->key.objectid) {
 105                         p = &(*p)->rb_left;
 106                 } else if (block_group->key.objectid > cache->key.objectid) {
 107                         p = &(*p)->rb_right;
 108                 } else {
 109                         spin_unlock(&info->block_group_cache_lock);
 110                         return -EEXIST;
 111                 }
 112         }
 113
 114         rb_link_node(&block_group->cache_node, parent, p);
 115         rb_insert_color(&block_group->cache_node,
 116                         &info->block_group_cache_tree);
 117         spin_unlock(&info->block_group_cache_lock);
 118
 119         return 0;
 120 }
 121
 122 /*
 123  * This will return the block group at or after bytenr if contains is 0, else
 124  * it will return the block group that contains the bytenr
 125  */
 126 static struct btrfs_block_group_cache *
 127 block_group_cache_tree_search(struct btrfs_fs_info *info, u64 bytenr,
 128                               int contains)
 129 {
 130         struct btrfs_block_group_cache *cache, *ret = NULL;
 131         struct rb_node *n;
 132         u64 end, start;
 133
 134         spin_lock(&info->block_group_cache_lock);
 135         n = info->block_group_cache_tree.rb_node;
 136
 137         while (n) {
 138                 cache = rb_entry(n, struct btrfs_block_group_cache,
 139                                  cache_node);
 140                 end = cache->key.objectid + cache->key.offset - 1;
 141                 start = cache->key.objectid;
 142
 143                 if (bytenr < start) {
 144                         if (!contains && (!ret || start < ret->key.objectid))
 145                                 ret = cache;
 146                         n = n->rb_left;
 147                 } else if (bytenr > start) {
 148                         if (contains && bytenr <= end) {
 149                                 ret = cache;
 150                                 break;
 151                         }
 152                         n = n->rb_right;
 153                 } else {
 154                         ret = cache;
 155                         break;
 156                 }
 157         }
 158         if (ret)
 159                 atomic_inc(&ret->count);
 160         spin_unlock(&info->block_group_cache_lock);
 161
 162         return ret;
 163 }
 164
 165 static int add_excluded_extent(struct btrfs_root *root,
 166                                u64 start, u64 num_bytes)
 167 {
 168         u64 end = start + num_bytes - 1;
 169         set_extent_bits(&root->fs_info->freed_extents[0],
 170                         start, end, EXTENT_UPTODATE, GFP_NOFS);
 171         set_extent_bits(&root->fs_info->freed_extents[1],
 172                         start, end, EXTENT_UPTODATE, GFP_NOFS);
 173         return 0;
 174 }
 175
 176 static void free_excluded_extents(struct btrfs_root *root,
 177                                   struct btrfs_block_group_cache *cache)
 178 {
 179         u64 start, end;
 180
 181         start = cache->key.objectid;
 182         end = start + cache->key.offset - 1;
 183
 184         clear_extent_bits(&root->fs_info->freed_extents[0],
 185                           start, end, EXTENT_UPTODATE, GFP_NOFS);
 186         clear_extent_bits(&root->fs_info->freed_extents[1],
 187                           start, end, EXTENT_UPTODATE, GFP_NOFS);
 188 }
 189
 190 static int exclude_super_stripes(struct btrfs_root *root,
 191                                  struct btrfs_block_group_cache *cache)
 192 {
 193         u64 bytenr;
 194         u64 *logical;
 195         int stripe_len;
 196         int i, nr, ret;
 197
 198         for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
 199                 bytenr = btrfs_sb_offset(i);
 200                 ret = btrfs_rmap_block(&root->fs_info->mapping_tree,
 201                                        cache->key.objectid, bytenr,
 202                                        0, &logical, &nr, &stripe_len);
 203                 BUG_ON(ret);
 204
 205                 while (nr--) {
 206                         cache->bytes_super += stripe_len;
 207                         ret = add_excluded_extent(root, logical[nr],
 208                                                   stripe_len);
 209                         BUG_ON(ret);
 210                 }
 211
 212                 kfree(logical);
 213         }
 214         return 0;
 215 }
 216
 217 static struct btrfs_caching_control *
 218 get_caching_control(struct btrfs_block_group_cache *cache)
 219 {
 220         struct btrfs_caching_control *ctl;
 221
 222         spin_lock(&cache->lock);
 223         if (cache->cached != BTRFS_CACHE_STARTED) {
 224                 spin_unlock(&cache->lock);
 225                 return NULL;
 226         }
 227
 228         ctl = cache->caching_ctl;
 229         atomic_inc(&ctl->count);
 230         spin_unlock(&cache->lock);
 231         return ctl;
 232 }
 233
 234 static void put_caching_control(struct btrfs_caching_control *ctl)
 235 {
 236         if (atomic_dec_and_test(&ctl->count))
 237                 kfree(ctl);
 238 }
 239
 240 /*
 241  * this is only called by cache_block_group, since we could have freed extents
 242  * we need to check the pinned_extents for any extents that can't be used yet
 243  * since their free space will be released as soon as the transaction commits.
 244  */
 245 static u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
 246                               struct btrfs_fs_info *info, u64 start, u64 end)
 247 {
 248         u64 extent_start, extent_end, size, total_added = 0;
 249         int ret;
 250
 251         while (start < end) {
 252                 ret = find_first_extent_bit(info->pinned_extents, start,
 253                                             &extent_start, &extent_end,
 254                                             EXTENT_DIRTY | EXTENT_UPTODATE);
 255                 if (ret)
 256                         break;
 257
 258                 if (extent_start == start) {
 259                         start = extent_end + 1;
 260                 } else if (extent_start > start && extent_start < end) {
 261                         size = extent_start - start;
 262                         total_added += size;
 263                         ret = btrfs_add_free_space(block_group, start,
 264                                                    size);
 265                         BUG_ON(ret);
 266                         start = extent_end + 1;
 267                 } else {
 268                         break;
 269                 }
 270         }
 271
 272         if (start < end) {
 273                 size = end - start;
 274                 total_added += size;
 275                 ret = btrfs_add_free_space(block_group, start, size);
 276                 BUG_ON(ret);
 277         }
 278
 279         return total_added;
 280 }
 281
 282 static int caching_kthread(void *data)
 283 {
 284         struct btrfs_block_group_cache *block_group = data;
 285         struct btrfs_fs_info *fs_info = block_group->fs_info;
 286         struct btrfs_caching_control *caching_ctl = block_group->caching_ctl;
 287         struct btrfs_root *extent_root = fs_info->extent_root;
 288         struct btrfs_path *path;
 289         struct extent_buffer *leaf;
 290         struct btrfs_key key;
 291         u64 total_found = 0;
 292         u64 last = 0;
 293         u32 nritems;
 294         int ret = 0;
 295
 296         path = btrfs_alloc_path();
 297         if (!path)
 298                 return -ENOMEM;
 299
 300         exclude_super_stripes(extent_root, block_group);
 301         spin_lock(&block_group->space_info->lock);
 302         block_group->space_info->bytes_super += block_group->bytes_super;
 303         spin_unlock(&block_group->space_info->lock);
 304
 305         last = max_t(u64, block_group->key.objectid, BTRFS_SUPER_INFO_OFFSET);
 306
 307         /*
 308          * We don't want to deadlock with somebody trying to allocate a new
 309          * extent for the extent root while also trying to search the extent
 310          * root to add free space.  So we skip locking and search the commit
 311          * root, since its read-only
 312          */
 313         path->skip_locking = 1;
 314         path->search_commit_root = 1;
 315         path->reada = 2;
 316
 317         key.objectid = last;
 318         key.offset = 0;
 319         key.type = BTRFS_EXTENT_ITEM_KEY;
 320 again:
 321         mutex_lock(&caching_ctl->mutex);
 322         /* need to make sure the commit_root doesn't disappear */
 323         down_read(&fs_info->extent_commit_sem);
 324
 325         ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0);
 326         if (ret < 0)
 327                 goto err;
 328
 329         leaf = path->nodes[0];
 330         nritems = btrfs_header_nritems(leaf);
 331
 332         while (1) {
 333                 smp_mb();
 334                 if (fs_info->closing > 1) {
 335                         last = (u64)-1;
 336                         break;
 337                 }
 338
 339                 if (path->slots[0] < nritems) {
 340                         btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 341                 } else {
 342                         ret = find_next_key(path, 0, &key);
 343                         if (ret)
 344                                 break;
 345
 346                         caching_ctl->progress = last;
 347                         btrfs_release_path(extent_root, path);
 348                         up_read(&fs_info->extent_commit_sem);
 349                         mutex_unlock(&caching_ctl->mutex);
 350                         if (btrfs_transaction_in_commit(fs_info))
 351                                 schedule_timeout(1);
 352                         else
 353                                 cond_resched();
 354                         goto again;
 355                 }
 356
 357                 if (key.objectid < block_group->key.objectid) {
 358                         path->slots[0]++;
 359                         continue;
 360                 }
 361
 362                 if (key.objectid >= block_group->key.objectid +
 363                     block_group->key.offset)
 364                         break;
 365
 366                 if (key.type == BTRFS_EXTENT_ITEM_KEY) {
 367                         total_found += add_new_free_space(block_group,
 368                                                           fs_info, last,
 369                                                           key.objectid);
 370                         last = key.objectid + key.offset;
 371
 372                         if (total_found > (1024 * 1024 * 2)) {
 373                                 total_found = 0;
 374                                 wake_up(&caching_ctl->wait);
 375                         }
 376                 }
 377                 path->slots[0]++;
 378         }
 379         ret = 0;
 380
 381         total_found += add_new_free_space(block_group, fs_info, last,
 382                                           block_group->key.objectid +
 383                                           block_group->key.offset);
 384         caching_ctl->progress = (u64)-1;
 385
 386         spin_lock(&block_group->lock);
 387         block_group->caching_ctl = NULL;
 388         block_group->cached = BTRFS_CACHE_FINISHED;
 389         spin_unlock(&block_group->lock);
 390
 391 err:
 392         btrfs_free_path(path);
 393         up_read(&fs_info->extent_commit_sem);
 394
 395         free_excluded_extents(extent_root, block_group);
 396
 397         mutex_unlock(&caching_ctl->mutex);
 398         wake_up(&caching_ctl->wait);
 399
 400         put_caching_control(caching_ctl);
 401         atomic_dec(&block_group->space_info->caching_threads);
 402         return 0;
 403 }
 404
 405 static int cache_block_group(struct btrfs_block_group_cache *cache)
 406 {
 407         struct btrfs_fs_info *fs_info = cache->fs_info;
 408         struct btrfs_caching_control *caching_ctl;
 409         struct task_struct *tsk;
 410         int ret = 0;
 411
 412         smp_mb();
 413         if (cache->cached != BTRFS_CACHE_NO)
 414                 return 0;
 415
 416         caching_ctl = kzalloc(sizeof(*caching_ctl), GFP_KERNEL);
 417         BUG_ON(!caching_ctl);
 418
 419         INIT_LIST_HEAD(&caching_ctl->list);
 420         mutex_init(&caching_ctl->mutex);
 421         init_waitqueue_head(&caching_ctl->wait);
 422         caching_ctl->block_group = cache;
 423         caching_ctl->progress = cache->key.objectid;
 424         /* one for caching kthread, one for caching block group list */
 425         atomic_set(&caching_ctl->count, 2);
 426
 427         spin_lock(&cache->lock);
 428         if (cache->cached != BTRFS_CACHE_NO) {
 429                 spin_unlock(&cache->lock);
 430                 kfree(caching_ctl);
 431                 return 0;
 432         }
 433         cache->caching_ctl = caching_ctl;
 434         cache->cached = BTRFS_CACHE_STARTED;
 435         spin_unlock(&cache->lock);
 436
 437         down_write(&fs_info->extent_commit_sem);
 438         list_add_tail(&caching_ctl->list, &fs_info->caching_block_groups);
 439         up_write(&fs_info->extent_commit_sem);
 440
 441         atomic_inc(&cache->space_info->caching_threads);
 442
 443         tsk = kthread_run(caching_kthread, cache, "btrfs-cache-%llu\n",
 444                           cache->key.objectid);
 445         if (IS_ERR(tsk)) {
 446                 ret = PTR_ERR(tsk);
 447                 printk(KERN_ERR "error running thread %d\n", ret);
 448                 BUG();
 449         }
 450
 451         return ret;
 452 }
 453
 454 /*
 455  * return the block group that starts at or after bytenr
 456  */
 457 static struct btrfs_block_group_cache *
 458 btrfs_lookup_first_block_group(struct btrfs_fs_info *info, u64 bytenr)
 459 {
 460         struct btrfs_block_group_cache *cache;
 461
 462         cache = block_group_cache_tree_search(info, bytenr, 0);
 463
 464         return cache;
 465 }
 466
 467 /*
 468  * return the block group that contains the given bytenr
 469  */
 470 struct btrfs_block_group_cache *btrfs_lookup_block_group(
 471                                                  struct btrfs_fs_info *info,
 472                                                  u64 bytenr)
 473 {
 474         struct btrfs_block_group_cache *cache;
 475
 476         cache = block_group_cache_tree_search(info, bytenr, 1);
 477
 478         return cache;
 479 }
 480
 481 void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
 482 {
 483         if (atomic_dec_and_test(&cache->count))
 484                 kfree(cache);
 485 }
 486
 487 static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 488                                                   u64 flags)
 489 {
 490         struct list_head *head = &info->space_info;
 491         struct btrfs_space_info *found;
 492
 493         rcu_read_lock();
 494         list_for_each_entry_rcu(found, head, list) {
 495                 if (found->flags == flags) {
 496                         rcu_read_unlock();
 497                         return found;
 498                 }
 499         }
 500         rcu_read_unlock();
 501         return NULL;
 502 }
 503
 504 /*
 505  * after adding space to the filesystem, we need to clear the full flags
 506  * on all the space infos.
 507  */
 508 void btrfs_clear_space_info_full(struct btrfs_fs_info *info)
 509 {
 510         struct list_head *head = &info->space_info;
 511         struct btrfs_space_info *found;
 512
 513         rcu_read_lock();
 514         list_for_each_entry_rcu(found, head, list)
 515                 found->full = 0;
 516         rcu_read_unlock();
 517 }
 518
 519 static u64 div_factor(u64 num, int factor)
 520 {
 521         if (factor == 10)
 522                 return num;
 523         num *= factor;
 524         do_div(num, 10);
 525         return num;
 526 }
 527
 528 u64 btrfs_find_block_group(struct btrfs_root *root,
 529                            u64 search_start, u64 search_hint, int owner)
 530 {
 531         struct btrfs_block_group_cache *cache;
 532         u64 used;
 533         u64 last = max(search_hint, search_start);
 534         u64 group_start = 0;
 535         int full_search = 0;
 536         int factor = 9;
 537         int wrapped = 0;
 538 again:
 539         while (1) {
 540                 cache = btrfs_lookup_first_block_group(root->fs_info, last);
 541                 if (!cache)
 542                         break;
 543
 544                 spin_lock(&cache->lock);
 545                 last = cache->key.objectid + cache->key.offset;
 546                 used = btrfs_block_group_used(&cache->item);
 547
 548                 if ((full_search || !cache->ro) &&
 549                     block_group_bits(cache, BTRFS_BLOCK_GROUP_METADATA)) {
 550                         if (used + cache->pinned + cache->reserved <
 551                             div_factor(cache->key.offset, factor)) {
 552                                 group_start = cache->key.objectid;
 553                                 spin_unlock(&cache->lock);
 554                                 btrfs_put_block_group(cache);
 555                                 goto found;
 556                         }
 557                 }
 558                 spin_unlock(&cache->lock);
 559                 btrfs_put_block_group(cache);
 560                 cond_resched();
 561         }
 562         if (!wrapped) {
 563                 last = search_start;
 564                 wrapped = 1;
 565                 goto again;
 566         }
 567         if (!full_search && factor < 10) {
 568                 last = search_start;
 569                 full_search = 1;
 570                 factor = 10;
 571                 goto again;
 572         }
 573 found:
 574         return group_start;
 575 }
 576
 577 /* simple helper to search for an existing extent at a given offset */
 578 int btrfs_lookup_extent(struct btrfs_root *root, u64 start, u64 len)
 579 {
 580         int ret;
 581         struct btrfs_key key;
 582         struct btrfs_path *path;
 583
 584         path = btrfs_alloc_path();
 585         BUG_ON(!path);
 586         key.objectid = start;
 587         key.offset = len;
 588         btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
 589         ret = btrfs_search_slot(NULL, root->fs_info->extent_root, &key, path,
 590                                 0, 0);
 591         btrfs_free_path(path);
 592         return ret;
 593 }
 594
 595 /*
 596  * Back reference rules.  Back refs have three main goals:
 597  *
 598  * 1) differentiate between all holders of references to an extent so that
 599  *    when a reference is dropped we can make sure it was a valid reference
 600  *    before freeing the extent.
 601  *
 602  * 2) Provide enough information to quickly find the holders of an extent
 603  *    if we notice a given block is corrupted or bad.
 604  *
 605  * 3) Make it easy to migrate blocks for FS shrinking or storage pool
 606  *    maintenance.  This is actually the same as #2, but with a slightly
 607  *    different use case.
 608  *
 609  * There are two kinds of back refs. The implicit back refs is optimized
 610  * for pointers in non-shared tree blocks. For a given pointer in a block,
 611  * back refs of this kind provide information about the block's owner tree
 612  * and the pointer's key. These information allow us to find the block by
 613  * b-tree searching. The full back refs is for pointers in tree blocks not
 614  * referenced by their owner trees. The location of tree block is recorded
 615  * in the back refs. Actually the full back refs is generic, and can be
 616  * used in all cases the implicit back refs is used. The major shortcoming
 617  * of the full back refs is its overhead. Every time a tree block gets
 618  * COWed, we have to update back refs entry for all pointers in it.
 619  *
 620  * For a newly allocated tree block, we use implicit back refs for
 621  * pointers in it. This means most tree related operations only involve
 622  * implicit back refs. For a tree block created in old transaction, the
 623  * only way to drop a reference to it is COW it. So we can detect the
 624  * event that tree block loses its owner tree's reference and do the
 625  * back refs conversion.
 626  *
 627  * When a tree block is COW'd through a tree, there are four cases:
 628  *
 629  * The reference count of the block is one and the tree is the block's
 630  * owner tree. Nothing to do in this case.
 631  *
 632  * The reference count of the block is one and the tree is not the
 633  * block's owner tree. In this case, full back refs is used for pointers
 634  * in the block. Remove these full back refs, add implicit back refs for
 635  * every pointers in the new block.
 636  *
 637  * The reference count of the block is greater than one and the tree is
 638  * the block's owner tree. In this case, implicit back refs is used for
 639  * pointers in the block. Add full back refs for every pointers in the
 640  * block, increase lower level extents' reference counts. The original
 641  * implicit back refs are entailed to the new block.
 642  *
 643  * The reference count of the block is greater than one and the tree is
 644  * not the block's owner tree. Add implicit back refs for every pointer in
 645  * the new block, increase lower level extents' reference count.
 646  *
 647  * Back Reference Key composing:
 648  *
 649  * The key objectid corresponds to the first byte in the extent,
 650  * The key type is used to differentiate between types of back refs.
 651  * There are different meanings of the key offset for different types
 652  * of back refs.
 653  *
 654  * File extents can be referenced by:
 655  *
 656  * - multiple snapshots, subvolumes, or different generations in one subvol
 657  * - different files inside a single subvolume
 658  * - different offsets inside a file (bookend extents in file.c)
 659  *
 660  * The extent ref structure for the implicit back refs has fields for:
 661  *
 662  * - Objectid of the subvolume root
 663  * - objectid of the file holding the reference
 664  * - original offset in the file
 665  * - how many bookend extents
 666  *
 667  * The key offset for the implicit back refs is hash of the first
 668  * three fields.
 669  *
 670  * The extent ref structure for the full back refs has field for:
 671  *
 672  * - number of pointers in the tree leaf
 673  *
 674  * The key offset for the implicit back refs is the first byte of
 675  * the tree leaf
 676  *
 677  * When a file extent is allocated, The implicit back refs is used.
 678  * the fields are filled in:
 679  *
 680  *     (root_key.objectid, inode objectid, offset in file, 1)
 681  *
 682  * When a file extent is removed file truncation, we find the
 683  * corresponding implicit back refs and check the following fields:
 684  *
 685  *     (btrfs_header_owner(leaf), inode objectid, offset in file)
 686  *
 687  * Btree extents can be referenced by:
 688  *
 689  * - Different subvolumes
 690  *
 691  * Both the implicit back refs and the full back refs for tree blocks
 692  * only consist of key. The key offset for the implicit back refs is
 693  * objectid of block's owner tree. The key offset for the full back refs
 694  * is the first byte of parent block.
 695  *
 696  * When implicit back refs is used, information about the lowest key and
 697  * level of the tree block are required. These information are stored in
 698  * tree block info structure.
 699  */
 700
 701 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
 702 static int convert_extent_item_v0(struct btrfs_trans_handle *trans,
 703                                   struct btrfs_root *root,
 704                                   struct btrfs_path *path,
 705                                   u64 owner, u32 extra_size)
 706 {
 707         struct btrfs_extent_item *item;
 708         struct btrfs_extent_item_v0 *ei0;
 709         struct btrfs_extent_ref_v0 *ref0;
 710         struct btrfs_tree_block_info *bi;
 711         struct extent_buffer *leaf;
 712         struct btrfs_key key;
 713         struct btrfs_key found_key;
 714         u32 new_size = sizeof(*item);
 715         u64 refs;
 716         int ret;
 717
 718         leaf = path->nodes[0];
 719         BUG_ON(btrfs_item_size_nr(leaf, path->slots[0]) != sizeof(*ei0));
 720
 721         btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 722         ei0 = btrfs_item_ptr(leaf, path->slots[0],
 723                              struct btrfs_extent_item_v0);
 724         refs = btrfs_extent_refs_v0(leaf, ei0);
 725
 726         if (owner == (u64)-1) {
 727                 while (1) {
 728                         if (path->slots[0] >= btrfs_header_nritems(leaf)) {
 729                                 ret = btrfs_next_leaf(root, path);
 730                                 if (ret < 0)
 731                                         return ret;
 732                                 BUG_ON(ret > 0);
 733                                 leaf = path->nodes[0];
 734                         }
 735                         btrfs_item_key_to_cpu(leaf, &found_key,
 736                                               path->slots[0]);
 737                         BUG_ON(key.objectid != found_key.objectid);
 738                         if (found_key.type != BTRFS_EXTENT_REF_V0_KEY) {
 739                                 path->slots[0]++;
 740                                 continue;
 741                         }
 742                         ref0 = btrfs_item_ptr(leaf, path->slots[0],
 743                                               struct btrfs_extent_ref_v0);
 744                         owner = btrfs_ref_objectid_v0(leaf, ref0);
 745                         break;
 746                 }
 747         }
 748         btrfs_release_path(root, path);
 749
 750         if (owner < BTRFS_FIRST_FREE_OBJECTID)
 751                 new_size += sizeof(*bi);
 752
 753         new_size -= sizeof(*ei0);
 754         ret = btrfs_search_slot(trans, root, &key, path,
 755                                 new_size + extra_size, 1);
 756         if (ret < 0)
 757                 return ret;
 758         BUG_ON(ret);
 759
 760         ret = btrfs_extend_item(trans, root, path, new_size);
 761         BUG_ON(ret);
 762
 763         leaf = path->nodes[0];
 764         item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
 765         btrfs_set_extent_refs(leaf, item, refs);
 766         /* FIXME: get real generation */
 767         btrfs_set_extent_generation(leaf, item, 0);
 768         if (owner < BTRFS_FIRST_FREE_OBJECTID) {
 769                 btrfs_set_extent_flags(leaf, item,
 770                                        BTRFS_EXTENT_FLAG_TREE_BLOCK |
 771                                        BTRFS_BLOCK_FLAG_FULL_BACKREF);
 772                 bi = (struct btrfs_tree_block_info *)(item + 1);
 773                 /* FIXME: get first key of the block */
 774                 memset_extent_buffer(leaf, 0, (unsigned long)bi, sizeof(*bi));
 775                 btrfs_set_tree_block_level(leaf, bi, (int)owner);
 776         } else {
 777                 btrfs_set_extent_flags(leaf, item, BTRFS_EXTENT_FLAG_DATA);
 778         }
 779         btrfs_mark_buffer_dirty(leaf);
 780         return 0;
 781 }
 782 #endif
 783
 784 static u64 hash_extent_data_ref(u64 root_objectid, u64 owner, u64 offset)
 785 {
 786         u32 high_crc = ~(u32)0;
 787         u32 low_crc = ~(u32)0;
 788         __le64 lenum;
 789
 790         lenum = cpu_to_le64(root_objectid);
 791         high_crc = crc32c(high_crc, &lenum, sizeof(lenum));
 792         lenum = cpu_to_le64(owner);
 793         low_crc = crc32c(low_crc, &lenum, sizeof(lenum));
 794         lenum = cpu_to_le64(offset);
 795         low_crc = crc32c(low_crc, &lenum, sizeof(lenum));
 796
 797         return ((u64)high_crc << 31) ^ (u64)low_crc;
 798 }
 799
 800 static u64 hash_extent_data_ref_item(struct extent_buffer *leaf,
 801                                      struct btrfs_extent_data_ref *ref)
 802 {
 803         return hash_extent_data_ref(btrfs_extent_data_ref_root(leaf, ref),
 804                                     btrfs_extent_data_ref_objectid(leaf, ref),
 805                                     btrfs_extent_data_ref_offset(leaf, ref));
 806 }
 807
 808 static int match_extent_data_ref(struct extent_buffer *leaf,
 809                                  struct btrfs_extent_data_ref *ref,
 810                                  u64 root_objectid, u64 owner, u64 offset)
 811 {
 812         if (btrfs_extent_data_ref_root(leaf, ref) != root_objectid ||
 813             btrfs_extent_data_ref_objectid(leaf, ref) != owner ||
 814             btrfs_extent_data_ref_offset(leaf, ref) != offset)
 815                 return 0;
 816         return 1;
 817 }
 818
 819 static noinline int lookup_extent_data_ref(struct btrfs_trans_handle *trans,
 820                                            struct btrfs_root *root,
 821                                            struct btrfs_path *path,
 822                                            u64 bytenr, u64 parent,
 823                                            u64 root_objectid,
 824                                            u64 owner, u64 offset)
 825 {
 826         struct btrfs_key key;
 827         struct btrfs_extent_data_ref *ref;
 828         struct extent_buffer *leaf;
 829         u32 nritems;
 830         int ret;
 831         int recow;
 832         int err = -ENOENT;
 833
 834         key.objectid = bytenr;
 835         if (parent) {
 836                 key.type = BTRFS_SHARED_DATA_REF_KEY;
 837                 key.offset = parent;
 838         } else {
 839                 key.type = BTRFS_EXTENT_DATA_REF_KEY;
 840                 key.offset = hash_extent_data_ref(root_objectid,
 841                                                   owner, offset);
 842         }
 843 again:
 844         recow = 0;
 845         ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 846         if (ret < 0) {
 847                 err = ret;
 848                 goto fail;
 849         }
 850
 851         if (parent) {
 852                 if (!ret)
 853                         return 0;
 854 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
 855                 key.type = BTRFS_EXTENT_REF_V0_KEY;
 856                 btrfs_release_path(root, path);
 857                 ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 858                 if (ret < 0) {
 859                         err = ret;
 860                         goto fail;
 861                 }
 862                 if (!ret)
 863                         return 0;
 864 #endif
 865                 goto fail;
 866         }
 867
 868         leaf = path->nodes[0];
 869         nritems = btrfs_header_nritems(leaf);
 870         while (1) {
 871                 if (path->slots[0] >= nritems) {
 872                         ret = btrfs_next_leaf(root, path);
 873                         if (ret < 0)
 874                                 err = ret;
 875                         if (ret)
 876                                 goto fail;
 877
 878                         leaf = path->nodes[0];
 879                         nritems = btrfs_header_nritems(leaf);
 880                         recow = 1;
 881                 }
 882
 883                 btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 884                 if (key.objectid != bytenr ||
 885                     key.type != BTRFS_EXTENT_DATA_REF_KEY)
 886                         goto fail;
 887
 888                 ref = btrfs_item_ptr(leaf, path->slots[0],
 889                                      struct btrfs_extent_data_ref);
 890
 891                 if (match_extent_data_ref(leaf, ref, root_objectid,
 892                                           owner, offset)) {
 893                         if (recow) {
 894                                 btrfs_release_path(root, path);
 895                                 goto again;
 896                         }
 897                         err = 0;
 898                         break;
 899                 }
 900                 path->slots[0]++;
 901         }
 902 fail:
 903         return err;
 904 }
 905
 906 static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans,
 907                                            struct btrfs_root *root,
 908                                            struct btrfs_path *path,
 909                                            u64 bytenr, u64 parent,
 910                                            u64 root_objectid, u64 owner,
 911                                            u64 offset, int refs_to_add)
 912 {
 913         struct btrfs_key key;
 914         struct extent_buffer *leaf;
 915         u32 size;
 916         u32 num_refs;
 917         int ret;
 918
 919         key.objectid = bytenr;
 920         if (parent) {
 921                 key.type = BTRFS_SHARED_DATA_REF_KEY;
 922                 key.offset = parent;
 923                 size = sizeof(struct btrfs_shared_data_ref);
 924         } else {
 925                 key.type = BTRFS_EXTENT_DATA_REF_KEY;
 926                 key.offset = hash_extent_data_ref(root_objectid,
 927                                                   owner, offset);
 928                 size = sizeof(struct btrfs_extent_data_ref);
 929         }
 930
 931         ret = btrfs_insert_empty_item(trans, root, path, &key, size);
 932         if (ret && ret != -EEXIST)
 933                 goto fail;
 934
 935         leaf = path->nodes[0];
 936         if (parent) {
 937                 struct btrfs_shared_data_ref *ref;
 938                 ref = btrfs_item_ptr(leaf, path->slots[0],
 939                                      struct btrfs_shared_data_ref);
 940                 if (ret == 0) {
 941                         btrfs_set_shared_data_ref_count(leaf, ref, refs_to_add);
 942                 } else {
 943                         num_refs = btrfs_shared_data_ref_count(leaf, ref);
 944                         num_refs += refs_to_add;
 945                         btrfs_set_shared_data_ref_count(leaf, ref, num_refs);
 946                 }
 947         } else {
 948                 struct btrfs_extent_data_ref *ref;
 949                 while (ret == -EEXIST) {
 950                         ref = btrfs_item_ptr(leaf, path->slots[0],
 951                                              struct btrfs_extent_data_ref);
 952                         if (match_extent_data_ref(leaf, ref, root_objectid,
 953                                                   owner, offset))
 954                                 break;
 955                         btrfs_release_path(root, path);
 956                         key.offset++;
 957                         ret = btrfs_insert_empty_item(trans, root, path, &key,
 958                                                       size);
 959                         if (ret && ret != -EEXIST)
 960                                 goto fail;
 961
 962                         leaf = path->nodes[0];
 963                 }
 964                 ref = btrfs_item_ptr(leaf, path->slots[0],
 965                                      struct btrfs_extent_data_ref);
 966                 if (ret == 0) {
 967                         btrfs_set_extent_data_ref_root(leaf, ref,
 968                                                        root_objectid);
 969                         btrfs_set_extent_data_ref_objectid(leaf, ref, owner);
 970                         btrfs_set_extent_data_ref_offset(leaf, ref, offset);
 971                         btrfs_set_extent_data_ref_count(leaf, ref, refs_to_add);
 972                 } else {
 973                         num_refs = btrfs_extent_data_ref_count(leaf, ref);
 974                         num_refs += refs_to_add;
 975                         btrfs_set_extent_data_ref_count(leaf, ref, num_refs);
 976                 }
 977         }
 978         btrfs_mark_buffer_dirty(leaf);
 979         ret = 0;
 980 fail:
 981         btrfs_release_path(root, path);
 982         return ret;
 983 }
 984
 985 static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans,
 986                                            struct btrfs_root *root,
 987                                            struct btrfs_path *path,
 988                                            int refs_to_drop)
 989 {
 990         struct btrfs_key key;
 991         struct btrfs_extent_data_ref *ref1 = NULL;
 992         struct btrfs_shared_data_ref *ref2 = NULL;
 993         struct extent_buffer *leaf;
 994         u32 num_refs = 0;
 995         int ret = 0;
 996
 997         leaf = path->nodes[0];
 998         btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 999
1000         if (key.type == BTRFS_EXTENT_DATA_REF_KEY) {
1001                 ref1 = btrfs_item_ptr(leaf, path->slots[0],
1002                                       struct btrfs_extent_data_ref);
1003                 num_refs = btrfs_extent_data_ref_count(leaf, ref1);
1004         } else if (key.type == BTRFS_SHARED_DATA_REF_KEY) {
1005                 ref2 = btrfs_item_ptr(leaf, path->slots[0],
1006                                       struct btrfs_shared_data_ref);
1007                 num_refs = btrfs_shared_data_ref_count(leaf, ref2);
1008 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
1009         } else if (key.type == BTRFS_EXTENT_REF_V0_KEY) {
1010                 struct btrfs_extent_ref_v0 *ref0;
1011                 ref0 = btrfs_item_ptr(leaf, path->slots[0],
1012                                       struct btrfs_extent_ref_v0);
1013                 num_refs = btrfs_ref_count_v0(leaf, ref0);
1014 #endif
1015         } else {
1016                 BUG();
1017         }
1018
1019         BUG_ON(num_refs < refs_to_drop);
1020         num_refs -= refs_to_drop;
1021
1022         if (num_refs == 0) {
1023                 ret = btrfs_del_item(trans, root, path);
1024         } else {
1025                 if (key.type == BTRFS_EXTENT_DATA_REF_KEY)
1026                         btrfs_set_extent_data_ref_count(leaf, ref1, num_refs);
1027                 else if (key.type == BTRFS_SHARED_DATA_REF_KEY)
1028                         btrfs_set_shared_data_ref_count(leaf, ref2, num_refs);
1029 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
1030                 else {
1031                         struct btrfs_extent_ref_v0 *ref0;
1032                         ref0 = btrfs_item_ptr(leaf, path->slots[0],
1033                                         struct btrfs_extent_ref_v0);
1034                         btrfs_set_ref_count_v0(leaf, ref0, num_refs);
1035                 }
1036 #endif
1037                 btrfs_mark_buffer_dirty(leaf);
1038         }
1039         return ret;
1040 }
1041
1042 static noinline u32 extent_data_ref_count(struct btrfs_root *root,
1043                                           struct btrfs_path *path,
1044                                           struct btrfs_extent_inline_ref *iref)
1045 {
1046         struct btrfs_key key;
1047         struct extent_buffer *leaf;
1048         struct btrfs_extent_data_ref *ref1;
1049         struct btrfs_shared_data_ref *ref2;
1050         u32 num_refs = 0;
1051
1052         leaf = path->nodes[0];
1053         btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
1054         if (iref) {
1055                 if (btrfs_extent_inline_ref_type(leaf, iref) ==
1056                     BTRFS_EXTENT_DATA_REF_KEY) {
1057                         ref1 = (struct btrfs_extent_data_ref *)(&iref->offset);
1058                         num_refs = btrfs_extent_data_ref_count(leaf, ref1);
1059                 } else {
1060                         ref2 = (struct btrfs_shared_data_ref *)(iref + 1);
1061                         num_refs = btrfs_shared_data_ref_count(leaf, ref2);
1062                 }
1063         } else if (key.type == BTRFS_EXTENT_DATA_REF_KEY) {
1064                 ref1 = btrfs_item_ptr(leaf, path->slots[0],
1065                                       struct btrfs_extent_data_ref);
1066                 num_refs = btrfs_extent_data_ref_count(leaf, ref1);
1067         } else if (key.type == BTRFS_SHARED_DATA_REF_KEY) {
1068                 ref2 = btrfs_item_ptr(leaf, path->slots[0],
1069                                       struct btrfs_shared_data_ref);
1070                 num_refs = btrfs_shared_data_ref_count(leaf, ref2);
1071 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
1072         } else if (key.type == BTRFS_EXTENT_REF_V0_KEY) {
1073                 struct btrfs_extent_ref_v0 *ref0;
1074                 ref0 = btrfs_item_ptr(leaf, path->slots[0],
1075                                       struct btrfs_extent_ref_v0);
1076                 num_refs = btrfs_ref_count_v0(leaf, ref0);
1077 #endif
1078         } else {
1079                 WARN_ON(1);
1080         }
1081         return num_refs;
1082 }
1083
1084 static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans,
1085                                           struct btrfs_root *root,
1086                                           struct btrfs_path *path,
1087                                           u64 bytenr, u64 parent,
1088                                           u64 root_objectid)
1089 {
1090         struct btrfs_key key;
1091         int ret;
1092
1093         key.objectid = bytenr;
1094         if (parent) {
1095                 key.type = BTRFS_SHARED_BLOCK_REF_KEY;
1096                 key.offset = parent;
1097         } else {
1098                 key.type = BTRFS_TREE_BLOCK_REF_KEY;
1099                 key.offset = root_objectid;
1100         }
1101
1102         ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
1103         if (ret > 0)
1104                 ret = -ENOENT;
1105 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
1106         if (ret == -ENOENT && parent) {
1107                 btrfs_release_path(root, path);
1108                 key.type = BTRFS_EXTENT_REF_V0_KEY;
1109                 ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
1110                 if (ret > 0)
1111                         ret = -ENOENT;
1112         }
1113 #endif
1114         return ret;
1115 }
1116
1117 static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans,
1118                                           struct btrfs_root *root,
1119                                           struct btrfs_path *path,
1120                                           u64 bytenr, u64 parent,
1121                                           u64 root_objectid)
1122 {
1123         struct btrfs_key key;
1124         int ret;
1125
1126         key.objectid = bytenr;
1127         if (parent) {
1128                 key.type = BTRFS_SHARED_BLOCK_REF_KEY;
1129                 key.offset = parent;
1130         } else {
1131                 key.type = BTRFS_TREE_BLOCK_REF_KEY;
1132                 key.offset = root_objectid;
1133         }
1134
1135         ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
1136         btrfs_release_path(root, path);
1137         return ret;
1138 }
1139
1140 static inline int extent_ref_type(u64 parent, u64 owner)
1141 {
1142         int type;
1143         if (owner < BTRFS_FIRST_FREE_OBJECTID) {
1144                 if (parent > 0)
1145                         type = BTRFS_SHARED_BLOCK_REF_KEY;
1146                 else
1147                         type = BTRFS_TREE_BLOCK_REF_KEY;
1148         } else {
1149                 if (parent > 0)
1150                         type = BTRFS_SHARED_DATA_REF_KEY;
1151                 else
1152                         type = BTRFS_EXTENT_DATA_REF_KEY;
1153         }
1154         return type;
1155 }
1156
1157 static int find_next_key(struct btrfs_path *path, int level,
1158                          struct btrfs_key *key)
1159
1160 {
1161         for (; level < BTRFS_MAX_LEVEL; level++) {
1162                 if (!path->nodes[level])
1163                         break;
1164                 if (path->slots[level] + 1 >=
1165                     btrfs_header_nritems(path->nodes[level]))
1166                         continue;
1167                 if (level == 0)
1168                         btrfs_item_key_to_cpu(path->nodes[level], key,
1169                                               path->slots[level] + 1);
1170                 else
1171                         btrfs_node_key_to_cpu(path->nodes[level], key,
1172                                               path->slots[level] + 1);
1173                 return 0;
1174         }
1175         return 1;
1176 }
1177
1178 /*
1179  * look for inline back ref. if back ref is found, *ref_ret is set
1180  * to the address of inline back ref, and 0 is returned.
1181  *
1182  * if back ref isn't found, *ref_ret is set to the address where it
1183  * should be inserted, and -ENOENT is returned.
1184  *
1185  * if insert is true and there are too many inline back refs, the path
1186  * points to the extent item, and -EAGAIN is returned.
1187  *
1188  * NOTE: inline back refs are ordered in the same way that back ref
1189  *       items in the tree are ordered.
1190  */
1191 static noinline_for_stack
1192 int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
1193                                  struct btrfs_root *root,
1194                                  struct btrfs_path *path,
1195                                  struct btrfs_extent_inline_ref **ref_ret,
1196                                  u64 bytenr, u64 num_bytes,
1197                                  u64 parent, u64 root_objectid,
1198                                  u64 owner, u64 offset, int insert)
1199 {
1200         struct btrfs_key key;
1201         struct extent_buffer *leaf;
1202         struct btrfs_extent_item *ei;
1203         struct btrfs_extent_inline_ref *iref;
1204         u64 flags;
1205         u64 item_size;
1206         unsigned long ptr;
1207         unsigned long end;
1208         int extra_size;
1209         int type;
1210         int want;
1211         int ret;
1212         int err = 0;
1213
1214         key.objectid = bytenr;
1215         key.type = BTRFS_EXTENT_ITEM_KEY;
1216         key.offset = num_bytes;
1217
1218         want = extent_ref_type(parent, owner);
1219         if (insert) {
1220                 extra_size = btrfs_extent_inline_ref_size(want);
1221                 path->keep_locks = 1;
1222         } else
1223                 extra_size = -1;
1224         ret = btrfs_search_slot(trans, root, &key, path, extra_size, 1);
1225         if (ret < 0) {
1226                 err = ret;
1227                 goto out;
1228         }
1229         BUG_ON(ret);
1230
1231         leaf = path->nodes[0];
1232         item_size = btrfs_item_size_nr(leaf, path->slots[0]);
1233 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
1234         if (item_size < sizeof(*ei)) {
1235                 if (!insert) {
1236                         err = -ENOENT;
1237                         goto out;
1238                 }
1239                 ret = convert_extent_item_v0(trans, root, path, owner,
1240                                              extra_size);
1241                 if (ret < 0) {
1242                         err = ret;
1243                         goto out;
1244                 }
1245                 leaf = path->nodes[0];
1246                 item_size = btrfs_item_size_nr(leaf, path->slots[0]);
1247         }
1248 #endif
1249         BUG_ON(item_size < sizeof(*ei));
1250
1251         ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
1252         flags = btrfs_extent_flags(leaf, ei);
1253
1254         ptr = (unsigned long)(ei + 1);
1255         end = (unsigned long)ei + item_size;
1256
1257         if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
1258                 ptr += sizeof(struct btrfs_tree_block_info);
1259                 BUG_ON(ptr > end);
1260         } else {
1261                 BUG_ON(!(flags & BTRFS_EXTENT_FLAG_DATA));
1262         }
1263
1264         err = -ENOENT;
1265         while (1) {
1266                 if (ptr >= end) {
1267                         WARN_ON(ptr > end);
1268                         break;
1269                 }
1270                 iref = (struct btrfs_extent_inline_ref *)ptr;
1271                 type = btrfs_extent_inline_ref_type(leaf, iref);
1272                 if (want < type)
1273                         break;
1274                 if (want > type) {
1275                         ptr += btrfs_extent_inline_ref_size(type);
1276                         continue;
1277                 }
1278
1279                 if (type == BTRFS_EXTENT_DATA_REF_KEY) {
1280                         struct btrfs_extent_data_ref *dref;
1281                         dref = (struct btrfs_extent_data_ref *)(&iref->offset);
1282                         if (match_extent_data_ref(leaf, dref, root_objectid,
1283                                                   owner, offset)) {
1284                                 err = 0;
1285                                 break;
1286                         }
1287                         if (hash_extent_data_ref_item(leaf, dref) <
1288                             hash_extent_data_ref(root_objectid, owner, offset))
1289                                 break;
1290                 } else {
1291                         u64 ref_offset;
1292                         ref_offset = btrfs_extent_inline_ref_offset(leaf, iref);
1293                         if (parent > 0) {
1294                                 if (parent == ref_offset) {
1295                                         err = 0;
1296                                         break;
1297                                 }
1298                                 if (ref_offset < parent)
1299                                         break;
1300                         } else {
1301                                 if (root_objectid == ref_offset) {
1302                                         err = 0;
1303                                         break;
1304                                 }
1305                                 if (ref_offset < root_objectid)
1306                                         break;
1307                         }
1308                 }
1309                 ptr += btrfs_extent_inline_ref_size(type);
1310         }
1311         if (err == -ENOENT && insert) {
1312                 if (item_size + extra_size >=
1313                     BTRFS_MAX_EXTENT_ITEM_SIZE(root)) {
1314                         err = -EAGAIN;
1315                         goto out;
1316                 }
1317                 /*
1318                  * To add new inline back ref, we have to make sure
1319                  * there is no corresponding back ref item.
1320                  * For simplicity, we just do not add new inline back
1321                  * ref if there is any kind of item for this block
1322                  */
1323                 if (find_next_key(path, 0, &key) == 0 &&
1324                     key.objectid == bytenr &&
1325                     key.type < BTRFS_BLOCK_GROUP_ITEM_KEY) {
1326                         err = -EAGAIN;
1327                         goto out;
1328                 }
1329         }
1330         *ref_ret = (struct btrfs_extent_inline_ref *)ptr;
1331 out:
1332         if (insert) {
1333                 path->keep_locks = 0;
1334                 btrfs_unlock_up_safe(path, 1);
1335         }
1336         return err;
1337 }
1338
1339 /*
1340  * helper to add new inline back ref
1341  */
1342 static noinline_for_stack
1343 int setup_inline_extent_backref(struct btrfs_trans_handle *trans,
1344                                 struct btrfs_root *root,
1345                                 struct btrfs_path *path,
1346                                 struct btrfs_extent_inline_ref *iref,
1347                                 u64 parent, u64 root_objectid,
1348                                 u64 owner, u64 offset, int refs_to_add,
1349                                 struct btrfs_delayed_extent_op *extent_op)
1350 {
1351         struct extent_buffer *leaf;
1352         struct btrfs_extent_item *ei;
1353         unsigned long ptr;
1354         unsigned long end;
1355         unsigned long item_offset;
1356         u64 refs;
1357         int size;
1358         int type;
1359         int ret;
1360
1361         leaf = path->nodes[0];
1362         ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
1363         item_offset = (unsigned long)iref - (unsigned long)ei;
1364
1365         type = extent_ref_type(parent, owner);
1366         size = btrfs_extent_inline_ref_size(type);
1367
1368         ret = btrfs_extend_item(trans, root, path, size);
1369         BUG_ON(ret);
1370
1371         ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
1372         refs = btrfs_extent_refs(leaf, ei);
1373         refs += refs_to_add;
1374         btrfs_set_extent_refs(leaf, ei, refs);
1375         if (extent_op)
1376                 __run_delayed_extent_op(extent_op, leaf, ei);
1377
1378         ptr = (unsigned long)ei + item_offset;
1379         end = (unsigned long)ei + btrfs_item_size_nr(leaf, path->slots[0]);
1380         if (ptr < end - size)
1381                 memmove_extent_buffer(leaf, ptr + size, ptr,
1382                                       end - size - ptr);
1383
1384         iref = (struct btrfs_extent_inline_ref *)ptr;
1385         btrfs_set_extent_inline_ref_type(leaf, iref, type);
1386         if (type == BTRFS_EXTENT_DATA_REF_KEY) {
1387                 struct btrfs_extent_data_ref *dref;
1388                 dref = (struct btrfs_extent_data_ref *)(&iref->offset);
1389                 btrfs_set_extent_data_ref_root(leaf, dref, root_objectid);
1390                 btrfs_set_extent_data_ref_objectid(leaf, dref, owner);
1391                 btrfs_set_extent_data_ref_offset(leaf, dref, offset);
1392                 btrfs_set_extent_data_ref_count(leaf, dref, refs_to_add);
1393         } else if (type == BTRFS_SHARED_DATA_REF_KEY) {
1394                 struct btrfs_shared_data_ref *sref;
1395                 sref = (struct btrfs_shared_data_ref *)(iref + 1);
1396                 btrfs_set_shared_data_ref_count(leaf, sref, refs_to_add);
1397                 btrfs_set_extent_inline_ref_offset(leaf, iref, parent);
1398         } else if (type == BTRFS_SHARED_BLOCK_REF_KEY) {
1399                 btrfs_set_extent_inline_ref_offset(leaf, iref, parent);
1400         } else {
1401                 btrfs_set_extent_inline_ref_offset(leaf, iref, root_objectid);
1402         }
1403         btrfs_mark_buffer_dirty(leaf);
1404         return 0;
1405 }
1406
1407 static int lookup_extent_backref(struct btrfs_trans_handle *trans,
1408                                  struct btrfs_root *root,
1409                                  struct btrfs_path *path,
1410                                  struct btrfs_extent_inline_ref **ref_ret,
1411                                  u64 bytenr, u64 num_bytes, u64 parent,
1412                                  u64 root_objectid, u64 owner, u64 offset)
1413 {
1414         int ret;
1415
1416         ret = lookup_inline_extent_backref(trans, root, path, ref_ret,
1417                                            bytenr, num_bytes, parent,
1418                                            root_objectid, owner, offset, 0);
1419         if (ret != -ENOENT)
1420                 return ret;
1421
1422         btrfs_release_path(root, path);
1423         *ref_ret = NULL;
1424
1425         if (owner < BTRFS_FIRST_FREE_OBJECTID) {
1426                 ret = lookup_tree_block_ref(trans, root, path, bytenr, parent,
1427                                             root_objectid);
1428         } else {
1429                 ret = lookup_extent_data_ref(trans, root, path, bytenr, parent,
1430                                              root_objectid, owner, offset);
1431         }
1432         return ret;
1433 }
1434
1435 /*
1436  * helper to update/remove inline back ref
1437  */
1438 static noinline_for_stack
1439 int update_inline_extent_backref(struct btrfs_trans_handle *trans,
1440                                  struct btrfs_root *root,
1441                                  struct btrfs_path *path,
1442                                  struct btrfs_extent_inline_ref *iref,
1443                                  int refs_to_mod,
1444                                  struct btrfs_delayed_extent_op *extent_op)
1445 {
1446         struct extent_buffer *leaf;
1447         struct btrfs_extent_item *ei;
1448         struct btrfs_extent_data_ref *dref = NULL;
1449         struct btrfs_shared_data_ref *sref = NULL;
1450         unsigned long ptr;
1451         unsigned long end;
1452         u32 item_size;
1453         int size;
1454         int type;
1455         int ret;
1456         u64 refs;
1457
1458         leaf = path->nodes[0];
1459         ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
1460         refs = btrfs_extent_refs(leaf, ei);
1461         WARN_ON(refs_to_mod < 0 && refs + refs_to_mod <= 0);
1462         refs += refs_to_mod;
1463         btrfs_set_extent_refs(leaf, ei, refs);
1464         if (extent_op)
1465                 __run_delayed_extent_op(extent_op, leaf, ei);
1466
1467         type = btrfs_extent_inline_ref_type(leaf, iref);
1468
1469         if (type == BTRFS_EXTENT_DATA_REF_KEY) {
1470                 dref = (struct btrfs_extent_data_ref *)(&iref->offset);
1471                 refs = btrfs_extent_data_ref_count(leaf, dref);
1472         } else if (type == BTRFS_SHARED_DATA_REF_KEY) {
1473                 sref = (struct btrfs_shared_data_ref *)(iref + 1);
1474                 refs = btrfs_shared_data_ref_count(leaf, sref);
1475         } else {
1476                 refs = 1;
1477                 BUG_ON(refs_to_mod != -1);
1478         }
1479
1480         BUG_ON(refs_to_mod < 0 && refs < -refs_to_mod);
1481         refs += refs_to_mod;
1482
1483         if (refs > 0) {
1484                 if (type == BTRFS_EXTENT_DATA_REF_KEY)
1485                         btrfs_set_extent_data_ref_count(leaf, dref, refs);
1486                 else
1487                         btrfs_set_shared_data_ref_count(leaf, sref, refs);
1488         } else {
1489                 size =  btrfs_extent_inline_ref_size(type);
1490                 item_size = btrfs_item_size_nr(leaf, path->slots[0]);
1491                 ptr = (unsigned long)iref;
1492                 end = (unsigned long)ei + item_size;
1493                 if (ptr + size < end)
1494                         memmove_extent_buffer(leaf, ptr, ptr + size,
1495                                               end - ptr - size);
1496                 item_size -= size;
1497                 ret = btrfs_truncate_item(trans, root, path, item_size, 1);
1498                 BUG_ON(ret);
1499         }
1500         btrfs_mark_buffer_dirty(leaf);
1501         return 0;
1502 }
1503
1504 static noinline_for_stack
1505 int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
1506                                  struct btrfs_root *root,
1507                                  struct btrfs_path *path,
1508                                  u64 bytenr, u64 num_bytes, u64 parent,
1509                                  u64 root_objectid, u64 owner,
1510                                  u64 offset, int refs_to_add,
1511                                  struct btrfs_delayed_extent_op *extent_op)
1512 {
1513         struct btrfs_extent_inline_ref *iref;
1514         int ret;
1515
1516         ret = lookup_inline_extent_backref(trans, root, path, &iref,
1517                                            bytenr, num_bytes, parent,
1518                                            root_objectid, owner, offset, 1);
1519         if (ret == 0) {
1520                 BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID);
1521                 ret = update_inline_extent_backref(trans, root, path, iref,
1522                                                    refs_to_add, extent_op);
1523         } else if (ret == -ENOENT) {
1524                 ret = setup_inline_extent_backref(trans, root, path, iref,
1525                                                   parent, root_objectid,
1526                                                   owner, offset, refs_to_add,
1527                                                   extent_op);
1528         }
1529         return ret;
1530 }
1531
1532 static int insert_extent_backref(struct btrfs_trans_handle *trans,
1533                                  struct btrfs_root *root,
1534                                  struct btrfs_path *path,
1535                                  u64 bytenr, u64 parent, u64 root_objectid,
1536                                  u64 owner, u64 offset, int refs_to_add)
1537 {
1538         int ret;
1539         if (owner < BTRFS_FIRST_FREE_OBJECTID) {
1540                 BUG_ON(refs_to_add != 1);
1541                 ret = insert_tree_block_ref(trans, root, path, bytenr,
1542                                             parent, root_objectid);
1543         } else {
1544                 ret = insert_extent_data_ref(trans, root, path, bytenr,
1545                                              parent, root_objectid,
1546                                              owner, offset, refs_to_add);
1547         }
1548         return ret;
1549 }
1550
1551 static int remove_extent_backref(struct btrfs_trans_handle *trans,
1552                                  struct btrfs_root *root,
1553                                  struct btrfs_path *path,
1554                                  struct btrfs_extent_inline_ref *iref,
1555                                  int refs_to_drop, int is_data)
1556 {
1557         int ret;
1558
1559         BUG_ON(!is_data && refs_to_drop != 1);
1560         if (iref) {
1561                 ret = update_inline_extent_backref(trans, root, path, iref,
1562                                                    -refs_to_drop, NULL);
1563         } else if (is_data) {
1564                 ret = remove_extent_data_ref(trans, root, path, refs_to_drop);
1565         } else {
1566                 ret = btrfs_del_item(trans, root, path);
1567         }
1568         return ret;
1569 }
1570
1571 #ifdef BIO_RW_DISCARD
1572 static void btrfs_issue_discard(struct block_device *bdev,
1573                                 u64 start, u64 len)
1574 {
1575         blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_KERNEL,
1576                              DISCARD_FL_BARRIER);
1577 }
1578 #endif
1579
1580 static int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
1581                                 u64 num_bytes)
1582 {
1583 #ifdef BIO_RW_DISCARD
1584         int ret;
1585         u64 map_length = num_bytes;
1586         struct btrfs_multi_bio *multi = NULL;
1587
1588         /* Tell the block device(s) that the sectors can be discarded */
1589         ret = btrfs_map_block(&root->fs_info->mapping_tree, READ,
1590                               bytenr, &map_length, &multi, 0);
1591         if (!ret) {
1592                 struct btrfs_bio_stripe *stripe = multi->stripes;
1593                 int i;
1594
1595                 if (map_length > num_bytes)
1596                         map_length = num_bytes;
1597
1598                 for (i = 0; i < multi->num_stripes; i++, stripe++) {
1599                         btrfs_issue_discard(stripe->dev->bdev,
1600                                             stripe->physical,
1601                                             map_length);
1602                 }
1603                 kfree(multi);
1604         }
1605
1606         return ret;
1607 #else
1608         return 0;
1609 #endif
1610 }
1611
1612 int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
1613                          struct btrfs_root *root,
1614                          u64 bytenr, u64 num_bytes, u64 parent,
1615                          u64 root_objectid, u64 owner, u64 offset)
1616 {
1617         int ret;
1618         BUG_ON(owner < BTRFS_FIRST_FREE_OBJECTID &&
1619                root_objectid == BTRFS_TREE_LOG_OBJECTID);
1620
1621         if (owner < BTRFS_FIRST_FREE_OBJECTID) {
1622                 ret = btrfs_add_delayed_tree_ref(trans, bytenr, num_bytes,
1623                                         parent, root_objectid, (int)owner,
1624                                         BTRFS_ADD_DELAYED_REF, NULL);
1625         } else {
1626                 ret = btrfs_add_delayed_data_ref(trans, bytenr, num_bytes,
1627                                         parent, root_objectid, owner, offset,
1628                                         BTRFS_ADD_DELAYED_REF, NULL);
1629         }
1630         return ret;
1631 }
1632
1633 static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
1634                                   struct btrfs_root *root,
1635                                   u64 bytenr, u64 num_bytes,
1636                                   u64 parent, u64 root_objectid,
1637                                   u64 owner, u64 offset, int refs_to_add,
1638                                   struct btrfs_delayed_extent_op *extent_op)
1639 {
1640         struct btrfs_path *path;
1641         struct extent_buffer *leaf;
1642         struct btrfs_extent_item *item;
1643         u64 refs;
1644         int ret;
1645         int err = 0;
1646
1647         path = btrfs_alloc_path();
1648         if (!path)
1649                 return -ENOMEM;
1650
1651         path->reada = 1;
1652         path->leave_spinning = 1;
1653         /* this will setup the path even if it fails to insert the back ref */
1654         ret = insert_inline_extent_backref(trans, root->fs_info->extent_root,
1655                                            path, bytenr, num_bytes, parent,
1656                                            root_objectid, owner, offset,
1657                                            refs_to_add, extent_op);
1658         if (ret == 0)
1659                 goto out;
1660
1661         if (ret != -EAGAIN) {
1662                 err = ret;
1663                 goto out;
1664         }
1665
1666         leaf = path->nodes[0];
1667         item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
1668         refs = btrfs_extent_refs(leaf, item);
1669         btrfs_set_extent_refs(leaf, item, refs + refs_to_add);
1670         if (extent_op)
1671                 __run_delayed_extent_op(extent_op, leaf, item);
1672
1673         btrfs_mark_buffer_dirty(leaf);
1674         btrfs_release_path(root->fs_info->extent_root, path);
1675
1676         path->reada = 1;
1677         path->leave_spinning = 1;
1678
1679         /* now insert the actual backref */
1680         ret = insert_extent_backref(trans, root->fs_info->extent_root,
1681                                     path, bytenr, parent, root_objectid,
1682                                     owner, offset, refs_to_add);
1683         BUG_ON(ret);
1684 out:
1685         btrfs_free_path(path);
1686         return err;
1687 }
1688
1689 static int run_delayed_data_ref(struct btrfs_trans_handle *trans,
1690                                 struct btrfs_root *root,
1691                                 struct btrfs_delayed_ref_node *node,
1692                                 struct btrfs_delayed_extent_op *extent_op,
1693                                 int insert_reserved)
1694 {
1695         int ret = 0;
1696         struct btrfs_delayed_data_ref *ref;
1697         struct btrfs_key ins;
1698         u64 parent = 0;
1699         u64 ref_root = 0;
1700         u64 flags = 0;
1701
1702         ins.objectid = node->bytenr;
1703         ins.offset = node->num_bytes;
1704         ins.type = BTRFS_EXTENT_ITEM_KEY;
1705
1706         ref = btrfs_delayed_node_to_data_ref(node);
1707         if (node->type == BTRFS_SHARED_DATA_REF_KEY)
1708                 parent = ref->parent;
1709         else
1710                 ref_root = ref->root;
1711
1712         if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) {
1713                 if (extent_op) {
1714                         BUG_ON(extent_op->update_key);
1715                         flags |= extent_op->flags_to_set;
1716                 }
1717                 ret = alloc_reserved_file_extent(trans, root,
1718                                                  parent, ref_root, flags,
1719                                                  ref->objectid, ref->offset,
1720                                                  &ins, node->ref_mod);
1721         } else if (node->action == BTRFS_ADD_DELAYED_REF) {
1722                 ret = __btrfs_inc_extent_ref(trans, root, node->bytenr,
1723                                              node->num_bytes, parent,
1724                                              ref_root, ref->objectid,
1725                                              ref->offset, node->ref_mod,
1726                                              extent_op);
1727         } else if (node->action == BTRFS_DROP_DELAYED_REF) {
1728                 ret = __btrfs_free_extent(trans, root, node->bytenr,
1729                                           node->num_bytes, parent,
1730                                           ref_root, ref->objectid,
1731                                           ref->offset, node->ref_mod,
1732                                           extent_op);
1733         } else {
1734                 BUG();
1735         }
1736         return ret;
1737 }
1738
1739 static void __run_delayed_extent_op(struct btrfs_delayed_extent_op *extent_op,
1740                                     struct extent_buffer *leaf,
1741                                     struct btrfs_extent_item *ei)
1742 {
1743         u64 flags = btrfs_extent_flags(leaf, ei);
1744         if (extent_op->update_flags) {
1745                 flags |= extent_op->flags_to_set;
1746                 btrfs_set_extent_flags(leaf, ei, flags);
1747         }
1748
1749         if (extent_op->update_key) {
1750                 struct btrfs_tree_block_info *bi;
1751                 BUG_ON(!(flags & BTRFS_EXTENT_FLAG_TREE_BLOCK));
1752                 bi = (struct btrfs_tree_block_info *)(ei + 1);
1753                 btrfs_set_tree_block_key(leaf, bi, &extent_op->key);
1754         }
1755 }
1756
1757 static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
1758                                  struct btrfs_root *root,
1759                                  struct btrfs_delayed_ref_node *node,
1760                                  struct btrfs_delayed_extent_op *extent_op)
1761 {
1762         struct btrfs_key key;
1763         struct btrfs_path *path;
1764         struct btrfs_extent_item *ei;
1765         struct extent_buffer *leaf;
1766         u32 item_size;
1767         int ret;
1768         int err = 0;
1769
1770         path = btrfs_alloc_path();
1771         if (!path)
1772                 return -ENOMEM;
1773
1774         key.objectid = node->bytenr;
1775         key.type = BTRFS_EXTENT_ITEM_KEY;
1776         key.offset = node->num_bytes;
1777
1778         path->reada = 1;
1779         path->leave_spinning = 1;
1780         ret = btrfs_search_slot(trans, root->fs_info->extent_root, &key,
1781                                 path, 0, 1);
1782         if (ret < 0) {
1783                 err = ret;
1784                 goto out;
1785         }
1786         if (ret > 0) {
1787                 err = -EIO;
1788                 goto out;
1789         }
1790
1791         leaf = path->nodes[0];
1792         item_size = btrfs_item_size_nr(leaf, path->slots[0]);
1793 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
1794         if (item_size < sizeof(*ei)) {
1795                 ret = convert_extent_item_v0(trans, root->fs_info->extent_root,
1796                                              path, (u64)-1, 0);
1797                 if (ret < 0) {
1798                         err = ret;
1799                         goto out;
1800                 }
1801                 leaf = path->nodes[0];
1802                 item_size = btrfs_item_size_nr(leaf, path->slots[0]);
1803         }
1804 #endif
1805         BUG_ON(item_size < sizeof(*ei));
1806         ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
1807         __run_delayed_extent_op(extent_op, leaf, ei);
1808
1809         btrfs_mark_buffer_dirty(leaf);
1810 out:
1811         btrfs_free_path(path);
1812         return err;
1813 }
1814
1815 static int run_delayed_tree_ref(struct btrfs_trans_handle *trans,
1816                                 struct btrfs_root *root,
1817                                 struct btrfs_delayed_ref_node *node,
1818                                 struct btrfs_delayed_extent_op *extent_op,
1819                                 int insert_reserved)
1820 {
1821         int ret = 0;
1822         struct btrfs_delayed_tree_ref *ref;
1823         struct btrfs_key ins;
1824         u64 parent = 0;
1825         u64 ref_root = 0;
1826
1827         ins.objectid = node->bytenr;
1828         ins.offset = node->num_bytes;
1829         ins.type = BTRFS_EXTENT_ITEM_KEY;
1830
1831         ref = btrfs_delayed_node_to_tree_ref(node);
1832         if (node->type == BTRFS_SHARED_BLOCK_REF_KEY)
1833                 parent = ref->parent;
1834         else
1835                 ref_root = ref->root;
1836
1837         BUG_ON(node->ref_mod != 1);
1838         if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) {
1839                 BUG_ON(!extent_op || !extent_op->update_flags ||
1840                        !extent_op->update_key);
1841                 ret = alloc_reserved_tree_block(trans, root,
1842                                                 parent, ref_root,
1843                                                 extent_op->flags_to_set,
1844                                                 &extent_op->key,
1845                                                 ref->level, &ins);
1846         } else if (node->action == BTRFS_ADD_DELAYED_REF) {
1847                 ret = __btrfs_inc_extent_ref(trans, root, node->bytenr,
1848                                              node->num_bytes, parent, ref_root,
1849                                              ref->level, 0, 1, extent_op);
1850         } else if (node->action == BTRFS_DROP_DELAYED_REF) {
1851                 ret = __btrfs_free_extent(trans, root, node->bytenr,
1852                                           node->num_bytes, parent, ref_root,
1853                                           ref->level, 0, 1, extent_op);
1854         } else {
1855                 BUG();
1856         }
1857         return ret;
1858 }
1859
1860
1861 /* helper function to actually process a single delayed ref entry */
1862 static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
1863                                struct btrfs_root *root,
1864                                struct btrfs_delayed_ref_node *node,
1865                                struct btrfs_delayed_extent_op *extent_op,
1866                                int insert_reserved)
1867 {
1868         int ret;
1869         if (btrfs_delayed_ref_is_head(node)) {
1870                 struct btrfs_delayed_ref_head *head;
1871                 /*
1872                  * we've hit the end of the chain and we were supposed
1873                  * to insert this extent into the tree.  But, it got
1874                  * deleted before we ever needed to insert it, so all
1875                  * we have to do is clean up the accounting
1876                  */
1877                 BUG_ON(extent_op);
1878                 head = btrfs_delayed_node_to_head(node);
1879                 if (insert_reserved) {
1880                         int mark_free = 0;
1881                         struct extent_buffer *must_clean = NULL;
1882
1883                         ret = pin_down_bytes(trans, root, NULL,
1884                                              node->bytenr, node->num_bytes,
1885                                              head->is_data, 1, &must_clean);
1886                         if (ret > 0)
1887                                 mark_free = 1;
1888
1889                         if (must_clean) {
1890                                 clean_tree_block(NULL, root, must_clean);
1891                                 btrfs_tree_unlock(must_clean);
1892                                 free_extent_buffer(must_clean);
1893                         }
1894                         if (head->is_data) {
1895                                 ret = btrfs_del_csums(trans, root,
1896                                                       node->bytenr,
1897                                                       node->num_bytes);
1898                                 BUG_ON(ret);
1899                         }
1900                         if (mark_free) {
1901                                 ret = btrfs_free_reserved_extent(root,
1902                                                         node->bytenr,
1903                                                         node->num_bytes);
1904                                 BUG_ON(ret);
1905                         }
1906                 }
1907                 mutex_unlock(&head->mutex);
1908                 return 0;
1909         }
1910
1911         if (node->type == BTRFS_TREE_BLOCK_REF_KEY ||
1912             node->type == BTRFS_SHARED_BLOCK_REF_KEY)
1913                 ret = run_delayed_tree_ref(trans, root, node, extent_op,
1914                                            insert_reserved);
1915         else if (node->type == BTRFS_EXTENT_DATA_REF_KEY ||
1916                  node->type == BTRFS_SHARED_DATA_REF_KEY)
1917                 ret = run_delayed_data_ref(trans, root, node, extent_op,
1918                                            insert_reserved);
1919         else
1920                 BUG();
1921         return ret;
1922 }
1923
1924 static noinline struct btrfs_delayed_ref_node *
1925 select_delayed_ref(struct btrfs_delayed_ref_head *head)
1926 {
1927         struct rb_node *node;
1928         struct btrfs_delayed_ref_node *ref;
1929         int action = BTRFS_ADD_DELAYED_REF;
1930 again:
1931         /*
1932          * select delayed ref of type BTRFS_ADD_DELAYED_REF first.
1933          * this prevents ref count from going down to zero when
1934          * there still are pending delayed ref.
1935          */
1936         node = rb_prev(&head->node.rb_node);
1937         while (1) {
1938                 if (!node)
1939                         break;
1940                 ref = rb_entry(node, struct btrfs_delayed_ref_node,
1941                                 rb_node);
1942                 if (ref->bytenr != head->node.bytenr)
1943                         break;
1944                 if (ref->action == action)
1945                         return ref;
1946                 node = rb_prev(node);
1947         }
1948         if (action == BTRFS_ADD_DELAYED_REF) {
1949                 action = BTRFS_DROP_DELAYED_REF;
1950                 goto again;
1951         }
1952         return NULL;
1953 }
1954
1955 static noinline int run_clustered_refs(struct btrfs_trans_handle *trans,
1956                                        struct btrfs_root *root,
1957                                        struct list_head *cluster)
1958 {
1959         struct btrfs_delayed_ref_root *delayed_refs;
1960         struct btrfs_delayed_ref_node *ref;
1961         struct btrfs_delayed_ref_head *locked_ref = NULL;
1962         struct btrfs_delayed_extent_op *extent_op;
1963         int ret;
1964         int count = 0;
1965         int must_insert_reserved = 0;
1966
1967         delayed_refs = &trans->transaction->delayed_refs;
1968         while (1) {
1969                 if (!locked_ref) {
1970                         /* pick a new head ref from the cluster list */
1971                         if (list_empty(cluster))
1972                                 break;
1973
1974                         locked_ref = list_entry(cluster->next,
1975                                      struct btrfs_delayed_ref_head, cluster);
1976
1977                         /* grab the lock that says we are going to process
1978                          * all the refs for this head */
1979                         ret = btrfs_delayed_ref_lock(trans, locked_ref);
1980
1981                         /*
1982                          * we may have dropped the spin lock to get the head
1983                          * mutex lock, and that might have given someone else
1984                          * time to free the head.  If that's true, it has been
1985                          * removed from our list and we can move on.
1986                          */
1987                         if (ret == -EAGAIN) {
1988                                 locked_ref = NULL;
1989                                 count++;
1990                                 continue;
1991                         }
1992                 }
1993
1994                 /*
1995                  * record the must insert reserved flag before we
1996                  * drop the spin lock.
1997                  */
1998                 must_insert_reserved = locked_ref->must_insert_reserved;
1999                 locked_ref->must_insert_reserved = 0;
2000
2001                 extent_op = locked_ref->extent_op;
2002                 locked_ref->extent_op = NULL;
2003
2004                 /*
2005                  * locked_ref is the head node, so we have to go one
2006                  * node back for any delayed ref updates
2007                  */
2008                 ref = select_delayed_ref(locked_ref);
2009                 if (!ref) {
2010                         /* All delayed refs have been processed, Go ahead
2011                          * and send the head node to run_one_delayed_ref,
2012                          * so that any accounting fixes can happen
2013                          */
2014                         ref = &locked_ref->node;
2015
2016                         if (extent_op && must_insert_reserved) {
2017                                 kfree(extent_op);
2018                                 extent_op = NULL;
2019                         }
2020
2021                         if (extent_op) {
2022                                 spin_unlock(&delayed_refs->lock);
2023
2024                                 ret = run_delayed_extent_op(trans, root,
2025                                                             ref, extent_op);
2026                                 BUG_ON(ret);
2027                                 kfree(extent_op);
2028
2029                                 cond_resched();
2030                                 spin_lock(&delayed_refs->lock);
2031                                 continue;
2032                         }
2033
2034                         list_del_init(&locked_ref->cluster);
2035                         locked_ref = NULL;
2036                 }
2037
2038                 ref->in_tree = 0;
2039                 rb_erase(&ref->rb_node, &delayed_refs->root);
2040                 delayed_refs->num_entries--;
2041
2042                 spin_unlock(&delayed_refs->lock);
2043
2044                 ret = run_one_delayed_ref(trans, root, ref, extent_op,
2045                                           must_insert_reserved);
2046                 BUG_ON(ret);
2047
2048                 btrfs_put_delayed_ref(ref);
2049                 kfree(extent_op);
2050                 count++;
2051
2052                 cond_resched();
2053                 spin_lock(&delayed_refs->lock);
2054         }
2055         return count;
2056 }
2057
2058 /*
2059  * this starts processing the delayed reference count updates and
2060  * extent insertions we have queued up so far.  count can be
2061  * 0, which means to process everything in the tree at the start
2062  * of the run (but not newly added entries), or it can be some target
2063  * number you'd like to process.
2064  */
2065 int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
2066                            struct btrfs_root *root, unsigned long count)
2067 {
2068         struct rb_node *node;
2069         struct btrfs_delayed_ref_root *delayed_refs;
2070         struct btrfs_delayed_ref_node *ref;
2071         struct list_head cluster;
2072         int ret;
2073         int run_all = count == (unsigned long)-1;
2074         int run_most = 0;
2075
2076         if (root == root->fs_info->extent_root)
2077                 root = root->fs_info->tree_root;
2078
2079         delayed_refs = &trans->transaction->delayed_refs;
2080         INIT_LIST_HEAD(&cluster);
2081 again:
2082         spin_lock(&delayed_refs->lock);
2083         if (count == 0) {
2084                 count = delayed_refs->num_entries * 2;
2085                 run_most = 1;
2086         }
2087         while (1) {
2088                 if (!(run_all || run_most) &&
2089                     delayed_refs->num_heads_ready < 64)
2090                         break;
2091
2092                 /*
2093                  * go find something we can process in the rbtree.  We start at
2094                  * the beginning of the tree, and then build a cluster
2095                  * of refs to process starting at the first one we are able to
2096                  * lock
2097                  */
2098                 ret = btrfs_find_ref_cluster(trans, &cluster,
2099                                              delayed_refs->run_delayed_start);
2100                 if (ret)
2101                         break;
2102
2103                 ret = run_clustered_refs(trans, root, &cluster);
2104                 BUG_ON(ret < 0);
2105
2106                 count -= min_t(unsigned long, ret, count);
2107
2108                 if (count == 0)
2109                         break;
2110         }
2111
2112         if (run_all) {
2113                 node = rb_first(&delayed_refs->root);
2114                 if (!node)
2115                         goto out;
2116                 count = (unsigned long)-1;
2117
2118                 while (node) {
2119                         ref = rb_entry(node, struct btrfs_delayed_ref_node,
2120                                        rb_node);
2121                         if (btrfs_delayed_ref_is_head(ref)) {
2122                                 struct btrfs_delayed_ref_head *head;
2123
2124                                 head = btrfs_delayed_node_to_head(ref);
2125                                 atomic_inc(&ref->refs);
2126
2127                                 spin_unlock(&delayed_refs->lock);
2128                                 mutex_lock(&head->mutex);
2129                                 mutex_unlock(&head->mutex);
2130
2131                                 btrfs_put_delayed_ref(ref);
2132                                 cond_resched();
2133                                 goto again;
2134                         }
2135                         node = rb_next(node);
2136                 }
2137                 spin_unlock(&delayed_refs->lock);
2138                 schedule_timeout(1);
2139                 goto again;
2140         }
2141 out:
2142         spin_unlock(&delayed_refs->lock);
2143         return 0;
2144 }
2145
2146 int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
2147                                 struct btrfs_root *root,
2148                                 u64 bytenr, u64 num_bytes, u64 flags,
2149                                 int is_data)
2150 {
2151         struct btrfs_delayed_extent_op *extent_op;
2152         int ret;
2153
2154         extent_op = kmalloc(sizeof(*extent_op), GFP_NOFS);
2155         if (!extent_op)
2156                 return -ENOMEM;
2157
2158         extent_op->flags_to_set = flags;
2159         extent_op->update_flags = 1;
2160         extent_op->update_key = 0;
2161         extent_op->is_data = is_data ? 1 : 0;
2162
2163         ret = btrfs_add_delayed_extent_op(trans, bytenr, num_bytes, extent_op);
2164         if (ret)
2165                 kfree(extent_op);
2166         return ret;
2167 }
2168
2169 static noinline int check_delayed_ref(struct btrfs_trans_handle *trans,
2170                                       struct btrfs_root *root,
2171                                       struct btrfs_path *path,
2172                                       u64 objectid, u64 offset, u64 bytenr)
2173 {
2174         struct btrfs_delayed_ref_head *head;
2175         struct btrfs_delayed_ref_node *ref;
2176         struct btrfs_delayed_data_ref *data_ref;
2177         struct btrfs_delayed_ref_root *delayed_refs;
2178         struct rb_node *node;
2179         int ret = 0;
2180
2181         ret = -ENOENT;
2182         delayed_refs = &trans->transaction->delayed_refs;
2183         spin_lock(&delayed_refs->lock);
2184         head = btrfs_find_delayed_ref_head(trans, bytenr);
2185         if (!head)
2186                 goto out;
2187
2188         if (!mutex_trylock(&head->mutex)) {
2189                 atomic_inc(&head->node.refs);
2190                 spin_unlock(&delayed_refs->lock);
2191
2192                 btrfs_release_path(root->fs_info->extent_root, path);
2193
2194                 mutex_lock(&head->mutex);
2195                 mutex_unlock(&head->mutex);
2196                 btrfs_put_delayed_ref(&head->node);
2197                 return -EAGAIN;
2198         }
2199
2200         node = rb_prev(&head->node.rb_node);
2201         if (!node)
2202                 goto out_unlock;
2203
2204         ref = rb_entry(node, struct btrfs_delayed_ref_node, rb_node);
2205
2206         if (ref->bytenr != bytenr)
2207                 goto out_unlock;
2208
2209         ret = 1;
2210         if (ref->type != BTRFS_EXTENT_DATA_REF_KEY)
2211                 goto out_unlock;
2212
2213         data_ref = btrfs_delayed_node_to_data_ref(ref);
2214
2215         node = rb_prev(node);
2216         if (node) {
2217                 ref = rb_entry(node, struct btrfs_delayed_ref_node, rb_node);
2218                 if (ref->bytenr == bytenr)
2219                         goto out_unlock;
2220         }
2221
2222         if (data_ref->root != root->root_key.objectid ||
2223             data_ref->objectid != objectid || data_ref->offset != offset)
2224                 goto out_unlock;
2225
2226         ret = 0;
2227 out_unlock:
2228         mutex_unlock(&head->mutex);
2229 out:
2230         spin_unlock(&delayed_refs->lock);
2231         return ret;
2232 }
2233
2234 static noinline int check_committed_ref(struct btrfs_trans_handle *trans,
2235                                         struct btrfs_root *root,
2236                                         struct btrfs_path *path,
2237                                         u64 objectid, u64 offset, u64 bytenr)
2238 {
2239         struct btrfs_root *extent_root = root->fs_info->extent_root;
2240         struct extent_buffer *leaf;
2241         struct btrfs_extent_data_ref *ref;
2242         struct btrfs_extent_inline_ref *iref;
2243         struct btrfs_extent_item *ei;
2244         struct btrfs_key key;
2245         u32 item_size;
2246         int ret;
2247
2248         key.objectid = bytenr;
2249         key.offset = (u64)-1;
2250         key.type = BTRFS_EXTENT_ITEM_KEY;
2251
2252         ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0);
2253         if (ret < 0)
2254                 goto out;
2255         BUG_ON(ret == 0);
2256
2257         ret = -ENOENT;
2258         if (path->slots[0] == 0)
2259                 goto out;
2260
2261         path->slots[0]--;
2262         leaf = path->nodes[0];
2263         btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
2264
2265         if (key.objectid != bytenr || key.type != BTRFS_EXTENT_ITEM_KEY)
2266                 goto out;
2267
2268         ret = 1;
2269         item_size = btrfs_item_size_nr(leaf, path->slots[0]);
2270 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
2271         if (item_size < sizeof(*ei)) {
2272                 WARN_ON(item_size != sizeof(struct btrfs_extent_item_v0));
2273                 goto out;
2274         }
2275 #endif
2276         ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
2277
2278         if (item_size != sizeof(*ei) +
2279             btrfs_extent_inline_ref_size(BTRFS_EXTENT_DATA_REF_KEY))
2280                 goto out;
2281
2282         if (btrfs_extent_generation(leaf, ei) <=
2283             btrfs_root_last_snapshot(&root->root_item))
2284                 goto out;
2285
2286         iref = (struct btrfs_extent_inline_ref *)(ei + 1);
2287         if (btrfs_extent_inline_ref_type(leaf, iref) !=
2288             BTRFS_EXTENT_DATA_REF_KEY)
2289                 goto out;
2290
2291         ref = (struct btrfs_extent_data_ref *)(&iref->offset);
2292         if (btrfs_extent_refs(leaf, ei) !=
2293             btrfs_extent_data_ref_count(leaf, ref) ||
2294             btrfs_extent_data_ref_root(leaf, ref) !=
2295             root->root_key.objectid ||
2296             btrfs_extent_data_ref_objectid(leaf, ref) != objectid ||
2297             btrfs_extent_data_ref_offset(leaf, ref) != offset)
2298                 goto out;
2299
2300         ret = 0;
2301 out:
2302         return ret;
2303 }
2304
2305 int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
2306                           struct btrfs_root *root,
2307                           u64 objectid, u64 offset, u64 bytenr)
2308 {
2309         struct btrfs_path *path;
2310         int ret;
2311         int ret2;
2312
2313         path = btrfs_alloc_path();
2314         if (!path)
2315                 return -ENOENT;
2316
2317         do {
2318                 ret = check_committed_ref(trans, root, path, objectid,
2319                                           offset, bytenr);
2320                 if (ret && ret != -ENOENT)
2321                         goto out;
2322
2323                 ret2 = check_delayed_ref(trans, root, path, objectid,
2324                                          offset, bytenr);
2325         } while (ret2 == -EAGAIN);
2326
2327         if (ret2 && ret2 != -ENOENT) {
2328                 ret = ret2;
2329                 goto out;
2330         }
2331
2332         if (ret != -ENOENT || ret2 != -ENOENT)
2333                 ret = 0;
2334 out:
2335         btrfs_free_path(path);
2336         return ret;
2337 }
2338
2339 #if 0
2340 int btrfs_cache_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
2341                     struct extent_buffer *buf, u32 nr_extents)
2342 {
2343         struct btrfs_key key;
2344         struct btrfs_file_extent_item *fi;
2345         u64 root_gen;
2346         u32 nritems;
2347         int i;
2348         int level;
2349         int ret = 0;
2350         int shared = 0;
2351
2352         if (!root->ref_cows)
2353                 return 0;
2354
2355         if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
2356                 shared = 0;
2357                 root_gen = root->root_key.offset;
2358         } else {
2359                 shared = 1;
2360                 root_gen = trans->transid - 1;
2361         }
2362
2363         level = btrfs_header_level(buf);
2364         nritems = btrfs_header_nritems(buf);
2365
2366         if (level == 0) {
2367                 struct btrfs_leaf_ref *ref;
2368                 struct btrfs_extent_info *info;
2369
2370                 ref = btrfs_alloc_leaf_ref(root, nr_extents);
2371                 if (!ref) {
2372                         ret = -ENOMEM;
2373                         goto out;
2374                 }
2375
2376                 ref->root_gen = root_gen;
2377                 ref->bytenr = buf->start;
2378                 ref->owner = btrfs_header_owner(buf);
2379                 ref->generation = btrfs_header_generation(buf);
2380                 ref->nritems = nr_extents;
2381                 info = ref->extents;
2382
2383                 for (i = 0; nr_extents > 0 && i < nritems; i++) {
2384                         u64 disk_bytenr;
2385                         btrfs_item_key_to_cpu(buf, &key, i);
2386                         if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
2387                                 continue;
2388                         fi = btrfs_item_ptr(buf, i,
2389                                             struct btrfs_file_extent_item);
2390                         if (btrfs_file_extent_type(buf, fi) ==
2391                             BTRFS_FILE_EXTENT_INLINE)
2392                                 continue;
2393                         disk_bytenr = btrfs_file_extent_disk_bytenr(buf, fi);
2394                         if (disk_bytenr == 0)
2395                                 continue;
2396
2397                         info->bytenr = disk_bytenr;
2398                         info->num_bytes =
2399                                 btrfs_file_extent_disk_num_bytes(buf, fi);
2400                         info->objectid = key.objectid;
2401                         info->offset = key.offset;
2402                         info++;
2403                 }
2404
2405                 ret = btrfs_add_leaf_ref(root, ref, shared);
2406                 if (ret == -EEXIST && shared) {
2407                         struct btrfs_leaf_ref *old;
2408                         old = btrfs_lookup_leaf_ref(root, ref->bytenr);
2409                         BUG_ON(!old);
2410                         btrfs_remove_leaf_ref(root, old);
2411                         btrfs_free_leaf_ref(root, old);
2412                         ret = btrfs_add_leaf_ref(root, ref, shared);
2413                 }
2414                 WARN_ON(ret);
2415                 btrfs_free_leaf_ref(root, ref);
2416         }
2417 out:
2418         return ret;
2419 }
2420
2421 /* when a block goes through cow, we update the reference counts of
2422  * everything that block points to.  The internal pointers of the block
2423  * can be in just about any order, and it is likely to have clusters of
2424  * things that are close together and clusters of things that are not.
2425  *
2426  * To help reduce the seeks that come with updating all of these reference
2427  * counts, sort them by byte number before actual updates are done.
2428  *
2429  * struct refsort is used to match byte number to slot in the btree block.
2430  * we sort based on the byte number and then use the slot to actually
2431  * find the item.
2432  *
2433  * struct refsort is smaller than strcut btrfs_item and smaller than
2434  * struct btrfs_key_ptr.  Since we're currently limited to the page size
2435  * for a btree block, there's no way for a kmalloc of refsorts for a
2436  * single node to be bigger than a page.
2437  */
2438 struct refsort {
2439         u64 bytenr;
2440         u32 slot;
2441 };
2442
2443 /*
2444  * for passing into sort()
2445  */
2446 static int refsort_cmp(const void *a_void, const void *b_void)
2447 {
2448         const struct refsort *a = a_void;
2449         const struct refsort *b = b_void;
2450
2451         if (a->bytenr < b->bytenr)
2452                 return -1;
2453         if (a->bytenr > b->bytenr)
2454                 return 1;
2455         return 0;
2456 }
2457 #endif
2458
2459 static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
2460                            struct btrfs_root *root,
2461                            struct extent_buffer *buf,
2462                            int full_backref, int inc)
2463 {
2464         u64 bytenr;
2465         u64 num_bytes;
2466         u64 parent;
2467         u64 ref_root;
2468         u32 nritems;
2469         struct btrfs_key key;
2470         struct btrfs_file_extent_item *fi;
2471         int i;
2472         int level;
2473         int ret = 0;
2474         int (*process_func)(struct btrfs_trans_handle *, struct btrfs_root *,
2475                             u64, u64, u64, u64, u64, u64);
2476
2477         ref_root = btrfs_header_owner(buf);
2478         nritems = btrfs_header_nritems(buf);
2479         level = btrfs_header_level(buf);
2480
2481         if (!root->ref_cows && level == 0)
2482                 return 0;
2483
2484         if (inc)
2485                 process_func = btrfs_inc_extent_ref;
2486         else
2487                 process_func = btrfs_free_extent;
2488
2489         if (full_backref)
2490                 parent = buf->start;
2491         else
2492                 parent = 0;
2493
2494         for (i = 0; i < nritems; i++) {
2495                 if (level == 0) {
2496                         btrfs_item_key_to_cpu(buf, &key, i);
2497                         if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
2498                                 continue;
2499                         fi = btrfs_item_ptr(buf, i,
2500                                             struct btrfs_file_extent_item);
2501                         if (btrfs_file_extent_type(buf, fi) ==
2502                             BTRFS_FILE_EXTENT_INLINE)
2503                                 continue;
2504                         bytenr = btrfs_file_extent_disk_bytenr(buf, fi);
2505                         if (bytenr == 0)
2506                                 continue;
2507
2508                         num_bytes = btrfs_file_extent_disk_num_bytes(buf, fi);
2509                         key.offset -= btrfs_file_extent_offset(buf, fi);
2510                         ret = process_func(trans, root, bytenr, num_bytes,
2511                                            parent, ref_root, key.objectid,
2512                                            key.offset);
2513                         if (ret)
2514                                 goto fail;
2515                 } else {
2516                         bytenr = btrfs_node_blockptr(buf, i);
2517                         num_bytes = btrfs_level_size(root, level - 1);
2518                         ret = process_func(trans, root, bytenr, num_bytes,
2519                                            parent, ref_root, level - 1, 0);
2520                         if (ret)
2521                                 goto fail;
2522                 }
2523         }
2524         return 0;
2525 fail:
2526         BUG();
2527         return ret;
2528 }
2529
2530 int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
2531                   struct extent_buffer *buf, int full_backref)
2532 {
2533         return __btrfs_mod_ref(trans, root, buf, full_backref, 1);
2534 }
2535
2536 int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
2537                   struct extent_buffer *buf, int full_backref)
2538 {
2539         return __btrfs_mod_ref(trans, root, buf, full_backref, 0);
2540 }
2541
2542 static int write_one_cache_group(struct btrfs_trans_handle *trans,
2543                                  struct btrfs_root *root,
2544                                  struct btrfs_path *path,
2545                                  struct btrfs_block_group_cache *cache)
2546 {
2547         int ret;
2548         struct btrfs_root *extent_root = root->fs_info->extent_root;
2549         unsigned long bi;
2550         struct extent_buffer *leaf;
2551
2552         ret = btrfs_search_slot(trans, extent_root, &cache->key, path, 0, 1);
2553         if (ret < 0)
2554                 goto fail;
2555         BUG_ON(ret);
2556
2557         leaf = path->nodes[0];
2558         bi = btrfs_item_ptr_offset(leaf, path->slots[0]);
2559         write_extent_buffer(leaf, &cache->item, bi, sizeof(cache->item));
2560         btrfs_mark_buffer_dirty(leaf);
2561         btrfs_release_path(extent_root, path);
2562 fail:
2563         if (ret)
2564                 return ret;
2565         return 0;
2566
2567 }
2568
2569 static struct btrfs_block_group_cache *
2570 next_block_group(struct btrfs_root *root,
2571                  struct btrfs_block_group_cache *cache)
2572 {
2573         struct rb_node *node;
2574         spin_lock(&root->fs_info->block_group_cache_lock);
2575         node = rb_next(&cache->cache_node);
2576         btrfs_put_block_group(cache);
2577         if (node) {
2578                 cache = rb_entry(node, struct btrfs_block_group_cache,
2579                                  cache_node);
2580                 atomic_inc(&cache->count);
2581         } else
2582                 cache = NULL;
2583         spin_unlock(&root->fs_info->block_group_cache_lock);
2584         return cache;
2585 }
2586
2587 int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans,
2588                                    struct btrfs_root *root)
2589 {
2590         struct btrfs_block_group_cache *cache;
2591         int err = 0;
2592         struct btrfs_path *path;
2593         u64 last = 0;
2594
2595         path = btrfs_alloc_path();
2596         if (!path)
2597                 return -ENOMEM;
2598
2599         while (1) {
2600                 if (last == 0) {
2601                         err = btrfs_run_delayed_refs(trans, root,
2602                                                      (unsigned long)-1);
2603                         BUG_ON(err);
2604                 }
2605
2606                 cache = btrfs_lookup_first_block_group(root->fs_info, last);
2607                 while (cache) {
2608                         if (cache->dirty)
2609                                 break;
2610                         cache = next_block_group(root, cache);
2611                 }
2612                 if (!cache) {
2613                         if (last == 0)
2614                                 break;
2615                         last = 0;
2616                         continue;
2617                 }
2618
2619                 cache->dirty = 0;
2620                 last = cache->key.objectid + cache->key.offset;
2621
2622                 err = write_one_cache_group(trans, root, path, cache);
2623                 BUG_ON(err);
2624                 btrfs_put_block_group(cache);
2625         }
2626
2627         btrfs_free_path(path);
2628         return 0;
2629 }
2630
2631 int btrfs_extent_readonly(struct btrfs_root *root, u64 bytenr)
2632 {
2633         struct btrfs_block_group_cache *block_group;
2634         int readonly = 0;
2635
2636         block_group = btrfs_lookup_block_group(root->fs_info, bytenr);
2637         if (!block_group || block_group->ro)
2638                 readonly = 1;
2639         if (block_group)
2640                 btrfs_put_block_group(block_group);
2641         return readonly;
2642 }
2643
2644 static int update_space_info(struct btrfs_fs_info *info, u64 flags,
2645                              u64 total_bytes, u64 bytes_used,
2646                              struct btrfs_space_info **space_info)
2647 {
2648         struct btrfs_space_info *found;
2649
2650         found = __find_space_info(info, flags);
2651         if (found) {
2652                 spin_lock(&found->lock);
2653                 found->total_bytes += total_bytes;
2654                 found->bytes_used += bytes_used;
2655                 found->full = 0;
2656                 spin_unlock(&found->lock);
2657                 *space_info = found;
2658                 return 0;
2659         }
2660         found = kzalloc(sizeof(*found), GFP_NOFS);
2661         if (!found)
2662                 return -ENOMEM;
2663
2664         INIT_LIST_HEAD(&found->block_groups);
2665         init_rwsem(&found->groups_sem);
2666         spin_lock_init(&found->lock);
2667         found->flags = flags;
2668         found->total_bytes = total_bytes;
2669         found->bytes_used = bytes_used;
2670         found->bytes_pinned = 0;
2671         found->bytes_reserved = 0;
2672         found->bytes_readonly = 0;
2673         found->bytes_delalloc = 0;
2674         found->full = 0;
2675         found->force_alloc = 0;
2676         *space_info = found;
2677         list_add_rcu(&found->list, &info->space_info);
2678         atomic_set(&found->caching_threads, 0);
2679         return 0;
2680 }
2681
2682 static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
2683 {
2684         u64 extra_flags = flags & (BTRFS_BLOCK_GROUP_RAID0 |
2685                                    BTRFS_BLOCK_GROUP_RAID1 |
2686                                    BTRFS_BLOCK_GROUP_RAID10 |
2687                                    BTRFS_BLOCK_GROUP_DUP);
2688         if (extra_flags) {
2689                 if (flags & BTRFS_BLOCK_GROUP_DATA)
2690                         fs_info->avail_data_alloc_bits |= extra_flags;
2691                 if (flags & BTRFS_BLOCK_GROUP_METADATA)
2692                         fs_info->avail_metadata_alloc_bits |= extra_flags;
2693                 if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
2694                         fs_info->avail_system_alloc_bits |= extra_flags;
2695         }
2696 }
2697
2698 static void set_block_group_readonly(struct btrfs_block_group_cache *cache)
2699 {
2700         spin_lock(&cache->space_info->lock);
2701         spin_lock(&cache->lock);
2702         if (!cache->ro) {
2703                 cache->space_info->bytes_readonly += cache->key.offset -
2704                                         btrfs_block_group_used(&cache->item);
2705                 cache->ro = 1;
2706         }
2707         spin_unlock(&cache->lock);
2708         spin_unlock(&cache->space_info->lock);
2709 }
2710
2711 u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
2712 {
2713         u64 num_devices = root->fs_info->fs_devices->rw_devices;
2714
2715         if (num_devices == 1)
2716                 flags &= ~(BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID0);
2717         if (num_devices < 4)
2718                 flags &= ~BTRFS_BLOCK_GROUP_RAID10;
2719
2720         if ((flags & BTRFS_BLOCK_GROUP_DUP) &&
2721             (flags & (BTRFS_BLOCK_GROUP_RAID1 |
2722                       BTRFS_BLOCK_GROUP_RAID10))) {
2723                 flags &= ~BTRFS_BLOCK_GROUP_DUP;
2724         }
2725
2726         if ((flags & BTRFS_BLOCK_GROUP_RAID1) &&
2727             (flags & BTRFS_BLOCK_GROUP_RAID10)) {
2728                 flags &= ~BTRFS_BLOCK_GROUP_RAID1;
2729         }
2730
2731         if ((flags & BTRFS_BLOCK_GROUP_RAID0) &&
2732             ((flags & BTRFS_BLOCK_GROUP_RAID1) |
2733              (flags & BTRFS_BLOCK_GROUP_RAID10) |
2734              (flags & BTRFS_BLOCK_GROUP_DUP)))
2735                 flags &= ~BTRFS_BLOCK_GROUP_RAID0;
2736         return flags;
2737 }
2738
2739 static u64 btrfs_get_alloc_profile(struct btrfs_root *root, u64 data)
2740 {
2741         struct btrfs_fs_info *info = root->fs_info;
2742         u64 alloc_profile;
2743
2744         if (data) {
2745                 alloc_profile = info->avail_data_alloc_bits &
2746                         info->data_alloc_profile;
2747                 data = BTRFS_BLOCK_GROUP_DATA | alloc_profile;
2748         } else if (root == root->fs_info->chunk_root) {
2749                 alloc_profile = info->avail_system_alloc_bits &
2750                         info->system_alloc_profile;
2751                 data = BTRFS_BLOCK_GROUP_SYSTEM | alloc_profile;
2752         } else {
2753                 alloc_profile = info->avail_metadata_alloc_bits &
2754                         info->metadata_alloc_profile;
2755                 data = BTRFS_BLOCK_GROUP_METADATA | alloc_profile;
2756         }
2757
2758         return btrfs_reduce_alloc_profile(root, data);
2759 }
2760
2761 void btrfs_set_inode_space_info(struct btrfs_root *root, struct inode *inode)
2762 {
2763         u64 alloc_target;
2764
2765         alloc_target = btrfs_get_alloc_profile(root, 1);
2766         BTRFS_I(inode)->space_info = __find_space_info(root->fs_info,
2767                                                        alloc_target);
2768 }
2769
2770 static u64 calculate_bytes_needed(struct btrfs_root *root, int num_items)
2771 {
2772         u64 num_bytes;
2773         int level;
2774
2775         level = BTRFS_MAX_LEVEL - 2;
2776         /*
2777          * NOTE: these calculations are absolutely the worst possible case.
2778          * This assumes that _every_ item we insert will require a new leaf, and
2779          * that the tree has grown to its maximum level size.
2780          */
2781
2782         /*
2783          * for every item we insert we could insert both an extent item and a
2784          * extent ref item.  Then for ever item we insert, we will need to cow
2785          * both the original leaf, plus the leaf to the left and right of it.
2786          *
2787          * Unless we are talking about the extent root, then we just want the
2788          * number of items * 2, since we just need the extent item plus its ref.
2789          */
2790         if (root == root->fs_info->extent_root)
2791                 num_bytes = num_items * 2;
2792         else
2793                 num_bytes = (num_items + (2 * num_items)) * 3;
2794
2795         /*
2796          * num_bytes is total number of leaves we could need times the leaf
2797          * size, and then for every leaf we could end up cow'ing 2 nodes per
2798          * level, down to the leaf level.
2799          */
2800         num_bytes = (num_bytes * root->leafsize) +
2801                 (num_bytes * (level * 2)) * root->nodesize;
2802
2803         return num_bytes;
2804 }
2805
2806 /*
2807  * Unreserve metadata space for delalloc.  If we have less reserved credits than
2808  * we have extents, this function does nothing.
2809  */
2810 int btrfs_unreserve_metadata_for_delalloc(struct btrfs_root *root,
2811                                           struct inode *inode, int num_items)
2812 {
2813         struct btrfs_fs_info *info = root->fs_info;
2814         struct btrfs_space_info *meta_sinfo;
2815         u64 num_bytes;
2816         u64 alloc_target;
2817         bool bug = false;
2818
2819         /* get the space info for where the metadata will live */
2820         alloc_target = btrfs_get_alloc_profile(root, 0);
2821         meta_sinfo = __find_space_info(info, alloc_target);
2822
2823         num_bytes = calculate_bytes_needed(root->fs_info->extent_root,
2824                                            num_items);
2825
2826         spin_lock(&meta_sinfo->lock);
2827         if (BTRFS_I(inode)->delalloc_reserved_extents <=
2828             BTRFS_I(inode)->delalloc_extents) {
2829                 spin_unlock(&meta_sinfo->lock);
2830                 return 0;
2831         }
2832
2833         BTRFS_I(inode)->delalloc_reserved_extents--;
2834         BUG_ON(BTRFS_I(inode)->delalloc_reserved_extents < 0);
2835
2836         if (meta_sinfo->bytes_delalloc < num_bytes) {
2837                 bug = true;
2838                 meta_sinfo->bytes_delalloc = 0;
2839         } else {
2840                 meta_sinfo->bytes_delalloc -= num_bytes;
2841         }
2842         spin_unlock(&meta_sinfo->lock);
2843
2844         BUG_ON(bug);
2845
2846         return 0;
2847 }
2848
2849 static void check_force_delalloc(struct btrfs_space_info *meta_sinfo)
2850 {
2851         u64 thresh;
2852
2853         thresh = meta_sinfo->bytes_used + meta_sinfo->bytes_reserved +
2854                 meta_sinfo->bytes_pinned + meta_sinfo->bytes_readonly +
2855                 meta_sinfo->bytes_super + meta_sinfo->bytes_root +
2856                 meta_sinfo->bytes_may_use;
2857
2858         thresh = meta_sinfo->total_bytes - thresh;
2859         thresh *= 80;
2860         do_div(thresh, 100);
2861         if (thresh <= meta_sinfo->bytes_delalloc)
2862                 meta_sinfo->force_delalloc = 1;
2863         else
2864                 meta_sinfo->force_delalloc = 0;
2865 }
2866
2867 static int maybe_allocate_chunk(struct btrfs_root *root,
2868                                  struct btrfs_space_info *info)
2869 {
2870         struct btrfs_super_block *disk_super = &root->fs_info->super_copy;
2871         struct btrfs_trans_handle *trans;
2872         bool wait = false;
2873         int ret = 0;
2874         u64 min_metadata;
2875         u64 free_space;
2876
2877         free_space = btrfs_super_total_bytes(disk_super);
2878         /*
2879          * we allow the metadata to grow to a max of either 5gb or 5% of the
2880          * space in the volume.
2881          */
2882         min_metadata = min((u64)5 * 1024 * 1024 * 1024,
2883                              div64_u64(free_space * 5, 100));
2884         if (info->total_bytes >= min_metadata) {
2885                 spin_unlock(&info->lock);
2886                 return 0;
2887         }
2888
2889         if (info->full) {
2890                 spin_unlock(&info->lock);
2891                 return 0;
2892         }
2893
2894         if (!info->allocating_chunk) {
2895                 info->force_alloc = 1;
2896                 info->allocating_chunk = 1;
2897                 init_waitqueue_head(&info->wait);
2898         } else {
2899                 wait = true;
2900         }
2901
2902         spin_unlock(&info->lock);
2903
2904         if (wait) {
2905                 wait_event(info->wait,
2906                            !info->allocating_chunk);
2907                 return 1;
2908         }
2909
2910         trans = btrfs_start_transaction(root, 1);
2911         if (!trans) {
2912                 ret = -ENOMEM;
2913                 goto out;
2914         }
2915
2916         ret = do_chunk_alloc(trans, root->fs_info->extent_root,
2917                              4096 + 2 * 1024 * 1024,
2918                              info->flags, 0);
2919         btrfs_end_transaction(trans, root);
2920         if (ret)
2921                 goto out;
2922 out:
2923         spin_lock(&info->lock);
2924         info->allocating_chunk = 0;
2925         spin_unlock(&info->lock);
2926         wake_up(&info->wait);
2927
2928         if (ret)
2929                 return 0;
2930         return 1;
2931 }
2932
2933 /*
2934  * Reserve metadata space for delalloc.
2935  */
2936 int btrfs_reserve_metadata_for_delalloc(struct btrfs_root *root,
2937                                         struct inode *inode, int num_items)
2938 {
2939         struct btrfs_fs_info *info = root->fs_info;
2940         struct btrfs_space_info *meta_sinfo;
2941         u64 num_bytes;
2942         u64 used;
2943         u64 alloc_target;
2944         int flushed = 0;
2945         int force_delalloc;
2946
2947         /* get the space info for where the metadata will live */
2948         alloc_target = btrfs_get_alloc_profile(root, 0);
2949         meta_sinfo = __find_space_info(info, alloc_target);
2950
2951         num_bytes = calculate_bytes_needed(root->fs_info->extent_root,
2952                                            num_items);
2953 again:
2954         spin_lock(&meta_sinfo->lock);
2955
2956         force_delalloc = meta_sinfo->force_delalloc;
2957
2958         if (unlikely(!meta_sinfo->bytes_root))
2959                 meta_sinfo->bytes_root = calculate_bytes_needed(root, 6);
2960
2961         if (!flushed)
2962                 meta_sinfo->bytes_delalloc += num_bytes;
2963
2964         used = meta_sinfo->bytes_used + meta_sinfo->bytes_reserved +
2965                 meta_sinfo->bytes_pinned + meta_sinfo->bytes_readonly +
2966                 meta_sinfo->bytes_super + meta_sinfo->bytes_root +
2967                 meta_sinfo->bytes_may_use + meta_sinfo->bytes_delalloc;
2968
2969         if (used > meta_sinfo->total_bytes) {
2970                 flushed++;
2971
2972                 if (flushed == 1) {
2973                         if (maybe_allocate_chunk(root, meta_sinfo))
2974                                 goto again;
2975                         flushed++;
2976                 } else {
2977                         spin_unlock(&meta_sinfo->lock);
2978                 }
2979
2980                 if (flushed == 2) {
2981                         filemap_flush(inode->i_mapping);
2982                         goto again;
2983                 } else if (flushed == 3) {
2984                         btrfs_start_delalloc_inodes(root);
2985                         btrfs_wait_ordered_extents(root, 0);
2986                         goto again;
2987                 }
2988                 spin_lock(&meta_sinfo->lock);
2989                 meta_sinfo->bytes_delalloc -= num_bytes;
2990                 spin_unlock(&meta_sinfo->lock);
2991                 printk(KERN_ERR "enospc, has %d, reserved %d\n",
2992                        BTRFS_I(inode)->delalloc_extents,
2993                        BTRFS_I(inode)->delalloc_reserved_extents);
2994                 dump_space_info(meta_sinfo, 0, 0);
2995                 return -ENOSPC;
2996         }
2997
2998         BTRFS_I(inode)->delalloc_reserved_extents++;
2999         check_force_delalloc(meta_sinfo);
3000         spin_unlock(&meta_sinfo->lock);
3001
3002         if (!flushed && force_delalloc)
3003                 filemap_flush(inode->i_mapping);
3004
3005         return 0;
3006 }
3007
3008 /*
3009  * unreserve num_items number of items worth of metadata space.  This needs to
3010  * be paired with btrfs_reserve_metadata_space.
3011  *
3012  * NOTE: if you have the option, run this _AFTER_ you do a
3013  * btrfs_end_transaction, since btrfs_end_transaction will run delayed ref
3014  * oprations which will result in more used metadata, so we want to make sure we
3015  * can do that without issue.
3016  */
3017 int btrfs_unreserve_metadata_space(struct btrfs_root *root, int num_items)
3018 {
3019         struct btrfs_fs_info *info = root->fs_info;
3020         struct btrfs_space_info *meta_sinfo;
3021         u64 num_bytes;
3022         u64 alloc_target;
3023         bool bug = false;
3024
3025         /* get the space info for where the metadata will live */
3026         alloc_target = btrfs_get_alloc_profile(root, 0);
3027         meta_sinfo = __find_space_info(info, alloc_target);
3028
3029         num_bytes = calculate_bytes_needed(root, num_items);
3030
3031         spin_lock(&meta_sinfo->lock);
3032         if (meta_sinfo->bytes_may_use < num_bytes) {
3033                 bug = true;
3034                 meta_sinfo->bytes_may_use = 0;
3035         } else {
3036                 meta_sinfo->bytes_may_use -= num_bytes;
3037         }
3038         spin_unlock(&meta_sinfo->lock);
3039
3040         BUG_ON(bug);
3041
3042         return 0;
3043 }
3044
3045 /*
3046  * Reserve some metadata space for use.  We'll calculate the worste case number
3047  * of bytes that would be needed to modify num_items number of items.  If we
3048  * have space, fantastic, if not, you get -ENOSPC.  Please call
3049  * btrfs_unreserve_metadata_space when you are done for the _SAME_ number of
3050  * items you reserved, since whatever metadata you needed should have already
3051  * been allocated.
3052  *
3053  * This will commit the transaction to make more space if we don't have enough
3054  * metadata space.  THe only time we don't do this is if we're reserving space
3055  * inside of a transaction, then we will just return -ENOSPC and it is the
3056  * callers responsibility to handle it properly.
3057  */
3058 int btrfs_reserve_metadata_space(struct btrfs_root *root, int num_items)
3059 {
3060         struct btrfs_fs_info *info = root->fs_info;
3061         struct btrfs_space_info *meta_sinfo;
3062         u64 num_bytes;
3063         u64 used;
3064         u64 alloc_target;
3065         int retries = 0;
3066
3067         /* get the space info for where the metadata will live */
3068         alloc_target = btrfs_get_alloc_profile(root, 0);
3069         meta_sinfo = __find_space_info(info, alloc_target);
3070
3071         num_bytes = calculate_bytes_needed(root, num_items);
3072 again:
3073         spin_lock(&meta_sinfo->lock);
3074
3075         if (unlikely(!meta_sinfo->bytes_root))
3076                 meta_sinfo->bytes_root = calculate_bytes_needed(root, 6);
3077
3078         if (!retries)
3079                 meta_sinfo->bytes_may_use += num_bytes;
3080
3081         used = meta_sinfo->bytes_used + meta_sinfo->bytes_reserved +
3082                 meta_sinfo->bytes_pinned + meta_sinfo->bytes_readonly +
3083                 meta_sinfo->bytes_super + meta_sinfo->bytes_root +
3084                 meta_sinfo->bytes_may_use + meta_sinfo->bytes_delalloc;
3085
3086         if (used > meta_sinfo->total_bytes) {
3087                 retries++;
3088                 if (retries == 1) {
3089                         if (maybe_allocate_chunk(root, meta_sinfo))
3090                                 goto again;
3091                         retries++;
3092                 } else {
3093                         spin_unlock(&meta_sinfo->lock);
3094                 }
3095
3096                 if (retries == 2) {
3097                         btrfs_start_delalloc_inodes(root);
3098                         btrfs_wait_ordered_extents(root, 0);
3099                         goto again;
3100                 }
3101                 spin_lock(&meta_sinfo->lock);
3102                 meta_sinfo->bytes_may_use -= num_bytes;
3103                 spin_unlock(&meta_sinfo->lock);
3104
3105                 dump_space_info(meta_sinfo, 0, 0);
3106                 return -ENOSPC;
3107         }
3108
3109         check_force_delalloc(meta_sinfo);
3110         spin_unlock(&meta_sinfo->lock);
3111
3112         return 0;
3113 }
3114
3115 /*
3116  * This will check the space that the inode allocates from to make sure we have
3117  * enough space for bytes.
3118  */
3119 int btrfs_check_data_free_space(struct btrfs_root *root, struct inode *inode,
3120                                 u64 bytes)
3121 {
3122         struct btrfs_space_info *data_sinfo;
3123         int ret = 0, committed = 0;
3124
3125         /* make sure bytes are sectorsize aligned */
3126         bytes = (bytes + root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
3127
3128         data_sinfo = BTRFS_I(inode)->space_info;
3129         if (!data_sinfo)
3130                 goto alloc;
3131
3132 again:
3133         /* make sure we have enough space to handle the data first */
3134         spin_lock(&data_sinfo->lock);
3135         if (data_sinfo->total_bytes - data_sinfo->bytes_used -
3136             data_sinfo->bytes_delalloc - data_sinfo->bytes_reserved -
3137             data_sinfo->bytes_pinned - data_sinfo->bytes_readonly -
3138             data_sinfo->bytes_may_use - data_sinfo->bytes_super < bytes) {
3139                 struct btrfs_trans_handle *trans;
3140
3141                 /*
3142                  * if we don't have enough free bytes in this space then we need
3143                  * to alloc a new chunk.
3144                  */
3145                 if (!data_sinfo->full) {
3146                         u64 alloc_target;
3147
3148                         data_sinfo->force_alloc = 1;
3149                         spin_unlock(&data_sinfo->lock);
3150 alloc:
3151                         alloc_target = btrfs_get_alloc_profile(root, 1);
3152                         trans = btrfs_start_transaction(root, 1);
3153                         if (!trans)
3154                                 return -ENOMEM;
3155
3156                         ret = do_chunk_alloc(trans, root->fs_info->extent_root,
3157                                              bytes + 2 * 1024 * 1024,
3158                                              alloc_target, 0);
3159                         btrfs_end_transaction(trans, root);
3160                         if (ret)
3161                                 return ret;
3162
3163                         if (!data_sinfo) {
3164                                 btrfs_set_inode_space_info(root, inode);
3165                                 data_sinfo = BTRFS_I(inode)->space_info;
3166                         }
3167                         goto again;
3168                 }
3169                 spin_unlock(&data_sinfo->lock);
3170
3171                 /* commit the current transaction and try again */
3172                 if (!committed && !root->fs_info->open_ioctl_trans) {
3173                         committed = 1;
3174                         trans = btrfs_join_transaction(root, 1);
3175                         if (!trans)
3176                                 return -ENOMEM;
3177                         ret = btrfs_commit_transaction(trans, root);
3178                         if (ret)
3179                                 return ret;
3180                         goto again;
3181                 }
3182
3183                 printk(KERN_ERR "no space left, need %llu, %llu delalloc bytes"
3184                        ", %llu bytes_used, %llu bytes_reserved, "
3185                        "%llu bytes_pinned, %llu bytes_readonly, %llu may use "
3186                        "%llu total\n", (unsigned long long)bytes,
3187                        (unsigned long long)data_sinfo->bytes_delalloc,
3188                        (unsigned long long)data_sinfo->bytes_used,
3189                        (unsigned long long)data_sinfo->bytes_reserved,
3190                        (unsigned long long)data_sinfo->bytes_pinned,
3191                        (unsigned long long)data_sinfo->bytes_readonly,
3192                        (unsigned long long)data_sinfo->bytes_may_use,
3193                        (unsigned long long)data_sinfo->total_bytes);
3194                 return -ENOSPC;
3195         }
3196         data_sinfo->bytes_may_use += bytes;
3197         BTRFS_I(inode)->reserved_bytes += bytes;
3198         spin_unlock(&data_sinfo->lock);
3199
3200         return 0;
3201 }
3202
3203 /*
3204  * if there was an error for whatever reason after calling
3205  * btrfs_check_data_free_space, call this so we can cleanup the counters.
3206  */
3207 void btrfs_free_reserved_data_space(struct btrfs_root *root,
3208                                     struct inode *inode, u64 bytes)
3209 {
3210         struct btrfs_space_info *data_sinfo;
3211
3212         /* make sure bytes are sectorsize aligned */
3213         bytes = (bytes + root->sectorsize - 1) & ~((u64)root->sectorsize - 1);
3214
3215         data_sinfo = BTRFS_I(inode)->space_info;
3216         spin_lock(&data_sinfo->lock);
3217         data_sinfo->bytes_may_use -= bytes;
3218         BTRFS_I(inode)->reserved_bytes -= bytes;
3219         spin_unlock(&data_sinfo->lock);
3220 }
3221
3222 /* called when we are adding a delalloc extent to the inode's io_tree */
3223 void btrfs_delalloc_reserve_space(struct btrfs_root *root, struct inode *inode,
3224                                   u64 bytes)
3225 {
3226         struct btrfs_space_info *data_sinfo;
3227
3228         /* get the space info for where this inode will be storing its data */
3229         data_sinfo = BTRFS_I(inode)->space_info;
3230
3231         /* make sure we have enough space to handle the data first */
3232         spin_lock(&data_sinfo->lock);
3233         data_sinfo->bytes_delalloc += bytes;
3234
3235         /*
3236          * we are adding a delalloc extent without calling
3237          * btrfs_check_data_free_space first.  This happens on a weird
3238          * writepage condition, but shouldn't hurt our accounting
3239          */
3240         if (unlikely(bytes > BTRFS_I(inode)->reserved_bytes)) {
3241                 data_sinfo->bytes_may_use -= BTRFS_I(inode)->reserved_bytes;
3242                 BTRFS_I(inode)->reserved_bytes = 0;
3243         } else {
3244                 data_sinfo->bytes_may_use -= bytes;
3245                 BTRFS_I(inode)->reserved_bytes -= bytes;
3246         }
3247
3248         spin_unlock(&data_sinfo->lock);
3249 }
3250
3251 /* called when we are clearing an delalloc extent from the inode's io_tree */
3252 void btrfs_delalloc_free_space(struct btrfs_root *root, struct inode *inode,
3253                               u64 bytes)
3254 {
3255         struct btrfs_space_info *info;
3256
3257         info = BTRFS_I(inode)->space_info;
3258
3259         spin_lock(&info->lock);
3260         info->bytes_delalloc -= bytes;
3261         spin_unlock(&info->lock);
3262 }
3263
3264 static void force_metadata_allocation(struct btrfs_fs_info *info)
3265 {
3266         struct list_head *head = &info->space_info;
3267         struct btrfs_space_info *found;
3268
3269         rcu_read_lock();
3270         list_for_each_entry_rcu(found, head, list) {
3271                 if (found->flags & BTRFS_BLOCK_GROUP_METADATA)
3272                         found->force_alloc = 1;
3273         }
3274         rcu_read_unlock();
3275 }
3276
3277 static int do_chunk_alloc(struct btrfs_trans_handle *trans,
3278                           struct btrfs_root *extent_root, u64 alloc_bytes,
3279                           u64 flags, int force)
3280 {
3281         struct btrfs_space_info *space_info;
3282         struct btrfs_fs_info *fs_info = extent_root->fs_info;
3283         u64 thresh;
3284         int ret = 0;
3285
3286         mutex_lock(&fs_info->chunk_mutex);
3287
3288         flags = btrfs_reduce_alloc_profile(extent_root, flags);
3289
3290         space_info = __find_space_info(extent_root->fs_info, flags);
3291         if (!space_info) {
3292                 ret = update_space_info(extent_root->fs_info, flags,
3293                                         0, 0, &space_info);
3294                 BUG_ON(ret);
3295         }
3296         BUG_ON(!space_info);
3297
3298         spin_lock(&space_info->lock);
3299         if (space_info->force_alloc)
3300                 force = 1;
3301         if (space_info->full) {
3302                 spin_unlock(&space_info->lock);
3303                 goto out;
3304         }
3305
3306         thresh = space_info->total_bytes - space_info->bytes_readonly;
3307         thresh = div_factor(thresh, 8);
3308         if (!force &&
3309            (space_info->bytes_used + space_info->bytes_pinned +
3310             space_info->bytes_reserved + alloc_bytes) < thresh) {
3311                 spin_unlock(&space_info->lock);
3312                 goto out;
3313         }
3314         spin_unlock(&space_info->lock);
3315
3316         /*
3317          * if we're doing a data chunk, go ahead and make sure that
3318          * we keep a reasonable number of metadata chunks allocated in the
3319          * FS as well.
3320          */
3321         if (flags & BTRFS_BLOCK_GROUP_DATA && fs_info->metadata_ratio) {
3322                 fs_info->data_chunk_allocations++;
3323                 if (!(fs_info->data_chunk_allocations %
3324                       fs_info->metadata_ratio))
3325                         force_metadata_allocation(fs_info);
3326         }
3327
3328         ret = btrfs_alloc_chunk(trans, extent_root, flags);
3329         spin_lock(&space_info->lock);
3330         if (ret)
3331                 space_info->full = 1;
3332         space_info->force_alloc = 0;
3333         spin_unlock(&space_info->lock);
3334 out:
3335         mutex_unlock(&extent_root->fs_info->chunk_mutex);
3336         return ret;
3337 }
3338
3339 static int update_block_group(struct btrfs_trans_handle *trans,
3340                               struct btrfs_root *root,
3341                               u64 bytenr, u64 num_bytes, int alloc,
3342                               int mark_free)
3343 {
3344         struct btrfs_block_group_cache *cache;
3345         struct btrfs_fs_info *info = root->fs_info;
3346         u64 total = num_bytes;
3347         u64 old_val;
3348         u64 byte_in_group;
3349
3350         /* block accounting for super block */
3351         spin_lock(&info->delalloc_lock);
3352         old_val = btrfs_super_bytes_used(&info->super_copy);
3353         if (alloc)
3354                 old_val += num_bytes;
3355         else
3356                 old_val -= num_bytes;
3357         btrfs_set_super_bytes_used(&info->super_copy, old_val);
3358
3359         /* block accounting for root item */
3360         old_val = btrfs_root_used(&root->root_item);
3361         if (alloc)
3362                 old_val += num_bytes;
3363         else
3364                 old_val -= num_bytes;
3365         btrfs_set_root_used(&root->root_item, old_val);
3366         spin_unlock(&info->delalloc_lock);
3367
3368         while (total) {
3369                 cache = btrfs_lookup_block_group(info, bytenr);
3370                 if (!cache)
3371                         return -1;
3372                 byte_in_group = bytenr - cache->key.objectid;
3373                 WARN_ON(byte_in_group > cache->key.offset);
3374
3375                 spin_lock(&cache->space_info->lock);
3376                 spin_lock(&cache->lock);
3377                 cache->dirty = 1;
3378                 old_val = btrfs_block_group_used(&cache->item);
3379                 num_bytes = min(total, cache->key.offset - byte_in_group);
3380                 if (alloc) {
3381                         old_val += num_bytes;
3382                         btrfs_set_block_group_used(&cache->item, old_val);
3383                         cache->reserved -= num_bytes;
3384                         cache->space_info->bytes_used += num_bytes;
3385                         cache->space_info->bytes_reserved -= num_bytes;
3386                         if (cache->ro)
3387                                 cache->space_info->bytes_readonly -= num_bytes;
3388                         spin_unlock(&cache->lock);
3389                         spin_unlock(&cache->space_info->lock);
3390                 } else {
3391                         old_val -= num_bytes;
3392                         cache->space_info->bytes_used -= num_bytes;
3393                         if (cache->ro)
3394                                 cache->space_info->bytes_readonly += num_bytes;
3395                         btrfs_set_block_group_used(&cache->item, old_val);
3396                         spin_unlock(&cache->lock);
3397                         spin_unlock(&cache->space_info->lock);
3398                         if (mark_free) {
3399                                 int ret;
3400
3401                                 ret = btrfs_discard_extent(root, bytenr,
3402                                                            num_bytes);
3403                                 WARN_ON(ret);
3404
3405                                 ret = btrfs_add_free_space(cache, bytenr,
3406                                                            num_bytes);
3407                                 WARN_ON(ret);
3408                         }
3409                 }
3410                 btrfs_put_block_group(cache);
3411                 total -= num_bytes;
3412                 bytenr += num_bytes;
3413         }
3414         return 0;
3415 }
3416
3417 static u64 first_logical_byte(struct btrfs_root *root, u64 search_start)
3418 {
3419         struct btrfs_block_group_cache *cache;
3420         u64 bytenr;
3421
3422         cache = btrfs_lookup_first_block_group(root->fs_info, search_start);
3423         if (!cache)
3424                 return 0;
3425
3426         bytenr = cache->key.objectid;
3427         btrfs_put_block_group(cache);
3428
3429         return bytenr;
3430 }
3431
3432 /*
3433  * this function must be called within transaction
3434  */
3435 int btrfs_pin_extent(struct btrfs_root *root,
3436                      u64 bytenr, u64 num_bytes, int reserved)
3437 {
3438         struct btrfs_fs_info *fs_info = root->fs_info;
3439         struct btrfs_block_group_cache *cache;
3440
3441         cache = btrfs_lookup_block_group(fs_info, bytenr);
3442         BUG_ON(!cache);
3443
3444         spin_lock(&cache->space_info->lock);
3445         spin_lock(&cache->lock);
3446         cache->pinned += num_bytes;
3447         cache->space_info->bytes_pinned += num_bytes;
3448         if (reserved) {
3449                 cache->reserved -= num_bytes;
3450                 cache->space_info->bytes_reserved -= num_bytes;
3451         }
3452         spin_unlock(&cache->lock);
3453         spin_unlock(&cache->space_info->lock);
3454
3455         btrfs_put_block_group(cache);
3456
3457         set_extent_dirty(fs_info->pinned_extents,
3458                          bytenr, bytenr + num_bytes - 1, GFP_NOFS);
3459         return 0;
3460 }
3461
3462 static int update_reserved_extents(struct btrfs_block_group_cache *cache,
3463                                    u64 num_bytes, int reserve)
3464 {
3465         spin_lock(&cache->space_info->lock);
3466         spin_lock(&cache->lock);
3467         if (reserve) {
3468                 cache->reserved += num_bytes;
3469                 cache->space_info->bytes_reserved += num_bytes;
3470         } else {
3471                 cache->reserved -= num_bytes;
3472                 cache->space_info->bytes_reserved -= num_bytes;
3473         }
3474         spin_unlock(&cache->lock);
3475         spin_unlock(&cache->space_info->lock);
3476         return 0;
3477 }
3478
3479 int btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans,
3480                                 struct btrfs_root *root)
3481 {
3482         struct btrfs_fs_info *fs_info = root->fs_info;
3483         struct btrfs_caching_control *next;
3484         struct btrfs_caching_control *caching_ctl;
3485         struct btrfs_block_group_cache *cache;
3486
3487         down_write(&fs_info->extent_commit_sem);
3488
3489         list_for_each_entry_safe(caching_ctl, next,
3490                                  &fs_info->caching_block_groups, list) {
3491                 cache = caching_ctl->block_group;
3492                 if (block_group_cache_done(cache)) {
3493                         cache->last_byte_to_unpin = (u64)-1;
3494                         list_del_init(&caching_ctl->list);
3495                         put_caching_control(caching_ctl);
3496                 } else {
3497                         cache->last_byte_to_unpin = caching_ctl->progress;
3498                 }
3499         }
3500
3501         if (fs_info->pinned_extents == &fs_info->freed_extents[0])
3502                 fs_info->pinned_extents = &fs_info->freed_extents[1];
3503         else
3504                 fs_info->pinned_extents = &fs_info->freed_extents[0];
3505
3506         up_write(&fs_info->extent_commit_sem);
3507         return 0;
3508 }
3509
3510 static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end)
3511 {
3512         struct btrfs_fs_info *fs_info = root->fs_info;
3513         struct btrfs_block_group_cache *cache = NULL;
3514         u64 len;
3515
3516         while (start <= end) {
3517                 if (!cache ||
3518                     start >= cache->key.objectid + cache->key.offset) {
3519                         if (cache)
3520                                 btrfs_put_block_group(cache);
3521                         cache = btrfs_lookup_block_group(fs_info, start);
3522                         BUG_ON(!cache);
3523                 }
3524
3525                 len = cache->key.objectid + cache->key.offset - start;
3526                 len = min(len, end + 1 - start);
3527
3528                 if (start < cache->last_byte_to_unpin) {
3529                         len = min(len, cache->last_byte_to_unpin - start);
3530                         btrfs_add_free_space(cache, start, len);
3531                 }
3532
3533                 spin_lock(&cache->space_info->lock);
3534                 spin_lock(&cache->lock);
3535                 cache->pinned -= len;
3536                 cache->space_info->bytes_pinned -= len;
3537                 spin_unlock(&cache->lock);
3538                 spin_unlock(&cache->space_info->lock);
3539
3540                 start += len;
3541         }
3542
3543         if (cache)
3544                 btrfs_put_block_group(cache);
3545         return 0;
3546 }
3547
3548 int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
3549                                struct btrfs_root *root)
3550 {
3551         struct btrfs_fs_info *fs_info = root->fs_info;
3552         struct extent_io_tree *unpin;
3553         u64 start;
3554         u64 end;
3555         int ret;
3556
3557         if (fs_info->pinned_extents == &fs_info->freed_extents[0])
3558                 unpin = &fs_info->freed_extents[1];
3559         else
3560                 unpin = &fs_info->freed_extents[0];
3561
3562         while (1) {
3563                 ret = find_first_extent_bit(unpin, 0, &start, &end,
3564                                             EXTENT_DIRTY);
3565                 if (ret)
3566                         break;
3567
3568                 ret = btrfs_discard_extent(root, start, end + 1 - start);
3569
3570                 clear_extent_dirty(unpin, start, end, GFP_NOFS);
3571                 unpin_extent_range(root, start, end);
3572                 cond_resched();
3573         }
3574
3575         return ret;
3576 }
3577
3578 static int pin_down_bytes(struct btrfs_trans_handle *trans,
3579                           struct btrfs_root *root,
3580                           struct btrfs_path *path,
3581                           u64 bytenr, u64 num_bytes,
3582                           int is_data, int reserved,
3583                           struct extent_buffer **must_clean)
3584 {
3585         int err = 0;
3586         struct extent_buffer *buf;
3587
3588         if (is_data)
3589                 goto pinit;
3590
3591         buf = btrfs_find_tree_block(root, bytenr, num_bytes);
3592         if (!buf)
3593                 goto pinit;
3594
3595         /* we can reuse a block if it hasn't been written
3596          * and it is from this transaction.  We can't
3597          * reuse anything from the tree log root because
3598          * it has tiny sub-transactions.
3599          */
3600         if (btrfs_buffer_uptodate(buf, 0) &&
3601             btrfs_try_tree_lock(buf)) {
3602                 u64 header_owner = btrfs_header_owner(buf);
3603                 u64 header_transid = btrfs_header_generation(buf);
3604                 if (header_owner != BTRFS_TREE_LOG_OBJECTID &&
3605                     header_transid == trans->transid &&
3606                     !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) {
3607                         *must_clean = buf;
3608                         return 1;
3609                 }
3610                 btrfs_tree_unlock(buf);
3611         }
3612         free_extent_buffer(buf);
3613 pinit:
3614         if (path)
3615                 btrfs_set_path_blocking(path);
3616         /* unlocks the pinned mutex */
3617         btrfs_pin_extent(root, bytenr, num_bytes, reserved);
3618
3619         BUG_ON(err < 0);
3620         return 0;
3621 }
3622
3623 static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
3624                                 struct btrfs_root *root,
3625                                 u64 bytenr, u64 num_bytes, u64 parent,
3626                                 u64 root_objectid, u64 owner_objectid,
3627                                 u64 owner_offset, int refs_to_drop,
3628                                 struct btrfs_delayed_extent_op *extent_op)
3629 {
3630         struct btrfs_key key;
3631         struct btrfs_path *path;
3632         struct btrfs_fs_info *info = root->fs_info;
3633         struct btrfs_root *extent_root = info->extent_root;
3634         struct extent_buffer *leaf;
3635         struct btrfs_extent_item *ei;
3636         struct btrfs_extent_inline_ref *iref;
3637         int ret;
3638         int is_data;
3639         int extent_slot = 0;
3640         int found_extent = 0;
3641         int num_to_del = 1;
3642         u32 item_size;
3643         u64 refs;
3644
3645         path = btrfs_alloc_path();
3646         if (!path)
3647                 return -ENOMEM;
3648
3649         path->reada = 1;
3650         path->leave_spinning = 1;
3651
3652         is_data = owner_objectid >= BTRFS_FIRST_FREE_OBJECTID;
3653         BUG_ON(!is_data && refs_to_drop != 1);
3654
3655         ret = lookup_extent_backref(trans, extent_root, path, &iref,
3656                                     bytenr, num_bytes, parent,
3657                                     root_objectid, owner_objectid,
3658                                     owner_offset);
3659         if (ret == 0) {
3660                 extent_slot = path->slots[0];
3661                 while (extent_slot >= 0) {
3662                         btrfs_item_key_to_cpu(path->nodes[0], &key,
3663                                               extent_slot);
3664                         if (key.objectid != bytenr)
3665                                 break;
3666                         if (key.type == BTRFS_EXTENT_ITEM_KEY &&
3667                             key.offset == num_bytes) {
3668                                 found_extent = 1;
3669                                 break;
3670                         }
3671                         if (path->slots[0] - extent_slot > 5)
3672                                 break;
3673                         extent_slot--;
3674                 }
3675 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
3676                 item_size = btrfs_item_size_nr(path->nodes[0], extent_slot);
3677                 if (found_extent && item_size < sizeof(*ei))
3678                         found_extent = 0;
3679 #endif
3680                 if (!found_extent) {
3681                         BUG_ON(iref);
3682                         ret = remove_extent_backref(trans, extent_root, path,
3683                                                     NULL, refs_to_drop,
3684                                                     is_data);
3685                         BUG_ON(ret);
3686                         btrfs_release_path(extent_root, path);
3687                         path->leave_spinning = 1;
3688
3689                         key.objectid = bytenr;
3690                         key.type = BTRFS_EXTENT_ITEM_KEY;
3691                         key.offset = num_bytes;
3692
3693                         ret = btrfs_search_slot(trans, extent_root,
3694                                                 &key, path, -1, 1);
3695                         if (ret) {
3696                                 printk(KERN_ERR "umm, got %d back from search"
3697                                        ", was looking for %llu\n", ret,
3698                                        (unsigned long long)bytenr);
3699                                 btrfs_print_leaf(extent_root, path->nodes[0]);
3700                         }
3701                         BUG_ON(ret);
3702                         extent_slot = path->slots[0];
3703                 }
3704         } else {
3705                 btrfs_print_leaf(extent_root, path->nodes[0]);
3706                 WARN_ON(1);
3707                 printk(KERN_ERR "btrfs unable to find ref byte nr %llu "
3708                        "parent %llu root %llu  owner %llu offset %llu\n",
3709                        (unsigned long long)bytenr,
3710                        (unsigned long long)parent,
3711                        (unsigned long long)root_objectid,
3712                        (unsigned long long)owner_objectid,
3713                        (unsigned long long)owner_offset);
3714         }
3715
3716         leaf = path->nodes[0];
3717         item_size = btrfs_item_size_nr(leaf, extent_slot);
3718 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
3719         if (item_size < sizeof(*ei)) {
3720                 BUG_ON(found_extent || extent_slot != path->slots[0]);
3721                 ret = convert_extent_item_v0(trans, extent_root, path,
3722                                              owner_objectid, 0);
3723                 BUG_ON(ret < 0);
3724
3725                 btrfs_release_path(extent_root, path);
3726                 path->leave_spinning = 1;
3727
3728                 key.objectid = bytenr;
3729                 key.type = BTRFS_EXTENT_ITEM_KEY;
3730                 key.offset = num_bytes;
3731
3732                 ret = btrfs_search_slot(trans, extent_root, &key, path,
3733                                         -1, 1);
3734                 if (ret) {
3735                         printk(KERN_ERR "umm, got %d back from search"
3736                                ", was looking for %llu\n", ret,
3737                                (unsigned long long)bytenr);
3738                         btrfs_print_leaf(extent_root, path->nodes[0]);
3739                 }
3740                 BUG_ON(ret);
3741                 extent_slot = path->slots[0];
3742                 leaf = path->nodes[0];
3743                 item_size = btrfs_item_size_nr(leaf, extent_slot);
3744         }
3745 #endif
3746         BUG_ON(item_size < sizeof(*ei));
3747         ei = btrfs_item_ptr(leaf, extent_slot,
3748                             struct btrfs_extent_item);
3749         if (owner_objectid < BTRFS_FIRST_FREE_OBJECTID) {
3750                 struct btrfs_tree_block_info *bi;
3751                 BUG_ON(item_size < sizeof(*ei) + sizeof(*bi));
3752                 bi = (struct btrfs_tree_block_info *)(ei + 1);
3753                 WARN_ON(owner_objectid != btrfs_tree_block_level(leaf, bi));
3754         }
3755
3756         refs = btrfs_extent_refs(leaf, ei);
3757         BUG_ON(refs < refs_to_drop);
3758         refs -= refs_to_drop;
3759
3760         if (refs > 0) {
3761                 if (extent_op)
3762                         __run_delayed_extent_op(extent_op, leaf, ei);
3763                 /*
3764                  * In the case of inline back ref, reference count will
3765                  * be updated by remove_extent_backref
3766                  */
3767                 if (iref) {
3768                         BUG_ON(!found_extent);
3769                 } else {
3770                         btrfs_set_extent_refs(leaf, ei, refs);
3771                         btrfs_mark_buffer_dirty(leaf);
3772                 }
3773                 if (found_extent) {
3774                         ret = remove_extent_backref(trans, extent_root, path,
3775                                                     iref, refs_to_drop,
3776                                                     is_data);
3777                         BUG_ON(ret);
3778                 }
3779         } else {
3780                 int mark_free = 0;
3781                 struct extent_buffer *must_clean = NULL;
3782
3783                 if (found_extent) {
3784                         BUG_ON(is_data && refs_to_drop !=
3785                                extent_data_ref_count(root, path, iref));
3786                         if (iref) {
3787                                 BUG_ON(path->slots[0] != extent_slot);
3788                         } else {
3789                                 BUG_ON(path->slots[0] != extent_slot + 1);
3790                                 path->slots[0] = extent_slot;
3791                                 num_to_del = 2;
3792                         }
3793                 }
3794
3795                 ret = pin_down_bytes(trans, root, path, bytenr,
3796                                      num_bytes, is_data, 0, &must_clean);
3797                 if (ret > 0)
3798                         mark_free = 1;
3799                 BUG_ON(ret < 0);
3800                 /*
3801                  * it is going to be very rare for someone to be waiting
3802                  * on the block we're freeing.  del_items might need to
3803                  * schedule, so rather than get fancy, just force it
3804                  * to blocking here
3805                  */
3806                 if (must_clean)
3807                         btrfs_set_lock_blocking(must_clean);
3808
3809                 ret = btrfs_del_items(trans, extent_root, path, path->slots[0],
3810                                       num_to_del);
3811                 BUG_ON(ret);
3812                 btrfs_release_path(extent_root, path);
3813
3814                 if (must_clean) {
3815                         clean_tree_block(NULL, root, must_clean);
3816                         btrfs_tree_unlock(must_clean);
3817                         free_extent_buffer(must_clean);
3818                 }
3819
3820                 if (is_data) {
3821                         ret = btrfs_del_csums(trans, root, bytenr, num_bytes);
3822                         BUG_ON(ret);
3823                 } else {
3824                         invalidate_mapping_pages(info->btree_inode->i_mapping,
3825                              bytenr >> PAGE_CACHE_SHIFT,
3826                              (bytenr + num_bytes - 1) >> PAGE_CACHE_SHIFT);
3827                 }
3828
3829                 ret = update_block_group(trans, root, bytenr, num_bytes, 0,
3830                                          mark_free);
3831                 BUG_ON(ret);
3832         }
3833         btrfs_free_path(path);
3834         return ret;
3835 }
3836
3837 /*
3838  * when we free an extent, it is possible (and likely) that we free the last
3839  * delayed ref for that extent as well.  This searches the delayed ref tree for
3840  * a given extent, and if there are no other delayed refs to be processed, it
3841  * removes it from the tree.
3842  */
3843 static noinline int check_ref_cleanup(struct btrfs_trans_handle *trans,
3844                                       struct btrfs_root *root, u64 bytenr)
3845 {
3846         struct btrfs_delayed_ref_head *head;
3847         struct btrfs_delayed_ref_root *delayed_refs;
3848         struct btrfs_delayed_ref_node *ref;
3849         struct rb_node *node;
3850         int ret;
3851
3852         delayed_refs = &trans->transaction->delayed_refs;
3853         spin_lock(&delayed_refs->lock);
3854         head = btrfs_find_delayed_ref_head(trans, bytenr);
3855         if (!head)
3856                 goto out;
3857
3858         node = rb_prev(&head->node.rb_node);
3859         if (!node)
3860                 goto out;
3861
3862         ref = rb_entry(node, struct btrfs_delayed_ref_node, rb_node);
3863
3864         /* there are still entries for this ref, we can't drop it */
3865         if (ref->bytenr == bytenr)
3866                 goto out;
3867
3868         if (head->extent_op) {
3869                 if (!head->must_insert_reserved)
3870                         goto out;
3871                 kfree(head->extent_op);
3872                 head->extent_op = NULL;
3873         }
3874
3875         /*
3876          * waiting for the lock here would deadlock.  If someone else has it
3877          * locked they are already in the process of dropping it anyway
3878          */
3879         if (!mutex_trylock(&head->mutex))
3880                 goto out;
3881
3882         /*
3883          * at this point we have a head with no other entries.  Go
3884          * ahead and process it.
3885          */
3886         head->node.in_tree = 0;
3887         rb_erase(&head->node.rb_node, &delayed_refs->root);
3888
3889         delayed_refs->num_entries--;
3890
3891         /*
3892          * we don't take a ref on the node because we're removing it from the
3893          * tree, so we just steal the ref the tree was holding.
3894          */
3895         delayed_refs->num_heads--;
3896         if (list_empty(&head->cluster))
3897                 delayed_refs->num_heads_ready--;
3898
3899         list_del_init(&head->cluster);
3900         spin_unlock(&delayed_refs->lock);
3901
3902         ret = run_one_delayed_ref(trans, root->fs_info->tree_root,
3903                                   &head->node, head->extent_op,
3904                                   head->must_insert_reserved);
3905         BUG_ON(ret);
3906         btrfs_put_delayed_ref(&head->node);
3907         return 0;
3908 out:
3909         spin_unlock(&delayed_refs->lock);
3910         return 0;
3911 }
3912
3913 int btrfs_free_extent(struct btrfs_trans_handle *trans,
3914                       struct btrfs_root *root,
3915                       u64 bytenr, u64 num_bytes, u64 parent,
3916                       u64 root_objectid, u64 owner, u64 offset)
3917 {
3918         int ret;
3919
3920         /*
3921          * tree log blocks never actually go into the extent allocation
3922          * tree, just update pinning info and exit early.
3923          */
3924         if (root_objectid == BTRFS_TREE_LOG_OBJECTID) {
3925                 WARN_ON(owner >= BTRFS_FIRST_FREE_OBJECTID);
3926                 /* unlocks the pinned mutex */
3927                 btrfs_pin_extent(root, bytenr, num_bytes, 1);
3928                 ret = 0;
3929         } else if (owner < BTRFS_FIRST_FREE_OBJECTID) {
3930                 ret = btrfs_add_delayed_tree_ref(trans, bytenr, num_bytes,
3931                                         parent, root_objectid, (int)owner,
3932                                         BTRFS_DROP_DELAYED_REF, NULL);
3933                 BUG_ON(ret);
3934                 ret = check_ref_cleanup(trans, root, bytenr);
3935                 BUG_ON(ret);
3936         } else {
3937                 ret = btrfs_add_delayed_data_ref(trans, bytenr, num_bytes,
3938                                         parent, root_objectid, owner,
3939                                         offset, BTRFS_DROP_DELAYED_REF, NULL);
3940                 BUG_ON(ret);
3941         }
3942         return ret;
3943 }
3944
3945 static u64 stripe_align(struct btrfs_root *root, u64 val)
3946 {
3947         u64 mask = ((u64)root->stripesize - 1);
3948         u64 ret = (val + mask) & ~mask;
3949         return ret;
3950 }
3951
3952 /*
3953  * when we wait for progress in the block group caching, its because
3954  * our allocation attempt failed at least once.  So, we must sleep
3955  * and let some progress happen before we try again.
3956  *
3957  * This function will sleep at least once waiting for new free space to
3958  * show up, and then it will check the block group free space numbers
3959  * for our min num_bytes.  Another option is to have it go ahead
3960  * and look in the rbtree for a free extent of a given size, but this
3961  * is a good start.
3962  */
3963 static noinline int
3964 wait_block_group_cache_progress(struct btrfs_block_group_cache *cache,
3965                                 u64 num_bytes)
3966 {
3967         struct btrfs_caching_control *caching_ctl;
3968         DEFINE_WAIT(wait);
3969
3970         caching_ctl = get_caching_control(cache);
3971         if (!caching_ctl)
3972                 return 0;
3973
3974         wait_event(caching_ctl->wait, block_group_cache_done(cache) ||
3975                    (cache->free_space >= num_bytes));
3976
3977         put_caching_control(caching_ctl);
3978         return 0;
3979 }
3980
3981 static noinline int
3982 wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
3983 {
3984         struct btrfs_caching_control *caching_ctl;
3985         DEFINE_WAIT(wait);
3986
3987         caching_ctl = get_caching_control(cache);
3988         if (!caching_ctl)
3989                 return 0;
3990
3991         wait_event(caching_ctl->wait, block_group_cache_done(cache));
3992
3993         put_caching_control(caching_ctl);
3994         return 0;
3995 }
3996
3997 enum btrfs_loop_type {
3998         LOOP_CACHED_ONLY = 0,
3999         LOOP_CACHING_NOWAIT = 1,
4000         LOOP_CACHING_WAIT = 2,
4001         LOOP_ALLOC_CHUNK = 3,
4002         LOOP_NO_EMPTY_SIZE = 4,
4003 };
4004
4005 /*
4006  * walks the btree of allocated extents and find a hole of a given size.
4007  * The key ins is changed to record the hole:
4008  * ins->objectid == block start
4009  * ins->flags = BTRFS_EXTENT_ITEM_KEY
4010  * ins->offset == number of blocks
4011  * Any available blocks before search_start are skipped.
4012  */
4013 static noinline int find_free_extent(struct btrfs_trans_handle *trans,
4014                                      struct btrfs_root *orig_root,
4015                                      u64 num_bytes, u64 empty_size,
4016                                      u64 search_start, u64 search_end,
4017                                      u64 hint_byte, struct btrfs_key *ins,
4018                                      u64 exclude_start, u64 exclude_nr,
4019                                      int data)
4020 {
4021         int ret = 0;
4022         struct btrfs_root *root = orig_root->fs_info->extent_root;
4023         struct btrfs_free_cluster *last_ptr = NULL;
4024         struct btrfs_block_group_cache *block_group = NULL;
4025         int empty_cluster = 2 * 1024 * 1024;
4026         int allowed_chunk_alloc = 0;
4027         struct btrfs_space_info *space_info;
4028         int last_ptr_loop = 0;
4029         int loop = 0;
4030         bool found_uncached_bg = false;
4031         bool failed_cluster_refill = false;
4032
4033         WARN_ON(num_bytes < root->sectorsize);
4034         btrfs_set_key_type(ins, BTRFS_EXTENT_ITEM_KEY);
4035         ins->objectid = 0;
4036         ins->offset = 0;
4037
4038         space_info = __find_space_info(root->fs_info, data);
4039
4040         if (orig_root->ref_cows || empty_size)
4041                 allowed_chunk_alloc = 1;
4042
4043         if (data & BTRFS_BLOCK_GROUP_METADATA) {
4044                 last_ptr = &root->fs_info->meta_alloc_cluster;
4045                 if (!btrfs_test_opt(root, SSD))
4046                         empty_cluster = 64 * 1024;
4047         }
4048
4049         if ((data & BTRFS_BLOCK_GROUP_DATA) && btrfs_test_opt(root, SSD)) {
4050                 last_ptr = &root->fs_info->data_alloc_cluster;
4051         }
4052
4053         if (last_ptr) {
4054                 spin_lock(&last_ptr->lock);
4055                 if (last_ptr->block_group)
4056                         hint_byte = last_ptr->window_start;
4057                 spin_unlock(&last_ptr->lock);
4058         }
4059
4060         search_start = max(search_start, first_logical_byte(root, 0));
4061         search_start = max(search_start, hint_byte);
4062
4063         if (!last_ptr)
4064                 empty_cluster = 0;
4065
4066         if (search_start == hint_byte) {
4067                 block_group = btrfs_lookup_block_group(root->fs_info,
4068                                                        search_start);
4069                 /*
4070                  * we don't want to use the block group if it doesn't match our
4071                  * allocation bits, or if its not cached.
4072                  */
4073                 if (block_group && block_group_bits(block_group, data) &&
4074                     block_group_cache_done(block_group)) {
4075                         down_read(&space_info->groups_sem);
4076                         if (list_empty(&block_group->list) ||
4077                             block_group->ro) {
4078                                 /*
4079                                  * someone is removing this block group,
4080                                  * we can't jump into the have_block_group
4081                                  * target because our list pointers are not
4082                                  * valid
4083                                  */
4084                                 btrfs_put_block_group(block_group);
4085                                 up_read(&space_info->groups_sem);
4086                         } else
4087                                 goto have_block_group;
4088                 } else if (block_group) {
4089                         btrfs_put_block_group(block_group);
4090                 }
4091         }
4092
4093 search:
4094         down_read(&space_info->groups_sem);
4095         list_for_each_entry(block_group, &space_info->block_groups, list) {
4096                 u64 offset;
4097                 int cached;
4098
4099                 atomic_inc(&block_group->count);
4100                 search_start = block_group->key.objectid;
4101
4102 have_block_group:
4103                 if (unlikely(block_group->cached == BTRFS_CACHE_NO)) {
4104                         /*
4105                          * we want to start caching kthreads, but not too many
4106                          * right off the bat so we don't overwhelm the system,
4107                          * so only start them if there are less than 2 and we're
4108                          * in the initial allocation phase.
4109                          */
4110                         if (loop > LOOP_CACHING_NOWAIT ||
4111                             atomic_read(&space_info->caching_threads) < 2) {
4112                                 ret = cache_block_group(block_group);
4113                                 BUG_ON(ret);
4114                         }
4115                 }
4116
4117                 cached = block_group_cache_done(block_group);
4118                 if (unlikely(!cached)) {
4119                         found_uncached_bg = true;
4120
4121                         /* if we only want cached bgs, loop */
4122                         if (loop == LOOP_CACHED_ONLY)
4123                                 goto loop;
4124                 }
4125
4126                 if (unlikely(block_group->ro))
4127                         goto loop;
4128
4129                 /*
4130                  * Ok we want to try and use the cluster allocator, so lets look
4131                  * there, unless we are on LOOP_NO_EMPTY_SIZE, since we will
4132                  * have tried the cluster allocator plenty of times at this
4133                  * point and not have found anything, so we are likely way too
4134                  * fragmented for the clustering stuff to find anything, so lets
4135                  * just skip it and let the allocator find whatever block it can
4136                  * find
4137                  */
4138                 if (last_ptr && loop < LOOP_NO_EMPTY_SIZE) {
4139                         /*
4140                          * the refill lock keeps out other
4141                          * people trying to start a new cluster
4142                          */
4143                         spin_lock(&last_ptr->refill_lock);
4144                         if (last_ptr->block_group &&
4145                             (last_ptr->block_group->ro ||
4146                             !block_group_bits(last_ptr->block_group, data))) {
4147                                 offset = 0;
4148                                 goto refill_cluster;
4149                         }
4150
4151                         offset = btrfs_alloc_from_cluster(block_group, last_ptr,
4152                                                  num_bytes, search_start);
4153                         if (offset) {
4154                                 /* we have a block, we're done */
4155                                 spin_unlock(&last_ptr->refill_lock);
4156                                 goto checks;
4157                         }
4158
4159                         spin_lock(&last_ptr->lock);
4160                         /*
4161                          * whoops, this cluster doesn't actually point to
4162                          * this block group.  Get a ref on the block
4163                          * group is does point to and try again
4164                          */
4165                         if (!last_ptr_loop && last_ptr->block_group &&
4166                             last_ptr->block_group != block_group) {
4167
4168                                 btrfs_put_block_group(block_group);
4169                                 block_group = last_ptr->block_group;
4170                                 atomic_inc(&block_group->count);
4171                                 spin_unlock(&last_ptr->lock);
4172                                 spin_unlock(&last_ptr->refill_lock);
4173
4174                                 last_ptr_loop = 1;
4175                                 search_start = block_group->key.objectid;
4176                                 /*
4177                                  * we know this block group is properly
4178                                  * in the list because
4179                                  * btrfs_remove_block_group, drops the
4180                                  * cluster before it removes the block
4181                                  * group from the list
4182                                  */
4183                                 goto have_block_group;
4184                         }
4185                         spin_unlock(&last_ptr->lock);
4186 refill_cluster:
4187                         /*
4188                          * this cluster didn't work out, free it and
4189                          * start over
4190                          */
4191                         btrfs_return_cluster_to_free_space(NULL, last_ptr);
4192
4193                         last_ptr_loop = 0;
4194
4195                         /* allocate a cluster in this block group */
4196                         ret = btrfs_find_space_cluster(trans, root,
4197                                                block_group, last_ptr,
4198                                                offset, num_bytes,
4199                                                empty_cluster + empty_size);
4200                         if (ret == 0) {
4201                                 /*
4202                                  * now pull our allocation out of this
4203                                  * cluster
4204                                  */
4205                                 offset = btrfs_alloc_from_cluster(block_group,
4206                                                   last_ptr, num_bytes,
4207                                                   search_start);
4208                                 if (offset) {
4209                                         /* we found one, proceed */
4210                                         spin_unlock(&last_ptr->refill_lock);
4211                                         goto checks;
4212                                 }
4213                         } else if (!cached && loop > LOOP_CACHING_NOWAIT
4214                                    && !failed_cluster_refill) {
4215                                 spin_unlock(&last_ptr->refill_lock);
4216
4217                                 failed_cluster_refill = true;
4218                                 wait_block_group_cache_progress(block_group,
4219                                        num_bytes + empty_cluster + empty_size);
4220                                 goto have_block_group;
4221                         }
4222
4223                         /*
4224                          * at this point we either didn't find a cluster
4225                          * or we weren't able to allocate a block from our
4226                          * cluster.  Free the cluster we've been trying
4227                          * to use, and go to the next block group
4228                          */
4229                         btrfs_return_cluster_to_free_space(NULL, last_ptr);
4230                         spin_unlock(&last_ptr->refill_lock);
4231                         goto loop;
4232                 }
4233
4234                 offset = btrfs_find_space_for_alloc(block_group, search_start,
4235                                                     num_bytes, empty_size);
4236                 if (!offset && (cached || (!cached &&
4237                                            loop == LOOP_CACHING_NOWAIT))) {
4238                         goto loop;
4239                 } else if (!offset && (!cached &&
4240                                        loop > LOOP_CACHING_NOWAIT)) {
4241                         wait_block_group_cache_progress(block_group,
4242                                         num_bytes + empty_size);
4243                         goto have_block_group;
4244                 }
4245 checks:
4246                 search_start = stripe_align(root, offset);
4247                 /* move on to the next group */
4248                 if (search_start + num_bytes >= search_end) {
4249                         btrfs_add_free_space(block_group, offset, num_bytes);
4250                         goto loop;
4251                 }
4252
4253                 /* move on to the next group */
4254                 if (search_start + num_bytes >
4255                     block_group->key.objectid + block_group->key.offset) {
4256                         btrfs_add_free_space(block_group, offset, num_bytes);
4257                         goto loop;
4258                 }
4259
4260                 if (exclude_nr > 0 &&
4261                     (search_start + num_bytes > exclude_start &&
4262                      search_start < exclude_start + exclude_nr)) {
4263                         search_start = exclude_start + exclude_nr;
4264
4265                         btrfs_add_free_space(block_group, offset, num_bytes);
4266                         /*
4267                          * if search_start is still in this block group
4268                          * then we just re-search this block group
4269                          */
4270                         if (search_start >= block_group->key.objectid &&
4271                             search_start < (block_group->key.objectid +
4272                                             block_group->key.offset))
4273                                 goto have_block_group;
4274                         goto loop;
4275                 }
4276
4277                 ins->objectid = search_start;
4278                 ins->offset = num_bytes;
4279
4280                 if (offset < search_start)
4281                         btrfs_add_free_space(block_group, offset,
4282                                              search_start - offset);
4283                 BUG_ON(offset > search_start);
4284
4285                 update_reserved_extents(block_group, num_bytes, 1);
4286
4287                 /* we are all good, lets return */
4288                 break;
4289 loop:
4290                 failed_cluster_refill = false;
4291                 btrfs_put_block_group(block_group);
4292         }
4293         up_read(&space_info->groups_sem);
4294
4295         /* LOOP_CACHED_ONLY, only search fully cached block groups
4296          * LOOP_CACHING_NOWAIT, search partially cached block groups, but
4297          *                      dont wait foR them to finish caching
4298          * LOOP_CACHING_WAIT, search everything, and wait if our bg is caching
4299          * LOOP_ALLOC_CHUNK, force a chunk allocation and try again
4300          * LOOP_NO_EMPTY_SIZE, set empty_size and empty_cluster to 0 and try
4301          *                      again
4302          */
4303         if (!ins->objectid && loop < LOOP_NO_EMPTY_SIZE &&
4304             (found_uncached_bg || empty_size || empty_cluster ||
4305              allowed_chunk_alloc)) {
4306                 if (found_uncached_bg) {
4307                         found_uncached_bg = false;
4308                         if (loop < LOOP_CACHING_WAIT) {
4309                                 loop++;
4310                                 goto search;
4311                         }
4312                 }
4313
4314                 if (loop == LOOP_ALLOC_CHUNK) {
4315                         empty_size = 0;
4316                         empty_cluster = 0;
4317                 }
4318
4319                 if (allowed_chunk_alloc) {
4320                         ret = do_chunk_alloc(trans, root, num_bytes +
4321                                              2 * 1024 * 1024, data, 1);
4322                         allowed_chunk_alloc = 0;
4323                 } else {
4324                         space_info->force_alloc = 1;
4325                 }
4326
4327                 if (loop < LOOP_NO_EMPTY_SIZE) {
4328                         loop++;
4329                         goto search;
4330                 }
4331                 ret = -ENOSPC;
4332         } else if (!ins->objectid) {
4333                 ret = -ENOSPC;
4334         }
4335
4336         /* we found what we needed */
4337         if (ins->objectid) {
4338                 if (!(data & BTRFS_BLOCK_GROUP_DATA))
4339                         trans->block_group = block_group->key.objectid;
4340
4341                 btrfs_put_block_group(block_group);
4342                 ret = 0;
4343         }
4344
4345         return ret;
4346 }
4347
4348 static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
4349                             int dump_block_groups)
4350 {
4351         struct btrfs_block_group_cache *cache;
4352
4353         spin_lock(&info->lock);
4354         printk(KERN_INFO "space_info has %llu free, is %sfull\n",
4355                (unsigned long long)(info->total_bytes - info->bytes_used -
4356                                     info->bytes_pinned - info->bytes_reserved -
4357                                     info->bytes_super),
4358                (info->full) ? "" : "not ");
4359         printk(KERN_INFO "space_info total=%llu, pinned=%llu, delalloc=%llu,"
4360                " may_use=%llu, used=%llu, root=%llu, super=%llu, reserved=%llu"
4361                "\n",
4362                (unsigned long long)info->total_bytes,
4363                (unsigned long long)info->bytes_pinned,
4364                (unsigned long long)info->bytes_delalloc,
4365                (unsigned long long)info->bytes_may_use,
4366                (unsigned long long)info->bytes_used,
4367                (unsigned long long)info->bytes_root,
4368                (unsigned long long)info->bytes_super,
4369                (unsigned long long)info->bytes_reserved);
4370         spin_unlock(&info->lock);
4371
4372         if (!dump_block_groups)
4373                 return;
4374
4375         down_read(&info->groups_sem);
4376         list_for_each_entry(cache, &info->block_groups, list) {
4377                 spin_lock(&cache->lock);
4378                 printk(KERN_INFO "block group %llu has %llu bytes, %llu used "
4379                        "%llu pinned %llu reserved\n",
4380                        (unsigned long long)cache->key.objectid,
4381                        (unsigned long long)cache->key.offset,
4382                        (unsigned long long)btrfs_block_group_used(&cache->item),
4383                        (unsigned long long)cache->pinned,
4384                        (unsigned long long)cache->reserved);
4385                 btrfs_dump_free_space(cache, bytes);
4386                 spin_unlock(&cache->lock);
4387         }
4388         up_read(&info->groups_sem);
4389 }
4390
4391 int btrfs_reserve_extent(struct btrfs_trans_handle *trans,
4392                          struct btrfs_root *root,
4393                          u64 num_bytes, u64 min_alloc_size,
4394                          u64 empty_size, u64 hint_byte,
4395                          u64 search_end, struct btrfs_key *ins,
4396                          u64 data)
4397 {
4398         int ret;
4399         u64 search_start = 0;
4400         struct btrfs_fs_info *info = root->fs_info;
4401
4402         data = btrfs_get_alloc_profile(root, data);
4403 again:
4404         /*
4405          * the only place that sets empty_size is btrfs_realloc_node, which
4406          * is not called recursively on allocations
4407          */
4408         if (empty_size || root->ref_cows) {
4409                 if (!(data & BTRFS_BLOCK_GROUP_METADATA)) {
4410                         ret = do_chunk_alloc(trans, root->fs_info->extent_root,
4411                                      2 * 1024 * 1024,
4412                                      BTRFS_BLOCK_GROUP_METADATA |
4413                                      (info->metadata_alloc_profile &
4414                                       info->avail_metadata_alloc_bits), 0);
4415                 }
4416                 ret = do_chunk_alloc(trans, root->fs_info->extent_root,
4417                                      num_bytes + 2 * 1024 * 1024, data, 0);
4418         }
4419
4420         WARN_ON(num_bytes < root->sectorsize);
4421         ret = find_free_extent(trans, root, num_bytes, empty_size,
4422                                search_start, search_end, hint_byte, ins,
4423                                trans->alloc_exclude_start,
4424                                trans->alloc_exclude_nr, data);
4425
4426         if (ret == -ENOSPC && num_bytes > min_alloc_size) {
4427                 num_bytes = num_bytes >> 1;
4428                 num_bytes = num_bytes & ~(root->sectorsize - 1);
4429                 num_bytes = max(num_bytes, min_alloc_size);
4430                 do_chunk_alloc(trans, root->fs_info->extent_root,
4431                                num_bytes, data, 1);
4432                 goto again;
4433         }
4434         if (ret == -ENOSPC) {
4435                 struct btrfs_space_info *sinfo;
4436
4437                 sinfo = __find_space_info(root->fs_info, data);
4438                 printk(KERN_ERR "btrfs allocation failed flags %llu, "
4439                        "wanted %llu\n", (unsigned long long)data,
4440                        (unsigned long long)num_bytes);
4441                 dump_space_info(sinfo, num_bytes, 1);
4442         }
4443
4444         return ret;
4445 }
4446
4447 int btrfs_free_reserved_extent(struct btrfs_root *root, u64 start, u64 len)
4448 {
4449         struct btrfs_block_group_cache *cache;
4450         int ret = 0;
4451
4452         cache = btrfs_lookup_block_group(root->fs_info, start);
4453         if (!cache) {
4454                 printk(KERN_ERR "Unable to find block group for %llu\n",
4455                        (unsigned long long)start);
4456                 return -ENOSPC;
4457         }
4458
4459         ret = btrfs_discard_extent(root, start, len);
4460
4461         btrfs_add_free_space(cache, start, len);
4462         update_reserved_extents(cache, len, 0);
4463         btrfs_put_block_group(cache);
4464
4465         return ret;
4466 }
4467
4468 static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
4469                                       struct btrfs_root *root,
4470                                       u64 parent, u64 root_objectid,
4471                                       u64 flags, u64 owner, u64 offset,
4472                                       struct btrfs_key *ins, int ref_mod)
4473 {
4474         int ret;
4475         struct btrfs_fs_info *fs_info = root->fs_info;
4476         struct btrfs_extent_item *extent_item;
4477         struct btrfs_extent_inline_ref *iref;
4478         struct btrfs_path *path;
4479         struct extent_buffer *leaf;
4480         int type;
4481         u32 size;
4482
4483         if (parent > 0)
4484                 type = BTRFS_SHARED_DATA_REF_KEY;
4485         else
4486                 type = BTRFS_EXTENT_DATA_REF_KEY;
4487
4488         size = sizeof(*extent_item) + btrfs_extent_inline_ref_size(type);
4489
4490         path = btrfs_alloc_path();
4491         BUG_ON(!path);
4492
4493         path->leave_spinning = 1;
4494         ret = btrfs_insert_empty_item(trans, fs_info->extent_root, path,
4495                                       ins, size);
4496         BUG_ON(ret);
4497
4498         leaf = path->nodes[0];
4499         extent_item = btrfs_item_ptr(leaf, path->slots[0],
4500                                      struct btrfs_extent_item);
4501         btrfs_set_extent_refs(leaf, extent_item, ref_mod);
4502         btrfs_set_extent_generation(leaf, extent_item, trans->transid);
4503         btrfs_set_extent_flags(leaf, extent_item,
4504                                flags | BTRFS_EXTENT_FLAG_DATA);
4505
4506         iref = (struct btrfs_extent_inline_ref *)(extent_item + 1);
4507         btrfs_set_extent_inline_ref_type(leaf, iref, type);
4508         if (parent > 0) {
4509                 struct btrfs_shared_data_ref *ref;
4510                 ref = (struct btrfs_shared_data_ref *)(iref + 1);
4511                 btrfs_set_extent_inline_ref_offset(leaf, iref, parent);
4512                 btrfs_set_shared_data_ref_count(leaf, ref, ref_mod);
4513         } else {
4514                 struct btrfs_extent_data_ref *ref;
4515                 ref = (struct btrfs_extent_data_ref *)(&iref->offset);
4516                 btrfs_set_extent_data_ref_root(leaf, ref, root_objectid);
4517                 btrfs_set_extent_data_ref_objectid(leaf, ref, owner);
4518                 btrfs_set_extent_data_ref_offset(leaf, ref, offset);
4519                 btrfs_set_extent_data_ref_count(leaf, ref, ref_mod);
4520         }
4521
4522         btrfs_mark_buffer_dirty(path->nodes[0]);
4523         btrfs_free_path(path);
4524
4525         ret = update_block_group(trans, root, ins->objectid, ins->offset,
4526                                  1, 0);
4527         if (ret) {
4528                 printk(KERN_ERR "btrfs update block group failed for %llu "
4529                        "%llu\n", (unsigned long long)ins->objectid,
4530                        (unsigned long long)ins->offset);
4531                 BUG();
4532         }
4533         return ret;
4534 }
4535
4536 static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
4537                                      struct btrfs_root *root,
4538                                      u64 parent, u64 root_objectid,
4539                                      u64 flags, struct btrfs_disk_key *key,
4540                                      int level, struct btrfs_key *ins)
4541 {
4542         int ret;
4543         struct btrfs_fs_info *fs_info = root->fs_info;
4544         struct btrfs_extent_item *extent_item;
4545         struct btrfs_tree_block_info *block_info;
4546         struct btrfs_extent_inline_ref *iref;
4547         struct btrfs_path *path;
4548         struct extent_buffer *leaf;
4549         u32 size = sizeof(*extent_item) + sizeof(*block_info) + sizeof(*iref);
4550
4551         path = btrfs_alloc_path();
4552         BUG_ON(!path);
4553
4554         path->leave_spinning = 1;
4555         ret = btrfs_insert_empty_item(trans, fs_info->extent_root, path,
4556                                       ins, size);
4557         BUG_ON(ret);
4558
4559         leaf = path->nodes[0];
4560         extent_item = btrfs_item_ptr(leaf, path->slots[0],
4561                                      struct btrfs_extent_item);
4562         btrfs_set_extent_refs(leaf, extent_item, 1);
4563         btrfs_set_extent_generation(leaf, extent_item, trans->transid);
4564         btrfs_set_extent_flags(leaf, extent_item,
4565                                flags | BTRFS_EXTENT_FLAG_TREE_BLOCK);
4566         block_info = (struct btrfs_tree_block_info *)(extent_item + 1);
4567
4568         btrfs_set_tree_block_key(leaf, block_info, key);
4569         btrfs_set_tree_block_level(leaf, block_info, level);
4570
4571         iref = (struct btrfs_extent_inline_ref *)(block_info + 1);
4572         if (parent > 0) {
4573                 BUG_ON(!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF));
4574                 btrfs_set_extent_inline_ref_type(leaf, iref,
4575                                                  BTRFS_SHARED_BLOCK_REF_KEY);
4576                 btrfs_set_extent_inline_ref_offset(leaf, iref, parent);
4577         } else {
4578                 btrfs_set_extent_inline_ref_type(leaf, iref,
4579                                                  BTRFS_TREE_BLOCK_REF_KEY);
4580                 btrfs_set_extent_inline_ref_offset(leaf, iref, root_objectid);
4581         }
4582
4583         btrfs_mark_buffer_dirty(leaf);
4584         btrfs_free_path(path);
4585
4586         ret = update_block_group(trans, root, ins->objectid, ins->offset,
4587                                  1, 0);
4588         if (ret) {
4589                 printk(KERN_ERR "btrfs update block group failed for %llu "
4590                        "%llu\n", (unsigned long long)ins->objectid,
4591                        (unsigned long long)ins->offset);
4592                 BUG();
4593         }
4594         return ret;
4595 }
4596
4597 int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
4598                                      struct btrfs_root *root,
4599                                      u64 root_objectid, u64 owner,
4600                                      u64 offset, struct btrfs_key *ins)
4601 {
4602         int ret;
4603
4604         BUG_ON(root_objectid == BTRFS_TREE_LOG_OBJECTID);
4605
4606         ret = btrfs_add_delayed_data_ref(trans, ins->objectid, ins->offset,
4607                                          0, root_objectid, owner, offset,
4608                                          BTRFS_ADD_DELAYED_EXTENT, NULL);
4609         return ret;
4610 }
4611
4612 /*
4613  * this is used by the tree logging recovery code.  It records that
4614  * an extent has been allocated and makes sure to clear the free
4615  * space cache bits as well
4616  */
4617 int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
4618                                    struct btrfs_root *root,
4619                                    u64 root_objectid, u64 owner, u64 offset,
4620                                    struct btrfs_key *ins)
4621 {
4622         int ret;
4623         struct btrfs_block_group_cache *block_group;
4624         struct btrfs_caching_control *caching_ctl;
4625         u64 start = ins->objectid;
4626         u64 num_bytes = ins->offset;
4627
4628         block_group = btrfs_lookup_block_group(root->fs_info, ins->objectid);
4629         cache_block_group(block_group);
4630         caching_ctl = get_caching_control(block_group);
4631
4632         if (!caching_ctl) {
4633                 BUG_ON(!block_group_cache_done(block_group));
4634                 ret = btrfs_remove_free_space(block_group, start, num_bytes);
4635                 BUG_ON(ret);
4636         } else {
4637                 mutex_lock(&caching_ctl->mutex);
4638
4639                 if (start >= caching_ctl->progress) {
4640                         ret = add_excluded_extent(root, start, num_bytes);
4641                         BUG_ON(ret);
4642                 } else if (start + num_bytes <= caching_ctl->progress) {
4643                         ret = btrfs_remove_free_space(block_group,
4644                                                       start, num_bytes);
4645                         BUG_ON(ret);
4646                 } else {
4647                         num_bytes = caching_ctl->progress - start;
4648                         ret = btrfs_remove_free_space(block_group,
4649                                                       start, num_bytes);
4650                         BUG_ON(ret);
4651
4652                         start = caching_ctl->progress;
4653                         num_bytes = ins->objectid + ins->offset -
4654                                     caching_ctl->progress;
4655                         ret = add_excluded_extent(root, start, num_bytes);
4656                         BUG_ON(ret);
4657                 }
4658
4659                 mutex_unlock(&caching_ctl->mutex);
4660                 put_caching_control(caching_ctl);
4661         }
4662
4663         update_reserved_extents(block_group, ins->offset, 1);
4664         btrfs_put_block_group(block_group);
4665         ret = alloc_reserved_file_extent(trans, root, 0, root_objectid,
4666                                          0, owner, offset, ins, 1);
4667         return ret;
4668 }
4669
4670 /*
4671  * finds a free extent and does all the dirty work required for allocation
4672  * returns the key for the extent through ins, and a tree buffer for
4673  * the first block of the extent through buf.
4674  *
4675  * returns 0 if everything worked, non-zero otherwise.
4676  */
4677 static int alloc_tree_block(struct btrfs_trans_handle *trans,
4678                             struct btrfs_root *root,
4679                             u64 num_bytes, u64 parent, u64 root_objectid,
4680                             struct btrfs_disk_key *key, int level,
4681                             u64 empty_size, u64 hint_byte, u64 search_end,
4682                             struct btrfs_key *ins)
4683 {
4684         int ret;
4685         u64 flags = 0;
4686
4687         ret = btrfs_reserve_extent(trans, root, num_bytes, num_bytes,
4688                                    empty_size, hint_byte, search_end,
4689                                    ins, 0);
4690         if (ret)
4691                 return ret;
4692
4693         if (root_objectid == BTRFS_TREE_RELOC_OBJECTID) {
4694                 if (parent == 0)
4695                         parent = ins->objectid;
4696                 flags |= BTRFS_BLOCK_FLAG_FULL_BACKREF;
4697         } else
4698                 BUG_ON(parent > 0);
4699
4700         if (root_objectid != BTRFS_TREE_LOG_OBJECTID) {
4701                 struct btrfs_delayed_extent_op *extent_op;
4702                 extent_op = kmalloc(sizeof(*extent_op), GFP_NOFS);
4703                 BUG_ON(!extent_op);
4704                 if (key)
4705                         memcpy(&extent_op->key, key, sizeof(extent_op->key));
4706                 else
4707                         memset(&extent_op->key, 0, sizeof(extent_op->key));
4708                 extent_op->flags_to_set = flags;
4709                 extent_op->update_key = 1;
4710                 extent_op->update_flags = 1;
4711                 extent_op->is_data = 0;
4712
4713                 ret = btrfs_add_delayed_tree_ref(trans, ins->objectid,
4714                                         ins->offset, parent, root_objectid,
4715                                         level, BTRFS_ADD_DELAYED_EXTENT,
4716                                         extent_op);
4717                 BUG_ON(ret);
4718         }
4719         return ret;
4720 }
4721
4722 struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
4723                                             struct btrfs_root *root,
4724                                             u64 bytenr, u32 blocksize,
4725                                             int level)
4726 {
4727         struct extent_buffer *buf;
4728
4729         buf = btrfs_find_create_tree_block(root, bytenr, blocksize);
4730         if (!buf)
4731                 return ERR_PTR(-ENOMEM);
4732         btrfs_set_header_generation(buf, trans->transid);
4733         btrfs_set_buffer_lockdep_class(buf, level);
4734         btrfs_tree_lock(buf);
4735         clean_tree_block(trans, root, buf);
4736
4737         btrfs_set_lock_blocking(buf);
4738         btrfs_set_buffer_uptodate(buf);
4739
4740         if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
4741                 set_extent_dirty(&root->dirty_log_pages, buf->start,
4742                          buf->start + buf->len - 1, GFP_NOFS);
4743         } else {
4744                 set_extent_dirty(&trans->transaction->dirty_pages, buf->start,
4745                          buf->start + buf->len - 1, GFP_NOFS);
4746         }
4747         trans->blocks_used++;
4748         /* this returns a buffer locked for blocking */
4749         return buf;
4750 }
4751
4752 /*
4753  * helper function to allocate a block for a given tree
4754  * returns the tree buffer or NULL.
4755  */
4756 struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
4757                                         struct btrfs_root *root, u32 blocksize,
4758                                         u64 parent, u64 root_objectid,
4759                                         struct btrfs_disk_key *key, int level,
4760                                         u64 hint, u64 empty_size)
4761 {
4762         struct btrfs_key ins;
4763         int ret;
4764         struct extent_buffer *buf;
4765
4766         ret = alloc_tree_block(trans, root, blocksize, parent, root_objectid,
4767                                key, level, empty_size, hint, (u64)-1, &ins);
4768         if (ret) {
4769                 BUG_ON(ret > 0);
4770                 return ERR_PTR(ret);
4771         }
4772
4773         buf = btrfs_init_new_buffer(trans, root, ins.objectid,
4774                                     blocksize, level);
4775         return buf;
4776 }
4777
4778 struct walk_control {
4779         u64 refs[BTRFS_MAX_LEVEL];
4780         u64 flags[BTRFS_MAX_LEVEL];
4781         struct btrfs_key update_progress;
4782         int stage;
4783         int level;
4784         int shared_level;
4785         int update_ref;
4786         int keep_locks;
4787         int reada_slot;
4788         int reada_count;
4789 };
4790
4791 #define DROP_REFERENCE  1
4792 #define UPDATE_BACKREF  2
4793
4794 static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
4795                                      struct btrfs_root *root,
4796                                      struct walk_control *wc,
4797                                      struct btrfs_path *path)
4798 {
4799         u64 bytenr;
4800         u64 generation;
4801         u64 refs;
4802         u64 last = 0;
4803         u32 nritems;
4804         u32 blocksize;
4805         struct btrfs_key key;
4806         struct extent_buffer *eb;
4807         int ret;
4808         int slot;
4809         int nread = 0;
4810
4811         if (path->slots[wc->level] < wc->reada_slot) {
4812                 wc->reada_count = wc->reada_count * 2 / 3;
4813                 wc->reada_count = max(wc->reada_count, 2);
4814         } else {
4815                 wc->reada_count = wc->reada_count * 3 / 2;
4816                 wc->reada_count = min_t(int, wc->reada_count,
4817                                         BTRFS_NODEPTRS_PER_BLOCK(root));
4818         }
4819
4820         eb = path->nodes[wc->level];
4821         nritems = btrfs_header_nritems(eb);
4822         blocksize = btrfs_level_size(root, wc->level - 1);
4823
4824         for (slot = path->slots[wc->level]; slot < nritems; slot++) {
4825                 if (nread >= wc->reada_count)
4826                         break;
4827
4828                 cond_resched();
4829                 bytenr = btrfs_node_blockptr(eb, slot);
4830                 generation = btrfs_node_ptr_generation(eb, slot);
4831
4832                 if (slot == path->slots[wc->level])
4833                         goto reada;
4834
4835                 if (wc->stage == UPDATE_BACKREF &&
4836                     generation <= root->root_key.offset)
4837                         continue;
4838
4839                 if (wc->stage == DROP_REFERENCE) {
4840                         ret = btrfs_lookup_extent_info(trans, root,
4841                                                 bytenr, blocksize,
4842                                                 &refs, NULL);
4843                         BUG_ON(ret);
4844                         BUG_ON(refs == 0);
4845                         if (refs == 1)
4846                                 goto reada;
4847
4848                         if (!wc->update_ref ||
4849                             generation <= root->root_key.offset)
4850                                 continue;
4851                         btrfs_node_key_to_cpu(eb, &key, slot);
4852                         ret = btrfs_comp_cpu_keys(&key,
4853                                                   &wc->update_progress);
4854                         if (ret < 0)
4855                                 continue;
4856                 }
4857 reada:
4858                 ret = readahead_tree_block(root, bytenr, blocksize,
4859                                            generation);
4860                 if (ret)
4861                         break;
4862                 last = bytenr + blocksize;
4863                 nread++;
4864         }
4865         wc->reada_slot = slot;
4866 }
4867
4868 /*
4869  * hepler to process tree block while walking down the tree.
4870  *
4871  * when wc->stage == UPDATE_BACKREF, this function updates
4872  * back refs for pointers in the block.
4873  *
4874  * NOTE: return value 1 means we should stop walking down.
4875  */
4876 static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
4877                                    struct btrfs_root *root,
4878                                    struct btrfs_path *path,
4879                                    struct walk_control *wc)
4880 {
4881         int level = wc->level;
4882         struct extent_buffer *eb = path->nodes[level];
4883         u64 flag = BTRFS_BLOCK_FLAG_FULL_BACKREF;
4884         int ret;
4885
4886         if (wc->stage == UPDATE_BACKREF &&
4887             btrfs_header_owner(eb) != root->root_key.objectid)
4888                 return 1;
4889
4890         /*
4891          * when reference count of tree block is 1, it won't increase
4892          * again. once full backref flag is set, we never clear it.
4893          */
4894         if ((wc->stage == DROP_REFERENCE && wc->refs[level] != 1) ||
4895             (wc->stage == UPDATE_BACKREF && !(wc->flags[level] & flag))) {
4896                 BUG_ON(!path->locks[level]);
4897                 ret = btrfs_lookup_extent_info(trans, root,
4898                                                eb->start, eb->len,
4899                                                &wc->refs[level],
4900                                                &wc->flags[level]);
4901                 BUG_ON(ret);
4902                 BUG_ON(wc->refs[level] == 0);
4903         }
4904
4905         if (wc->stage == DROP_REFERENCE) {
4906                 if (wc->refs[level] > 1)
4907                         return 1;
4908
4909                 if (path->locks[level] && !wc->keep_locks) {
4910                         btrfs_tree_unlock(eb);
4911                         path->locks[level] = 0;
4912                 }
4913                 return 0;
4914         }
4915
4916         /* wc->stage == UPDATE_BACKREF */
4917         if (!(wc->flags[level] & flag)) {
4918                 BUG_ON(!path->locks[level]);
4919                 ret = btrfs_inc_ref(trans, root, eb, 1);
4920                 BUG_ON(ret);
4921                 ret = btrfs_dec_ref(trans, root, eb, 0);
4922                 BUG_ON(ret);
4923                 ret = btrfs_set_disk_extent_flags(trans, root, eb->start,
4924                                                   eb->len, flag, 0);
4925                 BUG_ON(ret);
4926                 wc->flags[level] |= flag;
4927         }
4928
4929         /*
4930          * the block is shared by multiple trees, so it's not good to
4931          * keep the tree lock
4932          */
4933         if (path->locks[level] && level > 0) {
4934                 btrfs_tree_unlock(eb);
4935                 path->locks[level] = 0;
4936         }
4937         return 0;
4938 }
4939
4940 /*
4941  * hepler to process tree block pointer.
4942  *
4943  * when wc->stage == DROP_REFERENCE, this function checks
4944  * reference count of the block pointed to. if the block
4945  * is shared and we need update back refs for the subtree
4946  * rooted at the block, this function changes wc->stage to
4947  * UPDATE_BACKREF. if the block is shared and there is no
4948  * need to update back, this function drops the reference
4949  * to the block.
4950  *
4951  * NOTE: return value 1 means we should stop walking down.
4952  */
4953 static noinline int do_walk_down(struct btrfs_trans_handle *trans,
4954                                  struct btrfs_root *root,
4955                                  struct btrfs_path *path,
4956                                  struct walk_control *wc)
4957 {
4958         u64 bytenr;
4959         u64 generation;
4960         u64 parent;
4961         u32 blocksize;
4962         struct btrfs_key key;
4963         struct extent_buffer *next;
4964         int level = wc->level;
4965         int reada = 0;
4966         int ret = 0;
4967
4968         generation = btrfs_node_ptr_generation(path->nodes[level],
4969                                                path->slots[level]);
4970         /*
4971          * if the lower level block was created before the snapshot
4972          * was created, we know there is no need to update back refs
4973          * for the subtree
4974          */
4975         if (wc->stage == UPDATE_BACKREF &&
4976             generation <= root->root_key.offset)
4977                 return 1;
4978
4979         bytenr = btrfs_node_blockptr(path->nodes[level], path->slots[level]);
4980         blocksize = btrfs_level_size(root, level - 1);
4981
4982         next = btrfs_find_tree_block(root, bytenr, blocksize);
4983         if (!next) {
4984                 next = btrfs_find_create_tree_block(root, bytenr, blocksize);
4985                 reada = 1;
4986         }
4987         btrfs_tree_lock(next);
4988         btrfs_set_lock_blocking(next);
4989
4990         if (wc->stage == DROP_REFERENCE) {
4991                 ret = btrfs_lookup_extent_info(trans, root, bytenr, blocksize,
4992                                                &wc->refs[level - 1],
4993                                                &wc->flags[level - 1]);
4994                 BUG_ON(ret);
4995                 BUG_ON(wc->refs[level - 1] == 0);
4996
4997                 if (wc->refs[level - 1] > 1) {
4998                         if (!wc->update_ref ||
4999                             generation <= root->root_key.offset)
5000                                 goto skip;
5001
5002                         btrfs_node_key_to_cpu(path->nodes[level], &key,
5003                                               path->slots[level]);
5004                         ret = btrfs_comp_cpu_keys(&key, &wc->update_progress);
5005                         if (ret < 0)
5006                                 goto skip;
5007
5008                         wc->stage = UPDATE_BACKREF;
5009                         wc->shared_level = level - 1;
5010                 }
5011         }
5012
5013         if (!btrfs_buffer_uptodate(next, generation)) {
5014                 btrfs_tree_unlock(next);
5015                 free_extent_buffer(next);
5016                 next = NULL;
5017         }
5018
5019         if (!next) {
5020                 if (reada && level == 1)
5021                         reada_walk_down(trans, root, wc, path);
5022                 next = read_tree_block(root, bytenr, blocksize, generation);
5023                 btrfs_tree_lock(next);
5024                 btrfs_set_lock_blocking(next);
5025         }
5026
5027         level--;
5028         BUG_ON(level != btrfs_header_level(next));
5029         path->nodes[level] = next;
5030         path->slots[level] = 0;
5031         path->locks[level] = 1;
5032         wc->level = level;
5033         if (wc->level == 1)
5034                 wc->reada_slot = 0;
5035         return 0;
5036 skip:
5037         wc->refs[level - 1] = 0;
5038         wc->flags[level - 1] = 0;
5039
5040         if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) {
5041                 parent = path->nodes[level]->start;
5042         } else {
5043                 BUG_ON(root->root_key.objectid !=
5044                        btrfs_header_owner(path->nodes[level]));
5045                 parent = 0;
5046         }
5047
5048         ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent,
5049                                 root->root_key.objectid, level - 1, 0);
5050         BUG_ON(ret);
5051
5052         btrfs_tree_unlock(next);
5053         free_extent_buffer(next);
5054         return 1;
5055 }
5056
5057 /*
5058  * hepler to process tree block while walking up the tree.
5059  *
5060  * when wc->stage == DROP_REFERENCE, this function drops
5061  * reference count on the block.
5062  *
5063  * when wc->stage == UPDATE_BACKREF, this function changes
5064  * wc->stage back to DROP_REFERENCE if we changed wc->stage
5065  * to UPDATE_BACKREF previously while processing the block.
5066  *
5067  * NOTE: return value 1 means we should stop walking up.
5068  */
5069 static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
5070                                  struct btrfs_root *root,
5071                                  struct btrfs_path *path,
5072                                  struct walk_control *wc)
5073 {
5074         int ret = 0;
5075         int level = wc->level;
5076         struct extent_buffer *eb = path->nodes[level];
5077         u64 parent = 0;
5078
5079         if (wc->stage == UPDATE_BACKREF) {
5080                 BUG_ON(wc->shared_level < level);
5081                 if (level < wc->shared_level)
5082                         goto out;
5083
5084                 ret = find_next_key(path, level + 1, &wc->update_progress);
5085                 if (ret > 0)
5086                         wc->update_ref = 0;
5087
5088                 wc->stage = DROP_REFERENCE;
5089                 wc->shared_level = -1;
5090                 path->slots[level] = 0;
5091
5092                 /*
5093                  * check reference count again if the block isn't locked.
5094                  * we should start walking down the tree again if reference
5095                  * count is one.
5096                  */
5097                 if (!path->locks[level]) {
5098                         BUG_ON(level == 0);
5099                         btrfs_tree_lock(eb);
5100                         btrfs_set_lock_blocking(eb);
5101                         path->locks[level] = 1;
5102
5103                         ret = btrfs_lookup_extent_info(trans, root,
5104                                                        eb->start, eb->len,
5105                                                        &wc->refs[level],
5106                                                        &wc->flags[level]);
5107                         BUG_ON(ret);
5108                         BUG_ON(wc->refs[level] == 0);
5109                         if (wc->refs[level] == 1) {
5110                                 btrfs_tree_unlock(eb);
5111                                 path->locks[level] = 0;
5112                                 return 1;
5113                         }
5114                 }
5115         }
5116
5117         /* wc->stage == DROP_REFERENCE */
5118         BUG_ON(wc->refs[level] > 1 && !path->locks[level]);
5119
5120         if (wc->refs[level] == 1) {
5121                 if (level == 0) {
5122                         if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
5123                                 ret = btrfs_dec_ref(trans, root, eb, 1);
5124                         else
5125                                 ret = btrfs_dec_ref(trans, root, eb, 0);
5126                         BUG_ON(ret);
5127                 }
5128                 /* make block locked assertion in clean_tree_block happy */
5129                 if (!path->locks[level] &&
5130                     btrfs_header_generation(eb) == trans->transid) {
5131                         btrfs_tree_lock(eb);
5132                         btrfs_set_lock_blocking(eb);
5133                         path->locks[level] = 1;
5134                 }
5135                 clean_tree_block(trans, root, eb);
5136         }
5137
5138         if (eb == root->node) {
5139                 if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
5140                         parent = eb->start;
5141                 else
5142                         BUG_ON(root->root_key.objectid !=
5143                                btrfs_header_owner(eb));
5144         } else {
5145                 if (wc->flags[level + 1] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
5146                         parent = path->nodes[level + 1]->start;
5147                 else
5148                         BUG_ON(root->root_key.objectid !=
5149                                btrfs_header_owner(path->nodes[level + 1]));
5150         }
5151
5152         ret = btrfs_free_extent(trans, root, eb->start, eb->len, parent,
5153                                 root->root_key.objectid, level, 0);
5154         BUG_ON(ret);
5155 out:
5156         wc->refs[level] = 0;
5157         wc->flags[level] = 0;
5158         return ret;
5159 }
5160
5161 static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
5162                                    struct btrfs_root *root,
5163                                    struct btrfs_path *path,
5164                                    struct walk_control *wc)
5165 {
5166         int level = wc->level;
5167         int ret;
5168
5169         while (level >= 0) {
5170                 if (path->slots[level] >=
5171                     btrfs_header_nritems(path->nodes[level]))
5172                         break;
5173
5174                 ret = walk_down_proc(trans, root, path, wc);
5175                 if (ret > 0)
5176                         break;
5177
5178                 if (level == 0)
5179                         break;
5180
5181                 ret = do_walk_down(trans, root, path, wc);
5182                 if (ret > 0) {
5183                         path->slots[level]++;
5184                         continue;
5185                 }
5186                 level = wc->level;
5187         }
5188         return 0;
5189 }
5190
5191 static noinline int walk_up_tree(struct btrfs_trans_handle *trans,
5192                                  struct btrfs_root *root,
5193                                  struct btrfs_path *path,
5194                                  struct walk_control *wc, int max_level)
5195 {
5196         int level = wc->level;
5197         int ret;
5198
5199         path->slots[level] = btrfs_header_nritems(path->nodes[level]);
5200         while (level < max_level && path->nodes[level]) {
5201                 wc->level = level;
5202                 if (path->slots[level] + 1 <
5203                     btrfs_header_nritems(path->nodes[level])) {
5204                         path->slots[level]++;
5205                         return 0;
5206                 } else {
5207                         ret = walk_up_proc(trans, root, path, wc);
5208                         if (ret > 0)
5209                                 return 0;
5210
5211                         if (path->locks[level]) {
5212                                 btrfs_tree_unlock(path->nodes[level]);
5213                                 path->locks[level] = 0;
5214                         }
5215                         free_extent_buffer(path->nodes[level]);
5216                         path->nodes[level] = NULL;
5217                         level++;
5218                 }
5219         }
5220         return 1;
5221 }
5222
5223 /*
5224  * drop a subvolume tree.
5225  *
5226  * this function traverses the tree freeing any blocks that only
5227  * referenced by the tree.
5228  *
5229  * when a shared tree block is found. this function decreases its
5230  * reference count by one. if update_ref is true, this function
5231  * also make sure backrefs for the shared block and all lower level
5232  * blocks are properly updated.
5233  */
5234 int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref)
5235 {
5236         struct btrfs_path *path;
5237         struct btrfs_trans_handle *trans;
5238         struct btrfs_root *tree_root = root->fs_info->tree_root;
5239         struct btrfs_root_item *root_item = &root->root_item;
5240         struct walk_control *wc;
5241         struct btrfs_key key;
5242         int err = 0;
5243         int ret;
5244         int level;
5245
5246         path = btrfs_alloc_path();
5247         BUG_ON(!path);
5248
5249         wc = kzalloc(sizeof(*wc), GFP_NOFS);
5250         BUG_ON(!wc);
5251
5252         trans = btrfs_start_transaction(tree_root, 1);
5253
5254         if (btrfs_disk_key_objectid(&root_item->drop_progress) == 0) {
5255                 level = btrfs_header_level(root->node);
5256                 path->nodes[level] = btrfs_lock_root_node(root);
5257                 btrfs_set_lock_blocking(path->nodes[level]);
5258                 path->slots[level] = 0;
5259                 path->locks[level] = 1;
5260                 memset(&wc->update_progress, 0,
5261                        sizeof(wc->update_progress));
5262         } else {
5263                 btrfs_disk_key_to_cpu(&key, &root_item->drop_progress);
5264                 memcpy(&wc->update_progress, &key,
5265                        sizeof(wc->update_progress));
5266
5267                 level = root_item->drop_level;
5268                 BUG_ON(level == 0);
5269                 path->lowest_level = level;
5270                 ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
5271                 path->lowest_level = 0;
5272                 if (ret < 0) {
5273                         err = ret;
5274                         goto out;
5275                 }
5276                 WARN_ON(ret > 0);
5277
5278                 /*
5279                  * unlock our path, this is safe because only this
5280                  * function is allowed to delete this snapshot
5281                  */
5282                 btrfs_unlock_up_safe(path, 0);
5283
5284                 level = btrfs_header_level(root->node);
5285                 while (1) {
5286                         btrfs_tree_lock(path->nodes[level]);
5287                         btrfs_set_lock_blocking(path->nodes[level]);
5288
5289                         ret = btrfs_lookup_extent_info(trans, root,
5290                                                 path->nodes[level]->start,
5291                                                 path->nodes[level]->len,
5292                                                 &wc->refs[level],
5293                                                 &wc->flags[level]);
5294                         BUG_ON(ret);
5295                         BUG_ON(wc->refs[level] == 0);
5296
5297                         if (level == root_item->drop_level)
5298                                 break;
5299
5300                         btrfs_tree_unlock(path->nodes[level]);
5301                         WARN_ON(wc->refs[level] != 1);
5302                         level--;
5303                 }
5304         }
5305
5306         wc->level = level;
5307         wc->shared_level = -1;
5308         wc->stage = DROP_REFERENCE;
5309         wc->update_ref = update_ref;
5310         wc->keep_locks = 0;
5311         wc->reada_count = BTRFS_NODEPTRS_PER_BLOCK(root);
5312
5313         while (1) {
5314                 ret = walk_down_tree(trans, root, path, wc);
5315                 if (ret < 0) {
5316                         err = ret;
5317                         break;
5318                 }
5319
5320                 ret = walk_up_tree(trans, root, path, wc, BTRFS_MAX_LEVEL);
5321                 if (ret < 0) {
5322                         err = ret;
5323                         break;
5324                 }
5325
5326                 if (ret > 0) {
5327                         BUG_ON(wc->stage != DROP_REFERENCE);
5328                         break;
5329                 }
5330
5331                 if (wc->stage == DROP_REFERENCE) {
5332                         level = wc->level;
5333                         btrfs_node_key(path->nodes[level],
5334                                        &root_item->drop_progress,
5335                                        path->slots[level]);
5336                         root_item->drop_level = level;
5337                 }
5338
5339                 BUG_ON(wc->level == 0);
5340                 if (trans->transaction->in_commit ||
5341                     trans->transaction->delayed_refs.flushing) {
5342                         ret = btrfs_update_root(trans, tree_root,
5343                                                 &root->root_key,
5344                                                 root_item);
5345                         BUG_ON(ret);
5346
5347                         btrfs_end_transaction(trans, tree_root);
5348                         trans = btrfs_start_transaction(tree_root, 1);
5349                 } else {
5350                         unsigned long update;
5351                         update = trans->delayed_ref_updates;
5352                         trans->delayed_ref_updates = 0;
5353                         if (update)
5354                                 btrfs_run_delayed_refs(trans, tree_root,
5355                                                        update);
5356                 }
5357         }
5358         btrfs_release_path(root, path);
5359         BUG_ON(err);
5360
5361         ret = btrfs_del_root(trans, tree_root, &root->root_key);
5362         BUG_ON(ret);
5363
5364         if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
5365                 ret = btrfs_find_last_root(tree_root, root->root_key.objectid,
5366                                            NULL, NULL);
5367                 BUG_ON(ret < 0);
5368                 if (ret > 0) {
5369                         ret = btrfs_del_orphan_item(trans, tree_root,
5370                                                     root->root_key.objectid);
5371                         BUG_ON(ret);
5372                 }
5373         }
5374
5375         if (root->in_radix) {
5376                 btrfs_free_fs_root(tree_root->fs_info, root);
5377         } else {
5378                 free_extent_buffer(root->node);
5379                 free_extent_buffer(root->commit_root);
5380                 kfree(root);
5381         }
5382 out:
5383         btrfs_end_transaction(trans, tree_root);
5384         kfree(wc);
5385         btrfs_free_path(path);
5386         return err;
5387 }
5388
5389 /*
5390  * drop subtree rooted at tree block 'node'.
5391  *
5392  * NOTE: this function will unlock and release tree block 'node'
5393  */
5394 int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
5395                         struct btrfs_root *root,
5396                         struct extent_buffer *node,
5397                         struct extent_buffer *parent)
5398 {
5399         struct btrfs_path *path;
5400         struct walk_control *wc;
5401         int level;
5402         int parent_level;
5403         int ret = 0;
5404         int wret;
5405
5406         BUG_ON(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID);
5407
5408         path = btrfs_alloc_path();
5409         BUG_ON(!path);
5410
5411         wc = kzalloc(sizeof(*wc), GFP_NOFS);
5412         BUG_ON(!wc);
5413
5414         btrfs_assert_tree_locked(parent);
5415         parent_level = btrfs_header_level(parent);
5416         extent_buffer_get(parent);
5417         path->nodes[parent_level] = parent;
5418         path->slots[parent_level] = btrfs_header_nritems(parent);
5419
5420         btrfs_assert_tree_locked(node);
5421         level = btrfs_header_level(node);
5422         path->nodes[level] = node;
5423         path->slots[level] = 0;
5424         path->locks[level] = 1;
5425
5426         wc->refs[parent_level] = 1;
5427         wc->flags[parent_level] = BTRFS_BLOCK_FLAG_FULL_BACKREF;
5428         wc->level = level;
5429         wc->shared_level = -1;
5430         wc->stage = DROP_REFERENCE;
5431         wc->update_ref = 0;
5432         wc->keep_locks = 1;
5433         wc->reada_count = BTRFS_NODEPTRS_PER_BLOCK(root);
5434
5435         while (1) {
5436                 wret = walk_down_tree(trans, root, path, wc);
5437                 if (wret < 0) {
5438                         ret = wret;
5439                         break;
5440                 }
5441
5442                 wret = walk_up_tree(trans, root, path, wc, parent_level);
5443                 if (wret < 0)
5444                         ret = wret;
5445                 if (wret != 0)
5446                         break;
5447         }
5448
5449         kfree(wc);
5450         btrfs_free_path(path);
5451         return ret;
5452 }
5453
5454 #if 0
5455 static unsigned long calc_ra(unsigned long start, unsigned long last,
5456                              unsigned long nr)
5457 {
5458         return min(last, start + nr - 1);
5459 }
5460
5461 static noinline int relocate_inode_pages(struct inode *inode, u64 start,
5462                                          u64 len)
5463 {
5464         u64 page_start;
5465         u64 page_end;
5466         unsigned long first_index;
5467         unsigned long last_index;
5468         unsigned long i;
5469         struct page *page;
5470         struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
5471         struct file_ra_state *ra;
5472         struct btrfs_ordered_extent *ordered;
5473         unsigned int total_read = 0;
5474         unsigned int total_dirty = 0;
5475         int ret = 0;
5476
5477         ra = kzalloc(sizeof(*ra), GFP_NOFS);
5478
5479         mutex_lock(&inode->i_mutex);
5480         first_index = start >> PAGE_CACHE_SHIFT;
5481         last_index = (start + len - 1) >> PAGE_CACHE_SHIFT;
5482
5483         /* make sure the dirty trick played by the caller work */
5484         ret = invalidate_inode_pages2_range(inode->i_mapping,
5485                                             first_index, last_index);
5486         if (ret)
5487                 goto out_unlock;
5488
5489         file_ra_state_init(ra, inode->i_mapping);
5490
5491         for (i = first_index ; i <= last_index; i++) {
5492                 if (total_read % ra->ra_pages == 0) {
5493                         btrfs_force_ra(inode->i_mapping, ra, NULL, i,
5494                                        calc_ra(i, last_index, ra->ra_pages));
5495                 }
5496                 total_read++;
5497 again:
5498                 if (((u64)i << PAGE_CACHE_SHIFT) > i_size_read(inode))
5499                         BUG_ON(1);
5500                 page = grab_cache_page(inode->i_mapping, i);
5501                 if (!page) {
5502                         ret = -ENOMEM;
5503                         goto out_unlock;
5504                 }
5505                 if (!PageUptodate(page)) {
5506                         btrfs_readpage(NULL, page);
5507                         lock_page(page);
5508                         if (!PageUptodate(page)) {
5509                                 unlock_page(page);
5510                                 page_cache_release(page);
5511                                 ret = -EIO;
5512                                 goto out_unlock;
5513                         }
5514                 }
5515                 wait_on_page_writeback(page);
5516
5517                 page_start = (u64)page->index << PAGE_CACHE_SHIFT;
5518                 page_end = page_start + PAGE_CACHE_SIZE - 1;
5519                 lock_extent(io_tree, page_start, page_end, GFP_NOFS);
5520
5521                 ordered = btrfs_lookup_ordered_extent(inode, page_start);
5522                 if (ordered) {
5523                         unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
5524                         unlock_page(page);
5525                         page_cache_release(page);
5526                         btrfs_start_ordered_extent(inode, ordered, 1);
5527                         btrfs_put_ordered_extent(ordered);
5528                         goto again;
5529                 }
5530                 set_page_extent_mapped(page);
5531
5532                 if (i == first_index)
5533                         set_extent_bits(io_tree, page_start, page_end,
5534                                         EXTENT_BOUNDARY, GFP_NOFS);
5535                 btrfs_set_extent_delalloc(inode, page_start, page_end);
5536
5537                 set_page_dirty(page);
5538                 total_dirty++;
5539
5540                 unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
5541                 unlock_page(page);
5542                 page_cache_release(page);
5543         }
5544
5545 out_unlock:
5546         kfree(ra);
5547         mutex_unlock(&inode->i_mutex);
5548         balance_dirty_pages_ratelimited_nr(inode->i_mapping, total_dirty);
5549         return ret;
5550 }
5551
5552 static noinline int relocate_data_extent(struct inode *reloc_inode,
5553                                          struct btrfs_key *extent_key,
5554                                          u64 offset)
5555 {
5556         struct btrfs_root *root = BTRFS_I(reloc_inode)->root;
5557         struct extent_map_tree *em_tree = &BTRFS_I(reloc_inode)->extent_tree;
5558         struct extent_map *em;
5559         u64 start = extent_key->objectid - offset;
5560         u64 end = start + extent_key->offset - 1;
5561
5562         em = alloc_extent_map(GFP_NOFS);
5563         BUG_ON(!em || IS_ERR(em));
5564
5565         em->start = start;
5566         em->len = extent_key->offset;
5567         em->block_len = extent_key->offset;
5568         em->block_start = extent_key->objectid;
5569         em->bdev = root->fs_info->fs_devices->latest_bdev;
5570         set_bit(EXTENT_FLAG_PINNED, &em->flags);
5571
5572         /* setup extent map to cheat btrfs_readpage */
5573         lock_extent(&BTRFS_I(reloc_inode)->io_tree, start, end, GFP_NOFS);
5574         while (1) {
5575                 int ret;
5576                 write_lock(&em_tree->lock);
5577                 ret = add_extent_mapping(em_tree, em);
5578                 write_unlock(&em_tree->lock);
5579                 if (ret != -EEXIST) {
5580                         free_extent_map(em);
5581                         break;
5582                 }
5583                 btrfs_drop_extent_cache(reloc_inode, start, end, 0);
5584         }
5585         unlock_extent(&BTRFS_I(reloc_inode)->io_tree, start, end, GFP_NOFS);
5586
5587         return relocate_inode_pages(reloc_inode, start, extent_key->offset);
5588 }
5589
5590 struct btrfs_ref_path {
5591         u64 extent_start;
5592         u64 nodes[BTRFS_MAX_LEVEL];
5593         u64 root_objectid;
5594         u64 root_generation;
5595         u64 owner_objectid;
5596         u32 num_refs;
5597         int lowest_level;
5598         int current_level;
5599         int shared_level;
5600
5601         struct btrfs_key node_keys[BTRFS_MAX_LEVEL];
5602         u64 new_nodes[BTRFS_MAX_LEVEL];
5603 };
5604
5605 struct disk_extent {
5606         u64 ram_bytes;
5607         u64 disk_bytenr;
5608         u64 disk_num_bytes;
5609         u64 offset;
5610         u64 num_bytes;
5611         u8 compression;
5612         u8 encryption;
5613         u16 other_encoding;
5614 };
5615
5616 static int is_cowonly_root(u64 root_objectid)
5617 {
5618         if (root_objectid == BTRFS_ROOT_TREE_OBJECTID ||
5619             root_objectid == BTRFS_EXTENT_TREE_OBJECTID ||
5620             root_objectid == BTRFS_CHUNK_TREE_OBJECTID ||
5621             root_objectid == BTRFS_DEV_TREE_OBJECTID ||
5622             root_objectid == BTRFS_TREE_LOG_OBJECTID ||
5623             root_objectid == BTRFS_CSUM_TREE_OBJECTID)
5624                 return 1;
5625         return 0;
5626 }
5627
5628 static noinline int __next_ref_path(struct btrfs_trans_handle *trans,
5629                                     struct btrfs_root *extent_root,
5630                                     struct btrfs_ref_path *ref_path,
5631                                     int first_time)
5632 {
5633         struct extent_buffer *leaf;
5634         struct btrfs_path *path;
5635         struct btrfs_extent_ref *ref;
5636         struct btrfs_key key;
5637         struct btrfs_key found_key;
5638         u64 bytenr;
5639         u32 nritems;
5640         int level;
5641         int ret = 1;
5642
5643         path = btrfs_alloc_path();
5644         if (!path)
5645                 return -ENOMEM;
5646
5647         if (first_time) {
5648                 ref_path->lowest_level = -1;
5649                 ref_path->current_level = -1;
5650                 ref_path->shared_level = -1;
5651                 goto walk_up;
5652         }
5653 walk_down:
5654         level = ref_path->current_level - 1;
5655         while (level >= -1) {
5656                 u64 parent;
5657                 if (level < ref_path->lowest_level)
5658                         break;
5659
5660                 if (level >= 0)
5661                         bytenr = ref_path->nodes[level];
5662                 else
5663                         bytenr = ref_path->extent_start;
5664                 BUG_ON(bytenr == 0);
5665
5666                 parent = ref_path->nodes[level + 1];
5667                 ref_path->nodes[level + 1] = 0;
5668                 ref_path->current_level = level;
5669                 BUG_ON(parent == 0);
5670
5671                 key.objectid = bytenr;
5672                 key.offset = parent + 1;
5673                 key.type = BTRFS_EXTENT_REF_KEY;
5674
5675                 ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 0);
5676                 if (ret < 0)
5677                         goto out;
5678                 BUG_ON(ret == 0);
5679
5680                 leaf = path->nodes[0];
5681                 nritems = btrfs_header_nritems(leaf);
5682                 if (path->slots[0] >= nritems) {
5683                         ret = btrfs_next_leaf(extent_root, path);
5684                         if (ret < 0)
5685                                 goto out;
5686                         if (ret > 0)
5687                                 goto next;
5688                         leaf = path->nodes[0];
5689                 }
5690
5691                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
5692                 if (found_key.objectid == bytenr &&
5693                     found_key.type == BTRFS_EXTENT_REF_KEY) {
5694                         if (level < ref_path->shared_level)
5695                                 ref_path->shared_level = level;
5696                         goto found;
5697                 }
5698 next:
5699                 level--;
5700                 btrfs_release_path(extent_root, path);
5701                 cond_resched();
5702         }
5703         /* reached lowest level */
5704         ret = 1;
5705         goto out;
5706 walk_up:
5707         level = ref_path->current_level;
5708         while (level < BTRFS_MAX_LEVEL - 1) {
5709                 u64 ref_objectid;
5710
5711                 if (level >= 0)
5712                         bytenr = ref_path->nodes[level];
5713                 else
5714                         bytenr = ref_path->extent_start;
5715
5716                 BUG_ON(bytenr == 0);
5717
5718                 key.objectid = bytenr;
5719                 key.offset = 0;
5720                 key.type = BTRFS_EXTENT_REF_KEY;
5721
5722                 ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 0);
5723                 if (ret < 0)
5724                         goto out;
5725
5726                 leaf = path->nodes[0];
5727                 nritems = btrfs_header_nritems(leaf);
5728                 if (path->slots[0] >= nritems) {
5729                         ret = btrfs_next_leaf(extent_root, path);
5730                         if (ret < 0)
5731                                 goto out;
5732                         if (ret > 0) {
5733                                 /* the extent was freed by someone */
5734                                 if (ref_path->lowest_level == level)
5735                                         goto out;
5736                                 btrfs_release_path(extent_root, path);
5737                                 goto walk_down;
5738                         }
5739                         leaf = path->nodes[0];
5740                 }
5741
5742                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
5743                 if (found_key.objectid != bytenr ||
5744                                 found_key.type != BTRFS_EXTENT_REF_KEY) {
5745                         /* the extent was freed by someone */
5746                         if (ref_path->lowest_level == level) {
5747                                 ret = 1;
5748                                 goto out;
5749                         }
5750                         btrfs_release_path(extent_root, path);
5751                         goto walk_down;
5752                 }
5753 found:
5754                 ref = btrfs_item_ptr(leaf, path->slots[0],
5755                                 struct btrfs_extent_ref);
5756                 ref_objectid = btrfs_ref_objectid(leaf, ref);
5757                 if (ref_objectid < BTRFS_FIRST_FREE_OBJECTID) {
5758                         if (first_time) {
5759                                 level = (int)ref_objectid;
5760                                 BUG_ON(level >= BTRFS_MAX_LEVEL);
5761                                 ref_path->lowest_level = level;
5762                                 ref_path->current_level = level;
5763                                 ref_path->nodes[level] = bytenr;
5764                         } else {
5765                                 WARN_ON(ref_objectid != level);
5766                         }
5767                 } else {
5768                         WARN_ON(level != -1);
5769                 }
5770                 first_time = 0;
5771
5772                 if (ref_path->lowest_level == level) {
5773                         ref_path->owner_objectid = ref_objectid;
5774                         ref_path->num_refs = btrfs_ref_num_refs(leaf, ref);
5775                 }
5776
5777                 /*
5778                  * the block is tree root or the block isn't in reference
5779                  * counted tree.
5780                  */
5781                 if (found_key.objectid == found_key.offset ||
5782                     is_cowonly_root(btrfs_ref_root(leaf, ref))) {
5783                         ref_path->root_objectid = btrfs_ref_root(leaf, ref);
5784                         ref_path->root_generation =
5785                                 btrfs_ref_generation(leaf, ref);
5786                         if (level < 0) {
5787                                 /* special reference from the tree log */
5788                                 ref_path->nodes[0] = found_key.offset;
5789                                 ref_path->current_level = 0;
5790                         }
5791                         ret = 0;
5792                         goto out;
5793                 }
5794
5795                 level++;
5796                 BUG_ON(ref_path->nodes[level] != 0);
5797                 ref_path->nodes[level] = found_key.offset;
5798                 ref_path->current_level = level;
5799
5800                 /*
5801                  * the reference was created in the running transaction,
5802                  * no need to continue walking up.
5803                  */
5804                 if (btrfs_ref_generation(leaf, ref) == trans->transid) {
5805                         ref_path->root_objectid = btrfs_ref_root(leaf, ref);
5806                         ref_path->root_generation =
5807                                 btrfs_ref_generation(leaf, ref);
5808                         ret = 0;
5809                         goto out;
5810                 }
5811
5812                 btrfs_release_path(extent_root, path);
5813                 cond_resched();
5814         }
5815         /* reached max tree level, but no tree root found. */
5816         BUG();
5817 out:
5818         btrfs_free_path(path);
5819         return ret;
5820 }
5821
5822 static int btrfs_first_ref_path(struct btrfs_trans_handle *trans,
5823                                 struct btrfs_root *extent_root,
5824                                 struct btrfs_ref_path *ref_path,
5825                                 u64 extent_start)
5826 {
5827         memset(ref_path, 0, sizeof(*ref_path));
5828         ref_path->extent_start = extent_start;
5829
5830         return __next_ref_path(trans, extent_root, ref_path, 1);
5831 }
5832
5833 static int btrfs_next_ref_path(struct btrfs_trans_handle *trans,
5834                                struct btrfs_root *extent_root,
5835                                struct btrfs_ref_path *ref_path)
5836 {
5837         return __next_ref_path(trans, extent_root, ref_path, 0);
5838 }
5839
5840 static noinline int get_new_locations(struct inode *reloc_inode,
5841                                       struct btrfs_key *extent_key,
5842                                       u64 offset, int no_fragment,
5843                                       struct disk_extent **extents,
5844                                       int *nr_extents)
5845 {
5846         struct btrfs_root *root = BTRFS_I(reloc_inode)->root;
5847         struct btrfs_path *path;
5848         struct btrfs_file_extent_item *fi;
5849         struct extent_buffer *leaf;
5850         struct disk_extent *exts = *extents;
5851         struct btrfs_key found_key;
5852         u64 cur_pos;
5853         u64 last_byte;
5854         u32 nritems;
5855         int nr = 0;
5856         int max = *nr_extents;
5857         int ret;
5858
5859         WARN_ON(!no_fragment && *extents);
5860         if (!exts) {
5861                 max = 1;
5862                 exts = kmalloc(sizeof(*exts) * max, GFP_NOFS);
5863                 if (!exts)
5864                         return -ENOMEM;
5865         }
5866
5867         path = btrfs_alloc_path();
5868         BUG_ON(!path);
5869
5870         cur_pos = extent_key->objectid - offset;
5871         last_byte = extent_key->objectid + extent_key->offset;
5872         ret = btrfs_lookup_file_extent(NULL, root, path, reloc_inode->i_ino,
5873                                        cur_pos, 0);
5874         if (ret < 0)
5875                 goto out;
5876         if (ret > 0) {
5877                 ret = -ENOENT;
5878                 goto out;
5879         }
5880
5881         while (1) {
5882                 leaf = path->nodes[0];
5883                 nritems = btrfs_header_nritems(leaf);
5884                 if (path->slots[0] >= nritems) {
5885                         ret = btrfs_next_leaf(root, path);
5886                         if (ret < 0)
5887                                 goto out;
5888                         if (ret > 0)
5889                                 break;
5890                         leaf = path->nodes[0];
5891                 }
5892
5893                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
5894                 if (found_key.offset != cur_pos ||
5895                     found_key.type != BTRFS_EXTENT_DATA_KEY ||
5896                     found_key.objectid != reloc_inode->i_ino)
5897                         break;
5898
5899                 fi = btrfs_item_ptr(leaf, path->slots[0],
5900                                     struct btrfs_file_extent_item);
5901                 if (btrfs_file_extent_type(leaf, fi) !=
5902                     BTRFS_FILE_EXTENT_REG ||
5903                     btrfs_file_extent_disk_bytenr(leaf, fi) == 0)
5904                         break;
5905
5906                 if (nr == max) {
5907                         struct disk_extent *old = exts;
5908                         max *= 2;
5909                         exts = kzalloc(sizeof(*exts) * max, GFP_NOFS);
5910                         memcpy(exts, old, sizeof(*exts) * nr);
5911                         if (old != *extents)
5912                                 kfree(old);
5913                 }
5914
5915                 exts[nr].disk_bytenr =
5916                         btrfs_file_extent_disk_bytenr(leaf, fi);
5917                 exts[nr].disk_num_bytes =
5918                         btrfs_file_extent_disk_num_bytes(leaf, fi);
5919                 exts[nr].offset = btrfs_file_extent_offset(leaf, fi);
5920                 exts[nr].num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
5921                 exts[nr].ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
5922                 exts[nr].compression = btrfs_file_extent_compression(leaf, fi);
5923                 exts[nr].encryption = btrfs_file_extent_encryption(leaf, fi);
5924                 exts[nr].other_encoding = btrfs_file_extent_other_encoding(leaf,
5925                                                                            fi);
5926                 BUG_ON(exts[nr].offset > 0);
5927                 BUG_ON(exts[nr].compression || exts[nr].encryption);
5928                 BUG_ON(exts[nr].num_bytes != exts[nr].disk_num_bytes);
5929
5930                 cur_pos += exts[nr].num_bytes;
5931                 nr++;
5932
5933                 if (cur_pos + offset >= last_byte)
5934                         break;
5935
5936                 if (no_fragment) {
5937                         ret = 1;
5938                         goto out;
5939                 }
5940                 path->slots[0]++;
5941         }
5942
5943         BUG_ON(cur_pos + offset > last_byte);
5944         if (cur_pos + offset < last_byte) {
5945                 ret = -ENOENT;
5946                 goto out;
5947         }
5948         ret = 0;
5949 out:
5950         btrfs_free_path(path);
5951         if (ret) {
5952                 if (exts != *extents)
5953                         kfree(exts);
5954         } else {
5955                 *extents = exts;
5956                 *nr_extents = nr;
5957         }
5958         return ret;
5959 }
5960
5961 static noinline int replace_one_extent(struct btrfs_trans_handle *trans,
5962                                         struct btrfs_root *root,
5963                                         struct btrfs_path *path,
5964                                         struct btrfs_key *extent_key,
5965                                         struct btrfs_key *leaf_key,
5966                                         struct btrfs_ref_path *ref_path,
5967                                         struct disk_extent *new_extents,
5968                                         int nr_extents)
5969 {
5970         struct extent_buffer *leaf;
5971         struct btrfs_file_extent_item *fi;
5972         struct inode *inode = NULL;
5973         struct btrfs_key key;
5974         u64 lock_start = 0;
5975         u64 lock_end = 0;
5976         u64 num_bytes;
5977         u64 ext_offset;
5978         u64 search_end = (u64)-1;
5979         u32 nritems;
5980         int nr_scaned = 0;
5981         int extent_locked = 0;
5982         int extent_type;
5983         int ret;
5984
5985         memcpy(&key, leaf_key, sizeof(key));
5986         if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS) {
5987                 if (key.objectid < ref_path->owner_objectid ||
5988                     (key.objectid == ref_path->owner_objectid &&
5989                      key.type < BTRFS_EXTENT_DATA_KEY)) {
5990                         key.objectid = ref_path->owner_objectid;
5991                         key.type = BTRFS_EXTENT_DATA_KEY;
5992                         key.offset = 0;
5993                 }
5994         }
5995
5996         while (1) {
5997                 ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
5998                 if (ret < 0)
5999                         goto out;
6000
6001                 leaf = path->nodes[0];
6002                 nritems = btrfs_header_nritems(leaf);
6003 next:
6004                 if (extent_locked && ret > 0) {
6005                         /*
6006                          * the file extent item was modified by someone
6007                          * before the extent got locked.
6008                          */
6009                         unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
6010                                       lock_end, GFP_NOFS);
6011                         extent_locked = 0;
6012                 }
6013
6014                 if (path->slots[0] >= nritems) {
6015                         if (++nr_scaned > 2)
6016                                 break;
6017
6018                         BUG_ON(extent_locked);
6019                         ret = btrfs_next_leaf(root, path);
6020                         if (ret < 0)
6021                                 goto out;
6022                         if (ret > 0)
6023                                 break;
6024                         leaf = path->nodes[0];
6025                         nritems = btrfs_header_nritems(leaf);
6026                 }
6027
6028                 btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
6029
6030                 if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS) {
6031                         if ((key.objectid > ref_path->owner_objectid) ||
6032                             (key.objectid == ref_path->owner_objectid &&
6033                              key.type > BTRFS_EXTENT_DATA_KEY) ||
6034                             key.offset >= search_end)
6035                                 break;
6036                 }
6037
6038                 if (inode && key.objectid != inode->i_ino) {
6039                         BUG_ON(extent_locked);
6040                         btrfs_release_path(root, path);
6041                         mutex_unlock(&inode->i_mutex);
6042                         iput(inode);
6043                         inode = NULL;
6044                         continue;
6045                 }
6046
6047                 if (key.type != BTRFS_EXTENT_DATA_KEY) {
6048                         path->slots[0]++;
6049                         ret = 1;
6050                         goto next;
6051                 }
6052                 fi = btrfs_item_ptr(leaf, path->slots[0],
6053                                     struct btrfs_file_extent_item);
6054                 extent_type = btrfs_file_extent_type(leaf, fi);
6055                 if ((extent_type != BTRFS_FILE_EXTENT_REG &&
6056                      extent_type != BTRFS_FILE_EXTENT_PREALLOC) ||
6057                     (btrfs_file_extent_disk_bytenr(leaf, fi) !=
6058                      extent_key->objectid)) {
6059                         path->slots[0]++;
6060                         ret = 1;
6061                         goto next;
6062                 }
6063
6064                 num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
6065                 ext_offset = btrfs_file_extent_offset(leaf, fi);
6066
6067                 if (search_end == (u64)-1) {
6068                         search_end = key.offset - ext_offset +
6069                                 btrfs_file_extent_ram_bytes(leaf, fi);
6070                 }
6071
6072                 if (!extent_locked) {
6073                         lock_start = key.offset;
6074                         lock_end = lock_start + num_bytes - 1;
6075                 } else {
6076                         if (lock_start > key.offset ||
6077                             lock_end + 1 < key.offset + num_bytes) {
6078                                 unlock_extent(&BTRFS_I(inode)->io_tree,
6079                                               lock_start, lock_end, GFP_NOFS);
6080                                 extent_locked = 0;
6081                         }
6082                 }
6083
6084                 if (!inode) {
6085                         btrfs_release_path(root, path);
6086
6087                         inode = btrfs_iget_locked(root->fs_info->sb,
6088                                                   key.objectid, root);
6089                         if (inode->i_state & I_NEW) {
6090                                 BTRFS_I(inode)->root = root;
6091                                 BTRFS_I(inode)->location.objectid =
6092                                         key.objectid;
6093                                 BTRFS_I(inode)->location.type =
6094                                         BTRFS_INODE_ITEM_KEY;
6095                                 BTRFS_I(inode)->location.offset = 0;
6096                                 btrfs_read_locked_inode(inode);
6097                                 unlock_new_inode(inode);
6098                         }
6099                         /*
6100                          * some code call btrfs_commit_transaction while
6101                          * holding the i_mutex, so we can't use mutex_lock
6102                          * here.
6103                          */
6104                         if (is_bad_inode(inode) ||
6105                             !mutex_trylock(&inode->i_mutex)) {
6106                                 iput(inode);
6107                                 inode = NULL;
6108                                 key.offset = (u64)-1;
6109                                 goto skip;
6110                         }
6111                 }
6112
6113                 if (!extent_locked) {
6114                         struct btrfs_ordered_extent *ordered;
6115
6116                         btrfs_release_path(root, path);
6117
6118                         lock_extent(&BTRFS_I(inode)->io_tree, lock_start,
6119                                     lock_end, GFP_NOFS);
6120                         ordered = btrfs_lookup_first_ordered_extent(inode,
6121                                                                     lock_end);
6122                         if (ordered &&
6123                             ordered->file_offset <= lock_end &&
6124                             ordered->file_offset + ordered->len > lock_start) {
6125                                 unlock_extent(&BTRFS_I(inode)->io_tree,
6126                                               lock_start, lock_end, GFP_NOFS);
6127                                 btrfs_start_ordered_extent(inode, ordered, 1);
6128                                 btrfs_put_ordered_extent(ordered);
6129                                 key.offset += num_bytes;
6130                                 goto skip;
6131                         }
6132                         if (ordered)
6133                                 btrfs_put_ordered_extent(ordered);
6134
6135                         extent_locked = 1;
6136                         continue;
6137                 }
6138
6139                 if (nr_extents == 1) {
6140                         /* update extent pointer in place */
6141                         btrfs_set_file_extent_disk_bytenr(leaf, fi,
6142                                                 new_extents[0].disk_bytenr);
6143                         btrfs_set_file_extent_disk_num_bytes(leaf, fi,
6144                                                 new_extents[0].disk_num_bytes);
6145                         btrfs_mark_buffer_dirty(leaf);
6146
6147                         btrfs_drop_extent_cache(inode, key.offset,
6148                                                 key.offset + num_bytes - 1, 0);
6149
6150                         ret = btrfs_inc_extent_ref(trans, root,
6151                                                 new_extents[0].disk_bytenr,
6152                                                 new_extents[0].disk_num_bytes,
6153                                                 leaf->start,
6154                                                 root->root_key.objectid,
6155                                                 trans->transid,
6156                                                 key.objectid);
6157                         BUG_ON(ret);
6158
6159                         ret = btrfs_free_extent(trans, root,
6160                                                 extent_key->objectid,
6161                                                 extent_key->offset,
6162                                                 leaf->start,
6163                                                 btrfs_header_owner(leaf),
6164                                                 btrfs_header_generation(leaf),
6165                                                 key.objectid, 0);
6166                         BUG_ON(ret);
6167
6168                         btrfs_release_path(root, path);
6169                         key.offset += num_bytes;
6170                 } else {
6171                         BUG_ON(1);
6172 #if 0
6173                         u64 alloc_hint;
6174                         u64 extent_len;
6175                         int i;
6176                         /*
6177                          * drop old extent pointer at first, then insert the
6178                          * new pointers one bye one
6179                          */
6180                         btrfs_release_path(root, path);
6181                         ret = btrfs_drop_extents(trans, root, inode, key.offset,
6182                                                  key.offset + num_bytes,
6183                                                  key.offset, &alloc_hint);
6184                         BUG_ON(ret);
6185
6186                         for (i = 0; i < nr_extents; i++) {
6187                                 if (ext_offset >= new_extents[i].num_bytes) {
6188                                         ext_offset -= new_extents[i].num_bytes;
6189                                         continue;
6190                                 }
6191                                 extent_len = min(new_extents[i].num_bytes -
6192                                                  ext_offset, num_bytes);
6193
6194                                 ret = btrfs_insert_empty_item(trans, root,
6195                                                               path, &key,
6196                                                               sizeof(*fi));
6197                                 BUG_ON(ret);
6198
6199                                 leaf = path->nodes[0];
6200                                 fi = btrfs_item_ptr(leaf, path->slots[0],
6201                                                 struct btrfs_file_extent_item);
6202                                 btrfs_set_file_extent_generation(leaf, fi,
6203                                                         trans->transid);
6204                                 btrfs_set_file_extent_type(leaf, fi,
6205                                                         BTRFS_FILE_EXTENT_REG);
6206                                 btrfs_set_file_extent_disk_bytenr(leaf, fi,
6207                                                 new_extents[i].disk_bytenr);
6208                                 btrfs_set_file_extent_disk_num_bytes(leaf, fi,
6209                                                 new_extents[i].disk_num_bytes);
6210                                 btrfs_set_file_extent_ram_bytes(leaf, fi,
6211                                                 new_extents[i].ram_bytes);
6212
6213                                 btrfs_set_file_extent_compression(leaf, fi,
6214                                                 new_extents[i].compression);
6215                                 btrfs_set_file_extent_encryption(leaf, fi,
6216                                                 new_extents[i].encryption);
6217                                 btrfs_set_file_extent_other_encoding(leaf, fi,
6218                                                 new_extents[i].other_encoding);
6219
6220                                 btrfs_set_file_extent_num_bytes(leaf, fi,
6221                                                         extent_len);
6222                                 ext_offset += new_extents[i].offset;
6223                                 btrfs_set_file_extent_offset(leaf, fi,
6224                                                         ext_offset);
6225                                 btrfs_mark_buffer_dirty(leaf);
6226
6227                                 btrfs_drop_extent_cache(inode, key.offset,
6228                                                 key.offset + extent_len - 1, 0);
6229
6230                                 ret = btrfs_inc_extent_ref(trans, root,
6231                                                 new_extents[i].disk_bytenr,
6232                                                 new_extents[i].disk_num_bytes,
6233                                                 leaf->start,
6234                                                 root->root_key.objectid,
6235                                                 trans->transid, key.objectid);
6236                                 BUG_ON(ret);
6237                                 btrfs_release_path(root, path);
6238
6239                                 inode_add_bytes(inode, extent_len);
6240
6241                                 ext_offset = 0;
6242                                 num_bytes -= extent_len;
6243                                 key.offset += extent_len;
6244
6245                                 if (num_bytes == 0)
6246                                         break;
6247                         }
6248                         BUG_ON(i >= nr_extents);
6249 #endif
6250                 }
6251
6252                 if (extent_locked) {
6253                         unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
6254                                       lock_end, GFP_NOFS);
6255                         extent_locked = 0;
6256                 }
6257 skip:
6258                 if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS &&
6259                     key.offset >= search_end)
6260                         break;
6261
6262                 cond_resched();
6263         }
6264         ret = 0;
6265 out:
6266         btrfs_release_path(root, path);
6267         if (inode) {
6268                 mutex_unlock(&inode->i_mutex);
6269                 if (extent_locked) {
6270                         unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
6271                                       lock_end, GFP_NOFS);
6272                 }
6273                 iput(inode);
6274         }
6275         return ret;
6276 }
6277
6278 int btrfs_reloc_tree_cache_ref(struct btrfs_trans_handle *trans,
6279                                struct btrfs_root *root,
6280                                struct extent_buffer *buf, u64 orig_start)
6281 {
6282         int level;
6283         int ret;
6284
6285         BUG_ON(btrfs_header_generation(buf) != trans->transid);
6286         BUG_ON(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID);
6287
6288         level = btrfs_header_level(buf);
6289         if (level == 0) {
6290                 struct btrfs_leaf_ref *ref;
6291                 struct btrfs_leaf_ref *orig_ref;
6292
6293                 orig_ref = btrfs_lookup_leaf_ref(root, orig_start);
6294                 if (!orig_ref)
6295                         return -ENOENT;
6296
6297                 ref = btrfs_alloc_leaf_ref(root, orig_ref->nritems);
6298                 if (!ref) {
6299                         btrfs_free_leaf_ref(root, orig_ref);
6300                         return -ENOMEM;
6301                 }
6302
6303                 ref->nritems = orig_ref->nritems;
6304                 memcpy(ref->extents, orig_ref->extents,
6305                         sizeof(ref->extents[0]) * ref->nritems);
6306
6307                 btrfs_free_leaf_ref(root, orig_ref);
6308
6309                 ref->root_gen = trans->transid;
6310                 ref->bytenr = buf->start;
6311                 ref->owner = btrfs_header_owner(buf);
6312                 ref->generation = btrfs_header_generation(buf);
6313
6314                 ret = btrfs_add_leaf_ref(root, ref, 0);
6315                 WARN_ON(ret);
6316                 btrfs_free_leaf_ref(root, ref);
6317         }
6318         return 0;
6319 }
6320
6321 static noinline int invalidate_extent_cache(struct btrfs_root *root,
6322                                         struct extent_buffer *leaf,
6323                                         struct btrfs_block_group_cache *group,
6324                                         struct btrfs_root *target_root)
6325 {
6326         struct btrfs_key key;
6327         struct inode *inode = NULL;
6328         struct btrfs_file_extent_item *fi;
6329         u64 num_bytes;
6330         u64 skip_objectid = 0;
6331         u32 nritems;
6332         u32 i;
6333
6334         nritems = btrfs_header_nritems(leaf);
6335         for (i = 0; i < nritems; i++) {
6336                 btrfs_item_key_to_cpu(leaf, &key, i);
6337                 if (key.objectid == skip_objectid ||
6338                     key.type != BTRFS_EXTENT_DATA_KEY)
6339                         continue;
6340                 fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
6341                 if (btrfs_file_extent_type(leaf, fi) ==
6342                     BTRFS_FILE_EXTENT_INLINE)
6343                         continue;
6344                 if (btrfs_file_extent_disk_bytenr(leaf, fi) == 0)
6345                         continue;
6346                 if (!inode || inode->i_ino != key.objectid) {
6347                         iput(inode);
6348                         inode = btrfs_ilookup(target_root->fs_info->sb,
6349                                               key.objectid, target_root, 1);
6350                 }
6351                 if (!inode) {
6352                         skip_objectid = key.objectid;
6353                         continue;
6354                 }
6355                 num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
6356
6357                 lock_extent(&BTRFS_I(inode)->io_tree, key.offset,
6358                             key.offset + num_bytes - 1, GFP_NOFS);
6359                 btrfs_drop_extent_cache(inode, key.offset,
6360                                         key.offset + num_bytes - 1, 1);
6361                 unlock_extent(&BTRFS_I(inode)->io_tree, key.offset,
6362                               key.offset + num_bytes - 1, GFP_NOFS);
6363                 cond_resched();
6364         }
6365         iput(inode);
6366         return 0;
6367 }
6368
6369 static noinline int replace_extents_in_leaf(struct btrfs_trans_handle *trans,
6370                                         struct btrfs_root *root,
6371                                         struct extent_buffer *leaf,
6372                                         struct btrfs_block_group_cache *group,
6373                                         struct inode *reloc_inode)
6374 {
6375         struct btrfs_key key;
6376         struct btrfs_key extent_key;
6377         struct btrfs_file_extent_item *fi;
6378         struct btrfs_leaf_ref *ref;
6379         struct disk_extent *new_extent;
6380         u64 bytenr;
6381         u64 num_bytes;
6382         u32 nritems;
6383         u32 i;
6384         int ext_index;
6385         int nr_extent;
6386         int ret;
6387
6388         new_extent = kmalloc(sizeof(*new_extent), GFP_NOFS);
6389         BUG_ON(!new_extent);
6390
6391         ref = btrfs_lookup_leaf_ref(root, leaf->start);
6392         BUG_ON(!ref);
6393
6394         ext_index = -1;
6395         nritems = btrfs_header_nritems(leaf);
6396         for (i = 0; i < nritems; i++) {
6397                 btrfs_item_key_to_cpu(leaf, &key, i);
6398                 if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
6399                         continue;
6400                 fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
6401                 if (btrfs_file_extent_type(leaf, fi) ==
6402                     BTRFS_FILE_EXTENT_INLINE)
6403                         continue;
6404                 bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
6405                 num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
6406                 if (bytenr == 0)
6407                         continue;
6408
6409                 ext_index++;
6410                 if (bytenr >= group->key.objectid + group->key.offset ||
6411                     bytenr + num_bytes <= group->key.objectid)
6412                         continue;
6413
6414                 extent_key.objectid = bytenr;
6415                 extent_key.offset = num_bytes;
6416                 extent_key.type = BTRFS_EXTENT_ITEM_KEY;
6417                 nr_extent = 1;
6418                 ret = get_new_locations(reloc_inode, &extent_key,
6419                                         group->key.objectid, 1,
6420                                         &new_extent, &nr_extent);
6421                 if (ret > 0)
6422                         continue;
6423                 BUG_ON(ret < 0);
6424
6425                 BUG_ON(ref->extents[ext_index].bytenr != bytenr);
6426                 BUG_ON(ref->extents[ext_index].num_bytes != num_bytes);
6427                 ref->extents[ext_index].bytenr = new_extent->disk_bytenr;
6428                 ref->extents[ext_index].num_bytes = new_extent->disk_num_bytes;
6429
6430                 btrfs_set_file_extent_disk_bytenr(leaf, fi,
6431                                                 new_extent->disk_bytenr);
6432                 btrfs_set_file_extent_disk_num_bytes(leaf, fi,
6433                                                 new_extent->disk_num_bytes);
6434                 btrfs_mark_buffer_dirty(leaf);
6435
6436                 ret = btrfs_inc_extent_ref(trans, root,
6437                                         new_extent->disk_bytenr,
6438                                         new_extent->disk_num_bytes,
6439                                         leaf->start,
6440                                         root->root_key.objectid,
6441                                         trans->transid, key.objectid);
6442                 BUG_ON(ret);
6443
6444                 ret = btrfs_free_extent(trans, root,
6445                                         bytenr, num_bytes, leaf->start,
6446                                         btrfs_header_owner(leaf),
6447                                         btrfs_header_generation(leaf),
6448                                         key.objectid, 0);
6449                 BUG_ON(ret);
6450                 cond_resched();
6451         }
6452         kfree(new_extent);
6453         BUG_ON(ext_index + 1 != ref->nritems);
6454         btrfs_free_leaf_ref(root, ref);
6455         return 0;
6456 }
6457
6458 int btrfs_free_reloc_root(struct btrfs_trans_handle *trans,
6459                           struct btrfs_root *root)
6460 {
6461         struct btrfs_root *reloc_root;
6462         int ret;
6463
6464         if (root->reloc_root) {
6465                 reloc_root = root->reloc_root;
6466                 root->reloc_root = NULL;
6467                 list_add(&reloc_root->dead_list,
6468                          &root->fs_info->dead_reloc_roots);
6469
6470                 btrfs_set_root_bytenr(&reloc_root->root_item,
6471                                       reloc_root->node->start);
6472                 btrfs_set_root_level(&root->root_item,
6473                                      btrfs_header_level(reloc_root->node));
6474                 memset(&reloc_root->root_item.drop_progress, 0,
6475                         sizeof(struct btrfs_disk_key));
6476                 reloc_root->root_item.drop_level = 0;
6477
6478                 ret = btrfs_update_root(trans, root->fs_info->tree_root,
6479                                         &reloc_root->root_key,
6480                                         &reloc_root->root_item);
6481                 BUG_ON(ret);
6482         }
6483         return 0;
6484 }
6485
6486 int btrfs_drop_dead_reloc_roots(struct btrfs_root *root)
6487 {
6488         struct btrfs_trans_handle *trans;
6489         struct btrfs_root *reloc_root;
6490         struct btrfs_root *prev_root = NULL;
6491         struct list_head dead_roots;
6492         int ret;
6493         unsigned long nr;
6494
6495         INIT_LIST_HEAD(&dead_roots);
6496         list_splice_init(&root->fs_info->dead_reloc_roots, &dead_roots);
6497
6498         while (!list_empty(&dead_roots)) {
6499                 reloc_root = list_entry(dead_roots.prev,
6500                                         struct btrfs_root, dead_list);
6501                 list_del_init(&reloc_root->dead_list);
6502
6503                 BUG_ON(reloc_root->commit_root != NULL);
6504                 while (1) {
6505                         trans = btrfs_join_transaction(root, 1);
6506                         BUG_ON(!trans);
6507
6508                         mutex_lock(&root->fs_info->drop_mutex);
6509                         ret = btrfs_drop_snapshot(trans, reloc_root);
6510                         if (ret != -EAGAIN)
6511                                 break;
6512                         mutex_unlock(&root->fs_info->drop_mutex);
6513
6514                         nr = trans->blocks_used;
6515                         ret = btrfs_end_transaction(trans, root);
6516                         BUG_ON(ret);
6517                         btrfs_btree_balance_dirty(root, nr);
6518                 }
6519
6520                 free_extent_buffer(reloc_root->node);
6521
6522                 ret = btrfs_del_root(trans, root->fs_info->tree_root,
6523                                      &reloc_root->root_key);
6524                 BUG_ON(ret);
6525                 mutex_unlock(&root->fs_info->drop_mutex);
6526
6527                 nr = trans->blocks_used;
6528                 ret = btrfs_end_transaction(trans, root);
6529                 BUG_ON(ret);
6530                 btrfs_btree_balance_dirty(root, nr);
6531
6532                 kfree(prev_root);
6533                 prev_root = reloc_root;
6534         }
6535         if (prev_root) {
6536                 btrfs_remove_leaf_refs(prev_root, (u64)-1, 0);
6537                 kfree(prev_root);
6538         }
6539         return 0;
6540 }
6541
6542 int btrfs_add_dead_reloc_root(struct btrfs_root *root)
6543 {
6544         list_add(&root->dead_list, &root->fs_info->dead_reloc_roots);
6545         return 0;
6546 }
6547
6548 int btrfs_cleanup_reloc_trees(struct btrfs_root *root)
6549 {
6550         struct btrfs_root *reloc_root;
6551         struct btrfs_trans_handle *trans;
6552         struct btrfs_key location;
6553         int found;
6554         int ret;
6555
6556         mutex_lock(&root->fs_info->tree_reloc_mutex);
6557         ret = btrfs_find_dead_roots(root, BTRFS_TREE_RELOC_OBJECTID, NULL);
6558         BUG_ON(ret);
6559         found = !list_empty(&root->fs_info->dead_reloc_roots);
6560         mutex_unlock(&root->fs_info->tree_reloc_mutex);
6561
6562         if (found) {
6563                 trans = btrfs_start_transaction(root, 1);
6564                 BUG_ON(!trans);
6565                 ret = btrfs_commit_transaction(trans, root);
6566                 BUG_ON(ret);
6567         }
6568
6569         location.objectid = BTRFS_DATA_RELOC_TREE_OBJECTID;
6570         location.offset = (u64)-1;
6571         location.type = BTRFS_ROOT_ITEM_KEY;
6572
6573         reloc_root = btrfs_read_fs_root_no_name(root->fs_info, &location);
6574         BUG_ON(!reloc_root);
6575         btrfs_orphan_cleanup(reloc_root);
6576         return 0;
6577 }
6578
6579 static noinline int init_reloc_tree(struct btrfs_trans_handle *trans,
6580                                     struct btrfs_root *root)
6581 {
6582         struct btrfs_root *reloc_root;
6583         struct extent_buffer *eb;
6584         struct btrfs_root_item *root_item;
6585         struct btrfs_key root_key;
6586         int ret;
6587
6588         BUG_ON(!root->ref_cows);
6589         if (root->reloc_root)
6590                 return 0;
6591
6592         root_item = kmalloc(sizeof(*root_item), GFP_NOFS);
6593         BUG_ON(!root_item);
6594
6595         ret = btrfs_copy_root(trans, root, root->commit_root,
6596                               &eb, BTRFS_TREE_RELOC_OBJECTID);
6597         BUG_ON(ret);
6598
6599         root_key.objectid = BTRFS_TREE_RELOC_OBJECTID;
6600         root_key.offset = root->root_key.objectid;
6601         root_key.type = BTRFS_ROOT_ITEM_KEY;
6602
6603         memcpy(root_item, &root->root_item, sizeof(root_item));
6604         btrfs_set_root_refs(root_item, 0);
6605         btrfs_set_root_bytenr(root_item, eb->start);
6606         btrfs_set_root_level(root_item, btrfs_header_level(eb));
6607         btrfs_set_root_generation(root_item, trans->transid);
6608
6609         btrfs_tree_unlock(eb);
6610         free_extent_buffer(eb);
6611
6612         ret = btrfs_insert_root(trans, root->fs_info->tree_root,
6613                                 &root_key, root_item);
6614         BUG_ON(ret);
6615         kfree(root_item);
6616
6617         reloc_root = btrfs_read_fs_root_no_radix(root->fs_info->tree_root,
6618                                                  &root_key);
6619         BUG_ON(!reloc_root);
6620         reloc_root->last_trans = trans->transid;
6621         reloc_root->commit_root = NULL;
6622         reloc_root->ref_tree = &root->fs_info->reloc_ref_tree;
6623
6624         root->reloc_root = reloc_root;
6625         return 0;
6626 }
6627
6628 /*
6629  * Core function of space balance.
6630  *
6631  * The idea is using reloc trees to relocate tree blocks in reference
6632  * counted roots. There is one reloc tree for each subvol, and all
6633  * reloc trees share same root key objectid. Reloc trees are snapshots
6634  * of the latest committed roots of subvols (root->commit_root).
6635  *
6636  * To relocate a tree block referenced by a subvol, there are two steps.
6637  * COW the block through subvol's reloc tree, then update block pointer
6638  * in the subvol to point to the new block. Since all reloc trees share
6639  * same root key objectid, doing special handing for tree blocks owned
6640  * by them is easy. Once a tree block has been COWed in one reloc tree,
6641  * we can use the resulting new block directly when the same block is
6642  * required to COW again through other reloc trees. By this way, relocated
6643  * tree blocks are shared between reloc trees, so they are also shared
6644  * between subvols.
6645  */
6646 static noinline int relocate_one_path(struct btrfs_trans_handle *trans,
6647                                       struct btrfs_root *root,
6648                                       struct btrfs_path *path,
6649                                       struct btrfs_key *first_key,
6650                                       struct btrfs_ref_path *ref_path,
6651                                       struct btrfs_block_group_cache *group,
6652                                       struct inode *reloc_inode)
6653 {
6654         struct btrfs_root *reloc_root;
6655         struct extent_buffer *eb = NULL;
6656         struct btrfs_key *keys;
6657         u64 *nodes;
6658         int level;
6659         int shared_level;
6660         int lowest_level = 0;
6661         int ret;
6662
6663         if (ref_path->owner_objectid < BTRFS_FIRST_FREE_OBJECTID)
6664                 lowest_level = ref_path->owner_objectid;
6665
6666         if (!root->ref_cows) {
6667                 path->lowest_level = lowest_level;
6668                 ret = btrfs_search_slot(trans, root, first_key, path, 0, 1);
6669                 BUG_ON(ret < 0);
6670                 path->lowest_level = 0;
6671                 btrfs_release_path(root, path);
6672                 return 0;
6673         }
6674
6675         mutex_lock(&root->fs_info->tree_reloc_mutex);
6676         ret = init_reloc_tree(trans, root);
6677         BUG_ON(ret);
6678         reloc_root = root->reloc_root;
6679
6680         shared_level = ref_path->shared_level;
6681         ref_path->shared_level = BTRFS_MAX_LEVEL - 1;
6682
6683         keys = ref_path->node_keys;
6684         nodes = ref_path->new_nodes;
6685         memset(&keys[shared_level + 1], 0,
6686                sizeof(*keys) * (BTRFS_MAX_LEVEL - shared_level - 1));
6687         memset(&nodes[shared_level + 1], 0,
6688                sizeof(*nodes) * (BTRFS_MAX_LEVEL - shared_level - 1));
6689
6690         if (nodes[lowest_level] == 0) {
6691                 path->lowest_level = lowest_level;
6692                 ret = btrfs_search_slot(trans, reloc_root, first_key, path,
6693                                         0, 1);
6694                 BUG_ON(ret);
6695                 for (level = lowest_level; level < BTRFS_MAX_LEVEL; level++) {
6696                         eb = path->nodes[level];
6697                         if (!eb || eb == reloc_root->node)
6698                                 break;
6699                         nodes[level] = eb->start;
6700                         if (level == 0)
6701                                 btrfs_item_key_to_cpu(eb, &keys[level], 0);
6702                         else
6703                                 btrfs_node_key_to_cpu(eb, &keys[level], 0);
6704                 }
6705                 if (nodes[0] &&
6706                     ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
6707                         eb = path->nodes[0];
6708                         ret = replace_extents_in_leaf(trans, reloc_root, eb,
6709                                                       group, reloc_inode);
6710                         BUG_ON(ret);
6711                 }
6712                 btrfs_release_path(reloc_root, path);
6713         } else {
6714                 ret = btrfs_merge_path(trans, reloc_root, keys, nodes,
6715                                        lowest_level);
6716                 BUG_ON(ret);
6717         }
6718
6719         /*
6720          * replace tree blocks in the fs tree with tree blocks in
6721          * the reloc tree.
6722          */
6723         ret = btrfs_merge_path(trans, root, keys, nodes, lowest_level);
6724         BUG_ON(ret < 0);
6725
6726         if (ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
6727                 ret = btrfs_search_slot(trans, reloc_root, first_key, path,
6728                                         0, 0);
6729                 BUG_ON(ret);
6730                 extent_buffer_get(path->nodes[0]);
6731                 eb = path->nodes[0];
6732                 btrfs_release_path(reloc_root, path);
6733                 ret = invalidate_extent_cache(reloc_root, eb, group, root);
6734                 BUG_ON(ret);
6735                 free_extent_buffer(eb);
6736         }
6737
6738         mutex_unlock(&root->fs_info->tree_reloc_mutex);
6739         path->lowest_level = 0;
6740         return 0;
6741 }
6742
6743 static noinline int relocate_tree_block(struct btrfs_trans_handle *trans,
6744                                         struct btrfs_root *root,
6745                                         struct btrfs_path *path,
6746                                         struct btrfs_key *first_key,
6747                                         struct btrfs_ref_path *ref_path)
6748 {
6749         int ret;
6750
6751         ret = relocate_one_path(trans, root, path, first_key,
6752                                 ref_path, NULL, NULL);
6753         BUG_ON(ret);
6754
6755         return 0;
6756 }
6757
6758 static noinline int del_extent_zero(struct btrfs_trans_handle *trans,
6759                                     struct btrfs_root *extent_root,
6760                                     struct btrfs_path *path,
6761                                     struct btrfs_key *extent_key)
6762 {
6763         int ret;
6764
6765         ret = btrfs_search_slot(trans, extent_root, extent_key, path, -1, 1);
6766         if (ret)
6767                 goto out;
6768         ret = btrfs_del_item(trans, extent_root, path);
6769 out:
6770         btrfs_release_path(extent_root, path);
6771         return ret;
6772 }
6773
6774 static noinline struct btrfs_root *read_ref_root(struct btrfs_fs_info *fs_info,
6775                                                 struct btrfs_ref_path *ref_path)
6776 {
6777         struct btrfs_key root_key;
6778
6779         root_key.objectid = ref_path->root_objectid;
6780         root_key.type = BTRFS_ROOT_ITEM_KEY;
6781         if (is_cowonly_root(ref_path->root_objectid))
6782                 root_key.offset = 0;
6783         else
6784                 root_key.offset = (u64)-1;
6785
6786         return btrfs_read_fs_root_no_name(fs_info, &root_key);
6787 }
6788
6789 static noinline int relocate_one_extent(struct btrfs_root *extent_root,
6790                                         struct btrfs_path *path,
6791                                         struct btrfs_key *extent_key,
6792                                         struct btrfs_block_group_cache *group,
6793                                         struct inode *reloc_inode, int pass)
6794 {
6795         struct btrfs_trans_handle *trans;
6796         struct btrfs_root *found_root;
6797         struct btrfs_ref_path *ref_path = NULL;
6798         struct disk_extent *new_extents = NULL;
6799         int nr_extents = 0;
6800         int loops;
6801         int ret;
6802         int level;
6803         struct btrfs_key first_key;
6804         u64 prev_block = 0;
6805
6806
6807         trans = btrfs_start_transaction(extent_root, 1);
6808         BUG_ON(!trans);
6809
6810         if (extent_key->objectid == 0) {
6811                 ret = del_extent_zero(trans, extent_root, path, extent_key);
6812                 goto out;
6813         }
6814
6815         ref_path = kmalloc(sizeof(*ref_path), GFP_NOFS);
6816         if (!ref_path) {
6817                 ret = -ENOMEM;
6818                 goto out;
6819         }
6820
6821         for (loops = 0; ; loops++) {
6822                 if (loops == 0) {
6823                         ret = btrfs_first_ref_path(trans, extent_root, ref_path,
6824                                                    extent_key->objectid);
6825                 } else {
6826                         ret = btrfs_next_ref_path(trans, extent_root, ref_path);
6827                 }
6828                 if (ret < 0)
6829                         goto out;
6830                 if (ret > 0)
6831                         break;
6832
6833                 if (ref_path->root_objectid == BTRFS_TREE_LOG_OBJECTID ||
6834                     ref_path->root_objectid == BTRFS_TREE_RELOC_OBJECTID)
6835                         continue;
6836
6837                 found_root = read_ref_root(extent_root->fs_info, ref_path);
6838                 BUG_ON(!found_root);
6839                 /*
6840                  * for reference counted tree, only process reference paths
6841                  * rooted at the latest committed root.
6842                  */
6843                 if (found_root->ref_cows &&
6844                     ref_path->root_generation != found_root->root_key.offset)
6845                         continue;
6846
6847                 if (ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
6848                         if (pass == 0) {
6849                                 /*
6850                                  * copy data extents to new locations
6851                                  */
6852                                 u64 group_start = group->key.objectid;
6853                                 ret = relocate_data_extent(reloc_inode,
6854                                                            extent_key,
6855                                                            group_start);
6856                                 if (ret < 0)
6857                                         goto out;
6858                                 break;
6859                         }
6860                         level = 0;
6861                 } else {
6862                         level = ref_path->owner_objectid;
6863                 }
6864
6865                 if (prev_block != ref_path->nodes[level]) {
6866                         struct extent_buffer *eb;
6867                         u64 block_start = ref_path->nodes[level];
6868                         u64 block_size = btrfs_level_size(found_root, level);
6869
6870                         eb = read_tree_block(found_root, block_start,
6871                                              block_size, 0);
6872                         btrfs_tree_lock(eb);
6873                         BUG_ON(level != btrfs_header_level(eb));
6874
6875                         if (level == 0)
6876                                 btrfs_item_key_to_cpu(eb, &first_key, 0);
6877                         else
6878                                 btrfs_node_key_to_cpu(eb, &first_key, 0);
6879
6880                         btrfs_tree_unlock(eb);
6881                         free_extent_buffer(eb);
6882                         prev_block = block_start;
6883                 }
6884
6885                 mutex_lock(&extent_root->fs_info->trans_mutex);
6886                 btrfs_record_root_in_trans(found_root);
6887                 mutex_unlock(&extent_root->fs_info->trans_mutex);
6888                 if (ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
6889                         /*
6890                          * try to update data extent references while
6891                          * keeping metadata shared between snapshots.
6892                          */
6893                         if (pass == 1) {
6894                                 ret = relocate_one_path(trans, found_root,
6895                                                 path, &first_key, ref_path,
6896                                                 group, reloc_inode);
6897                                 if (ret < 0)
6898                                         goto out;
6899                                 continue;
6900                         }
6901                         /*
6902                          * use fallback method to process the remaining
6903                          * references.
6904                          */
6905                         if (!new_extents) {
6906                                 u64 group_start = group->key.objectid;
6907                                 new_extents = kmalloc(sizeof(*new_extents),
6908                                                       GFP_NOFS);
6909                                 nr_extents = 1;
6910                                 ret = get_new_locations(reloc_inode,
6911                                                         extent_key,
6912                                                         group_start, 1,
6913                                                         &new_extents,
6914                                                         &nr_extents);
6915                                 if (ret)
6916                                         goto out;
6917                         }
6918                         ret = replace_one_extent(trans, found_root,
6919                                                 path, extent_key,
6920                                                 &first_key, ref_path,
6921                                                 new_extents, nr_extents);
6922                 } else {
6923                         ret = relocate_tree_block(trans, found_root, path,
6924                                                   &first_key, ref_path);
6925                 }
6926                 if (ret < 0)
6927                         goto out;
6928         }
6929         ret = 0;
6930 out:
6931         btrfs_end_transaction(trans, extent_root);
6932         kfree(new_extents);
6933         kfree(ref_path);
6934         return ret;
6935 }
6936 #endif
6937
6938 static u64 update_block_group_flags(struct btrfs_root *root, u64 flags)
6939 {
6940         u64 num_devices;
6941         u64 stripped = BTRFS_BLOCK_GROUP_RAID0 |
6942                 BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10;
6943
6944         num_devices = root->fs_info->fs_devices->rw_devices;
6945         if (num_devices == 1) {
6946                 stripped |= BTRFS_BLOCK_GROUP_DUP;
6947                 stripped = flags & ~stripped;
6948
6949                 /* turn raid0 into single device chunks */
6950                 if (flags & BTRFS_BLOCK_GROUP_RAID0)
6951                         return stripped;
6952
6953                 /* turn mirroring into duplication */
6954                 if (flags & (BTRFS_BLOCK_GROUP_RAID1 |
6955                              BTRFS_BLOCK_GROUP_RAID10))
6956                         return stripped | BTRFS_BLOCK_GROUP_DUP;
6957                 return flags;
6958         } else {
6959                 /* they already had raid on here, just return */
6960                 if (flags & stripped)
6961                         return flags;
6962
6963                 stripped |= BTRFS_BLOCK_GROUP_DUP;
6964                 stripped = flags & ~stripped;
6965
6966                 /* switch duplicated blocks with raid1 */
6967                 if (flags & BTRFS_BLOCK_GROUP_DUP)
6968                         return stripped | BTRFS_BLOCK_GROUP_RAID1;
6969
6970                 /* turn single device chunks into raid0 */
6971                 return stripped | BTRFS_BLOCK_GROUP_RAID0;
6972         }
6973         return flags;
6974 }
6975
6976 static int __alloc_chunk_for_shrink(struct btrfs_root *root,
6977                      struct btrfs_block_group_cache *shrink_block_group,
6978                      int force)
6979 {
6980         struct btrfs_trans_handle *trans;
6981         u64 new_alloc_flags;
6982         u64 calc;
6983
6984         spin_lock(&shrink_block_group->lock);
6985         if (btrfs_block_group_used(&shrink_block_group->item) +
6986             shrink_block_group->reserved > 0) {
6987                 spin_unlock(&shrink_block_group->lock);
6988
6989                 trans = btrfs_start_transaction(root, 1);
6990                 spin_lock(&shrink_block_group->lock);
6991
6992                 new_alloc_flags = update_block_group_flags(root,
6993                                                    shrink_block_group->flags);
6994                 if (new_alloc_flags != shrink_block_group->flags) {
6995                         calc =
6996                              btrfs_block_group_used(&shrink_block_group->item);
6997                 } else {
6998                         calc = shrink_block_group->key.offset;
6999                 }
7000                 spin_unlock(&shrink_block_group->lock);
7001
7002                 do_chunk_alloc(trans, root->fs_info->extent_root,
7003                                calc + 2 * 1024 * 1024, new_alloc_flags, force);
7004
7005                 btrfs_end_transaction(trans, root);
7006         } else
7007                 spin_unlock(&shrink_block_group->lock);
7008         return 0;
7009 }
7010
7011
7012 int btrfs_prepare_block_group_relocation(struct btrfs_root *root,
7013                                          struct btrfs_block_group_cache *group)
7014
7015 {
7016         __alloc_chunk_for_shrink(root, group, 1);
7017         set_block_group_readonly(group);
7018         return 0;
7019 }
7020
7021 /*
7022  * checks to see if its even possible to relocate this block group.
7023  *
7024  * @return - -1 if it's not a good idea to relocate this block group, 0 if its
7025  * ok to go ahead and try.
7026  */
7027 int btrfs_can_relocate(struct btrfs_root *root, u64 bytenr)
7028 {
7029         struct btrfs_block_group_cache *block_group;
7030         struct btrfs_space_info *space_info;
7031         struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
7032         struct btrfs_device *device;
7033         int full = 0;
7034         int ret = 0;
7035
7036         block_group = btrfs_lookup_block_group(root->fs_info, bytenr);
7037
7038         /* odd, couldn't find the block group, leave it alone */
7039         if (!block_group)
7040                 return -1;
7041
7042         /* no bytes used, we're good */
7043         if (!btrfs_block_group_used(&block_group->item))
7044                 goto out;
7045
7046         space_info = block_group->space_info;
7047         spin_lock(&space_info->lock);
7048
7049         full = space_info->full;
7050
7051         /*
7052          * if this is the last block group we have in this space, we can't
7053          * relocate it unless we're able to allocate a new chunk below.
7054          *
7055          * Otherwise, we need to make sure we have room in the space to handle
7056          * all of the extents from this block group.  If we can, we're good
7057          */
7058         if ((space_info->total_bytes != block_group->key.offset) &&
7059            (space_info->bytes_used + space_info->bytes_reserved +
7060             space_info->bytes_pinned + space_info->bytes_readonly +
7061             btrfs_block_group_used(&block_group->item) <
7062             space_info->total_bytes)) {
7063                 spin_unlock(&space_info->lock);
7064                 goto out;
7065         }
7066         spin_unlock(&space_info->lock);
7067
7068         /*
7069          * ok we don't have enough space, but maybe we have free space on our
7070          * devices to allocate new chunks for relocation, so loop through our
7071          * alloc devices and guess if we have enough space.  However, if we
7072          * were marked as full, then we know there aren't enough chunks, and we
7073          * can just return.
7074          */
7075         ret = -1;
7076         if (full)
7077                 goto out;
7078
7079         mutex_lock(&root->fs_info->chunk_mutex);
7080         list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
7081                 u64 min_free = btrfs_block_group_used(&block_group->item);
7082                 u64 dev_offset, max_avail;
7083
7084                 /*
7085                  * check to make sure we can actually find a chunk with enough
7086                  * space to fit our block group in.
7087                  */
7088                 if (device->total_bytes > device->bytes_used + min_free) {
7089                         ret = find_free_dev_extent(NULL, device, min_free,
7090                                                    &dev_offset, &max_avail);
7091                         if (!ret)
7092                                 break;
7093                         ret = -1;
7094                 }
7095         }
7096         mutex_unlock(&root->fs_info->chunk_mutex);
7097 out:
7098         btrfs_put_block_group(block_group);
7099         return ret;
7100 }
7101
7102 static int find_first_block_group(struct btrfs_root *root,
7103                 struct btrfs_path *path, struct btrfs_key *key)
7104 {
7105         int ret = 0;
7106         struct btrfs_key found_key;
7107         struct extent_buffer *leaf;
7108         int slot;
7109
7110         ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
7111         if (ret < 0)
7112                 goto out;
7113
7114         while (1) {
7115                 slot = path->slots[0];
7116                 leaf = path->nodes[0];
7117                 if (slot >= btrfs_header_nritems(leaf)) {
7118                         ret = btrfs_next_leaf(root, path);
7119                         if (ret == 0)
7120                                 continue;
7121                         if (ret < 0)
7122                                 goto out;
7123                         break;
7124                 }
7125                 btrfs_item_key_to_cpu(leaf, &found_key, slot);
7126
7127                 if (found_key.objectid >= key->objectid &&
7128                     found_key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) {
7129                         ret = 0;
7130                         goto out;
7131                 }
7132                 path->slots[0]++;
7133         }
7134         ret = -ENOENT;
7135 out:
7136         return ret;
7137 }
7138
7139 int btrfs_free_block_groups(struct btrfs_fs_info *info)
7140 {
7141         struct btrfs_block_group_cache *block_group;
7142         struct btrfs_space_info *space_info;
7143         struct btrfs_caching_control *caching_ctl;
7144         struct rb_node *n;
7145
7146         down_write(&info->extent_commit_sem);
7147         while (!list_empty(&info->caching_block_groups)) {
7148                 caching_ctl = list_entry(info->caching_block_groups.next,
7149                                          struct btrfs_caching_control, list);
7150                 list_del(&caching_ctl->list);
7151                 put_caching_control(caching_ctl);
7152         }
7153         up_write(&info->extent_commit_sem);
7154
7155         spin_lock(&info->block_group_cache_lock);
7156         while ((n = rb_last(&info->block_group_cache_tree)) != NULL) {
7157                 block_group = rb_entry(n, struct btrfs_block_group_cache,
7158                                        cache_node);
7159                 rb_erase(&block_group->cache_node,
7160                          &info->block_group_cache_tree);
7161                 spin_unlock(&info->block_group_cache_lock);
7162
7163                 down_write(&block_group->space_info->groups_sem);
7164                 list_del(&block_group->list);
7165                 up_write(&block_group->space_info->groups_sem);
7166
7167                 if (block_group->cached == BTRFS_CACHE_STARTED)
7168                         wait_block_group_cache_done(block_group);
7169
7170                 btrfs_remove_free_space_cache(block_group);
7171
7172                 WARN_ON(atomic_read(&block_group->count) != 1);
7173                 kfree(block_group);
7174
7175                 spin_lock(&info->block_group_cache_lock);
7176         }
7177         spin_unlock(&info->block_group_cache_lock);
7178
7179         /* now that all the block groups are freed, go through and
7180          * free all the space_info structs.  This is only called during
7181          * the final stages of unmount, and so we know nobody is
7182          * using them.  We call synchronize_rcu() once before we start,
7183          * just to be on the safe side.
7184          */
7185         synchronize_rcu();
7186
7187         while(!list_empty(&info->space_info)) {
7188                 space_info = list_entry(info->space_info.next,
7189                                         struct btrfs_space_info,
7190                                         list);
7191
7192                 list_del(&space_info->list);
7193                 kfree(space_info);
7194         }
7195         return 0;
7196 }
7197
7198 int btrfs_read_block_groups(struct btrfs_root *root)
7199 {
7200         struct btrfs_path *path;
7201         int ret;
7202         struct btrfs_block_group_cache *cache;
7203         struct btrfs_fs_info *info = root->fs_info;
7204         struct btrfs_space_info *space_info;
7205         struct btrfs_key key;
7206         struct btrfs_key found_key;
7207         struct extent_buffer *leaf;
7208
7209         root = info->extent_root;
7210         key.objectid = 0;
7211         key.offset = 0;
7212         btrfs_set_key_type(&key, BTRFS_BLOCK_GROUP_ITEM_KEY);
7213         path = btrfs_alloc_path();
7214         if (!path)
7215                 return -ENOMEM;
7216
7217         while (1) {
7218                 ret = find_first_block_group(root, path, &key);
7219                 if (ret > 0) {
7220                         ret = 0;
7221                         goto error;
7222                 }
7223                 if (ret != 0)
7224                         goto error;
7225
7226                 leaf = path->nodes[0];
7227                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
7228                 cache = kzalloc(sizeof(*cache), GFP_NOFS);
7229                 if (!cache) {
7230                         ret = -ENOMEM;
7231                         break;
7232                 }
7233
7234                 atomic_set(&cache->count, 1);
7235                 spin_lock_init(&cache->lock);
7236                 spin_lock_init(&cache->tree_lock);
7237                 cache->fs_info = info;
7238                 INIT_LIST_HEAD(&cache->list);
7239                 INIT_LIST_HEAD(&cache->cluster_list);
7240
7241                 /*
7242                  * we only want to have 32k of ram per block group for keeping
7243                  * track of free space, and if we pass 1/2 of that we want to
7244                  * start converting things over to using bitmaps
7245                  */
7246                 cache->extents_thresh = ((1024 * 32) / 2) /
7247                         sizeof(struct btrfs_free_space);
7248
7249                 read_extent_buffer(leaf, &cache->item,
7250                                    btrfs_item_ptr_offset(leaf, path->slots[0]),
7251                                    sizeof(cache->item));
7252                 memcpy(&cache->key, &found_key, sizeof(found_key));
7253
7254                 key.objectid = found_key.objectid + found_key.offset;
7255                 btrfs_release_path(root, path);
7256                 cache->flags = btrfs_block_group_flags(&cache->item);
7257                 cache->sectorsize = root->sectorsize;
7258
7259                 /*
7260                  * check for two cases, either we are full, and therefore
7261                  * don't need to bother with the caching work since we won't
7262                  * find any space, or we are empty, and we can just add all
7263                  * the space in and be done with it.  This saves us _alot_ of
7264                  * time, particularly in the full case.
7265                  */
7266                 if (found_key.offset == btrfs_block_group_used(&cache->item)) {
7267                         exclude_super_stripes(root, cache);
7268                         cache->last_byte_to_unpin = (u64)-1;
7269                         cache->cached = BTRFS_CACHE_FINISHED;
7270                         free_excluded_extents(root, cache);
7271                 } else if (btrfs_block_group_used(&cache->item) == 0) {
7272                         exclude_super_stripes(root, cache);
7273                         cache->last_byte_to_unpin = (u64)-1;
7274                         cache->cached = BTRFS_CACHE_FINISHED;
7275                         add_new_free_space(cache, root->fs_info,
7276                                            found_key.objectid,
7277                                            found_key.objectid +
7278                                            found_key.offset);
7279                         free_excluded_extents(root, cache);
7280                 }
7281
7282                 ret = update_space_info(info, cache->flags, found_key.offset,
7283                                         btrfs_block_group_used(&cache->item),
7284                                         &space_info);
7285                 BUG_ON(ret);
7286                 cache->space_info = space_info;
7287                 spin_lock(&cache->space_info->lock);
7288                 cache->space_info->bytes_super += cache->bytes_super;
7289                 spin_unlock(&cache->space_info->lock);
7290
7291                 down_write(&space_info->groups_sem);
7292                 list_add_tail(&cache->list, &space_info->block_groups);
7293                 up_write(&space_info->groups_sem);
7294
7295                 ret = btrfs_add_block_group_cache(root->fs_info, cache);
7296                 BUG_ON(ret);
7297
7298                 set_avail_alloc_bits(root->fs_info, cache->flags);
7299                 if (btrfs_chunk_readonly(root, cache->key.objectid))
7300                         set_block_group_readonly(cache);
7301         }
7302         ret = 0;
7303 error:
7304         btrfs_free_path(path);
7305         return ret;
7306 }
7307
7308 int btrfs_make_block_group(struct btrfs_trans_handle *trans,
7309                            struct btrfs_root *root, u64 bytes_used,
7310                            u64 type, u64 chunk_objectid, u64 chunk_offset,
7311                            u64 size)
7312 {
7313         int ret;
7314         struct btrfs_root *extent_root;
7315         struct btrfs_block_group_cache *cache;
7316
7317         extent_root = root->fs_info->extent_root;
7318
7319         root->fs_info->last_trans_log_full_commit = trans->transid;
7320
7321         cache = kzalloc(sizeof(*cache), GFP_NOFS);
7322         if (!cache)
7323                 return -ENOMEM;
7324
7325         cache->key.objectid = chunk_offset;
7326         cache->key.offset = size;
7327         cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
7328         cache->sectorsize = root->sectorsize;
7329
7330         /*
7331          * we only want to have 32k of ram per block group for keeping track
7332          * of free space, and if we pass 1/2 of that we want to start
7333          * converting things over to using bitmaps
7334          */
7335         cache->extents_thresh = ((1024 * 32) / 2) /
7336                 sizeof(struct btrfs_free_space);
7337         atomic_set(&cache->count, 1);
7338         spin_lock_init(&cache->lock);
7339         spin_lock_init(&cache->tree_lock);
7340         INIT_LIST_HEAD(&cache->list);
7341         INIT_LIST_HEAD(&cache->cluster_list);
7342
7343         btrfs_set_block_group_used(&cache->item, bytes_used);
7344         btrfs_set_block_group_chunk_objectid(&cache->item, chunk_objectid);
7345         cache->flags = type;
7346         btrfs_set_block_group_flags(&cache->item, type);
7347
7348         cache->last_byte_to_unpin = (u64)-1;
7349         cache->cached = BTRFS_CACHE_FINISHED;
7350         exclude_super_stripes(root, cache);
7351
7352         add_new_free_space(cache, root->fs_info, chunk_offset,
7353                            chunk_offset + size);
7354
7355         free_excluded_extents(root, cache);
7356
7357         ret = update_space_info(root->fs_info, cache->flags, size, bytes_used,
7358                                 &cache->space_info);
7359         BUG_ON(ret);
7360
7361         spin_lock(&cache->space_info->lock);
7362         cache->space_info->bytes_super += cache->bytes_super;
7363         spin_unlock(&cache->space_info->lock);
7364
7365         down_write(&cache->space_info->groups_sem);
7366         list_add_tail(&cache->list, &cache->space_info->block_groups);
7367         up_write(&cache->space_info->groups_sem);
7368
7369         ret = btrfs_add_block_group_cache(root->fs_info, cache);
7370         BUG_ON(ret);
7371
7372         ret = btrfs_insert_item(trans, extent_root, &cache->key, &cache->item,
7373                                 sizeof(cache->item));
7374         BUG_ON(ret);
7375
7376         set_avail_alloc_bits(extent_root->fs_info, type);
7377
7378         return 0;
7379 }
7380
7381 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
7382                              struct btrfs_root *root, u64 group_start)
7383 {
7384         struct btrfs_path *path;
7385         struct btrfs_block_group_cache *block_group;
7386         struct btrfs_free_cluster *cluster;
7387         struct btrfs_key key;
7388         int ret;
7389
7390         root = root->fs_info->extent_root;
7391
7392         block_group = btrfs_lookup_block_group(root->fs_info, group_start);
7393         BUG_ON(!block_group);
7394         BUG_ON(!block_group->ro);
7395
7396         memcpy(&key, &block_group->key, sizeof(key));
7397
7398         /* make sure this block group isn't part of an allocation cluster */
7399         cluster = &root->fs_info->data_alloc_cluster;
7400         spin_lock(&cluster->refill_lock);
7401         btrfs_return_cluster_to_free_space(block_group, cluster);
7402         spin_unlock(&cluster->refill_lock);
7403
7404         /*
7405          * make sure this block group isn't part of a metadata
7406          * allocation cluster
7407          */
7408         cluster = &root->fs_info->meta_alloc_cluster;
7409         spin_lock(&cluster->refill_lock);
7410         btrfs_return_cluster_to_free_space(block_group, cluster);
7411         spin_unlock(&cluster->refill_lock);
7412
7413         path = btrfs_alloc_path();
7414         BUG_ON(!path);
7415
7416         spin_lock(&root->fs_info->block_group_cache_lock);
7417         rb_erase(&block_group->cache_node,
7418                  &root->fs_info->block_group_cache_tree);
7419         spin_unlock(&root->fs_info->block_group_cache_lock);
7420
7421         down_write(&block_group->space_info->groups_sem);
7422         /*
7423          * we must use list_del_init so people can check to see if they
7424          * are still on the list after taking the semaphore
7425          */
7426         list_del_init(&block_group->list);
7427         up_write(&block_group->space_info->groups_sem);
7428
7429         if (block_group->cached == BTRFS_CACHE_STARTED)
7430                 wait_block_group_cache_done(block_group);
7431
7432         btrfs_remove_free_space_cache(block_group);
7433
7434         spin_lock(&block_group->space_info->lock);
7435         block_group->space_info->total_bytes -= block_group->key.offset;
7436         block_group->space_info->bytes_readonly -= block_group->key.offset;
7437         spin_unlock(&block_group->space_info->lock);
7438
7439         btrfs_clear_space_info_full(root->fs_info);
7440
7441         btrfs_put_block_group(block_group);
7442         btrfs_put_block_group(block_group);
7443
7444         ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
7445         if (ret > 0)
7446                 ret = -EIO;
7447         if (ret < 0)
7448                 goto out;
7449
7450         ret = btrfs_del_item(trans, root, path);
7451 out:
7452         btrfs_free_path(path);
7453         return ret;
7454 }