]>
Commit | Line | Data |
---|---|---|
c1d7c514 | 1 | // SPDX-License-Identifier: GPL-2.0 |
e02119d5 CM |
2 | /* |
3 | * Copyright (C) 2008 Oracle. All rights reserved. | |
e02119d5 CM |
4 | */ |
5 | ||
6 | #include <linux/sched.h> | |
5a0e3ad6 | 7 | #include <linux/slab.h> |
c6adc9cc | 8 | #include <linux/blkdev.h> |
5dc562c5 | 9 | #include <linux/list_sort.h> |
c7f88c4e | 10 | #include <linux/iversion.h> |
9678c543 | 11 | #include "ctree.h" |
995946dd | 12 | #include "tree-log.h" |
e02119d5 CM |
13 | #include "disk-io.h" |
14 | #include "locking.h" | |
15 | #include "print-tree.h" | |
f186373f | 16 | #include "backref.h" |
ebb8765b | 17 | #include "compression.h" |
df2c95f3 | 18 | #include "qgroup.h" |
900c9981 | 19 | #include "inode-map.h" |
e02119d5 CM |
20 | |
21 | /* magic values for the inode_only field in btrfs_log_inode: | |
22 | * | |
23 | * LOG_INODE_ALL means to log everything | |
24 | * LOG_INODE_EXISTS means to log just enough to recreate the inode | |
25 | * during log replay | |
26 | */ | |
27 | #define LOG_INODE_ALL 0 | |
28 | #define LOG_INODE_EXISTS 1 | |
781feef7 | 29 | #define LOG_OTHER_INODE 2 |
e02119d5 | 30 | |
12fcfd22 CM |
31 | /* |
32 | * directory trouble cases | |
33 | * | |
34 | * 1) on rename or unlink, if the inode being unlinked isn't in the fsync | |
35 | * log, we must force a full commit before doing an fsync of the directory | |
36 | * where the unlink was done. | |
37 | * ---> record transid of last unlink/rename per directory | |
38 | * | |
39 | * mkdir foo/some_dir | |
40 | * normal commit | |
41 | * rename foo/some_dir foo2/some_dir | |
42 | * mkdir foo/some_dir | |
43 | * fsync foo/some_dir/some_file | |
44 | * | |
45 | * The fsync above will unlink the original some_dir without recording | |
46 | * it in its new location (foo2). After a crash, some_dir will be gone | |
47 | * unless the fsync of some_file forces a full commit | |
48 | * | |
49 | * 2) we must log any new names for any file or dir that is in the fsync | |
50 | * log. ---> check inode while renaming/linking. | |
51 | * | |
52 | * 2a) we must log any new names for any file or dir during rename | |
53 | * when the directory they are being removed from was logged. | |
54 | * ---> check inode and old parent dir during rename | |
55 | * | |
56 | * 2a is actually the more important variant. With the extra logging | |
57 | * a crash might unlink the old name without recreating the new one | |
58 | * | |
59 | * 3) after a crash, we must go through any directories with a link count | |
60 | * of zero and redo the rm -rf | |
61 | * | |
62 | * mkdir f1/foo | |
63 | * normal commit | |
64 | * rm -rf f1/foo | |
65 | * fsync(f1) | |
66 | * | |
67 | * The directory f1 was fully removed from the FS, but fsync was never | |
68 | * called on f1, only its parent dir. After a crash the rm -rf must | |
69 | * be replayed. This must be able to recurse down the entire | |
70 | * directory tree. The inode link count fixup code takes care of the | |
71 | * ugly details. | |
72 | */ | |
73 | ||
e02119d5 CM |
74 | /* |
75 | * stages for the tree walking. The first | |
76 | * stage (0) is to only pin down the blocks we find | |
77 | * the second stage (1) is to make sure that all the inodes | |
78 | * we find in the log are created in the subvolume. | |
79 | * | |
80 | * The last stage is to deal with directories and links and extents | |
81 | * and all the other fun semantics | |
82 | */ | |
83 | #define LOG_WALK_PIN_ONLY 0 | |
84 | #define LOG_WALK_REPLAY_INODES 1 | |
dd8e7217 JB |
85 | #define LOG_WALK_REPLAY_DIR_INDEX 2 |
86 | #define LOG_WALK_REPLAY_ALL 3 | |
e02119d5 | 87 | |
12fcfd22 | 88 | static int btrfs_log_inode(struct btrfs_trans_handle *trans, |
a59108a7 | 89 | struct btrfs_root *root, struct btrfs_inode *inode, |
49dae1bc FM |
90 | int inode_only, |
91 | const loff_t start, | |
8407f553 FM |
92 | const loff_t end, |
93 | struct btrfs_log_ctx *ctx); | |
ec051c0f YZ |
94 | static int link_to_fixup_dir(struct btrfs_trans_handle *trans, |
95 | struct btrfs_root *root, | |
96 | struct btrfs_path *path, u64 objectid); | |
12fcfd22 CM |
97 | static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans, |
98 | struct btrfs_root *root, | |
99 | struct btrfs_root *log, | |
100 | struct btrfs_path *path, | |
101 | u64 dirid, int del_all); | |
e02119d5 CM |
102 | |
103 | /* | |
104 | * tree logging is a special write ahead log used to make sure that | |
105 | * fsyncs and O_SYNCs can happen without doing full tree commits. | |
106 | * | |
107 | * Full tree commits are expensive because they require commonly | |
108 | * modified blocks to be recowed, creating many dirty pages in the | |
109 | * extent tree an 4x-6x higher write load than ext3. | |
110 | * | |
111 | * Instead of doing a tree commit on every fsync, we use the | |
112 | * key ranges and transaction ids to find items for a given file or directory | |
113 | * that have changed in this transaction. Those items are copied into | |
114 | * a special tree (one per subvolume root), that tree is written to disk | |
115 | * and then the fsync is considered complete. | |
116 | * | |
117 | * After a crash, items are copied out of the log-tree back into the | |
118 | * subvolume tree. Any file data extents found are recorded in the extent | |
119 | * allocation tree, and the log-tree freed. | |
120 | * | |
121 | * The log tree is read three times, once to pin down all the extents it is | |
122 | * using in ram and once, once to create all the inodes logged in the tree | |
123 | * and once to do all the other items. | |
124 | */ | |
125 | ||
e02119d5 CM |
126 | /* |
127 | * start a sub transaction and setup the log tree | |
128 | * this increments the log tree writer count to make the people | |
129 | * syncing the tree wait for us to finish | |
130 | */ | |
131 | static int start_log_trans(struct btrfs_trans_handle *trans, | |
8b050d35 MX |
132 | struct btrfs_root *root, |
133 | struct btrfs_log_ctx *ctx) | |
e02119d5 | 134 | { |
0b246afa | 135 | struct btrfs_fs_info *fs_info = root->fs_info; |
34eb2a52 | 136 | int ret = 0; |
7237f183 YZ |
137 | |
138 | mutex_lock(&root->log_mutex); | |
34eb2a52 | 139 | |
7237f183 | 140 | if (root->log_root) { |
0b246afa | 141 | if (btrfs_need_log_full_commit(fs_info, trans)) { |
50471a38 MX |
142 | ret = -EAGAIN; |
143 | goto out; | |
144 | } | |
34eb2a52 | 145 | |
ff782e0a | 146 | if (!root->log_start_pid) { |
27cdeb70 | 147 | clear_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state); |
34eb2a52 | 148 | root->log_start_pid = current->pid; |
ff782e0a | 149 | } else if (root->log_start_pid != current->pid) { |
27cdeb70 | 150 | set_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state); |
ff782e0a | 151 | } |
34eb2a52 | 152 | } else { |
0b246afa JM |
153 | mutex_lock(&fs_info->tree_log_mutex); |
154 | if (!fs_info->log_root_tree) | |
155 | ret = btrfs_init_log_root_tree(trans, fs_info); | |
156 | mutex_unlock(&fs_info->tree_log_mutex); | |
34eb2a52 Z |
157 | if (ret) |
158 | goto out; | |
ff782e0a | 159 | |
e02119d5 | 160 | ret = btrfs_add_log_tree(trans, root); |
4a500fd1 | 161 | if (ret) |
e87ac136 | 162 | goto out; |
34eb2a52 Z |
163 | |
164 | clear_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state); | |
165 | root->log_start_pid = current->pid; | |
e02119d5 | 166 | } |
34eb2a52 | 167 | |
2ecb7923 | 168 | atomic_inc(&root->log_batch); |
7237f183 | 169 | atomic_inc(&root->log_writers); |
8b050d35 | 170 | if (ctx) { |
34eb2a52 | 171 | int index = root->log_transid % 2; |
8b050d35 | 172 | list_add_tail(&ctx->list, &root->log_ctxs[index]); |
d1433deb | 173 | ctx->log_transid = root->log_transid; |
8b050d35 | 174 | } |
34eb2a52 | 175 | |
e87ac136 | 176 | out: |
7237f183 | 177 | mutex_unlock(&root->log_mutex); |
e87ac136 | 178 | return ret; |
e02119d5 CM |
179 | } |
180 | ||
181 | /* | |
182 | * returns 0 if there was a log transaction running and we were able | |
183 | * to join, or returns -ENOENT if there were not transactions | |
184 | * in progress | |
185 | */ | |
186 | static int join_running_log_trans(struct btrfs_root *root) | |
187 | { | |
188 | int ret = -ENOENT; | |
189 | ||
190 | smp_mb(); | |
191 | if (!root->log_root) | |
192 | return -ENOENT; | |
193 | ||
7237f183 | 194 | mutex_lock(&root->log_mutex); |
e02119d5 CM |
195 | if (root->log_root) { |
196 | ret = 0; | |
7237f183 | 197 | atomic_inc(&root->log_writers); |
e02119d5 | 198 | } |
7237f183 | 199 | mutex_unlock(&root->log_mutex); |
e02119d5 CM |
200 | return ret; |
201 | } | |
202 | ||
12fcfd22 CM |
203 | /* |
204 | * This either makes the current running log transaction wait | |
205 | * until you call btrfs_end_log_trans() or it makes any future | |
206 | * log transactions wait until you call btrfs_end_log_trans() | |
207 | */ | |
208 | int btrfs_pin_log_trans(struct btrfs_root *root) | |
209 | { | |
210 | int ret = -ENOENT; | |
211 | ||
212 | mutex_lock(&root->log_mutex); | |
213 | atomic_inc(&root->log_writers); | |
214 | mutex_unlock(&root->log_mutex); | |
215 | return ret; | |
216 | } | |
217 | ||
e02119d5 CM |
218 | /* |
219 | * indicate we're done making changes to the log tree | |
220 | * and wake up anyone waiting to do a sync | |
221 | */ | |
143bede5 | 222 | void btrfs_end_log_trans(struct btrfs_root *root) |
e02119d5 | 223 | { |
7237f183 | 224 | if (atomic_dec_and_test(&root->log_writers)) { |
093258e6 DS |
225 | /* atomic_dec_and_test implies a barrier */ |
226 | cond_wake_up_nomb(&root->log_writer_wait); | |
7237f183 | 227 | } |
e02119d5 CM |
228 | } |
229 | ||
230 | ||
231 | /* | |
232 | * the walk control struct is used to pass state down the chain when | |
233 | * processing the log tree. The stage field tells us which part | |
234 | * of the log tree processing we are currently doing. The others | |
235 | * are state fields used for that specific part | |
236 | */ | |
237 | struct walk_control { | |
238 | /* should we free the extent on disk when done? This is used | |
239 | * at transaction commit time while freeing a log tree | |
240 | */ | |
241 | int free; | |
242 | ||
243 | /* should we write out the extent buffer? This is used | |
244 | * while flushing the log tree to disk during a sync | |
245 | */ | |
246 | int write; | |
247 | ||
248 | /* should we wait for the extent buffer io to finish? Also used | |
249 | * while flushing the log tree to disk for a sync | |
250 | */ | |
251 | int wait; | |
252 | ||
253 | /* pin only walk, we record which extents on disk belong to the | |
254 | * log trees | |
255 | */ | |
256 | int pin; | |
257 | ||
258 | /* what stage of the replay code we're currently in */ | |
259 | int stage; | |
260 | ||
261 | /* the root we are currently replaying */ | |
262 | struct btrfs_root *replay_dest; | |
263 | ||
264 | /* the trans handle for the current replay */ | |
265 | struct btrfs_trans_handle *trans; | |
266 | ||
267 | /* the function that gets used to process blocks we find in the | |
268 | * tree. Note the extent_buffer might not be up to date when it is | |
269 | * passed in, and it must be checked or read if you need the data | |
270 | * inside it | |
271 | */ | |
272 | int (*process_func)(struct btrfs_root *log, struct extent_buffer *eb, | |
581c1760 | 273 | struct walk_control *wc, u64 gen, int level); |
e02119d5 CM |
274 | }; |
275 | ||
276 | /* | |
277 | * process_func used to pin down extents, write them or wait on them | |
278 | */ | |
279 | static int process_one_buffer(struct btrfs_root *log, | |
280 | struct extent_buffer *eb, | |
581c1760 | 281 | struct walk_control *wc, u64 gen, int level) |
e02119d5 | 282 | { |
0b246afa | 283 | struct btrfs_fs_info *fs_info = log->fs_info; |
b50c6e25 JB |
284 | int ret = 0; |
285 | ||
8c2a1a30 JB |
286 | /* |
287 | * If this fs is mixed then we need to be able to process the leaves to | |
288 | * pin down any logged extents, so we have to read the block. | |
289 | */ | |
0b246afa | 290 | if (btrfs_fs_incompat(fs_info, MIXED_GROUPS)) { |
581c1760 | 291 | ret = btrfs_read_buffer(eb, gen, level, NULL); |
8c2a1a30 JB |
292 | if (ret) |
293 | return ret; | |
294 | } | |
295 | ||
04018de5 | 296 | if (wc->pin) |
2ff7e61e JM |
297 | ret = btrfs_pin_extent_for_log_replay(fs_info, eb->start, |
298 | eb->len); | |
e02119d5 | 299 | |
b50c6e25 | 300 | if (!ret && btrfs_buffer_uptodate(eb, gen, 0)) { |
8c2a1a30 | 301 | if (wc->pin && btrfs_header_level(eb) == 0) |
2ff7e61e | 302 | ret = btrfs_exclude_logged_extents(fs_info, eb); |
e02119d5 CM |
303 | if (wc->write) |
304 | btrfs_write_tree_block(eb); | |
305 | if (wc->wait) | |
306 | btrfs_wait_tree_block_writeback(eb); | |
307 | } | |
b50c6e25 | 308 | return ret; |
e02119d5 CM |
309 | } |
310 | ||
311 | /* | |
312 | * Item overwrite used by replay and tree logging. eb, slot and key all refer | |
313 | * to the src data we are copying out. | |
314 | * | |
315 | * root is the tree we are copying into, and path is a scratch | |
316 | * path for use in this function (it should be released on entry and | |
317 | * will be released on exit). | |
318 | * | |
319 | * If the key is already in the destination tree the existing item is | |
320 | * overwritten. If the existing item isn't big enough, it is extended. | |
321 | * If it is too large, it is truncated. | |
322 | * | |
323 | * If the key isn't in the destination yet, a new item is inserted. | |
324 | */ | |
325 | static noinline int overwrite_item(struct btrfs_trans_handle *trans, | |
326 | struct btrfs_root *root, | |
327 | struct btrfs_path *path, | |
328 | struct extent_buffer *eb, int slot, | |
329 | struct btrfs_key *key) | |
330 | { | |
2ff7e61e | 331 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 CM |
332 | int ret; |
333 | u32 item_size; | |
334 | u64 saved_i_size = 0; | |
335 | int save_old_i_size = 0; | |
336 | unsigned long src_ptr; | |
337 | unsigned long dst_ptr; | |
338 | int overwrite_root = 0; | |
4bc4bee4 | 339 | bool inode_item = key->type == BTRFS_INODE_ITEM_KEY; |
e02119d5 CM |
340 | |
341 | if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) | |
342 | overwrite_root = 1; | |
343 | ||
344 | item_size = btrfs_item_size_nr(eb, slot); | |
345 | src_ptr = btrfs_item_ptr_offset(eb, slot); | |
346 | ||
347 | /* look for the key in the destination tree */ | |
348 | ret = btrfs_search_slot(NULL, root, key, path, 0, 0); | |
4bc4bee4 JB |
349 | if (ret < 0) |
350 | return ret; | |
351 | ||
e02119d5 CM |
352 | if (ret == 0) { |
353 | char *src_copy; | |
354 | char *dst_copy; | |
355 | u32 dst_size = btrfs_item_size_nr(path->nodes[0], | |
356 | path->slots[0]); | |
357 | if (dst_size != item_size) | |
358 | goto insert; | |
359 | ||
360 | if (item_size == 0) { | |
b3b4aa74 | 361 | btrfs_release_path(path); |
e02119d5 CM |
362 | return 0; |
363 | } | |
364 | dst_copy = kmalloc(item_size, GFP_NOFS); | |
365 | src_copy = kmalloc(item_size, GFP_NOFS); | |
2a29edc6 | 366 | if (!dst_copy || !src_copy) { |
b3b4aa74 | 367 | btrfs_release_path(path); |
2a29edc6 | 368 | kfree(dst_copy); |
369 | kfree(src_copy); | |
370 | return -ENOMEM; | |
371 | } | |
e02119d5 CM |
372 | |
373 | read_extent_buffer(eb, src_copy, src_ptr, item_size); | |
374 | ||
375 | dst_ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); | |
376 | read_extent_buffer(path->nodes[0], dst_copy, dst_ptr, | |
377 | item_size); | |
378 | ret = memcmp(dst_copy, src_copy, item_size); | |
379 | ||
380 | kfree(dst_copy); | |
381 | kfree(src_copy); | |
382 | /* | |
383 | * they have the same contents, just return, this saves | |
384 | * us from cowing blocks in the destination tree and doing | |
385 | * extra writes that may not have been done by a previous | |
386 | * sync | |
387 | */ | |
388 | if (ret == 0) { | |
b3b4aa74 | 389 | btrfs_release_path(path); |
e02119d5 CM |
390 | return 0; |
391 | } | |
392 | ||
4bc4bee4 JB |
393 | /* |
394 | * We need to load the old nbytes into the inode so when we | |
395 | * replay the extents we've logged we get the right nbytes. | |
396 | */ | |
397 | if (inode_item) { | |
398 | struct btrfs_inode_item *item; | |
399 | u64 nbytes; | |
d555438b | 400 | u32 mode; |
4bc4bee4 JB |
401 | |
402 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
403 | struct btrfs_inode_item); | |
404 | nbytes = btrfs_inode_nbytes(path->nodes[0], item); | |
405 | item = btrfs_item_ptr(eb, slot, | |
406 | struct btrfs_inode_item); | |
407 | btrfs_set_inode_nbytes(eb, item, nbytes); | |
d555438b JB |
408 | |
409 | /* | |
410 | * If this is a directory we need to reset the i_size to | |
411 | * 0 so that we can set it up properly when replaying | |
412 | * the rest of the items in this log. | |
413 | */ | |
414 | mode = btrfs_inode_mode(eb, item); | |
415 | if (S_ISDIR(mode)) | |
416 | btrfs_set_inode_size(eb, item, 0); | |
4bc4bee4 JB |
417 | } |
418 | } else if (inode_item) { | |
419 | struct btrfs_inode_item *item; | |
d555438b | 420 | u32 mode; |
4bc4bee4 JB |
421 | |
422 | /* | |
423 | * New inode, set nbytes to 0 so that the nbytes comes out | |
424 | * properly when we replay the extents. | |
425 | */ | |
426 | item = btrfs_item_ptr(eb, slot, struct btrfs_inode_item); | |
427 | btrfs_set_inode_nbytes(eb, item, 0); | |
d555438b JB |
428 | |
429 | /* | |
430 | * If this is a directory we need to reset the i_size to 0 so | |
431 | * that we can set it up properly when replaying the rest of | |
432 | * the items in this log. | |
433 | */ | |
434 | mode = btrfs_inode_mode(eb, item); | |
435 | if (S_ISDIR(mode)) | |
436 | btrfs_set_inode_size(eb, item, 0); | |
e02119d5 CM |
437 | } |
438 | insert: | |
b3b4aa74 | 439 | btrfs_release_path(path); |
e02119d5 | 440 | /* try to insert the key into the destination tree */ |
df8d116f | 441 | path->skip_release_on_error = 1; |
e02119d5 CM |
442 | ret = btrfs_insert_empty_item(trans, root, path, |
443 | key, item_size); | |
df8d116f | 444 | path->skip_release_on_error = 0; |
e02119d5 CM |
445 | |
446 | /* make sure any existing item is the correct size */ | |
df8d116f | 447 | if (ret == -EEXIST || ret == -EOVERFLOW) { |
e02119d5 CM |
448 | u32 found_size; |
449 | found_size = btrfs_item_size_nr(path->nodes[0], | |
450 | path->slots[0]); | |
143bede5 | 451 | if (found_size > item_size) |
2ff7e61e | 452 | btrfs_truncate_item(fs_info, path, item_size, 1); |
143bede5 | 453 | else if (found_size < item_size) |
2ff7e61e | 454 | btrfs_extend_item(fs_info, path, |
143bede5 | 455 | item_size - found_size); |
e02119d5 | 456 | } else if (ret) { |
4a500fd1 | 457 | return ret; |
e02119d5 CM |
458 | } |
459 | dst_ptr = btrfs_item_ptr_offset(path->nodes[0], | |
460 | path->slots[0]); | |
461 | ||
462 | /* don't overwrite an existing inode if the generation number | |
463 | * was logged as zero. This is done when the tree logging code | |
464 | * is just logging an inode to make sure it exists after recovery. | |
465 | * | |
466 | * Also, don't overwrite i_size on directories during replay. | |
467 | * log replay inserts and removes directory items based on the | |
468 | * state of the tree found in the subvolume, and i_size is modified | |
469 | * as it goes | |
470 | */ | |
471 | if (key->type == BTRFS_INODE_ITEM_KEY && ret == -EEXIST) { | |
472 | struct btrfs_inode_item *src_item; | |
473 | struct btrfs_inode_item *dst_item; | |
474 | ||
475 | src_item = (struct btrfs_inode_item *)src_ptr; | |
476 | dst_item = (struct btrfs_inode_item *)dst_ptr; | |
477 | ||
1a4bcf47 FM |
478 | if (btrfs_inode_generation(eb, src_item) == 0) { |
479 | struct extent_buffer *dst_eb = path->nodes[0]; | |
2f2ff0ee | 480 | const u64 ino_size = btrfs_inode_size(eb, src_item); |
1a4bcf47 | 481 | |
2f2ff0ee FM |
482 | /* |
483 | * For regular files an ino_size == 0 is used only when | |
484 | * logging that an inode exists, as part of a directory | |
485 | * fsync, and the inode wasn't fsynced before. In this | |
486 | * case don't set the size of the inode in the fs/subvol | |
487 | * tree, otherwise we would be throwing valid data away. | |
488 | */ | |
1a4bcf47 | 489 | if (S_ISREG(btrfs_inode_mode(eb, src_item)) && |
2f2ff0ee FM |
490 | S_ISREG(btrfs_inode_mode(dst_eb, dst_item)) && |
491 | ino_size != 0) { | |
1a4bcf47 | 492 | struct btrfs_map_token token; |
1a4bcf47 FM |
493 | |
494 | btrfs_init_map_token(&token); | |
495 | btrfs_set_token_inode_size(dst_eb, dst_item, | |
496 | ino_size, &token); | |
497 | } | |
e02119d5 | 498 | goto no_copy; |
1a4bcf47 | 499 | } |
e02119d5 CM |
500 | |
501 | if (overwrite_root && | |
502 | S_ISDIR(btrfs_inode_mode(eb, src_item)) && | |
503 | S_ISDIR(btrfs_inode_mode(path->nodes[0], dst_item))) { | |
504 | save_old_i_size = 1; | |
505 | saved_i_size = btrfs_inode_size(path->nodes[0], | |
506 | dst_item); | |
507 | } | |
508 | } | |
509 | ||
510 | copy_extent_buffer(path->nodes[0], eb, dst_ptr, | |
511 | src_ptr, item_size); | |
512 | ||
513 | if (save_old_i_size) { | |
514 | struct btrfs_inode_item *dst_item; | |
515 | dst_item = (struct btrfs_inode_item *)dst_ptr; | |
516 | btrfs_set_inode_size(path->nodes[0], dst_item, saved_i_size); | |
517 | } | |
518 | ||
519 | /* make sure the generation is filled in */ | |
520 | if (key->type == BTRFS_INODE_ITEM_KEY) { | |
521 | struct btrfs_inode_item *dst_item; | |
522 | dst_item = (struct btrfs_inode_item *)dst_ptr; | |
523 | if (btrfs_inode_generation(path->nodes[0], dst_item) == 0) { | |
524 | btrfs_set_inode_generation(path->nodes[0], dst_item, | |
525 | trans->transid); | |
526 | } | |
527 | } | |
528 | no_copy: | |
529 | btrfs_mark_buffer_dirty(path->nodes[0]); | |
b3b4aa74 | 530 | btrfs_release_path(path); |
e02119d5 CM |
531 | return 0; |
532 | } | |
533 | ||
534 | /* | |
535 | * simple helper to read an inode off the disk from a given root | |
536 | * This can only be called for subvolume roots and not for the log | |
537 | */ | |
538 | static noinline struct inode *read_one_inode(struct btrfs_root *root, | |
539 | u64 objectid) | |
540 | { | |
5d4f98a2 | 541 | struct btrfs_key key; |
e02119d5 | 542 | struct inode *inode; |
e02119d5 | 543 | |
5d4f98a2 YZ |
544 | key.objectid = objectid; |
545 | key.type = BTRFS_INODE_ITEM_KEY; | |
546 | key.offset = 0; | |
73f73415 | 547 | inode = btrfs_iget(root->fs_info->sb, &key, root, NULL); |
2e19f1f9 | 548 | if (IS_ERR(inode)) |
5d4f98a2 | 549 | inode = NULL; |
e02119d5 CM |
550 | return inode; |
551 | } | |
552 | ||
553 | /* replays a single extent in 'eb' at 'slot' with 'key' into the | |
554 | * subvolume 'root'. path is released on entry and should be released | |
555 | * on exit. | |
556 | * | |
557 | * extents in the log tree have not been allocated out of the extent | |
558 | * tree yet. So, this completes the allocation, taking a reference | |
559 | * as required if the extent already exists or creating a new extent | |
560 | * if it isn't in the extent allocation tree yet. | |
561 | * | |
562 | * The extent is inserted into the file, dropping any existing extents | |
563 | * from the file that overlap the new one. | |
564 | */ | |
565 | static noinline int replay_one_extent(struct btrfs_trans_handle *trans, | |
566 | struct btrfs_root *root, | |
567 | struct btrfs_path *path, | |
568 | struct extent_buffer *eb, int slot, | |
569 | struct btrfs_key *key) | |
570 | { | |
0b246afa | 571 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 | 572 | int found_type; |
e02119d5 | 573 | u64 extent_end; |
e02119d5 | 574 | u64 start = key->offset; |
4bc4bee4 | 575 | u64 nbytes = 0; |
e02119d5 CM |
576 | struct btrfs_file_extent_item *item; |
577 | struct inode *inode = NULL; | |
578 | unsigned long size; | |
579 | int ret = 0; | |
580 | ||
581 | item = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item); | |
582 | found_type = btrfs_file_extent_type(eb, item); | |
583 | ||
d899e052 | 584 | if (found_type == BTRFS_FILE_EXTENT_REG || |
4bc4bee4 JB |
585 | found_type == BTRFS_FILE_EXTENT_PREALLOC) { |
586 | nbytes = btrfs_file_extent_num_bytes(eb, item); | |
587 | extent_end = start + nbytes; | |
588 | ||
589 | /* | |
590 | * We don't add to the inodes nbytes if we are prealloc or a | |
591 | * hole. | |
592 | */ | |
593 | if (btrfs_file_extent_disk_bytenr(eb, item) == 0) | |
594 | nbytes = 0; | |
595 | } else if (found_type == BTRFS_FILE_EXTENT_INLINE) { | |
e41ca589 | 596 | size = btrfs_file_extent_ram_bytes(eb, item); |
4bc4bee4 | 597 | nbytes = btrfs_file_extent_ram_bytes(eb, item); |
da17066c | 598 | extent_end = ALIGN(start + size, |
0b246afa | 599 | fs_info->sectorsize); |
e02119d5 CM |
600 | } else { |
601 | ret = 0; | |
602 | goto out; | |
603 | } | |
604 | ||
605 | inode = read_one_inode(root, key->objectid); | |
606 | if (!inode) { | |
607 | ret = -EIO; | |
608 | goto out; | |
609 | } | |
610 | ||
611 | /* | |
612 | * first check to see if we already have this extent in the | |
613 | * file. This must be done before the btrfs_drop_extents run | |
614 | * so we don't try to drop this extent. | |
615 | */ | |
f85b7379 DS |
616 | ret = btrfs_lookup_file_extent(trans, root, path, |
617 | btrfs_ino(BTRFS_I(inode)), start, 0); | |
e02119d5 | 618 | |
d899e052 YZ |
619 | if (ret == 0 && |
620 | (found_type == BTRFS_FILE_EXTENT_REG || | |
621 | found_type == BTRFS_FILE_EXTENT_PREALLOC)) { | |
e02119d5 CM |
622 | struct btrfs_file_extent_item cmp1; |
623 | struct btrfs_file_extent_item cmp2; | |
624 | struct btrfs_file_extent_item *existing; | |
625 | struct extent_buffer *leaf; | |
626 | ||
627 | leaf = path->nodes[0]; | |
628 | existing = btrfs_item_ptr(leaf, path->slots[0], | |
629 | struct btrfs_file_extent_item); | |
630 | ||
631 | read_extent_buffer(eb, &cmp1, (unsigned long)item, | |
632 | sizeof(cmp1)); | |
633 | read_extent_buffer(leaf, &cmp2, (unsigned long)existing, | |
634 | sizeof(cmp2)); | |
635 | ||
636 | /* | |
637 | * we already have a pointer to this exact extent, | |
638 | * we don't have to do anything | |
639 | */ | |
640 | if (memcmp(&cmp1, &cmp2, sizeof(cmp1)) == 0) { | |
b3b4aa74 | 641 | btrfs_release_path(path); |
e02119d5 CM |
642 | goto out; |
643 | } | |
644 | } | |
b3b4aa74 | 645 | btrfs_release_path(path); |
e02119d5 CM |
646 | |
647 | /* drop any overlapping extents */ | |
2671485d | 648 | ret = btrfs_drop_extents(trans, root, inode, start, extent_end, 1); |
3650860b JB |
649 | if (ret) |
650 | goto out; | |
e02119d5 | 651 | |
07d400a6 YZ |
652 | if (found_type == BTRFS_FILE_EXTENT_REG || |
653 | found_type == BTRFS_FILE_EXTENT_PREALLOC) { | |
5d4f98a2 | 654 | u64 offset; |
07d400a6 YZ |
655 | unsigned long dest_offset; |
656 | struct btrfs_key ins; | |
657 | ||
3168021c FM |
658 | if (btrfs_file_extent_disk_bytenr(eb, item) == 0 && |
659 | btrfs_fs_incompat(fs_info, NO_HOLES)) | |
660 | goto update_inode; | |
661 | ||
07d400a6 YZ |
662 | ret = btrfs_insert_empty_item(trans, root, path, key, |
663 | sizeof(*item)); | |
3650860b JB |
664 | if (ret) |
665 | goto out; | |
07d400a6 YZ |
666 | dest_offset = btrfs_item_ptr_offset(path->nodes[0], |
667 | path->slots[0]); | |
668 | copy_extent_buffer(path->nodes[0], eb, dest_offset, | |
669 | (unsigned long)item, sizeof(*item)); | |
670 | ||
671 | ins.objectid = btrfs_file_extent_disk_bytenr(eb, item); | |
672 | ins.offset = btrfs_file_extent_disk_num_bytes(eb, item); | |
673 | ins.type = BTRFS_EXTENT_ITEM_KEY; | |
5d4f98a2 | 674 | offset = key->offset - btrfs_file_extent_offset(eb, item); |
07d400a6 | 675 | |
df2c95f3 QW |
676 | /* |
677 | * Manually record dirty extent, as here we did a shallow | |
678 | * file extent item copy and skip normal backref update, | |
679 | * but modifying extent tree all by ourselves. | |
680 | * So need to manually record dirty extent for qgroup, | |
681 | * as the owner of the file extent changed from log tree | |
682 | * (doesn't affect qgroup) to fs/file tree(affects qgroup) | |
683 | */ | |
a95f3aaf | 684 | ret = btrfs_qgroup_trace_extent(trans, |
df2c95f3 QW |
685 | btrfs_file_extent_disk_bytenr(eb, item), |
686 | btrfs_file_extent_disk_num_bytes(eb, item), | |
687 | GFP_NOFS); | |
688 | if (ret < 0) | |
689 | goto out; | |
690 | ||
07d400a6 YZ |
691 | if (ins.objectid > 0) { |
692 | u64 csum_start; | |
693 | u64 csum_end; | |
694 | LIST_HEAD(ordered_sums); | |
695 | /* | |
696 | * is this extent already allocated in the extent | |
697 | * allocation tree? If so, just add a reference | |
698 | */ | |
2ff7e61e | 699 | ret = btrfs_lookup_data_extent(fs_info, ins.objectid, |
07d400a6 YZ |
700 | ins.offset); |
701 | if (ret == 0) { | |
84f7d8e6 | 702 | ret = btrfs_inc_extent_ref(trans, root, |
07d400a6 | 703 | ins.objectid, ins.offset, |
5d4f98a2 | 704 | 0, root->root_key.objectid, |
b06c4bf5 | 705 | key->objectid, offset); |
b50c6e25 JB |
706 | if (ret) |
707 | goto out; | |
07d400a6 YZ |
708 | } else { |
709 | /* | |
710 | * insert the extent pointer in the extent | |
711 | * allocation tree | |
712 | */ | |
5d4f98a2 | 713 | ret = btrfs_alloc_logged_file_extent(trans, |
2ff7e61e | 714 | root->root_key.objectid, |
5d4f98a2 | 715 | key->objectid, offset, &ins); |
b50c6e25 JB |
716 | if (ret) |
717 | goto out; | |
07d400a6 | 718 | } |
b3b4aa74 | 719 | btrfs_release_path(path); |
07d400a6 YZ |
720 | |
721 | if (btrfs_file_extent_compression(eb, item)) { | |
722 | csum_start = ins.objectid; | |
723 | csum_end = csum_start + ins.offset; | |
724 | } else { | |
725 | csum_start = ins.objectid + | |
726 | btrfs_file_extent_offset(eb, item); | |
727 | csum_end = csum_start + | |
728 | btrfs_file_extent_num_bytes(eb, item); | |
729 | } | |
730 | ||
731 | ret = btrfs_lookup_csums_range(root->log_root, | |
732 | csum_start, csum_end - 1, | |
a2de733c | 733 | &ordered_sums, 0); |
3650860b JB |
734 | if (ret) |
735 | goto out; | |
b84b8390 FM |
736 | /* |
737 | * Now delete all existing cums in the csum root that | |
738 | * cover our range. We do this because we can have an | |
739 | * extent that is completely referenced by one file | |
740 | * extent item and partially referenced by another | |
741 | * file extent item (like after using the clone or | |
742 | * extent_same ioctls). In this case if we end up doing | |
743 | * the replay of the one that partially references the | |
744 | * extent first, and we do not do the csum deletion | |
745 | * below, we can get 2 csum items in the csum tree that | |
746 | * overlap each other. For example, imagine our log has | |
747 | * the two following file extent items: | |
748 | * | |
749 | * key (257 EXTENT_DATA 409600) | |
750 | * extent data disk byte 12845056 nr 102400 | |
751 | * extent data offset 20480 nr 20480 ram 102400 | |
752 | * | |
753 | * key (257 EXTENT_DATA 819200) | |
754 | * extent data disk byte 12845056 nr 102400 | |
755 | * extent data offset 0 nr 102400 ram 102400 | |
756 | * | |
757 | * Where the second one fully references the 100K extent | |
758 | * that starts at disk byte 12845056, and the log tree | |
759 | * has a single csum item that covers the entire range | |
760 | * of the extent: | |
761 | * | |
762 | * key (EXTENT_CSUM EXTENT_CSUM 12845056) itemsize 100 | |
763 | * | |
764 | * After the first file extent item is replayed, the | |
765 | * csum tree gets the following csum item: | |
766 | * | |
767 | * key (EXTENT_CSUM EXTENT_CSUM 12865536) itemsize 20 | |
768 | * | |
769 | * Which covers the 20K sub-range starting at offset 20K | |
770 | * of our extent. Now when we replay the second file | |
771 | * extent item, if we do not delete existing csum items | |
772 | * that cover any of its blocks, we end up getting two | |
773 | * csum items in our csum tree that overlap each other: | |
774 | * | |
775 | * key (EXTENT_CSUM EXTENT_CSUM 12845056) itemsize 100 | |
776 | * key (EXTENT_CSUM EXTENT_CSUM 12865536) itemsize 20 | |
777 | * | |
778 | * Which is a problem, because after this anyone trying | |
779 | * to lookup up for the checksum of any block of our | |
780 | * extent starting at an offset of 40K or higher, will | |
781 | * end up looking at the second csum item only, which | |
782 | * does not contain the checksum for any block starting | |
783 | * at offset 40K or higher of our extent. | |
784 | */ | |
07d400a6 YZ |
785 | while (!list_empty(&ordered_sums)) { |
786 | struct btrfs_ordered_sum *sums; | |
787 | sums = list_entry(ordered_sums.next, | |
788 | struct btrfs_ordered_sum, | |
789 | list); | |
b84b8390 | 790 | if (!ret) |
0b246afa | 791 | ret = btrfs_del_csums(trans, fs_info, |
5b4aacef JM |
792 | sums->bytenr, |
793 | sums->len); | |
3650860b JB |
794 | if (!ret) |
795 | ret = btrfs_csum_file_blocks(trans, | |
0b246afa | 796 | fs_info->csum_root, sums); |
07d400a6 YZ |
797 | list_del(&sums->list); |
798 | kfree(sums); | |
799 | } | |
3650860b JB |
800 | if (ret) |
801 | goto out; | |
07d400a6 | 802 | } else { |
b3b4aa74 | 803 | btrfs_release_path(path); |
07d400a6 YZ |
804 | } |
805 | } else if (found_type == BTRFS_FILE_EXTENT_INLINE) { | |
806 | /* inline extents are easy, we just overwrite them */ | |
807 | ret = overwrite_item(trans, root, path, eb, slot, key); | |
3650860b JB |
808 | if (ret) |
809 | goto out; | |
07d400a6 | 810 | } |
e02119d5 | 811 | |
4bc4bee4 | 812 | inode_add_bytes(inode, nbytes); |
3168021c | 813 | update_inode: |
b9959295 | 814 | ret = btrfs_update_inode(trans, root, inode); |
e02119d5 CM |
815 | out: |
816 | if (inode) | |
817 | iput(inode); | |
818 | return ret; | |
819 | } | |
820 | ||
821 | /* | |
822 | * when cleaning up conflicts between the directory names in the | |
823 | * subvolume, directory names in the log and directory names in the | |
824 | * inode back references, we may have to unlink inodes from directories. | |
825 | * | |
826 | * This is a helper function to do the unlink of a specific directory | |
827 | * item | |
828 | */ | |
829 | static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans, | |
830 | struct btrfs_root *root, | |
831 | struct btrfs_path *path, | |
207e7d92 | 832 | struct btrfs_inode *dir, |
e02119d5 CM |
833 | struct btrfs_dir_item *di) |
834 | { | |
835 | struct inode *inode; | |
836 | char *name; | |
837 | int name_len; | |
838 | struct extent_buffer *leaf; | |
839 | struct btrfs_key location; | |
840 | int ret; | |
841 | ||
842 | leaf = path->nodes[0]; | |
843 | ||
844 | btrfs_dir_item_key_to_cpu(leaf, di, &location); | |
845 | name_len = btrfs_dir_name_len(leaf, di); | |
846 | name = kmalloc(name_len, GFP_NOFS); | |
2a29edc6 | 847 | if (!name) |
848 | return -ENOMEM; | |
849 | ||
e02119d5 | 850 | read_extent_buffer(leaf, name, (unsigned long)(di + 1), name_len); |
b3b4aa74 | 851 | btrfs_release_path(path); |
e02119d5 CM |
852 | |
853 | inode = read_one_inode(root, location.objectid); | |
c00e9493 | 854 | if (!inode) { |
3650860b JB |
855 | ret = -EIO; |
856 | goto out; | |
c00e9493 | 857 | } |
e02119d5 | 858 | |
ec051c0f | 859 | ret = link_to_fixup_dir(trans, root, path, location.objectid); |
3650860b JB |
860 | if (ret) |
861 | goto out; | |
12fcfd22 | 862 | |
207e7d92 NB |
863 | ret = btrfs_unlink_inode(trans, root, dir, BTRFS_I(inode), name, |
864 | name_len); | |
3650860b JB |
865 | if (ret) |
866 | goto out; | |
ada9af21 | 867 | else |
e5c304e6 | 868 | ret = btrfs_run_delayed_items(trans); |
3650860b | 869 | out: |
e02119d5 | 870 | kfree(name); |
e02119d5 CM |
871 | iput(inode); |
872 | return ret; | |
873 | } | |
874 | ||
875 | /* | |
876 | * helper function to see if a given name and sequence number found | |
877 | * in an inode back reference are already in a directory and correctly | |
878 | * point to this inode | |
879 | */ | |
880 | static noinline int inode_in_dir(struct btrfs_root *root, | |
881 | struct btrfs_path *path, | |
882 | u64 dirid, u64 objectid, u64 index, | |
883 | const char *name, int name_len) | |
884 | { | |
885 | struct btrfs_dir_item *di; | |
886 | struct btrfs_key location; | |
887 | int match = 0; | |
888 | ||
889 | di = btrfs_lookup_dir_index_item(NULL, root, path, dirid, | |
890 | index, name, name_len, 0); | |
891 | if (di && !IS_ERR(di)) { | |
892 | btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location); | |
893 | if (location.objectid != objectid) | |
894 | goto out; | |
895 | } else | |
896 | goto out; | |
b3b4aa74 | 897 | btrfs_release_path(path); |
e02119d5 CM |
898 | |
899 | di = btrfs_lookup_dir_item(NULL, root, path, dirid, name, name_len, 0); | |
900 | if (di && !IS_ERR(di)) { | |
901 | btrfs_dir_item_key_to_cpu(path->nodes[0], di, &location); | |
902 | if (location.objectid != objectid) | |
903 | goto out; | |
904 | } else | |
905 | goto out; | |
906 | match = 1; | |
907 | out: | |
b3b4aa74 | 908 | btrfs_release_path(path); |
e02119d5 CM |
909 | return match; |
910 | } | |
911 | ||
912 | /* | |
913 | * helper function to check a log tree for a named back reference in | |
914 | * an inode. This is used to decide if a back reference that is | |
915 | * found in the subvolume conflicts with what we find in the log. | |
916 | * | |
917 | * inode backreferences may have multiple refs in a single item, | |
918 | * during replay we process one reference at a time, and we don't | |
919 | * want to delete valid links to a file from the subvolume if that | |
920 | * link is also in the log. | |
921 | */ | |
922 | static noinline int backref_in_log(struct btrfs_root *log, | |
923 | struct btrfs_key *key, | |
f186373f | 924 | u64 ref_objectid, |
df8d116f | 925 | const char *name, int namelen) |
e02119d5 CM |
926 | { |
927 | struct btrfs_path *path; | |
928 | struct btrfs_inode_ref *ref; | |
929 | unsigned long ptr; | |
930 | unsigned long ptr_end; | |
931 | unsigned long name_ptr; | |
932 | int found_name_len; | |
933 | int item_size; | |
934 | int ret; | |
935 | int match = 0; | |
936 | ||
937 | path = btrfs_alloc_path(); | |
2a29edc6 | 938 | if (!path) |
939 | return -ENOMEM; | |
940 | ||
e02119d5 CM |
941 | ret = btrfs_search_slot(NULL, log, key, path, 0, 0); |
942 | if (ret != 0) | |
943 | goto out; | |
944 | ||
e02119d5 | 945 | ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); |
f186373f MF |
946 | |
947 | if (key->type == BTRFS_INODE_EXTREF_KEY) { | |
1f250e92 FM |
948 | if (btrfs_find_name_in_ext_backref(path->nodes[0], |
949 | path->slots[0], | |
950 | ref_objectid, | |
f186373f MF |
951 | name, namelen, NULL)) |
952 | match = 1; | |
953 | ||
954 | goto out; | |
955 | } | |
956 | ||
957 | item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]); | |
e02119d5 CM |
958 | ptr_end = ptr + item_size; |
959 | while (ptr < ptr_end) { | |
960 | ref = (struct btrfs_inode_ref *)ptr; | |
961 | found_name_len = btrfs_inode_ref_name_len(path->nodes[0], ref); | |
962 | if (found_name_len == namelen) { | |
963 | name_ptr = (unsigned long)(ref + 1); | |
964 | ret = memcmp_extent_buffer(path->nodes[0], name, | |
965 | name_ptr, namelen); | |
966 | if (ret == 0) { | |
967 | match = 1; | |
968 | goto out; | |
969 | } | |
970 | } | |
971 | ptr = (unsigned long)(ref + 1) + found_name_len; | |
972 | } | |
973 | out: | |
974 | btrfs_free_path(path); | |
975 | return match; | |
976 | } | |
977 | ||
5a1d7843 | 978 | static inline int __add_inode_ref(struct btrfs_trans_handle *trans, |
e02119d5 | 979 | struct btrfs_root *root, |
e02119d5 | 980 | struct btrfs_path *path, |
5a1d7843 | 981 | struct btrfs_root *log_root, |
94c91a1f NB |
982 | struct btrfs_inode *dir, |
983 | struct btrfs_inode *inode, | |
f186373f MF |
984 | u64 inode_objectid, u64 parent_objectid, |
985 | u64 ref_index, char *name, int namelen, | |
986 | int *search_done) | |
e02119d5 | 987 | { |
34f3e4f2 | 988 | int ret; |
f186373f MF |
989 | char *victim_name; |
990 | int victim_name_len; | |
991 | struct extent_buffer *leaf; | |
5a1d7843 | 992 | struct btrfs_dir_item *di; |
f186373f MF |
993 | struct btrfs_key search_key; |
994 | struct btrfs_inode_extref *extref; | |
c622ae60 | 995 | |
f186373f MF |
996 | again: |
997 | /* Search old style refs */ | |
998 | search_key.objectid = inode_objectid; | |
999 | search_key.type = BTRFS_INODE_REF_KEY; | |
1000 | search_key.offset = parent_objectid; | |
1001 | ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0); | |
e02119d5 | 1002 | if (ret == 0) { |
e02119d5 CM |
1003 | struct btrfs_inode_ref *victim_ref; |
1004 | unsigned long ptr; | |
1005 | unsigned long ptr_end; | |
f186373f MF |
1006 | |
1007 | leaf = path->nodes[0]; | |
e02119d5 CM |
1008 | |
1009 | /* are we trying to overwrite a back ref for the root directory | |
1010 | * if so, just jump out, we're done | |
1011 | */ | |
f186373f | 1012 | if (search_key.objectid == search_key.offset) |
5a1d7843 | 1013 | return 1; |
e02119d5 CM |
1014 | |
1015 | /* check all the names in this back reference to see | |
1016 | * if they are in the log. if so, we allow them to stay | |
1017 | * otherwise they must be unlinked as a conflict | |
1018 | */ | |
1019 | ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
1020 | ptr_end = ptr + btrfs_item_size_nr(leaf, path->slots[0]); | |
d397712b | 1021 | while (ptr < ptr_end) { |
e02119d5 CM |
1022 | victim_ref = (struct btrfs_inode_ref *)ptr; |
1023 | victim_name_len = btrfs_inode_ref_name_len(leaf, | |
1024 | victim_ref); | |
1025 | victim_name = kmalloc(victim_name_len, GFP_NOFS); | |
3650860b JB |
1026 | if (!victim_name) |
1027 | return -ENOMEM; | |
e02119d5 CM |
1028 | |
1029 | read_extent_buffer(leaf, victim_name, | |
1030 | (unsigned long)(victim_ref + 1), | |
1031 | victim_name_len); | |
1032 | ||
f186373f MF |
1033 | if (!backref_in_log(log_root, &search_key, |
1034 | parent_objectid, | |
1035 | victim_name, | |
e02119d5 | 1036 | victim_name_len)) { |
94c91a1f | 1037 | inc_nlink(&inode->vfs_inode); |
b3b4aa74 | 1038 | btrfs_release_path(path); |
12fcfd22 | 1039 | |
94c91a1f | 1040 | ret = btrfs_unlink_inode(trans, root, dir, inode, |
4ec5934e | 1041 | victim_name, victim_name_len); |
f186373f | 1042 | kfree(victim_name); |
3650860b JB |
1043 | if (ret) |
1044 | return ret; | |
e5c304e6 | 1045 | ret = btrfs_run_delayed_items(trans); |
ada9af21 FDBM |
1046 | if (ret) |
1047 | return ret; | |
f186373f MF |
1048 | *search_done = 1; |
1049 | goto again; | |
e02119d5 CM |
1050 | } |
1051 | kfree(victim_name); | |
f186373f | 1052 | |
e02119d5 CM |
1053 | ptr = (unsigned long)(victim_ref + 1) + victim_name_len; |
1054 | } | |
e02119d5 | 1055 | |
c622ae60 | 1056 | /* |
1057 | * NOTE: we have searched root tree and checked the | |
bb7ab3b9 | 1058 | * corresponding ref, it does not need to check again. |
c622ae60 | 1059 | */ |
5a1d7843 | 1060 | *search_done = 1; |
e02119d5 | 1061 | } |
b3b4aa74 | 1062 | btrfs_release_path(path); |
e02119d5 | 1063 | |
f186373f MF |
1064 | /* Same search but for extended refs */ |
1065 | extref = btrfs_lookup_inode_extref(NULL, root, path, name, namelen, | |
1066 | inode_objectid, parent_objectid, 0, | |
1067 | 0); | |
1068 | if (!IS_ERR_OR_NULL(extref)) { | |
1069 | u32 item_size; | |
1070 | u32 cur_offset = 0; | |
1071 | unsigned long base; | |
1072 | struct inode *victim_parent; | |
1073 | ||
1074 | leaf = path->nodes[0]; | |
1075 | ||
1076 | item_size = btrfs_item_size_nr(leaf, path->slots[0]); | |
1077 | base = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
1078 | ||
1079 | while (cur_offset < item_size) { | |
dd9ef135 | 1080 | extref = (struct btrfs_inode_extref *)(base + cur_offset); |
f186373f MF |
1081 | |
1082 | victim_name_len = btrfs_inode_extref_name_len(leaf, extref); | |
1083 | ||
1084 | if (btrfs_inode_extref_parent(leaf, extref) != parent_objectid) | |
1085 | goto next; | |
1086 | ||
1087 | victim_name = kmalloc(victim_name_len, GFP_NOFS); | |
3650860b JB |
1088 | if (!victim_name) |
1089 | return -ENOMEM; | |
f186373f MF |
1090 | read_extent_buffer(leaf, victim_name, (unsigned long)&extref->name, |
1091 | victim_name_len); | |
1092 | ||
1093 | search_key.objectid = inode_objectid; | |
1094 | search_key.type = BTRFS_INODE_EXTREF_KEY; | |
1095 | search_key.offset = btrfs_extref_hash(parent_objectid, | |
1096 | victim_name, | |
1097 | victim_name_len); | |
1098 | ret = 0; | |
1099 | if (!backref_in_log(log_root, &search_key, | |
1100 | parent_objectid, victim_name, | |
1101 | victim_name_len)) { | |
1102 | ret = -ENOENT; | |
1103 | victim_parent = read_one_inode(root, | |
94c91a1f | 1104 | parent_objectid); |
f186373f | 1105 | if (victim_parent) { |
94c91a1f | 1106 | inc_nlink(&inode->vfs_inode); |
f186373f MF |
1107 | btrfs_release_path(path); |
1108 | ||
1109 | ret = btrfs_unlink_inode(trans, root, | |
4ec5934e | 1110 | BTRFS_I(victim_parent), |
94c91a1f | 1111 | inode, |
4ec5934e NB |
1112 | victim_name, |
1113 | victim_name_len); | |
ada9af21 FDBM |
1114 | if (!ret) |
1115 | ret = btrfs_run_delayed_items( | |
e5c304e6 | 1116 | trans); |
f186373f | 1117 | } |
f186373f MF |
1118 | iput(victim_parent); |
1119 | kfree(victim_name); | |
3650860b JB |
1120 | if (ret) |
1121 | return ret; | |
f186373f MF |
1122 | *search_done = 1; |
1123 | goto again; | |
1124 | } | |
1125 | kfree(victim_name); | |
f186373f MF |
1126 | next: |
1127 | cur_offset += victim_name_len + sizeof(*extref); | |
1128 | } | |
1129 | *search_done = 1; | |
1130 | } | |
1131 | btrfs_release_path(path); | |
1132 | ||
34f3e4f2 | 1133 | /* look for a conflicting sequence number */ |
94c91a1f | 1134 | di = btrfs_lookup_dir_index_item(trans, root, path, btrfs_ino(dir), |
f186373f | 1135 | ref_index, name, namelen, 0); |
34f3e4f2 | 1136 | if (di && !IS_ERR(di)) { |
94c91a1f | 1137 | ret = drop_one_dir_item(trans, root, path, dir, di); |
3650860b JB |
1138 | if (ret) |
1139 | return ret; | |
34f3e4f2 | 1140 | } |
1141 | btrfs_release_path(path); | |
1142 | ||
1143 | /* look for a conflicing name */ | |
94c91a1f | 1144 | di = btrfs_lookup_dir_item(trans, root, path, btrfs_ino(dir), |
34f3e4f2 | 1145 | name, namelen, 0); |
1146 | if (di && !IS_ERR(di)) { | |
94c91a1f | 1147 | ret = drop_one_dir_item(trans, root, path, dir, di); |
3650860b JB |
1148 | if (ret) |
1149 | return ret; | |
34f3e4f2 | 1150 | } |
1151 | btrfs_release_path(path); | |
1152 | ||
5a1d7843 JS |
1153 | return 0; |
1154 | } | |
e02119d5 | 1155 | |
bae15d95 QW |
1156 | static int extref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr, |
1157 | u32 *namelen, char **name, u64 *index, | |
1158 | u64 *parent_objectid) | |
f186373f MF |
1159 | { |
1160 | struct btrfs_inode_extref *extref; | |
1161 | ||
1162 | extref = (struct btrfs_inode_extref *)ref_ptr; | |
1163 | ||
1164 | *namelen = btrfs_inode_extref_name_len(eb, extref); | |
1165 | *name = kmalloc(*namelen, GFP_NOFS); | |
1166 | if (*name == NULL) | |
1167 | return -ENOMEM; | |
1168 | ||
1169 | read_extent_buffer(eb, *name, (unsigned long)&extref->name, | |
1170 | *namelen); | |
1171 | ||
1f250e92 FM |
1172 | if (index) |
1173 | *index = btrfs_inode_extref_index(eb, extref); | |
f186373f MF |
1174 | if (parent_objectid) |
1175 | *parent_objectid = btrfs_inode_extref_parent(eb, extref); | |
1176 | ||
1177 | return 0; | |
1178 | } | |
1179 | ||
bae15d95 QW |
1180 | static int ref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr, |
1181 | u32 *namelen, char **name, u64 *index) | |
f186373f MF |
1182 | { |
1183 | struct btrfs_inode_ref *ref; | |
1184 | ||
1185 | ref = (struct btrfs_inode_ref *)ref_ptr; | |
1186 | ||
1187 | *namelen = btrfs_inode_ref_name_len(eb, ref); | |
1188 | *name = kmalloc(*namelen, GFP_NOFS); | |
1189 | if (*name == NULL) | |
1190 | return -ENOMEM; | |
1191 | ||
1192 | read_extent_buffer(eb, *name, (unsigned long)(ref + 1), *namelen); | |
1193 | ||
1f250e92 FM |
1194 | if (index) |
1195 | *index = btrfs_inode_ref_index(eb, ref); | |
f186373f MF |
1196 | |
1197 | return 0; | |
1198 | } | |
1199 | ||
1f250e92 FM |
1200 | /* |
1201 | * Take an inode reference item from the log tree and iterate all names from the | |
1202 | * inode reference item in the subvolume tree with the same key (if it exists). | |
1203 | * For any name that is not in the inode reference item from the log tree, do a | |
1204 | * proper unlink of that name (that is, remove its entry from the inode | |
1205 | * reference item and both dir index keys). | |
1206 | */ | |
1207 | static int unlink_old_inode_refs(struct btrfs_trans_handle *trans, | |
1208 | struct btrfs_root *root, | |
1209 | struct btrfs_path *path, | |
1210 | struct btrfs_inode *inode, | |
1211 | struct extent_buffer *log_eb, | |
1212 | int log_slot, | |
1213 | struct btrfs_key *key) | |
1214 | { | |
1215 | int ret; | |
1216 | unsigned long ref_ptr; | |
1217 | unsigned long ref_end; | |
1218 | struct extent_buffer *eb; | |
1219 | ||
1220 | again: | |
1221 | btrfs_release_path(path); | |
1222 | ret = btrfs_search_slot(NULL, root, key, path, 0, 0); | |
1223 | if (ret > 0) { | |
1224 | ret = 0; | |
1225 | goto out; | |
1226 | } | |
1227 | if (ret < 0) | |
1228 | goto out; | |
1229 | ||
1230 | eb = path->nodes[0]; | |
1231 | ref_ptr = btrfs_item_ptr_offset(eb, path->slots[0]); | |
1232 | ref_end = ref_ptr + btrfs_item_size_nr(eb, path->slots[0]); | |
1233 | while (ref_ptr < ref_end) { | |
1234 | char *name = NULL; | |
1235 | int namelen; | |
1236 | u64 parent_id; | |
1237 | ||
1238 | if (key->type == BTRFS_INODE_EXTREF_KEY) { | |
1239 | ret = extref_get_fields(eb, ref_ptr, &namelen, &name, | |
1240 | NULL, &parent_id); | |
1241 | } else { | |
1242 | parent_id = key->offset; | |
1243 | ret = ref_get_fields(eb, ref_ptr, &namelen, &name, | |
1244 | NULL); | |
1245 | } | |
1246 | if (ret) | |
1247 | goto out; | |
1248 | ||
1249 | if (key->type == BTRFS_INODE_EXTREF_KEY) | |
1250 | ret = btrfs_find_name_in_ext_backref(log_eb, log_slot, | |
1251 | parent_id, name, | |
1252 | namelen, NULL); | |
1253 | else | |
1254 | ret = btrfs_find_name_in_backref(log_eb, log_slot, name, | |
1255 | namelen, NULL); | |
1256 | ||
1257 | if (!ret) { | |
1258 | struct inode *dir; | |
1259 | ||
1260 | btrfs_release_path(path); | |
1261 | dir = read_one_inode(root, parent_id); | |
1262 | if (!dir) { | |
1263 | ret = -ENOENT; | |
1264 | kfree(name); | |
1265 | goto out; | |
1266 | } | |
1267 | ret = btrfs_unlink_inode(trans, root, BTRFS_I(dir), | |
1268 | inode, name, namelen); | |
1269 | kfree(name); | |
1270 | iput(dir); | |
1271 | if (ret) | |
1272 | goto out; | |
1273 | goto again; | |
1274 | } | |
1275 | ||
1276 | kfree(name); | |
1277 | ref_ptr += namelen; | |
1278 | if (key->type == BTRFS_INODE_EXTREF_KEY) | |
1279 | ref_ptr += sizeof(struct btrfs_inode_extref); | |
1280 | else | |
1281 | ref_ptr += sizeof(struct btrfs_inode_ref); | |
1282 | } | |
1283 | ret = 0; | |
1284 | out: | |
1285 | btrfs_release_path(path); | |
1286 | return ret; | |
1287 | } | |
1288 | ||
0d836392 FM |
1289 | static int btrfs_inode_ref_exists(struct inode *inode, struct inode *dir, |
1290 | const u8 ref_type, const char *name, | |
1291 | const int namelen) | |
1292 | { | |
1293 | struct btrfs_key key; | |
1294 | struct btrfs_path *path; | |
1295 | const u64 parent_id = btrfs_ino(BTRFS_I(dir)); | |
1296 | int ret; | |
1297 | ||
1298 | path = btrfs_alloc_path(); | |
1299 | if (!path) | |
1300 | return -ENOMEM; | |
1301 | ||
1302 | key.objectid = btrfs_ino(BTRFS_I(inode)); | |
1303 | key.type = ref_type; | |
1304 | if (key.type == BTRFS_INODE_REF_KEY) | |
1305 | key.offset = parent_id; | |
1306 | else | |
1307 | key.offset = btrfs_extref_hash(parent_id, name, namelen); | |
1308 | ||
1309 | ret = btrfs_search_slot(NULL, BTRFS_I(inode)->root, &key, path, 0, 0); | |
1310 | if (ret < 0) | |
1311 | goto out; | |
1312 | if (ret > 0) { | |
1313 | ret = 0; | |
1314 | goto out; | |
1315 | } | |
1316 | if (key.type == BTRFS_INODE_EXTREF_KEY) | |
1317 | ret = btrfs_find_name_in_ext_backref(path->nodes[0], | |
1318 | path->slots[0], parent_id, | |
1319 | name, namelen, NULL); | |
1320 | else | |
1321 | ret = btrfs_find_name_in_backref(path->nodes[0], path->slots[0], | |
1322 | name, namelen, NULL); | |
1323 | ||
1324 | out: | |
1325 | btrfs_free_path(path); | |
1326 | return ret; | |
1327 | } | |
1328 | ||
5a1d7843 JS |
1329 | /* |
1330 | * replay one inode back reference item found in the log tree. | |
1331 | * eb, slot and key refer to the buffer and key found in the log tree. | |
1332 | * root is the destination we are replaying into, and path is for temp | |
1333 | * use by this function. (it should be released on return). | |
1334 | */ | |
1335 | static noinline int add_inode_ref(struct btrfs_trans_handle *trans, | |
1336 | struct btrfs_root *root, | |
1337 | struct btrfs_root *log, | |
1338 | struct btrfs_path *path, | |
1339 | struct extent_buffer *eb, int slot, | |
1340 | struct btrfs_key *key) | |
1341 | { | |
03b2f08b GB |
1342 | struct inode *dir = NULL; |
1343 | struct inode *inode = NULL; | |
5a1d7843 JS |
1344 | unsigned long ref_ptr; |
1345 | unsigned long ref_end; | |
03b2f08b | 1346 | char *name = NULL; |
5a1d7843 JS |
1347 | int namelen; |
1348 | int ret; | |
1349 | int search_done = 0; | |
f186373f MF |
1350 | int log_ref_ver = 0; |
1351 | u64 parent_objectid; | |
1352 | u64 inode_objectid; | |
f46dbe3d | 1353 | u64 ref_index = 0; |
f186373f MF |
1354 | int ref_struct_size; |
1355 | ||
1356 | ref_ptr = btrfs_item_ptr_offset(eb, slot); | |
1357 | ref_end = ref_ptr + btrfs_item_size_nr(eb, slot); | |
1358 | ||
1359 | if (key->type == BTRFS_INODE_EXTREF_KEY) { | |
1360 | struct btrfs_inode_extref *r; | |
1361 | ||
1362 | ref_struct_size = sizeof(struct btrfs_inode_extref); | |
1363 | log_ref_ver = 1; | |
1364 | r = (struct btrfs_inode_extref *)ref_ptr; | |
1365 | parent_objectid = btrfs_inode_extref_parent(eb, r); | |
1366 | } else { | |
1367 | ref_struct_size = sizeof(struct btrfs_inode_ref); | |
1368 | parent_objectid = key->offset; | |
1369 | } | |
1370 | inode_objectid = key->objectid; | |
e02119d5 | 1371 | |
5a1d7843 JS |
1372 | /* |
1373 | * it is possible that we didn't log all the parent directories | |
1374 | * for a given inode. If we don't find the dir, just don't | |
1375 | * copy the back ref in. The link count fixup code will take | |
1376 | * care of the rest | |
1377 | */ | |
f186373f | 1378 | dir = read_one_inode(root, parent_objectid); |
03b2f08b GB |
1379 | if (!dir) { |
1380 | ret = -ENOENT; | |
1381 | goto out; | |
1382 | } | |
5a1d7843 | 1383 | |
f186373f | 1384 | inode = read_one_inode(root, inode_objectid); |
5a1d7843 | 1385 | if (!inode) { |
03b2f08b GB |
1386 | ret = -EIO; |
1387 | goto out; | |
5a1d7843 JS |
1388 | } |
1389 | ||
5a1d7843 | 1390 | while (ref_ptr < ref_end) { |
f186373f | 1391 | if (log_ref_ver) { |
bae15d95 QW |
1392 | ret = extref_get_fields(eb, ref_ptr, &namelen, &name, |
1393 | &ref_index, &parent_objectid); | |
f186373f MF |
1394 | /* |
1395 | * parent object can change from one array | |
1396 | * item to another. | |
1397 | */ | |
1398 | if (!dir) | |
1399 | dir = read_one_inode(root, parent_objectid); | |
03b2f08b GB |
1400 | if (!dir) { |
1401 | ret = -ENOENT; | |
1402 | goto out; | |
1403 | } | |
f186373f | 1404 | } else { |
bae15d95 QW |
1405 | ret = ref_get_fields(eb, ref_ptr, &namelen, &name, |
1406 | &ref_index); | |
f186373f MF |
1407 | } |
1408 | if (ret) | |
03b2f08b | 1409 | goto out; |
5a1d7843 JS |
1410 | |
1411 | /* if we already have a perfect match, we're done */ | |
f85b7379 DS |
1412 | if (!inode_in_dir(root, path, btrfs_ino(BTRFS_I(dir)), |
1413 | btrfs_ino(BTRFS_I(inode)), ref_index, | |
1414 | name, namelen)) { | |
5a1d7843 JS |
1415 | /* |
1416 | * look for a conflicting back reference in the | |
1417 | * metadata. if we find one we have to unlink that name | |
1418 | * of the file before we add our new link. Later on, we | |
1419 | * overwrite any existing back reference, and we don't | |
1420 | * want to create dangling pointers in the directory. | |
1421 | */ | |
1422 | ||
1423 | if (!search_done) { | |
1424 | ret = __add_inode_ref(trans, root, path, log, | |
94c91a1f | 1425 | BTRFS_I(dir), |
d75eefdf | 1426 | BTRFS_I(inode), |
f186373f MF |
1427 | inode_objectid, |
1428 | parent_objectid, | |
1429 | ref_index, name, namelen, | |
5a1d7843 | 1430 | &search_done); |
03b2f08b GB |
1431 | if (ret) { |
1432 | if (ret == 1) | |
1433 | ret = 0; | |
3650860b JB |
1434 | goto out; |
1435 | } | |
5a1d7843 JS |
1436 | } |
1437 | ||
0d836392 FM |
1438 | /* |
1439 | * If a reference item already exists for this inode | |
1440 | * with the same parent and name, but different index, | |
1441 | * drop it and the corresponding directory index entries | |
1442 | * from the parent before adding the new reference item | |
1443 | * and dir index entries, otherwise we would fail with | |
1444 | * -EEXIST returned from btrfs_add_link() below. | |
1445 | */ | |
1446 | ret = btrfs_inode_ref_exists(inode, dir, key->type, | |
1447 | name, namelen); | |
1448 | if (ret > 0) { | |
1449 | ret = btrfs_unlink_inode(trans, root, | |
1450 | BTRFS_I(dir), | |
1451 | BTRFS_I(inode), | |
1452 | name, namelen); | |
1453 | /* | |
1454 | * If we dropped the link count to 0, bump it so | |
1455 | * that later the iput() on the inode will not | |
1456 | * free it. We will fixup the link count later. | |
1457 | */ | |
1458 | if (!ret && inode->i_nlink == 0) | |
1459 | inc_nlink(inode); | |
1460 | } | |
1461 | if (ret < 0) | |
1462 | goto out; | |
1463 | ||
5a1d7843 | 1464 | /* insert our name */ |
db0a669f NB |
1465 | ret = btrfs_add_link(trans, BTRFS_I(dir), |
1466 | BTRFS_I(inode), | |
1467 | name, namelen, 0, ref_index); | |
3650860b JB |
1468 | if (ret) |
1469 | goto out; | |
5a1d7843 JS |
1470 | |
1471 | btrfs_update_inode(trans, root, inode); | |
1472 | } | |
1473 | ||
f186373f | 1474 | ref_ptr = (unsigned long)(ref_ptr + ref_struct_size) + namelen; |
5a1d7843 | 1475 | kfree(name); |
03b2f08b | 1476 | name = NULL; |
f186373f MF |
1477 | if (log_ref_ver) { |
1478 | iput(dir); | |
1479 | dir = NULL; | |
1480 | } | |
5a1d7843 | 1481 | } |
e02119d5 | 1482 | |
1f250e92 FM |
1483 | /* |
1484 | * Before we overwrite the inode reference item in the subvolume tree | |
1485 | * with the item from the log tree, we must unlink all names from the | |
1486 | * parent directory that are in the subvolume's tree inode reference | |
1487 | * item, otherwise we end up with an inconsistent subvolume tree where | |
1488 | * dir index entries exist for a name but there is no inode reference | |
1489 | * item with the same name. | |
1490 | */ | |
1491 | ret = unlink_old_inode_refs(trans, root, path, BTRFS_I(inode), eb, slot, | |
1492 | key); | |
1493 | if (ret) | |
1494 | goto out; | |
1495 | ||
e02119d5 CM |
1496 | /* finally write the back reference in the inode */ |
1497 | ret = overwrite_item(trans, root, path, eb, slot, key); | |
5a1d7843 | 1498 | out: |
b3b4aa74 | 1499 | btrfs_release_path(path); |
03b2f08b | 1500 | kfree(name); |
e02119d5 CM |
1501 | iput(dir); |
1502 | iput(inode); | |
3650860b | 1503 | return ret; |
e02119d5 CM |
1504 | } |
1505 | ||
c71bf099 | 1506 | static int insert_orphan_item(struct btrfs_trans_handle *trans, |
9c4f61f0 | 1507 | struct btrfs_root *root, u64 ino) |
c71bf099 YZ |
1508 | { |
1509 | int ret; | |
381cf658 | 1510 | |
9c4f61f0 DS |
1511 | ret = btrfs_insert_orphan_item(trans, root, ino); |
1512 | if (ret == -EEXIST) | |
1513 | ret = 0; | |
381cf658 | 1514 | |
c71bf099 YZ |
1515 | return ret; |
1516 | } | |
1517 | ||
f186373f | 1518 | static int count_inode_extrefs(struct btrfs_root *root, |
36283658 | 1519 | struct btrfs_inode *inode, struct btrfs_path *path) |
f186373f MF |
1520 | { |
1521 | int ret = 0; | |
1522 | int name_len; | |
1523 | unsigned int nlink = 0; | |
1524 | u32 item_size; | |
1525 | u32 cur_offset = 0; | |
36283658 | 1526 | u64 inode_objectid = btrfs_ino(inode); |
f186373f MF |
1527 | u64 offset = 0; |
1528 | unsigned long ptr; | |
1529 | struct btrfs_inode_extref *extref; | |
1530 | struct extent_buffer *leaf; | |
1531 | ||
1532 | while (1) { | |
1533 | ret = btrfs_find_one_extref(root, inode_objectid, offset, path, | |
1534 | &extref, &offset); | |
1535 | if (ret) | |
1536 | break; | |
c71bf099 | 1537 | |
f186373f MF |
1538 | leaf = path->nodes[0]; |
1539 | item_size = btrfs_item_size_nr(leaf, path->slots[0]); | |
1540 | ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); | |
2c2c452b | 1541 | cur_offset = 0; |
f186373f MF |
1542 | |
1543 | while (cur_offset < item_size) { | |
1544 | extref = (struct btrfs_inode_extref *) (ptr + cur_offset); | |
1545 | name_len = btrfs_inode_extref_name_len(leaf, extref); | |
1546 | ||
1547 | nlink++; | |
1548 | ||
1549 | cur_offset += name_len + sizeof(*extref); | |
1550 | } | |
1551 | ||
1552 | offset++; | |
1553 | btrfs_release_path(path); | |
1554 | } | |
1555 | btrfs_release_path(path); | |
1556 | ||
2c2c452b | 1557 | if (ret < 0 && ret != -ENOENT) |
f186373f MF |
1558 | return ret; |
1559 | return nlink; | |
1560 | } | |
1561 | ||
1562 | static int count_inode_refs(struct btrfs_root *root, | |
f329e319 | 1563 | struct btrfs_inode *inode, struct btrfs_path *path) |
e02119d5 | 1564 | { |
e02119d5 CM |
1565 | int ret; |
1566 | struct btrfs_key key; | |
f186373f | 1567 | unsigned int nlink = 0; |
e02119d5 CM |
1568 | unsigned long ptr; |
1569 | unsigned long ptr_end; | |
1570 | int name_len; | |
f329e319 | 1571 | u64 ino = btrfs_ino(inode); |
e02119d5 | 1572 | |
33345d01 | 1573 | key.objectid = ino; |
e02119d5 CM |
1574 | key.type = BTRFS_INODE_REF_KEY; |
1575 | key.offset = (u64)-1; | |
1576 | ||
d397712b | 1577 | while (1) { |
e02119d5 CM |
1578 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); |
1579 | if (ret < 0) | |
1580 | break; | |
1581 | if (ret > 0) { | |
1582 | if (path->slots[0] == 0) | |
1583 | break; | |
1584 | path->slots[0]--; | |
1585 | } | |
e93ae26f | 1586 | process_slot: |
e02119d5 CM |
1587 | btrfs_item_key_to_cpu(path->nodes[0], &key, |
1588 | path->slots[0]); | |
33345d01 | 1589 | if (key.objectid != ino || |
e02119d5 CM |
1590 | key.type != BTRFS_INODE_REF_KEY) |
1591 | break; | |
1592 | ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); | |
1593 | ptr_end = ptr + btrfs_item_size_nr(path->nodes[0], | |
1594 | path->slots[0]); | |
d397712b | 1595 | while (ptr < ptr_end) { |
e02119d5 CM |
1596 | struct btrfs_inode_ref *ref; |
1597 | ||
1598 | ref = (struct btrfs_inode_ref *)ptr; | |
1599 | name_len = btrfs_inode_ref_name_len(path->nodes[0], | |
1600 | ref); | |
1601 | ptr = (unsigned long)(ref + 1) + name_len; | |
1602 | nlink++; | |
1603 | } | |
1604 | ||
1605 | if (key.offset == 0) | |
1606 | break; | |
e93ae26f FDBM |
1607 | if (path->slots[0] > 0) { |
1608 | path->slots[0]--; | |
1609 | goto process_slot; | |
1610 | } | |
e02119d5 | 1611 | key.offset--; |
b3b4aa74 | 1612 | btrfs_release_path(path); |
e02119d5 | 1613 | } |
b3b4aa74 | 1614 | btrfs_release_path(path); |
f186373f MF |
1615 | |
1616 | return nlink; | |
1617 | } | |
1618 | ||
1619 | /* | |
1620 | * There are a few corners where the link count of the file can't | |
1621 | * be properly maintained during replay. So, instead of adding | |
1622 | * lots of complexity to the log code, we just scan the backrefs | |
1623 | * for any file that has been through replay. | |
1624 | * | |
1625 | * The scan will update the link count on the inode to reflect the | |
1626 | * number of back refs found. If it goes down to zero, the iput | |
1627 | * will free the inode. | |
1628 | */ | |
1629 | static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans, | |
1630 | struct btrfs_root *root, | |
1631 | struct inode *inode) | |
1632 | { | |
1633 | struct btrfs_path *path; | |
1634 | int ret; | |
1635 | u64 nlink = 0; | |
4a0cc7ca | 1636 | u64 ino = btrfs_ino(BTRFS_I(inode)); |
f186373f MF |
1637 | |
1638 | path = btrfs_alloc_path(); | |
1639 | if (!path) | |
1640 | return -ENOMEM; | |
1641 | ||
f329e319 | 1642 | ret = count_inode_refs(root, BTRFS_I(inode), path); |
f186373f MF |
1643 | if (ret < 0) |
1644 | goto out; | |
1645 | ||
1646 | nlink = ret; | |
1647 | ||
36283658 | 1648 | ret = count_inode_extrefs(root, BTRFS_I(inode), path); |
f186373f MF |
1649 | if (ret < 0) |
1650 | goto out; | |
1651 | ||
1652 | nlink += ret; | |
1653 | ||
1654 | ret = 0; | |
1655 | ||
e02119d5 | 1656 | if (nlink != inode->i_nlink) { |
bfe86848 | 1657 | set_nlink(inode, nlink); |
e02119d5 CM |
1658 | btrfs_update_inode(trans, root, inode); |
1659 | } | |
8d5bf1cb | 1660 | BTRFS_I(inode)->index_cnt = (u64)-1; |
e02119d5 | 1661 | |
c71bf099 YZ |
1662 | if (inode->i_nlink == 0) { |
1663 | if (S_ISDIR(inode->i_mode)) { | |
1664 | ret = replay_dir_deletes(trans, root, NULL, path, | |
33345d01 | 1665 | ino, 1); |
3650860b JB |
1666 | if (ret) |
1667 | goto out; | |
c71bf099 | 1668 | } |
33345d01 | 1669 | ret = insert_orphan_item(trans, root, ino); |
12fcfd22 | 1670 | } |
12fcfd22 | 1671 | |
f186373f MF |
1672 | out: |
1673 | btrfs_free_path(path); | |
1674 | return ret; | |
e02119d5 CM |
1675 | } |
1676 | ||
1677 | static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans, | |
1678 | struct btrfs_root *root, | |
1679 | struct btrfs_path *path) | |
1680 | { | |
1681 | int ret; | |
1682 | struct btrfs_key key; | |
1683 | struct inode *inode; | |
1684 | ||
1685 | key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID; | |
1686 | key.type = BTRFS_ORPHAN_ITEM_KEY; | |
1687 | key.offset = (u64)-1; | |
d397712b | 1688 | while (1) { |
e02119d5 CM |
1689 | ret = btrfs_search_slot(trans, root, &key, path, -1, 1); |
1690 | if (ret < 0) | |
1691 | break; | |
1692 | ||
1693 | if (ret == 1) { | |
1694 | if (path->slots[0] == 0) | |
1695 | break; | |
1696 | path->slots[0]--; | |
1697 | } | |
1698 | ||
1699 | btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); | |
1700 | if (key.objectid != BTRFS_TREE_LOG_FIXUP_OBJECTID || | |
1701 | key.type != BTRFS_ORPHAN_ITEM_KEY) | |
1702 | break; | |
1703 | ||
1704 | ret = btrfs_del_item(trans, root, path); | |
65a246c5 TI |
1705 | if (ret) |
1706 | goto out; | |
e02119d5 | 1707 | |
b3b4aa74 | 1708 | btrfs_release_path(path); |
e02119d5 | 1709 | inode = read_one_inode(root, key.offset); |
c00e9493 TI |
1710 | if (!inode) |
1711 | return -EIO; | |
e02119d5 CM |
1712 | |
1713 | ret = fixup_inode_link_count(trans, root, inode); | |
e02119d5 | 1714 | iput(inode); |
3650860b JB |
1715 | if (ret) |
1716 | goto out; | |
e02119d5 | 1717 | |
12fcfd22 CM |
1718 | /* |
1719 | * fixup on a directory may create new entries, | |
1720 | * make sure we always look for the highset possible | |
1721 | * offset | |
1722 | */ | |
1723 | key.offset = (u64)-1; | |
e02119d5 | 1724 | } |
65a246c5 TI |
1725 | ret = 0; |
1726 | out: | |
b3b4aa74 | 1727 | btrfs_release_path(path); |
65a246c5 | 1728 | return ret; |
e02119d5 CM |
1729 | } |
1730 | ||
1731 | ||
1732 | /* | |
1733 | * record a given inode in the fixup dir so we can check its link | |
1734 | * count when replay is done. The link count is incremented here | |
1735 | * so the inode won't go away until we check it | |
1736 | */ | |
1737 | static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans, | |
1738 | struct btrfs_root *root, | |
1739 | struct btrfs_path *path, | |
1740 | u64 objectid) | |
1741 | { | |
1742 | struct btrfs_key key; | |
1743 | int ret = 0; | |
1744 | struct inode *inode; | |
1745 | ||
1746 | inode = read_one_inode(root, objectid); | |
c00e9493 TI |
1747 | if (!inode) |
1748 | return -EIO; | |
e02119d5 CM |
1749 | |
1750 | key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID; | |
962a298f | 1751 | key.type = BTRFS_ORPHAN_ITEM_KEY; |
e02119d5 CM |
1752 | key.offset = objectid; |
1753 | ||
1754 | ret = btrfs_insert_empty_item(trans, root, path, &key, 0); | |
1755 | ||
b3b4aa74 | 1756 | btrfs_release_path(path); |
e02119d5 | 1757 | if (ret == 0) { |
9bf7a489 JB |
1758 | if (!inode->i_nlink) |
1759 | set_nlink(inode, 1); | |
1760 | else | |
8b558c5f | 1761 | inc_nlink(inode); |
b9959295 | 1762 | ret = btrfs_update_inode(trans, root, inode); |
e02119d5 CM |
1763 | } else if (ret == -EEXIST) { |
1764 | ret = 0; | |
1765 | } else { | |
3650860b | 1766 | BUG(); /* Logic Error */ |
e02119d5 CM |
1767 | } |
1768 | iput(inode); | |
1769 | ||
1770 | return ret; | |
1771 | } | |
1772 | ||
1773 | /* | |
1774 | * when replaying the log for a directory, we only insert names | |
1775 | * for inodes that actually exist. This means an fsync on a directory | |
1776 | * does not implicitly fsync all the new files in it | |
1777 | */ | |
1778 | static noinline int insert_one_name(struct btrfs_trans_handle *trans, | |
1779 | struct btrfs_root *root, | |
e02119d5 | 1780 | u64 dirid, u64 index, |
60d53eb3 | 1781 | char *name, int name_len, |
e02119d5 CM |
1782 | struct btrfs_key *location) |
1783 | { | |
1784 | struct inode *inode; | |
1785 | struct inode *dir; | |
1786 | int ret; | |
1787 | ||
1788 | inode = read_one_inode(root, location->objectid); | |
1789 | if (!inode) | |
1790 | return -ENOENT; | |
1791 | ||
1792 | dir = read_one_inode(root, dirid); | |
1793 | if (!dir) { | |
1794 | iput(inode); | |
1795 | return -EIO; | |
1796 | } | |
d555438b | 1797 | |
db0a669f NB |
1798 | ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), name, |
1799 | name_len, 1, index); | |
e02119d5 CM |
1800 | |
1801 | /* FIXME, put inode into FIXUP list */ | |
1802 | ||
1803 | iput(inode); | |
1804 | iput(dir); | |
1805 | return ret; | |
1806 | } | |
1807 | ||
df8d116f FM |
1808 | /* |
1809 | * Return true if an inode reference exists in the log for the given name, | |
1810 | * inode and parent inode. | |
1811 | */ | |
1812 | static bool name_in_log_ref(struct btrfs_root *log_root, | |
1813 | const char *name, const int name_len, | |
1814 | const u64 dirid, const u64 ino) | |
1815 | { | |
1816 | struct btrfs_key search_key; | |
1817 | ||
1818 | search_key.objectid = ino; | |
1819 | search_key.type = BTRFS_INODE_REF_KEY; | |
1820 | search_key.offset = dirid; | |
1821 | if (backref_in_log(log_root, &search_key, dirid, name, name_len)) | |
1822 | return true; | |
1823 | ||
1824 | search_key.type = BTRFS_INODE_EXTREF_KEY; | |
1825 | search_key.offset = btrfs_extref_hash(dirid, name, name_len); | |
1826 | if (backref_in_log(log_root, &search_key, dirid, name, name_len)) | |
1827 | return true; | |
1828 | ||
1829 | return false; | |
1830 | } | |
1831 | ||
e02119d5 CM |
1832 | /* |
1833 | * take a single entry in a log directory item and replay it into | |
1834 | * the subvolume. | |
1835 | * | |
1836 | * if a conflicting item exists in the subdirectory already, | |
1837 | * the inode it points to is unlinked and put into the link count | |
1838 | * fix up tree. | |
1839 | * | |
1840 | * If a name from the log points to a file or directory that does | |
1841 | * not exist in the FS, it is skipped. fsyncs on directories | |
1842 | * do not force down inodes inside that directory, just changes to the | |
1843 | * names or unlinks in a directory. | |
bb53eda9 FM |
1844 | * |
1845 | * Returns < 0 on error, 0 if the name wasn't replayed (dentry points to a | |
1846 | * non-existing inode) and 1 if the name was replayed. | |
e02119d5 CM |
1847 | */ |
1848 | static noinline int replay_one_name(struct btrfs_trans_handle *trans, | |
1849 | struct btrfs_root *root, | |
1850 | struct btrfs_path *path, | |
1851 | struct extent_buffer *eb, | |
1852 | struct btrfs_dir_item *di, | |
1853 | struct btrfs_key *key) | |
1854 | { | |
1855 | char *name; | |
1856 | int name_len; | |
1857 | struct btrfs_dir_item *dst_di; | |
1858 | struct btrfs_key found_key; | |
1859 | struct btrfs_key log_key; | |
1860 | struct inode *dir; | |
e02119d5 | 1861 | u8 log_type; |
4bef0848 | 1862 | int exists; |
3650860b | 1863 | int ret = 0; |
d555438b | 1864 | bool update_size = (key->type == BTRFS_DIR_INDEX_KEY); |
bb53eda9 | 1865 | bool name_added = false; |
e02119d5 CM |
1866 | |
1867 | dir = read_one_inode(root, key->objectid); | |
c00e9493 TI |
1868 | if (!dir) |
1869 | return -EIO; | |
e02119d5 CM |
1870 | |
1871 | name_len = btrfs_dir_name_len(eb, di); | |
1872 | name = kmalloc(name_len, GFP_NOFS); | |
2bac325e FDBM |
1873 | if (!name) { |
1874 | ret = -ENOMEM; | |
1875 | goto out; | |
1876 | } | |
2a29edc6 | 1877 | |
e02119d5 CM |
1878 | log_type = btrfs_dir_type(eb, di); |
1879 | read_extent_buffer(eb, name, (unsigned long)(di + 1), | |
1880 | name_len); | |
1881 | ||
1882 | btrfs_dir_item_key_to_cpu(eb, di, &log_key); | |
4bef0848 CM |
1883 | exists = btrfs_lookup_inode(trans, root, path, &log_key, 0); |
1884 | if (exists == 0) | |
1885 | exists = 1; | |
1886 | else | |
1887 | exists = 0; | |
b3b4aa74 | 1888 | btrfs_release_path(path); |
4bef0848 | 1889 | |
e02119d5 CM |
1890 | if (key->type == BTRFS_DIR_ITEM_KEY) { |
1891 | dst_di = btrfs_lookup_dir_item(trans, root, path, key->objectid, | |
1892 | name, name_len, 1); | |
d397712b | 1893 | } else if (key->type == BTRFS_DIR_INDEX_KEY) { |
e02119d5 CM |
1894 | dst_di = btrfs_lookup_dir_index_item(trans, root, path, |
1895 | key->objectid, | |
1896 | key->offset, name, | |
1897 | name_len, 1); | |
1898 | } else { | |
3650860b JB |
1899 | /* Corruption */ |
1900 | ret = -EINVAL; | |
1901 | goto out; | |
e02119d5 | 1902 | } |
c704005d | 1903 | if (IS_ERR_OR_NULL(dst_di)) { |
e02119d5 CM |
1904 | /* we need a sequence number to insert, so we only |
1905 | * do inserts for the BTRFS_DIR_INDEX_KEY types | |
1906 | */ | |
1907 | if (key->type != BTRFS_DIR_INDEX_KEY) | |
1908 | goto out; | |
1909 | goto insert; | |
1910 | } | |
1911 | ||
1912 | btrfs_dir_item_key_to_cpu(path->nodes[0], dst_di, &found_key); | |
1913 | /* the existing item matches the logged item */ | |
1914 | if (found_key.objectid == log_key.objectid && | |
1915 | found_key.type == log_key.type && | |
1916 | found_key.offset == log_key.offset && | |
1917 | btrfs_dir_type(path->nodes[0], dst_di) == log_type) { | |
a2cc11db | 1918 | update_size = false; |
e02119d5 CM |
1919 | goto out; |
1920 | } | |
1921 | ||
1922 | /* | |
1923 | * don't drop the conflicting directory entry if the inode | |
1924 | * for the new entry doesn't exist | |
1925 | */ | |
4bef0848 | 1926 | if (!exists) |
e02119d5 CM |
1927 | goto out; |
1928 | ||
207e7d92 | 1929 | ret = drop_one_dir_item(trans, root, path, BTRFS_I(dir), dst_di); |
3650860b JB |
1930 | if (ret) |
1931 | goto out; | |
e02119d5 CM |
1932 | |
1933 | if (key->type == BTRFS_DIR_INDEX_KEY) | |
1934 | goto insert; | |
1935 | out: | |
b3b4aa74 | 1936 | btrfs_release_path(path); |
d555438b | 1937 | if (!ret && update_size) { |
6ef06d27 | 1938 | btrfs_i_size_write(BTRFS_I(dir), dir->i_size + name_len * 2); |
d555438b JB |
1939 | ret = btrfs_update_inode(trans, root, dir); |
1940 | } | |
e02119d5 CM |
1941 | kfree(name); |
1942 | iput(dir); | |
bb53eda9 FM |
1943 | if (!ret && name_added) |
1944 | ret = 1; | |
3650860b | 1945 | return ret; |
e02119d5 CM |
1946 | |
1947 | insert: | |
df8d116f FM |
1948 | if (name_in_log_ref(root->log_root, name, name_len, |
1949 | key->objectid, log_key.objectid)) { | |
1950 | /* The dentry will be added later. */ | |
1951 | ret = 0; | |
1952 | update_size = false; | |
1953 | goto out; | |
1954 | } | |
b3b4aa74 | 1955 | btrfs_release_path(path); |
60d53eb3 Z |
1956 | ret = insert_one_name(trans, root, key->objectid, key->offset, |
1957 | name, name_len, &log_key); | |
df8d116f | 1958 | if (ret && ret != -ENOENT && ret != -EEXIST) |
3650860b | 1959 | goto out; |
bb53eda9 FM |
1960 | if (!ret) |
1961 | name_added = true; | |
d555438b | 1962 | update_size = false; |
3650860b | 1963 | ret = 0; |
e02119d5 CM |
1964 | goto out; |
1965 | } | |
1966 | ||
1967 | /* | |
1968 | * find all the names in a directory item and reconcile them into | |
1969 | * the subvolume. Only BTRFS_DIR_ITEM_KEY types will have more than | |
1970 | * one name in a directory item, but the same code gets used for | |
1971 | * both directory index types | |
1972 | */ | |
1973 | static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans, | |
1974 | struct btrfs_root *root, | |
1975 | struct btrfs_path *path, | |
1976 | struct extent_buffer *eb, int slot, | |
1977 | struct btrfs_key *key) | |
1978 | { | |
bb53eda9 | 1979 | int ret = 0; |
e02119d5 CM |
1980 | u32 item_size = btrfs_item_size_nr(eb, slot); |
1981 | struct btrfs_dir_item *di; | |
1982 | int name_len; | |
1983 | unsigned long ptr; | |
1984 | unsigned long ptr_end; | |
bb53eda9 | 1985 | struct btrfs_path *fixup_path = NULL; |
e02119d5 CM |
1986 | |
1987 | ptr = btrfs_item_ptr_offset(eb, slot); | |
1988 | ptr_end = ptr + item_size; | |
d397712b | 1989 | while (ptr < ptr_end) { |
e02119d5 CM |
1990 | di = (struct btrfs_dir_item *)ptr; |
1991 | name_len = btrfs_dir_name_len(eb, di); | |
1992 | ret = replay_one_name(trans, root, path, eb, di, key); | |
bb53eda9 FM |
1993 | if (ret < 0) |
1994 | break; | |
e02119d5 CM |
1995 | ptr = (unsigned long)(di + 1); |
1996 | ptr += name_len; | |
bb53eda9 FM |
1997 | |
1998 | /* | |
1999 | * If this entry refers to a non-directory (directories can not | |
2000 | * have a link count > 1) and it was added in the transaction | |
2001 | * that was not committed, make sure we fixup the link count of | |
2002 | * the inode it the entry points to. Otherwise something like | |
2003 | * the following would result in a directory pointing to an | |
2004 | * inode with a wrong link that does not account for this dir | |
2005 | * entry: | |
2006 | * | |
2007 | * mkdir testdir | |
2008 | * touch testdir/foo | |
2009 | * touch testdir/bar | |
2010 | * sync | |
2011 | * | |
2012 | * ln testdir/bar testdir/bar_link | |
2013 | * ln testdir/foo testdir/foo_link | |
2014 | * xfs_io -c "fsync" testdir/bar | |
2015 | * | |
2016 | * <power failure> | |
2017 | * | |
2018 | * mount fs, log replay happens | |
2019 | * | |
2020 | * File foo would remain with a link count of 1 when it has two | |
2021 | * entries pointing to it in the directory testdir. This would | |
2022 | * make it impossible to ever delete the parent directory has | |
2023 | * it would result in stale dentries that can never be deleted. | |
2024 | */ | |
2025 | if (ret == 1 && btrfs_dir_type(eb, di) != BTRFS_FT_DIR) { | |
2026 | struct btrfs_key di_key; | |
2027 | ||
2028 | if (!fixup_path) { | |
2029 | fixup_path = btrfs_alloc_path(); | |
2030 | if (!fixup_path) { | |
2031 | ret = -ENOMEM; | |
2032 | break; | |
2033 | } | |
2034 | } | |
2035 | ||
2036 | btrfs_dir_item_key_to_cpu(eb, di, &di_key); | |
2037 | ret = link_to_fixup_dir(trans, root, fixup_path, | |
2038 | di_key.objectid); | |
2039 | if (ret) | |
2040 | break; | |
2041 | } | |
2042 | ret = 0; | |
e02119d5 | 2043 | } |
bb53eda9 FM |
2044 | btrfs_free_path(fixup_path); |
2045 | return ret; | |
e02119d5 CM |
2046 | } |
2047 | ||
2048 | /* | |
2049 | * directory replay has two parts. There are the standard directory | |
2050 | * items in the log copied from the subvolume, and range items | |
2051 | * created in the log while the subvolume was logged. | |
2052 | * | |
2053 | * The range items tell us which parts of the key space the log | |
2054 | * is authoritative for. During replay, if a key in the subvolume | |
2055 | * directory is in a logged range item, but not actually in the log | |
2056 | * that means it was deleted from the directory before the fsync | |
2057 | * and should be removed. | |
2058 | */ | |
2059 | static noinline int find_dir_range(struct btrfs_root *root, | |
2060 | struct btrfs_path *path, | |
2061 | u64 dirid, int key_type, | |
2062 | u64 *start_ret, u64 *end_ret) | |
2063 | { | |
2064 | struct btrfs_key key; | |
2065 | u64 found_end; | |
2066 | struct btrfs_dir_log_item *item; | |
2067 | int ret; | |
2068 | int nritems; | |
2069 | ||
2070 | if (*start_ret == (u64)-1) | |
2071 | return 1; | |
2072 | ||
2073 | key.objectid = dirid; | |
2074 | key.type = key_type; | |
2075 | key.offset = *start_ret; | |
2076 | ||
2077 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
2078 | if (ret < 0) | |
2079 | goto out; | |
2080 | if (ret > 0) { | |
2081 | if (path->slots[0] == 0) | |
2082 | goto out; | |
2083 | path->slots[0]--; | |
2084 | } | |
2085 | if (ret != 0) | |
2086 | btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); | |
2087 | ||
2088 | if (key.type != key_type || key.objectid != dirid) { | |
2089 | ret = 1; | |
2090 | goto next; | |
2091 | } | |
2092 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
2093 | struct btrfs_dir_log_item); | |
2094 | found_end = btrfs_dir_log_end(path->nodes[0], item); | |
2095 | ||
2096 | if (*start_ret >= key.offset && *start_ret <= found_end) { | |
2097 | ret = 0; | |
2098 | *start_ret = key.offset; | |
2099 | *end_ret = found_end; | |
2100 | goto out; | |
2101 | } | |
2102 | ret = 1; | |
2103 | next: | |
2104 | /* check the next slot in the tree to see if it is a valid item */ | |
2105 | nritems = btrfs_header_nritems(path->nodes[0]); | |
2a7bf53f | 2106 | path->slots[0]++; |
e02119d5 CM |
2107 | if (path->slots[0] >= nritems) { |
2108 | ret = btrfs_next_leaf(root, path); | |
2109 | if (ret) | |
2110 | goto out; | |
e02119d5 CM |
2111 | } |
2112 | ||
2113 | btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); | |
2114 | ||
2115 | if (key.type != key_type || key.objectid != dirid) { | |
2116 | ret = 1; | |
2117 | goto out; | |
2118 | } | |
2119 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
2120 | struct btrfs_dir_log_item); | |
2121 | found_end = btrfs_dir_log_end(path->nodes[0], item); | |
2122 | *start_ret = key.offset; | |
2123 | *end_ret = found_end; | |
2124 | ret = 0; | |
2125 | out: | |
b3b4aa74 | 2126 | btrfs_release_path(path); |
e02119d5 CM |
2127 | return ret; |
2128 | } | |
2129 | ||
2130 | /* | |
2131 | * this looks for a given directory item in the log. If the directory | |
2132 | * item is not in the log, the item is removed and the inode it points | |
2133 | * to is unlinked | |
2134 | */ | |
2135 | static noinline int check_item_in_log(struct btrfs_trans_handle *trans, | |
2136 | struct btrfs_root *root, | |
2137 | struct btrfs_root *log, | |
2138 | struct btrfs_path *path, | |
2139 | struct btrfs_path *log_path, | |
2140 | struct inode *dir, | |
2141 | struct btrfs_key *dir_key) | |
2142 | { | |
2143 | int ret; | |
2144 | struct extent_buffer *eb; | |
2145 | int slot; | |
2146 | u32 item_size; | |
2147 | struct btrfs_dir_item *di; | |
2148 | struct btrfs_dir_item *log_di; | |
2149 | int name_len; | |
2150 | unsigned long ptr; | |
2151 | unsigned long ptr_end; | |
2152 | char *name; | |
2153 | struct inode *inode; | |
2154 | struct btrfs_key location; | |
2155 | ||
2156 | again: | |
2157 | eb = path->nodes[0]; | |
2158 | slot = path->slots[0]; | |
2159 | item_size = btrfs_item_size_nr(eb, slot); | |
2160 | ptr = btrfs_item_ptr_offset(eb, slot); | |
2161 | ptr_end = ptr + item_size; | |
d397712b | 2162 | while (ptr < ptr_end) { |
e02119d5 CM |
2163 | di = (struct btrfs_dir_item *)ptr; |
2164 | name_len = btrfs_dir_name_len(eb, di); | |
2165 | name = kmalloc(name_len, GFP_NOFS); | |
2166 | if (!name) { | |
2167 | ret = -ENOMEM; | |
2168 | goto out; | |
2169 | } | |
2170 | read_extent_buffer(eb, name, (unsigned long)(di + 1), | |
2171 | name_len); | |
2172 | log_di = NULL; | |
12fcfd22 | 2173 | if (log && dir_key->type == BTRFS_DIR_ITEM_KEY) { |
e02119d5 CM |
2174 | log_di = btrfs_lookup_dir_item(trans, log, log_path, |
2175 | dir_key->objectid, | |
2176 | name, name_len, 0); | |
12fcfd22 | 2177 | } else if (log && dir_key->type == BTRFS_DIR_INDEX_KEY) { |
e02119d5 CM |
2178 | log_di = btrfs_lookup_dir_index_item(trans, log, |
2179 | log_path, | |
2180 | dir_key->objectid, | |
2181 | dir_key->offset, | |
2182 | name, name_len, 0); | |
2183 | } | |
8d9e220c | 2184 | if (!log_di || log_di == ERR_PTR(-ENOENT)) { |
e02119d5 | 2185 | btrfs_dir_item_key_to_cpu(eb, di, &location); |
b3b4aa74 DS |
2186 | btrfs_release_path(path); |
2187 | btrfs_release_path(log_path); | |
e02119d5 | 2188 | inode = read_one_inode(root, location.objectid); |
c00e9493 TI |
2189 | if (!inode) { |
2190 | kfree(name); | |
2191 | return -EIO; | |
2192 | } | |
e02119d5 CM |
2193 | |
2194 | ret = link_to_fixup_dir(trans, root, | |
2195 | path, location.objectid); | |
3650860b JB |
2196 | if (ret) { |
2197 | kfree(name); | |
2198 | iput(inode); | |
2199 | goto out; | |
2200 | } | |
2201 | ||
8b558c5f | 2202 | inc_nlink(inode); |
4ec5934e NB |
2203 | ret = btrfs_unlink_inode(trans, root, BTRFS_I(dir), |
2204 | BTRFS_I(inode), name, name_len); | |
3650860b | 2205 | if (!ret) |
e5c304e6 | 2206 | ret = btrfs_run_delayed_items(trans); |
e02119d5 CM |
2207 | kfree(name); |
2208 | iput(inode); | |
3650860b JB |
2209 | if (ret) |
2210 | goto out; | |
e02119d5 CM |
2211 | |
2212 | /* there might still be more names under this key | |
2213 | * check and repeat if required | |
2214 | */ | |
2215 | ret = btrfs_search_slot(NULL, root, dir_key, path, | |
2216 | 0, 0); | |
2217 | if (ret == 0) | |
2218 | goto again; | |
2219 | ret = 0; | |
2220 | goto out; | |
269d040f FDBM |
2221 | } else if (IS_ERR(log_di)) { |
2222 | kfree(name); | |
2223 | return PTR_ERR(log_di); | |
e02119d5 | 2224 | } |
b3b4aa74 | 2225 | btrfs_release_path(log_path); |
e02119d5 CM |
2226 | kfree(name); |
2227 | ||
2228 | ptr = (unsigned long)(di + 1); | |
2229 | ptr += name_len; | |
2230 | } | |
2231 | ret = 0; | |
2232 | out: | |
b3b4aa74 DS |
2233 | btrfs_release_path(path); |
2234 | btrfs_release_path(log_path); | |
e02119d5 CM |
2235 | return ret; |
2236 | } | |
2237 | ||
4f764e51 FM |
2238 | static int replay_xattr_deletes(struct btrfs_trans_handle *trans, |
2239 | struct btrfs_root *root, | |
2240 | struct btrfs_root *log, | |
2241 | struct btrfs_path *path, | |
2242 | const u64 ino) | |
2243 | { | |
2244 | struct btrfs_key search_key; | |
2245 | struct btrfs_path *log_path; | |
2246 | int i; | |
2247 | int nritems; | |
2248 | int ret; | |
2249 | ||
2250 | log_path = btrfs_alloc_path(); | |
2251 | if (!log_path) | |
2252 | return -ENOMEM; | |
2253 | ||
2254 | search_key.objectid = ino; | |
2255 | search_key.type = BTRFS_XATTR_ITEM_KEY; | |
2256 | search_key.offset = 0; | |
2257 | again: | |
2258 | ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0); | |
2259 | if (ret < 0) | |
2260 | goto out; | |
2261 | process_leaf: | |
2262 | nritems = btrfs_header_nritems(path->nodes[0]); | |
2263 | for (i = path->slots[0]; i < nritems; i++) { | |
2264 | struct btrfs_key key; | |
2265 | struct btrfs_dir_item *di; | |
2266 | struct btrfs_dir_item *log_di; | |
2267 | u32 total_size; | |
2268 | u32 cur; | |
2269 | ||
2270 | btrfs_item_key_to_cpu(path->nodes[0], &key, i); | |
2271 | if (key.objectid != ino || key.type != BTRFS_XATTR_ITEM_KEY) { | |
2272 | ret = 0; | |
2273 | goto out; | |
2274 | } | |
2275 | ||
2276 | di = btrfs_item_ptr(path->nodes[0], i, struct btrfs_dir_item); | |
2277 | total_size = btrfs_item_size_nr(path->nodes[0], i); | |
2278 | cur = 0; | |
2279 | while (cur < total_size) { | |
2280 | u16 name_len = btrfs_dir_name_len(path->nodes[0], di); | |
2281 | u16 data_len = btrfs_dir_data_len(path->nodes[0], di); | |
2282 | u32 this_len = sizeof(*di) + name_len + data_len; | |
2283 | char *name; | |
2284 | ||
2285 | name = kmalloc(name_len, GFP_NOFS); | |
2286 | if (!name) { | |
2287 | ret = -ENOMEM; | |
2288 | goto out; | |
2289 | } | |
2290 | read_extent_buffer(path->nodes[0], name, | |
2291 | (unsigned long)(di + 1), name_len); | |
2292 | ||
2293 | log_di = btrfs_lookup_xattr(NULL, log, log_path, ino, | |
2294 | name, name_len, 0); | |
2295 | btrfs_release_path(log_path); | |
2296 | if (!log_di) { | |
2297 | /* Doesn't exist in log tree, so delete it. */ | |
2298 | btrfs_release_path(path); | |
2299 | di = btrfs_lookup_xattr(trans, root, path, ino, | |
2300 | name, name_len, -1); | |
2301 | kfree(name); | |
2302 | if (IS_ERR(di)) { | |
2303 | ret = PTR_ERR(di); | |
2304 | goto out; | |
2305 | } | |
2306 | ASSERT(di); | |
2307 | ret = btrfs_delete_one_dir_name(trans, root, | |
2308 | path, di); | |
2309 | if (ret) | |
2310 | goto out; | |
2311 | btrfs_release_path(path); | |
2312 | search_key = key; | |
2313 | goto again; | |
2314 | } | |
2315 | kfree(name); | |
2316 | if (IS_ERR(log_di)) { | |
2317 | ret = PTR_ERR(log_di); | |
2318 | goto out; | |
2319 | } | |
2320 | cur += this_len; | |
2321 | di = (struct btrfs_dir_item *)((char *)di + this_len); | |
2322 | } | |
2323 | } | |
2324 | ret = btrfs_next_leaf(root, path); | |
2325 | if (ret > 0) | |
2326 | ret = 0; | |
2327 | else if (ret == 0) | |
2328 | goto process_leaf; | |
2329 | out: | |
2330 | btrfs_free_path(log_path); | |
2331 | btrfs_release_path(path); | |
2332 | return ret; | |
2333 | } | |
2334 | ||
2335 | ||
e02119d5 CM |
2336 | /* |
2337 | * deletion replay happens before we copy any new directory items | |
2338 | * out of the log or out of backreferences from inodes. It | |
2339 | * scans the log to find ranges of keys that log is authoritative for, | |
2340 | * and then scans the directory to find items in those ranges that are | |
2341 | * not present in the log. | |
2342 | * | |
2343 | * Anything we don't find in the log is unlinked and removed from the | |
2344 | * directory. | |
2345 | */ | |
2346 | static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans, | |
2347 | struct btrfs_root *root, | |
2348 | struct btrfs_root *log, | |
2349 | struct btrfs_path *path, | |
12fcfd22 | 2350 | u64 dirid, int del_all) |
e02119d5 CM |
2351 | { |
2352 | u64 range_start; | |
2353 | u64 range_end; | |
2354 | int key_type = BTRFS_DIR_LOG_ITEM_KEY; | |
2355 | int ret = 0; | |
2356 | struct btrfs_key dir_key; | |
2357 | struct btrfs_key found_key; | |
2358 | struct btrfs_path *log_path; | |
2359 | struct inode *dir; | |
2360 | ||
2361 | dir_key.objectid = dirid; | |
2362 | dir_key.type = BTRFS_DIR_ITEM_KEY; | |
2363 | log_path = btrfs_alloc_path(); | |
2364 | if (!log_path) | |
2365 | return -ENOMEM; | |
2366 | ||
2367 | dir = read_one_inode(root, dirid); | |
2368 | /* it isn't an error if the inode isn't there, that can happen | |
2369 | * because we replay the deletes before we copy in the inode item | |
2370 | * from the log | |
2371 | */ | |
2372 | if (!dir) { | |
2373 | btrfs_free_path(log_path); | |
2374 | return 0; | |
2375 | } | |
2376 | again: | |
2377 | range_start = 0; | |
2378 | range_end = 0; | |
d397712b | 2379 | while (1) { |
12fcfd22 CM |
2380 | if (del_all) |
2381 | range_end = (u64)-1; | |
2382 | else { | |
2383 | ret = find_dir_range(log, path, dirid, key_type, | |
2384 | &range_start, &range_end); | |
2385 | if (ret != 0) | |
2386 | break; | |
2387 | } | |
e02119d5 CM |
2388 | |
2389 | dir_key.offset = range_start; | |
d397712b | 2390 | while (1) { |
e02119d5 CM |
2391 | int nritems; |
2392 | ret = btrfs_search_slot(NULL, root, &dir_key, path, | |
2393 | 0, 0); | |
2394 | if (ret < 0) | |
2395 | goto out; | |
2396 | ||
2397 | nritems = btrfs_header_nritems(path->nodes[0]); | |
2398 | if (path->slots[0] >= nritems) { | |
2399 | ret = btrfs_next_leaf(root, path); | |
b98def7c | 2400 | if (ret == 1) |
e02119d5 | 2401 | break; |
b98def7c LB |
2402 | else if (ret < 0) |
2403 | goto out; | |
e02119d5 CM |
2404 | } |
2405 | btrfs_item_key_to_cpu(path->nodes[0], &found_key, | |
2406 | path->slots[0]); | |
2407 | if (found_key.objectid != dirid || | |
2408 | found_key.type != dir_key.type) | |
2409 | goto next_type; | |
2410 | ||
2411 | if (found_key.offset > range_end) | |
2412 | break; | |
2413 | ||
2414 | ret = check_item_in_log(trans, root, log, path, | |
12fcfd22 CM |
2415 | log_path, dir, |
2416 | &found_key); | |
3650860b JB |
2417 | if (ret) |
2418 | goto out; | |
e02119d5 CM |
2419 | if (found_key.offset == (u64)-1) |
2420 | break; | |
2421 | dir_key.offset = found_key.offset + 1; | |
2422 | } | |
b3b4aa74 | 2423 | btrfs_release_path(path); |
e02119d5 CM |
2424 | if (range_end == (u64)-1) |
2425 | break; | |
2426 | range_start = range_end + 1; | |
2427 | } | |
2428 | ||
2429 | next_type: | |
2430 | ret = 0; | |
2431 | if (key_type == BTRFS_DIR_LOG_ITEM_KEY) { | |
2432 | key_type = BTRFS_DIR_LOG_INDEX_KEY; | |
2433 | dir_key.type = BTRFS_DIR_INDEX_KEY; | |
b3b4aa74 | 2434 | btrfs_release_path(path); |
e02119d5 CM |
2435 | goto again; |
2436 | } | |
2437 | out: | |
b3b4aa74 | 2438 | btrfs_release_path(path); |
e02119d5 CM |
2439 | btrfs_free_path(log_path); |
2440 | iput(dir); | |
2441 | return ret; | |
2442 | } | |
2443 | ||
2444 | /* | |
2445 | * the process_func used to replay items from the log tree. This | |
2446 | * gets called in two different stages. The first stage just looks | |
2447 | * for inodes and makes sure they are all copied into the subvolume. | |
2448 | * | |
2449 | * The second stage copies all the other item types from the log into | |
2450 | * the subvolume. The two stage approach is slower, but gets rid of | |
2451 | * lots of complexity around inodes referencing other inodes that exist | |
2452 | * only in the log (references come from either directory items or inode | |
2453 | * back refs). | |
2454 | */ | |
2455 | static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, | |
581c1760 | 2456 | struct walk_control *wc, u64 gen, int level) |
e02119d5 CM |
2457 | { |
2458 | int nritems; | |
2459 | struct btrfs_path *path; | |
2460 | struct btrfs_root *root = wc->replay_dest; | |
2461 | struct btrfs_key key; | |
e02119d5 CM |
2462 | int i; |
2463 | int ret; | |
2464 | ||
581c1760 | 2465 | ret = btrfs_read_buffer(eb, gen, level, NULL); |
018642a1 TI |
2466 | if (ret) |
2467 | return ret; | |
e02119d5 CM |
2468 | |
2469 | level = btrfs_header_level(eb); | |
2470 | ||
2471 | if (level != 0) | |
2472 | return 0; | |
2473 | ||
2474 | path = btrfs_alloc_path(); | |
1e5063d0 MF |
2475 | if (!path) |
2476 | return -ENOMEM; | |
e02119d5 CM |
2477 | |
2478 | nritems = btrfs_header_nritems(eb); | |
2479 | for (i = 0; i < nritems; i++) { | |
2480 | btrfs_item_key_to_cpu(eb, &key, i); | |
e02119d5 CM |
2481 | |
2482 | /* inode keys are done during the first stage */ | |
2483 | if (key.type == BTRFS_INODE_ITEM_KEY && | |
2484 | wc->stage == LOG_WALK_REPLAY_INODES) { | |
e02119d5 CM |
2485 | struct btrfs_inode_item *inode_item; |
2486 | u32 mode; | |
2487 | ||
2488 | inode_item = btrfs_item_ptr(eb, i, | |
2489 | struct btrfs_inode_item); | |
4f764e51 FM |
2490 | ret = replay_xattr_deletes(wc->trans, root, log, |
2491 | path, key.objectid); | |
2492 | if (ret) | |
2493 | break; | |
e02119d5 CM |
2494 | mode = btrfs_inode_mode(eb, inode_item); |
2495 | if (S_ISDIR(mode)) { | |
2496 | ret = replay_dir_deletes(wc->trans, | |
12fcfd22 | 2497 | root, log, path, key.objectid, 0); |
b50c6e25 JB |
2498 | if (ret) |
2499 | break; | |
e02119d5 CM |
2500 | } |
2501 | ret = overwrite_item(wc->trans, root, path, | |
2502 | eb, i, &key); | |
b50c6e25 JB |
2503 | if (ret) |
2504 | break; | |
e02119d5 | 2505 | |
471d557a FM |
2506 | /* |
2507 | * Before replaying extents, truncate the inode to its | |
2508 | * size. We need to do it now and not after log replay | |
2509 | * because before an fsync we can have prealloc extents | |
2510 | * added beyond the inode's i_size. If we did it after, | |
2511 | * through orphan cleanup for example, we would drop | |
2512 | * those prealloc extents just after replaying them. | |
e02119d5 CM |
2513 | */ |
2514 | if (S_ISREG(mode)) { | |
471d557a FM |
2515 | struct inode *inode; |
2516 | u64 from; | |
2517 | ||
2518 | inode = read_one_inode(root, key.objectid); | |
2519 | if (!inode) { | |
2520 | ret = -EIO; | |
2521 | break; | |
2522 | } | |
2523 | from = ALIGN(i_size_read(inode), | |
2524 | root->fs_info->sectorsize); | |
2525 | ret = btrfs_drop_extents(wc->trans, root, inode, | |
2526 | from, (u64)-1, 1); | |
2527 | /* | |
2528 | * If the nlink count is zero here, the iput | |
2529 | * will free the inode. We bump it to make | |
2530 | * sure it doesn't get freed until the link | |
2531 | * count fixup is done. | |
2532 | */ | |
2533 | if (!ret) { | |
2534 | if (inode->i_nlink == 0) | |
2535 | inc_nlink(inode); | |
2536 | /* Update link count and nbytes. */ | |
2537 | ret = btrfs_update_inode(wc->trans, | |
2538 | root, inode); | |
2539 | } | |
2540 | iput(inode); | |
b50c6e25 JB |
2541 | if (ret) |
2542 | break; | |
e02119d5 | 2543 | } |
c71bf099 | 2544 | |
e02119d5 CM |
2545 | ret = link_to_fixup_dir(wc->trans, root, |
2546 | path, key.objectid); | |
b50c6e25 JB |
2547 | if (ret) |
2548 | break; | |
e02119d5 | 2549 | } |
dd8e7217 JB |
2550 | |
2551 | if (key.type == BTRFS_DIR_INDEX_KEY && | |
2552 | wc->stage == LOG_WALK_REPLAY_DIR_INDEX) { | |
2553 | ret = replay_one_dir_item(wc->trans, root, path, | |
2554 | eb, i, &key); | |
2555 | if (ret) | |
2556 | break; | |
2557 | } | |
2558 | ||
e02119d5 CM |
2559 | if (wc->stage < LOG_WALK_REPLAY_ALL) |
2560 | continue; | |
2561 | ||
2562 | /* these keys are simply copied */ | |
2563 | if (key.type == BTRFS_XATTR_ITEM_KEY) { | |
2564 | ret = overwrite_item(wc->trans, root, path, | |
2565 | eb, i, &key); | |
b50c6e25 JB |
2566 | if (ret) |
2567 | break; | |
2da1c669 LB |
2568 | } else if (key.type == BTRFS_INODE_REF_KEY || |
2569 | key.type == BTRFS_INODE_EXTREF_KEY) { | |
f186373f MF |
2570 | ret = add_inode_ref(wc->trans, root, log, path, |
2571 | eb, i, &key); | |
b50c6e25 JB |
2572 | if (ret && ret != -ENOENT) |
2573 | break; | |
2574 | ret = 0; | |
e02119d5 CM |
2575 | } else if (key.type == BTRFS_EXTENT_DATA_KEY) { |
2576 | ret = replay_one_extent(wc->trans, root, path, | |
2577 | eb, i, &key); | |
b50c6e25 JB |
2578 | if (ret) |
2579 | break; | |
dd8e7217 | 2580 | } else if (key.type == BTRFS_DIR_ITEM_KEY) { |
e02119d5 CM |
2581 | ret = replay_one_dir_item(wc->trans, root, path, |
2582 | eb, i, &key); | |
b50c6e25 JB |
2583 | if (ret) |
2584 | break; | |
e02119d5 CM |
2585 | } |
2586 | } | |
2587 | btrfs_free_path(path); | |
b50c6e25 | 2588 | return ret; |
e02119d5 CM |
2589 | } |
2590 | ||
d397712b | 2591 | static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, |
e02119d5 CM |
2592 | struct btrfs_root *root, |
2593 | struct btrfs_path *path, int *level, | |
2594 | struct walk_control *wc) | |
2595 | { | |
0b246afa | 2596 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 | 2597 | u64 root_owner; |
e02119d5 CM |
2598 | u64 bytenr; |
2599 | u64 ptr_gen; | |
2600 | struct extent_buffer *next; | |
2601 | struct extent_buffer *cur; | |
2602 | struct extent_buffer *parent; | |
2603 | u32 blocksize; | |
2604 | int ret = 0; | |
2605 | ||
2606 | WARN_ON(*level < 0); | |
2607 | WARN_ON(*level >= BTRFS_MAX_LEVEL); | |
2608 | ||
d397712b | 2609 | while (*level > 0) { |
581c1760 QW |
2610 | struct btrfs_key first_key; |
2611 | ||
e02119d5 CM |
2612 | WARN_ON(*level < 0); |
2613 | WARN_ON(*level >= BTRFS_MAX_LEVEL); | |
2614 | cur = path->nodes[*level]; | |
2615 | ||
fae7f21c | 2616 | WARN_ON(btrfs_header_level(cur) != *level); |
e02119d5 CM |
2617 | |
2618 | if (path->slots[*level] >= | |
2619 | btrfs_header_nritems(cur)) | |
2620 | break; | |
2621 | ||
2622 | bytenr = btrfs_node_blockptr(cur, path->slots[*level]); | |
2623 | ptr_gen = btrfs_node_ptr_generation(cur, path->slots[*level]); | |
581c1760 | 2624 | btrfs_node_key_to_cpu(cur, &first_key, path->slots[*level]); |
0b246afa | 2625 | blocksize = fs_info->nodesize; |
e02119d5 CM |
2626 | |
2627 | parent = path->nodes[*level]; | |
2628 | root_owner = btrfs_header_owner(parent); | |
e02119d5 | 2629 | |
2ff7e61e | 2630 | next = btrfs_find_create_tree_block(fs_info, bytenr); |
c871b0f2 LB |
2631 | if (IS_ERR(next)) |
2632 | return PTR_ERR(next); | |
e02119d5 | 2633 | |
e02119d5 | 2634 | if (*level == 1) { |
581c1760 QW |
2635 | ret = wc->process_func(root, next, wc, ptr_gen, |
2636 | *level - 1); | |
b50c6e25 JB |
2637 | if (ret) { |
2638 | free_extent_buffer(next); | |
1e5063d0 | 2639 | return ret; |
b50c6e25 | 2640 | } |
4a500fd1 | 2641 | |
e02119d5 CM |
2642 | path->slots[*level]++; |
2643 | if (wc->free) { | |
581c1760 QW |
2644 | ret = btrfs_read_buffer(next, ptr_gen, |
2645 | *level - 1, &first_key); | |
018642a1 TI |
2646 | if (ret) { |
2647 | free_extent_buffer(next); | |
2648 | return ret; | |
2649 | } | |
e02119d5 | 2650 | |
681ae509 JB |
2651 | if (trans) { |
2652 | btrfs_tree_lock(next); | |
2653 | btrfs_set_lock_blocking(next); | |
7c302b49 | 2654 | clean_tree_block(fs_info, next); |
681ae509 JB |
2655 | btrfs_wait_tree_block_writeback(next); |
2656 | btrfs_tree_unlock(next); | |
1846430c LB |
2657 | } else { |
2658 | if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) | |
2659 | clear_extent_buffer_dirty(next); | |
681ae509 | 2660 | } |
e02119d5 | 2661 | |
e02119d5 CM |
2662 | WARN_ON(root_owner != |
2663 | BTRFS_TREE_LOG_OBJECTID); | |
2ff7e61e JM |
2664 | ret = btrfs_free_and_pin_reserved_extent( |
2665 | fs_info, bytenr, | |
2666 | blocksize); | |
3650860b JB |
2667 | if (ret) { |
2668 | free_extent_buffer(next); | |
2669 | return ret; | |
2670 | } | |
e02119d5 CM |
2671 | } |
2672 | free_extent_buffer(next); | |
2673 | continue; | |
2674 | } | |
581c1760 | 2675 | ret = btrfs_read_buffer(next, ptr_gen, *level - 1, &first_key); |
018642a1 TI |
2676 | if (ret) { |
2677 | free_extent_buffer(next); | |
2678 | return ret; | |
2679 | } | |
e02119d5 CM |
2680 | |
2681 | WARN_ON(*level <= 0); | |
2682 | if (path->nodes[*level-1]) | |
2683 | free_extent_buffer(path->nodes[*level-1]); | |
2684 | path->nodes[*level-1] = next; | |
2685 | *level = btrfs_header_level(next); | |
2686 | path->slots[*level] = 0; | |
2687 | cond_resched(); | |
2688 | } | |
2689 | WARN_ON(*level < 0); | |
2690 | WARN_ON(*level >= BTRFS_MAX_LEVEL); | |
2691 | ||
4a500fd1 | 2692 | path->slots[*level] = btrfs_header_nritems(path->nodes[*level]); |
e02119d5 CM |
2693 | |
2694 | cond_resched(); | |
2695 | return 0; | |
2696 | } | |
2697 | ||
d397712b | 2698 | static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, |
e02119d5 CM |
2699 | struct btrfs_root *root, |
2700 | struct btrfs_path *path, int *level, | |
2701 | struct walk_control *wc) | |
2702 | { | |
0b246afa | 2703 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 | 2704 | u64 root_owner; |
e02119d5 CM |
2705 | int i; |
2706 | int slot; | |
2707 | int ret; | |
2708 | ||
d397712b | 2709 | for (i = *level; i < BTRFS_MAX_LEVEL - 1 && path->nodes[i]; i++) { |
e02119d5 | 2710 | slot = path->slots[i]; |
4a500fd1 | 2711 | if (slot + 1 < btrfs_header_nritems(path->nodes[i])) { |
e02119d5 CM |
2712 | path->slots[i]++; |
2713 | *level = i; | |
2714 | WARN_ON(*level == 0); | |
2715 | return 0; | |
2716 | } else { | |
31840ae1 ZY |
2717 | struct extent_buffer *parent; |
2718 | if (path->nodes[*level] == root->node) | |
2719 | parent = path->nodes[*level]; | |
2720 | else | |
2721 | parent = path->nodes[*level + 1]; | |
2722 | ||
2723 | root_owner = btrfs_header_owner(parent); | |
1e5063d0 | 2724 | ret = wc->process_func(root, path->nodes[*level], wc, |
581c1760 QW |
2725 | btrfs_header_generation(path->nodes[*level]), |
2726 | *level); | |
1e5063d0 MF |
2727 | if (ret) |
2728 | return ret; | |
2729 | ||
e02119d5 CM |
2730 | if (wc->free) { |
2731 | struct extent_buffer *next; | |
2732 | ||
2733 | next = path->nodes[*level]; | |
2734 | ||
681ae509 JB |
2735 | if (trans) { |
2736 | btrfs_tree_lock(next); | |
2737 | btrfs_set_lock_blocking(next); | |
7c302b49 | 2738 | clean_tree_block(fs_info, next); |
681ae509 JB |
2739 | btrfs_wait_tree_block_writeback(next); |
2740 | btrfs_tree_unlock(next); | |
1846430c LB |
2741 | } else { |
2742 | if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) | |
2743 | clear_extent_buffer_dirty(next); | |
681ae509 | 2744 | } |
e02119d5 | 2745 | |
e02119d5 | 2746 | WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID); |
2ff7e61e JM |
2747 | ret = btrfs_free_and_pin_reserved_extent( |
2748 | fs_info, | |
e02119d5 | 2749 | path->nodes[*level]->start, |
d00aff00 | 2750 | path->nodes[*level]->len); |
3650860b JB |
2751 | if (ret) |
2752 | return ret; | |
e02119d5 CM |
2753 | } |
2754 | free_extent_buffer(path->nodes[*level]); | |
2755 | path->nodes[*level] = NULL; | |
2756 | *level = i + 1; | |
2757 | } | |
2758 | } | |
2759 | return 1; | |
2760 | } | |
2761 | ||
2762 | /* | |
2763 | * drop the reference count on the tree rooted at 'snap'. This traverses | |
2764 | * the tree freeing any blocks that have a ref count of zero after being | |
2765 | * decremented. | |
2766 | */ | |
2767 | static int walk_log_tree(struct btrfs_trans_handle *trans, | |
2768 | struct btrfs_root *log, struct walk_control *wc) | |
2769 | { | |
2ff7e61e | 2770 | struct btrfs_fs_info *fs_info = log->fs_info; |
e02119d5 CM |
2771 | int ret = 0; |
2772 | int wret; | |
2773 | int level; | |
2774 | struct btrfs_path *path; | |
e02119d5 CM |
2775 | int orig_level; |
2776 | ||
2777 | path = btrfs_alloc_path(); | |
db5b493a TI |
2778 | if (!path) |
2779 | return -ENOMEM; | |
e02119d5 CM |
2780 | |
2781 | level = btrfs_header_level(log->node); | |
2782 | orig_level = level; | |
2783 | path->nodes[level] = log->node; | |
2784 | extent_buffer_get(log->node); | |
2785 | path->slots[level] = 0; | |
2786 | ||
d397712b | 2787 | while (1) { |
e02119d5 CM |
2788 | wret = walk_down_log_tree(trans, log, path, &level, wc); |
2789 | if (wret > 0) | |
2790 | break; | |
79787eaa | 2791 | if (wret < 0) { |
e02119d5 | 2792 | ret = wret; |
79787eaa JM |
2793 | goto out; |
2794 | } | |
e02119d5 CM |
2795 | |
2796 | wret = walk_up_log_tree(trans, log, path, &level, wc); | |
2797 | if (wret > 0) | |
2798 | break; | |
79787eaa | 2799 | if (wret < 0) { |
e02119d5 | 2800 | ret = wret; |
79787eaa JM |
2801 | goto out; |
2802 | } | |
e02119d5 CM |
2803 | } |
2804 | ||
2805 | /* was the root node processed? if not, catch it here */ | |
2806 | if (path->nodes[orig_level]) { | |
79787eaa | 2807 | ret = wc->process_func(log, path->nodes[orig_level], wc, |
581c1760 QW |
2808 | btrfs_header_generation(path->nodes[orig_level]), |
2809 | orig_level); | |
79787eaa JM |
2810 | if (ret) |
2811 | goto out; | |
e02119d5 CM |
2812 | if (wc->free) { |
2813 | struct extent_buffer *next; | |
2814 | ||
2815 | next = path->nodes[orig_level]; | |
2816 | ||
681ae509 JB |
2817 | if (trans) { |
2818 | btrfs_tree_lock(next); | |
2819 | btrfs_set_lock_blocking(next); | |
7c302b49 | 2820 | clean_tree_block(fs_info, next); |
681ae509 JB |
2821 | btrfs_wait_tree_block_writeback(next); |
2822 | btrfs_tree_unlock(next); | |
1846430c LB |
2823 | } else { |
2824 | if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) | |
2825 | clear_extent_buffer_dirty(next); | |
681ae509 | 2826 | } |
e02119d5 | 2827 | |
e02119d5 CM |
2828 | WARN_ON(log->root_key.objectid != |
2829 | BTRFS_TREE_LOG_OBJECTID); | |
2ff7e61e JM |
2830 | ret = btrfs_free_and_pin_reserved_extent(fs_info, |
2831 | next->start, next->len); | |
3650860b JB |
2832 | if (ret) |
2833 | goto out; | |
e02119d5 CM |
2834 | } |
2835 | } | |
2836 | ||
79787eaa | 2837 | out: |
e02119d5 | 2838 | btrfs_free_path(path); |
e02119d5 CM |
2839 | return ret; |
2840 | } | |
2841 | ||
7237f183 YZ |
2842 | /* |
2843 | * helper function to update the item for a given subvolumes log root | |
2844 | * in the tree of log roots | |
2845 | */ | |
2846 | static int update_log_root(struct btrfs_trans_handle *trans, | |
2847 | struct btrfs_root *log) | |
2848 | { | |
0b246afa | 2849 | struct btrfs_fs_info *fs_info = log->fs_info; |
7237f183 YZ |
2850 | int ret; |
2851 | ||
2852 | if (log->log_transid == 1) { | |
2853 | /* insert root item on the first sync */ | |
0b246afa | 2854 | ret = btrfs_insert_root(trans, fs_info->log_root_tree, |
7237f183 YZ |
2855 | &log->root_key, &log->root_item); |
2856 | } else { | |
0b246afa | 2857 | ret = btrfs_update_root(trans, fs_info->log_root_tree, |
7237f183 YZ |
2858 | &log->root_key, &log->root_item); |
2859 | } | |
2860 | return ret; | |
2861 | } | |
2862 | ||
60d53eb3 | 2863 | static void wait_log_commit(struct btrfs_root *root, int transid) |
e02119d5 CM |
2864 | { |
2865 | DEFINE_WAIT(wait); | |
7237f183 | 2866 | int index = transid % 2; |
e02119d5 | 2867 | |
7237f183 YZ |
2868 | /* |
2869 | * we only allow two pending log transactions at a time, | |
2870 | * so we know that if ours is more than 2 older than the | |
2871 | * current transaction, we're done | |
2872 | */ | |
49e83f57 | 2873 | for (;;) { |
7237f183 YZ |
2874 | prepare_to_wait(&root->log_commit_wait[index], |
2875 | &wait, TASK_UNINTERRUPTIBLE); | |
12fcfd22 | 2876 | |
49e83f57 LB |
2877 | if (!(root->log_transid_committed < transid && |
2878 | atomic_read(&root->log_commit[index]))) | |
2879 | break; | |
12fcfd22 | 2880 | |
49e83f57 LB |
2881 | mutex_unlock(&root->log_mutex); |
2882 | schedule(); | |
7237f183 | 2883 | mutex_lock(&root->log_mutex); |
49e83f57 LB |
2884 | } |
2885 | finish_wait(&root->log_commit_wait[index], &wait); | |
7237f183 YZ |
2886 | } |
2887 | ||
60d53eb3 | 2888 | static void wait_for_writer(struct btrfs_root *root) |
7237f183 YZ |
2889 | { |
2890 | DEFINE_WAIT(wait); | |
8b050d35 | 2891 | |
49e83f57 LB |
2892 | for (;;) { |
2893 | prepare_to_wait(&root->log_writer_wait, &wait, | |
2894 | TASK_UNINTERRUPTIBLE); | |
2895 | if (!atomic_read(&root->log_writers)) | |
2896 | break; | |
2897 | ||
7237f183 | 2898 | mutex_unlock(&root->log_mutex); |
49e83f57 | 2899 | schedule(); |
575849ec | 2900 | mutex_lock(&root->log_mutex); |
7237f183 | 2901 | } |
49e83f57 | 2902 | finish_wait(&root->log_writer_wait, &wait); |
e02119d5 CM |
2903 | } |
2904 | ||
8b050d35 MX |
2905 | static inline void btrfs_remove_log_ctx(struct btrfs_root *root, |
2906 | struct btrfs_log_ctx *ctx) | |
2907 | { | |
2908 | if (!ctx) | |
2909 | return; | |
2910 | ||
2911 | mutex_lock(&root->log_mutex); | |
2912 | list_del_init(&ctx->list); | |
2913 | mutex_unlock(&root->log_mutex); | |
2914 | } | |
2915 | ||
2916 | /* | |
2917 | * Invoked in log mutex context, or be sure there is no other task which | |
2918 | * can access the list. | |
2919 | */ | |
2920 | static inline void btrfs_remove_all_log_ctxs(struct btrfs_root *root, | |
2921 | int index, int error) | |
2922 | { | |
2923 | struct btrfs_log_ctx *ctx; | |
570dd450 | 2924 | struct btrfs_log_ctx *safe; |
8b050d35 | 2925 | |
570dd450 CM |
2926 | list_for_each_entry_safe(ctx, safe, &root->log_ctxs[index], list) { |
2927 | list_del_init(&ctx->list); | |
8b050d35 | 2928 | ctx->log_ret = error; |
570dd450 | 2929 | } |
8b050d35 MX |
2930 | |
2931 | INIT_LIST_HEAD(&root->log_ctxs[index]); | |
2932 | } | |
2933 | ||
e02119d5 CM |
2934 | /* |
2935 | * btrfs_sync_log does sends a given tree log down to the disk and | |
2936 | * updates the super blocks to record it. When this call is done, | |
12fcfd22 CM |
2937 | * you know that any inodes previously logged are safely on disk only |
2938 | * if it returns 0. | |
2939 | * | |
2940 | * Any other return value means you need to call btrfs_commit_transaction. | |
2941 | * Some of the edge cases for fsyncing directories that have had unlinks | |
2942 | * or renames done in the past mean that sometimes the only safe | |
2943 | * fsync is to commit the whole FS. When btrfs_sync_log returns -EAGAIN, | |
2944 | * that has happened. | |
e02119d5 CM |
2945 | */ |
2946 | int btrfs_sync_log(struct btrfs_trans_handle *trans, | |
8b050d35 | 2947 | struct btrfs_root *root, struct btrfs_log_ctx *ctx) |
e02119d5 | 2948 | { |
7237f183 YZ |
2949 | int index1; |
2950 | int index2; | |
8cef4e16 | 2951 | int mark; |
e02119d5 | 2952 | int ret; |
0b246afa | 2953 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 | 2954 | struct btrfs_root *log = root->log_root; |
0b246afa | 2955 | struct btrfs_root *log_root_tree = fs_info->log_root_tree; |
bb14a59b | 2956 | int log_transid = 0; |
8b050d35 | 2957 | struct btrfs_log_ctx root_log_ctx; |
c6adc9cc | 2958 | struct blk_plug plug; |
e02119d5 | 2959 | |
7237f183 | 2960 | mutex_lock(&root->log_mutex); |
d1433deb MX |
2961 | log_transid = ctx->log_transid; |
2962 | if (root->log_transid_committed >= log_transid) { | |
2963 | mutex_unlock(&root->log_mutex); | |
2964 | return ctx->log_ret; | |
2965 | } | |
2966 | ||
2967 | index1 = log_transid % 2; | |
7237f183 | 2968 | if (atomic_read(&root->log_commit[index1])) { |
60d53eb3 | 2969 | wait_log_commit(root, log_transid); |
7237f183 | 2970 | mutex_unlock(&root->log_mutex); |
8b050d35 | 2971 | return ctx->log_ret; |
e02119d5 | 2972 | } |
d1433deb | 2973 | ASSERT(log_transid == root->log_transid); |
7237f183 YZ |
2974 | atomic_set(&root->log_commit[index1], 1); |
2975 | ||
2976 | /* wait for previous tree log sync to complete */ | |
2977 | if (atomic_read(&root->log_commit[(index1 + 1) % 2])) | |
60d53eb3 | 2978 | wait_log_commit(root, log_transid - 1); |
48cab2e0 | 2979 | |
86df7eb9 | 2980 | while (1) { |
2ecb7923 | 2981 | int batch = atomic_read(&root->log_batch); |
cd354ad6 | 2982 | /* when we're on an ssd, just kick the log commit out */ |
0b246afa | 2983 | if (!btrfs_test_opt(fs_info, SSD) && |
27cdeb70 | 2984 | test_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state)) { |
86df7eb9 YZ |
2985 | mutex_unlock(&root->log_mutex); |
2986 | schedule_timeout_uninterruptible(1); | |
2987 | mutex_lock(&root->log_mutex); | |
2988 | } | |
60d53eb3 | 2989 | wait_for_writer(root); |
2ecb7923 | 2990 | if (batch == atomic_read(&root->log_batch)) |
e02119d5 CM |
2991 | break; |
2992 | } | |
e02119d5 | 2993 | |
12fcfd22 | 2994 | /* bail out if we need to do a full commit */ |
0b246afa | 2995 | if (btrfs_need_log_full_commit(fs_info, trans)) { |
12fcfd22 CM |
2996 | ret = -EAGAIN; |
2997 | mutex_unlock(&root->log_mutex); | |
2998 | goto out; | |
2999 | } | |
3000 | ||
8cef4e16 YZ |
3001 | if (log_transid % 2 == 0) |
3002 | mark = EXTENT_DIRTY; | |
3003 | else | |
3004 | mark = EXTENT_NEW; | |
3005 | ||
690587d1 CM |
3006 | /* we start IO on all the marked extents here, but we don't actually |
3007 | * wait for them until later. | |
3008 | */ | |
c6adc9cc | 3009 | blk_start_plug(&plug); |
2ff7e61e | 3010 | ret = btrfs_write_marked_extents(fs_info, &log->dirty_log_pages, mark); |
79787eaa | 3011 | if (ret) { |
c6adc9cc | 3012 | blk_finish_plug(&plug); |
66642832 | 3013 | btrfs_abort_transaction(trans, ret); |
0b246afa | 3014 | btrfs_set_log_full_commit(fs_info, trans); |
79787eaa JM |
3015 | mutex_unlock(&root->log_mutex); |
3016 | goto out; | |
3017 | } | |
7237f183 | 3018 | |
5d4f98a2 | 3019 | btrfs_set_root_node(&log->root_item, log->node); |
7237f183 | 3020 | |
7237f183 YZ |
3021 | root->log_transid++; |
3022 | log->log_transid = root->log_transid; | |
ff782e0a | 3023 | root->log_start_pid = 0; |
7237f183 | 3024 | /* |
8cef4e16 YZ |
3025 | * IO has been started, blocks of the log tree have WRITTEN flag set |
3026 | * in their headers. new modifications of the log will be written to | |
3027 | * new positions. so it's safe to allow log writers to go in. | |
7237f183 YZ |
3028 | */ |
3029 | mutex_unlock(&root->log_mutex); | |
3030 | ||
28a23593 | 3031 | btrfs_init_log_ctx(&root_log_ctx, NULL); |
d1433deb | 3032 | |
7237f183 | 3033 | mutex_lock(&log_root_tree->log_mutex); |
2ecb7923 | 3034 | atomic_inc(&log_root_tree->log_batch); |
7237f183 | 3035 | atomic_inc(&log_root_tree->log_writers); |
d1433deb MX |
3036 | |
3037 | index2 = log_root_tree->log_transid % 2; | |
3038 | list_add_tail(&root_log_ctx.list, &log_root_tree->log_ctxs[index2]); | |
3039 | root_log_ctx.log_transid = log_root_tree->log_transid; | |
3040 | ||
7237f183 YZ |
3041 | mutex_unlock(&log_root_tree->log_mutex); |
3042 | ||
3043 | ret = update_log_root(trans, log); | |
7237f183 YZ |
3044 | |
3045 | mutex_lock(&log_root_tree->log_mutex); | |
3046 | if (atomic_dec_and_test(&log_root_tree->log_writers)) { | |
093258e6 DS |
3047 | /* atomic_dec_and_test implies a barrier */ |
3048 | cond_wake_up_nomb(&log_root_tree->log_writer_wait); | |
7237f183 YZ |
3049 | } |
3050 | ||
4a500fd1 | 3051 | if (ret) { |
d1433deb MX |
3052 | if (!list_empty(&root_log_ctx.list)) |
3053 | list_del_init(&root_log_ctx.list); | |
3054 | ||
c6adc9cc | 3055 | blk_finish_plug(&plug); |
0b246afa | 3056 | btrfs_set_log_full_commit(fs_info, trans); |
995946dd | 3057 | |
79787eaa | 3058 | if (ret != -ENOSPC) { |
66642832 | 3059 | btrfs_abort_transaction(trans, ret); |
79787eaa JM |
3060 | mutex_unlock(&log_root_tree->log_mutex); |
3061 | goto out; | |
3062 | } | |
bf89d38f | 3063 | btrfs_wait_tree_log_extents(log, mark); |
4a500fd1 YZ |
3064 | mutex_unlock(&log_root_tree->log_mutex); |
3065 | ret = -EAGAIN; | |
3066 | goto out; | |
3067 | } | |
3068 | ||
d1433deb | 3069 | if (log_root_tree->log_transid_committed >= root_log_ctx.log_transid) { |
3da5ab56 | 3070 | blk_finish_plug(&plug); |
cbd60aa7 | 3071 | list_del_init(&root_log_ctx.list); |
d1433deb MX |
3072 | mutex_unlock(&log_root_tree->log_mutex); |
3073 | ret = root_log_ctx.log_ret; | |
3074 | goto out; | |
3075 | } | |
8b050d35 | 3076 | |
d1433deb | 3077 | index2 = root_log_ctx.log_transid % 2; |
7237f183 | 3078 | if (atomic_read(&log_root_tree->log_commit[index2])) { |
c6adc9cc | 3079 | blk_finish_plug(&plug); |
bf89d38f | 3080 | ret = btrfs_wait_tree_log_extents(log, mark); |
60d53eb3 | 3081 | wait_log_commit(log_root_tree, |
d1433deb | 3082 | root_log_ctx.log_transid); |
7237f183 | 3083 | mutex_unlock(&log_root_tree->log_mutex); |
5ab5e44a FM |
3084 | if (!ret) |
3085 | ret = root_log_ctx.log_ret; | |
7237f183 YZ |
3086 | goto out; |
3087 | } | |
d1433deb | 3088 | ASSERT(root_log_ctx.log_transid == log_root_tree->log_transid); |
7237f183 YZ |
3089 | atomic_set(&log_root_tree->log_commit[index2], 1); |
3090 | ||
12fcfd22 | 3091 | if (atomic_read(&log_root_tree->log_commit[(index2 + 1) % 2])) { |
60d53eb3 | 3092 | wait_log_commit(log_root_tree, |
d1433deb | 3093 | root_log_ctx.log_transid - 1); |
12fcfd22 CM |
3094 | } |
3095 | ||
60d53eb3 | 3096 | wait_for_writer(log_root_tree); |
7237f183 | 3097 | |
12fcfd22 CM |
3098 | /* |
3099 | * now that we've moved on to the tree of log tree roots, | |
3100 | * check the full commit flag again | |
3101 | */ | |
0b246afa | 3102 | if (btrfs_need_log_full_commit(fs_info, trans)) { |
c6adc9cc | 3103 | blk_finish_plug(&plug); |
bf89d38f | 3104 | btrfs_wait_tree_log_extents(log, mark); |
12fcfd22 CM |
3105 | mutex_unlock(&log_root_tree->log_mutex); |
3106 | ret = -EAGAIN; | |
3107 | goto out_wake_log_root; | |
3108 | } | |
7237f183 | 3109 | |
2ff7e61e | 3110 | ret = btrfs_write_marked_extents(fs_info, |
c6adc9cc MX |
3111 | &log_root_tree->dirty_log_pages, |
3112 | EXTENT_DIRTY | EXTENT_NEW); | |
3113 | blk_finish_plug(&plug); | |
79787eaa | 3114 | if (ret) { |
0b246afa | 3115 | btrfs_set_log_full_commit(fs_info, trans); |
66642832 | 3116 | btrfs_abort_transaction(trans, ret); |
79787eaa JM |
3117 | mutex_unlock(&log_root_tree->log_mutex); |
3118 | goto out_wake_log_root; | |
3119 | } | |
bf89d38f | 3120 | ret = btrfs_wait_tree_log_extents(log, mark); |
5ab5e44a | 3121 | if (!ret) |
bf89d38f JM |
3122 | ret = btrfs_wait_tree_log_extents(log_root_tree, |
3123 | EXTENT_NEW | EXTENT_DIRTY); | |
5ab5e44a | 3124 | if (ret) { |
0b246afa | 3125 | btrfs_set_log_full_commit(fs_info, trans); |
5ab5e44a FM |
3126 | mutex_unlock(&log_root_tree->log_mutex); |
3127 | goto out_wake_log_root; | |
3128 | } | |
e02119d5 | 3129 | |
0b246afa JM |
3130 | btrfs_set_super_log_root(fs_info->super_for_commit, |
3131 | log_root_tree->node->start); | |
3132 | btrfs_set_super_log_root_level(fs_info->super_for_commit, | |
3133 | btrfs_header_level(log_root_tree->node)); | |
e02119d5 | 3134 | |
7237f183 | 3135 | log_root_tree->log_transid++; |
7237f183 YZ |
3136 | mutex_unlock(&log_root_tree->log_mutex); |
3137 | ||
3138 | /* | |
3139 | * nobody else is going to jump in and write the the ctree | |
3140 | * super here because the log_commit atomic below is protecting | |
3141 | * us. We must be called with a transaction handle pinning | |
3142 | * the running transaction open, so a full commit can't hop | |
3143 | * in and cause problems either. | |
3144 | */ | |
eece6a9c | 3145 | ret = write_all_supers(fs_info, 1); |
5af3e8cc | 3146 | if (ret) { |
0b246afa | 3147 | btrfs_set_log_full_commit(fs_info, trans); |
66642832 | 3148 | btrfs_abort_transaction(trans, ret); |
5af3e8cc SB |
3149 | goto out_wake_log_root; |
3150 | } | |
7237f183 | 3151 | |
257c62e1 CM |
3152 | mutex_lock(&root->log_mutex); |
3153 | if (root->last_log_commit < log_transid) | |
3154 | root->last_log_commit = log_transid; | |
3155 | mutex_unlock(&root->log_mutex); | |
3156 | ||
12fcfd22 | 3157 | out_wake_log_root: |
570dd450 | 3158 | mutex_lock(&log_root_tree->log_mutex); |
8b050d35 MX |
3159 | btrfs_remove_all_log_ctxs(log_root_tree, index2, ret); |
3160 | ||
d1433deb | 3161 | log_root_tree->log_transid_committed++; |
7237f183 | 3162 | atomic_set(&log_root_tree->log_commit[index2], 0); |
d1433deb MX |
3163 | mutex_unlock(&log_root_tree->log_mutex); |
3164 | ||
33a9eca7 | 3165 | /* |
093258e6 DS |
3166 | * The barrier before waitqueue_active (in cond_wake_up) is needed so |
3167 | * all the updates above are seen by the woken threads. It might not be | |
3168 | * necessary, but proving that seems to be hard. | |
33a9eca7 | 3169 | */ |
093258e6 | 3170 | cond_wake_up(&log_root_tree->log_commit_wait[index2]); |
e02119d5 | 3171 | out: |
d1433deb | 3172 | mutex_lock(&root->log_mutex); |
570dd450 | 3173 | btrfs_remove_all_log_ctxs(root, index1, ret); |
d1433deb | 3174 | root->log_transid_committed++; |
7237f183 | 3175 | atomic_set(&root->log_commit[index1], 0); |
d1433deb | 3176 | mutex_unlock(&root->log_mutex); |
8b050d35 | 3177 | |
33a9eca7 | 3178 | /* |
093258e6 DS |
3179 | * The barrier before waitqueue_active (in cond_wake_up) is needed so |
3180 | * all the updates above are seen by the woken threads. It might not be | |
3181 | * necessary, but proving that seems to be hard. | |
33a9eca7 | 3182 | */ |
093258e6 | 3183 | cond_wake_up(&root->log_commit_wait[index1]); |
b31eabd8 | 3184 | return ret; |
e02119d5 CM |
3185 | } |
3186 | ||
4a500fd1 YZ |
3187 | static void free_log_tree(struct btrfs_trans_handle *trans, |
3188 | struct btrfs_root *log) | |
e02119d5 CM |
3189 | { |
3190 | int ret; | |
d0c803c4 CM |
3191 | u64 start; |
3192 | u64 end; | |
e02119d5 CM |
3193 | struct walk_control wc = { |
3194 | .free = 1, | |
3195 | .process_func = process_one_buffer | |
3196 | }; | |
3197 | ||
681ae509 JB |
3198 | ret = walk_log_tree(trans, log, &wc); |
3199 | /* I don't think this can happen but just in case */ | |
3200 | if (ret) | |
66642832 | 3201 | btrfs_abort_transaction(trans, ret); |
e02119d5 | 3202 | |
d397712b | 3203 | while (1) { |
d0c803c4 | 3204 | ret = find_first_extent_bit(&log->dirty_log_pages, |
55237a5f LB |
3205 | 0, &start, &end, |
3206 | EXTENT_DIRTY | EXTENT_NEW | EXTENT_NEED_WAIT, | |
e6138876 | 3207 | NULL); |
d0c803c4 CM |
3208 | if (ret) |
3209 | break; | |
3210 | ||
8cef4e16 | 3211 | clear_extent_bits(&log->dirty_log_pages, start, end, |
55237a5f | 3212 | EXTENT_DIRTY | EXTENT_NEW | EXTENT_NEED_WAIT); |
d0c803c4 CM |
3213 | } |
3214 | ||
7237f183 YZ |
3215 | free_extent_buffer(log->node); |
3216 | kfree(log); | |
4a500fd1 YZ |
3217 | } |
3218 | ||
3219 | /* | |
3220 | * free all the extents used by the tree log. This should be called | |
3221 | * at commit time of the full transaction | |
3222 | */ | |
3223 | int btrfs_free_log(struct btrfs_trans_handle *trans, struct btrfs_root *root) | |
3224 | { | |
3225 | if (root->log_root) { | |
3226 | free_log_tree(trans, root->log_root); | |
3227 | root->log_root = NULL; | |
3228 | } | |
3229 | return 0; | |
3230 | } | |
3231 | ||
3232 | int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans, | |
3233 | struct btrfs_fs_info *fs_info) | |
3234 | { | |
3235 | if (fs_info->log_root_tree) { | |
3236 | free_log_tree(trans, fs_info->log_root_tree); | |
3237 | fs_info->log_root_tree = NULL; | |
3238 | } | |
e02119d5 CM |
3239 | return 0; |
3240 | } | |
3241 | ||
e02119d5 CM |
3242 | /* |
3243 | * If both a file and directory are logged, and unlinks or renames are | |
3244 | * mixed in, we have a few interesting corners: | |
3245 | * | |
3246 | * create file X in dir Y | |
3247 | * link file X to X.link in dir Y | |
3248 | * fsync file X | |
3249 | * unlink file X but leave X.link | |
3250 | * fsync dir Y | |
3251 | * | |
3252 | * After a crash we would expect only X.link to exist. But file X | |
3253 | * didn't get fsync'd again so the log has back refs for X and X.link. | |
3254 | * | |
3255 | * We solve this by removing directory entries and inode backrefs from the | |
3256 | * log when a file that was logged in the current transaction is | |
3257 | * unlinked. Any later fsync will include the updated log entries, and | |
3258 | * we'll be able to reconstruct the proper directory items from backrefs. | |
3259 | * | |
3260 | * This optimizations allows us to avoid relogging the entire inode | |
3261 | * or the entire directory. | |
3262 | */ | |
3263 | int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans, | |
3264 | struct btrfs_root *root, | |
3265 | const char *name, int name_len, | |
49f34d1f | 3266 | struct btrfs_inode *dir, u64 index) |
e02119d5 CM |
3267 | { |
3268 | struct btrfs_root *log; | |
3269 | struct btrfs_dir_item *di; | |
3270 | struct btrfs_path *path; | |
3271 | int ret; | |
4a500fd1 | 3272 | int err = 0; |
e02119d5 | 3273 | int bytes_del = 0; |
49f34d1f | 3274 | u64 dir_ino = btrfs_ino(dir); |
e02119d5 | 3275 | |
49f34d1f | 3276 | if (dir->logged_trans < trans->transid) |
3a5f1d45 CM |
3277 | return 0; |
3278 | ||
e02119d5 CM |
3279 | ret = join_running_log_trans(root); |
3280 | if (ret) | |
3281 | return 0; | |
3282 | ||
49f34d1f | 3283 | mutex_lock(&dir->log_mutex); |
e02119d5 CM |
3284 | |
3285 | log = root->log_root; | |
3286 | path = btrfs_alloc_path(); | |
a62f44a5 TI |
3287 | if (!path) { |
3288 | err = -ENOMEM; | |
3289 | goto out_unlock; | |
3290 | } | |
2a29edc6 | 3291 | |
33345d01 | 3292 | di = btrfs_lookup_dir_item(trans, log, path, dir_ino, |
e02119d5 | 3293 | name, name_len, -1); |
4a500fd1 YZ |
3294 | if (IS_ERR(di)) { |
3295 | err = PTR_ERR(di); | |
3296 | goto fail; | |
3297 | } | |
3298 | if (di) { | |
e02119d5 CM |
3299 | ret = btrfs_delete_one_dir_name(trans, log, path, di); |
3300 | bytes_del += name_len; | |
3650860b JB |
3301 | if (ret) { |
3302 | err = ret; | |
3303 | goto fail; | |
3304 | } | |
e02119d5 | 3305 | } |
b3b4aa74 | 3306 | btrfs_release_path(path); |
33345d01 | 3307 | di = btrfs_lookup_dir_index_item(trans, log, path, dir_ino, |
e02119d5 | 3308 | index, name, name_len, -1); |
4a500fd1 YZ |
3309 | if (IS_ERR(di)) { |
3310 | err = PTR_ERR(di); | |
3311 | goto fail; | |
3312 | } | |
3313 | if (di) { | |
e02119d5 CM |
3314 | ret = btrfs_delete_one_dir_name(trans, log, path, di); |
3315 | bytes_del += name_len; | |
3650860b JB |
3316 | if (ret) { |
3317 | err = ret; | |
3318 | goto fail; | |
3319 | } | |
e02119d5 CM |
3320 | } |
3321 | ||
3322 | /* update the directory size in the log to reflect the names | |
3323 | * we have removed | |
3324 | */ | |
3325 | if (bytes_del) { | |
3326 | struct btrfs_key key; | |
3327 | ||
33345d01 | 3328 | key.objectid = dir_ino; |
e02119d5 CM |
3329 | key.offset = 0; |
3330 | key.type = BTRFS_INODE_ITEM_KEY; | |
b3b4aa74 | 3331 | btrfs_release_path(path); |
e02119d5 CM |
3332 | |
3333 | ret = btrfs_search_slot(trans, log, &key, path, 0, 1); | |
4a500fd1 YZ |
3334 | if (ret < 0) { |
3335 | err = ret; | |
3336 | goto fail; | |
3337 | } | |
e02119d5 CM |
3338 | if (ret == 0) { |
3339 | struct btrfs_inode_item *item; | |
3340 | u64 i_size; | |
3341 | ||
3342 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
3343 | struct btrfs_inode_item); | |
3344 | i_size = btrfs_inode_size(path->nodes[0], item); | |
3345 | if (i_size > bytes_del) | |
3346 | i_size -= bytes_del; | |
3347 | else | |
3348 | i_size = 0; | |
3349 | btrfs_set_inode_size(path->nodes[0], item, i_size); | |
3350 | btrfs_mark_buffer_dirty(path->nodes[0]); | |
3351 | } else | |
3352 | ret = 0; | |
b3b4aa74 | 3353 | btrfs_release_path(path); |
e02119d5 | 3354 | } |
4a500fd1 | 3355 | fail: |
e02119d5 | 3356 | btrfs_free_path(path); |
a62f44a5 | 3357 | out_unlock: |
49f34d1f | 3358 | mutex_unlock(&dir->log_mutex); |
4a500fd1 | 3359 | if (ret == -ENOSPC) { |
995946dd | 3360 | btrfs_set_log_full_commit(root->fs_info, trans); |
4a500fd1 | 3361 | ret = 0; |
79787eaa | 3362 | } else if (ret < 0) |
66642832 | 3363 | btrfs_abort_transaction(trans, ret); |
79787eaa | 3364 | |
12fcfd22 | 3365 | btrfs_end_log_trans(root); |
e02119d5 | 3366 | |
411fc6bc | 3367 | return err; |
e02119d5 CM |
3368 | } |
3369 | ||
3370 | /* see comments for btrfs_del_dir_entries_in_log */ | |
3371 | int btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans, | |
3372 | struct btrfs_root *root, | |
3373 | const char *name, int name_len, | |
a491abb2 | 3374 | struct btrfs_inode *inode, u64 dirid) |
e02119d5 | 3375 | { |
0b246afa | 3376 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 CM |
3377 | struct btrfs_root *log; |
3378 | u64 index; | |
3379 | int ret; | |
3380 | ||
a491abb2 | 3381 | if (inode->logged_trans < trans->transid) |
3a5f1d45 CM |
3382 | return 0; |
3383 | ||
e02119d5 CM |
3384 | ret = join_running_log_trans(root); |
3385 | if (ret) | |
3386 | return 0; | |
3387 | log = root->log_root; | |
a491abb2 | 3388 | mutex_lock(&inode->log_mutex); |
e02119d5 | 3389 | |
a491abb2 | 3390 | ret = btrfs_del_inode_ref(trans, log, name, name_len, btrfs_ino(inode), |
e02119d5 | 3391 | dirid, &index); |
a491abb2 | 3392 | mutex_unlock(&inode->log_mutex); |
4a500fd1 | 3393 | if (ret == -ENOSPC) { |
0b246afa | 3394 | btrfs_set_log_full_commit(fs_info, trans); |
4a500fd1 | 3395 | ret = 0; |
79787eaa | 3396 | } else if (ret < 0 && ret != -ENOENT) |
66642832 | 3397 | btrfs_abort_transaction(trans, ret); |
12fcfd22 | 3398 | btrfs_end_log_trans(root); |
e02119d5 | 3399 | |
e02119d5 CM |
3400 | return ret; |
3401 | } | |
3402 | ||
3403 | /* | |
3404 | * creates a range item in the log for 'dirid'. first_offset and | |
3405 | * last_offset tell us which parts of the key space the log should | |
3406 | * be considered authoritative for. | |
3407 | */ | |
3408 | static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans, | |
3409 | struct btrfs_root *log, | |
3410 | struct btrfs_path *path, | |
3411 | int key_type, u64 dirid, | |
3412 | u64 first_offset, u64 last_offset) | |
3413 | { | |
3414 | int ret; | |
3415 | struct btrfs_key key; | |
3416 | struct btrfs_dir_log_item *item; | |
3417 | ||
3418 | key.objectid = dirid; | |
3419 | key.offset = first_offset; | |
3420 | if (key_type == BTRFS_DIR_ITEM_KEY) | |
3421 | key.type = BTRFS_DIR_LOG_ITEM_KEY; | |
3422 | else | |
3423 | key.type = BTRFS_DIR_LOG_INDEX_KEY; | |
3424 | ret = btrfs_insert_empty_item(trans, log, path, &key, sizeof(*item)); | |
4a500fd1 YZ |
3425 | if (ret) |
3426 | return ret; | |
e02119d5 CM |
3427 | |
3428 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
3429 | struct btrfs_dir_log_item); | |
3430 | btrfs_set_dir_log_end(path->nodes[0], item, last_offset); | |
3431 | btrfs_mark_buffer_dirty(path->nodes[0]); | |
b3b4aa74 | 3432 | btrfs_release_path(path); |
e02119d5 CM |
3433 | return 0; |
3434 | } | |
3435 | ||
3436 | /* | |
3437 | * log all the items included in the current transaction for a given | |
3438 | * directory. This also creates the range items in the log tree required | |
3439 | * to replay anything deleted before the fsync | |
3440 | */ | |
3441 | static noinline int log_dir_items(struct btrfs_trans_handle *trans, | |
684a5773 | 3442 | struct btrfs_root *root, struct btrfs_inode *inode, |
e02119d5 CM |
3443 | struct btrfs_path *path, |
3444 | struct btrfs_path *dst_path, int key_type, | |
2f2ff0ee | 3445 | struct btrfs_log_ctx *ctx, |
e02119d5 CM |
3446 | u64 min_offset, u64 *last_offset_ret) |
3447 | { | |
3448 | struct btrfs_key min_key; | |
e02119d5 CM |
3449 | struct btrfs_root *log = root->log_root; |
3450 | struct extent_buffer *src; | |
4a500fd1 | 3451 | int err = 0; |
e02119d5 CM |
3452 | int ret; |
3453 | int i; | |
3454 | int nritems; | |
3455 | u64 first_offset = min_offset; | |
3456 | u64 last_offset = (u64)-1; | |
684a5773 | 3457 | u64 ino = btrfs_ino(inode); |
e02119d5 CM |
3458 | |
3459 | log = root->log_root; | |
e02119d5 | 3460 | |
33345d01 | 3461 | min_key.objectid = ino; |
e02119d5 CM |
3462 | min_key.type = key_type; |
3463 | min_key.offset = min_offset; | |
3464 | ||
6174d3cb | 3465 | ret = btrfs_search_forward(root, &min_key, path, trans->transid); |
e02119d5 CM |
3466 | |
3467 | /* | |
3468 | * we didn't find anything from this transaction, see if there | |
3469 | * is anything at all | |
3470 | */ | |
33345d01 LZ |
3471 | if (ret != 0 || min_key.objectid != ino || min_key.type != key_type) { |
3472 | min_key.objectid = ino; | |
e02119d5 CM |
3473 | min_key.type = key_type; |
3474 | min_key.offset = (u64)-1; | |
b3b4aa74 | 3475 | btrfs_release_path(path); |
e02119d5 CM |
3476 | ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0); |
3477 | if (ret < 0) { | |
b3b4aa74 | 3478 | btrfs_release_path(path); |
e02119d5 CM |
3479 | return ret; |
3480 | } | |
33345d01 | 3481 | ret = btrfs_previous_item(root, path, ino, key_type); |
e02119d5 CM |
3482 | |
3483 | /* if ret == 0 there are items for this type, | |
3484 | * create a range to tell us the last key of this type. | |
3485 | * otherwise, there are no items in this directory after | |
3486 | * *min_offset, and we create a range to indicate that. | |
3487 | */ | |
3488 | if (ret == 0) { | |
3489 | struct btrfs_key tmp; | |
3490 | btrfs_item_key_to_cpu(path->nodes[0], &tmp, | |
3491 | path->slots[0]); | |
d397712b | 3492 | if (key_type == tmp.type) |
e02119d5 | 3493 | first_offset = max(min_offset, tmp.offset) + 1; |
e02119d5 CM |
3494 | } |
3495 | goto done; | |
3496 | } | |
3497 | ||
3498 | /* go backward to find any previous key */ | |
33345d01 | 3499 | ret = btrfs_previous_item(root, path, ino, key_type); |
e02119d5 CM |
3500 | if (ret == 0) { |
3501 | struct btrfs_key tmp; | |
3502 | btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]); | |
3503 | if (key_type == tmp.type) { | |
3504 | first_offset = tmp.offset; | |
3505 | ret = overwrite_item(trans, log, dst_path, | |
3506 | path->nodes[0], path->slots[0], | |
3507 | &tmp); | |
4a500fd1 YZ |
3508 | if (ret) { |
3509 | err = ret; | |
3510 | goto done; | |
3511 | } | |
e02119d5 CM |
3512 | } |
3513 | } | |
b3b4aa74 | 3514 | btrfs_release_path(path); |
e02119d5 CM |
3515 | |
3516 | /* find the first key from this transaction again */ | |
3517 | ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0); | |
fae7f21c | 3518 | if (WARN_ON(ret != 0)) |
e02119d5 | 3519 | goto done; |
e02119d5 CM |
3520 | |
3521 | /* | |
3522 | * we have a block from this transaction, log every item in it | |
3523 | * from our directory | |
3524 | */ | |
d397712b | 3525 | while (1) { |
e02119d5 CM |
3526 | struct btrfs_key tmp; |
3527 | src = path->nodes[0]; | |
3528 | nritems = btrfs_header_nritems(src); | |
3529 | for (i = path->slots[0]; i < nritems; i++) { | |
2f2ff0ee FM |
3530 | struct btrfs_dir_item *di; |
3531 | ||
e02119d5 CM |
3532 | btrfs_item_key_to_cpu(src, &min_key, i); |
3533 | ||
33345d01 | 3534 | if (min_key.objectid != ino || min_key.type != key_type) |
e02119d5 CM |
3535 | goto done; |
3536 | ret = overwrite_item(trans, log, dst_path, src, i, | |
3537 | &min_key); | |
4a500fd1 YZ |
3538 | if (ret) { |
3539 | err = ret; | |
3540 | goto done; | |
3541 | } | |
2f2ff0ee FM |
3542 | |
3543 | /* | |
3544 | * We must make sure that when we log a directory entry, | |
3545 | * the corresponding inode, after log replay, has a | |
3546 | * matching link count. For example: | |
3547 | * | |
3548 | * touch foo | |
3549 | * mkdir mydir | |
3550 | * sync | |
3551 | * ln foo mydir/bar | |
3552 | * xfs_io -c "fsync" mydir | |
3553 | * <crash> | |
3554 | * <mount fs and log replay> | |
3555 | * | |
3556 | * Would result in a fsync log that when replayed, our | |
3557 | * file inode would have a link count of 1, but we get | |
3558 | * two directory entries pointing to the same inode. | |
3559 | * After removing one of the names, it would not be | |
3560 | * possible to remove the other name, which resulted | |
3561 | * always in stale file handle errors, and would not | |
3562 | * be possible to rmdir the parent directory, since | |
3563 | * its i_size could never decrement to the value | |
3564 | * BTRFS_EMPTY_DIR_SIZE, resulting in -ENOTEMPTY errors. | |
3565 | */ | |
3566 | di = btrfs_item_ptr(src, i, struct btrfs_dir_item); | |
3567 | btrfs_dir_item_key_to_cpu(src, di, &tmp); | |
3568 | if (ctx && | |
3569 | (btrfs_dir_transid(src, di) == trans->transid || | |
3570 | btrfs_dir_type(src, di) == BTRFS_FT_DIR) && | |
3571 | tmp.type != BTRFS_ROOT_ITEM_KEY) | |
3572 | ctx->log_new_dentries = true; | |
e02119d5 CM |
3573 | } |
3574 | path->slots[0] = nritems; | |
3575 | ||
3576 | /* | |
3577 | * look ahead to the next item and see if it is also | |
3578 | * from this directory and from this transaction | |
3579 | */ | |
3580 | ret = btrfs_next_leaf(root, path); | |
80c0b421 LB |
3581 | if (ret) { |
3582 | if (ret == 1) | |
3583 | last_offset = (u64)-1; | |
3584 | else | |
3585 | err = ret; | |
e02119d5 CM |
3586 | goto done; |
3587 | } | |
3588 | btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]); | |
33345d01 | 3589 | if (tmp.objectid != ino || tmp.type != key_type) { |
e02119d5 CM |
3590 | last_offset = (u64)-1; |
3591 | goto done; | |
3592 | } | |
3593 | if (btrfs_header_generation(path->nodes[0]) != trans->transid) { | |
3594 | ret = overwrite_item(trans, log, dst_path, | |
3595 | path->nodes[0], path->slots[0], | |
3596 | &tmp); | |
4a500fd1 YZ |
3597 | if (ret) |
3598 | err = ret; | |
3599 | else | |
3600 | last_offset = tmp.offset; | |
e02119d5 CM |
3601 | goto done; |
3602 | } | |
3603 | } | |
3604 | done: | |
b3b4aa74 DS |
3605 | btrfs_release_path(path); |
3606 | btrfs_release_path(dst_path); | |
e02119d5 | 3607 | |
4a500fd1 YZ |
3608 | if (err == 0) { |
3609 | *last_offset_ret = last_offset; | |
3610 | /* | |
3611 | * insert the log range keys to indicate where the log | |
3612 | * is valid | |
3613 | */ | |
3614 | ret = insert_dir_log_key(trans, log, path, key_type, | |
33345d01 | 3615 | ino, first_offset, last_offset); |
4a500fd1 YZ |
3616 | if (ret) |
3617 | err = ret; | |
3618 | } | |
3619 | return err; | |
e02119d5 CM |
3620 | } |
3621 | ||
3622 | /* | |
3623 | * logging directories is very similar to logging inodes, We find all the items | |
3624 | * from the current transaction and write them to the log. | |
3625 | * | |
3626 | * The recovery code scans the directory in the subvolume, and if it finds a | |
3627 | * key in the range logged that is not present in the log tree, then it means | |
3628 | * that dir entry was unlinked during the transaction. | |
3629 | * | |
3630 | * In order for that scan to work, we must include one key smaller than | |
3631 | * the smallest logged by this transaction and one key larger than the largest | |
3632 | * key logged by this transaction. | |
3633 | */ | |
3634 | static noinline int log_directory_changes(struct btrfs_trans_handle *trans, | |
dbf39ea4 | 3635 | struct btrfs_root *root, struct btrfs_inode *inode, |
e02119d5 | 3636 | struct btrfs_path *path, |
2f2ff0ee FM |
3637 | struct btrfs_path *dst_path, |
3638 | struct btrfs_log_ctx *ctx) | |
e02119d5 CM |
3639 | { |
3640 | u64 min_key; | |
3641 | u64 max_key; | |
3642 | int ret; | |
3643 | int key_type = BTRFS_DIR_ITEM_KEY; | |
3644 | ||
3645 | again: | |
3646 | min_key = 0; | |
3647 | max_key = 0; | |
d397712b | 3648 | while (1) { |
dbf39ea4 NB |
3649 | ret = log_dir_items(trans, root, inode, path, dst_path, key_type, |
3650 | ctx, min_key, &max_key); | |
4a500fd1 YZ |
3651 | if (ret) |
3652 | return ret; | |
e02119d5 CM |
3653 | if (max_key == (u64)-1) |
3654 | break; | |
3655 | min_key = max_key + 1; | |
3656 | } | |
3657 | ||
3658 | if (key_type == BTRFS_DIR_ITEM_KEY) { | |
3659 | key_type = BTRFS_DIR_INDEX_KEY; | |
3660 | goto again; | |
3661 | } | |
3662 | return 0; | |
3663 | } | |
3664 | ||
3665 | /* | |
3666 | * a helper function to drop items from the log before we relog an | |
3667 | * inode. max_key_type indicates the highest item type to remove. | |
3668 | * This cannot be run for file data extents because it does not | |
3669 | * free the extents they point to. | |
3670 | */ | |
3671 | static int drop_objectid_items(struct btrfs_trans_handle *trans, | |
3672 | struct btrfs_root *log, | |
3673 | struct btrfs_path *path, | |
3674 | u64 objectid, int max_key_type) | |
3675 | { | |
3676 | int ret; | |
3677 | struct btrfs_key key; | |
3678 | struct btrfs_key found_key; | |
18ec90d6 | 3679 | int start_slot; |
e02119d5 CM |
3680 | |
3681 | key.objectid = objectid; | |
3682 | key.type = max_key_type; | |
3683 | key.offset = (u64)-1; | |
3684 | ||
d397712b | 3685 | while (1) { |
e02119d5 | 3686 | ret = btrfs_search_slot(trans, log, &key, path, -1, 1); |
3650860b | 3687 | BUG_ON(ret == 0); /* Logic error */ |
4a500fd1 | 3688 | if (ret < 0) |
e02119d5 CM |
3689 | break; |
3690 | ||
3691 | if (path->slots[0] == 0) | |
3692 | break; | |
3693 | ||
3694 | path->slots[0]--; | |
3695 | btrfs_item_key_to_cpu(path->nodes[0], &found_key, | |
3696 | path->slots[0]); | |
3697 | ||
3698 | if (found_key.objectid != objectid) | |
3699 | break; | |
3700 | ||
18ec90d6 JB |
3701 | found_key.offset = 0; |
3702 | found_key.type = 0; | |
3703 | ret = btrfs_bin_search(path->nodes[0], &found_key, 0, | |
3704 | &start_slot); | |
3705 | ||
3706 | ret = btrfs_del_items(trans, log, path, start_slot, | |
3707 | path->slots[0] - start_slot + 1); | |
3708 | /* | |
3709 | * If start slot isn't 0 then we don't need to re-search, we've | |
3710 | * found the last guy with the objectid in this tree. | |
3711 | */ | |
3712 | if (ret || start_slot != 0) | |
65a246c5 | 3713 | break; |
b3b4aa74 | 3714 | btrfs_release_path(path); |
e02119d5 | 3715 | } |
b3b4aa74 | 3716 | btrfs_release_path(path); |
5bdbeb21 JB |
3717 | if (ret > 0) |
3718 | ret = 0; | |
4a500fd1 | 3719 | return ret; |
e02119d5 CM |
3720 | } |
3721 | ||
94edf4ae JB |
3722 | static void fill_inode_item(struct btrfs_trans_handle *trans, |
3723 | struct extent_buffer *leaf, | |
3724 | struct btrfs_inode_item *item, | |
1a4bcf47 FM |
3725 | struct inode *inode, int log_inode_only, |
3726 | u64 logged_isize) | |
94edf4ae | 3727 | { |
0b1c6cca JB |
3728 | struct btrfs_map_token token; |
3729 | ||
3730 | btrfs_init_map_token(&token); | |
94edf4ae JB |
3731 | |
3732 | if (log_inode_only) { | |
3733 | /* set the generation to zero so the recover code | |
3734 | * can tell the difference between an logging | |
3735 | * just to say 'this inode exists' and a logging | |
3736 | * to say 'update this inode with these values' | |
3737 | */ | |
0b1c6cca | 3738 | btrfs_set_token_inode_generation(leaf, item, 0, &token); |
1a4bcf47 | 3739 | btrfs_set_token_inode_size(leaf, item, logged_isize, &token); |
94edf4ae | 3740 | } else { |
0b1c6cca JB |
3741 | btrfs_set_token_inode_generation(leaf, item, |
3742 | BTRFS_I(inode)->generation, | |
3743 | &token); | |
3744 | btrfs_set_token_inode_size(leaf, item, inode->i_size, &token); | |
3745 | } | |
3746 | ||
3747 | btrfs_set_token_inode_uid(leaf, item, i_uid_read(inode), &token); | |
3748 | btrfs_set_token_inode_gid(leaf, item, i_gid_read(inode), &token); | |
3749 | btrfs_set_token_inode_mode(leaf, item, inode->i_mode, &token); | |
3750 | btrfs_set_token_inode_nlink(leaf, item, inode->i_nlink, &token); | |
3751 | ||
a937b979 | 3752 | btrfs_set_token_timespec_sec(leaf, &item->atime, |
0b1c6cca | 3753 | inode->i_atime.tv_sec, &token); |
a937b979 | 3754 | btrfs_set_token_timespec_nsec(leaf, &item->atime, |
0b1c6cca JB |
3755 | inode->i_atime.tv_nsec, &token); |
3756 | ||
a937b979 | 3757 | btrfs_set_token_timespec_sec(leaf, &item->mtime, |
0b1c6cca | 3758 | inode->i_mtime.tv_sec, &token); |
a937b979 | 3759 | btrfs_set_token_timespec_nsec(leaf, &item->mtime, |
0b1c6cca JB |
3760 | inode->i_mtime.tv_nsec, &token); |
3761 | ||
a937b979 | 3762 | btrfs_set_token_timespec_sec(leaf, &item->ctime, |
0b1c6cca | 3763 | inode->i_ctime.tv_sec, &token); |
a937b979 | 3764 | btrfs_set_token_timespec_nsec(leaf, &item->ctime, |
0b1c6cca JB |
3765 | inode->i_ctime.tv_nsec, &token); |
3766 | ||
3767 | btrfs_set_token_inode_nbytes(leaf, item, inode_get_bytes(inode), | |
3768 | &token); | |
3769 | ||
c7f88c4e JL |
3770 | btrfs_set_token_inode_sequence(leaf, item, |
3771 | inode_peek_iversion(inode), &token); | |
0b1c6cca JB |
3772 | btrfs_set_token_inode_transid(leaf, item, trans->transid, &token); |
3773 | btrfs_set_token_inode_rdev(leaf, item, inode->i_rdev, &token); | |
3774 | btrfs_set_token_inode_flags(leaf, item, BTRFS_I(inode)->flags, &token); | |
3775 | btrfs_set_token_inode_block_group(leaf, item, 0, &token); | |
94edf4ae JB |
3776 | } |
3777 | ||
a95249b3 JB |
3778 | static int log_inode_item(struct btrfs_trans_handle *trans, |
3779 | struct btrfs_root *log, struct btrfs_path *path, | |
6d889a3b | 3780 | struct btrfs_inode *inode) |
a95249b3 JB |
3781 | { |
3782 | struct btrfs_inode_item *inode_item; | |
a95249b3 JB |
3783 | int ret; |
3784 | ||
efd0c405 | 3785 | ret = btrfs_insert_empty_item(trans, log, path, |
6d889a3b | 3786 | &inode->location, sizeof(*inode_item)); |
a95249b3 JB |
3787 | if (ret && ret != -EEXIST) |
3788 | return ret; | |
3789 | inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
3790 | struct btrfs_inode_item); | |
6d889a3b NB |
3791 | fill_inode_item(trans, path->nodes[0], inode_item, &inode->vfs_inode, |
3792 | 0, 0); | |
a95249b3 JB |
3793 | btrfs_release_path(path); |
3794 | return 0; | |
3795 | } | |
3796 | ||
31ff1cd2 | 3797 | static noinline int copy_items(struct btrfs_trans_handle *trans, |
44d70e19 | 3798 | struct btrfs_inode *inode, |
31ff1cd2 | 3799 | struct btrfs_path *dst_path, |
16e7549f | 3800 | struct btrfs_path *src_path, u64 *last_extent, |
1a4bcf47 FM |
3801 | int start_slot, int nr, int inode_only, |
3802 | u64 logged_isize) | |
31ff1cd2 | 3803 | { |
3ffbd68c | 3804 | struct btrfs_fs_info *fs_info = trans->fs_info; |
31ff1cd2 CM |
3805 | unsigned long src_offset; |
3806 | unsigned long dst_offset; | |
44d70e19 | 3807 | struct btrfs_root *log = inode->root->log_root; |
31ff1cd2 CM |
3808 | struct btrfs_file_extent_item *extent; |
3809 | struct btrfs_inode_item *inode_item; | |
16e7549f JB |
3810 | struct extent_buffer *src = src_path->nodes[0]; |
3811 | struct btrfs_key first_key, last_key, key; | |
31ff1cd2 CM |
3812 | int ret; |
3813 | struct btrfs_key *ins_keys; | |
3814 | u32 *ins_sizes; | |
3815 | char *ins_data; | |
3816 | int i; | |
d20f7043 | 3817 | struct list_head ordered_sums; |
44d70e19 | 3818 | int skip_csum = inode->flags & BTRFS_INODE_NODATASUM; |
16e7549f | 3819 | bool has_extents = false; |
74121f7c | 3820 | bool need_find_last_extent = true; |
16e7549f | 3821 | bool done = false; |
d20f7043 CM |
3822 | |
3823 | INIT_LIST_HEAD(&ordered_sums); | |
31ff1cd2 CM |
3824 | |
3825 | ins_data = kmalloc(nr * sizeof(struct btrfs_key) + | |
3826 | nr * sizeof(u32), GFP_NOFS); | |
2a29edc6 | 3827 | if (!ins_data) |
3828 | return -ENOMEM; | |
3829 | ||
16e7549f JB |
3830 | first_key.objectid = (u64)-1; |
3831 | ||
31ff1cd2 CM |
3832 | ins_sizes = (u32 *)ins_data; |
3833 | ins_keys = (struct btrfs_key *)(ins_data + nr * sizeof(u32)); | |
3834 | ||
3835 | for (i = 0; i < nr; i++) { | |
3836 | ins_sizes[i] = btrfs_item_size_nr(src, i + start_slot); | |
3837 | btrfs_item_key_to_cpu(src, ins_keys + i, i + start_slot); | |
3838 | } | |
3839 | ret = btrfs_insert_empty_items(trans, log, dst_path, | |
3840 | ins_keys, ins_sizes, nr); | |
4a500fd1 YZ |
3841 | if (ret) { |
3842 | kfree(ins_data); | |
3843 | return ret; | |
3844 | } | |
31ff1cd2 | 3845 | |
5d4f98a2 | 3846 | for (i = 0; i < nr; i++, dst_path->slots[0]++) { |
31ff1cd2 CM |
3847 | dst_offset = btrfs_item_ptr_offset(dst_path->nodes[0], |
3848 | dst_path->slots[0]); | |
3849 | ||
3850 | src_offset = btrfs_item_ptr_offset(src, start_slot + i); | |
3851 | ||
0dde10be | 3852 | if (i == nr - 1) |
16e7549f JB |
3853 | last_key = ins_keys[i]; |
3854 | ||
94edf4ae | 3855 | if (ins_keys[i].type == BTRFS_INODE_ITEM_KEY) { |
31ff1cd2 CM |
3856 | inode_item = btrfs_item_ptr(dst_path->nodes[0], |
3857 | dst_path->slots[0], | |
3858 | struct btrfs_inode_item); | |
94edf4ae | 3859 | fill_inode_item(trans, dst_path->nodes[0], inode_item, |
f85b7379 DS |
3860 | &inode->vfs_inode, |
3861 | inode_only == LOG_INODE_EXISTS, | |
1a4bcf47 | 3862 | logged_isize); |
94edf4ae JB |
3863 | } else { |
3864 | copy_extent_buffer(dst_path->nodes[0], src, dst_offset, | |
3865 | src_offset, ins_sizes[i]); | |
31ff1cd2 | 3866 | } |
94edf4ae | 3867 | |
16e7549f JB |
3868 | /* |
3869 | * We set need_find_last_extent here in case we know we were | |
3870 | * processing other items and then walk into the first extent in | |
3871 | * the inode. If we don't hit an extent then nothing changes, | |
3872 | * we'll do the last search the next time around. | |
3873 | */ | |
3874 | if (ins_keys[i].type == BTRFS_EXTENT_DATA_KEY) { | |
3875 | has_extents = true; | |
74121f7c | 3876 | if (first_key.objectid == (u64)-1) |
16e7549f JB |
3877 | first_key = ins_keys[i]; |
3878 | } else { | |
3879 | need_find_last_extent = false; | |
3880 | } | |
3881 | ||
31ff1cd2 CM |
3882 | /* take a reference on file data extents so that truncates |
3883 | * or deletes of this inode don't have to relog the inode | |
3884 | * again | |
3885 | */ | |
962a298f | 3886 | if (ins_keys[i].type == BTRFS_EXTENT_DATA_KEY && |
d2794405 | 3887 | !skip_csum) { |
31ff1cd2 CM |
3888 | int found_type; |
3889 | extent = btrfs_item_ptr(src, start_slot + i, | |
3890 | struct btrfs_file_extent_item); | |
3891 | ||
8e531cdf | 3892 | if (btrfs_file_extent_generation(src, extent) < trans->transid) |
3893 | continue; | |
3894 | ||
31ff1cd2 | 3895 | found_type = btrfs_file_extent_type(src, extent); |
6f1fed77 | 3896 | if (found_type == BTRFS_FILE_EXTENT_REG) { |
5d4f98a2 YZ |
3897 | u64 ds, dl, cs, cl; |
3898 | ds = btrfs_file_extent_disk_bytenr(src, | |
3899 | extent); | |
3900 | /* ds == 0 is a hole */ | |
3901 | if (ds == 0) | |
3902 | continue; | |
3903 | ||
3904 | dl = btrfs_file_extent_disk_num_bytes(src, | |
3905 | extent); | |
3906 | cs = btrfs_file_extent_offset(src, extent); | |
3907 | cl = btrfs_file_extent_num_bytes(src, | |
a419aef8 | 3908 | extent); |
580afd76 CM |
3909 | if (btrfs_file_extent_compression(src, |
3910 | extent)) { | |
3911 | cs = 0; | |
3912 | cl = dl; | |
3913 | } | |
5d4f98a2 YZ |
3914 | |
3915 | ret = btrfs_lookup_csums_range( | |
0b246afa | 3916 | fs_info->csum_root, |
5d4f98a2 | 3917 | ds + cs, ds + cs + cl - 1, |
a2de733c | 3918 | &ordered_sums, 0); |
3650860b JB |
3919 | if (ret) { |
3920 | btrfs_release_path(dst_path); | |
3921 | kfree(ins_data); | |
3922 | return ret; | |
3923 | } | |
31ff1cd2 CM |
3924 | } |
3925 | } | |
31ff1cd2 CM |
3926 | } |
3927 | ||
3928 | btrfs_mark_buffer_dirty(dst_path->nodes[0]); | |
b3b4aa74 | 3929 | btrfs_release_path(dst_path); |
31ff1cd2 | 3930 | kfree(ins_data); |
d20f7043 CM |
3931 | |
3932 | /* | |
3933 | * we have to do this after the loop above to avoid changing the | |
3934 | * log tree while trying to change the log tree. | |
3935 | */ | |
4a500fd1 | 3936 | ret = 0; |
d397712b | 3937 | while (!list_empty(&ordered_sums)) { |
d20f7043 CM |
3938 | struct btrfs_ordered_sum *sums = list_entry(ordered_sums.next, |
3939 | struct btrfs_ordered_sum, | |
3940 | list); | |
4a500fd1 YZ |
3941 | if (!ret) |
3942 | ret = btrfs_csum_file_blocks(trans, log, sums); | |
d20f7043 CM |
3943 | list_del(&sums->list); |
3944 | kfree(sums); | |
3945 | } | |
16e7549f JB |
3946 | |
3947 | if (!has_extents) | |
3948 | return ret; | |
3949 | ||
74121f7c FM |
3950 | if (need_find_last_extent && *last_extent == first_key.offset) { |
3951 | /* | |
3952 | * We don't have any leafs between our current one and the one | |
3953 | * we processed before that can have file extent items for our | |
3954 | * inode (and have a generation number smaller than our current | |
3955 | * transaction id). | |
3956 | */ | |
3957 | need_find_last_extent = false; | |
3958 | } | |
3959 | ||
16e7549f JB |
3960 | /* |
3961 | * Because we use btrfs_search_forward we could skip leaves that were | |
3962 | * not modified and then assume *last_extent is valid when it really | |
3963 | * isn't. So back up to the previous leaf and read the end of the last | |
3964 | * extent before we go and fill in holes. | |
3965 | */ | |
3966 | if (need_find_last_extent) { | |
3967 | u64 len; | |
3968 | ||
44d70e19 | 3969 | ret = btrfs_prev_leaf(inode->root, src_path); |
16e7549f JB |
3970 | if (ret < 0) |
3971 | return ret; | |
3972 | if (ret) | |
3973 | goto fill_holes; | |
3974 | if (src_path->slots[0]) | |
3975 | src_path->slots[0]--; | |
3976 | src = src_path->nodes[0]; | |
3977 | btrfs_item_key_to_cpu(src, &key, src_path->slots[0]); | |
44d70e19 | 3978 | if (key.objectid != btrfs_ino(inode) || |
16e7549f JB |
3979 | key.type != BTRFS_EXTENT_DATA_KEY) |
3980 | goto fill_holes; | |
3981 | extent = btrfs_item_ptr(src, src_path->slots[0], | |
3982 | struct btrfs_file_extent_item); | |
3983 | if (btrfs_file_extent_type(src, extent) == | |
3984 | BTRFS_FILE_EXTENT_INLINE) { | |
e41ca589 | 3985 | len = btrfs_file_extent_ram_bytes(src, extent); |
16e7549f | 3986 | *last_extent = ALIGN(key.offset + len, |
0b246afa | 3987 | fs_info->sectorsize); |
16e7549f JB |
3988 | } else { |
3989 | len = btrfs_file_extent_num_bytes(src, extent); | |
3990 | *last_extent = key.offset + len; | |
3991 | } | |
3992 | } | |
3993 | fill_holes: | |
3994 | /* So we did prev_leaf, now we need to move to the next leaf, but a few | |
3995 | * things could have happened | |
3996 | * | |
3997 | * 1) A merge could have happened, so we could currently be on a leaf | |
3998 | * that holds what we were copying in the first place. | |
3999 | * 2) A split could have happened, and now not all of the items we want | |
4000 | * are on the same leaf. | |
4001 | * | |
4002 | * So we need to adjust how we search for holes, we need to drop the | |
4003 | * path and re-search for the first extent key we found, and then walk | |
4004 | * forward until we hit the last one we copied. | |
4005 | */ | |
4006 | if (need_find_last_extent) { | |
4007 | /* btrfs_prev_leaf could return 1 without releasing the path */ | |
4008 | btrfs_release_path(src_path); | |
f85b7379 DS |
4009 | ret = btrfs_search_slot(NULL, inode->root, &first_key, |
4010 | src_path, 0, 0); | |
16e7549f JB |
4011 | if (ret < 0) |
4012 | return ret; | |
4013 | ASSERT(ret == 0); | |
4014 | src = src_path->nodes[0]; | |
4015 | i = src_path->slots[0]; | |
4016 | } else { | |
4017 | i = start_slot; | |
4018 | } | |
4019 | ||
4020 | /* | |
4021 | * Ok so here we need to go through and fill in any holes we may have | |
4022 | * to make sure that holes are punched for those areas in case they had | |
4023 | * extents previously. | |
4024 | */ | |
4025 | while (!done) { | |
4026 | u64 offset, len; | |
4027 | u64 extent_end; | |
4028 | ||
4029 | if (i >= btrfs_header_nritems(src_path->nodes[0])) { | |
44d70e19 | 4030 | ret = btrfs_next_leaf(inode->root, src_path); |
16e7549f JB |
4031 | if (ret < 0) |
4032 | return ret; | |
4033 | ASSERT(ret == 0); | |
4034 | src = src_path->nodes[0]; | |
4035 | i = 0; | |
8434ec46 | 4036 | need_find_last_extent = true; |
16e7549f JB |
4037 | } |
4038 | ||
4039 | btrfs_item_key_to_cpu(src, &key, i); | |
4040 | if (!btrfs_comp_cpu_keys(&key, &last_key)) | |
4041 | done = true; | |
44d70e19 | 4042 | if (key.objectid != btrfs_ino(inode) || |
16e7549f JB |
4043 | key.type != BTRFS_EXTENT_DATA_KEY) { |
4044 | i++; | |
4045 | continue; | |
4046 | } | |
4047 | extent = btrfs_item_ptr(src, i, struct btrfs_file_extent_item); | |
4048 | if (btrfs_file_extent_type(src, extent) == | |
4049 | BTRFS_FILE_EXTENT_INLINE) { | |
e41ca589 | 4050 | len = btrfs_file_extent_ram_bytes(src, extent); |
da17066c | 4051 | extent_end = ALIGN(key.offset + len, |
0b246afa | 4052 | fs_info->sectorsize); |
16e7549f JB |
4053 | } else { |
4054 | len = btrfs_file_extent_num_bytes(src, extent); | |
4055 | extent_end = key.offset + len; | |
4056 | } | |
4057 | i++; | |
4058 | ||
4059 | if (*last_extent == key.offset) { | |
4060 | *last_extent = extent_end; | |
4061 | continue; | |
4062 | } | |
4063 | offset = *last_extent; | |
4064 | len = key.offset - *last_extent; | |
44d70e19 | 4065 | ret = btrfs_insert_file_extent(trans, log, btrfs_ino(inode), |
f85b7379 | 4066 | offset, 0, 0, len, 0, len, 0, 0, 0); |
16e7549f JB |
4067 | if (ret) |
4068 | break; | |
74121f7c | 4069 | *last_extent = extent_end; |
16e7549f | 4070 | } |
4ee3fad3 FM |
4071 | |
4072 | /* | |
4073 | * Check if there is a hole between the last extent found in our leaf | |
4074 | * and the first extent in the next leaf. If there is one, we need to | |
4075 | * log an explicit hole so that at replay time we can punch the hole. | |
4076 | */ | |
4077 | if (ret == 0 && | |
4078 | key.objectid == btrfs_ino(inode) && | |
4079 | key.type == BTRFS_EXTENT_DATA_KEY && | |
4080 | i == btrfs_header_nritems(src_path->nodes[0])) { | |
4081 | ret = btrfs_next_leaf(inode->root, src_path); | |
4082 | need_find_last_extent = true; | |
4083 | if (ret > 0) { | |
4084 | ret = 0; | |
4085 | } else if (ret == 0) { | |
4086 | btrfs_item_key_to_cpu(src_path->nodes[0], &key, | |
4087 | src_path->slots[0]); | |
4088 | if (key.objectid == btrfs_ino(inode) && | |
4089 | key.type == BTRFS_EXTENT_DATA_KEY && | |
4090 | *last_extent < key.offset) { | |
4091 | const u64 len = key.offset - *last_extent; | |
4092 | ||
4093 | ret = btrfs_insert_file_extent(trans, log, | |
4094 | btrfs_ino(inode), | |
4095 | *last_extent, 0, | |
4096 | 0, len, 0, len, | |
4097 | 0, 0, 0); | |
4098 | } | |
4099 | } | |
4100 | } | |
16e7549f JB |
4101 | /* |
4102 | * Need to let the callers know we dropped the path so they should | |
4103 | * re-search. | |
4104 | */ | |
4105 | if (!ret && need_find_last_extent) | |
4106 | ret = 1; | |
4a500fd1 | 4107 | return ret; |
31ff1cd2 CM |
4108 | } |
4109 | ||
5dc562c5 JB |
4110 | static int extent_cmp(void *priv, struct list_head *a, struct list_head *b) |
4111 | { | |
4112 | struct extent_map *em1, *em2; | |
4113 | ||
4114 | em1 = list_entry(a, struct extent_map, list); | |
4115 | em2 = list_entry(b, struct extent_map, list); | |
4116 | ||
4117 | if (em1->start < em2->start) | |
4118 | return -1; | |
4119 | else if (em1->start > em2->start) | |
4120 | return 1; | |
4121 | return 0; | |
4122 | } | |
4123 | ||
e7175a69 JB |
4124 | static int log_extent_csums(struct btrfs_trans_handle *trans, |
4125 | struct btrfs_inode *inode, | |
a9ecb653 | 4126 | struct btrfs_root *log_root, |
e7175a69 | 4127 | const struct extent_map *em) |
5dc562c5 | 4128 | { |
2ab28f32 JB |
4129 | u64 csum_offset; |
4130 | u64 csum_len; | |
8407f553 FM |
4131 | LIST_HEAD(ordered_sums); |
4132 | int ret = 0; | |
0aa4a17d | 4133 | |
e7175a69 JB |
4134 | if (inode->flags & BTRFS_INODE_NODATASUM || |
4135 | test_bit(EXTENT_FLAG_PREALLOC, &em->flags) || | |
8407f553 | 4136 | em->block_start == EXTENT_MAP_HOLE) |
70c8a91c | 4137 | return 0; |
5dc562c5 | 4138 | |
e7175a69 | 4139 | /* If we're compressed we have to save the entire range of csums. */ |
488111aa FDBM |
4140 | if (em->compress_type) { |
4141 | csum_offset = 0; | |
8407f553 | 4142 | csum_len = max(em->block_len, em->orig_block_len); |
488111aa | 4143 | } else { |
e7175a69 JB |
4144 | csum_offset = em->mod_start - em->start; |
4145 | csum_len = em->mod_len; | |
488111aa | 4146 | } |
2ab28f32 | 4147 | |
70c8a91c | 4148 | /* block start is already adjusted for the file extent offset. */ |
a9ecb653 | 4149 | ret = btrfs_lookup_csums_range(trans->fs_info->csum_root, |
70c8a91c JB |
4150 | em->block_start + csum_offset, |
4151 | em->block_start + csum_offset + | |
4152 | csum_len - 1, &ordered_sums, 0); | |
4153 | if (ret) | |
4154 | return ret; | |
5dc562c5 | 4155 | |
70c8a91c JB |
4156 | while (!list_empty(&ordered_sums)) { |
4157 | struct btrfs_ordered_sum *sums = list_entry(ordered_sums.next, | |
4158 | struct btrfs_ordered_sum, | |
4159 | list); | |
4160 | if (!ret) | |
a9ecb653 | 4161 | ret = btrfs_csum_file_blocks(trans, log_root, sums); |
70c8a91c JB |
4162 | list_del(&sums->list); |
4163 | kfree(sums); | |
5dc562c5 JB |
4164 | } |
4165 | ||
70c8a91c | 4166 | return ret; |
5dc562c5 JB |
4167 | } |
4168 | ||
8407f553 | 4169 | static int log_one_extent(struct btrfs_trans_handle *trans, |
9d122629 | 4170 | struct btrfs_inode *inode, struct btrfs_root *root, |
8407f553 FM |
4171 | const struct extent_map *em, |
4172 | struct btrfs_path *path, | |
8407f553 FM |
4173 | struct btrfs_log_ctx *ctx) |
4174 | { | |
4175 | struct btrfs_root *log = root->log_root; | |
4176 | struct btrfs_file_extent_item *fi; | |
4177 | struct extent_buffer *leaf; | |
4178 | struct btrfs_map_token token; | |
4179 | struct btrfs_key key; | |
4180 | u64 extent_offset = em->start - em->orig_start; | |
4181 | u64 block_len; | |
4182 | int ret; | |
4183 | int extent_inserted = 0; | |
8407f553 | 4184 | |
a9ecb653 | 4185 | ret = log_extent_csums(trans, inode, log, em); |
8407f553 FM |
4186 | if (ret) |
4187 | return ret; | |
4188 | ||
8407f553 FM |
4189 | btrfs_init_map_token(&token); |
4190 | ||
9d122629 | 4191 | ret = __btrfs_drop_extents(trans, log, &inode->vfs_inode, path, em->start, |
8407f553 FM |
4192 | em->start + em->len, NULL, 0, 1, |
4193 | sizeof(*fi), &extent_inserted); | |
4194 | if (ret) | |
4195 | return ret; | |
4196 | ||
4197 | if (!extent_inserted) { | |
9d122629 | 4198 | key.objectid = btrfs_ino(inode); |
8407f553 FM |
4199 | key.type = BTRFS_EXTENT_DATA_KEY; |
4200 | key.offset = em->start; | |
4201 | ||
4202 | ret = btrfs_insert_empty_item(trans, log, path, &key, | |
4203 | sizeof(*fi)); | |
4204 | if (ret) | |
4205 | return ret; | |
4206 | } | |
4207 | leaf = path->nodes[0]; | |
4208 | fi = btrfs_item_ptr(leaf, path->slots[0], | |
4209 | struct btrfs_file_extent_item); | |
4210 | ||
50d9aa99 | 4211 | btrfs_set_token_file_extent_generation(leaf, fi, trans->transid, |
8407f553 FM |
4212 | &token); |
4213 | if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) | |
4214 | btrfs_set_token_file_extent_type(leaf, fi, | |
4215 | BTRFS_FILE_EXTENT_PREALLOC, | |
4216 | &token); | |
4217 | else | |
4218 | btrfs_set_token_file_extent_type(leaf, fi, | |
4219 | BTRFS_FILE_EXTENT_REG, | |
4220 | &token); | |
4221 | ||
4222 | block_len = max(em->block_len, em->orig_block_len); | |
4223 | if (em->compress_type != BTRFS_COMPRESS_NONE) { | |
4224 | btrfs_set_token_file_extent_disk_bytenr(leaf, fi, | |
4225 | em->block_start, | |
4226 | &token); | |
4227 | btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, block_len, | |
4228 | &token); | |
4229 | } else if (em->block_start < EXTENT_MAP_LAST_BYTE) { | |
4230 | btrfs_set_token_file_extent_disk_bytenr(leaf, fi, | |
4231 | em->block_start - | |
4232 | extent_offset, &token); | |
4233 | btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, block_len, | |
4234 | &token); | |
4235 | } else { | |
4236 | btrfs_set_token_file_extent_disk_bytenr(leaf, fi, 0, &token); | |
4237 | btrfs_set_token_file_extent_disk_num_bytes(leaf, fi, 0, | |
4238 | &token); | |
4239 | } | |
4240 | ||
4241 | btrfs_set_token_file_extent_offset(leaf, fi, extent_offset, &token); | |
4242 | btrfs_set_token_file_extent_num_bytes(leaf, fi, em->len, &token); | |
4243 | btrfs_set_token_file_extent_ram_bytes(leaf, fi, em->ram_bytes, &token); | |
4244 | btrfs_set_token_file_extent_compression(leaf, fi, em->compress_type, | |
4245 | &token); | |
4246 | btrfs_set_token_file_extent_encryption(leaf, fi, 0, &token); | |
4247 | btrfs_set_token_file_extent_other_encoding(leaf, fi, 0, &token); | |
4248 | btrfs_mark_buffer_dirty(leaf); | |
4249 | ||
4250 | btrfs_release_path(path); | |
4251 | ||
4252 | return ret; | |
4253 | } | |
4254 | ||
31d11b83 FM |
4255 | /* |
4256 | * Log all prealloc extents beyond the inode's i_size to make sure we do not | |
4257 | * lose them after doing a fast fsync and replaying the log. We scan the | |
4258 | * subvolume's root instead of iterating the inode's extent map tree because | |
4259 | * otherwise we can log incorrect extent items based on extent map conversion. | |
4260 | * That can happen due to the fact that extent maps are merged when they | |
4261 | * are not in the extent map tree's list of modified extents. | |
4262 | */ | |
4263 | static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans, | |
4264 | struct btrfs_inode *inode, | |
4265 | struct btrfs_path *path) | |
4266 | { | |
4267 | struct btrfs_root *root = inode->root; | |
4268 | struct btrfs_key key; | |
4269 | const u64 i_size = i_size_read(&inode->vfs_inode); | |
4270 | const u64 ino = btrfs_ino(inode); | |
4271 | struct btrfs_path *dst_path = NULL; | |
4272 | u64 last_extent = (u64)-1; | |
4273 | int ins_nr = 0; | |
4274 | int start_slot; | |
4275 | int ret; | |
4276 | ||
4277 | if (!(inode->flags & BTRFS_INODE_PREALLOC)) | |
4278 | return 0; | |
4279 | ||
4280 | key.objectid = ino; | |
4281 | key.type = BTRFS_EXTENT_DATA_KEY; | |
4282 | key.offset = i_size; | |
4283 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
4284 | if (ret < 0) | |
4285 | goto out; | |
4286 | ||
4287 | while (true) { | |
4288 | struct extent_buffer *leaf = path->nodes[0]; | |
4289 | int slot = path->slots[0]; | |
4290 | ||
4291 | if (slot >= btrfs_header_nritems(leaf)) { | |
4292 | if (ins_nr > 0) { | |
4293 | ret = copy_items(trans, inode, dst_path, path, | |
4294 | &last_extent, start_slot, | |
4295 | ins_nr, 1, 0); | |
4296 | if (ret < 0) | |
4297 | goto out; | |
4298 | ins_nr = 0; | |
4299 | } | |
4300 | ret = btrfs_next_leaf(root, path); | |
4301 | if (ret < 0) | |
4302 | goto out; | |
4303 | if (ret > 0) { | |
4304 | ret = 0; | |
4305 | break; | |
4306 | } | |
4307 | continue; | |
4308 | } | |
4309 | ||
4310 | btrfs_item_key_to_cpu(leaf, &key, slot); | |
4311 | if (key.objectid > ino) | |
4312 | break; | |
4313 | if (WARN_ON_ONCE(key.objectid < ino) || | |
4314 | key.type < BTRFS_EXTENT_DATA_KEY || | |
4315 | key.offset < i_size) { | |
4316 | path->slots[0]++; | |
4317 | continue; | |
4318 | } | |
4319 | if (last_extent == (u64)-1) { | |
4320 | last_extent = key.offset; | |
4321 | /* | |
4322 | * Avoid logging extent items logged in past fsync calls | |
4323 | * and leading to duplicate keys in the log tree. | |
4324 | */ | |
4325 | do { | |
4326 | ret = btrfs_truncate_inode_items(trans, | |
4327 | root->log_root, | |
4328 | &inode->vfs_inode, | |
4329 | i_size, | |
4330 | BTRFS_EXTENT_DATA_KEY); | |
4331 | } while (ret == -EAGAIN); | |
4332 | if (ret) | |
4333 | goto out; | |
4334 | } | |
4335 | if (ins_nr == 0) | |
4336 | start_slot = slot; | |
4337 | ins_nr++; | |
4338 | path->slots[0]++; | |
4339 | if (!dst_path) { | |
4340 | dst_path = btrfs_alloc_path(); | |
4341 | if (!dst_path) { | |
4342 | ret = -ENOMEM; | |
4343 | goto out; | |
4344 | } | |
4345 | } | |
4346 | } | |
4347 | if (ins_nr > 0) { | |
4348 | ret = copy_items(trans, inode, dst_path, path, &last_extent, | |
4349 | start_slot, ins_nr, 1, 0); | |
4350 | if (ret > 0) | |
4351 | ret = 0; | |
4352 | } | |
4353 | out: | |
4354 | btrfs_release_path(path); | |
4355 | btrfs_free_path(dst_path); | |
4356 | return ret; | |
4357 | } | |
4358 | ||
5dc562c5 JB |
4359 | static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans, |
4360 | struct btrfs_root *root, | |
9d122629 | 4361 | struct btrfs_inode *inode, |
827463c4 | 4362 | struct btrfs_path *path, |
de0ee0ed FM |
4363 | struct btrfs_log_ctx *ctx, |
4364 | const u64 start, | |
4365 | const u64 end) | |
5dc562c5 | 4366 | { |
5dc562c5 JB |
4367 | struct extent_map *em, *n; |
4368 | struct list_head extents; | |
9d122629 | 4369 | struct extent_map_tree *tree = &inode->extent_tree; |
8c6c5928 | 4370 | u64 logged_start, logged_end; |
5dc562c5 JB |
4371 | u64 test_gen; |
4372 | int ret = 0; | |
2ab28f32 | 4373 | int num = 0; |
5dc562c5 JB |
4374 | |
4375 | INIT_LIST_HEAD(&extents); | |
4376 | ||
9d122629 | 4377 | down_write(&inode->dio_sem); |
5dc562c5 JB |
4378 | write_lock(&tree->lock); |
4379 | test_gen = root->fs_info->last_trans_committed; | |
8c6c5928 JB |
4380 | logged_start = start; |
4381 | logged_end = end; | |
5dc562c5 JB |
4382 | |
4383 | list_for_each_entry_safe(em, n, &tree->modified_extents, list) { | |
4384 | list_del_init(&em->list); | |
2ab28f32 JB |
4385 | /* |
4386 | * Just an arbitrary number, this can be really CPU intensive | |
4387 | * once we start getting a lot of extents, and really once we | |
4388 | * have a bunch of extents we just want to commit since it will | |
4389 | * be faster. | |
4390 | */ | |
4391 | if (++num > 32768) { | |
4392 | list_del_init(&tree->modified_extents); | |
4393 | ret = -EFBIG; | |
4394 | goto process; | |
4395 | } | |
4396 | ||
5dc562c5 JB |
4397 | if (em->generation <= test_gen) |
4398 | continue; | |
8c6c5928 | 4399 | |
31d11b83 FM |
4400 | /* We log prealloc extents beyond eof later. */ |
4401 | if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) && | |
4402 | em->start >= i_size_read(&inode->vfs_inode)) | |
4403 | continue; | |
4404 | ||
8c6c5928 JB |
4405 | if (em->start < logged_start) |
4406 | logged_start = em->start; | |
4407 | if ((em->start + em->len - 1) > logged_end) | |
4408 | logged_end = em->start + em->len - 1; | |
4409 | ||
ff44c6e3 | 4410 | /* Need a ref to keep it from getting evicted from cache */ |
490b54d6 | 4411 | refcount_inc(&em->refs); |
ff44c6e3 | 4412 | set_bit(EXTENT_FLAG_LOGGING, &em->flags); |
5dc562c5 | 4413 | list_add_tail(&em->list, &extents); |
2ab28f32 | 4414 | num++; |
5dc562c5 JB |
4415 | } |
4416 | ||
4417 | list_sort(NULL, &extents, extent_cmp); | |
2ab28f32 | 4418 | process: |
5dc562c5 JB |
4419 | while (!list_empty(&extents)) { |
4420 | em = list_entry(extents.next, struct extent_map, list); | |
4421 | ||
4422 | list_del_init(&em->list); | |
4423 | ||
4424 | /* | |
4425 | * If we had an error we just need to delete everybody from our | |
4426 | * private list. | |
4427 | */ | |
ff44c6e3 | 4428 | if (ret) { |
201a9038 | 4429 | clear_em_logging(tree, em); |
ff44c6e3 | 4430 | free_extent_map(em); |
5dc562c5 | 4431 | continue; |
ff44c6e3 JB |
4432 | } |
4433 | ||
4434 | write_unlock(&tree->lock); | |
5dc562c5 | 4435 | |
a2120a47 | 4436 | ret = log_one_extent(trans, inode, root, em, path, ctx); |
ff44c6e3 | 4437 | write_lock(&tree->lock); |
201a9038 JB |
4438 | clear_em_logging(tree, em); |
4439 | free_extent_map(em); | |
5dc562c5 | 4440 | } |
ff44c6e3 JB |
4441 | WARN_ON(!list_empty(&extents)); |
4442 | write_unlock(&tree->lock); | |
9d122629 | 4443 | up_write(&inode->dio_sem); |
5dc562c5 | 4444 | |
5dc562c5 | 4445 | btrfs_release_path(path); |
31d11b83 FM |
4446 | if (!ret) |
4447 | ret = btrfs_log_prealloc_extents(trans, inode, path); | |
4448 | ||
5dc562c5 JB |
4449 | return ret; |
4450 | } | |
4451 | ||
481b01c0 | 4452 | static int logged_inode_size(struct btrfs_root *log, struct btrfs_inode *inode, |
1a4bcf47 FM |
4453 | struct btrfs_path *path, u64 *size_ret) |
4454 | { | |
4455 | struct btrfs_key key; | |
4456 | int ret; | |
4457 | ||
481b01c0 | 4458 | key.objectid = btrfs_ino(inode); |
1a4bcf47 FM |
4459 | key.type = BTRFS_INODE_ITEM_KEY; |
4460 | key.offset = 0; | |
4461 | ||
4462 | ret = btrfs_search_slot(NULL, log, &key, path, 0, 0); | |
4463 | if (ret < 0) { | |
4464 | return ret; | |
4465 | } else if (ret > 0) { | |
2f2ff0ee | 4466 | *size_ret = 0; |
1a4bcf47 FM |
4467 | } else { |
4468 | struct btrfs_inode_item *item; | |
4469 | ||
4470 | item = btrfs_item_ptr(path->nodes[0], path->slots[0], | |
4471 | struct btrfs_inode_item); | |
4472 | *size_ret = btrfs_inode_size(path->nodes[0], item); | |
4473 | } | |
4474 | ||
4475 | btrfs_release_path(path); | |
4476 | return 0; | |
4477 | } | |
4478 | ||
36283bf7 FM |
4479 | /* |
4480 | * At the moment we always log all xattrs. This is to figure out at log replay | |
4481 | * time which xattrs must have their deletion replayed. If a xattr is missing | |
4482 | * in the log tree and exists in the fs/subvol tree, we delete it. This is | |
4483 | * because if a xattr is deleted, the inode is fsynced and a power failure | |
4484 | * happens, causing the log to be replayed the next time the fs is mounted, | |
4485 | * we want the xattr to not exist anymore (same behaviour as other filesystems | |
4486 | * with a journal, ext3/4, xfs, f2fs, etc). | |
4487 | */ | |
4488 | static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans, | |
4489 | struct btrfs_root *root, | |
1a93c36a | 4490 | struct btrfs_inode *inode, |
36283bf7 FM |
4491 | struct btrfs_path *path, |
4492 | struct btrfs_path *dst_path) | |
4493 | { | |
4494 | int ret; | |
4495 | struct btrfs_key key; | |
1a93c36a | 4496 | const u64 ino = btrfs_ino(inode); |
36283bf7 FM |
4497 | int ins_nr = 0; |
4498 | int start_slot = 0; | |
4499 | ||
4500 | key.objectid = ino; | |
4501 | key.type = BTRFS_XATTR_ITEM_KEY; | |
4502 | key.offset = 0; | |
4503 | ||
4504 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
4505 | if (ret < 0) | |
4506 | return ret; | |
4507 | ||
4508 | while (true) { | |
4509 | int slot = path->slots[0]; | |
4510 | struct extent_buffer *leaf = path->nodes[0]; | |
4511 | int nritems = btrfs_header_nritems(leaf); | |
4512 | ||
4513 | if (slot >= nritems) { | |
4514 | if (ins_nr > 0) { | |
4515 | u64 last_extent = 0; | |
4516 | ||
1a93c36a | 4517 | ret = copy_items(trans, inode, dst_path, path, |
36283bf7 FM |
4518 | &last_extent, start_slot, |
4519 | ins_nr, 1, 0); | |
4520 | /* can't be 1, extent items aren't processed */ | |
4521 | ASSERT(ret <= 0); | |
4522 | if (ret < 0) | |
4523 | return ret; | |
4524 | ins_nr = 0; | |
4525 | } | |
4526 | ret = btrfs_next_leaf(root, path); | |
4527 | if (ret < 0) | |
4528 | return ret; | |
4529 | else if (ret > 0) | |
4530 | break; | |
4531 | continue; | |
4532 | } | |
4533 | ||
4534 | btrfs_item_key_to_cpu(leaf, &key, slot); | |
4535 | if (key.objectid != ino || key.type != BTRFS_XATTR_ITEM_KEY) | |
4536 | break; | |
4537 | ||
4538 | if (ins_nr == 0) | |
4539 | start_slot = slot; | |
4540 | ins_nr++; | |
4541 | path->slots[0]++; | |
4542 | cond_resched(); | |
4543 | } | |
4544 | if (ins_nr > 0) { | |
4545 | u64 last_extent = 0; | |
4546 | ||
1a93c36a | 4547 | ret = copy_items(trans, inode, dst_path, path, |
36283bf7 FM |
4548 | &last_extent, start_slot, |
4549 | ins_nr, 1, 0); | |
4550 | /* can't be 1, extent items aren't processed */ | |
4551 | ASSERT(ret <= 0); | |
4552 | if (ret < 0) | |
4553 | return ret; | |
4554 | } | |
4555 | ||
4556 | return 0; | |
4557 | } | |
4558 | ||
a89ca6f2 FM |
4559 | /* |
4560 | * If the no holes feature is enabled we need to make sure any hole between the | |
4561 | * last extent and the i_size of our inode is explicitly marked in the log. This | |
4562 | * is to make sure that doing something like: | |
4563 | * | |
4564 | * 1) create file with 128Kb of data | |
4565 | * 2) truncate file to 64Kb | |
4566 | * 3) truncate file to 256Kb | |
4567 | * 4) fsync file | |
4568 | * 5) <crash/power failure> | |
4569 | * 6) mount fs and trigger log replay | |
4570 | * | |
4571 | * Will give us a file with a size of 256Kb, the first 64Kb of data match what | |
4572 | * the file had in its first 64Kb of data at step 1 and the last 192Kb of the | |
4573 | * file correspond to a hole. The presence of explicit holes in a log tree is | |
4574 | * what guarantees that log replay will remove/adjust file extent items in the | |
4575 | * fs/subvol tree. | |
4576 | * | |
4577 | * Here we do not need to care about holes between extents, that is already done | |
4578 | * by copy_items(). We also only need to do this in the full sync path, where we | |
4579 | * lookup for extents from the fs/subvol tree only. In the fast path case, we | |
4580 | * lookup the list of modified extent maps and if any represents a hole, we | |
4581 | * insert a corresponding extent representing a hole in the log tree. | |
4582 | */ | |
4583 | static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans, | |
4584 | struct btrfs_root *root, | |
a0308dd7 | 4585 | struct btrfs_inode *inode, |
a89ca6f2 FM |
4586 | struct btrfs_path *path) |
4587 | { | |
0b246afa | 4588 | struct btrfs_fs_info *fs_info = root->fs_info; |
a89ca6f2 FM |
4589 | int ret; |
4590 | struct btrfs_key key; | |
4591 | u64 hole_start; | |
4592 | u64 hole_size; | |
4593 | struct extent_buffer *leaf; | |
4594 | struct btrfs_root *log = root->log_root; | |
a0308dd7 NB |
4595 | const u64 ino = btrfs_ino(inode); |
4596 | const u64 i_size = i_size_read(&inode->vfs_inode); | |
a89ca6f2 | 4597 | |
0b246afa | 4598 | if (!btrfs_fs_incompat(fs_info, NO_HOLES)) |
a89ca6f2 FM |
4599 | return 0; |
4600 | ||
4601 | key.objectid = ino; | |
4602 | key.type = BTRFS_EXTENT_DATA_KEY; | |
4603 | key.offset = (u64)-1; | |
4604 | ||
4605 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
4606 | ASSERT(ret != 0); | |
4607 | if (ret < 0) | |
4608 | return ret; | |
4609 | ||
4610 | ASSERT(path->slots[0] > 0); | |
4611 | path->slots[0]--; | |
4612 | leaf = path->nodes[0]; | |
4613 | btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); | |
4614 | ||
4615 | if (key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY) { | |
4616 | /* inode does not have any extents */ | |
4617 | hole_start = 0; | |
4618 | hole_size = i_size; | |
4619 | } else { | |
4620 | struct btrfs_file_extent_item *extent; | |
4621 | u64 len; | |
4622 | ||
4623 | /* | |
4624 | * If there's an extent beyond i_size, an explicit hole was | |
4625 | * already inserted by copy_items(). | |
4626 | */ | |
4627 | if (key.offset >= i_size) | |
4628 | return 0; | |
4629 | ||
4630 | extent = btrfs_item_ptr(leaf, path->slots[0], | |
4631 | struct btrfs_file_extent_item); | |
4632 | ||
4633 | if (btrfs_file_extent_type(leaf, extent) == | |
4634 | BTRFS_FILE_EXTENT_INLINE) { | |
e41ca589 | 4635 | len = btrfs_file_extent_ram_bytes(leaf, extent); |
6399fb5a FM |
4636 | ASSERT(len == i_size || |
4637 | (len == fs_info->sectorsize && | |
4638 | btrfs_file_extent_compression(leaf, extent) != | |
4639 | BTRFS_COMPRESS_NONE)); | |
a89ca6f2 FM |
4640 | return 0; |
4641 | } | |
4642 | ||
4643 | len = btrfs_file_extent_num_bytes(leaf, extent); | |
4644 | /* Last extent goes beyond i_size, no need to log a hole. */ | |
4645 | if (key.offset + len > i_size) | |
4646 | return 0; | |
4647 | hole_start = key.offset + len; | |
4648 | hole_size = i_size - hole_start; | |
4649 | } | |
4650 | btrfs_release_path(path); | |
4651 | ||
4652 | /* Last extent ends at i_size. */ | |
4653 | if (hole_size == 0) | |
4654 | return 0; | |
4655 | ||
0b246afa | 4656 | hole_size = ALIGN(hole_size, fs_info->sectorsize); |
a89ca6f2 FM |
4657 | ret = btrfs_insert_file_extent(trans, log, ino, hole_start, 0, 0, |
4658 | hole_size, 0, hole_size, 0, 0, 0); | |
4659 | return ret; | |
4660 | } | |
4661 | ||
56f23fdb FM |
4662 | /* |
4663 | * When we are logging a new inode X, check if it doesn't have a reference that | |
4664 | * matches the reference from some other inode Y created in a past transaction | |
4665 | * and that was renamed in the current transaction. If we don't do this, then at | |
4666 | * log replay time we can lose inode Y (and all its files if it's a directory): | |
4667 | * | |
4668 | * mkdir /mnt/x | |
4669 | * echo "hello world" > /mnt/x/foobar | |
4670 | * sync | |
4671 | * mv /mnt/x /mnt/y | |
4672 | * mkdir /mnt/x # or touch /mnt/x | |
4673 | * xfs_io -c fsync /mnt/x | |
4674 | * <power fail> | |
4675 | * mount fs, trigger log replay | |
4676 | * | |
4677 | * After the log replay procedure, we would lose the first directory and all its | |
4678 | * files (file foobar). | |
4679 | * For the case where inode Y is not a directory we simply end up losing it: | |
4680 | * | |
4681 | * echo "123" > /mnt/foo | |
4682 | * sync | |
4683 | * mv /mnt/foo /mnt/bar | |
4684 | * echo "abc" > /mnt/foo | |
4685 | * xfs_io -c fsync /mnt/foo | |
4686 | * <power fail> | |
4687 | * | |
4688 | * We also need this for cases where a snapshot entry is replaced by some other | |
4689 | * entry (file or directory) otherwise we end up with an unreplayable log due to | |
4690 | * attempts to delete the snapshot entry (entry of type BTRFS_ROOT_ITEM_KEY) as | |
4691 | * if it were a regular entry: | |
4692 | * | |
4693 | * mkdir /mnt/x | |
4694 | * btrfs subvolume snapshot /mnt /mnt/x/snap | |
4695 | * btrfs subvolume delete /mnt/x/snap | |
4696 | * rmdir /mnt/x | |
4697 | * mkdir /mnt/x | |
4698 | * fsync /mnt/x or fsync some new file inside it | |
4699 | * <power fail> | |
4700 | * | |
4701 | * The snapshot delete, rmdir of x, mkdir of a new x and the fsync all happen in | |
4702 | * the same transaction. | |
4703 | */ | |
4704 | static int btrfs_check_ref_name_override(struct extent_buffer *eb, | |
4705 | const int slot, | |
4706 | const struct btrfs_key *key, | |
4791c8f1 | 4707 | struct btrfs_inode *inode, |
44f714da | 4708 | u64 *other_ino) |
56f23fdb FM |
4709 | { |
4710 | int ret; | |
4711 | struct btrfs_path *search_path; | |
4712 | char *name = NULL; | |
4713 | u32 name_len = 0; | |
4714 | u32 item_size = btrfs_item_size_nr(eb, slot); | |
4715 | u32 cur_offset = 0; | |
4716 | unsigned long ptr = btrfs_item_ptr_offset(eb, slot); | |
4717 | ||
4718 | search_path = btrfs_alloc_path(); | |
4719 | if (!search_path) | |
4720 | return -ENOMEM; | |
4721 | search_path->search_commit_root = 1; | |
4722 | search_path->skip_locking = 1; | |
4723 | ||
4724 | while (cur_offset < item_size) { | |
4725 | u64 parent; | |
4726 | u32 this_name_len; | |
4727 | u32 this_len; | |
4728 | unsigned long name_ptr; | |
4729 | struct btrfs_dir_item *di; | |
4730 | ||
4731 | if (key->type == BTRFS_INODE_REF_KEY) { | |
4732 | struct btrfs_inode_ref *iref; | |
4733 | ||
4734 | iref = (struct btrfs_inode_ref *)(ptr + cur_offset); | |
4735 | parent = key->offset; | |
4736 | this_name_len = btrfs_inode_ref_name_len(eb, iref); | |
4737 | name_ptr = (unsigned long)(iref + 1); | |
4738 | this_len = sizeof(*iref) + this_name_len; | |
4739 | } else { | |
4740 | struct btrfs_inode_extref *extref; | |
4741 | ||
4742 | extref = (struct btrfs_inode_extref *)(ptr + | |
4743 | cur_offset); | |
4744 | parent = btrfs_inode_extref_parent(eb, extref); | |
4745 | this_name_len = btrfs_inode_extref_name_len(eb, extref); | |
4746 | name_ptr = (unsigned long)&extref->name; | |
4747 | this_len = sizeof(*extref) + this_name_len; | |
4748 | } | |
4749 | ||
4750 | if (this_name_len > name_len) { | |
4751 | char *new_name; | |
4752 | ||
4753 | new_name = krealloc(name, this_name_len, GFP_NOFS); | |
4754 | if (!new_name) { | |
4755 | ret = -ENOMEM; | |
4756 | goto out; | |
4757 | } | |
4758 | name_len = this_name_len; | |
4759 | name = new_name; | |
4760 | } | |
4761 | ||
4762 | read_extent_buffer(eb, name, name_ptr, this_name_len); | |
4791c8f1 NB |
4763 | di = btrfs_lookup_dir_item(NULL, inode->root, search_path, |
4764 | parent, name, this_name_len, 0); | |
56f23fdb | 4765 | if (di && !IS_ERR(di)) { |
44f714da FM |
4766 | struct btrfs_key di_key; |
4767 | ||
4768 | btrfs_dir_item_key_to_cpu(search_path->nodes[0], | |
4769 | di, &di_key); | |
4770 | if (di_key.type == BTRFS_INODE_ITEM_KEY) { | |
4771 | ret = 1; | |
4772 | *other_ino = di_key.objectid; | |
4773 | } else { | |
4774 | ret = -EAGAIN; | |
4775 | } | |
56f23fdb FM |
4776 | goto out; |
4777 | } else if (IS_ERR(di)) { | |
4778 | ret = PTR_ERR(di); | |
4779 | goto out; | |
4780 | } | |
4781 | btrfs_release_path(search_path); | |
4782 | ||
4783 | cur_offset += this_len; | |
4784 | } | |
4785 | ret = 0; | |
4786 | out: | |
4787 | btrfs_free_path(search_path); | |
4788 | kfree(name); | |
4789 | return ret; | |
4790 | } | |
4791 | ||
e02119d5 CM |
4792 | /* log a single inode in the tree log. |
4793 | * At least one parent directory for this inode must exist in the tree | |
4794 | * or be logged already. | |
4795 | * | |
4796 | * Any items from this inode changed by the current transaction are copied | |
4797 | * to the log tree. An extra reference is taken on any extents in this | |
4798 | * file, allowing us to avoid a whole pile of corner cases around logging | |
4799 | * blocks that have been removed from the tree. | |
4800 | * | |
4801 | * See LOG_INODE_ALL and related defines for a description of what inode_only | |
4802 | * does. | |
4803 | * | |
4804 | * This handles both files and directories. | |
4805 | */ | |
12fcfd22 | 4806 | static int btrfs_log_inode(struct btrfs_trans_handle *trans, |
a59108a7 | 4807 | struct btrfs_root *root, struct btrfs_inode *inode, |
49dae1bc FM |
4808 | int inode_only, |
4809 | const loff_t start, | |
8407f553 FM |
4810 | const loff_t end, |
4811 | struct btrfs_log_ctx *ctx) | |
e02119d5 | 4812 | { |
0b246afa | 4813 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 CM |
4814 | struct btrfs_path *path; |
4815 | struct btrfs_path *dst_path; | |
4816 | struct btrfs_key min_key; | |
4817 | struct btrfs_key max_key; | |
4818 | struct btrfs_root *log = root->log_root; | |
16e7549f | 4819 | u64 last_extent = 0; |
4a500fd1 | 4820 | int err = 0; |
e02119d5 | 4821 | int ret; |
3a5f1d45 | 4822 | int nritems; |
31ff1cd2 CM |
4823 | int ins_start_slot = 0; |
4824 | int ins_nr; | |
5dc562c5 | 4825 | bool fast_search = false; |
a59108a7 NB |
4826 | u64 ino = btrfs_ino(inode); |
4827 | struct extent_map_tree *em_tree = &inode->extent_tree; | |
1a4bcf47 | 4828 | u64 logged_isize = 0; |
e4545de5 | 4829 | bool need_log_inode_item = true; |
9a8fca62 | 4830 | bool xattrs_logged = false; |
e02119d5 | 4831 | |
e02119d5 | 4832 | path = btrfs_alloc_path(); |
5df67083 TI |
4833 | if (!path) |
4834 | return -ENOMEM; | |
e02119d5 | 4835 | dst_path = btrfs_alloc_path(); |
5df67083 TI |
4836 | if (!dst_path) { |
4837 | btrfs_free_path(path); | |
4838 | return -ENOMEM; | |
4839 | } | |
e02119d5 | 4840 | |
33345d01 | 4841 | min_key.objectid = ino; |
e02119d5 CM |
4842 | min_key.type = BTRFS_INODE_ITEM_KEY; |
4843 | min_key.offset = 0; | |
4844 | ||
33345d01 | 4845 | max_key.objectid = ino; |
12fcfd22 | 4846 | |
12fcfd22 | 4847 | |
5dc562c5 | 4848 | /* today the code can only do partial logging of directories */ |
a59108a7 | 4849 | if (S_ISDIR(inode->vfs_inode.i_mode) || |
5269b67e | 4850 | (!test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, |
a59108a7 | 4851 | &inode->runtime_flags) && |
781feef7 | 4852 | inode_only >= LOG_INODE_EXISTS)) |
e02119d5 CM |
4853 | max_key.type = BTRFS_XATTR_ITEM_KEY; |
4854 | else | |
4855 | max_key.type = (u8)-1; | |
4856 | max_key.offset = (u64)-1; | |
4857 | ||
2c2c452b FM |
4858 | /* |
4859 | * Only run delayed items if we are a dir or a new file. | |
4860 | * Otherwise commit the delayed inode only, which is needed in | |
4861 | * order for the log replay code to mark inodes for link count | |
4862 | * fixup (create temporary BTRFS_TREE_LOG_FIXUP_OBJECTID items). | |
4863 | */ | |
a59108a7 NB |
4864 | if (S_ISDIR(inode->vfs_inode.i_mode) || |
4865 | inode->generation > fs_info->last_trans_committed) | |
4866 | ret = btrfs_commit_inode_delayed_items(trans, inode); | |
2c2c452b | 4867 | else |
a59108a7 | 4868 | ret = btrfs_commit_inode_delayed_inode(inode); |
2c2c452b FM |
4869 | |
4870 | if (ret) { | |
4871 | btrfs_free_path(path); | |
4872 | btrfs_free_path(dst_path); | |
4873 | return ret; | |
16cdcec7 MX |
4874 | } |
4875 | ||
781feef7 LB |
4876 | if (inode_only == LOG_OTHER_INODE) { |
4877 | inode_only = LOG_INODE_EXISTS; | |
a59108a7 | 4878 | mutex_lock_nested(&inode->log_mutex, SINGLE_DEPTH_NESTING); |
781feef7 | 4879 | } else { |
a59108a7 | 4880 | mutex_lock(&inode->log_mutex); |
781feef7 | 4881 | } |
e02119d5 CM |
4882 | |
4883 | /* | |
4884 | * a brute force approach to making sure we get the most uptodate | |
4885 | * copies of everything. | |
4886 | */ | |
a59108a7 | 4887 | if (S_ISDIR(inode->vfs_inode.i_mode)) { |
e02119d5 CM |
4888 | int max_key_type = BTRFS_DIR_LOG_INDEX_KEY; |
4889 | ||
4f764e51 FM |
4890 | if (inode_only == LOG_INODE_EXISTS) |
4891 | max_key_type = BTRFS_XATTR_ITEM_KEY; | |
33345d01 | 4892 | ret = drop_objectid_items(trans, log, path, ino, max_key_type); |
e02119d5 | 4893 | } else { |
1a4bcf47 FM |
4894 | if (inode_only == LOG_INODE_EXISTS) { |
4895 | /* | |
4896 | * Make sure the new inode item we write to the log has | |
4897 | * the same isize as the current one (if it exists). | |
4898 | * This is necessary to prevent data loss after log | |
4899 | * replay, and also to prevent doing a wrong expanding | |
4900 | * truncate - for e.g. create file, write 4K into offset | |
4901 | * 0, fsync, write 4K into offset 4096, add hard link, | |
4902 | * fsync some other file (to sync log), power fail - if | |
4903 | * we use the inode's current i_size, after log replay | |
4904 | * we get a 8Kb file, with the last 4Kb extent as a hole | |
4905 | * (zeroes), as if an expanding truncate happened, | |
4906 | * instead of getting a file of 4Kb only. | |
4907 | */ | |
a59108a7 | 4908 | err = logged_inode_size(log, inode, path, &logged_isize); |
1a4bcf47 FM |
4909 | if (err) |
4910 | goto out_unlock; | |
4911 | } | |
a742994a | 4912 | if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, |
a59108a7 | 4913 | &inode->runtime_flags)) { |
a742994a | 4914 | if (inode_only == LOG_INODE_EXISTS) { |
4f764e51 | 4915 | max_key.type = BTRFS_XATTR_ITEM_KEY; |
a742994a FM |
4916 | ret = drop_objectid_items(trans, log, path, ino, |
4917 | max_key.type); | |
4918 | } else { | |
4919 | clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC, | |
a59108a7 | 4920 | &inode->runtime_flags); |
a742994a | 4921 | clear_bit(BTRFS_INODE_COPY_EVERYTHING, |
a59108a7 | 4922 | &inode->runtime_flags); |
28ed1345 CM |
4923 | while(1) { |
4924 | ret = btrfs_truncate_inode_items(trans, | |
a59108a7 | 4925 | log, &inode->vfs_inode, 0, 0); |
28ed1345 CM |
4926 | if (ret != -EAGAIN) |
4927 | break; | |
4928 | } | |
a742994a | 4929 | } |
4f764e51 | 4930 | } else if (test_and_clear_bit(BTRFS_INODE_COPY_EVERYTHING, |
a59108a7 | 4931 | &inode->runtime_flags) || |
6cfab851 | 4932 | inode_only == LOG_INODE_EXISTS) { |
4f764e51 | 4933 | if (inode_only == LOG_INODE_ALL) |
183f37fa | 4934 | fast_search = true; |
4f764e51 | 4935 | max_key.type = BTRFS_XATTR_ITEM_KEY; |
5dc562c5 | 4936 | ret = drop_objectid_items(trans, log, path, ino, |
e9976151 | 4937 | max_key.type); |
a95249b3 JB |
4938 | } else { |
4939 | if (inode_only == LOG_INODE_ALL) | |
4940 | fast_search = true; | |
a95249b3 | 4941 | goto log_extents; |
5dc562c5 | 4942 | } |
a95249b3 | 4943 | |
e02119d5 | 4944 | } |
4a500fd1 YZ |
4945 | if (ret) { |
4946 | err = ret; | |
4947 | goto out_unlock; | |
4948 | } | |
e02119d5 | 4949 | |
d397712b | 4950 | while (1) { |
31ff1cd2 | 4951 | ins_nr = 0; |
6174d3cb | 4952 | ret = btrfs_search_forward(root, &min_key, |
de78b51a | 4953 | path, trans->transid); |
fb770ae4 LB |
4954 | if (ret < 0) { |
4955 | err = ret; | |
4956 | goto out_unlock; | |
4957 | } | |
e02119d5 CM |
4958 | if (ret != 0) |
4959 | break; | |
3a5f1d45 | 4960 | again: |
31ff1cd2 | 4961 | /* note, ins_nr might be > 0 here, cleanup outside the loop */ |
33345d01 | 4962 | if (min_key.objectid != ino) |
e02119d5 CM |
4963 | break; |
4964 | if (min_key.type > max_key.type) | |
4965 | break; | |
31ff1cd2 | 4966 | |
e4545de5 FM |
4967 | if (min_key.type == BTRFS_INODE_ITEM_KEY) |
4968 | need_log_inode_item = false; | |
4969 | ||
56f23fdb FM |
4970 | if ((min_key.type == BTRFS_INODE_REF_KEY || |
4971 | min_key.type == BTRFS_INODE_EXTREF_KEY) && | |
a59108a7 | 4972 | inode->generation == trans->transid) { |
44f714da FM |
4973 | u64 other_ino = 0; |
4974 | ||
56f23fdb | 4975 | ret = btrfs_check_ref_name_override(path->nodes[0], |
a59108a7 NB |
4976 | path->slots[0], &min_key, inode, |
4977 | &other_ino); | |
56f23fdb FM |
4978 | if (ret < 0) { |
4979 | err = ret; | |
4980 | goto out_unlock; | |
28a23593 | 4981 | } else if (ret > 0 && ctx && |
4a0cc7ca | 4982 | other_ino != btrfs_ino(BTRFS_I(ctx->inode))) { |
44f714da FM |
4983 | struct btrfs_key inode_key; |
4984 | struct inode *other_inode; | |
4985 | ||
4986 | if (ins_nr > 0) { | |
4987 | ins_nr++; | |
4988 | } else { | |
4989 | ins_nr = 1; | |
4990 | ins_start_slot = path->slots[0]; | |
4991 | } | |
a59108a7 | 4992 | ret = copy_items(trans, inode, dst_path, path, |
44f714da FM |
4993 | &last_extent, ins_start_slot, |
4994 | ins_nr, inode_only, | |
4995 | logged_isize); | |
4996 | if (ret < 0) { | |
4997 | err = ret; | |
4998 | goto out_unlock; | |
4999 | } | |
5000 | ins_nr = 0; | |
5001 | btrfs_release_path(path); | |
5002 | inode_key.objectid = other_ino; | |
5003 | inode_key.type = BTRFS_INODE_ITEM_KEY; | |
5004 | inode_key.offset = 0; | |
0b246afa | 5005 | other_inode = btrfs_iget(fs_info->sb, |
44f714da FM |
5006 | &inode_key, root, |
5007 | NULL); | |
5008 | /* | |
5009 | * If the other inode that had a conflicting dir | |
5010 | * entry was deleted in the current transaction, | |
5011 | * we don't need to do more work nor fallback to | |
5012 | * a transaction commit. | |
5013 | */ | |
8d9e220c | 5014 | if (other_inode == ERR_PTR(-ENOENT)) { |
44f714da FM |
5015 | goto next_key; |
5016 | } else if (IS_ERR(other_inode)) { | |
5017 | err = PTR_ERR(other_inode); | |
5018 | goto out_unlock; | |
5019 | } | |
5020 | /* | |
5021 | * We are safe logging the other inode without | |
5022 | * acquiring its i_mutex as long as we log with | |
5023 | * the LOG_INODE_EXISTS mode. We're safe against | |
5024 | * concurrent renames of the other inode as well | |
5025 | * because during a rename we pin the log and | |
5026 | * update the log with the new name before we | |
5027 | * unpin it. | |
5028 | */ | |
a59108a7 NB |
5029 | err = btrfs_log_inode(trans, root, |
5030 | BTRFS_I(other_inode), | |
5031 | LOG_OTHER_INODE, 0, LLONG_MAX, | |
5032 | ctx); | |
44f714da FM |
5033 | iput(other_inode); |
5034 | if (err) | |
5035 | goto out_unlock; | |
5036 | else | |
5037 | goto next_key; | |
56f23fdb FM |
5038 | } |
5039 | } | |
5040 | ||
36283bf7 FM |
5041 | /* Skip xattrs, we log them later with btrfs_log_all_xattrs() */ |
5042 | if (min_key.type == BTRFS_XATTR_ITEM_KEY) { | |
5043 | if (ins_nr == 0) | |
5044 | goto next_slot; | |
a59108a7 | 5045 | ret = copy_items(trans, inode, dst_path, path, |
36283bf7 FM |
5046 | &last_extent, ins_start_slot, |
5047 | ins_nr, inode_only, logged_isize); | |
5048 | if (ret < 0) { | |
5049 | err = ret; | |
5050 | goto out_unlock; | |
5051 | } | |
5052 | ins_nr = 0; | |
5053 | if (ret) { | |
5054 | btrfs_release_path(path); | |
5055 | continue; | |
5056 | } | |
5057 | goto next_slot; | |
5058 | } | |
5059 | ||
31ff1cd2 CM |
5060 | if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) { |
5061 | ins_nr++; | |
5062 | goto next_slot; | |
5063 | } else if (!ins_nr) { | |
5064 | ins_start_slot = path->slots[0]; | |
5065 | ins_nr = 1; | |
5066 | goto next_slot; | |
e02119d5 CM |
5067 | } |
5068 | ||
a59108a7 | 5069 | ret = copy_items(trans, inode, dst_path, path, &last_extent, |
1a4bcf47 FM |
5070 | ins_start_slot, ins_nr, inode_only, |
5071 | logged_isize); | |
16e7549f | 5072 | if (ret < 0) { |
4a500fd1 YZ |
5073 | err = ret; |
5074 | goto out_unlock; | |
a71db86e RV |
5075 | } |
5076 | if (ret) { | |
16e7549f JB |
5077 | ins_nr = 0; |
5078 | btrfs_release_path(path); | |
5079 | continue; | |
4a500fd1 | 5080 | } |
31ff1cd2 CM |
5081 | ins_nr = 1; |
5082 | ins_start_slot = path->slots[0]; | |
5083 | next_slot: | |
e02119d5 | 5084 | |
3a5f1d45 CM |
5085 | nritems = btrfs_header_nritems(path->nodes[0]); |
5086 | path->slots[0]++; | |
5087 | if (path->slots[0] < nritems) { | |
5088 | btrfs_item_key_to_cpu(path->nodes[0], &min_key, | |
5089 | path->slots[0]); | |
5090 | goto again; | |
5091 | } | |
31ff1cd2 | 5092 | if (ins_nr) { |
a59108a7 | 5093 | ret = copy_items(trans, inode, dst_path, path, |
16e7549f | 5094 | &last_extent, ins_start_slot, |
1a4bcf47 | 5095 | ins_nr, inode_only, logged_isize); |
16e7549f | 5096 | if (ret < 0) { |
4a500fd1 YZ |
5097 | err = ret; |
5098 | goto out_unlock; | |
5099 | } | |
16e7549f | 5100 | ret = 0; |
31ff1cd2 CM |
5101 | ins_nr = 0; |
5102 | } | |
b3b4aa74 | 5103 | btrfs_release_path(path); |
44f714da | 5104 | next_key: |
3d41d702 | 5105 | if (min_key.offset < (u64)-1) { |
e02119d5 | 5106 | min_key.offset++; |
3d41d702 | 5107 | } else if (min_key.type < max_key.type) { |
e02119d5 | 5108 | min_key.type++; |
3d41d702 FDBM |
5109 | min_key.offset = 0; |
5110 | } else { | |
e02119d5 | 5111 | break; |
3d41d702 | 5112 | } |
e02119d5 | 5113 | } |
31ff1cd2 | 5114 | if (ins_nr) { |
a59108a7 | 5115 | ret = copy_items(trans, inode, dst_path, path, &last_extent, |
1a4bcf47 FM |
5116 | ins_start_slot, ins_nr, inode_only, |
5117 | logged_isize); | |
16e7549f | 5118 | if (ret < 0) { |
4a500fd1 YZ |
5119 | err = ret; |
5120 | goto out_unlock; | |
5121 | } | |
16e7549f | 5122 | ret = 0; |
31ff1cd2 CM |
5123 | ins_nr = 0; |
5124 | } | |
5dc562c5 | 5125 | |
36283bf7 FM |
5126 | btrfs_release_path(path); |
5127 | btrfs_release_path(dst_path); | |
a59108a7 | 5128 | err = btrfs_log_all_xattrs(trans, root, inode, path, dst_path); |
36283bf7 FM |
5129 | if (err) |
5130 | goto out_unlock; | |
9a8fca62 | 5131 | xattrs_logged = true; |
a89ca6f2 FM |
5132 | if (max_key.type >= BTRFS_EXTENT_DATA_KEY && !fast_search) { |
5133 | btrfs_release_path(path); | |
5134 | btrfs_release_path(dst_path); | |
a59108a7 | 5135 | err = btrfs_log_trailing_hole(trans, root, inode, path); |
a89ca6f2 FM |
5136 | if (err) |
5137 | goto out_unlock; | |
5138 | } | |
a95249b3 | 5139 | log_extents: |
f3b15ccd JB |
5140 | btrfs_release_path(path); |
5141 | btrfs_release_path(dst_path); | |
e4545de5 | 5142 | if (need_log_inode_item) { |
a59108a7 | 5143 | err = log_inode_item(trans, log, dst_path, inode); |
9a8fca62 FM |
5144 | if (!err && !xattrs_logged) { |
5145 | err = btrfs_log_all_xattrs(trans, root, inode, path, | |
5146 | dst_path); | |
5147 | btrfs_release_path(path); | |
5148 | } | |
e4545de5 FM |
5149 | if (err) |
5150 | goto out_unlock; | |
5151 | } | |
5dc562c5 | 5152 | if (fast_search) { |
a59108a7 | 5153 | ret = btrfs_log_changed_extents(trans, root, inode, dst_path, |
a2120a47 | 5154 | ctx, start, end); |
5dc562c5 JB |
5155 | if (ret) { |
5156 | err = ret; | |
5157 | goto out_unlock; | |
5158 | } | |
d006a048 | 5159 | } else if (inode_only == LOG_INODE_ALL) { |
06d3d22b LB |
5160 | struct extent_map *em, *n; |
5161 | ||
49dae1bc FM |
5162 | write_lock(&em_tree->lock); |
5163 | /* | |
5164 | * We can't just remove every em if we're called for a ranged | |
5165 | * fsync - that is, one that doesn't cover the whole possible | |
5166 | * file range (0 to LLONG_MAX). This is because we can have | |
5167 | * em's that fall outside the range we're logging and therefore | |
5168 | * their ordered operations haven't completed yet | |
5169 | * (btrfs_finish_ordered_io() not invoked yet). This means we | |
5170 | * didn't get their respective file extent item in the fs/subvol | |
5171 | * tree yet, and need to let the next fast fsync (one which | |
5172 | * consults the list of modified extent maps) find the em so | |
5173 | * that it logs a matching file extent item and waits for the | |
5174 | * respective ordered operation to complete (if it's still | |
5175 | * running). | |
5176 | * | |
5177 | * Removing every em outside the range we're logging would make | |
5178 | * the next fast fsync not log their matching file extent items, | |
5179 | * therefore making us lose data after a log replay. | |
5180 | */ | |
5181 | list_for_each_entry_safe(em, n, &em_tree->modified_extents, | |
5182 | list) { | |
5183 | const u64 mod_end = em->mod_start + em->mod_len - 1; | |
5184 | ||
5185 | if (em->mod_start >= start && mod_end <= end) | |
5186 | list_del_init(&em->list); | |
5187 | } | |
5188 | write_unlock(&em_tree->lock); | |
5dc562c5 JB |
5189 | } |
5190 | ||
a59108a7 NB |
5191 | if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->vfs_inode.i_mode)) { |
5192 | ret = log_directory_changes(trans, root, inode, path, dst_path, | |
5193 | ctx); | |
4a500fd1 YZ |
5194 | if (ret) { |
5195 | err = ret; | |
5196 | goto out_unlock; | |
5197 | } | |
e02119d5 | 5198 | } |
49dae1bc | 5199 | |
a59108a7 NB |
5200 | spin_lock(&inode->lock); |
5201 | inode->logged_trans = trans->transid; | |
5202 | inode->last_log_commit = inode->last_sub_trans; | |
5203 | spin_unlock(&inode->lock); | |
4a500fd1 | 5204 | out_unlock: |
a59108a7 | 5205 | mutex_unlock(&inode->log_mutex); |
e02119d5 CM |
5206 | |
5207 | btrfs_free_path(path); | |
5208 | btrfs_free_path(dst_path); | |
4a500fd1 | 5209 | return err; |
e02119d5 CM |
5210 | } |
5211 | ||
2be63d5c FM |
5212 | /* |
5213 | * Check if we must fallback to a transaction commit when logging an inode. | |
5214 | * This must be called after logging the inode and is used only in the context | |
5215 | * when fsyncing an inode requires the need to log some other inode - in which | |
5216 | * case we can't lock the i_mutex of each other inode we need to log as that | |
5217 | * can lead to deadlocks with concurrent fsync against other inodes (as we can | |
5218 | * log inodes up or down in the hierarchy) or rename operations for example. So | |
5219 | * we take the log_mutex of the inode after we have logged it and then check for | |
5220 | * its last_unlink_trans value - this is safe because any task setting | |
5221 | * last_unlink_trans must take the log_mutex and it must do this before it does | |
5222 | * the actual unlink operation, so if we do this check before a concurrent task | |
5223 | * sets last_unlink_trans it means we've logged a consistent version/state of | |
5224 | * all the inode items, otherwise we are not sure and must do a transaction | |
01327610 | 5225 | * commit (the concurrent task might have only updated last_unlink_trans before |
2be63d5c FM |
5226 | * we logged the inode or it might have also done the unlink). |
5227 | */ | |
5228 | static bool btrfs_must_commit_transaction(struct btrfs_trans_handle *trans, | |
ab1717b2 | 5229 | struct btrfs_inode *inode) |
2be63d5c | 5230 | { |
ab1717b2 | 5231 | struct btrfs_fs_info *fs_info = inode->root->fs_info; |
2be63d5c FM |
5232 | bool ret = false; |
5233 | ||
ab1717b2 NB |
5234 | mutex_lock(&inode->log_mutex); |
5235 | if (inode->last_unlink_trans > fs_info->last_trans_committed) { | |
2be63d5c FM |
5236 | /* |
5237 | * Make sure any commits to the log are forced to be full | |
5238 | * commits. | |
5239 | */ | |
5240 | btrfs_set_log_full_commit(fs_info, trans); | |
5241 | ret = true; | |
5242 | } | |
ab1717b2 | 5243 | mutex_unlock(&inode->log_mutex); |
2be63d5c FM |
5244 | |
5245 | return ret; | |
5246 | } | |
5247 | ||
12fcfd22 CM |
5248 | /* |
5249 | * follow the dentry parent pointers up the chain and see if any | |
5250 | * of the directories in it require a full commit before they can | |
5251 | * be logged. Returns zero if nothing special needs to be done or 1 if | |
5252 | * a full commit is required. | |
5253 | */ | |
5254 | static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans, | |
aefa6115 | 5255 | struct btrfs_inode *inode, |
12fcfd22 CM |
5256 | struct dentry *parent, |
5257 | struct super_block *sb, | |
5258 | u64 last_committed) | |
e02119d5 | 5259 | { |
12fcfd22 | 5260 | int ret = 0; |
6a912213 | 5261 | struct dentry *old_parent = NULL; |
aefa6115 | 5262 | struct btrfs_inode *orig_inode = inode; |
e02119d5 | 5263 | |
af4176b4 CM |
5264 | /* |
5265 | * for regular files, if its inode is already on disk, we don't | |
5266 | * have to worry about the parents at all. This is because | |
5267 | * we can use the last_unlink_trans field to record renames | |
5268 | * and other fun in this file. | |
5269 | */ | |
aefa6115 NB |
5270 | if (S_ISREG(inode->vfs_inode.i_mode) && |
5271 | inode->generation <= last_committed && | |
5272 | inode->last_unlink_trans <= last_committed) | |
5273 | goto out; | |
af4176b4 | 5274 | |
aefa6115 | 5275 | if (!S_ISDIR(inode->vfs_inode.i_mode)) { |
fc64005c | 5276 | if (!parent || d_really_is_negative(parent) || sb != parent->d_sb) |
12fcfd22 | 5277 | goto out; |
aefa6115 | 5278 | inode = BTRFS_I(d_inode(parent)); |
12fcfd22 CM |
5279 | } |
5280 | ||
5281 | while (1) { | |
de2b530b JB |
5282 | /* |
5283 | * If we are logging a directory then we start with our inode, | |
01327610 | 5284 | * not our parent's inode, so we need to skip setting the |
de2b530b JB |
5285 | * logged_trans so that further down in the log code we don't |
5286 | * think this inode has already been logged. | |
5287 | */ | |
5288 | if (inode != orig_inode) | |
aefa6115 | 5289 | inode->logged_trans = trans->transid; |
12fcfd22 CM |
5290 | smp_mb(); |
5291 | ||
aefa6115 | 5292 | if (btrfs_must_commit_transaction(trans, inode)) { |
12fcfd22 CM |
5293 | ret = 1; |
5294 | break; | |
5295 | } | |
5296 | ||
fc64005c | 5297 | if (!parent || d_really_is_negative(parent) || sb != parent->d_sb) |
12fcfd22 CM |
5298 | break; |
5299 | ||
44f714da | 5300 | if (IS_ROOT(parent)) { |
aefa6115 NB |
5301 | inode = BTRFS_I(d_inode(parent)); |
5302 | if (btrfs_must_commit_transaction(trans, inode)) | |
44f714da | 5303 | ret = 1; |
12fcfd22 | 5304 | break; |
44f714da | 5305 | } |
12fcfd22 | 5306 | |
6a912213 JB |
5307 | parent = dget_parent(parent); |
5308 | dput(old_parent); | |
5309 | old_parent = parent; | |
aefa6115 | 5310 | inode = BTRFS_I(d_inode(parent)); |
12fcfd22 CM |
5311 | |
5312 | } | |
6a912213 | 5313 | dput(old_parent); |
12fcfd22 | 5314 | out: |
e02119d5 CM |
5315 | return ret; |
5316 | } | |
5317 | ||
2f2ff0ee FM |
5318 | struct btrfs_dir_list { |
5319 | u64 ino; | |
5320 | struct list_head list; | |
5321 | }; | |
5322 | ||
5323 | /* | |
5324 | * Log the inodes of the new dentries of a directory. See log_dir_items() for | |
5325 | * details about the why it is needed. | |
5326 | * This is a recursive operation - if an existing dentry corresponds to a | |
5327 | * directory, that directory's new entries are logged too (same behaviour as | |
5328 | * ext3/4, xfs, f2fs, reiserfs, nilfs2). Note that when logging the inodes | |
5329 | * the dentries point to we do not lock their i_mutex, otherwise lockdep | |
5330 | * complains about the following circular lock dependency / possible deadlock: | |
5331 | * | |
5332 | * CPU0 CPU1 | |
5333 | * ---- ---- | |
5334 | * lock(&type->i_mutex_dir_key#3/2); | |
5335 | * lock(sb_internal#2); | |
5336 | * lock(&type->i_mutex_dir_key#3/2); | |
5337 | * lock(&sb->s_type->i_mutex_key#14); | |
5338 | * | |
5339 | * Where sb_internal is the lock (a counter that works as a lock) acquired by | |
5340 | * sb_start_intwrite() in btrfs_start_transaction(). | |
5341 | * Not locking i_mutex of the inodes is still safe because: | |
5342 | * | |
5343 | * 1) For regular files we log with a mode of LOG_INODE_EXISTS. It's possible | |
5344 | * that while logging the inode new references (names) are added or removed | |
5345 | * from the inode, leaving the logged inode item with a link count that does | |
5346 | * not match the number of logged inode reference items. This is fine because | |
5347 | * at log replay time we compute the real number of links and correct the | |
5348 | * link count in the inode item (see replay_one_buffer() and | |
5349 | * link_to_fixup_dir()); | |
5350 | * | |
5351 | * 2) For directories we log with a mode of LOG_INODE_ALL. It's possible that | |
5352 | * while logging the inode's items new items with keys BTRFS_DIR_ITEM_KEY and | |
5353 | * BTRFS_DIR_INDEX_KEY are added to fs/subvol tree and the logged inode item | |
5354 | * has a size that doesn't match the sum of the lengths of all the logged | |
5355 | * names. This does not result in a problem because if a dir_item key is | |
5356 | * logged but its matching dir_index key is not logged, at log replay time we | |
5357 | * don't use it to replay the respective name (see replay_one_name()). On the | |
5358 | * other hand if only the dir_index key ends up being logged, the respective | |
5359 | * name is added to the fs/subvol tree with both the dir_item and dir_index | |
5360 | * keys created (see replay_one_name()). | |
5361 | * The directory's inode item with a wrong i_size is not a problem as well, | |
5362 | * since we don't use it at log replay time to set the i_size in the inode | |
5363 | * item of the fs/subvol tree (see overwrite_item()). | |
5364 | */ | |
5365 | static int log_new_dir_dentries(struct btrfs_trans_handle *trans, | |
5366 | struct btrfs_root *root, | |
51cc0d32 | 5367 | struct btrfs_inode *start_inode, |
2f2ff0ee FM |
5368 | struct btrfs_log_ctx *ctx) |
5369 | { | |
0b246afa | 5370 | struct btrfs_fs_info *fs_info = root->fs_info; |
2f2ff0ee FM |
5371 | struct btrfs_root *log = root->log_root; |
5372 | struct btrfs_path *path; | |
5373 | LIST_HEAD(dir_list); | |
5374 | struct btrfs_dir_list *dir_elem; | |
5375 | int ret = 0; | |
5376 | ||
5377 | path = btrfs_alloc_path(); | |
5378 | if (!path) | |
5379 | return -ENOMEM; | |
5380 | ||
5381 | dir_elem = kmalloc(sizeof(*dir_elem), GFP_NOFS); | |
5382 | if (!dir_elem) { | |
5383 | btrfs_free_path(path); | |
5384 | return -ENOMEM; | |
5385 | } | |
51cc0d32 | 5386 | dir_elem->ino = btrfs_ino(start_inode); |
2f2ff0ee FM |
5387 | list_add_tail(&dir_elem->list, &dir_list); |
5388 | ||
5389 | while (!list_empty(&dir_list)) { | |
5390 | struct extent_buffer *leaf; | |
5391 | struct btrfs_key min_key; | |
5392 | int nritems; | |
5393 | int i; | |
5394 | ||
5395 | dir_elem = list_first_entry(&dir_list, struct btrfs_dir_list, | |
5396 | list); | |
5397 | if (ret) | |
5398 | goto next_dir_inode; | |
5399 | ||
5400 | min_key.objectid = dir_elem->ino; | |
5401 | min_key.type = BTRFS_DIR_ITEM_KEY; | |
5402 | min_key.offset = 0; | |
5403 | again: | |
5404 | btrfs_release_path(path); | |
5405 | ret = btrfs_search_forward(log, &min_key, path, trans->transid); | |
5406 | if (ret < 0) { | |
5407 | goto next_dir_inode; | |
5408 | } else if (ret > 0) { | |
5409 | ret = 0; | |
5410 | goto next_dir_inode; | |
5411 | } | |
5412 | ||
5413 | process_leaf: | |
5414 | leaf = path->nodes[0]; | |
5415 | nritems = btrfs_header_nritems(leaf); | |
5416 | for (i = path->slots[0]; i < nritems; i++) { | |
5417 | struct btrfs_dir_item *di; | |
5418 | struct btrfs_key di_key; | |
5419 | struct inode *di_inode; | |
5420 | struct btrfs_dir_list *new_dir_elem; | |
5421 | int log_mode = LOG_INODE_EXISTS; | |
5422 | int type; | |
5423 | ||
5424 | btrfs_item_key_to_cpu(leaf, &min_key, i); | |
5425 | if (min_key.objectid != dir_elem->ino || | |
5426 | min_key.type != BTRFS_DIR_ITEM_KEY) | |
5427 | goto next_dir_inode; | |
5428 | ||
5429 | di = btrfs_item_ptr(leaf, i, struct btrfs_dir_item); | |
5430 | type = btrfs_dir_type(leaf, di); | |
5431 | if (btrfs_dir_transid(leaf, di) < trans->transid && | |
5432 | type != BTRFS_FT_DIR) | |
5433 | continue; | |
5434 | btrfs_dir_item_key_to_cpu(leaf, di, &di_key); | |
5435 | if (di_key.type == BTRFS_ROOT_ITEM_KEY) | |
5436 | continue; | |
5437 | ||
ec125cfb | 5438 | btrfs_release_path(path); |
0b246afa | 5439 | di_inode = btrfs_iget(fs_info->sb, &di_key, root, NULL); |
2f2ff0ee FM |
5440 | if (IS_ERR(di_inode)) { |
5441 | ret = PTR_ERR(di_inode); | |
5442 | goto next_dir_inode; | |
5443 | } | |
5444 | ||
0f8939b8 | 5445 | if (btrfs_inode_in_log(BTRFS_I(di_inode), trans->transid)) { |
2f2ff0ee | 5446 | iput(di_inode); |
ec125cfb | 5447 | break; |
2f2ff0ee FM |
5448 | } |
5449 | ||
5450 | ctx->log_new_dentries = false; | |
3f9749f6 | 5451 | if (type == BTRFS_FT_DIR || type == BTRFS_FT_SYMLINK) |
2f2ff0ee | 5452 | log_mode = LOG_INODE_ALL; |
a59108a7 | 5453 | ret = btrfs_log_inode(trans, root, BTRFS_I(di_inode), |
2f2ff0ee | 5454 | log_mode, 0, LLONG_MAX, ctx); |
2be63d5c | 5455 | if (!ret && |
ab1717b2 | 5456 | btrfs_must_commit_transaction(trans, BTRFS_I(di_inode))) |
2be63d5c | 5457 | ret = 1; |
2f2ff0ee FM |
5458 | iput(di_inode); |
5459 | if (ret) | |
5460 | goto next_dir_inode; | |
5461 | if (ctx->log_new_dentries) { | |
5462 | new_dir_elem = kmalloc(sizeof(*new_dir_elem), | |
5463 | GFP_NOFS); | |
5464 | if (!new_dir_elem) { | |
5465 | ret = -ENOMEM; | |
5466 | goto next_dir_inode; | |
5467 | } | |
5468 | new_dir_elem->ino = di_key.objectid; | |
5469 | list_add_tail(&new_dir_elem->list, &dir_list); | |
5470 | } | |
5471 | break; | |
5472 | } | |
5473 | if (i == nritems) { | |
5474 | ret = btrfs_next_leaf(log, path); | |
5475 | if (ret < 0) { | |
5476 | goto next_dir_inode; | |
5477 | } else if (ret > 0) { | |
5478 | ret = 0; | |
5479 | goto next_dir_inode; | |
5480 | } | |
5481 | goto process_leaf; | |
5482 | } | |
5483 | if (min_key.offset < (u64)-1) { | |
5484 | min_key.offset++; | |
5485 | goto again; | |
5486 | } | |
5487 | next_dir_inode: | |
5488 | list_del(&dir_elem->list); | |
5489 | kfree(dir_elem); | |
5490 | } | |
5491 | ||
5492 | btrfs_free_path(path); | |
5493 | return ret; | |
5494 | } | |
5495 | ||
18aa0922 | 5496 | static int btrfs_log_all_parents(struct btrfs_trans_handle *trans, |
d0a0b78d | 5497 | struct btrfs_inode *inode, |
18aa0922 FM |
5498 | struct btrfs_log_ctx *ctx) |
5499 | { | |
3ffbd68c | 5500 | struct btrfs_fs_info *fs_info = trans->fs_info; |
18aa0922 FM |
5501 | int ret; |
5502 | struct btrfs_path *path; | |
5503 | struct btrfs_key key; | |
d0a0b78d NB |
5504 | struct btrfs_root *root = inode->root; |
5505 | const u64 ino = btrfs_ino(inode); | |
18aa0922 FM |
5506 | |
5507 | path = btrfs_alloc_path(); | |
5508 | if (!path) | |
5509 | return -ENOMEM; | |
5510 | path->skip_locking = 1; | |
5511 | path->search_commit_root = 1; | |
5512 | ||
5513 | key.objectid = ino; | |
5514 | key.type = BTRFS_INODE_REF_KEY; | |
5515 | key.offset = 0; | |
5516 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
5517 | if (ret < 0) | |
5518 | goto out; | |
5519 | ||
5520 | while (true) { | |
5521 | struct extent_buffer *leaf = path->nodes[0]; | |
5522 | int slot = path->slots[0]; | |
5523 | u32 cur_offset = 0; | |
5524 | u32 item_size; | |
5525 | unsigned long ptr; | |
5526 | ||
5527 | if (slot >= btrfs_header_nritems(leaf)) { | |
5528 | ret = btrfs_next_leaf(root, path); | |
5529 | if (ret < 0) | |
5530 | goto out; | |
5531 | else if (ret > 0) | |
5532 | break; | |
5533 | continue; | |
5534 | } | |
5535 | ||
5536 | btrfs_item_key_to_cpu(leaf, &key, slot); | |
5537 | /* BTRFS_INODE_EXTREF_KEY is BTRFS_INODE_REF_KEY + 1 */ | |
5538 | if (key.objectid != ino || key.type > BTRFS_INODE_EXTREF_KEY) | |
5539 | break; | |
5540 | ||
5541 | item_size = btrfs_item_size_nr(leaf, slot); | |
5542 | ptr = btrfs_item_ptr_offset(leaf, slot); | |
5543 | while (cur_offset < item_size) { | |
5544 | struct btrfs_key inode_key; | |
5545 | struct inode *dir_inode; | |
5546 | ||
5547 | inode_key.type = BTRFS_INODE_ITEM_KEY; | |
5548 | inode_key.offset = 0; | |
5549 | ||
5550 | if (key.type == BTRFS_INODE_EXTREF_KEY) { | |
5551 | struct btrfs_inode_extref *extref; | |
5552 | ||
5553 | extref = (struct btrfs_inode_extref *) | |
5554 | (ptr + cur_offset); | |
5555 | inode_key.objectid = btrfs_inode_extref_parent( | |
5556 | leaf, extref); | |
5557 | cur_offset += sizeof(*extref); | |
5558 | cur_offset += btrfs_inode_extref_name_len(leaf, | |
5559 | extref); | |
5560 | } else { | |
5561 | inode_key.objectid = key.offset; | |
5562 | cur_offset = item_size; | |
5563 | } | |
5564 | ||
0b246afa | 5565 | dir_inode = btrfs_iget(fs_info->sb, &inode_key, |
18aa0922 FM |
5566 | root, NULL); |
5567 | /* If parent inode was deleted, skip it. */ | |
5568 | if (IS_ERR(dir_inode)) | |
5569 | continue; | |
5570 | ||
657ed1aa FM |
5571 | if (ctx) |
5572 | ctx->log_new_dentries = false; | |
a59108a7 | 5573 | ret = btrfs_log_inode(trans, root, BTRFS_I(dir_inode), |
18aa0922 | 5574 | LOG_INODE_ALL, 0, LLONG_MAX, ctx); |
2be63d5c | 5575 | if (!ret && |
ab1717b2 | 5576 | btrfs_must_commit_transaction(trans, BTRFS_I(dir_inode))) |
2be63d5c | 5577 | ret = 1; |
657ed1aa FM |
5578 | if (!ret && ctx && ctx->log_new_dentries) |
5579 | ret = log_new_dir_dentries(trans, root, | |
f85b7379 | 5580 | BTRFS_I(dir_inode), ctx); |
18aa0922 FM |
5581 | iput(dir_inode); |
5582 | if (ret) | |
5583 | goto out; | |
5584 | } | |
5585 | path->slots[0]++; | |
5586 | } | |
5587 | ret = 0; | |
5588 | out: | |
5589 | btrfs_free_path(path); | |
5590 | return ret; | |
5591 | } | |
5592 | ||
e02119d5 CM |
5593 | /* |
5594 | * helper function around btrfs_log_inode to make sure newly created | |
5595 | * parent directories also end up in the log. A minimal inode and backref | |
5596 | * only logging is done of any parent directories that are older than | |
5597 | * the last committed transaction | |
5598 | */ | |
48a3b636 | 5599 | static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, |
19df27a9 | 5600 | struct btrfs_inode *inode, |
49dae1bc FM |
5601 | struct dentry *parent, |
5602 | const loff_t start, | |
5603 | const loff_t end, | |
41a1eada | 5604 | int inode_only, |
8b050d35 | 5605 | struct btrfs_log_ctx *ctx) |
e02119d5 | 5606 | { |
f882274b | 5607 | struct btrfs_root *root = inode->root; |
0b246afa | 5608 | struct btrfs_fs_info *fs_info = root->fs_info; |
e02119d5 | 5609 | struct super_block *sb; |
6a912213 | 5610 | struct dentry *old_parent = NULL; |
12fcfd22 | 5611 | int ret = 0; |
0b246afa | 5612 | u64 last_committed = fs_info->last_trans_committed; |
2f2ff0ee | 5613 | bool log_dentries = false; |
19df27a9 | 5614 | struct btrfs_inode *orig_inode = inode; |
12fcfd22 | 5615 | |
19df27a9 | 5616 | sb = inode->vfs_inode.i_sb; |
12fcfd22 | 5617 | |
0b246afa | 5618 | if (btrfs_test_opt(fs_info, NOTREELOG)) { |
3a5e1404 SW |
5619 | ret = 1; |
5620 | goto end_no_trans; | |
5621 | } | |
5622 | ||
995946dd MX |
5623 | /* |
5624 | * The prev transaction commit doesn't complete, we need do | |
5625 | * full commit by ourselves. | |
5626 | */ | |
0b246afa JM |
5627 | if (fs_info->last_trans_log_full_commit > |
5628 | fs_info->last_trans_committed) { | |
12fcfd22 CM |
5629 | ret = 1; |
5630 | goto end_no_trans; | |
5631 | } | |
5632 | ||
f882274b | 5633 | if (btrfs_root_refs(&root->root_item) == 0) { |
76dda93c YZ |
5634 | ret = 1; |
5635 | goto end_no_trans; | |
5636 | } | |
5637 | ||
19df27a9 NB |
5638 | ret = check_parent_dirs_for_sync(trans, inode, parent, sb, |
5639 | last_committed); | |
12fcfd22 CM |
5640 | if (ret) |
5641 | goto end_no_trans; | |
e02119d5 | 5642 | |
19df27a9 | 5643 | if (btrfs_inode_in_log(inode, trans->transid)) { |
257c62e1 CM |
5644 | ret = BTRFS_NO_LOG_SYNC; |
5645 | goto end_no_trans; | |
5646 | } | |
5647 | ||
8b050d35 | 5648 | ret = start_log_trans(trans, root, ctx); |
4a500fd1 | 5649 | if (ret) |
e87ac136 | 5650 | goto end_no_trans; |
e02119d5 | 5651 | |
19df27a9 | 5652 | ret = btrfs_log_inode(trans, root, inode, inode_only, start, end, ctx); |
4a500fd1 YZ |
5653 | if (ret) |
5654 | goto end_trans; | |
12fcfd22 | 5655 | |
af4176b4 CM |
5656 | /* |
5657 | * for regular files, if its inode is already on disk, we don't | |
5658 | * have to worry about the parents at all. This is because | |
5659 | * we can use the last_unlink_trans field to record renames | |
5660 | * and other fun in this file. | |
5661 | */ | |
19df27a9 NB |
5662 | if (S_ISREG(inode->vfs_inode.i_mode) && |
5663 | inode->generation <= last_committed && | |
5664 | inode->last_unlink_trans <= last_committed) { | |
4a500fd1 YZ |
5665 | ret = 0; |
5666 | goto end_trans; | |
5667 | } | |
af4176b4 | 5668 | |
19df27a9 | 5669 | if (S_ISDIR(inode->vfs_inode.i_mode) && ctx && ctx->log_new_dentries) |
2f2ff0ee FM |
5670 | log_dentries = true; |
5671 | ||
18aa0922 | 5672 | /* |
01327610 | 5673 | * On unlink we must make sure all our current and old parent directory |
18aa0922 FM |
5674 | * inodes are fully logged. This is to prevent leaving dangling |
5675 | * directory index entries in directories that were our parents but are | |
5676 | * not anymore. Not doing this results in old parent directory being | |
5677 | * impossible to delete after log replay (rmdir will always fail with | |
5678 | * error -ENOTEMPTY). | |
5679 | * | |
5680 | * Example 1: | |
5681 | * | |
5682 | * mkdir testdir | |
5683 | * touch testdir/foo | |
5684 | * ln testdir/foo testdir/bar | |
5685 | * sync | |
5686 | * unlink testdir/bar | |
5687 | * xfs_io -c fsync testdir/foo | |
5688 | * <power failure> | |
5689 | * mount fs, triggers log replay | |
5690 | * | |
5691 | * If we don't log the parent directory (testdir), after log replay the | |
5692 | * directory still has an entry pointing to the file inode using the bar | |
5693 | * name, but a matching BTRFS_INODE_[REF|EXTREF]_KEY does not exist and | |
5694 | * the file inode has a link count of 1. | |
5695 | * | |
5696 | * Example 2: | |
5697 | * | |
5698 | * mkdir testdir | |
5699 | * touch foo | |
5700 | * ln foo testdir/foo2 | |
5701 | * ln foo testdir/foo3 | |
5702 | * sync | |
5703 | * unlink testdir/foo3 | |
5704 | * xfs_io -c fsync foo | |
5705 | * <power failure> | |
5706 | * mount fs, triggers log replay | |
5707 | * | |
5708 | * Similar as the first example, after log replay the parent directory | |
5709 | * testdir still has an entry pointing to the inode file with name foo3 | |
5710 | * but the file inode does not have a matching BTRFS_INODE_REF_KEY item | |
5711 | * and has a link count of 2. | |
5712 | */ | |
19df27a9 | 5713 | if (inode->last_unlink_trans > last_committed) { |
18aa0922 FM |
5714 | ret = btrfs_log_all_parents(trans, orig_inode, ctx); |
5715 | if (ret) | |
5716 | goto end_trans; | |
5717 | } | |
5718 | ||
12fcfd22 | 5719 | while (1) { |
fc64005c | 5720 | if (!parent || d_really_is_negative(parent) || sb != parent->d_sb) |
e02119d5 CM |
5721 | break; |
5722 | ||
19df27a9 NB |
5723 | inode = BTRFS_I(d_inode(parent)); |
5724 | if (root != inode->root) | |
76dda93c YZ |
5725 | break; |
5726 | ||
19df27a9 NB |
5727 | if (inode->generation > last_committed) { |
5728 | ret = btrfs_log_inode(trans, root, inode, | |
5729 | LOG_INODE_EXISTS, 0, LLONG_MAX, ctx); | |
4a500fd1 YZ |
5730 | if (ret) |
5731 | goto end_trans; | |
12fcfd22 | 5732 | } |
76dda93c | 5733 | if (IS_ROOT(parent)) |
e02119d5 | 5734 | break; |
12fcfd22 | 5735 | |
6a912213 JB |
5736 | parent = dget_parent(parent); |
5737 | dput(old_parent); | |
5738 | old_parent = parent; | |
e02119d5 | 5739 | } |
2f2ff0ee | 5740 | if (log_dentries) |
19df27a9 | 5741 | ret = log_new_dir_dentries(trans, root, orig_inode, ctx); |
2f2ff0ee FM |
5742 | else |
5743 | ret = 0; | |
4a500fd1 | 5744 | end_trans: |
6a912213 | 5745 | dput(old_parent); |
4a500fd1 | 5746 | if (ret < 0) { |
0b246afa | 5747 | btrfs_set_log_full_commit(fs_info, trans); |
4a500fd1 YZ |
5748 | ret = 1; |
5749 | } | |
8b050d35 MX |
5750 | |
5751 | if (ret) | |
5752 | btrfs_remove_log_ctx(root, ctx); | |
12fcfd22 CM |
5753 | btrfs_end_log_trans(root); |
5754 | end_no_trans: | |
5755 | return ret; | |
e02119d5 CM |
5756 | } |
5757 | ||
5758 | /* | |
5759 | * it is not safe to log dentry if the chunk root has added new | |
5760 | * chunks. This returns 0 if the dentry was logged, and 1 otherwise. | |
5761 | * If this returns 1, you must commit the transaction to safely get your | |
5762 | * data on disk. | |
5763 | */ | |
5764 | int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans, | |
e5b84f7a | 5765 | struct dentry *dentry, |
49dae1bc FM |
5766 | const loff_t start, |
5767 | const loff_t end, | |
8b050d35 | 5768 | struct btrfs_log_ctx *ctx) |
e02119d5 | 5769 | { |
6a912213 JB |
5770 | struct dentry *parent = dget_parent(dentry); |
5771 | int ret; | |
5772 | ||
f882274b NB |
5773 | ret = btrfs_log_inode_parent(trans, BTRFS_I(d_inode(dentry)), parent, |
5774 | start, end, LOG_INODE_ALL, ctx); | |
6a912213 JB |
5775 | dput(parent); |
5776 | ||
5777 | return ret; | |
e02119d5 CM |
5778 | } |
5779 | ||
5780 | /* | |
5781 | * should be called during mount to recover any replay any log trees | |
5782 | * from the FS | |
5783 | */ | |
5784 | int btrfs_recover_log_trees(struct btrfs_root *log_root_tree) | |
5785 | { | |
5786 | int ret; | |
5787 | struct btrfs_path *path; | |
5788 | struct btrfs_trans_handle *trans; | |
5789 | struct btrfs_key key; | |
5790 | struct btrfs_key found_key; | |
5791 | struct btrfs_key tmp_key; | |
5792 | struct btrfs_root *log; | |
5793 | struct btrfs_fs_info *fs_info = log_root_tree->fs_info; | |
5794 | struct walk_control wc = { | |
5795 | .process_func = process_one_buffer, | |
5796 | .stage = 0, | |
5797 | }; | |
5798 | ||
e02119d5 | 5799 | path = btrfs_alloc_path(); |
db5b493a TI |
5800 | if (!path) |
5801 | return -ENOMEM; | |
5802 | ||
afcdd129 | 5803 | set_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags); |
e02119d5 | 5804 | |
4a500fd1 | 5805 | trans = btrfs_start_transaction(fs_info->tree_root, 0); |
79787eaa JM |
5806 | if (IS_ERR(trans)) { |
5807 | ret = PTR_ERR(trans); | |
5808 | goto error; | |
5809 | } | |
e02119d5 CM |
5810 | |
5811 | wc.trans = trans; | |
5812 | wc.pin = 1; | |
5813 | ||
db5b493a | 5814 | ret = walk_log_tree(trans, log_root_tree, &wc); |
79787eaa | 5815 | if (ret) { |
5d163e0e JM |
5816 | btrfs_handle_fs_error(fs_info, ret, |
5817 | "Failed to pin buffers while recovering log root tree."); | |
79787eaa JM |
5818 | goto error; |
5819 | } | |
e02119d5 CM |
5820 | |
5821 | again: | |
5822 | key.objectid = BTRFS_TREE_LOG_OBJECTID; | |
5823 | key.offset = (u64)-1; | |
962a298f | 5824 | key.type = BTRFS_ROOT_ITEM_KEY; |
e02119d5 | 5825 | |
d397712b | 5826 | while (1) { |
e02119d5 | 5827 | ret = btrfs_search_slot(NULL, log_root_tree, &key, path, 0, 0); |
79787eaa JM |
5828 | |
5829 | if (ret < 0) { | |
34d97007 | 5830 | btrfs_handle_fs_error(fs_info, ret, |
79787eaa JM |
5831 | "Couldn't find tree log root."); |
5832 | goto error; | |
5833 | } | |
e02119d5 CM |
5834 | if (ret > 0) { |
5835 | if (path->slots[0] == 0) | |
5836 | break; | |
5837 | path->slots[0]--; | |
5838 | } | |
5839 | btrfs_item_key_to_cpu(path->nodes[0], &found_key, | |
5840 | path->slots[0]); | |
b3b4aa74 | 5841 | btrfs_release_path(path); |
e02119d5 CM |
5842 | if (found_key.objectid != BTRFS_TREE_LOG_OBJECTID) |
5843 | break; | |
5844 | ||
cb517eab | 5845 | log = btrfs_read_fs_root(log_root_tree, &found_key); |
79787eaa JM |
5846 | if (IS_ERR(log)) { |
5847 | ret = PTR_ERR(log); | |
34d97007 | 5848 | btrfs_handle_fs_error(fs_info, ret, |
79787eaa JM |
5849 | "Couldn't read tree log root."); |
5850 | goto error; | |
5851 | } | |
e02119d5 CM |
5852 | |
5853 | tmp_key.objectid = found_key.offset; | |
5854 | tmp_key.type = BTRFS_ROOT_ITEM_KEY; | |
5855 | tmp_key.offset = (u64)-1; | |
5856 | ||
5857 | wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key); | |
79787eaa JM |
5858 | if (IS_ERR(wc.replay_dest)) { |
5859 | ret = PTR_ERR(wc.replay_dest); | |
b50c6e25 JB |
5860 | free_extent_buffer(log->node); |
5861 | free_extent_buffer(log->commit_root); | |
5862 | kfree(log); | |
5d163e0e JM |
5863 | btrfs_handle_fs_error(fs_info, ret, |
5864 | "Couldn't read target root for tree log recovery."); | |
79787eaa JM |
5865 | goto error; |
5866 | } | |
e02119d5 | 5867 | |
07d400a6 | 5868 | wc.replay_dest->log_root = log; |
5d4f98a2 | 5869 | btrfs_record_root_in_trans(trans, wc.replay_dest); |
e02119d5 | 5870 | ret = walk_log_tree(trans, log, &wc); |
e02119d5 | 5871 | |
b50c6e25 | 5872 | if (!ret && wc.stage == LOG_WALK_REPLAY_ALL) { |
e02119d5 CM |
5873 | ret = fixup_inode_link_counts(trans, wc.replay_dest, |
5874 | path); | |
e02119d5 CM |
5875 | } |
5876 | ||
900c9981 LB |
5877 | if (!ret && wc.stage == LOG_WALK_REPLAY_ALL) { |
5878 | struct btrfs_root *root = wc.replay_dest; | |
5879 | ||
5880 | btrfs_release_path(path); | |
5881 | ||
5882 | /* | |
5883 | * We have just replayed everything, and the highest | |
5884 | * objectid of fs roots probably has changed in case | |
5885 | * some inode_item's got replayed. | |
5886 | * | |
5887 | * root->objectid_mutex is not acquired as log replay | |
5888 | * could only happen during mount. | |
5889 | */ | |
5890 | ret = btrfs_find_highest_objectid(root, | |
5891 | &root->highest_objectid); | |
5892 | } | |
5893 | ||
e02119d5 | 5894 | key.offset = found_key.offset - 1; |
07d400a6 | 5895 | wc.replay_dest->log_root = NULL; |
e02119d5 | 5896 | free_extent_buffer(log->node); |
b263c2c8 | 5897 | free_extent_buffer(log->commit_root); |
e02119d5 CM |
5898 | kfree(log); |
5899 | ||
b50c6e25 JB |
5900 | if (ret) |
5901 | goto error; | |
5902 | ||
e02119d5 CM |
5903 | if (found_key.offset == 0) |
5904 | break; | |
5905 | } | |
b3b4aa74 | 5906 | btrfs_release_path(path); |
e02119d5 CM |
5907 | |
5908 | /* step one is to pin it all, step two is to replay just inodes */ | |
5909 | if (wc.pin) { | |
5910 | wc.pin = 0; | |
5911 | wc.process_func = replay_one_buffer; | |
5912 | wc.stage = LOG_WALK_REPLAY_INODES; | |
5913 | goto again; | |
5914 | } | |
5915 | /* step three is to replay everything */ | |
5916 | if (wc.stage < LOG_WALK_REPLAY_ALL) { | |
5917 | wc.stage++; | |
5918 | goto again; | |
5919 | } | |
5920 | ||
5921 | btrfs_free_path(path); | |
5922 | ||
abefa55a | 5923 | /* step 4: commit the transaction, which also unpins the blocks */ |
3a45bb20 | 5924 | ret = btrfs_commit_transaction(trans); |
abefa55a JB |
5925 | if (ret) |
5926 | return ret; | |
5927 | ||
e02119d5 CM |
5928 | free_extent_buffer(log_root_tree->node); |
5929 | log_root_tree->log_root = NULL; | |
afcdd129 | 5930 | clear_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags); |
e02119d5 | 5931 | kfree(log_root_tree); |
79787eaa | 5932 | |
abefa55a | 5933 | return 0; |
79787eaa | 5934 | error: |
b50c6e25 | 5935 | if (wc.trans) |
3a45bb20 | 5936 | btrfs_end_transaction(wc.trans); |
79787eaa JM |
5937 | btrfs_free_path(path); |
5938 | return ret; | |
e02119d5 | 5939 | } |
12fcfd22 CM |
5940 | |
5941 | /* | |
5942 | * there are some corner cases where we want to force a full | |
5943 | * commit instead of allowing a directory to be logged. | |
5944 | * | |
5945 | * They revolve around files there were unlinked from the directory, and | |
5946 | * this function updates the parent directory so that a full commit is | |
5947 | * properly done if it is fsync'd later after the unlinks are done. | |
2be63d5c FM |
5948 | * |
5949 | * Must be called before the unlink operations (updates to the subvolume tree, | |
5950 | * inodes, etc) are done. | |
12fcfd22 CM |
5951 | */ |
5952 | void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans, | |
4176bdbf | 5953 | struct btrfs_inode *dir, struct btrfs_inode *inode, |
12fcfd22 CM |
5954 | int for_rename) |
5955 | { | |
af4176b4 CM |
5956 | /* |
5957 | * when we're logging a file, if it hasn't been renamed | |
5958 | * or unlinked, and its inode is fully committed on disk, | |
5959 | * we don't have to worry about walking up the directory chain | |
5960 | * to log its parents. | |
5961 | * | |
5962 | * So, we use the last_unlink_trans field to put this transid | |
5963 | * into the file. When the file is logged we check it and | |
5964 | * don't log the parents if the file is fully on disk. | |
5965 | */ | |
4176bdbf NB |
5966 | mutex_lock(&inode->log_mutex); |
5967 | inode->last_unlink_trans = trans->transid; | |
5968 | mutex_unlock(&inode->log_mutex); | |
af4176b4 | 5969 | |
12fcfd22 CM |
5970 | /* |
5971 | * if this directory was already logged any new | |
5972 | * names for this file/dir will get recorded | |
5973 | */ | |
5974 | smp_mb(); | |
4176bdbf | 5975 | if (dir->logged_trans == trans->transid) |
12fcfd22 CM |
5976 | return; |
5977 | ||
5978 | /* | |
5979 | * if the inode we're about to unlink was logged, | |
5980 | * the log will be properly updated for any new names | |
5981 | */ | |
4176bdbf | 5982 | if (inode->logged_trans == trans->transid) |
12fcfd22 CM |
5983 | return; |
5984 | ||
5985 | /* | |
5986 | * when renaming files across directories, if the directory | |
5987 | * there we're unlinking from gets fsync'd later on, there's | |
5988 | * no way to find the destination directory later and fsync it | |
5989 | * properly. So, we have to be conservative and force commits | |
5990 | * so the new name gets discovered. | |
5991 | */ | |
5992 | if (for_rename) | |
5993 | goto record; | |
5994 | ||
5995 | /* we can safely do the unlink without any special recording */ | |
5996 | return; | |
5997 | ||
5998 | record: | |
4176bdbf NB |
5999 | mutex_lock(&dir->log_mutex); |
6000 | dir->last_unlink_trans = trans->transid; | |
6001 | mutex_unlock(&dir->log_mutex); | |
1ec9a1ae FM |
6002 | } |
6003 | ||
6004 | /* | |
6005 | * Make sure that if someone attempts to fsync the parent directory of a deleted | |
6006 | * snapshot, it ends up triggering a transaction commit. This is to guarantee | |
6007 | * that after replaying the log tree of the parent directory's root we will not | |
6008 | * see the snapshot anymore and at log replay time we will not see any log tree | |
6009 | * corresponding to the deleted snapshot's root, which could lead to replaying | |
6010 | * it after replaying the log tree of the parent directory (which would replay | |
6011 | * the snapshot delete operation). | |
2be63d5c FM |
6012 | * |
6013 | * Must be called before the actual snapshot destroy operation (updates to the | |
6014 | * parent root and tree of tree roots trees, etc) are done. | |
1ec9a1ae FM |
6015 | */ |
6016 | void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans, | |
43663557 | 6017 | struct btrfs_inode *dir) |
1ec9a1ae | 6018 | { |
43663557 NB |
6019 | mutex_lock(&dir->log_mutex); |
6020 | dir->last_unlink_trans = trans->transid; | |
6021 | mutex_unlock(&dir->log_mutex); | |
12fcfd22 CM |
6022 | } |
6023 | ||
6024 | /* | |
6025 | * Call this after adding a new name for a file and it will properly | |
6026 | * update the log to reflect the new name. | |
6027 | * | |
d4682ba0 FM |
6028 | * @ctx can not be NULL when @sync_log is false, and should be NULL when it's |
6029 | * true (because it's not used). | |
6030 | * | |
6031 | * Return value depends on whether @sync_log is true or false. | |
6032 | * When true: returns BTRFS_NEED_TRANS_COMMIT if the transaction needs to be | |
6033 | * committed by the caller, and BTRFS_DONT_NEED_TRANS_COMMIT | |
6034 | * otherwise. | |
6035 | * When false: returns BTRFS_DONT_NEED_LOG_SYNC if the caller does not need to | |
6036 | * to sync the log, BTRFS_NEED_LOG_SYNC if it needs to sync the log, | |
6037 | * or BTRFS_NEED_TRANS_COMMIT if the transaction needs to be | |
6038 | * committed (without attempting to sync the log). | |
12fcfd22 CM |
6039 | */ |
6040 | int btrfs_log_new_name(struct btrfs_trans_handle *trans, | |
9ca5fbfb | 6041 | struct btrfs_inode *inode, struct btrfs_inode *old_dir, |
d4682ba0 FM |
6042 | struct dentry *parent, |
6043 | bool sync_log, struct btrfs_log_ctx *ctx) | |
12fcfd22 | 6044 | { |
3ffbd68c | 6045 | struct btrfs_fs_info *fs_info = trans->fs_info; |
d4682ba0 | 6046 | int ret; |
12fcfd22 | 6047 | |
af4176b4 CM |
6048 | /* |
6049 | * this will force the logging code to walk the dentry chain | |
6050 | * up for the file | |
6051 | */ | |
9a6509c4 | 6052 | if (!S_ISDIR(inode->vfs_inode.i_mode)) |
9ca5fbfb | 6053 | inode->last_unlink_trans = trans->transid; |
af4176b4 | 6054 | |
12fcfd22 CM |
6055 | /* |
6056 | * if this inode hasn't been logged and directory we're renaming it | |
6057 | * from hasn't been logged, we don't need to log it | |
6058 | */ | |
9ca5fbfb NB |
6059 | if (inode->logged_trans <= fs_info->last_trans_committed && |
6060 | (!old_dir || old_dir->logged_trans <= fs_info->last_trans_committed)) | |
d4682ba0 FM |
6061 | return sync_log ? BTRFS_DONT_NEED_TRANS_COMMIT : |
6062 | BTRFS_DONT_NEED_LOG_SYNC; | |
6063 | ||
6064 | if (sync_log) { | |
6065 | struct btrfs_log_ctx ctx2; | |
6066 | ||
6067 | btrfs_init_log_ctx(&ctx2, &inode->vfs_inode); | |
6068 | ret = btrfs_log_inode_parent(trans, inode, parent, 0, LLONG_MAX, | |
6069 | LOG_INODE_EXISTS, &ctx2); | |
6070 | if (ret == BTRFS_NO_LOG_SYNC) | |
6071 | return BTRFS_DONT_NEED_TRANS_COMMIT; | |
6072 | else if (ret) | |
6073 | return BTRFS_NEED_TRANS_COMMIT; | |
6074 | ||
6075 | ret = btrfs_sync_log(trans, inode->root, &ctx2); | |
6076 | if (ret) | |
6077 | return BTRFS_NEED_TRANS_COMMIT; | |
6078 | return BTRFS_DONT_NEED_TRANS_COMMIT; | |
6079 | } | |
6080 | ||
6081 | ASSERT(ctx); | |
6082 | ret = btrfs_log_inode_parent(trans, inode, parent, 0, LLONG_MAX, | |
6083 | LOG_INODE_EXISTS, ctx); | |
6084 | if (ret == BTRFS_NO_LOG_SYNC) | |
6085 | return BTRFS_DONT_NEED_LOG_SYNC; | |
6086 | else if (ret) | |
6087 | return BTRFS_NEED_TRANS_COMMIT; | |
12fcfd22 | 6088 | |
d4682ba0 | 6089 | return BTRFS_NEED_LOG_SYNC; |
12fcfd22 CM |
6090 | } |
6091 |