blob: f997ec0c1ba4b8efb6b46f88160de43cda3da4f2 [file] [log] [blame]
Chris Masone02119d2008-09-05 16:13:11 -04001/*
2 * Copyright (C) 2008 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
19#include <linux/sched.h>
Tejun Heo5a0e3ad2010-03-24 17:04:11 +090020#include <linux/slab.h>
Chris Masone02119d2008-09-05 16:13:11 -040021#include "ctree.h"
22#include "transaction.h"
23#include "disk-io.h"
24#include "locking.h"
25#include "print-tree.h"
26#include "compat.h"
Christoph Hellwigb2950862008-12-02 09:54:17 -050027#include "tree-log.h"
Chris Masone02119d2008-09-05 16:13:11 -040028
29/* magic values for the inode_only field in btrfs_log_inode:
30 *
31 * LOG_INODE_ALL means to log everything
32 * LOG_INODE_EXISTS means to log just enough to recreate the inode
33 * during log replay
34 */
35#define LOG_INODE_ALL 0
36#define LOG_INODE_EXISTS 1
37
38/*
Chris Mason12fcfd22009-03-24 10:24:20 -040039 * directory trouble cases
40 *
41 * 1) on rename or unlink, if the inode being unlinked isn't in the fsync
42 * log, we must force a full commit before doing an fsync of the directory
43 * where the unlink was done.
44 * ---> record transid of last unlink/rename per directory
45 *
46 * mkdir foo/some_dir
47 * normal commit
48 * rename foo/some_dir foo2/some_dir
49 * mkdir foo/some_dir
50 * fsync foo/some_dir/some_file
51 *
52 * The fsync above will unlink the original some_dir without recording
53 * it in its new location (foo2). After a crash, some_dir will be gone
54 * unless the fsync of some_file forces a full commit
55 *
56 * 2) we must log any new names for any file or dir that is in the fsync
57 * log. ---> check inode while renaming/linking.
58 *
59 * 2a) we must log any new names for any file or dir during rename
60 * when the directory they are being removed from was logged.
61 * ---> check inode and old parent dir during rename
62 *
63 * 2a is actually the more important variant. With the extra logging
64 * a crash might unlink the old name without recreating the new one
65 *
66 * 3) after a crash, we must go through any directories with a link count
67 * of zero and redo the rm -rf
68 *
69 * mkdir f1/foo
70 * normal commit
71 * rm -rf f1/foo
72 * fsync(f1)
73 *
74 * The directory f1 was fully removed from the FS, but fsync was never
75 * called on f1, only its parent dir. After a crash the rm -rf must
76 * be replayed. This must be able to recurse down the entire
77 * directory tree. The inode link count fixup code takes care of the
78 * ugly details.
79 */
80
81/*
Chris Masone02119d2008-09-05 16:13:11 -040082 * stages for the tree walking. The first
83 * stage (0) is to only pin down the blocks we find
84 * the second stage (1) is to make sure that all the inodes
85 * we find in the log are created in the subvolume.
86 *
87 * The last stage is to deal with directories and links and extents
88 * and all the other fun semantics
89 */
90#define LOG_WALK_PIN_ONLY 0
91#define LOG_WALK_REPLAY_INODES 1
92#define LOG_WALK_REPLAY_ALL 2
93
Chris Mason12fcfd22009-03-24 10:24:20 -040094static int btrfs_log_inode(struct btrfs_trans_handle *trans,
Chris Masone02119d2008-09-05 16:13:11 -040095 struct btrfs_root *root, struct inode *inode,
96 int inode_only);
Yan Zhengec051c02009-01-05 15:43:42 -050097static int link_to_fixup_dir(struct btrfs_trans_handle *trans,
98 struct btrfs_root *root,
99 struct btrfs_path *path, u64 objectid);
Chris Mason12fcfd22009-03-24 10:24:20 -0400100static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
101 struct btrfs_root *root,
102 struct btrfs_root *log,
103 struct btrfs_path *path,
104 u64 dirid, int del_all);
Chris Masone02119d2008-09-05 16:13:11 -0400105
106/*
107 * tree logging is a special write ahead log used to make sure that
108 * fsyncs and O_SYNCs can happen without doing full tree commits.
109 *
110 * Full tree commits are expensive because they require commonly
111 * modified blocks to be recowed, creating many dirty pages in the
112 * extent tree an 4x-6x higher write load than ext3.
113 *
114 * Instead of doing a tree commit on every fsync, we use the
115 * key ranges and transaction ids to find items for a given file or directory
116 * that have changed in this transaction. Those items are copied into
117 * a special tree (one per subvolume root), that tree is written to disk
118 * and then the fsync is considered complete.
119 *
120 * After a crash, items are copied out of the log-tree back into the
121 * subvolume tree. Any file data extents found are recorded in the extent
122 * allocation tree, and the log-tree freed.
123 *
124 * The log tree is read three times, once to pin down all the extents it is
125 * using in ram and once, once to create all the inodes logged in the tree
126 * and once to do all the other items.
127 */
128
129/*
Chris Masone02119d2008-09-05 16:13:11 -0400130 * start a sub transaction and setup the log tree
131 * this increments the log tree writer count to make the people
132 * syncing the tree wait for us to finish
133 */
134static int start_log_trans(struct btrfs_trans_handle *trans,
135 struct btrfs_root *root)
136{
137 int ret;
Yan, Zheng4a500fd2010-05-16 10:49:59 -0400138 int err = 0;
Yan Zheng7237f182009-01-21 12:54:03 -0500139
140 mutex_lock(&root->log_mutex);
141 if (root->log_root) {
Josef Bacikff782e02009-10-08 15:30:04 -0400142 if (!root->log_start_pid) {
143 root->log_start_pid = current->pid;
144 root->log_multiple_pids = false;
145 } else if (root->log_start_pid != current->pid) {
146 root->log_multiple_pids = true;
147 }
148
Yan Zheng7237f182009-01-21 12:54:03 -0500149 root->log_batch++;
150 atomic_inc(&root->log_writers);
151 mutex_unlock(&root->log_mutex);
152 return 0;
153 }
Josef Bacikff782e02009-10-08 15:30:04 -0400154 root->log_multiple_pids = false;
155 root->log_start_pid = current->pid;
Chris Masone02119d2008-09-05 16:13:11 -0400156 mutex_lock(&root->fs_info->tree_log_mutex);
157 if (!root->fs_info->log_root_tree) {
158 ret = btrfs_init_log_root_tree(trans, root->fs_info);
Yan, Zheng4a500fd2010-05-16 10:49:59 -0400159 if (ret)
160 err = ret;
Chris Masone02119d2008-09-05 16:13:11 -0400161 }
Yan, Zheng4a500fd2010-05-16 10:49:59 -0400162 if (err == 0 && !root->log_root) {
Chris Masone02119d2008-09-05 16:13:11 -0400163 ret = btrfs_add_log_tree(trans, root);
Yan, Zheng4a500fd2010-05-16 10:49:59 -0400164 if (ret)
165 err = ret;
Chris Masone02119d2008-09-05 16:13:11 -0400166 }
Chris Masone02119d2008-09-05 16:13:11 -0400167 mutex_unlock(&root->fs_info->tree_log_mutex);
Yan Zheng7237f182009-01-21 12:54:03 -0500168 root->log_batch++;
169 atomic_inc(&root->log_writers);
170 mutex_unlock(&root->log_mutex);
Yan, Zheng4a500fd2010-05-16 10:49:59 -0400171 return err;
Chris Masone02119d2008-09-05 16:13:11 -0400172}
173
174/*
175 * returns 0 if there was a log transaction running and we were able
176 * to join, or returns -ENOENT if there were not transactions
177 * in progress
178 */
179static int join_running_log_trans(struct btrfs_root *root)
180{
181 int ret = -ENOENT;
182
183 smp_mb();
184 if (!root->log_root)
185 return -ENOENT;
186
Yan Zheng7237f182009-01-21 12:54:03 -0500187 mutex_lock(&root->log_mutex);
Chris Masone02119d2008-09-05 16:13:11 -0400188 if (root->log_root) {
189 ret = 0;
Yan Zheng7237f182009-01-21 12:54:03 -0500190 atomic_inc(&root->log_writers);
Chris Masone02119d2008-09-05 16:13:11 -0400191 }
Yan Zheng7237f182009-01-21 12:54:03 -0500192 mutex_unlock(&root->log_mutex);
Chris Masone02119d2008-09-05 16:13:11 -0400193 return ret;
194}
195
196/*
Chris Mason12fcfd22009-03-24 10:24:20 -0400197 * This either makes the current running log transaction wait
198 * until you call btrfs_end_log_trans() or it makes any future
199 * log transactions wait until you call btrfs_end_log_trans()
200 */
201int btrfs_pin_log_trans(struct btrfs_root *root)
202{
203 int ret = -ENOENT;
204
205 mutex_lock(&root->log_mutex);
206 atomic_inc(&root->log_writers);
207 mutex_unlock(&root->log_mutex);
208 return ret;
209}
210
211/*
Chris Masone02119d2008-09-05 16:13:11 -0400212 * indicate we're done making changes to the log tree
213 * and wake up anyone waiting to do a sync
214 */
Chris Mason12fcfd22009-03-24 10:24:20 -0400215int btrfs_end_log_trans(struct btrfs_root *root)
Chris Masone02119d2008-09-05 16:13:11 -0400216{
Yan Zheng7237f182009-01-21 12:54:03 -0500217 if (atomic_dec_and_test(&root->log_writers)) {
218 smp_mb();
219 if (waitqueue_active(&root->log_writer_wait))
220 wake_up(&root->log_writer_wait);
221 }
Chris Masone02119d2008-09-05 16:13:11 -0400222 return 0;
223}
224
225
226/*
227 * the walk control struct is used to pass state down the chain when
228 * processing the log tree. The stage field tells us which part
229 * of the log tree processing we are currently doing. The others
230 * are state fields used for that specific part
231 */
232struct walk_control {
233 /* should we free the extent on disk when done? This is used
234 * at transaction commit time while freeing a log tree
235 */
236 int free;
237
238 /* should we write out the extent buffer? This is used
239 * while flushing the log tree to disk during a sync
240 */
241 int write;
242
243 /* should we wait for the extent buffer io to finish? Also used
244 * while flushing the log tree to disk for a sync
245 */
246 int wait;
247
248 /* pin only walk, we record which extents on disk belong to the
249 * log trees
250 */
251 int pin;
252
253 /* what stage of the replay code we're currently in */
254 int stage;
255
256 /* the root we are currently replaying */
257 struct btrfs_root *replay_dest;
258
259 /* the trans handle for the current replay */
260 struct btrfs_trans_handle *trans;
261
262 /* the function that gets used to process blocks we find in the
263 * tree. Note the extent_buffer might not be up to date when it is
264 * passed in, and it must be checked or read if you need the data
265 * inside it
266 */
267 int (*process_func)(struct btrfs_root *log, struct extent_buffer *eb,
268 struct walk_control *wc, u64 gen);
269};
270
271/*
272 * process_func used to pin down extents, write them or wait on them
273 */
274static int process_one_buffer(struct btrfs_root *log,
275 struct extent_buffer *eb,
276 struct walk_control *wc, u64 gen)
277{
Josef Bacik04018de2009-04-03 10:14:18 -0400278 if (wc->pin)
Yan Zheng11833d62009-09-11 16:11:19 -0400279 btrfs_pin_extent(log->fs_info->extent_root,
280 eb->start, eb->len, 0);
Chris Masone02119d2008-09-05 16:13:11 -0400281
282 if (btrfs_buffer_uptodate(eb, gen)) {
283 if (wc->write)
284 btrfs_write_tree_block(eb);
285 if (wc->wait)
286 btrfs_wait_tree_block_writeback(eb);
287 }
288 return 0;
289}
290
291/*
292 * Item overwrite used by replay and tree logging. eb, slot and key all refer
293 * to the src data we are copying out.
294 *
295 * root is the tree we are copying into, and path is a scratch
296 * path for use in this function (it should be released on entry and
297 * will be released on exit).
298 *
299 * If the key is already in the destination tree the existing item is
300 * overwritten. If the existing item isn't big enough, it is extended.
301 * If it is too large, it is truncated.
302 *
303 * If the key isn't in the destination yet, a new item is inserted.
304 */
305static noinline int overwrite_item(struct btrfs_trans_handle *trans,
306 struct btrfs_root *root,
307 struct btrfs_path *path,
308 struct extent_buffer *eb, int slot,
309 struct btrfs_key *key)
310{
311 int ret;
312 u32 item_size;
313 u64 saved_i_size = 0;
314 int save_old_i_size = 0;
315 unsigned long src_ptr;
316 unsigned long dst_ptr;
317 int overwrite_root = 0;
318
319 if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
320 overwrite_root = 1;
321
322 item_size = btrfs_item_size_nr(eb, slot);
323 src_ptr = btrfs_item_ptr_offset(eb, slot);
324
325 /* look for the key in the destination tree */
326 ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
327 if (ret == 0) {
328 char *src_copy;
329 char *dst_copy;
330 u32 dst_size = btrfs_item_size_nr(path->nodes[0],
331 path->slots[0]);
332 if (dst_size != item_size)
333 goto insert;
334
335 if (item_size == 0) {
336 btrfs_release_path(root, path);
337 return 0;
338 }
339 dst_copy = kmalloc(item_size, GFP_NOFS);
340 src_copy = kmalloc(item_size, GFP_NOFS);
liubo2a29edc2011-01-26 06:22:08 +0000341 if (!dst_copy || !src_copy) {
342 btrfs_release_path(root, path);
343 kfree(dst_copy);
344 kfree(src_copy);
345 return -ENOMEM;
346 }
Chris Masone02119d2008-09-05 16:13:11 -0400347
348 read_extent_buffer(eb, src_copy, src_ptr, item_size);
349
350 dst_ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
351 read_extent_buffer(path->nodes[0], dst_copy, dst_ptr,
352 item_size);
353 ret = memcmp(dst_copy, src_copy, item_size);
354
355 kfree(dst_copy);
356 kfree(src_copy);
357 /*
358 * they have the same contents, just return, this saves
359 * us from cowing blocks in the destination tree and doing
360 * extra writes that may not have been done by a previous
361 * sync
362 */
363 if (ret == 0) {
364 btrfs_release_path(root, path);
365 return 0;
366 }
367
368 }
369insert:
370 btrfs_release_path(root, path);
371 /* try to insert the key into the destination tree */
372 ret = btrfs_insert_empty_item(trans, root, path,
373 key, item_size);
374
375 /* make sure any existing item is the correct size */
376 if (ret == -EEXIST) {
377 u32 found_size;
378 found_size = btrfs_item_size_nr(path->nodes[0],
379 path->slots[0]);
380 if (found_size > item_size) {
381 btrfs_truncate_item(trans, root, path, item_size, 1);
382 } else if (found_size < item_size) {
Yan Zheng87b29b22008-12-17 10:21:48 -0500383 ret = btrfs_extend_item(trans, root, path,
384 item_size - found_size);
Chris Masone02119d2008-09-05 16:13:11 -0400385 BUG_ON(ret);
386 }
387 } else if (ret) {
Yan, Zheng4a500fd2010-05-16 10:49:59 -0400388 return ret;
Chris Masone02119d2008-09-05 16:13:11 -0400389 }
390 dst_ptr = btrfs_item_ptr_offset(path->nodes[0],
391 path->slots[0]);
392
393 /* don't overwrite an existing inode if the generation number
394 * was logged as zero. This is done when the tree logging code
395 * is just logging an inode to make sure it exists after recovery.
396 *
397 * Also, don't overwrite i_size on directories during replay.
398 * log replay inserts and removes directory items based on the
399 * state of the tree found in the subvolume, and i_size is modified
400 * as it goes
401 */
402 if (key->type == BTRFS_INODE_ITEM_KEY && ret == -EEXIST) {
403 struct btrfs_inode_item *src_item;
404 struct btrfs_inode_item *dst_item;
405
406 src_item = (struct btrfs_inode_item *)src_ptr;
407 dst_item = (struct btrfs_inode_item *)dst_ptr;
408
409 if (btrfs_inode_generation(eb, src_item) == 0)
410 goto no_copy;
411
412 if (overwrite_root &&
413 S_ISDIR(btrfs_inode_mode(eb, src_item)) &&
414 S_ISDIR(btrfs_inode_mode(path->nodes[0], dst_item))) {
415 save_old_i_size = 1;
416 saved_i_size = btrfs_inode_size(path->nodes[0],
417 dst_item);
418 }
419 }
420
421 copy_extent_buffer(path->nodes[0], eb, dst_ptr,
422 src_ptr, item_size);
423
424 if (save_old_i_size) {
425 struct btrfs_inode_item *dst_item;
426 dst_item = (struct btrfs_inode_item *)dst_ptr;
427 btrfs_set_inode_size(path->nodes[0], dst_item, saved_i_size);
428 }
429
430 /* make sure the generation is filled in */
431 if (key->type == BTRFS_INODE_ITEM_KEY) {
432 struct btrfs_inode_item *dst_item;
433 dst_item = (struct btrfs_inode_item *)dst_ptr;
434 if (btrfs_inode_generation(path->nodes[0], dst_item) == 0) {
435 btrfs_set_inode_generation(path->nodes[0], dst_item,
436 trans->transid);
437 }
438 }
439no_copy:
440 btrfs_mark_buffer_dirty(path->nodes[0]);
441 btrfs_release_path(root, path);
442 return 0;
443}
444
445/*
446 * simple helper to read an inode off the disk from a given root
447 * This can only be called for subvolume roots and not for the log
448 */
449static noinline struct inode *read_one_inode(struct btrfs_root *root,
450 u64 objectid)
451{
Yan Zheng5d4f98a2009-06-10 10:45:14 -0400452 struct btrfs_key key;
Chris Masone02119d2008-09-05 16:13:11 -0400453 struct inode *inode;
Chris Masone02119d2008-09-05 16:13:11 -0400454
Yan Zheng5d4f98a2009-06-10 10:45:14 -0400455 key.objectid = objectid;
456 key.type = BTRFS_INODE_ITEM_KEY;
457 key.offset = 0;
Josef Bacik73f73412009-12-04 17:38:27 +0000458 inode = btrfs_iget(root->fs_info->sb, &key, root, NULL);
Yan Zheng5d4f98a2009-06-10 10:45:14 -0400459 if (IS_ERR(inode)) {
460 inode = NULL;
461 } else if (is_bad_inode(inode)) {
Chris Masone02119d2008-09-05 16:13:11 -0400462 iput(inode);
463 inode = NULL;
464 }
465 return inode;
466}
467
468/* replays a single extent in 'eb' at 'slot' with 'key' into the
469 * subvolume 'root'. path is released on entry and should be released
470 * on exit.
471 *
472 * extents in the log tree have not been allocated out of the extent
473 * tree yet. So, this completes the allocation, taking a reference
474 * as required if the extent already exists or creating a new extent
475 * if it isn't in the extent allocation tree yet.
476 *
477 * The extent is inserted into the file, dropping any existing extents
478 * from the file that overlap the new one.
479 */
480static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
481 struct btrfs_root *root,
482 struct btrfs_path *path,
483 struct extent_buffer *eb, int slot,
484 struct btrfs_key *key)
485{
486 int found_type;
487 u64 mask = root->sectorsize - 1;
488 u64 extent_end;
489 u64 alloc_hint;
490 u64 start = key->offset;
Yan Zheng07d400a2009-01-06 11:42:00 -0500491 u64 saved_nbytes;
Chris Masone02119d2008-09-05 16:13:11 -0400492 struct btrfs_file_extent_item *item;
493 struct inode *inode = NULL;
494 unsigned long size;
495 int ret = 0;
496
497 item = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
498 found_type = btrfs_file_extent_type(eb, item);
499
Yan Zhengd899e052008-10-30 14:25:28 -0400500 if (found_type == BTRFS_FILE_EXTENT_REG ||
501 found_type == BTRFS_FILE_EXTENT_PREALLOC)
Chris Masone02119d2008-09-05 16:13:11 -0400502 extent_end = start + btrfs_file_extent_num_bytes(eb, item);
503 else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
Chris Masonc8b97812008-10-29 14:49:59 -0400504 size = btrfs_file_extent_inline_len(eb, item);
Chris Masone02119d2008-09-05 16:13:11 -0400505 extent_end = (start + size + mask) & ~mask;
506 } else {
507 ret = 0;
508 goto out;
509 }
510
511 inode = read_one_inode(root, key->objectid);
512 if (!inode) {
513 ret = -EIO;
514 goto out;
515 }
516
517 /*
518 * first check to see if we already have this extent in the
519 * file. This must be done before the btrfs_drop_extents run
520 * so we don't try to drop this extent.
521 */
522 ret = btrfs_lookup_file_extent(trans, root, path, inode->i_ino,
523 start, 0);
524
Yan Zhengd899e052008-10-30 14:25:28 -0400525 if (ret == 0 &&
526 (found_type == BTRFS_FILE_EXTENT_REG ||
527 found_type == BTRFS_FILE_EXTENT_PREALLOC)) {
Chris Masone02119d2008-09-05 16:13:11 -0400528 struct btrfs_file_extent_item cmp1;
529 struct btrfs_file_extent_item cmp2;
530 struct btrfs_file_extent_item *existing;
531 struct extent_buffer *leaf;
532
533 leaf = path->nodes[0];
534 existing = btrfs_item_ptr(leaf, path->slots[0],
535 struct btrfs_file_extent_item);
536
537 read_extent_buffer(eb, &cmp1, (unsigned long)item,
538 sizeof(cmp1));
539 read_extent_buffer(leaf, &cmp2, (unsigned long)existing,
540 sizeof(cmp2));
541
542 /*
543 * we already have a pointer to this exact extent,
544 * we don't have to do anything
545 */
546 if (memcmp(&cmp1, &cmp2, sizeof(cmp1)) == 0) {
547 btrfs_release_path(root, path);
548 goto out;
549 }
550 }
551 btrfs_release_path(root, path);
552
Yan Zheng07d400a2009-01-06 11:42:00 -0500553 saved_nbytes = inode_get_bytes(inode);
Chris Masone02119d2008-09-05 16:13:11 -0400554 /* drop any overlapping extents */
Yan, Zheng920bbbf2009-11-12 09:34:08 +0000555 ret = btrfs_drop_extents(trans, inode, start, extent_end,
556 &alloc_hint, 1);
Chris Masone02119d2008-09-05 16:13:11 -0400557 BUG_ON(ret);
558
Yan Zheng07d400a2009-01-06 11:42:00 -0500559 if (found_type == BTRFS_FILE_EXTENT_REG ||
560 found_type == BTRFS_FILE_EXTENT_PREALLOC) {
Yan Zheng5d4f98a2009-06-10 10:45:14 -0400561 u64 offset;
Yan Zheng07d400a2009-01-06 11:42:00 -0500562 unsigned long dest_offset;
563 struct btrfs_key ins;
Chris Masone02119d2008-09-05 16:13:11 -0400564
Yan Zheng07d400a2009-01-06 11:42:00 -0500565 ret = btrfs_insert_empty_item(trans, root, path, key,
566 sizeof(*item));
567 BUG_ON(ret);
568 dest_offset = btrfs_item_ptr_offset(path->nodes[0],
569 path->slots[0]);
570 copy_extent_buffer(path->nodes[0], eb, dest_offset,
571 (unsigned long)item, sizeof(*item));
572
573 ins.objectid = btrfs_file_extent_disk_bytenr(eb, item);
574 ins.offset = btrfs_file_extent_disk_num_bytes(eb, item);
575 ins.type = BTRFS_EXTENT_ITEM_KEY;
Yan Zheng5d4f98a2009-06-10 10:45:14 -0400576 offset = key->offset - btrfs_file_extent_offset(eb, item);
Yan Zheng07d400a2009-01-06 11:42:00 -0500577
578 if (ins.objectid > 0) {
579 u64 csum_start;
580 u64 csum_end;
581 LIST_HEAD(ordered_sums);
582 /*
583 * is this extent already allocated in the extent
584 * allocation tree? If so, just add a reference
585 */
586 ret = btrfs_lookup_extent(root, ins.objectid,
587 ins.offset);
588 if (ret == 0) {
589 ret = btrfs_inc_extent_ref(trans, root,
590 ins.objectid, ins.offset,
Yan Zheng5d4f98a2009-06-10 10:45:14 -0400591 0, root->root_key.objectid,
592 key->objectid, offset);
Yan Zheng07d400a2009-01-06 11:42:00 -0500593 } else {
594 /*
595 * insert the extent pointer in the extent
596 * allocation tree
597 */
Yan Zheng5d4f98a2009-06-10 10:45:14 -0400598 ret = btrfs_alloc_logged_file_extent(trans,
599 root, root->root_key.objectid,
600 key->objectid, offset, &ins);
Yan Zheng07d400a2009-01-06 11:42:00 -0500601 BUG_ON(ret);
602 }
603 btrfs_release_path(root, path);
604
605 if (btrfs_file_extent_compression(eb, item)) {
606 csum_start = ins.objectid;
607 csum_end = csum_start + ins.offset;
608 } else {
609 csum_start = ins.objectid +
610 btrfs_file_extent_offset(eb, item);
611 csum_end = csum_start +
612 btrfs_file_extent_num_bytes(eb, item);
613 }
614
615 ret = btrfs_lookup_csums_range(root->log_root,
616 csum_start, csum_end - 1,
617 &ordered_sums);
618 BUG_ON(ret);
619 while (!list_empty(&ordered_sums)) {
620 struct btrfs_ordered_sum *sums;
621 sums = list_entry(ordered_sums.next,
622 struct btrfs_ordered_sum,
623 list);
624 ret = btrfs_csum_file_blocks(trans,
625 root->fs_info->csum_root,
626 sums);
627 BUG_ON(ret);
628 list_del(&sums->list);
629 kfree(sums);
630 }
631 } else {
632 btrfs_release_path(root, path);
633 }
634 } else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
635 /* inline extents are easy, we just overwrite them */
636 ret = overwrite_item(trans, root, path, eb, slot, key);
637 BUG_ON(ret);
638 }
639
640 inode_set_bytes(inode, saved_nbytes);
Chris Masone02119d2008-09-05 16:13:11 -0400641 btrfs_update_inode(trans, root, inode);
642out:
643 if (inode)
644 iput(inode);
645 return ret;
646}
647
648/*
649 * when cleaning up conflicts between the directory names in the
650 * subvolume, directory names in the log and directory names in the
651 * inode back references, we may have to unlink inodes from directories.
652 *
653 * This is a helper function to do the unlink of a specific directory
654 * item
655 */
656static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans,
657 struct btrfs_root *root,
658 struct btrfs_path *path,
659 struct inode *dir,
660 struct btrfs_dir_item *di)
661{
662 struct inode *inode;
663 char *name;
664 int name_len;
665 struct extent_buffer *leaf;
666 struct btrfs_key location;
667 int ret;
668
669 leaf = path->nodes[0];
670
671 btrfs_dir_item_key_to_cpu(leaf, di, &location);
672 name_len = btrfs_dir_name_len(leaf, di);
673 name = kmalloc(name_len, GFP_NOFS);
liubo2a29edc2011-01-26 06:22:08 +0000674 if (!name)
675 return -ENOMEM;
676
Chris Masone02119d2008-09-05 16:13:11 -0400677 read_extent_buffer(leaf, name, (unsigned long)(di + 1), name_len);
678 btrfs_release_path(root, path);
679
680 inode = read_one_inode(root, location.objectid);
681 BUG_ON(!inode);
682
Yan Zhengec051c02009-01-05 15:43:42 -0500683 ret = link_to_fixup_dir(trans, root, path, location.objectid);
684 BUG_ON(ret);
Chris Mason12fcfd22009-03-24 10:24:20 -0400685
Chris Masone02119d2008-09-05 16:13:11 -0400686 ret = btrfs_unlink_inode(trans, root, dir, inode, name, name_len);
Yan Zhengec051c02009-01-05 15:43:42 -0500687 BUG_ON(ret);
Chris Masone02119d2008-09-05 16:13:11 -0400688 kfree(name);
689
690 iput(inode);
691 return ret;
692}
693
694/*
695 * helper function to see if a given name and sequence number found
696 * in an inode back reference are already in a directory and correctly
697 * point to this inode
698 */
699static noinline int inode_in_dir(struct btrfs_root *root,
700 struct btrfs_path *path,
701 u64 dirid, u64 objectid, u64 index,
702 const char *name, int name_len)
703{
704 struct btrfs_dir_item *di;
705 struct btrfs_key location;
706 int match = 0;
707
708 di = btrfs_lookup_dir_index_item(NULL, root, path, dirid,
709 index, name, name_len, 0);
710 if (di && !IS_ERR(di)) {
711 btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location);
712 if (location.objectid != objectid)
713 goto out;
714 } else
715 goto out;
716 btrfs_release_path(root, path);
717
718 di = btrfs_lookup_dir_item(NULL, root, path, dirid, name, name_len, 0);
719 if (di && !IS_ERR(di)) {
720 btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location);
721 if (location.objectid != objectid)
722 goto out;
723 } else
724 goto out;
725 match = 1;
726out:
727 btrfs_release_path(root, path);
728 return match;
729}
730
731/*
732 * helper function to check a log tree for a named back reference in
733 * an inode. This is used to decide if a back reference that is
734 * found in the subvolume conflicts with what we find in the log.
735 *
736 * inode backreferences may have multiple refs in a single item,
737 * during replay we process one reference at a time, and we don't
738 * want to delete valid links to a file from the subvolume if that
739 * link is also in the log.
740 */
741static noinline int backref_in_log(struct btrfs_root *log,
742 struct btrfs_key *key,
743 char *name, int namelen)
744{
745 struct btrfs_path *path;
746 struct btrfs_inode_ref *ref;
747 unsigned long ptr;
748 unsigned long ptr_end;
749 unsigned long name_ptr;
750 int found_name_len;
751 int item_size;
752 int ret;
753 int match = 0;
754
755 path = btrfs_alloc_path();
liubo2a29edc2011-01-26 06:22:08 +0000756 if (!path)
757 return -ENOMEM;
758
Chris Masone02119d2008-09-05 16:13:11 -0400759 ret = btrfs_search_slot(NULL, log, key, path, 0, 0);
760 if (ret != 0)
761 goto out;
762
763 item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]);
764 ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
765 ptr_end = ptr + item_size;
766 while (ptr < ptr_end) {
767 ref = (struct btrfs_inode_ref *)ptr;
768 found_name_len = btrfs_inode_ref_name_len(path->nodes[0], ref);
769 if (found_name_len == namelen) {
770 name_ptr = (unsigned long)(ref + 1);
771 ret = memcmp_extent_buffer(path->nodes[0], name,
772 name_ptr, namelen);
773 if (ret == 0) {
774 match = 1;
775 goto out;
776 }
777 }
778 ptr = (unsigned long)(ref + 1) + found_name_len;
779 }
780out:
781 btrfs_free_path(path);
782 return match;
783}
784
785
786/*
787 * replay one inode back reference item found in the log tree.
788 * eb, slot and key refer to the buffer and key found in the log tree.
789 * root is the destination we are replaying into, and path is for temp
790 * use by this function. (it should be released on return).
791 */
792static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
793 struct btrfs_root *root,
794 struct btrfs_root *log,
795 struct btrfs_path *path,
796 struct extent_buffer *eb, int slot,
797 struct btrfs_key *key)
798{
799 struct inode *dir;
800 int ret;
Chris Masone02119d2008-09-05 16:13:11 -0400801 struct btrfs_inode_ref *ref;
Chris Masone02119d2008-09-05 16:13:11 -0400802 struct inode *inode;
803 char *name;
804 int namelen;
805 unsigned long ref_ptr;
806 unsigned long ref_end;
liuboc622ae62011-03-26 08:01:12 -0400807 int search_done = 0;
Chris Masone02119d2008-09-05 16:13:11 -0400808
Chris Masone02119d2008-09-05 16:13:11 -0400809 /*
810 * it is possible that we didn't log all the parent directories
811 * for a given inode. If we don't find the dir, just don't
812 * copy the back ref in. The link count fixup code will take
813 * care of the rest
814 */
815 dir = read_one_inode(root, key->offset);
816 if (!dir)
817 return -ENOENT;
818
819 inode = read_one_inode(root, key->objectid);
Julia Lawall631c07c2009-07-27 13:57:00 -0400820 BUG_ON(!inode);
Chris Masone02119d2008-09-05 16:13:11 -0400821
822 ref_ptr = btrfs_item_ptr_offset(eb, slot);
823 ref_end = ref_ptr + btrfs_item_size_nr(eb, slot);
824
825again:
826 ref = (struct btrfs_inode_ref *)ref_ptr;
827
828 namelen = btrfs_inode_ref_name_len(eb, ref);
829 name = kmalloc(namelen, GFP_NOFS);
830 BUG_ON(!name);
831
832 read_extent_buffer(eb, name, (unsigned long)(ref + 1), namelen);
833
834 /* if we already have a perfect match, we're done */
835 if (inode_in_dir(root, path, dir->i_ino, inode->i_ino,
836 btrfs_inode_ref_index(eb, ref),
837 name, namelen)) {
838 goto out;
839 }
840
841 /*
842 * look for a conflicting back reference in the metadata.
843 * if we find one we have to unlink that name of the file
844 * before we add our new link. Later on, we overwrite any
845 * existing back reference, and we don't want to create
846 * dangling pointers in the directory.
847 */
liuboc622ae62011-03-26 08:01:12 -0400848
849 if (search_done)
850 goto insert;
851
Chris Masone02119d2008-09-05 16:13:11 -0400852 ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
853 if (ret == 0) {
854 char *victim_name;
855 int victim_name_len;
856 struct btrfs_inode_ref *victim_ref;
857 unsigned long ptr;
858 unsigned long ptr_end;
859 struct extent_buffer *leaf = path->nodes[0];
860
861 /* are we trying to overwrite a back ref for the root directory
862 * if so, just jump out, we're done
863 */
864 if (key->objectid == key->offset)
865 goto out_nowrite;
866
867 /* check all the names in this back reference to see
868 * if they are in the log. if so, we allow them to stay
869 * otherwise they must be unlinked as a conflict
870 */
871 ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
872 ptr_end = ptr + btrfs_item_size_nr(leaf, path->slots[0]);
Chris Masond3977122009-01-05 21:25:51 -0500873 while (ptr < ptr_end) {
Chris Masone02119d2008-09-05 16:13:11 -0400874 victim_ref = (struct btrfs_inode_ref *)ptr;
875 victim_name_len = btrfs_inode_ref_name_len(leaf,
876 victim_ref);
877 victim_name = kmalloc(victim_name_len, GFP_NOFS);
878 BUG_ON(!victim_name);
879
880 read_extent_buffer(leaf, victim_name,
881 (unsigned long)(victim_ref + 1),
882 victim_name_len);
883
884 if (!backref_in_log(log, key, victim_name,
885 victim_name_len)) {
886 btrfs_inc_nlink(inode);
887 btrfs_release_path(root, path);
Chris Mason12fcfd22009-03-24 10:24:20 -0400888
Chris Masone02119d2008-09-05 16:13:11 -0400889 ret = btrfs_unlink_inode(trans, root, dir,
890 inode, victim_name,
891 victim_name_len);
Chris Masone02119d2008-09-05 16:13:11 -0400892 }
893 kfree(victim_name);
894 ptr = (unsigned long)(victim_ref + 1) + victim_name_len;
895 }
896 BUG_ON(ret);
liuboc622ae62011-03-26 08:01:12 -0400897
898 /*
899 * NOTE: we have searched root tree and checked the
900 * coresponding ref, it does not need to check again.
901 */
902 search_done = 1;
Chris Masone02119d2008-09-05 16:13:11 -0400903 }
904 btrfs_release_path(root, path);
905
liuboc622ae62011-03-26 08:01:12 -0400906insert:
Chris Masone02119d2008-09-05 16:13:11 -0400907 /* insert our name */
908 ret = btrfs_add_link(trans, dir, inode, name, namelen, 0,
909 btrfs_inode_ref_index(eb, ref));
910 BUG_ON(ret);
911
912 btrfs_update_inode(trans, root, inode);
913
914out:
915 ref_ptr = (unsigned long)(ref + 1) + namelen;
916 kfree(name);
917 if (ref_ptr < ref_end)
918 goto again;
919
920 /* finally write the back reference in the inode */
921 ret = overwrite_item(trans, root, path, eb, slot, key);
922 BUG_ON(ret);
923
924out_nowrite:
925 btrfs_release_path(root, path);
926 iput(dir);
927 iput(inode);
928 return 0;
929}
930
Yan, Zhengc71bf092009-11-12 09:34:40 +0000931static int insert_orphan_item(struct btrfs_trans_handle *trans,
932 struct btrfs_root *root, u64 offset)
933{
934 int ret;
935 ret = btrfs_find_orphan_item(root, offset);
936 if (ret > 0)
937 ret = btrfs_insert_orphan_item(trans, root, offset);
938 return ret;
939}
940
941
Chris Masone02119d2008-09-05 16:13:11 -0400942/*
Chris Masone02119d2008-09-05 16:13:11 -0400943 * There are a few corners where the link count of the file can't
944 * be properly maintained during replay. So, instead of adding
945 * lots of complexity to the log code, we just scan the backrefs
946 * for any file that has been through replay.
947 *
948 * The scan will update the link count on the inode to reflect the
949 * number of back refs found. If it goes down to zero, the iput
950 * will free the inode.
951 */
952static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans,
953 struct btrfs_root *root,
954 struct inode *inode)
955{
956 struct btrfs_path *path;
957 int ret;
958 struct btrfs_key key;
959 u64 nlink = 0;
960 unsigned long ptr;
961 unsigned long ptr_end;
962 int name_len;
963
964 key.objectid = inode->i_ino;
965 key.type = BTRFS_INODE_REF_KEY;
966 key.offset = (u64)-1;
967
968 path = btrfs_alloc_path();
liubo2a29edc2011-01-26 06:22:08 +0000969 if (!path)
970 return -ENOMEM;
Chris Masone02119d2008-09-05 16:13:11 -0400971
Chris Masond3977122009-01-05 21:25:51 -0500972 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -0400973 ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
974 if (ret < 0)
975 break;
976 if (ret > 0) {
977 if (path->slots[0] == 0)
978 break;
979 path->slots[0]--;
980 }
981 btrfs_item_key_to_cpu(path->nodes[0], &key,
982 path->slots[0]);
983 if (key.objectid != inode->i_ino ||
984 key.type != BTRFS_INODE_REF_KEY)
985 break;
986 ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
987 ptr_end = ptr + btrfs_item_size_nr(path->nodes[0],
988 path->slots[0]);
Chris Masond3977122009-01-05 21:25:51 -0500989 while (ptr < ptr_end) {
Chris Masone02119d2008-09-05 16:13:11 -0400990 struct btrfs_inode_ref *ref;
991
992 ref = (struct btrfs_inode_ref *)ptr;
993 name_len = btrfs_inode_ref_name_len(path->nodes[0],
994 ref);
995 ptr = (unsigned long)(ref + 1) + name_len;
996 nlink++;
997 }
998
999 if (key.offset == 0)
1000 break;
1001 key.offset--;
1002 btrfs_release_path(root, path);
1003 }
Chris Mason12fcfd22009-03-24 10:24:20 -04001004 btrfs_release_path(root, path);
Chris Masone02119d2008-09-05 16:13:11 -04001005 if (nlink != inode->i_nlink) {
1006 inode->i_nlink = nlink;
1007 btrfs_update_inode(trans, root, inode);
1008 }
Chris Mason8d5bf1c2008-09-11 15:51:21 -04001009 BTRFS_I(inode)->index_cnt = (u64)-1;
Chris Masone02119d2008-09-05 16:13:11 -04001010
Yan, Zhengc71bf092009-11-12 09:34:40 +00001011 if (inode->i_nlink == 0) {
1012 if (S_ISDIR(inode->i_mode)) {
1013 ret = replay_dir_deletes(trans, root, NULL, path,
1014 inode->i_ino, 1);
1015 BUG_ON(ret);
1016 }
1017 ret = insert_orphan_item(trans, root, inode->i_ino);
Chris Mason12fcfd22009-03-24 10:24:20 -04001018 BUG_ON(ret);
1019 }
1020 btrfs_free_path(path);
1021
Chris Masone02119d2008-09-05 16:13:11 -04001022 return 0;
1023}
1024
1025static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans,
1026 struct btrfs_root *root,
1027 struct btrfs_path *path)
1028{
1029 int ret;
1030 struct btrfs_key key;
1031 struct inode *inode;
1032
1033 key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID;
1034 key.type = BTRFS_ORPHAN_ITEM_KEY;
1035 key.offset = (u64)-1;
Chris Masond3977122009-01-05 21:25:51 -05001036 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -04001037 ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
1038 if (ret < 0)
1039 break;
1040
1041 if (ret == 1) {
1042 if (path->slots[0] == 0)
1043 break;
1044 path->slots[0]--;
1045 }
1046
1047 btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
1048 if (key.objectid != BTRFS_TREE_LOG_FIXUP_OBJECTID ||
1049 key.type != BTRFS_ORPHAN_ITEM_KEY)
1050 break;
1051
1052 ret = btrfs_del_item(trans, root, path);
1053 BUG_ON(ret);
1054
1055 btrfs_release_path(root, path);
1056 inode = read_one_inode(root, key.offset);
1057 BUG_ON(!inode);
1058
1059 ret = fixup_inode_link_count(trans, root, inode);
1060 BUG_ON(ret);
1061
1062 iput(inode);
1063
Chris Mason12fcfd22009-03-24 10:24:20 -04001064 /*
1065 * fixup on a directory may create new entries,
1066 * make sure we always look for the highset possible
1067 * offset
1068 */
1069 key.offset = (u64)-1;
Chris Masone02119d2008-09-05 16:13:11 -04001070 }
1071 btrfs_release_path(root, path);
1072 return 0;
1073}
1074
1075
1076/*
1077 * record a given inode in the fixup dir so we can check its link
1078 * count when replay is done. The link count is incremented here
1079 * so the inode won't go away until we check it
1080 */
1081static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans,
1082 struct btrfs_root *root,
1083 struct btrfs_path *path,
1084 u64 objectid)
1085{
1086 struct btrfs_key key;
1087 int ret = 0;
1088 struct inode *inode;
1089
1090 inode = read_one_inode(root, objectid);
1091 BUG_ON(!inode);
1092
1093 key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID;
1094 btrfs_set_key_type(&key, BTRFS_ORPHAN_ITEM_KEY);
1095 key.offset = objectid;
1096
1097 ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
1098
1099 btrfs_release_path(root, path);
1100 if (ret == 0) {
1101 btrfs_inc_nlink(inode);
1102 btrfs_update_inode(trans, root, inode);
1103 } else if (ret == -EEXIST) {
1104 ret = 0;
1105 } else {
1106 BUG();
1107 }
1108 iput(inode);
1109
1110 return ret;
1111}
1112
1113/*
1114 * when replaying the log for a directory, we only insert names
1115 * for inodes that actually exist. This means an fsync on a directory
1116 * does not implicitly fsync all the new files in it
1117 */
1118static noinline int insert_one_name(struct btrfs_trans_handle *trans,
1119 struct btrfs_root *root,
1120 struct btrfs_path *path,
1121 u64 dirid, u64 index,
1122 char *name, int name_len, u8 type,
1123 struct btrfs_key *location)
1124{
1125 struct inode *inode;
1126 struct inode *dir;
1127 int ret;
1128
1129 inode = read_one_inode(root, location->objectid);
1130 if (!inode)
1131 return -ENOENT;
1132
1133 dir = read_one_inode(root, dirid);
1134 if (!dir) {
1135 iput(inode);
1136 return -EIO;
1137 }
1138 ret = btrfs_add_link(trans, dir, inode, name, name_len, 1, index);
1139
1140 /* FIXME, put inode into FIXUP list */
1141
1142 iput(inode);
1143 iput(dir);
1144 return ret;
1145}
1146
1147/*
1148 * take a single entry in a log directory item and replay it into
1149 * the subvolume.
1150 *
1151 * if a conflicting item exists in the subdirectory already,
1152 * the inode it points to is unlinked and put into the link count
1153 * fix up tree.
1154 *
1155 * If a name from the log points to a file or directory that does
1156 * not exist in the FS, it is skipped. fsyncs on directories
1157 * do not force down inodes inside that directory, just changes to the
1158 * names or unlinks in a directory.
1159 */
1160static noinline int replay_one_name(struct btrfs_trans_handle *trans,
1161 struct btrfs_root *root,
1162 struct btrfs_path *path,
1163 struct extent_buffer *eb,
1164 struct btrfs_dir_item *di,
1165 struct btrfs_key *key)
1166{
1167 char *name;
1168 int name_len;
1169 struct btrfs_dir_item *dst_di;
1170 struct btrfs_key found_key;
1171 struct btrfs_key log_key;
1172 struct inode *dir;
Chris Masone02119d2008-09-05 16:13:11 -04001173 u8 log_type;
Chris Mason4bef0842008-09-08 11:18:08 -04001174 int exists;
Chris Masone02119d2008-09-05 16:13:11 -04001175 int ret;
1176
1177 dir = read_one_inode(root, key->objectid);
1178 BUG_ON(!dir);
1179
1180 name_len = btrfs_dir_name_len(eb, di);
1181 name = kmalloc(name_len, GFP_NOFS);
liubo2a29edc2011-01-26 06:22:08 +00001182 if (!name)
1183 return -ENOMEM;
1184
Chris Masone02119d2008-09-05 16:13:11 -04001185 log_type = btrfs_dir_type(eb, di);
1186 read_extent_buffer(eb, name, (unsigned long)(di + 1),
1187 name_len);
1188
1189 btrfs_dir_item_key_to_cpu(eb, di, &log_key);
Chris Mason4bef0842008-09-08 11:18:08 -04001190 exists = btrfs_lookup_inode(trans, root, path, &log_key, 0);
1191 if (exists == 0)
1192 exists = 1;
1193 else
1194 exists = 0;
1195 btrfs_release_path(root, path);
1196
Chris Masone02119d2008-09-05 16:13:11 -04001197 if (key->type == BTRFS_DIR_ITEM_KEY) {
1198 dst_di = btrfs_lookup_dir_item(trans, root, path, key->objectid,
1199 name, name_len, 1);
Chris Masond3977122009-01-05 21:25:51 -05001200 } else if (key->type == BTRFS_DIR_INDEX_KEY) {
Chris Masone02119d2008-09-05 16:13:11 -04001201 dst_di = btrfs_lookup_dir_index_item(trans, root, path,
1202 key->objectid,
1203 key->offset, name,
1204 name_len, 1);
1205 } else {
1206 BUG();
1207 }
1208 if (!dst_di || IS_ERR(dst_di)) {
1209 /* we need a sequence number to insert, so we only
1210 * do inserts for the BTRFS_DIR_INDEX_KEY types
1211 */
1212 if (key->type != BTRFS_DIR_INDEX_KEY)
1213 goto out;
1214 goto insert;
1215 }
1216
1217 btrfs_dir_item_key_to_cpu(path->nodes[0], dst_di, &found_key);
1218 /* the existing item matches the logged item */
1219 if (found_key.objectid == log_key.objectid &&
1220 found_key.type == log_key.type &&
1221 found_key.offset == log_key.offset &&
1222 btrfs_dir_type(path->nodes[0], dst_di) == log_type) {
1223 goto out;
1224 }
1225
1226 /*
1227 * don't drop the conflicting directory entry if the inode
1228 * for the new entry doesn't exist
1229 */
Chris Mason4bef0842008-09-08 11:18:08 -04001230 if (!exists)
Chris Masone02119d2008-09-05 16:13:11 -04001231 goto out;
1232
Chris Masone02119d2008-09-05 16:13:11 -04001233 ret = drop_one_dir_item(trans, root, path, dir, dst_di);
1234 BUG_ON(ret);
1235
1236 if (key->type == BTRFS_DIR_INDEX_KEY)
1237 goto insert;
1238out:
1239 btrfs_release_path(root, path);
1240 kfree(name);
1241 iput(dir);
1242 return 0;
1243
1244insert:
1245 btrfs_release_path(root, path);
1246 ret = insert_one_name(trans, root, path, key->objectid, key->offset,
1247 name, name_len, log_type, &log_key);
1248
Stoyan Gaydarovc2934982009-04-02 17:05:11 -04001249 BUG_ON(ret && ret != -ENOENT);
Chris Masone02119d2008-09-05 16:13:11 -04001250 goto out;
1251}
1252
1253/*
1254 * find all the names in a directory item and reconcile them into
1255 * the subvolume. Only BTRFS_DIR_ITEM_KEY types will have more than
1256 * one name in a directory item, but the same code gets used for
1257 * both directory index types
1258 */
1259static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans,
1260 struct btrfs_root *root,
1261 struct btrfs_path *path,
1262 struct extent_buffer *eb, int slot,
1263 struct btrfs_key *key)
1264{
1265 int ret;
1266 u32 item_size = btrfs_item_size_nr(eb, slot);
1267 struct btrfs_dir_item *di;
1268 int name_len;
1269 unsigned long ptr;
1270 unsigned long ptr_end;
1271
1272 ptr = btrfs_item_ptr_offset(eb, slot);
1273 ptr_end = ptr + item_size;
Chris Masond3977122009-01-05 21:25:51 -05001274 while (ptr < ptr_end) {
Chris Masone02119d2008-09-05 16:13:11 -04001275 di = (struct btrfs_dir_item *)ptr;
Josef Bacik22a94d42011-03-16 16:47:17 -04001276 if (verify_dir_item(root, eb, di))
1277 return -EIO;
Chris Masone02119d2008-09-05 16:13:11 -04001278 name_len = btrfs_dir_name_len(eb, di);
1279 ret = replay_one_name(trans, root, path, eb, di, key);
1280 BUG_ON(ret);
1281 ptr = (unsigned long)(di + 1);
1282 ptr += name_len;
1283 }
1284 return 0;
1285}
1286
1287/*
1288 * directory replay has two parts. There are the standard directory
1289 * items in the log copied from the subvolume, and range items
1290 * created in the log while the subvolume was logged.
1291 *
1292 * The range items tell us which parts of the key space the log
1293 * is authoritative for. During replay, if a key in the subvolume
1294 * directory is in a logged range item, but not actually in the log
1295 * that means it was deleted from the directory before the fsync
1296 * and should be removed.
1297 */
1298static noinline int find_dir_range(struct btrfs_root *root,
1299 struct btrfs_path *path,
1300 u64 dirid, int key_type,
1301 u64 *start_ret, u64 *end_ret)
1302{
1303 struct btrfs_key key;
1304 u64 found_end;
1305 struct btrfs_dir_log_item *item;
1306 int ret;
1307 int nritems;
1308
1309 if (*start_ret == (u64)-1)
1310 return 1;
1311
1312 key.objectid = dirid;
1313 key.type = key_type;
1314 key.offset = *start_ret;
1315
1316 ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
1317 if (ret < 0)
1318 goto out;
1319 if (ret > 0) {
1320 if (path->slots[0] == 0)
1321 goto out;
1322 path->slots[0]--;
1323 }
1324 if (ret != 0)
1325 btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
1326
1327 if (key.type != key_type || key.objectid != dirid) {
1328 ret = 1;
1329 goto next;
1330 }
1331 item = btrfs_item_ptr(path->nodes[0], path->slots[0],
1332 struct btrfs_dir_log_item);
1333 found_end = btrfs_dir_log_end(path->nodes[0], item);
1334
1335 if (*start_ret >= key.offset && *start_ret <= found_end) {
1336 ret = 0;
1337 *start_ret = key.offset;
1338 *end_ret = found_end;
1339 goto out;
1340 }
1341 ret = 1;
1342next:
1343 /* check the next slot in the tree to see if it is a valid item */
1344 nritems = btrfs_header_nritems(path->nodes[0]);
1345 if (path->slots[0] >= nritems) {
1346 ret = btrfs_next_leaf(root, path);
1347 if (ret)
1348 goto out;
1349 } else {
1350 path->slots[0]++;
1351 }
1352
1353 btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
1354
1355 if (key.type != key_type || key.objectid != dirid) {
1356 ret = 1;
1357 goto out;
1358 }
1359 item = btrfs_item_ptr(path->nodes[0], path->slots[0],
1360 struct btrfs_dir_log_item);
1361 found_end = btrfs_dir_log_end(path->nodes[0], item);
1362 *start_ret = key.offset;
1363 *end_ret = found_end;
1364 ret = 0;
1365out:
1366 btrfs_release_path(root, path);
1367 return ret;
1368}
1369
1370/*
1371 * this looks for a given directory item in the log. If the directory
1372 * item is not in the log, the item is removed and the inode it points
1373 * to is unlinked
1374 */
1375static noinline int check_item_in_log(struct btrfs_trans_handle *trans,
1376 struct btrfs_root *root,
1377 struct btrfs_root *log,
1378 struct btrfs_path *path,
1379 struct btrfs_path *log_path,
1380 struct inode *dir,
1381 struct btrfs_key *dir_key)
1382{
1383 int ret;
1384 struct extent_buffer *eb;
1385 int slot;
1386 u32 item_size;
1387 struct btrfs_dir_item *di;
1388 struct btrfs_dir_item *log_di;
1389 int name_len;
1390 unsigned long ptr;
1391 unsigned long ptr_end;
1392 char *name;
1393 struct inode *inode;
1394 struct btrfs_key location;
1395
1396again:
1397 eb = path->nodes[0];
1398 slot = path->slots[0];
1399 item_size = btrfs_item_size_nr(eb, slot);
1400 ptr = btrfs_item_ptr_offset(eb, slot);
1401 ptr_end = ptr + item_size;
Chris Masond3977122009-01-05 21:25:51 -05001402 while (ptr < ptr_end) {
Chris Masone02119d2008-09-05 16:13:11 -04001403 di = (struct btrfs_dir_item *)ptr;
Josef Bacik22a94d42011-03-16 16:47:17 -04001404 if (verify_dir_item(root, eb, di)) {
1405 ret = -EIO;
1406 goto out;
1407 }
1408
Chris Masone02119d2008-09-05 16:13:11 -04001409 name_len = btrfs_dir_name_len(eb, di);
1410 name = kmalloc(name_len, GFP_NOFS);
1411 if (!name) {
1412 ret = -ENOMEM;
1413 goto out;
1414 }
1415 read_extent_buffer(eb, name, (unsigned long)(di + 1),
1416 name_len);
1417 log_di = NULL;
Chris Mason12fcfd22009-03-24 10:24:20 -04001418 if (log && dir_key->type == BTRFS_DIR_ITEM_KEY) {
Chris Masone02119d2008-09-05 16:13:11 -04001419 log_di = btrfs_lookup_dir_item(trans, log, log_path,
1420 dir_key->objectid,
1421 name, name_len, 0);
Chris Mason12fcfd22009-03-24 10:24:20 -04001422 } else if (log && dir_key->type == BTRFS_DIR_INDEX_KEY) {
Chris Masone02119d2008-09-05 16:13:11 -04001423 log_di = btrfs_lookup_dir_index_item(trans, log,
1424 log_path,
1425 dir_key->objectid,
1426 dir_key->offset,
1427 name, name_len, 0);
1428 }
1429 if (!log_di || IS_ERR(log_di)) {
1430 btrfs_dir_item_key_to_cpu(eb, di, &location);
1431 btrfs_release_path(root, path);
1432 btrfs_release_path(log, log_path);
1433 inode = read_one_inode(root, location.objectid);
1434 BUG_ON(!inode);
1435
1436 ret = link_to_fixup_dir(trans, root,
1437 path, location.objectid);
1438 BUG_ON(ret);
1439 btrfs_inc_nlink(inode);
1440 ret = btrfs_unlink_inode(trans, root, dir, inode,
1441 name, name_len);
1442 BUG_ON(ret);
1443 kfree(name);
1444 iput(inode);
1445
1446 /* there might still be more names under this key
1447 * check and repeat if required
1448 */
1449 ret = btrfs_search_slot(NULL, root, dir_key, path,
1450 0, 0);
1451 if (ret == 0)
1452 goto again;
1453 ret = 0;
1454 goto out;
1455 }
1456 btrfs_release_path(log, log_path);
1457 kfree(name);
1458
1459 ptr = (unsigned long)(di + 1);
1460 ptr += name_len;
1461 }
1462 ret = 0;
1463out:
1464 btrfs_release_path(root, path);
1465 btrfs_release_path(log, log_path);
1466 return ret;
1467}
1468
1469/*
1470 * deletion replay happens before we copy any new directory items
1471 * out of the log or out of backreferences from inodes. It
1472 * scans the log to find ranges of keys that log is authoritative for,
1473 * and then scans the directory to find items in those ranges that are
1474 * not present in the log.
1475 *
1476 * Anything we don't find in the log is unlinked and removed from the
1477 * directory.
1478 */
1479static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
1480 struct btrfs_root *root,
1481 struct btrfs_root *log,
1482 struct btrfs_path *path,
Chris Mason12fcfd22009-03-24 10:24:20 -04001483 u64 dirid, int del_all)
Chris Masone02119d2008-09-05 16:13:11 -04001484{
1485 u64 range_start;
1486 u64 range_end;
1487 int key_type = BTRFS_DIR_LOG_ITEM_KEY;
1488 int ret = 0;
1489 struct btrfs_key dir_key;
1490 struct btrfs_key found_key;
1491 struct btrfs_path *log_path;
1492 struct inode *dir;
1493
1494 dir_key.objectid = dirid;
1495 dir_key.type = BTRFS_DIR_ITEM_KEY;
1496 log_path = btrfs_alloc_path();
1497 if (!log_path)
1498 return -ENOMEM;
1499
1500 dir = read_one_inode(root, dirid);
1501 /* it isn't an error if the inode isn't there, that can happen
1502 * because we replay the deletes before we copy in the inode item
1503 * from the log
1504 */
1505 if (!dir) {
1506 btrfs_free_path(log_path);
1507 return 0;
1508 }
1509again:
1510 range_start = 0;
1511 range_end = 0;
Chris Masond3977122009-01-05 21:25:51 -05001512 while (1) {
Chris Mason12fcfd22009-03-24 10:24:20 -04001513 if (del_all)
1514 range_end = (u64)-1;
1515 else {
1516 ret = find_dir_range(log, path, dirid, key_type,
1517 &range_start, &range_end);
1518 if (ret != 0)
1519 break;
1520 }
Chris Masone02119d2008-09-05 16:13:11 -04001521
1522 dir_key.offset = range_start;
Chris Masond3977122009-01-05 21:25:51 -05001523 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -04001524 int nritems;
1525 ret = btrfs_search_slot(NULL, root, &dir_key, path,
1526 0, 0);
1527 if (ret < 0)
1528 goto out;
1529
1530 nritems = btrfs_header_nritems(path->nodes[0]);
1531 if (path->slots[0] >= nritems) {
1532 ret = btrfs_next_leaf(root, path);
1533 if (ret)
1534 break;
1535 }
1536 btrfs_item_key_to_cpu(path->nodes[0], &found_key,
1537 path->slots[0]);
1538 if (found_key.objectid != dirid ||
1539 found_key.type != dir_key.type)
1540 goto next_type;
1541
1542 if (found_key.offset > range_end)
1543 break;
1544
1545 ret = check_item_in_log(trans, root, log, path,
Chris Mason12fcfd22009-03-24 10:24:20 -04001546 log_path, dir,
1547 &found_key);
Chris Masone02119d2008-09-05 16:13:11 -04001548 BUG_ON(ret);
1549 if (found_key.offset == (u64)-1)
1550 break;
1551 dir_key.offset = found_key.offset + 1;
1552 }
1553 btrfs_release_path(root, path);
1554 if (range_end == (u64)-1)
1555 break;
1556 range_start = range_end + 1;
1557 }
1558
1559next_type:
1560 ret = 0;
1561 if (key_type == BTRFS_DIR_LOG_ITEM_KEY) {
1562 key_type = BTRFS_DIR_LOG_INDEX_KEY;
1563 dir_key.type = BTRFS_DIR_INDEX_KEY;
1564 btrfs_release_path(root, path);
1565 goto again;
1566 }
1567out:
1568 btrfs_release_path(root, path);
1569 btrfs_free_path(log_path);
1570 iput(dir);
1571 return ret;
1572}
1573
1574/*
1575 * the process_func used to replay items from the log tree. This
1576 * gets called in two different stages. The first stage just looks
1577 * for inodes and makes sure they are all copied into the subvolume.
1578 *
1579 * The second stage copies all the other item types from the log into
1580 * the subvolume. The two stage approach is slower, but gets rid of
1581 * lots of complexity around inodes referencing other inodes that exist
1582 * only in the log (references come from either directory items or inode
1583 * back refs).
1584 */
1585static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
1586 struct walk_control *wc, u64 gen)
1587{
1588 int nritems;
1589 struct btrfs_path *path;
1590 struct btrfs_root *root = wc->replay_dest;
1591 struct btrfs_key key;
Chris Masone02119d2008-09-05 16:13:11 -04001592 int level;
1593 int i;
1594 int ret;
1595
1596 btrfs_read_buffer(eb, gen);
1597
1598 level = btrfs_header_level(eb);
1599
1600 if (level != 0)
1601 return 0;
1602
1603 path = btrfs_alloc_path();
1604 BUG_ON(!path);
1605
1606 nritems = btrfs_header_nritems(eb);
1607 for (i = 0; i < nritems; i++) {
1608 btrfs_item_key_to_cpu(eb, &key, i);
Chris Masone02119d2008-09-05 16:13:11 -04001609
1610 /* inode keys are done during the first stage */
1611 if (key.type == BTRFS_INODE_ITEM_KEY &&
1612 wc->stage == LOG_WALK_REPLAY_INODES) {
Chris Masone02119d2008-09-05 16:13:11 -04001613 struct btrfs_inode_item *inode_item;
1614 u32 mode;
1615
1616 inode_item = btrfs_item_ptr(eb, i,
1617 struct btrfs_inode_item);
1618 mode = btrfs_inode_mode(eb, inode_item);
1619 if (S_ISDIR(mode)) {
1620 ret = replay_dir_deletes(wc->trans,
Chris Mason12fcfd22009-03-24 10:24:20 -04001621 root, log, path, key.objectid, 0);
Chris Masone02119d2008-09-05 16:13:11 -04001622 BUG_ON(ret);
1623 }
1624 ret = overwrite_item(wc->trans, root, path,
1625 eb, i, &key);
1626 BUG_ON(ret);
1627
Yan, Zhengc71bf092009-11-12 09:34:40 +00001628 /* for regular files, make sure corresponding
1629 * orhpan item exist. extents past the new EOF
1630 * will be truncated later by orphan cleanup.
Chris Masone02119d2008-09-05 16:13:11 -04001631 */
1632 if (S_ISREG(mode)) {
Yan, Zhengc71bf092009-11-12 09:34:40 +00001633 ret = insert_orphan_item(wc->trans, root,
1634 key.objectid);
Chris Masone02119d2008-09-05 16:13:11 -04001635 BUG_ON(ret);
Chris Masone02119d2008-09-05 16:13:11 -04001636 }
Yan, Zhengc71bf092009-11-12 09:34:40 +00001637
Chris Masone02119d2008-09-05 16:13:11 -04001638 ret = link_to_fixup_dir(wc->trans, root,
1639 path, key.objectid);
1640 BUG_ON(ret);
1641 }
1642 if (wc->stage < LOG_WALK_REPLAY_ALL)
1643 continue;
1644
1645 /* these keys are simply copied */
1646 if (key.type == BTRFS_XATTR_ITEM_KEY) {
1647 ret = overwrite_item(wc->trans, root, path,
1648 eb, i, &key);
1649 BUG_ON(ret);
1650 } else if (key.type == BTRFS_INODE_REF_KEY) {
1651 ret = add_inode_ref(wc->trans, root, log, path,
1652 eb, i, &key);
1653 BUG_ON(ret && ret != -ENOENT);
1654 } else if (key.type == BTRFS_EXTENT_DATA_KEY) {
1655 ret = replay_one_extent(wc->trans, root, path,
1656 eb, i, &key);
1657 BUG_ON(ret);
Chris Masone02119d2008-09-05 16:13:11 -04001658 } else if (key.type == BTRFS_DIR_ITEM_KEY ||
1659 key.type == BTRFS_DIR_INDEX_KEY) {
1660 ret = replay_one_dir_item(wc->trans, root, path,
1661 eb, i, &key);
1662 BUG_ON(ret);
1663 }
1664 }
1665 btrfs_free_path(path);
1666 return 0;
1667}
1668
Chris Masond3977122009-01-05 21:25:51 -05001669static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
Chris Masone02119d2008-09-05 16:13:11 -04001670 struct btrfs_root *root,
1671 struct btrfs_path *path, int *level,
1672 struct walk_control *wc)
1673{
1674 u64 root_owner;
Chris Masone02119d2008-09-05 16:13:11 -04001675 u64 bytenr;
1676 u64 ptr_gen;
1677 struct extent_buffer *next;
1678 struct extent_buffer *cur;
1679 struct extent_buffer *parent;
1680 u32 blocksize;
1681 int ret = 0;
1682
1683 WARN_ON(*level < 0);
1684 WARN_ON(*level >= BTRFS_MAX_LEVEL);
1685
Chris Masond3977122009-01-05 21:25:51 -05001686 while (*level > 0) {
Chris Masone02119d2008-09-05 16:13:11 -04001687 WARN_ON(*level < 0);
1688 WARN_ON(*level >= BTRFS_MAX_LEVEL);
1689 cur = path->nodes[*level];
1690
1691 if (btrfs_header_level(cur) != *level)
1692 WARN_ON(1);
1693
1694 if (path->slots[*level] >=
1695 btrfs_header_nritems(cur))
1696 break;
1697
1698 bytenr = btrfs_node_blockptr(cur, path->slots[*level]);
1699 ptr_gen = btrfs_node_ptr_generation(cur, path->slots[*level]);
1700 blocksize = btrfs_level_size(root, *level - 1);
1701
1702 parent = path->nodes[*level];
1703 root_owner = btrfs_header_owner(parent);
Chris Masone02119d2008-09-05 16:13:11 -04001704
1705 next = btrfs_find_create_tree_block(root, bytenr, blocksize);
liubo2a29edc2011-01-26 06:22:08 +00001706 if (!next)
1707 return -ENOMEM;
Chris Masone02119d2008-09-05 16:13:11 -04001708
Chris Masone02119d2008-09-05 16:13:11 -04001709 if (*level == 1) {
Yan, Zheng4a500fd2010-05-16 10:49:59 -04001710 wc->process_func(root, next, wc, ptr_gen);
1711
Chris Masone02119d2008-09-05 16:13:11 -04001712 path->slots[*level]++;
1713 if (wc->free) {
1714 btrfs_read_buffer(next, ptr_gen);
1715
1716 btrfs_tree_lock(next);
1717 clean_tree_block(trans, root, next);
Chris Masonb4ce94d2009-02-04 09:25:08 -05001718 btrfs_set_lock_blocking(next);
Chris Masone02119d2008-09-05 16:13:11 -04001719 btrfs_wait_tree_block_writeback(next);
1720 btrfs_tree_unlock(next);
1721
Chris Masone02119d2008-09-05 16:13:11 -04001722 WARN_ON(root_owner !=
1723 BTRFS_TREE_LOG_OBJECTID);
Chris Masond00aff02008-09-11 15:54:42 -04001724 ret = btrfs_free_reserved_extent(root,
1725 bytenr, blocksize);
Chris Masone02119d2008-09-05 16:13:11 -04001726 BUG_ON(ret);
1727 }
1728 free_extent_buffer(next);
1729 continue;
1730 }
1731 btrfs_read_buffer(next, ptr_gen);
1732
1733 WARN_ON(*level <= 0);
1734 if (path->nodes[*level-1])
1735 free_extent_buffer(path->nodes[*level-1]);
1736 path->nodes[*level-1] = next;
1737 *level = btrfs_header_level(next);
1738 path->slots[*level] = 0;
1739 cond_resched();
1740 }
1741 WARN_ON(*level < 0);
1742 WARN_ON(*level >= BTRFS_MAX_LEVEL);
1743
Yan, Zheng4a500fd2010-05-16 10:49:59 -04001744 path->slots[*level] = btrfs_header_nritems(path->nodes[*level]);
Chris Masone02119d2008-09-05 16:13:11 -04001745
1746 cond_resched();
1747 return 0;
1748}
1749
Chris Masond3977122009-01-05 21:25:51 -05001750static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans,
Chris Masone02119d2008-09-05 16:13:11 -04001751 struct btrfs_root *root,
1752 struct btrfs_path *path, int *level,
1753 struct walk_control *wc)
1754{
1755 u64 root_owner;
Chris Masone02119d2008-09-05 16:13:11 -04001756 int i;
1757 int slot;
1758 int ret;
1759
Chris Masond3977122009-01-05 21:25:51 -05001760 for (i = *level; i < BTRFS_MAX_LEVEL - 1 && path->nodes[i]; i++) {
Chris Masone02119d2008-09-05 16:13:11 -04001761 slot = path->slots[i];
Yan, Zheng4a500fd2010-05-16 10:49:59 -04001762 if (slot + 1 < btrfs_header_nritems(path->nodes[i])) {
Chris Masone02119d2008-09-05 16:13:11 -04001763 path->slots[i]++;
1764 *level = i;
1765 WARN_ON(*level == 0);
1766 return 0;
1767 } else {
Zheng Yan31840ae2008-09-23 13:14:14 -04001768 struct extent_buffer *parent;
1769 if (path->nodes[*level] == root->node)
1770 parent = path->nodes[*level];
1771 else
1772 parent = path->nodes[*level + 1];
1773
1774 root_owner = btrfs_header_owner(parent);
Chris Masone02119d2008-09-05 16:13:11 -04001775 wc->process_func(root, path->nodes[*level], wc,
1776 btrfs_header_generation(path->nodes[*level]));
1777 if (wc->free) {
1778 struct extent_buffer *next;
1779
1780 next = path->nodes[*level];
1781
1782 btrfs_tree_lock(next);
1783 clean_tree_block(trans, root, next);
Chris Masonb4ce94d2009-02-04 09:25:08 -05001784 btrfs_set_lock_blocking(next);
Chris Masone02119d2008-09-05 16:13:11 -04001785 btrfs_wait_tree_block_writeback(next);
1786 btrfs_tree_unlock(next);
1787
Chris Masone02119d2008-09-05 16:13:11 -04001788 WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID);
Chris Masond00aff02008-09-11 15:54:42 -04001789 ret = btrfs_free_reserved_extent(root,
Chris Masone02119d2008-09-05 16:13:11 -04001790 path->nodes[*level]->start,
Chris Masond00aff02008-09-11 15:54:42 -04001791 path->nodes[*level]->len);
Chris Masone02119d2008-09-05 16:13:11 -04001792 BUG_ON(ret);
1793 }
1794 free_extent_buffer(path->nodes[*level]);
1795 path->nodes[*level] = NULL;
1796 *level = i + 1;
1797 }
1798 }
1799 return 1;
1800}
1801
1802/*
1803 * drop the reference count on the tree rooted at 'snap'. This traverses
1804 * the tree freeing any blocks that have a ref count of zero after being
1805 * decremented.
1806 */
1807static int walk_log_tree(struct btrfs_trans_handle *trans,
1808 struct btrfs_root *log, struct walk_control *wc)
1809{
1810 int ret = 0;
1811 int wret;
1812 int level;
1813 struct btrfs_path *path;
1814 int i;
1815 int orig_level;
1816
1817 path = btrfs_alloc_path();
Tsutomu Itohdb5b4932011-03-23 08:14:16 +00001818 if (!path)
1819 return -ENOMEM;
Chris Masone02119d2008-09-05 16:13:11 -04001820
1821 level = btrfs_header_level(log->node);
1822 orig_level = level;
1823 path->nodes[level] = log->node;
1824 extent_buffer_get(log->node);
1825 path->slots[level] = 0;
1826
Chris Masond3977122009-01-05 21:25:51 -05001827 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -04001828 wret = walk_down_log_tree(trans, log, path, &level, wc);
1829 if (wret > 0)
1830 break;
1831 if (wret < 0)
1832 ret = wret;
1833
1834 wret = walk_up_log_tree(trans, log, path, &level, wc);
1835 if (wret > 0)
1836 break;
1837 if (wret < 0)
1838 ret = wret;
1839 }
1840
1841 /* was the root node processed? if not, catch it here */
1842 if (path->nodes[orig_level]) {
1843 wc->process_func(log, path->nodes[orig_level], wc,
1844 btrfs_header_generation(path->nodes[orig_level]));
1845 if (wc->free) {
1846 struct extent_buffer *next;
1847
1848 next = path->nodes[orig_level];
1849
1850 btrfs_tree_lock(next);
1851 clean_tree_block(trans, log, next);
Chris Masonb4ce94d2009-02-04 09:25:08 -05001852 btrfs_set_lock_blocking(next);
Chris Masone02119d2008-09-05 16:13:11 -04001853 btrfs_wait_tree_block_writeback(next);
1854 btrfs_tree_unlock(next);
1855
Chris Masone02119d2008-09-05 16:13:11 -04001856 WARN_ON(log->root_key.objectid !=
1857 BTRFS_TREE_LOG_OBJECTID);
Chris Masond00aff02008-09-11 15:54:42 -04001858 ret = btrfs_free_reserved_extent(log, next->start,
1859 next->len);
Chris Masone02119d2008-09-05 16:13:11 -04001860 BUG_ON(ret);
1861 }
1862 }
1863
1864 for (i = 0; i <= orig_level; i++) {
1865 if (path->nodes[i]) {
1866 free_extent_buffer(path->nodes[i]);
1867 path->nodes[i] = NULL;
1868 }
1869 }
1870 btrfs_free_path(path);
Chris Masone02119d2008-09-05 16:13:11 -04001871 return ret;
1872}
1873
Yan Zheng7237f182009-01-21 12:54:03 -05001874/*
1875 * helper function to update the item for a given subvolumes log root
1876 * in the tree of log roots
1877 */
1878static int update_log_root(struct btrfs_trans_handle *trans,
1879 struct btrfs_root *log)
1880{
1881 int ret;
1882
1883 if (log->log_transid == 1) {
1884 /* insert root item on the first sync */
1885 ret = btrfs_insert_root(trans, log->fs_info->log_root_tree,
1886 &log->root_key, &log->root_item);
1887 } else {
1888 ret = btrfs_update_root(trans, log->fs_info->log_root_tree,
1889 &log->root_key, &log->root_item);
1890 }
1891 return ret;
1892}
1893
Chris Mason12fcfd22009-03-24 10:24:20 -04001894static int wait_log_commit(struct btrfs_trans_handle *trans,
1895 struct btrfs_root *root, unsigned long transid)
Chris Masone02119d2008-09-05 16:13:11 -04001896{
1897 DEFINE_WAIT(wait);
Yan Zheng7237f182009-01-21 12:54:03 -05001898 int index = transid % 2;
Chris Masone02119d2008-09-05 16:13:11 -04001899
Yan Zheng7237f182009-01-21 12:54:03 -05001900 /*
1901 * we only allow two pending log transactions at a time,
1902 * so we know that if ours is more than 2 older than the
1903 * current transaction, we're done
1904 */
Chris Masone02119d2008-09-05 16:13:11 -04001905 do {
Yan Zheng7237f182009-01-21 12:54:03 -05001906 prepare_to_wait(&root->log_commit_wait[index],
1907 &wait, TASK_UNINTERRUPTIBLE);
1908 mutex_unlock(&root->log_mutex);
Chris Mason12fcfd22009-03-24 10:24:20 -04001909
1910 if (root->fs_info->last_trans_log_full_commit !=
1911 trans->transid && root->log_transid < transid + 2 &&
Yan Zheng7237f182009-01-21 12:54:03 -05001912 atomic_read(&root->log_commit[index]))
Chris Masone02119d2008-09-05 16:13:11 -04001913 schedule();
Chris Mason12fcfd22009-03-24 10:24:20 -04001914
Yan Zheng7237f182009-01-21 12:54:03 -05001915 finish_wait(&root->log_commit_wait[index], &wait);
1916 mutex_lock(&root->log_mutex);
1917 } while (root->log_transid < transid + 2 &&
1918 atomic_read(&root->log_commit[index]));
1919 return 0;
1920}
1921
Chris Mason12fcfd22009-03-24 10:24:20 -04001922static int wait_for_writer(struct btrfs_trans_handle *trans,
1923 struct btrfs_root *root)
Yan Zheng7237f182009-01-21 12:54:03 -05001924{
1925 DEFINE_WAIT(wait);
1926 while (atomic_read(&root->log_writers)) {
1927 prepare_to_wait(&root->log_writer_wait,
1928 &wait, TASK_UNINTERRUPTIBLE);
1929 mutex_unlock(&root->log_mutex);
Chris Mason12fcfd22009-03-24 10:24:20 -04001930 if (root->fs_info->last_trans_log_full_commit !=
1931 trans->transid && atomic_read(&root->log_writers))
Yan Zheng7237f182009-01-21 12:54:03 -05001932 schedule();
1933 mutex_lock(&root->log_mutex);
1934 finish_wait(&root->log_writer_wait, &wait);
1935 }
Chris Masone02119d2008-09-05 16:13:11 -04001936 return 0;
1937}
1938
1939/*
1940 * btrfs_sync_log does sends a given tree log down to the disk and
1941 * updates the super blocks to record it. When this call is done,
Chris Mason12fcfd22009-03-24 10:24:20 -04001942 * you know that any inodes previously logged are safely on disk only
1943 * if it returns 0.
1944 *
1945 * Any other return value means you need to call btrfs_commit_transaction.
1946 * Some of the edge cases for fsyncing directories that have had unlinks
1947 * or renames done in the past mean that sometimes the only safe
1948 * fsync is to commit the whole FS. When btrfs_sync_log returns -EAGAIN,
1949 * that has happened.
Chris Masone02119d2008-09-05 16:13:11 -04001950 */
1951int btrfs_sync_log(struct btrfs_trans_handle *trans,
1952 struct btrfs_root *root)
1953{
Yan Zheng7237f182009-01-21 12:54:03 -05001954 int index1;
1955 int index2;
Yan, Zheng8cef4e12009-11-12 09:33:26 +00001956 int mark;
Chris Masone02119d2008-09-05 16:13:11 -04001957 int ret;
Chris Masone02119d2008-09-05 16:13:11 -04001958 struct btrfs_root *log = root->log_root;
Yan Zheng7237f182009-01-21 12:54:03 -05001959 struct btrfs_root *log_root_tree = root->fs_info->log_root_tree;
Yan, Zheng8cef4e12009-11-12 09:33:26 +00001960 unsigned long log_transid = 0;
Chris Masone02119d2008-09-05 16:13:11 -04001961
Yan Zheng7237f182009-01-21 12:54:03 -05001962 mutex_lock(&root->log_mutex);
1963 index1 = root->log_transid % 2;
1964 if (atomic_read(&root->log_commit[index1])) {
Chris Mason12fcfd22009-03-24 10:24:20 -04001965 wait_log_commit(trans, root, root->log_transid);
Yan Zheng7237f182009-01-21 12:54:03 -05001966 mutex_unlock(&root->log_mutex);
1967 return 0;
Chris Masone02119d2008-09-05 16:13:11 -04001968 }
Yan Zheng7237f182009-01-21 12:54:03 -05001969 atomic_set(&root->log_commit[index1], 1);
1970
1971 /* wait for previous tree log sync to complete */
1972 if (atomic_read(&root->log_commit[(index1 + 1) % 2]))
Chris Mason12fcfd22009-03-24 10:24:20 -04001973 wait_log_commit(trans, root, root->log_transid - 1);
Chris Masone02119d2008-09-05 16:13:11 -04001974
Yan, Zheng86df7eb2009-10-14 09:24:59 -04001975 while (1) {
Yan Zheng7237f182009-01-21 12:54:03 -05001976 unsigned long batch = root->log_batch;
Yan, Zheng86df7eb2009-10-14 09:24:59 -04001977 if (root->log_multiple_pids) {
1978 mutex_unlock(&root->log_mutex);
1979 schedule_timeout_uninterruptible(1);
1980 mutex_lock(&root->log_mutex);
1981 }
Chris Mason12fcfd22009-03-24 10:24:20 -04001982 wait_for_writer(trans, root);
Yan Zheng7237f182009-01-21 12:54:03 -05001983 if (batch == root->log_batch)
Chris Masone02119d2008-09-05 16:13:11 -04001984 break;
1985 }
Chris Masond0c803c2008-09-11 16:17:57 -04001986
Chris Mason12fcfd22009-03-24 10:24:20 -04001987 /* bail out if we need to do a full commit */
1988 if (root->fs_info->last_trans_log_full_commit == trans->transid) {
1989 ret = -EAGAIN;
1990 mutex_unlock(&root->log_mutex);
1991 goto out;
1992 }
1993
Yan, Zheng8cef4e12009-11-12 09:33:26 +00001994 log_transid = root->log_transid;
1995 if (log_transid % 2 == 0)
1996 mark = EXTENT_DIRTY;
1997 else
1998 mark = EXTENT_NEW;
1999
Chris Mason690587d2009-10-13 13:29:19 -04002000 /* we start IO on all the marked extents here, but we don't actually
2001 * wait for them until later.
2002 */
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002003 ret = btrfs_write_marked_extents(log, &log->dirty_log_pages, mark);
Chris Masone02119d2008-09-05 16:13:11 -04002004 BUG_ON(ret);
Yan Zheng7237f182009-01-21 12:54:03 -05002005
Yan Zheng5d4f98a2009-06-10 10:45:14 -04002006 btrfs_set_root_node(&log->root_item, log->node);
Yan Zheng7237f182009-01-21 12:54:03 -05002007
2008 root->log_batch = 0;
2009 root->log_transid++;
2010 log->log_transid = root->log_transid;
Josef Bacikff782e02009-10-08 15:30:04 -04002011 root->log_start_pid = 0;
Yan Zheng7237f182009-01-21 12:54:03 -05002012 smp_mb();
2013 /*
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002014 * IO has been started, blocks of the log tree have WRITTEN flag set
2015 * in their headers. new modifications of the log will be written to
2016 * new positions. so it's safe to allow log writers to go in.
Yan Zheng7237f182009-01-21 12:54:03 -05002017 */
2018 mutex_unlock(&root->log_mutex);
2019
2020 mutex_lock(&log_root_tree->log_mutex);
2021 log_root_tree->log_batch++;
2022 atomic_inc(&log_root_tree->log_writers);
2023 mutex_unlock(&log_root_tree->log_mutex);
2024
2025 ret = update_log_root(trans, log);
Yan Zheng7237f182009-01-21 12:54:03 -05002026
2027 mutex_lock(&log_root_tree->log_mutex);
2028 if (atomic_dec_and_test(&log_root_tree->log_writers)) {
2029 smp_mb();
2030 if (waitqueue_active(&log_root_tree->log_writer_wait))
2031 wake_up(&log_root_tree->log_writer_wait);
2032 }
2033
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002034 if (ret) {
2035 BUG_ON(ret != -ENOSPC);
2036 root->fs_info->last_trans_log_full_commit = trans->transid;
2037 btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
2038 mutex_unlock(&log_root_tree->log_mutex);
2039 ret = -EAGAIN;
2040 goto out;
2041 }
2042
Yan Zheng7237f182009-01-21 12:54:03 -05002043 index2 = log_root_tree->log_transid % 2;
2044 if (atomic_read(&log_root_tree->log_commit[index2])) {
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002045 btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
Chris Mason12fcfd22009-03-24 10:24:20 -04002046 wait_log_commit(trans, log_root_tree,
2047 log_root_tree->log_transid);
Yan Zheng7237f182009-01-21 12:54:03 -05002048 mutex_unlock(&log_root_tree->log_mutex);
Chris Masonb31eabd2011-01-31 16:48:24 -05002049 ret = 0;
Yan Zheng7237f182009-01-21 12:54:03 -05002050 goto out;
2051 }
2052 atomic_set(&log_root_tree->log_commit[index2], 1);
2053
Chris Mason12fcfd22009-03-24 10:24:20 -04002054 if (atomic_read(&log_root_tree->log_commit[(index2 + 1) % 2])) {
2055 wait_log_commit(trans, log_root_tree,
2056 log_root_tree->log_transid - 1);
2057 }
Yan Zheng7237f182009-01-21 12:54:03 -05002058
Chris Mason12fcfd22009-03-24 10:24:20 -04002059 wait_for_writer(trans, log_root_tree);
2060
2061 /*
2062 * now that we've moved on to the tree of log tree roots,
2063 * check the full commit flag again
2064 */
2065 if (root->fs_info->last_trans_log_full_commit == trans->transid) {
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002066 btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
Chris Mason12fcfd22009-03-24 10:24:20 -04002067 mutex_unlock(&log_root_tree->log_mutex);
2068 ret = -EAGAIN;
2069 goto out_wake_log_root;
2070 }
Yan Zheng7237f182009-01-21 12:54:03 -05002071
2072 ret = btrfs_write_and_wait_marked_extents(log_root_tree,
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002073 &log_root_tree->dirty_log_pages,
2074 EXTENT_DIRTY | EXTENT_NEW);
Chris Masone02119d2008-09-05 16:13:11 -04002075 BUG_ON(ret);
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002076 btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
Chris Masone02119d2008-09-05 16:13:11 -04002077
2078 btrfs_set_super_log_root(&root->fs_info->super_for_commit,
Yan Zheng7237f182009-01-21 12:54:03 -05002079 log_root_tree->node->start);
Chris Masone02119d2008-09-05 16:13:11 -04002080 btrfs_set_super_log_root_level(&root->fs_info->super_for_commit,
Yan Zheng7237f182009-01-21 12:54:03 -05002081 btrfs_header_level(log_root_tree->node));
Chris Masone02119d2008-09-05 16:13:11 -04002082
Yan Zheng7237f182009-01-21 12:54:03 -05002083 log_root_tree->log_batch = 0;
2084 log_root_tree->log_transid++;
Chris Masone02119d2008-09-05 16:13:11 -04002085 smp_mb();
Yan Zheng7237f182009-01-21 12:54:03 -05002086
2087 mutex_unlock(&log_root_tree->log_mutex);
2088
2089 /*
2090 * nobody else is going to jump in and write the the ctree
2091 * super here because the log_commit atomic below is protecting
2092 * us. We must be called with a transaction handle pinning
2093 * the running transaction open, so a full commit can't hop
2094 * in and cause problems either.
2095 */
Chris Mason47226072009-10-13 12:55:09 -04002096 write_ctree_super(trans, root->fs_info->tree_root, 1);
Chris Mason12fcfd22009-03-24 10:24:20 -04002097 ret = 0;
Yan Zheng7237f182009-01-21 12:54:03 -05002098
Chris Mason257c62e2009-10-13 13:21:08 -04002099 mutex_lock(&root->log_mutex);
2100 if (root->last_log_commit < log_transid)
2101 root->last_log_commit = log_transid;
2102 mutex_unlock(&root->log_mutex);
2103
Chris Mason12fcfd22009-03-24 10:24:20 -04002104out_wake_log_root:
Yan Zheng7237f182009-01-21 12:54:03 -05002105 atomic_set(&log_root_tree->log_commit[index2], 0);
2106 smp_mb();
2107 if (waitqueue_active(&log_root_tree->log_commit_wait[index2]))
2108 wake_up(&log_root_tree->log_commit_wait[index2]);
Chris Masone02119d2008-09-05 16:13:11 -04002109out:
Yan Zheng7237f182009-01-21 12:54:03 -05002110 atomic_set(&root->log_commit[index1], 0);
2111 smp_mb();
2112 if (waitqueue_active(&root->log_commit_wait[index1]))
2113 wake_up(&root->log_commit_wait[index1]);
Chris Masonb31eabd2011-01-31 16:48:24 -05002114 return ret;
Chris Masone02119d2008-09-05 16:13:11 -04002115}
2116
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002117static void free_log_tree(struct btrfs_trans_handle *trans,
2118 struct btrfs_root *log)
Chris Masone02119d2008-09-05 16:13:11 -04002119{
2120 int ret;
Chris Masond0c803c2008-09-11 16:17:57 -04002121 u64 start;
2122 u64 end;
Chris Masone02119d2008-09-05 16:13:11 -04002123 struct walk_control wc = {
2124 .free = 1,
2125 .process_func = process_one_buffer
2126 };
2127
Chris Masone02119d2008-09-05 16:13:11 -04002128 ret = walk_log_tree(trans, log, &wc);
2129 BUG_ON(ret);
2130
Chris Masond3977122009-01-05 21:25:51 -05002131 while (1) {
Chris Masond0c803c2008-09-11 16:17:57 -04002132 ret = find_first_extent_bit(&log->dirty_log_pages,
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002133 0, &start, &end, EXTENT_DIRTY | EXTENT_NEW);
Chris Masond0c803c2008-09-11 16:17:57 -04002134 if (ret)
2135 break;
2136
Yan, Zheng8cef4e12009-11-12 09:33:26 +00002137 clear_extent_bits(&log->dirty_log_pages, start, end,
2138 EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS);
Chris Masond0c803c2008-09-11 16:17:57 -04002139 }
2140
Yan Zheng7237f182009-01-21 12:54:03 -05002141 free_extent_buffer(log->node);
2142 kfree(log);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002143}
2144
2145/*
2146 * free all the extents used by the tree log. This should be called
2147 * at commit time of the full transaction
2148 */
2149int btrfs_free_log(struct btrfs_trans_handle *trans, struct btrfs_root *root)
2150{
2151 if (root->log_root) {
2152 free_log_tree(trans, root->log_root);
2153 root->log_root = NULL;
2154 }
2155 return 0;
2156}
2157
2158int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
2159 struct btrfs_fs_info *fs_info)
2160{
2161 if (fs_info->log_root_tree) {
2162 free_log_tree(trans, fs_info->log_root_tree);
2163 fs_info->log_root_tree = NULL;
2164 }
Chris Masone02119d2008-09-05 16:13:11 -04002165 return 0;
2166}
2167
2168/*
Chris Masone02119d2008-09-05 16:13:11 -04002169 * If both a file and directory are logged, and unlinks or renames are
2170 * mixed in, we have a few interesting corners:
2171 *
2172 * create file X in dir Y
2173 * link file X to X.link in dir Y
2174 * fsync file X
2175 * unlink file X but leave X.link
2176 * fsync dir Y
2177 *
2178 * After a crash we would expect only X.link to exist. But file X
2179 * didn't get fsync'd again so the log has back refs for X and X.link.
2180 *
2181 * We solve this by removing directory entries and inode backrefs from the
2182 * log when a file that was logged in the current transaction is
2183 * unlinked. Any later fsync will include the updated log entries, and
2184 * we'll be able to reconstruct the proper directory items from backrefs.
2185 *
2186 * This optimizations allows us to avoid relogging the entire inode
2187 * or the entire directory.
2188 */
2189int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
2190 struct btrfs_root *root,
2191 const char *name, int name_len,
2192 struct inode *dir, u64 index)
2193{
2194 struct btrfs_root *log;
2195 struct btrfs_dir_item *di;
2196 struct btrfs_path *path;
2197 int ret;
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002198 int err = 0;
Chris Masone02119d2008-09-05 16:13:11 -04002199 int bytes_del = 0;
2200
Chris Mason3a5f1d42008-09-11 15:53:37 -04002201 if (BTRFS_I(dir)->logged_trans < trans->transid)
2202 return 0;
2203
Chris Masone02119d2008-09-05 16:13:11 -04002204 ret = join_running_log_trans(root);
2205 if (ret)
2206 return 0;
2207
2208 mutex_lock(&BTRFS_I(dir)->log_mutex);
2209
2210 log = root->log_root;
2211 path = btrfs_alloc_path();
Tsutomu Itoha62f44a2011-04-25 19:43:51 -04002212 if (!path) {
2213 err = -ENOMEM;
2214 goto out_unlock;
2215 }
liubo2a29edc2011-01-26 06:22:08 +00002216
Chris Masone02119d2008-09-05 16:13:11 -04002217 di = btrfs_lookup_dir_item(trans, log, path, dir->i_ino,
2218 name, name_len, -1);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002219 if (IS_ERR(di)) {
2220 err = PTR_ERR(di);
2221 goto fail;
2222 }
2223 if (di) {
Chris Masone02119d2008-09-05 16:13:11 -04002224 ret = btrfs_delete_one_dir_name(trans, log, path, di);
2225 bytes_del += name_len;
2226 BUG_ON(ret);
2227 }
2228 btrfs_release_path(log, path);
2229 di = btrfs_lookup_dir_index_item(trans, log, path, dir->i_ino,
2230 index, name, name_len, -1);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002231 if (IS_ERR(di)) {
2232 err = PTR_ERR(di);
2233 goto fail;
2234 }
2235 if (di) {
Chris Masone02119d2008-09-05 16:13:11 -04002236 ret = btrfs_delete_one_dir_name(trans, log, path, di);
2237 bytes_del += name_len;
2238 BUG_ON(ret);
2239 }
2240
2241 /* update the directory size in the log to reflect the names
2242 * we have removed
2243 */
2244 if (bytes_del) {
2245 struct btrfs_key key;
2246
2247 key.objectid = dir->i_ino;
2248 key.offset = 0;
2249 key.type = BTRFS_INODE_ITEM_KEY;
2250 btrfs_release_path(log, path);
2251
2252 ret = btrfs_search_slot(trans, log, &key, path, 0, 1);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002253 if (ret < 0) {
2254 err = ret;
2255 goto fail;
2256 }
Chris Masone02119d2008-09-05 16:13:11 -04002257 if (ret == 0) {
2258 struct btrfs_inode_item *item;
2259 u64 i_size;
2260
2261 item = btrfs_item_ptr(path->nodes[0], path->slots[0],
2262 struct btrfs_inode_item);
2263 i_size = btrfs_inode_size(path->nodes[0], item);
2264 if (i_size > bytes_del)
2265 i_size -= bytes_del;
2266 else
2267 i_size = 0;
2268 btrfs_set_inode_size(path->nodes[0], item, i_size);
2269 btrfs_mark_buffer_dirty(path->nodes[0]);
2270 } else
2271 ret = 0;
2272 btrfs_release_path(log, path);
2273 }
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002274fail:
Chris Masone02119d2008-09-05 16:13:11 -04002275 btrfs_free_path(path);
Tsutomu Itoha62f44a2011-04-25 19:43:51 -04002276out_unlock:
Chris Masone02119d2008-09-05 16:13:11 -04002277 mutex_unlock(&BTRFS_I(dir)->log_mutex);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002278 if (ret == -ENOSPC) {
2279 root->fs_info->last_trans_log_full_commit = trans->transid;
2280 ret = 0;
2281 }
Chris Mason12fcfd22009-03-24 10:24:20 -04002282 btrfs_end_log_trans(root);
Chris Masone02119d2008-09-05 16:13:11 -04002283
Andi Kleen411fc6b2010-10-29 15:14:31 -04002284 return err;
Chris Masone02119d2008-09-05 16:13:11 -04002285}
2286
2287/* see comments for btrfs_del_dir_entries_in_log */
2288int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans,
2289 struct btrfs_root *root,
2290 const char *name, int name_len,
2291 struct inode *inode, u64 dirid)
2292{
2293 struct btrfs_root *log;
2294 u64 index;
2295 int ret;
2296
Chris Mason3a5f1d42008-09-11 15:53:37 -04002297 if (BTRFS_I(inode)->logged_trans < trans->transid)
2298 return 0;
2299
Chris Masone02119d2008-09-05 16:13:11 -04002300 ret = join_running_log_trans(root);
2301 if (ret)
2302 return 0;
2303 log = root->log_root;
2304 mutex_lock(&BTRFS_I(inode)->log_mutex);
2305
2306 ret = btrfs_del_inode_ref(trans, log, name, name_len, inode->i_ino,
2307 dirid, &index);
2308 mutex_unlock(&BTRFS_I(inode)->log_mutex);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002309 if (ret == -ENOSPC) {
2310 root->fs_info->last_trans_log_full_commit = trans->transid;
2311 ret = 0;
2312 }
Chris Mason12fcfd22009-03-24 10:24:20 -04002313 btrfs_end_log_trans(root);
Chris Masone02119d2008-09-05 16:13:11 -04002314
Chris Masone02119d2008-09-05 16:13:11 -04002315 return ret;
2316}
2317
2318/*
2319 * creates a range item in the log for 'dirid'. first_offset and
2320 * last_offset tell us which parts of the key space the log should
2321 * be considered authoritative for.
2322 */
2323static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans,
2324 struct btrfs_root *log,
2325 struct btrfs_path *path,
2326 int key_type, u64 dirid,
2327 u64 first_offset, u64 last_offset)
2328{
2329 int ret;
2330 struct btrfs_key key;
2331 struct btrfs_dir_log_item *item;
2332
2333 key.objectid = dirid;
2334 key.offset = first_offset;
2335 if (key_type == BTRFS_DIR_ITEM_KEY)
2336 key.type = BTRFS_DIR_LOG_ITEM_KEY;
2337 else
2338 key.type = BTRFS_DIR_LOG_INDEX_KEY;
2339 ret = btrfs_insert_empty_item(trans, log, path, &key, sizeof(*item));
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002340 if (ret)
2341 return ret;
Chris Masone02119d2008-09-05 16:13:11 -04002342
2343 item = btrfs_item_ptr(path->nodes[0], path->slots[0],
2344 struct btrfs_dir_log_item);
2345 btrfs_set_dir_log_end(path->nodes[0], item, last_offset);
2346 btrfs_mark_buffer_dirty(path->nodes[0]);
2347 btrfs_release_path(log, path);
2348 return 0;
2349}
2350
2351/*
2352 * log all the items included in the current transaction for a given
2353 * directory. This also creates the range items in the log tree required
2354 * to replay anything deleted before the fsync
2355 */
2356static noinline int log_dir_items(struct btrfs_trans_handle *trans,
2357 struct btrfs_root *root, struct inode *inode,
2358 struct btrfs_path *path,
2359 struct btrfs_path *dst_path, int key_type,
2360 u64 min_offset, u64 *last_offset_ret)
2361{
2362 struct btrfs_key min_key;
2363 struct btrfs_key max_key;
2364 struct btrfs_root *log = root->log_root;
2365 struct extent_buffer *src;
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002366 int err = 0;
Chris Masone02119d2008-09-05 16:13:11 -04002367 int ret;
2368 int i;
2369 int nritems;
2370 u64 first_offset = min_offset;
2371 u64 last_offset = (u64)-1;
2372
2373 log = root->log_root;
2374 max_key.objectid = inode->i_ino;
2375 max_key.offset = (u64)-1;
2376 max_key.type = key_type;
2377
2378 min_key.objectid = inode->i_ino;
2379 min_key.type = key_type;
2380 min_key.offset = min_offset;
2381
2382 path->keep_locks = 1;
2383
2384 ret = btrfs_search_forward(root, &min_key, &max_key,
2385 path, 0, trans->transid);
2386
2387 /*
2388 * we didn't find anything from this transaction, see if there
2389 * is anything at all
2390 */
2391 if (ret != 0 || min_key.objectid != inode->i_ino ||
2392 min_key.type != key_type) {
2393 min_key.objectid = inode->i_ino;
2394 min_key.type = key_type;
2395 min_key.offset = (u64)-1;
2396 btrfs_release_path(root, path);
2397 ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0);
2398 if (ret < 0) {
2399 btrfs_release_path(root, path);
2400 return ret;
2401 }
2402 ret = btrfs_previous_item(root, path, inode->i_ino, key_type);
2403
2404 /* if ret == 0 there are items for this type,
2405 * create a range to tell us the last key of this type.
2406 * otherwise, there are no items in this directory after
2407 * *min_offset, and we create a range to indicate that.
2408 */
2409 if (ret == 0) {
2410 struct btrfs_key tmp;
2411 btrfs_item_key_to_cpu(path->nodes[0], &tmp,
2412 path->slots[0]);
Chris Masond3977122009-01-05 21:25:51 -05002413 if (key_type == tmp.type)
Chris Masone02119d2008-09-05 16:13:11 -04002414 first_offset = max(min_offset, tmp.offset) + 1;
Chris Masone02119d2008-09-05 16:13:11 -04002415 }
2416 goto done;
2417 }
2418
2419 /* go backward to find any previous key */
2420 ret = btrfs_previous_item(root, path, inode->i_ino, key_type);
2421 if (ret == 0) {
2422 struct btrfs_key tmp;
2423 btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]);
2424 if (key_type == tmp.type) {
2425 first_offset = tmp.offset;
2426 ret = overwrite_item(trans, log, dst_path,
2427 path->nodes[0], path->slots[0],
2428 &tmp);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002429 if (ret) {
2430 err = ret;
2431 goto done;
2432 }
Chris Masone02119d2008-09-05 16:13:11 -04002433 }
2434 }
2435 btrfs_release_path(root, path);
2436
2437 /* find the first key from this transaction again */
2438 ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0);
2439 if (ret != 0) {
2440 WARN_ON(1);
2441 goto done;
2442 }
2443
2444 /*
2445 * we have a block from this transaction, log every item in it
2446 * from our directory
2447 */
Chris Masond3977122009-01-05 21:25:51 -05002448 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -04002449 struct btrfs_key tmp;
2450 src = path->nodes[0];
2451 nritems = btrfs_header_nritems(src);
2452 for (i = path->slots[0]; i < nritems; i++) {
2453 btrfs_item_key_to_cpu(src, &min_key, i);
2454
2455 if (min_key.objectid != inode->i_ino ||
2456 min_key.type != key_type)
2457 goto done;
2458 ret = overwrite_item(trans, log, dst_path, src, i,
2459 &min_key);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002460 if (ret) {
2461 err = ret;
2462 goto done;
2463 }
Chris Masone02119d2008-09-05 16:13:11 -04002464 }
2465 path->slots[0] = nritems;
2466
2467 /*
2468 * look ahead to the next item and see if it is also
2469 * from this directory and from this transaction
2470 */
2471 ret = btrfs_next_leaf(root, path);
2472 if (ret == 1) {
2473 last_offset = (u64)-1;
2474 goto done;
2475 }
2476 btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]);
2477 if (tmp.objectid != inode->i_ino || tmp.type != key_type) {
2478 last_offset = (u64)-1;
2479 goto done;
2480 }
2481 if (btrfs_header_generation(path->nodes[0]) != trans->transid) {
2482 ret = overwrite_item(trans, log, dst_path,
2483 path->nodes[0], path->slots[0],
2484 &tmp);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002485 if (ret)
2486 err = ret;
2487 else
2488 last_offset = tmp.offset;
Chris Masone02119d2008-09-05 16:13:11 -04002489 goto done;
2490 }
2491 }
2492done:
Chris Masone02119d2008-09-05 16:13:11 -04002493 btrfs_release_path(root, path);
2494 btrfs_release_path(log, dst_path);
2495
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002496 if (err == 0) {
2497 *last_offset_ret = last_offset;
2498 /*
2499 * insert the log range keys to indicate where the log
2500 * is valid
2501 */
2502 ret = insert_dir_log_key(trans, log, path, key_type,
2503 inode->i_ino, first_offset,
2504 last_offset);
2505 if (ret)
2506 err = ret;
2507 }
2508 return err;
Chris Masone02119d2008-09-05 16:13:11 -04002509}
2510
2511/*
2512 * logging directories is very similar to logging inodes, We find all the items
2513 * from the current transaction and write them to the log.
2514 *
2515 * The recovery code scans the directory in the subvolume, and if it finds a
2516 * key in the range logged that is not present in the log tree, then it means
2517 * that dir entry was unlinked during the transaction.
2518 *
2519 * In order for that scan to work, we must include one key smaller than
2520 * the smallest logged by this transaction and one key larger than the largest
2521 * key logged by this transaction.
2522 */
2523static noinline int log_directory_changes(struct btrfs_trans_handle *trans,
2524 struct btrfs_root *root, struct inode *inode,
2525 struct btrfs_path *path,
2526 struct btrfs_path *dst_path)
2527{
2528 u64 min_key;
2529 u64 max_key;
2530 int ret;
2531 int key_type = BTRFS_DIR_ITEM_KEY;
2532
2533again:
2534 min_key = 0;
2535 max_key = 0;
Chris Masond3977122009-01-05 21:25:51 -05002536 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -04002537 ret = log_dir_items(trans, root, inode, path,
2538 dst_path, key_type, min_key,
2539 &max_key);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002540 if (ret)
2541 return ret;
Chris Masone02119d2008-09-05 16:13:11 -04002542 if (max_key == (u64)-1)
2543 break;
2544 min_key = max_key + 1;
2545 }
2546
2547 if (key_type == BTRFS_DIR_ITEM_KEY) {
2548 key_type = BTRFS_DIR_INDEX_KEY;
2549 goto again;
2550 }
2551 return 0;
2552}
2553
2554/*
2555 * a helper function to drop items from the log before we relog an
2556 * inode. max_key_type indicates the highest item type to remove.
2557 * This cannot be run for file data extents because it does not
2558 * free the extents they point to.
2559 */
2560static int drop_objectid_items(struct btrfs_trans_handle *trans,
2561 struct btrfs_root *log,
2562 struct btrfs_path *path,
2563 u64 objectid, int max_key_type)
2564{
2565 int ret;
2566 struct btrfs_key key;
2567 struct btrfs_key found_key;
2568
2569 key.objectid = objectid;
2570 key.type = max_key_type;
2571 key.offset = (u64)-1;
2572
Chris Masond3977122009-01-05 21:25:51 -05002573 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -04002574 ret = btrfs_search_slot(trans, log, &key, path, -1, 1);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002575 BUG_ON(ret == 0);
2576 if (ret < 0)
Chris Masone02119d2008-09-05 16:13:11 -04002577 break;
2578
2579 if (path->slots[0] == 0)
2580 break;
2581
2582 path->slots[0]--;
2583 btrfs_item_key_to_cpu(path->nodes[0], &found_key,
2584 path->slots[0]);
2585
2586 if (found_key.objectid != objectid)
2587 break;
2588
2589 ret = btrfs_del_item(trans, log, path);
2590 BUG_ON(ret);
2591 btrfs_release_path(log, path);
2592 }
2593 btrfs_release_path(log, path);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002594 return ret;
Chris Masone02119d2008-09-05 16:13:11 -04002595}
2596
Chris Mason31ff1cd2008-09-11 16:17:57 -04002597static noinline int copy_items(struct btrfs_trans_handle *trans,
2598 struct btrfs_root *log,
2599 struct btrfs_path *dst_path,
2600 struct extent_buffer *src,
2601 int start_slot, int nr, int inode_only)
2602{
2603 unsigned long src_offset;
2604 unsigned long dst_offset;
2605 struct btrfs_file_extent_item *extent;
2606 struct btrfs_inode_item *inode_item;
2607 int ret;
2608 struct btrfs_key *ins_keys;
2609 u32 *ins_sizes;
2610 char *ins_data;
2611 int i;
Chris Masond20f7042008-12-08 16:58:54 -05002612 struct list_head ordered_sums;
2613
2614 INIT_LIST_HEAD(&ordered_sums);
Chris Mason31ff1cd2008-09-11 16:17:57 -04002615
2616 ins_data = kmalloc(nr * sizeof(struct btrfs_key) +
2617 nr * sizeof(u32), GFP_NOFS);
liubo2a29edc2011-01-26 06:22:08 +00002618 if (!ins_data)
2619 return -ENOMEM;
2620
Chris Mason31ff1cd2008-09-11 16:17:57 -04002621 ins_sizes = (u32 *)ins_data;
2622 ins_keys = (struct btrfs_key *)(ins_data + nr * sizeof(u32));
2623
2624 for (i = 0; i < nr; i++) {
2625 ins_sizes[i] = btrfs_item_size_nr(src, i + start_slot);
2626 btrfs_item_key_to_cpu(src, ins_keys + i, i + start_slot);
2627 }
2628 ret = btrfs_insert_empty_items(trans, log, dst_path,
2629 ins_keys, ins_sizes, nr);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002630 if (ret) {
2631 kfree(ins_data);
2632 return ret;
2633 }
Chris Mason31ff1cd2008-09-11 16:17:57 -04002634
Yan Zheng5d4f98a2009-06-10 10:45:14 -04002635 for (i = 0; i < nr; i++, dst_path->slots[0]++) {
Chris Mason31ff1cd2008-09-11 16:17:57 -04002636 dst_offset = btrfs_item_ptr_offset(dst_path->nodes[0],
2637 dst_path->slots[0]);
2638
2639 src_offset = btrfs_item_ptr_offset(src, start_slot + i);
2640
2641 copy_extent_buffer(dst_path->nodes[0], src, dst_offset,
2642 src_offset, ins_sizes[i]);
2643
2644 if (inode_only == LOG_INODE_EXISTS &&
2645 ins_keys[i].type == BTRFS_INODE_ITEM_KEY) {
2646 inode_item = btrfs_item_ptr(dst_path->nodes[0],
2647 dst_path->slots[0],
2648 struct btrfs_inode_item);
2649 btrfs_set_inode_size(dst_path->nodes[0], inode_item, 0);
2650
2651 /* set the generation to zero so the recover code
2652 * can tell the difference between an logging
2653 * just to say 'this inode exists' and a logging
2654 * to say 'update this inode with these values'
2655 */
2656 btrfs_set_inode_generation(dst_path->nodes[0],
2657 inode_item, 0);
2658 }
2659 /* take a reference on file data extents so that truncates
2660 * or deletes of this inode don't have to relog the inode
2661 * again
2662 */
2663 if (btrfs_key_type(ins_keys + i) == BTRFS_EXTENT_DATA_KEY) {
2664 int found_type;
2665 extent = btrfs_item_ptr(src, start_slot + i,
2666 struct btrfs_file_extent_item);
2667
2668 found_type = btrfs_file_extent_type(src, extent);
Yan Zhengd899e052008-10-30 14:25:28 -04002669 if (found_type == BTRFS_FILE_EXTENT_REG ||
2670 found_type == BTRFS_FILE_EXTENT_PREALLOC) {
Yan Zheng5d4f98a2009-06-10 10:45:14 -04002671 u64 ds, dl, cs, cl;
2672 ds = btrfs_file_extent_disk_bytenr(src,
2673 extent);
2674 /* ds == 0 is a hole */
2675 if (ds == 0)
2676 continue;
2677
2678 dl = btrfs_file_extent_disk_num_bytes(src,
2679 extent);
2680 cs = btrfs_file_extent_offset(src, extent);
2681 cl = btrfs_file_extent_num_bytes(src,
Joe Perchesa419aef2009-08-18 11:18:35 -07002682 extent);
Chris Mason580afd72008-12-08 19:15:39 -05002683 if (btrfs_file_extent_compression(src,
2684 extent)) {
2685 cs = 0;
2686 cl = dl;
2687 }
Yan Zheng5d4f98a2009-06-10 10:45:14 -04002688
2689 ret = btrfs_lookup_csums_range(
2690 log->fs_info->csum_root,
2691 ds + cs, ds + cs + cl - 1,
2692 &ordered_sums);
2693 BUG_ON(ret);
Chris Mason31ff1cd2008-09-11 16:17:57 -04002694 }
2695 }
Chris Mason31ff1cd2008-09-11 16:17:57 -04002696 }
2697
2698 btrfs_mark_buffer_dirty(dst_path->nodes[0]);
2699 btrfs_release_path(log, dst_path);
2700 kfree(ins_data);
Chris Masond20f7042008-12-08 16:58:54 -05002701
2702 /*
2703 * we have to do this after the loop above to avoid changing the
2704 * log tree while trying to change the log tree.
2705 */
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002706 ret = 0;
Chris Masond3977122009-01-05 21:25:51 -05002707 while (!list_empty(&ordered_sums)) {
Chris Masond20f7042008-12-08 16:58:54 -05002708 struct btrfs_ordered_sum *sums = list_entry(ordered_sums.next,
2709 struct btrfs_ordered_sum,
2710 list);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002711 if (!ret)
2712 ret = btrfs_csum_file_blocks(trans, log, sums);
Chris Masond20f7042008-12-08 16:58:54 -05002713 list_del(&sums->list);
2714 kfree(sums);
2715 }
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002716 return ret;
Chris Mason31ff1cd2008-09-11 16:17:57 -04002717}
2718
Chris Masone02119d2008-09-05 16:13:11 -04002719/* log a single inode in the tree log.
2720 * At least one parent directory for this inode must exist in the tree
2721 * or be logged already.
2722 *
2723 * Any items from this inode changed by the current transaction are copied
2724 * to the log tree. An extra reference is taken on any extents in this
2725 * file, allowing us to avoid a whole pile of corner cases around logging
2726 * blocks that have been removed from the tree.
2727 *
2728 * See LOG_INODE_ALL and related defines for a description of what inode_only
2729 * does.
2730 *
2731 * This handles both files and directories.
2732 */
Chris Mason12fcfd22009-03-24 10:24:20 -04002733static int btrfs_log_inode(struct btrfs_trans_handle *trans,
Chris Masone02119d2008-09-05 16:13:11 -04002734 struct btrfs_root *root, struct inode *inode,
2735 int inode_only)
2736{
2737 struct btrfs_path *path;
2738 struct btrfs_path *dst_path;
2739 struct btrfs_key min_key;
2740 struct btrfs_key max_key;
2741 struct btrfs_root *log = root->log_root;
Chris Mason31ff1cd2008-09-11 16:17:57 -04002742 struct extent_buffer *src = NULL;
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002743 int err = 0;
Chris Masone02119d2008-09-05 16:13:11 -04002744 int ret;
Chris Mason3a5f1d42008-09-11 15:53:37 -04002745 int nritems;
Chris Mason31ff1cd2008-09-11 16:17:57 -04002746 int ins_start_slot = 0;
2747 int ins_nr;
Chris Masone02119d2008-09-05 16:13:11 -04002748
2749 log = root->log_root;
2750
2751 path = btrfs_alloc_path();
Tsutomu Itoh5df67082011-02-01 09:17:35 +00002752 if (!path)
2753 return -ENOMEM;
Chris Masone02119d2008-09-05 16:13:11 -04002754 dst_path = btrfs_alloc_path();
Tsutomu Itoh5df67082011-02-01 09:17:35 +00002755 if (!dst_path) {
2756 btrfs_free_path(path);
2757 return -ENOMEM;
2758 }
Chris Masone02119d2008-09-05 16:13:11 -04002759
2760 min_key.objectid = inode->i_ino;
2761 min_key.type = BTRFS_INODE_ITEM_KEY;
2762 min_key.offset = 0;
2763
2764 max_key.objectid = inode->i_ino;
Chris Mason12fcfd22009-03-24 10:24:20 -04002765
2766 /* today the code can only do partial logging of directories */
2767 if (!S_ISDIR(inode->i_mode))
2768 inode_only = LOG_INODE_ALL;
2769
Chris Masone02119d2008-09-05 16:13:11 -04002770 if (inode_only == LOG_INODE_EXISTS || S_ISDIR(inode->i_mode))
2771 max_key.type = BTRFS_XATTR_ITEM_KEY;
2772 else
2773 max_key.type = (u8)-1;
2774 max_key.offset = (u64)-1;
2775
Chris Masone02119d2008-09-05 16:13:11 -04002776 mutex_lock(&BTRFS_I(inode)->log_mutex);
2777
2778 /*
2779 * a brute force approach to making sure we get the most uptodate
2780 * copies of everything.
2781 */
2782 if (S_ISDIR(inode->i_mode)) {
2783 int max_key_type = BTRFS_DIR_LOG_INDEX_KEY;
2784
2785 if (inode_only == LOG_INODE_EXISTS)
2786 max_key_type = BTRFS_XATTR_ITEM_KEY;
2787 ret = drop_objectid_items(trans, log, path,
2788 inode->i_ino, max_key_type);
2789 } else {
2790 ret = btrfs_truncate_inode_items(trans, log, inode, 0, 0);
2791 }
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002792 if (ret) {
2793 err = ret;
2794 goto out_unlock;
2795 }
Chris Masone02119d2008-09-05 16:13:11 -04002796 path->keep_locks = 1;
2797
Chris Masond3977122009-01-05 21:25:51 -05002798 while (1) {
Chris Mason31ff1cd2008-09-11 16:17:57 -04002799 ins_nr = 0;
Chris Masone02119d2008-09-05 16:13:11 -04002800 ret = btrfs_search_forward(root, &min_key, &max_key,
2801 path, 0, trans->transid);
2802 if (ret != 0)
2803 break;
Chris Mason3a5f1d42008-09-11 15:53:37 -04002804again:
Chris Mason31ff1cd2008-09-11 16:17:57 -04002805 /* note, ins_nr might be > 0 here, cleanup outside the loop */
Chris Masone02119d2008-09-05 16:13:11 -04002806 if (min_key.objectid != inode->i_ino)
2807 break;
2808 if (min_key.type > max_key.type)
2809 break;
Chris Mason31ff1cd2008-09-11 16:17:57 -04002810
Chris Masone02119d2008-09-05 16:13:11 -04002811 src = path->nodes[0];
Chris Mason31ff1cd2008-09-11 16:17:57 -04002812 if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) {
2813 ins_nr++;
2814 goto next_slot;
2815 } else if (!ins_nr) {
2816 ins_start_slot = path->slots[0];
2817 ins_nr = 1;
2818 goto next_slot;
Chris Masone02119d2008-09-05 16:13:11 -04002819 }
2820
Chris Mason31ff1cd2008-09-11 16:17:57 -04002821 ret = copy_items(trans, log, dst_path, src, ins_start_slot,
2822 ins_nr, inode_only);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002823 if (ret) {
2824 err = ret;
2825 goto out_unlock;
2826 }
Chris Mason31ff1cd2008-09-11 16:17:57 -04002827 ins_nr = 1;
2828 ins_start_slot = path->slots[0];
2829next_slot:
Chris Masone02119d2008-09-05 16:13:11 -04002830
Chris Mason3a5f1d42008-09-11 15:53:37 -04002831 nritems = btrfs_header_nritems(path->nodes[0]);
2832 path->slots[0]++;
2833 if (path->slots[0] < nritems) {
2834 btrfs_item_key_to_cpu(path->nodes[0], &min_key,
2835 path->slots[0]);
2836 goto again;
2837 }
Chris Mason31ff1cd2008-09-11 16:17:57 -04002838 if (ins_nr) {
2839 ret = copy_items(trans, log, dst_path, src,
2840 ins_start_slot,
2841 ins_nr, inode_only);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002842 if (ret) {
2843 err = ret;
2844 goto out_unlock;
2845 }
Chris Mason31ff1cd2008-09-11 16:17:57 -04002846 ins_nr = 0;
2847 }
Chris Mason3a5f1d42008-09-11 15:53:37 -04002848 btrfs_release_path(root, path);
2849
Chris Masone02119d2008-09-05 16:13:11 -04002850 if (min_key.offset < (u64)-1)
2851 min_key.offset++;
2852 else if (min_key.type < (u8)-1)
2853 min_key.type++;
2854 else if (min_key.objectid < (u64)-1)
2855 min_key.objectid++;
2856 else
2857 break;
2858 }
Chris Mason31ff1cd2008-09-11 16:17:57 -04002859 if (ins_nr) {
2860 ret = copy_items(trans, log, dst_path, src,
2861 ins_start_slot,
2862 ins_nr, inode_only);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002863 if (ret) {
2864 err = ret;
2865 goto out_unlock;
2866 }
Chris Mason31ff1cd2008-09-11 16:17:57 -04002867 ins_nr = 0;
2868 }
2869 WARN_ON(ins_nr);
Chris Mason9623f9a2008-09-11 17:42:42 -04002870 if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->i_mode)) {
Chris Masone02119d2008-09-05 16:13:11 -04002871 btrfs_release_path(root, path);
2872 btrfs_release_path(log, dst_path);
2873 ret = log_directory_changes(trans, root, inode, path, dst_path);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002874 if (ret) {
2875 err = ret;
2876 goto out_unlock;
2877 }
Chris Masone02119d2008-09-05 16:13:11 -04002878 }
Chris Mason3a5f1d42008-09-11 15:53:37 -04002879 BTRFS_I(inode)->logged_trans = trans->transid;
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002880out_unlock:
Chris Masone02119d2008-09-05 16:13:11 -04002881 mutex_unlock(&BTRFS_I(inode)->log_mutex);
2882
2883 btrfs_free_path(path);
2884 btrfs_free_path(dst_path);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04002885 return err;
Chris Masone02119d2008-09-05 16:13:11 -04002886}
2887
Chris Mason12fcfd22009-03-24 10:24:20 -04002888/*
2889 * follow the dentry parent pointers up the chain and see if any
2890 * of the directories in it require a full commit before they can
2891 * be logged. Returns zero if nothing special needs to be done or 1 if
2892 * a full commit is required.
2893 */
2894static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
2895 struct inode *inode,
2896 struct dentry *parent,
2897 struct super_block *sb,
2898 u64 last_committed)
Chris Masone02119d2008-09-05 16:13:11 -04002899{
Chris Mason12fcfd22009-03-24 10:24:20 -04002900 int ret = 0;
2901 struct btrfs_root *root;
Josef Bacik6a912212010-11-20 09:48:00 +00002902 struct dentry *old_parent = NULL;
Chris Masone02119d2008-09-05 16:13:11 -04002903
Chris Masonaf4176b2009-03-24 10:24:31 -04002904 /*
2905 * for regular files, if its inode is already on disk, we don't
2906 * have to worry about the parents at all. This is because
2907 * we can use the last_unlink_trans field to record renames
2908 * and other fun in this file.
2909 */
2910 if (S_ISREG(inode->i_mode) &&
2911 BTRFS_I(inode)->generation <= last_committed &&
2912 BTRFS_I(inode)->last_unlink_trans <= last_committed)
2913 goto out;
2914
Chris Mason12fcfd22009-03-24 10:24:20 -04002915 if (!S_ISDIR(inode->i_mode)) {
2916 if (!parent || !parent->d_inode || sb != parent->d_inode->i_sb)
2917 goto out;
2918 inode = parent->d_inode;
2919 }
2920
2921 while (1) {
2922 BTRFS_I(inode)->logged_trans = trans->transid;
2923 smp_mb();
2924
2925 if (BTRFS_I(inode)->last_unlink_trans > last_committed) {
2926 root = BTRFS_I(inode)->root;
2927
2928 /*
2929 * make sure any commits to the log are forced
2930 * to be full commits
2931 */
2932 root->fs_info->last_trans_log_full_commit =
2933 trans->transid;
2934 ret = 1;
2935 break;
2936 }
2937
2938 if (!parent || !parent->d_inode || sb != parent->d_inode->i_sb)
2939 break;
2940
Yan, Zheng76dda932009-09-21 16:00:26 -04002941 if (IS_ROOT(parent))
Chris Mason12fcfd22009-03-24 10:24:20 -04002942 break;
2943
Josef Bacik6a912212010-11-20 09:48:00 +00002944 parent = dget_parent(parent);
2945 dput(old_parent);
2946 old_parent = parent;
Chris Mason12fcfd22009-03-24 10:24:20 -04002947 inode = parent->d_inode;
2948
2949 }
Josef Bacik6a912212010-11-20 09:48:00 +00002950 dput(old_parent);
Chris Mason12fcfd22009-03-24 10:24:20 -04002951out:
Chris Masone02119d2008-09-05 16:13:11 -04002952 return ret;
2953}
2954
Chris Mason257c62e2009-10-13 13:21:08 -04002955static int inode_in_log(struct btrfs_trans_handle *trans,
2956 struct inode *inode)
2957{
2958 struct btrfs_root *root = BTRFS_I(inode)->root;
2959 int ret = 0;
2960
2961 mutex_lock(&root->log_mutex);
2962 if (BTRFS_I(inode)->logged_trans == trans->transid &&
2963 BTRFS_I(inode)->last_sub_trans <= root->last_log_commit)
2964 ret = 1;
2965 mutex_unlock(&root->log_mutex);
2966 return ret;
2967}
2968
2969
Chris Masone02119d2008-09-05 16:13:11 -04002970/*
2971 * helper function around btrfs_log_inode to make sure newly created
2972 * parent directories also end up in the log. A minimal inode and backref
2973 * only logging is done of any parent directories that are older than
2974 * the last committed transaction
2975 */
Chris Mason12fcfd22009-03-24 10:24:20 -04002976int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
2977 struct btrfs_root *root, struct inode *inode,
2978 struct dentry *parent, int exists_only)
Chris Masone02119d2008-09-05 16:13:11 -04002979{
Chris Mason12fcfd22009-03-24 10:24:20 -04002980 int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL;
Chris Masone02119d2008-09-05 16:13:11 -04002981 struct super_block *sb;
Josef Bacik6a912212010-11-20 09:48:00 +00002982 struct dentry *old_parent = NULL;
Chris Mason12fcfd22009-03-24 10:24:20 -04002983 int ret = 0;
2984 u64 last_committed = root->fs_info->last_trans_committed;
2985
2986 sb = inode->i_sb;
2987
Sage Weil3a5e1402009-04-02 16:49:40 -04002988 if (btrfs_test_opt(root, NOTREELOG)) {
2989 ret = 1;
2990 goto end_no_trans;
2991 }
2992
Chris Mason12fcfd22009-03-24 10:24:20 -04002993 if (root->fs_info->last_trans_log_full_commit >
2994 root->fs_info->last_trans_committed) {
2995 ret = 1;
2996 goto end_no_trans;
2997 }
2998
Yan, Zheng76dda932009-09-21 16:00:26 -04002999 if (root != BTRFS_I(inode)->root ||
3000 btrfs_root_refs(&root->root_item) == 0) {
3001 ret = 1;
3002 goto end_no_trans;
3003 }
3004
Chris Mason12fcfd22009-03-24 10:24:20 -04003005 ret = check_parent_dirs_for_sync(trans, inode, parent,
3006 sb, last_committed);
3007 if (ret)
3008 goto end_no_trans;
Chris Masone02119d2008-09-05 16:13:11 -04003009
Chris Mason257c62e2009-10-13 13:21:08 -04003010 if (inode_in_log(trans, inode)) {
3011 ret = BTRFS_NO_LOG_SYNC;
3012 goto end_no_trans;
3013 }
3014
Yan, Zheng4a500fd2010-05-16 10:49:59 -04003015 ret = start_log_trans(trans, root);
3016 if (ret)
3017 goto end_trans;
Chris Mason12fcfd22009-03-24 10:24:20 -04003018
3019 ret = btrfs_log_inode(trans, root, inode, inode_only);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04003020 if (ret)
3021 goto end_trans;
Chris Mason12fcfd22009-03-24 10:24:20 -04003022
Chris Masonaf4176b2009-03-24 10:24:31 -04003023 /*
3024 * for regular files, if its inode is already on disk, we don't
3025 * have to worry about the parents at all. This is because
3026 * we can use the last_unlink_trans field to record renames
3027 * and other fun in this file.
3028 */
3029 if (S_ISREG(inode->i_mode) &&
3030 BTRFS_I(inode)->generation <= last_committed &&
Yan, Zheng4a500fd2010-05-16 10:49:59 -04003031 BTRFS_I(inode)->last_unlink_trans <= last_committed) {
3032 ret = 0;
3033 goto end_trans;
3034 }
Chris Masonaf4176b2009-03-24 10:24:31 -04003035
3036 inode_only = LOG_INODE_EXISTS;
Chris Masond3977122009-01-05 21:25:51 -05003037 while (1) {
Chris Mason12fcfd22009-03-24 10:24:20 -04003038 if (!parent || !parent->d_inode || sb != parent->d_inode->i_sb)
Chris Masone02119d2008-09-05 16:13:11 -04003039 break;
3040
Chris Mason12fcfd22009-03-24 10:24:20 -04003041 inode = parent->d_inode;
Yan, Zheng76dda932009-09-21 16:00:26 -04003042 if (root != BTRFS_I(inode)->root)
3043 break;
3044
Chris Mason12fcfd22009-03-24 10:24:20 -04003045 if (BTRFS_I(inode)->generation >
3046 root->fs_info->last_trans_committed) {
3047 ret = btrfs_log_inode(trans, root, inode, inode_only);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04003048 if (ret)
3049 goto end_trans;
Chris Mason12fcfd22009-03-24 10:24:20 -04003050 }
Yan, Zheng76dda932009-09-21 16:00:26 -04003051 if (IS_ROOT(parent))
Chris Masone02119d2008-09-05 16:13:11 -04003052 break;
Chris Mason12fcfd22009-03-24 10:24:20 -04003053
Josef Bacik6a912212010-11-20 09:48:00 +00003054 parent = dget_parent(parent);
3055 dput(old_parent);
3056 old_parent = parent;
Chris Masone02119d2008-09-05 16:13:11 -04003057 }
Chris Mason12fcfd22009-03-24 10:24:20 -04003058 ret = 0;
Yan, Zheng4a500fd2010-05-16 10:49:59 -04003059end_trans:
Josef Bacik6a912212010-11-20 09:48:00 +00003060 dput(old_parent);
Yan, Zheng4a500fd2010-05-16 10:49:59 -04003061 if (ret < 0) {
3062 BUG_ON(ret != -ENOSPC);
3063 root->fs_info->last_trans_log_full_commit = trans->transid;
3064 ret = 1;
3065 }
Chris Mason12fcfd22009-03-24 10:24:20 -04003066 btrfs_end_log_trans(root);
3067end_no_trans:
3068 return ret;
Chris Masone02119d2008-09-05 16:13:11 -04003069}
3070
3071/*
3072 * it is not safe to log dentry if the chunk root has added new
3073 * chunks. This returns 0 if the dentry was logged, and 1 otherwise.
3074 * If this returns 1, you must commit the transaction to safely get your
3075 * data on disk.
3076 */
3077int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
3078 struct btrfs_root *root, struct dentry *dentry)
3079{
Josef Bacik6a912212010-11-20 09:48:00 +00003080 struct dentry *parent = dget_parent(dentry);
3081 int ret;
3082
3083 ret = btrfs_log_inode_parent(trans, root, dentry->d_inode, parent, 0);
3084 dput(parent);
3085
3086 return ret;
Chris Masone02119d2008-09-05 16:13:11 -04003087}
3088
3089/*
3090 * should be called during mount to recover any replay any log trees
3091 * from the FS
3092 */
3093int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
3094{
3095 int ret;
3096 struct btrfs_path *path;
3097 struct btrfs_trans_handle *trans;
3098 struct btrfs_key key;
3099 struct btrfs_key found_key;
3100 struct btrfs_key tmp_key;
3101 struct btrfs_root *log;
3102 struct btrfs_fs_info *fs_info = log_root_tree->fs_info;
3103 struct walk_control wc = {
3104 .process_func = process_one_buffer,
3105 .stage = 0,
3106 };
3107
Chris Masone02119d2008-09-05 16:13:11 -04003108 path = btrfs_alloc_path();
Tsutomu Itohdb5b4932011-03-23 08:14:16 +00003109 if (!path)
3110 return -ENOMEM;
3111
3112 fs_info->log_root_recovering = 1;
Chris Masone02119d2008-09-05 16:13:11 -04003113
Yan, Zheng4a500fd2010-05-16 10:49:59 -04003114 trans = btrfs_start_transaction(fs_info->tree_root, 0);
Tsutomu Itoh98d5dc12011-01-20 06:19:37 +00003115 BUG_ON(IS_ERR(trans));
Chris Masone02119d2008-09-05 16:13:11 -04003116
3117 wc.trans = trans;
3118 wc.pin = 1;
3119
Tsutomu Itohdb5b4932011-03-23 08:14:16 +00003120 ret = walk_log_tree(trans, log_root_tree, &wc);
3121 BUG_ON(ret);
Chris Masone02119d2008-09-05 16:13:11 -04003122
3123again:
3124 key.objectid = BTRFS_TREE_LOG_OBJECTID;
3125 key.offset = (u64)-1;
3126 btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
3127
Chris Masond3977122009-01-05 21:25:51 -05003128 while (1) {
Chris Masone02119d2008-09-05 16:13:11 -04003129 ret = btrfs_search_slot(NULL, log_root_tree, &key, path, 0, 0);
3130 if (ret < 0)
3131 break;
3132 if (ret > 0) {
3133 if (path->slots[0] == 0)
3134 break;
3135 path->slots[0]--;
3136 }
3137 btrfs_item_key_to_cpu(path->nodes[0], &found_key,
3138 path->slots[0]);
3139 btrfs_release_path(log_root_tree, path);
3140 if (found_key.objectid != BTRFS_TREE_LOG_OBJECTID)
3141 break;
3142
3143 log = btrfs_read_fs_root_no_radix(log_root_tree,
3144 &found_key);
Tsutomu Itohdb5b4932011-03-23 08:14:16 +00003145 BUG_ON(IS_ERR(log));
Chris Masone02119d2008-09-05 16:13:11 -04003146
3147 tmp_key.objectid = found_key.offset;
3148 tmp_key.type = BTRFS_ROOT_ITEM_KEY;
3149 tmp_key.offset = (u64)-1;
3150
3151 wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key);
Chris Masone02119d2008-09-05 16:13:11 -04003152 BUG_ON(!wc.replay_dest);
3153
Yan Zheng07d400a2009-01-06 11:42:00 -05003154 wc.replay_dest->log_root = log;
Yan Zheng5d4f98a2009-06-10 10:45:14 -04003155 btrfs_record_root_in_trans(trans, wc.replay_dest);
Chris Masone02119d2008-09-05 16:13:11 -04003156 ret = walk_log_tree(trans, log, &wc);
3157 BUG_ON(ret);
3158
3159 if (wc.stage == LOG_WALK_REPLAY_ALL) {
3160 ret = fixup_inode_link_counts(trans, wc.replay_dest,
3161 path);
3162 BUG_ON(ret);
3163 }
Chris Masone02119d2008-09-05 16:13:11 -04003164
3165 key.offset = found_key.offset - 1;
Yan Zheng07d400a2009-01-06 11:42:00 -05003166 wc.replay_dest->log_root = NULL;
Chris Masone02119d2008-09-05 16:13:11 -04003167 free_extent_buffer(log->node);
Chris Masonb263c2c2009-06-11 11:24:47 -04003168 free_extent_buffer(log->commit_root);
Chris Masone02119d2008-09-05 16:13:11 -04003169 kfree(log);
3170
3171 if (found_key.offset == 0)
3172 break;
3173 }
3174 btrfs_release_path(log_root_tree, path);
3175
3176 /* step one is to pin it all, step two is to replay just inodes */
3177 if (wc.pin) {
3178 wc.pin = 0;
3179 wc.process_func = replay_one_buffer;
3180 wc.stage = LOG_WALK_REPLAY_INODES;
3181 goto again;
3182 }
3183 /* step three is to replay everything */
3184 if (wc.stage < LOG_WALK_REPLAY_ALL) {
3185 wc.stage++;
3186 goto again;
3187 }
3188
3189 btrfs_free_path(path);
3190
3191 free_extent_buffer(log_root_tree->node);
3192 log_root_tree->log_root = NULL;
3193 fs_info->log_root_recovering = 0;
3194
3195 /* step 4: commit the transaction, which also unpins the blocks */
3196 btrfs_commit_transaction(trans, fs_info->tree_root);
3197
3198 kfree(log_root_tree);
3199 return 0;
3200}
Chris Mason12fcfd22009-03-24 10:24:20 -04003201
3202/*
3203 * there are some corner cases where we want to force a full
3204 * commit instead of allowing a directory to be logged.
3205 *
3206 * They revolve around files there were unlinked from the directory, and
3207 * this function updates the parent directory so that a full commit is
3208 * properly done if it is fsync'd later after the unlinks are done.
3209 */
3210void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
3211 struct inode *dir, struct inode *inode,
3212 int for_rename)
3213{
3214 /*
Chris Masonaf4176b2009-03-24 10:24:31 -04003215 * when we're logging a file, if it hasn't been renamed
3216 * or unlinked, and its inode is fully committed on disk,
3217 * we don't have to worry about walking up the directory chain
3218 * to log its parents.
3219 *
3220 * So, we use the last_unlink_trans field to put this transid
3221 * into the file. When the file is logged we check it and
3222 * don't log the parents if the file is fully on disk.
3223 */
3224 if (S_ISREG(inode->i_mode))
3225 BTRFS_I(inode)->last_unlink_trans = trans->transid;
3226
3227 /*
Chris Mason12fcfd22009-03-24 10:24:20 -04003228 * if this directory was already logged any new
3229 * names for this file/dir will get recorded
3230 */
3231 smp_mb();
3232 if (BTRFS_I(dir)->logged_trans == trans->transid)
3233 return;
3234
3235 /*
3236 * if the inode we're about to unlink was logged,
3237 * the log will be properly updated for any new names
3238 */
3239 if (BTRFS_I(inode)->logged_trans == trans->transid)
3240 return;
3241
3242 /*
3243 * when renaming files across directories, if the directory
3244 * there we're unlinking from gets fsync'd later on, there's
3245 * no way to find the destination directory later and fsync it
3246 * properly. So, we have to be conservative and force commits
3247 * so the new name gets discovered.
3248 */
3249 if (for_rename)
3250 goto record;
3251
3252 /* we can safely do the unlink without any special recording */
3253 return;
3254
3255record:
3256 BTRFS_I(dir)->last_unlink_trans = trans->transid;
3257}
3258
3259/*
3260 * Call this after adding a new name for a file and it will properly
3261 * update the log to reflect the new name.
3262 *
3263 * It will return zero if all goes well, and it will return 1 if a
3264 * full transaction commit is required.
3265 */
3266int btrfs_log_new_name(struct btrfs_trans_handle *trans,
3267 struct inode *inode, struct inode *old_dir,
3268 struct dentry *parent)
3269{
3270 struct btrfs_root * root = BTRFS_I(inode)->root;
3271
3272 /*
Chris Masonaf4176b2009-03-24 10:24:31 -04003273 * this will force the logging code to walk the dentry chain
3274 * up for the file
3275 */
3276 if (S_ISREG(inode->i_mode))
3277 BTRFS_I(inode)->last_unlink_trans = trans->transid;
3278
3279 /*
Chris Mason12fcfd22009-03-24 10:24:20 -04003280 * if this inode hasn't been logged and directory we're renaming it
3281 * from hasn't been logged, we don't need to log it
3282 */
3283 if (BTRFS_I(inode)->logged_trans <=
3284 root->fs_info->last_trans_committed &&
3285 (!old_dir || BTRFS_I(old_dir)->logged_trans <=
3286 root->fs_info->last_trans_committed))
3287 return 0;
3288
3289 return btrfs_log_inode_parent(trans, root, inode, parent, 1);
3290}
3291