2 * Copyright (C) 2007 Oracle. All rights reserved.
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.
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.
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.
19 #include <linux/sched.h>
20 #include <linux/crc32c.h>
21 #include <linux/pagemap.h>
25 #include "print-tree.h"
26 #include "transaction.h"
28 #define BLOCK_GROUP_DATA EXTENT_WRITEBACK
29 #define BLOCK_GROUP_METADATA EXTENT_UPTODATE
30 #define BLOCK_GROUP_DIRTY EXTENT_DIRTY
32 static int finish_current_insert(struct btrfs_trans_handle *trans, struct
33 btrfs_root *extent_root);
34 static int del_pending_extents(struct btrfs_trans_handle *trans, struct
35 btrfs_root *extent_root);
37 static int cache_block_group(struct btrfs_root *root,
38 struct btrfs_block_group_cache *block_group)
40 struct btrfs_path *path;
43 struct extent_buffer *leaf;
44 struct extent_map_tree *free_space_cache;
54 root = root->fs_info->extent_root;
55 free_space_cache = &root->fs_info->free_space_cache;
57 if (block_group->cached)
60 path = btrfs_alloc_path();
65 first_free = block_group->key.objectid;
66 key.objectid = block_group->key.objectid;
69 btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
70 ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
75 if (ret && path->slots[0] > 0)
79 leaf = path->nodes[0];
80 slot = path->slots[0];
81 if (slot >= btrfs_header_nritems(leaf)) {
82 ret = btrfs_next_leaf(root, path);
92 btrfs_item_key_to_cpu(leaf, &key, slot);
93 if (key.objectid < block_group->key.objectid) {
94 if (btrfs_key_type(&key) != BTRFS_EXTENT_REF_KEY &&
95 key.objectid + key.offset > first_free)
96 first_free = key.objectid + key.offset;
100 if (key.objectid >= block_group->key.objectid +
101 block_group->key.offset) {
105 if (btrfs_key_type(&key) == BTRFS_EXTENT_ITEM_KEY) {
110 if (key.objectid > last) {
111 hole_size = key.objectid - last;
112 set_extent_dirty(free_space_cache, last,
113 last + hole_size - 1,
116 last = key.objectid + key.offset;
124 if (block_group->key.objectid +
125 block_group->key.offset > last) {
126 hole_size = block_group->key.objectid +
127 block_group->key.offset - last;
128 set_extent_dirty(free_space_cache, last,
129 last + hole_size - 1, GFP_NOFS);
131 block_group->cached = 1;
133 btrfs_free_path(path);
137 struct btrfs_block_group_cache *btrfs_lookup_block_group(struct
141 struct extent_map_tree *block_group_cache;
142 struct btrfs_block_group_cache *block_group = NULL;
148 block_group_cache = &info->block_group_cache;
149 ret = find_first_extent_bit(block_group_cache,
150 bytenr, &start, &end,
151 BLOCK_GROUP_DATA | BLOCK_GROUP_METADATA);
155 ret = get_state_private(block_group_cache, start, &ptr);
159 block_group = (struct btrfs_block_group_cache *)(unsigned long)ptr;
160 if (block_group->key.objectid <= bytenr && bytenr <
161 block_group->key.objectid + block_group->key.offset)
166 static u64 noinline find_search_start(struct btrfs_root *root,
167 struct btrfs_block_group_cache **cache_ret,
168 u64 search_start, int num,
169 int data, int full_scan)
172 struct btrfs_block_group_cache *cache = *cache_ret;
183 ret = cache_block_group(root, cache);
187 last = max(search_start, cache->key.objectid);
190 ret = find_first_extent_bit(&root->fs_info->free_space_cache,
191 last, &start, &end, EXTENT_DIRTY);
198 start = max(last, start);
200 if (last - start < num) {
201 if (last == cache->key.objectid + cache->key.offset)
205 if (data != BTRFS_BLOCK_GROUP_MIXED &&
206 start + num > cache->key.objectid + cache->key.offset)
211 cache = btrfs_lookup_block_group(root->fs_info, search_start);
213 printk("Unable to find block group for %Lu\n",
221 last = cache->key.objectid + cache->key.offset;
223 cache = btrfs_lookup_block_group(root->fs_info, last);
229 data = BTRFS_BLOCK_GROUP_MIXED;
234 if (cache_miss && !cache->cached) {
235 cache_block_group(root, cache);
237 cache = btrfs_lookup_block_group(root->fs_info, last);
239 cache = btrfs_find_block_group(root, cache, last, data, 0);
247 static u64 div_factor(u64 num, int factor)
256 struct btrfs_block_group_cache *btrfs_find_block_group(struct btrfs_root *root,
257 struct btrfs_block_group_cache
258 *hint, u64 search_start,
261 struct btrfs_block_group_cache *cache;
262 struct extent_map_tree *block_group_cache;
263 struct btrfs_block_group_cache *found_group = NULL;
264 struct btrfs_fs_info *info = root->fs_info;
278 block_group_cache = &info->block_group_cache;
283 if (data == BTRFS_BLOCK_GROUP_MIXED) {
284 bit = BLOCK_GROUP_DATA | BLOCK_GROUP_METADATA;
287 bit = BLOCK_GROUP_DATA;
289 bit = BLOCK_GROUP_METADATA;
292 struct btrfs_block_group_cache *shint;
293 shint = btrfs_lookup_block_group(info, search_start);
294 if (shint && (shint->data == data ||
295 shint->data == BTRFS_BLOCK_GROUP_MIXED)) {
296 used = btrfs_block_group_used(&shint->item);
297 if (used + shint->pinned <
298 div_factor(shint->key.offset, factor)) {
303 if (hint && (hint->data == data ||
304 hint->data == BTRFS_BLOCK_GROUP_MIXED)) {
305 used = btrfs_block_group_used(&hint->item);
306 if (used + hint->pinned <
307 div_factor(hint->key.offset, factor)) {
310 last = hint->key.objectid + hint->key.offset;
314 hint_last = max(hint->key.objectid, search_start);
316 hint_last = search_start;
322 ret = find_first_extent_bit(block_group_cache, last,
327 ret = get_state_private(block_group_cache, start, &ptr);
331 cache = (struct btrfs_block_group_cache *)(unsigned long)ptr;
332 last = cache->key.objectid + cache->key.offset;
333 used = btrfs_block_group_used(&cache->item);
336 free_check = cache->key.offset;
338 free_check = div_factor(cache->key.offset, factor);
339 if (used + cache->pinned < free_check) {
352 bit = BLOCK_GROUP_DATA | BLOCK_GROUP_METADATA;
360 static u64 hash_extent_ref(u64 root_objectid, u64 ref_generation,
361 u64 owner, u64 owner_offset)
363 u32 high_crc = ~(u32)0;
364 u32 low_crc = ~(u32)0;
367 lenum = cpu_to_le64(root_objectid);
368 high_crc = crc32c(high_crc, &lenum, sizeof(lenum));
369 lenum = cpu_to_le64(ref_generation);
370 low_crc = crc32c(low_crc, &lenum, sizeof(lenum));
373 lenum = cpu_to_le64(owner);
374 low_crc = crc32c(low_crc, &lenum, sizeof(lenum));
375 lenum = cpu_to_le64(owner_offset);
376 low_crc = crc32c(low_crc, &lenum, sizeof(lenum));
378 return ((u64)high_crc << 32) | (u64)low_crc;
381 static int match_extent_ref(struct extent_buffer *leaf,
382 struct btrfs_extent_ref *disk_ref,
383 struct btrfs_extent_ref *cpu_ref)
388 if (cpu_ref->objectid)
389 len = sizeof(*cpu_ref);
391 len = 2 * sizeof(u64);
392 ret = memcmp_extent_buffer(leaf, cpu_ref, (unsigned long)disk_ref,
397 static int noinline lookup_extent_backref(struct btrfs_trans_handle *trans,
398 struct btrfs_root *root,
399 struct btrfs_path *path, u64 bytenr,
401 u64 ref_generation, u64 owner,
402 u64 owner_offset, int del)
405 struct btrfs_key key;
406 struct btrfs_key found_key;
407 struct btrfs_extent_ref ref;
408 struct extent_buffer *leaf;
409 struct btrfs_extent_ref *disk_ref;
413 btrfs_set_stack_ref_root(&ref, root_objectid);
414 btrfs_set_stack_ref_generation(&ref, ref_generation);
415 btrfs_set_stack_ref_objectid(&ref, owner);
416 btrfs_set_stack_ref_offset(&ref, owner_offset);
418 hash = hash_extent_ref(root_objectid, ref_generation, owner,
421 key.objectid = bytenr;
422 key.type = BTRFS_EXTENT_REF_KEY;
425 ret = btrfs_search_slot(trans, root, &key, path,
429 leaf = path->nodes[0];
431 u32 nritems = btrfs_header_nritems(leaf);
432 if (path->slots[0] >= nritems) {
433 ret2 = btrfs_next_leaf(root, path);
436 leaf = path->nodes[0];
438 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
439 if (found_key.objectid != bytenr ||
440 found_key.type != BTRFS_EXTENT_REF_KEY)
442 key.offset = found_key.offset;
444 btrfs_release_path(root, path);
448 disk_ref = btrfs_item_ptr(path->nodes[0],
450 struct btrfs_extent_ref);
451 if (match_extent_ref(path->nodes[0], disk_ref, &ref)) {
455 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
456 key.offset = found_key.offset + 1;
457 btrfs_release_path(root, path);
464 * Back reference rules. Back refs have three main goals:
466 * 1) differentiate between all holders of references to an extent so that
467 * when a reference is dropped we can make sure it was a valid reference
468 * before freeing the extent.
470 * 2) Provide enough information to quickly find the holders of an extent
471 * if we notice a given block is corrupted or bad.
473 * 3) Make it easy to migrate blocks for FS shrinking or storage pool
474 * maintenance. This is actually the same as #2, but with a slightly
475 * different use case.
477 * File extents can be referenced by:
479 * - multiple snapshots, subvolumes, or different generations in one subvol
480 * - different files inside a single subvolume (in theory, not implemented yet)
481 * - different offsets inside a file (bookend extents in file.c)
483 * The extent ref structure has fields for:
485 * - Objectid of the subvolume root
486 * - Generation number of the tree holding the reference
487 * - objectid of the file holding the reference
488 * - offset in the file corresponding to the key holding the reference
490 * When a file extent is allocated the fields are filled in:
491 * (root_key.objectid, trans->transid, inode objectid, offset in file)
493 * When a leaf is cow'd new references are added for every file extent found
494 * in the leaf. It looks the same as the create case, but trans->transid
495 * will be different when the block is cow'd.
497 * (root_key.objectid, trans->transid, inode objectid, offset in file)
499 * When a file extent is removed either during snapshot deletion or file
500 * truncation, the corresponding back reference is found
503 * (btrfs_header_owner(leaf), btrfs_header_generation(leaf),
504 * inode objectid, offset in file)
506 * Btree extents can be referenced by:
508 * - Different subvolumes
509 * - Different generations of the same subvolume
511 * Storing sufficient information for a full reverse mapping of a btree
512 * block would require storing the lowest key of the block in the backref,
513 * and it would require updating that lowest key either before write out or
514 * every time it changed. Instead, the objectid of the lowest key is stored
515 * along with the level of the tree block. This provides a hint
516 * about where in the btree the block can be found. Searches through the
517 * btree only need to look for a pointer to that block, so they stop one
518 * level higher than the level recorded in the backref.
520 * Some btrees do not do reference counting on their extents. These
521 * include the extent tree and the tree of tree roots. Backrefs for these
522 * trees always have a generation of zero.
524 * When a tree block is created, back references are inserted:
526 * (root->root_key.objectid, trans->transid or zero, level, lowest_key_objectid)
528 * When a tree block is cow'd in a reference counted root,
529 * new back references are added for all the blocks it points to.
530 * These are of the form (trans->transid will have increased since creation):
532 * (root->root_key.objectid, trans->transid, level, lowest_key_objectid)
534 * Because the lowest_key_objectid and the level are just hints
535 * they are not used when backrefs are deleted. When a backref is deleted:
537 * if backref was for a tree root:
538 * root_objectid = root->root_key.objectid
540 * root_objectid = btrfs_header_owner(parent)
542 * (root_objectid, btrfs_header_generation(parent) or zero, 0, 0)
544 * Back Reference Key hashing:
546 * Back references have four fields, each 64 bits long. Unfortunately,
547 * This is hashed into a single 64 bit number and placed into the key offset.
548 * The key objectid corresponds to the first byte in the extent, and the
549 * key type is set to BTRFS_EXTENT_REF_KEY
551 int btrfs_insert_extent_backref(struct btrfs_trans_handle *trans,
552 struct btrfs_root *root,
553 struct btrfs_path *path, u64 bytenr,
554 u64 root_objectid, u64 ref_generation,
555 u64 owner, u64 owner_offset)
558 struct btrfs_key key;
559 struct btrfs_extent_ref ref;
560 struct btrfs_extent_ref *disk_ref;
563 btrfs_set_stack_ref_root(&ref, root_objectid);
564 btrfs_set_stack_ref_generation(&ref, ref_generation);
565 btrfs_set_stack_ref_objectid(&ref, owner);
566 btrfs_set_stack_ref_offset(&ref, owner_offset);
568 hash = hash_extent_ref(root_objectid, ref_generation, owner,
571 key.objectid = bytenr;
572 key.type = BTRFS_EXTENT_REF_KEY;
574 ret = btrfs_insert_empty_item(trans, root, path, &key, sizeof(ref));
575 while (ret == -EEXIST) {
576 disk_ref = btrfs_item_ptr(path->nodes[0], path->slots[0],
577 struct btrfs_extent_ref);
578 if (match_extent_ref(path->nodes[0], disk_ref, &ref))
581 btrfs_release_path(root, path);
582 ret = btrfs_insert_empty_item(trans, root, path, &key,
587 disk_ref = btrfs_item_ptr(path->nodes[0], path->slots[0],
588 struct btrfs_extent_ref);
589 write_extent_buffer(path->nodes[0], &ref, (unsigned long)disk_ref,
591 btrfs_mark_buffer_dirty(path->nodes[0]);
593 btrfs_release_path(root, path);
597 int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
598 struct btrfs_root *root,
599 u64 bytenr, u64 num_bytes,
600 u64 root_objectid, u64 ref_generation,
601 u64 owner, u64 owner_offset)
603 struct btrfs_path *path;
605 struct btrfs_key key;
606 struct extent_buffer *l;
607 struct btrfs_extent_item *item;
610 WARN_ON(num_bytes < root->sectorsize);
611 path = btrfs_alloc_path();
615 key.objectid = bytenr;
616 btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
617 key.offset = num_bytes;
618 ret = btrfs_search_slot(trans, root->fs_info->extent_root, &key, path,
627 item = btrfs_item_ptr(l, path->slots[0], struct btrfs_extent_item);
628 refs = btrfs_extent_refs(l, item);
629 btrfs_set_extent_refs(l, item, refs + 1);
630 btrfs_mark_buffer_dirty(path->nodes[0]);
632 btrfs_release_path(root->fs_info->extent_root, path);
634 ret = btrfs_insert_extent_backref(trans, root->fs_info->extent_root,
635 path, bytenr, root_objectid,
636 ref_generation, owner, owner_offset);
638 finish_current_insert(trans, root->fs_info->extent_root);
639 del_pending_extents(trans, root->fs_info->extent_root);
641 btrfs_free_path(path);
645 int btrfs_extent_post_op(struct btrfs_trans_handle *trans,
646 struct btrfs_root *root)
648 finish_current_insert(trans, root->fs_info->extent_root);
649 del_pending_extents(trans, root->fs_info->extent_root);
653 static int lookup_extent_ref(struct btrfs_trans_handle *trans,
654 struct btrfs_root *root, u64 bytenr,
655 u64 num_bytes, u32 *refs)
657 struct btrfs_path *path;
659 struct btrfs_key key;
660 struct extent_buffer *l;
661 struct btrfs_extent_item *item;
663 WARN_ON(num_bytes < root->sectorsize);
664 path = btrfs_alloc_path();
665 key.objectid = bytenr;
666 key.offset = num_bytes;
667 btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
668 ret = btrfs_search_slot(trans, root->fs_info->extent_root, &key, path,
673 btrfs_print_leaf(root, path->nodes[0]);
674 printk("failed to find block number %Lu\n", bytenr);
678 item = btrfs_item_ptr(l, path->slots[0], struct btrfs_extent_item);
679 *refs = btrfs_extent_refs(l, item);
681 btrfs_free_path(path);
685 u32 btrfs_count_snapshots_in_path(struct btrfs_root *root,
686 struct btrfs_path *count_path,
689 struct btrfs_root *extent_root = root->fs_info->extent_root;
690 struct btrfs_path *path;
693 u64 root_objectid = root->root_key.objectid;
698 struct btrfs_key key;
699 struct btrfs_key found_key;
700 struct extent_buffer *l;
701 struct btrfs_extent_item *item;
702 struct btrfs_extent_ref *ref_item;
705 path = btrfs_alloc_path();
708 bytenr = first_extent;
710 bytenr = count_path->nodes[level]->start;
713 key.objectid = bytenr;
716 btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
717 ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0);
723 btrfs_item_key_to_cpu(l, &found_key, path->slots[0]);
725 if (found_key.objectid != bytenr ||
726 found_key.type != BTRFS_EXTENT_ITEM_KEY) {
730 item = btrfs_item_ptr(l, path->slots[0], struct btrfs_extent_item);
733 nritems = btrfs_header_nritems(l);
734 if (path->slots[0] >= nritems) {
735 ret = btrfs_next_leaf(extent_root, path);
740 btrfs_item_key_to_cpu(l, &found_key, path->slots[0]);
741 if (found_key.objectid != bytenr)
744 if (found_key.type != BTRFS_EXTENT_REF_KEY) {
750 ref_item = btrfs_item_ptr(l, path->slots[0],
751 struct btrfs_extent_ref);
752 found_objectid = btrfs_ref_root(l, ref_item);
754 if (found_objectid != root_objectid) {
761 if (cur_count == 0) {
765 if (level >= 0 && root->node == count_path->nodes[level])
768 btrfs_release_path(root, path);
772 btrfs_free_path(path);
777 int btrfs_inc_root_ref(struct btrfs_trans_handle *trans,
778 struct btrfs_root *root, u64 owner_objectid)
784 struct btrfs_disk_key disk_key;
786 level = btrfs_header_level(root->node);
787 generation = trans->transid;
788 nritems = btrfs_header_nritems(root->node);
791 btrfs_item_key(root->node, &disk_key, 0);
793 btrfs_node_key(root->node, &disk_key, 0);
794 key_objectid = btrfs_disk_key_objectid(&disk_key);
798 return btrfs_inc_extent_ref(trans, root, root->node->start,
799 root->node->len, owner_objectid,
800 generation, level, key_objectid);
803 int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
804 struct extent_buffer *buf)
808 struct btrfs_key key;
809 struct btrfs_file_extent_item *fi;
818 level = btrfs_header_level(buf);
819 nritems = btrfs_header_nritems(buf);
820 for (i = 0; i < nritems; i++) {
823 btrfs_item_key_to_cpu(buf, &key, i);
824 if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
826 fi = btrfs_item_ptr(buf, i,
827 struct btrfs_file_extent_item);
828 if (btrfs_file_extent_type(buf, fi) ==
829 BTRFS_FILE_EXTENT_INLINE)
831 disk_bytenr = btrfs_file_extent_disk_bytenr(buf, fi);
832 if (disk_bytenr == 0)
834 ret = btrfs_inc_extent_ref(trans, root, disk_bytenr,
835 btrfs_file_extent_disk_num_bytes(buf, fi),
836 root->root_key.objectid, trans->transid,
837 key.objectid, key.offset);
843 bytenr = btrfs_node_blockptr(buf, i);
844 btrfs_node_key_to_cpu(buf, &key, i);
845 ret = btrfs_inc_extent_ref(trans, root, bytenr,
846 btrfs_level_size(root, level - 1),
847 root->root_key.objectid,
849 level - 1, key.objectid);
860 for (i =0; i < faili; i++) {
863 btrfs_item_key_to_cpu(buf, &key, i);
864 if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
866 fi = btrfs_item_ptr(buf, i,
867 struct btrfs_file_extent_item);
868 if (btrfs_file_extent_type(buf, fi) ==
869 BTRFS_FILE_EXTENT_INLINE)
871 disk_bytenr = btrfs_file_extent_disk_bytenr(buf, fi);
872 if (disk_bytenr == 0)
874 err = btrfs_free_extent(trans, root, disk_bytenr,
875 btrfs_file_extent_disk_num_bytes(buf,
879 bytenr = btrfs_node_blockptr(buf, i);
880 err = btrfs_free_extent(trans, root, bytenr,
881 btrfs_level_size(root, level - 1), 0);
889 static int write_one_cache_group(struct btrfs_trans_handle *trans,
890 struct btrfs_root *root,
891 struct btrfs_path *path,
892 struct btrfs_block_group_cache *cache)
896 struct btrfs_root *extent_root = root->fs_info->extent_root;
898 struct extent_buffer *leaf;
900 ret = btrfs_search_slot(trans, extent_root, &cache->key, path, 0, 1);
905 leaf = path->nodes[0];
906 bi = btrfs_item_ptr_offset(leaf, path->slots[0]);
907 write_extent_buffer(leaf, &cache->item, bi, sizeof(cache->item));
908 btrfs_mark_buffer_dirty(leaf);
909 btrfs_release_path(extent_root, path);
911 finish_current_insert(trans, extent_root);
912 pending_ret = del_pending_extents(trans, extent_root);
921 int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans,
922 struct btrfs_root *root)
924 struct extent_map_tree *block_group_cache;
925 struct btrfs_block_group_cache *cache;
929 struct btrfs_path *path;
935 block_group_cache = &root->fs_info->block_group_cache;
936 path = btrfs_alloc_path();
941 ret = find_first_extent_bit(block_group_cache, last,
942 &start, &end, BLOCK_GROUP_DIRTY);
947 ret = get_state_private(block_group_cache, start, &ptr);
951 cache = (struct btrfs_block_group_cache *)(unsigned long)ptr;
952 err = write_one_cache_group(trans, root,
955 * if we fail to write the cache group, we want
956 * to keep it marked dirty in hopes that a later
963 clear_extent_bits(block_group_cache, start, end,
964 BLOCK_GROUP_DIRTY, GFP_NOFS);
966 btrfs_free_path(path);
970 static int update_block_group(struct btrfs_trans_handle *trans,
971 struct btrfs_root *root,
972 u64 bytenr, u64 num_bytes, int alloc,
973 int mark_free, int data)
975 struct btrfs_block_group_cache *cache;
976 struct btrfs_fs_info *info = root->fs_info;
977 u64 total = num_bytes;
984 cache = btrfs_lookup_block_group(info, bytenr);
988 byte_in_group = bytenr - cache->key.objectid;
989 WARN_ON(byte_in_group > cache->key.offset);
990 start = cache->key.objectid;
991 end = start + cache->key.offset - 1;
992 set_extent_bits(&info->block_group_cache, start, end,
993 BLOCK_GROUP_DIRTY, GFP_NOFS);
995 old_val = btrfs_block_group_used(&cache->item);
996 num_bytes = min(total, cache->key.offset - byte_in_group);
998 if (cache->data != data &&
999 old_val < (cache->key.offset >> 1)) {
1004 bit_to_clear = BLOCK_GROUP_METADATA;
1005 bit_to_set = BLOCK_GROUP_DATA;
1006 cache->item.flags &=
1007 ~BTRFS_BLOCK_GROUP_MIXED;
1008 cache->item.flags |=
1009 BTRFS_BLOCK_GROUP_DATA;
1011 bit_to_clear = BLOCK_GROUP_DATA;
1012 bit_to_set = BLOCK_GROUP_METADATA;
1013 cache->item.flags &=
1014 ~BTRFS_BLOCK_GROUP_MIXED;
1015 cache->item.flags &=
1016 ~BTRFS_BLOCK_GROUP_DATA;
1018 clear_extent_bits(&info->block_group_cache,
1019 start, end, bit_to_clear,
1021 set_extent_bits(&info->block_group_cache,
1022 start, end, bit_to_set,
1024 } else if (cache->data != data &&
1025 cache->data != BTRFS_BLOCK_GROUP_MIXED) {
1026 cache->data = BTRFS_BLOCK_GROUP_MIXED;
1027 set_extent_bits(&info->block_group_cache,
1030 BLOCK_GROUP_METADATA,
1033 old_val += num_bytes;
1035 old_val -= num_bytes;
1037 set_extent_dirty(&info->free_space_cache,
1038 bytenr, bytenr + num_bytes - 1,
1042 btrfs_set_block_group_used(&cache->item, old_val);
1044 bytenr += num_bytes;
1048 static int update_pinned_extents(struct btrfs_root *root,
1049 u64 bytenr, u64 num, int pin)
1052 struct btrfs_block_group_cache *cache;
1053 struct btrfs_fs_info *fs_info = root->fs_info;
1056 set_extent_dirty(&fs_info->pinned_extents,
1057 bytenr, bytenr + num - 1, GFP_NOFS);
1059 clear_extent_dirty(&fs_info->pinned_extents,
1060 bytenr, bytenr + num - 1, GFP_NOFS);
1063 cache = btrfs_lookup_block_group(fs_info, bytenr);
1065 len = min(num, cache->key.offset -
1066 (bytenr - cache->key.objectid));
1068 cache->pinned += len;
1069 fs_info->total_pinned += len;
1071 cache->pinned -= len;
1072 fs_info->total_pinned -= len;
1080 int btrfs_copy_pinned(struct btrfs_root *root, struct extent_map_tree *copy)
1085 struct extent_map_tree *pinned_extents = &root->fs_info->pinned_extents;
1089 ret = find_first_extent_bit(pinned_extents, last,
1090 &start, &end, EXTENT_DIRTY);
1093 set_extent_dirty(copy, start, end, GFP_NOFS);
1099 int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
1100 struct btrfs_root *root,
1101 struct extent_map_tree *unpin)
1106 struct extent_map_tree *free_space_cache;
1107 free_space_cache = &root->fs_info->free_space_cache;
1110 ret = find_first_extent_bit(unpin, 0, &start, &end,
1114 update_pinned_extents(root, start, end + 1 - start, 0);
1115 clear_extent_dirty(unpin, start, end, GFP_NOFS);
1116 set_extent_dirty(free_space_cache, start, end, GFP_NOFS);
1121 static int finish_current_insert(struct btrfs_trans_handle *trans,
1122 struct btrfs_root *extent_root)
1126 struct btrfs_fs_info *info = extent_root->fs_info;
1127 struct extent_buffer *eb;
1128 struct btrfs_path *path;
1129 struct btrfs_key ins;
1130 struct btrfs_disk_key first;
1131 struct btrfs_extent_item extent_item;
1136 btrfs_set_stack_extent_refs(&extent_item, 1);
1137 btrfs_set_key_type(&ins, BTRFS_EXTENT_ITEM_KEY);
1138 path = btrfs_alloc_path();
1141 ret = find_first_extent_bit(&info->extent_ins, 0, &start,
1142 &end, EXTENT_LOCKED);
1146 ins.objectid = start;
1147 ins.offset = end + 1 - start;
1148 err = btrfs_insert_item(trans, extent_root, &ins,
1149 &extent_item, sizeof(extent_item));
1150 clear_extent_bits(&info->extent_ins, start, end, EXTENT_LOCKED,
1152 eb = read_tree_block(extent_root, ins.objectid, ins.offset);
1153 level = btrfs_header_level(eb);
1155 btrfs_item_key(eb, &first, 0);
1157 btrfs_node_key(eb, &first, 0);
1159 err = btrfs_insert_extent_backref(trans, extent_root, path,
1160 start, extent_root->root_key.objectid,
1162 btrfs_disk_key_objectid(&first));
1164 free_extent_buffer(eb);
1166 btrfs_free_path(path);
1170 static int pin_down_bytes(struct btrfs_root *root, u64 bytenr, u32 num_bytes,
1174 struct extent_buffer *buf;
1177 buf = btrfs_find_tree_block(root, bytenr, num_bytes);
1179 if (btrfs_buffer_uptodate(buf)) {
1181 root->fs_info->running_transaction->transid;
1182 if (btrfs_header_generation(buf) == transid) {
1183 free_extent_buffer(buf);
1187 free_extent_buffer(buf);
1189 update_pinned_extents(root, bytenr, num_bytes, 1);
1191 set_extent_bits(&root->fs_info->pending_del,
1192 bytenr, bytenr + num_bytes - 1,
1193 EXTENT_LOCKED, GFP_NOFS);
1200 * remove an extent from the root, returns 0 on success
1202 static int __free_extent(struct btrfs_trans_handle *trans, struct btrfs_root
1203 *root, u64 bytenr, u64 num_bytes,
1204 u64 root_objectid, u64 ref_generation,
1205 u64 owner_objectid, u64 owner_offset, int pin,
1208 struct btrfs_path *path;
1209 struct btrfs_key key;
1210 struct btrfs_fs_info *info = root->fs_info;
1211 struct btrfs_root *extent_root = info->extent_root;
1212 struct extent_buffer *leaf;
1214 struct btrfs_extent_item *ei;
1217 key.objectid = bytenr;
1218 btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY);
1219 key.offset = num_bytes;
1221 path = btrfs_alloc_path();
1225 ret = lookup_extent_backref(trans, extent_root, path,
1226 bytenr, root_objectid,
1228 owner_objectid, owner_offset, 1);
1230 ret = btrfs_del_item(trans, extent_root, path);
1232 btrfs_print_leaf(extent_root, path->nodes[0]);
1234 printk("Unable to find ref byte nr %Lu root %Lu "
1235 " gen %Lu owner %Lu offset %Lu\n", bytenr,
1236 root_objectid, ref_generation, owner_objectid,
1239 btrfs_release_path(extent_root, path);
1240 ret = btrfs_search_slot(trans, extent_root, &key, path, -1, 1);
1245 leaf = path->nodes[0];
1246 ei = btrfs_item_ptr(leaf, path->slots[0],
1247 struct btrfs_extent_item);
1248 refs = btrfs_extent_refs(leaf, ei);
1251 btrfs_set_extent_refs(leaf, ei, refs);
1252 btrfs_mark_buffer_dirty(leaf);
1259 ret = pin_down_bytes(root, bytenr, num_bytes, 0);
1265 /* block accounting for super block */
1266 super_used = btrfs_super_bytes_used(&info->super_copy);
1267 btrfs_set_super_bytes_used(&info->super_copy,
1268 super_used - num_bytes);
1270 /* block accounting for root item */
1271 root_used = btrfs_root_used(&root->root_item);
1272 btrfs_set_root_used(&root->root_item,
1273 root_used - num_bytes);
1275 ret = btrfs_del_item(trans, extent_root, path);
1279 ret = update_block_group(trans, root, bytenr, num_bytes, 0,
1283 btrfs_free_path(path);
1284 finish_current_insert(trans, extent_root);
1289 * find all the blocks marked as pending in the radix tree and remove
1290 * them from the extent map
1292 static int del_pending_extents(struct btrfs_trans_handle *trans, struct
1293 btrfs_root *extent_root)
1299 struct extent_map_tree *pending_del;
1300 struct extent_map_tree *pinned_extents;
1302 pending_del = &extent_root->fs_info->pending_del;
1303 pinned_extents = &extent_root->fs_info->pinned_extents;
1306 ret = find_first_extent_bit(pending_del, 0, &start, &end,
1310 update_pinned_extents(extent_root, start, end + 1 - start, 1);
1311 clear_extent_bits(pending_del, start, end, EXTENT_LOCKED,
1313 ret = __free_extent(trans, extent_root,
1314 start, end + 1 - start,
1315 extent_root->root_key.objectid,
1324 * remove an extent from the root, returns 0 on success
1326 int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_root
1327 *root, u64 bytenr, u64 num_bytes,
1328 u64 root_objectid, u64 ref_generation,
1329 u64 owner_objectid, u64 owner_offset, int pin)
1331 struct btrfs_root *extent_root = root->fs_info->extent_root;
1335 WARN_ON(num_bytes < root->sectorsize);
1336 if (!root->ref_cows)
1339 if (root == extent_root) {
1340 pin_down_bytes(root, bytenr, num_bytes, 1);
1343 ret = __free_extent(trans, root, bytenr, num_bytes, root_objectid,
1344 ref_generation, owner_objectid, owner_offset,
1346 pending_ret = del_pending_extents(trans, root->fs_info->extent_root);
1347 return ret ? ret : pending_ret;
1350 static u64 stripe_align(struct btrfs_root *root, u64 val)
1352 u64 mask = ((u64)root->stripesize - 1);
1353 u64 ret = (val + mask) & ~mask;
1358 * walks the btree of allocated extents and find a hole of a given size.
1359 * The key ins is changed to record the hole:
1360 * ins->objectid == block start
1361 * ins->flags = BTRFS_EXTENT_ITEM_KEY
1362 * ins->offset == number of blocks
1363 * Any available blocks before search_start are skipped.
1365 static int noinline find_free_extent(struct btrfs_trans_handle *trans,
1366 struct btrfs_root *orig_root,
1367 u64 num_bytes, u64 empty_size,
1368 u64 search_start, u64 search_end,
1369 u64 hint_byte, struct btrfs_key *ins,
1370 u64 exclude_start, u64 exclude_nr,
1373 struct btrfs_path *path;
1374 struct btrfs_key key;
1380 u64 orig_search_start = search_start;
1382 struct extent_buffer *l;
1383 struct btrfs_root * root = orig_root->fs_info->extent_root;
1384 struct btrfs_fs_info *info = root->fs_info;
1385 u64 total_needed = num_bytes;
1387 struct btrfs_block_group_cache *block_group;
1392 WARN_ON(num_bytes < root->sectorsize);
1393 btrfs_set_key_type(ins, BTRFS_EXTENT_ITEM_KEY);
1395 level = btrfs_header_level(root->node);
1397 if (num_bytes >= 32 * 1024 * 1024 && hint_byte) {
1398 data = BTRFS_BLOCK_GROUP_MIXED;
1401 if (search_end == (u64)-1)
1402 search_end = btrfs_super_total_bytes(&info->super_copy);
1404 block_group = btrfs_lookup_block_group(info, hint_byte);
1406 hint_byte = search_start;
1407 block_group = btrfs_find_block_group(root, block_group,
1408 hint_byte, data, 1);
1410 block_group = btrfs_find_block_group(root,
1412 search_start, data, 1);
1415 total_needed += empty_size;
1416 path = btrfs_alloc_path();
1419 block_group = btrfs_lookup_block_group(info, search_start);
1421 block_group = btrfs_lookup_block_group(info,
1424 search_start = find_search_start(root, &block_group, search_start,
1425 total_needed, data, full_scan);
1426 search_start = stripe_align(root, search_start);
1427 cached_start = search_start;
1428 btrfs_init_path(path);
1429 ins->objectid = search_start;
1434 ret = btrfs_search_slot(trans, root, ins, path, 0, 0);
1438 if (path->slots[0] > 0) {
1443 btrfs_item_key_to_cpu(l, &key, path->slots[0]);
1446 * walk backwards to find the first extent item key
1448 while(btrfs_key_type(&key) != BTRFS_EXTENT_ITEM_KEY) {
1449 if (path->slots[0] == 0) {
1450 ret = btrfs_prev_leaf(root, path);
1452 ret = btrfs_search_slot(trans, root, ins,
1456 if (path->slots[0] > 0)
1464 btrfs_item_key_to_cpu(l, &key, path->slots[0]);
1468 slot = path->slots[0];
1469 if (slot >= btrfs_header_nritems(l)) {
1470 ret = btrfs_next_leaf(root, path);
1476 search_start = max(search_start,
1477 block_group->key.objectid);
1479 aligned = stripe_align(root, search_start);
1480 ins->objectid = aligned;
1481 if (aligned >= search_end) {
1485 ins->offset = search_end - aligned;
1489 ins->objectid = stripe_align(root,
1490 last_byte > search_start ?
1491 last_byte : search_start);
1492 if (search_end <= ins->objectid) {
1496 ins->offset = search_end - ins->objectid;
1497 BUG_ON(ins->objectid >= search_end);
1500 btrfs_item_key_to_cpu(l, &key, slot);
1502 if (key.objectid >= search_start && key.objectid > last_byte &&
1504 if (last_byte < search_start)
1505 last_byte = search_start;
1506 aligned = stripe_align(root, last_byte);
1507 hole_size = key.objectid - aligned;
1508 if (key.objectid > aligned && hole_size >= num_bytes) {
1509 ins->objectid = aligned;
1510 ins->offset = hole_size;
1514 if (btrfs_key_type(&key) != BTRFS_EXTENT_ITEM_KEY) {
1515 if (!start_found && btrfs_key_type(&key) ==
1516 BTRFS_BLOCK_GROUP_ITEM_KEY) {
1517 last_byte = key.objectid;
1525 last_byte = key.objectid + key.offset;
1527 if (!full_scan && data != BTRFS_BLOCK_GROUP_MIXED &&
1528 last_byte >= block_group->key.objectid +
1529 block_group->key.offset) {
1530 btrfs_release_path(root, path);
1531 search_start = block_group->key.objectid +
1532 block_group->key.offset;
1540 /* we have to make sure we didn't find an extent that has already
1541 * been allocated by the map tree or the original allocation
1543 btrfs_release_path(root, path);
1544 BUG_ON(ins->objectid < search_start);
1546 if (ins->objectid + num_bytes >= search_end)
1548 if (!full_scan && data != BTRFS_BLOCK_GROUP_MIXED &&
1549 ins->objectid + num_bytes > block_group->
1550 key.objectid + block_group->key.offset) {
1551 search_start = block_group->key.objectid +
1552 block_group->key.offset;
1555 if (test_range_bit(&info->extent_ins, ins->objectid,
1556 ins->objectid + num_bytes -1, EXTENT_LOCKED, 0)) {
1557 search_start = ins->objectid + num_bytes;
1560 if (test_range_bit(&info->pinned_extents, ins->objectid,
1561 ins->objectid + num_bytes -1, EXTENT_DIRTY, 0)) {
1562 search_start = ins->objectid + num_bytes;
1565 if (exclude_nr > 0 && (ins->objectid + num_bytes > exclude_start &&
1566 ins->objectid < exclude_start + exclude_nr)) {
1567 search_start = exclude_start + exclude_nr;
1571 block_group = btrfs_lookup_block_group(info, ins->objectid);
1573 trans->block_group = block_group;
1575 ins->offset = num_bytes;
1576 btrfs_free_path(path);
1580 if (search_start + num_bytes >= search_end) {
1582 search_start = orig_search_start;
1589 total_needed -= empty_size;
1591 data = BTRFS_BLOCK_GROUP_MIXED;
1595 block_group = btrfs_lookup_block_group(info, search_start);
1597 block_group = btrfs_find_block_group(root, block_group,
1598 search_start, data, 0);
1602 btrfs_release_path(root, path);
1603 btrfs_free_path(path);
1607 * finds a free extent and does all the dirty work required for allocation
1608 * returns the key for the extent through ins, and a tree buffer for
1609 * the first block of the extent through buf.
1611 * returns 0 if everything worked, non-zero otherwise.
1613 int btrfs_alloc_extent(struct btrfs_trans_handle *trans,
1614 struct btrfs_root *root,
1615 u64 num_bytes, u64 root_objectid, u64 ref_generation,
1616 u64 owner, u64 owner_offset,
1617 u64 empty_size, u64 hint_byte,
1618 u64 search_end, struct btrfs_key *ins, int data)
1622 u64 super_used, root_used;
1623 u64 search_start = 0;
1625 struct btrfs_fs_info *info = root->fs_info;
1626 struct btrfs_root *extent_root = info->extent_root;
1627 struct btrfs_extent_item extent_item;
1628 struct btrfs_path *path;
1630 btrfs_set_stack_extent_refs(&extent_item, 1);
1632 new_hint = max(hint_byte, root->fs_info->alloc_start);
1633 if (new_hint < btrfs_super_total_bytes(&info->super_copy))
1634 hint_byte = new_hint;
1636 WARN_ON(num_bytes < root->sectorsize);
1637 ret = find_free_extent(trans, root, num_bytes, empty_size,
1638 search_start, search_end, hint_byte, ins,
1639 trans->alloc_exclude_start,
1640 trans->alloc_exclude_nr, data);
1645 /* block accounting for super block */
1646 super_used = btrfs_super_bytes_used(&info->super_copy);
1647 btrfs_set_super_bytes_used(&info->super_copy, super_used + num_bytes);
1649 /* block accounting for root item */
1650 root_used = btrfs_root_used(&root->root_item);
1651 btrfs_set_root_used(&root->root_item, root_used + num_bytes);
1653 clear_extent_dirty(&root->fs_info->free_space_cache,
1654 ins->objectid, ins->objectid + ins->offset - 1,
1657 if (root == extent_root) {
1658 set_extent_bits(&root->fs_info->extent_ins, ins->objectid,
1659 ins->objectid + ins->offset - 1,
1660 EXTENT_LOCKED, GFP_NOFS);
1665 WARN_ON(trans->alloc_exclude_nr);
1666 trans->alloc_exclude_start = ins->objectid;
1667 trans->alloc_exclude_nr = ins->offset;
1668 ret = btrfs_insert_item(trans, extent_root, ins, &extent_item,
1669 sizeof(extent_item));
1671 trans->alloc_exclude_start = 0;
1672 trans->alloc_exclude_nr = 0;
1675 path = btrfs_alloc_path();
1677 ret = btrfs_insert_extent_backref(trans, extent_root, path,
1678 ins->objectid, root_objectid,
1679 ref_generation, owner, owner_offset);
1682 btrfs_free_path(path);
1683 finish_current_insert(trans, extent_root);
1684 pending_ret = del_pending_extents(trans, extent_root);
1694 ret = update_block_group(trans, root, ins->objectid, ins->offset, 1, 0,
1701 * helper function to allocate a block for a given tree
1702 * returns the tree buffer or NULL.
1704 struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
1705 struct btrfs_root *root,
1707 u64 root_objectid, u64 hint,
1713 ref_generation = trans->transid;
1718 return __btrfs_alloc_free_block(trans, root, blocksize, root_objectid,
1719 ref_generation, 0, 0, hint, empty_size);
1723 * helper function to allocate a block for a given tree
1724 * returns the tree buffer or NULL.
1726 struct extent_buffer *__btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
1727 struct btrfs_root *root,
1736 struct btrfs_key ins;
1738 struct extent_buffer *buf;
1740 ret = btrfs_alloc_extent(trans, root, blocksize,
1741 root_objectid, ref_generation,
1742 level, first_objectid, empty_size, hint,
1746 return ERR_PTR(ret);
1748 buf = btrfs_find_create_tree_block(root, ins.objectid, blocksize);
1750 btrfs_free_extent(trans, root, ins.objectid, blocksize,
1751 root->root_key.objectid, ref_generation,
1753 return ERR_PTR(-ENOMEM);
1755 btrfs_set_buffer_uptodate(buf);
1756 set_extent_dirty(&trans->transaction->dirty_pages, buf->start,
1757 buf->start + buf->len - 1, GFP_NOFS);
1758 set_extent_bits(&BTRFS_I(root->fs_info->btree_inode)->extent_tree,
1759 buf->start, buf->start + buf->len - 1,
1760 EXTENT_CSUM, GFP_NOFS);
1761 buf->flags |= EXTENT_CSUM;
1762 btrfs_set_buffer_defrag(buf);
1763 trans->blocks_used++;
1767 static int noinline drop_leaf_ref(struct btrfs_trans_handle *trans,
1768 struct btrfs_root *root,
1769 struct extent_buffer *leaf)
1772 u64 leaf_generation;
1773 struct btrfs_key key;
1774 struct btrfs_file_extent_item *fi;
1779 BUG_ON(!btrfs_is_leaf(leaf));
1780 nritems = btrfs_header_nritems(leaf);
1781 leaf_owner = btrfs_header_owner(leaf);
1782 leaf_generation = btrfs_header_generation(leaf);
1784 for (i = 0; i < nritems; i++) {
1787 btrfs_item_key_to_cpu(leaf, &key, i);
1788 if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
1790 fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
1791 if (btrfs_file_extent_type(leaf, fi) ==
1792 BTRFS_FILE_EXTENT_INLINE)
1795 * FIXME make sure to insert a trans record that
1796 * repeats the snapshot del on crash
1798 disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
1799 if (disk_bytenr == 0)
1801 ret = btrfs_free_extent(trans, root, disk_bytenr,
1802 btrfs_file_extent_disk_num_bytes(leaf, fi),
1803 leaf_owner, leaf_generation,
1804 key.objectid, key.offset, 0);
1810 static void noinline reada_walk_down(struct btrfs_root *root,
1811 struct extent_buffer *node)
1821 nritems = btrfs_header_nritems(node);
1822 level = btrfs_header_level(node);
1823 for (i = 0; i < nritems; i++) {
1824 bytenr = btrfs_node_blockptr(node, i);
1825 blocksize = btrfs_level_size(root, level - 1);
1826 ret = lookup_extent_ref(NULL, root, bytenr, blocksize, &refs);
1830 mutex_unlock(&root->fs_info->fs_mutex);
1831 ret = readahead_tree_block(root, bytenr, blocksize);
1833 mutex_lock(&root->fs_info->fs_mutex);
1840 * helper function for drop_snapshot, this walks down the tree dropping ref
1841 * counts as it goes.
1843 static int noinline walk_down_tree(struct btrfs_trans_handle *trans,
1844 struct btrfs_root *root,
1845 struct btrfs_path *path, int *level)
1850 struct extent_buffer *next;
1851 struct extent_buffer *cur;
1852 struct extent_buffer *parent;
1857 WARN_ON(*level < 0);
1858 WARN_ON(*level >= BTRFS_MAX_LEVEL);
1859 ret = lookup_extent_ref(trans, root,
1860 path->nodes[*level]->start,
1861 path->nodes[*level]->len, &refs);
1867 * walk down to the last node level and free all the leaves
1869 while(*level >= 0) {
1870 WARN_ON(*level < 0);
1871 WARN_ON(*level >= BTRFS_MAX_LEVEL);
1872 cur = path->nodes[*level];
1874 if (*level > 0 && path->slots[*level] == 0)
1875 reada_walk_down(root, cur);
1877 if (btrfs_header_level(cur) != *level)
1880 if (path->slots[*level] >=
1881 btrfs_header_nritems(cur))
1884 ret = drop_leaf_ref(trans, root, cur);
1888 bytenr = btrfs_node_blockptr(cur, path->slots[*level]);
1889 blocksize = btrfs_level_size(root, *level - 1);
1890 ret = lookup_extent_ref(trans, root, bytenr, blocksize, &refs);
1893 parent = path->nodes[*level];
1894 root_owner = btrfs_header_owner(parent);
1895 root_gen = btrfs_header_generation(parent);
1896 path->slots[*level]++;
1897 ret = btrfs_free_extent(trans, root, bytenr,
1898 blocksize, root_owner,
1903 next = btrfs_find_tree_block(root, bytenr, blocksize);
1904 if (!next || !btrfs_buffer_uptodate(next)) {
1905 free_extent_buffer(next);
1906 mutex_unlock(&root->fs_info->fs_mutex);
1907 next = read_tree_block(root, bytenr, blocksize);
1908 mutex_lock(&root->fs_info->fs_mutex);
1910 /* we dropped the lock, check one more time */
1911 ret = lookup_extent_ref(trans, root, bytenr,
1915 parent = path->nodes[*level];
1916 root_owner = btrfs_header_owner(parent);
1917 root_gen = btrfs_header_generation(parent);
1919 path->slots[*level]++;
1920 free_extent_buffer(next);
1921 ret = btrfs_free_extent(trans, root, bytenr,
1929 WARN_ON(*level <= 0);
1930 if (path->nodes[*level-1])
1931 free_extent_buffer(path->nodes[*level-1]);
1932 path->nodes[*level-1] = next;
1933 *level = btrfs_header_level(next);
1934 path->slots[*level] = 0;
1937 WARN_ON(*level < 0);
1938 WARN_ON(*level >= BTRFS_MAX_LEVEL);
1940 if (path->nodes[*level] == root->node) {
1941 root_owner = root->root_key.objectid;
1942 parent = path->nodes[*level];
1944 parent = path->nodes[*level + 1];
1945 root_owner = btrfs_header_owner(parent);
1948 root_gen = btrfs_header_generation(parent);
1949 ret = btrfs_free_extent(trans, root, path->nodes[*level]->start,
1950 path->nodes[*level]->len,
1951 root_owner, root_gen, 0, 0, 1);
1952 free_extent_buffer(path->nodes[*level]);
1953 path->nodes[*level] = NULL;
1960 * helper for dropping snapshots. This walks back up the tree in the path
1961 * to find the first node higher up where we haven't yet gone through
1964 static int noinline walk_up_tree(struct btrfs_trans_handle *trans,
1965 struct btrfs_root *root,
1966 struct btrfs_path *path, int *level)
1970 struct btrfs_root_item *root_item = &root->root_item;
1975 for(i = *level; i < BTRFS_MAX_LEVEL - 1 && path->nodes[i]; i++) {
1976 slot = path->slots[i];
1977 if (slot < btrfs_header_nritems(path->nodes[i]) - 1) {
1978 struct extent_buffer *node;
1979 struct btrfs_disk_key disk_key;
1980 node = path->nodes[i];
1983 WARN_ON(*level == 0);
1984 btrfs_node_key(node, &disk_key, path->slots[i]);
1985 memcpy(&root_item->drop_progress,
1986 &disk_key, sizeof(disk_key));
1987 root_item->drop_level = i;
1990 if (path->nodes[*level] == root->node) {
1991 root_owner = root->root_key.objectid;
1993 btrfs_header_generation(path->nodes[*level]);
1995 struct extent_buffer *node;
1996 node = path->nodes[*level + 1];
1997 root_owner = btrfs_header_owner(node);
1998 root_gen = btrfs_header_generation(node);
2000 ret = btrfs_free_extent(trans, root,
2001 path->nodes[*level]->start,
2002 path->nodes[*level]->len,
2003 root_owner, root_gen, 0, 0, 1);
2005 free_extent_buffer(path->nodes[*level]);
2006 path->nodes[*level] = NULL;
2014 * drop the reference count on the tree rooted at 'snap'. This traverses
2015 * the tree freeing any blocks that have a ref count of zero after being
2018 int btrfs_drop_snapshot(struct btrfs_trans_handle *trans, struct btrfs_root
2024 struct btrfs_path *path;
2027 struct btrfs_root_item *root_item = &root->root_item;
2029 path = btrfs_alloc_path();
2032 level = btrfs_header_level(root->node);
2034 if (btrfs_disk_key_objectid(&root_item->drop_progress) == 0) {
2035 path->nodes[level] = root->node;
2036 extent_buffer_get(root->node);
2037 path->slots[level] = 0;
2039 struct btrfs_key key;
2040 struct btrfs_disk_key found_key;
2041 struct extent_buffer *node;
2043 btrfs_disk_key_to_cpu(&key, &root_item->drop_progress);
2044 level = root_item->drop_level;
2045 path->lowest_level = level;
2046 wret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
2051 node = path->nodes[level];
2052 btrfs_node_key(node, &found_key, path->slots[level]);
2053 WARN_ON(memcmp(&found_key, &root_item->drop_progress,
2054 sizeof(found_key)));
2057 wret = walk_down_tree(trans, root, path, &level);
2063 wret = walk_up_tree(trans, root, path, &level);
2071 for (i = 0; i <= orig_level; i++) {
2072 if (path->nodes[i]) {
2073 free_extent_buffer(path->nodes[i]);
2074 path->nodes[i] = NULL;
2078 btrfs_free_path(path);
2082 int btrfs_free_block_groups(struct btrfs_fs_info *info)
2089 ret = find_first_extent_bit(&info->block_group_cache, 0,
2090 &start, &end, (unsigned int)-1);
2093 ret = get_state_private(&info->block_group_cache, start, &ptr);
2095 kfree((void *)(unsigned long)ptr);
2096 clear_extent_bits(&info->block_group_cache, start,
2097 end, (unsigned int)-1, GFP_NOFS);
2100 ret = find_first_extent_bit(&info->free_space_cache, 0,
2101 &start, &end, EXTENT_DIRTY);
2104 clear_extent_dirty(&info->free_space_cache, start,
2110 static int noinline relocate_inode_pages(struct inode *inode, u64 start,
2116 u64 existing_delalloc;
2117 unsigned long last_index;
2120 struct btrfs_root *root = BTRFS_I(inode)->root;
2121 struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
2122 struct file_ra_state *ra;
2124 ra = kzalloc(sizeof(*ra), GFP_NOFS);
2126 mutex_lock(&inode->i_mutex);
2127 i = start >> PAGE_CACHE_SHIFT;
2128 last_index = (start + len - 1) >> PAGE_CACHE_SHIFT;
2130 file_ra_state_init(ra, inode->i_mapping);
2131 btrfs_force_ra(inode->i_mapping, ra, NULL, i, last_index);
2134 for (; i <= last_index; i++) {
2135 page = grab_cache_page(inode->i_mapping, i);
2138 if (!PageUptodate(page)) {
2139 btrfs_readpage(NULL, page);
2141 if (!PageUptodate(page)) {
2143 page_cache_release(page);
2147 page_start = (u64)page->index << PAGE_CACHE_SHIFT;
2148 page_end = page_start + PAGE_CACHE_SIZE - 1;
2150 lock_extent(em_tree, page_start, page_end, GFP_NOFS);
2152 delalloc_start = page_start;
2154 count_range_bits(&BTRFS_I(inode)->extent_tree,
2155 &delalloc_start, page_end,
2156 PAGE_CACHE_SIZE, EXTENT_DELALLOC);
2158 set_extent_delalloc(em_tree, page_start,
2159 page_end, GFP_NOFS);
2161 spin_lock(&root->fs_info->delalloc_lock);
2162 root->fs_info->delalloc_bytes += PAGE_CACHE_SIZE -
2164 spin_unlock(&root->fs_info->delalloc_lock);
2166 unlock_extent(em_tree, page_start, page_end, GFP_NOFS);
2167 set_page_dirty(page);
2169 page_cache_release(page);
2173 mutex_unlock(&inode->i_mutex);
2178 * note, this releases the path
2180 static int noinline relocate_one_reference(struct btrfs_root *extent_root,
2181 struct btrfs_path *path,
2182 struct btrfs_key *extent_key)
2184 struct inode *inode;
2185 struct btrfs_root *found_root;
2186 struct btrfs_key *root_location;
2187 struct btrfs_extent_ref *ref;
2194 ref = btrfs_item_ptr(path->nodes[0], path->slots[0],
2195 struct btrfs_extent_ref);
2196 ref_root = btrfs_ref_root(path->nodes[0], ref);
2197 ref_gen = btrfs_ref_generation(path->nodes[0], ref);
2198 ref_objectid = btrfs_ref_objectid(path->nodes[0], ref);
2199 ref_offset = btrfs_ref_offset(path->nodes[0], ref);
2200 btrfs_release_path(extent_root, path);
2202 root_location = kmalloc(sizeof(*root_location), GFP_NOFS);
2203 root_location->objectid = ref_root;
2205 root_location->offset = 0;
2207 root_location->offset = (u64)-1;
2208 root_location->type = BTRFS_ROOT_ITEM_KEY;
2210 found_root = btrfs_read_fs_root_no_name(extent_root->fs_info,
2212 BUG_ON(!found_root);
2213 kfree(root_location);
2215 if (ref_objectid >= BTRFS_FIRST_FREE_OBJECTID) {
2216 mutex_unlock(&extent_root->fs_info->fs_mutex);
2217 inode = btrfs_iget_locked(extent_root->fs_info->sb,
2218 ref_objectid, found_root);
2219 if (inode->i_state & I_NEW) {
2220 /* the inode and parent dir are two different roots */
2221 BTRFS_I(inode)->root = found_root;
2222 BTRFS_I(inode)->location.objectid = ref_objectid;
2223 BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY;
2224 BTRFS_I(inode)->location.offset = 0;
2225 btrfs_read_locked_inode(inode);
2226 unlock_new_inode(inode);
2229 /* this can happen if the reference is not against
2230 * the latest version of the tree root
2232 if (is_bad_inode(inode)) {
2233 mutex_lock(&extent_root->fs_info->fs_mutex);
2236 relocate_inode_pages(inode, ref_offset, extent_key->offset);
2237 /* FIXME, data=ordered will help get rid of this */
2238 filemap_fdatawrite(inode->i_mapping);
2240 mutex_lock(&extent_root->fs_info->fs_mutex);
2242 struct btrfs_trans_handle *trans;
2243 struct btrfs_key found_key;
2244 struct extent_buffer *eb;
2248 trans = btrfs_start_transaction(found_root, 1);
2249 eb = read_tree_block(found_root, extent_key->objectid,
2250 extent_key->offset);
2251 level = btrfs_header_level(eb);
2254 btrfs_item_key_to_cpu(eb, &found_key, 0);
2256 btrfs_node_key_to_cpu(eb, &found_key, 0);
2258 free_extent_buffer(eb);
2260 path->lowest_level = level;
2262 ret = btrfs_search_slot(trans, found_root, &found_key, path,
2264 path->lowest_level = 0;
2265 for (i = level; i < BTRFS_MAX_LEVEL; i++) {
2266 if (!path->nodes[i])
2268 free_extent_buffer(path->nodes[i]);
2269 path->nodes[i] = NULL;
2271 btrfs_release_path(found_root, path);
2272 btrfs_end_transaction(trans, found_root);
2279 static int noinline relocate_one_extent(struct btrfs_root *extent_root,
2280 struct btrfs_path *path,
2281 struct btrfs_key *extent_key)
2283 struct btrfs_key key;
2284 struct btrfs_key found_key;
2285 struct extent_buffer *leaf;
2290 key.objectid = extent_key->objectid;
2291 key.type = BTRFS_EXTENT_REF_KEY;
2295 ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0);
2303 leaf = path->nodes[0];
2304 nritems = btrfs_header_nritems(leaf);
2305 if (path->slots[0] == nritems)
2308 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
2309 if (found_key.objectid != extent_key->objectid)
2312 if (found_key.type != BTRFS_EXTENT_REF_KEY)
2315 key.offset = found_key.offset + 1;
2316 item_size = btrfs_item_size_nr(leaf, path->slots[0]);
2318 ret = relocate_one_reference(extent_root, path, extent_key);
2324 btrfs_release_path(extent_root, path);
2328 static int find_overlapping_extent(struct btrfs_root *root,
2329 struct btrfs_path *path, u64 new_size)
2331 struct btrfs_key found_key;
2332 struct extent_buffer *leaf;
2336 if (path->slots[0] == 0) {
2337 ret = btrfs_prev_leaf(root, path);
2346 leaf = path->nodes[0];
2347 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
2348 if (found_key.type == BTRFS_EXTENT_ITEM_KEY) {
2349 if (found_key.objectid + found_key.offset > new_size)
2358 int btrfs_shrink_extent_tree(struct btrfs_root *root, u64 new_size)
2360 struct btrfs_trans_handle *trans;
2361 struct btrfs_root *tree_root = root->fs_info->tree_root;
2362 struct btrfs_path *path;
2365 struct btrfs_fs_info *info = root->fs_info;
2366 struct extent_map_tree *block_group_cache;
2367 struct btrfs_key key;
2368 struct btrfs_key found_key = { 0, 0, 0 };
2369 struct extent_buffer *leaf;
2374 btrfs_set_super_total_bytes(&info->super_copy, new_size);
2375 block_group_cache = &info->block_group_cache;
2376 path = btrfs_alloc_path();
2377 root = root->fs_info->extent_root;
2382 key.objectid = new_size;
2383 cur_byte = key.objectid;
2388 ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
2392 leaf = path->nodes[0];
2393 if (key.objectid == new_size - 1) {
2394 ret = find_overlapping_extent(root, path, new_size);
2396 btrfs_release_path(root, path);
2397 ret = btrfs_search_slot(NULL, root, &key,
2403 nritems = btrfs_header_nritems(leaf);
2405 slot = path->slots[0];
2407 btrfs_item_key_to_cpu(leaf, &found_key, slot);
2408 if (slot == nritems ||
2409 btrfs_key_type(&found_key) != BTRFS_EXTENT_ITEM_KEY) {
2411 if (path->slots[0] >= nritems) {
2412 ret = btrfs_next_leaf(root, path);
2422 btrfs_item_key_to_cpu(leaf, &found_key, slot);
2423 if (found_key.objectid + found_key.offset <= cur_byte)
2426 cur_byte = found_key.objectid + found_key.offset;
2427 key.objectid = cur_byte;
2428 btrfs_release_path(root, path);
2429 ret = relocate_one_extent(root, path, &found_key);
2432 btrfs_release_path(root, path);
2434 if (total_found > 0) {
2435 trans = btrfs_start_transaction(tree_root, 1);
2436 btrfs_commit_transaction(trans, tree_root);
2438 mutex_unlock(&root->fs_info->fs_mutex);
2439 btrfs_clean_old_snapshots(tree_root);
2440 mutex_lock(&root->fs_info->fs_mutex);
2442 trans = btrfs_start_transaction(tree_root, 1);
2443 btrfs_commit_transaction(trans, tree_root);
2447 trans = btrfs_start_transaction(root, 1);
2448 key.objectid = new_size;
2454 ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
2458 leaf = path->nodes[0];
2459 nritems = btrfs_header_nritems(leaf);
2461 slot = path->slots[0];
2463 btrfs_item_key_to_cpu(leaf, &found_key, slot);
2464 if (slot == nritems ||
2465 btrfs_key_type(&found_key) != BTRFS_BLOCK_GROUP_ITEM_KEY) {
2466 if (slot < nritems) {
2467 printk("shrinker found key %Lu %u %Lu\n",
2468 found_key.objectid, found_key.type,
2472 if (path->slots[0] >= nritems) {
2473 ret = btrfs_next_leaf(root, path);
2483 btrfs_item_key_to_cpu(leaf, &found_key, slot);
2484 ret = get_state_private(&info->block_group_cache,
2485 found_key.objectid, &ptr);
2487 kfree((void *)(unsigned long)ptr);
2489 clear_extent_bits(&info->block_group_cache, found_key.objectid,
2490 found_key.objectid + found_key.offset - 1,
2491 (unsigned int)-1, GFP_NOFS);
2493 key.objectid = found_key.objectid + 1;
2494 btrfs_del_item(trans, root, path);
2495 btrfs_release_path(root, path);
2497 clear_extent_dirty(&info->free_space_cache, new_size, (u64)-1,
2499 btrfs_commit_transaction(trans, root);
2501 btrfs_free_path(path);
2505 int btrfs_grow_extent_tree(struct btrfs_trans_handle *trans,
2506 struct btrfs_root *root, u64 new_size)
2508 struct btrfs_path *path;
2513 struct btrfs_block_group_cache *cache;
2514 struct btrfs_block_group_item *item;
2515 struct btrfs_fs_info *info = root->fs_info;
2516 struct extent_map_tree *block_group_cache;
2517 struct btrfs_key key;
2518 struct extent_buffer *leaf;
2522 old_size = btrfs_super_total_bytes(&info->super_copy);
2523 block_group_cache = &info->block_group_cache;
2525 root = info->extent_root;
2527 cache = btrfs_lookup_block_group(root->fs_info, old_size - 1);
2529 cur_byte = cache->key.objectid + cache->key.offset;
2530 if (cur_byte >= new_size)
2533 key.offset = BTRFS_BLOCK_GROUP_SIZE;
2534 btrfs_set_key_type(&key, BTRFS_BLOCK_GROUP_ITEM_KEY);
2536 path = btrfs_alloc_path();
2540 while(cur_byte < new_size) {
2541 key.objectid = cur_byte;
2542 ret = btrfs_insert_empty_item(trans, root, path, &key,
2543 sizeof(struct btrfs_block_group_item));
2545 leaf = path->nodes[0];
2546 item = btrfs_item_ptr(leaf, path->slots[0],
2547 struct btrfs_block_group_item);
2549 btrfs_set_disk_block_group_used(leaf, item, 0);
2550 div_long_long_rem(nr, 3, &rem);
2552 btrfs_set_disk_block_group_flags(leaf, item,
2553 BTRFS_BLOCK_GROUP_DATA);
2555 btrfs_set_disk_block_group_flags(leaf, item, 0);
2559 cache = kmalloc(sizeof(*cache), GFP_NOFS);
2562 read_extent_buffer(leaf, &cache->item, (unsigned long)item,
2563 sizeof(cache->item));
2565 memcpy(&cache->key, &key, sizeof(key));
2568 cur_byte = key.objectid + key.offset;
2569 btrfs_release_path(root, path);
2571 if (cache->item.flags & BTRFS_BLOCK_GROUP_DATA) {
2572 bit = BLOCK_GROUP_DATA;
2573 cache->data = BTRFS_BLOCK_GROUP_DATA;
2575 bit = BLOCK_GROUP_METADATA;
2579 /* use EXTENT_LOCKED to prevent merging */
2580 set_extent_bits(block_group_cache, key.objectid,
2581 key.objectid + key.offset - 1,
2582 bit | EXTENT_LOCKED, GFP_NOFS);
2583 set_state_private(block_group_cache, key.objectid,
2584 (unsigned long)cache);
2586 btrfs_free_path(path);
2588 btrfs_set_super_total_bytes(&info->super_copy, new_size);
2592 int btrfs_read_block_groups(struct btrfs_root *root)
2594 struct btrfs_path *path;
2598 struct btrfs_block_group_cache *cache;
2599 struct btrfs_fs_info *info = root->fs_info;
2600 struct extent_map_tree *block_group_cache;
2601 struct btrfs_key key;
2602 struct btrfs_key found_key;
2603 struct extent_buffer *leaf;
2605 block_group_cache = &info->block_group_cache;
2607 root = info->extent_root;
2609 key.offset = BTRFS_BLOCK_GROUP_SIZE;
2610 btrfs_set_key_type(&key, BTRFS_BLOCK_GROUP_ITEM_KEY);
2612 path = btrfs_alloc_path();
2617 ret = btrfs_search_slot(NULL, info->extent_root,
2623 leaf = path->nodes[0];
2624 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
2625 cache = kmalloc(sizeof(*cache), GFP_NOFS);
2631 read_extent_buffer(leaf, &cache->item,
2632 btrfs_item_ptr_offset(leaf, path->slots[0]),
2633 sizeof(cache->item));
2634 memcpy(&cache->key, &found_key, sizeof(found_key));
2637 key.objectid = found_key.objectid + found_key.offset;
2638 btrfs_release_path(root, path);
2640 if (cache->item.flags & BTRFS_BLOCK_GROUP_MIXED) {
2641 bit = BLOCK_GROUP_DATA | BLOCK_GROUP_METADATA;
2642 cache->data = BTRFS_BLOCK_GROUP_MIXED;
2643 } else if (cache->item.flags & BTRFS_BLOCK_GROUP_DATA) {
2644 bit = BLOCK_GROUP_DATA;
2645 cache->data = BTRFS_BLOCK_GROUP_DATA;
2647 bit = BLOCK_GROUP_METADATA;
2651 /* use EXTENT_LOCKED to prevent merging */
2652 set_extent_bits(block_group_cache, found_key.objectid,
2653 found_key.objectid + found_key.offset - 1,
2654 bit | EXTENT_LOCKED, GFP_NOFS);
2655 set_state_private(block_group_cache, found_key.objectid,
2656 (unsigned long)cache);
2659 btrfs_super_total_bytes(&info->super_copy))
2663 btrfs_free_path(path);