SAFE public projects git trees. - safe/jmp/linux-2.6/blob - 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 "hash.h"
  23 #include "crc32c.h"
  24 #include "ctree.h"
  25 #include "disk-io.h"
  26 #include "print-tree.h"
  27 #include "transaction.h"
  28 #include "volumes.h"
  29 #include "locking.h"
  30 #include "ref-cache.h"
  31
  32 #define PENDING_EXTENT_INSERT 0
  33 #define PENDING_EXTENT_DELETE 1
  34 #define PENDING_BACKREF_UPDATE 2
  35
  36 struct pending_extent_op {
  37         int type;
  38         u64 bytenr;
  39         u64 num_bytes;
  40         u64 parent;
  41         u64 orig_parent;
  42         u64 generation;
  43         u64 orig_generation;
  44         int level;
  45         struct list_head list;
  46         int del;
  47 };
  48
  49 static int finish_current_insert(struct btrfs_trans_handle *trans, struct
  50                                  btrfs_root *extent_root, int all);
  51 static int del_pending_extents(struct btrfs_trans_handle *trans, struct
  52                                btrfs_root *extent_root, int all);
  53 static struct btrfs_block_group_cache *
  54 __btrfs_find_block_group(struct btrfs_root *root,
  55                          struct btrfs_block_group_cache *hint,
  56                          u64 search_start, int data, int owner);
  57 static int pin_down_bytes(struct btrfs_trans_handle *trans,
  58                           struct btrfs_root *root,
  59                           u64 bytenr, u64 num_bytes, int is_data);
  60 static int update_block_group(struct btrfs_trans_handle *trans,
  61                               struct btrfs_root *root,
  62                               u64 bytenr, u64 num_bytes, int alloc,
  63                               int mark_free);
  64
  65 static int block_group_bits(struct btrfs_block_group_cache *cache, u64 bits)
  66 {
  67         return (cache->flags & bits) == bits;
  68 }
  69
  70 /*
  71  * this adds the block group to the fs_info rb tree for the block group
  72  * cache
  73  */
  74 int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
  75                                 struct btrfs_block_group_cache *block_group)
  76 {
  77         struct rb_node **p;
  78         struct rb_node *parent = NULL;
  79         struct btrfs_block_group_cache *cache;
  80
  81         spin_lock(&info->block_group_cache_lock);
  82         p = &info->block_group_cache_tree.rb_node;
  83
  84         while (*p) {
  85                 parent = *p;
  86                 cache = rb_entry(parent, struct btrfs_block_group_cache,
  87                                  cache_node);
  88                 if (block_group->key.objectid < cache->key.objectid) {
  89                         p = &(*p)->rb_left;
  90                 } else if (block_group->key.objectid > cache->key.objectid) {
  91                         p = &(*p)->rb_right;
  92                 } else {
  93                         spin_unlock(&info->block_group_cache_lock);
  94                         return -EEXIST;
  95                 }
  96         }
  97
  98         rb_link_node(&block_group->cache_node, parent, p);
  99         rb_insert_color(&block_group->cache_node,
 100                         &info->block_group_cache_tree);
 101         spin_unlock(&info->block_group_cache_lock);
 102
 103         return 0;
 104 }
 105
 106 /*
 107  * This will return the block group at or after bytenr if contains is 0, else
 108  * it will return the block group that contains the bytenr
 109  */
 110 static struct btrfs_block_group_cache *
 111 block_group_cache_tree_search(struct btrfs_fs_info *info, u64 bytenr,
 112                               int contains)
 113 {
 114         struct btrfs_block_group_cache *cache, *ret = NULL;
 115         struct rb_node *n;
 116         u64 end, start;
 117
 118         spin_lock(&info->block_group_cache_lock);
 119         n = info->block_group_cache_tree.rb_node;
 120
 121         while (n) {
 122                 cache = rb_entry(n, struct btrfs_block_group_cache,
 123                                  cache_node);
 124                 end = cache->key.objectid + cache->key.offset - 1;
 125                 start = cache->key.objectid;
 126
 127                 if (bytenr < start) {
 128                         if (!contains && (!ret || start < ret->key.objectid))
 129                                 ret = cache;
 130                         n = n->rb_left;
 131                 } else if (bytenr > start) {
 132                         if (contains && bytenr <= end) {
 133                                 ret = cache;
 134                                 break;
 135                         }
 136                         n = n->rb_right;
 137                 } else {
 138                         ret = cache;
 139                         break;
 140                 }
 141         }
 142         spin_unlock(&info->block_group_cache_lock);
 143
 144         return ret;
 145 }
 146
 147 /*
 148  * this is only called by cache_block_group, since we could have freed extents
 149  * we need to check the pinned_extents for any extents that can't be used yet
 150  * since their free space will be released as soon as the transaction commits.
 151  */
 152 static int add_new_free_space(struct btrfs_block_group_cache *block_group,
 153                               struct btrfs_fs_info *info, u64 start, u64 end)
 154 {
 155         u64 extent_start, extent_end, size;
 156         int ret;
 157
 158         mutex_lock(&info->pinned_mutex);
 159         while (start < end) {
 160                 ret = find_first_extent_bit(&info->pinned_extents, start,
 161                                             &extent_start, &extent_end,
 162                                             EXTENT_DIRTY);
 163                 if (ret)
 164                         break;
 165
 166                 if (extent_start == start) {
 167                         start = extent_end + 1;
 168                 } else if (extent_start > start && extent_start < end) {
 169                         size = extent_start - start;
 170                         ret = btrfs_add_free_space_lock(block_group, start,
 171                                                         size);
 172                         BUG_ON(ret);
 173                         start = extent_end + 1;
 174                 } else {
 175                         break;
 176                 }
 177         }
 178
 179         if (start < end) {
 180                 size = end - start;
 181                 ret = btrfs_add_free_space_lock(block_group, start, size);
 182                 BUG_ON(ret);
 183         }
 184         mutex_unlock(&info->pinned_mutex);
 185
 186         return 0;
 187 }
 188
 189 static int cache_block_group(struct btrfs_root *root,
 190                              struct btrfs_block_group_cache *block_group)
 191 {
 192         struct btrfs_path *path;
 193         int ret = 0;
 194         struct btrfs_key key;
 195         struct extent_buffer *leaf;
 196         int slot;
 197         u64 last = 0;
 198         u64 first_free;
 199         int found = 0;
 200
 201         if (!block_group)
 202                 return 0;
 203
 204         root = root->fs_info->extent_root;
 205
 206         if (block_group->cached)
 207                 return 0;
 208
 209         path = btrfs_alloc_path();
 210         if (!path)
 211                 return -ENOMEM;
 212
 213         path->reada = 2;
 214         /*
 215          * we get into deadlocks with paths held by callers of this function.
 216          * since the alloc_mutex is protecting things right now, just
 217          * skip the locking here
 218          */
 219         path->skip_locking = 1;
 220         first_free = max_t(u64, block_group->key.objectid,
 221                            BTRFS_SUPER_INFO_OFFSET + BTRFS_SUPER_INFO_SIZE);
 222         key.objectid = block_group->key.objectid;
 223         key.offset = 0;
 224         btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
 225         ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 226         if (ret < 0)
 227                 goto err;
 228         ret = btrfs_previous_item(root, path, 0, BTRFS_EXTENT_ITEM_KEY);
 229         if (ret < 0)
 230                 goto err;
 231         if (ret == 0) {
 232                 leaf = path->nodes[0];
 233                 btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 234                 if (key.objectid + key.offset > first_free)
 235                         first_free = key.objectid + key.offset;
 236         }
 237         while(1) {
 238                 leaf = path->nodes[0];
 239                 slot = path->slots[0];
 240                 if (slot >= btrfs_header_nritems(leaf)) {
 241                         ret = btrfs_next_leaf(root, path);
 242                         if (ret < 0)
 243                                 goto err;
 244                         if (ret == 0)
 245                                 continue;
 246                         else
 247                                 break;
 248                 }
 249                 btrfs_item_key_to_cpu(leaf, &key, slot);
 250                 if (key.objectid < block_group->key.objectid)
 251                         goto next;
 252
 253                 if (key.objectid >= block_group->key.objectid +
 254                     block_group->key.offset)
 255                         break;
 256
 257                 if (btrfs_key_type(&key) == BTRFS_EXTENT_ITEM_KEY) {
 258                         if (!found) {
 259                                 last = first_free;
 260                                 found = 1;
 261                         }
 262
 263                         add_new_free_space(block_group, root->fs_info, last,
 264                                            key.objectid);
 265
 266                         last = key.objectid + key.offset;
 267                 }
 268 next:
 269                 path->slots[0]++;
 270         }
 271
 272         if (!found)
 273                 last = first_free;
 274
 275         add_new_free_space(block_group, root->fs_info, last,
 276                            block_group->key.objectid +
 277                            block_group->key.offset);
 278
 279         block_group->cached = 1;
 280         ret = 0;
 281 err:
 282         btrfs_free_path(path);
 283         return ret;
 284 }
 285
 286 /*
 287  * return the block group that starts at or after bytenr
 288  */
 289 struct btrfs_block_group_cache *btrfs_lookup_first_block_group(struct
 290                                                        btrfs_fs_info *info,
 291                                                          u64 bytenr)
 292 {
 293         struct btrfs_block_group_cache *cache;
 294
 295         cache = block_group_cache_tree_search(info, bytenr, 0);
 296
 297         return cache;
 298 }
 299
 300 /*
 301  * return the block group that contains teh given bytenr
 302  */
 303 struct btrfs_block_group_cache *btrfs_lookup_block_group(struct
 304                                                          btrfs_fs_info *info,
 305                                                          u64 bytenr)
 306 {
 307         struct btrfs_block_group_cache *cache;
 308
 309         cache = block_group_cache_tree_search(info, bytenr, 1);
 310
 311         return cache;
 312 }
 313
 314 static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 315                                                   u64 flags)
 316 {
 317         struct list_head *head = &info->space_info;
 318         struct list_head *cur;
 319         struct btrfs_space_info *found;
 320         list_for_each(cur, head) {
 321                 found = list_entry(cur, struct btrfs_space_info, list);
 322                 if (found->flags == flags)
 323                         return found;
 324         }
 325         return NULL;
 326 }
 327
 328 static u64 div_factor(u64 num, int factor)
 329 {
 330         if (factor == 10)
 331                 return num;
 332         num *= factor;
 333         do_div(num, 10);
 334         return num;
 335 }
 336
 337 static struct btrfs_block_group_cache *
 338 __btrfs_find_block_group(struct btrfs_root *root,
 339                          struct btrfs_block_group_cache *hint,
 340                          u64 search_start, int data, int owner)
 341 {
 342         struct btrfs_block_group_cache *cache;
 343         struct btrfs_block_group_cache *found_group = NULL;
 344         struct btrfs_fs_info *info = root->fs_info;
 345         u64 used;
 346         u64 last = 0;
 347         u64 free_check;
 348         int full_search = 0;
 349         int factor = 10;
 350         int wrapped = 0;
 351
 352         if (data & BTRFS_BLOCK_GROUP_METADATA)
 353                 factor = 9;
 354
 355         if (search_start) {
 356                 struct btrfs_block_group_cache *shint;
 357                 shint = btrfs_lookup_first_block_group(info, search_start);
 358                 if (shint && block_group_bits(shint, data)) {
 359                         spin_lock(&shint->lock);
 360                         used = btrfs_block_group_used(&shint->item);
 361                         if (used + shint->pinned + shint->reserved <
 362                             div_factor(shint->key.offset, factor)) {
 363                                 spin_unlock(&shint->lock);
 364                                 return shint;
 365                         }
 366                         spin_unlock(&shint->lock);
 367                 }
 368         }
 369         if (hint && block_group_bits(hint, data)) {
 370                 spin_lock(&hint->lock);
 371                 used = btrfs_block_group_used(&hint->item);
 372                 if (used + hint->pinned + hint->reserved <
 373                     div_factor(hint->key.offset, factor)) {
 374                         spin_unlock(&hint->lock);
 375                         return hint;
 376                 }
 377                 spin_unlock(&hint->lock);
 378                 last = hint->key.objectid + hint->key.offset;
 379         } else {
 380                 if (hint)
 381                         last = max(hint->key.objectid, search_start);
 382                 else
 383                         last = search_start;
 384         }
 385 again:
 386         while (1) {
 387                 cache = btrfs_lookup_first_block_group(root->fs_info, last);
 388                 if (!cache)
 389                         break;
 390
 391                 spin_lock(&cache->lock);
 392                 last = cache->key.objectid + cache->key.offset;
 393                 used = btrfs_block_group_used(&cache->item);
 394
 395                 if (block_group_bits(cache, data)) {
 396                         free_check = div_factor(cache->key.offset, factor);
 397                         if (used + cache->pinned + cache->reserved <
 398                             free_check) {
 399                                 found_group = cache;
 400                                 spin_unlock(&cache->lock);
 401                                 goto found;
 402                         }
 403                 }
 404                 spin_unlock(&cache->lock);
 405                 cond_resched();
 406         }
 407         if (!wrapped) {
 408                 last = search_start;
 409                 wrapped = 1;
 410                 goto again;
 411         }
 412         if (!full_search && factor < 10) {
 413                 last = search_start;
 414                 full_search = 1;
 415                 factor = 10;
 416                 goto again;
 417         }
 418 found:
 419         return found_group;
 420 }
 421
 422 struct btrfs_block_group_cache *btrfs_find_block_group(struct btrfs_root *root,
 423                                                  struct btrfs_block_group_cache
 424                                                  *hint, u64 search_start,
 425                                                  int data, int owner)
 426 {
 427
 428         struct btrfs_block_group_cache *ret;
 429         ret = __btrfs_find_block_group(root, hint, search_start, data, owner);
 430         return ret;
 431 }
 432
 433 /* simple helper to search for an existing extent at a given offset */
 434 int btrfs_lookup_extent(struct btrfs_root *root, u64 start, u64 len)
 435 {
 436         int ret;
 437         struct btrfs_key key;
 438         struct btrfs_path *path;
 439
 440         path = btrfs_alloc_path();
 441         BUG_ON(!path);
 442         key.objectid = start;
 443         key.offset = len;
 444         btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
 445         ret = btrfs_search_slot(NULL, root->fs_info->extent_root, &key, path,
 446                                 0, 0);
 447         btrfs_free_path(path);
 448         return ret;
 449 }
 450
 451 /*
 452  * Back reference rules.  Back refs have three main goals:
 453  *
 454  * 1) differentiate between all holders of references to an extent so that
 455  *    when a reference is dropped we can make sure it was a valid reference
 456  *    before freeing the extent.
 457  *
 458  * 2) Provide enough information to quickly find the holders of an extent
 459  *    if we notice a given block is corrupted or bad.
 460  *
 461  * 3) Make it easy to migrate blocks for FS shrinking or storage pool
 462  *    maintenance.  This is actually the same as #2, but with a slightly
 463  *    different use case.
 464  *
 465  * File extents can be referenced by:
 466  *
 467  * - multiple snapshots, subvolumes, or different generations in one subvol
 468  * - different files inside a single subvolume
 469  * - different offsets inside a file (bookend extents in file.c)
 470  *
 471  * The extent ref structure has fields for:
 472  *
 473  * - Objectid of the subvolume root
 474  * - Generation number of the tree holding the reference
 475  * - objectid of the file holding the reference
 476  * - number of references holding by parent node (alway 1 for tree blocks)
 477  *
 478  * Btree leaf may hold multiple references to a file extent. In most cases,
 479  * these references are from same file and the corresponding offsets inside
 480  * the file are close together.
 481  *
 482  * When a file extent is allocated the fields are filled in:
 483  *     (root_key.objectid, trans->transid, inode objectid, 1)
 484  *
 485  * When a leaf is cow'd new references are added for every file extent found
 486  * in the leaf.  It looks similar to the create case, but trans->transid will
 487  * be different when the block is cow'd.
 488  *
 489  *     (root_key.objectid, trans->transid, inode objectid,
 490  *      number of references in the leaf)
 491  *
 492  * When a file extent is removed either during snapshot deletion or
 493  * file truncation, we find the corresponding back reference and check
 494  * the following fields:
 495  *
 496  *     (btrfs_header_owner(leaf), btrfs_header_generation(leaf),
 497  *      inode objectid)
 498  *
 499  * Btree extents can be referenced by:
 500  *
 501  * - Different subvolumes
 502  * - Different generations of the same subvolume
 503  *
 504  * When a tree block is created, back references are inserted:
 505  *
 506  * (root->root_key.objectid, trans->transid, level, 1)
 507  *
 508  * When a tree block is cow'd, new back references are added for all the
 509  * blocks it points to. If the tree block isn't in reference counted root,
 510  * the old back references are removed. These new back references are of
 511  * the form (trans->transid will have increased since creation):
 512  *
 513  * (root->root_key.objectid, trans->transid, level, 1)
 514  *
 515  * When a backref is in deleting, the following fields are checked:
 516  *
 517  * if backref was for a tree root:
 518  *     (btrfs_header_owner(itself), btrfs_header_generation(itself), level)
 519  * else
 520  *     (btrfs_header_owner(parent), btrfs_header_generation(parent), level)
 521  *
 522  * Back Reference Key composing:
 523  *
 524  * The key objectid corresponds to the first byte in the extent, the key
 525  * type is set to BTRFS_EXTENT_REF_KEY, and the key offset is the first
 526  * byte of parent extent. If a extent is tree root, the key offset is set
 527  * to the key objectid.
 528  */
 529
 530 static int noinline lookup_extent_backref(struct btrfs_trans_handle *trans,
 531                                           struct btrfs_root *root,
 532                                           struct btrfs_path *path,
 533                                           u64 bytenr, u64 parent,
 534                                           u64 ref_root, u64 ref_generation,
 535                                           u64 owner_objectid, int del)
 536 {
 537         struct btrfs_key key;
 538         struct btrfs_extent_ref *ref;
 539         struct extent_buffer *leaf;
 540         u64 ref_objectid;
 541         int ret;
 542
 543         key.objectid = bytenr;
 544         key.type = BTRFS_EXTENT_REF_KEY;
 545         key.offset = parent;
 546
 547         ret = btrfs_search_slot(trans, root, &key, path, del ? -1 : 0, 1);
 548         if (ret < 0)
 549                 goto out;
 550         if (ret > 0) {
 551                 ret = -ENOENT;
 552                 goto out;
 553         }
 554
 555         leaf = path->nodes[0];
 556         ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_ref);
 557         ref_objectid = btrfs_ref_objectid(leaf, ref);
 558         if (btrfs_ref_root(leaf, ref) != ref_root ||
 559             btrfs_ref_generation(leaf, ref) != ref_generation ||
 560             (ref_objectid != owner_objectid &&
 561              ref_objectid != BTRFS_MULTIPLE_OBJECTIDS)) {
 562                 ret = -EIO;
 563                 WARN_ON(1);
 564                 goto out;
 565         }
 566         ret = 0;
 567 out:
 568         return ret;
 569 }
 570
 571 /*
 572  * updates all the backrefs that are pending on update_list for the
 573  * extent_root
 574  */
 575 static int noinline update_backrefs(struct btrfs_trans_handle *trans,
 576                                     struct btrfs_root *extent_root,
 577                                     struct btrfs_path *path,
 578                                     struct list_head *update_list)
 579 {
 580         struct btrfs_key key;
 581         struct btrfs_extent_ref *ref;
 582         struct btrfs_fs_info *info = extent_root->fs_info;
 583         struct pending_extent_op *op;
 584         struct extent_buffer *leaf;
 585         int ret = 0;
 586         struct list_head *cur = update_list->next;
 587         u64 ref_objectid;
 588         u64 ref_root = extent_root->root_key.objectid;
 589
 590         op = list_entry(cur, struct pending_extent_op, list);
 591
 592 search:
 593         key.objectid = op->bytenr;
 594         key.type = BTRFS_EXTENT_REF_KEY;
 595         key.offset = op->orig_parent;
 596
 597         ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 1);
 598         BUG_ON(ret);
 599
 600         leaf = path->nodes[0];
 601
 602 loop:
 603         ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_ref);
 604
 605         ref_objectid = btrfs_ref_objectid(leaf, ref);
 606
 607         if (btrfs_ref_root(leaf, ref) != ref_root ||
 608             btrfs_ref_generation(leaf, ref) != op->orig_generation ||
 609             (ref_objectid != op->level &&
 610              ref_objectid != BTRFS_MULTIPLE_OBJECTIDS)) {
 611                 printk(KERN_ERR "couldn't find %Lu, parent %Lu, root %Lu, "
 612                        "owner %u\n", op->bytenr, op->orig_parent,
 613                        ref_root, op->level);
 614                 btrfs_print_leaf(extent_root, leaf);
 615                 BUG();
 616         }
 617
 618         key.objectid = op->bytenr;
 619         key.offset = op->parent;
 620         key.type = BTRFS_EXTENT_REF_KEY;
 621         ret = btrfs_set_item_key_safe(trans, extent_root, path, &key);
 622         BUG_ON(ret);
 623         ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_ref);
 624         btrfs_set_ref_generation(leaf, ref, op->generation);
 625
 626         cur = cur->next;
 627
 628         list_del_init(&op->list);
 629         unlock_extent(&info->extent_ins, op->bytenr,
 630                       op->bytenr + op->num_bytes - 1, GFP_NOFS);
 631         kfree(op);
 632
 633         if (cur == update_list) {
 634                 btrfs_mark_buffer_dirty(path->nodes[0]);
 635                 btrfs_release_path(extent_root, path);
 636                 goto out;
 637         }
 638
 639         op = list_entry(cur, struct pending_extent_op, list);
 640
 641         path->slots[0]++;
 642         while (path->slots[0] < btrfs_header_nritems(leaf)) {
 643                 btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 644                 if (key.objectid == op->bytenr &&
 645                     key.type == BTRFS_EXTENT_REF_KEY)
 646                         goto loop;
 647                 path->slots[0]++;
 648         }
 649
 650         btrfs_mark_buffer_dirty(path->nodes[0]);
 651         btrfs_release_path(extent_root, path);
 652         goto search;
 653
 654 out:
 655         return 0;
 656 }
 657
 658 static int noinline insert_extents(struct btrfs_trans_handle *trans,
 659                                    struct btrfs_root *extent_root,
 660                                    struct btrfs_path *path,
 661                                    struct list_head *insert_list, int nr)
 662 {
 663         struct btrfs_key *keys;
 664         u32 *data_size;
 665         struct pending_extent_op *op;
 666         struct extent_buffer *leaf;
 667         struct list_head *cur = insert_list->next;
 668         struct btrfs_fs_info *info = extent_root->fs_info;
 669         u64 ref_root = extent_root->root_key.objectid;
 670         int i = 0, last = 0, ret;
 671         int total = nr * 2;
 672
 673         if (!nr)
 674                 return 0;
 675
 676         keys = kzalloc(total * sizeof(struct btrfs_key), GFP_NOFS);
 677         if (!keys)
 678                 return -ENOMEM;
 679
 680         data_size = kzalloc(total * sizeof(u32), GFP_NOFS);
 681         if (!data_size) {
 682                 kfree(keys);
 683                 return -ENOMEM;
 684         }
 685
 686         list_for_each_entry(op, insert_list, list) {
 687                 keys[i].objectid = op->bytenr;
 688                 keys[i].offset = op->num_bytes;
 689                 keys[i].type = BTRFS_EXTENT_ITEM_KEY;
 690                 data_size[i] = sizeof(struct btrfs_extent_item);
 691                 i++;
 692
 693                 keys[i].objectid = op->bytenr;
 694                 keys[i].offset = op->parent;
 695                 keys[i].type = BTRFS_EXTENT_REF_KEY;
 696                 data_size[i] = sizeof(struct btrfs_extent_ref);
 697                 i++;
 698         }
 699
 700         op = list_entry(cur, struct pending_extent_op, list);
 701         i = 0;
 702         while (i < total) {
 703                 int c;
 704                 ret = btrfs_insert_some_items(trans, extent_root, path,
 705                                               keys+i, data_size+i, total-i);
 706                 BUG_ON(ret < 0);
 707
 708                 if (last && ret > 1)
 709                         BUG();
 710
 711                 leaf = path->nodes[0];
 712                 for (c = 0; c < ret; c++) {
 713                         int ref_first = keys[i].type == BTRFS_EXTENT_REF_KEY;
 714
 715                         /*
 716                          * if the first item we inserted was a backref, then
 717                          * the EXTENT_ITEM will be the odd c's, else it will
 718                          * be the even c's
 719                          */
 720                         if ((ref_first && (c % 2)) ||
 721                             (!ref_first && !(c % 2))) {
 722                                 struct btrfs_extent_item *itm;
 723
 724                                 itm = btrfs_item_ptr(leaf, path->slots[0] + c,
 725                                                      struct btrfs_extent_item);
 726                                 btrfs_set_extent_refs(path->nodes[0], itm, 1);
 727                                 op->del++;
 728                         } else {
 729                                 struct btrfs_extent_ref *ref;
 730
 731                                 ref = btrfs_item_ptr(leaf, path->slots[0] + c,
 732                                                      struct btrfs_extent_ref);
 733                                 btrfs_set_ref_root(leaf, ref, ref_root);
 734                                 btrfs_set_ref_generation(leaf, ref,
 735                                                          op->generation);
 736                                 btrfs_set_ref_objectid(leaf, ref, op->level);
 737                                 btrfs_set_ref_num_refs(leaf, ref, 1);
 738                                 op->del++;
 739                         }
 740
 741                         /*
 742                          * using del to see when its ok to free up the
 743                          * pending_extent_op.  In the case where we insert the
 744                          * last item on the list in order to help do batching
 745                          * we need to not free the extent op until we actually
 746                          * insert the extent_item
 747                          */
 748                         if (op->del == 2) {
 749                                 unlock_extent(&info->extent_ins, op->bytenr,
 750                                               op->bytenr + op->num_bytes - 1,
 751                                               GFP_NOFS);
 752                                 cur = cur->next;
 753                                 list_del_init(&op->list);
 754                                 kfree(op);
 755                                 if (cur != insert_list)
 756                                         op = list_entry(cur,
 757                                                 struct pending_extent_op,
 758                                                 list);
 759                         }
 760                 }
 761                 btrfs_mark_buffer_dirty(leaf);
 762                 btrfs_release_path(extent_root, path);
 763
 764                 /*
 765                  * Ok backref's and items usually go right next to eachother,
 766                  * but if we could only insert 1 item that means that we
 767                  * inserted on the end of a leaf, and we have no idea what may
 768                  * be on the next leaf so we just play it safe.  In order to
 769                  * try and help this case we insert the last thing on our
 770                  * insert list so hopefully it will end up being the last
 771                  * thing on the leaf and everything else will be before it,
 772                  * which will let us insert a whole bunch of items at the same
 773                  * time.
 774                  */
 775                 if (ret == 1 && !last && (i + ret < total)) {
 776                         /*
 777                          * last: where we will pick up the next time around
 778                          * i: our current key to insert, will be total - 1
 779                          * cur: the current op we are screwing with
 780                          * op: duh
 781                          */
 782                         last = i + ret;
 783                         i = total - 1;
 784                         cur = insert_list->prev;
 785                         op = list_entry(cur, struct pending_extent_op, list);
 786                 } else if (last) {
 787                         /*
 788                          * ok we successfully inserted the last item on the
 789                          * list, lets reset everything
 790                          *
 791                          * i: our current key to insert, so where we left off
 792                          *    last time
 793                          * last: done with this
 794                          * cur: the op we are messing with
 795                          * op: duh
 796                          * total: since we inserted the last key, we need to
 797                          *        decrement total so we dont overflow
 798                          */
 799                         i = last;
 800                         last = 0;
 801                         total--;
 802                         if (i < total) {
 803                                 cur = insert_list->next;
 804                                 op = list_entry(cur, struct pending_extent_op,
 805                                                 list);
 806                         }
 807                 } else {
 808                         i += ret;
 809                 }
 810
 811                 cond_resched();
 812         }
 813         ret = 0;
 814         kfree(keys);
 815         kfree(data_size);
 816         return ret;
 817 }
 818
 819 static int noinline insert_extent_backref(struct btrfs_trans_handle *trans,
 820                                           struct btrfs_root *root,
 821                                           struct btrfs_path *path,
 822                                           u64 bytenr, u64 parent,
 823                                           u64 ref_root, u64 ref_generation,
 824                                           u64 owner_objectid)
 825 {
 826         struct btrfs_key key;
 827         struct extent_buffer *leaf;
 828         struct btrfs_extent_ref *ref;
 829         u32 num_refs;
 830         int ret;
 831
 832         key.objectid = bytenr;
 833         key.type = BTRFS_EXTENT_REF_KEY;
 834         key.offset = parent;
 835
 836         ret = btrfs_insert_empty_item(trans, root, path, &key, sizeof(*ref));
 837         if (ret == 0) {
 838                 leaf = path->nodes[0];
 839                 ref = btrfs_item_ptr(leaf, path->slots[0],
 840                                      struct btrfs_extent_ref);
 841                 btrfs_set_ref_root(leaf, ref, ref_root);
 842                 btrfs_set_ref_generation(leaf, ref, ref_generation);
 843                 btrfs_set_ref_objectid(leaf, ref, owner_objectid);
 844                 btrfs_set_ref_num_refs(leaf, ref, 1);
 845         } else if (ret == -EEXIST) {
 846                 u64 existing_owner;
 847                 BUG_ON(owner_objectid < BTRFS_FIRST_FREE_OBJECTID);
 848                 leaf = path->nodes[0];
 849                 ref = btrfs_item_ptr(leaf, path->slots[0],
 850                                      struct btrfs_extent_ref);
 851                 if (btrfs_ref_root(leaf, ref) != ref_root ||
 852                     btrfs_ref_generation(leaf, ref) != ref_generation) {
 853                         ret = -EIO;
 854                         WARN_ON(1);
 855                         goto out;
 856                 }
 857
 858                 num_refs = btrfs_ref_num_refs(leaf, ref);
 859                 BUG_ON(num_refs == 0);
 860                 btrfs_set_ref_num_refs(leaf, ref, num_refs + 1);
 861
 862                 existing_owner = btrfs_ref_objectid(leaf, ref);
 863                 if (existing_owner != owner_objectid &&
 864                     existing_owner != BTRFS_MULTIPLE_OBJECTIDS) {
 865                         btrfs_set_ref_objectid(leaf, ref,
 866                                         BTRFS_MULTIPLE_OBJECTIDS);
 867                 }
 868                 ret = 0;
 869         } else {
 870                 goto out;
 871         }
 872         btrfs_mark_buffer_dirty(path->nodes[0]);
 873 out:
 874         btrfs_release_path(root, path);
 875         return ret;
 876 }
 877
 878 static int noinline remove_extent_backref(struct btrfs_trans_handle *trans,
 879                                           struct btrfs_root *root,
 880                                           struct btrfs_path *path)
 881 {
 882         struct extent_buffer *leaf;
 883         struct btrfs_extent_ref *ref;
 884         u32 num_refs;
 885         int ret = 0;
 886
 887         leaf = path->nodes[0];
 888         ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_ref);
 889         num_refs = btrfs_ref_num_refs(leaf, ref);
 890         BUG_ON(num_refs == 0);
 891         num_refs -= 1;
 892         if (num_refs == 0) {
 893                 ret = btrfs_del_item(trans, root, path);
 894         } else {
 895                 btrfs_set_ref_num_refs(leaf, ref, num_refs);
 896                 btrfs_mark_buffer_dirty(leaf);
 897         }
 898         btrfs_release_path(root, path);
 899         return ret;
 900 }
 901
 902 static int noinline free_extents(struct btrfs_trans_handle *trans,
 903                                  struct btrfs_root *extent_root,
 904                                  struct list_head *del_list)
 905 {
 906         struct btrfs_fs_info *info = extent_root->fs_info;
 907         struct btrfs_path *path;
 908         struct btrfs_key key, found_key;
 909         struct extent_buffer *leaf;
 910         struct list_head *cur;
 911         struct pending_extent_op *op;
 912         struct btrfs_extent_item *ei;
 913         int ret, num_to_del, extent_slot = 0, found_extent = 0;
 914         u32 refs;
 915         u64 bytes_freed = 0;
 916
 917         path = btrfs_alloc_path();
 918         if (!path)
 919                 return -ENOMEM;
 920         path->reada = 1;
 921
 922 search:
 923         /* search for the backref for the current ref we want to delete */
 924         cur = del_list->next;
 925         op = list_entry(cur, struct pending_extent_op, list);
 926         ret = lookup_extent_backref(trans, extent_root, path, op->bytenr,
 927                                     op->orig_parent,
 928                                     extent_root->root_key.objectid,
 929                                     op->orig_generation, op->level, 1);
 930         if (ret) {
 931                 printk("Unable to find backref byte nr %Lu root %Lu gen %Lu "
 932                        "owner %u\n", op->bytenr,
 933                        extent_root->root_key.objectid, op->orig_generation,
 934                        op->level);
 935                 btrfs_print_leaf(extent_root, path->nodes[0]);
 936                 WARN_ON(1);
 937                 goto out;
 938         }
 939
 940         extent_slot = path->slots[0];
 941         num_to_del = 1;
 942         found_extent = 0;
 943
 944         /*
 945          * if we aren't the first item on the leaf we can move back one and see
 946          * if our ref is right next to our extent item
 947          */
 948         if (likely(extent_slot)) {
 949                 extent_slot--;
 950                 btrfs_item_key_to_cpu(path->nodes[0], &found_key,
 951                                       extent_slot);
 952                 if (found_key.objectid == op->bytenr &&
 953                     found_key.type == BTRFS_EXTENT_ITEM_KEY &&
 954                     found_key.offset == op->num_bytes) {
 955                         num_to_del++;
 956                         found_extent = 1;
 957                 }
 958         }
 959
 960         /*
 961          * if we didn't find the extent we need to delete the backref and then
 962          * search for the extent item key so we can update its ref count
 963          */
 964         if (!found_extent) {
 965                 key.objectid = op->bytenr;
 966                 key.type = BTRFS_EXTENT_ITEM_KEY;
 967                 key.offset = op->num_bytes;
 968
 969                 ret = remove_extent_backref(trans, extent_root, path);
 970                 BUG_ON(ret);
 971                 btrfs_release_path(extent_root, path);
 972                 ret = btrfs_search_slot(trans, extent_root, &key, path, -1, 1);
 973                 BUG_ON(ret);
 974                 extent_slot = path->slots[0];
 975         }
 976
 977         /* this is where we update the ref count for the extent */
 978         leaf = path->nodes[0];
 979         ei = btrfs_item_ptr(leaf, extent_slot, struct btrfs_extent_item);
 980         refs = btrfs_extent_refs(leaf, ei);
 981         BUG_ON(refs == 0);
 982         refs--;
 983         btrfs_set_extent_refs(leaf, ei, refs);
 984
 985         btrfs_mark_buffer_dirty(leaf);
 986
 987         /*
 988          * This extent needs deleting.  The reason cur_slot is extent_slot +
 989          * num_to_del is because extent_slot points to the slot where the extent
 990          * is, and if the backref was not right next to the extent we will be
 991          * deleting at least 1 item, and will want to start searching at the
 992          * slot directly next to extent_slot.  However if we did find the
 993          * backref next to the extent item them we will be deleting at least 2
 994          * items and will want to start searching directly after the ref slot
 995          */
 996         if (!refs) {
 997                 struct list_head *pos, *n, *end;
 998                 int cur_slot = extent_slot+num_to_del;
 999                 u64 super_used;
1000                 u64 root_used;
1001
1002                 path->slots[0] = extent_slot;
1003                 bytes_freed = op->num_bytes;
1004
1005                 mutex_lock(&info->pinned_mutex);
1006                 ret = pin_down_bytes(trans, extent_root, op->bytenr,
1007                                      op->num_bytes, op->level >=
1008                                      BTRFS_FIRST_FREE_OBJECTID);
1009                 mutex_unlock(&info->pinned_mutex);
1010                 BUG_ON(ret < 0);
1011                 op->del = ret;
1012
1013                 /*
1014                  * we need to see if we can delete multiple things at once, so
1015                  * start looping through the list of extents we are wanting to
1016                  * delete and see if their extent/backref's are right next to
1017                  * eachother and the extents only have 1 ref
1018                  */
1019                 for (pos = cur->next; pos != del_list; pos = pos->next) {
1020                         struct pending_extent_op *tmp;
1021
1022                         tmp = list_entry(pos, struct pending_extent_op, list);
1023
1024                         /* we only want to delete extent+ref at this stage */
1025                         if (cur_slot >= btrfs_header_nritems(leaf) - 1)
1026                                 break;
1027
1028                         btrfs_item_key_to_cpu(leaf, &found_key, cur_slot);
1029                         if (found_key.objectid != tmp->bytenr ||
1030                             found_key.type != BTRFS_EXTENT_ITEM_KEY ||
1031                             found_key.offset != tmp->num_bytes)
1032                                 break;
1033
1034                         /* check to make sure this extent only has one ref */
1035                         ei = btrfs_item_ptr(leaf, cur_slot,
1036                                             struct btrfs_extent_item);
1037                         if (btrfs_extent_refs(leaf, ei) != 1)
1038                                 break;
1039
1040                         btrfs_item_key_to_cpu(leaf, &found_key, cur_slot+1);
1041                         if (found_key.objectid != tmp->bytenr ||
1042                             found_key.type != BTRFS_EXTENT_REF_KEY ||
1043                             found_key.offset != tmp->orig_parent)
1044                                 break;
1045
1046                         /*
1047                          * the ref is right next to the extent, we can set the
1048                          * ref count to 0 since we will delete them both now
1049                          */
1050                         btrfs_set_extent_refs(leaf, ei, 0);
1051
1052                         /* pin down the bytes for this extent */
1053                         mutex_lock(&info->pinned_mutex);
1054                         ret = pin_down_bytes(trans, extent_root, tmp->bytenr,
1055                                              tmp->num_bytes, tmp->level >=
1056                                              BTRFS_FIRST_FREE_OBJECTID);
1057                         mutex_unlock(&info->pinned_mutex);
1058                         BUG_ON(ret < 0);
1059
1060                         /*
1061                          * use the del field to tell if we need to go ahead and
1062                          * free up the extent when we delete the item or not.
1063                          */
1064                         tmp->del = ret;
1065                         bytes_freed += tmp->num_bytes;
1066
1067                         num_to_del += 2;
1068                         cur_slot += 2;
1069                 }
1070                 end = pos;
1071
1072                 /* update the free space counters */
1073                 spin_lock_irq(&info->delalloc_lock);
1074                 super_used = btrfs_super_bytes_used(&info->super_copy);
1075                 btrfs_set_super_bytes_used(&info->super_copy,
1076                                            super_used - bytes_freed);
1077                 spin_unlock_irq(&info->delalloc_lock);
1078
1079                 root_used = btrfs_root_used(&extent_root->root_item);
1080                 btrfs_set_root_used(&extent_root->root_item,
1081                                     root_used - bytes_freed);
1082
1083                 /* delete the items */
1084                 ret = btrfs_del_items(trans, extent_root, path,
1085                                       path->slots[0], num_to_del);
1086                 BUG_ON(ret);
1087
1088                 /*
1089                  * loop through the extents we deleted and do the cleanup work
1090                  * on them
1091                  */
1092                 for (pos = cur, n = pos->next; pos != end;
1093                      pos = n, n = pos->next) {
1094                         struct pending_extent_op *tmp;
1095 #ifdef BIO_RW_DISCARD
1096                         u64 map_length;
1097                         struct btrfs_multi_bio *multi = NULL;
1098 #endif
1099                         tmp = list_entry(pos, struct pending_extent_op, list);
1100
1101                         /*
1102                          * remember tmp->del tells us wether or not we pinned
1103                          * down the extent
1104                          */
1105                         ret = update_block_group(trans, extent_root,
1106                                                  tmp->bytenr, tmp->num_bytes, 0,
1107                                                  tmp->del);
1108                         BUG_ON(ret);
1109
1110 #ifdef BIO_RW_DISCARD
1111                         ret = btrfs_map_block(&info->mapping_tree, READ,
1112                                               tmp->bytenr, &map_length, &multi,
1113                                               0);
1114                         if (!ret) {
1115                                 struct btrfs_bio_stripe *stripe;
1116                                 int i;
1117
1118                                 stripe = multi->stripe;
1119
1120                                 if (map_length > tmp->num_bytes)
1121                                         map_length = tmp->num_bytes;
1122
1123                                 for (i = 0; i < multi->num_stripes;
1124                                      i++, stripe++)
1125                                         blkdev_issue_discard(stripe->dev->bdev,
1126                                                         stripe->physical >> 9,
1127                                                         map_length >> 9);
1128                                 kfree(multi);
1129                         }
1130 #endif
1131                         list_del_init(&tmp->list);
1132                         unlock_extent(&info->extent_ins, tmp->bytenr,
1133                                       tmp->bytenr + tmp->num_bytes - 1,
1134                                       GFP_NOFS);
1135                         kfree(tmp);
1136                 }
1137         } else if (refs && found_extent) {
1138                 /*
1139                  * the ref and extent were right next to eachother, but the
1140                  * extent still has a ref, so just free the backref and keep
1141                  * going
1142                  */
1143                 ret = remove_extent_backref(trans, extent_root, path);
1144                 BUG_ON(ret);
1145
1146                 list_del_init(&op->list);
1147                 unlock_extent(&info->extent_ins, op->bytenr,
1148                               op->bytenr + op->num_bytes - 1, GFP_NOFS);
1149                 kfree(op);
1150         } else {
1151                 /*
1152                  * the extent has multiple refs and the backref we were looking
1153                  * for was not right next to it, so just unlock and go next,
1154                  * we're good to go
1155                  */
1156                 list_del_init(&op->list);
1157                 unlock_extent(&info->extent_ins, op->bytenr,
1158                               op->bytenr + op->num_bytes - 1, GFP_NOFS);
1159                 kfree(op);
1160         }
1161
1162         btrfs_release_path(extent_root, path);
1163         if (!list_empty(del_list))
1164                 goto search;
1165
1166 out:
1167         btrfs_free_path(path);
1168         return ret;
1169 }
1170
1171 static int __btrfs_update_extent_ref(struct btrfs_trans_handle *trans,
1172                                      struct btrfs_root *root, u64 bytenr,
1173                                      u64 orig_parent, u64 parent,
1174                                      u64 orig_root, u64 ref_root,
1175                                      u64 orig_generation, u64 ref_generation,
1176                                      u64 owner_objectid)
1177 {
1178         int ret;
1179         struct btrfs_root *extent_root = root->fs_info->extent_root;
1180         struct btrfs_path *path;
1181
1182         if (root == root->fs_info->extent_root) {
1183                 struct pending_extent_op *extent_op;
1184                 u64 num_bytes;
1185
1186                 BUG_ON(owner_objectid >= BTRFS_MAX_LEVEL);
1187                 num_bytes = btrfs_level_size(root, (int)owner_objectid);
1188                 mutex_lock(&root->fs_info->extent_ins_mutex);
1189                 if (test_range_bit(&root->fs_info->extent_ins, bytenr,
1190                                 bytenr + num_bytes - 1, EXTENT_WRITEBACK, 0)) {
1191                         u64 priv;
1192                         ret = get_state_private(&root->fs_info->extent_ins,
1193                                                 bytenr, &priv);
1194                         BUG_ON(ret);
1195                         extent_op = (struct pending_extent_op *)
1196                                                         (unsigned long)priv;
1197                         BUG_ON(extent_op->parent != orig_parent);
1198                         BUG_ON(extent_op->generation != orig_generation);
1199
1200                         extent_op->parent = parent;
1201                         extent_op->generation = ref_generation;
1202                 } else {
1203                         extent_op = kmalloc(sizeof(*extent_op), GFP_NOFS);
1204                         BUG_ON(!extent_op);
1205
1206                         extent_op->type = PENDING_BACKREF_UPDATE;
1207                         extent_op->bytenr = bytenr;
1208                         extent_op->num_bytes = num_bytes;
1209                         extent_op->parent = parent;
1210                         extent_op->orig_parent = orig_parent;
1211                         extent_op->generation = ref_generation;
1212                         extent_op->orig_generation = orig_generation;
1213                         extent_op->level = (int)owner_objectid;
1214                         INIT_LIST_HEAD(&extent_op->list);
1215                         extent_op->del = 0;
1216
1217                         set_extent_bits(&root->fs_info->extent_ins,
1218                                         bytenr, bytenr + num_bytes - 1,
1219                                         EXTENT_WRITEBACK, GFP_NOFS);
1220                         set_state_private(&root->fs_info->extent_ins,
1221                                           bytenr, (unsigned long)extent_op);
1222                 }
1223                 mutex_unlock(&root->fs_info->extent_ins_mutex);
1224                 return 0;
1225         }
1226
1227         path = btrfs_alloc_path();
1228         if (!path)
1229                 return -ENOMEM;
1230         ret = lookup_extent_backref(trans, extent_root, path,
1231                                     bytenr, orig_parent, orig_root,
1232                                     orig_generation, owner_objectid, 1);
1233         if (ret)
1234                 goto out;
1235         ret = remove_extent_backref(trans, extent_root, path);
1236         if (ret)
1237                 goto out;
1238         ret = insert_extent_backref(trans, extent_root, path, bytenr,
1239                                     parent, ref_root, ref_generation,
1240                                     owner_objectid);
1241         BUG_ON(ret);
1242         finish_current_insert(trans, extent_root, 0);
1243         del_pending_extents(trans, extent_root, 0);
1244 out:
1245         btrfs_free_path(path);
1246         return ret;
1247 }
1248
1249 int btrfs_update_extent_ref(struct btrfs_trans_handle *trans,
1250                             struct btrfs_root *root, u64 bytenr,
1251                             u64 orig_parent, u64 parent,
1252                             u64 ref_root, u64 ref_generation,
1253                             u64 owner_objectid)
1254 {
1255         int ret;
1256         if (ref_root == BTRFS_TREE_LOG_OBJECTID &&
1257             owner_objectid < BTRFS_FIRST_FREE_OBJECTID)
1258                 return 0;
1259         ret = __btrfs_update_extent_ref(trans, root, bytenr, orig_parent,
1260                                         parent, ref_root, ref_root,
1261                                         ref_generation, ref_generation,
1262                                         owner_objectid);
1263         return ret;
1264 }
1265
1266 static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
1267                                   struct btrfs_root *root, u64 bytenr,
1268                                   u64 orig_parent, u64 parent,
1269                                   u64 orig_root, u64 ref_root,
1270                                   u64 orig_generation, u64 ref_generation,
1271                                   u64 owner_objectid)
1272 {
1273         struct btrfs_path *path;
1274         int ret;
1275         struct btrfs_key key;
1276         struct extent_buffer *l;
1277         struct btrfs_extent_item *item;
1278         u32 refs;
1279
1280         path = btrfs_alloc_path();
1281         if (!path)
1282                 return -ENOMEM;
1283
1284         path->reada = 1;
1285         key.objectid = bytenr;
1286         key.type = BTRFS_EXTENT_ITEM_KEY;
1287         key.offset = (u64)-1;
1288
1289         ret = btrfs_search_slot(trans, root->fs_info->extent_root, &key, path,
1290                                 0, 1);
1291         if (ret < 0)
1292                 return ret;
1293         BUG_ON(ret == 0 || path->slots[0] == 0);
1294
1295         path->slots[0]--;
1296         l = path->nodes[0];
1297
1298         btrfs_item_key_to_cpu(l, &key, path->slots[0]);
1299         if (key.objectid != bytenr) {
1300                 btrfs_print_leaf(root->fs_info->extent_root, path->nodes[0]);
1301                 printk("wanted %Lu found %Lu\n", bytenr, key.objectid);
1302                 BUG();
1303         }
1304         BUG_ON(key.type != BTRFS_EXTENT_ITEM_KEY);
1305
1306         item = btrfs_item_ptr(l, path->slots[0], struct btrfs_extent_item);
1307         refs = btrfs_extent_refs(l, item);
1308         btrfs_set_extent_refs(l, item, refs + 1);
1309         btrfs_mark_buffer_dirty(path->nodes[0]);
1310
1311         btrfs_release_path(root->fs_info->extent_root, path);
1312
1313         path->reada = 1;
1314         ret = insert_extent_backref(trans, root->fs_info->extent_root,
1315                                     path, bytenr, parent,
1316                                     ref_root, ref_generation,
1317                                     owner_objectid);
1318         BUG_ON(ret);
1319         finish_current_insert(trans, root->fs_info->extent_root, 0);
1320         del_pending_extents(trans, root->fs_info->extent_root, 0);
1321
1322         btrfs_free_path(path);
1323         return 0;
1324 }
1325
1326 int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
1327                          struct btrfs_root *root,
1328                          u64 bytenr, u64 num_bytes, u64 parent,
1329                          u64 ref_root, u64 ref_generation,
1330                          u64 owner_objectid)
1331 {
1332         int ret;
1333         if (ref_root == BTRFS_TREE_LOG_OBJECTID &&
1334             owner_objectid < BTRFS_FIRST_FREE_OBJECTID)
1335                 return 0;
1336         ret = __btrfs_inc_extent_ref(trans, root, bytenr, 0, parent,
1337                                      0, ref_root, 0, ref_generation,
1338                                      owner_objectid);
1339         return ret;
1340 }
1341
1342 int btrfs_extent_post_op(struct btrfs_trans_handle *trans,
1343                          struct btrfs_root *root)
1344 {
1345         finish_current_insert(trans, root->fs_info->extent_root, 1);
1346         del_pending_extents(trans, root->fs_info->extent_root, 1);
1347         return 0;
1348 }
1349
1350 int btrfs_lookup_extent_ref(struct btrfs_trans_handle *trans,
1351                             struct btrfs_root *root, u64 bytenr,
1352                             u64 num_bytes, u32 *refs)
1353 {
1354         struct btrfs_path *path;
1355         int ret;
1356         struct btrfs_key key;
1357         struct extent_buffer *l;
1358         struct btrfs_extent_item *item;
1359
1360         WARN_ON(num_bytes < root->sectorsize);
1361         path = btrfs_alloc_path();
1362         path->reada = 1;
1363         key.objectid = bytenr;
1364         key.offset = num_bytes;
1365         btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
1366         ret = btrfs_search_slot(trans, root->fs_info->extent_root, &key, path,
1367                                 0, 0);
1368         if (ret < 0)
1369                 goto out;
1370         if (ret != 0) {
1371                 btrfs_print_leaf(root, path->nodes[0]);
1372                 printk("failed to find block number %Lu\n", bytenr);
1373                 BUG();
1374         }
1375         l = path->nodes[0];
1376         item = btrfs_item_ptr(l, path->slots[0], struct btrfs_extent_item);
1377         *refs = btrfs_extent_refs(l, item);
1378 out:
1379         btrfs_free_path(path);
1380         return 0;
1381 }
1382
1383 int btrfs_cross_ref_exist(struct btrfs_trans_handle *trans,
1384                           struct btrfs_root *root, u64 bytenr)
1385 {
1386         struct btrfs_root *extent_root = root->fs_info->extent_root;
1387         struct btrfs_path *path;
1388         struct extent_buffer *leaf;
1389         struct btrfs_extent_ref *ref_item;
1390         struct btrfs_key key;
1391         struct btrfs_key found_key;
1392         u64 ref_root;
1393         u64 last_snapshot;
1394         u32 nritems;
1395         int ret;
1396
1397         key.objectid = bytenr;
1398         key.offset = (u64)-1;
1399         key.type = BTRFS_EXTENT_ITEM_KEY;
1400
1401         path = btrfs_alloc_path();
1402         ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0);
1403         if (ret < 0)
1404                 goto out;
1405         BUG_ON(ret == 0);
1406
1407         ret = -ENOENT;
1408         if (path->slots[0] == 0)
1409                 goto out;
1410
1411         path->slots[0]--;
1412         leaf = path->nodes[0];
1413         btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
1414
1415         if (found_key.objectid != bytenr ||
1416             found_key.type != BTRFS_EXTENT_ITEM_KEY)
1417                 goto out;
1418
1419         last_snapshot = btrfs_root_last_snapshot(&root->root_item);
1420         while (1) {
1421                 leaf = path->nodes[0];
1422                 nritems = btrfs_header_nritems(leaf);
1423                 if (path->slots[0] >= nritems) {
1424                         ret = btrfs_next_leaf(extent_root, path);
1425                         if (ret < 0)
1426                                 goto out;
1427                         if (ret == 0)
1428                                 continue;
1429                         break;
1430                 }
1431                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
1432                 if (found_key.objectid != bytenr)
1433                         break;
1434
1435                 if (found_key.type != BTRFS_EXTENT_REF_KEY) {
1436                         path->slots[0]++;
1437                         continue;
1438                 }
1439
1440                 ref_item = btrfs_item_ptr(leaf, path->slots[0],
1441                                           struct btrfs_extent_ref);
1442                 ref_root = btrfs_ref_root(leaf, ref_item);
1443                 if (ref_root != root->root_key.objectid &&
1444                     ref_root != BTRFS_TREE_LOG_OBJECTID) {
1445                         ret = 1;
1446                         goto out;
1447                 }
1448                 if (btrfs_ref_generation(leaf, ref_item) <= last_snapshot) {
1449                         ret = 1;
1450                         goto out;
1451                 }
1452
1453                 path->slots[0]++;
1454         }
1455         ret = 0;
1456 out:
1457         btrfs_free_path(path);
1458         return ret;
1459 }
1460
1461 int btrfs_cache_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
1462                     struct extent_buffer *buf, u32 nr_extents)
1463 {
1464         struct btrfs_key key;
1465         struct btrfs_file_extent_item *fi;
1466         u64 root_gen;
1467         u32 nritems;
1468         int i;
1469         int level;
1470         int ret = 0;
1471         int shared = 0;
1472
1473         if (!root->ref_cows)
1474                 return 0;
1475
1476         if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
1477                 shared = 0;
1478                 root_gen = root->root_key.offset;
1479         } else {
1480                 shared = 1;
1481                 root_gen = trans->transid - 1;
1482         }
1483
1484         level = btrfs_header_level(buf);
1485         nritems = btrfs_header_nritems(buf);
1486
1487         if (level == 0) {
1488                 struct btrfs_leaf_ref *ref;
1489                 struct btrfs_extent_info *info;
1490
1491                 ref = btrfs_alloc_leaf_ref(root, nr_extents);
1492                 if (!ref) {
1493                         ret = -ENOMEM;
1494                         goto out;
1495                 }
1496
1497                 ref->root_gen = root_gen;
1498                 ref->bytenr = buf->start;
1499                 ref->owner = btrfs_header_owner(buf);
1500                 ref->generation = btrfs_header_generation(buf);
1501                 ref->nritems = nr_extents;
1502                 info = ref->extents;
1503
1504                 for (i = 0; nr_extents > 0 && i < nritems; i++) {
1505                         u64 disk_bytenr;
1506                         btrfs_item_key_to_cpu(buf, &key, i);
1507                         if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
1508                                 continue;
1509                         fi = btrfs_item_ptr(buf, i,
1510                                             struct btrfs_file_extent_item);
1511                         if (btrfs_file_extent_type(buf, fi) ==
1512                             BTRFS_FILE_EXTENT_INLINE)
1513                                 continue;
1514                         disk_bytenr = btrfs_file_extent_disk_bytenr(buf, fi);
1515                         if (disk_bytenr == 0)
1516                                 continue;
1517
1518                         info->bytenr = disk_bytenr;
1519                         info->num_bytes =
1520                                 btrfs_file_extent_disk_num_bytes(buf, fi);
1521                         info->objectid = key.objectid;
1522                         info->offset = key.offset;
1523                         info++;
1524                 }
1525
1526                 ret = btrfs_add_leaf_ref(root, ref, shared);
1527                 if (ret == -EEXIST && shared) {
1528                         struct btrfs_leaf_ref *old;
1529                         old = btrfs_lookup_leaf_ref(root, ref->bytenr);
1530                         BUG_ON(!old);
1531                         btrfs_remove_leaf_ref(root, old);
1532                         btrfs_free_leaf_ref(root, old);
1533                         ret = btrfs_add_leaf_ref(root, ref, shared);
1534                 }
1535                 WARN_ON(ret);
1536                 btrfs_free_leaf_ref(root, ref);
1537         }
1538 out:
1539         return ret;
1540 }
1541
1542 int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
1543                   struct extent_buffer *orig_buf, struct extent_buffer *buf,
1544                   u32 *nr_extents)
1545 {
1546         u64 bytenr;
1547         u64 ref_root;
1548         u64 orig_root;
1549         u64 ref_generation;
1550         u64 orig_generation;
1551         u32 nritems;
1552         u32 nr_file_extents = 0;
1553         struct btrfs_key key;
1554         struct btrfs_file_extent_item *fi;
1555         int i;
1556         int level;
1557         int ret = 0;
1558         int faili = 0;
1559         int (*process_func)(struct btrfs_trans_handle *, struct btrfs_root *,
1560                             u64, u64, u64, u64, u64, u64, u64, u64);
1561
1562         ref_root = btrfs_header_owner(buf);
1563         ref_generation = btrfs_header_generation(buf);
1564         orig_root = btrfs_header_owner(orig_buf);
1565         orig_generation = btrfs_header_generation(orig_buf);
1566
1567         nritems = btrfs_header_nritems(buf);
1568         level = btrfs_header_level(buf);
1569
1570         if (root->ref_cows) {
1571                 process_func = __btrfs_inc_extent_ref;
1572         } else {
1573                 if (level == 0 &&
1574                     root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
1575                         goto out;
1576                 if (level != 0 &&
1577                     root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID)
1578                         goto out;
1579                 process_func = __btrfs_update_extent_ref;
1580         }
1581
1582         for (i = 0; i < nritems; i++) {
1583                 cond_resched();
1584                 if (level == 0) {
1585                         btrfs_item_key_to_cpu(buf, &key, i);
1586                         if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
1587                                 continue;
1588                         fi = btrfs_item_ptr(buf, i,
1589                                             struct btrfs_file_extent_item);
1590                         if (btrfs_file_extent_type(buf, fi) ==
1591                             BTRFS_FILE_EXTENT_INLINE)
1592                                 continue;
1593                         bytenr = btrfs_file_extent_disk_bytenr(buf, fi);
1594                         if (bytenr == 0)
1595                                 continue;
1596
1597                         nr_file_extents++;
1598
1599                         ret = process_func(trans, root, bytenr,
1600                                            orig_buf->start, buf->start,
1601                                            orig_root, ref_root,
1602                                            orig_generation, ref_generation,
1603                                            key.objectid);
1604
1605                         if (ret) {
1606                                 faili = i;
1607                                 WARN_ON(1);
1608                                 goto fail;
1609                         }
1610                 } else {
1611                         bytenr = btrfs_node_blockptr(buf, i);
1612                         ret = process_func(trans, root, bytenr,
1613                                            orig_buf->start, buf->start,
1614                                            orig_root, ref_root,
1615                                            orig_generation, ref_generation,
1616                                            level - 1);
1617                         if (ret) {
1618                                 faili = i;
1619                                 WARN_ON(1);
1620                                 goto fail;
1621                         }
1622                 }
1623         }
1624 out:
1625         if (nr_extents) {
1626                 if (level == 0)
1627                         *nr_extents = nr_file_extents;
1628                 else
1629                         *nr_extents = nritems;
1630         }
1631         return 0;
1632 fail:
1633         WARN_ON(1);
1634         return ret;
1635 }
1636
1637 int btrfs_update_ref(struct btrfs_trans_handle *trans,
1638                      struct btrfs_root *root, struct extent_buffer *orig_buf,
1639                      struct extent_buffer *buf, int start_slot, int nr)
1640
1641 {
1642         u64 bytenr;
1643         u64 ref_root;
1644         u64 orig_root;
1645         u64 ref_generation;
1646         u64 orig_generation;
1647         struct btrfs_key key;
1648         struct btrfs_file_extent_item *fi;
1649         int i;
1650         int ret;
1651         int slot;
1652         int level;
1653
1654         BUG_ON(start_slot < 0);
1655         BUG_ON(start_slot + nr > btrfs_header_nritems(buf));
1656
1657         ref_root = btrfs_header_owner(buf);
1658         ref_generation = btrfs_header_generation(buf);
1659         orig_root = btrfs_header_owner(orig_buf);
1660         orig_generation = btrfs_header_generation(orig_buf);
1661         level = btrfs_header_level(buf);
1662
1663         if (!root->ref_cows) {
1664                 if (level == 0 &&
1665                     root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
1666                         return 0;
1667                 if (level != 0 &&
1668                     root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID)
1669                         return 0;
1670         }
1671
1672         for (i = 0, slot = start_slot; i < nr; i++, slot++) {
1673                 cond_resched();
1674                 if (level == 0) {
1675                         btrfs_item_key_to_cpu(buf, &key, slot);
1676                         if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
1677                                 continue;
1678                         fi = btrfs_item_ptr(buf, slot,
1679                                             struct btrfs_file_extent_item);
1680                         if (btrfs_file_extent_type(buf, fi) ==
1681                             BTRFS_FILE_EXTENT_INLINE)
1682                                 continue;
1683                         bytenr = btrfs_file_extent_disk_bytenr(buf, fi);
1684                         if (bytenr == 0)
1685                                 continue;
1686                         ret = __btrfs_update_extent_ref(trans, root, bytenr,
1687                                             orig_buf->start, buf->start,
1688                                             orig_root, ref_root,
1689                                             orig_generation, ref_generation,
1690                                             key.objectid);
1691                         if (ret)
1692                                 goto fail;
1693                 } else {
1694                         bytenr = btrfs_node_blockptr(buf, slot);
1695                         ret = __btrfs_update_extent_ref(trans, root, bytenr,
1696                                             orig_buf->start, buf->start,
1697                                             orig_root, ref_root,
1698                                             orig_generation, ref_generation,
1699                                             level - 1);
1700                         if (ret)
1701                                 goto fail;
1702                 }
1703         }
1704         return 0;
1705 fail:
1706         WARN_ON(1);
1707         return -1;
1708 }
1709
1710 static int write_one_cache_group(struct btrfs_trans_handle *trans,
1711                                  struct btrfs_root *root,
1712                                  struct btrfs_path *path,
1713                                  struct btrfs_block_group_cache *cache)
1714 {
1715         int ret;
1716         int pending_ret;
1717         struct btrfs_root *extent_root = root->fs_info->extent_root;
1718         unsigned long bi;
1719         struct extent_buffer *leaf;
1720
1721         ret = btrfs_search_slot(trans, extent_root, &cache->key, path, 0, 1);
1722         if (ret < 0)
1723                 goto fail;
1724         BUG_ON(ret);
1725
1726         leaf = path->nodes[0];
1727         bi = btrfs_item_ptr_offset(leaf, path->slots[0]);
1728         write_extent_buffer(leaf, &cache->item, bi, sizeof(cache->item));
1729         btrfs_mark_buffer_dirty(leaf);
1730         btrfs_release_path(extent_root, path);
1731 fail:
1732         finish_current_insert(trans, extent_root, 0);
1733         pending_ret = del_pending_extents(trans, extent_root, 0);
1734         if (ret)
1735                 return ret;
1736         if (pending_ret)
1737                 return pending_ret;
1738         return 0;
1739
1740 }
1741
1742 int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans,
1743                                    struct btrfs_root *root)
1744 {
1745         struct btrfs_block_group_cache *cache, *entry;
1746         struct rb_node *n;
1747         int err = 0;
1748         int werr = 0;
1749         struct btrfs_path *path;
1750         u64 last = 0;
1751
1752         path = btrfs_alloc_path();
1753         if (!path)
1754                 return -ENOMEM;
1755
1756         while(1) {
1757                 cache = NULL;
1758                 spin_lock(&root->fs_info->block_group_cache_lock);
1759                 for (n = rb_first(&root->fs_info->block_group_cache_tree);
1760                      n; n = rb_next(n)) {
1761                         entry = rb_entry(n, struct btrfs_block_group_cache,
1762                                          cache_node);
1763                         if (entry->dirty) {
1764                                 cache = entry;
1765                                 break;
1766                         }
1767                 }
1768                 spin_unlock(&root->fs_info->block_group_cache_lock);
1769
1770                 if (!cache)
1771                         break;
1772
1773                 cache->dirty = 0;
1774                 last += cache->key.offset;
1775
1776                 err = write_one_cache_group(trans, root,
1777                                             path, cache);
1778                 /*
1779                  * if we fail to write the cache group, we want
1780                  * to keep it marked dirty in hopes that a later
1781                  * write will work
1782                  */
1783                 if (err) {
1784                         werr = err;
1785                         continue;
1786                 }
1787         }
1788         btrfs_free_path(path);
1789         return werr;
1790 }
1791
1792 static int update_space_info(struct btrfs_fs_info *info, u64 flags,
1793                              u64 total_bytes, u64 bytes_used,
1794                              struct btrfs_space_info **space_info)
1795 {
1796         struct btrfs_space_info *found;
1797
1798         found = __find_space_info(info, flags);
1799         if (found) {
1800                 spin_lock(&found->lock);
1801                 found->total_bytes += total_bytes;
1802                 found->bytes_used += bytes_used;
1803                 found->full = 0;
1804                 spin_unlock(&found->lock);
1805                 *space_info = found;
1806                 return 0;
1807         }
1808         found = kzalloc(sizeof(*found), GFP_NOFS);
1809         if (!found)
1810                 return -ENOMEM;
1811
1812         list_add(&found->list, &info->space_info);
1813         INIT_LIST_HEAD(&found->block_groups);
1814         init_rwsem(&found->groups_sem);
1815         spin_lock_init(&found->lock);
1816         found->flags = flags;
1817         found->total_bytes = total_bytes;
1818         found->bytes_used = bytes_used;
1819         found->bytes_pinned = 0;
1820         found->bytes_reserved = 0;
1821         found->bytes_readonly = 0;
1822         found->full = 0;
1823         found->force_alloc = 0;
1824         *space_info = found;
1825         return 0;
1826 }
1827
1828 static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
1829 {
1830         u64 extra_flags = flags & (BTRFS_BLOCK_GROUP_RAID0 |
1831                                    BTRFS_BLOCK_GROUP_RAID1 |
1832                                    BTRFS_BLOCK_GROUP_RAID10 |
1833                                    BTRFS_BLOCK_GROUP_DUP);
1834         if (extra_flags) {
1835                 if (flags & BTRFS_BLOCK_GROUP_DATA)
1836                         fs_info->avail_data_alloc_bits |= extra_flags;
1837                 if (flags & BTRFS_BLOCK_GROUP_METADATA)
1838                         fs_info->avail_metadata_alloc_bits |= extra_flags;
1839                 if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
1840                         fs_info->avail_system_alloc_bits |= extra_flags;
1841         }
1842 }
1843
1844 static void set_block_group_readonly(struct btrfs_block_group_cache *cache)
1845 {
1846         spin_lock(&cache->space_info->lock);
1847         spin_lock(&cache->lock);
1848         if (!cache->ro) {
1849                 cache->space_info->bytes_readonly += cache->key.offset -
1850                                         btrfs_block_group_used(&cache->item);
1851                 cache->ro = 1;
1852         }
1853         spin_unlock(&cache->lock);
1854         spin_unlock(&cache->space_info->lock);
1855 }
1856
1857 u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
1858 {
1859         u64 num_devices = root->fs_info->fs_devices->rw_devices;
1860
1861         if (num_devices == 1)
1862                 flags &= ~(BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID0);
1863         if (num_devices < 4)
1864                 flags &= ~BTRFS_BLOCK_GROUP_RAID10;
1865
1866         if ((flags & BTRFS_BLOCK_GROUP_DUP) &&
1867             (flags & (BTRFS_BLOCK_GROUP_RAID1 |
1868                       BTRFS_BLOCK_GROUP_RAID10))) {
1869                 flags &= ~BTRFS_BLOCK_GROUP_DUP;
1870         }
1871
1872         if ((flags & BTRFS_BLOCK_GROUP_RAID1) &&
1873             (flags & BTRFS_BLOCK_GROUP_RAID10)) {
1874                 flags &= ~BTRFS_BLOCK_GROUP_RAID1;
1875         }
1876
1877         if ((flags & BTRFS_BLOCK_GROUP_RAID0) &&
1878             ((flags & BTRFS_BLOCK_GROUP_RAID1) |
1879              (flags & BTRFS_BLOCK_GROUP_RAID10) |
1880              (flags & BTRFS_BLOCK_GROUP_DUP)))
1881                 flags &= ~BTRFS_BLOCK_GROUP_RAID0;
1882         return flags;
1883 }
1884
1885 static int do_chunk_alloc(struct btrfs_trans_handle *trans,
1886                           struct btrfs_root *extent_root, u64 alloc_bytes,
1887                           u64 flags, int force)
1888 {
1889         struct btrfs_space_info *space_info;
1890         u64 thresh;
1891         int ret = 0;
1892
1893         mutex_lock(&extent_root->fs_info->chunk_mutex);
1894
1895         flags = btrfs_reduce_alloc_profile(extent_root, flags);
1896
1897         space_info = __find_space_info(extent_root->fs_info, flags);
1898         if (!space_info) {
1899                 ret = update_space_info(extent_root->fs_info, flags,
1900                                         0, 0, &space_info);
1901                 BUG_ON(ret);
1902         }
1903         BUG_ON(!space_info);
1904
1905         spin_lock(&space_info->lock);
1906         if (space_info->force_alloc) {
1907                 force = 1;
1908                 space_info->force_alloc = 0;
1909         }
1910         if (space_info->full) {
1911                 spin_unlock(&space_info->lock);
1912                 goto out;
1913         }
1914
1915         thresh = space_info->total_bytes - space_info->bytes_readonly;
1916         thresh = div_factor(thresh, 6);
1917         if (!force &&
1918            (space_info->bytes_used + space_info->bytes_pinned +
1919             space_info->bytes_reserved + alloc_bytes) < thresh) {
1920                 spin_unlock(&space_info->lock);
1921                 goto out;
1922         }
1923         spin_unlock(&space_info->lock);
1924
1925         ret = btrfs_alloc_chunk(trans, extent_root, flags);
1926         if (ret) {
1927 printk("space info full %Lu\n", flags);
1928                 space_info->full = 1;
1929         }
1930 out:
1931         mutex_unlock(&extent_root->fs_info->chunk_mutex);
1932         return ret;
1933 }
1934
1935 static int update_block_group(struct btrfs_trans_handle *trans,
1936                               struct btrfs_root *root,
1937                               u64 bytenr, u64 num_bytes, int alloc,
1938                               int mark_free)
1939 {
1940         struct btrfs_block_group_cache *cache;
1941         struct btrfs_fs_info *info = root->fs_info;
1942         u64 total = num_bytes;
1943         u64 old_val;
1944         u64 byte_in_group;
1945
1946         while(total) {
1947                 cache = btrfs_lookup_block_group(info, bytenr);
1948                 if (!cache)
1949                         return -1;
1950                 byte_in_group = bytenr - cache->key.objectid;
1951                 WARN_ON(byte_in_group > cache->key.offset);
1952
1953                 spin_lock(&cache->space_info->lock);
1954                 spin_lock(&cache->lock);
1955                 cache->dirty = 1;
1956                 old_val = btrfs_block_group_used(&cache->item);
1957                 num_bytes = min(total, cache->key.offset - byte_in_group);
1958                 if (alloc) {
1959                         old_val += num_bytes;
1960                         cache->space_info->bytes_used += num_bytes;
1961                         if (cache->ro) {
1962                                 cache->space_info->bytes_readonly -= num_bytes;
1963                                 WARN_ON(1);
1964                         }
1965                         btrfs_set_block_group_used(&cache->item, old_val);
1966                         spin_unlock(&cache->lock);
1967                         spin_unlock(&cache->space_info->lock);
1968                 } else {
1969                         old_val -= num_bytes;
1970                         cache->space_info->bytes_used -= num_bytes;
1971                         if (cache->ro)
1972                                 cache->space_info->bytes_readonly += num_bytes;
1973                         btrfs_set_block_group_used(&cache->item, old_val);
1974                         spin_unlock(&cache->lock);
1975                         spin_unlock(&cache->space_info->lock);
1976                         if (mark_free) {
1977                                 int ret;
1978                                 ret = btrfs_add_free_space(cache, bytenr,
1979                                                            num_bytes);
1980                                 if (ret)
1981                                         return -1;
1982                         }
1983                 }
1984                 total -= num_bytes;
1985                 bytenr += num_bytes;
1986         }
1987         return 0;
1988 }
1989
1990 static u64 first_logical_byte(struct btrfs_root *root, u64 search_start)
1991 {
1992         struct btrfs_block_group_cache *cache;
1993
1994         cache = btrfs_lookup_first_block_group(root->fs_info, search_start);
1995         if (!cache)
1996                 return 0;
1997
1998         return cache->key.objectid;
1999 }
2000
2001 int btrfs_update_pinned_extents(struct btrfs_root *root,
2002                                 u64 bytenr, u64 num, int pin)
2003 {
2004         u64 len;
2005         struct btrfs_block_group_cache *cache;
2006         struct btrfs_fs_info *fs_info = root->fs_info;
2007
2008         WARN_ON(!mutex_is_locked(&root->fs_info->pinned_mutex));
2009         if (pin) {
2010                 set_extent_dirty(&fs_info->pinned_extents,
2011                                 bytenr, bytenr + num - 1, GFP_NOFS);
2012         } else {
2013                 clear_extent_dirty(&fs_info->pinned_extents,
2014                                 bytenr, bytenr + num - 1, GFP_NOFS);
2015         }
2016         while (num > 0) {
2017                 cache = btrfs_lookup_block_group(fs_info, bytenr);
2018                 BUG_ON(!cache);
2019                 len = min(num, cache->key.offset -
2020                           (bytenr - cache->key.objectid));
2021                 if (pin) {
2022                         spin_lock(&cache->space_info->lock);
2023                         spin_lock(&cache->lock);
2024                         cache->pinned += len;
2025                         cache->space_info->bytes_pinned += len;
2026                         spin_unlock(&cache->lock);
2027                         spin_unlock(&cache->space_info->lock);
2028                         fs_info->total_pinned += len;
2029                 } else {
2030                         spin_lock(&cache->space_info->lock);
2031                         spin_lock(&cache->lock);
2032                         cache->pinned -= len;
2033                         cache->space_info->bytes_pinned -= len;
2034                         spin_unlock(&cache->lock);
2035                         spin_unlock(&cache->space_info->lock);
2036                         fs_info->total_pinned -= len;
2037                         if (cache->cached)
2038                                 btrfs_add_free_space(cache, bytenr, len);
2039                 }
2040                 bytenr += len;
2041                 num -= len;
2042         }
2043         return 0;
2044 }
2045
2046 static int update_reserved_extents(struct btrfs_root *root,
2047                                    u64 bytenr, u64 num, int reserve)
2048 {
2049         u64 len;
2050         struct btrfs_block_group_cache *cache;
2051         struct btrfs_fs_info *fs_info = root->fs_info;
2052
2053         while (num > 0) {
2054                 cache = btrfs_lookup_block_group(fs_info, bytenr);
2055                 BUG_ON(!cache);
2056                 len = min(num, cache->key.offset -
2057                           (bytenr - cache->key.objectid));
2058
2059                 spin_lock(&cache->space_info->lock);
2060                 spin_lock(&cache->lock);
2061                 if (reserve) {
2062                         cache->reserved += len;
2063                         cache->space_info->bytes_reserved += len;
2064                 } else {
2065                         cache->reserved -= len;
2066                         cache->space_info->bytes_reserved -= len;
2067                 }
2068                 spin_unlock(&cache->lock);
2069                 spin_unlock(&cache->space_info->lock);
2070                 bytenr += len;
2071                 num -= len;
2072         }
2073         return 0;
2074 }
2075
2076 int btrfs_copy_pinned(struct btrfs_root *root, struct extent_io_tree *copy)
2077 {
2078         u64 last = 0;
2079         u64 start;
2080         u64 end;
2081         struct extent_io_tree *pinned_extents = &root->fs_info->pinned_extents;
2082         int ret;
2083
2084         mutex_lock(&root->fs_info->pinned_mutex);
2085         while(1) {
2086                 ret = find_first_extent_bit(pinned_extents, last,
2087                                             &start, &end, EXTENT_DIRTY);
2088                 if (ret)
2089                         break;
2090                 set_extent_dirty(copy, start, end, GFP_NOFS);
2091                 last = end + 1;
2092         }
2093         mutex_unlock(&root->fs_info->pinned_mutex);
2094         return 0;
2095 }
2096
2097 int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
2098                                struct btrfs_root *root,
2099                                struct extent_io_tree *unpin)
2100 {
2101         u64 start;
2102         u64 end;
2103         int ret;
2104
2105         mutex_lock(&root->fs_info->pinned_mutex);
2106         while(1) {
2107                 ret = find_first_extent_bit(unpin, 0, &start, &end,
2108                                             EXTENT_DIRTY);
2109                 if (ret)
2110                         break;
2111                 btrfs_update_pinned_extents(root, start, end + 1 - start, 0);
2112                 clear_extent_dirty(unpin, start, end, GFP_NOFS);
2113                 if (need_resched()) {
2114                         mutex_unlock(&root->fs_info->pinned_mutex);
2115                         cond_resched();
2116                         mutex_lock(&root->fs_info->pinned_mutex);
2117                 }
2118         }
2119         mutex_unlock(&root->fs_info->pinned_mutex);
2120         return 0;
2121 }
2122
2123 static int finish_current_insert(struct btrfs_trans_handle *trans,
2124                                  struct btrfs_root *extent_root, int all)
2125 {
2126         u64 start;
2127         u64 end;
2128         u64 priv;
2129         u64 search = 0;
2130         u64 skipped = 0;
2131         struct btrfs_fs_info *info = extent_root->fs_info;
2132         struct btrfs_path *path;
2133         struct pending_extent_op *extent_op, *tmp;
2134         struct list_head insert_list, update_list;
2135         int ret;
2136         int num_inserts = 0, max_inserts;
2137
2138         path = btrfs_alloc_path();
2139         INIT_LIST_HEAD(&insert_list);
2140         INIT_LIST_HEAD(&update_list);
2141
2142         max_inserts = extent_root->leafsize /
2143                 (2 * sizeof(struct btrfs_key) + 2 * sizeof(struct btrfs_item) +
2144                  sizeof(struct btrfs_extent_ref) +
2145                  sizeof(struct btrfs_extent_item));
2146 again:
2147         mutex_lock(&info->extent_ins_mutex);
2148         while (1) {
2149                 ret = find_first_extent_bit(&info->extent_ins, search, &start,
2150                                             &end, EXTENT_WRITEBACK);
2151                 if (ret) {
2152                         if (skipped && all && !num_inserts) {
2153                                 skipped = 0;
2154                                 search = 0;
2155                                 continue;
2156                         }
2157                         mutex_unlock(&info->extent_ins_mutex);
2158                         break;
2159                 }
2160
2161                 ret = try_lock_extent(&info->extent_ins, start, end, GFP_NOFS);
2162                 if (!ret) {
2163                         skipped = 1;
2164                         search = end + 1;
2165                         if (need_resched()) {
2166                                 mutex_unlock(&info->extent_ins_mutex);
2167                                 cond_resched();
2168                                 mutex_lock(&info->extent_ins_mutex);
2169                         }
2170                         continue;
2171                 }
2172
2173                 ret = get_state_private(&info->extent_ins, start, &priv);
2174                 BUG_ON(ret);
2175                 extent_op = (struct pending_extent_op *)(unsigned long) priv;
2176
2177                 if (extent_op->type == PENDING_EXTENT_INSERT) {
2178                         num_inserts++;
2179                         list_add_tail(&extent_op->list, &insert_list);
2180                         search = end + 1;
2181                         if (num_inserts == max_inserts) {
2182                                 mutex_unlock(&info->extent_ins_mutex);
2183                                 break;
2184                         }
2185                 } else if (extent_op->type == PENDING_BACKREF_UPDATE) {
2186                         list_add_tail(&extent_op->list, &update_list);
2187                         search = end + 1;
2188                 } else {
2189                         BUG();
2190                 }
2191         }
2192
2193         /*
2194          * process the update list, clear the writeback bit for it, and if
2195          * somebody marked this thing for deletion then just unlock it and be
2196          * done, the free_extents will handle it
2197          */
2198         mutex_lock(&info->extent_ins_mutex);
2199         list_for_each_entry_safe(extent_op, tmp, &update_list, list) {
2200                 clear_extent_bits(&info->extent_ins, extent_op->bytenr,
2201                                   extent_op->bytenr + extent_op->num_bytes - 1,
2202                                   EXTENT_WRITEBACK, GFP_NOFS);
2203                 if (extent_op->del) {
2204                         list_del_init(&extent_op->list);
2205                         unlock_extent(&info->extent_ins, extent_op->bytenr,
2206                                       extent_op->bytenr + extent_op->num_bytes
2207                                       - 1, GFP_NOFS);
2208                         kfree(extent_op);
2209                 }
2210         }
2211         mutex_unlock(&info->extent_ins_mutex);
2212
2213         /*
2214          * still have things left on the update list, go ahead an update
2215          * everything
2216          */
2217         if (!list_empty(&update_list)) {
2218                 ret = update_backrefs(trans, extent_root, path, &update_list);
2219                 BUG_ON(ret);
2220         }
2221
2222         /*
2223          * if no inserts need to be done, but we skipped some extents and we
2224          * need to make sure everything is cleaned then reset everything and
2225          * go back to the beginning
2226          */
2227         if (!num_inserts && all && skipped) {
2228                 search = 0;
2229                 skipped = 0;
2230                 INIT_LIST_HEAD(&update_list);
2231                 INIT_LIST_HEAD(&insert_list);
2232                 goto again;
2233         } else if (!num_inserts) {
2234                 goto out;
2235         }
2236
2237         /*
2238          * process the insert extents list.  Again if we are deleting this
2239          * extent, then just unlock it, pin down the bytes if need be, and be
2240          * done with it.  Saves us from having to actually insert the extent
2241          * into the tree and then subsequently come along and delete it
2242          */
2243         mutex_lock(&info->extent_ins_mutex);
2244         list_for_each_entry_safe(extent_op, tmp, &insert_list, list) {
2245                 clear_extent_bits(&info->extent_ins, extent_op->bytenr,
2246                                   extent_op->bytenr + extent_op->num_bytes - 1,
2247                                   EXTENT_WRITEBACK, GFP_NOFS);
2248                 if (extent_op->del) {
2249                         list_del_init(&extent_op->list);
2250                         unlock_extent(&info->extent_ins, extent_op->bytenr,
2251                                       extent_op->bytenr + extent_op->num_bytes
2252                                       - 1, GFP_NOFS);
2253
2254                         mutex_lock(&extent_root->fs_info->pinned_mutex);
2255                         ret = pin_down_bytes(trans, extent_root,
2256                                              extent_op->bytenr,
2257                                              extent_op->num_bytes, 0);
2258                         mutex_unlock(&extent_root->fs_info->pinned_mutex);
2259
2260                         ret = update_block_group(trans, extent_root,
2261                                                  extent_op->bytenr,
2262                                                  extent_op->num_bytes,
2263                                                  0, ret > 0);
2264                         BUG_ON(ret);
2265                         kfree(extent_op);
2266                         num_inserts--;
2267                 }
2268         }
2269         mutex_unlock(&info->extent_ins_mutex);
2270
2271         ret = insert_extents(trans, extent_root, path, &insert_list,
2272                              num_inserts);
2273         BUG_ON(ret);
2274
2275         /*
2276          * if we broke out of the loop in order to insert stuff because we hit
2277          * the maximum number of inserts at a time we can handle, then loop
2278          * back and pick up where we left off
2279          */
2280         if (num_inserts == max_inserts) {
2281                 INIT_LIST_HEAD(&insert_list);
2282                 INIT_LIST_HEAD(&update_list);
2283                 num_inserts = 0;
2284                 goto again;
2285         }
2286
2287         /*
2288          * again, if we need to make absolutely sure there are no more pending
2289          * extent operations left and we know that we skipped some, go back to
2290          * the beginning and do it all again
2291          */
2292         if (all && skipped) {
2293                 INIT_LIST_HEAD(&insert_list);
2294                 INIT_LIST_HEAD(&update_list);
2295                 search = 0;
2296                 skipped = 0;
2297                 num_inserts = 0;
2298                 goto again;
2299         }
2300 out:
2301         btrfs_free_path(path);
2302         return 0;
2303 }
2304
2305 static int pin_down_bytes(struct btrfs_trans_handle *trans,
2306                           struct btrfs_root *root,
2307                           u64 bytenr, u64 num_bytes, int is_data)
2308 {
2309         int err = 0;
2310         struct extent_buffer *buf;
2311
2312         if (is_data)
2313                 goto pinit;
2314
2315         buf = btrfs_find_tree_block(root, bytenr, num_bytes);
2316         if (!buf)
2317                 goto pinit;
2318
2319         /* we can reuse a block if it hasn't been written
2320          * and it is from this transaction.  We can't
2321          * reuse anything from the tree log root because
2322          * it has tiny sub-transactions.
2323          */
2324         if (btrfs_buffer_uptodate(buf, 0) &&
2325             btrfs_try_tree_lock(buf)) {
2326                 u64 header_owner = btrfs_header_owner(buf);
2327                 u64 header_transid = btrfs_header_generation(buf);
2328                 if (header_owner != BTRFS_TREE_LOG_OBJECTID &&
2329                     header_owner != BTRFS_TREE_RELOC_OBJECTID &&
2330                     header_transid == trans->transid &&
2331                     !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) {
2332                         clean_tree_block(NULL, root, buf);
2333                         btrfs_tree_unlock(buf);
2334                         free_extent_buffer(buf);
2335                         return 1;
2336                 }
2337                 btrfs_tree_unlock(buf);
2338         }
2339         free_extent_buffer(buf);
2340 pinit:
2341         btrfs_update_pinned_extents(root, bytenr, num_bytes, 1);
2342
2343         BUG_ON(err < 0);
2344         return 0;
2345 }
2346
2347 /*
2348  * remove an extent from the root, returns 0 on success
2349  */
2350 static int __free_extent(struct btrfs_trans_handle *trans,
2351                          struct btrfs_root *root,
2352                          u64 bytenr, u64 num_bytes, u64 parent,
2353                          u64 root_objectid, u64 ref_generation,
2354                          u64 owner_objectid, int pin, int mark_free)
2355 {
2356         struct btrfs_path *path;
2357         struct btrfs_key key;
2358         struct btrfs_fs_info *info = root->fs_info;
2359         struct btrfs_root *extent_root = info->extent_root;
2360         struct extent_buffer *leaf;
2361         int ret;
2362         int extent_slot = 0;
2363         int found_extent = 0;
2364         int num_to_del = 1;
2365         struct btrfs_extent_item *ei;
2366         u32 refs;
2367
2368         key.objectid = bytenr;
2369         btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
2370         key.offset = num_bytes;
2371         path = btrfs_alloc_path();
2372         if (!path)
2373                 return -ENOMEM;
2374
2375         path->reada = 1;
2376         ret = lookup_extent_backref(trans, extent_root, path,
2377                                     bytenr, parent, root_objectid,
2378                                     ref_generation, owner_objectid, 1);
2379         if (ret == 0) {
2380                 struct btrfs_key found_key;
2381                 extent_slot = path->slots[0];
2382                 while(extent_slot > 0) {
2383                         extent_slot--;
2384                         btrfs_item_key_to_cpu(path->nodes[0], &found_key,
2385                                               extent_slot);
2386                         if (found_key.objectid != bytenr)
2387                                 break;
2388                         if (found_key.type == BTRFS_EXTENT_ITEM_KEY &&
2389                             found_key.offset == num_bytes) {
2390                                 found_extent = 1;
2391                                 break;
2392                         }
2393                         if (path->slots[0] - extent_slot > 5)
2394                                 break;
2395                 }
2396                 if (!found_extent) {
2397                         ret = remove_extent_backref(trans, extent_root, path);
2398                         BUG_ON(ret);
2399                         btrfs_release_path(extent_root, path);
2400                         ret = btrfs_search_slot(trans, extent_root,
2401                                                 &key, path, -1, 1);
2402                         if (ret) {
2403                                 printk(KERN_ERR "umm, got %d back from search"
2404                                        ", was looking for %Lu\n", ret,
2405                                        bytenr);
2406                                 btrfs_print_leaf(extent_root, path->nodes[0]);
2407                         }
2408                         BUG_ON(ret);
2409                         extent_slot = path->slots[0];
2410                 }
2411         } else {
2412                 btrfs_print_leaf(extent_root, path->nodes[0]);
2413                 WARN_ON(1);
2414                 printk("Unable to find ref byte nr %Lu root %Lu "
2415                        "gen %Lu owner %Lu\n", bytenr,
2416                        root_objectid, ref_generation, owner_objectid);
2417         }
2418
2419         leaf = path->nodes[0];
2420         ei = btrfs_item_ptr(leaf, extent_slot,
2421                             struct btrfs_extent_item);
2422         refs = btrfs_extent_refs(leaf, ei);
2423         BUG_ON(refs == 0);
2424         refs -= 1;
2425         btrfs_set_extent_refs(leaf, ei, refs);
2426
2427         btrfs_mark_buffer_dirty(leaf);
2428
2429         if (refs == 0 && found_extent && path->slots[0] == extent_slot + 1) {
2430                 struct btrfs_extent_ref *ref;
2431                 ref = btrfs_item_ptr(leaf, path->slots[0],
2432                                      struct btrfs_extent_ref);
2433                 BUG_ON(btrfs_ref_num_refs(leaf, ref) != 1);
2434                 /* if the back ref and the extent are next to each other
2435                  * they get deleted below in one shot
2436                  */
2437                 path->slots[0] = extent_slot;
2438                 num_to_del = 2;
2439         } else if (found_extent) {
2440                 /* otherwise delete the extent back ref */
2441                 ret = remove_extent_backref(trans, extent_root, path);
2442                 BUG_ON(ret);
2443                 /* if refs are 0, we need to setup the path for deletion */
2444                 if (refs == 0) {
2445                         btrfs_release_path(extent_root, path);
2446                         ret = btrfs_search_slot(trans, extent_root, &key, path,
2447                                                 -1, 1);
2448                         BUG_ON(ret);
2449                 }
2450         }
2451
2452         if (refs == 0) {
2453                 u64 super_used;
2454                 u64 root_used;
2455 #ifdef BIO_RW_DISCARD
2456                 u64 map_length = num_bytes;
2457                 struct btrfs_multi_bio *multi = NULL;
2458 #endif
2459
2460                 if (pin) {
2461                         mutex_lock(&root->fs_info->pinned_mutex);
2462                         ret = pin_down_bytes(trans, root, bytenr, num_bytes,
2463                                 owner_objectid >= BTRFS_FIRST_FREE_OBJECTID);
2464                         mutex_unlock(&root->fs_info->pinned_mutex);
2465                         if (ret > 0)
2466                                 mark_free = 1;
2467                         BUG_ON(ret < 0);
2468                 }
2469
2470                 /* block accounting for super block */
2471                 spin_lock_irq(&info->delalloc_lock);
2472                 super_used = btrfs_super_bytes_used(&info->super_copy);
2473                 btrfs_set_super_bytes_used(&info->super_copy,
2474                                            super_used - num_bytes);
2475                 spin_unlock_irq(&info->delalloc_lock);
2476
2477                 /* block accounting for root item */
2478                 root_used = btrfs_root_used(&root->root_item);
2479                 btrfs_set_root_used(&root->root_item,
2480                                            root_used - num_bytes);
2481                 ret = btrfs_del_items(trans, extent_root, path, path->slots[0],
2482                                       num_to_del);
2483                 BUG_ON(ret);
2484                 btrfs_release_path(extent_root, path);
2485                 ret = update_block_group(trans, root, bytenr, num_bytes, 0,
2486                                          mark_free);
2487                 BUG_ON(ret);
2488
2489 #ifdef BIO_RW_DISCARD
2490                 /* Tell the block device(s) that the sectors can be discarded */
2491                 ret = btrfs_map_block(&root->fs_info->mapping_tree, READ,
2492                                       bytenr, &map_length, &multi, 0);
2493                 if (!ret) {
2494                         struct btrfs_bio_stripe *stripe = multi->stripes;
2495                         int i;
2496
2497                         if (map_length > num_bytes)
2498                                 map_length = num_bytes;
2499
2500                         for (i = 0; i < multi->num_stripes; i++, stripe++) {
2501                                 blkdev_issue_discard(stripe->dev->bdev,
2502                                                      stripe->physical >> 9,
2503                                                      map_length >> 9);
2504                         }
2505                         kfree(multi);
2506                 }
2507 #endif
2508         }
2509         btrfs_free_path(path);
2510         finish_current_insert(trans, extent_root, 0);
2511         return ret;
2512 }
2513
2514 /*
2515  * find all the blocks marked as pending in the radix tree and remove
2516  * them from the extent map
2517  */
2518 static int del_pending_extents(struct btrfs_trans_handle *trans, struct
2519                                btrfs_root *extent_root, int all)
2520 {
2521         int ret;
2522         int err = 0;
2523         u64 start;
2524         u64 end;
2525         u64 priv;
2526         u64 search = 0;
2527         int nr = 0, skipped = 0;
2528         struct extent_io_tree *pending_del;
2529         struct extent_io_tree *extent_ins;
2530         struct pending_extent_op *extent_op;
2531         struct btrfs_fs_info *info = extent_root->fs_info;
2532         struct list_head delete_list;
2533
2534         INIT_LIST_HEAD(&delete_list);
2535         extent_ins = &extent_root->fs_info->extent_ins;
2536         pending_del = &extent_root->fs_info->pending_del;
2537
2538 again:
2539         mutex_lock(&info->extent_ins_mutex);
2540         while(1) {
2541                 ret = find_first_extent_bit(pending_del, search, &start, &end,
2542                                             EXTENT_WRITEBACK);
2543                 if (ret) {
2544                         if (all && skipped && !nr) {
2545                                 search = 0;
2546                                 continue;
2547                         }
2548                         mutex_unlock(&info->extent_ins_mutex);
2549                         break;
2550                 }
2551
2552                 ret = try_lock_extent(extent_ins, start, end, GFP_NOFS);
2553                 if (!ret) {
2554                         search = end+1;
2555                         skipped = 1;
2556
2557                         if (need_resched()) {
2558                                 mutex_unlock(&info->extent_ins_mutex);
2559                                 cond_resched();
2560                                 mutex_lock(&info->extent_ins_mutex);
2561                         }
2562
2563                         continue;
2564                 }
2565                 BUG_ON(ret < 0);
2566
2567                 ret = get_state_private(pending_del, start, &priv);
2568                 BUG_ON(ret);
2569                 extent_op = (struct pending_extent_op *)(unsigned long)priv;
2570
2571                 clear_extent_bits(pending_del, start, end, EXTENT_WRITEBACK,
2572                                   GFP_NOFS);
2573                 if (!test_range_bit(extent_ins, start, end,
2574                                     EXTENT_WRITEBACK, 0)) {
2575                         list_add_tail(&extent_op->list, &delete_list);
2576                         nr++;
2577                 } else {
2578                         kfree(extent_op);
2579
2580                         ret = get_state_private(&info->extent_ins, start,
2581                                                 &priv);
2582                         BUG_ON(ret);
2583                         extent_op = (struct pending_extent_op *)
2584                                                 (unsigned long)priv;
2585
2586                         clear_extent_bits(&info->extent_ins, start, end,
2587                                           EXTENT_WRITEBACK, GFP_NOFS);
2588
2589                         if (extent_op->type == PENDING_BACKREF_UPDATE) {
2590                                 list_add_tail(&extent_op->list, &delete_list);
2591                                 search = end + 1;
2592                                 nr++;
2593                                 continue;
2594                         }
2595
2596                         mutex_lock(&extent_root->fs_info->pinned_mutex);
2597                         ret = pin_down_bytes(trans, extent_root, start,
2598                                              end + 1 - start, 0);
2599                         mutex_unlock(&extent_root->fs_info->pinned_mutex);
2600
2601                         ret = update_block_group(trans, extent_root, start,
2602                                                 end + 1 - start, 0, ret > 0);
2603
2604                         unlock_extent(extent_ins, start, end, GFP_NOFS);
2605                         BUG_ON(ret);
2606                         kfree(extent_op);
2607                 }
2608                 if (ret)
2609                         err = ret;
2610
2611                 search = end + 1;
2612
2613                 if (need_resched()) {
2614                         mutex_unlock(&info->extent_ins_mutex);
2615                         cond_resched();
2616                         mutex_lock(&info->extent_ins_mutex);
2617                 }
2618         }
2619
2620         if (nr) {
2621                 ret = free_extents(trans, extent_root, &delete_list);
2622                 BUG_ON(ret);
2623         }
2624
2625         if (all && skipped) {
2626                 INIT_LIST_HEAD(&delete_list);
2627                 search = 0;
2628                 nr = 0;
2629                 goto again;
2630         }
2631
2632         return err;
2633 }
2634
2635 /*
2636  * remove an extent from the root, returns 0 on success
2637  */
2638 static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
2639                                struct btrfs_root *root,
2640                                u64 bytenr, u64 num_bytes, u64 parent,
2641                                u64 root_objectid, u64 ref_generation,
2642                                u64 owner_objectid, int pin)
2643 {
2644         struct btrfs_root *extent_root = root->fs_info->extent_root;
2645         int pending_ret;
2646         int ret;
2647
2648         WARN_ON(num_bytes < root->sectorsize);
2649         if (root == extent_root) {
2650                 struct pending_extent_op *extent_op = NULL;
2651
2652                 mutex_lock(&root->fs_info->extent_ins_mutex);
2653                 if (test_range_bit(&root->fs_info->extent_ins, bytenr,
2654                                 bytenr + num_bytes - 1, EXTENT_WRITEBACK, 0)) {
2655                         u64 priv;
2656                         ret = get_state_private(&root->fs_info->extent_ins,
2657                                                 bytenr, &priv);
2658                         BUG_ON(ret);
2659                         extent_op = (struct pending_extent_op *)
2660                                                 (unsigned long)priv;
2661
2662                         extent_op->del = 1;
2663                         if (extent_op->type == PENDING_EXTENT_INSERT) {
2664                                 mutex_unlock(&root->fs_info->extent_ins_mutex);
2665                                 return 0;
2666                         }
2667                 }
2668
2669                 if (extent_op) {
2670                         ref_generation = extent_op->orig_generation;
2671                         parent = extent_op->orig_parent;
2672                 }
2673
2674                 extent_op = kmalloc(sizeof(*extent_op), GFP_NOFS);
2675                 BUG_ON(!extent_op);
2676
2677                 extent_op->type = PENDING_EXTENT_DELETE;
2678                 extent_op->bytenr = bytenr;
2679                 extent_op->num_bytes = num_bytes;
2680                 extent_op->parent = parent;
2681                 extent_op->orig_parent = parent;
2682                 extent_op->generation = ref_generation;
2683                 extent_op->orig_generation = ref_generation;
2684                 extent_op->level = (int)owner_objectid;
2685                 INIT_LIST_HEAD(&extent_op->list);
2686                 extent_op->del = 0;
2687
2688                 set_extent_bits(&root->fs_info->pending_del,
2689                                 bytenr, bytenr + num_bytes - 1,
2690                                 EXTENT_WRITEBACK, GFP_NOFS);
2691                 set_state_private(&root->fs_info->pending_del,
2692                                   bytenr, (unsigned long)extent_op);
2693                 mutex_unlock(&root->fs_info->extent_ins_mutex);
2694                 return 0;
2695         }
2696         /* if metadata always pin */
2697         if (owner_objectid < BTRFS_FIRST_FREE_OBJECTID) {
2698                 if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
2699                         struct btrfs_block_group_cache *cache;
2700
2701                         /* btrfs_free_reserved_extent */
2702                         cache = btrfs_lookup_block_group(root->fs_info, bytenr);
2703                         BUG_ON(!cache);
2704                         btrfs_add_free_space(cache, bytenr, num_bytes);
2705                         update_reserved_extents(root, bytenr, num_bytes, 0);
2706                         return 0;
2707                 }
2708                 pin = 1;
2709         }
2710
2711         /* if data pin when any transaction has committed this */
2712         if (ref_generation != trans->transid)
2713                 pin = 1;
2714
2715         ret = __free_extent(trans, root, bytenr, num_bytes, parent,
2716                             root_objectid, ref_generation,
2717                             owner_objectid, pin, pin == 0);
2718
2719         finish_current_insert(trans, root->fs_info->extent_root, 0);
2720         pending_ret = del_pending_extents(trans, root->fs_info->extent_root, 0);
2721         return ret ? ret : pending_ret;
2722 }
2723
2724 int btrfs_free_extent(struct btrfs_trans_handle *trans,
2725                       struct btrfs_root *root,
2726                       u64 bytenr, u64 num_bytes, u64 parent,
2727                       u64 root_objectid, u64 ref_generation,
2728                       u64 owner_objectid, int pin)
2729 {
2730         int ret;
2731
2732         ret = __btrfs_free_extent(trans, root, bytenr, num_bytes, parent,
2733                                   root_objectid, ref_generation,
2734                                   owner_objectid, pin);
2735         return ret;
2736 }
2737
2738 static u64 stripe_align(struct btrfs_root *root, u64 val)
2739 {
2740         u64 mask = ((u64)root->stripesize - 1);
2741         u64 ret = (val + mask) & ~mask;
2742         return ret;
2743 }
2744
2745 /*
2746  * walks the btree of allocated extents and find a hole of a given size.
2747  * The key ins is changed to record the hole:
2748  * ins->objectid == block start
2749  * ins->flags = BTRFS_EXTENT_ITEM_KEY
2750  * ins->offset == number of blocks
2751  * Any available blocks before search_start are skipped.
2752  */
2753 static int noinline find_free_extent(struct btrfs_trans_handle *trans,
2754                                      struct btrfs_root *orig_root,
2755                                      u64 num_bytes, u64 empty_size,
2756                                      u64 search_start, u64 search_end,
2757                                      u64 hint_byte, struct btrfs_key *ins,
2758                                      u64 exclude_start, u64 exclude_nr,
2759                                      int data)
2760 {
2761         int ret = 0;
2762         struct btrfs_root * root = orig_root->fs_info->extent_root;
2763         u64 total_needed = num_bytes;
2764         u64 *last_ptr = NULL;
2765         u64 last_wanted = 0;
2766         struct btrfs_block_group_cache *block_group = NULL;
2767         int chunk_alloc_done = 0;
2768         int empty_cluster = 2 * 1024 * 1024;
2769         int allowed_chunk_alloc = 0;
2770         struct list_head *head = NULL, *cur = NULL;
2771         int loop = 0;
2772         int extra_loop = 0;
2773         struct btrfs_space_info *space_info;
2774
2775         WARN_ON(num_bytes < root->sectorsize);
2776         btrfs_set_key_type(ins, BTRFS_EXTENT_ITEM_KEY);
2777         ins->objectid = 0;
2778         ins->offset = 0;
2779
2780         if (orig_root->ref_cows || empty_size)
2781                 allowed_chunk_alloc = 1;
2782
2783         if (data & BTRFS_BLOCK_GROUP_METADATA) {
2784                 last_ptr = &root->fs_info->last_alloc;
2785                 empty_cluster = 64 * 1024;
2786         }
2787
2788         if ((data & BTRFS_BLOCK_GROUP_DATA) && btrfs_test_opt(root, SSD))
2789                 last_ptr = &root->fs_info->last_data_alloc;
2790
2791         if (last_ptr) {
2792                 if (*last_ptr) {
2793                         hint_byte = *last_ptr;
2794                         last_wanted = *last_ptr;
2795                 } else
2796                         empty_size += empty_cluster;
2797         } else {
2798                 empty_cluster = 0;
2799         }
2800         search_start = max(search_start, first_logical_byte(root, 0));
2801         search_start = max(search_start, hint_byte);
2802
2803         if (last_wanted && search_start != last_wanted) {
2804                 last_wanted = 0;
2805                 empty_size += empty_cluster;
2806         }
2807
2808         total_needed += empty_size;
2809         block_group = btrfs_lookup_block_group(root->fs_info, search_start);
2810         if (!block_group)
2811                 block_group = btrfs_lookup_first_block_group(root->fs_info,
2812                                                              search_start);
2813         space_info = __find_space_info(root->fs_info, data);
2814
2815         down_read(&space_info->groups_sem);
2816         while (1) {
2817                 struct btrfs_free_space *free_space;
2818                 /*
2819                  * the only way this happens if our hint points to a block
2820                  * group thats not of the proper type, while looping this
2821                  * should never happen
2822                  */
2823                 if (empty_size)
2824                         extra_loop = 1;
2825
2826                 if (!block_group)
2827                         goto new_group_no_lock;
2828
2829                 mutex_lock(&block_group->alloc_mutex);
2830                 if (unlikely(!block_group_bits(block_group, data)))
2831                         goto new_group;
2832
2833                 ret = cache_block_group(root, block_group);
2834                 if (ret) {
2835                         mutex_unlock(&block_group->alloc_mutex);
2836                         break;
2837                 }
2838
2839                 if (block_group->ro)
2840                         goto new_group;
2841
2842                 free_space = btrfs_find_free_space(block_group, search_start,
2843                                                    total_needed);
2844                 if (free_space) {
2845                         u64 start = block_group->key.objectid;
2846                         u64 end = block_group->key.objectid +
2847                                 block_group->key.offset;
2848
2849                         search_start = stripe_align(root, free_space->offset);
2850
2851                         /* move on to the next group */
2852                         if (search_start + num_bytes >= search_end)
2853                                 goto new_group;
2854
2855                         /* move on to the next group */
2856                         if (search_start + num_bytes > end)
2857                                 goto new_group;
2858
2859                         if (last_wanted && search_start != last_wanted) {
2860                                 total_needed += empty_cluster;
2861                                 empty_size += empty_cluster;
2862                                 last_wanted = 0;
2863                                 /*
2864                                  * if search_start is still in this block group
2865                                  * then we just re-search this block group
2866                                  */
2867                                 if (search_start >= start &&
2868                                     search_start < end) {
2869                                         mutex_unlock(&block_group->alloc_mutex);
2870                                         continue;
2871                                 }
2872
2873                                 /* else we go to the next block group */
2874                                 goto new_group;
2875                         }
2876
2877                         if (exclude_nr > 0 &&
2878                             (search_start + num_bytes > exclude_start &&
2879                              search_start < exclude_start + exclude_nr)) {
2880                                 search_start = exclude_start + exclude_nr;
2881                                 /*
2882                                  * if search_start is still in this block group
2883                                  * then we just re-search this block group
2884                                  */
2885                                 if (search_start >= start &&
2886                                     search_start < end) {
2887                                         mutex_unlock(&block_group->alloc_mutex);
2888                                         last_wanted = 0;
2889                                         continue;
2890                                 }
2891
2892                                 /* else we go to the next block group */
2893                                 goto new_group;
2894                         }
2895
2896                         ins->objectid = search_start;
2897                         ins->offset = num_bytes;
2898
2899                         btrfs_remove_free_space_lock(block_group, search_start,
2900                                                      num_bytes);
2901                         /* we are all good, lets return */
2902                         mutex_unlock(&block_group->alloc_mutex);
2903                         break;
2904                 }
2905 new_group:
2906                 mutex_unlock(&block_group->alloc_mutex);
2907 new_group_no_lock:
2908                 /* don't try to compare new allocations against the
2909                  * last allocation any more
2910                  */
2911                 last_wanted = 0;
2912
2913                 /*
2914                  * Here's how this works.
2915                  * loop == 0: we were searching a block group via a hint
2916                  *              and didn't find anything, so we start at
2917                  *              the head of the block groups and keep searching
2918                  * loop == 1: we're searching through all of the block groups
2919                  *              if we hit the head again we have searched
2920                  *              all of the block groups for this space and we
2921                  *              need to try and allocate, if we cant error out.
2922                  * loop == 2: we allocated more space and are looping through
2923                  *              all of the block groups again.
2924                  */
2925                 if (loop == 0) {
2926                         head = &space_info->block_groups;
2927                         cur = head->next;
2928                         loop++;
2929                 } else if (loop == 1 && cur == head) {
2930                         int keep_going;
2931
2932                         /* at this point we give up on the empty_size
2933                          * allocations and just try to allocate the min
2934                          * space.
2935                          *
2936                          * The extra_loop field was set if an empty_size
2937                          * allocation was attempted above, and if this
2938                          * is try we need to try the loop again without
2939                          * the additional empty_size.
2940                          */
2941                         total_needed -= empty_size;
2942                         empty_size = 0;
2943                         keep_going = extra_loop;
2944                         loop++;
2945
2946                         if (allowed_chunk_alloc && !chunk_alloc_done) {
2947                                 up_read(&space_info->groups_sem);
2948                                 ret = do_chunk_alloc(trans, root, num_bytes +
2949                                                      2 * 1024 * 1024, data, 1);
2950                                 down_read(&space_info->groups_sem);
2951                                 if (ret < 0)
2952                                         goto loop_check;
2953                                 head = &space_info->block_groups;
2954                                 /*
2955                                  * we've allocated a new chunk, keep
2956                                  * trying
2957                                  */
2958                                 keep_going = 1;
2959                                 chunk_alloc_done = 1;
2960                         } else if (!allowed_chunk_alloc) {
2961                                 space_info->force_alloc = 1;
2962                         }
2963 loop_check:
2964                         if (keep_going) {
2965                                 cur = head->next;
2966                                 extra_loop = 0;
2967                         } else {
2968                                 break;
2969                         }
2970                 } else if (cur == head) {
2971                         break;
2972                 }
2973
2974                 block_group = list_entry(cur, struct btrfs_block_group_cache,
2975                                          list);
2976                 search_start = block_group->key.objectid;
2977                 cur = cur->next;
2978         }
2979
2980         /* we found what we needed */
2981         if (ins->objectid) {
2982                 if (!(data & BTRFS_BLOCK_GROUP_DATA))
2983                         trans->block_group = block_group;
2984
2985                 if (last_ptr)
2986                         *last_ptr = ins->objectid + ins->offset;
2987                 ret = 0;
2988         } else if (!ret) {
2989                 printk(KERN_ERR "we were searching for %Lu bytes, num_bytes %Lu,"
2990                        " loop %d, allowed_alloc %d\n", total_needed, num_bytes,
2991                        loop, allowed_chunk_alloc);
2992                 ret = -ENOSPC;
2993         }
2994
2995         up_read(&space_info->groups_sem);
2996         return ret;
2997 }
2998
2999 static void dump_space_info(struct btrfs_space_info *info, u64 bytes)
3000 {
3001         struct btrfs_block_group_cache *cache;
3002         struct list_head *l;
3003
3004         printk(KERN_INFO "space_info has %Lu free, is %sfull\n",
3005                info->total_bytes - info->bytes_used - info->bytes_pinned -
3006                info->bytes_reserved, (info->full) ? "" : "not ");
3007
3008         down_read(&info->groups_sem);
3009         list_for_each(l, &info->block_groups) {
3010                 cache = list_entry(l, struct btrfs_block_group_cache, list);
3011                 spin_lock(&cache->lock);
3012                 printk(KERN_INFO "block group %Lu has %Lu bytes, %Lu used "
3013                        "%Lu pinned %Lu reserved\n",
3014                        cache->key.objectid, cache->key.offset,
3015                        btrfs_block_group_used(&cache->item),
3016                        cache->pinned, cache->reserved);
3017                 btrfs_dump_free_space(cache, bytes);
3018                 spin_unlock(&cache->lock);
3019         }
3020         up_read(&info->groups_sem);
3021 }
3022
3023 static int __btrfs_reserve_extent(struct btrfs_trans_handle *trans,
3024                                   struct btrfs_root *root,
3025                                   u64 num_bytes, u64 min_alloc_size,
3026                                   u64 empty_size, u64 hint_byte,
3027                                   u64 search_end, struct btrfs_key *ins,
3028                                   u64 data)
3029 {
3030         int ret;
3031         u64 search_start = 0;
3032         u64 alloc_profile;
3033         struct btrfs_fs_info *info = root->fs_info;
3034
3035         if (data) {
3036                 alloc_profile = info->avail_data_alloc_bits &
3037                                 info->data_alloc_profile;
3038                 data = BTRFS_BLOCK_GROUP_DATA | alloc_profile;
3039         } else if (root == root->fs_info->chunk_root) {
3040                 alloc_profile = info->avail_system_alloc_bits &
3041                                 info->system_alloc_profile;
3042                 data = BTRFS_BLOCK_GROUP_SYSTEM | alloc_profile;
3043         } else {
3044                 alloc_profile = info->avail_metadata_alloc_bits &
3045                                 info->metadata_alloc_profile;
3046                 data = BTRFS_BLOCK_GROUP_METADATA | alloc_profile;
3047         }
3048 again:
3049         data = btrfs_reduce_alloc_profile(root, data);
3050         /*
3051          * the only place that sets empty_size is btrfs_realloc_node, which
3052          * is not called recursively on allocations
3053          */
3054         if (empty_size || root->ref_cows) {
3055                 if (!(data & BTRFS_BLOCK_GROUP_METADATA)) {
3056                         ret = do_chunk_alloc(trans, root->fs_info->extent_root,
3057                                      2 * 1024 * 1024,
3058                                      BTRFS_BLOCK_GROUP_METADATA |
3059                                      (info->metadata_alloc_profile &
3060                                       info->avail_metadata_alloc_bits), 0);
3061                 }
3062                 ret = do_chunk_alloc(trans, root->fs_info->extent_root,
3063                                      num_bytes + 2 * 1024 * 1024, data, 0);
3064         }
3065
3066         WARN_ON(num_bytes < root->sectorsize);
3067         ret = find_free_extent(trans, root, num_bytes, empty_size,
3068                                search_start, search_end, hint_byte, ins,
3069                                trans->alloc_exclude_start,
3070                                trans->alloc_exclude_nr, data);
3071
3072         if (ret == -ENOSPC && num_bytes > min_alloc_size) {
3073                 num_bytes = num_bytes >> 1;
3074                 num_bytes = num_bytes & ~(root->sectorsize - 1);
3075                 num_bytes = max(num_bytes, min_alloc_size);
3076                 do_chunk_alloc(trans, root->fs_info->extent_root,
3077                                num_bytes, data, 1);
3078                 goto again;
3079         }
3080         if (ret) {
3081                 struct btrfs_space_info *sinfo;
3082
3083                 sinfo = __find_space_info(root->fs_info, data);
3084                 printk("allocation failed flags %Lu, wanted %Lu\n",
3085                        data, num_bytes);
3086                 dump_space_info(sinfo, num_bytes);
3087                 BUG();
3088         }
3089
3090         return ret;
3091 }
3092
3093 int btrfs_free_reserved_extent(struct btrfs_root *root, u64 start, u64 len)
3094 {
3095         struct btrfs_block_group_cache *cache;
3096
3097         cache = btrfs_lookup_block_group(root->fs_info, start);
3098         if (!cache) {
3099                 printk(KERN_ERR "Unable to find block group for %Lu\n", start);
3100                 return -ENOSPC;
3101         }
3102         btrfs_add_free_space(cache, start, len);
3103         update_reserved_extents(root, start, len, 0);
3104         return 0;
3105 }
3106
3107 int btrfs_reserve_extent(struct btrfs_trans_handle *trans,
3108                                   struct btrfs_root *root,
3109                                   u64 num_bytes, u64 min_alloc_size,
3110                                   u64 empty_size, u64 hint_byte,
3111                                   u64 search_end, struct btrfs_key *ins,
3112                                   u64 data)
3113 {
3114         int ret;
3115         ret = __btrfs_reserve_extent(trans, root, num_bytes, min_alloc_size,
3116                                      empty_size, hint_byte, search_end, ins,
3117                                      data);
3118         update_reserved_extents(root, ins->objectid, ins->offset, 1);
3119         return ret;
3120 }
3121
3122 static int __btrfs_alloc_reserved_extent(struct btrfs_trans_handle *trans,
3123                                          struct btrfs_root *root, u64 parent,
3124                                          u64 root_objectid, u64 ref_generation,
3125                                          u64 owner, struct btrfs_key *ins)
3126 {
3127         int ret;
3128         int pending_ret;
3129         u64 super_used;
3130         u64 root_used;
3131         u64 num_bytes = ins->offset;
3132         u32 sizes[2];
3133         struct btrfs_fs_info *info = root->fs_info;
3134         struct btrfs_root *extent_root = info->extent_root;
3135         struct btrfs_extent_item *extent_item;
3136         struct btrfs_extent_ref *ref;
3137         struct btrfs_path *path;
3138         struct btrfs_key keys[2];
3139
3140         if (parent == 0)
3141                 parent = ins->objectid;
3142
3143         /* block accounting for super block */
3144         spin_lock_irq(&info->delalloc_lock);
3145         super_used = btrfs_super_bytes_used(&info->super_copy);
3146         btrfs_set_super_bytes_used(&info->super_copy, super_used + num_bytes);
3147         spin_unlock_irq(&info->delalloc_lock);
3148
3149         /* block accounting for root item */
3150         root_used = btrfs_root_used(&root->root_item);
3151         btrfs_set_root_used(&root->root_item, root_used + num_bytes);
3152
3153         if (root == extent_root) {
3154                 struct pending_extent_op *extent_op;
3155
3156                 extent_op = kmalloc(sizeof(*extent_op), GFP_NOFS);
3157                 BUG_ON(!extent_op);
3158
3159                 extent_op->type = PENDING_EXTENT_INSERT;
3160                 extent_op->bytenr = ins->objectid;
3161                 extent_op->num_bytes = ins->offset;
3162                 extent_op->parent = parent;
3163                 extent_op->orig_parent = 0;
3164                 extent_op->generation = ref_generation;
3165                 extent_op->orig_generation = 0;
3166                 extent_op->level = (int)owner;
3167                 INIT_LIST_HEAD(&extent_op->list);
3168                 extent_op->del = 0;
3169
3170                 mutex_lock(&root->fs_info->extent_ins_mutex);
3171                 set_extent_bits(&root->fs_info->extent_ins, ins->objectid,
3172                                 ins->objectid + ins->offset - 1,
3173                                 EXTENT_WRITEBACK, GFP_NOFS);
3174                 set_state_private(&root->fs_info->extent_ins,
3175                                   ins->objectid, (unsigned long)extent_op);
3176                 mutex_unlock(&root->fs_info->extent_ins_mutex);
3177                 goto update_block;
3178         }
3179
3180         memcpy(&keys[0], ins, sizeof(*ins));
3181         keys[1].objectid = ins->objectid;
3182         keys[1].type = BTRFS_EXTENT_REF_KEY;
3183         keys[1].offset = parent;
3184         sizes[0] = sizeof(*extent_item);
3185         sizes[1] = sizeof(*ref);
3186
3187         path = btrfs_alloc_path();
3188         BUG_ON(!path);
3189
3190         ret = btrfs_insert_empty_items(trans, extent_root, path, keys,
3191                                        sizes, 2);
3192         BUG_ON(ret);
3193
3194         extent_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
3195                                      struct btrfs_extent_item);
3196         btrfs_set_extent_refs(path->nodes[0], extent_item, 1);
3197         ref = btrfs_item_ptr(path->nodes[0], path->slots[0] + 1,
3198                              struct btrfs_extent_ref);
3199
3200         btrfs_set_ref_root(path->nodes[0], ref, root_objectid);
3201         btrfs_set_ref_generation(path->nodes[0], ref, ref_generation);
3202         btrfs_set_ref_objectid(path->nodes[0], ref, owner);
3203         btrfs_set_ref_num_refs(path->nodes[0], ref, 1);
3204
3205         btrfs_mark_buffer_dirty(path->nodes[0]);
3206
3207         trans->alloc_exclude_start = 0;
3208         trans->alloc_exclude_nr = 0;
3209         btrfs_free_path(path);
3210         finish_current_insert(trans, extent_root, 0);
3211         pending_ret = del_pending_extents(trans, extent_root, 0);
3212
3213         if (ret)
3214                 goto out;
3215         if (pending_ret) {
3216                 ret = pending_ret;
3217                 goto out;
3218         }
3219
3220 update_block:
3221         ret = update_block_group(trans, root, ins->objectid, ins->offset, 1, 0);
3222         if (ret) {
3223                 printk("update block group failed for %Lu %Lu\n",
3224                        ins->objectid, ins->offset);
3225                 BUG();
3226         }
3227 out:
3228         return ret;
3229 }
3230
3231 int btrfs_alloc_reserved_extent(struct btrfs_trans_handle *trans,
3232                                 struct btrfs_root *root, u64 parent,
3233                                 u64 root_objectid, u64 ref_generation,
3234                                 u64 owner, struct btrfs_key *ins)
3235 {
3236         int ret;
3237
3238         if (root_objectid == BTRFS_TREE_LOG_OBJECTID)
3239                 return 0;
3240         ret = __btrfs_alloc_reserved_extent(trans, root, parent, root_objectid,
3241                                             ref_generation, owner, ins);
3242         update_reserved_extents(root, ins->objectid, ins->offset, 0);
3243         return ret;
3244 }
3245
3246 /*
3247  * this is used by the tree logging recovery code.  It records that
3248  * an extent has been allocated and makes sure to clear the free
3249  * space cache bits as well
3250  */
3251 int btrfs_alloc_logged_extent(struct btrfs_trans_handle *trans,
3252                                 struct btrfs_root *root, u64 parent,
3253                                 u64 root_objectid, u64 ref_generation,
3254                                 u64 owner, struct btrfs_key *ins)
3255 {
3256         int ret;
3257         struct btrfs_block_group_cache *block_group;
3258
3259         block_group = btrfs_lookup_block_group(root->fs_info, ins->objectid);
3260         mutex_lock(&block_group->alloc_mutex);
3261         cache_block_group(root, block_group);
3262
3263         ret = btrfs_remove_free_space_lock(block_group, ins->objectid,
3264                                            ins->offset);
3265         mutex_unlock(&block_group->alloc_mutex);
3266         BUG_ON(ret);
3267         ret = __btrfs_alloc_reserved_extent(trans, root, parent, root_objectid,
3268                                             ref_generation, owner, ins);
3269         return ret;
3270 }
3271
3272 /*
3273  * finds a free extent and does all the dirty work required for allocation
3274  * returns the key for the extent through ins, and a tree buffer for
3275  * the first block of the extent through buf.
3276  *
3277  * returns 0 if everything worked, non-zero otherwise.
3278  */
3279 int btrfs_alloc_extent(struct btrfs_trans_handle *trans,
3280                        struct btrfs_root *root,
3281                        u64 num_bytes, u64 parent, u64 min_alloc_size,
3282                        u64 root_objectid, u64 ref_generation,
3283                        u64 owner_objectid, u64 empty_size, u64 hint_byte,
3284                        u64 search_end, struct btrfs_key *ins, u64 data)
3285 {
3286         int ret;
3287
3288         ret = __btrfs_reserve_extent(trans, root, num_bytes,
3289                                      min_alloc_size, empty_size, hint_byte,
3290                                      search_end, ins, data);
3291         BUG_ON(ret);
3292         if (root_objectid != BTRFS_TREE_LOG_OBJECTID) {
3293                 ret = __btrfs_alloc_reserved_extent(trans, root, parent,
3294                                         root_objectid, ref_generation,
3295                                         owner_objectid, ins);
3296                 BUG_ON(ret);
3297
3298         } else {
3299                 update_reserved_extents(root, ins->objectid, ins->offset, 1);
3300         }
3301         return ret;
3302 }
3303
3304 struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
3305                                             struct btrfs_root *root,
3306                                             u64 bytenr, u32 blocksize)
3307 {
3308         struct extent_buffer *buf;
3309
3310         buf = btrfs_find_create_tree_block(root, bytenr, blocksize);
3311         if (!buf)
3312                 return ERR_PTR(-ENOMEM);
3313         btrfs_set_header_generation(buf, trans->transid);
3314         btrfs_tree_lock(buf);
3315         clean_tree_block(trans, root, buf);
3316         btrfs_set_buffer_uptodate(buf);
3317         if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
3318                 set_extent_dirty(&root->dirty_log_pages, buf->start,
3319                          buf->start + buf->len - 1, GFP_NOFS);
3320         } else {
3321                 set_extent_dirty(&trans->transaction->dirty_pages, buf->start,
3322                          buf->start + buf->len - 1, GFP_NOFS);
3323         }
3324         trans->blocks_used++;
3325         return buf;
3326 }
3327
3328 /*
3329  * helper function to allocate a block for a given tree
3330  * returns the tree buffer or NULL.
3331  */
3332 struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
3333                                              struct btrfs_root *root,
3334                                              u32 blocksize, u64 parent,
3335                                              u64 root_objectid,
3336                                              u64 ref_generation,
3337                                              int level,
3338                                              u64 hint,
3339                                              u64 empty_size)
3340 {
3341         struct btrfs_key ins;
3342         int ret;
3343         struct extent_buffer *buf;
3344
3345         ret = btrfs_alloc_extent(trans, root, blocksize, parent, blocksize,
3346                                  root_objectid, ref_generation, level,
3347                                  empty_size, hint, (u64)-1, &ins, 0);
3348         if (ret) {
3349                 BUG_ON(ret > 0);
3350                 return ERR_PTR(ret);
3351         }
3352
3353         buf = btrfs_init_new_buffer(trans, root, ins.objectid, blocksize);
3354         return buf;
3355 }
3356
3357 int btrfs_drop_leaf_ref(struct btrfs_trans_handle *trans,
3358                         struct btrfs_root *root, struct extent_buffer *leaf)
3359 {
3360         u64 leaf_owner;
3361         u64 leaf_generation;
3362         struct btrfs_key key;
3363         struct btrfs_file_extent_item *fi;
3364         int i;
3365         int nritems;
3366         int ret;
3367
3368         BUG_ON(!btrfs_is_leaf(leaf));
3369         nritems = btrfs_header_nritems(leaf);
3370         leaf_owner = btrfs_header_owner(leaf);
3371         leaf_generation = btrfs_header_generation(leaf);
3372
3373         for (i = 0; i < nritems; i++) {
3374                 u64 disk_bytenr;
3375                 cond_resched();
3376
3377                 btrfs_item_key_to_cpu(leaf, &key, i);
3378                 if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
3379                         continue;
3380                 fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
3381                 if (btrfs_file_extent_type(leaf, fi) ==
3382                     BTRFS_FILE_EXTENT_INLINE)
3383                         continue;
3384                 /*
3385                  * FIXME make sure to insert a trans record that
3386                  * repeats the snapshot del on crash
3387                  */
3388                 disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
3389                 if (disk_bytenr == 0)
3390                         continue;
3391
3392                 ret = __btrfs_free_extent(trans, root, disk_bytenr,
3393                                 btrfs_file_extent_disk_num_bytes(leaf, fi),
3394                                 leaf->start, leaf_owner, leaf_generation,
3395                                 key.objectid, 0);
3396                 BUG_ON(ret);
3397
3398                 atomic_inc(&root->fs_info->throttle_gen);
3399                 wake_up(&root->fs_info->transaction_throttle);
3400                 cond_resched();
3401         }
3402         return 0;
3403 }
3404
3405 static int noinline cache_drop_leaf_ref(struct btrfs_trans_handle *trans,
3406                                         struct btrfs_root *root,
3407                                         struct btrfs_leaf_ref *ref)
3408 {
3409         int i;
3410         int ret;
3411         struct btrfs_extent_info *info = ref->extents;
3412
3413         for (i = 0; i < ref->nritems; i++) {
3414                 ret = __btrfs_free_extent(trans, root, info->bytenr,
3415                                           info->num_bytes, ref->bytenr,
3416                                           ref->owner, ref->generation,
3417                                           info->objectid, 0);
3418
3419                 atomic_inc(&root->fs_info->throttle_gen);
3420                 wake_up(&root->fs_info->transaction_throttle);
3421                 cond_resched();
3422
3423                 BUG_ON(ret);
3424                 info++;
3425         }
3426
3427         return 0;
3428 }
3429
3430 int drop_snap_lookup_refcount(struct btrfs_root *root, u64 start, u64 len,
3431                               u32 *refs)
3432 {
3433         int ret;
3434
3435         ret = btrfs_lookup_extent_ref(NULL, root, start, len, refs);
3436         BUG_ON(ret);
3437
3438 #if 0 // some debugging code in case we see problems here
3439         /* if the refs count is one, it won't get increased again.  But
3440          * if the ref count is > 1, someone may be decreasing it at
3441          * the same time we are.
3442          */
3443         if (*refs != 1) {
3444                 struct extent_buffer *eb = NULL;
3445                 eb = btrfs_find_create_tree_block(root, start, len);
3446                 if (eb)
3447                         btrfs_tree_lock(eb);
3448
3449                 mutex_lock(&root->fs_info->alloc_mutex);
3450                 ret = lookup_extent_ref(NULL, root, start, len, refs);
3451                 BUG_ON(ret);
3452                 mutex_unlock(&root->fs_info->alloc_mutex);
3453
3454                 if (eb) {
3455                         btrfs_tree_unlock(eb);
3456                         free_extent_buffer(eb);
3457                 }
3458                 if (*refs == 1) {
3459                         printk("block %llu went down to one during drop_snap\n",
3460                                (unsigned long long)start);
3461                 }
3462
3463         }
3464 #endif
3465
3466         cond_resched();
3467         return ret;
3468 }
3469
3470 /*
3471  * helper function for drop_snapshot, this walks down the tree dropping ref
3472  * counts as it goes.
3473  */
3474 static int noinline walk_down_tree(struct btrfs_trans_handle *trans,
3475                                    struct btrfs_root *root,
3476                                    struct btrfs_path *path, int *level)
3477 {
3478         u64 root_owner;
3479         u64 root_gen;
3480         u64 bytenr;
3481         u64 ptr_gen;
3482         struct extent_buffer *next;
3483         struct extent_buffer *cur;
3484         struct extent_buffer *parent;
3485         struct btrfs_leaf_ref *ref;
3486         u32 blocksize;
3487         int ret;
3488         u32 refs;
3489
3490         WARN_ON(*level < 0);
3491         WARN_ON(*level >= BTRFS_MAX_LEVEL);
3492         ret = drop_snap_lookup_refcount(root, path->nodes[*level]->start,
3493                                 path->nodes[*level]->len, &refs);
3494         BUG_ON(ret);
3495         if (refs > 1)
3496                 goto out;
3497
3498         /*
3499          * walk down to the last node level and free all the leaves
3500          */
3501         while(*level >= 0) {
3502                 WARN_ON(*level < 0);
3503                 WARN_ON(*level >= BTRFS_MAX_LEVEL);
3504                 cur = path->nodes[*level];
3505
3506                 if (btrfs_header_level(cur) != *level)
3507                         WARN_ON(1);
3508
3509                 if (path->slots[*level] >=
3510                     btrfs_header_nritems(cur))
3511                         break;
3512                 if (*level == 0) {
3513                         ret = btrfs_drop_leaf_ref(trans, root, cur);
3514                         BUG_ON(ret);
3515                         break;
3516                 }
3517                 bytenr = btrfs_node_blockptr(cur, path->slots[*level]);
3518                 ptr_gen = btrfs_node_ptr_generation(cur, path->slots[*level]);
3519                 blocksize = btrfs_level_size(root, *level - 1);
3520
3521                 ret = drop_snap_lookup_refcount(root, bytenr, blocksize, &refs);
3522                 BUG_ON(ret);
3523                 if (refs != 1) {
3524                         parent = path->nodes[*level];
3525                         root_owner = btrfs_header_owner(parent);
3526                         root_gen = btrfs_header_generation(parent);
3527                         path->slots[*level]++;
3528
3529                         ret = __btrfs_free_extent(trans, root, bytenr,
3530                                                 blocksize, parent->start,
3531                                                 root_owner, root_gen,
3532                                                 *level - 1, 1);
3533                         BUG_ON(ret);
3534
3535                         atomic_inc(&root->fs_info->throttle_gen);
3536                         wake_up(&root->fs_info->transaction_throttle);
3537                         cond_resched();
3538
3539                         continue;
3540                 }
3541                 /*
3542                  * at this point, we have a single ref, and since the
3543                  * only place referencing this extent is a dead root
3544                  * the reference count should never go higher.
3545                  * So, we don't need to check it again
3546                  */
3547                 if (*level == 1) {
3548                         ref = btrfs_lookup_leaf_ref(root, bytenr);
3549                         if (ref && ref->generation != ptr_gen) {
3550                                 btrfs_free_leaf_ref(root, ref);
3551                                 ref = NULL;
3552                         }
3553                         if (ref) {
3554                                 ret = cache_drop_leaf_ref(trans, root, ref);
3555                                 BUG_ON(ret);
3556                                 btrfs_remove_leaf_ref(root, ref);
3557                                 btrfs_free_leaf_ref(root, ref);
3558                                 *level = 0;
3559                                 break;
3560                         }
3561                         if (printk_ratelimit()) {
3562                                 printk("leaf ref miss for bytenr %llu\n",
3563                                        (unsigned long long)bytenr);
3564                         }
3565                 }
3566                 next = btrfs_find_tree_block(root, bytenr, blocksize);
3567                 if (!next || !btrfs_buffer_uptodate(next, ptr_gen)) {
3568                         free_extent_buffer(next);
3569
3570                         next = read_tree_block(root, bytenr, blocksize,
3571                                                ptr_gen);
3572                         cond_resched();
3573 #if 0
3574                         /*
3575                          * this is a debugging check and can go away
3576                          * the ref should never go all the way down to 1
3577                          * at this point
3578                          */
3579                         ret = lookup_extent_ref(NULL, root, bytenr, blocksize,
3580                                                 &refs);
3581                         BUG_ON(ret);
3582                         WARN_ON(refs != 1);
3583 #endif
3584                 }
3585                 WARN_ON(*level <= 0);
3586                 if (path->nodes[*level-1])
3587                         free_extent_buffer(path->nodes[*level-1]);
3588                 path->nodes[*level-1] = next;
3589                 *level = btrfs_header_level(next);
3590                 path->slots[*level] = 0;
3591                 cond_resched();
3592         }
3593 out:
3594         WARN_ON(*level < 0);
3595         WARN_ON(*level >= BTRFS_MAX_LEVEL);
3596
3597         if (path->nodes[*level] == root->node) {
3598                 parent = path->nodes[*level];
3599                 bytenr = path->nodes[*level]->start;
3600         } else {
3601                 parent = path->nodes[*level + 1];
3602                 bytenr = btrfs_node_blockptr(parent, path->slots[*level + 1]);
3603         }
3604
3605         blocksize = btrfs_level_size(root, *level);
3606         root_owner = btrfs_header_owner(parent);
3607         root_gen = btrfs_header_generation(parent);
3608
3609         ret = __btrfs_free_extent(trans, root, bytenr, blocksize,
3610                                   parent->start, root_owner, root_gen,
3611                                   *level, 1);
3612         free_extent_buffer(path->nodes[*level]);
3613         path->nodes[*level] = NULL;
3614         *level += 1;
3615         BUG_ON(ret);
3616
3617         cond_resched();
3618         return 0;
3619 }
3620
3621 /*
3622  * helper function for drop_subtree, this function is similar to
3623  * walk_down_tree. The main difference is that it checks reference
3624  * counts while tree blocks are locked.
3625  */
3626 static int noinline walk_down_subtree(struct btrfs_trans_handle *trans,
3627                                       struct btrfs_root *root,
3628                                       struct btrfs_path *path, int *level)
3629 {
3630         struct extent_buffer *next;
3631         struct extent_buffer *cur;
3632         struct extent_buffer *parent;
3633         u64 bytenr;
3634         u64 ptr_gen;
3635         u32 blocksize;
3636         u32 refs;
3637         int ret;
3638
3639         cur = path->nodes[*level];
3640         ret = btrfs_lookup_extent_ref(trans, root, cur->start, cur->len,
3641                                       &refs);
3642         BUG_ON(ret);
3643         if (refs > 1)
3644                 goto out;
3645
3646         while (*level >= 0) {
3647                 cur = path->nodes[*level];
3648                 if (*level == 0) {
3649                         ret = btrfs_drop_leaf_ref(trans, root, cur);
3650                         BUG_ON(ret);
3651                         clean_tree_block(trans, root, cur);
3652                         break;
3653                 }
3654                 if (path->slots[*level] >= btrfs_header_nritems(cur)) {
3655                         clean_tree_block(trans, root, cur);
3656                         break;
3657                 }
3658
3659                 bytenr = btrfs_node_blockptr(cur, path->slots[*level]);
3660                 blocksize = btrfs_level_size(root, *level - 1);
3661                 ptr_gen = btrfs_node_ptr_generation(cur, path->slots[*level]);
3662
3663                 next = read_tree_block(root, bytenr, blocksize, ptr_gen);
3664                 btrfs_tree_lock(next);
3665
3666                 ret = btrfs_lookup_extent_ref(trans, root, bytenr, blocksize,
3667                                               &refs);
3668                 BUG_ON(ret);
3669                 if (refs > 1) {
3670                         parent = path->nodes[*level];
3671                         ret = btrfs_free_extent(trans, root, bytenr,
3672                                         blocksize, parent->start,
3673                                         btrfs_header_owner(parent),
3674                                         btrfs_header_generation(parent),
3675                                         *level - 1, 1);
3676                         BUG_ON(ret);
3677                         path->slots[*level]++;
3678                         btrfs_tree_unlock(next);
3679                         free_extent_buffer(next);
3680                         continue;
3681                 }
3682
3683                 *level = btrfs_header_level(next);
3684                 path->nodes[*level] = next;
3685                 path->slots[*level] = 0;
3686                 path->locks[*level] = 1;
3687                 cond_resched();
3688         }
3689 out:
3690         parent = path->nodes[*level + 1];
3691         bytenr = path->nodes[*level]->start;
3692         blocksize = path->nodes[*level]->len;
3693
3694         ret = btrfs_free_extent(trans, root, bytenr, blocksize,
3695                         parent->start, btrfs_header_owner(parent),
3696                         btrfs_header_generation(parent), *level, 1);
3697         BUG_ON(ret);
3698
3699         if (path->locks[*level]) {
3700                 btrfs_tree_unlock(path->nodes[*level]);
3701                 path->locks[*level] = 0;
3702         }
3703         free_extent_buffer(path->nodes[*level]);
3704         path->nodes[*level] = NULL;
3705         *level += 1;
3706         cond_resched();
3707         return 0;
3708 }
3709
3710 /*
3711  * helper for dropping snapshots.  This walks back up the tree in the path
3712  * to find the first node higher up where we haven't yet gone through
3713  * all the slots
3714  */
3715 static int noinline walk_up_tree(struct btrfs_trans_handle *trans,
3716                                  struct btrfs_root *root,
3717                                  struct btrfs_path *path,
3718                                  int *level, int max_level)
3719 {
3720         u64 root_owner;
3721         u64 root_gen;
3722         struct btrfs_root_item *root_item = &root->root_item;
3723         int i;
3724         int slot;
3725         int ret;
3726
3727         for (i = *level; i < max_level && path->nodes[i]; i++) {
3728                 slot = path->slots[i];
3729                 if (slot < btrfs_header_nritems(path->nodes[i]) - 1) {
3730                         struct extent_buffer *node;
3731                         struct btrfs_disk_key disk_key;
3732                         node = path->nodes[i];
3733                         path->slots[i]++;
3734                         *level = i;
3735                         WARN_ON(*level == 0);
3736                         btrfs_node_key(node, &disk_key, path->slots[i]);
3737                         memcpy(&root_item->drop_progress,
3738                                &disk_key, sizeof(disk_key));
3739                         root_item->drop_level = i;
3740                         return 0;
3741                 } else {
3742                         struct extent_buffer *parent;
3743                         if (path->nodes[*level] == root->node)
3744                                 parent = path->nodes[*level];
3745                         else
3746                                 parent = path->nodes[*level + 1];
3747
3748                         root_owner = btrfs_header_owner(parent);
3749                         root_gen = btrfs_header_generation(parent);
3750
3751                         clean_tree_block(trans, root, path->nodes[*level]);
3752                         ret = btrfs_free_extent(trans, root,
3753                                                 path->nodes[*level]->start,
3754                                                 path->nodes[*level]->len,
3755                                                 parent->start, root_owner,
3756                                                 root_gen, *level, 1);
3757                         BUG_ON(ret);
3758                         if (path->locks[*level]) {
3759                                 btrfs_tree_unlock(path->nodes[*level]);
3760                                 path->locks[*level] = 0;
3761                         }
3762                         free_extent_buffer(path->nodes[*level]);
3763                         path->nodes[*level] = NULL;
3764                         *level = i + 1;
3765                 }
3766         }
3767         return 1;
3768 }
3769
3770 /*
3771  * drop the reference count on the tree rooted at 'snap'.  This traverses
3772  * the tree freeing any blocks that have a ref count of zero after being
3773  * decremented.
3774  */
3775 int btrfs_drop_snapshot(struct btrfs_trans_handle *trans, struct btrfs_root
3776                         *root)
3777 {
3778         int ret = 0;
3779         int wret;
3780         int level;
3781         struct btrfs_path *path;
3782         int i;
3783         int orig_level;
3784         struct btrfs_root_item *root_item = &root->root_item;
3785
3786         WARN_ON(!mutex_is_locked(&root->fs_info->drop_mutex));
3787         path = btrfs_alloc_path();
3788         BUG_ON(!path);
3789
3790         level = btrfs_header_level(root->node);
3791         orig_level = level;
3792         if (btrfs_disk_key_objectid(&root_item->drop_progress) == 0) {
3793                 path->nodes[level] = root->node;
3794                 extent_buffer_get(root->node);
3795                 path->slots[level] = 0;
3796         } else {
3797                 struct btrfs_key key;
3798                 struct btrfs_disk_key found_key;
3799                 struct extent_buffer *node;
3800
3801                 btrfs_disk_key_to_cpu(&key, &root_item->drop_progress);
3802                 level = root_item->drop_level;
3803                 path->lowest_level = level;
3804                 wret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
3805                 if (wret < 0) {
3806                         ret = wret;
3807                         goto out;
3808                 }
3809                 node = path->nodes[level];
3810                 btrfs_node_key(node, &found_key, path->slots[level]);
3811                 WARN_ON(memcmp(&found_key, &root_item->drop_progress,
3812                                sizeof(found_key)));
3813                 /*
3814                  * unlock our path, this is safe because only this
3815                  * function is allowed to delete this snapshot
3816                  */
3817                 for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
3818                         if (path->nodes[i] && path->locks[i]) {
3819                                 path->locks[i] = 0;
3820                                 btrfs_tree_unlock(path->nodes[i]);
3821                         }
3822                 }
3823         }
3824         while(1) {
3825                 wret = walk_down_tree(trans, root, path, &level);
3826                 if (wret > 0)
3827                         break;
3828                 if (wret < 0)
3829                         ret = wret;
3830
3831                 wret = walk_up_tree(trans, root, path, &level,
3832                                     BTRFS_MAX_LEVEL);
3833                 if (wret > 0)
3834                         break;
3835                 if (wret < 0)
3836                         ret = wret;
3837                 if (trans->transaction->in_commit) {
3838                         ret = -EAGAIN;
3839                         break;
3840                 }
3841                 atomic_inc(&root->fs_info->throttle_gen);
3842                 wake_up(&root->fs_info->transaction_throttle);
3843         }
3844         for (i = 0; i <= orig_level; i++) {
3845                 if (path->nodes[i]) {
3846                         free_extent_buffer(path->nodes[i]);
3847                         path->nodes[i] = NULL;
3848                 }
3849         }
3850 out:
3851         btrfs_free_path(path);
3852         return ret;
3853 }
3854
3855 int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
3856                         struct btrfs_root *root,
3857                         struct extent_buffer *node,
3858                         struct extent_buffer *parent)
3859 {
3860         struct btrfs_path *path;
3861         int level;
3862         int parent_level;
3863         int ret = 0;
3864         int wret;
3865
3866         path = btrfs_alloc_path();
3867         BUG_ON(!path);
3868
3869         BUG_ON(!btrfs_tree_locked(parent));
3870         parent_level = btrfs_header_level(parent);
3871         extent_buffer_get(parent);
3872         path->nodes[parent_level] = parent;
3873         path->slots[parent_level] = btrfs_header_nritems(parent);
3874
3875         BUG_ON(!btrfs_tree_locked(node));
3876         level = btrfs_header_level(node);
3877         extent_buffer_get(node);
3878         path->nodes[level] = node;
3879         path->slots[level] = 0;
3880
3881         while (1) {
3882                 wret = walk_down_subtree(trans, root, path, &level);
3883                 if (wret < 0)
3884                         ret = wret;
3885                 if (wret != 0)
3886                         break;
3887
3888                 wret = walk_up_tree(trans, root, path, &level, parent_level);
3889                 if (wret < 0)
3890                         ret = wret;
3891                 if (wret != 0)
3892                         break;
3893         }
3894
3895         btrfs_free_path(path);
3896         return ret;
3897 }
3898
3899 static unsigned long calc_ra(unsigned long start, unsigned long last,
3900                              unsigned long nr)
3901 {
3902         return min(last, start + nr - 1);
3903 }
3904
3905 static int noinline relocate_inode_pages(struct inode *inode, u64 start,
3906                                          u64 len)
3907 {
3908         u64 page_start;
3909         u64 page_end;
3910         unsigned long first_index;
3911         unsigned long last_index;
3912         unsigned long i;
3913         struct page *page;
3914         struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
3915         struct file_ra_state *ra;
3916         struct btrfs_ordered_extent *ordered;
3917         unsigned int total_read = 0;
3918         unsigned int total_dirty = 0;
3919         int ret = 0;
3920
3921         ra = kzalloc(sizeof(*ra), GFP_NOFS);
3922
3923         mutex_lock(&inode->i_mutex);
3924         first_index = start >> PAGE_CACHE_SHIFT;
3925         last_index = (start + len - 1) >> PAGE_CACHE_SHIFT;
3926
3927         /* make sure the dirty trick played by the caller work */
3928         ret = invalidate_inode_pages2_range(inode->i_mapping,
3929                                             first_index, last_index);
3930         if (ret)
3931                 goto out_unlock;
3932
3933         file_ra_state_init(ra, inode->i_mapping);
3934
3935         for (i = first_index ; i <= last_index; i++) {
3936                 if (total_read % ra->ra_pages == 0) {
3937                         btrfs_force_ra(inode->i_mapping, ra, NULL, i,
3938                                        calc_ra(i, last_index, ra->ra_pages));
3939                 }
3940                 total_read++;
3941 again:
3942                 if (((u64)i << PAGE_CACHE_SHIFT) > i_size_read(inode))
3943                         BUG_ON(1);
3944                 page = grab_cache_page(inode->i_mapping, i);
3945                 if (!page) {
3946                         ret = -ENOMEM;
3947                         goto out_unlock;
3948                 }
3949                 if (!PageUptodate(page)) {
3950                         btrfs_readpage(NULL, page);
3951                         lock_page(page);
3952                         if (!PageUptodate(page)) {
3953                                 unlock_page(page);
3954                                 page_cache_release(page);
3955                                 ret = -EIO;
3956                                 goto out_unlock;
3957                         }
3958                 }
3959                 wait_on_page_writeback(page);
3960
3961                 page_start = (u64)page->index << PAGE_CACHE_SHIFT;
3962                 page_end = page_start + PAGE_CACHE_SIZE - 1;
3963                 lock_extent(io_tree, page_start, page_end, GFP_NOFS);
3964
3965                 ordered = btrfs_lookup_ordered_extent(inode, page_start);
3966                 if (ordered) {
3967                         unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
3968                         unlock_page(page);
3969                         page_cache_release(page);
3970                         btrfs_start_ordered_extent(inode, ordered, 1);
3971                         btrfs_put_ordered_extent(ordered);
3972                         goto again;
3973                 }
3974                 set_page_extent_mapped(page);
3975
3976                 btrfs_set_extent_delalloc(inode, page_start, page_end);
3977                 if (i == first_index)
3978                         set_extent_bits(io_tree, page_start, page_end,
3979                                         EXTENT_BOUNDARY, GFP_NOFS);
3980
3981                 set_page_dirty(page);
3982                 total_dirty++;
3983
3984                 unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
3985                 unlock_page(page);
3986                 page_cache_release(page);
3987         }
3988
3989 out_unlock:
3990         kfree(ra);
3991         mutex_unlock(&inode->i_mutex);
3992         balance_dirty_pages_ratelimited_nr(inode->i_mapping, total_dirty);
3993         return ret;
3994 }
3995
3996 static int noinline relocate_data_extent(struct inode *reloc_inode,
3997                                          struct btrfs_key *extent_key,
3998                                          u64 offset)
3999 {
4000         struct btrfs_root *root = BTRFS_I(reloc_inode)->root;
4001         struct extent_map_tree *em_tree = &BTRFS_I(reloc_inode)->extent_tree;
4002         struct extent_map *em;
4003         u64 start = extent_key->objectid - offset;
4004         u64 end = start + extent_key->offset - 1;
4005
4006         em = alloc_extent_map(GFP_NOFS);
4007         BUG_ON(!em || IS_ERR(em));
4008
4009         em->start = start;
4010         em->len = extent_key->offset;
4011         em->block_len = extent_key->offset;
4012         em->block_start = extent_key->objectid;
4013         em->bdev = root->fs_info->fs_devices->latest_bdev;
4014         set_bit(EXTENT_FLAG_PINNED, &em->flags);
4015
4016         /* setup extent map to cheat btrfs_readpage */
4017         lock_extent(&BTRFS_I(reloc_inode)->io_tree, start, end, GFP_NOFS);
4018         while (1) {
4019                 int ret;
4020                 spin_lock(&em_tree->lock);
4021                 ret = add_extent_mapping(em_tree, em);
4022                 spin_unlock(&em_tree->lock);
4023                 if (ret != -EEXIST) {
4024                         free_extent_map(em);
4025                         break;
4026                 }
4027                 btrfs_drop_extent_cache(reloc_inode, start, end, 0);
4028         }
4029         unlock_extent(&BTRFS_I(reloc_inode)->io_tree, start, end, GFP_NOFS);
4030
4031         return relocate_inode_pages(reloc_inode, start, extent_key->offset);
4032 }
4033
4034 struct btrfs_ref_path {
4035         u64 extent_start;
4036         u64 nodes[BTRFS_MAX_LEVEL];
4037         u64 root_objectid;
4038         u64 root_generation;
4039         u64 owner_objectid;
4040         u32 num_refs;
4041         int lowest_level;
4042         int current_level;
4043         int shared_level;
4044
4045         struct btrfs_key node_keys[BTRFS_MAX_LEVEL];
4046         u64 new_nodes[BTRFS_MAX_LEVEL];
4047 };
4048
4049 struct disk_extent {
4050         u64 ram_bytes;
4051         u64 disk_bytenr;
4052         u64 disk_num_bytes;
4053         u64 offset;
4054         u64 num_bytes;
4055         u8 compression;
4056         u8 encryption;
4057         u16 other_encoding;
4058 };
4059
4060 static int is_cowonly_root(u64 root_objectid)
4061 {
4062         if (root_objectid == BTRFS_ROOT_TREE_OBJECTID ||
4063             root_objectid == BTRFS_EXTENT_TREE_OBJECTID ||
4064             root_objectid == BTRFS_CHUNK_TREE_OBJECTID ||
4065             root_objectid == BTRFS_DEV_TREE_OBJECTID ||
4066             root_objectid == BTRFS_TREE_LOG_OBJECTID)
4067                 return 1;
4068         return 0;
4069 }
4070
4071 static int noinline __next_ref_path(struct btrfs_trans_handle *trans,
4072                                     struct btrfs_root *extent_root,
4073                                     struct btrfs_ref_path *ref_path,
4074                                     int first_time)
4075 {
4076         struct extent_buffer *leaf;
4077         struct btrfs_path *path;
4078         struct btrfs_extent_ref *ref;
4079         struct btrfs_key key;
4080         struct btrfs_key found_key;
4081         u64 bytenr;
4082         u32 nritems;
4083         int level;
4084         int ret = 1;
4085
4086         path = btrfs_alloc_path();
4087         if (!path)
4088                 return -ENOMEM;
4089
4090         if (first_time) {
4091                 ref_path->lowest_level = -1;
4092                 ref_path->current_level = -1;
4093                 ref_path->shared_level = -1;
4094                 goto walk_up;
4095         }
4096 walk_down:
4097         level = ref_path->current_level - 1;
4098         while (level >= -1) {
4099                 u64 parent;
4100                 if (level < ref_path->lowest_level)
4101                         break;
4102
4103                 if (level >= 0) {
4104                         bytenr = ref_path->nodes[level];
4105                 } else {
4106                         bytenr = ref_path->extent_start;
4107                 }
4108                 BUG_ON(bytenr == 0);
4109
4110                 parent = ref_path->nodes[level + 1];
4111                 ref_path->nodes[level + 1] = 0;
4112                 ref_path->current_level = level;
4113                 BUG_ON(parent == 0);
4114
4115                 key.objectid = bytenr;
4116                 key.offset = parent + 1;
4117                 key.type = BTRFS_EXTENT_REF_KEY;
4118
4119                 ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 0);
4120                 if (ret < 0)
4121                         goto out;
4122                 BUG_ON(ret == 0);
4123
4124                 leaf = path->nodes[0];
4125                 nritems = btrfs_header_nritems(leaf);
4126                 if (path->slots[0] >= nritems) {
4127                         ret = btrfs_next_leaf(extent_root, path);
4128                         if (ret < 0)
4129                                 goto out;
4130                         if (ret > 0)
4131                                 goto next;
4132                         leaf = path->nodes[0];
4133                 }
4134
4135                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
4136                 if (found_key.objectid == bytenr &&
4137                     found_key.type == BTRFS_EXTENT_REF_KEY) {
4138                         if (level < ref_path->shared_level)
4139                                 ref_path->shared_level = level;
4140                         goto found;
4141                 }
4142 next:
4143                 level--;
4144                 btrfs_release_path(extent_root, path);
4145                 cond_resched();
4146         }
4147         /* reached lowest level */
4148         ret = 1;
4149         goto out;
4150 walk_up:
4151         level = ref_path->current_level;
4152         while (level < BTRFS_MAX_LEVEL - 1) {
4153                 u64 ref_objectid;
4154                 if (level >= 0) {
4155                         bytenr = ref_path->nodes[level];
4156                 } else {
4157                         bytenr = ref_path->extent_start;
4158                 }
4159                 BUG_ON(bytenr == 0);
4160
4161                 key.objectid = bytenr;
4162                 key.offset = 0;
4163                 key.type = BTRFS_EXTENT_REF_KEY;
4164
4165                 ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 0);
4166                 if (ret < 0)
4167                         goto out;
4168
4169                 leaf = path->nodes[0];
4170                 nritems = btrfs_header_nritems(leaf);
4171                 if (path->slots[0] >= nritems) {
4172                         ret = btrfs_next_leaf(extent_root, path);
4173                         if (ret < 0)
4174                                 goto out;
4175                         if (ret > 0) {
4176                                 /* the extent was freed by someone */
4177                                 if (ref_path->lowest_level == level)
4178                                         goto out;
4179                                 btrfs_release_path(extent_root, path);
4180                                 goto walk_down;
4181                         }
4182                         leaf = path->nodes[0];
4183                 }
4184
4185                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
4186                 if (found_key.objectid != bytenr ||
4187                                 found_key.type != BTRFS_EXTENT_REF_KEY) {
4188                         /* the extent was freed by someone */
4189                         if (ref_path->lowest_level == level) {
4190                                 ret = 1;
4191                                 goto out;
4192                         }
4193                         btrfs_release_path(extent_root, path);
4194                         goto walk_down;
4195                 }
4196 found:
4197                 ref = btrfs_item_ptr(leaf, path->slots[0],
4198                                 struct btrfs_extent_ref);
4199                 ref_objectid = btrfs_ref_objectid(leaf, ref);
4200                 if (ref_objectid < BTRFS_FIRST_FREE_OBJECTID) {
4201                         if (first_time) {
4202                                 level = (int)ref_objectid;
4203                                 BUG_ON(level >= BTRFS_MAX_LEVEL);
4204                                 ref_path->lowest_level = level;
4205                                 ref_path->current_level = level;
4206                                 ref_path->nodes[level] = bytenr;
4207                         } else {
4208                                 WARN_ON(ref_objectid != level);
4209                         }
4210                 } else {
4211                         WARN_ON(level != -1);
4212                 }
4213                 first_time = 0;
4214
4215                 if (ref_path->lowest_level == level) {
4216                         ref_path->owner_objectid = ref_objectid;
4217                         ref_path->num_refs = btrfs_ref_num_refs(leaf, ref);
4218                 }
4219
4220                 /*
4221                  * the block is tree root or the block isn't in reference
4222                  * counted tree.
4223                  */
4224                 if (found_key.objectid == found_key.offset ||
4225                     is_cowonly_root(btrfs_ref_root(leaf, ref))) {
4226                         ref_path->root_objectid = btrfs_ref_root(leaf, ref);
4227                         ref_path->root_generation =
4228                                 btrfs_ref_generation(leaf, ref);
4229                         if (level < 0) {
4230                                 /* special reference from the tree log */
4231                                 ref_path->nodes[0] = found_key.offset;
4232                                 ref_path->current_level = 0;
4233                         }
4234                         ret = 0;
4235                         goto out;
4236                 }
4237
4238                 level++;
4239                 BUG_ON(ref_path->nodes[level] != 0);
4240                 ref_path->nodes[level] = found_key.offset;
4241                 ref_path->current_level = level;
4242
4243                 /*
4244                  * the reference was created in the running transaction,
4245                  * no need to continue walking up.
4246                  */
4247                 if (btrfs_ref_generation(leaf, ref) == trans->transid) {
4248                         ref_path->root_objectid = btrfs_ref_root(leaf, ref);
4249                         ref_path->root_generation =
4250                                 btrfs_ref_generation(leaf, ref);
4251                         ret = 0;
4252                         goto out;
4253                 }
4254
4255                 btrfs_release_path(extent_root, path);
4256                 cond_resched();
4257         }
4258         /* reached max tree level, but no tree root found. */
4259         BUG();
4260 out:
4261         btrfs_free_path(path);
4262         return ret;
4263 }
4264
4265 static int btrfs_first_ref_path(struct btrfs_trans_handle *trans,
4266                                 struct btrfs_root *extent_root,
4267                                 struct btrfs_ref_path *ref_path,
4268                                 u64 extent_start)
4269 {
4270         memset(ref_path, 0, sizeof(*ref_path));
4271         ref_path->extent_start = extent_start;
4272
4273         return __next_ref_path(trans, extent_root, ref_path, 1);
4274 }
4275
4276 static int btrfs_next_ref_path(struct btrfs_trans_handle *trans,
4277                                struct btrfs_root *extent_root,
4278                                struct btrfs_ref_path *ref_path)
4279 {
4280         return __next_ref_path(trans, extent_root, ref_path, 0);
4281 }
4282
4283 static int noinline get_new_locations(struct inode *reloc_inode,
4284                                       struct btrfs_key *extent_key,
4285                                       u64 offset, int no_fragment,
4286                                       struct disk_extent **extents,
4287                                       int *nr_extents)
4288 {
4289         struct btrfs_root *root = BTRFS_I(reloc_inode)->root;
4290         struct btrfs_path *path;
4291         struct btrfs_file_extent_item *fi;
4292         struct extent_buffer *leaf;
4293         struct disk_extent *exts = *extents;
4294         struct btrfs_key found_key;
4295         u64 cur_pos;
4296         u64 last_byte;
4297         u32 nritems;
4298         int nr = 0;
4299         int max = *nr_extents;
4300         int ret;
4301
4302         WARN_ON(!no_fragment && *extents);
4303         if (!exts) {
4304                 max = 1;
4305                 exts = kmalloc(sizeof(*exts) * max, GFP_NOFS);
4306                 if (!exts)
4307                         return -ENOMEM;
4308         }
4309
4310         path = btrfs_alloc_path();
4311         BUG_ON(!path);
4312
4313         cur_pos = extent_key->objectid - offset;
4314         last_byte = extent_key->objectid + extent_key->offset;
4315         ret = btrfs_lookup_file_extent(NULL, root, path, reloc_inode->i_ino,
4316                                        cur_pos, 0);
4317         if (ret < 0)
4318                 goto out;
4319         if (ret > 0) {
4320                 ret = -ENOENT;
4321                 goto out;
4322         }
4323
4324         while (1) {
4325                 leaf = path->nodes[0];
4326                 nritems = btrfs_header_nritems(leaf);
4327                 if (path->slots[0] >= nritems) {
4328                         ret = btrfs_next_leaf(root, path);
4329                         if (ret < 0)
4330                                 goto out;
4331                         if (ret > 0)
4332                                 break;
4333                         leaf = path->nodes[0];
4334                 }
4335
4336                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
4337                 if (found_key.offset != cur_pos ||
4338                     found_key.type != BTRFS_EXTENT_DATA_KEY ||
4339                     found_key.objectid != reloc_inode->i_ino)
4340                         break;
4341
4342                 fi = btrfs_item_ptr(leaf, path->slots[0],
4343                                     struct btrfs_file_extent_item);
4344                 if (btrfs_file_extent_type(leaf, fi) !=
4345                     BTRFS_FILE_EXTENT_REG ||
4346                     btrfs_file_extent_disk_bytenr(leaf, fi) == 0)
4347                         break;
4348
4349                 if (nr == max) {
4350                         struct disk_extent *old = exts;
4351                         max *= 2;
4352                         exts = kzalloc(sizeof(*exts) * max, GFP_NOFS);
4353                         memcpy(exts, old, sizeof(*exts) * nr);
4354                         if (old != *extents)
4355                                 kfree(old);
4356                 }
4357
4358                 exts[nr].disk_bytenr =
4359                         btrfs_file_extent_disk_bytenr(leaf, fi);
4360                 exts[nr].disk_num_bytes =
4361                         btrfs_file_extent_disk_num_bytes(leaf, fi);
4362                 exts[nr].offset = btrfs_file_extent_offset(leaf, fi);
4363                 exts[nr].num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
4364                 exts[nr].ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
4365                 exts[nr].compression = btrfs_file_extent_compression(leaf, fi);
4366                 exts[nr].encryption = btrfs_file_extent_encryption(leaf, fi);
4367                 exts[nr].other_encoding = btrfs_file_extent_other_encoding(leaf,
4368                                                                            fi);
4369                 BUG_ON(exts[nr].offset > 0);
4370                 BUG_ON(exts[nr].compression || exts[nr].encryption);
4371                 BUG_ON(exts[nr].num_bytes != exts[nr].disk_num_bytes);
4372
4373                 cur_pos += exts[nr].num_bytes;
4374                 nr++;
4375
4376                 if (cur_pos + offset >= last_byte)
4377                         break;
4378
4379                 if (no_fragment) {
4380                         ret = 1;
4381                         goto out;
4382                 }
4383                 path->slots[0]++;
4384         }
4385
4386         WARN_ON(cur_pos + offset > last_byte);
4387         if (cur_pos + offset < last_byte) {
4388                 ret = -ENOENT;
4389                 goto out;
4390         }
4391         ret = 0;
4392 out:
4393         btrfs_free_path(path);
4394         if (ret) {
4395                 if (exts != *extents)
4396                         kfree(exts);
4397         } else {
4398                 *extents = exts;
4399                 *nr_extents = nr;
4400         }
4401         return ret;
4402 }
4403
4404 static int noinline replace_one_extent(struct btrfs_trans_handle *trans,
4405                                         struct btrfs_root *root,
4406                                         struct btrfs_path *path,
4407                                         struct btrfs_key *extent_key,
4408                                         struct btrfs_key *leaf_key,
4409                                         struct btrfs_ref_path *ref_path,
4410                                         struct disk_extent *new_extents,
4411                                         int nr_extents)
4412 {
4413         struct extent_buffer *leaf;
4414         struct btrfs_file_extent_item *fi;
4415         struct inode *inode = NULL;
4416         struct btrfs_key key;
4417         u64 lock_start = 0;
4418         u64 lock_end = 0;
4419         u64 num_bytes;
4420         u64 ext_offset;
4421         u64 first_pos;
4422         u32 nritems;
4423         int nr_scaned = 0;
4424         int extent_locked = 0;
4425         int extent_type;
4426         int ret;
4427
4428         memcpy(&key, leaf_key, sizeof(key));
4429         first_pos = INT_LIMIT(loff_t) - extent_key->offset;
4430         if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS) {
4431                 if (key.objectid < ref_path->owner_objectid ||
4432                     (key.objectid == ref_path->owner_objectid &&
4433                      key.type < BTRFS_EXTENT_DATA_KEY)) {
4434                         key.objectid = ref_path->owner_objectid;
4435                         key.type = BTRFS_EXTENT_DATA_KEY;
4436                         key.offset = 0;
4437                 }
4438         }
4439
4440         while (1) {
4441                 ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
4442                 if (ret < 0)
4443                         goto out;
4444
4445                 leaf = path->nodes[0];
4446                 nritems = btrfs_header_nritems(leaf);
4447 next:
4448                 if (extent_locked && ret > 0) {
4449                         /*
4450                          * the file extent item was modified by someone
4451                          * before the extent got locked.
4452                          */
4453                         unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
4454                                       lock_end, GFP_NOFS);
4455                         extent_locked = 0;
4456                 }
4457
4458                 if (path->slots[0] >= nritems) {
4459                         if (++nr_scaned > 2)
4460                                 break;
4461
4462                         BUG_ON(extent_locked);
4463                         ret = btrfs_next_leaf(root, path);
4464                         if (ret < 0)
4465                                 goto out;
4466                         if (ret > 0)
4467                                 break;
4468                         leaf = path->nodes[0];
4469                         nritems = btrfs_header_nritems(leaf);
4470                 }
4471
4472                 btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
4473
4474                 if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS) {
4475                         if ((key.objectid > ref_path->owner_objectid) ||
4476                             (key.objectid == ref_path->owner_objectid &&
4477                              key.type > BTRFS_EXTENT_DATA_KEY) ||
4478                             (key.offset >= first_pos + extent_key->offset))
4479                                 break;
4480                 }
4481
4482                 if (inode && key.objectid != inode->i_ino) {
4483                         BUG_ON(extent_locked);
4484                         btrfs_release_path(root, path);
4485                         mutex_unlock(&inode->i_mutex);
4486                         iput(inode);
4487                         inode = NULL;
4488                         continue;
4489                 }
4490
4491                 if (key.type != BTRFS_EXTENT_DATA_KEY) {
4492                         path->slots[0]++;
4493                         ret = 1;
4494                         goto next;
4495                 }
4496                 fi = btrfs_item_ptr(leaf, path->slots[0],
4497                                     struct btrfs_file_extent_item);
4498                 extent_type = btrfs_file_extent_type(leaf, fi);
4499                 if ((extent_type != BTRFS_FILE_EXTENT_REG &&
4500                      extent_type != BTRFS_FILE_EXTENT_PREALLOC) ||
4501                     (btrfs_file_extent_disk_bytenr(leaf, fi) !=
4502                      extent_key->objectid)) {
4503                         path->slots[0]++;
4504                         ret = 1;
4505                         goto next;
4506                 }
4507
4508                 num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
4509                 ext_offset = btrfs_file_extent_offset(leaf, fi);
4510
4511                 if (first_pos > key.offset - ext_offset)
4512                         first_pos = key.offset - ext_offset;
4513
4514                 if (!extent_locked) {
4515                         lock_start = key.offset;
4516                         lock_end = lock_start + num_bytes - 1;
4517                 } else {
4518                         if (lock_start > key.offset ||
4519                             lock_end + 1 < key.offset + num_bytes) {
4520                                 unlock_extent(&BTRFS_I(inode)->io_tree,
4521                                               lock_start, lock_end, GFP_NOFS);
4522                                 extent_locked = 0;
4523                         }
4524                 }
4525
4526                 if (!inode) {
4527                         btrfs_release_path(root, path);
4528
4529                         inode = btrfs_iget_locked(root->fs_info->sb,
4530                                                   key.objectid, root);
4531                         if (inode->i_state & I_NEW) {
4532                                 BTRFS_I(inode)->root = root;
4533                                 BTRFS_I(inode)->location.objectid =
4534                                         key.objectid;
4535                                 BTRFS_I(inode)->location.type =
4536                                         BTRFS_INODE_ITEM_KEY;
4537                                 BTRFS_I(inode)->location.offset = 0;
4538                                 btrfs_read_locked_inode(inode);
4539                                 unlock_new_inode(inode);
4540                         }
4541                         /*
4542                          * some code call btrfs_commit_transaction while
4543                          * holding the i_mutex, so we can't use mutex_lock
4544                          * here.
4545                          */
4546                         if (is_bad_inode(inode) ||
4547                             !mutex_trylock(&inode->i_mutex)) {
4548                                 iput(inode);
4549                                 inode = NULL;
4550                                 key.offset = (u64)-1;
4551                                 goto skip;
4552                         }
4553                 }
4554
4555                 if (!extent_locked) {
4556                         struct btrfs_ordered_extent *ordered;
4557
4558                         btrfs_release_path(root, path);
4559
4560                         lock_extent(&BTRFS_I(inode)->io_tree, lock_start,
4561                                     lock_end, GFP_NOFS);
4562                         ordered = btrfs_lookup_first_ordered_extent(inode,
4563                                                                     lock_end);
4564                         if (ordered &&
4565                             ordered->file_offset <= lock_end &&
4566                             ordered->file_offset + ordered->len > lock_start) {
4567                                 unlock_extent(&BTRFS_I(inode)->io_tree,
4568                                               lock_start, lock_end, GFP_NOFS);
4569                                 btrfs_start_ordered_extent(inode, ordered, 1);
4570                                 btrfs_put_ordered_extent(ordered);
4571                                 key.offset += num_bytes;
4572                                 goto skip;
4573                         }
4574                         if (ordered)
4575                                 btrfs_put_ordered_extent(ordered);
4576
4577                         extent_locked = 1;
4578                         continue;
4579                 }
4580
4581                 if (nr_extents == 1) {
4582                         /* update extent pointer in place */
4583                         btrfs_set_file_extent_disk_bytenr(leaf, fi,
4584                                                 new_extents[0].disk_bytenr);
4585                         btrfs_set_file_extent_disk_num_bytes(leaf, fi,
4586                                                 new_extents[0].disk_num_bytes);
4587                         btrfs_mark_buffer_dirty(leaf);
4588
4589                         btrfs_drop_extent_cache(inode, key.offset,
4590                                                 key.offset + num_bytes - 1, 0);
4591
4592                         ret = btrfs_inc_extent_ref(trans, root,
4593                                                 new_extents[0].disk_bytenr,
4594                                                 new_extents[0].disk_num_bytes,
4595                                                 leaf->start,
4596                                                 root->root_key.objectid,
4597                                                 trans->transid,
4598                                                 key.objectid);
4599                         BUG_ON(ret);
4600
4601                         ret = btrfs_free_extent(trans, root,
4602                                                 extent_key->objectid,
4603                                                 extent_key->offset,
4604                                                 leaf->start,
4605                                                 btrfs_header_owner(leaf),
4606                                                 btrfs_header_generation(leaf),
4607                                                 key.objectid, 0);
4608                         BUG_ON(ret);
4609
4610                         btrfs_release_path(root, path);
4611                         key.offset += num_bytes;
4612                 } else {
4613                         BUG_ON(1);
4614 #if 0
4615                         u64 alloc_hint;
4616                         u64 extent_len;
4617                         int i;
4618                         /*
4619                          * drop old extent pointer at first, then insert the
4620                          * new pointers one bye one
4621                          */
4622                         btrfs_release_path(root, path);
4623                         ret = btrfs_drop_extents(trans, root, inode, key.offset,
4624                                                  key.offset + num_bytes,
4625                                                  key.offset, &alloc_hint);
4626                         BUG_ON(ret);
4627
4628                         for (i = 0; i < nr_extents; i++) {
4629                                 if (ext_offset >= new_extents[i].num_bytes) {
4630                                         ext_offset -= new_extents[i].num_bytes;
4631                                         continue;
4632                                 }
4633                                 extent_len = min(new_extents[i].num_bytes -
4634                                                  ext_offset, num_bytes);
4635
4636                                 ret = btrfs_insert_empty_item(trans, root,
4637                                                               path, &key,
4638                                                               sizeof(*fi));
4639                                 BUG_ON(ret);
4640
4641                                 leaf = path->nodes[0];
4642                                 fi = btrfs_item_ptr(leaf, path->slots[0],
4643                                                 struct btrfs_file_extent_item);
4644                                 btrfs_set_file_extent_generation(leaf, fi,
4645                                                         trans->transid);
4646                                 btrfs_set_file_extent_type(leaf, fi,
4647                                                         BTRFS_FILE_EXTENT_REG);
4648                                 btrfs_set_file_extent_disk_bytenr(leaf, fi,
4649                                                 new_extents[i].disk_bytenr);
4650                                 btrfs_set_file_extent_disk_num_bytes(leaf, fi,
4651                                                 new_extents[i].disk_num_bytes);
4652                                 btrfs_set_file_extent_ram_bytes(leaf, fi,
4653                                                 new_extents[i].ram_bytes);
4654
4655                                 btrfs_set_file_extent_compression(leaf, fi,
4656                                                 new_extents[i].compression);
4657                                 btrfs_set_file_extent_encryption(leaf, fi,
4658                                                 new_extents[i].encryption);
4659                                 btrfs_set_file_extent_other_encoding(leaf, fi,
4660                                                 new_extents[i].other_encoding);
4661
4662                                 btrfs_set_file_extent_num_bytes(leaf, fi,
4663                                                         extent_len);
4664                                 ext_offset += new_extents[i].offset;
4665                                 btrfs_set_file_extent_offset(leaf, fi,
4666                                                         ext_offset);
4667                                 btrfs_mark_buffer_dirty(leaf);
4668
4669                                 btrfs_drop_extent_cache(inode, key.offset,
4670                                                 key.offset + extent_len - 1, 0);
4671
4672                                 ret = btrfs_inc_extent_ref(trans, root,
4673                                                 new_extents[i].disk_bytenr,
4674                                                 new_extents[i].disk_num_bytes,
4675                                                 leaf->start,
4676                                                 root->root_key.objectid,
4677                                                 trans->transid, key.objectid);
4678                                 BUG_ON(ret);
4679                                 btrfs_release_path(root, path);
4680
4681                                 inode_add_bytes(inode, extent_len);
4682
4683                                 ext_offset = 0;
4684                                 num_bytes -= extent_len;
4685                                 key.offset += extent_len;
4686
4687                                 if (num_bytes == 0)
4688                                         break;
4689                         }
4690                         BUG_ON(i >= nr_extents);
4691 #endif
4692                 }
4693
4694                 if (extent_locked) {
4695                         unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
4696                                       lock_end, GFP_NOFS);
4697                         extent_locked = 0;
4698                 }
4699 skip:
4700                 if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS &&
4701                     key.offset >= first_pos + extent_key->offset)
4702                         break;
4703
4704                 cond_resched();
4705         }
4706         ret = 0;
4707 out:
4708         btrfs_release_path(root, path);
4709         if (inode) {
4710                 mutex_unlock(&inode->i_mutex);
4711                 if (extent_locked) {
4712                         unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
4713                                       lock_end, GFP_NOFS);
4714                 }
4715                 iput(inode);
4716         }
4717         return ret;
4718 }
4719
4720 int btrfs_reloc_tree_cache_ref(struct btrfs_trans_handle *trans,
4721                                struct btrfs_root *root,
4722                                struct extent_buffer *buf, u64 orig_start)
4723 {
4724         int level;
4725         int ret;
4726
4727         BUG_ON(btrfs_header_generation(buf) != trans->transid);
4728         BUG_ON(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID);
4729
4730         level = btrfs_header_level(buf);
4731         if (level == 0) {
4732                 struct btrfs_leaf_ref *ref;
4733                 struct btrfs_leaf_ref *orig_ref;
4734
4735                 orig_ref = btrfs_lookup_leaf_ref(root, orig_start);
4736                 if (!orig_ref)
4737                         return -ENOENT;
4738
4739                 ref = btrfs_alloc_leaf_ref(root, orig_ref->nritems);
4740                 if (!ref) {
4741                         btrfs_free_leaf_ref(root, orig_ref);
4742                         return -ENOMEM;
4743                 }
4744
4745                 ref->nritems = orig_ref->nritems;
4746                 memcpy(ref->extents, orig_ref->extents,
4747                         sizeof(ref->extents[0]) * ref->nritems);
4748
4749                 btrfs_free_leaf_ref(root, orig_ref);
4750
4751                 ref->root_gen = trans->transid;
4752                 ref->bytenr = buf->start;
4753                 ref->owner = btrfs_header_owner(buf);
4754                 ref->generation = btrfs_header_generation(buf);
4755                 ret = btrfs_add_leaf_ref(root, ref, 0);
4756                 WARN_ON(ret);
4757                 btrfs_free_leaf_ref(root, ref);
4758         }
4759         return 0;
4760 }
4761
4762 static int noinline invalidate_extent_cache(struct btrfs_root *root,
4763                                         struct extent_buffer *leaf,
4764                                         struct btrfs_block_group_cache *group,
4765                                         struct btrfs_root *target_root)
4766 {
4767         struct btrfs_key key;
4768         struct inode *inode = NULL;
4769         struct btrfs_file_extent_item *fi;
4770         u64 num_bytes;
4771         u64 skip_objectid = 0;
4772         u32 nritems;
4773         u32 i;
4774
4775         nritems = btrfs_header_nritems(leaf);
4776         for (i = 0; i < nritems; i++) {
4777                 btrfs_item_key_to_cpu(leaf, &key, i);
4778                 if (key.objectid == skip_objectid ||
4779                     key.type != BTRFS_EXTENT_DATA_KEY)
4780                         continue;
4781                 fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
4782                 if (btrfs_file_extent_type(leaf, fi) ==
4783                     BTRFS_FILE_EXTENT_INLINE)
4784                         continue;
4785                 if (btrfs_file_extent_disk_bytenr(leaf, fi) == 0)
4786                         continue;
4787                 if (!inode || inode->i_ino != key.objectid) {
4788                         iput(inode);
4789                         inode = btrfs_ilookup(target_root->fs_info->sb,
4790                                               key.objectid, target_root, 1);
4791                 }
4792                 if (!inode) {
4793                         skip_objectid = key.objectid;
4794                         continue;
4795                 }
4796                 num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
4797
4798                 lock_extent(&BTRFS_I(inode)->io_tree, key.offset,
4799                             key.offset + num_bytes - 1, GFP_NOFS);
4800                 btrfs_drop_extent_cache(inode, key.offset,
4801                                         key.offset + num_bytes - 1, 1);
4802                 unlock_extent(&BTRFS_I(inode)->io_tree, key.offset,
4803                               key.offset + num_bytes - 1, GFP_NOFS);
4804                 cond_resched();
4805         }
4806         iput(inode);
4807         return 0;
4808 }
4809
4810 static int noinline replace_extents_in_leaf(struct btrfs_trans_handle *trans,
4811                                         struct btrfs_root *root,
4812                                         struct extent_buffer *leaf,
4813                                         struct btrfs_block_group_cache *group,
4814                                         struct inode *reloc_inode)
4815 {
4816         struct btrfs_key key;
4817         struct btrfs_key extent_key;
4818         struct btrfs_file_extent_item *fi;
4819         struct btrfs_leaf_ref *ref;
4820         struct disk_extent *new_extent;
4821         u64 bytenr;
4822         u64 num_bytes;
4823         u32 nritems;
4824         u32 i;
4825         int ext_index;
4826         int nr_extent;
4827         int ret;
4828
4829         new_extent = kmalloc(sizeof(*new_extent), GFP_NOFS);
4830         BUG_ON(!new_extent);
4831
4832         ref = btrfs_lookup_leaf_ref(root, leaf->start);
4833         BUG_ON(!ref);
4834
4835         ext_index = -1;
4836         nritems = btrfs_header_nritems(leaf);
4837         for (i = 0; i < nritems; i++) {
4838                 btrfs_item_key_to_cpu(leaf, &key, i);
4839                 if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
4840                         continue;
4841                 fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
4842                 if (btrfs_file_extent_type(leaf, fi) ==
4843                     BTRFS_FILE_EXTENT_INLINE)
4844                         continue;
4845                 bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
4846                 num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
4847                 if (bytenr == 0)
4848                         continue;
4849
4850                 ext_index++;
4851                 if (bytenr >= group->key.objectid + group->key.offset ||
4852                     bytenr + num_bytes <= group->key.objectid)
4853                         continue;
4854
4855                 extent_key.objectid = bytenr;
4856                 extent_key.offset = num_bytes;
4857                 extent_key.type = BTRFS_EXTENT_ITEM_KEY;
4858                 nr_extent = 1;
4859                 ret = get_new_locations(reloc_inode, &extent_key,
4860                                         group->key.objectid, 1,
4861                                         &new_extent, &nr_extent);
4862                 if (ret > 0)
4863                         continue;
4864                 BUG_ON(ret < 0);
4865
4866                 BUG_ON(ref->extents[ext_index].bytenr != bytenr);
4867                 BUG_ON(ref->extents[ext_index].num_bytes != num_bytes);
4868                 ref->extents[ext_index].bytenr = new_extent->disk_bytenr;
4869                 ref->extents[ext_index].num_bytes = new_extent->disk_num_bytes;
4870
4871                 btrfs_set_file_extent_disk_bytenr(leaf, fi,
4872                                                 new_extent->disk_bytenr);
4873                 btrfs_set_file_extent_disk_num_bytes(leaf, fi,
4874                                                 new_extent->disk_num_bytes);
4875                 btrfs_mark_buffer_dirty(leaf);
4876
4877                 ret = btrfs_inc_extent_ref(trans, root,
4878                                         new_extent->disk_bytenr,
4879                                         new_extent->disk_num_bytes,
4880                                         leaf->start,
4881                                         root->root_key.objectid,
4882                                         trans->transid, key.objectid);
4883                 BUG_ON(ret);
4884                 ret = btrfs_free_extent(trans, root,
4885                                         bytenr, num_bytes, leaf->start,
4886                                         btrfs_header_owner(leaf),
4887                                         btrfs_header_generation(leaf),
4888                                         key.objectid, 0);
4889                 BUG_ON(ret);
4890                 cond_resched();
4891         }
4892         kfree(new_extent);
4893         BUG_ON(ext_index + 1 != ref->nritems);
4894         btrfs_free_leaf_ref(root, ref);
4895         return 0;
4896 }
4897
4898 int btrfs_free_reloc_root(struct btrfs_trans_handle *trans,
4899                           struct btrfs_root *root)
4900 {
4901         struct btrfs_root *reloc_root;
4902         int ret;
4903
4904         if (root->reloc_root) {
4905                 reloc_root = root->reloc_root;
4906                 root->reloc_root = NULL;
4907                 list_add(&reloc_root->dead_list,
4908                          &root->fs_info->dead_reloc_roots);
4909
4910                 btrfs_set_root_bytenr(&reloc_root->root_item,
4911                                       reloc_root->node->start);
4912                 btrfs_set_root_level(&root->root_item,
4913                                      btrfs_header_level(reloc_root->node));
4914                 memset(&reloc_root->root_item.drop_progress, 0,
4915                         sizeof(struct btrfs_disk_key));
4916                 reloc_root->root_item.drop_level = 0;
4917
4918                 ret = btrfs_update_root(trans, root->fs_info->tree_root,
4919                                         &reloc_root->root_key,
4920                                         &reloc_root->root_item);
4921                 BUG_ON(ret);
4922         }
4923         return 0;
4924 }
4925
4926 int btrfs_drop_dead_reloc_roots(struct btrfs_root *root)
4927 {
4928         struct btrfs_trans_handle *trans;
4929         struct btrfs_root *reloc_root;
4930         struct btrfs_root *prev_root = NULL;
4931         struct list_head dead_roots;
4932         int ret;
4933         unsigned long nr;
4934
4935         INIT_LIST_HEAD(&dead_roots);
4936         list_splice_init(&root->fs_info->dead_reloc_roots, &dead_roots);
4937
4938         while (!list_empty(&dead_roots)) {
4939                 reloc_root = list_entry(dead_roots.prev,
4940                                         struct btrfs_root, dead_list);
4941                 list_del_init(&reloc_root->dead_list);
4942
4943                 BUG_ON(reloc_root->commit_root != NULL);
4944                 while (1) {
4945                         trans = btrfs_join_transaction(root, 1);
4946                         BUG_ON(!trans);
4947
4948                         mutex_lock(&root->fs_info->drop_mutex);
4949                         ret = btrfs_drop_snapshot(trans, reloc_root);
4950                         if (ret != -EAGAIN)
4951                                 break;
4952                         mutex_unlock(&root->fs_info->drop_mutex);
4953
4954                         nr = trans->blocks_used;
4955                         ret = btrfs_end_transaction(trans, root);
4956                         BUG_ON(ret);
4957                         btrfs_btree_balance_dirty(root, nr);
4958                 }
4959
4960                 free_extent_buffer(reloc_root->node);
4961
4962                 ret = btrfs_del_root(trans, root->fs_info->tree_root,
4963                                      &reloc_root->root_key);
4964                 BUG_ON(ret);
4965                 mutex_unlock(&root->fs_info->drop_mutex);
4966
4967                 nr = trans->blocks_used;
4968                 ret = btrfs_end_transaction(trans, root);
4969                 BUG_ON(ret);
4970                 btrfs_btree_balance_dirty(root, nr);
4971
4972                 kfree(prev_root);
4973                 prev_root = reloc_root;
4974         }
4975         if (prev_root) {
4976                 btrfs_remove_leaf_refs(prev_root, (u64)-1, 0);
4977                 kfree(prev_root);
4978         }
4979         return 0;
4980 }
4981
4982 int btrfs_add_dead_reloc_root(struct btrfs_root *root)
4983 {
4984         list_add(&root->dead_list, &root->fs_info->dead_reloc_roots);
4985         return 0;
4986 }
4987
4988 int btrfs_cleanup_reloc_trees(struct btrfs_root *root)
4989 {
4990         struct btrfs_root *reloc_root;
4991         struct btrfs_trans_handle *trans;
4992         struct btrfs_key location;
4993         int found;
4994         int ret;
4995
4996         mutex_lock(&root->fs_info->tree_reloc_mutex);
4997         ret = btrfs_find_dead_roots(root, BTRFS_TREE_RELOC_OBJECTID, NULL);
4998         BUG_ON(ret);
4999         found = !list_empty(&root->fs_info->dead_reloc_roots);
5000         mutex_unlock(&root->fs_info->tree_reloc_mutex);
5001
5002         if (found) {
5003                 trans = btrfs_start_transaction(root, 1);
5004                 BUG_ON(!trans);
5005                 ret = btrfs_commit_transaction(trans, root);
5006                 BUG_ON(ret);
5007         }
5008
5009         location.objectid = BTRFS_DATA_RELOC_TREE_OBJECTID;
5010         location.offset = (u64)-1;
5011         location.type = BTRFS_ROOT_ITEM_KEY;
5012
5013         reloc_root = btrfs_read_fs_root_no_name(root->fs_info, &location);
5014         BUG_ON(!reloc_root);
5015         btrfs_orphan_cleanup(reloc_root);
5016         return 0;
5017 }
5018
5019 static int noinline init_reloc_tree(struct btrfs_trans_handle *trans,
5020                                     struct btrfs_root *root)
5021 {
5022         struct btrfs_root *reloc_root;
5023         struct extent_buffer *eb;
5024         struct btrfs_root_item *root_item;
5025         struct btrfs_key root_key;
5026         int ret;
5027
5028         BUG_ON(!root->ref_cows);
5029         if (root->reloc_root)
5030                 return 0;
5031
5032         root_item = kmalloc(sizeof(*root_item), GFP_NOFS);
5033         BUG_ON(!root_item);
5034
5035         ret = btrfs_copy_root(trans, root, root->commit_root,
5036                               &eb, BTRFS_TREE_RELOC_OBJECTID);
5037         BUG_ON(ret);
5038
5039         root_key.objectid = BTRFS_TREE_RELOC_OBJECTID;
5040         root_key.offset = root->root_key.objectid;
5041         root_key.type = BTRFS_ROOT_ITEM_KEY;
5042
5043         memcpy(root_item, &root->root_item, sizeof(root_item));
5044         btrfs_set_root_refs(root_item, 0);
5045         btrfs_set_root_bytenr(root_item, eb->start);
5046         btrfs_set_root_level(root_item, btrfs_header_level(eb));
5047         btrfs_set_root_generation(root_item, trans->transid);
5048
5049         btrfs_tree_unlock(eb);
5050         free_extent_buffer(eb);
5051
5052         ret = btrfs_insert_root(trans, root->fs_info->tree_root,
5053                                 &root_key, root_item);
5054         BUG_ON(ret);
5055         kfree(root_item);
5056
5057         reloc_root = btrfs_read_fs_root_no_radix(root->fs_info->tree_root,
5058                                                  &root_key);
5059         BUG_ON(!reloc_root);
5060         reloc_root->last_trans = trans->transid;
5061         reloc_root->commit_root = NULL;
5062         reloc_root->ref_tree = &root->fs_info->reloc_ref_tree;
5063
5064         root->reloc_root = reloc_root;
5065         return 0;
5066 }
5067
5068 /*
5069  * Core function of space balance.
5070  *
5071  * The idea is using reloc trees to relocate tree blocks in reference
5072  * counted roots. There is one reloc tree for each subvol, and all
5073  * reloc trees share same root key objectid. Reloc trees are snapshots
5074  * of the latest committed roots of subvols (root->commit_root).
5075  *
5076  * To relocate a tree block referenced by a subvol, there are two steps.
5077  * COW the block through subvol's reloc tree, then update block pointer
5078  * in the subvol to point to the new block. Since all reloc trees share
5079  * same root key objectid, doing special handing for tree blocks owned
5080  * by them is easy. Once a tree block has been COWed in one reloc tree,
5081  * we can use the resulting new block directly when the same block is
5082  * required to COW again through other reloc trees. By this way, relocated
5083  * tree blocks are shared between reloc trees, so they are also shared
5084  * between subvols.
5085  */
5086 static int noinline relocate_one_path(struct btrfs_trans_handle *trans,
5087                                       struct btrfs_root *root,
5088                                       struct btrfs_path *path,
5089                                       struct btrfs_key *first_key,
5090                                       struct btrfs_ref_path *ref_path,
5091                                       struct btrfs_block_group_cache *group,
5092                                       struct inode *reloc_inode)
5093 {
5094         struct btrfs_root *reloc_root;
5095         struct extent_buffer *eb = NULL;
5096         struct btrfs_key *keys;
5097         u64 *nodes;
5098         int level;
5099         int shared_level;
5100         int lowest_level = 0;
5101         int ret;
5102
5103         if (ref_path->owner_objectid < BTRFS_FIRST_FREE_OBJECTID)
5104                 lowest_level = ref_path->owner_objectid;
5105
5106         if (!root->ref_cows) {
5107                 path->lowest_level = lowest_level;
5108                 ret = btrfs_search_slot(trans, root, first_key, path, 0, 1);
5109                 BUG_ON(ret < 0);
5110                 path->lowest_level = 0;
5111                 btrfs_release_path(root, path);
5112                 return 0;
5113         }
5114
5115         mutex_lock(&root->fs_info->tree_reloc_mutex);
5116         ret = init_reloc_tree(trans, root);
5117         BUG_ON(ret);
5118         reloc_root = root->reloc_root;
5119
5120         shared_level = ref_path->shared_level;
5121         ref_path->shared_level = BTRFS_MAX_LEVEL - 1;
5122
5123         keys = ref_path->node_keys;
5124         nodes = ref_path->new_nodes;
5125         memset(&keys[shared_level + 1], 0,
5126                sizeof(*keys) * (BTRFS_MAX_LEVEL - shared_level - 1));
5127         memset(&nodes[shared_level + 1], 0,
5128                sizeof(*nodes) * (BTRFS_MAX_LEVEL - shared_level - 1));
5129
5130         if (nodes[lowest_level] == 0) {
5131                 path->lowest_level = lowest_level;
5132                 ret = btrfs_search_slot(trans, reloc_root, first_key, path,
5133                                         0, 1);
5134                 BUG_ON(ret);
5135                 for (level = lowest_level; level < BTRFS_MAX_LEVEL; level++) {
5136                         eb = path->nodes[level];
5137                         if (!eb || eb == reloc_root->node)
5138                                 break;
5139                         nodes[level] = eb->start;
5140                         if (level == 0)
5141                                 btrfs_item_key_to_cpu(eb, &keys[level], 0);
5142                         else
5143                                 btrfs_node_key_to_cpu(eb, &keys[level], 0);
5144                 }
5145                 if (nodes[0] &&
5146                     ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
5147                         eb = path->nodes[0];
5148                         ret = replace_extents_in_leaf(trans, reloc_root, eb,
5149                                                       group, reloc_inode);
5150                         BUG_ON(ret);
5151                 }
5152                 btrfs_release_path(reloc_root, path);
5153         } else {
5154                 ret = btrfs_merge_path(trans, reloc_root, keys, nodes,
5155                                        lowest_level);
5156                 BUG_ON(ret);
5157         }
5158
5159         /*
5160          * replace tree blocks in the fs tree with tree blocks in
5161          * the reloc tree.
5162          */
5163         ret = btrfs_merge_path(trans, root, keys, nodes, lowest_level);
5164         BUG_ON(ret < 0);
5165
5166         if (ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
5167                 ret = btrfs_search_slot(trans, reloc_root, first_key, path,
5168                                         0, 0);
5169                 BUG_ON(ret);
5170                 extent_buffer_get(path->nodes[0]);
5171                 eb = path->nodes[0];
5172                 btrfs_release_path(reloc_root, path);
5173                 ret = invalidate_extent_cache(reloc_root, eb, group, root);
5174                 BUG_ON(ret);
5175                 free_extent_buffer(eb);
5176         }
5177
5178         mutex_unlock(&root->fs_info->tree_reloc_mutex);
5179         path->lowest_level = 0;
5180         return 0;
5181 }
5182
5183 static int noinline relocate_tree_block(struct btrfs_trans_handle *trans,
5184                                         struct btrfs_root *root,
5185                                         struct btrfs_path *path,
5186                                         struct btrfs_key *first_key,
5187                                         struct btrfs_ref_path *ref_path)
5188 {
5189         int ret;
5190
5191         ret = relocate_one_path(trans, root, path, first_key,
5192                                 ref_path, NULL, NULL);
5193         BUG_ON(ret);
5194
5195         if (root == root->fs_info->extent_root)
5196                 btrfs_extent_post_op(trans, root);
5197
5198         return 0;
5199 }
5200
5201 static int noinline del_extent_zero(struct btrfs_trans_handle *trans,
5202                                     struct btrfs_root *extent_root,
5203                                     struct btrfs_path *path,
5204                                     struct btrfs_key *extent_key)
5205 {
5206         int ret;
5207
5208         ret = btrfs_search_slot(trans, extent_root, extent_key, path, -1, 1);
5209         if (ret)
5210                 goto out;
5211         ret = btrfs_del_item(trans, extent_root, path);
5212 out:
5213         btrfs_release_path(extent_root, path);
5214         return ret;
5215 }
5216
5217 static struct btrfs_root noinline *read_ref_root(struct btrfs_fs_info *fs_info,
5218                                                 struct btrfs_ref_path *ref_path)
5219 {
5220         struct btrfs_key root_key;
5221
5222         root_key.objectid = ref_path->root_objectid;
5223         root_key.type = BTRFS_ROOT_ITEM_KEY;
5224         if (is_cowonly_root(ref_path->root_objectid))
5225                 root_key.offset = 0;
5226         else
5227                 root_key.offset = (u64)-1;
5228
5229         return btrfs_read_fs_root_no_name(fs_info, &root_key);
5230 }
5231
5232 static int noinline relocate_one_extent(struct btrfs_root *extent_root,
5233                                         struct btrfs_path *path,
5234                                         struct btrfs_key *extent_key,
5235                                         struct btrfs_block_group_cache *group,
5236                                         struct inode *reloc_inode, int pass)
5237 {
5238         struct btrfs_trans_handle *trans;
5239         struct btrfs_root *found_root;
5240         struct btrfs_ref_path *ref_path = NULL;
5241         struct disk_extent *new_extents = NULL;
5242         int nr_extents = 0;
5243         int loops;
5244         int ret;
5245         int level;
5246         struct btrfs_key first_key;
5247         u64 prev_block = 0;
5248
5249
5250         trans = btrfs_start_transaction(extent_root, 1);
5251         BUG_ON(!trans);
5252
5253         if (extent_key->objectid == 0) {
5254                 ret = del_extent_zero(trans, extent_root, path, extent_key);
5255                 goto out;
5256         }
5257
5258         ref_path = kmalloc(sizeof(*ref_path), GFP_NOFS);
5259         if (!ref_path) {
5260                ret = -ENOMEM;
5261                goto out;
5262         }
5263
5264         for (loops = 0; ; loops++) {
5265                 if (loops == 0) {
5266                         ret = btrfs_first_ref_path(trans, extent_root, ref_path,
5267                                                    extent_key->objectid);
5268                 } else {
5269                         ret = btrfs_next_ref_path(trans, extent_root, ref_path);
5270                 }
5271                 if (ret < 0)
5272                         goto out;
5273                 if (ret > 0)
5274                         break;
5275
5276                 if (ref_path->root_objectid == BTRFS_TREE_LOG_OBJECTID ||
5277                     ref_path->root_objectid == BTRFS_TREE_RELOC_OBJECTID)
5278                         continue;
5279
5280                 found_root = read_ref_root(extent_root->fs_info, ref_path);
5281                 BUG_ON(!found_root);
5282                 /*
5283                  * for reference counted tree, only process reference paths
5284                  * rooted at the latest committed root.
5285                  */
5286                 if (found_root->ref_cows &&
5287                     ref_path->root_generation != found_root->root_key.offset)
5288                         continue;
5289
5290                 if (ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
5291                         if (pass == 0) {
5292                                 /*
5293                                  * copy data extents to new locations
5294                                  */
5295                                 u64 group_start = group->key.objectid;
5296                                 ret = relocate_data_extent(reloc_inode,
5297                                                            extent_key,
5298                                                            group_start);
5299                                 if (ret < 0)
5300                                         goto out;
5301                                 break;
5302                         }
5303                         level = 0;
5304                 } else {
5305                         level = ref_path->owner_objectid;
5306                 }
5307
5308                 if (prev_block != ref_path->nodes[level]) {
5309                         struct extent_buffer *eb;
5310                         u64 block_start = ref_path->nodes[level];
5311                         u64 block_size = btrfs_level_size(found_root, level);
5312
5313                         eb = read_tree_block(found_root, block_start,
5314                                              block_size, 0);
5315                         btrfs_tree_lock(eb);
5316                         BUG_ON(level != btrfs_header_level(eb));
5317
5318                         if (level == 0)
5319                                 btrfs_item_key_to_cpu(eb, &first_key, 0);
5320                         else
5321                                 btrfs_node_key_to_cpu(eb, &first_key, 0);
5322
5323                         btrfs_tree_unlock(eb);
5324                         free_extent_buffer(eb);
5325                         prev_block = block_start;
5326                 }
5327
5328                 if (ref_path->owner_objectid >= BTRFS_FIRST_FREE_OBJECTID &&
5329                     pass >= 2) {
5330                         /*
5331                          * use fallback method to process the remaining
5332                          * references.
5333                          */
5334                         if (!new_extents) {
5335                                 u64 group_start = group->key.objectid;
5336                                 new_extents = kmalloc(sizeof(*new_extents),
5337                                                       GFP_NOFS);
5338                                 nr_extents = 1;
5339                                 ret = get_new_locations(reloc_inode,
5340                                                         extent_key,
5341                                                         group_start, 1,
5342                                                         &new_extents,
5343                                                         &nr_extents);
5344                                 if (ret)
5345                                         goto out;
5346                         }
5347                         btrfs_record_root_in_trans(found_root);
5348                         ret = replace_one_extent(trans, found_root,
5349                                                 path, extent_key,
5350                                                 &first_key, ref_path,
5351                                                 new_extents, nr_extents);
5352                         if (ret < 0)
5353                                 goto out;
5354                         continue;
5355                 }
5356
5357                 btrfs_record_root_in_trans(found_root);
5358                 if (ref_path->owner_objectid < BTRFS_FIRST_FREE_OBJECTID) {
5359                         ret = relocate_tree_block(trans, found_root, path,
5360                                                   &first_key, ref_path);
5361                 } else {
5362                         /*
5363                          * try to update data extent references while
5364                          * keeping metadata shared between snapshots.
5365                          */
5366                         ret = relocate_one_path(trans, found_root, path,
5367                                                 &first_key, ref_path,
5368                                                 group, reloc_inode);
5369                 }
5370                 if (ret < 0)
5371                         goto out;
5372         }
5373         ret = 0;
5374 out:
5375         btrfs_end_transaction(trans, extent_root);
5376         kfree(new_extents);
5377         kfree(ref_path);
5378         return ret;
5379 }
5380
5381 static u64 update_block_group_flags(struct btrfs_root *root, u64 flags)
5382 {
5383         u64 num_devices;
5384         u64 stripped = BTRFS_BLOCK_GROUP_RAID0 |
5385                 BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10;
5386
5387         num_devices = root->fs_info->fs_devices->rw_devices;
5388         if (num_devices == 1) {
5389                 stripped |= BTRFS_BLOCK_GROUP_DUP;
5390                 stripped = flags & ~stripped;
5391
5392                 /* turn raid0 into single device chunks */
5393                 if (flags & BTRFS_BLOCK_GROUP_RAID0)
5394                         return stripped;
5395
5396                 /* turn mirroring into duplication */
5397                 if (flags & (BTRFS_BLOCK_GROUP_RAID1 |
5398                              BTRFS_BLOCK_GROUP_RAID10))
5399                         return stripped | BTRFS_BLOCK_GROUP_DUP;
5400                 return flags;
5401         } else {
5402                 /* they already had raid on here, just return */
5403                 if (flags & stripped)
5404                         return flags;
5405
5406                 stripped |= BTRFS_BLOCK_GROUP_DUP;
5407                 stripped = flags & ~stripped;
5408
5409                 /* switch duplicated blocks with raid1 */
5410                 if (flags & BTRFS_BLOCK_GROUP_DUP)
5411                         return stripped | BTRFS_BLOCK_GROUP_RAID1;
5412
5413                 /* turn single device chunks into raid0 */
5414                 return stripped | BTRFS_BLOCK_GROUP_RAID0;
5415         }
5416         return flags;
5417 }
5418
5419 int __alloc_chunk_for_shrink(struct btrfs_root *root,
5420                      struct btrfs_block_group_cache *shrink_block_group,
5421                      int force)
5422 {
5423         struct btrfs_trans_handle *trans;
5424         u64 new_alloc_flags;
5425         u64 calc;
5426
5427         spin_lock(&shrink_block_group->lock);
5428         if (btrfs_block_group_used(&shrink_block_group->item) > 0) {
5429                 spin_unlock(&shrink_block_group->lock);
5430
5431                 trans = btrfs_start_transaction(root, 1);
5432                 spin_lock(&shrink_block_group->lock);
5433
5434                 new_alloc_flags = update_block_group_flags(root,
5435                                                    shrink_block_group->flags);
5436                 if (new_alloc_flags != shrink_block_group->flags) {
5437                         calc =
5438                              btrfs_block_group_used(&shrink_block_group->item);
5439                 } else {
5440                         calc = shrink_block_group->key.offset;
5441                 }
5442                 spin_unlock(&shrink_block_group->lock);
5443
5444                 do_chunk_alloc(trans, root->fs_info->extent_root,
5445                                calc + 2 * 1024 * 1024, new_alloc_flags, force);
5446
5447                 btrfs_end_transaction(trans, root);
5448         } else
5449                 spin_unlock(&shrink_block_group->lock);
5450         return 0;
5451 }
5452
5453 static int __insert_orphan_inode(struct btrfs_trans_handle *trans,
5454                                  struct btrfs_root *root,
5455                                  u64 objectid, u64 size)
5456 {
5457         struct btrfs_path *path;
5458         struct btrfs_inode_item *item;
5459         struct extent_buffer *leaf;
5460         int ret;
5461
5462         path = btrfs_alloc_path();
5463         if (!path)
5464                 return -ENOMEM;
5465
5466         ret = btrfs_insert_empty_inode(trans, root, path, objectid);
5467         if (ret)
5468                 goto out;
5469
5470         leaf = path->nodes[0];
5471         item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_inode_item);
5472         memset_extent_buffer(leaf, 0, (unsigned long)item, sizeof(*item));
5473         btrfs_set_inode_generation(leaf, item, 1);
5474         btrfs_set_inode_size(leaf, item, size);
5475         btrfs_set_inode_mode(leaf, item, S_IFREG | 0600);
5476         btrfs_set_inode_flags(leaf, item, BTRFS_INODE_NODATASUM |
5477                                           BTRFS_INODE_NOCOMPRESS);
5478         btrfs_mark_buffer_dirty(leaf);
5479         btrfs_release_path(root, path);
5480 out:
5481         btrfs_free_path(path);
5482         return ret;
5483 }
5484
5485 static struct inode noinline *create_reloc_inode(struct btrfs_fs_info *fs_info,
5486                                         struct btrfs_block_group_cache *group)
5487 {
5488         struct inode *inode = NULL;
5489         struct btrfs_trans_handle *trans;
5490         struct btrfs_root *root;
5491         struct btrfs_key root_key;
5492         u64 objectid = BTRFS_FIRST_FREE_OBJECTID;
5493         int err = 0;
5494
5495         root_key.objectid = BTRFS_DATA_RELOC_TREE_OBJECTID;
5496         root_key.type = BTRFS_ROOT_ITEM_KEY;
5497         root_key.offset = (u64)-1;
5498         root = btrfs_read_fs_root_no_name(fs_info, &root_key);
5499         if (IS_ERR(root))
5500                 return ERR_CAST(root);
5501
5502         trans = btrfs_start_transaction(root, 1);
5503         BUG_ON(!trans);
5504
5505         err = btrfs_find_free_objectid(trans, root, objectid, &objectid);
5506         if (err)
5507                 goto out;
5508
5509         err = __insert_orphan_inode(trans, root, objectid, group->key.offset);
5510         BUG_ON(err);
5511
5512         err = btrfs_insert_file_extent(trans, root, objectid, 0, 0, 0,
5513                                        group->key.offset, 0, group->key.offset,
5514                                        0, 0, 0);
5515         BUG_ON(err);
5516
5517         inode = btrfs_iget_locked(root->fs_info->sb, objectid, root);
5518         if (inode->i_state & I_NEW) {
5519                 BTRFS_I(inode)->root = root;
5520                 BTRFS_I(inode)->location.objectid = objectid;
5521                 BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY;
5522                 BTRFS_I(inode)->location.offset = 0;
5523                 btrfs_read_locked_inode(inode);
5524                 unlock_new_inode(inode);
5525                 BUG_ON(is_bad_inode(inode));
5526         } else {
5527                 BUG_ON(1);
5528         }
5529
5530         err = btrfs_orphan_add(trans, inode);
5531 out:
5532         btrfs_end_transaction(trans, root);
5533         if (err) {
5534                 if (inode)
5535                         iput(inode);
5536                 inode = ERR_PTR(err);
5537         }
5538         return inode;
5539 }
5540
5541 int btrfs_relocate_block_group(struct btrfs_root *root, u64 group_start)
5542 {
5543         struct btrfs_trans_handle *trans;
5544         struct btrfs_path *path;
5545         struct btrfs_fs_info *info = root->fs_info;
5546         struct extent_buffer *leaf;
5547         struct inode *reloc_inode;
5548         struct btrfs_block_group_cache *block_group;
5549         struct btrfs_key key;
5550         u64 skipped;
5551         u64 cur_byte;
5552         u64 total_found;
5553         u32 nritems;
5554         int ret;
5555         int progress;
5556         int pass = 0;
5557
5558         root = root->fs_info->extent_root;
5559
5560         block_group = btrfs_lookup_block_group(info, group_start);
5561         BUG_ON(!block_group);
5562
5563         printk("btrfs relocating block group %llu flags %llu\n",
5564                (unsigned long long)block_group->key.objectid,
5565                (unsigned long long)block_group->flags);
5566
5567         path = btrfs_alloc_path();
5568         BUG_ON(!path);
5569
5570         reloc_inode = create_reloc_inode(info, block_group);
5571         BUG_ON(IS_ERR(reloc_inode));
5572
5573         __alloc_chunk_for_shrink(root, block_group, 1);
5574         set_block_group_readonly(block_group);
5575
5576         btrfs_start_delalloc_inodes(info->tree_root);
5577         btrfs_wait_ordered_extents(info->tree_root, 0);
5578 again:
5579         skipped = 0;
5580         total_found = 0;
5581         progress = 0;
5582         key.objectid = block_group->key.objectid;
5583         key.offset = 0;
5584         key.type = 0;
5585         cur_byte = key.objectid;
5586
5587         trans = btrfs_start_transaction(info->tree_root, 1);
5588         btrfs_commit_transaction(trans, info->tree_root);
5589
5590         mutex_lock(&root->fs_info->cleaner_mutex);
5591         btrfs_clean_old_snapshots(info->tree_root);
5592         btrfs_remove_leaf_refs(info->tree_root, (u64)-1, 1);
5593         mutex_unlock(&root->fs_info->cleaner_mutex);
5594
5595         while(1) {
5596                 ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
5597                 if (ret < 0)
5598                         goto out;
5599 next:
5600                 leaf = path->nodes[0];
5601                 nritems = btrfs_header_nritems(leaf);
5602                 if (path->slots[0] >= nritems) {
5603                         ret = btrfs_next_leaf(root, path);
5604                         if (ret < 0)
5605                                 goto out;
5606                         if (ret == 1) {
5607                                 ret = 0;
5608                                 break;
5609                         }
5610                         leaf = path->nodes[0];
5611                         nritems = btrfs_header_nritems(leaf);
5612                 }
5613
5614                 btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
5615
5616                 if (key.objectid >= block_group->key.objectid +
5617                     block_group->key.offset)
5618                         break;
5619
5620                 if (progress && need_resched()) {
5621                         btrfs_release_path(root, path);
5622                         cond_resched();
5623                         progress = 0;
5624                         continue;
5625                 }
5626                 progress = 1;
5627
5628                 if (btrfs_key_type(&key) != BTRFS_EXTENT_ITEM_KEY ||
5629                     key.objectid + key.offset <= cur_byte) {
5630                         path->slots[0]++;
5631                         goto next;
5632                 }
5633
5634                 total_found++;
5635                 cur_byte = key.objectid + key.offset;
5636                 btrfs_release_path(root, path);
5637
5638                 __alloc_chunk_for_shrink(root, block_group, 0);
5639                 ret = relocate_one_extent(root, path, &key, block_group,
5640                                           reloc_inode, pass);
5641                 BUG_ON(ret < 0);
5642                 if (ret > 0)
5643                         skipped++;
5644
5645                 key.objectid = cur_byte;
5646                 key.type = 0;
5647                 key.offset = 0;
5648         }
5649
5650         btrfs_release_path(root, path);
5651
5652         if (pass == 0) {
5653                 btrfs_wait_ordered_range(reloc_inode, 0, (u64)-1);
5654                 invalidate_mapping_pages(reloc_inode->i_mapping, 0, -1);
5655                 WARN_ON(reloc_inode->i_mapping->nrpages);
5656         }
5657
5658         if (total_found > 0) {
5659                 printk("btrfs found %llu extents in pass %d\n",
5660                        (unsigned long long)total_found, pass);
5661                 pass++;
5662                 if (total_found == skipped && pass > 2) {
5663                         iput(reloc_inode);
5664                         reloc_inode = create_reloc_inode(info, block_group);
5665                         pass = 0;
5666                 }
5667                 goto again;
5668         }
5669
5670         /* delete reloc_inode */
5671         iput(reloc_inode);
5672
5673         /* unpin extents in this range */
5674         trans = btrfs_start_transaction(info->tree_root, 1);
5675         btrfs_commit_transaction(trans, info->tree_root);
5676
5677         spin_lock(&block_group->lock);
5678         WARN_ON(block_group->pinned > 0);
5679         WARN_ON(block_group->reserved > 0);
5680         WARN_ON(btrfs_block_group_used(&block_group->item) > 0);
5681         spin_unlock(&block_group->lock);
5682         ret = 0;
5683 out:
5684         btrfs_free_path(path);
5685         return ret;
5686 }
5687
5688 int find_first_block_group(struct btrfs_root *root, struct btrfs_path *path,
5689                            struct btrfs_key *key)
5690 {
5691         int ret = 0;
5692         struct btrfs_key found_key;
5693         struct extent_buffer *leaf;
5694         int slot;
5695
5696         ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
5697         if (ret < 0)
5698                 goto out;
5699
5700         while(1) {
5701                 slot = path->slots[0];
5702                 leaf = path->nodes[0];
5703                 if (slot >= btrfs_header_nritems(leaf)) {
5704                         ret = btrfs_next_leaf(root, path);
5705                         if (ret == 0)
5706                                 continue;
5707                         if (ret < 0)
5708                                 goto out;
5709                         break;
5710                 }
5711                 btrfs_item_key_to_cpu(leaf, &found_key, slot);
5712
5713                 if (found_key.objectid >= key->objectid &&
5714                     found_key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) {
5715                         ret = 0;
5716                         goto out;
5717                 }
5718                 path->slots[0]++;
5719         }
5720         ret = -ENOENT;
5721 out:
5722         return ret;
5723 }
5724
5725 int btrfs_free_block_groups(struct btrfs_fs_info *info)
5726 {
5727         struct btrfs_block_group_cache *block_group;
5728         struct rb_node *n;
5729
5730         spin_lock(&info->block_group_cache_lock);
5731         while ((n = rb_last(&info->block_group_cache_tree)) != NULL) {
5732                 block_group = rb_entry(n, struct btrfs_block_group_cache,
5733                                        cache_node);
5734                 rb_erase(&block_group->cache_node,
5735                          &info->block_group_cache_tree);
5736                 spin_unlock(&info->block_group_cache_lock);
5737
5738                 btrfs_remove_free_space_cache(block_group);
5739                 down_write(&block_group->space_info->groups_sem);
5740                 list_del(&block_group->list);
5741                 up_write(&block_group->space_info->groups_sem);
5742                 kfree(block_group);
5743
5744                 spin_lock(&info->block_group_cache_lock);
5745         }
5746         spin_unlock(&info->block_group_cache_lock);
5747         return 0;
5748 }
5749
5750 int btrfs_read_block_groups(struct btrfs_root *root)
5751 {
5752         struct btrfs_path *path;
5753         int ret;
5754         struct btrfs_block_group_cache *cache;
5755         struct btrfs_fs_info *info = root->fs_info;
5756         struct btrfs_space_info *space_info;
5757         struct btrfs_key key;
5758         struct btrfs_key found_key;
5759         struct extent_buffer *leaf;
5760
5761         root = info->extent_root;
5762         key.objectid = 0;
5763         key.offset = 0;
5764         btrfs_set_key_type(&key, BTRFS_BLOCK_GROUP_ITEM_KEY);
5765         path = btrfs_alloc_path();
5766         if (!path)
5767                 return -ENOMEM;
5768
5769         while(1) {
5770                 ret = find_first_block_group(root, path, &key);
5771                 if (ret > 0) {
5772                         ret = 0;
5773                         goto error;
5774                 }
5775                 if (ret != 0)
5776                         goto error;
5777
5778                 leaf = path->nodes[0];
5779                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
5780                 cache = kzalloc(sizeof(*cache), GFP_NOFS);
5781                 if (!cache) {
5782                         ret = -ENOMEM;
5783                         break;
5784                 }
5785
5786                 spin_lock_init(&cache->lock);
5787                 mutex_init(&cache->alloc_mutex);
5788                 INIT_LIST_HEAD(&cache->list);
5789                 read_extent_buffer(leaf, &cache->item,
5790                                    btrfs_item_ptr_offset(leaf, path->slots[0]),
5791                                    sizeof(cache->item));
5792                 memcpy(&cache->key, &found_key, sizeof(found_key));
5793
5794                 key.objectid = found_key.objectid + found_key.offset;
5795                 btrfs_release_path(root, path);
5796                 cache->flags = btrfs_block_group_flags(&cache->item);
5797
5798                 ret = update_space_info(info, cache->flags, found_key.offset,
5799                                         btrfs_block_group_used(&cache->item),
5800                                         &space_info);
5801                 BUG_ON(ret);
5802                 cache->space_info = space_info;
5803                 down_write(&space_info->groups_sem);
5804                 list_add_tail(&cache->list, &space_info->block_groups);
5805                 up_write(&space_info->groups_sem);
5806
5807                 ret = btrfs_add_block_group_cache(root->fs_info, cache);
5808                 BUG_ON(ret);
5809
5810                 set_avail_alloc_bits(root->fs_info, cache->flags);
5811                 if (btrfs_chunk_readonly(root, cache->key.objectid))
5812                         set_block_group_readonly(cache);
5813         }
5814         ret = 0;
5815 error:
5816         btrfs_free_path(path);
5817         return ret;
5818 }
5819
5820 int btrfs_make_block_group(struct btrfs_trans_handle *trans,
5821                            struct btrfs_root *root, u64 bytes_used,
5822                            u64 type, u64 chunk_objectid, u64 chunk_offset,
5823                            u64 size)
5824 {
5825         int ret;
5826         struct btrfs_root *extent_root;
5827         struct btrfs_block_group_cache *cache;
5828
5829         extent_root = root->fs_info->extent_root;
5830
5831         root->fs_info->last_trans_new_blockgroup = trans->transid;
5832
5833         cache = kzalloc(sizeof(*cache), GFP_NOFS);
5834         if (!cache)
5835                 return -ENOMEM;
5836
5837         cache->key.objectid = chunk_offset;
5838         cache->key.offset = size;
5839         spin_lock_init(&cache->lock);
5840         mutex_init(&cache->alloc_mutex);
5841         INIT_LIST_HEAD(&cache->list);
5842         btrfs_set_key_type(&cache->key, BTRFS_BLOCK_GROUP_ITEM_KEY);
5843
5844         btrfs_set_block_group_used(&cache->item, bytes_used);
5845         btrfs_set_block_group_chunk_objectid(&cache->item, chunk_objectid);
5846         cache->flags = type;
5847         btrfs_set_block_group_flags(&cache->item, type);
5848
5849         ret = update_space_info(root->fs_info, cache->flags, size, bytes_used,
5850                                 &cache->space_info);
5851         BUG_ON(ret);
5852         down_write(&cache->space_info->groups_sem);
5853         list_add_tail(&cache->list, &cache->space_info->block_groups);
5854         up_write(&cache->space_info->groups_sem);
5855
5856         ret = btrfs_add_block_group_cache(root->fs_info, cache);
5857         BUG_ON(ret);
5858
5859         ret = btrfs_insert_item(trans, extent_root, &cache->key, &cache->item,
5860                                 sizeof(cache->item));
5861         BUG_ON(ret);
5862
5863         finish_current_insert(trans, extent_root, 0);
5864         ret = del_pending_extents(trans, extent_root, 0);
5865         BUG_ON(ret);
5866         set_avail_alloc_bits(extent_root->fs_info, type);
5867
5868         return 0;
5869 }
5870
5871 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
5872                              struct btrfs_root *root, u64 group_start)
5873 {
5874         struct btrfs_path *path;
5875         struct btrfs_block_group_cache *block_group;
5876         struct btrfs_key key;
5877         int ret;
5878
5879         root = root->fs_info->extent_root;
5880
5881         block_group = btrfs_lookup_block_group(root->fs_info, group_start);
5882         BUG_ON(!block_group);
5883         BUG_ON(!block_group->ro);
5884
5885         memcpy(&key, &block_group->key, sizeof(key));
5886
5887         path = btrfs_alloc_path();
5888         BUG_ON(!path);
5889
5890         btrfs_remove_free_space_cache(block_group);
5891         rb_erase(&block_group->cache_node,
5892                  &root->fs_info->block_group_cache_tree);
5893         down_write(&block_group->space_info->groups_sem);
5894         list_del(&block_group->list);
5895         up_write(&block_group->space_info->groups_sem);
5896
5897         spin_lock(&block_group->space_info->lock);
5898         block_group->space_info->total_bytes -= block_group->key.offset;
5899         block_group->space_info->bytes_readonly -= block_group->key.offset;
5900         spin_unlock(&block_group->space_info->lock);
5901         block_group->space_info->full = 0;
5902
5903         /*
5904         memset(shrink_block_group, 0, sizeof(*shrink_block_group));
5905         kfree(shrink_block_group);
5906         */
5907
5908         ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
5909         if (ret > 0)
5910                 ret = -EIO;
5911         if (ret < 0)
5912                 goto out;
5913
5914         ret = btrfs_del_item(trans, root, path);
5915 out:
5916         btrfs_free_path(path);
5917         return ret;
5918 }