]>
Commit | Line | Data |
---|---|---|
e02119d5 CM |
1 | /* |
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> | |
5a0e3ad6 | 20 | #include <linux/slab.h> |
c6adc9cc | 21 | #include <linux/blkdev.h> |
5dc562c5 | 22 | #include <linux/list_sort.h> |
e02119d5 CM |
23 | #include "ctree.h" |
24 | #include "transaction.h" | |
25 | #include "disk-io.h" | |
26 | #include "locking.h" | |
27 | #include "print-tree.h" | |
f186373f | 28 | #include "backref.h" |
b2950863 | 29 | #include "tree-log.h" |
f186373f | 30 | #include "hash.h" |
e02119d5 CM |
31 | |
32 | /* magic values for the inode_only field in btrfs_log_inode: | |
33 | * | |
34 | * LOG_INODE_ALL means to log everything | |
35 | * LOG_INODE_EXISTS means to log just enough to recreate the inode | |
36 | * during log replay | |
37 | */ | |
38 | #define LOG_INODE_ALL 0 | |
39 | #define LOG_INODE_EXISTS 1 | |
40 | ||
12fcfd22 CM |
41 | /* |
42 | * directory trouble cases | |
43 | * | |
44 | * 1) on rename or unlink, if the inode being unlinked isn't in the fsync | |
45 | * log, we must force a full commit before doing an fsync of the directory | |
46 | * where the unlink was done. | |
47 | * ---> record transid of last unlink/rename per directory | |
48 | * | |
49 | * mkdir foo/some_dir | |
50 | * normal commit | |
51 | * rename foo/some_dir foo2/some_dir | |
52 | * mkdir foo/some_dir | |
53 | * fsync foo/some_dir/some_file | |
54 | * | |
55 | * The fsync above will unlink the original some_dir without recording | |
56 | * it in its new location (foo2). After a crash, some_dir will be gone | |
57 | * unless the fsync of some_file forces a full commit | |
58 | * | |
59 | * 2) we must log any new names for any file or dir that is in the fsync | |
60 | * log. ---> check inode while renaming/linking. | |
61 | * | |
62 | * 2a) we must log any new names for any file or dir during rename | |
63 | * when the directory they are being removed from was logged. | |
64 | * ---> check inode and old parent dir during rename | |
65 | * | |
66 | * 2a is actually the more important variant. With the extra logging | |
67 | * a crash might unlink the old name without recreating the new one | |
68 | * | |
69 | * 3) after a crash, we must go through any directories with a link count | |
70 | * of zero and redo the rm -rf | |
71 | * | |
72 | * mkdir f1/foo | |
73 | * normal commit | |
74 | * rm -rf f1/foo | |
75 | * fsync(f1) | |
76 | * | |
77 | * The directory f1 was fully removed from the FS, but fsync was never | |
78 | * called on f1, only its parent dir. After a crash the rm -rf must | |
79 | * be replayed. This must be able to recurse down the entire | |
80 | * directory tree. The inode link count fixup code takes care of the | |
81 | * ugly details. | |
82 | */ | |
83 | ||
e02119d5 CM |
84 | /* |
85 | * stages for the tree walking. The first | |
86 | * stage (0) is to only pin down the blocks we find | |
87 | * the second stage (1) is to make sure that all the inodes | |
88 | * we find in the log are created in the subvolume. | |
89 | * | |
90 | * The last stage is to deal with directories and links and extents | |
91 | * and all the other fun semantics | |
92 | */ | |
93 | #define LOG_WALK_PIN_ONLY 0 | |
94 | #define LOG_WALK_REPLAY_INODES 1 | |
dd8e7217 JB |
95 | #define LOG_WALK_REPLAY_DIR_INDEX 2 |
96 | #define LOG_WALK_REPLAY_ALL 3 | |
e02119d5 | 97 | |
12fcfd22 | 98 | static int btrfs_log_inode(struct btrfs_trans_handle *trans, |
e02119d5 CM |
99 | struct btrfs_root *root, struct inode *inode, |
100 | int inode_only); | |
ec051c0f YZ |
101 | static int link_to_fixup_dir(struct btrfs_trans_handle *trans, |
102 | struct btrfs_root *root, | |
103 | struct btrfs_path *path, u64 objectid); | |
12fcfd22 CM |
104 | static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans, |
105 | struct btrfs_root *root, | |
106 | struct btrfs_root *log, | |
107 | struct btrfs_path *path, | |
108 | u64 dirid, int del_all); | |
e02119d5 CM |
109 | |
110 | /* | |
111 | * tree logging is a special write ahead log used to make sure that | |
112 | * fsyncs and O_SYNCs can happen without doing full tree commits. | |
113 | * | |
114 | * Full tree commits are expensive because they require commonly | |
115 | * modified blocks to be recowed, creating many dirty pages in the | |
116 | * extent tree an 4x-6x higher write load than ext3. | |
117 | * | |
118 | * Instead of doing a tree commit on every fsync, we use the | |
119 | * key ranges and transaction ids to find items for a given file or directory | |
120 | * that have changed in this transaction. Those items are copied into | |
121 | * a special tree (one per subvolume root), that tree is written to disk | |
122 | * and then the fsync is considered complete. | |
123 | * | |
124 | * After a crash, items are copied out of the log-tree back into the | |
125 | * subvolume tree. Any file data extents found are recorded in the extent | |
126 | * allocation tree, and the log-tree freed. | |
127 | * | |
128 | * The log tree is read three times, once to pin down all the extents it is | |
129 | * using in ram and once, once to create all the inodes logged in the tree | |
130 | * and once to do all the other items. | |
131 | */ | |
132 | ||
e02119d5 CM |
133 | /* |
134 | * start a sub transaction and setup the log tree | |
135 | * this increments the log tree writer count to make the people | |
136 | * syncing the tree wait for us to finish | |
137 | */ | |
138 | static int start_log_trans(struct btrfs_trans_handle *trans, | |
8b050d35 MX |
139 | struct btrfs_root *root, |
140 | struct btrfs_log_ctx *ctx) | |
e02119d5 | 141 | { |
8b050d35 | 142 | int index; |
e02119d5 | 143 | int ret; |
7237f183 YZ |
144 | |
145 | mutex_lock(&root->log_mutex); | |
146 | if (root->log_root) { | |
ff782e0a JB |
147 | if (!root->log_start_pid) { |
148 | root->log_start_pid = current->pid; | |
149 | root->log_multiple_pids = false; | |
150 | } else if (root->log_start_pid != current->pid) { | |
151 | root->log_multiple_pids = true; | |
152 | } | |
153 | ||
2ecb7923 | 154 | atomic_inc(&root->log_batch); |
7237f183 | 155 | atomic_inc(&root->log_writers); |
8b050d35 MX |
156 | if (ctx) { |
157 | index = root->log_transid % 2; | |
158 | list_add_tail(&ctx->list, &root->log_ctxs[index]); | |
d1433deb | 159 | ctx->log_transid = root->log_transid; |
8b050d35 | 160 | } |
7237f183 YZ |
161 | mutex_unlock(&root->log_mutex); |
162 | return 0; | |
163 | } | |
e87ac136 MX |
164 | |
165 | ret = 0; | |
e02119d5 | 166 | mutex_lock(&root->fs_info->tree_log_mutex); |
e87ac136 | 167 | if (!root->fs_info->log_root_tree) |
e02119d5 | 168 | ret = btrfs_init_log_root_tree(trans, root->fs_info); |
e87ac136 MX |
169 | mutex_unlock(&root->fs_info->tree_log_mutex); |
170 | if (ret) | |
171 | goto out; | |
172 | ||
173 | if (!root->log_root) { | |
e02119d5 | 174 | ret = btrfs_add_log_tree(trans, root); |
4a500fd1 | 175 | if (ret) |
e87ac136 | 176 | goto out; |
e02119d5 | 177 | } |
e87ac136 MX |
178 | root->log_multiple_pids = false; |
179 | root->log_start_pid = current->pid; | |
2ecb7923 | 180 | atomic_inc(&root->log_batch); |
7237f183 | 181 | atomic_inc(&root->log_writers); |
8b050d35 MX |
182 | if (ctx) { |
183 | index = root->log_transid % 2; | |
184 | list_add_tail(&ctx->list, &root->log_ctxs[index]); | |
d1433deb | 185 | ctx->log_transid = root->log_transid; |
8b050d35 | 186 | } |
e87ac136 | 187 | out: |
7237f183 | 188 | mutex_unlock(&root->log_mutex); |
e87ac136 | 189 | return ret; |
e02119d5 CM |
190 | } |
191 | ||
192 | /* | |
193 | * returns 0 if there was a log transaction running and we were able | |
194 | * to join, or returns -ENOENT if there were not transactions | |
195 | * in progress | |
196 | */ | |
197 | static int join_running_log_trans(struct btrfs_root *root) | |
198 | { | |
199 | int ret = -ENOENT; | |
200 | ||
201 | smp_mb(); | |
202 | if (!root->log_root) | |
203 | return -ENOENT; | |
204 | ||
7237f183 | 205 | mutex_lock(&root->log_mutex); |
e02119d5 CM |
206 | if (root->log_root) { |
207 | ret = 0; | |
7237f183 | 208 | atomic_inc(&root->log_writers); |
e02119d5 | 209 | } |
7237f183 | 210 | mutex_unlock(&root->log_mutex); |
e02119d5 CM |
211 | return ret; |
212 | } | |
213 | ||
12fcfd22 CM |
214 | /* |
215 | * This either makes the current running log transaction wait | |
216 | * until you call btrfs_end_log_trans() or it makes any future | |
217 | * log transactions wait until you call btrfs_end_log_trans() | |
218 | */ | |
219 | int btrfs_pin_log_trans(struct btrfs_root *root) | |
220 | { | |
221 | int ret = -ENOENT; | |
222 | ||
223 | mutex_lock(&root->log_mutex); | |
224 | atomic_inc(&root->log_writers); | |
225 | mutex_unlock(&root->log_mutex); | |
226 | return ret; | |
227 | } | |
228 | ||
e02119d5 CM |
229 | /* |
230 | * indicate we're done making changes to the log tree | |
231 | * and wake up anyone waiting to do a sync | |
232 | */ | |
143bede5 | 233 | void btrfs_end_log_trans(struct btrfs_root *root) |
e02119d5 | 234 | { |
7237f183 YZ |
235 | if (atomic_dec_and_test(&root->log_writers)) { |
236 | smp_mb(); | |
237 | if (waitqueue_active(&root->log_writer_wait)) | |
238 | wake_up(&root->log_writer_wait); | |
239 | } | |
e02119d5 CM |
240 | } |
241 | ||
242 | ||
243 | /* | |
244 | * the walk control struct is used to pass state down the chain when | |
245 | * processing the log tree. The stage field tells us which part | |
246 | * of the log tree processing we are currently doing. The others | |
247 | * are state fields used for that specific part | |
248 | */ | |
249 | struct walk_control { | |
250 | /* should we free the extent on disk when done? This is used | |
251 | * at transaction commit time while freeing a log tree | |
252 | */ | |
253 | int free; | |
254 | ||
255 | /* should we write out the extent buffer? This is used | |
256 | * while flushing the log tree to disk during a sync | |
257 | */ | |
258 | int write; | |
259 | ||
260 | /* should we wait for the extent buffer io to finish? Also used | |
261 | * while flushing the log tree to disk for a sync | |
262 | */ | |
263 | int wait; | |
264 | ||
265 | /* pin only walk, we record which extents on disk belong to the | |
266 | * log trees | |
267 | */ | |
268 | int pin; | |
269 | ||
270 | /* what stage of the replay code we're currently in */ | |
271 | int stage; | |
272 | ||
273 | /* the root we are currently replaying */ | |
274 | struct btrfs_root *replay_dest; | |
275 | ||
276 | /* the trans handle for the current replay */ | |
277 | struct btrfs_trans_handle *trans; | |
278 | ||
279 | /* the function that gets used to process blocks we find in the | |
280 | * tree. Note the extent_buffer might not be up to date when it is | |
281 | * passed in, and it must be checked or read if you need the data | |
282 | * inside it | |
283 | */ | |
284 | int (*process_func)(struct btrfs_root *log, struct extent_buffer *eb, | |
285 | struct walk_control *wc, u64 gen); | |
286 | }; | |
287 | ||
288 | /* | |
289 | * process_func used to pin down extents, write them or wait on them | |
290 | */ | |
291 | static int process_one_buffer(struct btrfs_root *log, | |
292 | struct extent_buffer *eb, | |
293 | struct walk_control *wc, u64 gen) | |
294 | { | |
b50c6e25 JB |
295 | int ret = 0; |
296 | ||
8c2a1a30 JB |
297 | /* |
298 | * If this fs is mixed then we need to be able to process the leaves to | |
299 | * pin down any logged extents, so we have to read the block. | |
300 | */ | |
301 | if (btrfs_fs_incompat(log->fs_info, MIXED_GROUPS)) { | |
302 | ret = btrfs_read_buffer(eb, gen); | |
303 | if (ret) | |
304 | return ret; | |
305 | } | |
306 | ||
04018de5 | 307 | if (wc->pin) |
b50c6e25 JB |
308 | ret = btrfs_pin_extent_for_log_replay(log->fs_info->extent_root, |
309 | eb->start, eb->len); | |
e02119d5 | 310 | |
b50c6e25 | 311 | if (!ret && btrfs_buffer_uptodate(eb, gen, 0)) { |
8c2a1a30 JB |
312 | if (wc->pin && btrfs_header_level(eb) == 0) |
313 | ret = btrfs_exclude_logged_extents(log, eb); | |
e02119d5 CM |
314 | if (wc->write) |
315 | btrfs_write_tree_block(eb); | |
316 | if (wc->wait) | |
317 | btrfs_wait_tree_block_writeback(eb); | |
318 | } | |
b50c6e25 | 319 | return ret; |
e02119d5 CM |
320 | } |
321 | ||
322 | /* | |
323 | * Item overwrite used by replay and tree logging. eb, slot and key all refer | |
324 | * to the src data we are copying out. | |
325 | * | |
326 | * root is the tree we are copying into, and path is a scratch | |
327 | * path for use in this function (it should be released on entry and | |
328 | * will be released on exit). | |
329 | * | |
330 | * If the key is already in the destination tree the existing item is | |
331 | * overwritten. If the existing item isn't big enough, it is extended. | |
332 | * If it is too large, it is truncated. | |
333 | * | |
334 | * If the key isn't in the destination yet, a new item is inserted. | |
335 | */ | |
336 | static noinline int overwrite_item(struct btrfs_trans_handle *trans, | |
337 | struct btrfs_root *root, | |
338 | struct btrfs_path *path, | |
339 | struct extent_buffer *eb, int slot, | |
340 | struct btrfs_key *key) | |
341 | { | |
342 | int ret; | |
343 | u32 item_size; | |
344 | u64 saved_i_size = 0; | |
345 | int save_old_i_size = 0; | |
346 | unsigned long src_ptr; | |
347 | unsigned long dst_ptr; | |
348 | int overwrite_root = 0; | |
4bc4bee4 | 349 | bool inode_item = key->type == BTRFS_INODE_ITEM_KEY; |
e02119d5 CM |
350 | |
351 | if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) | |
352 | overwrite_root = 1; | |
353 | ||
354 | item_size = btrfs_item_size_nr(eb, slot); | |
355 | src_ptr = btrfs_item_ptr_offset(eb, slot); | |
356 | ||
357 | /* look for the key in the destination tree */ | |
358 | ret = btrfs_search_slot(NULL, root, key, path, 0, 0); | |
4bc4bee4 JB |
359 | if (ret < 0) |
360 | return ret; | |
361 | ||
e02119d5 CM |
362 | if (ret == 0) { |
363 | char *src_copy; | |
364 | char *dst_copy; | |
365 | u32 dst_size = btrfs_item_size_nr(path->nodes[0], | |
366 | path->slots[0]); | |
367 | if (dst_size != item_size) | |
368 | goto insert; | |
369 | ||
370 | if (item_size == 0) { | |
b3b4aa74 | 371 | btrfs_release_path(path); |
e02119d5 CM |
372 | return 0; |
373 | } | |
374 | dst_copy = kmalloc(item_size, GFP_NOFS); | |
375 | src_copy = kmalloc(item_size, GFP_NOFS); | |
2a29edc6 | 376 | if (!dst_copy || !src_copy) { |
b3b4aa74 | 377 | btrfs_release_path(path); |
2a29edc6 | 378 | kfree(dst_copy); |
379 | kfree(src_copy); | |
380 | return -ENOMEM; | |
381 | } | |
e02119d5 CM |
382 | |
383 | read_extent_buffer(eb, src_copy, src_ptr, item_size); | |
384 | ||
385 | dst_ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); | |
386 | read_extent_buffer(path->nodes[0], dst_copy, dst_ptr, | |
387 | item_size); | |
388 | ret = memcmp(dst_copy, src_copy, item_size); | |
389 | ||
390 | kfree(dst_copy); | |
391 | kfree(src_copy); | |
392 | /* | |
393 | * they have the same contents, just return, this saves | |
394 | * us from cowing blocks in the destination tree and doing | |
395 | * extra writes that may not have been done by a previous | |
396 | * sync | |
397 | */ | |
398 | if (ret == 0) { | |
b3b4aa74 | 399 | btrfs_release_path(path); |
e02119d5 CM |
400 | return 0; |
401 | } | |
402 | ||
4bc4bee4 JB |
403 | /* |
404 | * We need to load the old nbytes into the inode so when we | |
405 | * replay the extents we've logged we get the right nbytes. | |
406 | */ | |
407 | if (inode_item) { | |
408 | struct btrfs_inode_item *item; | |
409 | u64 nbytes; | |
d555438b | 410 | u32 mode; |
4bc4bee4 JB |
411 | |
412 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
413 | struct btrfs_inode_item); | |
414 | nbytes = btrfs_inode_nbytes(path->nodes[0], item); | |
415 | item = btrfs_item_ptr(eb, slot, | |
416 | struct btrfs_inode_item); | |
417 | btrfs_set_inode_nbytes(eb, item, nbytes); | |
d555438b JB |
418 | |
419 | /* | |
420 | * If this is a directory we need to reset the i_size to | |
421 | * 0 so that we can set it up properly when replaying | |
422 | * the rest of the items in this log. | |
423 | */ | |
424 | mode = btrfs_inode_mode(eb, item); | |
425 | if (S_ISDIR(mode)) | |
426 | btrfs_set_inode_size(eb, item, 0); | |
4bc4bee4 JB |
427 | } |
428 | } else if (inode_item) { | |
429 | struct btrfs_inode_item *item; | |
d555438b | 430 | u32 mode; |
4bc4bee4 JB |
431 | |
432 | /* | |
433 | * New inode, set nbytes to 0 so that the nbytes comes out | |
434 | * properly when we replay the extents. | |
435 | */ | |
436 | item = btrfs_item_ptr(eb, slot, struct btrfs_inode_item); | |
437 | btrfs_set_inode_nbytes(eb, item, 0); | |
d555438b JB |
438 | |
439 | /* | |
440 | * If this is a directory we need to reset the i_size to 0 so | |
441 | * that we can set it up properly when replaying the rest of | |
442 | * the items in this log. | |
443 | */ | |
444 | mode = btrfs_inode_mode(eb, item); | |
445 | if (S_ISDIR(mode)) | |
446 | btrfs_set_inode_size(eb, item, 0); | |
e02119d5 CM |
447 | } |
448 | insert: | |
b3b4aa74 | 449 | btrfs_release_path(path); |
e02119d5 CM |
450 | /* try to insert the key into the destination tree */ |
451 | ret = btrfs_insert_empty_item(trans, root, path, | |
452 | key, item_size); | |
453 | ||
454 | /* make sure any existing item is the correct size */ | |
455 | if (ret == -EEXIST) { | |
456 | u32 found_size; | |
457 | found_size = btrfs_item_size_nr(path->nodes[0], | |
458 | path->slots[0]); | |
143bede5 | 459 | if (found_size > item_size) |
afe5fea7 | 460 | btrfs_truncate_item(root, path, item_size, 1); |
143bede5 | 461 | else if (found_size < item_size) |
4b90c680 | 462 | btrfs_extend_item(root, path, |
143bede5 | 463 | item_size - found_size); |
e02119d5 | 464 | } else if (ret) { |
4a500fd1 | 465 | return ret; |
e02119d5 CM |
466 | } |
467 | dst_ptr = btrfs_item_ptr_offset(path->nodes[0], | |
468 | path->slots[0]); | |
469 | ||
470 | /* don't overwrite an existing inode if the generation number | |
471 | * was logged as zero. This is done when the tree logging code | |
472 | * is just logging an inode to make sure it exists after recovery. | |
473 | * | |
474 | * Also, don't overwrite i_size on directories during replay. | |
475 | * log replay inserts and removes directory items based on the | |
476 | * state of the tree found in the subvolume, and i_size is modified | |
477 | * as it goes | |
478 | */ | |
479 | if (key->type == BTRFS_INODE_ITEM_KEY && ret == -EEXIST) { | |
480 | struct btrfs_inode_item *src_item; | |
481 | struct btrfs_inode_item *dst_item; | |
482 | ||
483 | src_item = (struct btrfs_inode_item *)src_ptr; | |
484 | dst_item = (struct btrfs_inode_item *)dst_ptr; | |
485 | ||
486 | if (btrfs_inode_generation(eb, src_item) == 0) | |
487 | goto no_copy; | |
488 | ||
489 | if (overwrite_root && | |
490 | S_ISDIR(btrfs_inode_mode(eb, src_item)) && | |
491 | S_ISDIR(btrfs_inode_mode(path->nodes[0], dst_item))) { | |
492 | save_old_i_size = 1; | |
493 | saved_i_size = btrfs_inode_size(path->nodes[0], | |
494 | dst_item); | |
495 | } | |
496 | } | |
497 | ||
498 | copy_extent_buffer(path->nodes[0], eb, dst_ptr, | |
499 | src_ptr, item_size); | |
500 | ||
501 | if (save_old_i_size) { | |
502 | struct btrfs_inode_item *dst_item; | |
503 | dst_item = (struct btrfs_inode_item *)dst_ptr; | |
504 | btrfs_set_inode_size(path->nodes[0], dst_item, saved_i_size); | |
505 | } | |
506 | ||
507 | /* make sure the generation is filled in */ | |
508 | if (key->type == BTRFS_INODE_ITEM_KEY) { | |
509 | struct btrfs_inode_item *dst_item; | |
510 | dst_item = (struct btrfs_inode_item *)dst_ptr; | |
511 | if (btrfs_inode_generation(path->nodes[0], dst_item) == 0) { | |
512 | btrfs_set_inode_generation(path->nodes[0], dst_item, | |
513 | trans->transid); | |
514 | } | |
515 | } | |
516 | no_copy: | |
517 | btrfs_mark_buffer_dirty(path->nodes[0]); | |
b3b4aa74 | 518 | btrfs_release_path(path); |
e02119d5 CM |
519 | return 0; |
520 | } | |
521 | ||
522 | /* | |
523 | * simple helper to read an inode off the disk from a given root | |
524 | * This can only be called for subvolume roots and not for the log | |
525 | */ | |
526 | static noinline struct inode *read_one_inode(struct btrfs_root *root, | |
527 | u64 objectid) | |
528 | { | |
5d4f98a2 | 529 | struct btrfs_key key; |
e02119d5 | 530 | struct inode *inode; |
e02119d5 | 531 | |
5d4f98a2 YZ |
532 | key.objectid = objectid; |
533 | key.type = BTRFS_INODE_ITEM_KEY; | |
534 | key.offset = 0; | |
73f73415 | 535 | inode = btrfs_iget(root->fs_info->sb, &key, root, NULL); |
5d4f98a2 YZ |
536 | if (IS_ERR(inode)) { |
537 | inode = NULL; | |
538 | } else if (is_bad_inode(inode)) { | |
e02119d5 CM |
539 | iput(inode); |
540 | inode = NULL; | |
541 | } | |
542 | return inode; | |
543 | } | |
544 | ||
545 | /* replays a single extent in 'eb' at 'slot' with 'key' into the | |
546 | * subvolume 'root'. path is released on entry and should be released | |
547 | * on exit. | |
548 | * | |
549 | * extents in the log tree have not been allocated out of the extent | |
550 | * tree yet. So, this completes the allocation, taking a reference | |
551 | * as required if the extent already exists or creating a new extent | |
552 | * if it isn't in the extent allocation tree yet. | |
553 | * | |
554 | * The extent is inserted into the file, dropping any existing extents | |
555 | * from the file that overlap the new one. | |
556 | */ | |
557 | static noinline int replay_one_extent(struct btrfs_trans_handle *trans, | |
558 | struct btrfs_root *root, | |
559 | struct btrfs_path *path, | |
560 | struct extent_buffer *eb, int slot, | |
561 | struct btrfs_key *key) | |
562 | { | |
563 | int found_type; | |
e02119d5 | 564 | u64 extent_end; |
e02119d5 | 565 | u64 start = key->offset; |
4bc4bee4 | 566 | u64 nbytes = 0; |
e02119d5 CM |
567 | struct btrfs_file_extent_item *item; |
568 | struct inode *inode = NULL; | |
569 | unsigned long size; | |
570 | int ret = 0; | |
571 | ||
572 | item = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item); | |
573 | found_type = btrfs_file_extent_type(eb, item); | |
574 | ||
d899e052 | 575 | if (found_type == BTRFS_FILE_EXTENT_REG || |
4bc4bee4 JB |
576 | found_type == BTRFS_FILE_EXTENT_PREALLOC) { |
577 | nbytes = btrfs_file_extent_num_bytes(eb, item); | |
578 | extent_end = start + nbytes; | |
579 | ||
580 | /* | |
581 | * We don't add to the inodes nbytes if we are prealloc or a | |
582 | * hole. | |
583 | */ | |
584 | if (btrfs_file_extent_disk_bytenr(eb, item) == 0) | |
585 | nbytes = 0; | |
586 | } else if (found_type == BTRFS_FILE_EXTENT_INLINE) { | |
514ac8ad | 587 | size = btrfs_file_extent_inline_len(eb, slot, item); |
4bc4bee4 | 588 | nbytes = btrfs_file_extent_ram_bytes(eb, item); |
fda2832f | 589 | extent_end = ALIGN(start + size, root->sectorsize); |
e02119d5 CM |
590 | } else { |
591 | ret = 0; | |
592 | goto out; | |
593 | } | |
594 | ||
595 | inode = read_one_inode(root, key->objectid); | |
596 | if (!inode) { | |
597 | ret = -EIO; | |
598 | goto out; | |
599 | } | |
600 | ||
601 | /* | |
602 | * first check to see if we already have this extent in the | |
603 | * file. This must be done before the btrfs_drop_extents run | |
604 | * so we don't try to drop this extent. | |
605 | */ | |
33345d01 | 606 | ret = btrfs_lookup_file_extent(trans, root, path, btrfs_ino(inode), |
e02119d5 CM |
607 | start, 0); |
608 | ||
d899e052 YZ |
609 | if (ret == 0 && |
610 | (found_type == BTRFS_FILE_EXTENT_REG || | |
611 | found_type == BTRFS_FILE_EXTENT_PREALLOC)) { | |
e02119d5 CM |
612 | struct btrfs_file_extent_item cmp1; |
613 | struct btrfs_file_extent_item cmp2; | |
614 | struct btrfs_file_extent_item *existing; | |
615 | struct extent_buffer *leaf; | |
616 | ||
617 | leaf = path->nodes[0]; | |
618 | existing = btrfs_item_ptr(leaf, path->slots[0], | |
619 | struct btrfs_file_extent_item); | |
620 | ||
621 | read_extent_buffer(eb, &cmp1, (unsigned long)item, | |
622 | sizeof(cmp1)); | |
623 | read_extent_buffer(leaf, &cmp2, (unsigned long)existing, | |
624 | sizeof(cmp2)); | |
625 | ||
626 | /* | |
627 | * we already have a pointer to this exact extent, | |
628 | * we don't have to do anything | |
629 | */ | |
630 | if (memcmp(&cmp1, &cmp2, sizeof(cmp1)) == 0) { | |
b3b4aa74 | 631 | btrfs_release_path(path); |
e02119d5 CM |
632 | goto out; |
633 | } | |
634 | } | |
b3b4aa74 | 635 | btrfs_release_path(path); |
e02119d5 CM |
636 | |
637 | /* drop any overlapping extents */ | |
2671485d | 638 | ret = btrfs_drop_extents(trans, root, inode, start, extent_end, 1); |
3650860b JB |
639 | if (ret) |
640 | goto out; | |
e02119d5 | 641 | |
07d400a6 YZ |
642 | if (found_type == BTRFS_FILE_EXTENT_REG || |
643 | found_type == BTRFS_FILE_EXTENT_PREALLOC) { | |
5d4f98a2 | 644 | u64 offset; |
07d400a6 YZ |
645 | unsigned long dest_offset; |
646 | struct btrfs_key ins; | |
647 | ||
648 | ret = btrfs_insert_empty_item(trans, root, path, key, | |
649 | sizeof(*item)); | |
3650860b JB |
650 | if (ret) |
651 | goto out; | |
07d400a6 YZ |
652 | dest_offset = btrfs_item_ptr_offset(path->nodes[0], |
653 | path->slots[0]); | |
654 | copy_extent_buffer(path->nodes[0], eb, dest_offset, | |
655 | (unsigned long)item, sizeof(*item)); | |
656 | ||
657 | ins.objectid = btrfs_file_extent_disk_bytenr(eb, item); | |
658 | ins.offset = btrfs_file_extent_disk_num_bytes(eb, item); | |
659 | ins.type = BTRFS_EXTENT_ITEM_KEY; | |
5d4f98a2 | 660 | offset = key->offset - btrfs_file_extent_offset(eb, item); |
07d400a6 YZ |
661 | |
662 | if (ins.objectid > 0) { | |
663 | u64 csum_start; | |
664 | u64 csum_end; | |
665 | LIST_HEAD(ordered_sums); | |
666 | /* | |
667 | * is this extent already allocated in the extent | |
668 | * allocation tree? If so, just add a reference | |
669 | */ | |
670 | ret = btrfs_lookup_extent(root, ins.objectid, | |
671 | ins.offset); | |
672 | if (ret == 0) { | |
673 | ret = btrfs_inc_extent_ref(trans, root, | |
674 | ins.objectid, ins.offset, | |
5d4f98a2 | 675 | 0, root->root_key.objectid, |
66d7e7f0 | 676 | key->objectid, offset, 0); |
b50c6e25 JB |
677 | if (ret) |
678 | goto out; | |
07d400a6 YZ |
679 | } else { |
680 | /* | |
681 | * insert the extent pointer in the extent | |
682 | * allocation tree | |
683 | */ | |
5d4f98a2 YZ |
684 | ret = btrfs_alloc_logged_file_extent(trans, |
685 | root, root->root_key.objectid, | |
686 | key->objectid, offset, &ins); | |
b50c6e25 JB |
687 | if (ret) |
688 | goto out; | |
07d400a6 | 689 | } |
b3b4aa74 | 690 | btrfs_release_path(path); |
07d400a6 YZ |
691 | |
692 | if (btrfs_file_extent_compression(eb, item)) { | |
693 | csum_start = ins.objectid; | |
694 | csum_end = csum_start + ins.offset; | |
695 | } else { | |
696 | csum_start = ins.objectid + | |
697 | btrfs_file_extent_offset(eb, item); | |
698 | csum_end = csum_start + | |
699 | btrfs_file_extent_num_bytes(eb, item); | |
700 | } | |
701 | ||
702 | ret = btrfs_lookup_csums_range(root->log_root, | |
703 | csum_start, csum_end - 1, | |
a2de733c | 704 | &ordered_sums, 0); |
3650860b JB |
705 | if (ret) |
706 | goto out; | |
07d400a6 YZ |
707 | while (!list_empty(&ordered_sums)) { |
708 | struct btrfs_ordered_sum *sums; | |
709 | sums = list_entry(ordered_sums.next, | |
710 | struct btrfs_ordered_sum, | |
711 | list); | |
3650860b JB |
712 | if (!ret) |
713 | ret = btrfs_csum_file_blocks(trans, | |
07d400a6 YZ |
714 | root->fs_info->csum_root, |
715 | sums); | |
07d400a6 YZ |
716 | list_del(&sums->list); |
717 | kfree(sums); | |
718 | } | |
3650860b JB |
719 | if (ret) |
720 | goto out; | |
07d400a6 | 721 | } else { |
b3b4aa74 | 722 | btrfs_release_path(path); |
07d400a6 YZ |
723 | } |
724 | } else if (found_type == BTRFS_FILE_EXTENT_INLINE) { | |
725 | /* inline extents are easy, we just overwrite them */ | |
726 | ret = overwrite_item(trans, root, path, eb, slot, key); | |
3650860b JB |
727 | if (ret) |
728 | goto out; | |
07d400a6 | 729 | } |
e02119d5 | 730 | |
4bc4bee4 | 731 | inode_add_bytes(inode, nbytes); |
b9959295 | 732 | ret = btrfs_update_inode(trans, root, inode); |
e02119d5 CM |
733 | out: |
734 | if (inode) | |
735 | iput(inode); | |
736 | return ret; | |
737 | } | |
738 | ||
739 | /* | |
740 | * when cleaning up conflicts between the directory names in the | |
741 | * subvolume, directory names in the log and directory names in the | |
742 | * inode back references, we may have to unlink inodes from directories. | |
743 | * | |
744 | * This is a helper function to do the unlink of a specific directory | |
745 | * item | |
746 | */ | |
747 | static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans, | |
748 | struct btrfs_root *root, | |
749 | struct btrfs_path *path, | |
750 | struct inode *dir, | |
751 | struct btrfs_dir_item *di) | |
752 | { | |
753 | struct inode *inode; | |
754 | char *name; | |
755 | int name_len; | |
756 | struct extent_buffer *leaf; | |
757 | struct btrfs_key location; | |
758 | int ret; | |
759 | ||
760 | leaf = path->nodes[0]; | |
761 | ||
762 | btrfs_dir_item_key_to_cpu(leaf, di, &location); | |
763 | name_len = btrfs_dir_name_len(leaf, di); | |
764 | name = kmalloc(name_len, GFP_NOFS); | |
2a29edc6 | 765 | if (!name) |
766 | return -ENOMEM; | |
767 | ||
e02119d5 | 768 | read_extent_buffer(leaf, name, (unsigned long)(di + 1), name_len); |
b3b4aa74 | 769 | btrfs_release_path(path); |
e02119d5 CM |
770 | |
771 | inode = read_one_inode(root, location.objectid); | |
c00e9493 | 772 | if (!inode) { |
3650860b JB |
773 | ret = -EIO; |
774 | goto out; | |
c00e9493 | 775 | } |
e02119d5 | 776 | |
ec051c0f | 777 | ret = link_to_fixup_dir(trans, root, path, location.objectid); |
3650860b JB |
778 | if (ret) |
779 | goto out; | |
12fcfd22 | 780 | |
e02119d5 | 781 | ret = btrfs_unlink_inode(trans, root, dir, inode, name, name_len); |
3650860b JB |
782 | if (ret) |
783 | goto out; | |
ada9af21 FDBM |
784 | else |
785 | ret = btrfs_run_delayed_items(trans, root); | |
3650860b | 786 | out: |
e02119d5 | 787 | kfree(name); |
e02119d5 CM |
788 | iput(inode); |
789 | return ret; | |
790 | } | |
791 | ||
792 | /* | |
793 | * helper function to see if a given name and sequence number found | |
794 | * in an inode back reference are already in a directory and correctly | |
795 | * point to this inode | |
796 | */ | |
797 | static noinline int inode_in_dir(struct btrfs_root *root, | |
798 | struct btrfs_path *path, | |
799 | u64 dirid, u64 objectid, u64 index, | |
800 | const char *name, int name_len) | |
801 | { | |
802 | struct btrfs_dir_item *di; | |
803 | struct btrfs_key location; | |
804 | int match = 0; | |
805 | ||
806 | di = btrfs_lookup_dir_index_item(NULL, root, path, dirid, | |
807 | index, name, name_len, 0); | |
808 | if (di && !IS_ERR(di)) { | |
809 | btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location); | |
810 | if (location.objectid != objectid) | |
811 | goto out; | |
812 | } else | |
813 | goto out; | |
b3b4aa74 | 814 | btrfs_release_path(path); |
e02119d5 CM |
815 | |
816 | di = btrfs_lookup_dir_item(NULL, root, path, dirid, name, name_len, 0); | |
817 | if (di && !IS_ERR(di)) { | |
818 | btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location); | |
819 | if (location.objectid != objectid) | |
820 | goto out; | |
821 | } else | |
822 | goto out; | |
823 | match = 1; | |
824 | out: | |
b3b4aa74 | 825 | btrfs_release_path(path); |
e02119d5 CM |
826 | return match; |
827 | } | |
828 | ||
829 | /* | |
830 | * helper function to check a log tree for a named back reference in | |
831 | * an inode. This is used to decide if a back reference that is | |
832 | * found in the subvolume conflicts with what we find in the log. | |
833 | * | |
834 | * inode backreferences may have multiple refs in a single item, | |
835 | * during replay we process one reference at a time, and we don't | |
836 | * want to delete valid links to a file from the subvolume if that | |
837 | * link is also in the log. | |
838 | */ | |
839 | static noinline int backref_in_log(struct btrfs_root *log, | |
840 | struct btrfs_key *key, | |
f186373f | 841 | u64 ref_objectid, |
e02119d5 CM |
842 | char *name, int namelen) |
843 | { | |
844 | struct btrfs_path *path; | |
845 | struct btrfs_inode_ref *ref; | |
846 | unsigned long ptr; | |
847 | unsigned long ptr_end; | |
848 | unsigned long name_ptr; | |
849 | int found_name_len; | |
850 | int item_size; | |
851 | int ret; | |
852 | int match = 0; | |
853 | ||
854 | path = btrfs_alloc_path(); | |
2a29edc6 | 855 | if (!path) |
856 | return -ENOMEM; | |
857 | ||
e02119d5 CM |
858 | ret = btrfs_search_slot(NULL, log, key, path, 0, 0); |
859 | if (ret != 0) | |
860 | goto out; | |
861 | ||
e02119d5 | 862 | ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); |
f186373f MF |
863 | |
864 | if (key->type == BTRFS_INODE_EXTREF_KEY) { | |
865 | if (btrfs_find_name_in_ext_backref(path, ref_objectid, | |
866 | name, namelen, NULL)) | |
867 | match = 1; | |
868 | ||
869 | goto out; | |
870 | } | |
871 | ||
872 | item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]); | |
e02119d5 CM |
873 | ptr_end = ptr + item_size; |
874 | while (ptr < ptr_end) { | |
875 | ref = (struct btrfs_inode_ref *)ptr; | |
876 | found_name_len = btrfs_inode_ref_name_len(path->nodes[0], ref); | |
877 | if (found_name_len == namelen) { | |
878 | name_ptr = (unsigned long)(ref + 1); | |
879 | ret = memcmp_extent_buffer(path->nodes[0], name, | |
880 | name_ptr, namelen); | |
881 | if (ret == 0) { | |
882 | match = 1; | |
883 | goto out; | |
884 | } | |
885 | } | |
886 | ptr = (unsigned long)(ref + 1) + found_name_len; | |
887 | } | |
888 | out: | |
889 | btrfs_free_path(path); | |
890 | return match; | |
891 | } | |
892 | ||
5a1d7843 | 893 | static inline int __add_inode_ref(struct btrfs_trans_handle *trans, |
e02119d5 | 894 | struct btrfs_root *root, |
e02119d5 | 895 | struct btrfs_path *path, |
5a1d7843 JS |
896 | struct btrfs_root *log_root, |
897 | struct inode *dir, struct inode *inode, | |
5a1d7843 | 898 | struct extent_buffer *eb, |
f186373f MF |
899 | u64 inode_objectid, u64 parent_objectid, |
900 | u64 ref_index, char *name, int namelen, | |
901 | int *search_done) | |
e02119d5 | 902 | { |
34f3e4f2 | 903 | int ret; |
f186373f MF |
904 | char *victim_name; |
905 | int victim_name_len; | |
906 | struct extent_buffer *leaf; | |
5a1d7843 | 907 | struct btrfs_dir_item *di; |
f186373f MF |
908 | struct btrfs_key search_key; |
909 | struct btrfs_inode_extref *extref; | |
c622ae60 | 910 | |
f186373f MF |
911 | again: |
912 | /* Search old style refs */ | |
913 | search_key.objectid = inode_objectid; | |
914 | search_key.type = BTRFS_INODE_REF_KEY; | |
915 | search_key.offset = parent_objectid; | |
916 | ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0); | |
e02119d5 | 917 | if (ret == 0) { |
e02119d5 CM |
918 | struct btrfs_inode_ref *victim_ref; |
919 | unsigned long ptr; | |
920 | unsigned long ptr_end; | |
f186373f MF |
921 | |
922 | leaf = path->nodes[0]; | |
e02119d5 CM |
923 | |
924 | /* are we trying to overwrite a back ref for the root directory | |
925 | * if so, just jump out, we're done | |
926 | */ | |
f186373f | 927 | if (search_key.objectid == search_key.offset) |
5a1d7843 | 928 | return 1; |
e02119d5 CM |
929 | |
930 | /* check all the names in this back reference to see | |
931 | * if they are in the log. if so, we allow them to stay | |
932 | * otherwise they must be unlinked as a conflict | |
933 | */ | |
934 | ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
935 | ptr_end = ptr + btrfs_item_size_nr(leaf, path->slots[0]); | |
d397712b | 936 | while (ptr < ptr_end) { |
e02119d5 CM |
937 | victim_ref = (struct btrfs_inode_ref *)ptr; |
938 | victim_name_len = btrfs_inode_ref_name_len(leaf, | |
939 | victim_ref); | |
940 | victim_name = kmalloc(victim_name_len, GFP_NOFS); | |
3650860b JB |
941 | if (!victim_name) |
942 | return -ENOMEM; | |
e02119d5 CM |
943 | |
944 | read_extent_buffer(leaf, victim_name, | |
945 | (unsigned long)(victim_ref + 1), | |
946 | victim_name_len); | |
947 | ||
f186373f MF |
948 | if (!backref_in_log(log_root, &search_key, |
949 | parent_objectid, | |
950 | victim_name, | |
e02119d5 | 951 | victim_name_len)) { |
8b558c5f | 952 | inc_nlink(inode); |
b3b4aa74 | 953 | btrfs_release_path(path); |
12fcfd22 | 954 | |
e02119d5 CM |
955 | ret = btrfs_unlink_inode(trans, root, dir, |
956 | inode, victim_name, | |
957 | victim_name_len); | |
f186373f | 958 | kfree(victim_name); |
3650860b JB |
959 | if (ret) |
960 | return ret; | |
ada9af21 FDBM |
961 | ret = btrfs_run_delayed_items(trans, root); |
962 | if (ret) | |
963 | return ret; | |
f186373f MF |
964 | *search_done = 1; |
965 | goto again; | |
e02119d5 CM |
966 | } |
967 | kfree(victim_name); | |
f186373f | 968 | |
e02119d5 CM |
969 | ptr = (unsigned long)(victim_ref + 1) + victim_name_len; |
970 | } | |
e02119d5 | 971 | |
c622ae60 | 972 | /* |
973 | * NOTE: we have searched root tree and checked the | |
974 | * coresponding ref, it does not need to check again. | |
975 | */ | |
5a1d7843 | 976 | *search_done = 1; |
e02119d5 | 977 | } |
b3b4aa74 | 978 | btrfs_release_path(path); |
e02119d5 | 979 | |
f186373f MF |
980 | /* Same search but for extended refs */ |
981 | extref = btrfs_lookup_inode_extref(NULL, root, path, name, namelen, | |
982 | inode_objectid, parent_objectid, 0, | |
983 | 0); | |
984 | if (!IS_ERR_OR_NULL(extref)) { | |
985 | u32 item_size; | |
986 | u32 cur_offset = 0; | |
987 | unsigned long base; | |
988 | struct inode *victim_parent; | |
989 | ||
990 | leaf = path->nodes[0]; | |
991 | ||
992 | item_size = btrfs_item_size_nr(leaf, path->slots[0]); | |
993 | base = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
994 | ||
995 | while (cur_offset < item_size) { | |
996 | extref = (struct btrfs_inode_extref *)base + cur_offset; | |
997 | ||
998 | victim_name_len = btrfs_inode_extref_name_len(leaf, extref); | |
999 | ||
1000 | if (btrfs_inode_extref_parent(leaf, extref) != parent_objectid) | |
1001 | goto next; | |
1002 | ||
1003 | victim_name = kmalloc(victim_name_len, GFP_NOFS); | |
3650860b JB |
1004 | if (!victim_name) |
1005 | return -ENOMEM; | |
f186373f MF |
1006 | read_extent_buffer(leaf, victim_name, (unsigned long)&extref->name, |
1007 | victim_name_len); | |
1008 | ||
1009 | search_key.objectid = inode_objectid; | |
1010 | search_key.type = BTRFS_INODE_EXTREF_KEY; | |
1011 | search_key.offset = btrfs_extref_hash(parent_objectid, | |
1012 | victim_name, | |
1013 | victim_name_len); | |
1014 | ret = 0; | |
1015 | if (!backref_in_log(log_root, &search_key, | |
1016 | parent_objectid, victim_name, | |
1017 | victim_name_len)) { | |
1018 | ret = -ENOENT; | |
1019 | victim_parent = read_one_inode(root, | |
1020 | parent_objectid); | |
1021 | if (victim_parent) { | |
8b558c5f | 1022 | inc_nlink(inode); |
f186373f MF |
1023 | btrfs_release_path(path); |
1024 | ||
1025 | ret = btrfs_unlink_inode(trans, root, | |
1026 | victim_parent, | |
1027 | inode, | |
1028 | victim_name, | |
1029 | victim_name_len); | |
ada9af21 FDBM |
1030 | if (!ret) |
1031 | ret = btrfs_run_delayed_items( | |
1032 | trans, root); | |
f186373f | 1033 | } |
f186373f MF |
1034 | iput(victim_parent); |
1035 | kfree(victim_name); | |
3650860b JB |
1036 | if (ret) |
1037 | return ret; | |
f186373f MF |
1038 | *search_done = 1; |
1039 | goto again; | |
1040 | } | |
1041 | kfree(victim_name); | |
3650860b JB |
1042 | if (ret) |
1043 | return ret; | |
f186373f MF |
1044 | next: |
1045 | cur_offset += victim_name_len + sizeof(*extref); | |
1046 | } | |
1047 | *search_done = 1; | |
1048 | } | |
1049 | btrfs_release_path(path); | |
1050 | ||
34f3e4f2 | 1051 | /* look for a conflicting sequence number */ |
1052 | di = btrfs_lookup_dir_index_item(trans, root, path, btrfs_ino(dir), | |
f186373f | 1053 | ref_index, name, namelen, 0); |
34f3e4f2 | 1054 | if (di && !IS_ERR(di)) { |
1055 | ret = drop_one_dir_item(trans, root, path, dir, di); | |
3650860b JB |
1056 | if (ret) |
1057 | return ret; | |
34f3e4f2 | 1058 | } |
1059 | btrfs_release_path(path); | |
1060 | ||
1061 | /* look for a conflicing name */ | |
1062 | di = btrfs_lookup_dir_item(trans, root, path, btrfs_ino(dir), | |
1063 | name, namelen, 0); | |
1064 | if (di && !IS_ERR(di)) { | |
1065 | ret = drop_one_dir_item(trans, root, path, dir, di); | |
3650860b JB |
1066 | if (ret) |
1067 | return ret; | |
34f3e4f2 | 1068 | } |
1069 | btrfs_release_path(path); | |
1070 | ||
5a1d7843 JS |
1071 | return 0; |
1072 | } | |
e02119d5 | 1073 | |
f186373f MF |
1074 | static int extref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr, |
1075 | u32 *namelen, char **name, u64 *index, | |
1076 | u64 *parent_objectid) | |
1077 | { | |
1078 | struct btrfs_inode_extref *extref; | |
1079 | ||
1080 | extref = (struct btrfs_inode_extref *)ref_ptr; | |
1081 | ||
1082 | *namelen = btrfs_inode_extref_name_len(eb, extref); | |
1083 | *name = kmalloc(*namelen, GFP_NOFS); | |
1084 | if (*name == NULL) | |
1085 | return -ENOMEM; | |
1086 | ||
1087 | read_extent_buffer(eb, *name, (unsigned long)&extref->name, | |
1088 | *namelen); | |
1089 | ||
1090 | *index = btrfs_inode_extref_index(eb, extref); | |
1091 | if (parent_objectid) | |
1092 | *parent_objectid = btrfs_inode_extref_parent(eb, extref); | |
1093 | ||
1094 | return 0; | |
1095 | } | |
1096 | ||
1097 | static int ref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr, | |
1098 | u32 *namelen, char **name, u64 *index) | |
1099 | { | |
1100 | struct btrfs_inode_ref *ref; | |
1101 | ||
1102 | ref = (struct btrfs_inode_ref *)ref_ptr; | |
1103 | ||
1104 | *namelen = btrfs_inode_ref_name_len(eb, ref); | |
1105 | *name = kmalloc(*namelen, GFP_NOFS); | |
1106 | if (*name == NULL) | |
1107 | return -ENOMEM; | |
1108 | ||
1109 | read_extent_buffer(eb, *name, (unsigned long)(ref + 1), *namelen); | |
1110 | ||
1111 | *index = btrfs_inode_ref_index(eb, ref); | |
1112 | ||
1113 | return 0; | |
1114 | } | |
1115 | ||
5a1d7843 JS |
1116 | /* |
1117 | * replay one inode back reference item found in the log tree. | |
1118 | * eb, slot and key refer to the buffer and key found in the log tree. | |
1119 | * root is the destination we are replaying into, and path is for temp | |
1120 | * use by this function. (it should be released on return). | |
1121 | */ | |
1122 | static noinline int add_inode_ref(struct btrfs_trans_handle *trans, | |
1123 | struct btrfs_root *root, | |
1124 | struct btrfs_root *log, | |
1125 | struct btrfs_path *path, | |
1126 | struct extent_buffer *eb, int slot, | |
1127 | struct btrfs_key *key) | |
1128 | { | |
03b2f08b GB |
1129 | struct inode *dir = NULL; |
1130 | struct inode *inode = NULL; | |
5a1d7843 JS |
1131 | unsigned long ref_ptr; |
1132 | unsigned long ref_end; | |
03b2f08b | 1133 | char *name = NULL; |
5a1d7843 JS |
1134 | int namelen; |
1135 | int ret; | |
1136 | int search_done = 0; | |
f186373f MF |
1137 | int log_ref_ver = 0; |
1138 | u64 parent_objectid; | |
1139 | u64 inode_objectid; | |
f46dbe3d | 1140 | u64 ref_index = 0; |
f186373f MF |
1141 | int ref_struct_size; |
1142 | ||
1143 | ref_ptr = btrfs_item_ptr_offset(eb, slot); | |
1144 | ref_end = ref_ptr + btrfs_item_size_nr(eb, slot); | |
1145 | ||
1146 | if (key->type == BTRFS_INODE_EXTREF_KEY) { | |
1147 | struct btrfs_inode_extref *r; | |
1148 | ||
1149 | ref_struct_size = sizeof(struct btrfs_inode_extref); | |
1150 | log_ref_ver = 1; | |
1151 | r = (struct btrfs_inode_extref *)ref_ptr; | |
1152 | parent_objectid = btrfs_inode_extref_parent(eb, r); | |
1153 | } else { | |
1154 | ref_struct_size = sizeof(struct btrfs_inode_ref); | |
1155 | parent_objectid = key->offset; | |
1156 | } | |
1157 | inode_objectid = key->objectid; | |
e02119d5 | 1158 | |
5a1d7843 JS |
1159 | /* |
1160 | * it is possible that we didn't log all the parent directories | |
1161 | * for a given inode. If we don't find the dir, just don't | |
1162 | * copy the back ref in. The link count fixup code will take | |
1163 | * care of the rest | |
1164 | */ | |
f186373f | 1165 | dir = read_one_inode(root, parent_objectid); |
03b2f08b GB |
1166 | if (!dir) { |
1167 | ret = -ENOENT; | |
1168 | goto out; | |
1169 | } | |
5a1d7843 | 1170 | |
f186373f | 1171 | inode = read_one_inode(root, inode_objectid); |
5a1d7843 | 1172 | if (!inode) { |
03b2f08b GB |
1173 | ret = -EIO; |
1174 | goto out; | |
5a1d7843 JS |
1175 | } |
1176 | ||
5a1d7843 | 1177 | while (ref_ptr < ref_end) { |
f186373f MF |
1178 | if (log_ref_ver) { |
1179 | ret = extref_get_fields(eb, ref_ptr, &namelen, &name, | |
1180 | &ref_index, &parent_objectid); | |
1181 | /* | |
1182 | * parent object can change from one array | |
1183 | * item to another. | |
1184 | */ | |
1185 | if (!dir) | |
1186 | dir = read_one_inode(root, parent_objectid); | |
03b2f08b GB |
1187 | if (!dir) { |
1188 | ret = -ENOENT; | |
1189 | goto out; | |
1190 | } | |
f186373f MF |
1191 | } else { |
1192 | ret = ref_get_fields(eb, ref_ptr, &namelen, &name, | |
1193 | &ref_index); | |
1194 | } | |
1195 | if (ret) | |
03b2f08b | 1196 | goto out; |
5a1d7843 JS |
1197 | |
1198 | /* if we already have a perfect match, we're done */ | |
1199 | if (!inode_in_dir(root, path, btrfs_ino(dir), btrfs_ino(inode), | |
f186373f | 1200 | ref_index, name, namelen)) { |
5a1d7843 JS |
1201 | /* |
1202 | * look for a conflicting back reference in the | |
1203 | * metadata. if we find one we have to unlink that name | |
1204 | * of the file before we add our new link. Later on, we | |
1205 | * overwrite any existing back reference, and we don't | |
1206 | * want to create dangling pointers in the directory. | |
1207 | */ | |
1208 | ||
1209 | if (!search_done) { | |
1210 | ret = __add_inode_ref(trans, root, path, log, | |
f186373f MF |
1211 | dir, inode, eb, |
1212 | inode_objectid, | |
1213 | parent_objectid, | |
1214 | ref_index, name, namelen, | |
5a1d7843 | 1215 | &search_done); |
03b2f08b GB |
1216 | if (ret) { |
1217 | if (ret == 1) | |
1218 | ret = 0; | |
3650860b JB |
1219 | goto out; |
1220 | } | |
5a1d7843 JS |
1221 | } |
1222 | ||
1223 | /* insert our name */ | |
1224 | ret = btrfs_add_link(trans, dir, inode, name, namelen, | |
f186373f | 1225 | 0, ref_index); |
3650860b JB |
1226 | if (ret) |
1227 | goto out; | |
5a1d7843 JS |
1228 | |
1229 | btrfs_update_inode(trans, root, inode); | |
1230 | } | |
1231 | ||
f186373f | 1232 | ref_ptr = (unsigned long)(ref_ptr + ref_struct_size) + namelen; |
5a1d7843 | 1233 | kfree(name); |
03b2f08b | 1234 | name = NULL; |
f186373f MF |
1235 | if (log_ref_ver) { |
1236 | iput(dir); | |
1237 | dir = NULL; | |
1238 | } | |
5a1d7843 | 1239 | } |
e02119d5 CM |
1240 | |
1241 | /* finally write the back reference in the inode */ | |
1242 | ret = overwrite_item(trans, root, path, eb, slot, key); | |
5a1d7843 | 1243 | out: |
b3b4aa74 | 1244 | btrfs_release_path(path); |
03b2f08b | 1245 | kfree(name); |
e02119d5 CM |
1246 | iput(dir); |
1247 | iput(inode); | |
3650860b | 1248 | return ret; |
e02119d5 CM |
1249 | } |
1250 | ||
c71bf099 YZ |
1251 | static int insert_orphan_item(struct btrfs_trans_handle *trans, |
1252 | struct btrfs_root *root, u64 offset) | |
1253 | { | |
1254 | int ret; | |
3f870c28 KN |
1255 | ret = btrfs_find_item(root, NULL, BTRFS_ORPHAN_OBJECTID, |
1256 | offset, BTRFS_ORPHAN_ITEM_KEY, NULL); | |
c71bf099 YZ |
1257 | if (ret > 0) |
1258 | ret = btrfs_insert_orphan_item(trans, root, offset); | |
1259 | return ret; | |
1260 | } | |
1261 | ||
f186373f MF |
1262 | static int count_inode_extrefs(struct btrfs_root *root, |
1263 | struct inode *inode, struct btrfs_path *path) | |
1264 | { | |
1265 | int ret = 0; | |
1266 | int name_len; | |
1267 | unsigned int nlink = 0; | |
1268 | u32 item_size; | |
1269 | u32 cur_offset = 0; | |
1270 | u64 inode_objectid = btrfs_ino(inode); | |
1271 | u64 offset = 0; | |
1272 | unsigned long ptr; | |
1273 | struct btrfs_inode_extref *extref; | |
1274 | struct extent_buffer *leaf; | |
1275 | ||
1276 | while (1) { | |
1277 | ret = btrfs_find_one_extref(root, inode_objectid, offset, path, | |
1278 | &extref, &offset); | |
1279 | if (ret) | |
1280 | break; | |
c71bf099 | 1281 | |
f186373f MF |
1282 | leaf = path->nodes[0]; |
1283 | item_size = btrfs_item_size_nr(leaf, path->slots[0]); | |
1284 | ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
1285 | ||
1286 | while (cur_offset < item_size) { | |
1287 | extref = (struct btrfs_inode_extref *) (ptr + cur_offset); | |
1288 | name_len = btrfs_inode_extref_name_len(leaf, extref); | |
1289 | ||
1290 | nlink++; | |
1291 | ||
1292 | cur_offset += name_len + sizeof(*extref); | |
1293 | } | |
1294 | ||
1295 | offset++; | |
1296 | btrfs_release_path(path); | |
1297 | } | |
1298 | btrfs_release_path(path); | |
1299 | ||
1300 | if (ret < 0) | |
1301 | return ret; | |
1302 | return nlink; | |
1303 | } | |
1304 | ||
1305 | static int count_inode_refs(struct btrfs_root *root, | |
1306 | struct inode *inode, struct btrfs_path *path) | |
e02119d5 | 1307 | { |
e02119d5 CM |
1308 | int ret; |
1309 | struct btrfs_key key; | |
f186373f | 1310 | unsigned int nlink = 0; |
e02119d5 CM |
1311 | unsigned long ptr; |
1312 | unsigned long ptr_end; | |
1313 | int name_len; | |
33345d01 | 1314 | u64 ino = btrfs_ino(inode); |
e02119d5 | 1315 | |
33345d01 | 1316 | key.objectid = ino; |
e02119d5 CM |
1317 | key.type = BTRFS_INODE_REF_KEY; |
1318 | key.offset = (u64)-1; | |
1319 | ||
d397712b | 1320 | while (1) { |
e02119d5 CM |
1321 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); |
1322 | if (ret < 0) | |
1323 | break; | |
1324 | if (ret > 0) { | |
1325 | if (path->slots[0] == 0) | |
1326 | break; | |
1327 | path->slots[0]--; | |
1328 | } | |
e93ae26f | 1329 | process_slot: |
e02119d5 CM |
1330 | btrfs_item_key_to_cpu(path->nodes[0], &key, |
1331 | path->slots[0]); | |
33345d01 | 1332 | if (key.objectid != ino || |
e02119d5 CM |
1333 | key.type != BTRFS_INODE_REF_KEY) |
1334 | break; | |
1335 | ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); | |
1336 | ptr_end = ptr + btrfs_item_size_nr(path->nodes[0], | |
1337 | path->slots[0]); | |
d397712b | 1338 | while (ptr < ptr_end) { |
e02119d5 CM |
1339 | struct btrfs_inode_ref *ref; |
1340 | ||
1341 | ref = (struct btrfs_inode_ref *)ptr; | |
1342 | name_len = btrfs_inode_ref_name_len(path->nodes[0], | |
1343 | ref); | |
1344 | ptr = (unsigned long)(ref + 1) + name_len; | |
1345 | nlink++; | |
1346 | } | |
1347 | ||
1348 | if (key.offset == 0) | |
1349 | break; | |
e93ae26f FDBM |
1350 | if (path->slots[0] > 0) { |
1351 | path->slots[0]--; | |
1352 | goto process_slot; | |
1353 | } | |
e02119d5 | 1354 | key.offset--; |
b3b4aa74 | 1355 | btrfs_release_path(path); |
e02119d5 | 1356 | } |
b3b4aa74 | 1357 | btrfs_release_path(path); |
f186373f MF |
1358 | |
1359 | return nlink; | |
1360 | } | |
1361 | ||
1362 | /* | |
1363 | * There are a few corners where the link count of the file can't | |
1364 | * be properly maintained during replay. So, instead of adding | |
1365 | * lots of complexity to the log code, we just scan the backrefs | |
1366 | * for any file that has been through replay. | |
1367 | * | |
1368 | * The scan will update the link count on the inode to reflect the | |
1369 | * number of back refs found. If it goes down to zero, the iput | |
1370 | * will free the inode. | |
1371 | */ | |
1372 | static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans, | |
1373 | struct btrfs_root *root, | |
1374 | struct inode *inode) | |
1375 | { | |
1376 | struct btrfs_path *path; | |
1377 | int ret; | |
1378 | u64 nlink = 0; | |
1379 | u64 ino = btrfs_ino(inode); | |
1380 | ||
1381 | path = btrfs_alloc_path(); | |
1382 | if (!path) | |
1383 | return -ENOMEM; | |
1384 | ||
1385 | ret = count_inode_refs(root, inode, path); | |
1386 | if (ret < 0) | |
1387 | goto out; | |
1388 | ||
1389 | nlink = ret; | |
1390 | ||
1391 | ret = count_inode_extrefs(root, inode, path); | |
1392 | if (ret == -ENOENT) | |
1393 | ret = 0; | |
1394 | ||
1395 | if (ret < 0) | |
1396 | goto out; | |
1397 | ||
1398 | nlink += ret; | |
1399 | ||
1400 | ret = 0; | |
1401 | ||
e02119d5 | 1402 | if (nlink != inode->i_nlink) { |
bfe86848 | 1403 | set_nlink(inode, nlink); |
e02119d5 CM |
1404 | btrfs_update_inode(trans, root, inode); |
1405 | } | |
8d5bf1cb | 1406 | BTRFS_I(inode)->index_cnt = (u64)-1; |
e02119d5 | 1407 | |
c71bf099 YZ |
1408 | if (inode->i_nlink == 0) { |
1409 | if (S_ISDIR(inode->i_mode)) { | |
1410 | ret = replay_dir_deletes(trans, root, NULL, path, | |
33345d01 | 1411 | ino, 1); |
3650860b JB |
1412 | if (ret) |
1413 | goto out; | |
c71bf099 | 1414 | } |
33345d01 | 1415 | ret = insert_orphan_item(trans, root, ino); |
12fcfd22 | 1416 | } |
12fcfd22 | 1417 | |
f186373f MF |
1418 | out: |
1419 | btrfs_free_path(path); | |
1420 | return ret; | |
e02119d5 CM |
1421 | } |
1422 | ||
1423 | static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans, | |
1424 | struct btrfs_root *root, | |
1425 | struct btrfs_path *path) | |
1426 | { | |
1427 | int ret; | |
1428 | struct btrfs_key key; | |
1429 | struct inode *inode; | |
1430 | ||
1431 | key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID; | |
1432 | key.type = BTRFS_ORPHAN_ITEM_KEY; | |
1433 | key.offset = (u64)-1; | |
d397712b | 1434 | while (1) { |
e02119d5 CM |
1435 | ret = btrfs_search_slot(trans, root, &key, path, -1, 1); |
1436 | if (ret < 0) | |
1437 | break; | |
1438 | ||
1439 | if (ret == 1) { | |
1440 | if (path->slots[0] == 0) | |
1441 | break; | |
1442 | path->slots[0]--; | |
1443 | } | |
1444 | ||
1445 | btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); | |
1446 | if (key.objectid != BTRFS_TREE_LOG_FIXUP_OBJECTID || | |
1447 | key.type != BTRFS_ORPHAN_ITEM_KEY) | |
1448 | break; | |
1449 | ||
1450 | ret = btrfs_del_item(trans, root, path); | |
65a246c5 TI |
1451 | if (ret) |
1452 | goto out; | |
e02119d5 | 1453 | |
b3b4aa74 | 1454 | btrfs_release_path(path); |
e02119d5 | 1455 | inode = read_one_inode(root, key.offset); |
c00e9493 TI |
1456 | if (!inode) |
1457 | return -EIO; | |
e02119d5 CM |
1458 | |
1459 | ret = fixup_inode_link_count(trans, root, inode); | |
e02119d5 | 1460 | iput(inode); |
3650860b JB |
1461 | if (ret) |
1462 | goto out; | |
e02119d5 | 1463 | |
12fcfd22 CM |
1464 | /* |
1465 | * fixup on a directory may create new entries, | |
1466 | * make sure we always look for the highset possible | |
1467 | * offset | |
1468 | */ | |
1469 | key.offset = (u64)-1; | |
e02119d5 | 1470 | } |
65a246c5 TI |
1471 | ret = 0; |
1472 | out: | |
b3b4aa74 | 1473 | btrfs_release_path(path); |
65a246c5 | 1474 | return ret; |
e02119d5 CM |
1475 | } |
1476 | ||
1477 | ||
1478 | /* | |
1479 | * record a given inode in the fixup dir so we can check its link | |
1480 | * count when replay is done. The link count is incremented here | |
1481 | * so the inode won't go away until we check it | |
1482 | */ | |
1483 | static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans, | |
1484 | struct btrfs_root *root, | |
1485 | struct btrfs_path *path, | |
1486 | u64 objectid) | |
1487 | { | |
1488 | struct btrfs_key key; | |
1489 | int ret = 0; | |
1490 | struct inode *inode; | |
1491 | ||
1492 | inode = read_one_inode(root, objectid); | |
c00e9493 TI |
1493 | if (!inode) |
1494 | return -EIO; | |
e02119d5 CM |
1495 | |
1496 | key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID; | |
1497 | btrfs_set_key_type(&key, BTRFS_ORPHAN_ITEM_KEY); | |
1498 | key.offset = objectid; | |
1499 | ||
1500 | ret = btrfs_insert_empty_item(trans, root, path, &key, 0); | |
1501 | ||
b3b4aa74 | 1502 | btrfs_release_path(path); |
e02119d5 | 1503 | if (ret == 0) { |
9bf7a489 JB |
1504 | if (!inode->i_nlink) |
1505 | set_nlink(inode, 1); | |
1506 | else | |
8b558c5f | 1507 | inc_nlink(inode); |
b9959295 | 1508 | ret = btrfs_update_inode(trans, root, inode); |
e02119d5 CM |
1509 | } else if (ret == -EEXIST) { |
1510 | ret = 0; | |
1511 | } else { | |
3650860b | 1512 | BUG(); /* Logic Error */ |
e02119d5 CM |
1513 | } |
1514 | iput(inode); | |
1515 | ||
1516 | return ret; | |
1517 | } | |
1518 | ||
1519 | /* | |
1520 | * when replaying the log for a directory, we only insert names | |
1521 | * for inodes that actually exist. This means an fsync on a directory | |
1522 | * does not implicitly fsync all the new files in it | |
1523 | */ | |
1524 | static noinline int insert_one_name(struct btrfs_trans_handle *trans, | |
1525 | struct btrfs_root *root, | |
1526 | struct btrfs_path *path, | |
1527 | u64 dirid, u64 index, | |
1528 | char *name, int name_len, u8 type, | |
1529 | struct btrfs_key *location) | |
1530 | { | |
1531 | struct inode *inode; | |
1532 | struct inode *dir; | |
1533 | int ret; | |
1534 | ||
1535 | inode = read_one_inode(root, location->objectid); | |
1536 | if (!inode) | |
1537 | return -ENOENT; | |
1538 | ||
1539 | dir = read_one_inode(root, dirid); | |
1540 | if (!dir) { | |
1541 | iput(inode); | |
1542 | return -EIO; | |
1543 | } | |
d555438b | 1544 | |
e02119d5 CM |
1545 | ret = btrfs_add_link(trans, dir, inode, name, name_len, 1, index); |
1546 | ||
1547 | /* FIXME, put inode into FIXUP list */ | |
1548 | ||
1549 | iput(inode); | |
1550 | iput(dir); | |
1551 | return ret; | |
1552 | } | |
1553 | ||
1554 | /* | |
1555 | * take a single entry in a log directory item and replay it into | |
1556 | * the subvolume. | |
1557 | * | |
1558 | * if a conflicting item exists in the subdirectory already, | |
1559 | * the inode it points to is unlinked and put into the link count | |
1560 | * fix up tree. | |
1561 | * | |
1562 | * If a name from the log points to a file or directory that does | |
1563 | * not exist in the FS, it is skipped. fsyncs on directories | |
1564 | * do not force down inodes inside that directory, just changes to the | |
1565 | * names or unlinks in a directory. | |
1566 | */ | |
1567 | static noinline int replay_one_name(struct btrfs_trans_handle *trans, | |
1568 | struct btrfs_root *root, | |
1569 | struct btrfs_path *path, | |
1570 | struct extent_buffer *eb, | |
1571 | struct btrfs_dir_item *di, | |
1572 | struct btrfs_key *key) | |
1573 | { | |
1574 | char *name; | |
1575 | int name_len; | |
1576 | struct btrfs_dir_item *dst_di; | |
1577 | struct btrfs_key found_key; | |
1578 | struct btrfs_key log_key; | |
1579 | struct inode *dir; | |
e02119d5 | 1580 | u8 log_type; |
4bef0848 | 1581 | int exists; |
3650860b | 1582 | int ret = 0; |
d555438b | 1583 | bool update_size = (key->type == BTRFS_DIR_INDEX_KEY); |
e02119d5 CM |
1584 | |
1585 | dir = read_one_inode(root, key->objectid); | |
c00e9493 TI |
1586 | if (!dir) |
1587 | return -EIO; | |
e02119d5 CM |
1588 | |
1589 | name_len = btrfs_dir_name_len(eb, di); | |
1590 | name = kmalloc(name_len, GFP_NOFS); | |
2bac325e FDBM |
1591 | if (!name) { |
1592 | ret = -ENOMEM; | |
1593 | goto out; | |
1594 | } | |
2a29edc6 | 1595 | |
e02119d5 CM |
1596 | log_type = btrfs_dir_type(eb, di); |
1597 | read_extent_buffer(eb, name, (unsigned long)(di + 1), | |
1598 | name_len); | |
1599 | ||
1600 | btrfs_dir_item_key_to_cpu(eb, di, &log_key); | |
4bef0848 CM |
1601 | exists = btrfs_lookup_inode(trans, root, path, &log_key, 0); |
1602 | if (exists == 0) | |
1603 | exists = 1; | |
1604 | else | |
1605 | exists = 0; | |
b3b4aa74 | 1606 | btrfs_release_path(path); |
4bef0848 | 1607 | |
e02119d5 CM |
1608 | if (key->type == BTRFS_DIR_ITEM_KEY) { |
1609 | dst_di = btrfs_lookup_dir_item(trans, root, path, key->objectid, | |
1610 | name, name_len, 1); | |
d397712b | 1611 | } else if (key->type == BTRFS_DIR_INDEX_KEY) { |
e02119d5 CM |
1612 | dst_di = btrfs_lookup_dir_index_item(trans, root, path, |
1613 | key->objectid, | |
1614 | key->offset, name, | |
1615 | name_len, 1); | |
1616 | } else { | |
3650860b JB |
1617 | /* Corruption */ |
1618 | ret = -EINVAL; | |
1619 | goto out; | |
e02119d5 | 1620 | } |
c704005d | 1621 | if (IS_ERR_OR_NULL(dst_di)) { |
e02119d5 CM |
1622 | /* we need a sequence number to insert, so we only |
1623 | * do inserts for the BTRFS_DIR_INDEX_KEY types | |
1624 | */ | |
1625 | if (key->type != BTRFS_DIR_INDEX_KEY) | |
1626 | goto out; | |
1627 | goto insert; | |
1628 | } | |
1629 | ||
1630 | btrfs_dir_item_key_to_cpu(path->nodes[0], dst_di, &found_key); | |
1631 | /* the existing item matches the logged item */ | |
1632 | if (found_key.objectid == log_key.objectid && | |
1633 | found_key.type == log_key.type && | |
1634 | found_key.offset == log_key.offset && | |
1635 | btrfs_dir_type(path->nodes[0], dst_di) == log_type) { | |
1636 | goto out; | |
1637 | } | |
1638 | ||
1639 | /* | |
1640 | * don't drop the conflicting directory entry if the inode | |
1641 | * for the new entry doesn't exist | |
1642 | */ | |
4bef0848 | 1643 | if (!exists) |
e02119d5 CM |
1644 | goto out; |
1645 | ||
e02119d5 | 1646 | ret = drop_one_dir_item(trans, root, path, dir, dst_di); |
3650860b JB |
1647 | if (ret) |
1648 | goto out; | |
e02119d5 CM |
1649 | |
1650 | if (key->type == BTRFS_DIR_INDEX_KEY) | |
1651 | goto insert; | |
1652 | out: | |
b3b4aa74 | 1653 | btrfs_release_path(path); |
d555438b JB |
1654 | if (!ret && update_size) { |
1655 | btrfs_i_size_write(dir, dir->i_size + name_len * 2); | |
1656 | ret = btrfs_update_inode(trans, root, dir); | |
1657 | } | |
e02119d5 CM |
1658 | kfree(name); |
1659 | iput(dir); | |
3650860b | 1660 | return ret; |
e02119d5 CM |
1661 | |
1662 | insert: | |
b3b4aa74 | 1663 | btrfs_release_path(path); |
e02119d5 CM |
1664 | ret = insert_one_name(trans, root, path, key->objectid, key->offset, |
1665 | name, name_len, log_type, &log_key); | |
3650860b JB |
1666 | if (ret && ret != -ENOENT) |
1667 | goto out; | |
d555438b | 1668 | update_size = false; |
3650860b | 1669 | ret = 0; |
e02119d5 CM |
1670 | goto out; |
1671 | } | |
1672 | ||
1673 | /* | |
1674 | * find all the names in a directory item and reconcile them into | |
1675 | * the subvolume. Only BTRFS_DIR_ITEM_KEY types will have more than | |
1676 | * one name in a directory item, but the same code gets used for | |
1677 | * both directory index types | |
1678 | */ | |
1679 | static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans, | |
1680 | struct btrfs_root *root, | |
1681 | struct btrfs_path *path, | |
1682 | struct extent_buffer *eb, int slot, | |
1683 | struct btrfs_key *key) | |
1684 | { | |
1685 | int ret; | |
1686 | u32 item_size = btrfs_item_size_nr(eb, slot); | |
1687 | struct btrfs_dir_item *di; | |
1688 | int name_len; | |
1689 | unsigned long ptr; | |
1690 | unsigned long ptr_end; | |
1691 | ||
1692 | ptr = btrfs_item_ptr_offset(eb, slot); | |
1693 | ptr_end = ptr + item_size; | |
d397712b | 1694 | while (ptr < ptr_end) { |
e02119d5 | 1695 | di = (struct btrfs_dir_item *)ptr; |
22a94d44 JB |
1696 | if (verify_dir_item(root, eb, di)) |
1697 | return -EIO; | |
e02119d5 CM |
1698 | name_len = btrfs_dir_name_len(eb, di); |
1699 | ret = replay_one_name(trans, root, path, eb, di, key); | |
3650860b JB |
1700 | if (ret) |
1701 | return ret; | |
e02119d5 CM |
1702 | ptr = (unsigned long)(di + 1); |
1703 | ptr += name_len; | |
1704 | } | |
1705 | return 0; | |
1706 | } | |
1707 | ||
1708 | /* | |
1709 | * directory replay has two parts. There are the standard directory | |
1710 | * items in the log copied from the subvolume, and range items | |
1711 | * created in the log while the subvolume was logged. | |
1712 | * | |
1713 | * The range items tell us which parts of the key space the log | |
1714 | * is authoritative for. During replay, if a key in the subvolume | |
1715 | * directory is in a logged range item, but not actually in the log | |
1716 | * that means it was deleted from the directory before the fsync | |
1717 | * and should be removed. | |
1718 | */ | |
1719 | static noinline int find_dir_range(struct btrfs_root *root, | |
1720 | struct btrfs_path *path, | |
1721 | u64 dirid, int key_type, | |
1722 | u64 *start_ret, u64 *end_ret) | |
1723 | { | |
1724 | struct btrfs_key key; | |
1725 | u64 found_end; | |
1726 | struct btrfs_dir_log_item *item; | |
1727 | int ret; | |
1728 | int nritems; | |
1729 | ||
1730 | if (*start_ret == (u64)-1) | |
1731 | return 1; | |
1732 | ||
1733 | key.objectid = dirid; | |
1734 | key.type = key_type; | |
1735 | key.offset = *start_ret; | |
1736 | ||
1737 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
1738 | if (ret < 0) | |
1739 | goto out; | |
1740 | if (ret > 0) { | |
1741 | if (path->slots[0] == 0) | |
1742 | goto out; | |
1743 | path->slots[0]--; | |
1744 | } | |
1745 | if (ret != 0) | |
1746 | btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); | |
1747 | ||
1748 | if (key.type != key_type || key.objectid != dirid) { | |
1749 | ret = 1; | |
1750 | goto next; | |
1751 | } | |
1752 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
1753 | struct btrfs_dir_log_item); | |
1754 | found_end = btrfs_dir_log_end(path->nodes[0], item); | |
1755 | ||
1756 | if (*start_ret >= key.offset && *start_ret <= found_end) { | |
1757 | ret = 0; | |
1758 | *start_ret = key.offset; | |
1759 | *end_ret = found_end; | |
1760 | goto out; | |
1761 | } | |
1762 | ret = 1; | |
1763 | next: | |
1764 | /* check the next slot in the tree to see if it is a valid item */ | |
1765 | nritems = btrfs_header_nritems(path->nodes[0]); | |
1766 | if (path->slots[0] >= nritems) { | |
1767 | ret = btrfs_next_leaf(root, path); | |
1768 | if (ret) | |
1769 | goto out; | |
1770 | } else { | |
1771 | path->slots[0]++; | |
1772 | } | |
1773 | ||
1774 | btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); | |
1775 | ||
1776 | if (key.type != key_type || key.objectid != dirid) { | |
1777 | ret = 1; | |
1778 | goto out; | |
1779 | } | |
1780 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
1781 | struct btrfs_dir_log_item); | |
1782 | found_end = btrfs_dir_log_end(path->nodes[0], item); | |
1783 | *start_ret = key.offset; | |
1784 | *end_ret = found_end; | |
1785 | ret = 0; | |
1786 | out: | |
b3b4aa74 | 1787 | btrfs_release_path(path); |
e02119d5 CM |
1788 | return ret; |
1789 | } | |
1790 | ||
1791 | /* | |
1792 | * this looks for a given directory item in the log. If the directory | |
1793 | * item is not in the log, the item is removed and the inode it points | |
1794 | * to is unlinked | |
1795 | */ | |
1796 | static noinline int check_item_in_log(struct btrfs_trans_handle *trans, | |
1797 | struct btrfs_root *root, | |
1798 | struct btrfs_root *log, | |
1799 | struct btrfs_path *path, | |
1800 | struct btrfs_path *log_path, | |
1801 | struct inode *dir, | |
1802 | struct btrfs_key *dir_key) | |
1803 | { | |
1804 | int ret; | |
1805 | struct extent_buffer *eb; | |
1806 | int slot; | |
1807 | u32 item_size; | |
1808 | struct btrfs_dir_item *di; | |
1809 | struct btrfs_dir_item *log_di; | |
1810 | int name_len; | |
1811 | unsigned long ptr; | |
1812 | unsigned long ptr_end; | |
1813 | char *name; | |
1814 | struct inode *inode; | |
1815 | struct btrfs_key location; | |
1816 | ||
1817 | again: | |
1818 | eb = path->nodes[0]; | |
1819 | slot = path->slots[0]; | |
1820 | item_size = btrfs_item_size_nr(eb, slot); | |
1821 | ptr = btrfs_item_ptr_offset(eb, slot); | |
1822 | ptr_end = ptr + item_size; | |
d397712b | 1823 | while (ptr < ptr_end) { |
e02119d5 | 1824 | di = (struct btrfs_dir_item *)ptr; |
22a94d44 JB |
1825 | if (verify_dir_item(root, eb, di)) { |
1826 | ret = -EIO; | |
1827 | goto out; | |
1828 | } | |
1829 | ||
e02119d5 CM |
1830 | name_len = btrfs_dir_name_len(eb, di); |
1831 | name = kmalloc(name_len, GFP_NOFS); | |
1832 | if (!name) { | |
1833 | ret = -ENOMEM; | |
1834 | goto out; | |
1835 | } | |
1836 | read_extent_buffer(eb, name, (unsigned long)(di + 1), | |
1837 | name_len); | |
1838 | log_di = NULL; | |
12fcfd22 | 1839 | if (log && dir_key->type == BTRFS_DIR_ITEM_KEY) { |
e02119d5 CM |
1840 | log_di = btrfs_lookup_dir_item(trans, log, log_path, |
1841 | dir_key->objectid, | |
1842 | name, name_len, 0); | |
12fcfd22 | 1843 | } else if (log && dir_key->type == BTRFS_DIR_INDEX_KEY) { |
e02119d5 CM |
1844 | log_di = btrfs_lookup_dir_index_item(trans, log, |
1845 | log_path, | |
1846 | dir_key->objectid, | |
1847 | dir_key->offset, | |
1848 | name, name_len, 0); | |
1849 | } | |
269d040f | 1850 | if (!log_di || (IS_ERR(log_di) && PTR_ERR(log_di) == -ENOENT)) { |
e02119d5 | 1851 | btrfs_dir_item_key_to_cpu(eb, di, &location); |
b3b4aa74 DS |
1852 | btrfs_release_path(path); |
1853 | btrfs_release_path(log_path); | |
e02119d5 | 1854 | inode = read_one_inode(root, location.objectid); |
c00e9493 TI |
1855 | if (!inode) { |
1856 | kfree(name); | |
1857 | return -EIO; | |
1858 | } | |
e02119d5 CM |
1859 | |
1860 | ret = link_to_fixup_dir(trans, root, | |
1861 | path, location.objectid); | |
3650860b JB |
1862 | if (ret) { |
1863 | kfree(name); | |
1864 | iput(inode); | |
1865 | goto out; | |
1866 | } | |
1867 | ||
8b558c5f | 1868 | inc_nlink(inode); |
e02119d5 CM |
1869 | ret = btrfs_unlink_inode(trans, root, dir, inode, |
1870 | name, name_len); | |
3650860b | 1871 | if (!ret) |
ada9af21 | 1872 | ret = btrfs_run_delayed_items(trans, root); |
e02119d5 CM |
1873 | kfree(name); |
1874 | iput(inode); | |
3650860b JB |
1875 | if (ret) |
1876 | goto out; | |
e02119d5 CM |
1877 | |
1878 | /* there might still be more names under this key | |
1879 | * check and repeat if required | |
1880 | */ | |
1881 | ret = btrfs_search_slot(NULL, root, dir_key, path, | |
1882 | 0, 0); | |
1883 | if (ret == 0) | |
1884 | goto again; | |
1885 | ret = 0; | |
1886 | goto out; | |
269d040f FDBM |
1887 | } else if (IS_ERR(log_di)) { |
1888 | kfree(name); | |
1889 | return PTR_ERR(log_di); | |
e02119d5 | 1890 | } |
b3b4aa74 | 1891 | btrfs_release_path(log_path); |
e02119d5 CM |
1892 | kfree(name); |
1893 | ||
1894 | ptr = (unsigned long)(di + 1); | |
1895 | ptr += name_len; | |
1896 | } | |
1897 | ret = 0; | |
1898 | out: | |
b3b4aa74 DS |
1899 | btrfs_release_path(path); |
1900 | btrfs_release_path(log_path); | |
e02119d5 CM |
1901 | return ret; |
1902 | } | |
1903 | ||
1904 | /* | |
1905 | * deletion replay happens before we copy any new directory items | |
1906 | * out of the log or out of backreferences from inodes. It | |
1907 | * scans the log to find ranges of keys that log is authoritative for, | |
1908 | * and then scans the directory to find items in those ranges that are | |
1909 | * not present in the log. | |
1910 | * | |
1911 | * Anything we don't find in the log is unlinked and removed from the | |
1912 | * directory. | |
1913 | */ | |
1914 | static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans, | |
1915 | struct btrfs_root *root, | |
1916 | struct btrfs_root *log, | |
1917 | struct btrfs_path *path, | |
12fcfd22 | 1918 | u64 dirid, int del_all) |
e02119d5 CM |
1919 | { |
1920 | u64 range_start; | |
1921 | u64 range_end; | |
1922 | int key_type = BTRFS_DIR_LOG_ITEM_KEY; | |
1923 | int ret = 0; | |
1924 | struct btrfs_key dir_key; | |
1925 | struct btrfs_key found_key; | |
1926 | struct btrfs_path *log_path; | |
1927 | struct inode *dir; | |
1928 | ||
1929 | dir_key.objectid = dirid; | |
1930 | dir_key.type = BTRFS_DIR_ITEM_KEY; | |
1931 | log_path = btrfs_alloc_path(); | |
1932 | if (!log_path) | |
1933 | return -ENOMEM; | |
1934 | ||
1935 | dir = read_one_inode(root, dirid); | |
1936 | /* it isn't an error if the inode isn't there, that can happen | |
1937 | * because we replay the deletes before we copy in the inode item | |
1938 | * from the log | |
1939 | */ | |
1940 | if (!dir) { | |
1941 | btrfs_free_path(log_path); | |
1942 | return 0; | |
1943 | } | |
1944 | again: | |
1945 | range_start = 0; | |
1946 | range_end = 0; | |
d397712b | 1947 | while (1) { |
12fcfd22 CM |
1948 | if (del_all) |
1949 | range_end = (u64)-1; | |
1950 | else { | |
1951 | ret = find_dir_range(log, path, dirid, key_type, | |
1952 | &range_start, &range_end); | |
1953 | if (ret != 0) | |
1954 | break; | |
1955 | } | |
e02119d5 CM |
1956 | |
1957 | dir_key.offset = range_start; | |
d397712b | 1958 | while (1) { |
e02119d5 CM |
1959 | int nritems; |
1960 | ret = btrfs_search_slot(NULL, root, &dir_key, path, | |
1961 | 0, 0); | |
1962 | if (ret < 0) | |
1963 | goto out; | |
1964 | ||
1965 | nritems = btrfs_header_nritems(path->nodes[0]); | |
1966 | if (path->slots[0] >= nritems) { | |
1967 | ret = btrfs_next_leaf(root, path); | |
1968 | if (ret) | |
1969 | break; | |
1970 | } | |
1971 | btrfs_item_key_to_cpu(path->nodes[0], &found_key, | |
1972 | path->slots[0]); | |
1973 | if (found_key.objectid != dirid || | |
1974 | found_key.type != dir_key.type) | |
1975 | goto next_type; | |
1976 | ||
1977 | if (found_key.offset > range_end) | |
1978 | break; | |
1979 | ||
1980 | ret = check_item_in_log(trans, root, log, path, | |
12fcfd22 CM |
1981 | log_path, dir, |
1982 | &found_key); | |
3650860b JB |
1983 | if (ret) |
1984 | goto out; | |
e02119d5 CM |
1985 | if (found_key.offset == (u64)-1) |
1986 | break; | |
1987 | dir_key.offset = found_key.offset + 1; | |
1988 | } | |
b3b4aa74 | 1989 | btrfs_release_path(path); |
e02119d5 CM |
1990 | if (range_end == (u64)-1) |
1991 | break; | |
1992 | range_start = range_end + 1; | |
1993 | } | |
1994 | ||
1995 | next_type: | |
1996 | ret = 0; | |
1997 | if (key_type == BTRFS_DIR_LOG_ITEM_KEY) { | |
1998 | key_type = BTRFS_DIR_LOG_INDEX_KEY; | |
1999 | dir_key.type = BTRFS_DIR_INDEX_KEY; | |
b3b4aa74 | 2000 | btrfs_release_path(path); |
e02119d5 CM |
2001 | goto again; |
2002 | } | |
2003 | out: | |
b3b4aa74 | 2004 | btrfs_release_path(path); |
e02119d5 CM |
2005 | btrfs_free_path(log_path); |
2006 | iput(dir); | |
2007 | return ret; | |
2008 | } | |
2009 | ||
2010 | /* | |
2011 | * the process_func used to replay items from the log tree. This | |
2012 | * gets called in two different stages. The first stage just looks | |
2013 | * for inodes and makes sure they are all copied into the subvolume. | |
2014 | * | |
2015 | * The second stage copies all the other item types from the log into | |
2016 | * the subvolume. The two stage approach is slower, but gets rid of | |
2017 | * lots of complexity around inodes referencing other inodes that exist | |
2018 | * only in the log (references come from either directory items or inode | |
2019 | * back refs). | |
2020 | */ | |
2021 | static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, | |
2022 | struct walk_control *wc, u64 gen) | |
2023 | { | |
2024 | int nritems; | |
2025 | struct btrfs_path *path; | |
2026 | struct btrfs_root *root = wc->replay_dest; | |
2027 | struct btrfs_key key; | |
e02119d5 CM |
2028 | int level; |
2029 | int i; | |
2030 | int ret; | |
2031 | ||
018642a1 TI |
2032 | ret = btrfs_read_buffer(eb, gen); |
2033 | if (ret) | |
2034 | return ret; | |
e02119d5 CM |
2035 | |
2036 | level = btrfs_header_level(eb); | |
2037 | ||
2038 | if (level != 0) | |
2039 | return 0; | |
2040 | ||
2041 | path = btrfs_alloc_path(); | |
1e5063d0 MF |
2042 | if (!path) |
2043 | return -ENOMEM; | |
e02119d5 CM |
2044 | |
2045 | nritems = btrfs_header_nritems(eb); | |
2046 | for (i = 0; i < nritems; i++) { | |
2047 | btrfs_item_key_to_cpu(eb, &key, i); | |
e02119d5 CM |
2048 | |
2049 | /* inode keys are done during the first stage */ | |
2050 | if (key.type == BTRFS_INODE_ITEM_KEY && | |
2051 | wc->stage == LOG_WALK_REPLAY_INODES) { | |
e02119d5 CM |
2052 | struct btrfs_inode_item *inode_item; |
2053 | u32 mode; | |
2054 | ||
2055 | inode_item = btrfs_item_ptr(eb, i, | |
2056 | struct btrfs_inode_item); | |
2057 | mode = btrfs_inode_mode(eb, inode_item); | |
2058 | if (S_ISDIR(mode)) { | |
2059 | ret = replay_dir_deletes(wc->trans, | |
12fcfd22 | 2060 | root, log, path, key.objectid, 0); |
b50c6e25 JB |
2061 | if (ret) |
2062 | break; | |
e02119d5 CM |
2063 | } |
2064 | ret = overwrite_item(wc->trans, root, path, | |
2065 | eb, i, &key); | |
b50c6e25 JB |
2066 | if (ret) |
2067 | break; | |
e02119d5 | 2068 | |
c71bf099 YZ |
2069 | /* for regular files, make sure corresponding |
2070 | * orhpan item exist. extents past the new EOF | |
2071 | * will be truncated later by orphan cleanup. | |
e02119d5 CM |
2072 | */ |
2073 | if (S_ISREG(mode)) { | |
c71bf099 YZ |
2074 | ret = insert_orphan_item(wc->trans, root, |
2075 | key.objectid); | |
b50c6e25 JB |
2076 | if (ret) |
2077 | break; | |
e02119d5 | 2078 | } |
c71bf099 | 2079 | |
e02119d5 CM |
2080 | ret = link_to_fixup_dir(wc->trans, root, |
2081 | path, key.objectid); | |
b50c6e25 JB |
2082 | if (ret) |
2083 | break; | |
e02119d5 | 2084 | } |
dd8e7217 JB |
2085 | |
2086 | if (key.type == BTRFS_DIR_INDEX_KEY && | |
2087 | wc->stage == LOG_WALK_REPLAY_DIR_INDEX) { | |
2088 | ret = replay_one_dir_item(wc->trans, root, path, | |
2089 | eb, i, &key); | |
2090 | if (ret) | |
2091 | break; | |
2092 | } | |
2093 | ||
e02119d5 CM |
2094 | if (wc->stage < LOG_WALK_REPLAY_ALL) |
2095 | continue; | |
2096 | ||
2097 | /* these keys are simply copied */ | |
2098 | if (key.type == BTRFS_XATTR_ITEM_KEY) { | |
2099 | ret = overwrite_item(wc->trans, root, path, | |
2100 | eb, i, &key); | |
b50c6e25 JB |
2101 | if (ret) |
2102 | break; | |
2da1c669 LB |
2103 | } else if (key.type == BTRFS_INODE_REF_KEY || |
2104 | key.type == BTRFS_INODE_EXTREF_KEY) { | |
f186373f MF |
2105 | ret = add_inode_ref(wc->trans, root, log, path, |
2106 | eb, i, &key); | |
b50c6e25 JB |
2107 | if (ret && ret != -ENOENT) |
2108 | break; | |
2109 | ret = 0; | |
e02119d5 CM |
2110 | } else if (key.type == BTRFS_EXTENT_DATA_KEY) { |
2111 | ret = replay_one_extent(wc->trans, root, path, | |
2112 | eb, i, &key); | |
b50c6e25 JB |
2113 | if (ret) |
2114 | break; | |
dd8e7217 | 2115 | } else if (key.type == BTRFS_DIR_ITEM_KEY) { |
e02119d5 CM |
2116 | ret = replay_one_dir_item(wc->trans, root, path, |
2117 | eb, i, &key); | |
b50c6e25 JB |
2118 | if (ret) |
2119 | break; | |
e02119d5 CM |
2120 | } |
2121 | } | |
2122 | btrfs_free_path(path); | |
b50c6e25 | 2123 | return ret; |
e02119d5 CM |
2124 | } |
2125 | ||
d397712b | 2126 | static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, |
e02119d5 CM |
2127 | struct btrfs_root *root, |
2128 | struct btrfs_path *path, int *level, | |
2129 | struct walk_control *wc) | |
2130 | { | |
2131 | u64 root_owner; | |
e02119d5 CM |
2132 | u64 bytenr; |
2133 | u64 ptr_gen; | |
2134 | struct extent_buffer *next; | |
2135 | struct extent_buffer *cur; | |
2136 | struct extent_buffer *parent; | |
2137 | u32 blocksize; | |
2138 | int ret = 0; | |
2139 | ||
2140 | WARN_ON(*level < 0); | |
2141 | WARN_ON(*level >= BTRFS_MAX_LEVEL); | |
2142 | ||
d397712b | 2143 | while (*level > 0) { |
e02119d5 CM |
2144 | WARN_ON(*level < 0); |
2145 | WARN_ON(*level >= BTRFS_MAX_LEVEL); | |
2146 | cur = path->nodes[*level]; | |
2147 | ||
fae7f21c | 2148 | WARN_ON(btrfs_header_level(cur) != *level); |
e02119d5 CM |
2149 | |
2150 | if (path->slots[*level] >= | |
2151 | btrfs_header_nritems(cur)) | |
2152 | break; | |
2153 | ||
2154 | bytenr = btrfs_node_blockptr(cur, path->slots[*level]); | |
2155 | ptr_gen = btrfs_node_ptr_generation(cur, path->slots[*level]); | |
2156 | blocksize = btrfs_level_size(root, *level - 1); | |
2157 | ||
2158 | parent = path->nodes[*level]; | |
2159 | root_owner = btrfs_header_owner(parent); | |
e02119d5 CM |
2160 | |
2161 | next = btrfs_find_create_tree_block(root, bytenr, blocksize); | |
2a29edc6 | 2162 | if (!next) |
2163 | return -ENOMEM; | |
e02119d5 | 2164 | |
e02119d5 | 2165 | if (*level == 1) { |
1e5063d0 | 2166 | ret = wc->process_func(root, next, wc, ptr_gen); |
b50c6e25 JB |
2167 | if (ret) { |
2168 | free_extent_buffer(next); | |
1e5063d0 | 2169 | return ret; |
b50c6e25 | 2170 | } |
4a500fd1 | 2171 | |
e02119d5 CM |
2172 | path->slots[*level]++; |
2173 | if (wc->free) { | |
018642a1 TI |
2174 | ret = btrfs_read_buffer(next, ptr_gen); |
2175 | if (ret) { | |
2176 | free_extent_buffer(next); | |
2177 | return ret; | |
2178 | } | |
e02119d5 | 2179 | |
681ae509 JB |
2180 | if (trans) { |
2181 | btrfs_tree_lock(next); | |
2182 | btrfs_set_lock_blocking(next); | |
2183 | clean_tree_block(trans, root, next); | |
2184 | btrfs_wait_tree_block_writeback(next); | |
2185 | btrfs_tree_unlock(next); | |
2186 | } | |
e02119d5 | 2187 | |
e02119d5 CM |
2188 | WARN_ON(root_owner != |
2189 | BTRFS_TREE_LOG_OBJECTID); | |
e688b725 | 2190 | ret = btrfs_free_and_pin_reserved_extent(root, |
d00aff00 | 2191 | bytenr, blocksize); |
3650860b JB |
2192 | if (ret) { |
2193 | free_extent_buffer(next); | |
2194 | return ret; | |
2195 | } | |
e02119d5 CM |
2196 | } |
2197 | free_extent_buffer(next); | |
2198 | continue; | |
2199 | } | |
018642a1 TI |
2200 | ret = btrfs_read_buffer(next, ptr_gen); |
2201 | if (ret) { | |
2202 | free_extent_buffer(next); | |
2203 | return ret; | |
2204 | } | |
e02119d5 CM |
2205 | |
2206 | WARN_ON(*level <= 0); | |
2207 | if (path->nodes[*level-1]) | |
2208 | free_extent_buffer(path->nodes[*level-1]); | |
2209 | path->nodes[*level-1] = next; | |
2210 | *level = btrfs_header_level(next); | |
2211 | path->slots[*level] = 0; | |
2212 | cond_resched(); | |
2213 | } | |
2214 | WARN_ON(*level < 0); | |
2215 | WARN_ON(*level >= BTRFS_MAX_LEVEL); | |
2216 | ||
4a500fd1 | 2217 | path->slots[*level] = btrfs_header_nritems(path->nodes[*level]); |
e02119d5 CM |
2218 | |
2219 | cond_resched(); | |
2220 | return 0; | |
2221 | } | |
2222 | ||
d397712b | 2223 | static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, |
e02119d5 CM |
2224 | struct btrfs_root *root, |
2225 | struct btrfs_path *path, int *level, | |
2226 | struct walk_control *wc) | |
2227 | { | |
2228 | u64 root_owner; | |
e02119d5 CM |
2229 | int i; |
2230 | int slot; | |
2231 | int ret; | |
2232 | ||
d397712b | 2233 | for (i = *level; i < BTRFS_MAX_LEVEL - 1 && path->nodes[i]; i++) { |
e02119d5 | 2234 | slot = path->slots[i]; |
4a500fd1 | 2235 | if (slot + 1 < btrfs_header_nritems(path->nodes[i])) { |
e02119d5 CM |
2236 | path->slots[i]++; |
2237 | *level = i; | |
2238 | WARN_ON(*level == 0); | |
2239 | return 0; | |
2240 | } else { | |
31840ae1 ZY |
2241 | struct extent_buffer *parent; |
2242 | if (path->nodes[*level] == root->node) | |
2243 | parent = path->nodes[*level]; | |
2244 | else | |
2245 | parent = path->nodes[*level + 1]; | |
2246 | ||
2247 | root_owner = btrfs_header_owner(parent); | |
1e5063d0 | 2248 | ret = wc->process_func(root, path->nodes[*level], wc, |
e02119d5 | 2249 | btrfs_header_generation(path->nodes[*level])); |
1e5063d0 MF |
2250 | if (ret) |
2251 | return ret; | |
2252 | ||
e02119d5 CM |
2253 | if (wc->free) { |
2254 | struct extent_buffer *next; | |
2255 | ||
2256 | next = path->nodes[*level]; | |
2257 | ||
681ae509 JB |
2258 | if (trans) { |
2259 | btrfs_tree_lock(next); | |
2260 | btrfs_set_lock_blocking(next); | |
2261 | clean_tree_block(trans, root, next); | |
2262 | btrfs_wait_tree_block_writeback(next); | |
2263 | btrfs_tree_unlock(next); | |
2264 | } | |
e02119d5 | 2265 | |
e02119d5 | 2266 | WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID); |
e688b725 | 2267 | ret = btrfs_free_and_pin_reserved_extent(root, |
e02119d5 | 2268 | path->nodes[*level]->start, |
d00aff00 | 2269 | path->nodes[*level]->len); |
3650860b JB |
2270 | if (ret) |
2271 | return ret; | |
e02119d5 CM |
2272 | } |
2273 | free_extent_buffer(path->nodes[*level]); | |
2274 | path->nodes[*level] = NULL; | |
2275 | *level = i + 1; | |
2276 | } | |
2277 | } | |
2278 | return 1; | |
2279 | } | |
2280 | ||
2281 | /* | |
2282 | * drop the reference count on the tree rooted at 'snap'. This traverses | |
2283 | * the tree freeing any blocks that have a ref count of zero after being | |
2284 | * decremented. | |
2285 | */ | |
2286 | static int walk_log_tree(struct btrfs_trans_handle *trans, | |
2287 | struct btrfs_root *log, struct walk_control *wc) | |
2288 | { | |
2289 | int ret = 0; | |
2290 | int wret; | |
2291 | int level; | |
2292 | struct btrfs_path *path; | |
e02119d5 CM |
2293 | int orig_level; |
2294 | ||
2295 | path = btrfs_alloc_path(); | |
db5b493a TI |
2296 | if (!path) |
2297 | return -ENOMEM; | |
e02119d5 CM |
2298 | |
2299 | level = btrfs_header_level(log->node); | |
2300 | orig_level = level; | |
2301 | path->nodes[level] = log->node; | |
2302 | extent_buffer_get(log->node); | |
2303 | path->slots[level] = 0; | |
2304 | ||
d397712b | 2305 | while (1) { |
e02119d5 CM |
2306 | wret = walk_down_log_tree(trans, log, path, &level, wc); |
2307 | if (wret > 0) | |
2308 | break; | |
79787eaa | 2309 | if (wret < 0) { |
e02119d5 | 2310 | ret = wret; |
79787eaa JM |
2311 | goto out; |
2312 | } | |
e02119d5 CM |
2313 | |
2314 | wret = walk_up_log_tree(trans, log, path, &level, wc); | |
2315 | if (wret > 0) | |
2316 | break; | |
79787eaa | 2317 | if (wret < 0) { |
e02119d5 | 2318 | ret = wret; |
79787eaa JM |
2319 | goto out; |
2320 | } | |
e02119d5 CM |
2321 | } |
2322 | ||
2323 | /* was the root node processed? if not, catch it here */ | |
2324 | if (path->nodes[orig_level]) { | |
79787eaa | 2325 | ret = wc->process_func(log, path->nodes[orig_level], wc, |
e02119d5 | 2326 | btrfs_header_generation(path->nodes[orig_level])); |
79787eaa JM |
2327 | if (ret) |
2328 | goto out; | |
e02119d5 CM |
2329 | if (wc->free) { |
2330 | struct extent_buffer *next; | |
2331 | ||
2332 | next = path->nodes[orig_level]; | |
2333 | ||
681ae509 JB |
2334 | if (trans) { |
2335 | btrfs_tree_lock(next); | |
2336 | btrfs_set_lock_blocking(next); | |
2337 | clean_tree_block(trans, log, next); | |
2338 | btrfs_wait_tree_block_writeback(next); | |
2339 | btrfs_tree_unlock(next); | |
2340 | } | |
e02119d5 | 2341 | |
e02119d5 CM |
2342 | WARN_ON(log->root_key.objectid != |
2343 | BTRFS_TREE_LOG_OBJECTID); | |
e688b725 | 2344 | ret = btrfs_free_and_pin_reserved_extent(log, next->start, |
d00aff00 | 2345 | next->len); |
3650860b JB |
2346 | if (ret) |
2347 | goto out; | |
e02119d5 CM |
2348 | } |
2349 | } | |
2350 | ||
79787eaa | 2351 | out: |
e02119d5 | 2352 | btrfs_free_path(path); |
e02119d5 CM |
2353 | return ret; |
2354 | } | |
2355 | ||
7237f183 YZ |
2356 | /* |
2357 | * helper function to update the item for a given subvolumes log root | |
2358 | * in the tree of log roots | |
2359 | */ | |
2360 | static int update_log_root(struct btrfs_trans_handle *trans, | |
2361 | struct btrfs_root *log) | |
2362 | { | |
2363 | int ret; | |
2364 | ||
2365 | if (log->log_transid == 1) { | |
2366 | /* insert root item on the first sync */ | |
2367 | ret = btrfs_insert_root(trans, log->fs_info->log_root_tree, | |
2368 | &log->root_key, &log->root_item); | |
2369 | } else { | |
2370 | ret = btrfs_update_root(trans, log->fs_info->log_root_tree, | |
2371 | &log->root_key, &log->root_item); | |
2372 | } | |
2373 | return ret; | |
2374 | } | |
2375 | ||
8b050d35 MX |
2376 | static void wait_log_commit(struct btrfs_trans_handle *trans, |
2377 | struct btrfs_root *root, int transid) | |
e02119d5 CM |
2378 | { |
2379 | DEFINE_WAIT(wait); | |
7237f183 | 2380 | int index = transid % 2; |
e02119d5 | 2381 | |
7237f183 YZ |
2382 | /* |
2383 | * we only allow two pending log transactions at a time, | |
2384 | * so we know that if ours is more than 2 older than the | |
2385 | * current transaction, we're done | |
2386 | */ | |
e02119d5 | 2387 | do { |
7237f183 YZ |
2388 | prepare_to_wait(&root->log_commit_wait[index], |
2389 | &wait, TASK_UNINTERRUPTIBLE); | |
2390 | mutex_unlock(&root->log_mutex); | |
12fcfd22 | 2391 | |
d1433deb | 2392 | if (root->log_transid_committed < transid && |
7237f183 YZ |
2393 | atomic_read(&root->log_commit[index])) |
2394 | schedule(); | |
12fcfd22 | 2395 | |
7237f183 YZ |
2396 | finish_wait(&root->log_commit_wait[index], &wait); |
2397 | mutex_lock(&root->log_mutex); | |
d1433deb | 2398 | } while (root->log_transid_committed < transid && |
7237f183 | 2399 | atomic_read(&root->log_commit[index])); |
7237f183 YZ |
2400 | } |
2401 | ||
143bede5 JM |
2402 | static void wait_for_writer(struct btrfs_trans_handle *trans, |
2403 | struct btrfs_root *root) | |
7237f183 YZ |
2404 | { |
2405 | DEFINE_WAIT(wait); | |
8b050d35 MX |
2406 | |
2407 | while (atomic_read(&root->log_writers)) { | |
7237f183 YZ |
2408 | prepare_to_wait(&root->log_writer_wait, |
2409 | &wait, TASK_UNINTERRUPTIBLE); | |
2410 | mutex_unlock(&root->log_mutex); | |
8b050d35 | 2411 | if (atomic_read(&root->log_writers)) |
e02119d5 | 2412 | schedule(); |
7237f183 YZ |
2413 | mutex_lock(&root->log_mutex); |
2414 | finish_wait(&root->log_writer_wait, &wait); | |
2415 | } | |
e02119d5 CM |
2416 | } |
2417 | ||
8b050d35 MX |
2418 | static inline void btrfs_remove_log_ctx(struct btrfs_root *root, |
2419 | struct btrfs_log_ctx *ctx) | |
2420 | { | |
2421 | if (!ctx) | |
2422 | return; | |
2423 | ||
2424 | mutex_lock(&root->log_mutex); | |
2425 | list_del_init(&ctx->list); | |
2426 | mutex_unlock(&root->log_mutex); | |
2427 | } | |
2428 | ||
2429 | /* | |
2430 | * Invoked in log mutex context, or be sure there is no other task which | |
2431 | * can access the list. | |
2432 | */ | |
2433 | static inline void btrfs_remove_all_log_ctxs(struct btrfs_root *root, | |
2434 | int index, int error) | |
2435 | { | |
2436 | struct btrfs_log_ctx *ctx; | |
2437 | ||
2438 | if (!error) { | |
2439 | INIT_LIST_HEAD(&root->log_ctxs[index]); | |
2440 | return; | |
2441 | } | |
2442 | ||
2443 | list_for_each_entry(ctx, &root->log_ctxs[index], list) | |
2444 | ctx->log_ret = error; | |
2445 | ||
2446 | INIT_LIST_HEAD(&root->log_ctxs[index]); | |
2447 | } | |
2448 | ||
e02119d5 CM |
2449 | /* |
2450 | * btrfs_sync_log does sends a given tree log down to the disk and | |
2451 | * updates the super blocks to record it. When this call is done, | |
12fcfd22 CM |
2452 | * you know that any inodes previously logged are safely on disk only |
2453 | * if it returns 0. | |
2454 | * | |
2455 | * Any other return value means you need to call btrfs_commit_transaction. | |
2456 | * Some of the edge cases for fsyncing directories that have had unlinks | |
2457 | * or renames done in the past mean that sometimes the only safe | |
2458 | * fsync is to commit the whole FS. When btrfs_sync_log returns -EAGAIN, | |
2459 | * that has happened. | |
e02119d5 CM |
2460 | */ |
2461 | int btrfs_sync_log(struct btrfs_trans_handle *trans, | |
8b050d35 | 2462 | struct btrfs_root *root, struct btrfs_log_ctx *ctx) |
e02119d5 | 2463 | { |
7237f183 YZ |
2464 | int index1; |
2465 | int index2; | |
8cef4e16 | 2466 | int mark; |
e02119d5 | 2467 | int ret; |
e02119d5 | 2468 | struct btrfs_root *log = root->log_root; |
7237f183 | 2469 | struct btrfs_root *log_root_tree = root->fs_info->log_root_tree; |
bb14a59b | 2470 | int log_transid = 0; |
8b050d35 | 2471 | struct btrfs_log_ctx root_log_ctx; |
c6adc9cc | 2472 | struct blk_plug plug; |
e02119d5 | 2473 | |
7237f183 | 2474 | mutex_lock(&root->log_mutex); |
d1433deb MX |
2475 | log_transid = ctx->log_transid; |
2476 | if (root->log_transid_committed >= log_transid) { | |
2477 | mutex_unlock(&root->log_mutex); | |
2478 | return ctx->log_ret; | |
2479 | } | |
2480 | ||
2481 | index1 = log_transid % 2; | |
7237f183 | 2482 | if (atomic_read(&root->log_commit[index1])) { |
d1433deb | 2483 | wait_log_commit(trans, root, log_transid); |
7237f183 | 2484 | mutex_unlock(&root->log_mutex); |
8b050d35 | 2485 | return ctx->log_ret; |
e02119d5 | 2486 | } |
d1433deb | 2487 | ASSERT(log_transid == root->log_transid); |
7237f183 YZ |
2488 | atomic_set(&root->log_commit[index1], 1); |
2489 | ||
2490 | /* wait for previous tree log sync to complete */ | |
2491 | if (atomic_read(&root->log_commit[(index1 + 1) % 2])) | |
d1433deb | 2492 | wait_log_commit(trans, root, log_transid - 1); |
48cab2e0 | 2493 | |
86df7eb9 | 2494 | while (1) { |
2ecb7923 | 2495 | int batch = atomic_read(&root->log_batch); |
cd354ad6 CM |
2496 | /* when we're on an ssd, just kick the log commit out */ |
2497 | if (!btrfs_test_opt(root, SSD) && root->log_multiple_pids) { | |
86df7eb9 YZ |
2498 | mutex_unlock(&root->log_mutex); |
2499 | schedule_timeout_uninterruptible(1); | |
2500 | mutex_lock(&root->log_mutex); | |
2501 | } | |
12fcfd22 | 2502 | wait_for_writer(trans, root); |
2ecb7923 | 2503 | if (batch == atomic_read(&root->log_batch)) |
e02119d5 CM |
2504 | break; |
2505 | } | |
e02119d5 | 2506 | |
12fcfd22 | 2507 | /* bail out if we need to do a full commit */ |
5c902ba6 MX |
2508 | if (ACCESS_ONCE(root->fs_info->last_trans_log_full_commit) == |
2509 | trans->transid) { | |
12fcfd22 | 2510 | ret = -EAGAIN; |
2ab28f32 | 2511 | btrfs_free_logged_extents(log, log_transid); |
12fcfd22 CM |
2512 | mutex_unlock(&root->log_mutex); |
2513 | goto out; | |
2514 | } | |
2515 | ||
8cef4e16 YZ |
2516 | if (log_transid % 2 == 0) |
2517 | mark = EXTENT_DIRTY; | |
2518 | else | |
2519 | mark = EXTENT_NEW; | |
2520 | ||
690587d1 CM |
2521 | /* we start IO on all the marked extents here, but we don't actually |
2522 | * wait for them until later. | |
2523 | */ | |
c6adc9cc | 2524 | blk_start_plug(&plug); |
8cef4e16 | 2525 | ret = btrfs_write_marked_extents(log, &log->dirty_log_pages, mark); |
79787eaa | 2526 | if (ret) { |
c6adc9cc | 2527 | blk_finish_plug(&plug); |
79787eaa | 2528 | btrfs_abort_transaction(trans, root, ret); |
2ab28f32 | 2529 | btrfs_free_logged_extents(log, log_transid); |
79787eaa JM |
2530 | mutex_unlock(&root->log_mutex); |
2531 | goto out; | |
2532 | } | |
7237f183 | 2533 | |
5d4f98a2 | 2534 | btrfs_set_root_node(&log->root_item, log->node); |
7237f183 | 2535 | |
7237f183 YZ |
2536 | root->log_transid++; |
2537 | log->log_transid = root->log_transid; | |
ff782e0a | 2538 | root->log_start_pid = 0; |
7237f183 | 2539 | /* |
8cef4e16 YZ |
2540 | * IO has been started, blocks of the log tree have WRITTEN flag set |
2541 | * in their headers. new modifications of the log will be written to | |
2542 | * new positions. so it's safe to allow log writers to go in. | |
7237f183 YZ |
2543 | */ |
2544 | mutex_unlock(&root->log_mutex); | |
2545 | ||
d1433deb MX |
2546 | btrfs_init_log_ctx(&root_log_ctx); |
2547 | ||
7237f183 | 2548 | mutex_lock(&log_root_tree->log_mutex); |
2ecb7923 | 2549 | atomic_inc(&log_root_tree->log_batch); |
7237f183 | 2550 | atomic_inc(&log_root_tree->log_writers); |
d1433deb MX |
2551 | |
2552 | index2 = log_root_tree->log_transid % 2; | |
2553 | list_add_tail(&root_log_ctx.list, &log_root_tree->log_ctxs[index2]); | |
2554 | root_log_ctx.log_transid = log_root_tree->log_transid; | |
2555 | ||
7237f183 YZ |
2556 | mutex_unlock(&log_root_tree->log_mutex); |
2557 | ||
2558 | ret = update_log_root(trans, log); | |
7237f183 YZ |
2559 | |
2560 | mutex_lock(&log_root_tree->log_mutex); | |
2561 | if (atomic_dec_and_test(&log_root_tree->log_writers)) { | |
2562 | smp_mb(); | |
2563 | if (waitqueue_active(&log_root_tree->log_writer_wait)) | |
2564 | wake_up(&log_root_tree->log_writer_wait); | |
2565 | } | |
2566 | ||
4a500fd1 | 2567 | if (ret) { |
d1433deb MX |
2568 | if (!list_empty(&root_log_ctx.list)) |
2569 | list_del_init(&root_log_ctx.list); | |
2570 | ||
c6adc9cc | 2571 | blk_finish_plug(&plug); |
79787eaa JM |
2572 | if (ret != -ENOSPC) { |
2573 | btrfs_abort_transaction(trans, root, ret); | |
2574 | mutex_unlock(&log_root_tree->log_mutex); | |
2575 | goto out; | |
2576 | } | |
5c902ba6 MX |
2577 | ACCESS_ONCE(root->fs_info->last_trans_log_full_commit) = |
2578 | trans->transid; | |
4a500fd1 | 2579 | btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark); |
2ab28f32 | 2580 | btrfs_free_logged_extents(log, log_transid); |
4a500fd1 YZ |
2581 | mutex_unlock(&log_root_tree->log_mutex); |
2582 | ret = -EAGAIN; | |
2583 | goto out; | |
2584 | } | |
2585 | ||
d1433deb MX |
2586 | if (log_root_tree->log_transid_committed >= root_log_ctx.log_transid) { |
2587 | mutex_unlock(&log_root_tree->log_mutex); | |
2588 | ret = root_log_ctx.log_ret; | |
2589 | goto out; | |
2590 | } | |
8b050d35 | 2591 | |
d1433deb | 2592 | index2 = root_log_ctx.log_transid % 2; |
7237f183 | 2593 | if (atomic_read(&log_root_tree->log_commit[index2])) { |
c6adc9cc | 2594 | blk_finish_plug(&plug); |
8cef4e16 | 2595 | btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark); |
8b050d35 | 2596 | wait_log_commit(trans, log_root_tree, |
d1433deb | 2597 | root_log_ctx.log_transid); |
2ab28f32 | 2598 | btrfs_free_logged_extents(log, log_transid); |
7237f183 | 2599 | mutex_unlock(&log_root_tree->log_mutex); |
8b050d35 | 2600 | ret = root_log_ctx.log_ret; |
7237f183 YZ |
2601 | goto out; |
2602 | } | |
d1433deb | 2603 | ASSERT(root_log_ctx.log_transid == log_root_tree->log_transid); |
7237f183 YZ |
2604 | atomic_set(&log_root_tree->log_commit[index2], 1); |
2605 | ||
12fcfd22 CM |
2606 | if (atomic_read(&log_root_tree->log_commit[(index2 + 1) % 2])) { |
2607 | wait_log_commit(trans, log_root_tree, | |
d1433deb | 2608 | root_log_ctx.log_transid - 1); |
12fcfd22 CM |
2609 | } |
2610 | ||
2611 | wait_for_writer(trans, log_root_tree); | |
7237f183 | 2612 | |
12fcfd22 CM |
2613 | /* |
2614 | * now that we've moved on to the tree of log tree roots, | |
2615 | * check the full commit flag again | |
2616 | */ | |
5c902ba6 MX |
2617 | if (ACCESS_ONCE(root->fs_info->last_trans_log_full_commit) == |
2618 | trans->transid) { | |
c6adc9cc | 2619 | blk_finish_plug(&plug); |
8cef4e16 | 2620 | btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark); |
2ab28f32 | 2621 | btrfs_free_logged_extents(log, log_transid); |
12fcfd22 CM |
2622 | mutex_unlock(&log_root_tree->log_mutex); |
2623 | ret = -EAGAIN; | |
2624 | goto out_wake_log_root; | |
2625 | } | |
7237f183 | 2626 | |
c6adc9cc MX |
2627 | ret = btrfs_write_marked_extents(log_root_tree, |
2628 | &log_root_tree->dirty_log_pages, | |
2629 | EXTENT_DIRTY | EXTENT_NEW); | |
2630 | blk_finish_plug(&plug); | |
79787eaa JM |
2631 | if (ret) { |
2632 | btrfs_abort_transaction(trans, root, ret); | |
2ab28f32 | 2633 | btrfs_free_logged_extents(log, log_transid); |
79787eaa JM |
2634 | mutex_unlock(&log_root_tree->log_mutex); |
2635 | goto out_wake_log_root; | |
2636 | } | |
8cef4e16 | 2637 | btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark); |
c6adc9cc MX |
2638 | btrfs_wait_marked_extents(log_root_tree, |
2639 | &log_root_tree->dirty_log_pages, | |
2640 | EXTENT_NEW | EXTENT_DIRTY); | |
2ab28f32 | 2641 | btrfs_wait_logged_extents(log, log_transid); |
e02119d5 | 2642 | |
6c41761f | 2643 | btrfs_set_super_log_root(root->fs_info->super_for_commit, |
7237f183 | 2644 | log_root_tree->node->start); |
6c41761f | 2645 | btrfs_set_super_log_root_level(root->fs_info->super_for_commit, |
7237f183 | 2646 | btrfs_header_level(log_root_tree->node)); |
e02119d5 | 2647 | |
7237f183 | 2648 | log_root_tree->log_transid++; |
7237f183 YZ |
2649 | mutex_unlock(&log_root_tree->log_mutex); |
2650 | ||
2651 | /* | |
2652 | * nobody else is going to jump in and write the the ctree | |
2653 | * super here because the log_commit atomic below is protecting | |
2654 | * us. We must be called with a transaction handle pinning | |
2655 | * the running transaction open, so a full commit can't hop | |
2656 | * in and cause problems either. | |
2657 | */ | |
5af3e8cc | 2658 | ret = write_ctree_super(trans, root->fs_info->tree_root, 1); |
5af3e8cc SB |
2659 | if (ret) { |
2660 | btrfs_abort_transaction(trans, root, ret); | |
2661 | goto out_wake_log_root; | |
2662 | } | |
7237f183 | 2663 | |
257c62e1 CM |
2664 | mutex_lock(&root->log_mutex); |
2665 | if (root->last_log_commit < log_transid) | |
2666 | root->last_log_commit = log_transid; | |
2667 | mutex_unlock(&root->log_mutex); | |
2668 | ||
12fcfd22 | 2669 | out_wake_log_root: |
8b050d35 MX |
2670 | /* |
2671 | * We needn't get log_mutex here because we are sure all | |
2672 | * the other tasks are blocked. | |
2673 | */ | |
2674 | btrfs_remove_all_log_ctxs(log_root_tree, index2, ret); | |
2675 | ||
d1433deb MX |
2676 | mutex_lock(&log_root_tree->log_mutex); |
2677 | log_root_tree->log_transid_committed++; | |
7237f183 | 2678 | atomic_set(&log_root_tree->log_commit[index2], 0); |
d1433deb MX |
2679 | mutex_unlock(&log_root_tree->log_mutex); |
2680 | ||
7237f183 YZ |
2681 | if (waitqueue_active(&log_root_tree->log_commit_wait[index2])) |
2682 | wake_up(&log_root_tree->log_commit_wait[index2]); | |
e02119d5 | 2683 | out: |
8b050d35 MX |
2684 | /* See above. */ |
2685 | btrfs_remove_all_log_ctxs(root, index1, ret); | |
2686 | ||
d1433deb MX |
2687 | mutex_lock(&root->log_mutex); |
2688 | root->log_transid_committed++; | |
7237f183 | 2689 | atomic_set(&root->log_commit[index1], 0); |
d1433deb | 2690 | mutex_unlock(&root->log_mutex); |
8b050d35 | 2691 | |
7237f183 YZ |
2692 | if (waitqueue_active(&root->log_commit_wait[index1])) |
2693 | wake_up(&root->log_commit_wait[index1]); | |
b31eabd8 | 2694 | return ret; |
e02119d5 CM |
2695 | } |
2696 | ||
4a500fd1 YZ |
2697 | static void free_log_tree(struct btrfs_trans_handle *trans, |
2698 | struct btrfs_root *log) | |
e02119d5 CM |
2699 | { |
2700 | int ret; | |
d0c803c4 CM |
2701 | u64 start; |
2702 | u64 end; | |
e02119d5 CM |
2703 | struct walk_control wc = { |
2704 | .free = 1, | |
2705 | .process_func = process_one_buffer | |
2706 | }; | |
2707 | ||
681ae509 JB |
2708 | ret = walk_log_tree(trans, log, &wc); |
2709 | /* I don't think this can happen but just in case */ | |
2710 | if (ret) | |
2711 | btrfs_abort_transaction(trans, log, ret); | |
e02119d5 | 2712 | |
d397712b | 2713 | while (1) { |
d0c803c4 | 2714 | ret = find_first_extent_bit(&log->dirty_log_pages, |
e6138876 JB |
2715 | 0, &start, &end, EXTENT_DIRTY | EXTENT_NEW, |
2716 | NULL); | |
d0c803c4 CM |
2717 | if (ret) |
2718 | break; | |
2719 | ||
8cef4e16 YZ |
2720 | clear_extent_bits(&log->dirty_log_pages, start, end, |
2721 | EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS); | |
d0c803c4 CM |
2722 | } |
2723 | ||
2ab28f32 JB |
2724 | /* |
2725 | * We may have short-circuited the log tree with the full commit logic | |
2726 | * and left ordered extents on our list, so clear these out to keep us | |
2727 | * from leaking inodes and memory. | |
2728 | */ | |
2729 | btrfs_free_logged_extents(log, 0); | |
2730 | btrfs_free_logged_extents(log, 1); | |
2731 | ||
7237f183 YZ |
2732 | free_extent_buffer(log->node); |
2733 | kfree(log); | |
4a500fd1 YZ |
2734 | } |
2735 | ||
2736 | /* | |
2737 | * free all the extents used by the tree log. This should be called | |
2738 | * at commit time of the full transaction | |
2739 | */ | |
2740 | int btrfs_free_log(struct btrfs_trans_handle *trans, struct btrfs_root *root) | |
2741 | { | |
2742 | if (root->log_root) { | |
2743 | free_log_tree(trans, root->log_root); | |
2744 | root->log_root = NULL; | |
2745 | } | |
2746 | return 0; | |
2747 | } | |
2748 | ||
2749 | int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans, | |
2750 | struct btrfs_fs_info *fs_info) | |
2751 | { | |
2752 | if (fs_info->log_root_tree) { | |
2753 | free_log_tree(trans, fs_info->log_root_tree); | |
2754 | fs_info->log_root_tree = NULL; | |
2755 | } | |
e02119d5 CM |
2756 | return 0; |
2757 | } | |
2758 | ||
e02119d5 CM |
2759 | /* |
2760 | * If both a file and directory are logged, and unlinks or renames are | |
2761 | * mixed in, we have a few interesting corners: | |
2762 | * | |
2763 | * create file X in dir Y | |
2764 | * link file X to X.link in dir Y | |
2765 | * fsync file X | |
2766 | * unlink file X but leave X.link | |
2767 | * fsync dir Y | |
2768 | * | |
2769 | * After a crash we would expect only X.link to exist. But file X | |
2770 | * didn't get fsync'd again so the log has back refs for X and X.link. | |
2771 | * | |
2772 | * We solve this by removing directory entries and inode backrefs from the | |
2773 | * log when a file that was logged in the current transaction is | |
2774 | * unlinked. Any later fsync will include the updated log entries, and | |
2775 | * we'll be able to reconstruct the proper directory items from backrefs. | |
2776 | * | |
2777 | * This optimizations allows us to avoid relogging the entire inode | |
2778 | * or the entire directory. | |
2779 | */ | |
2780 | int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans, | |
2781 | struct btrfs_root *root, | |
2782 | const char *name, int name_len, | |
2783 | struct inode *dir, u64 index) | |
2784 | { | |
2785 | struct btrfs_root *log; | |
2786 | struct btrfs_dir_item *di; | |
2787 | struct btrfs_path *path; | |
2788 | int ret; | |
4a500fd1 | 2789 | int err = 0; |
e02119d5 | 2790 | int bytes_del = 0; |
33345d01 | 2791 | u64 dir_ino = btrfs_ino(dir); |
e02119d5 | 2792 | |
3a5f1d45 CM |
2793 | if (BTRFS_I(dir)->logged_trans < trans->transid) |
2794 | return 0; | |
2795 | ||
e02119d5 CM |
2796 | ret = join_running_log_trans(root); |
2797 | if (ret) | |
2798 | return 0; | |
2799 | ||
2800 | mutex_lock(&BTRFS_I(dir)->log_mutex); | |
2801 | ||
2802 | log = root->log_root; | |
2803 | path = btrfs_alloc_path(); | |
a62f44a5 TI |
2804 | if (!path) { |
2805 | err = -ENOMEM; | |
2806 | goto out_unlock; | |
2807 | } | |
2a29edc6 | 2808 | |
33345d01 | 2809 | di = btrfs_lookup_dir_item(trans, log, path, dir_ino, |
e02119d5 | 2810 | name, name_len, -1); |
4a500fd1 YZ |
2811 | if (IS_ERR(di)) { |
2812 | err = PTR_ERR(di); | |
2813 | goto fail; | |
2814 | } | |
2815 | if (di) { | |
e02119d5 CM |
2816 | ret = btrfs_delete_one_dir_name(trans, log, path, di); |
2817 | bytes_del += name_len; | |
3650860b JB |
2818 | if (ret) { |
2819 | err = ret; | |
2820 | goto fail; | |
2821 | } | |
e02119d5 | 2822 | } |
b3b4aa74 | 2823 | btrfs_release_path(path); |
33345d01 | 2824 | di = btrfs_lookup_dir_index_item(trans, log, path, dir_ino, |
e02119d5 | 2825 | index, name, name_len, -1); |
4a500fd1 YZ |
2826 | if (IS_ERR(di)) { |
2827 | err = PTR_ERR(di); | |
2828 | goto fail; | |
2829 | } | |
2830 | if (di) { | |
e02119d5 CM |
2831 | ret = btrfs_delete_one_dir_name(trans, log, path, di); |
2832 | bytes_del += name_len; | |
3650860b JB |
2833 | if (ret) { |
2834 | err = ret; | |
2835 | goto fail; | |
2836 | } | |
e02119d5 CM |
2837 | } |
2838 | ||
2839 | /* update the directory size in the log to reflect the names | |
2840 | * we have removed | |
2841 | */ | |
2842 | if (bytes_del) { | |
2843 | struct btrfs_key key; | |
2844 | ||
33345d01 | 2845 | key.objectid = dir_ino; |
e02119d5 CM |
2846 | key.offset = 0; |
2847 | key.type = BTRFS_INODE_ITEM_KEY; | |
b3b4aa74 | 2848 | btrfs_release_path(path); |
e02119d5 CM |
2849 | |
2850 | ret = btrfs_search_slot(trans, log, &key, path, 0, 1); | |
4a500fd1 YZ |
2851 | if (ret < 0) { |
2852 | err = ret; | |
2853 | goto fail; | |
2854 | } | |
e02119d5 CM |
2855 | if (ret == 0) { |
2856 | struct btrfs_inode_item *item; | |
2857 | u64 i_size; | |
2858 | ||
2859 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
2860 | struct btrfs_inode_item); | |
2861 | i_size = btrfs_inode_size(path->nodes[0], item); | |
2862 | if (i_size > bytes_del) | |
2863 | i_size -= bytes_del; | |
2864 | else | |
2865 | i_size = 0; | |
2866 | btrfs_set_inode_size(path->nodes[0], item, i_size); | |
2867 | btrfs_mark_buffer_dirty(path->nodes[0]); | |
2868 | } else | |
2869 | ret = 0; | |
b3b4aa74 | 2870 | btrfs_release_path(path); |
e02119d5 | 2871 | } |
4a500fd1 | 2872 | fail: |
e02119d5 | 2873 | btrfs_free_path(path); |
a62f44a5 | 2874 | out_unlock: |
e02119d5 | 2875 | mutex_unlock(&BTRFS_I(dir)->log_mutex); |
4a500fd1 YZ |
2876 | if (ret == -ENOSPC) { |
2877 | root->fs_info->last_trans_log_full_commit = trans->transid; | |
2878 | ret = 0; | |
79787eaa JM |
2879 | } else if (ret < 0) |
2880 | btrfs_abort_transaction(trans, root, ret); | |
2881 | ||
12fcfd22 | 2882 | btrfs_end_log_trans(root); |
e02119d5 | 2883 | |
411fc6bc | 2884 | return err; |
e02119d5 CM |
2885 | } |
2886 | ||
2887 | /* see comments for btrfs_del_dir_entries_in_log */ | |
2888 | int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans, | |
2889 | struct btrfs_root *root, | |
2890 | const char *name, int name_len, | |
2891 | struct inode *inode, u64 dirid) | |
2892 | { | |
2893 | struct btrfs_root *log; | |
2894 | u64 index; | |
2895 | int ret; | |
2896 | ||
3a5f1d45 CM |
2897 | if (BTRFS_I(inode)->logged_trans < trans->transid) |
2898 | return 0; | |
2899 | ||
e02119d5 CM |
2900 | ret = join_running_log_trans(root); |
2901 | if (ret) | |
2902 | return 0; | |
2903 | log = root->log_root; | |
2904 | mutex_lock(&BTRFS_I(inode)->log_mutex); | |
2905 | ||
33345d01 | 2906 | ret = btrfs_del_inode_ref(trans, log, name, name_len, btrfs_ino(inode), |
e02119d5 CM |
2907 | dirid, &index); |
2908 | mutex_unlock(&BTRFS_I(inode)->log_mutex); | |
4a500fd1 YZ |
2909 | if (ret == -ENOSPC) { |
2910 | root->fs_info->last_trans_log_full_commit = trans->transid; | |
2911 | ret = 0; | |
79787eaa JM |
2912 | } else if (ret < 0 && ret != -ENOENT) |
2913 | btrfs_abort_transaction(trans, root, ret); | |
12fcfd22 | 2914 | btrfs_end_log_trans(root); |
e02119d5 | 2915 | |
e02119d5 CM |
2916 | return ret; |
2917 | } | |
2918 | ||
2919 | /* | |
2920 | * creates a range item in the log for 'dirid'. first_offset and | |
2921 | * last_offset tell us which parts of the key space the log should | |
2922 | * be considered authoritative for. | |
2923 | */ | |
2924 | static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans, | |
2925 | struct btrfs_root *log, | |
2926 | struct btrfs_path *path, | |
2927 | int key_type, u64 dirid, | |
2928 | u64 first_offset, u64 last_offset) | |
2929 | { | |
2930 | int ret; | |
2931 | struct btrfs_key key; | |
2932 | struct btrfs_dir_log_item *item; | |
2933 | ||
2934 | key.objectid = dirid; | |
2935 | key.offset = first_offset; | |
2936 | if (key_type == BTRFS_DIR_ITEM_KEY) | |
2937 | key.type = BTRFS_DIR_LOG_ITEM_KEY; | |
2938 | else | |
2939 | key.type = BTRFS_DIR_LOG_INDEX_KEY; | |
2940 | ret = btrfs_insert_empty_item(trans, log, path, &key, sizeof(*item)); | |
4a500fd1 YZ |
2941 | if (ret) |
2942 | return ret; | |
e02119d5 CM |
2943 | |
2944 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
2945 | struct btrfs_dir_log_item); | |
2946 | btrfs_set_dir_log_end(path->nodes[0], item, last_offset); | |
2947 | btrfs_mark_buffer_dirty(path->nodes[0]); | |
b3b4aa74 | 2948 | btrfs_release_path(path); |
e02119d5 CM |
2949 | return 0; |
2950 | } | |
2951 | ||
2952 | /* | |
2953 | * log all the items included in the current transaction for a given | |
2954 | * directory. This also creates the range items in the log tree required | |
2955 | * to replay anything deleted before the fsync | |
2956 | */ | |
2957 | static noinline int log_dir_items(struct btrfs_trans_handle *trans, | |
2958 | struct btrfs_root *root, struct inode *inode, | |
2959 | struct btrfs_path *path, | |
2960 | struct btrfs_path *dst_path, int key_type, | |
2961 | u64 min_offset, u64 *last_offset_ret) | |
2962 | { | |
2963 | struct btrfs_key min_key; | |
e02119d5 CM |
2964 | struct btrfs_root *log = root->log_root; |
2965 | struct extent_buffer *src; | |
4a500fd1 | 2966 | int err = 0; |
e02119d5 CM |
2967 | int ret; |
2968 | int i; | |
2969 | int nritems; | |
2970 | u64 first_offset = min_offset; | |
2971 | u64 last_offset = (u64)-1; | |
33345d01 | 2972 | u64 ino = btrfs_ino(inode); |
e02119d5 CM |
2973 | |
2974 | log = root->log_root; | |
e02119d5 | 2975 | |
33345d01 | 2976 | min_key.objectid = ino; |
e02119d5 CM |
2977 | min_key.type = key_type; |
2978 | min_key.offset = min_offset; | |
2979 | ||
2980 | path->keep_locks = 1; | |
2981 | ||
6174d3cb | 2982 | ret = btrfs_search_forward(root, &min_key, path, trans->transid); |
e02119d5 CM |
2983 | |
2984 | /* | |
2985 | * we didn't find anything from this transaction, see if there | |
2986 | * is anything at all | |
2987 | */ | |
33345d01 LZ |
2988 | if (ret != 0 || min_key.objectid != ino || min_key.type != key_type) { |
2989 | min_key.objectid = ino; | |
e02119d5 CM |
2990 | min_key.type = key_type; |
2991 | min_key.offset = (u64)-1; | |
b3b4aa74 | 2992 | btrfs_release_path(path); |
e02119d5 CM |
2993 | ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0); |
2994 | if (ret < 0) { | |
b3b4aa74 | 2995 | btrfs_release_path(path); |
e02119d5 CM |
2996 | return ret; |
2997 | } | |
33345d01 | 2998 | ret = btrfs_previous_item(root, path, ino, key_type); |
e02119d5 CM |
2999 | |
3000 | /* if ret == 0 there are items for this type, | |
3001 | * create a range to tell us the last key of this type. | |
3002 | * otherwise, there are no items in this directory after | |
3003 | * *min_offset, and we create a range to indicate that. | |
3004 | */ | |
3005 | if (ret == 0) { | |
3006 | struct btrfs_key tmp; | |
3007 | btrfs_item_key_to_cpu(path->nodes[0], &tmp, | |
3008 | path->slots[0]); | |
d397712b | 3009 | if (key_type == tmp.type) |
e02119d5 | 3010 | first_offset = max(min_offset, tmp.offset) + 1; |
e02119d5 CM |
3011 | } |
3012 | goto done; | |
3013 | } | |
3014 | ||
3015 | /* go backward to find any previous key */ | |
33345d01 | 3016 | ret = btrfs_previous_item(root, path, ino, key_type); |
e02119d5 CM |
3017 | if (ret == 0) { |
3018 | struct btrfs_key tmp; | |
3019 | btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]); | |
3020 | if (key_type == tmp.type) { | |
3021 | first_offset = tmp.offset; | |
3022 | ret = overwrite_item(trans, log, dst_path, | |
3023 | path->nodes[0], path->slots[0], | |
3024 | &tmp); | |
4a500fd1 YZ |
3025 | if (ret) { |
3026 | err = ret; | |
3027 | goto done; | |
3028 | } | |
e02119d5 CM |
3029 | } |
3030 | } | |
b3b4aa74 | 3031 | btrfs_release_path(path); |
e02119d5 CM |
3032 | |
3033 | /* find the first key from this transaction again */ | |
3034 | ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0); | |
fae7f21c | 3035 | if (WARN_ON(ret != 0)) |
e02119d5 | 3036 | goto done; |
e02119d5 CM |
3037 | |
3038 | /* | |
3039 | * we have a block from this transaction, log every item in it | |
3040 | * from our directory | |
3041 | */ | |
d397712b | 3042 | while (1) { |
e02119d5 CM |
3043 | struct btrfs_key tmp; |
3044 | src = path->nodes[0]; | |
3045 | nritems = btrfs_header_nritems(src); | |
3046 | for (i = path->slots[0]; i < nritems; i++) { | |
3047 | btrfs_item_key_to_cpu(src, &min_key, i); | |
3048 | ||
33345d01 | 3049 | if (min_key.objectid != ino || min_key.type != key_type) |
e02119d5 CM |
3050 | goto done; |
3051 | ret = overwrite_item(trans, log, dst_path, src, i, | |
3052 | &min_key); | |
4a500fd1 YZ |
3053 | if (ret) { |
3054 | err = ret; | |
3055 | goto done; | |
3056 | } | |
e02119d5 CM |
3057 | } |
3058 | path->slots[0] = nritems; | |
3059 | ||
3060 | /* | |
3061 | * look ahead to the next item and see if it is also | |
3062 | * from this directory and from this transaction | |
3063 | */ | |
3064 | ret = btrfs_next_leaf(root, path); | |
3065 | if (ret == 1) { | |
3066 | last_offset = (u64)-1; | |
3067 | goto done; | |
3068 | } | |
3069 | btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]); | |
33345d01 | 3070 | if (tmp.objectid != ino || tmp.type != key_type) { |
e02119d5 CM |
3071 | last_offset = (u64)-1; |
3072 | goto done; | |
3073 | } | |
3074 | if (btrfs_header_generation(path->nodes[0]) != trans->transid) { | |
3075 | ret = overwrite_item(trans, log, dst_path, | |
3076 | path->nodes[0], path->slots[0], | |
3077 | &tmp); | |
4a500fd1 YZ |
3078 | if (ret) |
3079 | err = ret; | |
3080 | else | |
3081 | last_offset = tmp.offset; | |
e02119d5 CM |
3082 | goto done; |
3083 | } | |
3084 | } | |
3085 | done: | |
b3b4aa74 DS |
3086 | btrfs_release_path(path); |
3087 | btrfs_release_path(dst_path); | |
e02119d5 | 3088 | |
4a500fd1 YZ |
3089 | if (err == 0) { |
3090 | *last_offset_ret = last_offset; | |
3091 | /* | |
3092 | * insert the log range keys to indicate where the log | |
3093 | * is valid | |
3094 | */ | |
3095 | ret = insert_dir_log_key(trans, log, path, key_type, | |
33345d01 | 3096 | ino, first_offset, last_offset); |
4a500fd1 YZ |
3097 | if (ret) |
3098 | err = ret; | |
3099 | } | |
3100 | return err; | |
e02119d5 CM |
3101 | } |
3102 | ||
3103 | /* | |
3104 | * logging directories is very similar to logging inodes, We find all the items | |
3105 | * from the current transaction and write them to the log. | |
3106 | * | |
3107 | * The recovery code scans the directory in the subvolume, and if it finds a | |
3108 | * key in the range logged that is not present in the log tree, then it means | |
3109 | * that dir entry was unlinked during the transaction. | |
3110 | * | |
3111 | * In order for that scan to work, we must include one key smaller than | |
3112 | * the smallest logged by this transaction and one key larger than the largest | |
3113 | * key logged by this transaction. | |
3114 | */ | |
3115 | static noinline int log_directory_changes(struct btrfs_trans_handle *trans, | |
3116 | struct btrfs_root *root, struct inode *inode, | |
3117 | struct btrfs_path *path, | |
3118 | struct btrfs_path *dst_path) | |
3119 | { | |
3120 | u64 min_key; | |
3121 | u64 max_key; | |
3122 | int ret; | |
3123 | int key_type = BTRFS_DIR_ITEM_KEY; | |
3124 | ||
3125 | again: | |
3126 | min_key = 0; | |
3127 | max_key = 0; | |
d397712b | 3128 | while (1) { |
e02119d5 CM |
3129 | ret = log_dir_items(trans, root, inode, path, |
3130 | dst_path, key_type, min_key, | |
3131 | &max_key); | |
4a500fd1 YZ |
3132 | if (ret) |
3133 | return ret; | |
e02119d5 CM |
3134 | if (max_key == (u64)-1) |
3135 | break; | |
3136 | min_key = max_key + 1; | |
3137 | } | |
3138 | ||
3139 | if (key_type == BTRFS_DIR_ITEM_KEY) { | |
3140 | key_type = BTRFS_DIR_INDEX_KEY; | |
3141 | goto again; | |
3142 | } | |
3143 | return 0; | |
3144 | } | |
3145 | ||
3146 | /* | |
3147 | * a helper function to drop items from the log before we relog an | |
3148 | * inode. max_key_type indicates the highest item type to remove. | |
3149 | * This cannot be run for file data extents because it does not | |
3150 | * free the extents they point to. | |
3151 | */ | |
3152 | static int drop_objectid_items(struct btrfs_trans_handle *trans, | |
3153 | struct btrfs_root *log, | |
3154 | struct btrfs_path *path, | |
3155 | u64 objectid, int max_key_type) | |
3156 | { | |
3157 | int ret; | |
3158 | struct btrfs_key key; | |
3159 | struct btrfs_key found_key; | |
18ec90d6 | 3160 | int start_slot; |
e02119d5 CM |
3161 | |
3162 | key.objectid = objectid; | |
3163 | key.type = max_key_type; | |
3164 | key.offset = (u64)-1; | |
3165 | ||
d397712b | 3166 | while (1) { |
e02119d5 | 3167 | ret = btrfs_search_slot(trans, log, &key, path, -1, 1); |
3650860b | 3168 | BUG_ON(ret == 0); /* Logic error */ |
4a500fd1 | 3169 | if (ret < 0) |
e02119d5 CM |
3170 | break; |
3171 | ||
3172 | if (path->slots[0] == 0) | |
3173 | break; | |
3174 | ||
3175 | path->slots[0]--; | |
3176 | btrfs_item_key_to_cpu(path->nodes[0], &found_key, | |
3177 | path->slots[0]); | |
3178 | ||
3179 | if (found_key.objectid != objectid) | |
3180 | break; | |
3181 | ||
18ec90d6 JB |
3182 | found_key.offset = 0; |
3183 | found_key.type = 0; | |
3184 | ret = btrfs_bin_search(path->nodes[0], &found_key, 0, | |
3185 | &start_slot); | |
3186 | ||
3187 | ret = btrfs_del_items(trans, log, path, start_slot, | |
3188 | path->slots[0] - start_slot + 1); | |
3189 | /* | |
3190 | * If start slot isn't 0 then we don't need to re-search, we've | |
3191 | * found the last guy with the objectid in this tree. | |
3192 | */ | |
3193 | if (ret || start_slot != 0) | |
65a246c5 | 3194 | break; |
b3b4aa74 | 3195 | btrfs_release_path(path); |
e02119d5 | 3196 | } |
b3b4aa74 | 3197 | btrfs_release_path(path); |
5bdbeb21 JB |
3198 | if (ret > 0) |
3199 | ret = 0; | |
4a500fd1 | 3200 | return ret; |
e02119d5 CM |
3201 | } |
3202 | ||
94edf4ae JB |
3203 | static void fill_inode_item(struct btrfs_trans_handle *trans, |
3204 | struct extent_buffer *leaf, | |
3205 | struct btrfs_inode_item *item, | |
3206 | struct inode *inode, int log_inode_only) | |
3207 | { | |
0b1c6cca JB |
3208 | struct btrfs_map_token token; |
3209 | ||
3210 | btrfs_init_map_token(&token); | |
94edf4ae JB |
3211 | |
3212 | if (log_inode_only) { | |
3213 | /* set the generation to zero so the recover code | |
3214 | * can tell the difference between an logging | |
3215 | * just to say 'this inode exists' and a logging | |
3216 | * to say 'update this inode with these values' | |
3217 | */ | |
0b1c6cca JB |
3218 | btrfs_set_token_inode_generation(leaf, item, 0, &token); |
3219 | btrfs_set_token_inode_size(leaf, item, 0, &token); | |
94edf4ae | 3220 | } else { |
0b1c6cca JB |
3221 | btrfs_set_token_inode_generation(leaf, item, |
3222 | BTRFS_I(inode)->generation, | |
3223 | &token); | |
3224 | btrfs_set_token_inode_size(leaf, item, inode->i_size, &token); | |
3225 | } | |
3226 | ||
3227 | btrfs_set_token_inode_uid(leaf, item, i_uid_read(inode), &token); | |
3228 | btrfs_set_token_inode_gid(leaf, item, i_gid_read(inode), &token); | |
3229 | btrfs_set_token_inode_mode(leaf, item, inode->i_mode, &token); | |
3230 | btrfs_set_token_inode_nlink(leaf, item, inode->i_nlink, &token); | |
3231 | ||
3232 | btrfs_set_token_timespec_sec(leaf, btrfs_inode_atime(item), | |
3233 | inode->i_atime.tv_sec, &token); | |
3234 | btrfs_set_token_timespec_nsec(leaf, btrfs_inode_atime(item), | |
3235 | inode->i_atime.tv_nsec, &token); | |
3236 | ||
3237 | btrfs_set_token_timespec_sec(leaf, btrfs_inode_mtime(item), | |
3238 | inode->i_mtime.tv_sec, &token); | |
3239 | btrfs_set_token_timespec_nsec(leaf, btrfs_inode_mtime(item), | |
3240 | inode->i_mtime.tv_nsec, &token); | |
3241 | ||
3242 | btrfs_set_token_timespec_sec(leaf, btrfs_inode_ctime(item), | |
3243 | inode->i_ctime.tv_sec, &token); | |
3244 | btrfs_set_token_timespec_nsec(leaf, btrfs_inode_ctime(item), | |
3245 | inode->i_ctime.tv_nsec, &token); | |
3246 | ||
3247 | btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode), | |
3248 | &token); | |
3249 | ||
3250 | btrfs_set_token_inode_sequence(leaf, item, inode->i_version, &token); | |
3251 | btrfs_set_token_inode_transid(leaf, item, trans->transid, &token); | |
3252 | btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token); | |
3253 | btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token); | |
3254 | btrfs_set_token_inode_block_group(leaf, item, 0, &token); | |
94edf4ae JB |
3255 | } |
3256 | ||
a95249b3 JB |
3257 | static int log_inode_item(struct btrfs_trans_handle *trans, |
3258 | struct btrfs_root *log, struct btrfs_path *path, | |
3259 | struct inode *inode) | |
3260 | { | |
3261 | struct btrfs_inode_item *inode_item; | |
a95249b3 JB |
3262 | int ret; |
3263 | ||
efd0c405 FDBM |
3264 | ret = btrfs_insert_empty_item(trans, log, path, |
3265 | &BTRFS_I(inode)->location, | |
a95249b3 JB |
3266 | sizeof(*inode_item)); |
3267 | if (ret && ret != -EEXIST) | |
3268 | return ret; | |
3269 | inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
3270 | struct btrfs_inode_item); | |
3271 | fill_inode_item(trans, path->nodes[0], inode_item, inode, 0); | |
3272 | btrfs_release_path(path); | |
3273 | return 0; | |
3274 | } | |
3275 | ||
31ff1cd2 | 3276 | static noinline int copy_items(struct btrfs_trans_handle *trans, |
d2794405 | 3277 | struct inode *inode, |
31ff1cd2 | 3278 | struct btrfs_path *dst_path, |
16e7549f | 3279 | struct btrfs_path *src_path, u64 *last_extent, |
31ff1cd2 CM |
3280 | int start_slot, int nr, int inode_only) |
3281 | { | |
3282 | unsigned long src_offset; | |
3283 | unsigned long dst_offset; | |
d2794405 | 3284 | struct btrfs_root *log = BTRFS_I(inode)->root->log_root; |
31ff1cd2 CM |
3285 | struct btrfs_file_extent_item *extent; |
3286 | struct btrfs_inode_item *inode_item; | |
16e7549f JB |
3287 | struct extent_buffer *src = src_path->nodes[0]; |
3288 | struct btrfs_key first_key, last_key, key; | |
31ff1cd2 CM |
3289 | int ret; |
3290 | struct btrfs_key *ins_keys; | |
3291 | u32 *ins_sizes; | |
3292 | char *ins_data; | |
3293 | int i; | |
d20f7043 | 3294 | struct list_head ordered_sums; |
d2794405 | 3295 | int skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; |
16e7549f JB |
3296 | bool has_extents = false; |
3297 | bool need_find_last_extent = (*last_extent == 0); | |
3298 | bool done = false; | |
d20f7043 CM |
3299 | |
3300 | INIT_LIST_HEAD(&ordered_sums); | |
31ff1cd2 CM |
3301 | |
3302 | ins_data = kmalloc(nr * sizeof(struct btrfs_key) + | |
3303 | nr * sizeof(u32), GFP_NOFS); | |
2a29edc6 | 3304 | if (!ins_data) |
3305 | return -ENOMEM; | |
3306 | ||
16e7549f JB |
3307 | first_key.objectid = (u64)-1; |
3308 | ||
31ff1cd2 CM |
3309 | ins_sizes = (u32 *)ins_data; |
3310 | ins_keys = (struct btrfs_key *)(ins_data + nr * sizeof(u32)); | |
3311 | ||
3312 | for (i = 0; i < nr; i++) { | |
3313 | ins_sizes[i] = btrfs_item_size_nr(src, i + start_slot); | |
3314 | btrfs_item_key_to_cpu(src, ins_keys + i, i + start_slot); | |
3315 | } | |
3316 | ret = btrfs_insert_empty_items(trans, log, dst_path, | |
3317 | ins_keys, ins_sizes, nr); | |
4a500fd1 YZ |
3318 | if (ret) { |
3319 | kfree(ins_data); | |
3320 | return ret; | |
3321 | } | |
31ff1cd2 | 3322 | |
5d4f98a2 | 3323 | for (i = 0; i < nr; i++, dst_path->slots[0]++) { |
31ff1cd2 CM |
3324 | dst_offset = btrfs_item_ptr_offset(dst_path->nodes[0], |
3325 | dst_path->slots[0]); | |
3326 | ||
3327 | src_offset = btrfs_item_ptr_offset(src, start_slot + i); | |
3328 | ||
16e7549f JB |
3329 | if ((i == (nr - 1))) |
3330 | last_key = ins_keys[i]; | |
3331 | ||
94edf4ae | 3332 | if (ins_keys[i].type == BTRFS_INODE_ITEM_KEY) { |
31ff1cd2 CM |
3333 | inode_item = btrfs_item_ptr(dst_path->nodes[0], |
3334 | dst_path->slots[0], | |
3335 | struct btrfs_inode_item); | |
94edf4ae JB |
3336 | fill_inode_item(trans, dst_path->nodes[0], inode_item, |
3337 | inode, inode_only == LOG_INODE_EXISTS); | |
3338 | } else { | |
3339 | copy_extent_buffer(dst_path->nodes[0], src, dst_offset, | |
3340 | src_offset, ins_sizes[i]); | |
31ff1cd2 | 3341 | } |
94edf4ae | 3342 | |
16e7549f JB |
3343 | /* |
3344 | * We set need_find_last_extent here in case we know we were | |
3345 | * processing other items and then walk into the first extent in | |
3346 | * the inode. If we don't hit an extent then nothing changes, | |
3347 | * we'll do the last search the next time around. | |
3348 | */ | |
3349 | if (ins_keys[i].type == BTRFS_EXTENT_DATA_KEY) { | |
3350 | has_extents = true; | |
3351 | if (need_find_last_extent && | |
3352 | first_key.objectid == (u64)-1) | |
3353 | first_key = ins_keys[i]; | |
3354 | } else { | |
3355 | need_find_last_extent = false; | |
3356 | } | |
3357 | ||
31ff1cd2 CM |
3358 | /* take a reference on file data extents so that truncates |
3359 | * or deletes of this inode don't have to relog the inode | |
3360 | * again | |
3361 | */ | |
d2794405 LB |
3362 | if (btrfs_key_type(ins_keys + i) == BTRFS_EXTENT_DATA_KEY && |
3363 | !skip_csum) { | |
31ff1cd2 CM |
3364 | int found_type; |
3365 | extent = btrfs_item_ptr(src, start_slot + i, | |
3366 | struct btrfs_file_extent_item); | |
3367 | ||
8e531cdf | 3368 | if (btrfs_file_extent_generation(src, extent) < trans->transid) |
3369 | continue; | |
3370 | ||
31ff1cd2 | 3371 | found_type = btrfs_file_extent_type(src, extent); |
6f1fed77 | 3372 | if (found_type == BTRFS_FILE_EXTENT_REG) { |
5d4f98a2 YZ |
3373 | u64 ds, dl, cs, cl; |
3374 | ds = btrfs_file_extent_disk_bytenr(src, | |
3375 | extent); | |
3376 | /* ds == 0 is a hole */ | |
3377 | if (ds == 0) | |
3378 | continue; | |
3379 | ||
3380 | dl = btrfs_file_extent_disk_num_bytes(src, | |
3381 | extent); | |
3382 | cs = btrfs_file_extent_offset(src, extent); | |
3383 | cl = btrfs_file_extent_num_bytes(src, | |
a419aef8 | 3384 | extent); |
580afd76 CM |
3385 | if (btrfs_file_extent_compression(src, |
3386 | extent)) { | |
3387 | cs = 0; | |
3388 | cl = dl; | |
3389 | } | |
5d4f98a2 YZ |
3390 | |
3391 | ret = btrfs_lookup_csums_range( | |
3392 | log->fs_info->csum_root, | |
3393 | ds + cs, ds + cs + cl - 1, | |
a2de733c | 3394 | &ordered_sums, 0); |
3650860b JB |
3395 | if (ret) { |
3396 | btrfs_release_path(dst_path); | |
3397 | kfree(ins_data); | |
3398 | return ret; | |
3399 | } | |
31ff1cd2 CM |
3400 | } |
3401 | } | |
31ff1cd2 CM |
3402 | } |
3403 | ||
3404 | btrfs_mark_buffer_dirty(dst_path->nodes[0]); | |
b3b4aa74 | 3405 | btrfs_release_path(dst_path); |
31ff1cd2 | 3406 | kfree(ins_data); |
d20f7043 CM |
3407 | |
3408 | /* | |
3409 | * we have to do this after the loop above to avoid changing the | |
3410 | * log tree while trying to change the log tree. | |
3411 | */ | |
4a500fd1 | 3412 | ret = 0; |
d397712b | 3413 | while (!list_empty(&ordered_sums)) { |
d20f7043 CM |
3414 | struct btrfs_ordered_sum *sums = list_entry(ordered_sums.next, |
3415 | struct btrfs_ordered_sum, | |
3416 | list); | |
4a500fd1 YZ |
3417 | if (!ret) |
3418 | ret = btrfs_csum_file_blocks(trans, log, sums); | |
d20f7043 CM |
3419 | list_del(&sums->list); |
3420 | kfree(sums); | |
3421 | } | |
16e7549f JB |
3422 | |
3423 | if (!has_extents) | |
3424 | return ret; | |
3425 | ||
3426 | /* | |
3427 | * Because we use btrfs_search_forward we could skip leaves that were | |
3428 | * not modified and then assume *last_extent is valid when it really | |
3429 | * isn't. So back up to the previous leaf and read the end of the last | |
3430 | * extent before we go and fill in holes. | |
3431 | */ | |
3432 | if (need_find_last_extent) { | |
3433 | u64 len; | |
3434 | ||
3435 | ret = btrfs_prev_leaf(BTRFS_I(inode)->root, src_path); | |
3436 | if (ret < 0) | |
3437 | return ret; | |
3438 | if (ret) | |
3439 | goto fill_holes; | |
3440 | if (src_path->slots[0]) | |
3441 | src_path->slots[0]--; | |
3442 | src = src_path->nodes[0]; | |
3443 | btrfs_item_key_to_cpu(src, &key, src_path->slots[0]); | |
3444 | if (key.objectid != btrfs_ino(inode) || | |
3445 | key.type != BTRFS_EXTENT_DATA_KEY) | |
3446 | goto fill_holes; | |
3447 | extent = btrfs_item_ptr(src, src_path->slots[0], | |
3448 | struct btrfs_file_extent_item); | |
3449 | if (btrfs_file_extent_type(src, extent) == | |
3450 | BTRFS_FILE_EXTENT_INLINE) { | |
514ac8ad CM |
3451 | len = btrfs_file_extent_inline_len(src, |
3452 | src_path->slots[0], | |
3453 | extent); | |
16e7549f JB |
3454 | *last_extent = ALIGN(key.offset + len, |
3455 | log->sectorsize); | |
3456 | } else { | |
3457 | len = btrfs_file_extent_num_bytes(src, extent); | |
3458 | *last_extent = key.offset + len; | |
3459 | } | |
3460 | } | |
3461 | fill_holes: | |
3462 | /* So we did prev_leaf, now we need to move to the next leaf, but a few | |
3463 | * things could have happened | |
3464 | * | |
3465 | * 1) A merge could have happened, so we could currently be on a leaf | |
3466 | * that holds what we were copying in the first place. | |
3467 | * 2) A split could have happened, and now not all of the items we want | |
3468 | * are on the same leaf. | |
3469 | * | |
3470 | * So we need to adjust how we search for holes, we need to drop the | |
3471 | * path and re-search for the first extent key we found, and then walk | |
3472 | * forward until we hit the last one we copied. | |
3473 | */ | |
3474 | if (need_find_last_extent) { | |
3475 | /* btrfs_prev_leaf could return 1 without releasing the path */ | |
3476 | btrfs_release_path(src_path); | |
3477 | ret = btrfs_search_slot(NULL, BTRFS_I(inode)->root, &first_key, | |
3478 | src_path, 0, 0); | |
3479 | if (ret < 0) | |
3480 | return ret; | |
3481 | ASSERT(ret == 0); | |
3482 | src = src_path->nodes[0]; | |
3483 | i = src_path->slots[0]; | |
3484 | } else { | |
3485 | i = start_slot; | |
3486 | } | |
3487 | ||
3488 | /* | |
3489 | * Ok so here we need to go through and fill in any holes we may have | |
3490 | * to make sure that holes are punched for those areas in case they had | |
3491 | * extents previously. | |
3492 | */ | |
3493 | while (!done) { | |
3494 | u64 offset, len; | |
3495 | u64 extent_end; | |
3496 | ||
3497 | if (i >= btrfs_header_nritems(src_path->nodes[0])) { | |
3498 | ret = btrfs_next_leaf(BTRFS_I(inode)->root, src_path); | |
3499 | if (ret < 0) | |
3500 | return ret; | |
3501 | ASSERT(ret == 0); | |
3502 | src = src_path->nodes[0]; | |
3503 | i = 0; | |
3504 | } | |
3505 | ||
3506 | btrfs_item_key_to_cpu(src, &key, i); | |
3507 | if (!btrfs_comp_cpu_keys(&key, &last_key)) | |
3508 | done = true; | |
3509 | if (key.objectid != btrfs_ino(inode) || | |
3510 | key.type != BTRFS_EXTENT_DATA_KEY) { | |
3511 | i++; | |
3512 | continue; | |
3513 | } | |
3514 | extent = btrfs_item_ptr(src, i, struct btrfs_file_extent_item); | |
3515 | if (btrfs_file_extent_type(src, extent) == | |
3516 | BTRFS_FILE_EXTENT_INLINE) { | |
514ac8ad | 3517 | len = btrfs_file_extent_inline_len(src, i, extent); |
16e7549f JB |
3518 | extent_end = ALIGN(key.offset + len, log->sectorsize); |
3519 | } else { | |
3520 | len = btrfs_file_extent_num_bytes(src, extent); | |
3521 | extent_end = key.offset + len; | |
3522 | } | |
3523 | i++; | |
3524 | ||
3525 | if (*last_extent == key.offset) { | |
3526 | *last_extent = extent_end; | |
3527 | continue; | |
3528 | } | |
3529 | offset = *last_extent; | |
3530 | len = key.offset - *last_extent; | |
3531 | ret = btrfs_insert_file_extent(trans, log, btrfs_ino(inode), | |
3532 | offset, 0, 0, len, 0, len, 0, | |
3533 | 0, 0); | |
3534 | if (ret) | |
3535 | break; | |
3536 | *last_extent = offset + len; | |
3537 | } | |
3538 | /* | |
3539 | * Need to let the callers know we dropped the path so they should | |
3540 | * re-search. | |
3541 | */ | |
3542 | if (!ret && need_find_last_extent) | |
3543 | ret = 1; | |
4a500fd1 | 3544 | return ret; |
31ff1cd2 CM |
3545 | } |
3546 | ||
5dc562c5 JB |
3547 | static int extent_cmp(void *priv, struct list_head *a, struct list_head *b) |
3548 | { | |
3549 | struct extent_map *em1, *em2; | |
3550 | ||
3551 | em1 = list_entry(a, struct extent_map, list); | |
3552 | em2 = list_entry(b, struct extent_map, list); | |
3553 | ||
3554 | if (em1->start < em2->start) | |
3555 | return -1; | |
3556 | else if (em1->start > em2->start) | |
3557 | return 1; | |
3558 | return 0; | |
3559 | } | |
3560 | ||
5dc562c5 JB |
3561 | static int log_one_extent(struct btrfs_trans_handle *trans, |
3562 | struct inode *inode, struct btrfs_root *root, | |
827463c4 MX |
3563 | struct extent_map *em, struct btrfs_path *path, |
3564 | struct list_head *logged_list) | |
5dc562c5 JB |
3565 | { |
3566 | struct btrfs_root *log = root->log_root; | |
70c8a91c JB |
3567 | struct btrfs_file_extent_item *fi; |
3568 | struct extent_buffer *leaf; | |
2ab28f32 | 3569 | struct btrfs_ordered_extent *ordered; |
70c8a91c | 3570 | struct list_head ordered_sums; |
0b1c6cca | 3571 | struct btrfs_map_token token; |
5dc562c5 | 3572 | struct btrfs_key key; |
2ab28f32 JB |
3573 | u64 mod_start = em->mod_start; |
3574 | u64 mod_len = em->mod_len; | |
3575 | u64 csum_offset; | |
3576 | u64 csum_len; | |
70c8a91c JB |
3577 | u64 extent_offset = em->start - em->orig_start; |
3578 | u64 block_len; | |
5dc562c5 | 3579 | int ret; |
70c8a91c | 3580 | bool skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; |
1acae57b | 3581 | int extent_inserted = 0; |
09a2a8f9 | 3582 | |
70c8a91c | 3583 | INIT_LIST_HEAD(&ordered_sums); |
0b1c6cca | 3584 | btrfs_init_map_token(&token); |
70c8a91c | 3585 | |
1acae57b FDBM |
3586 | ret = __btrfs_drop_extents(trans, log, inode, path, em->start, |
3587 | em->start + em->len, NULL, 0, 1, | |
3588 | sizeof(*fi), &extent_inserted); | |
09a2a8f9 | 3589 | if (ret) |
70c8a91c | 3590 | return ret; |
1acae57b FDBM |
3591 | |
3592 | if (!extent_inserted) { | |
3593 | key.objectid = btrfs_ino(inode); | |
3594 | key.type = BTRFS_EXTENT_DATA_KEY; | |
3595 | key.offset = em->start; | |
3596 | ||
3597 | ret = btrfs_insert_empty_item(trans, log, path, &key, | |
3598 | sizeof(*fi)); | |
3599 | if (ret) | |
3600 | return ret; | |
3601 | } | |
70c8a91c JB |
3602 | leaf = path->nodes[0]; |
3603 | fi = btrfs_item_ptr(leaf, path->slots[0], | |
3604 | struct btrfs_file_extent_item); | |
124fe663 | 3605 | |
0b1c6cca JB |
3606 | btrfs_set_token_file_extent_generation(leaf, fi, em->generation, |
3607 | &token); | |
70c8a91c JB |
3608 | if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { |
3609 | skip_csum = true; | |
0b1c6cca JB |
3610 | btrfs_set_token_file_extent_type(leaf, fi, |
3611 | BTRFS_FILE_EXTENT_PREALLOC, | |
3612 | &token); | |
70c8a91c | 3613 | } else { |
0b1c6cca JB |
3614 | btrfs_set_token_file_extent_type(leaf, fi, |
3615 | BTRFS_FILE_EXTENT_REG, | |
3616 | &token); | |
ed9e8af8 | 3617 | if (em->block_start == EXTENT_MAP_HOLE) |
70c8a91c JB |
3618 | skip_csum = true; |
3619 | } | |
3620 | ||
3621 | block_len = max(em->block_len, em->orig_block_len); | |
3622 | if (em->compress_type != BTRFS_COMPRESS_NONE) { | |
0b1c6cca JB |
3623 | btrfs_set_token_file_extent_disk_bytenr(leaf, fi, |
3624 | em->block_start, | |
3625 | &token); | |
3626 | btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, block_len, | |
3627 | &token); | |
70c8a91c | 3628 | } else if (em->block_start < EXTENT_MAP_LAST_BYTE) { |
0b1c6cca JB |
3629 | btrfs_set_token_file_extent_disk_bytenr(leaf, fi, |
3630 | em->block_start - | |
3631 | extent_offset, &token); | |
3632 | btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, block_len, | |
3633 | &token); | |
70c8a91c | 3634 | } else { |
0b1c6cca JB |
3635 | btrfs_set_token_file_extent_disk_bytenr(leaf, fi, 0, &token); |
3636 | btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, 0, | |
3637 | &token); | |
3638 | } | |
3639 | ||
3640 | btrfs_set_token_file_extent_offset(leaf, fi, | |
3641 | em->start - em->orig_start, | |
3642 | &token); | |
3643 | btrfs_set_token_file_extent_num_bytes(leaf, fi, em->len, &token); | |
cc95bef6 | 3644 | btrfs_set_token_file_extent_ram_bytes(leaf, fi, em->ram_bytes, &token); |
0b1c6cca JB |
3645 | btrfs_set_token_file_extent_compression(leaf, fi, em->compress_type, |
3646 | &token); | |
3647 | btrfs_set_token_file_extent_encryption(leaf, fi, 0, &token); | |
3648 | btrfs_set_token_file_extent_other_encoding(leaf, fi, 0, &token); | |
70c8a91c | 3649 | btrfs_mark_buffer_dirty(leaf); |
0aa4a17d | 3650 | |
70c8a91c | 3651 | btrfs_release_path(path); |
70c8a91c JB |
3652 | if (ret) { |
3653 | return ret; | |
3654 | } | |
0aa4a17d | 3655 | |
70c8a91c JB |
3656 | if (skip_csum) |
3657 | return 0; | |
5dc562c5 | 3658 | |
2ab28f32 JB |
3659 | /* |
3660 | * First check and see if our csums are on our outstanding ordered | |
3661 | * extents. | |
3662 | */ | |
827463c4 | 3663 | list_for_each_entry(ordered, logged_list, log_list) { |
2ab28f32 JB |
3664 | struct btrfs_ordered_sum *sum; |
3665 | ||
3666 | if (!mod_len) | |
3667 | break; | |
3668 | ||
2ab28f32 JB |
3669 | if (ordered->file_offset + ordered->len <= mod_start || |
3670 | mod_start + mod_len <= ordered->file_offset) | |
3671 | continue; | |
3672 | ||
3673 | /* | |
3674 | * We are going to copy all the csums on this ordered extent, so | |
3675 | * go ahead and adjust mod_start and mod_len in case this | |
3676 | * ordered extent has already been logged. | |
3677 | */ | |
3678 | if (ordered->file_offset > mod_start) { | |
3679 | if (ordered->file_offset + ordered->len >= | |
3680 | mod_start + mod_len) | |
3681 | mod_len = ordered->file_offset - mod_start; | |
3682 | /* | |
3683 | * If we have this case | |
3684 | * | |
3685 | * |--------- logged extent ---------| | |
3686 | * |----- ordered extent ----| | |
3687 | * | |
3688 | * Just don't mess with mod_start and mod_len, we'll | |
3689 | * just end up logging more csums than we need and it | |
3690 | * will be ok. | |
3691 | */ | |
3692 | } else { | |
3693 | if (ordered->file_offset + ordered->len < | |
3694 | mod_start + mod_len) { | |
3695 | mod_len = (mod_start + mod_len) - | |
3696 | (ordered->file_offset + ordered->len); | |
3697 | mod_start = ordered->file_offset + | |
3698 | ordered->len; | |
3699 | } else { | |
3700 | mod_len = 0; | |
3701 | } | |
3702 | } | |
3703 | ||
3704 | /* | |
3705 | * To keep us from looping for the above case of an ordered | |
3706 | * extent that falls inside of the logged extent. | |
3707 | */ | |
3708 | if (test_and_set_bit(BTRFS_ORDERED_LOGGED_CSUM, | |
3709 | &ordered->flags)) | |
3710 | continue; | |
2ab28f32 | 3711 | |
23c671a5 MX |
3712 | if (ordered->csum_bytes_left) { |
3713 | btrfs_start_ordered_extent(inode, ordered, 0); | |
3714 | wait_event(ordered->wait, | |
3715 | ordered->csum_bytes_left == 0); | |
3716 | } | |
2ab28f32 JB |
3717 | |
3718 | list_for_each_entry(sum, &ordered->list, list) { | |
3719 | ret = btrfs_csum_file_blocks(trans, log, sum); | |
827463c4 | 3720 | if (ret) |
2ab28f32 | 3721 | goto unlocked; |
2ab28f32 | 3722 | } |
2ab28f32 JB |
3723 | |
3724 | } | |
2ab28f32 JB |
3725 | unlocked: |
3726 | ||
3727 | if (!mod_len || ret) | |
3728 | return ret; | |
3729 | ||
488111aa FDBM |
3730 | if (em->compress_type) { |
3731 | csum_offset = 0; | |
3732 | csum_len = block_len; | |
3733 | } else { | |
3734 | csum_offset = mod_start - em->start; | |
3735 | csum_len = mod_len; | |
3736 | } | |
2ab28f32 | 3737 | |
70c8a91c JB |
3738 | /* block start is already adjusted for the file extent offset. */ |
3739 | ret = btrfs_lookup_csums_range(log->fs_info->csum_root, | |
3740 | em->block_start + csum_offset, | |
3741 | em->block_start + csum_offset + | |
3742 | csum_len - 1, &ordered_sums, 0); | |
3743 | if (ret) | |
3744 | return ret; | |
5dc562c5 | 3745 | |
70c8a91c JB |
3746 | while (!list_empty(&ordered_sums)) { |
3747 | struct btrfs_ordered_sum *sums = list_entry(ordered_sums.next, | |
3748 | struct btrfs_ordered_sum, | |
3749 | list); | |
3750 | if (!ret) | |
3751 | ret = btrfs_csum_file_blocks(trans, log, sums); | |
3752 | list_del(&sums->list); | |
3753 | kfree(sums); | |
5dc562c5 JB |
3754 | } |
3755 | ||
70c8a91c | 3756 | return ret; |
5dc562c5 JB |
3757 | } |
3758 | ||
3759 | static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans, | |
3760 | struct btrfs_root *root, | |
3761 | struct inode *inode, | |
827463c4 MX |
3762 | struct btrfs_path *path, |
3763 | struct list_head *logged_list) | |
5dc562c5 | 3764 | { |
5dc562c5 JB |
3765 | struct extent_map *em, *n; |
3766 | struct list_head extents; | |
3767 | struct extent_map_tree *tree = &BTRFS_I(inode)->extent_tree; | |
3768 | u64 test_gen; | |
3769 | int ret = 0; | |
2ab28f32 | 3770 | int num = 0; |
5dc562c5 JB |
3771 | |
3772 | INIT_LIST_HEAD(&extents); | |
3773 | ||
5dc562c5 JB |
3774 | write_lock(&tree->lock); |
3775 | test_gen = root->fs_info->last_trans_committed; | |
3776 | ||
3777 | list_for_each_entry_safe(em, n, &tree->modified_extents, list) { | |
3778 | list_del_init(&em->list); | |
2ab28f32 JB |
3779 | |
3780 | /* | |
3781 | * Just an arbitrary number, this can be really CPU intensive | |
3782 | * once we start getting a lot of extents, and really once we | |
3783 | * have a bunch of extents we just want to commit since it will | |
3784 | * be faster. | |
3785 | */ | |
3786 | if (++num > 32768) { | |
3787 | list_del_init(&tree->modified_extents); | |
3788 | ret = -EFBIG; | |
3789 | goto process; | |
3790 | } | |
3791 | ||
5dc562c5 JB |
3792 | if (em->generation <= test_gen) |
3793 | continue; | |
ff44c6e3 JB |
3794 | /* Need a ref to keep it from getting evicted from cache */ |
3795 | atomic_inc(&em->refs); | |
3796 | set_bit(EXTENT_FLAG_LOGGING, &em->flags); | |
5dc562c5 | 3797 | list_add_tail(&em->list, &extents); |
2ab28f32 | 3798 | num++; |
5dc562c5 JB |
3799 | } |
3800 | ||
3801 | list_sort(NULL, &extents, extent_cmp); | |
3802 | ||
2ab28f32 | 3803 | process: |
5dc562c5 JB |
3804 | while (!list_empty(&extents)) { |
3805 | em = list_entry(extents.next, struct extent_map, list); | |
3806 | ||
3807 | list_del_init(&em->list); | |
3808 | ||
3809 | /* | |
3810 | * If we had an error we just need to delete everybody from our | |
3811 | * private list. | |
3812 | */ | |
ff44c6e3 | 3813 | if (ret) { |
201a9038 | 3814 | clear_em_logging(tree, em); |
ff44c6e3 | 3815 | free_extent_map(em); |
5dc562c5 | 3816 | continue; |
ff44c6e3 JB |
3817 | } |
3818 | ||
3819 | write_unlock(&tree->lock); | |
5dc562c5 | 3820 | |
827463c4 | 3821 | ret = log_one_extent(trans, inode, root, em, path, logged_list); |
ff44c6e3 | 3822 | write_lock(&tree->lock); |
201a9038 JB |
3823 | clear_em_logging(tree, em); |
3824 | free_extent_map(em); | |
5dc562c5 | 3825 | } |
ff44c6e3 JB |
3826 | WARN_ON(!list_empty(&extents)); |
3827 | write_unlock(&tree->lock); | |
5dc562c5 | 3828 | |
5dc562c5 | 3829 | btrfs_release_path(path); |
5dc562c5 JB |
3830 | return ret; |
3831 | } | |
3832 | ||
e02119d5 CM |
3833 | /* log a single inode in the tree log. |
3834 | * At least one parent directory for this inode must exist in the tree | |
3835 | * or be logged already. | |
3836 | * | |
3837 | * Any items from this inode changed by the current transaction are copied | |
3838 | * to the log tree. An extra reference is taken on any extents in this | |
3839 | * file, allowing us to avoid a whole pile of corner cases around logging | |
3840 | * blocks that have been removed from the tree. | |
3841 | * | |
3842 | * See LOG_INODE_ALL and related defines for a description of what inode_only | |
3843 | * does. | |
3844 | * | |
3845 | * This handles both files and directories. | |
3846 | */ | |
12fcfd22 | 3847 | static int btrfs_log_inode(struct btrfs_trans_handle *trans, |
e02119d5 CM |
3848 | struct btrfs_root *root, struct inode *inode, |
3849 | int inode_only) | |
3850 | { | |
3851 | struct btrfs_path *path; | |
3852 | struct btrfs_path *dst_path; | |
3853 | struct btrfs_key min_key; | |
3854 | struct btrfs_key max_key; | |
3855 | struct btrfs_root *log = root->log_root; | |
31ff1cd2 | 3856 | struct extent_buffer *src = NULL; |
827463c4 | 3857 | LIST_HEAD(logged_list); |
16e7549f | 3858 | u64 last_extent = 0; |
4a500fd1 | 3859 | int err = 0; |
e02119d5 | 3860 | int ret; |
3a5f1d45 | 3861 | int nritems; |
31ff1cd2 CM |
3862 | int ins_start_slot = 0; |
3863 | int ins_nr; | |
5dc562c5 | 3864 | bool fast_search = false; |
33345d01 | 3865 | u64 ino = btrfs_ino(inode); |
e02119d5 | 3866 | |
e02119d5 | 3867 | path = btrfs_alloc_path(); |
5df67083 TI |
3868 | if (!path) |
3869 | return -ENOMEM; | |
e02119d5 | 3870 | dst_path = btrfs_alloc_path(); |
5df67083 TI |
3871 | if (!dst_path) { |
3872 | btrfs_free_path(path); | |
3873 | return -ENOMEM; | |
3874 | } | |
e02119d5 | 3875 | |
33345d01 | 3876 | min_key.objectid = ino; |
e02119d5 CM |
3877 | min_key.type = BTRFS_INODE_ITEM_KEY; |
3878 | min_key.offset = 0; | |
3879 | ||
33345d01 | 3880 | max_key.objectid = ino; |
12fcfd22 | 3881 | |
12fcfd22 | 3882 | |
5dc562c5 | 3883 | /* today the code can only do partial logging of directories */ |
5269b67e MX |
3884 | if (S_ISDIR(inode->i_mode) || |
3885 | (!test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, | |
3886 | &BTRFS_I(inode)->runtime_flags) && | |
3887 | inode_only == LOG_INODE_EXISTS)) | |
e02119d5 CM |
3888 | max_key.type = BTRFS_XATTR_ITEM_KEY; |
3889 | else | |
3890 | max_key.type = (u8)-1; | |
3891 | max_key.offset = (u64)-1; | |
3892 | ||
94edf4ae JB |
3893 | /* Only run delayed items if we are a dir or a new file */ |
3894 | if (S_ISDIR(inode->i_mode) || | |
3895 | BTRFS_I(inode)->generation > root->fs_info->last_trans_committed) { | |
3896 | ret = btrfs_commit_inode_delayed_items(trans, inode); | |
3897 | if (ret) { | |
3898 | btrfs_free_path(path); | |
3899 | btrfs_free_path(dst_path); | |
3900 | return ret; | |
3901 | } | |
16cdcec7 MX |
3902 | } |
3903 | ||
e02119d5 CM |
3904 | mutex_lock(&BTRFS_I(inode)->log_mutex); |
3905 | ||
827463c4 | 3906 | btrfs_get_logged_extents(inode, &logged_list); |
2ab28f32 | 3907 | |
e02119d5 CM |
3908 | /* |
3909 | * a brute force approach to making sure we get the most uptodate | |
3910 | * copies of everything. | |
3911 | */ | |
3912 | if (S_ISDIR(inode->i_mode)) { | |
3913 | int max_key_type = BTRFS_DIR_LOG_INDEX_KEY; | |
3914 | ||
3915 | if (inode_only == LOG_INODE_EXISTS) | |
3916 | max_key_type = BTRFS_XATTR_ITEM_KEY; | |
33345d01 | 3917 | ret = drop_objectid_items(trans, log, path, ino, max_key_type); |
e02119d5 | 3918 | } else { |
5dc562c5 JB |
3919 | if (test_and_clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC, |
3920 | &BTRFS_I(inode)->runtime_flags)) { | |
e9976151 JB |
3921 | clear_bit(BTRFS_INODE_COPY_EVERYTHING, |
3922 | &BTRFS_I(inode)->runtime_flags); | |
5dc562c5 JB |
3923 | ret = btrfs_truncate_inode_items(trans, log, |
3924 | inode, 0, 0); | |
a95249b3 | 3925 | } else if (test_and_clear_bit(BTRFS_INODE_COPY_EVERYTHING, |
6cfab851 JB |
3926 | &BTRFS_I(inode)->runtime_flags) || |
3927 | inode_only == LOG_INODE_EXISTS) { | |
183f37fa LB |
3928 | if (inode_only == LOG_INODE_ALL) |
3929 | fast_search = true; | |
a95249b3 | 3930 | max_key.type = BTRFS_XATTR_ITEM_KEY; |
5dc562c5 | 3931 | ret = drop_objectid_items(trans, log, path, ino, |
e9976151 | 3932 | max_key.type); |
a95249b3 JB |
3933 | } else { |
3934 | if (inode_only == LOG_INODE_ALL) | |
3935 | fast_search = true; | |
3936 | ret = log_inode_item(trans, log, dst_path, inode); | |
3937 | if (ret) { | |
3938 | err = ret; | |
3939 | goto out_unlock; | |
3940 | } | |
3941 | goto log_extents; | |
5dc562c5 | 3942 | } |
a95249b3 | 3943 | |
e02119d5 | 3944 | } |
4a500fd1 YZ |
3945 | if (ret) { |
3946 | err = ret; | |
3947 | goto out_unlock; | |
3948 | } | |
e02119d5 CM |
3949 | path->keep_locks = 1; |
3950 | ||
d397712b | 3951 | while (1) { |
31ff1cd2 | 3952 | ins_nr = 0; |
6174d3cb | 3953 | ret = btrfs_search_forward(root, &min_key, |
de78b51a | 3954 | path, trans->transid); |
e02119d5 CM |
3955 | if (ret != 0) |
3956 | break; | |
3a5f1d45 | 3957 | again: |
31ff1cd2 | 3958 | /* note, ins_nr might be > 0 here, cleanup outside the loop */ |
33345d01 | 3959 | if (min_key.objectid != ino) |
e02119d5 CM |
3960 | break; |
3961 | if (min_key.type > max_key.type) | |
3962 | break; | |
31ff1cd2 | 3963 | |
e02119d5 | 3964 | src = path->nodes[0]; |
31ff1cd2 CM |
3965 | if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) { |
3966 | ins_nr++; | |
3967 | goto next_slot; | |
3968 | } else if (!ins_nr) { | |
3969 | ins_start_slot = path->slots[0]; | |
3970 | ins_nr = 1; | |
3971 | goto next_slot; | |
e02119d5 CM |
3972 | } |
3973 | ||
16e7549f JB |
3974 | ret = copy_items(trans, inode, dst_path, path, &last_extent, |
3975 | ins_start_slot, ins_nr, inode_only); | |
3976 | if (ret < 0) { | |
4a500fd1 YZ |
3977 | err = ret; |
3978 | goto out_unlock; | |
16e7549f JB |
3979 | } if (ret) { |
3980 | ins_nr = 0; | |
3981 | btrfs_release_path(path); | |
3982 | continue; | |
4a500fd1 | 3983 | } |
31ff1cd2 CM |
3984 | ins_nr = 1; |
3985 | ins_start_slot = path->slots[0]; | |
3986 | next_slot: | |
e02119d5 | 3987 | |
3a5f1d45 CM |
3988 | nritems = btrfs_header_nritems(path->nodes[0]); |
3989 | path->slots[0]++; | |
3990 | if (path->slots[0] < nritems) { | |
3991 | btrfs_item_key_to_cpu(path->nodes[0], &min_key, | |
3992 | path->slots[0]); | |
3993 | goto again; | |
3994 | } | |
31ff1cd2 | 3995 | if (ins_nr) { |
16e7549f JB |
3996 | ret = copy_items(trans, inode, dst_path, path, |
3997 | &last_extent, ins_start_slot, | |
31ff1cd2 | 3998 | ins_nr, inode_only); |
16e7549f | 3999 | if (ret < 0) { |
4a500fd1 YZ |
4000 | err = ret; |
4001 | goto out_unlock; | |
4002 | } | |
16e7549f | 4003 | ret = 0; |
31ff1cd2 CM |
4004 | ins_nr = 0; |
4005 | } | |
b3b4aa74 | 4006 | btrfs_release_path(path); |
3a5f1d45 | 4007 | |
3d41d702 | 4008 | if (min_key.offset < (u64)-1) { |
e02119d5 | 4009 | min_key.offset++; |
3d41d702 | 4010 | } else if (min_key.type < max_key.type) { |
e02119d5 | 4011 | min_key.type++; |
3d41d702 FDBM |
4012 | min_key.offset = 0; |
4013 | } else { | |
e02119d5 | 4014 | break; |
3d41d702 | 4015 | } |
e02119d5 | 4016 | } |
31ff1cd2 | 4017 | if (ins_nr) { |
16e7549f JB |
4018 | ret = copy_items(trans, inode, dst_path, path, &last_extent, |
4019 | ins_start_slot, ins_nr, inode_only); | |
4020 | if (ret < 0) { | |
4a500fd1 YZ |
4021 | err = ret; |
4022 | goto out_unlock; | |
4023 | } | |
16e7549f | 4024 | ret = 0; |
31ff1cd2 CM |
4025 | ins_nr = 0; |
4026 | } | |
5dc562c5 | 4027 | |
a95249b3 | 4028 | log_extents: |
f3b15ccd JB |
4029 | btrfs_release_path(path); |
4030 | btrfs_release_path(dst_path); | |
5dc562c5 | 4031 | if (fast_search) { |
827463c4 MX |
4032 | ret = btrfs_log_changed_extents(trans, root, inode, dst_path, |
4033 | &logged_list); | |
5dc562c5 JB |
4034 | if (ret) { |
4035 | err = ret; | |
4036 | goto out_unlock; | |
4037 | } | |
d006a048 | 4038 | } else if (inode_only == LOG_INODE_ALL) { |
06d3d22b LB |
4039 | struct extent_map_tree *tree = &BTRFS_I(inode)->extent_tree; |
4040 | struct extent_map *em, *n; | |
4041 | ||
bbe14267 | 4042 | write_lock(&tree->lock); |
06d3d22b LB |
4043 | list_for_each_entry_safe(em, n, &tree->modified_extents, list) |
4044 | list_del_init(&em->list); | |
bbe14267 | 4045 | write_unlock(&tree->lock); |
5dc562c5 JB |
4046 | } |
4047 | ||
9623f9a3 | 4048 | if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->i_mode)) { |
e02119d5 | 4049 | ret = log_directory_changes(trans, root, inode, path, dst_path); |
4a500fd1 YZ |
4050 | if (ret) { |
4051 | err = ret; | |
4052 | goto out_unlock; | |
4053 | } | |
e02119d5 | 4054 | } |
3a5f1d45 | 4055 | BTRFS_I(inode)->logged_trans = trans->transid; |
46d8bc34 | 4056 | BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->last_sub_trans; |
4a500fd1 | 4057 | out_unlock: |
827463c4 MX |
4058 | if (unlikely(err)) |
4059 | btrfs_put_logged_extents(&logged_list); | |
4060 | else | |
4061 | btrfs_submit_logged_extents(&logged_list, log); | |
e02119d5 CM |
4062 | mutex_unlock(&BTRFS_I(inode)->log_mutex); |
4063 | ||
4064 | btrfs_free_path(path); | |
4065 | btrfs_free_path(dst_path); | |
4a500fd1 | 4066 | return err; |
e02119d5 CM |
4067 | } |
4068 | ||
12fcfd22 CM |
4069 | /* |
4070 | * follow the dentry parent pointers up the chain and see if any | |
4071 | * of the directories in it require a full commit before they can | |
4072 | * be logged. Returns zero if nothing special needs to be done or 1 if | |
4073 | * a full commit is required. | |
4074 | */ | |
4075 | static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans, | |
4076 | struct inode *inode, | |
4077 | struct dentry *parent, | |
4078 | struct super_block *sb, | |
4079 | u64 last_committed) | |
e02119d5 | 4080 | { |
12fcfd22 CM |
4081 | int ret = 0; |
4082 | struct btrfs_root *root; | |
6a912213 | 4083 | struct dentry *old_parent = NULL; |
de2b530b | 4084 | struct inode *orig_inode = inode; |
e02119d5 | 4085 | |
af4176b4 CM |
4086 | /* |
4087 | * for regular files, if its inode is already on disk, we don't | |
4088 | * have to worry about the parents at all. This is because | |
4089 | * we can use the last_unlink_trans field to record renames | |
4090 | * and other fun in this file. | |
4091 | */ | |
4092 | if (S_ISREG(inode->i_mode) && | |
4093 | BTRFS_I(inode)->generation <= last_committed && | |
4094 | BTRFS_I(inode)->last_unlink_trans <= last_committed) | |
4095 | goto out; | |
4096 | ||
12fcfd22 CM |
4097 | if (!S_ISDIR(inode->i_mode)) { |
4098 | if (!parent || !parent->d_inode || sb != parent->d_inode->i_sb) | |
4099 | goto out; | |
4100 | inode = parent->d_inode; | |
4101 | } | |
4102 | ||
4103 | while (1) { | |
de2b530b JB |
4104 | /* |
4105 | * If we are logging a directory then we start with our inode, | |
4106 | * not our parents inode, so we need to skipp setting the | |
4107 | * logged_trans so that further down in the log code we don't | |
4108 | * think this inode has already been logged. | |
4109 | */ | |
4110 | if (inode != orig_inode) | |
4111 | BTRFS_I(inode)->logged_trans = trans->transid; | |
12fcfd22 CM |
4112 | smp_mb(); |
4113 | ||
4114 | if (BTRFS_I(inode)->last_unlink_trans > last_committed) { | |
4115 | root = BTRFS_I(inode)->root; | |
4116 | ||
4117 | /* | |
4118 | * make sure any commits to the log are forced | |
4119 | * to be full commits | |
4120 | */ | |
4121 | root->fs_info->last_trans_log_full_commit = | |
4122 | trans->transid; | |
4123 | ret = 1; | |
4124 | break; | |
4125 | } | |
4126 | ||
4127 | if (!parent || !parent->d_inode || sb != parent->d_inode->i_sb) | |
4128 | break; | |
4129 | ||
76dda93c | 4130 | if (IS_ROOT(parent)) |
12fcfd22 CM |
4131 | break; |
4132 | ||
6a912213 JB |
4133 | parent = dget_parent(parent); |
4134 | dput(old_parent); | |
4135 | old_parent = parent; | |
12fcfd22 CM |
4136 | inode = parent->d_inode; |
4137 | ||
4138 | } | |
6a912213 | 4139 | dput(old_parent); |
12fcfd22 | 4140 | out: |
e02119d5 CM |
4141 | return ret; |
4142 | } | |
4143 | ||
4144 | /* | |
4145 | * helper function around btrfs_log_inode to make sure newly created | |
4146 | * parent directories also end up in the log. A minimal inode and backref | |
4147 | * only logging is done of any parent directories that are older than | |
4148 | * the last committed transaction | |
4149 | */ | |
48a3b636 ES |
4150 | static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, |
4151 | struct btrfs_root *root, struct inode *inode, | |
8b050d35 MX |
4152 | struct dentry *parent, int exists_only, |
4153 | struct btrfs_log_ctx *ctx) | |
e02119d5 | 4154 | { |
12fcfd22 | 4155 | int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL; |
e02119d5 | 4156 | struct super_block *sb; |
6a912213 | 4157 | struct dentry *old_parent = NULL; |
12fcfd22 CM |
4158 | int ret = 0; |
4159 | u64 last_committed = root->fs_info->last_trans_committed; | |
4160 | ||
4161 | sb = inode->i_sb; | |
4162 | ||
3a5e1404 SW |
4163 | if (btrfs_test_opt(root, NOTREELOG)) { |
4164 | ret = 1; | |
4165 | goto end_no_trans; | |
4166 | } | |
4167 | ||
12fcfd22 CM |
4168 | if (root->fs_info->last_trans_log_full_commit > |
4169 | root->fs_info->last_trans_committed) { | |
4170 | ret = 1; | |
4171 | goto end_no_trans; | |
4172 | } | |
4173 | ||
76dda93c YZ |
4174 | if (root != BTRFS_I(inode)->root || |
4175 | btrfs_root_refs(&root->root_item) == 0) { | |
4176 | ret = 1; | |
4177 | goto end_no_trans; | |
4178 | } | |
4179 | ||
12fcfd22 CM |
4180 | ret = check_parent_dirs_for_sync(trans, inode, parent, |
4181 | sb, last_committed); | |
4182 | if (ret) | |
4183 | goto end_no_trans; | |
e02119d5 | 4184 | |
22ee6985 | 4185 | if (btrfs_inode_in_log(inode, trans->transid)) { |
257c62e1 CM |
4186 | ret = BTRFS_NO_LOG_SYNC; |
4187 | goto end_no_trans; | |
4188 | } | |
4189 | ||
8b050d35 | 4190 | ret = start_log_trans(trans, root, ctx); |
4a500fd1 | 4191 | if (ret) |
e87ac136 | 4192 | goto end_no_trans; |
e02119d5 | 4193 | |
12fcfd22 | 4194 | ret = btrfs_log_inode(trans, root, inode, inode_only); |
4a500fd1 YZ |
4195 | if (ret) |
4196 | goto end_trans; | |
12fcfd22 | 4197 | |
af4176b4 CM |
4198 | /* |
4199 | * for regular files, if its inode is already on disk, we don't | |
4200 | * have to worry about the parents at all. This is because | |
4201 | * we can use the last_unlink_trans field to record renames | |
4202 | * and other fun in this file. | |
4203 | */ | |
4204 | if (S_ISREG(inode->i_mode) && | |
4205 | BTRFS_I(inode)->generation <= last_committed && | |
4a500fd1 YZ |
4206 | BTRFS_I(inode)->last_unlink_trans <= last_committed) { |
4207 | ret = 0; | |
4208 | goto end_trans; | |
4209 | } | |
af4176b4 CM |
4210 | |
4211 | inode_only = LOG_INODE_EXISTS; | |
12fcfd22 CM |
4212 | while (1) { |
4213 | if (!parent || !parent->d_inode || sb != parent->d_inode->i_sb) | |
e02119d5 CM |
4214 | break; |
4215 | ||
12fcfd22 | 4216 | inode = parent->d_inode; |
76dda93c YZ |
4217 | if (root != BTRFS_I(inode)->root) |
4218 | break; | |
4219 | ||
12fcfd22 CM |
4220 | if (BTRFS_I(inode)->generation > |
4221 | root->fs_info->last_trans_committed) { | |
4222 | ret = btrfs_log_inode(trans, root, inode, inode_only); | |
4a500fd1 YZ |
4223 | if (ret) |
4224 | goto end_trans; | |
12fcfd22 | 4225 | } |
76dda93c | 4226 | if (IS_ROOT(parent)) |
e02119d5 | 4227 | break; |
12fcfd22 | 4228 | |
6a912213 JB |
4229 | parent = dget_parent(parent); |
4230 | dput(old_parent); | |
4231 | old_parent = parent; | |
e02119d5 | 4232 | } |
12fcfd22 | 4233 | ret = 0; |
4a500fd1 | 4234 | end_trans: |
6a912213 | 4235 | dput(old_parent); |
4a500fd1 | 4236 | if (ret < 0) { |
4a500fd1 YZ |
4237 | root->fs_info->last_trans_log_full_commit = trans->transid; |
4238 | ret = 1; | |
4239 | } | |
8b050d35 MX |
4240 | |
4241 | if (ret) | |
4242 | btrfs_remove_log_ctx(root, ctx); | |
12fcfd22 CM |
4243 | btrfs_end_log_trans(root); |
4244 | end_no_trans: | |
4245 | return ret; | |
e02119d5 CM |
4246 | } |
4247 | ||
4248 | /* | |
4249 | * it is not safe to log dentry if the chunk root has added new | |
4250 | * chunks. This returns 0 if the dentry was logged, and 1 otherwise. | |
4251 | * If this returns 1, you must commit the transaction to safely get your | |
4252 | * data on disk. | |
4253 | */ | |
4254 | int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans, | |
8b050d35 MX |
4255 | struct btrfs_root *root, struct dentry *dentry, |
4256 | struct btrfs_log_ctx *ctx) | |
e02119d5 | 4257 | { |
6a912213 JB |
4258 | struct dentry *parent = dget_parent(dentry); |
4259 | int ret; | |
4260 | ||
8b050d35 MX |
4261 | ret = btrfs_log_inode_parent(trans, root, dentry->d_inode, parent, |
4262 | 0, ctx); | |
6a912213 JB |
4263 | dput(parent); |
4264 | ||
4265 | return ret; | |
e02119d5 CM |
4266 | } |
4267 | ||
4268 | /* | |
4269 | * should be called during mount to recover any replay any log trees | |
4270 | * from the FS | |
4271 | */ | |
4272 | int btrfs_recover_log_trees(struct btrfs_root *log_root_tree) | |
4273 | { | |
4274 | int ret; | |
4275 | struct btrfs_path *path; | |
4276 | struct btrfs_trans_handle *trans; | |
4277 | struct btrfs_key key; | |
4278 | struct btrfs_key found_key; | |
4279 | struct btrfs_key tmp_key; | |
4280 | struct btrfs_root *log; | |
4281 | struct btrfs_fs_info *fs_info = log_root_tree->fs_info; | |
4282 | struct walk_control wc = { | |
4283 | .process_func = process_one_buffer, | |
4284 | .stage = 0, | |
4285 | }; | |
4286 | ||
e02119d5 | 4287 | path = btrfs_alloc_path(); |
db5b493a TI |
4288 | if (!path) |
4289 | return -ENOMEM; | |
4290 | ||
4291 | fs_info->log_root_recovering = 1; | |
e02119d5 | 4292 | |
4a500fd1 | 4293 | trans = btrfs_start_transaction(fs_info->tree_root, 0); |
79787eaa JM |
4294 | if (IS_ERR(trans)) { |
4295 | ret = PTR_ERR(trans); | |
4296 | goto error; | |
4297 | } | |
e02119d5 CM |
4298 | |
4299 | wc.trans = trans; | |
4300 | wc.pin = 1; | |
4301 | ||
db5b493a | 4302 | ret = walk_log_tree(trans, log_root_tree, &wc); |
79787eaa JM |
4303 | if (ret) { |
4304 | btrfs_error(fs_info, ret, "Failed to pin buffers while " | |
4305 | "recovering log root tree."); | |
4306 | goto error; | |
4307 | } | |
e02119d5 CM |
4308 | |
4309 | again: | |
4310 | key.objectid = BTRFS_TREE_LOG_OBJECTID; | |
4311 | key.offset = (u64)-1; | |
4312 | btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY); | |
4313 | ||
d397712b | 4314 | while (1) { |
e02119d5 | 4315 | ret = btrfs_search_slot(NULL, log_root_tree, &key, path, 0, 0); |
79787eaa JM |
4316 | |
4317 | if (ret < 0) { | |
4318 | btrfs_error(fs_info, ret, | |
4319 | "Couldn't find tree log root."); | |
4320 | goto error; | |
4321 | } | |
e02119d5 CM |
4322 | if (ret > 0) { |
4323 | if (path->slots[0] == 0) | |
4324 | break; | |
4325 | path->slots[0]--; | |
4326 | } | |
4327 | btrfs_item_key_to_cpu(path->nodes[0], &found_key, | |
4328 | path->slots[0]); | |
b3b4aa74 | 4329 | btrfs_release_path(path); |
e02119d5 CM |
4330 | if (found_key.objectid != BTRFS_TREE_LOG_OBJECTID) |
4331 | break; | |
4332 | ||
cb517eab | 4333 | log = btrfs_read_fs_root(log_root_tree, &found_key); |
79787eaa JM |
4334 | if (IS_ERR(log)) { |
4335 | ret = PTR_ERR(log); | |
4336 | btrfs_error(fs_info, ret, | |
4337 | "Couldn't read tree log root."); | |
4338 | goto error; | |
4339 | } | |
e02119d5 CM |
4340 | |
4341 | tmp_key.objectid = found_key.offset; | |
4342 | tmp_key.type = BTRFS_ROOT_ITEM_KEY; | |
4343 | tmp_key.offset = (u64)-1; | |
4344 | ||
4345 | wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key); | |
79787eaa JM |
4346 | if (IS_ERR(wc.replay_dest)) { |
4347 | ret = PTR_ERR(wc.replay_dest); | |
b50c6e25 JB |
4348 | free_extent_buffer(log->node); |
4349 | free_extent_buffer(log->commit_root); | |
4350 | kfree(log); | |
79787eaa JM |
4351 | btrfs_error(fs_info, ret, "Couldn't read target root " |
4352 | "for tree log recovery."); | |
4353 | goto error; | |
4354 | } | |
e02119d5 | 4355 | |
07d400a6 | 4356 | wc.replay_dest->log_root = log; |
5d4f98a2 | 4357 | btrfs_record_root_in_trans(trans, wc.replay_dest); |
e02119d5 | 4358 | ret = walk_log_tree(trans, log, &wc); |
e02119d5 | 4359 | |
b50c6e25 | 4360 | if (!ret && wc.stage == LOG_WALK_REPLAY_ALL) { |
e02119d5 CM |
4361 | ret = fixup_inode_link_counts(trans, wc.replay_dest, |
4362 | path); | |
e02119d5 CM |
4363 | } |
4364 | ||
4365 | key.offset = found_key.offset - 1; | |
07d400a6 | 4366 | wc.replay_dest->log_root = NULL; |
e02119d5 | 4367 | free_extent_buffer(log->node); |
b263c2c8 | 4368 | free_extent_buffer(log->commit_root); |
e02119d5 CM |
4369 | kfree(log); |
4370 | ||
b50c6e25 JB |
4371 | if (ret) |
4372 | goto error; | |
4373 | ||
e02119d5 CM |
4374 | if (found_key.offset == 0) |
4375 | break; | |
4376 | } | |
b3b4aa74 | 4377 | btrfs_release_path(path); |
e02119d5 CM |
4378 | |
4379 | /* step one is to pin it all, step two is to replay just inodes */ | |
4380 | if (wc.pin) { | |
4381 | wc.pin = 0; | |
4382 | wc.process_func = replay_one_buffer; | |
4383 | wc.stage = LOG_WALK_REPLAY_INODES; | |
4384 | goto again; | |
4385 | } | |
4386 | /* step three is to replay everything */ | |
4387 | if (wc.stage < LOG_WALK_REPLAY_ALL) { | |
4388 | wc.stage++; | |
4389 | goto again; | |
4390 | } | |
4391 | ||
4392 | btrfs_free_path(path); | |
4393 | ||
abefa55a JB |
4394 | /* step 4: commit the transaction, which also unpins the blocks */ |
4395 | ret = btrfs_commit_transaction(trans, fs_info->tree_root); | |
4396 | if (ret) | |
4397 | return ret; | |
4398 | ||
e02119d5 CM |
4399 | free_extent_buffer(log_root_tree->node); |
4400 | log_root_tree->log_root = NULL; | |
4401 | fs_info->log_root_recovering = 0; | |
e02119d5 | 4402 | kfree(log_root_tree); |
79787eaa | 4403 | |
abefa55a | 4404 | return 0; |
79787eaa | 4405 | error: |
b50c6e25 JB |
4406 | if (wc.trans) |
4407 | btrfs_end_transaction(wc.trans, fs_info->tree_root); | |
79787eaa JM |
4408 | btrfs_free_path(path); |
4409 | return ret; | |
e02119d5 | 4410 | } |
12fcfd22 CM |
4411 | |
4412 | /* | |
4413 | * there are some corner cases where we want to force a full | |
4414 | * commit instead of allowing a directory to be logged. | |
4415 | * | |
4416 | * They revolve around files there were unlinked from the directory, and | |
4417 | * this function updates the parent directory so that a full commit is | |
4418 | * properly done if it is fsync'd later after the unlinks are done. | |
4419 | */ | |
4420 | void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans, | |
4421 | struct inode *dir, struct inode *inode, | |
4422 | int for_rename) | |
4423 | { | |
af4176b4 CM |
4424 | /* |
4425 | * when we're logging a file, if it hasn't been renamed | |
4426 | * or unlinked, and its inode is fully committed on disk, | |
4427 | * we don't have to worry about walking up the directory chain | |
4428 | * to log its parents. | |
4429 | * | |
4430 | * So, we use the last_unlink_trans field to put this transid | |
4431 | * into the file. When the file is logged we check it and | |
4432 | * don't log the parents if the file is fully on disk. | |
4433 | */ | |
4434 | if (S_ISREG(inode->i_mode)) | |
4435 | BTRFS_I(inode)->last_unlink_trans = trans->transid; | |
4436 | ||
12fcfd22 CM |
4437 | /* |
4438 | * if this directory was already logged any new | |
4439 | * names for this file/dir will get recorded | |
4440 | */ | |
4441 | smp_mb(); | |
4442 | if (BTRFS_I(dir)->logged_trans == trans->transid) | |
4443 | return; | |
4444 | ||
4445 | /* | |
4446 | * if the inode we're about to unlink was logged, | |
4447 | * the log will be properly updated for any new names | |
4448 | */ | |
4449 | if (BTRFS_I(inode)->logged_trans == trans->transid) | |
4450 | return; | |
4451 | ||
4452 | /* | |
4453 | * when renaming files across directories, if the directory | |
4454 | * there we're unlinking from gets fsync'd later on, there's | |
4455 | * no way to find the destination directory later and fsync it | |
4456 | * properly. So, we have to be conservative and force commits | |
4457 | * so the new name gets discovered. | |
4458 | */ | |
4459 | if (for_rename) | |
4460 | goto record; | |
4461 | ||
4462 | /* we can safely do the unlink without any special recording */ | |
4463 | return; | |
4464 | ||
4465 | record: | |
4466 | BTRFS_I(dir)->last_unlink_trans = trans->transid; | |
4467 | } | |
4468 | ||
4469 | /* | |
4470 | * Call this after adding a new name for a file and it will properly | |
4471 | * update the log to reflect the new name. | |
4472 | * | |
4473 | * It will return zero if all goes well, and it will return 1 if a | |
4474 | * full transaction commit is required. | |
4475 | */ | |
4476 | int btrfs_log_new_name(struct btrfs_trans_handle *trans, | |
4477 | struct inode *inode, struct inode *old_dir, | |
4478 | struct dentry *parent) | |
4479 | { | |
4480 | struct btrfs_root * root = BTRFS_I(inode)->root; | |
4481 | ||
af4176b4 CM |
4482 | /* |
4483 | * this will force the logging code to walk the dentry chain | |
4484 | * up for the file | |
4485 | */ | |
4486 | if (S_ISREG(inode->i_mode)) | |
4487 | BTRFS_I(inode)->last_unlink_trans = trans->transid; | |
4488 | ||
12fcfd22 CM |
4489 | /* |
4490 | * if this inode hasn't been logged and directory we're renaming it | |
4491 | * from hasn't been logged, we don't need to log it | |
4492 | */ | |
4493 | if (BTRFS_I(inode)->logged_trans <= | |
4494 | root->fs_info->last_trans_committed && | |
4495 | (!old_dir || BTRFS_I(old_dir)->logged_trans <= | |
4496 | root->fs_info->last_trans_committed)) | |
4497 | return 0; | |
4498 | ||
8b050d35 | 4499 | return btrfs_log_inode_parent(trans, root, inode, parent, 1, NULL); |
12fcfd22 CM |
4500 | } |
4501 |